An experimental test of the fermi-teller «Z-law»

Fermi and Teller predicted, in their classic paper on the slowing-down process of negative mesons, the relative probabilities with which the latter should be captured by the different atoms of a chemical compound. They concluded that theseatomic capture probabilities should be proportional to the nuclear charges,Z,i of the atoms considered. The experiments here described were carried out to check this prediction. They are based on the fact that negative muons reach rapidly theK-orbits around the nuclei of the atoms of the material in which they are slowed down, and disappear from these orbits (either by decay or absorption) at rates which are characteristic for the nuclei in question. Thus the time distribution of decay electrons emerging from a chemical compound in which muons are brought to rest presents in general as many exponential components as there are constituents ; their intercepts at time zero (entry of the muons into the material) are directly proportional ton,ipi where ni=fraction of component of atomic numberZ,i pi=capture probability into the mesicK-shell of atom of same constituent. Neglecting the possibility of transfer of muons between neighbouring atoms, thesepi are equivalent to the capture probabilities predicted by Fermi and Teller. The present experiments show that at least in insulators (A12O3, P2O5, SiO2, CC14, KOH, KHF2, p-dichlorobenzene) the predictions of Fermi and Teller do not hold. No weighting ofni withZi seems to arise, the lighter element being even preferred in some cases. As an explanation, it is pointed out that the energy loss to lattice vibrations, not considered by Fermi and Teller, may be an important mechanism in insulators for negative particles of nearly zero kinetic energy, i.e. at the point of being captured.