A review of microstructure optimization design for flow fields of metallic bipolar plates in hydrogen fuel cells

Abstract

Metallic bipolar plates (BPPs) are critical components in proton exchange membrane fuel cells (PEMFCs), accounting for the majority of stack mass, volume, and cost. The microstructure of the flow field directly governs reactant distribution, water management, and interfacial contact resistance, thereby determining overall cell performance and durability. This review examines recent advances in the optimization of flow field microstructures for metallic BPPs, encompassing conventional channel configurations, bio-inspired and fractal geometries, and topology-optimized architectures. The analysis extends to manufacturing technologies, including stamping, hydroforming, and additive manufacturing, which dictate the geometric fidelity and cost-effectiveness of metallic plates. Furthermore, protective coating strategies—such as diamond-like carbon, nitride films, and metal oxide layers—are discussed in relation to corrosion resistance and electrical conductivity in acidic fuel cell environments. Multiphysics simulation methods that couple fluid dynamics, electrochemical reactions, and structural mechanics are highlighted as essential tools for rationalizing microstructural design. Finally, future research directions are identified, focusing on multi-objective optimization, in situ characterization, and the integration of advanced manufacturing with intelligent design frameworks to accelerate the commercialization of high-performance PEMFC stacks.

1 Introduction

This review synthesizes peer-reviewed literature (2005–2025) from Web of Science, Scopus, and ScienceDirect, focusing on studies with quantitative performance metrics and excluding conference abstracts and non-metallic plate investigations. Metallic bipolar plates have emerged as the preferred alternative to graphite and composite plates for automotive PEMFC applications due to their superior mechanical strength, thin-gauge formability, and excellent electrical and thermal conductivity (Hermann et al., 2005; Antunes et al., 2010). Stainless steels, titanium alloys, and aluminum alloys represent the primary candidate materials, with austenitic stainless steel 316L being particularly attractive because of its balance between corrosion resistance, formability, and cost (Wang and Northwood, 2007; Antunes et al., 2010). The bipolar plate performs multiple critical functions simultaneously: it distributes hydrogen and oxygen uniformly across the active area, conducts electrical current between adjacent cells, removes generated heat, and provides structural support to the membrane electrode assembly (Li and Sabir, 2005; Peng et al., 2020; Wilberforce et al., 2016). These multifunctional requirements are reflected in stringent U.S. Department of Energy targets that specify interfacial contact resistance below 10 mΩ cm², corrosion current density less than 1 μA/cm², and flexural strength exceeding 25 MPa (Peng et al., 2020; Wind et al., 2014). The flow field microstructure—comprising channel geometry, rib-to-channel ratio, depth, width, and surface morphology—constitutes the dominant design variable influencing pressure drop, reactant uniformity, liquid water removal, and current density distribution across the catalyst layer (Li and Sabir, 2005; Manso et al., 2012). Conventional configurations such as parallel, serpentine, and interdigitated channels have been extensively studied, yet they often suffer from inherent trade-offs between pressure drop and flow uniformity (Li and Sabir, 2005; Chen et al., 2014). In response, researchers have pursued unconventional microstructures including bio-inspired fractal networks, three-dimensional baffled geometries, wire-mesh architectures, and topology-optimized flow pathways that challenge traditional manufacturing constraints (Kloess et al., 2009; Xia et al., 2024). This review synthesizes the state-of-the-art in flow field microstructure optimization for metallic bipolar plates, addressing geometric design innovations, precision manufacturing processes, surface engineering strategies, and computational optimization frameworks that collectively define the current frontier of PEMFC technology (Wang, 2012). To provide a navigational scaffold for the technically intricate discourse that follows, Figure 1 presents a systematic six-panel framework for microstructure optimization of metallic bipolar plate flow fields: (a) the multifunctional bipolar plate integrating reactant distribution, electrical conduction, heat removal, and structural support with DOE performance targets; (b) conventional flow field configurations—parallel, serpentine, and interdigitated—illustrating the pressure drop versus flow uniformity trade-off; (c) advanced bio-inspired leaf venation, fractal branching, and topology-optimized designs achieving 20%–30% power density improvement (Xia et al., 2024); (d) manufacturing technology comparison across stamping, hydroforming, photochemical etching, laser machining, and SLM(selective laser melting) additive manufacturing; (e) protective coating performance radar comparing DLC, nitride films, metal oxides, and conductive polymers; and (f) multiphysics CFD (computational fluid dynamics) simulation resolving current density hotspots and water flooding risk zones across the active area. Readers are encouraged to refer to this figure throughout the subsequent sections to contextualize specific technical advances within the broader PEMFC performance optimization ecosystem (Sevinc and Hazar, 2026).

