Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

arXiv:cond-mat/9612134

Instability of the Non-Fermi-Liquid Fixed Point in the Dissipative Gauge Theory of Fermions (I)Impurity Effects

Hiroshi Takano, Masaru Onoda, Ikuo Ichinose, Tetsuo Matsui

Published Dec 14, 1996• 1 citation
arXivHTMLPDF

Abstract

We study a dissipative gauge theory of nonrelativistic fermions in 2+1 dimensions at zero temperature by the Wilsonian renormalization-group method. In this theory, we incorporate in the fermion propagator a new term of the form $ i κ\cdot\sign(ω)$ (where $κ$ is a parameter and $ω$ is the fermion frequency), which is usually induced by impurity effects. In the previous papers, we studied this system for $κ= 0$, and showed that there exists a non-Fermi-liquid infrared fixed point. In this paper, we address the question whether this non-Fermi-liquid behavior remains stable or not in the presence of impurity effects, i.e., the $κ$ term. Our results show that the non-Fermi-liquid fixed point is unstable for $κ\neq 0$ and an effective gauge coupling constant tends to vanish at low energies. However, in intermediate energy scales, the behavior of correlation functions for $κ\neq 0$ is controlled by the non-Fermi-liquid fixed point at $κ=0$, i.e., a crossover phenomenon appears. Physical implications of this phenomenon are discussed.

DOI
10.48550/arxiv.cond-mat/9612134

Related papers

Deviations from Fermi-liquid behavior in (2+1)-dimensional quantum electrodynamics and the normal phase of high- Tc superconductors

I. J. R. Aitchison, Nick E. Mavromatos

We argue that the gauge-fermion interaction in multiflavor quantum electrodynamics in (2+1) dimensions is responsible for non-Fermi-liquid behavior in the infrared, in the sense of leading to the existence of a nontrivial (quasi)fixed point that lies between the trivial fixed point (at infinite momenta) and the region where dynamical symmetry breaking and mass generation occurs. This quasifixed-point structure implies slowly varying, rather than fixed, couplings in the intermediate regime of momenta, a situation which resembles that of (four-dimensional) "walking technicolor" models of particle physics. The inclusion of wave-function renormalization yields marginal $O(\frac{1}{N})$ corrections to the "bulk" non-Fermi-liquid behavior caused by the gauge inter-action in the limit of infinite flavor number. Such corrections lead to the appearance of modified critical exponents. In particular, at low temperatures there appear to be logarithmic scaling violations of the linear resistivity of the system of order $O(\frac{1}{N})$. The connection with the anomalous normal-state properties of certain condensed-matter systems relevant for high-temperature superconductivity is briefly discussed. The relevance of the large (flavor) $N$ expansion to the Fermi-liquid problem is emphasized. As a partial result of our analysis, we point out the absence of charge-density-wave instabilities from the effective low-energy theory, as a consequence of gauge invariance.

arXiv:cond-mat/9603198•Physical review. B, Condensed matter•Apr 1, 1996•56 citations

Non-Fermi-liquid fixed point in a Wilsonian theory of quantum critical metals

A. Liam Fitzpatrick, Shamit Kachru A, Jared Kaplan A +1 more

We study the problem of disorder-free metals near a continuous quantum critical point. We depart from the standard paradigm [Hertz, Phys. Rev. B 14, 1165 (1976) and Millis, Phys. Rev. B 48, 7183 (1993)] and treat both fermions and bosons (i.e., order parameter fields) on equal footing. We construct a Wilsonian effective field theory that integrates out only high-energy boson and fermion modes. Below the upper critical dimension of the theory ($d=3$ spatial dimensions), we find new fixed points in which the bosons are described by the Wilson-Fisher fixed point and are coupled to a non-Fermi-liquid metal. We describe subtleties with the renormalization group flow of four-Fermi interactions, which can be surmounted in a controlled large-$N$ limit. In this limit, we find that the theory has no superconducting instability.

