Membranes and Membrane Technologies

, Volume 1, Issue 3, pp 127–136 | Cite as

Russian Multilayer Metal–Ceramic Membranes: Structure, Application, and Perspectives

  • V. I. Novikov
  • V. V. Kryachko
  • Yu. I. Tarasov
  • Sainsanaa Tserenchimed
  • A. Yu. AlentievEmail author


The main trends in modifying the structure of bilayer metal-ceramic membranes developed in Russia to fabricate next-generation multilayer metal–ceramic membranes (MMCM) with a wide range of pore sizes from 0.4 to 100 nm and a narrow pore size distribution have been considered. The examples of MMCM structures of various morphologies, the possibilities of further approach to their modification, and possible applications have been shown. Schematic diagrams of existing and prospective membrane units based on metal–ceramic membranes have been demonstrated. The advantages and promising areas of application of such units are discussed.


metal–ceramic membranes nanoporous materials ultrafiltration nanofiltration membrane installations 


Membrane technologies of separation and purification of substances are characterized by high efficiency at low energy and material consumption, mobility, modular structure of membrane units, and ease of management, properties that facilitate their widespread implementation and development. Most membrane processes have been realized using polymer membranes; however, polymer membranes cannot be used in agressive environments leading to rapid degradation of polymeric materials, i.e. in organic solvents, at high temperatures, in acidic or alkaline solutions, and in the presence of strong oxidizing agents or radioactive substances. Under these conditions, ceramic membranes can be effective, but their widespread use is limited due to their brittleness, short service life, low abrasion resistance, and limited morphological diversity. Flexible metal–ceramic membranes, produced according to the technology developed in the early 1990s in Russia on the basis of analysis of domestic and foreign experience, are free from these disadvantages. A bilayer metal–ceramic membrane, designed by Russian researchers under the leadership of the Lenin Prize winner V.N. Lapovok [1, 2], consists of a porous metal or alloy (stainless steel, titanium) layer and a ceramic layer based on oxides, carbides, nitrides, or their composites [3, 4, 5, 6]. The technical characteristics of the new membrane cover the ultra- and microfiltration spectrum and allow solving many modern problems of separation of mixtures, including the processing of liquid radioactive waste and widespread use in the food industry, medicine, biotechnology, and other real sectors of the market. Since the beginning of the 2000s, the ASPECT Association [7, 8] has been producing and introducing ultra- and microfiltration metal–ceramic membranes under the trade name Trumem and Rusmem.

For the first time, the structure of metal–ceramic membranes was created using the technologies of fabricating plastic ceramics and combining porous metals with nanoceramic porous layers [1, 2, 3, 4, 5, 6], which made it possible to develop unique equipment for cleaning solutions of complex chemical and radiochemical composition [8, 9, 10, 11, 12]. These processes and equipment were patented in Russia and other countries, including the United States [2, 3]. Positive opinions by leading foreign research centers and industrial companies that had tested the metal–ceramic membranes determined a significant export potential of the new Russian product [7].

The uniqueness of metal–ceramic membranes is due to the combination of the best qualities of ceramic membranes (thermal and corrosion resistance, resistance to aggressive media, durability) and polymer membranes (flexibility, high performance). The metal–ceramic membrane fabrication technology is capable of creating filter elements of almost any shape (bend radius, 5 mm), including the corrugated one; welding and soldering of filtering elements is also possible. Finally, metal–ceramic membranes are characterized by a narrow pore size distribution in the application range from ultra- to microfiltration [7, 8, 12].

At present, AO Krasnaya Zvezda produces flexible bilayer metal–ceramic membranes under the brand name MKM and membrane units based on them [13]. MKM membranes on a porous stainless steel substrate (nickel or titanium) have a thickness of 200–250 μm with a ceramic layer of about 15 μm in thickness. The ceramic layer consists of TiO2, TiO2/Al2O3, ZrO2, or a composite of two or more of these ceramics and has a calibrated pore size of 60, 140, 200, or 400 nm. The pure water flux of MKM membranes varies from 2800 to 18 000 L/(h m2) with a pressure differential of 2 atm. These membranes have unique radiation resistance and chemical stability and are used for ultrafiltration processes at elevated temperatures (up to 300°C) and in corrosive environments (pH 2–14). The MKM membranes withstand welding and mechanical, chemical, or electrochemical cleaning; have high abrasion resistance; and are capable of working at high pressures up to 200 atm.


