Graphite-based Composite Bipolar Plates for Flow Batteries

Fengjing Jiang; Hanchen Song

doi:10.7536/PC210839

Progress in Chemistry ›› 2022, Vol. 34 ›› Issue (6) : 1290-1297 DOI: 10.7536/PC210839

research-article

Scan to read & Share

Graphite-based Composite Bipolar Plates for Flow Batteries

Fengjing Jiang ^*
, Hanchen Song

Author information +

History +

PDF

Abstract

Flow battery is considered a promising technology for large-scale energy storage, because of its high safety, long service life, and scalability. It can assist the power grid in peak shaving and energy storage, which will help to improve the energy utilization. Bipolar plate is a key component in a flow battery, which can separate single cells, connect the adjacent electrodes in series, help to conduct electric current and provide structural support for the stack. However, the cost of bipolar plate accounts for a big proportion of the total stack cost. It is of great significance to develop high-performance and low-cost bipolar plates, which will significantly accelerate the commercialization of flow batteries. Thus, its development has become an urgent need in the flow battery industry. Although much work related to the development of bipolar plates has been carried out, there is still lack of ideal bipolar plate products in the market, taking both their performance and cost into consideration. This article focuses on the introduction of current research status on graphite-based composite bipolar plates for flow batteries and the influence of material selection and processing techniques on the performance of bipolar plates, including electrical conductivity, mechanical strength, barrier properties, and corrosion resistance. Based on the review, some suggestions are made for the further developments of bipolar plates for flow batteries.

Contents

1 Introduction

2 Key performance and requirements of graphite-based composite bipolar plates

2.1 Electrical conductivity

2.2 Mechanical strength

2.3 Barrier properties

2.4 Corrosion resistance

2.5 Economical requirements

3 Conclusion and outlook

Graphical abstract

Keywords

bipolar plate / flow battery / carbon materials / carbon polymer composites

Cite this article

Download citation ▾

Fengjing Jiang, Hanchen Song. Graphite-based Composite Bipolar Plates for Flow Batteries. Progress in Chemistry, 2022, 34(6): 1290-1297 DOI:10.7536/PC210839

登录浏览全文

4963

注册一个新账户忘记密码

1 Introduction

Energy is an important foundation for human survival and development. With the increasing demand for energy, the increasing depletion of fossil energy and the increasing emphasis on environmental pollution and climate change, it has become an inevitable trend to shift from traditional fossil energy to non-fossil energy. Renewable energy is affected by seasonal climate, day and night alternation and other factors, and its power generation often has strong volatility and intermittency, which will have a certain impact on the stability and peak shaving capacity of the traditional power grid. The development of large-scale energy storage system can overcome this obstacle. Flow battery has the characteristics of high safety, high cycle life and scalability, which is an ideal large-scale energy storage technology.

As an important part of flow battery, bipolar plate plays a role in separating, connecting batteries in series, conducting current, and providing structural support for the stack. Therefore, the bipolar plate needs to have high conductivity, mechanical strength, air tightness, and corrosion resistance, as well as the characteristics of easy processing and low cost. The research of low-cost and high-performance bipolar plates is of great significance to accelerate the commercial application of the above batteries.

From the current types of flow batteries, most of the bipolar plates are in corrosive electrolytes (such as all-vanadium flow batteries), so only graphite bipolar plates and graphite-based composite bipolar plates can meet the requirements (neutral flow batteries are not within the scope of this paper). Although graphite plate has the advantages of high chemical stability, good corrosion resistance, high conductivity and low density, it can not meet the needs of large-scale application of flow batteries in the future because of its complex processing, high material cost (the space for material cost reduction is limited) and low production efficiency. Graphite-based composite bipolar plate is a kind of bipolar plate with graphite as the main conductive material and other auxiliary conductive carbon materials, reinforcing materials and resins as binders. Graphite is generally considered to be the best conductive material for the preparation of composite bipolar plates because of its excellent conductivity, chemical stability and low price.

It can be seen that the graphite-based composite bipolar plate is a promising type of bipolar plate for large-scale application of flow batteries in the future. However, it is still necessary to solve the problem of mutual restriction among key factors such as conductivity, mechanical strength, production cost and production efficiency. It can be found from the literature that many research results have been able to meet the requirements of conductivity and mechanical strength at the same time, but most of the work has not considered the importance of production cost and production efficiency of the technical route.This is also an important reason why many technologies can not be commercialized, and it is also an aspect that needs special attention in the future development of bipolar plate technology.

