80 Trends in activity and stability
how the thickness d varies from the original values under a different assumption:
the Scofield cross sections are now considered as inaccurate and they have been
corrected in order to produce the stoichiometric ratio of 5 on the as-prepared sample.
Figure 4.11: Thickness d of the Pt overlayer for different values of t (thickness of the C-OH
contamination layer). The originally calculated values (dotted lines) have been corrected
by assuming that Scofield cross sections are inaccurate (full lines): Scofield cross sections
have been modified in order to provide a Pt to Gd ratio of 5 in the measurements on a
sputter-cleaned sample (see Figure 4.9 a).
With this correction, the two lines corresponding to the Gd 4d and the Gd
3d5=2 peaks move down and predict smaller thicknesses, particularly in the case of
the Gd 4d peak which is less sensitive to the presence of contaminations but more
sensitive to errors in the quantification of the Pt to Gd ratios. The two lines now
cross each other at a value of t 4.5 Å, reasonably close to the estimated value
6.7 Å. Most importantly at t = 6.7 Å the choice of the two peaks lead to a similar
estimate of the Pt thickness, within the experimental uncertainty: d 9.1 Å and
d 8.2 Å respectively. This is a further evidence of the fact that the measured
discrepancies from the measured ratios on sputter-cleaned samples are mainly due
to inaccuracies in the Scofield cross sections or in their association to the measured
intensities. This also suggests that method 1 does not provide the best estimate of
the Pt overlayer thickness and that method 2 should be preferred.