Scientists have sought to a new artificial intelligence approach in a bid to design new catalysts. The approach is to identify catalytic properties of new catalysts that convert carbon dioxide into methane.
The approach involves tracking structure, size, and chemistry of catalytic particles in real reaction conditions. This helps scientists to identify the properties that correspond to best catalytic performance. The use of this information is for design of more efficient catalysts.
The ability to convert carbon dioxide into methane serves two objectives: unveil a sustainable non-fossil source of energy that is easy to store and transport.
The team of researchers involved are developing a machine-learning method to identify catalytic properties. The analysis of catalytic properties of newly designed catalysts is presented in a paper published in the Journal of Chemical Physics.
Use of Copper Atom Clusters to test Change of State
For the research part, chemists at the Argonne National Laboratory prepared clusters of copper atoms. Following this, the approach involved use of x-rays and mass spectrometry at Argonne’s Advanced Photon Source. This was to study the performance of various clusters in the reaction. The study also involved how oxidation state of copper atoms evolved at the time of reaction of carbon dioxide with hydrogen.
Meanwhile, copper displays promise to be a legit catalyst. It can reduce the temperature during carbon dioxide to methane reaction. The use of size-selective copper clusters serves to drive the reaction efficiently for desired outcome. Copper atom selectively produce only water vapor and methane sans channelizing of reactants.
Broadly, the idea involves two major challenges for implementation. Firstly, lack of knowledge of structure of prepared clusters. By the law of chemistry, small size of prepared clusters are likely to exhibit more variations in shapes and structures. The variation is exhibited even if the number of atoms in each cluster remains same.