Compressed-air energy storage, or CAES, is one of the cheapest ways to store really massive amounts of energy for long periods of time. In fact, beyond pumped hydro storage, compressing air when power’s cheap and plentiful and then using it to boost natural gas-fired power turbines during times of peak demand is really one of the only ways to shift hundreds of megawatts of load from one hour to the next.

So why have so few CAES-backed power plants been built? One problem is cost: while they’re much cheaper and easier to site than pumped hydro projects, CAES projects are still massive infrastructure projects at the $100-million-and-up scale.

They also have their geographic restrictions: specifically, the availability of vast underground cavern structures to use as reservoirs for all that compressed air. Then, of course, there are regulatory and economic factors that have limited energy storage’s appeal in general, which apply to CAES as well.

Put caves and power plants together, however, and the energy storage that results is cost-effective and reliable, as the world’s two first CAES projects -- the 290-megawatt plant in Huntorf, Germany, built in 1978, and the 110-megawatt McIntosh, Alabama plant, built in 1991 -- have proven over their combined decades of operation. That’s led to a resurgence of CAES projects over the past half-decade or so, driven by technology advances, as well as grid operators’ growing need for energy storage to balance intermittent wind and solar power with ever-shifting demand.

The latest comes in Texas, where Dresser-Rand and Apex Compressed Air Energy Storage announced last week that they’re building the first big CAES project in the United States in decades. Known as the Bethel Energy Center, the 317-megawatt project will serve Texas grid operator ERCOT, and is the first of more to come from the partners, both in the United States and elsewhere, the companies said.

The project is being built near Tennessee Colony, Texas, a rural crossroads about 100 miles southeast of Dallas featuring a church, a general store, and a Calpine natural gas facility. That part of the state has its share of giant underground “salt dome” caverns, suitable for storing natural gas, as well as pumped air.

Dresser-Rand, the Houston-based supplier of “high-speed rotating equipment and services solutions” for the oil and gas industries, is providing the equipment for the project at an estimated cost of $200 million. Dresser-Rand also built the equipment for the Mcintosh, Ala. CAES plant in 1991. Since then, it’s developed an integrated energy storage technology, called SMARTCAES, that it’s using in the Bethel project.

There are many different ways to compress air and then use it to provide energy, but in the simplest terms, most CAES projects use it to boost natural gas-fired power turbines. That essentially ties the storage capacity to helping the turbines do whatever they’re doing, only at higher efficiency, and without burning nearly as much natural gas to get there. In other words, it can’t be turned on and off like a battery.

At the same time, compressed air is a lot cheaper energy storage medium than the electrochemical mixtures contained inside batteries. At $200 million for 317 megawatts, the Apex/Dresser-Rand project adds up to about $630 per kilowatt, or cheaper than any battery technology out there except lead-acid batteries, which have been used for decades on the grid but are short-lived and expensive to maintain. As for scale, the biggest lithium-ion battery grid storage projects out there (AES Energy Storage, BYD) are no bigger than 40 megawatts, a little more than one-tenth the size of the Bethel project.

Apex and Dresser-Rand didn’t provide details on how many megawatt-hours of storage the plant would provide, or any other measures of duration for the 317 megawatts of capacity they’re planning. Of course, CAES is an inherent part of a natural gas-fired power plant’s overall efficiency, which makes it hard to measure against other forms of storage, like pumped hydro or batteries. That efficiency can also come in multiple ways, such as boosting top running efficiency, or helping turbines ramp up or ramp down to match wind and solar power fluctuations -- as well as in terms of how much natural gas it’s saving over time.

Texas has a lot of caves, depleted oil wells and other convenient underground empty spaces to fill with compressed gas, and it’s also got a lot of wind power that it’s struggling to integrate more smoothly into its statewide grid. Both of those features make it a good candidate for CAES projects. General Compression, an ARPA-E-backed startup that’s working on a fuel-free CAES technology, announced in April that it’s working with ConocoPhillips on a CAES pilot project, also in Texas, in hopes of expanding to commercial scale in the years to come. 

In California, Pacific Gas & Electric is studying the idea of building its own 300-megawatt, 10-hour CAES project, meant to use wind power to compress air at night when it’s cheap and plentiful, then release it to manage peak afternoon loads. The project costs about $50 million, half of it provided by a Department of Energy smart grid stimulus grant, which means results of its performance will be made public to guide future projects. That site would use porous rock, rather than caverns, to store air, an as-yet unproven technology.

In the meantime, some startups are trying to get rid of the need for conveniently located caves, porous rock formations and other geological anomalies to store air. LightSail Energy, a grid-scale energy storage startup, is working on CAES via aboveground tanks, and is backed by investors including French oil giant Total, Peter Thiel, Bill Gates and Khosla Ventures.

Dresser-Rand has made its own investments in cutting-edge energy storage and recovery technologies. In 2011, for example, it invested $10 million in Echogen Power Systems to take a stake in the Akron, Ohio startup, formerly known as reXorce Thermionics, working on supercritical carbon dioxide (ScCO2)-based thermodynamic waste heat recovery systems.