Project Name: Machine Learning Assisted Enhancement of Perovskite Stability and Performance
Funding Opportunity: Solar Energy Technologies Office Fiscal Year 2018 Funding Program (SETO FY2018)
SETO Research Area: Photovoltaics
Location: Seattle, WA
SETO Award Amount: $1,500,000
Awardee Cost Share: $375,000
Principal Investigator: Hugh Hillhouse

-- Award and cost share amounts are subject to change pending negotiations --

This project team will research the physical properties and stability of different perovskite systems, build an open-source database to store large perovskite-centered data sets, and develop a machine learning model using this data to predict the degradation, long-term feasibility, and future use of perovskite solar cells. This work will help discover materials that can be used to make better perovskite solar cells, and the resulting machine learning algorithms will help predict overall system stability.


This project will develop and use a machine-learning algorithm to rapidly characterize perovskite systems and bridge knowledge gaps related to perovskite stability in different oxygen-, moisture-, and light-rich environments. The team will develop experimental protocols to test perovskite efficiency via photoluminescence tests that determine how well light is absorbed. They will then compare the results against existing material-property data sets to determine the relationship between perovskite composition and performance degradation.


The team will develop characterization tools that help identify stable perovskite materials that can be used to make more efficient, resilient solar cells. This project will enable machine learning to be used in materials research to improve perovskite lifetime prediction and solar cell stability.