2 Performance characteristics and requirements of graphite-based composite bipolar plate

Graphite-based composite bipolar plate uses resin as binder, which can improve the mechanical strength and airtightness, but because resin is generally a non-conductive material, its conductivity is not as good as that of metal bipolar plate and pure graphite bipolar plate. In order to maximize the conductivity of graphite-based composite bipolar plates, it is necessary to connect the conductive fillers to form a conductive network during the preparation process, which is also the focus of many research works. Common conductive fillers are graphite, conductive carbon black, carbon nanotubes and so on. Commonly used polymer materials can be divided into thermoplastic resin and thermosetting resin according to the characteristics of molding. Different types of resin will cause great differences in dispersion, preparation process, equipment requirements, bipolar plate performance and so on. Thermoplastic resin is characterized by secondary processing and large selectivity of production process, mainly including polyvinylidene fluoride, polypropylene, polyphenylene sulfide, polyethylene and so on. Thermosetting resin will undergo curing reaction in the heating process, and will not soften when heated again after molding, so only one molding process can be used. Thermosetting resins mainly include epoxy resin, phenolic resin, etc.

Generally speaking, researchers mainly prepare high-performance and low-cost graphite-based bipolar plates for flow batteries through the selection of conductive materials and resins and the optimization of processes. However, from the current situation, there is no clear index for the performance and economic requirements of flow battery bipolar plate in the industry, which causes some confusion for the development of flow battery bipolar plate, and this paper will also discuss the performance requirements of flow cell bipolar plate.

2.1 Conductivity

Conductivity is one of the most important attributes of bipolar plates, which directly affects the performance of the stack. When the stack is in operation, the current density tends to be large, and if the resistance is high, a large amount of ohmic loss will be generated. The resistance of bipolar plates can be divided into two categories, bulk resistance and contact resistance. The volume resistivity can be measured by a four-probe low resistance meter, from which the conductivity of the bipolar plate can be calculated. Contact resistance is usually measured using Kumar et al.'S modified test method, as shown in Figure 1, which is also cited by the US Department of Energy^[1].

The conductivity, size, surface energy of conductive particles, the content and distribution of fillers and other factors will affect the conductivity of bipolar plates. Kang et al. Studied the effect of graphite particle size on the conductivity. When the graphite with smaller particles was used, the number of particles increased, and the conductivity of the composite decreased due to the higher contact resistance between particles^[2]. Rigail-Cede Cedeño et al. Used epoxy resin, expanded graphite and secondary fillers to prepare composite bipolar plates, and pointed out that the conductivity of bipolar plates was better when the particle size of expanded graphite was larger^[3]. Derieth et al. Pointed out that when the filling content is the same, the conductivity of flake graphite is significantly higher than that of spheroidal graphite^[4]. Clingerman et al. Pointed out that the difference in surface energy between the polymer and the conductive filler would affect the conductivity of the composite bipolar plate^[5]. When the filler has a similar surface energy to the resin, the filler is more easily wetted and achieves higher conductivity. According to the percolation threshold theory, when the filler content in the composite bipolar plate is less than the percolation threshold, the conductivity changes little with the increase of the filler content, and the conductivity of the composite is equal to or slightly higher than that of the polymer^[6]. When the filler content exceeds the penetration threshold, the conductive particles are able to form an interconnected conductive network, thereby greatly improving the conductivity. The penetration threshold is related to the particle size and geometry of the conductive filler, the type of resin, the degree of mixing, the crystallinity of the polymer, and the arrangement of the filler. For example, packings with a low aspect ratio tend to have a higher penetration threshold^[7]. Kang et al. Proposed that when the content of conductive filler is high, the viscosity of the mixture also increases, and if the wettability is poor, the filler and the matrix will produce pores, thereby reducing the conductivity^[8]. In this study, methanol was used to improve the wettability of phenolic resin on graphite and reduce the viscosity of the resin.

In the selection of conductive fillers, natural graphite has excellent conductivity and chemical stability, but its hardness is low, and its wettability with the resin matrix is small, so it is easy to form pores at the interface of the composite. The composition of synthetic graphite is mainly graphitic carbon, which is usually spherical, with uniform stress distribution and small strain, so the mechanical strength of the composite is higher when synthetic graphite is used. In addition, the surface energy of synthetic graphite is higher than that of natural graphite, which helps to reduce the porosity of the composite^[9]. Expanded graphite, also known as flexible graphite, has a loose and porous worm-like structure, which can be obtained by treating natural flake graphite with acidic oxidant^[10]. Expanded graphite has good conductivity, large specific surface area and is easy to process, but its conductivity is anisotropic, and the performance of in-plane direction is much better than that of vertical direction. Some studies prepared composite bipolar plates by impregnating expanded graphite blocks in organic solution containing resin, and Du et al. Compared the effects of three preparation methods: compression-impregnation method, impregnation-compression method, and compression-impregnation-compression method^[11,12]^[13]. The results show that impregnation followed by compression causes more resin molecules to enter the exfoliated graphite sheets and makes the sheets more difficult to compress, resulting in thicker bipolar plates and higher mechanical strength. When the method of compression first is used, the conductive network in the expanded graphite sheet is denser and the conductivity is better. The study also points out the appropriate concentration of resin solution for different methods: when the compression-impregnation method is used, a high concentration of resin solution (40%) is required; When the impregnation-compression method is used, the concentration shall be less than 10%; When the compression-impregnation-compression method is used, the concentration shall be kept between 25% and 40%. Carbon black has a large specific surface area and small particles, which can be used as a secondary filler to fill the pores between graphite particles and form conductive channels, thus improving the conductivity of the composite bipolar plate^[14]. Gautam et al. Used expanded graphite (35 wt%), carbon black (5 wt%), graphite powder (3 wt%) and phenolic resin to prepare a low-density composite bipolar plate with a conductivity of 374.42 S/cm^[15]. Mesocarbon microbeads (MCMB), graphene, and graphene nanosheets (GNP) are also commonly used as conductive fillers. Liao et al. Used graphene and graphite powder to prepare composite bipolar plates with good conductivity at low carbon content (less than 20 wt%)^[16⇓~18]^[19⇓~21]^[22]^[23].

