Separation and Purification Technology 338 (2024) 126441

Contents lists available at ScienceDirect

Separation and Purification Technology

journal homepage: www.elsevier.com/locate/seppur

Co-sintered ceramic membranes for separation applications: Where are we

and where to go?
Dong Zou a, Zhaoxiang Zhong a, b, *, Yiqun Fan b, *
a
School of Environmental Science and Engineering, Nanjing Tech University, Nanjing 211816, China
b
National Engineering Research Center for Special Separation Membrane, College of Chemical Engineering, Nanjing Tech University, Nanjing 211816, China

A R T I C L E I N F O A B S T R A C T

Keywords: Owing to their high mechanical strength and thermal/chemical resistance, ceramic membranes play an impor­
Ceramic membrane tant role in water treatment, gas separation, gas/solid filtration, and other applications. However, compared with
Co-sintering process the market share of polymer membranes, that of ceramic membranes is lower in industrial applications owing to
Industrial application
high costs of materials and sintering consumption, which account for ~ 20 % and ~ 60 % of the total cost,
respectively. Reducing the sintering consumption would widen the usage range of ceramic membranes. The co-
sintering process, which combines several sintering processes into a single process, can lower the sintering
consumption by drastically shortening the fabrication time. However, the mismatched sintering stress and sin­
tering temperature between the two layers in the co-sintering process degrade the microstructure, integrity, and
performance of the product. This review systemically investigated the three evolutionary stages of sintering
processes (free sintering, constrained sintering, and co-sintering) and analyzed the sintering consumption,
interfacial strength, and application areas of co-sintering. To provide deep insights regarding the co-sintering
process of ceramic membranes, this review analyzed and discussed applications of co-sintering to solid oxide
fuel cells and low-temperature co-fired ceramics processes. Finally, it discussed co-sintering processes for
fabricating ceramic membranes, such as ceramic microfiltration and ultrafiltration membranes. Overall, this
review can provide an important reference on ceramic membranes with low energy consumption for further
industrial applications.

1. Introduction MPa. The intermediate layers formed on the ceramic support prevent the
penetration of ceramic particles from the membrane layers into the
Membrane separation technology is widely used for treating waste­ support. The penetration phenomenon, called the layer effect, consid­
water, dust-laden gas, and other contaminated materials [1,2]. Based on erably increases the filtration resistance and decreases the water flux.
the membrane material, separation membranes are broadly divided into The separation layer deposited on the intermediate layers provides
polymer and inorganic membranes [3]. Among the inorganic mem­ selectivity to the membrane [8].
branes, ceramic membranes deliver satisfactory performance and can be Publications on ceramic membranes comprised a rapidly increasing
easily scaled-up in future for industrial application. Polymer membranes proportion of Web of Science publications from 2003 to 2022 (Fig. 1b).
can be quickly and easily fabricated via (for example) the phase- More than 1000 ceramic-membrane publications were recorded in Web
inversion method. Unlike the fabrication methods of polymer mem­ of Science in 2022. Publications on low-cost membranes also increased,
branes, ceramic membranes must be sintered at high temperatures to from 10 in 2003 to 130 in 2022. Therefore, identifying strategies that
maintain a satisfactory bending strength [4,5]. Most ceramic mem­ will lower the production costs of ceramic membranes is vitally
branes exhibit an asymmetric structure comprising a support, interme­ important. Co-sintering is the most promising technique for decreasing
diate layers, and a separation layer (Fig. 1a). Each layer requires a the energy consumption, not only by decreasing the fabrication cost, but
repeated coating–drying–sintering process, requiring high energy con­ also by shortening the fabrication period [9–11].
sumption and long fabrication time [6,7]. The ceramic support provides Recently, some reviews on the development of ceramic membrane
the high bending strength of the final ceramic membrane, usually > 50 have been reported. Dong et al. [12] systemically investigated the cost

* Corresponding authors.
E-mail addresses: zhongzx@njtech.edu.cn (Z. Zhong), yiqunfan@njtech.edu.cn (Y. Fan).

https://doi.org/10.1016/j.seppur.2024.126441
Received 24 November 2023; Received in revised form 3 January 2024; Accepted 15 January 2024
Available online 20 January 2024
1383-5866/© 2024 Elsevier B.V. All rights reserved.
D. Zou et al. Separation and Purification Technology 338 (2024) 126441

and efficiency of ceramic membranes and their applications in water processes of membranes prepared by ceramic microfiltration (MF) and
treatment. They pointed out that the ceramic membrane could be ultrafiltration (UF) are then analyzed. The review also mentions the
decreased by the strategies of membrane formation, including the ma­ main research foci of the co-sintering process, which have not been
terial cost and processing cost. Wang et al. [2] reported the development discussed in previous reviews.
and water treatment of alumina ceramic membranes. The lowing cost
strategies and improvement of membrane performance were reviewed 2. Analysis of the sintering process
of alumina membranes. Hubadillah et al. [13] reported the development
of membranes by using low-cost kaolin as materials. The future devel­ Sintering is a key step in the preparation of porous ceramic mem­
opment of kaolin based membranes was also illustrated. Mestre et al. branes [16,17]. The transportation mechanism of the mass and the
[14] reported the industrial applications by using low-cost ceramic growth of grains and pores play crucial roles in the ceramic sintering
membranes. They mainly discussed the raw materials, ceramic compo­ kinetics and resulting properties [18]. The membrane layers shrink
sitions and variables of the producing processes in the formation of during the sintering process, generating membrane sintering stress that
ceramic membranes. Although these published reviews focused on the degrades the membrane integrity. Because the microstructures and
cost issues of the ceramic membranes and related applications, this properties of a sintered membrane are anisotropic, the development of a
reviewed systemically introduced the co-sintering technology for the sintering theory for stress analysis can guide the preparation process of
fabrication of low-cost membranes. It is the first review that deeply asymmetric ceramic membranes. The following content concisely re­
analyzed the stress evolution of the ceramic membrane with various views the constrained sintering process of single membrane layers and
structures. Comprehensively introducing the fabrication of ceramic the co-sintering process of bi-layer membranes.
membrane by using co-sintering technique. This provides new and deep
insights of sintering consumption related to the fabrication cost.
This paper systematically reviews co-sintering processes for fabri­ 2.1. Constrained sintering process of single ceramic-membrane layers
cating ceramic membranes. First, the sintering behaviors of membranes
are analyzed, the theory of constrained sintering of monolayer mem­ A wet membrane layer is usually prepared via dip coating, spray
branes is introduced, and the multi-layer co-sintering process and its coating, or spin coating on a rigid ceramic support [19–21]. After a
mature applications to solid oxide fuel cells (SOFCs) and low- series of drying and sintering processes, the membrane layer becomes an
temperature co-fired ceramics (LTCCs) are discussed. The co-sintering asymmetric ceramic membrane. During the sintering process, the single
membrane layer is subjected to the tensile stress of the rigid support

Fig. 1. (a) The cross-sectional structure of the asymmetric ceramic membrane [15], and (b) the publications of ceramic membrane in recent 20 years from 2003 to
2022. Note that the data records were obtained from Web of Science with the topic of “ceramic” and “membrane”.

2
D. Zou et al. Separation and Purification Technology 338 (2024) 126441

while the rigid support receives compressive stress from the membrane
ε˙z = εfree
˙ [(1 + N)/(1 − N)] (6)
layer. The constrained sintering theory was originally proposed in the
Scherer–Garino model to explain the effects of sintering stress on the
σ z = σ y = − F εfree
˙ /(1 − N) (7)
densification rate of ceramic materials [22]. Since then, this theory has
been widely applied to the sintering processes of ceramic and metal Meanwhile, the sintering driving force of porous materials is related
membranes on porous rigid supports. On rigid ceramic supports, to the shrinkage rate and is calculated as follows:
ceramic-membrane materials shrink only along the z-axis direction; ∑ / ( )
shrinkages along the x- and y-axis directions are zero (εx = εy = 0). In free
= F [3(1 − 2N)]⋅3⋅ − εfree
˙ (8)
addition, the sintering stresses generated along the x- and y-axes are
∑ ∑
equal (σx = σ y) [23,24]. = − σ (9)
The tensile stress induced in a ceramic-membrane layer on a rigid
constrained free

support due to sintering is related to the thermal expansion coefficients Combining εfree
˙ and a set of constitutive equations, we obtain Eqs.
of the materials (Fig. 2a). Considering this fact, Scherer et al. [25] (10)–(13) as follows:
investigated the relation between the shrinkage rate and sintering stress ρ
of ceramic materials and derived Eqs. (1)–(3): = 3πx2 (1 − cx) (10)
ρs
)]
˙ + F − 1 [σ x − N(σ z + σy
ε˙x = εfree (1) √̅̅̅
F(x)s = 2 2x/π (11)
˙ + F − 1 [σy − N(σx + σ z )]
ε˙y = εfree (2)
3π1/3 2 − 3cx
ε˙x = − K (12)
ε˙z = εfree
˙ +F − 1
[σz − N(σx + σ y )] (3) 6 x1/3 (1 − cx)1/3

where εfree
˙ and ε̇i represent the shrinkage rates of the ceramics during K = (γ/ηl0 )(ρs /ρ0 )1/3 (13)
free and constrained (along direction i) sintering, respectively, and F, N,
and σi denote the uniaxial viscosity, Poisson’s ratio, and sintering stress where l0, γ, and η represent the initial particle size, surface tension, and
(along direction i), respectively, of the ceramic materials during the viscosity of the ceramic powder at high temperature, respectively, x
constrained sintering process. represents the sintering degree of the ceramic material (on a scale from
An asymmetric ceramic membrane sintered on a rigid support is 0 to 0.5), c is a constant (1.2), ρ0 and ρs denote the initial bulk density of
constrained by the support. No deformation occurs in the x–y direction, the material and the density of the ceramic material, respectively, and σ
which is parallel to the membrane surface, and the deformation along is the average sintering stress of the single membrane layer, obtained by
the z direction is slight. The stress is limited to the x–y direction. Eqs (4) superposing the stresses in three directions. As the shrinkage in the z
and (5) hold during the sintering process: direction is free, the stress in that direction is zero and the average
driving force is given by
ε˙x = ε˙y = 0, ε˙z = 0 (4)
σ = (σx + σy )/3 (14)
σx = σy ∕
= 0, σz = 0 (5)
The sintering stress and driving force of the single membrane layer
Combining Eqs. (1)–(3) with Eqs. (4) and (5), the shrinkage rate in during the constrained sintering process largely affect the appropriate
the z direction and the sintering stress in the x–y direction are respec­ sintering temperature. Therefore, quantifying the sintering stress in the
tively given by membrane layer and the driving force of the top layer on the bottom
membrane during the co-sintering process would provide a theoretical

Fig. 2. Sintering stress analysis of ceramic membranes. (a) Constrained sintering process on the rigid supports, and (b) co-sintering process of the “AB”, “ABA”, and
“ABC” structures [27].

