Spinning Electrons and the Structure of Spectra

Abstract

GOUDSMIT and Uhlenbeck have suggested that the structure of multiple lines in spectra can be made much more comprehensible if it is assumed that the electron itself has a magnetic moment of one Bohr magneton (eh/4mc) and an angular momentum of one-half unit (h/4), which can orient itself in various ways with respect to the normal to the electron orbits. Goudsmit and Uhlenbeck point out (Naturwissenschaften, Nov. 20, 1925) that to obtain such a magnetic moment from the rotation of a Lorentz electron about its axis, there would have to be a peripheral velocity greatly exceeding the speed of light. It can also be easily seen that the magnetic energy with its resulting mass would be much greater than the electrical energy. These difficulties do not seem to be easy to overcome by any small modification of the classical model of the electron. It appears, however, that an entirely different picture of electronic structure, based on the idea of light quanta, is capable of giving a consistent explanation of the energy of the electron, and also of predicting the values of some of the electronic constants.