In the fall of 2010, Kristin Persson had been at the Lab for two years. Hired to do lithium-ion battery modelling, she had her eye on a project to conduct data-driven research and design across chemistries to facilitate materials learning across systems. She decided to apply for an LDRD award (her second application, the first having been unsuccessful). She proposed creating a public database of computed properties of materials and making the database freely available, in the hopes of advancing materials design and democratizing materials data.
And she got the award.
According to Horst Simon, who had then just been appointed Deputy Laboratory Director for Research and Chief Research Officer and who led the LDRD selection process, the proposal presented an out-of-the-box idea. The idea wasn’t without risk, and indeed one judge was unsure about the modelling techniques, but it was worthy of further exploration.
“It wasn’t a huge amount of money, but it made all the difference,” said Kristin. “And Berkeley Lab was the right place for it, because it required interdisciplinary skills, from materials science to high performance computing to database expertise. Doing this work at a national lab made sense, because it’s the kind of work that, if successful, would require broader interdisciplinary expertise, and staff support that a university would find challenging to provide.”
“The LDRD helped me to try something that was creative, a bit risky, but exciting,” said Kristin.
The “skunkworks team,” as Kristin calls it, consisted of herself, a postdoc (Michael Kocher), and a gang of volunteer lab scientists as well as collaborators from MIT who were interested and enthusiastic to help out with the project on their own time. They built out the first platform, which was launched online in October 2011.
The rest is, as they say, history. Following the initial LDRD project, Kristin applied for funding from the DOE, and was granted $11 million over five years to create the Materials Project Center for Functional Electronic Materials Design and to scale up the project. Kristin ran the Center for five years, during this time establishing Berkeley Lab as the leading materials data provider in the world.
Today the Materials Project has 140,000 registered users worldwide, and is eclipsing the Alexa ranking (i.e., the data usage) of other large experimental databases, for example, the Inorganic Crystal Structure Database (ICSD). Three to five thousand users log on every day, and the platform delivers, on average, two million data points daily. The initiative has accelerated the discovery of new battery materials, transparent conducting oxides, and thermoelectric materials, among many others. The Materials Project is now a core program in the Lab’s Materials Science Division. And Kristin herself has continued to be instrumental with the Project and beyond; she was recently named director of the Molecular Foundry.
By all measures, the small “skunkworks” project funded by the LDRD award back in 2011 has grown into a huge success. “If it weren’t for that initial funding from the LDRD Award, we wouldn’t be here today,” said Kristin.
Kristin’s tips for other researchers who are considering applying for the LDRD? “I encourage researchers to be fearless when applying for LDRD funding. Unlike most government funding that favor projects that have already had some early results or success, LDRD funding allows you to explore out-there ideas,” said Kristin. “Don’t waste this kind of opportunity on funding that you could get elsewhere.”
The LDRD, which has seeded many wonderful projects at Berkeley Lab, will issue the next call for proposals by the end of December; proposals will be due in March, 2021.
For more information about the LDRD, email ldrd@lbl.gov.