Long Term Electricity Storage
Silicon Valley veterans view Sacramento’s obsession with renewables mandates with pragmatic detachment. Blessed with disposable income sufficient to make them indifferent to the price of gasoline or electricity, they view life on the bleeding edge as an opportunity for California to lead the world into the electric age. They’re not wrong. Heartless, perhaps. But not wrong.
If a more appropriate balance can be struck between a few thousand strategic investors using our state as a petri dish from which to birth an electric future — and the less glorified but more compelling aspirations of tens of millions of state residents who are just tired of politically enforced downward mobility — we might set an example of broad based prosperity along with bleeding edge innovation. Imagine decentralized, private, rooftop photovoltaics, with utility scale storage decentralized and privatized as well, thanks to millions of EV owners selling electricity using vehicle-to-grid home hookups. No more duck curve. Abundant baseload power. Affordable utility bills.
There is nothing wild or crazy about that scenario. Solar panels are cheaper every year, robotic recycling is on the way, and battery technology for both stationary and vehicle applications is becoming less resource intensive, more practical, and more affordable. Even without subsidies, writing off the competitive potential of photovoltaics and batteries is probably shortsighted.
But a solar energy solution still fails in one critical area. Even in subtropical California, the short days in winter result in a photovoltaic yield that averages 16 percent of panel capacity, versus an average yield of 33 percent in the summer months. For photovoltaic and battery systems to generate sufficient electricity from our winter sun, they need to be built to literally twice what would be enough to power the state during the summer.
This calls for long-term electricity storage, and the quantities required are daunting. Let’s assume the California Energy Commission’s stated goal of 500,000 gigawatt-hours per year of electricity production (in 2023 we consumed 281,140 GWH, producing 215,623 GWH in-state and importing the rest) is achieved exclusively with photovoltaics. That would equate to an average output of 57 gigawatts. With an average year-round yield of 25 percent, that would require a photovoltaic array generating 228 gigawatts in full sun.
Here’s where it gets fun. For the sake of argument, let’s assume 20 watts per square foot of PV and 90 percent space utilization. That’s 500 megawatts of output per square mile in full sun, or 1,000 gigawatt-hours per square mile per year. That is not an unrealistic projection when taking into account ongoing advances in PV and inverter efficiencies and may ultimately be too low. But using these assumptions, with a footprint of 500 square miles – preferably privately financed on rooftops – photovoltaics can generate 500,000 gigawatt-hours of electricity per year, and they can use millions of connected EV batteries to balance daily fluctuation.
The problem is that this generation is seasonally uneven. To be grossly simplistic but nonetheless to accurately determine the basic scale of this challenge, assume photovoltaics deliver a 16 percent yield in winter, 33 percent in summer, and 25 percent in spring and fall. That means that in the spring and fall, production of 125,000 gigawatt-hours is equal to demand, but in summer a stupendous 165,000 gigawatt-hours are generated, and in winter only 80,000 gigawatt-hours are generated. So how do we save 40,000 gigawatt-hours from summer, to be discharged in winter?
This is the quandary facing photovoltaic power. Batteries will balance night and day. But can they balance summer and winter?
One intriguing solution to this quandary is synthetic geothermal power. A company pioneering this technology, Premier Resource Management (PRM), aims to construct a pilot project in Kern County over the next few years. The concept rests on the potential of underground formations of porous rock to retain vast quantities of heat for extended periods of time. Thus it is possible to charge the rocks with heat during the summer and harvest it in the winter to generate electricity. Because Kern County’s oil industry has already evaluated the underground strata in order to recover oil, there are hundreds of known sites where the technology can be developed.
As a video produced by PRM’s technology partner, Ramsgate Engineering, LLC, explains, the project calls for a single-axis parabolic trough array to heat water in a closed loop to avoid introducing contaminants. Then in a separate closed loop, water is pumped out of the underground formation, screened for contaminants, then passed through a heat exchanger that is heated by the water circulating in the parabolic trough system. The heated water is pumped back into the underground formation, slowly building its temperature up to over 400 degrees. Then, as needed, the water circulating in and out of the underground formation is itself redirected through another heat exchanger which is used to boil water that passes through a third independent loop. This boiling water is used to drive a steam turbine to generate electricity.
With apologies to the PRM team, that’s an oversimplified explanation of a complicated design (watch the video, lingering on the site diagram at 3:40), but this concept offers something batteries cannot. A system that can store gigawatt-hours of energy for months at a time. PRM estimates a 2,000 acre parabolic trough array, positioned over an underground formation with a volume roughly equivalent to a regular cube 100 meters on a side, would have the potential to discharge 400 megawatts continuously for up to 1,000 hours. The project is expected to have an operating life of about 100 years.
To put 400,000 megawatt-hours (400 GWH) into perspective, it means that you would need 100 of these installations (40,000 GWH required / 400 GWH per system) to balance summer and winter in an all-PV, 500,000 GWH per year California. But the estimated potential of this system to also deliver night-time electricity year-round more than doubles its estimated annual power storage capacity, to about 1,000 gigawatt-hours per year. Do the economics work?
At an estimated cost of $2.0 billion, financed at 4 percent, 30 year terms, the annual financing repayment would be $115 million, equating to $0.12 per kilowatt-hour. The ultimate break-even price per kilowatt-hour would have to increase, of course, to cover operating costs and profit, and the annual loan repayments could be lowered to the extent financing is via equity or other non-interest-bearing instruments. Total construction costs could drop as more plants are built. And even if daily fluctuations in the price for electricity are eliminated with batteries, winter power may still sell at a premium that lifts the business to profitability. Moreover, there is potential to recover additional oil from these wells with minimal additions to the plant investment.
We can embrace that hybrid approach that makes long-term electricity storage using synthetic geothermal power more attractive economically, or we can see this promising technology migrate to Texas, a state equally blessed with an abundance of depleted, well mapped underground formations. Perhaps our Silicon Valley innovators can commercialize a chemical process that extracts energy from oil without emissions, or they can help overhaul the refining technologies that give us products made from oil derivatives. Or they might just recognize that until we aren’t importing a single drop of crude oil from overseas, we may as well produce it here.
Edward Ring is the Director of Water and Energy Policy at the California Policy Center, which he co-founded in 2013. Ring is the author of Fixing California: Abundance, Pragmatism, Optimism (2021) and The Abundance Choice: Our Fight for More Water in California(2022).