Research
My research is in the field of condensed matter theory, with particular focus on many-body theories of strongly-correlated electron systems. Below is a summary of some recent research projects. A full list of publications can be found here.
Kondo correlations in real-space
Most fundamental aspects of the Kondo effect are by now very well understood, with various detailed theoretical predictions having been confirmed directly by experiments on impurity systems or quantum dot devices. Key insights into the underlying physics have been provided by the renormalization group (RG) concept, where progressive reduction of the temperature or energy scale results in RG flow between 'fixed points' (FPs) that can be easily identified for a given model. Surprisingly however, the basic physics in real-space is still somewhat controversial, although the emergence of a fundamental length-scale RK ~ 1/TK (with TK the Kondo temperature) has long been predicted. At low temperatures T << TK, the standard interpretation is that a spin-1/2 impurity is screened by a surrounding 'Kondo cloud' of spatial extent RK. We argue in Phys. Rev. B 84, 115120 (2011) that RG flow between any two FPs results in a characteristic length-scale, observed in real-space as a crossover between physical behavior typical of each FP. We show explicitly that 'free orbital', 'local moment' and 'strong coupling' regions of space can be identified for a single Anderson impurity embedded in a 1d host. These distinct regions are separated by two crossover length-scales RLM and RK, with the latter diverging as the Kondo effect is destroyed on increasing temperature through TK. One implication is that moment formation occurs inside the Kondo cloud', while the screening process itself occurs on flowing to the strong coupling FP at distances ~RK. In ongoing work, we hope to generalize these results to more complex impurity systems and 2d or 3d hosts.Critical non-Fermi liquid physics in quantum dot devices
The most basic quantum impurity model exhibiting non-Fermi liquid (NFL) behavior is arguably the two-channel Kondo (2CK) model, describing the symmetric antiferromagnetic coupling of a local spin-1/2 impurity to two equivalent but independent conduction channels. The resulting ground state possesses various intriguing properties, including notably a residual entropy of (kB/2) ln(2) and conductance that approaches its T=0 value as the square root of T.Similar behavior is predicted at the critical point of the two-impurity Kondo (2IK) model. The tendency to form a trivial local singlet state is favored by an exchange coupling acting directly between the impurities; while the coupling of each impurity to its own metallic lead favors separate single-channel Kondo screening. The resulting competition gives rise to an unusual quantum critical point.
In Phys. Rev. Lett. 108, 086405 (2012), we study the NFL quantum critical state of the generalized spin-S 2IK model, and its potential realization in a quantum dot device. Using conformal field theory (CFT) and the numerical renormalization group (NRG), we show the critical point to be identical to that of the 2CK model with additional potential scattering, for any spin-S. Distinct conductance signatures are shown to arise as a function of device asymmetry; with the 'smoking gun' square-root behavior, commonly believed to arise at low-energies, dominant only in certain asymmetric regimes. Exact conductance lineshapes are calculated numerically for our proposed realization.
However, the central difficulty in observing experimentally the NFL physics of either 2CK or 2IK models is their extreme sensitivity to various symmetry-breaking perturbations. For example, channel asymmetry, magnetic field and inter-lead charge transfer processes destabilize the critical point and destroy NFL behavior in both cases. In Phys. Rev. Lett. 106, 147202 (2011) we study the T=0 crossover from NFL to more standard Fermi-liquid physics which inevitably results at low energies. We exploit a connection between the crossover arising in the 2IK model due to a detuning perturbation and the simpler 1d quantum critical boundary Ising model, to calculate the full crossover Green function analytically. A large SO(8) emergent symmetry at the critical point also allows us to generalize the results to other perturbations. Indeed, the same crossover must also appear in the 2CK model due to the connection between the 2IK and 2CK models. We thus obtain a single exact universal crossover Green function that applies for an arbitrary mixture of any relevant perturbations in either model. The results are in remarkable agreement with full NRG calculations.
To generalize these results to finite temperature T = 1/beta, we require exact results for the 1d quantum critical boundary Ising model at this temperature. The model is equivalent to a classical 2d Ising model on a semi-infinite cylindrical geometry, with a magnetic field applied at the circular boundary of circumference proportional to beta. Using CFT methods, we re-examine this system in J. Phys. Stat. Mech. P04006 (2012), obtaining the full scaling function for the local magnetization crossover analytically in the continuum limit. The validity of our result as the continuum limit of the 1d lattice model is confirmed numerically, exploiting a modified Jordan-Wigner representation.
In Phys. Rev. B 85, 235127 (2012), we then proceed to derive an exact expression for the electron Green function in two-channel Kondo models with one and two impurities, describing the crossover from NFL behavior at intermediate temperatures to standard Fermi liquid physics at low temperatures. Symmetry-breaking perturbations generically present in experiment ensure the standard low-energy FL description, but the full crossover is shown to be wholly characteristic of the unstable NFL state. Distinctive conductance lineshapes in quantum dot devices are shown to result. Excellent agreement is demonstrated between exact results and full NRG calculations.
Finally, in Phys. Rev. B 84, 035119 (2011) and Phys. Rev. B 81, 075126 (2010), we consider emergent 2CK physics in generic odd-membered quantum dot chains and rings. In chains, two overscreening mechanisms are found to occur depending on coupling strength, with distinct signatures in physical properties. 2CK physics is shown to be wholly robust to variable dot filling; in particular the single-particle spectrum at the Fermi level, and hence the low-temperature zero-bias conductance, is always pinned to half-unitarity. We derive a Friedel-Luttinger sum rule and from it show that, in contrast to a Fermi liquid, the Luttinger integral is non-zero and determined solely by the 'excess' dot charge as controlled by gate voltage. In rings, a frustration-induced quantum phase transition between non-Fermi liquid phases is supported.
