Chemistry Reference
In-Depth Information
(Smith, 1993) and (DCNQI)
2
Cu (Inoue
et al.
, 1993). However, one has to be ex-
tremely cautious when the photoemission data show that
N
(
E
F
)
0, because,
since the photoemission process is extremely surface-sensitive, a 3D metal may ex-
hibit low emission at
E
F
when the surface and bulk exhibit different stoichiometries.
Perhaps the best example to illustrate this problem is K
0
.
3
MoO
3
. It is well-known
that transition metal oxides desorb oxygen in UHV when irradiated with X-ray
photons with energies greater than the absorption edges of the transition metal core
levels. This photon stimulated desorption is due to an interatomic Auger process
(Knotek & Feibelman, 1978). In the case of K
0
.
3
MoO
3
for
h
→
larger than the Mo3
d
threshold energy,
c
. 230 eV, the surface becomes oxygen-deficient and thus semi-
conducting, and hence
N
(
E
F
)
ω
0. Emission at
E
F
thus strongly depends on the
surface stoichiometry. The case of high-temperature superconductors is also re-
vealing and edifying since the first photoemission experiments performed by 1988
showed an absence of emission at
E
F
and many articles were produced justifying
this experimental fact as a result of strong electronic correlations. Those experi-
ments were performed on polycrystalline samples. However, when single crystals
were available a finite intensity at
E
F
was observed with ARUPS and the supercon-
ducting gap could be unambiguously measured (Imer
et al.
, 1989). In conclusion,
special care has to be taken with the stoichiometry and morphology, and thus with
the quality of the surface.
With these advisory ideas in mind let us spend some time on the electronic
structure of one of the most extensively studied MOMs: the quasi-1D metal TTF-
TCNQ.
=
Electronic structure of TTF-TCNQ
Theoretical band structure close to E
F
Because of the molecular nature and the relatively large unit cell, the correct de-
scription of the electronic structure of TTF-TCNQ, as for most of the molecular
conductors, is still a challenge from the computational viewpoint even with current
computational capabilities. Figure 6.3 shows the calculated band structure near
E
F
(
E
F
=
0) for the RT and ambient pressure structure of TTF-TCNQ (Kistenmacher
et al.
, 1974) along the
a
∗
-(
X),
b
∗
-(
Y) and
c
∗
-(
Z) directions of the recipro-
cal lattice. The calculations shown here were performed using a numerical atomic
orbitals DFT approach (see Appendix B), using the generalized gradient approxima-
tion (GGA) with atomic orbitals and pseudopotentials, developed and designed for
efficient calculations in large systems and implemented in the SIESTA code (Soler
et al.
, 2002). The predicted band structure is almost identical to those found in
Ishibashi & Kohyama (2000) and Sing
et al.
(2003b) despite technical differences.
−
−
−
