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:cond-mat/9802189 · Physical review. B, Condensed matter

Absence of string order in the anisotropic S=2 Heisenberg antiferromagnet

Harald N. Aschauer, Ulrich Schollwöck

Published Jul 1, 1998• 26 citations
arXivHTMLPDF

Abstract

In this paper we study an antiferromagnetic Heisenberg $S=2$ quantum spin chain at $T=0$ with both interaction and on-site anisotropy, $H={\ensuremath{\sum}}_{i}\frac{1}{2}{(S}_{i}^{+}{S}_{i+1}^{\ensuremath{-}}{+S}_{i}^{\ensuremath{-}}{S}_{i+1}^{+}{)+J}^{z}{S}_{i}^{z}{S}_{i+1}^{z}{+D(S}_{i}^{z}{)}^{2}$. While the phase diagram of this chain for $S=1$ and the isotropic point for $S=2$ are by now excellently understood, contradictory scenarios exist for the $S=2$ anisotropic phase diagram, which imply completely different mechanisms of the emergence of the classical $\stackrel{\ensuremath{\rightarrow}}{S}\ensuremath{\infty}$ limit, hence the importance of the question. One main scenario predicts the emergence of a cascade of phase transitions not seen in the $S=1$ case, based on an analysis using the Affleck-Kennedy-Lieb-Tasaki (AKLT) model. Another scenario is in favor of an almost classical phase diagram for $S=2$, from which the $S=\ensuremath{\infty}$ phase diagram evolves quite trivially; the $S=1$ case then emerges as very special with its dominant quantum effects. Numerical studies have so far not been conclusive. Using the density matrix renormalization group, we study gaps and correlation functions in the anisotropic $S=2$ chain. The question of the existence of hidden topological order in the anisotropic $S=2$ chain is emphasized, as it distinguishes between the proposed scenarios. Careful extrapolation allows us to show that the topological order is zero in the thermodynamical limit in all disordered phases, in particular in the new phase interposed between the Haldane and large-$D$ phases. The decay of the hidden order is substantially faster than that of ${S}^{+}{S}^{\ensuremath{-}}$ correlations and excellently fitted by power laws typical of massless phases. This excludes the AKLT-model-based scenario in favor of an almost classical phase diagram for the $S=2$ and thus higher spin chains.

DOI
10.1103/physrevb.58.359

Related papers

S=2 antiferromagnetic quantum spin chain

Ulrich Schollwöck, O. Golinelli, Th. Jolicœur

We have investigated Haldane's conjecture for the S=2 antiferromagnetic quantum spin chain with nearest-neighbor exchange J. Using a density matrix renormalization group algorithm for chains up to L=350 spins, we find in the thermodynamic limit a finite gap of \ensuremath{\Delta}=0.085(5)J and a finite spin-spin correlation length \ensuremath{\xi}=49(1) lattice spacings. We have confirmed the gap value by a zero-temperature quantum Monte Carlo study. We show that the ground state has a hidden topological order that is revealed in a nonlocal string correlation function which saturates to a nonzero value in the thermodynamic limit. We investigate the behavior of the spin-2 chain under an easy-plane anisotropy D(${\mathit{S}}_{\mathit{i}}^{\mathit{z}}$${)}^{2}$, D\ensuremath{\gtrsim}0, and find that the Haldane and the large-D phase are separated by an XY phase. The string correlation function vanishes only in the D\ensuremath{\rightarrow}\ensuremath{\infty} limit and does not distinguish between the Haldane phase and the perturbative large-D phase. An analysis of the transition mechanism and of the S=2 phase diagram in the presence of easy-plane and exchange anisotropy, markedly different from the S=1 phase diagram, allow us to conjecture how the classical limit is reached from increasing integer spins. \textcopyright{} 1996 The American Physical Society.