FIGURE 1

Peng et al. (2020) and Wind et al. (2014). Panel (b) schematic of conventional designs based on Li and Sabir (2005) and Chen et al. (2014). Panel (c) bio-inspired and topology-optimized designs; 20%–30% power density improvement under H₂/air, 70 °C–80 °C, 1–3 atm from Xia et al. (2024). Panel (d) manufacturing technologies reviewed by Hu et al. (2014), Mahabunphachai and Koc (2010), and Porstmann et al. (2020). Panel (e) coating performance synthesized from Wang et al. (2006), Lee and Park (2015), and Ren et al. (2021). Panel (f) CFD simulation methodologies from Andersson et al. (2016) and Park et al. (2016).

While existing reviews have examined PEMFC bipolar plates from isolated perspectives—flow field geometry, coating technologies, or manufacturing processes—an integrated treatment of microstructure optimization that couples geometric design with manufacturing constraints and surface engineering remains lacking. The specific contribution of this review is to establish a microstructure-centric framework connecting flow field design innovations (conventional, bio-inspired, fractal, and topology-optimized) with their manufacturability and coating requirements. By identifying critical gaps at this intersection—particularly the absence of design-for-manufacturing guidelines for topology-optimized geometries and coating uniformity models for additively manufactured surfaces—this work aims to guide future research toward co-optimized rather than isolated solutions.