The selection of resin material also has a certain influence on the conductivity of the composite bipolar plate. Due to the lack of polar groups in the main chain of polypropylene, the conductive filler is difficult to achieve uniform dispersion, and the conductivity of the composite is general. Dweiri et al. Increased the conductivity by adding carbon black and pointed out that higher conductivity could be obtained by solution blending than by melt blending^[24]. The interfacial compatibilizer can increase the interfacial reaction of fillers such as polypropylene and graphite, thereby improving the dispersibility^[25,26]. Adloo et al. selected graphene and nano-carbon black to improve the conductivity, and added 5 wt% maleic anhydride grafted polypropylene (pp-MAH) to improve the dispersion of graphene in the resin matrix. The optimal filling mass fractions of graphene and nano-carbon black were 1% and 7%, respectively, and the conductivity of the composite could reach 104.63 S/cm^[27].

In addition, the parameters such as molding pressure and molding temperature can also affect the conductivity of the composite bipolar plate. The determination of these parameters needs to take into account the curing and rheological properties of the resin material used. Increasing the molding pressure can make the contact between conductive fillers closer, and increase the conductive path and the contact area between conductive fillers. However, when the molding pressure is too high, the structure of the filler itself may be destroyed, and the conductivity of the composite bipolar plate may decrease^[28]. Increasing the molding temperature can enhance the fluidity of the resin material and make it distribute evenly. San et al. Used response surface analysis to study the appropriate process parameters such as molding pressure, temperature and time^[29]. The results show that when the molding temperature increases, the molecular mobility increases, which is beneficial to the formation of conductive network. When the molding temperature is constant, the conductivity increases first and then decreases with the increase of molding pressure. Prolonging the molding time can reduce the porosity and improve the surface properties of the bipolar plate.

During the molding process of the composite bipolar plate at high temperature, the polymer material has smaller surface energy and is easier to migrate to the surface, which leads to the formation of a resin-rich layer on the surface of the composite bipolar plate^[30]. For graphite-based composite bipolar plates, the contact resistance is often more significant than the bulk resistance. In order to reduce the contact resistance, the resin-rich layer on the surface of the composite bipolar plate can be removed by surface treatment^{[31⇓⇓⇓⇓~36]}. Lee et al. Used polyester fabric to absorb the excessive resin on the surface of the composite bipolar plate during the resin curing process, which significantly reduced the surface specific resistance of the bipolar plate^[37]. Liu et al. Compared the effects of plasma treatment, flake graphite coating and carbon nanotube coating, and the results showed that the contact resistance of the composite bipolar plate with carbon nanotube coating was the lowest, followed by the effect of flake graphite coating^[38]. The contact resistance is related to the bipolar plate material, compaction pressure, surface roughness and other factors^[39]. Higher surface roughness reduces the actual contact area and increases the contact resistance. Increasing the compaction pressure of the stack can increase the contact area, but excessive compaction pressure will deform the proton exchange membrane and may lead to leakage and internal short circuit. Xiong et al. Studied the effect of compaction pressure on the electrochemical performance and mechanical reliability of the battery^[40]. The results show that when the compaction pressure increases, the overall polarization of the cell decreases, while the power density at a certain current density increases. However, when the compaction pressure is too high, the probability of battery failure due to mechanical failure will also increase, and 500 kN is a more appropriate compaction pressure. Table 1 shows the conductivity data of some composite bipolar plates for flow batteries in recent years.

2.2 Mechanical strength

In order to meet the sealing requirements, the stack is usually compressed during operation, so the mechanical strength of the bipolar plate, especially the bending strength, is required. Flexural strength is often measured by the three-point bending method. The mechanical strength of the composite bipolar plate is related to the resin type, content and process. Thermosetting resins tend to have higher strength, creep resistance and thermal stability at higher temperatures than thermoplastic resins, but are more brittle^[45]. When the content of conductive filler increases, more pores and defects will be produced inside the bipolar plate, which will lead to the decrease of mechanical strength. Increasing the resin content can fully infiltrate the conductive filler particles, and the mechanical strength of the composite bipolar plate can be enhanced by the acting force at the interface between the conductive filler and the resin material. However, the increase of resin content often leads to the decrease of conductivity, so how to balance the conductivity and mechanical properties is a key issue in the design of composite bipolar plates. In addition, increasing the molding pressure can promote the flow of the resin material between the conductive fillers and improve the wetting effect of the fillers, thereby improving the mechanical properties of the composite bipolar plate. The mechanical strength is also affected by the type and size of the conductive filler. Kang et al. Pointed out that the ratio of surface area to volume of small graphite particles is larger, and the mechanical strength of the prepared composite is higher^[2]. At the same time, the surface energy of natural graphite is lower than that of synthetic graphite, and it is more difficult to be completely wetted by resin matrix, so the flexural strength of composites using natural graphite is lower. In order to improve the mechanical strength of the composite bipolar plate, fillers that can be used as reinforcement materials have been used in many studies.

