Topological Entanglement and Spectrum Diagnostics

Entanglement diagnostics interrogate a ground-state wavefunction without creating physical quasiparticles or edges. Topological entanglement entropy estimates the total quantum dimension; the entanglement spectrum can display sector and edge-CFT counting. Both are powerful consistency tests, but finite correlation length, cut geometry, state selection, and small Hilbert spaces can imitate or obscure their signatures.

Required background. Intrinsic topological order supplies long-range entanglement and quantum dimensions; Hall edges supplies the boundary counting compared with the spectrum.

Cancelling the boundary law

For a simply connected region with smooth boundary,

$$S(A)=\alpha L-\gamma+\cdots, \qquad \gamma=\ln\mathcal D.$$

Kitaev–Preskill or Levin–Wen combinations of overlapping regions cancel the nonuniversal $\alpha L$ terms. For the tripartite combination,

$$S_A+S_B+S_C-S_{AB}-S_{AC}-S_{BC}+S_{ABC}=-\gamma$$

when the regions and separations are large compared with the correlation length Kitaev and Preskill 2006. Corners add subleading terms; criticality, Goldstone modes, and finite circumference can spoil the constant. Several sizes and shapes are needed, with covariance retained in the fit.

On a cylinder, different minimally entangled ground states correspond to different anyon flux sectors and can have sector-dependent constant terms. Failing to control the selected linear combination can produce an apparent drift or wrong $\gamma$.

Entanglement Hamiltonian and counting

Schmidt decomposition across a cut gives $\rho_A=e^{-H_E}/Z$ and entanglement levels $\xi_i=-\ln\lambda_i$. For fractional Hall trial states, the low-lying levels organized by momentum and charge often reproduce the counting of the corresponding chiral edge theory Li and Haldane 2008. An “entanglement gap” separating this universal-like branch from generic high levels can support an adiabatic-continuity claim.

The numerical values of $\xi_i$ and the gap are cut dependent and not physical edge energies. Level counting must be performed in the correct conserved sectors and compared across particle cuts, orbital cuts, sizes, and aspect ratios. A few matching levels can occur accidentally.

A phase-identification protocol

For a candidate Hall state, combine:

1. a separated many-body energy manifold with flux spectral flow;
2. topology- and momentum-sector counting;
3. extrapolated entanglement entropy or minimally entangled states;
4. low-lying spectrum counting and its size stability;
5. many-body Chern response and quasihole data;
6. structure factors excluding charge, spin, or valley order.

The conclusion should not exceed the weakest controlled link. A stable counting pattern supports a candidate universality class; it does not directly demonstrate real-time braiding in a material.

Exercise

For a Laughlin $1/3$ state, what topological constant should a large-region entropy approach?

Solution

There are three Abelian sectors with $d_a=1$, so $\mathcal D=\sqrt3$ and $\gamma=\ln\sqrt3=\tfrac12\ln3$. The entropy is $S=\alpha L-\tfrac12\ln3+\cdots$; extracting it requires cancellation or extrapolation of the area term.

References

• Alexei Kitaev and John Preskill, “Topological Entanglement Entropy,” Physical Review Letters 96 (2006) 110404, doi:10.1103/PhysRevLett.96.110404.
• Hui Li and F. Duncan M. Haldane, “Entanglement Spectrum as a Generalization of Entanglement Entropy: Identification of Topological Order in Non-Abelian Fractional Quantum Hall Effect States,” Physical Review Letters 101 (2008) 010504, doi:10.1103/PhysRevLett.101.010504.