3
D. Zou et al. Separation and Purification Technology 338 (2024) 126441

basis for controlling the sintering stress. co-sintering technology have received widespread attention. Among
them, LTCCs have recently emerged as a remarkable interdisciplinary
2.2. Co-sintering of multi-layers during formation of the ceramic technology involving multiple fields such as circuitry, multi-layer wir­
membrane ing, multi-layer substrates, and co-firing [28–30]. The LTCC technology
is a new material preparation technology developed by Hughes Com­
The structures of co-sintered bi- or multi-layered ceramic membranes pany in 1982 for fabricating precise and dense raw ceramic belts using
can be simply divided into symmetric ABA structures or asymmetric AB low-temperature-sintered ceramic powders. The required circuit
structures (Fig. 2b). In a co-sintered ABA structure, the intermediate B graphics are made by conducting laser punching, micropore grouting,
layer simultaneously experiences the same stress from the upper and precision conductor paste printing, and other processes on raw ceramic
lower layers. Therefore, the overall ABA structure is not deformed by belts. Multiple passive components (such as low-capacity capacitors,
bending. By contrast, a co-sintered AB structure experiences different resistors, filters, impedance converters, and couplers) are embedded in
shrinkage rates and sintering stresses between the A and B layers, the multi-layer ceramic substrate. The inner and outer electrodes can be
causing bending deformation during the co-sintering process. Two cases co-sintered with silver, copper, gold and other metals below 900 ◦ C,
are observed depending on the inconsistency of the shrinkage rates forming the passive integrated components of three-dimensional circuit
layers A and B. When the B layer has a higher shrinkage rate than the A networks.
layer in the co-sintering process, the bending direction is towards the B The LTCC process mainly includes the preparation, blanking,
layer. Conversely, when the A layer has a higher shrinkage rate than the punching, pore filling, and printing of conductive media and lamination,
B layer, the bending direction will be towards to the A layer. The co- hot pressing, slicing, and co-firing of the tape. After sintering, the multi-
sintering behaviors of bi-layer membranes on rigid supports differ layered substrate can be welded or electroplated. As the dielectric ma­
from those of ABA and AB structures because the rigid support cannot terials (e.g., capacitors and substrates), magnetic dielectric materials (e.
shrink during sintering and limits the shrinkage of the membrane layer g., inductors), conductive materials (including the silver electrodes),
parallel to the membrane surface. This structure can be simplified as the and other materials must be fired at one time, co-sintering technology
ABC structure, which is the traditional asymmetric ceramic-membrane has become the bottleneck of the LTCC process. At present, LTCC ma­
structure and is most commonly used in the co-sintering process. Dur­ terials are usually obtained by (1) adding appropriate sintering aids
ing sintering, the AB layers in an ABC structure exhibit different (low-melting-point oxides or low-melting-point glass) for liquid-phase
shrinkage properties (sintering activities) caused by inconsistencies in sintering, (2) chemically preparing powders with high surface activity,
their materials and particle sizes. Thus, the two membrane layers restrict (3) employing (as far as possible) materials with fine particle size and a
each other and produce sintering stress. The total bending deformation main crystal phase that can be synthesized at low temperatures, and (4)
and stress of an ABC structure on a rigid support are zero. The stresses in utilizing glass–ceramic or amorphous glass.
each membrane layer of an ABC structure are calculated using [26] Eqs. Zou et al. [31] analyzed the generation and influence of interface
(15)–(18): stress during the co-sintering process of multi-layer electronic ceramic
components and elaborated technological measures for improving the
1
σA = ( ) ηA(ε˙A − ε˙B ) (15) interface stress. They found that interfacial stress in the co-sintering
1 + mn process was caused by mismatch of the thermal properties between
hA the different materials. Increasing the thickness or thickness ratio of the
σB = − σB (16) ceramic layer to the electrode layer can decrease the bending degree of
hB
the co-sintered material. Jean et al. [23,24,32] conducted in-depth
hA research and theoretical calculations on the LTCC process. They
m= (17) showed that co-sintering generated less sintering stress than sintering
hB
each layer at the same temperature. In addition, the deformation degree
ηA (1 − NB ) and stress strength are directly affected by differences among the sin­
n= (18)
ηB (1− NA) tering rates of the materials and the thickness of each layer. The thinner
the material, the greater is the bending degree during the co-sintering
where σ B (σA) represents the stress of layer A (layer B) on layer B (layer process.
A), h represents the membrane thickness, η represents the viscosity of
the material, and N is the viscous Poisson’s ratio. 3.2. Solid oxide fuel cells
In general, the sintering behaviors of each membrane layer must be
considered before co-sintering a multi-layered membrane. If the mate­ SOFCs are promising clean-energy conversion devices that convert
rials have similar sintering temperatures, employing different methods the chemical energy in various fuels (such as gases, liquids, and even
can alleviate the sintering stress of co-sintering. However, if the sinter­ solid fuels) into electricity through electrochemical reactions [33,34]. A
ing temperatures and shrinkage rates of the materials are very different, conventional SOFC comprises a cathode, an electrolyte, and an anode,
the co-sintering process is difficult or impossible. which are usually operated at temperatures above 900 ◦ C to overcome
the low conductivity of yttria-stabilized zirconia (YSZ). Co-sintering
3. Evolution and applications of co-sintering processes for technology is mainly used to prepare the anode/cathode battery of an
ceramic materials SOFC by sintering an electrolyte material with an anode or cathode
material.
Co-sintering has attracted an increasing share of attention because it The anode battery is usually prepared from a NiO/YSZ composite,
fabricates multi-layered ceramics using simple and efficient procedures which provides high electrical conductivity and thermal stability in a
at low production cost. Co-sintering technology was originally employed reducing atmosphere. The co-sintering temperature considerably affects
in the preparation of LTCCs and SOFCs for fuel cells. the anode (Ni/YSZ) microstructure, electrolyte film microstructure, and
cell performance. Co-sintering a cell at 1380 ◦ C maximizes the power
3.1. Low-temperature co-fired ceramics density and lowers the polarization concentration of the anodes. By
applying the phase-inversion co-extrusion/co-sintering technology,
Rapid development of modern information industry has increased Laguna-Bercero [35,36] prepared a high-temperature microtubular
the demand for high-performance, small-sized electronic components. SOFC with a NiO–YSZ/YSZ/LSM–YSZ structure. After varying the co-
Multi-layered structural ceramic components based on tape casting and sintering temperature, YSZ cathode size, and anode/electrolyte layers,

4
D. Zou et al. Separation and Purification Technology 338 (2024) 126441

they obtained cells with favorable microstructures and long-term enhance the ductility of the membrane layers [45].
stability. For example, Dong et al. [41] discussed the sintering characteristics
Although cathode-supported SOFCs are more stable than anode- and particle-size distributions of two cordierite powders in detail. When
supported SOFCs, cathode material batteries are difficult to prepare sintering was performed at 1200 ◦ C, the shrinkage difference of the two
because the cathode and electrolyte are chemically incompatible at high ceramic-membrane layers prepared from the cordierite powders was
temperatures. The cathode layer in a cathode-supported SOFC is much only 0.59 %, allowing the formation of a crack-free membrane layer
thicker and more liable to shrinkage than the electrolyte and anode with a narrow pore-size distribution. Zou et al. [43] designed a
layers, leading to densification of the electrolyte film during the co- sandwich-structured ceramic membrane with a SiC whisker layer and an
sintering process. To better match the cathode and electrolyte mate­ alumina layer on a macroporous support (mean pore size ≈ 3 μm)
rials, Dhrubaet al. [37] added sintering additives (Fe2O3 and NiO) to the (Fig. 4a). The intermediate layer of SiC whiskers prevents the infiltration
YSZ electrolyte material. The additives facilitate shrinkage of the elec­ of alumina particles while the alumina layer decreases the membrane
trolyte during the co-sintering process, thus obtaining a dense electro­ roughness. After co-sintering at 1000 ◦ C, a fibrous hybrid alumina
lyte film. membrane with a mean pore size of 100 nm and a water permeance of
645 L⋅m− 2⋅h− 1⋅bar− 1 was obtained. Interestingly, the sandwich-
4. Co-sintering process of ceramic membranes structured ceramic membrane obtained via co-sintering can be
extended to other applications. Zhang et al. [52] reported a hierar­
An asymmetric ceramic membrane with a hierarchical structure is chically porous interlayer for high-permeance ceramic membranes,
typically composed of a ceramic support, intermediate layers, and a which is similar to the whisker layer in Zou’s work (Fig. 4b). They coated
separation layer. Each layer is usually prepared through repeated sin­ a TiO2 layer containing polystyrene beads and a TiO2 layer not con­
tering–coating–drying processes, which prolong the preparation period taining beads on a rigid ceramic support. After co-sintering at 850 ◦ C,
and increase the energy consumption of co-sintering. A previous study the polystyrene in the TiO2 layer decomposed, which considerably
[38] estimated that sintering accounts for 60 % of the total cost of increased the porosity of the layer to 58 %.
asymmetric membrane preparation. Therefore, reducing the energy The MF–MF layered ceramic support was the earliest co-sintered
consumption and preparation period of sintering are essential when structure [39,53]. Co-sintered ceramic MF membranes are formed by
fabricating cost-effective ceramic membranes for engineering applica­ two main strategies: one involving two direct dry-pressing steps on a
tions. At present, the main asymmetric ceramic membranes prepared by rigid macroporous support and the other involving dip coating or spray
co-sintering are ceramic MF and ceramic UF membranes. Co-sintered coating. The first strategy effectively alleviates the penetration of the
ceramic nanofiltration (NF) membranes are not yet reported because membrane layer into the ceramic support or intermediate layer but
the shrinkage difference between the NF and MF layers is very large. The tends to increase the membrane thickness (to > 200 μm) [54] and hence
milestones of ceramic membrane via co-sintering process is shown in the filtration resistance. Moreover, this strategy is inappropriate for
Fig. 3. industrial preparations. In the second (conventional) strategy, the
ceramic was mixed with organic additives to form a well-dispersed
ceramic slurry under ultrasonic treatment, ball milling or other treat­
4.1. Ceramic microfiltration membranes ment methods [55,56]. Then, the dispersion was directly coated on the
ceramic support. The intermediate MF layer was “green” before the co-
4.1.1. Co-sintering of microfiltration layer and microfiltration layer sintering process. However, when a separation layer was coated on the
The fabrication of ceramic MF membranes by co-sintering two MF intermediate layer, the formation of membrane layer was difficult
layers on a macroporous support is similar to the fabrication process of because the polymer additives in the intermediate layer affected the
LTCCs and SOFCs. For fabricating a homogeneous and continuous capillary force. Hence, when coating the top layer, a tradeoff exists
membrane layer via co-sintering, the expansion and shrinkage behavior between the membrane-layer thickness and integrity, necessitating
of each ceramic layer must be considered. Researchers have explored control of the separation-layer thickness.
several effective strategies that alleviate the sintering stress caused by Co-sintering MF layer and MF layer on the rigid support is the most
different materials: adjusting the sintering temperature of the two layers mature co-sintering procedure. In the future, more efforts should
[49], adding rigid ceramic whiskers/particles to alleviate layer decrease the transferring resistance and increase the permeance of the
shrinkage [43,50], adding sintering aids to promote the sintering pro­ ceramic membrane. For example, decrease the “layer effect” in the
cess [51], and introducing intermediate layers or nano-sized metals to

Fig. 3. The milestones for the co-sintering of ceramic membranes [39–48].

