PEM electrolyzer bipolar plate corrosion resistance

The Corrosion Challenge Inside a PEM Electrolyzer

Bipolar plates in proton exchange membrane (PEM) water electrolyzers operate under conditions that few engineering materials can withstand without degradation: pH below 2, temperatures of 50–80°C, current densities above 2 A/cm², and oxidative potentials exceeding 1.8 V vs. RHE at the anode. This combination drives rapid corrosion of unprotected metals and gradual passivation of others, both of which increase interfacial contact resistance (ICR) and reduce stack efficiency over time.

The standard response to this challenge has been to apply platinum group metal (PGM) coatings — primarily platinum or gold — to titanium substrates. While effective, this approach carries two compounding cost penalties: the PGM material itself, and the expense of titanium at $15–20/kg. As noted by research cited in WIPO-tracked patent filings and peer-reviewed literature, the commercial cost target for the entire bipolar plate — material, coating, and manufacturing combined — is below $10 per kilowatt. Achieving that target demands a fundamentally different approach.

The solution space is not a single silver bullet. It is a framework of four complementary strategies: advanced non-noble metal coating systems; substrate material optimisation and surface pre-treatment; multi-layer and gradient coating architectures; and novel deposition techniques. Used in combination, these strategies can meet or exceed the performance of PGM coatings while delivering cost reductions of 50–80% compared to coated titanium substrates.

Advanced Non-Noble Metal Coatings That Replace PGMs

The most direct path to eliminating PGM cost is replacing those coatings with alternative materials that match their electrochemical performance. Four material families have demonstrated the necessary combination of corrosion resistance and electrical conductivity for PEM electrolyzer service.

Transition Metal Nitrides

Chromium nitride (CrN) and its modified variants are among the most extensively validated alternatives. CrN achieves ICR values below 10 mΩ·cm² while maintaining corrosion current densities below 1 μA/cm² in simulated PEMWE environments — meeting both the DOE conductivity target and the corrosion threshold in a single coating. Stoichiometry optimisation (targeting a slight nitrogen excess in the N/Cr ratio) produces a more compact structure with fewer pinholes, while doping with small amounts of Al, Si, or Ti to form Cr-Al-N or Cr-Si-N further improves oxidation resistance and reduces grain boundary corrosion pathways. Post-deposition annealing at 400–500°C in an inert atmosphere enhances crystallinity and reduces residual stress, improving both corrosion resistance and electrical conductivity simultaneously.

Niobium (Nb) coatings have demonstrated particularly strong long-term credentials. Research shows that Nb-coated stainless steel bipolar plates can operate for over 10,000 hours in actual electrolyzer environments with minimal degradation, achieving ICR below 15 mΩ·cm². The native niobium oxide (Nb₂O₅) that forms on the surface is both highly corrosion-resistant and sufficiently conductive — a combination that most passive oxide layers fail to achieve. Titanium nitride (TiN) offers excellent electrical conductivity (resistivity approximately 20–25 μΩ·cm), though hybrid approaches combining TiN with other nitrides or using N/Ti co-doped amorphous carbon coatings are required for the most aggressive anode-side conditions.

“Niobium-coated 316L stainless steel bipolar plates have demonstrated stable operation exceeding 10,000 hours with voltage degradation rates below 20 μV/h — at a substrate cost roughly 75–85% lower than titanium.”

Doped Oxide and Carbon-Based Coatings

Pure titanium dioxide (TiO₂) is an excellent corrosion barrier but an insulator. Nitrogen doping resolves this by creating oxygen vacancies and introducing mid-gap electronic states. Recent studies show N-doped TiO₂ (N-TiO₂) coatings achieving ICR values below 10 mΩ·cm² with corrosion current densities below 0.5 μA/cm² in 0.5 M H₂SO₄ at 80°C under anodic polarisation — performance that meets the DOE 2025 target on both metrics simultaneously. Fluorine, niobium, or tantalum doping of TiO₂ offers additional pathways to conductivity enhancement. Ti₄O₇ (Magnéli phase titanium oxide) exhibits metallic conductivity while retaining oxide-level corrosion resistance, making it well-suited to electrolyzer applications.

