The Green Hydrogen Organisation (GH2) sets the headline emissions threshold at ≤1 kg CO2e per kg H2 averaged over a 12-month period, a number considerably tighter than competing "clean" or "low-carbon" hydrogen definitions that accommodate fossil pathways [S2].
Green hydrogen is hydrogen produced by electrolysis of water using 100% or near-100% renewable electricity (hydropower, wind, solar, geothermal, tidal, wave), with limited flexibility for backup power so long as the CO2e cap holds [S2]. Standardised impurity ceilings for product H2 now list 14 contaminants, including H2O, total hydrocarbon (as CH4), O2, He, N2, Ar, CO2, CO, total sulfide (as H2S), HCHO, HCOOH, NH3, total halide (as halide ions), and maximum particulates [S10].
GH2 Green Hydrogen Standard: Definition, Threshold, and Boundary
GH2 launched the standard in Barcelona in May 2022 as the first global label allowing projects to mark output as "GH2 Green Hydrogen" and trade certificates of origin for H2 and derivatives such as green ammonia [S2]. The standard currently uses a "well to gate" system boundary and is expected to extend to a full life-cycle boundary in later revisions [S1].
GH2's definition requires compliance with the ≤1 kg CO2e/kg H2 ceiling, the use of renewable electricity at near-100% level, and a quantified reporting scope covering scope 1 emissions (production, water treatment, desalination) and scope 2 emissions (on-site or purchased renewable electricity), built on the IPHE methodology [S2]. Requirement 5E expects operators to additionally calculate and report emissions tied to the storage, conversion, and delivery of H2 and its derivatives, including synthetic methane (e-NG / green methane) when used as a carrier [S1].
Selection Criteria: What Specifiers Should Verify
Specifiers evaluating a green hydrogen project against the GH2 label need to verify three layered criteria: the renewable electricity input, the emissions accounting, and the product purity. The renewable input must be hydropower, wind, solar, geothermal, tidal, wave, or comparable ocean energy, with a narrow tolerance window for backup [S2]. The emissions accounting must use the IPHE methodology, averaging emissions over a rolling 12-month window and reporting scope 1 plus scope 2, with scope 3 reporting encouraged for downstream delivery [S2].
The purity gate is the 14-contaminant rule: H2O, total hydrocarbon (measured as methane), O2, He, N2, Ar, CO2, CO, total sulfide (as H2S), formaldehyde (HCHO), formic acid (HCOOH), ammonia (NH3), total halide (as halide ions), and maximum particulates must each stay below their individual ceiling, with one source noting the limits were inherited from an existing hydrogen quality standard [S10]. Independent third-party verification of compliance, plus stakeholder consultation evidence, are mandatory under the GH2 accreditation flow [S1].
Technology Comparison: Alkaline vs PEM vs AEM at the Plant Gate
Three electrolyser families compete for green hydrogen duty. Alkaline units (ALK) use a 25%–30% potassium hydroxide electrolyte, run at 60°C–85°C, deliver up to 30 bar cell-side pressure as standard, and operate at 1.8–2.2 V per cell, with a 3-phase 415 V AC / 50 Hz supply and a 1–500 Nm³/hr module envelope; product purity from the cited vendor reaches ≥99.8% [S5]. PEM units trade higher current density and dynamic response for higher catalyst cost and stricter water purity; the comparison matters when the upstream renewable is intermittent and the stack must ramp frequently.
Multi-megawatt skid/container architectures are now common: alkaline skids from 250 kW to 5 MW scale, with containerised units from 5–4,000 Nm³/hr and a 1-to-N multi-module layout, stack capacity 0.1–5 MW, and a stated 20 MW compact unit built from 20 × 5 MW modules [S3]. For power-to-X routes such as green ammonia, the high-pressure nitrogen side uses cryogenic distillation, pressure swing adsorption (PSA), or membrane separation, then the N2/H2 mix is compressed to the Haber-Bosch loop pressure before the electrically driven synthesis loop [S7]. Operators planning a new build in 2026 should size the electrolyser on a path from pilot to GW: 30 MW assembly plants and 100 MW stack plants are the current Indian PLI-era scale, and the same project can run a 1 MW alkaline skid with integrated 200 bar compression feeding a galvanisation line today [S3].
Use Cases and Who the Standard Is For
The standard is built for projects that need an internationally recognised label to underwrite investment, offtake, and certificate-of-origin trading. It is not a fit for grey or blue hydrogen producers, since the ≤1 kg CO2e/kg H2 cap is incompatible with unabated steam-methane reforming or with fossil pathways that exceed the threshold by an order of magnitude [S2]. It is also not a fit for facilities that cannot demonstrate additionality, temporal matching, or geographic correlation between the renewable generator and the electrolyser, because the IPHE-based accounting reads these gaps directly into the CO2e number [S2].
Real use cases include industrial decarbonisation (replacing grey hydrogen in ammonia and methanol synthesis), heavy-duty transport fuel for fuel cell electric vehicles in long-haul trucking and buses, renewable energy storage via Power-to-X, and export blending [S4][S5]. A typical Indian microgrid reference couples a 300 kW alkaline electrolyser with a 70 kW PEM fuel cell plus balance-of-plant, while industrial sites run 1 MW alkaline skids with integrated compression to 200 bar for galvanisation duty [S3]. Plant designers note that intermittent renewable power shortens electrolyser life if left unmanaged, and the field-proven fix is buffer storage plus advanced control at the BoP interface, a finding directly relevant to anyone sizing flow meters and pressure transmitters on the hydrogen side [S4].
Limitations, Failure Modes, and Open Variables
Several constraints are still live in 2026. The system boundary is "well to gate", not full life-cycle, and downstream delivery emissions are reported on an expected basis rather than capped; until the boundary expands, an H2 molecule that travels by truck or pipeline accumulates un-capped emissions [S1]. The renewable electricity mix allows backup sources, so the renewable percentage can dip below 100% provided the rolling 12-month CO2e average stays under 1 kg/kg H2, a flexibility that some buyers read as a loophole [S2].
Product purity at 99.8% from a single alkaline pass is adequate for many industrial users but not for all fuel-cell stacks, and PSA polishing is typically required to hit fuel-cell-grade 99.99% H2 [S4][S5]. Process control on intermittent input remains a leading failure mode: rapid ramp of an alkaline stack without a buffer accelerates membrane and electrode degradation, so the BoP must include industrial valves sized for turndown and power quality analyzers to flag harmonic distortion from the rectifier front-end [S4]. The 14-contaminant ceiling is enforced, but each individual limit and test method is inherited from the underlying hydrogen quality standard, not redefined by GH2 [S10].
Trackable Signals for the Next Reporting Window
Watch the GH2 standard revision track: the planned extension from "well to gate" to full life-cycle will reset the boundary used for emissions accounting and could move the ≤1 kg CO2e/kg H2 cap downstream into delivery and conversion [S1]. Watch the 30 MW and 100 MW Indian electrolyser assembly plants reaching commissioning, because PLI-driven capacity will set the price benchmark for alkaline stacks through 2026–2027 [S3].
For related hydrogen-economy context, see the spec-side procurement map for fuel cell stack production capacity planning, which lines up stack and BoP sourcing against the same purity and emissions gates that drive the GH2 label.