1.

11 Ceramic Membranes
Minghui Qiu, Xianfu Chen, Yiqun Fan, and Weihong Xing, Nanjing Tech University, Nanjing, China
r 2017 Elsevier B.V. All rights reserved.

1.11.1 Introduction 270

1.11.2 Fabrication Methods for Ceramic Membranes 274
1.11.2.1 Pore-Forming Agent Method 274
1.11.2.2 Template Method 274
1.11.2.2.1 Organic microspheres as templates 275
1.11.2.2.2 Surfactants as templates 276
1.11.2.3 Fiber Construction Method 276
1.11.2.4 Sol–Gel Method 276
1.11.2.4.1 Fabrication of ceramic UF membranes 277
1.11.2.4.2 Fabrication of ceramic NF membranes 277
1.11.2.5 Chemical Deposition Methods 279
1.11.2.5.1 Chemical vapor deposition method 279
1.11.2.5.2 Supercritical fluid deposition (SCFD) method 279
1.11.2.5.3 Other novel chemical deposition methods 279
1.11.2.6 Low-Cost Fabrication Methods 279
1.11.2.6.1 Low-cost and easy to be sintered raw materials 280
1.11.2.6.2 Low-cost sintering methods 280
1.11.3 Influence Factors of Ceramic Membrane Performance 281
1.11.3.1 Effects of Microstructure 281
1.11.3.2 Effects of Membrane Surface Properties 282
1.11.3.3 Effects of Solution Properties 283
1.11.3.4 Effect of Operation Parameters 285
1.11.4 Application and Case Study of Ceramic Membrane 286
1.11.4.1 Ceramic Membrane Reactors 286
1.11.4.1.1 Example: Cyclohexanone oxime production 287
1.11.4.1.2 Example: p-Aminophenol production 287
1.11.4.2 Oily Wastewater Treatments 287
1.11.4.2.1 Example: cold-rolling emulsion wastewater treatment 288
1.11.4.3 Biological Fermentation Treatments 288
1.11.4.3.1 Example: Lactic acid production 288
1.11.4.3.2 Example: Cephalosprin production 289
1.11.4.4 Plant Extraction 289
1.11.4.4.1 Example: Anthocyanin production 290
1.11.4.4.2 Example: Liqueur production 290
1.11.4.5 Brine Refinements 291
1.11.5 Final Remarks 293
Acknowledgments 293
References 293

1.11.1 Introduction

Ceramic membranes are typically made of aluminum oxide, zirconium oxide, or titanium oxide, silicon carbide, which all belong
to the group of inorganic materials. They are resistant to mechanical, chemical, and thermal stress and have high porosity and a
hydrophilic surface.1 Since the 1980s, ceramic membranes were gradually being used in civilian areas and developed rapidly.2
With their excellent properties, they are not only used in the field of water treatment, but also suitable for applications under harsh
environments, offering reliable performance over long periods of time.
The structure of membranes dictates the appropriate separation process. Membrane separation applications and the corre-
sponding sizes are given in Fig. 1. Ceramic membranes include pores with a defined average size, ranging from macrometers to
nanometers, and they can be used for liquid filtration, gas separation, and pervaporation. But ceramic membranes suitable for the
separation of gases are either microporous or dense, such as microporous silica membranes or zeolite membranes. The driving
force for gas transport is a concentration of chemical potential gradient across the membrane. In this article, we focus on ceramic

270 Comprehensive Membrane Science and Engineering II, Volume 1 doi:10.1016/B978-0-12-409547-2.12243-7

 Ceramic Membranes 271

Fig. 1 Membrane applications and corresponding separation processes.

Fig. 2 Schematic of an asymmetrical ceramic membrane cross section.

membranes for filtration of liquid from microfiltration (MF; 50 nm–1 mm), ultrafiltration (UF; 2–50 nm) to nanofiltration (NF;
o2 nm), which is the main application of ceramic membranes.
In general, ceramic membranes tend to have asymmetric structures composed of a thin, selective layer that fulfills the
separation requirements, and a permeable support structure comprising one to five layers, as shown in Fig. 2. Based on what cutoff
rate shall be achieved, several membrane layers are fixed on the support, starting from coarse intermediate layers to fine separation
layers, with pore sizes decreasing until the designated pore sizes are reached. The aperture ratio of the intermediate layer is less than
that of the support layer, and the purpose of the layer is to prevent particles from permeating the porous support layer while
preparing the function layer.3 Typically, the higher the membrane selectivity required, the more support layers are needed.
Moreover, the thinner top layer of asymmetric membranes enables greater permeability.
The support layer aperture ranges from 1 to 20 mm, with porosity of 30–65%, and its major role is to ensure the mechanical
strength of the overall membrane. Supports define the geometries of the ceramic membrane elements, and they can be flat sheet,
hollow fiber, tubular, or multichannel. Multichannel ceramic membranes have been widely used in industrial applications. While
hollow-fiber, tubular, and flat-sheet ceramic membranes are mostly used in laboratory studies, their applications in industrial
settings are in the initial stages. A picture of a ceramic membrane with a multichannel design is displayed in Fig. 3.
As already mentioned, according to the pore size of the separation layer, ceramic membranes for liquid filtration can be
classified into MF (pore size 450 nm), UF (pore size 2–50 nm), and NF (pore size o2 nm). The fabrication methods for ceramic
membranes strongly depend on their targeted structures, especially the range of the pore size. Generally, solid-state sintering and
sol–gel are the two common techniques for the preparation of microfiltration and ultrafiltration membranes in the large-scale
continuous process. These two methods have a similar process, starting from the well-dispersed suspension or sol coating on the
top of a membrane substrate, followed by heat treatment to achieve the desired final selectivity for MF, UF, or NF.
272 Ceramic Membranes

Fig. 3 Pictures of ceramic multichannel membrane. Photo courtesy of Jiuwu.

Table 1 Summary of the manufacturers that supply of ceramic membranes

Manufacturer Separation Process Materials Molecular Weight/Pore Size

Pall, USA MF a-Al2O3, 0.1–1.4 mm

UF ZrO2 20–100 nm
NF TiO2, SiO2 600–1000 Da
NGK, Japan MF a-Al2O3 0.1–1.2 mm
UF TiO2, SiO2 10–150 kDa
Jiuwu, China MF a-Al2O3, ZrO2 0.1–0.8 mm
UF a-Al2O3, TiO2, ZrO2 5–50 nm
NF TiO2, ZrO2 1000 Da
Atech, Germany MF a-Al2O3, TiO2, ZrO2 0.1–1.2 mm
UF TiO2, ZrO2, Al2O3 1–150 kDa, 50 nm
Tami, France MF a-Al2O3 0.14–1.4 mm
UF ZrO2/TiO2 3000–300,000 Da
NF TiO2, 1000 Da
Novasep, France MF ZrO2/TiO2 0.1–0.8 mm
UF ZrO2/TiO2 15–300 kDa
Inopor, Germany MF a-Al2O3, TiO2, ZrO2 0.07–1.0 mm
UF g-Al2O3, TiO2, ZrO2 3–30 nm
NF TiO2, SiO2 450–750 Da
CreaMem, USA MF SiC, a-Al2O3 0.1–0.5 mm
UF SiO2, TiO2 5–50 nm

The ceramic MF membrane is the first to be used successfully in industrial manufacture and commercial applications. Due to
their various advantages, which are important for the treatment and filtration of foods and beverages, most ceramic membranes
were used for the MF of milk, as well as prefiltration of wines and juices. After successful application in the dairy and wine
industries, the application of ceramic MF membranes gradually extended to the food industry, environmental engineering,
biological engineering, electronics industry, purification, and other fields. Commonly used materials for the separation layer
include Al2O3, ZrO2. and TiO2, supported on a porous a-Al2O3 substrate or multilayer composite asymmetric support. There are
currently dozens of vendors in the world that provide such ceramic membrane components, mainly in the United States, France,
Germany, Japan, and China (Table 1).
The ceramic UF membrane has a pore size of generally between 2 and 50 nm. They provide enhanced separation performance,
including high-separation precision and resistance to fouling, due to their smaller pore size and narrow distribution compared to
ceramic MF membranes. In the early stage, the UF membranes were also fabricated by a solid-state sintering process, but using
nanoparticles or nanofibers with smaller particle sizes. The typical applications for ceramic UF membranes are oily wastewater
treatment, beverage and juice purification, biological fermentation treatment and organic solvents filtration. To fabricate UF
membranes with ultrafine pore sizes (2–10 nm), the sol–gel technique starting from the colloidal or polymeric sol was introduced
and developed. Leenaars et al.4 applied the sol–gel technique successfully to prepare the Al2O3 membrane, and the AlOOH sol
synthesis processes and membrane permeation and retention performance were reviewed in detail in a series of papers. Subse-
quently, researchers in the United States, Japan, Germany, France, China, and other countries have been carrying out studies about
 Ceramic Membranes 273

this technology. The sol–gel approach has a lot of advantages over other methods of forming UF membranes. By means of sol–gel
processing, a wide variety of chemical compositions and large range of pore sizes (depend on the particle size of sol) can be
achieved. Also, the precursor materials are inexpensive and the processing costs are low as well. Currently, UF g-alumina, titania, or
zirconia membranes are already commercialized, basically by the sol–gel method, and are used for applications such as separation
of dye particles from water and to create intermediate or support layers for other membranes.
Ceramic NF has a molecular weight cutoff (MWCO) between 200 and 1000 Da. There are two types of solutes separated by
different mechanisms: ions based chiefly on their valence state in water (if solutes charged) and sieving based on molecular weight
(if solutes uncharged).5 With the development of nanotechnology, NF membranes have gradually assumed an important position
in a variety of processing applications, including the petrochemical, food processing, wastewater treatment, energy, and medical
industries, taking the place of traditional small molecule separation methods and achieving sustainable development of generic
technology. Ceramic NF membranes are fabricated in a large number of different ceramic materials, such as g-Al2O3, TiO2, ZrO2,
SiO2, TiO2–ZrO2, and SiO2–ZrO2. In recent years, ceramic NF membranes are moving from the laboratory research to industrial-
scale products, though only a small amount of ceramic NF membranes were applied in the industrial projects. The development of
preparation methods and optimization of membrane structures and performance for ceramic NF membranes were reviewed in the
latter of this article.
Critical to advancement in ceramic membrane technology are high membrane permeability and selectivity, and the use of
inexpensive and controllable fabrication methods. During the membrane processes, the permeability and selectivity are related to
the materials; microstructures such as pore size and distribution, porosity, and membrane thickness; and other surface properties.
By research into the quantitative relationships between controlling parameters and membrane microstructure, some mathematical
models for the process of membrane preparation have been found and the quantity control of membrane preparation can be
achieved.6 On the other hand, the surface properties of ceramic membranes play a significant role in their separation performance
and fouling tendency.7,8 It has been well known that a modified membrane with special surface properties promotes the ceramic
membrane applications in separating water-in-oil emulsions, oil purification, solvent recovery and protein purification.
Combining high performance in terms of flux stability, separation efficiency, and long lifetime with the continued decrease in
fabrication cost as market volumes increase, ceramic membranes are becoming competitive now and could even cause a radical
shift in the membrane market over the next decade. BlueTech Research has modeled ceramic membrane UF adoption for water and
wastewater applications over the previous 20 years, as illustrated in Fig. 4. Adoption of ceramic membranes has been fairly modest,
with an estimated 200 þ treatment plants representing a population equivalent of roughly 5 million people utilizing the tech-
nology over the past 20 years. The past 5 years have seen a small uptick in the application of ceramic membranes for treatment,
with two large-scale plants developed by PWN and Metawater (4100,000 m3/d) opening in 2014. Up to now, more than 10,000
items of ceramic membrane equipment have been established in the world, including about 2000 by Jiangsu Jiuwu High-Tech
Co. Ltd. in China.
An overview of membrane structure, preparation methods, and current applications is given in the following sections of
this article.

Fig. 4 Ceramic membrane UF adoption for water and wastewater applications over the previous 20 years.
274 Ceramic Membranes

1.11.2 Fabrication Methods for Ceramic Membranes

Membrane preparation is a core issue in the fields of membrane science and technology. Many scientists focused their attention on
this subject, while numerous engineers have developed new equipment to improve the technology of membrane preparation. In
the past several decades, fabrication technologies of ceramic membrane have made significant progress. Some new methods, such
as the pore-forming agent method,9 template method,10 fiber construction method,11 sol–gel method,12 and chemical deposition
method, as well as low-cost fabrication methods, have been successfully applied for the preparation of ceramic membranes to
enhance the permeability, selectivity, and separation precision, as well as antifouling properties. For example, to obtain highly
permeable ceramic membranes, pore forming and fiber construction methods are proposed. The pore forming approach is to
enlarge the number of pores and to increase the porosity of ceramic membranes by adding pore-forming agents. The template
method is a special type of the pore-forming agent method, in which the agent can create pores with specific sizes, as well as
ordered shapes. The fiber construction method uses ceramic fiber as raw material set up through the layers of fiber channels to
diversify the hole shapes, with the goal of improving porosity.
To develop ceramic membranes with smaller pores and higher separation precision for the NF process, the sol–gel technique is
currently the principal method used to prepare mesoporous and ceramic membrane. It usually employs metal alkoxides as a
precursor to forming polymeric or colloidal sols, which contain numerous nanosized particles. In addition, some chemical
deposition methods are also applied to improve the separation properties of ceramic membranes by decreasing pore size, patching
the defects, and changing the surface properties. Last but not least, to satisfy the requirements for commercial applications of
ceramic membranes, some fabrication methods are used that employ low-cost raw materials, a low-cost sintering process, and
some advanced fabrication technology. In this article, these methods will be discussed extensively.

1.11.2.1 Pore-Forming Agent Method

The permeability of ceramic membrane depends mainly on its porosity, pore tortuosity, and pore morphology. The pore-forming
agent method has been considered as a simple, efficient, and economical route to increase the porosity of the ceramic. The pore-
forming agent can be divided into two types: inorganic and organic substances. Inorganic pore-forming agents include inorganic
salts and carbon material, which are decomposable under high temperature, such as ammonium carbonate, ammonium bicar-
bonate, ammonium chloride, graphite, and coal powders. Organic pore-forming agents include natural fibers and polymers, such
as sawdust, starch, polystyrene (PS), and polymethyl methacrylate (PMMA).
To increase open porosity, carbon particles such as graphite or activated carbon are usually mixed with the ceramic powders
and then burned out by calcination or sintering. Yao et al.13 prepared mullite-corundum porous ceramic through an in situ
reaction sintering process by using kaolin and alumina as raw materials. It was found that the porosity increased from 29.7% to
50.6% by adding 20 wt.% graphite as the pore-forming agent. The use of graphite was effective in increasing open porosity and in
creating relatively large pores. But when a further amount was added, porosity decreased, likely because of the excessive pore-
forming agent creating more shrinkage during the sintering, which might partially offset the increase of porosity caused by the
pore-forming agent. Liu et al.9 studied the effects of graphite particle size and content on the properties of porous, cordierite-
connected SiC. The results shown that the porosity increased from 28.1% to 44.5% by using graphite with a particle size of 10 mm
as the pore-forming agent, when the adding amount was 25 wt.%. The degree of porosity of ceramic membranes created by this
method depended on the volume of the pore-forming agent, but it did not increase much when the particle size increased from 5
to 20 mm. Dong et al.14 investigated the effect of adding activated carbon on the porosity and pore size. It was found that the
porosity increased from 36% to 45% when the amount of activated carbon added increased from 0 to 17 wt.%, while the three
point bending strength was almost constant, at about 50 MPa. They also found that the pore size increased with sintering
temperature when the amount of activated carbon that was added was constant. The reason for this could be that the connection
between pores was promoted due to the bridge phenomenon of activated carbon at high temperatures.15 She et al.16 studied the
fabrication of porous SiC by using graphite as a pore-forming agent. When the amount of graphite increased from 25% to 60%,
the porous ceramic porosity increased linearly from 36.4% to 75.4%. Collier et al.17 obtained an alumina membrane with a high
porosity of about 64.6% after being sintered at 16001C by adding 15 wt.% starch as a pore-forming agent. As prepared, the
membrane showed an average pore size of about 10.1 mm and very high pure water permeability of about 122 m3/(m2 h bar).
Yang et al.18 fabricated a porous alumina membrane by using bentonite and cornstarch as the sintering aid and pore-forming
agent, respectively. Both the porosity and gas permeability of the porous alumina membrane were increased by adding 10 wt.%
cornstarch. The porosity increased from 24% to 38%, and the gas permeability increased from 1.68 to 6.86 m2.
Researchers have used the pore-forming agent to enhance the permeability of porous ceramic membranes, mainly by increasing
their porosity. During these processes, the porosity of ceramic membrane was quantitatively controlled, while keeping the average
pore size and its distribution at their initial levels. However, with this method, it is difficult to obtain membranes with fine pores.
However, it is suitable for the preparation of highly permeable membranes with large pores and high porosity.