2 Flow field microstructure design and optimization

The optimization of flow field microstructures has evolved from empirical channel modifications to physics-driven computational design methodologies that maximize power density while minimizing parasitic losses. Conventional parallel channels offer low pressure drop but frequently exhibit reactant maldistribution and poor water drainage, whereas serpentine configurations ensure superior reactant delivery and water purging at the expense of elevated pressure losses and significant concentration gradients along the flow path (Li and Sabir, 2005; Chen et al., 2014). Interdigitated designs promote forced convection through the gas diffusion layer by creating dead-end channels that expel reactants under pressure, thereby enhancing mass transfer coefficients and mitigating flooding; however, the associated high pressure drop demands increased auxiliary power, reducing net system efficiency (Al-Baghdadi, 2011; Peng et al., 2020). Hybrid configurations that combine serpentine channels with embedded pin or obstacle arrays have been proposed to mitigate the trade-off between pressure drop and mass transfer inherent in pure serpentine or interdigitated designs. Al-Baghdadi (2011) numerically investigated a hybrid serpentine–pin flow field under variable operating conditions, demonstrating that localized turbulence generated at pin obstacles enhances convective transport toward the gas diffusion layer without imposing the extreme parasitic losses of fully interdigitated layouts. Such designs maintain relatively uniform reactant distribution even at reduced stoichiometry, offering operational flexibility that pure serpentine channels cannot achieve. However, pin placement density must be optimized to avoid excessive pressure accumulation, and manufacturing feasibility depends on stamping or etching resolution for sub-millimeter pin features. To overcome these dichotomies, bio-inspired flow fields derived from natural transport networks—such as leaf venation, lung alveoli, and river delta branching—have been proposed to achieve homogeneous reactant distribution with minimal hydraulic resistance (Kloess et al., 2009; Cho et al., 2012). Fractal branching geometries, in particular, generate hierarchical flow networks that scale self-similarly from inlet to terminal branches, reducing pressure drop by up to 50% relative to serpentine configurations under comparable flow rates (Kloess et al., 2009; Al-Baghdadi, 2011) while improving current density uniformity compared to conventional serpentine plates (Kloess et al., 2009; Al-Baghdadi, 2011). Systematic performance assessment of bio-inspired designs against conventional benchmarks under identical operating conditions has validated their advantages in reactant uniformity, though with important caveats. Kloess et al. (2009) compared leaf-venation and fractal branching networks directly with parallel and serpentine configurations, confirming hierarchical designs achieve lower pressure drop and improved current density uniformity. However, these gains are sensitive to manufacturing fidelity—particularly the reproduction of branching angles and terminal channel widths below 300 μm. This implies that bio-inspired optimization must incorporate manufacturability constraints during the design phase rather than as a post hoc consideration, a requirement that becomes more stringent as geometric complexity increases. Three-dimensional microstructures, including sub-millimeter baffles, dimples, and stepped channel profiles, disrupt the boundary layer within the channel, enhancing convective mass transfer toward the gas diffusion layer and accelerating liquid water detachment from hydrophobic surfaces (Park et al., 2016; Fu et al., 2008). Computational fluid dynamics models that resolve the Navier-Stokes equations coupled with electrochemical reaction kinetics and water transport have become indispensable tools for evaluating novel microstructures prior to experimental fabrication (Andersson et al., 2016; Park et al., 2016). Topology optimization represents the most advanced design paradigm, employing adjoint methods and density-based algorithms to generate non-intuitive channel layouts that maximize reactant concentration uniformity across the catalyst layer while constraining pressure losses and manufacturing complexity (Xia et al., 2024; Park et al., 2016). Recent studies demonstrate that topology-optimized flow fields can improve power density by 20%–30% relative to conventional designs, though their geometric complexity often necessitates additive manufacturing rather than conventional stamping (Xia et al., 2024). Furthermore, the cross-sectional morphology of channels—whether rectangular, trapezoidal, triangular, or semicircular—significantly influences the velocity profile, secondary flow generation, and contact area with the gas diffusion layer, with trapezoidal and wavy cross-sections showing particular promise for enhancing net power output in metallic plates (Khazaee et al., 2012; Wang et al., 2008). The integration of these microstructural innovations with metallic substrates demands careful consideration of electrical contact resistance at the channel-land interface, as reduced contact area in complex 3D geometries can offset the mass-transfer benefits unless surface coatings and compression loads are co-optimized (Peng et al., 2020; Aichele et al., 2021). Quantitative comparison across representative studies reveals consistent performance trade-offs under comparable operating conditions (H₂/air, 70 °C–80 °C, 1–3 atm). Parallel channels achieve pressure drops below 2 kPa but power densities remain below 0.4 W/cm² due to severe reactant maldistribution (Li and Sabir, 2005). Serpentine configurations improve power density to approximately 0.52 W/cm² at the cost of 15–25 kPa pressure losses and significant concentration gradients along the flow path (Chen et al., 2014). Interdigitated designs further enhance mass transfer to 0.61 W/cm² but impose parasitic loads of 35–50 kPa, reducing net system efficiency (Al-Baghdadi, 2011). Fractal branching networks reduce pressure drop by up to 50% relative to serpentine while maintaining comparable power density (0.55 W/cm²), though terminal branch widths below 300 μm necessitate additive manufacturing (Kloess et al., 2009). Topology-optimized geometries achieve the highest reported power density (0.68 W/cm²) by adaptively redistributing reactant, yet their non-intuitive layouts with variable cross-sections remain constrained by SLM resolution limits and post-processing requirements (Xia et al., 2024). To facilitate systematic comparison, Table 1 summarizes key geometric, operational, and performance parameters across representative studies. Fractal branching geometries, in particular, generate hierarchical flow networks that scale self-similarly from inlet to terminal branches, reducing pressure drop by up to 50% while improving current density uniformity compared to conventional serpentine plates (Kloess et al., 2009; Al-Baghdadi, 2011).

