Encyclopedia of Membranes PDF - PDFCOFFEE.COM (2024)

C

CHA Zeolite Membrane Tsuneji Sano Department of Applied Chemistry, Hiroshima University, Graduate School of Engineering, Higashi-Hiroshima, Japan

Zeolites are a class of crystalline aluminosilicates with highly regular and open microporous structures. More than 200 types of zeolite frameworks have been identified by the Structure Commission of the International Zeolite Association. Zeolite membranes combine the great advantages of inorganic membranes, such as temperature stability and resistance against solvents, with the molecular sieving effect. Zeolite chabazite (CHA, where the three characters indicate the framework type) with Si/Al ratios of 2 ~ 3 is known to possess a threedimensional pore system with large ellipsoidal ˚ ) that are accessible via eightcages (6.7 10 A ˚ ). Figure 1 membered ring windows (3.8 3.8 A shows the framework structure of CHA zeolite (IZA web. 2013). High-silica CHA (Si/Al ratio > 5) has attracted great interest owing to its thermal and acid stabilities, and hence, application of this material to membrane has been widely investigated. The high-silica CHA zeolite membrane can separate light-gas mixtures of CO/N2, CO2/ CH4, H2/CH4, and H2/n-C4H10 with notably higher selectivity than that allowed by the # Springer-Verlag Berlin Heidelberg 2014 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_1-1

Knudsen mechanism (Kalipcilar et al. 2002). The high-silica CHA zeolite membrane also exhibits excellent dehydration performance for water/alcohol mixtures (Hasegawa et al. 2010). As the high-silica CHA zeolite has only been synthesized using expensive N,N,N-trimethyl-1adamantammonium cation (TMAda+) as a structure-directing agent (SDA). Very recently, success was achieved in synthesizing high-silica CHA zeolites with Si/Al ratios of 5–21 from FAU zeolite using the benzyltrimethylammonium cation (BTMA+) instead of the expensive TMAda+. The high-silica CHA zeolite synthesized by the interzeolite conversion of FAU zeolite has superior acid stability (structural and compositional stabilities) as compared to the CHA zeolite synthesized using TMAda+ (Yamanaka et al. 2012). Polycrystalline high-silica CHA zeolite membranes can be formed by the secondary growth of seed crystals on the outer surface of porous supports such as a porous a-alumina tube. Seed crystals are applied to the outer surface of the support tube by rubbing in order to implant the seed crystals for nucleation. The secondary growth solution with a chemical composition of SiO2:0.03Al2O3:0.2BTMA:0.1NaCl:10H2O is prepared from the dealuminated FAU, BTMAOH, NaCl, and distilled water. Thereafter, a hydrothermal reaction is carried out at 130 C for 7 days using a Teflon ®-lined autoclave. After cooling the autoclave, the support tube is recovered, washed with distilled water, and dried

2

CHA Zeolite Membrane

overnight in air at room temperature. Finally, the support tube is calcined at 550 C for 20 h. Figure 2 shows scanning electron micrographs (SEM) of the surface and cross section of the CHA membrane on the a-alumina support (Yamanaka et al. 2012). The high-silica CHA-type zeolite membrane prepared on the a-alumina tube is connected with the stainless steel tube using heat shrink tubing, and the tube is subsequently set in the conventional batch-type pervaporation apparatus. An acetic acid aqueous

solution (50/50 wt%) is used as the feed at 75 C. The compositions of the feed and the permeate are determined by FID-gas chromatography (5MS capillary column). The permeation flux and separation factor, a(H2O/ CH3COOH), are calculated from the following equations: Flux ðkg=m2 hÞ ¼ ðweight of permeate, kgÞ ðmembrane area, m2 Þ ðpermeation time, hÞ (1) Separation factor aðH2 O=CH3 COOHÞ ¼

½CH2 O =CCH3 COOH Permeate ½CH2 O =CCH3 COOH Feed

(2)

where the CCH3 COOH and CH2 O are the weight fractions of acetic acid and water, respectively. The permeate flux and separation factor, a(H2O/CH3COOH), are 7.9 kg/m2 h and ca. 2500, respectively, and the membrane performance is identical to that of commercially available NaLTA zeolite membranes used for the dehydration of alcohol solution (Sato et al. 2008). The long-range time courses of both the separation factor and the flux are listed

CHA Zeolite Membrane, Fig. 1 Framework structure of CHA zeolite (IZA web. 2013) CHA Zeolite Membrane, Fig. 2 SEM images of (a) outer surface and (b) cross section of CHA zeolite membrane (Reproduced from Yamanaka et al. (2012) with the permission of Elsevier)

a

b

CHA Al2O3 support 5 µm

5 µm

CHA Zeolite Membrane

3

CHA Zeolite Membrane, Table 1 Time course of CHA-type zeolite membrane performance for dehydration of 50 wt% acetic acid aqueous solution at 75 C (Reproduced from Yamanaka et al. (2012) with the permission of Elsevier) Permeation time/h 1 22 77 125 170

a(H2O/CH3COOH) 2500 2480 2500 2380 2480

Flux/kg/m2 h 7.96 7.88 7.91 7.80 7.80

in Table 1 (Yamanaka et al. 2012). The CHA membrane prepared by the interzeolite conversion of FAU zeolites has high potential for use in the separation of water from acidic organic solvents and is not limited to acetic acid.

References Hasegawa Y, Abe C, Nishioka M, Sato K, Nagase T, Hanaoka T (2010) Formation of high flux CHA-type zeolite membranes and their application to the dehydration of alcohol solutions. J Membr Sci 363:318–324 International Zeolite Association Web site (2013) http:// www.iza-online.org/ Kalipcilar H, Bowen TC, Noble RD, Falconer JL (2002) Synthesis and separation performance of SSZ-13 zeolite membranes on tubular supports. Chem Mater 14:3458–3464 Sato K, Sugimoto K, Nakane T (2008) Preparation of higher flux NaA zeolite membrane on asymmetric porous support and permeation behavior at higher temperatures up to 145 C in vapor permeation. J Membr Sci 307:181–195 Yamanaka N, Itakura M, Kiyozumi Y, Ide Y, Sadakane M, Sano T (2012) Acid stability evaluation of CHA-type zeolites synthesized by interzeolite conversion of FAU-type zeolite and their membrane application for dehydration of acetic acid aqueous solution. Microporous Mesoporous Mater 158:141–147

A

Aerobic Membrane Bioreactor A. Achilli1 and R. W. Holloway2 1 Environmental Resources Engineering Department, Humboldt State University, Arcata, CA, USA 2 Department of Civil and Environmental Engineering, Colorado School of Mines, Golden, CO, USA

Synonyms Cross-flow membrane bioreactor; Membrane bioreactor; Microfiltration membrane bioreactor; Submerged membrane bioreactor; Ultrafiltration membrane bioreactor

Introduction Aerobic membrane bioreactors (MBRs) are one of the leading technologies to achieve sustainability in wastewater treatment through reuse, decentralization, and low energy consumption (Fane and Fane 2005; Fawehinmi et al. 2005). In aerobic MBRs, aerated activated sludge is coupled with membrane process to remove dissolved contaminants (carbon and ammonia) and separate solids from the treated municipal or industrial wastewater. Carbon is removed by microorganisms that metabolize the carbon in the presence of dissolved oxygen for microbial # Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_7-1

growth and respiration (organic carbon reduced to carbon dioxide). Ammonia is removed through ammonia oxidation (nitrification). Nitrification is a microbially mitigated reduction process that occurs in an aquatic environment that contains moderate to high concentrations of ammonia and dissolved oxygen and low concentrations of organic carbon. In submerged MBRs (Fig. 1), microporous (microfiltration (MF) or ultrafiltration (UF) (▶ Ultrafiltration)) membranes are immersed in a bioreactor, and water is filtered (▶ Permeate) through the membranes using vacuum; suspended solids are retained in the system; and high levels of treatment (including nutrient removal) can be achieved (Judd 2006). The MBR replaces the two-stage conventional activated sludge process (biotreatment and clarification) with a single, integrated process (▶ Wastewater Treatment in Membrane Bioreactors). The advantages of MBRs over conventional treatment have been thoroughly reviewed (Stephenson et al. 2000), and they include product consistency, reduced footprint, reduced sludge production, and nearly complete suspended solid separation from the effluent. Additionally, MBR effluent may be suitable for use as irrigation water, as process water, or as a pretreatment for potable reuse applications (▶ Membrane Bioreactors for Reuse; ▶ Potable Water Production) (Lawrence et al. 2002). However, the establishment of membrane bioreactor technology (▶ Anaerobic Membrane

2 Aerobic Membrane Bioreactor, Fig. 1 Schematic representation of a submerged aerobic MBR system

Aerobic Membrane Bioreactor Membrane Cassette Membrane Channels

Activated Sludge

Product Water

Membranes

Bioreactor; ▶ Attached Growth Membrane Bioreactor; ▶ Membrane Bioreactors (MBR) for Plants; ▶ Submerged Membrane Bioreactor) has been slower than expected because decisionmakers view MBRs as high risk and costly compared to conventional technology (Judd 2006). To date, MBRs have been used to treat municipal and industrial wastewater where water reuse is desired, a small footprint is required, or stringent discharge standards exist (Kang et al. 2007; Yang et al. 2006).

Current Limitations One of the major limitations to widespread application of MBR technologies is to control membrane fouling with modest energy and chemical input (▶ Irreversible Fouling; ▶ Reversible Fouling) (Le-Clech et al. 2006). Membrane fouling (▶ Biological Fouling; ▶ Cake Layer; ▶ Inorganic Scaling) has been investigated from various perspectives, including the causes, characteristics, and mechanisms of fouling and methods, to prevent or reduce membrane fouling (Achilli et al. 2011; Judd 2005; Le-Clech et al. 2006; Meng et al. 2009; Wang and Wu 2009). Fouling markedly affects membrane cleaning (Membrane cleaning, 369690) and replacement intervals, system productivity, and membrane integrity; all of which are factors that affect energy requirements and costs (Judd 2006; Le-Clech et al. 2006).

Operation In order to operate conventional MBRs at constant flux, physical membrane-cleaning techniques are utilized; they include air scouring, backwashing, relaxation, or a combination of the three, depending on the membrane configuration (hollow fibers, flat sheet, or tubular). Air scouring is required for submerged MBR configurations to gas lift fresh sludge through the membrane bundle or cassette and to scour solids from the membrane surface. During backwashing, permeate is pumped in the opposite direction through the membrane, effectively removing most of the reversible fouling (▶ Backwashing). The efficiency of backwashing has been studied in detail, and the key parameters have been found to be frequency, duration, and intensity (▶ Backwashing Frequency) (Bouhabila et al. 2001; Psoch and Schiewer 2005, 2006). During membrane relaxation, permeate suction is stopped, and the back transport of foulants is naturally enhanced as reversibly attached foulants diffuse away from the membrane surface. Membrane backwashing and relaxation are regularly used for tubular and hollow fiber membranes to control fouling (Bouhabila et al. 2001; Hong et al. 2002; Psoch and Schiewer 2005, 2006; Smith et al. 2005). This is not the case for flat-sheet membranes that cannot be backwashed due to their inability to withstand pressure in the opposite direction of the operating flow; for this

Aerobic Membrane Bioreactor

reason, relaxation is used to control the fouling of these membranes (Le-Clech et al. 2006). Regardless of the membrane configuration, chemicals must be used at regular intervals to enhance physical cleaning (Le-Clech et al. 2005).

Cross-References ▶ Anaerobic Membrane Bioreactor ▶ Attached Growth Membrane Bioreactor ▶ Backwashing ▶ Backwashing Frequency ▶ Biological Fouling ▶ Cake Layer ▶ Inorganic Scaling ▶ Irreversible Fouling ▶ Membrane Bioreactor for Reuse ▶ Membrane Bioreactor (MBR) Plants ▶ Membrane Cleaning ▶ Permeate ▶ Potable Water Production ▶ Reversible Fouling ▶ Submerged Membrane Bioreactor ▶ Ultrafiltration ▶ Wastewater Treatment in Membrane Bioreactors

References Achilli A, Marchand EA, Childress AE (2011) A performance evaluation of three membrane bioreactor systems: aerobic, anaerobic, and attachedgrowth. Water Sci Technol 63:2999–3005 Bouhabila EH, Ben-Aim R, Buisson H (2001) Fouling characterisation in membrane bioreactors. Sep Purif Technol 22–23:123–132 Fane AG, Fane SA (2005) The role of membrane technology in sustainable decentralized wastewater systems. Water Sci Technol 51(10):317–325 Fawehinmi F, Lens P, Stephenson T, Rogalla F, Jefferson B (2005) The influence of operating conditions on extracellular polymeric substances (EPS), soluble

3 microbial products (SMP) and bio-fouling in anaerobic membrane bioreactors. In: IWA specialized conference on water environment Hong SP, Bae TH, Tak TM, Hong S, Randall A (2002) Fouling control in activated sludge submerged hollow fiber membrane bioreactors. Desalination 143:219–228 Judd S (2005) Fouling control in submerged membrane bioreactors. Water Sci Technol 51(6–7):27–34 Judd S (2006) The MBR book: principles and applications of membrane bioreactors in water and wastewater treatment. Elsevier Science, Oxford Kang Y-T, Cho Y-H, Chung E-H (2007) Development of the wastewater reclamation and reusing system with a submerged membrane bioreactor combined bio-filtration. Desalination 202:68–76 Lawrence P, Adham S, Barro L (2002) Ensuring water re-use projects succeed – institutional and technical issues for treated wastewater re-use. Desalination 152:291–298 Le-Clech P, Fane A, Leslie G, Childress A (2005) MBR focus: the operators’ perspective. Filtr Sep 42(5):20–23 Le-Clech P, Chen V, Fane TAG (2006) Fouling in membrane bioreactors used in wastewater treatment. J Membr Sci 284:17–53 Meng FG, Chae SR, Drews A, Kraume M, Shin HS, Yang FL (2009) Recent advances in membrane bioreactors (MBRs): membrane fouling and membrane material. Water Res 43(6):1489–1512 Psoch C, Schiewer S (2005) Critical flux aspect of air sparging and backflushing on membrane bioreactors. Desalination 175:61–71 Psoch C, Schiewer S (2006) Resistance analysis for enhanced wastewater membrane filtration. J Membr Sci 280:284–297 Smith PJ, Vigneswaran S, Ngo HH, Ben-Aim R, Nguyen H (2005) Design of a generic control system for optimising back flush durations in a submerged membrane hybrid reactor. J Membr Sci 255:99–106 Stephenson T, Judd S, Jefferson B, Brindle K (2000) Membrane bioreactors for wastewater treatment. IWA Publishing, London Wang ZW, Wu ZC (2009) A review of membrane fouling in MBRs: characteristics and role of sludge cake formed on membrane surfaces. Sep Sci Technol 44(15):3571–3596 Yang W, Cicek N, Ilg J (2006) State of the art of membrane bioreactors: worldwide research and commercial applications in north America. J Membr Sci 270:201–211

A

Affinity Membranes Francesca Militano and Lidietta Giorno Institute on Membrane Technology, National Research Council of Italy, ITM-CNR, National Research Council of Italy, Rende (CS), Calabria, Italy

Introduction Affinity separation methods rely on a “molecular recognition” phenomenon between species. A molecule, known as the ligand, is permanently bounded onto an inert matrix and specifically recognizes the molecule of interest, known as the ligate, that can be separated. The ligand can be a naturally occurring molecule, an engineered macromolecule, or a synthetic molecule linked to the matrix by covalent coupling. The ligandligate interaction is selective and reversible, enabling the separation and fine purification of biological substances such as proteins, peptides, and nucleic acids on the basis of its individual chemical structure or biological function (Wilson and Poole 2009). Among the separation techniques based on the affinity method, affinity chromatography is the most widely used. Due to the limitation associated to the traditional affinity chromatography with porous bead-packed columns (i.e., limited flow rate by pore diffusion), the membrane-based separation technique is gaining an increasing # Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_8-1

importance. Indeed, affinity membranes combine the specificity of affinity adsorption (of the common affinity resins) with the high productivity associated with filtration membranes. They provide low pressure separation systems without diffusional limitation, as the mass transfer is mainly governed by convection (Klein 2000). Microporous membranes, both in flat sheet and hollow fiber configuration, coupled with biological or biomimetic ligands have been used as affinity membrane chromatography supports (Zou et al. 2001).

Affinity Membrane Preparation The realization of affinity membranes usually involves three steps: (1) preparation of the basic membrane, (2) activation (functionalization) of the basic membrane, and (3) coupling of affinity ligands to the activated membrane. Membrane Material The membrane materials should possess some characteristics: (i) hydrophilicity to minimize the nonspecific adsorption of bioactive species, (ii) chemical and physical stability under harsh conditions used during ligand coupling and ligand-ligate complex formation, (iii) large surface area relative to membrane volume, (iv) biocompatibility when the membrane is used for blood treatment, and (v) presence of

2

Affinity Membranes

Affinity Membranes, Table 1 Some commonly used activation agents for affinity membrane preparation Activation agent Cyanogen bromide

Structure

N

C

Functional groups of the basic membrane –OH

Br

Carbonyldiimidazole

–OH, –NH2

O N

N

N

N

2-Amino-4,6-trichloro-s-triazine

–OH Cl N Cl

N N

NH2

Glutaraldehyde

–OH, –SH, –NH2 O

O

Divinyl sulfone

–OH, –SH O

CH2

S H2C

O

Epichlorohydrin

–OH, NH2

O Cl

functional groups required for the coupling of the ligands (such as –OH, –NH, –SH, –COOH). The commercially available materials used for affinity membrane preparation include organic (natural or synthetic polymer), inorganic, and composite materials. Among the first materials used for affinity membrane preparation, cellulose and its derivatives (regenerated cellulose, cellulose acetate) were the most common. These present a hydrophilic and biocompatible surface, low nonspecific adsorption, and abundant reactive hydroxyl groups that can be easily activated by different strategies for the ligand coupling. To improve the density of functional groups and increase the mechanical strength, composite cellulose membranes have been also prepared by chemical grafting with acrylic polymers. Another suitable membrane material is polyvinyl alcohol, thanks to its hydrophilicity and

biocompatibility. Like cellulose, it contains hydroxyl groups that can be easily activated. Polyamide and nylon have been also used for the preparation of affinity membranes, thanks to their good mechanical and chemical stability. Nylon membranes have a low concentration of amino groups to serve as functional groups for ligand coupling. Also, due to the hydrophobic surface, the membrane presents a high nonspecific adsorption of biomolecules during affinity separation process. Therefore, the hydrolysis of the membrane is generally performed to increase the density of reactive groups and to prevent electrostatic interaction with proteins. To overcome these problems, composite nylon membranes have been also prepared. Other materials that have been used for affinity membranes are poly(methyl methacrylate), polysulfone and its derivative, polycaprolactam, polyvinylidene difluoride, poly(ether-urethaneurea), and inorganic materials such as glass.

Affinity Membranes

3

Affinity Membranes, Table 2 Some examples of the use of affinity membranes for isolation and purification of biomolecules Membrane Cellulose and regenerated cellulose

Ligand Cibacron Blue F3GA Protein A/G Histidine

Ligate Alkaline phosphatase IgG Endotoxin

Poly(ethylene-co-vinyl alcohol) Nylon

Histidine

IgG

Cibacron Blue Protein G Histidine

Glucose-6-phosphate dehydrogenase IgG Endotoxin

Protein A

IgG

Purification Removal of endotoxin Purification

Trypsin Iminodiacetic acid (IDA) IDA-Cu2+

Trypsin inhibitor Histidine/tryptophan

Purification Purification

Lysozyme, cytochrome c, ribonuclease A

Purification

Polyvinylidene difluoride (PVDF) Polysulfone (PS)

Glass

Membrane Activation and Ligand Coupling If the basic membrane does not possess the functional groups for ligand coupling, it can be activated. In Table 1 are reported the commonly used activation agents with respect to the functional groups present on the basic membrane. The coupling of the ligand directly on the activated membranes may result in low binding specificity due to the low steric availability (in particular when the ligand is a small molecule). This problem is generally overcome by the introduction of a spacer molecule to the membrane prior to attaching the ligand, improving the ligand accessibility for the molecule to be separated. Spacer arms are bifunctional molecules able to react with both the membrane and ligand.

Affinity Ligands Affinity ligands can be classified into biospecific and pseudo-biospecific ligands (Klein 1991). Biospecific ligands are biomolecules such as antibodies, antigens, and proteins A and G that show

Application Purification Purification Removal of endotoxin Purification Purification

specificity for only one complementary biomolecule. Because of their selectivity, biomolecules have been the most widely used affinity ligands on affinity membrane separation technology. One of the most common applications is the use of immobilized monoclonal antibodies for immunoaffinity separation. Another important example is membranes with covalently coupled protein A or protein G for immunoglobulin purification from plasma, serum, or cell culture supernatants. Although the biospecific ligands possess high specificity for proteins, they have some limitation for large-scale application due to their poor stability and high price. The alternative approach to biospecific ligands involves the use of pseudo-biospecific ligands. These are usually molecules with higher chemical and physical stability than biomolecules. Pseudo-biospecific ligands can be distinguished by biological (amino acids, specially histidine, lysine, tryptophan) or non-biological molecules (dyes, chelated metal ions). The working principle of pseudo-biospecific ligands relies on the complementarity of

4

structural features of ligand and ligate rather than on a biological function. Immobilized dyes have been found to act as affinity ligands for a wide variety of biological molecules. For example, triazine-linked dyes have been used to mimic coenzymes that bind a number of dehydrogenases, hexokinases, and alkaline phosphatases; the reactive triazine groups can be linked to any matrix containing hydroxyl groups by mixing the two together. Cibacron Blue F3GA (a textile dye) has been employed as ligand in affinity membranes for the purification of over 80 enzymes and proteins. Another important example of pseudobiospecific affinity ligands is chelated metal ions used for the purification of histidine-tagged fusion proteins.

Molecular Imprinted Membranes A different approach for the preparation of affinity membranes is the use of molecularly imprinted polymeric materials. These are produced by entrapping a template molecule (the molecule to be separated) in a polymer matrix during polymerization and subsequent extraction. In this way, binding sites are introduced in the polymer that are complementary in shape and functionality to the target molecule.

Affinity Membranes

Applications of Affinity Membranes Affinity membranes are used for several different applications such as purification of biomolecules, removal of unwanted substances from biological fluids, and also small-scale analytical separations. The most common application is the separation and purification of biomolecules and especially proteins for large-scale production. In Table 2 are reported some typical examples of their use for the separation and purification of biomolecules.

References Klein E (ed) (1991) Affinity membranes: their chemistry and performance in adsorptive separation processes. Wiley, New York Klein E (2000) Affinity membranes: a 10-year review. J Membr Sci 179:1–27 Wilson ID, Poole CF (2009) Handbook of methods and instrumentation in separation science, vol 1. Academic, Boston Zou H, Luo Q, Zhou D (2001) Affinity membrane chromatography for the analysis and purification of proteins. J Biochem Biophys Methods 49:199–240

A

Air Dehydration by Membrane Technology Wei Liu Energy and Environmental Technology, Pacific Northwest National Laboratory, Richland, WA, USA

Synonym Membrane air dehydration Air dehydration refers to removal of moisture from humid air and is also termed as air dehumidification and air drying. Due to the ubiquitous presence of moisture in air, air dehydration represents one of the largest membrane applications. Air dehydration through a membrane is essentially a vapor-phase separation process (Fig. 1), which can be carried out in two ways. In one way, moisture in the feed air diffuses across the membrane and is swept out of the membrane unit with a sweep gas stream (Fig. 1a). In another way, the permeated moisture is pulled away from the membrane unit by vacuum (Fig. 1b). In the sweeping process, degree of air dehumidification will be limited by moisture content in the sweep gas and operation pressures, as explained by the following equations:

# Springer-Verlag Berlin Heidelberg 2014 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_11-1

Dp ¼ xF pF xS pP > 0 xF >

x P pP pF

Low molar fraction of moisture in sweep gas (xP), low permeate pressure (pP), and high feed pressure (pF) favor for deep dehumidification, i.e., allowing low moisture content in feed air (xF). If no sweep gas is used, the permeate pressure has to be decreased by vacuum to obtain a positive partial pressure gradient of water vapor across the membrane: pP < x F pF Membrane dehydration is commonly used in industries for removal of moisture from pressurized air. In these applications, ambient air at atmospheric pressure is often used as sweep gas. Though moisture content (such as molar fraction) in the sweep air may be the same as in the process air, a significant partial pressure gradient of water vapor can be managed due to high pressures in the process air. For example, hollow fiber membranes are used for drying of the air pressurized to about 10 bar (Morgan et al. 1996). Air-sweep dehumidification can also be found in some laboratory membrane devices, such as Nafion membrane tubes for moisture exchange (Ye and LeVan 2003).

2

Air Dehydration by Membrane Technology

Air Dehydration by Membrane Technology, Fig. 1 Removal of moisture from humid air through a membrane

WATER VAPOR HUMID AIR

1

LOW HUMIDITY AIR

DEHUMIDIFICATION UNIT

2

3

CONDENSER 4

“COMPRESSOR”

5

NON-CONDENSABLE GAS

VACUUM PUMP

LIQUID RESERVOIR

LIQUID PUMP

LIQUID

Air Dehydration by Membrane Technology, Fig. 2 Process flow diagram of a membrane dehumidification system

When humid air is at low or atmospheric pressures, using ambient air as a sweep stream is no longer practical, and pulling vacuum in the permeate side becomes necessary. In this case, several other pieces of equipment in addition to the membrane separator are required to make an integrated dehumidification system (Fig. 2). Such a membrane dehumidifier has been recently proposed and is still under development (Xing et al. 2013). An array of membrane units can be assembled together that the permeate sides of all the membrane units are connected to one common vacuum line. The vacuum is generated by the use of a vacuum pump or gas compressor. The permeated moisture is compressed to a pressure above water dew point at environmental temperature so that water vapor is condensed into liquidphase water by rejecting heat of condensation

into environment. The condensed water is collected in a reservoir. Water in the reservoir may be discharged using a liquid pump, while residual air – noncondensable gas – is discharged into environment using a secondary vacuum pump.

References Morgan WH, Bleikamp LK, Kalthod DG (1996) Hollow fiber membrane dryer with internal sweep, US patent no. 5,525,143 Xing R, Rao Y, TeGrotenhuis W, Canfield N, Zheng F, Winiarski DW, Liu W (2013) Advanced thin zeolite/ metal flat sheet membrane for energy efficient air dehumidification and conditioning. Chem Eng Sci 104:596–609 Ye X, LeVan MD (2003) Water transport properties of Nafion membranes: part I. Single-tube membrane module for air drying. J Membr Sci 221:147–161

A

Air Enrichment, by Polymeric Magnetic Membranes Anna Strzelewicz Faculty of Chemistry, Silesian University of Technology, Gliwice, Poland

The idea of using magnetic membranes for enrichment of air by oxygen is based on the observation that oxygen and nitrogen have quite different magnetic properties, i.e., oxygen is paramagnetic whereas nitrogen diamagnetic. The oxygen molecule is paramagnetic with a magnetic moment of mO2 ¼ 2:73 1023 JT 1 : Magnetic susceptibility of nitrogen is equal to w ¼ 150:8 106 mol1 ¼ 2:5 1028 molecule1 which corresponds only to m ¼ 2:5 1028 JT 1 in a magnetic field of 1 T, a value five orders of magnitude smaller than the O2 magnetic moment (Morrish 1965; Borys et al. 2011). Magnetic membranes are polymeric membranes (ethyl cellulose (EC) or poly (2,6-dimethyl-1,4-phenylene oxide) (PPO)) with dispersed magnetic powder (ferrite, praseodymium, and neodymium). The membranes are made by casting of appropriate polymer solution with dispersed magnetic powder in an external magnetic field of a specially designed coil (stable magnetic field with range of induction 0–40 mT). Removed from Petri dish membranes # Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_12-4

are dried in 40 C for at least 2 days and then are stored in an exsiccator under the vacuum conditions (p = 3 mmHg). Collection of permeation quantities both for individual gases (O2, N2) and for their mixture (air 21 %O2/79 % N2) is measured in experimental setup (Fig. 1). The measurements are carried out in room temperature for membranes with dispersed magnetic powder before and after magnetization in a field magnet with magnetic induction about 2.5 T. The setup furnished with a gas chromatograph allows to measure the oxygen and the nitrogen concentration in permeate. The main part of the experimental setup is diffusive chamber, where the membrane is put in the form of disk. The setup is used for a low-pressure (from 0.1 to 1.0 MPa) analysis of gas permeation. Transport coefficients can be calculated using flow rate data and percentage of air enrichment. The flow rate of the permeate can be recorded using a flowmeter. Integration of flux with respect to time allows to sketch a downstream permeation curve. Typical plot of Qa(l,t) versus time is presented in the Fig. 2. Time-lag method and D1-D8 system allow to get some insight into the nature of the considered transport process. In papers (Strzelewicz and Grzywna 2007, 2008; Rybak et al. 2009a, b, 2012; Grzywna et al. 2010) a concept of magnetic membranes for air enrichment is explained. The authors observed an increase in the oxygen flux with respect to the nitrogen flux and the enrichment in the oxygen content of the permeate up to

2

Air Enrichment, by Polymeric Magnetic Membranes

Air Enrichment, by Polymeric Magnetic Membranes, Fig. 1 Scheme of the experimental setup

Air Enrichment, by Polymeric Magnetic Membranes, Fig. 2 Downstream absorption permeation curve

Q

a

stationary state

point for checking oxygen content

a

Q (I, t)

La(L)

t

– C0 1 6

55.6 % for ethyl cellulose magnetic membranes. Further permeation measurements done in polyphenylene oxide (PPO) magnetic membranes provided higher enrichments, up to 61.9 % (Table 1). The transport of the molecules through the magnetic membranes can be modeled by a diffusion with a position-dependent diffusion coefficient. Such a diffusion coefficient reflects the changes in the membrane composition along the permeation axis. One can observe on the scanning electron microscope photograph (Borys

et al. 2011) of the magnetic membrane cross section that the feed side of the membrane is composed of a pure polymer while the output side consists of the polymer with dispersed magnetic granules. When the magnetic field is too strong, then magnetic aggregates are created, which influence the permeation of gases. Detailed analysis of the available data and microscopy images allowed to arrive four conclusions (Borys et al. 2011):

Air Enrichment, by Polymeric Magnetic Membranes

3

Air Enrichment, by Polymeric Magnetic Membranes, Table 1 Air enrichment for various membranes (Strzelewicz and Grzywna 2007) Membrane EC + 1.38 g Nd EC + 1.49 g Nd PPO + 1.80 g Nd PPO + 1.80 g Nd

B[mT] 0.79 1.25 1.70 2.70

1. There are magnetic channels formed around the magnetic granules. 2. The channels provide high permeability “highways” for the diffusion of permeating molecules. 3. The oxygen molecules, due to their paramagnetic properties, stick to these “highways” for a longer time than nitrogen, which is probably based on the interaction with the Weiss molecular field of the permanent magnet. 4. The magnetic field induces aggregation between oxygen and nitrogen which enhances the transport of both nitrogen and oxygen by prolonging their residence in the channel. The method of air enrichment by magnetic membranes seems to be effective and efficient.

References Borys P, Pawelek K, Grzywna ZJ (2011) On the magnetic channels in polymer membranes. Phys Chem Chem Phys 13:17122–17129

Oxygen content in permeate [%] 40.7 1.1 43.8 1.1 54.1 1.4 61.9 1.5

Grzywna ZJ, Rybak A, Strzelewicz A (2010) Air enrichment by polymeric magnetic membrane. In: Yampolskii Y, Freeman B (eds) Membrane gas separation. Wiley, Chichester, pp 159–182 Morrish AH (1965) The physical principles of magnetism, 1st edn. Wiley, New York Rybak A, Krasowska M, Strzelewicz A, Grzywna ZJ (2009a) “Smoluchowski type” equations for modelling of air separation by membranes with various structure. Acta Phys Pol B 40:1447–1454 Rybak A, Grzywna ZJ, Kaszuwara W (2009b) On the air enrichment by polymer magnetic membranes. J Membr Sci 336:79–85 Rybak A, Strzelewicz A, Krasowska M, Dudek G, Grzywna ZJ (2012) On the air separation process by magnetic membranes influence of various parameters. Sep Sci Technol 47:1395–1404 Strzelewicz A, Grzywna ZJ (2007) Studies on the air membrane separation in the presence of magnetic field. J Membr Sci 294:60–67 Strzelewicz A, Grzywna ZJ (2008) On the permeation time lag for different transport equations by Frisch method. J Membr Sci 322:460–465

A

Air Separation A. F. Ismail1,2 and M. H. D. Othman1,2 1 Advanced Membrane Technology Research Centre (AMTEC), Universiti Teknologi Malaysia, Skudai, Johor, Malaysia 2 Faculty of Chemical and Energy Engineering, Universiti Teknologi Malaysia, Skudai, Johor, Malaysia

Air separation technology is used for the production of oxygen, nitrogen, and rare gases that are present in air such as argon and neon. There are two fundamental approaches to air separation, which are cryogenic and non-cryogenic processes. The cryogenic process which is carried out in distillation column has the capability to deliver large and high purities of products, while the non-cryogenic which is based on absorption and membrane technologies is more suitable for on-site production, which is most common for small and medium throughputs. Membrane technology for air separation has developed rapidly in recent years. Polymeric and ceramic membranes have been used commercially for oxygen production. Polymeric membranes operate based on the difference in rates of diffusion of oxygen and nitrogen through a membrane. Due to the smaller size of the oxygen molecule, most membrane materials are more permeable to oxygen than to nitrogen. Materials such as polysulfone or acetate membranes make # Springer-Verlag Berlin Heidelberg 2014 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_13-2

it possible to permeable oxygen five times than of nitrogen. Membrane units capable of producing nearly 600 tonnes per day nitrogen have been built (Castle 2002). A major benefit of polymeric membrane separation is the simple, continuous nature of the process and operation at near ambient conditions (Smith and Klosek 2001); however, the low separation factor of two to six limits the polymeric membrane to produce oxygen-enriched air rather than pure oxygen (Zhu et al. 2008). Production of pure oxygen from air can be achieved by using ceramic membrane system at elevated temperatures, typically in the range of 800–900 C (Hashim et al. 2011). The oxygen transporting through this type of membrane is in the form of oxygen ions instead of oxygen molecules; therefore, the pure oxygen is obtained. Enormous efforts have been directed to ceramic membranes with mixed ionic–electronic conducting (MIEC) characteristics. Among them, perovskite-type (ABO3) ceramic membranes exhibit the highest oxygen permeability due to their high ionic and electronic conductivity. The perovskite oxide based on La1 xAxCo1 yFeyO3 d achieved very high oxygen permeation fluxes as reported by Teraoka et al. (1985). Interestingly, the oxygen separation from air in MIEC ceramic membrane system requires neither electrodes nor an external circuit to operate. As depicted in Fig. 1, the electronic conductivity itself performs as an internal short circuit that

2 Air Separation, Fig. 1 Schematic representation of the oxygen transport in MIEC ceramic membrane

Air Separation

Pressurized air feed MIEC ceramic membrane

O2 Oxygen rich stream

involves oxygen partial pressure gradient. The oxygen molecule permeates from the high oxygen partial pressure side to the low oxygen partial pressure side, while the overall charge neutrality is maintained by counterbalancing the flux of electrons (Liu and Gavalas 2005). Several industrial gas companies are working on developing ceramic membranes for oxygen separation from air at high temperatures. Air Products and Chemicals has developed an ion transport membrane (ITM) system, which is based on patented, high-temperature ceramic membranes for the production of oxygen from air separation (Hashim et al. 2011). Praxair is also working on oxygen-conducting ceramic membrane systems that are specially designed to separate oxygen from air at elevated temperature environment (Hashim et al. 2011).

½O 2 + 2e − → O2−

e-

O2-

O2− → ½O 2 + 2e −

Oxygen depleted air

eSweep gas

References Castle WF (2002) Air separation and liquefaction: recent developments and prospects for the beginning of the new millennium. Int J Refrig 25:158–172 Hashim SS, Mohamed AR, Bhatia S (2011) Oxygen separation from air using ceramic-based membrane technology for sustainable fuel production and power generation. Renew Sustain Energy Rev 15:1284–1293 Liu S, Gavalas GR (2005) Oxygen selective ceramic hollow fiber membranes. J Membr Sci 246:103–108 Smith AR, Klosek J (2001) A review of air separation technologies and their integration with energy conversion processes. Fuel Process Technol 70:115–134 Teraoka Y, Zhang HM, Furukawa S, Yamazoe N (1985) Oxygen permeation through perovskite-type oxides. Chem Lett 167:1743–1746 Zhu X, Sun S, He Y, Cong Y, Yang W (2008) New concept on air separation. J Membr Sci 323:221–224

A

Alamine 336 Karel Friess Department of Physical Chemistry, University of Chemistry and Technology Prague, Prague, Czech Republic

Synonyms Alamine 3365; Alamine 336S; Farmin08; Octanamine; Tricaprylamine; Tridioctylamine; Tri-n-octylamine; Tri-n-caprylylamine; Trioctylamine The IUPAC systematic name is N,N-dioctyl-1octanamine (Fig. 1). Alamine 336 is colorless (C24H51N, CAS Reg. No. 1116-76-3) or light yellow liquid (mixture of Tri C8-10 Alkyl Amines, C27H57N – CAS Reg. No. 57176-40-6, produced by Cognis Corp., now part of BASF). If released to air, estimated vapor pressure is about

# Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_15-1

0.007 Pa at 25 C. It is moderately toxic by ingestion and intraperitoneal routes; when heated to decomposition, it emits toxic vapors of NOx (Zhu et al. 2012). Manufacturing of Alamine 336 is possible by catalytic amination of octanol (Li et al. 2011) or by catalytic hydrogenation of caprylonitrile (Lazier 1940). It is used as the extractant for reactor fuel processing (Moyer and McDowell 1981), for dye identification (Puttemans et al. 1982), and for metal adsorption (Tasker et al. 2003) and recovery from diluted aqueous solutions (Kislik 2012; Coca et al. 1990; Sun and Lee 2011). Mixtures of Alamine 336 with meta-xylene can be successfully applied for extraction metal ions from their strong acidic aqueous chloride solutions (Sayar et al. 2007). Alamine 336 can be also used for specific extractions in biotechnology, e.g., separation of carboxylic acids (Tamada et al. 1990; Yordanov and Boyadzhiev 2004), and in supported liquid membranes (San Román et al. 2010; Dżygiel and Wieczorek 2010) (Table 1).

2

Alamine 336

Alamine 336, Fig. 1 Schematic structure of Alamine 336

N

Alamine 336, Table 1 Properties of Alamine 336 (Steele et al. 1996) Molar mass Melting point Boiling point Density Viscosity Surface tension Constant pressure heat capacity of liquid Solubility in chloroform Solubility in water

Cross-References ▶ Extraction ▶ Metal Adsorption and Recovery ▶ Supported Liquid Membranes ▶ Tertiary Amine ▶ Water Treatment

References ChemSpider (CSID). http://www.chemspider.com/ Chemical-Structure.13591.html. Accessed 1 June 2012 Coca J, Díez FV, Morís MA (1990) Solvent extraction of molybdenum and tungsten by Alamine 336 and DEHPA. Hydrometallurgy 25:125–135 Dżygiel P, Wieczorek PP (2010) Supported liquid membranes and their modifications: definition, classification, theory, stability, application and perspectives. In: Kislik VS (ed) Liquid membranes. Elsevier, Amsterdam, pp 73–140

353.67 34.6 363.5 0.818 7.8610 3.4810 750.8 0.1 0.050

2 2

gmol 1 C C gcm 3 at 25 C Pas at 34.6 C Nm 1 at 34.6 C Jmol 1K 1 at 25 C gcm 3 at 25 C mgdm 3 at 25 C

Kislik VS (2012) Examples of application of solvent extraction techniques in chemical, radiochemical, biochemical, pharmaceutical, analytical separations, and wastewater treatment. In: Kislik VS (ed) Solvent extraction – classical and novel approaches. Elsevier, Amsterdam, pp 185–314 Lazier WA (1940) Process for catalytic hydrogenation of higher aliphatic nitriles. US Patent 2225059 Li Y, Li Q, Zhi L, Zhang M (2011) Catalytic amination of octanol for synthesis of trioctylamine and catalyst characterization. Catal Lett 141:1635–1642 Moyer BA, McDowell WJ (1981) Factors influencing phase disengagement rates in solvent extraction systems employing tertiary amine extractants. Sep Sci Technol 16:1261–1289 National Institute of Standards and Technology (NIST) database. http://webbook.nist.gov. Accessed 1 June 2012 Puttemans ML, Dryon L, Massart DL (1982) Evaluation of thin layer-, paper- and high performance liquid chromatography for the identification of dyes extracted as ion-pairs with tri-n-octylamine. J Assoc Off Anal Chem 65:730–736 San Román MF, Bringas E, Iban˜ez R, Ortiz I (2010) Liquid membrane technology: fundamentals and

Alamine 336 review of its applications. J Chem Technol Biotechnol 85:2–10 Sayar NA, Filiz M, Sayar AA (2007) Extraction of Zn (II) from aqueous hydrochloric acid solutions into Alamine 336–m-xylene systems. Modeling considerations to predict optimum operational conditions. Hydrometallurgy 86:27–36 Steele WV, Chirico RD, Knipmeyer SE, Nguyen A, Smith NK, Tasker IR (1996) Thermodynamic properties and ideal-gas enthalpies of formation for cyclohexene, phthalan (2,5-dihydrobenzo-3,4-furan), isoxazole, octylamine, dioctylamine, trioctylamine, phenyl isocyanate, and 1,4,5,6-tetrahydropyrimidine. J Chem Eng Data 41:1269-1284

3 Sun PP, Lee MS (2011) Separation of Pt(IV) and Pd (II) from the loaded Alamine 336 by stripping. Hydrometallurgy 109:181–184 Tamada JA, Kertes AS, King CJ (1990) Extraction of carboxylic acids with amine extractants. Ind Eng Chem Res 29:1319–1326 Tasker PA, Plieger PG, West LC (2003) Metal complexes for hydrometallurgy and extraction. In: McCleverty JA, Meyer TJ (eds) Comprehensive coordination chemistry II, vol 9. Elsevier: Oxford, UK,. pp 759–808 Yordanov B, Boyadzhiev L (2004) Pertraction of citric acid by means of emulsion liquid membranes. J Membr Sci 238:191–197 Zhu Y, Cao M, Ma X, Xu C, Wang X, Ren L, Hu C (2012) Dalton Trans 41(10):2935–40

A

Alcohol and Water Separation Tadashi Uragami Organization for Research and Development of Innovative Science and Technology (ORDIST), Kansai University, Suita, Osaka, Japan

Cross-linked poly(vinyl alcohol) (PVA) composite membranes have been used in commercial PV plants for dehydration of ethanol beyond the azeotrope. However aqueous ethanol solutions that can be produced by bio-fermentation are dilute (about 10 wt.%). Therefore, if ethanol/ water-selective membranes with high efficiency can be prepared, the distillation process in the first stage to obtain an azeotrope can be replaced ethanol-/water-selective membrane which is very advantageous for reduction of energy cost. There are fewer reports on ethanol-/water-selective membranes compared with those of water-/ ethanol-selective membranes. One reason why the development of efficient high-performance ethanol-/water-selective membranes is difficult can be attributed to the fact that ethanol has a larger molecular size than water and must be preferentially permeated through the membrane. In fact, permeation and separation in a

# Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_16-2

pervaporation (PV) process through dense membranes are based on the solution-diffusion mechanism (Binding et al. 1961; Aptel et al. 1974). Therefore, when it is required that ethanol molecules with larger molecular size preferentially permeate from an aqueous ethanol solution, it cannot be expected to be separated by the diffusion process. Consequently, only a difference of solubility selectivity in the solution process in which both ethanol and water components are dissolved can contribute to the separation. Figure 1 shows the ethanol concentration in the permeate through a poly(dimethylsiloxane) (PDMS) membrane during PV and that sorbed into a PDMS membrane. These results support the hypothesis that the difference in the solubility of the permeants contributes to the ethanol/water selectivity. PDMS membranes show high ethanol/water selectivity, but their mechanical strength is weak, and it is difficult to prepare thin membranes from PDMS. In order to obtain both ethanol/water selectivity and mechanical strength, graft copolymers composed of PDMS macromonomer and vinyl monomers were synthesized. Graft copolymer membranes, which were either ethanol or water selective, were prepared by copolymerization of a dimethylsiloxane

2

Alcohol and Water Separation

EtOH in permeate (wt%)

ve

80

ium

id liqu

ibr qul

cur

80

e

or-

60

vap

60

40

40

20

20

0 0

20 40 60 80 EtOH in feed solution (wt%)

EtOH in membrane (wt%)

100

100

100

Alcohol and Water Separation, Fig. 1 Permeation and separation characteristics of aqueous ethanol solutions through a PDMS membrane during PV

Alcohol and Water Separation, Fig. 2 Effects of the DMS content on the normalized permeation rate (○) and separation factor (●) through the PMMA-g-PDMS membrane during PV. Feed: aqueous solution of 10 wt.% ethanol. Dashed line is the feed composition

(DMS) macromonomer with methyl methacrylate (MMA) (Miyata et al. 1995, 1996). Two glass transition temperatures (Tg) were observed at about 120 C and 127 C in the graft copolymer membranes. Transmission electron micrograph (TEM) demonstrated that the PMMA-g-PDMS membranes showed microphase-separated structures. When an aqueous solution of 10 wt.% ethanol was permeated through the PMMA-g-PDMS membranes by PV, the ethanol concentration in the permeate and the permeation rate increased drastically with the DMS content in the copolymer. In particular, at a DMS content of less than 40 mol%, water permeates preferentially from an aqueous solution of 10 wt.% ethanol, whereas membranes with more than about 40 mol% of DMS are ethanol/water selective, as shown in Fig. 2. The change in the ethanol/water selectivity of the PMMA-g-PDMS membranes can be explained by a microphase-separated polymer structure using Maxwell’s model and a combined model consisting of both parallel and series expressions. Furthermore, image processing of TEMs allowed

the determination of the percolation transition of the PDMS phase at a DMS content of about 40 mol%. These results suggest that the continuity of the PDMS phases in the microphaseseparated PMMA-g-PDMS membranes directly affects their ethanol/water selectivity for aqueous ethanol solutions (Miyata et al. 1995, 1996). In Table 1, the performance of the ethanol-/ water-selective polymer membranes is compared. As can be seen in this Table, the addition of PFA-g-PDMS to the PTMSP membrane in PV was very effective, the application of TDEV method to the membrane separation technique was also very interesting for the enhancement of the ethanol/water selectivity for the ethanol/ water mixtures, and in particular the application of porous PDMS membrane to temperaturedifference controlled evapomeation (TDEV) (Uragami 1998, 2005, 2006a, b, 2008, 2010, 2011; Uragami and Shinomiya 1991, 1992; Uragami and Tanaka 1991, 1993, 1994; Uragami et al. 2002; Uragami and Morikawa 1989) was a very excellent performance for the ethanol/water mixtures.

Alcohol and Water Separation

3

Alcohol and Water Separation, Table 1 Performance of ethanol-/water-selective membranes

Membrane PDMS PTMSP PTMSP PEA-g-PDMS/ PTMSPb PPP-g-PDMS PSt-g-PhdFDA (7.6/ 12.4) TFE/i-OcVE/C18VE terpolymer (50/25/25) Modified silicone

Feed (wt %) 7 7 10 10

Method PV PV PV PV

Applied temperature ( C) 25 25 30 40

aEtOH/

NPRa (kgmm (m2h) 1) 2.1 1.1 4.5 24.1

5.5 0.6

References Eustache and Histi (1981) Ishihara et al. (1986) Masuda et al. (1986) Uragami et al. (2000); Uragami and Shinomiya (1991) Nagase et al. (1989) Ishihara and Matsui (1987)

7.13

5

Kashiwagi et al. (1988)

40

3.65

11

30/40 40 30/40

19.3 7.44 85.7

16.6 6.4 0.9

40

7.1

4.8

Uragami and Shinomiya (1991) Miyata et al. (1996) Miyata et al. (1995, 1996) Uragami and Shinomiya (1992) Miyata et al. (1996)

PV

40

8

5.1

Miyata et al. (1999a)

10

PV

40

6.8

3.5

Miyata et al. (1999b)

10 10

TDEV TDEV

0/40 0.5

77.5 23.1

7.28 8

PV PV

30 30

15

PV

50

10

PV

Modified silicone PDMS PDMS

10 10 10

TDEV PV TDEV

PMMA-g-PDMS (34/66) PMMA-g-PDMS (27/73) PMMA-g-PDMS (38/62)c PTMST Porous PDMS

10

PV

10

H2O

11.8 11.2 12 20

22.5 45.9

38 1,250

Uragami (2010) Uragami (2008)

a

Normalized permeation rate PFA-g-PDMS is 0.2 wt.% c Annealing is 120 C, 2 h b

References Aptel P, Cuny J, Jozenfonvice J, Morel G, Neel J (1974) Liquid transport through membranes prepared by grafting of polar monomer onto poly(tetrafluoroethylene) films. II. Some factors determining pervaporation rate and selectivity. J Appl Polym Sci 18:365 Binding RC, Lee RJ, Jennings JF, Mertic EC (1961) Separation of liquid mixtures by pervaporation. Ind Eng Chem 53:47 Eustache H, Histi G (1981) Separation of aqueous organic mixtures by pervaporation and analysis by mass spectrometry or a coupled gas chromatograph-mass spectrometer. J Membr Sci 8:105

Ishihara K, Matsui K (1987) Ethanol permselective polymer membranes 3. Pervaporation of ethanol water mixture through composite membranes composited of styrene fluoroalkyl acrylate graft copolymers and cross-linked polydimethylsiloxane membrane. J Appl Polym Sci 34:437 Ishihara K, Nagase Y, Matsui K (1986) Pervaporation of alcohol/water mixtures through poly [1-(trimethylsilyl)-1-propyne] membrane. Macromol Chem Rapid Commun 7:43 Kashiwagi T, Okabe K, Okita K (1988) Separation of ethanol from ethanol/water mixtures by plasmapolymerized membranes from silicone compounds. J Membr Sci 36:353

4 Masuda T, Tang B-Z, Higashimura T (1986) Ethanolwater separation by pervaporation through substituted-polyacetylene membranes. Polym J 18:565 Miyata T, Takagi T, Kadota T, Uragami T (1995) Characteristics of permeation and separation for aqueous ethanol solutions through methyl methacrylatedimethylsiloxane graft copolymer membranes. Macromol Chem Phys 196:1211 Miyata T, Takagi T, Uragami T (1996) Microphase separation in graft copolymer membranes with pendant oligodimethylsiloxanes and their permselectivity for aqueous ethanol solutions. Macromolecules 29:7787 Miyata T, Obata S, Uragami T (1999a) Morphological effects of microphase separation on the permselectivity for aqueous ethanol solutions of block and graft copolymer membranes containing poly(dimethylsiloxane). Macromolecules 32:3712 Miyata T, Obata S, Uragami T (1999b) Annealing effect of microphase-separated membranes containing poly (dimethylsiloxane) on their permselectivity for aqueous ethanol solutions. Macromolecules 32:8465 Nagase Y, Mori S, Matsui K (1989) Chemical modification of poly(substituted-acetylene). 4. Pervaporation of organic liquid water mixture through poly(1-phenyl-1propylene) polydimethylsiloxane graft copolymer membrane. J Appl Polym Sci 37:1259 Uragami T (1993) Separation of aqueous organic liquid solutions through polymer membranes. Desalination 90:325 Uragami T (1998) Structures and properties of membranes from polysaccharide derivatives. In: Dumitriu S (ed) Polysaccharide, structural diversity and functional versatility. Marcer Dekker, New York/Basel/Hong Kong, pp 887–924 Uragami T (2005) Structures and functionalities of membranes from polysaccharide derivatives. In: Dumitriu S (ed) Polysaccharide, structural diversity and functional versatility, 2nd edn. Marcer Dekker, New York, pp 1087–1122 Uragami T (2006a) Separation materials derived from chitin and chitosan. In: Uragami T, Tokura S (eds) Material science of chitin and chitosan. KODANSHA, Springer, Berlin/Heiderberg/New York/Tokyo, pp 113–163 Uragami T (2006b) Polymer membranes for separation of organic liquid mixtures. In: Yampolskii Y, Pinau I, Freeman BD (eds) Materials science of membranes for gas and vapor separation. Wiley, Chichester, pp 355–372 Uragami T (2008) Structural design of polymer membranes for concentration of bio-ethanol. Polym J 40:485 Uragami T (2010) Selective membranes for purification and separation of organic liquid mixtures. In: Drioli E,

Alcohol and Water Separation Giorno L (eds) Comprehensive membrane science and engineering, vol 2, Membrane operations in molecular separations. Elsevier, Amsterdam/Boston/Heidelberg/ London/New York/Oxford/Paris/San Diego/San Francisco/Singapore/Sydney/Tokyo, pp 273–324 Uragami T (2011) Concentration of bio-ethanol through cellulose ester membranes during temperaturedifference controlled evapomeation. Mater Sci Appl 2:1499 Uragami T (2012) Concentration of bio-ethanol through poly[1-(trimethylsilyl)-1-propyne] membranes during temperature-difference controlled evapomeation. 11th World Filtration Congress & Exhibition, p 313 Uragami T, Morikawa T (1989) Permeation of ethanol through polydimethyl siloxane membranes using temperature-difference in membrane process of the evapomeation method. Makromol Chem Rapid Commun 10:287 Uragami T, Morikawa T (1992) Permeation and separation characteristics of alcohol-water mixtures through dimethylsiloxane membrane by pervaporation and evapomeation. J Appl Polym Sci 44:2009 Uragami T, Shinomiya H (1991) Concentration of aqueous alcoholic solutions through a modified silicone rubber membrane by pervaporation and evapomeation. Makromol Chem 192:2293 Uragami T, Tanaka Y (1991) Method of separating liquid component from a solution containing two or more liquid component. European Patent 0,346,739 Uragami T, Shinomiya H (1992) Concentration of aqueous dimethyl sulfoxide solutions through a chitosan membrane by permeation with a temperature difference. J Membr Sci J Membr Sci 7:183 Uragami T, Tanaka Y (1993) Method of separating liquid component from a solution containing two or more liquid component. U.S. Patent 5,271,846 Uragami T, Tanaka Y (1994) Separation method for mixed solutions. Japanese Patent 1,906,854 Uragami T, Doi T, Miyata T (1999) Control of permselectivity with surface modifications of poly [1-(trimethylsilyl)-1-propyne] membranes. Inter J Adhes Adhes 19:405 Uragami T, Doi T, Miyata T (2000) Pervaporation property of surface modified poly[1-(trimethylsilyl)1propyne] membranes. In: Pinnau I, Freeman BD (eds) Membrane formation and modification, ACS Symposium Series 744. American Chemical Society, Washington, DC, pp 263–279 Uragami T, Tanaka Y, Nishida S (2002) Permeation and separation under high temperature and high pressure for ethanol/water vapors through cross-linked quaternized chitosan composite membranes. Desalination 147:449

A

Alkaline Fuel Cells (AFCs) Tanja Vidakovic-Koch Max Planck Institute for Dynamics of Complex Technical Systems, Magdeburg, Germany

Alkaline fuel cell (AFC) is a fuel cell type which utilizes alkaline electrolyte, usually potassium hydroxide. It consumes hydrogen and oxygen producing only water, heat, and electricity. Depending on a concentration of potassium hydroxide, AFC can operate at temperatures between 60 C and 250 C. The fuel cell reactions are as follows: Anode : H2 þ 2OH ⇄ 2H2 O þ 2e Cathode : 1⁄2 O2 þ 2e þ 2H2 O ⇄ 2OH Overall reaction : H2 þ 1⁄2 O2 ⇄ H2 O The main ionic charge carriers are OH ions which mitigate from the cathode to the anode. Water is formed at the anode side and has to be removed from the system in order to prevent KOH dilution. The AFC has improved cathode performance compared to acidic fuel cells due to more favorable oxygen reduction reaction kinetics. For this reason, AFC can achieve higher efficiency, i.e., higher voltage at comparable current densities than other fuel cell types. It has very long operating life time, e.g., 15,000 h # Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_17-4

have been demonstrated (Cifrain and Kordesch 2003). Furthermore, alkaline conditions allow application of nonprecious metal catalysts which could reduce significantly the material costs of this fuel cell. One of major disadvantages of AFC is its low CO2 tolerance due to formation of carbonates according to: CO2 þ 2OH ⇄ CO3 2 þ H2 O The carbonates have low solubility in strong alkaline environments forming crystals, capable of blocking of electrolyte pathways. The low CO2 tolerance and the application of liquid electrolyte are two main hurdles for the broader commercialization of these systems. However, due to its high efficiency and high power density, AFC found applications in aerospace industry, e.g., they were employed on the Apollo missions as well as on the Space Shuttle orbitals. For other examples of developed AFC systems, please see Cifrain and Kordesch (2003) and G€ulzow (2012). Hydrogen and oxygen gases in the AFC are separated by a membrane. Usually permeable membranes also called diaphragms have been used (G€ulzow 2012). A common diaphragm material up to 1980s was asbestos which due to health and environmental concerns is nowadays abandoned. Alternative materials are different polymer materials like porous polyethylene plates, nonwoven polypropylene, and similar. Potassium hydroxide electrolyte wets the

2

diaphragm pores in order to ensure its ionic conductivity. For a good fuel cell performance, the electrolyte resistance induced by diaphragm has to be minimized. This resistance is influenced by the thickness of the diaphragm material, its pore size and tortuosity, and the KOH concentration. Since liquid electrolyte causes some practical problems in AFC usage, new developments go into direction of alkaline polymer electrolyte membranes which, in addition to separation, possesses ionic conductivity in the absence of liquid electrolyte. Currently there is no long-term stable membrane functioning without a liquid electrolyte phase.

Alkaline Fuel Cells (AFCs)

References Cifrain M, Kordesch K (2003) Hydrogen/oxygen (Air) fuel cells with alkaline electrolytes. In: Vielstich W, Gasteiger HA, Lamm A (eds) Handbook of fuel cellsfundamentals, technology and applications, vol 1, Fundamentals and survey of systems. Wiley, Chichester, pp 267–280 G€ ulzow E (2012) Alkaline fuel cells. In: Stolten D, Emonts B (eds) Fuel cell science and engineering: materials, processes, systems and technology, vol 1. Wiley-VCH, Weinheim, pp 97–129

A

Alumina Membranes Bo Wang Department of Chemical Engineering, Imperial College London, London, UK

Alumina membranes are membranes made of high-purity (mostly >97 %) aluminum oxides (alumina). They belong to the categories of ceramic membranes and in most cases, microfiltration and ultrafiltration membranes. Pure alumina membranes are predominant in all kinds of commercial ceramic membranes, and most other commercial ceramic membranes, if using other materials as the separation coating, are based on alumina membrane support.

Advantages and Disadvantages Alumina membranes are chemically inert and mechanically strong, and also thermally stable, therefore they can be used under very harsh conditions such as aggressive chemical environments and elevated temperatures, making them versatile in various industrial processes. Their strong mechanical properties allow them to be operated under high pressures and high feed flow rates to achieve high productivity and to tolerate high solid contents in the feed.

# Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_18-1

The high fabrication cost is a major obstacle for alumina membranes to expand its market share. Although alumina is widely available, the complicated fabrication process and high energy consumption for sintering lift significantly the total membrane cost, making it about one order of magnitude higher than its polymeric counterparts. However, because of much longer lifetimes and higher fluxes of alumina membranes, the longterm operating cost of alumina membrane plants is comparable with polymeric membrane plants, no mention there are many scenarios in which only alumina/ceramic membranes can be used.

Membrane Designs and Fabrication Most alumina membranes use asymmetric membrane architectures. Coarse alumina particles are used for the supporting layer to provide mechanical strength, and the transport resistance in this layer is usually low due to its big pore size; fine alumina particles are coated on top of the support to form a thin separating layer to reach desired pore sizes and high fluxes; and depending on the pore size difference between the top and the supporting layer, one or few transition layers might be used between the top and supporting layers to avoid coating defects and to increase adhesion between layers.

2

Alumina membranes may adopt a tubular shape, plate shape, or monolithic multichannel design. The latest is the most commonly used design in industry because it provides higher membrane area per unit volume, therefore reduces the size of the separation plant and operating costs. Some commercial products can achieve an area to volume ratio of 800 m2/m3 with the monolithic design, which is comparable with spiral wound polymeric membrane modules. Fabrication of alumina membranes usually involves several forming and sintering steps due to their multilayer asymmetric membrane architectures, and the total production process is often longer than 1 week. The supporting layer can be formed by extrusion (for tubular or monolith designs) or by pressing (for plate or disk designs), followed by partial sintering at a high temperature (up to 1600 C) to achieve a high strength and porous structures. Transition layers and the top layer then can be coated on the support, also followed by sintering after each coating. Depending on the pore size, two types of alumina could be used for the top layer coating: for microfiltration membranes whose pore size is larger than 100 nm, a-alumina is used; for ultrafiltration membranes whose pore size is smaller than 100 nm, g-alumina or a mixture of a and g types are often used. There are also other fabrication techniques that can produce unique alumina membranes, for example, the anodic alumina membranes that have straight cylindrical pores and asymmetric alumina membranes made by single-step phase inversion/sintering methods. These membranes have shown good potential in filtration but have not been widely accepted as a feasible replacement of conventional alumina membranes yet.

Alumina Membranes

Applications Alumina membranes are widely used in water and wastewater treatment, pharmaceuticals, food processing, chemical processing, and others. Production of drinking water and wastewater treatment are the most important applications, which occupy about half of the market share of alumina membranes. The stable pore structure of alumina membranes even under high pressures and abrasive conditions guarantees constant quality of the permeate and makes it preferable in the scenarios where quality control is crucial, such as in drinking water production and food industry. And the natural hydrophilic surface of alumina offers better anti-fouling property than polymeric membranes; therefore its usage in wastewater treatment is expanding rapidly; and in some other heavy-duty filtration processes such as treating produced water from oil fields, alumina membranes have been dominant.

Further Reading Benfer S, Árki P, Tomandl G (2004) Ceramic membranes for filtration applications – preparation and characterization. Adv Eng Mater 6(7):495–500 Lehman SG, Adham S, Liu L (2008) Performance of new generation ceramic membranes using hybrid coagulation pretreatment. J Environ Eng Manage 18(4):257–60 Ebrahimi M, Willershausen D, Ashaghi KS, Engel L, Placido L, Mund P, Bolduan P, Czermak P (2010) Investigations on the use of different ceramic membranes for efficient oil-field produced water treatment. Desalination 250(3):991–996

A

Anion-Exchange Membrane (AEM) Mitsuru Higa Graduate School of Science and Engineering, Yamaguchi University, Ube, Yamaguchi, Japan

The anion-exchange membrane (AEM) is a thin film with anion-exchange groups (positively charged groups) and permeates anions selectively. AEMs are classified according to the species of the ion-exchange groups and materials constituting the membrane and microstructure. Anion-exchange groups are positively charged groups: primary, secondary, and tertiary amino groups, quaternary ammonium groups, tertiary sulfonium groups, quaternary phosphonium groups, cobalticinum groups, and other groups that provide a positive fixed charge in aqueous or mixed water and organic solvent solutions. Based on the materials constituting membranes, AEMs can be classified as (i) organic membranes, (ii) inorganic membranes, and (iii) composite membranes of inorganic ion exchangers and organic polymers. AEMs are also classified into two types by their microstructure: heterogeneous and homogeneous. Heterogeneous AEMs consist of finely powdered anion exchanger and an inert binder polymer. Various methods have been reported for preparing homogeneous AEMs. A typical example of a hydrocarbon AEMs is a copolymer membrane composed of styrene and divinylbenzene with benzyltrimethylammonium # Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_23-1

groups. These membranes will be prepared, for example, by the reaction of trimethylamine with a copolymer membrane prepared from chloromethylstyrene and divinylbenzene or by alkylation with alkyl halide of a copolymer membrane prepared from vinylpyridine and divinylbenzene. Though AEMs have been used in many fields, most are used in electrodialysis, separation of electrolysis, and solid polymer electrolytes for fuel cells. The properties required depend on the intended application of AEMs. Generally required properties are (1) low electrical resistance, (2) high transport number of anions, (3) low diffusion coefficient of electrolytes, (4) low osmotic water and low electroosmotic water, (5) antifouling properties, (6) mechanical strength, (7) dimensional stability, (8) high chemical stability and durability, and (9) low cost (Sata 2004). AEMs with low sulfate ion permeability have been industrially used to prevent precipitation of calcium sulfate in the AEMs and electrodialyzer. A nitrate ion permselective AEM has been developed and has contributed to human health because the concentrations of nitrate ions are greatly increased in groundwater. To prepare these functionalized AEMs, ordinary AEMs are modified by suitable chemical or physical methods. For example, most commercially available homogeneous AEMs are mainly crosslinked with divinylbenzene. When the contents of divinylbenzene increase or compact layers are

2

formed on the AEMs, the pore size of the membrane decreases and the transport number of sulfate ions, which are bulky relative to chloride ions, decreases. Fouling of separation membranes is a common problem. AEMs are fouled by ionic materials of medium molecular weight such as ionic surfactants having the charge opposite to the fixed charged of the membrane. Almost all of organic foulants in many effluent streams have negative charges; hence, fouling of AEMs due to deposition and/or adsorption of the foulants on/in the AEM is one of the serious problems in their applications. The pore size of the AEM is generally recognized to be about 1 nm; therefore, ions of medium molecular weight permeate with difficulty through the membrane. Consequently, the fouling can lead to an unacceptably high stack resistance and replacement of membranes in an electrodialysis system due to clogging of the

Anion-Exchange Membrane (AEM)

membrane pores with the medium molecular weight ions. To alleviate the problem of organic fouling, there are basically two methods on the membrane side: first, to increase the pore size of the membrane to allow easy permeation of large ionic materials and, second, to prevent penetration of the materials into the membrane at the membrane surfaces. There are two methods to prevent penetration of large organic materials into the membrane matrix: forming a thin charged layer opposite in sign to the ion-exchange groups of the membrane or forming a very thin and dense layer on the membrane surfaces.

Reference Sata T (2004) Ion exchange membrane. The Royal Society of Chemistry, Cambridge

A

Antoine Equation

Phenomenon

Denis Roizard Laboratoire Réactions et Génie des Procédés, CNRS- Université de Lorraine, Nancy, France

Vapor Pressure-Temperature Relationship

The vapor pressure over a liquid is due to the thermodynamic equilibrium between the gas and the liquid states of the component, which depends on the cohesive forces linking the molecules. In a closed cell, the vapor pressure of a pure component is a nonlinear relation of the temperature: the more the component is volatile, the more the vapor pressure is high.

History

Antoine Equation

Until the mid-nineteenth century, the prediction of the saturated vapor pressure of a liquid mixture or even of a pure liquid in relation with the temperature was not accurate despite the phenomena being studied long before the Middle ages. The Antoine equation, which solves this problem for pure components, is due to a French engineer (Louis Charles Antoine, 1825–1897) and was first published in “Annales de Physique et de Chimie” in 1891 (Antoine 1891). Antoine introduced an equation able to predict the vapor pressure of pure liquids (vaporization) and solids (sublimation). It is worth noting that this equation is still widely used today because of its accuracy. Wisniak (2001) has recently reviewed the historical development of the vapor pressure equations from Dalton to Antoine.

For a given pure component in a closed cell, it calculates the saturated vapor pressure P (mmHg) with the temperature T ( C) as follows:

Synonyms

# Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_26-1

log10 P ¼ A B=ðT þ CÞ

(1)

with A, B, and C being the Antoine coefficients, in mmHg and C units, which are componentspecific constants; e.g., for water : A = 8.07131; B = 1730.63 and C = 233.426 in the temperature range 1–99 C. Other forms of the Antoine equation can be deduced: P ¼ 10ðAB=ðTþCÞÞ

(2)

2

Antoine Equation

Antoine Equation, Fig. 1 Curves of P(sat) H2O versus temperature; range 50–150 C calculated with the Antoine equation

180000 160000 600

P mmHg

140000 120000

400

100000

200

ZOOM 0–100°C

00

80000

100

60000 40000 20000 0 0

100

200

300

400

T °C

T ¼ B=ðA log10 PÞ C

(3)

These equations fit very well for experimental vapor pressure data. It is worth noting that the Antoine equation can also be used at high pressure or near the critical point. Indeed, for each component, two sets of parameters can be used according to the considered temperature range, i.e., either up to the normal boiling point or from normal boiling point to the critical point; in this last range, the Clausius-Clapeyron relation (Reid 1990) does not apply properly. In the case of pure water, in the temperature range 99–374 C, the dedicated Antoine coefficients are A = 8.140191, B = 1810.94, and C = 244.485. The extension (Wagner 1973) of the Antoine equation has been worked out to cover with a single set of parameters the whole temperature range going from the critical point to the triple point. This expression is required when computational techniques must be used. On the other hand, a simplified equation using only two parameters was known to evaluate the vapor pressure; it was the August equation in which the parameter C is set to zero, thus

assuming a temperature-independent heat of vaporization.

Antoine Coefficients Two systems of units can be used, either based on temperature and pressure, respectively, in C and in mmHg or on K and Pa (SI system). Note that for historic reasons, the Antoine coefficients are still normally given based on the CGS system. Conversion from the historical system unit to the SI one can be made easily: – B coefficient is the same in both systems; – To get A coefficient in SI, add 2.124903 to the historical A; this value corresponds the pressure unit modification, i.e., to log10 (101325/ 760). – For C coefficient, subtract 273.15 to take into account the modification of the temperature unit. The Antoine coefficients have been tabulated for most of pure compounds. They can be obtained from various sources, including web

Antoine Equation

3

Antoine Equation, Table 1 Example calculation of P(sat) H2O; range 50–150 C T C 50 60 70 80 90 100 A = 8.07131

log10 (P) 1.9652012 2.1732982 2.3676788 2.5496558 2.7203796 2.8808629 B = 1730.63

P mmHg 92.30 149.04 233.17 354.53 525.27 760.09 C = 233.426

T C 100 110 120 130 140 150 A = 8.140191

log10 (P) 2.8832575 3.0320002 3.1745848 3.3093227 3.4368443 3.5577142 B = 1810.94

P mmHg 764.29 1076.47 1494.81 2038.56 2734.29 3611.72 C = 244.485

access Data banks of Antoine Coefficients (visited June 2013).

▶ Clausius Clapeyron Equation ▶ Vapor Pressure

Example Calculation

References

This calculation has been done for water with the coefficients: A = 8.07131, B = 1730.63, and C = 233.426. The temperature validity range is from 1 C to100 C.

Antoine C (1891) Annales de Physique et de Chimie 22: 281; ibid, Annales de Physique et de Chimie (1892) 26: 426; Comptes Rendus Acad Sci (Paris) (1888) 107: 1143 Data banks of Antoine Coefficients (visited June 2013) http://webbook.nist.gov/ – http://booksite.elsevier. com/9780080966595/content/Appendices/Appendix %20C.pdf (free download) – http://www.eqi.ethz.ch/ fmi/xsl/eqi/eqi_property_details_en.xsl?node_id= 983 (directory for Physical Properties Sources) – Vapor-Liquid Equilibrium Data Collection, DECHEMA Chemistry Data Series, Jurgen Gmehling et al. (Frankfurt) Reid CE (1990) Chemical thermodynamics. McGrawHill, New York, p 73 Wagner W (1973) New vapour pressure measurements for argon and nitrogen and an new method for establishing rational vapour pressure equations. Cryogenics 13(8):470–482 Wisniak J (2001) Historical development of the vapor pressure equation from Dalton to Antoine. J Phase Equilib 22–6:622–630. doi:10.1007/s11669-0010026-x1143

– At T = 50 C: log10 (P) = 8.07131–1730.63/ (50 + 233.426) = 1.9652 – Hence, P(sat) H2O = 92 mmHg The results shown in Fig. 1 and Table 1 correspond to water vapor pressure calculated with the respective sets of parameters of each domain with the Eq. 1. The discontinuity at 100 C can be seen in the Table 1.

Cross-References ▶ August Equation

A

Aptamers Thomas Scha¨fer Institute for Polymer Materials (Polymat), University of the Basque Country (UPV/EHU), Donostia-San Sebastián, Spain

Aptamers (from Latin aptus, to fit) are artificial ligands binding with high affinity and specificity to their cognate target, selected through a Darwinian-like evolution method referred to as SELEX. Aptamers generally consist of structured single-strand nucleic acid molecules such as RNA and ssDNA; however, dsDNA and peptide aptamers have been described as well (Colas et al. 1996; Patel et al. 1997). Aptamers can be selected against virtually any target and under nonphysiological conditions because no animal host is required. Hitherto, aptamers were selected against a broad range of target molecules, such as amino acids, peptides, proteins, drugs, organic and inorganic molecules, or even whole cells, with affinities often comparable to those of monoclonal antibodies. The binding proprieties of aptamers are due to the formation of specific aptamer-target complexes stabilized by non-covalent interactions. The latter are a combination of van der Waals forces, hydrogen bonds, and electrostatic interactions. Binding of an aptamer to its cognate target can trigger an adaptive folding in which the target promotes and stabilizes the secondary and # Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_27-3

tertiary structures of aptamers. The dynamic conformations of free aptamers, consisting of labile knots, loops, and stems, turn into stable architectures within the aptamer-target complexes where noncanonical base pairing along with intermolecular interactions creates unique binding pockets. Hence, along with a remarkable affinity, aptamers are generally characterized by high specificity. One of the most noteworthy examples is the anti-theophylline aptamer which can discriminate between theophylline and caffeine, two related molecules that only differ by a methyl group of the imidazole ring, having about 11,000 times higher affinity for the former. Similarly, aptamers have been reported to discriminate between enantiomers, macromolecules, proteins, and whole cells. Apart from their high affinity and specificity, aptamers possess unique chemical and biochemical characteristics which clearly set them apart from other receptors. For example, aptamers are exceptionally stable: they may undergo denaturation, but the process is fully reversible within minutes; hence, temperature changes or longterm storage does not affect their functionality. Furthermore, the well-known chemistry of aptamers allows for site-directed chemical modifications and competitive production costs. New functional groups can be introduced a priori – within the SELEX process – using modified nucleic acid libraries or a posteriori modifying a selected DNA/RNA through chemical synthesis. While the first approach relies on the

2

compatibility of the modification adopted with the enzymatic amplification necessary for selection, the latter approach is streamlined by detailed knowledge of the structure of both the aptamer and the aptamer-target complex. Since their first appearance in the early 1990s, aptamers have received constantly increasing attention, reflected in their use as diagnostic reagents and therapeutic compounds.

Aptamers

References Colas P, Cohen B, Jessen T, Grishina I, McCoy J, Brent R (1996) Genetic selection of peptide aptamers that recognize and inhibit cyclin-dependent kinase 2. Nature 380(6574):548–550 Patel DJ, Suri AK, Jiang F, Jiang L, Fan P, Kumar RA, Nonin S (1997) Structure, recognition and adaptive binding in RNA aptamer complexes. J Mol Biol 272(5):645–664

A

Aptamer Membrane Functionalization Thomas Schäfer Institute for Polymer Materials (Polymat), University of the Basque Country (UPV/EHU), Donostia-San Sebastián, Spain

Aptamer Membrane Functionalization. Aptamers can be incorporated into adequate porous membrane structures in order to obtain stimulusresponsive membranes whose permeability is modulated via a molecular recognition event. The concept relies on the fact that aptamers can recognize very specifically a molecular target, upon which a significant conformational change can occur if the aptamer is designed accordingly. In this sense, aptamer-modified membranes follow the concept “structure determines separation.” The stimulus-responsive membranes so obtained therefore do not require a bulk stimulus, such as a change in temperature or pH, which makes them particularly interesting for being employed in biomedical separations or DNA-based nanodevices (Bhattacharyya et al. 2013). Incorporating the aptamer into an adequate porous structure, the aptamer conformational change upon target recognition can give rise to a hindered pore flow, modulating in this way the overall membrane permeability. Aptamertarget interactions can be highly specific, target concentration dependent, and do not involve the # Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_28-1

formation of chemical bonds but are physical. As a consequence, membranes functionalized with aptamers change their permeability depending on the target concentration and in a reversible manner. Figure 1 depicts the concept of a membrane pore functionalized with a DNA-aptamer hairpin structure which changes its conformation upon a molecular stimulus such as adenosine 50 -monophosphate (AMP). In the absence of the target AMP, the aptamer-functionalized membrane pores remain open and the membrane permeability maximum. Upon interaction with AMP, the aptamer undergoes a conformational change which significantly reduces the pore flow and, hence, overall membrane permeability. For the conformational change to take effect, the pore diameter needs to be of the same order of magnitude. Therefore, mesoporous structures are preferably employed that furthermore possess a high degree of isoporosity. Aptamer-modulated pore flow has been thoroughly studied for mesoporous particles that are used as controlled delivery devices (Özalp and Schäfer 2011). Figure 2 shows the immobilization strategy for modifying mesoporous silica particles with aptamers and Fig. 3 the concept of how such an aptamer can serve as a reversibly opening “lid” in order to liberate cargo molecules upon target recognition (here: ATP). A key parameter for the functioning of aptamer-functionalized membranes is the finetuning of the pore size within which the aptamer conformational change takes place. Pores too

2

Aptamer Membrane Functionalization

Aptamer Membrane Functionalization, Fig. 1 Scheme of a stimulus-responsive aptamer-functionalized membrane

c

a

3. Amino functionalized Aptamer

HN

10 nm

100 nm

O

b 120

Intensity

100

2. GMBS crosslinker O

80

N O

60 40

S

1. Thiol Addition via silanization

20 0 150

160

170

180

190

200

210

Average Size (nm)

220

O

Si

O

O

Silica Surface

Aptamer Membrane Functionalization, Fig. 2 Mesoporous particles (a), their size distribution (b), and the functionalization of mesoporous silica with an aptamer (c)

large would not result in any aptamer-modulated permeability, while pores too narrow would hinder the aptamer from freely changing its conformation. It could be shown that the thickness change of a DNA-aptamer hairpin film amounts up to about 2 nm upon interaction with the target

(Serrano-Santos et al. 2012). Hence, mesoporous structures with a pore diameter between 2 and 3 nm have been found to be a suitable base material for aptamer-functionalized particles and membranes (Özalp and Schäfer 2011).

Aptamer Membrane Functionalization

3

Aptamer Membrane Functionalization, Fig. 3 Sequence of a DNA-aptamer hairpin that binds selectively to ATP (a), scheme of its conformational

change upon target recognition (b), and function to release in a reversible, specific, and concentration-dependent fashion cargo molecules (c)

References

Özalp VC, Schäfer T (2011) Aptamer-based switchable nanovalves for stimuli-responsive drug delivery. Chem Eur J 11:9893–9896 Serrano-Santos MB, Llobet E, Özalp VC, Schäfer T (2012) Characterization of structural changes in aptamer films for controlled release nanodevices. Chem Commun 48:10087–10089

Bhattacharyya D, Schäfer T, Wickramasinghe RR, Daunert S (eds) (2013) Responsive membranes and materials. Wiley, Hoboken

A

Aptamer Screening Thomas Scha¨fer Institute for Polymer Materials (Polymat), University of the Basque Country (UPV/EHU), Donostia-San Sebastián, Spain

Aptamer screening or selection, also called systematic evolution of ligands by exponential enrichment (SELEX), is a combinatorial chemistry method by which aptamers (single-stranded DNA or RNA) are selected. SELEX is an evolutionary method driven by the binding of oligonucleotide sequences to a specific target based on their particular tridimentional structure, which confers high affinity and specificity. The selection process starts with a random oligonucleotide library consisting of a random region flanked by fixed primer regions required for PCR amplification. The random region confers outstanding variability, derived from the number of possible sequences in the library. The variability of the library can be calculated with the expression 4n, where 4 is the number of possibilities at each position (A, T, C, or G) and n is the length of the random region. SELEX allows the screening of oligonucleotide libraries against a given target (i.e., protein, small

# Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_29-3

molecules, or even whole cells) by an iterative process that involves (i) library/target interaction, (ii) selection of the sequences that bind to the target and removal of no binders, (iii) enrichment of the selected sequences by PCR, and (iv) after several screening rounds, cloning and sequencing to identify individual sequences. Finally, bioinformatic analysis is required to determine the most promising motifs or sequences. Next, the candidate sequences are evaluated to confirm their binding capability. SELEX technology could be compared to antibody production; however, SELEX offers interesting features that overcome some limitations of the antibody production, such as fast in vitro performance, no need for cells or animals, ease of selection under physiological or nonphysiological conditions, no limitation for target selection, and batch-to-batch reproducibility. Since the first report of SELEX in 1992 by two independent groups (Tuerk and Gold 1990; Ellington and Szostak1990), several modifications of the standard method have been reported to improve its performance. Nowadays, a significant number of aptamers have been selected and used as recognition molecules for environmental, food, diagnosis, and therapeutic applications (Fig. 1).

2 Aptamer Screening, Fig. 1

Aptamer Screening

Library Design by regions fixed

random

fixed

Cloning, Sequencing and Bioinformatics

Library Target

iteration

Interaction Library/Target

Selection of Target Binders Removal of no binders

Enrichment of selected binders

References Ellington AD, Szostak W (1990) In vitro selection of RNA molecules that bind specific ligands. Nature 346:818–822

Tuerk C, Gold L (1990) Systematic evolution of ligands by exponential enrichment: RNA ligands to bacteriophage T4 DNA polymerase. Science 249:505–510

A

Aquaporins (AQPs) or Water Channels Fabio Bazzarelli and Lidietta Giorno Institute on Membrane Technology, ITM, National Research Council of Italy, Rende (CS), Calabria, Italy

Aquaporins (AQPs) or water channels are a family of integral membrane proteins that form hydrophilic pores in the cellular membrane. They are involved in the water transport through the membrane. All cells depend on their ability to maintain water homeostasis. This is achieved through the action of aquaporins, membrane-bound water channels that facilitate water flow across cellular membrane along osmotic gradients, while excluding the passage of ions and protons. It is required for maintenance of the membrane potential and intracellular pH. As with any membrane transport facilitator, aquaporins have evolved to be highly selective for their transported substrate without binding water so strongly that transport is inhibited. On the basis of their selectivity, aquaporins can be divided into two groups: the ordinary aquaporins, permeable to water only, and aquaglyceroporins which also permit transport of small solutes such as glycerol and urea. A number of other compounds have also been reported to be transported through aquaporins, including CO2, NH3, and arsenite (Kreida and # Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_30-1

Tornroth-Horsefield 2015). AQPs are transmembrane channels; thus the ability of a molecule to cross an AQP channel depends on its own characteristics (size, polarity, charge) and on the features of the AQP involved (Di Giorgio et al. 2014). They have a similar basic structure; AQPs are monomers of about 30 kDa and, in general, contain six membrane-spanning helical segments and two shorter helical segments that do not span the entire membrane. The AQPs generally form stable tetramers in membranes, although each monomer contains a separate water pore. High-resolution structural data show that the membrane-spanning helical domains surround cytoplasmic and extracellular vestibules that are connected by a narrow aqueous pore. Structural data and molecular dynamics simulations suggest that water molecules move through this narrow aqueous pore and that steric and electrostatic factors are responsible for the water selectivity of AQPs. The pore is less constricted in the aquaglyceroporins than in the water˚ versus 2.8 A ˚, selective AQPs (diameter of 3.4 A respectively) and is lined by more hydrophobic residues (Papadopoulos and Verkman 2013). An important property of aquaporin-mediated water transport is its ability to be regulated in response to cellular or environmental signals. This is achieved by controlling water transport at the individual protein level through a conformational change, so-called gating, or by altering the aquaporin density of a particular membrane. These proteins are present in all kingdoms of

2

life, demonstrating their central role in maintaining normal physiology of all organisms. The first member of this family, AQP1, was identified in erythrocytes in 1991. This discovery led to homology cloning of hundreds of AQPs homologues from throughout the animal and plant kingdoms, as well as from lower organisms. In humans there have been identified 13 aquaporins (AQP0–12) with specific organ, tissue, and cellular localization. Thus, different members of the AQPs family are expected to function in virtually all physiological processes that involve water transport across the membrane. The AQPs are expressed in many cell types involved in fluid transport, including epithelia and endothelia in the kidney, lung, exocrine glands, eyes, and gastrointestinal tract. However, aquaporins are also expressed in cells that do not have an obvious role in fluid transport, such as erythrocytes and some leukocytes, adipocytes, and muscle. In addition, these are also expressed in astrocytes throughout the central nervous system and in supportive cells. Aquaporins have been linked to a number of pathological conditions, including brain edema, renal disease, obesity, and cancer, raising their attractiveness as drug targets. Given the key role played by aquaporins in the kidney, for recovering water permeated together

Aquaporins (AQPs) or Water Channels

with other ions and molecules through the first part of the glomerulus, they have been investigated for developing biohybrid membranes able to desalinate seawater. The biohybrid membranes containing aquaporins showed very high water selectivity and permeability. Technological challenges for productive application include biohybrid membrane preparation on a large scale, aquaporin stability under operating conditions or during membrane module cleaning, and maintenance operation. Development of synthetic water channels mimicking aquaporins is a new strategy under investigation. Highly selective and permeable water channels are very interesting for the development of seawater desalination plant operating with very low energy input.

References Di Giorgio J, Soto G, Alleva K, Jozefkowicz C, Amodeo G, Muschietti JP, Ayub ND (2014) Prediction of aquaporin function by integrating evolutionary and functional analyses. J Membr Biol 247:107–125 Kreida S, Tornroth-Horsefield S (2015) Structural insights into aquaporin selectivity and regulation. Curr Opin Struct Biol 33:126–134 Papadopoulos MC, Verkman AS (2013) Aquaporin water channels in the nervous system. Nat Rev Neurosci 14:265–277

A

Atomic Force Microscopy (AFM) Nidal Hilal and Daniel Johnson Centre for Water Advanced Technologies and Environmental Research (CWATER), College of Engineering, Swansea University, Swansea, Wales, UK

The atomic force microscope is a versatile tool increasingly used for the physical characterization of surfaces and is of great interest for the visualization and analysis of process surfaces including those of membranes used for filtration. It is capable of resolving features from the micrometer down to the subnanometer scale and can operate in air and liquid environments, allowing membranes to be studied in environments matching those encountered during their operation (Hilal et al. 2004; Hilal and Johnson 2010), which allows assessment of effects of, for instance, pH, ionic strength, and effects of additives on membrane structure, a feature not available with other high resolution imaging applications. In addition, the surface needs no special preparation, providing it remains clear of unwanted contamination, and does not need to be electrically conductive, limitations found with some imaging techniques.

# Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_34-1

The AFM consists of a sharpened probe mounted at the end of a flexible cantilever. The tip of the probe is then used to “feel” the underlying membrane surface, producing a threedimensional map of the sample topography. There are three basic imaging modes: contact mode, where the tip maintains constant contact with the surface, while feedback loops adjust its height to maintain cantilever deflection and hence force; tapping mode, where the cantilever is vibrated close to the resonant frequency and intermittently contacts the surface with decreased lateral forces compared with contact mode; and noncontact mode in which the probe tip interacts with attractive forces very close to the surface but is kept away from hard repulsive interactions. Noncontact mode is capable of extremely high resolution but is the most difficult to attain in practice. From the imaging data, several quantitative parameters of interest to membrane technologists can be obtained including: surface roughness, pore size, pore-size distribution, as well as showing in fine detail the morphology of fouled and unfouled membrane surfaces and the effects of chemical modification of membrane surfaces (Johnson et al. 2012; Kochkodan et al. 2013). Figure 1 shows a typical image of a cyclopore microfiltration membrane and a resulting pore-size distribution (Bowen et al. 1996; Bowen and Hilal 2009). There are a

2 Atomic Force Microscopy (AFM), Fig. 1 3D image of cyclopore microfiltration membrane and pore-size distribution for same membrane obtained by AFM

Atomic Force Microscopy (AFM)

mm 0.20 0.10 0.00 3

2 mm 1

3 2 1

mm

25

% of pores

20

15

10

5

0 0.10

large number of expansions for AFM instruments allowing many other modes allowing mapping of electrical and electrochemical, mechanical, adhesive, frictional and magnetic properties of surfaces of interest. As well as providing this information, the AFM probe can be used as a force sensor by moving the probe into and out of contact with the sample surface and monitoring deflection in the microcantilever arm. Through careful calibration of the system this deflection can be converted into force, directly measuring the interaction force as a function of probe tip – sample separation distance (Gibson et al. 2004). This allows the detection of DLVO, hydrophobic, hydrostatic and steric interactions, membrane stiffness, and adhesion forces. The sharp tip may be replaced by a colloidal particle, which may be functionalized

0.15

0.20 0.25 Pore size, µm

0.30

0.35

to simulate any of a number of potential membrane foulants allowing quantitative measurements of foulant rejection and attachment forces. This makes possible the direct quantification of membrane fouling properties of different materials under a range of environmental conditions and with only a relatively small sample of membrane needed.

References Bowen WR, Hilal N (eds) (2009) Atomic force microscopy in process engineering. Butterworth-Heineman, Oxford Bowen WR, Hilal N et al (1996) Atomic force microscope studies of membranes: surface pore structures of cyclopore and anopore membranes. J Membr Sci 110(2):233–238

Atomic Force Microscopy (AFM) Gibson CT, Johnson DJ et al (2004) Method to determine the spring constant of atomic force microscope cantilevers. Rev Sci Instrum 75(2):565–567 Hilal N, Johnson DJ (2010) The use of atomic force microscopy in membrane characterisation. In: Drioli E, Giorno L (eds) Comprehensive membrane science and engineering. Elsevier Science, Oxford, pp 521–538 Hilal N, Al-Zoubi H et al (2004) A comprehensive review of nanofiltration membranes: treatment, pretreatment,

3 modelling and atomic force microscopy. Desalination 170:281–308 Johnson DJ, Al-Malek SA et al (2012) Atomic force microscopy of nanofiltration membranes: effect of imaging mode and environment. J Membr Sci 389:486–498 Kochkodan V, Johnson DJ et al (2013) Polymeric membranes: surface modification for minimizing (bio)colloidal fouling. Adv Colloid Interface Sci (in press). doi:10.1016/j.cis.2013.05.005

A

Atomistic Simulations Methods Elena Tocci Institute on Membrane Technology ITM-CNR, University of Calabria, Rande (CS), Italy

Atomistic Simulations Atomistic simulations are theoretical and computational modeling tools for interpreting what happens at the atomic scale in solids, liquids, molecules and plasmas. Atomistic simulations, such as phonon calculations, free-energy optimizations (molecular mechanics), molecular dynamics (MD), Monte Carlo simulations, and crystal structure prediction, are used to interpret existing experimental data and predict new phenomena and to provide a way forward where experiments are not yet possible, e.g., under extreme conditions or at atomistic size- and timescales which are difficult to detect directly (Allen and Tildesley 1989; Haile 1992; Frenkel and Smit 2002; Leach 2001; Brenner 2000; Allen 2004; Tocci and Pullumbi 2011). Molecular modeling is primarily a tool for calculating the energy of a given molecular structure, and the goal is to understand and model the motion of each atom in the material. Different levels of atomistic simulations exist, ranging from quantum mechanical models to statistical methods. This means solving numerically the classical or quantum mechanical microscopic # Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_36-1

equations for the motion of interacting atoms or even deeper – electrons and nuclei. Quantum mechanical (QM) or ab initio methods describe matter at the electronic level, considering the fundamental particles, electrons and protons. The equation from which molecular properties can be derived is the Schrodinger equation, and various approximations must be introduced in order to extend the utility of the method to polyatomic systems. Atomistic methods are used to compute molecular properties, which do not depend on electronic effects; the whole atom is modeled just as a soft sphere and obeys the laws of statistical mechanics. Atomistic simulation utilizes analytic potential energy expressions (sometimes referred to as empirical or classical potentials) to describe the systems. The analytic potential energy functions are simplified mathematical expressions that attempt to model interatomic forces arising from the quantum mechanical interaction of electrons and nuclei. Their use is dictated by the need to model systems with sizes and/or timescales that exceed available computing resources, required for quantum calculations, which give no account of the complex electronic structure of atoms. Forces between atoms are derived from empirical interatomic potentials that are obtained from fitting material properties (e.g., lattice constant, elastic constants, vacancy formation energy, etc.) from experimental data or QM calculations. They may depend on the distance between atoms,

2

Atomistic Simulations Methods

angles between bonds, angles between planes, etc. The general form of the total potential of the N-atom system describes types of interactions, bonded and non-bonded, and can be written as:

X ! ! ! ! V r 1 , r 2 , . .. , r N ¼ V 1 r i X X ! ! ! ! ! þ V2 r i, r j þ V3 r i, r j, r k þ . . . i, j⊳i i, j⊳i, k⊳j

The part of the potential energy V representing bonding interactions will include terms of the following kind: 2 1X r K r ij r eq 2 bonds ij1 2 1 X y þ K ijk yijk yeq 2

! ! ! V nonbonded r 1 , r 2 , . . . , r N X ! XX ! ! ¼ v ri þ w r i, r j þ . . . i

!

i

The v r i term represents an externally applied potential field and describes external force fields (e.g., gravitational field) and external constraining fields (e.g., the “wall function” for particles in a chamber). The pair potential ! ! w r i , r j ¼ w r ij neglect three-body (and higher order) interactions. The Lennard-Jones potential is the most commonly used form:

V intramolecular ¼

bend

angles

þ

1 X X f, m K ijkl 1 þ cos mfijkl gm 2 m torsion

j⊳i

wLJ ðr Þ ¼ 4e

s 12 s6 r r

where s is the diameter and e is the depth of the potential energy well. If electrostatic charges are present, the appropriate Coulomb potentials are added:

angles

The “bonds” typically involve the separation ! ! r ij ¼ r i r j between adjacent pairs of atoms in a molecular framework, and a harmonic form with specified equilibrium separation has been used, although this is not the only possible type. The “bend angles” yijk are between successive ! ! ! ! bond vectors such as r i r j and r j r k and involve three atom coordinates. Usually this bending term is quadratic in the angular displacement from the equilibrium value, although periodic functions are also used. The “torsion angles” fijkl are defined in terms of three connected bonds; hence four atomic coordinates are used. The part of the potential energy V representing non-bonded interactions between atoms is traditionally split into one-body, two-body, threebody terms:

wCoulomb ðr Þ ¼

Q1 Q2 4pe0 r

where Q1, Q2 are the charges and e0 the permittivity of the free space. To be effective, an analytic potential energy function must possess the following critical properties: Flexibility: A potential energy function must be sufficiently flexible that it accommodates as wide a range as possible of fitting data. Accuracy: A potential should be able to accurately reproduce properties of interest as closely as possible. Transferability: A potential function should be able to study a variety of properties for which it was not fit. Computational efficiency: Evaluation of the function should be relatively efficient depending on quantities such as system sizes and timescales of interest, as well as available computing resources.

Atomistic Simulations Methods

The major methods are molecular mechanics (MM), molecular dynamics (MD), Monte Carlo (MC), and additionally, there is a whole range of hybrid techniques which combine features from both MD and MC methods.

References Allen MP (2004) Introduction to molecular dynamics simulation computational soft matter: from synthetic polymers to proteins. In: Attig N, Binder K, Grubmuller H, Kremer K (eds) Lecture notes, John von Neumann Institute for computing, Julich, NIC series, vol 23, ISBN 3-00-012641-4, pp 1–28

3 Allen MP, Tildesley DJ (1989) Computer simulation of liquids. Clarendon Press, Oxford Brenner DW (2000) The art and science of an analytic potential. Phys Status Solidi B 217:23 Frenkel D, Smit B (2002) Understanding molecular simulation: from algorithm to applications, 2nd edn. Academic, San Diego Haile JM (1992) Molecular dynamics simulation. Wiley, Chichester Leach AR (2001) Molecular modelling: principles and applications, 2nd edn. Prentice Hall, Harlow Tocci E, Pullumbi P (2011) Chapter 1: Multi-scale molecular modeling approaches for designing/selecting polymers used for developing novel membranes. In: Drioli E, Barbieri G (eds) Membrane engineering for the treatment of gases: gas-separation problems with membranes. The Royal Society of Chemistry, Cambridge, UK, pp 1–28

A

Autothermal Reforming Fausto Gallucci Faculty of Chemical Engineering and Chemistry, Eindhoven University of Technology, Eindhoven, The Netherlands

Autothermal reforming or oxidative steam reforming is a combination of conventional steam reforming of the fuel (endothermic reaction) with the partial oxidation of a small amount of the fuel (exothermic reaction) in order to achieve an autothermal reaction that proceeds without external input of energy (Chang et al. 2010; Tiemersma et al. 2012). The most studied autothermal reforming is the conversion of methane to hydrogen (Gallucci et al. 2009). The overall chemical reactions taking place in the autothermal reforming of methane include steam reforming (Eq. 1), water gas shift (Eq. 2), and total oxidation (Eq. 3). The energy generated by the oxidation reaction and WGS is used for the SMR: CH4 þ H2 O Ð CO þ 3H2

(1)

CO þ H2 O Ð CO2 þ H2

(2)

CH4 þ 2O2 ! CO2 þ 2H2 O

(3)

This reaction system can be carried out efficiently in a membrane reactor as the extraction of # Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_39-3

hydrogen during the reaction shifts the equilibrium reactions toward completion at moderate temperatures, and thus the extent of oxidation reaction to achieve autothermal reforming is moderate. One of the problems of autothermal reforming carried out in membrane reactors is the mismatch between the oxidation reaction rate and the reforming reaction rate. The oxidation is often much faster than the reforming, and for this reason in packed bed membrane reactors, a hightemperature region is obtained at the beginning of the bed followed by a low-temperature region at the end of the bed. This could cause problems to the membranes that could be damaged by high temperatures while not working properly (low flux – see Richardson equation) at lower temperatures (Tiemersma et al. 2006; Gallucci et al. 2010). To circumvent these problems, fluidized bed membrane reactors are often proposed for this kind of reaction system, as the solid circulation inside the reactor allows a virtually isothermal condition even in case of highly exothermic reactions. Examples of autothermal reforming (ATR) (or oxidative steam reforming) reactions also include the ATR of ethanol and methanol or the ATR of naphtha (Lin et al. 2010; Tosti et al. 2010; Moreno and Wilhite 2009). All these reactions have been successfully tested in membrane reactors for pure hydrogen production.

2

References Chang H-F, Pai W-J, Chen Y-J, Lin W-H (2010) Autothermal reforming of methane for producing high-purity hydrogen in a Pd/Ag membrane reactor. Int J Hydrog Energy 35(23):12986–12992, Retrieved from http://www.scopus.com/inward/ record.url?eid=2-s2.0-78049471869&partnerID=40 &md5=631e9a956f70441c60cc3fb3987ed76c Gallucci F, Van Sint Annaland M, Kuipers JAM (2009) Autothermal reforming of methane with integrated CO2 capture in novel fluidized bed membrane reactors. Asia Pac J Chem Eng 4(3):334–344. doi:10.1002/apj Gallucci F, Van Sintannaland M, Kuipers JAM (2010) Theoretical comparison of packed bed and fluidized bed membrane reactors for methane reforming. Int J Hydrog Energy 35(13):7142–7150, Retrieved from http://www.scopus.com/inward/record.url?eid= 2-s2.0-77954826477&partnerID=40&md5=569ec26 bf526471832579c3f965b1d90 Lin W-H, Liu Y-C, Chang H-F (2010) Autothermal reforming of ethanol in a Pd-Ag/Ni composite membrane reactor. Int J Hydrog Energy 35(23):12961–12969, Retrieved from http://www.scopus.com/inward/record. url?eid=2-s2.0-78049467904&partnerID=40&md5= 0048616ca9b8d29ade703c2eea0ab1bf

Autothermal Reforming Moreno AM, Wilhite BA (2009) Autothermal hydrogen generation from methanol in a ceramic microchannel network. In: Conference proceedings – 2009 AIChE annual meeting, 09AIChE. Retrieved from http://www. scopus.com/inward/record.url?eid=2-s2.0-779522829 00&partnerID=40&md5=0650ebd8cd9440d3ac32e03 b4f859c7d Tiemersma TP, Patil CS, Sint Annaland MV, Kuipers JAM (2006) Modelling of packed bed membrane reactors for autothermal production of ultrapure hydrogen. Chem Eng Sci 61(5):1602–1616, Retrieved from http://www.scopus.com/inward/record.url?eid=2-s2.030344455632&partnerID=40&md5=9fb77a0611e8e6 a89745a4b8625bbebf Tiemersma TP, Kolkman T, Kuipers JAM, van Sint Annaland M (2012) A novel autothermal reactor concept for thermal coupling of the exothermic oxidative coupling and endothermic steam reforming of methane. Chem Eng J 203:223–230, Retrieved from http://www. scopus.com/inward/record.url?eid=2-s2.0-848656377 97&partnerID=40&md5=47095694b70255a4e72794 5b6ad24016 Tosti S, Borelli R, Santucci A, Scuppa L (2010) Pd-Ag membranes for auto-thermal ethanol reforming. Asia Pac J Chem Eng 5(1):207–212, Retrieved from http:// www.scopus.com/inward/record.url?eid=2-s2.0-77649 116340&partnerID=40&md5=50f41a519d60bbaa07 aa2f96c86792bc

A

Azeotropic Distillation Alessandra Criscuoli Institute of Membrane Technology (ITM-CNR), Rende, CS, Italy

Azeotropic distillation is a particular type of distillation by which it is possible to separate azeotropes (Perry and Green 1984). Azeotropes are mixtures of two or more substances that boil together at a constant temperature. It is, therefore, impossible to separate them directly by distillation, because a distillate of the same composition of the liquid feed is produced. Azeotropic distillation is based on the addition of a compound that acts on the volatility of the substances contained into the azeotrope, so that a new azeotrope, made of the substances present in the starting azeotrope and the added compound, is formed. The new

# Springer-Verlag Berlin Heidelberg 2013 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_40-3

azeotrope can be, then, removed by distillation, and a residual highly rich in one of the substances is obtained. A typical example is the distillation of mixtures of water-ethanol. By classic distillation, the distillate contains 95 % of alcohol (pure boiling temperature, 78.4 C) and 5 % of water (pure boiling temperature, 100 C), and no further increments in ethanol purity can be reached, because this mixture boils, as a unique compound, at 78.17 C. However, if benzene or cyclohexane is added to the mixture, a ternary azeotrope is formed that incorporates all the water and has a lower boiling point than ethanol. Therefore, if the distillation is now carried out, the new azeotrope will be recovered as distillate, while practically pure ethanol is produced as residue (see Fig. 1). Besides azeotropic distillation, other methods can be also employed for the separation of azeotropes and, in particular, membrane operations like pervaporation and membrane distillation.

2

Azeotropic Distillation

Ternary azeotrope Benzene or cyclohexane

+

Ternary azeotrope + ethanol

Distillation

Water-ethanol azeotrope Ethanol

Azeotropic Distillation, Fig. 1 Azeotropic distillation of water-ethanol

References Perry RH, Green D (1984) Perry’s chemical engineers’ handbook. McGraw-Hill, Singapore

B

Bacteria and Spore Removal Karin Schroen Laboratory of Food Process Engineering, Wageningen University and Research Centre, Wageningen, The Netherlands

Bacteria can be of great influence on many processes, be it in, for example, food, pharma, fermentation, or water production. Obviously, bacteria (and their spores) can be removed by heat treatment or addition of specific components that inhibit their growth or even kill them, but the downside of this is that the product properties will be influenced and this is mostly undesirable. In this respect, membrane microfiltration could be an interesting alternative when used for cold sterilization. If the bacterial count is not too high and other components are sufficiently smaller, the removal of bacteria and spores can rather easily be carried out using dead-end microfiltration. When the size of bacteria and other components overlap, this separation is far from straightforward. An illustrative example can be found in dairy separation, in which cold sterilization of “milk” has been reported. The components that are most important for this separation are the milk fat globules (cream; typical sizes from 0.1 to 15 mm), bacteria (0.5–5 mm), and casein micelles (20–300 nm). Since cream and bacteria overlap in size, the cream is first removed by centrifugation. The resulting skim milk receives the cold # Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_41-5

sterilization treatment. Various researchers have investigated how far the bacterial count could be reduced as listed in Table 1. From the table, it is clear that a variety of membranes and process conditions have been used, ranging from very high to low cross-flow velocities, and application of a uniform transmembrane pressure or frequent back pulsing. These conditions are clearly aiming at processing under very different conditions, where frequent back pulsing will control the amount of deposited material, the uniform transmembrane pressure concept aims at stable filtration conditions along the length of the membrane. All approaches have shown interesting results; the log reductions that can be achieved are around 4 (10,000-fold reduction), although it should be noted that these values are not as high as obtained after regular heat treatment. The highest log reduction (6.6, which is higher than for regular pasteurization) was claimed for microsieves, which are silicon plates with very uniform pores prepared by laser interference lithography (Van Rijn and Elwenspoek 1995). Although the bacterial reduction was measured for dead-end filtration using a 0.5 mm microsieve and SMUF (simulated ultrafiltrate) spiked with Bacillus subtilis, it is expected that the high log count reduction is a result of its narrow pore size distribution of the microsieve. In case of the uniform transmembrane pressure concept, reduction of bacteria and spores by microfiltration is carried out near the critical

2

Bacteria and Spore Removal

Bacteria and Spore Removal, Table 1 Comparison of cold sterilization results from various sources Process conditions cross-flow/pressure, UTP, back pulsing 50 kPa, 7.2 m/s UTP, skim milk

Log reduction Above 3.5

Reversed asymmetric 0.87 mm; 1.4•10 4 m/s Microsieve 0.5 mm

0.5–1 m/s; back pulsing frequency 0.2–1 s 1, skim milk Dead-end filtration of spiked SMUF

Between 4 and 5 6.6

Bactocatch: ceramic membranes

6–8 m/s, skim milk, UTP

Membrane type and flux Ceramic 1.4 mm; 1.4•10

4

m/s

pressure, at which the amount of particles that is carried toward that membrane by permeate is counterbalanced by the amount of particles diffusing away from the membrane to the feed solution. We expect that this can be even taken one step further, as indicated in the work of van Dinther et al. (2011), in which particles that size-wise correspond to milk fat globules and bacteria were reported to be separated by fluid skimming and lift effects. For this to occur, first a nonporous channel is used to induce particle migration after which in a porous area the small particles that are situated close to the wall can specifically be removed. This process has not been demonstrated at large scale, but it holds great promise since it would allow for direct separation of bacteria from full milk without the need of centrifugation. Besides, the separation is no longer determined by the pore size of the membrane; metal sieves from SPG Veco with uniform pores of around 20 mm were used, but by the process conditions that determine which part of the feed is removed.