1.11.2.2 Template Method

The template method is a kind of special pore-forming agent method that uses a regular or uniform pore-forming agent
to effectively control the morphology of the synthesized material to obtain an ordered pore structure and uniform pore size
 Ceramic Membranes 275

distribution. The template method has attractive characteristics, such as variability in the selection of template agents and
flexibility in the adjusting of pore structure. Hence, the template method has attracted a good deal of interest from researchers in
recent years.

1.11.2.2.1 Organic microspheres as templates

Velev et al.10 prepared three-dimensional (3D) ordered macroporous SiO2 materials by using colloidal crystals as a template. The
ordered macroporous SiO2 materials, with pore size in the range of 0.15–1 mm, could be obtained via changing the particle size of
the template. As prepared, macroporous SiO2 materials exhibited a large porosity (about 78%). Park et al.19 prepared 3D ordered
macroporous membranes with an average pore size of about 100 nm by using the organic PS microspheres as template via an
ultraviolet polymerization process. As shown in Fig. 5, the membrane exhibited a highly ordered, 3D porous structure. The surface
density of pores is about 2  109 cm2 and the diameter of each pore on the surface is about 100 nm. Sadakane et al.20 fabricated a
3D ordered macroporous ceramic by using PMMA as the template. The obtained porous ceramic showed great porosity, in the
range of 66–81%.
Many efforts have been made to increase the membrane porosity and improve the membrane flux. Zhao et al.21,22 obtained 3D
ordered macroporous ZrO2, SiO2, and Al2O3 symmetric ceramic membranes by using PMMA as the template. As prepared, ZrO2
and SiO2 ceramic membranes showed a great porosity of about 60% and 48.1%, respectively. Fig. 6 shows scanning electron
microscopy (SEM) images of the ordered macroporous Al2O3 membranes prepared using negatively charged PMMA (680 nm) and
positively charged Al2O3 particles at a pH of 4. It indicates that the membranes prepared with this template method have high
porosity. Furthermore, this structure has modified pore sizes and low sintering temperatures. They also obtained Al2O3 asymmetric
via a dip-coating method. Xu et al.23 mixed PS microspheres with TiO2–SiO2 composite sol to prepare a TiO2–SiO2 composite

Fig. 5 SEM images (A, top view; B, cross-sectional view) of 3D ordered macroporous membranes.

Fig. 6 SEM images of 3D ordered macroporous Al2O3 membranes, using PMMA as the template.
276 Ceramic Membranes

membrane on porous substrate via a spin-coating process. As prepared, the composite membrane showed an average pore size of
about 200–300 nm.

1.11.2.2.2 Surfactants as templates

Since the surfactants can self-assemble in solution to form micelles, microemulsion, liquid crystal, vesicles, and other self-
assembled substances, surfactants are usually applied as the organic template agents. In 1992, scientists from the Mobil Company
fabricated the ordered mesoporous MCM-41 for the first time by using cetyltrimethyl ammonium bromide (CTAB) as an organic
template.24 Such mesoporous materials display various symmetric pores, with the pore size in the range of 2–50 nm. Ji et al.25
fabricated mesoporous MCM-48 silica membranes on porous supports under hydrothermal conditions by using CTAB as the
template surfactant. In their study, the solvent extraction method was applied to remove the templates instead of calcination. The
results showed that the solvent of 1 M of HCl/EtOH solution was sufficient to extract the templates in the MCM-48 materials, and
over 90% of the templates were extracted at room temperature for 24 h. Xu et al.26 further studied the preparation of mesoporous
MCM-48 membranes by using H2O as a pore-filling substance to prevent the sol solution from infiltrating the support pores.
The H2/N2 separation factor was 3.47, with a transmembrane pressure (TMP) of 0.05 MPa.
Different template agents could be applied for the fabrication of ordered silica membranes. Kumar et al.27 prepared meso-
porous MCM-48 membranes on porous substrates via a hydrothermal method using CTAB as the template. Two kinds of substrate
were applied, which have average pore sizes of about 200 and 300 nm, respectively. The test results showed that the average pore
size and porosity of this membrane were 2.6 nm and 40%, respectively. Meanwhile, as prepared, the membrane layer exhibited
a highly ordered pore structure, which was very beneficial for improving membrane performance. Zhang et al.28 prepared an
unsupported ordered mesoporous SiO2 membrane and a supported ordered mesoporous SiO2 membrane on Al2O3 substrate by
using cetyltrimethyl ammonium chloride as the template agent. As prepared mesoporous SiO2 membrane showed a homo-
geneous pore size distribution and small pore size (2–3 nm).
In addition to ordered silica membranes, Choi et al.29 obtained TiO2–Al2O3 composite ceramic membranes with gradient pore
structures and improved permeability by using Tween80 as the template. The gradient pore sizes were 2-6, 3-8, and 5-11 nm from
the top to the bottom layer, and the corresponding porosities were as high as 46.2%, 56.7%, and 69.3%, respectively. Cao et al.30
fabricated a TiO2 UF membrane by using a P123-templated sol–gel method. As prepared, the UF membrane exhibited a cutoff
molecular weight of 7500 Da and a high pure water flux of 170 L/(m h bar).

1.11.2.3 Fiber Construction Method

Due to its slim configuration, ceramic fiber could quickly stack and bridge on the surface of macroporous substrate, so as to reduce
the phenomenon of penetration. Meanwhile, it is easy to obtain ceramic membranes with a high porosity and surface area, which
significantly affects the improvement of permeability for ceramic membranes.11
Lei et al.31 obtained a ceramic MF membrane with an average pore size of 2.6 mm through a dip-coating method by using TiO2
fibers as raw materials. This membrane performed great pure water permeability, which is almost twofold of that prepared from
particle raw materials. Ke et al.11,32,33 fabricated Al2O3 and TiO2 nanofiber membranes on different porous supports by using large
TiO2 fibers and small boehmite fibers as raw materials. The randomly oriented titanate nanofibers can completely cover the rough
surface of the porous substrate, leaving no pinholes or cracks. On the top of this titanate fiber layer, a layer of alumina fibers was
formed using boehmite nanofibers. They found that the differences in the porous supports have no substantial influences on the
texture of resulting membranes. The resulting UF membrane showed an average pore size of about 50 nm and a high rejection
ratio of more than 95% for nanospherical particles with an average particle size of 60 nm. Meanwhile, the membrane provided a
large porosity (more than 70%), leading to a high permeability of 900 L/(m2 h). They further synthesized boehmite nanofibers by
a hydrothermal reaction from alumina colloidal gels for the fabrication of nanofiber membranes with smaller pores.34 As
prepared, alumina nanofiber membranes showed an average pore size of about 11 nm and excellent separation efficiency, with an
MWCO of 70 kDa.
The fiber construction method promotes the permeability of ceramic membrane by increasing the porosity. However, the
strength of membrane layer is weakened. Therefore, several methods were proposed to strengthen the neck connection between
fibers. Fernando et al.35 added phosphate as a binder agent to promote the neck connection of Al2O3 fibers during the fabrication
of Al2O3 fiber membranes with improved strength. Qiu et al.36 added TiO2 sol particles to TiO2 fibers to fabricate TiO2 fiber UF
membranes. The addition of TiO2 sol particles could promote the formation of a strong neck connection between fibers at low
sintering temperatures. Hence, the breaking and aggregation phenomenon of ceramic fibers at high temperatures could be avoided
efficiently.

1.11.2.4 Sol–Gel Method

Since the fabrication of ceramic MF and UF membranes has matured, recent studies on the promoting of ceramic membranes are
focusing on two main aspects. One is to fabricate ceramic membranes with large pores for gas cleaning at high temperatures.37–39
The other is to fabricate ceramic membranes with small pores for NF and even gas separation processes. For the latter purpose, the
sol–gel method has be considered as an effective and the most promising one for commercial production of ceramic membranes
with a nanoporous structure.
 Ceramic Membranes 277

The sol–gel method controls the separation precision of membrane layers, mainly by adjusting the particle size of colloids in
sols. The colloid size could be controlled to be as small as several nanometers, so as to prepare UF and NF ceramic membranes
with small pore size, narrow pore size distribution, and high separation precision.

1.11.2.4.1 Fabrication of ceramic UF membranes

The g-Al2O3 UF membrane was one of the earliest ceramic UF membranes fabricated by the sol–gel technique.4 Subsequently,
numerous studies have been carried out that focus on sol synthesis processes to improve membrane permeation and retention
performance. Das et al.12 fabricated g-Al2O3 UF membranes on porous substrate by using a sol–gel method. An MF membrane
with an average pore size of 0.1–0.7 mm was applied as support for the deposition of the UF membrane layer. Colloidal sol with an
average particle size of 30–40 nm was used to form the UF membrane layer. As prepared, the UF membrane displayed a high
rejection ratio of almost 100% to remove colibacillus from water.
Compared to g-Al2O3 membranes, ZrO2 and TiO2 are more popular for the fabrication of ceramic UF membranes due to their
better chemical stability. Vacassy et al.40 prepared ZrO2 UF membranes via a polymeric sol–gel method. The ZrO2 UF showed
rejection ratios of 54% and 73% for sucrose (MW¼ 342 g mol1) and vitamin B12 (MW ¼1355 g mol1), respectively. Ju et al.41
obtained ZrO2 UF membranes with smaller pores at lower sintering temperatures by using a colloidal sol–gel method. The
sintering temperature for the ZrO2 UF membrane decreased from 11001C to 5001C Meanwhile, the average pore size of for ZrO2
UF membrane decreased from 50 to 20 nm. Consequently, the separation precision improved significantly. Manjumol et al.42
obtained TiO2 UF membranes with an average pore size of about 5 nm by using a colloidal sol–gel method. As prepared, the UF
membrane had a high Bovine serum albumin (BSA) (MW ¼ 67,000 g mol1) retention of more than 98%. Fan et al.43 fabricated a
multichannel TiO2 UF membrane layer by using a colloidal sol–gel method, where multichannel substrate with an average pore
size of about 200 nm was applied. The MWCO of this TiO2 UF membrane was tested by using dextran as the standard substance
and showed a value of about 9000 Da. Subsequently, as prepared, the multichannel TiO2 UF membrane was used for the removal
of direct black (MW ¼ 909 g mol1) and polyethylene glycol (PEG) (MW ¼70,000 g mol1) from wastewater. The retention ratios
for both these two substance were more than 99%.

1.11.2.4.2 Fabrication of ceramic NF membranes

Ceramic NF membranes with higher separation accuracy can be used for selective separation of oligosaccharides, dyes, multivalent
ions, and other elements. The sol–gel process is one of the most appropriate methods for the preparation of ceramic NF
membranes. Two sol–gel routes are generally described in the literature.44 One utilizes the chemistry of metal-organic precursors in
organic solvents; the other is based on colloidal chemistry in aqueous media. These two processes are termed the polymeric sol–gel
route and the colloidal sol–gel route, respectively.
The polymeric sol–gel route is the most frequently used to prepare ceramic NF membranes. It could preserve individual grains
of less than 10 nm due to the mineral core surrounded by an organic shell, which prevents aggregation. By using a-Al2O3
macroporous membrane with an average pore size of about 1 mm as the substrate, Tsuru et al.45 fabricated TiO2 NF membranes
with an average pore size of 1.2 nm by repeating several coating processes with a polymeric sol–gel method. As prepared, the TiO2
NF membrane showed an MWCO of 600 Da and a rejection ratio of 60% for NaCl. In a later study, they prepared TiO2 NF
membranes with controllable pore size in the range of 0.7–2.5 nm by using a polymeric sol–gel method.46 Such membranes
showed MWCO in the range from 500 to 2000 Da based on the retention of PEG. Qi et al.47 prepared a polymeric TiO2 sol with an
average particle size of about 1.2 nm. Subsequently, TiO2 NF membranes with an MWCO of 890 Da were obtained by using this
polymeric TiO2 sol. As prepared, the TiO2 NF membrane showed high rejection ratios of 96.5% and 92.8% for Ca2 þ and Mg2 þ
ions, respectively, at a salt concentration of 0.025 mol/L, a pH value of 4.0 and a TMP of 5  105 Pa.
With respect to the application of ceramic NF membranes in harsh conditions, ZrO2 has been considered a promising material.
Van Gestel et al.48 reported a ZrO2 disc NF membrane with an MWCO of o300 Da using alkoxide as the precursor modified by
acetylacetone via the polymeric sol–gel method. Benfer et al.49 obtained ZrO2 NF membranes through the polymeric sol–gel
method by using propanol zirconium as a precursor. The rejection ratio of direct red (MW ¼990.8 g mol1) tested for this ZrO2 NF
membrane was as high as 99.2%.
Meanwhile, it is noted that TiO2–ZrO2 composite NF membranes showed a much higher phase transition temperature and
thermal stability than pure NF membranes or NF membranes.50 Aust et al.51 prepared TiO2–ZrO2 composite NF membranes via
the polymeric sol–gel route. The pore size and separation precision could be controlled by adjusting the mole ratios of the
zirconium salt precursor and titanium salt precursor. All the prepared TiO2–ZrO2 composite NF membranes showed good
retention for direct red dye (MW ¼ 990.8 g mol1), with rejection ratios as high as 95%.
Unlike the polymeric sol–gel route, as an environmentally friendly alternative to a process involving a variety of organic and
hazardous solvents, the colloidal sol–gel route has attracted more and more attention due to using water as solvent, which is
physiologically harmless, nontoxic, not very volatile, low-cost, and available in large quantities. Convincingly, the colloidal sol–gel
route will be preferred in industrial-scale production. However, a difficulty in the fabrication of TiO2 NF membranes via the
colloidal sol–gel technique might be the prevention of particle aggregation at the sol stage in order to yield microporous ceramics.
For this reason, a highly stable particulate sol with particle size below 10 nm should be synthesized first. A modified colloidal
sol–gel process, with organic additives involved, as shown in Fig. 7, was proposed for the fabrication of TiO2 NF membranes, with
an MWCO of about 820 Da and a pure water permeability of about 8 L/(m2 h bar) achieved for the membranes.52
278 Ceramic Membranes

Fig. 7 A modified colloidal sol–gel process.

Fig. 8 Pure water permeability (a) and PEG retention (b) of the YSZ NF membrane.

Fig. 9 Field-emission SEM images of TiO2–ZrO2 NF membrane: (a) surface and (b) cross section.

