Direct Measurements of the Mean Lifetime of Atomic States

Determining the mean duration of an excited state is of great importance for a number of questions in fundamental physics. At present we already have a fairly rich arsenal of methods, both direct and indirect, which enable us to calculate the mean lifetime; but each of these methods has its own advantages and shortcomings and can be applied with some success only in a definite range of numerical values of \(\tau\). For this reason the development of a new direct method for determining the duration of an excited state of an atom in the thermal diffusion of a gas should be welcomed in every way. The idea of this method is not new. As early as 1913 Dunoyer attempted to determine in this way the duration of the excited state for the \(D\) line of Na. It is known that this attempt was unsuccessful, and the authors believe that the chief cause of the failure of Dunoyer’s experiment was the unsuccessful choice of the object for the given method.

In order that an atom, during thermal diffusion, may traverse between absorption and emission any appreciable length of path, it must have a sufficiently large \(\tau\). This is precisely not the case for the resonance lines of sodium, and therefore the authors chose cadmium as the object—namely, its \(2^3P\) state, for which Ellet had obtained, by the depolarization method, the relatively large value \(\tau = 2.3 \cdot 10^{-6}\) sec.

It is true that cadmium has another, more intense line corresponding to the excited state \(2^1P\), but it has a much larger absorption coefficient and, under the conditions of the experiment, was, as the authors indicate, completely absorbed in the thin layer of cadmium glass formed in the quartz vessel in which the resonance radiation was taken.

The basic arrangement of the apparatus is the same as in Dunoyer’s experiment; i.e., the stream of atoms, the exciting resonance radiation, and the line of observation are arranged along three mutually perpendicular directions.

At first the qualitative effect of the extension of the glow beyond the boundary of the excitation region could be noted on a photographic plate already in a comparatively crude preliminary setup. Subsequently the authors undertook the construction of an improved apparatus suitable for purposes of quantitative calculation. The photographs obtained with this apparatus were processed on a microphotometer, and by simple calculations the value \(\tau = 2.5 \cdot 10^{-6}\) sec was obtained, which agrees very well with the above-mentioned value obtained by the depolarization method. (Koenig and Ellett, Direct Measurement of Mean Lives of Atomic States, “Phys. Rev.,” vol. 39, 1932.)

B. Sveshnikov