11.2 Valence: Integral, Mixed, 605: Charge Density Wave

Download as pdf or txt
Download as pdf or txt
You are on page 1of 1

11.

2 Valence: Integral, Mixed, Nearly Integral 605

Thus the t - J- alternation of the antiferromagnetic ground state of the


positive-U model is mapped onto an alternating 6s2 - 63' charge order of
the negative4 model. We conclude that the ground state of BaBiOs is a
charge density wave. Since it amounts to an alternation of Bi3+ and Bi5+
valence states, we may also speak about valence disproportionation. The ar-
gument implies that (at sufficiently large IUl) the charge order is stabilized by
a coupling t2/lUl.
N

The negative-U concept has a nice corolfary. Let us notice that the binding
into singlet pairs (d-sites) is energetically favoured whether the band is half-
filled or not. However, away from half-filling, the singlet pairs are free to move.
This seems to be the recipe for a superconductorwith a small coherence length.
And indeed, doped BaBiOs (notably, Bal-,K,BiOs [263]) is a representative
of one of the classes of high-T, superconductors.

11.2.2 Valence Mixing in Rare Earth Compounds


For the rest of this chapter, we return to the study of positive-U systems
whose zeroth order Hamiltonian is (11.1). In the mixed valent regimes,
the ground state of 7d0)is macroscopically degenerate. Taking the
example of 4fo-4f' mixing, there are Cinf ways to distribute the 4f1
sites over the lattice, and the ground state entropy is further increased
by the fact that each 4f1 site has a degenerate ionic ground state.
The most significant omission in writing down (11.1) is that it ne-
glects any quantum-mechanical mixing between the f-electrons and the
conduction electrons. In reality, the two sorts of electron states are
hybridized. The complete Hamiltonian should include at least the one-
electron hybridization processes'' which turn an f-electron into a con-
duction electron, or vice versa

To see the effect of switching on %hyb, let us consider the two con-
secutive hybridization processes shown in Fig. 11.3. We envisage a
"The notation Vk& for the hybridization amplitude is very slipshod. Properly
written, it should make it clear that hybridization satisfies the requirements of sym-
metry. Furthermore, some off-diagonal Coulomb processes also hybridize. However,
we do not need the detailed form here.

You might also like