Source Saboya and Maubois 2000 Guerra et al. 1997 Van Rijn and Kromkamp 2001 Holm et al. 1989

References Guerra A, Jonsson G, Rasmussen A, Waagner Nielsen E, Edelsten D (1997) Low cross flow velocity microfiltration of skim milk for removal of bacterial spores. Int Dairy J 7:849 Holm S, Malmberg R, Svensson K (1989) Method and plant producing milk with low bacterial content. WO Patent 8:601–687 Saboya LV, Maubois JL (2000) Current developments of microfiltration technology in the dairy industry. Lait 80:541 Van Dinther AMC, Schroe¨n CGPH, Boom RM (2011) High-flux membrane separation using fluid skimming dominated convective fluid flow. J Membr Sci 371(1–2):20 Van Rijn CJM, Elwenspoek MC (1995) Micro filtration membrane sieve with silicon micro machining for industrial and biomedical applications. In: Proceedings of micro electro mechanical systems (MEMS), Amsterdam, p 83 Van Rijn CJM, Kromkamp J (2001) Method for filtering milk. WO Patent 0,209,527. http://www.spgveco.com/ precision+metal/applications/filtration. Accessed July 2012

B

Bacterial Biofilm Formation Lidietta Giorno and Napoleone D’Agostino Institute on Membrane Technology, National Research Council of Italy, ITM-CNR, University of Calabria, Rende (CS), Italy

Synonyms Bacterial biofilm; Bacterial adhesion; Biofouling Biofilms were observed as early as 1674, when Antonie van Leeuwenhoek used his primitive but effective microscope to describe aggregates of “animalcules” that he scraped from human tooth surface (Costerton 1999). Since then, more accurate descriptions of biofilms are made. Bacteria generally exist in one of two types of population: planktonic, freely existing in bulk solution, and sessile, as a unit attached to a surface or within the confines of a biofilm (Garrett et al. 2008). Biofilm is a result of many complicated steps. It includes the formation of a conditioning film on a material’s surface, the movement of bacteria, an attachment process, the growth on material surfaces, and the breakdown finally. For bacteria, the advantages of biofilm formation are numerous. These advantages include: protection from antibiotics (Godberg 2002), disinfectants (Peng et al. 2002), and dynamic environments (Chen et al. 1998). Over the past few decades, biofilm growth has been observed in many industrial and domestic # Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_42-1

domains. Many industries suffer the ill effects of biofilm growth of one type or another, which can result in heavy costs in cleaning and maintenance. Biofilms occurring in membrane systems may cause severe loss performance and the use of costly cleaning procedures to maintain output and quality. The fouling is often so severe that acceptable operation cannot be maintained and membrane replacements are needed. It is necessary to understand the biofilm formation mechanism with the aim to propose a solution to contrast this fouling. Bacteria are capable of colonizing almost any surface and have been found at extreme conditions such as temperatures from 12 C to 110 C and pH values between 0.5 and 13. Biofilm growth occurs by physical, chemical, and biological processes. Fletcher described the accumulation of microorganisms on a collecting surface as a process of three stages: (i) adsorption, or the accumulation of an organism on a collector surface, i.e., substrate (deposition); (ii) attachment, or the consolidation of the interface between an organism and a collector, often involving the formation of polymer bridges between the organism and collector; and (iii) colonization, or growth and division of organisms on the collector’s surface. Although useful as a snapshot of biofilm growth, this type of profile is limited when considering the intimate processes of cell–substrate/cell–cell interaction. Characklis and Marshal later described an eight-step process which included the formation

2

of an initial conditioning layer, reversible and irreversible adhesion of bacteria, and the eventual detachment of cells from a mature biofilm for subsequent colonization. Anything that may be present within the bulk fluid can through gravitational force or movement of flow settle onto a substrate and become part of a conditioning layer. This layer modifies substrata facilitating accessibility to bacteria. Surface charge, potential, and tensions can be altered favorably by the interactions between the conditioning layer and substrate. The substrate provides anchorage and nutrients augmenting growth of the bacterial community. Initially, planktonic microbial cells are transported from bulk liquid to the conditioned surface either by physical forces or by bacterial appendages such as flagella. The reversible adsorption of a fraction of the cells reaching the surface normally occurs. Local environmental variables which contribute to bacterial adhesion are factors such as available energy, surface functionality, bacterial orientation, temperature, and pressure conditions. If repulsive forces are greater than the attractive forces, the bacteria will detach from the surface. The probability of this phenomenon occurs is higher before the formation of the conditioning layer. The activation energy for desorption of bacteria is low and so it is likely to occur, underlining the weakness of the bonds. Physical forces associated to bacterial adhesion include the van der Waals forces, steric interactions, and electrostatic (double layer) interaction, collectively known as the DVLO (Derjaguin, Verwey, Landau, and Overbeek) forces, which originally described the interaction of a colloidal particle with a surface (Rutter and Vincent 1980). According to this theory, the total interaction between a surface and a particle is the summation of their van der Waals and Coulomb interactions. Since the van der Waals attractive force is dominant in the vicinity of a surface, particles adhere irreversibly because they cannot separate from the surface by Brownian motion. In contrast, the Coulomb interaction becomes dominant at a distance away from the surface because the van der Waals force decreases sharply with distance. Other interactions that DVLO theory

Bacterial Biofilm Formation

takes into consideration are hydrophobic–hydrophilic and osmotic (Chang and Chang 2002) and have also been described in terms of thermodynamic interaction (Gallardo-Moreno et al. 2002). In real time, a number of the reversibly adsorbed cells remain immobilized and become irreversibly adsorbed. It has been argued that the physical appendages of bacteria (flagella, fimbriae, and pili) overcome the physical repulsive forces of the electrical double layer (De Weger 1987). Subsequently, the appendages make contact with the bulk lattice of the conditioning layer stimulating chemical reactions such as oxidation and hydration and consolidating the bacteria–surface bond. Some evidence has shown that microbial adhesion strongly depends on the hydrophobic–hydrophilic properties of interacting surfaces (Liu et al. 2004). As the stationary cells divide (binary division), daughter cells spread outward and upward from the attachment point to form clusters. Typically, such interactions and growth within the developing biofilm form into a mushroom-like structure. This structure is believed to allow the passage of nutrients to bacteria deep within a biofilm. After an initial stage, a rapid increase in population is observed, otherwise described as the exponential growth phase. This depends on the nature of the environment, both physically and chemically. The rapid growth occurs at the expense of the surrounding nutrients from the bulk fluid and the substrate. At this stage the physical and chemical contribution to the initial attachment ends and the biological processes begin to dominate. Excretion of polysaccharide intercellular adhesin (PIA) polymers and the presence of divalent cations interact to form stronger bonding between cells (Dunne 2002). Differential gene expression between the two bacterial states (planktonic/sessile) is in part associated to the adhesive needs of the population. For example, the production of surface appendages is inhibited in sessile species as motility is restricted and no longer necessary. Simultaneously, expression of a number of genes for the production of cell surface proteins and excretion products increases. Surface

Bacterial Biofilm Formation

proteins (porins), such as Opr C and Opr E, allow the transport of extracellular products into the cell and excretion materials out of the cell, e.g., polysaccharides (Hancock et al. 1990). The structure of many Gram-negative bacterial polysaccharides is relatively simple, comprising either homopolysaccharides or heteropolysaccharides (Sutherland 2001). These molecules impart mechanical stability and are pivotal to biofilm adhesion and cohesion and evasion from harsh dynamic environmental conditions. They consolidate the biofilm structure. Hall-Stoodley and Stoodley identified the differences in gene expression of planktonic and sessile cells, and as many as 57 biofilm-associated proteins were not found in the planktonic profile. The stationary phase of growth describes a phase where the rate of cell division equals the rate of cell death. At high cell concentration, a series of cell signaling mechanisms are employed by the biofilm, and this is collectively termed quorum sensing (Bassler 1999). Quorum sensing describes a process where a number of autoinducers (chemical and peptide signals in high concentrations, e.g., homoserine lactones) are used to stimulate genetic expression of both mechanical and enzymatic processors of alginates, which form a fundamental part of the extracellular matrix. The death phase sees the breakdown of the biofilm. Enzymes are produced by the community itself which break down polysaccharides holding the biofilm together, actively releasing surface bacteria for colonization of fresh substrates. Alginate lyase produced by Pseudomonas fluorescens and Pseudomonas aeruginosa, N-acetyl-heparosan lyase by Escherichia coli, and hyaluronidase by Streptococcus equi are examples of the enzymes used in the breakdown of the biofilm matrix (Sutherland 1999). Simultaneously, the operons coding for flagella proteins are upregulated so that the organisms have the apparatus for motility and the genes coding for a number of porins are downregulated, thus completing a genetic cycle for biofilm adhesion and cohesion. Changes in pH can have a marked effect on bacterial growth and as such are frequently exploited in the production of detergents and

3

disinfectants used to kill bacteria. Bacteria possess membrane-bound proton pumps which extrude protons from the cytoplasm to generate a transmembrane electrochemical gradient (Rowland 2003), i.e., the proton motive force. The passive influx of protons in response to the proton motive force can be a problem for cells attempting to regulate their cytoplasmic pH (Booth 1985). Large variations in external pH can overwhelm such mechanisms and have a biocidal effect on the microorganisms. Bacteria respond to changes in internal and external pH by adjusting the activity and synthesis of proteins associated with many different cellular processes (Olsen 1993). Studies have shown that a gradual increase in acidity increases the chances of cell survival in comparison to a sudden increase by rapid addition of HCl (Li 2001). This suggests that bacteria contain mechanisms in place which allow the bacterial population to adapt to small environmental changes in pH. However, there are cellular processes which do not adapt to pH fluctuations so easily. One such process is the excretion of exopolymeric substances (polysaccharides). Optimum pH for polysaccharide production depends on the individual species, but it is around pH 7 for most bacteria. Both mixed species and pure culture biofilms behave like viscoelastic fluids. Biofilms exhibit both irreversible viscous deformation and reversible elastic response and recoil (Ohashi and Harada 2004). Extracellular polymeric substances like alginate, xanthan, and gellan gum aggregate due to hydrogen bonding to form highly hydrated viscoelastic gels (Stoodley et al. 1999). The presence of acetylated uronic acids in the bacterial alginate of P. aeruginosa biofilms increases its hydration capacity. These properties provide the biofilm with mechanical stability (Stoodley et al. 2002). The matrix formed by EPS responds to stress by exhibiting (i) elastic tension due to a combination of polymeric entanglement, entropic, and weak hydrogen bonding forces; (ii) viscous damping due to polymeric friction and hydrogen bond breakage; and (iii) alignment of the polymers in the shear direction (Klapper et al. 2002). Such properties change with increased

4

temperature. Increasing the temperature of polysaccharides produces a gel-like substance which gradually increases in strength until a critical point is reached. At the critical point the gel forms a solution (Villain-Simonnet et al. 2000). Such behavior affects the viscosity of the polysaccharides which can affect biofilm adherence. The optimum temperature for a microorganism is associated with an increase in nutrient intake resulting in a rapid formation of biofilm (Stepanovic et al. 2003). Enzymes are responsible of nutrient metabolism; then the formation of a biofilm is dependent on the presence and reaction rates of enzymes. Temperature influences the reaction rate of enzymes having an impact on the development of the cells. Optimum temperatures result in the healthy growth of bacterial populations. Temperatures away from the optimum negatively influence bacterial enzyme reaction rates, and a reduction of bacterial growth efficiency occurs. Fletcher reported the effect of temperature on attachment of stationary phase cells. Findings showed that a decrease in temperature reduced the adhesive properties of a marine Pseudomonad. It is believed that the effect was due to a decrease in the bacterial surface polymer at lower temperatures as well as effects such as reduced surface area. However, Herald and Zottola observed that the presence of bacterial surface appendages was dependent on temperature. At 35 C cells were shown to have a single flagellum, while at 21 C they had two to three flagella, and at 10 C, cells exhibited several flagella. This may suggest that the initial interaction between the bacteria and substrate may increase with a lowering of temperature, increasing the likelihood of adhesion. Perhaps the more uniform properties of polysaccharides at lower temperatures increase the possibility of biofilm adhesion, because many microbial polysaccharides undergo transition from an ordered state at lower temperatures and in the presence of ions to a disordered state at elevated temperature under low ionic environments (Nisbet et al. 1984). Although there is plenty of information describing the effect of temperature on bacterial growth in culture, the effect of temperature on the removal of adhered microorganisms is not so well

Bacterial Biofilm Formation

documented. The reports available describe fairly radical effects of temperature on adhered bacteria. Marion-Ferey et al. observed the effect of high temperatures (80–90 C) on the removal of biofilms. It was discovered that these temperatures were not effective for biofilm removal due to “baking effects” at high temperature, apparently increasing the adherent nature of the biofilm to the surface.

References Bassler BL (1999) How bacteria talk to each other: regulation of gene expression by quorum sensing. Curr Opin Microbiol 2:582–587 Booth IR (1985) Regulation of cytoplasmic pH in bacteria. Microbiol Rev 49:359–378 Chang YI, Chang PK (2002) The role of hydration force on the stability of the suspension of Saccharomyces cerevisiae – application of the extended DLVO theory. Colloids Surf A Physicochem Eng Asp 211:67–77 Characklis WG, Marshal KC (1990) Biofilms. Wiley, New York Chen MJ, Zhang Z, Bott TR (1998) Direct measurement of the adhesive strength of biofilms in pipes by micromanipulation. Biotechnol Tech 12:875–880 Costerton JW (1999) Introduction to biofilm. Int J Antimicrob Agents 11:217–221 De Weger LA, van der Vlugt C, Wijfjes AHM et al (1987) Flagella of a plant-growth-stimulating Pseudomonas fluorescens strain are required for colonization of potato roots. J Bacteriol 169:2769–2773 Dunne WM (2002) Bacterial adhesion: seen any good biofilms lately? Clin Microbiol Rev 15:155–166 Fletcher M (1977) The effects of culture concentration and age, time, and temperature on bacterial attachment to polystyrene. Can J Microbiol 23:1–6 Fletcher M (1980) Microbial adhesion to surfaces. Ellis Horwood, Chichester Gallardo-Moreno AM, Gonzalez-Martin ML, PerezGiraldo C et al (2002) The measurement temperature: an important factor relating physicochemical and adhesive properties of yeast cells to biomaterials. J Colloid Interface Sci 271:351–358 Garrett TR, Bhakoo M, Zhang Z (2008) Bacterial adhesion and biofilms on surfaces. Prog Nat Sci 18:1049–1056 Godberg J (2002) Biofilms and antibiotic resistance: a genetic linkage. Trends Microbiol 10:264 Hall-Stoodley L, Stoodley P (2002) Developmental regulation of microbial biofilms. Curr Opin Biotechnol 13:228–233 Hancock REW, Siehnel R, Martin N (1990) Outer membrane proteins of Pseudomonas. Mol Microbiol 4:1069–1075

Bacterial Biofilm Formation Herald PJ, Zottola EA (1988) Attachment of Listeria monocytogenes to stainless steel surfaces at various temperatures and pH values. J Food Sci 53:1549–1552 Hiemenz PC, Rajagopalan R (1997) Principles of colloid and surface chemistry. Marcel Dekker, New York Klapper I, Rupp CJ, Cargo R et al (2002) Viscoelastic fluid description of bacterial biofilm material properties. Biotechnol Bioeng 80:289–296 Li Y (2001) Cell density modulates acid adaptation in Streptococcus mutans. J Bacteriol 183:6875–6884 Liu Y, Yang S, Xu H et al (2004) The influence of cell and substratum surface hydrophobicities on microbial attachment. J Biotechnol 110:251–256 Marion-Ferey K, Pasmore M, Stoodley P et al (2002) Biofilm removal from silicone tubing: an assessment of the efficacy of dialysis machine decontamination procedures using an in vitro model. J Hosp Infect 53:64–71 Marshall KC, Stout R, Mitchell R (1971) Mechanism of the initial events in the sorption of marine bacteria to surfaces. J Gen Microbiol 68:337–348 Nisbet BA, Sutherland IW, Bradshaw IJ et al (1984) XM-6: a new gel-forming bacterial polysaccharide. Carbohydr Polym 4:377–394 Ohashi A, Harada H (2004) Adhesion strength of biofilm developed in an attached-growth reactor. Water Sci Technol 29:281–288 Olsen ER (1993) Influence of pH on bacterial gene expression. Mol Microbiol 8:5–14

5 Peng JS, Tsai WC, Chou CC (2002) Inactivation and removal of Bacillus cereus by sanitizer and detergent. Int J Food Microbiol 77:11–18 Rowland BM (2003) Bacterial contamination of dental unit waterlines: what is your dentist spraying into your mouth? Clin Microbiol Newsl 25:73–77 Rutter PR, Vincent B (1980) Microbial adhesion to surfaces. Ellis Horwood, London Stepanovic S, Cirkovic I, Mijac V et al (2003) Influence of the incubation temperature, atmosphere and dynamic conditions on biofilm formation by Salmonella spp. Food Microbiol 20:339–343 Stoodley P, Lewandowski Z, Boyle J et al (1999) Structural deformation of bacterial biofilms caused by shortterm fluctuations in fluid shear: an in situ investigation of biofilm rheology. Biotechnol Bioeng 65:83–92 Stoodley P, Cargo R, Rupp CJ et al (2002) Biofilm material properties as related to shear-induced deformation and detachment phenomena. J Ind Microbiol Biotechnol 29:361–367 Sutherland IW (1999) Polysaccharases for microbial exopolysaccharides. Carbohydr Polym 38:319–328 Sutherland IW (2001) Microbial polysaccharides from Gram-negative bacteria. Int Dairy J 11:663–674 Villain-Simonnet A, Milas M, Rinaudo MA (2000) New bacterial exopolysaccharide (YAS34). II. Influence of thermal treatments on the conformation and structure. Int J Biol Macromol 27:77–87

S

Silicalite Membrane Tsuneji Sano Department of Applied Chemistry, Hiroshima University, Graduate School of Engineering, Higashi-Hiroshima, Japan

Zeolites are a class of crystalline aluminosilicates with highly regular and open microporous structures. More than 200 types of zeolite frameworks have been identified by the Structure Commission of the International Zeolite Association. Zeolite membranes combine the great advantages of inorganic membranes, such as temperature stability and resistance against solvents, with the molecular sieving effect. Zeolite silicalite (MFI, where the three characters indicate the framework type) with Si/Al ratio of 1 is well known as a hydrophobic zeolite. Figure 1 shows the framework structure of MFI zeolite (International Zeolite Association Web site 2013). There are two channel systems: a straight Flux kg=m2 h ¼

channel running to (010) with ten-membered ring ˚ and a sinusoidal channel openings of 5.3 5.6 A parallel to the (100) with ten-membered ring ˚. openings of 5.1 5.5 A Silicalite membrane was prepared as a selfstanding polycrystalline film at first and was very fragile (Sano et al. 1991). Therefore, the silicalite membranes are crystallized on porous supports such as sintered stainless steel disc or alumina disc(tube) with an average pore diameter of 0.5 2 mm. Colloidal silica is added to a stirred mixture of tetrapropylammonium bromide (TPABr) and sodium hydroxide in solution, to give a hydrogel with a composition of 0.1 TPABr-0.05 Na2OSiO2-80H2O. The hydrogel is transferred to a stainless steel autoclave and kept at 170 C for 48 h. The separation performances can be evaluated by pervaporation measurements using various aqueous alcohol solutions as a feed. Flux and separation factor a(ROH/H2O) are calculated from following equations:

ðWeight of permeate, kgÞ ðMembrane area, m2 Þ ðPermeation time, hÞ

Separation factor aðROH=H2 OÞ ¼

# Springer-Verlag Berlin Heidelberg 2015 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_43-1

½CROH =CH2 O Permeate ½CROH =CH2 O Feed

(1)

(2)

2

Silicalite Membrane

where the CROH and CH2 O are the volume fractions of alcohols and water, respectively. Figure 2 shows scanning electron micrographs (SEM) of the outer surface and cross-section of the membrane on the stainless steel support (Sano et al. 1994). The surface is formed of an aggregate of crystals of 10 30 mm and the growth on the support led to a randomly grown crystalline layer. The average thickness of the silicalite layer was confirmed by Si line analysis using energydispersive X-ray analysis (EDX). The organic amine used in the silicalite synthesis as a template remains in the channels of silicalite crystals. In order to use the membrane as the separation membrane, the amine must be removed from the channels by certain procedures. As the silicalite

Silicalite Membrane, Fig. 1 Framework structure of silicalite (MFI) zeolite (IZA web. 2013) Silicalite Membrane, Fig. 2 SEM images of (a) outer surface and (b) crosssection of silicalite membrane on stainless steel support (Reproduced from Sano et al. (1994) with the permission of Elsevier)

membrane experiences the irregular stresses that arise from a difference in the thermal expansion between the support and the silicalite crystals or from removal of volatile materials from the zeolite framework, cracks are easily formed within the membrane during the treatment process. Therefore, pretreatment conditions of membranes affect strongly the separation performance. As listed in Table 1 (Sano et al. 1994), the silicalite membrane after air-drying at 100 C (containing TPA+) shows a very low flux combined with a low separation factor below 1, indicating that there is no cracks and pores between silicalite grains within the membrane before the pretreatment. However, in the case of the membrane after the thermal treatment, the flux and the separation factor increase with an increase in the treatment temperature and period. The membrane calcined at 500 C to remove the template completely shows the high flux combined with the higher separation factor. The membrane changes from the water-selective membrane to the ethanol-selective one by decreasing the amount of TPA+ occluded in the zeolite framework, and the separation of ethanol/water takes place by transport through the zeolite channels. Figure 3 displays effects of the feed temperature on the separation factor and the flux for various alcohol/water mixtures (Sano et al. 1994). The high separation factor is obtained for 1-propanol/ water mixture, although the flux is very low. On

a

b

25 μm

250 μm

Silicalite Membrane

3

Silicalite Membrane, Table 1 Influence of pretreatment conditions on pervaporation performance of silicate membrane (Reproduced from Sano et al. (1994) with the permission of Elsevier) Treatment condition Air-drying In vacuum In vacuum Calcination

100 C 300 C 380 C 500 C

Separation factor a (EtOH/H2O) 0.38 0.58 7.8 58

Flux (kg/m2 h) 0.00303 0.00840 0.0394 0.760

12 h 6h 6h 20 h

Feed temperature: 60 C Feed ethanol concentration: 5 vol.%

100

2

Separation factor α (ROH/H2O)

MeOH

Flux (kg/m2h)

1.5

1 EtOH

0.5 2-PrOH

0 20

1-PrOH 30

40

50

60

70

1-PrOH

80

60

EtOH

2-PrOH

40

20

0 20

Feed temperature (°C)

MeOH

30

40

50

60

70

Feed temperature (°C)

Silicalite Membrane, Fig. 3 Pervaporation flux and separation factor a for various alcohol/water mixtures (Reproduced from Sano et al. (1994) with the permission of Elsevier). Feed alcohol concentration: 1 mol%

the other hand, the higher flux and the lower separation factor are obtained for methanol/ water mixture. This can be explained by the differences in the molecular size and the interaction between alcohol and silicalite, methanol being the smallest and the most polar molecule in the group of alcohols tested here.

References International Zeolite Association Web site (2013) http:// www.iza-online.org/ Sano T, Kiyozumi Y, Kawamura M, Mizukami F, Takaya H, Mouri T, Inaoka W, Toida Y, Watanabe M, Toyoda K (1991) Preparation and characterization of ZSM-5 zeolite film. Zeolites 11:842–845 Sano T, Hasegawa M, Kawakami Y, Kiyozumi Y, Yanagishita H, Kitamoto D, Mizukami F (1994) Potentials of silicalite membrane for the separation of alcohol/water mixture. Stud Surf Sci Catal 84:1175–1182

B

Batch Diafiltration Zoltán Kovács Department of Food Engineering, Institue of Bioengineering and Process Engineering, Szent Istvan University, Budapest, Hungary

Batch diafiltration refers to a pressure-driven membrane filtration process in which a diluant (pure solvent) is added into the feed tank in order to enhance the degree of separation of macrosolutes from microsolutes. In batch diafiltration, in contrast to continuous diafiltration, the retentate stream is recirculated to the feed tank, and only the permeate stream is collected separately. During the operation, solute-free diluant is introduced into the feed tank to replace solvent losses as schematically illustrated in Fig. 1. The requirement for an effective separation is the utilization of a membrane which highly retains the macrosolute but permeable for the microsolute. Thus, depending on the application, microfiltration, ultrafiltration, nanofiltration, or even reverse osmosis membranes can be applied. In its strict, original sense, the term batch diafiltration refers to a process that aims at removing the microsolutes from the process liquor. The standard way of achieving this purification goal is to employ a constant-volume diafiltration (CVD) process that is probably the most common type of batch diafiltration. In CVD, # Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_45-6

the feed volume is kept constant by continuously adding a diluant at a rate equal to the permeation rate. It should be pointed out that the removal of microsolutes and the concentration of macrosolutes (i.e., the reduction of the volume of process liquor) are both required for most applications. The term batch diafiltration, in its broader sense, may stand for a batch filtration process that is designed to achieve the twin objectives of concentrating and purifying a multisolute system according to a specific diluant utilization strategy. In this context, batch diafiltration is a complex process that may involve a sequence of consecutive operational steps. A straightforward way of achieving the dual objectives of concentration and fractionation is to combine CVD with concentration mode operational steps (i.e., in which no diluant is added into the feed tank). A so constructed typical three-step process, also referred to as traditional diafiltration (TD), involves the following phases: (i) concentration mode to achieve an intermediate macrosolute concentration, (ii) constant-volume diafiltration to “wash out” the microsolute by a pure solvent introduced into the system, and (iii) further concentration to the final desired macrosolute concentration. Beside TD, a number of alternative strategies have been proposed. These include the sequential dilution diafiltration (SDD), intermittent feed diafiltration (IFD), variable-volume diafiltration (VVD),

2

Batch Diafiltration

Batch Diafiltration, Fig. 1 Schematic representation of batch diafiltration configuration

Diluant u(t) Retentate

Permeate q(t)

Feed

pre-concentration combined with variablevolume diafiltration (CVVD), and dynamicvolume diafiltration (DVD). Diafiltration techniques differ in controlling the addition of the diluant in terms of quantity and duration. The differences between the various operations are best described by the proportionality factor a (i.e., the ratio of diluant flow d(t) to permeate flow q(t)) as a function of operation time (Foley 2006). For instance, TD process is characterized with a sequence a(t) = {0, 1, 0} with two unknown switching times at the end of the first and of the second time interval. Similarly, CVVD process has two phases with constant a levels a(t) = {0, a1} and an unknown switching time. Table 1 shows the diluant control strategies applied in batch processing. Note that the best time-varying profile of diluant addition needs not necessarily be one of the arbitrarily predefined profiles. The diafiltration process, that is, designed by the evaluation of the optimal time-varying profile of the diluant flow, is referred to as dynamic-volume diafiltration (Paulen et al. 2012). The governing differential equations for a generalized batch diafiltration process are given as dci ci q ¼ ðRi aÞ, ci ð0Þ ¼ ci0 , V dt dV ¼ ða 1Þq, V ð0Þ ¼ V 0 : dt

i ¼ 1, 2

where ci is the solute concentration in the feed tank, Ri the rejection of component i, and

V represents the feed tank volume. This initial value problem describes the evolution in time of the feed volume and the feed concentration of any solute under the assumption that the diluant is solute-free and the feed tank is well-mixed. Note that the time-dependent variables (i.e., permeate flux and the solute rejections) are, in a general case, a function of feed concentrations and may vary with operation conditions (temperature, pressure, cross-flow velocity, etc.). The unique feature of realizing both concentration and fractionation puts membrane filtration in an attractive position and compares favorably with other separation methods or even with a sequence of consecutive unit operations. In comparison with continuous processes, batch operations are especially suited to small-scale operations, require less expensive automatic controls, and enable a reduced membrane area in order to reach the target (Baker 2004). Most batch plants operate under constant mechanistic membrane pressure adjusted simply by the retentate valve. There exist, however, other types of process control strategies in engineering practice, such as constant flux or constant wall concentration control (van Reis et al. 1997). These are normally employed when unfavorable side effects (e.g., enhanced fouling or product quality deterioration) occur that can be associated with the high concentration at the membrane wall.

Batch Diafiltration

3

Batch Diafiltration, Table 1 Diluant utilization strategies in batch diafiltration Name Constant-volume dilution

a - strategy a¼1

Traditional diafiltration

a ¼ f0, 1, 0g

Variable-volume diafiltration

a ¼ const, 0 < a < 1

Preconcentration with variable-volume diafiltration Sequential dilution diafiltration

a ¼ f0, a1 g, a1 ¼ const, 0 < a1 < 1 a ¼ ð0, amax Þn , 0 a ¼ ðamax , 0Þn

Dynamic-volume diafiltration

a ¼ aðtÞ 0 aðtÞ amax

Intermittent feed diafiltration

n number of repetition

References Baker R (2004) Membrane technology and applications. Wiley, Chichester Foley G (2006) Ultrafiltration with variable volume diafiltration: a novel approach to water saving in diafiltration processes. Desalination 199(1–3):220–221

Paulen R, Fikar M, Foley G, Kovács Z, Czermak P (2012) Optimal feeding strategy of diafiltration buffer in batch membrane processes. J Membr Sci 411–412:160–172 van Reis R, Goodrich EM, Yson CL, Frautschy LN, Whiteley R, Zydney AL (1997) Constant Cwall ultrafiltration process control. J Membr Sci 130(1–2):123–140

B

Beer Clarification Frank Lipnizki Alfa Laval Copenhagen, Søborg, Copenhagen, Denmark

In the traditional brewing process, the beer is clarified after fermentation and before maturation to remove mainly the remaining yeast but also microorganisms and haze. The conventional process for beer clarification is the combination of a high-speed separator followed by diatomaceous earth (DE)/kieselguhr filtration which is complemented in some cases by second filtration with PVP (polyvinylpyrrolidone) to obtain a very clear beer. The major challenge of the conventional process is related to the DE filtration because the DE can vary in quality and is problematic for handling and disposal since it is hazardous plus DE filtration generates large amounts of effluent. Alternatively, from its introduction in 2001 (Buttrick 2007) cross-flow microfiltration has established itself in the brewing industry with over 50 breweries worldwide adopting this DE-free beer filtration. Generally, three concepts are currently used in the industry: 1. Beer Membrane Filtration (BMF) by Pentair (previously Norit) – a system based on hollow fiber microfiltration cartridges without a highspeed separator as pretreatment # Springer-Verlag Berlin Heidelberg 2014 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_46-1

2. PROFi Membrane System by Pall and GEA Westfalia – a combination of hollow fiber microfiltration system with a high-speed separator as pretreatment 3. AlfaBright system by Alfa Laval and Sartorius – a process based on the combination of cassette microfiltration system with a highspeed separator as pretreatment The membranes established for beer filtration are all polymeric microfiltration membranes based on polyethersulfone with 0.5–0.65 mm pores and beer capacities of 0.5 1.0 hl/(m2 h) (Buttrick 2007). The resulting beer quality from the membranes is similar or improved compared to the DE filtration; in Table 1 a comparison between beer filtered by DE and microfiltration membranes is given. Other membranes, e.g., ceramic membranes, have been tested for beer filtration but so far have not established themselves on the market. The key difference between the three concepts is the use of a high-speed separator as pretreatment. In the BMF system a retentate tank is used to collect the beer solids, while both the PROFi and the AlfaBright systems are using high-speed separators as pretreatment before the membrane to remove the beer solids and thus eliminating the need for a retentate tank. The systems can be typically run in batch or continuous operation depending on the size and requirements of the individual brewery. In batch operation, the complete system shifts from filtration into cleaning mode after each batch. Plants

2

Beer Clarification

Beer Clarification, Table 1 Comparison of key parameters of beer filtered with DE filtration and microfiltration (Lipnizki 2005) Original extract [%] Alcohol [%] Color [EBC] Viscosity [MPas] Turbidity at 0 C [EBC]

Beer before filtration 11.40 3.84 7.20 1.62 32.00

with continuous operation are arranged in skids/ blocks and use sequential cleaning which allows for some skids/blocks to be in standby cleaning mode, while other blocks are in filtration mode. The major challenges for cross-flow membrane beer filtration are the relatively high investment costs and complexity of the process when compared to DE filtration and the increasing availability of alternative DE-free filter aids.

Beer after DE filtration 11.37 3.83 6.70 1.57 0.53

Beer after microfiltration 11.39 3.84 7.00 1.56 0.41

References Buttrick P (2007) Filtration – the facts. Brewer Distiller Int 3(12):12–19 Lipnizki F (2005) Optimisation and integration of membrane processes in the beverage industry, 10th edn. Aachener Membran Kolloquium, Aachen, 16.17.03.2005

B

Beer Dealcoholization Frank Lipnizki Alfa Laval Copenhagen, Søborg, Copenhagen, Denmark

In the last decades the demand for low alcohol and alcohol-free drinks increased, e.g., in Germany the annual consumption of alcohol-free drinks nearly doubled from 130.4 l per person in 1980 to 248.2 l person in 1999, while the annual consumption of alcoholic drinks declined from 179.5 to 156.3 l during the same period (Gebhardt 2001). Conventionally beer can be dealcoholized by distillation, but additionally the membrane process reverse osmosis and dialysis have established themselves for the partial dealcoholization of beer by eight to ten times. The key advantage of membrane processes over distillation is that beer can be dealcoholized at low temperatures typically 78 C to minimize the effect of temperature on the beer flavor resulting in high-quality low alcohol beer, which can be bottled after final sterile filtration. Reverse osmosis is typically carried out in spiral wound modules, and the dealcoholization process based on reverse osmosis can be divided into four individual operations which are typically carried out in batch mode: 1. Pre-concentration: In this step the volume of the feed beer is reduced. The beer is passed # Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_47-1

through the membrane modules and is then recycled to the batch tank. The permeate – water and alcohol – is removed from the process, while retentate, concentrated beer and flavors, is returned to the batch tank. 2. Diafiltration: This step is similar to the pre-concentration step but diafiltration water – desalted and deoxygenized water – is added to wash out the alcohol. The amount of diafiltration water added balances the amount of permeate removed from the process, and thus, the level in the batch tank remains constant. This operation is continued until the desired alcohol concentrations in the beer are achieved. 3. Alcohol adjustment: In this step, the taste and alcohol content is fine-tuned by addition of desalted and deoxygenized water. 4. Posttreatment: In order to give the beer its specific character and to balance taste losses due to removal of the taste carrier alcohol, the CO2 levels can be adjusted, and hop extract, syrup, or other flavor enhancers are added. As an alternative to reverse osmosis, dialysis can be used for the dealcoholization of beer. Commonly, hollow fiber modules are used for dialysis allowing the beer to flow on one side of the membrane and water as dialysate on the other side. The process is normally operated in countercurrent flow to maximize the concentration gradient over the dialysis membrane and thus the

2

driving force of the dialysis process. The dialysate is constantly recycled over a steam stripping column to remove the alcohols, thus maintaining the driving force of the process while minimizing the dialysate consumption. In addition, in order to minimize CO2 losses, the feed pressure should be selected close to the CO2 saturation pressure, and small amounts of the CO2 should be added to the dialysate (Branyik et al. 2012). Furthermore, osmotic distillation for beer dealcoholization (Russo et al. 2013) and pervaporation for aroma recovery can be considered as part of the beer dealcoholization process (del Olmo et al. 2012), but so far these processes are not commercialized.

Beer Dealcoholization

References Branyik T, Silva DP, Baszczynski M, Lehnert R, Almeida e Silva JB (2012) A review of methods of low alcohol and alcohol-free beer production. J Food Eng 2012(108):493–506 del Olmo A, Blanco CA, Palacio L, Prádanos P, Hernández A (2012) Setting up of a method of pervaporation for improving alcohol-free beer. Euromembrane, London, pp 23–27 Gebhardt W (2001) Weltforum der Wein- und Saftbereitung. F&S Filtrieren Separieren 15(5):229 Russo P, Liguori L, Albanese D, Crescitelli A, Di Matteo M (2013) Investigation of osmotic distillation technique for beer dealcoholization. CEt Chem Eng Trans 32:1735–1840

B

Beer Maturation Frank Lipnizki Alfa Laval Copenhagen, Søborg, Copenhagen, Denmark

In the beer production process, the clarified and cooled wort from the brewhouse is transferred together with yeast to the fermentation tanks for the primary fermentation which converts the fermentable sugar into alcohol and CO2. The resulting “green beer” undergoes a second fermentation – beer maturation – under addition of sugar or fresh yeasted wort. During the maturation, the residual fermentable sugars in the “green beer” will be converted to alcohol and the beer will be saturated with CO2. After the fermentation, the beer is clarified and stored in the bright beer cellar. Remaining in the fermentation tanks are tank bottoms – a mixture of settled yeast cells and beer – which are equal approx. 1.5–2.0 % of the fermentation tank volume. In the past this beer in the tank bottom would be considered lost since it could not be added to the main beer stream. However, using microfiltration with tubular ceramic membranes or plate-and-frame modules with polymeric

# Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_48-2

membranes, it is possible to recover a highquality beer which can be blended with the main beer stream going toward clarification and storage. Using microfiltration membranes with pores of 0.4–0.5 mm, it is possible to retain the yeast and allow the beer to pass the membrane without any major impact on the quality of the beer. In this process, the yeast in the tank bottoms is concentrated from 7 % to 10 % DM to approx. 20 % DM and thus 50–70 % of the beer in the tank bottoms can be recovered. The yield of recovered extract and alcohols can be further maximized if diafiltration water is added. Recovering beer from tank bottoms can increase the output of an average brewery by 1 % of its annual production or 24,000 hl extra for a brewery with an annual output of 2 million hl (Lipnizki 2005). The amortization for a microfiltration beer recovery unit is typically 1–2 years.

Reference Lipnizki F (2005) Optimisation and integration of membrane processes in the beverage industry, 10. Aachener Membran Kolloquium, 16.-17.03.2005: preprints. Aachen, Mainz, 2005. - 3-86130-409-0.

B

Benzene and Cyclohexane Separation Tadashi Uragami Kansai University, Organization for Research and Development of Innovative Science and Technology (ORDIST), Suita, Osaka, Japan

Benzene/Cyclohexane Separation In the petrochemical industry, the separation of benzene (Bz) and cyclohexane (Chx) is the most important and difficult processes. Chx is produced in benzene hydrogenation units under Ni or Pd catalyst. The unreacted Bz is remained in the reaction mixture and must be removed to produce pure Chx. The separation of benzene and Chx is very difficult by a conventional distillation because close-boiling point mixtures are formed in the entire range of their compositions. At present, azeotropic distillation and extractive distillation are applied to this separation. These distillations, however, are complex and need high energy consumption. In the industry of Chx production, the conventional Bz/Chx separation processes are strongly required. Therefore, many studies have investigated the PV properties of polymer membranes for Bz/Chx separation. Pervaporation (PV) is a promising membrane technique for the separation of organic/organic mixtures, as PV can be used to separate organic liquid mixtures such as azeotropic and close# Springer-Verlag Berlin Heidelberg 2014 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_49-1

boiling point mixtures. The separation mechanism in PV is not based on only relative volatility of components in distillation but on the difference in sorption and diffusion properties of the feed substances. Figure 1 illustrates the principles of PV. In this separation process, when a liquid mixture is fed to the upstream side of a polymer membrane and the downstream side is evacuated, a component in the feed mixture can preferentially permeate through the membrane. In a PV process, differences between the solubility and diffusivity of the mixture components in the polymer membrane and the relative volatility of the permeants determine the permeability and selectivity (Binding et al. 1961; Aptel et al. 1974). In general, PV exhibits the following characteristics (Uragami 2006, 2010): 1. Selective transport across the nonporous membrane is achieved by a three-step process of solution, diffusion, and evaporation. 2. Because the driving force for permeation is the vapor pressure for each component rather than total system pressure, this method is effective for separation of organic liquid mixtures with high osmotic pressure. 3. PV can be applied to the separation and concentration of mixtures that are difficult to separate by distillation. For example, it is useful for the separations of azeotropic mixtures, close-boiling point mixtures, and structural isomers.

2

Benzene and Cyclohexane Fig. 1 Principle of pervaporation (PV)

Benzene and Cyclohexane Separation

Separation,

4. PV can be used for the removal of certain components in equilibrium reactions. 5. Polymer membrane compaction, a frequent problem in high-pressure gas separations, is not encountered in PV because the feed pressure is typically low. A side-chain liquid-crystalline polymer (LCP) was synthesized by the addition of mesogenic monomer to poly(methylsiloxane) with a Pt catalyst. When Bz/Chx mixtures were permeated through the LCP membranes by PV at various temperatures, the permeation rate increased with increasing benzene concentration in the feed solution and permeation temperature. Although the LCP membranes exhibited Bz/Chx selectivity, the mechanism responsible for the permeation and separation of the Bz/Chx mixtures was different in the glassy, liquid-crystalline state versus the isotropic state of the LCP membranes. These results suggest that the Bz/Chx selectivity was moderately influenced by the change in LCP membrane structure (i.e., a state transformation). The balance between the orientation of the mesogenic groups and the flexibility of the siloxane chain is very important with respect to permeability and Bz/Chx selectivity (Inui et al. 1997, 1998). When benzene/cyclohexane, toluene/cyclohexane, and o-xylene/cyclohexane mixtures were subjected to PV through an LCP membrane in the liquid-crystalline state, the permeation rate increased with increasing

temperature and the LCP membrane exhibited selectivity for the aromatic hydrocarbons. The permeation rate and selectivity of the LCP membrane for each mixture decreased with increasing molecular size of the aromatic hydrocarbon in the binary feed mixture (Inui et al. 1998). When Bz/Chx mixtures were permeated through nematic and smectic side-chain liquid-crystalline polymer (n- and s-LCP) membranes under various conditions during PV, the n- and s-LCP membranes exhibited Bz/Chx selectivity. The selectivity of the n-LCP membrane changed from solubility-selectivity controlled to diffusion-selectivity controlled upon the state transformation of the membrane, induced by an increase in the permeation temperature. In contrast, the selectivity of the s-LCP membrane was governed by diffusion selectivity regardless of the state of this membrane. At low permeation temperatures, the n-LCP membrane in the liquidcrystalline state exhibited lower permeability but higher selectivity than the s-LCP membrane. However, at high permeation temperatures, the relationship between the permeability and Bz/Chx selectivity of the n-LCP and s-LCP membranes in the liquid-crystalline state was reversed. These results are a result of differences in the chemical and physical structure of the n-LCP and s-LCP membranes (Inui et al. 1998). The PV properties of a series of cross-linked 4,40 -hexafluoro-isopropylidene dianhydride (6FDA)-based copolyimide membranes for the separation of Bz/Chx mixtures were investigated (Ren et al. 2001). The glassy, highly rigid copolyimides were obtained by polycondensation of 6FDA with various diamines. To obtain high permeability as well as high selectivity, a combination of the diamines 2,3,5,6-tetramethyl1,4-phenylene diamine (4MPD), 4,40 hexafluoroisopropylidiene dianiline (6FpDA), and 3,5-diaminobenzoic acid (DABA) as monomers with a crosslinkable group was used. Crosslinking is necessary to prevent undesirable swelling effects, which generally occur with non-cross-linked polyimides, especially if high benzene concentrations are reached during PV. The degree of cross-linking was kept constant at 20 %, whereas the ratio of the diamine

Benzene and Cyclohexane Separation

monomers 6FpDA and 4MPD was varied. The PV experiments were performed at 60 C, using Bz/Chx mixtures with benzene concentrations covering the entire concentration range. All of the cross-linked polymers had excellent chemical and thermal stability in the PV experiments. In all cases, conditioning of the membrane samples with pure benzene was a suitable pretreatment to enhance the permeation rate without decreasing the Bz/Chx selectivity significantly. For the most promising membrane material, 6FDA4MPD/DABA of 4:1 cross-linked with ethylene glycol, the PV experiments were performed with a benzene/cyclohexane feed mixture of 50/50 (w/w) over a temperature range between 60 C and 110 C to determine the effect of temperature on the separation characteristics.

References Aptel P, Cuny J, Jozenfonvice J, Morel G, Neel J (1974) Liquid transport through membranes prepared by grafting of polar monomer onto poly(tetrafluoroethylene) films. II. Some factors determining pervaporation rate and selectivity. J Appl Polym Sci 18:365

3 Binding RC, Lee RJ, Jennings JF, Mertic EC (1961) Separation of liquid mixtures by pervaporation. Ind Eng Chem 53:47 Inui K, Miyata T, Uragami T (1997) Permeation and separation of benzene/cyclohexane mixtures through liquid-crystalline polymer membranes. J Polym Sci Part B Polym Phys 35:699 Inui K, Miyata T, Uragami T (1998a) Effect of permeation temperature on permeation and separation of a benzene/cyclohexane mixture through liquid-crystalline polymer membranes. J Polym Sci Part B Polym Phys 36:281 Inui K, Miyata T, Uragami T (1998b) Permeation and separation of binary organic mixtures through a liquid-crystalline polymer membrane. Macromol Chem Phys 199:589 Inui K, Okazaki K, Miyata T, Uragami T (1998c) Effect of mesogenic groups on characteristics of permeation and separation for benzene/cyclohexane mixtures of sidechain liquid-crystalline polymer membranes. J Membr Sci 143:93 Ren J, Standt-Bickel C, Lichtenthaler R (2001) Separation of aromatics/aliphatics with crosslinked 6FDA-based copolyimides. Sep Purif Technol 22–3:31 Uragami T (2006) Polymer membranes for separation of organic liquid mixtures. In: Yanpolskii Y, Pinnau I, Freeman BD (eds) Materials science of membranes for gas and vapor separation. Wiley, Chichester, pp 355–372 Uragami T (2010) Selective membranes for purification and separation of organic liquid mixtures. In: Drioli E, Georno L (eds) Comprehensive membrane science and engineering, volume 2 membrane operations in molecular separations. Elsevier, Amsterdam, pp 273–324

B

Biodegradable Membrane Isabel Coelhoso1, Filomena Freitas2, Vitor D. Alves3 and Maria A. M. Reis2 1 LAQV- REQUIMTE, Departamento de Quı´mica, Faculdade de Ciencias e Tecnologia, Universidade Nova de Lisboa, Caparica, Portugal 2 UCIBIO-REQUIMTE, Departamento de Quı´mica, Faculdade de Cieˆncias e Tecnologia, Universidade Nova de Lisboa, Caparica, Portugal 3 LEAF – Linking Landscape, Environment, Agriculture and Food, Instituto Superior de Agronomia, Universidade de Lisboa, Lisboa, Portugal

Biodegradable membranes consist of thin-filmlike structures usually applied as separating selective barriers and support devices. These structures are generally composed of biodegradable polymers, complemented with additives, such as plasticizers, emulsifiers, and crosslinking agents. To be considered biodegradable, all membrane components must be degraded by the action of microorganisms and converted into water, carbon dioxide and/or methane, and new cell biomass. The wide range of polymers used in the development of biodegradable membranes enables the production of structures with quite diverse properties, finding applications in different areas. It is interesting to note that some of them, beyond being biodegradable, are also biocompatible # Springer-Verlag Berlin Heidelberg 2012 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_51-2

and/or edible, extending their use to the fields of biomedicine and edible coatings for food products. Biodegradable polymers obtained by chemical synthesis have been used, for example, in food packaging materials (Siracusa et al. 2008) and medical applications (e.g., fracture fixation, dental orthopedic implants, artificial skin, suture anchors, drug delivery) (Yen et al. 2009; Armentano et al. 2010; Bettahalli et al. 2011). They have also been applied for separation of organic mixtures by pervaporation (Zereshki et al. 2010, 2011). They include the following: (i) Polyglycolic acid (PGA), an aliphatic–aromatic copolymer, which combines the excellent material properties of aromatic polyethylene terephthalate and the biodegradability of aliphatic polyesters. It is produced by a polycondensation reaction of glycol and aliphatic dicarboxylic acids, which may be obtained from renewable resources. (ii) Polylactic acid (PLA), a thermoplastic aliphatic polyester obtained from polymerization of the lactic acid monomer produced by microbial fermentation. (iii) Polycaprolactone (PCL), a thermoplastic polymer obtained by chemical synthesis using nonrenewable resources (petrochemical derivatives).

2

Among the biodegradable polymers produced by microbial fermentation, polysaccharides (e.g., pullulan, hyaluronan, gellan, GalactoPol, curdlan, bacterial alginate, bacterial cellulose) (Freitas et al. 2011) and polyesters (e.g., polyhydroxyalkanoates) have been applied in films and edible coatings (e.g., pullulan, gellan) (Nieto 2009; Alves et al. 2011). Both microbial polysaccharides and polyesters show a wide range of properties that may be tuned by manipulating the bioreaction conditions. Polyhydroxyalkanoates show a wide range of applications, such as in industry (e.g., packaging, waterproof paperboard), medicine (e.g., bone plates, osteosynthetic materials, surgical sutures, and dressing materials for surgery), and agriculture (e.g., mulch films) (Philip et al. 2007). The polymers recovered from natural products generally used to produce biodegradable membranes include polysaccharides (e.g., starch, cellulose, pectin, alginate, carrageenan, chitosan) and proteins (e.g., gelatin/collagen, soy protein, gluten). These polymers are widely used to develop edible coatings for food products and biodegradable films intended for food packaging (Nieto 2009). Chitosan, alginate, and collagen are also referred to be applied in tissue engineering (Eisenbarth 2007). Polysaccharide-based membranes, such as chitosan and sodium alginate, have received much attention for solvent dehydration by pervaporation, due to their good film-forming properties, chemical resistance, and high permselectivity for water (Chapman et al. 2008; Yu et al. 2006).

References Alves VD, Ferreira AR, Costa N, Freitas F, Reis MAM, Coelhoso IM (2011) Characterization of

Biodegradable Membrane biodegradable films from the extracellular polysaccharide produced by Pseudomonas oleovorans grown on glycerol byproduct. Carbohydr Polym 83:1582–1590 Armentano I, Dottori M, Fortunati E, Mattioli S, Kenny JM (2010) Biodegradable polymer matrix nanocomposites for tissue engineering: a review. Polym Degrad Stab 95:2126–2146 Bettahalli N, Steg H, Wessling M, Stamatialis D (2011) Development of poly(L-lactic acid) hollow fiber membranes for artificial vasculature in tissue engineering scaffolds. J Membr Sci 371:117–126 Chapman PD, Oliveira T, Livingston AG, Li K (2008) Membranes for dehydration of solvents by pervaporation. J Membr Sci 318:5–37 Eisenbart E (2007) Biomaterials for tissue engineering. Adv Eng Mat 9:1051–1060 Freitas F, Alves VD, Reis MAM (2011) Advances in bacterial exopolysaccharides: from production to biotechnological applications. Trends Biotechnol 29:388–398 Nieto MB (2009) Structure and function of polysaccharide gum-based edible films and coatings. In: Embuscado ME, Huber KC (eds) Edible films and coatings for food applications. Springer, New York, pp 57–112 Philip S, Keshavarz T, Roy I (2007) Polyhydroxyalkanoates: biodegradable polymers with a range of applications. J Chem Technol Biotechnol 82:233–247 Siracusa V, Rocculi P, Romani S, Rosa MD (2008) Biodegradable polymers for food packaging: a review. Trends Food Sci Technol 19:634–643 Yen C, He H, Lee LJ, Winston Ho WS (2009) Synthesis and characterization of nanoporous polycaprolactone membranes via thermally- and nonsolvent-induced phase separations for biomedical device application. J Membr Sci 343:180–188 Yu L, Dean K, Li L (2006) Polymer blends and composites from renewable resources. Prog Polym Sci 31:576–602 Zereshki S, Figoli A, Madaeni SS, Simone S, Jansen JC, Esmailinezhad M, Drioli E (2010) Poly(lactic acid)/ poly(vinyl pyrrolidone) blend membranes: effect of membrane composition on pervaporation separation of ethanol/cyclohexane mixture. J Membr Sci 362:105–112 Zereshki S, Figoli A, Madaeni SS, Galiano F, Esmailinezhad M, Drioli E (2011) Pervaporation separation of ethanol/ETBE mixture using poly(lactic acid)/poly(vinyl pyrrolidone) blend membranes. J Membr Sci 373:29–35

B

Biodegradable Organic Matter Isabel Coelhoso1, Filomena Freitas2, Vitor D. Alves3 and Maria A. M. Reis2 1 LAQV-REQUIMTE, Departamento de Química, Faculdade de Ciencias e Tecnologia, Universidade Nova de Lisboa, Caparica, Portugal 2 UCIBIO-REQUIMTE, Departamento de Química, Faculdade de Cieˆncias e Tecnologia, Universidade Nova de Lisboa, Caparica, Portugal 3 LEAF – Linking Landscape, Environment, Agriculture and Food, Instituto Superior de Agronomia, Universidade de Lisboa, Lisboa, Portugal

Biodegradable organic matter is organic material, plant, and animal matter with origin in living organisms, which can be converted by the action of microorganisms to water, carbon dioxide, and/or methane and biomass. Organic materials can be used to obtain biodegradable polymers which are classified according to the method of production or their source (Fig. 1): – Polymers directly extracted or removed from biomass such as polysaccharides and proteins – Polymers produced by classical chemical synthesis starting from renewable bio-based monomers such as polylactic acid (PLA) – Polymers produced by microorganisms or genetically modified bacteria such as # Springer-Verlag Berlin Heidelberg 2012 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_52-1

polyhydroxyalkanoates, bacterial cellulose, xanthan, and pullulan (Mensitier et al. 2011) Polysaccharides are the most abundant macromolecules in the biosphere. These complex carbohydrates constituted of monosaccharides joined together by glycosidic bonds are often one of the main structural elements of plant and animal exoskeleton (e.g., cellulose, carrageenan, chitin) or have a key role in the plant energy storage (e.g., starch). Cellulose and starch are of prime interest as biopolymers because of their availability and rather low cost. A variety of polysaccharides and their derivatives, besides starch and cellulose derivatives, have been used as biodegradable membrane-forming matrixes, including alginate, pectin, carrageenan, chitin, and various gums. Several protein sources have been proposed for the preparation of biopolymers, in particular, cereal proteins which are available in large amounts as by-products arising from agricultural and biofuel processing activities such as ethanol production. These protein-rich products include spent grain from the brewing and distilling industries, cereal bran streams from milling, and protein residues from starch extraction activities (Mensitier et al. 2011). Microbial biopolymers are naturally synthesized by microorganisms with different functions in the microbial cell, including intracellular carbon or energy storage reserves (e.g., glycogen,

2

Biodegradable Organic Matter

Biodegradable Polymers

Biopolymers (renewable resources)

Extracted from Biomass

Biomonomer Synthesis

Polymers (fossil resources)

Produced by Microorganisms

PCL PGA

PHA

PLA Proteins

PBSA

Polysaccharides Pullulan Whey protein

Pectin

Casein

Carrageenan

Soy protein

Starch

Gluten

Cellulose

Glucan

Biodegradable Organic Matter, Fig. 1 Biodegradable polymers

polyesters), structural cell wall components (e.g., chitin, b-glucans), and extracellular biopolymers (e.g., exopolysaccharides), often secreted as protective mechanisms in response to environmental conditions (Rehm 2010). A wide range of agro-food and industrial wastes/by-products have been proposed as alternative substrates for the production of microbial biopolymers, including molasses, cheese whey, palm date syrup, olive mill wastewater (OMW), glycerol by-product from the biodiesel industry, corn-steep liquor, spent malt grains, apple and grape pomaces, citrus peels, peach pulp, used oils, and several acid-hydrolysate wastes (e.g., melon, watermelon, cucumber, tomato, rice), among others (Freitas et al. 2011; Verlinden et al. 2011). The microbial biopolymers produced include polysaccharides, polyamides, polyesters, and polyanhydrides. Depending on their composition and molecular weight, they have properties

ranging from rheology modifiers to bioplastics, which makes them useful in many industrial applications (e.g., agro-food, cosmetics, pharmaceutical, textile) (Rehm 2010).

References Freitas F, Alves VD, Reis MAM (2011) Advances in bacterial exopolysaccharides: from production to biotechnological applications. Trends Biotechnol 29(8):388–398 Mensitieri G, Di Maio E, Buonocore G, Nedi I, Oliviero M, Sansone L, Iannace S (2011) Processing and shelf life issues of selected food packaging materials and structures from renewable resources. Trends Food Sci Technol 22:72–80 Rehm BHA (2010) Bacterial polymers: biosynthesis, modifications and applications. Nat Rev Microbiol 8:578–592 Verlinden RAJ, Hill DJ, Kenward MA, Williams CD, Piotrowska-Seget Z, Radecka IK (2011) Production of polyhydroxyalkanoates from waste frying oil by Cupriavidus necator. AMB Express 2011:1–11

B

Biogas Ivo Vankelecom1 and Muhammad Waqas Anjum2 1 Centre for Surface Chemistry and Catalysis, KU Leuven, Heverlee, Leuven, Belgium 2 School of Chemical & Materials Engineering, National University of Science & Technology, Islamabad, Pakistan

Fossil fuels are still the primary source of energy by preference. However, as fossil fuels become more and more expensive with the possibility of depletion of resources, the quest for alternate sources of energy is gaining attention. In this whole situation, anaerobic digestion of biological resources and biological waste could be a promising alternative energy carrier. Natural gas normally consists of 90–95 % methane, but in biogas this composition is reduced to 50–65 % making it a low-grade natural gas which is the product of neutral decomposition of organic substance of animal or plant origin due to anaerobic bacterial activity. The plants used for biogas production are normally

# Springer-Verlag Berlin Heidelberg 2013 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_54-1

referred to as anaerobic digesters or anaerobic fermenters. The resources used for biogas production include kitchen waste, dry poultry droppings and animal excrements, remnants of food processing, and slaughterhouse leftovers. Four ingredients needed for biogas production are organic matter, bacteria, anaerobic conditions, and heat. In a controlled reaction system, the gaseous mixture thus produced can contain up to 70 % of biohydrogen and biomethane, respectively, that can be used for commercial applications (Harold 2007). The anaerobic digestion process can be classified into different sets of complex reactions, as shown in Fig. 1. Production of biohydrogen and biomethane from organic wastes consists of mainly three steps including hydrolysis, acetogenesis, and methanogenesis. Organic substrates can be converted to biogas by a diverse group of microbes using multienzyme (cellulases, amylases, lipases, proteases, etc.) systems. Organic material is fed into digesters after grinding to an appropriate size. In the digesters, these substrates are heated and agitated leading to

2

Biogas

Biogas, Fig. 1 Schematic diagram of the anaerobic digestion process for biogas generation (Modified from Basu et al. 2010)

Organic matter (OM)

Hydrolysis (Soluble OM)

Fermentation (Fatty acids)

Acetate

Methanogenesis

H2 + C2

CH4 + CO2

production of biogas which is collected in a biogas container. This gas is fed into an electric generator which produces electricity and heat. Biohydrogen generated in fermentation processes (e.g., anaerobic fermentation, photofermentation, dark fermentation) has hydrogen and carbon dioxide as major ingredients. On the other hand, biomethane produced by anaerobic digestion of biological wastes has 38–40 % carbon dioxide with smaller amounts of hydrogen sulfide along with trace amounts (ppm) of hydrogen, nitrogen, oxygen, and volatiles with 55–60 % methane as a major part (Rasi et al. 2007). Biogas has a calorific value of 35–44 kJ g 1, which is comparatively higher than other energy resources like petrol, diesel, or LPG and solid fuels like coal, charcoal wood, etc. Biogas is a potential source of environmentally benign, clean, and

cheap alternative energy. However, the presence of incombustible and acid gases, like CO2, not only reduces its calorific value, but their corrosive nature restricts its transportation. One of the many trace components includes silicone containing compounds. Commonly occurring siloxanes in biogas are known as volatile methyl siloxanes (VMS) that include cyclic tri-, tetra-, and pentasiloxane, as well as linear di-, tri-, tetrasiloxane. After combustion, undesirable microcrystalline quartz and pentasiloxane are produced as they cause the wear & tear of engines and turbines. Purified biogas can be used as a feed to fuel cells or for domestic applications and power generation. Figure 2 represents the possible applications of biogas (Basu et al. 2010).

Biogas

3

Hydrogen production

Fuel cell

Bio Gas

Gasification Combined Cycle

ELectricity Generation

Thermal power

Anaerobic digesstion/ Hydro gasifiction

Methane

Gas engines/Fuel cell

Biogas, Fig. 2 Possible applications of biogas as energy resource (Modified from Basu et al. 2010)

References S. Basu, A.L. Khan, A. Cano-Odena, C. Liu, I.F.J. Vankelecom, Membrane-based technologies for biogas separations. Chem Soc Rev 39, 750–768 (2010) Harold House (2007) Proceeding for London Swine conference-today’s challenges. Tomorrow’s

opportunities 3–4. http://www.londonswineconference. ca/proceedings/2007/LSC2007_HHouse.pdf. Accessed 29 Jan 2013. S. Rasi, A. Veijanen, J. Rintala, Trace compounds of biogas from different biogas production plants. Energy 32, 1375–1380 (2007)

B

Biogas Recovery Ivo Vankelecom Centre for Surface Chemistry and Catalysis, KU Leuven, Leuven (Heverlee), Belgium

The constant increase in energy demand, coupled with the depletion of fossil fuels, is a concern that cannot be ignored in the industrial and highly populated world of today (Basu et al. 2010). Biogas production is a very promising source of renewable energy that still offers further exploitation of its potential (Scholz et al. 2013). Essentially, microbially controlled biogas production is an already existing part of the global carbon cycle, releasing an estimated 590–800 million tons of methane to the atmosphere (Bond and Templeton 2011). Current biogas recovery systems seek to exploit these processes in order to produce energy from sewage wastewater, animal manure, crop straws, or mixed agricultural wastes (Chen et al. 2012; Rasi et al. 2011). Usually, a system for biogas recovery consists of four components: a collection system which helps transfer the biogas source to the anaerobic digester, the anaerobic digester in which the methane production takes place, a biogas collection system providing piping of biogas to a combustion device, and a gas use device where biogas is combusted to produce heat or electricity (Agstar 2011). Alternatively, the produced biogas can be used on site as fuel for automotives or can be injected # Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_55-3

in the natural gas grid. In this case, the gas must be upgraded to natural gas standards, namely, 98 % methane content (AEBIOM 2009). Although the composition of biogas varies significantly depending on many factors such as the type of digester, the average values are reported to range between 50 % and 70 % methane and 30–50 % carbon dioxide, as well as hydrogen sulfide, sulfur compounds, siloxanes, and aromatic and halogenated compounds (Rasi et al. 2007). Upgrading (purification) of biogas is highly beneficial in terms of increasing the amount of methane per unit volume of biogas, which equals an increase in its calorific value. In addition to the obvious need for removal of carbon dioxide as the major pollutant, the trace compounds have the potential to trigger ozone depletion and the greenhouse effect, reduce the quality of local air due to formation of volatile organic compounds (VOCs), and so on. Moreover, sulfur compounds corrode pipelines and combustion engines, while silicon compounds oxidize and are deposited on engine parts (Rasi et al. 2007). Currently, biogas upgrading is largely achieved through amine scrubbing where carbon dioxide is absorbed into an amine solution, water scrubbing where carbon dioxide is absorbed to water at elevated pressures, and pressure swing adsorption where pressure of the gas mixture is changed to induce adsorption and desorption of one gas species. These technologies are in general more energy intensive and polluting than membrane technology (Basu et al. 2010;

2

AEBIOM 2009; Scholz et al. 2013). Polymeric membranes have gained a share in the separations market, as opposed to the inorganic membranes which are expensive and brittle. Some commercial polymeric membranes such as SEPAREX (cellulose acetate) of UOP and polydimethyl siloxane (PDMS) have already proven or were reported to operate successfully in separation of carbon dioxide or siloxanes and VOCs, respectively. However, the separation of hydrogen sulfide still remains a difficult and interesting issue for membrane research. Mixed matrix membranes composed of inorganic particles distributed throughout a polymeric matrix are promising alternatives thoroughly investigated for these separations (Basu et al. 2010).

References Agstar-EPA (2011). Market opportunities for biogas recovery systems at U.S. Livestock Facilities. http://

Biogas Recovery epa.gov/agstar/documents/biogas_recovery_systems_ screenres.pdf. Accessed 31 Jan 2013 Basu S, Khan AL, Cano-Odena A, Lui C, Vankelecom IFJ (2010) Membrane-based technologies for biogas separations. Chem Soc Rev 39(2):750–768 Bond T, Templeton MR (2011) History and future of domestic biogas plants in the developing world. Energy Sustain Dev 15(4):347–354 Chen L, Lixin Z, Changshan R, Fei W (2012) The progress and prospects of rural biogas production in China. Energy Policy 51:58–63 European Biogas Association-AEBIOM (2009). A biogas road map for Europe. http://www.aebiom.org/IMG/ pdf/Brochure_BiogasRoadmap_WEB.pdf. Accessed 31 Jan 2013 Rasi S, Veijanen A, Rintala J (2007) Trace compounds of biogas from different biogas production plants. Energy 32(8):1375–1380 Rasi S, La¨ntela¨ J, Rintala J (2011) Trace compounds affecting biogas energy utilization – a review. Energy Convers Manag 52(12):3369–3375 Scholz M, Melin T, Wessling M (2013) Transforming biogas into biomethane using membrane technology. Renew Sust Energ Rev 17:199–212

B

Biohybrid Artificial Liver (BAL) Sabrina Morelli National Research Council of Italy, Institute on Membrane Technology, ITM-CNR, Rende, CS, Italy

A biohybrid artificial liver (BAL) is a bioartificial device which consists of functional liver cells supported by an artificial cell culture material. It incorporates hepatocytes into a bioreactor in which the cells are immobilized, cultured, and induced to perform the hepatic functions by processing the blood or plasma of liver failure patients. BAL provides temporary support for patients waiting for an allogeneic liver transplant, and since the liver can regenerate, the temporary support provided by BAL may allow time for liver regeneration. The bioreactor is an important component of BAL, because it determines the viability and function of the hepatocytes within it. A successful and clinically effective bioreactor should mimic the structure of the liver and provide an in vivo-like microenvironment for the growth of hepatocytes, thereby maintaining the cells’ viability and function to the maximum extent. The important issues are the choice of cell sources and the design of the bioreactor (Ding and Shi 2011). The cell sources provide liver-specific functions, such as detoxification, drug metabolism, and protein synthesis, while the bioreactors maintain the # Springer-Verlag Berlin Heidelberg 2014 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_56-3

viability and function of cells. More efforts are now underway in search for the best cell resource and best design of bioreactors. Considering the several functions that the liver performs, the bioreactor for BAL devices has to ensure the rapid detoxification of neural and hepatic toxins, the return of liver-specific hepatotrophic factors, as well as liver-specific coagulation factors, back into patient’s blood, and the maintenance of liver cell detoxification and synthetic functions until liver tissue regeneration or organ transplantation. One of the most promising bioreactors is the membrane bioreactor. Polymeric membranes in flat and hollow fiber configuration with different morphology and chemical–physical properties have been used in BAL devices (De Bartolo and Bader 2001; Kamlot et al., 1996; Kasuya and Tanishita 2012). Most of the extracorporeal BALs have not only used cellulose and polysulfone derivatives but also native and modified polypropylene membranes. Morphological (e.g., pore size, pore size distribution, and roughness) and physicochemical membrane properties (e.g., surface charge, wettability, and surface free energy) affect all the adhesion and metabolic functions of hepatocytes. Hepatocytes have been cultured in membrane bioreactors in different configurations: between flat sheet membranes in a sandwich configuration; in the lumen of hollow fiber membranes entrapped in a collagen layer; in the shell of hollow fiber membranes in monolayer,

2

aggregate, or spheroid structure and attached to microcarriers; in a network of hollow fiber membranes with different functions; in a spirally wound device in which hollow fibers are used to provide oxygen to the cells; in multibore capillaries; microencapsulated and in an oxygenpermeable membrane rotating system under microgravity conditions. Several designs of BAL devices that are different in configuration, cell source, and culture technique have currently undergone clinical trials (Morelli et al. 2010).

Biohybrid Artificial Liver (BAL)

References De Bartolo L, Bader A (2001) Review of a flat membrane bioreactor as a bioartificial liver. Ann Transplant 6:40–46 Ding YT, Shi XL (2011) Bioartificial liver devices: perspectives on the state of the art. Front Med 5:15–19 Kamlot A, Rozga J, Watanable FD, Demetriou AA (1996) Review: Artificial liver support systems. Biotechnol Bioeng 50:382–391 Kasuya J, Tanishita K (2012) Microporous membranebased liver tissue engineering for the reconstruction of three-dimensional functional liver tissues in vitro. Biomatter 2:290–295 Morelli S, Salerno S, Piscioneri A, Campana C, Drioli E, De Bartolo L (2010) Membrane bioreactors for regenerative medicine: an example of the bioartificial liver. Asia Pac J Chem Eng 5:146–159

B

Biohybrid Artificial Liver (BAL) Systems Sabrina Morelli National Research Council of Italy, Institute on Membrane Technology, ITM-CNR, Rende, CS, Italy

A biohybrid artificial liver (BAL) system is an artificial extracorporeal supportive device which represents an important therapeutic strategy for patients with acute liver failure. Generally, a BAL system consists of functional liver cells supported by an artificial cell culture material. In particular, it incorporates hepatocytes into a bioreactor in which the cells are immobilized, cultured, and induced to perform the hepatic functions by processing the blood or plasma of liver failure patients. The BAL system acts as a bridge for the patients until a donor organ is available for transplantation or until liver regeneration. The development of a BAL system involves many design considerations. It must provide (1) an adhesion support to the cells; (2) adequate mass transfer of oxygen, nutrients, and toxic substances from the blood or plasma of patients to the cell compartments and proteins, catabolites, and other specific compounds produced by cells from the cell compartment to the blood or plasma; (3) immunoprotection of cells; and (4) biocompatibility. BAL devices are classified by the cell # Springer-Verlag Berlin Heidelberg 2014 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_57-3

source, the type of culture system for the hepatocytes, and the configuration of the bioreactor. Several BAL systems have been evaluated preclinically in in vitro experiments and in large animal models of liver failure (Morelli et al. 2010). Currently, different types of BAL devices are in various stages of clinical evaluation, and some of them are listed in Table 1 (van de Kerkhove et al. 2004). Many of these devices use hollow fiber membranes (HFMs) as supports for the cultured hepatocytes and as immunoselective barriers between the plasma of the patients and the hepatocytes used in the bioreactor. Membranes also permit the transport of nutrients and metabolites to cells and the transport of catabolites and specific metabolic products to the blood. In the membrane bioreactors, mass transfer is determined by the molecular weight cutoff (MWCO) or pore diameter of the membrane and occurs by diffusion and/or convection in response to existing transmembrane concentration or pressure gradients. Most of the bioreactors for BAL systems use membranes with MWCO ranging from 70 to 100 kDa that allow the transport of serum albumin but exclude proteins with high MW such as immunoglobulins and cells. One of the first clinical devices using HFMs was developed by Sussman and coworkers, namely, the extracorporeal liver-assist device (ELAD) in which the human hepatocytes were located outside the hollow fiber and blood flows through the lumen of the hollow fibers. This

2

Biohybrid Artificial Liver (BAL) Systems

Biohybrid Artificial Liver (BAL) Systems, Table 1 Membrane BAL systems in clinical evaluation BAL system Kiil dialyzer bioartificial liver ELAD Amphioxus Cell Technology LLS Charite, Humboldt University, Germany HepatAssist Circe Biomedical AMC-BAL University of Amsterdam BLSS Excorp Medical Inc. BAL TECA Corp.

Bioreactor configuration Plate

Membrane Cellulose

Hollow fiber

Cellulose acetate

Hollow fiber

Polyamide Polyethersulfone Polypropylene Polysulfone

Hollow fiber

References Matsumura et al. (1987) Sussman et al. (1992) Gerlach et al. (1994) Demetriou et al. (1995)

Spirally wound

Nonwoven polyester matrix, polypropylene

Flendrig et al. (1997)

Hollow fiber

Cellulose acetate

Patzer et al. (2002)

Hollow fiber

Polysulfone

Ding et al. (2003)

device was commercialized by Amphioxus Cell Technologies (Sussman et al. 1992). HepatAssist Circe Biomedical is the most clinically advanced system of its kind. It is an extracorporeal cellbased bioartificial liver device, based on the use of an open membrane hollow fiber bioreactor (Demetriou et al. 1995). In this system, hepatocytes are loaded into the extracapillary space, and the patient’s plasma flows through the capillary lumina of membranes. A more complex system is the liver support system (LSS) or the modular extracorporeal liver system (MELS) which consists of a bioreactor with four interwoven independent capillary membrane systems that serve different functions (Gerlach et al. 1994). The BLSS is a hollow fiber device that uses porcine hepatocytes embedded in a collagen matrix (Patzer et al. 2002). The Academic Medical Center Bioartificial Liver (AMC-BAL) developed by Flendrig et al. uses a three-dimensional, spirally wound, nonwoven polyester matrix for hepatocyte attachment with integrated hollow fibers for oxygen delivery to the cells (Flendrig et al. 1997). Another BAL system that is currently in clinical testing is a bioreactor from TECA Corp. in which a polysulfone membrane compartmentalizes porcine hepatocytes (Ding et al. 2003).

References Demetriou AA, Rozga J, Podesta L, Lepage E, Woolf G, Vierling J, Makowka LE, Moscioni AD, Hoffman A, McGrath M, Kong L, Rosen H (1995) Early clinical experience with a hybrid bioartificial liver. Scand J Gastroenterol 208:111–117 Ding YT, Qiu YD, Chen Z, Xu QX, Zhang HY, Tang Q, Yu DC (2003) The development of a new bioartificial liver and its application in 12 acute liver failure patients. World J Gastroenterol 9:829–832 Flendrig LM, la Soe JW, Jorning GG, Steenbeek A, Karlsen OT, Bovee WM, Ladiges NC, Te Velde AA, Chamuleau RA (1997) In vitro evaluation of a novel bioreactor based on an integral oxygenator and a spirally wound nonwoven polyester matrix for hepatocyte culture as small aggregates. J Hepatol 26:1379–1392 Gerlach JC, Encke J, Hole O, Muller C, Ryan CJ, Neuhaus P (1994) Bioreactor for larger scale hepatocyte in vitro perfusion. Transplantation 58:984–988 Matsumura KN, Guevara GR, Huston H, Hamilton WL, Rikimaru M, Yamasaki G, Matsumura MS (1987) Hybrid bioartificial liver in hepatic failure: preliminary clinical report. Surgery 101:99–103 Morelli S, Salerno S, Piscioneri A, Campana C, Drioli E, De Bartolo L (2010) Membrane bioreactors for regenerative medicine: an example of the bioartificial liver. Asia Pac J Chem Eng 5:146–159 Patzer JF, Mazariegos GV, Lopez R (2002) Preclinical evaluation of the Excorp Medical, Inc, bioartificial liver support system. J Am Coll Surg 195:299–310

Biohybrid Artificial Liver (BAL) Systems Sussman NL, Chong MG, Koussayer T, He DE, Shang TA, Whisennand HH, Kelly JH (1992) Reversal of fulminant hepatic failure using an extracorporeal liver assist device. Hepatology 16:60–65

3 van de Kerkhove MP, Hoekstra R, Chamuleau RAFM, van Gulik TM (2004) Clinical application of bioartificial liver support systems. Ann Surg 240:216–230

B

Biohybrid Membrane Systems Sabrina Morelli Institute on Membrane Technology, ITM-CNR, National Research Council of Italy, Rende (CS), Italy

Biohybrid membrane systems are engineered systems based on the combination of biological units, cells, or tissues, immobilized on an artificial structure, the membrane. In these systems, membranes act as instructive materials which are capable of supporting tissue/organ formation. Cells have to make an intimate contact with the surface of the membrane but also to develop close cell-cell connections, which is a precondition for their survival and high functional activity. Among polymeric materials, membranes in flat and hollow fiber configuration are the most attractive in the use of biohybrid systems for their characteristics of stability, biocompatibility, and selective permeability. Polymeric membranes could mimic the extracellular matrix with which cells interact allowing the organization of the cells into a three-dimensional architecture. The membranes are able to modulate the adhesion, proliferation, and differentiation of cells which are fundamental processes for tissue regeneration by governing the mass transfer of molecules that generate a precisely controlled microenvironment that mimic the specific features of in vivo environments. # Springer-Verlag Berlin Heidelberg 2014 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_58-3

Biohybrid membrane systems are successfully applied in the field of tissue engineering and regenerative medicine (Morelli et al. 2009). For the development of functional biohybrid membrane systems, a number of issues need to be addressed: morphological, physicochemical, mechanical, and transport properties of the membrane, the optimal density of immobilized cells, the interaction of cells with the membrane, the differentiation of cells, as well as the maintenance of viability and metabolic functions in vitro membrane constructs. Different types of biohybrid membrane systems have been proposed for the reconstruction and/or regeneration of many organs and tissues (e.g., the pancreas, liver, kidney, skin, and bone Scharp et al. 1994; Saito et al. 2006; De Bartolo et al. 2009; Ding and Shi 2011; Gentile et al. 2011). Currently, biohybrid membrane systems are also developed for the creation of a biomimetic microenvironment for neural tissue engineering since they may be used for the in vitro simulation of human brain functions. Semipermeable hollow fiber membranes are widely used as guidance channels in promoting in vitro and in vivo neuronal regeneration (Zhang et al. 2005; Morelli et al. 2010, 2012). Generally, biohybrid membrane systems could not only have a role in the replacement of injured organ or tissue but also accelerate the development of new drugs that may cure patients as an alternative to animal experimentation.

2

References De Bartolo L, Salerno S, Curcio E, Piscioneri A, Rende M, Morelli S, Tasselli F, Bader A, Drioli E (2009) Human hepatocyte functions in a crossed hollow fiber membrane bioreactor. Biomaterials 30:2531–2543 Ding YT, Shi XL (2011) Bioartificial liver devices: perspectives on the state of the art. Front Med 5:15–19 Gentile P, Chiono V, Tonda-Turo C, Ferreira AM, Ciardelli G (2011) Polymeric membranes for guided bone regeneration. Biotechnol J 6:1187–1197 Morelli S, Salerno S, Piscioneri A, Rende M, Campana C, Drioli E, De Bartolo L (2009) Membranes in regenerative medicine and tissue engineering. In: Drioli E, Giorno L (eds) Membrane operations: innovative separations and transformations. Wiley VCH, Verlag GmbH & Co. KGaA, Weinheim, pp 433–446 Morelli S, Salerno S, Piscioneri A, Papenburg BJ, Di Vito A, Giusi G, Canonaco M, Stamatialis D, Drioli E, De Bartolo L (2010) Influence of micro-

Biohybrid Membrane Systems patterned PLLA membranes on outgrowth and orientation of hippocampal neurites. Biomaterials 31:7000–7011 Morelli S, Piscioneri A, Salerno S, Rende M, Campana C, Tasselli F, di Vito A, Giusi G, Canonaco M, Drioli E, De Bartolo L (2012) Flat and tubular membrane systems for the reconstruction of hippocampal neuronal network. J Tissue Eng Regen Med 6:299–313 Saito A, Aung T, Sekiguchi K, Sato Y, Vu DM, Inagaki M, Kanai G, Tanaka R, Suzuki H, Kakuta T (2006) Present status and perspectives of bioartificial kidneys. J Artif Organs 9:130–135 Scharp DW, Swanson CJ, Olack BJ, Latta PP, Hegra OD, Doherty EJ, Gentile FT, Flavin KS, Ansara MF, Lacy PE (1994) Protection of encapsulated human islets implanted without immunosuppression in patients with type I or type II diabetes and in nondiabetic control subjects. Diabetes 43:1167–1170 Zhang N, Yan H, Wen X (2005) Tissue-engineering approaches for axonal guidance. Brain Res Rev 49:48–64

B

Biosurfactant Frederico de Araujo Kronemberger COPPE – Chemical Engineering Program, Federal University of Rio de Janeiro (UFRJ), Rio de Janeiro, RJ, Brazil

A biosurfactant (also called microbial surfactant) can be defined as a surfactant compound produced by microorganisms. Since surfactants are amphiphilic compounds, presenting both hydrophilic and hydrophobic moieties, they tend to move toward the interfaces, reducing the surface and/or interfacial tensions. These properties make surfactants excellent detergency, emulsifier, foaming, and dispersing agents (Mukherjee and Das 2010). Biosurfactants can be categorized mainly by their chemical composition and their microbial origin, being classified as glycolipids, lipopeptides and lipoproteins, phospholipids and fatty acids, polymeric surfactants, and particulate surfactants, and they are produced by a great variety of microorganisms, either secreted extracellularly or attached to parts of cells (Desai and Banat 1997). The biosurfactants have several advantages over chemically synthesized surfactants such as lower toxicity, higher biodegradability, and effectiveness at extreme temperatures or pH values, besides presenting high surface activity and low critical micelle concentration values emerging as promising substitutes of the latter (Vaz et al. 2012). # Springer-Verlag Berlin Heidelberg 2014 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_61-1

Concerning the oil industry, biosurfactants can be used in the bioremediation of oil pollutants, in the enhanced oil recovery (reducing the capillary forces that retain the oil in the reservoir rock), in the treatment of oily sludges, and in the cleanup of storage tanks. They can also be applied in the pharmaceutical (due to their antimicrobial activity) and agricultural sectors (for the hydrophilization of heavy soil), as a pesticide, and in the food industry (Mukherjee and Das 2010; Banat 1995; Freire et al. 2009). They can also be used in the effluent treatment, for metal ion removal from aqueous solutions (Ramani et al. 2012) and to enhance the water/oil interaction in the degradation of high fat content effluents (Damasceno et al. 2012). Regarding membrane separation processes, biosurfactants can be used in the micellar enhanced ultrafiltration, in which small contaminants, like heavy metal ions, are bound onto larger surfactant micelle complexes. These ions associated with surfactant macromolecules can be easily retained by an ultrafiltration membrane module. El Zeftawy and Mulligan (2011) investigated the use of rhamnolipid biosurfactants (glycolipids) in the micellar enhanced ultrafiltration of cadmium, copper, nickel, lead, and zinc ions with polysulfone hollow-fiber ultrafiltration membranes with molecular weight cutoff of 10,000 and 30,000. The authors reported nearly complete rejection of the ions, considering a feed concentration of 5 mg.L 1, when using a rhamnolipid solution above its critical micelle

2

concentration with both membranes. Even when increasing the metal ion content up to 50 mg.L 1, the biosurfactant concentration could be adjusted to enable their complete rejection, with a permeate flux up to 200 L.h 1.m 2.bar 1. Hong et al. (1998) also reported the ultrafiltration of copper, zinc, cadmium, and nickel ions using a polycarboxylic acid-type biosurfactant. The author used flat-sheet cellulose acetate membranes with molecular weight cutoff of 1,000 and 3,000, but the biosurfactant was not as effective as the rhamnolipid reported by El Zeftawy and Mulligan (2011), leading to lower rejection values. Another interesting application of biosurfactant was reported by Qin et al. (2012). These authors used rhamnolipid biosurfactants to enhance the frying oil degradation and to reduce the membrane fouling in a submerged membrane bioreactor, and an increase from 66 % up to 91 % in the oil removal efficiency was reported. The antifouling property of the biosurfactant was also confirmed. Besides, the biosurfactants could be used to replace their chemically synthesized counterparts in several other membrane processes, like the micellar enhanced ultrafiltration of aromatic alcohols and in the emulsion liquid membranes, used to remove and/or purify metal ions, dyes, lignosulfonate, and lactic acid.

References Banat IM (1995) Biosurfactants production and possible uses in microbial enhanced oil recovery and oil pollution remediation: a review. Bioresour Technol 51:1–12

Biosurfactant Damasceno FRC, Cammarota MC, Freire DMG (2012) The combined use of a biosurfactant and an enzyme preparation to treat an effluent with a high fat content. Colloid Surf B 95:241–246 Desai JD, Banat IM (1997) Microbial production of surfactants and their commercial potential. Microbiol Mol Biol R 61:47–64 El Zeftawy MAM, Mulligan CN (2011) Use of rhamnolipid to remove heavy metals from wastewater by micellar-enhanced ultrafiltration (MEUF). Sep Purif Technol 77:120–127 Freire DMG, Araujo LV, Kronemberger FA, Nitschke M (2009) Biosurfactants as emerging additives in food processing. In: Passos ML, Ribeiro CP (eds) Innovation in food engineering: new techniques and products, Contemporary Food Engineering Series. CRC Press, Boca Raton, pp 685–705 Hong J, Yang S, Lee C, Choi Y, Kajiuchi T (1998) Ultrafiltration of divalent metal cations from aqueous solution using polycarboxylic acid type biosurfactant. J Colloid Interf Sci 202:63–73 Mukherjee AK, Das K (2010) Microbial surfactants and their potential applications: an overview. In: Sen R (ed) Biosurfactants, vol 672, Advances in experimental medicine and biology series. Springer Science +Business Media, LLC, New York, pp 54–64 Qin L, Zhang G, Meng Q, Zhang H, Xu L, Lv B (2012) Enhanced submerged membrane bioreactor combined with biosurfactant rhamnolipids: performance for frying oil degradation and membrane fouling reduction. Bioresour Technol 126:314–320 Ramani K, Chandan Jain S, Mandal AB, Sekaran G (2012) Microbial induced lipoprotein biosurfactant from slaughterhouse lipid waste and its application to the removal of metal ions from aqueous solution. Colloid Surf B 97:254–263 Vaz DA, Gudin˜a EJ, Alameda EJ, Teixeira JA, Rodrigues LR (2012) Performance of a biosurfactant produced by a Bacillus subtilis strain isolated from crude oil samples as compared to commercial chemical surfactants. Colloid Surf B 89:167–174

B

Biosurfactant Production Frederico de Araujo Kronemberger COPPE - Chemical Engineering Program, Federal University of Rio de Janeiro (UFRJ), Rio de Janeiro, RJ, Brazil

A biosurfactant can be defined as a surfactant compound produced by microorganisms. The biosurfactants have several advantages over chemically synthesized surfactants such as lower toxicity, higher biodegradability, and effectiveness at extreme temperatures or pH values, besides presenting high-surface activity and low-critical micelle concentration values emerging as promising substitutes of the latter (Vaz et al. 2012). In the present moment, the biosurfactants still present high production costs in comparison to the chemically synthesized surfactants. That is mainly the result of the low productivity of the microbial strains and the inefficient methodology of the bioprocessing. The technological improvement of the production process is essential (Kronemberger et al. 2008). In order to decrease the production costs, the scale-up of the whole production process, including upstream and downstream, should be developed. One of the bottlenecks of the biosurfactant production relies on the fact that most of them are obtained through aerobic bioreactions. The use of the conventional submerged aeration can # Springer-Verlag Berlin Heidelberg 2014 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_62-1

lead to the formation of very stable foams, causing serious operational problems. In order to overcome that difficulty, a nondispersive oxygenation process using membrane contactors can be applied (Kronemberger et al. 2008, 2012; Gruber and Chmiel 1991). Kronemberger et al. (2008) investigated the rhamnolipid-type biosurfactant production in bioreactors with a nondispersive oxygenation device, obtaining productivities higher to the ones observed in shake flasks. This system was then successfully used in a fed-batch experiment, in order to assess the potential of a long-term production (Kronemberger et al. 2010). Coutte et al. (2010) reported a similar system for the production of lipopeptide biosurfactants, comparing internal and external nondispersive oxygenation. Several authors have been investigating the recovery of biosurfactants, mainly surfactin, using ultrafiltration. Chen et al. (2008a) reported the flux decline in the ultrafiltration of surfactin using cellulose ester and polyethersulfone membranes with 100,000 Da of molecular weight cutoff, and the latter was recommended as the best one for this kind of experiment, even though the biosurfactant recovery was a little lower. Isa et al. (2007) investigated a two-step ultrafiltration recovery system for surfactin. In the first step ultrafiltration, surfactin was retained by polyethersulfone or regenerated cellulose membranes at above its critical micelle concentration. In the second step, with the same kind of membranes, after the disruption of the micelles by the

2

addition of methanol, the purified surfactin was recovered in the permeate. Chen et al. (2008b) also reported the surfactin recovery, but using ammonium sulfate salting out, ultrafiltration, nanofiltration, and their hybrid process. The combination of salting out and ultrafiltration was selected due to the reduction of the fouling in the membranes used. Another point of view should be the whole integrated production system supported by membrane processes. A system described by Coutte et al. (2013) comprises the nondispersive oxygenation of a bioreactor for the production of surfactin and the continuous cell removal and product separation using microfiltration and ultrafiltration modules, respectively. A pilot scale system, designed for the production of rhamnolipid-type biosurfactants, with 200 L of useful volume was also described (Kronemberger et al. 2012). It comprises microfiltration modules for fresh medium sterilization, nondispersive oxygenation, another set of microfiltration modules with self-backwashing for cell retention, and a reverse osmosis unit used to concentrate the product and to recover the water as the permeate stream, enabling its reuse and minimizing the effluents.

References Chen H, Chen Y, Juang R (2008a) Flux decline and membrane cleaning in cross-flow ultrafiltration of treated

Biosurfactant Production fermentation broths for surfactin recovery. Sep Purif Technol 62:47–55 Chen H, Chen Y, Juang R (2008b) Recovery of surfactin from fermentation broths by a hybrid salting-out and membrane filtration process. Sep Purif Technol 59:244–252 Coutte F, Lecouturier D, Yahia SA, Lecle`re V, Be´chet M, Jacques P, Dhulster P (2010) Production of surfactin and fengycin by Bacillus subtilis in a bubbleless membrane bioreactor. Appl Microbiol Biotechnol 87:499–507 Coutte F, Lecouturier D, Lecle`re V, Be´chet M, Jacques P, Dhulster P (2013) New integrated bioprocess for the continuous production, extraction and purification of lipopeptides produced by Bacillus subtilis in membrane bioreactor. Process Biochem 48:25–32 Gruber T, Chmiel H (1991) Aerobic production of biosurfactants avoiding foam problems. In: Reuss M, Knackmuss HJ, Chmiel H, Gilles ED (eds) Biochemical Engineering-Stuttgart. Fischer Verlag, Stuttgart, pp. 212–215 Isa MHM, Coraglia DE, Frazier RA, Jauregi P (2007) Recovery and purification of surfactin from fermentation broth by a two-step ultrafiltration process. J Membr Sci 296:51–57 Kronemberger FA, Santa Anna LMM, Fernandes ACLB, Menezes RR, Borges CP, Freire DMG (2008) Oxygencontrolled biosurfactant production in a bench scale bioreactor. Appl Biochem Biotechnol 147:33–45 Kronemberger FA, Borges CP, Freire DMG (2010) Fed-batch biosurfactant production in a bioreactor. Int Rev Chem Eng 2:513–518 Kronemberger FA, Freire DMG, Castro AM, Santa Anna LMM, Borges CP (2012) System for obtaining biological products. Patent WO2012/079138 A1 Vaz DA, Gudin˜a EJ, Alameda EJ, Teixeira JA, Rodrigues LR (2012) Performance of a biosurfactant produced by a Bacillus subtilis strain isolated from crude oil samples as compared to commercial chemical surfactants. Colloids Surf B 89:167–174

E

Electrodialysis with Bipolar Membranes Heiner Strathmann Institute for Chemical Technology, Stuttgart University, Stuttgart, Germany

The conventional electrodialysis can be combined with bipolar membranes and utilized to produce acids and bases from the corresponding salts. A bipolar membrane is a laminate of an anion on a cation-exchange layer. In this process monopolar cation- and anion-exchange membranes are installed together with bipolar membranes in alternating series in an electrodialysis stack as illustrated in Fig. 1 which shows a typical repeating unit of an electrodialysis stack with bipolar membranes is composed of three cells, two monopolar membranes and a bipolar membrane. The outer cells of the repeating unit are fed with a salt solution, the inner cells with water, or a diluted acid and base. When an electrical potential gradient is applied across a repeating unit, protons and hydroxide ions which are generated in the bipolar membrane generate with the cations and anions removed from the salt solution, an acid and a base on either side of the bipolar membrane. The process design is closely related to that of the conventional electrodialysis using the sheet flow stack concept (Liu et al. 1977; Simons 1993). # Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_64-1

The utilization of electrodialysis with bipolar membranes to produce acids and bases from the corresponding salts is economically very attractive and has a multitude of interesting potential applications in the chemical industry as well as in biotechnology and water treatment processes. Its key component is the bipolar membrane. The bipolar membrane schematically illustrated in Fig. 2 consists of a laminate of an anion- and a cation-exchange membrane with a 4–5 nm thick catalytic transition layer in between. In Fig. 2 this transition layer has been artificially magnified. Water is diffusing through both membrane layers into the transition layer where it gets electrocatalytically dissociated into H+- and OH-ions, which migrate toward cathode and anode into the outer solutions. The energy required for the water dissociation can be calculated from the Nernst equation for a concentration chain between solutions of different pH-values. It is given by: DG ¼ FD’ ¼ 2:3RTDpH Here DG is the Gibbs free energy and DpH and D’ are the pH-value and the potential difference between the two solutions separated by the bipolar membrane. For 1 mol/L acid and base in the two phases separated by the bipolar membrane, DG is 0,022 kWh/mol and D’ is ca. 0,83 V at 25 C. Compared to the ohmic potential drop over the membranes, the required potential drop for water splitting in the transition layer is much

2

Electrodialysis with Bipolar Membranes Base

Acid

Base

Acid

Repeating cell unit bpm cm + – – + – – + – – – + + – M + – – + – – + – – + – – + – – − + – – OH + – – + – – + – – + – – + – – + – –

+

Salt solution

am cm + – + – + – + – M+ – M+ H+ + + – + – + – OH− − − + X – X + – + – + – + – + – + – + –

Salt solution

Electrodialysis with Bipolar Membranes, Fig. 1 Schematic drawing illustrating the principle of the electrodialytic production of an acid and a base from the corresponding salt with bipolar membranes. Repeating

Anion-exchange layer

Cation-exchange layer

Cathode

H2O H+

– – – – – – – –

+ + + + + + + +

H2O Anode OH–

Bipolar membrane

Electrodialysis with Bipolar Membranes, Fig. 2 Schematic drawing illustrating the electrodialytic water dissociation in a bipolar membrane with water diffusing into the reaction region between the cation- and anion-exchange layers of the membrane and protons and hydroxide migrate to the corresponding electrode

more pronounced. The determination of the costs for the production of acids and bases from the corresponding salts follows the same general procedure as applied for the cost calculation in electrodialysis desalination. The overall costs are the investment-related costs and the operating costs. The investment-related costs are dominated by the membrane costs and are proportional to the required membrane area for a given capacity plant. They are a function of the current density applied in a given stack operation. A unit cell

bpm + – + – + – + – + – + – + – + – + – + – + – + – + – + – + – + –

H+

X

am + + + + + + + + + + + + + + + +

Salt solution

cell unit consisting of a cation-exchange membrane (cm), a bipolar membrane (bpm), and an anion-exchange membrane (am)

contains a bipolar membrane, a cation- and an anion-exchange membrane. The bipolar membrane is rather expensive, and its useful life time as well as that of the anion-exchange membrane is rather limited in strong bases. The operating costs in electrodialysis with bipolar membranes are strongly determined by the energy requirements which are composed of the energy required for the water dissociation in the bipolar membrane and the energy necessary to transfer the salt ions from the feed solution and protons and hydroxide ions from the transition region of the bipolar membrane into the acid and base solutions. The energy consumption due to the pumping of the solutions through the stack can generally be neglected. Since bipolar membranes became available as commercial products, a large number of applications have been identified and studied on a laboratory or pilot plant scale. However, in spite of the obvious technical and economical advantages of the technology, largescale industrial plants are still quite rare (Gineste et al. 1996). The main reasons for the reluctant use of bipolar membrane electrodialysis are poor membrane stability at very high or low pH-values and insufficient permselectivity at high ion concentrations, which results in a substantial product salt contamination, low current

Electrodialysis with Bipolar Membranes

efficiency, and short membrane life. Nevertheless, there are a number of smaller-scale applications in the chemical process industry, in biotechnology, in food processing, and in wastewater treatment.

References Gineste JL, Pourecelly G, Lorrain Y, Presin F, Gavach C (1996) Analysis of factors limiting the use of BPM: a

3 simplified model to determine trends. J Membr Sci 112:199–208 Liu KJ, Chlanda FP, Nagasubramanian KJ (1977) Use of bipolar membranes for generation of acid and base: an engineering and economic analysis. J Membr Sci 2:109–124 Simons R (1993) Preparation of high performance bipolar membranes. J Membr Sci 78:13–23

B

Blood Separation A. Higuchi Department of Chemical and Materials Engineering, National Central University, Chung-Li, Taoyuan, Taiwan

Blood separation can be performed by centrifugation, magnetic cell selection system (MACS), fluorescence-activated cell sorting (FACS), and membrane filtration method. Blood is a living tissue composed of several blood cells in plasma. The cellular elements of red blood cells (RBCs), platelets, and white blood cells make up 45 % of the volume of whole blood. Another 55 % is plasma, which contains 7–8 % of the plasma proteins and 92–93 % of water (Higuchi 2010). Figure 1 shows typical blood after centrifugation with and without addition of Ficoll-Paque (Ficoll-Hypaque) solution. After centrifugation of blood without addition of Ficoll-Paque (Ficoll-Hypaque) solution (native blood centrifugation), blood can be separated into plasma layer, platelet and leukocyte layer, and RBC layer (Fig. 1a). In this case, each layer contains other contaminant cells, e.g., RBC layer contains 96 % of RBCs, 3 % of leukocytes, and 1 % of platelets in blood cells. Platelets and leukocytes are also included in plasma layers. Platelet-rich plasma is necessary to use for the evaluation of biocompatibility of biomaterials (Higuchi et al. 2003). Platelet-rich plasma is obtained by centrifugation # Springer-Verlag Berlin Heidelberg 2014 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_72-7

of peripheral blood or umbilical cord blood at 3,000 rpm. Platelet-poor plasma is used for plasma protein adsorption on biomaterials for the evaluation of biocompatibility of the biomaterials (Higuchi et al. 2003). Platelet-poor plasma is obtained by centrifugation of peripheral blood or umbilical cord blood at a relatively high speed of 3,000 rpm. Mononuclear cells including hematopoietic stem cells (HSCs) cannot easily be obtained by centrifugation of native blood. Therefore, Ficoll-Paque (or Ficoll-Hypaque) solution was injected into blood sample, and the mixed solution was centrifuged at 400 g for 30–40 min at 20 C (Fig. 1b). The upper layer contains plasma and platelets. Mononuclear cells including lymphocytes (T cells, B cells, and NK (natural killer) cells) and HSCs can be isolated from the upper second layer (Fig. 1b). When specific cells such as HSCs, T cells, N cells, or NK cells should be isolated, MACS or FACS are applied. Direct application of MACS and FACS to isolate the specific blood cells is difficult due to large quantity of RBCs in blood samples. After the mononuclear cells were isolated by FicollPaque method, residual RBCs were removed by the treatment of lysing solution and then HSCs (CD34+ cells) can be isolated by MACS or FACS treatment using antibody of CD34+ (Chen et al. 2012). MACS is a sophisticated cell separation method, in which magnetic beads attaching a monoclonal antibody as the cell-surface marker are mixed with cells. Figure 2 shows the

2 Blood Separation, Fig. 1 Blood components after centrifugation of native blood (a) and before and after centrifugation of blood with Ficoll-Paque solution (b)

Blood Separation

a

Blood contains 55% of plasma. Plasma

Concentration of plasma protein is 7 - 8%

Platelets Lymphocytes

Blood contains 45% of blood cells (96% of red blood cell, 3% of leukocytes, and 1% of platelet)

Red Blood cell

b Blood sample

Plasma, Platelets

Centrifugation Lymphocytes Ficoll-Paque Ficoll-Paque

Blood Separation, Fig. 2 Schematic mechanism of the separation method of cells by a magnetic cell selection system

a

Magnet

schematic mechanism of the separation method by an MACS. The magnetic beads attaching the monoclonal antibody are separated by magnetic force to collect the specific marked cells. The MACS needs to use an expensive antibody conjugated with magnetic beads to bind to the target cells for the detection of the cells. Both cell

Red blood cells, Granulocytes

b

Magnet

Magnet

Magnet

separation methods using FACS and MACS are not applicable if the antibodies to the specific markers on the surface of the target cells have not been established. Blood cell separation through membrane filtration was recently reported by several researchers (Komai et al. 1998; Yasutake

Blood Separation

et al. 2001; Higuchi et al. 2004, 2008). Typical blood cell separation membranes are leukocyte removal filter (membrane) and HSC purification membranes. HSC separation from peripheral blood and umbilical cord blood through surface-modified polyurethane membranes by membrane filtration method was reported (Higuchi et al. 2004, 2008, 2010). Peripheral blood or umbilical cord blood was permeated though the surface-modified membranes by filtration. HSCs are more adhesive cells than RBCs, platelets, and lymphocytes. Therefore, HSCs remained to adhere on the membranes during permeation of blood. The membrane-adhering HSCs were rinsed with phosphate buffer saline and subsequently human serum albumin or dextran solution as surfactant solution was permeated through the membranes. The HSCs can be harvested in the recovery solution of human serum albumin or dextran solution. The membrane filtration method of blood separation should be useful, because centrifugation instrument is not necessary to use and antibodies targeting specific cells are not used in the method, which cause contamination of antibodies in the blood cell samples.

Cross-References ▶ Blood Cell Origins ▶ Blood Filtration ▶ Blood Treatment Membranes

3

References Chen LY, Chang Y, Shiao JS, Ling QD, Chang Y, Chen YH, Chen DC, Hsu ST, Lee H, Higuchi A (2012) Effect of the surface density of nanosegments immobilized on culture dishes on ex vivo expansion of hematopoietic stem and progenitor cells from umbilical cord blood. Acta Biomater 8:1749–1758 Higuchi A (2010) Separation and purification of stem and blood cells by porous polymeric membranes. In: Driolli E, Giorno L (eds) Comprehensive membrane science and engineering. Elsevier, Cambridge Higuchi A, Sugiyama K, Yoon BO, Sakurai M, Hara M, Sumita M, Sugahara S, Shirai T (2003) Serum protein and platelet adsorption on pluronic-coated polysulfone membranes. Biomaterials 24:3235–3245 Higuchi A, Yamamiya S, Yoon BO, Sakurai M, Hara M (2004) Peripheral blood cell separation through surface-modified polyurethane membranes. J Biomed Mater Res Part A 68A:34–42 Higuchi A, Sekiya M, Gomei Y, Sakurai M, Chen WY, Egashira S, Matsuoka Y (2008) Separation of hematopoietic stem cells from human peripheral blood through modified polyurethane foaming membranes. J Biomed Mater Res Part A 85A:853–861 Higuchi A, Yang ST, Li PT, Tamai M, Tagawa T, Chang Y, Chang Y, Ling QD, Hsu ST (2010) Direct ex vivo expansion of hematopoietic stem cells from umbilical cord blood on membranes. J Membr Sci 351:104–111 Komai H, Naito Y, Fujiwara K, Takagaki Y, Noguchi Y, Nishimura Y (1998) The protective effect of a leukocyte removal filter on the lung in open-heart surgery for ventricular septal defect. Perfusion 13:27–34 Yasutake M, Sumita M, Terashima S, Tokushima Y, Nitadori Y, Takahashi TA (2001) Stem cell collection filter system for human placental/umbilical cord blood processing. Vox Sang 80:101–105

B

Blood Treatment Membrane A. Higuchi Department of Chemical and Materials Engineering, National Central University, Chung-Li, Taoyuan, Taiwan

Blood treatment membrane is categorized as dialysis membrane, leucocyte removal filter, and plasma separation membrane.

Dialysis Membrane Dialysis is a process for removing waste and excess water from the blood and is used primarily to provide an artificial replacement for lost kidney function in people with renal failure. Dialysis was used for those with an acute disturbance in kidney function (acute kidney injury), or progressive but chronically worsening kidney function. Dialysis works on the principles of the diffusion of solutes and ultrafiltration of fluid across a semipermeable membrane. Blood flows by one side of a semipermeable membrane, and a special dialysis fluid flows by the opposite side. A semipermeable membrane (dialysis membrane, dialyzer) is a thin layer of material that contains the appropriate size of pores. Smaller solutes (urea, NaCl) and fluid pass through the membrane, but the membrane blocks the passage of larger substances such as red blood cells and # Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_73-3

large proteins (albumin, globulin). This is the filtering process taking place in the kidneys, when the blood enters the kidneys and the larger substances are separated from the smaller ones in the glomerulus (Daugirdas et al. 2006). In hemodialysis, the patient’s blood is pumped through the blood compartment of a dialyzer, exposing it to a dialysis membrane. The dialyzer is composed of thousands of tiny synthetic hollow fibers. The fiber wall acts as the semipermeable membrane. Blood flows through the fibers, dialysis solution flows around the outside of the fibers, and water and wastes move between these two solutions (Daugirdas et al. 2006). The cleansed blood is then returned via the circuit back to the body. Ultrafiltration occurs by increasing the hydrostatic pressure across the dialyzer membrane. This usually is done by applying a negative pressure to the dialysate compartment of the dialyzer. This pressure gradient causes water and dissolved solutes to move from blood to dialysate and allows the removal of several liters of excess fluid during a typical 3- to 5-h treatment. Hemodialysis treatments are typically given in a dialysis center three times per week (Daugirdas et al. 2006). Hemodialysis membranes are typically prepared from cellulose materials or polysulfonepolyvinylpyrrolidone (PVP)-blended materials. Cellulose is hydrophilic and can be used as a hemocompatible material, whereas polysulfone is one of the engineering plastics and needs to add a hydrophilic and hemocompatible material

2

as a blending material. PVP shows relatively good hemocompatibility and the more important fact is that PVP can be blended well with polysulfone. There are a lot of materials reported excellent biocompatibility. However, these biocompatible materials cannot be used as blending materials in polysulfone dialysis membranes due to low mixing with polysulfone. PVP is also used as a porogen in dialysis membranes of polysulfone. There is a recent demand for hemodialysis membranes that remove the low-molecular-weight proteins such as b2myoglobin (MW 11,500) and endotoxin (subunit of MW = 5,000–20,000) and useful albumin in the plasma should be recovered by the membranes. Polysulfone hollow fibers blended with PVP have been widely used as suitable hemodialysis membranes which satisfy this requirement (Higuchi et al. 2002).

Leukocyte Removal Filter White blood cells (leukocytes) generate many adverse reactions during blood-transfusion therapy, which are graft-versus-host disease (GVHD), platelet refractoriness, nonhemolytic febrile transfusion reaction, and infection of viruses, such as human T-lymphotropic virus (HTLV), cytomegalovirus (CMV), and human immunodeficient virus (HIV) (Higuchi 2010). It was found that most of the viruses infect specific type of leukocytes, such as granulocytes, monocytes, lymphocytes, lymphocytes-B, T helper cell (CD4+ cell), and T-cell suppressor/cytotoxic cells (CD8+ cell). HTLV-1 and HIV mainly infect T helper cell, while CMV mainly infects granulocytes, monocytes, and lymphocytes. GVHD was mainly generated by T helper cell and T-cell suppressor/cytotoxic cells (CD8+ cell) (Higuchi 2010). Therefore, removal of leukocytes in RBC and platelet concentrates as well as whole blood component are essential to prevent the adverse effect of contaminated leukocytes. Leukocytes can be removed using a filter comprised of nonwoven fabric or sponge materials as a filter medium. The mechanism of leukocyte removal on the filters comprised of

Blood Treatment Membrane

nonwoven fabric is based on the adsorption of leukocytes, while that comprised of sponge materials is based on the sieving effect and adsorption. Filtration methods have several advantages compared to other methods of removing leukocytes such as centrifugation. Virus contamination is lower in blood components during the process in the filtration method than in the centrifugation method due to mild operation and the ease of operation under sterilized conditions. Leukocyte removal filters were typically made of polyurethane (PU) foaming membranes where the pore was made by salt leaching method and nonwoven fabric. The pore structure of both filters is found to be completely different, although the pore size of those filters was almost the same from capillary flow porometer measurements. The mechanism of leukocyte removal (i.e., separation of leukocyte from plasma and other blood cells) in leukocyte removal filters is based on leukocyte adsorption on the filters. The adsorption of leukocytes was affected significantly by filter materials, pore structure, and pore size.

Plasma Separation Membrane Plasma-exchange therapy has been increasingly applied clinically over the past few years. Membrane plasma separation has been used since 1979, which is similar to hemodialysis and hemofiltration. Plasma separation from whole blood is now performed routinely. The materials of plasma separation membranes are typically made of nitrocellulose, polysulfone, and polypropylene.

Cross-References ▶ Blood Cell Origins ▶ Blood Filtration ▶ Blood Separation ▶ Diafiltration ▶ Diffusion

Blood Treatment Membrane

References Daugirdas JT, Van Stone JC, Boag JT (2006) Hemodialysis apparatus. In: Daugirdas JT, Peter G, Blake PG, Todd S, Ing TS (eds) Handbook of dialysis. Lippincott Williams & Wilkins, Riverwoods Higuchi A (2010) Separation and purification of stem and blood cells by porous polymeric membranes. In:

3 Driolli E, Giorno L (eds) Comprehensive membrane science and engineering. Elsevier, Cambridge Higuchi A, Shirano K, Harashima M, Yoon BO, Hara M, Hattori M, Imamura K (2002) Chemically modified polysulfone hollow fibers with vinylpyrrolidone having improved blood compatibility. Biomaterials 23:2659–2666

B

Boron Reduction Piotr Dydo Faculty of Chemistry, Silesian University of Technology, Gliwice, Poland

Synonyms Boron removal

Characteristics The boron element has an average concentration in the Earth’s crust equal to about 10 mg/kg. Due to its strong affinity toward oxygen, it exists in nature mainly in the form of boric acids or borates. Boric acid and borates are mainly used in the glass and ceramic industry to obtain borosilicate glass, insulation fiberglass, flame retardant fiberglass, ceramic glazes, and porcelain enamels. Boron compounds are also used as a flame retardant in plastics and cellulosic insulation, neutron absorbers, herbicides (when at high concentrations) or fertilizers (when at low doses), and components of washing powders and soaps. According to the WHO (2009) and (2011), average boron concentration in surface waters does not exceed the value of 0.5 mg/L. However, the amount of boron in these waters depends upon the presence of boron-bearing minerals in the proximity of water reservoirs. It may also be # Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_74-10

affected by the discharge of municipal and industrial effluents into the environment. On the other hand, naturally occurring boron is present in ground waters at a wide concentration range from 100 mg/L. This includes strongly mineralized, naturally carbonized geothermal waters. Considerable amounts of boron are also present in oceans, with an average concentration of 4.5 mg/L as reported by the WHO in 2009. Despite the fact that boron is an important nutrient, it manifests toxic action against plants and animals when found at high concentration in irrigation or drinking water. The effects of boron on plants and mammals were summarized by the WHO (2009), Kabay et al. (2010), and Hilal et al. (2011). Among the reported effects, the adverse impact of boron on the male reproductive system in rats, mice, and dogs was presented. Based on analysis of toxicokintetic of boron, the WHO proposed a guideline value for boron in drinking water of 2.4 mg/L in 2011. In areas with high natural boron levels, local regulatory and health authorities are, however, advised to consider values in excess of 2.4 mg/L by assessing exposure from other sources (e.g., food). The low drinking water boron content limit implies the necessity for a reduction of boron content in drinking water, especially when seawater or boron-rich underground waters (e.g., geothermal waters) are to be treated. There could also be a need for boron removal from industrial and municipal effluents that contain

2

more than the WHO guideline value. Conventional methods for water treatment do not significantly remove boron. The methods proven to be efficient in a reduction of boron content in waters are either adsorption or membrane based. Adsorption-based methods for boron removal include coagulation with metal hydroxides, adsorption on clays, fly ashes, and activated carbon (WHO 2009). Also, successful boron reduction with conventional anion exchange resins or resins functionalized with N-methyl D-glucamine (Chillon Arias et al. 2011) was reported. Among the membrane-based methods, boron removal by reverse osmosis is reported most commonly. Also some achievements in boron removal with ion-exchange membranes as well as boron removal by electrodialysis are presented. In addition, some hybrid systems, which combine membrane filtration with sorption, are proposed, e.g., the adsorption-membrane filtration (AMF) system. In this system, microparticulate boron selective resin suspension is recirculated in the retentate. Boron species remains adsorbed on the resin, while the borondepleted water is continuously separated by microfiltration. According to Kabay et al. 2008, the main advantage of the AMF over other methods of boron separation lies in the fast removal of boron, ease of sorbent separation, and low operating pressure. According to the WHO (2009), the abovementioned methods for boron removal are likely to be prohibitively expensive, and blending with waters of low boron content may be the only

Boron Reduction

economically feasible option to reduce high concentrations of boron in water.

Cross-References ▶ Boron Removal by Electrodialysis ▶ Boron Removal by Reverse Osmosis ▶ Boron Removal with Ion-Exchange Membranes

References Chillon Arias MF, Valero i Bru L, Prats Rico D, Varo Galvan P (2011) Comparison of ion exchange resins used in reduction of boron in desalinated water for human consumption. Desalination 278:244–249 Hilal N, Kim GJ, Somerfield C (2011) Boron removal from saline water: a comprehensive review. Desalination 273:23–35 Kabay N, Bryjak M, Schlosser S, Kits M, Avlonitis S, Matejka Z, Al-Mutaz I, Yuksel M (2008) Adsorptionmembrane filtration (AMF) hybrid process for boron removal from sweater: an overview. Desalination 223:38–48 Kabay N, Guler E, Bryjak M (2010) Boron in seawater and methods for its separation – a review. Desalination 261:212–217 WHO (2009) Boron in drinking-water. Background document for development of WHO guidelines for drinkingwater quality. World Health Organization. http:// whqlibdoc.who.int/hq/2009/WHO_HSE_WSH_09.01_ 2_eng.pdf. Accessed 17 June 2012 WHO (2011) Guidelines for drinking-water quality, 4th edn. World Health Organization. http://whqlibdoc. who.int/publications/2011/9789241548151_eng.pdf. Accessed 17 June 2012

B

Boron Removal by Electrodialysis Piotr Dydo Faculty of Chemistry, Silesian University of Technology, Gliwice, Poland

As in the case of boron removal by reverse osmosis, the effectiveness of boron removal by electrodialysis (ED) is strongly affected by the aqueous chemistry of the boron species. In aqueous solutions, boric acid may exist in the form of boric acid, metaborate, and polyborates. Boric acid as a weak electrolyte (pKa = 9.2) dominates in diluted aqueous solutions of a pH equal to or less than 9, while at higher pH borate ions dominate. Therefore, the effectiveness of boron transport across ion-exchange membranes (IEM) in the electrodialysis process needs to be discussed with regard to the type of the species that dominates in the dilute.

