The Nobel Prize in chemistry was awarded this morning to three theoretical chemists who created computer models that combined quantum and classical physics to describe chemical reactions. Martin Karplus from the University of Strasbourg in France and Harvard University, Michael Levitt from Stanford University and Arieh Warshel from the University of Southern California in Los Angeles, published their work in the 1970s, and since then, chemists have been using their models to predict how chemical reactions occur.
Before theoretical chemistry, experiments told chemists what products existed in the beginning of a reaction and the end of a reaction, but not much about what happened in between. “It’s like seeing the actors before Hamlet and all of the dead bodies after, and then you wonder what happened in the middle,” said Sven Lidin, chairman of the Nobel Committee for Chemistry 2013.
That’s why, in the 1970s, scientists began to model chemical reactions as a way to predict what happened in the middle. However, they ran into one important barrier: Quantum physics accurately described the process, but the computers couldn’t handle the complex calculations that were required. Using classical mechanics made those chemical interactions easier to calculate, but it didn’t provide accurate predictions.
In order to get a really good look at those in-between moments, Karplus, Warshel and Levitt combined the accuracy of quantum mechanics with the computational ease of classical mechanics. Karplus and Warshel created a computer program that used quantum physics to predict the behavior of free electrons and classical physics to predict other electrons and all the atomic nuclei. They published their work in 1972, and according to Nobel Committee member Gunnar Karlström , it was “the first time that anyone ever merged the classical world with quantum chemistry.” Warshel and Levitt expanded on these efforts in the mid-1970s to model how proteins functioned.
In both cases, the Nobel laureates used wave-based quantum calculations for the central part of the reaction, and particle-based classical calculations to model the electrons and nuclei that weren’t interacting.
This strategy “gives a greater resolution to parts that are directly involved and less resolution to farther away parts of the reaction,” said Karlström. That way, scientists can get an accurate reading of the bits that matter, and don’t have to worry about the bits that don’t.
These days, all kinds of chemists use this hybrid modeling, from materials scientists to pharmaceutical developers. According to Lidin, every pharmaceutical company has a theoretical division that uses theory to predict what kind of molecules will target proteins. This allows chemists to predict what experiments will show the most important results, says Lidin. “We save a lot of money, a lot of time, and a lot of effort by doing the theoretical homework first.”