1983 Volume 26 Issue 11 Pages 846-858
Among the several possible factors, that is, 1) primary energy of the incident electron, 2) geometrical distribution of the incident electron on the specimens, 3) geometrical alignment between the electron and the analyzer, 4) stray magnetic and electric fields, 5) modulation energy of the analyzer, 6) response of electron multiplier, 7) energy sweep rate of the analyzer, 8) time constant of lock-in amplifier, and 9) response of recording system, which may affect a shape of the measured AES spectrum, the energy sweep rate and the time constant are investigated for finding improved operation procedures. It is shown that the spectrum shape measured is determined uniquely by introducing the trace factor which is derived from the energy sweep rate and the time constant. The optimum value of the trace factor is determined and examined for performing a reliable measurement with a proper accuracy. If the apparatus function is written as a function of the trace factor including the time constant in exponential decay form, it is possible to estimate the original profile of the spectrum by a numerical procedure. Concerning with this deconvolution procedure, an improvement of S/N is examined. We discuss a turning effect on the noise, its reduction, and a distortion due to time constant in repeated measurement.