Hitherto, the experiments which have been carried out oil the Raman effect at low temperatures have been performed on substances of which the Raman spectra were unobtainable, or only obtainable with difficulty, at ordinary temperatures. Thus Daure* has obtained the Raman spectra of liquid methane, ethylene, ethane, propane, and ammonia, while McLennan and his co-workers have studied those of liquid hydrogen, oxygen, nitrogen, methane, helium, nitrous oxide and solid carbon dioxide. The primary object in these researches was to obtain the spectrum of the substance, and the influence of temperature on the spectrum has received very little attention. While the present work began as an attempt to obtain the Raman spectrum of nitrogen tetroxide, it has become increasingly evident in the course of it that the study of Raman spectra at very low temperatures may well prove to be a very fruitful field of research. It is clear that there will be at least two distinct possible effects of a change in temperature of the scattering substance on its Raman spectrum. If the change in temperature causes a change in the molecular structure of the substance (such as association of simple molecules into more complex aggregates, or a change in the crystalline form of the substance), then one may expect the appearance of new lines in the spectrum. Secondly, a change in temperature will result in a new distribution of the molecules in the various rotational and vibrational energy levels ; the effect of this will be to alter the character of the individual lines and bands in the spectrum without, however, giving rise to any new lines. The following paper gives a description of a new and very simple apparatus for the observation of Raman spectra at low temperatures, together with the results and discussion of the preliminary experiments on a few simple substances.
Experiments on the raman effect at very low temperatures
