Abstract
By timing on the earliest component of the anode current pulse of the photomultiplier, the time resolution obtained with NaI(TI) at 300°K can be made somewhat superior to that reported for pure NaI at 77°K. For example, the time distribution for x rays (31. 6 keV) relative to the conversion electrons from a Cs137 source, shows a full width at half maximum of 1.6 nsec and a slope corresponding to a half life T4324091 = 0.63 nsec. This implies that τ /R is much smaller than commonly believed, where τ is the mean life of the scintillation and R is the average number of photoelectrons per event. This small value is not due to a small τ because only the usual slow component is observed when the shape of the scintillation is measured by a probability technique. Therefore R must be larger than expected; the observed time resolution requires at least 8.7 photoelectrons/ keV. This is consistent with the value 9.3 p.e./ keV inferred from the pulse-height resolution obtained with artificial light pulses which are equivalent in amplitude to a 662-keV event in NaI(Tl). It is also consistent with the value (9. 1 p.e./keV) obtained by combining the absolute efficiency of NaI(Tl) measured by Van Sciver and Bogart with the quantum efficiency that RCA lists for the bi-alkali photocathode.