The Effectiveness of Boric Acid Transport in ED Boric acid, H3BO3, is a small and electrically neutral species. The reported boric acid removal efficiencies and electric current efficiencies in ED are poor when compared to ionic species (Melnik et al. 1999; Turek et al. 2005, 2007, 2008a; Kabay et al. 2008; Banasiak and Schafer 2009; WHO 2009). In fact, boric acid fluxes # Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_75-4

across IEMs are so low that the possibility of boric acid separation from ionic species, strong acids (Melnik et al. 2005, 2007) and salts (Turek et al. 2005, 2007; Bandura-Zalska et al. 2009; Dydo 2012a), by the use of electrodialysis is considered. Employing this method, ionic species are effectively transported across IEMs while boric acid remains in the ion-depleted dilute solution. The type of the membrane, the dilute boron concentration, the presence of ions in the dilute, and the electric current density were shown to affect the rate of boric acid transport across IEM to a great extent (Melnik et al. 1999; Yazicigil and Oztekin 2006; Kabay et al. 2008; Dydo 2012a). The reported mechanism for boric acid transport across IEMs is diffusion (Dydo 2012b); however, the flux of boric acid was found to increase with an increase in the flux of ionic species. This was suggested to be the result of a kind of ion-coupled transport of boric acid across IEMs. The effectiveness of such a transport was found to decrease in the following ion series: sulfate nitrate chloride. It was also reported that most of the boric acid was transported across anion-exchange membrane (AEM). Turek et al. 2008b reported high boron fluxes and exceptionally high electric current efficiencies of up to 220 % when boric acid was transported from ion-depleted, neutral or acidic, dilute into the alkaline concentrate of pH >11. Such a behavior was identified to be the result of Donnan dialysis, in which hydroxyl acts as a

2

carrier for boric acid. However, in the light of recent reports, simple boric acid diffusion should explain the results as well. Tentative results on boron removal by electrodialysis units equipped with ion-exchange spacers were also presented by Oren et al. in 2006. The possibility of up to an 80 % reduction of boron from water containing 4.5 mg/L was reported.

The Effectiveness of Borate Transport in ED Melnik et al. (1999), Yazicigil et al. (2006), and Kabay et al. (2008) reported that as the pH of boron-containing water was brought up to just above the value of 9, a significant increase in the rate of boron transport across IEMs is observed. Moreover, Melnik et al. 1999 reported that in the case of heterogeneous membranes, an increase in the boron transport rate across anionexchange membrane at pH >9 is accompanied by a decrease in the rate of boron transport across cation-exchange membrane (CEM). It is clear that under these conditions (pH >9), borates are the boron species that dominates and their transport across AEMs rather than CEMs should be discussed. The effectiveness of borate transport across AEMs in ED systems was found to depend upon the pH of the dilute, the type of the membrane, dilute boron concentration, and the kind of the ion cotransported across the membrane. Yazicigil et al. in 2006 reported that at a dilute pH of approx. 9, there is a maximum boron transport rate, and at higher pH the rate of boron transport decreases. These results are contradictory to those presented by Melnik et al. (1999) (heterogeneous membranes), Turek et al. (2007), and Kabay et al. (2008), according to which there is a continuous increase in the rate of boron transport with dilute pH even when above 9.0. However, the Yazicigil et al. (2006) results seem to be justified by high boron concentration (0.1 mol/L) in their experiments. Similar behavior was observed by Ayyildiz and Kara in 2005. It is agreed that an increase in dilute boron concentration causes an increase in boron (borate) flux

Boron Removal by Electrodialysis

(Yazicigil et al. 2006; Kabay et. al 2008; Turek et al. 2008a). It is also agreed that boron (borate) is transported faster in the presence of chloride ions than in the presence of sulfates (Yazicigil et al. 2006; Kabay et al. 2008) or nitrates. In 2005 and 2008a, Turek et al. analyzed the effect of salinity on the electric current efficiency of boron (borate) transport. They found that the observed electric current efficiencies are low as long as dilute salinity remains high. During the initial part of the experiments, no, or almost no, boron was transported from the dilute. Then, i.e., when more than 90 % of the Cl was removed from the dilute, a dramatic increase in boron (borate) electric current efficiency was observed. This increase was accompanied by an increase in boron flux. Such an effect can be identified as the result of low borate mobility (diffusivity) when compared to other ions present in the waters, i.e., Cl . As long as there are ions of higher mobility than borates in the dilute, boron (borate) flux remains low. However, even in an ion-depleted dilute, the electric current efficiency of boron (borate) transport has not exceeded the value of 30 %. The remaining percentage of the electric current was probably utilized for hydroxyl ion transport. A considerable drop in the pH of the dilute was reported afterwards. This resulted in a drop in dilute pH and in a consequent reduction in the flux of boron since all the borate present was converted into boric acid. So the low electric current efficiency of borate transport in an ED system and the necessity for deep dilute demineralization make the economic feasibility of borate removal by ED questionable.

Cross-References ▶ Boron Reduction ▶ Boron Removal by Electrodialysis ▶ Boron Removal by Reverse Osmosis

References Ayyildiz HF, Kara H (2005) Boron removal by ion exchange membranes. Desalination 180:99–108

Boron Removal by Electrodialysis Banasiak LJ, Schafer AI (2009) Removal of boron, fluoride and nitrate by electrodialysis in the presence of organic matter. J Membr Sci 334:101–109 Bandura-Zalska B, Dydo P, Turek M (2009) Desalination of boron-containing wastewater at no boron transport. Desalination 241:133–137 Dydo P (2012a) The effect of process parameters on boric acid transport during the electrodialytic desalination of aqueous solutions containing selected salts. Desalination. In press. doi:10.1016/j.bbr.2011.03.031 Dydo P (2012b) The mechanism of boric acid transport during an electrodialytic desalination process. J Membr Sci 407–408:202–210 Kabay N, Arar O, Acara F, Ghazal A, Yuksel U, Yuksel M (2008) Removal of boron from water by electrodialysis: effect of feed characteristics and interfering ions. Desalination 223:63–72 Melnik L, Vysotskaja O, Kornilovich B (1999) Boron behavior during desalination of sea and underground water by electrodialysis. Desalination 124:125–130 Melnik L, Goncharuk V, Butnyk I, Tsapiuk E (2005) Boron removal from natural and wastewaters using combined sorption/membrane proces. Desalination 185:147–157 Melnik L, Goncharuk V, Butnyk I, Tsapiuk E (2007) Development of the sorption-membrane “green”

3 technology for boron removal from natural and wastewaters. Desalination 205:206–213 Oren Y, Linder C, Daltrophe N, Mirsky Y, Skorka J, Kedem O (2006) Boron removal from desalinated seawater and brackish water by improved electrodialysis. Desalination 199:52–54 Turek M, Dydo P, Ciba J, Trojanowska J, Kluczka J, Palka-Kupczak B (2005) Electrodialytic treatment of boron-containing wastewater with univalent permselective membranes. Desalination 185:139–145 Turek M, Dydo P, Trojanowska J, Bandura B (2007) Electrodialytic treatment of boron-containing wastewater. Desalination 205:185–191 Turek M, Bandura B, Dydo P (2008a) Electrodialytic boron removal from SWRO permeate. Desalination 223:17–22 Turek M, Bandura B, Dydo P (2008b) The influence of concentrate alkalinity on electrodialytic boron transport. Desalination 223:119–125 WHO (2009) Boron in drinking-water. Background document for development of WHO guidelines for drinking-water quality. World Health Organization. http://whqlibdoc.who.int/hq/2009/WHO_HSE_WSH _09.01_2_eng.pdf. Accessed 17 Jun 2012 Yazicigil Z, Oztekin Y (2006) Boron removal by electrodialysis with anion-exchange membranes. Desalination 190:71–78

B

Boron Removal by Reverse Osmosis Piotr Dydo Faculty of Chemistry, Silesian University of Technology, Gliwice, Poland

Reverse osmosis (RO) is a commonly applied membrane technique of desalination capable of producing permeates of drinking-water quality. In most cases, the boron content in the product should not exceed the WHO 2009 and 2011 guideline value of 2.4 mg/L. The effectiveness of boron rejection by reverse osmosis membranes is, however, strongly affected by its concentration in the feedwater and the aqueous chemistry of boron specie (Kabay et al. 2010; Hilal et al. 2011). In aqueous solutions, boric acid may exist in the form of boric acid, metaborates, and polyborates. Boric acid as a weak electrolyte (pKa = 9.2) dominates in diluted aqueous solutions of a pH less than 9. Since the pH of most naturally occurring waters is lower than 9, the rate of boric acid transport determines the effectiveness of boron removal by RO. According to Kabay et al. 2010, the rejection of boric acid by RO membranes is, however, low due to its small size and lack of electric charge. The reported boron rejection coefficients range from around 20 % in the case of old RO membranes to approx. 90 % in the case of modern RO membranes (WHO 2009; Kabay et al. 2010; Hilal et al. 2011). # Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_76-11

The effectiveness of boron removal from seawater by RO can be enhanced by adjusting the feedwater pH to above the value of 9.25. The monoborate anion, which dominates in diluted solutions at pH > 9.5, was found to be rejected more effectively than boric acid due to its larger size and discrete charge (Kabay et al. 2010; Hilal et al. 2011). The reported boron rejection coefficients at pH = 11 exceeds 99 % in the case of dense seawater reverse osmosis membranes. Also, in the case of boron-rich geothermal waters and industrial wastewaters, high boron rejection is observed only at pH > 10.5 (Koseoglu et al. 2010; Dydo et al. 2005). Furthermore, it was shown that the effectiveness of boron removal can be further enhanced by creating borate complexes (esters) with polyhydroxyl alcohols (Geffen et al. 2006; Dydo et al. 2012), although again, under alkaline conditions only. The results of mechanistic studies on boron rejection by reverse osmosis membranes conducted by Sagiv and Semiat (2004), Hyung and Kim (2006), Hun et al. (2009), and Tu et al. (2010) showed that permeabilities of boric acid are at least one order of magnitude larger than those of monoborate ion. It proves more intensive diffusion of boric acid than of borate across RO membranes. Also, the reflection coefficients for boric acid (0.995) as reported by Hyung and Kim (2006). This, in turn, indicates strong boric acid–water interaction.

2

Feedwater pH is said to have a dominant impact on boron rejection by RO membranes (Hilal et al. 2011). Among other RO parameters that affect boron rejection are: operating pressure (increase), feed temperature (decrease), feedwater salinity (decrease), and recovery (decrease). However, RO systems cannot be directly operated at high feedwater pH due to the possibility of membrane scaling with insoluble calcium and magnesium basic compounds. Therefore, several modifications to the operation of RO and its design have been proposed. In general, a cascade design for RO systems in which a fraction of the first stage permeate is treated in the following RO stages (if possible at elevated pH’s) and a blending of all the permeates produces water of the desired quality (Hilal et al. 2011; Faigon and Hefer 2008). This design was successfully applied, e.g., in the full-scale seawater treatment plant in Eliat.

Cross-References ▶ Boron Reduction ▶ Boron Removal by Electrodialysis ▶ Boron Removal with Ion-Exchange Membranes

References Dydo P, Turek M, Ciba J, Trojanowska J, Kluczka J (2005) Boron removal from landfill leachate by

Boron Removal by Reverse Osmosis means of nanofiltration and reverse osmosis. Desalination 185:131–137 Dydo P, Nems´ I, Turek M (2012) Boron removal and its concentration by reverse osmosis in the presence of polyol compounds. Sep Purif Technol 89:171–180 Faigon M, Hefer D (2008) Boron rejection in SWRO at high pH conditions versus cascade design. Desalination 223:10–16 Geffen N, Semiat R, Eisen MS, Balazs Y, Katz I, Dosoretz CG (2006) Boron removal from water by complexation with polyol compounds. J Membr Sci 286:45–51 Hilal N, Kim GJ, Somerfield C (2011) Boron removal from saline water: a comprehensive review. Desalination 273:23–35 Hun PVX, Cho S-H, Moon S-H (2009) Prediction of boron transport through seawater reverse osmosis membranes using solution-diffusion model. Desalination 247:33–44 Hyung H, Kim J-H (2006) A mechanistic study on boron rejection by sea water reverse osmosis membranes. J Membr Sci 286:269–278 Kabay N, Guler E, Bryjak M (2010) Boron in seawater and methods for its separation – a review. Desalination 261:212–217 Koseoglu H, Harman BI, Yigit NO, Guler E, Kabay N, Kitis M (2010) The effects of operating conditions on boron removal from geotermal waters by membranes processes. Desalination 258:72–78 Sagiv A, Semiat R (2004) Analysis of parameters affecting boron permeation through reverse osmosis membranes. J Membr Sci 243:79–87 Tu KL, Nghiem LD, Chivas AR (2010) Boron removal by reverse osmosis membranes in seawater desalination applications. Sep Purif Technol 75:87–101 WHO (2009) Boron in drinking-water. Background document for development of WHO guidelines for drinkingwater quality. World Health Organization. http:// whqlibdoc.who.int/hq/2009/WHO_HSE_WSH_09.01_2_ eng.pdf. Accessed 17 June 2012 WHO (2011) Guidelines for drinking-water quality, 4th edn. World Health Organization. http://whqlibdoc. who.int/publications/2011/9789241548151_eng.pdf. Accessed 17 June 2012

B

Boron Removal with Ion-Exchange Membranes Piotr Dydo Faculty of Chemistry, Silesian University of Technology, Gliwice, Poland

Apart from boron removal by electrodialysis, ion-exchange membranes are able to transport boron in the so-called Donnan dialysis (DD) or diffusion dialysis processes (Fig. 1). Ayyildiz and Kara (2005) examined the effectiveness of boron transport across anionexchange membranes in a DD process. They found that boron flux depends upon the membrane, concentration of boron in the feed solution, pH of the feed and receiving solution, presence of the accompanying ions in the feed solution, and the type of the carrier anion in the receiving solutions. The effect of pH of the feed solution was found to be complex. At high boron concentration (0.1 mol/L), maximum boron flux was observed at around the pH of 9.5, while in the case of a diluted solution (0.001 mol/L), maximum boron flux was observed at the maximum examined feedwater pH of 11.5. Such behavior was explained by the formation of polyborate ions at high boron concentrations and their absence in diluted solutions. The maximum rate of boron transport at the pH of around 9.5 in the case of concentrated boron solution was later confirmed by Yazicigil et al. in 2006. Ayyildiz # Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_77-5

and Kara in 2005 reported also that the pH of the receiving phase affects boron flux with its maximum at around a pH of 9.5. Moreover, the boronaccompanying anions, chlorides, bicarbonates, and sulfates, were found to affect the rate of boron transport in DD with a maximum observed in the presence of bicarbonate. On the other hand, the highest boron transport rates were observed with sodium chloride in the receiving solution. Neosepta AHA and AMH membranes produced similar fluxes of boron, while the superiority of AFN was clearly seen. In 2011, Kir et al. showed that plasma modification of the existing anionexchange membranes may result in a significant enhancement of the rate of boron transport in a DD process. The mechanistic study on boric acid transport across cation-exchange membranes (CEMs) in a DD process was presented by Goli et al. in 2010. It was reported that the membrane diffusion coefficients for boric acid depend not only upon the manufacturer of the CEM but also upon the type of the cation present in the membrane. The same behavior is reported for arsenite. In general, in the presence of monovalent cations in the membrane, higher fluxes of boron were observed than in the presence of divalent cations. Such a behavior was later confirmed by Dydo in 2012. The existing differences in boric acid membrane diffusion coefficients were rationalized by Goli et al. (2010) as being the result of the change in the mean viscosity of the solution confined in membrane pores.

2

Boron Removal with Ion-Exchange Membranes

Boron Removal with Ion-Exchange Membranes, Fig. 1 Boron Removal with Ion-Exchange Membranes

In 2007, Bryjak et al. proposed a Donnan dialysis-based method for the regeneration of finely divided boron selective resin (BSR) DOWEX XUS 43594.00. In this process, BSR slurry with boron adsorbed on it is fed into the DD feed compartment. It was assumed that there are always some minute amounts of borate in the feedwater at equilibrium with the BSR. These amounts were subject to transport into the receiving solution as a result of Donnan dialysis, which should ultimately result in complete boron removal from the BSR. The net effect of the process would be a regenerated resin in the chloride form and boron (borate)-rich receiving solution. As reported by Bryjak et al. (2007), considerably high boron fluxes were observed during the course of a DD regeneration of boron containing BSR. Boron desorption was found to be the phenomena that governs kinetics of the boron transport in the process. Unfortunately the tentative, mostly mechanistic report on DD boron removal presented in this chapter does not provide much information about

the effectiveness of boron removal from water nor its final concentration. It seems that a lot of work needs to be done to enhance the kinetics of the diffusive transport of borate during DD.

Cross-References ▶ Boron Reduction ▶ Boron Removal by Electrodialysis ▶ Boron Removal by Reverse Osmosis

References Ayyildiz HF, Kara H (2005) Boron removal by ion exchange membranes. Desalination 180:99–108 Bryjak M, Pozniak G, Kabay N (2007) Donnan dialysis of borate anions through anion exchange membranes: a new method for regeneration of boron selective resins. React Funct Polym 67:1635–1642 Dydo P (2012) The mechanism of boric acid transport during an electrodialytic desalination process. J Membr Sci 407–408:202–210

Boron Removal with Ion-Exchange Membranes Goli E, Hiemstra T, Van Riemsdijk WH, Rahnemaie R, Malakouti MJ (2010) Diffusion of neutral and ionic species in charged membranes: boric acid, arsenite, and water. Anal Chem 82:8438–8445 Kir E, Gurler B, Gulec A (n.d.) Boron removal from aqueous solution by using plasma-modified and

3 unmodified anion-exchange membranes. Desalination 267: 114–177 Yazicigil Z, Oztekin Y (2006) Boron removal by electrodialysis with anion-exchange membranes. Desalination 190:71–78

B

Bulk Biotech Industry Frank Lipnizki Alfa Laval Copenhagen, Søborg, Copenhagen, Denmark

The term “bulk biotech industry” is referring to the use of biological processes on industrial scale to produce bulk products. Examples of products from the bulk biotech industry are antibiotics, enzymes, organic and amino acids, vitamins, bioalcohols, and biopolymers. One of the earliest biotechnological processes adopted by humans is the production of alcohol by fermenting fruits around 5000–6000 years ago. The production of lactic acid by Pasteur in 1857 is often considered to be the beginning of modern biotechnology followed by the industrial scale production of citric acid by Pfizer in 1923 (Chotani et al. 2007). The first wave of biotechnology started with the discovery of penicillin by Fleming in 1928 paving the way for the industrial scale production of antibiotics and amino acids. The discovery and increased understanding of the DNA created the foundation to use molecular engineering to recombine DNA leading to the second wave of biotechnology processes in the 1980s. The current third and so far final wave of biotechnology – the white biotechnology – aims to replace the C2/C3 chemistry based on fossil fuels such as oil and gas by biotechnological processes. It is foreseen that in the near future, # Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_82-1

up to 20 % of all chemical products with a market value of approx 250 € billion will be produced by biotechnology. Approximately 60 % of the products produced by white biotechnology will be intermediated chemicals used in the pharmaceutical industry, and the remaining 40 % will be biopolymers and special chemicals for various industries (Festel et al. 2004). Since the 1970s, cross-flow membrane processes have established themselves in the downstream processing of the biotechnology industry for the recovery and purification of the products. It is foreseen that membrane processes will also play an important role in white biotechnology and the related concept of biorefineries. Similar to petroleum refineries, biorefineries are aiming for the full utilization of biomass for the simultaneous production of biofuels, biochemicals, heat, and power (Axega˚rd 2005). The integrated production of biomaterials can be based on, e.g., sugar, starch, and cellulose-based feedstock and as such extend current sugar, starch, and pulp factories. Membrane processes as highly selective and energy-saving processes are well suited to play an important part in biorefineries and thus the white biotechnology. The key membrane technologies for the bulk biotech industry are microfiltration (MF), ultrafiltration (UF), nanofiltration (NF), and reverse osmosis (RO). Other membrane technologies such as membrane contactors (MC), electrodialysis (ED), pervaporation (PV), and vapor permeation (VP) plus membrane bioreactors for

2

continuous fermentation are less established but have, nevertheless, the potential to become increasingly important.

References Axega˚rd P (2005) The future pulp mill – a biorefinery? Presentation at the 1st international biorefinery workshop, Washington, DC

Bulk Biotech Industry Chotani GK, Dodge TC, Gaertner AL, Arbige MV (2007) Industrial biotechnology: discovery to delivery. In: Kent JA (ed) Kent and Riegel’s handbook of industrial chemistry and biotechnology, 11th edn. Springer, Berlin Festel G, Kno¨ll J, Go¨tz H, Zinke H (2004) Der einfluss der biotechnologie auf produktionsverfahren in der chemieindustrie. Chem Ing Technol 76:307–312

B

Bulk Liquid Membrane Vladimir S. Kislik Campus Givat Ram, Casali Institute of Applied Chemistry, The Hebrew University of Jerusalem, Jerusalem, Israel

Bulk liquid membrane (BLM) consists of a bulk aqueous feed and receiving phases separated by a bulk organic, water-immiscible liquid phase. The feed and receiving phases may be separating from the LM by microporous supports or may be without supports (layered BLM). Many technologies that were developed and tested in the last two decades have to be included in the BLM group. These are hybrid liquid membrane (HLM), hollow-fiber liquid membrane (HFLM), hollow-fiber-contained liquid membrane (HFCLM), pertraction, flowing liquid membranes (FLM), membrane-based extraction and stripping, multimembrane hybrid system (MHS), and membrane contactor systems (see entry “▶ Liquid Membranes”). All these systems are based on membrane-based nondispersive (as the means for blocking the organic reagent from mixing with the aqueous feed and strip solutions) selective extraction coupled to permselective diffusion of soluteextractant complexes and selective stripping of the solute in one continuous dynamic process (see Fig. 1). A great number of terms for similar bulk LM processes confuse the readers. The terms vary by membrane type used (hollow fiber, flat neutral, # Springer-Verlag Berlin Heidelberg 2012 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_83-1

ion-exchange sheets) or by module design. All abovementioned bulk LM processes with waterimmiscible organic liquid membrane solutions may be unified under the term bulk organic hybrid liquid membrane (BOHLM) systems (for more details, see Kislik 2010a). Bulk LM processes with water-soluble carriers are defined as bulk aqueous hybrid liquid membrane (BAHLM) systems (for more details, see Kislik 2010b). Regenerable water-soluble polyionic complexants are used as suitable aqueous liquid membrane carriers. These polyelectrolytes typically have a very high effective concentration of charged groups and could constitute highly selective complexants. The BAHLM technology is based on a combination of liquid membrane (LM) process and dialysis (D) in the case of neutral hydrophilic membranes or Donnan dialysis (DD) in the case of ion-exchange membranes used. BOHLM and BAHLM technologies achieve the necessary transport and selectivity characteristics to have potential for commercial applications. Applications of the BOHLM processes are mainly in metal separation, wastewater treatment, biotechnologies, drugs recoveryseparation, organic compounds, and gas separation. Selective separations of alkali, alkali earth, rare earth, heavy metal ions, precious metals, etc., are studied by many authors using all abovedescribed techniques. Recovery and separation of carboxylic and amino acids from fermentation broth have been tested using layered BLM, rotating, creeping, spiral-type FLM, HFLM, HLM,

2

Bulk Liquid Membrane

Bulk Liquid Membrane, Fig. 1 Schematic transport models of a bulk organic hybrid liquid membrane (BOHLM) system with a, d hydrophobic membranes and b, e hydrophilic or ion-exchange membranes

and MHS-PV techniques of the BOHLM processes. Separation of ethylene, benzene, propanol, olefin, and aromatic amines from organic liquid mixtures and of volatile organic compounds (VOC) and phenol from wastewater was investigated using a rotating film module, spiral-type FLM, and hollow-fiber and layered LM techniques. High separation factors (>1,000) in pilot- and industrial-scale experiments were found. In the last two decades, BOHLM techniques were intensively used in analytical chemistry for separation and preconcentration of metals, organic acids, organic, and pharmaceutical compounds. BAHLM is a new technology and few studies are known up to date. In metal separations coppercadmium recovery from chloride aqueous solutions and cadmium, copper, and zinc separation from wet-process phosphoric acid are studied. BAHLM systems were tested for the separation of carboxylic acids such as lactic, citric, and acetic or their anions. Continuous separation of different isomeric mixtures of organic compounds has been studied by means of a hollow-fiber-contained liquid membrane, HFCLM.

In recent years, integrated hybrid systems incorporating two or more functions in one module, for example, biotransformation and separation, become of great interest to researchers. The BOHLM systems, integrating reaction, separation, and concentration functions in one apparatus (bioreactor) attracted great interest in the last few years. Bioreactors combine the use of specific biocatalyst for the desired chemical reactions, with repeated or continuous application of it under very specific conditions. Such techniques were termed hybrid membrane reactors.

References Kislik V (2010a) Bulk hybrid liquid membrane with organic water-immiscible carriers: application to chemical, biochemical, pharmaceutical, and gas separations. In: Kislik V (ed) Liquid membranes principles and applications in chemical separations & wastewater treatment, 1st edn. Elsevier, Amsterdam, pp 201–276, Ch. 5 Kislik V (2010b) Bulk Aqueous Hybrid Liquid Membrane (BAHLM) processes with water-soluble carriers: application in chemical and biochemical separations. In: Kislik V (ed) Liquid membranes principles and applications in chemical separations, 1st edn, Elsevier, Ch. 6, pp 277–326

C

Capillary Andras Koris Corvinus University of Budapest, Budapest, Hungary

The word capillary originated from the Latin adjective capillaris, which means “pertaining to the hair.” Possibly, the scientific phenomenon was first observed between contiguous hairs, for example, within a paintbrush. In medicine and biology, capillary is the smallest of a body’s blood vessel and is part of the microcirculation. The word capillary, in the nonmedical sense, means narrow tube. In membrane science the noun capillary is both used to name the inner spaces of a porous media and to specify a typical membrane (see ▶ Capillary Membrane). The classification of porous membrane filter types is mainly based on the nominal diameter of the capillaries in the membrane. Artificial capillaries in porous media

# Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_98-1

are produced by interfacial polymerization, or the natural porosity in solid materials is adjusted to the desirable nominal diameter. The shape and size of the pores in the membrane are very diverse. The pore theory of membrane transport through the capillaries considers that both convection and diffusion contribute to solute transport across the porous membrane, the two processes being impeded by steric hindrance at the entrance of the pores and by frictional forces within the pores; the same steric hindrance and frictional resistance terms were used in the convection and diffusion terms of the solute transport equation (Pappenheimer et al. 1951).

Reference Pappenheimer JR, Renkin EM, Borrero LM (1951) Filtration, diffusion and molecular sieving through peripheral capillary membranes. Am J Physiol 167:13

C

Capillary Flow

Andras Koris Corvinus University of Budapest, Budapest, Hungary

Capillary flow means the movement of the fluid in the internal gaps and on the surface of the solid material, where the driving force is the molecular interaction between the solid and fluid material (see “▶ Capillary Force”). If the fluid is incompressible and Newtonian, the flow is laminar through a pipe of constant circular cross section that is substantially longer than its diameter, and since there is no acceleration of fluid in the pipe, the total flow rate (Q) could be estimated according to Poiseuille’s law (Briant et al. 1989): Q¼

d4 p DP 128 m L

where d represents capillary diameter, DP is the total pressure drop along the channel, m is the fluid viscosity, and L represents the capillary length. For velocity calculation, let us consider a laminar horizontal flow through a straight circular channel of length L and a radius r. The mean flow velocity u through the channel is given by (Landau and Lifshitz 1987):

# Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_99-3

r 2 DP 8m L

where DP is the total pressure drop along the channel and m is the fluid viscosity. In the general case of two immiscible fluids: u¼

r 2 DP Pc 8meff L

where x Lx ma meff ¼ mw þ L L is the effective viscosity; mw and ma are viscosities for phase w and phase a fluids, respectively; x is the location of the interface measured from the inlet; Pc is the capillary pressure across the interface; and a0 is a constant associated with the channel cross-sectional shape. Assuming that the fluid system is in static equilibrium, the capillary pressure at an interface in a circular channel is (Young-Laplace eq.): Pc ¼

4g cos y dc

where g represents the interfacial tension, y is the contact angle, and dm represents the capillary diameter. Since the determination of the number of active capillaries during membrane filtration is

2

Capillary Flow

very difficult, statistical approach from Darcy can be applied basically. Darcy’s law is a simple proportional relationship between the instantaneous discharge rate through a porous medium, the viscosity of the fluid, and the pressure drop over a given distance: Q¼

k A DP mL

where Q is overall flow rate, k represents permeability of the medium, A is the cross-sectional area to flow, DP is the pressure drop, m represents viscosity, and L is the length over which the pressure drop is taking place. Due to the concentration polarization phenomenon, gel layer forming, membrane fouling, and pore blocking, the global model was further developed for different membrane techniques, see, e.g., resistances-in-series model or solutiondiffusion model. The diffusion of gasses, liquids, and solids in solids is quite important in mass transfer operations. The diffusion is greatly affected by the size and shape of pores and capillaries. For example, in porous solids, the effective diffusion of salts (A) in water (B) is described as:

D_f_A, eff g ¼ DAB

ϵ k2

where DAB represents the diffusivity of salt in water, e is the porosity, and k is a correction factor for the distance (when it is nonlinear; general value = 1.2–2.5). The diffusion of a liquid in capillaries has three types: Knudsen diffusion, molecular (Fick’s) diffusion, and transition region diffusion (Geankoplis 1972). The power-law is an improvement to the Poiseuille’s equation by generalizing the flow to include non-Newtonian effect. The beauty of this model is its simplycity. However there are other aspects of the flow properties of fluids which this model fails to examine, one of them being the presence of yield stress (Mazumdar 1992)

References Briant J, Guy Parc JD (1989) Rheological properties of Lubricants, ch. 3., p. 69-71 Geankoplis CJ (1972) Mass transport phenomena. Rinehart and Winston, Holt Landau LD, Lifshitz EM (1987) Fluid mechanics, 2nd edn. Pergamon Press, Oxford Mazumdar J (1992) Biofluid Mechanics, p. 86-87, World Scientific

C

Capillary Force Andras Koris Corvinus University of Budapest, Budapest, Hungary

Capillary force is the force which generates movement of a liquid along the surface of a solid caused by the attraction of molecules of the liquid to the molecules of the solid. For example, molecules of water are naturally attracted to each other and form temporary hydrogen bonds with each other; their attraction for each other on the surface of a liquid, for example, gives rise to surface tension. But they are also attracted in a similar way to other molecules, called hydrophilic molecules, such as those in the sides of a narrow glass tube inserted into water. These attractive forces can draw water up against the force of gravity to a certain degree. Where three different fluid interfaces are contacting each other along one line, equilibrium exists only when the vector polygon of the stresses is closed. This requirement determines the angle between the edges of disjunctive surfaces (Pattantyu´s 1961) (Fig. 1). Sometimes |C1–3| > |C1–2| + |C2–3| occurs. Because vector C1–3 is always larger, equilibrium is not possible; therefore, fluid no. 2 is permanently moving on the no. 1 and 3 fluid

# Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_100-2

interface. For example, mineral oil unfolds the water surface even to molecule thick layer when the available area is sufficient. One angle of the vector triangle is given in cases where instead one of the liquids a solid material presents. In Fig. 2 this angle is 180 . The requirement of equilibrium: C1–3 = C2–3 + C1–2*cosa. In case of solid particle or material, |C1–3| > |C1–2| + |C2–3| can also exist sometimes, and no. 2 liquid could totally cover the solid surface (1). This movement is usually limited by impurities on the solid surface. For mercury–air–glass system, |C1–3| < |C2–3| and a > 90 . The Young–Laplace equation is a nonlinear partial differential equation that describes the capillary pressure difference or capillary force sustained across the interface between two static fluids, due to the phenomenon of surface tension or wall tension, although usage on the latter is only applicable if assuming that the wall is very thin. The Young–Laplace equation relates the pressure difference to the shape of the surface or wall, and it is fundamentally important in the study of static capillary surfaces. It is a statement of normal stress balance for static fluids meeting at an interface, where the interface is treated as a surface (scpacp).

2

Capillary Force

References Sályi I (ed) (1961) Pattantyu´s gépész és villamosmérno¨ko¨k kéziko¨nyve, 2. Alaptudományok – Anyagismeret. Mu˝szaki Ko¨nyvkiado´

Capillary Force, Fig. 1 Contact of three fluids

Capillary Force, Fig. 2 Contact when one solid phase is present next to two fluids

C

Capillary Membranes Andras Koris Corvinus University of Budapest, Budapest, Hungary

Artificial capillary membranes are one type of the tubular-shaped membrane filters. With capillary membranes, the membrane serves as a selective barrier, which is sufficiently strong to resist filtration pressure. Because of this, the flow through capillary membranes can be both inside out and outside in. The diameter of capillary membranes is much smaller than that of tubular membranes, namely, 0.5–5 mm. Because of the smaller diameter, the chances of plugging are much higher with a capillary membrane. A benefit is that the packing density is much greater. The first capillary membrane was fabricated from polymers, but nowadays the technology development enables the production of inorganic capillary modules (Fig. 1).

Description of a Capillary Membrane Module A complete description of a membrane module requires the simultaneous solution of local transport equations that describe the flow and transport conditions. In a capillary membrane module with # Springer-Verlag Berlin Heidelberg 2015 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_101-2

a permeable membrane wall, three regions of flow should be considered: flow in the lumen, flow within the membrane matrix, and flow in the extracapillary space. The pressure distribution in the membrane is determined by application of the overall balance of linear momentum. The utilization of this conservation principle is complicated by the fact that an external force must be applied to keep the polymer membrane stationary. In order to describe the behavior of a membrane module, three submodels are required: two that describe the transport on either side of the membrane and a third model that characterizes the separation properties of the membrane and any porous support material (Nagy 2012).

Membrane Development Besides polymeric and ceramic capillary membranes, other types are also developing due to its benefits. For example, supported zeolite membranes have been synthesized under microwave heating in order to reduce synthesis time in the work of Sebastian et al. (2010) to prevent support dissolution and to reproducibly obtain a thin defect-free zeolite layer. The MFI-type zeolite membranes were synthesized on ceramic capillaries, with a high membrane surface area-tovolume ratio (>1,000 m2 m 3), which is by far higher than that of classical tubular supports (10 for gases and >1000 for liquids) have been obtained with these membranes. However, the small area (typically in the order of few mm2), the long and complex fabrication process, and the poor mechanical stability of these type of CNT membranes, limit the practical applicability of these interesting nanostructured systems. On the contrary, more easy is the scale up of CNT membranes in which the CNTs are mixed with a polymer in the form of mixed matrix membranes. The main limitations of these membranes are related to the poor dispersion of the CNTs in the polymeric matrix and the moderate increase of the membrane performance. It is interesting to note that graphitic carbon nanomaterials like CNTs are usually introduced at lower content (2 wt%) in mixed matrix membranes in comparison with three-dimensional inorganic nanofillers like TiO2 and ZrO2 (usually blended at loading 5 wt%, up to 60 wt%), thank to their high specific surface, elevated aspect ratio, and the intrinsic properties of graphitized structure. In addition, the relatively easy functionalization of the surface of carbon nanomaterials render them ideal candidate to tailor the polymer/nanofiller interface. CNTs were succesfully entrapped in mixed matrix membranes made of various polymeric materials by several techniques including:

2

Carbon Nanotube Membranes (CNT Membranes)

Carbon Nanotube Membranes (CNT Membranes), Fig. 1 Schematic honeycomb structure of a graphene sheet (sp2 hybridized carbon network) (a) and image of single-, double-, and multiwalled carbon nanotubes (SWCNTs, DWCNTs, MWCNTs), (b)

a

m

n

b

SWCNT

dispersion in the casting solution and successive phase separation, entrapping in the membrane pores using a polymer binder, in situ crosslinking of a polymer matrix around oriented CNTs (Ismail et al. 2009). The resulting mixed matrix CNT membranes offer, in many cases, relevant advantages in comparison with the polymeric samples. Poly(vinyl alcohol) (PVA)/MWCNTs membranes were realized for pervaporation application, obtaining significant improvement in Young’s modulus and thermal stability, as compared to pure PVA membranes (Peng et al. 2007). The entrapment of MWCNTs in polyethersulfone (PES) membranes reduced fouling problems in water treatment (Celik et al. 2011). Mixed matrix membranes consisting of sulfonated carbon nanotubes (sCNTs) and sulfonated poly (ethersulfone ether ketone ketone) (SPESEKK) were also fabricated via the solution casting method (Zhou et al. 2011). The proton conductivity of the SPESEKK membrane increased while the methanol permeability decreased as the sCNTs content increased. MWCNTs were covalently linked to aromatic polyamide (PA) membranes by a polymer grafting process (Shawky et al. 2011). Measurements of mechanical properties of this composite showed an increased

DWCNT

MWCNT

membrane mechanical strength relative to the PA polymeric membranes. Mixed matrix polyimide membranes were also prepared using functionalized MWCNTs (aminated or oxidized). The MWCNTs functionalization improved their dispersion in the casting solution in comparison to pristine MWCNTs and, as a consequence, in the formed membranes. The membranes containing functionalized MWCNTs showed better performance in the rejection of dyes (enhanced flux and reduced fouling, with similar or higher rejection), with respect to reference polymeric membranes (without MWCNTs). These results were attributed to the formation of low-resistance pathways for solvent transport at the interface between the MWCNTs and the polymeric chains. Moreover, the MWCNTs reduced the severe membrane fouling caused by the absorption of the positively charged dye Safranine (used as a model of organic pollutant) with respect the polymeric membrane, by a screening effect of the attractive electrical interactions between the Safranine and the membrane surface characterized by a negative value of zeta potential (Grosso et al. 2014). The presence of oxygen-containing polar groups on oxidized MWCNTs, resulted also in a

Carbon Nanotube Membranes (CNT Membranes)

3

Carbon Nanotube Membranes (CNT Membranes), Table 1 Different forms of a SWCNT and its electrical properties (n, m) (n, 0)

Form Zigzag

Electrical conductivity Metallic when n is multiple of 3, otherwise, semiconducting

(n, m) with n = m

Armchair

Metallic

(n, m) with m 6¼ 0 and n

Chiral

Metallic when (n-m)/3 is an integer, otherwise, semiconducting

good dispersion in polyvinylidene difluoride (PVDF) membranes, allowing the formation of mixed matrix membranes with a lower fouling tendency in comparison with the bare polymeric samples (Fontananova et al. 2014). MWCNTs were also immobilized in the pores of a hydrophobic membrane improving the watermembrane interactions to promote vapor permeability in membrane distillation process (Gethard et al. 2011). In this case, the CNTs dispersion was forced under vacuum into the pores of a polypropylene (PP) membrane, using PVDF as binder.

References Celik E, Park H, Choi H, Choi H (2011) Carbon nanotube blended polyethersulfone membranes for fouling control in water treatment. Water Res 45:274–282 Dai H (2002) Carbon nanotubes: synthesis, integration, and properties. Acc Chem Res 35:1035–1044 Fontananova E, Bahattab MA, Aljlil SA, Alowairdy M, Rinaldi G, Vuono D, Nagy BJ, Drioli E, Di Profio G (2014) From hydrophobic to hydrophilic polyvinylidenefluoride (PVDF) membranes by gaining new insight into material’s properties. RSC Adv 5:56219–56231 Gethard K, Sea-Know O, Mitra S (2011) Water desalination using carbon-nanotube-enhanced membrane distillation. Appl Mater Interfaces 3:110–114 Grosso V, Vuono D, Bahattab MA, Di Profio G, Curcio E, Al-Jilil SA, Alsubaie F, Alfife M, Nagy BJ, Drioli E, Fontananova E (2014) Polymeric and mixed matrix polyimide membranes. Sep Purif Technol 132:684–696

4 Hummer G, Rasaiah JC, Noworyta JP (2001) Water conduction through the hydrophobic channel of a carbon nanotube. Nature 414:188–190 Ismail AF, Goh PS, Sanip SM, Aziz M (2009) Transport and separation properties of carbon nanotube-mixed matrix membrane. Sep Purif Technol 70:12–26 Kalra A, Garde S, Hummer G (2003) Osmotic water transport through carbon nanotube membranes. Proc Natl Acad Sci U S A 100:10175–10180 Majumder M, Chopra N, Bruce JH (2011) Mass transport through carbon nanotube membranes in three different regimes: ionic diffusion and gas and liquid flow. ACS Nano 5:3867–3877

Carbon Nanotube Membranes (CNT Membranes) Peng F, Hu C, Jiang Z (2007) Novel ploy(vinyl alcohol)/ carbon nanotube hybrid membranes for pervaporation separation of benzene/cyclohexane mixtures. J Membr Sci 297:236–242 Shawky HA, Chae S-R, Lin S, Wiesner MR (2011) Synthesis and characterization of a carbon nanotube/polymer nanocomposite membrane for water treatment. Desalination 272:46–50 Zhou W, Xiao J, Chen Y, Zeng R, Xiao S, Nie H, Li F, Song C (2011) Sulfonated carbon nanotubes/sulfonated poly (ether sulfone ether ketone ketone) composites for polymer electrolyte membranes. Polym Adv Technol 22:1747–1752

C

Carbon Porous Membranes Modified electropolymerization at +0.9 V vs. Ag/AgCl. with Enzymes The oxidation of pyrrole allowed the entrapment Christophe Innocent cc 407, UMR 5635 CNRS - ENSCM - UM II, Institut Europe´en des Membranes, Montpellier Cedex 5, France

The carbon porous tubes (from Novasep-OrelisFrance) have been characterized by average pore diameter of 3 mm and an inner diameter of 0.6 cm (Fig. 1). The immobilization of enzyme on the carbon tube surface has been carried out by dipping in enzyme solution and then addition of pyrrole monomer just before starting the anodic

# Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_112-5

of enzyme in polypyrrole coated on the electrode surface. A glucose/O2 biofuel cell is based on tubular cathode (coated with laccase/ABTS on the external surface for dioxygen reduction) and the tubular anode (coated with glucose oxidase enzyme (GOD) and hydroxyquinoline sulfonic acid as mediator) on the internal surface for glucose oxidation. The two electrodes were soaked in an unstirred 10 mM glucose solution. Glucose diffused through the polypyrrole film to be oxidized by the GOD at the anode. In order to prevent the presence of dioxygen in the vicinity of GOD, dissolved oxygen was supplied to the system by

2

Carbon Porous Membranes Modified with Enzymes

Carbon Porous Membranes Modified with Enzymes, Fig. 1 Photography of carbon tube (photo IEM)

circulating through the inner cavity of the biocathode, then before being reduced at the external surface by the BOD. Cylindrical and porous carbon tubes were used as original conducting support for the compartment of the bioelectrodes, for enzyme immobilization and transport of dissolved oxygen via diffusive flow through the porosity (Fig. 2). The resulting glucose/oxygen biofuel cell is operated at a maximum power density of 42 mWcm 2at 0.3 V and 37 C in phosphate buffer pH 7.4.

Carbon Porous Membranes Modified with Enzymes, Fig. 2 Scheme of concentric biofuel cell (thesis of G. MERLE, University Montpellier 2, 2008)

C

Catalytic Membranes Jose M. Sousa Chemistry Department, School of Life and Environment Sciences, University of Tras-osMontes e Alto Douro, Vila Real, Portugal LEPABE - Laboratory for Process Engineering, Environment, Biotechnology and Energy, Chemical Engineering Department, Faculty of Engineering, University of Porto, Rua Roberto Frias, Porto, Portugal

A catalytic membrane is a membrane with catalytic properties. The catalytic activity can be intrinsic to the material itself, as in the case of membranes made of Pd, TiO2, and H-ZSM-5 zeolite, which are catalytic for specific reactions, or can be obtained by coating the external or the internal (porous) surfaces of the membrane with the catalyst (metal or oxides) or even by occluding the catalyst (metal nanoclusters, zeolites, activated carbon, metal complexes) inside a dense polymer matrix (Gryaznov 1986, 1992; Irusta et al. 1998; Gobina and Hughes 1994; Saracco and Specchia 2000; Itoh 2000; Dittmeyer et al. 2001; Piera et al. 2001; Julbe 2005; van Dyk et al. 2003; Marcano and Tsotsis 2002; Basile 2012; Fritsch and Peinemann 1995). Catalytic membranes can be inorganic (metallic, ceramic, or carbon made) or polymeric in their nature. Examples of dense inorganic catalytic membranes are the ones made of # Springer-Verlag Berlin Heidelberg 2015 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_116-6

palladium, which are catalytically active for hydrogenation reactions (though presenting low activity), while porous catalytic inorganic membranes can be made of a variety of materials, namely, alumina, silica, titania, and zeolites among others. Porous polymeric membranes are usually made, for example, of polysulfone, polyacrylonitrile, or polypropylene, while dense polymeric membranes are prepared from PDMS (polydimethylsiloxane) and other silicones, PVA (polyvinyl alcohol), perfluoropolymers, polyimides, or polyamides among other polymers (Basile and Gallucci 2011; Basile 2012). Membranes in general can be permselective or nonpermselective. Catalytic permselective membranes are characterized by two important parameters concerning their separation ability: Permselectivity, which describes the ability of the membrane to transport the different components of a mixture at different rates, and the permeability, which quantifies the total amount permeated by the membrane when subjected to specific operation conditions. Nonpermselective membranes can be described by their permeability, but they are not able to discriminate between the different components of a mixture (Marcano and Tsotsis 2002). The mass transport mechanism suitable to describe the permeation rate through a (catalytic) membrane is primarily a function of the type of membrane. For dense membranes, the accepted model is the sorption (solution) diffusion (the permeant species in gas or liquid phase

2

Catalytic Membranes

Catalytic Membranes, Fig. 1 Porous catalytic membrane for wastewater treatment (Reprinted with permission from Raeder 2010)

sorbs into the membrane at the higher chemical potential side, diffuses through the membrane subjected to the driving force gradient, and desorbs on the opposite side). These membranes are usually permselective. Concerning porous membranes, the total mass transport results from different contributions, namely, a diffusive component (activated, surface, Knudsen and bulk) and a viscous component (Poiseuille flow). The contribution of each one depends on the operating conditions and on the pore’s size. For pore radius between 1 and 50 nm, Knudsen diffusion is normally the predominant mass transfer mechanism (Knudsen diffusion is characterized by the meanfree path of the traveling species being much larger than the pore radius and is independent of any pressure gradient along the pore), and the membranes present some permselectivity, related with the ratio of the species molar mass. For membranes with larger pores, namely, macroporous, the mass transfer mechanism is a viscous flow if a total pressure gradient across the porous membrane is present or bulk diffusion if not. Neither of these situations allows any permselectivity. When the membrane has micropores and the species adsorb significantly, as in the case of zeolites, surface diffusion has a relevant contribution. Also in these cases the membranes are permselective (Basile and Nunes 2011). A catalytic membrane is used to carry out different types of reactions, namely, chemical, biochemical, electrochemical, or photocatalytic, in a so-called membrane reactor. The type of

membrane to be used, porous or dense, polymeric or inorganic, etc., depends on several factors, namely, the type of reaction. For example, dense polymeric catalytic membranes are used for low-temperature reactions where the permselectivity is important for shifting the conversion beyond the thermodynamic equilibrium value based on feed conditions (e.g., selective removal of water in esterification reactions) or to enhance the reaction selectivity (e.g., hydration, epoxidation, isomerization, and hydrogenation reactions). Dense polymeric catalytic membranes can also be used as contactors for promoting reactions between two immiscible phases with segregated feed (e.g., oxyfunctionalization of hydrocarbons with hydrogen peroxide reactions) (Basile and Gallucci 2009). Another type of reaction that has been studied is the photocatalytic removal of contaminants from water, either using porous or dense polymeric catalytic membranes. The catalyst typically used is TiO2, occluded or coated, but also occluded Fe in an H2O2 oxidant medium (photoFenton process) has been considered (Basile and Nunes 2011). Catalytic nonpermselective porous membranes, polymeric or inorganic, can be used to improve the contact between the reactants and the catalyst, in order to obtain higher reaction conversions. Particularly in the case of triphasic (gas/liquid/solid) reactions, usually limited by the diffusion of the gaseous reactant, porous membrane reactors show conversion advantages. By controlling the pressure of the gas and liquid

Catalytic Membranes

flows, it is possible to shift the reactants to meet the catalyst zone (Fig. 1). The same segregated feeding strategy can also be used in cases of reactions where a strict control of the reactants is important, namely, in the case of very fast or highly exothermic reactions. In the case of reactions where a complete conversion of some components, like VOCs, is of high importance, porous catalytic membranes, permselective or not, are usually used in a “flowthrough” mode, that is, all the reactants are fed conjointly. In these cases, it is critical controlling the residence time of the reactants to guarantee a complete conversion (Westermann and Melin 2009).

References Basile A (ed) (2012) Handbook of membrane reactors. Woodhead, Cambridge, UK (in press) Basile A, Gallucci F (eds) (2009) Modeling and simulation of membrane reactors. Nova, New York Basile A, Gallucci F (eds) (2011) Membranes for membrane reactors: preparation, optimization and selection. Wiley, New York Basile A, Nunes S (eds) (2011) Advanced membrane science and technology for sustainable energy and environmental applications. Woodhead, Cambridge, UK Dittmeyer R, Hollein V, Daub K (2001) Membrane reactors for hydrogenation and dehydrogenation processes based on supported palladium. J Mol Catal A Chem 173:135–184 Fritsch D, Peinemann K-V (1995) Catalysis with homogeneous membranes loaded with nanoscale metallic clusters and their preparation. Catal Today 25:277–283 Gobina E, Hughes R (1994) Ethane dehydrogenation using a high-temperature catalytic membrane reactor. J Membr Sci 90:11

3 Gryaznov VM (1986) Hydrogen permeable palladium membrane catalysts. Platinum Met Rev 30:68 Gryaznov VM (1992) Platinum metals as components of catalyst-membrane systems. Platinum Met Rev 36:70 Irusta S, Pina MP, Menendez M, Santamaria J (1998) Development and application of perovskite-based catalytic membrane reactors. Catal Lett 54:69 Itoh N, Haraya K (2000) A carbon membrane reactor. Catal Today 56:103 Julbe A (2005) Zeolite membranes – a short overview. In: Cejka J, van Bekkum H (eds) Zeolites and ordered mesoporous materials: progress and prospects. Studies in surface science and catalysis, Elsevier science & technology, vol 157. p 135 Marcano JS, Tsotsis TT (2002) Catalytic membranes and membrane reactors. Wiley, Weinheim Piera E, Tellez C, Coronas J, Menendez M, Santamaria J (2001) Use of zeolite membrane reactors for selectivity enhancement: application to the liquid-phase oligomerization of i-butene. Catal Today 67:127 Raeder H (2010) Wastewater oxidation using catalytic contactor – a revolutionary catalytic membrane reactor for wastewater treatment. SINTEF Materials and Chemistry. https://www.sintef.no/globalassets/ upload/materialer_kjemi/energikonvertering-ogmaterialer/dokumenter/watercatox-screen.pdf Saracco G, Specchia V (2000) Catalytic combustion of propane in a membrane reactor with separate feed of reactants. IV. Transition from the kinetics- to the transport-controlled regime. Chem Eng Sci 55:3979 van Dyk L, Miachon S, Lorenzen L, Torres M, Fiaty K, Dalmon JA (2003) Comparison of microporous MFI and dense Pd membrane performances in an extractortype CMR. Catal Today 82:167–177 Westermann T, Melin T (2009) Flow-through catalytic membrane reactors – principles and applications. Chem Eng Process 48:17–28

Further Reading Seidel-Morgenstern A (ed) (2010) Membrane reactors: distributing reactants to improve selectivity and yield. Wiley, New York

C

Catalytic Reactions, Membrane Operations of Adelio Mendes Faculty of Engineering, University of Porto, Porto, Portugal

Catalytic reactions (chemical, biochemical, photochemical, electrochemical, and photoelectrochemical) are promoted or enhanced in their rate by a catalyst. Most of the chemical catalytic reactions are heterogeneous; the catalyst is normally a solid, while reactants are in fluid phase. In the petrochemical industry, for example, the catalyst is usually composed of micro- or nanoclusters of a noble metal in a support, usually a metal oxide. In the fine chemical and organic synthesis, on the other hand, it is common that the reactions be homogeneous. Membrane is a permselective medium or interface between two fluid phases. Membrane processes can be synergistically combined with catalytic reactors targeting conversion, selectivity, or safety enhancements. These effects may be obtained by selective product extraction and purification, normally originating conversion and selectivity enhancements, or by segregated or distributed feed of reactants, aiming selectivity and/or safety enhancements. When a reactor is synergistically combined with a membrane process, the unit is called membrane reactor. Usually, # Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_118-3

membrane and reactor are integrated in the same housing (Marcano and Tsotsis 2002). Membranes can be inorganic (metallic, ceramic, carbon) or polymeric in their nature. Membranes can be permselective or non-permselective. Permselective membranes are characterized by two important parameters: permselectivity, which describes the ability of the membrane to transport the different components of a mixture at different rates, and permeability, which quantifies the total amount permeated by the membrane when subjected to specific operation conditions, normalized by the membrane thickness. The mass transport mechanism is described by the sorption – diffusion models for dense and microporous membranes. For porous, different contributions should be considered: diffusive (activated, surface, Knudsen, and bulk) and viscous (Poiseuille flow). The contribution of each one depends on the operating conditions, pore size, penetrant mass and size and the surface, and penetrant nature (Basile and Gallucci 2011). There are various potential industrial applications that take advantage of combining a catalytic reactor with a membrane process, integrated or not in the same device, operating either in liquid and/or gas phases. One of the main membrane studied functions has been the selective removal of components from the reaction medium. Biorefining and biofuel production, for example, take advantage of process integration of a bioreactor and a permselective membrane, as is the

2

case of bioethanol and acetic acid production: the continuous removal of the main product from the reaction medium decreases or even eliminates the potential reaction inhibition (Ma et al. 2009; He et al. 2012). Also the removal of fermentation inhibitors generated during the pretreatment process of lignocellulosic material for the secondgeneration bioethanol production is important, because of the negative impact in the ethanol yield and productivity and in the cell growth inhibition (He et al. 2012). Still in the selective removal of a reaction product from the reaction medium, it can be referred to as the selective removal of water in esterification reactions and the selective removal of hydrogen in dehydrogenation and water-gas shift reactions for a conversion shifting beyond the thermodynamic equilibrium based on feed conditions. In case of using palladium-silver membranes in the dehydrogenation and water-gas shift reactions, highpurity hydrogen can be obtained (Marcano and Tsotsis 2002; Basile and Nunes 2011). The membrane can also play a role of a selective distributor, dosing a reactant along the reaction medium. This approach has been extensively used in consecutive-parallel reaction schemes, especially in partial oxy-dehydrogenation or oxidation of alkanes and oxidative coupling of methane, using dense ceramic or metallic membranes permselective to oxygen (silver, yttriumstabilized zirconia, perovskites and related oxides, or composite membranes involving a mixture of ionic and electronic conducting materials, usually oxides and metals). Separating the

Catalytic Reactions, Membrane Operations of

hydrocarbon and oxygen feed, the reaction is carried on in much safer conditions, and the possibility of thermal optimization and oxygen concentration control along the reaction medium can be used to improve the selectivity for the intermediate desired oxygenated product. Though in less extension, also proton-conducting membranes have been considered as distributors for hydrogenation reactions (Marcano and Tsotsis 2002; Seidel-Morgenstern 2010).

References Basile A, Gallucci F (eds) (2011) Membranes for membrane reactors: preparation, optimization and selection. Wiley, New York Basile A, Nunes S (eds) (2011) Advanced membrane science and technology for sustainable energy and environmental applications. Woodhead Publishing Limited, Cambridge He Y, Bagley DM, Leung KT, Liss SN, Liao BQ (2012) Recent advances in membrane technologies for biorefining and bioenergy production. Biotechnol Adv 30:817–858 Ma Y, Wang J, Tsuru T (2009) Pervaporation of water/ ethanol mixtures through microporous silica membranes. Sep Purif Technol 66:479–485 Marcano JS, Tsotsis TT (2002) Catalytic membranes and membrane reactors. Wiley-VCH Verlag GmbH, Weinheim Seidel-Morgenstern A (ed) (2010) Membrane reactors: distributing reactants to improve selectivity and yield. Wiley, New York

Further Reading Basile A (ed) (in press) Handbook of membrane reactors. Woodhead Publishing Limited, Cambridge

I

Ion-Exchange Membranes Heiner Strathmann Institute for Chemical Technology, Stuttgart University, Stuttgart, Germany

Ion-exchange membranes. Most of today’s ion-exchange membranes used in commercially relevant processes are 0.2–1 mm thin sheet of hydrocarbon or fluorinated hydrocarbon polymers which carry positively or negatively charged ions fixed to the polymer structure (Xu 2005; Bergsma and Kruissink 1961; Molau 1981). The type and the concentration of the fixed ions in a membrane structure determine permselectivity and the electrical resistance of a membrane, while the chemical stability and the mechanical properties of the membrane are determined mainly by the matrix polymer. There are three types of ion-exchange membranes: • Cation-exchange membranes which contain fixed negatively charged ions and which have a selective permeability for cations • Anion-exchange membranes which contain fixed positively charged ions and which have a selective permeability for anions • Bipolar membranes which consist of a cationand an anion-exchange membrane laminated together

# Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_119-1

The ions often used as fixed charges in cation exchange membranes are SO 3 and COO . Fixed charges used in anion-exchange membranes are Nþ HR2 and Nþ R3 . The structure of a cation-exchange membrane is illustrated in Fig. 1 which shows the polymer matrix with the fixed negative ions and the mobile counterions as well as their pathway through the membrane. The most desired properties for ion-exchange membranes are: • High permselectivity – an ion-exchange membrane should be highly permeable for counterions, but should be impermeable to co-ions. • Low electrical resistance – the permeability of an ion-exchange membrane for the counterions under the driving force of an electrical potential gradient should be as high as possible. • Good mechanical and form stability – the membrane should be mechanically strong and should have a low degree of swelling or shrinking in transition from dilute to concentrated ion solutions. • High chemical stability – the membrane should be stable over the entire pH range from 1 to 14 and in the presence of oxidizing agents. For the practical preparation of ion-exchange membranes, two rather different procedures are

2

Ion-Exchange Membranes

Ion-Exchange Membranes, Fig. 1 Schematic drawing illustrating the structure of a cation-exchange membrane showing fixed negative ions and mobile positive counterions in the polymer matrix

used. The first procedure results in a homogeneous ion-exchange membrane structure and is closely related to the preparation of ion-exchange resins. Homogeneous ion-exchange membranes are produced by either a polymerization of monomers that carry anionic or cationic moieties or by introducing these moieties into a polymer which may be in an appropriate solution or a solid preformed film. The second widely used ion-exchange membrane preparation technique which leads to a rather heterogeneous structure is based on mixing an ion-exchange resin powder with a binder polymer, such as polyvinylchloride or polyethylene, and extruding the mixture as a film at a temperature close to the melting point of the polymer. A cation-exchange membrane with exceptional good chemical and thermal stability which is widely used in the electrolytic chlorine-alkaline production and as polymer

electrolyte in low-temperature fuel cell consists of a polyfluorocarbon material. This membrane is often referred to as “Nafion ® membrane” which is the trade name of DuPont. There are several more ion-exchange membranes with special properties commercially available such as the bipolar membrane which is composed of an anion- and a cation-exchange layer laminated together (Simons 1993).

References Bergsma F, Kruissink CH (1961) Ion-exchange membranes. Fortschr Hochpolym Forsch 21:307–362 Molau GE (1981) Heterogeneous ion-exchange membranes. J Membr Sci 8:309–330 Simons R (1993) Preparation of high performance bipolar membranes. J Membr Sci 78:13–23 Xu TW (2005) Ion-exchange membranes: state of their development and perspective. J Membr Sci 263:1–29

C

Cell Adhesion Loredana De Bartolo National Research Council of Italy, Institute on Membrane Technology, CNR-ITM, Rende, CS, Italy

Cell adhesion is the binding of a cell to a surface, extracellular matrix (ECM), or another cell using cell adhesion molecules, which are integral membrane proteins that have cytoplasmic, transmembrane, and extracellular domains. The extracellular domains of adhesion molecules extend from the cell and bind to other cells or matrix by binding to other adhesion molecules of the same type (homophilic binding), binding to other adhesion molecules of a different type (heterophilic binding), or binding to an intermediary “linker” which itself binds to other adhesion molecules. Adhesion molecules belong to four major families: cadherins, immunoglobulin-like adhesion molecules, integrins, and selectins. Cadherins cause adhesion via homophilic binding to other cadherins in a calcium-dependent manner. As is the case for their role in desmosomes and adherens junctions, cadherins ultimately anchor cells through cytoplasmic actin and intermediate filaments. Immunoglobulinlike adhesion molecules are involved in both homophilic and heterophilic binding. The well-studied members of this group are the neural cell adhesion molecules (N-CAMs), # Springer-Verlag Berlin Heidelberg 2014 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_120-6

which are expressed predominantly in nervous tissue, and the intercellular cell adhesion molecules (ICAMs). Integrins are a diverse and large group of heterodimeric glycoproteins. The two subunits, designated as alpha and beta, both participate in binding. Integrins participate in cell–cell adhesion and are of great importance in binding and interactions of cells with components of the extracellular matrix such as fibronectin. Importantly, integrins facilitate “communication” between the cytoskeleton and extracellular matrix, allow each to influence the orientation and structure of the other. Selectins are expressed primarily on leukocytes and endothelial cells and, like integrins, are important in many host defense mechanisms involving those cells. In contrast to other cell adhesion molecules, selectins bind to carbohydrate ligands on cells, and the resulting binding forces are relatively weak. Cell adhesion is believed to be the first and dominant step for cell growth. Cells adhere strongly to some materials, but not to others (Hynes 2002). This is determined by the special structure of individual cell membranes and material surface properties. The initial cellular events that take place at the biomaterials interface mimic to a certain extent the natural adhesive interaction of cells with the extracellular matrix (Griffin and Naughton 2002). Cell adhesion on a substrate such as a membrane is a multistep process that involves, in sequence, adsorption of ECM proteins onto the surface, recognition of ECM components by cell

2

receptors, cytoskeletal rearrangements, and cell spreading. In particular, immediately after the biomaterial comes into contact with cell environment, protein adsorption to its surface occurs. This happens within seconds, long before the first cells reach the surface. Consequently, cells almost never come into direct contact with the material surface; they rather interact with the layer of adsorbed proteins. This layer mediates the cell adhesion and also provides signals to the cell through the cell adhesion receptors. The membrane properties (roughness, physicochemical, mechanical, and transport) especially the surface characteristics play an important role in the adhesion process (De Bartolo et al. 2004, 2008). Surface free energy, electric charge, and morphology might all affect cell attachment and its behavior either indirectly, e.g., by controlling adsorption of the proteins present in the culture medium (or secreted by the cells), or directly, e.g., by guiding cell spreading with suitable surface topography (De Bartolo et al. 2002). Such properties resulted in being critical to cell–substratum interaction and have to be

Cell Adhesion

considered in the choice of material suitable for biomedical device.

References De Bartolo L, Morelli S, Bader A, Drioli E (2002) Evaluation of cell behaviour related to physicochemical properties of polymeric membranes to be used in bioartificial organs. Biomaterials 23(12):2485–2497 De Bartolo L, Gugliuzza A, Morelli S, Cirillo B, Gordano A, Drioli E (2004) Novel PEEK-WC membranes with low plasma protein affinity related to surface free energy parameters. J Mater Sci Mater Med 15:877–883 De Bartolo L, Rende M, Morelli S, Giusi G, Salerno S, Piscioneri A, Gordano A, Di Vito A, Canonaco M, Drioli E (2008) Influence of membrane surface properties on the growth of neuronal cells isolated form hippocampus. J Membr Sci 325:139–149 Griffin L, Naughton G (2002) Tissue engineering – current challenges and expanding opportunities. Science 259:1009–1014 Hynes RO (2002) Integrins: bidirectional, allosteric signaling machines. Cell 10:673–687

C

Cell Adhesion in Bio Artificial Organs receptors recognize and interact with either Loredana De Bartolo Institute on Membrane Technology, ITM-CNR, National Research Council of Italy, Rende (CS), Italy

Binding of cells to a surface, extracellular matrix, or other cells in an artificial device is used to replace a nonfunctioning organ. Correct cellular adhesion is essential in maintaining multicellular structure and allows cells to interact dynamically with adjacent cells and the extracellular matrix. In bio artificial organs the cell adhesion process involves cell–cell and cell–natural or artificial matrix/material that plays essential roles in overall tissue architecture and proper physiological functions of the device. The functional units of cell adhesion are typically multiprotein complexes made up of three general classes of proteins: the cell adhesion molecules/adhesion receptors, the extracellular matrix (ECM) proteins, and the cytoplasmic plaque/peripheral membrane proteins. The cell adhesion receptors are usually transmembrane glycoproteins that mediate binding interactions at the extracellular surface and determine the specificity of cell–cell and cell–ECM recognition. They include members of the integrin, cadherin, immunoglobulin, selectin, and proteoglycan (for example, syndecans) superfamilies. At the extracellular surface, the cell adhesion # Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_121-5

other cell adhesion receptors on neighboring cells or with proteins of the ECM. ECM proteins are typically large glycoproteins, including the collagens, fibronectins, laminins, and proteoglycans that assemble into fibrils or other complex macromolecular arrays. Owing to their binding to adhesion receptors, they can also be tightly associated with the cell surface. At the intracellular surface of the plasma membrane, cell adhesion receptors associate with cytoplasmic plaque or peripheral membrane proteins. Cytoplasmic plaque proteins serve to link the adhesion systems to the cytoskeleton, to regulate the functions of the adhesion molecules, and to transmit signals initiated at the cell surface by the adhesion receptors. Several substrates different for the shape and physico-chemical properties are used for cell adhesion in artificial devices. Natural substrates such as ECM proteins (collagen, fibronectin, vitronectin, laminin) or polymers (chitosan, polylactic acid, polylysine, etc.) and synthetic substrates such as polystyrene, polycarbonate in the shape of scaffolds, gels, sponge, and membranes can be used in devices. Semipermeable membranes made from polyethersulfone, polysulfone, polycarbonate, polytetrafluoroethylene, modified polyetheretherketone, chitosan, polycaprolactone, polylactic acid, polyglycolic acid, and copolymers in flat and hollow fiber configurations are used for adhesion of different type of anchorage-dependent cells (hepatocytes, endothelial cells, neuronal cells,

2

keratinocytes, progenitor cells, osteoblasts). In the case of bioartificial organs, isolated cells are compartmentalized in a polymeric membrane which provides a number of important functions for the success of these devices. Membranes should act as barriers to immunocompetent species present in the patient’s blood and should permit the rapid passage of key metabolites such as nutrients and oxygen from the surrounding to the cell compartment (Curcio et al. 2005). In such devices cells are contacting with membranes, the surface properties of membrane could affect the various bioresponses. Thus, one major approach of the materials scientists has been to try to influence the extent and the character of the cell response by modifying the surface composition and properties of the polymer. The response of cells to different material properties is a complex process and even minute changes in composition of the substrate produce amplified differences in cell responses. Although surface properties are often derived from the bulk properties of the materials, the bulk materials do not entirely define them, because the used substrates are coated with proteins almost immediately after implantation in the body or immersion in culture media. Surface chemistry and topography determine the identity, quantity, and conformational change of these adsorbed proteins. In particular, the roughness and pore size of polymeric membranes seem to play an important role since they have been shown to influence the viability and metabolic rates of cells (De Bartolo et al. 2008). Modification of surface chemistry including grafting of functional groups, peptides, and proteins represents a strategy to control cell responses in in vitro and in vivo systems. The most common peptide immobilized onto surfaces

Cell Adhesion in Bio Artificial Organs

is RGD amino acid (arginine-glycine-aspartic acid) sequence that stimulates cell adhesion and growth since this peptide represents the minimal adhesion domains of the most ECM proteins (De Bartolo et al. 2005). The immobilization of specific moieties that interact with specific receptors of cell membrane is a challenge to enhance the selectivity of the membrane with respect to a cell type in order to employ it in tissue engineering for the reconstruction of a specific tissue. For example the immobilization of galactose motifs on the surface could enhance the specific interaction with hepatocytes owing to the specific binding between the galactose moiety and the asialoglycoprotein receptor present on hepatocyte cytoplasmatic membrane (De Bartolo et al. 2006).

References Curcio E, De Bartolo L, Barbieri G, Rende M, Giorno L, Morelli S, Drioli E (2005) Diffusive and convective transport through hollow fiber membranes for liver cell culture. J Biotechnol 117:309–321 De Bartolo L, Morelli S, Lopez L, Giorno L, Campana C, Salerno S, Rende M, Favia P, Detomaso L, Gristina R, d’Agostino R, Drioli E (2005) Biotransformation and liver specific functions of human hepatocytes in culture on RGD-immobilised plasma-processed membranes. Biomaterials 26(21):4432–4441 De Bartolo L, Morelli S, Rende M, Salerno S, Giorno L, Lopez LC, Favia P, d’Agostino R, Drioli E (2006) Galactose derivative immobilized glow discharge processed PES membranes maintain the metabolic activity of human and pig liver cells. J Nanosci Nanotechnol 6:2344–2353 De Bartolo L, Rende M, Morelli S, Giusi G, Salerno S, Piscioneri A, Gordano A, Di Vito A, Canonaco M, Drioli E (2008) Influence of membrane surface properties on the growth of neuronal cells isolated form hippocampus. J Membr Sci 325:139–149

C

Cell Separation Loredana De Bartolo National Research Council of Italy, Institute on Membrane Technology, CNR-ITM, Rende(CS), Italy

Cell separation is the process aimed at separating cells. Cells can be separated on the basis of different size, shape, physicochemical characteristics, specific molecules, or receptors present over their plasmatic membrane (Orfao and RuizArguelles 1996). Traditionally cells are separated through centrifugation technique for the different density and size. This technique only requires the resuspension of the cells in an appropriate buffer and the knowledge of their approximate composition and density or size. Cells of different masses and densities are pelleted accordingly, with the densest cells pelleting first and at comparatively low centrifugation speeds, while the smallest, lightest cells require much faster speeds. The pellets can then be collected and resuspended in the desired buffer. Another method consists to use antibody which are able to recognize cell membrane proteins. All cells have an array of proteins on their surface membrane, some of which are found only on particular cells. The antibodies are often tagged with a fluorescent compound (fluorophore) so that the whole cell suspension can then undergo flow cytometry, which will identify and then isolate # Springer-Verlag Berlin Heidelberg 2013 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_122-4

each cell individually. The antibodies can be also combined with a tiny magnet. The cells are then applied to a magnetic column that is capable of retaining these tagged cells when a magnetic field is generated within it. Cells can be separated through a biochemical method, which involves perturbing a biochemical process that is required by the cell for its growth and/or survival, and it should only be performed if a drastic manipulation will not detrimentally affect other steps in the experiment. Biochemical separations include blocking DNA synthesis, e.g., with hydroxyurea, and serum deprivation, i.e., growing cells in serum-free media for a specific amount of time. Membrane processes are used for separation of blood cells from other blood components through filtration. Cells are isolated by using membranes with suitable molecular weight cutoff that permit the passage of all components excluding only cells. Membrane filtration has also been used for the industrial separation of blood cells. Blood for transplantation is typically passed through membrane filters to eliminate leukocytes, which can help prevent infection by viruses such as human immunodeficiency virus and hepatitis C virus. Compared with other cell separation methods, membrane filtration is simple and inexpensive, and it is easy to maintain sterility during the process. Various porous polymeric membranes have been used for separation by filtration of different types of marrow stromal cells (KUSAA1 osteoblasts and H-1/A preadipocytes), due to the different cell size (Higuchi et al. 2005).

2

Separation of hepatocytes and fibroblasts has been realized through surface-modified polyurethane membranes combining the filtration process with the use of negatively charged membranes (Higuchi and Tsukamoto 2004).

References Higuchi A, Tsukamoto Y (2004) Cell separation of hepatocytes and fibroblasts through surface-modified

Cell Separation polyurethane membranes. J Biomed Mater Res A 71A(3):470–479 Higuchi A, Shindo Y, Gomei Y, Mori T, Uyama T, Umezawa A (2005) Cell separation between mesenchymal progenitor cells through porous polymeric membranes. J Biomed Mater Res B Appl Biomater 74(1):511–519 Orfao A, Ruiz-Arguelles A (1996) General concepts about cell sorting techniques. Clin Biochem 29(1):5–9

C

CH4/N2 Separation A. Baudot Physics and Analysis Division, IFP Energies nouvelles, Solaize, France

The US pipeline specification requires an inert content lower than 4 % in natural gas. It is estimated that 14 % of US present proven gas reserves contain more than 4 % nitrogen (Baker and Lokhandwala 2008; Lokhandwala et al. 2010). The most trivial solution consists in diluting small flow rates of nitrogen-concentrated natural gas with main streams of natural gas containing a low concentration in inert gases. However, if this is not feasible (for instance, when all the surrounding gas fields contain high concentrations in nitrogen), it is necessary to install a nitrogen removal plant to comply with the regulatory natural gas composition specifications. Cryogenic distillation is the dedicated technology for large-scale nitrogen removal operations (60,000–600,000 Nm3/h capacity for operation life higher than 10 years). At the very end of the 1990s, 26 such plants were in operation on US ground (Lokhandwala et al. 2010). Two technologies have been recently introduced in order to solve this challenge at small and medium scale: pressure-swing adsorption (PSA) with molecular sieves (Molecular Gates ® by the Engelhard company (Mitariten 2001, 2004) and gas permeation, which has been promoted by # Springer-Verlag Berlin Heidelberg 2014 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_127-2

MTR Inc. since the end of the 1990s (Lokhandwala et al. 2010). In the mid-1990s, MTR Inc. evaluated at lab scale (on 1–2 m2 membrane modules) a wide array of composite membranes which selective membranes were made of various glassy or rubbery polymers (Lokhandwala et al. 2010). Intuitively, rubbery polymers proved to be more selective toward methane (due to its higher sorption, induced by its more condensable behavior than nitrogen), while glassy polymers, especially fluorinated ones, proved to be more selective toward nitrogen (due to diffusive selectivity). MTR Inc. evaluated the economic feasibility of a membrane-based nitrogen removal operation on a 12,000 Nm3/h natural gas stream entering the process at 32 bar and containing 10 % nitrogen. The target of the membrane operation was to reach the 4 % nitrogen specification in the product gas while the waste gas was containing at least 50 % nitrogen and the overall methane recovery yield was to be higher than 93 %. Based on the performances of the various tested membrane materials, the study led to the conclusion that a single-stage membrane operation was not viable, whether the membrane material was nitrogen-selective or methane-selective. Therefore, two-stage recycle schemes were studied involving only nitrogen-selective membranes or only methane-selective membranes or a combination of both. Only the use of methane-selective membranes proved to offer interesting performances (Lokhandwala et al. 2010), though the

2

CH4/N2 Separation

Nitrogen removal (4 to 8% N2) (NitrosepTM, MTR Inc.) Flow rate (103 Nm3/h) Pressure (bar) Composition (% mol.)

1

2

3

4

2

5

12

9.7

2.3

1.5

0,8

70

14

?

?

5.6

N2

8

4

26

50

12.6

CH4

92

96

92.5

50

87.4

1

3

CH4 recovery = 86 % 550 m² membrane area 690 kW (theoretical compression power) stream 5 used as compressor fuel stream 4 to flare

Nitrogen removal (8 to 15% N2 ) (NitrosepTM, MTR Inc.) Flow rate (103 Nm3/h) Pressure (bar) Composition (% mol.)

4 2 1

2

3

4

5

3 6

12

9

?

?

1,9

1,1

70

14

?

?

?

3.5

N2

15

4

8.7

50

65

26

CH4

85

96

81.3

50

35

74

1

CH4 recovery = 86 % 1,800 m² membrane area 1,860 kW (theoretical compression power) stream 6 used as compressor fuel stream 5 to flare

Nitrogen removal (15 to 30% N2 ) (NitrosepTM, MTR Inc.) Flow rate (103 Nm3/h) Pressure (bar) Composition (% mol.)

5

1

6

4 4

3 2

3

5

4

12

9

?

?

70

14

?

?

N2

15

80

11

4

CH4

85

20

89

96

CH4 recovery = 96 % 3,500 m² membrane area 2,680 kW (theoretical compression power) part of stream 1 used as compressor fuel stream 2 to flare

1

2

CH4/N2 Separation, Fig. 1 Permeation-based nitrogen removal from natural gas (After Lokhandwala et al. 2010)

resulting process schemes were leading to significant extra cost due to the presence of two compressors: the first compression unit was used to pressurize the methane-enriched permeate produced by the first membrane stage to the pipeline pressure, while the second one was aiming at boosting the pressure of the permeate stream issued from the second-stage in order to recycle it at the input of the first membrane stage. With this scenario, the total equipment investment cost was in a 4–8 million US dollar range (in 2009), while the required membrane area was approximately equal to 4,000 m2. This appears to be acceptable by gas processors as the payback

time of this type of installation was estimated to be 1 year. The first field test of these membranes was therefore conducted with a 40 in-long module equipped with few m2 of methane-selective membranes for more than 1 year on a shut-in gas well containing 15 % nitrogen operated by Butcher Energy in Southern Ohio. The performances of the spiral-wound module were very similar to those observed at lab scale, and its permeation properties remained unchanged for the first 6-month period. After this proof of principle, MTR Inc. designed different membranebased nitrogen removal process schemes, trade-

CH4/N2 Separation

named Nitrosep™, with various module/compressor configurations depending on the nitrogen concentration to be addressed (Fig. 1). Up till now, 12 Nitrosep™ units have been installed in the industry. Few examples depicting the applications addressed by this membranes process are listed hereafter: – A very small unit was installed in Southern Kentucky in order to upgrade 200 Nm3/h of natural gas containing 7 % nitrogen. This two-membrane module unit was able to recover 80 % of the natural gas with a nitrogen content of 3.8 %. Part of the residue gas was used as compressor engine fuel, while the rest was vented. – A much larger unit was installed in Rio Vista, California (14,000 Nm3/h treatment capacity). The membrane operation was aiming at upgrading a 16 % nitrogen containing natural

3

gas to a gas stream offering 10 % more heating value (lowering thus the nitrogen content to 9 %). The membrane system involved three stages in series and allowed a methane recovery yield of 95 % for pipeline delivery.

References Baker RW, Lokhandwala K (2008) Natural gas processing with membranes: an overview. Ind Eng Chem Res 47:2109–2121 Lokhandwala KA, Pinnau I, He ZJ, Amo KD, DaCosta AR, Wijmans JG, Baker RW (2010) Membrane separation of nitrogen from natural gas: a case study from membrane synthesis to commercial deployment. J Membr Sci 346:270–279 Mitariten M (2001) New technology improves nitrogenremoval economics. Oil Gas J 99:42–44 Mitariten M (2004) Economic nitrogen removal. Hydrocarb Eng 9:53–57

C

Charged Ultrafiltration Membrane Andrew Zydney Department of Chemical Engineering, The Pennsylvania State University, University Park, PA, USA

Electrically charged ultrafiltration membranes can be used to reduce the fouling and increase the retention of like-charged species. Most polymeric membranes have a net negative charge in solution due to the presence of trace anionic groups (e.g., carboxylic acids) and/or the preferential adsorption of negatively charged ions from the aqueous solution. It is also possible to cast membranes from polymers containing fixed charge groups, e.g., a positively charged membrane can be developed using polymers containing fixed amine groups. However, the most common method for generating charged ultrafiltration membranes is by surface modification of a base polymer through the attachment of appropriate anionic (e.g., carboxylic or sulfonic acid) or cationic (e.g., quaternary amine) groups. The overall performance of these membranes is determined by the density of the charge groups, the chemistry of the ligand and the covalent linkage, and the properties of the spacer arm used to attach the ligand to the membrane (Zydney 2011). One of the first applications of charged ultrafiltration membranes was in the recovery of # Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_129-4

electropaint. Electrodeposition of a charged paint resin is used extensively in the industrial painting of metallic surfaces in automobiles and large household appliances. The paint resins are organic polymers with attached anionic or cationic groups. Ultrafiltration has been used to recover the charged electropaint from the dilute solution produced by washing the excess paint off of the metal part. Ultrafiltration can also be applied directly to the paint in the electrodeposition bath, with the permeate (essentially water) used for subsequent rinsing steps (Zeman and Zydney 1996). The use of charged membranes (having the same polarity as the electropaint) provides significantly greater filtrate flux with less fouling than neutral or oppositely charged membranes. Charged ultrafiltration membranes can be used to significantly improve the inherent tradeoff between the permeability and selectivity of an ultrafiltration membrane (Mehta and Zydney 2005). The rate of solute transport through a charged ultrafiltration membrane is determined by a combination of steric (size based) and electrostatic interactions (Mehta and Zydney 2006). Electrostatic interactions strongly effect the partitioning of charged solutes into the membrane pores. For example, the presence of positively charged groups on the pore surface causes a strong electrostatic exclusion of positively charged species from the membrane pores significantly increasing the selectivity of the membrane. This makes it possible to employ

2

Charged Ultrafiltration Membrane

Charged Ultrafiltration Membrane, Fig. 1 Positively charged membrane provides high retention of positively charged protein while allowing transmission of neutral proteins

charged ultrafiltration membranes with relatively large pore size and thus with very high permeability, with the required selectivity achieved by electrostatic exclusion of the protein. There is also considerable interest in performing protein separations by charged UF membranes. In this case, the charged ultrafiltration membrane provides very high retention of like-charged proteins, enabling uncharged proteins and smaller impurities to be washed into the permeate by a diafiltration process (see Fig. 1). The solution pH and ionic strength can be adjusted to obtain high-resolution protein separations (van Reis et al. 1999). Van Reis and Zydney (2007) have discussed a number of separation processes using charged ultrafiltration membranes, including a nonaffinity process for purifying a monoclonal antibody from harvested cell culture fluid.

References Mehta A, Zydney AL (2005) Permeability and selectivity analysis for ultrafiltration membranes. J Membr Sci 249:245–249 Mehta A, Zydney AL (2006) Effect of membrane charge on flow and protein transport during ultrafiltration. Biotechnol Prog 22:484–492 van Reis R, Zydney AL (2007) Bioprocess membrane technology. J Membr Sci 297:16–50 van Reis R, Brake JM, Charkoudian J, Burns DB, Zydney AL (1999) High performance tangential flow filtration using charged membranes. J Membr Sci 159:133–142 Zeman LJ, Zydney AL (1996) Microfiltration and ultrafiltration: principles and applications. Marcel Dekker, New York Zydney AL (2011) High performance ultrafiltration membranes: pore geometry and charge effects. In: Oyama ST, Stagg-Williams SM (eds) Inorganic, polymeric, and composite membranes: structure, function, and other correlations. Elsevier, Amsterdam, pp 333–352

C

Chemical Binding of Biomolecules to Membranes Christophe Innocent Institut Europe´en des Membranes, University of Montpellier, Montpellier Cedex 5, France

The immobilization in electropolymerized polymers such as polypyrrole has been developed for a variety of biomolecules and provides very stable environment for the biocatalyst. The advantages of the polypyrrole are the control of film thickness and its contribution in electron transfer between the biocatalysts and the conductive support.

# Springer-Verlag Berlin Heidelberg 2015 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_130-5

Amino pyrrole is synthetized according the following scheme (Fig. 1). Electropolymerization of 1-(2-cyanoethyl) pyrrole monomer (1) in acetonitrile 0.1 M NBu4PF6 was performed by controlled-potential oxidation of the pyrrole at +1.01 V vs. SCE. Reduction of the nitrile to amine function was carried out using a large excess of LiAlH4 in dried ether at room temperature. Enzyme is grafted on the modified tube (with amino polypyrrole) by using glutaraldehyde as coupling agent (Fig. 2). Fixation of laccase (enzyme which catalyzed the oxygen reduction) on the modified electrode has been investigated and applied to the fabrication of biocathode for enzymatic biofuel cell (Servat et al. 2007).

2

Chemical Binding of Biomolecules to Membranes

Et2O, AlLiH4 N

N

NH2

N 1

2

Chemical Binding of Biomolecules to Membranes, Fig. 1 Scheme of synthesis of amino propyl pyrrole

Chemical Binding of Biomolecules to Membranes, Fig. 2 Laccase immobilization on modified electrode with polyaminopyrrole; (a) nucleophilic attack on the glutaraldehyde, (b) reduction of imine function (IEM picture)

a O

O NH2

H2N

laccase

b

References Servat K, Tingry S, Brunel L, Querelle S, Cretin M, Innocent C, Jolivalt C, Rolland M (2007) Modification

N

N

laccase

N H

N H

laccase

of porous carbon tubes with enzymes: application for biofuel cells. J Appl Electrochem 37:121–127

C

Chlorine-Resistant Polymeric Membranes Pejman Ahmadiannamini Biomedical Engineering, University of Arkansas, Fayetteville, AR, USA

A chlorine-resistant polymeric membrane is mainly referred to a polymeric reverse osmosis (RO) desalination membrane that can withstand exposure to chlorine and preserve its separation characteristics under such a harsh condition. Currently, commercially available RO membranes are derived from two classes of polymers: cellulose acetate (CA) and aromatic polyamide (PA) (Li and Wang 2010). CA membranes are relatively low cost and tolerant to limited free chlorine. However, CA membranes suffer from some disadvantages such as a narrow operating pH range (4.5–7.5), susceptibility to biological attack, structural compaction under high pressure, and low upper temperature limit. On the other hand, PA thin film composite (TFC) membranes feature thin highly selective interfacially polymerized layers, which exhibit superior flux and salt rejections, wider operating temperature and pH range, and higher stability to biological attack, as compared with CA membranes (Park et al. 2008). Thus, PA TFC membranes used in RO are most preferred by the desalination industry. However, PA membranes encounter significant drawbacks in desalination processes, # Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_131-14

namely, membrane degradation under continuous exposure to trace amount of chlorine and deterioration of their performances. Chlorine is the most common industrial oxidizing biocide in water treatment that is used for disinfection of domestic water and for the removal of tastes and odors from water. It is also typically used in RO treatment processes in order to control microorganisms that biofoul and clog the membrane. Consequently, water to be purified is often chlorinated, to disinfect it and ultimately inhibit biofouling of the membranes, and then dechlorinated before being fed to membrane desalination units. After passing through the membranes, the water is then rechlorinated before being sent to the distribution network. However, this requires additional equipments and chemicals and increased operating cost for the plant, accordingly. When chlorine gas is added into water, it is hydrolyzed to form hydrogen ion, chloride, and hydrochlorous acids. The hydrochlorous can be further ionized to produce hypochlorite ions according to the following reactions (Geise et al. 2010): Cl2 + H2O HCIO

HCIO + H+ + ClH+ + CIO-

Cl2, HOCl, and OCl are in equilibrium, and depending on the pH, different distributions of aqueous chlorine species are observed (Deborde

2

Chlorine-Resistant Polymeric Membranes

and von Gunten 2008). The established chlorinating strength of these species is Cl2 ~ HOCl > OCl (Soice et al. 2003). The amide nitrogen of the membrane is highly vulnerable to chlorine attack because of electronwithdrawing effect of carbonyl group. Upon exposure to free chlorine, N-H group is chlorinated to N-Cl group which can reversibly form the initial amid by treatment with reducing agents. The aromatic rings are also susceptible to attack by chlorine because it is an electron-rich region. Two possible chlorination mechanisms are proposed for aromatic rings, i.e., direct chlorination of the aromatic ring and Orton rearrangement, which involves initial chlorination of amid nitrogen followed by an intermolecular rearrangement, forming various aromatic substitution products (Fig. 1) (Raval et al. 2010). According to the chlorination mechanisms, chlorine resistance of PA membranes largely depends on the chemical structures of the diamine components used. Aliphatic PA reversibly reacts with chlorine to yield N-chlorinated amide. Tertiary PAs are inactive towards

O

O

H

H

C

C

N

N

oxidative chlorine. The Orton rearrangement takes place only when amide linkage is directly connected with benzene ring. Generally, the chlorine resistance increases in the order of PA synthesized from aromatic, cycloaliphatic, and aliphatic diamines, respectively. Thus, the following modifications can be considered as potential strategies to enhance the chlorine resistance of PA membranes (Geise et al. 2010): 1. Replacing chlorine-sensitive amidic hydrogen on the amide linkages with other moieties, e.g., methyl ( CH3) or phenyl ( C6H5) 2. Replacing the aromatic ring bonded to the amide nitrogen with aliphatic chain or cyclics 3. Prevention of Orton rearrangement by adding protective groups at the possible chlorination sites on the aromatic rings Other than PA, polysulfone has much better chlorine resistance as it has stronger chemical bonds. However, due to hydrophobic nature of polysulfone, introduction of controlled levels of hydrophilic groups, e.g., SO3H, while retaining

Cl2

O

O

H

H

C

C

N

N

Irreversible direct aromatic chlorination

n

Cl

2,

Re

ve

rs

HC

IO

ibl

Cl

e

d

ch

lor

e

e ev

Irr

Cl

O-

ina

tio

on

t

ibl

rs

an

en

m

e ng

ra

ar

re

n

t

Or

n O

O

Cl

H

C

C

N

N

n

Chlorine-Resistant Polymeric Membranes, Fig. 1 Proposed mechanism of chlorination of aromatic PA (Reprinted from Desalination, 250, Raval HD, Trivedi JJ, Joshi SV, Devmurari CV, Flux enhancement of thin

film composite RO membrane by controlled chlorine treatment, 945–949, Copyright (2010) with permission from Elsevier)

Chlorine-Resistant Polymeric Membranes

its physical properties is necessary for polysulfone RO membranes. The polymer chain cleavage and side reactions, which can make it difficult to control the degree of sulfonation and molecular weight and consequently drop the mechanical and the thermal properties of polymer, can occur in the post-sulfonation method. The direct synthesis of the functionalized polysulfone from a sulfonated monomer is found to be more advantageous since the associated challenges can be avoided (Park et al. 2008).

References Deborde M, von Gunten U (2008) Reactions of chlorine with inorganic and organic compounds during water

3 treatment—kinetics and mechanisms: a critical review. Water Res 42(1–2):13–51 Geise GM, Lee H, Miller DJ, Freeman BD, McGrath JE, Paul DR (2010) Water purification by membranes: the role of polymer science. J Polym Sci B: Polym Phys 48(15):1685–1718 Li D, Wang H (2010) Recent developments in reverse osmosis desalination membranes. J Mater Chem 20:4551–4566 Park HB, Freeman BD, Zhang Z, Sankir M, McGrath JE (2008) Highly chlorine-tolerant polymers for desalination. Angew Chem Int Ed 47(32):6019–6024 Raval HD, Trivedi JJ, Joshi SV, Devmurari CV (2010) Flux enhancement of thin film composite RO membrane by controlled chlorine treatment. Desalination 250(3):945–949 Soice NP, Maladono AC, Takigawa DY, Norman AD, Krantz WB, Greenberg AR (2003) Oxidative degradation of polyamide reverse osmosis membranes: studies of molecular model compounds and selected membranes. J Appl Polym Sci 90(5):1173–1184

C

Citric Acid Recovery by Electrodialysis Gerald Pourcelly Institut Europeen des Membranes, CC 047, Universite Montpellier II, Place Eugene Bataillon, Montpellier Cedex 5, France

Citric Acid Recovery by Electrodialysis [G. Pourcelly] Carboxylic acids such as lactic, succinic, gluconic, citric, and tartaric are widely used in food processing, detergent manufacture, and biodegradable plastic production (Bailly et al. 2001). Their production at the industrial scale is mainly achieved by mean of fermentation from molasses, starch hydrolysates, or sugars. Traditional processes to obtain these carboxylic acids consist on the precipitation of both the acids and their salts, so they can be isolated from the rest of the components of the raw material. Then, they are placed in an acid medium (sulfuric acid) to generate the acid form. A latter concentration step is carried out by evaporation followed by a crystallization. This generates large volumes of effluents with high salt contents. For example, typical yields of 1 kg of citric acid are obtained per 2 kg of gypsum which is very difficult to dispose

# Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_133-1

(Bialey and Ollis 1986). In order to reduce this environmental impact, the design of alternative production scheme was investigated. Extraction, adsorption, and membrane technologies, like electrodialysis (ED), were proposed to replace precipitation (Novalic et al. 1995). A complete scheme for carboxylic acid recovery is depicted in Fig. 1. For EDC step of citric acid salts to the three dissociation constants of citric acid, the conductivity is mainly influenced by the pH value and the concentration. More ion dissociation leads to formation of further citric-acid-charged complexes in the solution which in turn causes electrical resistance reduction of the solution. The maximum ion density and consequently the minimum electrical resistance of the solution is obtained at pH 8 (Novalic et al. 1995). The EDBM step, which is water splitting by membrane electrodialysis, provides an attractive complement to the fermentation technology by removing the product acid while simultaneously providing an equivalent amount of base for use in adjusting the pH in the fermentor (Fig.1). Moreover, the produced citric acid is usually at a relatively high concentration (0.5 M) so that the subsequent purification via crystallization or other techniques is relatively inexpensive (Xu and Yang 2002).

2

Citric Acid Recovery by Electrodialysis

Citric Acid Recovery by Electrodialysis, Fig. 1 Scheme of a process of carboxylic acid production from fermentation. (EDC) concentration ED step, (EDBM) ED with bipolar membranes

References Bailly M, Roux-de-Balman H, Aimar P, Lutin F, Cheyran M (2001) Production processes of fermented organic acids targeted around membrane operations: design of the concentration step by conventional electrodialysis. J Membr Sci 191:129–142

Bialey JE, Ollis DF (1986) Biochemical engineering fundamentals. MacGraw-Hill, Singapore Novalic S, Jagschits F, Okwor J, Kulbe KD (1995) Behaviour of citric acid during electrodialysis. J Membr Sci 108:201–205 Xu T, Yang W (2002) Citric acid production by electrodialysis with bipolar membranes. Chem Eng Process 41:519–524

C

Composite Membrane with Inorganic Fillers: Electrolyser Application Antonino Salvatore Arico CNR-ITAE Institute, Messina, Italy

Composite Membrane with Inorganic Fillers: Electrolyzer Application Composite recast Nafion ® membranes containing inorganic fillers have been primarily employed in fuel cells for high-temperature operation (Arico` et al. 1998) and self-humidification purposes (Watanabe et al. 1996). Composite membrane with inorganic fillers for electrolyzer application has been developed in order to extend the operating temperature range of polymer electrolyte membrane (PEM) electrolyzer and to reduce gas crossover effects (Antonucci et al. 2008). Generally, Nafion® membrane is used as conducting polymer electrolyte in PEM electrolyzer systems. An increase of the operation temperature of an electrolyzer should enhance the oxygen evolution reaction rate that is the rate-determining step of this process allowing to obtain high current and high conversion efficiency. However, commercial Nafion membranes loose conductivity at temperature above 100 C due to membrane dehydration. Perfluorosulfonic acid (PFSA) composite membranes containing hygroscopic ceramic oxide fillers that require water for proton # Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_140-2

conduction appear suitable for this application especially in the light of the high operating pressure of PEM electrolyzers (up to 100 bars) that allows to maintain a good fraction of liquid water even at temperatures above 100 C. The inorganic fillers enhance the water retention inside the composite membrane allowing to operate properly at high temperatures. A composite Nafion–SiO2 membrane for SPE electrolyzers has shown promising properties for hightemperature operation allowing to achieve significantly higher performances with respect to a bare commercial Nafion. This effect is mainly due to a significantly better water retention than the bare perfluorosulfonic membrane and lower gas crossover as a result by the increased tortuosity effect produced by the inorganic filler inside the membrane. The performance of the electrolyzer based on Nafion–SiO2 membrane increased as a function of the temperature up to 120 C and pressure. A maximum current density of about 2.1 A cm 2 versus 0.7 A cm 2 at 1.9 V, 120 C, and 3 bar abs was recorded for the composite membrane compared to Nafion 115 (Fig. 1). An increase of electrical efficiency was recorded at low current densities for the hightemperature SPE electrolyzer compared to conventional membrane-based devices (Antonucci et al. 2008).

2

a

2

1.8 Terminal voltage / V

Composite Membrane with Inorganic Fillers: Electrolyser Application, Fig. 1 Polarization measurements for a PEM water electrolysis cell based on conventional Nafion 115 and composite PFSA–SiO2 membrane at various temperature and 3.0 bar abs pressure

Composite Membrane with Inorganic Fillers: Electrolyser Application

80°C 90°C 100°C 110°C 120°C

1.6 Nafion 115 1.4

1.2

1 −0.2

0.2

0.4

0.6

1

0.8

1.2

1.4

Current density / A cm-2

b

2

Terminal voltage / V

1.8 80°C 90°C

1.6

100°C 110°C

Composite Membrane

1.4

120°C

1.2

1 −0.2

0.2

0.6

1

1.4

1.8

2.2

-2

Current density / A cm

References Antonucci V, Di Blasi A, Baglio V, Ornelas R, Matteucci F, Ledesma-Garcia J, Arriaga LG, Arico` AS (2008) High temperature operation of a composite membrane-based solid polymer electrolyte water electrolyser. Electrochim Acta 53:7350

Arico` AS, Cretı` P, Antonucci PL, Antonucci V (1998) Comparison of ethanol and methanol oxidation in a liquid-feed solid polymer electrolyte fuel cell at high temperature. Electrochem Solid-State Lett 1:66–68 Watanabe M, Uchida H, Seki Y, Emori M, Stonehart P (1996) Self-Humidifying Polymer Electrolyte Membranes for Fuel Cells. J Electrochem Soc 143:3847

C

Composite Membrane with Inorganic Fillers: Fuel Cell Application Antonino Salvatore Arico CNR-ITAE Institute, Messina, Italy Composite recast Nafion ® membranes containing inorganic fillers have been employed in hightemperature (~150 C) direct alcohol (Arico` et al. 1998) and H2-air fuel cells (Watanabe et al.1996). These composite membranes were originally developed for reduced humidification operation in polymer electrolyte fuel cells (Watanabe et al.1996) due to the enhanced water retention inside the membrane by the effect of the inorganic filler (Arico` et al. 1998). A further advantage of composite membranes relies in the barrier effect given by the inorganic filler for methanol cross over (Ren et al. 1996) which is of particular relevance at high temperature. It is well known that the physical adsorption of water by materials such as silica (one of the most used inorganic fillers) is mainly determined by their surface properties; similar considerations can be made for other hygroscopic inorganic oxides such as alumina. Functional groups on the surface of these oxides are believed to act as water coordination centers (Arico` et al. 2003). FTIR analysis of various silica materials suggests that oxygen surface functionalities play a prevailing role in the adsorption of water. The # Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_141-2

surface characteristics of an inorganic oxide can be modified by thermal treatments in inert or oxidizing atmosphere through reactions with strong inorganic acids (Arico` et al. 2003). Most of the works on composite membranes have addressed the technical aspects related to the use of these materials as electrolytes in hightemperature fuel cells. Accordingly, performance, conductivity, and stability characteristics have been investigated in-depth. Parallel work has been concerned with the investigation of relevant filler properties for application in composite membranes such as surface area analysis, surface chemistry studies, and surface acid-base investigations of the fillers. The conductivity of perfluorosulfonic acid (PFSA) composite membranes and fuel cell power density at high temperature have been found to be related to the characteristics of the water adsorbed on the filler particles. Inorganic fillers characterized by acidic properties undergo a strong interaction with water and enhance the DMFC performance at high temperature. Appropriate selection of the surface properties for the inorganic fillers thus allows to enhance proton conductivity and fuel cell performance and extends the operating temperature range of composite membranes (Arico` et al. 2003; Fig. 1).

Composite Membrane with Inorganic Fillers: Fuel Cell Application

surface OH groups stretching/ cm-1

a

3300

3550

3200

3500

3100

3450

3000

2

3

4

5

6

7

8

9

H2O stretching vibration / cm-1

2

3400

pH filler slurry

b

1650

1640

300

200 1630

Power Density / mW cm-2

H2O bending vibration / cm-1

400

100

1620

2

4

6

8

0 10

pH filler slurry

Composite Membrane with Inorganic Fillers: Fuel Cell Application, Fig. 1 Relationships between water adsorption characteristics of inorganic fillers and

corresponding performance of a composite membranebased direct methanol fuel cell at 145 C

References

performance of composite membranes in direct methanol fuel cells. Solid State Ion 161:251–265 Ren X, Wilson MS, Gottesfeld S (1996) High performance direct methanol polymer electrolyte fuel cells. J Electrochem Soc 143:L12 Watanabe M, Uchida H, Seki Y, Emori M, Stonehart P (1996) Self-Humidifying Polymer Electrolyte Membranes for Fuel Cells. J Electrochem Soc 143:3847

Arico` AS, Cretı` P, Antonucci PL, Antonucci V (1998) Comparison of ethanol and methanol oxidation in a liquid-feed solid polymer electrolyte fuel cell at high temperature. Electrochem Solid-State Lett 1:66–68 Arico` AS, Baglio V, Di Blasi A, Creti P, Antonucci PL, Antonucci V (2003) Influence of the acid-base characteristics of inorganic fillers on the high temperature

C

Computational Fluid Dynamics (CFD) about the wall shear stress distribution at the and Membranes surface of the membrane. The study of hydrodyPhilippe Moulin Laboratoire de Me´canique, Mode´lisation et Proce´de´s Propres (M2P2-UMR 7340), Equipe Proce´de´s Membranaires (EPM), Aix Marseille Universite´, AIX en PROVENCE, France

The increasing number of CFD membrane studies is clearly related to the recent developments in computer power and to the use of finer grid meshes in the vicinity of the membrane. Two approaches have been particularly considered: the comprehension of the hydrodynamics and of the mass transfer. The hydrodynamics allows the increase of the shear stress near the wall or the transmembrane pressure thus allowing the enhancement of permeate flux and the membrane processes. CFD allows determining the hydrodynamics, i.e., the pressure and velocity fields, taking into account the geometry of the module and the membranes, the membrane permeability and compactness, as well as the operating entry values such as filtration or backwash pressures, filtration mode, and gravity. For example, it is possible to determine the pressure and velocity fields in (i) a hollow fiber module containing more than 40,000 fibers (small diameter (di = 0.93 mm) and large flow rate (50 m3.h 1) or in (ii) ceramic multichannel membranes or module to obtain information # Springer-Verlag Berlin Heidelberg 2014 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_142-1

namics reveals the optimum operating conditions and the most suitable geometry characteristics to determine a compromise between membrane area, channel (geometry and number), and energy consumption to optimize membrane processes. A large number of studies relate to the diphasic flows, to the turbulence promoters, and to the geometries of membranes capable of generating secondary flows. Usually a good agreement is obtained between CFD and experimental data obtained for the transfer of solvent. Some recent studies consider turbulent flows and the significant number of turbulence models in numerical simulations, and their comparison with experimental results in the case of membrane processes is not excellent, which limits the use of CFD in turbulent regime. Fouling remains a major problem in membrane processes: this phenomenon limits the process efficiency and is difficult to predict and anticipate. These difficulties are linked to the complexity of this phenomenon which implies different interdependent mechanisms occurring at pore scale. The relative importance of each fouling mechanism has been determined according to the particle size, the pore size, and the surface density of pores. By CFD progress it is now possible (a) to pursue the description of the different fouling mechanisms by integrating the complexity of the real membrane structure in the numerical simulation and (b) to simulate

2

different kinds of experimental deposit structure. The limitation is the membrane reconstruction technique limiting this description at the MF and UF membranes. The filtration in a cylindrical pore (i.e., microchannels) can be simulated for different sizes of particles or pores, tortuosities, and hydrodynamic conditions. For these CFD

Computational Fluid Dynamics (CFD) and Membranes

studies, the difficulties are to well describe particle/particle colloidal interactions and resuspension of particles after capture. Thus, CFD is without any doubt an important tool for understanding mass transfer in membrane processes, and opportunities for the development of new membrane geometries are numerous.

C

Computer-Aided Methods and Tools Giorgio De Luca Institute on Membrane Technology ITM-CNR, University of Calabria, Rende (CS), Italy

Computational methods can be divided into approaches using adjustable or empirical parameters and those which do not use them. Concerning the first methods, several procedures have been developed for the optimization of the adjustable parameters. For example, design of experiments is an approach defining the minimum number of experiments required to obtain the fitting parameters. Instead, in the ab initio methods, these quantities are obtained by direct experimental measurements or from other simulations carried out in smaller time-space scale (Steinhauser 2008), until the sub-nanometer scale is reached. The choice of the computational approaches strictly depend on the proprieties to be studied. Figure 1 shows the various computational methods and properties that can be obtained. Binding energies, molecular electrostatic properties must be evaluated by quantum mechanics (QM) approaches (Veszprémi and Fehér 1999). These generally require significant computational resources, although the development of parallel supercomputers and efficient algorithms have allowed to carry out QM calculations unthinkable few years ago. The quantities, # Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_143-2

obtained at QM level, can be used in subsequent molecular dynamics (MD), Monte Carlo (MC), or semiempirical calculations to describe properties related to a huge number of molecules and atoms, like polymers. MD methodologies are based on the simultaneous solution of Newton’s equation of motion referred to the atoms of the physical system (Allen and Tildesley 2003). The potential energy surface (interaction potential), also called force field (FF), used in the MD simulations is defined by QM or parameterized over measurements. The temperature, volume, pressure, and number of particles define the statistical ensemble in which the MD can be performed. The classic MC method is generally based on the von Neumann, Metropolis, and Ulam algorithm (Metropolis and Ulam 1949; Wood 1986). Later, the method has grown to the point where it leads to several methods all belonging to the MC family. In the grand canonical MC ensemble, the simulations are performed at chemical potential, volume, and temperature constant, whereas the number of molecules varies. The potential surface energy is always defined by means of a specific FF. Interesting MC approaches are based on the quantum MC methodology. The semiempirical methods use Hamiltonians to describe the system, but some contributions of these operators are obtained empirically (Clementi and Corongiu 1995). These approaches allow to evaluate the target properties more quickly than the QM methods. In Table 1,

2

Computer-Aided Methods and Tools

Time, s

D,α, J,θ, Κeq. FEM, Lattice Bolztmanm

Macro

10–3

Phase distri., Morphologies, Micro fluids.

Micro

Lattice Bolztmanm 10–9 FFV,FAV, d-spa,Tg, Conformationa analysis, Molecular and sorption diffusion in large system Semiempirical, Monte Carlo

Nano 10–15

Bond energies ,Dipoles, Electr. Charges, Accurate geometries, Spectroscopic analysis. Quantum Mechanics

Sub Nano 10–9

10–7

10–4

100

Space, m Computer-Aided Methods and Tools, Fig. 1 Computational methods and achievable properties

Computer-Aided Methods and Tools, Table 1 Advantages and disadvantages of some computational methods Method Molecular mechanics and Monte Carlo dynamics Semiempirical

Ab initio, density functional theory quantum mechanics

Advantages Systems of thousands of atoms, low computational costs

Less computational demand than quantum mechanics approaches, systems of hundreds of atoms Do not depend on experimental data, useful for a broad range of molecules without available experimental data, general

the illustrated computational methods are summarized with their advantages and disadvantages. The information provided by smaller scales can be utilized in mesoscale calculations: coarse grain MD or Lattice Boltzmann (LB) (Swift et al. 1996). In particular, LB is a powerful method for simulating fluid confined in microsystems, as may be precisely a membrane. This method allows to solve the Navier-Stokes equations in a simple way and with a notable reduction of computational time. LB allows an easy description of the interfaces and especially without the use of adjustable parameters as made by the conventional finite elements methods (FEM). Finally, the description of the properties

Disadvantages Experimental data or values from quantum mechanics, absence of bond breaking/ forming. Noncovalent bonds are not well described, less general Experimental data or values from quantum mechanics, less general Computationally very expensive Small systems

of macrosystems can be carried out using methods based on classic mechanics or dynamics. The differential equations describing these physical systems are evaluated numerically by means of FEM or finite volume procedures (De Luca et al. 2014). There is a notable number of codes implemented for each method just described. These are divided into programs in which a single methodology is implemented or those in which several methods, radically different, can be found. Therefore, it is possible to find codes in which only QM (based on different theories) or MD are implemented separately and others in which both the methodologies are present. Particular attention should be given to

Computer-Aided Methods and Tools

parallel algorithms since parallel supercomputers allow the use of huge number of processors (or computers interconnected by the net). These possibilities allow to perform complex calculations in short computational time.

References Allen MP, Tildesley DJ (2003) Computer simulation of liquids. Oxford University Press, New York Clementi E, Corongiu G (eds) (1995) Methods and techniques in computational chemistry METECC-95. STEF, Cagliari. ISBN 88-86327-02-1, Club Europeen MOTECC

3 De Luca G, Bisignano F, Paone F, Curcio S (2014) Multiscale modeling of protein fouling in ultrafiltration process. J Membr Sci 452:400–414 Metropolis N, Ulam S (1949) The Monte Carlo method. J Am Stat Assoc 44:335–341 Steinhauser MO (2008) Computational multiscale modeling of fluids and solids. Springer, Heidelberg Swift MR, Orlandini E, Osborn WR, Yeomans JM (1996) Lattice Boltzmann simulations of liquid-gas and binary fluid systems. Phys Rev E 54(5):5041–5052 Veszprémi T, Fehér M (1999) Quantum chemistry. Kluwer Academic/Plenum Publishing, New York Wood WW (1986) Early history of computer simulations in statistical mechanics molecular dynamics simulation of statistical systems. Proceedings of the Enrico Fermi Summer School. Varenna, pp 3–13

C

Computer-Aided Models Giorgio De Luca Institute on Membrane Technology ITM-CNR, University of Calabria, Rende (CS), Italy

Models can be divided into mathematical and structural (chemical models), both connected to each other. Mathematical models and procedures can be numerical algorithms, in which each single step is simple arithmetic and logical relations, or they can be defined by analytical relationships, regardless of how they are evaluated. In either case, mathematical models and procedures are the result of some assumptions or approximations based on structural or chemical models (Allen and Tildesley 2003; De Luca et al. 2006, 2008). For example, the description of a droplet formation during membrane emulsification by means of analytical force-balance relationships requires some approximations about the droplet shape and its evolution along the membrane membrane pore. Moreover, the shape of the membrane pores should be also modeled. Therefore, any mathematical model can be correlated with structural or chemical models. By using computational procedures, starting from quantum mechanical calculations, molecular dynamics (MD) or Monte Carlo (MC), in fact particular attention should be paid to the molecular (chemical) models. Molecular models are closely dependent on the # Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_144-2

target properties to be assessed as well as on the computational time required to get these properties. Computational time, in turn, depends on the level of theory used, that is, the mathematical approaches. For example, at the moment, quantum mechanical methods are not applicable to optimize the geometry of macromolecules as polymers or systems containing thousands of atoms like biological systems. Molecular dynamics methodologies or coarse grain Monte Carlo can be used in these cases. However, molecular dynamics approaches cannot be used to study systems in which the breaking and formation of bonds or noncovalent bonds are decisive. Thus, in the latter case, the choice of a chemical model of macromolecules is crucial. These structural models, also called analogues, inevitably lead to neglect some aspects; nevertheless in some cases these may be irrelevant if the choice of the analogues is done correctly. In fact, albeit molecular models certainly introduce approximations in the evaluation of the macromolecular or biological proprieties, some functions of these only depend on a limited part of the whole structures. Thus, chemical models can mimic very well the function of complex systems (Gademann et al. 2007; Z€urcher et al. 2006; Saxer et al. 2010). Some examples, concerning models of carbon nanotubes (CNT), have been presented in Fig. 1. In summary, mathematical models, computational methods, and structural (chemical) models

2

Computer-Aided Models

Computer-Aided Models, Fig. 1 CNT structural models

carefully tune as a function of the type of calculation which is required to be done. Huge literature exists about studies on any kind of chemical models (periodic surfaces, slabs, or clusters) and correlated computational procedures (periodic calculations, embedded clusters, and quantum mechanics/molecular mechanics methods, etc.).

