For the Bethe ansatz equations (BAE) of the periodic spin-1/2 isotropic Heisenberg (XXX) chain, Hao, Nepomechie and Sommese [Phys. Rev. E 88, 052113 (2013)] used numerical homotopy continuation (all solutions found for chain lengths N≤14N\le14) to conjecture the completeness ("Pauli-principle") relation N(N,M)−Ns(N,M)+Nsp(N,M)=(NM)−(NM−1)\mathcal{N}(N,M)-\mathcal{N}_s(N,M)+\mathcal{N}_{sp}(N,M)=\binom{N}{M}-\binom{N}{M-1} among the numbers of solutions with pairwise distinct roots, singular solutions (containing the pair {i/2,−i/2}\{i/2,-i/2\}), and physical singular solutions (those passing the Nepomechie-Wang criterion). Later rigorous work settled the count of physical solutions in the polynomial TQTQ/QQ-system formulation, but the conjecture as a statement about the BAE's own solution set is still open - it is now equivalent, modulo the proven results, to excluding physical solutions with repeated roots ("strange" solutions), which is proven only for the Lieb-Liniger analogue (the one-dimensional Bose gas with contact interactions) and is supported numerically for the XXX chain. Also open are closed-form expressions for the individual counts for M≥3M\ge3 (known only for M≤2M\le2) and the thermodynamic-limit (N→∞N\to\infty) fate of the unphysical singular solutions, whose number at finite NN with M∼N/2M\sim N/2 is comparable to the number of highest-weight states - in tension with the string hypothesis underlying all classical thermodynamic counting. The problem asks for a complete rigorous accounting of the finite-NN solution set of the XXX Bethe equations and of its thermodynamic limit.

Background

The isotropic Heisenberg chain (XXX chain) is the periodic quantum spin chain of NN spin-1/21/2 degrees of freedom with Hamiltonian

H=14∑n=1N(σ⃗n⋅σ⃗n+1−1),σ⃗N+1≡σ⃗1,H=\frac{1}{4}\sum_{n=1}^{N}\left(\vec\sigma_n\cdot\vec\sigma_{n+1}-1\right),\qquad \vec\sigma_{N+1}\equiv\vec\sigma_1,

where σ⃗n=(σnx,σny,σnz)\vec\sigma_n=(\sigma_n^x,\sigma_n^y,\sigma_n^z) are Pauli matrices acting on site nn. It is the prototypical exactly solvable ("quantum integrable") many-body system: Bethe's 1931 solution replaces diagonalizing the 2N×2N2^N\times 2^N matrix HH by an algebraic problem. In the sector with MM reversed spins (0≤M≤N/20\le M\le N/2; MM is called the magnon number), eigenvectors are parametrized by MM complex numbers {λ1,…,λM}\{\lambda_1,\dots,\lambda_M\}, called Bethe roots, satisfying the Bethe ansatz equations (BAE)

(λk+i2)N∏j≠k(λk−λj−i)=(λk−i2)N∏j≠k(λk−λj+i),k=1,…,M,\left(\lambda_k+\frac{i}{2}\right)^N\prod_{j\neq k}\left(\lambda_k-\lambda_j-i\right)=\left(\lambda_k-\frac{i}{2}\right)^N\prod_{j\neq k}\left(\lambda_k-\lambda_j+i\right),\qquad k=1,\dots,M,

with energy E=−12∑k=1M(λk2+14)−1E=-\frac{1}{2}\sum_{k=1}^{M}\left(\lambda_k^2+\frac{1}{4}\right)^{-1}.