Carbon fiber is a common reinforcing material with high strength, high modulus, good conductivity and low density, which can improve the mechanical properties of graphite-based composite bipolar plates^[46⇓~48]. The interfacial bonding between carbon fiber and polymer matrix, the content and length of carbon fiber can affect the properties of composites. Ghosh et al. Pointed out that the length of carbon fiber should be 1 mm, and when the length exceeds 2 mm, the tendency of agglomeration increases, and it is difficult to achieve uniform dispersion^[19]. When the carbon fiber is poorly bonded to the polymer, the porosity of the composite increases and results in a decrease in flexural strength. Because of the surface inertness of carbon fiber, plasma treatment, gas-liquid double effect treatment, chemical treatment and other methods are often used to modify carbon fiber to improve the interfacial properties between carbon fiber and polymer^{[49⇓⇓~52]}. Carbon nanotubes are also commonly used as fillers to improve mechanical properties, which bend elastically when subjected to external forces and absorb impact energy^[53]^[54]. The higher the aspect ratio of carbon nanotubes, the smaller the penetration threshold^[55]. When the content of carbon nanotubes is low, the carbon nanotubes can be uniformly dispersed in the resin matrix, but when the content is high, the conductivity of the composite bipolar plate will decrease due to agglomeration. Dhakate et al. Used phenolic resin, natural graphite and multi-walled carbon nanotubes (MWNT) to prepare composite bipolar plates, and the results showed that when the MWNT content exceeded 1 vol%, agglomeration was prone to occur^[56]. Functionalized carbon nanotubes can improve their dispersion and penetration threshold. Jin et al. Used a mixture of concentrated sulfuric acid and nitric acid to treat carbon nanotubes, and then functionalized them with dodecylamine. The treated carbon nanotubes could be more uniformly dispersed in the resin matrix^[57]. Lee et al. compared the effects of multi-walled carbon nanotubes (MWNT) and carbon fiber on improving the flexural strength of composites. The experimental results showed that the optimal content of MWNT was 2 vol%, and the optimal content of carbon fiber was 7 vol%, and the effect was better than that of adding MWNT^[58]. In recent years, we have sorted out and summarized the mechanical properties of composite bipolar plates for graphite-based flow batteries, as shown in Table 2.

2.3 Barrier property

The gas permeation mechanism is determined by the properties of the substrate material of the composite bipolar plate. With the increase of resin content in the composite bipolar plate, the porosity decreases, the structure becomes more dense, and the barrier property is also improved. When selecting the conductive filler, the filler with higher aspect ratio is more conducive to reducing the gas permeability and improving the barrier property^[61]. The type of conductive filler also has an effect on the barrier properties. Choe et al. Compared the permeability when using pyrolytic graphite and flaky exfoliated graphite, and pointed out that pyrolytic graphite has a highly ordered structure in the in-plane direction, so it has better barrier properties^[62]. In the process of preparing the composite bipolar plate, increasing the molding pressure can reduce the porosity and improve the barrier property of the bipolar plate. Kim et al. Pointed out that the surface of pure graphite bipolar plate prepared by expanded graphite has micro-cracks, and sulfuric acid solution will diffuse into the interior of the bipolar plate along these cracks. Adding more than 6% polytetrafluoroethylene (PTFE) can significantly reduce the surface cracks and enhance the barrier property of the bipolar plate^[63].

In addition, the aging of the polymer in the battery environment can also lead to the decline of the barrier performance of the composite bipolar plate^[7]. At present, in order to achieve higher energy density of the stack, the thickness of the bipolar plate is often reduced as much as possible, which also leads to higher permeability. When testing the barrier property of the bipolar plate, the sample is usually fixed between two gas chambers by a fixture, the gas is introduced into the gas chamber and the pressure difference between the two sides is maintained, and the concentration of the measured gas at the outlet is measured by a gas chromatograph.