The further development of the generation of metal–ceramic membranes is associated with the appearance of three-layer and multilayer metal–ceramic membranes (MMCMs) with an average pore size in the range from 0.4 to 100 nm for nano- and ultrafiltration of liquid media; reverse osmosis, gas and vapor separation, catalytic membrane processes; etc. [3, 6, 14, 15, 16, 17, 18, 19, 20, 21].

The main idea of creating multilayer membranes based on nanoporous materials is the sequential deposition on the substrate of various metal and ceramic layers differing in thickness and pore size (Fig. 1).

Fig. 1.

Schematic representation of the structure of a three-layer metal–ceramic membrane (MMCM). (1) Substrate. Pore size: 2–5 μm. Thickness: 200 μm. Materials: nickel; SS 316L steel, SS 304 steel. (2) Intermediate layer. Pore size: 50–500 nm. Thickness: 5–20 μm. Materials: ZrO2, TiO2, Al2O3. (3) Selective layer. Pore size: 0.4–10 nm. Thickness: 0.1–5 μm. Materials: Ru, Rh, Pt, Pd, Ag, oxides, zeolites, polymers.

Particles that are to form the third layer can be deposited from liquid media on the MKM surface either by introducing nanosized powders into the solution or by using the sol–gel process. In the former case, an electric potential can be used to control the migration of electrically charged particles in electrophoresis and electrodialysis processes [14, 16, 18, 19]. In the latter case, membrane impregnation and the subsequent deposition of nanosized particles, formed during the chemical reaction, in the pores or on the surface of the membrane are generally used [15, 17, 20, 21, 22]. Regardless of the choice of the option, this procedure allows controlling the chemical reaction with the aim to influence the modification of pores on the working side of the substrate. Procedures based on the chemical vapor deposition (CVD) or magnetron sputtering techniques can also be used [23, 24]. The selective layer on the substrate can be a thin (~0.1–5 μm) porous membrane [23] with a calibrated pore size depending both on the size of the particles forming the layer and on the characteristics of the coating material. The selective layer on the substrate may be nonporous as well [24].

Metal–ceramic membranes can also be used as a substrate for fabricating MMCMs with a nonporous selective polymer layer [25]. In this case, the selective layer is applied onto the substrate from a polymer solution.


For testing various methods of modification and fabrication of multilayer metal–ceramic membranes MMCMs, a 316L–TiO2 MKM substrate with a TiO2 ceramic layer and a pore size of 140 nm was used (Fig. 2).

Fig. 2.

Images of the surface of the metal–ceramic membrane MKM 316L–TiO2 at different magnifications: (a) 10 000; (b) 50 000.

The surface of the porous ceramic layer of this membrane is hydrophilic, since the surface of oxide ceramics inevitably contains hydroxyl groups, making it possible to use various methods of chemical modification of the membrane surface and pores to create the third MMCM layer of the required thickness and functionality.

Today, one of the important problems of nuclear energy is the recovery of heavy hydrogen isotopes from water. Currently, there are several processes for recovering heavy isotopes from water: isotopic exchange in the presence of palladium and platinum, electrolysis of water in combination with catalytic isotope exchange between water and hydrogen, column distillation, vacuum freezing of cold vapor followed by thawing, etc. Along with these processes, attempts to search for cheaper methods of recovery are being currently made. At the Novo-Voronezh NPP, studies were carried out on the sorption of tritium and deuterium on a 316L–TiO2–Ti three-layer MMCM (Fig. 3) [23] with a porous titanium layer deposited by magnetron sputtering. Since titanium, along with platinum and palladium, is an effective selective sorbent for these isotopes, the 316L–TiO2–Ti membrane has demonstrated a high sorption capacity for deuterium and tritium. These studies will be continued using platinum and palladium as the third MMCM layer.

