Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

arXiv:2312.08349 · SciPost Physics

Kernel polynomial method for linear spin wave theory

Harry Lane, Hao Zhang, David Dahlbom, Sam Quinn, Rolando D. Somma, Martin Mourigal, Christian D. Batista, Kipton M. Barros

Published Nov 26, 2024• 8 citations
arXivHTMLPDF

Abstract

Calculating dynamical spin correlations is essential for matching model magnetic exchange Hamiltonians to momentum-resolved spectroscopic measurements. A major numerical bottleneck is the diagonalization of the dynamical matrix, especially in systems with large magnetic unit cells, such as those with incommensurate magnetic structures or quenched disorder. In this paper, we demonstrate an efficient scheme based on the kernel polynomial method for calculating dynamical correlations of relevance to inelastic neutron scattering experiments. This method reduces the scaling of numerical cost from cubic to linear in the magnetic unit cell size.

DOI
10.21468/scipostphys.17.5.145

Related papers

Linear-scaling electronic structure theory: Electronic temperature in the Kernel Polynomial Method

Eunan J. McEniry, Ralf Drautz

Linear-scaling electronic structure methods based on the calculation of moments of the underlying electronic Hamiltonian offer a computationally efficient and numerically robust scheme to drive large-scale atomistic simulations, in which the quantum-mechanical nature of the electrons is explicitly taken into account. We compare the kernel polynomial method to the Fermi operator expansion method and establish a formal connection between the two approaches. We show that the convolution of the kernel polynomial method may be understood as an effective electron temperature. The results of a number of possible kernels are formally examined, and then applied to a representative tight-binding model.

arXiv:1701.01568•Jan 6, 2017•1 citation

The kernel polynomial method

Alexander Weiße, Gerhard Wellein, Andreas Alvermann +1 more

Efficient and stable algorithms for the calculation of spectral quantities and correlation functions are some of the key tools in computational condensed-matter physics. In this paper basic properties and recent developments of Chebyshev expansion based algorithms and the kernel polynomial method are reviewed. Characterized by a resource consumption that scales linearly with the problem dimension these methods enjoyed growing popularity over the last decade and found broad application not only in physics. Representative examples from the fields of disordered systems, strongly correlated electrons, electron-phonon interaction, and quantum spin systems are discussed in detail. In addition, an illustration on how the kernel polynomial method is successfully embedded into other numerical techniques, such as cluster perturbation theory or Monte Carlo simulation, is provided.

arXiv:cond-mat/0504627•Reviews of Modern Physics•Mar 24, 2006•1,084 citations

Efficient memory equation algorithm for reduced dynamics in spin-boson models

Andrei A. Golosov, Richard A. Friesner, Philip Pechukas

The dynamics of a one-dimensional quantum system coupled to a harmonic bath can be expressed through Feynman’s path integral expression for the reduced density matrix. In this expression the influence of the environment is seen in correlations between positions of the system that are nonlocal in time. Makri and Makarov [J. Chem. Phys. 102, 4600 (1995)] showed that for many practical problems correlations over only a few time steps, Δkmax, need to be taken into account, which led to an efficient iterative scheme. However, this algorithm scales as the size of the system to the power of 2(Δkmax+1), which restricts the size of the system that can be studied with this method. In this work we present an efficient algorithm which scales linearly with Δkmax. In our method the reduced density matrix is written as a convolution of its past values with an integral equation kernel. The calculation of that kernel is based on a perturbative expansion of the discretized quasiadiabatic path integral expression for the reduced density matrix. The expansion ignores certain types of correlations.

The Journal of Chemical Physics•Jan 1, 1999•60 citations

Dynamical correlation functions for linear spin systems

Gerhard H. Muller, H Beck

Dynamical spin correlation functions in (q, omega )-space are calculated numerically for linear quantum-mechanical Heisenberg systems containing up to ten spins with s= 1 / 2 and s=1. The authors consider ferro- and antiferromagnets including single-site and exchange anisotropies. They compare their results with various other theoretical treatments and draw conclusions for the dynamics of infinite chains. In the case of the s=1 planar Heisenberg ferromagnet direct comparison is made with inelastic neutron scattering cross sections of CsNiF 3 . The theoretical and experimental results are in good agreement.

