Can nematic transitions be predicted by atomistic simulations? A computational study of the odd-even effect

In this work we attack the problem of predicting the nematic-isotropic transition temperature (TNI) of mesogenic molecules with molecular dynamics (MD) computer simulations and we set ourselves two objectives. The first is to examine the feasibility of calculating TNI using computer simulations, a task until recently judged impossible in the literature and we choose as test case the first three homologues of the phenylalkyl-4-(4'-cyanobenzylidene)--aminocinnamates mesogenic series. The second is to understand the large odd-even effect demonstrated many years ago by G.W. Gray et al. in the above series using the simulation results. We have succeeded in showing that the nematic transition temperature of liquid crystals can be predicted from scratch within 10-15 degrees of the experimental result. We have showed that this require simulation runs of various tens of nanoseconds, an order of magnitude more than typical simulations currently published, in view of the relatively long molecular tumbling times in these systems. We have also clarified the origin of the large odd-even effect in terms of shape distributions.