Fig. 3.

Structure of the transverse fracture of a three-layer 316L–TiO2–Ti membrane with an average pore size of the titanium layer of 5 nm [23].

Another pressing problem today is the production of ultrahigh-purity hydrogen, including that for fuel cells, from hydrogen-containing mixtures. The most effective method of producing hydrogen with a purity of up to 99.9999% is its purification on membranes made of palladium foil or silver–palladium alloys. To date, palladium foils manufactured by cold rolling have a minimum thickness of ~30 μm. To reduce the palladium membrane thickness, an MMCM with a nonporous palladium layer of 2–3 μm (Fig. 4) was prepared using magnetron sputtering in cooperation with the Voronezh State University; the use of this membrane led to a significant increase in the performance of the hydrogen purification process and also affected the product purity, increasing it to 99.99999% [24]. Such membranes are promising for use in catalytic processes for producing hydrogen from hydrocarbons and alcohols.

Fig. 4.

Structure of the transverse section of a three-layer membrane 316L–TiO2–Pd with a thin (~2–3 μm) nonporous palladium coating [24].

The MKM membrane substrate was also used to fabricate zeolite membranes [17, 20, 21, 22], which hold promise for obtaining pure water, selective separation of aqueous organic media, gas and vapor separation, and membrane catalysis. Promising results were obtained in the case of fabrication of membranes from synthetic zeolite NaA. Its crystal lattice forms a three-dimensional system of pores of a 0.41 nm diameter; in combination with high hydrophilicity (∆\(H_{{{\text{ads}}}}^{^\circ }\)(H2O) = −125 kJ/mol), this system makes the zeolite a promising membrane material for the selective recovery of water from various mixtures. To obtain high-performance zeolite membranes, it is necessary to grow thin layers of fine NaA crystals on a strong and smooth substrate. In collaboration with OOO Unisit (Moscow State University), a procedure was developed for fabricating metal–ceramic membranes with a selective NaA zeolite layer (Fig. 5). The resulting membranes exhibit high selectivity for water, possess chemical and mechanical stability, and have been used to separate water from water–alcohol mixtures [21, 22].

Fig. 5.

Scanning electron microscope image of the cross section of a three-layer zeolite membrane 316L–TiO2–Zeolite NaA with an average pore size of 0.41 nm. The upper light part is the zeolite layer; in bottom left corner, there is a cross section of a polished thin section of a 316L–TiO2 membrane.

Due to the presence of hydroxyl groups on the surface of membrane pores, it is possible to use simple methods for modifying the porous system of metal–ceramic membranes.

For example, joint research was commenced at the Topchiev Institute of Petrochemical Synthesis and the Enikolopov Institute of Synthetic Polymer Materials, Russian Academy of Sciences, on working out a membrane pore modification technology using impregnation with a solution of silica sol nanoparticles followed by heat treatment, which resulted in creation of an aggregated spherical silica nanoparticles on the membrane surface and inside the pores. By selecting the solution concentration, impregnation time, and heat treatment conditions, the desired pore size and pore size distribution can be achieved. One of the silica sol-treated MKM membrane samples was studied by means of X-ray microtomography with numerical analysis, scanning electron–ion microscopy, X-ray photoelectron spectroscopy, and high-resolution transmission electron microscopy. Thus, XPS analysis showed that the SiO2 content in the membrane is maximum on the surface down to 25 nm in depth and decreases with an increase in the scan depth to 125 nm. Micrographs of the surface demonstrate that the pore space between TiO2 particles on the membrane surface is almost completely filled with silica SiO2 (Fig. 6). At the same time, according to numerical modeling data, the connected porosity of the void space is preserved: at a total porosity of 17.3%, the connected porosity is as high as 16.9%. Thus, the membrane obtained according to the impregnation and heat treatment procedure, retain connected porosity. In this case, deposition of silica in pores on the membrane surface is observed, confirming the effectiveness of the inexpensive method for creating new three-layer metal–ceramic membranes with a calibrated pore diameter on the order of several nanometers for operation in aggressive media in nanofiltration, gas and vapor separation, and membrane catalysis processes. The pore size in the silica phase on the membrane surface does not exceed 2 nm according to image analysis of transmission electron microscopy.

