Researchers Use Artificial Intelligence To Optimize Spray-On Solar Cell Technology
Why are scientists so interested in perovskites, the organic-inorganic halide materials that can convert sunlight into electricity? Because they are cheap to produce, flexible, and, at least in theory, could be sprayed on to just about any surface — the exterior of buildings, windows, roofs, parking lots, bridges, highway barriers — to capture more energy from the sun than any other product known to science.
You really need to have a doctorate in chemistry and electrical engineering to understand what makes perovskites so fascinating. Suffice it to say to people of ordinary intelligence like myself, if the quest is to electrify everything in order to prevent the world from becoming a burned out cinder, perovskites may be exactly what we need to reach that goal.
But perovskites have several drawbacks. They are not as durable or efficient as traditional silicon solar cells. Yet any lack of efficiency could be more than made up for if the area available for producing electricity is increased by several orders of magnitude. Imagine if every building in every city around the world were able to able to supply electricity to the building occupants or to the local grid during daylight hours every day of every year. You see where this is going, right?
Chemistry is dauntingly complex. Take all the elements and all the compounds that can be made with those elements, mix them all together, and you are left with millions of possible combinations. Now add in liquid versus solid configurations, efficiency, longevity, and cost and the total number of combinations are in the billions. Sorting through them by hand would take decades if not centuries to complete.
Researchers at the University of Central Florida think artificial intelligence could help move that process forward faster. The team’s work so impressed the editors of the journal Advanced Energy Materials, they decided to make it the cover story of their December 13 edition.
The research was led by Jayan Thomas, an associate professor at the UCF’s NanoScience Technology Center. The team reviewed more than 2,000 peer-reviewed publications about perovskites and collected more than 300 data points and then fed into the AI system they created. The system was able to analyze the information and predict which perovskites recipe would work best, according to UCF.
“This can be a guide to design new materials as evidenced by our experimental demonstration,” Thomas says. “Our results demonstrate that machine learning tools can be used for crafting perovskite materials and investigating the physics behind developing highly efficient perovskite solar cells.” If Thomas and his team are correct, AI could help identify the best formula to create a world standard, leading to spray-on solar cells within a generation or two.
“This is a promising finding because we use data from real experiments to predict and obtain a similar trend from the theoretical calculation, which is new for PSCs. We also predicted the best recipe to make PSC with different bandgap perovskites,” says Thomas and his graduate student, Jinxin Li, who is the first author of this paper. “Perovskites have been a hot research topic for the past 10 years, but we think we really have something here that can move us forward.” And not a moment too soon.
Have a tip for CleanTechnica? Want to advertise? Want to suggest a guest for our CleanTech Talk podcast? Contact us here.
Latest CleanTechnica.TV Video
CleanTechnica uses affiliate links. See our policy here.
