When people learn what we make, they usually ask which industry we're in. We make proton exchange membranes, which are thin polymer films most famously used inside fuel cells and electrolyzers, so the natural assumption is that we're a hydrogen company. We're not, exactly, and the reason is interesting enough that it's worth working through.

A proton exchange membrane is a piece of plastic about the thickness of a sheet of paper. Its job is to let protons through and block everything else. That sounds narrow, but it turns out that a surprising number of industrial processes amount to moving protons through a barrier while keeping other things separated. Fuel cells do it. Electrolyzers do it in reverse. Redox flow batteries do it. So do electrochemical compressors, lithium extraction systems, chlor-alkali production, membrane dehumidifiers and humidifiers, submarine oxygen generators, and CO₂ reduction reactors. Some of these are industries today. Some are research projects. All of them share the same bottleneck.

That's the structure of a platform technology. It's a piece of capability sitting upstream of a bunch of applications that don't share customers and don't share a thesis but happen to share a chemistry. Silicon is the canonical example, and the parallel to silicon turns out to be unusually exact. So is the transistor. In a smaller way, so is lithium-ion. None of these started as platforms. They started as parts for one specific thing — silicon for solar cells, the transistor for hearing aids, lithium for camcorder batteries — and the platform nature only became obvious once people started using them for other stuff.

The interesting question is why nobody seems to think of the membrane this way. The answer, I think, is that the membrane hasn't moved in thirty years. Nafion was invented at DuPont in the late 1960s, and the last meaningful improvement to the basic structure was in 1994, when W.L.Gore added a porous mesh to make it mechanically more robust. When a substrate doesn't change, nobody competes on it, and when nobody competes on it, nobody talks about it as a substrate. They talk about it as a part. Silicon hasn't really moved structurally for fifty years but it's good enough, and every chip company in the world competes around it. The membrane hasn't moved either but unlike silicon, it's a constraint.

The thing that makes us not a hydrogen company is that first and foremost, we're competing on the membrane. Hydrogen happens to be the application wherewe have the most paying customers right now, and probably will for the next several years, but it's not where the value of a better membrane stops. If you can get the polymer chemistry right, you can pull iridium loading down by 80%, which matters in hydrogen production but matters even more in any other use case where the catalyst is the limiting material cost. You can push durability and current density up, which matters for the size of a fuel cell, the size of an electrolyzer, and the size of a flow battery cell — all very different products with very different buyers, all gated by the same polymer.

The mistake I see people make, including some investors, is to look at our customer list, notice that most of them are doing something with hydrogen, and conclude that we're a hydrogen company. By the same logic, Intel was a calculator company in 1975. Their biggest customer at the time was a Japanese calculator manufacturer named Busicom. If you'd looked at their revenue, you would have said they were in the calculator business. They weren't. They were in the transistor business, and calculators were just the first thing transistors turned out to be useful for.

I'm not saying we're Intel. The point is just that the framing matters. If you treat a company as a hydrogen company, the question is whether hydrogen is going to work. If you treat a company as a platform company, the question is whether the platform is general — whether the same chemistry that makes a better fuel cell makes a better electrolyzer makes a better flow battery makes a better separator across electrochemistry more broadly. It does. We've shipped material into all three already, and we're getting organic inbounds for the others.

The reason this distinction is worth being precise about is that platform companies and application companies have different shapes. An application company lives or dies on a single market. A platform company has the option to walk across markets as they emerge, which is both a strength and a risk — a strength because you're not betting on any one thing, a risk because you have to resist the temptation to chase every new application before the first ones are profitable.

So far we've been disciplined about that. The membrane stays the same. The manufacturing stays the same. The applications change. The platform compounds.

That, more than anything else, is what's interesting about this category. Not the hydrogen narrative, which goes in and out of fashion, but the fact that there's a substrate underneath the hydrogen narrative — and underneath several other narratives — that almost nobody is paying attention to, and whose chemistry has barely moved in a generation.

It's worth paying attention to now.

‍