by Julian Spector
June 27, 2017

When people fight about deep decarbonization, they’re really fighting about how to store surplus renewable energy.

Last week saw plenty of fighting over the right way to cut carbon from the grid, as a 21-scientist cohort published a stinging, peer-reviewed rebuttal to the widely circulated 100 percent renewable roadmap put forth by Stanford Professor Mark Jacobson.

My account of that intellectual conflict is here, but for this week’s Storage+ I wanted to dig into how energy storage plays the role of kingmaker in the jockeying to design a low-carbon grid.

Jacobson’s vision has wind and solar shouldering most of the work in powering the grid, with excess production going into several forms of thermal storage, as well as pumped hydro and hydrogen production. When generation and stored energy can’t meet the grid’s needs, Jacobson's model calls up hydropower (at an instantaneous capacity 15 times what the U.S. currently musters).

You may notice that lithium-ion did not appear in that sketch, despite its outright domination in storage deployments today. That’s not an oversight; Jacobson steered clear of lithium-ion because it is expensive, and he was searching for a “low-cost” solution.

In its stead, he and his coauthors rely heavily on underground thermal storage, which captures heat from the sun and pumps it deep underground for safekeeping until the buildings aboveground get cold and need that heat back.

This has been practiced for more than a decade at Drake Landing, a somewhat obscure green housing development in Canada that has nonetheless captured the imagination of 100 percent renewables advocates. That community powered all of its heating needs for the 2015-2016 winter from the underground storage system.

The 21 authors argue that this single demonstration does not provide enough proof of concept to justify Jacobson’s proposal to install it for any building in the U.S. that uses heating. Jacobson asserts that the technology works for the role he assigns it; the barrier to wide-scale deployment has more to do with motivation than science.

If underground storage works so well, though, why isn’t it everywhere?

I’ll emphasize that just because the UTES market does not exist now doesn’t mean it can’t exist sometime in the future. Any storage evangelist worth their weight in value stacks knows that the ability to perform a useful grid function does not guarantee real-world deployment.

It takes time even for proven and widely demonstrated storage technologies to work their way into the utility planning process. The people responsible for keeping the lights on need to feel secure that the newfangled tool won’t jeopardize the system that’s been working a certain way for years.

UTES has to cross an additional hurdle, because it sits on and below customer property. It’s one thing to buy an Ice Cub when you're up for a replacement air-conditioner; UTES requires carving up your property to bury a mechanical apparatus deep underground. (And how well has that worked for geothermal heat pumps so far?) Not only that, but it works best as a form of district heating serving multiple homes, which requires community buy-in.

That makes the proposition more attractive for new builds like Drake Landing, rather than retrofits. The same holds true for battery storage paired with rooftop solar; just ask Sonnen or Tesla.

When drafting a system of high renewable penetration, success hangs on the methods chosen for storing surplus generation. The renewables part of the equation grabs the headlines, but it's storage that makes the whole thing hum.

When evaluating the prospects for Jacobson’s scenario or any of the other decarbonization plans, test the believability of the storage component.

Taking Jacobson’s plan at its word would require abandoning the storage industry as we know it, and pivoting to a handful of technologies that have seen vanishingly little real-world usage. States would have to birth a supply chain out of sheer will, like Athena emerging ready for battle from the head of Zeus.

Then again, sole reliance on lithium-ion storage wouldn't work.

"Once you are talking 100 percent renewables, the type of storage we need to get there is very different from lithium-ion," said Ravi Manghani, energy storage director at GTM Research. "You're essentially talking about much longer-duration storage assets that can shift your energy for days, weeks, months or seasons."

Under the right circumstances, good technology can achieve surprisingly swift adoption. Any policymaker considering a 100 percent renewable plan should make sure there’s a viable mechanism for delivering the core storage technologies it relies on.

The (state) house doesn’t always win

A few weeks ago I reported that Nevada had surged to the forefront of storage policy formation thanks to a prolific legislative season. That process hit a snag when the governor vetoed the major renewable portfolio standard update.

The bill would have raised the RPS to 40 percent and included energy storage, with a 2X modifier if it delivered clean energy at peak times or helped integrate renewable generation.

That storage bit has never been done and took even ardent storage fans by surprise. It’s not actually clear if it would have succeeded in spurring more storage adoption; that depends on whether the companies covered by the RPS could meet their obligation more cheaply with solar or wind, or if the double-counting makes storage cost-competitive.

Now we won’t have a chance to find out.

Instead, stay tuned for the Nevada PUC’s investigation of the cost-effectiveness of storage for ratepayers. Governor Brian Sandoval did sign that bill, which gives the PUC the power to set a mandate if the cost-benefit analysis turns out positive.

West side storage

New York just passed a storage target too. Still awaiting a signature from pro-renewable Governor Andrew Cuomo, the bill calls on the Public Service Commission to set a 2030 target for energy storage systems by January 1.

After that, the New York State Energy Research and Development Authority and the Long Island Power Authority will start implementing a deployment program to meet the target. The program must consider both customer-sited and front-of-the-meter storage, evaluating its use for transmission upgrade deferral and peak load reduction in constrained areas.

This puts New York in the running to be the fourth state to set a storage target. California and Oregon have them; Massachusetts owes its legislature a target by July; and Nevada’s deadline isn’t until October 2018.

Australia goes giga-scale

Two giga-scale lithium-ion storage factories could be coming to Australia in the next few years.

The business case for manufacturing there is driven by the country’s healthy storage market, proximity to Asian markets, and the domestic mining of critical ingredients like lithium, cobalt and graphite, Renew Economy has reported.

The upstart manufacturing consortium Energy Renaissance includes Massachusetts-based 24M, which claims it can cut battery costs in half through a faster, more efficient production process. 24M will have a chance to prove its claims when the facility begins production in Darwin, which is expected to occur in late 2018.

The factory will start with four 150-megawatt-hour production lines, which are expected to grow to seven in total over time.

Meanwhile, former Macquarie Bank executive Bill Moss is working to open a 15-gigawatt-hour storage factory in Townsville.


Automakers want to sell more electric vehicles, and utilities want to sell more power. They have a lot to gain from working together on this endeavor.

BMW and PG&E showed what that friendship might look like in their i ChargeForward pilot, which saw 100 customer-owned i3 cars and a stationary storage reserve at BMW’s Mountain View office deliver 100 kilowatts to the grid for an hour when called upon.

BMW turned car charging into a demand response event, delaying customer charging by an hour if needed, and discharging its stationary second-life battery system to make up the difference. This is a key first step toward treating EVs like full-fledged grid assets.

Julia Pyper has the details here, but a few points jump out:

  • The cars themselves averaged 20 percent of the kilowatts contributed over 209 demand response events. On average, only seven vehicles participated in a given event. That means the stationary storage is still doing the heavy lifting, which makes this look more like a test for grid storage as a DR asset.
  • That said, the pilot did verify that real consumer vehicles can receive and respond to utility signals in real time.
  • Vehicle response rate was correlated to PG&E’s time-of-use rates. When a DR event fell between the optimal charging hours of 11 p.m. and 2 a.m., more cars participated. That means owners are responding to the financial incentives the utility has employed to make charging behavior work for the grid.
  • 98 percent of participants were satisfied with the program, so the grid benefit did not come at a cost to customer experience.

Now BMW is expanding to a 250-vehicle demonstration, which will focus on how to make smart charging useful for the grid and how to get customers to participate in such events.