An interesting way in which an excited atom can emit its excess energy has been brought to light by the experiments of Robinson and of Auger. If, for example, an atom is ionised in the K state, then it may emit a quantum of radiation of some line of its K X-ray spectrum by means of a transition of an electron to the K level, but as an alternative method it may emit an electron instead, thus leaving the atom doubly ionised. One such process might be represented as [L
I
→ K, L
II
→ ∝] and the energy E of the ejected electron would be given by E = K
abs
— L
Iabs
— L
IIabs
— δ, where δ is a small correcting term to take into account that the work required to remove an electron from an ionised atom is slightly greater than that necessary in the case of a normal atom. Processes of this kind are essentially different from those giving rise to radiation since two electrons instead of one are concerned in the transition. The entire process must be considered as occurring simultaneously, and, to take as an example the case already mentioned, it has no meaning to attempt to state whether it is an L
I
electron which goes to the K state, and an L
II
electron which is ejected or
vice versa
. Two points of interest in this phenomenon are the investigation of the magnitude of the correction term δ, and of the relative probabilities of the different types of transition. It will be seen later that the possible transitions are considerably more numerous than with single electron transitions which give rise to radiation. This phenomenon has been studied by Robinson by analysing the ejected electrons with a magnetic field. A thin layer of the element under investigation is placed in the position of the source in the well-known semi-circular focussing apparatus, and is irradiated with X-rays of sufficiently high frequency to be able to eject electrons from the K state. There then follows a further electronic emission from these ionised atoms in the manner already described. Both sets of electrons are recorded photographically, and the various groups show up as lines or narrow bands on the photographic plate. A difficulty inherent in the nature of the experiment is that the groups of homogeneous electrons become slightly diffuse in emerging from the target which must have a certain thickness in order to yield groups of reasonable intensity.
The corpuscular X-ray spectra of the raio-elements
