74,99 €
Mehr erfahren.
- Herausgeber: Wiley-VCH
- Kategorie: Fachliteratur
- Sprache: Englisch
This textbook gives a clear and coherent overview of ceramic membranes, from preparation methods all the way to applications and economics. The authors, who are known for their clear writing style, combine their expertise in environmental engineering and porous materials to cover a wide range of examples, with over 1000 references. Chapters 1, 2 and 3 give a detailed introduction to membrane synthesis, transport mechanisms, and characterisation. Building on this, Chapter 4 outlines the state-of-the-art in ceramic membrane applications, including fuel cells, water purification, gas separation, and the making of cheeses, fruit juice, wine and beer. The final chapter deals with the economics of ceramic membrane processes, using industrial case studies to examine market barriers and opportunities.
Ceramics are known throughout history, but now, after thousands of years, they're making a comeback. Indeed, they may hold the key for addressing three of today's biggest challenges: clean energy, drinking water and air pollution. This book is a must-have for anyone who wants to enter the ceramic membranes field, or keep up-to-date with the latest developments and applications.
This textbook gives a clear and coherent overview of ceramic membranes, from preparation methods all the way to applications and economics. The authors, who are known for their clear writing style, combine their expertise in environmental engineering and porous materials to cover a wide range of examples, with over 1000 references. Chapters 1, 2 and 3 give a detailed introduction to membrane synthesis, transport mechanisms, and characterisation. Building on this, Chapter 4 outlines the state-of-the-art in ceramic membrane applications, including fuel cells, water purification, gas separation, and the making of cheeses, fruit juice, wine and beer. The final chapter deals with the economics of ceramic membrane processes, using industrial case studies to examine market barriers and opportunities.
Ceramics are known throughout history, but now, after thousands of years, they're making a comeback. Indeed, they may hold the key for addressing three of today's biggest challenges: clean energy, drinking water and air pollution. This book is a must-have for anyone who wants to enter the ceramic membranes field, or keep up-to-date with the latest developments and applications.
Sie lesen das E-Book in den Legimi-Apps auf:
Seitenzahl: 826
Veröffentlichungsjahr: 2016
Ähnliche
Contents
Cover
Series Page
Title Page
Copyright
Dedication
Preface
Chapter 1: The Basics
1.1 General Introduction and Historical Perspective
1.2 The Basics of Membrane Separation
1.3 Membrane Separation Processes
1.4 The Morphology of Membranes
1.5 Membrane Modules
1.6 Fouling and Cleaning
1.7 Ceramic versus Polymer Membranes
1.8 Raw Materials for Ceramic Membranes
1.9 Preparation of Ceramic Membranes
1.10 Intermediate and Top Layers
1.11 Industrial Applications of Ceramic Membranes
1.12 Further Reading
Exercises
References
Chapter 2: Fundamentals of Membrane Separation
2.1 A Short Introduction to Mass Transfer Phenomena
2.2 Fick's Law
2.3 The Mass Diffusivity DAB
2.4 Integral and Differential Expressions of Mass Balance Equation
2.5 Convective Mass Transfer
2.6 Fluxes of Liquids through Porous Membranes
2.7 Fluxes of Gases through Porous Membranes
2.8 Fluxes through Non-porous Membranes
Exercises
References
Chapter 3: Characterization of Ceramic Membranes
3.1 Introduction
3.2 Pore Size and Pore Size Distribution
3.3 Visualization of Membrane Surfaces
3.4 Chemical Methods for Membrane Characterization
3.5 Physical Parameters of Ceramic Membranes
3.6 Conclusions
Exercises
References
Chapter 4: Applications
4.1 Classical Applications of Ceramic Membranes
4.2 Gas Separation with Ceramic Membranes
4.3 Ceramic Membrane Reactors
4.4 Liquid Separation and Purification
4.5 Cleaning of Wastewater with Ceramic Membranes
4.6 Ceramic Membranes in Food Applications
Exercises
References
Chapter 5: Economics
5.1 Introduction
5.2 A Layman Scientist's Guide to Project Appraisal: SWOT, PEST and LCA
5.3 Economic Considerations in the Manufacturing and Application of Ceramic Membranes
5.4 Discussion
5.5 Outlook
Exercises
References
Index
End User License Agreement
List of Tables
Table 1.1
Table 1.2
Table 1.3
Table 1.4
Table 1.5
Table 1.6
Table 1.7
Table 1.8
Table 1.9
Table 2.1
Table 2.2
Table 2.3
Table 2.4
Table 3.1
Table 3.2
Table 3.3
Table 3.4
Table 3.5
Table 3.6
Table 4.1
Table 4.2
Table 4.3
Table 4.4
Table 4.5
Table 4.6
Table 4.7
Table 4.8
Table 4.9
Table 4.10
Table 4.11
Table 4.12
Table 4.13
Table 5.1
Table 5.2
List of Illustrations
Figure 1.1
Figure 1.2
Figure 1.3
Figure 1.4
Figure 1.5
Figure 1.6
Figure 1.7
Figure 1.8
Figure 1.9
Figure 1.10
Figure 1.11
Figure 1.12
Figure 1.13
Figure 1.14
Figure 1.15
Figure 1.16
Figure 1.17
Figure 1.18
Figure 1.19
Figure 1.20
Figure 1.21
Figure 1.22
Figure 1.23
Figure 1.24
Figure 1.25
Figure 1.26
Figure 1.27
Figure 1.28
Figure 2.1
Figure 2.2
Figure 2.3
Figure 2.4
Figure 2.5
Figure 2.6
Figure 2.7
Figure 2.8
Figure 2.9
Figure 2.10
Figure 2.11
Figure 2.12
Figure 2.13
Figure 2.14
Figure 2.15
Figure 2.16
Figure 2.17
Figure 2.18
Figure 2.19
Figure 2.20
Figure 2.21
Figure 2.22
Figure 3.1
Figure 3.2
Figure 3.3
Figure 3.4
Figure 3.5
Figure 3.6
Figure 3.7
Figure 3.8
Figure 3.9
Figure 3.10
Figure 3.11
Figure 3.12
Figure 3.13
Figure 3.14
Figure 3.15
Figure 3.16
Figure 3.17
Figure 3.18
Figure 3.19
Figure 3.20
Figure 3.21
Figure 3.22
Figure 3.23
Figure 3.24
Figure 4.1
Figure 4.2
Figure 4.3
Figure 4.4
Figure 4.5
Figure 4.6
Figure 4.7
Figure 4.8
Figure 4.9
Figure 4.10
Figure 4.11
Figure 4.12
Figure 4.13
Figure 4.14
Figure 4.15
Figure 4.16
Figure 4.17
Figure 4.18
Figure 4.19
Figure 4.20
Figure 4.21
Figure 4.22
Figure 4.23
Figure 4.24
Figure 4.25
Figure 4.26
Figure 4.27
Figure 4.28
Figure 4.29
Figure 4.30
Figure 4.31
Figure 4.32
Figure 4.33
Figure 4.34
Figure 4.35
Figure 4.36
Figure 4.37
Figure 4.38
Figure 4.39
Figure 4.40
Figure 4.41
Figure 4.42
Figure 4.43
Figure 5.1
Figure 5.2
Figure 5.3
Figure 5.4
Figure 5.5
Figure 5.6
Figure 5.7
Figure 5.8
Figure 5.9
Figure 5.10
Figure 5.11
Figure 5.12
Guide
Cover
Table of Contents
Begin Reading
Chapter 1
Pages
i
ii
iii
iv
v
vi
xi
xii
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
Also published by one of the Authors:
Rothenberg, G.
