Tank storage is challenging because at that scale you're competing with batteries which have much better round-trip efficiency. There was a DNV GL paper earlier

Tank storage is challenging because at that scale you're competing with batteries which have much better round-trip efficiency. There was a DNV GL paper earlier this year which quoted a time of about 24 hrs as the cross-over point between H2 and battery tech. Underground cavern storage lets you do inter-seasonal storage at a very low cost per MWh. It's critical to understand that liquid and gas fuels have uncoupled per MW and per MWh costs in a way that batteries don't. You can adjust battery chemistry of course but fundamentally you're stuck with a narrow range of peak MW / MWh performance which means that if you want to build a massive battery for inter-seasonal storage (needs MWh >>> MW) you end up "buying" discharge capacity in MW that you don't need. Hydrogen decouples the size of the production (in MW), the storage (MWh), and the fuel cell (MW) which lets you size them independently of each other. I am also sceptical of the need for home use or vehicular hydrogen (certainly in passenger cars). You get a lot of efficiency improvement if you are able to use the waste heat from fuel cells. Yes that can be done domestically, but putting them in large urban centres and driving district heating schemes from them seems like a much more natural fit. Especially in places like Europe where total energy demand (heat and electricity both) peak at the same time in the winter as renewable generation decreases. N.b. there are already times in the UK grid when prices are negative wholesale. Even one of the retail energy suppliers here had a few half hourly periods of negative prices a few months ago. The subsidy for hydrogen electrolysers in The Netherlands was set assuming 2,000 hours a year of operation (using the cleanest and cheapest 2,000 hrs) since that maximises the environmental benefit with the current Dutch grid.