Completeness asks whether the Bethe roots account for all 2N2^N eigenstates. Because the chain is invariant under global spin rotations, the algebraic Bethe ansatz - a construction of eigenvectors as products of commuting creation operators B(λ1)⋯B(λM)B(\lambda_1)\cdots B(\lambda_M) acting on the fully polarized reference state - produces only highest-weight states (states annihilated by the total spin raising operator S+S^+); the remaining states follow by applying the lowering operator S−S^-. In the sector with MM reversed spins the number of highest-weight states is (NM)−(NM−1)\binom{N}{M}-\binom{N}{M-1}, so a complete solution should produce exactly that many "physical" solutions of the BAE per sector. The subtlety, known since the 1980s, is that the BAE also admit singular (or exceptional) solutions containing the pair {i/2,−i/2}\{i/2,-i/2\} (an "exact two-string"): for such solutions the Bethe vector and the energy formula are naively ill-defined (both numerator and denominator vanish), and only a subset correspond to genuine eigenvectors. Using a regularized algebraic Bethe ansatz, Nepomechie and Wang (2013) derived a criterion: a singular solution {i/2,−i/2,λ3,…,λM}\{i/2,-i/2,\lambda_3,\dots,\lambda_M\} is physical if and only if the remaining roots satisfy

[−∏k=3Mλk+i2λk−i2]N=1.\left[-\prod_{k=3}^{M}\frac{\lambda_k+\frac{i}{2}}{\lambda_k-\frac{i}{2}}\right]^N=1.

Hao, Nepomechie and Sommese (2013) computed all solutions of the BAE with pairwise distinct roots - counted up to permutation of the roots - for N≤14N\le14 and all M≤N/2M\le N/2, using homotopy continuation, a method of numerical algebraic geometry that tracks every isolated solution of a polynomial system along a deformation from a trivially solvable start system. Writing N(N,M)\mathcal{N}(N,M) for the number of such distinct-root solutions, Ns(N,M)\mathcal{N}_s(N,M) for the number of them that are singular, and Nsp(N,M)\mathcal{N}_{sp}(N,M) for the number of singular ones that are physical, they found that the combination N−Ns+Nsp\mathcal{N}-\mathcal{N}_s+\mathcal{N}_{sp} equals (NM)−(NM−1)\binom{N}{M}-\binom{N}{M-1} in every case (for example, for N=14N=14, M=7M=7: (N,Ns,Nsp)=(1701,1287,15)(\mathcal{N},\mathcal{N}_s,\mathcal{N}_{sp})=(1701,1287,15), and 1701−1287+15=429=(147)−(146)1701-1287+15=429=\binom{14}{7}-\binom{14}{6}), and conjectured this as the completeness ("Pauli-principle") relation. The numbers are striking: for M∼N/2M\sim N/2 the equations have roughly four times more distinct-root solutions than physical ones, and most singular solutions are unphysical.

The conjecture is entangled with the string hypothesis - the assumption, used in essentially all thermodynamic (N→∞N\to\infty) computations since Bethe and Takahashi, that the Bethe roots of the dominant solutions organize into regular arithmetic patterns in the complex plane ("strings") λ+im,λ+i(m−1),…,λ−im\lambda+im,\lambda+i(m-1),\dots,\lambda-im. Unphysical singular solutions often violate the string hypothesis (many are not even invariant under complex conjugation), yet the classical string-hypothesis computations still recover the correct number of states; the source paper suggests, heuristically, that the string hypothesis may "effectively project out" the unphysical solutions, and asks for this to be understood.

