Green Hydrogen’s Supply Chain Risk Lives Upstream of the Electrolyser

The Nel ASA framework agreement with Hydrasun, announced today, is being read across the industry as a European assembly milestone. What most of that coverage will skip past is the materials layer underneath it.

Every megawatt of proton exchange membrane (PEM) electrolyser capacity assembled in Aberdeen, Connecticut, or anywhere else runs on platinum and iridium. More than three-quarters of the world’s supply of both sits under South African soil.

Green hydrogen has drawn over US$130 billion in global investment, according to figures cited around today’s announcement, and the build-out is accelerating. Yet the supply chain risk conversation has stayed fixed on project finance and grid integration. The upstream materials layer, specifically the platinum group metals (PGMs) that make PEM electrolysis work, has drawn far less scrutiny from investors.

That gap is where both the commercial opportunity and the risk live. What this article gives you is a framework for assessing the PGM angle inside the green hydrogen build-out: where the data supports a structural thesis, and where it demands caution.

How Nel ASA’s European expansion reveals the shape of the PEM supply chain

On 10 September 2026, alongside a Hydrogen Council webinar, Norwegian technology firm Nel ASA confirmed from Oslo that it had signed a framework agreement with Hydrasun. The deal establishes European assembly and integration capability for Nel’s modular MC Series PEM electrolyser platform.

The structural detail matters more than the headline. Nel’s PEM stack manufacturing stays put at its Connecticut facility in the United States. Hydrasun takes on assembly and integration of the containerised systems in Aberdeen, Scotland.

According to Tushar Ghuwalewala, Nel’s senior vice president of PEM operations, the rationale is rising demand for standardised, modular hydrogen systems. Reports suggest the Aberdeen upgrade may be supported by the Scottish Government’s Just Transition Fund, potentially creating up to 12 new jobs while securing 11 existing positions, though these specifics are not independently confirmed.

The commercial logic is straightforward: regional assembly cuts delivery times, standardises output, and puts integration closer to European demand. That is why agreements of this shape are multiplying.

Where platinum and iridium fit in the PEM stack

To read the announcement correctly, you need to know what a containerised PEM system actually contains.

• The electrolyser stack: the core where water is split into hydrogen and oxygen. This is where the PGM catalysts concentrate. Platinum works as the cathode catalyst, iridium as the anode catalyst.
• Balance of plant: the pumps, power electronics, and water treatment surrounding the stack.
• Containerisation: the modular housing that makes the unit shippable and deployable.

PGM-catalysed water splitting at the electrolyser stack level is where the physical chemistry constrains any substitution timeline, because the electrochemical stability requirements at the anode under acidic PEM operating conditions are precisely what makes iridium difficult to displace with cheaper alternatives.

Here is the point most coverage misses. Nel manufactures the stack in Connecticut regardless of where the container is assembled. Regional assembly partnerships shift the geography of integration, but they leave catalyst sourcing geography untouched.

So the proliferation of Nel-Hydrasun style agreements does not diversify the upstream materials supply chain. It concentrates demand pressure on the same PGM producers. Read the agreement as a hardware story and you are looking at the wrong layer of the value chain.

South Africa’s grip on global PGM supply

The numbers set the terms of the debate before anyone interprets them.

According to the U.S. Geological Survey (USGS) Mineral Commodity Summaries 2026, South Africa’s PGM reserves stand at roughly 63,000,000 kg against global resources exceeding 100,000,000 kg. That works out to approximately 77-78% of tracked global reserves in one country. These specific figures are reported by USGS and not independently verified here.

On production, USGS lists South African platinum mine output at around 126,000 kg in 2024, with a forecast near 120,000 kg for 2025. The reserve dominance is real, but so is a structural ceiling: domestic mine closures since 2016 have cut capacity even as the reserve position holds.

That combination matters. A demand surge cannot be met quickly if the mines to meet it have already been shut.

South African PGM supply risk is further compounded by the geological reality that the Bushveld Complex, while vast, is not uniformly accessible, and the cost curves of marginal shafts already idled since 2016 make rapid restarts economically difficult even at elevated prices.

For PEM-specific supply chains, iridium is the sharper signal.

The iridium concentration is where the risk sharpens. A 2026 supply-chain topology profile calculates a Herfindahl-Hirschman Index (HHI) of 0.80 for iridium mine supply.

What the HHI tells you The Herfindahl-Hirschman Index measures market concentration on a scale from 0 to 1. Anything above 0.25 is treated as highly concentrated. At 0.80, iridium supply is close to the top of that scale, meaning it is genuinely vulnerable to disruption in a single country.

For an investor weighing PGM exposure through the hydrogen channel, iridium is the near-term variable to watch, not platinum. Its concentration sits at a level that no European assembly facility can engineer around. The only real levers are catalyst substitution or a sustained expansion of South African mine capacity, and neither happens fast.

Structural demand driver or transitional catalyst load?

This is the section where the thesis either holds or gets qualified. The disagreement is not about whether hydrogen demand grows. It is about how much PGM each unit of growth actually requires, and two institutions read it very differently.

Start with the structural case. The World Platinum Investment Council (WPIC) frames hydrogen as a major end-market in the making.