Simulation Limitations. Single-phase CFD neglects liquid water blockage, potentially causing significant ICR underestimation (Aichele et al., 2021). Two-phase models require empirical relations for droplet behavior, introducing uncertainty in pressure drop predictions (Park et al., 2016). Electrochemical coupling assumes uniform catalyst utilization, yet experiments show 40%–60% variation due to maldistribution (Peng et al., 2020). Contact resistance models ignore localized rib pressure, causing 2–3× ICR underestimation (Aichele et al., 2021). Mechanical compression reduces channel depth, shifting flow distribution (Peng et al., 2020). Degradation mechanisms—coating delamination, corrosion, catalyst poisoning—are rarely integrated (Ren et al., 2021). Validation against neutron imaging or transparent cell data remains sparse (Fu et al., 2008). Computational cost escalates from hours (single-phase) to weeks (topology optimization); machine learning surrogates offer seconds-scale prediction but require extensive training data and lack physical interpretability (Xia et al., 2024).

3 Manufacturing technologies for metallic bipolar plates

The manufacturability of optimized flow field microstructures is constrained by the geometric complexity of channel layouts. Stamping formability limits achievable aspect ratios and corner radii that define flow field microstructure; hydroforming precision determines channel depth uniformity that affects reactant distribution; additive manufacturing resolution dictates the minimum feature size of topology-optimized or fractal geometries. This section examines how each manufacturing technology imposes specific constraints on flow field microstructure design and the resulting trade-offs between geometric fidelity and performance (Hu et al., 2014; Mahabunphachai and Koc, 2010). Numerical simulations of the stamping process reveal that optimized punch velocity, blank holder force, and die clearance are essential to prevent tearing in channel corners and to maintain uniform channel depth across large active areas (Hu et al., 2014; Lee et al., 2004). Micro-hydroforming enables the fabrication of more complex channel profiles with improved surface finish and reduced tool wear compared to conventional stamping, though cycle times are longer and equipment costs are higher (Mahabunphachai and Koc, 2010). Photochemical etching and laser machining provide alternative routes for prototyping and small-batch production, delivering channel resolutions down to 100 μm without mechanical tool contact, yet throughput limitations and surface roughness from thermal ablation restrict their scalability (Wind et al., 2014). Additive manufacturing has emerged as a transformative technology for realizing topology-optimized and bio-inspired flow fields that are inaccessible to conventional forming processes. Selective laser melting of stainless steel and titanium powders enables the direct fabrication of lattice-based, mesh-type, and fractal channel geometries with dimensional tolerances within ±50 μm, while electron beam melting offers superior energy efficiency for titanium alloys though with coarser surface finishes (Porstmann et al., 2020). Post-processing steps—including sandblasting, electropolishing, and laser surface remelting—are invariably required for additively manufactured metallic plates to reduce surface roughness, eliminate partially fused particles, and achieve the interfacial contact resistance targets stipulated by the DOE (Porstmann et al., 2020). The economic viability of metallic BPP manufacturing depends critically on cycle time, tool longevity, and material yield; consequently, hybrid approaches that combine high-speed stamping for simple channel arrays with precision machining or coating for functional surface modifications are gaining traction as cost-effective pathways toward mass production (Wind et al., 2014; Aichele et al., 2021).

Critical comparison reveals distinct trade-offs for thin metallic bipolar plates. Stamping achieves lowest cost (5–15/plate, 2–5 s) but is limited to aspect ratios <3:1; springback in thin foils causes ±10–20 μm depth variation (Hu et al., 2014). Hydroforming enables higher aspect ratios (up to 5:1) with better surface finish (Ra 0.4–0.8 μm), yet 3–5× higher tool cost and 30–60 s cycles restrict scalability (Mahabunphachai and Koc, 2010). Etching/laser machining offer prototype flexibility without hard tooling, but undercuts and recast layers (Ra >5 μm) require electropolishing (+8–12/plate) (Wind et al., 2014). SLM uniquely realizes topology-optimized geometries (±50 μm), yet as-built roughness (Ra 10–20 μm) mandates sandblasting and electropolishing, and thermal distortion limits thin-wall compatibility (Porstmann et al., 2020). Hybrid approaches balance cost and precision (±40 μm, $20–50/plate) but introduce alignment risks (Wind et al., 2014). For automotive production, stamping dominates conventional designs; SLM is reserved for topology-optimized plates where significant power density gain may justify the higher manufacturing cost associated with SLM (Porstmann et al., 2020; Wind et al., 2014).

