FRONTIER SIGNAL | DEEP DIVE | ENERGY & INVESTING | 6 MIN READ | MONDAY, 26 OCTOBER 2026
IN THIS DEEP DIVE
What the Groningen team reports: about double the hydrogen, 500+ hours, over 95% of current into hydrogen
How a ceramic cell that passes protons works, in plain English
The catch: nobody has said "double" compared with what
Who could gain, and what to watch next
Making hydrogen from water takes a lot of electricity. One way to cut the bill is to feed the system steam and let waste heat do part of the work. The catch is finding materials that survive being hot, wet and electrically busy.
A team at the University of Groningen says it has found a combination that does. Its protonic ceramic cell runs between 400 and 650 degrees Celsius. A ceramic membrane passes only protons, so the hydrogen comes out separate from the steam. A self-assembled layer of praseodymium oxide nanoparticles on the electrode, followed by an acid wash, helps.
The paper is in Energy & Environmental Science. We could not open it. Everything here comes from one trade-press report of the paper, so treat the figures as reported, not confirmed.
400 TO 650 °C · 500+ HOURS STABLE · OVER 95% OF CURRENT TO HYDROGEN
Reported by Interesting Engineering on 23 October 2026 from the Groningen paper. The article gives no absolute baseline for the "about double" production rate.
That is a lab result. It is not a plant, a stack or a product. The commercial question is whether this cell chemistry ever reaches the scale where it competes with electrolysers that already ship.
Key takeaways
A different kind of electrolyser. Protonic ceramic cells move protons, not oxygen ions, through the membrane. That keeps the hydrogen separate and avoids a steam-removal step.
Reported performance. More than 500 hours of stable operation, over 95% of the electric current converted to hydrogen, and about double the hydrogen output of earlier designs.
The electrode is the trick. A perovskite oxide matrix grows tiny praseodymium oxide nanoparticles when heated in air, and an acid wash then multiplies the active sites. Lead researcher Vassilis Kyriakou describes "a previously overlooked synergy between membrane and electrode materials."
No extra power. The coverage says the higher output came without more electricity. That is the claim to test.
Key facts are missing. The coverage gives no baseline for "double," no voltage or current density, no cell size and no degradation rate.
Timeline is open. No company, funder or commercial plan is named in what we read.
Here is what happened
The paper. Chemical engineers at the University of Groningen published in Energy & Environmental Science (DOI 10.1039/D6EE03199D). Vassilis Kyriakou, at the university's ENTEG institute, is the lead named in coverage.
The coverage. Interesting Engineering reported it on 23 October 2026 under the headline "Ceramic cell delivers 500 hours of doubled clean hydrogen yield without extra power." A TechXplore headline, "Doubling hydrogen production through a synergy of materials," also appeared, but we could not open it. We found no Groningen release and no independent expert reaction.
The build. Heating in air moves ions across the boundary between electrode and membrane, which makes the praseodymium oxide nanoparticles form on the surface. The report says an acid wash then multiplied the number of active catalyst sites, without giving figures.
How it works
This section is general background on the technology, not details from the paper.
Conventional high-temperature cells. In a standard solid oxide electrolyser, oxygen ions carry the charge. Hydrogen comes out mixed with leftover steam and has to be dried.
Protonic ceramic cells. Protons pass through the ceramic instead. Hydrogen forms on the dry side, separate from the steam. Because the membrane passes protons, the hydrogen stays apart from the rest.
Why heat helps. Steam as feedstock and waste heat as an input mean part of the energy arrives as heat, not electricity. That is the standard argument for high-temperature electrolysis.
Why the electrode matters. The cell is only as good as the place where steam turns into hydrogen. More active sites mean more hydrogen from the same current. The report says protons pass through the membrane and recombine into hydrogen on the other side. The exact mechanism is in the paper, which we could not open.
Why this matters
Hydrogen is stuck on cost. Green hydrogen projects have struggled with the price of electricity and equipment. Anything that makes each unit of electricity produce more hydrogen is directly useful, in principle. Clean-energy hardware keeps hitting the same gap between a demonstration and a product: the hillside battery we covered in September reached its design power but held it for about 15 minutes, not the planned four hours.
Intermediate temperatures are a sweet spot. Cooler than conventional solid oxide systems, which typically run hotter, but hot enough to use industrial waste heat. That may be easier on materials. That reasoning is general, not from the paper.
A material result, not a system result. Cell chemistry decides what a stack could do. Stacks, plants and bankable warranties decide whether anyone buys it.
Who captures the value
Bloom Energy (NYSE: BE). The best-known listed solid oxide company, but its current growth is not electrolysis. Third-quarter 2026 revenue was $1,235.8 million, up 138.1%, and full-year guidance was raised to $4.4 billion to $4.6 billion. The release does not mention electrolysers. The growth is about powering AI data centres. Nothing links Bloom to this research.