5
D. Zou et al. Separation and Purification Technology 338 (2024) 126441

Fig. 4. Co-sintering ceramic MF layer and MF layer on the rigid supports for the fabrication of ceramic MF membrane. Take sand-witch liked ceramic membrane for
example. (a) Ceramic fibers acted as intermediate layer for co-sintering [43] and (b) holey porous intermediate layer for co-sintering [52].

interface between the support and membrane layer or decrease the MF membrane with an average pore size and permeability of approxi­
resistance of the intermediate layer is useful. Therefore, structure design mately 100 nm and 450 L⋅m− 2⋅h− 1⋅bar− 1, respectively, showed a high
by using whiskers or 2D materials will be an important direction. separation efficiency in oil-in-water emulsions. Zhu et al. [20] prepared
a green ceramic support by dry-pressing method, and then coated a pre-
4.1.2. Co-sintering of the ceramic support and microfiltration layer membrane layer on it. After co-sintering process, the resulting mem­
Ceramic MF membranes formed from a support and a membrane brane obtained an ultrahigh water permeance and high selectivity
layer have emerged with the development of co-sintering techniques for (Fig. 5a). They added silica sol as a sintering aid to the alumina support,
ceramic MF membranes [57]. This co-sintering method requires lower thereby reducing the sintering temperature and promoting adhesion
energy consumption and fewer preparation cycles than co-sintering two between the support and the membrane. After co-sintering at 1300 ◦ C,
different MF layers, but the thickness of the ceramic support (2–3 mm) they obtained an alumina ceramic membrane with an average pore size
far exceeds the thickness of the membrane layer (10–50 μm). Owing to of 249 nm and a water permeability of 5040 L⋅m− 2⋅h− 1⋅bar− 1.
the large shrinkage difference generated by sintering stress between the Membrane layers on ceramic green supports are mainly formed via
support and the membrane layer, a homogeneous MF membrane is thermal spraying and transfer coating, which are difficult to realize in
difficult to form. Strategies to alleviate this shrinkage include selecting engineering applications. The simple and feasible dip coating method
and optimizing the co-sintering temperature to minimize cracks in the has become the main method for preparing commercial asymmetric
membrane layer, and adding sintering additives and rigid fibers that ceramic membranes. However, as dry-pressed compact green materials
inhibit shrinkage of the ceramic support and membrane layer. The first lack a pore structure and flexural strength, a membrane layer with
strategy is mainly adopted. capillary functionality is not easily formed. Therefore, membrane-layer
To mitigate the large shrinkage difference between a green support formation through conventional dip coating on green compacts is rarely
made from fly ash and an Al2O3 layer, Zou et al. [44,50] doped the fly considered. To solve this problem, Gao et al. [58,59] integrated dip
ash with rigid mullite whiskers. A fly ash–based ceramic MF membrane coating with gel casting to prepare a piezoelectric quartz MF membrane
was successfully prepared by co-sintering at 1050 ◦ C. The Al2O3/fly-ash (Fig. 5b). The green quartz–ceramic body prepared via aqueous gel

6
D. Zou et al. Separation and Purification Technology 338 (2024) 126441

Fig. 5. Co-sintering ceramic support and MF layer for the fabrication of ceramic MF membranes. (a) Transfer method for alumina MF membranes [20], (b) dip-
coating method for quartz membranes [59], (c) dry-pressing method for mullite membranes, and (d) permeability performance of the membranes [57].

casting exhibited an obvious pore structure and a low flexure strength, the sintering energy consumption and preparation time. However, few
enabling dip coating of a thin and uniform membrane layer. Moreover, researchers have investigated the sintering shrinkage control of green
as the green support was negligibly shrunk at sintering temperatures of supports and the formation process of membranes on green supports
> 1000 ◦ C, the membrane layer was completely formed. When co- [38]. These topics must be addressed in the future work.
sintered at 1200 ◦ C, the piezoelectric quartz MF membrane exhibited
an average pore size of 190 nm, a membrane thickness of 10 μm, a pure- 4.1.3. Co-sintering of three layers for MF applications
water permeability of 351 L⋅m− 2⋅h− 1⋅bar− 1, and excellent anti-fouling Above we summarized the co-sintering processes of bi-layer mem­
performance in an oil-in-water emulsion owing to in-situ piezoelectric branes and membrane–support systems. One-step fabrication of ceramic
vibrations [60]. MF membranes on macroporous supports commonly requires the co-
Co-sintering has been used to synthesize not only Al2O3 and sintering of three or more membrane layers, which complicates the
quartz–ceramic MF membranes but also silicon carbide [61] and mullite sintering shrinkage behavior. A novel method that co-sinters three layers
ceramic membranes [62]. Similar to the co-sintering of Al2O3 ceramic (a ceramic support, a mullite whisker layer, and an alumina whisker
membranes, co-sintering of SiC MF membranes requires control of the layer) for MF membrane applications has been recently reported [47]
sintering shrinkage behavior between the support formed from coarse (Fig. 6 for details). The authors analyzed the contribution of the whisker
SiC powders and the membrane layer formed from fine SiC powder. Liu membrane to the co-sintering process and pore-structure regulation. The
et al. [63] added 1.5- wt% B4C powder to a support to prepare a porous mullite transition layer was found to form mullite–alumina sintering
SiC ceramic membrane; the addition of B4C decreased the co-sintering necks with the alumina support and alumina whisker-separation layer.
temperature from 2350 ◦ C to 2200 ◦ C, achieving a good sintering tem­ The pore structure established a gradient derived from the aspect ratio of
perature match between the support and the membrane. The prepared the mullite fibers and alumina whiskers, with the alumina whiskers
SiC ceramic membrane with an average pore size of 9.93 µm exhibited forming a smaller pore size than the fibers. The resulting ceramic
high mechanical strength (38.77 MPa) and an excellent nitrogen gas flux membranes with a pore size and water permeance of 120 nm and 4253
(19406 m3⋅m− 2⋅h− 1⋅bar− 1). Previously, an MF membrane layer on a L⋅m− 2⋅h− 1⋅bar− 1, respectively, treated 500 and 1000 ppm of oil in oil-in-
green SiC support was co-sintered (Fig. 5c). In this process, mullite water emulsions. The authors did not theoretically analyze the sintering
supports and membranes were formed from the same SiC materials using shrinkage behaviors of the three membrane materials; rather, they
two drying methods. Si–Al sol and catalysts such as AlF3 and MoO3 were assumed that co-sintering formed a nearly complete membrane layer.
added to the mullite layer. After sintering at high temperatures, the Nevertheless, the study confirmed that whisker materials alleviate sin­
alumina and silicon resources from the Si–Al sol were grown in-situ into tering stress and can be employed as the intermediate layer in mem­
the mullite whiskers in the pore channels of the mullite MF layer, branes showing large sintering shrinkage behavior.
decorating the pore structures and decreasing the pore sizes. The Darcy’s In general, co-sintering multi-layered (above 2 layers) ceramic
permeability coefficient [64] of the resulting membrane exceeded those membrane is a great challenge as the sintering stress is difficult to
of all previously reported membranes, and the membrane showed great control and analyze in the whole process. From sintering stress pro­
potential in filtering dust-laden gas (See Fig. 5d). spective, the thickness of co-sintered membrane should be lower to
Compared with the co-sintering of top and intermediate layers, the guarantee the integrity of the membranes. In addition, it can be visu­
co-sintering of membrane and green supports can considerably decrease alized that multi-co-sintering should be promoted in the ceramic MF

7
D. Zou et al. Separation and Purification Technology 338 (2024) 126441

Fig. 6. The co-sintering process for ceramic MF membrane with three co-sintered layers. (a) The co-sintering procedures and (b) detailed ceramic structures [47].

membrane. After the sintering theory and procedure being mature, this sintering of the MF layer. This mutual doping method largely matched
process could be expanded to the ceramic UF membranes. the sintering properties of the two layers. In the UF membrane with a 40-
μm-thick MF layer and a 1-μm-thick UF layer co-sintered at 1000 ◦ C, the
pure-water permeance and pore size reached 70 L⋅m− 2⋅ h− 1⋅bar− 1 and 5
4.2. Ceramic ultrafiltration membranes
nm, respectively. Although this membrane effectively removed dye from
dye-containing wastewater, the high isoelectric point (IEP = 8–9) of the
The first co-sintered ceramic UF membrane was reported by Liu et al.
alumina membranes can cause serious fouling on the membrane surface
[65]. They found that the decomposition of organic compounds can
when the wastewater pollutants are negatively charged. They continued
relieve the sintering stress between two membrane layers. They suc­
to put forward new methods for developing UF membranes. For
cessfully prepared a zirconia UF membrane with an average pore size of
example, when the UF layer was impregnated with nano-sized silver
30 nm. Zou et al. [66] co-sintered a tubular bi-layer Al2O3/ZrO2 mem­
particles, the IEP of the UF layer reduced from 9 to 3, enhancing the anti-
brane on a macroporous support, controlling the co-sintering tempera­
fouling performance of the resulting membranes. The silver-based
ture at 1050 ◦ C to regulate the sintering stress (sintering temperatures
nanoparticles also provided sufficient ductility to alleviate shrinkage
above 1050 ◦ C caused cracks on the membrane surface). They first
of the membrane layer. Fig. 7a shows the anti-fouling mechanism of this
deposited an Al2O3 layer (with a pore size of 100 nm) on the substrate to
membrane. The pure-water permeance of this membrane was 62
prevent top-layer penetration issues. Subsequently, they coated a ZrO2
L⋅m− 2⋅h− 1⋅bar− 1, the molecular weight cut-off was 9000 Da, and the
layer (with a pore size of 50 nm) on the wet Al2O3 layer for co-sintering.
rejection rate for bovine serum albumin (BSA) exceeded 99.5 %. The
The water permeance of the ZrO2/Al2O3 UF membrane reached 650
anti-fouling performance of this membrane far exceeded that of the
L⋅m− 2⋅h− 1⋅bar− 1, and the membrane showed good separation perfor­
conventional ceramic UF membrane (without silver nanoparticles) after
mance in treating the turbidity of a kaolin suspension. Wen et al. [67]
three fouling cycles. To prevent sol penetration, Shi et al. [71] prepared
prepared a bi-layer ZrO2 ceramic mesoporous membrane via a co-
a tight α-Al2O3 UF membrane with a pre-sintered intermediate layer
sintering process. To match the sintering behaviors of the sub-layer
(Fig. 7b). Owing to its sufficient strength and low surface roughness, the
and top layer, they added ZrO2 nanoparticles to the sub-layer to
pre-sintered zirconia layer resisted a certain level of sintering shrinkage
decrease the sintering temperature to 700 ◦ C. The mesoporous mem­
stress and reduced the membrane thickness. The prepared α-alumina UF
brane achieved a high water permeance of 280 L⋅m− 2⋅h− 1⋅bar− 1, a
membrane was ultrathin (thickness ≈ 140 nm) and exhibited a pure-
molecular weight cut-off of 40–50 kDa, and high application prospects
water permeance of 230 L⋅m− 2⋅h− 1⋅bar− 1, an average pore size of ~ 7
in water treatment.
nm, and high separation performance and stable flux during treatment
Ceramic UF membranes with small pore sizes are commonly fabri­
of wastewater containing polyvinyl alcohol (PVA) and Congo red dye.
cated using sol–gel technology as this technology could prepare a ho­
Most co-sintered ceramic UF membranes have an intermediate layer
mogenous sol with small particle size. For example, ZrO2, Al2O3, TiO2,
that increases the transfer resistance during the filtration process. The
and CeO2 sols were commonly prepared [68]. During the sintering
literature describes two effective approaches for alleviating the resis­
process, the UF membrane layers derived from sols are considerably
tance phenomenon: fabricating sandwich-like ceramic UF membranes
shrunk owing to the decomposition of organic additives. Therefore, the
with an intermediate layer of ceramic nanofibers, and removing the
preparation of UF membranes via co-sintering is limited owing to the
intermediate layer [72]. The nanofibers in the intermediate layers of
large shrinkage difference between the sub-micron-particle intermedi­
sandwich structures alleviate the stress between both membrane layers
ate layer and the nano-particle UF layer. To alleviate the high shrinkage
[73]. Qiu et al. [40] prepared bi-layer TiO2 UF membranes with an in­
rate between the MF and UF layers, Zou et al. adopted the mutual doping
termediate layer of sol-coated nanofibers. After co-sintering the sol and
method [69] and incorporated nano-sized metal particles in the UF layer
nanofiber layers at 480 ◦ C, they obtained a defect-free membrane with a
[70], forming a UF membrane layer with high integrity during water
molecular weight cut-off of 32000 Da and a pure-water flux of 1100
treatment. More specifically, the large alumina particles from the MF
L⋅m− 2⋅h− 1⋅bar− 1. Recently, a graphene oxide (GO) layer has been co-
layer doped into the UF layer alleviated shrinkage of the UF layer, while
sintered with an alumina layer [72]. The mean pore size of the GO
the sol from the UF layer incorporated into the MF layer promoted

