Spinor spectra on holographic screens: a polar-decomposition WKB test of screen thermodynamics

Some approaches to gravity treat it as thermodynamics. In these approaches, a surface of constant gravitational potential around a black hole, called a holographic screen, carries a temperature and an entropy, and Einstein's equations follow from a first law of thermodynamics written on that surface. The temperature, the entropy $S = A/4$ (with $A$ the screen area) and the phase transitions on these screens are all postulated at the macroscopic level. Nobody has computed them from the spectrum of an actual quantum field living near the screen. This note proposes that computation for charged anti-de Sitter (AdS) black holes, for both integer-spin (scalar) and half-integer-spin (Dirac) fields. The tool is WKB analysis, a semiclassical method that writes a wave as a slowly varying amplitude times a rapidly oscillating phase and then reads off eigenvalues from a quantization condition.

The main technical step concerns the Dirac field. Its spin direction is part of the "phase", so a single complex number cannot describe it. We write the semiclassical spinor using the native polar decomposition of Clifford algebra (the algebra of vectors and their oriented products): a spinor factors as a unitary multivector, which acts as a rotation of the spin frame, times a positive amplitude. The WKB phase is then promoted from a scalar to a rotation. This keeps the sign of the energy, which squaring the Dirac operator would discard, and that sign is what couples the fermion spectrum to the electrostatic potential $\Phi$. $\Phi$ is also the parameter that controls where screen phase transitions occur. The deliverable is a falsifiable comparison. Do the heat-capacity divergences computed from the field's spectrum sit at the locations predicted by macroscopic screen thermodynamics, and do bosons and fermions agree?

Core Claim

Consider a charged AdS black hole. Bound a region by the equipotential screen $e^{2\phi} = c$ on the inside and by the AdS boundary on the outside. AdS acts as a box, so a field confined to this region has a discrete spectrum, and no horizon cutoff is needed. A uniform WKB calculation, matched across turning points, can produce this spectrum accurately enough to locate divergences in the heat capacity of the field system. For the Dirac field, the right semiclassical object is

$$\psi(x) \approx P(x)\, W(x)\, e^{I\, S(x)/\epsilon},$$

with the following pieces. $P = P^\dagger$ is a positive multivector amplitude. $W$ satisfies $W^\dagger W = 1$ and rotates the local spin frame. $I$ is the unit pseudoscalar of the spacetime algebra, playing the role of the imaginary unit. $S$ is the classical action. $\epsilon$ is the small semiclassical parameter, roughly the inverse mode number. Substituting this form into the Dirac equation gives three things at successive orders:

  • an eikonal (Hamilton-Jacobi) equation for $S$;
  • a transport equation for $P$;
  • a separate transport equation for $W$, whose accumulated rotation adds a spin-dependent shift $\gamma_{\text{spin}}$ to the quantization condition.

That quantization condition is

$$\oint p_r\, dr = \pi\left(n + \tfrac{1}{2}\right) + \gamma_{\text{spin}}(\omega, q\Phi, c),$$

where $p_r$ is the local radial momentum, $n$ is the mode index and $\omega$ is the mode frequency.

The claim has two parts. The first is that this rotor-valued WKB reproduces high-precision numerical Dirac eigenvalues, including the sign asymmetry between particle and antiparticle modes induced by $q\Phi$. The second is that the resulting spectral heat capacity is a quantitative probe of whether the macroscopic phase boundaries of screen thermodynamics have a microscopic counterpart. We commit to reporting either outcome.

Where the pieces come from

Macroscopic screen results (what we test against). Three papers define the target.

  • First law of thermodynamics on holographic screens in entropic force frame builds a first law $dM = T\,dA + \sum \Phi_i dQ_i$ on equipotential screens around spherically symmetric black holes. It defines temperature through the acceleration of observers sitting on the screen and assumes $S = A/4$.
  • Phase structures of holographic screen themodynamics extends this to Reissner-Nordström-AdS (charged AdS) black holes. It shows the screen Smarr formula is equivalent to the radial Einstein equation. It locates heat-capacity divergences:
  • at $0 \le c < 1$ for neutral screens;
  • at $0 \le c + 16\tilde\Gamma^2\Phi^2 < 1$ at fixed electrostatic potential, where $\tilde\Gamma$ is a dimension-dependent constant defined in that paper.

Both papers stay entirely macroscopic, and that is the gap we fill. We do not try to fix their acknowledged ambiguity in rewriting the metric function $f(r)$ in thermodynamic terms. We also do not attempt rotating screens.

