arXiv:2610.16915
Mutual information across the commensurate--incommensurate crossover and near the Kosterlitz--Thouless point of the spin-$\tfrac12$ $J_1$--$J_2$ Heisenberg chain
Wuttichai Pankeaw, Pruet Kalasuwan
Abstract
What quantum mutual information adds to conventional correlation functions is often unclear. We study the spin-1/2 $J_1$-$J_2$ Heisenberg chain, using DMRG on infinite chains and periodic rings and an SU(2) decomposition of the mutual information into correlation channels. For singlet ground states the two-site mutual information is an exact function of the spin correlator, so any additional content must come from blocks. In the dimerized phase, blocks resolve bond-energy correlations in the SU(2)-scalar channel without an independent asymptotic length: singlet and triplet correlation lengths agree to about 1%. The scalar channel dominates block mutual information at large distances in the commensurate regime up to $J_2/J_1=0.40$, but falls to a few percent beyond the disorder point, which coincides numerically with the Majumdar-Ghosh point $J_2/J_1=1/2$; the structure-factor peak splits only at the Lifshitz point $J_2/J_1=0.520622(2)$. On the gapless side we compare block mutual information on rings extrapolated to infinite size with the two-interval prediction of SU(2)$_1$ conformal field theory across the Kosterlitz-Thouless point $J_2/J_1=0.241167$. There, for block lengths $l\le6$, the correction decays as $l^{-p}$ with $p=1.96$-$2.07$, and any logarithmic term is at least 60 times smaller than in the Heisenberg chain. Expanding in the marginal coupling (from level spectroscopy) the first-order coefficient is consistent with $l^{-2}$ scaling and the second-order coefficient is positive for $x\le1/4$, suggesting that the mutual information relative to the field-theory value becomes stationary at the KT point for large blocks. The scalar-channel share approaches 1/4 at small cross ratio $x$, as expected from SO(4) symmetry; under two stated assumptions its first-order response to the coupling is $(3/8)\ln x-5/8$ at small $x$, matched by the $l=4$ data within 1-3% at $x\le1/9$.
AI Summary
The paper asks what block quantum mutual information reveals beyond spin correlations in the spin-$\tfrac12$ $J_1$–$J_2$ Heisenberg chain, across the commensurate–incommensurate crossover and near the Kosterlitz–Thouless (KT) transition. It combines infinite-chain DMRG, transfer-matrix spectra, periodic-ring DMRG and exact-diagonalization level spectroscopy with an SU(2)-channel decomposition based on the quadratic relative-entropy expansion. For singlet states, two-site mutual information is exactly determined by the spin correlator. Blocks additionally resolve predominantly bond-energy correlations: the scalar channel dominates at the largest distances studied for $0.32\le J_2/J_1\le0.40$, but becomes small beyond the disorder point near $1/2$. Singlet and triplet correlation lengths agree to approximately 1%, indicating no additional asymptotic length. Transfer-matrix analysis places the incommensurate onset within $5\times10^{-4}$ of the Majumdar–Ghosh point; structure-factor curvature gives the distinct Lifshitz point $J_2/J_1=0.520622(2)$. At the KT point $J_2/J_1=0.241167$, ring-extrapolated block mutual information approaches the SU(2)$_1$ CFT prediction with fitted corrections $l^{-p}$, $p=1.96$–$2.07$, and a fitted logarithmic amplitude at least 60 times smaller than in the Heisenberg chain. The scalar share approaches $1/4$ at small cross ratio. An independently calibrated marginal-coupling expansion finds a linear coefficient consistent with $l^{-2}$ decay and positive quadratic coefficients for $x\le1/4$, suggesting large-block stationarity at KT. Limitations include blocks restricted to $l\le6$, unresolved large-block extrapolations, correlated error bounds potentially underestimated by up to a factor 1.5, and weak-correlation restrictions on channel shares. Small-cross-ratio response formulas additionally require two unproven replica/channel-identification assumptions.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.
- Comments
- 57 pages, 19 figures, 16 tables