arXiv:1307.0004•Physical Review B•Sep 10, 2013•111 citations

Non-Fermi Liquid Behavior in 3+1 Dimensions with PT Invariant Gauge Field: An Renormalization Group Approach

Yan Gu

We introduce a Hamiltonian coupled between a normal Fermi surface and a polarized Maxwell type gauge field.We adopt a {\it calibrated scaling } approach in order to be consistent with the results obtained at $2+1$ dimensions as well as the Weinberg's theo rem. Renormalization group equations are calculated by using a dimension regulator. It turns out that the gauge interaction is controlled by a non-trivial fixed point. The two-point Green's function can thus be calculated accurately in the low temperature limit and possesses a non-quasi particle pole which signals the break down of the Landau Fermi liquid theory. In addition,we show that the anomalous dimension of fermion field is in general gauge dependent. But it will survive under generic gauge choice due to the renormalization of the gauge fixing parameter.

arXiv:cond-mat/9411124•Nov 29, 1994•0 citations

Deconfinement in the presence of a Fermi surface

Ki‐Seok Kim

U(1) gauge theory of nonrelativistic fermions interacting via compact U(1) gauge fields in the presence of a Fermi surface appears as an effective field theory in low dimensional quantum antiferromagnetism and heavy fermion liquids. We investigate deconfinement of fermions near the Fermi surface in the effective U(1) gauge theory. Our present analysis benchmarks the recent investigation of quantum electrodynamics in two space and one time dimensions $({\mathrm{QED}}_{3})$ by [Hermele et al.Phys. Rev. B 70, 214437 (2004)]. Utilizing a renormalization group analysis, we show that the effective U(1) gauge theory with a Fermi surface has a stable charged fixed point. Remarkably, the renormalization group equation for an internal charge $e$ [the coupling strength between nonrelativistic fermions and U(1) gauge fields] reveals that the conductivity $\ensuremath{\sigma}$ of fermions near the Fermi surface plays the same role as the flavor number $N$ of massless Dirac fermions in ${\mathrm{QED}}_{3}$. This leads us to the conclusion that if the conductivity of fermions is sufficiently large, instanton excitations of U(1) gauge fields can be suppressed owing to critical fluctuations of the nonrelativistic fermions at the charged fixed point. As a result a critical field theory of nonrelativistic fermions interacting via noncompact U(1) gauge fields is obtained at the charged fixed point.

arXiv:cond-mat/0502652•Physical Review B•Dec 7, 2005•23 citations

Non-Fermi liquid induced by U(1) gauge-field interactions: A functional renormalization group analysis

Thomas P. Sheerin, C. Hooley

We study the non-Fermi-liquid state formed by an isotropic, degenerate Fermi gas in two spatial dimensions interacting with a U(1) gauge field. Our calculation uses the functional renormalization group (fRG) with a soft frequency cutoff for the fermions. The fRG scheme we employ takes account of the gauge symmetry, which imposes relations (modified Ward-Takahashi identities) between the couplings which constrain the RG flow. The critical exponents and couplings we find for the resulting non-Fermi liquid are mostly insensitive to whether we enforce the gauge symmetry constraints, which signifies either that the constraints are superfluous or that the frequency-cutoff scheme is particularly robust. The exception is the gauge-boson mass term, which is RG-relevant about the fixed point without the constraints but is irrelevant when they are enforced. The latter is physically accurate, as a gauge symmetry cannot be spontaneously broken. In addition, we find z = 2 and Σ ( ω , k F ) ∼ ω 1 / 2 for the boson dynamical exponent and scaling of the fermion self-energy, respectively. These results differ considerably from those of past works on the model, though we argue for their plausibility.

