Search This Blog

Saturday, December 31, 2011

What If Electric Cars Were Better?



Improving the energy density of batteries is the key to mass-market electric vehicles.

  • BY DAVID ROTMAN
Electric vehicles are still too expensive and have too many limitations to compete with regular cars, except in a few niche markets. Will that ever change? The answer has everything to do with battery technology. Batteries carrying more charge for a lower price could extend the range of electric cars from today's 70 miles to hundreds of miles, effectively challenging the internal-combustion motor. 

To get there, many experts agree, a major shift in battery technology may be needed. Electric vehicles such as the all-electric Nissan Leaf and the Chevrolet Volt, a plug-in hybrid from GM, rely on larger versions of the lithium-ion batteries that power smart phones, iPads, and ultrathin laptops. Such gadgets are possible only because lithium-ion batteries have twice the energy density of the nickel–metal hydride batteries used in the brick-size mobile phones and other bulky consumer electronics of the 1980s. 
Using lithium-ion batteries, companies like Nissan, which has sold 20,000 Leafs globally (the car is priced at $33,000 in the U.S.), are predicting that they've already hit upon the right mix of vehicle range and sticker price to satisfy many commuters who drive limited distances.
The problem, however, is that despite several decades of optimization, lithium-ion batteries are still expensive and limited in performance, and they will probably not get much better. Assembled battery packs for a vehicle like the Volt cost roughly $10,000 and deliver about 40 miles before an internal-combustion engine kicks in to extend the charge. The battery for the Leaf costs about $15,000 (according to estimates from the Department of Energy) and delivers about 70 miles of driving, depending on various conditions. According to an analysis by the National Academy of Sciences, plug-in hybrid electric vehicles with a 40-mile electric range are "unlikely" to be cost competitive with conventional cars before 2040, assuming gasoline prices of $4 per gallon.
Estimates of the cost of assembled lithium-ion battery packs vary widely (see "Will Electric Vehicles Finally Succeed?"). The NAS report put the cost at about $625 to $850 per kilowatt-hour of energy; a Volt-like car requires a battery capacity of 16 kilowatts. But the bottom line is that batteries need to get far cheaper and provide far greater range if electric vehicles are ever to become truly popular. 
Whether that's possible with conventional lithium-ion technology is a matter of debate. Though some involved in battery manufacturing say the technology still has room for improvement, the NAS report, for one, notes that although lithium-ion batteries have been getting far cheaper over the last decade, those reductions seem to be leveling off. It concludes that even under optimistic assumptions, lithium-ion batteries are likely to cost around $360 per kilowatt-hour in 2030.
The U.S. Department of Energy, however, has far more ambitious goals for electric-vehicle batteries, aiming to bring the cost down to $125 per kilowatt-hour by 2020. For that, radical new technologies will probably be necessary. As part of its effort to encourage battery innovation, the DOE's ARPA-E program has funded 10 projects, most of them involving startup companies, to find "game-changing technologies" that will deliver an electric car with a range of 300 to 500 miles.
The department has put $57 million toward efforts to develop a number of very different technologies, including metal-air, lithium-sulfur, and solid-state batteries. Among the funding recipients is Pellion Technologies, a Cambridge, Massachusetts-based startup working on magnesium-ion batteries that could provide twice the energy density of lithium-ion ones; another ARPA-E-funded startup, Sion Power in Tucson, Arizona, promises a lithium-sulfur battery that has an energy density three times that of conventional lithium-ion batteries and could power electric vehicles for more than 300 miles.
The ARPA-E program is meant to support high-risk projects, so it's hard to know whether any of the new battery technologies will succeed. But if the DOE meets its ambitious goals, it will truly change the economics of electric cars. Improving the energy density of batteries has already changed how we communicate. Someday it could change how we commute.

The Year in Materials



Vibrant displays head to market, invisibility cloaks become more practical, and batteries store more energy.

