AI-Guided Processing of Advanced Materials and Devices
Project Description
In the proposed project, the involved researcher will start from 2D and 3D sections of granular materials for electronics and energy applications to: 1) generate 2D and 3D synthetic descriptions that enable to identify those with that deliver the highest (power and energy) performance and reliability as applied in solid state systems; 2) infer the volumetric and interfacial properties directly from the images in terms of processing maps and physically consistent differential equations. The project will integrate a wide variety of data analytics and machine, learning strategies, as a steppingstone to automatically simulate and then design the analyzed material. The developed, automated understanding will be used to develop a feedback loop between processing and simulation that allows the automated optimization of the material description.
Start Date
Fall 2026
Postdoc Qualifictions
The ideal candidate will have a background on image analysis, machine learning, and CUDA. It is desirable (but not necessary) to have a background on granular science and numerical methods as it pertains to the microstructural properties and kinetics of materials.
Co-advisors
Edwin Garcia (redwing@purdue.edu; School of Materials Engineering; https://engineering.purdue.edu/ComputationalMaterials/)
Bedrich Benes (bbenes@purdue.edu; Computer Science; https://cs.purdue.edu/homes/bbenes/)
Bibliography
Deva A, Krs V, Robinson LD, Adorf CS, Benes B, Glotzer SC, García RE. Data driven analytics of porous battery microstructures. Energy & Environmental Science. 2021;14(4):2485-93.