Deixis Online

To solve this mystery, they turned to Lin-Wang Wang, a computational scientist in Berkeley’s Materials Science Division. Wang has spent many years developing and testing algorithms that employ density functional theory (DFT), a method that simulates the electronic properties of systems starting from fundamental physics. The battery system perfectly fit those that DFT simulates well, Wang says. It was an ordered system with structures built on the nanoscale, including the polymer, silicon nanoparticles and lithium ions.

According to Wang, several possible scenarios could account for the increased conductance and the X-ray absorption findings. By using codes that implement the density functional theory, and the computational resources at the Berkeley Lab-based National Energy Research Science Computing Center (NERSC), Wang and postdoctoral fellow Nenad Vukmirovic modeled lithium interaction with the silicon-polymer electrode.

“Of the many possible explanations of the experimental results, density functional theory pointed out how the whole thing works,” Wang says.

The theoretical calculations dovetailed nicely with the experimental calculations, providing a detailed look at how and why the polymer works so well.

Lithium ions – rather than handing off electrons to the embedded nanoparticle silicon – donate electrons to the polymer first, providing it with its conducting properties. Only afterward do lithium ions adhere to the silicon. Because of its conductivity, the polymer acts as an electrical bridge, allowing the positively charged lithium ions to move into the silicon and combine with the negatively charged electrons flowing through the polymer, something that no other previously tested polymer could do. The findings, published in the September 2011 issue of the journal Advanced Materials, provided an experimental and theoretical foundation for further refinement and advancement of battery components.

The ability to incorporate DFT calculations into the new materials development process was a relatively new phenomenon, Wang says. Ten years ago, DFT algorithms were untested and computationally intensive. It took many years to demonstrate that results obtained by DFT were reliable. But in recent years, the method has proven to be a valuable tool to provide insight into the underlying physics, particularly of nanostructural materials, Wang says.

The next generation of DFT algorithms, which can provide more accurate results and can calculate bigger systems, combined with the computational resources at NERSC and other facilities, will allow DFT to be almost a routine tool to explain and predict properties of new nanomaterials. Wang expects DFT to contribute to the development of new quantum dots and organic polymer arrays by tracking individual electrons as they move through the system.

In the case of the battery simulation, DFT provided a complete picture of the lithium-polymer interaction.

“At the beginning of this project, we discussed many wild pictures that were all possible,” Wang says. “It is really only through this calculation that we pinned down the actual mechanisms of how the system works.”

The picture that finally emerged, Wang adds, was “very beautiful.”