Physical review. B, Condensed matter•Aug 1, 1996•78 citations

Variational and density-matrix renormalization-group studies of the frustrated antiferromagnetic Heisenberg S=1 quantum spin chain

Alexei K. Kolezhuk, R. Roth, Ulrich Schollwöck

In this paper we study a frustrated antiferromagnetic isotropic Heisenberg S=1 quantum spin chain H=${\ensuremath{\sum}}_{\mathrm{i}}$${\mathbf{S}}_{\mathrm{i}}$${\mathbf{S}}_{\mathrm{i}+1}$+\ensuremath{\alpha}${\ensuremath{\sum}}_{\mathrm{i}}$${\mathbf{S}}_{\mathrm{i}}$${\mathbf{S}}_{\mathrm{i}+2}$, using a variational ansatz starting from valence bond states and the density-matrix renormalization group. We find both methods to give results in very good qualitative and good quantitative agreement, which clarify the phase diagram as follows: At ${\mathrm{\ensuremath{\alpha}}}_{\mathrm{D}}$=0.284(1), there is a disorder point of the first kind, marking the onset of incommensurate spin-spin correlations in the chain. At ${\mathrm{\ensuremath{\alpha}}}_{\mathrm{L}}$=0.3725(25) there is a Lifshitz point, at which the excitation spectrum is found to develop a particular doubly degenerate structure. These points are the quantum remnants of the transition from antiferromagnetic to spiral order in the classical frustrated chain. At ${\mathrm{\ensuremath{\alpha}}}_{\mathrm{T}}$=0.7444(6) there is a first-order phase transition from an Affleck-Kennedy-Lieb-Tasaki (AKLT) phase characterized by nonvanishing string order to a phase which can be understood as a next-nearest-neighbor generalization of the AKLT model. At the transition, the string order parameter shows (to numerical precision) a discontinuous jump of 0.085 to zero; the correlation length and the gap are both finite at the transition. The problem of edge states in open frustrated chains is discussed at length.

arXiv:cond-mat/9609064•Physical review. B, Condensed matter•Apr 1, 1997•51 citations

First Order Transition in the Frustrated Antiferromagnetic Heisenberg S=1 Quantum Spin Chain

Alexei K. Kolezhuk, Roland Roth, Ulrich Schollwöck

We study a frustrated Heisenberg $S\phantom{\rule{0ex}{0ex}}=\phantom{\rule{0ex}{0ex}}1$ quantum spin chain with next-nearest neighbor (NNN) coupling $\ensuremath{\alpha}$ using a variational ansatz and the density matrix renormalization group. We find as quantum remnants of the phase transition in the classical chain a disorder point at ${\ensuremath{\alpha}}_{D}\phantom{\rule{0ex}{0ex}}=\phantom{\rule{0ex}{0ex}}0.284(1)$ and a Lifshitz point at ${\ensuremath{\alpha}}_{L}\phantom{\rule{0ex}{0ex}}=\phantom{\rule{0ex}{0ex}}0.3725(25)$. Our main finding is a first-order transition from an Affleck-Kennedy-Lieb-Tasaki (AKLT) phase to a NNN generalization of the AKLT phase at ${\ensuremath{\alpha}}_{T}\phantom{\rule{0ex}{0ex}}=\phantom{\rule{0ex}{0ex}}0.7444(6)$. At the transition, string order jumps discontinuously by $\ensuremath{\approx}0.085$ to 0; correlation length and gap are finite.

Physical Review Letters•Dec 16, 1996•89 citations

Haldane Gap and Hidden Order in the S = 2 Antiferromagnetic Quantum Spin Chain

Ulrich Schollwöck, Th. Jolicœur

We have investigated Haldane's conjecture for the S = 2 antiferromagnetic quantum spin chain with nearest-neighbour exchange J . Using a density matrix renormalization group algorithm for chains up to L = 350 spins, we show that the ground state has hidden topological order: the physics of the S = 2 chain can be captured by a valence bond solid description. We also observe effective free spin-1 states at the ends of an open S = 2 chain. We find in the thermodynamic limit a gap Δ = 0.085(5) J and a spin correlation length ξ = 49(1) lattice spacings in the isotropic case. We give the phase diagram in the presence of single-ion and exchange anisotropy, showing how the Haldane phase is squeezed out when the spin S increases.

arXiv:cond-mat/9501115•Europhysics Letters (EPL)•Jun 10, 1995•130 citations

Phase diagram of the anisotropic spin-2 XXZ model: Infinite-system density matrix renormalization group study

