Unlike some time-consuming and complex models, the new model is fast, easy to use
The key to developing more efficient and user-friendlyfor the user of the model, it was the replacement of complex and computationally expensive parameters requiring quantum mechanical calculations with simpler and chemically interpretable descriptors of the signatures of the analyzed molecules. They provide important data on the most significant chemical moieties in materials that affect the PCE by generating information. In the future, it can be used to develop improved materials.
A new approach could help significantly speed upthe process of developing more efficient solar cells at a time when the demand for renewable energy and its importance in reducing carbon emissions has never been greater. The results were published in the journal NatureComputational Materials.
After decades of using silicon, whichis relatively expensive and not flexible enough, more and more attention is being paid to organic photovoltaic cells (OPV, organic photovoltaics), which are cheaper to produce, also more versatile and easier to recycle.
The main problem is sortinga huge amount of potentially suitable chemical compounds that can be synthesized (adapted by scientists) for use in OPV. Researchers have tried to use machine learning to solve this problem before. However, many of these models were time consuming, required significant computing power, and were difficult to reproduce. And crucially, they did not provide sufficient guidance to experimental scientists who were working on new devices for green energy.
Now the work led by Dr. NastaranMeftahi and Professor Salvi Russo from RMIT University, together with the team of Professor Udo Bach from Monash University, have successfully solved many of these problems.
Most other models useelectronic descriptors, which are complex, computationally intensive, and defy chemical interpretation. This means that the experimental chemist or scientist cannot draw ideas from these models to design and synthesize materials in the laboratory. The collaboration of scientists led to the creation of the BioModeller program, which formed the basis of a new open source model. Using it, the researchers obtained reliable and predictable results and, among other things, quantified the relationship between the molecular signatures under study and the effectiveness of future OPV devices.
Read more
Unknown animals resembling sponges found in the ice of Antarctica
The Brunt shelf in Antarctica is collapsing at a speed of 5 meters per day
Abortion and science: what will happen to the children who will give birth