• WPIC structural thesis: hydrogen platinum demand rising from 40,000 oz in 2023 toward roughly 875,000 oz by 2030, feeding average annual platinum deficits near 430,000 oz between 2025 and 2028, and reaching 3.5 million oz by 2040, when WPIC expects hydrogen to be platinum’s largest end-market. Argus Media reported a 123% year-on-year rise in hydrogen-linked platinum demand in 2024.
• IEA and DOE transitional view: current loadings are high but falling, and the explicit policy goal is to engineer iridium out of the anode catalyst layer over time through substitution and recycling.

The tension resolves in the loading data. The International Energy Agency (IEA) reports current PEM catalysts using around 0.3 kg of platinum and 0.7 kg of iridium per MW. Commercial state-of-the-art systems in early 2026 were already reported at roughly 0.2 kg/MW of iridium, and a 2024 UK Foresight Electrolysers study projects a potential floor near 0.07 kg/MW. All these figures are reported by the cited bodies and not independently verified.

The table is the whole argument in one frame. A market that triples in installed capacity but cuts its iridium intensity from 0.7 to 0.07 kg/MW delivers a very different outcome for producers than one that scales without efficiency gains.

Iridium loading reduction through nanotechnology-based catalyst architectures represents the most aggressive pathway to the Foresight study’s 0.07 kg/MW target, with 2026 research reporting efficiency gains that, if commercialised, would substantially alter the demand projections underpinning the WPIC thesis.

Market sizing adds a second axis of uncertainty. PEM electrolyser market estimates for 2025-2026 range from around US$1.26 billion to US$6.1 billion depending on the source, with forecast CAGRs spanning 13.8% to 43.4% through the early 2030s. These are third-party projections and should be treated as illustrative rather than settled.

What this tells you is that the loading trajectory matters more than the headline demand volume. The WPIC and IEA are not just disagreeing about scale. They are disagreeing about the pace of technological change, and that pace is the variable to track.

Where the supply chain risk actually concentrates, and what the industry is doing about it

Strip away the demand debate and one chokepoint remains. South African mine capacity cannot respond quickly to a demand spike, no matter how many European assembly lines come online. The risk is upstream, and it is physical.

Against that constraint, three strategic responses are in play, ranked here by how operationally mature they are today.

1. Scenario planning by producers. The Minerals Council South Africa has modelled combined fuel-cell electric vehicle (FCEV) and electrolyser demand reaching 1.6 million oz of annual platinum demand by 2030. This is planning and advocacy, not committed capacity.
2. Closed-loop recycling. The U.S. Department of Energy (DOE) recommends developing PGM recovery from end-of-life PEM components. The intent is clear, but recycling infrastructure at scale remains an emerging channel rather than a live secondary supply.
3. Catalyst substitution. Research into ruthenium as an iridium alternative is active, flagged by the Minerals Council South Africa, but it sits firmly at the research stage.

The DOE’s position is worth reading as a signal in itself.

DOE recommendation The Department of Energy explicitly recommends substituting iridium-based anode catalysts and building PGM recycling from end-of-life PEM components. In other words, the current catalyst architecture is under deliberate institutional pressure to change.

Here is the caution the data forces. Detailed evidence of individual PGM miners restructuring operations specifically to serve PEM catalyst demand remains sparse. There is producer-backed advocacy, and there is demonstrated operational repositioning, and those are not the same thing.

For you, the practical read is this. The near-term investment case rests on price exposure to existing producers, not on identifying companies that have built a specialised strategic position in this end-market. That position, so far, is largely rhetorical.

For investors seeking a specific producer case study, our full explainer on Sibanye-Stillwater’s strategic positioning examines how one of the sector’s largest PGM operators is navigating the gap between producer-backed advocacy and demonstrated operational repositioning toward hydrogen end-markets.

Positioning in a supply chain story that is still being written

The build-out is structurally real. US$130 billion in global hydrogen investment is flowing toward downstream demand, and PGM demand from PEM systems is growing with it. What remains genuinely uncertain is the magnitude and timing of the uplift, because the loading reduction trajectory cuts directly against the demand curve.

That uncertainty is not a reason to dismiss the thesis. It is the reason to track the right variables. WPIC’s 3.5 million oz forecast for 2040 marks the outer bound of the structural case, while the Foresight target of 0.07 kg/MW marks the outer bound of the efficiency case. The truth will land somewhere between them.

Watch these indicators over the next 12 to 24 months:

• Catalyst loading trends in newly deployed electrolysers, the single most decisive variable.
• South African mine output measured against demand projections.
• PEM recycling infrastructure investment announcements.
• Any regulatory reclassification of iridium or ruthenium.

Investors exposed to South African PGM producers sit at one pole of this thesis; those backing catalyst technology alternatives sit at the other. The balance between them depends on a technology timeline that is genuinely uncertain but actively being written.

Every Nel-Hydrasun style agreement is a leading indicator here. Each expansion of European or North American assembly capacity tightens the window in which upstream PGM supply must respond. Watching that catalyst transition is more actionable than tracking project-level hydrogen headlines.

This article is for informational purposes only and should not be considered financial advice. Investors should conduct their own research and consult with financial professionals before making investment decisions.

Past performance does not guarantee future results. Financial projections are subject to market conditions and various risk factors. These statements are speculative and subject to change based on market developments and company performance.