Subsequently, the crack-free thin g-Al2O3,53 YSZ,54 ZrO2,55 and TiO2–ZrO256 NF membranes were prepared by a colloidal sol–gel
route. The YSZ NF membrane exhibited great water permeability of 28 L/(m h bar) and a low MWCO of 800 Da (Fig. 8).54 When the
filtration was operated at 601C, the fluorescent brightener recovery value of 99% was obtained, while NaCl removal was greater than
98%, indicating that ceramic NF membranes were the technology of choice for the treatment of high-salinity dye wastewater.
The combined utilization of different metal oxides to fabricate composite ceramic membranes has attracted more and more
interest in increasing the diversity and performance of NF membranes. Tsuru et al.57 prepared a series of SiO2–ZrO2 colloidal
composite sols with various particle size distributions. Based on these composite sols, they obtained SiO2–ZrO2 composite NF
membranes with pore sizes in the order of 2.9 nm, 1.6 nm, and 1.0 nm, which decreased with the particle size of colloidal
composite sols. Lu et al.56 fabricated TiO2–ZrO2 composite NF membranes via a modified colloidal sol–gel process using
a mixture of Zr-inorganic salts and Ti-alkoxides. The FESEM images of the TiO2–ZrO2 NF membrane is shown in Fig. 9.
 Ceramic Membranes 279

It demonstrated that a well-defined and crack-free top layer with a thickness of about 200 nm was formed. The composite NF
membrane exhibited a low MWCO of about 500 Da and a high pure water permeability of 35–40 L/(m h bar) and showed desired
retention rates of 99.6% for 1 ppm Co2 þ , 99.2% for 13 ppm Sr2 þ , and 75.5% for 7 ppm Cs þ in simulative radioactive wastewater
treatment (pH 3, T ¼ 251C), suggesting a great potential for ceramic NF membranes in radioactive effluent disposal.

1.11.2.5 Chemical Deposition Methods

Since the sol–gel method has been commercialized for the fabrication of ceramic, tight UF membranes, and even ceramic NF
membranes, researchers are developing chemical deposition methods to further improve the permeability and enhance the
selectivity of ceramic membranes by reducing the pore size and refining the pore size distribution. Several developed modification
techniques are chemical vapor deposition (CVD), supercritical fluid deposition (SCFD), atomic layer deposition (ALD) tech-
nology, and surface grafting. These chemical deposition methods can not only fix the defects and improve the stability of
membranes,58 but also further reduce the pore size to improve the accuracy of the separation.

1.11.2.5.1 Chemical vapor deposition method

The CVD method is an efficient way to optimize the pore structure and improve the permeability of ceramic membranes by
depositing silicon oxide or metal oxide on porous substrate.59 Labropoulos et al.60 reduced the pore size of SiO2 membranes from
1 to 0.56 nm by recycling the CVD method at 573 K. Lin et al.61 modified g-Al2O3 ceramic membranes with an average pore size of
4 nm by the CVD method. As a result, SiO2 membranes with an average pore size of 0.4–0.6 nm and a membrane thickness of
1.5 mm were obtained. Fernandes et al.62 deposited silanized silicon tetrachloride solution on porous silica glass by using the
CVD method. The pore size was reduced from 4.4 to 2 nm. However, CVD processes typically require a high temperature and
vacuum environment, and the precursors should be volatile. Hence, the most CVD methods are usually applied in the research in
laboratories.

1.11.2.5.2 Supercritical fluid deposition (SCFD) method

SCFD technology is based on supercritical fluid solvents (such as the typical SC-CO2), which carry precursors to deposit in the
pores of porous ceramic. This method can be used for the modification of ceramic membranes.63 In this process, the solubility of
the ceramic precursor in the supercritical fluid is reduced by reducing the pressure. As a result, the precursor deposited in the pores
of ceramic substrates. Hence, the pore size of ceramic membranes was reduced. Wang et al.64 established a mathematical model to
describe the supercritical fluid penetration process based on a set of kinetic equations, a phase equilibrium model of supercritical
solution, and the classical nucleation theory. Subsequently, they applied this mathematical model in experiments and reduced the
average pore size from 110 to 80 nm. Brasseur et al.65 proposed the SCFD method by using supercritical isopropanol as a solvent
for the deposition of titanium alkoxide precursor onto the alumina substrate. It was found that the pore size of substrate decreased
significantly, from 110 to 5 nm. Tatsuda et al.66 modified mesoporous SiO2 through a SCFD method by using titanium tetra-
isopropoxide as precursor, and using SC-CO2 as solvent. The result showed that the pore size was reduced significantly due the
penetration of precursor into the pore of mesoporous SiO2.

1.11.2.5.3 Other novel chemical deposition methods

ALD is a thin film deposition technique that is based on the sequential use of a gas phase chemical process. It is considered a
subclass of CVD. The majority of ALD reactions use two chemicals, typically called precursors, which react with the surface of a
material one at a time in a sequential, self-limiting, manner. Through repeated exposure to separate precursors, a thin film is slowly
deposited. Li et al.67 deposited an alumina layer onto substrate with an average pore size of 50 nm through ALD. After the ALD
process was repeated for 600 cycles, the BSA rejection ratio of this membrane increased from 2.9% to 97.1%.
Similar to the chemical deposition methods, the surface grafting technique is currently used to adjust the surface properties of
membrane materials by forming an ultrathin polymer layer (even smaller than 1 nm) instead of an inorganic deposition
layer.68–74 Meanwhile, for membranes with small pores, the pore structure could also be reduced during the grafting process. The
ceramic membrane surface generally adsorbed water to form hydroxyl groups, and a layer of organic molecules can be modified by
grafting the organosilane method onto the mesoporous membrane surface. The pore size and surface properties of membrane
could be controlled by varying the grafting substance with different functional groups and molecular chain length so as to satisfy
the various demands of different application systems.75 Sah et al.70 successfully reduced the pore size from 2.3 to 2.0 nm by using
trimethylchlorosilane as the grafting substance. Faibish et al.72 reduced the pore size by 25–28% to increase the retention
performance of ceramic membranes by grafting polyvinylpyrrolidone onto ceramic UF membranes through a two-step reaction.

1.11.2.6 Low-Cost Fabrication Methods

The high production cost of ceramic membranes limits their applications. The preparation cost of ceramic membranes mainly
came from raw materials and the sintering process. Therefore, it has been a research focus to reduce the fabrication cost of ceramic
membranes, such as reducing the sintering temperature with sintering aids, applying low-cost raw materials, and advanced
sintering techniques.
280 Ceramic Membranes

1.11.2.6.1 Low-cost and easy to be sintered raw materials

To create sintering ceramic membranes at low temperatures and reduce the fabrication cost, liquid-state sintering aids and solid-
state sintering aids were often applied to promote sintering at low temperatures. The liquid-state sintering aids include former
kaolin, potassium feldspar, and other natural silicate clay minerals, which could melt and form liquids at low temperatures. The
liquid phase would wet and coat the particles under the drive of capillary forces. Subsequently, the sintering aids connect the
particles and enhance the strength of the ceramic. The solid-state sintering aids include titania, zirconia, yttrium oxide, and other
melt oxides, which could form a solid solution system with alumina to reduce the sintering temperature.
Hu et al.76 added a-Al2O3 powder with a smaller particle size of 0.5 mm to a-Al2O3 raw materials (22 mm) as a sintering
aid. Meanwhile, ammonium polymethacrylate and polyethyleneimine were used to achieve the coating of small particles on
the surface of large particles. The sintering temperature decreased from 17001C to 15501C. As prepared, a-Al2O3 ceramic mem-
brane showed a strength of 34.2 MPa, a porosity of 34%, an average pore size of 2.34 mm, and a pure water permeability of
205 m3/(m2/h) (1 MPa).
Falamaki et al.77 studied the dual behavior of CaCO3 as a porosifier and sintering aid in the manufacture of disk alumina
supports. It was found that there is an optimum sintering temperature of 13501C corresponding to the highest permeability when
the porosifier initial content was in the range of 1–7 wt.%. Meanwhile, the maximum permeability is obtained using 5 wt.% initial
porosifier content. They also demonstrated that the pore growth mechanism was not the reason for such enhanced permeability.
Instead, the decrease in tortuosity was the main reason, caused by the “round pore shaping” mechanism during the establishment
of a liquid-phase sintering process.
Wang et al.78 realized the reaction sintering of porous alumina ceramic membrane support at 13001C when the commercial
titanium white powder (rutile) was used as a sintering aid. It was also found that the microstructures of a sintered body were
significantly affected by the amount of rutile, organic pore-forming agents, and sintering techniques. These results indicated that
the porosity dramatically decreased, as did the pore radius (from 1.6 to 0.8 mm) when less than 10.0 wt.% of rutile was added.
For energy saving purposes, the shorter sintering time (2–4 h) at 14001C is long enough to obtain ceramic membrane support with
a high open porosity and desired mechanical strength, as well as better resistance to caustic corrosion.
Cheap, natural, nonmetallic minerals, such as kaolin, industrial mullite, fly ash, and other industrial waste, were also used as
raw materials to achieve low-cost fabrication of ceramic membranes.79–81 Almandoz et al.82 made nonsupported MF ceramic
membranes from various aluminosilicate paste formulations. The optimum sintering temperature for cast green dopes was
between 12001C and 13001C. Dong et al.83 prepared cordierite-based porous ceramic MF membranes on tubular macroporous
cordierite supports by using waste fly ash and basic magnesium carbonate as the starting materials. Majouli et al.84 fabricated the
flat support from natural perlite, of which the powder was crushed, sieved to 200 mm, and mixed with organic additives and water.
The firing temperature for the support was optimized at 10001C. As obtained, elaborated support showed an average pore
diameter of 6.64 mm and a porosity of 41.8%. Meanwhile, it also performed a pure water permeability of 1797 L/(h m2 bar).
In addition, the chemical resistance was also tested, and the results showed that the perlite support presents a better chemical
resistance in an acidic medium than in a basic medium.

1.11.2.6.2 Low-cost sintering methods

To obtain a multilayer asymmetric structure, the sintering process must be repeated several times, which may cause a long
fabrication period and high cost. Hence, the sintering process has become the bottleneck for low-cost fabrication of ceramic
membranes. To reduce the cost of ceramic membranes, cutting down the sintering period has become another efficient method to
reduce the fabrication cost of ceramic membranes, besides the methods using low-cost and easy-to-sinter raw materials.
Some researchers use rapid sintering technology to shorten the sintering time and reducing the cost of sintering. Among the
methods to do so, microwave sintering technology is mostly used. The technology is a noncontact technique in which heat transfer
in the form of electromagnetic waves. In this way, the heat could access the internal area of materials directly. The nonuniformity
of sintering could be avoided efficiently. As a result, both the sintering time and temperature could be reduced.85 Microwave
techniques are always used for the preparation of near-dense ceramic composites, such as Al2O3–ZrO2 86 and Al2O3–TiO2.87 In
addition, microwave techniques could improve the tissue of ceramic membranes and enhance the properties. Therefore, these
microwave techniques could be used for the fabrication of porous ceramic membranes. Oh et al.88 prepared porous composites of
Al2O3/ZrO2 with improved mechanical properties by microwave sintering. The results demonstrated the neck growth between the
initially touching particles was achieved by the selective heating of ZrO2 between Al2O3 particles in the microwave heating, which
because of the high dielectric loss factor and good microwave-energy absorption ability of ZrO2. As a result, the composite
prepared by the microwave sintering exhibited higher elastic modulus and fracture strength than that prepared by conventional
sintering.
For purposes of shortening the sintering period, some researchers introduced cosintering technology from the area of packaging
multilayer ceramic devices for the cutting down of the fabrication cost of ceramic membranes by reducing the cycles of sintering.
Feng et al.89,90 proposed a cosintering process to reduce the production time and cost for the fabrication of tubular bilayer
a-alumina membrane. It was found that the first problem to be solved was the shrinkage mismatch between two membrane layers
during cosintering. The selection of the cosintering temperature was the main issue. Otherwise, defects such as cracks and big
pores could be caused by the too-large mismatch in the sintering shrinkage of two membrane layers. The membrane prepared by
cosintering at 13001C was defect-free and the interfacial bonding was strong. These advantages were beneficial from the moderate
shrinkage mismatch between two membrane layers during cosintering. Meanwhile, the cosintered membranes performed a
 Ceramic Membranes 281

stable permeability and microstructure after ultrasonic treatment. In addition, during the filtration of the CaCO3 suspension, the
membrane treated by cosintering had a higher steady permeate flux than the membrane prepared by the conventional process and
can withstand a high backflushing pressure.
In a subsequent study,90 a cosintering process was applied to fabricate bilayer ZrO2/Al2O3 membranes on Al2O3 substrates. It
was found that good bonding between the membrane layers and the support after cosintering could be achieved by using thinner
sublayer membrane without defects. However, the top-layer membrane should be not very thin. Otherwise, it would be insuf-
ficient to promote the sintering of the sublayer membrane during the cosintering process due to the compressive stress caused by
fast shrinkage of the top-layer membrane. The thicknesses for the sublayer and the top-layer membranes were optimized as 15 and
10 mm, respectively. The bilayer ZrO2/Al2O3 membrane with an average pore size of about 0.28 mm and a pure water flux of about
2.82  103 L/(m2 h bar) was obtained after being cosintered at 12001C.
To use the cosintering method for the fabrication of ceramic membranes with smaller pores, Qiu et al.36 prepared titania UF
membranes by using sol-coated nanofibers as the material for the sublayer during the cosintering process. In this respect, titania
nanofibers were applied to cover the porous substrate to produce a uniform layer with high porosity and flux. Meanwhile, titania
nanoparticles from sol were used to improve the mechanical strength of the titania nanofiber membrane. The enhance mechanism
could be explained by the formation of sintering neck between nanofibers with colloidal particles (sol) at lower sintering
temperatures. Subsequently, titania colloidal particulate sol is deposited on the top of the titania nanofiber layer and followed by a
cosintering process at a suitable sintering temperature of 4801C to reduce the pore size and achieve high separation efficiency. As
prepared, the UF membrane has a homogeneous surface without obvious defects. Meanwhile, the UF membrane showed high
pure water flux permeability of about 1100 L/(m2 h bar), while the MWCO is 32,000. Dong et al.91 prepared porous cordierite MF
membranes by a cosintering process. Meanwhile, the low-cost, industrial-grade powder was used as the raw material to reduce
fabrication costs and process time. The results demonstrated that the cosintering of two-layer cordierite membranes (45.0 and 15.0
mm thick) was feasible due to their low difference in sintering shrinkage behavior.

1.11.3 Influence Factors of Ceramic Membrane Performance

Application-oriented consideration for the fabrication of ceramic membranes mainly includes the designs of microstructure,
material properties, and operation parameters for ceramic membranes. The separation performance of ceramic membranes closely
depends on their structure parameters and material properties. The structure parameters include average pore size, pore size
distribution, membrane thickness, porosity, pore shape, and tortuosity, which are the main factors in the permeability of ceramic
membranes. The material properties include chemical stability, thermal stability, mechanical strength, and surface properties,
which not only affect the permeability of the membrane separation performance but also have a close relationship with membrane
lifespan. The operation parameters include flow velocity on membrane surface, TMP, and temperature, which affect the degree of
concentration polarization, membrane pollution, and permeability and rejection properties.

1.11.3.1 Effects of Microstructure

With the development of ceramic membrane preparation technology, the specifications of ceramic membranes increased a good
deal. Meanwhile, the controllability of preparation technique made it possible to adjust the microstructure of ceramic membranes.
Hence, considering the different performance of membranes with different pore sizes in a similar application system, more and
more research has focused on the effects of membrane microstructures on the performance of ceramic membranes.
Pore size is the major factor affecting membrane permeability and separation performance. For pure solvent systems, the bigger
the pore size, the larger the permeability of ceramic membrane is.92 However, due to the effects of absorption, concentration
polarization, and membrane-fouling phenomena such as clogging, the actual filtration systems rarely show comparable permeate
flux values to the pure water permeate flux of membranes. In the treatment of oilfield-produced water93 and apple cider,94 it was
found that the flux declined severely with the increase in membrane pore size. Only in the case of membrane systems with
suitable pore size could they achieve their highest permeate flux when the total resistance of membrane and fouling layer reach the
minimum value.95 Therefore, in membrane applications, the membrane should have a suitable pore size distribution to achieve as
great a permeability as possible. Meanwhile, the rejection properties should be maintained at the required level. As shown in
Fig. 10, the permeability did not linearly increase with the pore size of membranes in the separation of ovalbumin from dextran
solutions. The highest permeability was obtained when the membrane pore size was 0.8 mm. It could be that the clogging
phenomenon caused the increase of transport resistance when the membranes with larger pore size were used. Only when the
membrane pore size matches the particle size of the target substance can the membrane system reach the highest permeate flux.
Effects of membrane layer thickness on the performance are mainly expressed as permeability, since the path length of fluid
passing through membrane obviously increases with the membrane thickness. As a result, the transport resistance increases and the
permeability decreases with the increase of membrane-layer thickness. However, membrane thickness can also affect the retention
properties of membrane. Cheng et al.96 studied positively charged microporous ceramic membranes for the removal of Titan
Yellow through electrostatic adsorption, and found that with the removal rate of Titan Yellow increased with membrane thickness.
The researchers explained that both the filtration path and the effective adsorption area grew with the increase of membrane
thickness, so the removal rate increased as well.
282 Ceramic Membranes

Fig. 10 Effect of pore size on the permeability of ceramic membrane in the filtration of ovalbumin and dextran solutions.