Ceres Power (LSE: CWR). Licenses solid oxide technology. First-half 2026 revenue was £22.8 million, up 8%, with an adjusted EBITDA loss of £6.8 million, narrowed from £11.3 million. Its electrolysis progress includes about ¥35 billion of Japanese government funding to DENSO for a demonstration with JERA, and a 1 megawatt demonstration at Shell's Bangalore centre. Protonic ceramics would be a rival or an extension, not a replacement.
Nel ASA (Oslo: NEL) and Plug Power (NASDAQ: PLUG). Sell mature alkaline and PEM electrolysers. Nel's third-quarter revenue fell 63% to NOK 112 million while order intake rose 339% to NOK 253 million, with a backlog of NOK 1,314 million. Plug Power cites a funnel of about $8 billion and expects full third-quarter results on 9 November. A 500-hour cell moves none of these numbers. It sets the future bar they must clear.
ITM Power (LSE: ITM). Reported interim revenue of £18 million for the six months to October 2025, the latest figures we could verify, and reiterated full-year guidance of £35 million to £40 million. Treat the figures as dated.
THE PRIVATE-COMPANY WATCHLIST
This is an industry read-through. We did not verify facts for these companies.
Topsoe, Sunfire and Elcogen all develop solid oxide electrolysis. If protonic ceramics prove out, they are the likeliest to license, acquire or race the technology. Watch for patent filings and partner announcements from Groningen.
The honest catch
The baseline is missing. "About double" is relative to the group's earlier designs, per the coverage. Without the metric, voltage and temperature, it cannot be judged. We saw the same gap in the crystal membrane for carbon capture, where the headline speed was real and the cost-deciding numbers were not published.
500 hours is short. Commercial electrolysers are expected to run for tens of thousands of hours. We have no degradation rate.
Scale is unknown. We found no cell size, stack or system information. It is likely a small laboratory cell, but no source we read says so. Scale is where good lab results usually stall: fusion would need 315 million perfect fuel capsules a year, and nobody has built that supply chain either.
No cost data. Praseodymium is a rare-earth element. Materials cost and supply are not addressed in the coverage.
Source limits. The journal page and DOI could not be opened on 25 October, and we found no press release. All study figures come from one secondary article.
The one question every reader should ask
Double what, measured how, and for how long?
A hydrogen cell that doubles output for 500 hours is a promising chemistry result. A cell that holds a published current density for 40,000 hours in a stack is a business. The paper will say which of those this is.
OUR VIEW
This is a credible lab advance with an unproven headline number. The mechanism, protons moving through a ceramic so hydrogen comes out separate, is attractive, and the electrode finding is worth watching.
For investors, it is a long-dated option on a technology family, not a catalyst for any stock. Hydrogen companies are still fighting cost, orders and project cancellations. A better cell chemistry does not fix those this year.
What to watch next
The paper itself. The baseline, voltage, current density and degradation rate.
Longer tests. Thousands of hours, not hundreds.
Scale-up. A larger cell, then a small stack.
A partner. A licence or collaboration with a solid oxide or electrolyser maker.
Cost. Materials and manufacturing estimates.
Quick questions
What did the Groningen team make?
A protonic ceramic electrolysis cell that splits steam into hydrogen at 400 to 650 degrees Celsius, with a praseodymium oxide nanoparticle electrode.
How much better is it?
Reported as about double the hydrogen output of earlier designs, with over 95% of current converted to hydrogen and more than 500 hours of stable operation. No absolute baseline was given in the coverage.
Why does separate hydrogen matter?
In a protonic cell, hydrogen forms on the dry side, apart from the steam, which can remove a drying step.
Is this ready for industry?
No. It is a laboratory result with a short test, no reported stack and no stated commercial partner.
Which stocks does it affect?
None directly. Solid oxide and electrolyser companies such as Ceres Power, Bloom Energy, Nel and Plug Power are the nearest, but nothing links them to this work.
READ MORE: ENERGY SIGNALS
A Membrane That Is 75% Crystal Passes CO2 at Very High Speed · 14 Oct 2026 · A carbon-capture film with fast numbers and missing ones.
A Hillside Battery That Does Not Need a Mountain · 18 Sep 2026 · A grid-storage demo that hit power but not duration.
Fusion Needs 315 Million Perfect Fuel Capsules a Year · 10 Sep 2026 · Why manufacturing, not physics, may limit laser fusion.
China Just Broke Ground on Fusion Nobody Else Will Touch · 30 Sep 2026 · A hydrogen-boron fusion bet, and what has not fused yet.
Sources
Interesting Engineering report, 23 October 2026: study details, reported figures and Kyriakou quotes.
Energy & Environmental Science paper (DOI 10.1039/D6EE03199D): journal reference; page could not be opened.
Bloom Energy Q3 2026 results (SEC filing): revenue and guidance.
Ceres Power H1 2026 summary: revenue, EBITDA and electrolysis progress, as relayed by trade press.
Nel ASA Q3 2026 summary: revenue, orders and backlog, as relayed by trade press.
Plug Power update: funnel and preliminary results, as relayed by trade press.
ITM Power interim results: interim revenue and guidance.
Frontier Signal explains frontier technology in plain English. Study figures come from one trade-press report; the paper was not reviewed. This is general information, not investment advice.