MIT Technology Review

Tuesday, June 29, 2010

A Guide to Recent Battery Advances

Battery breakthroughs could lower costs and improve performance for electric vehicles and renewable energy storage--but commercializing these new technologies will be challenging.

Electric vehicles, hybrids, and renewable energy have at least one thing in common--if they're ever going to be more widely used, representing the majority of cars on the road or a large share of electricity supply, batteries need to get significantly better. Batteries will need to store more energy, deliver it faster and more reliably, and ultimately, cost far less. The specific ways batteries need to improve vary by the application, but in all these areas, researchers have been making significant headway.

Last week, MIT researchers led by Yang-Shao Horn , a professor of materials science and engineering and mechanical engineering, and Paula Hammond, a professor of chemical engineering, announced a new approach to high-power lithium-ion batteries, the type that's useful for hybrid vehicles or for stabilizing the electricity grid. High-power batteries accept and deliver charge rapidly. In hybrids, the goal is to supplement the gasoline engine, allowing it to run at its most efficient. The battery drives the car at low speeds for short distances and boosts acceleration, lowering demand on the engine. It also captures energy from braking that would otherwise be lost as heat. For the electricity grid, such batteries could buffer changes in supply and demand of electricity--something that's becoming more important as more variable sources of electricity are introduced, such as wind and solar power.

The MIT researchers demonstrated a new battery electrode, based on specially treated carbon nanotubes, that last for thousands of cycles without any loss in performance. Batteries made from these electrodes could deliver enough power to propel large delivery vans or garbage trucks, for example, without the batteries being too heavy to be practical. (The researchers need to increase the thickness of the electrodes for them to be practical in these applications.) Companies such as A123 Systems, based in Watertown, MA, have also developed very high-power lithium-ion batteries, and other academic groups and startups are developing carbon nanotube-based ultracapacitors, which store energy using a different mechanism than batteries that's particularly useful for high power and long life.

While the new electrodes could eventually be useful for hybrids, and for stabilizing the grid, they aren't particularly good for other applications such as all-electric vehicles. For electric vehicles, the total amount of energy that batteries store is more important than how fast that energy can be delivered, since it's the total amount that determines how far these cars can travel between charges. The MIT researchers who developed the new carbon nanotube electrodes are also developing a different type of battery to store large amounts of energy. Called a lithum-air battery, where one of a battery's two electrodes is replaced by an interface with the air, the technology has recently attracted large amounts of government funding and interest from companies such as IBM. In theory, such batteries could store three times as much energy as conventional lithium-ion batteries. But the design has a number of problems that make it hard to commercialize, among the vulnerability of its active materials to moisture (the lithium metal it uses can catch fire if it gets wet) and the batteries' tendency to stop working after being recharged just a few times.

Like lithium-air batteries, other potential high-energy battery technologies face a number of hurdles, which could help explain why hybrids with their high-power rather than high-energy batteries have been more successful than electric vehicles. Many of the most promising battery chemistries are too difficult to make at a large scale, fall apart after a few cycles, or are too expensive. According to the U.S. Department of Energy, complete battery packs today cost between $800 and $1,200 a kilowatt hour, and store about 100 to 120 watt-hours per kilogram. To make electric vehicles practical and affordable, the DOE would like to see costs drop to $250 per kilowatt hour and increase storage capacity to over 200 watt-hours per kilogram. (Reaching these goals will require even higher storage capacities for the individual battery cells that make up battery packs--about 400 watt hours per kilogram.)

While improving batteries for hybrids and electric vehicles is difficult, one of the biggest long-term challenges for battery researchers is making batteries that can cheaply store vast amounts of energy generated by solar panels and wind turbines, so that electricity from these sources is available when the sun isn't shining or the wind isn't blowing. For now, such batteries aren't needed--there's enough power from conventional sources to take up the slack. But if solar and wind are ever to provide the majority of electricity, storage will be needed, and batteries today are far too expensive. The DOE goal for such batteries is less than $100 per kilowatt-hour, less than half its goal for electric vehicles. It's cheaper today to build a natural gas power plant as a backup source of power, or to store energy by pumping water uphill, where it can later flow downhill to spin a generator. One experimental approach to such low-cost batteries is something called a "liquid" battery, which uses inexpensive battery materials that assemble themselves.

Even if problems with batteries are overcome in the lab, these technologies face obstacles to being commercialized. To drive down costs, battery makers are turning to applications other than electric vehicles and the grid to get new technologies off the ground, applications such as microelectronics, power tools, and race cars. Plug-in hybrids can also help serve as a bridge to electric vehicles. Plug-ins use back-up gas-powered generators to help extend their range, allowing automakers to use smaller, less expensive battery packs than they'd need for electric vehicles. Automakers such as GM, with its Chevrolet Volt due out this year, are taking this approach. The electric vehicles on sale now, and that will be going on sale in the next few years, are either expensive sports cars and luxury vehicles, where costs can be high, or their upfront costs are being decreased using creative financing, such as leasing battery packs or offering per-mile plans something like the per-minute plans offered by cell phone companies.