The ceramic membranes with high porosity mean having more open pores and high permeability. Generally, the porosity of
membrane layers is about 20–60%, while the porosity of substrates is much larger than that of membrane layers. For MF
membranes, the porosity is usually larger than 30%.97
Considering the abovementioned phenomenon, some researchers have established theoretical relationships between structures
and performance, so as to build application-oriented structure design methods for ceramic membranes. Xu et al.98–100 proposed a
research framework of application-oriented ceramic membrane design. A new, crossflow MF model of particle suspensions was
built, in which a blocking factor (k) was proposed to describe membrane fouling.98 The correlation between membrane per-
meability and membrane structure parameters was described in this model. Furthermore, the variation of permeate flux with
filtration time was obtained, and the influences of membrane structure parameters such as pore size, thickness, and porosity on
permeate flux could be theoretically predicted under the ideal conditions through this model. The simulated results showed that
there was an optimum pore size in the filtration of suspensions to obtain maximal permeate flux. Meanwhile, the optimum pore
size was influenced by the size and size distribution of the particulates.99 Finally, the novel membrane design and application
methods were well verified in the nanosuspensions system of titania, where the model calculation showed good agreement with
the experimental results.100

1.11.3.2 Effects of Membrane Surface Properties

The hydrophilicity and charging properties of ceramic membranes play an important role in membrane applications. The elec-
trochemical characters of the membrane surface affect the interactions between the substance of fluids and membranes. Conse-
quently, the permeability and rejection properties of ceramic membranes were influenced. Zeta potential and isoelectric point
(IEP) are always used for the characterization of membrane electrochemical properties.
Elzo et al.101 studied the effects of the membrane-charging state on the performance of ceramic membranes. They performed
zeta potential measurements on the a-alumina particles to represent the zeta potential of the membrane surface. Moritz et al.8
investigated ceramic UF membranes, which were made of TiO2, ZrO2, and a- and g-Al2O3 by streaming potential measurements. It
was demonstrated that the interaction between the membrane surface charge and the charge of the rejected substances has an
increasing influence on the filtration results. Meanwhile, the results also illustrated that the electrochemical interactions with the
membrane surface possibly have effects on the permeate flux, retention, fouling tendency, efficiency, and even the cleaning
conditions in the application process. Zhang et al.102 investigated the microporous ceramic membrane for the rejection of organic
dyes. Due to the coating of nano-Y2O3, the ceramic membrane surface changed from negatively charged to positively charged. As a
result, such positively charged ceramic membranes exhibited high rejection to Titan Yellow and eriochrome black T, which were
both small molecule dyes with negative charges in water.
Surface properties of porous ceramic membranes, such as surface wettability and surface charge, also play important roles in the
oil-water filtration process.103,104 Lu et al.104 deposited ceramic UF membranes with a number of metal oxides, including TiO2,
Fe2O3, MnO2, CuO, and CeO2. Consequently, ceramic membranes with different hydrophilicities were obtained. It was found that
the differences in hydrophilicity lead to different adhesions of membrane surfaces toward oil droplets. Highly hydrophilic Fe2O3
showed the lowest fouling tendency. To improve the steady flux of ceramic membrane in the oily emulsions systems, surface
modification by doping have also been investigated.105,106 The TiO2–doped Al2O3 composite MF membranes with an average
pore size of 0.2 mm have been prepared by the solid-state sintering method. The doping of TiO2 into Al2O3 results in a stronger
negative surface charge and a shift of IEP toward lower pH. Moreover, the Al2O3–TiO2 composite membrane has a lower initial
contact angle and a more rapid rate of decrease in contact angle than the Al2O3 membrane (Fig. 11). Consequently, a higher and
 Ceramic Membranes 283

Fig. 11 Time dependence of the contact angle for Al2O3 and Al2O3–TiO2 membrane.

Fig. 12 Variation of the permeate flux for Al2O3 and Al2O3–TiO2 membranes versus filtration time.

more stable flux was observed through the Al2O3–TiO2 composite membrane (Fig. 12), owing to more hydrophilic groups, to
reduce the adsorption of oil droplets. Moreover, the composite membrane had the same type of charge (i.e., a negative charge)
with oil droplets, which benefits the control of membrane fouling during filtration of oily emulsion wastewater.
On the other hand, surface hydrophobic modification of ceramic membranes was accomplished with a self-assembled
monolayer of organosilane.107,108 The time dependence of the variation in contact angles for the ZrO2 membrane and the
modified ZrO2 membrane are shown in Fig. 13. The contact angle of the original ZrO2 membrane is less than 17 degrees and
decreased as time passes. However, the modified ZrO2 membrane has a much larger contact angle of about 134 degrees, and the
contact angle stays stable. The result proves that this modification can endow the ZrO2 membrane surface with much better
hydrophobicity, which means a better fouling resistance in the oil system. As shown in Fig. 14, the water rejection of the modified
membranes was always greater than that of the unmodified membranes. For the unmodified membranes, the water rejection was
around 88%, and there was a slight decrease when the TMP increased from 0.04 to 0.13 MPa, while for the modified membranes,
the water rejection was around 98%, and there was no clear difference between the water rejections when the TMP increased.

1.11.3.3 Effects of Solution Properties

The natural properties of the solution include viscosity, pH, ionic strength, and electrolytes. These properties directly affect
the surface properties of contacted membranes. Meanwhile, the behaviors of particles or macromolecule solutes, which would be
separated from solvents by membranes, may also be influenced by changing the solution properties. Consequently, the membrane
performance changed due to changes in solution properties.
In general, the changes in pH value and ionic strength will affect the behavior of solution. For example, the colloidal particles
would precipitate and the sol system would be unstable when the pH value of solutions was around the IEP. In addition, the
proteins also show minimum solvability at their IEP and tend to absorb on the surface of membranes. As a result, the membrane
284 Ceramic Membranes

Fig. 13 Time dependence of contact angles for the ZrO2 membranes and the modified ZrO2 membranes.

Fig. 14 Comparison of water rejections between the ZrO2 membranes and the modified ZrO2 membranes in the filtration of water-in-kerosene
emulsions at different TMP values.

systems always performed at minimum permeability during the separation of proteins, microorganisms, and macromolecules
when the pH value of solutions was around the IEP. However, for the systems with inorganic particles, the permeability of
membranes increased when the pH value of solutions was around the IEP due to aggregation of inorganic particles. In addition,
the change of pH values in solution would also affect the surface charge of membranes. In general, ceramic membranes
are negatively charged in neutral aqueous solution.109 Therefore, the selectivity of proteins for mixtures could be improved
dramatically by adjusting the zeta potential of membranes.110–113 Meanwhile, the change in zeta potential of membranes could
also affect electroviscous effects when fluid passes through membrane pores, consequently affecting the permeability of
membranes.114,115
Inorganic ions in the solution also have a significant impact on membrane separation performance, especially in the treatment
of protein systems. On the one hand, some inorganic salt composites could directly absorb and deposit on the membrane surface
and in membrane pores, or increase the absorbance of protein onto the membrane surface. On the other hand, inorganic salts
affect the ionic strength, which have strong effects on the solvability and structure of proteins and even the density of protein-
fouling cakes. As a result, these changes caused by inorganic ions significantly affect the permeability of ceramic membranes.
Alventosa et al.116 investigated the UF of model solutions containing NaCl and a reactive dye using ceramic membranes.
Higher flux decline and total resistance, together with lower color removal, were observed when adding NaCl to the solution,
which affected the repulsion–attraction phenomenon by changing the membrane charge.
Yoon et al.117 studied the fouling mechanism of inorganic membranes in the system of the membrane-coupled anaerobic
bioreactor (MCAB) by using alcohol distillery wastewater as a digester feed. The results demonstrated that the fouling mechanism
of inorganic membranes was significantly different from that of conventional membrane filtration processes. The inorganic
 Ceramic Membranes 285

precipitate, struvite (MgNH4PO4  6H2O), rather than anaerobic microbial floes, was found to be the main foulant. In this process,
the struvite appears to be precipitated not only on the membrane surface, but also inside the membrane pores, and it was not
easily removed by general physical cleaning. Therefore, chemical cleaning by acid was still necessary to recover the membrane flux.
Hoogland et al.118 studied the effect of pH on the permeability of alumina membrane, for both pure water and the filtration of
mineral slurries. The maximum pure water flux was found in a pH value below 2, which equaled the IEP value of membrane. In
contrast, the membrane resistance was greatest at high pH, which was far from the IEP. This could be explained by an electro-
osmosis phenomenon at high pH. However, higher permeabilities for alumina membranes were obtained at both low and high
pH in the filtration of mineral slurries (silica particles). This is because at high pH, both the membrane and the particles exhibit
negative charges, which lead to repulsive forces at the membrane-solution interface and depolarization of the membrane.
Moosemiller et al.119 found that both the supported g-Al2O3 and TiO2 ceramic membranes showed maximum permeate flux
for particle-free electrolyte solution when the pH value was around the IEP. Nazzal et al.120 studied the effects of pH and ionic
strength on the performance of ceramic MF membranes. However, the maximum permeate flux was not found around the IEP.
Meanwhile, the flux decreased as ionic strength increased. The results suggested that counter-ions exert an influence on the
permeate flux at distances substantially greater than a Debye length from the pore walls. Huisman et al.121 found that the increases
in water fluxes of 2–8% for both MF and UF membranes were caused by the increasing of salt concentration from 30 mM to 0.1 M.
These results were explained by electroviscous effects: increased salt concentrations led to lower zeta-potentials and thinner double
layers, offering less resistance to water passage.
Kwon et al.122 studied the effect of particle size and ionic strength of the feed suspension on critical flux. It was demonstrated
that the ionic strength of suspension had a significant effect on the critical flux. For an ionic strength less than 1  101.5 M, there
was a decrease in the critical flux. This could be due to the dense layer of deposit, which is the result of a less diffuse layer thickness
of particles. Above this ionic strength, a significant increase in critical flux was noticed, which may be due to the aggregation of
particles.
Chevereau et al.123 studied the effect of the presence of single salts in solution on the behavior of UF TiO2 membranes. It was
found that the IEP of TiO2 remains stable in the presence of NaCl, while it shifts to a higher pH value in the presence of MgCl2. At
native pH (close to 7), the adsorption of Mg2 þ on TiO2 surface leads to an inversion of the surface charge, as well as to an
important increase of the surface charge density.
Elzo et al.101 studied the effects of pH, ionic strength, and salts on the crossflow MF performance of 0.5 mm silica particles
using an inorganic membrane of 0.2 mm pore size. High permeate fluxes are obtained at high pH and low salt concentrations. In
contrast, low filtration fluxes are measured to have a high salt concentration, low pH, and a CaCl2 electrolyte. These results were
explained by the repulsion between the silica particles, which could be affected by the conditions.
Chiu124 studied the influence of time or aging on the electrokinetic characteristics of ceramic MF and their subsequent impact
on the MF performance of whey suspension. It was found that the zeta potential of the membrane increases as the aging period
increases, and then plateaus toward the end of the aging period. Consequently, a higher permeate flux was found over the aging
period.
Huisman et al.125 investigated the critical flux of MF while filtering a suspension of silica particles. It was found that neither the
zeta potential of the membrane nor the zeta potential of the particles influenced the observed critical flux. The critical flux
increased linearly with the wall shear stress and decreased with increasing particle concentration. Two different particle transport
mechanisms, particle rolling (the torque-balance model) and shear-induced diffusion, were applied for the expression of
critical flux.
Due to the complexity of the solution system, there is still no uniform law for the effect of solution properties on the
performance of membrane processes. There are even some conflicting results. Therefore, further studies should be carried out
focusing on these issues.

1.11.3.4 Effect of Operation Parameters

Optimization of operating conditions is an important aspect of the study of membrane applications. Different operating con-
ditions (eg. pressure difference across the membrane, the membrane surface flow rate and operation temperature) have significant
effects on the membrane separation performance. All membrane applications should work in the optimized operating conditions.
The critical values of permeate flux were concerned by many researchers. Wang et al.126 investigated the feasibility of treating
waste-rolling emulsion using zirconia MF membranes. It was found that the flux increases with the increase of TMP if it is less than
0.2 MPa. The flux may decline with the increase of TMP if it is greater than 0.2 MPa. In other words, a critical TMP was observed to
obtain maximum permeate flux, which was also observed by Zhao et al.127 by using Al2O3 MF membranes to recover fine TiO2
particles in acid waste streams. Alpatova et al.128 investigated the performance of ceramic UF membranes for the removal of
inorganic and organic compounds from oil-sand-process-affected water. While permeate flux increased with increasing TMP from
1.4 to 3.5 bar, a greater permeate flux decline was observed at a TMP of 3.5 bar due to increased accumulation of foulants at the
membrane surface.
In addition, Cui et al.129 studied the effect of crossflow velocity on the critical flux of ceramic membrane filtration as a
pretreatment for seawater desalination. The results demonstrated that crossflow velocity has no significant effect on the critical flux
when it varies in the laminar flow region only or in the turbulent flow region only, but the effect is distinct when the crossflow
velocity varies in the transition region. They concluded that the critical flux at a low crossflow velocity can be evaluated by
286 Ceramic Membranes

measuring it at a high crossflow velocity for a laminar flow. Consequently, they obtained the critical flux at a crossflow velocity of
0.015 m/s, which is slightly less than 185.4 L/(m2 h).
In general, the viscosity of solution decreases with the increase of temperature. Meanwhile, the solubility and mass transfer
coefficient increase with temperature. These events could accelerate the diffusion of solutes from the membrane surface to the body
of fluid, so as to reduce the thickness of the concentration polarization layer. As a result, the membrane permeate flux and
separation efficiency improved. For a dilute solution, the relationship between the permeate flux and temperature could be
predicted by the relationship between viscosity and temperature. Chang et al.130 found that the permeate flux almost doubled over
its original value when the temperature increased from 201C to 601C in the separation of a stable oil-in-water emulsion. Da et al.54
studied the effect of temperature on the performance of ceramic NF membranes in the treatment of dye wastewater. The flux value
was about 44 L/(m2 h) at 301C and increased to 84 L/(m2 h) at 601C. However, higher temperatures inescapably caused a decrease
in solute retention. This observation was explained by two reasons: (i) higher temperatures accelerated the fluorescent brightener
diffusion through the membrane; and (ii) the layer of adsorbed water molecules on the pore walls would be thinner at higher
temperatures, decreasing the hydrodynamic drag forces inside the pores as the effective pore diameter increased with the
temperature.

1.11.4 Application and Case Study of Ceramic Membrane

The unique thermal stability, chemical resistance, and mechanical properties of ceramic membranes give them significant
advantages and make them suitable for application in hash environments. Typical applications of ceramic membranes include
biopharmaceuticals, chemicals, electronics, food and beverage, and effluents. In this article, five industrial applications, including
ceramic membrane reactors, oily wastewater treatments, biological fermentation treatments, plant extracts, and brine refinements,
are introduced.