References Allen MP, Tildesley DJ (2003) Computer simulation of liquids. Oxford University Press, New York Gademann K, Bethuel Y, Locher HH, Hubschwerlen C (2007) Biomimetic total synthesis and antimicrobial evaluation of anachelin H. J Org Chem 72:8361–8370

De Luca G, Drioli E (2006) Force balance conditions for droplet formation in cross-flow membrane emulsifications. Journal of Colloid and Interface Science 294:436–448 De Luca G, Di Maio FP, Di Renzo A, Drioli E (2008) Droplet detachment in cross-flow membrane emulsification: Comparison among torque- and force-based models. Chemical Engineering and Processing 47:1150–1158 Saxer S, Portmann C, Tosatti S, Gademann K, Zurcher S, Textor M (2010) Surface assembly of catecholfunctionalized poly(L-lysine)-graft poly(ethylene glycol) copolymer on titanium exploiting combined electrostatically driven self-organization and biomimetic strong adhesion. Macromolecules 43:1050–1060 Z€ urcher S, Wackerlin D, Bethuel Y, Malisova B, Textor M, Tosatti S, Gademann K (2006) Biomimetic surface modifications based on the cyanobacterial iron chelator anachelin. J Am Chem Soc 128:1064–1065

C

Constant-Volume Diafiltration Zoltán Kovács Department of Food Engineering, Institue of Bioengineering and Process Engineering, Szent Istvan University, Budapest, Hungary

Constant-volume diafiltration (also called constant-volume dilution mode) is a batch diafiltration process for separating microsolutes from macrosolutes, in which the volume of the process liquor is kept constant during filtration by continuously adding a diluant into the feed tank at a rate equal to the permeation rate. The schematic representation of its configuration is shown in Fig. 1. The constant tank volume can be maintained by the use of a ball float valve or by means of liquid level controller (Beaton and Klinkowski 1983). As filtration progresses, the concentration of membrane-permeating microsolutes in the feed tank continuously decreases, while that of the macrosolutes remains ideally unchanged (or close to constant in case of incomplete rejection). Note that some literature sources misleadingly refer to “constant-volume diafiltration” as “continuous diafiltration.” Although the addition of diluant is performed in a continuous manner, constant-volume diafiltration should not be

# Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_149-3

considered as a continuous process. It is a true batch process. It is advised, however, to use the term “constant-volume dilution mode” which is less common in the literature. The term “dilution mode” reveals that this technique is actually an operational mode of a membrane filtration process in which the process liquor is diluted with pure solvent. Depending on the technological goal, either the permeate or the retentate represents the phase of primary economic importance. Thus, the product of the operation is either formed in the feed tank as a purified mixture of macrosolutes or collected in the permeate tank where the microsolutes are accumulated. Both treatments pose a dynamic modeling problem. A mass balance on component i in the feed tank gives the following initial value problem for its concentration cf,i(t): 8 < dcf , i ðtÞ ¼ cf , i ðtÞqðtÞ½Ri ðtÞ 1 Vf dt : ½3 mmcf , i ð0Þ ¼ c0f , i where q(t) and Ri(t) are the permeate flow and the solute rejection that are subject to change during operation. The constants Vf and cf,i denote the volume of the feed tank and the initial feed concentration of component i. The equation describes the evolution in time of the feed

2

Constant-Volume Diafiltration

Constant-Volume Diafiltration, Fig. 1 Schematic representation of constantvolume diafiltration settings

Retentate

Diluant u(t)

Permeate q(t)

Membrane module

Feed diafiltration level

Feed tank

0 Microsolute reduction in percentage [%]

Constant-Volume Diafiltration, Fig. 2 Dependence of microsolute removal on applied diavolumes in constant-volume diafiltration

R = 0.9 R = 0.8 R = 0.7 90 R = 0.6 R = 0.5 R = 0.4

99

R = 0.3 R = 0.2 R = 0 R = 0.1 0

1

concentration cf,i(t) assuming that the diluant consists of no component i and the feed tank is well mixed. In many applications, the flux and the rejections are concentration-(inter)dependent quantities (Kovács et al. 2009) and may vary with operating conditions such as temperature, applied pressure, and hydrodynamics. In such cases, no closed form solution of the set of resulting complex differential algebraic equations exists; thus, numerical techniques are required to solve the model equations. Under the simple assumptions of constant permeate flow and rejections,

2

3

4 5 6 Diavolumes [-]

7

8

9

10

however, the problem can be reduced to the following algebraic expression:

cf , i ðtÞ ln ¼ DðRi 1Þ% cf , i ð0Þ

for

i ¼ 1, 2, N

where D is the diafiltration factor (also called diavolume) that is defined as the ratio of applied volume of diluant to feed volume. For applications where the objective is to reduce the microsolute concentration by a fixed amount, the necessary diavolumes can be determined based on Eq.2 as illustrated in Fig. 2.

Constant-Volume Diafiltration

In practice, constant-volume dilution mode is frequently preceded and/or followed by batch concentration operational steps in order to reduce the initial feed volume to a desired level and, thus, to concentrate the macrosolutes (for further details, see entry on “traditional diafiltration”).

3

References Beaton NC, Klinkowski PR (1983) Industrial ultrafiltration design and application of diafiltration processes. J Sep Process Technol 4(2):1–10 Kovács Z, Discacciati M, Samhaber W (2009) Modeling of batch and semi-batch membrane filtration processes. J Membr Sci 327(1–2):164–173

C

Contactor-Type Catalytic Membrane Reactor Juergen Caro Institute of Physical Chemistry and Electrochemistry, Leibniz University, Hannover, Germany

Synonyms Catalytic diffuser; Pore-through-flow catalytic membrane The term “catalytic membrane contactor” refers to a device in which a membrane containing a catalytically active phase is used to provide the reaction zone for conversion of one or more reactants from one or more fluid phases (Dittmeyer and Caro 2008). The membrane not always has a separation function, it provides the surface-rich reaction zone for a gas-liquid, but also for a gas-gas and a liquid-liquid reaction. Applications of membranes with built-in catalysts for gas-liquid reactions have been also reviewed in Dittmeyer et al. (2004). The catalytic diffuser concept can be utilized for solid-catalyzed gas/liquid reactions in various

# Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_151-3

ways. If the membrane is not wetted, the operation principle is similar to that of a membrane contactor. The gas is on the support side, and the liquid is pumped through the shell side. The gaseous reactants enter the liquid phase at the gas/liquid contact plane which is established at the pore mouth towards the external membrane surface. The catalyst is deposited on this surface and gets in contact with the reactants. On the contrary, if the membrane is wetted the gas and not the liquid is at overpressure as described in Dittmeyer and Caro (2008). Gas–liquid contactors without catalytic function can be used in gas adsorption. Examples are carbon dioxide removal from gas mixtures using monoethanolamine (Simons et al. 2009) or ionic liquids (Albo et al. 2010) as carbon dioxide absorbing solutions. For liquid-liquid and gas-gas catalytic membrane contactors, see Dittmeyer and Caro (2008) (Fig. 1). Catalytic diffuser with wettable membrane: The active material is placed solely into the surface layer of an asymmetric membrane. By applying overpressure on the gas side, above the bubble point pressure of the intermediate layer but below that of the surface layer, the gas/liquid contacting plane is established inside the

2

Contactor-Type Catalytic Membrane Reactor

membrane close to the surface layer. In this way, a short diffusion path for the liquid and for the gas reactant is achieved (after Dittmeyer and Caro (2008)).

References

Contactor-Type Catalytic Membrane Reactor, Fig. 1 Catalytic diffuser with wettable membrane as an example for a Contactor-Type Catalytic Membrane Reactor for a solid-catalyzed gas-liquid reaction. By applying overpressure on the gas side, the gas-liquid contacting plane is established inside the membrane close to the surface. In this way, a hort diffusion path for the liquid and for the gas is achieved. Reproduced from Roland Dittmeyer, Karel Svajda, Martin ReifA review of catalytic membrane layers for gas/liquid reactionsTopics in Catalysis, 29 (2004), 3-27 reprinted with permission from Elsevier)

Albo J, Luis P, Irabien A (2010) Ind Eng Chem Res 49:11045–11051 Dittmeyer R, Caro J (2008) Catalytic membrane reactors. In: Ertl G, Kno¨zinger H, Sch€ uth F, Weitkamp J (eds) Handbook of heterogeneous catalysis. Wiley-VCH, Weinheim, pp 2198–2248 Dittmeyer R, Svajda K, Reif M (2004) A review of catalytic membrane layers for gas/liquid reactions. Top Catal 29:3–27 Simons K, Nijmeijer K, Wessling M (2009) J Membr Sci 340:214–220

C

Inlet Outlet þ Generation ¼ Accumulation

Continuous Stirred Tank Membrane Reactor (CST-MR) Giuseppe Barbieri Institute on Membrane Technology, Italian National Research Council, Rende (CS), Italy

The mass balance for the component i-th considering a single reaction will be: FInlet i

Before introducing the continuous stirred tank membrane reactor (CST-MR), it is useful/helpful to report about the continuous stirred tank reactor (CSTR). The continuous stirred tank reactor (CSTR) is an ideal reactor model assuming perfect mixing, with no spatial gradients of any variable such as species concentration, temperature, pressure, etc. The effect of the perfect mixing is the same value for, e.g., concentration, temperature, etc., in any point of the whole reactor volume; and their values are equal to those of the stream exiting the reactor. In addition, the reaction rate has the same value in any point of the reactor. Therefore, this reactor model operating at the lowest reactant concentration and the highest product concentration results in the lowest reaction rate. However, an important effect of perfect mixing is an easy temperature and reaction rate control, which results quite simply. Figure 1 shows a CSTR scheme. The generic balance equation on the reactor volume will be:

(1)

FOutlet i

þ

reactions NX

ni, j r j V Reaction ¼ CSTR

j¼1

dN i (2) dt

Mass balances provide a set of ordinary differential equations (ODEs), in which the number of moles of component i-th is time dependent. In Eq. 2, FiInlet and FiOutlet are, respectively, the inlet and outlet molar flow of the component i-th, ri is the reaction rate, Ni is the number of moles, and V is the reactor/reaction volume. In steady-state condition, the mass balance equations fall into algebraic equations: FInlet FOutlet þ i i

reactions NX

ni, j r j V Reaction ¼0 CSTR

(3)

j¼1

Equation 3 provides the design equation Eq. 4 for a CSTR in steady-state conditions: V Reaction ¼ CSTR

FInlet FOutlet i i reactions NX ni, j r j

(4)

j¼1

An increase in reaction rate ri causes a reduction of reactor volume (Eq. 4). The design equation # Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_152-1

2

Continuous Stirred Tank Membrane Reactor (CST-MR)

Continuous Stirred Tank Membrane Reactor (CST-MR), Fig. 1 Continuous stirred tank reactor (CSTR) scheme

transfer coefficient, A is the heat exchange area, and DH is the enthalpy of the chemical reaction. This equipment is very commonly used in continuous industrial processes as well as in the plug flow reactor. For reactions of greater than zero order, CSTR always requires a volume larger than that of a plug flow reactor to achieve the same conversion; this is owing to the lower reactant concentrations, as said, at which it operates. Therefore, the use of CSTR is recommended when the desired reaction rate is smaller than that of the side reaction one in order to limit by-product formation.

Continuous Stirred Tank Membrane Reactor (CST-MR), Fig. 2 Continuous stirred tank membrane reactor (CST-MR)

also provides the reactor volume necessary for obtaining the exit flow rate FiOutlet, from the feed conditions FiInlet, and the reaction rate knowledge. The energy balance, coupled to mass balance, is given by Eq. 5: Conversion ¼ Cp T Outlet T Inlet UA T Outlet T External DH Reaction (5) where Cp is the mean heat capacity, TExternal is the external temperature, U is the overall heat

CST Membrane Reactor A membrane reactor is a device that combines in one unit a chemical reaction with selective product separation by means of a permselective membrane. The selective removal of products from the reaction side to the permeate side also provides an increase, e.g., of the reaction rate and equilibrium conversion (Fig. 2). A continuous stirred tank membrane reactor [1, 2] (CST-MR) is a device characterized (1) in addition to the same properties described above for the CSTR (perfect mixing that has no spatial gradients of species concentration, temperature, pressure, etc.) on both reaction and separation sides (2) by the selective removal of reaction product by the membrane allowing improved performance (see later on) with respect to a CSTR.

Continuous Stirred Tank Membrane Reactor (CST-MR)

3

The generic mass balance (Eq. 1) is still valid and has to be written down for both reaction and permeation even though another term has to be included: the one taking into account the permeation through the membrane. Therefore, Eqs. 6 and 7 include the permeating flow rate (AMembrane JPermeating, the product of the membrane area and permeating flux). This term is negative (being mass leaving the reaction volume) on the retentate side and positive (entering the permeation volume) on the permeate one. Reaction side:

FPermeate þ AMembrane J i Permeating FSweep i i ¼

dN Permeate i dt

(7)

The mass balance equations for steady state have an algebraic form: Reaction side:

FFeed FRetentate AMembrane J i Permeating i i þ

reactions NX

ni, j r j V Reaction CSTMR

j¼1

¼

dN Retentate i dt

(6)

Permeation side:

FFeed i

FRetentate i

A

Membrane

Ji

Permeating

þ

reactions NX

ni, j r j V Reaction CSTMR ¼ 0

j¼1

V Reaction CSTMR

¼

FFeed FRetentate AMembrane J i Permeating i i reactions NX

(8)

ni , j r j

j¼1

performance of a continuous stirred tank reactor (CSTR), specifically, for the following points:

Permeation side: FSweep FPermeate þ AMembrane J i Permeating ¼ 0 i i FSweep AMembrane J i Permeating ¼ FPermeate i i (9) Equation 8 also identifies the reaction volume. It results lower than the CSTR reaction volume owing to the permeation through the membrane. Therefore, a continuous stirred tank membrane reactor (CST-MR) has a higher

1. Higher conversion owing to the selective permeation through the membrane 2. Higher reaction rate owing to the lower concentration of the products removed by the selective membrane 3. Lower reaction volume at the same productivity owing to the species permeation

4

References Barbieri G.; Scura F.; Brunetti A. Series Membrane Science and Technology, Volume 13 Inorganic Membranes: synthesis, Characterization and Applications;

Continuous Stirred Tank Membrane Reactor (CST-MR) Chapter 9-Mathematical modelling of pdalloy membrane reactors, 325–400 Raich, B. A., and Foley, H.C (1995). Supra-equilibrium conversion in palladium membrane reactors: Kinetic sensitivity and time dependence. Applied Catalysis A.: General 129, 167–188

C

Copolyimide Precursors Tauqir A. Sherazi Department of Chemistry, COMSATS Institute of Information Technology, Abbottabad, Pakistan

Polyimide Precursors Polyimides are a class of thermally stable polymers that are often based on stiff aromatic backbones derived from aromatic dianhydrides and aromatic diamines. Polyimides due to their unusual properties are finding a wide range of applications and thus used as precursor (a precursor is a compound that participates in the chemical reaction that produces another compound) to develop materials for various applications. The chemistry of polyimides is in itself a vast area with a large variety of monomers available and several methodologies available for synthesis. The general formula and chemical structure of polyimide is shown in Fig. 1. Few of the specialty properties of polyimides include: High thermal and thermo-oxidative stability up to 400 C (750 F) Excellent mechanical properties, both at room temperature and elevated temperatures Film- and fiber-forming ability # Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_153-2

Excellent adhesive properties, both at room temperature and elevated temperature Nonflammability – will not support combustion

Applications of Polyimide Precursors Owing to its excellent chemical and thermal stability, polyimides have wide range of applications such as in electronics, aircraft, automobile, medical, machining, gas purification, aerospace, and military applications. For application in electronic and microelectronics industry, fluorine may be introduced to polyimides which reduced the dielectric constant along with retaining the other characteristics mentioned above (Hermciuc et al. 2000). Formation of nanofoam polyimides is a novel approach for reducing the dielectric constant. Basically in nanofoam formation, polyimide polymer replaces with air which reduces the dielectric constant (Hedrick and Charlier 1994). The membrane developed by polyimide containing functional or pendant groups in the backbone of the aromatic dianhydride and/or diamine exhibits good as well as selective gas permeability. Hyperbranched polyimide (HBPI) precursors were also studied to develop high-performance carbon molecular sieve membranes for improved gas separation applications. It was found that the unique hyperbranched network structure found in HBPI possesses great potential to produce carbon molecular sieve membranes with superior

2

Copolyimide Precursors

Copolyimide Precursors, Fig. 1 Chemical structure of polyimide

development of polyimide oligomers for improved processability (Smith and Connell 2000; Simone and Scola 2000).

performance (Sim et al. 2013). It is also reported that the membrane produced from the pyrolysis of a hollow-fiber polyimide precursor under suitable conditions has good separation properties when applied to mixed gas pairs, such as O2/N2, CO2/CH4, and H2/CH4 (Jones and Koros 1994). Polyimide-based membranes have been extensively studied for removal of CO2 particularly from natural gas and found few of them quite efficient (Xiao et al. 2009). Soluble and optically transparent fluorine-containing photoreactive polyimide precursors were developed. These precursors offer high-resolution patterns with aspect ratio of more than 2.0. In these polyimide precursors, the polymers which have a benzophenone segment in the polymer backbone are selfsensitized and show interesting photochemical reactions (Omote et al. 1989). Polyimide processability is one of the issues associated which cause hurdle to manufacture polyimide parts, such as composites, at costs competitive to other metal parts. Improvement in polyimide processability is essential which could be achieved by reduction in its melt viscosity. Low-molecular-weight end-capped oligomers are potential candidates for the

References Hedrick JL, Charlier Y (1994) High temperature polyimide nanofoams. Polym Prepr 35:245–346 Hermciuc S, Hamciuc E, Sava I, Diaconu I, Bruma M (2000) New fluorinated poly(imide-ether-amide)s. High Perform Polym 12:205–276 Jones CW, Koros WJ (1994) Carbon molecular sieve gas separation membranes-I. Preparation and characterization based on polyimide precursors. Carbon 32:1419–1425 Omote T, Koseki K, Yamaoka T (1989) Soluble and optically transparent fluorine-containing photoreactive polyimide precursors: Spectral sensitization by organic peroxide and organic dye combination. Polym Eng Sci 29(14):945–949 Sim YH, Wang H, Li FY, Chua ML, Chung T-S, Toriida M, Tamai S (2013) High performance carbon molecular sieve membranes derived from hyperbranched polyimide precursors for improved gas separation applications. Carbon 53:101–111 Simone C, Scola DA (2000) Novel fluorinated polyimidess. Proc Fluoro Polym Am Chem Soc 15–18 Smith JG Jr, Connell JW (2000) Chemistry and properties of imide oligomers fom phenylethynyl containing diamines. High Perform Polym 12:213–223 Xiao Y, Low BT, Hosseini SS, Chung TS, Paul DR (2009) The strategies of molecular architecture and modification of polyimide-based membranes for CO2 removal from natural gas – a review. Prog Polym Sci 34:561–580

C

Cost-Effective Gas Separations Pushpinder S. Puri PuriMem, LLC, Emmaus, PA, USA

The separation of gases is an essential unit operation for a large number of chemical processes. The membrane-based gas separation processes are one of the several options available to the process engineers. Therefore, a membrane gas separation process has to compete with the more established gas separation processes such as cryogenic distillation, absorption, and adsorption. Although gas separation membranes can be used for a wide range of applications, they have generally found their applications only in a few niche markets. These applications are listed in Table 1. In these markets membranes may not be the most cost-effective gas separation process but offer other advantages which make a compelling case for their use. These non-tangible benefits of the membrane gas separation processes are listed in Table 2. Two necessary but not sufficient properties of the gas separation membranes are its selectivity for a given pair of gases and the gas permeation rate of the faster gas. The membrane selectivity is associated with the energy usage and/or product loss (energy cost), and the gas permeation rate is a measure of the membrane area needed (capital cost) for a given separation. Thus, these two factors have a major impact on the economic # Springer-Verlag Berlin Heidelberg 2015 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_154-3

viability of a given membrane for a separation of interest. In addition, these two properties dictate the geometry of the membrane (tubular, hollow fiber of flat sheet), design of the membrane module, configuration of the membrane cascade (single stage or multiple stages), and process flow schemes (parallel, series, recycle, sweep, etc.) (Agrawal and Xu 1996). They also determine the operational mode of the membrane such as pressurized feed gas with or without intermediate compression or membrane module operation with vacuum on the permeate side while keeping the feed gas a little over the atmospheric pressure. All these factors contribute to the bottom line economics of the gas separation process. The life of the membrane and the replacement cost of the membrane module are a major contributor to the cost effectiveness of the membrane gas separation. For commercial membrane gas separations, the membrane life varies from application to application. A major contributor to the life of the membrane is its chemical compatibility with the feed gases. A membrane material may be chemically inert to the major gases in the feed gas, but the trace impurities present in it may damage the membrane over a period of time by causing compaction or other defects. In those cases, the feed gas may be pretreated to remove trace impurities before the gases are fed to the membrane separator. The type and extent of cleaning of the feed gases has a major bearing on the overall cost of gas separation by membranes.

2

Cost-Effective Gas Separations

Cost-Effective Gas Separations, Table 1 Commercial applications of gas separation membranes Nitrogen production from air Air-drying H2/CO adjustment in synthesis gas H2 recovery in ammonia synthesis H2 recovery in petrochemical processes Acid gas removal from natural gas Recovery and recycle of olefins in olefin polymerization process

Cost-Effective Gas Separations, Table 2 Benefits of membrane gas separation process

Lower foot print and weight No moving parts (except for compressors) Remote operation Easy turnup/turndown Portability No open flames

Hybrid processes consisting of a membrane separation process with a conventional gas separation process sometimes offer a more costeffective solution for gas separation. Two examples of these processes which are commercially used are in production of high purity nitrogen and in the treatment of natural gas. Air separation membranes have selectivity limitations;

therefore, they are not best suited to produce high purity nitrogen. However, when a membrane air separation process is combined with an adsorption process, membrane separation produces 95–99 % nitrogen from which final traces of oxygen is removed by adsorption process (Choe et al. 1987). The second application of hybrid process is for the removal of acid gases from the natural gas. Here, a hybrid process consisting of acid gas selective membrane followed by an amine scrubbing process offers a more economical process for natural gas upgrading (Baker and Lokhandwala 2008).

References Agrawal R, Xu J (1996) Gas-separation membrane cascades utilizing limited numbers of compressors. AIChE J 42:2141 Baker RW, Lokhandwala K (2008) Natural gas processing with membranes: an overview. Ind Eng Chem Res 47(7):2109–2121 Choe JC et al (1987) Process for separating components of a gas stream, US Patent 4,701,187

Further Reading Drioli E, Barbieri G (eds) (2011). Membrane engineering for the treatment of gases, volume 1: gas-separation problems with membranes. RSC Publishing, Cambridge, UK Paul DR, Yampol’skii YP (1994) Polymer gas separation membranes. CRC Press, Boca Raton, Fl, USA

C

Crown Ethers P. K. Mohapatra Radiochemistry Division, Bhabha Atomic Research Centre, Mumbai, Maharashtra, India

Pedersen (1967) synthesized the crown ethers (see Fig. 1a) for the first time in 1967 which formed stable complexes with alkali/alkaline earth metal ions and displayed a unique selectivity based on the size compatibility of the ligand cavity size and the ionic size of these metal ions. The crown ethers mimic the biological receptors such as valinomycin (see Fig. 1b) which selectively transports K+ ion as compared to Na+ ion (by a factor of 105) across cell membranes. The transport properties of the crown ethers depend on the size of the crown ring, number of donor atoms, nature of donor atoms, nature of lipophilic/ionizable side arms, etc. Crown ethers have been overwhelmingly used for the transport of alkali metal/alkaline earth ions, though there have been quite a large number of reports on the transport of Ag+, Hg2+ (with thia-crown ethers), and Pb2+ which are proposed for heavy metal removal from wastewaters (Lamb et al. 1980). There have also been several reports on the use of crown ethers for the selective transport of lanthanides and actinides though size selective factors are far less pronounced in such cases. Another major application of crown ethers for metal ion transport includes the recovery of # Springer-Verlag Berlin Heidelberg 2014 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_162-4

radio-cesium and radio-strontium from radioactive wastes which can significantly reduce the load on the vitrified glass blocks (Dozol et al. 1995; Dozol and Casas 1994). Crown ether-based supported liquid membranes have shown large amounts of acid cotransport when attempts have been made for the transport of Cs-137 and Sr-90 from nitric acid feeds. This has affected the overall efficiency of the transport process, and appropriate selection of the diluent has helped in overcoming this issue (Raut et al. 2012). Though crown ethers have been employed for the transport of Cs(I) in laboratory-scale studies, calix crowns (where crown ether structure has been appended to a calix[4]arene) have been proposed as one of the most efficient carrier extractants for the effective transport of radio-cesium (Casnati et al. 1995). Transport of organic/biological receptors like cytochrome C (Paul et al. 2003) and amino acids (Yamaguchi et al. 1988) has also been facilitated by crown ethers. Usually, the neutral macrocyclic ionophores like crown ether, cryptand, calixarenes, cavitands, etc. require a large lipophilic counter anion like picrate, tetraphenyl borate, etc. for the effective transport of the cationic species from the aqueous phase to the membrane phase. The ion pairs are subsequently transported across the membrane phase which usually contains a polar diluent such as chloroform, long-chain alcohols, ethers, etc. for the stabilization of the charged metal-carrier (crown ether) complex. However,

2

Crown Ethers

O O

O

HN

O

O

H N

O O

O

O

O

O

O HN

O

OCH2COH

O

O

O

O

NH

O O

O

O

H

O

O

O O

O

O N H

O O

NH

O

Crown Ethers, Fig. 1 Structural formulae of (a) a typical crown ether, (b) valinomycin, and (c) an ionizable lariat crown ether

the large molar volume counter anions can make the diffusion of the ion pair rather slow. Appending an ionizable pendent arm to the crown ether ring (the ligand is called an ionizable lariat ether; see Fig. 1c) can eliminate the need for the counter anions making it a more efficient transport system (Strzelbicki et al. 1989). Other types of functionalization have been found to impart exotic properties to the receptors. For example, crown ethers with suitable functional groups based on redox-switched (Shinkai et al. 1985a), thermosensitive (Shinkai et al. 1985b), and photoresponsive (Shinkai et al. 1981) properties have also been used for the transport of receptors. Functionalized crown ether-type ligands with carboxylate groups have been suggested for the recovery of U from seawater which is yet another exotic application of the crown ether-based membrane transport systems.

References Casnati A, Pochini A, Ungaro R, Ugozzoli F, Arnaud F, Fanni S, Schwing MJ, Egberink RJM, De Jong F, Reinhoudt DN (1995) Synthesis, complexation and membrane transport studies of 1,3-alternate calix[4] arene-crown-6 conformers: a new class of cesium selective ionophores. J Am Chem Soc 117:2767–2777 Dozol JF, Casas J (1994) Influence of the extractant on strontium transport from reprocessing concentrate solutions through flat-sheet supported liquid membranes. Sep Sci Technol 29:1999–2018 Dozol JF, Casas J, Sastre AM (1995) Transport of cesium from reprocessing concentrate solutions through flat-

sheet supported liquid membranes: influence of the extractant. Sep Sci Technol 30:435–448 Lamb JD, Izatt RM, Robertson PA, Christensen JJ (1980) Highly selective membrane transport of Pb2+ from aqueous metal ion mixtures using macrocyclic carriers. J Am Chem Soc 102:2452–2454 Paul D, Suzumura A, Sugimoto H, Teraoka J, Shinoda S, Tsukube H (2003) Chemical activation of cytochrome c proteins via crown ether complexation: cold-active synzymes for enantiomer-selective sulfoxide oxidation in methanol. J Am Chem Soc 2003(125):11478–11479 Pedersen CJ (1967) Cyclic polyethers and their complexes with metal salts. J Am Chem Soc 20:7017–7022 Raut DR, Mohapatra PK, Manchanda VK (2012) A highly efficient supported liquid membrane system for selective strontium separation leading to radioactive waste remediation. J. Membr. Sci. 390–391: 76–83 Shinkai S, Nakaji T, Ogawa T, Shigematsu K, Manabe O (1981) Photoresponsive crown ethers. 2. Photocontrol of ion extraction and ion transport by a bis(crown ether) with a butterfly-like motion. J Am Chem Soc 103:111–115 Shinkai S, Inuzuka K, Miyazaki O, Manabe O (1985a) Redox-switchable crown ethers. 3. Cyclic-acyclic interconversion coupled with redox between dithiol and disulfide and its application to membrane transport. J Am Chem Soc 107:3950–3955 Shinkai S, Nakamura S, Tachiki S, Manabe O, Kajiyama T (1985b) Thermocontrol of ion permeation through ternary composition membranes composed of polymer/liquid crystal/amphiphilic crown ethers. J Am Chem Soc 107:3363–3365 Strzelbicki J, Charewicz WA, Liu Y, Bartsch RA (1989) Solvent extraction and bulk liquid membrane transport of Co(II) and Ni(II) ammine cations by proton-ionizable crown ethers. J Incl Phenom Mol Recognit Chem 7:349–361 Yamaguchi T, Nishimura K, Shinbo T, Sugiura M (1988) Amino acid transport through supported liquid membranes: mechanism and its application to enantiomeric resolution. Bioelectrochem Bioenerg 20:109–123

C

Cryogels Nilay Bereli, Handan Yavuz and Adil Denizli Department of Chemistry, Biochemistry Division, Hacettepe University, Ankara, Turkey

Chromatography is the most powerful technology in the downstream applications for the separation of proteins both in the analytical and large scale. Conventional packed bed columns have been used for many applications; however, they have some important drawbacks such as the slow diffusional mass transfer and the large void volume between the beads (Gun’ko et al. 2013). In order to resolve these problems, nonporous beads and perfusion chromatography packing have been designed and used as a carrier, but these adsorbents are not sufficient to resolve these limitations in essence. New-generation stationary phases i.e., polymeric cryogels, are found to have an increasing use in the separation science due to their easy preparations, excellent flow properties, and high performances compared with the conventional beads. Cryogels are

# Springer-Verlag Berlin Heidelberg 2014 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_163-1

mega-porous three-dimensional networks formed under freezing conditions. The pore size of the cryogels varies from 10 to 250 mm (Fig. 1), which can be changed by optimizing the freezing regime and type and concentrations of polymerization precursors. The unique properties of cryogels like osmotic, chemical, and mechanical stability, large pores, short diffusion path, low-pressure drop (Fig. 2), and short residence time for both binding and elution stages make them attractive matrices for affinity chromatography of large molecules such as proteins, plasmids even whole cells, as well as small molecules (Lozinsky et al. 2001; Stela and Valentina 2013; Bereli et al. 2008, 2010; Tamahkar et al. 2011; Derazshamshir et al. 2010). Therefore, cryogels can be used in the various affinity chromatography applications such as protein A affinity chromatography, histidine affinity chromatography, thiophilic affinity chromatography, metal-chelate affinity chromatography, dye affinity chromatography, ion-exchange chromatography, DNA affinity chromatography, cell affinity chromatography, molecular imprinting technique, etc.

2

Cryogels

Cryogels, Fig. 1 SEM images of the PHEMA cryogel

Pressure (MPa)

0.3

0.2

0.1

0.0

100

300 200 Flow rate (cm/h)

400

Cryogels, Fig. 2 Pressure drop at different flow rates

References Bereli N, Andac M, Baydemir G, Say R, Galaev IY, Denizli A (2008) Protein recognition via ion

coordinated molecularly imprinted supermacroporous cryogels. J Chromatogr A 1190:18 Bereli N, S¸ener G, Altintas¸ EB, Yavuz H, Denizli A (2010) Poly(glycidyl methacrylate) beads embedded cryogels for pseudo-specific affinity depletion of albumin and immunoglobulin G. Mater Sci Eng 30:323 Derazshamshir A, Baydemir G, Andac¸ M, Say R, Galaev IY, Denizli A (2010) Molecularly imprinted PHEMAbased cryogel for depletion of hemoglobin from human blood. Macromol Chem Phys 211:657 Gun’ko VM, Savina IN, Mikhalovsky SV (2013) Cryogels: morphological, structural and adsorption characterization. Adv Colloid Interface Sci 187–188:1 Lozinsky VI, Plieva FM, Galaev IY, Mattiasson B (2001) The potential of polymeric cryogels in bioseparation. Bioseparation 10:163 Stela DE, Valentina DM (2013) Design, synthesis and interaction with Cu2+ ions of ice templated composite hydrogels. Res J Chem Environ 17:4 Tamahkar E, Bereli N, Say R, Denizli A (2011) Molecularly imprinted supermacroporous cryogels for cytochrome c recognition. J Sep Sci 34:3433

D

Dehydration Wei Liu Energy and Environmental Technology, Pacific Northwest National Laboratory, Richland, WA, USA

Dehydration is a common terminology but often has different meanings in different fields. In chemistry, dehydration is a chemical reaction process about conversion of one molecule into another one by removing H and O atoms as a water molecule, such as conversion of ethanol (C2H5OH) into ethylene (C2H4). In physiology and medicine, dehydration means the excessive loss of body water. Dehydration in food processing involves removal of water from various types of food for long-term preservation. In the membrane field, dehydration generally refers to removal of water molecules from a water-containing fluid or mixture, i.e., it is a physical process. Two common membrane dehydration processes are illustrated in Fig. 1. In pervaporation, water molecules are removed from water-containing liquid as water vapor and thus, a phase change of water occurs during the process. Due to significant heat of vaporization, the liquid feed temperature will be reduced without thermal energy inputs. Continuous supply of thermal energy is necessary to conduct pervaporation under a constant temperature. In the vapor-phase separation process, water vapor # Springer-Verlag Berlin Heidelberg 2014 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_165-1

is separated out of a water vapor-containing fluid, and no phase changes are involved. Vapor-phase membrane separation is nearly an isothermal process. The driving force for water molecules to move across the membrane is typically partial pressure gradient of water vapor. For a given feed fluid, partial pressure of water vapor in the permeate side can be lowered by pulling vacuum and/or introducing a sweep gas stream. Water transport across the membrane can also be driven by chemical potential gradient of water. For example, the forward osmosis process involves water transport from water-containing liquid of a lower solute concentration to liquid of a higher solute concentration, and a membrane gas/liquid contactor for gas drying involves transport of water from water vapor-containing gases into water-absorbing liquid. However, those membrane processes are viewed as different technologies from membrane dehydration. There are a variety of applications for membrane dehydration. Dehydration is necessary for production of pure or anhydrous alcohols, because a water-alcohol mixture is often produced by fermentation or catalytic reactions from feedstock of sugars, corn, cellulose, or syngas. Ethanol fuel production represents one major application of dehydration technologies, and its worldwide production capacity reaches about 85 billion liters/year in 2012. Successful development of cellulosic ethanol technologies is expected to lead more growth of ethanol fuels.

2

Dehydration

a

T2 2000) Eq. 1: t¼

lrn2 2

(1)

• In stirred ME, the shear stress (t) [Pa] depends on the angular velocity (o) [s1] of the stirrer. The shear over the whole membrane area can be calculated with the Eqs. 2 and 3:

tmax ¼ 0:825mc or trans t ¼ 0:825c or trans

1 d

for r < r trans

r 0:6 1 trans d r

(2)

for r < r trans (3)

where mc [Pa s] is the continuous phase viscosity, rtrans is the transitional radius, and d is the boundary layer thickness the Landau given qffiffiffiffiffiffiffiffifrom Lifshitz equation d ¼ m=or . The shear stress is not uniformly distributed over the membrane surface, and it can be assumed that the maximal shear (tmax) is reached at distance rtrans from the center of the membrane; rtrans is the transitional radius in which the rotation changed from a free vortex to a forced vortex. • In rotating ME, the shear stress is directly proportional to the membrane rotational speed (n) [rpm] but depends also on the width of the annular gap between the rotating membrane and the stationary vessel. The shear at the surface of the membrane is calculated using Eq. 4: I? ¼

I?R2 n 2 1 2 15 R2 R1

(4)

where R1 is the radius of the rotating membrane and R2 is the radius of the stationary vessel. • Pulsed ME and vibrating and azimuthally oscillating ME are based on the generation of the shear stress by oscillation of either the continuous phase or the membrane. Thus, there are two parameters affecting shear stress on the membrane surface: frequency (f) [Hz] and amplitude (a) [m] of the oscillation. During the oscillation, the shear is variable and the emerging drop detaches when it experiences the maximum shear stress which is calculated using Eq. 5: 3=

tmax ¼ 2aðpf Þ 2 ðmrÞ

1= 2

(5)

where m is the viscosity and r is the density of the continuous phase. A peak shear event occurs twice per oscillation: once in each direction that the

Dynamic Membrane Emulsification

wave is moving, for a regular wave form such as a sine wave.

References Holdich RG, Dragosavac MM, Vladisavljevic GT (2010) Membrane emulsification with oscillating and stationary membranes. Ind Eng Chem Res 49(8):3810–3817 Holdich RG, Dragosavac MM, Vladisavljevic GT, Piacentini E (2013) Continuous membrane emulsification with pulsed (oscillatory) flow. Ind Eng Chem Res 52:507–515 Peng SJ, Fellow RAW (1998) Controlled production of emulsions using a crossflow membrane: part II:

7 industrial scale manufacture. Chem Eng Res Des 76(8):902–910 Piacentini E, Drioli E, Giorno L (2014) Pulsed back-andforward cross-flow batch membrane emulsification with high productivity to obtain highly uniform and concentrate emulsions. J Membr Sci 453:119–125 Silva PS, Dragosavac MM, Vladisavljevic´ GT, Bandulasena HCH, Holdich RG, Stillwell M, Williams B (2015) Azimuthally oscillating membrane emulsification for controlled droplet production. AIChE J. doi:10.1002/aic.14894 Stillwell MT, Holdich RG, Kosvintsev SR, Gasparini G, Cumming IW (2007) Stirred cell membrane emulsification and factors influencing dispersion drop size and uniformity. Ind Eng Chem Res 46(3):965–972 Vladisavljevic´ GT, Williams RA (2006) Manufacture of large uniform droplets using rotating membrane emulsification. J Colloid Interface Sci 299(1):396–402

E

Electrochemical Processing Tanja Vidakovic-Koch Max Planck Institute for Dynamics of Complex Technical Systems, Magdeburg, Germany

In electrochemical processing, electrical energy is supplied to or obtained from the electrochemical system in order for chemical production or energy conversion to take place (Bard and Stratmann 2007). The first group of processes, also called electrolytic, is not spontaneous. The second group of processes, called galvanic, is spontaneous and it delivers electrical energy. Electrolytic processes can be further divided into two categories: inorganic and organic. In inorganic electrochemical processing, important commodity chemicals such as sodium hydroxide, chlorine, and pure metals are produced. The major inorganic electrochemical processing technologies are chlor-alkali electrolysis and electrowinning of metals like aluminum or copper. Nowadays, hydrogen production by water electrolysis gets more on importance in context of chemical storage of renewable electrical energy (wind and photovoltaic) in hydrogen. The most significant commercial electroorganic synthesis is Monsanto’s electrohydrodimerization (EHD) of acrylonitrile to adiponitrile. Adiponitrile has an importance in production of nylon 6-6. Examples of galvanic systems are fuel cells and batteries. The main “product” of galvanic systems is electrical energy. # Springer-Verlag Berlin Heidelberg 2014 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_198-2

Electrochemical systems have some intrinsic advantages over other types of chemical systems like better control of a reaction rate, operation at lower temperatures, and less environmental impact. They take place in an electrochemical reactor. The design of an electrochemical reactor is influenced by the state of aggregation of reactants (gas, liquid, or solid), necessity of reactants and/or products separation, required mass transport conditions, and electrode materials. If product or reactant separation is required, an electrochemical reactor must contain a separator, which is a membrane. Major requirements on a membrane are good separation efficiency, low electrical resistance, no electron conductivity, low cost, long operating life time, good dimensional stability, and resistance to plugging and fouling. In general permeable and semipermeable membranes have been applied in electrochemical processing. Permeable membranes are porous materials filled with liquid electrolyte which permit the bulk flow of liquid through their structure and are thus nonselective regarding transport of ions or neutral molecules. In electrochemical processes, these are also referred to as diaphragms. Permeable membranes can be made of inorganic and organic materials and composites. Examples of these materials are asbestos (chlor-alkali electrolysis), polymers like polyethylene and polypropylene (batteries), or composites like polymer (polypropylene)-modified asbestos. Semipermeable membranes permit the selective passage of certain species by virtue of molecular

2

size or charge. In electrochemical processes, ion-conducting membranes (see solid electrolyte) are broadly applied. In general, ion-conducting membranes have higher separation efficiency and lower electrical resistance than diaphragms, but they are also more costly and impose higher requirements on system purity.

Electrochemical Processing

References Bard AJ, Stratmann M (eds) (2007) Encyclopedia of electrochemistry. Macdonald DD, Schmuki P (eds) Electrochemical engineering, vol 5. Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

E

Electrochemical Regeneration Tanja Vidakovic-Koch Max Planck Institute for Dynamics of Complex Technical Systems, Magdeburg, Germany

In electrochemical regeneration electrical energy is applied to restore some important property, like adsorption capacity or catalyst activity of a technical system. Electrochemical regeneration relies on principles of electrochemistry and relates to electrochemical processing. Electrochemical regeneration is conveniently conducted in situ with an electron as only reagent requiring simple handling and equipment. A technical setup for electrochemical regeneration requires in general two electrodes, an electrolyte and a power supply. In addition a membrane can be added to the setup in order to separate anode and cathode department. Some examples of electrochemical regeneration are electrochemical regeneration of activated carbon-based adsorbents in wastewater treatment and regeneration of enzymatic cofactors in electroenzymatic processes. Organic pollutants in wastewaters can be removed by adsorption using, e.g., activated carbon as an adsorbent. This process is normally operated using a batch of adsorbent with sufficient capacity to operate for many months before reaching saturation. Once loaded, adsorbent must be disposed or regenerated. One option for adsorbent regeneration is electrochemical regeneration # Springer-Verlag Berlin Heidelberg 2014 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_199-2

(Brown et al. 2004). The loaded adsorbent is located in a form of a packed or fluidized bed in the anode (anodic regeneration) or cathode (cathodic regeneration) compartment of the reactor. The efficiency of the regeneration depends on the processing time, voltage gradient, an electrolyte, and a compartment. According to literature the efficiency of cathodic regeneration is higher than of anodic regeneration. The mechanism of electrochemical regeneration is ascribed at the first place to local pH changes close to anode or cathode. At the anode side due to oxygen evolution reaction a pH decrease can be expected, while at the cathode side due to hydrogen evolution pH value will increase. This pH changes induce organic pollutants desorption. In the next step, dissolved pollutants can be oxidized at the anode. In the case of the cathodic regeneration they have first to mitigate from the cathode to the anode. This might be mass transfer controlled leaving some residues in the cathode, unless very large currents or long regeneration times are employed. Further example of electrochemical regeneration is regeneration of enzymatic cofactors in electroenzymatic processes (Wichmann and Vasic-Racki 2005). Redox enzymes are very selective and specific catalysts, which can enable a number of partial oxidation or reduction reactions for industrial applications at mild conditions. Broader industrial application of redox enzymes has been so far hindered by their dependence on expensive cofactors (e.g., nicotinamide

2

Electrochemical Regeneration

adenine dinucleotide (NAD)), which are consumed in the reaction (e.g., Eq. 1) and have to be regenerated for a process to be economical:

NADþ þ Medred ⇄ NADH þ Medox þ

þ

CO2 þ NADH þ H ⇄ HCOOH þ NADþ

(1)

Electrochemical regeneration offers a possibility of cofactor regeneration. In this respect especially regeneration of NAD has been studied since NAD-dependent oxidoreductases are of great industrial interest. The electrochemical regeneration can be represented by this reaction: þ

þ

NAD þ H þ 2e ⇄ NADH

Another strategy is to add an additional mediator according to

Medox þ þ Hþ þ 2e ⇄ Medred

(3) (4)

Electrochemical cofactor regeneration is still not a mature technology, and further improvements in electrode materials are needed to make this option feasible.

References

(2)

This reaction is however not selective enough and the kinetics is very sluggish on most known electrode materials. Some improvements have been achieved by using surface-modified electrodes.

Brown NW, Roberts EPL, Garforth AA, Dryfe RAW (2004) Electrochemical regeneration of a carbonbased adsorbent loaded with crystal violet dye. Electrochim Acta 49:3269–3281 Wichmann R, Vasic-Racki D (2005) Cofactor regeneration at the lab scale. Adv Biochem Eng Biotechnol 92:225–260

E

Electrochemistry Tanja Vidakovic-Koch Max Planck Institute for Dynamics of Complex Technical Systems, Magdeburg, Germany

Electrochemistry is a branch of chemistry which studies charge transfer processes across an electrified interface also called an electrochemical double layer (Bockris and Reddy 1988, Hamann et al. 2007). Applications of electrochemistry are broad including electrochemical processing, electroanalysis electrochemical sensors, electrochemical regeneration, and corrosion. In addition, many important processes in biological systems like photosynthesis and cell respiration are inherently electrochemical processes. The main feature of an electrochemical system is a separation of ionic- and electronic flows. Ions are flowing through the electrolyte which is exclusively an ionic conductor, while electrons flow through an outer electrical circuit which is exclusively an electron conductor. These two flows are interconverted at the electrode/electrolyte interface across the electrochemical double layer by means of an electrochemical reaction The potential difference in the electrochemical double layer is related to thermodynamics (Nernst equation) and kinetics (Butler-Volmer or Tafel equations) of an electrochemical reaction and it is a driving force for the electrochemical reaction to take place. This unique feature of # Springer-Verlag Berlin Heidelberg 2014 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_200-2

electrochemistry makes easy to control the reaction rate by electrons at different energies. Electrochemical processes can be spontaneous (Gibbs free energy, DG 0, called electrolytic. Instead in terms of Gibbs free energy, spontaneity of an electrochemical process can be expressed in terms of cell voltage, where a positive value stands for a galvanic system and a negative for an electrolytic. The relationship between the cell voltage and Gibbs energy is given by equation DG = nFUr, where n stands for number of exchanged electrons, F for a Faraday constant, and Ur for an equilibrium cell voltage. Many electrochemical systems require presence of separators. This is usually a membrane which can be a permeable, termed diaphragm, or semipermeable, termed membrane. The latter type usually in addition to separation serves as an electrolyte, so-called solid electrolyte in electrochemical systems. An example is ceramic yttria-stabilized zirconia (YSZ) membrane which has found an application in solid oxide fuel cells. The ionic conductivity of this material is provided by O2 ions.

References Bockris JO’M, Reddy AKN (1988) Modern electrochemistry. Plenum Press, New York

2 Hamann CH, Hamnett A, Vielstich W (2007) Electrochemistry, 2nd edn. Wiley-VCH Verlag GmBH, Weinheim

Electrochemistry

E

Electrodeionization Karel Bouzek Faculty of Chemical Technology, University of Chemistry and Technology Prague, Technická 5, Prague 6, Czech Republic

Electrodeionization represents a variant of electrodialysis, modified in order to allow treatment of low-salinity and low-conductivity media. This technique is typically applied to produce highpurity water suitable for use, for example, in energetics. It combines the advantages of ion exchange with those of electrodialysis. This technology is based on an electrodialysis unit with a diluate and also, in selected cases, a concentrate chamber filled with ion-exchanger particles. They can be arranged as monopolar beds (formed by particles of one polarity ion exchanger), as layered beds (cation- and anionexchanger particles filled separately in several alternating layers), or as a mixed bed (uniform mixture of both types of ion-changer particles). The ion-exchange phase takes on the role of electroconductive media, thus reducing ohmic drop in the dilute chamber. At the same time it provides a three-dimensional interface for the removal of traces of ions present in the solution. Two regions are typically distinguished in the electrodeionization operation: (i) ions removal and (ii) solvent splitting. Within the first region the electrodeionization unit works bellow mass # Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_201-1

transfer limitation. It means, flux of ions to the solution – ion exchanger interface driven by the current load used has a value well below mass transfer limitation in a dilute chamber. The function of the ion-exchange bed consists in providing a pathway for ions trapped in the dilute channel to the ion-selective membranes separating dilute/ concentrate chambers. In a second domain, however, the current load exceeds limiting current density, i.e. limiting flux of ions present from solution to the solution – ion exchanger interface. In such a case sufficient number of ions to transport corresponding electrical charge is provided by decomposition (dissociation) of the solvent (typically water). In contrast to electrodialysis, this splitting does not take place only at the solution-membrane interface but also at the contact of the cation- and anion-selective phase (Alvarado and Chen 2014). In the case of a concentrate chamber, the role of the ion-exchange phase again consists in reducing ohmic drop in the channel while maintaining the concentration of the ions in the liquid phase at a minimum to reduce back diffusion from the concentrate to the dilute chamber. The quality of the stream produced is comparable to that of the ion-exchange process. The advantage of electrodeionization is that it is a continuous process that does not require a regular regeneration phase of operation. This feature of electrodeionization has a further important advantage. It saves a significant amount of corresponding chemicals and reduces the salinity

2

of the waste streams produced. Such technology is thus a suitable component for closed loop technologies which, on ending, discharge removed salts in the solid form and avoids production of contaminated liquid streams.

Electrodeionization

References Alvarado L, Chen A, (2014) Electrodeionization: Principles, Strategies and Applications, Electrochim. Acta 132:583

E

Electrofiltration Karel Bouzek University of Chemistry and Technology Prague, Technická 5, Prague 6, Czech Republic

Electrofiltration represents a modification of dead-end membrane micro- or ultrafiltration. It targets a significant reduction of the filtration time and focuses especially on the filtration and/or concentration of colloidal substances that otherwise rapidly build up a deposit of colloidal particles on the surface of the membrane, which strongly hinders permeation of the fluid phase. The basic principle is that colloidal particles usually carry an electric charge. Hence, by applying an appropriate electric field, colloidal particles can be moved in the direction opposite to the fluid flow, thus keeping the surface of the filtrating membrane free of the deposit [1]. A schematic sketch of this arrangement is shown in the Fig. 1. As the filtrating membrane remains unimpeded by a deposit of colloidal particles,

# Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_202-1

the pressure drop across it remains relatively low. In contrast, the separator covered by the layer of colloidal particles attracted by the electric field does not permit a significant fluid flow. This results in a reduction of the shear stress forces in the deposited film of separated colloid. These facts make electrofiltration especially promising for the separation of biotechnologyderived products, the reason being that such products are typically sensitive to high shear stress forces while at the same time they are electrically charged. The mild conditions of electrofiltration thus enable their properties to be preserved during the process of their separation [2].

References Henry jr. JD, Lawler LF, Kuo CHA (1977) A solid/liquid separation process based on cross flow and electrofiltration. AIChE Journal 23:851 Go¨zke G, Posten C (2010) Electrofiltration of Biopolymers. Food Eng. Rev. 2:131

2 Electrofiltration, Fig. 1 Schematic sketch of electrofiltration principle, Fw stands for driving force due to the friction between the particle and flowing solvent molecules and Fe stands for driving force resulting from action of the applied electric field E on the particle carrying electrical charge

Electrofiltration

E

Electrolyzers Antonino Salvatore Arico CNR-ITAE Institute, Messina, Italy

One of the main processes occurring in an electrolyzer device is the water electrolysis. Electrolysis of water is the dissociation of water molecules into hydrogen and oxygen gases. For this process, in the presence of liquid water at 298 K and 1 bar, DG is 237 kJ mol 1 (corresponding to ~1.23 V), DS is 163 J mol 1 K 1 (TDS ~0.25 V), whereas DH is 286 kJ mol 1. The thermoneutral potential at which this reaction occurs in the absence of external heat supply is Eth,DH = 1.48 V (upper heating value 3.54 kWhNm 3 H2) (Millet et al. 2011). If steam is fed to the device, the reaction enthalpy is reduced by ~40 kJ mol 1 corresponding to the vaporization enthalpy. Water electrolysis is traditionally carried out in alkaline media with several commercial electrolyzers available on the market. Water electrolyzers using a solid polymer electrolyte are less common and generally use expensive materials such as noble metal electrocatalysts and Nafion membranes (Barbir 2005; Siracusano et al. 2010). Polymer

# Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_203-2

electrolyte membrane (PEM) electrolyzers represent a viable alternative to alkaline electrolyzer using KOH or NaOH as electrolytes for hydrogen generation. The advantages of SPE water electrolyzers especially concern with increased safety, high energy density, and low maintenance. In the PEM water electrolyzer, water is usually supplied to the anodic compartment where oxygen evolution occurs, whereas hydrogen is produced at the cathode by protons transported through the protonic membrane (Fig. 1). The electrodes are usually composed of a platinum electrocatalyst for hydrogen evolution, whereas metal oxides (e.g., IrO2, RuO2, etc.) are used for the anode due to their enhanced activity and stability than Pt for this reaction (Marshall et al. 2007; Siracusano et al. 2010). The performance of an SPE electrolyzer is strongly related to the characteristics of the membrane and electrode assembly (MEA) where the electrochemical reactions take place at triplephase boundary. Therefore, the interface between solid polymer electrolyte and electrocatalyst layers should be characterized by a suitable extension; furthermore, the contact resistance between the catalytic layer and the membrane should be as low as possible. Generally, Nafion®

2

Electrolyzers

Anode

H2O « 2H+ + 2e- + 0.5O2 Erev° = 1.23 V vs. RHE Metal Oxides

Cathode

2H+ + 2e- « H2 Erev° = 0.00 V vs. RHE Pt/C

O2

H2 Solid Polymer Electrolyte

A N O D E

C A T H O D E

H+ H+ H+

Electron Flow H2O H2O

H2O

« H2 + 0.5O2 E rev° = 1.23 V

Electrolyzers, Fig. 1 Principle of operation of a PEM water electrolysis cell

membrane is used as conducting polymer electrolyte in PEM electrolyzer systems. However, low levels of H2 and O2 crossover are necessary for PEMWE application due to the high-pressure operation that may reach 50–100 bars. Thus, a

proper thickness is necessary for the polymer electrolyte separator (around 100 mm). For highpressure operation in PEM electrolyzers, reinforced PFSA membranes provide a proper combination of good conductivity and high mechanical strength.

References Barbir F (2005) PEM electrolysis for production of hydrogen from renewable energy sources. Sol Energy 78:661 Marshall A, Børresen B, Hagen G, Tsypkin M, Tunold R (2007) Hydrogen production by advanced proton exchange mebrane (PEM) water electrolysers – Reduced energy consumption by improved electrocatalysis. Energy 32:431 Millet P, Mbemba N, Grigoriev SA, Fateev VN, Aukauloo A, Etie´vant C (2011) Electrochemical performances of PEM water electrolysis cells and perspectives. Int J Hydrog Energy 36:4134 Siracusano S, Baglio V, Di Blasi A, Briguglio N, Stassi A, Ornelas R, Trifoni E, Antonucci V, Arico AS (2010) Electrochemical characterization of single cell and short stack PEM electrolyzers based on a nanosized IrO2 anode electrocatalyst. Int J Hydrog Energy 35:5558

E

Electromembrane Gerarld Pourcelly Ge´rald Pourcelly, Institut Europe´en des Membranes, CC 047, Universite´ Montpellier II, Place E.Bataillon, Montpellier, France

Electromembrane or “charged membrane” stands for ion-exchange membrane [IEM]. They are used in a number of processes which are rather different in their basic concept, their practical applications, and their technical relevance (Strathmann 2004). All IEM separation processes are based on the same fundamental principle which is the coupling of the transport of electrical charges, i.e., an electrical current with a transport of mass, i.e., cations or anions, through a permselective membrane due to an externally applied or internally generated potential gradient. There are two types of IEM: (i) monopolar and (ii) bipolar membranes. Monopolar membranes are either cationexchange membranes which contain negatively charged groups fixed to the polymer matrix or anion-exchange membranes which contain positively charged groups fixed to the polymer matrix. In a cation-exchange membrane, the fixed negative charges are in electrical equilibrium with mobile cations (counterions) in the interstices of the polymer as shown in Fig. 1 (Strathmann 2010). In this case, the mobile anions are referred to as coions. They are more # Springer-Verlag Berlin Heidelberg 2015 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_204-1

or less excluded from the polymer matrix because of their electrical charge which is identical to that of the fixed ions (Donnan exclusion (Donnan 1911)). Thus, the selectivity of an IEM results from the exclusion of coions from the membrane phase. The properties of IEM are determined by different parameters such as the density of the polymer network, the hydrophobic/hydrophilic character of the polymer matrix, the nature and the ratio of fixed ion-exchange groups, the crosslinking ratio, etc. The most desired properties of IEM are (i) high chemical and thermal stabilities, (ii) high mechanical and dimension stabilities, (iii) high permselectivity, (iv) low electrical resistance, (v) and a low cost. Bipolar membranes (BPMs) are composed of two layers of ion exchangers joined by a hydrophilic junction (Pourcelly et al. 2009). The diffusion of water from both sides of the BPM allows its dissociation under the electrical field to generate protons and hydroxyl ions, which further migrate from the junction layer through the cation- and anion-exchange layers of the BPM as depicted in Fig. 2. The requirements for suitability of BPM include that for monopolar membranes but also an experimental potential to achieve the water-splitting capability as close as possible as the theoretical value equal to 0.83 V at 25 C. Nowadays, superior styrene-divinylbenzene copolymer membranes can be easily purchased, perfluorinated membranes with great chemical

2 Electromembrane, Fig. 1 (a) Cationexchange membrane with a homogeneous structure; (b) ion-exchange membrane with a heterogeneous structure prepared from an ion-exchange resin powder in a binder polymer (From Strathmann 2010)

Electromembrane

a Counter-ion pathway

Counter-ion

Co-ion

Fixed ion

Polymer matrix

b Counter-ion pathway

Solution filled gaps Ion-exchange resin

Binder polymer

Electromembrane, Fig. 2 Principle of a bipolar membrane. Left hand: water dissociation under electrical field. Right hand: the two ion-exchange layers bearing fixed anion- or cation- exchange groups

stability are on the market, and BPM with an industrial-scale lifetime (>20,000 h) is available.

References Donnan FG (1911) Theory of membrane equilibrium and membrane potential in the presence of non-dialysing electrolyte. Z Electrokem Angew Phys Chem 17:572–581

Pourcelly G, Bazinet L (2009) Developments of BPM technology in food and bio-industries. In: Pabby AK, Rizvi SSH, Sastre AM (eds) Handbook of membrane separations, CRC Press, Boca Raton, pp 581–634 Strathmann H (ed) (2004) In: Ion exchange membrane separation processes. Membrane technologies series, Elsevier, Amsterdam Strathmann H (2010) Electrodialysis: a mature technology with a multitude of new applications. Desalination 264:268–288

E

Electrophoresis Catherine Charcosset Université Lyon 1, Lyon, Villeurbanne, France

The term electrophoresis refers to the motion of suspended particles in an applied electric field. Among separation techniques, electrophoresis is widely used in research and development and quality control in disciplines such as biochemistry, immunology, genetics, and molecular biology (Westermeier 2001). Electrophoresis is based on the differential migration of charged species in a semiconductive medium under the influence of an electric field. Separation of many different kinds of species including proteins, DNA, nucleotides, drugs, and many other biochemicals is obtained upon differences in size, charge, and hydrophobicity. The technique was first reported in 1937 by Arne Tiselius who won the Nobel Prize in Chemistry in 1948 for the separation of different serum proteins by a method called “moving-boundary electrophoresis.” Since then, a number of improved techniques have been introduced such as gel electrophoresis, capillary electrophoresis, and two-dimensional electrophoresis. Gel electrophoresis uses an electric current passed through an agarose or polyacrylamide gel (SDS-PAGE) to separate the molecules in a sample on the basis of their differences in molecular size and charge. As the sample migrates in # Springer-Verlag Berlin Heidelberg 2012 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_206-1

the gel in response to the electric current, the smaller species move more quickly than the larger species, which results in a distinct banded pattern in the gel. This banded pattern may be visualized via the application of staining agents, such as ethidium bromide, which reveals the gel bands under UV light, or silver stain, which is typically used to detect proteins. The silver stain is compatible with mass spectrometry techniques for further analysis of the protein composition. Capillary electrophoresis (CE) involves a combination of both polyacrylamide gel electrophoresis (SDS-PAGE) and high-performance liquid chromatography (HPLC) (Ahuja and Jimidar 2008). High voltages of 500 V/cm or greater are generated within narrow capillaries (20–200 mm). The high voltages cause electroosmotic and electrophoretic movement of buffer solutions and ions, respectively, within the capillary. Two-dimensional gel electrophoresis (2-D electrophoresis) separates species in two steps, according to two independent properties. In a common technique, the first dimension is isoelectric focusing, which separates proteins according to their isoelectric points; the second dimension is SDS polyacrylamide gel electrophoresis (SDS-PAGE), which separates proteins according to their molecular size. The method involves placing the sample in gel with a pH gradient and applying a potential difference across it. Cellulose acetate membranes are other current supporting media for electrophoresis separation

2

(Westermeier 2001). They are used for routine clinical analysis and related applications, as well as for the analysis of molecules in physiological fluids. These membranes have large pores and therefore have a low sieving effect on molecules. The electrophoretic separation is thus entirely based on charge density. The matrix exerts little effect on diffusion so that the separated zones are relatively wide and the resolution and limit of detection area is low. For these reasons, cellulose acetate membranes are often replaced by gel electrophoresis. Other supporting membranes for electrophoresis include Nafion membranes, a type of perfluorosulfonic acid membrane, and cationexchange membranes, which are chemically resistant and consist of a pore-structure cluster network (Fang et al. 2004). These membranes are widely used in the field of chloralkali industry and in fuel cells. A Nafion membrane contains ˚ and 50–60 A ˚ in size) hydrophilic pores (10–20 A acting as very narrow electrophoresis channels. The fixed-charge sites ( SO3 ) on the hydrophilic pore surface provide a strong charged

Electrophoresis

background. Nafion membrane electrophoresis is a potentially attractive technique for the separation of small organic molecules like amino acids or ions.

Cross-References ▶ Cellulose Acetate (CA) Membrane ▶ Ion Exchange Membrane ▶ Perfluorosulfonic Acid Polymer Membrane

References Ahuja S, Jimidar M (2008) Capillary electrophoresis methods for pharmaceutical analysis. Academic, Amsterdam Fang C, Wu B, Zhou X (2004) Nafion membrane electrophoresis with direct and simplified end-column pulse electrochemical detection of amino acids. Electrophoresis 25:375–380 Westermeier R (2001) Electrophoresis in practice, 3rd edn. Wiley-VCH, Weinheim

E

Electrophoretic Deposition Catherine Charcosset Universite´ Lyon 1, Villeurbanne, France

Electrophoretic deposition (EPD), also called electrocoating, e-coating, cathodic electrodeposition, or electrophoretic coating, is a simple and effective technique for coating of charged particles on substrates (Besra and Liu 2007). It has several advantages including continuous processing, uniform deposition and control of the thickness, and morphology of a deposited film by adjustment of the deposition time and applied potential. In EPD, charged powder particles, dispersed or suspended in a liquid medium, are attracted and deposited onto a conductive substrate of opposite charge on application of a DC electric field. There are two types of electrophoretic deposition (Fig. 1). When the particles are positively charged, the deposition happens on the negative electrode (cathode) and the process is termed cathodic electrophoretic deposition. The deposition of negatively charged particles on positive electrode (anode) is called anodic electrophoretic deposition. By suitable modification of the surface charge on the particles, any of the two modes of deposition is possible. This technique is convenient for stable suspensions containing charged particles free to move when an electric field is applied. Therefore, EPD can be applied to any material that is available as a fine # Springer-Verlag Berlin Heidelberg 2014 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_207-2

powder (e.g., 90 %), but it is not very appropriate for more dilute gas streams. The main advantage of the cryogenic gas separation is that it enables direct production of liquid gas, which is often very useful for certain transport options, such as transport by ship. A major disadvantage is connected with the high amount of energy required for the refrigeration especially for dilute gas streams.

Separation with Membranes Separation of gases with membranes relies on the different affinities of one or more gases toward the membrane material, causing one gas to permeate faster (or slower) than others. It is one of the fastest growing field for gas separation techniques, especially due to the high variety of materials which the membrane could be composed of, including microporous organic polymers,

Gas Separation, Fig. 3 Schematic representation of membranes for gas separation

zeolites, ceramic, and metal-containing materials (for a more in-depth reading, see Yampolskii and Freeman (Yampolskii et al. 2010)). The gas mixture is directed into a vessel and put in contact to the membrane material which is at the interface with another vessel (Fig. 3). The mixture

Gas Separation

3

Gas Separation, Fig. 4 An example of Robeson plot, in this case O2/N2 (Carta et al. 2013). The black line represents the 1991 (Robeson 1991) upper bound, whereas the red line is the current (2008) upper bound (Robeson 2008)

is allowed to diffuse into the second vessel under a pressure gradient which promotes the mass transport through the membrane separating the retentate (slower gas) from the permeate (faster gas). The use of membranes for gas separation offers several benefits, probably the most valuable is the high cost efficiency (both for the mechanical simplicity of the system and for low-energy regeneration). In fact, they do not require thermal regeneration, a phase change, or active moving parts in their operation. Probably the greatest limitation of membranes for gas separation is derived from their trade-off relationship between permeability and selectivity for a required gas component. This means that high permeable membranes have low selectivity, requiring several run for a good separation, and highly selective membranes have low permeability, meaning long operational times. This trade-off was well addressed by Robeson in two wellknown articles (Robeson 1991, 2008) in which he studied the gas separation performance of several membrane-forming materials in terms of permeability of a particular species (PA) and selectivity toward one component of a gas pair (aA/B = PA/ PB), organizing the data in double logarithmic plots for a series of commercially selected important gas pairs such as H2/CH4, H2/CO2, and O2/N2. He

confirmed that highly selective membranes generally exhibit low permeability and vice versa. The most important outcome of this study is represented by the so-called Robeson upper bound, an empirical line which is drawn for every gas pair plot that is meant to define how good a material for gas separation is. In Fig. 4, there is a typical example (Carta et al. 2013) in which the red line represents the 2008 upper bound for the gas pair O2/N2. Supposedly, if we plot the selectivity aA/B versus permeability PA for a new membrane and the data point fall close or go over the upper bound, it is widely accepted that the material has an excellent compromise between P (rate of separation) and a (goodness of separation).

References Carta M, Malpass-Evans R, Croad M, Rogan Y, Jansen JC, Bernardo P, Bazzarelli F, McKeown NB (2013) An efficient polymer molecular sieve for membrane gas separations. Science 339(6117):303–307 Robeson LM (1991) Correlation of separation factor versus permeability for polymeric membranes. J Membr Sci 62(2):165–185 Yampolskii Y, Freeman B (eds) (2010) Membrane gas separation. Wiley, Chichester, UK, 370 pp Robeson LM (2008) The upper bound revisited. J Membr Sci 320(1+2):390–400

G

Gas Separation by Membrane Operations Mariolino Carta1 and Paola Bernardo2 1 School of Chemistry, University of Edinburgh, Edinburgh, UK 2 Research Institute on Membrane Technology, ITM-CNR, Rende (CS), Italy

The separation of mixtures of gases and vapors is required in manufacturing processes across various industries. In the last years, membrane systems are gaining a larger acceptance in industry for gas separation and are recognized as a costefficient separation able to compete with consolidated processes such as pressure swing absorption and cryogenic distillation (Bernardo et al. 2009; Sanders et al. 2013). Membrane processes have several advantages over conventional separation techniques (e.g., distillation, extraction, absorption, and adsorption), including modularity and compactness, operational flexibility, and no need for energy-intensive phase changes or potentially expensive adsorbents and/or difficult to handle solvents. The features of membrane operations allow implementing the process intensification strategy in different production cycles. Their versatility represents a decisive factor to impose membrane processes in most gas separation fields. The first membrane units were installed in ammonia plants for the separation of hydrogen # Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_262-1

from nitrogen more than 30 years ago. Today, the production of nitrogen from air is the largest application of membrane systems, owing to the demand for nitrogen to inert fuel tanks, also aboard aircrafts, and for blanketing chemical and liquefied gas shipments. Membrane systems are also applied to enrich oxygen for medical uses, for hydrogen recovery and purification in refineries, for air and gas dehydration, and for ratio adjustment of gas mixtures. Natural gas processing represents an important emerging application field (Baker and Lokhandwala 2008). The relatively low volume flow and the relatively high inlet carbon dioxide content are strong drivers for the implementation of the membrane technology in the biogas upgrading that it is at a developing stage (Makaruk et al. 2010). The challenging olefin/paraffin separation, not yet commercial, is attracting a lot of interest from the scientific community (Rungta et al. 2013). Membrane separation allows recovering and recycling valuable compounds, such as hydrogen and light hydrocarbons (ethylene, propylene, and LPG), present in different off-gas streams (Baker et al. 1998). Polymeric membranes, cheap and with an easy processability, are typically used in the commercially available membrane system for gas separation (Yampolskii 2012). Commercial modules employ composite membranes (Pinnau et al. 1988), mainly in the form of compact hollow fibers. These membranes typically operate the separation based on a solution-diffusion

2

transport mechanism: sorption of the permeant into the membrane, permeation by diffusion through the membrane, and desorption at the low-pressure side of the membrane. The experimentally observed upper bound, based on various polymeric membranes, was reported by Robeson in 1991 and then updated in 2008 (Robeson 1991, 2008), thanks to the efforts to improve the gas separation performance of ultrahigh free volume and perfluoropolymers. Glassy polymers are chosen for their sizeselective behavior (e.g., in O2/N2 or H2 separations). However, when applied to mixtures and/or at high gas activities, these materials are prone to plasticization, which causes swelling of the polymer matrix and results in a higher permeability coupled with a loss of selectivity. Strategies to overcome plasticization include thermal curing and chemical cross-linking, which reduce the polymer free volume (Wind et al. 2002). The addition of nanofillers to a polymer matrix represents an interesting solution to overcome the trade-off of the polymeric membranes and the inherent brittleness issues of inorganic membranes (Goh et al. 2011). Rubbery polymers, instead, present a solubility-controlled permeation and preferentially allow the permeation of large gas or vapor molecules in a gaseous mixture containing also smaller molecules (Grinevich et al. 2011). Their permeability, much higher than in conventional glassy polymers, increases with the critical volume of the penetrant (Matteucci et al. 2006). These materials are applied to the separation of organic vapors from non-condensable gases, treating petrochemical vent and process streams to recover valuable feedstocks (Baker 1999). An interesting new concept is the use of waterswollen thin film composite membranes for biogas purification, taking advantage of the large difference in solubility in water to become selective for CO2 (Kárászová et al. 2012). Facilitated transport membranes contain carrier agents that can react reversibly with the target gas component. Therefore, the reaction in the membrane creates another transport mechanism, in addition to the simple solution–diffusion

Gas Separation by Membrane Operations

mechanism (Huang et al. 2008). However, carrier poisoning and short life span of the polymeric membranes are typically reported (Rungta et al. 2013). Ionic liquids were considered as additives for facilitated transport membranes. Indeed, their negligible vapor pressure avoids solvent losses by evaporation, providing stability to the metallic cation dissolved inside, and acting as a medium for facilitated transport with mobile carrier (Fallanza et al. 2013). Ionic liquid gel membranes based on conventional polymers (Jansen et al. 2011) or on polymer ionic liquids (Bara et al. 2008) were proposed to increase the stability compared to supported liquid membranes. The key for new applications of membranes in challenging and harsh environments (e.g., petrochemistry) is the development of new tough, high-performance materials. In the field of inorganic membranes, metal organic frameworks were recently considered for preparing membranes to be applied to the olefin/paraffin separation (Bux et al. 2011) or as additive to a polymer matrix (Bushell et al. 2013). High free volume polymers have been investigated as gas separation membranes, combining their ease of processing and mechanical stability with the potential to surpass the polymeric upper bound for different gas pairs (Budd and McKeown 2010). Novel PIMs, characterized by a significant shape persistence, were developed, showing interesting performance for the O2/N2 separation (Carta et al. 2013). Properly designed hybrid processes, combining a membrane system with a conventional one (e.g., PSA or absorption), represent technically and economically viable solutions, able to reduce energy consumption and total costs (Esteves and Mota 2007).

References Baker R (1999) Recent developments in membrane vapour separation systems. Membr Technol 1999(114):9–12 Baker RW, Lokhandwala K (2008) Natural gas processing with membranes: an overview. Ind Eng Chem Res 47:2109–2121

Gas Separation by Membrane Operations Baker RW, Wijmans JG, Kaschemekat JH (1998) The design of membrane vapour-gas separation systems. J Membr Sci 151:55–62 Bara JE, Hatakeyama SE, Gin DL, Noble RD (2008) Improving CO2 permeability in polymerized room-temperature ionic liquid gas separation membranes through the formation of a solid composite with a room-temperature ionic liquid. Polym Adv Technol 19:1415–1420 Bernardo P, Drioli E, Golemme G (2009) Membrane gas separation. A review/state of the art. Ind Eng Chem Res 48(10):4638–4663 Budd PM, McKeown NB (2010) Highly permeable polymers for gas separation membranes. Polym Chem 1(1):63–68 Bushell AF, Attfield MP, Mason CR, Budd PM, Yampolskii YP, Starannikova L, Rebrov A, Bazzarelli F, Bernardo P, Jansen JC, Lancˇ M, Friess K, Shantarovic V, Gustov V, Isaeva V (2013) Gas permeation parameters of mixed matrix membranes based on the polymer of intrinsic microporosity PIM-1 and the zeolitic imidazolate framework ZIF-8. J Membr Sci 427:48–62 Bux H, Chmelik C, Krishna R, Caro J (2011) Ethene/ ethane separation by the MOF membrane ZIF-8: molecular correlation of permeation, adsorption, diffusion. J Membr Sci 369:284–289 Carta M, Malpass-Evans R, Croad M, Rogan Y, Jansen JC, Bernardo P, Bazzarelli F, McKeown NB (2013) An efficient polymer-based molecular sieve membranes for membrane gas separations. Science 339:303–307 Esteves IAAC, Mota JPB (2007) Gas separation by a novel hybrid membrane/pressure swing adsorption process. Ind Eng Chem Res 46(17):5723–5733 Fallanza M, Ortiz A, Gorri D, Ortiz I (2013) Polymer–ionic liquid composite membranes for propane/propylene separation by facilitated transport. J Membr Sci 444:164–172 Goh PS, Ismail AF, Sanip SM, Ng BC, Aziz M (2011) Recent advances of inorganic fillers in mixed matrix membrane for gas separation. Sep Purif Technol 81:243–264 Grinevich Yu, Starannikova L, Yampolskii Yu, Gringolts M, Finkelshtein E (2011) Solubility

3 controlled permeation of hydrocarbons in novel highly permeable polymers. J Membr Sci 378:250–256 Huang J, Zou J, Ho WSW (2008) Carbon dioxide capture using a CO2-selective facilitated transport membrane. Ind Eng Chem Res 47(4):1261–1267 Jansen JC, Friess K, Clarizia G, Schauer J, Izák P (2011) High ionic liquid content polymeric gel membranes: preparation and performance. Macromolecules 44:39–45 Kárászová M, Vejražka J, Veselý V, Friess K, Randová A, Hejtmánek V, Brabec L, Izák P (2012) A waterswollen thin film composite membrane for effective upgrading of raw biogas by methane. Sep Purif Technol 89:212–216 Makaruk A, Miltner M, Harasek M (2010) Membrane biogas upgrading processes for the production of natural gas substitute. Sep Purif Technol 74:83–92 Matteucci S, Yampolskii Y, Freeman B, Pinnau I (2006) Transport of gases and vapors in glassy and rubbery polymers. In: Yampolskii Y, Pinnau I, Freeman B (eds) Material science of membranes for gas and vapor separation. Wiley, Chichester, pp 1–48 Pinnau I, Wijmans JG, Blume I, Kuroda T, Peinemann KV (1988) Gas permeation through composite membranes. J Membr Sci 37(1):81–88 Robeson LM (1991) Correlation of separation factor versus permeability for polymeric membranes. J Membr Sci 62(2):165–185 Robeson LM (2008) The upper bound revisited. J Membr Sci 320(1–2):390–400 Rungta M, Zhang C, Koros WJ, Xu L (2013) Membranebased ethylene/ethane separation: the upper bound and beyond. AIChE J 59(9):3475–3489 Sanders D, Smith ZP, Guo R, Robeson LM, McGrath JE, Paul DR, Freeman BD (2013) Energy-efficient polymeric gas separation membranes for a sustainable future: a review. Polymer 54(4):729–4761 Wind JD, Staudt-Bickel C, Paul DR, Koros WJ (2002) The effects of crosslinking chemistry on CO2 plasticization of polyimide gas separation membranes. Ind Eng Chem Res 41(24):6139–6148 Yampolskii Y (2012) Polymeric gas separation membranes. Macromolecules 45(8):3298–3311

G

Gas/Vapor Transport Johannes Carolus Jansen1 and Marek Lanč2 1 Institute on Membrane Technology, ITM-CNR, Rende (CS), Italy 2 University of Chemistry and Technology Prague, Prague 6, Czech Republic

General Introduction Transport of gases and vapors in membranes depends first of all on their micro- and macroscopic structure. There is a fundamental difference between dense membranes, where transport takes place through the material of the membrane itself, and porous membranes, where the transport takes place through the open space of the pores in the membrane. In the latter, different transport modes exist, depending on the size of the pores and on the interaction of the gases and vapors with the membrane material and with themselves. The most representative examples of transport mechanisms are shown in Fig. 1. (A) In large pores, convective or viscous flow occurs. Such membranes find application as filters to remove particulate matter from gas and liquid streams, but they are not able to separate gases, which move as a homogeneous mixture through the pores. (B) If the average pore diameter is smaller than the mean free path between the molecules in # Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_263-1

the gas mixture at the given pressure, so-called Knudsen diffusion takes place (Knudsen 1909; Datta et al. 1992). In this case, the transport rate is inversely proportional to the square root of the molecular weight of the gas species and the selectivity is only a function of their molecular weight ratios. (C) For even smaller pores, where the size of the pores is in the range of the size of the gas molecules themselves, molecular sieving can occur. Molecules that are larger than the pore size are completely excluded, and only smaller molecules may diffuse through the pores of the membrane. Such membranes can have very high selectivities, in the case of very narrow pore size distributions. Typical examples are carbon membranes (Vu et al. 2001) or zeolite membranes (Caro and Noack 2008; Rangnekar et al. 2015). Also dense polymer membranes, when they have a combination of very high free volume and high stiffness of the polymer chains, may exhibit behavior that comes close to molecular sieving (Carta et al. 2013). (D) In the case of strong interaction of the membrane material with the permeating species or in case of readily condensable species, the latter condenses on the pore wall. In this case, what is the most permeable species depends on various factors, including the molecular dimensions of the permeating species of the mixture and on the remaining

2

Gas/Vapor Transport

Gas/Vapor Transport in Dense Membranes Fick’s Law of Diffusion The penetrant flow, J, through a nonporous membrane can be described by Fick’s first law of diffusion (Fick 1855), derived analogically to Fourier’s law for the description of heat transfer. J i ¼ Di

dci dx

(1)

where D is the diffusion coefficient and dc/dx the concentration gradient. The concentration in nonstationary conditions depends not only on the position in the continuous phase but also on time. Assuming one-dimensional diffusion, the transport behavior can be described by Fick’s second law of diffusion: Gas/Vapor Transport, Fig. 1 Schematic representation of different transport mechanisms in porous and dense membranes

aperture of the pores, as well as the mobility of the condensed species on the pore surface. (E) In the extreme situation of the previous, capillary condensation takes place and the whole pore is filled with the condensed liquid. The selectivity of these membranes for gas mixtures may be completely different from the ideal selectivity for single gases. Based on the Kundsen diffusion mechanism, small molecules would always be much more permeable than larger molecules. However, if the larger molecules condense inside the pore, the condensed phase obstructs the permeation of smaller molecules. Thus, the mixed gas selectivity may be opposite to the ideal selectivity. (F) In dense membranes, the molecules move through the bulk of the membrane material itself, or more precisely through its free volume. The rest of this chapter will focus mostly on the transport in dense membranes because of their relevance for gas separation applications.

@ci @ 2 ci ¼ Di 2 @t @x

(2)

Both equations assume that the diffusion coefficient is independent on concentration, which is true at low penetrant activity. Figure 2 describes the gas concentration profile at different times, for a previously evacuated polymer membrane after single-sided and double-sided exposure to a gas, based on Eq. 2 (Crank 1975). These profiles correspond to those in a typical permeation and sorption experiment, respectively. The Solution-Diffusion Mechanism The transport of gases and vapors in dense polymer membranes is governed by the solutiondiffusion model (Wijmans and Baker 1995, 2006). The driving force of this process in dense membranes is the partial pressure gradient of the permeating species across the membrane, or more correctly its chemical potential gradient. The process consists of three steps: first, the gas is absorbed by the membrane at the polymer/gas interface on the feed side, then it diffuses across the polymer bulk, and finally it desorbs from the membrane at the permeate side. The permeability coefficient, P, is the product of the diffusion coefficient, D, and the solubility coefficient, S:

Gas/Vapor Transport

3

Gas/Vapor Transport, Fig. 2 Concentration profiles in a flat sheet membrane (100 mm) calculated by Eq. 2, for D = 1012 m2 s-1 (Crank 1975) in the case of: (a) ideal time lag permeation experiment (Relative feed concentration = 1, permeate concentration = 0); (b) ideal

sorption experiment (Relative feed concentration = 1) Arrows represent the increasing time during the experiment. Dashed lines represent the concentration profile in time equal to permeation time lag (see below)

P¼DS

matrix, and they can be found in an excellent review (Matteucci et al. 2006). In phenomenological terms, the membrane productivity is expressed by means of its permeance, defined as an amount of permeate per unit membrane area, time, and driving force. A commonly used unit is the gas permeation unit, GPU:

(3)

where S is defined as the ratio of the equilibrium gas concentration, C, and the gas pressure, p: S ¼ C=p

(4)

The selectivity aA/B of two species A and B is defined as the ratio of the two individual permeability coefficients and this can be decomposed into a diffusion selectivity term and a solubility selectivity term: aA=B ¼

PA D A S A ¼ PB D B S B

(5)

In ideal systems, such as for the transport of light gases in rubbers, P, D, and S are constants, but this is rather an exception and in many cases D and S depend on the concentration of the permeating species in the membrane and thus on the gas pressure. Numerous correlations have been proposed, relating the gas and vapor transport properties of polymeric membranes to the molecular properties of the penetrating species and to the chemical and physical properties of the polymer

cm3STP cm2 s cm Hg m3 ¼ 2:70 109 2 STP m h bar

1 GPU ¼ 106

(6)

The permeance is a property of the membrane and depends on its effective thickness. The permeability coefficient is an intrinsic property of the material. The most commonly used unit to describe the permeability coefficient is the Barrer: 1 Barrer ¼ 1010

cm3STP cm cm2 s cm Hg

(7)

Diffusion is an activated process and solubility is a thermodynamic property. The temperature dependence of diffusivity and solubility can therefore be described by the following Arrhenius and

4

Gas/Vapor Transport

van’t Hoff relationships, respectively (Van Amerongen 1946; Costello and Koros 1992). D ¼ D0 eED =RT

(8)

S ¼ S0 eDHs =RT

(9)

where DHs is the enthalpy of sorption of the penetrant in the polymer, ED activation energy of diffusion, and D0 and S0 are the preexponential factors. Assuming solution-diffusion model (Eq. 3), temperature dependency of permeability is: P ¼ P0 eEP =RT

(10)

where P0 is preexponential factor and EP activation energy of permeation which is equal to: EP ¼ DH S þ ED

(11)

As a temperature-activated process, diffusion usually accelerates with temperature. Dissolution of the gas can be considered as a two-step process of condensation of the gas phase, followed by mixing with the polymer matrix. For light gases, the solubility therefore increases with increasing temperature, because the negative enthalpy of condensation is negligible with respect to the positive enthalpy of mixing. On the other hand, enthalpy of sorption of more condensable gases and vapors is negative due to the high negative enthalpy of condensation and the solubility decreases with increasing temperature. Temperature dependency of diffusivity is usually stronger than that of solubility, and therefore the permeability usually increases with increasing temperature (Ghosal and Freeman 1994).

Gas/Gas, Gas/Vapor, and Vapor/Vapor Separation The most important industrial applications of gas and vapor separations vary from (A) gas/gas separations to (B) gas/vapor separations, where the membrane is in contact with highly condensable

species. In the extreme case of pervaporation (C), the membrane is at one side in contact with a liquid phase and at the downstream side it is in contact with a gas phase. The type of separation process dictates the possible operation conditions and the choice of the membrane materials. (A) For simple gas/gas separations, for instance, O2/N2 separation from air for pure nitrogen production or for O2 enrichment, in principle many membrane materials can be safely used (Baker and Low 2014). The choice depends mainly on the need to achieve a high separation factor at relatively low flux or if a high flux is needed and the separation factor is less important. (B) For gas/vapor separations, the transport properties of the different species vary widely. Often the vapor consists of readily condensable large molecules, which have a high solubility in combination with a low diffusion coefficient, in contrast to the light gas, with a lower solubility and a high diffusion coefficient. In this case, the species in the mixture are likely to influence each other, directly via competitive sorption in the limited free volume available and indirectly via plasticization of the polymer matrix by the condensable species. The same situation occurs in gas/gas separations, where one of the two gases readily condenses at higher pressure, for instance, CO2. Typical examples are volatile organic compounds (VOC) removal from air (Leemann et al. 1996), or CO2 removal from natural gas (Adewole et al. 2013) or biogas, air dehydration, etc. For such separations, it may be convenient to use rubbery membranes, which are less prone to plasticization and which are solubility selective rather than diffusivity selective. (C) Pervaporation is the extreme case of vapor/ vapor separation, with condensed vapors (=liquid mixture) at the feed side of the membrane and gaseous species at the permeate side, either by application of a vacuum or by the use of a sweeping gas (Mulder and Smolders 1991). Pervaporation is particularly

Gas/Vapor Transport

Methods for Analysis of the Transport Parameters in Membranes Time Lag Method for Pure Gases The most common way to determine the basic gas transport parameters is the so-called time lag method (Crank 1975), in which the membrane is fully evacuated inside a closed permeation cell, and after exposure of the membrane to the gas at the feed side, the pressure at the permeate side is recorded as a function of time. For ideal systems, the concentration profile in the early stage of the experiment takes the form of Fig. 2a and the resulting permeation curve takes the form of Fig. 3, which is described by the following equation (Jansen et al. 2011):

4

Permeate pressure (mbar)

advantageous in the case of azeotropic liquid mixtures, where the membrane can break the azeotrope. Typical examples of practical separations are the dehydration of alcohol with hydrophilic membranes, ethanol recovery from hydroalcoholic solutions, or VOC removal from wastewater with hydrophobic membranes. In terms of the transport properties, these membranes are often strongly affected by the swelling of the polymer by the permeating species. An exception is the alcohol dehydration with glassy perfluoropolymer membranes (Scholes et al. 2015a, b). A further complication in pervaporation with respect to gas/vapor separation is the existence of the Schroder’s paradox, according to which the membrane material often behaves differently in contact with a liquid phase or in contact with the saturated vapor phase (Vallieres et al. 2006). A curiosity is that pervaporation membranes can be porous and nonselective in the dry state and become dense and selective in contact with the feed mixture (Van Der Bruggen et al. 2004, 2006).

5

3

2

1

Θ 0

25

50 75 Time (s)

100

125

Gas/Vapor Transport, Fig. 3 Typical permeation curve for analysis of pure gas permeability by the time lag method, with indication of time lag Y determined via the tangent method

pt ¼ p0 þ ðdp=dtÞ0 t RT A l Dt 1 þ pf S ser VP Vm 6 l2 with n 1 2 X ð1Þ D n2 p2 t ser ¼ 2 exp p n¼1 n2 l2 (12) in which pt is the permeate pressure at time t and p0 is the starting pressure, typically as close to zero as the vacuum pump allows. In a leak-proof instrument, the baseline slope (dp/dt)0 is usually negligible for a defect-free and well-evacuated membrane. R is the universal gas constant, T the absolute temperature, A the exposed membrane area, VP the permeate volume, Vm the molar volume of a gas at standard conditions (0 C and 1 atm), pf the feed pressure, and l the membrane thickness. The permeability coefficient, P, is calculated in the regime of quasi steady state permeation, which is defined by the simplified Eq. 13, describing the tangent to the linear part of the permeation curve:

6

Gas/Vapor Transport

pt ¼ p0 þ ðdp=dtÞ0 t þ

RT A pf P ðt Y Þ VP Vm l

(13)

The last term is the so-called permeation time lag, Y, which is usually determined from the intersection of the tangents before the onset of permeation and after reaching the quasi steady state: l2 Y¼ 6D

(14)

For a membrane of known thickness, it allows the determination of the diffusion coefficient of the gas. The gas solubility coefficient, S, can then be obtained indirectly as the ratio of the permeability to the diffusion coefficient, using Eq. 3. While the transport parameters P and D are usually obtained by using the tangents to the permeation curve, the permeation curve can also be fitted directly with Eq. 12, after expansion into a sufficient number of terms (Scheichl et al. 2005; Jansen et al. 2011). This yields the values of P, D, and S directly. An inaccurate fit in the case of a deviating curve shape is a direct indication of nonideal behavior. This happens, for instance, in the case of clustering or in the case of strong dual mode sorption (DMS) behavior. Constant Pressure: Variable Volume Method In this method, the pure feed gas or the feed gas mixture flows through the membrane cell in crossflow mode and the permeate is either collected as such or it is transported by a sweeping gas to the gas analyzer. The total permeation rate can be determined directly, for instance, by a bubble flow meter or electronic flow meters, measuring the volumetric permeate flow rate, JPermeate. The permeate flux, QPermeate, is the volumetric flow rate per unit area: QPermeate ¼

J Permeate A

(15)

When using a sweeping gas, the permeate flux can also be calculated from the known sweeping gas flow rate and the gas concentration in the

permeate/sweeping gas mixture. The individual gas permeance, P, of the ith species in a gas mixture is obtained as the ratio of its volumetric permeate flux, QPermeate, to the partial pressure difference between the feed and permeate sides: Pi ¼

xi Permeate QPermeate Feed Feed xi p xi Permeate pPermeate

(16)

in which xi is the volume fraction or mole fraction of the ith species, pFeed and pPermeate are the total feed and permeate pressures, respectively. The mixed gas selectivity of species A and B, aA/B, is then calculated as the ratio of their individual permeances: aA=B ¼

PA PB

(17)

Sorption Analysis Direct sorption analysis is the most reliable way to determine the solubility of the gas in the polymer matrix. Sorption can be determined volumetrically, gravimetrically, and with the pressure decay method (Keller and Staudt 2005), or by inverse gas chromatography (IGC) (Danner et al. 1998). Equilibrium sorption in polymers is one of the basic characteristics describing the interaction between a penetrant and a polymer. In all these methods, the polymer sample is placed inside the test cell and exposed to a given penetrant pressure. In gravimetric measurements, the mass uptake can be measured electromagnetically (Mamaliga et al. 2004), by a quartz crystal microbalance (QCM) (Mikkilineni et al. 1995) or McBain’s quartz spiral balance (Friess et al. 2011; Vopička et al. 2013). Since the sorption coefficient (solubility) is usually a function of pressure or activity, knowledge of the sorption isotherm shape is important. There are many different types of sorption isotherms (Rouquerolt et al. 1994) depending on the polymer structure and the relative difference between penetrant/polymer and penetrant/penetrant interaction. Sorption of light gases in rubbery polymers increases linearly with pressure according to

Gas/Vapor Transport

7

Henry’s law. On the other hand, vapor’s sorption can be described by numerous equations such as Flory-Huggins theory (Flory 1953), Flory-Rehner (Flory and Rehner 1943a; Flory and Rehner Jr. 1943b; Izák et al. 2003), KoningsveldKleintjens equations (Koningsveld and Kleintjens 1971), or the ENSIC model (Favre et al. 1996). In the case of dense glassy polymeric membranes, three models are the most often used. Permanent gases behave almost linearly, at low pressures following Henry’s law. Alternatively, the dual-mode sorption model, Eq. 18, gives usually a satisfactory description of the behavior (Barrer et al. 1958). This model is a combination of Henry’s law and the Langmuir sorption isotherm, assuming monolayer sorption at existing sorption sites. C ¼ CD þ C H ¼ k d p þ

ch p b 1þbp

vm h a f ð 1 f aÞ ð 1 f a þ h f aÞ

ð p

v m h p p pÞ ð h p þ p pÞ

(20)

where p* is a pressure independent constant which has the meaning of a reference pressure. When p* is equal to saturated vapor pressure, the model is equivalent to the BET model (Brunauer et al. 1938) (Fig. 4). For samples with a well-defined geometry, sorption kinetics measurements allow the determination of the diffusion coefficient by equations based on Fick’s second law (Crank 1975). The relative sorbed amount Qt/Q1 in a flat film with thickness l is given as a function of time by the following equation: 1 Dð2nþ1Þ2 p2 t Qt 8X 1 l2 ¼1 2 e (21) p n¼0 ð2n þ 1Þ2 Q1

(18)

where C is the gas concentration in the polymer, CD and CH are the Henry and Langmuir concentration contributions, respectively, kd is the linear sorption parameter, ch is the monolayer capacity, and b is the affinity parameter, reflecting the interaction strength between polymer and penetrant. In the case of vapor sorption in glassy polymers, the sorption curves often have a typical S-shape and the DMS model cannot be used. In such cases, the Guggenheim, Anderson, and de Boer (GAB) model (Guggenheim 1966) gives a better description: v¼

The corresponding concentration-time profile in a flat membrane during a sorption experiment is shown schematically in Fig. 2b. Under real conditions, a finite time is needed to charge the gas in the sorption apparatus and a correction for the assumed step-pressure-increase is necessary (Vopička et al. 2009). An example of a typical sorption kinetics curve obtained with this model is shown in Fig. 5.

(19)

where v is the mass of adsorbed vapor per mass of polymer adsorbent, vm is the capacity of the first adsorption monolayer, h defines the ratio of the adsorption strength in the first and the subsequent layers, a is the vapor activity, and f is a constant defining the deviation of the saturated vapor pressure from a chosen reference pressure. For gases, the following form of the GAB model was proposed (Vopička and Friess 2014).

Gas/Vapor Transport, Fig. 4 Gravimetric sorption isotherm of CO2 in Amine-PIM-1 (Mason et al. 2014), extrapolated with GAB and DMS models

8

Gas/Vapor Transport

separation, models to analyze and predict the mass transport in pervaporation require different approaches (Lipnizki and Tr€agårdh 2001), such as the UNIQUAC model (Heintz and Stephan 1994).

Transport in Heterogeneous and Homogeneous Mixtures

Gas/Vapor Transport, Fig. 5 Typical gravimetric sorption kinetics curve. CO2 sorption fitted with the model proposed by Vopička (Vopička et al. 2009) in microporous Tröger’s base polymer EA-TB-PIM (Carta et al. 2013) after MeOH treatment

Modeling of Transport With the increasing computational power of modern computers, modeling of structural (Heuchel et al. 2008) and transport properties of gases (Hofmann et al. 2000; Frentrup et al. 2015) and vapors (Giacinti Baschetti and De Angelis 2015) in polymeric membranes has gained a prominent position in membrane research. Especially in the description of the free volume distribution of membrane materials, computational methods offer a level of insight that no single experimental method can give. The transport can be studied at the atomistic level (Hofmann et al. 2000; Theodorou 2006), showing, for instance, the “hopping” mechanism of a gas molecule from one free volume element to the next, confirming the activated mechanism seen experimentally. Although there is often a large discrepancy between the calculated sorption and diffusion coefficients and the experimental values, the trends between different gas species are usually reproduced well in the simulations (Macchione et al. 2007; Jansen et al. 2010). Whereas molecular dynamics simulations work fairly well for small molecules at low concentration, like in gas

The description of the transport in polymers with homogeneously or heterogeneously dispersed additives or in polymer blends is much more complex than that in neat polymers. Mixed matrix membranes are currently receiving much attention (Aroon and Ismail 2010; Rezakazemi et al. 2014) because they have the potential to combine the high permeability and selectivity of inorganic (e.g., zeolites (Miller et al. 2007)), carbonaceous (Vu et al. 2003), and organometallic filler particles (metal organic frameworks, MOFs (Bushell et al. 2013; Zornoza et al. 2013; Adatoz et al. 2015)) with the good mechanical properties of polymers. There is a large number of predictive models to describe the performance of MMMs (Vinh-Thang and Kaliaguine 2013). One of the simplest and most commonly used models to describe the transport in MMMs is the Maxwell model (Shimekit et al. 2011), valid for a low concentration of spherical particles dispersed in the continuous phase: Peff ¼

Pd þ 2Pc 2fd ðPd Pc Þ Pd þ 2Pc þ 2fd ðPd Pc Þ

(22)

where the Peff is the effective permeability of the mixed matrix membrane, Pc and Pd represent the gas permeabilities in the continuous and dispersed phase, respectively, and fd is the volume fraction of the dispersed phase. Generally, the permeability of the gases through the dispersed phase depends on the overall void volume, its distribution in the filler particles, and on the channel size, which affect the free volume of the overall system. More sophisticated models take into account also the particle shape of the dispersed phase (Cussler 1990). Such particles may have a pronounced

Gas/Vapor Transport

9

effect in the case of high aspect ratios (Rodenas et al. 2014) due to the strong effect on the diffusion path length (Falla et al. 1996). Interestingly, also impermeable fumed silica (Merkel et al. 2002) or graphene (Althumayri et al. 2016) filler particles, with intrinsic barrier properties, can have a positive effect on the permeability of the membranes, when additional free volume is created at the polymer-particle interface. When the dispersed phase is another polymer, in the case of immiscible polymer blends, the transport can be described by fundamentally the same equations as the mixed matrix materials (e.g., Eq. 22). Instead, for miscible polymer blends, the permeability, Pb, is reported to obey the following equation (Robeson 2010): lnPb ¼ f1 lnP1 þ f2 lnP2

(23)

in which f1 and f2 are the volume fractions of the two polymers, respectively, and P1 and P2 are their permeabilities. It shows a linear trend when the permeabilities are plotted on a logarithmic scale and deviates from linearity in the case of (partial) immiscibility of the two polymers (Jansen et al. 2013).

Overall Performance Robeson Trade-Off Behavior Although membrane separations may have many advantages compared to traditional separation processes such as distillation or pressure swing adsorption, a limitation is the trade-off behavior between selectivity and permeability. This trend was firstly discussed by Robeson in 1991, who suggested a linear so-called upper bound for many relevant gas pairs (Fig. 6) (Robeson 1991), which were subsequently updated and extended (Robeson et al. 1994; Robeson 2008). In 2015, a new upper bound was set for O2/N2, H2/N2, and H2/CH4, based on mostly the development of polymers of intrinsic microporosity (Swaidan et al. 2015). Freeman discussed the basis of the upper bound (Freeman 1999) and concluded that a combination of high free volume and extreme rigidity of the polymer chains is needed to exceed

Gas/Vapor Transport, Fig. 6 O2/N2 1991 and 2008 upper bound curves for the selectivity versus permeability trade-off relation (Robeson 1991, 2008) with the latest upper bound suggested by Swaidan et al. (2015). The oval represents the approximate cloud of experimental data

the current upper bound (Robeson et al. 2009). This was confirmed by McKeown et al. with a novel polymer of intrinsic microporosity based on Troger’s base and ethanoanthracene (Carta et al. 2013) or benzotriptycene units (Rose et al. 2015). Alentiev presented a similar approach for the trade-off in diffusion coefficient and diffusion selectivity (Alentiev and Yampolskii 2013). Effect of Physical Aging on the Transport Properties The global nonequilibrium state of glassy polymeric membranes tends to relax over time. This process, where no chemical changes occur, is called physical aging (Struik 1978) and affects different properties of a polymer. One of these is the free volume distribution, which, in turn, is reflected in the gas transport properties of the membrane. Physical aging therefore has a strong impact on the performance of amorphous glassy gas separation membranes (Pfromm 2006), and gas diffusion is a very sensitive method to probe changes in the free volume of a polymer membrane (Jansen et al. 2009). Different aging mechanisms have been proposed. Harms claims that free volume elements diffuse towards the surface

10

Gas/Vapor Transport

of the polymer (Harms et al. 2012). However, due to the very low expected diffusion coefficient of this process, it is significant only in the case of thin layers. McCaig et al. proposed that the aging consists of two distinct processes (McCaig and Paul 2000): (i) thickness independent lattice contraction and (ii) thickness dependent diffusion of free volume. Usually, the polymer chains pack more efficiently during physical aging and the polymer becomes denser, resulting in a decrease in permeability and an increase in selectivity. Many approaches have been used to overcome physical aging and to prepare time-stable material, such as crosslinking and addition of fillers. Lau et al. showed that addition of an ultraporous additive can prevent the effect of aging on the transport properties of super glassy polymer membranes (Lau et al. 2014). Anomalous Transport As already anticipated, Fick’s first and second laws give a rather simplified representation of diffusion in polymers, which usually applies only to permanent gases at low pressures. In the majority of cases, the diffusion coefficient is not a constant, because of mutual interactions between the permeating species, giving rise to clustering (Jansen et al. 2011) and interaction between the polymer and the permeating species, “immobilizing” the latter (Mason et al. 2014) or plasticizing the polymer (Lo et al. 2010) and favoring diffusion. The solubility of gases and vapors in the polymer matrix is usually only constant at very low gas pressure or vapor activity. As a result, the simple expression for the solution diffusion model in Eq. 3 becomes concentration dependent: Pð c Þ ¼ D ð c Þ S ð c Þ

(24)

This anomalous behavior becomes even more complex in the case of mixed gas or vapor permeation, where there may also exist a coupling effect between the different species in the mixture.

