Inadequate design, deterioration through ageing and poor setting-up procedu
res may cause the intensity measuring systems in XPS instruments, and very
possibly in all simple particle counting systems, to show non-linearities.
Even in correctly designed and used equipment smaller non-linearities occur
but these, being predictable, may be corrected to a high level of accuracy
. Non-linearities are shown for both the high and low count rate regimes fo
r both single channel and multidetector systems. To monitor these defects f
or XPS systems three simple methods are described which require no addition
al equipment. The first two, for assessing the effects at high count rates,
make use of 'ratio plots' which involve the ratios of intensities measured
at two X-ray source emission currents. The first of these methods uses the
survey spectrum from a copper or stainless steel sample measured in either
the constant Delta E/E or constant Delta E mode. The second method uses th
e Cu 2p(3/2) peak to diagnose other effects. For very low count rates a sig
nal generator simulating the detector pulses, but with pulses evenly spaced
in time, may be used for checking the electronic part of the circuit. Usin
g these methods, non-linearities of up to 50% are illustrated for commercia
l equipment for electron counting in common analytical use. (C) 1999 Elsevi
er Science B.V. All rights reserved.