2.4 Corrosion resistance

When the polarization potential of the composite bipolar plate is greater than the polarization potential of oxygen evolution reaction, carbon materials including graphite, carbon black, carbon nanotubes, etc. May be corroded, thereby affecting the performance of the battery. In addition, at high state of charge (SOC), the side reactions of the positive and negative electrolytes can lead to the generation of gases such as O₂ and CO₂, and damage the bipolar plate^[64]. Al-Fetlawi et Al. Pointed out that the produced oxygen will reduce the active surface area and effective diffusion coefficient of the electrode, and reduce the coulombic efficiency of the battery, while the increase of temperature will enhance the evolution of oxygen^[65]. Different types of carbon materials have different corrosion resistance, such as carbon black, carbon nanotubes and other secondary fillers often show higher chemical instability than natural graphite^[66]. At the same time, the porosity related to the type and content of filler will also affect the corrosion resistance of composites^[67]. Liu et al. Pointed out that the porosity in the composite bipolar plate can be reduced and the corrosion resistance can be improved by increasing the molding pressure and adding smaller secondary fillers^[68]. In the experiment of Hu et al., the corrosion current density of the composite bipolar plate first decreased and then increased with the increase of graphite content from 35 wt% to 60 wt%, and it was pointed out that the increase of graphite content would reduce the corrosion resistance of the bipolar plate, while the decrease of the corrosion current density in the early stage might be due to the increase of hydrophobicity of the bipolar plate and the decrease of electrolyte adsorption^[69].

Corrosion resistance is also related to the type of resin material. Xu Dongqing et al pointed out that the carbon-hydrogen bonds in polypropylene and polyethylene may be oxidized and broken in the flow battery environment. In comparison, polyvinyl chloride and polyvinylidene fluoride (PVDF) have better chemical stability^[70]. Lee et al pointed out that epoxy resin has high electrochemical stability, so that the corrosion current density of the composite bipolar plate is lower than that of the traditional graphite bipolar plate, but when the content of carbon black in the composite material is too high, the compatibility between carbon black and epoxy resin becomes poor, and the corrosion resistance of the composite bipolar plate will be reduced^[41]. When testing the corrosion resistance, the bipolar plate is usually placed in a simulated battery environment for potential scanning, and the polarization curve is fitted by Tafel.

2.5 Economy

Zheng et al. divided the cost of flow battery system into six parts: battery cost (membrane, electrode, bipolar plate, etc.), stacking cost (end plate, collector plate, bolt, etc.), pipeline cost (pipeline and pump), system cost (control system and temperature management system), installation cost and electrolyte cost.Four kinds of flow batteries were studied, including vanadium redox battery (VRB), iron-chromium battery (Fe-Cr), zinc-bromine battery (ZB) and sodium polysulfide/bromine flow battery^[71]. The study shows that the most expensive part of the four flow battery systems is the battery cost, in which the membrane accounts for the highest proportion, followed by the bipolar plate, accounting for about 15% (VRB) to 27% (ZBB). Viswanathan et al. compared the cost and composition of iron-vanadium flow battery and all-vanadium flow battery, and pointed out that the cost of ion exchange membrane, active material and pumping cost accounted for a large proportion^[72]. For the 1 MW-0.25 H all-vanadium redox flow battery system, the bipolar plate accounts for 6% of the total cost. When the application time increases to 4 H, the active material cost increases significantly, and the proportion of the bipolar plate in the total cost decreases to 4%. For the iron-vanadium system with the same power and capacity, the cost of the bipolar plate accounts for 15% and 9% of the total cost, respectively, as shown in Figure 2. Zeng et al. analyzed the energy storage system of 1 MW-8 H, and proposed that the cost of the all-vanadium flow battery is about $229/kWh, of which the cost of the bipolar plate accounts for about 2%, while the iron-chromium flow battery has a low cost of active materials.The cost is $194/kWh, and the bipolar plate accounts for about 4%. The cost of these two flow batteries is still far from the goal of controlling the cost of energy storage system at $100/kWh proposed by the U.S. Department of Energy^[73]^[74]. Considering the cost of flow battery, the lower the cost of bipolar plate, the better, but there is no clear requirement in the industry. Viswanathan et al. Pointed out that the cost of bipolar plate is about $55 /m² at this stage, which is expected to be reduced to $35 /m² in the short term, while the study believes that the ideal cost of bipolar plate should eventually reach $25 /m²^[72]. The authors believe that the goal of the first phase should be to reduce the price of high-performance composite bipolar plates for flow batteries (at a current density of 100 mA/cm², the energy efficiency of flow batteries can approach and reach the level of using traditional graphite plates) to less than 200 yuan /m².

The cost of graphite-based composite bipolar plate mainly includes raw material cost and production cost. According to the current public information, the materials selected for the graphite-based composite bipolar plate are mostly common graphite materials, carbon materials and resins, and the cost of the materials can meet the requirements of cost reduction. However, the production cost mainly depends on the production mode of the bipolar plate, which in turn depends on the material composition of the graphite-based composite bipolar plate. At present, the most commonly used preparation process is compression molding, but the production efficiency of this method is low. Injection molding has higher production efficiency and is suitable for mass production, but it requires relatively high fluidity of the material, and high shear force may break the filler particles with high aspect ratio^[75]. Minke et al. Provided a method to quantify the cost of composite bipolar plates, evaluated three composite bipolar plates with different resin matrices, and compared the costs associated with compression molding and injection molding^[76]. The method involves raw material costs, manufacturing costs (machine parameters, energy parameters, etc.). The results show that injection molding has more advantages in terms of cost and output, but it is not suitable for all materials due to viscosity and other limitations. In the injection molding process, the investment of material cost, machine and mold accounts for a slightly higher share. The study also points out that the most potential way to reduce costs is to promote standardization and mass production. Yang et al. Compared the carbon fiber/polyethylene composite bipolar plates prepared by injection molding and compression molding, and pointed out that the bipolar plates prepared by compression molding had higher tensile strength and conductivity^[43].

