Making hydrogen from natural gas inevitably means hydrogen is more expensive than natural gas. But if hydrogen is made using electricity from renewables, it could become cheaper than natural gas, at least during periods of surplus low carbon electricity supply.
This is the third of three posts about hydrogen in a low carbon economy.
Making hydrogen from natural gas, the main approach at present, inevitably means that low-carbon hydrogen is more expensive than natural gas per unit of energy. This is because there are additional costs involved in making the hydrogen, and these will remain even if there is substantial technological progress. The major costs are:
- The capital and operating costs of the reformer that converts natural gas to hydrogen
- Energy losses in the reforming process.
- The additional cost of CCS, essential to make the hydrogen low carbon in this way.
- Some emissions will remain, even with CCS, imposing an additional cost from carbon pricing, and requiring measures to absorb carbon in a net-zero economy.
These additional costs mean that hydrogen produced in this way is inevitably more expensive than natural gas – typically by a factor of two or more, even allowing for technological progress. This is likely to be a barrier to displacing natural gas with hydrogen.
The other route for making low carbon hydrogen, electrolysis, is now more expensive than using reformers. As a result, it accounts for only about 4% of total manufacture. When the UK’s Committee on Climate Change looked at the potential role of hydrogen in a net zero emissions economy in the UK it concluded that reforming is likely to continue to predominate, because electrolysis is likely to remain more expensive, and would require very large amount of low carbon electricity[i].
But could the costs of electrolysis come down by enough to make it competitive?
The costs of electrolysers have already come down markedly, by around 40% in developed economies according to an estimate from BNEF, with costs in China already lower still. The potential for further cost reductions from experience is likely to be very large, because the size of the market for hydrogen is likely to grow to many times its current size, and electrolysis could take a larger share of this larger market. This could lead to large cost reductions of the type already seen for wind power, solar power and batteries.
The main barrier to reducing the cost of electrolysis to below that of reforming is the price of low carbon electricity. Electricity is typically more expensive per unit of energy than natural gas, and this makes is difficult to compete as a source of hydrogen. However, the costs of renewables continue to fall, and as they become a larger part of the system, periods of surplus will become more common. In these periods electricity is likely to become very cheap, perhaps with a price at or close to zero. Hydrogen manufacture becomes a means of storing the energy in this surplus electricity.
This may give opportunities for lower cost hydrogen manufacture using cheap renewable electricity, provided the electrolyser is sufficiently cheap and flexible to enable economic low load factor operation. Eventually electrolysis could become cheaper than reforming, at least at times.
It is even possible to that low carbon hydrogen from electrolysis could become cheaper than natural gas. This would require very low cost electricity, most probably during periods of substantial surplus on the grid. However, as renewables costs continue to fall, especially for solar, electrolysis could even be competitive when electricity systems are not in surplus.
However, the materiality of this will depend on the amount of surplus and very low cost renewables relative to the scale of hydrogen demand. In the UK at least there is unlikely to be enough surplus renewables power to make the large amounts of hydrogen required for a net zero emissions economy.
Whatever the eventual outcome, policy should recognise the uncertainties. It should allow for the possibility of cheaper hydrogen from electrolysis, and the impact this might have.
Adam Whitmore – 23rd January 2020