Journal of Physics C Solid State Physics•Feb 14, 1978•17 citations

Spin-dynamics simulations of the magnetic dynamics of RbMnF3 and direct comparison with experiment

Shan-Ho Tsai, Alex Bunker, David Paul Landau

Spin-dynamics techniques have been used to perform large-scale simulations of the dynamic behavior of the classical Heisenberg antiferromagnet in simple cubic lattices with linear sizes $L<~60.$ This system is widely recognized as an appropriate model for the magnetic properties of ${\mathrm{RbMnF}}_{3}.$ Time evolutions of spin configurations were determined numerically from coupled equations of motion for individual spins using an algorithm implemented by Krech et al., which is based on fourth-order Suzuki-Trotter decompositions of exponential operators. The dynamic structure factor was calculated from the space- and time-displaced spin-spin correlation function. The crossover from hydrodynamic to critical behavior of the dispersion curve and spin-wave half-width was studied as the temperature was increased towards the critical temperature. The dynamic critical exponent was estimated to be $z=(1.43\ifmmode\pm\else\textpm\fi{}0.03),$ which is slightly lower than the dynamic scaling prediction, but in good agreement with a recent experimental value. Direct, quantitative comparisons of both the dispersion curve and the line shapes obtained from our simulations with very recent experimental results for ${\mathrm{RbMnF}}_{3}$ are presented.

arXiv:cond-mat/9910429•Physical review. B, Condensed matter•Jan 1, 2000•28 citations

Spin-dynamics study of the dynamic critical behavior of the three-dimensional classical Heisenberg ferromagnet

Kun Chen, David Paul Landau

Using spin-dynamics techniques we have performed large-scale computer simulations of the dynamic behavior of the L\ifmmode\times\else\texttimes\fi{}L\ifmmode\times\else\texttimes\fi{}L body-centered-cubic classical Heisenberg ferromagnet with L\ensuremath{\le}40 in the vicinity of the critical point ${\mathit{T}}_{\mathit{c}}$. The temporal evolutions of spin configurations were determined numerically from coupled equations of motion for individual spins by a fourth-order predictor-corrector method with initial spin configurations generated by Monte Carlo methods. The space- and time-displaced spin-spin correlation functions and their space-time Fourier transforms were calculated to determine the neutron-scattering functions. We developed a dynamic finite-size scaling theory for the neutron-scattering function at ${\mathit{T}}_{\mathit{c}}$ and used it to extract the dynamic critical exponent z. Within our resolution limit, the value of z was estimated to be 2.478(28), in excellent agreement with the dynamic scaling prediction, and dynamic scaling was found to be valid for momentum transfer q up to 0.4\ensuremath{\pi} in the (1,0,0) direction.