4 Protective coatings and surface engineering

Flow field microstructure directly dictates coating performance and durability. Ribs and sharp corners in stamped or hydroformed channels create localized stress concentrations that compromise coating adhesion; deep channels with high aspect ratios (>3:1) impede uniform deposition, leading to potentially causing significant thickness variation between land top and channel bottom (Ren et al., 2021); additively manufactured surfaces with roughness (Ra 10–20 μm) require additional coating thickness to ensure conformal coverage (Porstmann et al., 2020), increasing ICR, increasing ICR. This section examines how protective coating strategies must be co-designed with flow field microstructure to simultaneously achieve corrosion resistance, low interfacial contact resistance, and conformal coverage across geometrically complex channel-land interfaces (Antunes et al., 2010; Wang and Northwood, 2007). Protective coatings must therefore simultaneously provide excellent corrosion resistance, low electrical contact resistance below 10 mΩ cm², strong adhesion to metallic substrates, and minimal pinhole defects that would permit localized pitting corrosion (Ren et al., 2021; Ji et al., 2024). Carbon-based coatings, particularly diamond-like carbon and amorphous carbon films, have attracted significant attention due to their chemical inertness, low coefficient of friction, and tunable electrical conductivity through sp²/sp³ hybridization ratio control; recent investigations demonstrate that 100 nm DLC (diamond-like carbon) coatings on SS316L achieve corrosion current densities as low as 0.11 μA/cm² while maintaining hydrophobicity conducive to water management (Wang et al., 2006; Wang et al., 2025). Nitride coatings—including titanium nitride, chromium nitride, and multilayered TiAlN—offer exceptional hardness and wear resistance alongside acceptable electrical conductivity, with cathodic arc physical vapor deposition producing dense, adherent films that effectively seal substrate porosity (Lee and Park, 2015; Ramakrishnan et al., 2025). Metal oxide coatings such as TiO₂, Nb-doped TiO₂, and fluorine-doped tin oxide present cost-effective alternatives that leverage sol-gel, magnetron sputtering, or plasma-enhanced chemical vapor deposition techniques, though their inherently higher resistivities require careful stoichiometric control and doping strategies to meet conductivity targets (Hwang et al., 2011; Lee et al., 2010). Conductive polymer coatings, including polyaniline and polypyrrole, provide corrosion protection through passivation and barrier effects while offering facile electrodeposition; however, their long-term thermal stability and mechanical durability under cyclic compression remain insufficient for automotive lifetime requirements (Ren et al., 2021). Surface texturing and micro-patterning prior to coating deposition—creating sub-micron roughness or hierarchical structures—have been shown to improve coating adhesion and extend service life by distributing mechanical stresses and disrupting crack propagation pathways (Ji et al., 2024). The selection of coating technology must be harmonized with the flow field microstructure, as complex 3D geometries produced by additive manufacturing or hydroforming may exhibit non-uniform coating thicknesses in recesses and sharp corners, necessitating advanced deposition techniques such as atomic layer deposition or directed vapor deposition to ensure conformal coverage (Ren et al., 2021; Porstmann et al., 2020).

5 Conclusions and future perspectives

In synthesizing the state-of-the-art reviewed herein, Table 2 provides a consolidated reference that juxtaposes the dominant flow field configurations, coating strategies, manufacturing processes, and simulation tools discussed in the preceding sections. This simplified comparison highlights the persistent dichotomies—between pressure drop and uniformity, cost and performance, geometric complexity and manufacturability—that must be navigated in practical bipolar plate design, while identifying the convergent trends toward topology-optimized geometries, multi-layer coatings, hybrid manufacturing, and machine learning-accelerated simulation that define the future research frontier.

