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

Physical review. B, Condensed matter

Finite-size scaling in the ground state of spin-1/2antiferromagneticXXZrings

T. A. Kaplan, Peter Horsch, J. Borysowicz

Published Feb 1, 1987• 19 citations
Publisher

Abstract

Convincing evidence that the correlation function \ensuremath{\omega}(l,N)=4〈${S}_{i}^{x}$${S}_{i+l}^{x}$〉 obeys finite-size scaling for the XY model is obtained from large rings (number of spins N up to \ensuremath{\approxeq}1000). \ensuremath{\omega}(l,N) for the Heisenberg model behaves much like the XY model for the small N's available (up to 18), suggesting finite-size scaling in the Heisenberg model. The scaling function, f(l/N)\ensuremath{\approxeq}\ensuremath{\omega}(l,N)/\ensuremath{\omega}(l,\ensuremath{\infty}), seems to show the remarkable property ${f}_{\mathrm{Heis}(\mathrm{y})=[{f}_{\mathrm{XY}}(\mathrm{y}{)]}^{\ensuremath{\alpha}}}$, with \ensuremath{\alpha}\ensuremath{\approxeq}1.75. Also, \ensuremath{\alpha} appears to vary smoothly with the exchange-parameter ratio \ensuremath{\gamma}=${J}_{\mathrm{xx}}$/${J}_{\mathrm{zz}}$. The scaling assumption is used to estimate \ensuremath{\omega}(l,\ensuremath{\infty}).

DOI
10.1103/physrevb.35.1877

Related papers

Ground state of the one-dimensional antiferromagnetic Heisenberg model

J. Borysowicz, T. A. Kaplan, Peter Horsch

We have calculated the two-point correlation functions ${\ensuremath{\omega}}_{\mathrm{il}}$(N)=(4/3)〈S${\ensuremath{\rightarrow}}_{i}$\ensuremath{\cdot}S${\ensuremath{\rightarrow}}_{i+l}$ 〉 and their averages over i,${\ensuremath{\omega}}_{l}$ (N), in the ground state of the one-dimensional antiferromagnetic Heisenberg model for N=4,6,8,...,16 spins. Both periodic (rings) and free-end (chains) boundary conditions are considered. Surprisingly tight lower and upper bounds have been obtained for ${\ensuremath{\omega}}_{l}$(\ensuremath{\infty}) under reasonable assumptions. In addition to showing the rather strong even-l--odd-l alternation in \ensuremath{\Vert}${\ensuremath{\omega}}_{l}$(N)\ensuremath{\Vert}, known from earlier results of Bonner and Fisher for rings with N up to 10, our bounds indicate a smooth behavior in l\ensuremath{\Vert}${\ensuremath{\omega}}_{l}$(\ensuremath{\infty})\ensuremath{\Vert} for l odd and l even, with, surprisingly, a broad maximum attained within the odd-l values. The bounds obtained from the chain results were essential to seeing this maximum (because of the larger l values available for given N). The quantity l\ensuremath{\Vert}${\ensuremath{\omega}}_{l}$(N)\ensuremath{\Vert} for chains with fixed N also shows such a maximum, and in addition shows a similar maximum for even l's. If the indicated trends for large l and N continue in ${\ensuremath{\omega}}_{l}$(\ensuremath{\infty}) and in ${S}_{N}$, the structure factor at wave vector \ensuremath{\pi}, then finite-size contributions to ${\ensuremath{\omega}}_{l}$(N) will have to contribute to the (seemingly) logarithmic divergence of ${S}_{N}$ as N\ensuremath{\rightarrow}\ensuremath{\infty}. We are not aware of any models where a similarly weak divergence shows such a finite-size contribution.

Physical review. B, Condensed matter•Feb 1, 1985•40 citations

Ground state of the 1-dimensional antiferromagnetic Heisenberg model (abstract)

