February 27, 2008, was a bad day for renewable energy. A cold front moved through West Texas, and the winds died in the evening just as electricity demand was peaking. Generation from wind power in the region rapidly plummeted from 1.7 gigawatts to only 300 megawatts (1 megawatt is enough to power about 250 average-size houses). The sudden loss of electricity supply forced grid operators to cut power to some offices and factories for several hours to prevent statewide blackouts.
By the next day everything was back to normal, but the Texas event highlights a huge, rarely discussed challenge to the adoption of wind and solar power on a large scale. Unlike fossil fuel plants, wind turbines and photovoltaic cells cannot be switched on and off at will: The wind blows when it blows and the sun shines when it shines, regardless of demand. Even though Texas relies on wind for just over 3 percent of its electricity, that is enough to inject uncertainty into the state’s power supplies. The problem is sure to grow more acute as states and utilities press for the expanded use of zero-carbon energy. Wind is the fastest-growing power source in the United States, solar is small but also building rapidly, and California is gearing up to source 20 percent of its power from renewables by 2017.
Experts reckon that when wind power provides a significant portion of the electricity supply (with “significant” defined as about 10 percent of grid capacity), some form of energy storage will be essential to keeping the grid stable. “Without storage, renewables will find it hard to make it big,” says Imre Gyuk, manager of energy systems research at the U.S. Department of Energy.
Fortunately, there is a promising solution on the horizon: an obscure piece of technology known as the vanadium redox flow battery. This unusual battery was invented more than 20 years ago by Maria Skyllas-Kazacos, a tenacious professor of electrochemistry at the University of New South Wales in Sydney, Australia. The vanadium battery has a marvelous advantage over lithium-ion and most other types of batteries. It can absorb and release huge amounts of electricity at the drop of a hat and do so over and over, making it ideal for smoothing out the flow from wind turbines and solar cells.
Skyllas-Kazacos’s invention, in short, could be the thing that saves renewable energy’s bacon.
To the engineers who maintain the electrical grid, one of the greatest virtues of a power supply is predictability, and that is exactly why renewable energy gives them the willies. Nuclear- and fossil fuel–powered plants produce electricity that is, in industry speak, “dispatchable”; that means it can be controlled from second to second to keep the grid balanced, so the amount of energy being put into the wires exactly matches demand. If the grid goes out of balance, power surges can damage transmission lines and equipment. Generators are therefore designed to protect themselves by going off-line if the grid becomes unstable. Sometimes this can amplify a small fluctuation into a cascading disaster, which is what happened in the northeastern United States and eastern Canada in August 2003, plunging 50 million people into a blackout. Unless the reliability of renewable energy sources can be improved, as these sources contribute more and more electricity to the grid, engineers will have an increasingly difficult time keeping the system balanced. This raises the specter of more blackouts, which nobody would tolerate. “We want to make renewables truly dispatchable so we can deliver given amounts of electricity at a given time,” Gyuk says.
The way to make renewables more reliable is to store the excess electricity generated during times of plenty (when there are high winds, for instance, or strong sun) and release it later to match the actual demand. Utilities have been using various storage techniques for decades. Hydroelectric plants, for instance, often draw on reservoirs to generate additional electricity at peak times, and then pump some of the water back uphill in off-peak periods. Compressed air is another, less common form of large-scale energy storage. It can be pumped into underground cavities and tapped later. These technologies have been suggested as ways of storing renewable energy, but both approaches rely on unusual geographical conditions.
“For most of us right now, the real key to effective storage is batteries,” says Jim Kelly, senior vice president of transmission and distribution at Southern California Edison. Specifically, what is needed is a battery that can store enough energy to pull an entire power station through a rough patch, can be charged and discharged over and over, and can release large amounts of electricity at a moment’s notice. Several promising battery technologies are already in early-stage commercialization, but the vanadium battery may have the edge in terms of scalability and economy.
Like the battery in your cell phone or car, vanadium batteries are rechargeable, but chemically and structurally they go their own way. A vanadium battery consists of three main components: a stack where the electricity is generated and two tanks that hold liquid electrolytes. An electrolyte is any substance containing atoms or molecules that have positive or negative electrical charges. These charged atoms or molecules are known as ions, and the amount of charge on an ion is known as its oxidation state. In a battery, electrolytes are used as an energy storage medium. When two electrolytes, each containing ions with different oxidation states, are allowed to exchange charges, the result is an electric current. The technical term for this kind of charge exchange is a redox reaction, which is why the vanadium battery is formally known as the vanadium redox battery.