References Adatoz E, Avci AK, Keskin S (2015) Opportunities and challenges of MOF-based membranes in gas separations. Sep Purif Technol 152:207–237. doi:10.1016/j. seppur.2015.08.020 Adewole JK, Ahmad AL, Ismail S, Leo CP (2013) Current challenges in membrane separation of CO2 from natural gas: a review. Int J Greenh Gas Control 17:46–65. doi:10.1016/j.ijggc.2013.04.012Review Alentiev A, Yampolskii Y (2013) Correlation of gas permeability and diffusivity with selectivity: orientations of the clouds of the data points and the effects of temperature. Ind Eng Chem Res 52:8864–8874. doi:10.1021/ie302680r Althumayri K, Harrison WJ, Shin Y, et al (2016) The influence of few-layer graphene on the gas permeability of the high-free-volume polymer PIM-1. Philos Trans R Soc London A Math Phys Eng Sci 374:20150031:in press. doi:10.1098/rsta.2015.0031 Aroon MA, Ismail AF (2010) Performance studies of mixed matrix membranes for gas separation: a review. Sep Purif Technol 75:229–242. doi:10.1016/j. seppur.2010.08.023 Baker RW, Low BT (2014) Gas separation membrane materials: a perspective. Macromolecules 47:6999–7013. doi:10.1021/ma501488s Barrer RM, Barrie JA, Slater J (1958) Sorption and diffusion in ethyl cellulose. Part III. Comparison between ethyl cellulose and rubber. J Polym Sci 27:177–197. doi:10.1016/0010-4361(90)90120-L Brunauer S, Emmett PH, Teller E (1938) Adsorption of gases in multimolecular layers. J Am Chem Soc 60:309–319. doi:10.1021/ja01269a023 Bushell AF, Attfield MP, Mason CR et al (2013) Gas permeation parameters of mixed matrix membranes based on the polymer of intrinsic microporosity PIM-1 and the zeolitic imidazolate framework ZIF-8. J Membr Sci 427:48–62. doi:10.1016/j. memsci.2012.09.035 Caro J, Noack M (2008) Zeolite membranes – recent developments and progress. Microporous Mesoporous Mater 115:215–233 Carta M, Malpass-Evans R, Croad M et al (2013) An efficient polymer molecular sieve for membrane gas separations. Science 339:303–307. doi:10.1126/ science.1228032 Costello LM, Koros WJ (1992) Temperature dependence of gas sorption and transport properties in polymers: measurement and applications. Ind Eng Chem Res 31:2708–2714. doi:10.1021/ie00012a012 Crank J (1975) The mathematics of diffusion, 2nd edn. Clarendon, Oxford Cussler EL (1990) Membranes containing selective flakes. J Membr Sci 52:275–288. doi:10.1016/S03767388(00)85132-7 Danner RP, Tihminlioglu F, Surana RK, Duda JL (1998) Inverse gas chromatography applications in

Gas/Vapor Transport polymer-solvent systems. Fluid Phase Equilib 148:171–188. doi:10.1016/S0378-3812(98)00203-9 Datta R, Dechapanichkul S, Kim JS et al (1992) A generalized model for the transport of gases in porous, non-porous, and leaky membranes. I. Application to single gases. J Membr Sci 75:245–263. doi:10.1016/0376-7388(92)85067-S Falla WR, Mulski M, Cussler EL (1996) Estimating diffusion through flake-filled membranes. J Membr Sci 119:129–138. doi:10.1016/0376-7388(96)00106-8 Favre E, Nguyen QT, Clément R, Néel J (1996) The engaged species induced clustering (ENSIC) model: a unified mechanistic approach of sorption phenomena in polymers. J Membr Sci 117:227–236. doi:10.1016/ 0376-7388(96)00060-9 Fick A (1855) Ueber diffusion. Ann Phys 170:59–86. doi:10.1002/andp.18551700105 Flory PJ (1953) Principles of polymer chemistry. Cornell University Press, Ithaca Flory PJ, Rehner JJ (1943a) Statistical mechanics of crosslinked polymer networks. J Chem Phys 11:512–520. doi:10.1063/1.1723792 Flory PJ, Rehner J Jr (1943b) Statistical mechanics of cross-linked polymer networks II. Swelling. J Chem Phys 11:521–526. doi:10.1063/1.1723792 Freeman BD (1999) Basis of permeability/selectivity tradeoff relations in polymeric gas separation membranes. Macromolecules 32:375–380. doi:10.1021/ ma9814548 Frentrup H, Hart K, Colina C, M€ uller E (2015) In silico determination of gas permeabilities by non-equilibrium molecular dynamics: CO2 and He through PIM-1. Membranes 5:99–119. doi:10.3390/ membranes5010099 Friess K, Hynek V, Šípek M et al (2011) Permeation and sorption properties of poly(ether-block-amide) membranes filled by two types of zeolites. Sep Purif Technol 80:418–427. doi:10.1016/j.seppur.2011.04.012 Ghosal K, Freeman BD (1994) Gas separation using polymer membranes: an overview. Polym Adv Technol 5:673–697. doi:10.1002/pat.1994.220051102 Giacinti Baschetti M, De Angelis MG (2015) Vapour permeation modelling. In: Basile A, Khayet AF (eds) Pervaporation, vapour permeation and membrane distillation: principles and applications. Woodhead Publishing, Oxford, pp 203–246 Guggenheim EA (1966) Applications of statistical mechanics. Clarendon, Oxford, pp 186–207 Harms S, R€atzke K, Faupel F et al (2012) Aging and free volume in a polymer of intrinsic microporosity (PIM-1). J Adhes 88:608–619. doi:10.1080/ 00218464.2012.682902 Heintz A, Stephan W (1994) A generalized solution – diffusion model of the pervaporation process through composite membranes Part I. Prediction of mixture solubilities in the dense active layer using the UNIQUAC model. J Membr Sci 89:143–151. doi:10.1016/0376-7388(93)E0222-6

11 Heuchel M, Fritsch D, Budd PM et al (2008) Atomistic packing model and free volume distribution of a polymer with intrinsic microporosity (PIM-1). J Membr Sci 318:84–99. doi:10.1016/j.memsci.2008.02.038 Hofmann D, Fritz L, Ulbrich J et al (2000) Detailed-atomistic molecular modeling of small molecule diffusion and solution processes in polymeric membrane materials. Macromol Theory Simul 9:293–327. doi:10.1002/15213919(20000701)9:63.0.CO;2-1 Izák P, Bartovská L, Friess K et al (2003) Comparison of various models for transport of binary mixtures through dense polymer membrane. Polymer 44:2679–2687. doi:10.1016/S0032-3861(03)00137-X Jansen JC, Macchione M, Tocci E et al (2009) Comparative study of different probing techniques for the analysis of the free volume distribution in amorphous glassy perfluoropolymers. Macromolecules 42:7589–7604. doi:10.1021/ma901244d Jansen JC, Friess K, Tocci E, Macchione M (2010) Amorphous glassy perfluoropolymer membranes of hyflon AD ®: free volume distribution by photochromic probing and vapour transport properties. In: Yampolskii Y, Freeman B (eds) Membrane gas separation. Wiley, Chichester, pp 59–83 Jansen JC, Friess K, Drioli E (2011) Organic vapour transport in glassy perfluoropolymer membranes: a simple semi-quantitative approach to analyze clustering phenomena by time lag measurements. J Membr Sci 367:141–151. doi:10.1016/j.memsci.2010.10.063 Jansen JC, Darvishmanesh S, Tasselli F et al (2013) Influence of the blend composition on the properties and separation performance of novel solvent resistant polyphenylsulfone/polyimide nanofiltration membranes. J Membr Sci 447:107–118. doi:10.1016/j. memsci.2013.07.009 Keller JU, Staudt R (2005) Gas adsorption equilibria. Springer, Boston Knudsen M (1909) Die Gesetze der Molekularströmung und der inneren Reibungsströmung der Gase durch Röhren. Ann Phys 333:75–130. doi:10.1002/ andp.19093330106 Koningsveld R, Kleintjens LA (1971) Liquid-liquid phase separation in multicomponent polymer systems. X. Concentration dependence of the pair-interaction parameter in the system cyclohexane-polystyrene. Macromolecules 4:637–641. doi:10.1021/ ma60023a026 Lau CH, Nguyen PT, Hill MR et al (2014) Ending aging in super glassy polymer membranes. Angew Chem Int Ed 53:5322–5326. doi:10.1002/anie.201402234 Leemann M, Eigenberger G, Strathmann H (1996) Vapour permeation for the recovery of organic solvents from waste air streams: separation capacities and process optimization. J Membr Sci 113:313–322. doi:10.1016/ 0376-7388(95)00130-1 Lipnizki F, Tr€agårdh G (2001) Modelling of pervaporation: models to analyze and predict the mass transport in pervaporation. Sep Purif Rev 30:49–125. doi:10.1081/SPM-100102985

12 Lo C-H, Hung W-S, De Guzman M et al (2010) Investigation on CO2-induced plasticization in polycarbonate membrane using positron annihilation lifetime spectroscopy. J Membr Sci 363:302–308. doi:10.1016/j. memsci.2010.07.049 Macchione M, Jansen JC, De Luca G et al (2007) Experimental analysis and simulation of the gas transport in dense Hyflon ® AD60X membranes: influence of residual solvent. Polymer 48:2619–2635. doi:10.1016/j. polymer.2007.02.068 Mamaliga I, Schabel W, Kind M (2004) Measurements of sorption isotherms and diffusion coefficients by means of a magnetic suspension balance. Chem Eng Process Process Intensif 43:753–763. doi:10.1016/S02552701(03)00077-1 Mason CR, Maynard-Atem L, Heard KWJ et al (2014) Enhancement of CO2 affinity in a polymer of intrinsic microporosity by amine modification. Macromolecules 47:1021–1029. doi:10.1021/ma401869p Matteucci S, Yampolskii Y, Freeman BD, Pinnau I (2006) Transport of gases and vapors in glassy and rubbery polymers. In: Yampolskii Y, Pinnau I, Freeman BD (eds) Materials science of membranes for gas and vapor separation. John Wiley & Sons, Ltd, Chichester, UK, doi: 10.1002/047002903X.ch1 McCaig MS, Paul DR (2000) Effect of film thickness on the changes in gas permeability of a glassy polyarylate due to physical aging: part I. Experimental observations. Polymer 41:629–637. doi:10.1016/S00323861(99)00172-X Merkel TC, Freeman BD, Spontak RJ et al (2002) Ultrapermeable, reverse-selective nanocomposite membranes. Science 296:519–522. doi:10.1126/ science.1069580 Mikkilineni SPVN, Tree DA, High MS (1995) Thermophysical properties of penetrants in polymers via a piezoelectric quartz crystal microbalance. J Chem Eng Data 40:750–755 Miller SJ, Koros WJ, Vu DQ (2007) Mixed matrix membrane technology: enhancing gas separations with polymer/molecular sieve composites. Elsevier, Amsterdam/ New York, 170, pp 1590–1596 Mulder MHV, Smolders CA (1991) Mass transport phenomena in pervaporation processes. Sep Sci Technol 26:85–95. doi:10.1080/01496399108050458 Pfromm PH (2006) The impact of physical aging of amorphous glassy polymers on gas separation membranes. In: Yampolskii Y, Pinnau I, Freeman BD (eds.) Materials science of membranes for gas and vapor separation. John Wiley & Sons, Ltd, Chichester, UK Rangnekar N, Mittal N, Elyassi B et al (2015) Zeolite membranes – a review and comparison with MOFs. Chem Soc Rev 44:7128–7154. doi:10.1039/ C5CS00292C Rezakazemi M, Ebadi Amooghin A, Montazer-Rahmati MM et al (2014) State-of-the-art membrane based CO2 separation using mixed matrix membranes (MMMs): an overview on current status and future

Gas/Vapor Transport directions. Prog Polym Sci 39:817–861. doi:10.1016/ j.progpolymsci.2014.01.003 Robeson LM (1991) Correlation of separation factor versus permeability for polymeric membranes. J Membr Sci 62:165–185. doi:10.1016/0376-7388(91)80060-J Robeson LM (2008) The upper bound revisited. J Membr Sci 320:390–400. doi:10.1016/j.memsci.2008.04.030 Robeson LM (2010) Polymer blends in membrane transport processes. Ind Eng Chem Res 49:11859–11865. doi:10.1021/ie100153q Robeson LM, Burgoyne WF, Langsam M et al (1994) High performance polymers for membrane separation. Polymer 35:4970–4978. doi:10.1016/0032-3861(94) 90651-3 Robeson LM, Freeman BD, Paul DR, Rowe BW (2009) An empirical correlation of gas permeability and permselectivity in polymers and its theoretical basis. J Membr Sci 341:178–185. doi:10.1016/j. memsci.2009.06.005 Rodenas T, Luz I, Prieto G et al (2014) Metal–organic framework nanosheets in polymer composite materials for gas separation. Nat Mater 14:48–55. doi:10.1038/ nmat4113 Rose I, Carta M, Malpass-Evans R et al (2015) Highly permeable benzotriptycene-based polymer of intrinsic microporosity. ACS Macro Lett 4:912–915. doi:10.1021/acsmacrolett.5b00439 Rouquerolt J, Avnir D, Fairbridge CW et al (1994) Recommendations for the characterization of porous solids. Pure Appl Chem 66:1739–1758. doi:10.1351/ pac199466081739 Scheichl R, Klopffer M, Benjelloundabaghi Z, Flaconneche B (2005) Permeation of gases in polymers: parameter identification and nonlinear regression analysis. J Membr Sci 254:275–293. doi:10.1016/j. memsci.2005.01.019 Scholes CA, Kanehashi S, Stevens GW, Kentish SE (2015a) Water permeability and competitive permeation with CO2 and CH4 in perfluorinated polymeric membranes. Sep Purif Technol 147:203–209. doi:10.1016/j.seppur.2015.04.023 Scholes CA, Kentish SE, Stevens GW et al (2015b) Thinfilm composite membrane contactors for desorption of CO2 from monoethanolamine at elevated temperatures. Sep Purif Technol 156:841–847. doi:10.1016/j. seppur.2015.11.010 Shimekit B, Mukhtar H, Murugesan T (2011) Prediction of the relative permeability of gases in mixed matrix membranes. J Membr Sci 373:152–159. doi:10.1016/ j.memsci.2011.02.038 Struik LCE (1978) Physical aging in amorphous polymers and other materials. Elsevier Scientific, New York Swaidan R, Ghanem B, Pinnau I (2015) Fine-tuned intrinsically ultramicroporous polymers redefine the permeability/selectivity upper bounds of membrane-based air and hydrogen separations. ACS Macro Lett 4:947–951. doi:10.1021/acsmacrolett.5b00512 Theodorou DN (2006) Principles of molecular simulation of gas transport in polymers. In: Yampolskii Y,

Gas/Vapor Transport Pinnau I, Freeman B (eds) Materials science of membranes for gas and vapor separation. Wiley, Chichester, pp 49–94 Vallieres C, Winkelmann D, Roizard D et al (2006) On schroeder’s paradox. J Membr Sci 278:357–364. doi:10.1016/j.memsci.2005.11.020 Van Amerongen GJ (1946) The permeability of different rubbers to gases and its relation to diffusivity and solubility. J Appl Phys 17:972. doi:10.1063/1.1707667 Van Der Bruggen B, Jansen JC, Figoli A et al (2004) Determination of parameters affecting transport in polymeric membranes: parallels between pervaporation and nanofiltration. J Phys Chem B 108:13273–13279. doi:10.1021/jp048249g Van Der Bruggen B, Jansen JC, Figoli A et al (2006) Characteristics and performance of a “universal” membrane suitable for gas separation, pervaporation, and nanofiltration applications. J Phys Chem B 110:13799–13803. doi:10.1021/jp0608933 Vinh-Thang H, Kaliaguine S (2013) Predictive models for mixed-matrix membrane performance: a review. Chem Rev 113:4980–5028. doi:10.1021/cr3003888 Vopička O, Friess K (2014) Analysis of gas sorption in glassy polymers with the GAB model: an alternative to the dual mode sorption model. J Polym Sci B 52:1490–1495. doi:10.1002/polb.23588 Vopička O, Hynek V, Zgažar M et al (2009) A new sorption model with a dynamic correction for the determination

13 of diffusion coefficients. J Membr Sci 330:51–56. doi:10.1016/j.memsci.2008.12.037 Vopička O, Friess K, Hynek Vet al (2013) Equilibrium and transient sorption of vapours and gases in the polymer of intrinsic microporosity PIM-1. J Membr Sci 434:148–160. doi:10.1016/j.memsci.2013.01.040 Vu DQ, Koros WJ, Miller SJ (2001) High pressure CO2/ CH4 separation using carbon molecular sieve hollow fiber membranes. Ind Eng Chem Res 41:367–380. doi:10.1021/ie010119w Vu DQ, Koros WJ, Miller SJ (2003) Mixed matrix membranes using carbon molecular sieves II. Modeling permeation behavior. J Membr Sci 211:311–334. doi:10.1016/S0376-7388(02)00429-5 Wijmans JG, Baker RW (1995) The solution-diffusion model: a review. J Membr Sci 107:1–21. doi:10.1016/ 0376-7388(95)00102-I Wijmans JG, Baker RW (2006) The solution-diffusion model: a unified approach to membrane permeation. In: Yampolskii Y, Pinnau I, Freeman BD (eds.) Materials science of membranes for gas and vapor separation, John Wiley & Sons, Ltd, Chichester, UK. doi: 10.1002/047002903X.ch5 Zornoza B, Tellez C, Coronas J et al (2013) Metal organic framework based mixed matrix membranes: an increasingly important field of research with a large application potential. Microporous Mesoporous Mater 166:67–78. doi:10.1016/j.micromeso.2012.03.012

G

Gas–Liquid Membrane Contactor Denis Roizard Laboratoire Réactions et Génie des Procédés, CNRS- Université de Lorraine, Nancy, France

Synonym Gas–liquid exchangers using membrane as interface; Nondispersive gas–liquid contactor

History Gas–liquid contactors are devices which are designed to promote mass transfer between a gas phase and a liquid phase, thanks to gas–liquid contact. Among the various types of existing contactors, e.g., valve trays, random or structured packing, demister, vacuum towers, etc., and membranes, one can clearly distinct two categories: firstly, contactors requiring a straight mixing between the gas and the liquid phases and, secondly, contactors where the direct physical contact between the two phases does not exist, i.e., a contactor which does not need dispersion of one phase into the other one to be efficient (Fig. 1). Up to now, membrane contactors are the only example of the second category, i.e., systems which are simultaneously able to avoid phase mixing while promoting mass transfer. Obviously the higher the membrane permeability, the # Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_265-1

better is mass transfer efficacy. Currently, most of the industrial applications are using microporous membrane contactors (Liqui-Cell ® 2014). Like with other gas/liquid contactors, the mass transfer obtained by means of contactors can be directed to carry out either separation and purification operations of a gas mixture by the selective removal of a given gas component or, conversely, absorption of gas (e.g., N2, O2, O3, etc.) in the liquid phase. The driving force is the gas partial pressure according to the Henry law: pi = Hxi with p the partial pressure, x the concentration of gas at equilibrium, and H the Henry law coefficient. Note also that the driving force of the mass transfer can also be due to a chemical reaction between the gas species and the liquid phase.

Phenomenon In the 1980s, Qi and Cussler (1985) achieved pioneering work devoted to the understanding of mass transfer in gas–liquid membrane contactors. Within this specific type of contactors, the membrane is primarily a physical barrier between a gas phase and a liquid phase. So to get an efficient device, a proper choice of membrane and operating conditions must be done to ensure a high level of mass transfer. For the membrane selection, it turned out that a microporous structure looked to be the best one to gather appropriate mechanical properties and high mass transfer coefficient. Most of the time,

2

Gas–Liquid Membrane Contactor

Dp ¼ 4gL : cos y=dmax

Gas–Liquid Membrane Contactor, Fig. 1 Schematic representation of a G/L membrane contactor

(1)

with gL the surface tension of water, y the contact angle, and dmax the diameter of the biggest micropores. As a guide, using pure water (gL = 72.8 mN/m at 20 C) and PP (with pore dmax = 1 mm and y = 115 ), the breakthrough pressure is 1.23 bar. However in reality, the measured breakthrough pressure is often much lower due to the presence of impurities or organic solutes in the water; any alteration of the polymer surface properties can also lower this value. On the other hand, to avoid any gas bubbling in the liquid phase, a slight overpressure is usually applied at the membrane liquid interface.

Mass Transfer Theory

Gas–Liquid Membrane Fig. 2 Microporous polypropylene

Contactor,

the liquid phase used is an aqueous phase; hence, it logically orientated the selection of membranes to hydrophobic rigid polymers, i.e., glassy or semicrystalline structures, like polypropylene (PP) (Fig. 2), polyvinyl difluoride (PVDF), or Teflon. Thus the aqueous phase cannot enter spontaneously the small pores (0.1–1.106 m) of the membrane structure, as predicted by the Young–Laplace equation given below. Nevertheless, the main concern with a microporous membrane contactor is to avoid pore wetting by the liquid phase, because pore wetting would induce a detrimental decrease of the overall mass transfer coefficient (Fig. 3a, b). In case of pore flooding, the mass transfer coefficient can drop by a factor 1,000–10,000. Hence there is a breakthrough pressure to be respected; this limiting pressure (Dp) difference between the two phases can be calculated by the Young–Laplace equation (Kim and Harriott 1987):

As one can expect, the mass transfer extent is closely related to the operating parameters, i.e., mainly gas partial pressure in each phase, temperature, gas and liquid flow rates, hydrodynamic conditions, and the specific transfer area. To achieve a given purification target, the mechanism of the gas/liquid transfer must be known and used to model the mass transfer and thus to predict adequate operating parameters. The performance of the mass transfer can be related to the variation of gas composition either between the inlet and the outlet of the gas phase or between the inlet-dissolved gas and the outletdissolved gas of the liquid phase. For a fast reaction with a steady gas phase velocity, the same basic equations, which are used to predict transferring gas in conventional columns, also apply for membrane contactors as recalled below: CGout ¼ ek:a:L=ug CGin

(2)

where CGout and CGin are respectively the outlet and inlet gas concentrations, k is the overall mass transfer coefficient, L is length of the contactor, a is the contact area between the two phases, and ug the gas phase velocity.

Gas–Liquid Membrane Contactor

a

3

b

Gas phase Æ Æ Liquid phase

Cgas

Cgas-Mb

Gas phase Æ Æ Liquid phase

CMb-Liquid

Cliquid 1/kov = 1/kgas + 1/kmembrane + H/kliquid Gas–Liquid Membrane Contactor, Fig. 3 (a) In a nonwetted porous membrane, the membrane transport occurs in gas-filled pore. (b) In a wetted porous membrane, the membrane transport in liquid-filled pore is much slower

Noting one can define the gas transfer efficiency Z = (CGin–CGout)/CGin, thus one can write: CGout ¼ 1 Z ¼ ek:a:L=ug CGin

(3)

It is worth noting that modeling the mass transfer properties of a membrane contactor is even easier than with conventional contactors because the gas/liquid contact area is well known and remains constant whatever the gas or liquid flow rates. As shown in Fig. 3, the overall mass transfer resistance is due to the successive resistances of the gas phase, of the membrane, and of the liquid phase. Clearly the analogy can be made with electrical resistances, and the reciprocal of the overall mass transfer coefficient (kov) can be written as the sum of the individual mass transfer coefficients: 1=kov ¼ 1=kgas þ 1=kmembrane þ H=kliquid

of drawbacks due to phase dispersion such as solvent loss, foaming, unknown specified area, and column flooding. It is worth also to underline that as polymer is the core of membrane contactors, one can get a strong reduction in weight of the equipment compared to conventional ones. At last one can note as well that a membrane contactor can be installed horizontally or vertically without problems and that its efficiency is not dependent of roll and pitch marine. Note that liquid–liquid extraction can also be achieved using membrane contactors, hence avoiding any hazard of emulsion formation. All these advantages are linked to the mixing nondispersive feature of a membrane contactor. On the other side, drawbacks can come from the facts that the pores of a membrane do not have the same size, that pore fouling can occur and block the transfer, and that the hydrophobic properties of the surface can be altered inducing a dramatic decrease of the breakthrough pressure.

(4)

H being the Henry law constant between the gas and liquid phases.

Pros and Cons Compared to conventional gas/liquid contactors (tower, packed columns, mixer settler), membrane contactors are known to avoid a number

Intensification Potential of Mass Transfer As with any membrane modules, the membrane contactors can be prepared with different geometries, going from plate and frame modules, spiral modules, or hollow fiber modules. The last type has received the most attention because it allows the creation of very large interfacial areas, up to

4

10,000 m2/m3 that is up to 20-fold the interfacial area of a structured packed column (Gabelman and Hwang 1999; http://docnum.univ-lorraine.fr/ public/INPL/2011_NGUYEN_P_T.pdf6). Thus the factor k.a., which is one of the key parameters predicting the mass transfer efficiency (Eq. 3), indicates clearly that for a constant value of k, a strong intensification of the transfer can be reached with membrane contactors. This high value of area will be obtained with fibers having very low inner diameter, typically in the 50–100 mm range. Hence, a limit of the transfer with membrane contactor can now be foreseen: the increase of the specific area shall correspond to an increase of the pressure drop in the fibers of smaller diameter. This drawback will be amplified by using fluid of high viscosity. Thus the potential of intensification is also strongly linked to the hydrodynamic conditions prevailing in the contactor. This shall depend merely on the nature of the fluid circulating in the lumen of the fibers.

Applications As example of separations, acid impurities of flue gases like SOx, NOx, or even CO2 can be removed by contacting the gas feed flow with a liquid properly chosen to trap the acidic species. The gas removal can be due to a physical chemical dissolution of the acid gas into the liquid or to a chemical reaction with the liquid; in this case, the liquid must be endowed of basic properties or contains a solute which is itself a base. When the principle of gas removal is linked to physical chemical affinities, it is wise to operate under pressure to promote higher solubility of the gas into the liquid. On the other case, if a chemical reaction is involved between the gas and the liquid, pressure is not a key parameter.

Gas–Liquid Membrane Contactor

As example of gas dissolution, one can cite nitrogenation in the beverage industry or blood oxygenation. It is worth to note that blood oxygenation has been one of the very first examples of using membrane contactors in 1975 (Esato and Eiseman 1975); currently the total annual market is above €500 million. Some examples of use of membrane contactors are listed here: – Liquid degassing: O2, CO2, and N2 removal from liquids, used for carbonation (food and beverage industry), nitrogenation (microelectronics), deoxygenation, etc. – Bubble-free gas/liquid mass transfer primarily for ozonation of semiconductor cleaning water – Dehydration – Blood oxygenator (health sector)

Cross-References ▶ Gas Stripping ▶ Henry Law

References Liqui-Cell ® web site (visited Apr 2014). http://www. liquicel.com/ Esato K, Eiseman B (1975) Experimental evaluation of Gore-Tex, membrane oxygenator. J Thorac Cardiovascular Surg 69(5):690–697 Gabelman A, Hwang S-T (1999) Hollow fiber membrane contactors. JMS 159:61–106 Kim B, Harriott P (1987) Critical entry pressure for liquids in hydrophobic membranes. J Colloid Interface Sci 115(1):1–8 Qi Z, Cussler EL (1985) Microporous hollow fibers for gas absorption. I, II. Mass transfer in the liquid. J Membr Sci 23(3):321–332, and ibid 23(3):333–345

G

Glass Transition Temperature (Tg) Johannes Carolus Jansen Institute on Membrane Technology, ITM-CNR, Rende (CS), Italy

General Introduction At the macroscopic scale, the glass transition temperature, Tg, represents the temperature above which a material changes from a stiff glass into a viscous fluid or a rubbery material. Besides polymers, which are the most common materials with a glass transition temperature, also various amorphous solids, organic liquids, alloys, or inorganic glasses may exhibit a glass transition. At the molecular scale, the Tg of a polymer is the temperature above which large segmental motions of the polymer chains become possible within the time scale of the experiment. The glass transition temperature depends on the molecular architecture. Substituents that restrict the backbone rotation of a very simple polymer, such as polyethylene, will increase its Tg, and the presence of polar groups will have an even stronger effect (Table 1). Besides the substituents on a flexible polymer backbone, the chemical structure of the backbone itself obviously has a major impact on the glass transition temperature. The presence of sterically hindered groups, conjugated bonds, fused rings, etc., increases the glass transition temperature significantly. In the extreme case, # Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_267-1

the mobility is so much restricted that a Tg is no longer observed. For instance, polymers of intrinsic microporosity (PIMs), a novel class of polymers with growing interest for their high gas permeability (McKeown and Budd 2010), consist of ladder structures with such a high rigidity that large-scale motions are impossible and they do not exhibit a glass transition below their degradation temperature. The glass transition temperature of polymer blends depends on their miscibility. In the rare case of miscibility at the molecular level, the glass transition of the blend has a value between those of the neat polymers. Different equations describe the behavior of polymer blends more or less satisfactorily. In the Fox equation, the Tg of the blend depends exclusively on its composition: 1 w1 w2 ¼ þ T g T g ,1 T g ,2

(1)

where w1 and w2 are the weight fractions and Tg,1 and Tg,2 are the glass transition temperatures of pure polymer 1 and polymer 2, respectively. The Gordon-Taylor equation is able to describe slightly asymmetric dependencies of the Tg on the blend composition, by means of an adjustable parameter, KGT:

2

Glass Transition Temperature (Tg)

Glass Transition Temperature (Tg), Table 1 Glass transition temperature of polymers -[-CH2CH(R)-]n- with the same backbone and different substituents R (Brandrup et al. 1999) Polymer Polyethylene Polypropylene Poly(vinyl fluoride) Poly(vinyl chloride) Poly(vinyl alcohol) Polystyrene Poly(vinyl acetate)

R H CH3 F Cl OH C6H5 CH3COO

Tg 155 258–270 314 354 358 373 305

Glass Transition Temperature Analysis

Glass Transition Temperature (Tg), Fig. 1 Schematic representation of a DSC trace (bottom) and of the specific volume (top) of an amorphous polymer

Tg ¼

w1 T g,1 þ K GT w2 T g,2 w1 þ K GT w2

(2)

The Kwei equation introduces a binary parameter, q, which represents the interaction between the two polymers, and can be used to describe even more asymmetric Tg versus composition profiles (ElMiloudi et al. 2009): Tg ¼

w1 T g,1 þ K Kwei w2 T g,2 þ qw1 w2 w1 þ K Kwei w2

Besides a change in the mechanical properties, the Tg is also accompanied by a fairly abrupt change in the specific heat of the sample (Fig. 1). Various other properties undergo more or less pronounced changes at the Tg, such as the density, the specific heat, the elasticity coefficient or Young’s modulus, the rate of diffusion of gases or liquids through the polymer, the thermal expansion coefficient, etc. There is a clear correlation between the glass transition temperature and the transport properties in dense membranes (Matteucci et al. 2006). The gas and vapor permeability is much higher in the rubbery state than in the glassy state, and the selectivity is usually lower. Therefore, the value of the Tg is one of the main parameters influencing the membrane performance. For most other membrane applications, also those using porous membranes, the glass transition temperature is of large interest too. Since porous membranes may collapse upon softening, it is of fundamental importance that they are operated at a temperature sufficiently far below the Tg.

References (3) Andrews RJ, Grulke EA (1999) Glass Transition Temperatures of Polymers. In: Brandrup J, Immergut EH, Grulke EA (eds) Polymer handbook, 4th edn. Wiley, Hoboken ElMiloudi K, Djadoun S, Sbirrazzuoli N, Geribaldi S (2009) Miscibility and phase behaviour of binary and ternary homoblends of poly(styrene-Co-Acrylic acid), poly(styrene-Co-N, N-Dimethylacrylamide) and Poly

Glass Transition Temperature (Tg) (styrene-co-4-Vinylpyridine). Thermochim Acta 483(1–2):49–54 Matteucci S, Yampolskii Y, Freeman BD, Pinnau I (2006), Transport of gases and vapors in glassy and rubbery polymers, In: Yampolskii Y, Pinnau I, Freeman BD (eds.) Materials science of membranes for gas and

3 vapor separation, John Wiley & Sons, Ltd, Chichester, UK. doi: 10.1002/047002903X.ch1 McKeown NB, Budd PM (2010) Exploitation of intrinsic microporosity in polymer-based materials. Macromolecules 43(12):5163–5176

G

Glass Transition Temperature Depression Johannes Carolus Jansen Institute on Membrane Technology ITM-CNR, Consiglio Nazionale delle Ricerche, Rende (CS), Italy

Glass transition temperature depression is the phenomenon which describes the reduction of the glass transition temperature by external factors, usually by the presence of solvent molecules or of other additives in the polymer matrix. Such additives have the capacity to enhance the mobility of the polymer chains, thus enabling long

# Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_268-1

range motions at lower temperature than in the neat polymer. Although this often goes hand in hand with the phenomenon of plasticization, a reduction of the elastic modulus at room temperature, the two concepts should not be confused. Plasticization strictly refers to the mechanical properties of the polymer and the molecules, which reduce the elastic modulus of a polymer by an increase of the chain mobility generally also reduce their glass transition temperature. The opposite is not necessarily the case and it may happen that an additive, which lowers the Tg, at low concentration increases the elastic or Young’s modulus at room temperature. This is then referred to as anti-plasticization.

G

Glassy Membranes John Jansen Institute on Membrane Technology ITM-CNR, University of Calabria, Rande(CS), Italy

Glassy membranes are membranes consisting of amorphous polymers which are in their glassy state at room temperature or under the normal operation conditions, i.e., their glass transition temperature is above room temperature or the operating temperature. The high stiffness of the glassy polymer provides sufficient mechanical strength to the membranes to exist also as porous

# Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_269-1

or dense integrally skinned flat films or hollow fibers. The polymer stiffness is an essential aspect in the membrane formation process by non-solvent induced phase inversion, where at a certain point of the process solidification of the polymer is required to consolidate the morphology of the membrane. In glassy polymers, this is typically by vitrification of the polymer when diffusion of the solvent from the polymer rich phase into the coagulation bath leads to a gradual increase of the glass transition temperature until this exceeds the coagulation bath temperature.

G

Glassy Polymer Johannes Carolus Jansen Institute on Membrane Technology ITM-CNR, Consiglio Nazionale delle Ricerche, Rende (CS), Italy

Strictly, any amorphous polymer is glassy at temperatures below its Tg, but in practice a glassy polymer refers to those polymers which are in their glassy state at room temperature, thus to amorphous polymers with a glass transition temperature above room temperature, in contrast to rubbery polymers, which have a Tg below room temperature, and (semi-)crystalline polymers with a melting point above room temperature. Under

# Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_270-1

particular circumstances also semi-crystalline polymers can become glassy after quenching, if the crystallization kinetics and the crystal nucleation rate are sufficiently slow to prevent crystallization upon rapid cooling. A typical example of such polymer is poly(ethylene terephthalate), PET. Glassy polymers are characterized by their high stiffness and amorphous non-crystalline structure. These properties make them size selective in their dense form and also mechanically sufficiently resistant to exist as porous flat film or hollow fibre membranes. Examples of the most commonly used polymers in commercial membranes are cellulose acetate, polysulfone, poly (ether sulfone), polyimide, and polycarbonate.

G

Gold Recovery by Supported Liquid Membranes Argurio Pietro Department of Environmental and Chemical Engineering, University of Calabria, Arcavacata di Rende (CS), Italy

Gold is the chemical element which symbol is Au (from Latin aurum) and the atomic number is 79. The oxidation states in its compounds range from 1 to +5, and Au(III) is the most common. Typical Au(I) complex is Au(CN)2 (in cyanide media) which is the soluble form of gold encountered in mining, while in chloride media Au(III) complexes (Au2Cl6) are the typical ones. Gold production by means of its extraction from mining can represent an important contribution to environmental pollution. Metal ores, which generally contain less than 1 ppm of gold, are ground and mixed with sodium or potassium cyanide for gold chemical extraction. Precious metal and heavy metal impurities such as cadmium, lead, zinc, copper, nickel, and arsenic are usually present in an anionic form (i.e., cyanide salts) after their extraction from metal ores. These salts are toxic to the liver and kidneys because of both cyanide and metal content. Thus, it would be desirable to be able to selectively remove these complexes for the recovery of precious metals. # Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_272-6

Gold, over its use in jewelry, also has a wide use in various industries, thanks to its physical and chemical properties. Thus, the recovery of this metal from the different wastewater generated by these industries is also of growing interest (Alguacil 2004). Gold can be recovered from different aqueous solutions by various physicochemical separation techniques as precipitation, ion exchange, carbon adsorption, cementation, solvent extraction, etc. In the case of convectional solvent extraction, metal ion-containing solution is placed in contact with a large amount of an appropriate organic phase (Kargari et al. 2004). This extraction step is followed by a stripping one. The main drawback of solvent extraction is the large amount of solvent required when dilute solutions were processed, making this process not very cheap and safe, since the used solvents are often chlorinated and sometimes carcinogenic. Liquid membrane (LM)-based processes have become an attractive alternative to conventional techniques for selective separation and concentration of both organic and inorganic compounds from dilute aqueous solutions since they combine extraction and stripping into a single process, thus reducing the solvent inventory requirement and then cost significantly (Molinari and Argurio 2011). They also allow the use of expensive and highly selective extractants, which otherwise would be uneconomic in solvent extractions. LM systems include nonsupported liquid membranes (bulk liquid membrane (BLM) and

2

Gold Recovery by Supported Liquid Membranes

emulsion liquid membrane (ELM)) and supported liquid membrane (SLM). SLMs consist of an organic LM phase impregnated in a thin hydrophobic microfiltration membrane. This LM phase generally contains an extractant (carrier) which binds very selectively the target component in the donor phase (feed), transporting it into the acceptor phase (strip), resulting in the so-called facilitated transport (Molinari et al. 2009a, b). Referring to gold transport, SLM systems have been tested in the separation of this precious metal mainly from cyanide (Au(I)) or chloride (Au(III)) media. Highly acidic conditions are required for the extraction and transport of Au (III) because of its easily reducible nature. Various carriers for Au transport across a SLM were reported in literature. Among them, the commercially available extractant Cyanex ® 921 has been applied in the carrier-facilitated transport of both gold(I) and gold(III) recovery from cyanide and chloride media, respectively (Alguacil et al. 2005). Gold(I) is transported from alkaline pH values (6–11). Gold (I) extraction takes place in that pH range by the following equilibrium reaction: Mþ þ AuðCNÞ2aq þ 3 Lorg , Mþ AuðCNÞ2 3Lorg

where the subscripts aq and org denote the species contained in the aqueous and organic phase, respectively, and L represents the extractant. In the case of gold(III), the extraction is governed by the following pH-dependent equilibrium reaction: Hþ aq þ AuCl4 þ nLorg , Hþ AuCl4 nLorg where n = 1, 2.

References Alguacil FJ (2004) Carrier-mediated gold transport in the system Cyanex 921–HCl–Au(III). Hydrometallurgy 71:363–369 Alguacil FJ, Alonso M, Sastre AM (2005) Facilitated supported liquid membrane transport of gold (I) and gold (III) using Cyanex ® 921. J Membr Sci 252:237–244 Kargari A, Kaghazchi T, Soleimani M (2004) Role of emulsifier in the extraction of gold (III) ions from aqueous solutions using the emulsion liquid membrane technique. Desalination 162:237–247 Molinari R, Argurio P (2011) Recent progress in supported liquid membrane technology: stabilization and feasible applications. Membr Water Treat 2(4):207–223 Molinari R, Argurio P, Poerio T (2009a) Studies of various solid membrane supports to prepare stable sandwich liquid membranes and testing copper(II) removal from aqueous media. Sep Purif Technol 70:166–172 Molinari R, Argurio P, Poerio T (2009b) Flux enhancement of stagnant sandwich compared to supported liquid membrane systems in the removal of Gemfibrozil from waters. J Membr Sci 340:26–34

G

Graft Polymerization Tauqir A. Sherazi Department of Chemistry, COMSATS Institute of Information Technology, Abbottabad, Pakistan

Graft polymerization is a process in which monomers are covalently bonded and polymerized as side chains onto the main polymer chain (the backbone).

Grafting is an attractive approach to impart a variety of functional groups to a polymer. Graft polymers are also known as graft copolymer since it contains at least two different kinds of monomer units such as the grafted side chains that are structurally distinct from the main chain. The monomer to be grafted may be of one or more than one type; thus, the graft chains in grafted copolymer may be homo-polymers or copolymers as illustrated in Fig. 1 (A, B, and C are representing different types of monomers). Grafting can be accomplished by either “grafting to” or “grafting from” approaches. In

# Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_274-2

“grafting to” approaches, functionalized monomers react with the backbone polymer to form the grafted one. On the other hand, “grafting from” is achieved by treating a substrate with some method to generate immobilized initiators followed by polymerization. High grafting density polymer also can be accomplished using this technique (Bhattacharya et al. 2009). Graft copolymerization can be initiated by various methods including chemical treatment, photochemical treatment, ionizing radiation (such as gamma radiation, electron beam radiation, etc.), photo-irradiation, plasma-induced techniques and enzymatic grafting, etc. Grafted polymers can be very useful as they can be tailored to the requirements of particular applications by appropriate selection of backbone and monomers to be grafted. Grafted polymers have wide range of application such as in the field of biomedical, textiles, automobiles, cable technology, separation and purification, electrolyte membranes, coatings, adhesives, laminates, commodity plastics, etc.

2

Graft Polymerization graft homo-polymer

graft copolymer

B

B

B B

C C

C

C C C C base polmer C A B C A A A A A A C A A A A A A A A A A A A A A A A A A A A A A A B A C C B C B B B C B B C C B B B B C graft homo-polymer B B B graft copolymer B B

B

B

Graft Polymerization, Fig. 1 Structural representation of graft copolymer

References Bhattacharya A, Rawlins JW, Ray P (2009) Polymer grafting and crosslinking. John Wiley & Sons, Inc., Hoboken, New Jersey

H

Hexavalent Chromium Separation by 0.05 mg/L. Usually Cr(VI) concentrations in Supported Liquid Membranes industrial wastewaters range from 0.5 to Argurio Pietro Department of Environmental and Chemical Engineering, University of Calabria, Arcavacata di Rende (CS), Italy

The removal and/or recovery of heavy metals from industrial wastewater is a major topic of research. Chromium is a unique, toxic heavy metal released in aqueous environment in both +3 and +6 oxidation states. Hexavalent chromium (Cr(VI)) receives particular attention because of its muta-, terato- and carcinogenic properties. Cr (VI) exists as anionic species, such as HCrO4, Cr2O72, and CrO42, which are highly mobile on subsurface environment (Kumbasar 2008). These anionic species are bioaccumulative, and their oxidizing potentials make them highly toxic to biological systems. The major industries that contribute to water pollution by chromium are mining, leather tanning, textile dyeing, electroplating, metal finishing, and corrosion inhibition (Rajasimman and Karthic 2010). The World Health Organization recommends the toxic limits of Cr(VI) in wastewaters at the level of 0.005 mg/L. Many countries have regulations of the maximum permissible concentration of Cr(VI) in natural or drinking water which typical tolerance limit for discharge into inland surface waters is 0.1 mg/L and in potable water is # Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_276-9

270,000 mg/L. Then chromium-bearing wastewaters must be discharged into aquatic environments or onto land after appropriate treatments to drastically reduce Cr(VI) content. The methodologies for Cr(VI) recovery, from industrial wastewater, range from ion exchange to solvent extraction, non-dispersive solvent extraction, precipitation, and adsorption. Solvent extraction has been widely used for the removal and/or recovery of chromium in hydrometallurgy since this technique allows the Cr ions recovery, but it involves high capital and operating costs due to large inventory of solvent, especially in the case of dilute solutions. Conventionally the treatment chromium-bearing wastewaters consists of the reduction of Cr(VI) to Cr(III) with an adequate chemical-reducing agent. Cr(III) is then easily precipitated by the addition of an alkali compound (generally calcium hydroxide) to the liquid effluent. The main drawback of this method is the production of a large amount of sludge-containing chromium often in high concentration, which disposal/treatment is a very costly affair and it is not eco-friendly. Supported liquid membrane (SLM) process offers a technology with a low solvent/extractant consumption since it involves in a single stage the extraction and stripping processes, which are generally performed in two separate steps in conventional solvent extraction. SLM consists of a LM phase, impregnated in the pores of a thin

2

Hexavalent Chromium Separation by Supported Liquid Membranes Feed

HCrO4−

Liquid membrane

Strip HCrO4−

H2CrO4 TOPO

H+

H+

TOPO HCrO4− H+

HCrO4− H+

Hexavalent Chromium Separation by Supported Liquid Membranes, Fig. 1

hydrophobic microfiltration membrane, kept there by capillary forces. It combines the typical advantages of liquid membrane with the mechanical resistance of solid membranes. The transport across a SLM is mediated by a mobile extractant (carrier) which binds very selectively the target component in the donor phase (feed), transporting it into the acceptor phase (strip), resulting in the so-called facilitated transport (Molinari et al. 2009a, b). The selection of an appropriate carrier provides higher selectivity and enrichment factor as compared to the other separation methodologies. Despite of their advantages with respect to the traditional separation techniques, SLM is not widely applied in treatment of chromium-bearing wastewaters. Cr(VI) compounds could be removed from dilute aqueous solutions by using trioctylphosphine oxide (TOPO), Alamine 336, tri-n-octylamine (TOA), and tributyl phosphate (TBP) as ionic carrier. The extraction of HCrO4 with TOPO from acidic solutions could be expressed by the following equation (Kumbasar 2009; Hasan et al. 2009): HCrO4 ðaqÞ þ Hþ ðaqÞ þ TOPOðorgÞ , H2 CrO4 TOPOðorgÞ The so-formed complex diffuses through the membrane toward the stripping basic solution where the de-complexation reaction takes place

and HCrO4 and H+ ions are released. The so-regenerated carrier molecule diffuses back to the feed and the transport cycle begins again (Fig. 1). This transport mechanism is the so-called facilitated coupled co-transport, typical when a basic carrier like amines or phosphates is used to transport negatively charged species (in this case HCrO4) and usually H+ as counterion across the membrane in the same direction (Fig. 1). Cr(VI) complexes can be efficiently removed from acidic chloride aqueous solutions by facilitated transport with TOA into a basic (NaOH 0.1 M) acceptor phase (Kozlowski and Walkowiak 2002). In agreement with the transport mechanism that is similar to that one previously reported for TOPO, the permeability coefficient and then initial flux values decrease linearly by increasing the feed pH. Cr (VI) concentration can be successfully reduced in the feed phase from 1.0 to 0.0028 mg/L, thus respecting the World Health Organization’s recommendations.

References Hasan MA, Selim YT, Mohamed KM (2009) Removal of chromium from aqueous waste solution using liquid emulsion membrane. J Hazard Mater 168:1537–1541 Kozlowski CA, Walkowiak W (2002) Removal of chromium (VI) from aqueous solutions by polymer inclusion membranes. Water Res 36:4870–4876 Kumbasar RA (2008) Studies on extraction of chromium (VI) from acidic solutions containing various metal ions by emulsion liquid membrane using Alamine 336 as extractant. J Membr Sci 325:460–466 Kumbasar RA (2009) Extraction of chromium (VI) from multicomponent acidic solutions by emulsion liquid membranes using TOPO as extractant. J Hazard Mater 167:1141–1147 Molinari R, Argurio P, Poerio T (2009a) Studies of various solid membrane supports to prepare stable sandwich liquid membranes and testing copper(II) removal from aqueous media. Sep Purif Technol 70:166–172 Molinari R, Argurio P, Poerio T (2009b) Flux enhancement of stagnant sandwich compared to supported liquid membrane systems in the removal of Gemfibrozil from waters. J Membr Sci 340:26–34 Rajasimman M, Karthic P (2010) Application of response surface methodology for the extraction of chromium (VI) by emulsion liquid membrane. J Taiwan Inst Chem Eng 41:105–110

H

Heterogeneous Ion-Exchange Membranes Mitsuru Higa Graduate School of Science and Engineering, Yamaguchi University, Ube, Yamaguchi, Japan

Heterogeneous ion-exchange membrane consists of finely powdered ion exchanger and a binder polymer, and, in many cases, the membrane is reinforced by woven cloth or net to improve its mechanical properties. In general, heterogeneous ion-exchange membranes (IEMs) are prepared by the following method (Sata 2004): finely powdered organic and/or inorganic ion exchanger is homogeneously mixed and heated with a thermoplastic polymer such as poly(vinyl chloride), polyethylene, polypropylene, or other engineering plastics, and then the mixture is formed into the membrane by pressing and/or heating. Heterogeneous IEMs have slightly lower electrochemical properties: lower permselectivity for

# Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_278-1

counterions and/or higher membrane resistance than homogeneous ion-exchange membrane. However, heterogeneous IEMs are easily prepared and have high mechanical strength. Moreover, the IEMs of various kinds of shapes can be easily prepared by pressing and/or heating as shown in Fig. 1: (a) commercial flat-sheet IEMs such as Ralex CMH and AMH (Mega a.s., Czech Republic), (b) profiled IEM for reverse electrolyte applications (Vermaas et al. 2011), (c) hollow fiber-type IEMs (Kiyono et al. 2004), and (d) commercial tubular-type IEMs (EDCORE, Astom. Co., Ltd., Japan). EDCORE is a membrane electrode apparatus, with smoothsurfaced seamless and tubular IEMs (Fig. 2), and is used in industries such as electro-deposition painting of automobile, building materials, house appliance, and other applications due to their high mechanical strength and ease of handling.

2

Heterogeneous Ion-Exchange Membranes

Heterogeneous Ion-Exchange Membranes, Fig. 1 Schematic diagram of heterogeneous IEMs with various kinds of shapes

References http://www.astom-corp.jp/en/en-main2-edcore.html Kiyono R, Koops GH, Wessling M, Strathmann H (2004) Mixed matrix microporous hollow fibers with ion-exchange functionality, J Membr Sci 231: 109–115 Sata T (2004) Ion exchangemembrane. The Royal Society of Chemistry, Cambridge Vermaas DA, Saakes M, Nijmeijer K (2011) Power generation using profiled membranes in reverse electrodialysis, J Membr Sci 385-386. 234-242 Heterogeneous Ion-Exchange Membranes, Fig. 2 Tubular-type heterogeneous IEM, EDCORE

H

High Free Volume Polymer Peter M. Budd Organic Materials Innovation Centre, School of Chemistry, University of Manchester, Manchester, UK

Free volume is the unoccupied space between molecules. The concept of free volume is used to explain molecular motion in liquids and solids. In a liquid or rubber, free volume increases with increasing temperature. A flexible polymer will flow or behave as a rubber at temperatures at which there is sufficient free volume for largescale movements of polymer segments but will behave as a glass when the temperature is reduced to the point where there is not enough free volume for such movements. Different assumptions may be made about what constitutes “occupied” and what constitutes “free” volume in a material. Thus, in different contexts, different values may be quoted for the amount of free volume in a polymer. In membrane science, it is common to define fractional free volume as fv ¼

V 1:3V w V

where V is the specific volume (reciprocal of density) and Vw is the specific van der Waals volume, which for many polymers may be # Springer-Verlag Berlin Heidelberg 2013 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_279-5

estimated using group contribution methods. The factor 1.3 takes account of the fact that molecules cannot completely fill space even in a perfectly ordered crystal at absolute zero and is an average value for crystalline materials. By this definition, most glassy polymers have fv < 0.2. Some glassy polymers, however, have much higher fractional free volumes and consequently exhibit high permeabilities to gases and vapors (Budd and McKeown 2010; Yampolskii 2012; Pinnau and Toy 1996; Starannikova et al. 2008; Thomas et al. 2009). The structures of some polymers with exceptionally high free volume are shown in Fig. 1. A common feature of high free volume polymers is that they have relatively inflexible, twisted backbones, which cannot change conformation in order to fill space efficiently. In substituted polyacetylenes such as PTMSP, the bulky side group inhibits rotation about single bonds in the backbone and forces the backbone into a twisted shape. In perfluoropolymers such as Teflon AF2400, neighboring dioxolane rings cannot easily rotate past each other. In substituted polynorbornenes prepared by addition polymerization, such as PTMSN, ring structures and bulky side groups restrict rotation about backbone bonds. The ultimate extension of this idea is found in polymers of intrinsic microporosity, such as PIM-1, which have no single bonds in the backbone about which rotation can occur but which incorporate sites of contortion (the spiro

2

High Free Volume Polymer

High Free Volume Polymer, Fig. 1 Molecular structures and fractional free volumes of poly(1-trimethylsilyl-1propyne) (PTMSP), a copolymer of 2,2-bistrifluoromethyl4,5-difluoro-1,3-dioxole and tetrafluoroethylene (Teflon AF2400), addition-type poly(trimethylsilyl norbornene)

(PTMSN), and a polymer of intrinsic microporosity prepared from 5,50 ,6,60 -tetrahydroxy-3,3,30 ,30 -tetramethyl1, 10 -spirobisindane and 1,4-dicyanotetrafluorobenzene (PIM-1)

centers in PIM-1) to force the backbone to twist and turn. The polymers shown in Fig. 1 are soluble and can readily be processed from solution to form membranes. However, there are other high free volume polymers which cannot be prepared in soluble form. Sometimes, it is possible to form a membrane from a soluble precursor and subsequently convert it to the desired structure through chemical or thermal treatment. For example, polybenzoxazole structures can be prepared by thermal rearrangement from aromatic hydroxylcontaining polyimides (Park et al. 2007). The permeation of gases and vapors through a polymer depends not only on the amount of free

volume but also on the size, distribution, and connectivity of free volume elements. Computer simulation is useful for visualizing the free volume distribution in amorphous polymers (Hofmann et al. 2002). A number of experimental techniques, notably positron annihilation lifetime spectroscopy (PALS), have been employed to obtain information about the size and distribution of free volume elements. In a high free volume polymer, there may be sufficient connectivity between free volume elements for the polymer to behave like a molecular sieve or microporous material (pore size hydrophobic acids > transphilic acids > hydrophilic charged. However, it may be that the hydrophilic neutral fraction exhibited the greatest fouling potential because of its size (>30 kD) relative to the membrane pore size. The components of NOM can also be fractionated into the following four fractions by pyrolysis-GC/MS: polysaccharides, proteins, polyhydroxy-aromatics, and amino sugars. It has been found that the fouling potential could be ranked in the order of polyhydroxy-aromatics > proteins > polysaccharides and amino sugars. The polyhydroxy aromatics were thought to be the main foulants for negatively charged NF membrane surfaces, and are probably hydrophobic acids with phenolic groups, exhibiting no negative charge at a neutral pH. However, this could change in the presence of calcium, and similar divalent cations, that appear to exacerbate NF fouling by NOM due to binding between the negative membranes and negative components of the NOM. Controlling NOM fouling can significantly reduce the cost of membrane water treatment, extend membrane life, and reduce energy demand. Techniques developed to minimize NOM fouling include hybrid membrane processes, pretreatment to reduce the NOM in the raw water, optimization of hydrodynamic parameters, careful membrane selection, and cleaning of the membrane system. Hybrid membrane processes involve the combination of coagulation/flocculation, or sorbents, such as powdered activated carbon (PAC), heated iron oxide particles (HIOPs), or an ion-exchange resin (such as MIEX ®) with membrane

THMP Removal (%) 15–20 ~60 ~40 ~80 90–99

processes. MIEX ® is an anion-exchange resin capable of removing relatively low molecular weight negative organics from NOM. It has been shown that using MIEX ® to remove organics leads to very low fouling of a UF membrane which is hydrophilic and positively charged. The performance was particularly enhanced when calcium was removed. Pretreatment strategies include biologically activated carbon, where a bed of granular carbon with developed biofilms treats the feed water and reduces the biodegradable components of NOM. Hydrophilic membranes have been found to be less prone to fouling than hydrophobic membranes when treating natural water containing NOM. Unfortunately, most commercially available UF and MF membranes are relatively hydrophobic materials with low surface energies such as polypropylene, polysulfone, polyethersulfone, and polyvinylidene fluoride. Some membranes (such as PVDF) can be rendered more hydrophilic by blending or surface treatment. A recent development involves the use of ceramic UF membranes with ozone pretreatment to oxidize NOM components. This method appears to limit fouling, even at relatively high fluxes.

References Fane AG, Wei Xi, Wang R (2006) Membrane filtration processes and fouling, chapter 10. In: Newcombe G, Dixon D (eds) Interface science in drinking water treatment: fundamentals and applications. Academic Mallevialle J, Odendaal PE, Wiesner MR (eds) (1996) Water treatment membrane processes. McGraw-Hill, New York

H

Hybrid Membranes Mario Aparicio Instituto de Ceramica y Vidrio (CSIC), Madrid, Spain

The term “hybrid membrane” more commonly refers to a membrane formed by at least two components of different chemical nature (considering chemical-bond modes) from the groups of metals, organic materials and their polymers, and inorganic materials. Hybrid membranes can be classified into two main types, depending on the nature of the interaction between components: (1) systems where there are no covalent or iono-covalent bonds between components, only Van der Waals, hydrogen bonding, or electrostatic forces, and (2) systems where at least parts of the components are linked through strong covalent or ionocovalent bonds (Sanchez and Go´mez-Romero 2004). The first type can also be named as

# Springer-Verlag Berlin Heidelberg 2013 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_292-3

composites, or nanocomposites, where at least one of the components’ domains has a dimension ranging from a few angstroms to several nanometers. The main objective in the synthesis of hybrid membranes is the performance improvement of the material for different applications. It is obvious that the properties of these membranes are not only the sum of the individual contributions of both components, and interfaces can play a significant role. The high number of parameters involves in the design and preparation of hybrid membranes: the number of components, composition, components ratio, size, shape, and kind of interaction between components results in an almost infinite number of combinations.

References Sanchez C, Go´mez-Romero P (2004) Functional hybrid materials. Wiley VCH, Weinheim

H

Hybrid Organic-Inorganic Nanostructured Membranes Mario Aparicio Instituto de Ceramica y Vidrio (CSIC), Madrid, Spain

A hybrid organic-inorganic nanostructured membrane is a specific kind of hybrid membrane formed by at least two components from organic materials and their polymers and inorganic materials. The components of this kind of hybrid membranes have dimensions up to several nanometers, and components are linked through covalent or iono-covalent bonds. Improved and also new properties are expected in this specific kind of hybrid membranes because of the combination of two very different materials and their chemical interactions. For example, it is possible to obtain membranes with a high flexibility and processability, such as a polymer, but with the improved thermal and chemical stability, and mechanical strength of inorganic materials (Sanchez et al. 2005). The possible applications of these hybrid membranes are increasing continuously. For example, an antimicrobial drug (substituted 1,3,4-oxadiazole) with functionalized silica was successfully incorporated into an organic phase by sol-gel to achieve a highly stable and antibiofouling membrane for water treatment (Singh et al. 2012). Hybrid organic-inorganic # Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_293-3

nanostructured membranes have also found applications in the medical field for advanced separation of heavy metals from blood or other physiological liquids, such as new polymericcarbon nanotube composite membranes based on polysulfone with different types of nanotubes (Nechifor et al. 2009). Another important area where these hybrid membranes may have a significant relevance is the energy sector, especially as new membranes for proton exchange membrane fuel cells (PEMFC) and solid-state Li-ion batteries. There membranes share several common characteristics, such as high ion conductivity (proton and lithium ion, respectively), low electronic conductivity, high thermal stability, and high chemical/electrochemical stability. Decreasing the membrane thickness but preserving the properties described above would improve the performance of the systems. Hybrid organic-inorganic nanostructured membranes can be designed to incorporate all these properties (Mosa and Aparicio 2012).

References Mosa J, Aparicio M (2012) Hybrid materials for high ionic conductivity. In: Aparicio M, Jitianu A, Klein LC (eds) Sol–gel processing for conventional and alternative energy, 1st edn. Springer, New York, pp 99–122 Nechifor G, Voicu SI, Nechifor AC, Garea S (2009) Nanostructured hybrid membrane polysulfone-carbon nanotubes for hemodialysis. Desalination 241:342–348

2 Sanchez C, Julian B, Belleville P, Popall M (2005) Applications of hybrid organic–inorganic nanocomposites. J Mater Chem 15:3559–3592

Hybrid Organic-Inorganic Nanostructured Membranes Singh AK, Singh P, Mishrab S, Shahi VK (2012) Antibiofouling organic–inorganic hybrid membrane for water treatment. J Mater Chem 22:1834–1844

H

Hydrocarbon Branching A. Baudot Physics and Analysis Division, IFP Energies nouvelles, Solaize, France

For more than 20 years now, regulations have imposed increasingly tight limitations on the content in gasoline of aromatic octane number boosters produced by the reforming of straightrun gasoline (alkanes). Among the available alternative technologies designed to enhance the octane number of straight-run gasoline, hydroisomerization is a catalytic technology that upgrades low-octane-number linear paraffins into higher-octane-number branched paraffins. Since the rate of conversion of linear paraffins in the isomerization units is limited by a thermodynamic equilibrium, an option for increasing the production yield of dibranched paraffins consists of separating the linear and monobranched paraffins from the isomerization unit effluent and recycling them in the input of the isomerization reactor. The more conventional solution consists of fractionating the output stream from the isomerization reactor through a continuous distillation column (a deisohexanizer or DIH) into three effluents: – A sidestream, mainly containing unconverted normal hexane and the monobranched # Springer-Verlag Berlin Heidelberg 2014 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_296-2

paraffins with six carbon atoms. This stream is recycled to the isomerization unit. – The bottom stream, containing the heaviest alkanes (paraffins) with six carbon atoms and naphthenes (cyclic paraffins) with six carbon atoms, with an RON of 82, is sent directly to the gasoline pool. – At the top of the column, a head stream rich in dibranched paraffins with six carbon atoms, isopentane, and normal pentane. This stream, which corresponds to 90 % by weight of the fresh feed, has an RON (research octane number) of 87 though it contains about 16 % by weight of normal pentane, which has a low RON of 61. With current distillation processes, it is not economically feasible to separate normal pentane from the other components in the top stream of the deisohexanizer since their respective boiling points are very close. This type of separation, however, can be achieved with molecular sieves, such as zeolites, implemented in cyclic adsorption processes such as a “simulated moving bed” (UOP Molex process) or “cycled pressurization/ depressurization” (IFP IPSORB process or ExxonMobil ISOSIEVE process). With such processes, normal paraffins are preferentially adsorbed inside the microporosity of the zeolites and therefore separated from their branched isomers. Though offering excellent separation performance, this type of technology exhibits

2

several drawbacks: high investment costs, sophisticated sequential operation (adsorptiondesorption cycle), the use of large quantities of solvents (desorbants), and lack of modularity. Since the beginning of the 1990s, much attention has been paid to overcoming the drawbacks of conventional zeolite adsorbents through the development of zeolite membranes that combine the technical advantages of membranes (modularity, continuous operation) with the high separation performances of zeolites (due to their sieving properties). Most of the published R&D work on that topic was carried out at lab scale mainly by academic laboratories, though a few companies, like ExxonMobil or NGK Insulators, have also published results on that topic. One of the most studied topics in this research field was the separation of normal short (C4–C6) paraffins from their branched isomers through MFI-type zeolite membranes (Arruebo et al. 2006; Bakker et al. 1996; Coronas et al. 1998). Indeed, MFI zeolites are crystalline aluminosilicates with a microporous structure that is composed of two intricate micropore networks: elliptical straight channels with openings of 0.51 0.55 nm and zigzag channels that are almost cylindrical with a diameter of 0.53 0.56 nm, as measured by X-ray diffraction at ambient temperature (Flanigen et al. 1978). In such a confined porous system wherein the diameter of the micropores and the kinetic diameter of the diffusing molecules are close, the higher the kinetic diameter of a permeating molecule, the higher the friction of the molecule alongside the micropore wall and therefore the lower its diffusion coefficient inside the microporosity of the MFI zeolite. Therefore,

Hydrocarbon Branching

the diffusion coefficient of normal alkanes in MFI zeolites is higher (Courthial et al. 2008) than the diffusion coefficient of their monobranched isomers. Moreover, these materials prove to be hardly permeable to dibranched paraffins. As an illustration, it was shown experimentally that a MFI zeolite membrane operated under close to industrial operating conditions (2–4 bar feed pressure, membrane temperature between 200 C and 400 C) was able to produce a permeate composed of 95 % normal pentane and 5 % isopentane from a vapor feed composed of 20 % 2,2-dimethylbutane, 55 % isopentane, and only 25 % normal pentane (Baudot and Bournay 2009).

References Arruebo M, Falconer JL, Noble RD (2006) Separation of binary C-5 and C-6 hydrocarbon mixtures through MF1 zeolite membranes. J Membr Sci 269:171–176 Bakker WJW, Kapteijn F, Poppe J, Moulijn JA (1996) Permeation characteristics of a metal-supported silicalite1 zeolite membrane. J Membr Sci 117:57–78 Baudot A, Bournay L (2009) Integration of MFI zeolite membranes in the light gasoline isomerisation process. Oil Gas Sci Technol 64:759–771 Coronas J, Noble RD, Falconer JL (1998) Separations of C-4 and C-6 isomers in ZSM-5 tubular membranes. Ind Eng Chem Res 37:166–176 Courthial L, Baudot A, Tayakout-Fayolle M, Jallut C (2008) A transient permeation-based method for composite zeolite membrane characterization. AIChE J 54:2527–2538 Flanigen EM, Bennett JM, Grose RW, Cohen JP, Patton RL, Kirchner RM, Smith JV (1978) Silicalite, a new hydrophobic crystalline silica molecular-sieve. Nature 271:512–516

H

Hydrogen from Bioethanol Jose M. Sousa Chemistry Department, School of Life & Environment Sciences, University of Tras-osMontes e Alto Douro, Vila Real Codex, Portugal LEPABE - Laboratory for Process Engineering, Environment, Biotechnology and Energy Chemical Engineering Department Faculty of Engineering, University of Porto Rua Roberto Frias, Porto, Portugal

The production of hydrogen from bioethanol has been considered an attractive way for exploring sustainable renewable energy sources, from an environmentally friendly point of view. Bioethanol consists of an aqueous solution containing 8–12 wt.% of ethanol, besides other by-products such as glycerol, acetaldehyde, diethyl ether, methanol, etc. (Ni et al. 2007; Iulianelli and Basile 2011). Hydrogen production from ethanol is essentially carried out by steam reforming, according to the main reaction described by C2 H5 OH þ 3H2 O ! 6H2 þ 2CO2 , DH ð298 KÞ ¼ 348 kJ mol1 . Ethanol steam reforming is a very attractive way to locally produce hydrogen, comparatively to other fuels such as methanol, glycerol, acetic acid, diethyl ether, etc. For example, ethanol is easily obtained by fermentation from renewable sources such as sugars and starches (e.g., # Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_297-3

sugarcane and corn – first-generation bioethanol) and from lignocelluloses (agricultural, industrial, and forest residues – second-generation bioethanol), among other sources; ethanol is easy to transport and store, it is biodegradable and shows low toxicity; ethanol is relatively easy to dehydrogenate by steam reforming, and it does not contain catalyst poison such as sulfur (Ni et al. 2007). Steam reforming of ethanol is a highly endothermic reaction, which limits its industrial application for hydrogen production. The oxidative reforming, oxidation of a fraction of ethanol to provide part of the energetic needs, is a possible way to minimize such impact. If the amount of oxygen is sufficient for balancing the reforming enthalpy needs, the process is named autothermal reforming (Song 2012). Other ways to produce hydrogen from ethanol are fermentation processes using metabolically engineered microorganisms, solar photocatalytic processes using suitable semiconductors, CO2 dry reforming, plasma reforming, partial oxidation, and aqueous phase reforming (Song 2012). Concerning the catalysts for the hydrogen production from (bio)ethanol, current state of the art includes the noble metal-based (Pt, Pd, Ru, Rh) and non-noble metal-based (Cu, Co, Zn, Ni) ones (Iulianelli and Basile 2011). The ethanol conversion and hydrogen yield and selectivity are strongly dependent on the catalyst type, its support (e.g., ZnO, MgO, Al2O3, SiO2, La2O3,

Co

x

Co

x

Cx Cx C

Cx

O

O

O

OH

scheme 6

OH

Deactivation

Co

Co

Co

O O O

H

O

H

OH

OH

H2 O

OH O

OH

3

CH CHO

scheme 3

OH

H

O

O

O

scheme 5

OH

H2

scheme 4

O

O

OH

H2O

scheme 1

O

CH4CHO

O

O Co

Co

O C O

Co

H3C

O O

O

scheme 8

scheme 9

H2

O

O

O

O

O scheme 7

OH + OH O

CO

O

O

O

Sufficient oxygen accessiblity

Support (metal oxide)

OH

CH4

CH3COCH3

Support (metal oxide)

O

O

O

CO2

Support (metal oxide)

O

Route 1

O

O O

CO2 Co

O

O O

O C O

O

Co

H

Co

Co

H-CHx

Support (metal oxide)

OH OH O O

H2

Support (metal oxide)

OH + O O O O O

H2

Support (metal oxide)

O

Support (metal oxide)

O

O

Hydrogen from Bioethanol, Fig. 1 Proposed reaction mechanism for ethanol steam reforming over supported Co catalysts (Song 2012)

Support (metal oxide)

OH

Support (metal oxide)

OH

Support (metal oxide)

O

H

Support (metal oxide)

H3CC

Deficient oxygen or Co site accessiblity

Support (metal oxide)

OH

H

scheme 2

H

2 H2O

H

e

ac

CH3CH2O

Cx

Support (metal oxide)

H

C2H4

urf

cs

idi

Ac

Co

CH3CH O

C C x

CH3CH2OH

OC

CH3CH2O HC

O

O

C

O

O

O

O

O

O

O

O

H2

O scheme 11

O

O scheme 12

O

O scheme 13

O

O scheme 10

OH + OH O

CO

O

O

CO2

OC

O

2 Hydrogen from Bioethanol

Route 2

Hydrogen from Bioethanol

3

Water

Air

Feed FBR

HT WGS

LT WGS Cooler

H2Purification (PSA)

CO PROX Cooler

H2

Cooler

Hydrogen from Bioethanol, Fig. 2 High-purity hydrogen production in a conventional multistage system

Hydrogen from Bioethanol, Fig. 3 Highpurity hydrogen production in a membrane reactor with a Pd-based membrane

CeO2), and the reaction conditions (Costa-Serra et al. 2010; Song 2012). Besides the ethanol steam reforming main reaction referred above, the process follows a complex reaction system with several possible consecutive parallel reactions, such as partial reforming to CO, water gas shift, methanation, coke formation from intermediate products, Boudouard reaction, CO reduction, methane cracking, dehydration/hydrogenation, and dehydrogenation. In addition to H2 and CO2, reformate stream may contain also CO, methane, aldehydes, ketones, ethylene, ethane, and high alcohols, among others (Vizcaíno et al. 2007; Song 2012) (Fig. 1). The main drawback of using bioethanol to produce hydrogen via steam reforming is the high cost associated to the downstream distillation and purification steps of the crude ethanol obtained from fermentation. Feeding directly the crude bioethanol to the reformer would reduce drastically the costs of the produced hydrogen. Besides the unnecessary expensive distillation process for water and other compounds elimination, the reforming of other oxygenated hydrocarbons contained in the fermentation broth could contribute to generate extra hydrogen. The main challenge for the implementation of this approach at an industrial level remains in the tolerance and stability of the catalyst to the

impurities present in the crude ethanol solution, especially high linear and branched alcohols (Le Valant et al. 2011; Song 2012). In the viewpoint of hydrogen production for supplying polymer electrolyte membrane fuel cells (PEMFCs), the reformate stream, which comprises a complex mixture of compounds, needs a separation/purification, especially due to the maximum allowed CO concentration (0.2 ppm). A conventional steam reformer system is composed by the reformer (fixed bed reactor (FBR)), two water gas shift reactors (high- and low-temperature WGS), a CO partial oxidation reactor (PROX), and pressure swing adsorption (PSA) units (Fig. 2). This complex process may be replaced by a much simpler membrane reactor (MR) holding hydrogen permselective membranes. This new reactor is able to perform both the steam reforming of bioethanol and the separation/purification of the produced hydrogen in the same device (Fig. 3). Moreover, this kind of MR makes possible the in situ removal of hydrogen from the reaction side, allowing the conversion to overcome the thermodynamic equilibrium value (which is not possible in the FBR). Furthermore, if Pd or Pd-based membranes are used, a pure hydrogen stream is collected in the permeate side, suitable for direct PEMFC supplying (Iulianelli and Basile 2011).

4

References Costa-Serra JF, Guil-Lo´pez R, Chica A (2010) Co/ZnO and Ni/ZnO catalysts for hydrogen production by bioethanol steam reforming. Influence of ZnO support morphology on the catalytic properties of Co and Ni active phases. Int J Hydrog Energy 35:6709–6716 Iulianelli A, Basile A (2011) Hydrogen production from ethanol via inorganic membrane reactors technology: a review. Catal Sci Technol 1:366–379 Le Valant A, Garron A, Bion N, Duprez D, Epron F (2011) Effect of higher alcohols on the performances of a 1 % Rh/MgAl2O4/Al2O3 catalyst for hydrogen production by crude bioethanol steam reforming. Int J Hydrog Energy 36:311–318 Ni M, Leung DYC, Leung MKH (2007) A review on reforming bio-ethanol for hydrogen production. Int J Hydrog Energy 32:3238–3247

Hydrogen from Bioethanol Song H (2012) Catalytic hydrogen production from bioethanol. In: Lima MAP, Natalense APP (eds) Bioethanol. InTech Publishing, Janeza Trdine 9, Rijeka, Croatia Vizcaíno AJ, Carrero A, Calles JA (2007) Hydrogen production by ethanol steam reforming over Cu–Ni supported catalysts. Int J Hydrog Energy 32:1450–1461

Further Reading Curcio S (2011) Membranes for advanced biofuels production. In: Basile A, Nunes S (eds) Advanced membrane science and technology for sustainable energy and environmental applications. Woodhead Publishing Limited, Cambridge, UK

H

Hydroprocessor Purge Gases A. Baudot Physics and Analysis Division, IFP Energies nouvelles, Solaize, France

The demand for hydrogen is constantly increasing in refineries due to more and more stringent sulfur content specifications for fuels (leading to an increasing hydrogen consumption in hydrodesulfurization processes) and a growing heavy crude consumption, which results in a higher worldwide demand for highly hydrogenconsuming upgrading processes, such as hydrocracking. In this changing landscape, permeation membranes constitute an elegant option for the recovery of hydrogen that is nowadays wasted in an array of refinery off-gases, such as fuel gas, PSA tail gas, FCC gas, catalytic reformer off-gases, or hydrocracker/hydrotreater off-gases. Nowadays, three main membrane providers offer hydrogen purification permeators: – Air Products with the Prism ® silicon-coated polysulfone membranes (issued from Monsanto). Air Products claims that the lifetime of the Prism ® modules can be more than 15 years. – Ube Industries with polyimide hollow fiber membranes. # Springer-Verlag Berlin Heidelberg 2012 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_298-1

– Air Liquide with the MEDAL polyimide and high selectivity polyaramide membranes. All those membranes are based on glassy polymers and offer a diffusion-based hydrogen selectivity. Today’s membranes have high H2/ CH4 selectivities (from 35 to 200 at 80 C (Roman et al. 2001)). For instance, Air Products claims that a single-stage array of Prism ® modules is able to raise the concentration of gases from 10–30 % up to 70–90 % (Air Products website), while MEDAL membranes are able to raise the concentration of a gas at 51 bar from 86 % in hydrogen up to 98 % with a permeate pressure at 30 bar and a residue containing 52 % of hydrogen at 50 bar (Medal website). In 2003, it was reported that more than 400 hydrogen permeators were installed worldwide (Monereau 2003) while approximately 100 were operated in refineries (Baker 2002). As there are more than 500 refineries in the world, it is clear that the potential market for this type of membrane applications is far from saturated. Nevertheless, three main limits still hinder the wide acceptance of permeation-based hydrogen purification in the refining industry: – The purified hydrogen is recovered at low pressure in the permeate side and requires compression in order to feed it back to reactors. As such, PSA (pressure swing adsorption) is a more attractive process, as the

2

Hydroprocessor Purge Gases

Hydroprocessor Purge Gases, Table 1 Ube Industries membrane material compatibility against contaminants (Ube Industries website) Contaminants Water vapor H2S NH3 and amines Methanol Methyl ether Benzene Toluene C5+ hydrocarbons

produced purified hydrogen is delivered directly at high pressure. – The sensitivity of membrane to contaminants, such as water vapor, higher hydrocarbons, or acid gases. The installation of permeators generally requires at first a very detailed analysis of the contaminants present in the feed, even at traces level, in order to design pretreatment operations. A wide array of solutions can be used in order to remove poisonous compounds: coalescing filters in order to remove aerosols, sorbent beds, or even complete PSA (pressure swing adsorption) or TSA (temperature swing adsorption) units (Monereau 2003). This can lead to a significant increase of investment and operating costs. It should be reminded here that recent hydrogen purification membranes are still relatively tolerant to contaminants (Table 1). This is particularly true if membrane modules are operated at higher temperatures (from 80 C to 110 C), which results in lowering of the sorption of contaminants. – The membrane’s mechanical integrity can be damaged in transient operating conditions, especially in the case of an emergency blow-

Maximum allowable content Up to saturation 3 % vol 100 ppm vol 5 % vol 5 % vol 1 % vol 2000 ppm vol Up to saturation

down of the membrane-based process. In certain cases, the membrane module can be submitted to pressures differences larger than its mechanical tolerance. Solution is nowadays proposed by membrane providers in order to monitor automatically the pressure balance between the feed and the permeate compartment when operating condition limits are reached (Monereau 2003).

References Air

Products website. http://www. airproducts. com/Products/Equipment/PRISMMembranes/page08. htm Baker RW (2002) Future directions of membrane gas separation technology. Ind Eng Chem Res 41:1393–1411 Medal Internet website. www.medal.airliquide.com Monereau C (2003) Perméation hydroge`ne: de la périphérie des procédé vers le coeur des procédés. Intégration des membranes dans les procédés 2:89:275–282 Roman IC, Ubersax RW, Fleming GK (2001) New directions in membrane for gas separation. Chim Industria 83:1–3 Ube Industries website. http://www.ube.com/content. php?pageid=45

H

Hyperbranched Polyimides Ryohei Shindo, Shinji Kanehashi and Kazukiyo Nagai Department of Applied Chemistry, Meiji University, Tama-ku, Kawasaki, Japan

Hyperbranched polyimides are formed by repeated division of branches of comb polyimides. Hyperbranched polymers are synthesized by polymerization of AB2-type monomers and consist of branched structures and mixed straight chains (Fig. 1). Rigorous structural analysis cannot be performed because branching does not regularly occur. However, hyperbranched polymers show properties differing from a normal linear polymer because entanglement of the intermolecular chains is difficult. Flory (1952) showed that gelation of polymerization of ABxtype monomer cannot be statistically performed. Kricheldorf et al. (1982) reported the use of AB2type monomer as one component in copolymers, but their results lacked research attention. Hyperbranched polymers containing various repeating units have been reported since the synthesis of hyperbranched polyphenylene by one-step polymerization of AB2-type monomer was reported as a simple synthesis method of polymers, which were similar in structure to dendrimers by Kim and Webster (1990) of Du Pont. However, dendrimers as structurallycontrolled polymers have been actively studied # Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_300-1

in the field of medicine and for use in catalytic reaction and photoreaction. Hyperbranched polymers have an obvious advantage in synthesis compared with dendrimers. Thus, these polymers can be used as an alternative to dendrimers and a low-viscosity polymer in a wide range of areas. Polyimide is a condensation polymer synthesized from dicarboxylic anhydride and primary diamine. Aromatic heterocyclic polyimides show good mechanical property and superior thermal and oxidation stability. These polyimides are widely used in place of metal and glass. They are also used for high-functional application in electrical engineering, electronic engineering, automobiles, aircraft, and packaging industry. Linear aromatic polyimides are known as polymers, which have poor workability because they are insoluble and infusible in the rigidity of the main chain structure. However, solubility for organic solvent can remarkably improve by the introduction of a multiple branching structure. Hyperbranched polyimides with 4-methylphthalimide as end groups show low dielectric constant, birefringence, and high optical transparency. These qualities result from the improvement in isotropy of molecular chains by the introduction of a multiple branching structure and inhibition of the formation of charge-transfer complex, causing coloration on linear polyimides. For aromatic amine to react easily with acid anhydride at room temperature, isolation of ABx-type monomer with these functional groups in a molecule is difficult because of their instability. Poly(amic acid ester) is

2

Hyperbranched Polyimides

Hyperbranched Polyimides, Fig. 1 Architecture of polymers

synthesized by AB2-type monomer, which has a carboxylic acid ester and two amino groups in a molecule and a condensation agent. Hyperbranched polyimides are synthesized by chemical imidization of poly(amic acid ester) (Yamanaka et al. 2000). Hyperbranched polyimides can be synthesized by selfpolycondensation of ABx-type monomers with imide ring in the monomer framework. Thompson et al. (1999) reported that hyperbranched polyetherimides are synthesized by thermal polycondensation of AB2-type monomers with fluorine (A functional group) and silylated phenolic hydroxyl group (B functional group), which can be detached in a molecule. When A2- and B3-type monomers are used as starting materials in polymerization, AB2-type monomers need not be synthesized. Various hyperbranched polyimides are synthesized because they can be obtained by polymerization using a commercial A2-type monomer and synthesized B3-type monomer. Hyperbranched polyimides have attracted attention as materials for gas separation membranes

since the early 2000s, and most of them are synthesized by A2- and B3-type monomers. The gas permeability of hyperbranched polyimides is almost equal to or higher than that of other glassy polymers such as polysulfone or polycarbonate.

References Flory PJ (1952) Molecular size distribution in threedimensional polymers. VI. Branched polymer containing A-R-Bf-1-type units. J Am Chem Soc 74:2718–2723 Kim YH, Webster OW (1990) Water soluble hyperbranched polyphenylene: “a unimolecular micelle”. J Am Chem Soc 112:4592–4593 Kricheldorf HR, Zang QZ, Schwarz G (1982) New polymer syntheses. 6. Linear and branched poly (3-hydroxybenzoates). Polymer 23:1821–1829 Thompson DS, Markoski LJ, Moore JS (1999) Rapid synthesis of hyperbranched aromatic polyetherimides. Macromolecules 32:4764–4768 Yamanaka K, Jikei M, Kakimoto M (2000) Synthesis of hyperbranched aromatic polyimides via polyamic acid methyl ester precursor. Macromolecules 33:1111–1114

I

Ideal Gas Selectivity Johannes Carolus Jansen Institute on Membrane Technology ITM-CNR, Consiglio Nazionale delle Ricerche, Rende (CS), Italy

The ideal gas selectivity of a membrane, aij, is defined as the ratio of the permeability of two pure gases, measured separately under the same conditions: aij ¼

Pi Pj

(1)

where Pi and Pj are the permeability (or the permeance) of the two pure gases, respectively, with i being the most permeable gas. Rarely the real selectivity is equal to the ideal gas selectivity. Most commonly the ideal selectivity of a membrane is lower than the real selectivity, especially when the more permeable gas species plasticizes the polymer matrix, making it relatively more permeable for the slower species. In some cases, in particular in high free-volume polymers, strong

# Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_301-1

sorption of the more permeable species may obstruct the transport of the less permeable species, making the mixed gas selectivity higher than the ideal selectivity. In dense membranes, where transport occurs by the solution-diffusion mechanism, the permeability is the product of the diffusivity and the solubility: P ¼ D S

(2)

where D is the diffusion coefficient and S is the solubility coefficient. Similarly, the selectivity can be expressed in a diffusion term and a solubility term: aij ¼

D i Si D j Sj

(3)

The ideal gas selectivity is an intrinsic property, specific for the membrane material and the particular gas pair. However, it is not a constant but it depends on the operation conditions temperature and pressure because both D and S depend on the temperature and on the operating pressure.

I

Ideal Separation Factor Johannes Carolus Jansen Institute on Membrane Technology ITM-CNR, Consiglio Nazionale delle Ricerche, Rende (CS), Italy

The separation factor, SF, is a measure of the efficiency of the separation process and is determined from the ratio of the concentrations of the more permeable gas species i and the less permeable gas species j in the permeate divided by the ratio of the same gases i and j in the feed stream: SF ¼

xi, p =xj, p xi, f =xj, f

(1)

# Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_302-1

where xi,p and xj,p are the fractions of components i and j in the permeate and xi,f and xj,f are the fractions of components i and j in the feed. The separation factor is not a material property, but it also depends on the conditions of the separation process. It depends both on the membrane properties and on the driving force, which in turn depends on the pressure and on, for instance, the presence of concentration polarization phenomena, nonideal behaviour such as plasticization, coupling effect, etc. Analogously, the ideal separation factor is the separation factor under ideal conditions. It can be calculated from the pure gas permeabilities.

I

IgG Purification Nilay Bereli, Deniz T€ urkmen, Handan Yavuz and Adil Denizli Department of Chemistry, Biochemistry Division, Hacettepe University, Ankara, Turkey

Owing to their use in the treatment of various diseases, such as primary and secondary immune deficiencies, infections, and inflammatory and autoimmune diseases, large number of immunoglobulin G (IgG) products is under clinical development. This requires certain protocols for purification and standardization. Affinity chromatography is the most popular technique to reach these requirements (Low et al. 2007). Staphylococcal protein A is one of the first affinity ligands with a very high specificity for IgG purification. It interacts with IgG through hydrophobic interactions and some hydrogen bonds and electrostatic interactions. Main disadvantages of protein A-containing carriers are the possible ligand leakage that contaminates the therapeutic product, and also they are expensive and difficult to handle, sterilize, and preserve (F€ uglistaller 1989). Pseudo-specific ligands, such as histidine, tryptophan, phenylalanine, etc., can be used for the IgG purification. They are small molecules with high physical and chemical stability and low cost (Altıntas¸ and Denizli 2009; T€ urkmen et al. 2008). The interaction of histidine with IgG has been shown to be # Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_303-1

water mediated involving the combined electrostatic, hydrophobic, and charge-transfer interactions between histidine and the specific amino acid residues available on the protein surface (Bhattacharyya et al. 2003). In immobilized metal ion affinity chromatography (IMAC), the separation is based on the interaction of a Lewis acid (electron pair acceptor), i.e., a chelated metal ion, with electron donor atoms (N, O, and S) on the side groups of the surface histidine, tryptophan, and cysteine of the protein. Histidine-rich sequence-containing IgGs show an innate affinity for metal ions, and IMAC allows one-step separation of IgG (Altıntas¸ et al. 2007). Textile dyes bind proteins in a selective and reversible manner and can be used for antibody purification (Denizli and Pis¸kin 2001). Dye ligands can engage in ionic, hydrophobic, charge-transfer, and hydrogen bonding with proteins. In the thiophilic adsorption of proteins, electron donor-acceptor interactions between both functional groups present in the ligand structure and the adjacent sulfone group are the driving force for selective recognition (Bakhspour et al. 2014). In general, specificity, rapid processing, mild operation conditions, conventional equipment, and reusability determine which technique to be used for IgG purification. The use of membranes has become indispensible for chromatographic applications in both research and industry area for the last few decades due to their relatively wide configuration for the size-, charge-, and affinity-based protein separation and purification. The pressure

2

drop across the membranes is very low due to the large pore size. Owing to the continuous pore structure, mass transport occurs by convection rather than by diffusion. Chromatographic membranes are generally cost effective, and their scale-up is easier than the packed-bed chromatography (Charcosset 1998). Membrane operations including ultrafiltration (Mohanty and Ghosh 2008; Rosenberg et al. 2009), dialysis (Bruce et al. 2002), and affinity membrane chromatography (Boi et al. 2009) have been demonstrated for their potential for IgG purification.

References Altıntas¸ EB, Denizli A (2009) Monosize magnetic hydrophobic beads for lysozyme purification under magnetic field. Mater Sci Eng C 29:1627 Altıntas¸ EB, T€uzmen N, Uzun L, Denizli A (2007) Immobilized metal affinity adsorption for antibody depletion from human serum with monosize beads. Ind Eng Chem Res 46:7802 Bakhspour M, Bereli N, S¸enel S (2014) Preparation and characterization of thiophilic cryogels with 2mercaptoethanol as the ligand for IgG purification. Colloid Surf B 113:261 Bhattacharyya R, Saha RP, Samana U, Chakrabarti P (2003) Geometry of interaction of the histidine ring

IgG Purification with other planar and basic residues. J Proteome Res 2:255 Boi C, Busini V, Salvalaglio M, Cavallotti C, Sarti GC (2009) Understanding ligand-protein interactions in affinity membrane chromatography for antibody purification. J Chromatogr A 1216:8687–8696 Bruce MP, Boyd V, Duch C, White JR (2002) Dialysisbased bioreactor systems for the production of monoclonal antibodies-alternatives to ascites production in mice. J Immunol Methods 264:59–68 Charcosset C (1998) Purification of proteins by membrane chromatography. J Chem Technol Biotechnol 71:95 Denizli A, Pis¸kin E (2001) Dye-ligand affinity systems. J Biochem Biophys Methods 49:391 F€ uglistaller P (1989) Comparison of immunoglobulin binding capacities and ligand leakage using eight different protein A affinity matrices. J Immunol Methods 124:171 Low D, O’Leary R, Pujar NS (2007) Future of antibody purification. J Chromatogr B 848:48 Mohanty K, Ghosh R (2008) Novel tangential-flow countercurrent cascade ultrafiltration configuration for continuous purification of humanized monoclonal antibody. J Membr Sci 307:117–125 Rosenberg E, Hepbildikler S, Kuhne W, Winter G (2009) Ultrafiltration concentration of monoclonal antibody solutions: development of an optimized method minimizing aggregation. J Membr Sci 342:50–59 ¨ zt€ T€ urkmen D, O urk N, Elkak A, Akgo¨l S, Denizli A (2008) Phenylalanine containing hydrophobic nanospheres for antibody purification. Biotechnol Prog 24:1297

I

Immunoaffinity Membranes Nilay Bereli, Handan Yavuz and Adil Denizli Department of Chemistry, Biochemistry Division, Hacettepe University, Ankara, Turkey

Immunoaffinity chromatography is a process in which the specific binding of an antigen to its specific antibody is utilized (Subramanian 2002). The specificity of the binding makes this technique a very useful tool for the applications in which selective and strong antigen-antibody binding is advantageous. Immunoadsorption, in general, can be used for the purpose of therapy as well as preparative chromatography. Normally, the human immune system works to recognize, respond, and destroy pathogenic substances. When the ability of the immune systems to recognize foreign antigens versus healthy cells or tissues is failed, arising immune complexes, so-called autoantibodies, cause many kinds of autoimmune diseases (Massey and McPherson 2007). For example, myasthenia gravis, autoimmune hemolytic anemia and immune thrombocytopenic purpura, rheumatoid arthritis, systemic lupus erythematosus, thyroiditis, and insulindependent diabetes mellitus are such diseases. The immunoadsorption columns have been used for the treatment of immune diseases since the mid-1970s, in a study performed for the removal of DNA antibodies (Terman et al. 1974). Since then immunoadsorption therapy with affinity # Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_305-1

adsorbents using target specific antibodies has been increasingly utilized to remove pathogenic autoantibodies from patients’ plasma (Uzun et al. 2010). Besides their use in the treatment of autoimmune diseases, immunoaffinity membranes can be used for the purification of antibodies or antigens with a high purity and also used for the selective and specific removal of toxic substances from human plasma (Denizli 2002). In such applications, membrane-based columns have advantages over traditional columns in terms of compressibility of the particles, the fouling and slow flow rate through the column. Especially in contact with blood, stacked membrane system is desirable because of high convective transport rates without cell damage. The other desirable properties of affinity membranes are high porosity; large internal surface area; high chemical, biological, and mechanical stabilities; hydrophilicity; low nonspecific adsorption of blood proteins; and the presence of functional groups for derivatization (Denizli 2011).

References Denizli A (2002) Preparation of immuno-affinity membranes for cholesterol removal from human plasma. J Chromatogr B 772:357 Denizli A (2011) Autoimmune diseases and immunoadsorption therapy. Hacettepe J Biol Chem 39(3):213 Massey HD, McPherson RA (2007) Human leukocyte antigen: the major histocompatibility complex of

2 man. In: Henry’s clinical diagnosis and management by laboratory methods. 21st edn, McPherson RA, Pincus AR (eds); Saunders-Elsevier publishes, Philadelphia pp. 876-893 Subramanian A (2002) Immunoaffinity chromatography. Mol Biotechnol 20:41

Immunoaffinity Membranes Terman DS, Stewart I, Hofmann A, Carr R, Harbeck R (1974) Specific removal of DNA antibody with an immunoadsorbent. Experientia 30:1493 Uzun L, Yavuz H, Osman B, C¸elik H, Denizli A (2010) PHEMA based affinity membranes for in-vitro removal of anti-dsDNA antibodies from SLE plasma. Int J Biol Macromol 47:44

I

Imprinted Polysulfone-Aldehyde Derivatized Nanofiber Membranes Masakazu Yoshikawa Department of Biomolecular Engineering, Kyoto Institute of Technology, Kyoto, Japan

The functionalization of aromatic polysulfones for tailoring properties in membrane applications is of great interest. Polysulfone has overall thermal and chemical stability combined with good mechanical and membrane-making qualities. Polysulfone is a stable platform for functional group attachment and a good candidate polymeric material for membranes with tailored functionalities (Guiver et al. 1999). To this end, modified polysulfones have been intensively studied in connection with chiral separation (Yoshikawa et al. 1998, 2005, 2006, 2007; Mizushima et al. 2011; Sueyoshi et al. 2012), pervaporation separation (Yoshikawa et al. 1992a, b, 1999), and selective separation of CO2 (Yoshikawa et al. 2000). In membrane separation, both flux and permselectivity are important factors. The enhancement of permselectivity would be relatively easily attained by application of molecular imprinting so that molecular recognition sites, which specifically incorporate target substrate into the membrane, can be introduced into a

# Springer-Verlag Berlin Heidelberg 2012 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_306-1

given membrane. However, the enhancement of flux without a concurrent reduction in permselectivity is perceived to be an unsolved problem or an unsolvable problem in membrane separation. In other words, flux and permselectivity often show a trade-off relationship. Membranes with high surface area and high porosity would be expected to break such a tradeoff relationship between flux and permselectivity. Nanofiber membranes are expected to simultaneously give both high flux and high permselectivity. To this end, nanofiber membranes with molecular recognition sites, which are called molecularly imprinted nanofiber membranes, were fabricated by simultaneously applying an alternative molecular imprinting and an electrospray deposition (Sueyoshi et al. 2010; Yoshikawa et al. 2007). Those studies revealed that molecularly imprinted nanofiber membranes gave high flux without a concurrent reduction in permselectivity. A breakthrough in membrane separation was attained; in other words, membrane morphology in the form of molecularly imprinted nanofiber fabric was one of the suitable membrane forms to solve a tradeoff relationship in membrane separation. Molecularly imprinted nanofiber membranes and usual molecularly imprinted membranes were fabricated from polysulfone with aldehyde group (PSf-CHO) and print molecules, and their membrane performances, such as adsorption

2

Imprinted Polysulfone-Aldehyde Derivatized Nanofiber Membranes

Imprinted PolysulfoneAldehyde Derivatized Nanofiber Membranes, Fig. 1 Chemical structures of polysulfone with aldehyde group (PSf-CHO) and print molecule (Z-D-Glu or Z-L-Glu) (Cited from Sueyoshi et al. 2012 with permission. Copyright 2012 Elsevier Inc)

Imprinted PolysulfoneAldehyde Derivatized Nanofiber Membranes, Fig. 2 Schematic illustration for the fabrication of molecularly imprinted nanofiber membranes, where PSf-CHO and Z-Glu were simultaneously electrosprayed

selectivity, permselectivity, and flux, were studied (Sueyoshi et al. 2012). PSf-CHO with degree of substitution of 0.50 and 1.00 were adopted as candidate materials, and the derivative of D- or L-glutamic acid was

applied as a print molecule to obtain molecularly imprinted nanofiber membranes and usual molecularly imprinted membranes for optical resolution (Figs. 1 and 2).

Imprinted Polysulfone-Aldehyde Derivatized Nanofiber Membranes

3

105 Δ[Glu]R / mol dm−3

2.0 D-Glu L-Glu NaN3

1.5

NaN3

L R membrane

1.0 a L/D = 1.20 0.5 D-Glu L-Glu 0

2.0

4.0

6.0

8.0

10.0

Time / h Imprinted Polysulfone-Aldehyde Derivatized Nanofiber Membranes, Fig. 3 Adsorption isotherm of D-Glu and L-Glu in the nanofiber membrane molecularly imprinted by Z-D-Glu. (PSf-CHO-10 was adopted as a candidate material) (Cited from Sueyoshi et al. 2012 with permission. Copyright 2012 Elsevier Inc)

Imprinted Polysulfone-Aldehyde Derivatized Nanofiber Membranes, Fig. 4 Time-transport curves of racemic Glus through the nanofiber membrane molecularly imprinted by Z-D-Glu (PSf-CHO-10 was adopted as a candidate material) (Cited from Sueyoshi et al. 2012 with permission. Copyright 2012 Elsevier Inc)

The membranes molecularly imprinted by Z-D-Glu incorporated the D-isomer of glutamic acid (Glu) in preference to the corresponding L-isomer and vice versa. In other words, the membrane imprinted by the L-isomer selectively adsorbed the L-isomer. Figure 3 shows the adsorption isotherms of D-Glu and L-Glu for the Z-D-Glu molecularly imprinted nanofiber membrane as an example of adsorption isotherms. The adsorption isotherm of D-Glu, which was preferentially adsorbed in the membrane, shows a dual adsorption isotherm. It consists of nonspecific adsorption and specific adsorption on the specific recognition site, which was constructed by the presence of a print molecule during the membrane preparation process. Contrary to this, L-Glu, which was nonspecifically adsorbed in the membrane, gives a straight line passing through the origin. Time-transport curves of racemic Glu through the D-isomer molecularly imprinted nanofiber membrane are shown in Fig. 4. As often observed, the transport of the enantiomer

preferentially incorporated into the membrane was retarded by a relatively strong interaction between the enantiomer and the membrane. As a result, the antipode was selectively transported. Such discrepancy between adsorption selectivity and permselectivity is often observed in chiral separation. Table 1 summarizes membrane performances for two types of molecularly imprinted membrane. As can be seen, the fluxes through the molecularly imprinted nanofiber membranes gave one to two orders of magnitude higher than those of usual molecularly imprinted membranes without depression of permselectivity. As proved in the previous studies (Sueyoshi et al. 2010; Yoshikawa et al. 2007), the present study revealed that molecularly imprinted nanofiber membranes gave high flux without depression of permselectivity. The emergence of molecularly imprinted nanofiber membrane would solve a trade-off relationship in membrane separation (Yoshikawa et al. 2011).

4

Imprinted Polysulfone-Aldehyde Derivatized Nanofiber Membranes

Imprinted Polysulfone-Aldehyde Derivatized Nanofiber Membranes, Table 1 Results of chiral separation with molecularly imprinted nanofiber (MINFM’s) and molecularly imprinted (MIPM’s) membranes Mmebrane MINFM-10a MIPM-10a MINFM-05b MIPM-05b

Z-D-Glu imprinted membrane aL/D u0 1.24 1.15 10 9 1.20 4.20 10 11 1.12 7.00 10 9 1.25 6.64 10 11

(28) (1) (231) (2.2)

Z-L-Glu imprinted membrane aD/L u0 1.20 1.67 10 9 1.20 4.10 10 11 1.20 2.20 10 9 1.16 3.05 10 11

(41) (1) (72) (1)

a

Figures in parentheses are the relative values; the U Value for MIPM-10 imprinted by Z-L-Glu being set as unity Figures in parentheses are the relative values; the U Value for MIPM-05 imprinted by Z-L-Glu being set as unity c U = (–J/C)/(dm/dx) [{(mol cm cm-2 h-1)/(mol cm-3)}/(J mol-1 cm-1) = mol cm cm-2 J-1 h-1]. (Cited from ref. Sueyoshi et al. 2012 with permission. Copyright 2012 Elsevier lnc.) b

References Guiver MD, Robertson GP, Yoshikawa M, Tan CM (1999) Functionalized polysulfones: methods for chemical modification and membrane applications. In: Pinnau I, Freeman BD (eds) Membrane formation and modification. ACS symposium series, vol 744. ACS, Washington, DC, pp 137–161 Mizushima H, Yoshikawa M, Robertson GP, Guiver MD (2011) Optical resolution membranes from polysulfones bearing alanine derivatives as chiral selectors. Makromol Mater Eng 296:562–567 Sueyoshi Y, Fukushima C, Yoshiakwa M (2010) Molecularly imprinted nanofiber membranes from cellulose acetate aimed for chiral separation. J Membr Sci 357:90–97 Sueyoshi Y, Utsunomiya A, Yoshiakwa M, Robertson GP, Guiver MD (2012) Chiral separation with molecularly imprinted polysulfone-aldehyde derivatized nanofiber membranes. J Membr Sci 401–402:89–96 Yoshikawa M, Hara H, Tanigaki M, Guiver M, Matsuura T (1992a) Modified polysulfone membranes: 1. Pervaporation of water/alcohol mixtures through modified polysulfone membranes having methyl ester moiety. Polymer 33:4805–4813 Yoshikawa M, Hara H, Tanigaki M, Guiver M, Matsuura T (1992b) Modified polysulfone membranes II. Pervaporation of aqueous ethanol solution through modified polysulfone membranes bearing various hydroxyl groups. Polym J 24:1049–1055 Yoshikawa M, Izumi J, Ooi T, Kitao T, Guiver MD, Robertson GP (1998) Carboxylated polysulfone

membranes having a chiral recognition site induced by an alternative molecular imprinting technique. Polym Bull 40:517–524 Yoshikawa M, Tsubouchi K, Guiver MD, Robertson GP (1999) Modified polysulfone membranes. III. Pervaporation separation of benzene-cyclohexane mixtures through carboxylated polysulfone membranes. J Appl Polym Sci 74:407–412 Yoshikawa M, Niimi A, Guiver MD, Robertson GP (2000) Modified polysulfone membranes. IV. Gas separation with aminated polysulfone membranes. Sen’i Gakkaishi 56:272–281 Yoshikawa M, Hanaoka K, Guiver MD, Robertson GP (2005) Chiral separation of racemic amino acids through membranes derived from modified polysulfone having perillaldehyde moiety as a side group. Membrane 30:219–225 Yoshikawa M, Murakoshi K, Kogita T, Hanaoka K, Guiver MD, Robertson GP (2006) Chiral separation membranes from modified polysulfone having myrtenal-derived terpenoid side groups. Eur Polym J 42:2532–2539 Yoshikawa M, Nakai K, Matsumoto H, Tanioka A, Guiver MD, Robertson GP (2007) Molecularly imprinted nanofiber membranes from carboxylated polysulfone by electrospray deposition. Macromol Rapid Commun 28:2100–2105 Yoshikawa M, Tanioka A, Matsumoto H (2011) Molecularly imprinted nanofiber membranes. Curr Opin Chem Eng 1:18–26

I

Inert Membrane Rune Bredesen Sustainable Energy Technology, SINTEF Materials and Chemistry, Oslo, Norway

The term inert membrane denotes that no change in chemical reaction occurs due to contact between the membrane material and the surrounding constituents. A reaction occurring between constituents A and B to form C and D can be used as an example: AþB¼CþD

(1)

If the membrane is catalytically inactive with respect to the reaction, the term inert membrane is used to describe the membrane. The presence of a catalyst, deemed not to be part of the membrane material, may yield a change in reaction 1, and the integration of such a combination of membrane and catalyst is referred to as an ▶ inert membrane reactor (IMR). The term inert membrane is therefore commonly used in connection with membrane reactors (Koros et al. 1996) to distinguish the membrane properties from those of a catalytic membrane, the latter being catalytically active. In such reactors, the inert membrane may be used for selectively separating reaction products from the reaction for which the catalyst serves to activate. Alternatively, the inert membrane may be employed as # Springer-Verlag Berlin Heidelberg 2012 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_309-1

a distributor of reactants to ensure controlled delivery to the catalytic reaction site. Classification of a membrane as inert may also depend on the operation conditions and surface properties since the actual catalytic activity depends on parameters such as temperature, surface area, and surrounding chemical composition. Thus the same membrane may, or may not, be an inert membrane depending on the conditions under which it is operated. The interactions between an inert membrane and its surroundings typically involve surface adsorption/desorption reactions, which may be followed by other reactions necessary for transport of matter within the membrane. For example, in the case of dense polymeric membranes, incorporation of the permeant gas molecule is required on the feed side, while the reverse process is required on the permeate side (Mulder 1996). Dense inorganic membranes, in addition, require transformation of the adsorbed molecule to atomic (in the case of metal membranes (Ward and Dao 1999)) or ionic and electronic (in the case of ceramic membranes (Sirman 2006) species at the feed side, which are then able to diffuse through the bulk membrane phase. At the permeate side, the recombination of species to the same molecular form as on the feed side takes place before desorption to the gas phase. To enhance the transformation of adsorbed gas molecules to diffusing species within the membrane and, thus, contribute to higher fluxes, catalytic surface properties are usually required. Nevertheless,

2

such membranes which although they incorporate catalytic surface reactions, they are regarded as inert membranes since the reaction is present solely as a means of sustaining transport of the gas molecule from one membrane side to the other. Another example is inert membranes for liquid separation applications, where surface hydrophilicity and hydrophobicity may completely determine the membrane transport properties. The term inert membrane is rarely used for conventional membrane separation processes; however, one may find the term inert membrane used in such phrases as “chemically inert membrane” or “bio-inert membrane.” These expressions refer to a specific property of the membrane such as chemical stability or compatibility in the case of chemical inertness or biological inactivity in the case of bio-inertness. In these

Inert Membrane

cases the term inert membrane has a somewhat different meaning than that related to the IMRs.

References Koros WJ, Ma YH, Shimidzu T (1996) Terminology for membranes and membrane processes. Pure Appl Chem 68:1479–1489 Mulder M (1996) Basic principles of membrane technology. Kluwer Academic Publishers, Dordrecht, The Netherlands Sirman J (2006) The evolution of materials and architecture for oxygen transport membranes. In: Sammells AF, Mundschau MV (eds) Nonporous inorganic membranes. Wiley-VCH, Weinheim, pp 165–184 Ward TL, Dao T (1999) Model of hydrogen permeation behavior in palladium membranes. J Membr Sci 153:211–231

I

Inert Membrane Reactors Rune Bredesen Sustainable Energy Technology, SINTEF Materials and Chemistry, Oslo, Norway

A membrane, defined as a barrier between two phases through which transport of one or several species occurs, can be made from virtually any solid or liquid material or combinations of both. Membranes are commonly divided into inorganic, polymeric, hybrid inorganic and polymeric, or as dual phase consisting of a solid phase and a liquid phase. Furthermore, the membrane may either be dense or porous with a continuous network of pores. A membrane reactor (MR) is a device for simultaneously carrying out a reaction and membrane-based separation in the same physical enclosure (Koros et al. 1996). In an inert membrane reactor (IMR), the inert membrane (link) and the catalyst in the form of a separate solid or a liquid phase are contained in the reactor (Fontananova and Drioli 2010; Julbe et al. 2001; Coronas and Santamaria 1999). Figure 1 illustrates different catalyst and inert membrane combinations in an IMR. Due to the separation of membrane and catalyst in the IMR, the membrane-based process and catalytic reactions occur in sequence. This decoupling of the processes can be advantageous with respect to operation and replacement of membrane and/or # Springer-Verlag Berlin Heidelberg 2012 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_310-1

catalyst compared to MRs where the membrane serves as both catalyst and membrane. The two main functions of the membrane applied in IMRs are as an extractor or a distributor (Julbe et al. 2001; Dalmon 1997) (see Fig. 2). As an extractor, the inert membrane selectively separates a reaction product or intermediate product. The advantage may be higher conversion (equilibrium-limited reactions) or/and higher selectivity (e.g., via extraction of an intermediate that would otherwise lead to subsequent unwanted reactions). Extraction of hydrogen from hydrocarbon dehydrogenation, reforming, or water gas shift reactions, employing hydrogen selective membranes in combination with packed or fluidized catalyst beds, has been widely studied (Sanchez Marcano and Tsotsis 2002). Dehydrogenation and reforming are typically equilibrium-limited endothermic reactions, and membrane reactor applications may benefit from the use of lower operating temperature and/or higher pressure without sacrificing yield. Esterification is yet another example of catalyzed equilibrium-limited reactions where water extraction by an inert membrane is used to increase yield (Van der Bruggen 2010). As a distributor, the inert membrane delivers the reactant in a controlled manner to the catalyzed reaction taking place in the reactor compartment (Julbe et al. 2001). The typical aim is to enhance selectivity through careful addition and temperature control of exothermic oxidation and hydrogenation reactions. Some common examples are

2

Inert Membrane Reactors

Inert Membrane Reactors, Fig. 1 Inert membrane (a) with solid catalyst, (b) with liquid containing the catalyst, (c) encapsulating solid catalyst, (d) encapsulating liquid containing the catalyst

Inert Membrane Reactors, Fig. 2 (a) Membrane extractor operation, (b) membrane distributor operation

IMR with packed catalyst bed for oxidative coupling of methane to C2, or oxidative dehydrogenation of hydrocarbons, by addition of oxygen and hydrogenation of alkenes by hydrogen addition. For some oxidation reactions employing IMRs, the separation of the bulk of the reactants by the membrane wall lowers the explosion potential (Coronas and Santamaria 1999). As most membrane materials are inert, the incorporation of many different types into IMRs has been studied. Cheap polymeric membranes are advantageous with respect to capital cost, high packing density, and simple sealing technology in modules. However, their low operating temperature, typically less than 100 C, and limited chemical stability narrow the range of applications. More expensive inorganic membranes enable high temperature operation, but since stability and transport properties are usually very temperature dependent, the different inorganic membranes have a limited window of operation. Several decades of research have shown that a

number of challenges still exist with respect to high temperature applications, and commercial use of IMRs is still a future prospect. To aid reaching this future prospect, mathematical modeling and simulation is required to develop IMR design and provide an understanding of the optimal operating conditions.

References Coronas J, Santamaria J (1999) Catalytic reactors based on porous ceramic membranes. Catal Today 51:377–389 Dalmon JA (1997) Catalytic membrane reactors. In: Ertl G, Kno¨zinger H, Weitkamp J (eds) Handbook of heterogeneous catalysis, vol 3. Wiley-VCH Weinheim, Germany, pp 1387–1398 Fontananova E, Drioli E (2010) Catalytic membranes and membrane reactors. In: Drioli E, Giorno L (eds) Comprehensive membrane science and engineering, vol 3. Elsevier, Amsterdam, pp 109–133 Julbe A, Farrusseng D, Guizard C (2001) Porous ceramic membranes for catalytic reactors – overview and new ideas. J Membr Sci 181:3–20

Inert Membrane Reactors Koros WJ, Ma YH, Shimidzu T (1996) Terminology for membranes and membrane processes. Pure Appl Chem 68:1479–1489 Sanchez Marcano JG, Tsotsis TT (2002) Catalytic membranes and membrane reactors. Wiley-VCH, Weinheim

3 Van der Bruggen B (2010) Pervaporation membrane reactors. In: Drioli E, Giorno L (eds) Comprehensive membrane science and engineering, vol 3. Elsevier, Amsterdam, pp 135–163

F

Filler in Membranes Antonino Salvatore Arico CNR-ITAE Institute, Messina, Italy

Synonyms Inorganic fillers Composite perfluorosulfonic acid membranes containing different types of inorganic fillers such as hygroscopic oxides, surface-modified oxides, zeolites, inorganic proton conductors, etc. have shown an increased conductivity with respect to the bare perfluorosulfonic membranes at high temperature (Arico` et al. 1998, 2003). The presence of hygroscopic inorganic oxides inside the composite membrane besides extending the operation of perfluorosulfonic membranes (e.g., Nafion®) in the high-temperature range reduces the crossover effects by increasing the “tortuosity factor” in the permeation path. Such effects are particularly serious at high temperature in DMFC systems. An appropriate tailoring of the surface chemistry in these nanoparticles is a key step to enhance water retention at high temperature. Composite recast Nafion ® membranes containing inorganic fillers have been employed in high-

# Springer-Verlag Berlin Heidelberg 2012 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_313-1

temperature (~150 C) direct alcohol (Arico` et al. 1998) and H2-air fuel cells (Watanabe et al.1996). These composite membranes were originally developed for reduced humidification operation in polymer electrolyte fuel cells (Watanabe et al.1996) due to the enhanced water retention inside the membrane by the effect of the inorganic filler (Arico` et al. 1998). A further advantage of composite membranes relies in the barrier effect given by the inorganic filler for methanol crossover (Ren et al. 1996).

References Arico` AS, Cretı` P, Antonucci PL, Antonucci V (1998) Comparison of ethanol and methanol oxidation in a liquid-feed solid polymer electrolyte fuel cell at high temperature. Electrochem Solid-State Lett 1:66–68 Arico` AS, Baglio V, Di Blasi A, Creti P, Antonucci PL, Antonucci V (2003) Influence of the acid–base characteristics of inorganic fillers on the high temperature performance of composite membranes in direct methanol fuel cells. Solid State Ionics 161:251–265 Ren X, Wilson MS, Gottesfeld S (1996) High performance direct methanol polymer electrolyte fuel cells. J Electrochem Soc 143:L12 Watanabe M, Uchida H, Seki Y, Emori M, Stonehart P (1996) Self-humidifying polymer electrolyte membranes for fuel cells. J Electrochem Soc 143:3847

I

Inorganic Catalytic Membrane Rune Bredesen Sustainable Energy Technology, SINTEF Materials and Chemistry, Oslo, Norway

An inorganic catalytic membrane is an inorganic membrane that is catalytically active. Inorganic membranes are made from metals, ceramics, glass, and carbon and may be porous or dense. The term inorganic catalytic membrane is typically meant in the context of membrane reactors where yield is enhanced by combining a membrane separation and a catalyzed chemical reaction. Inorganic catalytic membranes can either be composed of an inherently catalytically active material or an inert membrane structure to which a catalytically active phase is added on the outer surface or inside pores (Specchia et al. 2006), see Fig. 1. In the latter case, integration of catalyst material and inorganic membrane structure is carried out by common deposition and impregnation methods such as wet (electro) chemical deposition, chemical vapor deposition, or physical vapor deposition. Since the membrane is catalytic, reaction and separation occur in parallel a feature which distinguishes it from the sequential processes of inert membrane reactors where the inert membrane and catalyst material are separated. The function of the membrane may serve as extractor, distributor, or contactor depending on application (Dalmon 1997) (see # Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_314-2

Fig. 2) and the reaction mechanism can be (electro)catalytic, photo-catalytic, or bio-catalytic. The many different concepts and envisaged applications are described comprehensively in various text books and scientific papers (Gryaznov et al. 2006; Sanchez Marcano and Tsotsis 2002; Caro 2010; Fontananova and Drioli 2010). Inorganic catalytic membranes can operate at higher temperatures than their polymeric counterpart, which opens a broader window of operation as high temperature is also required for many important reactions which are limited by low yield in traditional reactors. The assembling of inorganic catalytic membranes in membrane modules may, however, give too low a catalytic surface area and limited conversion and therefore render operation less cost efficient compared to more conventional reactors. To mitigate this problem, additional catalyst can be added to the reactor volume. In order to increase efficiency, some current development activities are aimed at manufacturing inorganic thin capillary, multichannel, and hollow fiber membranes with high surface area/volume ratio. The dual operational nature of inorganic catalytic membranes is particularly challenging since both membrane and catalytic properties must work satisfactory. Commonly encountered issues related to degradation processes such as adsorption, poisoning, clogging, and fouling are all critical and membrane lifetime may suffer when trying to achieve the optimal trade-off between

2

a

Inorganic Catalytic Membrane

b

c d

e

Inorganic Catalytic Membrane, Fig. 1 Inherently catalytic (a) porous, (b) dense membrane. Inert membrane with catalyst deposited (c) inside pores, on outer surface of (d) porous, and (e) dense membrane

Inorganic Catalytic Membrane, Fig. 2 Inorganic catalytic membrane in (a) extractor operation, (b) distributor operation, (c) flow-through contactor operation, (d) liquid–gas contactor operation

membrane and catalyst properties. Common examples of applications include hydrocarbon dehydrogenation, hydrocarbon reforming, and hydrogenation of alkenes for which hydrogen selective inorganic catalytic membranes can be used (Sanchez Marcano and Tsotsis 2002). Combination of endothermic (e.g., water splitting to produce hydrogen) and exothermic (e.g., partial oxidation of methane) reactions for integrated mass and heat transport has been demonstrated by using catalytically active oxygen permeable membranes (Caro 2010). By feeding oxygen in a controlled manner, partial oxidation of hydrocarbons to form synthesis gas, oxidative coupling of methane, and oxidative dehydrogenation reactions have been demonstrated using dense and porous inorganic catalytic membranes (Sanchez Marcano and Tsotsis 2002; Coronas and Santamaria 1999; Sammells et al. 2006). When porous inorganic catalytic membranes are used as flow-through contactors, a forced close spatial contact between reactants and catalyst can give improved conversion efficiency compared to

reactions in catalyst powder beds (Westermann and Melin 2009). Additionally, higher selectivity may be achieved due to the short and welldefined contact time between reactants and membrane catalyst. Yet in another contactor mode, the porous inorganic catalytic membrane provides a defined catalytic region for reaction and controls transport of reactants from both sides of the membrane. For example, in liquid–gas catalytic contactors, both hydrogenation and oxidative reactions have been demonstrated.

