Chemistry Reference
In-Depth Information
Fig. 8.23 (a) Example of a
Pd-Ni mixed-metal chain
segment. (b) Schematically
drawn spin susceptibility as a
function of x ( solid line ). The
dotted line is the hypothetical
linear behavior. Vertical
arrow specifies the value in
the pure Ni case, namely, that
of the Mott insulator
the two phases by the adjustments of model parameters via the control of environ-
mental conditions such as temperature and pressure. However, in the Pt and Ni
compounds, they are deep inside each phase space, being very difficult to have any
relationship with the other phase. Here comes the Pd complex. It is always a CDW
insulator, except for the compound synthesized very recently, which shows a
CDW-Mott-insulator at a certain temperature [ 42 ]. Although most of the Pd
compounds are genuine CDW insulators, they have smaller CDW gaps, for exam-
ple, 0.7 eV for [Pd(chxn) 2 Br]Br 2 [ 43 ]. This small gap energy tells us that this
material is close to the phase boundary and therefore has the potential to cause a
phase transition, being a “bridge” between the two phases.
8.4.1 Pd and Ni Mixed-Metal Systems
We start from the beginning of this story, that is, the synthesis of Pd/Ni mixed-metal
compounds. Regarding the details of this synthesis, we recommend the readers to
read the experimental section of this topic. We here only mention three important
facts, that are, the behaviors of spin susceptibilities [ 44 ], and IR and Raman signals
[ 45 ]. Before those, we make the meaning of the mixed metal clear. The metal sites
are confirmed experimentally to be occupied by the Ni and Pd atoms replacing each
other randomly (see Fig. 8.23a ), that is, no interstitial site at all, which finding made
a theoretical analysis simpler because we only need to use site-dependent parameter
values. Returning back to the experimental facts, the magnitude of the spin suscep-
tibility was found to be not linear as a function of the mixing degree x as expressed
as Ni 1 x Pd x (chxn) 2 Br 3 . As we have already seen, the CDW, in which all the sites
are doubly occupied, is nonmagnetic, while the Mott insulator has many active
spins and has a finite magnitude of susceptibility. If the selection of the states
occurred independently at each site, the x -dependency would be linear, but it is not
this case. Moreover, its functional form schematically drawn in Fig. 8.23b shows
that even at 20% of Ni
inclusion ( x ¼
0.2) yields appreciable intensity of
Search WWH ::

Custom Search