Two weeks after a green coalition called for the US to adopt hourly matching of renewable power with electrolyser use in the country's forthcoming definition of green hydrogen, analyst Wood Mackenzie has revealed that, according to its calculations, annual matching could not only be a net-zero option but would also result in lower cost H2.

Hydrogen: hype, hope and the hard truths around its role in the energy transition
Will hydrogen be the skeleton key to unlock a carbon-neutral world? Subscribe to the weekly Hydrogen Insight newsletter and get the evidence-based market insight you need for this rapidly evolving global market

This might sound like a trivial bureaucratic question, but the matter is crucial in determining when and how renewable energy can be used to produce green hydrogen — and therefore the cost of production, the resulting greenhouse gas emissions, and which projects will be eligible for tax credits of up to $3/kg.

The definition of green H2, which is to be drawn up by the US Treasury, must ensure that using renewable energy to produce hydrogen does not increase overall emissions from the grid — as the aforementioned coalition fears will happen — while at the same time not unnecessarily increasing costs of green H2 production. After all, the more hours per day an electrolyser is in operation, the cheaper the hydrogen.

One of the critical decisions to be made is the extent to which grid power can be used to run electrolysers — the machines that split water molecules into hydrogen and oxygen — at times when the wind is not blowing and the sun isn’t shining, and there is not enough stored renewable electricity available.

Most developers would like to use grid power at such moments to keep electrolysers in operation, and then compensate at a later date by sending the same amount of wind or solar energy back to the network when supply is higher than demand.

But the fear is that subsidised grid-connected electrolysers could gobble up renewable power that would otherwise be used to decarbonise the electricity supply, leaving grid operators no choice but to call on coal-, oil- and gas-fired generators to replace the green energy.

The coalition of scientists, environmental campaigners and energy companies argues that this could at least double net emissions on a per-kilo basis compared to grey hydrogen made with unabated fossil gas.

This is why the coalition is calling for additionality — that only new renewables projects can produce green hydrogen — with hourly matching, which requires that electrolysis plants prove every hour that they have sourced their electricity from a qualified renewables facility, even if that power has come via the local grid using Renewable Energy Certificates (RECs) to prove the purchase of green electrons.

Annual matching

But according to analysis by Wood Mackenzie, annual matching could not only result in net-zero emissions but also produce far cheaper green hydrogen.

“Hourly matching requirements, depending on their implementation, could result in unfavourable economics for green hydrogen adoption, by limiting operating hours to those when renewable resources are available, ultimately reducing the electrolyser capacity factor,” wrote analysts Melany Vargas, Kara McNutt and Chris Seiple in an article published by Forbes magazine.

“The result is that operators must distribute their costs over a smaller volume of hydrogen production, requiring a higher price to recover their capital for each kilogram of hydrogen sold.”

Hourly matching would result in electrolyser capacity factors of 46-72%, according to WoodMac calculations, resulting in increases of 68-175% in the levelised cost of hydrogen compared to “an annual matching scenario that allows operators to reach a capacity factor of 100%”. The latter figure means that an electrolyser would continuously be operating at maximum capacity around the clock, compared to operation at 46-72% of its maximum capacity over the course of each year with hourly matching.

With annual matching, a 250MW electrolysis project in Arizona would produce green hydrogen for about $2/kg in 2025 and $1.50/kg in 2030, when including the $3/kg tax credit, the analysts write.

In an hourly matching scenario, the cost of hydrogen production for the same project would roughly be doubled, at about $4-5/kg.

“This degree of cost increase could delay the ability to produce green hydrogen at cost parity to lower-cost blue or grey hydrogen, ultimately hindering the economic competitiveness and adoption of both grid-connected and 100% renewable green hydrogen as a low carbon fuel.”

Impact on emissions

However, running an electrolyser at maximum capacity around the clock would require grid electricity 19-35% of the time in the Arizona case, the analysts write, leading to a carbon intensity of 7.9 kilograms of CO2 per kilo of green hydrogen in 2025 and 4.7kgCO2/kgH2 in 2030. Neither of these figures, on their own, would allow the hydrogen to qualify for any tax credits, let alone the top rate of $3/kg (which requires 0.45kg of lifecycle CO2-equivalent emissions per kg of H2).

As the coalition fears, this would lead to an absolute emissions increase in the Arizona grid, and also in a second study area in Texas.

“The analysis found that a slight overbuilding of renewables, or strategic curtailment of hydrogen production during peak thermal hours could be effective tools to minimise these unintended emissions impacts in the 2020s,” said the WoodMac analysts.

However, achieving a net-zero carbon intensity would require “REC offsets”.

In other words, it would have to buy more Renewable Energy Certificates. These RECs are tradeable market-based instruments that represent 1MWh of renewable electricity, which enable companies to buy green energy via the grid without being directly connected to a wind or solar farm.

It is exactly this method that enables consumers to buy renewable energy at home from their electricity supplier, even though all their power comes directly from the local grid.

But while this could result in a net-zero carbon intensity for hydrogen production, the impact on absolute grid emissions is not clear from the article.

And the renewable power generating these RECs would have to come from new wind or solar projects that otherwise would not have been built — ie, the additionality principle — otherwise it too could increase grid emissions by needing to be replaced by fossil fuel generation.

The WoodMac analysts recognise the challenges facing the Treasury in coming up with an optimum renewable hydrogen definition.

“Policy makers and regulators are in the tough position of navigating the trade-off between carbon emissions and green hydrogen economics within the context of rapidly changing US power markets,” they write.

“This early analysis demonstrates that on an economic basis, annual matching could be the catalyst the green hydrogen industry needs to support the early adoption and growth of the nascent low-carbon hydrogen industry.”

However, they also point out that their analysis only focused on Texas and Arizona, which have relatively high levels of renewable resource potential.

“There is more investigation required in these and other markets to fully assess the economic and emissions trade-offs being considered here. It is expected that the outcomes would vary significantly on a regional basis and may also vary as hydrogen production scales well past the addition of a 250 MW electrolyzer in a region.”

The analysts conclude: “Beyond 2030, as the build out of wind, solar and storage generation assets support lower carbon grids throughout the US, and electrolyzer costs come down, hourly matching could become a more appropriate mechanism to support 100% renewable green hydrogen production and power grid decarbonization in tandem.”

The European Commission would seem to agree. In its own definition of renewable hydrogen, contained in a long-awaited Delegated Act published on 13 February, monthly matching of renewables supply with electrolyser usage will be required until 31 December 2029, with only hourly correlation allowed after that.

However, a review on the introduction of the hourly correlation will be carried out in 2028, so it might never happen.