8
D. Zou et al. Separation and Purification Technology 338 (2024) 126441

Fig. 7. Co-sintering process for the ceramic UF membranes. (a) Co-sintered alumina ceramic UF membrane with the addition of silver nano-particles [45], (b) a pre-
sintered approach for co-sintered alumina UF membrane [71], and (c) GO-layer pyrolyzed approach for co-sintered UF membranes [72].

layer deposited on the ceramic support was 100 nm. Alumina sol with a prevent the formation of a defect-free and homogeneous surface
particle size of 3.7 nm was then coated on the GO layer. At a co-sintering morphology. Hence, the sintering properties of different materials and
temperature of 1000 ◦ C, the GO layer was pyrolyzed (Fig. 7c) and the the membrane thickness of each layer must be thoroughly analyzed
alumina sol converted into an α-Al2O3 layer that acted as a UF separation prior to co-sintering.
layer. The resulting membrane was defect-free, as evidenced by its clear As we mentioned above, the current co-sintered membrane are
cross-sectional structure, and delivered an excellent permeability of 390 mainly composed of MF and UF membranes based on the mean pore
L⋅m− 2⋅h− 1⋅bar− 1 owing to the missing intermediate layer. Its average size. Fig. 8. shows the main application areas of these membranes. It can
pore size was 3.6 nm. be seen that the co-sintered ceramic MF are mainly employed for the
However, co-sintering fabrication is more difficult for ceramic UF treatment of O/W emulsions, dust-laden gas, particle suspensions and
membranes than for ceramic MF membranes. Sintering shrinkage etc. The co-sintered ceramic UF membrane are mainly employed to
behavior mainly depends on the membrane thickness, material type, and separate the bovine serum albumin (BSA) solution, dextran solution,
particle size. For instance, two membrane layers formed from the same dye/salt waste water, PVA waste water, ultrafine particle dispersion and
materials with different particle sizes exhibit different sintering behav­ others.
iors. The larger particles in the support generally require higher sin­ In addition, in order to show the detailed co-sintered membrane
tering temperatures to form sintering necks. The membrane thickness materials and application prospects, Table 1 is list regarding the mem­
and particle size usually decrease in the order of support membrane > brane properties, pore size, permeance, separation performance and
MF membrane > UF membrane. The consequent stress differences other important parameters. It was demonstrated that these membranes

Fig. 8. The main application areas of co-sintered ceramic membranes.

9
D. Zou et al. Separation and Purification Technology 338 (2024) 126441

Table 1
The performance of the co-sintered ceramic membranes and their applications.
Materials Structure Co-sintering Porosity Bending Pore size Permeance/ Application Ref.
temperature (◦ C) (%) strength PWF (pure water flux)/
(MPa) GF (gas flux)

α-Alumina Tube 1300 ℃ – – 200 nm PWF: 2250 (0.1 MPa), 3250 CaCO3 suspension: 650 [74]
(0.15 MPa), 4250 (0.2 MPa) L⋅m− 2⋅h− 1⋅bar− 1
L⋅m− 2⋅h− 1⋅bar− 1
ZrO2/Al2O3 Tube 1200 ℃ – – 280 nm PWF: 2.82 × 103 L⋅m− 2⋅h− 1⋅bar− 1
oil–water separation [39]
600
Cordierite Tube 1200 ℃, 1300 ℃ – – 1.55 μm, – – [41]
2.17 μm
− 2 − 1 − 1
Titania Sheet 480 ℃ – – 10 nm PWF: >3000 L⋅m ⋅h ⋅bar Molecule retention (32000 [40]
Da): 1100 L⋅m− 2⋅h− 1⋅bar− 1
α-Alumina Sheet 1000 ℃ – – 5 nm PWF: 70 L⋅m− 2⋅h− 1⋅bar− 1
Dye waste water (10000 Da): [42]
30 ~ 60 L⋅m− 2⋅h− 1⋅bar− 1
Alumina-SiC Sheet 1000 ℃ – – 100 nm PWF: 645 L⋅m− 2⋅h− 1⋅bar− 1
Molecule retention (16 ~ 21 [43]
whisker kDa): 110 ~ 130
L⋅m− 2⋅h− 1⋅bar− 1
SiC Sheet 2200 ~ 2350 ℃ – 32.15 ~ 9.93 μm Nitrogen gas flux: – [63]
38.77 MPa 194.06 m3⋅m-2⋅h− 1⋅ kPa− 1
Alumina Sheet 1300 ℃ ~37.5 % 23.5 MPa 249 nm PWF: 198,000 L⋅m− 2⋅h− 1⋅bar− 1
Carbon ink (237 nm): 5040 [20]
L⋅m− 2⋅h− 1⋅bar− 1
Fly ash Sheet 1050 ℃ ~30 % 22 MPa 100 nm PWF: 450 L⋅m− 2⋅h− 1⋅bar− 1
O/W emulsion (200 mg/L): [44]
165 L⋅m− 2⋅h− 1⋅bar− 1
− 2 − 1 − 1
Alumina- Sheet 1000 ℃ 43.1 % – 5–10 nm PWF: 62 L⋅m ⋅h ⋅bar Molecular retention (9000 [45]
AgNPs Da): 15 ~ 30 L⋅m− 2⋅h− 1⋅bar− 1
Fly ash Sheet 1000 ℃ 36 ~ 45 % 38.4 MPa 3 ~ 4 μm PWF: 1532 ~ 6113 O/W emulsion (SP): 127 ~ [75]
L⋅m− 2⋅h− 1⋅bar− 1 225 L⋅m− 2⋅h− 1⋅bar− 1
Zirconia- Hollow 1200–1500 ℃ 33.831 ~ – ~100 nm PWF: 33 ~ 477 L⋅m− 2⋅h− 1⋅bar− 1
Wastewater (POME, TI, dye): [48]
kaolin fiber 39.113 % 10 ~ 60 L⋅m− 2⋅h− 1⋅bar− 1
SiC Sheet 1900 ℃ 26–29 % 45 MPa 280 nm – – [61]
Fly ash Sheet 1250 ℃ 37 % 50.7 MPa 320 nm PWF: 3650 L⋅m− 2⋅h− 1⋅bar− 1
O/W emulsions: 530 [76]
L⋅m− 2⋅h− 1⋅bar− 1
Alumina Sheet 1000 ℃ – – 5 nm PWF: 72 L⋅m− 2⋅h− 1⋅bar− 1
Dye/salt mixture: 30 ~ 40 [77]
L⋅m− 2⋅h− 1⋅bar− 1
ZrO2 Sheet 600 ~ 900 ℃ – – Mesoporous PWF: 280 L⋅m− 2⋅h− 1⋅bar− 1
Molecule retention (40–50 [78]
kDa)
TiO2 Sheet 850 ℃ 58 % – 70 ~ 110 nm PWF: 1000 ~ 2000 Sodium alginate (50 mg/L) [52]
L⋅m− 2⋅h− 1⋅bar− 1
Alumina Sheet 1100 ~ 1400 ℃ 40 ~ 55 % 10 ~ 70 80 ~ 98 nm PWF: 3640 L⋅m− 2⋅h− 1⋅bar− 1
O/W emulsions (100, 200 mg/ [79]
MPa L): 1000 L⋅m− 2⋅h− 1⋅bar− 1
Alumina Gradient 1050 ℃ – – 108 nm PWF: 1220 L⋅m− 2⋅h− 1⋅bar− 1
CeO2 suspension (170 nm): [80]
pore 850 L⋅m− 2⋅h− 1⋅bar− 1
Alumina/ Sheet 1300 ℃ 24.5–28 % 32.5 ~ 75 250 nm PWF: 9754 L⋅m− 2⋅h− 1⋅bar− 1
O/W emulsion (1000 mg/L): [81]
whisker MPa 1000 L⋅m− 2⋅h− 1⋅bar− 1
− 2 − 1 − 1
Alumina Sheet 950 ℃ – – 7 nm PWF: 230 L⋅m ⋅h ⋅bar PVA wastewater: 60 [71]
L⋅m− 2⋅h− 1⋅bar− 1
− 2 − 1 − 1
Alumina- Sheet 1300 ℃ 24.92 % 97 MPa 120 nm PWF: 4235 L⋅m ⋅h ⋅bar O/W emulsion (1000 mg/L): [47]
mullite 547 L⋅m− 2⋅h− 1⋅bar− 1
Mullite Sheet 1400 ℃ 59.0 %, 14.9 MPa, 2.7 μm GF: 69.25 ~ 150.19 m3⋅m-2⋅h− 1⋅ Dust-laden gas (0.3, 2.5 μm): [57]
whisker 61.1 % 13.1 MPa kPa− 1 1 ~ 2 kPa
Mullite – 1250 ~ 1400 ℃ 47 % – – – – [62]
Al2O3/ – 1200 ℃ – – 190 nm PWF: 351 L⋅m− 2⋅h− 1⋅bar− 1 O/W emulsion (500 mg/L): [59]
α-quartz 190 L⋅m− 2⋅h− 1⋅bar− 1