Semiclassical method (what we reuse). Two papers supply the technique.

Both treat scalar problems only. Our extension to a spin-valued phase is new.

Algebraic vocabulary (what makes the spinor phase well defined). On SVD and Polar Decomposition in Real and Complexified Clifford Algebras proves that any multivector factors as $M = WP$, with $W$ unitary and $P$ positive semi-definite. The factorization holds in any signature and uses no matrix representation. This lets us define "amplitude" and "phase" of a spinor without choosing components.

That paper offers existence proofs only, with no algorithms. For the low-dimensional spacetime algebra we need only a closed-form polar factor along the WKB path, and On Rank of Multivectors in Geometric Algebras gives the basis-free characteristic polynomial of $M^\dagger M$. Its positive square root yields $P$, and $W = MP^{-1}$. The same characteristic polynomial detects rank drops where $P$ becomes singular. Those points are caustics, where the WKB form breaks down.

Prior work that constrains the design. Entropy Modifications of Charged Accelerating Anti-de Sitter Black Hole uses one-dimensional radial WKB tunneling across a null horizon. It finds that boson entropy corrections are algebraic, while fermions acquire a logarithmic term $\hbar\beta_1 \ln S$. This motivates treating spin explicitly. It also differs from our setup: it works on the horizon, not on a timelike screen, and it uses no spectrum.

Universal Relation between Corrections to Entropy and Extremality proves that $\Delta M_{\text{ext}} \approx -T_0\,\Delta S$ for any small correction, where $\Delta M_{\text{ext}}$ is the shift in the minimum mass at fixed charge and $T_0$ is the uncorrected temperature. Any spectrally computed entropy correction must obey this sign rule, so it serves as an internal consistency check.

Research Program

The work proceeds in three stages.

Stage 1: the rotor-valued WKB. We derive the polar-decomposition WKB for the Dirac operator in Clifford algebra $Cl_{1,3}$ on the static charged AdS background in $3+1$ dimensions, with screen boundary conditions at $e^{2\phi} = c$. The Dirac field uses bag-type conditions, which forbid current through the screen; the scalar uses Dirichlet conditions. Away from turning points we solve the eikonal and the two transport equations. Near the turning points set by the AdS potential and the screen wall, we match to Airy or Bessel solutions, following the turning-point matching of the truncated-Hilbert-transform analysis. The output is the quantization condition above, with an explicit $\gamma_{\text{spin}}$.

Stage 2: spectra to thermodynamics. We sum the WKB spectra into a grand-canonical free energy,

$$F(\beta, c, \Phi) = \mp \beta^{-1} \sum_{n,\ell,\pm} \ln\left(1 \pm e^{-\beta(\omega_{n\ell} - q\Phi)}\right),$$

taking the upper signs for fermions and the lower signs for bosons. Here $\beta$ is the inverse of the local screen temperature defined by Chen and Li, and $\ell$ is angular momentum. From $F$ we obtain entropy and heat capacity as functions of black-hole mass $M$ at fixed $c$, and separately at fixed $\Phi$ or fixed $Q$.

Stage 3: ground truth and comparison. We compute reference spectra by high-precision Chebyshev collocation of the same operators. This lets us measure WKB error mode by mode and propagate it into the heat-capacity curves.

Baselines

Naive reference. The do-nothing reference is the macroscopic screen thermodynamics of Phase structures of holographic screen themodynamics. We supplement it with the leading Weyl-law density of states, the standard high-frequency count of modes by phase-space volume with no turning-point structure. This baseline asks whether the field contributes anything beyond volume-scaling terms that shift no phase boundary. Our calculation beats it if it predicts boundary shifts that the Weyl law misses and that the numerical spectra confirm.

Strongest competitor. The strongest competitor is standard multicomponent WKB. In that method the Dirac Hamiltonian's leading-order matrix symbol is diagonalized pointwise in a chosen 4-component representation, and the Berry-type correction is computed from the eigenvector derivatives. We implement it on the same radial grid, with the same turning-point matching and the same number of retained correction orders, and compare it mode by mode against the collocation reference. The two methods should agree at leading order, because they encode the same physics. A gain for our formulation would mean one of two things:

  • lower eigenvalue error at small $n$ and near caustics, where pointwise diagonalization requires continuous frame choices that can jump;
  • the same accuracy obtained with no representation-dependent gauge fixing.

