TECHNOLOGY EXPLAINER
AEM Water Electrolysis
— a “third method” without precious metals, whose life is set by membrane chemistry
AEM (anion exchange membrane) electrolysis uses a solid polymer membrane that carries hydroxide ions (OH⁻). It may combine a thin membrane like PEM’s with nickel-based electrodes like those of alkaline electrolysis, and is hoped to avoid both precious metals and titanium. Whether the membrane and the cationic groups inside it survive alkali and oxidation, however, decides its life. This article separates research, kW-class commercial products and MW-class products (manufacturer’s statements).
- What AEM electrolysis is (the short version)
- Reactions and cell structure
- A materials engineer’s view (1): membrane chemistry — backbone and cationic group
- A materials engineer’s view (2): “alkali-stable” is not enough — the pure-water anode
- Electrodes and parts — nickel-based, and a fluoropolymer binder
- Stages of maturity — reading research, kW-class and MW-class apart
- Our calculation: checking the manufacturer’s figures for consistency
- Reading the numbers, evaluation and standards
- Challenges and limits / Glossary / Primary sources / Claim-to-source audit
Sourced = a value or statement in published material (link given)
Our calculation = a figure this article derived, with the assumptions spelled out
Not yet confirmed = a plan, target or proposal with no confirmed result or adoption
Structural readings and materials-design interpretations are marked separately as Commentary.
Performance figures are given together with the level they apply to (single cell, stack or system; feed liquid; current density) and their status (research, commercial, manufacturer’s statement).
Hydrogen is a flammable gas, and its handling is governed by laws and standards. This article explains materials, principles, evaluation and policy; it is not advice on handling or operating hydrogen equipment. It does not rank any product, company or country.
1. What AEM electrolysis is (the short version)
AEM stands for anion exchange membrane. It is a method in which a solid polymer membrane bearing cationic groups carries OH⁻: the mirror image of PEM, which carries protons.
- The hope: IRENA says AEM’s potential lies in combining the milder environment of alkaline electrolysers with the simplicity and efficiency of PEM, allowing non-noble catalysts, titanium-free components and, as with PEM, operation under differential pressureSourced
- The reality: the same IRENA report (2020) says the AEM membrane has chemical and mechanical stability problems, lifetimes are unstable and performance is not yet as good as expected, because of low membrane conductivity, poor electrode architecture and slow catalyst kineticsSourced
- The stage: IRENA described AEM as the latest technology, with only a few companies commercialising it and limited deployment. As of October 2026, kW-class and MW-class products exist (both as the manufacturer states them), but this article has not confirmed any third-party verificationSourced
The life of AEM electrolysis tends to be set less by catalysts and electrodes than by the chemistry of the membrane, its cationic groups and its backbone, and being “alkali-stable” does not by itself mean lasting long. When the cell is fed pure water, the anode sits at a near-neutral pH and a high potential, and it is electrochemical oxidation, not alkaline attack, that damages the cationic groups. In one experiment the cation regarded as alkali-stable failed first (our commentary; Chapter 4).
PEM electrolysis is covered in our explainer on PEM water electrolysis, alkaline electrolysis in our explainer on alkaline water electrolysis and solid oxide electrolysis cells in our explainer on solid oxide electrolysis cells. Our explainer on fluoropolymers and PFAS is also relevant.
2. Reactions and cell structure
At the cathode water becomes hydrogen and OH⁻; OH⁻ passes through the membrane to the anode, where it becomes oxygen and water (the reactions are textbook material, our commentary). In IRENA’s 2020 table, AEM operates at 40–60 °C and below 35 bar, with an electrolyte described as “DVB polymer support with KOH or NaHCO₃ 1 mol/L”Sourced (some cells run on pure water alone).
