UC Davis Researchers Use Machine Learning to Develop Efficient Solar Cells

The University of California, Davis College of Engineering has made an exciting breakthrough in solar energy research. Researchers have been able to use machine learning to identify new materials for high-efficiency solar cells. They have found that by using high-throughput experiments and machine learning-based algorithms, it is possible to forecast the materials' dynamic behavior with very high accuracy, without the need to perform as many experiments. These findings are featured on the cover of the April issue of ACS Energy Letters. Over the past 10 years, hybrid perovskites have received a lot of attention for their potential use in renewable energy, according to Marina Leite, associate professor of materials science and engineering at UC Davis and senior author on the paper. Perovskites are organic-inorganic molecules that are cheaper to make and lighter than silicon, potentially allowing a wide range of applications, including light-emitting devices. However, a primary challenge in the field is that perovskite devices tend to degrade way more readily than silicon when exposed to moisture, oxygen, light, heat, and voltage. The researchers aim to find which perovskites combine high-efficiency performance with resilience to environmental conditions. The researchers built an automated, high-throughput system to measure the photoluminescence efficiency of five different perovskite films against the conditions of summer days in Sacramento. They were able to collect over 7,000 measurements in a week, accumulating enough data for a reliable training set. They used this data to train three different machine learning algorithms: a linear regression model, a neural network, and a statistical model called SARIMAX. The SARIMAX model showed the best performance, with a 90 percent match to observed results during a window of 50-plus hours. These results demonstrate that we can use machine learning in identifying candidate materials and suitable conditions to prevent degradation in perovskites. The next step will be to expand the experiments to quantify combinations of multiple environmental factors.