In addition, we believe that roll forming is one of the most efficient production methods, with low input, high output and low production cost. However, the rolling process requires very high processing characteristics of the material. At present, the efficient production of bipolar plates is an important problem to be solved in industrialization technology. In the process of designing the technical route of graphite-based composite bipolar plate, it is essential to consider whether efficient production methods can be adopted on the premise of meeting the performance requirements.

3 Conclusion and prospect

Graphite-based composite bipolar plates have great potential for development due to their low density and good corrosion resistance, which will have a significant impact on the large-scale application of flow batteries. In this paper, the research progress of graphite-based composite bipolar plates in recent years is introduced based on the performance and economic requirements of bipolar plates. In terms of material selection, many studies have started from the properties of the resin matrix itself to make up for the lack of performance, and compared or even combined a variety of secondary fillers (carbon black, carbon nanotubes, carbon fibers, graphene, etc.). In the aspect of preparation process, the mixing method of materials, molding parameters and process optimization were studied, and some surface treatment methods were proposed to remove the influence of resin-rich layer of composite bipolar plate and reduce the contact resistance. At present, the production cost of high-performance graphite-based composite bipolar plates is still high, which is difficult to meet the needs of large-scale applications. Performance, production efficiency and cost are the main problems to be solved in the future for bipolar plates of flow batteries. The commonly used compression molding method has high cost and low production rate; Injection molding has certain advantages in terms of cost and production rate, but it is limited in the choice of material system. The impregnated graphite plate also has the problems of high cost and long preparation time; Compared with the previous processes, roll forming has the advantages of high production efficiency and low cost, but usually the carbon content in the resin is low and the performance is insufficient. In the future, it is necessary to carry out innovative design on the material composition of bipolar plates, develop graphite-based composite bipolar plates with low carbon content and high conductivity, and adopt efficient production methods for large-scale production. In addition, the research on the corrosion resistance and stability of graphite-based composite bipolar plates is relatively limited. Considering that the electrolyte of flow batteries often has strong acidity and chemical corrosion, more exploration in this area is needed in the future, especially the oxidation resistance and acid resistance of non-graphite carbon materials and resins need to be further verified. In the face of the above challenges, it is necessary to continue to study new material systems, further study the synergistic effect of conductive fillers, adjust the composition and preparation process to promote the formation and optimization of conductive networks, reduce the production cost of graphite-based composite bipolar plates and improve the production efficiency on the premise of taking into account the performance, so as to reduce the cost of energy storage for flow batteries.

References

Publishing order | Descend order by publishing year | Descend order by cited within

[1]	Kumar A, Ricketts M, Hirano S. J. Power Sources, 2010, 195(5): 1401.

[2]	Kang K, Park S, Ju H. Solid State Ion., 2014, 262: 332.

[3]	Rigail-Cedeño A F, Espinoza-Andaluz M, Vera J, Orellana-Valarezo M, Villacis-Balbuca M. Mater. Today Proc., 2020, 33: 2003.

[4]	Derieth T, Bandlamudi G, Beckhaus P, Kreuz C, Mahlendorf F, Heinzel A. J. New Mat. Electr. Sys., 2008, 11: 21.

[5]	Clingerman M L, King J A, Schulz K H, Meyers J D. J. Appl. Polym. Sci., 2002, 83(6): 1341.

[6]	Mamunya Y P, Davydenko V V, Pissis P, Lebedev E V. Eur. Polym. J., 2002, 38(9): 1887.

[7]	Blunk R, Zhong F, Owens J. J. Power Sources, 2006, 159(1): 533.

[8]	Kang S J, Kim D O, Lee J H, Lee P C, Lee M H, Lee Y, Lee J Y, Choi H R, Lee J H, Oh Y S, Nam J D. J. Power Sources, 2010, 195(12): 3794.

[9]	Mathur R B, Dhakate S R, Gupta D K, Dhami T L, Aggarwal R K. J. Mater. Process. Technol., 2008, 203(1/3): 184.

[10]	Rao J, Zhang P, He S, Li Z H, Ma H W, Shen Z P, Miao S D. Sci. Sin. Technol., 2017, 47(1): 13.

[11]	( 饶娟, 张盼, 何帅, 李植淮, 马鸿文, 沈兆普, 苗世顶. 中国科学: 技术科学, 2017, 47(1): 13.)

[12]	Li W W, Jing S, Wang S B, Wang C, Xie X F. Int. J. Hydrog. Energy, 2016, 41(36): 16240.