J. Borysowicz, T. A. Kaplan, Peter Horsch

We have calculated the 2-point correlation functions ωil(N)=(4/3)〈Si ⋅ Si+l〉 and their averages over i, ωl(N), in the ground state of the 1-dimensional antiferromagnetic Heisenberg model for N=4(2)16 spins. Both periodic (rings) and free-ends (chains) boundary conditions are considered. Surprisingly tight lower and upper bounds have been obtained for ωl(∞) under reasonable assumptions. In addition to showing the rather strong even-l-odd-l alternation in ‖ωl(N)‖, already known, our bounds indicate a smooth behavior in l‖ωl(∞)‖ for l odd and for l even, with, surprisingly, a broad maximum attained within the odd-l values at l≂6 to 8. The bounds obtained from the chain results were essential to seeing this maximum (because the l-values available for given N are larger than for rings). The quantity l‖ωl(N)‖ for chains with fixed N also shows such a maximum, and in addition shows a similar maximum for even l’s. If the trends for large l and N which we have found continue in ωl(∞) and in SN, the structure factor at wave vector π, then finite-size contributions to ωl(N) will have to contribute to the (seemingly) logarithmic divergence of SN as N→∞. We are not aware of any models where a similarly weak divergence shows such a finite-size contribution. Earlier results, including ‖ωl(∞)‖≊A/l for l→∞, gave no hint of the decrease in l‖ωl(∞)‖ described above. Recently ‘‘logarithmic corrections’’ have been mentioned, wherein ‖ωl(∞)‖≊(A/l) (ln l+B)−1. This might be related to the decrease in l‖ωl(∞)‖ which we found. If finite-size corrections are neglected, then this 1/l ln l behavior would predict SN≊ln ln N for N→∞, as opposed to the trend SN∼ln N that we found; continuation of the latter trend would then have to be due entirely to the finite-size contributions. Some insight into various surprising aspects of ωil(N) is gained by considering a single-band model of noninteracting electrons, which is a special case of the Hubbard model, as is the Heisenberg model.

Journal of Applied Physics•Apr 15, 1985•0 citations

The Phase Transition of the Spin-1/2 Heisenberg Model with a Spatially Staggered Anisotropy on the Square Lattice

Fu-Jiun Jiang

Puzzled by the indication of a new critical theory for the spin-1/2 Heisenberg model with a spatially staggered anisotropy on the square lattice as suggested in \cite{Wenzel08}, we re-investigate the phase transition of this model induced by dimerization. We focus on studying the finite-size scaling of the observables $ρ_{s1} L$ and $ρ_{s2} L$, where $L$ stands for the spatial box sizes used in the simulations and $ρ_{si}$ with $i \in \{1,2\}$ is the spin-stiffness in $i$-direction. We find by performing finite-size scaling using the observable $ρ_{s2} L$, which corresponds to the spatial direction with a fixed antiferromagnetic coupling, one would suffer a much less severe correction compared to that of using $ρ_{s1} L$. Therefore $ρ_{s2} L$ is a better quantity than $ρ_{s1} L$ for finite-size scaling analysis concerning the limitation for the availability of large volumes data in our study. Remarkably, by employing the method of fixing the aspect-ratio of spatial winding numbers squared in the simulations, even from $ρ_{s1} L$ which receives the most serious correction among the observables considered in this study, we arrive at a value for the critical exponent $ν$ which is consistent with the expected O(3) value by using only up to $L = 64$ data points.

arXiv:0911.0653•Nov 3, 2009•3 citations

Finite-size scaling of the ground-state parameters of the two-dimensional Heisenberg model

Anders W. Sandvik

The ground-state parameters of the two-dimensional $S=1/2$ antiferromagnetic Heisenberg model are calculated using the stochastic series expansion quantum Monte Carlo method for $L\ifmmode\times\else\texttimes\fi{}L$ lattices with $L$ up to $16$. The finite-size results for the energy $E$, the sublattice magnetization $M$, the long-wavelength susceptibility ${\ensuremath{\chi}}_{\ensuremath{\perp}}(q=2\ensuremath{\pi}/L)$, and the spin stiffness ${\ensuremath{\rho}}_{s},$ are extrapolated to the thermodynamic limit using fits to polynomials in $1/L$, constrained by scaling forms previously obtained from renormalization-group calculations for the nonlinear $\ensuremath{\sigma}$ model and chiral perturbation theory. The results are fully consistent with the predicted leading finite-size corrections, and are of sufficient accuracy for extracting also subleading terms. The subleading energy correction $(\ensuremath{\sim}{1/L}^{4})$ agrees with chiral perturbation theory to within a statistical error of a few percent, thus providing numerical confirmation of the finite-size scaling forms to this order. The extrapolated ground- state energy per spin is $E=\ensuremath{-}0.669437(5)$. The result from previous Green's function Monte Carlo (GFMC) calculations is slightly higher than this value, most likely due to a small systematic error originating from ``population control'' bias in GFMC. The other extrapolated parameters are $M=0.3070(3)$, ${\ensuremath{\rho}}_{s}=0.175(2)$, ${\ensuremath{\chi}}_{\ensuremath{\perp}}=0.0625(9)$, and the spin-wave velocity $c=1.673(7)$. The statistical errors are comparable with those of previous estimates obtained by fitting loop algorithm quantum Monte Carlo data to finite-temperature scaling forms. Both $M$ and ${\ensuremath{\rho}}_{s}$ obtained from the finite-$T$ data are, however, a few error bars higher than the present estimates. It is argued that the $T=0$ extrapolations performed here are less sensitive to effects of neglected higher-order corrections, and therefore should be more reliable.

