As the renewable energy industry grows and becomes a larger part of our energy mix, the concept of energy storage has made its way into the spotlight and has created some important pressing questions: Do renewables' inherent intermittency require some kind of energy storage, and should it be at the endpoint of use or closer to the utility (or both)? Do we even need an "energy storage" application, or can grid flexibility and responsiveness assume this role? If energy storage is indeed embraced, how should we weigh the options?

Pike Research says nearly 56 gigawatts (GW) of "long-duration" bulk energy storage for the grid (ESG) will be installed from 2012-2022. Installations of energy storage for "ancillary services" alone (things like scheduling and dispatch, reactive power and voltage control, system protection) will increase more than ten-fold to surpass 3.5 GW in that timeframe.

We recently ran a story looking at several of these options, and some companies with new technologies in each area. Pumped hydro has been the go-to energy storage option proven to work at fully-deployed grid scale. Compressed energy, meanwhile, is fairly cheap where it can be deployed appropriate to grid-scale applications, using geological formations (caves or caverns) that can be relatively well sealed off.

Batteries are getting some utility-scale attention now, too. Duke Energy's 153-MW Notrees wind power project has a 36-MW battery storage system courtesy of Xtreme Power to deploy reserve power and help both the system operator and the grid (ERCOT) balance supply and demand. Elsewhere in Texas, Xtreme is working with Samsung SDI to provide a 1 MW/1MWh lithium-ion-based battery energy storage system as part of a $27 million "Smart Grid Demonstration Project". In the UK, S&C Electric and Scottish and Southern Energy Power Distribution have commissioned a pilot project with three single-phase 25 kWh lithium-ion batteries.

Battery technology is fast becoming "one of the favored options for grid-scale energy storage," says Aaron Feaver, CTO of EnerG2. It "has been deployed semi-successfully in grid-scale installations" such as backup power on hospitals, data centers, and renewable energy sites. It's extremely low-cost/kWh, though shortfalls in cycling and power mean batteries need to replaced every few years or even months. The beauty of batteries, though, is that the technology is ripe for cost and performance improvements, and generally speaking it can be added anywhere. Pumped hydro "works only when you have a hill," he says, while compressed storage needs leak-proof caves or caverns (though some new entrants claim to use pipes instead).

(EnerG2 has skin in the battery game, of course. Most carbon-based batteries require some kind of precursor; cheap, but more complicated applications require precision in material quality. EnerG2 engineers specialty carbon for batteries and ultracapacitors with high purity and surface area, removing the need for bioorganic precursor materials made from polymers to coconut husks to tires, substantially improves the battery's performance. And it's not the only one seeking to improve battery materials, either.)

To those who say energy storage isn't needed on the grid, Feaver counters that its current absence is precisely one of the supporting factors for having it. How can one rely on existing grid reliability as the answer to energy storage, given California's constant brownouts, the "ridiculous" blackout in the U.S. Northeast in 2003 (which kicked offline an estimated 45 million people across eight states, plus another 10 million in Canada, for up to two days), and numerous smaller events, including extremely high temperatures across the south (e.g. Texas) that burden the grid? "No doubt we have a fundamental stability issue with the grid," Feaver sums up. And adding solar and wind capacity will strain that even more.

One way to compensate for that instability is natural gas peaker plants, which utilities can ramp up fairly quickly, but their utilization is very low (only used when needed) and there are some inherent inefficiencies associated with turn-on and shutoff. Energy storage is a better option, Feaver notes, especially for more remote areas where power is less reliable. (He pointed to cell-phone towers in India, which rely on lead-acid batteries as well as diesel fuel-guzzling generators.)

But to truly get to grid-scale energy storage in batteries, "we probably need to move to a slightly different paradigm," Feaver explains. A few dozen pouch cells can be dropped into a Chevy Volt, but tens of thousands of them linked up isn't necessarily a cost-effective practice. To fundamentally drive down costs at grid scale, there needs to be a different formfactor developed. "We need either much higher energy density materials, or change the way the battery is constructed, to make it cheaper. Or both." Going the materials route with things like advanced chemicals will be more expensive initially; end-users will pay more for 20 percent improvement, and companies will charge that premium as long as they can.

Once these advancements are proven out initially at the smaller scales (e.g. consumer electronics, and automotive applications) then the technology will make greater inroads into grid-scale applications, which means more targeted development and competition to improve products and get prices down. "As the grid becomes more compelling as a business opportunity, people will make different batteries to serve that niche," Feaver said. He pointed to two examples already emerging specifically appealing to the grid scale: molten batteries offering very high-density and stability, while flow batteries that "take up a lot of space but are cheap," he said.

Feaver also highlights ultracapacitors, which are getting more attention amid concerns of grid load-leveling and making energy supplies available on a variety of different timetables. He points specifically to Japan — a nation historically more embracing of advanced technologies, and willing to invest in their deployment — where hospitals and industrials plants now use very large ultracapacitor installations (think tractor-trailer sized) for backup power with high reliability and quality. "Most of the time ultracapacitors are too expensive to see in those big deployments," Feaver said, but it's enough to generate a few minutes of power until a backup generator comes online without suffering a blip in power supply (a few minutes), and they'll last "15 years real easily." But that's only part of the portfolio; one might deploy batteries (lead-acid or lithium-ion) for periods of minutes to hours, and some remote locations might expand the battery/ultracapacitor installation to hold for 24 hours, overnight, or several days.

Gradually these technologies are evolving (and costs are lowering) to the point that there's enough energy storage put in to simulate a power plant for a short period of time. And that's really where it can address broad-scale deployment of renewable energy generation, he notes; by definition they overproduce during peak periods, and the key will be harnessing that excess and giving it back to the grid at a later time.

Feaver admits most battery-based grid-scale energy storage is at the concept stage, with pilot projects literally moving tractor trailers around to different areas as needed. Nevertheless, "people are deploying large-scale energy storage with batteries and capacitors," he said. "We're starting to see it grow."

Lead image: Renewable wind energy with battery, via Shutterstock