Chemistry Reference
In-Depth Information
Fig. 28 Sketch of a system composed of molecular layers. The
arrows
represent the molecules
(being polarizable entities) and the planes with
arrows
represent a layer of molecules, with equal
ʸ
and
. The molecules on the
upper plane
show a change in the tilt and twist angle with respect to
those of the
lower plane
.x(x
0
), y (y
0
), and z (z
0
) are the coordinates of the system (molecule). a
1
,
a
2
, and a
3
represent the primitive lattice vectors (From Mendoza and V
´
zquez-Nava [
67
].
Reprinted with permission. Copyright 2005 by the American Physical Society.)
˕
The results show that for a system with more than one molecular layer, if the
upper layers are allowed to change orientation with respect to the lower ones, the
final spectrum simulates efficiently the particular oscillating line shape of RAS
spectra (Fig.
29
). In this way, it is then possible to correlate the experimental line
shape to a well-defined change of structure due to the addition of a layer at a
particular threshold value (in experiments 8-10 ML). In particular, tilting the
dipoles of the upper layers towards the surface normal, a derivative-like line
shape is obtained that goes from negative to positive values of the anisotropy as
ʻ
increases. If instead the tilt change is away from the surface normal, the oscillating
spectrum goes from positive to negative, as
ʻ
increases. This means that the idea of
correlating the RAS line shape to a particular defined structure in the layer, the outer
layer different from the inner layers, is in principle correct. Nevertheless, the
method is of questionable utility in real systems: (i) Too many parameters should
be determined, some of them practically impossible to measure; (ii) from the point
of view of calculations, the high number of free parameters reduces the possible
impact of the model onto the effective and realistic comprehension of LB layers. In
conclusion, although the model by Mendoza represents formally a confirmation and
consequently an explanation of the experimental RAS results, its application is too
intricate, unless the system is reduced to extremely reduced thickness, in practice
not more than one or two layers on top of the (possibly isotropic) substrate.
However, this also means that a different deposition method must be used to control
more closely the structural details of such a thin layer, as well as its uniformity on
the whole sample surface. Moreover, the possibility to vary the thickness continu-
ously even at values lower than a true full layer could help the experimental
investigation of the development of the line shape. LB and LS deposition methods,
on the contrary, suffer limitations under both aspects: (i) A certain presence of
defects is unavoidable, mainly for LS; (ii) the possibility to deposit less (and even
