Spin-basis wavefunctions for the one-dimensional Kitaev model

Alwyn Jose Raja, Rajesh Narayanan, R. Ganesh · 2025

Read Spin-basis wavefunctions for the one-dimensional Kitaev model by Alwyn Jose Raja, Rajesh Narayanan, R. Ganesh (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2509-21771.txt.

Introduction

Alwyn Jose Raja, Rajesh Narayanan, and R. Ganesh write down ground-state wavefunctions for a chain of quantum magnets in this 2025 paper. The chain follows the one-dimensional Kitaev model, a standard test bed for entangled magnetism. Most known states for such models are written with helper particles called partons. States written directly in the spin basis are rare, and this paper supplies them.

The authors build their states from bonds between spins, classed as singlet-like or triplet-like. They check the states with a step-by-step method that starts from a lopsided limit, and the key features survive every step. For half-integer spins, many ground states appear together, under a rule that the triplet count stays even. For integer spins, one ground state remains, made only of triplets. Computer checks against exact results agree well.

Catalog record

CREATORAlwyn Jose Raja, Rajesh Narayanan, R. Ganesh
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE60 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2509-21771

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Spin-basis wavefunctions for the one-dimensional Kitaev model Preprint: APS/123-QED Alwyn Jose Raja, Rajesh Narayanan, R. Ganesh Abstract Magnetic phases with quantum entanglement are often expressed in terms of parton wavefunctions. Relatively few examples are known where wavefunctions can be directly written down in the spin basis. In this article, we consider the spin- S S Kitaev model in one dimension. For S = 1 / 2 S=1/2 , its eigenstates can be written using a Jordan-Wigner fermionic representation. Here, we present ground state wavefunctions for any S S directly in the spin basis. The states we propose are valence bond arrangements, with bonds having singlet or triplet character for S = 1 / 2 S=1/2 . For S > 1 / 2 S>1/2 , we use bond-states that serve as analogues of singlets and triplets. We establish the validity of our wavefunctions using a perturbative approach starting from an anisotropic limit, with key features surviving to all orders in perturbation theory. For half-integer S S and periodic boundaries, we have exponential ground state degeneracy. The ground states have topological character, with an even number of ‘triplets’ superposed on a background of ‘singlets’. For integer S S , a unique ground state emerges, composed purely of ‘triplets’. Our spin-basis wavefunctions, while not exact, capture the dominant weight of the ground state(s). We obtain good agreement against exact diagonalization wavefunctions and Jordan-Wigner spectra. I Introduction Quantum magnets are a fertile playground for entanglement and topology Fazekas 1999 ; Auerbach 2012 ; Kitaev 2006 . To explore these ideas, wavefunctions are typically written down by invoking fractionalization. The resonating valence bond (RVB) phase of Anderson and coworkers is a case in point Anderson et al. 1987 ; Liang et al. 1988 ; Zhou et al. 2017 . It proposes a wavefunction in the spin basis, consisting of a superposition of singlet covers. In practice, the RVB wavefunction is written in terms of fractionalized bosonic or fermionic operators Lee et al. 2006 . A second cornerstone is the Affleck, Kennedy, Lieb and Tasaki (AKLT) state Affleck et al. 1988 , expressed in terms of fractionalized spin- 1/2 degrees of freedom. A third example is the 1D XY model, solved using Jordan Wigner fermionization Lieb et al. 1961 . The ground state is expressed as that of a sea of non-interacting fermions. In all three examples, the wavefunction, even if it were known exactly, cannot be easily expressed in terms of the original spin operators. In contrast, the Majumdar-Ghosh Majumdar and Ghosh 1969 and Shastry-Sutherland Shastry 1988 models provide exact solutions in terms of the original spin operators. In this article, we present wavefunctions for the Kitaev chain directly in terms of the constituent spins. This model has been extensively studied as the one-dimensional version of Kitaev’s exactly solvable model on the honeycomb lattice Kitaev 2006 . It can also be viewed as the one-dimensional compass model Brzezicki et al. 2007 . Its frustrated character can be seen in the