A metal-organic framework is a material that has small pores. These pores give MOFs record-breaking internal surface areas, which make them extremely versatile for a number of applications, such as separating petrochemicals and gases, mimicking DNA, producing hydrogen, and removing heavy metals from water are just a few examples.
Professor Berend Smit's group uses machine learning to make breakthrough discoveries in the discovery, design, and even categorization of the ever-increasing number of MOFs.
Smit and his colleagues have developed a machine- learning model that predicts the heat capacity of MOFs. Smit says that it's about very classical thermodynamics. The amount of energy needed to heat up a material is not known. Engineering calculations have assumed that all MOFs have the same heat capacity because there is no data available.
If there is no data, how can one make a machine- learning model? That looks hard to believe.
The most innovative part of the work is a machine-learning model that predicts how the local chemical environment will change in the future.
Smit says that the heat capacity can be related to the vibrating things. A very expensive quantum calculation used to give us a single heat capacity for a single material, but now we get up to 200 data points on the vibrations. We used 40,000 data points to train the model on how these vibrations depend on their chemical environment.
The model was checked against experimental data. The results were disappointing until we realized that the experiments had been done with solvent in their pores. The results were in agreement with our model's predictions after we re-sized some MOFs and removed the synthesis solvent.
Moosavi says that their research shows how artificial intelligence can solve problems. Artificial intelligence gives us the ability to think about our problems in a different way.
Engineers at Heriot-Watt University were able to demonstrate the impact of the work.
Smit says that they used quantum simulations, machine learning, and chemical engineering. The results show that the energy cost of the carbon capture process can be lower than we thought. Our work has a huge impact on the viability of solutions to tackle climate change.
Nature Materials contains the research.
More information: Seyed Moosavi, A data-science approach to predict the heat capacity of nanoporous materials, Nature Materials (2022). DOI: 10.1038/s41563-022-01374-3. www.nature.com/articles/s41563-022-01374-3 Journal information: Nature Materials
All Rights Reserved © 2024