A parallel rigorous reformulation exists. The transfer matrices T(u)T(u) - a one-parameter commuting family of operators on the chain's Hilbert space, constructed as a trace over an auxiliary space of a product of local Lax operators acting site by site, whose simultaneous eigenvalues encode the spectrum of HH - satisfy an intertwining TQTQ relation with a polynomial Q(u)Q(u); solving the spin chain is equivalent to finding pairs (T,Q)(T,Q) for which the second, dual solution P(u)P(u) of the same relation is also a polynomial. Requiring polynomiality of both solutions is equivalent (Granet-Jacobsen 2020) to a finite set of zero-remainder conditions - divisibility checks of explicit polynomials - whose iteration defines the rational Q-system (Marboe-Volin 2017; Hou-Jiang-Zhu 2024). In this formulation the counting of physical solutions is rigorous: Mukhin, Tarasov and Varchenko (2009) proved that the Bethe algebra (the algebra generated by the commuting transfer matrices) of the homogeneous XXX chain has simple spectrum on highest-weight states, with exactly (NM)−(NM−1)\binom{N}{M}-\binom{N}{M-1} polynomial solution pairs of the Wronskian form of the TQTQ relation (the form in which the transfer matrix is eliminated so that the two solutions QQ and PP enter only through their Wronskian determinant) per sector; Tarasov (2018) proved the corresponding completeness for the periodic isotropic Heisenberg model; and Granet-Jacobsen (2020) showed these polynomial solutions are precisely the physical solutions of the BAE. What this chain of results does not settle is the conjecture as a statement about the BAE's own solution set, because it does not exclude physical solutions with repeated roots - which do occur ("strange" solutions) in the higher-spin XXX chains (Hao-Nepomechie-Sommese 2014), and which the source paper itself could not preclude for spin 1/21/2.

Problem Statement

Consider the Bethe ansatz equations (BAE) of the periodic spin-1/2 isotropic Heisenberg (XXX) chain of NN sites in the sector with MM reversed spins (0≤M≤N/20\le M\le N/2),

(λk+i2)N∏j≠k(λk−λj−i)=(λk−i2)N∏j≠k(λk−λj+i),k=1,…,M,\left(\lambda_k+\frac{i}{2}\right)^N\prod_{j\neq k}\left(\lambda_k-\lambda_j-i\right)=\left(\lambda_k-\frac{i}{2}\right)^N\prod_{j\neq k}\left(\lambda_k-\lambda_j+i\right),\qquad k=1,\dots,M,

for the Bethe roots {λ1,…,λM}\{\lambda_1,\dots,\lambda_M\}. Let N(N,M)\mathcal{N}(N,M) be the number of solutions with pairwise distinct roots, counted up to permutation of the roots; let Ns(N,M)\mathcal{N}_s(N,M) be the number of these that are singular, i.e. contain the pair {i/2,−i/2}\{i/2,-i/2\}; and let Nsp(N,M)\mathcal{N}_{sp}(N,M) be the number of singular ones that are physical, i.e. satisfy the Nepomechie-Wang criterion

[−∏k=3Mλk+i2λk−i2]N=1\left[-\prod_{k=3}^{M}\frac{\lambda_k+\frac{i}{2}}{\lambda_k-\frac{i}{2}}\right]^N=1

on the remaining roots λ3,…,λM\lambda_3,\dots,\lambda_M. Give a complete rigorous accounting of the finite-NN solution set of these equations and of its thermodynamic limit, as follows.

(1) Prove, as a statement about the solutions of the BAE themselves, that every physical solution has pairwise distinct Bethe roots (no "strange" solutions), and hence the completeness/Pauli-principle relation

N(N,M)−Ns(N,M)+Nsp(N,M)=(NM)−(NM−1)\mathcal{N}(N,M)-\mathcal{N}_s(N,M)+\mathcal{N}_{sp}(N,M)=\binom{N}{M}-\binom{N}{M-1}

for all NN and all 0≤M≤N/20\le M\le N/2. An explicit physical solution with repeated roots for some (N,M)(N,M) would refute this component and is an admissible answer to it.

(2) Determine closed-form expressions, with proof, for the individual counts N(N,M)\mathcal{N}(N,M), Ns(N,M)\mathcal{N}_s(N,M) and Nsp(N,M)\mathcal{N}_{sp}(N,M) for general M≥3M\ge3; only N(N,1)=N−1\mathcal{N}(N,1)=N-1 and N(N,2)=12(N2+3N+1+(−1)N)\mathcal{N}(N,2)=\frac{1}{2}\left(N^2+3N+1+(-1)^N\right) are currently known.