| Part | Example material (IRENA 2020) | Difference from PEM and alkaline |
|---|---|---|
| Electrolyte (membrane) | Anion exchange membrane (a polymer bearing cationic groups) | PEM uses a PFSA membrane; alkaline uses a porous diaphragm |
| Anode catalyst | High-surface-area nickel, or nickel-iron-cobalt alloys | PEM uses iridium oxide |
| Cathode catalyst | High-surface-area nickel | PEM uses platinum |
| Porous transport layer | Nickel foam on the anode; nickel foam or carbon cloth on the cathode | PEM uses platinum-coated titanium |
| Bipolar plate | Nickel-coated stainless steel (both sides) | PEM uses platinum- and gold-coated titanium |
| Frames and seals | PTFE, silicone | — |
All Sourced (IRENA [Source 1] Table 1). IRENA notes that for AEM and solid oxide, conditions and components vary significantly between companies and research institutes.
3. A materials engineer’s view (1): membrane chemistry — backbone and cationic group
An AEM consists of a backbone (the polymer skeleton) and cationic groups (such as N⁺) hanging from it. It can fail in two places: Wang et al. write that the overall stability can be limited by both the polymer backbone and the organic cationSourced.
- The backbone: the aryl-ether linkages common in engineering plastics (poly(arylene ether), poly(phenylene oxide) and the like) are not chemically stable once cationic groups are attached, and cause chain scission and loss of mechanical properties, say Wang et al. Ether-free aromatic backbones are therefore being studiedSourced
- The cationic group: quaternary ammonium, imidazolium, guanidinium, phosphonium and many others have been studied, but the paper says it has been a challenge to find a cation with both high alkaline stability (especially near 100 °C) and a small size that allows high ion exchange capacity (IEC)Sourced. It cites Marino and Kreuer for piperidinium being 21 times more stable than the benchmark benzyltrimethylammoniumSourced
The poly(aryl piperidinium) (PAP) of Wang et al. puts piperidinium on a rigid, ether-free backbone. After 2000 h in 1 M KOH at 100 °C, its IEC fell from 2.37 to 2.29 meq/g, about 3%Sourced (a laboratory immersion test of a membrane for hydroxide exchange membrane fuel cells, not the life of an electrolyzer in operation). Hydroxide conductivity at room temperature is reported as 60–80 mS/cmSourced.
IRENA explains that OH⁻ is intrinsically three-fold slower than the proton in PEM, forcing AEM developers either to make membranes thinner or to raise charge density (IEC)Sourced. Wang et al. likewise point to the trade-off between conductivity and dimensional stability: a higher IEC increases water uptake and worsens mechanical properties. AEM membrane design therefore has to meet conductivity, alkali resistance, dimensional stability and mechanical strength at the same time (our commentary).
To obtain a high molecular weight and low water uptake, the PAP of Wang et al. uses a monomer containing a –CF₃ group, 2,2,2-trifluoroacetophenoneSourced. The anode in the experiments of Jin et al. also contains a PTFE binder (15 wt%) in the catalystSourced. AEM uses no PFSA and so seems unrelated to the PFAS debate, but fluorinated monomers, binders and seals can still be present (how the OECD definition applies is covered in our explainer on fluoropolymers and PFAS; this is our reading). The updated EU PFAS restriction proposal suggests a 13.5-year transition after entry into force for “fuel cells and electrolysers”, broadened to all electrolysers (at the proposal stage; see our explainer on PEM water electrolysis)Not yet confirmed.
4. A materials engineer’s view (2): “alkali-stable” is not enough — the pure-water anode
There are two broad ways to run an AEM electrolyzer: fed with KOH solution or fed with pure water only. Jin et al. say KOH-fed AEM electrolysis has been reported in the literature to run stably for over 10,000 h, but that the corrosiveness of KOH and carbonate precipitation complicate the system. Pure-water feed is simpler, but at near-neutral pH and a high anode overpotential it often lasts only one-tenth to one-hundredth as long as KOH feed, they writeSourced.
At a pure-water anode the near-neutral environment weakens alkaline attack. But limited OH⁻ transport raises resistance and catalyst activity falls, so the anode potential rises. Jin et al. explain that this high potential can oxidise the cations directly and generate hydroxyl radicals (HO·) and the likeSourced.