Fig. 6.

Types of transverse fracture structures of a three-layer composite membrane 316L–TiO2–SiO2 with different magnifications: (a) 10 000, (b) 50 000.

A promising method for modifying metal–ceramic membranes is the synthesis of heat-resistant hyperbranched polymers, in particular, polyimides, in membrane pores. Having wide possibilities of their synthesis and possibilities for varying the size of free volume voids in them, hyperbranched polymers are not film-forming and, therefore, are inapplicable as membrane materials. The synthesis of such polymers within the pores of metal–ceramic membranes to be covalently bonded with surface hydroxyl groups is a promising method for fabricating of gas or vapor separation membranes and nanofiltration membranes with a separation mechanism intermediate between the solution–diffusion and molecular sieve mechanisms. Preliminary joint studies have demonstrated this approach to have promise and are continued on the basis of the Topchiev and Enikolopov Institutes of the Russian Academy of Sciences.


Krasnaya Zvezda (Red Star) currently produces and operates membrane units based on two-layer metal–ceramic membranes. These units are universal equipment that can be used for treating almost any kind of waste. For a particular task, it is just necessary to replace membranes in the unit with the ones having the appropriate pore size. The degree of purification of a feed fluid (ultrafiltration) and the performance of the equipment always depend on the properties of the filter material. Metal–ceramic membranes made it possible to create high-performance, durable, inexpensive and compact devices that can either completely replace bulky facilities, which require permanent costs for maintenance, or supplement the existing treatment facilities so that significant savings on the maintenance of the already working equipment can be achieved. One of the main advantages is self-cleaning of membranes during operation; i.e., the filtration process is not to be stopped for regenerating or replacing the filter element.

Currently, consumers are offered two types of units, the centrifugal membrane filtration devices (MFC) [10, 11, 13, 26] (Fig. 7) and plate-and-frame membrane units (PMUs) [27] (Fig. 8). The proposed membrane devices are designed for high-performance filtration, microfiltration, and ultrafiltration. With certain design changes and the use of MMCMs, these units can be used for nanofiltration and reverse osmosis.

Fig. 7.

Multidisk crossflow filtration unit FMC using the MKM membrane [13].

Fig. 8.

Schematic diagram of a PMU membrane module using MKM membranes [27].

The membrane filtration centrifuge (Fig. 8) is intended for the remediation of industrial wastewaters (liquid radioactive waste, industrial effluents, water conditioning, etc.) [10, 11, 13]. The unit uses a multilayer metal–ceramic membrane with a pore size from 0.4 to 100 nm as a filter element. The membrane rotor is made in the form of double-sided discs with sealing around the periphery and mounted on a shaft inside a cylindrical, hermetic centrifuge housing. The feed liquid under pressure is fed into the housing, the clarified liquid passes through the membrane and is withdrawn through the central shaft. The filtration area of a single disk is 0.1 m2. The MFC unit is easy to use and does not require the replacement of filter elements. The productivity of MFC depends on the number of disks in the membrane rotor. Ultrafiltration MFC units have been successfully used to clarify boron-containing waters of the Kola NPP spent fuel pool since 2003 [26].

Figure 9 shows a plate-and-frame membrane unit PMU operating in the crossflow filtration mode [27]. The unit consists of membrane modules, the number of which depends on the required productivity of the unit; valves; measurement components; and control system. The key advantage of this unit over the MFC is that the membrane modules can ensure a larger filtration flux up to 50–70 m3/h.

Fig. 9.

Multidisk immersion crossflow filtration unit MFC-M using MKM membranes [28].

The main advantages of MFC and PMU over analogues are as follows:

• The properties of the filter elements based on a multilayer metal–ceramic membrane allow the cleaning of liquids without prior preparation.

• The regeneration of filter elements is based on the shear effect, which allows the filter element to be cleaned during operation, and the unit, having reached the established productivity, operates continuously and does not need to be shut down to replace or regenerate the filter element.

• The units combine two properties, high performance and compactness.