Catalysis
Concepts and Green Applications
2008
Print ISBN: 978-3-527-31824-7
Vitaly Gitis and
Gadi Rothenberg
Ceramic Membranes
New Opportunities and Practical Applications
All books published by Wiley-VCH are carefully produced. Nevertheless, authors, editors, and publisher do not warrant the information contained in these books, including this book, to be free of errors. Readers are advised to keep in mind that statements, data, illustrations, procedural details or other items may inadvertently be inaccurate.
Library of Congress Card No.: applied for
British Library Cataloguing-in-Publication DataA catalogue record for this book is available from the British Library.
Bibliographic information published by the Deutsche NationalbibliothekThe Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available on the Internet at <http://dnb.d-nb.de>.
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA,Boschstr. 12, 69469 Weinheim, Germany
All rights reserved (including those of translation into other languages). No part of this book may be reproduced in any form – by photoprinting, microfilm, or any other means – nor transmitted or translated into a machine language without written permission from the publishers. Registered names, trademarks, etc. used in this book, even when not specifically marked as such, are not to be considered unprotected by law.
Print ISBN: 978-3-527-33493-3ePDF ISBN: 978-3-527-69657-4ePub ISBN: 978-3-527-69658-1Mobi ISBN: 978-3-527-69656-7oBook ISBN: 978-3-527-69655-0
Dedication
To our wives, Diana and Live, and our children,Abigail, Ariel, Avital, Daniel and Emil, who wondered what we did all this time.
Preface
This textbook tells the story of ceramic membranes. Ceramics are making a comeback as hi-tech materials, and the membranes are used in many processes, from uranium purification to orange juice finishing. Moreover, they can help us address urgent environmental issues air pollution and shortage of clean drinking water. It is definitely the time to study them.
The book is divided into five thematic chapters. Chapter 1 gives a general introduction to the membrane field and then covers the making of ceramic membranes. The physics of flow models and mass transfer through membranes are discussed in Chapter 2. Chapter 3 deals with membrane characterization, from the most common methods to the state-of-the-art. We then move to applications in Chapter 4, containing a range of industrial sectors, including chemicals, food and beverages and fuel cells. Finally, Chapter 5 looks at ceramic membranes from a different viewpoint: We examine the economics of membrane manufacturing and predict, on the basis of case studies, the future market for ceramic membrane technologies.
We wrote this book for both researchers and students. Thus, each chapter includes detailed references, as well as exercises. To help you master the membrane jargon, key terms are printed in italics and defined the first time they appear in the text. The book contains over 1000 references, whereof >90% are original papers in peer-reviewed journals. We cite reviews when introducing a subject, and articles when discussing specific topics (plus some patents in the industrial examples). Chapter 1 also contains a “Further Reading” list of books on related specialized subjects.
We thank our families for their patience and support during the past 3 years, and our colleagues Vittorio Boffa, Hessel Castricum, David Dubbeldam, Andre ten Elshof, Giovanni Catania, David Farrusseng, Slava Freger, Freek Kapteijn, Chiho Kojima, Erik Kossin, Daniel Sereth Larsen, Marc Pera-Titus, Ehsan Salehi, Jesus Santamaría, Lev Tsapovski, Ning Yan and Gennady Ziskind for their help and advice. V.G. thanks Rabbi Daniel Lewenstein for his support and encouragement. We also thank Itamar Daube for drawing the cover picture and Wiley-VCH editors Waltraud Wüst and Gudrun Walter for their encouraging and professional attitude.
January 2016
Vitaly Gitis
Beer-Sheva, Israel
Gadi Rothenberg
Amsterdam, The Netherlands
1The Basics
1.1 General Introduction and Historical Perspective
This chapter covers the basic definitions, main features and engineering and design of ceramic membranes. We give a brief history of the development of ceramic membranes, define key terms in membrane science, outline the popular separation processes (ultrafiltration (UF), nanofiltration (NF), pervaporation and gas separation) and explain the main module designs (plate-and-frame, spiral-wound, tubular, honeycomb and hollow fibres). The historical overview shows how membranes started, when the big breakthrough occurred, where membranes are now and how the near future will look like. The actual making of ceramic membranes is in itself an interesting story, and a good part of the chapter is devoted to the synthesis of various layers of the membrane. We give an overview of the main methods and materials used for preparing such membranes and characterizing them, as well as their key advantages and limitations. The discussion covers both isotropic and anisotropic membranes, prepared from a range of materials (zirconia, titania, alumina, hafnia, tin oxide, mixed oxides, zeolite membranes, silica, hybrid organic–ceramic membranes and metallo-organic frameworks). We analyse in detail the formation of support layer and list some rules of thumb collected by many researchers in numerous trials. A key aspect here is the gradual transition from the support layer through the intermediate layers and ultimately to the top layer. The development of top layer is reviewed through the basics of chemical vapour deposition (CVD), sol–gel technology and zeolite modifications. The chapter concludes with a list of books for further reading, qualitative and quantitative exercises and references.
A membrane is a semipermeable active or passive barrier that permits the passage of one or more components in the initial mix and limits the passage of others. Although Graham in 1848 used a sort of membrane in the development of diffusion law, and although the first membranes were synthesized more than a century ago, the development and implementation of membranes really turned into a scientific discipline in the second half of the twentieth century. Today's membranes, with their modest energy demands and small footprint, have become even more attractive and are often compared favourably with conventional separation processes such as distillation, adsorption, absorption, extraction and crystallization. There are many books on the development, characterization and implementation of polymer membranes. Ceramic membranes are much less in the focus, and this book will hopefully rectify this a little, by shedding light on this important subfield of membrane science.
