Squeeze the breeze

New Scientist
29/09/2007 Page: 44

JUST A SHORT drive west out of Des Moines, Iowa, amid fields of corn and soya, there's a dip in Route 44. Here, near the small community of Dallas Center, a short gravel road runs north to a cluster of houses and across the street there's a farm machinery dealer's yard. It seems to be an unremarkable corner of the Midwest, yet almost a kilometre beneath that dip in the road is something that could change the way we use wind energy. If all goes to plan, it could allow the world's most appealing renewable energy source to compete head-to-head with fossil fuels as a way of generating electricity.

My guide for the day is one of the architects of this project, the Iowa Stored Energy Park (ISEP), and he is happy to pull off the road for me to take a few snaps of the gently rolling terrain. Even his name seems to promote the project: Thomas A. Wind. No, really, that's his name. A consultant engineer whose family farm is close to Jefferson, some 50 kilometres away, Tom Wind now leases out his land so that he can devote his time to ISEP and other energy projects. He is a consultant for the Iowa Association of Municipal Utilities (IAMU), a consortium of more than 600 utility companies from across the state.

IAMU plans to transform a sandstone aquifer beneath Route 44 into a giant battery for storing energy from the wind. At night, when wind turbines produce power nobody needs, the electricity will be used to compress air and pump it into the aquifer, creating a huge pressurised bubble. During the day, when demand for power rises, the compressed air will be piped backed to the surface where its energy will be converted into electricity.

If the project comes to fruition it will be a world first, capable of delivering some 268 megawatts of electricity for 16 hours each day. That's enough energy to satisfy the needs of about 75,000 homes. The technology aims to tackle the big complaint that wind energy always faces: the wind doesn't necessarily blow when you want it to. With compressed air storage, it will be possible to store power from Iowa's growing wind generation capacity and then turn it on and off like water from a reservoir, available to customers when needed - and when they are prepared to pay the highest price for it. A power source that the energy industry has till now viewed as fickle will become firm and reliable.

The energy park project grew out of a study Tom Wind conducted for IAMU in 2002 to assess what sort of generating capacity the utilities would need to serve future demand. The study found they would need more power to fill in between daily peaks and valleys in usage - so-called "intermediate load". "There's a certain amount of power you need 24 hours a day, 365 days a year just to keep everything running," Kent Holst tells me. He is development director for the Iowa Stored Energy Park Agency, which has the task of managing the project and raising the funds needed to make it happen.

In Iowa, as in most places in the US, coal plants supply the 24-hour-a-day "baseload" power. To maximise the efficiency of these plants, utilities try to keep them running at a constant rate. When demand peaks - as it does on hot days when everybody flips on their air conditioning - many utilities have to fire up expensive diesel generators. That happens for about 200 hours a year. "In between," Hoist says, "something has to meet a variable load for about 1000 hours a year."

The usual strategy is to hold "spinning reserve" power for that intermediate load some form of generation such as gas turbines that can be cycled up or down quickly to meet spikes in demand. But what form of generation to use? IAMU and Tom Wind realised that what Iowa would have more and more of is wind turbines. Driven by a combination of federal tax breaks, concern over global warming and favourable wind conditions, wind farms are sprouting all over the state. But wind energy is notoriously unreliable, and as a result the industry operates on a rule of thumb that you shouldn't have more than about 20 per cent of generating capacity as wind. Something else generally coal and natural gas - has to fill the gaps when the wind isn't blowing. Install too much wind capacity and problems can arise. If the wind drops unexpectedly, for example, energy output can fall rapidly and the grid must be able to compensate for any variations in, say, voltage or frequency that this causes.

The way round this problem, Tom Wind's report argued, is a technique known as compressed air energy storage (CAES). "We figured we were going to end up with a lot of wind energy in Iowa, so we thought we would be needing something like this to use wind energy more effectively." In the jargon of the power industry, CAES makes wind "dispatchable". IAMU would use as much wind energy as possible at night to compress air, store it underground, and then tap it during the day to meet fluctuating demand. "What storage really does is let you use more wind than you could otherwise," Tom Wind says. What's more, CAES can be used to store cheap off-peak electricity from any source and sell it on the market for a higher price when demand rises.

