The Temperature Dependence of σw
Of all the possible aspects of σw and its changes, the dependence upon temperature takes a special place. It is the only effect which can (and does) also effect the shift of an isolated molecule. Since σ w is defined as σw = σ liq - σ0 - σb, not only must the bulk susceptibility term be calculated at the appropriate temperatures, but in addition d σ 0 /dt must be considered. There is a special difficulty here, since until now it has been impossible to measure accurate absolute temperature dependencies. Virtually all known experimental data are relative to some chosen molecule so that one measures only the difference of two temperature dependencies. (For apparent exceptions see the end of this section) The first paper on the subject is that by Petrakis and Sederholm , who measured dσw/dt for eleven gases relative to internal CH4. Relative to this standard they found negative dσw/dt values (except for C2H4) of magnitudes varying from zero (HBr and C2H6) to about -0.02 ppm/100 °C (for SiMe4). Buckingham  has commented on these data; he pointed out that Petra- kis and Sederholm’s data refer to samples at 5–10 atm pressure, and that under these circumstances the intermolecular temperature effects are in the same order of magnitude as those reported by Petrakis and Sederholm. Petrakis and Sederholm tried to explain their results in terms of vibrational excitation. They indicated that stretching modes can never contribute much, (explaining the zero-effect for HBr) and that only certain low frequency torsonial modes could possibly be responsible.
KeywordsBulk Susceptibility White Phosphorus Mass Distribution Function Spin Side Band Intermolecular Effect
Unable to display preview. Download preview PDF.