1.11.4.1 Ceramic Membrane Reactors

More than 70% of all existing industrial scale processes rely on catalysis. More than 99% of the world gasoline production
comes from catalytic cracking of oil fractions and other catalytic processes. Among the various catalysis methods, heterogeneous
catalysts with ultrafine or nanoparticle sizes exhibit very high activity and selectivity for numerous reactions.131 However,
separation of catalysis from the reaction slurry has retarded the large-scale promotion of heterogeneous catalysts with ultrafine or
nanoparticle sizes.
The membrane reactor has emerged as a reasonable alternative, where membrane separation processes were carried out to
remove products of the reaction, retain catalysts, or add reactants. What is most important is that membrane reactors could make
the reaction process continuous by retaining the catalysts in situ. In addition, membrane reactor systems have characteristics such
as being easy to operate, control, and scale up. However, due to the harsh conditions in the reaction processes, tough materials are
usually necessary for membrane to withstand such conditions. Hence, ceramic membranes with high thermal, chemical, and
mechanical resistance might be the most promising type.132,133
The synergistic effects between membrane separation and the heterogeneous catalytic reaction are the main factors affecting
the performance and long-term operational stability of porous ceramic membrane reactors. For membrane separation processes,
the efficiency relies strongly on the microstructure of the used membrane in relation to the application systems. Consistently, the
control of the porous membrane structure is of special importance in the development of stable membrane reactors,133 the
methods of which have been discussed earlier in this article. To enhance the efficiency of the membrane reactors, the con-
siderations of optimization are usually significant for ceramic membrane reactors, such as synergistic control of separation and
reaction, optimization of process parameters, and the establishment of a mathematical model. Consequently, almost 100%
catalyst retention has been achieved by using ceramic membrane reactors.134,135
Moreover, ceramic membranes can be used as a distributor during the addition of reactants. Jiang et al.136 developed a dual-
membrane reactor, in which one tubular, porous ceramic membrane was employed as a reactant distributor to control the supply
of reactant, and a second tubular, porous ceramic membrane was employed as a membrane separator to remove catalysts from the
products in situ. The results demonstrated that both the conversion of raw material and the selectivity of target compound
improved.
Membrane fouling is an important consideration, as it affects cleaning requirements and operating costs. For heterogeneous
catalysis, the physicochemical interactions between the reaction suspension and ceramic membranes are the main factors of
membrane fouling in porous ceramic membrane reactors. However, so far, few studies have been carried out to develop mem-
brane-fouling models for the systems of membrane reactors.137,138 It was likely due to the nature and extent of fouling in
membrane reactors is strongly influenced by many factors: (1) reaction suspension characteristics, including dissolved matter,
particle size and particle size distribution; (2) membrane characteristics, such as pore size, porosity, roughness, and hydrophilicity/
hydrophobicity; and (3) operating conditions related to crossflow velocity and TMP.139,140
In previous studies, two examples were examined to explain the fouling mechanisms of porous ceramic membrane reactors
for heterogeneous catalysis, involving ultrafine TS-1 catalysts and nanosized nickel catalysts.134,141,142 Based on this research, the
membrane-fouling problem was well controlled by optimizing membrane microstructures and operation parameters, and the
 Ceramic Membranes 287

Fig. 15 The industrialization apparatus of ceramic membrane reactors. Photo courtesy of Jiuwu.

ceramic membrane reactor system could be continually operated for more than 6 months without membrane cleaning. Conse-
quently, some industrial applications of ceramic membrane reactors have ever been realized. Fig. 15 shows an industrialization
apparatus of ceramic membrane reactors. Next, two industrial application examples for heterogeneous catalysis are presented:
cyclohexanone oxime production from the ammoxidation of cyclohexanone over an ultrafine TS-1 catalyst; and p-aminophenol
production from p-nitrophenol hydrogenation over nanosized nickel.

1.11.4.1.1 Example: Cyclohexanone oxime production

Cyclohexanone oxime is the intermediate used to produce caprolactam, of which 90% is obtained by the rearrangement
of cyclohexanone oxime. A novel method of catalytic direct ammoxidation of cyclohexanone with NH3/H2O2 to the oxime using
TS-1 as the catalyst was developed. However, the separation of the ultrafine catalyst became another problem to solve in practical
applications of this process. A sidestream membrane reactor was developed for the cyclohexanone oxime by Jiangsu Jiuwu High-
Tech Co. Ltd., with a processing capacity of about 350,000 t/a. The recovered catalysts were recycled continuously back to the
reactor and reused in the next reaction. The cleaning of fouled membranes can be achieved by a sequential use of strongly basic
and acidic solutions. The catalyst in the permeation was as low as 1 mg/L. Meanwhile, both the conversion and selectivity of this
reaction were all higher than 99.5%.

1.11.4.1.2 Example: p-Aminophenol production

This section discusses p-aminophenol, an important intermediate that is mainly used in the rubber industry, photosensitive
material industry, and other industries. Because of the growing demand for p-aminophenol, direct catalytic hydrogenation of
p-nitrophenol to p-aminophenol is becoming more and more important because it could be an efficient and greener route.143
Based on the research on many aspects of the system, such as membrane reactor design, membrane fouling control, and mem-
brane regeneration, Jiangsu Jiuwu High-Tech Co. Ltd. has successfully employed ceramic membrane reactors for the production of
p-aminophenol and constructed a project of 5000 t/a for p-aminophenol production. In this process, ceramic membrane units
were used to remove nanosized catalysts from the reaction slurry. In this project, the yield of p-aminophenol in unit reactor
volume significantly increased, and the product cost per unit markedly decreased.

1.11.4.2 Oily Wastewater Treatments

Large quantities of oily wastewater were produced during the process where the oil was produced or applied. If the effluent is
discharged untreated, it can certainly pose significant threats to the environment and human health due to the hazardous nature
of its oil contents, especially mineral oil.144 Therefore, to remove oil from wastewater is an important target of pollution control in
many industries.
The conventional treatment techniques, which are employed in oil-water separation and treatment, include gravity settling,
chemical emulsion breaking, coagulation and flocculation, and dissolved air flotation methods.145,146 The membrane separation
technique has attracted a good deal of interest in the treatment of oily wastewater due to its high efficiency and precision in
separation, especially when the diameter of oil droplets is less than 20 mm in the o/w emulsion.
Hua et al.147 studied the treatment of vegetable oil wastewater by using a ceramic membrane with an average pore size of
50 nm. The influence of parameters on the separation behaviors was investigated, and it was found the total organic carbon
removal rate was higher than 92.4% under all experimental conditions. Yeom et al.148 fabricated crack-free, alumina-coated, clay
diatomite composite membranes for the treatment of oily wastewater. In their study, a simple burnout process was subjected to
the used membranes in air to recover the steady-state flux, and oil rejection rate. The recycled membranes still showed an
exceptionally high oil rejection rate (99.9%) with a feed oil concentration of 600 mg/L at an applied pressure of 101 kPa.
288 Ceramic Membranes

However, the high purity requirements for discharge are hard to reach by using a single membrane process due to the much
stricter requirements of resource conservation and environmental protection. Therefore, the integrated process has been considered
an ideal method for the treatment of oily wastewater.149 Abbasi et al.150 investigated the treatment of oily wastewater by using a
hybrid coagulation-MF process. Four coagulants (i.e., ferrous chloride, ferrous sulfate, aluminum chloride, and aluminum sulfate)
with the addition of calcium hydroxide were evaluated in the coagulation-MF hybrid process at different concentrations. In
addition, they also studied the hybrid MF-powdered activated carbon process.151 In the two hybrid processes, Hermia's models
were applied to study the fouling mechanisms of mullite-alumina ceramic membranes in the treatment of oily wastewater. It was
found that the cake filtration model can be applied to predict permeation flux decline. Nidal et al.152 achieved the treatment of
metal working fluids and assessed the feasibility of permeate reuse by combining UF and NF membrane processes. Maria et al.153
investigated the feasibility of bilge water treatment in the integrated UF and reverse osmosis system. It was found that the permeate
from the first stage of bilge water treatment had oil content below 10 mg/L and was free of suspended solids, and the obtained
reverse osmosis permeate was free of oil. Yu et al.154 investigated the possibility and feasibility of vegetable oily wastewater
treatment with a process that integrated MF and reverse osmosis. The operating conditions for permeate flux and rejection were
optimized in the MF process. The operating temperature should be kept in the range of 30–401C, while the proper crossflow
velocity and the TMP should be kept at about 4 m/s, and 0.15 MPa, respectively. The results demonstrated that the two stage
membranes combining MF and reverse osmosis constitute the most feasible and reliable process for vegetable oily wastewater
treatment.

1.11.4.2.1 Example: cold-rolling emulsion wastewater treatment

Cold-rolling emulsion wastewater produced during cold rolling in the steel industry is one typically oily wastewater, and it is
difficult to handle. Yang et al.155 investigated the effects of membrane material and pore size on the performance of ceramic
membranes in the treatment of an oil-water emulsion. It was found that the ZrO2/a-Al2O3 MF membrane with an average pore size
of about 0.2 mm was the preferred membrane. Zhang et al.156 investigated the treatment of a cool-rolling emulsion with inorganic
ceramic membranes. It was also demonstrated that ZrO2 membranes have advantages over Al2O3 membranes. The flux through
ZrO2 membrane was greater than that of Al2O3 membranes. Zhang et al.106 studied the effects of the membrane material
characteristics and solution pH on permeate flux for oily wastewater treatment. The surface properties of ceramic membrane were
further optimized. Based on research, Jiangsu Jiuwu High-Tech Co. Ltd. successfully employed ceramic membranes in the treat-
ment of rolling emulsion wastewater. Up to now, the ceramic membrane technique has been widely applied in most iron and steel
plants to treat cool-rolling emulsion. In China, a million tons of rolling emulsion wastewater were treated every year. Fig. 16 shows
a ceramic membrane set for the treatment of oily wastewater.

1.11.4.3 Biological Fermentation Treatments

Biological fermentation is the main process used in the fields of biologic medical production and food processing. Important
issues of concern are the cost of production (low-volume, high-purity products are desirable) and microbial contamination (by
bacteria, viruses, and mycoplasma). Membrane filtration is a major separation technique in the biological fermentation industry.

1.11.4.3.1 Example: Lactic acid production

Lactic acid (HL) is a commodity chemical produced by fermentation, which is widely utilized in the food, chemical, and
pharmaceutical fields.157 HL is typically produced via fermentation, where a suitable pH value for microbe growth must be
maintained at around 5–6 by the addition of Ca(OH)2 to ensure the productivity of HL. Subsequently, a filtration method
was applied to separate calcium lactate from the microbial cells and followed by acidification with sulfuric acid to obtain HL.

Fig. 16 Ceramic membrane sets for the treatment of oily wastewater. Photo courtesy of Jiuwu.
 Ceramic Membranes 289

To improve the purity of HL, activated-carbon adsorption and ion exchange processes were carried out. Finally, HL was con-
centrated by removing the water through the multieffect evaporation method. In this traditional production process, a mass of
byproduct (gypsum) was produced at a high rate (1 ton per ton of HL).158
To reduce the industry waste residue of gypsum, NaOH or NH4OH is added as a neutralizer to adjust the pH and avoid the
inhibition of cell growth. However, the recovery of high-purity product from the fermentation broth becomes a critical technical
problem by using the traditional production process. Therefore, numerous studies have been carried out to propose new
separation techniques for HL recovery from the fermentation broth, such as extraction,159 adsorption,160 and membrane
separation.161 Over the past two decades, membrane techniques have been extensively investigated because they can be integrated
easily with conventional fermenters.162 The crossflow ceramic MF membrane can retain cells and proteins more efficiently than
plate-and-frame filtration in the conventional process.163
Pal et al.164 reviewed recent developments in the production of monomer-grade HL by using membrane-based processes. They
suggested an integrated production scheme using MF and NF membranes. However, this continuous fermentative HL production
scheme with two-stage integration of membranes cannot ensure the separation of microbial cells for their recycling without
additional provision of cells. To overcome this problem, a promising method is to combine the membrane process with batch
fermentation. Therefore, efforts should be directed toward developing a membrane-integrated process for HL production.
Wang et al.165 investigated the feasibility of an integrated ceramic MF/UF, NF, and bipolar membrane electrodialysis (BMED)
process followed by fermentation broth neutralization with sodium hydroxide. The scheme of the setup is presented in Fig. 17. A
ceramic membrane was used to remove the cells. The permeate from the ceramic membrane was then introduced to an NF process
in which the sugar, protein, and bivalent salts were separated from sodium lactate. The BMED process was applied to the
transformation of sodium lactate to HL and sodium hydroxide. Sodium hydroxide was circulated to the fermentation system to
adjust the pH. Based on this study, a pilot-scale HL-producing process was successfully carried out by Jiangsu Jiuwu High-Tech Co.
Ltd., with an annual production of 1000 t/a.

1.11.4.3.2 Example: Cephalosprin production

Since cephalosporins were discovered in 1945 by the Italian pharmacologist Giuseppe Brotzu, they have been used to treat a wide
range of bacterial infections due to their high efficiency, broad-spectrum antibacterial activity, low toxicity, and resistance to
enzymes.
During the production of cephalosporins, hyphas, suspended solids, and macromolecular proteins first must be removed from
fermentation broth to obtain a clarified solution of cephalosporins. Subsequently, the solution was subjected to refining and
drying processes to obtain cephalosporin powders. Due to the thermosensitivity of cephalosporins, the refining process must be
operated at a low temperature (0–51C), which is out of the suitable range for polymer membranes. Therefore, ceramic membrane
systems become a promising alternative to obtain a clarified solution of cephalosporins due to their excellent thermal stability.
Fig. 18 is an industrial ceramic membrane separation system for the production of cephalosporins, which, with a total membrane
area of 422 m2, was developed by Jiangsu Jiuwu High-Tech Co. Ltd. In this process, the yield of cephalosporins could exceed 92%.
So far, more than 400 industrial ceramic membrane systems have been constructed for biological fermentation treatments in
China. It has been found that ceramic membrane systems showed many advantages over plate and frame filter systems. A detailed
list of items is shown in Table 2.

1.11.4.4 Plant Extraction

Plant extraction separates and purifies effective constituents from plants by using a physical, chemical, or biological method. The
extracted effective constituents include glycoside, acid, polyphenols, polysaccharide, terpenoids, flavonoids, and alkaloids. The
separation and purification of active ingredients in plants are the key technology connected to plant extracts, which directly affect

Fig. 17 Scheme of the integrated membrane process setup: (1) fermenter; (2) UF feed tank; (3) ceramic membrane module; (4) NF feed tank; (5)
NF membrane module; (6) acid tank; (7) BMED stack; (8) caustic tank; (9–13) pump, (P1–P4) pressure gauges, (T1–T4) temperature gauges,
(F1–F4) flow meters.
290 Ceramic Membranes

Fig. 18 Industrial ceramic membrane separation system for the production of cephalosporins. Photo courtesy of Jiuwu.

Table 2 Comparison of ceramic membrane system with plate and frame filter systems in the
treatments of biological fermentation

Items Plate and frame filter Ceramic membrane

Flux (L/(m2 h)) 40–80 150–200

Removal of mycelium (%) 70–85 99
Removal of turbidity (%) 80–85 99
Membrane cleaning Hard Easy
Stability Poor Good
Equipment investment Low High
Lifetime/year 43 3–5

the benefits of enterprise. Generally, the active ingredients in plant extraction solutions come in low levels. Meanwhile, the
extraction solutions often contain impurities, including starches, proteins, carbohydrates, pigments, and pectin. The molecular
weight of active ingredients usually ranges from several hundred thousand to millions of Daltons.
Nowadays, various types of membrane separation technology has been widely used in the separation and purification,
decolorization, desalination, and concentration of plant extracts.166,167 In these processes, a complete set of integration membrane
technologies was usually applied to remove the macromolecular substances effectively. Many advantages have been shown,
including avoiding the backmixing phenomenon, improving transmittance, decreasing the salt content in the product, relieving
the burden of next stage (decolorization and desalination), and even the replacement of solvents.

