What is hydrogen best used for?

In theory, hydrogen is a very flexible energy vector – it can be used for power generation, heating and for transport. However, there may be limits and some say its use needs to be restricted to those parts of the economy where electrification/decarbonize is going to be hard, such as aviation and steel-making. For example, Michael Liebreich’s Hydrogen Ladder highlights those areas where hydrogen would be most useful and those areas where it would be uncompetitive. He has made it clear that he doesn’t see hydrogen cars as viable- we should stick to EVs. Provocatively, he claims that the Oil sector is lobbying for inefficient hydrogen cars ‘because it wants to delay electrification’.

He feels the same about home heating- hydrogen in not a sensible option. There has certainly been much debate over its viability for direct domestic space heating, with electric heat pumps widely seen as much more efficient, at least for some parts of the annual heating cycle. The UK government’s recent Hydrogen Strategy (see my last post) seemed to agree, although that was not immediately apparent from the BEIS press release, which talked about hydrogen replacing natural gas in powering around 3 million UK homes each year. As Jan Rosenow, of the Regulatory Assistance Project, pointed out, the 3 million households figure is misleading- BEIS is actually expecting less than 70,000 homes to use hydrogen for heating by 2030. He said ‘The government's strategy shows that less than 0.2% of all homes are expected to use hydrogen to keep warm in the next decade’.  So hydrogen wasn’t going to be a big player in this sector either. 

Although there are some disagreements, as a useful policy review by Safe Energy indicates, something of a consensus on hydrogen’s limits does seem to have emerged. It was reflected in an open letter that a group of academics sent to the government to warn of the inefficiencies and added expense of using hydrogen in sectors where electric alternatives were cheaper, arguing that ‘hydrogen for heating and road transport is not efficient and does not make economic sense’. 

That may be too forcefully put, in that there may be exceptions, but a recent report from the Potsdam Institute in Germany comes to the same general conclusion. It provides a merit order of energy usages, showing that, in some cases, decarbonisation by the use hydrogen is costly or even impossible. On this basis it suggests that hydrogen is not suited to cars or space heating systems - they are better run directly with (green) electricity. But, as it shows, there are areas offering important potential roles for hydrogen- e.g. steel, cement, aircraft. 

Hydrogen for grid balancing

What about the idea of using surplus renewable power to make and store hydrogen, ready to be converted back to power again (in a gas turbine or fuel cell) when there was a lull in renewable availability? Well, a new US study suggests that hydrogen might not be best used for grid balancing of variable renewable energy (VRE) generation. It says that using H2 for grid-scale energy balancing, via ‘power-to-gas-to-power (P2G2P)’ conversion steps, has low relative energy conversion efficiency. And focusing on that approach ignores the cost-sharing and CO2 emission benefits that would be gained from using the deployed H2 assets directly to decarbonize end-use sectors where electrification is challenging. 

Of course you do not have to use hydrogen exclusively for any one thing-  and the logistics using (effectively free) surplus renewable power differ from using renewables full time to generate hydrogen. However, the study found that, in general, the value of power-to-H2 (P2G) was higher than for P2G2P routes. Specifically, ‘P2G provides grid flexibility to support VRE integration without the round-trip efficiency penalty and additional cost incurred by P2G2P routes. This form of sector coupling leads to: (a) VRE generation increase by 13–56%, and (b) total system cost (and levelized costs of energy) reduction by 7–16% under deep decarbonization scenarios. Both effects increase as H2 demand for other end-uses increases, more than doubling for a 97% decarbonization scenario as H2 demand quadruples’. 

In addition, it found that ‘the grid flexibility enabled by sector coupling makes deployment of carbon capture and storage (CCS) for power generation less cost-effective than its use for low-carbon H2 production’, but also that the economics of green hydrogen production by electrolysis improved when carbon costs were high. 

Some mixed messages then. It’s good to hear that green hydrogen’s cost could be competitive, since there will be carbon emissions from the production of so-called blue hydrogen produced from fossil gas. We don’t want to go that way. However, it is clear that the multiple conversion steps with P2G2P will add significantly to its costs.  That may be offset to some extent, since it makes available power for direct grid balancing, and we will need a lot of that, if the use of VRE expands. Moreover, if and when that happens, there will, at times, be a lot of surplus renewables power, which otherwise would have to be dumped, wastefully.  However, the study says that there is higher value in meeting hard to decarbonize end-uses with hydrogen direct, at least in the context of the US North East . Maybe, but there are some unknowns. For example, will P2G, meeting non-power demand in hard to electrify areas, provide enough flexibility to balance the grid?  

The UK government’s advisory Climate Change Committee has emphasised that hydrogen use should be restricted to ‘areas less suited to electrification, particularly shipping and parts of industry’ and providing flexibility to the power system, but it is not yet clear exactly how that will work and what the balance should be. Long duration bulk hydrogen storage (in salt caverns), topped up from surplus renewable power, used to meet demand peaks during the occasional long lulls in renewable availability, may still be attractive, if the cycling costs can be reduced.  Indeed, there seem to be few other long duration storage/balancing options, other than compressed air storage and perhaps liquid air storage and maybe heat storage. 

The later two are probably best for intermediate storage periods, while batteries, and even large pumped hydro, are only good for a day or so at the most.  Indeed, in his earlier BNEF hydrogen overview, Liebreich concluded that, since it can be stored in unlimited quantities, ‘hydrogen is therefore the only solution that can provide deep resilience to the highly electrified net-zero economy of the future’. Well, reducing/delaying demand peaks and importing top up power might sometimes be able to help with short term balancing, but not for longer-term renewable supply lulls. Green hydrogen may be costly at present, but it does seem likely that its costs will continue to fall and Liebreich may be right: hydrogen storage may be important and so may its use for balancing.