Researchers from Nankai University in China have delved into the potential of ChatGPT in offering insights for material science, specifically in improving perovskite solar cells.
The study, led by corresponding author T. Jesper Jacobsson, highlighted the ability of ChatGPT 3.5 to propose hypotheses for enhancing the efficiency of perovskite solar cells by reducing surface recombination.
The research team utilized ChatGPT as a tool for generating suggestions on molecules that could aid in surface passivation for hybrid perovskite solar cells with a p-i-n architecture. By inputting specific criteria such as accessibility, affordability, and safety into the model, the scientists received recommendations, including the water-soluble polymer polyallylamine (PAA).
Following the suggestions, the team conducted real-world experiments, applying PAA to perovskite films before assessing the impact on device performance. The results revealed an increase in efficiency by approximately 2 percent units, showcasing the potential of human-AI collaboration in scientific research.
The findings were detailed in a publication titled The use of ChatGPT to generate experimentally testable hypotheses for improving the surface passivation of perovskite solar cells. The research involved collaboration between Nankai University in China and Linköping University in Sweden.
This study underscores the growing potential of artificial intelligence in advancing materials science research, offering scientists valuable insights and avenues for exploration in enhancing solar cell technologies.
