arXiv:2610.05521
Nonlinear quantum multi-spin dynamics without entanglement
Ivan P. Miranda, Erik Sjöqvist
Abstract
We introduce a distinct route for extending classical atomistic spin dynamics into the quantum regime for arbitrary spin $s$, based on a variational product-state \textit{ansatz} that includes both pure and mixed local states. By construction, intersite entanglement -- expected to remain negligible even on very short time scales in many real materials -- is excluded, while local quantum effects are retained through the nonclassical structure of $s\geq 1$ density operators. Nonlinear dynamics arise from both the mean-field interdependence of the local states and the state-dependent dissipative terms. The framework recovers classical spin dynamics for $s=\frac{1}{2}$ while allowing genuinely quantum local dynamics for higher spins. It clarifies the variational and geometric origin of the inequivalence between quantum Landau-Lifshitz and Landau-Lifshitz-Gilbert dynamics, provides a criterion for their equivalence under a common time rescaling, establishes connections with the classical theory through two qualitatively distinct limits ($s\to\infty$ and a large number of quantum spins $N$), and allows realistic $\mathcal{O}(N)$ simulations of quantum spin dynamics in materials.
AI Summary
The paper asks how to extend atomistic Landau–Lifshitz (LL) and Landau–Lifshitz–Gilbert (LLG) dynamics while retaining local quantum degrees of freedom but excluding intersite entanglement. It restricts the state to a product of arbitrary spin-$s$ density operators, each evolving on a fixed-spectrum unitary orbit. Dirac–Frenkel variation gives self-consistent mean-field precession; Rayleigh–Lagrange–d’Alembert and Onsager constructions supply distinct phenomenological damping laws, yielding local quantum LLG and LL equations. A spectral-block calculation gives an exact criterion for the two equations’ velocities to differ only by a common time factor: $1+\kappa_j^2(p_{j,a}-p_{j,b})^2$ must coincide across every Hamiltonian-active eigenvalue pair and site. Two-eigenvalue local spectra admit a single local factor; generic mixed higher-dimensional states need not. Pure states with uniform damping have factor $1+\kappa^2$. An exactly solved spin-1 anisotropy example shows a pure state developing zero dipole moment without becoming mixed. Numerical benchmarks separate nonlinearities from mean-field coupling and damping. For selected coherent preparations, initial-velocity and post-transient quantum–classical discrepancies scale as $1/s$. Lattice simulations reach $N=4913$ spins: with symmetry-enforced zero mean discrepancy and short-range correlations, spatially averaged mismatch scales approximately as $N^{-1/2}$ despite finite local mismatch. Applied fields and biased preparations instead leave a nonzero macroscopic discrepancy. The fixed-$s$ simulation cost is linear in $N$ for the local interactions considered. The framework omits all intersite correlations, preserves local spectra, and represents environmental damping phenomenologically rather than microscopically. Its large-$s$ correspondence is preparation-dependent, while large-$N$ agreement relies on cancellation and sufficiently short-ranged correlations.
Error found: The general large-spin classical-recovery statement exceeds Benchmark 2's evidence, which establishes selected diagnostics for selected coherent initial preparations rather than arbitrary local states. The stated O(N) cost requires an additional assumption that all mean fields can be evaluated in O(N), which holds for the simulated fixed-coordination lattice but not generic dense interactions. Benchmark 3 sets the classical Gilbert parameter to α=κ at polarization r=0.7, whereas matching the paper's mixed-qubit q-LLG equation requires α=κr, introducing an avoidable discrepancy into its quantum–classical comparison.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.