[13]	Luo X K, Hou M, Fu Y F, Hou Z J, Ming P W, Yi B L. Chinese Journal of Power Sources, 2008(03): 174.

[14]	( 罗晓宽, 侯明, 傅云峰, 侯中军, 明平文, 衣宝廉. 电源技术, 2008( 03): 174.).

[15]	Du C, Ming P W, Hou M, Fu J, Fu Y F, Luo X K, Shen Q, Shao Z G, Yi B L. J. Power Sources, 2010, 195(16): 5312.

[16]	Park M, Jung Y J, Ryu J, Cho J. J. Mater. Chem. A, 2014, 2(38): 15808.

[17]	Gautam R K, Kar K K. Fuel Cells, 2016, 16(2): 179.

[18]	Yang T, Shi P F. J. Power Sources, 2008, 175(1): 390.

[19]	Li F N, Li H J, Wang J. Carbon, 2004, 42: 2989.

[20]	Yang T, Shi P. Chinese J. Power Sources, 2008, 32(5): 306.

[21]	( 杨涛, 史鹏飞. 电源技术, 2008, 32(5): 306.)

[22]	Ghosh A, Goswami P, Mahanta P, Verma A. J. Solid State Electrochem., 2014, 18(12): 3427.

[23]	Liao W N, Jiang F J, Zhang Y, Zhou X J, He Z Q. Renew. Energy, 2020, 152: 1310.

[24]	Phuangngamphan M, Okhawilai M, Hiziroglu S, Rimdusit S. J. Appl. Polym. Sci., 2019, 136(11): 47183.

[25]	Jiang X, Drzal L T. J. Power Sources, 2012, 218: 297.

[26]	Liao W N, Zhang Y, Zhou X J, Zhuang M D, Guo D Y, Jiang F J, Yu Q C. ChemistrySelect, 2019, 4(8): 2421.

[27]	Dweiri R, Sahari J. J. Power Sources, 2007, 171(2): 424.

[28]	Mirzazadeh H, Katbab A A, Hrymak A N. Polym. Adv. Technol., 2011, 22(6): 863.

[29]	Wang W B, Wang J H, Wang S B, Xie X F, Lv Y F, Qi L, Yao K J. Chem. Ind. Eng., 2011, 62: 203.

[30]	( 王文嫔, 王金海, 王树博, 谢晓峰, 吕亚非, 齐亮, 姚克俭. 化工学报, 2011, 62: 203.).

[31]	Adloo A, Sadeghi M, Masoomi M, Pazhooh H N. Renew. Energy, 2016, 99: 867.

[32]	Zhao Z D, Men Y, Hou S Y, Wang Y L, Liu J G, Yan C W. Chinese Journal of Power Sources, 2011, 35(11): 1376.

[33]	( 赵正德, 门阅, 侯绍宇, 王远乐, 刘建国, 严川伟. 电源技术, 2011, 35(11): 1376.)

[34]	San F G B, Okur O. Int. J. Hydrog. Energy, 2017, 42(36): 23054.

[35]	Liu Z H, Wang B G, Yu L X. J. Energy Chem., 2018, 27(5): 1369.

[36]	Yu H N, Lim J W, Suh J D, Lee D G. J. Power Sources, 2011, 196(23): 9868.

[37]	Yu H N, Lim J W, Kim M K, Lee D G. Compos. Struct., 2012, 94(5): 1911.

[38]	Kim K H, Kim B G, Lee D G. Compos. Struct., 2014, 109: 253.

[39]	Lee H E, Han S H, Song S A, Kim S S. Compos. Struct., 2015, 134: 44.

[40]	Lee D, Lim J W, Nam S, Choi I, Lee D G. Compos. Struct., 2015, 134: 1.

[41]	Jiang F J, Liao W N, Ayukawa T, Yoon S H, Nakabayashi K, Miyawaki J. J. Power Sources, 2021, 482: 228903.

[42]	Lee D, Lee D G, Lim J W. J. Intell. Mater. Syst. Struct., 2018, 29(17): 3386.

[43]	Liu Z H, Wang B G, Yu L X. J. Energy Chem., 2018, 27(5): 1369.

[44]	Avasarala B, Haldar P. J. Power Sources, 2009, 188(1): 225.

[45]	Xiong J, Song Y X, Wang S L, Li X R, Liu J G, Yan C W, Tang A. J. Power Sources, 2019, 431: 170.

[46]	Lee N J, Lee S W, Kim K J, Kim J H, Park M S, Jeong G, Kim Y J, Byun D. Bull. Korean Chem. Soc., 2012, 33(11): 3589.

[47]	Satola B, Kirchner C N, Komsiyska L, Wittstock G. J. Electrochem. Soc., 2016, 163(10): A2318.

[48]	Yang L, Zhou Y, Wang S, Lin Y, Huang T, Yu A. Int. J. Electrochem. Sci., 2017, 12: 7031.

[49]	Nam S, Lee D, Lee D G, Kim J. Compos. Struct., 2017, 159: 220.