classical S\rightarrow\infty limit, which yields high ground state degeneracy. Classical ground states can be constructed starting from ‘cartesian states’, where every alternating bond is maximally satisfied Baskaran et al. 2008 ; Khatua et al. 2021 . Additional ground states appear, which can be viewed as pathways that connect pairs of cartesian states. Below, we demonstrate that the quantum version of the model also yields high degeneracy as long as the spin quantum number, S , is a half-integer. We build upon the approach of Ref. Gordon and Kee 2022 to propose wavefunctions for the spin- S model. They are not exact solutions, rather they are approximations that carry a significant weight of the true ground state(s). Two interesting features emerge: (i) There is a qualitative difference between integer and half-integer values of S . This is reminiscent of the Heisenberg chain, where such a difference originates from a topological term Tasaki 2020 . (ii) For half-integer S , ground state wavefunctions have topological character. They are obtained as dimer covers, where each dimer is assigned a ‘singlet’ or a ‘triplet’ wavefunction. (For S>1/2 , we use suitably defined analogues of the S=1/2 singlet and triplet states). Crucially, a global constraint applies that forces the total number of triplets to be even. We argue that this is a topological effect that cannot be altered by local perturbations. We describe the quantum spin- S problem using various approaches below. We define the model in Sec. II . In Sec. III , we discuss exact diagonalization results for small system sizes. Sec. IV presents a Jordan-Wigner treatment for S=1/2 . In Sec. V , we propose spin-basis wavefunctions for ground states. In Sec. VI , we justify the form of wavefunctions using a perturbative approach starting from the anisotropic K_{x}\gg K_{y} limit. In Sec. VII , we discuss the isotropic K_{x}=K_{y} point. We conclude with a discussion in Sec. VIII . II Model: Spin- S Kitaev Chain We consider a one-dimensional spin system with a two-site unit cell and a spin- S moment at each site. We have N spins that are coupled by nearest-neighbour bonds that alternate between X and Y character. The system is described by the Hamiltonian H=K_{x}\sum_{i=1}^{N/2}S_{A_{i}}^{x}S_{B_{i}}^{x}+K_{y}\sum_{i=1}^{N/2}S_{B_{i}}^{y}S_{A_{i+1}}^{y}. (1) We may enforce periodic boundary conditions by identifying the i^{\mathrm{th}} and (i+N/2)^{\mathrm{th}} unit cells. We assume that N is even and N\geq 4 , so as to allow for alternating X and Y bonds. For concreteness, we consider K_{x} and K_{y} to be positive coupling constants throughout this manuscript. Our results can be easily adapted to cases where one or both are negative, using basis changes generated by local rotations. Figure 1: Kitaev chains with periodic boundary conditions for (a) N=4 , (b) N=6 and (c) N=8 . With open boundary conditions, we fix the bonds at the ends to be of the X type, for concreteness. This is shown for N=6 in (d). This model describes a much wider class of systems than appears at first glance. As shown in Ref. Sen et al. 2010 , the isotropic limit ( K_{x}=K_{y}=1 ) of this model can be rewritten as H=\sum_{i}^{N}S_{i}^{x}S_{i+1}^{y}. (2) It can also be mapped to a 3-unit cell ‘Kitaev-XYZ’ chain with a repeating sequence of X-Y-Z bonds Yang et al. 2025 , H=\sum_{i}^{\frac{N}{3}}S_{A_{i}}^{x}S_{B_{i}}^{x}+\sum_{i}^{\frac{N}{3}}S_{B_{i}}^{y}S_{C_{i}}^{y}+\sum_{i}^{\frac{N}{3}}S_{C_{i}}^{z}S_{A_{i+1}}^{z}. (3) In Appendix A , we demonstrate that this Hamiltonian describes any chain with bonds of X , Y and Z types, with a constraint that two adjacent bonds cannot be of the same type. These chains are analogous to the Hägg code representation of close-packed structures Hägg 1943 ; Krishna, P. and Pandey, D. 1981 . To describe eigenstates of the Hamiltonian in Eq. 1 , we may use conserved quantities defined on bonds Sen et al. 2010 , \displaystyle W_{i,X}=e^{i\pi S_{A_{i}}^{y}}e^{i\pi S_{B_{i}}^{y}},~~W_{i,Y}=e^{i\pi S_{B_{i}}^{x}}e^{i\pi S_{A_{i+1}}^{x}}, (4) defined on X and Y bonds respectively. All W ’s square to unity and commute with the Hamiltonian. For integer values of S , the W ’s commute with one another. With periodic boundaries, this leads to N independent conserved quantities. For half-integer values of S , W ’s on neighbouring bonds do not commute with one another. This leads to N/2 conserved quantities, say corresponding to W ’s on the X bonds

Read the complete text