  • BY KEVIN BULLIS


Tiny crystals called quantum dots emit intense, sharply defined colours. Now researchers have made LED displays that use quantum dots. QD Vision demonstrated its first rudimentary one-colour displays using nanoscale crystals five years ago. This year it demonstrated a full-colour display capable of showing video. The company says it could be another five years before the technology appears in commercial displays. Samsung might get there first—it's also developing quantum-dot displays and demonstrated a full-colour one in February.
Quantum-dot displays could use far less energy than LCDs. Another ingenious way to reduce energy use is to make displays that emit no light but instead reflect ambient light, an approach Qualcomm is taking with its full-colour Mirasol displays, which use only a tenth of the energy of an LCD. The technology has started to appear in tablet computers in South Korea.
No display looks good after it's covered with fingerprints. A new coating based on soot from a candle flame could provide a cheap oil-repelling layer that could eliminate smudges.
Novel nanostructured materials could significantly enhance the power output of solar panels and make them cheaper by capturing light that would have otherwise been reflected. They could also achieve these goals by converting near-infrared light into colours that conventional silicon solar cells can absorb. Another material could render stealth aircraft invisible at night—and invisible to radar night and day.
Metamaterials offer another approach to invisibility: instead of absorbing light, metamaterials bend it around an object. Until this year, researchers have only been able to make metamaterials on a small scale—less than a millimetre across. Now they've made them big enough to be practical. They don't work yet for all wavelengths of light, but they could render objects invisible to night vision equipment.
Stanford researchers built a battery electrode that can be recharged 40,000 times—compared to the 1,000 charges you'd get with a typical laptop battery. Since the electrode lasts so long and is made of abundant materials, it could provide an inexpensive way to store power from wind turbines and solar panels.
Other researchers have developed inexpensive materials that can store 10 times as much energy as conventional graphite electrodes in lithium-ion batteries. Paired with an equally high-capacity opposite electrode, these could transform portable electronics and electric vehicles. One technology from Lawrence Berkeley National Laboratory seems promising because it uses a conductive polymer that can be incorporated into existing manufacturing lines instead of requiring the expensive new technology for making nanostructures required by others.
New tools could speed up the next materials breakthroughs. A modelling program developed at Harvard has led to one of the best organic semiconductors ever made. And a robotic system for making thousands of battery cells with unique electrode chemistries has discovered materials that could boost lithium-ion battery storage capacity by 25 per cent.

The Year in Numbers





A look back at 2011's biggest technology stories, by the numbers.

  • BY MIKE ORCUTT
Japan's nuclear catastrophe
The disaster at Tokyo Electric Power's Fukushima Daiichi nuclear power plant, set off by a 9.0 magnitude earthquake off Japan's east coast, scored a 7 out of 7 on the International Atomic Energy Agency's International Nuclear and Radiological Events Scale. The metric ranks severity based on many parameters, including an incident's effects on humans and the environment. According to the Japanese government, decommissioning the plant will take 30 to 40 years, and cost an estimated $15 billion.
Many nations reacted by scaling back their nuclear ambitions. Germany led the way, announcing that by 2022, all 17 of its nuclear reactors will be shut down. Those reactors, according to the World Nuclear Association, generated 133 billion kilowatt-hours in 2010, or28.4 percent of the country's electricity.

As of December 2011, there are still 433 operating nuclear power plants in the world. The United States has the most, with 104. Meanwhile, 499 more reactors are either planned or proposed globally—171 of those are in China, which currently has 26 operable reactors.

The solar industry in transition
The average nominal (not adjusted for inflation) price of crystalline silicon solar panels fell from $1.90/watt to $1.35/watt from January to November 2011, according to GTM Research. The solar industry is facing a large oversupply of panels, fueled largely by manufacturers in China, home to four of the top five largest solar-panel manufacturers in the world.
The world's solar capacity continues to grow quickly. GTM Research estimates that more than 20,563 megawatts of solar power were installed globally in 2011 (13,553 megawatts in Europe, 2,083 in North America, 3,938 in Asia,710 in Australia, and 279 megawatts in the rest of the world). That is 2,960 more than were installed last year, and brings the total global solar capacity to 59,152 megawatts.
(See The Chinese Solar Machine, by Kevin Bullis, and Can We Build Tomorrow's Breakthroughs? by David Rotman.)
Data deluge
The amount of data we create, replicate, and store in gadgets and the cloud is growing at a staggering rate. According to IDC, the total "digital universe," or all the digital information that has been created or replicated, grew to 1.8 zettabytes in 2011 (a zettabyte is a trillion gigabytes) in 500 quadrillion files. IDC says the total size has grown by a factor of nine over the past five years. Handily enough, in August, researchers at IBM unveiled the largest harddrive ever, capable of holding 120 petabytes (a petabyte is a million gigabytes), or about 24 billion five-megabyte mp3 songs.
(See IBM Builds Biggest Data Drive Everby Tom Simonite, and The Cloud Imperative, by Simson Garfinkel.)
Google tries social (again)
Google took a second shot at social networking with the release of Google+, which is a lot more like Facebook than its failed first attempt, Buzz, which was more comparable to Twitter.
Sign-ups skyrocketed in the days immediately after the introduction of the new service on June 28, hitting the 10 million mark about two weeks later. By October, the last time Google released user numbers, total sign-ups stood at 40 million. Active use of Google+ has been harder to gauge, but whatever the number, it's still dwarfed by Facebook's claim of 800 millionactive users.
(See Tom Simonite's Q&A with Bradley Horowitz, the man building Google+, and How Google+ Will Balkanize Your Social Life, by Paul Boutin.)

