Holographic Superconductors

⚡ Key Takeaways

Holographic superconductors are gravitational models in an asymptotically Anti-de Sitter (AdS) spacetime that, via the AdS/CFT Correspondence, describe a strongly-coupled boundary field theory undergoing a phase transition to a superconducting (or superfluid) state. The key result: below a critical temperature $T_c$ , a charged scalar field condenses around a planar black hole, spontaneously breaking a $U(1)$ symmetry and producing an infinite DC conductivity and a gap in the optical conductivity. They apply to strongly-correlated electron systems (high- $T_c$ cuprates, quantum critical matter) where conventional BCS theory fails. Introduced by Gubser (2008) and Hartnoll, Herzog & Horowitz (2008).

💡 The Core Insight

Superconductivity is the spontaneous breaking of a $U(1)$ symmetry signaled by a charged condensate. In the bulk, a global symmetry of the boundary becomes a gauged $U(1)$ (Maxwell field). Near a charged black hole, a sufficiently charged scalar field experiences an effective negative mass-squared (from the gauge potential) that, at low temperature, drops below the Breitenlohner–Freedman stability bound near the horizon. This destabilizes the scalar-free "normal phase" black hole, and the scalar acquires a nontrivial profile — hair — that asymptotes to a nonzero condensate value on the boundary. The same gravitational instability that nucleates the hair produces, in the conductivity computation, a delta-function in $\mathrm{Re}\,\sigma(\omega)$ at $\omega=0$ : an infinite DC conductivity.

DefinitionThe Abelian Higgs model in AdS

The minimal holographic superconductor (Hartnoll–Herzog–Horowitz, "HHH") is defined by the bulk action in $3+1$ dimensions (dual to a $2+1$ field theory):

1

\quad S = \int d^4x\,\sqrt{-g}\left[\frac{1}{2\kappa^2}\left(R + \frac{6}{L^2}\right) - \frac{1}{4}F_{\mu\nu}F^{\mu\nu} - |\partial_\mu\psi - iqA_\mu\psi|^2 - m^2|\psi|^2\right]

Derivation hint: take Einstein–Hilbert with negative cosmological constant $\Lambda=-3/L^2$ giving the $+6/L^2$ term in 4D, add a Maxwell kinetic term, and a complex scalar $\psi$ of charge $q$ and mass $m$ minimally coupled via the gauge-covariant derivative. Here $L$ is the AdS radius, $R$ the Ricci scalar, $\kappa^2=8\pi G$ , $F=dA$ , $\psi$ the charged scalar dual to the order-parameter operator $\mathcal{O}$ , and $q$ its charge. The scalar mass sets the conformal dimension of $\mathcal{O}$ via

2

\quad \Delta_\pm = \frac{d}{2} \pm \sqrt{\frac{d^2}{4} + m^2 L^2}, \qquad d=3

Derivation hint: solve the free scalar wave equation in pure AdS near the boundary $z\to 0$ , where $\psi \sim \psi_- z^{\Delta_-} + \psi_+ z^{\Delta_+}$ ; the exponents satisfy $\Delta(\Delta-d)=m^2L^2$ .

DefinitionProbe limit and planar black hole

In the probe limit ( $q\to\infty$ with $q\psi$ , $qA$ fixed, equivalently $\kappa^2\to 0$ ) the matter fields do not backreact, and one works on the fixed AdS–Schwarzschild planar black brane:

3

\quad ds^2 = \frac{L^2}{z^2}\left(-f(z)\,dt^2 + \frac{dz^2}{f(z)} + dx^2 + dy^2\right), \qquad f(z) = 1 - \left(\frac{z}{z_h}\right)^3

Derivation hint: solve the vacuum Einstein equations with $\Lambda<0$ demanding planar horizon topology; $z=z_h$ is the horizon, $z=0$ the AdS boundary, and the blackening factor $f$ has a simple zero at $z_h$ . The Hawking temperature, identified with the field theory temperature, is

4

\quad T = \frac{|f'(z_h)|}{4\pi} = \frac{3}{4\pi z_h}

Derivation hint: Wick-rotate $t\to -i\tau$ , demand regularity (absence of conical singularity) of the Euclidean metric at $z_h$ ; periodicity of $\tau$ is $\beta = 1/T = 4\pi/|f'(z_h)|$ .