References Caro J (2010) Basic aspects of membrane reactors. In: Drioli E, Giorno L (eds) Comprehensive membrane science and engineering, vol 3. Elsevier, Amsterdam, pp 1–24 Coronas J, Santamaria J (1999) Catalytic reactors based on porous ceramic membranes. Catal Today 51:377–389 Dalmon JA (1997) Catalytic membrane reactors. In: Ertl G, Kno¨zinger H, Weitkamp J (eds) Handbook of heterogeneous catalysis, vol 3. Wiley-VCH, Weinheim, pp 1387–1398

Inorganic Catalytic Membrane Fontananova E, Drioli E (2010) Catalytic membranes and membrane reactors. In: Drioli E, Giorno L (eds) Comprehensive membrane science and engineering, vol 3. Elsevier, Amsterdam, pp 109–133 Gryaznov VM, Ermilova MM, Orekhova NV, Teresschenko GF (2006) Reactors with metal and metal-containing membranes. In: Cybulski A, Moulijn JA (eds) Structured catalysts and reactors. Chemical industries, 2nd edn. Taylor and Francis Group, Boca Raton, pp 579–614 Sammells AF, White JH, Makay R (2006) Membranes for promoting partial oxidation chemistries. In: Sammells

3 AF, Mundschau MV (eds) Nonporous inorganic membranes. Wiley-VCH, Weinheim, pp 185–214 Sanchez Marcano JG, Tsotsis TT (2002) Catalytic membranes and membrane reactors. Wiley-VCH, Weinheim Specchia S, Fino D, Saracco G, Specchia V (2006) Reactors with metal and metal-containing membranes. In: Cybulski A, Moulijn JA (eds) Structured catalysts and reactors. Chemical industries, 2nd edn. Taylor and Francis Group, Boca Raton, pp 615–661 Westermann T, Melin T (2009) Flow-through catalytic membrane reactors-principles and applications. Chem Eng Process 48:17–28

I

Inorganic Scaling Tony Fane School of Chemical Engineering, UNSW, The University of New South Wales, Sydney, NSW, Australia

Synonyms Inorganic fouling Inorganic scaling occurs in reverse osmosis (RO) applications and is caused by the retention of sparingly soluble mineral salts such as calcium carbonate, calcium sulfate, calcium phosphate, barium sulfate, magnesium salts, silica, etc. Due to water removal and concentration polarization (CP), the concentrations build up and can exceed the saturation level and cause scaling of the membrane surface. Membrane scaling has two pathways, namely, surface crystallization and bulk crystallization. For surface crystallization, CP causes the sparingly soluble salt concentration at the membrane surface to exceed the solubility limit. Consequently, the surface is gradually covered by the lateral growth of a crystal deposit that reduces the effective area for permeation. For fixed imposed flux, the local flux increases in order to compensate for loss of area, and this increases the local CP level which exacerbates the scale formation. In the other pathways, crystals form in the bulk phase when concentrations # Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_317-1

exceed saturation level due to high recovery (defined as fraction of feedwater removed as permeate). Crystals in suspension are then deposited on the membrane and eventually form a porous cake layer, which provides a hydraulic resistance. There are many membrane elements connected in series in a RO plant, and concentration builds up from the feed to the tail end. Scale formation is therefore more prevalent in the tail end elements of the plant. It is likely that both scaling mechanisms, surface and bulk, could occur simultaneously in a RO system. This is because a crystal cake causes “cake-enhanced concentration polarization” (CECP) (see “Irreversible Fouling Resistance”), and the increased CP of the scale formers leads to precipitation on the membrane and within the cake. There are several strategies to control scaling in membrane systems. Bulk crystallization can be mitigated by controlling the supersaturation level of the salts by limiting the maximum recovery in the RO system. The effect of CP on surface concentration can be manipulated by the ratio of flux to crossflow velocity, where lower CP is favored by modest flux and raised crossflow velocity. Most scale formers are more soluble at lower pH, so the injection of acid into the feed stream can provide some control. The exception to this is silica scaling which is more serious at lower pH and requires an increase in pH. If silica and calcium scalants are both present, it may be necessary to soften the feed by ion exchange to remove the calcium and then raise the pH. Scale

2

formation can also be controlled by the use of antiscalant chemicals. These are proprietary materials, but include various molecular weight polycarboxylates and polyacrylates, usually dosed at a few ppm level. Care needs to be taken with antiscalant selection as some tend to promote biofouling.

Inorganic Scaling

References Fane AG, Chong TH, Le-Clech P (2009) Fouling in membrane processes, Chapter 6. In: Drioli E, Giorno L (eds) Membrane operations, innovative separations and transformations. Wiley –VCH. ISBN 978-3-52732038-7

I

Ion Exchange Membrane for Fuel Cells Antonino Salvatore Arico CNR-ITAE Institute, Messina, Italy

Low-temperature fuel cells can be equipped with a proton or anion exchange membrane in alternative to liquid electrolytes (Arico` et al. 2001). The core of a polymer electrolyte fuel cell (PEMFC) is the ion exchange membrane. The electrodes (anode and cathode) are in intimate contact with the membrane faces. The membrane determines the fuel cell resistance and the fuel permeation rate, and it influences the reaction rate. It is well known that the use of non-noble metal catalysts is possible in the presence of alkaline electrolytes. Protons conducting electrolytes have been preferred to alkaline electrolytes for several decades for practical reasons, e.g., to avoid carbonation. The standard electrolyte membrane is usually a perfluorosulfonic acid membrane such as Nafion. Most of the electrolytes alternative to Nafion both proton conducting and alkaline type, e.g., hydrocarbon type, are significantly cheaper, and in some cases, they are also characterized by smaller hydrogen and methanol crossover. However, lifetime characteristics similar to those shown by Nafion-type membranes in fuel cells have not yet been demonstrated for the alternative membranes. Concerning with the conductivity, only recently, membrane alternative to # Springer-Verlag Berlin Heidelberg 2012 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_326-2

Nafion type have shown similar levels of performance. One critical aspect is related to the fact that the presence of water is a requirement of low-temperature PEMFCs and DMFCs for the occurrence of the electrochemical reactions and to promote ion conductivity (water-assisted conductivity mechanism). High ionic conductivities are often associated to the presence of large water uptake by the membrane, but this property often causes poor mechanical characteristics such as large swelling and relevant crossover (especially methanol). Phosphoric acid-doped polybenzoimidazole membranes use a Grotthus mechanism of proton transport and do not require water (Wang et al.1995). These membranes operate at about 180–200 C. Whereas composite perfluorosulfonic acid membranes or sulfonated hydrocarbon including inorganic fillers such as silica rely on the water-assisted mechanism, they can operate up to 145 C, under particular conditions (3 bar abs pressure), due to the enhanced water retention of the filler (Arico` et al. 2003). Composite recast Nafion® membranes containing inorganic fillers have been employed in high temperature (150 C) direct alcohol (Arico` et al. 1998) and H2-air fuel cells (Watanabe et al.1996). These composite membranes were originally developed for reduced humidification operation in polymer electrolyte fuel cells (Watanabe et al.1996) due to the enhanced water retention inside the membrane by the effect of the inorganic filler (Arico` et al. 1998). A further advantage of composite

2

membranes relies in the barrier effect given by the inorganic filler for methanol crossover (Ren et al. 1996).

References Arico` AS, Cretı` P, Antonucci PL, Antonucci V (1998) Comparison of ethanol and methanol oxidation in a liquid-feed solid polymer electrolyte fuel cell at high temperature. Electrochem Solid-State Lett 1:66–68 Arico` AS, Srinivasan S, Antonucci V (2001) DMFCs: from fundamental aspects to technology development. Fuel Cells 1:133

Ion Exchange Membrane for Fuel Cells Arico` AS, Baglio V, Di Blasi A, Creti P, Antonucci PL, Antonucci V (2003) Influence of the acid–base characteristics of inorganic fillers on the high temperature performance of composite membranes in direct methanol fuel cells. Solid State Ionics 161:251–265 Ren X, Wilson MS, Gottesfeld S (1996) High performance direct methanol polymer electrolyte fuel cells. J Electrochem Soc 143:L12 Wang J, Wasmus S, Savinell RF (1995) Evaluation of ethanol, 1-propanol, and 2-propanol in a direct oxidation polymer-electrolyte fuel cell. J ElectrochemSoc 142:4218 Watanabe M, Uchida H, Seki Y, Emori M, Stonehart P (1996) Self-humidifying polymer electrolyte membranes for fuel cells. J Electrochem Soc 143:3847

I

Irreversible Flux Decline Tony Fane School of Chemical Engineering, The University of New South Wales, UNSW, Sydney, NSW, Australia Singapore Membrane Technology Centre, Nanyang Technological University, Singapore, Singapore

Membranes are either operated at constant pressure (P) or constant flux (J); the former is more typical of lab scale studies and the latter of commercial operation. Fouling leads to flux decline under constant pressure or a rise in the required transmembrane pressure DP for constant flux. These trends are illustrated in Fig. 1. It should be noted that fouling under constant pressure becomes self-limiting; the lower the flux, the slower the fouling. For constant flux operation, the fouling can be self-accelerating, often leading to a sudden rise in TMP (DP) as shown for case (ii) in Fig. 1b. This “TMP jump” could be associated with a critical consolidation of the fouling layer and rapid rise in fouling resistance. Irreversible flux decline is the drop in flux due to irreversible fouling. The effect of irreversible fouling can also be expressed as irreversible permeability

# Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_327-1

decline (given by the change in J/DP), and this can be used for both constant pressure and constant flux operation. However, it is important to note that fouling is flux driven. As a result it is more appropriate to compare membranes at the same initial flux for constant pressure tests. Figure 1a illustrates two membranes, M1 and M2, where M1 has an initially higher permeability and starts at a higher flux. Fluxes for M1 and M2 both decline due to the increasing fouling resistance (see ▶ Irreversible Fouling Resistance), and both membranes asymptote to similar declined fluxes. This is due to the eventual dominance of the fouling resistance (RF) over the membrane resistance (Rm) (see Eqs. 1 and 2 in ▶ Irreversible Fouling Resistance). If membranes M1 and M2 were tested at the same constant flux, it would be easier to identify if one was intrinsically less prone to fouling.

References Fane AG, Chong TH, Le-Clech P (2009) Fouling in membrane processes, Chapter 6. In: Drioli E, Giorno L (eds) Membrane operations, innovative separations and transformations. Wiley-VCH, Weinheim. ISBN 978-3-527-32038-7

2

Irreversible Flux Decline

a

b M1 TMP

FLUX

M2

(i) (ii)

Time

Time

Flux vs time for Constant Pressure operation

Pressure vs time for Constant Flux operation

Irreversible Flux Decline, Fig. 1 Constant pressure and constant flux operation: fouling trends

I

Irreversible Fouling Tony Fane School of Chemical Engineering, UNSW, The University of New South Wales, Sydney, NSW, Australia Singapore Membrane Technology Centre, Nanyang Technological University, Singapore

Membrane fouling is the accumulation of material on the surface of or within the membrane structure. The foulant may provide a cake or surface deposit layer, and if the membrane is microporous, the foulant could cause pore restriction/closure or pore plugging. These fouling mechanisms are depicted in Fig. 1. Membrane fouling is prevalent in all the liquid-phase membrane processes. Fouling species can be organic macromolecules (proteins, polysaccharides, etc.), inorganics (scale precipitates), and colloidal or biological (biofilms). Fouling is linked to, but different from, concentration polarization (CP). CP is a buildup of retained species (solutes) adjacent to the membrane surface and can be depicted as a boundary layer concentration profile based on a balance between convection (flux driven) and back transport (diffusion). CP increases with flux but is totally reversible when flux is ceased; fouling is not totally reversible. In many situations fouling has both reversible and irreversible components. The reversible # Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_328-1

component may be removed by hydrodynamic means, such as backwash or increased shear at the membrane surface. This type of foulant is usually loosely deposited or bound on the membrane and subject to removal by raised liquid velocities. However, it is not removed by simply dropping flux to zero. The irreversible fouling is material that is tightly held by the membrane. It could be colloids plugging pores, macrosolutes bound by adsorption, insoluble salts spreading as crystals, and biofilms comprising bacteria held in a matrix of extracellular polymeric substances (EPS). Irreversible fouling can usually be mitigated by feed pretreatment, suitable hydrodynamics, and careful membrane selection. For example, pretreatment could remove colloids or bacteria, or involve the addition of antiscalants. Hydrodynamic control could match crossflow velocity to the imposed flux to operate below the critical flux, where the critical flux is the flux below which fouling is negligible. Membrane selection would consider pore size, surface charge, hydrophobicity/hydrophilicity, etc. Irreversible fouling is usually partially removable by cleaning regimes, which could range from daily to monthly, with chemical or physical agents, depending on the situation. However, the gradual buildup of residual irreversible fouling, over multiple cleaning cycles, may eventually reach a critical level, and membrane replacement is required.

2

Irreversible Fouling

1 Pore Closure

2 Pore Plugging

3 ‘Cake’ (surface) deposition

Irreversible Fouling, Fig. 1 Membrane fouling mechanisms

For more information on membrane fouling, see “▶ Irreversible Flux Decline” and “▶ Irreversible Fouling Resistance.”

References Fane AG, Chong TH, Le-Clech P (2009) Fouling in membrane processes, Chapter 6. In: Drioli E, Giorno L (eds) Membrane operations, innovative separations and transformations, Wiley-VCH Weinheim. ISBN: 978-3-527-32038-7

I

Irreversible Fouling Resistance Tony Fane School of Chemical Engineering, The University of New South Wales, Sydney, NSW, Australia Singapore Membrane Technology Centre, Nanyang Technological University, Singapore, Singapore

Flux can be related to the driving force (DPDP) and the overall resistance, RT, by, DP DP J¼ mðRT Þ

J¼ (1)

The osmotic pressure term (DP) can be ignored for non-osmotic operation (i.e., MF applications). The resistance components are assumed to be in series, so, RT ¼ Rm þ RF

(2)

where Rm is the membrane resistance and RF is the fouling resistance, which may comprise a reversible (Rrr) and irreversible (Rir) component, i.e., RF ¼ Rrr þ Rir

RT – Rm gives RF. A water flush should remove Rrr, leaving resistances Rm + Rir. Chemical cleaning should remove Rir, but usually a residual DRir remains as explained in irreversible fouling. For constant flux, the DP data can be used to estimate the various resistances. For processes like reverse osmosis, the osmotic pressure term in Eq. 1 is important. The magnitude of DP is increased in practice by a factor M, the polarization modulus; this is due to concentration polarization. Equation 1 becomes:

(3)

For a given DP, the flux vs. time data gives the various resistances. The initial clean water flux gives Rm, the final “fouled” flux gives RT, and # Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_329-1

DP MDP mðRT Þ

(4)

M is the ratio of solute concentration at the membrane surface to the solute in the bulk, and in a well-operated RO desalination process, without fouling, M = 1.1–1.2. The polarization modulus, M, can be increased if there is a “cake” layer on the RO membrane surface. The layer provides an unstirred region where the solute (salt) concentration builds up. Figure 1 depicts concentration polarization for a clean and cake fouled RO membrane. The increased polarization (Fig. 1b) is “cake-enhanced” concentration polarization (CECP), and the increased osmotic pressure is the “cakeenhanced” osmotic pressure (CEOP) (Hoek and Elimelech 2003). Both fouling resistance and CEOP contribute to loss of flux (for fixed DP) or increased DP for fixed flux.

2 Irreversible Fouling Resistance, Fig. 1 (a) Concentration polarization – clean membrane. (b) Cakeenhanced concentration polarization – fouled membrane

Irreversible Fouling Resistance

a

CB

b

CB

Shear Membrane

Cake CW1 CP1

CW2 CP2

References

Further Reading

Hoek EMV, Elimelech M (2003) Cake enhanced concentration polarization: a new fouling mechanism for saltrejecting membranes. Environ Sci Technol 37:5581–5588

Fane AG, Chong TH, Le-Clech P (2009) Fouling in membrane processes, chapter 6. In: Drioli E, Giorno L (eds) Membrane operations, innovative separations and transformations. Wiley –VCH, Weinheim. ISBN 978-3-527-32038-7

H

Hybrid Regenerated Cellulose/ Loaded Lipid Nanoparticle Membranes: Preparation and Characterization Juana Benavente1 and Juan Manuel Lo´pezRomero2 1 Departamento de Fisica Aplicada I, Universidad de Malaga, Facultad de Ciencias, Malaga, Spain 2 Departamento de Quimica Organica, Universidad de Malaga, Facultad de Ciencias, Malaga, Spain

New membrane systems related to medical applications (drug-release devices, mimetic membranes, or patches) are receiving great attention (M€ uller et al. 2000). Among them, lipid nanoparticles (LNPs) prepared using biocompatible components and with tunable properties are of significant interest (Gupta and Kompella 2006; Huynh et al. 2009); although, their use is still limited due to stability problems during contact with biological fluids, storage, or administration (Korting and Scha¨fer–Korting 2010). To overcome these limitations, polymeric nanosphere gels were proposed for topical delivery of lipophilic molecules (Martins et al. 2007). In this context, the inclusion of functionalized lipid nanoparticles (FLNPs) in support to biocompatible membranes (such as regenerated cellulose membranes) offers an attractive route for controlled release of pharmacologic agents. # Springer-Verlag Berlin Heidelberg 2014 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_338-3

The LNPs used in this study were prepared by the ultrasound method (Feng and Huang 2001) using L-a-phosphatidylcholine and Tween ® 80 as surfactants and glyceryl tristearate as the main lipid component, while sunscreen DHB (2,4-dihydroxybenzophenone) was the active organic component. The hybrid membrane was obtained by embedding the DHBLNPs in a dense highly hydrophilic regenerated cellulose (RC) support by immersion in a water dispersion of DHBLNPs (membrane RC/DHBLNPs). The incorporation of the loaded LNPs was characterized by AFM, brilliant field microscopy (Fig. 1a), TEM (Fig. 1b), and Raman spectroscopy (Vázquez et al. 2011; Hierrezuelo et al. 2012). Reductions in material characteristic parameters (conductivity and dielectric constant) for dry RC/DHBLNP samples when compared with the original RC and changes associated to thermal effects were also obtained. The stability of the hybrid membrane as a result of both contact time with NaCl solutions and osmotic pressure gradients was established by comparing diffusional permeability (Ps) for original RC and RC/LNP membranes (Vázquez et al. 2011; Hierrezuelo et al. 2012). The presence of the DHBLNPs reduces in approximately 20 % Ps values for the whole interval of feed concentrations studied (0.001 Cf(M) 0.4), and a constancy in the NaCl flow for at least 20 h was also observed. The cellulose structure also

2

Hybrid Regenerated Cellulose/Loaded Lipid Nanoparticle Membranes

Hybrid Regenerated Cellulose/Loaded Lipid Nanoparticle Membranes: Preparation and Characterization, Fig. 1 Hybrid membrane micrographs: (a) brilliant filed microscopy (b) TEM

reduces the DHB delivery from the loaded LNPs, being the value for the hybrid membrane 5 % of that for the DHBLNPs (after 30 min).

References Feng SS, Huang G (2001) Effects of emulsifiers on the controlled release of paclitaxel (Taxol ®) from nanospheres of biodegradable polymers. J Control Release 71:53–69 Gupta RB, Kompella UB (2006) Nanoparticle technology for drug delivery. Taylor and Francis/CRC Press, New York Hierrezuelo J, Benavente J, Lo´pez–Romero JM, Martı´nez de Yuso MV, Rodrı´guez–Castello´n E (2012) Preparation, chemical and electrical characterizations of lipid nanoparticles loaded with dihydroxybenzophenone. Med Chem 8:541–548

Huynh NT, Passirani C, Saulnier P, Benoit JP (2009) Lipid nanocapsules: a new platform for nanomedicine. Int J Pharm 379:201–219 Korting HC, Scha¨fer–Korting M (2010) Carriers in the topical treatment of skin disease. Handb Exp Pharmacol 197:435–468 Martins S, Sarmento B, Ferreira DC, Souto EB (2007) Lipid–based colloidal carriers for peptide and protein delivery–liposomes versus lipid nanoparticles. Int J Nanomedicine 2:595–607 M€ uller RH, Ma¨der K, Gohla S (2000) Solid lipid nanoparticles for controlled drug delivery – a review of the state of the art. Eur J Pharm Biopharm 50:161–177 Vázquez MI, Peláez L, Benavente J, Lo´pez-Romero JM, Rico R, Hierrezuelo J, Guille´n E, Lo´pez-Ramı´rez MR (2011) Functionalized lipid nanoparticles-cellophane hybrid films for molecular delivery. J Pharm Sci 100:4815–4822

L

Liquid Membrane Separation Vladimir S. Kislik Campus Givat Ram, Casali Institute of Applied Chemistry, The Hebrew University of Jerusalem, Jerusalem, Israel

Liquid membrane separation combines the solvent extraction and stripping processes (re-extraction) in a single step. This entry has the objective of introducing the reader to the basic definitions of the liquid membrane field, with classification. The term liquid membrane transport includes processes incorporating liquid-liquid extraction (LLX) and membrane separation in one continuously operating device. It utilizes an extracting reagent solution, immiscible with water, stagnant or flowing between two aqueous solutions (or gases), the source or feed, and receiving or strip phases. In most cases, the source and receiving phases are aqueous, and the membrane is organic, but the reverse configuration can also be used. A polymeric or inorganic microporous support (membrane) may be used as bearer (as in SLM) or barrier (as in BLM technologies) or not used, as in ELM and layered BLM (see respective entries: ▶ SLM, ▶ ELM, ▶ BLM). The great potential for energy saving, low capital and operating cost, and the possibility to use expensive extractants, due to the small amounts of the membrane phase, make LMs an # Springer-Verlag Berlin Heidelberg 2012 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_340-1

area deserving special attention. Liquid membrane systems are being studied extensively by researchers in such fields as analytical, inorganic, and organic chemistry; chemical engineering, biotechnology, and biomedical engineering; and wastewater treatment. Research and development activities within these disciplines involve diverse applications of liquid membrane technology, such as gas separations, recovery of valued or toxic metals, removal of organic compounds, development of sensing devices, recovery of fermentation products, and some other biological systems. The general properties of liquid membrane systems have been a subject of extensive theoretical and experimental studies. Some general characteristics of LM processes are as follows: 1. Liquid membrane separation is a rate process, and the separation occurs due to a chemical potential gradient, not by equilibrium between phases. 2. LM is defined based on its function, not the material used in fabrication. Permeation is a general term for the concentration-driven membrane-based mass transport process. Differences in the permeability produce a separation between solutes at constant driving force. Because the diffusion coefficients in liquids are typically orders of magnitude higher than in polymers, a larger flux can be obtained with liquid membranes. Application of

2 Liquid Membrane Separation, Fig. 1 Three configurations of liquid membrane systems: bulk (BLM), supported (immobilized) (SLM or ILM), and emulsion (ELM). F is the source or feed phase, E is the liquid membrane, and R is the receiving phase

Liquid Membrane Separation

BLM

F

E

R Porous Support

Porous Support

SLM E

F

R Porous Support

ELM E

F

a pressure difference, an electric field, or temperature considerably intensifies the process. There are several different directions in LM separation classifications: according to module design configurations (see ▶ SLM, ▶ ELM, ▶ BLM entries), according to transport mechanisms (see ▶ LM Transport Mechanisms), according to applications, according to carrier type, and according to membrane support type. Below, these types of classifications are described and discussed briefly. According to configuration definition, three groups of liquid membranes are usually considered (see Fig. 1): bulk (BLM), supported or immobilized (SLM or ILM), and emulsion (ELM) liquid membrane transport. Some authors add to these definitions polymeric inclusion membranes, gel membranes, and dual-module hollow-fiber membranes, but, to my opinion, the

R

F

first two types are the modifications of the SLM, and the third is the modification of BLM. According to the transport mechanisms, the LM techniques may be divided into simple transport, facilitated or carrier-mediated transport, coupled counter- or cotransport, and active transport. According to applications, the LM techniques may be divided into (1) metal separationconcentration, (2) biotechnological product recovery-separation, (3) pharmaceutical product recovery-separation, (4) organic compound separation and organic pollutant recovery from wastewaters, (5) gas separations, (6) fermentation or enzymatic conversion-recovery-separation (bioreactors), (7) analytical applications, and (8) wastewater treatment including biodegradationseparation techniques. Classification according to carrier type is as follows: (1) water-immiscible, organic carriers,

Liquid Membrane Separation

(2) water-soluble polymers, (3) electrostatic, ion-exchange carriers, and (4) neutral, but polarizable carriers. Classification according to membrane support type is as follows: (1) neutral hydrophobic, hydrophilic membranes, (2) charged (ion-exchange) membranes, (3) flat sheet, spiral module membranes, (4) hollow-fiber membranes, and (5) capillary hollow-fiber membranes.

3

Module design configurations are used as a rule as basic classification. Practically, there are a lot of opportunities for liquid membrane separation processes in many areas of industry. The most interesting developments for industrial membrane technologies are related to the possibility of integrating various membrane operations in the same industrial cycle, with overall important benefits in terms of product quality and plant compactness.

M

Macrosolute Radoslav Paulen Department of Biochemical and Chemical Engineering, Technische Universität Dortmund, Dortmund, Germany

Macrosolute represents an operational term from the field of membrane filtration which denotes solute(s) of particle size(s) larger than ones that pass through a membrane (microsolute) of a specified pore size or permeability limit. Membrane filtration may be used to treat such solution (system with macrosolute, microsolute, and solvent) to concentrate valuable macrosolute in this solution and to get rid of (or dilute) lower molecular weight impurities (Fig. 1).

# Springer-Verlag Berlin Heidelberg 2013 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_346-5

Distinction of species considered as macrosolute depends on membrane process which is used to treat the solution (Cheryan 1998). In case of: • Microfiltration – macrosolute is represented by suspended particles and bacteria bodies. • Ultrafiltration – macrosolute presents macromolecules (typically proteins) and species with larger particle sizes • Nanofiltration – dissociated acids and divalent salts, sugars, and species with larger particle sizes can be concentrated in solution. • Reverse osmosis – only water passes through the membrane, and thus the term macrosolute stands for any other species present in the solution.

2 1,50E-04 CONCENTRATION OF THE SOLUTE IN THE PERMEATE [mM/ml]

Macrosolute, Fig. 1 Comparison of cross-flow and dead-end filtration in terms of permeate concentration

Macrosolute

dead-end cross-flow with optimum surface renewal

0,00E+00 0

100

200 TIME [s]

References Cheryan M (1998) Ultrafiltration and microfiltration handbook. CRC Press, Boca Raton

300

M

Manganese Removal by Liquid Membranes Argurio Pietro Depatment of Environmental and Chemical Engineering, University of Calabria, Arcavacata di Rende (CS), Italy

Manganese is the chemical element having the symbol Mn and atomic number 25. It is an essential trace nutrient in all forms of life: it serves as a necessary constituent of metalloproteins including enzymes since it optimize enzyme and membrane transport function. Although the toxicity of manganese compounds is lower than those of other widespread metals, such as nickel and copper, both its excess and deficiency in the body can cause serious impairment of vital physiological and biochemical processes. Excessive manganese intake is most frequently associated with the so-called manganism, a rare central nervous system disorder, characterized by symptoms similar to Parkinson’s disease (weakness, monotone and slowed speech, tremor, disorientation, memory impairment, anxiety, and hallucinations). Manganese compounds, in which it has various oxidation states, are widely used in the industry. Manganese dioxide is used in dry cell batteries as a depolarizer and in glassmaking as a drying agent. Depending on their oxidation state, manganese ions have various colors and are used industrially as pigments. The permanganates of alkali # Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_347-7

and alkaline earth metals are powerful oxidizers, so that they are used as a bactericide and algaecide in water treatment and as an oxidant in organic chemical synthesis. Since manganese is used in these and other different industrial processes, its separation and/or recovery from the various liquid effluents or wastewater is of practical value. A variety of separation processes for metal ions have been developed for industrial needs, including precipitation, inorganic and polymeric adsorption, evaporation, and reverse osmosis. Such techniques produce water within international health standards but entail some drawbacks. Solvent extraction procedures have proven to be very useful in the separation and/or recovery of metal ions from aqueous media, but they involve high capital and operating costs due to large inventory of solvent, especially in the case of dilute solutions. Thus liquid membrane has been proven to be a very powerful technology. A liquid membrane (LM) is a layer of an organic phase (the so-called LM phase) that separates two aqueous solutions. An extractant (carrier), promoting the so-called facilitated transport from the donor phase (feed) to the acceptor phase (strip), may be dissolved in the organic phase (Molinari et al. 2009a, b). LM-based processes offer a technology with a lot of advantages over conventional separation techniques. The main advantages are (i) combination of extraction and stripping processes in a single stage, resulting in a low solvent consumption; (ii) uphill transport against concentration gradient; and (iii) small

2

amounts of extractant needed, resulting in the possibility to use very selective carriers, which sometimes are not very cheap. A widely used important acidic extractant in hydrometallurgy in metal separation and/or recovery is di(2-ethylhexyl)phosphoric acid (D2EHPA). It is extensively used also in the extraction and recovery of manganese from neutral and weakly acidic solutions using supported liquid membrane and emulsion liquid membrane (Mohapatra and Kanungo 1992). Some studies (Yongtao et al. 1992) considered the simultaneous extraction and concentration of cadmium and manganese from aqueous solutions obtaining recoveries of cadmium and manganese in the range of 92–100 %. Recently a novel method for Mn(II) extraction from sulfuric acid solutions has been proposed (Sadyrbaeva 2011) which involves electrodialysis and a bulk liquid membrane containing D2EHPA as the carrier. To accelerate the ion transfer through the liquid ion-exchange membrane, an electric field was applied. Operating in this way, the electrodialytic process (unselective) was coupled with the liquid membrane-based process, which provides greater selectivity and permeability with respect to the traditional solid ion-exchange membrane. Complete removal of Mn(II) from a feed solution containing 0.01 mol/L of MnSO4 and a maximum extraction degree of 88 % was obtained under optimized conditions.

Liquid Membrane as Pre-concentration Technique Manganese could be accumulated in tissues and body fluids, like blood serum and urine.

Manganese Removal by Liquid Membranes

Considering the very low concentration of trace elements (manganese in our case) in these fluids, a pre-concentration step before analysis is required. According to results from a recent study (Soko et al. 2003), the D2EHPA-based supported liquid membrane technique can be used to extract and pre-concentrate Mn(II) from water, milk, and blood serum.

References Mohapatra R, Kanungo SB (1992) Kinetics of Mn (II) transport from aqueous sulphate solution through a supported liquid membrane containing di (2-ethylhexyl) phosphoric acid in kerosene. Sep Sci Technol 27:1759–1773 Molinari R, Argurio P, Poerio T (2009a) Studies of various solid membrane supports to prepare stable sandwich liquid membranes and testing copper(II) removal from aqueous media. Sep Purif Technol 70:166–172 Molinari R, Argurio P, Poerio T (2009b) Flux enhancement of stagnant sandwich compared to supported liquid membrane systems in the removal of Gemfibrozil from waters. J Membr Sci 340:26–34 Sadyrbaeva TZ (2011) Hybrid liquid membrane – electrodialysis process for extraction of manganese(II). Desalination 274:220–225 Soko L, Chimuka L, Cukrowska E, Pole S (2003) Extraction and preconcentration of manganese(II) from biological fluids (water, milk and blood serum) using supported liquid membrane and membrane probe methods. Anal Chim Acta 485:25–35 Yongtao L, Aixia W, Van Loon JC, Barefoot RR (1992) Extraction and enrichment of cadmium and manganese from aqueous solution using a liquid membrane. Talanta 39:1337–1341

M

Mass Intensity Adele Brunetti National Research Council, Institute for Membrane Technology (ITM-CNR), The University of Calabria, Rende, CS, Italy

Membrane operations are well known for their modularity, compactness, and flexibility; therefore, they can be considered as new operations developed in the logic of process intensification. Recently, new metrics for comparing membrane performance with those of conventional units have been introduced (Curzon et al. 2001; Brunetti et al. 2011, 2014). These new metrics take into account the size, the weight, the flexibility, the yield, and the modularity of the plants. They are useful for having an immediate indication of the eventual gain that a membrane operation can offer with respect to a traditional one. In this sense, they are useful also for selecting the proper separation technology for a specific process. Among them, mass intensity is a measure of the exploitation of the raw material with respect to final production or recovery of the desired product, and it can be intended as a measure of the efficiency of the process. In order to compare different systems, it can have different definitions: it can be calculated taking into account the product and its fraction # Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_348-1

recovered (Eq. 1) as valuable product with respect to the total mass fed. It can also be calculated as the ratio between the amount of product and its fraction recovered with respect to the steam and cooling water necessary to carry out the process (Eq. 2). In the ideal situation, the mass intensity would approach 1 which means that all the inlet mass has been converted or recovered as valuable product. Mass Intensity_1 ¼ Total inlet mass Mass of desired product recovered

(1)

Mass Intensity_2 ¼ Steam and cooling water required Mass of desired product recovered

(2)

References Brunetti A, Barbieri G, Drioli E (2011) New metrics in membrane gas separation, chapter 20. In: Drioli E, BarbCieri G (eds) Membrane engineering for the treatment of gases, vol 2. The Royal Society of Chemistry, Cambridge, UK, pp 279–301. ISBN 978-1-84973-239-0 Brunetti A, Drioli E, Barbieri G (2014) Energy and mass intensities in hydrogen upgrading by a membrane reactor. Fuel Process Technol 118:278–286 Curzons AD, Constable DJC, Mortimera DN, Cunningham VL (2001) So you think your process is green, how do you know?—Using principles of sustainability to determine what is green–a corporate perspective. Green Chem 3:1–6

M

Matrimid ® Membranes Ahmad Fauzi Ismail1 and Juhana Jaafar2 1 Faculty of Petroleum and Renewable Energy Engineering, Universiti Teknologi Malaysia, Advanced Membrane Technology Research Centre (AMTEC), Johor Bahru, Johor, Malaysia 2 Faculty of Chemical and Energy Engineering, Universiti Teknologi Malaysia, Advanced Membrane Technology Research Centre (AMTEC), Johor Bahru, Johor, Malaysia

Matrimid membrane can be developed from different types of commercially available polyimide resins such as Matrimid 5218 (3,30 ,4,40 -benzophenonetetracarboxylic dianhydride and diaminophenylindane), Kapton, Matrimid 5292A (4,4-bismaleimidodiphenylmethane), Matrimid 5292B, Matrimid P84, and Matrimid 9725 (micropulverized version Matrimid 5218) (www.ciba.com; www.huntsman.com). The employment of commercially available polyimide resin was found to be more practical from the economic point of view due to the high cost of synthesized polyimides on the laboratory scale (Xiao et al. 2005). The Matrimid membranes are commonly prepared via solvent evaporation (Sridhar et al. 2007; Mosleh et al. 2012) and phase inversion (Basu et al. 2010) techniques. The asymmetric (Nistor et al. 2008; Basu et al. 2010) or dense # Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_349-6

(Shishatskiy et al. 2006) Matrimid membrane can be in flat sheet (Shishatskiy et al. 2006; Aziz and Ismail 2010) or hollow fiber (Dong et al. 2011; Ding et al. 2008; Jiang et al. 2004) form. Matrimid especially the Matrimid 5218, an extremely popular polyimide, has been extensively studied for gas separation, pervaporation processes, and carbon membrane due to the combination of relatively high gas permeability coefficients and separation factors as well as excellent mechanical properties, high solubility in non-hazard organic solvents, and high glass transition temperature in comparison to polycarbonate, polysulfone, and other materials (Wind 2002; Nistor et al. 2008). The typical permeation properties of Matrimid 5218 are tabulated in Table 1. The overall performance (in terms of permeability and selectivity) of Matrimid membrane in gas separation processes can be diverse as reported in the open literatures which was due to the several factors such as: (i) Operating conditions of gas permeation test (ii) Effective (skin layer) membrane thickness (the thinner the selective skin layer is, the better the gas flux will be) (iii) Membrane morphology as a consequence of the different conditions of phase inversions of membrane preparation technique (iv) Properties of the gas tested The weakness of Matrimid membrane is always pronounced by the membrane transport

Matrimid ® Membranes

2

Matrimid ® Membranes, Table 1 Typical permeation properties of Matrimid 5218 (David et al. 2011) Permeability of pure gases (Barrera) Po2 1.32–1.34b

Ideal selectivity O2/N2 7.2b

PN2 0.185b

1 Barrer = 1010 cm3 (STP) cm cm2 S1 cmHg1 = 7.5 1018 m3 (STP) m m2 S1 Pa1, Pure gas Permeation measurement for O2 and N2 at 3447.3785 kPa, 25 C

a

b

O

O O

C

CH3

C C

N

N C O

C

n H3C

CH3

O

Matrimid ® Membranes, Fig. 1 Chemical structure of Matrimid

properties that are strongly affected by the gas separation operating conditions such as pressure and temperature that are likely relied to the changes in density, and consequently in the free volume, of the polymer itself (Rowe et al. 2009; David et al. 2011). Hence, the presence of minor components in the feed flow, such as impurities or water vapor, can strongly affect the gas transport mechanism, thus altering the separation performances (Chen et al. 2011). Moreover, its glassy nature may lead to plasticization phenomena at high fugacity of gases and vapors and to aging phenomena that decrease the performances upon time (Tin et al. 2003). These shortcomings limit the Matrimid membrane application particularly in gas separation processes. Due to this awareness, numbers of modification have been reported by researchers to improve the Matrimid membrane properties in order to fulfill the gas separation performance requirement particularly to surpass the Robeson boundary line (Basu et al. 2010). The idea to modify the Matrimid membrane was initiated by the good processability of the polymer itself owing to its carbonyl functional groups (see Fig. 1) that may lead to the donation of electrons to form good interaction with certain electron acceptors (Chung et al. 2003). The

modifications were conducted on the Matrimid membrane including the incorporation of mesoporous silica molecular sieve (Kenneth et al. 2002), bromination (Guiver et al. 2002; Xiao et al. 2005), blending with other polymers (Chung et al. 2003), thermal treatment (Krol et al. 2001), and chemical cross-linking (Tin et al. 2003). Due to the excellent properties shown by Matrimid membrane, which is desirable for gas separation process, therefore it is worth to accelerate relevant modifications on this promising membrane for better separation performance.

References Aziz F, Ismail AF (2010) Preparation and characterization of cross-linked Matrimid ® membranes using paraphenylenediamine for O2/N2 separation. Sep Purif Technol 73:421–428 Basu S, Cano-Odena A, Vankelecom IFJ (2010) Asymmetric Matrimid ®/[Cu3(BTC)2] mixed-matrix membranes for gas separations. J Membr Sci 362:478–487 Chen QG, Scholes CA, Qiao GG, Kentish SE (2011) Water vapor permeation in polyimide membranes. J Membr Sci 379:479–487 Chung TS, Chan SS, Wang R, Lu Z, He C (2003) Characterization of Permeability and Sorption in

Matrimid ® Membranes Matrimid/C60 Mixed Matrix Membranes. J Membr Sci 211:91–99 David OC, Gorri D, Urtiaga A, Ortiz I (2011) Mixed gas separation study for the hydrogen recovery from H2/ CO/N2/CO2 post combustion mixtures using a Matrimid membrane. J Membr Sci 378:359–368 Ding X, Cao Y, Zhao H, Wang L, Yuan (2008) Fabrication of high performance Matrimid/polysulfone dual-layer hollow fiber membranes for O2/N2 separation. J Membr Sci 323:352–361 Dong G, Li H, Chen V (2011) Plasticization mechanisms and effects of thermal annealing of Matrimid hollow fiber membranes for CO2 removal. J Membr Sci 369:206–220 Guiver MD, Thi NL, Robertson GP (2002) Composite Gas Separation Membranes. U S Patent 20020062737. Jiang L, Chung TS, Li DF, Cao C, Kulprathipanj S (2004) Fabrication of Matrimid/polyethersulfone dual-layer hollow fiber membranes for gas separation. J Membr Sci 240:91–103 Kenneth J, Balkus Jr, Cattanach K, Musselman IH, Ferraris JP (2002) Selective matrimid membranes containing mesoporous molecular sieves. MRS Proc. doi:10.1557/PROC-752-AA4.3. Accessed 2011 Krol JJ, Boerrigter M, Koops GH (2001) Polyimide hollow fiber gas separation membranes: preparation and the suppression of plasticization in propane/propylene environments. J Membr Sci 184:275–286 Mosleh S, Khosravi T, Bakhtiari O, Mohammadi T (2012) Zeolite filled polyimide membranes for dehydration of

3 isopropanol through pervaporation process. Chem Eng Res Des 90:433–441 Nistor C, Shishatskiy S, Popa M, Nunes SP (2008) Composite membranes with cross-linked matrimid selective layer for gas separation. Environ Eng Manag J7:653–659 Rowe BR, Freeman BD, Paul DR (2009) Physical aging of ultrathin glassy polymer films tracked by gas permeability. Polymer 50:5565–5575 Shishatskiy S, Nistor C, Popa M, Nunes SP, Peinemann KV (2006) Comparison of asymmetric and thin-film composite membranes having Matrimid 5218 selective layer. Desalination 199:193–194 Sridhar S, Veerapur RS, Patil MB, Gudasi KB, Aminabhavi TM (2007) Matrimid polyimide membranes for the separation of carbon dioxide from methane. J Appl Polym Sci 106:1585–1594 Tin PS, Liu Y, Wang R, Liu SL, Pramoda KP (2003) Effects of cross-linking modification on gas separation performance of Matrimid membranes. J Membr Sci 225:77–90 Wind JD (2002) Improving polyimide membrane resistance to carbon dioxide plasticization in natural gas separations. PhD Thesis, The University of Texas at Austin www.ciba.com www.huntsman.com Xiao Y, Dai Y, Chung T-S, Guiver MD (2005) Effects of brominating matrimid polyimide on the physical and gas transport properties of derived carbon membranes. Macromolecules 38:10042–10049

M

Membrane Artificial Organs Loredana De Bartolo ITM-CNR, National Research Council of Italy, Rende, CS, Italy

Synonyms Artificial functional organ Membrane artificial organs are membrane devices that replace the function of natural organs like artificial kidneys/dialyzers and artificial lungs/blood oxygenators for the purpose of restoring the specific organic functions. Polymeric semipermeable membranes and membrane processes play a pivotal role in replacement therapy for acute and chronic organ failure and in the management of immunological disease (De Bartolo and Drioli 1998). In these devices semipermeable membranes act as selective barriers for the removal of endogenous and exogenous toxins from patient’s blood (hemodialysis, hemofiltration, etc.) or for gas exchange with blood (blood oxygenation) (Kawakami 2008). Hemodialysis is an extracorporeal treatment of patients affected by kidney failure. The treatment is intermittent, generally three times weekly for periods between 3 and 5 h depending upon the patient clinical requirements. In hemodialyzer, selective membranes play the same function of kidneys allowing the # Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_351-4

extracorporeal removal of waste products such as creatinine and urea and free water from the blood when the kidneys are in a state of renal failure. It involves diffusion of small solutes across the semipermeable membrane retaining protein and molecules with high molecular weight. Membranes used in the hemodialyzer device have been engineered in order to have the similar structure of the renal glomerular membrane to ensure a similar performance. Membrane hollow fiber and glomerular capillary employ a cylindrical cross section in order to minimize perimembrane boundary layers and to maximize transport efficiency, have same ratio of wall thickness/cross-sectional diameter, and are composed of linear hydrophilic polymers that rely on van der Waals forces and islands of crystallinity to retain their integrity. Difference is scale: the glomerule has a diameter of 4–8 mm, and synthetic hollow fiber membranes have diameters in the range of 200 mm. For example, the structure of polyethersulfone membrane that is widely used in hemodialyzer is similar to glomerular membrane. Both have asymmetric structure with a selective skin layer supported by a sponge layer (Fig. 1). This structure confers to the membrane an optimal selectivity and high mechanical resistance. The first hollow fiber hemodialyzers made of unmodified cellulose were used clinically in the 1960s. Since the first cellulose membranes, significant efforts have been focused on the development of more biocompatible and selective

2

Membrane Artificial Organs

Membrane Artificial Organs, Fig. 1 Comparison of hemodialysis polyethersulfone membrane (a) with glomerular membrane (b)

membranes with respect to the clearance of “middle” molecules like b2-microglobulin that is known to be associated with many dialysisrelated disorders. A variety of synthetic polymers and modified cellulosic materials including polysulfone (PSF), polyethersulfone (PES), polyamide (PA), polyacrylonitrile (PAN), cellulose triacetate (CTA), and hemophan (HP) have been used for preparing hemodialysis membranes (Humes et al. 2006; Su et al. 2008). Extracorporeal blood oxygenators are used to oxygenate the blood during open-heart surgery. Today, the vast majority of these blood oxygenators use hydrophobic microporous hollow fiber membranes to separate the blood and gas phases. Oxygen diffuses from the gas phase through the gas-filled membrane pores into the blood. Oxygen in the blood plasma binds to hemoglobin present in the red blood cells. Consequently the rate of oxygen transfer is enhanced compared to the oxygenation of water where the oxygen does not react in the liquid phase (Lewandowski 2000). Membrane oxygenators in current use utilize microporous, silicon, or polypropylene membranes. Currently, there are three principal types of oxygenator: (a) Plaque oxygenators are built with microporous membranes of expanded polypropylene, folded in a Z shape. In these apparatuses, blood and gas flow in opposite sides of the

membrane. (b) Spiral oxygenators utilized silicon membranes that are rolled around a central axis. (c) Hollow fiber oxygenators are manufactured with microporous polypropylene membranes, constituted by hollow fibers or capillary. This is the most common type of oxygenator used currently. The most common membranes for blood oxygenator have been engineered in order to have morphology close to the pulmonary alveoli. For example oxygenator polypropylene membranes have massive reticulated surfaces deployed as fine, porous, open-cell foams, which remind the saccular form of the alveoli and the very thin alveoli septa (Fig. 2). In the natural lung, blood flows through fine capillaries in the alveoli and exchanges O2 and CO2 across a barrier made up of capillary endothelial cells; in oxygenators blood flows on the outside of a large bore fiber, with diameter of 300 mm, and exchanges respiratory gases across a meniscus that forms across the membrane pore at the interface of blood and gas. The diffusion of these gases inside the oxygenator depends on the type of material of the membrane and on its thickness and porosity, but it is also influenced by the thickness of the layer of blood in contact with the membrane and by the characteristics of the blood flow (Wickramasinghe et al. 2005).

Membrane Artificial Organs

3

Membrane Artificial Organs, Fig. 2 Comparison of oxygenator polypropylene membrane (a) with pulmonary alveoli (b)

References De Bartolo L, Drioli E (1998) Membranes in artificial organs. In: Haris PI, Chapman D (eds) Biomedical and health research, vol 16, New Biomedical Materials – Basic and Applied Studies. IOS Press, Amsterdam, pp 167–181 Humes HD, Fissell WH, Tiranathanagul K (2006) The future of hemodialysis membranes. Kidney Int 69:1115–1119 Kawakami H (2008) Polymeric membrane materials for artificial organs. J Artif Organs 11:177–181

Lewandowski K (2000) Extracorporeal membrane oxygenation for severe acute respiratory failure. Crit Care 4:156–168 Su BH, Fu P, Li Q, Tao Y, Li Z, Zao HS, Zhao CS (2008) Evaluation of polyethersulfone highflux hemodialysis membrane in vitro and in vivo. J Mater Sci Med 19(2):745–751 Wickramasinghe SR, Han B, Garcia JD, Specht R (2005) Microporous membrane blood oxygenators. AIChE J 51:656–670

M

Membrane Bioartificial Organs Loredana De Bartolo National Research Council of Italy, Institute on Membrane Technology, ITM-CNR, Rende (CS), Italy

Membrane bioartificial organ is a membrane device containing living cells, which is implanted or integrated into a human to replace for short or long term a natural organ. This device is realized as alternative to organ transplantation or as bridge for supporting patients until the organ transplantation. Every year, many patients die while waiting for an organ transplant. Organs such as the liver, kidney, and lungs are often in very high demand by patients with severe illnesses. Historically, the number of available organ donors has been insufficient to meet the needs of every patient. For a traditional organ transplant to succeed, the donor and patient must be a close biological match. Even when a replacement organ is available, the immune system of the recipient may reject the transplant. Membrane bioartificial organs have been designed to solve this problem. Some examples are represented by membrane bioartificial liver for the treatment of acute and chronic liver disease, bioartificial pancreas as alternative approach to exogenous insulin administration in the case of insulin-dependent diabetes mellitus, bioartificial kidney for people with poorly # Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_355-3

functioning kidneys, and bioartificial lung for the treatment of patients with end-stage lung disease. Membrane capsules containing dopaminesecreting cells are also being explored for the treatment of Parkinson’s disease, a progressive brain disorder characterized by a deficiency of the neurotransmitter dopamine. Immunoprotective membrane cell transplants are being investigated to treat other nervous system disorders. Other membrane bioreactors are used for the adoptive cell therapy for the treatment of malignant diseases or viral infections in the expansion of T lymphocytes (JagurGrodzinski 2006). Membrane bioartificial organs involve the design, modification, growth, and maintenance of living tissues embedded in natural or synthetic scaffolds to enable them to perform complex biochemical functions, including adaptive control and the replacement of normal living tissues. In membrane bioartificial organs, cells are compartmentalized by means of semipermeable membranes that permit the transport of nutrients and metabolites to cells and the transport of catabolites and specific metabolic products to the blood (Drioli and De Bartolo 2006). The membrane must avoid the contact between cells and patient’s blood to prevent immunological response and consequent rejection. Membranes act also as means for cell oxygenation and in the case of anchorage-dependent cells as substrata for cell attachment and culture. In these systems, cells come into contact with the membrane

2

surface. Therefore, the response of the cell behavior depends on the surface properties of the used membrane (Morelli et al. 2010). For this reason, membranes should be chosen not only on the basis of their separation properties but also on the basis of physicochemical and morphological surface properties. Membrane bioartificial organs are engineered to be used as extracorporeal devices or implantable systems. These devices can be distinguished on the basis of membrane material configuration, molecular weight cutoff or pore size, and cell culture technique. Cells can be compartmentalized in the lumen or shell of hollow fiber membranes (HFMs) or between flatsheet membranes, in a network of HF membranes, spirally wound module, encapsulated or attached to microcarriers. The most common membranes used in the membrane bioartificial organs are hydrophilic membranes with molecular weight cutoff ranging from 20,000 to 120,000 daltons, which are chosen on the basis of the molecules that have to provide to the cells and removed from them. The creation

Membrane Bioartificial Organs

of a physiological environment requires the use of membranes with specific physicochemical, morphological, and transport properties on the basis of the targeted tissue or organ (De Bartolo et al. 2012).

References De Bartolo L, Leindlein A, Hofmann D, Bader A, de Grey A, Curcio E, Drioli E (2012) Bio-hybrid organs and tissues for patient therapy: a future vision for 2030. Chem Eng Process: Proc Intensif 51:79–87 Drioli E, De Bartolo L (2006) Membrane bioreactor for cell tissues and organoids. Artif Organs 30(10):793–802 Jagur-Grodzinski J (2006) Polymer for tissue engineering, medical devices and regenerative medicine. Concise general review of recent studies Polym. Adv Technol 17:395–418 Morelli S, Salerno S, Piscioneri A, Campana C, Drioli E, De Bartolo L (2010) Membrane bioreactors for regenerative medicine: an example of the bioartificial liver. Asia Pac J Chem Eng 5(1):146–159

M

Membrane Biocompatibility Loredana De Bartolo Institute on Membrane Technology, ITM-CNR, National Research Council of Italy, Rende(CS), Italy

Synonyms Biocompatibility of membranes Membrane biocompatibility is the ability of membrane to perform its intended function without eliciting any host undesirable local or systemic effects. Membrane biocompatibility is a general term that must be subcategorized on the basis of the membrane applications in order to be able to make more narrow definitions. In the case of membranes used in the hemodialyzers, hemofilters, hemodiafilters, and blood oxygenator, the more appropriate term is blood compatibility or hemocompatibility because of the main interactions between membrane and blood. When blood comes into contact with an artificial surface, protein adsorption occurs in the first seconds. Proteins like albumin, immunoglobulin G, fibrinogen, fibronectin, factor XII, and high molecular weight kininogen are adsorbed forming a layer that supports the platelet adherence. The adsorption of thrombin may activate the clotting cascade (Ulhenbusch et al. 2004). The protein adsorption process is strongly # Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_356-3

dependent on the physicochemical (e.g., surface charge, functional groups) and morphological (e.g., porosity, roughness) surface properties of the membranes (De Bartolo et al. 2004). On the other hand, the artificial surface may activate the complement cascade through the membranebound complement factors such as C3b and iC3b. After the protein adhesion, platelet changes their morphology, release factors, and aggregate. The adsorbed proteins activate also the coagulation cascade with the fibrin and thrombus formation, which occur within several minutes. The coagulation pathway consists of a series of reactions, each requiring the formation of a surfacebound enzyme complex. In these reactions, inactive precursor proteins are transformed into active protease. Several studies indicated that platelet activation increases in the positively charged membrane and decreases in the membrane having a microdomain structure in which hydrophilic and hydrophobic groups coexist randomly as molecules. Also, the complement activation, which in the case of biomaterials proceeds via alternative pathway, is mainly noted with cellulose membranes; free hydroxyl groups on the membrane surface are bonded with C3b and further with factor B that promotes the activation. Improvements of biocompatibility of cellulose membrane can be obtained by surface modification devoted to graft, for example, alkyl group to the hydroxyl group.

2

When membranes are used in bioartificial systems (bioartificial organs and tissues), it is important to consider the cytocompatibility of membrane on which depends the response of the biological components. In bioartificial systems, cells come into contact with the membrane surface. Therefore, their morphological and functional response depends on the surface and transport membrane properties. Physicochemical properties, including surface composition, charge, energy, and morphology, may affect cell adhesion and behavior. In vivo cells are surrounded by the extracellular matrix (ECM) that provides physical architecture and mechanical strength to the tissue. The native ECM exhibits from macro- to nanoscale patterns of chemistry and topography. For this reason, the cells can respond to various chemically and/or topographically patterned features. When cells are cultured in vitro, they receive very different physical, chemical, and mechanical stimuli from the unfamiliar surrounding environment. Microand nanoscale mechanical properties of the membranes are critical because the cells not only adhere to the surface but also pull on the surface substrate and on adjacent cells. The surface chemistry of the membrane affects the adhesion of cells through the ECM protein adsorption and stereospecific chemical interactions. Several approaches have been undertaken to improve the cytocompatibility of membranes by increasing the wettability or by surface modification through grafting of functional groups or immobilization of peptides, and proteins, which interact with the cell receptors (De Bartolo et al. 2005, 2006, Salerno et al. 2009). It is known that the

Membrane Biocompatibility

cytocompatibility of the membrane can be increased by modulating its surface free energy and roughness (De Bartolo et al. 2002). It has been shown a significant enhancement of the hepatocyte adhesion and metabolic functions on membrane with high value of surface free energy base parameters.

References De Bartolo L, Morelli S, Bader A, Drioli E (2002) Evaluation of cell behaviour related to physico-chemical properties of polymeric membranes to be used in bioartificial organs. Biomaterials 23(12):2485–2497 De Bartolo L, Gugliuzza A, Morelli S, Cirillo B, Gordano A, Drioli E (2004) Novel PEEK-WC membranes with low plasma protein affinity related to surface free energy parameters. J Mater Sci Mater Med 15:877–883 De Bartolo L, Morelli S, Lopez L, Giorno L, Campana C, Salerno S, Rende M, Favia P, Detomaso L, Gristina R, d’Agostino R, Drioli E (2005) Biotransformation and liver specific functions of human hepatocytes in culture on RGD-immobilised plasma-processed membranes. Biomaterials 26(21):4432–4441 De Bartolo L, Morelli S, Rende M, Salerno S, Giorno L, Lopez LC, Favia P, d’Agostino R, Drioli E (2006) Galactose derivative immobilized glow discharge processed PES membranes maintain the metabolic activity of human and pig liver cells. J Nanosci Nanotechnol 6:2344–2353 Salerno S, Piscioneri A, Laera S, Morelli S, Favia P, Bader A, Drioli E, De Bartolo L (2009) Improved functions of human hepatocytes on NH3 Plasma–grafted PEEK-WC-PU membranes. Biomaterials 30:4348–4356 Ulhenbusch I, Bonnie-Schorn E, Grassmann A, Vienken J (2004) Understanding membranes and dialysers. Pabst Science Publishers, Lengerich

S

Solubility Maria Grazia De Angelis Dipartimento di Ingegneria Civile, Chimica, Ambientale e dei Materiali, Università di Bologna, Bologna, Italy

Solubility (C) is a thermodynamic quantity characterizing the distribution of chemical species between different phases in equilibrium. In particular, the solubility (of a solute in a solvent) expresses the amount of chemical species (solute), dissolved in a fixed amount of an uptaking species (solvent), at equilibrium. The equilibrium is reached at the end of a sorption (or absorption) process, in which the solute molecules spontaneously move from the original phase to the solvent phase and form a homogenous mixture (solution) of molecules interacting with noncovalent bonds. Often the term “solubility” is also used to indicate the sorption process. The solubility value is measured when the solute net mass transfer is zero and an equal chemical potential of the solute species in the two phases is attained. The time required to reach such equilibrium state depends on the diffusivity of the solute in the solvent in the state of the solution, on the geometry of the system, and the boundary and initial conditions of the mass transfer problem. The solubility of a solute in a solvent depends on the respective chemical nature and thermodynamic state but also on # Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_357-1

binary interactions and can be obtained by solving the phase equilibrium problem, with an appropriate model that allows to represent the chemical potential. The solubility is thus a measure of solute concentration at equilibrium and can be expressed in mass, molar, or volume terms and using different units as reported in Table 1. In membrane processes governed by the solution-diffusion framework like gas and vapor separation, pervaporation, and liquid/liquid separation, the membrane is the solvent in equilibrium with the feed and permeate fluid phases on the upstream and downstream side, respectively. The concentration of solute i absorbed onto the membrane boundaries is equal to Ci,up and Ci,down, respectively. The difference DCi ¼ Ci, up Ci, down ensures solute mass flux across the membrane, according to Fick’s law. Solubility can be measured via gravimetric, manometric, optical methods, etc. (Czichos et al. 2006) and is usually expressed through curves (solubility isotherms) reporting isothermal data. According to the phase rule, in the system formed by two phases (the original solute phase and the solvent phase), one solvent and NS solute species, the number of variables required to univocally define the state of the system is NS+1. The solubility isotherm of a solute i in a solvent reports the solubility of that solute in the solvent versus the composition of the solute in

2 Solubility, Table 1 Units used for solubility C of fluids in membranes. 1 solute, 2 solvent g1/g2 g1/cm3a 2 mol1/mol2 mol1/g2 mol1/cm3a 2 cm31(STP)/cm32 STP standard temperature and pressure, 1 atm and 25 C, 1 cm3(STP) = 4.461 105 mol a Usually cm3 of pure solvent

the original solute phase, which can be expressed with different quantities depending on the state of aggregation of such phase. Some examples are given below. For a gaseous solute phase, the fugacity fi or the partial pressure pi, if the gas phase is ideal, is used. For a vapor solute phase, the activity (ai) is preferentially used, i.e., the ratio between the fugacity in the current state and the fugacity in a reference state, that is usually chosen to be the

Solubility

vapor/liquid saturation point of the pure solute at the same temperature:

ai ¼

f i ðT, p, yi Þ f 0i T, pi, SAT ðT Þ

low

!

pressure

pyi pi, SAT ðT Þ

(1)

For pure liquid solute phases, the solubility, as many other liquid-state properties, is to all practical purposes a weak function of pressure, and its value can be assumed unique at a fixed temperature. For multicomponent liquid solute phases, the activity ai can be used, being gi the activity coefficient: ai ¼ x i g i

ideal

mixture

xi

(2)

References Czichos H, Saito T, Smith L (2006) Experimental measurement of gas and vapor sorption. In: Springer handbook of materials measurement methods. Springer, Berlin, pp 381–385

M

Membrane Biosorption A. B. Koltuniewicz Biotechnology and Bioprocess Engineering Division, Faculty of Chemical and Process Engineering, Warsaw University of Technology, Warsaw, Poland

Biosorption, in the biological sense, is the ability of cells (usually microorganisms) to bind various substances in a passive way. Bioaccumulation of cells is the active retrieval from the environment of certain components needed for life (i.e., metabolism). Biosorption can be achieved through a variety of specific mechanisms, where physical adsorption, ion exchange, and chemical reactions play a dominant role. Table 1 (below) shows the sorption capacity of various microorganisms such as bacteria, fungi, and algae. Biosorption is also a new separation process (Volesky 1990) which depends on the concentration, temperature, pH, and sorbent properties that determine the equilibrium and kinetics. Cell debris may be also regarded as sorbents, which have even better sorption properties than whole cells because of greater sorption capacity (see Table 1) and rate of sorption (see Table 2). The smaller the particle size, the greater the surface areas and shorter diffusion paths for the molecules of xenobiotic (Koltuniewicz and Bezak 2002; Koltuniewicz and Witek 2004). Due to the special ability to bind heavy metals, biosorption process # Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_358-7

is currently used mainly in metalworking operations such as processing of ores, the production of batteries, power, thermal, and nuclear power generation, and some industrial wastewater treatment systems (Koltuniewicz and Drioli 2008; Koltuniewicz 2010). General examples of applications of biosorption are (I) removal of harmful substances from the environment and (II) separation of valuable substances from multicomponent mixtures. In the first case, the desired feature and the main value of biosorption are its low cost and widespread availability of biosorbents. Dealing with large amounts of contaminated water, which must be processed and, moreover, contains rather diluted substances to remove, such as heavy metals, organic substances or radioactive materials, is the biggest challenge for process engineering. In these cases, almost all methods of separation fail because their costs are rising dramatically. Therefore, the use of cheap biosorbents, which are often waste materials (algae, fungi, bacteria, sawdust, bark, etc.), may be helpful, especially for environmental protection. In the second case (II), selective and specific sorbents with a high affinity for precious substances are sought. This method can also be useful to concentrate the atoms of noble metals such as gold, platinum and uranium, from industrial waste water, or even diluted in sea water. In both cases, the use of membrane-assisted biosorption, namely, the “membrane biosorption,” may be recommended

2

Membrane Biosorption

Membrane Biosorption, Table 1 Maximum uptake of heavy metals on various microorganisms Type of biosorbent (microorganisms) Bacillus subtilis bacteria Bacillus licheniformis bacteria Aureobasidium pullulans bacteria Citrobacter sp bacteria Sargassum natans algae Chlorella vulgaris algae Sargassum natans algae Chondrus crispus algae Ascophyllum nodosum algae Rhizopus arrhizus fungus Rhizopus arrhizus fungus Streptomyces noursei fungus Saccharomyces cerevisiae fungus Saccharomyces cerevisiae fungus Aspergillus niger fungus Penicillium chrysogenum fungus

Heavy metal Au Fe Pb U Au Au Au Au Cd Ag Au Ag Ag Cd Au Cd

Maximum uptake q (mg metal/g biomass) 79 45 220–360 800 400 80 400 76 215 54 164 38 5 20–40 176 56

Membrane Biosorption, Table 2 The isotherms of biosorption equilibrium (Pagnanelli et al. 2003) Isotherm Langmuir

Equation

Freundlich

q ¼ k Cn

Langmuir–Freundlich

q ¼ qmax

Radke and Prausnitz

1 q

Reddlich–Peterson

aC q ¼ 1þbC n

(Koltuniewicz and Bezak 2002). The membranes allow for easy concentration of finely ground biosorbents (Fig. 1) or cells (Fig. 2), which is impossible in the thick layer of packed columns. Very fine grinding is desirable because of the possibility of increasing the surface of sorption and shortening of the diffusion paths to active sites. This increases the efficiency of sorption, even in dilute suspensions. However, the costs of milling and additional operations such as drying, sorting, etc. should always be taken into account. So in order to make real use of biosorption, the calculations must be performed in any case. The

bC q ¼ qmax 1þbC 1

1 ¼ aC þ

1

bCn 1 1þbCn 1 bCn

next part shows the equations needed to calculate the membrane biosorption process.

Calculations of Membrane Biosorption Process The obvious drawback of membrane biosorption is the quick saturation of the thin layer of biosorbent on membrane surface is rapidly progressing with time, until equilibrium is reached. The equations describing the equilibrium and kinetics for biosorption process can be

Membrane Biosorption

3

Membrane Biosorption, Fig. 1 Equilibrium adsorption of copper ions to the pine sawdust (a). Effect of temperature (b) and pH (C). (A) Photomicrograph of pine sawdust used for the biosorption of copper ions. (B) Effect of temperature on biosorption of copper ions on the sawdust. (C) Effect of pH on biosorption of copper ions on the sawdust

a

b

24 22 20 18

q [mg/g]

16 14 12 10 8 6 4

T = 20 T = 30 T = 40

2 0 0

20

40

60

80

100

C [mg/I]

c

24 22 20 18

q [mg/g]

16 14 12 10 8 6 pH = 5.0 pH = 4.0 pH = 3.0

4 2 0

20

40

60 C [mg/I]

80

100

4

Membrane Biosorption

q* – equilibrium value of solute concentration in the biosorbent (maximum uptake see Table 1) q0 – initial solute concentration in the biosorbent, e.g., solute concentration in the sorbent which is supplied with the feed

Membrane Biosorption, Fig. 2 Microorganisms on the membrane surface

During dead-end mode of operation, when the unaffected biosorbent lies on the membrane surface, and the liquid passes through its layer, the concentration in the fluid changes according to equilibrium. Change of concentration over time in the permeate stream can be then calculated from the mass balance (see Fig. 3) (Koltuniewicz and Bezak 2002; Koltuniewicz and Witek 2004). The balance of xenobiotic in an open system, during the flow of fluid through a layer of biosorbent,

Membrane Biosorption, Table 3 Kinetic equation of the biosorption (Pagnanelli et al. 2003) Reaction type First-order reaction

Equation

Pseudo-second order reaction

q ð tÞ ¼

qðtÞ ¼ q 1 ekt 1

1 þt kq q

Membrane Biosorption, Table 4 Experimentally determined constants m and n in correlation between rate of surface renewal and average cross-flow velocity in different types of modules (Koltuniewicz 1995) Module type Plate and frame (slot 0.15 0.001 m) Tubular ceramic (Membralox ® 19 tubes) Capillary (Romicon ®, d = 1 mm)

determined on the basis of experimental data (see Tables 1, 2, 3, and 4 in Fig. 1). The concentration in the biosorbent (q) varies with time, due to its saturation with solute (i.e., xenobiotic), which can be expressed in the formula: qðtÞ ¼ q ðq0 q Þ ekt Where: q(t) [kgsolute/kgbiosorbent] – solute (xenobiotic) concentration in the biosorbent

Constant – m 0.00074 0.0020 0.0035

Constant – n 0.75 0.80 0.66

permeates flow results in a superposition of two effects: (1) the tendency to reduce the concentration of xenobiotic in the permeate due to its biosorption and (2) the tendency to increase the concentration as a result of inflow of xenobiotic, with the liquid. Thus, the relationship has a minimum of the xenobiotic concentration in the permeate after biosorption:

Membrane Biosorption

5

CP ðtÞ ¼ CR

i k Xm d ðq qR Þ h kt J e e dt Jkd

Where: CP(t) – instantaneous concentration of solute in the permeate CR [kgsolute/m3retentate] – constant concentration of solute in the retentate inflow, Xm [kgbiomass/m3retentate] – average biomass concentration on the membrane surface d [m] – thickness of the sorbent layer on the membrane k [s1] – biosorption kinetic constant qR [kgsolute/kgbiomass] – solute concentration in the sorbent in the retentate q* – maximum uptake, i.e., equilibrium value of the solute concentration in the sorbent J [m3/m2s] – permeate flux t [s] – real time of the process It should be underlined that sorption process in dead-end mode is unpractical because of very small volume of active biosorbent layer at the membrane surface. Therefore, to maintain stationary conditions of the process, some way to remove the thin active layer of sorbent from the membrane surface after appropriate process time must be found. Duration, after which the concentration of solute in the permeate is the minimum, may be selected as a time of the periodic back flushing and can be calculated from the formula:

Methods to reduce the concentration polarization layer are well known in the membrane processes to maximize the permeate flux. They are turbulences or shear forces, shear-induced diffusion, lifting forces, pinch effects, lateral migrations, scouring effects, gas sparging, etc. We can now apply them for controlled renewal of a biosorbent film from the surface of the membrane by using hydrodynamic effects during cross flow. Application of Danckwerts model allows to determine the optimal rate of surface renewal (Koltuniewicz and Witek 2004).

sopt

rffiffiffiffiffiffiffiffiffiffi kJ dCm P ¼0 ¼ for Cmin P when d ds

The optimal conditions for renewing the surface we get when the lowest concentration of the xenobiotic in the permeate (i.e., the highest possible purity) of cross-flow conditions can be achieved. This is when the frequency of renewal of the sorbent film(s) is synchronized with the (i) flow of permeate “J,” (ii) thickness of the sorbent layer on the membrane “d,” and (iii) the kinetics of sorption “k” (see Fig. 1). Then the average concentration in the permeate outflow from the entire surface of the membrane (i.e., the sum of the concentrations in the streams flowing from all of its components) can be expressed by the formula (see also Fig. 3 at continuous line):

kd ln min J t CP ¼ J k d

2

ðdJ þsÞtp

ðkþsÞtp 1e 61 e Cm P ¼ CR 4 J kþs þs d

3 s 7 k X m d ð q qR Þ 5 Jkd 1 estp

6

Membrane Biosorption

CONCENTRATION OF THE SOLUTE IN THE PERMEATE [mM/ml]

1,50E–04

dead-end cross-flow with optimum surface renewal

0,00E+00 100

200

300

TIME [S]

Membrane Biosorption, Fig. 3 Purity of permeate after the process of membrane biosorption in dead-end and crossflow modes of operation

After the long duration of process, this concentration becomes constant, leveling off on the asymptotic value independent on time of the process tp (see Fig. 1 dashed line). Cm P ¼ C R X m ð q qR Þ

kds ð J þ s dÞ ð k þ s Þ

The rate of the surface renewal depends on many hydrodynamic parameters and must be determined experimentally. However, rough estimates assuming close to the water systems can use a simple exponential dependence on average velocity retentate at different membrane modules, which were presented in Koltuniewicz (1992). s ¼ m un All the above data and formulas can be very helpful in practical use of opportunities and potential of membrane biosorption as a new separation process.

References Koltuniewicz AB (1992) Predicting permeate flux in ultrafiltration on the basis of surface renewal concept. J Membr Sci 68:107–118 Koltuniewicz AB (1995) Yield of the pressure-driven membrane processes in the light of the surface renewal theory. Oficyna Wydawnicza PWr, Wroclaw (in Polish) Koltuniewicz AB (2010) Integrated membrane operations in various industrial sectors. In: Drioli E, Giorno L (eds) Comprehensive membrane science and engineering, 1st edn. Elsevier, Oxford Koltuniewicz AB, Bezak K (2002) Engineering of membrane biosorption. Desalination 144:219–226 Koltuniewicz AB, Drioli E (2008) Membranes in clean technologies-theory and practice. Wiley&KGaA, Weinheim Koltuniewicz AB, Witek A (2004) Efficiency of membrane-sorption integrated processes. J Membr Sci 239:129–141 Pagnanelli F, Beolchini F, Biase A, Di Veglio V (2003) Effect of equilibrium models in the simulation of heavy metals biosorption in single and two-stage UF/MF membrane reactor systems. Biochem Eng J 15:27–35 Volesky B (1990) Biosorption of heavy metals. Ann Arbor, Boston

M

Membrane Capacitive Deionization Maarten Biesheuvel Wetsus, European Centre for Excellence for Sustainable Water Technology, Leeuwarden, Netherlands

Membrane capacitive deionization, MCDI, is a technology for water desalination based on the use of an assembly of a spacer channel, two ion exchange membranes, and two porous carbon electrodes. Of the two electrodes, one is the cathode, in which cations are adsorbed during desalination, and the other is the anode. In front of the cathode, a cation exchange membrane is placed, while an anion exchange membrane is placed in front of the anode. MCDI is a cyclic process, in which for some time freshwater is produced, followed by a period in which a concentrate stream leaves the device. To increase water recovery, the second period (ion release) must be short relative to the first period (ion adsorption). In MCDI, upon transferring electrical charge from cathode to anode, against a cell voltage difference opposing this electrical charge transfer, the ions are removed from the water flowing through the spacer channel and are adsorbed in their counter electrode. In this way water is obtained which is desalinated to a certain degree. After some time, the electrodes have reached their adsorption capacity, and the electrical # Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_359-5

current is reversed, thereby releasing the ions from the electrodes, leading to a stream of brine (concentrate) flowing from the device. As each layer in the assembly is about 100–300 mm of thickness, a complete cell, including also “graphoil” current collectors on either side, is about 1.0–2.0 mm of total thickness. Thus, in a “stack” of 10 cm of height, it is possible to pack up to 100 of such cells. The current collectors serve to inject the electrical current from the external circuit into the porous electrodes, which are accessible for the water and the ions. Ion storage is based on the formation of electrostatic double layers within the nanopores in the carbon electrode particles, where electrical charge is locally charged compensated by counterions. The advantage of MCDI over the related technology without membranes, CDI, is that during the charging or desalination step, co-ions (ions of the same charge sign as the electrode) are blocked from leaving the electrode region. This effect does occur in CDI and limits the charge efficiency, i.e., the amount of salt molecules removed from the water per amount of charge transferred. A second advantage of MCDI is that upon reversing the voltage during cell discharge, it is possible in MCDI to completely remove all counterions from their electrode, thereby increasing the desalination capacity in a next cycle.

M

Membrane Contactor (MC) A. Criscuoli Institute on Membrane Technology (ITM-CNR), Rende (CS), Italy

Membrane contactors are membrane-based devices that are used to carry out many unit operations like gas-liquid mass transfers, liquidliquid extractions, and distillation (Drioli et al. 2006). The membranes employed can be both hydrophobic and hydrophilic and their pore size usually ranges from 0.05 to 1 mm. The role of the membrane is to put in contact the involved phases, and the separation is not due to its selectivity, as in the other membrane operations, but the mass transport occurs at the interface, where the two phases are in contact. In particular, hydrophobic membranes are able to block at their surface polar phases, whereas hydrophilic membranes prevent the passage of nonpolar ones (Fig. 1a, b). Being each micropore a point of contact for the phases, and being the number of micropores in one membrane extremely high, one of the main

# Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_360-2

advantages of this system, with respect to conventional operations, is the high interfacial area available in small volumes. Furthermore, the possibility of having the interface established at each micropore results in a more stable and better contact between phases. However, for ensuring that no mixing of phases occurs, it is mandatory to do not exceed the so-called breakthrough pressure, that is, the value of pressure at which the phase blocked at the pore mouths starts to penetrate inside the pores. The value of the breakthrough pressure depends on different parameters, like the membrane pore size, the degree of hydrophobicity of the membrane surface, and the fluid properties and can be calculated by Laplace’s equation: DP ¼ ð2scosyÞ=r with s, surface tension of liquid; y, contact angle between the liquid and the membrane; and r, membrane pore radius. Different are the membrane operations that belong to the “membrane contactors family” like membrane strippers and scrubbers, supported liquid membranes, phase transfer catalysis, and membrane and osmotic distillation.

2 Membrane Contactor (MC), Fig. 1 Interface in a hydrophobic (a) and hydrophilic (b) membrane contactor and transport of the species i through the membrane micropore

Membrane Contactor (MC)

a

b

i

Polar phase

i

Apolar phase

Polar phase

Apolar phase

Membrane Contactor (MC), Table 1 Main advantages and drawbacks of membrane contactors Main advantages High compactness Known and stable interfacial area No need of units downstream for phases separation Flexibility and modularity

Main drawbacks Membrane resistance to the mass transport Membrane fouling and loss of membrane properties (like hydrophobicity) Operating pressures dependent on the breakthrough pressure value

Table 1 summarizes the main advantages and drawbacks of membrane contactors.

References Drioli E, Criscuoli A, Curcio E (2006) Membrane contactors: fundamentals, applications and potentialities. Membrane science and technology series 11, ISBN:0-444-52203-4 Elsevier, Amsterdam

M

Membrane Distillation (MD) Francesca Macedonio Institute on Membrane Technology, National Research Council of Italy, ITM-CNR, University of Calabria, Renede (CS), Italy

Membrane distillation (MD) is a thermal membrane operation known for more than 50 years (first patent was filed by Bodell on 3 June 1963 (Findley 1967); first MD paper was published 4 years later by Findley (Bodell 1963)). During the last years, the number of papers published on MD and the research groups focusing on MD studies have been increasing (Fig. 1). MD has great potential as a concentration process at low temperature and energy with respect to conventional processes, such as distillation and reverse osmosis (RO). MD allows the separation of volatile components from solutions. If the solutions contain nonvolatile components, it is possible to remove solvent by concentrating the solutions. In MD, one side (feed side) of a hydrophobic membrane is brought into contact with a heated, aqueous feed solution. The hydrophobic nature of the membrane prevents penetration of the aqueous solution into the pores, resulting in a vapor–liquid interface at each pore entrance. Here, volatile compounds evaporate, diffuse and/or convect across the pores, and are condensed on the opposite side (permeate) of the # Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_361-2

system (Fig. 2). The driving force of the process is linked to the vapor-pressure gradient between the two membrane sides. The documented and expected benefits resulting from MD are as follows: – The nature of the driving force and the hydrorepellent character of the membrane allow the theoretically complete rejection of nonvolatile components such as macromolecules, colloidal species, and ions. – Lower temperatures with respect to those usually used in conventional distillation column are sufficient to establish interesting transmembrane flux, with consequent reduction of energy costs, thus allowing the efficient recycle of low-grade waste heat streams as well as the use of alternative energy sources (solar, wind, or geothermal). – Lower pressures with respect to those usually utilized in RO because the required operating pressures are of the order of few hundred kPa. Lower operating pressure translates to lower equipment costs, increased process safety, and possibility of using plastic equipment, thus reducing or avoiding erosion problems. – If compared to RO process, MD permeate flux is only slightly affected by the concentration of the feedwater, and thus, productivity and performance remain roughly the same for high concentration feedwaters. This means that MD can be preferentially employed whenever elevated permeate recovery factors or high

2

Membrane Distillation (MD)

Number of article on MD published in Journal of membrane Science

25

20

15

10

5

2011

2009

2007

2005

2003

2001

1999

1997

1995

1993

1991

1989

1987

Year

Membrane Distillation (MD), Fig. 1 Number of the papers published in Journal of Membrane Science on MD field each year up to September 2011 (Drioli et al. 2012)

Membrane Distillation (MD), Fig. 2 General scheme of the MD process: aqueous solution on feed side whereas four different solutions can be realized on permeate side (aqueous solution or air gap or vacuum or sweeping gas) (Lawson and Lloyd 1997)

retentate concentrations are requested. On the contrary, in MD temperature polarization, similar to concentration polarization, arises from heat transfer through the membrane and it is often the rate-limiting step for mass transfer. – Since MD membranes act merely as a support for a vapor–liquid interface, they do not

distinguish between solution components on a chemical basis, do not act as a sieve, and do not react electrochemically with the solution; they can be fabricated from almost any chemically resistant polymers with hydrophobic intrinsic properties, such as polytetrafluoroethylene (PTFE), polypropylene (PP), and polyvinylidene difluoride (PVDF). This

Membrane Distillation (MD)

3

Membrane Distillation (MD), Fig. 3 Heat transfer resistances in MD

1/hms

1/hmg 1/hf

1/hp 1/hV

Tf

characteristic increases membrane life. New amorphous perfluoro polymers (e.g., Hyflon, Teflon) can be also utilized neglecting their still high costs. On the other hand, MD performance (in particular the transmembrane flux and the heat loss by conduction through the membrane) is intrinsically affected by the structure of the membrane in terms of thickness, porosity, mean pore size, pore distribution, and geometry. – Membrane fouling is less problematic in MD than in other membrane separations because (i) the pores are relatively large compared to RO/UF pores; (ii) the process liquid cannot wet the membrane – therefore, fouling layers can be deposited only on the membrane surface but not in the membrane pores; and (iii) due to the low operating pressure of the process, the deposition of aggregates on the membrane surface would be less compact and only slightly affect the transport resistance. – On the contrary, one problem that can arise in MD is the membrane wetting which occurs when the liquid penetrates into the membrane pores. Once a pore has been penetrated, it is said to be “wetted” and the membrane must be completely dried and cleaned before the wetted pores can once again support a vapor–liquid interface.

Tfm

Tpm

Tp

In MD process both heat and mass transfer phenomena occur through the membrane. Figure 3 illustrates the possible heat transfer resistances in MD with an electrical analogy. Heat is first transferred from the heated feed solution of uniform temperature Tf, across the thermal boundary layer to the membrane surface at a rate Q = hf ∙ DTf. At the surface of the membrane, liquid is vaporized and heat is transferred across the membrane at a rate QV = hV ∙ DTm = N ∙ DHV (where N is the rate of mass transfer and DHV is the heat of vaporization). Additionally, heat is conducted through the membrane material and the vapor that fills the pores at a rate Qm = hm ∙ DTm where hm ¼ e hmg þ ð1 eÞhms (e is the membrane porosity; hmg and hms represent the heat transfer coefficients of the vapor within the membrane pores and the solid membrane material, respectively). Conduction is considered a heat loss mechanism because no corresponding mass transfer takes place. Total heat transfer across the membrane is Q = QV + Qm. Finally, as vapor condenses at the vapor–liquid interface, heat is removed from the cold-side membrane surface through the thermal boundary layer at a rate Q = hp ∙ DTp. The overall heat transfer coefficient of the MD process is given by

4

Membrane Distillation (MD)

1 1 1 1 ¼ þ þ U hf hm þ hv hp ¼

1 hf þ

1 Kg e þ Km ð1 eÞ N DHV þ d Tfm Tpm

1 þ hp (1)

In literature, the recommended range of TPC is from 0.4 to 0.7 for well-designed systems (Lawson and Lloyd 1997). The boundary layer heat transfer coefficients are almost always estimated from empirical correlations such as the following: • Sieder–Tate correlation for turbulent liquid flow inside circular tubes Nu KT , Nu ¼ 0, 023 Re0, 8 Pr1=3 d 0, 8 1=3 cp m hd dG 0, 14 fm or T ¼ 0, 023 m KT K 0:14 h¼

where each h and each T represent the corresponding heat transfer coefficients and temperatures shown in Fig. 3. The total heat transferred across the membrane is given by Q ¼ U DT

(2)

Equation 1 illustrates the importance of minimizing the boundary layer resistances (maximizing the boundary layer heat transfer coefficients). A commonly used measure of the magnitudes of the boundary layer resistances relative to the total heat transfer resistance of the system is given by the temperature polarization coefficient (TPC): TPC ¼

Tfm Tpm Tf Tp

(3)

• If TPC ! 1, the MD system is well designed, and it is limited by mass transfer. • If TPC ! 0f , the MD system is poorly designed and it is limited by heat transfer through the boundary layers.

de ¼ 4 rH ¼ 4

m mw

(4) where d is the tube diameter, KT is the thermal conductivity of the liquid, G is the mass velocity equal to w/S = , m is the bulk liquid viscosity, mw is the liquid viscosity at the wall, cp is the liquid heat capacity, and fm is the heating/ cooling correction factor. Equation 4 should be used for Re > 6000 and for tubes with large ratios L/d (where L is the tube length). For short tubes (L/d < 50), several corrections are available, including d 0, 7 h ¼ 1 þ where h1 is the heat transfer h1 L coefficient given by Eq. 4. For the case of a noncircular flow channel, these correlations can still be used if the equivalent diameter de of the flow channel is substituted:

S across sectional area of the flow channel ¼4 ðrH ¼ hydraulic radiusÞ LP lenght of the wetted perimeter of the flow channel

Membrane Distillation (MD)

5

Membrane Distillation (MD), Fig. 4 Mass transfer resistances in MD

viscous

Knudsen

1/hf

molecular

1/hp

surface

• Sarti correlation for laminar liquid flow in circular tubes with constant wall temperature:

Nu ¼ 3:66 þ _ p mc Gz ¼ T K L

0:067 Gz 1 þ 0:04 Gz2=3

The resistances shown in Fig. 4 are arranged as described by the dusty gas model (DGM), which is a general model for mass transport in porous media.

where

(5) where Gz is the Graetz number, m_ is the mass flow rate, cp is the liquid heat capacity, KT is the liquid thermal conductivity, and L is the length of the tubes. However, several empirical correlations exist which allow to estimate the boundary layer heat transfer coefficients for other geometries and heat transfer mechanisms. The heat transfer across the membrane has been already described. For what concerns the heat transferred by convection within the membrane pores, this can be also considered but is negligible because convection accounts for, at most, 6 % of the total heat lost through the membrane and only 0.6 % of the total heat transferred across the membrane (Lawson and Lloyd 1997). Regarding mass transfer, Fig. 4 illustrates the possible mass transfer resistances in MD using an electrical analogy.

• Mass transfer across boundary layers A mass balance across the feed side boundary layer yields the relationship between molar flux N, the mass transfer coefficient kx, and the solute concentrations cm and cb at the interface and in the bulk, respectively (Curcio and Drioli 2005; Mya Tun et al. 2005): N cm ¼ kx ln r cb

(6)

where r is the solution density. The method that is used in literature to determine the mass transfer coefficient is to employ an analogy between heat and mass transfer. Therefore, Eqs. 4 and 5 can be used to estimate boundary layer mass transfer coefficients by substituting the Sherwood number for the Nusselt number, the Schmidt for the Prandtl, and the mass transfer Graetz number for its heat transfer form. In general, the used correlations are as follows:

6

Membrane Distillation (MD)

Sh ¼ a Reb Scg

(7)

where Sh = Sherwood number = (kxdh)/D (dh hydraulic diameter, D diffusion coefficient in the liquid), Sc = Schmidt number = m/(rD) (m is the bulk liquid viscosity; r is the solution density), and GzM = mass transfer Graetz _ _ number = GzM ¼ rDmAB L (m is the mass flow rate; L is the tube length). As a result of the solvent transmembrane flux across the membrane, when aqueous solutions containing nonvolatile solutes are considered, the concentration of the nonvolatile solutes at the membrane surface (CBm) becomes higher than that at the bulk feed (CBb) with time as long as the separation process is taking place. Almost 100 % of separation is obtained. In this case, care must be taken as supersaturation states may eventually be achieved affecting the efficiency of the membrane process. The term concentration polarization coefficient (CPC) is defined to quantify the mass transport resistance within the concentration boundary layer at the feed side as follows: CPC ¼

cBm cBb

(8)

The increased concentration of nonvolatile compounds next to the membrane surface would have the influence of reducing the transmembrane flux due to the establishment of concentration polarization (CP) layer at the feed side that acts as a mass transfer resistance to the volatile molecule species (water). Fortunately, in MD process, the low to moderate flow rates and high heat transfer coefficients reduce the impact of concentration polarization, which is lower than that of the temperature polarization effect (Laganà et al. 2000; El-Bourawi et al. 2006; Drioli et al. 1999; Srisurichan et al. 2006). In fact, boundary layers next to the membrane can contribute substantially to the overall transfer resistance; heat transfer across the boundary layers is often the rate-limiting step for mass transfer in MD because a large quantity of heat must be

supplied to the membrane surface to vaporize the liquid and because the membrane fabrication technology has improved in the last decades so much that MD process has shifted away from being limited by mass transfer across the membrane to being limited by heat transfer through the boundary layers on either side of the membrane.

• Mass transport through the membrane pores As stated earlier, the mass transfer process in MD is driven by the imposed vaporpressure gradient between both sides of the membrane. The mass transport mechanism is governed by three basic mechanisms known as Knudsen diffusion, Poiseuille flow, and molecular diffusion or the combinations between them known as transition mechanism (excluding surface diffusion, negligible in MD because, by definition of the MD phenomenon, molecule–membrane interaction is low and the surface diffusion area in MD membranes is small compared to the pore area). The dusty gas model is usually used as a general model taking into account the latter basic mechanisms (Lawson and Lloyd 1997; Curcio and Drioli 2005; El-Bourawi et al. 2006): D n X pj N D ND 1 i pi Nj i ∇p þ ¼ RT i D0ije Dkie j¼16¼i

Nvi ¼

Dkie

e r2 p i ∇P 8RTtm

2er ¼ 3t D0ije ¼

rffiffiffiffiffiffiffiffiffiffiffi 8RT p Mi

e 0 D t ij

(9)

(10)

(11)

(12)

where ND is the diffusive flux, NV is the viscous flux, Dk is Knudsen diffusion coefficient, D0 is the ordinary diffusion coefficient, pi is the partial pressure of the component i, P is the total pressure, Mi is the molecular weight of component i, r

Membrane Distillation (MD) Membrane Distillation (MD), Fig. 5 Common configurations of the membrane distillation process that may be utilized to establish the required driving force (El-Bourawi et al. 2006)

7

Feed in

permeate out

Coolant out

Feed in

Air gap Membrane Membrane Condensing plate Feed out

permeate in

Feed out

Coolant in

Product AGMD Configuration

DCMD Configuration

Feed in

Sweep gas out

Feed in

Vacuum

Membrane

Condenser

Membrane

Condenser

Permeate Feed out

Sweep gas in

SGMD Configuration

is the membrane pore radius, e is the membrane porosity (assuming the membrane consists of uniform cylindrical pores), m is the fluid viscosity, and t is the membrane tortuosity. The subscript “e” is indicative of the effective diffusion coefficient function of the membrane structure. There is only one problem with the application of the DGM to MD, and that lies in the fact that MD is a non-isothermal process. The DGM was derived for isothermal flux, but has been successfully applied to non-isothermal systems via the inclusion of terms for thermal diffusion and thermal transpiration. However, it is easily shown (Lawson and Lloyd 1997) that these terms are negligible in the MD operating regime, and Tavg in the membrane is used in place of T in the DGM equations. Regardless of which mechanism is involved in the mass transportation process, the molar flux, N, must be proportional to the vaporpressure difference across the membrane:

Permeate Feed out VMD Configuration

N ¼ C DP where DP is the vapor-pressure difference across the membrane (function of temperatures and compositions at the membrane surface) and C is the membrane distillation coefficient that can be obtained experimentally. C is a function of temperature, pressure, and composition within the membrane as well as membrane structure and depends on the MD configuration employed as well as on the Knudsen number (Kn, ratio of the mean free path of the transported gas molecules (l) through the membrane pores to the mean pore diameter of the membrane (d)). In fact, Kn number determines the physical nature of flow through membrane pores, and since the membranes used in MD exhibit pore size distribution, more than one mechanism may occur through the membrane. Whereas on feed side only an aqueous solution can be present, the nature of the permeate can be different and gives origin to the four basic MD configurations:

distillate channel condenser foil TH

coolant

hot feed

evaporator

T1 P1

condenser

hydrophobic membrane

membrane

Membrane Distillation (MD), Fig. 6 Basic channel arrangement and temperature profile for PGMD (Winter et al. 2011)

Membrane Distillation (MD) distillate channel

8

T0 TC P0

distillate

– Direct contact MD (DCMD), in which the permeate is in liquid phase and, therefore, the membrane is on both sides directly in contact with aqueous solutions – Vacuum MD (VMD), in which the vaporized solvent is recovered by vacuum and condensed, if needed, in a separate device – Air gap MD (AGMD), in which an air gap is interposed between the membrane and a condensation surface – Sweeping gas MD (SGMD), in which a stripping gas is used as a carrier for the removal of the produced vapor (Fig. 5) The type of employed MD depends upon permeate composition, flux, and volatility: • SGMD and VMD are typically used to remove volatile organic or dissolved gas from an aqueous solution. • Because AGMD and DCMD do not need an external condenser, they are best suited for applications where water is the permeating flux. • The DCMD configuration, which requires the least equipment and is simplest to operate, is best suited for applications such as desalination or the concentration of aqueous solutions (orange juice), in which water is the major permeate component. • AGMD, which is the most versatile MD configuration, can be applied to almost any applications.

Some new configurations with improved energy efficiency, better permeation flux, or smaller footprint have been proposed such as material gap membrane distillation, multi-effect membrane distillation, multi-effect vacuum membrane distillation, permeate gap membrane distillation, and hollow fiber multi-effect membrane distillation. Permeate gap membrane distillation (PGMD) is an enhancement of DCMD in which a third channel is introduced by an additional non-permeable foil (Fig. 6). One significant advantage of PGMD is the separation of the distillate from the coolant. Therefore the coolant can be any other liquid, such as cold feedwater. This offers the opportunity to integrate an efficient heat recovery system. Multi-effect membrane distillation (MEMD) is based on the concepts of multi-stage and multi-effect distillation for seawater desalination. The cold feed solution is placed beneath the condensation surface as a coolant to condense the permeated vapors as well as to gain heat at the same time. The pre-heated feed solution is further heated before it enters the feed channel. Vacuum-multi-effect membrane distillation (V-MEMD) is a modified form of VMD that integrates the concept of multi-effect distillation into the VMD. As a general principle of the process, the vapors produced in each stage are condensed during the subsequent stages. Vapors are generated in steam raiser working under vacuum by exchanging the heat provided by external source. The vapors are introduced in first stage

Membrane Distillation (MD)

where these are condensed by exchanging the heat with feed via a foil. The vapors generated in first stage are transported through the membrane and collected on the foil in the second stage. It has been claimed that these modules have excellent gained to output ratio which is crucial parameter for industrial applications (Zhao et al. 2013). A condenser is used to condense the vapors generated in final stage. The vapor pressure in each stage is less than its preceding stage. Material gap membrane distillation (MGMD) consists of filling the gap between the membrane and the condensation plate with different materials having different characteristics such as polyurethane (sponge), polypropylene mesh, sand, and deionized water in order to increase transmembrane flux with respect to AGMD. Osmotic distillation (OD) represents another extension of the MD concept: a microporous hydrophobic membrane separates two aqueous solutions that are kept in contact at different solute concentrations; this difference in activity causes a vapor-pressure difference that activates mass transport through the membrane. Because OD operates essentially at room temperature, it is appropriate for applications in the agro-food industry (such as in integrated membrane system for the clarification and the concentration of citrus and carrot juices that has been proposed as an alternative and efficient approach to the traditional techniques currently in operation), in pharmaceutical biotechnology, and medicine (more information can be found in Drioli et al. (2006), Drioli et al. (2015)). To date, the slow progress of MD has been related with the unavailability of appropriate membranes and modules for MD applications, high energy consumption with respect to RO that increases the overall energy demand, membrane wetting, and low flux. However, thanks to the recent and growing extensive research activities carried out in various areas of MD, the process has become much more attractive due to the

9

availability of better membranes and to the possibility to utilize alternative energy sources in niche applications.

References Bodell BR (1963) Silicone rubber vapor diffusion in saline water distillation. US Patent 285,032 Curcio E, Drioli E (2005) Membrane distillation and related operations-a review. Sep Purif Rev 34:35–85 Drioli E, Lagana F, Criscuoli A, Barbieri G (1999) Integrated membrane operations in desalination processes. Desalination 122:141–145 Drioli E, Criscuoli A, Curcio E (2006) Membrane contactors: fundamentals, applications and potentialities, vol 11, Membrane science and technology series. Elsevier, Amsterdam/Boston Drioli E, Macedonio F, Ali A (ed) (2012) Membrane distillation: basic aspects and applications – Virtual Special Issue of Journal of Membrane Science. Elsevier Drioli E, Ali A, Macedonio F (2015) Membrane distillation: recent developments and perspectives. Desalination 356:56–84 El-Bourawi MS, Ding Z, Ma R, Khayet M (2006) Review. A framework for better understanding membrane distillation separation process. J Membr Sci 285:4–29 Findley ME (1967) Vaporization through porous membranes. Ind Eng Chem Process Des Dev 6(2):226–230 Laganà F, Barbieri G, Drioli E (2000) Direct contact membrane distillation: modeling and concentration experiments. J Membr Sci 166:1–11 Lawson KW, Lloyd DR (1997) Membrane distillation. J Membr Sci 124:1–25 Mya Tun C, Fane AG, Matheickal JT, Sheikholeslami R (2005) Membrane distillation crystallization of concentrated salts – flux and crystal formation. J Membr Sci 257:144–155 Srisurichan S, Jiraratananon R, Fane AG (2006) Mass transfer mechanisms and transport resistances in direct contact membrane distillation process. J Membr Sci 277:186–194 Winter D, Koschikowski J, Wieghaus M (2011) Desalination using membrane distillation: experimental studies on full scale spiral wound modules. J Membr Sci 370:104–112 Zhao K, Heinzl W, Wenzel M, B€ uttner S, Bollen F, Lange G, Heinzl S, Sarda N (2013) Experimental study of the memsys vacuum-multi-effect-membranedistillation (V-MEMD) module. Desalination 323:150–160

M

Membrane Distillation Applications Francesca Macedonio Institute on Membrane Technology, National Research Council of Italy, ITM-CNR, Renede (CS), Italy

Membrane distillation (MD) is a non-isothermal process that allows the separation of volatile components from solutions. If the solutions contain nonvolatile components, it is possible to remove solvent by concentrating the solutions. MD has been applied for separation of nonvolatile components from water like ions, colloids, and macromolecules (El-Bourawi et al. 2006; Lawson and Lloyd 1996, 1997; Mengual and Pena 1997; Khayet et al. 2003, 2006; Sudoh et al. 1997; Drioli et al. 1986; Zolotarev et al. 1994); for the removal of trace volatile organic compounds from water such as benzene, chloroform, and trichloroethylene (Lawson and Lloyd 1997; Duan et al. 2001; Banat and Simandl 1996, 2000; Sarti et al. 1993; Qureshi et al. 1994; Banat and Al-Shannag 2000); or for the extraction of other organic compounds such as alcohols from dilute aqueous solutions (Lawson and Lloyd 1997; Garcia-Payo et al. 2000; Banat and Simandl 1999; Bandini et al. 1997; Bandini and Sarti 1999). As a consequence, MD is suited for both concentration of aqueous solutions and water production. In fact, MD has been applied for water desalination where near 100 % rejection # Springer-Verlag Berlin Heidelberg 2015 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_362-1

of nonvolatile ionic solutes is easily achieved (an example can be found in the European-funded project MEDINA (Drioli et al. 2011)), wastewater treatment and food processing (concentration of juice and raw cane sugar), and biomedical applications (such as water removal from blood and treatment of protein solutions) (El-Bourawi et al. 2006). Moreover, thanks to its characteristic to work at lower temperatures with respect to those usually used in thermal processes and thus allowing the efficient use of alternative energy sources (solar, wind, or geothermal), MD has been considered a valid desalination operation in arid areas with abundant solar energy available. However, a lot of other interesting applications of MD have been explored due to less fouling tendency of the process and the potential to treat complex feed solutions. Separation of azeotropic aqueous mixtures such as alcohol–water mixtures, concentration of radioactive solutions and application for nuclear desalination, wastewater treatment in which a less hazardous waste can be discharged to the environment specially in textile waste treatment that is contaminated with dyes, concentration of coolant (glycol) aqueous solutions, treatment of humic acid solutions, pharmaceutical wastewater treatment, and in areas where high-temperature applications lead to degradation of process fluids can be attractive (Banat and Simandl 1996). In addition, due to the chemical stability of the employed membranes, MD can also be applied for

2

Membrane Distillation Applications

Membrane Distillation Applications, Table 1 Applications of MD mentioned in different recent studies (Reprinted with permission from Drioli et al. 2015, Elsevier) Feed Seawater Simulated water

Target Boron removal Chromium removal

Used membranes PVDF PTFE

MD configuration DCMD DCMD

Produced water

Desalination

PTFE

DCMD

Aqueous solution of N-methyl-2-pyrrolidone

Concentration of N-methyl-2-pyrrolidone solution Desalination

PP

VMD

References Hou et al. (2013) Bhattacharya et al. (2014) Singh and Sirkar (2012) Shao et al. (2014)

FS PP

DCMD

Yu et al. (2013)

Removal of ammonia

PVDF capillary

Qu et al. (2013)

FS PTEF

Produced water

Concentration of phenolic compounds Desalination

DCMD and MDCMD DCMD

FS PTFE

AGMD

Model lactose solution

Ethanol production

PP capillary membrane

DCMD

Synthetic solution of trace OC

FS PTFE

DCMD

PP hollow fiber

Multi-effect MD DCMD

Li et al. (2012)

Water from Great Salt Lake

Removal of complex trace organic compounds Concentration of H2SO4 solution Improvement of water RF and salt crystallization Recovery of minerals

Alkhudhiri et al. (2013) Tomaszewska and Białon´czyk (2013) Wijekoon et al. (2014)

FS PTFE and PP

DCMD

Zabłocka Thermal Brine

Brine concentration

DCMD

Glycerol fermentation broth

Separation of acetic acid from the broth Removal of radioactive elements

PP hollow fiber accrual Accurel PP hollow fiber Hydrophobically modified FS PS or PES PP and PVDF FS

Hickenbottom and Cath (2014) Gryta (2013)

Cooling tower blow-down water Aqueous ammonia solution Olive oil waste mill water

Aqueous H2SO4 solution Retentate of NF and RO

Synthetic radioactive wastewater Wastewater containing arsenic in different concentrations

Removal of arsenic

the concentration of acids (Tomaszewska 1993; Tomaszewska et al. 1995; Tang et al. 2003). A list of innovative and potential uses of MD for various applications mentioned in the recent literature is provided in Table 1. More recently membrane distillation has been also used in membrane bioreactor configuration

PVDF hollow fibers

DCMD DCMD

VMD

Xie et al. (2013)

Tun and Groth (2011)

Gryta et al. (2013) Khayet (2013)

Criscuoli et al. (2013)

(MDBR) for the treatment of industrial and municipal used waters, in order to retain effectively small size and persistent contaminants (Phattaranawik et al. 2008). The possibility to combine MD with the production of high-quality crystals extracted from the brine of nanofiltration (NF) and reverse osmosis (RO) is particularly

Membrane Distillation Applications

interesting and promising as suggested in Drioli et al. (2011), Macedonio et al. (2007), and Macedonio and Drioli (2010). Moreover, membrane distillation can be also utilized in integrated system with pressure retarded osmosis (PRO) or reverse electrodialysis (RED) for utilizing salinity gradient for energy production (the so-called blue energy). An example can be found in the Megaton project (Kurihara and Hanakawa 2013) in Japan and in the SeaHero project (Kim et al. 2009, 2011) in South Korea. In the last part of these two projects, hybrid systems with MD and PRO units have been proposed for the extraction of valuable resources from the brine, the minimization of the environmental impact of the brine, and the recovery of energy.

List of Symbols AGMD Air gap membrane distillation DCMD Direct contact membrane distillation FS PP Flat sheet polypropylene FS PS or PES Flat sheet polysulfone or polyethersulfone FS PTEF Flat sheet polytetrafluoroethylene MDCMD Modified direct contact membrane distillation PP and PVDF FS Polypropylene and polyvinylidene difluoride flat sheet PP Polypropylene PTFE Polytetrafluoroethylene PVDF Polyvinylidene difluoride VMD Vacuum membrane distillation

References Alkhudhiri A, Darwish N, Hilal N (2013) Produced water treatment: application of air gap membrane distillation. Desalination 309:46–51 Banat FA, Al-Shannag M (2000) Recovery of dilute acetone–butanol–ethanol (ABE) solvents from aqueous solutions via membrane distillation. Bioprocess Eng 23(6):643–649 Banat FA, Simandl J (1996) Removal of benzene traces from contaminated water by vacuum membrane distillation. Chem Eng Sci 51(8):1257–1265

3 Banat FA, Simandl J (1999) Membrane distillation for dilute ethanol separation from aqueous streams. J Membr Sci 163:333–348 Banat FA, Simandl J (2000) Membrane distillation for propane removal from aqueous streams. J Chem Technol Biotechnol 75(2):168–178, AGMD Bandini S, Sarti GC (1999) Heat and mass transfer resistances in vacuum membrane distillation per drop. AIChE J 45(7):1422–1433 Bandini S, Saavedra A, Sarti GC (1997) Vacuum membrane distillation: experiments and modeling. AIChE J 43(2):398–408 Bhattacharya M, Dutta SK, Sikder J, Mandal MK (2014) Computational and experimental study of chromium (VI) removal in direct contact membrane distillation. J Membr Sci 450:447–456 Criscuoli A, Bafaro P, Drioli E (2013) Vacuum membrane distillation for purifying waters containing arsenic. Desalination 323:17–21 Drioli E, Calabro` V, Wu Y (1986) Microporous membranes in membrane distillation. Pure Appl Chem 58(12):1657–1662 Drioli E, Criscuoli A, Macedonio F (eds) (2011) Membrane-based Desalination: an Integrated Approach (MEDINA). Iwa Publishing, London Drioli E, Ali A, Macedonio F (2015) Membrane distillation: recent developments and perspectives. Desalination 356:56–84. doi:10.1016/j.desal.2014.10.028 Duan SH, Ito A, Ohkawa A (2001) Removal of trichloroethylene from water by aeration, pervaporation and membrane distillation. J Chem Eng Jpn 34(8):1069–1073 El-Bourawi MS, Ding Z, Ma R, Khayet M (2006) A framework for better understanding membrane distillation separation process. J Membr Sci 285:4–29 Garcia-Payo MC, Izquierdo-Gil MA, Fernandez-Pineda C (2000) Air gap membrane distillation of aqueous alcohol solutions. J Membr Sci 169:61–80 Gryta M (2013) The concentration of geothermal brines with iodine content by membrane distillation. DES 325:16–24 Gryta M, Markowska-Szczupak A, Bastrzyk J, Tomczak W (2013) The study of membrane distillation used for separation of fermenting glycerol solutions. J Membr Sci 431:1–8 Hickenbottom KL, Cath TY (2014) Sustainable operation of membrane distillation for enhancement of mineral recovery from hypersaline solutions. J Membr Sci 454:426–435 Hou D, Dai G, Wang J, Fan H, Luan Z, Fu C (2013) Boron removal and desalination from seawater by PVDF flatsheet membrane through direct contact membrane distillation. Desalination 326:115–124 Khayet M (2013) Treatment of radioactive wastewater solutions by direct contact membrane distillation using surface modified membranes. Desalination 321:60–66 Khayet M, Godino MP, Mengual JI (2003) Theoretical and experimental studies on desalination using the

4 sweeping gas membrane distillation. Desalination 157:297–305 Khayet M, Mengual JI, Zakrzewska-Trznadel G (2006) Direct contact membrane distillation for nuclear desalination Part II. Experiments with radioactive solutions. Int J Nucl Desalinat (IJND) 56:56–73 Kim S, Cho D, Lee M-S, Oh BS, Kim JH, Kim IS (2009) SEAHERO R%26D program and key strategies for the scale-up of a seawater reverse osmosis (SWRO) system. Desalination 238(1–3):1–9 Kim S, Oh BS, Hwang M-H, Hong S, Kim JH, Lee S, Kim IS (2011) An ambitious step to the future desalination technology: SEAHERO R%26D program (2007–2012). Appl Water Sci 1(1–2):11–17 Kurihara M, Hanakawa M (2013) Mega-ton Water System: Japanese national research and development project on seawater desalination and wastewater reclamation. Desalination 308:131–137 Lawson KW, Lloyd DR (1996) Membrane distillation. I. Module design and performance evaluation using vacuum membrane distillation. J Membr Sci 120:111–121 Lawson KW, Lloyd DR (1997) Membrane distillation. J Membr Sci 124:1–25 Li X, Qin Y, Liu R, Zhang Y, Yao K (2012) Study on concentration of aqueous sulfuric acid solution by multiple-effect membrane distillation. Desalination 307:34–41 Macedonio F, Drioli E (2010) Hydrophobic membranes for salts recovery from desalination plants. Desalin Water Treat 18:224–234 Macedonio F, Curcio E, Drioli E (2007) Integrated membrane systems for seawater desalination: energetic and exergetic analysis, economic evaluation, experimental study. Desalination 203:260–276 Mengual JI, Pena L (1997) Membrane distillation. Colloid Interf Sci 1:17–29 Phattaranawik J, Fane AG, Pasquier ACS, Bing W (2008) A novel membrane bioreactor based on membrane distillation. Desalination 223(1–3):386–395 Qu D, Sun D, Wang H, Yun Y (2013) Experimental study of ammonia removal from water by modified direct contact membrane distillation. Desalination 326:135–140 Qureshi N, Meagher MM, Hutkins RW (1994) Recovery of 2,3-butanediol by vacuum membrane distillation. Sep Sci Technol 29(13):1733–1748 Sarti GC, Gostoli C, Bandini S (1993) Extraction of organic-compounds from aqueous streams by vacuum membrane distillation. J Membr Sci 80:21–33

Membrane Distillation Applications Shao F, Hao C, Ni L, Zhang Y, Du R, Meng J, Liu Z, Xiao C (2014) Experimental and theoretical research on N-methyl-2-pyrrolidone concentration by vacuum membrane distillation using polypropylene hollow fiber membrane. J Membr Sci 452:157–164 Singh D, Sirkar KK (2012) Desalination of brine and produced water by direct contact membrane distillation at high temperatures and pressures. J Membr Sci 389:380–388 Sudoh M, Takuwa K, Iizuka H, Nagamatsuya K (1997) Effects of thermal and concentration boundary layers on vapor permeation in membrane distillation of aqueous lithium bromide solution. J Membr Sci 131:1–7 Tang JJ, Zhou KG, Zhao FG, Li RX, Zhang QX (2003) Hydrochloric acid recovery from rare earth chloride solutions by vacuum membrane distillation (1) Study on the possibility. J Rare Earths 21:78–82 Tomaszewska M (1993) Concentration of the extraction of fluid from sulphuric acid treatment of phosphogypsum by membrane distillation. J Membr Sci 78:277–282 Tomaszewska M, Białon´czyk L (2013) Production of ethanol from lactose in a bioreactor integrated with membrane distillation. Desalination 323:114–119 Tomaszewska M, Gryta M, Morawski AW (1995) Study on the concentration of acids by membrane distillation. J Membr Sci 102:113–122 Tun CM, Groth AM (2011) Sustainable integrated membrane contactor process for water reclamation, sodium sulfate salt and energy recovery from industrial effluent. Desalination 283:187–192 Wijekoon KC, Hai FI, Kang J, Price WE, Cath TY, Nghiem LD (2014) Rejection and fate of trace organic compounds (TrOCs) during membrane distillation. J Membr Sci 453:636–642 Xie M, Nghiem LD, Price WE, Elimelech M (2013) A forward osmosis-membrane distillation hybrid process for direct sewer mining: system performance and limitations. Environ Sci Technol 47(23):13486–13493 Yu X, Yang H, Lei H, Shapiro A (2013) Experimental evaluation on concentrating cooling tower blowdown water by direct contact membrane distillation. Desalination 323:134–141 Zolotarev PP, Ugrosov VV, Volkina IB, Nikulin VN (1994) Treatment of waste-water for removing heavy-metals by membrane distillation. J Hazard Mater 37(1):77–82

M

Membrane Distillation Bioreactor (MDBR) Tony Fane School of Chemical Engineering, UNSW, The University of New South Wales, Sydney, NSW, Australia Singapore Membrane Technology Centre, Nanyang Technological University, Singapore

The Membrane Distillation Bioreactor (MDBR) combines a wastewater bioreactor with membrane distillation (MD) (Fane et al. 2012). It is shown schematically in Fig. 1. The MDBR is a member of the Membrane Bioreactor (MBR) family but with important differences from the conventional MBR. The conventional MBR uses microporous membranes, either microfiltration (MF) or ultrafiltration (UF), to retain biomass, or mixed liquor, within the reactor. The MDBR employs a MD membrane to retain the mixed liquor and provide the treated wastewater. The MDBR membranes can be either submerged in the bioreactor or located in a sidestream. An elevated temperature, typically 50–60 C, is used to drive water across the membrane and this means that thermophilic biomass are required. In addition the hydrophobic membrane retains nonvolatile salts that accumulate in the reactor. Therefore the MDBR biomass need to be salt tolerant (halotolerant). The MDBR concept provides a “high retention” MBR, wherein # Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_363-1

low molecular weight molecules do not leave the reactor through the membrane, as they would through MF or UF membranes in the conventional MBR. This means that the organic retention time (ORT) in the MDBR can be significantly longer than the hydraulic retention time (HRT) and recalcitrant organics are given longer to degrade. In contrast the conventional MBR has ORT equal to HRT. The MDBR, in common with other MBRs, typically has a small “bleed”, or wastage, stream to maintain a steadystate concentration of mixed liquor suspended solids (MLSS). The proportion wasted daily determines the “sludge retention time” (SRT); for example 5 % removal per day gives SRT of 20 days, 10 % removal gives SRT of 10 days etc. The accumulation of nonvolatile and nonbiodegradable species in the reactor is determined by the ratio of SRT/HRT, which represents the concentration factor and could exceed 20x. Retained salts can reach significant levels, but with acclimatization the biomass can cope. The MDBR can provide a high-quality treated water, suitable for reuse. This is due to the unique separation properties of the typical MD membrane. In common with other membrane processes, the MDBR can experience fouling, and this can be controlled by bubbling and occasional cleaning. Experience indicates that the fouling is usually a combination of inorganic scale (calcium salts) and biofouling. Fluxes of the order of 10 l/m2h can be sustained.