Carbon-based coatings offer a different cost-reduction vector. Reduced graphene oxide (rGO) deposited on 316L stainless steel via solution-based methods followed by thermal reduction creates a dense, adherent carbon layer that reduces ICR to 5–8 mΩ·cm² while providing corrosion current below 1 μA/cm². Amorphous carbon (a-C) and diamond-like carbon (DLC) coatings can be engineered with varying sp²/sp³ ratios to balance conductivity and barrier properties. Nanostructured carbon material (NCM) coatings using carbon nanotubes, nanofibers, or graphene nanoplatelets in polymer binders can be applied by spray, dip-coating, or screen printing — methods that significantly reduce manufacturing costs compared to vacuum deposition.

Substrate Selection and Surface Pre-Treatment

Coating performance is inseparable from the substrate it protects. Selecting a lower-cost substrate with inherently better corrosion resistance reduces the performance burden on the coating, enabling thinner — and cheaper — protective layers.

Moving Beyond Titanium

Titanium is the gold standard for PEMWE bipolar plates, but at $15–20/kg it represents a significant cost driver. Stainless steel alternatives — particularly 316L, 904L, and 2205 duplex stainless steel — cost $2–5/kg. The primary challenge with stainless steel is the formation of insulating passive oxide layers (primarily Cr₂O₃ and Fe₂O₃/Fe₃O₄) that increase contact resistance. However, properly coated stainless steel can achieve performance comparable to coated titanium: 316L with a niobium coating has shown stable performance over 10,000 hours with voltage degradation rates below 20 μV/h. The 904L grade offers higher Cr and Mo content for enhanced baseline corrosion resistance, reducing the performance requirements on the protective coating. The ferritic-austenitic dual-phase structure of 2205 duplex stainless steel provides improved pitting resistance and mechanical strength.

Copper-based substrates represent a viable option for cathode-side bipolar plates specifically. The cathode operates under reducing conditions and lower potentials, making properly coated copper viable where it would fail on the anode side. Copper’s exceptional electrical conductivity reduces ohmic losses, and its cost is significantly below titanium — but this approach requires hermetic coating systems with zero pinhole density and robust edge sealing to prevent copper ion migration. Aluminium alloys (particularly 6061 and 5083) offer excellent formability and very low cost; surface modification strategies including chemical conversion coatings, anodisation, and subsequent conductive coating deposition have shown promise for electrolyzer adaptation.

Surface Pre-Treatment Strategies

Surface pre-treatment before coating deposition can dramatically improve adhesion, reduce coating defect density, and add an intrinsic corrosion-resistance layer that protects the substrate even if coating defects occur. Gas nitridation of stainless steel or iron-based alloy substrates at 400–550°C for 4–8 hours in NH₃ or N₂-H₂ atmospheres creates a 5–20 μm case-hardened layer without dimensional changes — making it compatible with pre-stamped bipolar plates. This nitrogen-enriched layer enhances substrate hardness, improves coating adhesion through better lattice matching, and acts as a diffusion barrier preventing substrate elements from migrating into the coating.

Plasma surface modification offers a lower-temperature alternative (300–400°C for plasma nitriding versus 400–550°C for gas nitridation) and can be performed in-line with coating deposition, eliminating a separate processing step and reducing total manufacturing cost. Chemical surface treatments — including chromate-free conversion coatings using Zr, Ti, or Ce-based systems, controlled acid pickling, and brief electrochemical surface activation — provide complementary adhesion promotion and surface standardisation before coating.

Multi-Layer and Gradient Coating Architectures

Rather than relying on a single thick coating, multi-layer systems leverage the complementary properties of different materials while maintaining overall coating thickness below that of equivalent PGM coatings — reducing material cost and internal stress simultaneously.

Barrier Layer + Conductive Layer Architecture

The most common and validated multi-layer approach uses a thin, dense barrier layer (typically 100–500 nm) that provides the primary corrosion protection, topped with a highly conductive layer (200–1,000 nm) that ensures low contact resistance with the gas diffusion layer. TiN as the barrier layer combined with CrN as the conductive surface layer is one established combination: TiN provides excellent barrier properties while CrN offers superior conductivity and surface hardness. Niobium or niobium oxide as a base layer paired with a doped oxide top layer offers another route. Carbon-based barriers combined with thin metal nitride surfaces ensure both impermeability and wear resistance in a single stack.

