Why the intermittency problem can't be solved

I often ask renewables enthusiasts to explain what we are supposed to do when the wind isn’t blowing if we can’t fall back on fossil fuels. The other day, I pressed James Murray, the editor of Business Green magazine, what forms of storage he thought we could use, and this is what he said:

… a portfolio of nuclear, demand response, grid scale batteries, other emerging forms of energy storage technologies, hydrogen, and gas, ultimately in conjunction with CCS.

Clearly, we were talking somewhat at cross purposes; my question was specifically about storage, but even if we broaden the scope to cover the general question of “what do we do when the wind isn’t blowing”, his answer suggests that he hasn’t grasped the fundamental economic problem.

That problem is that, with wind dominating the grid, for anyone looking to make money in the lulls, the economics look grim. There are two major kinds of lull that need to be filled. The first is a dunkelflaute, a lull in the winter, when solar is generating little or nothing. We get a dunkelflaute most years, and sometimes more than one. They can last from 1-3 weeks. The second is the long summer lull, with low wind generation right through the summer month, although perhaps with occasional windy interruptions. This happens every year of course, and a large amount of energy needs to be stored to cover the gap: perhaps as much as 50 days’ demand.

If we are talking about storage then, most of it will barely be used; it's required just once a year to deal with the summer wind lull. Most of it will be filled in autumn, and will then sit there waiting for the summer, when it will be emptied to meet demand, before sitting empty again until the winds pick up again as the nights draw in.

Making money on this basis is impossible. A kilowatt hour of lithium ion battery storage might cost £350. If, optimistically, it gets perform two charge-discharge cycles per year, it will complete just 20 cycles over its lifetime. That means it needs to charge £17.50 per kilowatt hour, just to cover its capital costs; the electricity is extra! That is perhaps thirty times the level seen at the peak of the crisis last year, and 300 times the prices we used to enjoy before the advent of “cheap renewables”.

Of course, cheaper storage systems may be on the horizon, so it’s worth looking at these. The best bet on the horizon seems to be liquid air storage, which has a 25-year lifespan, so might be expected to complete 50 recharging cycles. Its capital costs are also much lower, but it will still need £1.68 to cover its capital costs. That’s three times the peak price last year, and thirty times what they were in the good old days.

Needing a large amount of electricity just a couple of times a year makes the economics impossible. It's not just storage technologies that are affected - James' idea that we could use nuclear to plug the gap therefore doesn't stack up. Who is going to build a nuclear power station that only gets to run for 50 days a year? The idea is preposterous. In essence, the intermittency problem can't be solved. The costs of doing so make it impossible, for any technology, even on the most optimistic assumptions about cost trajectories.

As a coda, it’s interesting to note that the Committee on Climate Change’s model for a net zero energy system has a vast fleet of gas turbines (122 GW of them!) burning hydrogen to deal with the intermittency of its vast fleets of wind and solar. But the power stations get run very rarely – they deliver just 2% of their capacity each year. By my calculations this means they will deliver power at around £1/kWh, or 40 times the prices from the good old days. However, the CCC, perhaps wisely, has accidentally missed the bill for these units out of the final reckoning of the cost of net zero.