(3) Determine analytically the fate of the unphysical singular solutions as N→∞N\to\infty, in two parts. (3a) In sectors with M∼N/2M\sim N/2, determine the asymptotic behaviour, as N→∞N\to\infty, of the ratio

Ns(N,M)−Nsp(N,M)(NM)−(NM−1)\frac{\mathcal{N}_s(N,M)-\mathcal{N}_{sp}(N,M)}{\binom{N}{M}-\binom{N}{M-1}}

between the number of unphysical singular solutions and the number of highest-weight states: decide whether it tends to zero (the excess of unphysical singular solutions is negligible compared with the number of highest-weight states), tends to a finite positive constant, or diverges; if no limit exists, give the sharpest proved bounds on its lim inf⁡\liminf and lim sup⁡\limsup (finite-NN data suggest the ratio is not small). (3b) Explain rigorously - by proof, or by a controlled argument that states its assumptions and their justification - why string-hypothesis-based thermodynamic counting obtains the correct number of states despite the excess of distinct-root solutions; that is, reconcile the classical thermodynamic completeness arguments with the finite-NN census. All three components are mandatory, and the task is complete only when each of (1), (2) and (3) - the latter being the conjunction of (3a) and (3b) - is resolved; a solution of (2) that determines the closed forms for Ns(N,M)\mathcal{N}_s(N,M) and Nsp(N,M)\mathcal{N}_{sp}(N,M) also discharges the quantitative clause (3a), so that only the explanatory clause (3b) remains as a separate requirement. The three components share one objective - a complete accounting of the solution set of the prototype integrable chain - and are linked: (1) identifies the physical subset, (2) quantifies the census, and (3) is its thermodynamic limit.