TheoremCondensation criterion / instability

With ansatz $\psi=\psi(z)$ real, $A_t=\phi(z)$ , $A_i=0$ , the scalar equation of motion in the probe limit is

5

\quad \psi'' + \left(\frac{f'}{f} - \frac{2}{z}\right)\psi' + \left(\frac{q^2\phi^2}{f^2} - \frac{m^2 L^2}{z^2 f}\right)\psi = 0

Derivation hint: vary the action [1] in the matter sector on background [3], use $\sqrt{-g}=L^4/z^4$ , and take static, homogeneous, real fields. The term $q^2\phi^2/f^2$ acts as a position-dependent negative contribution to the effective mass. A nontrivial regular solution with $\psi_-=0$ (source-free, spontaneous) exists for $T<T_c$ . The transition is continuous (second order) with mean-field exponent in the probe/large- $N$ limit:

6

\quad \langle \mathcal{O}_{\Delta}\rangle \propto T_c^{\Delta}\left(1 - \frac{T}{T_c}\right)^{1/2}, \qquad T\to T_c^-

Derivation hint: expand [5] near $T_c$ where $\psi$ is infinitesimal; the linearized problem is an eigenvalue equation fixing $T_c$ , and Landau expansion of the free energy gives the $1/2$ exponent.

ProofInstability of the normal phase

Step 1 — Normal phase solution. Set $\psi=0$ . The Maxwell equation on background [3] becomes $\phi'' - \tfrac{2}{z}\phi'\cdot(\text{coeff})=0$ ; solving with regularity ( $\phi(z_h)=0$ ) and boundary data fixes

\phi(z) = \mu - \rho\, z = \mu\left(1 - z/z_h\right),

where $\mu$ is the chemical potential and $\rho$ the charge density (read off from $\phi \sim \mu - \rho z$ near $z\to0$ , with $\rho = \mu/z_h$ for the simplest 4D case).

Step 2 — Effective mass near horizon. Linearize [5] in $\psi$ . The effective mass squared is

m^2_{\rm eff}(z) = m^2 - \frac{q^2\phi(z)^2 L^2}{z^2 f(z)}\cdot z^2 .

The negative gauge contribution is largest deep in the bulk. As $T$ decreases ( $z_h$ grows), $\phi$ near the horizon is larger relative to the curvature scale, and $m^2_{\rm eff}$ violates the near-horizon Breitenlohner–Freedman bound of the emergent $AdS_2\times \mathbb{R}^2$ geometry that governs the extremal/near-extremal limit.

Step 3 — BF bound violation. Near a charged horizon the geometry develops an $AdS_2$ throat with radius $L_2$ . A charged scalar there has effective mass

m^2_{\rm eff}L_2^2 = m^2 L_2^2 - q^2 e_d^2,

with $e_d$ the near-horizon electric field. The $AdS_2$ BF bound is $m^2_{\rm eff}L_2^2 \ge -1/4$ . When

m^2 L_2^2 - q^2 e_d^2 < -\tfrac14,

the scalar is unstable, scalar hair forms, and the system superconducts. Large charge $q$ guarantees instability at low $T$ .

Step 4 — Endpoint. The instability saturates in a new branch with $\psi\ne0$ , regular at the horizon, with $\psi_-=0$ at the boundary (no explicit symmetry-breaking source). This is the superconducting ground state, free-energetically favored below $T_c$ . ∎

∎

SG

Steven Gubser, Sean Hartnoll, Christopher Herzog, Gary Horowitz

2008–

Steven Gubser (1972–2019) proposed in early 2008 that a charged scalar could condense outside an AdS black hole, evading no-hair theorems, and argued this would describe a superconductor. Hartnoll, Herzog and Horowitz then constructed the explicit model, computed the condensate and the frequency-dependent conductivity, and demonstrated the gap and infinite DC conductivity — establishing the field of "AdS/CMT" (applied holography for condensed matter). Gubser, a leading figure in AdS/CFT since its inception, was central to the holographic c-theorem and the application of holography to QCD and condensed matter; his early death cut short a decisive career.

