arXiv subjects
Publications and source records attributed to Markus Garst.
We show that the comment [cond-mat/0408217] by Continentino on our recent paper [PRL 91, 066404 (2003), cond-mat/0212335] reaches incorrect conclusions as the comment wrongly extrapolates from results valid close to a classical phase transition into the quantum critical regime.
We investigate a model of two Kondo impurities coupled via an Ising interaction. Exploiting the mapping to a generalized single-impurity Anderson model, we establish that the model has a singlet and a (pseudospin) doublet phase separated by a Kosterlitz-Thouless quantum phase transition. Based on a strong-coupling analysis and renormalization group arguments, we show that at this transition the conductance G through the system either displays a zero-bias anomaly, G ~ |V|^{-2(\sqrt{2}-1)}, or takes a universal value, G = e^2/(π\hbar) cos^2[π/(2\sqrt{2})], depending on the experimental setup. Close to the Toulouse point of the individual Kondo impurities, the strong-coupling analysis allows to obtain the location of the phase boundary analytically. For general model parameters, we determine the phase diagram and investigate the thermodynamics using numerical renormalization group calculations. In the singlet phase close to the quantum phase transtion, the entropy is quenched in two steps: first the two Ising-coupled spins form a magnetic mini-domain which is, in a second step, screened by a Kondoesque collective resonance in an effective solitonic Fermi sea. In addition, we present a flow equation analysis which provides a different mapping of the two-impurity model to a generalized single-impurity Anderson model in terms of fully renormalized couplings, which is applicable for the whole range of model parameters.
At a generic quantum critical point, the thermal expansion $α$ is more singular than the specific heat $c_p$. Consequently, the "Grüneisen ratio'', $\GE=α/c_p$, diverges. When scaling applies, $\GE \sim T^{-1/(νz)}$ at the critical pressure $p=p_c$, providing a means to measure the scaling dimension of the most relevant operator that pressure couples to; in the alternative limit $T\to0$ and $p \ne p_c$, $\GE \sim \frac{1}{p-p_c}$ with a prefactor that is, up to the molar volume, a simple {\it universal} combination of critical exponents. For a magnetic-field driven transition, similar relations hold for the magnetocaloric effect $(1/T)\partial T/\partial H|_S$. Finally, we determine the corrections to scaling in a class of metallic quantum critical points.