arXiv:cond-mat/9707123•Physical review. B, Condensed matter•Nov 1, 1997•503 citations

Numerical renormalization-group study of the correlation functions of the antiferromagnetic spin-1/2 Heisenberg chain

K. Hallberg, Peter Horsch, Gerardo Martínez

We use the density-matrix renormalization-group technique developed by White to calculate the spin correlation functions 〈${\mathit{S}}_{\mathit{n}+\mathit{l}}^{\mathit{z}}$ ${\mathit{S}}_{\mathit{n}}^{\mathit{z}}$〉=(-1${)}^{\mathit{l}}$\ensuremath{\omega}(l,N) for isotropic Heisenberg rings up to N=70 sites. The correlation functions for large l and N are found to obey the scaling relation \ensuremath{\omega}(l,N)=\ensuremath{\omega}(l,\ensuremath{\infty})${\mathit{f}}_{\mathit{XY}}^{\mathrm{\ensuremath{\alpha}}}$(l/N) proposed by Kaplan et al., which is used to determine \ensuremath{\omega}(l,\ensuremath{\infty}). The asymptotic correlation function \ensuremath{\omega}(l,\ensuremath{\infty}) and the magnetic structure factor S(q=\ensuremath{\pi}) show logarithmic corrections consistent with \ensuremath{\omega}(l,\ensuremath{\infty})\ensuremath{\sim}a\ensuremath{\surd}lncl /l, where c is related to the cut-off dependent coupling constant ${\mathit{g}}_{\mathrm{eff}}$(${\mathit{l}}_{0}$)=1/ln(${\mathit{cl}}_{0}$), as predicted by field theoretical treatments.

arXiv:cond-mat/9505132•Physical review. B, Condensed matter•Jul 1, 1995•66 citations

Ground-state properties of Heisenberg spin chains

Adriana Moreo

The ground-state properties of some finite antiferromagnetic Heisenberg rings with N spins at the isotropic point have been investigated using a modified Lanczos method. Rings with spin 1 and (3/2 have been studied and new results are presented for the correlation functions and the structure factor up to N=16 (12) for S=1 ((3/2). Evidence in support of Haldane's conjecture is found. The correlation functions for S=(3/2 behave as predicted by Affleck.

