Characterization of the noise in secondary ion mass spectrometry depth profiles

The noise in the depth profiles of secondary ion mass spectrometry (SIMS) is studied using different samples under various experimental conditions. Despite the noise contributions from various parts of the dynamic SIMS process, its overall character agrees very well with the Poissonian rather than the Gaussian distribution in all circumstances, The Poissonian relation between the measured mean-square error and mean can be used to describe our data in the range of four orders. The departure from this relation at high counts is analyzed and found to be due to the saturation of the channeltron used. Once saturated, the detector was found to exhibit hysteresis between rising and falling input Bur and output counts. (C) 1996 American Institute of Physics.