can also be applied successfully, showing great potential application different temperatures. The shrinkage of co-sintered layers can be
prospects. characterized using differential thermal dilatometry. Zou et al. [44]
systematically analyzed the shrinkage difference between a fly-ash
4.3. Main research foci of co-sintered ceramic membranes support and an alumina MF membrane layer during the co-sintering
process. As shown in Fig. 9a, the shrinkage rate of the fly-ash support
Based on the above discussion of ceramic MF and UF membranes exceeded that of the α-Al2O3 MF layer, although both rates increased
formed through the co-sintering process, this review identifies the sin­ with increasing temperature. Accordingly, the ceramics underwent
tering properties, interfacial strength, and sintering consumption as the bending deformation in the direction of the support after co-sintering
main research foci of co-sintered ceramic membranes. These three foci (Fig. 9b), leading to membrane cracks (Fig. 9c). To better match the
are discussed below. sintering shrinkage between the support and MF layers and improve the
feasibility of co-sintering, Zou et al. [44] added rigid mullite fibers as a
4.3.1. Sintering properties of co-sintered layers hindrance agent to the fly-ash support. After co-sintering at 1050 ◦ C, the
Compatible sintering properties of the layers are prerequisite to the MF membrane was defect-free with an average pore size of 100 nm. Gao
co-sintering process. Two factors are especially important: sintering et al. [59] prepared piezoelectric quartz MF membrane by co-sintering a
temperature and sintering shrinkage. Sintering shrinkage is caused by green porous quartz support and a thin alumina membrane layer.
phase transformation and the formation of sintering necks at high sin­ Aqueous gel casting yielded a quartz support with an obviously porous
tering temperatures. Therefore, to obtain defect-free membranes via co- structure and high flexure strength, and simple dip coating formed a thin
sintering, we must study the sintering behaviors of different materials at alumina membrane on the green support. Owing to the rigid alumina

10
D. Zou et al. Separation and Purification Technology 338 (2024) 126441

Fig. 9. Shrinkage difference analysis of co-sintered layers for membranes with different pore sizes. Co-sintering of fly ash support and alumina MF layer for
composite membranes [44]: (a) Shrinkage difference, (b) analysis of sintering stress, (c) membrane cracks after co-sintering; Co-sintering of fly ZrO2 UF layer and
alumina MF layer on the rigid support [49]: (d) Shrinkage difference, (e) analysis of sintering stress, (f) membrane cracks after co-sintering; Co-sintering of bi-layer
alumina UF membrane with pores size of 5 nm on the rigid support [42]: (g) Shrinkage difference, (h) schematic diagram of co-sintering process, (i) membrane cracks
after co-sintering.

particles and the three-dimensional network generated by gel casting, while submicron alumina particles are added to the UF layer, promoting
the shrinkage behavior of the membrane reduced as the sintering tem­ transition from the γ phase to the α phase at low temperatures and
perature increased to 1250 ◦ C, indicating a rigid support. The sintering alleviating the shrinkage. Fig. 9g plots the shrinkage-rate difference
stress generated by the green quartz support minimally affected the between the top and sub-layer materials versus sintering temperature.
membrane layer, indicating that the sintering was constrained. How­ The sintering shrinkages of the top-layer and sub-layer materials showed
ever, the shrinkage rate between the support and membrane layer must large differences. The sintering process of the top-layer material could
be appropriately controlled. Co-sintering a thin alumina layer and a be divided into three stages: decomposition of bound water and organic
quartz support at 1200 ◦ C successfully formed a piezoelectric quartz MF matter, phase transformation, and alumina diffusion, which accounts for
membrane with a narrow pore-size distribution (180–237 nm). the large shrinkage rate. Mutual doping effectively narrowed the
The sintering shrinkage behavior of UF layers largely differs from shrinkage rate between the top-layer and sub-layer materials. Fig. 9h is a
that of MF layers, as the types and particle sizes of the materials differ schematic of the sintering analysis. The boehmite sol dopant in the MF
between MF and UF layers. Zou et al. [66] analyzed the shrinkage be­ layer enhances the bonding strength between the support and MF layer,
haviors and stress distributions of a ZrO2 UF layer and an Al2O3 MF layer whereas the submicron alumina particles doped in the UF layer improve
during the co-sintering process. As shown in Fig. 9d, the shrinkage rate the shrinkage resistance and promote the γ-to-α phase transition at low
of the UF layer exceeded that of the MF layer from 950 ◦ C to 1300 ◦ C, sintering temperature (1000 ◦ C). After doping the MF layer with 2- wt%
leading to higher stress in the UF layer. The top layer was subjected to boehmite sol and the top layer with 4- wt% submicron alumina particles,
tensile stress caused by the rigid substrate, whereas the sub-layer was the UF membrane presented a pore size of 5 nm and a smooth and
subjected to compressive stress induced by the shrinkage-rate difference uniform surface morphology (Fig. 9i).
during the co-sintering process (Fig. 9e). Tension stress degrades the In conclusion, the sintering properties are the most important pa­
integrity of the membrane surface. Obvious cracks on the membrane rameters in the co-sintering process of ceramic MF membranes. Differ­
surface appeared at sintering temperatures around 1150 ◦ C (Fig. 9f). ential thermal dilatometry can effectively characterize the sintering
However, the shrinkage behaviors of the MF layer and support were not properties of co-sintered membrane layers. A shrinkage analysis can
large. Therefore, the integrity of the resulting membrane can be finely assess the possibility of the co-sintering technique. Based on the
tuned by controlling the sintering temperature. When co-sintered at shrinkage rate as a function of sintering temperatures, various strategies
1050 ◦ C, the mean pore size of the bi-layer membrane was 50 nm. could be employed depending on the acceleration or the alleviation of
Decreasing the pore size of the UF layer enlarges the shrinkage dif­ the sintering process. After that, scanning electron microscopy (SEM)
ference between the membrane layers. Zou et al [69] fabricated a bi- imaging and separation performance can assess the performances of
layer α-Al2O3 UF membrane using a new co-sintering process with membranes fabricated via the co-sintering process.
mutual doping. In the mutual doping process, a certain amount of
boehmite sol is added to the MF layer to accelerate the sintering process

11
D. Zou et al. Separation and Purification Technology 338 (2024) 126441

4.3.2. Interfacial strength between the two layers of a co-sintered ceramic scratch techniques [88,89]. A nano-scratch tester is schematized in
membrane Fig. 10a and 10b. The nano-scratch test is an auxiliary nano-indentation
As mentioned above, the interfacial strength between the co-sintered method that determines the adhesion strength of an interface by moving
layers is another important factor of co-sintered ceramic membranes. the sample perpendicularly to the scratch probe. The shear stress ac­
The interfacial strength is always lower between two co-sintered layers cumulates until the membrane layer peels. Beyond the critical load, the
than between a membrane and a rigid support in constrained sintering, membrane layer begins to fail. The interfacial strength between the
because the sintering stress is unevenly distributed through the layered membrane layer and ceramic support can be evaluated by comparing the
structure. The interfacial strengths of co-sintered layers are influenced force–displacement curves recorded during a scratch test. The charac­
by the nature of the bonding along the joined interface, residual stresses teristic critical load, considered as the interfacial strength between the
evolved during processing, and the sintered properties of the individual membrane layer and support, manifests as a discontinuity in the force­
layers [82]. Finite element simulations by Suresh et al. [83] confirmed a –displacement curve when the membrane layer peels from the support
non-uniform stress distribution across multi-layered structures. The or membrane surface. Zou et al. [70] investigated how the sintering
stress was maximized at the interface between the co-sintered layers. method influences the interfacial strength between MF and UF layers.
Meanwhile, the membrane thickness and sintering temperature affected Using a nano-scratch tester, they obtained critical loads of 59.5 and 56
the interfacial stress between two co-sintered layers and decisively mN for Ag-nanoparticle doped and conventional ceramic UF mem­
determined the strength of the interface. The characterization and branes, respectively (Fig. 10c–10g), signifying that ceramic UF mem­
modulation of the interfacial strength are detailed below. branes prepared by co-sintering have high interfacial strength. Shi et al.
Jong et al. [84] prepared multi-layered α-alumina hollow fibers by [72] characterized the interfacial strength of α-alumina UF membranes
combining polymer wet spinning with co-sintering. Compared to the prepared by co-sintering an intermediate layer (GO sacrificial layer or
inner support, the outer separation layer showed considerable reduction gel interlayer) and an ultrathin α-alumina separation layer. The
in geometric size and layer thickness, facilitating the increase in scratching-depth signal of the ceramic membrane signified a high
compressive stress. Hence, the bonding between the outer layer and interfacial strength with a critical load of 17.3 mN.
inner support strengthened, conferring a surprisingly high interfacial In summary, the interfacial strength between co-sintered layers can
strength. However, the interfacial strength between the two layers was be effectively determined from the mass loss of a ceramic membrane
only qualitatively determined from cross-sectional SEM images and a after ultrasonic treatment or by analyzing in-situ nano-indentation/
quantitative analysis was lacking. To quantitatively evaluate the inter­ scratch results. However, if the thickness of the membrane layer is
facial strength, Feng et al. [39,53] measured the mass loss and charac­ much higher, the in-situ nano-indentation/scratch was not appropriate
terized the surface-morphology changes of the co-sintered membrane as the probe was easily destroyed. Therefore, it is meaningful to design a
before and after 5 min of ultrasonic treatment (160 W and 40 kHz). If the universality research approach to calculate the interfacial strength for
interfacial strength between the co-sintered layers was sufficiently high, co-sintered ceramic membranes. In addition, how to enhance the
the mass-loss and surface-morphology changes after ultrasonic treat­ interfacial strength between the support and membrane layer is pre­
ment could be neglected. Conversely, a co-sintered layer with low requisite as the harsh application conditions in industrial wastewaters.
interfacial strength exhibited large mass losses and morphological dif­
ferences after ultrasonication. Based on quantitative measurements, 4.3.3. Analysis of sintering energy consumption
they also systematically analyzed the effects of sintering temperature The integrities and stabilities of membranes prepared through the co-
and membrane-layer thickness on the stability and integrity of a co- sintering process have been analyzed in detail. However, cost is an
sintered ZrO2/Al2O3 bi-layer membrane. First, they observed that the essential factor when designing ceramic membranes for engineering
membrane layers co-sintered at 1150 ◦ C peeled off from the support applications. The cost calculation of separation membranes for practical
after ultrasonic treatment, indicating an unsatisfactory interfacial industrial applications involves both the membrane cost (material and
strength owing to low shrinkage of the sub-layer Al2O3 membrane. The processing costs) and maintenance cost. The energy consumption of
small compressive stress generated by sintering shrinkage could not sintering commonly accounts for 60 % of the total ceramic processing
promote sintering of the membrane. After co-sintering at 1200 ◦ C and cost. Multi-layered ceramic membranes, which require repeated high-
1250 ◦ C, the interfacial strength of the bi-layer ZrO2/Al2O3 membrane temperature sintering procedures for each membrane layer, increase
was slightly improved because the larger sintering shrinkage induced the sintering energy consumption so their applications are limited. Co-
sufficient compressive strength. The membrane thickness is also a crit­ sintering technology simplifies the preparation of asymmetric ceramic
ical factor that influences the interfacial strength between the mem­ membranes, especially at the sintering stage. Co-sintering is a state-of-
brane layers and support. According to a theoretical model proposed by the-art method that reduces the sintering energy consumption and
Cai et al. [85,86] and Chartier et al. [87], a thicker top-layer membrane preparation period. The cost reduction of the co-sintering process is
exerts a larger compressive stress on the sub-layer membrane than a thin commonly questioned by researchers but is not difficult to calculate. For
top layer, thereby increasing the interfacial strength between the example, the production cost of an alumina-based ceramic membrane is
membrane layers. They et al. [39,53] observed a similar mass-loss $500 m− 2, including a $300 m− 2 (60 %) allocation to sintering energy
change in the bi-layer membranes before and after the ultrasonic consumption. The multiple sintering requirements of an asymmetric
treatment. The mass loss decreased with increasing top-layer thickness membrane will exponentially increase the energy consumption cost. If
but increased with increasing sub-layer thickness. The compressive multi-layer membranes could be produced using a one-step sintering
stress was distributed across the cross section of the sub-layer membrane process, the energy consumption cost could be limited to realize a
and was maximized at the interface between the membrane layers. This potentially cost-effective ceramic-membrane preparation method for
beneficial compressive stress decreased gradually along the direction engineering applications.
perpendicular to the rigid support. Hence, a thin sub-layer is expected to
ensure the integrity of the top-layer surface morphology. This expecta­ 5. Conclusions and recommendations
tion was confirmed by Feng’s experimental results. Mass-loss and
surface-morphology changes before and after ultrasonic treatment have Co-sintering technology has facilitated the development of func­
become the main methods for characterizing the interfacial strengths of tional ceramics such as SOFCs, LTCCs, and ceramic membranes. A deep
co-sintered ceramic composite membranes, as mentioned by Zou et al. literature analysis uncovers some experimental methods and analytical
[69] and Gao et al. [59]. ideas for preparing ceramic membranes. Co-sintering was performed on
The interfacial adhesion strengths of ceramic-supported composite porous alumina supports and low-temperature-sintered supports. Bi-
membranes have also been probed using in-situ nano-indentation/ layer α-Al2O3 and ZrO2/α-MF membranes have been realized on

