A. Datar et Sd. Prasad, ROLE OF ADSORBATE INTERACTIONS IN SURFACE DYNAMICS AND PHASE-TRANSFORMATIONS - MEAN-FIELD AND QUASI-CHEMICAL APPROXIMATION APPROACHES, The Journal of chemical physics, 100(2), 1994, pp. 1742-1755
The significant influence of adsorbate interactions in surface dynamic
s is quantified using mean-field approximation (MFA) and quasichemical
approximation (QCA) approaches and two typical situations (i) T > T-c
(critical temperature for surface phase transformation) and (ii) T <
T-c are analyzed. The formulation involves transition state theory (TS
T) and the key parameters involved are: (1) the sign and magnitude of
the pairwise adsorbate interaction energy (w > 0, w < 0 meaning repuls
ive and attractive interactions, respectively) (2) W-A#, the interacti
on energy between a molecule in the ground state and the activated com
plex. W-A#(A) is in turn related to w by a coupling parameter sigma. s
igma=0, sigma=1 are shown to result in extreme divergence of the rate
behavior for both repulsive and attractive interactions. First T > T-c
is considered. For sigma=0, attractive interactions retard and repuls
ive interactions enhance the surface rates. The rates display nonmonot
onic behavior for attractive interactions and steady increase with sur
face coverage for repulsive interactions. However, when sigma=1, the r
ates monotonically increase for both types of forces. In addition the
attractive forces show an instability of the slope due to a cooperativ
e catalytic effect. Both attractive and repulsive forces display maxim
a when plotted against temperature, the maxima being sharper for the f
ormer case. The case T < T-c is more interesting, as a discontinuous p
hase separation can occur for attractive interactions. The density and
internal energy differences between the coexisting phases are compute
d proceeding from closed-form expressions of the canonical ensemble pa
rtition functions and employing standard methods of statistical mechan
ics. Since repulsive forces can only show continuous order-disorder tr
ansitions, they are not considered for T < T-c. The surface rate expre
ssions (both corrected and uncorrected for ground-state internal energ
y differences between the phases) display a symmetric rate curve (symm
etric about theta=0.5) vs surface coverage with a maximum at theta=0.5
. A certain type of hole-particle symmetry is present in the rate expr
ession as the rate expression is invariant with respect to the exchang
e of an occupied and vacant site. This conclusion is valid for both si
gma=0, sigma=1. The appearance of symmetry in the rate curve is sugges
tive of the phase separation. The qualitative differences between the
rate predictions of MFA and QCA are significant enough to warrant refi
nement in the analysis of surface dynamics.