In a recent paper, an account was given of new measurements on the excitation potentials of lithium metal. Here we shall describe the corresponding results for the next element, beryllium, attention being again mainly focussed on the excitation of the K-radiation. The purpose of these investigations may be defined as an attempt to determine the effect of the metallic binding on the atomic constants of these light elements. Thus, in Paper I, it was shown that the minimum energy needed for the excitation of the K-radiation of lithium metal is considerably less than the K-ionization potential of the free atom. This low value seemed at first sight inconsistent with the facts of the excitation of X-rays in heavier elements. But, if we adopt a more direct comparison, it is easy to see that the observation that this minimum excitation potential agrees numerically with the energy required for a K→L switch in the free atom implies simply that the effect of the binding here is small. We shall show that for beryllium the effect of the binding is definitely observable; its magnitude, about 20 volts, is considerable compared with the absolute K-excitation energy, about 100 volts. In the attempt to correlate these observations, it became apparent that the theory of Paper I is incomplete. We shall therefore give a corrected treatment in its application to lithium and beryllium metals, and it will be seen that, on certain assumptions, the new theory seems capable of accounting in detail for the observed break-potentials. It has also been found that there is an appreciable difference between the minimum K-excitation potentials of beryllium in the form of a metal and of a polar compound. These problems are closely related to work done on the influence of chemical binding on the absorption-edges and lines in the region of harder X-rays. Thus Bäcklin† studied the K
α
line of sulphur, and other elements when used in the form of various compounds, and found shifts of the order of 10 volts. Such investigations open up the question of how far the spectroscopic data of X-rays from solid targets are representative of atomic constants.
The excitation potentials of light metals. II.—Beryllium
