Mitsubishi Power has put out a press release, comparing a number of salt caverns in Utah that can hold hydrogen to existing energy storage from lithium ion, and says the deal, which was originally cut over a year ago, will eventually make it the king of energy storage in the US – if it ever gets used and it isn’t just a dramatic waste of $1.9 billion.
The release claims that it effectively has orders for 151,000 MWh of energy storage, which is not quite true. What it admits to are orders for 920 MWh of lithium ion short term storage, which is considerable given that Rethink has estimated that some 4,456 MWh of battery energy storage was actually installed during 2020, doubling existing BES in the US.
But this much does not make Mitsubishi a dominant force in lithium ion energy storage, that honor has to go to Fluence which had deals as early as June last year for 2,300 MWh of battery energy storage placed in 2020.
But Mitsubishi Power is trying to compare this to salt caverns which may or may not end up being used for hydrogen storage, claiming it would be as much as 151,000 MWh simply because the amount of hydrogen would fit in them. Surely that would be all that the state of California could ever want, amounting to a gold mine’s worth of hydrogen.
This goes back to a deal that Mitsubishi signed with local developer Magnum in 2019, to develop 1,000 MW of clean energy to satisfy California that it could have the best of both worlds, the flexibility of natural gas turbines with the zero carbon of solar and wind. The idea is that Mitsubishi has 5 years to build a wind resource in Utah, near the caverns, and contract with a cheap electrolyzer maker to install multiple units, and then convert and store the wind energy as hydrogen, presumably using water delivered to it by the Los Angeles Department of Water and Power (LADWP) which is the customer here. It stores that hydrogen in the caverns waiting for a day which stalks California’s nightmares – when the solar and wind facilities it is building now fail in some way, along with all the battery energy storage that CAISO has contracted for – so far some 3.3 GWh and rising.
Think about the value that a source of energy would have at that moment, the ultimate spinning reserve. And on that day if this was an investor owned facility (it isn’t it’s public money) it could charge whatever it liked for electricity, because it would be an emergency. The hydrogen is routed to the turbines, which run up until 2025 in natural gas mode, then Mitsubishi “adjusts them” to run with 30% hydrogen, and then finally some time before 2045, they are converted to run ONLY on hydrogen.
We know that power to hydrogen is fairly efficient, perhaps running at 70%, and that hydrogen to power, like any thermal turbine is closer to 45%, yielding a final output of something like 30% of what went in from the original wind turbines. In all this is a $1.9 billion project which the LADWP pays for, making California energy even more expensive than it already is (It is among the most expensive cost per MWh in the US).
All of this was to replace a 1.9 GW coal-fired generator at the Intermountain Power Plant (IPP) site in Delta, Utah, for the city of Los Angeles and other parts of California.
The gas turbines will enter operation fueled by a 30% blend of green hydrogen, with the remainder natural gas. But the 840MW of turbine capacity is due to complete a phased switch to 100% green hydrogen by 2045.
Each salt cavern will hold 5,500 tonnes of H2, enough to fill up a million fuel cell vehicles. The gas turbine will not go live before the end of 2025 at a cost of $865 million so presumably the other parts of the system – wind, electrolysis and storage – will cost the best part of $1 billion, and even then this is another order book item, not actually delivered and not to be compared with lithium ion.
There is a big issue of how well we expect a natural gas turbine, come 2025 to compete. It is likely it will struggle to ever make electricity cheaply enough to sell to CAISO, competing with wind and solar which is 5 years cheaper than today, using this turbine. It could end up simply becoming one really expensive insurance cost.
We understand the idea, not unique to Mitsubishi (GE is doing the same) of promising turbine customers that if hydrogen emerges in large volumes, then it can move their turbine over to a fuel that will not trigger a carbon tax. They sell it as a future-proof transition strategy, but one which we suspect is nailing down the wrong aspect.
In the US we estimate that existing natural gas turbines will be economically viable until 2035, but no later. New ones that have to repay their installation costs are already not viable. It will be long after 2040 before hydrogen running through could possibly be economical.
Long before anyone wants to convert this turbine to hydrogen, it is likely the electricity it produces will be undermined by renewables, in price terms and untradeable – so for emergencies only.
As for comparing this with Battery Energy Storage, that was just a whim of someone writing the press release we assume, because those BES are what will prevent this system ever being called upon.