Automakers electrify
General Motors and Nissan released their anticipated, and much-hyped, electric models at the very end of 2010. As of December 1, GM had sold 6,468 Volts, well short of its goal of 10,000. Nissan, meanwhile, claims it has reached its 2010 global sales goal of 20,000 Leafs.
(See Will Electric Vehicles Finally Succeed?, by Peter Fairley, and A Wish List for the Next GM Voltby Kevin Bullis.)
Stem-cell trial halted
In November, biotechnology firm Geron halted its stem-cell research program, including the first U.S.-approved clinical trial of human embryonic stem cells. The company, which spent $45 million just to gain approval from the U.S. Food and Drug Administration for the landmark trial, only treated four patients.
(See Stem-Cell Gamble and Geron Shuts Down Pioneering Stem-Cell Program, both by Antonio Regalado.)

The Chinese Solar Machine



Chinese manufacturers have ­dominated the international ­market for conventional solar ­panels by building bigger ­factories faster. Now they will need to ­innovate to maintain their lead.

  • BY KEVIN BULLIS


T
en years ago, solar panels were made mostly in the United States, Germany, and Japan. Chinese manufacturers made almost none. But by 2006, the Chinese company Suntech Power had the capacity to make over a million silicon-based solar panels a year and was already the world's third-largest producer. Today Chinese manufacturers make about 50 million solar panels a year—over half the world's supply in 2010—and include four of the world's top five solar-panel manufacturers. What makes this particularly impressive is that the industry elsewhere has been doubling in size every two years, and Chinese manufacturers have done even better, doubling their production roughly every year.
This dominance isn't due to cheap labor in Chinese factories: making solar cells requires such expensive equipment and materials that labor contributes just a small fraction of the overall cost. Nor is it because the Chinese companies have introduced cells that last longer or produce more power: by and large, they make the same type of silicon-based solar panels as many of their competitors around the world, using the same equipment. They have succeeded in large part because it's faster and cheaper for them to build factories, thanks to inexpensive, efficient construction crews and China's streamlined permitting process. The new factories have the latest, most efficient equipment, which helps cut costs. So do the efficiencies that come with size. As a result, Chinese manufacturers have been able to undercut other makers of silicon solar panels and dash the hopes of many upstarts hoping to introduce novel technology.
But the solar market is rapidly evolving, and technological innovations are becoming increasingly essential. Though demand for solar power continues to grow around the world, the market is flooded with photovoltaic panels: worldwide production capacity more than doubled from 2009 to 2010 and continued to increase in 2011. The overcapacity was so great that last fall, Chinese manufacturers had trouble selling solar panels for more than it cost to make them. In such a market, the way to differentiate your product—and charge enough to stay afloat—is to make it better than your competitors'. 
For solar manufacturers today, that means inventing cells that are more efficient at converting light to electricity. As the price of solar panels has fallen, installation costs have come to account for a greater percentage of solar power's cost. Customers want panels that are more powerful, so that they can install fewer of them. From now on, the best way for Chinese manufacturers to lower the cost per watt of solar power may not be by lowering manufacturing costs but, instead, by increasing the number of watts each panel generates. "The game is now changing," says Mark Pinto, executive vice president of energy and environment solutions at Applied Materials in Santa Clara, California, the world's largest supplier of solar manufacturing equipment. "Before, it was all about scale. Now it is about conversion efficiency while keeping the cost down."
This might sound like bad news for Chinese manufacturers that have focused on scaling up standard technology. But their experience in building conventional solar panels could help them implement new designs that significantly boost the performance of silicon solar cells. Over the years, these manufacturers have lowered costs in part by developing better ways to manufacture the cells. That's given them an understanding of what works and what doesn't on the factory floor. They also have the capital and the engineers to help them translate newer technologies into mass production. They might not have initially set out to commercialize those technologies, but now, having mastered the market for conventional solar panels, they're poised to do just that.