Jonas A. Kjäll, Michael P. Zaletel, Roger S. K. Mong +2 more

We study the ground-state phase diagram of the quantum spin-2 XXZ chain in the presence of on-site anisotropy using a matrix-product state based infinite-system density matrix renormalization group (iDMRG) algorithm. One of the interests in this system is in connecting the highly quantum-mechanical spin-1 phase diagram with the classical $S=\ensuremath{\infty}$ phase diagram. Several of the recent advances within DMRG make it possible to perform a detailed analysis of the whole phase diagram. We consider different types of on-site anisotropies, which allows us to establish the validity of the following statements: (1) the spin-2 model can be tuned into a phase, which is equivalent to the ``topologically nontrivial'' spin-1 Haldane phase, and (2) the spin-2 Haldane phase at the isotropic Heisenberg point is adiabatically connected to the ``trivial'' large-$D$ phase, with a continuous change of the Hamiltonian parameters. Furthermore, we study the spin-3 XXZ chain to help explain the development of the classical phase diagram. We present details on how to use the iDMRG method to map out the phase diagram and include an extensive discussion of the numerical methods.

arXiv:1212.6255•Physical Review B•Jun 6, 2013•137 citations

Numerical study of the hidden antiferromagnetic order in the Haldane phase

Yasuhiro Hatsugai, Mahito Kohmoto

We investigate the string correlation functions proposed by den Nijs and Rommelse for S=1 Heisenberg antiferromagnets in one dimension. The Hamiltonian with D${\mathrm{\ensuremath{\Sigma}}}_{\mathit{i}}$(${\mathit{S}}_{\mathit{i}}^{\mathit{z}}$${)}^{2}$ term is diagonalized by the Lanczos method to obtain the ground state. We calculate both the usual spin-spin correlation functions and the string correlation functions not only in the z direction (quantized direction) but also in the x direction to investigate the ${\mathit{Z}}_{2}$\ifmmode\times\else\texttimes\fi{}${\mathit{Z}}_{2}$ symmetry breaking recently proposed by Kennedy and Tasaki. We find that the long-range string correlation, which is argued to exist in the Haldane disordered phase, in fact, exists at the Heisenberg point D=0 by a finite-size analysis. We can show explicitly that the string correlation in the x direction signifies the difference between the Haldane phase and the N\'eel phase, which appears for the D0, \ensuremath{\Vert}D\ensuremath{\Vert}\ensuremath{\gtrsim}1 case. In the large-D (D\ensuremath{\gtrsim}1) phase, all spin-spin correlations are of short range as expected. There is a significant enhancement in the usual and string correlations in the x direction at the boundary between the Haldane phase and the large-D phase.