A traditional battery, such as the familiar AA dry cell, holds electrolytes in its own sealed container. But the vanadium battery is a flow system—that is, liquid electrolytes are pumped from external tanks into the stack, where the electricity-generating redox reaction takes place. Want to store more power? Use bigger tanks. The bigger the tanks, the more energy-rich electrolytes they can store. The downside is that flow batteries tend to be big. It takes a flow battery the size of a refrigerator, incorporating a 160-gallon tank of electrolytes, to store 20,000 *watt-hours of electricity, enough to power a full-size HDTV for about three days. This is because the energy density in the liquid electrolytes is relatively low compared with that of the chemicals in lithium-ion batteries. (Energy density is a measure of the amount of energy that can be extracted from a given volume or mass of a battery.) For this reason, flow batteries are unlikely to be found in mobile applications, like laptops or electric cars. In those cases the battery of choice remains lithium-ion, which has an energy density five times that of vanadium.
For large-scale energy storage, the rules are very different. Typical rechargeable batteries are unsuitable because it is difficult to get a lot of energy out of them quickly; when the grid is on the verge of crashing, you want an energy infusion now. Ordinary rechargeables also wear out easily. A typical laptop battery will die after a few hundred charge-discharge cycles. In contrast, flow batteries can be charged and discharged many thousands of times.
The vanadium battery’s indefatigable nature echoes that of its creator, Skyllas-Kazacos, a single-minded researcher whose no-nonsense manner is frequently punctuated by an unexpected easy laugh. Her path to the vanadium battery began quite by accident in 1978 at Bell Laboratories in Murray Hill, New Jersey, where she was a member of the technical staff. She had applied to work on solar energy. At the time, Bell Labs was developing liquid-junction photovoltaics (a type of solar cell that employs liquid electrolytes), which seemed like a nice fit for her electrochemical training. But the director of the lab’s battery section picked up her job application first and liked what he saw. Much to her surprise, when Skyllas-Kazacos arrived she was assigned to do research on batteries, which she had never worked on before.
Her serendipitous experience in batteries was put to good use five years later after her return to Sydney, where she had grown up after immigrating with her family from Greece in 1954. She took a position at the University of New South Wales. A colleague there asked her to co-supervise a student who wanted to investigate ways of storing solar energy. The project sounded interesting, so she agreed.
Skyllas-Kazacos started her research by building on the foundational work on flow batteries done by NASA in the mid-1970s. The space agency’s scientists recognized that flow batteries could store solar power on a spacecraft, but they gave up on them after hitting a snag known as cross-contamination. When two liquid electrolytes made of different substances are separated by a membrane, sooner or later the membrane is permeated and the two substances mix, rendering the battery useless. The early NASA flow batteries, which used iron and chromium, quickly ran down as a result.
“We thought the way to solve this problem was to find an element that could be used on both sides,” Skyllas-Kazacos says. Technically, cross-contamination would still occur, but with essentially the same substance doing double duty, the problem would be moot. The key was to pick an element that could exist in a variety of electrical, or oxidation, states.
Skyllas-Kazacos chose vanadium, a soft, bright white, relatively abundant metal named for Vanadis, the Scandinavian goddess of beauty and youth. Vanadium has four oxidation states, known as V(+2), V(+3), V(+4), and V(+5); in each state the element carries a different amount of electric charge. Often oxidation states are hard to tell apart, but in this case nature was kind: V(+2) is purple, V(+3) green, V(+4) blue, and V(+5) yellow.
Simply having different oxidation states is not enough to make an element work for a liquid battery. The element has to be soluble, too. NASA had considered and rejected vanadium because the technical literature insisted that the solubility—and hence energy density—of the useful V(+5) form of the element was extremely low. Skyllas-Kazacos recognized, however, that just because something appears in print does not necessarily mean it is true. Previous studies had started by leaving a compound of vanadium, vanadium pentoxide, to dissolve in solution. This was a very slow process that could take days, and it never produced more than a tiny amount of V(+5) in solution. Skyllas-Kazacos approached the problem from a less direct route. “I started off with a highly soluble form, V(+4), then oxidized it up to produce a supersaturated solution of V(+5). I found that I could get much higher concentrations. From then on it became clear that the battery would actually work.”
In 1986 came a major milestone: Her university filed for a patent on the Skyllas-Kazacos vanadium battery. But proving the concept turned out to be the easy part. “We thought we would take the device to a certain level, and then some industry group would come and take it off our hands,” Skyllas-Kazacos says with her laugh. “What we didn’t realize was that the task was enormous. We had to develop the membranes, the conducting plastic for the electrodes, the structures, the materials, the designs, the control systems—everything!” In 1987 Agnew Clough, an Australian vanadium mining company, took out a license on the technology. But nothing came of the deal.