By the layman's definition, ceramics are materials made of pottery (κέραμoζ in Greek) that is then hardened by heat. A more scientific definition (from the Ceramic Tile Institute of America) describes ceramic material as an inorganic, non-metallic solid prepared by the action of heat and subsequent cooling [1]. This definition explores an older Sanskrit meaning of the Greek keramos – to be burned (unlike glass that is amorphous, ceramics are crystalline materials). Ceramics are compounds of metallic and non-metallic elements such as aluminium and oxygen (Al2O3), zirconium and oxygen (ZrO2) or silicon and carbon (SiC). These compounds occur naturally in clays and other minerals and are processed in supported forms. With such available ingredients, simple recipes and long-term robustness, no wonder that archaeologists have found man-made ceramics that date back to at least 24,000 BC [2]. The durability of ceramic artefacts has given them prominence in archaeology [3]. Ceramics were one of the remarkable keystones that marked the transition from Stone to Bronze Age when humans first started using man-made tools instead of sharpened stones. In this sense, ceramics are the oldest of three large classes of solid materials (ceramics, metals and polymers) on the main development route of industrial products. The first ceramics, found in former Czechoslovakia, were made of animal fat and bone mixed with bone ash and clays [4]. The initial mix was hardened at kilns dug in the ground at temperatures between 500 and 800 °C. We do not know how these ceramics were then used. The first use of ceramics as containers for holding and storing grains and other food dates back to 9000 BC. Heating the sand that contained calcium oxide combined with soda resulted in a coloured glaze on ceramic containers in Upper Egypt about 8000 BC [5]. One of the earliest civilizations, the Sumerians who lived in Southern Mesopotamia (modern Iraq) more than 5000 years ago, wrote on ceramic stone plaques. The ceramic amphora, which was invented in Greece, became a standard for the transport and storage of liquids (mostly wine and olive oil) in the Roman Empire. The need to purify the water transported in air-open aqueducts [6] expanded the use of ceramics in the Empire. Figure 1.1 shows one of the first ceramic filters, which dates back to Israel Iron Age II – 800 BC (an artefact from the Israeli National Museum).
Figure 1.1 One of the first ceramic filters dated back to the second Iron Age, circa 800 BC. A clay vessel that is probably used for serving beer. (https://www.pinterest.com/pearsonaf/pottery-of-the-past/.)
So ceramics have been with us for thousands of years, but ceramic technology has really developed only in the last century. Today's ceramics are no longer just dinnerware, bricks and toilets. Technical ceramics are used in space shuttles, engines, artificial bones and teeth, computers and other electronic devices and of course membranes. The first modern industrial application of ceramic membranes was in the separation of U-238 and U-235 isotopes for making nuclear weapons and fuels in the 1940s and 1950s [7]. This separation was performed at high temperatures by forcing highly corrosive UF6 through semipermeable membranes. The only membrane materials that could withstand such harsh environments were oxides such as Al2O3, TiO2 and ZrO2. Many aspects of that work, carried out by the Western Bloc during the Second World War (the so-called Manhattan Project), are still classified [8]. The only information on these comes from several patents filed in the 1970s. Trials using the same membranes in purification of liquids met with limited success, mainly due to low separation efficiency and low flux. The idea of dividing a membrane into a skin and a porous substructure, proposed by Loeb and Sourirajan [9] in 1962 for polymer membranes, boosted the development of a new generation of ceramic membranes. It appeared that ceramic membranes could also be made in a number of layers like onions. In this new anisotropic membrane, the skin layer determines the separation and the support layer gives the mechanical strength and uninterrupted flux. Technical questions on fusion of layers made from different materials were significantly facilitated by Burggraaf and Cot [10] who developed in the 1980s a concept and procedures for intermediate membrane layers. This opened the door to applications in food and beverage industries [11,12], gas separation [13,14] and biotechnology [15], albeit in small installations.
In the past two decades, ceramic membranes have become a valuable component of fuel cells and play a central part in the hydrogen economy. Full-scale installations for water and wastewater purification started in Japan in 1998, and have recently started spreading to Europe and the United States. The separation of uranium isotopes, that started more than half a century ago, reached its maximum in the 1970s when nuclear energy was considered a valuable replacement of fossil fuels. However, after the Chernobyl disaster in 1986, reassessment of true amount of fossil fuels available, and development of more cost-effective uranium enrichment techniques such as centrifuge and laser, the use of ceramic membranes for uranium enrichment halted. Companies such as Atech Innovations, Orelis, Veolia Water, Hyflux, Kubota, TAMI Industries, Inoceramic GmbH, Metawater, Mitsui, Meidensha, Jiangsu Jiuwu, Pervatech and Ceraver [16,17] [acquired by Alcoa in 1986, then Societe des Ceramiques Techniques as USFilter in 1992, and (since April 2002) Pall Corporation] now advance ceramic membranes in new fields such as the water and wastewater treatment, food and beverages, chemical, pharmaceutical, electronic, petrochemical and energy sectors. Figure 1.2 sketches a brief of ceramic membrane history and their entry into various industrial sectors.
Figure 1.2 A timeline of ceramic membrane applications.
Today, ceramic membranes are established in modern separation techniques. As we will show in this book, in the future ceramic membranes with their clear advantages in chemical and thermal stability, longer lifetime, higher flux and higher recoveries will be employed in more applications. This is supported by recent reports on large-scale piloting with ceramic membranes and several full-scale installations. Here we will introduce the main developments in ceramic membranes, starting with a brief introduction into the general field of membranes that will help us to discuss technical details of membrane preparation and operation.
1.2 The Basics of Membrane Separation
Here we give a very brief introduction to the general membrane field emphasizing the difference between ceramic and other membranes, where appropriate. For readers wishing to delve deeper into the principles of membrane separations, there are several books that give a good introduction to the subject [16–18]. Other books with a special emphasis on the ceramic membranes are also briefly discussed in the ‘Further Reading’ section at the end of this chapter (Section 1.10).
Ceramic membranes, as any other membranes, are used for separating suspensions, aerosols and mixtures. They leave particles, organic molecules, dissolved salts or even gases and liquids on one side and transfer purified gases and liquids to the other. Thus, the ceramic membrane is a semipermeable barrier that separates purified and concentrated streams out of a mixture. Figure 1.3 depicts the essentials of membrane separation where the initial feed is separated into permeate and retentate streams. If the separation is performed for purification purposes, the permeate stream is the final product and the retentate stream is the by-product. If the separation is performed to concentrate a component in the mixture, the retentate stream is the product and the permeate is the by-product.
Figure 1.3 The basic membrane separation set-up showing the feed tank, the permselective membrane and the feed, permeate and retentate flows.