Sorting commercial ionomers by cationic group, DMP and imidazolium types show competitive durability in alkaline solution but degrade badly in pure water, whereas TMA, regarded as relatively less alkali-stable, shows a better durability trend in pure water: a reversalSourced (a compilation of literature values for commercial products). In their experiment the same cell, with only the anode ionomer changed, was run on pure water at 1 A/cm². The TMA cell kept functioning for over 360 h, the DMP cell failed within 150 h, and a commercial DMP ionomer lasted about 100 hSourced (laboratory single cells). Post-mortem analysis showed more quaternary ammonium groups retained in the TMA electrodes.
What this suggests is that a cationic group that passes an “alkaline stability” test (immersion in hot, strong alkali) does not necessarily last at an electrolysis anode. In materials screening, always check whether the real operating conditions (potential, pH, feed liquid) have been reproduced (our commentary; Jin et al. also point out that many degradation assessments rarely replicate the electrochemical conditions of an operating cellSourced).
As for how degradation proceeds, Jin et al. observed lower-order amines and formaldehyde from oxidation, and describe a vicious circle: once the cations break down, charge density falls, the ionomer aggregates, the reaction interface collapses and voltage risesSourced.
5. Electrodes and parts — nickel-based, and a fluoropolymer binder
In IRENA’s summary the AEM electrodes and parts are nickel-based (nickel or nickel-iron-cobalt alloy, nickel foam, nickel-coated stainless steel), with neither precious metals nor titaniumSourced. Avoiding the titanium and precious-metal coatings that are one reason PEM stacks are expensive (see our explainer on PEM water electrolysis) is the heart of the hope placed in AEM.
But the catalyst layer is a mixture of catalyst and anion exchange resin (ionomer), and, as the previous chapter showed, the ionomer breaks down first, which limits durability. The anode in Jin et al. used a laboratory-synthesised nickel-cobalt (NiCo₂) catalyst with 15 wt% PTFE binderSourced. Even without precious metals, the design of the catalyst-layer materials (ionomer and binder) remains: a different kind of difficulty from PEM’s precious metals (our commentary).
6. Stages of maturity — reading research, kW-class and MW-class apart
| Level | What is published | Source |
|---|---|---|
| Research | Single cells on pure water at 1 A/cm² lasted from 100 to over 360 h depending on the cationic group of the anode ionomer (laboratory). A literature summary reports over 10,000 h with KOH feed | Jin et al. (2026) [Source 2] |
| Commercial (kW class) | Enapter “AEM Electrolyzer EL 4”: hydrogen production of 500 NL/h (1.0785 kg/24 h), outlet pressure up to 35 barg, hydrogen purity 99.9% (above 99.999% with an optional dryer), “starting from 2.4 kW”. The company states “over 7,500 units deployed worldwide” | Enapter product page [Source 4] (manufacturer’s stated values; viewed 6 October 2026) |
| Commercial (MW class) | Enapter “AEM Nexus”: from a 500 kW version to a 2.5 MW version. The 2.5 MW version gives 491.6 Nm³/h (44.2 kg/h), a system efficiency of 51.3 kWh/kg, an operating range of 3.5–100%, outlet pressure up to 31 barg and purity up to 99.999%. The company calls it the “world’s first megawatt-class AEM Electrolyzer” | Enapter product page [Source 5] (manufacturer’s stated values) |
| Overall | The company’s home page says “more than 12,000 Enapter AEM Stacks produce green hydrogen worldwide” | Enapter home page [Source 6] (manufacturer’s stated value) |
All Sourced (but company statements). “World’s first” and the number of units are the company’s claims, and this article has not confirmed third-party verification. The operating point of the 51.3 kWh/kg efficiency (load, temperature, voltage basis) is not stated.