Nanolaminate and Composition-Graded Structures

Alternating nanoscale layers (individual layer thickness 5–50 nm) of two different materials create interfaces that impede corrosion pathways through the coating. The N/Ti co-doped amorphous carbon pseudo-multilayer architecture alternates nitrogen-doped and titanium-doped amorphous carbon layers to create a tortuous path for electrolyte penetration while maintaining conductivity through the Ti-rich layers. CrN/AlN nanolaminates use the interfaces between CrN and AlN layers to deflect crack propagation and provide redundant barrier layers. Metal/ceramic nanocomposites — embedding TiO₂, Al₂O₃, or ZrO₂ nanoparticles in a metallic matrix — combine metal conductivity with ceramic corrosion resistance in a single co-deposited layer.

Functionally graded materials (FGMs) with continuously varying composition from substrate to surface eliminate sharp interfaces that can serve as delamination sites. A gradient that transitions from pure substrate to pure coating material over 200–500 nm creates excellent adhesion, while varying composition to maximise barrier properties at the coating-electrolyte interface and maximise conductivity at the coating-GDL interface optimises performance at every depth. These graded structures can be deposited using magnetron sputtering with time-varying target power, pulsed laser deposition with sequential target ablation, or chemical vapour deposition with programmed precursor flow rates — all established industrial processes.

Deposition Techniques That Deliver More With Less Material

The deposition method directly determines coating density, defect content, adhesion, and manufacturing cost. Advances in deposition technology mean that equivalent or superior corrosion protection can now be achieved with significantly less coating material — cutting both PGM-free and non-PGM material costs.

High-Power Impulse Magnetron Sputtering (HiPIMS)

HiPIMS represents a significant advancement over conventional DC or RF magnetron sputtering. The higher ionisation fraction of sputtered species results in denser coatings with fewer defects, better adhesion due to ion bombardment during deposition, smoother surfaces that reduce contact resistance with the gas diffusion layer, and the ability to deposit at lower substrate temperatures compatible with pre-assembled components. Critically, the higher density and lower defect content of HiPIMS coatings means equivalent corrosion protection can be achieved with 30–50% less coating thickness compared to conventional sputtering — directly offsetting the somewhat higher equipment costs of the process. For non-PGM coatings where material cost is already low, this thickness reduction primarily translates to faster throughput and lower deposition time per plate.

Atomic Layer Deposition and Plasma-Enhanced ALD

Atomic layer deposition (ALD) provides unparalleled conformality and thickness control, enabling ultra-thin coatings (10–100 nm) with zero pinholes. This dramatically reduces material costs while achieving perfect step coverage on complex flow field geometries — a geometry challenge that line-of-sight PVD techniques struggle with. Plasma-enhanced ALD (PEALD) operates at lower temperatures (150–300°C versus 300–500°C for thermal ALD) and enables deposition of conductive nitrides and oxynitrides not accessible by thermal ALD. PEALD-deposited zirconium oxynitride (ZrOₓNᵧ) coatings (10–50 nm) combine the corrosion resistance of zirconia with conductivity enhancement from nitrogen incorporation, achieving ICR below 5 mΩ·cm² — the lowest of any non-PGM system reviewed. Recent developments in spatial ALD and roll-to-roll ALD systems are reducing equipment costs toward commercial viability for large-area bipolar plate manufacturing, as tracked by patent databases including those indexed by EPO.

Solution-Based and Electrochemical Deposition

Solution-based methods eliminate vacuum equipment entirely, dramatically reducing capital costs and enabling continuous roll-to-roll processing that can coat both sides of a bipolar plate simultaneously. Graphene oxide or rGO dispersions can be spray-coated, dip-coated, or spin-coated onto substrates and then thermally or chemically reduced to create conductive graphene coatings. Metal-organic precursor solutions enable deposition of oxide coatings by spray pyrolysis or spin coating followed by thermal conversion. Electrodeposition — including pulse electrodeposition for finer grain structures and composite electrodeposition co-depositing hard ceramic nanoparticles (SiC, Al₂O₃, TiO₂) with metals like Ni, Co, or Ni-P — requires no vacuum equipment, scales to large areas, and can deposit complex alloys in a single step. Research on green hydrogen manufacturing cost reduction, tracked by organisations including IRENA, consistently identifies manufacturing process cost as a key lever alongside materials cost.