Mathematically, we express the above definition as the feed flow that approaches the membrane splits into permeate and retentate flows:
where Qf, Qr and Qp are the feed, the retentate and the permeate flows.
The efficiency of separation is evaluated using two parameters: the quantity of purified gas–liquid on the permeate side and the degree of purification. With different membrane areas and measurement periods, the quantity is unified by the transmembrane flux defined as the volume of gas–liquid passing through a unit of membrane area per a period of time:
where J is the volume flux, Qp is the flow of the fluid that passes through a membrane, and Am is the membrane surface area. Fluxes in liquid–liquid separations are typically reported in litres per square metre of the membrane surface per hour (l/(m2 h)) or gallons per square foot per day (gallons/(ft2 day)). Fluxes in gas and vapour separation are reported in cubic centimetres of gas per second per square centimetre of membrane area (cm3/(cm2 s)). Transmembrane pressure and temperature significantly change gas fluxes and reported gas flux values assume standard conditions of 0 °C and 1 atm. The volume flux J can be converted into mass flux or molar flux by using the density and the molecular weight of the feed, respectively. Some processes, for example DNA or protein purification, require high separation efficiency [19]. In others, the membranes must provide a predetermined flux (e.g. in dialysis or controlled drug release). Membranes used in controlled drug delivery need to provide a certain flux of a drug from a reservoir to the body. The ratio of permeate to the feed flux/flow is called the recovery ratio, RR, and defined as
A fluid passes through a membrane by the shortest path. Intuitively, the fluid should be forced through a membrane perpendicularly to its surface following this general concept of dead-end filtration. The concept of filtration perpendicular to the filter surface was developed for granular filters in France in the middle of the eighteenth century. An additional tangential or a cross-flow filtration (CFF), also known as tangential flow filtration (TFF), was developed in the middle of the twentieth century. In this mode, a fluid flows in parallel to the membrane surface. A pressure difference across the membrane drives the fluid through the membrane. Figure 1.4 shows the dead-end and cross-flow filtration modes.
Figure 1.4 Schematics of dead-end (a) and cross-flow (b) filtration modes.
The transmembrane pressure (TMP) in dead-end filtration is calculated as
The TMP in cross-flow filtration is an average between the pressure on permeate and retentate sides:
Here Pf, Pr and Pp are the feed, retentate and permeate pressures, respectively. In a single-stage installation, the permeate pipe is open to the air and, therefore, Ppequals 1 bar. The TMP thus is the additional pressure (above the atmospheric pressure) needed to pass a fluid through a membrane. Dead-end filtration is more suitable for treating dilute suspensions. Conversely, cross-flow filtration is used for more concentrated suspensions when the deposits are swept away from the membrane surface by the shear stress that is exerted by the flow.
A fluid passes through the membrane overcoming its resistance. This resistance has two components: the intrinsic resistance of the membrane itself, and the resistance of materials accumulated within the membrane during the filtration operation:
where Rm, Ro and Rt are the membrane resistance, the operational resistance and the total resistance, respectively. The increase in the total resistance is the result of changes in the operational resistance when the membrane resistance remains constant.
A flux J (Eq. (1.2)) through the membrane system is related to the TMP (Eq. (1.4)) through the total resistance Rt (see Eq. (1.6)) as
where μ is the viscosity of the fluid at a given temperature. A ratio of flux to TMP for a given membrane depends on the total membrane resistance and the viscosity but not on the operational parameters. The membrane permeability M ((Eq. 1.8)) is independent of the applied pressure and permits the comparison of the performances of different membranes operating under various conditions:
A permeability coefficient through a gas separation membrane takes into account the membrane thickness:
Here P* is the permeance through a membrane of thickness Δl and ΔP is the partial pressure difference of a gas across the membrane. A transmembrane gas flux J depends on the membrane thickness but does not depend on fluid viscosity as in (Eq. 1.7).
The degree of purification or the membrane selectivity is often evaluated using its retention ratio R or the separation factor α. In dilute solutions, it is more convenient to report the selectivity by R assuming that the solute is partially retained by the membrane when the solvent passes freely through the membrane:
where Cp and Cf are the concentrations of the separable compound in the permeate and in the feed, respectively. The retention ratio is dimensionless and does not depend on the units in which concentration is expressed. Its value ranges from 0 for a free penetration up to 1 for complete retention. The membrane selectivity in separation of gases and organic liquids, αAB, is expressed by the ratio of pure gas permeabilities for the individual components A and B:
Permeation through membranes occurs in two stages. A sorption of gas molecules onto and into membrane surface is followed by the diffusion of a gas through the membrane. Thus, the permeability P can be expressed as a multiplication of a thermodynamic component K related to sorption and kinetic component related to diffusion D:
In this equation, the diffusion coefficients DA and DB reflect mobilities of individual molecules in the membrane material, while the gas sorption coefficients KA and KB (cm3 of component in cm3 of membrane) express the number of molecules A and B adsorbed or dissolved in the membrane material. The KA/KB ratio can be viewed as the sorption or solubility selectivity of gases A and B [18]. As defined in (Eqs. 1.11) and (1.12), αAB is the ideal coefficient that does not account for mutual interactions of gases as they pass through a membrane. In binary mixtures with significant concentrations of A and B, the coefficient is calculated as the molar retention ratio, αAB:
where [yA, yB] are concentrations of gases A and B in the permeate, and [xA, xB] are their concentrations in the feed.
The selectivity is always >= 1. If the concentration of A in the permeate is higher than the concentration of B, the separation is denoted as αAB. If the concentration of B in the permeate is higher than that of A, the separation is denoted as αBA. If αAB = αBA, no separation is achieved [16].
1.3 Membrane Separation Processes
Membrane separation is a field that embraces many processes. These are subdivided by the origin of the applied driving force, phases of feed and permeate and pore size. The division by the driving force describes the origin of the force needed to transfer the fluid from the feed to the permeate side. Pressure, temperature, concentration or electrical potential are the driving forces available. Table 1.1 lists some common membrane processes, separation phases, driving forces, sizes of retained compounds and their types.
Table 1.1 Common membrane separation processes.