1.11.4.4.1 Example: Anthocyanin production

Membrane technology is often used for the purification and concentration of food extracts instead of traditional methods, such as
centrifugal separation and frame filtering. Anthocyanin that belongs to flavonoids of phenolic compounds is a water-soluble
pigment that can eliminate free radicals, whose capacity was 20 times that of vitamin C and 50 times that of vitamin E, and which
was widely used in food, pharmaceuticals, cosmetics, and other fields. In the process of extracting anthocyanin, ceramic membrane
serves as a refined filtration technology to remove macromolecular impurities from anthocyanin extract and reduce the load of the
next resin technology. Subsequently, anthocyanin extract is concentrated by NF membranes, which can remove small molecules
and inorganic salts and return ethanol. A membrane system for anthocyanin production by integrated membrane technology is
presented in Fig. 19. Integrated membrane technology applied for the production of anthocyanin can save 35% of the content of
resin and enhance 45% of the content of anthocyanin. Fig. 20 shows ceramic membrane equipment for an anthocyanin project.

1.11.4.4.2 Example: Liqueur production

In pharmaceutical extracts, active ingredients from more kinds of Chinese medicines are extracted and separated using integrated
membrane technology by Jiangsu Jiuwu High-Tech Co. Ltd., and then blended into the wine, which form a “digital extraction
technology” of tonic wine. First, nonmedicinal ingredients such as suspended solids, macromolecular protein, and polysaccharide
are removed by using inorganic ceramic UF membrane technology, which makes pharmacological substances to leak from
membrane pores. Second, the permeate of UF membranes is concentrated by NF membranes. On the purpose of purifying active
ingredients, the use of membrane technology can save more than 80% of the cost instead of using traditional Chinese medicine
extract technology. Fig. 21 shows ceramic membrane equipment for a liqueur project.
 Ceramic Membranes 291

Fig. 19 Flow sheet of the production process of anthocyanin by integrated membrane technology.

Fig. 20 Ceramic membrane equipment for an anthocyanin project. Photo courtesy of Jiuwu.

Fig. 21 Ceramic membrane equipment for a liqueur project. Photo courtesy of Jiuwu.

1.11.4.5 Brine Refinements

The chloralkali process is an industrial process for the electrolysis of NaCl, of which the principal products are chlorine and
sodium hydroxide. The chloralkali process has been one of the largest electrochemical operations in the world. The purity of the
brine used in the process is one of the most important factors in the efficiency of the electrolyzers. Therefore, the brine refinement
is usually the first step. Some common impurities, including Mg2 þ , Ca2 þ , and SO4 22 , are generally contained in the saturated
brine, which can affect the performance of the electrolyzers. To remove these impurities from brine, the chemical precipitation is
commonly used in the purification process and followed by coagulation, flocculation, sedimentation, sand filtration, or cen-
trifugation to separate the precipitates. Recently, the membrane technique has been employed as a simple and efficient separation
method to further increase the purity of brine.168
However, fouling in membrane separation brings a lot of negative impacts on the membrane process, such as reducing
the permeate flux, increasing the operating pressure, decreasing the product quality, and ultimately shortening the membrane
life.169 In some membrane bioreactor processes, inorganic fouling occurring on MF membrane surfaces can be eliminated
by acid cleaning.170,171 In membrane processes for brine purification developed by W. L. Gore & Associates, Inc., and
Hyflux Ltd., polytetrafluoroethylene membranes fouled by inorganic salt sludge can be recovered by acid cleaning following
a backpulse. Gu et al.172 investigated the cleaning efficiency and kinetics of multichannel ceramic membranes fouled during
brine purification, and noted that mainly BaSO4 crystals were deposited on the membrane surface and caused the fouling of
membrane. To recover the membrane flux, a cleaning solution composed of diethylenetrinitrilopentaacetic acid (DTPA), oxalic
acid, and NaOH was developed to clean BaSO4 crystals on the membrane surface. The membranes could be completely recovered
292 Ceramic Membranes

with the cleaning solution. In addition, a dissolution kinetics model associated with both the concentration factor and
the temperature factor was developed, which fitted well with the experimental results. Based on this model, it was found that
the activation energy of BaSO4 dissolution using the cleaning solution consisted of DTPA, oxalic acid, and NaOH was less
than that of using pure DTPA solution, which means the compound solution provided a better cleaning performance than
pure DTPA solution. The flow sheet of the ceramic membrane system for brine purification running in a chloralkali plant is shown
in Fig. 22.
Until now, dozens of industrial devices for the brine refinements have been constructed and successfully been in
smooth operation in the chloralkali enterprises in China. Several industrial devices with different handling capacities are shown
in Figs. 23–25.

Fig. 22 Flow sheet of the ceramic membrane system for brine purification.

Fig. 23 Brine refinement device for the production of caustic soda (10,0000 t/a). Photo courtesy of Jiuwu.

Fig. 24 Brine refinement device for the production of caustic soda (15,0000 t/a). Photo courtesy of Jiuwu.
 Ceramic Membranes 293

Fig. 25 Brine refinement device for the production of caustic soda (30,0000 t/a). Photo courtesy of Jiuwu.

1.11.5 Final Remarks

The study of the fabrication and application of ceramic membranes has stretched over the past 40 years. The research and
development has been supported by the government and some large companies. Remarkable improvements have been performed
in the development of application-oriented structure design methods that achieve a breakthrough ceramic membrane preparation
technology and promote the industrialization of ceramic membrane products. Although many novel fabrication techniques have
been developed to improve the separation performance and decrease the cost of ceramic membranes, further research is required
to produce membranes with a wide range of unique microstructures and surface properties, which can be adapted for the
application and increase their market volumes.
The emphases of future development regarding ceramic membrane fabrication and application will cover the following aspects:
(1) to develop a low-cost, high-performance ceramic membrane element based on the application-oriented design process and
preparation method; (2) to develop ceramic NF membranes for molecular level separation from lab scale to industrial scale and
open the window to extend their application fields; (3) to enrich surface properties of the ceramic membranes to improve their
fouling resistance and extend their lifetime in practical applications; (4) to strengthen the basic theoretical study of the transport
mechanism in confined nanochannels of ceramic membranes.

Acknowledgments

The authors would like to acknowledge the financial support from the National High Technical Research Program of China
(2007aa030303, 2012AA03A606), the National Basic Research Program of China (2003CB61570700, 2009CB623400), the
National Nature Science Foundation of China (21490580, 91534108), the Project of Priority Academic Program Development
of Jiangsu Higher Education Institutions (PAPD), and the program for Changjiang Scholars and Innovative Research Team in
University (IRT13070). The authors would also like to thank Jiangsu Jiuwu Hi-Tech Corporation and Dr. Wenbo Peng for valuable
contributions on the information of industrial applications.

See also: 1.12 Microstructured Ceramic Hollow Fiber Membranes and Their Applications

References

1. Burggraaf, A. J.; Cot, L. Fundamentals of Inorganic Membrane Science and Technology. Elsevier 1996.
2. Rao, L. J. M. Handbook of Membrane Separations; Wiley, 2009.
3. Van Gestel, T.; Vandecasteele, C.; Buekenhoudt, A.; et al. Alumina and Titania Multilayer Membranes for Nanofiltration: Preparation, Characterization and Chemical
Stability. J. Membr. Sci. 2002, 207 (1), 73–89.
4. Leenaars, A. F. M.; Keizer, K.; Burggraaf, A. J. The Preparation and Characterization of Alumina Membranes With Ultrafine Pores. J. Mater. Sci. 1984, 19,
1077–1088.
5. Hilal, N.; Al-Zoubi, H.; Darwish, N. A.; et al. A Comprehensive Review of Nanofiltration Membranes: Treatment, Pretreatment, Modelling, and Atomic Force Microscopy.
Desalination 2004, 170 (3), 281–308.
6. Drioli, E.; Giorno, L. Comprehensive Membrane Science and Engineering; Elsevier, 2010.
7. Su, C. H.; Xu, Y. Q.; Zhang, W.; et al. Porous Ceramic Membrane With Superhydrophobic and Superoleophilic Surface for Reclaiming Oil From Oily Water. Appl. Surf.
Sci. 2012, 258 (7), 2319–2323.
294 Ceramic Membranes

8. Moritz, T.; Benfer, S.; Arki, P.; et al. Influence of the Surface Charge on the Permeate Flux in the Dead-End Filtration With Ceramic Membranes. Sep. Purif. Technol.
2001, 25 (1–3), 501–508.
9. Liu, S.; Zeng, Y.-P.; Jiang, D. Fabrication and Characterization of Cordierite-Bonded Porous SiC Ceramics. Ceram. Int. 2009, 35 (2), 597–602.
10. Velev, O. D.; Jede, T. A.; Lobo, R. F.; et al. Microstructured Porous Silica Obtained via Colloidal Crystal Templates. Chem. Mater. 1998, 10 (11), 3597–3602.
11. Ke, X.; Zhu, H.; Gao, X.; et al. High-Performance Ceramic Membranes With a Separation Layer of Metal Oxide Nanofibers. Adv. Mater. 2007, 19 (6), 785–790.
12. Das, N.; Maiti, H. S. Ceramic Membrane by Tape Casting and Sol–Gel Coating for Microfiltration and Ultrafiltration Application. J. Phys. Chem. Solids 2009, 70 (11),
1395–1400.
13. Yao, A.; Yu, B.; Yang, K.; et al. Fabrication and Properties of Mullite-Alumina Ceramic Support. Rare Metal Mater. Eng. 2005, 34, 255–258.
14. Dong, G.; Qi, H.; Xu, N. Effect of Active Carbon Doping on Structure and Property of Porous Alumina Support. J. Chin. Ceram. Soc. 2012, 40 (6), 844–850.
15. Gregorová, E.; Pabst, W. Porosity and Pore Size Control in Starch Consolidation Casting of Oxide Ceramics—Achievements and Problems. J. Eur. Ceram. Soc. 2007,
27 (2–3), 669–672.
16. She, J.; Deng, Z.; Daniel-Doni, J.; et al. Oxidation Bonding of Porous Silicon Carbide Ceramics. J. Mater. Sci. 2002, 37 (17), 3615–3622.
17. Collier, A. K.; Liu, W.; Wang, J. G.; et al. Alpha-Alumina Inorganic Membrane Support and Method of Making the Same. US patent 20110045971, 2011.
18. Yang, G. C. C.; Tsai, C.-M. Effects of Starch Addition on Characteristics of Tubular Porous Ceramic Membrane Substrates. Desalination 2008, 233 (1–3), 129–136.
19. Park, S. H.; Xia, Y. N. Macroporous Membranes With Highly Ordered and Three-Dimensionally Interconnected Spherical Pores. Adv. Mater. 1998, 10 (13), 1045–1048.
20. Sadakane, M.; Horiuchi, T.; Kato, N.; et al. Facile Preparation of Three-Dimensionally Ordered Macroporous Alumina, Iron Oxide, Chromium Oxide, Manganese Oxide, and
Their Mixed-Metal Oxides With High Porosity. Chem. Mater. 2007, 19 (23), 5779–5785.
21. Zhao, K.; Fan, Y.; Xu, N. Preparation of Three-Dimensionally Ordered Macroporous SiO2 Membranes With Controllable Pore Size. Chem. Lett. 2007, 36 (3), 464–465.
22. Zhao, K.; Fan, Y.; Wang, R.; et al. Preparation of Closed Macroporous Al2O3 Membranes With a Three-Dimensionally Ordered Structure. Chem. Lett. 2008, 37 (4),
420–421.
23. Xu, J.; Xiang, W.; Hu, F. Preparation of Monodisperse Polystyrene Spheres and Inorganic Porous Films. Rare Metal Mater. Eng. 2008, 37, 196–200.
24. Beck, J. S.; Vartuli, J. C.; Roth, W. J.; et al. A New Family of Mesoporous Molecular-Sieves Prepared With Liquid-Crystal Templates. J. Am. Chem. Soc. 1992,
114 (27), 10834–10843.
25. Ji, H.; Fan, Y.; Jin, W.; et al. Synthesis of Si-MCM-48 Membrane by Solvent Extraction of the Surfactant Template. J. Non-Cryst. Solids 2008, 354 (18), 2010–2016.
26. Xu, D. K.; Fan, Y. Q. Mesoporous Si-MCM-48 Membrane Prepared by Pore-Filling Method. Sci. China Technol. Sci. 2010, 53 (4), 1064–1068.
27. Kumar, P.; Ida, J.; Kim, S.; et al. Ordered Mesoporous Membranes: Effects of Support and Surfactant Removal Conditions on Membrane Quality. J. Membr. Sci. 2006,
279 (1–2), 539–547.
28. Zhang, J.; Li, W.; Meng, X.; et al. Synthesis of Mesoporous Silica Membranes Oriented by Self-Assembles of Surfactants. J. Membr. Sci. 2003, 222 (1–2), 219–224.
29. Choi, H.; Sofranko, A. C.; Dionysiou, D. D. Nanocrystalline TiO2 Photocatalytic Membranes With a Hierarchical Mesoporous Multilayer Structure: Synthesis,
Characterization, and Multifunction. Adv. Funct. Mater. 2006, 16 (8), 1067–1074.
30. Cao, X. P.; Jing, W. H.; Xing, W. H.; et al. Fabrication of a Visible-Light Response Mesoporous TiO2 Membrane With Superior Water Permeability via a Weak Alkaline
Sol–Gel Process. Chem. Commun. 2011, 47 (12), 3457–3459.
31. Lei, W.; Fan, Y. Effect of Sintering Temperature on Microfiltration Membrane Prepared With TiO2 Ceramic-Fiber. Membr. Sci. Technol. 2009, 29 (5), 54–57.
32. Ke, X.; Zheng, Z.; Liu, H.; et al. High-Flux Ceramic Membranes With a Nanomesh of Metal Oxide Nanofibers. J. Phys. Chem. B 2008, 112 (16), 5000–5006.
33. Ke, X.; Zheng, Z.; Zhu, H.; et al. Metal Oxide Nanofibres Membranes Assembled by Spin-Coating Method. Desalination 2009, 236 (1–3), 1–7.
34. Ke, X. B.; Huang, Y. M.; Dargaville, T. R.; et al. Modified Alumina Nanofiber Membranes for Protein Separation. Sep. Purif. Technol. 2013, 120, 239–244.
35. Fernando, J. A.; Chung, D. D. L. Improving an Alumina Fiber Filter Membrane for Hot Gas Filtration Using an Acid Phosphate Binder. J. Mater. Sci. 2001, 36 (21),
5079–5085.
36. Qiu, M.; Fan, S.; Cai, Y.; et al. Co-Sintering Synthesis of Bi-Layer Titania Ultrafiltration Membranes With Intermediate Layer of Sol-Coated Nanofibers. J. Membr. Sci.
2010, 365 (1–2), 225–231.
37. Pastila, P.; Helanti, V.; Nikkila, A. P.; et al. Environmental Effects on Microstructure and Strength of SiC-Based Hot Gas Filters. J. Eur. Ceram. Soc. 2001, 21 (9),
1261–1268.
38. Li, J.; Lin, H.; Li, J. Factors that Influence the Flexural Strength of SiC-Based Porous Ceramics Used for Hot Gas Filter Support. J. Eur. Ceram. Soc. 2011, 31 (5),
825–831.
39. Ding, S.; Zeng, Y.; Jiang, D. In-Situ Reaction Bonding of Porous SiC Ceramics. Mater. Charact. 2008, 59 (2), 140–143.
40. Vacassy, R.; Guizard, C.; Thoraval, V.; et al. Synthesis and Characterization of Microporous Zirconia Powders: Application in Nanofilters and Nanofiltration Characteristics.
J. Membr. Sci. 1997, 132 (1), 109–118.
41. Ju, X.; Huang, P.; Xu, N.; et al. Study of Factors Influencing Pore Size of Zirconia Ultrafiltration Membrane. J. Chem. Eng. Chin. Univ. 2000, 14 (2), 103–105.
42. Manjumol, K. A.; Shajesh, P.; Baiju, K. V.; et al. An ‘Eco-friendly’ All Aqueous Sol Gel Process for Multi Functional Ultrafiltration Membrane on Porous Tubular Alumina
Substrate. J. Membr. Sci. 2011, 375 (1–2), 134–140.
43. Fan, S.; Qiu, M.; Zhou, X.; et al. Preparation of Multichannel TiO2 Ultrafiltration Membrane and Its Application in Dyeing Wastewater. J. Nanjing Univ. Technol. 2010,
33 (1), 44–47.
44. Sakka, S. Handbook of Sol–Gel Science and Technology: Processing, Characterization and Applications; Kluwer Academic Publishers: Holland, 2005.
45. Tsuru, T.; Hironaka, D.; Yoshioka, T.; et al. Titania Membranes for Liquid Phase Separation: Effect of Surface Charge on Flux. Sep. Purif. Technol. 2001, 25 (1–3),
307–314.
46. Tsuru, T.; Ogawa, K.; Kanezashi, M.; et al. Permeation Characteristics of Electrolytes and Neutral Solutes Through Titania Nanofiltration Membranes at High Temperatures.
Langmuir 2010, 26 (13), 10897–10905.
47. Qi, H.; Li, S.; Jiang, X.; et al. Preparation and Ions Retention Properties of TiO2 Nanofiltration Membranes. J. Inorg. Mater. 2011, 26 (3), 305–310.
48. Van Gestel, T.; Sebold, D.; Kruidhof, H.; et al. ZrO2 and TiO2 Membranes for Nanofiltration and Pervaporation. J. Membr. Sci. 2008, 318 (1–2), 413–421.
49. Benfer, S.; Popp, U.; Richter, H.; et al. Development and Characterization of Ceramic Nanofiltration Membranes. Sep. Purif. Technol. 2001, 22–23, 231–237.
50. Spijksma, G. I.; Huiskes, C.; Benes, N. E.; et al. Microporous Zirconia-Titania Composite Membranes Derived From Diethanolamine-Modified Precursors. Adv. Mater.
2006, 18 (16), 2165–2168.
51. Aust, U.; Benfer, S.; Dietze, M.; et al. Development of Microporous Ceramic Membranes in the System TiO2/ZrO2. J. Membr. Sci. 2006, 281 (1–2), 463–471.
52. Cai, Y.; Wang, Y.; Chen, X.; et al. Modified Colloidal Sol–Gel Process for Fabrication of Titania Nanofiltration Membranes With Organic Additives. J. Membr. Sci. 2015,
476, 432–441.
53. Chen, X.; Zhang, W.; Lin, Y.; et al. Preparation of High-Flux g-Alumina Nanofiltration Membranes by Using a Modified Sol–Gel Method. Microporous Mesoporous Mater.
2015, 214, 195–203.
54. Da, X.; Wen, J.; Lu, Y.; et al. An Aqueous Sol–Gel Process for the Fabrication of High-Flux YSZ Nanofiltration Membranes as Applied to the Nanofiltration of Dye
Wastewater. Sep. Purif. Technol. 2015, 152, 37–45.
55. Da, X.; Chen, X.; Sun, B.; et al. Preparation of Zirconia Nanofiltration Membranes Through an Aqueous Sol–Gel Process Modified by Glycerol for the Treatment of
Wastewater With High Salinity. J. Membr. Sci. 2016, 504, 29–39.
56. Lu, Y.; Chen, T.; Chen, X.; et al. Fabrication of TiO2-Doped ZrO2 Nanofiltration Membranes by Using a Modified Colloidal Sol–Gel Process and Its Application in
Simulative Radioactive Effluent. J. Membr. Sci. 2016, 514, 476–486.
 Ceramic Membranes 295