🧠 Intuition

Think of the black hole as a hot, strongly-interacting plasma. The chemical potential $\mu$ (set by the boundary value of $A_t$ ) charges the system. At high temperature the thermal black hole "eats" any scalar fluctuation — entropy wins, no condensate. As you cool, the horizon retreats and the electric field near it strengthens; a charged scalar quantum feels the gauge potential as an effective attractive potential well, strong enough to bind it into a static cloud — hair — hovering outside the horizon. The hair's tail reaching the boundary is the order parameter $\langle\mathcal{O}\rangle$ : a condensate. Because this condensate carries the broken $U(1)$ charge coherently, a DC current can flow without dissipation — the boundary is superconducting. The "gap" in the optical conductivity is the energy needed to break a constituent of the condensate, mirroring the BCS gap but with no Cooper-pair microphysics — only geometry.

Figure

Condensate

\langle\mathcal O\rangle

and onset of scalar hair as temperature drops below

T_c

```svg

```

ExampleCritical temperature scaling from charge density

Given: A 2+1 dimensional holographic superconductor in the probe limit with operator dimension $\Delta=2$ ( $m^2L^2=-2$ ), unit charge, fixed charge density $\rho$ .

Steps: 1. The only scale at the transition is set by $\rho$ (and $q$ ). By dimensional analysis in the conformal boundary theory, $[\rho]=2$ (mass dimension 2 in 2+1d) so $T_c \propto \rho^{1/2}$ . 2. The numerical solution of the linearized eigenvalue problem [5] at $\psi\to0$ gives, for $\Delta=2$ ,

T_c \approx 0.118\,\sqrt{\rho}.

3. Suppose $\rho = (100\ \text{meV})^2$ in natural units. Then $\sqrt{\rho}=100$ meV.

Answer:

\boxed{T_c \approx 0.118 \times 100\ \text{meV} = 11.8\ \text{meV} \approx 137\ \text{K}}

(using $1\ \text{meV} \approx 11.6\ \text{K}$ ). The scaling $T_c\propto\sqrt\rho$ — not the exponential BCS suppression — is a signature of strong coupling.

ExampleCondensate near the transition

Given: Near $T_c$ , the $\Delta=2$ condensate behaves as $\langle\mathcal{O}_2\rangle \approx 144\, T_c^2 (1-T/T_c)^{1/2}$ (HHH numerical coefficient). Take $T_c = 137$ K $=11.8$ meV and evaluate at $T = 0.96\,T_c$ .

Steps: 1. $T_c^2 = (11.8\ \text{meV})^2 = 139.2\ \text{meV}^2$ . 2. $1 - T/T_c = 0.04$ , so $(0.04)^{1/2}=0.2$ . 3. $\langle\mathcal{O}_2\rangle \approx 144 \times 139.2 \times 0.2\ \text{meV}^2$ .

Answer:

\boxed{\langle\mathcal{O}_2\rangle \approx 4.01\times10^{3}\ \text{meV}^2 \;(\text{i.e. }\sqrt{\langle\mathcal O_2\rangle}\approx 63\ \text{meV})}

The square-root onset is the mean-field exponent, exact in the large- $N$ (classical gravity) limit.

⚠️ Common mistake

The bulk $U(1)$ is gauged; the boundary $U(1)$ is global. A genuine superconductor breaks a gauged electromagnetic symmetry; the standard holographic model breaks a global $U(1)$ , so it is strictly a charged superfluid. To make it a true superconductor one must gauge the boundary symmetry (dynamical photon), which is a subleading $1/N$ effect. The Meissner effect, London penetration depth, and vortices require this additional step or external treatment.

⚠️ Common mistake

Infinite DC conductivity here is partly kinematic. Even the normal phase shows a $\delta(\omega)$ in $\mathrm{Re}\,\sigma$ because translation invariance forbids momentum relaxation — a charged fluid in a clean, homogeneous medium accelerates freely. The superconducting $\delta(\omega)$ is distinguished by the appearance of a gap and by the Ferrell–Glover–Tinkham sum rule transferring spectral weight from finite frequency to the zero-frequency pole. To isolate true superconductivity one must break translations (lattices, momentum relaxation).

