Spectral statistics and energy-gap scaling in $k-$local spin Hamiltonians
Sasanka Dowarah
TL;DR
This work analyzes spectral statistics and energy-gap formation in all-to-all interacting $k$-local spin Hamiltonians with Gaussian couplings $J_{j_1\cdots j_k}\sim N(\mu,\sigma^2)$. For $\mu=0$ the energy-level statistics are universal and determined by the parity of the system size $L$ and locality $k$, yielding GOE, GUE, or GSE ensembles; for $\mu\neq 0$ the model maps to deformed random matrices, enabling analytic predictions for the ground-state gap and its closing at a critical disorder $\sigma_c$. In the regime $k \gg \sqrt{L}$ the gap closes at $\sigma_c \approx |\mu|$, with a universal quadratic scaling $\Delta(\sigma)/L \sim (\sigma_c - \sigma)^2$ near criticality, while for $k \ll \sqrt{L}$ the transition is smoother and $\sigma_c$ scales with $|\mu|$. The paper provides a semi-solvable framework linking random-matrix statistics and gap formation in all-to-all interacting quantum systems, and outlines extensions to more realistic connectivity and symmetry constraints.
Abstract
We investigate the spectral properties of all-to-all interacting spin Hamiltonians acting on exactly $k$ spins, whose coupling coefficients are drawn from a normal distribution with mean $μ$ and variance $σ^2$. For $μ= 0$, we demonstrate that the random matrix ensemble -- Gaussian Orthogonal Ensemble (GOE), Gaussian Unitary Ensemble (GUE), or Gaussian Symplectic Ensemble (GSE) -- is determined by the parity of system size $L$ and locality $k$, following standard time-reversal symmetry classification. For couplings with a nonzero mean, we map the Hamiltonians to deformed random matrix ensembles and analyze conditions for an energy gap between the ground state and the first excited state. For $μ< 0$, we find two distinct regimes: for $k \gg \sqrt{L}$, the gap closes at critical disorder $σ_{c} \approx |μ|$. Near this transition the energy gap $Δ$ exhibits universal quadratic scaling $Δ/L \sim (σ- σ_{c})^{2}$. When $k \ll \sqrt{L}$, $σ_{c}$ scales with $|μ|$, but lacks a sharp transition. Our work introduces a semi-solvable model that captures universal features of random-matrix statistics, and spectral gap formation, providing a foundation for systematic extensions to more general many-body systems.
