Our group investigates new physical phenomena the emerge in nanoscale systems, at surfaces and interfaces of different materials. Our overarching goal is to develop fundamental understanding of the effects of confinement, interfaces, the resulting emerging interactions, and strongly nonequilibrium physical states that become possible to achieve only at nanoscale, and use this knowledge to develop nanoscale devices with new functionalities. Our laboratory is located in rooms W104, W106, and E108 of the Math and Science Building on the beautiful campus of Emory University.

Research highlights

Heisenberg vs Stoner magnetism as a function of electron interaction “Orbital correlations in ultrathin films of late transition metals”, PRM (2023). It may be surprising to some people that the mechanism of ferromagnetism in common transition metal ferromagnets such as cobalt and nickel is still unresolved. In a simple Stoner model of band ferromagnetism usually applied to these materials, electrons with the same spin avoid each other due to the Pauli principle, which lowers their Coolomb repulsion energy, stabilizing spin ordering. We utilize a Hubbard many-particle field-theoretical approach to show that the model is not applicable to thin transition metal films, and instead the effective Heisenberg model of quasi-localized d-electrons provides a good approximation, as can be seen from the figure for the realistic electron interation parameter U0. In this regime, ferromagnetism is stabilized by the orbital liquid state - a correlated singlet orbital state of electrons on neighboring sites that does involve a finite orbital moment but is expected to affect phenomena involving spin-orbit coupling.

Possible coexistence of Mott and BCS correlations in strontium titanate “Effects of spin-orbit interaction and electron correlations in strontium titanate”, PRB (2022). Electron-doped strontium titanate (STOI) exhibits superconductivity at record-low carrier densities. The conventional Bardeen-Cooper-Schrieffer theory of superconductivity is inapplicable in this regime, puzzling scientists for over 50 years. We show that STO exhibits highly anisotropic orbitally-selective electronic properties, such that each of its three conduction subbands is nearly dispersionless along one of the principal crystal directions. In other words, electrons can occupy states with different crystal momentum along this direction, without paying a kinetic energy cost. As a result, we predict that electrons form Mott-like spin singlet pairs allowing them to avoid each other, and lowering their Coulomb repulsion energy. Such pairs are similar to the the resonating valence bond pairs in cuprate superconductors, and may coexist with Cooper pairs, as shown in the figure. The mechanism revealed in our work is highly sensitive to crystal distortions, which may explain observations of strong dependence of superconducting properties of STO on strain.

“Electronic properties of the mean-field resonance valence band model of cuprates; arXiv:2106.09924 . We theoretically address the electronic properties of high temperature cuprate superconductors. This topic has been extensively researched by the condensed matter community for 36 years, but a widely accepted theory has not yet emerged. We demonstrate that a simple mean-field resonance valence bond model of cuprates proposed in 1987 (35 years ago) by P.W. Anderson and co-workers describes gapless spin singlet pairs of electrons localized in the reciprocal space to two special points at the boundary of the Brillouin zone. We show that the existence of such pairs explains all the main puzzling properties of cuprates, including the "d-wave symmetry" of pairing, the "strange metal" behaviors in the normal state, the pseudo-gap observed in the normal state, and the common charge density modulations. These features stem from the interplay between spin excitations (spinons) and singlet pairs of electrons localized at the intersection between the Brillouin zone boundary and the nodal lines of the spinon spectrum. Our results may provide a pathway for room-temperature superconductivity at ambient pressure.

nonlocal resistance vs current “Transport and relaxation of nonequilibrium phonons generated by current”, PRB (2022). Current-generated Joule heat is a major roadblock for the miniaturization and the increase of speed of electronic nanodevices. The present understanding of this phenomenon and the approaches to its mitigation are based on the assumption that the phonons generated by current form a thermal distribution. We perform nonlocal electronic measurements utilizing an electrically-biased metallic nanowire as a phonon source, and a separate nanowire serving as the phonon detector, to demonstrate that contrary to the 150 year-old paradigm, the distribution of phonons generated by current is highly non-thermal. We analyze the dependence on the thickness of the spacer separating the nanowires, to show that these non-equilibrium phonons relax via strongly anharmonic processes that cannot be described in terms of the usual few-phonon scattering. Our findings provide insight into the mechanisms of current-driven phonon generation, transport, and relaxation at nanoscale, which will likely facilitate new approaches to efficient Joule heat dissipation in nanodevices.

<< Nanolab research highlight archive