No abstract

No abstract

No abstract

Accurate measurements were made of the detailed shapes of y-ray yield curves from semithick aluminum targets over the 992-keV resonance. Detailed information from these targets was possible through the use of a new electronic device which allowed data to be taken while incrementing the target voltage to provide energy changes and simultaneously correcting for energy fluctuations in the beam. To provide pure targets, fast evaporation rates were used with pressures usually 3X10 -7 Torr or less. During runs the pressure was usually 3X10 -9 Torr or less. No contamination effects appeared. Experimental results were compared to curves computed by Monte Carlo techniques for an assumed collision spectrum in which the contributions of electrons from each shell were computed using the Born approximation. These calculated curves gave closer fits to the experimental data than previous curves for which all electrons were assumed to be free and stationary. The experimental results departed somewhat from those calculated because nonuniformities in target thickness were not completely eliminated. Some additional results on thick targets show that surface contamination can take place if O2 and N2 gases are introduced into the vacuum system.

Problems encountered in measuring narrow resonance widths have been studied utilizing several wellknown resonances in the Al 27 (^,7)Si 28 reaction. Previous measurements had been limited by imperfections of the gap in the electrostatic analyzer. To correct these defects, the surface of each deflection electrode was lapped to a mirror finish and its surface contour was examined by observing the fringe pattern of an optical interferometric device. Another similar device was used to position and measure the gap of the assembled analyzer. The result is a gap constant to ±0.22 fi over a 4-mm vertical height for 9/10 of the analyzer length. The proton beam from the analyzer was calibrated for energy spread by using the measured profile of the gap and the resonance of known width near 774 keV. Thick-and thin-target-yield curves taken over the resonances near 992 and 1317 keV were analyzed, using a Monte Carlo technique, which accurately ac counts for the proton energy loss. Despite significant improvement in target-chamber vacuum conditions and resolution of the electrostatic analyzer, target and analyzer uncertainties limited the accuracy of the measured widths to Г = 90±35 eV and Г = 85±45 eV for the 992-and 1317-keV resonances, respectively. This study shows that three problems are likely to cause errors in resonance-width and resonance-position measurements. They are: (1) target contamination, (2) treatment of energy loss process, and (3) beamanalyzer imperfections. Examination of methods used previously indicate that these errors tend to com pensate. Since compensation under all conditions cannot be expected, this work indicates need for improved measurement methods.