Oak Ridge National Laboratory supercomputers support Nobel Prize-winning research
Newswise – In October, a scientist whose research was supported by supercomputer modeling and simulation efforts at the U.S. Department of Energy’s (DOE) Oak Ridge National Laboratory (ORNL) shared the Nobel Prize for chemistry 2021.
The award-winning duo, Benjamin List of the Max-Planck-Institut für Kohlenforschung and David MacMillan of Princeton University, have developed a new, highly selective way of constructing chiral molecules, molecules that mirror each other others. This was achieved through asymmetric organocatalysis, a process in which an organic molecule serves as a catalyst that results in a chemical transformation into a desired product.
Typically, the synthesis of catalysts is tedious experimental work. This requires careful planning and a working hypothesis on the reaction mechanism to fine-tune the activity of the catalyst. However, the process is much more complex in real life than on paper because the details of the reaction mechanism are not known.
The team led by List at the Max-Planck-Institut needed computer modeling of the phenomenon to complement their ongoing physical experiments in the laboratory. That’s where the Oak Ridge Leadership Computing Facility (OLCF) and ORNL computing researcher Dmytro Bykov came in. Bykov used the Summit 200 petaflops, OCLF’s flagship supercomputer, to simulate the team’s new catalysts that turn molecules into specific end products. The modeling helped the team determine if the new catalysts would be effective.
“With Summit, I could build a molecule, optimize it, and converge to a stable conformation,” Bykov said. “Then I could take a molecule that the team had in their reaction, place it in this catalyst, and see how it passed this barrier in my simulation.”
In particular, the team studied the role of catalysts in the conversion of inert molecules called olefins. Olefins are among the most abundant chemical compounds in nature and are usually obtained from crude oil. They are of interest in biochemistry because they are ideal for use as precursors to drug compounds and other chemicals.
Catalysts work by lowering the energy barriers of a chemical transformation and thus directing a molecule towards particular products. However, products can be chiral, meaning they share the same composition and similar structure, but are reversed left or right of each other. Catalysts can force molecules into a left or right conformation, and these different positionings can have completely different effects, for example, in a human body.
“In a pairing, the molecule on the left could be a drug, and the right molecule could even be a poison in some extreme examples,” Bykov said. “For this reason, it is important to synthesize them in a specific direction.”
Without the right catalysts, scientists could end up with a mix of left and right molecules that could have deleterious effects. Organocatalysis makes it possible to circumvent this problem.
“The team had theorized the most promising catalysis candidates, but they needed details about reaction mechanisms that are difficult to determine in the lab,” Bykov said. “For example, computer modeling can find transition state structures, reaction barriers and alternative pathways and ultimately illuminate the nature of substrate activation.”
Bykov’s simulations gave the team a way to figure out how to stress a molecule in a specific way to bring it closer to the desired product.
“In an experiment, the original assumption might not work, and you might get something that you think should have responsiveness, but it doesn’t,” Bykov said. “Computational modeling can help direct experimental efforts so that we can synthesize these new and important organic catalysts in a more predictive and rational way.”
The OLCF is a user facility of the DOE Office of Science located at ORNL.
Related Post: Nobuya Tsuji, Jennifer L. Kennemur, Thomas Buyck, Sunggi Lee, Sébastien Prévost, Philip SJ Kaib, Dmytro Bykov, Christophe Farès and Benjamin List, “Olefin Activation via Brønsted Asymmetric Acid Catalysis”, Science 359, no. 6383 (2018): 1501–05, doi: 10.1126/science.aaq0445.