2

Membrane Distillation Bioreactor (MDBR) W/Water Negligible TOC Waste or Solar Heat

QIN

QOUT

Biomass (thermophilic)

MD Module FS or HF

S

Air

Membrane Distillation Bioreactor (MDBR), Fig. 1 Membrane distillation bioreactor

The MDBR is a developmental concept and not widely used. However it has been demonstrated in the petrochemical industry (Khaing et al. 2010) providing a highly treated water, and driven by waste heat. Other target industries include food and pharmaceutical wastewaters. Another opportunity could be water reclamation where an analysis shows a 30 % lower electrical energy usage than a combined MBR and RO process. Recently the MDBR has been operated under anaerobic conditions with the generation of biogas.

References Fane AG, Phattaranawik J, Wong FS (2012) Contaminated inflow treatment with membrane distillation bioreactor. US Patent 8,318,017B2 (PCT June 2006) Khaing T-H, Li J, Li Y, Wai N, Wong FS (2010) Feasibility study on petrochemical wastewater treatment and reuse using a novel submerged membrane distillation bioreactor. Sep Purif Technol 74:138–143

F

Fouling Lidietta Giorno and Napoleone D’Agostino Institute on Membrane Technology, National Research Council of Italy, ITM-CNR, University of Calabria, Rende(CS), Calabria, Italy

Synonyms Fouling in membranes; Membrane fouling An intrinsic phenomenon of all membrane separation processes is the decline of the flux through the membrane as a function of time, due to concentration polarization effects and the formation of cake or gel layers by feed solution constituents retained by the membrane. Equally devastating for the performance of a process is membrane fouling. Membrane fouling is a general term. It may be the result of concentration polarization; it may also be the consequence of adsorption of feed solution constituents at the membrane surface and especially in microfiltration also within the membrane structure. The control of concentration polarization and membrane fouling is a major engineering aspect in the design of membrane separation processes and equipment (Brian 1966; Blatt et al. 1970; Jonsson and Boesen 1984; Bian et al. 2000; Hoek and Elimelech 2003; Cornelissen et al. 2004). Concentration polarization effects will occur in all membrane separation processes. Its # Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_366-1

consequences, however, are especially severe in pressure-driven membrane processes. When in a mass separation procedure a molecular mixture is brought to a membrane surface, some components will permeate the membrane under a given driving force, while others are retained. This leads to an accumulation of retained material and to a depletion of the permeating components in the boundary layers adjacent to the membrane surface. This phenomenon is referred to as concentration polarization. The causes and consequences of concentration polarization may be rather different in different membrane processes. Often the adverse effects of concentration polarization are intensified by an adsorption of certain feed mixture constituents at the membrane surface. This phenomenon is referred to as membrane fouling. Concentration polarization can be minimized by hydrodynamic means such as the feed flow velocity and the membrane module design. The control of membrane fouling, however, is more difficult. The transition between concentration polarization and fouling can be expressed by the concept of “critical flux.” This concept was introduced in the 1990s (Field et al. 1995; Howell 1995; Bacchin 2004), and it is the flux value below which a decline of flux with time does not occur, above it fouling is observed. Critical flux values depend on numerous parameters, such as properties of solutions to be treated (particle size, concentration, ionic strength, pH), surface interactions (zeta potential, surface tension), and

2 pure water flux flux

Fouling, Fig. 1 Schematic diagram illustrating the difference between the flux decline due to concentration polarization and due to membrane fouling

Fouling

concentration polarization

membrane fouling

time

hydrodynamic conditions (axial velocity) (Bromley et al. 2002; Chen 1998; Ingmar et al. 1999; Belfort et al. 1994). When operating below the critical flux, it is possible to observe a linear correlation between flux and transmembrane pressure. Above it, further increase in transmembrane pressures lead to additional layer deposit on the membrane surface, until a point where the deposit fully compensates the increase in pressure drop. At this stage the liming flux is reached, which represents the maximum stationary permeation flux. After operating above the critical flux value, decreasing the transmembrane pressure will not lead to the previous flux behavior. However, experimental evidences show that, although operating at subcritical flux, gradual fouling develops in membrane materials, and it proves to be hydraulically irreversible after a long period experimentation (Ognier et al. 2004). The formation of a gel or cake layer is one cause for membrane fouling. Gel or cake layer formation may be caused by a variety of materials including inorganic precipitates such as CaSO4, Fe(OH), and other metal hydroxides; organic materials such as proteins, humic acids, and other macromolecular materials; and biological components such as microorganisms and products of their metabolism (biofouling). Membrane fouling may also occur without concentration polarization, i.e., a direct transport to the membrane surface. The attachment of the substances to the membrane surface may be caused by adsorption due to hydrophobic interactions, van der Waals force attractions, or electrostatic forces. The fouling layer itself may be rather

porous and thus permeable for aqueous solutions as some inorganic precipitants or highly impermeable as some films of mineral oils or hydrophobic surfactants. The fouling mechanism depends also on the membrane process. In electrodialysis fouling is caused mainly by the precipitation of polyelectrolytes or sparingly soluble salts such as CaSO4 or CaCO3. Membrane fouling in electrodialysis affects mainly the anionexchange membranes because most of the colloidal and macromolecular polyelectrolytes present in natural waters such as humic acids or proteins are negatively charged. In ultra- and microfiltration of biological solutions, but also in reverse osmosis of sea water, biological fouling is a severe problem affecting the economics of the processes. In biomedical applications protein adsorption and protein denaturation at the membrane surface is often impairing the performance of the membranes. In membrane distillation, adsorption of molecules can change surface energy and hydrophobic properties. The difference between concentration polarization and membrane fouling or scaling is illustrated schematically in Fig. 1. Concentration polarization is a reversible process based on diffusion and takes place over a few seconds; it can be described adequately by a simple mathematical model and easily be controlled by the proper process design. Fouling is generally irreversible and the flux decline takes place over many minutes, hours, or even days. A constant flux is generally not reached at all. Membrane fouling is more difficult to describe and to control by experimental means. Membrane fouling is determined by a variety of different

Fouling

3

Methods to reduce concentration polarization

Reduce pressure

Mixing perpendicular to membrane

Paddle mixers

Static mixers

High velocity gradient Move membrane

Move liquid

Reduce concentration at membrane surface

Low concentration factor

Reduce solids in feed

Remove concentration at membrane surface

Boundary layer skimming

Increase diffusivity

Increase mass transfer of solids back

Mechanical scouring

Thin channels

Short flow length

Increase temperature

Fouling, Fig. 2 Methods to reduce concentration polarization

parameters including the feed solution constituents and their concentration, membrane material, and the fluid dynamic system design. Membrane fouling can be caused by simple precipitation of insoluble materials or reversible and irreversible adsorption of components at the membrane surface and within the membrane pores. The means of preventing or at least controlling membrane fouling effects are as heterogeneous as the different material and mechanisms causing the fouling (Ridgway et al. 1983; Nilson 1998). The main procedures to avoid or control concentration polarization and fouling involve: • Pretreatment of the feed solution • Membrane surface modifications • Hydrodynamic optimization of the membrane module • Membrane cleaning with the proper chemical agents A list of various methods to reduce concentration polarization is reported in Fig. 2. A pretreatment of the feed solution may include chemical precipitation, prefiltration, pH adjustment, chlorination, or carbon adsorption. In some membrane module, design concepts as, for

instance, in hollow fiber modules, the elimination of all particulate materials is of great importance for the proper function of the membrane. Membrane surface modifications include the introduction of hydrophilic moieties or charged groups in the membrane surface by chemical means or plasma deposition. Increasing the shear rate imposed by the feed solution on the membrane surface will in many cases reduce the membrane fouling. High feed flow velocities and the proper module design are efficient tools in controlling membrane fouling (Fig. 3). When in spite of an adequate membrane and module design the membrane flux is decreasing with operation time to an unacceptable low value, it is necessary to clean the membrane to restore the flux in part or completely. Typical cleaning agents are acids and bases, such as HNO3 and NaOH, complexing agents, enzymes, and detergents. Another very effective method to minimize the effects of membrane fouling in microfiltration is backflushing (Fig. 4). In backflushing, the applied pressure is reversed and the permeate is pushed through the membrane, lifting off fouling material that had been precipitated on the feed side membrane surface and washing it out of the filtration device.

4

Fouling

Fouling, Fig. 3 Effect of axial velocity on concentration polarization Fouling, Fig. 4 Backflushing method to reduce fouling

Suspension

Suspension

Backflushing Jv With backflushing

Without backflushing t

Backflushing is done in certain time intervals for a couple of seconds.

References Bacchin P (2004) A possible link between critical and limiting flux for colloidal systems: consideration of critical deposit formation along a membrane. J Membr Sci 235:111–122 Belfort G, Davis RH, Zydney AL (1994) The behaviour of suspensions and macromolecular solution in crossflow microfiltration. J Membr Sci 98:1–58 Bian R, Yamamoto K, Watanabe Y (2000) The effect of shear rate on controlling the concentration polarization and membrane fouling. Desalination 131:225–236 Blatt WF, Dravid A, Michaels AS, Nelsen LM (1970) Solute polarization and cake formation in membrane

ultrafiltration: cause, consequences and control techniques. In: Flinn JE (ed) Membrane science and technology. Plenum Press, New York, pp 47–97 Brian PTL (1966) Mass transport in reverse osmosis. In: Merten U (ed) Desalination by reverse osmosis. MIT Press, Cambridge, MA, p 161 Bromley AJ, Holdich RG, Cumming IW (2002) Particulate fouling of surface microfilters with slotted and circular pore geometry. J Membr Sci 196:27 Chen V (1998) Performance of partially permeable microfiltration membranes under low fouling conditions. J Membr Sci 147:265 Cornelissen ER, Harmsen D, de Korte KF, Ruiken CJ, Qin J-J, Oo H, Wessels LP (2004) Membrane fouling and process performance of forward osmosis membranes on activated sludge. J Membr Sci 319:158–168 Field RW, Wu D, Howell JA, Gupta BB (1995) Critical flux concept for microfiltration fouling. J Membr Sci 100:259

Fouling Hoek EMV, Elimelech M (2003) Cake-enhanced concentration polarization: a new fouling mechanism for saltrejecting membranes. Environ Sci Technol 24:5581–5588 Howell J (1995) Subcritical flux operation of microfiltration. J Membr Sci 107:165 Ingmar HH, Eert V, Gun T, Christian T (1999) The influence of the membrane zeta potential on the critical flux for crossflow microfiltration of particle suspensions. J Membr Sci 156:153 Jonsson G, Boesen CE (1984) Polarization phenomena in membrane process. In: Belfort G (ed) Synthetic membrane processes. Academic, New York

5 Nilson JL (1998) Fouling of an ultrafiltration membrane by dissolved whey protein concentrate. J Membr Sci 36:147 Ognier S, Wisniewski C, Grasmick A (2004) Membrane bioreactor fouling in sub-critical filtration conditions: a local critical flux concept. J Membr Sci 229:171 Ridgway HF, Justice C, Kelly A, Olson BH (1983) Microbial fouling of reverse osmosis membranes used in advanced wastewater treatment technology: chemical, bacteriological and ultrastructural analyses. Appl Environ Microbiol 45:1066

M

Membrane Reactor Equilibrium Conversion Giuseppe Barbieri Institude on Membrane Technology, Italian National Research Council, Rende(CS), Italy

The equilibrium constant of any reaction is function of temperature only. The equilibrium of a conventional/traditional reacting system is function only of the thermodynamic variables (temperature, pressure, and mixture composition). It is independent from the reactor model (continuous stirred tank reactor, plug flow reactor, batch reactor, etc.), fluid dynamics (mixing, diffusion, etc.), heat exchange (isothermal, adiabatic, or no-isothermal reactor), reactor size, etc. since the equilibrium is reached when variation in temperature, pressure, and composition is no longer present in the chemical system. This condition is mathematically expressed by the following equations set valid for a reacting closed (no mass transfer with the environment) system consisting of N reactions. N Gas uij Kpj ¼ ∏ PReaction xi ðX1 , X2 ,::, XN Þ i¼1

j ¼ 1,::, N

(1)

Reaction

The formulation in terms of Kp constants enables for the evaluation of reaction conversion and # Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_368-1

hence the equilibrium composition of the reacting system kept under a set reaction pressure. As well known, the equilibrium expresses the highest limit for reaction conversion, and it will be here indicated as TREC (Traditional Reactor Equilibrium Conversion). The conversion of a real chemical reactor (of finite size) depends on several factors, e.g., reactor model, feed flow rate, residence time, and heat transfer with the environment in addition to the variables before indicated. Thus, its conversion is always lower than the equilibrium one. A membrane reactor (MR) has, in addition to the traditional reactor, another chamber/volume: the one named permeate. Therefore, it consists of two (reaction and permeation) volumes separated by a selective membrane. One can image each chamber is provided with a piston moving without friction, for keeping a constant pressure; consequently, the reaction and permeation volumes change according to the pistons movement. In addition, the reaction pressure can be different from the permeate pressure. During the reaction, the chamber volumes change for achieving the chemical and permeating equilibria. As the reactive equilibrium, the permeative one has a dynamic character such as the reactive one; this means molecules (if any) permeating the membrane from reaction side to permeate side are equal in number to the molecules permeating in the opposite direction. Temperature, pressure, and composition of the permeate can be different from those or reaction

2

Membrane Reactor Equilibrium Conversion

Membrane Reactor Equilibrium Conversion, Table 1 Thermodynamics table for water gas shift reaction (CO + H2O = CO2 + H2) in an MR. Both reaction and permeating equilibrium are taken into account Reaction Initial state Reaction conversion Reactive equilibrium state Permeation Reactive and permeative equilibrium state Total mole number

CO H2O CO2 n0 m n0 m1 n0 x n0 x n0 x n0 n0 (1x) n0 (mx) n0 (m1 + x) – – – n0 (1x) n0 (mx) n0 (m1 + x) n0[1 + m + m1 + m2 z (m2 + x)]

H2 m2 n0 x n0 n0 (m2 + x) z [n0 (m2 + x)] n0[(m2 + x) (1z)]

Where, m > =x; x = MREC; z permeate hydrogen fraction m, m1, and m2 are the feed molar ratio with respect to CO of H2O, CO2, and H2, respectively

volume. These variables have also to be taken into account for evaluating the equilibrium of an MR. Nevertheless, another constrain is required for the equilibrium of an MR in addition to the one defined by Eq. 1 for the traditional reactor.

This is the permeation equilibrium expressed as the equality of the chemical potentials (e.g., partial pressures for a gas phase reaction) (Barbieri et al. 2001, 2005; Marigliano et al. 2003) on both membrane sides of any permeable species Eq. 2.

ChemicalPotentialEquilibrium ¼ ChemicalPotentialReaction ¼ ChemicalPotentialPermeation i i i Equilibrium Reaction Permeation Pressurei ¼ Pressurei ¼ Pressurei 8 permeable species

This condition is independent from the membrane type and permeation rate. As chemical equilibrium in a conventional reactor, and in an MR as well, is independent from the reaction path, similarly the MR equilibrium is independent from permeation rate. The hydrogen permeates Pd-based membranes following the Sieverts law. Even though, the membrane characteristics (e.g., Pd, Pd-Ag, thickness, etc.) influence the permeation rate and hence the time necessary to reach the equilibrium, the value of these parameters does not affect the MR equilibrium. The equilibrium depends on the extraction capacity of the membrane system which is a function of the temperature, pressure, and composition of the MR permeate volume that directly gives the condition of Eq. 2. The MREC (Membrane Reactor Equilibrium Conversion) can be easily calculated using the constrains of Eqs. 1 and 2. MREC can be also calculated using the thermodynamics table for

(2)

the reaction(s) adding to it the terms related to the permeative equilibrium. CO þ H2 O ¼ CO2 þ H2

DH0298

¼ 41 kJ mol1 Table 1 reports the thermodynamics table for calculating the equilibrium conversion of the water gas shift reaction in a Pd-based MR, all the figures are in moles. The procedure for calculating the number of moles at the equilibrium is the one known since long time from chemistry as the fourth row “reactive equilibrium state” of Table 1 shows. In fact, in absence of a membrane this conversion is the one of a traditional reactor and here is identified as TREC. The presence of a selective membrane contributes to product (hydrogen) removal and hence to increase the equilibrium conversion because of the mass law effect or the Le Ch^atelier–Braun principle. Therefore, the equilibrium conversion of a

Membrane Reactor Equilibrium Conversion

3

Membrane Reactor Equilibrium Conversion, Table 2 Number of moles determination at reactive and permeative equilibrium in a membrane reactor Moles Initial state Reactive state of reaction 1

n-butane n0 n0 x

Reactive state of reaction 2

n-butene – n0x

Isobutene – –

n0 x w

n0 x w

Hydrogen – n0 x

Permeation state

n0 x z

Permeative and reactive equilibrium state

n 0 ð 1 xÞ

n0 x ð 1 w Þ

n0 x w

n0 x ð1 zÞ

n0 ð1 þ x xzÞ

Total number of moles

Where, x = MREC for reaction 1, w = MREC for reaction 2, z permeate hydrogen fraction The initial concentration of n-butene, isobutene, and hydrogen was chosen as zero

reactor conversion (here named MREC) is always higher than TREC. MREC is equal to TREC in absence of any permeation. The fifth row includes the term (z) for hydrogen permeation through a Pd-based membrane, whereas the sixth row reports the moles present in the reaction volume under the reactive and permeative equilibrium condition. The total mole number in an MR owing to permeation is lowered by the reducing term {z [n0 (m2 + x)]} with respect to traditional reactor. The equilibrium constant of the reaction (reactive constrain, see Eq. 1) for WGS reaction considering the reactive and permeative equilibrium terms can be expressed as KP ¼

ðm1 þ xÞ½ðm2 þ xÞð1 zÞ ð1 x Þðm x Þ

n butane ¼ n butene þ H2 n butene ¼ isobutene

The other, permeative constrain Eq. 2 for the hydrogen, the only permeating species is ¼ PEquilibrium H2 n0 ½ðm2 þ xÞð1 zÞ PReaction n0 ½1 þ m þ m1 þ m2 zðm2 xÞ CO conversion and the hydrogen fraction permeate the membrane are calculated resolving this set of two equation in two unknowns: the conversion (MREC) and the hydrogen permeated the membrane. Another example of using thermodynamics table is reported for the n-butane dehydroisomerization (Al-Megren et al. 2013). This reaction (reaction 3) involves the dehydrogenation of n-butane (reaction 1) and successive isomerization to isobutene (reaction 2). The main products measured were normal butenes, isobutane, and isobutene.

DHReaction ð@25 CÞ ¼ 130 kJ mol1

DHReaction ð@25 CÞ ¼ 17 kJ mol1

n butane ¼ isobutene þ H2

DHReaction ð@25 CÞ ¼ 1137 kJ mol1

From the thermodynamics table (Table 2), after some algebra, the equilibrium constants for

(reaction1) (reaction2) (reaction3)

reaction 1 and reaction 2 and the permeation equilibrium condition are:

4

Membrane Reactor Equilibrium Conversion

K P1 ¼

x2 ð1 zÞð1 wÞ Reaction side P ð1 þ x xzÞð1 xÞ K P2 ¼

w ð1 wÞ

PEquilibrium ¼ PReaction side hydrogen

x ð1 zÞ ð1 þ x xzÞ

This is a set of three equation in three unknowns that resolved for x, w, and z gives the conversion values (MRECs) of the reaction 1 and reaction 2 and the permeate hydrogen fraction.

References Al-Megren HA, Barbieri G, Mirabelli I, Brunetti A, Drioli E, Al-Kinany MC (2013) Direct conversion of n-butane to isobutene in a membrane reactor: a thermodynamic analysis. Ind Eng Chem Res 52:10380–10386. doi:10.1021/ie400006c Barbieri G, Marigliano G, Perri G, Drioli E (2001) Conversion–temperature diagram for a palladium membrane reactor. Analysis of an endothermic reaction: methane steam reforming. Ind Eng Chem Res 40:2017–2026. doi:10.1021/ie0006211 Barbieri G, Brunetti A, Granato T, Bernardo P, Drioli E (2005) Engineering evaluations of a catalytic membrane reactor for water gas shift reaction. Ind Eng Chem Res 44:7676–7683. doi:10.1021/ie050357h Marigliano G, Barbieri G, Drioli E (2003) Equilibrium conversion for a palladium membrane reactor. Dependence of the temperature and pressure. Chem Eng Process 42:231–236. doi:10.1016/S0255-2701(02) 00092-2

M

Membrane Wettability

gLW ¼ s

Annarosa Gugliuzza Institute on Membrane Technology - Research National Council, ITM-CNR, Rende (CS), Italy

The concept of wettability (S) is referred to the ability of a liquid to wet a surface (Hsu et al. 2011). The degree of wettability, also known as a wetting or spreading, is controlled by interfacial forces established between a liquid (L), solid (S), and vapor (V) phase at the minimum equilibrium distance. These interaction types are usually expressed as surface tensions (g) and can be correlated to the wettability coefficient by a mathematical equation: S ¼ gSV ðgSL þ gLV Þ

(1)

The surface tension, expressed as a force per unit length [N/m] or energy per unit surface area [J/m2], is a direct measurement of the cohesive energy required for minimizing the free surface area at the interface, where the number of similar species is more restricted than bulk and attractive forces are unequal and unbalanced in all directions (Fig. 1). This parameter and related polar and nonpolar components can be estimated by a set of mathematical equations (Good and Van Oss 1992):

# Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_370-6

gLW ð1 þ cos YÞ2 l 4

(2)

gl ð1 þ cos YÞ ¼ 2 qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffi þ LW LW þ gs gl þ gs gl þ gs gl (3) gAB s ¼ 2

pffiffiffiffiffiffiffiffiffiffi þ gþ s gs

(4)

where the Lifshitz–van der Waals forces gLW[mJ/m2], polar gAB[mJ/m2], acid (electron acceptor) g+[mJ/m2], and base (electron donor) g[mJ/m2] are the components of the surface free energy g [mJ/m2] and can be calculated from mean values of the apparent contact angle (ya), expressed in a somewhat simplified form by the Young’s equation: cos ðYa Þ ¼

gSV gSL gLV

(5)

It is a convention to consider y = 90 as a boundary value between hydrophilicity and hydrophobicity. Values less than 90 indicate hydrophilic membrane surfaces with a positive liquid spreading, whereas angle values higher than 90 are measured on hydrophobic membrane surfaces with negative spreading values (Fig. 2). Rolling water droplets can be observed on superhydrophobic surfaces with y > 150 and S Feþ2 þ 2e :

2

Corrosion

Corrosion, Fig. 1 The mechanism of corrosion

Both water and oxygen are required for the sequence of reactions. Iron (+2) ions are further oxidized to form ferric ions (iron+3) ions:

that rusting occurs much more rapidly in moist environment as compared to dry one. The mechanism of corrosion is shown in Fig. 1 below.

Feþ2 > Feþ3 þ 1e

Influence of Salt on Corrosion The presence of salt enhances the rate of corrosion. As the dissolved salt increases, the conductivity of the aqueous solution at the metal surface enhances the rate of electrochemical corrosion. Hence iron and steel corrode more rapidly when exposed to salt or moist salty air near the ocean.

The electrons provided from both oxidation steps are used to reduce oxygen as shown. Cathodic reactions involve electrochemical reduction wherein electrons appear on the left side: O2 ðgÞ þ 2 H2 O þ 4e_

> 4 OH

The ferric ions then combine with oxygen to form ferric oxide [iron (III) oxide] which is then hydrated with varying amounts of water. The reduction of oxygen at an electrode will cause a rise in pH due to production of hydroxide ion. The equation for rust formation is 4Feþ2 ðaqÞ þ O2 ðgÞ þ ½4 þ 2 H2 OðlÞ > 2Fe2 O3 H2 O þ 8 Hþ ðaqÞ : The formation of rust can occur at some distance from the actual pitting or erosion of iron. The electrons produced via the initial oxidation of iron can be conducted through the metal, and the iron ions can diffuse through the water layer to another point on the metal surface where oxygen is available. This process results in an electrochemical cell in which iron serves as anode, oxygen gas as the cathode, and iron solution as “salt bridge”. The involvement of water accounts for the fact

Factors Associated Mainly with the Metal • Effective electrode potential of a metal in a solution • Overvoltage of hydrogen on the metal • Chemical and physical homogenity of the metal surface • Inherent ability to form an insoluble protective film Factors Varying with the Environment • pH of the solution • Influence of oxygen in solution adjacent to the metal Forms of Corrosion Galvanic Corrosion

This is an electrochemical action of two dissimilar metals in the presence of an electrolyte and an electron conductive path. It occurs when

Corrosion

3

dissimilar metals are in contact. It is recognizable by the presence of a buildup of corrosion at the joint between the dissimilar metals. When aluminum or magnesium alloys are in contact with carbon or stainless steel, galvanic corrosion occurs and accelerates the corrosion of aluminum or magnesium (Fig. 2). Galvanic Series in Sea Water (Least Active) Gold Graphite Silver 18-8-3 stainless steel, type 316 (passive) 18-8 stainless steel, type 304 (passive) Titanium Thirteen percent Cr stainless steel, type 410 (passive) 7 Ni- 33 Cu alloy 75Ni – 33 Cu alloy 75Ni – 16 Cr – 7 Fe alloy (passive) Nickel (passive) Silver solder M-Bronze G- Bronze 70-30 cupro-nickel Silicon bronze Copper Red brass Aluminum bronze Admiralty brass Yellow brass 76 Ni-16 Cr-7 Fe alloy (active) Nickel (active) Naval brass Manganese bronze Muntz Tin Lead 18-8-3 stainless steel, type 316 (active) 18-8 stainless steel, type 304 (active) Thirteen percent chrome stainless steel, type 410 (active) Cast iron Mild steel Aluminum 2024 Cadmium Aluminum 6053

Corrosion, Fig. 2 Galvanic corrosion

Galvanized steel Zinc Magnesium alloys Magnesium Anodic (Most Active) The natural differences in metal potentials produce galvanic differences, such as the galvanic series in sea water. If electrical contact is made between any two of these materials in the presence of an electrolyte, current must flow between them. The farther apart the metals are in the galvanic series, the greater the galvanic corrosion effect or the rate. Metals or alloys at the upper end are noble while those at the lower end are active. The more active metal is the anode or the one that will corrode. The anode must be chosen from the above material list which is lower on the list than the material to be protected. Control of galvanic corrosion is achieved by using metals closer to each other in the galvanic series or by electrically isolating metals from each other. Cathodic protection can be used to control galvanic corrosion effects. The following practices are recommended to keep galvanic corrosion to a minimum.

4

Corrosion

Corrosion, Fig. 3 Isolation flanges to prevent galvanic

• Avoid the use of widely dissimilar metals in direct contact. • When dissimilar metals must come into contact, they should be separated by using nonconductive barrier materials, a paint coating, or by plating. • The anode should be as large as feasible in relation to cathode. • Coat both the anode and the cathode with the same material. • Install fasteners that have been dipped in epoxy mastic coatings. • Seal threaded inserts with epoxy mastic coatings prior to insertion into castings. • Avoid the use of lock or toothed washers over plated or anodized surfaces. The scuba tank above suffered galvanic corrosion when the brass valve and the steel tank were wetted by condensation. Electrical isolation flanges (Fig. 3) like those shown on the right are used to prevent galvanic corrosion. Insulating polymeric gaskets are inserted between the flanges, and insulating sleeves and washers isolate the bolted connections.

deposit is cleaned away, tiny pits or holes can be seen in the surface. Passive metals such as stainless steel resist corrosive media and can perform well over long periods of time. However, if corrosion does occur, it forms at random pits. Pitting may be a serious type of corrosion because it tends to penetrate rapidly into the metal section. It is most likely to occur in the presence of chloride ions, combined with such depolarizers as oxygen or oxidizing salts. Methods that can be used to control pitting include maintaining clean surfaces, application of a protective coating, and use of inhibitors or cathodic protection for immersion service. Sometimes pitting corrosion can be quite small on the surface and very large below the surface. The figure below left shows this effect common on stainless steel and other filmprotected metals. Pitting shown on the right (white arrow) led to the stress corrosion fracture shown by the black arrows (Fig. 4). Pitting corrosion can lead to unexpected catastrophic system failure. The split tubing below left was caused by pitting corrosion of stainless steel. A typical pit on this tubing is shown below right (Fig. 5).

Pitting Corrosion

The most common effect of corrosion on aluminum and magnesium alloys is pitting. First is noticeable as a white or gray powder, similar to dust, which blotches the surface. When the

Stress Corrosion Cracking

Stress corrosion cracking (SCC) is caused by the simultaneous effects of tensile stress and corrosion. Stress may be internally or externally

Corrosion

5

Corrosion, Fig. 4 Same images for pitting corrosion

Corrosion, Fig. 5 Pitting corrosion for stainless steel tube

applied. Internal stresses are produced by nonuniform deformation during cold working, by an unequal cooling from high temperatures, and by internal structural rearrangement involving volume changes. Stresses induced when a piece is deformed, those induced by press and shrink fits, and those in rivets and bolts are internal stresses. Concealed stress is more important than design stress, especially because stress corrosion is difficult to recognize before it has overcome the design factor. Few guidelines in avoiding the problem are: • Use metal alloys at no greater than 75 % of their yield strength and use exotic materials only where they are actually required. • Avoid assemblies where high-tensile loads are concentrated in a small area.

• Remove stress risers from counter bores, grooves, etc. Crevice Corrosion

Crevice or contact corrosion is produced at the region of contact of metals with metals or metals with nonmetals. It may occur at washers, under barnacles, at sand grains, under applied protective films, and at pockets formed by threaded joints. Whether or not stainless steels are free of pit nuclei, they are always susceptible to this kind of corrosion because a nucleus is not necessary. Crevice corrosion may begin through the action of an oxygen concentration cell and continue to form pitting. Cleanliness, the proper use of sealants, and protective coatings are effective means of controlling this problem.

6

Intergranular Corrosion

This is an attack on the grain boundaries of a metal or alloy. A highly magnified cross section of any commercial alloy will show its granular structure. This structure consists of quantities of individual grains, and each of these tiny grains has a clearly defined boundary that chemically differs from the metal within the grain center. One example of this type of corrosion is in unstabilized 300-series stainless steels sensitized by welding or brazing and subsequently subjected to a severe corrosion environment. Another example of intergranular or grain boundary corrosion is that which occurs when aluminum alloys are in contact with steel in the presence of an electrolyte. The aluminum alloy grain boundaries are anodic to both the aluminum alloy grain and the steel. In the later case, intergranular corrosion of the aluminum alloy occurs. Some austenitic steels are unstable when heated in the temperature range 470–915 C. Decreased corrosion resistance in austenitic stainless steels is due to depletion of chromium in the area near the grain boundaries, caused by the precipitation of chromium carbide. This condition can be eliminated by the use of stabilized stainless steel, such as columbium-, tantalum-, or titanium-stabilized stainless steels (types 321/347), or by the use of low-carbon stainless steels. Molybdenum additions as in type 316 stainless steels decrease the sensitivity to and the severity of the intergranular attack. Intergranular corrosion can be prevented by: • Select an alloy type that is resistant to intergranular corrosion. • Avoid heat treatments or service exposure that makes a material susceptible. Normally this occurs with austenitic stainless steels when they are held for some time in the sensitizing temperature range of 470–915 C (800–1600 F). • Apply a protective coating. Corrosion Removal Abrasive blasting is the preferred method of removing corrosion; other mechanical methods

Corrosion

used are grinding, chipping, sanding, or wire brushing. Remove corrosion by mechanical method such as wire brushing or abrasive blasting as appropriate. Failure to adequately clean all residues will permit corrosion to continue. Light corrosion may be removed from thin members – ducts and tubing – with a nonwoven, nonmetallic abrasive mat in accordance with MIL-A-9962 or number 400 aluminum oxide or silicon carbide grit abrasive paper or cloth. Do not use steel wool. Chemical Method

Chemical corrosion-removal methods can be used when no danger exists that the chemical used will become entrapped in recesses and when there is no danger that adjacent material will be attacked. The chemical method should be used on complex shapes and machined surfaces. Chemical rust removers are of two types: acid or alkaline. The acid type is used in removing rust and black oxide by immersion or brush application. This is a phosphoric acid-type remover and must not be used on high-strength steel heat treated above 1.24 gigapascals (GPa) tensile strength due to possible stress corrosion or hydrogen embrittlement problems. The alkaline type (sodium hydroxide base) is suitable for use by immersion only. Use on machined surfaces where a dimensional change would be objectionable. Scale conditioners are used as necessary to facilitate oxide scale removal by acid cleaning. The use of scale conditioners shall not cause pitting, intergranular attack, or reduction of mechanical properties below the minimum values as specified in the applicable material specification for the alloy, gage, and heat-treat condition. When acid cleaning is used to remove heattreat scale, flux, corrosive media, stains, and other contaminants, it shall be within the limits specified. Acid cleaning shall not result in intergranular attack that would be detrimental to the fabrication or use of the material or part. Acid cleaning shall not result in pitting or smutting, which will not be readily removed by subsequent

Corrosion

processing. Cleaning shall be accomplished in the following bath: 1. Nitric acid (HNO3) (42 Baume) 225–375 kg/m3 2. Hydrofluoric acid (HF) or NH4HF2 9–52 kg/m3 3. Temperature: room 60 C (140 F) 4. Metal content, HF ratio = 1: 1.8 (replenish bath when the metal concentration >1 part of metal to 1.8 parts of HF) Stainless Steel Alloys

Stainless steels owe their inherent corrosion resistance to a condition known as passivity – as a result of the presence of their oxide films called “passive films.” Under favorable conditions, such films are protective; however, unfavorable conditions deficient in oxygen will destroy the films and leave the surface in an “active” state with corrosion resistance comparable to carbon steel. The presence of hygroscopic salt deposits, dirt, dust, and other foreign matter all serve to destroy passivity. Underground exposure of bare stainless steel will result in unacceptable quality. Where localized corrosion occurs, rapid penetration (pitting) at the point of initiation can occur as an electrolytic cell is formed between the large cathodic (passive) area and the small anodic area under attack. This attack is particularly severe in the presence of halide salts. Localized attack will also occur in crevices, such as under sleeves on tube fittings. Superior resistance to pitting is attainable with type 904 L unified numbering system (UNS) N08904 stainless steel over other commonly used steels. However, this is only a matter of degree and localized attack can still occur. Maintaining clean surfaces will greatly reduce the opportunity for corrosion, regardless of the alloy employed. Typical Problem Areas

• Sharp edges. Sharp edges of metal structures will often be deficient of proper coating thickness; sharp edges should be rounded when possible with the National Association of Corrosion Engineers (NACE). A stripe coat or

7

brush coat of primer prior to spray will assist in getting adequate coverage. • Nuts and bolts. Premature coating failure and corrosion on nuts and bolt heads are common and can be reduced by conscientious surface preparation prior to application of a protective coating. A brush coat of primer prior to spray will ensure adequate coverage. • Tube clamps. Carbon steel clamps for interior applications are either zinc plated or painted. Stainless clamps shall be used for all exterior applications. Corrosion at the interface between the stainless steel tubing and the clamp can occur due to dissimilar metal and crevice corrosion. Corrosion at the interface between tubing and clamps is controlled by application of protective coatings. • Unistrut channels. Use of unistrut channels should be avoided in exterior locations. When the exterior use of unistrut cannot be avoided, selection of appropriate material shall be utilized, such as stainless steel or fiber glass. Where corrosion is noted, mechanically clean the member to remove corrosion products followed by application of a zinc-rich coating. Stainless Steel Components

General Stainless steel, although very corrosion resistant, is susceptible to localized corrosion (e.g., pitting, crevice corrosion, etc.) when exposed to marine environment. A mistake frequently made is to conclude that the corrosion noted on stainless steel tubing and bellows is only superficial. This conclusion is improperly reached when removal of external corrosion products leaves the surface in a condition that appears almost like new except for what appears to be a very tiny pit. A cross section taken through such a typical pit frequently discloses a void considerably greater in diameter than the surface pit diameter. Failure to arrest the apparent superficial corrosion will result in ultimate penetration of thin wall members. Application of Protective Coatings

Stainless steel tubings shall be treated in the following way:

8

• Accumulated dirt and oil shall be removed by rinsing with water followed by rinsing with methyl ethyl ketone (MEK). • Remove corrosion products by mechanical means, such as power tool cleaning or handtool cleaning. • Clean surfaces with methyl ethyl ketone using clean rags. • Apply by spraying, brushing, or dipping 75 mm (3 mils) minimum of the coating Aerocoat AR-7 manufactured by B.F. Goodrich. Tubing assemblies shall be abrasive blasted. When tubing assemblies are in close proximity to carbon steel structural members that are to be abrasive blasted and coated with inorganic zincrich primer, the tubing assemblies shall be similarly treated. • Using clean rags, accumulated dirt and oil shall be removed with water followed by wiping with methyl ethyl ketone. • Apply a zinc-rich coating in accordance with the manufacturer’s recommendations to a DFT (dry film thickness) of 100–150 mm (4–6 mils). Cathodic Protection

Cathodic protection (CP) is a technique used to control the corrosion of a metal surface by making it the cathode of an electrochemical cell. CP is carried out by connecting the metal to be protected with a piece of another more easily corroded “sacrificial metal” to act as the anode of the electrochemical cell. The sacrificial metal then corrodes instead of the protected metal. For structures where passive galvanic CP is not adequate, like in long pipelines, an external DC electrical current is applied. Cathodic protection systems are used to protect a wide range of metallic structures in various environments. Common applications are steel water or fuel pipelines and storage tanks such as home water heaters, steel pier piles, ship and boat hulls, offshore oil platforms and onshore oil well casings, and metal reinforcement bars in concrete buildings and

Corrosion

structures. A common application is in galvanized steel, in which a sacrificial coating of zinc on steel protects them from rust. Cathodic protection can, in some cases, prevent stress corrosion cracking. Applications Pipelines are routinely protected by a coating supplemented with cathodic protection. An ICCP – impressed current cathodic protection – system for a pipeline would consist of a DC power source, which is often an AC-powered rectifier and an anode, or array of anodes buried in the ground (the anode ground head) Ships: cathodic protection on ships is often implemented by galvanic anodes attached to the hull, rather than using ICCP. Since ships are regularly removed from the water for inspections and maintenance, it is a simple task to replace the galvanic anodes. Galvanic anodes are generally shaped to reduced drag in the water and fitted flush to the hull to also try to minimize drag. How to Prevent Metal Corrosion

• Choose products that are made of noncorrosive metals like stainless steel and aluminum. • Maintain a dry environment using appropriate moisture barriers. • Ensure the electrical connections are clean. • On a car or truck, apply a thin coating of petroleum jelly after cleaning the terminal. • Coat metals with oil, paint, grease, or varnish because it can prevent corrosion. • Make use of cleaning agents like soaps, solvents, emulsion compounds, and chemicals to efficiently get rid of oil, grease, dirt, and other unwanted foreign deposits and follow the correct procedures in applying them. • A mixture of cola and baking soda paste will remove metal corrosion on car batteries. • To prevent soil corrosion, install correctly copper or copper alloy plumbing underground. The main causes of copper corrosion are poor drainage and moisture. A loose layer of backfill such as limestone or pea level must be put down in the trench before laying copper pipes.

Corrosion

• Galvanizing also provides metal corrosion protection. This is the process of giving a thin coating of zinc or steel material by immersing the object in a bath primarily of molten zinc. Galvanizing is an efficient way to protect steel because even if the surface is scratched, the zinc still protects the underlying layer. This process is widely used in the auto industry.

9

environment, and temperature. Accordingly adopt the right protection techniques out of the available: • Removal of oxidizing agent • Prevention or inhibition of surface reaction • Application of protective coatings – organic/ metallic/nonmetallic • Modification of the metal or the surface conditions

Summary References Corrosion is effectively controlled by cathodic protection or by appropriate selection of inhibitors, provided the chemical and electrical conditions are scientifically monitored. Understanding of the mechanism is the key in handling the condition of the metals and structures. Every scenario is site specific and needs to be addressed on its associated factors of pH of the medium,

VIII- Met J Corros Prot TM 584 Revision C. http://www.corrosioncontrol.com. Accessed Nov 1994 Corrosion Technology Laboratory – Galvanic corrosion. http://corrosion.ksc.nasa.gov/galcorr.htm. http://corro sion.ksc.nasa.gov/pitcor.htm. http://corrosion.ksc. nasa.gov/stresscor.htm. http://corrosion.ksc.nasa.gov/ crevcor.htm

I

Imprinting Masakazu Yoshikawa1 and Kalsang Tharpa2 1 Department of Biomolecular Engineering, Kyoto Institute of Technology, Kyoto, Japan 2 Department of Chemistry, University of Mysore, Mysore, India

Imprinting, which is often called “molecular imprinting,” is a facile way to introduce molecular recognition sites into polymeric membranes (materials) (Sellergren 2001; Komiyama et al. 2003; Alexander et al. 2006). In other words, the molecular memory, such as a shape of the target molecule and an alignment of the functional moieties to interact with those in target molecule, is memorized in the polymeric membranes (materials) for the recognition or separation of target molecule from others during the formation of polymeric membranes (materials). Such molecularly imprinted materials are prepared by adopting two ways; one is “molecular imprinting,” the other “alternative molecular imprinting.” The former is a pioneering method to prepare polymeric materials with molecular recognition sites from functional monomer, crosslinker, and print molecule (template) (Wulff and Sarhan 1972; Arshady and Mosbach 1981); the molecular imprinting is further divided into two methods, covalent molecular imprinting (Wulff and Sarhan 1972) and non-covalent molecular imprinting (Arshady and Mosbach # Springer-Verlag Berlin Heidelberg 2012 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_786-1

1981) as schematically shown in Fig. 1. The latter is an alternative way to obtain polymeric membranes bearing molecular recognition sites directly from candidate polymeric materials and print molecule (Yoshikawa et al. 1995, 2011). The scheme of the alternative molecular imprinting is shown in Fig. 2. In Step 1, the polymeric material, which is a candidate material to construct molecular recognition sites, is interacted with a print molecule by specific interaction before and during the formation process of molecular recognition materials so that molecular memory can be introduced into the polymeric materials. In Step 2, the print molecule is extracted from the molecularly imprinted materials. When the molecularly imprinted material thus constructed is in contact with the print molecule or print molecule analogue, the molecular recognition sites preferentially interact with them or incorporate them into the molecular recognition sites (Step 3 and Step 4). Contrary to the pioneering molecular imprinting method, molecular recognition sites are formed at the same time as the molecularly imprinted materials are prepared from polymer solution or polymer melt. In other words, any polymeric materials, such as synthetic polymers, oligopeptide derivatives, derivatives of natural polymer, and natural polymers, can be directly converted into molecular recognition material by applying the alternative molecular imprinting (Yoshikawa 2001). The similar approach was proposed by Michaels et al. in 1962 (Michaels et al. 1962).

2

Imprinting Covalent Molecular Imprinting Synthesis of Polymerizable Print Molecule

Polymerization

Removal by Chemical Cleavage

Molecular Recognition

Noncovalent Molecular Imprinting

Self-assembly

Polymerization

Removal by Solvent Extraction

Molecular Recognition

Imprinting, Fig. 1 Schemes of the covalent and non-covalent molecular imprinting (Cited from Yoshikawa et al. 2011 with permission. Copyright 2012 Elsevier Inc.)

Alternative Molecular Imprinting

Step 1 Interaction between print molecule and candidate material

Step 4 Recognition of print molecule (print molecule analogue)

Step 2 Removal of print molecule

Step 3 Formation of recognition site and permeation path

Imprinting, Fig. 2 Scheme of the alternative molecular imprinting (Cited from Yoshikawa et al. 2011 with permission. Copyright 2012 Elsevier Inc.)

This study is the first report on the alternative molecular imprinting and the first application of molecularly imprinted polymeric membranes to membrane separation. Michaels’ paper is the commemorable paper in molecular imprinting and membrane separation. In addition to this, molecularly imprinted polymeric membranes prepared by non-covalent molecular imprinting was reported in 1990 (Piletskii et al. 1990). Since

then, various molecularly imprinted membranes were studied by adopting non-covalent molecular imprinting. A wet phase inversion process was applied to an alternative molecular imprinting to prepare asymmetric membranes (Trotta et al. 2002). As described above, applying molecular imprinting, such as conventional molecular imprinting or alternative molecular imprinting,

Imprinting

molecular recognition sites are easily introduced into polymeric membranes (Ulbricht 2004). From this, it is easy to enhance permselectivity of a given membrane by applying those molecular imprinting techniques. The enhancement of flux without a reduction in permselectivity is indispensable so that molecularly imprinted membranes can be applicable to industries. Molecularly imprinted membranes with a higher surface area and a higher porosity are required to give higher flux and permselectivity. Electrospun nanofiber membranes with molecular recognition sites is a suitable or the best membrane morphology to attain high flux and high permselectivity. Possibility of the enhancement of flux without a concurrent reduction in permselectivity was proved by molecularly imprinted nanofiber membranes, which were fabricated by simultaneously applying an electrospray deposition and an alternative molecular imprinting (Yoshikawa et al. 2011).

References Allexander C, Andersson HS, Andersson LI, Ansell RJ, Kirsch N, Nicholls IA, O’Mahony J, Whitcombe MJ (2006) Molecular imprinting science and technology: a survey of the literature for the years up to and including 2003. J Mol Recognit 19:106–180

3 Arshady R, Mosbach K (1981) Synthesis of substrateselective polymers by host-guest polymerization. Makromol Chem 182:687–692 Komiyama M, Takeuchi T, Mukawa T, Asanuma H (2003) Molecular imprinting. Wiley-VCH, Weinheim Michaels AS, Baddour RF, Bixler HJ, Choo CT (1962) Conditioned polyethylene as a permselective membrane. Separation of isomeric xylenes. Ind Eng Chem Process Des Dev 1:14–25 Piletskii SA, Dubei IY, Fedoryak DM, Kukhar VP (1990) Substrate-selective polymeric membranes Selective transfer of nucleic acid components. Biopolim Kletka 6:55–58 Sellergren B (ed) (2001) Molecularly imprinted polymers man made mimics of anti bodies and their applications in analytical chemistry. Elsevier, Amsterdam Trotta F, Drioli E, Baggiani C, Lacopo D (2002) J Membr Sci 201:77–84 Ulbricht M (2004) Membrane separations using molecularly imprinted polymers. J Chromatogr B 804:113–125 Wulff G, Sarhan A (1972) The use of polymers with enzyme-analogous structures for the resolution of ¨ ber die racemates. Angew Chem Int Ed 14:341 [U Anwendung von enzymanalog gebauten Polymeren zur Racemattrennung. Angew Chem 84: 364] Yoshikawa M (2001) Molecularly imprinted polymeric membranes. Bioseparation 10:277–286 Yoshikawa M, Izumi J, Kitao T, Koya S, Sakamoto S (1995) Molecularly imprinted polymeric membranes for optical resolution. J Membr Sci 108:171–175 Yoshikawa M, Tanioka A, Matsumoto H (2011) Molecularly imprinted nanofiber membranes. Curr Opin Chem Eng 1:18–26

N

Nanotechnology Membrane Abaynesh Yihdego Gebreyohannes and Lidietta Giorno Institute on Membrane Technology, National Research Council of Italy, ITM-CNR, University of Calabria, Rende (CS), Calabria, Italy

Over the last decade, nanotechnology has rapidly evolved from an academic research to commercial reality. The concept of nanotechnology led to the development of innovative nanotechnologybased membranes that surpass the state-of-the-art performance and enable new functionality such as high permselectivity, catalytic reactivity, and fouling resistance. Nanotechnology is used to enhance performance of traditional ceramic and polymeric membrane materials through various strategies (Pendergast and Hoek 2011). Nanotechnology membranes include: • Zeolite and catalytic nanoparticle-coated ceramic membranes • Hybrid inorganic-organic nanocomposites membranes • Bio-inspired nanotechnology membranes • Bio-hybrid immobilized enzyme membranes • Bio-hybrid magnetic-responsive membranes • Aquaporin membranes # Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_787-1

• Vertically aligned nanotube membranes • Isoporous block copolymer membranes

Hybrid Inorganic-Organic Nanocomposite Membranes Mixed matrix membranes present the synergistic advantage of both the low cost and ease of fabrication of organic polymeric membranes and mechanical strength and functional properties of inorganic materials (Dong et al. 2013). The concept was first introduced in the 1990s by Zimmerman et al., as a strategy to overcome the limitations of polymeric membranes for gas separation. The most commonly exploited inorganic fillers are alumina, carbon nanotubes, ironpalladium particles, silica, and titania, which are embedded within the polymeric matrix. The particle size covers broader ranges from 0.5 to 300 nm, while the general composition ranges from 0.01 to 40 wt.% (Pendergast and Hoek 2011). Generally, inclusion of the inorganic particle imparts change in the polymeric membrane surface property (hydrophilicity/hydrophobicity, roughness, increased or decreased water permeability and selectivity depending on the type, size, and weight fraction of the inorganic particle, improved thermal and mechanical stability, as well as enhanced antifouling property).

2

In addition to microparticles, nanoparticles have been added to polymeric membranes, e.g., the inclusions to the thin films of thin-film composite (TFC) reverse osmosis membrane have been of great interest to impart the properties of the nanomaterials. Embedding smaller nanoparticles increases permeability more by increasing the characteristic pore size. One of the main reasons attributing to the enhanced permeability of the TFC membranes with embedded nanoparticles is the presence of limited crosslinking than pure polyamide TFC counterparts. Overall, the addition of nanoparticles can be tailored to particular membrane applications with the selection of nanoparticle size, shape, and type. Furthermore, inclusion of the nanoparticles such as silver in the TFC could inhibit biofouling growth due to the antimicrobial property of nanosilver (Basri et al. 2010). Nevertheless, improved permselectivity through inclusion of nanoparticle to the TFC may not be a general fact. It is theoretically predicted that the permeability of TFC, which have impermeable nanoparticles like titania, reduces, while those containing permeable nanoparticles like SOD-zeolite increases. A nanoparticle with higher water permeability relative to the polymer matrix can increase the permeability of the resulting nanocomposite membrane, while impermeable nanoparticles can only reduce the water permeability of a membrane because they reduce the area available for permeation through the polymer film. The only time impermeable particle can enhance permeability is when they cause defect in the thin-film layer which also compromises their solute rejection capacity (Pendergast and Hoek 2011). For the impermeable particles, even with the presence of loss in permeability, inclusion of the nanofillers could be desirable to induce hydrophilic or antimicrobial properties that significantly reduce membrane fouling. In addition to performance, one also has to take the cost of the nanofillers into consideration for practical application. For instance, antimicrobial and zeolite nanoparticles are expensive, yet zeolite nanofillers can bring about a considerable flux

Nanotechnology Membrane

improvement at extremely low loadings such that the cost increase may be minimal. Other interesting features one can induce by inclusion of nanofillers into polymeric membranes are stimuli-responsive properties. One of the main strategies that follows the research on stimuli-responsive organic-inorganic (O/I) hybrid membranes is the incorporation of superparamagnetic nanoparticles (NPsSP). It is an emerging field that holds numerous unexplored interesting avenues (Daraei et al. 2013; Sanchez et al. 2011). It leads to stimuli-responsive “smart” polymeric membrane with properties that can be modulated in a reversible manner. For example, recently magneticresponsive micro-mixing nanofiltration membrane is developed. The membrane was prepared by grafting magnetic-responsive nanolayers consisting of hydrophilic poly(2-hydroxyethyl methacrylate) (polyHEMA) flexible polymer chains with NPSP attached to the chain ends. The chain oscillates in an oscillating magnetic field, induced mixing at the membrane-fluid interface that maximized the disruption of concentration polarization layer (Himstedt et al. 2011). NPSP has also been used as inorganic nanofillers in a PVDF membrane to impart desired surface functionality (Huang et al. 2012). The hybrid membrane has controlled porosity, texture, and chemical composition that can provide real prospect to an efficient membrane filtration. The presence of the NPSP also induces magnetic property in the membrane that can be controlled by an external magnetic field through remote on-off switches (Daraei et al. 2013; Thevenot et al. 2013). This responsive behavior, for example, is useful to manipulate the deposition of magnetic nanoparticle on the membrane surface to prevent direct membrane-foulant interaction (Gebreyohannes et al. 2015). In addition to inducing magnetic properties, inclusion of the NPSP is reported to enhance the membranes’ hydrophilicity, mechanical strength, compaction resistance, improved permselectivity, and antifouling property (Daraei et al. 2013; Huang et al. 2012). Therefore, these membranes have

Nanotechnology Membrane

broad function ranging from environmental remediation to smart product manufacturing. Particularly, the synergistic antifouling property and improved permselectivity give the membrane a big prospect to be utilized in water purification. NPSP-coated membranes, for example, have been used for effective removal of natural organic matter (Yao et al. 2009), arsenic (Sabbatini et al. 2010), and copper (Daraei et al. 2012) during water treatment.

Bioinspired Nanotechnology Membranes Biomimetic nanostructured membranes are formed either through directly embedding biomolecule into synthetic materials or by using functional molecules to modify synthetic materials to impart specific biological property. These hybrid biomimetic membranes combine the accurate structure of a biological pore with the durability, robustness, and the possibility to control the pore size and shape of solid-state nanopore membrane (Shen et al. 2014). Biohybrid Magnetic-Responsive Membranes Superparamagnetic nanoparticles (NPsSP) are most often superparamagnetic iron oxide with zero memory of their magnetic property in the absence of an external magnetic field (Yeon et al. 2009). Polymer coatings and the introduction of various surface functional groups facilitate the anchorage of biomolecules on the surface of these particles (Miguel-Sancho et al. 2012; Brullot et al. 2012; Xiao-Ming and Wainer 1993). The resulting bionanocomposites represent an important material with a versatile application in the biotechnology, fine chemicals, drug delivery, cell transplantation, or cell immobilization. These nano-sized particles exhibit biocompatibility, high surface-to-volume ratio which makes them a good candidate for enzyme immobilization to make biohybrids. The high surfaceto-volume ratio assists with higher enzyme loading capacity and enhanced mass transfer efficiency. Hence this material with its excellent mechanical, optical, electrical, ionic, and

3

catalytic properties is a good candidate to reversibly immobilize enzyme on the surface of a polymeric membrane to mimic hybrid proteinpolymer biomimetic membranes. Alternative to direct integration of the enzyme within the polymer matrix, stimulus-responsive programmable layers on membrane can be formed through the attraction of NPSP through reversible magnetic force. These biomimetic smart layers will enable development of adaptive enzyme membrane reactors. For instance, once an enzyme is immobilized on NPSP, they can be easily dispersed in a reaction medium in the absence of an external magnetic field. However, by applying an external magnetic field, it is possible to relocate the bionanocomposites toward the surface of a membrane using an external magnetic field. The method is promising to resolve pertinent issues related to the direct integration of enzyme within the polymeric matrix. Overall, tuning reversible enzyme immobilization using NPSP helps in: 1. Easy recovery, recycling, and removal of immobilized enzyme 2. Formation of dynamic layer between membrane and feed preventing membranepollutant interaction 3. Activation of the monolayer with biocatalyst to achieve biocatalysis at the membranesolution interface 4. Easy regeneration of membrane whenever the membrane is oversaturated with substrate This approach has recently been demonstrated to form an enzyme membrane reactor (Gebreyohannes et al. 2015). The method is novel for it has used NPSP as a carrier to anchorage various enzymes as well as nanofiller to form magnetic-responsive hybrid membrane. When an external magnetic field has been applied to the system, the magnetic-responsive polymeric membrane gets magnetized. Subsequently, the NPSP with the immobilized enzyme was dispersed in the upper stream. These particles were then attracted toward the surface of the membrane. The magnetic-responsive membrane acts as magnetic field actuators that eventually help

4

with uniform dispersion of the NPSP on the surface of the membrane. Hence it has potentially mimicked the micro-nanoarchitecture of the direct integration of enzyme into the membrane. The method has shown an interesting performance when it was applied to treat a feed rich in polysaccharides. After depositing 1–2 g/m2 of pectinase activated NPSP with an average diameter of 8 nm, 40 % to 100 % reduction in the pressure required to keep flux constant at 17 L/ m2 h was observed. Moreover, thanks to the easy recycling of enzyme, both membrane and immobilized enzyme were used over several cycles, adding a considerably extended lifetime to the membrane as well as the enzyme (Gebreyohannes et al. 2015). Modulating reversible enzyme immobilization through combinatorial stimulus-responsive enzyme immobilization technique holds a great potential in bioseparations, antifouling surfaces, and creating self-cleaning membranes. Hence, the future innovations on the basis of stimulusresponsive enzyme immobilization holds a bigger prospect to “reengineer” membrane biocatalysis that will lead to the design of more complex membrane systems that are capable of mimicking the nature. Aquaporin Membranes Aquaporins are the protein channels that control water flux across biological membranes. Water movement in aquaporins is mediated by selective, rapid diffusion caused by osmotic gradients (Agre 2006; Meinild et al. 1998). According to Zhu et al. (nd), the two major factors that control water transport are: (a) Osmotic permeability: molecular movement due to concentration differences resulting in net mass transfer (b) Diffusion permeability: random movement of molecules with no net mass transfer In the first case, the water molecules transport in a single file through a narrow aquaporin channel, while in the second case, permeation occurs due to the movement of two molecules between opposite pools. Hence, the ratio of the osmotic

Nanotechnology Membrane

permeability to diffusion permeability is the number of effective steps a water molecule shall move in order to permeate through a channel. Since aquaporin channels have high water selectivity, a membrane with 75 % aquaporin coverage is predicted to have an order of magnitude higher water permeability compared to commercial seawater RO membranes (Kaufman et al. 2010). There already are interesting research activities on the inclusion of aquaporins in polymeric membranes. For instance, the use of aquaporin-Z from E. coli bacterial cells used to form a proteinpolymer membrane exhibited an order of magnitude higher water permeability as compared to a counterpure polymeric membrane. However, practical application of these proteins is highly limited by lack of large quantities of protein production. Nevertheless, with a continued research and effort, techniques to simplify the fabrication and production of mechanically sound membranes aquaporin-based biomimetic membranes are inevitable. Vertically Aligned Carbon Nanotubes Carbon nanotubes (CNTs) exhibit the fast mass transfer resemblance to aquaporin water channels in which water transport is two to five times higher than theoretically predicted by HagenPoiseuille equation (Holt et al. 2006). Molecular dynamic simulation studies attributed this outstanding flow rate to atomic smoothness and molecular ordering. In particular, water molecules permeate through the CNTs in a one-dimensional single file resulting in very small driving force use. Moreover, CNT-based membrane represents excellent mechanical properties that impart longer lifetime than conventional membrane materials. The most commonly utilized mechanism used to produce uniformly aligned nanotube arrays is chemical vapor deposition (CVD). The CNTs are adapted by fixing surface functional groups to mimic aquaporin structure or by CVD and a simultaneous solid-state reaction (Holt et al. 2006). The pore densities of the CNTs made by fixing negative surface functional groups are as high as 0.25*1012/cm2 while representing three order of magnitude higher

Nanotechnology Membrane

water flux relative to what is predicted theoretically. However, CNTs’ alignment via CVD has a number of limitations such as its costliness, sensitivity, and difficulty for large-scale application. Alternative to CVD, magnetic alignment, selfassembly, and passing single-walled CNT suspensions through polymeric filters are approaches to form CNT-aligned membranes. Although CNT mimicked biological aquaporin channels with a material producible in large amount, no large-scale CNT-aligned membranes have been produced yet. However, one can resolve these challenges similar to the challenges faced by reverse osmosis membranes 50 years ago when performance improvements lead to the practical necessity of CNT-aligned membrane application.

References Agre P (2006) The aquaporin water channels. Proc Am Thorac Soc 3:5–13 Basri H, Ismail AF, Aziz M, Nagai K, Matsuura T, Abdullah MS, Ng BC (2010) Silver-filled polyethersulfone membranes for antibacterial applications – effect of PVP and TAP addition on silver dispersion. Desalination 261:264–271 Brullot W, Reddy NK, Wouters J, Valev VK, Goderis B, Vermant J, Verbiest T (2012) Versatile ferrofluids based on polyethylene glycol coated iron oxide nanoparticles. J Magn Magn Mater 324:1919–1925 Daraei P, Madaeni SS, Ghaemi N, Salehi E, Khadivi MA, Moradian R, Astinchap B (2012) Novel polyethersulfone nanocomposite membrane prepared by PANI/Fe3O4 nanoparticles with enhanced performance for Cu(II) removal from water. J Membr Sci 415–416:250–259 Daraei P, Madaeni SS, Ghaemi N, Khadivi MA, Astinchap B, Moradian R (2013) Fouling resistant mixed matrix polyethersulfone membranes blended with magnetic nanoparticles: study of magnetic field induced casting. Sep Purif Technol 109:111–121 Dong G, Li H, Chen V (2013) Challenges and opportunities for mixed-matrix membranes for gas separation. J Mater Chem A 1:4610–4630 Gebreyohannes AY, Bilad MR, Verbiest T, Courtin CM, Dornez E, Giorno L, Curcio E, Vankelecom IFJ (2015) Nanoscale tuning of enzyme localization for enhanced reactor performance in a novel magneticresponsive biocatalytic membrane reactor. J Membr Sci 487:209–220

5 Himstedt HH, Yang Q, Dasi LP, Qian X, Wickramasinghe SR, Ulbricht M (2011) Magnetically activated micromixers for separation membranes. Langmuir 27:5574–5581 Holt JK, Park HG, Wang Y, Stadermann M, Artyukhin AB, Grigoropoulos CP, Noy A, Bakajin O (2006) Fast mass transport through sub-2-nanometer carbon nanotubes. Science 312:1034–1037 Huang Z-Q, Zheng F, Zhang Z, Xu H-T, Zhou K-M (2012) The performance of the PVDF-Fe3O4 ultrafiltration membrane and the effect of a parallel magnetic field used during the membrane formation. Desalination 292:64–72 Kaufman Y, Berman A, Freger V (2010) Supported lipid bilayer membranes for water purification by reverse osmosis. Langmuir 26:7388–7395 Meinild AK, Klaerke DA, Zeuthen T (1998) Bidirectional water fluxes and specificity for small hydrophilic molecules in aquaporins 0–5. J Biol Chem 273:32446–32451 Miguel-Sancho N, Bomati-Miguel O, Roca AG, Martinez G, Arruebo M, Santamaria J (2012) Synthesis of magnetic nanocrystals by thermal decomposition in glycol media: effect of process variables and mechanistic study. Ind Eng Chem Res 51:8348–8357 Pendergast MM, Hoek EMV (2011) A review of water treatment membrane nanotechnologies. Energy Environ Sci 4:1946–1971 Sabbatini P, Yrazu F, Rossi F, Thern G, Marajofsky A, Fidalgo de Cortalezzi MM (2010) Fabrication and characterization of iron oxide ceramic membranes for arsenic removal. Water Res 44:5702–5712 Sanchez C, Belleville P, Popall M, Nicole L (2011) Applications of advanced hybrid organic-inorganic nanomaterials: from laboratory to market. Chem Soc Rev 40:696–753 Shen Y-x, Saboe PO, Sines IT, Erbakan M, Kumar M (2014) Biomimetic membranes: a review. J Membr Sci 454:359–381 Thevenot J, Oliveira H, Sandre O, Lecommandoux S (2013) Magnetic responsive polymer composite materials. Chem Soc Rev 42:7099–7116 Xiao-Ming Z, Wainer IW (1993) On-line determination of lipase activity and enantioselectivity using an immobilized enzyme reactor coupled to a chiral stationary phase. Tetrahedron Lett 34:4731–4734 Yao P, Choo KH, Kim MH (2009) A hybridized photocatalysis-microfiltration system with iron oxidecoated membranes for the removal of natural organic matter in water treatment: effects of iron oxide layers and colloids. Water Res 43:4238–4248 Yeon KM, Lee CH, Kim J (2009) Magnetic enzyme carrier for effective biofouling control in the membrane bioreactor based on enzymatic quorum quenching. Environ Sci Technol 43:7403–7409 Zhu, F, Tajkhorshid, E, Schulten, K (nd) Theory and simulation of water permeation in aquaporin-1. Biophys J 86:50–57

B

Boric Acid Separation by Membrane Contactor A. Criscuoli Institute on Membrane Technology (ITM-CNR), Rende (CS), Italy

Boron is usually found in water as boric acid, and its concentration ranges from few ppb in river water up to around 7 ppm in seawater and can be even higher in wastewaters discharged from some industrial plants. Due to its toxic effects on both humans and plants, the World Health Organization (WHO) has fixed at 0.3 ppm the maximum allowed concentration of boron in water. Different are the techniques that can be used for controlling the boric acid content of water, like adsorption, ion-exchange resins, solvent extraction, electrodialysis, multiple stages of reverse osmosis units (working at different pH), and membrane contactors. In particular, the effectiveness of membrane contactors has been confirmed using various configurations to treat waters containing boric acid. Boric acid was selectively separated through supported liquid membranes containing inside micropores 1,3-diols (carrier) dissolved into

# Springer-Verlag Berlin Heidelberg 2012 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_791-1

o-dichlorobenzene. The membranes were hydrophobic and separated the aqueous stream containing boric acid from an alkaline aqueous phase, representing the so-called receiving phase; see Fig. 1 (Bachelier et al. 1996). The removal of boric acid was also obtained by employing hydrophilic microporous membranes, sending distilled water as extractant phase. In this case, the membrane pores were filled with the aqueous feed, and, by properly acting on the pressures of the two phases, the interface was located at the membrane surfaceextractant side, so that the boric acid removal occurred by simply diffusion from the feed to the extractant (Fig. 2). For practical implementations, an integrated membrane system where the extractant was continuously regenerated and recycled to the membrane contactor was also proposed and designed (Criscuoli et al. 2010a, b). Successful performances of hydrophilic microporous membranes using distilled water as extractant were also observed by Park and Lee (1995). Moreover, they found good efficiencies of anion exchange membrane contactors for the removal of boric acid from liquid radioactive wastes.

2 Boric Acid Separation by Membrane Contactor, Fig. 1 Transport of boric acid through a supported liquid membrane

Boric Acid Separation by Membrane Contactor

Organic-filled micropore

Boric acid

Feed water

Alkaline aqueous phase

Boric Acid Separation by Membrane Contactor, Fig. 2 Transport of boric acid in a hydrophilic membrane

Boric acid

Distilled water Feed water

References Bachelier N, Chappey C, Langevin D, Me`tayer M, Verche`re J-F (1996) Facilitated transport of boric acid through supported liquid membranes. J Membr Sci 119:285–294 Criscuoli A, Rossi E, Cofone F, Drioli E (2010a) Boron removal by membrane contactors: the water that purifies water. Clean Technol Environ Pol 12:53–61

Criscuoli A, Rossi E, Cofone F, Drioli E (2010b) Boron removal by membrane contactors: the water that purifies water. Additional information. Clean Technol Environ Pol 12:63 Park JP, Lee KJ (1995) Separation of boric acid in liquid waste with anion exchange membrane contactor. Waste Manag 15:283–291

D

Direct Fluorination of Polymer Membranes: Gas Separation Properties A. P. Kharitonov Talrose Branch of the Institute of Energy Problems of Chemical Physics of the Russian Academy of Sciences, Moscow Region, Russia

Polymeric membranes can be used for the separation of gas mixtures such as He-CH4, H2-CH4, CO2-CH4, H2-N2, etc. There is, however, a common problem in a gas separation when a polymeric membrane is used: membranes with high gas permeability often have low gas separation factor, and on the contrary, membranes with high separation factor have low permeability (Fig. 1) (Robeson 2008). The direct fluorination can be effectively used to improve gas separation properties of polymer membranes when the gas mixture consists of gases with markedly different gas kinetic diameters. In this case substantial increase of separation selectivity (up to several tens times for the case of He/CH4 mixture) is accompanied with a relatively small decrease (or no change) of permeability of a gas with smaller gas kinetic diameter (He, H2, etc.). Direct fluorination of polymers is a heterogeneous reaction of gaseous F2 mixtures with a polymer surface. This is a method of the surface modification: only upper surface layer is modified (~0.01 to several microns in thickness), but the bulk properties # Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_794-5

(e.g., tensile strength) remain unchanged. The direct fluorination proceeds spontaneously at room temperature with sufficient for industrial applications rate. Fluorination results in a substitution of H-atoms for F-atoms, saturation of double (conjugated) bonds with fluorine, and disruption of majority of C-N and C-Si bonds followed with formation of C-F bonds. The chemical composition of fluorinated layer depends on composition and pressure of fluorinating mixture and treatment duration. Treatment at mild fluorination conditions does not cause disruption of C-C bonds in the main polymer chain. Direct fluorination is a dry technology. Polymer hollow fibers, fabricated membrane modules, and composite membranes can be treated. For the case of hollow fibers and composite membranes, only the dense separation layer can be fluorinated and the porous support will remain untouched, so the tensile strength of membrane element will not be decreased. There are safe and reliable methods to neutralize (by converting into the solid phase) unused F2 and the end product HF. The direct fluorination was used to enhance gas separation properties of several polymer membranes (both homogeneous and composite) and hollow fiber modules: polyimide (PI), polyvinyltrimethylsilane (PVTMS), poly (1-trimethylsilylpropyne) (PTMSP), poly (phenylene oxide), polysulfone, poly(4-methylpentene), polycarbonatesiloxane, etc. (Langsam et al. 1988; Le Roux et al. 1994; Amirkhanov

2

Direct Fluorination of Polymer Membranes: Gas Separation Properties

Direct Fluorination of Polymer Membranes: Gas Separation Properties, Fig. 1 Separation selectivity a for the He/CH4 mixture vs. permeability of He for various polymer membranes in logarithmic scale. Filled points – literature data (Robeson 2008). Empty triangle, square, and diamond correspond to pristine polyamide Matrimid ® 5218 hollow fiber module (points correspond to different treatment conditions) and PVTMS (Kharitonov 2007, 2008) and PTMSP (Langsam et al. 1988) flat membranes. Filled symbols represent transport properties of fluorine treated membranes

et al. 1998; Kharitonov 2007, 2008). Figure 1 illustrates the influence of direct fluorination on separation selectivity for He/CH4 mixture. As it is evidenced by Fig. 1, direct fluorination results in a very remarkable increase (by a factor of several tens times or more than hundred times) of separation selectivity. The permeability of He and H2 is not practically changed after fluorination. Hence the direct fluorination of PVTMS and polyimide Matrimid 5218 provides the possibility to “overjump” the Robeson boundary (straight line in Fig. 1). The direct fluorination can substantially improve the separation selectivity of CO2/CH4, He/N2 and He/CH4 mixtures (Fig. 2).

References Amirkhanov DM, Kotenko AA, Tul’skii MN (1998) Technological characteristics of the manufacture and use of

Direct Fluorination of Polymer Membranes: Gas Separation Properties, Fig. 2 Influence of treatment conditions of PVTMS flat membrane on the separation selectivity of CO2/CH4, He/N2, and He/CH4 mixtures. Treatment condition (from left to right in each group at the plot): virgin PVTMS, treatment with 2 % F2 + 98 % He mixture, treatment with 33 % F2 + 67 %He mixture, treatment with 2 % F2 + 98 % He mixture followed by a grafting of acrylonitrile, treatment with 60 % F2 + 40 % O2 mixture fluorine-modified graviton hollow gas-separation fibres. Fibre Chem 30:318–324 Kharitonov AP (2007) Chapter 2: direct fluorination of polymers – from fundamental research to industrial applications. In: Gardiner IV (ed) Fluorine chemistry research advances. Nova Science Publishers, New York, pp 35–103 Kharitonov AP (2008) Direct fluorination of polymers. Nova Science Publishers, New York Langsam M, Anand M, Karwacki EJ (1988) Chemical surface modification of poly[1-(trimethylsilyl) propyne] for gas separation membranes. Gas Sep Purif 2:162–170 Le Roux JD, Paul DR, Kampa J, Lagow RJ (1994) Modification of asymmetric polysulfone membranes by mild surface fluorination. Part I. Transport properties. J Membr Sci 94:121–141 Robeson LM (2008) Upper bound revisited. J Membr Sci 320:390–400

P

Protein Purification by Membrane Operations Andrew L. Zydney Department of Chemical Engineering, The Pennsylvania State University, University Park, PA, USA

Synonyms Proteins separation by membrane operations Membrane systems are used extensively in the purification of high-value proteins from natural sources (e.g., milk and blood plasma) and from the biotechnology industry. This includes the use of normal flow microfiltration for sterile filtration and bioburden reduction, depth filtration for the removal of cell debris and aggregated material, tangential flow microfiltration for initial product recovery, and virus removal filtration (van Reis and Zydney 2007). Ultrafiltration is used for protein concentration and buffer exchange, both for the conditioning of the feed between other unit operations and in the final product formulation. However, actual protein purification, which typically refers to the separation of a desired protein product from other protein impurities, is accomplished using either membrane chromatography or high-performance tangential flow filtration. Membrane chromatography provides highly selective separations by exploiting differences # Springer-Verlag Berlin Heidelberg 2014 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_801-3

in protein adsorption/binding interactions, analogous to what is done in column chromatography (Ghosh 2002). The membrane adsorbers used in these systems typically have relatively large pore size with the functional ligands attached to the inner pore surface throughout the membrane. Separations can be accomplished using ion exchange (i.e., charged) ligands, affinity ligands, and hydrophobic interactions. Zeng and Ruckenstein (1999) have reviewed the different base materials and surface modification chemistries that can be used to generate chromatographic membranes. Transport through membrane adsorbers is typically dominated by the pressure-driven convective flow; diffusional limitations are much less pronounced than in conventional chromatographic beads. This can be particularly important in the purification of large biomolecules and viruses that can have difficulty accessing the internal pore space in chromatographic beads. Membrane adsorbers can be operated at much higher flow rates and with lower pressure drops than conventional chromatography columns, potentially reducing the processing time (Zhou and Tressel 2006). There has been significant interest in using membrane chromatography for bioprocessing for more than 20 years, but there have been relatively few large-scale commercial installations. However, recent improvements in membrane materials and chemistries, coupled with a greater appreciation of appropriate target applications,

2

have generated renewed interest in this technology. This is particularly true for flow-through applications in which impurities are bound to the membrane while the product flows directly through the membrane (Zhou and Tressel 2006). This includes the removal of trace DNA, viruses, and endotoxins. Flow-through applications can take full advantage of the high linear velocities possible with membranes. Single-use membrane adsorbers can be very attractive in these applications because of the small amount of impurities that need to be captured. This eliminates the need for reuse studies at small scale and for regeneration and sanitization at large scale, both of which facilitate the development of flexible manufacturing processes (van Reis and Zydney 2007). High-performance tangential flow filtration (HPTFF) is an emerging technology that uses ultrafiltration membranes for the separation of proteins without limit to their relative size (van Reis and Zydney 2007). This is in sharp contrast to conventional ultrafiltration processes that are generally thought to require a tenfold difference in size between the product and impurity for effective separation. HPTFF has been used to separate monomers from oligomers based on their difference in size, protein variants differing at only a single amino acid residue based on the difference in charge, an antigen-binding fragment from a similar size impurity, and a monoclonal antibody from host cell proteins (van Reis and Zydney 2007). The high selectivity in HPTFF processes is obtained by exploiting a number of different

Protein Purification by Membrane Operations

phenomena. As originally described, HPTFF devices are operated in the pressure-dependent regime at or below the “transition point” in a plot of filtrate flux versus transmembrane pressure. This minimizes membrane fouling and takes advantage of concentration polarization effects to increase performance (Zeman and Zydney, 1996). Further increases in selectivity are obtained by using charged ultrafiltration membranes and by adjusting the solution pH and ionic strength to exploit differences in protein charge between the product and impurity. High degrees of purification and yield are obtained using a diafiltration process in which the impurities are washed through the membrane and away from the highly retained product. It is also possible to collect the product in the permeate solution with the impurities retained by the membrane.

References Ghosh R (2002) Protein separation using membrane chromatography: opportunities and challenges. J Chromatogr A 952:13–27 van Reis R, Zydney AL (2007) Bioprocess membrane technology. J Membr Sci 297:16–50 Zeman LJ, Zydney AL (1996) Microfiltration and ultrafiltration: principles and applications. Marcel Dekker, New York Zeng X, Ruckenstein E (1999) Membrane chromatography: preparation and applications to protein separation. Biotechnol Prog 15:1003 Zhou JX, Tressel T (2006) Basic concepts in q membrane chromatography for large-scale antibody production. Biotechnol Prog 22:341

P

Protein Recovery by Membrane Operations Andrew Zydney Department of Chemical Engineering, The Pennsylvania State University, University Park, PA, USA

# Springer-Verlag Berlin Heidelberg 2012 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_802-2

Protein recovery by membrane operations is synonymous with protein purification by membrane operations and is discussed under that listing.

P

Protein Separation by Charged UF Membranes Andrew Zydney Department of Chemical Engineering, The Pennsylvania State University, University Park, PA, USA

Charged ultrafiltration membranes can be used to significantly enhance the membrane selectivity, making it possible to use these membranes for the separation of similarly sized proteins (Zydney and van Reis 2011). Charged ultrafiltration membranes can provide very high retention of likecharged proteins, enabling uncharged proteins and smaller impurities to be removed in the permeate (Zeman and Zydney 1996). Separations are typically performed at relatively low ionic strength (100 L/m2/h Salt reverse flux: 7.5. It should be noted that, besides the ability to perform multielectron redox transformations,

Multielectron Redox Catalysts

HPAs also have a very strong Bro¨nsted acidity approaching superacid region. Their acid-base and redox properties can be varied over a wide range by changing the chemical composition (Damjanovic´ et al. 2005). HPAs catalyze a wide variety of reactions in homogeneous liquid phase. Being stronger acids, HPAs will have significantly higher catalytic activity than mineral acids. HPA catalysis lacks side reactions such as sulfonation, chlorination, nitration, etc., which occur with mineral acids (Kozhevnikov 1998).

References Bard AJ, Inzelt G, Scholz F (eds) (2008) Electrochemical dictionary, 2nd edn. Springer, Berlin Damjanovic´ L, Rakic´ V, Miocˇ UB, Auroux A (2005) Influence of cations on active sites of the alkaline earth salts of 12-tungstophosphoric acid:microcalorimetric study. Thermochim Acta 434:81–87 Fendler J, Bolton J (1986) Panel discussion on sensitization and immobilization of catalysts on various supports. Homog Heterog Photocatal NATO ASI Ser 174:699–701

3 Go´mez-Romero P, Casan˜ -Pastor N (1996) Photoredox chemistry in oxide clusters. Photochromic and redox properties of polyoxometalates in connection with analog solid state colloidal systems. J Phys Chem 100:12448–12454 Holclajtner-Antunovic´ I, Bajuk-Bogdanovic´ D, Todorovic´ MR, Miocˇ UB, Zakarevska J, Uskokovic´-Markovic´ S (2008) Spectroscopic study of stability and molecular species of 12-tungstophosphoric acid in aqueous solution. Can J Chem 86:996–1004 Huynh MHV, Meyer TJ (2007) Proton-coupled electron transfer. Chem Rev 107:5004–5064 Kozhevnikov IV (1998) Catalysis by heteropoly acids and multicomponent polyoxometalates in liquid-phase reactions. Chem Rev 98:171–198 Sadakane M, Steckhan E (1998) Electrochemical properties of polyoxometalates as electrocatalysts. Chem Rev 98:219–237 Takashima T, Yamaguchi A, Hashimoto K, Nakamura R (2012) Multielectron-transfer reactions at single Cu (II) centers embedded in polyoxotungstates driven by photo-induced metal-to-metal charge transfer from anchored Ce(III) to framework W(VI). Chem Commun 48:2964–2966 Wang T, Brudvig GW, Batista VS (2010) Study of proton coupled electron transfer in a biomimetic dimanganese water oxidation catalyst with terminal water ligands. J Chem Theory Comput 6:2395–2401

O

Organic Dehydration Tadashi Uragami Organization for Research and Development of Innovative Science and Technology (ORDIST), Kansai University, Suita, Osaka, Japan

Water/organic-selective membranes are effective for the dehydration of water/organic mixtures. The dehydrated organic solvents can be useful as industrial reaction solvents, washing solvents, and analytical solvents. In Fig. 1, the permeation and separation characteristics for an aqueous dimethyl sulfoxide (DMSO) solution through a dense chitosan (Chito) membrane by TDEV (see “▶ Temperature-Difference Controlled Evapomeation”) are shown. In this figure, the feed was an aqueous solution of 50 wt.% dimethyl sulfoxide, the temperature of the feed solution was kept constant at 40 C, and the temperature of the membrane surroundings was lowered to less than the temperature of the feed solution (Uragami and Shinomiya 1992). Both the total permeation rate and the separation factor increased with dropping temperature of the membrane surroundings. This increase in the total permeation rate may be due to the increase in the solubility of the vapor in the membrane with decreasing temperature of the membrane surroundings, according to Henry’s law. The increase of the separation factor, i.e., the # Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_1036-1

improvement of the H2O/DMSO selectivity, can be explained by the illustration shown in Fig. 2. When the dimethyl sulfoxide and water molecules which had vaporized from the feed mixture come close to the membrane surroundings, the dimethyl sulfoxide vapor aggregates much easier than the water vapor (because the freezing points of dimethyl sulfoxide and water are 18.4 C and 0 C, respectively) and tends to liquefy as the temperature of the surrounding membrane becomes lower. This aggregation of the dimethyl sulfoxide molecules is responsible for the increase in the H2O/DMSO selectivity for water through the Chito membrane. The increase in the separation factor with the TDEV method, in which the temperature of the membrane surroundings is lower than the temperature of the feed solution, can be attributed to the degree of aggregation of the DMSO molecule, which is significantly governed by the temperature of the membrane surroundings. The high H2O/DMSO selectivity of Chito membrane for an aqueous solution dimethyl sulfoxide in TDEV is significantly enhanced by both the high affinity for water of the Chito membrane and the decrease in the solubility selectivity for dimethyl sulfoxide molecules into the Chito membrane, based on their aggregation on the membrane surroundings (Uragami and Shinomiya 1992; Uragami 2008). The mechanism of permeation and separation of a dimethyl sulfoxide/water mixture through a dense Chito membrane during TDEV in Fig. 2 is effective for understanding that of an ethanol/

2

Organic Dehydration

Permation rate (kg/m2hr)

0.2

500 400 300

0.1 200 100 0

10

20

30

40

Separtion factor ; aH2O/DMSO

Organic Dehydration, Fig. 1 Effects of the temperature of the membrane surroundings on the characteristics of permeation and separation for an aqueous solution of 50 wt.% dimethyl sulfoxide (DMSO) through the Chito membrane in TDEV. Feed temperature: 40 C

Temperature of membrane surroundings (˚C) To vacuum

Membrane Lower temperature

Aqueous solution CH3COOH o: H2O

Higher temperature

:CH3COOH or (CH3)2SO

Organic Dehydration, Fig. 2 Tentative separation mechanism for aqueous dimethyl sulfoxide solution through a Chito membrane in TDEV

water mixture through a dense PDMS membrane during TDEV (Uragami 2008). The characteristics of permeation and separation of acetic acid/water mixtures through 85/15 (v/v) PVA/malic acid (MA) membranes were investigated by EV (see “▶ Evapomeation”) and TDEV. The permeation rates increased, but separation factors for H2O/CH3COOH selectivity decreased with increasing permeation temperature during EV. When the temperature of feed liquid was kept constant and the temperature of the membrane surrounding was dropped, the

permeation rate and separation factor for H2O/ CH3COOH selectivity were significantly influenced by the temperature of membrane surroundings. The increase in the acetic acid concentration in the feed vapor mixture decreased the permeation rate and increased the separation factor for H2O/CH3COOH selectivity except 40 wt.% acetic acid content. The best separation factors were 800 in the EV and 860 in the TDEV for 90 wt.% acetic acid. The separation index of TDEV was higher than that of EV for an azeotropic mixture of acetic acid/water. TDEV in the separation of acetic acid/water mixtures through the PVA/MA membranes was more effective than EV (Isiklan and Sanli 2005). Graft-copolymer membranes grafted 4-vinyl pyridine, acrylonitrile, and hydroxylmethyl methacrylate onto poly(vinyl alcohol) and also were applied to the dehydration of acetic acid in EV and TDEV, and those membranes showed high dehydration performance (Asman and Sanli 2006; Al-Ghezawi and Sanli 2006). Water/alcohol-selective membranes are effective for the following scenario. For example, when an aqueous solution of dilute ethanol (about 10 wt.%) produced by the bio-fermentation is concentrated by distillation, since an aqueous solution of 96.5 wt.% ethanol is an azeotropic mixture, ethanol cannot be concentrated anymore by distillation, and consequently, ethanol is concentrated by azeotropic distillation with the addition of benzene. If membranes that can preferentially permeate only water at 3.5 wt.

Organic Dehydration

10000

8

8000 6 6000 4 4000 2

2000 0 0.0

0.5

1.0 1.5 Molar ratio (−COO−Na+/−N(CH3)3 +CI−)

Permeation rate (10−2 kg/m2 hr)

10

12000 Separation factor, aH2O/EtOH

3

0 2.0

Organic Dehydration, Fig. 3 Permeation and separation characteristics for an azeotropic mixture of ethanol/ water through q-Chito/PEO acid 4000 polyion complex/ PES composite membranes during EV as a function of the molar ratio between the carboxylate groups in PEO acid 4000 and the ammonium groups in q-Chito at 40 C

% in an azeotropic mixture of aqueous ethanol solution can be developed, significant energy savings would be achieved. The permeation and separation mechanisms in PV, EV, and TDEV through dense membranes consist of the dissolution of the permeants into the membrane, the diffusion of the permeants in the membrane, and the evaporation of the permeants from the membrane. Therefore, the separation of permeants in the membrane separation techniques depends on the differences in the solubility and diffusivity of the permeants in the feed mixture. When the structure of water/alcohol- and water/organic liquid-selective membranes is domically designed, hydrophilic materials can be recommended as membrane materials. Therefore, an increase in the solubility of water molecules into the membrane during the solution process can be expected. In order to raise the affinity of membranes for water molecules, membranes with dissociation groups introduced into their structure are used for dehydration from organic solvents. The dehydration of an ethanol/water azeotrope during EV using polyion complex crosslinked chitosan composite (q-Chito-PEO acid PIC/PES composite) membranes, constructed

from quaternized chitosan (q-Chito) and polyethylene oxydiglycilic acid (PEO acid) on a porous polyethersulfone (PES) support, was investigated (Uragami and Yamada 1999, 2003). Both the q-Chito/PES composite and the q-Chito-PEO acid polyion complex/PES composite membranes showed high H2O/EtOH selectivity for an ethanol/water azeotrope. Both the permeation rate and the H2O/EtOH selectivity were enhanced by increasing the degree of quaternization of the chitosan molecule, because the affinity of the q-Chito/PES composite membranes for water was increased by introducing a quaternized ammonium group into the chitosan molecule. Q-Chito-PEO acid PIC/PES composite membranes prepared from an equimolar ratio of carboxylate groups in the PEO acid versus quaternized ammonium groups in the q-Chito showed the best separation factor for H2O/EtOH selectivity without lowering the permeation rate, as shown in Fig. 3. With an increasing molecular weight of PEO acid, the separation factor for H2O/EtOH selectivity increased, but the permeation rate almost did not change. The separation factor for aqueous solutions of 1- and 2-propanol was also maximized at an equimolar ratio of carboxylate groups and ammonium groups and was greater than for an ethanol/water azeotrope.

References Al-Ghezawi N, Sanli O (2006) Permeation and separation characteristics of acetic acid-water mixtures by pervaporation through acrylonitrile and hydroxy ethyl methacrylate grafted poly(vinyl alcohol) membrane. Sep Sci Technol 41:2913 Asman G, Sanli O (2006) Separation characteristics of acetic acid-water mixtures using poly(vinyl alcoholg-4-vinyl pyridine) membranes by pervaporation and temperature difference evapomeation techniques. J Appl Polym Sci 100:1385, 199 Isiklan N, Sanli O (2005) Permeation and separation characteristics of acetic acid-water mixtures through poly (vinyl alcohol)/malic acid membranes by evapomeation and temperature difference controlled evapomeation. Sep Sci Technol 40:1083 Uragami T (2008) Structural design of polymer membranes for concentration of bio-ethanol. Polym J 40:485 Uragami T, Shinomiya H (1992) Concentration of aqueous dimethyl sulfoxide solutions through a chitosan

4 membrane by permeation with a temperature difference. J Membr Sci 74:183 Uragami T, Yamada H (1999) Removal of volatile organic compounds from dilute aqueous solutions by pervaporation. Netw Polym 20:203

Organic Dehydration Uragami T, Yamada H (2003) Dehydration from alcohols by polyion complex cross-linked chitosan composite membranes during evapomeation. Biomacromolecules 4:137

P

Pervaporation for Chloroform Separation Tadashi Uragami Organization for Research and Development of Innovative Science and Technology (ORDIST), Kansai University, Suita, Osaka, Japan

Chloroform/water-selective membranes are effective for the removal of chloroform in water. These membranes can be contributed to the environmental problem. Removal of chloroform in an aqueous chloroform solution through the poly (methylmethacrylate)-poly(dimethylsiloxane) (PMMA-g-PDMS), poly(ethylmethacrylate)PDMS (PEMA-g-PDMS), and poly(nbutylmethacrylate)-PDMS (PBMA-g-PDMS) graft copolymer membranes was investigated by pervaporation (PV). Figure 1 shows the results of PV for an aqueous solution of 0.02 wt.% chloroform through the graft copolymer membranes. The chloroform permselectivity of PMMA-g-PDMS and PEMAg-PDMS membranes increased dramatically at a

# Springer-Verlag Berlin Heidelberg 2014 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_1037-1

DMS content of more than about 40 and 70 mol%, respectively. It is noticeable that the benzene and chloroform permselectivity and the normalized permeation rates for the graft copolymer membranes changed remarkably at the same DMS content. These results suggest that the PMMAg-PDMS and PEMA-g-PDMS membrane structure changed at the corresponding DMS content. On the other hand, the PBMA-g-PDMS membrane showed a gradual increase in chloroform permselectivity and normalized permeation rate with increasing DMS content. From the above results, it was found that the structural change in the PBMA-g-PDMS membrane with variations in DMS content was quite different from that of the PMMA-g-PDMS and PEMA-g-PDMS membranes (Uragami et al. 2001). The permeation and separation characteristics of chloroform from water by PV through crosslinked PDMS membranes prepared from PDMSDMMA and divinyl compounds, such as EGDM, DVB, DVS, and DVF, were studied (Ohshima et al. 2005). Those membranes showed high chloroform selectivity and permeability. Both chloroform/water selectivity and permeability were affected significantly by the divinyl

2

Pervaporation for Chloroform Separation

compound. Furthermore cross-linked PDMSDMMA membranes showed the highest chloroform/water selectivity. The chloroform/ water selectivity was mainly governed by the solubility rather than the diffusion selectivity. With increasing downstream pressure, the chloroform/water selectivity of all cross-linked PDMSDMMA membranes increased, but the permeability decreased. A PDMSDMMA-DVF membrane exhibited a normalized permeation rate of 1.9 10 5 kgm (m2h) 1 It and a separation factor for chloroform/water of 4,850, yielding a separation index of 9,110. PV experiments were carried out using poly (vinylidene fluoride) (PVF2) flat-sheet membranes with different thickness for chloroform/ water mixtures (Khayet and Matsuura 2004). In Table 1 the permeation rates for chloroform and water and the separation factor, aCH3Cl/H2O, are listed. With increasing membrane thickness, the separation factor for the chloroform/water increased, and the permeation rates for water and chloroform decreased. In fact, the selectivity of the PVF2 membranes is associated not only with the surface characteristics but also with the diffusion through the membrane bulk structure.

Normalized permeation rate (10−6kg m/m2 hr)

4

3

2

1

Chloroform in permeate (wt%)

0 100

80

60

40

20

0 0

20

40

60

80

100

DMS content (mol%)

Pervaporation for Chloroform Separation, Fig. 1 Effect of the DMS content on the chloroform concentration in the permeate and normalized permeation rate for an aqueous solution of 0.02 wt.% chloroform through the PMMA-g-PDMS (○), PEMA-g-PDMS (●), and PBMA-g-PDMS (□) membranes by pervaporation

Pervaporation for Chloroform Separation, Table 1 Permeation separation characteristics for chloroform/water mixture during PV (Khayet and Matsuura 2004) Membrane thickness (mm) 38.81 49.77 59.70 83.15 110.77

Permeation rate for water (10 4kgm 2s 1) 1.69 1.28 1.17 0.87 0.66

Permeation rate for chloroform (10 5kgm 2s 1) 2.47 1.81 1.91 1.40 1.14

Initial chloroform concentration 1 kg/m3, feed temperature 25 C, reduced pressure 1,666.5 Pa

Separation factor aCH3Cl/H2O 146.02 141.69 163.15 160.47 173.32

Pervaporation for Chloroform Separation

References Khayet M, Matsuura T (2004) Pervaporation and vacuum membrane distillation processes: modeling and experiments. AIChE J 50:1697 Ohshima T, Kogami Y, Miyata T, Uragami T (2005) Pervaporation characteristics of cross-linked poly (dimethylsiloxane) membranes for removal of various

3 volatile organic compounds from water. J Membr Sci 260:156 Uragami T, Yamada H, Miyata T (2001) Removal of dilute volatile organic compounds in water through graft copolymer membranes consisting of poly (alkylmethacrylate) and poly(dimethylsiloxane) by pervaporation and their membrane morphology. J Membr Sci 187:255

R

Removal of Dissolved VOCs from Aqueous Solutions (Pervaporation Application) Tadashi Uragami Organization for Research and Development of Innovative Science and Technology (ORDIST), Kansai University, Suita, Osaka, Japan

Organic liquid/water selective membranes are effective for the removal of organics in water and recovery of organic solvents from water. These membranes can contribute to the environmental problem and effective use of organic solvents. The removal of volatile organic compounds (VOCs) such as benzene and chloroform from aqueous benzene and chloroform solutions using poly(methyl methacrylate)-poly (dimethylsiloxane) (PMMA-g-PDMS), poly (ethyl methacrylate)-PDMS (PEMA-g-PDMS), and poly(n-butyl methacrylate)-PDMS (PBMAg-PDMS) graft copolymer membranes was investigated by pervaporation (PV). When aqueous solutions of dilute VOCs were permeated through the PMMA-g-PDMS and PEMA-g-PDMS membranes, these membranes were Bz/H2O- and CHCl3/H2O selective. The permeation and separation characteristics of the PMMA-g-PDMS and PEMA-g-PDMS membranes changed drastically at a DMS content of about 40 mol% and 70 mol%, respectively, as shown in Fig. 1. # Springer-Verlag Berlin Heidelberg 2014 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_1038-3

The permeation rate and VOC/water selectivity of the PBMA-g-PDMS membranes, however, increased gradually with increasing DMS content, unlike those of PMMA-g-PDMS and PEMA-g-PDMS membranes. Furthermore, TEM observations revealed that the PMMAg-PDMS and PEMA-g-PDMS membranes had microphase-separated structures, consisting of a PDMS phase and a poly(alkyl methacrylate) phase. On the other hand, the PBMA-g-PDMS membrane was homogeneous. It was found that the permeability and selectivity of these graft copolymer membranes for treatment of aqueous VOC solutions by PV were significantly related to a PDMS continuous layer in the phaseseparated structure (Uragami et al. 1999, 2001a). Hydrophobically surface-modified membranes were prepared by adding a fluorinecontaining graft copolymer to a microphaseseparated membrane consisting of PDMS and PMMA to improve organic component selectivity. Contact angle measurements and X-ray photoelectron spectroscopy (XPS) revealed that the addition of a fluorine-containing copolymer produced a hydrophobic surface at the air side of the membrane due to surface localization of the fluorinated copolymer. It was apparent from TEM that adding a fluorine-containing copolymer of less than small amount did not affect the morphology of the microphase-separated membrane. However, adding a fluorine-containing copolymer over a certain amount resulted in a morphological change, from a continuous PDMS phase

2

Removal of Dissolved VOCs from Aqueous Solutions (Pervaporation Application) 100

Renzene in permeate (wt%)

Normalized pemeation rate (10 −6kg m/m2 h)

4

3

2

1

80

60

40

20

0 0

20

40

60

80

100

DMS content (mol%)

20

40

60

80

100

DMS content (mol%)

Removal of Dissolved VOCs from Aqueous Solutions (Pervaporation Application), Fig. 1 Effects of the DMS content on the benzene concentration in the permeate and normalized permeate for an aqueous solution of

0.05 wt.% benzene through PMMA-g-PDMS (○), PEMAg-PDMS (●), and PBMA-g-PDMS (□) membranes during PV

to a discontinuous PDMS phase. The addition of a small amount of fluorine-containing copolymer to the microphase-separated membranes enhanced both their permeability and Bz/Chx selectivity for a dilute aqueous solution of benzene during PV because of their hydrophobic surfaces and microphase-separated structures. Specifically, the microphase-separated membrane with a small amount of fluorine-containing copolymer concentrated an aqueous solution from 0.05 to 70 wt.% benzene and, therefore, removed the benzene from water very effectively (Miyata et al. 2001). PMMA-g-PDMS and PMMA-b-PDMS membranes containing tert-butylcalix[4]arene (CA) (CA/PMMA-g-PDMS and CA/PMMAb-PDMS) were applied to the removal of benzene from a dilute aqueous solution of benzene by PV (Uragami et al. 2006). When an aqueous solution of 0.05 wt.% benzene was permeated through CA/PMMA-g-PDMS and CA/PMMA-b-PDMS membranes, these membranes showed high Bz/Chx selectivity. Both the permeability and Bz/H2O selectivity of the CA/PMMA-g-PDMS and CA/PMMA-b-PDMS membranes were enhanced by increasing the CA content, due to the affinity of CA for benzene. The permeability

and Bz/H2O selectivity of CA/PMMA-b-PDMS membranes were much greater than those of CA/PMMA-g-PDMS membranes. TEM observations revealed that both the CA/PMMA-g-PDMS and CA/PMMA-b-PDMS membranes had microphase-separated structures consisting of a PMMA phase and a PDMS phase-containing CA. The microphase-separated structure of the latter membranes was much clearer than that of the former and was lamellar. The distribution of CA in the microphase-separated structure of the CA/PMMA-g-PDMS and CA/PMMA-b-PDMS membranes was analyzed by differential scanning calorimetry (DSC), and CA was distributed in a PDMS continuous layer in microphaseseparated structure (Uragami et al. 2001b, 2006). It was found that a continuous PDMS layer in PMMA-g-PDMS and PMMA-b-PDMS membranes plays an important role for the removal of VOCs from water. For the purpose of constructing the membrane matrix from PDMS component mainly, polydimethylsiloxane dimethylmethacrylate macromonomer (PDMSDMMA) was selected as a membrane material. The effects of cross-linkers of the cross-linked PDMS membranes derived from PDMSDMMA and divinyl compounds were

Removal of Dissolved VOCs from Aqueous Solutions (Pervaporation Application)

3

Removal of Dissolved VOCs from Aqueous Solutions (Pervaporation Application), Table 1 Performance for Bz/H2O of various membranes containing PDMS component asep. Bz/ H2O 53 620 1,772 4,492

asorp. Bz/ H2O 422 739 1,267

adiff. Bz/ H2O 0.13 0.86 1.40

3,171

1,436

2.21

PDMSDMMA-DVSh

2,886

1,270

PDMSDMMA-DVFi CA/PDMSDMMA-DVBj CA/PDMSDMMA-DVSk CA/PDMSDMMA-DVFl

4,316 4,021 3,866 5,027

1,804 1,689 1,620 1,998

Various PDMS membranesa PMMA PMMA-g-PDMSd CA/PMMA-g-PDMSe PFA-g-PDMS/PMMAg-PDMSf PDMSDMMA-DVBg

NPRb 0.29 0.13 0.71 0.64

PSIc 16 226 1,240 2,879

Refs Uragami et al. (2001a) Uragami et al. (2006) Uragami et al. (2006) Miyata et al. (2001)

1.46

4,629

2.46

1.96

5,656

2.49 2.18 2.39 2.52

1.72 1.75 1.97 1.86

7,423 7,037 7,616 9,350

Uragami and Ohshima (2003) Uragami and Ohshima (2003) Ohshima et al. (2005) Ohshima et al. (2005) Ohshima et al. (2005) Ohshima et al. (2005)

PV experimental conditions: feed solution, an aqueous solution of 0.05 wt.% benzene; permeation temperature, 40 C; pressure of permeation side, 1.33 Pa b Normalized permeation rate [10 5 mkg(m2h) 1] c PV separation index (NPR asep. Bz/H2O) d PDMS content 74 mol% e PDMS content 74 mol%; CA content 40 mol% f PDMS content 74 mol%; PFA-g-PDMS content 1.2 wt.% g DVB content 80 mol% h DVS content 90 mol% i DVF content 90 mol% j DVB content 80 mol%; CA content 0.5 wt.% k DVS content 90 mol%; CA content 0.5 wt.% l DVF content; CA content a

investigated, on the PV characteristics of the removal of benzene from an aqueous solution of dilute benzene. When an aqueous solution of 0.05 wt.% benzene was permeated through the cross-linked PDMSDMMA membranes, they showed high Bz/H2O selectivity. Both the permeability and Bz/H2O selectivity of the membranes were enhanced with increasing divinyl compound content as the cross-linker and were significantly influenced by the kind of divinyl compound. PDMSDMMA membranes crosslinked with divinyl siloxane (DVS) showed very high membrane performance during PV. The best normalized permeation rate, separation factor for Bz/H2O selectivity, and PV separation index (PSI) (Huang and Rhim 1991) which is the product of the permeation rate and the separation factor and can be used as a measure of the

membrane performance during PV, of a PDMSDMMA-DVS membrane, were 1.96 10 5 mkg (m2h)-1, 98, and 192, respectively (Uragami et al. 2003). When divinyl perfluoron-hexane (DVF), which is much more hydrophobic, was employed as a cross-linker of PDMSDMMA, the best normalized permeation rate, separation factor for Bz/H2O selectivity, and PSI of a PDMSDMMA-DVF membrane were 1.72 10 5 mkg(m2h) 1, 4,316, and 7,423, respectively (Ohshima et al. 2005a). In Table 1, the permeation and separation characteristics of various polymer membranes consisting of the PDMS components are compared under the same PV condition: feed solution, an aqueous solution of 0.05 wt.% benzene; permeation temperature, 40 C; and pressure of permeation side, 1.33 Pa. As can be seen in

4

Removal of Dissolved VOCs from Aqueous Solutions (Pervaporation Application)

Table 1, both the normalized permeation rate and the Bz/H2O selectivity of each of the CA/ PDMSDMMA-DVB, CA/PDMSDMMA-DVS, and CA/PDMSDMMA-DVF membranes were improved as compared to each of the PDMSDMMA-DVB, PDMSDMMA-DVS, and PDMSDMMA-DVF membranes. Although the separation factors of the CA/PDMSDMMADVB and CA/PDMSDMMA-DVS membranes were lower than that of the PFA-g-PDMS/ PMMA-g-PDMS membranes, the PSI of the former membranes was much greater than that of the latter one. In Table 1, it was found that the addition of CA into the cross-linked PDMSDMMA membranes cross-linked with a suitable crosslinker is significantly effective to obtain higher permeation and separation characteristics. A CA/ PDMSDMMA-DVF membrane with DVF of 90 mol% and CA of 0.4 wt.% showed the best membrane performance, i.e., the normalized permeation rate, separation factor for Bz/H2O selectivity, and PSI were 1.86 10 5mkg(m2h) 1, 5,027, and 9,350, respectively. The air stripping removal of VOCs such as toluene and phenol from water by microporous polypropylene (PP) hollow fibers was studied. The VOC stream passed through the lumen side of the module, while air (stripping gas) flowed across the shell side. Experiments were performed at different liquid flow rates (8–16 cm3min 1), gas flow rates (60–180 cm3min 1), feed VOC concentrations (100–1,000 ppm), and temperatures (24–35 C). The removal was more effective when feed VOC level and liquid or gas flow rate increased. The applicability of a mass transfer model that considers diffusion in the liquid layer, membrane, and gas layer under steady state was checked. Unlike phenol with a very small dimensionless Henry’s law constant (equilibrium gas concentration divided by liquid concentration) and a relatively low amount of sorption on PP fibers, the measured overall mass transfer coefficients for toluene reasonably agreed with those predicted from the model. The large deviation observed for phenol indicated unsteady state nature, likely due to its small concentration

difference between air and liquid phase/fiber matrix (Juang et al. 2005).

References Huang RYM, Rhim JW (1991) Separation characteristics of pervaporation membrane separation processes. In: Huang RYM (ed) Pervaporation membrane separation processes, Membrane science technology series 1. Elsevier, Amsterdam, pp 111–180, Chapter 3 Juang RS, Lin SH, Yang MC (2005) Mass transfer analysis on air stripping of VOCs from water in microporous hollow fibers. J Memb Sci 255:79 Miyata T, Yamada H, Uragami T (2001) Surface modification of microphase-separated membranes by fluorine-containing polymer additive and removal of dilute benzene in water through these membranes. Macromolecules 34:8026 Ohshima T, Miyata T, Uragami T (2005a) Cross-linked smart polydimethylsiloxane membranes for removal of volatile organic compounds from water. J Mol Struct 739:47 Ohshima T, Miyata T, Uragami T (2005b) Selective removal of dilute benzene from water by various cross-linked poly(dimethylsiloxane) membranes containing tertbutylcalix [4] arene. Macromol Chem Phys 206:2521 Uragami T, Ohshima T (2003) Removal of benzene from an aqueous solution of dilute benzene by various crosslinked poly(dimethylsiloxane) membranes during pervaporation. Macromolecules 36:9430 Uragami T, Yamada H, Miyata T (1999) Removal of volatile organic compounds from dilute aqueous solutions by pervaporation. Trans Mater Res Soc Jpn 24:165 Uragami T, Yamada H, Miyata T (2001a) Removal of dilute volatile organic compounds (VOCs) in water through graft copolymer membranes consisting of poly(alkylmethacrylate) and poly(dimethylsiloxane) by pervaporation and their membrane morphology. J Memb Sci 187:255 Uragami T, Meotoiwa T, Miyata T (2001b) Effects of the addition of calixarene to microphase-separated membranes for the removal of volatile organic compounds from dilute aqueous solutions. Macromolecules 34:6806 Uragami T, Meotoiwa T, Miyata T (2003) Effects of morphology of multicomponent polymer membranes containing calixarene on permselective removal of benzene from a dilute aqueous solution of benzene. Macromolecules 36:2041 Uragami T, Yamada H, Miyata T (2006) Effects of fluorine-containing graft- and block-copolymer additives on removal characteristics of dilute benzene in water by microphase-separated membranes modified with these additives. Macromolecules 39:1890

D

Dissolved Oxygen (DO) in Water by Membrane Operations, Removal of V. V. Volkov A.V.Topchiev Institute of Petrochemical Synthesis Russian Academy of Sciences (TIPS RAS), Moscow, Russian Federation

Dissolved oxygen (DO) is one of the principal components which should be removed for the production of ultrapure water. Even though concentration of DO in water is very low (approximately 8 ppm under ambient conditions), DO leads to oxidation of dissolved components and materials, for example, boiler piping in (nuclear) power plants. In such systems, the desired DO concentration in the boiler feed water is 7–10 ppb. Therefore, DO removal from water presents a challenging problem for various industries such as production of semiconductors, power plants, pharmaceuticals, and biotechnology. In the production of semiconductors, standard DO level in ultrapure water is the most stringent (below 1 ppb). Removal of DO from water by membrane technology can be accomplished by using gas-liquid membrane contactors and membrane reactors. Gas-liquid membrane contactors generally incorporate a porous hollow fiber membrane for direct contacting of two immiscible phases, namely, a gas and a liquid, for the purpose of absorption or stripping without dispersion of # Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_1039-4

one phase into another. The DO removal from water on a membrane contactor requires generation of a driving force for oxygen transmembrane transfer, and this process demands reduction in oxygen concentration in a gas phase. This goal can be accomplished in one of three ways: (1) by applying vacuum to the water, (2) by supplying an appropriate sweep gas to the gas side, and (3) by a combination of the first two where a small flow of sweep gas is used to improve separation driving force but the gas side is still maintained under vacuum (Sengupta et al. 1998). In industrial (commercial) LiquiCel ® gas-liquid membrane contactors, polypropylene porous hollow fiber membranes are used. These membrane contactors offer a serious alternative to the traditional two-stage vacuum degassing towers in the high-purity water treatment systems in the semiconductor industry. DO removal from water can be performed on a membrane-UV reactor because DO can react with hydrogen to yield water under the action of UV irradiation with a wavelength of 185 nm. This reaction requires dissolution of hydrogen in water, and this process is provided by gas-liquid membrane contactors based on hollow fiber membranes which are permeable for hydrogen and UV irradiation (Li and Tan 2001). Most efficient existing methods for deep removal of DO from water are based on the principle of oxygen reduction by hydrogen on a palladium catalyst which yields water. The existing catalytic processes for water

2

Dissolved Oxygen (DO) in Water by Membrane Operations, Removal of

deoxygenation via hydrogenation reaction involve two stages: (1) absorption of hydrogen in water and (2) passage through a fixed-bed catalytic reactor. In contrast to existing processes, the DO removal on a catalytic membrane reactor can be accomplished in one stage. There are two configurations of catalytic membrane reactors. The first design of the membrane reactor is based on a polypropylene porous hollow fiber membrane module packed with a palladium catalyst, namely, spherical beds of palladiumdeposited granules of anion-exchange resin in the void space of the shell side (Li et al. 1995). An alternative design of a catalytic membrane reactor provides water deoxygenation by catalytic membranes, namely, by palladium-loaded hollow fibers (Volkov et al. 2009). Water and hydrogen are supplied into a shell side of the reactor and into a fiber lumen, respectively. Hence, DO removal is accomplished by chemical reaction between dissolved oxygen and dissolved hydrogen in the presence of the palladium-based

catalyst. Using a catalytic membrane reactor, concentration of DO in water of ten parts per billion (ppb) and lower is feasible.

