09.12.2012 Views

Plenarvorträge - DPG-Tagungen

Plenarvorträge - DPG-Tagungen

Plenarvorträge - DPG-Tagungen

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

Tiefe Temperaturen Dienstag<br />

H. Schoeller, and G. Schön, Phys. Rev. Lett. 78, 4482 (1997)<br />

[2] R. Schäfer et al., cond-mat/0205223; K. W. Lehnert et al., Phys. Rev.<br />

Lett. 91, 106801 (2003).<br />

TT 11.2 Di 09:45 H19<br />

The Coulomb blockade in two island systems with high conductive<br />

junctions — •Roland Schäfer, Bernhard Limbach, Peter<br />

vom Stein, and Christoph Wallisser — Forschungszentrum Karlsruhe,<br />

Institut für Festkörperphysik, Postfach 3640, 76021 Karlsruhe<br />

We report measurements on single electron pumps consisting of two<br />

metallic islands formed by three tunnel junctions in a row. We focus<br />

on the linear response conductance as a function of gate voltages<br />

and temperature of three samples with varying system parameters. In<br />

all cases strong quantum fluctuation phenomena are observed by a<br />

log (kBT/(2Eco))-reduction of the maximal conductance where Eco measures<br />

the coupling strength between the islands. The samples display<br />

a rich phenomenology culminating in a non-monotonic behavior of the<br />

maximal conductance as a function of temperature.<br />

TT 11.3 Di 10:00 H19<br />

Coulomb blockade and Non-Fermi-liquid behavior in quantum<br />

dots — •Frithjof Anders 1 , Eran Lebanon 2 , and Avraham<br />

Schiller 2 — 1 Insititut für Theoretische Physik, Universität Bremen,<br />

Postfach 330 440, D-28334 Bremen — 2 Racah Institute of Physics, The<br />

Hebrew University, Jerusalem 91904, Israel<br />

The non-Fermi-liquid properties of an ultrasmall quantum dot coupled<br />

to a lead and to a quantum box are investigated using a new variant of<br />

Wilson’s numerical renormalization group. Below the charging energy of<br />

the quantum box, a second screening channel is dynamically generated.<br />

Tuning the ratio of the tunneling amplitudes to the lead and box, we<br />

find a two-channel Kondo fixed point for arbitrary Coulomb repulsion on<br />

the dot, proving that the two-channel Kondo effect is far more generic<br />

to this setting than the original scenario of Oreg and Goldhaber-Gordon.<br />

At T = 0, a step-like structure is found in the conductance of a twolead<br />

setting, the height of which depends on the dot occupancy. The<br />

temperature scale below which the two-channel Kondo effect sets in is<br />

greatly enhanced away from the local-moment regime, making this effect<br />

accessible in realistic devices.<br />

TT 11.4 Di 10:15 H19<br />

Thermopower of metallic single electron devices — •Marko<br />

Turek, Jens Siewert, and Klaus Richter — Institut für theoretische<br />

Physik, Universität Regensburg, D-93040 Regensburg<br />

While charge transport in single-electron devices has been thoroughly<br />

investigated during the past decade, heat transport did not attract much<br />

attention, apart from basic considerations, e.g. [1 − 3]. This is surprising<br />

since heat transport exhibits interesting behavior as a function of<br />

external gate voltages and, in particular, renders transport regimes accessible<br />

which are difficult to study by means of charge transport. Here<br />

we present new results for the linear thermopower as a function of the<br />

applied gate voltage in single-electron transistors with normal leads and<br />

normal-conducting islands, as well as for devices with superconducting<br />

islands.<br />

[1] C. W. J. Beenakker and A. A. M. Staring, Phys. Rev. B 46, 9667<br />

(1992)<br />

[2] Y. M. Blanter, C. Bruder, R. Fazio and H. Schoeller, Phys. Rev. B 55,<br />

4069 (1997)<br />

[3] D. Boese and R. Fazio, Europhys. Lett. 56, 576 (2001)<br />

TT 11.5 Di 10:30 H19<br />

Weak charge quantization and full current statistics — •Dmitry<br />

Bagrets 1 and Yuli Nazarov 2 — 1 Institut für Theoretische<br />

Festkörperphysik, Universität Karlsruhe, D-76128 Karlsruhe, Germany<br />

— 2 Delft University of Technology, Lorentzweg 1, 2628 CJ Delft, The<br />

Netherlands<br />

We evaluate the full statistics of the current via a Coulomb island that<br />

is strongly coupled to the leads [1]. This strong coupling weakens the<br />

Coulomb interaction. We show that at voltages below the inverse RCtime<br />

in the circuit all effects of interaction can be incorporated into the<br />

energy renormalization of transmission eigenvalues of the arbitrary scatterers<br />

that connect the island and the leads. At moderate conductance<br />

G of the island below some critical value Gc the above renormalization<br />

results in the onset of an effective Coulomb blockade gap in the currentvoltage<br />