Membrane process
Feedphase – permeate phase
Driving force
Size of retained compounds
Type of retained compounds
Microfiltration (MF)
L → L
Δ
P
0.1–100 µm
Bacteria, fine solids
Ultrafiltration (UF)
L → L
Δ
P
5 nm to 100 µm
Viruses, total suspended solids, natural organic matter
Nanofiltration (NF)
L → L
Δ
P
1 nm to 100 µm
Inorganics, sugars, dyes, surfactants
Reverse osmosis (RO)
L → L
Δ
P
0.1 nm to 100 µm
Salts, metal ions, minerals
Gas separation
G → G
Δ
P
0.5 nm to 100 µm
Gases
Vapour permeation
G → G
Δ
P
0.5 nm to 100 µm
Liquids
Pervaporation
L → G
Δ
P
0.5 nm to 100 µm
Liquids
Electrodialysis
L → L
ΔΦ
Ions
Dialysis
L → L
Δ
C
Membrane distillation
L → L
Δ
T
Liquids
Note:
G and L stand for gas and liquid phases, respectively, Δ
P
is the pressure difference, ΔΦ is the electrical potential difference, Δ
C
is the concentration difference and Δ
T
is the temperature difference.
Except for membrane distillation (see Section 4.4.5) that uses a temperature difference as the driving force, ceramic membrane separations use pressure. The level of applied pressure varies as a function of size of solutes separated by a membrane. For the same flux J, small pores require high pressure and offer a retention of small solutes. A relation between pore size, solute size, flux and applied pressure resulted in an additional subclassification of pressure-driven membrane processes mostly applicable in particle–liquid separation. Particles in liquids can be quite big and reach the maximum size of 100 µm. Membranes are generally not applied for the retention of particles larger than 100 µm. These separations are typically done using either sedimentation or filtration. Membranes separate particles using pressure as the driving force and micro-, ultra-, nano- and subnanopores incorporated in microfiltration (MF), ultrafiltration, nanofiltration and reverse osmosis (RO) membranes. Table 1.1 details proper implementation of pressure-driven membranes in particle–liquid separation. Reverse osmosis and nanofiltration are processes used for separating solute and solvent components on the nanoscale. Water desalination is the most famous example of RO technology. The separation is so sensitive that while water molecules with a radius of 1.3 Å diffuse through the membrane, electrolytes and organic solutes with several hydrophilic groups cannot pass. Nanofiltration is similar to RO, and uses the same principles. The pores of NF membranes are slightly larger than in RO membranes and they can separate multivalent ions.
Two main advantages of NF over RO membranes are the lower operational costs due to a lower required TMP and a wider choice of membrane materials. Both processes are commercially performed with polymer membranes, where the dense polymer layer needed for separation of monovalent ions can be synthesized from cellulose acetate (CA) or polyamide (PA). Similarly, NF membranes are made from cellulose acetate blends or polyamide composites but can also be synthesized from more stable polymers such as polysulfone or polypiperazine. Ceramic NF membranes are prepared from alumina, titania, hafnia, silica–zirconia and zeolites, although higher cost and lower mechanical strength are currently limiting their wide commercialization.
Ultrafiltration and microfiltration are another popular subclass of membrane separation processes. Although RO, NF, UF and MF processes are all pressure-driven, a significant difference in the pore size determines different applications and features of these membranes. The UF/MF membranes are used in the food and beverage industries, in water and wastewater purification, in pharmacology and in medicine. A typical size of separable colloidal particles and high-molecular-weight solutes ranges from single nanometre to micrometres, so UF/MF membranes cannot be used in desalination. On the positive side, the bigger pore size means that a much lower TMP is needed to achieve a reasonable flux. Polymeric MF/UF membranes are synthesized from polyacrylonitrile (PAN), polysulfone (PS), polyethersulfone (PES) and polyvinylidene fluoride (PVDF). These polymers are more mechanically, thermally and chemically stable than cellulose acetate and polyamide and the resulting UF/MF membranes are employed under harsh conditions. There is no clear distinction between UF and MF (basically, the same particles can be retained by both membranes with the same applied TMP and with higher separation efficiency of UF membranes). The recent trend is therefore to use more UF membranes with a smaller pore size and a wider range of separated materials. Ceramic UF membranes synthesized from alumina, titania and zirconia are used in separations performed at high temperatures or with non-aqueous solvents such as benzene or toluene.
Gas separation, pervaporation and membrane distillation membranes deal with small molecules. The separation of gas molecules on the basis of their size requires small pores that can be described as low NF pores. The subdivision into MF, UF or NF membranes is therefore not applicable here and it is not implemented. Unsurprisingly, gas separation is performed with gas separation membranes. The basic role in the implementation of a certain type of membranes is rather simple – a preset degree of purification is to be achieved at a lowest possible cost. Larger membrane pores produce less resistance to the transmembrane flow and demand less pump energy. Thus, the membrane operational costs increase in the order MF → UF → NF → RO but so does the retention. This subdivision is somewhat arbitrary and the same membrane may be described as UF, MF or NF (although RO membranes are rarely mixed with others). We will get back to definitions and methods of detection of membrane pores in Chapter 3.
1.4 The Morphology of Membranes
From a morphological point of view, membranes are divided into two large groups. Porous membranes transport the solutes in a continuous fluid phase through the voids within the membrane structure. Dense membranes transport solutes by dissolution and diffusion across the membrane. Most ceramic membranes are porous. The examples of non-porous ones include Pd membranes for hydrogen separation and mixed (electronic, ionic) conducting oxides for oxygen separation [20]. Both porous and dense membranes can be prepared from polymers, ceramics, paper, glass and metals. The polymer membranes are also called ‘organic’ ones, while ceramic, glass and metal membranes are called ‘inorganic’. Polymer membranes are synthesized from different polymers, including cellulose, polyacrylonitrile, polyamide, polysulfone, polyethersulfone, polycarbonate, polyethylene, polypropylene and polyvinylidene fluoride. In addition, many polymers are grafted, custom-tailored, blended or used in a form of copolymers [21,22]. These modifications are made to increase the flux and retention of certain compounds or to avoid the flux drop due to accumulation of compounds on the membrane surface. Metal membranes are manufactured from palladium, nickel, silver, zirconium and their alloys, while ceramic membranes are made from metal oxides (alumina, titanium, zirconia), silica, zeolites and other mixed oxides.
There are various membrane preparation techniques, each with its own pros and cons. Typically, polymer membranes are prepared by phase inversion [9], track etching [23] and stretching [24]. Inorganic membranes are prepared by calcination and sintering and coated using sol–gel processes, chemical vapour deposition or hydrothermal methods. A detailed discussion on preparation techniques of ceramic membranes is given in Section 1.9.
