Accelerating Battery Materials Discovery by Multi-Property Optimization

PI:  Evan Reed, Materials Science & Engineering;
Students: Brandi Ransom, Eder Lomeli, Materials Science and Engineering

We aimed to develop statistical models that combine available experimental and computational data with physics and chemistry to identify promising new pairings of solid electrolyte and cathodes that exhibit the spectrum of desired properties for lithium ion batteries. For cathodes, these include simultaneously low chemical expansion, high energy density, high voltage, and high ionic conductivities. For solid electrolytes, these include a large electrochemical window, high ionic conductivity, low electronic conductivity. For the interface between the two, one would like either no chemistry, or breakdown into suitable products. Our models will be used to screen billions of possible combinations of cathode and electrolyte to identify the most promising new spaces in which to drive experimental synthetic and characterization efforts in collaboration with the Cui and Chueh groups.