IRENA in 2020 gave AEM stacks as 2.5 kW, an electrode area under 300 cm² and a lifetime over 5,000 hSourced. By 2026 companies advertise MW-class products. But this article found no source, within its scope, that confirms the lifetime or degradation rate of these products from primary information.
7. Our calculation: checking the manufacturer’s figures for consistency
We calculate from the figures on the product pagesOur calculation.
- Unit consistency: 491.6 Nm³/h × 0.08988 kg/Nm³ = 44.2 kg/h (matches the page). For the EL 4, 500 NL/h × 0.08988 g/NL = 44.9 g/h = 1.079 kg/24 h (matches the page’s 1.0785 kg/24 h)
- AEM Nexus 2500: 44.2 kg/h × 51.3 kWh/kg = about 2,267 kW, about 91% of the nominal 2.5 MW. Conversely, 44.2 kg/h at 2.5 MW would be about 56.6 kWh/kg. The page does not let us read whether 51.3 kWh/kg applies at rated power
- EL 4: if “2.4 kW” is the power at 500 NL/h, then 2.4 kW ÷ 0.04494 kg/h = about 53.4 kWh/kg
Assumptions and limits: hydrogen density of 0.08988 kg/Nm³ (0 °C, 1 atm) is our assumption. The pages do not say at which operating point “2.4 kW” applies, or whether it covers the whole system. These values therefore cannot be set beside DOE targets (stack) or Nel’s (PEM, 100% capacity) values to rank them.
8. Reading the numbers, evaluation and standards
| Item | IRENA summary (December 2020) | Enapter AEM Nexus (manufacturer’s stated values) |
|---|---|---|
| What the number is | A summary of the 2020 situation. Differences between companies are large for AEM | The statement on an MW-class product page |
| Current density and voltage | Nominal current density 0.2–2 A/cm², voltage range 1.4–2.0 V | Not stated |
| Electricity use | Stack 51.5–66, system 57–69 kWh/kg | System efficiency 51.3 kWh/kg (operating point not stated) |
| Operating conditions | 40–60 °C, pressure below 35 bar, load range 5–100%, hydrogen purity 99.9–99.999% | Outlet pressure up to 31 barg, operating range 3.5–100% (2.5 MW version), purity up to 99.999% |
| Life and stack size | Stack lifetime over 5,000 h, stack unit 2.5 kW, electrode area under 300 cm² | Not stated |
| Capital cost | “Unknown” (the target is under 100 and 200 $/kW for stack and system) | Not stated |
All Sourced (IRENA [Source 1] Table 6, Enapter [Source 5]). A 2020 summary and a manufacturer’s statement are set side by side; this is not a ranking or a measure of progress. Within the scope of this article we could not find a DOE page of technical targets for AEM electrolysis.
- The international safety standard: ISO 22734:2019 has been withdrawn and replaced by ISO 22734-1:2025 (Part 1: Safety), published in July 2025Sourced (ISO, checked 5 October 2026). The old edition’s scope explicitly included AEM (solid polymer with basic functional groups). The text is paid, so we limit ourselves to the title and scope
- Test methods for performance and durability: the JRC report (EUR 30752 EN, 2021) sets out harmonised operating conditions and testing protocols for low-temperature water electrolysis: PEM, AEM and alkalineSourced
9. Challenges and limits
(1) Life and degradation rate of commercial systems could not be confirmed
Enapter’s product pages give the number of units and an efficiency but no stack lifetime or degradation rate. Within the scope of this article we could not confirm the lifetime of a commercial AEM electrolyzer from primary information, so no figure is given.
(2) The research numbers are not on the same footing
The 360 h of Jin et al. (pure water, 1 A/cm²) and the over 10,000 h with KOH feed (a literature summary) come from tests that differ in current density, temperature, catalyst and membrane, and are not a like-for-like comparison. The membrane of Wang et al. is for fuel cells and does not show electrolyzer life.