12
D. Zou et al. Separation and Purification Technology 338 (2024) 126441

Fig. 10. The characterization of interfacial strength between the co-sintered layers. (a) Set-up of nano-indentation/scratch, (b) diagram of the nano-indentation/
scratch [88], (c) scratch depth–displacement curve and scratch load–displacement curve [45], (d) and (e): cross-sectional structure of the ceramic membranes,
(f) and (g): scratch morphology of the ceramic membrane. Note that the CCUM refers to the conventional ceramic UF membrane; ACUF refers to the co-sintered
ceramic UF membrane.

macroporous ceramic supports. The preparation of Al2O3 MF mem­ ceramic MF membrane products.
branes via co-sintering and the influence of the limited change behavior The future development of co-sintered ceramic membranes must
of the membrane layer on the microstructure during the heat treatment address the following problems:
process were investigated, along with the effects of binding strength on
the sintering system. A quantitative relation model between the 1. The present co-sintered ceramic membranes are mainly limited to
sintering-process parameters and microstructures was established to MF and UF membranes. NF membranes have not been fabricated by
quantitatively predict and control microstructural changes in the loaded the co-sintering technique because the sintering stress between the
membrane during the restricted sintering process. Based on the in-depth NF and UF layers is excessively large. Matching the sintering per­
study of restricted sintering, a co-sintering route that exploits the pres­ formances of UF and NF materials is an important future project.
sure stress between the layers during restricted sintering was proposed, 2. Conventional research studies have focused on co-sintered ceramic
and co-sintering of ceramic UF membranes was realized by optimizing membranes with low energy consumption. The next-generation co-
the design of the transition layer material. To reduce the cost of pre­ sintered ceramic membranes should exhibit high anti-fouling per­
paring ceramic membranes, many researchers adopted low-cost raw formance. For this purpose, lead–zirconate–titanate and other
materials and co-sintering technology. Finally, the ceramic-membrane ceramic membranes must be investigated in detail.
co-sintering technology must be industrialized for preparing low-cost

13
D. Zou et al. Separation and Purification Technology 338 (2024) 126441

3. The sintering stress among multiple membrane layers should be [9] Y. Han, D. Zou, Y. Kang, Z. Zhong, W. Xing, One-step sintering process for high-
performance SiC membranes for efficient filtration of dust-laden gas, J. Membr. Sci.
investigated as the theory is immature and has not been applied to
692 (2024) 122265.
co-sintered membranes, especially to more than three membrane [10] S.H. Lee, G.L. Messing, E.R. Twiname, A. Mohanram, C.A. Randall, D.J. Green, T.T.
layers. P. Ttp, Co-sintering of multilayer ceramics, Euro Ceramics Vii, Pt 1-32002, pp. 257-
4. Relevant standards for the interfacial strength of co-sintered mem­ 260. https://doi.org/10.4028/www.scientific.net/KEM.206-213.257.
[11] C. Pascal, A. Thomazic, A. Antoni-Zdziobek, J.-M. Chaix, Co-sintering and
brane layers should be constructed. For example, what constitutes microstructural characterization of steel/cobalt base alloy bimaterials, J. Mater.
the optimal mass loss and interfacial strength should be unambigu­ Sci. 47 (4) (2012) 1875–1886, https://doi.org/10.1007/s10853-011-5976-8.
ously defined. New methods for defining the interfacial strength [12] Y. Dong, H. Wu, F. Yang, S. Gray, Cost and efficiency perspectives of ceramic
membranes for water treatment, Water Res. 220 (2022) 118629, https://doi.org/
should also be developed. 10.1016/j.watres.2022.118629.
5. Last but not least, the co-sintering technique should be industrial­ [13] S.K. Hubadillah, M.H.D. Othman, T. Matsuura, A.F. Ismail, M.A. Rahman,
ized. At present, co-sintering is limited to plate or tubular ceramic Z. Harun, J. Jaafar, M. Nomura, Fabrications and applications of low cost ceramic
membrane from kaolin: a comprehensive review, Ceram. Int. 44 (5) (2018)
membranes. The fabrication of multi-channel ceramic membranes or 4538–4560, https://doi.org/10.1016/j.ceramint.2017.12.215.
hollow fiber ceramic membranes should be considered using the co- [14] S. Mestre, A. Gozalbo, M.M. Lorente-Ayza, E. Sánchez, Low-cost ceramic
sintering technique for potential industrialization. membranes: a research opportunity for industrial application, J. Eur. Ceram. Soc.
39 (12) (2019) 3392–3407, https://doi.org/10.1016/j.jeurceramsoc.2019.03.054.
6. The industrial applications of the co-sintered ceramic membranes are [15] M.H. Qiu, X.L. Kong, K.Y. Fu, S.X. Han, X.Y. Gao, X.F. Chen, Y.Q. Fan, Optimization
not mature. Therefore, more efforts should be emphasized on the of microstructure and geometry of hydrophobic ceramic membrane for SO2
application processes to improve the prospects of the co-sintered absorption from ship exhaust, AIChE J 65 (1) (2019) 409–420, https://doi.org/
10.1002/aic.16416.
ceramic membranes.
[16] J. Opfermann, J. Blumm, W.D. Emmerich, Simulation of the sintering behavior of a
ceramic green body using advanced thermokinetic analysis, Thermochim Acta 318
CRediT authorship contribution statement (1) (1998) 213–220.
[17] P. Wang, P. Huang, N. Xu, J. Shi, Y.S. Lin, Effects of sintering on properties of
alumina microfiltration membranes, J. Membr. Sci. 155 (155) (1999) 309–314.
Dong Zou: Writing – review & editing, Writing – original draft, [18] G. Kerbart, C. Maniere, C. Harnois, S. Marinel, Master sintering curve with
Investigation, Funding acquisition, Formal analysis, Data curation, dissimilar grain growth trajectories: a case study on MgAl2O4, J. Eur. Ceram. Soc.
41 (1) (2021) 1048–1051, https://doi.org/10.1016/j.jeurceramsoc.2020.09.003.
Conceptualization. Zhaoxiang Zhong: Writing – review & editing, Su­ [19] W. Qin, C. Peng, J. Wu, A sacrificial-interlayer technique for single-step coating
pervision, Project administration. Yiqun Fan: Writing – review & edit­ preparation of highly permeable alumina membrane, Ceram. Int. 43 (1) (2017)
ing, Supervision. 901–904, https://doi.org/10.1016/j.ceramint.2016.09.206.
[20] W. Zhu, Y. Liu, K. Guan, C. Peng, W. Qiu, J. Wu, Integrated preparation of alumina
microfiltration membrane with super permeability and high selectivity, J. Eur.
Ceram. Soc. 39 (4) (2019) 1316–1323, https://doi.org/10.1016/j.
Declaration of competing interest jeurceramsoc.2018.10.022.
[21] W. Zhu, Y. Liu, K. Guan, C. Peng, J. Wu, Design and optimization of ceramic
membrane structure: from the perspective of flux matching between support and
The authors declare that they have no known competing financial membrane, Ceram. Int. 47 (9) (2021) 12357–12365, https://doi.org/10.1016/j.
interests or personal relationships that could have appeared to influence ceramint.2021.01.088.
the work reported in this paper. [22] T.J. Garino, H.K. Bowen, Kinetics of constrained-film sintering, J. Am. Ceram. Soc.
73 (2) (1990) 251–257, https://doi.org/10.1111/j.1151-2916.1990.tb06502.x.
[23] J.C. Chang, J.H. Jean, Self-constrained sintering of mixed low-temperature-cofired
Acknowledgements ceramic laminates, J. Am. Ceram. Soc. 89 (3) (2006) 829–835, https://doi.org/
10.1111/j.1551-2916.2005.00820.x.
[24] S.Y. Tzeng, J.H. Jean, Stress development during constrained sintering of alumina/
This work was supported by National Key Research and Develop­ glass/alumina sandwich structure, J. Am. Ceram. Soc. 85 (2) (2002) 335–340.
ment Project of China (2021YFB3801300), National Natural Science [25] G.W. Scherer, T. Garino, Viscous sintering on a rigid substrate, J. Am. Ceram. Soc.
Foundation of China (22208145), Natural Science Foundation of 68 (4) (1985) 216–220, https://doi.org/10.1111/j.1151-2916.1985.tb15300.x.
[26] P.Z. Cai, D.J. Green, G.L. Messing, Constrained densification of alumina/zirconia
Jiangsu Province (Grant No : BK20220350) and Natural Science Foun­ hybrid laminates.1. Experimental observations of processing defects, J. Am. Ceram.
dation of the Jiangsu Higher Education institutions of China Soc. 80 (8) (1997) 1929–1939.
(22KJB530002). [27] M. Qiu, J. Feng, Y. Fan, N. Xu, Pore evolution model of ceramic membrane during
constrained sintering, J. Mater. Sci. 44 (3) (2009) 689–699, https://doi.org/
10.1007/s10853-008-3202-0.
References [28] A. Hajian, A. Artemenko, A. Kromka, S. Schwarz, M. Schneider, K. Dragounova,
M. Adaikkan, C. Zellner, U. Schmid, Impact of sintering temperature on phase
composition, microstructure, and porosification behavior of LTCC substrates,
[1] P. Jarvis, I. Carra, M. Jafari, S.J. Judd, Ceramic vs polymeric membrane
J. Eur. Ceram. Soc. 42 (13) (2022) 5789–5800, https://doi.org/10.1016/j.
implementation for potable water treatment, Water Res. 215 (2022), https://doi.
jeurceramsoc.2022.05.049.
org/10.1016/j.watres.2022.118269.
[29] B. Huang, T. Xia, F. Shang, G. Chen, A new BaB2O4 microwave dielectric ceramic
[2] Y. Wang, B. Ma, M. Ulbricht, Y. Dong, X. Zhao, Progress in alumina ceramic
for LTCC application, J. Eur. Ceram. Soc. 43 (14) (2023) 6107–6111, https://doi.
membranes for water purification: status and prospects, Water Res. 226 (2022),
org/10.1016/j.jeurceramsoc.2023.07.004.
https://doi.org/10.1016/j.watres.2022.119173.
[30] Z. Xiong, W. Xue, M. Li, F. Tan, Y. Chen, H. Yu, Microwave dielectric
[3] D. Zou, Y.M. Lee, Design strategy of poly(vinylidene fluoride) membranes for water
characterization and densification mechanism analysis of CaO-B-SiO2 glass-
treatment, Prog. Polym. Sci. 128 (2022) 101535, https://doi.org/10.1016/j.
ceramic Al2O3 composites for LTCC applications, J. Am. Ceram. Soc. 107 (1)
progpolymsci.2022.101535.
(2024) 234–243, https://doi.org/10.1111/jace.19436.
[4] C. Li, W. Sun, Z. Lu, X. Ao, S. Li, Ceramic nanocomposite membranes and
[31] R. Zuo, E. Aulbach, J. Rödel, Shrinkage-free sintering of low-temperature cofired
membrane fouling: A review, Water Res. 175 (2020), https://doi.org/10.1016/j.
ceramics by loading dilatometry, J. Am. Ceram. Soc. 87 (3) (2004) 526–528.
watres.2020.115674.
[32] C.C. Huang, J.H. Jean, Stress required for constrained sintering of a ceramic-filled
[5] R.B. Merlet, M.-A. Pizzoccaro-Zilamy, A. Nijmeijer, L. Winnubst, Hybrid ceramic
glass composite, J. Am. Ceram. Soc. 87 (8) (2004) 1454–1458, https://doi.org/
membranes for organic solvent nanofiltration: State-of-the-art and challenges,
10.1111/j.1551-2916.2004.01454.x.
J. Membr. Sci. 599 (2020), https://doi.org/10.1016/j.memsci.2020.117839.
[33] S.Y. Gomez, D. Hotza, Current developments in reversible solid oxide fuel cells,
[6] Q. Gu, T.C.A. Ng, W. Poh, C.H. Kirk, Z. Lyu, L. Zhang, J. Wang, H.Y. Ng, 3D spray-
Renew. Sustain. Energy Rev. 61 (2016) 155–174, https://doi.org/10.1016/j.
coated gradient profile ceramic membranes enables improved filtration
rser.2016.03.005.
performance in aerobic submerged membrane bioreactor, Water Res 220 (2022)
[34] S.Z. Golkhatmi, M.I. Asghar, P.D. Lund, A review on solid oxide fuel cell durability:
118661, https://doi.org/10.1016/j.watres.2022.118661.
Latest progress, mechanisms, and study tools, Renew. Sustain. Energy Rev. 161
[7] M.B. Asif, Z. Zhang, Ceramic membrane technology for water and wastewater
(2022), https://doi.org/10.1016/j.rser.2022.112339.
treatment: A critical review of performance, full-scale applications, membrane
[35] M.A. Laguna-Bercero, A. Larrea, J.I. Peña, R.I. Merino, V.M. Orera, Structured
fouling and prospects, Chem. Eng. J. 418 (2021), https://doi.org/10.1016/j.
porous Ni- and Co-YSZ cermets fabricated from directionally solidified eutectic
cej.2021.129481.
composites, J. Eur. Ceram. Soc. 25 (8) (2005) 1455–1462, https://doi.org/
[8] T.C.A. Ng, Z. Lyu, Q. Gu, L. Zhang, W.J. Poh, Z. Zhang, J. Wang, H.Y. Ng, Effect of
10.1016/j.jeurceramsoc.2005.01.025.
gradient profile in ceramic membranes on filtration characteristics: Implications
for membrane development, J. Membr. Sci. 595 (2020) 117576, https://doi.org/
10.1016/j.memsci.2019.117576.

