Skip to main navigation Skip to search Skip to main content

Magnetic quantum phase transitions in Kondo lattices

Research output: Contribution to journalArticlepeer-review

20 Scopus citations

Abstract

The identification of magnetic quantum critical points in heavy fermion metals has provided an ideal setting for experimentally studying quantum criticality. Motivated by these experiments, considerable theoretical efforts have recently been devoted to re-examining the interplay between Kondo screening and magnetic interactions in Kondo lattice systems. A local quantum critical picture has emerged, in which magnetic interactions suppress Kondo screening precisely at the magnetic quantum critical point (QCP). The Fermi surface undergoes a large reconstruction across the QCP and the coherence scale of the Kondo lattice vanishes at the QCP. The dynamical spin susceptibility exhibits ω/T scaling and non-trivial exponents describe the temperature and frequency dependences of various physical quantities. These properties are to be contrasted with the conventional spin density wave picture, in which the Kondo screening is not suppressed at the QCP and the Fermi surface evolves smoothly across the phase transition. In this article we discuss recent microscopic studies of Kondo lattices within an extended dynamical mean field theory (EDMFT). We summarize the earlier work based on an analytical ε-expansion renormalization group method, and expand on the more recent numerical results. We also discuss the issues that have been raised concerning the magnetic phase diagram. We show that the zero-temperature magnetic transition is second order when double counting of the Ruderman-Kittel-Kasuya-Yosida interactions is avoided in EDMFT.

Original languageEnglish
Pages (from-to)R1025-R1040
Number of pages16
JournalJournal of Physics: Condensed Matter
Volume17
Issue number37
DOIs
StatePublished - Sep 21 2005

Fingerprint

Dive into the research topics of 'Magnetic quantum phase transitions in Kondo lattices'. Together they form a unique fingerprint.

Cite this