Purdue Profs Weigh In On Moore's Law Comparisons

As the excitement over electric vehicles (EVs) rises, we’re repeatedly hearing comparisons between the future of electric cars and the historical price-performance of electronics.

            The logic usually goes like this: Look how the price of computer technology has plummeted over the years. The same will happen with EV batteries.

            Recently, we’ve heard comments along these lines from some pretty high-powered people. Mark Reuss, North American president of General Motors Corp. recently compared electric cars to cell phones in a Wall Street Journal opinion piece. Noted New York Times writer Thomas Friedman wrote about a “Moore’s Law of electric cars,” which says that “the cost per mile of the electric car battery will be cut in half every 18 months.” Shai Agassi of Better Place told CNNMoney.com that “every five years, you need about half as much lithium to create the same car battery.”

            At the same time, though, Bill Gates has expressed a wholly different opinion. He said we’ve been spoiled by Moore’s Law, and we shouldn’t expect the same kinds of advancements from battery technology. “They haven’t improved hardly at all,” he warned, in describing the state of battery technology. “There are deep physical limits.”

            It’s hard to imagine how the opinions of experts could be so wildly divergent, so we called on three engineering professors from Purdue to shed some light on this subject. We’re posting their comments below:

James Caruthers, Professor of Chemical Engineering, Purdue University:  “Moore’s Law is for silicon-based semiconductors, where the basic materials science has been basically unchanged since the original work in the 1950s.The key advances over the last 60 years have been in the manufacturing processes, where (i) ever-smaller units were etched into the silicon (i.e., better lithographic methods) and (ii) purer and purer silicon was manufactured (i.e., less defects that were etched could be smaller). The improvement in manufacturing processes often can be predicted to have steady improvement – i.e., it is engineering vs. fundamental materials science…”

“…In contrast, the improvement in batteries will require development of new materials. No one knows exactly when or where these improvements will occur, but it will not be a smooth steady improvement like Moore’s Law predicts for the manufacturing processes involved in chip manufacturing.”

Joseph Pekny, Professor of Chemical Engineering and Interim Head of Industrial Engineering at Purdue, together with Eric Dietz, Associate Professor of Computer and Information Technology at Purdue, and James Caruthers, Professor of Chemical Engineering at Purdue: “In the long term, batteries have the potential to provide the same or nearly the same amount of energy density as gasoline – but not much more, for example, rechargeable lithium-oxygen batteries or other metal-oxygen batteries can ultimately compete with gasoline. The lithium-ion batteries of today are much better than the batteries of the 1990s. Unfortunately, the path from today to the future will not be as predictable as Moore’s Law. Recall that in Moore’s Law, computers improved their cost and performance by a factor of two nearly every eighteen months. The battery improvements will surely come with research and development investment and increased sales volume of electrified drive trains with the starting point being today’s batteries and a long term goal being lithium-oxygen batteries that are as good or nearly as good as gasoline, rechargeable, long-lasting, safe, cheap and recyclable. How we get from today to the future will be more unpredictable than Moore’s Law in the frequency of advance, the amount of advance, and the timing of advance. For a society accustomed to Moore’s Law, the chemistry and engineering of battery improvement will be a bit more harrowing, but over time applied human ingenuity will result in batteries miraculous by today’s standards. Keep in mind that battery research and development has so far received much less funding than the semiconductor industry powering Moore’s Law in electronics. Also, the uptick in battery R&D funding will take some time to yield results, just as semiconductor research had a lag time.”

Finally, another comment regarding the future of lithium-oxygen batteries:

Joseph Pekny, Professor of Chemical Engineering and Interim Head of Industrial Engineering at Purdue: “Lithium-oxygen batteries will take a decade or two or three to perfect and other less theoretically capable batteries will improve with techniques such as on-board sensors…. The future for batteries should be bright, but people need to understand that the advances will be a bit lumpier since we are not fortunate to have a single guiding principle such as feature size reduction.”