Designing the best molecular building blocks for battery components is like trying to create a recipe for a new kind of cake when you have billions of potential ingredients and now researchers are building better batteries with machine learning.

The challenge involves determining which ingredients work best together or, more, produce an edible (or, in the case of batteries, a safe) product.

But even with state-of-the-art supercomputers, scientists cannot model the chemical characteristics of every molecule that could prove to be the basis of next-generation battery material.

Instead, researchers at the U.S. Department of Energy’s (DOE) Argonne National Laboratory have turned to the power of machine learning and artificial intelligence to speed up the process of battery discovery.

As described in two new papers, Argonne researchers first created an accurate database of roughly 133,000 small organic molecules that could form the basis of battery electrolytes.

To do so, they used an intensive model called G4MP2. This collection of molecules, but, represented only a small subset of 166 billion larger molecules that scientists wanted to probe for electrolyte candidates.

Because using G4MP2 to resolve each of the 166 billion molecules would have required an impossible amount of computing time and power, the research team used a machine-learning algorithm to relate the known structures from the smaller data set too much more modeled structures from the larger data set.

When it comes to determining how these molecules work, there are big tradeoffs between accuracy and the time it takes to compute a result.

To provide a basis for the machine learning model, Foster, and his colleagues used a less taxing modeling framework based on density functional theory, a quantum mechanical modeling framework used to calculate electronic structure in large systems.

Density functional theory provides a good approximation of molecular properties but is less accurate than G4MP2.

Refining the algorithm to better learn information about the broader class of organic molecules involved comparing the atomic positions of the molecules computed with the accurate G4MP2 versus those analyzed using only density functional theory.

The machine learning algorithm gives us a way to look at the relationship between the atoms in a large molecule and their neighbors, to see how they bond and interact, and look for similarities between those molecules and others we know quite well.

This will help us to make predictions about the energies of these larger molecules or the differences between the low- and high-accuracy calculations.