download report - Sapienza

Scientific Report 2007-2009
Condensed matter physics and biophysics
C35. High-pressure optical spectroscopy on strongly electron
correlated systems: the Metal Insulator transition
A deep understanding of the physics of strongly correlated
systems still represents one of the most challenging
tasks of condensed-matter research. Generlly speaking,
these systems show a variety extremely interesting physical
behaviors (e.g. high temperature superconductivity
or colossal magnetoresistance) and a high sensitivity
of their properties to external parameters which makes
them highly appealing for a wide range of technological
applications. The latter characteristic is ascribed to
the rather small extension of the electron bandwidth in
comparison with other relevant energy scales as the electron
correlation U or the charge-transfer (CT) energy
gap. Under these conditions the independent electron
approximation breaks down and, for example, materials
at half filling can be insulators despite the opposite prediction
of band theory. Materials at half-filling can become
insulating also in the presence of electronphonon
coupling triggered by spontaneous symmetry breaking
such as Peierls and Jahn-Teller lattice distortions as in
the cases of mixed-valence manganites and vanadium
dioxide.
Vanadium oxides have attracted a considerable interest
because of the abrupt and often huge change of conductivity
at the MIT. In particular, we carried out Infrared
and Raman measurements on HP VO 2 with the
aim of clarifying the microscopic mechanisms at the origin
of the spectacular temperature-driven MIT which
involves a jump of five order of magnitude in conductivity
and a simultaneous structural transition from a
monoclinic (M1) insulating to rutile (R) metallic phase
for T > 341 K. HP experiments show that the MIT is
basically due to electron correlation and that, above 10
GPa, the onset of a metallization process accompanied
by a sluggish structural transition to a new monoclinic
Peierls distorted phase are apparent.
Gasket
Force
Ruby
Sample
Figure 2: Left: Pressure dependence of the spectral weight
of VO 2 ; a metallization proces is observed for P>10 GPa [1]
. Right: spectral weight vs. lattice constant for NiS 2−x Se x .
A metallization process is observed either expanding (Se alloying)
or compressing (Pressure) the NiS 2 lattice [4].
Figure 1: phase diagram of the vanadium oxides investigated
The Metal Insulator Transition (MIT) is thus one of
the most important phenomena to be investigated in
these systems. To this purpose infrared and Raman spectroscopy
are the ideal tools since the former monitors
the charge delocalization proces, and the latter provides
information on the relevant lattice dynamic. Moreover
both the techniques can be efficiently coupled with diamond
anvil cells to carry out high pressure (HP) experiments
which can be indeed very effective in shedding
light on the complex physics lying behind the peculiar
properties shown by these systems. Progressive and controlled
lattice compression, indeed, tunes the strength of
the interactions, simultaneously at work in these systems,
at different extent. It becomes therefore possible
to disentangle the interactions and, in some cases, to enhance
the coupling mechanisms relevant to the physics
of the system which are otherwise weak at ambient pressure.
The research activity of our group in this field
has focused in the last years on different vanadium oxides
(V 3 O 5 ,V 2 O 3 , VO 2 [1,2,3]) belonging to the Magneli
phase and Ni pyrite compounds belonging to the
NiS 2−x Se x family [4].
Force
The cubic pyrite NiS2, is a CT insulator and is
considered, together with vanadium sesquioxide V 2 O 3 ,
a textbook example of strongly correlated materials.
NiS 2 easily forms a solid solution (NiS 2−x Se x ), with
NiSe 2 , which, while being isoelectronic and isostructural
to NiS 2 is nevertheless a good metal. A MIT is thus
observed at room temperature for x > 0.6 as well as
in pure NiS 2 on applying hydrostatic pressure above
P=4 GPa. We find that optical results are not compatible
with the previously claimed equivalence between
Se-alloying and pressure pointing out the different
microscopic origin of the MIT.
References
1. E. Arcangeletti et al., Phys. Rev. Lett. 98, 196406
(2007).
2. L. Baldassarre et al., Phys. Rev. B 75, 24510 (2007).
3. L. Baldassarre et al., Phys. Rev. B 77, 113107 (2008).
4. A. Perucchi et al., Phys. Rev. B 80, 073101 (2009).
Authors
P. Postorino, P. Dore, S. Lupi, E. Arcangeletti, L. Baldassarre,
D. Di Castro, C. Marini, M. Valentini
http://www.phys.uniroma1.it/gr/HPS/HPS.htm
Sapienza Università di Roma 88 Dipartimento di Fisica