Catalysis is fundamental to chemical engineering, speeding up processes at large scales.
Image Credit: ArtemisDiana/Shutterstock.com
Many breakthroughs in industrial chemistry have been made due to innovation with catalysts, and catalysts will likely be the key to current challenges. Researchers have found ways of using catalysts as novel solutions to problems facing industry, medicine and the environment. In the coming years, these studies may develop into full-scale applications.
Machine LearningDesigning new catalysts has always been a trial-and-error process as they tend to be incredibly complex structures with lots of varying properties. Due to this complexity, it is incredibly difficult for researchers to predict how small changes to the structure will affect performance. Instead, they must depend on practical trials of each variation, which makes the research process expensive and time-consuming as many possibilities are manufactured, tested and then dismissed.
However, advances in machine learning offer a new approach to catalyst design. As machine learning matures, many industries are finding ways to problem-solve with AI, and catalysis research is no exception. Specially trained programs are able to spot patterns and correlations in ways that humans never could and extrapolate data to make predictions. When applied to catalysis, machine learning can narrow down potential designs without practical testing, saving time and money.
Researchers from China recently demonstrated the potential for machine learning to be used for catalysis research. The team was researching solid-oxide fuel cells and aimed to design a perovskite oxide catalyst that could speed up oxide reduction in these cells. Having trained the AI with a curated data set, they could use it to find correlations between certain key properties and catalyst performance. The researchers hope that this method will be used for similar projects, accelerating the development of catalysts in the future.1
Enzyme Engineering Semiconductors eBook Compilation of the top interviews, articles, and news in the last year. Download a free copy
Machine learning has also seen use in enzyme engineering. Enzymes are biological catalysts, and in recent years companies have employed genetic engineering to design enzymes. By manipulating amino acids in microorganisms, they can produce enzymes that catalyze specific reactions. Despite being a young technology, enzyme engineering sees use in medicine and industrial processes.
However, enzyme engineering has its limitations. Unnatural enzymes tend to be less robust than unmodified enzymes, and manipulating genes can be an unpredictable process. Some enzyme designs have escaped researchers, but machine learning could change this. Moderna has recently developed a method for enhancing medical enzyme engineering with machine learning. 2 This technique could potentially be adapted for use in industry, allowing biological catalysis to break new frontiers.3
Alternatives to Precious MetalsMany chemical processes depend on precious metals as catalysts. An obvious example is using platinum in catalytic converters, but many manufacturing processes also need precious metals to catalyze reactions. This is clearly a problem because these metals will become more expensive and harder to source as the supply dwindles and demand keeps increasing as new applications are found. This will become a major barrier to catalysis in the future, so researchers have begun the search for an alternative. Progress is slow, but there have been some breakthroughs.
A collaboration between Princeton University, Rice University and Syzygy Plasmonics Inc. has shown that photocatalysis can be used in place of precious metal catalysts for some reactions. Photocatalysts are catalysts that can be accelerated by light, as the energy from photons excites electrons and creates free radicals. The concept of photocatalysts has existed for around a century, though it has yet to see commercial use.
This project used an iron-based photocatalyst to convert ammonia into hydrogen for use in fuel cells. Previously, this reaction used ruthenium, which is far harder to acquire than iron. The achievements of this group are significant not only for this specific process but precious metal catalysis in general. It brings the industry one step closer to replacing precious metals and developing photocatalysis.4
ElectrocatalysisElectrocatalysts are catalysts used in electrochemical reactions (reactions induced by an electric current). Many electrochemical processes already utilize electrocatalysts, but researchers are trying to improve these catalysts by making them more efficient and stable.
One example of their use is hydrogen fuel cells. The separation of hydrogen and oxygen molecules is an electrochemical process, and it currently requires too much energy to make fuel cells a reliable form of battery. For decades, electrocatalysts that can increase efficiency have been studied in the hopes of achieving a breakthrough in fuel cell technology. This would be a massive step forward for green energy by solving the issue of energy storage.5
Catalytic Waste ManagementPlastic waste is another environmental issue that catalysts may be the key to solving. It is well known that plastic’s incredibly long degradation time makes it problematic to dispose of. The methods currently in use, such as landfills, incineration and mechanical recycling, all come with significant drawbacks, and efforts to reduce plastic use have seen no success as waste production continues to increase. To prevent this plastic waste crisis, some researchers have turned to catalysis.
Numerous studies have found methods of dealing with various plastics by using catalysis. Not only do these methods help eliminate waste, they also produce beneficial products such as fuel or chemicals that can be reused for industrial processes. This justifies investment in such technology from both an economic and environmental standpoint.6Continue reading: A New Way to Visualize Electrochemical Reactions at the Nanoscale
References and Further ReadingHongliang, X. (2022) Catalyst design with machine learning. Nat Energy, 7
, pp. 790–791. https://doi.org/10.1038/s41560-022-01112-8
Giessel, A. Dousis, A. Ravichandran, K. et al. (2022) Therapeutic enzyme engineering using a generative neural network. Sci Rep., 12, p. 1536. https://doi.org/10.1038/s41598-022-05195-x
Scherer, M. Fleishman, S.J. Jones, P.R. Dandekar, T. Bencurova, E. (2021) Computational Enzyme Engineering Pipelines for Optimized Production of Renewable Chemicals. Front. Bioeng. Biotechnol., Sec. Synthetic Biology. 10.3389/fbioe.2021.673005/full
Yuan, Y. Zhou, L. Robatjazi, H. Bao, J.L. Zhou, J. Bayles, A. et al. (2022) Earth-abundant photocatalyst for H2 generation from NH3 with light-emitting diode illumination. Science, 378, p. 6622. https://www.science.org/doi/10.1126/science.abn5636
Ren, X. Lv, Q. Liu, L. Liu, B. Wang, Y. et al. (2020) Current progress of Pt and Pt-based electrocatalysts used for fuel cells. Sustainable Energy and Fuels, 1. https://pubs.rsc.org/en/content/articlelanding/2020/se/c9se00460b
Martin, A.J. Mondelli, C. Jaydev, S.D. Perez-Ramirez, J. (2021) Catalytic processing of plastic waste on the rise. Chem, 6(7). https://www.sciencedirect.com/science/article/pii/S2451929420306380
Disclaimer: The views expressed here are those of the author expressed in their private capacity and do not necessarily represent the views of AZoM.com Limited T/A AZoNetwork the owner and operator of this website. This disclaimer forms part of the Terms and conditions
of use of this website.