57. Tsuru, T.; Wada, S.; Izumi, S.; et al. Silica–Zirconia Membranes for Nanofiltration. J. Membr. Sci. 1998, 149 (1), 127–135.
58. Lin, Y. S. A Theoretical Analysis on Pore Size Change of Porous Ceramic Membranes After Modification. J. Membr. Sci. 1993, 79 (1), 55–64.
59. Xomeritakis, G.; Lin, Y. S. Chemical-Vapor-Deposition of Solid Oxides in Porous-Media for Ceramic Membrane Preparation—Comparison of Experimental Results With
Semianalytical Solutions. Ind. Eng. Chem. Res. 1994, 33 (11), 2607–2617.
60. Labropoulos, A. I.; Romanos, G. E.; Karanikolos, G. N.; et al. Comparative Study of the Rate and Locality of Silica Deposition During the CVD Treatment of Porous
Membranes With TEOS and TMOS. Microporous Mesoporous Mater. 2009, 120 (1–2), 177–185.
61. Lin, C. L.; Flowers, D. L.; Liu, P. K. T. Characterization of Ceramic Membranes. 2. Modified Commercial Membranes With Pore-Size Under 40 Angstrom. J. Membr. Sci.
1994, 92 (1), 45–58.
62. Fernandes, N. E.; Gavalas, G. R. Gas Transport in Porous Vycor Glass Subjected to Gradual Pore Narrowing. Chem. Eng. Sci. 1998, 53 (5), 1049–1058.
63. Sarrade, S.; Guizard, C.; Rios, G. M. New Applications of Supercritical Fluids and Supercritical Fluids Processes in Separation. Sep. Purif. Technol. 2003, 32 (1–3),
57–63.
64. Wang, Z.; Dong, J.; Xu, N.; et al. Pore Modification Using the Supercritical Solution Infiltration Method. AICHE J. 1997, 43 (9), 2359–2367.
65. Brasseur-Tilmant, J.; Chhor, K.; Jestin, P.; et al. Ceramic Membrane Elaboration Using Supercritical Fluid. Mater. Res. Bull. 1999, 34 (12–13), 2013–2025.
66. Tatsuda, N.; Fukushima, Y.; Wakayama, H. Penetration of Titanium Tetraisopropoxide Into Mesoporous Silica Using Supercritical Carbon Dioxide. Chem. Mater. 2004,
16 (9), 1799–1805.
67. Li, F.; Yang, Y.; Fan, Y.; et al. Modification of Ceramic Membranes for Pore Structure Tailoring: The Atomic Layer Deposition Route. J. Membr. Sci. 2012, 397, 17–23.
68. Van Gestel, T.; Van der Bruggen, B.; Buekenhoudt, A.; et al. Surface Modification of g-Al2O3/TiO2 Multilayer Membranes for Applications in Non-Polar Organic Solvents.
J. Membr. Sci. 2003, 224 (1–2), 3–10.
69. Singh, R. P.; Way, J. D.; Dec, S. F. Silane Modified Inorganic Membranes: Effects of Silane Surface Structure. J. Membr. Sci. 2005, 259 (1–2), 34–46.
70. Sah, A.; Castricum, H. L.; Bliek, A.; et al. Hydrophobic Modification of g-Alumina Membranes With Organochlorosilanes. J. Membr. Sci. 2004, 243 (1–2), 125–132.
71. Leger, C.; Lira, H. D. L.; Paterson, R. Preparation and Properties of Surface Modified Ceramic Membranes. Part II. Gas and Liquid Permeabilities of 5 nm Alumina
Membranes Modified by a Monolayer of Bound Polydimethylsiloxane (PDMS) Silicone Oil. J. Membr. Sci. 1996, 120 (1), 135–146.
72. Faibish, R. S.; Cohen, Y. Fouling-Resistant Ceramic-Supported Polymer Membranes for Ultrafiltration of Oil-in-Water Microemulsions. J. Membr. Sci. 2001, 185 (2),
129–143.
73. Rovira-Bru, M.; Giralt, F.; Cohen, Y. Protein Adsorption Onto Zirconia Modified With Terminally Grafted Polyvinylpyrrolidone. J. Colloid Interface Sci. 2001, 235 (1),
70–79.
74. Khemakhem, S.; Ben, Amar R. Grafting of Fluoroalkylsilanes on Microfiltration Tunisian Clay Membrane. Ceram. Int. 2011, 37 (8), 3323–3328.
75. Lin, Y. S.; Burggraaf, A. J. Experimental Studies on Pore Size Change of Porous Ceramic Membranes After Modification. J. Membr. Sci. 1993, 79 (1), 65–82.
76. Hu, J.; Qi, H.; Fan, Y.; et al. Porous Ceramic Support of Coated Alumina Prepared by Low-Temperature Sintering. J. Chin. Ceram. Soc. 2009, 37 (11), 1818–1823.
77. Falamaki, C.; Naimi, M.; Aghaie, A. Dual Behavior of CaCO3 as a Porosifier and Sintering Aid in the Manufacture of Alumina Membrane/Catalyst Supports. J. Eur. Ceram.
Soc. 2004, 24 (10–11), 3195–3201.
78. Wang, Y.; Zhang, Y.; Liu, X.; et al. Microstructure Control of Ceramic Membrane Support From Corundum-Rutile Powder Mixture. Powder Technol. 2006, 168 (3),
125–133.
79. Vasanth, D.; Pugazhenthi, G.; Uppaluri, R. Fabrication and Properties of Low Cost Ceramic Microfiltration Membranes for Separation of Oil and Bacteria From Its
Solution. J. Membr. Sci. 2011, 379 (1–2), 154–163.
80. Dong, Y.; Hampshire, S.; Zhou, J.; et al. Sintering and Characterization of Flyash-Based Mullite With MgO Addition. J. Eur. Ceram. Soc. 2011, 31 (5), 687–695.
81. Fang, J.; Qin, G.; Wei, W.; et al. Preparation and Characterization of Tubular Supported Ceramic Microfiltration Membranes From Fly Ash. Sep. Purif. Technol. 2011,
80 (3), 585–591.
82. Almandoz, M. C.; Marchese, J.; Prádanos, P.; et al. Preparation and Characterization of Non-Supported Microfiltration Membranes From Aluminosilicates. J. Membr. Sci.
2004, 241 (1), 95–103.
83. Dong, Y.; Liu, X.; Ma, Q.; et al. Preparation of Cordierite-Based Porous Ceramic Micro-Filtration Membranes Using Waste Fly Ash as the Main Raw Materials. J. Membr.
Sci. 2006, 285 (1–2), 173–181.
84. Majouli, A.; Younssi, S. A.; Tahiri, S.; et al. Characterization of Flat Membrane Support Elaborated From Local Moroccan Perlite. Desalination 2011, 277 (1–3), 61–66.
85. Kingman, S. W. Recent Developments in Microwave Processing of Minerals. Int. Mater. Rev. 2006, 51 (1), 1–12.
86. Menezes, R. R.; Kiminami, R. Microwave Sintering of Alumina-Zirconia Nanocomposites. J. Mater. Process. Technol. 2008, 203 (1–3), 513–517.
87. Bian, H. M.; Yang, Y.; Wang, Y.; et al. Alumina-Titania Ceramics Prepared by Microwave Sintering and Conventional Pressure-Less Sintering. J. Alloys Compd. 2012,
525, 63–67.
88. Oh, S. T.; Tajima, K.; Ando, M.; et al. Fabrication of Porous Al2O3 by Microwave Sintering and Its Properties. Mater. Lett. 2001, 48 (3-4), 215–218.
89. Feng, J.; Fan, Y.; Qi, H.; et al. Co-Sintering Synthesis of Tubular Bilayer Alpha-Alumina Membrane. J. Membr. Sci. 2007, 288 (1–2), 20–27.
90. Feng, J.; Qiu, M.; Fan, Y.; et al. The Effect of Membrane Thickness on the Co-Sintering Process of Bi-Layer ZrO2/Al2O3 Membrane. J. Membr. Sci. 2007, 305 (1–2),
20–26.
91. Dong, Y.; Lin, B.; Wang, S.; et al. Cost-Effective Tubular Cordierite Micro-Filtration Membranes Processed by Co-Sintering. J. Alloys Compd. 2009, 477 (1–2), L35–L40.
92. Chowdhury, S. R.; Keizer, K.; ten Elshof, J. E.; et al. Effect of Trace Amounts of Water on Organic Solvent Transport Through Gamma-Alumina Membranes With Varying
Pore Sizes. Langmuir 2004, 20 (11), 4548–4552.
93. Zhang, H.; Zhong, Z.; Xing, W. Application of Ceramic Membranes in the Treatment of Oilfield-Produced Water: Effects of Polyacrylamide and Inorganic Salts.
Desalination 2013, 309, 84–90.
94. Zhao, D.; Lau, E.; Huang, S.; et al. The Effect of Apple Cider Characteristics and Membrane Pore Size on Membrane Fouling. LWT Food Sci. Technol. 2015, 64 (2),
974–979.
95. Matsumoto, Y.; Nakao, S.; Kimura, S. Cross-Flow Filtration Solutions of Polymers Using Ceramic Microfiltration. Int. Chem. Eng. 1988, 28, 677–683.
96. Cheng, X.; Li, N.; Zhu, M.; et al. Positively Charged Microporous Ceramic Membrane for the Removal of Titan Yellow Through Electrostatic Adsorption. J. Environ. Sci.
2016, 44, 204–212.
97. Yang, C.; Zhang, G.; Xu, N.; et al. Preparation and Application in Oil–Water Separation of ZrO2/a-Al2O3 MF Membrane. J. Membr. Sci. 1998, 142 (2), 235–243.
98. Xu, N.; Li, W.; Zhao, Y.; et al. Theory and Method of Application-Oriented Ceramic Membranes Design (I). J. Chem. Ind. Eng. (China) 2003, 54 (9), 1284–1289.
99. Li, W.; Zhao, Y.; Liu, F.; et al. Theory and Method of Application-Oriented Ceramic Membranes Design (II). J. Chem. Ind. Eng. (China) 2003, 54 (9), 1290–1294.
100. Zhao, Y.; Li, W.; Zhang, W.; et al. Theory and Method of Application-Oriented Ceramic Membranes Design (III). J. Chem. Ind. Eng. (China) 2003, 54 (9), 1295–1299.
101. Elzo, D.; Huisman, I.; Middelink, E.; et al. Charge Effects on Inorganic Membrane Performance in a Cross-Flow Microfiltration Process. Colloids Surf. A Physicochem.
Eng. Asp. 1998, 138 (2–3), 145–159.
102. Zhang, L.; Li, N.; Zhu, M.; et al. Nano-Structured Surface Modification of Microporous Ceramic Membrane With Positively Charged Nano-Y2O3 Coating for Organic Dyes
Removal. RSC Adv. 2015, 5 (98), 80643–80649.
103. Fievet, P.; Szymczyk, A.; Aoubiza, B.; et al. Evaluation of Three Methods for the Characterisation of the Membrane–Solution Interface: Streaming Potential, Membrane
Potential and Electrolyte Conductivity Inside Pores. J. Membr. Sci. 2000, 168 (1–2), 87–100.
104. Lu, D.; Zhang, T.; Gutierrez, L.; et al. Influence of Surface Properties of Filtration-Layer Metal Oxide on Ceramic Membrane Fouling During Ultrafiltration of Oil/Water
Emulsion. Environ. Sci. Technol. 2016, 50 (9), 4668–4674.
296 Ceramic Membranes