Physical review. B, Condensed matter•Feb 1, 1994•88 citations
Calculating dynamical spin correlations is essential for matching model magnetic exchange Hamiltonians to momentum-resolved spectroscopic measurements. A major numerical bottleneck is the diagonalization of the dynamical matrix, especially in systems with large magnetic unit cells, such as those with incommensurate magnetic structures or quenched disorder. In this paper, we demonstrate an efficient scheme based on the kernel polynomial method for calculating dynamical correlations of relevance to inelastic neutron scattering experiments. This method reduces the scaling of numerical cost from cubic to linear in the magnetic unit cell size.
Linear-scaling electronic structure methods based on the calculation of moments of the underlying electronic Hamiltonian offer a computationally efficient and numerically robust scheme to drive large-scale atomistic simulations, in which the quantum-mechanical nature of the electrons is explicitly taken into account. We compare the kernel polynomial method to the Fermi operator expansion method and establish a formal connection between the two approaches. We show that the convolution of the kernel polynomial method may be understood as an effective electron temperature. The results of a number of possible kernels are formally examined, and then applied to a representative tight-binding model.
Efficient and stable algorithms for the calculation of spectral quantities and correlation functions are some of the key tools in computational condensed-matter physics. In this paper basic properties and recent developments of Chebyshev expansion based algorithms and the kernel polynomial method are reviewed. Characterized by a resource consumption that scales linearly with the problem dimension these methods enjoyed growing popularity over the last decade and found broad application not only in physics. Representative examples from the fields of disordered systems, strongly correlated electrons, electron-phonon interaction, and quantum spin systems are discussed in detail. In addition, an illustration on how the kernel polynomial method is successfully embedded into other numerical techniques, such as cluster perturbation theory or Monte Carlo simulation, is provided.
The dynamics of a one-dimensional quantum system coupled to a harmonic bath can be expressed through Feynman’s path integral expression for the reduced density matrix. In this expression the influence of the environment is seen in correlations between positions of the system that are nonlocal in time. Makri and Makarov [J. Chem. Phys. 102, 4600 (1995)] showed that for many practical problems correlations over only a few time steps, Δkmax, need to be taken into account, which led to an efficient iterative scheme. However, this algorithm scales as the size of the system to the power of 2(Δkmax+1), which restricts the size of the system that can be studied with this method. In this work we present an efficient algorithm which scales linearly with Δkmax. In our method the reduced density matrix is written as a convolution of its past values with an integral equation kernel. The calculation of that kernel is based on a perturbative expansion of the discretized quasiadiabatic path integral expression for the reduced density matrix. The expansion ignores certain types of correlations.
Dynamical spin correlation functions in (q, omega )-space are calculated numerically for linear quantum-mechanical Heisenberg systems containing up to ten spins with s= 1 / 2 and s=1. The authors consider ferro- and antiferromagnets including single-site and exchange anisotropies. They compare their results with various other theoretical treatments and draw conclusions for the dynamics of infinite chains. In the case of the s=1 planar Heisenberg ferromagnet direct comparison is made with inelastic neutron scattering cross sections of CsNiF 3 . The theoretical and experimental results are in good agreement.
Spin-dynamics techniques have been used to perform large-scale simulations of the dynamic behavior of the classical Heisenberg antiferromagnet in simple cubic lattices with linear sizes L< 60.L<~60.L< 60. This system is widely recognized as an appropriate model for the magnetic properties of RbMnF3.{\mathrm{RbMnF}}_{3}.RbMnF3​. Time evolutions of spin configurations were determined numerically from coupled equations of motion for individual spins using an algorithm implemented by Krech et al., which is based on fourth-order Suzuki-Trotter decompositions of exponential operators. The dynamic structure factor was calculated from the space- and time-displaced spin-spin correlation function. The crossover from hydrodynamic to critical behavior of the dispersion curve and spin-wave half-width was studied as the temperature was increased towards the critical temperature. The dynamic critical exponent was estimated to be $z=(1.43\ifmmode\pm\else\textpm\fi{}0.03),$ which is slightly lower than the dynamic scaling prediction, but in good agreement with a recent experimental value. Direct, quantitative comparisons of both the dispersion curve and the line shapes obtained from our simulations with very recent experimental results for RbMnF3{\mathrm{RbMnF}}_{3}RbMnF3​ are presented.
Using spin-dynamics techniques we have performed large-scale computer simulations of the dynamic behavior of the L\ifmmode\times\else\texttimes\fi{}L\ifmmode\times\else\texttimes\fi{}L body-centered-cubic classical Heisenberg ferromagnet with L\ensuremath{\le}40 in the vicinity of the critical point Tc{\mathit{T}}_{\mathit{c}}Tc​. The temporal evolutions of spin configurations were determined numerically from coupled equations of motion for individual spins by a fourth-order predictor-corrector method with initial spin configurations generated by Monte Carlo methods. The space- and time-displaced spin-spin correlation functions and their space-time Fourier transforms were calculated to determine the neutron-scattering functions. We developed a dynamic finite-size scaling theory for the neutron-scattering function at Tc{\mathit{T}}_{\mathit{c}}Tc​ and used it to extract the dynamic critical exponent z. Within our resolution limit, the value of z was estimated to be 2.478(28), in excellent agreement with the dynamic scaling prediction, and dynamic scaling was found to be valid for momentum transfer q up to 0.4\ensuremath{\pi} in the (1,0,0) direction.