The stakes are considerable. The United States Department of Energy has set technical targets for bipolar plates ahead of anticipated fuel cell stack commercialization by 2030, demanding an in-plane electrical conductivity above 100 siemens per centimeter, a flexural strength above 40 megapascals, and robust chemical resistance, processability, and long-term reliability. Meeting all of those requirements simultaneously has proven difficult. Metallic plates offer strength and easy machining but corrode without expensive surface treatments. Pure graphite plates conduct beautifully and resist chemicals, yet they are brittle and hard to form into the thin, intricate shapes that modern stack designs demand. Resin-based carbon composites, which blend moldable polymer with conductive graphite fillers, have emerged as a compelling middle path, but their performance depends exquisitely on the microstructure that forms during fabrication.

Researchers led by Seon Ho Lee and Song Mi Lee of Chungnam National University and the Korea Institute of Energy Research attacked the problem from three directions at once: the loading of multi-walled carbon nanotubes, the particle size of the graphite filler, and the mixing method used to combine everything with epoxy resin. Carbon nanotubes are the stars of the conductive composite world. Roughly 20 nanometers in diameter and up to 50 micrometers long, they possess extraordinary intrinsic conductivity and tensile strength, and they readily form interconnected electron highways through a material. But they carry a notorious flaw. Strong van der Waals attractions between their nonpolar surfaces cause them to clump into agglomerates, and those clumps act as defects rather than reinforcements, degrading both the electrical and mechanical performance of the finished composite.

The team’s solution was elegantly practical. In the conventional dry mixing route, graphite and nanotubes are simply stirred into the resin, leaving stubborn clumps intact. In the wet mixing route, the epoxy resin, curing agent, and accelerator are first dissolved in acetone, and the graphite and nanotubes are then blended into this dilute solution before the solvent is removed by rotary evaporation and oven drying. The dilution gives each nanotube room to unbundle and allows the resin to coat the graphite particles uniformly. When the coated powders were hot-pressed into plates at 160 to 200 degrees Celsius, the difference was visible to the naked eye. Dry-mixed samples collapsed at low molding pressures of 20 megapascals, while wet-mixed counterparts held their shape, and at 60 megapascals both routes produced dense, mirror-smooth surfaces. Quantitatively, wet mixing delivered 15 to 25 percent higher flexural strength at every pressure tested, a benefit the authors attribute to fewer microvoids and better mechanical continuity.

Scanning electron microscopy revealed exactly why the solvent makes such a difference. In cross-sections of dry-mixed composites, nanotubes sit in obvious clumps on the graphite surfaces, blocking resin infiltration and weakening the bonds between particles. In wet-mixed samples, the nanotubes are spread evenly along the interfaces between graphite grains, where they can do their intended job: bridging gaps, transferring stress, and stitching together a continuous conductive network. The images also showed that adding 7 weight percent nanotubes fills the voids between graphite particles with conductive bridges, explaining the sharp gains in conductivity and strength relative to nanotube-free plates.

Graphite particle size turned out to be the other decisive variable. The researchers compared a fine synthetic graphite with particles around 6 to 7 micrometers against a coarser grade reaching 48 to 65 micrometers. Counterintuitively, the coarse graphite won on nearly every electrical metric. Larger particles provide broader contact areas with their neighbors, so electrons hop more easily from grain to grain, and the wider interparticle spaces give well-dispersed nanotubes room to span the gaps. Composites built on large graphite particles consistently showed higher density, higher conductivity, and lower interfacial contact resistance than their fine-particle counterparts, although the finer particles offered a slight hardness advantage at low nanotube loadings, where the sheer number of particle contact points matters more.

The optimum formulation, labeled W-LG-CNT7 for wet mixing, large graphite, and 7 weight percent nanotubes, proved to be a genuine sweet spot. Its electrical conductivity reached 151.6 siemens per centimeter, comfortably beating the Department of Energy target of 100. Its flexural strength hit 56.7 megapascals, well above the 40 megapascal requirement, and its interfacial contact resistance measured just 4.6 milliohm square centimeters, far below the 10 milliohm square centimeter benchmark and better than the 8 to 12 milliohm square centimeters typical of commercial SIGRACELL graphite plates. Push the nanotube content to 9 weight percent, however, and everything deteriorates. The excess nanotubes re-agglomerate, creating resistive dead zones and microvoids that undermine both electron transport and mechanical bonding. More is not better; well-dispersed and just enough is better.

Durability testing delivered perhaps the most commercially significant result. Bipolar plates in a working fuel cell are exposed to phosphoric acid that leaches from the membrane electrode assembly, so the team immersed their samples in 95 percent phosphoric acid for 1,000 hours to simulate years of operation. Every composite lost less than 15 percent of its flexural strength, but the optimized wet-mixed plate lost only 8.2 percent, its dense, well-coated microstructure effectively blocking acid and moisture penetration. That kind of chemical resilience, combined with the corrosion immunity that composites enjoy over metals, addresses one of the last major obstacles to long-life fuel cell stacks.

The ultimate proof came in a real single-cell fuel cell test. The researchers assembled a 25 square centimeter cell with a Nafion membrane, platinum catalysts, and carbon paper gas diffusion layers, operated it at 70 degrees Celsius on humidified hydrogen and oxygen, and swept the current to map its performance. The optimized bipolar plate delivered a stable open-circuit voltage of about 0.97 volts and a peak power density of 0.86 watts per square centimeter, more than 30 percent higher than an identical cell built with a nanotube-free plate. The gains were most pronounced at high current densities between 1.0 and 1.6 amperes per square centimeter, precisely the region where internal resistance losses normally throttle a stack, confirming that the nanotube network was genuinely improving electron transfer under load.

What makes this study resonate beyond the laboratory is its message about manufacturing. No exotic materials were invented; the breakthrough lies in a solvent, a particle size, and a dosage. As hydrogen economies scale up worldwide, such process-level refinements, cheap to implement in existing compression-molding lines, may matter as much as catalyst breakthroughs. A bipolar plate that is stronger, more conductive, more acid-resistant, and demonstrably capable in a working cell brings the 2030 commercialization horizon measurably closer, one well-stirred batch of acetone and nanotubes at a time.

Subject of Research: Optimization of carbon nanotube loading, graphite particle size, and wet mixing for carbon composite bipolar plates in proton exchange membrane fuel cells

Article Title: Enhanced mechanical and electrical properties of carbon composite bipolar plates in PEMFCs via wet mixing and CNT optimization

Article References: Lee, S. H., Lee, S. M., Park, S., Jung, D.-H., Song, W.-J., & Lee, Y.-S. (2025). Enhanced mechanical and electrical properties of carbon composite bipolar plates in PEMFCs via wet mixing and CNT optimization. Advances in Industrial and Engineering Chemistry, 1(1), Article 27. https://doi.org/10.1007/s44405-025-00020-1

Image Credits: AI Generated

DOI: 10.1007/s44405-025-00020-1

Keywords: bipolar plates, PEMFC, carbon nanotubes, wet mixing, graphite composites, epoxy resin, electrical conductivity, flexural strength, interfacial contact resistance, fuel cell durability, hydrogen energy, compression molding