The optimization of flow field microstructures for metallic bipolar plates represents a multidisciplinary endeavor integrating fluid dynamics, electrochemistry, materials science, and advanced manufacturing. This review has highlighted that conventional channel configurations are progressively being supplanted by bio-inspired fractal networks, three-dimensional baffled geometries, and topology-optimized architectures that enhance reactant uniformity and water management while reducing parasitic pressure losses. The realization of these sophisticated microstructures depends upon the maturation of additive manufacturing technologies—particularly selective laser melting—and their integration with high-throughput stamping for cost-effective mass production. Concurrently, protective coating systems must evolve beyond single-layer films toward multifunctional, self-healing, and conformally deposited coatings that maintain low interfacial contact resistance and corrosion immunity over automotive lifetime targets (DOE, 2020; Peng et al., 2020). Future research should prioritize multi-objective optimization frameworks that simultaneously address electrochemical performance, mechanical durability, manufacturability, and cost, rather than optimizing flow field geometry in isolation. Digital twin technologies, coupling real-time operational data with high-fidelity multiphysics simulations, offer promising avenues for predictive maintenance and adaptive control of stack performance. Furthermore, the development of in situ diagnostic techniques—such as neutron imaging and transparent cell visualization—will provide unprecedented insights into two-phase flow dynamics within metallic microchannels, enabling data-driven refinement of computational models. Standardized accelerated testing protocols that replicate automotive drive-cycle conditions are essential to qualify novel microstructures and coatings for commercial deployment. Ultimately, the convergence of topology optimization, precision additive manufacturing, and advanced surface engineering will define the next-generation of metallic bipolar plates, accelerating the widespread adoption of PEMFCs in heavy-duty transportation and stationary power generation.

References

• 1

  AicheleA.MannußO.SauerA. (2021). Application of the dynamic tolerancing approach to the fuel cell stack assembly. CIRP Ann.70 (1), 419–422. 10.1016/j.cirp.2021.04.095

  • CrossRef
  • Google Scholar
  • View reference in article

• 2

  Al-BaghdadiM. A. R. S. (2011). CFD and experimental study of proton exchange membrane fuel cell using conventional and non-conventional flow field designs. Int. J. Hydrogen Energy36 (15), 9091–9104. 10.1016/j.ijhydene.2011.04.196

  • CrossRef
  • Google Scholar
  • View reference in article

• 3

  AnderssonM.BealeS. B.EspositoA.LehnertW.YongK. M. (2016). A review of cell-scale multiphase flow modeling in fuel cells. Electrochimica Acta216, 198–211. 10.1016/j.electacta.2016.08.135

  • CrossRef
  • Google Scholar
  • View reference in article

• 4

  AntunesR. A.OliveiraM. C. L.EttG.EttV. (2010). Corrosion of metallic bipolar plates for PEM fuel cells—A review. Int. J. Hydrogen Energy35 (8), 3632–3647. 10.1016/j.ijhydene.2010.01.059

  • CrossRef
  • Google Scholar
  • View reference in article

• 5

  ChenF.ChangM.HwangA. (2014). Design and optimization of the flow field pattern for proton exchange membrane fuel cells. Energy Convers. Manag.86, 699–708. 10.1016/j.enconman.2014.05.066

  • CrossRef
  • Google Scholar
  • View reference in article

• 6

  ChoH.MinK.LeeE.NamJ. H. (2012). Three-dimensional numerical simulation for the all-porous flow field of a unit proton exchange membrane fuel cell. Int. J. Hydrogen Energy37 (1), 710–723. 10.1016/j.ijhydene.2011.09.071

  • CrossRef
  • Google Scholar
  • View reference in article

• 7

  FuY.HouM.XuH.HouJ.ShaoZ. (2008). Transparent flow field for direct observation of the liquid water behavior in the channel of a proton exchange membrane fuel cell. J. Power Sources185 (2), 1040–1046. 10.1016/j.jpowsour.2008.07.025

  • CrossRef
  • Google Scholar
  • View reference in article

• 8

  HermannA.ChaudhuriT.SpagnolP. (2005). Bipolar plates for PEM fuel cells: a review. Int. J. Hydrogen Energy30 (12), 1297–1302. 10.1016/j.ijhydene.2005.04.008

