Session X1: Nanostructure Studies of Strongly Correlated Materials

Probing colossal magnetoresistance in manganites at the nanoscale

Although a complete understanding remains elusive, phase separation is often cited as being the underlying cause of colossal magnetoresistance (CMR) in the manganites. In the low T$_c$ manganites, a combination of transport and scanning probe experiments clearly reveal microscale insulating and metallic regions coexisting within a certain temperature range that coincides with CMR. We present results from recent experiments that explore phase separation and CMR in manganite nanostructures. When nanometer wide bridges are patterned from thin films of the prototypical phase separated manganite (La,Pr,Ca)MnO$_3$, a few alternating insulating and metallic regions can form along the length of the wire. For the purposes of electronic transport across the wire, the phase separated regions can be thought of as microscopic analogs of insulating and metallic multilayers. Transport measurements across the wires reveal that several distinct mechanisms are at play and act simultaneously to give the collective CMR effect observed in bulk. We identify signatures of tunneling magnetoresistance between two ferromagnetic metallic regions separated by an insulating region, exchange bias between the (antiferromagnetic) insulating and ferromagnetic metallic regions, colossal electroresistance jumps and finally, spin dependent tunneling across insulating regions that form at domain walls when the films are predominantly ferromagnetic and conducting at low temperatures. We discuss the implications of our results in understanding CMR in manganites and the potential for practical applications using manganite nanostructures.   

Describing nonequilibrium behavior in strongly correlated materials via dynamical mean-field theory

Dynamical mean-field theory was introduced in 1989 and has become one of the most successful methods for solving models of strongly correlated electrons in equilibrium (it becomes exact in the infinite-dimensional limit). In this talk, I show how to generalize dynamical mean-field theory to nonequilibrium situations. For transient response, one discretizes the Kadanoff-Baym-Keldysh contour then solves the discrete problem directly. For steady-state response, one can formulate a theory directly in the long-time limit for the retarded Green's functions. These techniques are applied to the problem of the quenching of Bloch oscillations due to electron-electron interactions and to the problem of time-resolved pump/probe photoemission spectroscopy of strongly correlated electrons when a system is driven to a nonequilibrium steady state and cannot be described by the quasiequilibrium approximation with an effective temperature. This work was completed in collaboration with Tom Devereaux, Sasha Joura, Hulikal Krishnamurthy, Brian Moritz, Thomas Pruschke, Volodomyr Turkowski, and Velko Zlati\'c. Recent references include: J. K. Freericks, V. M. Turkowski, and V. Zlati\'c, Phys. Rev. Lett. {\bf 97}, 266408 (2006); J. K. Freericks, Phys. Rev. B {\bf 77}, 075109 (2008); A. V.Joura, J. K. Freericks, and Th. Pruschke, Phys. Rev. Lett. {\bf 101}, 196401 (2008); J. K. Freericks, H. R. Krishnamurthy and Th. Pruschke, Phys. Rev. Lett. {\bf 102}, 136401 (2009); and B. Moritz, T. P. Devereaux, and J. K. Freericks, arXiv:0908.1807.   

Time-dependent DMRG studies of strongly correlated systems out of equilibrium

The recent development of time-dependent density-matrix renormalization group (tDMRG) has opened the door for studying several interesting problems that involve the nonequilibrium real-time dynamics of strongly interacting 1D lattice models [1]. We describe briefly one tDMRG approach, the Suzuki-Trotter algorithm. Then we discuss applying tDMRG to study the conductance of strongly correlated nanostructures [2] and describe a method to mitigate finite-size effects which may arise in such studies [3]. We present a few examples including quantum dots in the Kondo regime, and dielectric breakdown of a Mott insulator. Another class of problems involves the time-evolution of excitations in cold atoms and strongly interacting electronic materials. We present the results of a tDMRG study of an electron-hole pair in a 1D Mott insulator [4]. We finally present other possible applications and future directions. \\[4pt] [1] S. R. White and A. E. Feiguin, Phys. Rev. Lett. 93, 076401 (2004).\\[0pt] [2] K. A. Al-Hassanieh et al., Phys. Rev. B 73, 195304 (2006).\\[0pt] [3] Luis G. G. V. Dias da Silva et al., Phys. Rev. B 78 195317 (2008).\\[0pt] [4] K. A. Al-Hassanieh et al., Phys. Rev. Lett. 100 166403 (2008).   

