Climate change is posing a critical threat to the planet. Innovative approaches are required to reduce the emissions of carbon dioxide (CO2) that leads to climate change. The Lawrence Livermore National Laboratory (LLNL) has lately come up with a significant innovation in the field. It has developed a machine learning model to unravel the complexities of CO2 capture at an atomic level.

The LLNL team has developed a specific machine-learning model. It provides a detailed understanding of how amine-based sorbents capture CO2. It is learned that the research has the potential to enhance the efficiency of direct air capture (DAC) technologies. It may help in reducing the excess CO2 that fuels global warming.

The U.S. Department of Energy forecasts that non-renewable sources will continue to dominate national energy production until 2050 despite the current global efforts to transition to renewable energy. It means that there is an urgent need for developing renewable energy sources. It is simultaneously also important to improve technologies that capture and store CO2 emissions.

Amine-based sorbents have shown great promise. The substances are cost-effective and efficient too at binding CO2 even in very low concentrations. It is being considered as a viable solution DAC implementation in large-scale.

The machine-learning model developed by LLNL scientists lately has of course shed some lights on the fundamental processes involved. It has revealed that CO2 capture by amines involves forming a carbon-nitrogen chemical bond between the amino group and CO2. The proton transfer reactions are crucial for forming the most stable CO2-bound species. It is also significantly influenced by quantum proton fluctuations.

Lead author of the study, Marcos Calegari Andrade, said that the new method can be extended to amines with various chemical compositions and machine learning can help in understanding the fundamental chemistry involved in CO2 capture under the realistic and challenging conditions.