Covariant gaussian approximation in Ginzburg–Landau model

J. F. Wang, D. P. Li, H. C. Kao, B. Rosenstein

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

Condensed matter systems undergoing second order transition away from the critical fluctuation region are usually described sufficiently well by the mean field approximation. The critical fluctuation region, determined by the Ginzburg criterion, |T/Tc−1|≪Gi, is narrow even in high Tc superconductors and has universal features well captured by the renormalization group method. However recent experiments on magnetization, conductivity and Nernst effect suggest that fluctuations effects are large in a wider region both above and below Tc. In particular some “pseudogap” phenomena and strong renormalization of the mean field critical temperature Tmf can be interpreted as strong fluctuations effects that are nonperturbative (cannot be accounted for by “gaussian fluctuations”). The physics in a broader region therefore requires more accurate approach. Self consistent methods are generally “non-conserving” in the sense that the Ward identities are not obeyed. This is especially detrimental in the symmetry broken phase where, for example, Goldstone bosons become massive. Covariant gaussian approximation remedies these problems. The Green's functions obey all the Ward identities and describe the fluctuations much better. The results for the order parameter correlator and magnetic penetration depth of the Ginzburg–Landau model of superconductivity are compared with both Monte Carlo simulations and experiments in high Tc cuprates.

Original languageEnglish
Pages (from-to)228-254
Number of pages27
JournalAnnals of Physics
Volume380
DOIs
Publication statusPublished - 2017 May 1

Keywords

  • Covariant gaussian approximation
  • Ginzburg–Landau
  • Magnetic penetration depth
  • Superconducting thermal fluctuations
  • Ward identity

ASJC Scopus subject areas

  • Physics and Astronomy(all)

Fingerprint Dive into the research topics of 'Covariant gaussian approximation in Ginzburg–Landau model'. Together they form a unique fingerprint.

  • Cite this