[50]	Saadat N, Dhakal H N, Tjong J, Jaffer S, Yang W M, Sain M. Renew. Sustain. Energy Rev., 2021, 138: 110535.

[51]	Kakati B K, Sathiyamoorthy D, Verma A. Int. J. Hydrog. Energy, 2010, 35(9): 4185.

[52]	Kim N H, Kuila T, Kim K M, Nahm S H, Lee J H. Polym. Test., 2012, 31(4): 537.

[53]	Ruban E, Stepashkin A, Gvozdik N, Konev D, Kartashova N, Antipov A, Lyange M, Usenko A. Mater. Today Commun., 2021, 26: 101967.

[54]	Boudou J P, Paredes J I, Cuesta A, Martínez-Alonso A, TascÓn J M D. Carbon, 2003, 41(1): 41.

[55]	Montes-Morán M A, Martínez-Alonso A, TascÓn J M D, Paiva M C, Bernardo C A. Carbon, 2001, 39(7): 1057.

[56]	Dai Z S, Shi F H, Zhang B Y, Li M, Zhang Z G. Appl. Surf. Sci., 2011, 258(5): 1894.

[57]	Ozkan C, Karsli N G, Aytac A, Deniz V. Compos. B Eng., 2014, 62: 230.

[58]	Chaiwan P, Pumchusak J. Electrochimica Acta, 2015, 158: 1.

[59]	Zhang M Y, Sui S, Chen J Y, Wu S L, Sun R, Guo N. Trans. China Electrotech. Soc., 2014, 29(4): 97.

[60]	( 张明艳, 隋珊, 陈金玉, 吴淑龙, 孙睿, 郭宁. 电工技术学报, 2014, 29(4): 97.)

[61]	Celzard A, McRae E, Deleuze C, Dufort M, Furdin G, MarêchÉ J F. Phys. Rev. B, 1996, 53(10): 6209.

[62]	Dhakate S R, Sharma S, Chauhan N, Seth R K, Mathur R B. Int. J. Hydrog. Energy, 2010, 35(9): 4195.

[63]	Jin F L, Ma C J, Park S J. Mater. Sci. Eng. A, 2011, 528(29/30): 8517.

[64]	Lee J H, Jang Y K, Hong C E, Kim N H, Li P, Lee H K. J. Power Sources, 2009, 193(2): 523.

[65]	Caglar B, Fischer P, Kauranen P, Karttunen M, Elsner P. J. Power Sources, 2014, 256: 88.

[66]	Lee D, Choe J, Nam S, Lim J W, Choi I, Lee D G. Compos. Struct., 2017, 160: 976.

[67]	Jeong K I, Oh J, Song S A, Lee D, Lee D G, Kim S S. Compos. Struct., 2021, 262: 113617.

[68]	Choe J, Lim J W, Kim M, Kim J, Lee D G. Compos. Struct., 2015, 134: 106.

[69]	Kim S, Yoon Y, Narejo G M, Jung M, Kim K J, Kim Y J. Int. J. Energy Res., 2021, 45(7): 11098.

[70]	Satola B, Komsiyska L, Wittstock G. J. Electrochem. Soc., 2018, 165(5): A963.

[71]	Al-Fetlawi H, Shah A A, Walsh F C. Electrochimica Acta, 2010, 55(9): 3192.

[72]	Antunes R A, Lopes de Oliveira M C, Ett G. Int. J. Hydrog. Energy, 2011, 36(19): 12474.

[73]	Wang M X, Liu Q, Sun H F, Ogbeifun N, Xu F, Stach E A, Xie J. Mater. Chem. Phys., 2010, 123(2/3): 761.

[74]	Liu H J, Yang L X, Xu Q, Yan C W. RSC Adv., 2015, 5(8): 5928.

[75]	Hu B, Chang F L, Xiang L Y, He G J, Cao X W, Yin X C. Int. J. Hydrog. Energy, 2021, 46(50): 25666.

[76]	Xu D Q, Fan Y S, Liu P, Wang B G. J. Chem. Eng. Chin. Univ., 2011, 25(2): 308.

[77]	( 徐冬清, 范永生, 刘平, 王保国. 高校化学工程学报, 2011, 25(2): 308.)

[78]	Zheng M L, Sun J, Meinrenken C J, Wang T. J. Electrochem. Energy Convers. Storage, 2019, 16(2): 021001.

[79]	Viswanathan V, Crawford A, Stephenson D, Kim S, Wang W, Li B, Coffey G, Thomsen E, Graff G, Balducci P, Kintner-Meyer M, Sprenkle V. J. Power Sources, 2014, 247: 1040.

[80]	Zeng Y K, Zhao T S, An L, Zhou X L, Wei L. J. Power Sources, 2015, 300: 438.

[81]	Mongird K, Viswanathan V, Alam J, Vartanian C, Sprenkle V, Baxter R. 2020 Grid Energy Storage Technology Cost and Performance Assessment (2020-12-17). [2021-11-1]. https://www.energy.gov/energy-storage-grand-challenge/downloads/2020-grid-energy-storage-technology-cost-and-performance/.