KEEPING PACE
In 2010, when the U.S. secretary of energy, Steven Chu, gave a speech to the National Press Club laying out his case that the United States was falling behind in advanced manufacturing, Suntech Power was his Exhibit A. He had toured its factory, and he was impressed by what he'd seen. "It's a high-tech, automated factory," he said. "It's not succeeding because of cheap labor." Not only that, he noted, but Suntech had developed a type of solar cell with world-record efficiencies.
Chu's assessment might have surprised some observers, but Suntech's record-setting solar cells are impressive. The technology that goes into them takes advantage of changes in both design and manufacturing technique. The conductive metal lines that collect electric charge from the silicon aren't created with screen-printing methods, as is standard. Instead, Suntech uses a proprietary process to deposit much thinner, more closely spaced lines that are more efficient at extracting electricity from the cells. The changes have allowed the company to reach efficiency levels and cost reductions that an industry road map released in 2011 had set as targets for 2020. "When you put all those things together, we are not only doing better than what people are doing now," says Stuart Wenham, the chief technology officer at Suntech. "We are also doing better than what they think they could be doing in 10 years."
So far, Suntech has made relatively few solar panels based on the new technology. Instead, it has focused its resources on tweaking manufacturing processes to decrease the cost of making conventional silicon solar panels. But that could soon change. This year Suntech has started to increase production of the new cells, and now it can make enough of them annually to generate 500 megawatts of power—roughly 2.5 million solar panels. That achievement owes much to the company's success as a producer of the conventional products.
The technology behind the new cells was developed in the 1990s at the University of New South Wales, Australia, but the techniques used in the lab were too expensive for commercial production. It was a "horribly sophisticated process" including photolithography, vacuum deposition of "quite exotic metals," and "all sorts of chemical processes," says Wenham, who is also head of the photovoltaics research program at UNSW and was formerly a professor of Suntech's CEO and founder, Zhengrong Shi. According to Wenham, the technology remained a lab curiosity for decades until Suntech's researchers figured out how to adapt it to an assembly line. "They came up with a simple, low-cost way to replace all of that while achieving the same results," he says. The new technology could increase the power output of a standard-sized solar panel from 205 watts to 220 watts or more—and the cells costs less to produce than conventional ones.
Individual parts of the technology were quickly successful. Suntech introduced these into its standard manufacturing lines, with an eye to keeping just ahead of its competitors in terms of cost and efficiency. Scaling up the complete process, however, was a challenge. A pilot manufacturing line was up and running in 2009, but the company had to develop and implement new equipment to get yields and production rates to the point that the process was economical. Here Suntech's position as a market leader with experience in developing new manufacturing equipment proved critical. Not only did the company have the expertise it needed to improve the process; it also had the funds to keep working on the technology for years without its bringing in significant revenue.
Suntech isn't the only Chinese solar manufacturer to identify promising new technology and find ways to produce it at a large scale. Last September, Yingli Green Energy, based in Baoding, announced that a partnership with a Dutch research center, ECN, had yielded solar panels that could convert 17.6 percent of the energy in sunlight into electricity; the average is just over 14 percent. "ECN made the technology available to anyone in the world who wanted it," ­Wenham says. "Yet it's only been Yingli that's taken that technology and worked out how to make it in large-scale production, at low cost."
MATERIAL ADVANTAGE
Even now that Chinese solar manufacturers are shifting focus from production to innovation, there may be limits to what they can do with their chosen material, crystalline silicon. This material is attractive because the industry knows how to work with it, thanks in part to decades of research in silicon microchips. But compared with some other semiconductors, it's lousy at absorbing sunlight. Some alternatives, like gallium arsenide, can be made into films of material that can generate as much electricity as a typical silicon cell but are just a hundredth as thick, potentially reducing material costs. Such thin films can also be flexible: they could be rolled up, reducing packaging and shipping costs, and they could be built into roofing shingles to reduce installation costs.  
Yet despite their potential advantages, it has been difficult for thin-film solar cells to compete with the ever decreasing costs and improving efficiency of crystalline silicon ones. One company, Arizona-based First Solar, has succeeded in developing low-cost manufacturing techniques for thin-film solar panels, but these methods use a material—cadmium telluride—that results in panels less efficient than silicon ones. Other companies have tried to compete with silicon by using higher-efficiency thin-film panels of copper indium gallium selenide. Some of them, however, have had to declare bankruptcy and close their factories after failing to lower manufacturing costs fast enough.
Despite these struggles, Wenham believes that thin-film technology will eventually challenge conventional solar panels. If that's true, Chinese makers of crystalline silicon solar cells may not dominate the market forever. But the strategy of first scaling up conventional technology and then introducing innovative designs to keep lowering the cost per watt of solar power has put them in a good position to maintain their lead for years. In the meantime, some, like Suntech, are working to produce thin-film panels of their own. When thin films do replace crystalline silicon, it could be Chinese manufacturers that make them.
Kevin Bullis is Technology Review's Senior Editor for Energy.