Abstract
An investigation has been made of the molecular polarisation of t. butyl chloride and methyl cyanide in solution in non-polar solvents. The small range of dielectric constants encountered required the accuracy of their determination to be as high as possible and particular attention was given to the design of apparatus and the reduction of experimental errors. The polarisation at infinite dilution is independent of the method of extrapolation used provided the accuracy of the experimental results is high but a method which treats the separate physical constants of each solution is preferable to any dependent on a derived function (e. g. molecular polarisation). That proposed by Halverstadt and Kumler is satisfactory and is also very critical of the data examined. Based on the Debye Mosotti Clausius relationship and if the vapour phase data are correct, the apparent molecular polarisation of t. butyl chloride is greater in solution than in the vapour phase while that for methyl cyanide is less. This is ths only effect directly attributable to molecular shape and according to the theories of Frank and Hiagasi implies that with t. butyl chloride the axis of maximum palarisability is at right angles to that of the dipole. However, the results for t. butyl chloride must be considered anomalous since they do not confirm these authors' predictions that the total molecular polarisation at infinite dilution should increase with increasing dielectric constant of the solvent. The apparent specific volumes of the solutes vary from solvent to solvent and although the variations do not correlate with the variations of the polarisations they are similar for each solute, those for methyl cyanide being the greater. The experimental results have been examined by the application of equations proposed by various authors but in no case are the results correlated for all solvents. All the expressions agree in giving the highest polarisations for solutions in or containing dioxan which is probably due to hydrogen bonding, the exhaltation being associated with the orientation polarisation only. The normal paraffins are anomalous giving polarisations lower than those found in solvents of higher dielectric constant while mixed solvents give different polarisations from that in a pure solvent with the same dielectric constant. In general none of the equations can alone predict the experimental results and it seems probable that each solution must be treated as a separate problem.