Catalysts are materials that facilitate chemical reactions and are essential for the widespread adoption of renewable energy technologies.
A way to identify high-performing catalysts is through molecular simulations which model the interaction of a catalyst with molecules involved in electrochemical reactions. These simulations (DFTs) can help predict the catalyst’s effect on the rate of a chemical reaction by estimating the strength of the interaction between the catalyst and the molecules known as the absorption energy.
The absorption energy is determined by simulating the interaction between the molecule and the catalyst surface to find the relaxed energy, or how tightly the molecule binds to the surface.
The rate of the chemical reaction is often approximated using a function of the absorption energy.
The aim is to develop machine learning models that can accurately estimate the relaxed energy of a catalyst by predicting the the atomic forces and positions during the relaxation process.
<aside> 🌍 Three key tasks the ML model needs to be able to do:
Effective and affordable solutions for storing renewable energy are crucial to meeting the world’s growing energy needs and combating climate change. As we increasingly rely on renewable energy sources like wind and solar, which generate intermittent power, we need to find ways to store the energy for times when demand is high.
A solution to this is to convert renewable energy into other fuels like Hydrogen to be stored and used when needed. This process requires efficient and low cost catalysts to drive the necessary chemical reactions.
Currently, quantum mechanical simulations (Density Functional Theory, DFT) are used to test and evaluate potential catalyst structures.
DFTs work well for predicting the electronic structure and properties of materials and molecules. It is based on the idea that the density of electrons in a system determines the behavior of the system, rather than the individual electrons themselves.
DFT calculations involve solving the Schrödinger equation, which describes the quantum mechanical behavior of a system using approximations to make the calculations more tractable. The final equations can then be used to predict various properties of the system like its energy, charge density, and electron distribution.
in summary →