dependence, the value of this gap being exponentially suppressed<br />

as compared to the classical charging energy E = e 2 /2C of the island.<br />

In the opposite case, G > Gc, the Coulomb blockade does not develop<br />

and the interaction correction to transport saturates at low voltages below<br />

the Thouless energy of the island. Remarkably, our renormalization<br />

group approach agrees with the previous equilibrium instanton calculations<br />

[2], i.e. both methods give the same energy scale for the effective<br />

charging energy.<br />

[1] D.A. Bagrets and Yu. V. Nazarov, cond-matt/0304347<br />

[2] Yuli V. Nazarov, Phys. Rev. Lett. 82, 1245 (1999).<br />

TT 11.6 Di 10:45 H19<br />

Negative differential resistance due to the resonance coupling of<br />

a quantum-dot dimer — •Shidong Wang 1,2 , Zouzou Sun 1 , Nelson<br />

Cue 1 , and Xiangrong Wang 1 — 1 Department of Physics, Hong<br />

Kong University of science and technolgy, HKSAR, China — 2 Institut<br />

für Theoretische Physik, Universität Regensburg, 93040 Regensburg<br />

We investigate the electron tunneling through a coupled quantum-dot<br />

dimer (a two quantum-dot system) under a dc bias. We find a new mechanism<br />

for the negative differential resistance (NDR) in a system with more<br />

than one quantum dots. We show that a peak in the I-V curves, that is,<br />

a NDR, occurs at low temperature when two discrete electronic states in<br />

the two quantum dots are aligned with each other. We call this coupling<br />

of two states the resonance coupling. We study also the dependence of<br />

the height and width of the I-V peak on the dot-dot coupling, e-e interaction,<br />

and the temperature. We find that the peak disappears due to<br />

thermal smearing effects at high temperature. We expect that this new<br />

NDR is useful in STM experiments. We hope that a better understanding<br />

of this resonance coupling effect may enhance the STM as a powerful<br />

probe not just for a regular surface, but also for a cluster structure.<br />

TT 11.7 Di 11:00 H19<br />

Nonlinear Transport through Quantum Wires: Functional<br />

Renormalization Group in Nonequilibrium — •Severin<br />

Jakobs 1 , Volker Meden 2 , Herbert Schoeller 1 , and Kurt<br />

Schönhammer 2 — 1 Institut für theoretische Physik A, RWTH Aachen,<br />

D-52056 Aachen, Germany — 2 Institut für Theoretische Physik,<br />

Universität Göttingen, Bunsenstr. 9, D-37073 Göttingen, Germany<br />

We study nonlinear transport at finite temperature through a quantum<br />

wire with impurities. The wire is modelled by a one-dimensional<br />

quantum lattice including nearest neighbour interaction. It is coupled<br />

via tunneling barriers to two reservoirs to which a finite bias voltage is<br />

applied. For calculating the nonequilibrium properties we generalize a<br />

functional renormalization group method (applied recently to the equilibrium<br />

case [1]) to the nonequilibrium situation by utilizing real-time<br />

Feynman diagrams. This enables us to handle weak as well as strong tunneling<br />

and arbitrary potentials on the wire. The coulomb interaction is<br />

treated within perturbative renormalization group, i.e. only the Hartree-<br />

Fock type of diagrams are included. For a single impurity in an infinite<br />

quantum wire we recover the exponents for the current vs. temperature<br />

and for the current vs. bias voltage predicted from bosonization [2]. For<br />

a finite wire however, the interplay of tunneling barriers and impurity<br />

produces exponents depending on the position of the impurity.<br />

[1] V.Meden et al., Phys.Rev.B 65, 045318 (2002)<br />

[2] C.L.Kane, M.P.A.Fisher, Phys.Rev.B 46, 15233 (1992)<br />

TT 11.8 Di 11:15 H19<br />

Electron transport through interacting quantum dots —<br />

•Andrei Zaikin 1 and Dmitri Golubev 1,2 — 1 Institut ür Nanotechnologie,<br />

Forschungszentrum Karlsruhe, 76021 Karlsruhe — 2 Institut für<br />

Theoretische Festkörperphysik, Universität Karlsruhe, 76128 Karlsruhe<br />

We present a detailed theoretical investigation of the effect of Coulomb<br />

interactions on electron transport through quantum dots and double<br />

barrier structures connected to a voltage source via an arbitrary linear<br />

impedance. Combining real-time path integral techniques with the scattering<br />

matrix approach we derive the effective action and evaluate the<br />

current-voltage characteristics of quantum dots with large conductances.<br />

Our analysis reveals a reach variety of different regimes which we specify<br />

in detail for the case of chaotic quantum dots. At sufficiently low energies<br />

the interaction correction to the current depends logarithmically on<br />

temperature and voltage. We identify two different logarithmic regimes<br />

with the crossover between them occurring at energies of order of the<br />

inverse dwell time of electrons in the dot. We also analyze the frequencydependent<br />

shot noise in chaotic quantum dots and elucidate its direct<br />

relation to interaction effects in mesoscopic electron transport.

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!