• The lifetime of the units is limited by the total wear of the equipment (at least 1 year), the term during which there is no need to replace/regenerate the filter element.

• The units are mobile and easy to integrate into the production scheme.

Units of the MFC type are currently being developed by OOO Ecotransmission. Figure 9 shows an example of a MFC-M modified submersible unit [28] for filtering industrial and domestic wastewater. The principle of operation of the unit is similar to that of MFC; however, the MFC-M has no cooling loop, since the centrifuge body is immersed in the feed liquid, and the feed solution is pumped upwards by a paddle pressure pump. The weight of such a unit does not exceed 15 kg. The advantage of the MFC-M unit is its ease of transportation and the possibility of wastewater treatment at the place of their use.

A further development of MFC-type units is also the centrifugal filtration unit CFU [29, 30], the schematic diagram of which is shown in Fig. 10.

Fig. 10.

Schematic of the centrifugal filtration unit CFU [29].

The device is a rotating cartridge consisting of a perforated steel drum (support), on which an MKM metal–ceramic membrane is mounted. The feed stream under pressure flows tangentially with respect to the rotating membrane cartridge into the adjustable gap between the body of the unit and the cartridge. Due to the rotation of the cartridge, membrane self-cleaning is provided as in the case of MFC units.


Thus, MKM bilayer membranes are being used at present in MFC and PMU units as well as in promising MFC-M and CFU units. Three-layer metal–ceramic membranes can be used un devices of a similar configuration for:

—treatment of municipal and industrial wastewaters of complex composition, including that for disposal of liquid radioactive waste and chemical production waste;

—producing technical and drinking water;

—cleaning of petroleum products;

—regeneration of transformer oils.

Promising areas of application of multilayer metal–ceramic membranes are processes for:

—separation of hydrogen isotopes and production of ultrapure hydrogen;

—dehydration of organic liquids demanded by the industry;

—membrane electrolysis and fabrication of MMCM-based fuel cells;

—membrane catalysis;

—separation of associated and petroleum gas hydrocarbons.

Medical applications of multilayer metal–ceramic membranes are also possible, for example, for the separation of blood components.

Thus, studies in the field of modification of the structure of flexible metal–ceramic membranes are promising and have good prospects.


The work was performed within the framework of the State Task of the Topchiev Institute of Petrochemical Synthesis of the Russian Academy of Sciences.