14
D. Zou et al. Separation and Purification Technology 338 (2024) 126441

[36] M.A. Laguna-Bercero, R. Campana, A. Larrea, J.A. Kilner, V.M. Orera, Performance [60] Y. Zhao, Y. Gu, B. Liu, Y. Yan, C. Shan, J. Guo, S. Zhang, C.D. Vecitis, G. Gao,
and aging of microtubular YSZ-based solid oxide regenerative fuel cells, Fuel Cells Pulsed hydraulic-pressure-responsive self-cleaning membrane, Nature 608 (7921)
11 (1) (2011) 116–123, https://doi.org/10.1002/fuce.201000069. (2022) 69-+, https://doi.org/10.1038/s41586-022-04942-4.
[37] D. Panthi, B. Choi, Y. Du, A. Tsutsumi, Lowering the co-sintering temperature of [61] S. Li, Y. Li, C. Wei, P. Wang, P. Gao, L. Zhou, G. Wen, One step co-sintering of
cathode-electrolyte bilayers for micro-tubular solid oxide fuel cells, Ceram. Int. 43 silicon carbide ceramic membrane with the aid of boron carbide, J. Eur. Ceram.
(14) (2017) 10698–10707, https://doi.org/10.1016/j.ceramint.2017.05.003. Soc. 41 (2) (2021) 1181–1188, https://doi.org/10.1016/j.
[38] D. Zou, Y. Fan, State-of-the-art developments in fabricating ceramic membranes jeurceramsoc.2020.09.065.
with low energy consumption, Ceram. Int. 47 (11) (2021) 14966–14987, https:// [62] G. Chen, Y. Yangyan, L. Guo, J. Guo, X. Zhang, Ge, Co-sintered synthesis of needle-
doi.org/10.1016/j.ceramint.2021.02.195. like mullite porous membrane materials, Desalin. Water Treat. 76 (2017) 77–82,
[39] J. Feng, M. Qiu, Y. Fan, N. Xu, The effect of membrane thickness on the co- https://doi.org/10.5004/dwt.2017.20569.
sintering process of bi-layer ZrO2/Al2O3 membrane, J. Membr. Sci. 305 (1–2) [63] J. Liu, C. Tian, H. Xiao, W. Guo, P. Gao, J. Liang, Effect of B4C on co-sintering of
(2007) 20–26, https://doi.org/10.1016/j.memsci.2007.06.052. SiC ceramic membrane, Ceram. Int. 45 (3) (2019) 3921–3929, https://doi.org/
[40] M. Qiu, S. Fan, Y. Cai, Y. Fan, N. Xu, Co-sintering synthesis of bi-layer titania 10.1016/j.ceramint.2018.11.065.
ultrafiltration membranes with intermediate layer of sol-coated nanofibers, [64] D. Hotza, M. Di Luccio, M. Wilhelm, Y. Iwamoto, S. Bernard, J.C.D. da Costa,
J. Membr. Sci. 365 (1–2) (2010) 225–231, https://doi.org/10.1016/j. Silicon carbide filters and porous membranes: A review of processing, properties,
memsci.2010.09.005. performance and application, J. Membr. Sci. 610 (2020), https://doi.org/10.1016/
[41] Y. Dong, B. Lin, S. Wang, K. Xie, D. Fang, X. Zhang, H. Ding, X. Liu, G. Meng, Cost- j.memsci.2020.118193.
effective tubular cordierite micro-filtration membranes processed by co-sintering, [65] J. Liu, W. Riess, C. Munch, G. Ziegler, Ttp, Development of a co-firing technique for
J. Alloy. Compd. 477 (1–2) (2009) L35–L40, https://doi.org/10.1016/j. the production process of ceramic ultrafiltration membranes, Euro Ceramics Vii, Pt
jallcom.2008.11.061. 1-32002, pp. 1989-1992.
[42] D. Zou, M.H. Qiu, X.F. Chen, Y.Q. Fan, One-step preparation of high-performance [66] D. Zou, Z.Y. Meng, E. Drioli, X.W. Da, X.F. Chen, M.H. Qu, Y.Q. Fan, Design and
bilayer alpha-alumina ultrafiltration membranes via co-sintering process, efficient construction of bilayer Al2O3/ZrO2 mesoporous membranes for effective
J. Membr. Sci. 524 (2017) 141–150, https://doi.org/10.1016/j. treatment of suspension systems, Ind. Eng. Chem. Res. 59 (10) (2020) 4721–4731,
memsci.2016.11.025. https://doi.org/10.1021/acs.iecr.9b06568.
[43] D. Zou, X.B. Ke, M.H. Qiu, X.F. Chen, Y.Q. Fan, Design and fabrication of whisker [67] J.J. Wen, M.H. Qiu, X.F. Chen, X.W. Da, Y.Q. Fan, Fabrication of a dual-layer
hybrid ceramic membranes with narrow pore size distribution and high ceramic mesoporous membrane with high flux via a co-sintering process,
permeability via co-sintering process, Ceram. Int. 44 (17) (2018) 21159–21169, Microporous Mesoporous Mater. 334 (2022), https://doi.org/10.1016/j.
https://doi.org/10.1016/j.ceramint.2018.08.161. micromeso.2022.111764.
[44] D. Zou, M.H. Qiu, X.F. Chen, E. Drioli, Y.Q. Fan, One step co-sintering process for [68] Z. Jin, Y. Shen, X. Da, X. Chen, M. Qiu, Y. Fan, Construction of high-performance
low-cost fly ash based ceramic microfiltration membrane in oil-in-water emulsion CeO2 ultrafiltration membrane for high-temperature dye/salt separation,
treatment, Sep. Purif. Technol. 210 (2019) 511–520, https://doi.org/10.1016/j. J. Membr. Sci. 637 (2021), https://doi.org/10.1016/j.memsci.2021.119608.
seppur.2018.08.040. [69] D. Zou, M. Qiu, X. Chen, Y. Fan, One-step preparation of high-performance bilayer
[45] D. Zou, X. Chen, E. Drioli, X. Ke, M. Qiu, Y. Fan, Facile co-sintering process to α-alumina ultrafiltration membranes via co-sintering process, J. Membr. Sci. 524
fabricate sustainable antifouling silver nanoparticles (AgNPs)-enhanced tight (2017) 141–150, https://doi.org/10.1016/j.memsci.2016.11.025.
ceramic ultrafiltration membranes for protein separation, J. Membr. Sci. 593 [70] D. Zou, X.F. Chen, E. Drioli, X.B. Ke, M.H. Qiu, Y.Q. Fan, Facile co-sintering process
(2020) 117402, https://doi.org/10.1016/j.memsci.2019.117402. to fabricate sustainable antifouling silver nanoparticles (AgNPs)-enhanced tight
[46] J. Wen, C. Yang, X. Chen, M. Qiu, Y. Fan, Effective and efficient fabrication of high- ceramic ultrafiltration membranes for protein separation, J. Membr. Sci. 593
flux tight ZrO2 ultrafiltration membranes using a nanocrystalline precursor, (2020), https://doi.org/10.1016/j.memsci.2019.117402.
J. Membr. Sci. 634 (2021), https://doi.org/10.1016/j.memsci.2021.119378. [71] W. Shi, X. Hu, M. Qiu, Z. Jin, X. Chen, Y. Fan, Low temperature preparation of
[47] Y. Gao, G. Xu, P. Zhao, L. Liu, E. Zhang, One step co-sintering synthesis of gradient high-flux α-alumina tight ultrafiltration membrane by modified co-sintering
ceramic microfiltration membrane with mullite/alumina whisker bi-layer for high process, Sep. Purif. Technol. 306 (2023), https://doi.org/10.1016/j.
permeability oil-in-water emulsion treatment, Sep. Purif. Technol. 305 (2023), seppur.2022.122524.
https://doi.org/10.1016/j.seppur.2022.122400. [72] W. Shi, C. Yang, M. Qiu, X. Chen, Y. Fan, A new method for preparing α-alumina
[48] M.A.B. Pauzan, M.H.D. Othman, N.J. Ismail, M.H. Puteh, A.F. Ismail, M. ultrafiltration membrane at low sintering temperature, J. Membr. Sci. 642 (2022)
A. Rahman, J. Jaafar, Fabrication of zirconia-kaolin dual layer hollow fiber 119992, https://doi.org/10.1016/j.memsci.2021.119992.
membrane: Physical and performance study for industrial wastewater treatment, [73] X.B. Ke, H.Y. Zhu, X.P. Gao, J.W. Liu, Z.F. Zheng, High-performance ceramic