In principle, it ought to be possible to use the compressed air to spin an electric generator directly, but in practice that is not the most efficient way of exploiting it. To make the most of the stored energy, the energy park will install two 134-megawatt gas turbines adapted from conventional units used in gas-fired power stations. In a conventional gas turbine, compressor fans squeeze air into the combustion chamber at high pressure, where fuel is burned to produce hot exhaust gases that spin a set of turbine blades at high speed. The turbine in turn drives the electric generator, and also the compressor that squeezes air into the combustion chamber.

Though the compressor typically consumes between a half and two-thirds of the power available from the turbine, the high-pressure environment makes the unit more efficient overall. In the CAES plant, the compressed air from the underground store creates high pressure in the combustion chamber without the need for a power sapping compressor. As a result, the turbine generates two to three times as much power from a given amount of fuel. Although CAES is not widely used, two large plants have between them built up decades of operating experience. The first came on stream in 1978 in Huntorf, Germany.

The 290-megawatt plant stores compressed air in two deep salt caverns. Eight hours of compressed-air "charge" is enough to run the generators at full power for 2 hours. The second plant, in McIntosh, Alabama, was commissioned in 1991 by the Alabama Electric Cooperative. It stores its compressed air in a mined-out salt dome 8o metres across and 300 metres tall, lying 450 metres below ground, and can use the air to supply a turbine generating no megawatts of electrical power continuously for some 26 hours.

Giant Bubble
At the Iowa plant, the compressed air will be stored in a porous sandstone aquifer rather than a cavern. This has the advantage that the pressure of the stored air is kept constant, regardless of whether the reservoir is full or almost empty. As air is pumped into the aquifer it displaces water around it, and because this doesn't change the hydrostatic pressure of the water the pressure of the air remains constant too. "You can optimise your equipment for better efficiency if you have a constant pressure," Tom Wind says. There is a downside to using an aquifer, though: the porous water-bearing rock needs to be deep enough underground to provide the pressure needed to run a turbine, and be contained by a dome-shaped cap rock that retains the bubble of compressed air.

Fortunately for the ISEP team, these conditions are identical to those needed for storing natural gas. Northern Natural Gas and other utility companies have made detailed maps of Iowa's geology that have allowed the search to be narrowed to three candidate formations. Even so, the ISEP agency has had to invest hundreds of thousands of dollars - provided by power companies and the US Department of Energy - to narrow down the potential sites. Two proved unsuitable because they lacked a containing cap over the water-bearing sandstone. Preliminary seismic back surveys of the third one - the aquifer below Dallas Center - look good, says Hoist. In early 2007, the team got confirmation that the site is large enough and deep enough to be useful. It is also capped by a suitable rock structure.

Computer models based on the porosity of the rock show that 13 boreholes into the aquifer should be enough to get the compressed air in and out fast enough. The next step will be to make sure the aquifer doesn't contain minerals such as pyrites that could combine with oxygen in the stored air, and so inhibit combustion in the gas turbines. As with so many renewable energy projects, funding remains the sticking point.

A huge 2700-megawatt CAES project proposed in Norton, Ohio - using an abandoned limestone mine as the air storage reservoir - has already stalled for lack of finance. ISEP has a $200,000 government grant to keep the project moving, but failed to get a grant from the 2007 federal budget to help cover the $1.5 million funding it needs to complete the study of the aquifer.