The vanadium battery finally got its first chance to shine in 1991, when Kashima-Kita Electric Power, a Mitsubishi subsidiary located north of Tokyo, took out a new license on the technology. Kashima-Kita powers its generators with Venezuelan pitch, a fuel rich in vanadium. Skyllas-Kazacos’s battery was a perfect fit. Here was a technology that allowed the company to recycle the vanadium from its soot and flatten out fluctuations in demand for its electricity at the same time. The world’s first large-scale vanadium battery went into operation in 1995, able to deliver 200 kilowatts for four hours—enough to power about 100 homes. It was a success, but Kashima-Kita sold the license and didn’t build another.
The buyer, Sumitomo Electric Industries, a giant Osaka-based company, had been working on NASA-style iron-chromium flow batteries since the early 1980s. Things looked up for Skyllas-Kazacos’s invention when Sumitomo switched to vanadium and licensed the technology in 1997. Three years later Sumitomo began selling vanadium batteries, including a 1.5-megawatt model that provides backup power to a Japanese liquid crystal display factory. By maintaining power during blackouts and thus preventing production losses, the battery reportedly paid for itself in six months.
Sumitomo has since demonstrated vanadium technology in at least 15 other implementations, including a 170-kilowatt battery at a wind farm in Hokkaido. All are located in Japan, their development subsidized by the government. Sumitomo doesn’t sell outside Japan, possibly due to the battery’s high manufacturing cost.
One company is now taking up the vanadium banner worldwide: VRB Power Systems, a Vancouver, British Columbia, start-up that bought most of the early intellectual property rights to the technology. The company is targeting the market for hybrid systems used to power remote, off-grid telecom applications. “In places like Africa, cell phone towers are typically powered by little putt-putt diesel engines that run 24/7,” VRB CEO Tim Hennessy says. By adding a vanadium battery to the system, one can run the diesel generator while charging the battery, turn the diesel off, run the battery, then repeat the cycle nonstop. “The beauty of the battery is that you can cycle it as many times as you like,” Hennessy says. “The electrolyte doesn’t wear out.”
VRB has installed 5-kilowatt batteries at two sites in Kenya. Hennessy claims that these can produce “at least a 50 percent reduction in the burning of diesel fuel, plus the diesels will need less maintenance and last much longer. It promises to make a huge difference to our customers’ operating expenses.” The firm’s other recent sales include a 20-kilowatt system, worth $300,000, that will deliver nine hours of backup power for an undisclosed major telecom company in Sacramento, California. These customers are learning firsthand what Skyllas-Kazacos learned two decades ago. The vanadium battery really works.
For all of vanadium’s promise, it still faces skeptics—including, surprisingly, some in the wind-power business who think the energy storage problem is not such a big deal. One big sticking point is price. Vanadium batteries currently cost about $500 per kilowatt-hour. So to run a city of 250,000 for 24 hours off a vanadium battery, the price tag would come to $2.4 billion. “Storage is not needed for wind, and it is unlikely to be cost effective in the next decade,” argues Rob Gramlich, policy director of the American Wind Energy Association. Gramlich points out that a recent U.S. Department of Energy report, “20% Wind Energy by 2030,” hardly mentions storage. He notes, too, that Denmark, the world’s most enthusiastic user of wind power, gets by without storage.
How do the Danes do it? The grid in western Denmark is strongly interconnected with those of Norway, Sweden, and Germany, which act as giant energy sponges for their neighbor. They sop up cheap surplus power from Denmark when the wind is blowing and return expensive hydroelectric power during peak periods. The result is that, although 17 percent of the electricity the Danes generate comes from wind, they use only 7 or 8 percent, according to Hugh Sharman of Incoteco, a Denmark-based energy consultancy and development company whose clients include VRB. The rest is exported.
That situation will not be sustainable if the countries add more renewable power—and the Danes propose building another 4.5 gigawatts’ worth of offshore wind farms. That leaves two ways of meeting electricity demand when the wind drops. Either build lots of small, fast-acting, fossil-fueled backup turbines, or go for storage. As the price of natural gas soars, battery storage is rapidly becoming a more economically appealing option. Researchers at the Riso National Laboratory for Sustainable Energy in Roskilde, Denmark, are currently evaluating a 15-kilowatt VRB battery.