J. Water Process Eng. 41 (2021), https://doi.org/10.1016/j.jwpe.2021.102031. membranes with a separation layer of metal oxide nanofibers, Adv. Mater. 19 (6)
[49] D. Zou, Z. Meng, E. Drioli, X. Da, X. Chen, M. Qiu, Y. Fan, Design and efficient (2007) 785–790, https://doi.org/10.1002/adma.200601984.
construction of bilayer Al2O3/ZrO2 mesoporous membranes for effective treatment [74] J. Feng, Y. Fan, H. Qi, N. Xu, Co-sintering synthesis of tubular bilayer α-alumina
of suspension systems, Ind. Eng. Chem. Res. 59 (10) (2020) 4721–4731, https:// membrane, J. Membr. Sci. 288 (1–2) (2007) 20–27.
doi.org/10.1021/acs.iecr.9b06568. [75] D. Das, N. Kayal, G.A. Marsola, D.G. Parra, M.D.D. Innocentini, Recycling of coal
[50] D. Zou, J. Xu, X. Chen, E. Drioli, M. Qiu, Y. Fan, A novel thermal spraying fly ash for fabrication of elongated mullite rod bonded porous SiC ceramic
technique to fabricate fly ash/alumina composite membranes for oily emulsion and membrane and its application in filtration, J. Eur. Ceram. Soc. 40 (54) (2020)
spent tin wastewater treatment, Sep. Purif. Technol. 219 (2019) 127–136, https:// 2163–2172, https://doi.org/10.1016/j.jeurceramsoc.2020.01.034.
doi.org/10.1016/j.seppur.2019.02.051. [76] D. Zou, W. Fan, J. Xu, E. Drioli, X. Chen, M. Qiu, Y. Fan, One-step engineering of
[51] Q. Hong, X. Weihong, F. Yiqun, Preparation of high purity alumina macroporous low-cost kaolin/fly ash ceramic membranes for efficient separation of oil-water
support at low sintering temperature, J. Chin. Silic. Soc. 38 (2) (2010), pp. emulsions, J. Membr. Sci. 621 (2020) 118954, https://doi.org/10.1016/j.
283–288,293. memsci.2020.118954.
[52] Z. Zhang, Q. Gu, T.C.A. Ng, J. Zhang, X. Zhang, L. Zhang, X. Zhang, H. Wang, H. [77] D. Zou, S. Ni, H. Yao, C. Hu, Z.-X.-N. Low, Z. Zhong, Co-sintering of ceramic
Y. Ng, J. Wang, Hierarchically porous interlayer for highly permeable and fouling- ultrafiltration membrane with gradient pore structures for separation of dye/salt
resistant ceramic membranes in water treatment, Sep. Purif. Technol. 293 (2022), wastewater, Sep. Purif. Technol. 302 (2022), https://doi.org/10.1016/j.
https://doi.org/10.1016/j.seppur.2022.121092. seppur.2022.122030.
[53] J. Feng, Y. Fan, H. Qi, N. Xu, Co-sintering synthesis of tubular bilayer alpha- [78] J. Wen, M. Qiu, X. Chen, X. Da, Y. Fan, Fabrication of a dual-layer ceramic
alumina membrane, J. Membr. Sci. 288 (1–2) (2007) 20–27, https://doi.org/ mesoporous membrane with high flux via a co-sintering process, Microporous
10.1016/j.memsci.2006.09.034. Mesoporous Mater. 334 (2022) 111764, https://doi.org/10.1016/j.
[54] H. Mao, Preparation of piezoelectric ceramic-based porous membrane and its self- micromeso.2022.111764.
cleaning performance, Nanjing Tech University, 2020. [79] J. Ma, W. Chen, J. Qian, A. Shui, B. Du, C. He, Co-pressing and co-sintering
[55] L. Yu, M. Kanezashi, H. Nagasawa, T. Tsuru, Phase inversion/sintering-induced preparation of cost-effective and high-performance asymmetric ceramic membrane
porous ceramic microsheet membranes for high-quality separation of oily for oily wastewater treatment, Sep. Purif. Technol. 323 (2023), https://doi.org/
wastewater, J. Membr. Sci. 595 (2020) 117477, https://doi.org/10.1016/j. 10.1016/j.seppur.2023.124373.
memsci.2019.117477. [80] T. Chen, P. Xu, X. Chen, T. Wang, K. Fu, M. Qiu, Y. Fan, High-flux ceramic
[56] W. Qin, Y. Zhang, J. Wu, Preparation of high-permeance ceramic microfiltration membrane derived from UV-curable slurry for efficient separation of nanoparticles
membranes using a pore-sealing method, RSC Adv. 10 (10) (2020) 5560–5565, suspension, Sep. Purif. Technol. 304 (2023), https://doi.org/10.1016/j.
https://doi.org/10.1039/c9ra09805d. seppur.2022.122331.
[57] D. Zou, Y. Gong, Y. Liu, Z.-X.-N. Low, Z. Zhong, W. Xing, One-step co-sintering of [81] Y. Gao, G. Xu, P. Zhao, L. Liu, E. Zhang, S. Jin, High-flux whisker layer ceramic
hierarchical mullite whisker/fiber membranes with gradient pore structures for membrane prepared by gel spin-coating method for low-pressure oil/water
effective filtration of dust-laden gas, J. Membr. Sci. (2023), https://doi.org/ emulsion filtration, Ceram. Int. 49 (20) (2023) 32727–32738, https://doi.org/
10.1016/j.memsci.2022.121143. 10.1016/j.ceramint.2023.07.241.
[58] J. Gao, H. Cao, X. Hu, H. Mao, X. Chen, M. Qiu, H. Verweij, Y. Fan, Piezoelectric [82] Y. Boonyongmaneerat, C.A. Schuh, Contributions to the interfacial adhesion in Co-
porous α-quartz membrane by aqueous gel-casting with enhanced antifouling and sintered bilayers, Metall. Mater. Trans. A-Phys. Metall. Mater. Sci. 37A (5) (2006)
mechanical properties, J. Eur. Ceram. Soc. 43 (1) (2023) 109–120, https://doi.org/ 1435–1442, https://doi.org/10.1007/s11661-006-0088-9.
10.1016/j.jeurceramsoc.2022.09.038. [83] S. Suresh, A.E. Giannakopoulos, M. Olsson, Elastoplastic analysis of thermal
[59] J. Gao, M. Qiu, X. Chen, H. Verweij, Y. Fan, One-step sintering for anti-fouling cycling: Layered materials with sharp interfaces, J. Mech. Phys. Solids 42 (6)
piezoelectric α-quartz and thin layer of alumina membrane, J. Membr. Sci. 667 (1994) 979–1018, https://doi.org/10.1016/0022-5096(94)90081-7.
(2023), https://doi.org/10.1016/j.memsci.2022.121188.

15
D. Zou et al. Separation and Purification Technology 338 (2024) 126441

[84] J. de Jong, N.E. Benes, G.H. Koops, M. Wessling, Towards single step production of [87] T. Chartier, D. Merle, J.L. Besson, Laminar ceramic composites, J. Eur. Ceram. Soc.
multi-layer inorganic hollow fibers, J. Membr. Sci. 239 (2) (2004) 265–269, 15 (2) (1995) 101–107, https://doi.org/10.1016/0955-2219(95)93055-8.
https://doi.org/10.1016/j.memsci.2004.02.039. [88] Y. Hang, G. Liu, K. Huang, W. Jin, Mechanical properties and interfacial adhesion
[85] P.Z. Cai, D.J. Green, G.L. Messing, Constrained densification of alumina/zirconia of composite membranes probed by in-situ nano-indentation/scratch technique,
hybrid laminates, ii: Viscoelastic stress computation, J. Am. Ceram. Soc. 80 (8) J. Membr. Sci. 494 (2015) 205–215, https://doi.org/10.1016/j.
(1997) 1940–1948, https://doi.org/10.1111/j.1151-2916.1997.tb03076.x. memsci.2015.05.076.
[86] P.Z. Cai, D.J. Green, G.L. Messing, Constrained densification of alumina/zirconia [89] W. Wei, S. Xia, G. Liu, X. Gu, W. Jin, N. Xu, Interfacial adhesion between polymer
hybrid laminates, i: Experimental observations of processing defects, J. Am. Ceram. separation layer and ceramic support for composite membrane, AIChE J 56 (6)
Soc. 80 (8) (1997) 1929–1939, https://doi.org/10.1111/j.1151-2916.1997. (2010) 1584–1592, https://doi.org/10.1002/aic.12086.
tb03075.x.

16