We’ve previously described in this space the Materials Genome Initiative (MGI) — a $100 million initiative announced last June to drastically accelerate the discovery, development, and manufacturing of new and advanced materials — describing the critical role to be played by the computer and information sciences and engineering research community, including via predictive modeling, simulation, and visualization capabilities. Now there’s an interesting news focus (subscription required) in this week’s Science, noting, “Supercomputing power now makes it possible to compute the properties of thousands of crystalline materials in a flash and is expected to guide experimentalists where to search for the next best things.”

According to the article (following the link):

…Today’s machines aren’t yet able to simulate all types of materials. But a broad array of researchers think steady progress in technology has now made supercomputers powerful and available enough to make the task worth starting. “The scaling of computing is really making this possible,” says Gerbrand Ceder, a materials scientist at the Massachusetts Institute of Technology in Cambridge. Richard Hennig, a computational materials scientist at Cornell University, agrees. “The field of computational materials science is ready to take off,” Hennig says…

 

[The Materials Genome Initiative] has been a long time coming. Ever since the advent of quantum mechanics in the early 1900s, researchers have known how to compute the properties of materials, at least in principle. But most materials are so complex that only those with a small number of electrons can be fully analyzed at the quantum level. Because complex materials have on the order of 10²³ electrons, researchers must rely heavily on computational methods to simplify the problem. These include a wide array of simulation techniques, such as density functional theory calculations, finite-element methods, and Monte Carlo simulations, all of which have been around for years. “The codes really haven’t changed that much” over time, says Sadasivan Shankar, a computational materials scientist at Intel Corp. in Santa Clara, California. “What has changed is that the computing power has caught up.”

 

That progress opens the door to computing the properties of a wide range of materials that once seemed unapproachably complex, Ceder says. Among the more tractable problems should be advances in catalysts, battery materials, and thermoelectrics, which convert heat to electricity. And it should be relatively straightforward to make a big impact on materials research quickly. There are between 50,000 and 100,000 known inorganic compounds, depending on whose figures you believe, Ceder notes. Crunching the numbers for all the computable properties of a single known material — including crystal structure, stability, and ionic mobility — takes the equivalent of 1 day for a standard computer chip, known as a CPU. To make that calculation for all known inorganic compounds would take between 2 million and 3 million CPU hours, a job one of the most advanced supercomputers could carry out in just a day and a half. Examining a good swath of the possible unknown materials out there would still take only half a billion CPU hours, Ceder predicts. “That’s just a drop in the bucket” of the computing power available, Ceder says. “We don’t know most things about most materials,” he says. “Materials scientists are hungry for this data.”

 

Computational methods won’t supplant experimental synthesis of materials anytime soon, Ceder and others say. Rather, they will help focus the work. Because it’s not possible to synthesize and test all possible combinations of elements, computational tools are helping experimentalists decide where best to try cooking up new winners…

 

From Materials Scientists Look to a Data-Intensive Future, Robert F. Service, Science 23 March 2012: 335 (6075), 1434-1435. [DOI:10.1126/science.335.6075.1434]. Reprinted with permission from AAAS.”

Check out the full story here (subscription required) — and learn more about the MGI here.

(Contributed by Erwin Gianchandani, CCC Director)