The morphology of ceramic membranes is closely related to the membranes' pores. Pore size and size distribution, structure and tortuosity, interconnectivity and density (i.e. the number of pores per unit area) are the physical parameters that affect flux and separation efficiency. Membrane pore sizes are subdivided into macropores (diameter >50 nm), mesopores (between 50 and 2 nm) and micropores (<2 nm). The International Union of Pure and Applied Chemistry (IUPAC) also distinguishes between supermicropores (<2 nm) and ultramicropores (<0.7 nm) [25]. Pore size distribution indicates the presence of pores of different sizes within the membrane. Pore density is described by the porosity – the membrane surface or the membrane volume occupied by pores versus the total membrane surface or volume, respectively. Detailed definition of pore densities and their definitions are given in Chapter 3. Less porous structures are stronger, but also more resistant to flow, so the optimal porosity is a trade-off between the stability and the flux. There is no one-to-one relation between the porosity and the separation efficiency. After the initial packing of particles, the colloidal or polymer soils are heated at high temperatures. The sintering that occurs during the heating results in changes in porosity and pore size. Yet, the higher packing density of smaller initial grains, that is a low initial porosity, embeds more uniform distribution of grains during the sintering and will result in denser membranes with smaller pores [26].
Membrane structures are divided according to the type of their pores. Membranes with finger-like pores are called isotropic (having symmetrical pores going from one to another membrane side with the same width). Membranes with sponge-like pores are called anisotropic (having asymmetrical pores, see Figure 1.5).
Figure 1.5 Symmetric and asymmetric membrane pores.
The synthesis of membranes with symmetric pores is relatively simple, and symmetric nitrocellulose MF and UF membranes were successfully prepared in Germany a century ago [27]. These membranes were later commercialized by Sartorius and used by the German army during the Second World War for bacteriological water quality tests in cities where the water supply system was destroyed. A symmetric membrane has a rigid void structure with randomly distributed interconnected pores. Such a membrane acts as a molecular sieve, retaining solutes that are larger than its pore size and transferring those having similar or smaller dimensions. The pore size itself, however, can vary significantly from 100 µm all the way down to 3–5 Å. It can be so small that these membranes are sometimes described as non-porous [17]. The transition through such pores is by diffusion, driven by either concentration or electrical potential gradient. Porous membranes can be electrically charged when the pore walls carry either a positive charge (anion exchange membranes) or a negative one (cation exchange membranes). The main problem of symmetric membranes is their inherent high resistance to the flow due to the pore width uniformity. Moreover, the selectivity of symmetric membranes is determined already at the skin membrane part and does not change through the pore.
In 1962, Loeb and Sourirajan solved the problem of unnecessary resistance to the flow, inventing asymmetric polymeric membranes [9]. Here, the separation is determined by the upper membrane layer, and the mechanical strength and support are provided by the lower layers. The upper separation layer has the smallest pores within the membrane structure, while the supporting system has larger pores with lower hydraulic resistance to the permeate flow. The surface and support layers can be prepared simultaneously or sequentially. The membrane can be homogeneous, that is made from one material, or a composite made from different materials. In the latter case, the pore size and structures are conveniently determined by each constituent. Most ceramic membranes are asymmetric composites made from four or even five different layers. The composite structure of ceramic membranes is depicted in Figure 1.6, together with a cross-sectional scanning electron micrograph of a γ-alumina thin top layer of small pore sizes on top of an α-alumina support layer with gradually increasing pores towards the permeate side.
Figure 1.6 (a) Pictorial representation of an asymmetric composite ceramic membrane that consists of a nanofiltration-modified separation layer of 50 nm depth with pores less than 2 nm wide (A), an ultrafiltration layer of 100–500 nm depth with 10 nm pores (B), a 1–10 µm microfiltration intermediate layer with pores 100–200 nm wide (C) and a porous support of 1–1.5 mm width (D). (b) Scanning electron micrograph of a cross section of a ceramic composite membrane: γ-alumina on top of an α-alumina support [28].
The membrane support layer D, often called simply ‘the support’, has to provide the maximum mechanical strength at the minimum membrane resistance. The support is therefore often over 1 mm thick and macroporous. Such thick supports are very stable but are also resistant to transmembrane flow. An intrinsic deficiency of membrane supports is their high average pore size, high surface roughness and high void defect density. The ideal membrane support layer should be strong, homogeneous, stable and possess minimum flow resistance but not the separation ability [29]. It must also be chemically compatible with the intermediate and filtration layers, and mechanically and thermally stable.
The intermediate layers B and C must provide good chemical and thermal stability, and must have a narrow pore size distribution and a smooth homogeneous surface. While the former is a general requirement for the entire ceramic membrane, the smooth surface relates to the main function of intermediate layers. It is almost impossible to coat the separation layer A on top of the support layer D with macroporous voids. Therefore, the intermediate layers B and C are used to gradually decrease the pore size of the support, thus preventing the penetration of the very fine particles used for the formation of top layer. Typical intermediate layers are thick enough to increase the mechanical strength of a ceramic membrane. The thickness of a single intermediate layer is usually a few hundred micrometres. Pore widths are in a mesoporous range of 2–50 nm in diameter. The number of intermediate layers varies, depending upon the difference in grain sizes between layers A and D, and the intended membrane use. Membranes for water and wastewater treatment might possess a support and maybe one intermediate layer. Gas separation membranes will contain four to five layers with macropores in the support and ultramicropores in the top layer. Generally, the bigger the difference in the pore widths between the support and the top layer, the higher the number of layers. An insufficient number of intermediate layers will result in penetration of small particles into the next-layer pores, leading to an increase of flow resistance and low mechanical stability of the sintered membranes [20].
The actual separation is performed in the top layer A. This layer is typically coated last on top of an existing membrane from different materials. It is responsible for the separation and therefore contains the smallest pores in the membrane structure. Note that this layer is not responsible for the mechanical strength of the membrane and thus it is relatively thin. A typical thickness of the separation layer is between 10 and 20 µm, and the intermediate and support layer are of 1–2 mm in total. Similar to layers B and C, the top layer A should be chemically and thermally stable and must possess a narrow pore size distribution and a smooth homogeneous surface. Importantly, the top layer may not have any large pores – even a few macropores will render the membrane useless as the entire flux will be directed through those pores. In fact, a good top layer should also compensate for any structural defects of the intermediate layers.
Together, this multilayer configuration provides the membrane its separation and flux properties. Every membrane layer is functional and purposeful, yet the total number of membrane layers varies depending upon the separation processes. For a precise separation such as gas separation or water desalination, the membrane will contain all A + B + C + D layers. Conversely, the concentration of proteins in food and dairy industries or sterilization of beverages does not require the microporous separation and can be performed with layers C and D only. Table 1.2 shows the link between a number of layers in a ceramic membrane and its designated separation process [30].
Table 1.2 The layer structure of composite ceramic membranes.
Separation process
Number of layers
Average pore size
Microfiltration (MF)
1
5 µm
2
0.25 µm
Ultrafiltration (UF)
3
100 nm
Nanofiltration (NF)
4
2 nm
Reverse osmosis (RO), gas separation, pervaporation, vapour permeation
5
10 Å
Source:
After Bonekamp [30].
