Grid Scale Energy Storage Deployment By Technology Type: Part 2

Rare and exotic energy storage types!

In the previous post, we looked at the types of grid-scale energy storage that are being deployed most in current and planned projects over this decade.

That is: lithium-ion batteries, pumped hydro, compressed air, green hydrogen, and thermal hydro.

Lithium-ion batteries are the commonest. The other top energy storage types have a lot in common — they involve storing energy in mechanical or thermal form, using a fairly cheap/abundant material (like water or air), in huge reservoirs or caverns.

All of the top non-lithium-ion storage technologies are cheaper per kilowatt-hour than lithium-ion batteries today, which means they may be more suitable for the ultimate goal of complementing an all-solar energy grid. Most BESS operators are only aiming to meet power demand for stretches of 4 hours or less; constructing enough lithium-ion battery capacity to meet overnight power demands in an all-solar grid would be nearly 3x more expensive, and you don’t see anyone doing that.

The current price of lithium-ion batteries is barely higher than their raw materials cost, so lithium-ion can’t get much cheaper (certainly not by a factor of 3) unless the price of lithium drops. That could happen if lithium mining expands! But if not, widespread adoption of all-solar electric grids will probably require the use of cheaper energy storage techniques.

Pumped hydro, compressed air, green hydrogen, and thermal hydro are all old technologies (though some new startups are claiming recent innovations as well.)

What about newer or less common non-lithium-ion storage tech? What’s been developed enough to see any grid-scale adoption planned in the next ten years?

As before, I’m working from the dataset I’ve compiled from the past two years of Energy Storage News articles announcing planned or current energy storage projects.

Vanadium Redox Flow Batteries

• Planned Deployment: 16 projects totaling 367 MW and 1645 MWh

• Cost: $475/kWh, according to this PNNL report

• Duration: 2-8 hours

• Round Trip Efficiency: 65-80%

• Cycle Life: 12,000-14,000 cycles

• Energy Density: 10-20 Wh/kg

Vanadium redox flow batteries work by having vanadium ions flow between a negative electrolyte tank and a positive electrolyte tank. They are quite a bit more expensive than lithium-ion batteries per kWh, due to expensive raw materials, so the odds are against them replacing lithium-ion batteries.

Molten Salt Thermal

• Planned Deployment: 1 project, totaling 100 MW and 1 GWh; counting thermal solar plants, 21 projects, totaling 3.4 GW and 28 GWh.

• Cost: $30/kWh

• Duration: 10-20 hours

• Round Trip Efficiency: ?

• Cycle Life: n/a but can last decades

• Energy Density: 0.14 Wh/kg

Thermal molten salt storage is what it sounds like. Heat generated by solar power is stored in molten salt and turned into electricity by heating up steam to spin a turbine.

The only standalone molten salt thermal energy storage project reported in the dataset is being implemented by Malta Inc, a spinout of Google X’s “moonshot factory”, whose technology incorporates a heat pump to turn electricity into a temperature gradient between hot and cold salt containers.

Molten salt thermal energy storage is more often incorporated into thermal concentrated solar power stations, where it provides overnight energy storage. Wikipedia lists 3.4 GW worth of planned thermal solar plants worldwide, with at least 28 GWh of molten salt storage.

Molten salt thermal energy storage, at $30/kWh, is much cheaper than lithium-ion batteries, but thermal concentrated solar plants are more expensive in capital costs than photovoltaic solar plants ($119-251 per MWh for thermal solar vs. $50-60 per MWh for photovoltaic solar.)

This suggests that thermal solar plants are strictly cheaper per watt-hour than photovoltaic solar + battery storage, for the purpose of providing baseload power.

Thermal solar is still more expensive than coal (in levelized, per-kilowatt-hour terms), but its costs are on a downward trend:

Concentrated solar power plants have been lagging in deployment behind photovoltaic solar, because “photovoltaics are cheaper” and regulations require utilities to purchase power at the cheapest cost per kilowatt, not considering the fact that photovoltaic plants produce intermittent power (unless supplemented with energy storage) while CSP plants come with their own energy storage and are basically “dispatchable” power that’s always available, just like coal, gas, nuclear, and hydroelectric power.

These types of perverse regulatory incentives cause underinvestment in all kinds of energy storage; if grid-scale energy storage operators aren’t allowed to profit off of providing reliable energy, we aren’t going to get reliable clean energy.

Liquid Air

• Planned Deployment: 2 projects, totaling 100 MW and 750 MWh

• Cost: ?

• Duration: 10-20 hours

• Round Trip Efficiency: 60%

• Cycle Life: n/a but can last decades

• Energy Density: 150-250 Wh/kg

Liquid air energy storage uses electricity to cool air until it liquifies, and then releases electricity by allowing the air to warm up to ambient temperature and spin a turbine.

Both planned projects are being conducted by Highview Power, a UK startup which appears to be the only company building this technology.

Liquid air energy storage is cheaper (per kWh) than lithium-ion batteries and doesn’t have the same geological constraints as pumped hydro, compressed air, hydrogen, or thermal hydro.

Thermal Aluminum

• Planned Deployment: 1 project, totaling 65 MW

• Cost: ?

• Duration: 10-12 hours

• Round Trip Efficiency: 90%

• Cycle Life: n/a but lasts 30 years

• Energy Density: ?

The sole planned project for aluminum thermal energy storage is from Azelio, a Swedish startup that uses electricity to heat up an aluminum alloy until it melts; to get energy out, “heat is transferred … through a heat transfer fluid (HTF) to the Stirling engine. A working gas is heated and cooled off by ambient air, and runs the engine.”

The high promised round-trip efficiency is impressive (the only storage technology so far with an efficiency comparable to lithium-ion batteries) but without publicly available price information it’s impossible to evaluate whether this tech can see adoption at scale.

Iron-Air Battery

• Planned Deployment: 1 project, totaling 15 MW and 1500 MWh

• Cost: $20/kWh

• Duration: 100 hours

• Round Trip Efficiency: ?

• Cycle Life: 10,000 cycles

• Energy Density: 250 Wh/kg

The sole planned project using an iron-air battery is by Form Energy, whose battery turns iron to rust to discharge, and charges by using electricity to turn rust back into iron.

Iron-air batteries are very low cost — the only thing that comes close is molten salt thermal energy storage.

Geomechanical Pumped Storage

• Planned Deployment: 1 project, 15 MW, 10 MWh

• Cost: $50/kWh

• Duration: 10 hours

• Round Trip Efficiency: 75%

• Cycle Life: n/a but can last decades

• Energy Density: n/a

The sole project using geomechanical pumped storage is from Quidnet Energy, a startup with a more versatile spin on pumped hydro. Instead of pumping water up a hill, you pump it into a well. The energy is stored in the form of pressure, and released when the water is allowed to flow through a turbine to generate electricity.

Because it doesn’t need to be sited on a hill, it can be used in more contexts than pumped hydro, and appears to be cheaper per kWh, with higher efficiency than hydrogen, compressed air, thermal hydro, or liquid air.

Coming soon: Part 3, even more types of energy storage technology!