  • CrossRef
  • Google Scholar
  • View reference in article

• 9

  HuQ.ZhangD.FuH.HuangK. (2014). Investigation of stamping process of metallic bipolar plates in PEM fuel cell—Numerical simulation and experiments. Int. J. Hydrogen Energy39 (25), 13770–13776. 10.1016/j.ijhydene.2014.07.027

  • CrossRef
  • Google Scholar
  • View reference in article

• 10

  HwangJ. J.HuJ.LinC. H.WuK. Y. (2011). Development of metal bipolar plates for proton exchange membrane fuel cells. Energy36 (5), 2862–2868. 10.1016/j.energy.2011.02.018

  • CrossRef
  • Google Scholar
  • View reference in article

• 11

  JiX.WangH.LiuY.ChenZ. (2024). Advancements and challenges in corrosion-resistant coatings for metallic bipolar plates: enhancing durability and sustainability in proton exchange membrane fuel cells. Discov. Appl. Sci.7 (1), 100. 10.1007/s44251-025-00100-w

  • CrossRef
  • Google Scholar
  • View reference in article

• 12

  KhazaeeI.GhazikhaniM.RavaA. (2012). Experimental investigation of the effect of channel shape on the performance of a PEM fuel cell. Int. J. Hydrogen Energy37 (22), 17106–17113. 10.1016/j.ijhydene.2012.08.147

  • CrossRef
  • Google Scholar
  • View reference in article

• 13

  KloessJ. P.WangX.LiuJ.ShiZ.GuessousL. (2009). Investigation of bio-inspired flow channel designs for bipolar plates in proton exchange membrane fuel cells. J. Power Sources188 (1), 132–140. 10.1016/j.jpowsour.2008.11.082

  • CrossRef
  • Google Scholar
  • View reference in article

• 14

  LeeC. H.ParkH. S. (2015). Evaluation of TiN-coated type 316 stainless steel bipolar plate. Prog. Nat. Sci. Mater. Int.25 (4), 365–371. 10.1016/j.pnsc.2015.06.001

  • CrossRef
  • Google Scholar
  • View reference in article

• 15

  LeeS. J.LeeD. J.YangY. S. (2004). A study on the forming of aluminum bipolar plates with micro-channels for fuel cell. J. Mater. Process. Technol.147 (3), 354–359. 10.1016/j.jmatprotec.2003.10.014

  • CrossRef
  • Google Scholar
  • View reference in article

• 16

  LeeS. J.HuangC. H.ChenY. P.ChiuK. F. (2010). Corrosion-resistant component for proton exchange membrane fuel cells. J. Power Sources195 (21), 7338–7343. 10.1016/j.jpowsour.2010.05.017

  • CrossRef
  • Google Scholar
  • View reference in article

• 17

  LiX.SabirI. (2005). Review of bipolar plates in PEM fuel cells: Flow-Field designs. Int. J. Hydrogen Energy30 (4), 359–371. 10.1016/j.ijhydene.2004.09.016

  • CrossRef
  • Google Scholar
  • View reference in article

• 18

  MahabunphachaiS.KocM. (2010). Investigation of micro/meso scale sheet hydroforming for bipolar plate production. J. Mater. Process. Technol.210 (2), 356–364. 10.1016/j.jmatprotec.2009.09.014

  • CrossRef
  • Google Scholar
  • View reference in article

• 19

  MansoA. P.MarzoF. F.BarrancoJ.GarikanoI.ValinoA. A. (2012). Influence of geometric parameters of the flow fields on the performance of a PEM fuel cell. Int. J. Hydrogen Energy37 (22), 17278–17287. 10.1016/j.ijhydene.2012.08.144

  • CrossRef
  • Google Scholar
  • View reference in article

• 20

  ParkS.KimJ.KwonS. (2016). Design optimization of the flow field pattern for a fuel cell bipolar plate using computational fluid dynamics. Int. J. Hydrogen Energy41 (4), 2359–2368. 10.1016/j.ijhydene.2015.12.112

  • CrossRef
  • Google Scholar
  • View reference in article

• 21

  PengL.HuP.CaoX.YiP. (2020). Review on bipolar plates and flow fields for proton exchange membrane fuel cells. Int. J. Hydrogen Energy45 (43), 23581–23601. 10.1016/j.ijhydene.2020.06.220