Many-body theory of electric and thermal transport in single-molecule heterojunctions

Electron transport in single-molecule junctions (SMJ) is a key example of a {\it strongly-correlated system far from equilibrium}, with myriad potential applications in nanotechnology. When macroscopic leads are attached to a single molecule, a SMJ is formed, transforming the ``few-body'' molecular problem into a true ``many-body'' problem. Until recently, a theory of transport that properly accounts for both the particle and wave character of the electron has been lacking, so that the Coulomb blockade and coherent transport regimes were considered ``complementary.'' We have developed a nonequilibrium many-body theory\footnote{J.\ P.\ Bergfield and C.\ A.\ Stafford, Phys.\ Rev.\ B {\bf 79}, 245125 (2009).} that reproduces the key features of both the Coulomb blockade and coherent transport regimes simultaneously. Our approach is based on nonequilibrium Green's functions, enabling physically motivated approximations that sum terms to all orders. The junction Green's functions are calculated exactly in the sequential-tunneling limit, and the corrections to the electron self-energy due to finite tunneling width are included via Dyson-Keldysh equations. In this talk, I will present a brief overview of our many-body theory of SMJ and discuss the simulated linear and nonlinear response of a benzenedithiol-gold junction. I will also outline our derivation of an exact expression for the heat current in an interacting nanostructure, highlighting our prediction\footnote{J.~P.~Bergfield and C.~A.~Stafford, Nano Letters {\bf 9}, 3072 (2009).} of a dramatic quantum-induced enhancement of thermoelectric effects in the vicinity of a transmission node. Finally, I will provide several striking examples where the predictions of our many-body theory differ drastically from those of mean-field (density functional) theory.   

Electric field driven transition in magnetite

Magnetite, Fe$_{3}$O$_{4}$, is a strongly electronically correlated system and thus exhibits remarkable electrical and magnetic properties, including the Verwey transition at T$_{V} \sim $122 K, which has attracted much attention since its 1939 discovery. Fe$_{3}$O$_{4}$ has recently revealed a new effect. By performing experiments at the nanoscale, we have discovered a novel \textit{electric-field} driven transition (EFT) in magnetite below T$_{V}$, from high- to low-resistance states driven by high electric field. The EFT is detected both in Fe$_ {3}$O$_{4}$ nanoparticles and thin films, is hysteretic in voltage under continuous biasing, and is not caused by self- heating (S. Lee \textit{et. al.}, Nature Mater. 7, 130 (2008)). In this work we report on a thorough investigation of this new EFT. First, we unveil the origin of hysteresis observed in $I-V$ curves. By applying voltage in a \textit{pulsed} manner with controlled parameters we unambiguously demonstrate that while the transition is field-driven, hysteresis results from Joule heating in the low-resistance state. A simple relaxation-time thermal model captures the essentials of the hysteresis mechanism (A. Fursina \textit{et al.}, Phys. Rev. B 79, 245131 (2009)). Second, by doing multilead electrical measurements, we quantitatively separate the contributions of the Fe$_{3}$O$_{4}$ channel and each electrode interfaces and explore the contact effects upon testing several different contact metals. On the onset of the transition, contact resistances at \textit{both} source and drain electrodes and the resistance of Fe$_{3}$O$_{4} $ channel decrease abruptly. This behavior is consistent with a theoretically predicted transition mechanism of charge gap closure by electric field. Finally, we report recent measurements of the distribution of switching voltages and its evolution with temperature. These studies demonstrate that nanoscale, nonequilibrium probes can reveal much about the underlying physics of strongly correlated materials.