arXiv:2411.18245•Physical Review Research•Jun 2, 2025•1 citation

Prospects for non-Fermi-liquid behavior of a two-level impurity in a metal

Aris L. Moustakas, Daniel S. Fisher

It has been claimed that an impurity hopping between two sites in a metal can lead to non-Fermi-liquid behavior if the bias between the two sites is tuned to zero. Here it is shown that several extra tunneling processes are allowed; those involving bare impurity tunneling as well as two-electron-assisted tunneling play important roles. These terms generally drive the system away from the intermediate-coupling non-Fermi-liquid fixed point, which has the character of the two-channel Kondo problem. In order to find the non-Fermi-liquid behavior a combination of these terms must be small over a range of energy scales. The prospects for obtaining these conditions are studied by analyzing both the weak-coupling and the intermediate-coupling behavior, together with the connecting crossover region. Unfortunately, the conclusions are rather pessimistic. The phase diagram and the important crossover are analyzed via bosonization and refermionization and some of the results compared to those obtained from conformal field theory. Finally, an additional non-Fermi-liquid fixed point is found, a multicritical point with two relevant directions even in the unbiased case. \textcopyright{} 1996 The American Physical Society.

arXiv:cond-mat/9508011•Physical review. B, Condensed matter•Feb 15, 1996•21 citations
We study a dissipative gauge theory of nonrelativistic fermions in 2+1 dimensions at zero temperature by the Wilsonian renormalization-group method. In this theory, we incorporate in the fermion propagator a new term of the form iκ⋅\sign(ω) i κ\cdot\sign(ω)iκ⋅\sign(ω) (where κκκ is a parameter and ωωω is the fermion frequency), which is usually induced by impurity effects. In the previous papers, we studied this system for κ=0κ= 0κ=0, and showed that there exists a non-Fermi-liquid infrared fixed point. In this paper, we address the question whether this non-Fermi-liquid behavior remains stable or not in the presence of impurity effects, i.e., the κκκ term. Our results show that the non-Fermi-liquid fixed point is unstable for κ≠0κ\neq 0κ=0 and an effective gauge coupling constant tends to vanish at low energies. However, in intermediate energy scales, the behavior of correlation functions for κ≠0κ\neq 0κ=0 is controlled by the non-Fermi-liquid fixed point at κ=0κ=0κ=0, i.e., a crossover phenomenon appears. Physical implications of this phenomenon are discussed.
We argue that the gauge-fermion interaction in multiflavor quantum electrodynamics in (2+1) dimensions is responsible for non-Fermi-liquid behavior in the infrared, in the sense of leading to the existence of a nontrivial (quasi)fixed point that lies between the trivial fixed point (at infinite momenta) and the region where dynamical symmetry breaking and mass generation occurs. This quasifixed-point structure implies slowly varying, rather than fixed, couplings in the intermediate regime of momenta, a situation which resembles that of (four-dimensional) "walking technicolor" models of particle physics. The inclusion of wave-function renormalization yields marginal O(1N)O(\frac{1}{N})O(N1​) corrections to the "bulk" non-Fermi-liquid behavior caused by the gauge inter-action in the limit of infinite flavor number. Such corrections lead to the appearance of modified critical exponents. In particular, at low temperatures there appear to be logarithmic scaling violations of the linear resistivity of the system of order O(1N)O(\frac{1}{N})O(N1​). The connection with the anomalous normal-state properties of certain condensed-matter systems relevant for high-temperature superconductivity is briefly discussed. The relevance of the large (flavor) NNN expansion to the Fermi-liquid problem is emphasized. As a partial result of our analysis, we point out the absence of charge-density-wave instabilities from the effective low-energy theory, as a consequence of gauge invariance.
We study the problem of disorder-free metals near a continuous quantum critical point. We depart from the standard paradigm [Hertz, Phys. Rev. B 14, 1165 (1976) and Millis, Phys. Rev. B 48, 7183 (1993)] and treat both fermions and bosons (i.e., order parameter fields) on equal footing. We construct a Wilsonian effective field theory that integrates out only high-energy boson and fermion modes. Below the upper critical dimension of the theory (d=3d=3d=3 spatial dimensions), we find new fixed points in which the bosons are described by the Wilson-Fisher fixed point and are coupled to a non-Fermi-liquid metal. We describe subtleties with the renormalization group flow of four-Fermi interactions, which can be surmounted in a controlled large-NNN limit. In this limit, we find that the theory has no superconducting instability.
