Chemistry Reference
In-Depth Information
6.2.3 Pt/Au
The third column in Fig.
8
shows MD snapshots for Pt deposition on Au. A signi-
ficantly higher percentage of displacement of {100} surface Au atoms by deposited
Pt atoms is observed compared with Pd/Au. Pt and Au, which are mostly immis-
cible in the bulk, display a greater degree of surface mixing than Pd and Au, which
are completely miscible. This situation, which is analogous to that for Rh/Au, can
be attributed to the greater difference of the surface energies between the substrate
and the deposited element in the case of Pt/Au than in the case of Pd/Au and is
consistent with previous simulations by Haftel et al. who modelled the early stages
of growth of Pt/Au{100} and Au/Pt{100} surfaces [
90
]. The apparent mixing in the
Pt/Au{100} system is characterised by a certain degree of clustering, although not
as pronounced as for Rh/Au. Raising the temperature from 300 to 500 K results in a
smoother Pt/Au{100} interface, which is not observed for Rh/Au{100}. Pt depos-
ited onto the Au{111} substrate forms 3D clusters, similarly to Rh/Au{111}, but
displays greater wetting of the substrate. Two main differences are evident when
comparing the effect of temperature on the Pt/Au{111} and Rh/Au{111} systems.
First, there are a greater number of displaced Au surface atoms at 500 K for Pt/Au,
which shows that this temperature is sufficient to overcome the energy barrier for
this elementary process even on the close-packed {111} surface, which supports the
hypothesis put forward for the morphological changes in AuPt nanorods discussed
below (Sect.
6.3
). Second, the growth at 500 K is characterised by smaller surface
clusters than at 300 K. This tendency is opposite to that observed for Rh/Au{100}
and results from the very different kinetic picture of the active processes taking
place on the surface in both cases. For instance, several Pt atoms implanted in the
Au{111} surface at 500 K become nucleation centres, around which 3D surface
clusters form, while the geometries of the Rh/Au{100} surface clusters are mainly
controlled by the rate of surface diffusion. This also illustrates that the character-
istics of surfaces formed by kinetic effects may change in a non-uniform way with
temperature, especially in heteroatomic systems such as those considered here.
It should be emphasised that these MD simulations correspond to the physical
progress of vapour deposition. They reveal only certain internal properties of the
system and tendencies of the deposition process and do not give a comprehensive
picture of the growth kinetics, which may be different in the case of chemical
deposition. Also, modelling deposition on flat surfaces does not account for the
effect of curvature of the nanorod surface, which may result in different strain
conditions when the coating film becomes sufficiently thick. Nevertheless, the
results from these simulations underline the importance of taking special care in
kinetic modelling of such systems to determine the key kinetic processes and to
verify whether or not they change within the temperature range under study.