After that it will need $200 million to build the plant. The initial design studies are planned for early 2008. Current plans call for 268 megawatts of generating capacity drawing on power from new wind generators rated at -ILA j 75 megawatts. The wind turbines do not have to be on site. The electricity they generate can be imported over the grid from anywhere in Iowa or beyond, and ISEP will also buy cheap off-peak power from non-wind sources. The team suggest that for every megawatt-hour of wind energy used to compress air and store it underground, about 850 kilowatt-hours are recovered when the air is used to operate the turbines. They say ISEP could deliver electricity to consumers for about 4.5 cents per kilowatt-hour - about the price of electricity from conventional power stations.

Though there are several other CAES projects under consideration across the US, ISEP remains a one-off: no one else is contemplating storing wind energy in this way. Though Tom Wind remains enthusiastic about the project, even he admits that going it alone can sometimes lead to doubts. "We ask ourselves all the time: if this is such a good thing, how come nobody else is doing it?" he confesses. "You start to wonder when nobody is lining up behind you." Perhaps it is simply a case of nobody wanting to be first to take the plunge, as a number of recent analyses suggest that wind farms combined with CAES should compete favourably with conventional energy generation systems.

One report on the potential for wind energy with CAES in Texas, Oklahoma and New Mexico calculates that the operational costs of a CAES plant in the region could be less than that of a conventional gas or coal-fired unit. The economics could be even more attractive in future if the government starts to tax carbon emissions. "That's where our project starts to shine," Tom Wind says.

An analysis published in Energy Policy (vol 35, P 1474), suggests that if emitting greenhouse gases is made costly enough, wind energy combined with CAES will become an economical way to supply baseload power to the grid. It estimates that this combination could provide over 8o percent of the energy on the grid, while cutting greenhouse gas emissions by three-quarters compared with typical gas-fired power stations.

Aquifer storage might not be a suitable solution everywhere. In densely populated areas, where the need for energy is greatest, the water stored in aquifers is a precious resource in its own right, especially in the American Midwest. Fortunately, this is not a factor for ISEP because it is using an aquifer that is not required for drinking or irrigation, at least for now. However, Hoist says there could be problems elsewhere if the CAES portion of an aquifer were placed close to existing wells. When air is stored or removed, it could affect flow in the wells. Faced with plans to use local aquifers for a CAES scheme, communities might end up having to weigh their demand for power against the need to safeguard water supplies.

Before I leave Dallas Center, Tom Wind drives me north to a line of seven wind turbines that have recently gone up near his hometown of Jefferson. Standing more than loo metres tall from the ground to the tip of the blades, they are visible across the fields from a full to kilometres away. He then points to a concrete cistern next to an old farmhouse.

Farmers, he tells me, used small windmills initially to pump up groundwater and, more recently, to power radios. They stored the water in cisterns for use in times when the wind didn't blow, and charged lead-acid batteries to keep their radios running. So ISEP, he says, is just a modern take on what farmers have been doing for more than a century. This time round, perhaps, the idea could provide the gateway to clean energy for all.


Mega-batteries
Off-peak electricity is cheap and plentiful, so it pays to store it for use when supplies are scarce and it can be sold at a good price. There are several tried and tested ways of doing this:

• Pumped Hydro Storage: Off-peak electricity is used to pump water into reservoirs. When demand peaks the water drives generator turbines. A single site can store gigawatt-hours of energy.
• Flow Batteries: Electricity is stored as chemical energy in solutions held in large tanks. The technology is scalable and can store more than 100 megawatt-hours of energy at a single site.
• Flywheel Energy Storage: Using electric motors to spin up a flywheel to as much as 80,000 rpm can store up to 150 kilowatt-hours as kinetic energy.
• Superconducting Magnetic Energy Storage: Energy is stored as a magnetic field, generated by large currents circulating in a superconducting coil. Superconductors need to be held at low temperatures so the technology remains expensive to build. Less than 100 megawatt-hours of SMES storage is installed worldwide.
• Hydrogen: Electricity is used to split water molecules to produce hydrogen, which is then burned to generate electricity when needed. Projects in the UK and on Prince Edward Island in Canada will use wind turbines to generate Hydrogen, which will then be stored in tanks for use as fuel.