Counterexample

Not every charged scalar condenses. If the scalar charge $q$ is too small and $m^2$ too large (close to the upper, $\Delta_+$ window), the near-horizon BF bound is never violated and no superconducting instability occurs down to $T=0$ ; the extremal black hole remains hairless. Concretely, a neutral scalar ( $q=0$ ) with $m^2 L^2 > -3/2$ on a Reissner–Nordström-AdS background does not condense via this mechanism — demonstrating that the gauge coupling, not mere tachyonicity, drives the transition.

connection

Holographic superconductors sit at the intersection of gravitational physics and strongly-correlated matter. They are an application of the AdS/CFT Correspondence and rest on the Holographic Principle. The instability mechanism uses the Breitenlohner-Freedman Bound for scalars in AdS. The dual interpretation of bulk fields as boundary operators uses the Holographic Dictionary and the GKP-Witten Relation. The normal-phase background is the Reissner-Nordström-AdS Black Hole, and the thermodynamics relies on Hawking Temperature and Black Hole Thermodynamics. The order parameter and broken symmetry connect to Spontaneous Symmetry Breaking and, in the boundary theory, to Ginzburg-Landau Theory. The conductivity computation uses Linear Response Theory and Kubo Formula. As a model of non-BCS pairing it contrasts with BCS Theory and contacts Quantum Phase Transitions and Non-Fermi Liquids.

🔬 Where this appears

⏳ History

- 1997–1998 — Maldacena, GKP, Witten: the AdS/CFT Correspondence established, providing the gauge/gravity dictionary that makes all of this possible.

- 1999 — Hawking–Hertog, Breitenlohner–Freedman revisited: stability of scalars in AdS and the BF bound clarified for negative

m^2

.

- 2008 (March) — S. Gubser: "Breaking an Abelian gauge symmetry near a black hole horizon" proposes that charged scalar hair forms and describes a superconductor; identifies the near-horizon instability.

- 2008 (July) — Hartnoll, Herzog, Horowitz: "Building a Holographic Superconductor" constructs the explicit HHH model, computes the condensate, second-order transition, and the optical conductivity with a gap; founds AdS/CMT.

- 2008 — Hartnoll, Herzog, Horowitz (longer paper): full backreacted analysis, fermion response, and detailed conductivity sum rules.

- 2009 — Gubser–Pufu; Roberts–Hartnoll: p-wave (vector) holographic superconductors via condensing a charged

SU(2)

gauge field; anisotropic conductivity.

- 2010 — Horowitz–Roberts; Faulkner et al.: zero-temperature limits (emergent

AdS_2

, Lifshitz, conformal-to-AdS

_4

ground states); marginal Fermi liquid behavior from

AdS_2

.

- 2012 — Horowitz, Santos, Tong: holographic lattices break translation invariance, yielding finite DC conductivity and clean separation of the superfluid pole.

- 2013–2015 — Donos–Gauntlett: momentum relaxation via massive gravity / linear axions; analytic DC conductivities and incoherent metal physics.

open-problem

Beyond the probe and large- $N$ limits. The clean mean-field exponents and the $\eta/s=1/4\pi$ universality are artifacts of classical gravity (infinite $N$ , infinite coupling). Computing $1/N$ (quantum gravity / loop) corrections to $T_c$ , critical exponents, and conductivity remains technically formidable. What is the order parameter microscopically? Holography gives no Cooper-pair picture; whether a "pairing glue" interpretation exists for strange-metal superconductivity is open. True superconductor vs. superfluid: consistently gauging the boundary photon to recover the Meissner effect and finite penetration depth at leading order is subtle. Translation symmetry and DC transport: realistic resistivity requires momentum relaxation; the precise mechanism reproducing the observed linear- $T$ resistivity over decades of temperature, and its connection to a possible bound on diffusivity / Planckian dissipation $\tau\gtrsim \hbar/k_BT$ , is actively debated. Stability and competing orders: which ground state truly minimizes the free energy when multiple condensates (s, p, d-wave, charge-density-wave) are allowed, and whether holographic strange metals are intrinsically unstable to such orders, is unsettled.

Formal Statement

Why It Holds / Derivation

Building Intuition

Worked Examples

Common Misconceptions

Connections

Real-World Applications

Historical Development

Subtleties & Limits

See Also