We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cond-mat

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo

Condensed Matter > Strongly Correlated Electrons

Title: Non-analytic corrections to the Fermi-liquid behavior

Authors: A. V. Chubukov (1), D. L. Maslov (2) ((1) University of Wisconsin (2) University of Florida)
Abstract: The issue of non-analytic corrections to the Fermi-liquid behavior is revisited. Previous studies have indicated that the corrections to the Fermi-liquid forms of the specific heat and the static spin susceptibility scale as $T^{D}$ and $T^{D-1}$, respectively (with extra logarithms for $D=1,3$). In addition, the non-uniform spin susceptibility is expected to depend on the bosonic momentum $Q$ in a non-analytic way, i.e., as $Q^{D-1}$ (again with extra logarithms for $D=1,3$). It is shown that these non-analytic corrections originate from the universal singularities in the dynamical bosonic response functions of a generic Fermi liquid. In contrast to the leading, Fermi-liquid forms which depend on the interaction averaged over the Fermi surface, the non-analytic corrections are parameterized by only two coupling constants, which are the components of the interaction potential at momentum transfers $q=0$ and $q=2k_F$. For 3D systems, a recent result of Belitz, Kirkpatrick and Vojta for the spin susceptibility is reproduced and the issue why a non-analytic momentum dependence of the non-uniform spin susceptibility ($Q^{2}\ln |Q|$) is \emph{not}paralleled by a non-analyticity in the $T-$ dependence ($T^2$) is clarified. For the case of a 2D system with a finite-range interaction, explicit forms of the corrections to the specific heat ($\propto T^2$), uniform ($\propto T$) and non-uniform ($\propto |Q|$) spin susceptibilities are obtained. It is shown that previous calculations of the temperature dependences of these quantities in 2D were incomplete. Some of the results and conclusions of this paper have recently been announced in a short communication [A. V. Chubukov and D. L. Maslov, cond-mat/0304381].
Comments: 43 pages, 10 figures
Subjects: Strongly Correlated Electrons (cond-mat.str-el); Mesoscale and Nanoscale Physics (cond-mat.mes-hall)
DOI: 10.1103/PhysRevB.68.155113
Cite as: arXiv:cond-mat/0305022 [cond-mat.str-el]
  (or arXiv:cond-mat/0305022v1 [cond-mat.str-el] for this version)

Submission history

From: Dmitrii L. Maslov [view email]
[v1] Thu, 1 May 2003 18:54:12 GMT (551kb)

Link back to: arXiv, form interface, contact.