Hybrid physical-chemical vapour deposition combines PVD deposition of a dense, adherent base layer with CVD deposition of a conformal, pinhole-free top layer in a single system, with in-situ plasma treatment between layers for interface optimisation. This approach is particularly effective for complex flow field geometries where pure PVD may produce shadowing effects that leave recessed areas undercoated.

Implementation Pathway: Near-, Medium-, and Long-Term Recommendations

Selecting the right coating strategy requires matching technology readiness to deployment timeline. The following tiered pathway, grounded in published research and patent evidence, provides a structured route from immediate deployment to breakthrough performance.

Near-Term (1–2 Years): Proven Technologies

The primary near-term recommendation is niobium-coated 316L stainless steel. Niobium coating by magnetron sputtering to a thickness of 0.5–1.5 μm (thinner than typical PGM coatings) delivers ICR below 15 mΩ·cm², a demonstrated lifetime exceeding 10,000 hours, and a 60–70% cost reduction versus coated titanium substrates. The alternative near-term option is nitrogen-doped TiO₂ on stainless steel at 200–500 nm thickness, delivering ICR below 10 mΩ·cm² with excellent long-term stability at a 50–60% cost reduction versus coated titanium. Both technologies are deployable with existing magnetron sputtering infrastructure and have been validated in actual electrolyzer environments.

Manufacturing strategies that further reduce assembly cost include selective coating — applying highest-performance coatings only to the active area flow fields and sealing regions while using lower-cost protective treatments for peripheral areas — and asymmetric anode/cathode coating that recognises the anode side faces significantly harsher conditions. Pre-stamping coating (applying coating to flat sheets before flow field forming) enables continuous roll-to-roll processing but requires ductile coating systems, such as thin niobium coatings, that can withstand forming strains without cracking. Research published in Nature-family journals on electrochemical materials science confirms that coating ductility and adhesion after forming are achievable design targets for these material systems.

Medium-Term (2–4 Years): Advanced Coatings

The primary medium-term recommendation is CrN/AlN nanolaminate or Cr-Al-C gradient coating on nitrided stainless steel, targeting total coating thickness of 0.8–1.2 μm, ICR below 8 mΩ·cm², and a projected lifetime exceeding 20,000 hours. The Cr-Al-C system offers a self-healing mechanism: when micro-defects occur, preferential oxidation of aluminium forms a thin, protective Al₂O₃ barrier that seals the defect without significantly increasing contact resistance. The reduced graphene oxide composite coating alternative targets ICR below 8 mΩ·cm² with good corrosion resistance at a 70–80% cost reduction versus coated titanium if solution-based manufacturing can be scaled — a significant “if” that makes this the higher-risk, higher-reward option.

Long-Term (4+ Years): Breakthrough Technologies

PEALD-deposited ZrOₓNᵧ or doped oxide ultra-thin coatings (50–200 nm) target ICR below 5 mΩ·cm² with excellent durability from zero-defect coating morphology. The cost impact depends on the commercialisation of spatial ALD or roll-to-roll ALD systems, which are currently in active development. Amorphous metal substrates — such as Ni₄₀Ti₄₀Nb₂₀ — offer inherently superior corrosion resistance due to the absence of grain boundaries (which are preferential corrosion sites), excellent formability in the supercooled liquid region, and the potential for thinner coatings (0.2–0.5 μm) or even uncoated operation in less aggressive regions. Amorphous metal bipolar plates project a lifetime exceeding 30,000 hours, though current manufacturing costs remain high. Composite substrate designs — thin corrosion-resistant outer layers metallurgically bonded to lower-cost structural cores via explosive bonding, roll bonding, or co-extrusion — offer a near-breakthrough path to using expensive alloys only where needed. The IEA has identified electrolyzer cost reduction as a critical pathway to green hydrogen cost parity, and bipolar plate economics are a central component of that challenge.

“A systems engineering approach that considers not just the coating itself, but the substrate, surface preparation, deposition method, flow field design, and operating conditions as an integrated whole is what separates incremental improvement from transformative cost reduction.”

Operational strategies provide a complementary lever: ultra-pure water with resistivity above 10 MΩ·cm reduces ionic contamination that accelerates corrosion; operating at the lower end of the temperature range (50–60°C versus 70–80°C) significantly reduces corrosion rates; and avoiding unnecessary over-potentials during start-up and shutdown reduces oxidative stress on coatings. These system-level measures extend coating life without increasing coating thickness or cost, and are compatible with all coating strategies described above.