Layer A is sintered; layers B, C and D are produced by sequential calcination at high temperatures making the manufacturing of ceramic membranes a complicated procedure. Layer A is often produced by CVD or by sol–gel techniques [31] and requires a high variability to suit the separation needs.
1.5 Membrane Modules
Laboratory-scale ceramic membranes are typically produced in a plain form, often as small discs suitable for cylindrical benchscale filtration units. This geometry reduces production expenses and makes further examination of the membrane surface easier. Small solid ceramic discs are also fun to play with. Typical available membrane surface ranges from a few to a few dozen square centimetres. Practical, real-life industrial membrane applications require hundreds or even thousands of square metres of membrane surface, raising the question of packing density. The membrane packing density is defined as the total membrane surface available per a module volume. Each module needs space and the packing density is an expression of membrane footprint. Commercial membranes are produced in plate or tubular form. The membranes are packaged in modules that represent the ‘smallest discrete separation unit in a membrane system’ [32]. Each module contains at least several square metres of the membrane surface potted or sealed into an assembly. Modules are assembled in larger production units, also called skids, racks or trains. A production unit shares feed, retentate and filtrate valves, and allows the isolation of single modules. Plate membranes are assembled in plate-and-frame and spiral-wound modules, and tubular membranes are assembled in tubular and hollow fibre modules. Plate-and-frame and spiral-wound modules are often employed in cross-flow mode (see Figure 1.4). Tubular and hollow fibres are often used in dead-end and semi-dead end installations. Plate ceramic membranes cannot be bent, and therefore relevant ceramic membrane skid geometries are plate-and-frame, tubular (also called honeycomb) and hollow fibres.
The development of membrane modules started soon after the membranes entered large-scale industrial processes. Plate-and-frame modules developed in the 1960s are probably the oldest configurations used in commercial applications [33]. The design of modules was inspired by the filtration technology and is similar to a simple filter press. A module consists of multiple flat sheet units packed together as a multilayered sandwich. Each unit includes a support plate, a flat membrane sheet and feed and permeate spacers. A flat sheet placed on the permeate spacer is bent over the support plate, forming an envelope open to the feed from both sides. The edges of the membrane are sealed to the support. Many of these units, called cassettes, are stacked in parallel to form the module. Figure 1.7 shows an example of industrial tubular membranes made of alumina, the essence of the module. The feed fills the entire empty volume and is either released through a central permeate channel as the permeate or collected at the exit as the retentate. Two main advantages of plate-and-frame modules are the ease of cleaning and replacement of defective membranes and the ability to handle viscous feeds. A low packing density is the main disadvantage of plate-and-frame modules. The packing density of plate-and-frame modules can be increased with alumina multichannel monolithic elements [34] or by stacking many membrane sheets together [35].
Figure 1.7 Stacks of tubular alumina membranes after the sintering stage. (Photo courtesy of Atech Innovations GmbH.)
The packing density of flat sheets was increased in spiral-wound modules invented at the end of 1960s [36,37], a few years after the plate-and-frame modules. The module is arranged exactly as you would expect – several pairs of membrane sheets are placed back-to-back and then wound up. The edges of each pair of sheets are sealed to each other on three sides. On the fourth side, they are attached to a central perforated permeate channel of a pressure vessel. The sheets in a pair are separated by a fabric spacer that allows a permeate flow. A single spiral-wound module may consist of up to 20 pairs of sheets, each separated by a plastic mesh called a feed spacer. Viewed from the side, the path of a fluid on either feed or permeate side looks like an Archimedean spiral. The feed is usually pumped into the space outside the envelope through the feed spacers. Similarly, the permeate flows through a permeate spacer to the permeate channel. The advantages of the spiral-wound module are the high packing density and relatively low manufacturing cost. Its disadvantages are the difficulties in cleaning and repair of damaged membranes.
A parallel approach for increasing the packing density is to synthesize narrow hollow tubes, each only a few millimetres in diameter [38]. Although the diameter of each hollow fibre is small, a bundle of fibres packed inside a pressure vessel will have several square metres in total membrane surface. A typical commercial hollow fibre module consists of a few hundred to several thousand fibres. The fibres are often glued with a resin on the end that is far from the pump, so the feed must pass through the membrane. In this case, the hollow-fibre module works in a dead-end mode, having only two streams: one for the feed and one for the permeate. The retentate is held inside the fibres and released only during washing. If the feed can go through the membrane and be released as the retentate, the membrane works in a cross-flow mode. The modules are typically mounted vertically (although the option of a horizontal mounting does exist). If the skin membrane layer A (Figure 1.6) is synthesized near the lumen part of the fibre (the inside), the separation is performed in inside-out mode. If the skin layer is on the outer part of the fibre, the separation is performed in outside-in mode. The advantages of a hollow-fibre module include high packing density, a relative ease of cleaning and replacement/shutting down of single fibres and low dead volumes. Reported packing densities calculated as surface area to volume ratios are 30–250 m2/m3 for tubes, 130–500 m2/m3 for plate and frame and up to 9000 m2/m3 for hollow fibres [39]. The main disadvantage is the fragility of single fibres.
Hollow tubes can be small in diameter and thus fragile, or larger and more robust. Originally, tubular modules consisted of several single tubes with large inside diameters between 0.3 and 2.5 cm [40]. The wall thickness of a single element was about 2 mm. The tubes are bundled together similarly to hollow fibres and placed inside a plastic or stainless steel vessel to form a cartridge. These tubular membranes can have either circular or elliptic cross sections. Such membranes are usually cast in place within a support tube made of fibreglass, ceramic, plastic or stainless steel. Higher mechanical strength and relative ease of cleaning and replacement of single tubes are the advantages of tubular modules. The tubes are also less prone to clogging than fibres and spirals [41]. Low packing density, high capital cost and high dead end volume are the main disadvantages. A high internal hold-up volume of each tube allows creating turbulent flow regimes with intensive pumping and indeed tubular membranes are used in feeds with high solid content. Although there are some suppliers that still produce those elements, majority of them have switched to multichannel or honeycomb configurations. The channels of millimetres have become a part of one large element. The elements were produced commercially by SCT-Exekia and Orelis and called Membralox® and Kerasep® membrane modules, respectively. Further development of the elements changed the cylindrical shape of the channels for non-cylindrical flower-like geometries [42]. These elements with cross-sectional diameters of 10 or 25 mm and excellent packing density were developed by Metawater (Japan) and Tami Industries (France). The total dimensions of a single element can be up to 0.2 m in diameter and 1.5 m in length, and their total available membrane area is above 10 m2. The typical membrane element size cannot be increased any further due to limited hydraulic resistance of top layer and a chance for peeling off the top layer. Table 1.3 compares the relative advantages of each configuration, emphasizing the exploitation advantages of each module.
