As the battle against excess atmospheric CO₂ soldiers on, the leading alternative technologies are wind- and sun-powered. As wonderful as they are, their intermittent character forces us to re-think how the electrical grid needs to operate. Existing fossil fuel plants are the dominant source that can run continuously to fill the gap — at least for now. But unless we can reliably replace that technology, we still have the CO₂ problem to deal with. Most everyone agrees that we need a way of storing energy during the off-peak energy generation times.

To understand the scope of the issue let’s just focus on the U.S. consumption profile. The average electrical consumption in the U.S. is 500 million GWatts (G = giga = 10⁹ ) continuously. About 20% of that is covered by hydroelectric and nuclear, which are free of carbon emissions, leaving 400 million GW for other forms of power. If that additional power was supplied by intermittent sources such as wind and solar, it is estimated that about 20% storage is sufficient to cover backup situations when neither wind nor sun is available.

Importantly, this requires a nationwide smart grid to ensure that everybody has access to that storage. Finishing up the math, the backup requirement would need to be 80 million GW continuous. Power sources like batteries are rated not only by their rate of delivery (watts) but how long that delivery can be sustained (watt-hours/Wh). In a worst-case scenario perhaps 100 hours (four days) of power would be required, so 8,000 GWh connected to a national grid would be enough to cover that situation. For illustration purposes, based on standard Powerwall specifications: storage = 7 kWh. To meet our theoretical storage requirement through batteries alone it would take 1,142 million Powerwall batteries or about 9.2 per each and every household in the U.S. Clearly an unwieldy and impractical number. Experts agree that what is needed is fewer but larger storage facilities.

This brings us back to the previously mentioned hydroelectric power with a variation known as pumped hydro. The concept is quite simple. At times when the demand for power is low, excess power can be used to pump water from the outlet reservoir up to the storage reservoir. This is an existing and proven solution. There is already about 20 GW of pumped hydropower in the U.S. already, which theoretically gets us one-quarter of the way toward a solution. For comparison, about 2.5% of existing hydropower in the U.S. goes through reservoirs with this capability already. In Japan, it is 10%. Solutions are capital intensive and not all sites are economical to build, but it is the leading storage capacity for now.

The other promising large scale storage requirement just might be solved by storing compressed air in underground reservoirs. This Compressed Air Energy Storage (CAES) works in principle in a way that is analogous to pumped hydro. During peak production, compressed air can be stored in huge natural underground caverns. When that excess supply is needed, that compressed air can be run back through turbines to generate electricity. There are a lot of potential caverns, and layered rock formations that could withstand the 750 psi storage pressure. Of course, not all natural large storage cavities are situated near power distribution stations. Also, at present, it does require a natural gas burner to heat the cavern air for use in power-generating turbines to maximize efficiency. Of course, the initial compressing of air for storage generates heat, and engineers are working on ways to capture that heat and redeploy it during power generation, eliminating the need for burning natural gas.

You can see the pattern for storage solutions emerge from these examples: A large-scale backup technology situated near a substation on a national grid, to deliver the shortfall. And since the intermittent source is subject to the whims of weather, it should be at least 20% more than the primary source. So, what other technologies might fit the bill? It all depends on the intended application.

In the compressed gas category, besides compressed air as mentioned before, there is liquid air, compressed carbon dioxide, and green hydrogen. “Green” in this case means hydrogen separated from oxygen through electrolysis using photovoltaic solar energy. Hydrogen could be used either in a fuel cell or burned with the oxygen in the air and turned back into water. Storing hydrogen can be done using caverns, mines, existing gas pipelines or abandoned oil wells. Hydrogen is a high density fuel and so very highly explosive. After all, it is being used as rocket fuel.

Batteries as currently envisioned and driven to market by the requirements of electric vehicles (EVs) may have a role to play. However, they are burdened by the automotive market requirements and for large scale solutions, some very different battery versions are available or under development. One promising battery type is called a flow battery, which has a very rapid response, and a version is being used in Puerto Rico to smooth out dips in the power curve. There are, in fact, dozens of utility scale battery-based storage facilities around the globe with as much as 400 MWh (100 MW for four hours). Some consideration has been given to grid-connected EVs, but each individual source (one vehicle) would need to be managed so as to preserve battery life and to ensure available capacity. In some cases, companies take cast-off EV batteries that still have a useful life and reconstruct them into grid scale battery banks.

Different problems require different solutions. In places that get lots of heat, powered by the sun, it makes sense to focus that heat to a central location to heat a boiler which turns a steam turbine. These systems use molten salt as the heat transfer fluid. This salt is stored in an underground reservoir to maintain heat for storage. This reservoir can be tapped after the sun goes down, when the demand is lower. This is a viable solution specifically for desert areas.

And then there are some critical solutions that must be seamless, like hospitals and data centers. For fast power on demand, some installations have turned to flywheels. Usually, the rotating mass of the flywheel is kept under a vacuum and uses magnetic bearings to minimize friction. As power is applied, the flywheel gains speed. Once up to speed, it only needs a trickle of current to cover the losses. When power is needed, a circuit is closed and the power is harvested from the flywheel, slowing it down. The advantage is that flywheels are ready to share that stored energy very quickly. This makes them more suitable for load leveling applications and specialty applications like high energy physics experiments that need a huge burst of power for a very short time.

This overview might be enough to trigger your inner mad scientist. Just about any way of making and holding on to some power for use later, that you can dream up, seems to be up for grabs. In addition to the examples here, there are hybrid electrical/chemical systems, ammonia power, and rail cars stored on inclined tracks, attached to generators. Release the brakes and let the power flow! It’s a whole wild, wild west of different power arrangements. There is little doubt that by whittling away at the problem more solutions will be found, funded, and functional in the coming years. Perhaps even by the self-imposed year, 2050, as planned.

About the author

Scott Orlosky has an MS in Manufacturing and Control Theory from the University of California at Berkeley and has worked over 30 years designing, developing, marketing and selling sensors and actuators for industrial and commercial industries. He has written numerous articles and application notes for speed and position sensors used in industrial and hazardous area environments including an author credit in “Encoders for Dummies.” Scott authored an industrial newsletter for nearly 15 years and is also co-inventor on a number of patents involving design and manufacturing of inertial sensors.