There is a picture of the cold quantum world and high-temperature metal. Jose A. Garrido Torres/Columbia Engineering is credited.
It's important to extract metals from oxides at high temperatures for both production and recycling. Researchers have been looking at new approaches to developing "greener" processes because current processes are very carbon intensive. This work requires expensive reactor and it has been difficult to do in the lab. Computational methods that can accurately predict oxide reactions at high temperatures are currently not available, so building and running computer simulations would be an alternative.
A Columbia Engineering team has developed a new computation technique that can accurately predict the temperature of metal oxides to their base metals. Their approach is more accurate than simulations of temperature effects using quantum chemistry methods and is as efficient as conventional calculations at zero temperature. Alexander Urban is an assistant professor of chemical engineering.
Developing alternatives for established industrial processes is very cost-intensive and time-Consuming if we are to transition to a more sustainable future. A bottom-up computational process design that doesn't require initial experimental input is an attractive alternative but has not been realized. This is the first time that a hybrid approach, combining computational calculations with artificial intelligence, has been attempted for an application. It's the first demonstration that quantum-mechanics-based calculations can be used for high-temperature processes.
At very low temperatures, quantum-mechanics-based calculations can accurately predict the energy that chemical reactions require or release. The zero-temperature theory was augmented with a machine-learning model that learned the temperature dependence from publicly available high-temperature measurements. They designed their approach to extract metal at high temperatures to predict the change of the "free energy'' with the temperature.
The paper's first author, who was a research scientist in Urban's lab, said that free energy is a key quantity of thermodynamics and can be derived from it. We expect that our approach will be useful to predict melting temperatures and solubilities for the design of clean electrolytic metal extraction processes that are powered by renewable electric energy.
Nick Birbilis, the deputy dean of the Australian National University College of Engineering and Computer Science and an expert for materials design with a focus on corrosion durability, was not involved in the study. Over the past century, most of the human effort and sunken capital has been in the development of materials that we use every day, and that we rely on for our power, flight, and entertainment. Machine learning is a critical development for future materials design because of the slow and costly nature of materials development. Machine learning and artificial intelligence need models that are mechanistically relevant and interpretable. This is what the work of Urban and Garrido shows. One of the first times, the work links atomistic simulations on one end engineering applications on the other.
The team is working on extending the approach to other temperature dependent materials properties that are needed to design electrolytic metal extraction processes that are carbon-free and powered by clean electric energy.
The study uses machine learning to predict chemical reactions at high temperatures.
More information: "Augmenting zero-Kelvin quantum mechanics with machine learning for the prediction of chemical reactions at high temperatures," Nature Communications.
Nature Communications is a journal.
The new approach predicts chemical reactions at high temperatures on December 1st.
The document is copyrighted. Any fair dealing for the purpose of private study or research cannot be reproduced without written permission. The content is not intended to be used for anything other than information purposes.