Table 1.3 Advantages and disadvantages of tubular and plate-and-frame modules.
Hollow fibre
Plate-and-frame
Tubular
Cost/area
Low
High
Low
Membrane replacement cost
Moderate
Low
Moderate/low
Flux (l/(m
2
h))
Good
Low
Low
Packing density (m
2
/m
3
)
Excellent
Good/fair
Good
Hold-up volume
Low
Medium
Medium
Cleaning in place
Good
Fair/poor
Fair/poor
Source:
Modified after Ref. [43].
Here the hold-up volume is defined as the volume of the fluid retained inside the filter during the filtration process. An additional parameter significant in the selection of a membrane module is the feed channel height. Selecting the correct height can prevent channel blockage. A rule of thumb in the industry is that the channel height should be at least 10 times larger than the diameter of the largest particle that can ever enter the channel. That ratio increases up to 25 : 1 in spiral-wound modules [44]. Companies resort to the lowest possible ratio to prevent intensive pumping. Larger pump capacities are needed in filtration with modules with high channel heights.
The above analysis of relative advantages and disadvantages of different modules has been known to membrane manufacturers for a long time. Still, companies use all types of modules when the exact design is a proprietary of the membrane manufacturer. For example, Asahi Kasei (Japan), GE Healthcare (United Kingdom) and inge AG (Germany) use hollow fibres. Pall Corporation (USA), Sartorius (Germany), GE Healthcare (UK) and Microdyn-Nadire (Germany) use plate-and-frame modules. Pall Corporation (USA), NovaSep Process (France), Tami Industries (France) and IBMEM (Germany) are using tubular modules. Koch membrane systems (USA) uses all types of modules.
1.6 Fouling and Cleaning
1.6.1 Fouling
A membrane module can be operated in either constant flux or constant pressure mode. In a constant flux mode, the flux through the membrane remains constant, but the TMP rises as the operation continues. Conversely, in a constant pressure mode, the pressure remains constant, but the flux decreases. In both cases, we understand that the main operational parameters change during the operation and membrane performance deteriorates. This happens because of fouling, defined as ‘The process resulting in loss of performance of a membrane due to the deposition of suspended or dissolved substances on its external surfaces, at its pore openings, or within its pores’ [45]. The deterioration is so severe that the membrane operation must be stopped periodically for cleaning. Both fouling and cleaning are two integral parts of the membrane operation that now can be properly described as a sequence of four stages: normal operation – fouling – cleaning – integrity test. Many books and thousands of research papers have been written on fouling and cleaning. Here we provide only a brief overview. Readers who are especially interested in fouling are referred to ‘Fundamentals of fouling’ by Field [46], ‘Fouling and cleaning of ultrafiltration membranes: a review’ by Shi et al. [47] or ‘Membrane chemical cleaning: from art to science’ by Liu et al. [48].
The loss of membrane performance can be due to two separate phenomena, but only one of them is called fouling. The other is the pre-concentration of solutes near the membrane surface due to a preferential passage of a solvent through the membrane. These solutes, however, are located near the membrane surface but not attached to the membrane in any way. They reduce the solvent activity and the flow. This is known as concentration–polarization, a natural phenomenon that can be mitigated by TMP and flux. Conversely, fouling is a phenomenon of attachment of solutes to the membrane surface that reduces the available membrane surface or blocks membrane pores. Two other forms of fouling are cake formation and gel formation. A ‘cake’ is formed when the fouling layers are built up on each other. In this case, the main fouling cause is the solute–solute attachment. In a case of the extreme concentration–polarization, certain macromolecules such as proteins can form a gel layer near the membrane surface.
Fouling is subdivided into four categories – inorganic, colloidal, organic and biofouling [49]. Inorganic fouling or scaling is caused by the accumulation of inorganic precipitates on membrane surface or within pore structure. Precipitates are formed when the concentration of chemical species exceeds their solubility limit. For example, if a reverse osmosis plant is operated at 0.75 recovery ratio (RR), the concentration of sparingly soluble salts in the retentate will be four times larger than their concentration in the feed. Some salts such as CaSO4, CaCO3 and silica are only slightly soluble and that pre-concentration level could be enough to cause scaling. Other salts such as CaF2, BaSO4 and SrSO4 have a scaling potential and their concentration should be carefully inspected not to exceed the solubility limits. The inorganic fouling due to concentration polarization in MF and UF membranes is much less common, but can occur due to chemical interactions between ions and other fouling materials, such as organic polymers.
Colloidal fouling can occur with different kinds of particles. Algae, bacteria and certain natural organic matters fall in the size range of particle and colloids. However, they are different from inert particles and colloids such as silts and clays. In most cases, particles and colloids do not really foul the membrane. This is because the flux decline caused by their accumulation on the membrane surface is largely reversible by hydraulic cleaning measures such as backwash and air scrubbing. A rare case of irreversible fouling by particles and colloids happens when they are small relative to the membrane's pores. Such particles and colloids can enter and get trapped within the pores. Removing these by hydraulic cleaning is difficult.
Organic fouling is caused by organic solutes with molecular weights ranging between few thousands and 1 million Da. For example, proteins present the main challenge in biopurification and food industry. The fouling is caused by complex solute–solvent–membrane interactions such as electrostatic, lyophilic/lyophobic, steric and covalent bonding effects. Many factors can affect organic fouling: the properties of the feed constituents such as size, hydrophobicity, charge density and isoelectric point; the properties of the membrane (hydrophobicity, charge density, surface roughness and porosity) and the properties of the solution phase, such as pH, ionic strength, and concentration of metals. Other factors of importance are the hydrodynamics of the membrane system (characterized by the solution flux) and surface shear.
Microbial fouling is a result of formation of biofilms on membrane surfaces. Once bacteria attach to the membrane, they start to multiply and produce extracellular polymeric substances (EPS) to form a viscous, slimy, hydrated gel. The EPS typically consists of heteropolysaccharides which have a high negative charge density. This gel structure protects bacterial cells from both hydraulic shearing and chemical attacks of biocides such as chlorine.
In gas separations, membrane fouling has often been neglected, mainly because feeds are protected by upstream filtration and are therefore relatively clean. The remaining foulants can damage the membrane or the seals but cannot cause fouling. There are a few reports on the organic fouling in vapour permeation [50]. We will discuss some typical fouling models in detail in Chapter 2.
