Last week a National Science Foundation-sponsored workshop addressed multidisciplinary approaches to the Materials Genome Initiative. From left: Gregory Rohrer, Abby Kavner, Young-Shin Jun, and Amy Walker. Credit: ACerS.
Genome (noun; origin 1930s: blend of gene and chromosome): the complete set of genes or genetic material present in a cell or organism.
The Materials Genome Initiative has gotten lots of attention since it was announced by the White House Office of Science and Technology Policy (OSTP) in June 2011. Its stated goal (pdf) is simple: “… to discover, develop, manufacture, and deploy advanced materials at least twice as fast as possible today, at a fraction of the cost.”
In my opinion, “genome” is an unfortunate choice of word. Why? A genome is a constrained set. Perhaps a very large set, but the genome is limited to the genes that define the organism. In the human organism, that is about 23,000 genes. The Human Genome Project (HGP) sequenced and mapped the 23,000 genes in the human genome, which ended up involving about 3.3 billion base pairs.
The HGP narrowed the description of what we are made of by painting in the fine-strokes details. It was never about expanding the diversity of our basis or of life—no new organisms were “invented,” no additions to the genome were desired (for example, adding feathers to the anatomy), nor were any such innovations intended.
The idea of the MGI, in contrast, is to start with basic building blocks of matter—the fine structure—and discover new combinations of elements, i.e., materials, with end properties and functionalities in mind right from the start. The MGI is all about expanding the possibilities beyond that which is already mapped. And, with 116 elements comprising the periodic table of the elements, the possible combinations make 3.3 billion look paltry. (According to my “go to” mathematician, the number of ways you can combine up to just six elements or less of the 116 known elements exceeds three billion.)
Chances are you have seen the OSTP’s schematic of the “materials development continuum.” The beginning point is discovery of new materials, and this is where science researchers must take the lead in finding approaches to discovering the few hundred or thousand materials that are worth developing out of the many billion possibilities.
Last week, to address this issue, a group of researchers who work on ceramic materials gathered for a National Science Foundation-sponsored workshop, “The Materials Genome Initiative in Ceramics, Geosciences, and Solid-State Chemistry.” The workshop grew out of two previous “future directions” workshops: a 2011 workshop on “Materials by Design” at the University of California, Santa Barbara and the 2012 workshop on “Emerging Research Areas in Ceramic Science” in Arlington, Va.
About 20 researchers, mostly from academia, participated. They were an interesting mix of well-established scientists and new young-bloods, all working with ceramic materials, whether as materials scientists or as researchers from the geoscience, earth science, and solid-state chemistry and physics communities.
Alexandra Navrotsky, the brilliant and witty professor from the University of California, Davis, set the stage with her opening presentation by observing that there are striking commonalities between earth science and materials science. “The Materials Genome Initiative,” she said, “means different things to different people. It is up to the communities to define.” In her assessment, MGI means starting from first principles and CalPhaD-type calculations, mining data from databases, and designing new, smarter experiments. “Putting these together in a holistic fashion is the MGI,” she says.
Much common ground exists between the geosciences and materials science—both need structural, thermodynamic, and physical property data. Phase diagrams are very important in both fields. Both fields need kinetic and mechanistic information for modeling of impossible-to-observe phenomena, like the geoscience of Jupiter, for example.
Narrowing the idea somewhat, Krishna Rajan from Iowa State University, said, “MGI is about doing new science to solve pressing issues,” implying that new science should be driving solutions to new and urgent issues, not to incrementally improving existing technology. Incremental improvements may not be compelling to the business world, anyhow. For example, why would a company that specializes in refurbishing thermal barrier coatings be interested in a coating that lasts twice as long? To them, that looks like half as much work (read revenue). Similarly, a materials improvement that would require expensive retooling of a fabrication line may not be worth the trade-off.
The materials science and geoscience community share some common frustrations, for example, with the difficulty of modeling across multiple scales, especially length scales. This is true whether modeling the transition from the nanoscale to the mesoscale, or from meters to kilometers. Time scales matter, too. Modeling of processes that occur in picoseconds is challenging, but so is modeling of processes that occur in light years.
Given the challenges, what can modeling offer? Ram Seshadri from the University of California, Santa Barbara (and coorganizer of the workshop with Carnegie Mellon’s Gregory Rohrer) says, “looking at large data sets tells you where you will be wasting your time.” And the role of computation, according to Michelle Johannes of the Naval Research Laboratory, is to give experimentalists a “rough directional map, an explanation of trends, and suggestions for optimization of properties.” In return, the materials mathematicians need property data and characterization information (such as crystal structures) from the experimentalists.
All in attendance seemed to agree that access to data (that they liked) was a challenge with no easy (or cheap) solutions. There also seemed to be consensus that multidisciplinary dialog is extremely valuable, and this is an area where technical societies such as ACerS or the American Geophysical Union can help, perhaps by organizing multidisciplinary symposia, special issues of journals, or workshop-like meetings. (Last October, ACerS, took a step in this direction by setting up a structure for “Technical Interest Groups.” The idea of the TIGS is to provide a way for Society members and non-members to come together on a topic of mutual interest that does not really have a “home” in any particular society.)
Much of the discussion on data focused on the need for single crystal property data. The single crystal data gets you as close as possible to values for intrinsic properties. However, with a few exceptions, most engineered materials are not single crystals. Does that mean “junky” data is without value? Hardly! As one participant noted, much of the functionality of engineered materials we use today arises from their “junkiness,” whether from impure compositions or process effects or elsewhere.
There was much more, of course. The group will publish a report on the workshop in the ACerS Bulletin, which will capture in detail the issues, challenges, opportunities—and payoffs—of multidisciplinary approaches to materials discovery.