(3) Other companies’ products and plans are not covered
For a commercial example this article used Enapter’s product pages. Other companies’ AEM products (including membrane suppliers such as Ionomr, Versogen and Dioxide Materials), projects and any DOE AEM targets were not checked. No ranking of companies or products is made.
(4) What is not covered
Cost comparison with other methods, carbonate precipitation (with KOH feed), catalyst synthesis, membrane manufacturing and the details of polymer molecular design, and the handling of hydrogen and electrolyte are not covered.
- AEM electrolysis is a method that may avoid precious metals and titanium, built from an anion exchange membrane that carries OH⁻ and nickel-based electrodesSourced
- Life tends to be set by membrane chemistry: backbone and cationic group. Aryl-ether links are fragile, and ether-free backbones are being studiedSourced
- At a pure-water anode, “alkali-stable” is not enough. In the same laboratory cell, TMA lasted over 360 h and DMP within 150 h (1 A/cm²)Sourced
- Read the stages apart. Research (laboratory), kW-class commercial (EL 4) and MW-class (AEM Nexus) carry numbers of different meaning; the commercial numbers are the manufacturer’s statementsSourced
- The life and degradation rate of commercial systems could not be confirmed.
10. Glossary
- AEM
- Anion exchange membrane: a solid polymer membrane bearing cationic groups that carries OH⁻.
- Ionomer
- An ion-conducting polymer of the same kind as the membrane, used inside the catalyst layer.
- Cationic group
- N⁺ and the like attached to the polymer, attracting and carrying OH⁻.
- TMA / DMP
- Trimethylammonium / N,N-dimethylpiperidinium: types of cationic group.
- IEC
- Ion exchange capacity (meq/g): the amount of ion-exchange groups per gram of polymer.
- Aryl-ether link
- A bond joining aromatic rings through oxygen. It tends to break in strong alkali.
- PAP
- Poly(aryl piperidinium): a polymer with piperidinium groups on an ether-free backbone.
- Pure-water feed
- Operation on pure water alone, with no supporting electrolyte such as KOH.
- HO·
- The hydroxyl radical: strongly oxidising and damaging to polymers.
- NL / Nm³
- Litres / cubic metres converted to standard conditions (0 °C, 1 atm).
- Manufacturer’s stated value
- A value in the manufacturer’s own material, separate from third-party verification.
- mS/cm
- A unit of conductivity. The larger, the more readily ions pass.
11. Primary sources
- IRENA “Green hydrogen cost reduction: Scaling up electrolysers to meet the 1.5°C climate goal”, December 2020 (PDF) — irena.org
- Jin Y et al. “Oxidative stability as a guiding principle for durable ionomer cations in pure-water electrolysis”, Nature Communications 17, 6543 (2026, peer-reviewed) — doi.org/10.1038/s41467-026-73209-7
- Wang J et al. “Poly(aryl piperidinium) membranes and ionomers for hydroxide exchange membrane fuel cells”, Nature Energy 4, 392–398 (2019, peer-reviewed) — doi.org/10.1038/s41560-019-0372-8
- Enapter “AEM Electrolyzer EL 4” product page (viewed 6 October 2026) — enapter.com
- Enapter “AEM Nexus” product page (viewed 6 October 2026) — enapter.com
- Enapter home page (viewed 6 October 2026) — enapter.com
- ISO “ISO 22734-1:2025 Hydrogen generators using water electrolysis — Part 1: Safety” — iso.org
- ISO “ISO 22734:2019 Hydrogen generators using water electrolysis — Industrial, commercial, and residential applications” (withdrawn) — iso.org
- European Commission JRC Tsotridis G, Pilenga A, “EU harmonised protocols for testing of low temperature water electrolysers”, EUR 30752 EN (2021) — publications.jrc.ec.europa.eu
- KemI (Swedish Chemicals Agency) and the other four authorities “What you need to know about the updated PFAS restriction dossier”, updated 23 October 2025 (PDF) — kemi.se
12. Claim-to-source audit
