Executive Summary

Fuel-cell stack sealing is a multi-interface fluid-isolation problem. Bipolar-plate geometry, gasket compression, stack alignment, load uniformity, hydrogen containment, coolant isolation, thermal and pressure cycling, chemical compatibility and leakage verification jointly determine reliability. The stack must separate anode hydrogen, cathode air or oxygen, coolant and the environment while maintaining contact around channels, manifolds and active areas.

Anode-to-cathode cross-leakage, gas-to-coolant leakage, external hydrogen leakage, coolant ingress and hydrogen permeation are different observations. A single external pressure-hold test cannot prove internal isolation. A gas concentration change can result from an internal path, membrane-electrode crossover, sampling conditions or sensor error. Reliable diagnosis therefore combines differential pressure, gas concentration, coolant condition, flow behavior, compression history and structural inspection.

Fuel Cell Stack Sealing Architecture

A stack is built from bipolar plates, membrane-electrode assemblies, gaskets, end plates and a compression structure. Bipolar plates form anode, cathode and coolant passages; the gasket groove defines the perimeter and manifold boundaries; end plates and tie rods distribute compression through the stack. Hydrogen inlet and outlet manifolds, air inlet and outlet manifolds and coolant inlet and outlet paths must remain isolated even when their boundaries are closely spaced.

Gas-channel perimeter gaskets separate the anode and cathode flow fields from the environment. Coolant-boundary gaskets separate the coolant channels from both gas paths. Manifold seals prevent one inlet or outlet from bypassing into another path. The bipolar-plate surface is part of the sealing interface: flatness, roughness, coating integrity and local damage determine whether nominal gasket compression becomes continuous contact. The membrane-electrode assembly is relevant to diagnosis because gas crossover through the electrochemical region is not automatically gasket leakage.

Table I: Fuel-Cell Stack Sealing Zones and Leakage Risks

The sealing zone, not the component name, determines the failure mechanism. A gasket can appear intact while a local plate gap causes leakage, and a plate crack can create a path that external inspection cannot reveal.

Gasket Compression and Stack Assembly

Gasket compression creates contact pressure, but the useful margin depends on groove geometry, material recovery, plate flatness and load distribution. Insufficient compression may leave a continuous or intermittent path. Excessive compression can accelerate compression-set loss, extrusion, tearing or local material damage. A nominal compression ratio therefore cannot be interpreted without the actual stack design and assembly condition.

Fastener sequence, end-plate stiffness and stack alignment influence compression uniformity. Local gaps may form near manifolds, corners or plate transitions even when the average stack load appears acceptable. Surface roughness can create micro-paths, while a damaged coating can change friction and contact. A displaced gasket may be compressed in one region and unsupported in another. Fixed bolt torque, compression or leakage values will not be treated as universal standards; they must follow the stack architecture, gasket material, pressure, temperature and manufacturer requirements.

Hydrogen Cross-Leakage and Coolant Isolation

The principal internal risks are anode-to-cathode leakage, anode-to-coolant leakage and cathode-to-coolant leakage. Anode hydrogen may cross through a failed gasket, manifold interface, bipolar-plate crack or local gap. Gas-to-coolant leakage can introduce hydrogen or air into the coolant loop without producing an external wet mark. Coolant ingress reverses the direction of concern: liquid enters a gas channel, potentially contaminating the flow path and changing local operation.

Hydrogen entering the cathode side can change gas composition and operating stability. Gas entering coolant can affect safety assessment, flow behavior and coolant condition. Coolant crossover can change composition, conductivity or contamination indicators. External hydrogen leakage is a boundary-to-environment event; internal cross-leakage is a boundary-to-boundary event. They require different isolation tests and should not be combined into one leakage category.

A gas concentration change must be interpreted with pressure, flow, sampling location, humidity and sensor response. Hydrogen permeation and membrane-electrode crossover may produce a gradual signal without an open gasket path. Conversely, a pressure-dependent step change may support a discrete path. Plate crack, manifold leakage, gasket damage and sensor or test-system error remain separate hypotheses until evidence identifies the path.

Table II: Compression, Thermal and Pressure Influence Matrix

The matrix supports test planning, but it does not replace a design-specific definition of media, pressure, temperature, compression state and measurement method.

Thermal, Humidity and Pressure-Cycle Effects

Temperature gradients create differential expansion between bipolar plates, gaskets, coatings and surrounding hardware. Humidity and wet-dry exposure can alter dimensions, modulus, surface wetting and chemical interaction. Start-stop cycling repeatedly changes pressure and compression state, while pressure cycling changes local contact loading. Where freeze-thaw exposure is relevant, expansion and contraction can impose additional stress on grooves, manifolds and damaged interfaces.

Gasket stress relaxation and compression-set development reduce recovery after operation. Plate deformation can concentrate load at one edge and reduce it at another. A stack that passes one state may develop intermittent cross-leakage after thermal, humidity or pressure changes. The record should connect cycle history, differential pressure, gas composition, coolant condition and post-test morphology.

Materials and Interface Compatibility

Material selection should follow the complete interface. EPDM, FKM, FVMQ and HNBR may be considered under different hydrogen, air-side and coolant conditions, but compatibility depends on formulation, temperature, humidity, pressure and compression history. PTFE-based elements and engineered polymer gaskets can offer useful chemical or friction behavior in suitable geometries, but support, cold flow, sealing recovery and manufacturing tolerance require confirmation.

