Abstract

The study of molecule structures is a fundamental step for understanding a chemical process, which involves the rearrangement of atoms between reactants. In addition, conformational analysis can help study and predict the properties of compounds. For examples, protein folding is one illustration of its function. And the conformational behaviors of polymers could determine materials’ properties such as elasticity and strength. The geometry of a molecule is also closely associate with its molecular potential energy, which is a key factor influencing its stability and reactivity. Exploring potential energy surface(PES) of a molecule and locating its minima, which is called conformational searching problem, has been hold of great interest for a long time.

Conformational searching methods have been developed so far can be broadly divided on the basis of the conformers generating strategy employed: systematic and stochastic. Systematic method obtains new conformations by changing a molecule exhaustively according to setting criteria. A major drawback of systematic searching is that the number of structures generated increases dramatically as well as the running time, with the increase of the rotatable bonds, which is normally called combinatorial explosion. Stochastic methods prioritise speed to look for representatives but lack of full scanning of potential energy surface, and can then miss the global minimum. How to do fast and accurate conformational searching is a major challenge.

A new systematic conformational searching method was developed, based on an assumption that the structures of local minima are similar to diamond lattice, and then these conformers may be found by rotating structures lying on diamond lattice systematically. The conformations’ coordinates and energy can then be calculated with integer arithmetic and evaluated with simple energy equations, separately, this should fasten systematic searching. A series tests were conducted on acyclic alkanes and compared with a commonly used commercial software Maestro, and the results showed that this method was able to do systematic conformational searching exhaustively, accurately and fast.