Why hydrogen may be the principal media for energy storage
Over the course of the last week (December 12th – December 19th), South Australia produced more renewable power than its entire needs for electricity.[1] The state proudly claims that this is the first time that a large electricity market has met all its requirements from wind and solar. It also suggests that by March of next year, aggregate needs may be met for over a month. This is a striking picture of how renewables will look at many places around the world within a few years.
The state is connected into Australia’s National Energy Market (NEM) which can buffer periods of over-supply or shortages. In addition, South Australia has some gas generation used to make electricity when wind and solar outputs are insufficient. But high wind speeds over the last weeks have meant that electricity generation from turbines has been able to meet almost all instantaneous demand. The previous week saw gas only provide about 9% of all electricity produced in the state, with wind generating almost ten times as much.
This is the context for two important decisions announced this week.
· A plan by the state of South Australia to invest in one of the world’s first power-to-gas-to-power (P2G2P) plants reached the stage of requesting tenders from suppliers.[2]
· The national government of Australia proposed to build a set of large batteries, partly funded by its renewable energy agency (ARENA).[3]
Both of these schemes are a response to the growing need to productively store electricity surpluses to meet demand when wind and solar outputs are not sufficient.
The P2G2P plant
South Australia intends to build a 250 MW electrolyser to generate hydrogen, combined with a 200 MW hydrogen gas turbine power plant. I think this is by far the largest P2G2P proposal in the world. The intention seems to be to use 100% hydrogen from the opening in late 2025, unlike other plants, such as Intermountain in Utah, which are intending to gradually phase in increasing percentages of H2
When electricity is abundant, the electrolyser will make hydrogen, storing it until it is needed. The power plant will then use the gas to generate power. The proposed storage facility will hold 3,600 tonnes of hydrogen, which will have an energy value of about 120 gigawatt hours. This is approximately equal to 1% of South Australia’s annual electricity use.
South Australia says it has budgeted AU$593 (US$400) million for this proposal. The state itself will own and operate the facility. At today’s approximate figures of US$600 per kilowatt, the electrolyser will cost about US$150 million. The storage facility will probably be the next largest element.
We cannot know yet whether this plant can be built for US$400 million but the government seems optimistic. At least 60 external companies have registered interest in bidding for elements of the project.
The plan is to use land near the city of Whyalla, one of the centres of South Australia’s steel making industry. Steel will need huge quantities of hydrogen and may also be a customer from the hydrogen.
Large batteries for the grid
Australia’s ARENA announced eight locations for grid storage batteries, including one in South Australia. These will be constructed so that they can respond nearly instantly to the need for power and will assist in maintaining the correct frequency and voltage of the AC grid. (Apparently this is called ‘grid-forming’ capacity).
ARENA expects battery suppliers to quote costs of around US$1.8bn for these installations. They which will provide about 2.0 gigawatts of power and a total energy storage capacity of 4.2 gigawatt hours. the biggest will be 900 MWh, making it prospectively one of the largest batteries in the world. ARENA offered grants of around US$120 million towards the cost of putting these 8 systems in place.
Some of the implications of these projects
One thing immediately stands out comparing these two sets of projects. The single P2G2G installation will offer about 118 GWh of storage compared to about 4.2 GWh for the eight batteries. But the cost of the whole P2G2P project should be around US400m, compared to US$1,800 for the batteries. If these numbers are right, storing energy in the form of hydrogen is therefore less than one hundredth of the cost of the same energy value of electricity in a battery.
This is why I think hydrogen will be the means by which renewable-dominated grids are balanced. P2G2G systems are widely criticised for being ‘inefficient’ because the energy losses from making hydrogen and then converting it back to electricity are at least 50%. (I assume 75% efficiency in the electrolyser and 60% of energy going into a turbine being converted to electricity). But the capital cost of a large storage system for hydrogen is absolutely tiny compared to a battery, which might achieve 90% round-trip energy efficiency.
We will need batteries for very rapid response to changes in grid conditions. But the large bulk of energy storage needs will be provided by hydrogen systems such as the Whyalla power station.
If we assume that the Australian battery systems last 18 years (optimistic), charge and discharge 400 times a year (very optimistic indeed) and that the cycle goes from 100% charged to 0% charges (an impossible assumption), the capital cost per megawatt hour of charging is just under $60, without including any interest burden.
The same calculation for hydrogen storage produces a figure of about 50 US cents for each megawatt hour. If these Australian figures are realistic and typical, in normal conditions a hydrogen storage system is vastly cheaper to own and operate. Batteries will only be better than hydrogen in financial terms if the electricity to be stored is very expensive and therefore the P2G2G efficiency losses are costly.
This is unlikely. In a system dominated by renewables, periods of surplus electricity will see very low prices, and curtailment of wind and solar production. So the efficiency losses do not make hydrogen systems inherently uncompetitive. Put simply, the high capital cost of operating batteries will be greater than the cost of the losing energy via the hydrogen storage process.
Of course, prices may change. Batteries may get cheaper and hydrogen may prove more expensive to create and store than South Australia believes. But at the moment it looks as if large scale storage to deal with the growing number of hours and days of surplus electricity, such as South Australia is already seeing, will be accommodated using electrolysers and hydrogen turbines. Europe should sponsor a similar P2G2P plant now.
[2] https://www.whyalla.sa.gov.au/our-city/news-and-events/latest-news/hydrogen-power-plant-progresses-to-next-milestone
[3] https://arena.gov.au/blog/arena-backs-eight-big-batteries-to-bolster-grid/