  • CrossRef
  • Google Scholar
  • View reference in article

• 22

  PorstmannS.WannemacherT.DrosselW. G. (2020). A comprehensive comparison of state-of-the-art manufacturing methods for fuel cell bipolar plates including anticipated future industry trends. J. Manuf. Process.60, 366–383. 10.1016/j.jmapro.2020.10.040

  • CrossRef
  • Google Scholar
  • View reference in article

• 23

  RamakrishnanS.RajalakshmiN.ValletiK.OmarS.RamyaK. (2025). Microscopic analysis on high corrosion-resistant metallic nitride-coated metallic bipolar plate for PEM fuel cell application. Mater. Today Commun.42, 110–118. 10.1016/j.mtcomm.2025.110118

  • CrossRef
  • Google Scholar
  • View reference in article

• 24

  RenC.LiM.ZhangL. (2021). Recent advances in coatings for metallic bipolar plates in PEM fuel cells. Surf. Coatings Technol.421, 127478. 10.1016/j.surfcoat.2021.127478

  • CrossRef
  • Google Scholar
  • View reference in article

• 25

  SevincH.HazarH. (2026). Performance enhancement of PEM fuel cells using novel porous flow field designs. Energy Convers. Manag.353, 121201. 10.1016/j.enconman.2026.121201

  • CrossRef
  • Google Scholar
  • View reference in article

• 26

  WangJ. (2012). Barriers of scaling-up fuel cells: cost, durability and reliability. J. Power Sources203, 1–6. 10.1016/j.jpowsour.2011.12.016

  • CrossRef
  • Google Scholar
  • View reference in article

• 27

  WangY.NorthwoodD. O. (2007). Effects of O₂ and H₂ on the corrosion of SS316L metallic bipolar plate materials in a simulated fuel cell environment. Int. J. Hydrogen Energy32 (16), 3893–3901. 10.1016/j.ijhydene.2007.03.036

  • CrossRef
  • Google Scholar
  • View reference in article

• 28

  WangS. H.PengJ.LuiW. B. (2006). Properties of diamond-like carbon films coating on SS316L as bipolar plate for proton exchange membrane fuel cells. Int. J. Hydrogen Energy31 (6), 774–779. 10.1016/j.ijhydene.2005.06.014

  • CrossRef
  • Google Scholar
  • View reference in article

• 29

  WangX. D.DuanY. Y.YanW. M. (2008). Numerical study of cell performance and local transport phenomena in PEMFCs with conventional and wave-like flow field designs. J. Power Sources175 (1), 397–407. 10.1016/j.jpowsour.2007.09.047

  • CrossRef
  • Google Scholar
  • View reference in article

• 30

  WangS.LiM.ZhangY.ChenH. (2025). Diamond-like-carbon coated metallic bipolar plates for PEM fuel cells: an assessment of coating thickness effect. Fuel Cells25 (1), e70036. 10.1002/fuce.70036

  • CrossRef
  • Google Scholar
  • View reference in article

• 31

  WilberforceT.AlaswadA.PalumboA.DassistiM.OlabiA. G. (2016). Advances in stationary and portable fuel cell applications. Int. J. Hydrogen Energy41 (37), 16429–16440. 10.1016/j.ijhydene.2016.06.223

  • CrossRef
  • Google Scholar
  • View reference in article

• 32

  WindJ.LaCroixA.BraeuningerS.HedrichP.HellerC.SchudyM. (2014). Impact of materials and manufacturing processes on the cost of fuel cell bipolar plates. Int. J. Hydrogen Energy39 (27), 15242–15251. 10.1016/j.ijhydene.2014.03.028

  • CrossRef
  • Google Scholar
  • View reference in article

• 33

  XiaY.GuoC.DongE.ChenL.TaoW. (2024). Topology optimization of gas channels in proton exchange membrane fuel cells. Int. J. Heat Mass Transf.222, 125147. 10.1016/j.ijheatmasstransfer.2024.125147

  • CrossRef
  • Google Scholar
  • View reference in article