| Claim in the text | Basis | Label |
|---|---|---|
| That AEM’s potential lies in combining the milder environment of alkaline electrolysers with PEM’s simplicity and efficiency, allowing non-noble catalysts, titanium-free components and differential-pressure operation. That the AEM membrane has chemical and mechanical stability problems and performance is below expectation (low conductivity, electrode architecture, slow catalyst kinetics). That few companies commercialise it and deployment is limited. That OH⁻ is intrinsically three-fold slower than the proton in PEM, forcing thinner membranes or higher charge density. The table values (operating temperature 40–60 °C, pressure below 35 bar, electrolyte “DVB polymer support with KOH or NaHCO₃ 1 mol/L”, nickel-based catalysts, porous layers and bipolar plates, frames and seals of PTFE and silicone) and the note that AEM and solid oxide entries vary widely between companies. The 2020 summary values (nominal current density 0.2–2 A/cm², voltage 1.4–2.0 V, load range 5–100%, purity 99.9–99.999%, stack 51.5–66 and system 57–69 kWh/kg, lifetime over 5,000 h, stack unit 2.5 kW, electrode area under 300 cm², capital cost “unknown”). That OH⁻ attack on the backbone collapses the membrane | IRENA (December 2020)[Source 1] https://www.irena.org/-/media/Files/IRENA/Agency/Publication/2020/Dec/IRENA_Green_hydrogen_cost_2020.pdf | Sourced |
| That KOH-fed AEM electrolysis has been reported in the literature to run stably for over 10,000 h; that KOH’s corrosiveness and carbonate precipitation complicate the system; that pure-water feed often lasts one-tenth to one-hundredth as long. The classification of commercial ionomers by cationic group (TMA, DMP, imidazolium) and the reversal of durability trends between alkaline solution and pure water. That electrochemical oxidation (HO·, HO₂· and similar radicals) at the anode is the dominant degradation pathway. The experiment: TMA over 360 h at 1 A/cm² on pure water, laboratory-made DMP within 150 h, a commercial DMP ionomer (PiperION®) about 100 h, and more quaternary ammonium retained in TMA electrodes afterwards. The oxidation products (lower-order amines, formaldehyde) and the vicious circle of cation loss, lower charge density, aggregation and rising voltage. That the anode catalyst was NiCo₂ with 15 wt% PTFE binder. The remark that many degradation assessments do not reproduce operating conditions | Jin et al. (2026, Nature Communications)[Source 2] https://doi.org/10.1038/s41467-026-73209-7 | Sourced |
| That stability is limited by both backbone and cation; that aryl-ether links cause chain scission when cationic groups are attached and ether-free backbones are being studied; that piperidinium is 21 times as stable as benzyltrimethylammonium (citing Marino and Kreuer); that a PAP membrane lost about 3% of IEC (2.37 to 2.29 meq/g) in 1 M KOH at 100 °C over 2000 h and has a room-temperature hydroxide conductivity of 60–80 mS/cm (a fuel-cell membrane); the trade-off between conductivity and IEC, water uptake and mechanical properties; the use of a 2,2,2-trifluoroacetophenone monomer | Wang et al. (2019, Nature Energy)[Source 3] https://doi.org/10.1038/s41560-019-0372-8 | Sourced |
| EL 4: 500 NL/h (1.0785 kg/24 h), outlet pressure up to 35 barg, purity 99.9% (above 99.999% with a dryer), “starting from 2.4 kW”, and over 7,500 units deployed according to the company | Enapter product page (manufacturer’s stated value, viewed 6 October 2026)[Source 4] https://enapter.com/en/product/aem-electrolyzer-el-4/ | Sourced |
| AEM Nexus: 500 kW to 2.5 MW versions; the 2.5 MW version at 491.6 Nm³/h (44.2 kg/h), system efficiency 51.3 kWh/kg, operating range 3.5–100%, up to 31 barg, purity up to 99.999%, and the company’s description “world’s first megawatt-class AEM Electrolyzer” | Enapter product page (manufacturer’s stated value, viewed 6 October 2026)[Source 5] https://enapter.com/en/product/aem-nexus/ | Sourced |
