Computer-Aided Solvent Screening

Solvent effects play an essential role in determining chemical reaction rates and selectivities, but evaluating different solvents for a particular reaction has largely been a labor-intensive experimental process. A group of researchers led by Claire S. Adjiman and Amparo Galindo of Imperial College London has developed a computer-aided molecular design process that could save time by using a kinetic model to screen thousands of organic solvents for their effects on reaction rate constants (Nat. Chem. 2013, DOI: 10.1038/nchem.1755). Using a bimolecular nucleophilic substitution reaction as a test case, Adjiman, Galindo, and colleagues first calibrated the program using rate constants calculated with quantum mechanics for six solvents. They then looked at the effects of more than 1,300 solvents on the rate constant, performing full quantum mechanical computations for only nine solvents. The approach correctly predicted, for example, that nitromethane increased the reaction rate constant by 40% compared with acetonitrile. The screening framework can be adapted to evaluate reaction selectivity, as well as to incorporate factors such as solubility and toxicity, the researchers say.