105. Zhang, Q.; Jing, W.; Fan, Y.; et al. An Improved Parks Equation for Prediction of Surface Charge Properties of Composite Ceramic Membranes. J. Membr. Sci. 2008,
318 (1–2), 100–106.
106. Zhang, Q.; Fan, Y.; Xu, N. Effect of the Surface Properties on Filtration Performance of Al2O3–TiO2 Composite Membrane. Sep. Purif. Technol. 2009, 66 (2), 306–312.
107. Gao, N.; Li, M.; Jing, W.; et al. Improving the Filtration Performance of ZrO2 Membrane in Non-Polar Organic Solvents by Surface Hydrophobic Modification. J. Membr.
Sci. 2011, 375 (1–2), 276–283.
108. Gao, N. W.; Fan, Y. Q.; Quan, X. J.; et al. Modified Ceramic Membranes for Low Fouling Separation of Water-in-Oil Emulsions. J. Mater. Sci. 2016, 51 (13),
6379–6388.
109. Hsieh, H. P. Inorganic Membranes for Separation and Reaction; Elsevier Science: Amsterdam, 1996.
110. Almecija, M. C.; Ibanez, R.; Guadix, A.; et al. Effect of pH on the Fractionation of Whey Proteins With a Ceramic Ultrafiltration Membrane. J. Membr. Sci. 2007,
288 (1–2), 28–35.
111. De Angelis, L.; de Cortalezzi, M. M. F. Ceramic Membrane Filtration of Organic Compounds: Effect of Concentration, pH, and Mixtures Interactions on Fouling. Sep.
Purif. Technol. 2013, 118, 762–775.
112. De la Casa, E. J.; Guadix, A.; Ibanez, R.; et al. Influence of pH and Salt Concentration on the Cross-Flow Microfiltration of BSA Through a Ceramic Membrane. Biochem.
Eng. J. 2007, 33 (2), 110–115.
113. Fakhfakh, S.; Baklouti, S.; Baklouti, S.; et al. Preparation, Characterization and Application in BSA Solution of Silica Ceramic Membranes. Desalination 2010, 262 (1–3),
188–195.
114. Farsi, A.; Boffa, V.; Christensen, M. L. Electroviscous Effects in Ceramic Nanofiltration Membranes. ChemPhysChem 2015, 16 (16), 3397–3407.
115. Sbai, M.; Fievet, P.; Szymczyk, A.; et al. Streaming Potential, Electroviscous Effect, Pore Conductivity and Membrane Potential for the Determination of the Surface
Potential of a Ceramic Ultrafiltration Membrane. J. Membr. Sci. 2003, 215 (1–2), 1–9.
116. Alventosa-DeLara, E.; Barredo-Damas, S.; Zuriaga-Agusti, E.; et al. Ultrafiltration Ceramic Membrane Performance During the Treatment of Model Solutions Containing
Dye and Salt. Sep. Purif. Technol. 2014, 129, 96–105.
117. Yoon, S. H.; Kang, I. J.; Lee, C. H. Fouling of Inorganic Membrane and Flux Enhancement in Membrane-Coupled Anaerobic Bioreactor. Sep. Sci. Technol. 1999, 34 (5),
709–724.
118. Hoogland, M. R.; Fane, A. G.; Fell, C. J. D. The Effect of pH on the Cross-Flow Filtration of Mineral Slurries Using Ceramic Membranes. In Proceedings of the First
International Conference on Inorganic Membranes. Montpellier, France, 1989; p. 153.
119. Moosemiller, M. D.; Hill, C. G.; Anderson, M. A. Physicochemical Properties of Supported g-Al2O3 and TiO2 Ceramic Membranes. Sep. Sci. Technol. 1989, 24 (9–10),
641–657.
120. Nazzal, F. F.; Wiesner, M. R. pH and Ionic Strength Effects on the Performance of Ceramic Membranes in Water Filtration. J. Membr. Sci. 1994, 93 (1), 91–103.
121. Huisman, I. H.; Dutré, B.; Persson, K. M.; et al. Water Permeability in Ultrafiltration and Microfiltration: Viscous and Electroviscous Effects. Desalination 1997, 113 (1),
95–103.
122. Grabow, W. O. K.; Dohmann, M.; Haas, C.; et al. Influence of Particle Size and Surface Charge on Critical Flux of Crossflow Microfiltration. Water Quality International
‘98Water Sci. Technol. 1998, 38 (4), 481–488.
123. Chevereau, E.; Zouaoui, N.; Limousy, L.; et al. Surface Properties of Ceramic Ultrafiltration TiO2 Membranes: Effects of Surface Equilibriums on Salt Retention.
Desalination 2010, 255 (1–3), 1–8.
124. Chiu, T. Y. Effect of Ageing on the Microfiltration Performance of Ceramic Membranes. Sep. Purif. Technol. 2011, 83, 106–113.
125. Huisman, I. H.; Vellenga, E.; Trägårdh, G.; et al. The Influence of the Membrane Zeta Potential on the Critical Flux for Crossflow Microfiltration of Particle Suspensions. J.
Membr. Sci. 1999, 156 (1), 153–158.
126. Wang, P.; Xu, N.; Shi, J. A Pilot Study of the Treatment of Waste Rolling Emulsion Using Zirconia Microfiltration Membranes. J. Membr. Sci. 2000, 173 (2), 159–166.
127. Zhao, Y.; Zhong, J.; Li, H.; et al. Fouling and Regeneration of Ceramic Microfiltration Membranes in Processing Acid Wastewater Containing Fine TiO2 Particles. J.
Membr. Sci. 2002, 208 (1–2), 331–341.
128. Alpatova, A.; Kim, E.-S.; Dong, S.; et al. Treatment of Oil Sands Process-Affected Water With Ceramic Ultrafiltration Membrane: Effects of Operating Conditions on
Membrane Performance. Sep. Purif. Technol. 2014, 122, 170–182.
129. Cui, Z.; Peng, W.; Fan, Y.; et al. Effect of Cross-Flow Velocity on the Critical Flux of Ceramic Membrane Filtration as a Pre-treatment for Seawater Desalination. Chin. J.
Chem. Eng. 2013, 21 (4), 341–347.
130. Chang, Q.; Zhou, J.; Wang, Y.; et al. Application of Ceramic Microfiltration Membrane Modified by Nano-TiO2 Coating in Separation of a Stable Oil-in-Water Emulsion. J.
Membr. Sci. 2014, 456, 128–133.
131. Hutchings, G. J. New Approaches to Rate Enhancement in Heterogeneous Catalysis. Chem. Commun. 1999, 4, 301–306.
132. Julbe, A.; Farrusseng, D.; Guizard, C. Porous Ceramic Membranes for Catalytic Reactors—Overview and New Ideas. J. Membr. Sci. 2001, 181 (1), 3–20.
133. Jiang, H.; Meng, L.; Chen, R. Z.; et al. Progress on Porous Ceramic Membrane Reactors for Heterogeneous Catalysis Over Ultrafine and Nano-Sized Catalysts. Chin. J.
Chem. Eng. 2013, 21 (2), 205–215.
134. Zhong, Z.; Xing, W.; Jin, W.; et al. Adhesion of Nanosized Nickel Catalysts in the Nanocatalysis/UF System. AICHE J. 2007, 53 (5), 1204–1210.
135. Chen, R.; Du, Y.; Wang, Q.; et al. Effect of Catalyst Morphology on the Performance of Submerged Nanocatalysis/Membrane Filtration System. Ind. Eng. Chem. Res.
2009, 48 (14), 6600–6607.
136. Jiang, H.; Meng, L.; Chen, R. Z.; et al. A Novel Dual-Membrane Reactor for Continuous Heterogeneous Oxidation Catalysis. Ind. Eng. Chem. Res. 2011, 50 (18),
10458–10464.
137. Kim, M. J.; Kang, O. Y.; Rao, B. S.; et al. Proposing a New Fouling Index in a Membrane Bioreactor (MBR) Based on Mechanistic Fouling Model. Desalin. Water Treat.
2011, 33 (1–3), 209–217.
138. Wu, J.; He, C.; Jiang, X.; et al. Modeling of the Submerged Membrane Bioreactor Fouling by the Combined Pore Constriction, Pore Blockage and Cake Formation
Mechanisms. Desalination 2011, 279 (1–3), 127–134.
139. Chang, I. S.; Le Clech, P.; Jefferson, B.; et al. Membrane Fouling in Membrane Bioreactors for Wastewater Treatment. J. Environ. Eng.-ASCE 2002, 128 (11),
1018–1029.
140. Le-Clech, P.; Chen, V.; Fane, T. A. G. Fouling in Membrane Bioreactors Used in Wastewater Treatment. J. Membr. Sci. 2006, 284 (1–2), 17–53.
141. Zhong, Z.; Xing, W.; Liu, X.; et al. Fouling and Regeneration of Ceramic Membranes Used in Recovering Titanium Silicalite-1 Catalysts. J. Membr. Sci. 2007, 301 (1–2),
67–75.
142. Zhong, Z.; Li, D.; Liu, X.; et al. The Fouling Mechanism of Ceramic Membranes Used for Recovering TS-1 Catalysts. Chin. J. Chem. Eng. 2009, 17 (1), 53–57.
143. Vaidya, M. J.; Kulkarni, S. M.; Chaudhari, R. V. Synthesis of p-Aminophenol by Catalytic Hydrogenation of p-Nitrophenol. Org. Process Res. Dev. 2003, 7 (2), 202–208.
144. Koltuniewicz, A. B.; Field, R. W.; Arnot, T. C. Engineering of Membrane Processes II: Environmental Applications Cross-Flow and Dead-End Microfiltration of Oily-Water
Emulsion. Part I: Experimental Study and Analysis of Flux Decline. J. Membr. Sci. 1995, 102, 193–207.
145. Marchese, J.; Ochoa, N. A.; Pagliero, C.; et al. Pilot-Scale Ultrafiltration of an Emulsified Oil Wastewater. Environ. Sci. Technol. 2000, 34 (14), 2990–2996.
146. Bensadok, K.; Belkacem, M.; Nezzal, G. Treatment of Cutting Oil/Water Emulsion by Coupling Coagulation and Dissolved Air Flotation. Desalination 2007, 206 (1),
440–448.
147. Hua, F. L.; Tsang, Y. F.; Wang, Y. J.; et al. Performance Study of Ceramic Microfiltration Membrane for Oily Wastewater Treatment. Chem. Eng. J. 2007, 128 (2–3),
169–175.
 Ceramic Membranes 297

148. Yeom, H. J.; Kim, S. C.; Kim, Y. W.; et al. Processing of Alumina-Coated Clay-Diatomite Composite Membranes for Oily Wastewater Treatment. Ceram. Int. 2016,
42 (4), 5024–5035.
149. Bhave, R. R.; Fleming, H. L. Removal of Oily Contaminants in Wastewater With Microporous Alumina Membranes. AlChE Symp. Ser. 1988, 84, 19–27.
150. Abbasi, M.; Taheri, A. Modeling of Coagulation-Microfiltration Hybrid Process for Treatment of Oily Wastewater Using Ceramic Membranes. J. Water Chem. Technol.
2014, 36 (2), 80–89.
151. Abbasi, M.; Taheri, A. Selecting Model for Treatment of Oily Wastewater by MF-PAC Hybrid Process Using Mullite-Alumina Ceramic Membranes. J. Water Chem.
Technol. 2016, 38 (3), 173–180.
152. Hilal, N.; Busca, G.; Hankins, N.; et al. Desalination Strategies in South Mediterranean Countries: The Use of Ultrafiltration and Nanofiltration Membranes in the Treatment
of Metal-Working Fluids. Desalination 2004, 167, 227–238.
153. Tomaszewska, M.; Orecki, A.; Karakulski, K. Desalination and the Environment Treatment of Bilge Water Using a Combination of Ultrafiltration and Reverse Osmosis.
Desalination 2005, 185 (1), 203–212.
154. Yu, X.; Zhong, Z.; Xing, W. Treatment of Vegetable Oily Wastewater Using an Integrated Microfiltration-Reverse Osmosis System. Water Sci. Technol. 2010, 61 (2),
455–462.
155. Yang, C.; Zhang, G.; Xu, N.; et al. Preparation and Application in Oil7Water Separation of ZrO2/a-Al2O3 MF Membrane. J. Membr. Sci. 1998, 142 (2), 235–243.
156. Zhang, G.; Gu, H.; Xing, W. Treatment of Cool Rolling Emulsion With Inorganic Ceramic Membranes. J. Chem. Eng. Chin. Univ. 1998, 12 (3), 288.
157. Kamble, S. P.; Barve, P. P.; Joshi, J. B.; et al. Purification of Lactic Acid via Esterification of Lactic Acid Using a Packed Column, Followed by Hydrolysis of Methyl
Lactate Using Three Continuously Stirred Tank Reactors (CSTRs) in Series: A Continuous Pilot Plant Study. Ind. Eng. Chem. Res. 2012, 51 (4), 1506–1514.
158. Datta, R.; Henry, M. Lactic Acid: Recent Advances in Products, Processes and Technologies—A Review. J. Chem. Technol. Biotechnol. 2006, 81 (7), 1119–1129.
159. Kyuchoukov, G.; Yankov, D. Lactic Acid Extraction by Means of Long Chain Tertiary Amines: A Comparative Theoretical and Experimental Study. Ind. Eng. Chem. Res.
2012, 51 (26), 9117–9122.
160. Bayazit, S. S.; Inci, I.; Uslu, H. Adsorption of Lactic Acid From Model Fermentation Broth Onto Activated Carbon and Amber Lite IRA-67. J. Chem. Eng. Data 2011,
56 (5), 1751–1754.
161. Bouchoux, A.; Roux-de Balmann, H.; Lutin, F. Investigation of Nanofiltration as a Purification Step for Lactic Acid Production Processes Based on Conventional and
Bipolar Electrodialysis Operations. Sep. Purif. Technol. 2006, 52 (2), 266–273.
162. Li, W.; Xing, W. Advances in Refinement of Lactic Acid From Fermentation Broths. Chem. Ind. Eng. Prog. 2009, 28 (3), 491–495.
163. Persson, A.; Jonsson, A. S.; Zacchi, G. Separation of Lactic Acid-Producing Bacteria From Fermentation Broth Using a Ceramic Microfiltration Membrane With Constant
Permeate Flow. Biotechnol. Bioeng. 2001, 72 (3), 269–277.
164. Pal, P.; Sikder, J.; Roy, S.; et al. Process Intensification in Lactic Acid Production: A Review of Membrane Based Processes. Chem. Eng. Process. 2009, 48 (11–12),
1549–1559.
165. Wang, K.; Li, W.; Fan, Y.; et al. Integrated Membrane Process for the Purification of Lactic Acid From a Fermentation Broth Neutralized With Sodium Hydroxide. Ind. Eng.
Chem. Res. 2013, 52 (6), 2412–2417.
166. Aspelund, M. T.; Glatz, C. E. Purification of Recombinant Plant-Made Proteins From Corn Extracts by Ultrafiltration. J. Membr. Sci. 2010, 353 (1–2), 103–110.
167. Roman, G. P.; Neagu, E.; Moroeanu, V.; et al. The Ultrafiltration Performance of Composite Membranes for the Concentration of Plant Extracts. Rom. Biotech. Lett. 2009,
14 (5), 4620–4624.
168. Xu, N.; Xing, W. Process for Refining Salt-Water by Inorganic-Membrane Filtering. Patent CN1868878, 2006.
169. Shirazi, S.; Lin, C.-J.; Chen, D. Inorganic Fouling of Pressure-Driven Membrane Processes—A Critical Review. Desalination 2010, 250 (1), 236–248.
170. Kim, J.; Yoon, T. I. Direct Observations of Membrane Scale in Membrane Bioreactor for Wastewater Treatment Application. Water Sci. Technol. 2010, 61 (9), 2267–2272.
171. Lee, M.; Kim, J. Membrane Autopsy to Investigate CaCO3 Scale Formation in Pilot-Scale, Submerged Membrane Bioreactor Treating Calcium-Rich Wastewater. J. Chem.
Technol. Biotechnol. 2009, 84 (9), 1397–1404.
172. Gu, J.; Zhang, H.; Zhong, Z.; et al. Conditions Optimization and Kinetics for the Cleaning of Ceramic Membranes Fouled by BaSO4 Crystals in Brine Purification Using a
DTPA Complex Solution. Ind. Eng. Chem. Res. 2011, 50 (19), 11245–11251.