| That the company’s home page says more than 12,000 AEM stacks produce hydrogen worldwide | Enapter home page (manufacturer’s stated value, viewed 6 October 2026)[Source 6] https://enapter.com/en/ | Sourced |
| That ISO 22734:2019 was withdrawn and replaced by ISO 22734-1:2025 (Part 1: Safety, July 2025); that the old edition’s scope explicitly included AEM (solid polymer with basic functional groups) | ISO pages (checked 5 October 2026)[Source 7] https://www.iso.org/standard/82766.html [Source 8] https://www.iso.org/standard/69212.html | Sourced |
| That the JRC published a report (EUR 30752 EN, 2021) on harmonised operating conditions and testing protocols for low-temperature water electrolysis (PEM, AEM, alkaline) | JRC Publications Repository[Source 9] https://publications.jrc.ec.europa.eu/repository/handle/JRC122565 | Sourced |
| That the updated background document of the EU PFAS restriction proposal suggests a 13.5-year transition after entry into force for “fuel cells and electrolysers” (6.k), with the scope broadened to all fuel cells and electrolysers | Overview by KemI and the other four authorities (updated 23 October 2025). The proposers’ draft; not adopted or applied[Source 10] https://www.kemi.se/download/18.64e3bd9619a0f45982cfff/1761213262828/What%20you%20need%20to%20know%20about%20the%20updated%20PFAS%20restriction%20dossier-Updated-2025-10-23.pdf | Not yet confirmed |
| 491.6 Nm³/h × 0.08988 kg/Nm³ = 44.2 kg/h; 500 NL/h × 0.08988 g/NL = 44.9 g/h = 1.079 kg/24 h. 44.2 kg/h × 51.3 kWh/kg = about 2,267 kW (about 91% of the nominal 2.5 MW), and 2.5 MW ÷ 44.2 kg/h = about 56.6 kWh/kg. 2.4 kW ÷ 0.04494 kg/h = about 53.4 kWh/kg. 360 ÷ 150 = 2.4 times and 360 ÷ 100 = 3.6 times (lower bounds) | Our calculation. The hydrogen density of 0.08988 kg/Nm³ at 0 °C and 1 atm is our assumption; the operating points of 51.3 kWh/kg and 2.4 kW are not stated on the product pages | Our calculation |
| The life and degradation rate of commercial AEM electrolyzers, other companies’ AEM products and plans, and DOE technical targets for AEM | Could not be confirmed from primary sources within the scope of this article, so none is stated | Commentary |
| The reactions and the movement of OH⁻. Thinking of backbone and cationic group separately; the reading that passing an alkaline-stability test does not guarantee lasting at the anode; the reading that AEM materials are not necessarily fluorine-free (a –CF₃ monomer and a PTFE binder); the framing that catalyst-layer material design remains | Our summary and commentary based on published content. Not views expressed by any of the institutions | Commentary |
| That Fig. 2, Fig. 3, Fig. 5 and Fig. 6 are explanatory drawings rather than real design or measurement diagrams, and that Fig. 1 and Fig. 4 are AI-generated images | Our note | Commentary |
Last updated 6 October 2026. Sources are limited to primary material (official documents of international bodies, standards bodies and manufacturers, and peer-reviewed papers). Because the article includes structural readings and materials-design interpretations, those are marked as Commentary and kept separate from sourced fact. The life and degradation rate of commercial AEM electrolyzers, other companies’ products and plans, and DOE’s AEM targets could not be confirmed and are not stated. Commercial figures are the manufacturer’s statements and have not been independently verified; research figures (laboratory single cells) are not the life of a commercial system. Fig. 2, Fig. 3, Fig. 5 and Fig. 6 are vector drawings and Fig. 1 and Fig. 4 are AI-generated images; none of them shows a real cross-section, equipment or product.