Making large batches of a complex material consistently in a factory almost always requires processes different from those used to make small batches in the lab. That means more money, time, and risk. For example, say a research lab has made one-inch-square working solar cells whose active layer is created by printing a nanoparticle ink. Commercializing such a technology requires a company to develop several manufacturing techniques. First it has to figure out how to make the nanoparticles in large batches; then it must find an equipment maker to provide a customized machine for printing those inks over square meters, or develop that equipment itself. But it may not even get to that stage. What if, when researchers try to make large numbers of these solar cells, they can’t get the nanoparticles arranged in a consistent way, and the cells don’t work? At any stage, a fatal flaw might be uncovered.
The Materials Genome Initiative aims to predict such manufacturing problems and steer scientists and engineers away from them earlier in the development phase. The problems related to scaling up from lab bench to factory aren’t anything special, says Ceder. The main challenge right now is that individual groups and companies have been developing snippets of code and amassing data on new and existing materials, but they have no way to share this information. They file a patent, get a paper published, and it stops there. The Materials Genome will gather all such data into a central database.
Academic culture is more amenable to sharing data than corporate culture, but Wadia, who has been talking with representatives of the major materials companies about this initiative over the last few years, believes corporate labs will also contribute. Indeed, it would be difficult for such a project to succeed without them. “It will start in pockets of communities, but we have to get a critical mass for this to work,” he says. Companies that make advanced materials are already generating a large amount of data through daily monitoring of manufacturing operations, and he hopes they will share this kind of information with the Materials Genome Initiative.
“We think a key role industry can play is providing our perspective on how materials are used, designed, and evaluated for industrial product applications,” says Christine Furstoss, technical director of manufacturing and materials technologies at GE Global Research. “We use a large number of materials that are applied across multiple industries and have a keen interest in helping to advance the performance and manufacturability of such materials.”
The initial $100 million will be distributed among four government agencies: the National Institute of Standards and Technology, the Department of Energy, the National Science Foundation, and the Department of Defense. White House representatives would not comment on how much money would go to each agency and for what specific projects, but the emphasis, says Wadia, is on building computational infrastructure. Just what that infrastructure should look like will be hashed out over the next year. Funding will also go to educational initiatives.
“Novel materials are key enablers for manufacturing,” says Ceder. “If you’re going to increase manufacturing in the U.S., you’re not going to do that on old technologies.”