Chemistry Reference
In-Depth Information
a
b
K
MXM
=
6
U
M
=6
6
β
CDW
U
M
4
CP
4
2
AV
0
2
0
0.2
0.4
0.6
0.8
1
t
MXM
Fig. 12.3 Ground-state phase diagram of the 12-site model for infinitely large
K
MXM
(a) in the
t
MXM
-
b
-
U
M
space and (b) its cross section at
U
M
¼
6. The parameters are
t
MM
¼
1,
a ¼
0,
K
MX
¼
6,
V
MM
¼
0,
V
MXM
¼
0, and
V
2
¼
0[
23
]
competition among the charge-ordered phases until later, we simply ignore the
long-range interactions. Exactly diagonalizing the present model, we show a phase
diagram in the space spanned by
t
MXM
,
b
, and
U
M
in Fig.
12.3
, where
t
MM
is set to
2
be unity. Because the parameter sets (
a
,
b
,
K
MX
,
K
MXM
) and (
la
,
lb
,
l
K
MX
,
2
l
K
MXM
) are related by the scaling of
y
a
,
and
y
b
and thus equivalent, the specific
value of
K
MX
is insignificant. First of all,
t
MXM
is found to stabilize the AV phase. In
R
4
[Pt
2
(pop)
4
I]
n
H
2
O, where electrons are the most delocalized among X
Cl, Br,
and I, the AV phase may appear for small
d
MXM
because
t
MXM
is expected to
become large. Meanwhile
t
MXM
does not much affect the boundary between the
CDW and CP phases.
¼
12.4 Electron-Lattice Versus Electron-Electron Interactions
In Fig.
12.3
, the CDW phase is stabilized by the site-diagonal electron-lattice
coupling
, while the CP phase is stabilized by the on-site repulsion
U
M
. Thus,
the competition between the electron-lattice and electron-electron interactions
determines the relative stability of these two charge-ordered phases. This fact is
easily understood with the help of the second-order perturbation theory from the
strong-coupling limit
b
0 . The energies of the CDW and CP
phases are degenerate in this limit as long as the long-range interactions are absent.
Namely, their energies are both given by
t
MM
¼ t
MXM
¼ a ¼
bjyjþU
M
þ K
MX
y
2
per binuclear unit.
Search WWH ::
Custom Search