We introduce a Hamiltonian coupled between a normal Fermi surface and a polarized Maxwell type gauge field.We adopt a {\it calibrated scaling } approach in order to be consistent with the results obtained at 2+12+12+1 dimensions as well as the Weinberg's theo rem. Renormalization group equations are calculated by using a dimension regulator. It turns out that the gauge interaction is controlled by a non-trivial fixed point. The two-point Green's function can thus be calculated accurately in the low temperature limit and possesses a non-quasi particle pole which signals the break down of the Landau Fermi liquid theory. In addition,we show that the anomalous dimension of fermion field is in general gauge dependent. But it will survive under generic gauge choice due to the renormalization of the gauge fixing parameter.
U(1) gauge theory of nonrelativistic fermions interacting via compact U(1) gauge fields in the presence of a Fermi surface appears as an effective field theory in low dimensional quantum antiferromagnetism and heavy fermion liquids. We investigate deconfinement of fermions near the Fermi surface in the effective U(1) gauge theory. Our present analysis benchmarks the recent investigation of quantum electrodynamics in two space and one time dimensions (QED3)({\mathrm{QED}}_{3})(QED3​) by [Hermele et al.Phys. Rev. B 70, 214437 (2004)]. Utilizing a renormalization group analysis, we show that the effective U(1) gauge theory with a Fermi surface has a stable charged fixed point. Remarkably, the renormalization group equation for an internal charge eee [the coupling strength between nonrelativistic fermions and U(1) gauge fields] reveals that the conductivity $\ensuremath{\sigma}$ of fermions near the Fermi surface plays the same role as the flavor number NNN of massless Dirac fermions in QED3{\mathrm{QED}}_{3}QED3​. This leads us to the conclusion that if the conductivity of fermions is sufficiently large, instanton excitations of U(1) gauge fields can be suppressed owing to critical fluctuations of the nonrelativistic fermions at the charged fixed point. As a result a critical field theory of nonrelativistic fermions interacting via noncompact U(1) gauge fields is obtained at the charged fixed point.
We study the non-Fermi-liquid state formed by an isotropic, degenerate Fermi gas in two spatial dimensions interacting with a U(1) gauge field. Our calculation uses the functional renormalization group (fRG) with a soft frequency cutoff for the fermions. The fRG scheme we employ takes account of the gauge symmetry, which imposes relations (modified Ward-Takahashi identities) between the couplings which constrain the RG flow. The critical exponents and couplings we find for the resulting non-Fermi liquid are mostly insensitive to whether we enforce the gauge symmetry constraints, which signifies either that the constraints are superfluous or that the frequency-cutoff scheme is particularly robust. The exception is the gauge-boson mass term, which is RG-relevant about the fixed point without the constraints but is irrelevant when they are enforced. The latter is physically accurate, as a gauge symmetry cannot be spontaneously broken. In addition, we find z = 2 and Σ ( ω , k F ) ∼ ω 1 / 2 for the boson dynamical exponent and scaling of the fermion self-energy, respectively. These results differ considerably from those of past works on the model, though we argue for their plausibility.
It has been claimed that an impurity hopping between two sites in a metal can lead to non-Fermi-liquid behavior if the bias between the two sites is tuned to zero. Here it is shown that several extra tunneling processes are allowed; those involving bare impurity tunneling as well as two-electron-assisted tunneling play important roles. These terms generally drive the system away from the intermediate-coupling non-Fermi-liquid fixed point, which has the character of the two-channel Kondo problem. In order to find the non-Fermi-liquid behavior a combination of these terms must be small over a range of energy scales. The prospects for obtaining these conditions are studied by analyzing both the weak-coupling and the intermediate-coupling behavior, together with the connecting crossover region. Unfortunately, the conclusions are rather pessimistic. The phase diagram and the important crossover are analyzed via bosonization and refermionization and some of the results compared to those obtained from conformal field theory. Finally, an additional non-Fermi-liquid fixed point is found, a multicritical point with two relevant directions even in the unbiased case. \textcopyright{} 1996 The American Physical Society.