A demo of 1000-hour thermal energy storage in depleted oil wells is one of the breakthrough new climate technologies to have received funding from the US Department of Energy (with $6 million) this year.
This thermal energy storage, GeoTES (Geologic Thermal Energy Storage), would store concentrated solar heat for very long durations – able to supply 40 consecutive 24-hour days or 80 consecutive nights at any one time, and even while simultaneously charging with daily sunshine.
A National Renewable Energy Laboratory (NREL) paper (Using Concentrating Solar Power to Create a Geological Thermal Energy Reservoir for Seasonal Storage and Flexible Power Plant Operation) suggests that the Levelized Cost of Electricity (LCOE) of Geologic Thermal Energy Storage could be as low as 6 cents/kWh and the LCOS; Levelized Cost of Storage could be as low as 1 cents/kWh
Next year, at a five-acre test site near Bakersfield, California, parabolic trough solar collectors will gather the sun’s heat daily and accumulate it in a depleted oil reservoir underground, creating a closed-loop geologic form of energy storage.
When discharging the storage, the GeoTES system will use flash steam from stored pressured hot water to drive a steam turbine (or Organic Rankine Cycle, alternatively) to deliver electricity to the grid like a geothermal energy system. This would make GeoTES a hybrid of concentrated solar thermal with the power cycle of a geothermal system, enabling a far greater volume and duration of thermal energy storage than a typical CSP plant.
In today’s CSP plants, theat from the solar collectors is stored in a thermal energy storage tank, but for only up-to-24-hour charge and discharge cycles. By contrast, GeoTES would store this heat in a multi-acre underground sandstone reservoir, enabling charging and discharging over 1000-hours – or longer.
Water is not good at holding heat for a long time. However, these depleted oil wells are not open caves filled with water. They are actually primarily sandstone, a porous sponge-like rock. Hot water delivers the heat into the sandstone, which stores the heat.
“We’re circulating hot water through the system to exchange heat with the porous, permeable geologic formation of sandstone, so the whole formation gets hot,” said Mike Umbro, the business director at PRM, the startup developing the commercial technology.
“So we’re heating all this sandstone with the water in a closed loop, and that huge volume of the sandstone formation is how it allows us to have 1,000 hours of storage capacity.”
PRM was an oil developer, originally. But in 2017, the CEO re-birthed the company as a GeoTES developer, based on patents held by their technology partner Ramsgate Engineering, to enable long-duration thermal energy storage there instead. They have secured locations and leases amongst formerly prolific oil fields with 130 years of development in the southern San Joaquin. All its investors are California-based, in Kern County.
“I met the folks at NREL in 2021 when I attended Geothermal Rising,” Umbro said.
“NREL’s Amanda Kolker put me in touch with Guangdong because he’s an expert in concentrated solar thermal. We – and I think NREL, too – are so excited about GeoTES because you can operate this system for as long as you have the setup there.”
In collaboration with PRM, NREL researchers analyzed two potential regions for GeoTES, involving 568 depleted oil fields in central California and 198 in Texas, to provide technical insights and promote the adoption of GeoTES in future energy markets. (For the Texas projects, the water would be heated by electricity, like in an electric kettle, instead of concentrated solar thermal.)
NREL’s suitability analysis is described in two papers, Using Concentrating Solar Power to Create a Geological Thermal Energy Reservoir for Seasonal Storage and Flexible Power Plant Operation and Geological Thermal Energy Storage (GeoTES) Charged with Solar Thermal Technology Using Depleted Oil:Gas Reservoirs and Carnot-Battery Technique Using Shallow Reservoirs
The researchers analyzed potential sites using data on porosity, permeability, thermal conductivity, temperature, pressure, mineralogy, depth, thickness, brine salinity, and productive area. They calculated the solar field size, potential hourly energy generation, and when it is best to store or generate energy. They determined that shallow reservoirs would be most efficient and profitable due to pumping costs (the only significant cost in operation).
The challenges of investigating nature-based thermal energy storage come from its being a natural system, such as predicting the reservoir boundaries. But so do the benefits: there are no costs to build containers that already exist. Even so, the initial build cost would exceed that of CSP with storage as well as short-term batteries. However, the long lifecycle once built and the climate and grid benefits of much longer seasonal storage of 100% solar energy balance this out.
“In a seasonal weather event where it’s cloudy for 60 days straight, we can produce power for 1000 hours because we’re just extracting the heat from the reservoir. And though we would be depleting our battery, we’re also charging it most of the year,” said Umbro.
“That’s something we’re going to need to find out, and the question, for a given reservoir, you identify the allowable amount that it’s elevated,” said NREL lead researcher Guangdong Zhu.
“Though nature can handle high-temperature water or steam, we’re shooting for 200°C because we worry about too much higher temperature impacting the ground. Geothermal can be up to 400°C, but most ORC power plants are running between 150°C to 250°C. They typically have two different cycles. One is the flash steam cycle, which uses steam to drive the turbine; that’s just a lower-temperature part of all the safety power blocks. The second cycle for geothermal is the Organic Rankine Cycle – ORC cycle. Instead of using steam, you can produce power using other organic fluids like this.”
Zhu dismissed the efficiency disadvantage of low-temperature geothermal power cycle ORC.“It’s not about power cycle efficiency,” he said.
“The number we’re looking into is – will it be economically viable or not viable? The key thing we’re talking about here is overall energy cost. If you want to look into the storage cost, you should look into energy storage cost. The storage cost only has a huge advantage when storing the power over a long duration.”
PRM has invested over $3 million in documentation to permit 560 acres in California for GeoTES. This particular region of old abandoned oil wells has not been operational for four years, so they’ve been working on permits to re-inject this cleaned water out of the same reservoir it came from.
“There’s all sorts of things related to the environmental work that you need to do,” Umbro noted.
“Drilling permits, injection permits, environmental surveys, surface flora, fauna, air, impact, you name it. It’s very onerous to permit any project in California. It’s necessary to meet NEPA on the federal side and the California Environmental Quality Act, CEQA at the state level, which is extremely stringent. The salinity is around 16,000 total dissolved solids, parts per billion. So that’s like almost ocean water salinity. It is not water that anyone could ever consume. So you’re not taking a potential source for agriculture or potable water. It’s too brackish.”
Next year, PRM will complete the planning and start constructing a 30,000 MWh demonstration project, about the area of four football fields.
“We’ll use five acres of surface for the technology demonstration,” said Umbro.
“Our demonstration is just two acres of solar trough collectors, plus eight wells on half an acre, and some facilities. So, at most, five acres. We have a 560-acre lease position, so that’s just 1%. But you have to build the whole system to prove it, so there’s no real way to do it on a much smaller scale than that. We were awarded a $6 million grant through the Department of Energy from SETO, the Solar Energy Technologies Office. We’ll match that with $12 million of our capital. We have existing investors, all in Kern County and Bakersfield, and we’ll raise a little bit of capital.”