Cost is not the only obstacle that the vanadium battery has to overcome. Reliability may also be an issue, following the shutdown last year of a vanadium battery showcase, a 200-kilowatt backup system that was installed in 2003 at a wind farm on King Island, off the northern coast of Tasmania. A problem with the plant’s battery (which was not supplied by VRB) caused the electrolyte to overheat, damaging the stack. Still, other demonstration vanadium batteries, such as a 250-kilowatt installation at Castle Rock, Utah, have been operating reliably for years.
One vote of confidence comes from China. A group led by Huamin Zhang at the Dalian Institute of Chemical Physics in northern China has finished testing 2-, 5-, and 10-kilowatt vanadium battery modules and is currently evaluating a 100-kilowatt system. Vanadium “will have a potential market in China with the increasing development of renewable energy supported by the Chinese government,” Zhang wrote in an e-mail message. “Furthermore, large-scale energy storage systems are strongly needed in China [as backup during] frequent natural disasters” such as the recent Sichuan earthquake.
The greatest challenge to the vanadium battery may come from other advanced battery technologies, most seriously from sodium-sulfur batteries made by the Japanese ceramic specialist NGK Insulators. Though less scalable, sodium-sulfur has attracted investors because it is a more mature technology. Installations include the town of Rokkasho in northern Japan, where 34 megawatts of sodium-sulfur storage backs up 51 megawatts of wind turbines.
In the end, the vanadium battery has some uniquely appealing traits that may make it the best partner for renewable energy—not just for giant wind farms, but also for small-scale turbines and solar cells that bring renewable power directly into consumers’ homes. Currently, sodium-sulfur technology doesn’t work well at sizes below 1 megawatt. For smaller applications, such as regulating the flow of electricity from a house’s solar panels, vanadium-based systems look more cost-effective. They can be fit to more modest demands by using smaller tanks.
These smaller applications are where Skyllas-Kazacos is currently focusing her efforts. Three years ago she, along with her husband Michael and sons Nick and George, founded V-Fuel to develop and commercialize a second-generation vanadium battery. The impetus to found V-Fuel came when the University of New South Wales sold the rights to first-generation vanadium battery technology to VRB Power Systems. Two years later, with nothing left to develop, her battery lab—which at its height had 18 members—closed. Yet people kept contacting Skyllas-Kazacos about vanadium batteries, and she kept thinking up ideas for a better version. In 2005, at age 54, her husband wanted to retire. She told him, “No, you can’t—we’re starting again!”
“I could see so many opportunities,” Skyllas-Kazacos says, “but a lot of this interest wasn’t translating into real sales because the cost was just too expensive.” The key to cutting cost, she notes, is finding a replacement for the flow battery’s most expensive part, the membrane. Following a worldwide search for a suitable material, V-Fuel designed a polymer membrane that Skyllas-Kazacos claims is durable and less than half the price of conventional materials. A second challenge is making a smaller battery, one that does not need a warehouse to store electrolyte tanks. To do this, Skyllas-Kazacos has found an electrolyte that allows more vanadium to dissolve, thus doubling the energy storage density.
Atop a bench in V-Fuel’s cramped workshop in Sydney sits a prototype 5-kilowatt battery stack. The size of a filing-cabinet drawer, the stack is designed to be rack-mounted above a square block consisting of two electrolyte tanks. The resultant package would be compact enough to fit in a household closet. Configured as part of a home-based generation system, it could absorb power from rooftop solar panels and discharge electricity during peak periods. Skyllas-Kazacos estimates that such a consumer-use vanadium battery might eventually sell for around $5,000. At that price it could pay for itself in a few years.
So the vanadium battery may play a big role both invisibly at the electric utility and very visibly in the home, smoothing out Mother Nature’s rough edges so that renewable power works just as well as coal or natural gas. Stabilizing a future national grid that draws the majority of its power from renewable sources may seem like a tall order for a technology that delivers megawatts, not gigawatts, of power as it is used today, but some industry insiders are confident batteries can rise to the challenge. “At this point, [a 1.2-megawatt battery] is fairly large-scale, but we are at the front end of this curve,” Jim Kelly of Southern California Edison says. “Five years from now that will seem so trivial. It’s like comparing the first personal computer you had with the ones we have today. You look back and laugh. I think we’ll see that same thing happen with the battery industry. We are taking baby steps, in part because the industry is not mature, the technology winners have not been determined, and the costs are still high. But these are all the things you expect as a revolution happens.”
Click here to see the rest of DISCOVERmagazine.com's special energy coverage.
Correction, the article was originally attributed to Robert rather than Bob Johnstone
*Correction, the article originally stated kilowatt-hours rather than watt-hours.