Table of Contents
Fetching ...

Semi-Holographic Fermi Liquids

Thomas Faulkner, Joseph Polchinski

TL;DR

The paper develops a semi-holographic framework in which a dynamical boundary fermion couples to a strongly interacting CFT with a gravity dual, capturing universal IR features of holographic non-Fermi liquids. By deriving the coupled χ–Ψ system, it shows how the IR sector, often described by an AdS$_2$ fixed point, governs non-Fermi-liquid behavior and how the model can be generalized to AdS$_4$ and Lifshitz geometries, including lattice and impurity effects. It analyzes a broad set of relevant operators—fermionic bilinears, density multilinears, and spin-orbit couplings—studying their impact on stability, fixed points, and potential phase transitions, including the possibility of alternate quantization near the Fermi surface. The work also discusses UV–IR matching: mapping UV AdS$_4$ data to IR AdS$_2$ physics, and demonstrates how double-trace deformations can tune or even erase the Fermi surface, highlighting a tunable, doping-like control of low-energy metallic states in a controlled holographic setting.

Abstract

We show that the universal physics of recent holographic non-Fermi liquid models is captured by a semi-holographic description, in which a dynamical boundary field is coupled to a strongly coupled conformal sector having a gravity dual. This allows various generalizations, such as a dynamical exponent and lattice and impurity effects. We examine possible relevant deformations, including multi-trace terms and spin-orbit effects. We discuss the matching onto the UV theory of the earlier work, and an alternate description in which the boundary field is integrated out.

Semi-Holographic Fermi Liquids

TL;DR

The paper develops a semi-holographic framework in which a dynamical boundary fermion couples to a strongly interacting CFT with a gravity dual, capturing universal IR features of holographic non-Fermi liquids. By deriving the coupled χ–Ψ system, it shows how the IR sector, often described by an AdS fixed point, governs non-Fermi-liquid behavior and how the model can be generalized to AdS and Lifshitz geometries, including lattice and impurity effects. It analyzes a broad set of relevant operators—fermionic bilinears, density multilinears, and spin-orbit couplings—studying their impact on stability, fixed points, and potential phase transitions, including the possibility of alternate quantization near the Fermi surface. The work also discusses UV–IR matching: mapping UV AdS data to IR AdS physics, and demonstrates how double-trace deformations can tune or even erase the Fermi surface, highlighting a tunable, doping-like control of low-energy metallic states in a controlled holographic setting.

Abstract

We show that the universal physics of recent holographic non-Fermi liquid models is captured by a semi-holographic description, in which a dynamical boundary field is coupled to a strongly coupled conformal sector having a gravity dual. This allows various generalizations, such as a dynamical exponent and lattice and impurity effects. We examine possible relevant deformations, including multi-trace terms and spin-orbit effects. We discuss the matching onto the UV theory of the earlier work, and an alternate description in which the boundary field is integrated out.

Paper Structure

This paper contains 11 sections, 53 equations, 4 figures.

Figures (4)

  • Figure 1: Fermi momentum as a function of radius, shown in the WKB approximation. a) The minimal ingredients to give rise to non-Fermi liquid behavior, a domain wall Fermi sea plus $AdS_2$ horizon. b) Over most of the parameter space of Ref. Faulkner:2009wj there is also a Fermi liquid in the IR bulk (entire shaded region). Backreaction converts the geometry to Lifshitz form HPST, whose Fermi sea is shown in dark shading.
  • Figure 2: a) The geometric sum leading to the fermion correlator (\ref{['gcorr']}). The solid line represents $G_0$ and the dashed line represents ${\cal G}_0$. b) The geometric sum (\ref{['gof']}), where an $\times$ represents the double-trace perturbation.
  • Figure 3: Left: flows represented by the $AdS_4$ RN geometry in the presence of a probe fermion when $\Delta_{\vec{k}} < 1$. The vertical direction is an energy scale and $\Lambda_k > \Lambda_k'$ for the two different flows. The standard (alternative) quantized theory is denoted $AdS_2^0$ ($AdS_2^\infty$). Right: the fermi surface with $\Delta_{\vec{k}} < 1$ is a quantum critical point in $k$ space, with $AdS_2^\infty$ controlling the physics.
  • Figure 4: Size of the fermi surfaces $k_\star^{\alpha=2}$ (solid) and $k_\star^{\alpha=1}$ (dashed) as a function of the relevant double-trace coupling $\texttt{x}$ in units of $\mu$ for three choices of $q$ and $\Delta_4 = 5/4$. The splitting due to spin orbit coupling increases with $q$, and is absent here for $q=0$. Note that the IR scaling dimension $\Delta_{\vec{k}_\star}$ increases with "doping" because it increases with $k_\star$.