The US Department of Energy's paper, "Grid Operational Impacts of Widespread Storage Deployment" analyzes requirements to 2050. It variously calculates 1.3 TWh required to just over 6.0 TWh in its 94% renewable electricity, Zero Carbon scenario. Earlier, other researchers calculated 80% wind and solar will require 5.4 TWh of energy storage so it is obvious that huge amounts of storage are needed. Even that is with massive over-production and therefore curtailment - electricity only being generated 10% to 20% of the time on average. Polluting peaker plants only have 11% capacity factor so, on that model, bad replaces bad in that respect.
6 TWh would cost about $2 trillion at current battery prices when US gross domestic product is $23 trillion. Indeed, some lithium-ion battery prices are rising, driven by huge price increases for their three key metals and there are frequent shortages. It is obvious we cannot go on like this and the challenge is global.
The report does not analyze storage of 12 hours of output and longer but notes it will be an additional requirement as 100% clean electricity is approached. Think solar dead at night and no more than 25% as powerful in winter and note wind can be absent for up to months. More trillions of dollars?
In contrast, the IDTechEx report, "Future Stationary Energy Storage: Hydrogen, Batteries, Gravity, Gas, Other 2022-2042" gives reason for optimism, citing many less-intermittent zero-emission sources, better storage methods, and mitigating options coming along.
Overcapacity of ever-more-affordable solar makes sense to get useful power even at dawn and dusk but, as the DOE recognizes, you do not need to switch off the over-capacity at peak times. Make hydrogen or ammonia as industrial feedstock, preferably not suffering the inefficiency of reverting it to electricity at a later date. IDTechEx adds that, alternatively, you store it in new, low-cost forms of delayed electricity.
IDTechEx finds you can greatly reduce the intermittency of supply in the first place by adopting some of the emerging ocean wave, tidal, and distributed geothermal power in the years to come. Tethered drones and kites and even giant conventional wind turbines tap wind that is stronger and less intermittent higher up. Both grid inputs and storage are increasingly available from electric vehicles and microgrids. Add to that extending your major grid across a greater variety of weather patterns, as planned in Europe.
More immediately, we are questioning all-solar or all-wind facilities because the two together can offset at source rather than through some distant grid connection. In many locations, wind is stronger at night when solar is dead. An example is the Goyder Renewables Zone wind-solar-storage project in South Australia in 2022 with 1200MW of wind, 600MW of solar PV, and 900MW of battery energy storage.
The IDTechEx report assesses many forms of storage arriving that will displace lithium-ion batteries, massively reducing cost and even addressing the long-term storage requirement, something they will never provide. For storage up to weeks, there is very good news, with many forms of redox flow batteries, compressed air, liquid air, hot rock, gravity lifting weights, reinvented pumped hydro, and other options, some having GW levels of commitment already. Some can repurpose fossil fuel generation plants, even using their turbines.
China continues giant pumped-hydro schemes and the US is doing one at 24GWh. Underwater and heavy water versions may later lead to hydro in previously-impossible locations without the environmental black marks and maybe slow hydro losing share to batteries.
Further, sodium-ion batteries are now coming to market promise to beat lithium-ion on cycle life, low-temperature performance, and cost without problems from valuable metals, shortages, flammability, recycling, and so on. Significantly, this year, lithium-ion battery leader CATL has announced one, and Indian giant Reliance has bought the European sodium-ion leader Faradion, citing use for grid-scale storage and backup power. The more speculative options such as iron batteries are coming later but IDTechEx does not see advanced lithium batteries such as lithium-sulfur having appropriate attributes for future mainstream stationary storage.
Lithium-ion battery sales are growing rapidly for stationary energy storage, with bigger and bigger units. Nonetheless, IDTechEx suspects that recent announcements such as PG&E planning a massive 6.4GWh of four-hour lithium-ion battery storage to 2024 in California may not represent unstoppable progress of lithium-ion but rather an approach to peak adoption in stationary storage from 2026-2030. Far superior alternatives are coming in so rapidly that lithium-ion may have as little as ten years' window of opportunity in stationary storage, the end coming rapidly, accelerated by their diversion to tens of millions of solar houses and electric vehicles yearly where they are a much better fit. The good news is that you can make sodium-ion batteries in your lithium-ion factory.
There is far less choice emerging for months to seasonal storage. Six options are assessed by IDTechEx. Some can perform the shorter-term storage as well. IDTechEx believes that the world may spend around one trillion dollars on all energy storage in 2042 but the stationary storage part will buy much more and be much more impactful than is currently appreciated.
As IDTechEx CEO Raghu Das points out, "The future of stationary energy storage is one of continual disruption. Extrapolations based on what we have now will be very misleading."