Physical review. B, Condensed matter•Dec 1, 1991•60 citations
In this paper we study an antiferromagnetic Heisenberg S=2S=2S=2 quantum spin chain at T=0T=0T=0 with both interaction and on-site anisotropy, H=\ensuremath∑i12(Si+Si+1\ensuremath−+Si\ensuremath−Si+1+)+JzSizSi+1z+D(Siz)2H={\ensuremath{\sum}}_{i}\frac{1}{2}{(S}_{i}^{+}{S}_{i+1}^{\ensuremath{-}}{+S}_{i}^{\ensuremath{-}}{S}_{i+1}^{+}{)+J}^{z}{S}_{i}^{z}{S}_{i+1}^{z}{+D(S}_{i}^{z}{)}^{2}H=\ensuremath∑i​21​(Si+​Si+1\ensuremath−​+Si\ensuremath−​Si+1+​)+JzSiz​Si+1z​+D(Siz​)2. While the phase diagram of this chain for S=1S=1S=1 and the isotropic point for S=2S=2S=2 are by now excellently understood, contradictory scenarios exist for the S=2S=2S=2 anisotropic phase diagram, which imply completely different mechanisms of the emergence of the classical S\ensuremath→\ensuremath∞\stackrel{\ensuremath{\rightarrow}}{S}\ensuremath{\infty}S\ensuremath→​\ensuremath∞ limit, hence the importance of the question. One main scenario predicts the emergence of a cascade of phase transitions not seen in the S=1S=1S=1 case, based on an analysis using the Affleck-Kennedy-Lieb-Tasaki (AKLT) model. Another scenario is in favor of an almost classical phase diagram for S=2S=2S=2, from which the S=\ensuremath∞S=\ensuremath{\infty}S=\ensuremath∞ phase diagram evolves quite trivially; the S=1S=1S=1 case then emerges as very special with its dominant quantum effects. Numerical studies have so far not been conclusive. Using the density matrix renormalization group, we study gaps and correlation functions in the anisotropic S=2S=2S=2 chain. The question of the existence of hidden topological order in the anisotropic S=2S=2S=2 chain is emphasized, as it distinguishes between the proposed scenarios. Careful extrapolation allows us to show that the topological order is zero in the thermodynamical limit in all disordered phases, in particular in the new phase interposed between the Haldane and large-DDD phases. The decay of the hidden order is substantially faster than that of S+S\ensuremath−{S}^{+}{S}^{\ensuremath{-}}S+S\ensuremath− correlations and excellently fitted by power laws typical of massless phases. This excludes the AKLT-model-based scenario in favor of an almost classical phase diagram for the S=2S=2S=2 and thus higher spin chains.
We have investigated Haldane's conjecture for the S=2 antiferromagnetic quantum spin chain with nearest-neighbor exchange J. Using a density matrix renormalization group algorithm for chains up to L=350 spins, we find in the thermodynamic limit a finite gap of \ensuremath{\Delta}=0.085(5)J and a finite spin-spin correlation length \ensuremath{\xi}=49(1) lattice spacings. We have confirmed the gap value by a zero-temperature quantum Monte Carlo study. We show that the ground state has a hidden topological order that is revealed in a nonlocal string correlation function which saturates to a nonzero value in the thermodynamic limit. We investigate the behavior of the spin-2 chain under an easy-plane anisotropy D(Siz{\mathit{S}}_{\mathit{i}}^{\mathit{z}}Siz​)2{)}^{2})2, D\ensuremath{\gtrsim}0, and find that the Haldane and the large-D phase are separated by an XY phase. The string correlation function vanishes only in the D\ensuremath{\rightarrow}\ensuremath{\infty} limit and does not distinguish between the Haldane phase and the perturbative large-D phase. An analysis of the transition mechanism and of the S=2 phase diagram in the presence of easy-plane and exchange anisotropy, markedly different from the S=1 phase diagram, allow us to conjecture how the classical limit is reached from increasing integer spins. \textcopyright{} 1996 The American Physical Society.
In this paper we study a frustrated antiferromagnetic isotropic Heisenberg S=1 quantum spin chain H=${\ensuremath{\sum}}_{\mathrm{i}}$Si{\mathbf{S}}_{\mathrm{i}}Si​Si+1{\mathbf{S}}_{\mathrm{i}+1}Si+1​+\ensuremath{\alpha}${\ensuremath{\sum}}_{\mathrm{i}}$Si{\mathbf{S}}_{\mathrm{i}}Si​Si+2{\mathbf{S}}_{\mathrm{i}+2}Si+2​, using a variational ansatz starting from valence bond states and the density-matrix renormalization group. We find both methods to give results in very good qualitative and good quantitative agreement, which clarify the phase diagram as follows: At ${\mathrm{\ensuremath{\alpha}}}_{\mathrm{D}}$=0.284(1), there is a disorder point of the first kind, marking the onset of incommensurate spin-spin correlations in the chain. At ${\mathrm{\ensuremath{\alpha}}}_{\mathrm{L}}$=0.3725(25) there is a Lifshitz point, at which the excitation spectrum is found to develop a particular doubly degenerate structure. These points are the quantum remnants of the transition from antiferromagnetic to spiral order in the classical frustrated chain. At ${\mathrm{\ensuremath{\alpha}}}_{\mathrm{T}}$=0.7444(6) there is a first-order phase transition from an Affleck-Kennedy-Lieb-Tasaki (AKLT) phase characterized by nonvanishing string order to a phase which can be understood as a next-nearest-neighbor generalization of the AKLT model. At the transition, the string order parameter shows (to numerical precision) a discontinuous jump of 0.085 to zero; the correlation length and the gap are both finite at the transition. The problem of edge states in open frustrated chains is discussed at length.
