The researchers from Chalmers University of Technology (Sweden) with participation of their colleagues from University of Birmingham (Great Britain) discovered the structure of one of the most high-potential material for solar energy industry – formamidine-leaded iodide created on the basis of созданного на базе halide perovskite. To do that, they used machine learning and advanced computer simulation allowing for the first detailed description of the phase of material, which earlier even the most perfect experiments could not explain. This discovery may become the key to developing a new generation of solar elements.
Perovskites are believed to be the main candidate for replacing the traditional silicon materials: they are lighter, cheaper and their light absorbing performance is better. At the same time, their major problem is instability – the material degrades fast when exposed to moisture, heat or light. To eliminate this deficiency, the scientists had to understand the atomic structure of the material and its behavior at the atomic level, especially under low temperatures.
The Chalmers team used the machine learning techniques to simulate the processes inside perovskite. Combining the computational physics with artificial intelligence, the researchers succeeded in simulations, which previously were impossible: the calculations lasted thousand times longer and engaged millions of atoms making the simulations very close to reality. Thanks to that, it became possible to observe the transition of formamidine-leaded iodide to the low-temperature phase – a special state, when atoms and molecules line up in a new way. Until now, scientists could only assume such a phase, but now the scientists could describe its shape, sustainability and dynamics.
To test the simulation results, the researchers cooled the samples down to minus 200 °C. The experiment confirmed the accuracy of the digital models: the molecules of the materials indeed maintained the semi-stable state, which explains their disposition towards degrading and helps to understand how to cope with it.
Now, that the structure of this phase is known, the scientists will be able to design the perovskite materials more accurately, e.g., to create sustainable mixes of different compounds combining their advantages. In the long term, it will speed up the development of solar batteries, which will be not only more efficient, but more flexible and time-roof. They will be fit for using in different appliances – from mobile gadgets to buildings’ facades.