Physical review. B, Condensed matter•Jun 1, 1987•64 citations
Convincing evidence that the correlation function \ensuremath{\omega}(l,N)=4〈{S}_{i}^{x}$${S}_{i+l}^{x}〉 obeys finite-size scaling for the XY model is obtained from large rings (number of spins N up to \ensuremath{\approxeq}1000). \ensuremath{\omega}(l,N) for the Heisenberg model behaves much like the XY model for the small N's available (up to 18), suggesting finite-size scaling in the Heisenberg model. The scaling function, f(l/N)\ensuremath{\approxeq}\ensuremath{\omega}(l,N)/\ensuremath{\omega}(l,\ensuremath{\infty}), seems to show the remarkable property fHeis(y)=[fXY(y)]\ensuremathα{f}_{\mathrm{Heis}(\mathrm{y})=[{f}_{\mathrm{XY}}(\mathrm{y}{)]}^{\ensuremath{\alpha}}}fHeis(y)=[fXY​(y)]\ensuremathα​, with \ensuremath{\alpha}\ensuremath{\approxeq}1.75. Also, \ensuremath{\alpha} appears to vary smoothly with the exchange-parameter ratio \ensuremath{\gamma}=Jxx{J}_{\mathrm{xx}}Jxx​/Jzz{J}_{\mathrm{zz}}Jzz​. The scaling assumption is used to estimate \ensuremath{\omega}(l,\ensuremath{\infty}).
We have calculated the two-point correlation functions ${\ensuremath{\omega}}_{\mathrm{il}}$(N)=(4/3)〈S${\ensuremath{\rightarrow}}_{i}$\ensuremath{\cdot}S${\ensuremath{\rightarrow}}_{i+l}$ 〉 and their averages over i,${\ensuremath{\omega}}_{l}$ (N), in the ground state of the one-dimensional antiferromagnetic Heisenberg model for N=4,6,8,...,16 spins. Both periodic (rings) and free-end (chains) boundary conditions are considered. Surprisingly tight lower and upper bounds have been obtained for ${\ensuremath{\omega}}_{l}$(\ensuremath{\infty}) under reasonable assumptions. In addition to showing the rather strong even-l--odd-l alternation in \ensuremath{\Vert}${\ensuremath{\omega}}_{l}$(N)\ensuremath{\Vert}, known from earlier results of Bonner and Fisher for rings with N up to 10, our bounds indicate a smooth behavior in l\ensuremath{\Vert}${\ensuremath{\omega}}_{l}$(\ensuremath{\infty})\ensuremath{\Vert} for l odd and l even, with, surprisingly, a broad maximum attained within the odd-l values. The bounds obtained from the chain results were essential to seeing this maximum (because of the larger l values available for given N). The quantity l\ensuremath{\Vert}${\ensuremath{\omega}}_{l}$(N)\ensuremath{\Vert} for chains with fixed N also shows such a maximum, and in addition shows a similar maximum for even l's. If the indicated trends for large l and N continue in ${\ensuremath{\omega}}_{l}$(\ensuremath{\infty}) and in SN{S}_{N}SN​, the structure factor at wave vector \ensuremath{\pi}, then finite-size contributions to ${\ensuremath{\omega}}_{l}$(N) will have to contribute to the (seemingly) logarithmic divergence of SN{S}_{N}SN​ as N\ensuremath{\rightarrow}\ensuremath{\infty}. We are not aware of any models where a similarly weak divergence shows such a finite-size contribution.
We have calculated the 2-point correlation functions ωil(N)=(4/3)〈Si ⋅ Si+l〉 and their averages over i, ωl(N), in the ground state of the 1-dimensional antiferromagnetic Heisenberg model for N=4(2)16 spins. Both periodic (rings) and free-ends (chains) boundary conditions are considered. Surprisingly tight lower and upper bounds have been obtained for ωl(∞) under reasonable assumptions. In addition to showing the rather strong even-l-odd-l alternation in ‖ωl(N)‖, already known, our bounds indicate a smooth behavior in l‖ωl(∞)‖ for l odd and for l even, with, surprisingly, a broad maximum attained within the odd-l values at l≂6 to 8. The bounds obtained from the chain results were essential to seeing this maximum (because the l-values available for given N are larger than for rings). The quantity l‖ωl(N)‖ for chains with fixed N also shows such a maximum, and in addition shows a similar maximum for even l’s. If the trends for large l and N which we have found continue in ωl(∞) and in SN, the structure factor at wave vector π, then finite-size contributions to ωl(N) will have to contribute to the (seemingly) logarithmic divergence of SN as N→∞. We are not aware of any models where a similarly weak divergence shows such a finite-size contribution. Earlier results, including ‖ωl(∞)‖≊A/l for l→∞, gave no hint of the decrease in l‖ωl(∞)‖ described above. Recently ‘‘logarithmic corrections’’ have been mentioned, wherein ‖ωl(∞)‖≊(A/l) (ln l+B)−1. This might be related to the decrease in l‖ωl(∞)‖ which we found. If finite-size corrections are neglected, then this 1/l ln l behavior would predict SN≊ln ln N for N→∞, as opposed to the trend SN∼ln N that we found; continuation of the latter trend would then have to be due entirely to the finite-size contributions. Some insight into various surprising aspects of ωil(N) is gained by considering a single-band model of noninteracting electrons, which is a special case of the Hubbard model, as is the Heisenberg model.