References

  1. Wenrui Hao, Rafael I. Nepomechie, Andrew J. Sommese, "Completeness of solutions of Bethe's equations", Phys. Rev. E 88, 052113 (2013). DOI: 10.1103/PhysRevE.88.052113, https://doi.org/10.1103/PhysRevE.88.052113 (preprint arXiv:1308.4645, https://arxiv.org/abs/1308.4645)
  2. Rafael I. Nepomechie, Chunguang Wang, "Algebraic Bethe ansatz for singular solutions", J. Phys. A 46, 325002 (2013). DOI: 10.1088/1751-8113/46/32/325002, https://doi.org/10.1088/1751-8113/46/32/325002 (preprint arXiv:1304.7978, https://arxiv.org/abs/1304.7978)
  3. Wenrui Hao, Rafael I. Nepomechie, Andrew J. Sommese, "Singular solutions, repeated roots and completeness for higher-spin chains", J. Stat. Mech. (2014) P03024. DOI: 10.1088/1742-5468/2014/03/P03024, https://doi.org/10.1088/1742-5468/2014/03/P03024 (preprint arXiv:1312.2982, https://arxiv.org/abs/1312.2982)
  4. Rafael I. Nepomechie, Chunguang Wang, "Twisting singular solutions of Bethe's equations", J. Phys. A 47, 505004 (2014). DOI: 10.1088/1751-8113/47/50/505004, https://doi.org/10.1088/1751-8113/47/50/505004 (preprint arXiv:1409.7382, https://arxiv.org/abs/1409.7382)
  5. H. Bethe, "Zur Theorie der Metalle", Z. Phys. 71, 205-226 (1931). DOI: 10.1007/BF01341708, https://doi.org/10.1007/BF01341708
  6. Minoru Takahashi, "One-Dimensional Heisenberg Model at Finite Temperature", Prog. Theor. Phys. 46, 401-415 (1971). DOI: 10.1143/PTP.46.401, https://doi.org/10.1143/PTP.46.401
  7. E. Mukhin, V. Tarasov, A. Varchenko, "Bethe Algebra of Homogeneous XXX Heisenberg Model Has Simple Spectrum", Commun. Math. Phys. 288, 1-42 (2009). DOI: 10.1007/s00220-009-0733-4, https://doi.org/10.1007/s00220-009-0733-4 (preprint arXiv:0706.0688, https://arxiv.org/abs/0706.0688)
  8. V. Tarasov, "Completeness of the Bethe Ansatz for the Periodic Isotropic Heisenberg Model", Rev. Math. Phys. 30, 1840018 (2018). DOI: 10.1142/S0129055X18400184, https://doi.org/10.1142/S0129055X18400184
  9. Anatol N. Kirillov, Reiho Sakamoto, "Singular Solutions to the Bethe Ansatz Equations and Rigged Configurations", J. Phys. A 47, 205207 (2014). DOI: 10.1088/1751-8113/47/20/205207, https://doi.org/10.1088/1751-8113/47/20/205207 (preprint arXiv:1402.0651, https://arxiv.org/abs/1402.0651)
  10. Anatol N. Kirillov, Reiho Sakamoto, "Some Remarks on Nepomechie-Wang Eigenstates For Spin 1/2 XXX Model", Moscow Math. J. 15, 337-352 (2015). DOI: 10.17323/1609-4514-2015-15-2-337-352, https://doi.org/10.17323/1609-4514-2015-15-2-337-352 (preprint arXiv:1406.1958, https://arxiv.org/abs/1406.1958)
  11. Reiho Sakamoto, "Rigged Configurations Approach For The Spin-1/2 Isotropic Heisenberg Model", J. Phys. A 48, 165201 (2015). DOI: 10.1088/1751-8113/48/16/165201, https://doi.org/10.1088/1751-8113/48/16/165201 (preprint arXiv:1501.00532, https://arxiv.org/abs/1501.00532)
  12. Anatol N. Kirillov, Reiho Sakamoto, "Bethe's Quantum Numbers And Rigged Configurations", Nucl. Phys. B 905, 359-372 (2016). DOI: 10.1016/j.nuclphysb.2016.02.020, https://doi.org/10.1016/j.nuclphysb.2016.02.020 (preprint arXiv:1509.02305, https://arxiv.org/abs/1509.02305)
  13. Pulak Ranjan Giri, Tetsuo Deguchi, "Singular eigenstates in the even(odd) length Heisenberg spin chain", J. Phys. A 48, 175207 (2015). DOI: 10.1088/1751-8113/48/17/175207, https://doi.org/10.1088/1751-8113/48/17/175207 (preprint arXiv:1411.5839, https://arxiv.org/abs/1411.5839)
  14. Bernie Nickel, "Observations on the Bethe ansatz solutions of the spin-1/2 isotropic anti-ferromagnetic Heisenberg chain", J. Phys. Commun. 3, 025007 (2019). DOI: 10.1088/2399-6528/ab030b, https://doi.org/10.1088/2399-6528/ab030b
