MIT model explains how the brain can learn novel tasks while still remembering what it has already learned.
CAMBRIDGE, Mass-The main obstacle to the widespread use of electric cars is the lack of a suitable battery. And the problem with the battery is the lack of materials to make it cheap, light, and powerful.
So concluded a team of MIT researchers, who then set about identifying such materials. As reported in the April 16 Nature, the team has succeeded, and in a novel way. They first predicted the composition of a promising material via computer models, then produced and tested it. It met their expectations.
The work is a step toward the researchers' overall goal. "Our dream as a team is to make a battery that will enable the dawn of the electric vehicle age," said Donald R. Sadoway, Professor of Materials Chemistry and MacVicar Faculty Fellow.
The work also paves the way for a new approach to materials science. "To our knowledge, this is the first time a novel material has been predicted [on the computer] and actually made," said Gerbrand Ceder, Associate Professor of Materials Science and Engineering and first author of the Nature paper.
Professor Ceder and Sadoway's coauthors are Yet-Ming Chiang, Kyocera Professor of Ceramics; MIT affiliate Mehmet K. Aydinol; Young-Il Jang, a graduate student in the Department of Materials Science and Engineering (MSE), and MSE Postdoctoral Associate Biying Huang.
A battery has three parts: two electrodes (a cathode and anode) separated by an electrolyte. Chemical reactions at the electrodes produce an electronic current that can be made to flow through an appliance connected to the battery. In a rechargeable battery, once the reactions have run their course, they can be reversed by the action of a power supply or "charger."
The material reported in Nature was developed for the cathode of a rechargeable lithium battery, which boasts the highest energy density of any rechargeable (it will run longer between charges). Currently the key cathode material in such batteries is a compound of lithium, cobalt, and oxygen. But cobalt is extremely expensive. As a result, these batteries are used only in small devices like cellular phones and laptop computers. A lithium battery for an electric car "would cost you about $20,000," Professor Sadoway said, "and still the vehicle would travel only about 120 miles on a full charge."
With the assistance of Dr. Aydinol, Professor Ceder began the hunt for a replacement material by first determining via computer calculations how the cathode in a rechargeable lithium battery works. That led to an important surprise: he found that the cobalt is not as important to the chemical reactions in the cathode as had been thought. "Gerd's deeper understanding of how lithium cobalt oxide works allowed us to identify an entire class of materials that had previously been ignored for this application," Professor Sadoway said.
The calculations showed that aluminum, which is cheap and light, or other metals in the same class could be used in place of cobalt. The work further predicted that a compound containing aluminum would generate a higher cell voltage (resulting in a more powerful battery) than the conventional cobalt version.
'SHOW ME THE EVIDENCE'
"This was all well and good, but now show me the evidence," said Professor Sadoway, who noted that the researchers still had another problem. The predicted compound, lithium aluminum oxide, in its pure form does not conduct electrons. "In a cathode, this material would be all voltage and no current. It wouldn't work." Some modification was in order.
So the team widened the circle again, drafting Professor Chiang, a ceramist, who pointed out that adding back some cobalt would solve the conductivity problem. In other words, a mixture of lithium aluminum oxide and lithium cobalt oxide should meet the requirements.
The next challenge was making the new material. This wasn't trivial. Among other things, Professor Ceder's models showed that the aluminum and cobalt had to be incorporated at very specific sites to get the predicted properties. A synthesis technique refined by Professor Chiang worked perfectly, however, and his graduate student, Mr. Jang, produced the "designer materials" to specification. (Different samples contained different proportions of aluminum and cobalt.)
Dr. Huang, working with Professor Sadoway, took the resulting samples--then powders--and made each into a functioning cathode. Next step: create a battery by coupling each new cathode with a conventional anode and electrolyte. "We tested them in my lab and they worked like a charm," Professor Sadoway said. "We got, as predicted, substantial increases in cell voltage."
In fact, the cathodes with higher levels of aluminum worked a bit too well. "It got to a point where the cell voltage became so high that the electrolyte became unstable," Professor Sadoway said.
Another key part of the battery work is the development of a new electrolyte. A fourth faculty member on the MIT team recently succeeded in doing just that (the work was not reported in the Nature article).
Introducing Anne M. Mayes, Class of '48 Associate Professor of Polymer Physics in the Department of Materials Science and Engineering. Professor Mayes's new electrolyte, a polymer, has a variety of novel properties. For one, it's a solid. Conventional electrolytes are liquid, which limits the shape of the battery (you need some kind of vessel to hold the liquid).
But the new electrolyte conceived by Professor Mayes is flexible. A battery made of it "would be something with the consistency of a potato-chip bag," Professor Sadoway said. "You could fold it up to make any number of different configurations." Imagine a car powered by a battery that's incorporated into the body panels.
The researchers recently created a battery with the new electrolyte, the new cathode and a conventional anode. This realizes the MIT team's first solid-state lithium battery, or one made completely of solids. And that brings the researchers closer to their ultimate goal.
"The American marketplace will not be interested in an electric car that can't go at least 200 miles before recharging," Professor Sadoway said. "In my judgment the only technology that has a prayer of propelling a car that far is the solid-state lithium battery. That's based on theoretical studies and on our knowledge of materials' energy densities."
Work continues on the solid-state lithium battery. For example, with their new insight on how the cathode works the researchers are exploring new materials that might work even better than those reported in Nature. They are also looking at new anode materials, a task that was always less of a priority because present anode materials don't have as many limitations.
Professor Ceder stressed the value of tackling the work as a team. "We couldn't have done this with any one of us missing," he said. "It's an unusually close and good collaboration.
"I think getting four high-strung professors together is not always that easy," he concluded with a smile.
The work has been funded in part by Furukawa Electric Company and the DOE's Idaho National Engineering Laboratory University Research Consortium. Pacific Lithium, Ltd., has licensed a number of the patent applications submitted by the MIT inventors.