If the two agree to numerical precision everywhere, we report that the native formulation is a notational gain only, and the thermodynamic result stands on its own.

Key ablation. The ablation that isolates the spin rotor is scalar WKB on the squared Dirac operator, with $\gamma_{\text{spin}}$ dropped. Squaring mixes the energy dependence of the electrostatic coupling and discards the sign of $\omega - q\Phi$. We therefore expect this ablation to fail specifically in the fixed-$\Phi$ ensemble.

Direct numerical diagonalization. Direct diagonalization is the reference, not a competitor. Its role in the comparison is cost and stability: we record runtime and loss of precision as mode number grows. The truncated-Hilbert-transform analysis argues that exponentially small spectral quantities are where WKB earns its keep.

Reported numbers. No reported result was found on Lacuna for numerical spinor or scalar spectra on holographic screens in charged AdS. The only quantitative targets in the file are:

  • the analytic phase conditions $0 \le c < 1$ and $0 \le c + 16\tilde\Gamma^2\Phi^2 < 1$ from the phase-structures paper;
  • the truncated-Hilbert-transform singular-value asymptotics $\sigma_n = 2e^{-\eta\pi K_+/K_-}(1 + O(n^{-1/2+\delta}))$, which we use as a regression test for the scalar WKB pipeline before applying it to spinors.

Why not just…?

Why not just square the Dirac operator and use scalar WKB? For neutral fields this probably works nearly as well, and we expect our ablation to confirm it. With charge, squaring produces an energy-dependent effective potential and loses the distinction between particle and antiparticle modes. The fixed-$\Phi$ phase condition depends on precisely this distinction. The ablation decides the question. If the squared-operator heat capacity matches the collocation reference in the fixed-$\Phi$ ensemble to within the same tolerance as the rotor WKB, the rotor step is unnecessary and we will say so.

Why not just use standard multicomponent matrix WKB? This is the real threat, because it carries the same information. The native polar decomposition from On SVD and Polar Decomposition in Real and Complexified Clifford Algebras gives a unique amplitude-rotation split wherever the spinor is invertible, and it flags caustics through rank drops. Matrix WKB instead needs an adiabatic frame choice. Whether this matters numerically is an open question, and the head-to-head comparison in the Baselines section settles it. We do not claim an advantage in advance.

Why not just diagonalize numerically and skip WKB? Collocation gives the spectrum but no formula. The scientific question is how the heat-capacity divergence depends on $c$ and $\Phi$, and that needs an analytic handle like the quantization condition. Numerics remain the arbiter of accuracy at every step.

Why not just use horizon thermodynamics? Horizon tunneling, as in Entropy Modifications of Charged Accelerating Anti-de Sitter Black Hole, says nothing about the $c$-dependence of screen phase structure, and that dependence is what we test. In addition, Dynamic and thermodynamic stability of relativistic, perfect fluid stars shows that thermodynamic structure in general relativity does not require a horizon.

Proposed Experiments and Decisive Checks

Pipeline validation. First, we reproduce the truncated-Hilbert-transform singular-value asymptotics with our scalar WKB-plus-matching code. Second, we check that relative eigenvalue error against collocation decays at least as fast as the $O(n^{-1/2+\delta})$ rate that paper reports.

Accuracy of the spinor WKB. Using spinor spectra on a grid of $(c, Q, \Lambda)$ values, where $\Lambda$ is the cosmological constant, we report two metrics for rotor WKB, matrix WKB and the squared-operator ablation:

  • relative eigenvalue error versus $n$ and $\ell$;
  • the error in the inferred location of heat-capacity peaks.

First decisive check. We compare the $c$-values at which the field's spectral heat capacity diverges, or peaks most sharply, with the neutral condition $c < 1$ and the fixed-$\Phi$ condition from the phase-structures paper. There are two possible outcomes:

  • Agreement within WKB error for both spins would be the first microscopic evidence supporting the macroscopic screen phase diagram.
  • A systematic offset would show that matter fields shift the screen phase boundary. A boson-fermion split would point to a spin-dependent correction of the kind the tunneling paper found on horizons.

Second decisive check. We extract the entropy correction $\Delta S$ from the spectral free energy at the extremal limit and confirm that the shift in extremal mass obeys $\Delta M_{\text{ext}} \approx -T_0\,\Delta S$ from Universal Relation between Corrections to Entropy and Extremality. A sign violation would expose an error in our spectral bookkeeping before any physics is claimed.