Published on May 27, 2020
The draft “EU green recovery plan” proposes an audacious development of clean hydrogen as a mean of storage and transmission of energy. But what will be the CO2 impact, and what are the alternatives?
Simplistically put, the European electrical system is characterized by 2 modes :
– “Fossil mode” : renewables do not saturate the grid, and the load variation are provided by fossils. Any additional demand will increase the production of the last entrant on the grid. The electricity spot price is equal to the marginal cost of this last entrant (for the sake of the argument hereafter: a gas power plant)
– “Renewable mode”: renewables saturate the grid, fossil production is brought to its technical minimum or shutdown. Spot prices are low (or even negative).
The latter mode, in line with the Paris Agreement, occurs more and more often due to the planned increase of renewable capacity, and lately as a consequence Covid’s economic downturn, but does not exceed as of today a few hundred hours a year. Because of grid bottlenecks (in particular cross-border), the 2 modes can appear simultaneously in Europe.
1) Hydrogen CO2 footprint:
1.a) In “fossil mode”
The conversion factors for “green hydrogen” are as follows:
– 1MWh of hydrogen requires 1,5 MWh electricity,
– 1,5 MWh electricity requires 3 MWh of natural gas.
=> 1 MWh of hydrogen requires 3 MWh of natural gas
(Efficiency assumptions: gas to power 50%, electrolysis 70%, Steam Methane Reforming 70% (“grey hydrogen”), energy content of 1kg of H2 is 40kWh, CO2 emission of natural gas is 0,2tCO2/MWh).
Consequently, one energy unit of electrolysed hydrogen will cost at least 3 energy units of natural gas. As alternative, one energy unit of grey hydrogen will cost 1,5 energy units of natural gas.
Hence electrolyzed hydrogen emits twice as much CO2 as grey hydrogen and cannot qualify as “green” (even 4 times more if marginal production is on coal instead of natural gas!).
No commercial concepts such as Certificates of Origin, exclusive agreement between a renewable supplier and an electrolyser, or contract for difference will improve the physics of the system.
Even a strong prevalence of renewable in the mix leading to a low average CO2 emission per kWh does not change the marginal contribution of the last fossil entrant on the grid.
In any event, it is more climate-efficient to displace fossil power than grey hydrogen.
1.b) In the “Renewable mode”
The marginal cost of hydrogen is close to zero, and the direct emission zero. Hydrogen then physically qualifies as “green”.
So the 1million ton clean hydrogen initiative described in the “EU green recovery plan” will require an extra 60TWh from the grid, which will be as of today predominantly extra fossil power (at 0,4 to 0,8MtCO2/TWh).
2) Cross-border limitations effect on production modes
We observe that excess renewables in north western Europe induces locally low and even negative prices, whereas prices remain positive in central and southern Europe. This illustrate physical and regulatory limitations to trading excess renewables. For argument’s sake, the EU power grid is then operating in “renewable mode” in north western Europe, and a “fossil mode” elsewhere. If such limitations were lifted, Atlantic windfarms could sell their excess on east European markets against the local fossil-set prices (instead of dumping) and the local consumer could benefit from cheaper renewable power.
Episodes of negative pricing are used as argument for more subsidies instead as a call for free and unhindered power trading with coal-dependent neighbours.
This confirms that the cross-border transmission is a cornerstone of the power system decarbonation, indispensable for renewable penetration and price smoothening. Investing in such infrastructure is a state prerogative which should take precedence over other forms of subsidies. It also makes the case that TSO’s should be accountable for avoiding renewable dumping (negative pricing) whilst eastern European neighbors predominantly produce fossil power.
3) How urgent is green hydrogen?
Renewable intermittency can be addressed by following means:
– Develop Demand Side Response (smart grids, consumer behavior, electrolysis…)
– Decarbonize adjustable production (CCS, nuclear, biomass)
– Develop storage (batteries, hydrogen)
– Improve cross border transmission
Without evaluating first two options here, one can observe that the third (hydrogen storage) competes with the fourth (improving cross border transmission), until such time that the EU grid has become a “copper plate”. Allegedly, transmission improvement costs are much higher than hydrogen infrastructure, however operational losses of hydrogen conversion exceed 50%. There is no consensus on the pay-back time for a high investment-high efficiency solution (electrical grid improvement) over a lower investment-lower efficiency solution (adapting existing gas infrastructure to hydrogen).
To guide investments in electrolysis, following scenarios should be considered:
– Will cross border transmission and local distribution improve sufficiently to export and eliminate excess renewables in north western Europe?
– Will carbon accounting distinguish average from marginal CO2 footprint in power consumption?
– Will subsidies cover only the investment costs of infrastructure, or also bridge marginal cost differences with alternative technologies?
– Will policy makers allow vertical integration (windfarms exclusively dedicated to electrolysis) or oblige renewable producers to serve an open market and displace the heaviest polluters (fossil fuel and not grey hydrogen)?
Summing up, we face a contradiction: hydrogen is indispensable in the upcoming decarbonized energy system but is a huge indirect CO2 emitter as long as fossil production is still required. The main question is when and where the “renewable mode” will prevail over the “fossil mode” to make electrolysis a CO2 reducing technology, beyond the current occasional occurrences. If green hydrogen production is rolled out faster than the available excess renewables, additional CO2 emissions will be observed.