  1. 1.
    V. N. Lapovok, V. I. Novikov, and L. I. Trusov, RU Patent No. 2 040 371 (1995).Google Scholar
  2. 2.
    L. I. Trusov, V. N. Lapovok, and V. I. Novikov, US Patent No. 5 364 586 (1994).Google Scholar
  3. 3.
    L. I. Trusov, V. P. Fedotov, and V. I. Novikov, US Patent No. 5 830 340 (1998).Google Scholar
  4. 4.
    V. I. Novikov, V. S. Vasil’kovskii, A. B. Senyavin, and A. B. Petunin, RU Patent No. 2 424 083 (2011).Google Scholar
  5. 5.
    V. I. Novikov and E. M. Solov’ev, RU Patent No. 2 579 713 (2016).Google Scholar
  6. 6.
    V. I. Novikov, A. I. Sharapaev, A. B. Petunin, and A. G. Muradova, Khim. Tekhnol. 16, 667 (2015).Google Scholar
  7. 7. Scholar
  8. 8.
    L. I. Trusov, Krit. Tekhnol. Membr., No. 9, 20 (2001).Google Scholar
  9. 9.
    B. G. Ershov, V. M. Gelis, V. V. Milyutin, et al., Vopr. Radiat. Bezopasn., No. 4, 36 (2009).Google Scholar
  10. 10.
    R. A. Penzin, V. M. Gelis, L. I. Trusov, et al., RU Patent No. 2 172 032 (2000).Google Scholar
  11. 11.
    V. M. Gelis, Yu. V. Glagolenko, N. N. Davidenko, et al., RU Patent No. 2 223 923 (2002).Google Scholar
  12. 12.
    E. V. Khataibe, A. N. Nechaev, L. I. Trusov, et al., Krit. Tekhnol. Membr., No. 16, 3 (2002).Google Scholar
  13. 13. Scholar
  14. 14.
    E. V. Khataibe, A. N. Nechaev, V. V. Berezkin, et al., Krit. Tekhnol. Membr., No. 17, 3 (2003).Google Scholar
  15. 15.
    M. V. Tsodikov, V. V. Teplyakov, M. I. Magsumov, et al., Kinet. Catal. 47, 25 (2006).CrossRefGoogle Scholar
  16. 16.
    V. L. Tarasov, L. I. Trusov, and V. P. Fedotov, RU Patent No. 2 381 824 (2007).Google Scholar
  17. 17.
    I. I. Ivanova, L. I. Trusov, E. E. Knyazeva, et al., RU Patent No. 2 322 390 (2006).Google Scholar
  18. 18.
    L. I. Trusov and V. P. Fedotov, RU Patent No. 2 312 702 (2006).Google Scholar
  19. 19.
    L. I. Trusov and V. P. Fedotov, RU Patent No. 2 312 703 (2006).Google Scholar
  20. 20.
    I. I. Ivanova, L. I. Trusov, E. E. Knyazeva, et al., RU Patent No. 2 382 671 (2008).Google Scholar
  21. 21.
    D. A. Fedosov, A. V. Smirnov, E. E. Knyazeva, et al., Krit. Tekhnol. Membr., No. 44, 28 (2009).Google Scholar
  22. 22.
    D. A. Fedosov, A. V. Smirnov, E. E. Knyazeva, and I. I. Ivanova, Pet. Chem. 51, 657 (2011).CrossRefGoogle Scholar
  23. 23.
    V. S. Mitin, V. I. Novikov, A. I. Sharapaev, and A. G. Muradova, Pet. Chem. 56, 920 (2016).CrossRefGoogle Scholar
  24. 24.
    V. M. Ievlev, A. I. Dontsov, V. I. Novikov, et al., Metally, No. 5, 70 (2018).Google Scholar
  25. 25.
    G. A. Dibrov, V. V. Volkov, V. P. Vasilevsky, et al., J. Membr. Sci. 470, 439 (2014).CrossRefGoogle Scholar
  26. 26.
    V. V. Omel’chuk, O. V. Dorodnoi, L. F. Barmin, et al., in Proceedings of the 4th International Scientific-and-Technical Conference on Safety of VVER Nuclear Power Plants (Gidropress, Podolsk, 2006), p. 68.Google Scholar
  27. 27.
    V. I. Novikov, A. S. Romanov, A. S. Filimonenko, and V. S. Vasil’kovskii, RU Patent No. 1 23 343 (2012).Google Scholar
  28. 28.
    V. I. Novikov, A. S. Romanov, A. S. Filimonenko, et al., RU Patent No. 102 531 (2011).Google Scholar
  29. 29.
    E. M. Solov’ev, V. I. Novikov, and V. V. Chelnokov, et al., RU Patent No. 145 027 (2014).Google Scholar
  30. 30.
    M. G. Berengarten, E. A. Belyaev, V. V. Vorob’ev, and V. I. Novikov, Demanding Scientific and Technical Problems of Ensuring Chemical Safety of Russia: For the 80th Anniversary of the Birth of the Lenin Prize Winner, Academician of the Russian Academy of Sciences, Lieutenant General Anatoly Demyanovich Kuntsevich: Proceedings of II Russian Conference with International Participation, Ed. by A. V. Roshchin (MTsNIP, Kirov, 2017), p. 78 [in Russian].Google Scholar

Copyright information

© Pleiades Publishing, Ltd. 2019

Authors and Affiliations

  • V. I. Novikov
    • 1
  • V. V. Kryachko
    • 1
  • Yu. I. Tarasov
    • 1
  • Sainsanaa Tserenchimed
    • 1
  • A. Yu. Alentiev
    • 2
    Email author
  1. 1.OOO EcotransmissionMoscowRussia
  2. 2.Topchiev Institute of Petrochemical Synthesis, Russian Academy of SciencesMoscowRussia

Personalised recommendations