We study a frustrated Heisenberg S=1S\phantom{\rule{0ex}{0ex}}=\phantom{\rule{0ex}{0ex}}1S=1 quantum spin chain with next-nearest neighbor (NNN) coupling $\ensuremath{\alpha}$ using a variational ansatz and the density matrix renormalization group. We find as quantum remnants of the phase transition in the classical chain a disorder point at ${\ensuremath{\alpha}}_{D}\phantom{\rule{0ex}{0ex}}=\phantom{\rule{0ex}{0ex}}0.284(1)$ and a Lifshitz point at ${\ensuremath{\alpha}}_{L}\phantom{\rule{0ex}{0ex}}=\phantom{\rule{0ex}{0ex}}0.3725(25)$. Our main finding is a first-order transition from an Affleck-Kennedy-Lieb-Tasaki (AKLT) phase to a NNN generalization of the AKLT phase at ${\ensuremath{\alpha}}_{T}\phantom{\rule{0ex}{0ex}}=\phantom{\rule{0ex}{0ex}}0.7444(6)$. At the transition, string order jumps discontinuously by $\ensuremath{\approx}0.085$ to 0; correlation length and gap are finite.
We have investigated Haldane's conjecture for the S = 2 antiferromagnetic quantum spin chain with nearest-neighbour exchange J . Using a density matrix renormalization group algorithm for chains up to L = 350 spins, we show that the ground state has hidden topological order: the physics of the S = 2 chain can be captured by a valence bond solid description. We also observe effective free spin-1 states at the ends of an open S = 2 chain. We find in the thermodynamic limit a gap Δ = 0.085(5) J and a spin correlation length ξ = 49(1) lattice spacings in the isotropic case. We give the phase diagram in the presence of single-ion and exchange anisotropy, showing how the Haldane phase is squeezed out when the spin S increases.
We study the ground-state phase diagram of the quantum spin-2 XXZ chain in the presence of on-site anisotropy using a matrix-product state based infinite-system density matrix renormalization group (iDMRG) algorithm. One of the interests in this system is in connecting the highly quantum-mechanical spin-1 phase diagram with the classical $S=\ensuremath{\infty}$ phase diagram. Several of the recent advances within DMRG make it possible to perform a detailed analysis of the whole phase diagram. We consider different types of on-site anisotropies, which allows us to establish the validity of the following statements: (1) the spin-2 model can be tuned into a phase, which is equivalent to the ``topologically nontrivial'' spin-1 Haldane phase, and (2) the spin-2 Haldane phase at the isotropic Heisenberg point is adiabatically connected to the ``trivial'' large-DDD phase, with a continuous change of the Hamiltonian parameters. Furthermore, we study the spin-3 XXZ chain to help explain the development of the classical phase diagram. We present details on how to use the iDMRG method to map out the phase diagram and include an extensive discussion of the numerical methods.
We investigate the string correlation functions proposed by den Nijs and Rommelse for S=1 Heisenberg antiferromagnets in one dimension. The Hamiltonian with D${\mathrm{\ensuremath{\Sigma}}}_{\mathit{i}}$(Siz{\mathit{S}}_{\mathit{i}}^{\mathit{z}}Siz​)2{)}^{2})2 term is diagonalized by the Lanczos method to obtain the ground state. We calculate both the usual spin-spin correlation functions and the string correlation functions not only in the z direction (quantized direction) but also in the x direction to investigate the Z2{\mathit{Z}}_{2}Z2​\ifmmode\times\else\texttimes\fi{}Z2{\mathit{Z}}_{2}Z2​ symmetry breaking recently proposed by Kennedy and Tasaki. We find that the long-range string correlation, which is argued to exist in the Haldane disordered phase, in fact, exists at the Heisenberg point D=0 by a finite-size analysis. We can show explicitly that the string correlation in the x direction signifies the difference between the Haldane phase and the N\'eel phase, which appears for the D0, \ensuremath{\Vert}D\ensuremath{\Vert}\ensuremath{\gtrsim}1 case. In the large-D (D\ensuremath{\gtrsim}1) phase, all spin-spin correlations are of short range as expected. There is a significant enhancement in the usual and string correlations in the x direction at the boundary between the Haldane phase and the large-D phase.