Deep within the Department of Energy is a small agency devoted to supporting cutting-edge energy research: the Advanced Research Projects Agency-Energy, or ARPA-E. It’s only about 10 years old and not widely known or appreciated by the public — but among energy geeks, it is beloved.

By all accounts, ARPA-E is a rousing success. The National Academy of Sciences conducted an extensive assessment in 2017 and concluded as much. Of the roughly 500 grants the agency had given out at that point, about half had resulted in peer-reviewed research, about a quarter went on to leverage funding from the private sector, and around 13 percent resulted in new patents. And that’s with a deliberate focus on “high risk, high reward” investments.

The agency — originally created in 2007 by a bipartisan group of US lawmakers, fully funded by President Barack Obama’s stimulus bill in 2009, and put on firmer footing by Congress in 2011 — was consciously designed to mimic the Defense Advanced Research Projects Agency (DARPA), created way back in 1958 to do advanced research for the Department of Defense.

ARPA-E’s purpose is to identify promising advanced energy technologies and help them bridge the “valley of death” between basic research and commercialization — oh, and “to bring a freshness, excitement, and sense of mission to energy research that will attract the U.S.’s best and brightest minds.”

[embedded content]

In its modest way, it has done that. Naturally, because it is a successful agency associated with Obama, Donald Trump hates it.

Both of Trump’s budgets have proposed getting rid of the agency entirely. Both times, Congress quietly ignored him and maintained ARPA-E funding. In 2017, they actually increased it; in 2018, they ended up holding it steady (despite a House proposal to cut it by 8 percent). Whether it continues to survive will depend a great deal on the outcome of the 2018 midterm elections.

Anyway, despite all that drama, whiz kids at ARPA-E are still plugging away, funneling money to interesting energy projects.

The latest news from the agency is that is has chosen 10 recipients to receive the $30 million the Department of Energy set aside back in May for the Duration Addition to electricitY Storage (DAYS) program. (Nobody said energy nerds were good at acronyms.) DAYS will fund a set of exploratory projects and experiments in long-duration (between 10 and 100 hours) energy storage.

Long-duration storage is, as I’ve written before, a kind of a holy grail in energy. Let’s quickly review why.

More wind and solar are coming on the grid, which is a good thing and reduces carbon emissions. But renewable energy is variable — it comes and goes with the weather. The more of it you add, the more you need other flexible resources to balance it out. Natural gas works well for that, but getting to a zero-carbon grid means getting rid of natural gas (unless it adds carbon capture and sequestration).

What can serve as flexible, zero-carbon resources, taking the place of natural gas in balancing variable renewables? Well, lots of things, but mainly energy storage. Being able to store the renewable energy when the sun and wind come and discharge it when they go smooths out the steep swings.

Wind+storage in Australia.

Tesla’s battery installation in South Australia.

We have the storage technology to smooth out hourly swings, but we still don’t have anything that could cover a fallow period of wind and sun that lasted days, months, or even years. If we want to get variable renewable energy up to 60 percent, 80 percent, or even more of our electricity, we need long-term energy storage. It is the missing puzzle piece, the holy grail.

Nonetheless, long-duration storage companies have had trouble surviving in a market dominated by cheap lithium-ion batteries. Thus the need for some technology long shots.

So what’s cooking in advanced long-duration energy-storage research? Looking through the 10 recipients of ARPA-E’s grants offers an intriguing glimpse. Some of these technologies are meant to cycle energy both daily and in longer durations, and some are meant primarily as long-term storage, to be drawn upon when daily storage is exhausted.

The projects are varied, but a few categories stand out. (Interestingly, despite the considerable hype around electrolyzed hydrogen as a storage mechanism, there’s only one electrolysis project in the group.)

The first and biggest category, with five out of 10 of the projects, is thermal storage, i.e., storing electricity as heat. There are all kinds of variants on this, but the basic idea is: Use surplus electricity to heat something up; when you need electricity, use that heat to run a turbine. It turns out to be much easier to store heat than to store electricity.

amazon chillers

One way to store heat: as warm water.
Jordan Stead/Amazon

It’s all in what you heat and how much of the energy you get back out. At Michigan State University, they will heat “a bed of magnesium manganese oxide (Mg-Mn-O) particles.” Brayton Energy, in Hampton, New Hampshire, will heat molten salt. Echogen Power Systems, in Akron, Ohio, will heat “a ‘reservoir’ of low cost materials such as sand or concrete.” The National Renewable Energy Laboratory, in Golden, Colorado, will heat “inexpensive solid particles to temperatures greater than 1100°C” and then get the energy out using “a high performance heat exchanger and closed loop Brayton cycle turbine,” which certainly sounds cool.

Antora Energy, in Fremont, California, will heat “inexpensive carbon blocks” (to 2000° C!). Antora is somewhat unique in that it will get the energy back out not through a turbine, but with “thermophotovoltaic” solar panels “specifically designed to efficiently use the heat radiated by the blocks.”

Thermal storage doesn’t get a lot of press in the energy world — heat is somehow less sexy than electricity — but it has enormous potential to speed decarbonization. It would be awesome to see one of these techs catch on.

Another interesting category is flow batteries. Unlike the lithium-ion batteries most consumers are familiar with from their cellphones and electric cars, flow batteries store energy in liquid form. Specifically, they involve two separate tanks of fluids containing chemical components, which are circulated past a membrane, where ions and electric current are exchanged.

Flow batteries are heavy and not useful for vehicles, but as stationary storage, they have great advantages: They can scale to basically any size, charge and discharge tens of thousands of times without loss, and last well over 20 years. It’s just two tanks of stable fluids at room temperature. You can store those fluids as long as you like, or recharge them as often as you like, and make the tanks as big as you like.

flow battery

The basics of a flow battery.

Actually, Primus Power in Hayward, California, is developing a zinc-bromine flow battery that doesn’t require the two reactants to be separated at all, so they can be stored in a single tank together, saving hardware costs. The United Technologies Research Center, in East Hartford, Connecticut, is working on a flow battery using “inexpensive and readily available sulfur-manganese based active materials,” which, I mean, duh.

The one electrolysis project is out of the University of Tennessee, in Knoxville, which is combining an electrolyzer (which separates hydrogen from water) with a fuel cell that produces hydrogen peroxide. I’m not even going to pretend to understand this one.

And the last one worth mentioning, just because it seems peculiar, is Quidnet Energy, in San Francisco, California. It’s developing a kind of inverse version of “pumped hydro.” In the normal version, water is pumped uphill to store energy and then run downhill through a turbine to release energy. Instead, to store energy, Quidnet will pump pressurized water underground; to release energy, it will stop pumping and “the induced strain in the surrounding rock will force water back through a generator to produce electricity.” Wacky!

The DAYS recipients represent only a fraction of the research that’s going into long-duration energy storage. And honestly, their grants represent only a fraction of the money that ought to be directed to such research. Given how significant long-duration storage is to a sustainable power system, $30 million is beans. We ought to be spending like we take the climate crisis seriously.

doe research budget


But still. It’s cool to have a collection of energy nerds running a competent, forward-looking agency that supports and enables other energy nerds. Somehow this little miracle of good governance has survived the hurricane of Trumpian nihilism.

Who knows how long it can last. But for now, ARPA-E serves as a kind of echo of the Obama years, and of the idea that America can confidently reach for its future rather than clinging to its past.