  15. Christian Marboe, Dmytro Volin, "Fast analytic solver of rational Bethe equations", J. Phys. A 50, 204002 (2017). DOI: 10.1088/1751-8121/aa6b88, https://doi.org/10.1088/1751-8121/aa6b88 (preprint arXiv:1608.06504, https://arxiv.org/abs/1608.06504)
  16. Yunfeng Jiang, Yang Zhang, "Algebraic geometry and Bethe ansatz. Part I. The quotient ring for BAE", J. High Energy Phys. 2018, 087 (2018). DOI: 10.1007/JHEP03(2018)087, https://doi.org/10.1007/JHEP03(2018)087 (preprint arXiv:1710.04693, https://arxiv.org/abs/1710.04693)
  17. Jon Links, "Completeness of the Bethe states for the rational, spin-1/2 Richardson-Gaudin system", SciPost Phys. 3, 007 (2017). DOI: 10.21468/SciPostPhys.3.1.007, https://doi.org/10.21468/SciPostPhys.3.1.007 (preprint arXiv:1603.03542, https://arxiv.org/abs/1603.03542)
  18. Azat M. Gainutdinov, Wenrui Hao, Rafael I. Nepomechie, Andrew J. Sommese, "Counting solutions of the Bethe equations of the quantum group invariant open XXZ chain at roots of unity", J. Phys. A 48, 494003 (2015). DOI: 10.1088/1751-8113/48/49/494003, https://doi.org/10.1088/1751-8113/48/49/494003 (preprint arXiv:1505.02104, https://arxiv.org/abs/1505.02104)
  19. Dmitry Chernyak, Sébastien Leurent, Dmytro Volin, "Completeness of Wronskian Bethe equations for rational gl(m|n) spin chains", Commun. Math. Phys. 391, 969-1045 (2022). DOI: 10.1007/s00220-021-04275-9, https://doi.org/10.1007/s00220-021-04275-9 (preprint arXiv:2004.02865, https://arxiv.org/abs/2004.02865)
  20. Etienne Granet, Jesper Lykke Jacobsen, "On zero-remainder conditions in the Bethe ansatz", J. High Energy Phys. 2020, 178 (2020). DOI: 10.1007/JHEP03(2020)178, https://doi.org/10.1007/JHEP03(2020)178 (preprint arXiv:1910.07797, https://arxiv.org/abs/1910.07797)
  21. Hongfei Shu, Peng Zhao, Rui-Dong Zhu, Hao Zou, "Bethe-State Counting and the Witten Index", SciPost Phys. 15, 103 (2023). DOI: 10.21468/SciPostPhys.15.3.103, https://doi.org/10.21468/SciPostPhys.15.3.103 (preprint arXiv:2210.07116, https://arxiv.org/abs/2210.07116)
  22. Jue Hou, Yunfeng Jiang, Rui-Dong Zhu, "Spin-s Rational Q-system", SciPost Phys. 16, 113 (2024). DOI: 10.21468/SciPostPhys.16.4.113, https://doi.org/10.21468/SciPostPhys.16.4.113 (preprint arXiv:2303.07640, https://arxiv.org/abs/2303.07640)
  23. Jue Hou, Yunfeng Jiang, Yuan Miao, "Rational Q-systems at Root of Unity I. Closed Chains", SciPost Phys. 16, 129 (2024). DOI: 10.21468/SciPostPhys.16.5.129, https://doi.org/10.21468/SciPostPhys.16.5.129 (preprint arXiv:2310.14966, https://arxiv.org/abs/2310.14966)
  24. Hongfei Shu, Peng Zhao, Rui-Dong Zhu, Hao Zou, "Counting Bethe States in Twisted Spin Chains", J. High Energy Phys. 2025, 087 (2025). DOI: 10.1007/JHEP03(2025)087, https://doi.org/10.1007/JHEP03(2025)087 (preprint arXiv:2409.20338, https://arxiv.org/abs/2409.20338)
  25. Yi-Jun He, Jue Hou, Yi-Chao Liu, Zi-Xi Tan, "Spin-s Q-systems: Twist and Open Boundaries", SciPost Phys. Core 8, 057 (2025). DOI: 10.21468/SciPostPhysCore.8.3.057, https://doi.org/10.21468/SciPostPhysCore.8.3.057 (preprint arXiv:2502.15636, https://arxiv.org/abs/2502.15636)
  26. Yunfeng Jiang, Yi-Chao Liu, Yuan Miao, Zi-Xi Tan, "Rational Q-systems for integrable spin chains without U(1) symmetry", SciPost Phys. 20, 175 (2026). DOI: 10.21468/SciPostPhys.20.6.175, https://doi.org/10.21468/SciPostPhys.20.6.175 (preprint arXiv:2512.01551, https://arxiv.org/abs/2512.01551)
  27. Shinsuke Iwao, Kohei Motegi, "Bethe roots for periodic TASEP and algebraic curve", arXiv:2504.19690 (2025), https://arxiv.org/abs/2504.19690