Chemistry Reference
In-Depth Information
truly realistic because we must have a Mott-insulator state, in which spins are
strongly fluctuating, so as to reproduce the temperature dependence of magnetic
susceptibility in this system, that is, that of a Bonner-Fischer type. In the LDA
calculations, on the other hand, we find a long-range order of antiferromagnetic
type, because of the mean-field nature of those calculations. Such deficiencies are
solved in the model calculations.
Iwano et al. performed density-functional theory (DFT) calculations for [Pt
(en) 2 ][Pt(en) 2 Cl 2 ](ClO 4 ) 4 (hereafter PtCl), using a quantum-chemical molecular-
orbital method [ 15 ]. They also found a similar property, namely, the importance of
5d z 2 orbitals and its strong hybridization with the Cl 2p z orbitals. What was a matter
of interest was the atomic valencies of the Pt ions because people in this field (also
in this topic) often mention their nonequivalent two valencies, namely, +2 and +4.
Unfortunately, the result of the usual Mulliken analysis was disappointing. They
found +2.3 and +2.1, instead of +4 and +2, respectively. Nevertheless, they thought
that the whole result was sound and consistent with the CDW picture, since the
electronic excitations were correctly the CT excitations from5d z 2 (allegedly Pt 2+ )to
5d z 2 (allegedly Pt 4+ ). The above valency of +2.3 rather seems to indicate the strong
hybridization with the halogen orbitals approaching to that, in a way like
ðþ
4
Þþ
2
ð
. As one more comment, the hybridization of d z 2 orbitals with the
molecular orbitals of en is rather small. It was once argued that that hybridization
was a origin to reduce the on-site Coulombic repulsion effectively potentially
yielding the electron density spread to the ligand, but their result demonstrated
that such an effect was not a dominant one.
1
Þðþ
2
Þ
8.2 CDW (M
¼
Pt and Pd) Systems
In this section, we discuss the CDW systems from various aspects. From the
viewpoint of electron correlation, this system might be a little “boring” except for
the case in the vicinity of the phase boundary to the Mott insulator. However, this
has a great significance from the view point of e-l interaction and its manifestation
both in the optical spectra and the photoexcitations of nonlinear excitations like
solitons. The solitons are defined as a domain wall that separates the two equivalent
phases of the CDW. The formation of a soliton pair thus means a domain formation
in a uniform background. As far as we know, such a dynamical process has not been
analyzed completely even until now, not only in this field but also in other fields, in
spite of its importance as a many-body problem. The largest reason lies of course in
its difficulty mainly due to the quantum nature of the lattice. In this sense, this
section is also not a complete one and we are still on the way. We nevertheless
believe that the results are correct at least as a rough sketch and that they
will provide further motivation not only for the author himself but also for the
readers.
Search WWH ::




Custom Search