SearcharxivSearch

arXiv subjects

Sizhuo Yu

Publications and source records attributed to Sizhuo Yu.

2 recordsLinked to original sources

N-representable one-electron reduced density matrix reconstruction with frozen core electrons

Recent advances in quantum crystallography have shown that, beyond conventional charge density refinement, a one-electron reduced density matrix (1-RDM) satisfying N-representability conditions can be reconstructed using jointly experimental X-ray structure factors (XSF) and directional Compton profiles (DCP) through semi-definite programming. So far, such reconstruction methods for 1-RDM, not constrained to idempotency, had been tested only on a toy model system (CO$_2$). In this work, a new method is assessed on crystalline urea (CO(NH$_2$)$_2$) using static (0 K) and dynamic (50 K) artificial-experimental data. An improved model, including symmetry constraints and frozen-core electron contribution, is introduced to better handle the increasing system complexity. Reconstructed 1-RDMs, deformation densities and DCP anisotropy are analyzed, and it is demonstrated that the changes in the model significantly improve the reconstruction's quality against insufficient information and data corruption. The robustness of the model and the strategy are thus shown to be well-adapted to address the reconstruction problem from actual experimental scattering data.

quant-ph

Learning Effective Spin Hamiltonian of Quantum Magnet

Interacting spins in quantum magnet can cooperate and exhibit exotic states like the quantum spin liquid. To explore the materialization of such intriguing states, the determination of effective spin Hamiltonian of the quantum magnet is thus an important, while at the same time, very challenging inverse many-body problem. To efficiently learn the microscopic spin Hamiltonian from the macroscopic experimental measurements, here we propose an unbiased Hamiltonian searching approach that combines various optimization strategies, including the automatic differentiation and Bayesian optimization, etc, with the exact diagonalization and many-body thermal tensor network calculations. We showcase the accuracy and powerfulness by applying it to training thermal data generated from a given spin Hamiltonian, and then to realistic experimental data measured in the spin-chain compound Copper Nitrate and triangular-lattice materials TmMgGaO4. This automatic Hamiltonian searching constitutes a very promising approach in the studies of the intriguing spin liquid candidate magnets and correlated electron materials in general.

cond-mat.str-el