Coated bipolar plates, graphite plates and composite plates present different surface and structural risks. Coating damage may create a local sealing discontinuity or change contact friction. Graphite or composite plates may respond differently to load, surface damage and thermal gradients. Metal or polymer manifold interfaces add another tolerance and stiffness combination. Particle generation, extractables, swelling, hardening and permeation can affect the interface without immediately producing external leakage.

No gasket, bipolar-plate material or coating is a universal solution. Material screening must be followed by assembled-stack testing because plate flatness, groove geometry, compression uniformity, pressure differential, humidity and cycling can dominate the observed leakage behavior.

Inspection, Testing and Maintenance

External pressure-hold testing checks the selected exterior boundary, but it does not replace internal verification. Hydrogen detection can identify an external release or sampling signal, while helium testing can screen small paths under its own conditions. Neither result alone represents long-term hydrogen service. Differential-pressure and gas-to-coolant crossover tests address internal isolation.

Gas monitoring must retain pressure, flow, humidity, sampling location and calibration. Coolant analysis can identify contamination or gas interaction, but cannot prove every gas channel is isolated. Pressure-decay testing identifies loss under a defined condition. Thermal and start-stop cycling expose history-dependent changes. Compression, gasket, plate-surface and manifold inspections help localize a path.

Post-test teardown should examine gasket indentation, compression set, extrusion, tearing, swelling, hardening, displacement, bipolar-plate deformation, cracks, coating damage and surface contamination. After stack disassembly or gasket replacement, requalification should confirm material identity, groove cleanliness, gasket orientation, stack alignment, compression condition and internal isolation before return to service.

Table III: Fuel-Cell Stack Cross-Leakage Verification Guide

Each result should retain stack identity, plate and gasket records, assembly sequence, compression condition, anode and cathode pressure, coolant pressure, temperature, humidity, cycle history, gas data, coolant analysis and corrective action.

Data Interpretation and Maintenance Planning

A useful record connects stack architecture with bipolar-plate material, gasket material and geometry, groove condition, compression, assembly sequence, anode pressure, cathode pressure, coolant pressure, temperature, humidity, cycle history, hydrogen concentration, pressure-decay result, coolant analysis, leakage trend, maintenance history and teardown morphology. This prevents a gas signal from being assigned to gasket ageing before plate, manifold, membrane-electrode and test-system causes are considered.

Maintenance may be calendar-based, operating-cycle-based, leakage-trend-based, gas-concentration-based, coolant-condition-based, post-shutdown or risk-based. Post-repair requalification should verify alignment, cleanliness, compression, gasket orientation and relevant internal isolation tests. Laboratory cycle counts and single-test results should not be converted directly into a universal fuel-cell system lifetime.

FMEA Risk Analysis

Table IV: Fuel-Cell Stack Sealing FMEA and RPN Analysis

The RPN values are illustrative engineering risk rankings, not field statistics or experimental results. They prioritize review; they do not replace design evidence or failure analysis.

Conclusion

Fuel-cell stack sealing depends on the combined behavior of bipolar-plate interfaces, gasket compression, stack alignment, anode-cathode-coolant separation, hydrogen containment and the compression structure. External hydrogen leakage, internal cross-leakage, gas-to-coolant leakage, coolant ingress, hydrogen permeation, membrane-electrode crossover, plate cracks and sensor error are different diagnostic categories.

Reliable verification must combine external and internal leakage tests, differential pressure, gas concentration, coolant condition, thermal and pressure-cycle history, compression inspection and post-test teardown. No single gasket, bipolar-plate material or compression arrangement eliminates every cross-leakage path. Maintenance reliability is established only after alignment, cleanliness, gasket condition, compression and internal isolation are requalified.

Engineering FAQ

Q:How does a fuel-cell stack gasket separate hydrogen, air and coolant?

A:The gasket follows dedicated grooves around gas fields, coolant channels and manifolds. Its compressed contact against the bipolar-plate surfaces forms separate boundaries, while end plates and the compression structure maintain load. Flatness, surface condition and compression uniformity determine whether those boundaries remain continuous.

Q:What is the difference between external hydrogen leakage and internal cross-leakage?

A:External hydrogen leakage releases hydrogen from the stack to the environment. Internal cross-leakage transfers gas between anode, cathode or coolant boundaries without necessarily producing an external signal. The two conditions require different port-isolation and detection methods.

Q:Why can a stack pass an external leak test but still have gas-to-coolant leakage?

A:An external test challenges the selected outside boundary. A local internal path between a gas channel and the coolant circuit may remain isolated from the exterior and therefore escape that test. Dedicated gas-to-coolant crossover testing and coolant analysis are required.

Q:How do temperature and pressure cycling affect bipolar-plate gasket sealing?

A:They change gasket modulus, dimensions, recovery and contact pressure while also deforming plates and interfaces. Repeated operation can create local gaps, compression-set loss, extrusion or intermittent leakage that is absent at one static condition.

Q:Which tests are needed after replacing fuel-cell stack gaskets?

A:Confirm gasket material, geometry, orientation, groove cleanliness, plate condition, alignment and compression. Then repeat the applicable external, internal differential-pressure, gas-to-coolant, pressure-decay and functional checks before release.

Q:How can hydrogen cross-leakage be distinguished from membrane-electrode gas crossover?

A:Compare pressure dependence, sampling location, flow behavior, coolant condition and stack-interface evidence. Membrane-electrode crossover may produce a gradual transport signal without a discrete gasket or plate path, whereas a pressure-dependent boundary change may indicate an interface or structural leak. Sensor calibration and test-fixture integrity must also be checked.