Puzzled by the indication of a new critical theory for the spin-1/2 Heisenberg model with a spatially staggered anisotropy on the square lattice as suggested in \cite{Wenzel08}, we re-investigate the phase transition of this model induced by dimerization. We focus on studying the finite-size scaling of the observables ρs1Lρ_{s1} Lρs1​L and ρs2Lρ_{s2} Lρs2​L, where LLL stands for the spatial box sizes used in the simulations and ρsiρ_{si}ρsi​ with i∈{1,2}i \in \{1,2\}i∈{1,2} is the spin-stiffness in iii-direction. We find by performing finite-size scaling using the observable ρs2Lρ_{s2} Lρs2​L, which corresponds to the spatial direction with a fixed antiferromagnetic coupling, one would suffer a much less severe correction compared to that of using ρs1Lρ_{s1} Lρs1​L. Therefore ρs2Lρ_{s2} Lρs2​L is a better quantity than ρs1Lρ_{s1} Lρs1​L for finite-size scaling analysis concerning the limitation for the availability of large volumes data in our study. Remarkably, by employing the method of fixing the aspect-ratio of spatial winding numbers squared in the simulations, even from ρs1Lρ_{s1} Lρs1​L which receives the most serious correction among the observables considered in this study, we arrive at a value for the critical exponent ννν which is consistent with the expected O(3) value by using only up to L=64L = 64L=64 data points.
The ground-state parameters of the two-dimensional S=1/2S=1/2S=1/2 antiferromagnetic Heisenberg model are calculated using the stochastic series expansion quantum Monte Carlo method for $L\ifmmode\times\else\texttimes\fi{}L$ lattices with LLL up to 161616. The finite-size results for the energy EEE, the sublattice magnetization MMM, the long-wavelength susceptibility ${\ensuremath{\chi}}_{\ensuremath{\perp}}(q=2\ensuremath{\pi}/L)$, and the spin stiffness ${\ensuremath{\rho}}_{s},$ are extrapolated to the thermodynamic limit using fits to polynomials in 1/L1/L1/L, constrained by scaling forms previously obtained from renormalization-group calculations for the nonlinear $\ensuremath{\sigma}$ model and chiral perturbation theory. The results are fully consistent with the predicted leading finite-size corrections, and are of sufficient accuracy for extracting also subleading terms. The subleading energy correction $(\ensuremath{\sim}{1/L}^{4})$ agrees with chiral perturbation theory to within a statistical error of a few percent, thus providing numerical confirmation of the finite-size scaling forms to this order. The extrapolated ground- state energy per spin is $E=\ensuremath{-}0.669437(5)$. The result from previous Green's function Monte Carlo (GFMC) calculations is slightly higher than this value, most likely due to a small systematic error originating from ``population control'' bias in GFMC. The other extrapolated parameters are M=0.3070(3)M=0.3070(3)M=0.3070(3), ${\ensuremath{\rho}}_{s}=0.175(2)$, ${\ensuremath{\chi}}_{\ensuremath{\perp}}=0.0625(9)$, and the spin-wave velocity c=1.673(7)c=1.673(7)c=1.673(7). The statistical errors are comparable with those of previous estimates obtained by fitting loop algorithm quantum Monte Carlo data to finite-temperature scaling forms. Both MMM and ${\ensuremath{\rho}}_{s}$ obtained from the finite-TTT data are, however, a few error bars higher than the present estimates. It is argued that the T=0T=0T=0 extrapolations performed here are less sensitive to effects of neglected higher-order corrections, and therefore should be more reliable.
We use the density-matrix renormalization-group technique developed by White to calculate the spin correlation functions 〈Sn+lz{\mathit{S}}_{\mathit{n}+\mathit{l}}^{\mathit{z}}Sn+lz​ Snz{\mathit{S}}_{\mathit{n}}^{\mathit{z}}Snz​〉=(-1)l{)}^{\mathit{l}})l\ensuremath{\omega}(l,N) for isotropic Heisenberg rings up to N=70 sites. The correlation functions for large l and N are found to obey the scaling relation \ensuremath{\omega}(l,N)=\ensuremath{\omega}(l,\ensuremath{\infty})${\mathit{f}}_{\mathit{XY}}^{\mathrm{\ensuremath{\alpha}}}$(l/N) proposed by Kaplan et al., which is used to determine \ensuremath{\omega}(l,\ensuremath{\infty}). The asymptotic correlation function \ensuremath{\omega}(l,\ensuremath{\infty}) and the magnetic structure factor S(q=\ensuremath{\pi}) show logarithmic corrections consistent with \ensuremath{\omega}(l,\ensuremath{\infty})\ensuremath{\sim}a\ensuremath{\surd}lncl /l, where c is related to the cut-off dependent coupling constant geff{\mathit{g}}_{\mathrm{eff}}geff​(l0{\mathit{l}}_{0}l0​)=1/ln(cl0{\mathit{cl}}_{0}cl0​), as predicted by field theoretical treatments.