References Li K, Tan X (2001) Development of membrane-UV reactor for dissolved oxygen removal from water. Chem Eng Sci 56(17):5073–5083 Li K, Chua I, Ng WJ, Teo WK (1995) Removal of dissolved oxygen in ultrapure water production using a membrane reactor. Chem Eng Sci 50(22):3547–3556 Sengupta A, Peterson PA, Miller BD, Schneider J, Fulk CW (1998) Large-scale application of membrane contactors for gas transfer from or to ultrapure water. Sep Purif Technol 14:189–200 Volkov VV, Petrova IV, Lebedeva VI, Plyasova LM, Rudina NA, van Erkel J, van der Vaart R, Tereshchenko GF (2009) Catalytic nanoclusters of palladium on the surface of polypropylene hollow fiber membranes: removal of dissolved oxygen from water. In: Starov VM (ed) Nanoscience: colloidal and interfacial aspects. Taylor & Francis Group, New York, pp 1173–1188

H

H2 Permeation Through Pd-Based Membranes V. V. Volkov A.V.Topchiev Institute of Petrochemical Synthesis Russian Academy of Sciences (TIPS RAS), Moscow, Russian Federation

Palladium is known to be one of the most effective metals for hydrogen adsorption, dissociation, and recombination even though the permeability of this metal is by an order of magnitude lower than that of refractory metals such as tantalum, vanadium, and niobium. In 1863, Sainte-Claire Deville and Troost discovered that hydrogen is able to permeate through palladium; later, in 1866, Graham found that palladium can absorb huge amounts of hydrogen (several times with respect to its initial volume) at room temperature. The palladium-based membrane technology for hydrogen production has come into life in the late 1950s. In the 1960s, pioneering studies by Gryaznov led to the development of a breakthrough membrane reactor when palladiumbased membranes were used for hydrogen supply or withdrawal from the reaction zone, thus providing improved yield and high catalyst effectiveness (Pagliery and Way 2002). Hydrogen permeation through solid palladium proceeds via the solution–diffusion mechanism, and this mechanism involves the following steps: # Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_1040-2

(a) diffusion of molecular hydrogen to the surface of the palladium-based membrane, (b) a reversible dissociative adsorption on the palladium surface, (c) dissolution of atomic hydrogen in the bulk metal, (d) diffusion of atomic hydrogen through the bulk metal, (e) association of hydrogen atom on the palladium surface, (f) desorption of molecular hydrogen from the surface, and (g) diffusion of molecular hydrogen away from the surface (Yun and Oyama 2011). On the palladium surfaces, the dissociative adsorption of hydrogen molecules proceeds with a low or zero activation energy barrier. This is the first step in bulk absorption toward the formation of metal hydrides. Generally, hydrogen permeation is described by the following equation: J ¼ Pðpn h pn l Þ=L; where J is the hydrogen flux, P is the permeability coefficient, L is the thickness of the palladium layer, ph and pl are the partial pressures of hydrogen on the high-pressure (feed) side and the low-pressure (permeate) side, respectively, and n is the pressure exponent. This exponent generally ranges from 0.5 to 1 depending on the ratecontrolling step [steps (a–g)]. According to Sievert’s law, when the rate-controlling step is the bulk diffusion through the palladium layer [step (c)], the n value is equal to 0.5 because the diffusion rate is proportional to the concentration of hydrogen atoms on the opposite sides of the

2

metal surface and this hydrogen concentration is proportional to the square root of the hydrogen pressure. When mass transport to or from the surface (a, g) or dissociative adsorption (b) or associative desorption (e) becomes ratecontrolling steps, the expected value of n is 1 because these processes depend linearly on the concentration of molecular hydrogen. In the case of the Pd-based composite membranes with thick (>5 mm) Pd layers, the n values are higher than 0.5, and this fact can be explained by defects or pinholes through which substantial amounts of hydrogen can permeate. This process proceeds via the Knudsen or Poiseuille flow mechanisms, and the corresponding exponents are higher than 0.5. In the case of binary palladium alloys, hydrogen permeation is generally proportional to the average bond distance of the alloys. Permeability is the product of diffusivity and solubility. In the

H2 Permeation Through Pd-Based Membranes

case of Pd–Ag alloys, hydrogen solubility increases with increasing Ag content up to 20–40 wt.%; hydrogen diffusivity decreases with increasing Ag content. Concomitant changes in solubility and diffusivity lead to higher hydrogen permeability (1.7 times higher) as compared with that of pure Pd at 23 wt.% of Ag and at 623K (Ma et al. 2003).

References Ma YH, Mardilovich IP, Engwal EE (2003) Thin composite palladium and palladium/alloy membranes for hydrogen separation. Ann N Y Acad Sci 984:346–360 Paglieri SN, Way JD (2002) Innovations in palladium membrane research. Sep Purif Rev 31:1–169 Yun S, Oyama ST (2011) Correlations in palladium membranes for hydrogen separation: a review. J Membr Sci 375:28–45

H

Hydrogen Removal by Membranes V. V. Volkov A.V.Topchiev Institute of Petrochemical SynthesisRussian Academy of Sciences (TIPS RAS), Moscow, Russian Federation

Pure hydrogen is known to be a valuable industrial material, and its annual consumption comes to billions of cubic meters. Over 90 % of hydrogen is generated from fossil fuel sources (mainly, steam reforming of natural gas) and the other 10 % is produced by water electrolysis. Hydrogen is widely used in diverse large-scale processes in metallurgical, chemical, petrochemical, pharmaceutical, and textile industries for the production of a wide range of products – from semiconductors and steel alloys to vitamins and raw chemical materials such as ammonia, methanol, and hydrogen peroxide. Recently, hydrogen has attracted a keen attention as an alternative energy carrier for the solution of environmental problems (so-called concept of “hydrogen economy”). In contrast to fossil fuels, hydrogen combustion does not generate carbon dioxide but only water vapors. Large-scale hydrogen production requires significant capital investments for separation and purification processes; thus, the cost of hydrogen markedly increases. Conventional technologies for hydrogen separation include solvent absorption, pressure swing adsorption (PSA), fractional/cryogenic distillation, and # Springer-Verlag Berlin Heidelberg 2012 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_1041-1

membrane separation. Membrane separation seems to be the most promising method because of its low energy consumption, continuous operation, low investment costs, and easy operation (Ockwig and Nenoff 2007). Among all basic large-scale applications of polymer membranes, hydrogen recovery is the most important. In the 1970s, commercial success of hydrogen-selective hollow-fiber membrane systems for the in-process recycling of hydrogen from ammonia purge gases has triggered large-scale application of the membrane gas separation. This technology has been extended to other situations for recovery of hydrogen from gas mixtures (H2/CO or H2/CH4 ratio adjustment for syngas production) and has been successfully competing with cryogenic distillation and PSA processes. In the petrochemical industry, hydrogen recovery from refinery streams is an emerging field for membrane separation; it is a key approach to meet the increased demand of hydrogen (for hydrotreating, hydrocracking, or hydrodesulfurization processes) owing to new environmental regulations (Bernardo et al. 2009). The main liability of hydrogen-selective membranes is that recompression of the permeated hydrogen is usually required. Therefore, an alternative method involves contaminate preferential permeation, and this method is treated as a new approach for hydrogen purification, primarily by using the carbon-based membranes. For example, hydrogen production by natural gas

2

steam reforming yields a gaseous mixture containing hydrogen and also carbon dioxide and carbon monoxide. Solubility-controlled membranes (e.g., carbon-based membranes) are preferentially permeable for big-sized gas molecules (e.g., CO2) in relation to hydrogen. This peculiarity offers evident economic advantages that maintaining/collecting hydrogen in the retentate at nearly bulk feed pressure mitigates the demands for costly hydrogen recompression steps, even though a CO2 compression step is required (Ockwig and Nenoff 2007). Hydrogen-separating membranes made of palladium alloys have been developed over the past 50 years into a technology that in some instances is used in practice. In an early work in the United States and in the former Soviet Union, relatively thick-walled tubes were used. This design has been advanced for ultra-purification of hydrogen and for its application in the semiconductor manufacturing processes and in the hydrogen generators for remote or small-scale usage. Current research topics in this area are concerned with membrane reactors for hydrogen gas supply for fuel cells or for the chemical process industry.

Hydrogen Removal by Membranes

Palladium-alloy diffusers present a key component for the recovery of hydrogen radioisotopes which are used and produced by the nuclear fission in the fusion reactors. Hydrogen recovery from waste gases or purge streams (e.g., hydrotreater off-gas) presents a potentially large application of the palladium-based membrane technology. Coal gasification or natural gas reforming coupled to a palladium membrane reactor can offer an alternative huge source of hydrogen. Production of pure hydrogen for its use in fuel cells also seems to be another important mission of the palladium-based membrane reactor (Paglieri and Way 2002).

References Bernardo P, Drioli E, Golemme G (2009) Membrane gas separation: a review/state of the art. Ind Eng Chem Res 48:4638–4663 Ockwig NW, Nenoff TM (2007) Membranes for hydrogen separation. Chem Rev 107:4078–4110 Paglieri SN, Way JD (2002) Innovations in palladium membrane research. Sep Purif Rev 31:1–169

H

Hydrogen Selective Membranes V. V. Volkov A.V. Topchiev Institute of Petrochemical Synthesis Russian Academy of Sciences (TIPS RAS), Moscow, Russian Federation

As compared with other known fuels, hydrogen is abundantly available in the universe and possesses the highest energy content per weight unit. Moreover, in contrast to fossil fuels, the use of hydrogen as an energy source yields water as the only by-product. Hence, in recent years, the demand for hydrogen energy and production has been growing. Membrane separation process offers an attractive alternative to mature technologies such as pressure swing adsorption (PSA) and cryogenic distillation. Hydrogen selective membranes are designed such that hydrogen concentration increases in the permeate. Based on the materials used, hydrogen selective membranes can be classified into four types: (i) polymer (organic), (ii) metallic, (iii) carbon, and (iv) ceramic. Metallic, carbon, and ceramic membranes are referred to as inorganic membranes. Depending on the type of the raw material, inorganic membranes can be classified into two groups: metal membranes and ceramic membranes. In addition, they could be divided into porous (microporous) and nonporous (dense) membranes (Adhikari and Fernando 2006). Microporous inorganic membranes include # Springer-Verlag Berlin Heidelberg 2012 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_1042-2

carbon and ceramic (amorphous or crystalline) membranes, and their separation characteristics are governed by the molecular sieving transport mechanism. Usually, dense inorganic membranes are based on a metal or on a polycrystalline ceramic. In these dense membranes, the fundamental operating mechanism involves the conduction of free electrons and the presence of specific catalytic surfaces in order to dissociate H2 on the raw feed stream side and reassociate both protons and electrons on the product side. In such systems, hydrogen selectivity is typically very high because their dense structure prevents the transport of larger atoms and molecules (e.g., CO, CO2, O2, N2, etc.). This high selectivity provides the production of high-purity hydrogen (Ockwig and Nenoff 2007). Historically, hydrogen separation is accomplished on the Pd-based membranes because they catalyze surface dissociation/reassociation processes and they are highly permeable to hydrogen. Palladium and Pd-alloy membranes can produce hydrogen gas on the ppb impurity levels. Since the late 1950s, small-scale Pd-based membrane modules have produced high-purity hydrogen at remote sites and for industrial, laboratory, or military purposes. There are various types of metallic membrane materials for hydrogen separation: (i) pure metals (Pd, V, Ta, Nb, and Ti); (ii) binary alloys of Pd such as Pd-Cu, Pd-Ag, Pd-Y, and also Pd-based alloys with Ni, Au, Ce, and Fe; (iii) complex alloys (Pd alloyed with three to five other metals); (iv) amorphous

2

alloys (typically Group IV and Group V metals); and (v) coated metals such as Pd on Ta, V, etc. (Adhikari and Fernando 2006). Dense ceramic membranes (e.g., perovskite, bismuth oxide, and solid electrolyte) have been developed and commercialized. Dense inorganic membranes are generally designed as a thin film on a porous inorganic support. To overcome the relatively low permeance and high cost of dense metallic membranes, researchers are exploring the utility of highpermeance, less costly, and less selective microporous inorganic (silica or carbon or zeolite) membranes. The structure of these membranes presents a network of interconnected micropores with a diameter of ~0.5 nm. In general, microporous silicas show the highest hydrogen selectivities, and they exhibit the best H2/N2 selectivity which exceeds 10,000 for the membranes prepared by chemical vapor deposition (Ritter and Ebner 2007). Since the 1970s, gas separation polymer membranes have been used industrially for hydrogen separation from gaseous mixtures in the United States (Ockwig and Nenoff 2007) and in the former Soviet Union (Yampolskii and Volkov 1991). Membranes made of glassy polymers (i.e., polymers with glass transition temperatures above the operating temperature) are used for

Hydrogen Selective Membranes

removing hydrogen from gas mixtures. Examples of membrane materials for commercial polymer membranes include polyimide and polysulfone (asymmetric hollow fiber membranes) or polyvinyl trimethylsilane and cellulose acetate (asymmetric flat membranes). Such polymer membranes operate according to the solutiondiffusion mechanism. In general, hydrogen selectivity is low, moderate, and very high for polymeric, microporous inorganic, and metallic membranes, respectively. Polymeric membranes primarily operate at temperatures below 373 K, whereas carbon, silica, and dense ceramic membranes can function at higher temperatures (773–1,173 K). Metallic membranes can be used at ~573–873 K.

References Adhikari S, Fernando S (2006) Hydrogen membrane separation techniques. Ind Eng Chem Res 45:875–888 Ockwig NW, Nenoff TM (2007) Membranes for hydrogen separation. Chem Rev 107:4078–4110 Ritter JA, Ebner AD (2007) State-of-the-art adsorption and membrane separation processes for hydrogen production in the chemical and petrochemical industries. Sep Sci Technol 42:1123–1193 Yampolskii YP, Volkov VV (1991) Studies in gas permeability and membrane gas separation in the Soviet Union. J Membr Sci 64:191–228

H

Hydrogenation Contactors and Reactors V. V Volkov A.V.Topchiev Institute of Petrochemical Synthesis Russian Academy of Sciences (TIPS RAS), Moscow, Russian Federation

Among all known catalysts for hydrogenation, metals of platinum family (Pt, Pd, Ni) are known to be most efficient and advantageous. Palladium catalysts for selective hydrogenation hold the unique position. Palladium and its alloys show an excellent potential to dissolve huge amounts of hydrogen, and as membrane materials, dense palladium and palladium alloy membranes are permeable only for hydrogen. Pioneering studies by Gryaznov and his coworkers have encouraged the use of palladium and palladium alloy membranes in catalytic membrane reactors, including fundamental studies on selective hydrogenation of various unsaturated hydrocarbons, for example, acetylene, 1,3-pentadiene, and cyclopentadiene. The general concept can be formulated as follows: hydrogen is supplied to one side of the Pd-based catalytic membrane, and then it diffuses selectively through the membrane and approaches the other side of the membrane in its highly reactive form. Fundamental concepts on membrane catalysis and catalytic membrane reactors have been advanced in the 1960s, and the phenomenon of # Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_1043-1

conjugating of chemical reactions with the evolution (dehydrogenation reaction) and consumption (hydrogenation reaction) of hydrogen on palladium membranes has been discovered in 1964 (Gryaznov et al. 2003). This concept is illustrated in Fig. 1. Early studies in this direction have been focused on the use of relatively thick nonporous Pd or Pd-alloy membranes. Their thickness can be reduced by preparation of composite membranes on porous or nonporous supports. A critically new step providing reduced consumption of expensive noble metals is concerned with the development of catalytic membranes based on the metal–polymer composites. In this case, nonporous polymer matrix, for example, silicon rubber, is loaded with palladium nanoparticles (Gryaznov et al. 2003). Pd-based catalytic nanoparticles can be immobilized on the surface of porous inorganic or polymeric membranes, thus providing a new type of catalytic membrane reactor, called catalytic contactor. According to the proposed classification (Miachon and Dalmon 2004), depending on the mode of supply of reagents, two types of catalytic contactors exist: an interfacial contactor and a flow-through contactor. In this case, a key point is concerned with localization of reactants in the very volume where the catalyst is deposited. The typical example of hydrogenation on an interfacial catalytic contactor is related to removal of dissolved oxygen (DO) from water

2

Hydrogenation Contactors and Reactors

ΔH < 0

Substrate Liquid

Hydrogenated products

Palladium alloy

‹H›

q

~ 50 – 100 μm

R (opt.)

Gas H2 or R-H2

ΔH > 0

Hydrogenation Contactors and Reactors, Fig. 1 Principle of a dense and self-supporting palladium alloy membrane (a thin-walled tube or a foil) for liquid phase hydrogenation. Hydrogen or hydrogen-containing source R–H2 is supplied to the gas side; a to-behydrogenated substrate is supplied to the liquid side. Dehydrogenation of R–H2 occurs on the gas-side surface

and this process is endothermic; both reactions are accompanied by hydrogen transfer. Due to excellent thermal conductivity of the membrane, the heat released by hydrogenation can be utilized to facilitate endothermic dehydrogenation (autothermic operation) (Dittmeyer et al. 2004)

(Volkov et al. 2011). Hydrogen and DO-containing water are supplied to the opposite sides of a hydrophobic porous catalytic membrane. In this case, membrane material should be non-wettable for water (hydrophobic) in order to provide liquid-free pores and high-rate hydrogen transmembrane mass transfer. The Pd catalyst is placed onto the water-contacting membrane surface. Water deoxygenation by catalytic membranes, including palladium-loaded hollow fibers, is accomplished by the chemical reaction between dissolved oxygen and hydrogen in the presence of the palladium catalyst. In a flow-through contactor, reactants are forced to flow through a porous catalytic membrane, i.e., through the Pd-loaded pores. This enables to control and rule residence time of reacting species in the active zone of the catalytic membrane.

References Dittmeyer R, Svajda K, Reif M (2004) A review of catalytic membrane layers for gas/liquid reactions. Top Catal 29(1–2):3–27 Gryaznov VM, Ermilova MM, Orekhova NV, Tereschenko GF (2003) Reactors with metal and metal-containing membranes. In: Cybulski A, Moulijn JA (eds) Structured catalysts and reactors, 2nd edn. Taylor & Francis, New York, pp 579–614 Miachon S, Dalmon J-A (2004) Catalysis in membrane reactors: what about the catalyst. Top Catal 29:59–65 Volkov VV, Petrova IV, Lebedeva VI, Roldughin VI, Tereshchenko GF (2011) Palladium-loaded polymeric membranes for hydrogenation in catalytic membrane reactors. In: Basile A, Gallucci F (eds) Membranes for membrane reactors: preparation, optimization and selection. Wiley, New York, pp 531–548

P

Palladium-Loaded Hollow Fibers: rate, and it provides a nearly dense palladium Application in Water Deoxygenation layer with thickness of 2–3 mm. V. V. Volkov A.V.Topchiev Institute of Petrochemical Synthesis, Russian Academy of Sciences (TIPS RAS), Moscow, Russian Federation

Palladium nanoparticles can be deposited on a hydrophobic porous polypropylene hollow fiber membrane, which preserves its hydrophobic nature (Lebedeva et al. 2006). Prior to the deposition of palladium onto the membrane surface, hollow fibers should be cleaned with surfactants and organic solvents. Then, their outer surface is etched either by strong inorganic acids or by strong inorganic bases. The next step includes electroless deposition of palladium by one of the two following methods: (1) reduction of palladium tetraaminochloride by hydrazine hydrate (initial membranes are Accurel Q3/2 and Accurel S6/2) and (2) reduction of palladium chloride or palladium acetate by methanol (Accurel Q3/2, Celgard X50) (Volkov et al. 2009). According to the computational analysis of the SEM images, surface porosity of the initial membranes (both Accurel Q3/2 and Celgard X50) slightly increases after pretreatment. All palladium appears to be localized on the outer surface of the hollow fiber membranes. The deposition method (1) is characterized by a high deposition # Springer-Verlag Berlin Heidelberg 2012 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_1044-2

The slow wet-chemical deposition method includes the use of aliphatic alcohols (e.g., methanol) which serve as a reducing agent and as a solvent for palladium salts, and this process makes it possible to provide deposition of welldistributed fine palladium nanoparticles. In this case, loading of palladium is well below 20 mg/ cm2, and dimensions of nanoparticles tend to decrease with decreasing level of loading. The EDX and EXAFS observations reveal no other palladium-containing phases (e.g., oxides) and crystallinity of metallic palladium is proved by the XRD analysis. Dimensions of primary particles in their free state and in aggregates are estimated by the methods of X-ray analysis and SEM observations. Dimensions of primary palladium particles range from 10 to 40 nm, and dimensions of their aggregates vary from 200 nm to tens of microns. Palladium nanoclusters can catalyze DO hydrogenation, thus providing water deoxygenation by catalytic membranes, even though the loading of palladium is as low as 5 mg/cm2. The estimated surface porosity is (12 1) %, (17 2) %, and (17 3) % for initial Celgard X50, pretreated membrane, and Pd-loaded membrane (5.4 mg/cm2 Pd), respectively. For the sample containing 39 mg/cm2 of Pd, surface porosity cannot be estimated by the computer-aided SEM image processing because the deposited palladium totally occupies the entire structure of

2

Palladium-Loaded Hollow Fibers: Application in Water Deoxygenation

porous membranes. Palladium can be deposited either onto individual hollow fiber membranes or onto the surface of membranes within a membrane module. Hence, commercially available membrane module (Liqui-Cel ®) can be coated integrally as delivered, without any disassembly. Reduction of DO down to its concentration of 10 parts per billion (ppb) and lower is feasible using catalytic membrane reactors based on palladium-loaded porous hollow fibers.

References Lebedeva VI, Gryaznov VM, Petrova IV, Volkov VV, Tereshchenko GF, Shkol’nikov EI, Plyasova LM, Kochubey DI, van der Vaart R, van SoestVerecammen ELJ (2006) Porous Pd-containing polypropylene membranes for catalytic water deoxygenation. Kinet Catal 47:867–872 Volkov VV, Petrova IV, Lebedeva VI, Plyasova LM, Rudina NA, van Erkel J, van der Vaart R, Tereshchenko GF (2009) Catalytic nanoclusters of palladium on the surface of polypropylene hollow fiber membranes: removal of dissolved oxygen from water. In: Starov VM (ed) Nanoscience: colloidal and interfacial aspects. Taylor & Francis Group, New York, pp 1173–1188

W

Water Deoxygenation by Catalytic Membranes V. V. Volkov A.V.Topchiev Institute of Petrochemical Synthesis Russian Academy of Sciences (TIPS RAS), Moscow, Russian Federation

The process of catalytic removal of dissolved oxygen (DO) from water presents an encouraging example for the industrial application of palladium catalysts supported on ion-exchange resins. Noteworthy is that the existing catalytic processes for deoxygenation of water via hydrogenation reaction involve two stages: (1) absorption of hydrogen in water and (2) passage of water containing dissolved hydrogen and DO through a fixed-bed catalytic reactor. Catalytic membranes, i.e., palladium-loaded hollow fibers, can be prepared by deposition of palladium nanoparticles onto an outer surface of the hydrophobic porous polypropylene (PP) hollow fiber membranes and allow the DO removal from water in one stage (van der Vaart et al. 2001).

# Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_1045-2

The principle of DO removal using a catalytic membrane is illustrated in Fig. 1. The hydrophobic porous catalytic membrane serves three key functions: (1) a well-defined and easily controlled location of a hydrogen–water interface, (2) accessibility of a catalyst for reagents (hydrogen and oxygen), and (3) high overall mass transfer coefficient. DO-containing water flows over the outer surface of the Pd-loaded hydrophobic hollow fiber membrane, whereas hydrogen, as a reducing agent, is supplied into the lumen side of hollow fibers and approaches the working surface of a catalyst through the pores of the membranes. Due to the catalytic activation of hydrogen adsorbed on the palladium surface, a heterogeneous reaction of DO reduction takes place. Consequently, this design allows good access of both the gas phase and liquid phase reactants to the catalyst placed on the outer membrane surface, and this system is well suited for hydrogenation reactions in aqueous media. Moreover, the reaction proceeds at room temperature. The process of water deoxygenation based on catalytic membranes is capable of achieving residual oxygen concentration below 10 ppb.

2

Water Deoxygenation by Catalytic Membranes

References H2O

H2

2

1

Water Deoxygenation by Catalytic Membranes, Fig. 1 A scheme illustrating the process of DO removal from water using a Pd-loaded porous hollow fiber membrane: 1 deposited Pd, 2 polypropylene porous support (Volkov et al. 2011)

van der Vaart R, Hafkamp B, Koele PJ, Querreveld M, Jansen AE, Volkov VV, Lebedeva VI, Gryaznov VM (2001) Oxygen removal from water by two innovative membrane techniques. Ultrapure Water 18:27–32 Volkov VV, Lebedeva VI, Petrova IV, Bobyl AV, Konnikov SG, Roldughin VI, van Erkel J, Tereshchenko GF (2011) Adlayers of palladium particles and their aggregates on porous polypropylene hollow fiber membranes as hydrogenization contractors/reactors. Adv Colloid Interface 164:144–155

S

SPS/PES Asymmetric Blend Nanofiltration Membrane Yuzhong Zhang School of Material Science and Engineering, Tianjin Polytechnic University, Tianjin, China

An SPS/PES asymmetric blend nanofiltration (NF) membrane is a novel semipermeable membrane used in nanofiltration process, such as water treatment for drinking water production, desalination, concentration, and purification and pharmaceutical and chemical industries. SPS/PES asymmetric blend nanofiltration membranes are developed in 2011 by Senuo Filtration Technology (Tianjin) Co., Ltd (Chen et al. 2011). A nanofiltration membrane is a type of pressure-driven membrane with properties in between reverse osmosis (RO) and ultrafiltration (UF) membranes, with a membrane pore size between 0.5 and 2 nm, nominal molecular weight cutoffs from 200 to 1000 Da, and operating pressures between 5 and 40 bar (Hilal et al. 2004). Most nanofiltration membranes are composite membranes produced via sophisticated fabrication processes, such as coating or interfacial polymerization, where the top selective layer and bottom porous substrate of the membrane can be independently modified and optimized (Lau et al. 2012; Mulder 2009). The fabrication processes

# Springer-Verlag Berlin Heidelberg 2014 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_1049-6

for composite membrane in general involve various preparation conditions, making the entire process very labor intensive. Asymmetric blend nanofiltration membranes are made by non-solvent-induced phase separation, which is a single-step process. Non-solventinduced phase separation technology is a relatively simple preparation technique compared to composite membrane fabrication technology. And also, the blending technique has the advantage of combining the positive features of each component while being very simple. SPS/PES asymmetric blend nanofiltration membranes are fabricated by blending polyethersulfone (PES) with sulfonated polysulfone (SPS) as membrane material. PES is one of the important polymer materials in membrane fabrication, due to its excellent mechanical strength and thermal and chemical stability. PES also allows easy manufacturing of membranes, with reproducible properties and controllable size of pores down to several decade nanometers. Polyethersulfone repeating unit is shown in Fig. 1. PES is also an inherently hydrophobic polymer. A wide range of evidence shows that membrane with adequate hydrophilicity can prevent deposition of many solutes such as protein, which can increase resistance to fouling. However, hydrophobicity of PES has limited its application in NF membrane preparations. Introducing

2

SPS/PES Asymmetric Blend Nanofiltration Membrane

sulfonic groups to membrane is an efficient way to improve the hydrophilicity of PES membrane, which can be carried out through the introduction of sulfonic groups directly into PES or by blending with another sulfonated material, such as sulfonated polysulfones (SPSs). Sulfonated polysulfone (repeating unit shown in Fig. 2) with hydrophilic group, sulfonic group, has adequate hydrophilicity. The high hydrophilicity can improve the water flux and pollution resistibility. But sulfonated polysulfones possess weak strength, which will cause bad stability for operation. The miscibility of PES with SPS is better enough. Sulfonated polysulfone can be induced to increase hydrophilicity of the PES nanofiltration membrane. Besides increasing hydrophilicity, sulfonated polysulfone offers another advantage, negative

O O

S n

O

SPS/PES Asymmetric Blend Nanofiltration Membrane, Fig. 1 Polyethersulfone repeating unit SPS/PES Asymmetric Blend Nanofiltration Membrane, Fig. 2 Sulfonated polysulfone repeating unit

charge. Membranes with fixed charges are predicted to exclude solutes or colloids bearing the same sign of charge and to decrease membrane fouling. SPS/PES asymmetric blend nanofiltration membranes with functional group, sulfonic group, can gain negative charges and give them great selectivity with respect to ions or charged molecules. SPS/PES asymmetric blend nanofiltration membranes are primarily characterized by pure water flux, rejection, and hydrophilicity. An SPS/PES asymmetric blend nanofiltration membrane is a combination of good hydrophilicity and mechanical strength, which can improve the water flux, antipollution ability, and stability of blend membrane. The negative charge due to sulfonic group leads to a high selectivity over ions or molecules with positive charges. The SPS/PES asymmetric blend nanofiltration membrane has the asymmetric morphology, as shown in Fig. 3. Blending of PES with SPS in the casting dope leads to interesting asymmetric structure. There are fingerlike macropores and spongelike pores in SPS/PES membranes. A dense layer supported on a porous layer, as shown in Fig. 4. SO3H O

CH3 O

C CH3

O

S O

n

SPS/PES Asymmetric Blend Nanofiltration Membrane, Fig. 3 Typical cross section morphology of SPS/PES asymmetric blend nanofiltration membrane

SPS/PES Asymmetric Blend Nanofiltration Membrane

3

References Chen YH, Ma F, Li DW, Li Ran (2011) A preparation method for PES/SPS asymmetric blend nanofiltration membrane. China Patent CN101979132A Hilal N, AL-Zoubi H, Darwish NA, Mohammad AW, Abu Arabi M (2004) A comprehensive review of nanofiltration membranes: treatment, pretreatment, modelling, and atomic force microscopy. Desalination 170:281–308 Lau WJ, Ismail AF, Misdan N, Kassim MA (2012) A recent progress in thin film composite membrane: a review. Desalination 287:190–199 Mulder M (2009) Basic principles of membrane technology. Kluwer, London SPS/PES Asymmetric Blend Nanofiltration Membrane, Fig. 4 Typical morphology of the top surface

M

Microscale Modeling and Membranes Vasilis Burganos and Eugene Skouras Institute of Chemical Engineering Sciences, Foundation for Research and Technology, Hellas, Patras, Hellas

Microscale modeling refers to the modeling of membrane processes at a resolution level where fundamental phenomena can be described with minimum assumptions using first principles. Modeling tools at the microscopic level usually encompass molecular dynamics (MD) and/or Monte Carlo (MC) techniques, which incorporate many degrees of freedom for the detailed analysis of transport, sorption, reaction, and other process phenomena in membranes. Because of the tremendous significance of the internal structure of membranes on their performance, especially in separation or preferential transport applications (e.g., fuel cells and membrane reactors), models of the membrane structure at the microscopic level have been developed. In polymeric, “rubbery” membranes, models of gas transport in their interior are mainly based on free volume

# Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_1050-2

concepts. In the free volume approaches, the concentration dependence of the penetrant diffusivity is described by considering the average spaces between chains (Flory 1969). Such models commonly relate the mutual diffusion coefficients for a gas/polymer system to the free volume of the system. Significant efforts have been made to explain the mechanisms of gas transport in polymers by means of molecular theories. Molecular theories attempt to analyze the diffusion process in terms of specific postulated motions of the polymer chain relative to each other and the motion of penetrant molecules. In order to move into the polymer matrix, the gas molecule pushes the polymer chain and jumps into a new position (Pace and Datyner 1979). Αt around the glass transition temperature and below, thermal fluctuations of the polymer configurations are limited and penetrants are assumed to jump between cavities with a motion that is likely to entail a significant activation energy barrier (Crank and Park 1968). Simulations in glassy polymers have been attempted with the help of a dual-mode sorption model and some free volume model description. Attempts have also been made to explain gas diffusion in polymers by various molecular mechanisms and models that involve

2

Microscale Modeling and Membranes, Fig. 1 Atomistic reconstruction of SiO2 (Accelrys Material Studio)

the calculation of the energy for an assumed specific simplified polymer conformation after an energy-activated jump of the penetrant. Most models are leading to the concept of available free volume as diffusing channels (Kotelyanskii and Theodorou 2004). In fact, the mechanism of gas separation by dense, nonporous membranes is different from that by porous ones, since diffusion occurs via a solution-diffusion mechanism; the permeants dissolve in the membrane material and then diffuse through the membrane down a concentration or chemical potential gradient. Separation between the different permeants is achieved due to differences in the amount of

Microscale Modeling and Membranes

material diffusing through the membranes. The microstructure of membranes, both polymeric and inorganic, used in gas separation is usually reconstructed at the atomistic level, to study the interaction effect of solute or diffusing molecules with the membrane material. The reconstruction of inorganic membrane structures at the molecular scale is usually realized using MD and MC techniques. Initially, the atoms of the membrane materials are positioned at predefined bulk lattice sites, taken from literature, experimental data, or diffraction techniques. The interatomic potential energy of the configuration is dynamically calculated using some force field description. The force field approach involves an analytical expression that gives the energy of a molecular system in terms of the positions of all its atoms. Both intramolecular (short) and non-bonded (long) atomic interactions are usually taken into account. These data are obtained using either empirical or quantum mechanical (QM) calculations. Using a force field, useful quantities such as momenta, interaction energies, conformational energy barriers, and free energies can be estimated. The atomistic structure is the basis for MD and MC calculations to probe the locations, conformations, and motions of molecules, predicting the actual supply of sorbates to the active sites of membrane (Allen and Tildesley 1987). Mesoscopic and macroscopic membrane process parameters, such as permeabilities, diffusivities, sorption isotherms, reaction and decay coefficients, as well as (perm)selectivities, used in mesoscopic simulations derived from kinetic theory, Fickian, or Stefan-Maxwell (dusty gas) approaches, can then be derived based on proper integration of their microscopic origins (Figs. 1 and 2).

Microscale Modeling and Membranes

3

Microscale Modeling and Membranes, Fig. 2 FAU structure with equilibrated Na+ cations (Accelrys Material Studio)

References Allen MP, Tildesley DJ (1987) Computer simulation of liquids. Clarendon, Oxford Crank J, Park GS (1968) Diffusion in polymer. Academic, NY

Flory PJ (1969) Statistical mechanics of chain molecules. Wiley, NY Kotelyanskii MJ, Theodorou DN (2004) Simulation methods for polymers. Marcel Dekker, NY Pace RJ, Datyner A (1979) Statistical mechanical model for diffusion of simple penetrants in polymers. 1. Theory. J Polym Sci Part B: Polym Phys 17(3):437–451

M

Mesoscopic Transport Simulation Vasilis Burganos Institute of Chemical Engineering Sciences, Foundation for Research and Technology, Hellas, Patras, Hellas

Mesoscopic transport simulation is a type of conceptual or computational simulation of transport phenomena at the mesoscopic scale. The latter is defined as an intermediate scale between the microscopic scale, which considers molecular configurational details, and macroscopic scale, at which certain types of inhomogeneities or fluctuations have been averaged out. In the area of porous media, mesoscopic simulation of transport phenomena may refer to the pore scale itself, that is, to the scale of a few hundreds of pores or grains or to the scale at which a pore neighborhood is viewed as a thermodynamic point in space. In the former case, the characteristic length of a mesoscopic transport simulator is significantly larger than the atomic dimensions, and, as such, it contains or refers to a large number of molecules of the transported species. As a result, local values of system properties, like density, pressure, temperature, etc., can be defined in a consistent yet statistical manner. In the latter case, a continuum formulation is used to employ spatial averages over length scales much larger than the typical pore or grain scale. The concept of the thermodynamic point or of a representative # Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_1051-2

volume element (RVE) has contributed critically in this direction, sidestepping the need for treating spatial heterogeneities at smaller scales. These are reflected in pore-scale simulations, which are utilized to offer effective values for constitutive parameters that are functions of the local conditions and local structure (Bear 1972). In the context of transport phenomena in membranes, mesoscopic simulations of diffusion, flow, two-phase flow, dispersion, and ionic transport are typical examples of simulations that are performed with reference to finite samples of the membrane material. Once a mesoscopic description of the membrane sample is available, transport simulations can lead to the determination of transport coefficients at that scale. The mesoscopic description of the membrane material may involve a pore model to represent the void space of the membrane, a fiber or grain model to represent the solid phase, or a digitized model of the structure following threedimensional reconstruction. Different types of numerical tools and simulation techniques have been developed for the solution of problems involving transport phenomena at this scale. Notable examples include the classical numerical solvers for the direct solution of the transport equations complemented by appropriate boundary conditions at the faces of the working domain and at the interface between void and solid or, more generally, between the various phases that comprise the membrane. Typical examples of such equations include the diffusion equation,

2

Mesoscopic Transport Simulation

Mesoscopic Transport Simulation, Fig. 1 Representative element of porous medium

Representative element of porous medium the heat conduction equation, the Stokes or Navier–Stokes equations for flow problems, the convection–diffusion equation, etc. Another category involves the simulation of the motion of a number of molecule deputies that carry, in an efficient manner, the properties of the actual fluid molecules that are obviously too many to follow individually. Such methods include the direct simulation Monte Carlo (Bird 1963, 1976, 1994), the dissipative particle dynamics (Hoogerbrugge and Koelman 1992; Moeendarbary et al. 2009), the lattice gas method (Frisch et al. 1986; Wolfram 1986, Rothman and Keller 1988), and the lattice Boltzmann method (Benzi et al. 1992; Chen et al. 1992), which, rather than the fluid molecules themselves, follow the evolution of their probability density function across the lattice. The lattice Boltzmann method is a mesoscopic method that originated from the classical statistical physics and is based on simplified kinetic equations. A rigorous mathematical analysis that starts from the lattice Boltzmann equation has been shown to recover mesoscopic continuity and momentum equations using well-defined features of propagation–collision dynamics. Thanks to the explicit relation with details of the geometry and physics of the problem, complex boundaries and various

physicochemical phenomena can be treated. Single- and two-phase flow phenomena with or without phase transition can be studied in the context of membrane separations, including gas separation, filtration, membrane distillation, membrane emulsification, etc. (Fig. 1).

References Bear J (1972) Dynamics of fluids in porous media. American Elsevier, New York Benzi R, Succi S, Vergassola M (1992) The lattice Boltzmann-equation – theory and applications. Phys Rep 222:145–197 Bird GA (1963) Approach to translational equilibrium in a rigid sphere gas. Phys Fluids 6:1518–1519 Bird GA (1976) Molecular gas dynamics. Clarendon, Oxford Bird GA (1994) Molecular gas dynamics and the direct simulation of gas flows. Claredon, Oxford Chen S, Wang Z, Shan X, Doolen G (1992) Lattice Boltzmann computational fluid-dynamics in 3 dimensions. J Stat Phys 68:379–400 Frisch U, Hasslacher B, Pomeau Y (1986) Lattice-gas automata for the Navier–Stokes equation. Phys Rev Lett 56:1505–1508 Hoogerbrugge PJ, Koelman JMVA (1992) Simulating microscopic hydrodynamic phenomena with dissipative particle dynamics. Europhys Lett 19:155–160 Moeendarbary E, Ng T, Zangeneh M (2009) Dissipative particle dynamics: introduction, methodology and

Mesoscopic Transport Simulation complex fluid applications – a review. Int J Appl Mech 1:737–763 Rothman D, Keller JM (1988) Immiscible cellularautomaton fluids. J Stat Phys 52:1119–1127

3 Wolfram S (1986) Theory and applications of cellular automata. World Scientific, Singapore

D

Digitized Structure Model Vasilis Burganos Institute of Chemical Engineering Sciences, Foundation for Research and Technology Hellas, FORTH/ICE-HT, Patras, Hellas

Digital Reconstruction It is a procedure to represent the internal structure of materials in digital form, usually, binary, in three dimensions using data from microphotographs or other sources of spatial correlation data. It has contributed significantly to the visual comprehension of the internal configuration of porous materials, human tissues, and composite materials. In the case of porous materials, like porous membranes, catalyst supports, rocks, etc., a binary representation in three dimensions assigns the value of 0 (or 1) to the voxels that lie in the void space and the value of 1 (or 0) to the voxels in the complementary space (solid) (Torquato 2002, 2010). In this way, one obtains a fully three-dimensional array of flags (0 or 1) that defines in a discretized form the pore space inside the material. This array can then be utilized in simulations of sorption and transport phenomena (Gelb 2009), typically, diffusion, flow, two-phase flow, and dispersion that are of utmost importance in the design and modeling of membrane materials and membrane separation processes. Raw data for the initiation of the # Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_1053-2

reconstruction procedure can be obtained in various ways. Physical serial sectioning along planes that are normal to a fixed direction has repeatedly been used using special microtomes or surface grinders following impregnation of the void phase with some suitable substance (resin, Wood’s metal, etc.). Digitization of the photographs of each and every section of the material provides a set of two-dimensional arrays to be combined in a deterministic fashion into a threedimensional array that discretizes the working sample of the medium. Computer tomography provides the same type of information in an automated or semiautomated manner and enjoys tremendous use in medicine and some types of porous media, most typically in hydrocarbon reservoirs. More recently, stochastic techniques are attracting the increasing interest of investigators in this area inasmuch as they require a single section only rather than a whole series of sections. Provided that the sample is sufficiently isotropic and homogeneous, one can extract valuable geometric and topological data from the section and reconstruct in three dimensions a discretized form of the material that respects the features that characterize the two-dimensional image of the section (Quiblier 1984; Adler 1992). Such features include various moments of the correlation function, typically porosity and autocorrelation function, but also higherorder moments or lineal-length distribution, chord-length distribution, etc. The latter is succeeded by the simulated annealing technique

Digitized Structure Model Digitized Structure Model, Fig. 1 Stochastic reconstruction of porous material using the fractional Brownian motion method, showing individual volume elements

(Kirkpatrick et al. 1983) that, theoretically, is capable of reproducing three-dimensional descriptions of materials from an arbitrary number of moments of the correlation function at the expense, of course, of heavy requirements in computational time. A simpler procedure that has proven efficient for several types of porous membranes including asymmetric ones is the so-called fractional Brownian motion technique that respects the first two moments of the correlation function and, in addition, offers an interweaving option that allows three-dimensional reconstruction (Fig. 1) at much greater scale than the correlation length of the structure (Kikkinides and Burganos 2000). For materials that have evolved from powders through some agglomeration or sintering process, ballistic or random placement methods can be used that allow mass transfer between coalescing particles at a controlled fashion so as to reproduce the actual grain-size distribution and porosity or solid fraction. Reconstruction of fiber-type materials can proceed in a similar manner. In the case of porous membranes that contain nanoscale

pores, scattering techniques can be used that offer in an implicit manner the autocorrelation function, thus sidestepping the stage of microscopy and the subsequent phase of image analysis for the extraction of two-dimensional features.

References Adler PM (1992) Porous media- geometry and transports. Butterworth-Heinemann, Stoneham Gelb LD (2009) Modeling amorphous porous materials and confined fluids. MRS Bull 34:592–601 Kikkinides ES, Burganos VN (2000) Permeation properties of three-dimensional self-affine reconstructions of porous materials. Phys Rev E 62:6906–6915 Kirkpatrick S, Gelatt CD, Vecchi MP (1983) Optimization by simulated annealing. Science 220:671–680 Quiblier JA (1984) A new 3-dimensional modeling technique for studying porous-media. J Colloid Interface Sci 98:84–102 Torquato S (2002) Random heterogeneous materials – microstructure and macroscopic properties. Springer-Verlag, New York Torquato S (2010) Optimal design of heterogeneous materials. Annu Rev Mater Res 40:101–129

P

Pore Model Vasilis Burganos Institute of Chemical Engineering Sciences, Foundation for Research and Technology, Patras, Hellas

A model that involves a set or a network of pores to represent the void phase in the interior of porous materials. It is used in the context of pore structure characterization and as a structural basis for the prediction of transport properties. Traditionally used pore model types are the parallel pore model (Wheeler 1955), the random pore model (Wakao and Smith 1962) or dual porosity model, the cavity-neck model (Conner et al. 1983), and numerous variations. Because of the explicit representation of the void space that is offered by a pore model, major topological and geometrical features of the pore structure can be revealed or identified. The former relate mainly to pore connectivity as quantified by quantities like coordination number, genus per unit volume, etc., whereas the latter include pore shape, pore volume, pore size distribution, internal surface area per unit of volume, pore roughness, etc. These features can be extracted from individual or combined experimental techniques that can be divided into static (mercury porosimetry, gas adsorption/desorption, calorimetry, nuclear magnetic resonance, radiation scattering, wave propagation, etc.) and dynamic (liquid displacement, # Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_1055-2

fluid flow). In fact, pore models play an important role in the physically meaningful and explicit interpretation of the data that are produced by these techniques. Once a pore model is set up, investigations of transport phenomena in the interior of porous materials in a direct manner are greatly facilitated for the prediction of effective transport properties and for the identification of underlying mechanisms. Notable examples are diffusion of single species and multicomponent mixtures, single and multiphase flow, heat and ion conduction, convective diffusion and conduction, transport in catalytic or non-catalytic fluidsolid reaction systems accompanied by pore structure evolution, etc. In the presence of chemical reaction, as, for instance, in membrane reactors, access to and removal from the reaction front of reactants and products are affected significantly by the pore structure features, and the use of a pore model that is temporally changing is highly appreciated. The study of wetting phenomena is also greatly facilitated by the employment of pore models both for the purpose of structure characterization and for the theoretical modeling of capillary phenomena during immiscible and miscible displacement in porous media with the objective to predict percolation and breakthrough conditions. Thanks to this direct usefulness of pore models in the identification of structural features and in the prediction of transport properties and the fact that membrane separation is based on selective transport, pore models enjoy strong utilization in membrane

2

Pore Model, Fig. 1 Pore segment identification process to produce a cavity-neck model

science and engineering. More specifically, pore models are used for the understanding of the role of the membrane pore structure in the target separation process, for the elucidation of the relationship between structure and transport, and for the design of new or modified membranes for specific separations. Pore models are a major design component in numerous membrane processes, including, most notably, gas separation, ultrafiltration, microfiltration, pervaporation, membrane contactors, membrane distillation, etc. Flux equations at the individual pore scale and at the pore network scale have been developed (Maxwell-Stefan equations, dusty gas

Pore Model

equations) that describe the relationship between driving forces and resulting fluxes of the different species (Jackson 1977). The former include gradients of species concentration, partial pressure, chemical potential, pressure, temperature, etc. For pore networks that can be characterized adequately by a constant or a mean coordination number, effective medium treatments are available that homogenize the transport property of a distributed pore size network and with the help of the smooth field approximation can offer an excellent estimate of the overall transport of the pore structure (Burganos and Sotirchos 1987). In the case of variable connectivity of the pore network, standard network solutions are available in the form of numerical algorithms that practically express the continuity condition at pore intersections (Fig. 1).

References Burganos VN, Sotirchos SV (1987) Diffusion in pore networks. Effective medium theory and smooth field approximation. AIChE J 33:1678–1689 Conner WC, Lane AM, Ng KM, Goldblatt M (1983) Measurement of the morphology of high surface area solids: porosimetry of agglomerated particles. J Catal 83:336–345 Jackson R (1977) Transport in porous catalysts. Elsevier, New York Wakao N, Smith JM (1962) Diffusion in catalyst pellets. Chem Eng Sci 17:825–834 Wheeler A (1955) Catalysis, vol II. Reinhold, New York, pp 105–106

F

Fiber Models Vasilis Burganos Institute of Chemical Engineering Sciences, Foundation for Research and Technology, Hellas, Patras, Hellas

A fiber model is a model that uses straight or curved fibers, either finite in length or infinitely long, to represent the solid phase in a porous material or a reinforcement component in a composite material. Fiber models are the appropriate choice for modeling fibrous media, woven or nonwoven, typically synthetic but occasionally also natural ones. Typical examples include the representation of membranes and porous media for fuel cells (Mathias et al. 2003), filters for the separation or sieving of particulate matter, or filters for the exclusion of bubbles in diverse applications of microfluidics. There is a recent rapid growth of interest in the utilization of fiber models for the description of the structure of gas diffusion layers in fuel cells but also of modern textiles and fabrics for specialized applications (Thiedmann et al. 2009; Gaiselmann et al. 2012). The typical features that characterize a fiber model include the diameter and length of the fiber, the number or length density per unit volume, the solid fraction, the shape of the fibers,

# Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_1056-2

and the elastic properties of the fibers that determine the macroscopic mechanical properties of the material. Carbon cloth, electrospun polyacrylonitrile, and polyester are common examples of fibrous materials that can be represented by fiber models. To convert to a pore model, various attempts have been made to infer some effective pore sizes from the fiber model, usually with the help of inscribed spheres among neighboring fiber segments. Skeletonization of a fiber model is often part of the analysis routine to facilitate the comprehension of the fiber cluster articulation and the eventual determination of the topology of the system. Useful concepts from straight-line path statistics or randomness of secant distribution through convex bodies (Coleman 1969) are incorporated in this type of models. The fibers can be hollow or solid, randomly oriented or ordered to arbitrary degree, and either charged or neutral depending on the application. Fiber models of membrane materials and, more generally, fibrous media lend themselves to the numerical simulation of transport phenomena through their structure, usually diffusion, single or, two-phase flow, dispersion, particle attraction and deposition, combined phase transition and flow, heat conduction, electrical conduction, and light transmission (Torquato 2002; Tomadakis and Sotirchos 1993).

2

References Coleman R (1969) Random paths through convex bodies. J Appl Probab 6:430–441 Gaiselmann G, Thiedmann R, Manke I, Lehnert W, Schmidt V (2012) Stochastic 3D modeling of fiberbased materials. Comput Mater Sci 59:75–86 Mathias MF, Roth J, Fleming J, Lehnert W (2003) Diffusion media materials and characterisation. In: Vielstich W, Lamm A (eds) Handbook of fuel cells, volume III, chapter 42, 517–537, J. Wiley & Sons, London

Fiber Models Thiedmann R, Hartnig C, Manke I, Schmidt V, Lehnert W (2009) Local structural characteristics of pore space in GDLs of PEM fuel cells based on geometric 3D graphs. J Electrochem Soc 156:B1339–B1347 Tomadakis MM, Sotirchos SV (1993) Ordinary and transition regime diffusion in random fiber structures. AIChE J 39:397–412 Torquato S (2002) Random heterogeneous materials – microstructure and macroscopic properties. SpringerVerlag, New York

G

Grain Models Vasilis Burganos and Eugene Skouras Institute of Chemical Engineering Sciences, Foundation for Research and Technology, Hellas, Patras, Hellas

Grain models involve a set of grain-shaped objects to represent the solid phase in the interior of porous materials. Model representation of the structure of porous membranes is very important for visualizing and understanding the structure of amorphous, porous, and random membrane materials and can be implemented, among other ways, by deterministic or stochastic modeling of the outcome of the actual fabrication process (e.g., random assembly or sequential deposition of particles) with or without reference to the detailed physics of the process (Preparata and Shamos 1985). Typically, such reconstructions involve some random packing, optionally combined with ballistic deposition of hard or soft spheres, disks, or ellipsoids of prolate or oblate geometry (Rogers 1964). This stage is usually followed by simulation of grain sintering as the result of thermal or viscous sintering. In the majority of packing procedures, random sequential deposition of overlapping or non-overlapping particles takes place (Jaeger and Nagel 1992). The particles position themselves either randomly or under the influence of some unidirectional force (ballistic) or, alternatively, toward # Springer-Verlag Berlin Heidelberg 2014 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_1057-2

some fixed center of attraction (Finney 1970). Ballistic random sphere packs are usually driven by gravity and are considered to be more representative of the compaction process. Ballistic deposition and central attraction methods provide packings with similar structural properties in terms of the autocorrelation function. Depending on the type of fabrication process as well as on the characteristic dimensions of the grains, other forces, either local/short ranged (van der Waals, double layer, hydrodynamic) or long ranged (electrostatic, electromagnetic), can also be taken into account during simulations. One, two, or three points of contact can be assumed to discontinue the downhill motion of the particle. In energy-based models, downhill motion of a particle continues until a position of local energy minimum is reached that is considered stable. Such a procedure can be systematically followed through the use of a steepest descent method followed by a conjugate gradient algorithm (Conway and Sloane 1993). Procedures based on Monte Carlo or similar statistical methods have also been implemented, where each time a number of test grains are inserted, but only selected displacements are allowed that lead to minima of position or energy. Motion and interaction of granular particles have also been studied mechanistically with computational methods, such as the discrete element method (DEM), also called a distinct element method (Zhu et al. 2007, 2008), for computing the motion and effect of a large number of small particles. Although DEM

2

Grain Models

Grain Models, Fig. 1 Ballistic reconstruction followed by partial sintering

is closely related to molecular dynamics, the method is generally differentiated by its inclusion of rotational degrees of freedom, as well as compacted and other complex granular geometries. Grain models have also been used for the description of micro- or nanoparticle additives, also called nanofillers (at least one of the particles’ dimensions is in the nanometer range), into a polymer membrane to form a polymeric composite membrane or a hybrid, mixed matrix membrane (Sanchez et al. 2005). This is a promising idea to improve the separation properties of the neat polymer and produce the so-called nanobarriers with numerous applications in gas separation, fuel cells, preservation of sensitive products, etc. The enhancement of barrier properties through the addition of inorganic nanoparticles in polymer matrices can be studied with the help of grain models and elucidate their complex dependence on interfacial interactions, filler shape and orientation with respect to the transport direction, filler size, distribution of

inter-filler distances, and degree of agglomeration (Fig. 1).

References Conway JH, Sloane NJA (1993) Sphere packings, lattices, and groups, 2nd edn. Springer, New York Finney JL (1970) Random packings and the structure of simple liquids. I. The geometry of random close packing, Proc R Soc Lond A-Math Phys Sci 319:479–493 Jaeger HM, Nagel SR (1992) Physics of granular states. Science 255(5051):1523–1531 Preparata FP, Shamos MI (1985) Computational geometry: an introduction. Springer, New York Rogers CA (1964) Packing and covering. Cambridge University Press, Cambridge Sanchez C, Julian B, Belleville P, Popall M (2005) Applications of hybrid organic-inorganic nanocomposites. J Mater Chem 15:3559–3592 Zhu HP, Zhou ZY, Yang RY, Yu AB (2007) Discrete particle simulation of particulate systems: theoretical developments. Chem Eng Sci 62:3378–3392 Zhu HP, Zhou ZY, Yang RY, Yu AB (2008) Discrete particle simulation of particulate systems: a review of major applications and findings. Chem Eng Sci 63(23):5728–5770

A

Amorphous Polymers and the Amorphous Region Petr Sysel Department of Polymers, Institute of Chemical Technology, Prague, Czech Republic

Polymers consist of the large molecules (polymer chains). An arrangement of these chains in the bulk polymer is given by their chemical composition, spatial orientation (e.g., an isotactic vs atactic polypropylene), molecular weight, and processing conditions (e.g., a cooling rate). Based on these factors, bulk polymers show an amorphous or partially crystalline (semicrystalline) phase state. In the amorphous state, an ordering of the (entangled) polymer chains is random. The amorphous polymer does not include a long-range order, but it is characterized by the existence of some regularity on a short-range order. As a consequence of it, an X-ray diffraction provides a diffuse ring only (a set of discrete rings is typical for well-ordered structures). If the amorphous polymer is exposed to a gradually increased temperature, it passes from a glass (hard and rigid) to a rubber (soft and flexible) and finally to a molten consistency. Boundaries between the glass/rubber and rubber/ molten consistency are given by a glass transition temperature (Tg) and flow temperature, respectively (see a thermomechanical curve of an amorphous polymer, Fig. 1). # Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_1059-5

The glass transition temperature of the polymer is also an important indicator of its application temperature region (e.g., polydimethylsiloxane (Tg = ca 120 C) is rubbery and polystyrene (Tg = ca +100 C) is a glassy polymer at room temperature). A character of the main backbone (flexible/ rigid), its substitution (small vs bulk substituents), and a level of molecular (chain) interactions (e.g., van der Waals forces vs hydrogen bonds) are important factors influencing the position (value) of this transition. A flexibility of the chains (or segments) of an amorphous polymer is restricted at temperatures below its Tg. Under heating, molecular interactions (secondary forces) are disrupted, and a free volume among more flexible chains increases (Fig. 2). An extent of this (segmental) chain motion and free volume influences a lot of the polymer properties, e.g., a transport of the different low-molecular-weight media through the membranes made of these polymers (Strathmann 2011). For example, a nonporous (dense) membrane made of a rubbery polysiloxane [( Si(R)2-O-)n, where R is often CH3] shows a much higher gas permeation (and also diffusivity) in comparison with those made of common glassy polymers at room temperature (Pandey 2001; Bernardo et al. 2009). But their selectivity is traditionally low (Robeson 2008). Some polymers (e.g., polyethylene, polyamides) include both amorphous and crystalline fractions, and they are called semicrystalline

2

Amorphous Polymers and the Amorphous Region

deformation

glassy state

rubbery state

plastic state

flow temperature

glass transition temperature

temperature

Amorphous Polymers and the Amorphous Region, Fig. 1 Thermomechanical curve of an amorphous polymer

Amorphous Polymers and the Amorphous Region, Fig. 2 Temperature dependence of the specific volume of an amorphous polymer

specific volume

glass transition temperature temperature

ones. Chains of such polymers may be organized in spherulites having a well-ordered lamellar structure (i.e., a folded chain) with an incorporated disordered amorphous fraction (Fig. 3). Amorphous polymers are usually transparent; semicrystalline ones are opaque. A segmental mobility of the amorphous regions in

semicrystalline polymers varies from that of the amorphous polymer. These regions are affected by the presence of crystalline phase, and this phenomenon influences also transport characteristics of the penetrating media. Traditionally, the “more dense” crystalline phase decreases a chain mobility and prolongs a transport trajectory of

Amorphous Polymers and the Amorphous Region Amorphous Polymers and the Amorphous Region, Fig. 3 Schematic arrangement of the spherulite

3

amorphous region

lamellar structure

penetrating media, and – as a consequence of this – their permeability decreases (Pandey 2001). Nevertheless, some exceptions were also found, e.g., an increase of gas permeability in a polylactide membrane with its increasing crystallinity (Colomines et al. 2010).

References Bernardo P, Drioli E, Golemme G (2009) Membrane gas separation. Ind Eng Chem Res 48:4638–4663

Colomines G, Ducruet V, Courgneau C, Guinault A, Domenek S (2010) Barrier properties of poly(lactic acid) and its morphological changes induced by aroma compound sorption. Polym Int 59:818–826 Pandey P, Chauhan RS (2001) Membranes for gas separation. Prog Polym Sci 26:853–893 Robeson LM (2008) The upper bound revisited. J Membr Sci 320:390–400 Strathmann H (2011) Introduction to membrane science and technology. Wiley-VCH, Weinheim

D

Density Functional Theory Modeling of Membrane Systems Giorgio De Luca Institute on Membrane Technology ITM-CNR, University of Calabria, Rende (CS), Italy

Density functional theory is a new methodology in the frame of quantum mechanics (ab initio). The assessment of material features depending on the electron interactions or electron density polarization necessarily needs the use of quantum mechanics. Large molecular systems cannot be described with accurate quantum mechanical approaches, such as correlated Hartree-Fock (Szabo and Ostlund 1994), due to the huge computational time required by these. Instead, the density functional theory (Parr and Yang 1989) allows to get results with similar precision in relatively shorter computational time. Thus, density functional theory is a powerful tool to investigate large chemical systems like the nanostructures involved in the membrane preparation. In density functional theory, the total energy of an electronic system is evaluated through a total functional which depends on the electron density of the quantum system, r(r), and external potential v(r). It is defined as follows:

# Springer-Verlag Berlin Heidelberg 2012 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_1062-1

EDFT ½rðrÞ ¼ Ts ½rðrÞ þ J½rðrÞ þ Exc ½rðrÞ ð þ vðrÞrðrÞdr where Ts[r(r)] is the kinetic energy of an isoelectronic noninteracting system, while J[r(r)] describes the Coulomb electrostatic energy. The Exc[r(r)] is the exchange and correlation functional, which takes into account the difference between the kinetic energies of the isoelectronic interacting and noninteracting systems in addition to the difference between the quantum electron-electron and Coulomb electrostatic energy. r(r) can be evaluated by the nonlinear Kohn-Sham equations, defined by means of an effective potential. The electron density allows to calculate all the properties of the quantum system in addition to the total energy. Computational approaches, based on density functional theory, are applied in the study of catalysts used in membranes and in the study of noncovalent interactions, such as hydrogen-bonding and London dispersion interactions (De Luca et al. 2009). In fact, hydrogen-bonding, electrostatic interactions and London force are important for membranes at the basis of fundamental properties such as molecular adsorption and sorption, recognition, and self-assembly. Also size, shape, and electrostatic features of supramolecular architectures can be studied using density functional theory. All these properties control the selectivity of the materials, used in membranes, the permission as well as the antifouling or anti-embrittlement

2

Density Functional Theory Modeling of Membrane Systems

features. Density functional theory studies of the catalysis in membrane reactors require the definition of structural models of the catalysts. For the most part, they are modeled by infinite surfaces, slabs, or different types of adsorbed or absorbed atomic clusters. These studies could be treated independently from the aforementioned analysis. However, it is important to emphasize that the merging of the results obtained by the different investigations would be advisable. For example, the kinetic constants characterizing a reaction path, related to a particular catalyst, evaluated by the density functional theory, should be compared with the diffusion coefficients characterizing the permeability of reagents

and products through the membrane derived by the adsorption/absorption of different molecules.

References De Luca G, Gugliuzza A, Drioli E (2009) Competitive hydrogen-bonding interactions in modified polymer membranes: a density functional theory investigation. J Phys Chem B 113:5473–5477 Parr RG, Yang W (1989) Density functional theory of atoms and molecules. Oxford University Press, New York Szabo A, Ostlund NS (1994) Modern quantum chemistry. Macmillan Publishing CO, New York

E

Ethanol-Water Mixtures: Separation by Pervaporation Kew-Ho Lee Department of Chemical and Biomolecular Engineering, National Research Laboratory for Environmental Catalysis, Korea Research Institute of Chemical Technology (KRICT), Yuseong-gu, Daejeon, South Korea

Pervaporation is an important membrane process in chemical industries in which valuables are isolated from the liquid mixture. Liquid and vapor separation by thermal processes has always been highly energy intensive, and new separation processes taking advantage of mass transfer through dense membranes have already shown they enable very significant energy savings as compared to more classic technologies (Anne et al. 2002). Membranes can be used for the selective removal of water from aqueous organic mixtures. Pervaporation (PV) is a separation process that involves separation of liquid mixtures, in contact with a membrane. With feed solution on one side, permeate is removed as a vapor from the other side (Brian et al. 2011); pervaporation (PV) is a very well-known membrane process for the separation of liquid and vapor mixtures due to its energetic aspects (EP 909209A1 1999; EP 944575A1 1999; EP 880400A1 1998). Pervaporation mostly allows a variety of possible application areas: dewatering of organic fluids # Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_1063-1

like alcohols, ketones, ethers, etc. (EP 765682A1 1997); separation of mixtures from narrow boiling temperatures to constant (azeotrope) boiling temperatures (EP 811420A1 1997); removal of organic pollutants from water and air streams (EP 749351A1 1996); separation of fermentation products; and separation of organic-organic liquid mixtures (Kujawski 2000). Pervaporation is also considered as so-called clean technologies, especially well suited for the treatment and recycling of volatile organic compounds and pollution prevention (Anne et al. 2002). Transport mechanism of PV through polymeric membranes was studied by many research groups, and it was explained by the solutiondiffusion model (Binning et al. 1961; Paul and Paciotti 1975; Lee 1975; Mulder and Smolders 1984; Kataoka et al. 1991a, b). According to the solution-diffusion model, each component of the permeation molecules dissolves into the membrane and diffuses through the membrane due to the concentration gradient (Mikihiro et al. 1998) (Fig. 1). Transport through the membrane is driven by the vapor pressure difference between the feed solution and the permeate vapor. The vapor pressure difference can be maintained by applying a vacuum on the permeate side or by cooling the permeate vapor so that it condenses, thus creating a partial vacuum. Commercial systems for the dehydration of concentrated alcohol and other solutions have been developed since the 1980s, much of the push coming from interest in the

2 Ethanol-Water Mixtures: Separation by Pervaporation, Fig. 1 Solution-diffusion mechanism (Graham 1866)

Ethanol-Water Mixtures: Separation by Pervaporation

1. Sorption Micrivoids Polymer

2. Diffusion Microchannels

3. Desorption

production of pure ethanol as an alternative liquid fuel, where PV can be used to dehydrate (Brian et al. 2011).

Pervaporation Membranes Polymeric Membranes For dehydration, where the small molar volume favors the preferential sorption of water, materials have to be selected with a higher affinity for water than for the other component. The polymeric materials can be broadly classified into three categories: glassy polymers, rubbery or elastomeric polymers, and ionic polymers. In general, the glassy and ionic polymers are more suited for making water-selective membranes for dehydration. For water-selective membranes, the most important factor responsible for the separation is the specific interaction between water and the polymer. To obtain high selectivities, it is necessary to use polymers, which contain specific groups/active centers, capable of strong interactions with water. The highest fluxes are those for the hydrophilic membranes based on cellulose and Nafion and grafts of hydrophilic poly(vinylpyrrolidone) on Teflon and polyacrylonitrile. The PVA/TEOS

membranes are exceptions in that they are hydrophilic but exhibit low fluxes (Brian et al. 2011) (Table 1). The emphasis is on selectivity, postulated to be determined by selective sorption and selective diffusion. Selective sorption is governed by the presence in the membrane of active centers such as charged sites which are capable of specific interaction with water, while selective diffusion is governed by the rigidity and regularity of the polymer structure and the nature of the polymer interspace, exemplified by the degree of swelling and the frequency of the cross-links. The results for a series of membranes made by grafting neutral or charged polymers onto supporting membranes are reported in (Table 2). Polysalts, formed from anionic and cationic polyelectrolytes, would be appropriate for obtaining both highly permeable and highly selective membranes (Semenova et al. 1997) (Table 3). The best performers in terms of flux, which at a maximum of 5 kg/m2h never achieve high values, are charged polymers of one type or another, including polysalts. Anionic and polysalt membranes are superior. For anionic polymers, the proton form has a significantly higher flux than the metal or quaternary ammonium salt versions, owing to the greater free space within the polymer network (Table 4).

Ethanol-Water Mixtures: Separation by Pervaporation

3

Ethanol-Water Mixtures: Separation by Pervaporation, Table 1 PV dehydration of ethanol through various polymeric membranes (Brian et al. 2011) Polymer Regenerated cellulose Cellulose acetate Teflon-g-polyvinylpyrrolidone Perfluorinated polymer on PAN support Nafion-H+ Polyacrylonitrile-polyvinylpyrrolidone Poly(maleimide-co-acrylonitrile) Poly(acrylic acid-co-acrylonitrile) Polystyrene Poly(vinyl chloride) Alginic acid Chitosan Chitosan acetate salt Chitosan/glutaraldehyde PVA/25 % TEOS, annealed at 160 C PVA/25 % TEOS, annealed at 130 C

Feed (wt% water) 50 4 4 1.3 4 4 15 18 4 4 4 5 4 4 4 15 15

Temp. ( C) 45 60 25 50 70 20 15 15 40 40 40 60 40 40 40 40 40

Separation factor 5.0 5.9 2.9 387 2.5 3.2 33 877 101 63 8.8 13 2208 2556 202 329 893

Flux (g/m2h) 2060 200 2200 1650 5000 2200 8 13 5 3 48 2800 4 2 7 5 4

Ethanol-Water Mixtures: Separation by Pervaporation, Table 2 PV dehydration of 20 % aqueous ethanol at 70 C using graft polymer membrane having different charges (Brian et al. 2011) Host polymer Polyvinylidene fluoride Polyvinylidene fluoride Polyvinyl fluoride Polyvinyl fluoride Polyacrylonitrile Polyacrylonitrile Polyvinyl fluoride Polyvinylidene fluoride Polyvinylidene fluoride Polyvinylidene fluoride Polyvinyl fluoride

Grafted polymer 4-Vinylpyridine N-Vinyl-imidazole N-Vinylmethyl-acetamide N-Vinyl-pyrrolidone Acrylic acid K+ acrylate K+ acrylate Quaternized 4-vinylpyridine Quaternized N-vinylimidazole 4-Vinylpyridine/BrCH2COOH Vinylimidazole/BrCH2COOH

Hybrid Membranes Many a times, the polymeric membranes may fail to meet the desired separation requirements. In such cases, it becomes necessary to add filler materials such as ceramics and zeolites to improve the separation properties of the membrane. There

Graft site charge Neutral Neutral Neutral Neutral Neutral Anionic Anionic Cationic Cationic Zwitterionic Zwitterionic

Separation factor 9 10 4 7 10 500 156 175 61 76 63

are several reports showing good separation performance for ethanol/water mixture using zeolite membranes (Kita et al. 1995; Sano et al. 1994). Kita et al. made NaA-type zeolite membrane by hydrothermal synthesis. NaA zeolite membrane is a water-selective membrane, and the PV

4

Ethanol-Water Mixtures: Separation by Pervaporation

Ethanol-Water Mixtures: Separation by Pervaporation, Table 3 PV dehydration of aqueous ethanol with membrane based on various polysalts (Brian et al. 2011) Polyanion Poly(acrylonitrile-coacrylic acid) Cellulose-SO3-Na+ Cellulose-SO3-Na+ Cellulose-SO3-Na+ Cellulose-SO3-Na+ Aromatic polyamide sulphonate Poly(acrylic acid) On polysulphone No supporting Na+ polystyrene sulphonate Na+ CMC Na+ CMC Anionic PVA DS 2.3 % DS 5.0 %

Polycation Poly(acrylonitrile-co-vinyl pyridine) Polyethyleneimine PolyDADMAC, linear Same, but branched Poly-N, N-dimethyl-3, 5dimethylenepiperidine chloride Polyethyleneimine Chitosan supporting membrane membrane Polyallylamine HC1 Chitosan N-Ethyl-4-vinyl-pyridinium bromide Cationic PVA DS 2.9 % DS 5.2 %

Feed (wt% water) 10

Temp. ( C) –

Separation factor 5000

Flux (g/m2h) 400

16 16 16 16

50 50 50 50

295 140 123 123

1900 3200 4900 2700

20

60

15

300

5 5 6.2

30 30 70

1008 2216 70

132 33 230

10 10

70 70

1062 782

1140 1320

4.6 4.6

75 75

2250 1910

378 284

Ethanol-Water Mixtures: Separation by Pervaporation, Table 4 Highest fluxes for PV dehydration of aqueous ethanol (Brian et al. 2011) Membrane polymer Nafion-H+ Cellulose-SO3-Na+ and polyDADMAC, branched K+ acrylate graft on poly(vinyl fluoride) PEI/PAA on RO membrane Cellulose-SO3-Na+ and polyDADMAC, linear K+ acrylate graft on PAN

Mem. type Anionic Polysalt

Feed (wt% water) 4 16

Temp. ( C) 70 50

Flux (g/m2h) 5000 4900

Separation factor 2.5 125

Anionic

20

70

4700

156

Polysalt Polysalt

10+ 16

70 50

4050 3200

1075 140

Anionic

20

70

3000

500

separation factor of water/ethanol system was over 10,000 at 348 K. For ethanol permselective membranes, Sano et al. (1994) prepared polycrystalline silicalite membrane by the hydrothermal synthesis.

The silicalite membrane showed high ethanol permselectivity, and a separation factor of 58 was realized at 333 K by PV. Silicalite membranes seem to have great potential for the ethanol recovery by PV (Table 5).

Ethanol-Water Mixtures: Separation by Pervaporation

5

Ethanol-Water Mixtures: Separation by Pervaporation, Table 5 PV dehydration of ethanol using PVA/inorganic hybrid membranes (Brian et al. 2011) Crosslinker TEOS (160 C) TEOS (130 C) PEG blend and TEOS No PEG Poly(acrylic acid) copolymer and TEOS g-Aminopropyl-triethoxysilane Sulphated zirconia

Feed (wt% water) 15 15 15 15 15 5 5 10 20 30

References Anne J, Robert C, Pierre LND, Bruno C (2002) Industrial state-of-the-art of pervaporation and vapour permeation in the western countries. J Membr Sci 206:87–117 Binning RC, Lee RJ, Jennings JF, Martin EC (1961) Separation of liquid mixtures by permeation. Ind Eng Chem 53:45 Brian B, Manh H, Zongli X (2011) A review of membrane selection for the dehydration of aqueous ethanol by pervaporation. Chem Eng Process 50:227–235 EP 749351A1 (1996) Device for separating mixtures or for purifying substances by pervaporation EP 765682A1 (1997) Apparatus for separating liquid media with two membranes having their primary sides connected by an intermediate space EP 811420A1 (1997) Composite membrane for selective separating organic substances by pervaporation EP 880400A1 (1998) Composite membrane with a support membrane made in particular of a microporous material EP 909209A1 (1999) Pervaporisation and module for carrying out said process EP 944575A1 (1999) Esterification of fermentationderived acids via pervaporation Graham T (1866) On the absorption and dialytic separation of gases by colloid septa. Philos Mag J Sci 32:401–420 Kataoka T, Tsuru T, Nakao S, Kimura S (1991a) Permeation equations developed for prediction of membrane performance in pervaporation, vapor permeation and reverse osmosis based on the solution diffusion model. J Chem Eng Jpn 24:326

Temp. ( C) 40 40 50 50 40 50 50 50 50 50

Separation factor 329 893 300 160 250 537 263 142 86 61

Flux (g/m2h) 50 40 46 500 18 36 10 105 183 1036

Kataoka T, Tsuru T, Nakao S, Kimura S (1991b) Membrane transport properties of pervaporation in ethanolwater system using polyacrylonitrile and cellulose acetate membranes. J Chem Eng Jpn 24:334 Kita H, Horii K, Ohtoshi Y, Tanaka K, Okamoto K (1995) Synthesis of a zeolite NaA membranefor pervaporation of water/organic liquid mixtures. J Mater Sci Lett 14:206 Kujawski W (2000) Application of pervaporation and vapor permeation in environmental protection. Pol J Environ Stud 91:13–26 Lee CH (1975) Theory of reverse osmosis and some other membrane permeation operations. J Appl Polym Sci 1983 Mikihiro N, Takeo Y, Sin-ichi N (1998) Ethanol/water transport through silicalite membranes. J Membr Sci 144:161–171 Mulder MHV, Smolders CA (1984) On the mechanism of separation of ethanol/water mixtures by pervaporation I. Calculations of concentration profiles. J Membr Sci 17:289 Paul DR, Paciotti JD (1975) Driving force for hydraulic and pervaporative transport in homogeneous membranes. J Polym Sci 13:1201 Sano T, Yanagishita H, Kiyozumi Y, Mizukami F, Haraya K (1994) Separation of ethanol/water mixture by silicalite membrane on pervaporation. J Membr Sci 95:221 Semenova SI, Ohya H, Soontarapa K (1997) Hydrophilic membranes for pervaporation: an analytical review. Desalination 110:251–286

P

Pd–Cu Alloys for Hydrogen Separations Kew-Ho Lee Center for Membranes, Korea Research Institute of Chemical Technology (KRICT), Yuseong-gu, Daejeon, South Korea

Pd–Cu alloys have been widely studied as a hydrogen separation membrane because they do not show embrittlement even at low temperatures (Kulprathipanja et al. 2005; Miller et al. 2008). Studies by Morreale et al. have shown that the face-centered cubic (fcc) phase is more resistant to sulfur than the body-centered cubic (bcc) phase (Morreale et al. 2004). In transient experiments, the fcc Pd–Cu composition showed a decline of 0–10 % when exposed to 1,000 ppm of sulfur, while a bcc Pd–Cu composition had a decline of 99 %. Studies by Pomerantz and Ma (Pomerantz and Ma 2009) confirm these results for Pd–Cu compositions of 8, 18, and 19 wt.% Cu, with permeance losses of 80 % at 500 C (773 K). They further showed that the loss was partially reversible by hydrogen treatment. A recent study by Howard and coworkers (O’Brien et al. 2010) shows that Pd exposed to 1,000 ppm sulfur at 350 C (623 K) corrodes over a period of hours to form a thick (6.6 mm) PdS4 layer, probably by an autocatalytic process. In contrast a Pd47Cu53 alloy forms a thin (3 nm) Pd–Cu–S layer. Although this layer cannot dissociate hydrogen # Springer-Verlag Berlin Heidelberg 2015 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_1064-1

or is impermeable to hydrogen, it does protect the bulk from sulfidation and could be removed by a hydrogen treatment. The alloying of Cu with Pd increases the resistance for the a to b phase transition in Pd–H system. Moreover the phase transition in the Pd–H system is eliminated even at room temperature by adding more than 8 wt.% Cu in Pd (Karpova and Tverdovskii 1959). McKinley (McKinley et al. 1969) studied the permeation characteristics of Pd–Cu alloys with different compositions. The H2 permeabilities of various alloys are shown in Table 1. Out of all other Pd–Cu alloys, the Pd60Cu40 shows the highest permeability which corresponds to the ordered b-phase in Pd–Cu system (McKinley and US. Patent 3, 439, 474 1969). In contrast, it is reported that Pd47Cu53 (mol%) alloy has the highest hydrogen permeability at 350 C among the Pd–Cu alloys (Yuan et al. 2007; Yang et al. 2007; McKinley et al. 1969; Roa et al. 2002) and is comparable in its hydrogen permeability to pure Pd at the same temperature. The solubility of Pd–Cu (20 % Cu) is five times smaller than that of pure Pd, which corresponds to the permeability behavior of that alloy (Sonwane et al. 2006). In summary Pd–Cu (particularly Pd60Cu40) is the most promising composition; has the characteristics of having high hydrogen permeance, good sulfur resistance, robustness w.r.t. thermal cycling, and an excellent dimensional stability (small degree of swelling); is cheaper; eliminates

2

Pd–Cu Alloys for Hydrogen Separations

Pd–Cu Alloys for Hydrogen Separations, Table 1 Permeabilities of Pd–Cu alloys with different wt.% of Cu at 350 C, 300 psig wt.% 90Pd–10Cu 70Pd–30Cu 60Pd–40Cu 550Pd–45Cu 45Pd–55Cu 90Pd–10Cu

Permeability (Cm3/cm2.s) 0.69 0.12 1.52 0.25 0.01 0.69

warping and cracking; and avoids a to b phase transition in pure Pd (Gabitto and Tsouris 2009).

References Gabitto JF, Tsouris C (2009) Sulfur poisoning of metal membranes for hydrogen separation. Int J Chem Eng 1:394–411 Karpova RA, Tverdovskii IP (1959) Sorption of hydrogen by disperse Pd-Cu alloys. Zhur Fiz Khim 33:1393 Kulprathipanja A, Alptekin GO, Falconer JL, Way JD (2005) Pd and Pd–Cu membranes: inhibition of H2 permeation by H2S. J Membr Sci 254:49–62 McKinley DL (1969) US. Patent 3,439,474

McKinley DL (1969) Method for hydrogen separation and purification, USA Miller JB, Morreale BD, Gellman AJ (2008) The effect of adsorbed sulfur on surface segregation in a polycrystalline Pd70Cu30 alloy. Surf Sci 602:1819–1825 Morreale BD, Ciocco MV, Howard BH, Killmeyer RP, Cugini AV, Enick RM (2004) Effect of hydrogensulfide on the hydrogen permeance of palladiumcopper alloys at elevated temperatures. J Membr Sci 241:219–224 O’Brien CP, Howard BH, Miller JB, Morreale BD, Gellman AJ (2010) Inhibition of hydrogen transport through Pd and Pd47Cu53 membranes by H2S at 350 C. J Membr Sci 349:380–384 Pomerantz N, Ma YH (2009) Effect of H2S on the performance and long-term stability of Pd/Cu membranes. Ind Eng Chem Res 48:4030–4039 Roa F, Block MJ, Way JD (2002) The influence of alloy composition on the H2 flux of composite Pd–Cu membranes. Desalination 147:411–416 Sonwane CG, Wilcox J, Ma YH (2006) Solubility of hydrogen in PdAg and PdAu binary alloys using density functional theory. J Phys Chem B 110:24549–24558 Yang JY, Nishimura C, Komaki M (2007) Effect of H2S on hydrogen permeation of Pd60Cu40/V–15Ni composite membrane. J Alloys Compd 446–447 Yuan L, Goldbach A, Xu H (2007) Segregation and H2 transport rate control in body centered cubic PdCu membranes. J Phys Chem B 111:10952–10958

E

Emulsion Emma Piacentini Institute on Membrane Technology. (ITM-CNR), University of Calabria, Rende CS, Italy

An emulsion consists of two immiscible liquids (usually oil and water) with one of the liquids (dispersed phase or internal) dispersed as a form of spherical droplets in the other (continuous phase or external) (Israelachvili 1994). Depending upon the nature of the dispersed phase, the emulsions are classified as (i) oil-in-water emulsions (O/W) consisting of oil droplets dispersed in an aqueous phase and (ii) water-in-oil emulsions (W/O) consisting of aqueous droplets dispersed in an oil phase. It is also possible to prepare various types of multiple emulsions, for example, water-in-oil-inwater emulsions (W/O/W), in which water droplets are dispersed within larger oil droplets which are themselves dispersed in an aqueous phase and oil-in-water-in-oil emulsions (O/W/O) consisting of oil droplets dispersed in larger water droplets which are themselves dispersed in an oil phase. The preparation of an emulsion is termed emulsification and the agents used for this purpose are termed emulsifiers. Other agents, such as emulsion promoters or stabilizers, are often added to an emulsion to promote the emulsifying process, for example, by increasing the viscosity or providing a protective colloid action. The preparation of emulsions involves breaking up # Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_1066-1

the internal phase by supplying mechanical or chemical energy. When an emulsion is formed, the interface between the phases is considerably increased as a result of the droplet formation. The liquid always tends to reduce its surface or interface to a minimum; therefore, an increase in interface is possible only if energy is supplied. The work that must be expended on drop division is: dA ¼ g dI where dA is the work to be expended and dI is the increase in interface. The proportionality factor is the interfacial tension g between the phases to be emulsified. Thus, if the interfacial tension between the two phases is high, considerable mechanical energy is required for emulsification unless an emulsifier is added; if the interfacial tension is low, little mechanical energy is consumed. According to the droplet size, emulsions are classified as follows: • Macroemulsions: these usually have a size range of 0.1–5 mm. • Nanoemulsions: these usually have a size range of 20–100 nm. • Micellar emulsions or microemulsions: these usually have a size range of 5–50 nm.

OSTWALD RIPENING

FLOCCULATION

CREAMING

SEDIMENTATION

If the droplet size exhibits a wide statistical distribution, the emulsion is described as polydisperse, in contrast to monodisperse systems with a uniform droplet size. The ideal particle size depends on the available methods of preparation and industrial application in each case. Another important emulsion property is the ratio of the volume of the dispersed phase (Vi) to the volume of the continuous phase or (Ve) is called the phase volume ratio (F). If F < 0.43 (Vi = 30 % of total volume), the flow properties of the emulsion are determined primarily by the continuous phase. If F > 0.43, the viscosity of the emulsion generally increases. As F increases, either phase reversal or cream formation occurs. Emulsion stability should match its application. Thus, for a number of applications, the emulsion should be stable under very specific conditions, but it should break after its purpose has been achieved according to a specific condition (such as temperature, pH, or salts, or the like). An emulsion is stable if fusion of the droplets is prevented by a sufficiently high energy barrier (Tadros 2013). In general, the energy barrier is built up by the film of emulsifier that forms at the surface of the droplets. Several breakdown

EMULSION

COALESCENCE

Change in droplet size and size distribution

Emulsion, Fig. 1 The various breakdown processes in emulsions

Emulsion

No change in droplet size and size distribution

2

PHASE INVERSION

processes may occur on storage depending on particle size distribution and density difference between the droplets and the medium (Fig. 1). In sedimentation, the uniform dispersion of the droplets is disturbed by aggregation, which leads to settling or creaming of the internal phase. This process results from external forces usually gravitational or centrifugal. When such forces exceed the thermal motion of the droplets (Brownian motion), a concentration gradient builds up in the system with the larger droplets moving faster to the top (if their density is lower than that of the medium) or to the bottom (if their density is larger than that of the medium) of the container. To keep an emulsion stable, such aggregation must be prevented since droplet aggregates sediment faster than individual small droplets. Sedimentation is not always necessarily accompanied by coalescence. Although the distribution has been altered, the original dispersion can be restored by shaking or stirring. Flocculation refers to aggregation of the droplets (without any change in primary droplet size) into larger units. It is the result of the van der Waals attraction that is universal with all disperse systems. Flocculation occurs when there is no sufficient

Emulsion

repulsion to keep the droplets apart to distances where the van der Waals attraction is weak. Flocculation may be “strong” or “weak,” depending on the magnitude of the attractive energy involved. One way to overcome the van der Waals attraction is by electrostatic stabilization using ionic surfactants, which results in the formation of electrical double layers that introduce a repulsive energy that overcomes the attractive energy. The second and most effective method of overcoming flocculation is by “steric stabilization” using nonionic surfactants or polymers. Ostwald ripening (disproportionation) results from the finite solubility of the liquid phases. Liquids that are referred to as being immiscible often have mutual solubilities that are not negligible. With emulsions, which are usually polydisperse, the smaller droplets will have larger solubility when compared with the larger ones (due to curvature effects). With time, the smaller droplets disappear and their molecules diffuse to the bulk and become deposited on the larger droplets. With time, the droplet size distribution shifts to larger values. Several methods may be applied to reduce Ostwald ripening: (i) Addition of a second dispersed phase component that is insoluble in the continuous medium. In this case, partitioning between different droplet sizes occurs, with the component having low solubility expected to be concentrated in the smaller droplets. During Ostwald ripening in a two-component system, equilibrium is established when the difference in chemical potential between different size droplets (which results from curvature effects) is balanced by the difference in chemical potential resulting from partitioning of the two components. This effect reduces further growth of droplets. (ii) Modification of the interfacial film at emulsion interface. By using surfactants that are strongly adsorbed at the emulsion interface (i.e., polymeric surfactants) and that do not desorb during ripening (by choosing a molecule that is insoluble in the continuous phase), the rate could be significantly reduced. In coalescence, the individual droplets fuse together. First, the smaller droplets are absorbed by the larger droplets, and then increasingly larger drops merge together

3

until two continuous phases are finally formed. The driving force for coalescence is the surface or film fluctuations which results in close approach of the droplets whereby the van der Waals forces is strong thus preventing their separation. Two droplets can only coalesce if the intervening layer of liquid is pierced when they approach each other. Therefore, coalescence is opposed in two ways by the emulsifier film surrounding the droplets. First, as in the case of aggregation, the like charges of the electrical double layer prevent them from approaching each other. Second, the buildup of an elastic surface film causes the emulsion droplets to bounce off each other when they collide. Coalescence is always followed by accelerated settling or creaming, which destroys the emulsion completely. The emulsion is then broken and cannot be reconstituted by shaking or stirring. The driving force for prevention of coalescence is to produce a stable film that can be achieved by two mechanisms and their combination: (i) increased repulsion both electrostatic and steric and (ii) dampening of the fluctuation. In general, smaller droplets are less susceptible to surface fluctuations and hence coalescence is reduced. This explains the high stability of nanoemulsions. The phase inversion refers to the process whereby there will be an exchange between the disperse phase and the medium. For example, an O/W emulsion may with time or change of conditions invert to a W/O emulsion. In many cases, phase inversion passes through a transition state whereby multiple emulsions are produced. Phase inversion of emulsions can be one of two types: transitional inversion induced by changing the facers that affect the HLB of the system, for example, temperature and/or electrolyte concentration, and catastrophic inversion, which is induced by increasing the volume fraction of the disperse phase. Emulsions have application in several industrial systems such as food emulsion, for example, mayonnaise, salad creams, deserts, and beverages; personal care and cosmetics, for example, hand creams, lotions, hair sprays, and sunscreens; and pharmaceuticals, paints, and bitumen emulsions.

4

References Israelachvili J (1994) The science and applications of emulsions - an overview, Colloids and Surfaces A: Physicochemical and Engineering Aspects 91: 1–8.

Emulsion Tadros T F (2013) Emulsion Formation, Stability, and Rheology, in Emulsion Formation and Stability (ed T. F. Tadros), Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, Germany.

B

Bimetallic Electrocatalyst for Fuel Cells Dmitri Bessarabov Faculty of Engineering, DST HySA Infrastructure Center of Competence, NorthWest University, Potchefstroom, South Africa

General Requirements for Fuel Cells Most of the known today fuel cell systems (e.g., automotive, portable applications, backup power, etc.) have common requirements that include reduction in costs, improvement in performance and durability, increase in tolerance toward impurities in the feed fuel, etc. (Vielstich et al. 2003a; Debe 2012).

Pt-based Electrocatalyst: Key Component The platinum-based electrocatalyst used in PEM (proton-exchange membrane, also called polymer electrolyte membrane) fuel cells, PAFC (phosphoric acid fuel cell), and DMFC (direct methanol fuel cell) is one of the key components and directly influences performance, durability, costs, etc. More specifically, in the PEM fuel cell, the oxygen reduction reaction (ORR) occurring at the cathode is known to have slow kinetics, leading to large cathodic overpotential losses under # Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_1083-3

typical operating conditions. In the case of specific application of PEM fuel cells, when hydrogen is produced from a reformate, an additional requirement such as tolerance of the anode electrocatalyst toward the traces of CO in the feed fuel stream must also be met. In order to reduce costs of the fuel cells, an increased mass activity of the cathode electrocatalyst is required that would allow a decrease in the amount of the electrocatalyst in a fuel cell device.

Advanced Electrocatalysts A search for better ORR electrocatalysts for PEM fuel cells that meet advance requirements has resulted in the development of various Pt-based bimetallic compounds (these can include alloys or intermetallic systems). It is believed that the improved performance (i.e., activity enhancement) of bimetallic alloys as electrocatalysts could be explained by the structural modification of Pt 5d orbital, coordination number of Pt, and modification in adsorption of oxygenated species from the electrolyte to the Pt or the alloying metal. For example, alloying results in a lattice contraction, leading to a more favorable Pt-Pt distance for the dissociative adsorption of oxygen (Vielstich et al. 2003b; Zhang 2008). Pt+Cr, Pt+Ni, and Pt+Co bimetallic electrocatalysts have been demonstrated to show two- to threefold increase in their activities for ORR. The activation energy for oxygen reduction

2

was also shown to be lower than that in Pt. However, it is also well recognized that these bimetallic electrocatalysts are subject to significant degradation rates (sintering/dissolution) as well as carbon-support corrosion under specific voltage cycling conditions that are common for automotive drive cycles during fuel cell startup and shutdown. A lot of efforts have also been made during the last two decades to improve anode electrocatalyst in fuel cells that are designed to use hydrogen obtained by means of reforming other fuels. In that case an anode electrocatalyst is required for hydrogen oxidation reaction to take place in the presence of CO, as well as to improve performance of the methanol fuel cells. The focus has been on the development of bimetallic platinumbased electrocatalysts to reduce the amount of adsorbed CO and/or to improve performance of electrooxidation of CO. Many metals have been considered for modifying the activity of the platinum catalyst, but only a few of them (Ru, Ir, Sn, Mo, etc.) lead to improved performances. The most studied bimetallic electrocatalyst is the family of Pt/Ru alloys, which enhance greatly the rate of oxidation of many alcohols (methanol, ethanol, etc.).

Current Trends Research and development in the area of bimetallic and ternary electrocatalysts is ongoing. The

Bimetallic Electrocatalyst for Fuel Cells

trend is to make use of unique nanoscale structural effects that can be observed in structurally modified alloys (e.g., de-alloyed Pt-based catalysts) and intermetallic systems to enhance greatly ORR. These include novel structures observed in thin film extended surfaces (Stamenkovic et al. 2007; Debe 2012) as well as a multilayer Pt-skin surface (Wang et al. 2011, 2012).

References Debe MK (2012) Electrocatalyst approaches and challenges for automotive fuel cells. Nature 486:43–51 Stamenkovic VR, Mun BS, Arenz M, Mayrhofer KJJ, Lucas CA, Wang G, Ross PN, Markovic NM (2007) Trends in electrocatalysis on extended and nanoscale Pt-bimetallic alloy surfaces. Nat Mater 6:241–247 Vielstich W, Lamm A, Gasteiger HA (eds) (2003a) Handbook of fuel cells, vol 1, Fundamentals and survey of systems. Wiley, New York Vielstich W, Lamm A, Gasteiger HA (eds) (2003b) Handbook of fuel cells, vol 2, Electrocatalysis. Wiley, New York Wang C, Chi M, Li D, Strmcnik D, van der Vliet D, Wang G, Komanicky V, Chang K-C, Paulikas AP, Tripkovic D, Pearson J, More KL, Markovic NM, Stamenkovic VR (2011) Design and synthesis of bimetallic electrocatalyst with multilayered Pt-skin surfaces. J Am Chem Soc 133:14396–14403 Wang C, Markovic N, Stamenkovic VR (2012) Advanced platinum alloy electrocatalysts for the oxygen reduction reaction. ACS: Catalysis 2:891–898 Zhang J (ed) (2008) PEM fuel cell electrocatalysts and catalyst layers. Springer, London

F

Fuel Cell Components Rajnish Kaur Calay Energy Technology Research Group, Narvik University College, Navrik, Norway

All fuel cells essentially consist of two electrodes – an anode (negative side) and a cathode (positive side) – and an electrolyte to allow charges to move between the two sides of the fuel cell. Electrons are drawn from the anode to the cathode through an external circuit, producing direct current electricity. On both sides of the electrolyte and two electrodes (cathode, anode) are the catalyst layers. A catalyst lowers the activation energy to undergo a reaction and helps the reaction to take place at a faster rate. The fuel cell type is defined based on the nature of the electrolyte. The six types are alkaline fuel cell (AFC), molten carbonate fuel cell (MCFC), phosphoric acid fuel cell (PAFC), proton exchange membrane or polymer electrolyte membrane (PEM) fuel cell, and solid oxide fuel cell (SOFC). The PEM and SOFC have solid electrolyte, while other three fuel cell types have liquid electrolyte. In a solid electrolyte fuel cell, for example, a PEMFC, the catalyst layer is spread either on the electrolyte (as in PEM or SOFC) or on the electrode. In addition to these basic components,

# Springer-Verlag Berlin Heidelberg 2015 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_1084-1

other components are gas diffusion layer, electric connections, current collectors, separator plates, and seals. In PEM fuel cell bipolar plates made up of metal or conductive polymer (or carbon composites) have more than one function. Bipolar plates allow the transfer of electrons from the anode side of one cell to the cathode side of the adjacent cell, provide structural rigidity of the stack, distribute reactant gas to each cell within a stack, and distribute reactant gas within the cell in the stack through proprietary flow field topology (shape designs). In some cases, heat management can be integrated in BPP design which typically involves a thermally conductive medium being supplied through the stack to remove heat from the stack. Thermally conductive mediums typically can be air, de-ionized water, or a coolant with high specific heat capacity. Liquid electrolyte fuel cells will have electrodes immersed in the liquid electrolyte and often benefit from simplified design and also less expensive catalyst materials. As a result, liquid fuel cells, such as PAFC and MCFC, are well-established technologies and have been widely used for medium-scale ( Pd > Ir > Rh, which is in agreement with experimental results. Calculations have also predicted that Pt–M (M = Fe, Co, Ni, etc.) bimetallic alloys should have higher catalytic activity than pure Pt, which has again been proven by experiments (Nørskov et al. 2004).

current density, jd is hydrodynamic diffusion limiting current density, and jf is film diffusion limiting current density. RDE experiments should be designed to maximize jf. In that case, 1j ¼ j1 þ j1 k d and after rearrangement jk ¼ j jd =ðjd jÞ. For ORR the value of jd is normally taken at 0.4 V during RDE experiments. Once jk is known, activity is often reported in practical values as mass activity (A/mg Pt) by normalization to the value of Pt loading of the sample on the disk electrode. Another practically important way of reporting the electrocatalyst activity is to use area-specific activity. The area-specific activity, expressed in the following units, uA/cm2 Pt, is reported by normalization of jk to the electrochemical surface area (ECSA) of the catalyst.

Practical Measurements

References

Practically, electrocatalytic activity of various electrocatalysts can be compared by measuring electrical currents produced at a given overpotential. The electrocatalyst that generates higher current at lower overpotential is generally a better electrocatalyst. The rotating disk electrode (RDE) technique is often used to measure ORR activity of electrocatalysts for hydrogen PEM fuel cell applications. Typically, current density j is measured at 0.9 V at the positive voltage sweep at a certain electrode rotating speed. The measured current density j is expressed by the following equation: 1j ¼ j1 þ j1 þ j1 , where jk is kinetic

Bligaard T, Nørskov JK, Dahl S, Matthiesen J, Christensen CH, Sehested S (2004) The Brønsted–evans–polanyi relation and the volcano curve in heterogeneous catalysis. J Catal 224:206–217 Nørskov JK, Rossmeisl J, Logadottir A, Lindqvist L, Kitchin JR, Bligaard T, Jonsson H (2004) Origin of the overpotential for oxygen reduction at a fuel-cell cathode. J Phys Chem B 108:17886–17892 Zhang J (ed) (2008) PEM fuel cell electrocatalysts and catalyst layers. Springer, London

k

d

f

Further Reading Vielstich W, Lamm A, Gasteiger HA (eds) (2003) Handbook of fuel cells, vol 2, Electrocatalysis. Wiley, New York

F

Fruit Juice Processing by Integrated Membrane Operations Alfredo Cassano Institute on Membrane Technology, ITM-CNR, Rende, CS, Italy

Fruit juice clarification, stabilization, depectinization, and concentration are typical steps where membrane processes such as microfiltration (MF), ultrafiltration (UF), reverse osmosis (RO), osmotic distillation (OD), and membrane distillation (MD) can be utilized as alternative technologies to the conventional transformation technologies. Particularly, MF and UF are valid approaches for the clarification of fruit juices as alternative to the use of fining agents, such as gelatin, diatomaceous earth, bentonite, and silica sol, which cause problems of environmental impact due to their disposal (Echavarria et al. 2011). Clarified juices coming from MF or UF processes can be commercialized or submitted to a

# Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_1088-1

concentration process in order to obtain a product suitable for the preparation of juices and beverages. RO, MD, and OD can be used as concentration techniques as alternative systems to thermal evaporation or cryoconcentration (Jiao et al. 2004). Integrated membrane operations have been suggested for replacement of conventional juice processing unit operations for the clarification and concentration of different fruit juices as well as for the recovery of aroma compounds. A typical example of integrated membrane system for the clarification and concentration of fruit juices is depicted in Fig. 1. The fresh juice is clarified by UF; the clarified juice is pre-concentrated by RO and then concentrated by OD. Pervaporation (PV) is used to recover aroma compounds from the fresh juice.

2

Fruit Juice Processing by Integrated Membrane Operations

pasteurization pulp

Fresh juice (10-11 °Brix)

permeate

UF

RO

preconcentrated juice (25-26 °Brix)

concentrated stripping solution

diluted stripping solution

OD

PV

aromatic compounds

concentrated juice (64-65 °Brix)

Fruit Juice Processing by Integrated Membrane Operations, Fig. 1 Integrated membrane process for the production of concentrated fruit juice

References Echavarria AP, Torras C, Pagan J, Ibarz A (2011) Fruit juice processing and membrane technology application. Food Eng Rev 3:136–158

Jiao B, Cassano A, Drioli E (2004) Recent advances on membrane processes for the concentration of fruit juices: a review. J Food Eng 63:303–324

R

Recovery of Polyphenols from Olive Mill Wastewaters by Membrane Operations Alfredo Cassano Institute on Membrane Technology, ITM-CNR, Rende (CS), Italy

Olives contain an appreciable amount of bioactive compounds such as polyphenols with remarkable health benefits. They are potent antioxidants and exhibit various other physiological activities including antiinflammatory, antimicrobial, antiallergic, anticarcinogenic, and antihypertensive activities (Obied et al. 2005). The biophenolic fraction of olive oil comprises only 1–2 % of the total phenolic content of the fruits; the remaining 53 % and 45 % are lost in olive mill wastewaters (OMWs) and the olive cake, respectively. Typical biophenols occurring in OMWs are benzoic acid derivatives (4-hydroxybenzoic, protocatechuic, vanillic acids), hydroxycinnamic acid derivatives (ferulic, caffeic acids), tyrosol, homovanillyl alcohol, hydroxytyrosol, and oleuropein. These compounds represent a precious resource of potentially useful chemical substances (after their direct recovery or chemical transformation) for cosmetic and pharmaceutical industries and in food processing and food product conservation. Several techniques have been proposed individually or in integrated forms to recover # Springer-Verlag Berlin Heidelberg 2014 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_1089-1

phenolic compounds from OMWs including solvent extraction, supercritical fluid extraction, and chromatographic separation. Membrane operations such as microfiltration (MF), ultrafiltration (UF), nanofiltration (NF), and reverse osmosis (RO), mostly in sequential form, represent useful technologies for the recovery, purification, and concentration of polyphenols with regard to their molecular weight cutoff values. They offer significant advantages over conventional methodologies in terms of low energy consumption, no additive requirements, and no phase change. Integrated membrane processes based on the use of these operations permit to obtain purified water which can be discharged in aquatic systems according to national regulations or to be used for irrigation. NF is typically employed for the separation of most part of phenolic compounds (Paraskeva et al. 2007).

References Obied HK, Allen MS, Bedgood DR, Prenzler PD, Robards K, Stockmann R (2005) Bioactivity and analysis of biophenols recovered from olive mill waste. J Agric Food Chem 53:823–837 Paraskeva CA, Papadakis VG, Kanellopoulou DG, Koutsoukos PG, Angelopoulos KC (2007) Membrane filtration of olive mill wastewater (OMW) and OMW fractions exploitation. Water Environ Res 79:421–429

R

Recovery of Polyphenols from Wine Wastewaters by Membrane Operations Alfredo Cassano Institute on Membrane Technology, ITM-CNR, Rende (CS), Italy

Grape seeds and pomace (a solid material consisting of skins and grape seeds) are typical by-products of the winemaking process containing phenolic compounds. Grape seeds contain basically (w/w) 40 % fiber, 16 % essential oil, 11 % protein, 7 % complex phenolic compounds, sugars, minerals, and other substances (Campos et al. 2008). Grape skin is a source of anthocyanidins and anthocyanins, natural pigments with antioxidant properties.

# Springer-Verlag Berlin Heidelberg 2014 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_1090-1

These phenolic compounds have an extremely high market value as food additives, nutraceuticals, and cosmeceuticals, due to their biological activity (Crespo and Brazinha 2010). A membrane-based process scheme for the purification, fractionation, and concentration of phenolic compounds from wine wastewaters is depicted in Fig. 1. The purification step devoted to the removal of undesirable compounds (such as fats, proteins, and sugars) and microorganisms is based on the use of ultrafiltration (UF) membranes. The use of appropriate nanofiltration (NF) membranes allows for obtaining fractions enriched in phenolic compounds. The final concentration step with removal of the extracting solvent (as permeate) is performed by reverse osmosis (RO).

2 Recovery of Polyphenols from Wine Wastewaters by Membrane Operations, Fig. 1 Purification, fractionation, and concentration of polyphenols from wine wastewaters by integrated membrane process

Recovery of Polyphenols from Wine Wastewaters by Membrane Operations microorganisms, fats, proteins, sugars

Raw material (pomace, seeds)

extraction

extracting solvent

UF

RO

NF

phenolic fractions

food additives, cosmeceuticals, nutraceuticals

References Campos LMAS, Leimann FV, Pedrosa RC, Ferreira SRS (2008) Free radical scavenging of grape pomace

extracts from Cabernet Sauvingnon (Vitis vinifera). Bioresour Technol 99:8413–8420 Crespo JG, Brazinha C (2010) Membrane processing: natural antioxidants from winemaking by-products. Filtr Sep 47:32–35

O

Olive Mill Wastewater Treatment by Membrane Operations Alfredo Cassano Institute on Membrane Technology, ITM-CNR, Rende (CS), Italy

The extraction of olive oil generates huge quantities of wastes having a great impact on land and water environments because of their high phytotoxicity. Pressure and three-phase centrifugation systems produce a liquid effluent called olive mill wastewater (OMW). Several waste management approaches, including physicochemical treatments (natural evaporation, treatment with lime and clay, oxidation), agronomic methods (land spreading), animal breeding, and biological treatments (both aerobic and anaerobic) have been proposed to reduce the polluting load and, consequently, the final waste disposal. The efficiency of the process, the complexity, and the costs involved may vary remarkably. In addition, different legislations existing in olive-oil producing countries play an important role in the selection of appropriated technologies. Pressure-driven membrane operations, including microfiltration (MF), ultrafiltration (UF), nanofiltration (NF), and reverse osmosis

# Springer-Verlag Berlin Heidelberg 2014 E. Droli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_1091-1

(RO) have been proposed and combined in integrated processes to obtain effluent streams from OMWs of acceptable quality for safe disposal into the environment (surface water or soil) for irrigation or even for recycling and use in the olive mill. Basically, MF and UF processes are used for primary treatment purpose, while NF and RO are used for final treatment (Cassano et al. 2013; Paraskeva et al. 2007). Integrated systems based on the use of these processes permit to obtain a COD reduction of about 99 %, the recovery of high percentage of purified water (60–70 %) (permeate of RO membranes), a production of an organic fraction (retentate of MF and UF membranes) which can be submitted to anaerobic digestion for the production of biogas, the recovery of a phenolic fraction (retentate of NF and RO membranes) of potential interest for food, phytotherapic, or cosmetic applications (Fig. 1).

References Cassano A, Conidi C, Giorno L, Drioli E (2013) Fractionation of olive mill wastewaters by membrane separation techniques. J Hazard Mater 248–249:185–193 Paraskeva CA, Papadakis VG, Tsarouchi E, Kanellopoulou DG, Koutsoukos PG (2007) Membrane processing for olive mill wastewater fractionation. Desalination 213:218–229

2

Olive Mill Wastewater Treatment by Membrane Operations Formulations for food, cosmetic and phytoterapic industry anaerobic digestion

purification

organic fractions

phenolic fractions

OMWs

pretreatment

MF

UF

NF

RO

purified water

Olive Mill Wastewater Treatment by Membrane Operations, Fig. 1 Integrated membrane process for the treatment of OMWs

O

Olive Wastewater Alfredo Cassano Institute on Membrane Technology, ITM-CNR, Rende, CS, Italy

Olive oil extraction processes are usually grouped into press extraction and centrifugation extraction systems (two-phase and three-phase centrifugal olive oil extraction). The extraction of olive oil generates huge quantities of wastes having a great impact on land and water environments because of their high phytotoxicity. Pressure and three-phase centrifugation systems produce a liquid effluent called olive mill wastewater (OMW) or vegetation water. Typically, 0.5–1.5 m3 of OMWs per 1,000 kg of olives are produced depending on the process used. OMWs appear like violet-dark brown liquids with acid reaction (pH from 3 to 6) containing great quantity of suspended matter,

# Springer-Verlag Berlin Heidelberg 2015 E. Drioli, L. Giorno (eds.), Encyclopedia of Membranes, DOI 10.1007/978-3-642-40872-4_1092-1

high degree of organic pollution (COD, 40–220 g/L; BOD, 35–110 g/L), high electrical conductivity, reducing sugars up to 60 % of the dry substance and polyphenols (0.5–24 g/L) (Takac and Karakaya 2009). The organic content is mainly represented by polyphenols, carbohydrates, polysaccharides, sugars, nitro compounds, polyalcohols, and oil. Ultrafiltration (UF) membranes allow to separate macromolecules and colloids having molecular weight between 500 and 250,000 Da; they can be used to remove most parts of COD from pretreated OMWs producing a permeate stream which can be submitted to a biological treatment to comply with law limits relevant to BOD and COD (Borsani and Ferrando 1996). A scheme of OMW treatment including a concentration step via UF is depicted in Fig. 1.

2 Olive Wastewater, Fig. 1 Scheme of OMWs treatment with UF membranes

Olive Wastewater OMWs Neutralization Sedimentation

Storage

Biological treatment

Settled waters (35°C)

UF

discharge in superficial water

References Borsani R, Ferrando B (1996) Ultrafiltration plant for olive vegetation waters by polymeric membrane batteries. Desalination 108:281–286

Takac S, Karakaya A (2009) Recovery of phenolic antioxidants from olive mill wastewater. Rec Pat Chem Eng 2:230–237

D

Dynamic Mechanical Analysis Chi Hoon Park Gyeongnam National University of Science and Technology (GNTECH), Jinju-si, Gyeongsangnam-do, Republic Of Korea (South Korea)

Dynamic mechanical analysis (DMA) is a technique to study viscoelastic properties and modulus of e