Thread by @JesseJenkins

Investors are pouring billions into novel 'long-duration #energystorage technologies. What techs are most promising?

Dept of @Energy's Storage Earthshot wants to cut storage costs 90%. Is that enough?

How long is 'long duration'?

My new paper w/@nsepulvedam has answers... 🧵

https://twitter.com/JesseJenkins/status/1435720013944610827

Our new @Joule_CP commentary (authors.elsevier.com/a/1dj0d925JEG7…) builds on a recent @NatureEnergyJnl research paper (nature.com/articles/s4156…) to sift through the broad ‘‘design space’’ for potential long-duration energy storage (LDES) technologies & offer insights to guide innovation...

In short: to meet the cost & performance targets identified in our research, LDES techs need:

1. Ultra-low energy capacity costs (~$1-10 per kilowatt-hour the device is capable of storing)
2. Suitably high efficiency (particularly for discharge)
3. >100 hrs sustained discharge.

Using these insights, we can then filter through the diverse set of potential long-duration storage technologies to help focus efforts on those candidates most likely to succeed. This detailed figure sums it up, but below I'll walk through several candidate technologies...

There are 4 main types of energy storage devices:

1. Mechanical (e.g., compressed air or pumped hydro),
2. Electrochemical (e.g., flow batteries or metal air batteries),
3. Thermal (e.g. ceramic bricks),
4. Chemical storage technologies (e.g., hydrogen).

Which work for LDES?

Mechanical options...

Pumped hydro: Very large reservoirs (e.g. Snowy 2.0 in Australia) that can get costs down to ~$20-30/kWh + >100 hrs duration, but most are sized for daily cycles (8-24 hrs) & too costly (>$100/kWh). Geographically constrained.
LDES Verdict: 🤷‍♂️

Compressed air: In very large saline aquifers or depleted gas fields could result in costs as low as $1/kWh with 100s of hours of duration, while more conventional projects in salt caverns have higher costs and shorter durations. Geographically constrained.
LDES Verdict: ✅

Gravity based energy storage (e.g. Energy Vault): stacking concrete blocks (or running trains full of concrete uphill) is too costly ($200-250/kWh) to be competive for long-duration storage.
LDES Verdict: ❌

Geomechanical energy storage: pumps water undergound to store energy in the elastic compression of rock formations (e.g. Quidnet). May achieve costs in the $50/kWh range. Suitable for diurnal, but not 100+ hours.
LDES Verdict: ❌

Wildcard not included in paper, but subject of research w/Wilson Ricks of ZERO Lab + @JackNorbeck of Fervo: use similar geomechanics to store energy in reservoir of an advanced geothermal energy system, making a combo clean firm power plant + LDES. Very promising. Stay tuned...

Electrochemical options...

Flow batteries (conventional): most widely studied flow battery variants, vanadium redox and zinc bromine, have been estimated in the $100 range per kWh, too high to serve as a cost-effective LDES tech.
LDES Verdict: ❌

Flow batteries (novel): alternative flow batteries using ultra-cheap materials could get costs into $10-20/kWh range and 100+ hours. See e.g. aqueous sulfur storage tech advanced by @MIT's Fikilie Brushett, Yet Ming Chang & others. energy.mit.edu/publication/ai…
LDES Verdict: ✅

Metal-air batteries: @FormEnergyInc has been making headlines (e.g. wsj.com/articles/batte…) for their low-cost iron-air battery that they claim could get costs into the $20/kWh range with 100-150 hours sustained discharge. Cost could fall with deployment.
LDES Verdict: ✅

Liquid metal batteries: Not included in our lit review, but decade+ old @Ambri_Inc is back in the news and has raised more $ for manufacturing its liquid metal battery. They're targeting diurnal (24-hr) market & costs in $50-75/kWh range last I heard (Ambri?).
LDES Verdict: ❌

Chemical storage options...

Chemical energy storage candidates such as hydrogen, synthetic natural gas (SNG) & ammonia have potential to achieve very low energy storage capacity cost and uniquely exploit additional revenue streams due to value of chemical fuels in other sectors.

H2 & SNG: w/large underground storage (saline or depleted gas fields), costs could be <$1/kWh! Round trip (esp. discharge) efficiency low, which negates some of that advantage. Metal tanks cost $10-15/kWh. Electrolysis & fuel cell costs need to fall a lot.
LDES Verdict: ✅

Thermal storage options...

Molten salt storage: The most mature thermal storage option for electricity, used at concentrating solar PV sites. OK for diurnal, but too costly for LDES ($30–80/kW + costly charge/discharge power).
LDES Verdict: ✅

High-temp sensible heat storage: firebrick resistance-heated energy storage (FIRES) stores heat at >1,000C in ceramic bricks; could cost $5-10/kWh with very cheap charge capacity. Uses CT or CCGT for discharge ($$$) but could be economic. news.mit.edu/2017/firebrick…
LDES Verdict: ✅

Reversible heat pump energy storage: very high charge efficiency (>100% due to moving heat not creating it), low discharge efficiency (low temps mean low thermodynamic efficiency). Energy storage capacity cost $15-25/kWh. On the borderline.
LDES Verdict: 🤷‍♂️

LDES could be an important new tool in the decarbonization toolbox. There are myriad options. My research w/@nsepulvedam @dhariksm et al can help focus investment, R&D and innovation on most promising of these by providing concrete performance & cost targets. 4 recommendations...

1. Given limited $ and urgent need to decarbonize, investment & effort should carefully focus on a handful of specific techs that have the greatest potential to deliver requisite cost/performance goals. Not all techs capable of storing energy for long periods are viable options

2. True 'long-duration' energy storage w/greatest impact means storage capable of sustaining discharge 100+ hours (4+ days). This is longer than targeted by @Energy Storage Earthshot or @ARPAE DAYS program. Diurnal storage is a distinct opportunity (& less of a 'breakthrough').

3. While expanding R&D $$ for LDES techs is important, we know from experience with solar PV, lithium-ion and many more techs that cost declines & innovation are also driven by successive deployment. Commercial-scale demos & proactive policy support for early deployment are key.

3b. Technology policy efforts for LDES should include initial market creation via cost-shared demos, competitive procurement, and/or subsidies to ensure proactive deployment ahead of wide-scale market needs. (Good news: Bipartisan Infra. Bill & Biden budget plan has $ for this).

4. Proactive support for LDES technologies should NOT come at the expense of (or be viewed as a reliable alternative) to development and deployment of firm low-carbon power generation technologies (for more, see doi.org/10.1016/j.egyc…).

4b. Development of competitive LDES techs remains uncertain + our research indicates complete substitution of LDES for firm power generation is unlikely even if cost targets reached. Betting narrowly on LDES is too risky. Best outcome is wind/solar + Li-ion + LDES + clean firm.

If you've made it this far in this mega-thread, congrats. You dont have to read our paper! You've probably got the gist of it. But for more (and better prose!), please read and share the full paper at authors.elsevier.com/a/1dj0d925JEG7…

</END>