"Tunable spin and charge Seebeck effects in magnetic molecular junctions"
The increasing interest in the thermoelectric properties of materials and the
quest for high Seebeck coefficients is motivated by the promise of more
efficient solid state refrigerators and the conversion of waste heat into
electricity. The Seebeck effect refers to the generation of a charge current
(or a voltage drop) by a temperature gradient applied across a metal, and the
spin-Seebeck effect, concerns the thermal generation of pure spin currents. The
recent experimental observation of the Seebeck effect in different
nano-structures, in particular in molecular junctions and quantum dots, opened
new routes to study these phenomena.
On the one hand, the high sensitivity of these systems to external fields, their
scalability and tunability make them potential candidates for a variety of
technological applications. On the other hand, thermoelectric and thermomagnetic
effects provide a unique probe of electron correlation effects and are a useful
tool to gain further insight on fundamental problems like the Kondo regime where
the energy transfer is dominated by spin fluctuations.
I'll present results for the charge and spin Seebeck effects of a spin-1
molecular junction as a function of temperature (T), applied magnetic field (H),
and molecular magnetic anisotropy (D) obtained using Wilson's numerical
renormalization group [1]. A hard-axis magnetic anisotropy produces a large
enhancement of the charge Seebeck coefficient Sc (~ kB/|e|) whose value only
depends on the residual interaction between quasiparticles in the low
temperature Fermi-liquid regime. In the underscreened spin-1 Kondo regime, the
high sensitivity of the system to magnetic fields makes it possible to obtain a
sizable value for the spin Seebeck coefficient even for magnetic fields much
smaller than the Kondo temperature. Similar effects can be obtained in C60
junctions where the control parameter is the gap between a singlet and a triplet
molecular state.
I'll also discuss briefly the thermoelectric properties of an SU(4) Kondo
resonance, that describes the low temperature transport through clean C
nanotubes [2].
[1] Pablo S. Cornaglia, G. Usaj, and C. A. Balseiro, Phys Rev. B 86, 041107(R)
(2012)
[2] P. Roura-Bas, L. Tossi, A. A. Aligia, and Pablo S. Cornaglia (submitted)