Energy distributions of 1.12-MeV gamma rays Compton scattered byK-shell electrons of gold, lead, and thorium

Abstract

Measurements are reported of the pulse-height distributions of photons of initial energy 1.12 MeV after Compton scattering through 60' and 100' by K-shell electrons of Au, Pb, and Th. The energy distributions of scattered photons, obtained in the case of Au and Pb targets by a deconvolution of the detector response function, show a rise for energies lower than about 0.25 MeV, a broadening of the Compton line, and a negligible shift of the Compton peak. On account of natural radioactivity of the Th target, the corresponding pulse-height distributions are not precise enough to make the deconvolution procedure worthwhile. The energy distributions are compared with theoretical calculations based on the nonrelativistic impulse approximation. It is clear that a relativistic calculation incorporating the e6ects of electron binding in intermediate states is required. In the present experiment, a fast-slow coinci- dence system was used with resolving times of 30 nsec and 2 p, sec, respectively. Pulse-height analysis of anode pulses from the y detector was made with the help of a 20-channel analyzer gated by the output of the coincidence circuit. The lin- earity of pulse height was checked not only through the singles spectra of the y rays of different en- ergies but also through the spectra obtained in the coincidence mode. Annihilation quanta from a Na source, and 1.17- and 1.33-MeV y rays from a ~Co source were used for the spectral measure- ments in the coincidence mode. The photopeaks obtained in the coincidence mode agreed 'with the positions of the corresponding peaks in the singles spectra. Therefore, useful linearity and gain stability checks could be made more frequently in the singles mode than would have been possible in the coincidence mode. If a check showed a pulse- height shift of more than 2%, the corresponding data were discarded. Pulse-height distributions of the scattered y rays were studied at 60' and 100' scattering angles. In each case, the pulse-height distribution of false events was determined and subtracted from the measured distribution in order to obtain the true distribution. In all cases, the chance coineidenee rate was less than 1% of the total rate and was con- sidered negligible. The rate in each channel cor- responding to Nz of Eq. (2) in the preceding paper was negligible in the case of the high-Z targets used here. For the reasons mentioned in Sec. II of the pre- ceding paper, the distribution of false events was determined with an equivalent aluminum target in the case of targets of thickness less than 30 mg/cm'. In the case of the thicker lead targets, an equivalent copper target was also used. A cor- rection-had to be made in the case of thorium for its radioactivity. As a result, the errors in the thorium measurements are large.