Status of Haldane's Luttinger relation $v_S=(v_Nv_J)^{1/2}$ for Bethe-ansatz models with phase-shift kernels not of difference form

The Luttinger relation $v_S=(v_Nv_J)^{1/2}$ -- identifying the gapless-mode velocity with the geometric mean of the compressibility velocity $v_N$ and the current-stiffness (Drude-weight) velocity $v_J$ -- was proved in F. D. M. Haldane's 1981 note 'Demonstration of the "Luttinger liquid" character of Bethe-ansatz-soluble models of 1-D quantum fluids' only for Bethe-ansatz phase-shift kernels of difference form $\Theta(p,q)=\Theta(\alpha(p)-\alpha(q))$, and the note also deferred the spinful generalization and the explicit Luttinger-parameter evaluations for the Hubbard chain and the sine-Gordon/massive Thirring model. Later literature completed the deferred evaluations (dressed-charge-matrix exponents for the Hubbard model; exact exponents of the incommensurate sine-Gordon/U(1)-Thirring phase), so the surviving open core is the relation itself beyond difference form. A matrix generalization -- the squares of the gapless velocities are the eigenvalues of the product of the compressibility and Drude-weight matrices -- has been derived from the Bethe ansatz for a two-component system whose phase shifts are of difference form (1994) and, beyond the difference-form class, only for the two-component Bose-Fermi mixture with a single nested kernel (2026); for two-kernel systems such as spin-1/2 fermion models (Yang-Gaudin, Hubbard) and boson-spin-1/2-fermion mixtures, and for general antisymmetric non-difference kernels, no proof or counterexample exists, and the 2026 literature explicitly poses this as open. The problem: prove, refute, or delimit by a necessary and sufficient criterion on the Bethe-ansatz kernel the Luttinger relation for one-dimensional quantum fluids whose kernels do not admit the difference-form reduction.

Background

Particles confined to one spatial dimension interact so strongly that the Landau Fermi liquid fails: the generic low-energy state of an interacting one-dimensional (1D) quantum fluid is the Luttinger liquid, a critical state whose entire low-energy Hamiltonian is quadratic in bosonic density-wave operators. Haldane's 1981 papers introduced this terminology and connected it to exact solutions. In the notation of Haldane's 1981 note, the low-energy Hamiltonian and momentum of a 1D quantum fluid on a ring of length $L$ are $H=v_S\sum_{q\neq0}|q|b_q^\dagger b_q+\frac{\pi}{2L}\left[v_N(N-N_0)^2+v_JJ^2\right],$ $P=\left[k_F+\frac{\pi}{L}(N-N_0)\right]J+\sum_{q\neq0}q\,b_q^\dagger b_q,$ where $b_q$ are operators of bosonic density waves, $N$ the particle number, $J$ the total-current (winding) quantum number, and $k_F$ the Fermi momentum. The three velocities have independent microscopic meanings: $v_S$ is the velocity of the gapless density modes (a dynamical quantity, read off the excitation spectrum), $v_N$ is fixed by the inverse compressibility $\partial\mu/\partial n$ (a thermodynamic quantity, $\mu$ the chemical potential and $n$ the density), and $v_J$ by the current stiffness -- equivalently the Drude weight, the coefficient of ballistic transport, obtainable as the response of the ground-state energy to twisted boundary conditions (a flux through the ring). The Luttinger relation $v_S=(v_Nv_J)^{1/2}$ states that the dynamical velocity is the geometric mean of the two static-response velocities. Together with the parametrization $v_N=v_Se^{-2\varphi}$, $v_J=v_Se^{2\varphi}$ it guarantees that the low-energy theory is a consistent one-parameter Gaussian family, the coupling $\varphi$ alone controlling the power-law decay exponents of all correlation functions. In a one-component Galilean-invariant fluid it reduces to $v^2=(n/m)\,\partial\mu/\partial n$ ($m$ the particle mass), realized by Lieb's exact excitation spectrum of the interacting Bose gas.

A broad class of interacting 1D models -- the Lieb-Liniger gas (bosons with contact repulsion), spin-1/2 chains, the Hubbard chain (spin-1/2 fermions on a lattice, with nearest-neighbour hopping and on-site repulsion), the massive Thirring model (a relativistic field theory of fermions with a current-current interaction, equivalent to the sine-Gordon model, a scalar field with a periodic cosine interaction potential) -- is solved exactly by the Bethe ansatz: eigenstates are built from plane waves with quasimomenta $k_j$ satisfying quantization equations containing a two-body scattering phase shift $\Theta(p,q)$, and in the thermodynamic limit these become linear Fredholm integral equations (integral equations of the second kind) for the pseudomomentum density, with kernel $\partial\Theta/\partial p$. Haldane's note proved that whenever the kernel has difference form, i.e. $\Theta(p,q)=\Theta(\alpha(p)-\alpha(q))$ for an odd reparametrization $\alpha(p)$ of the pseudomomentum, the Bethe-ansatz data imply the complete Luttinger-liquid structure, including the Luttinger relation, which follows from the identity $X[\sigma]\equiv[1-\sigma(\Lambda)-\sigma(-\Lambda)][1-\sigma(\Lambda)+\sigma(-\Lambda)]=1$ for the auxiliary function $\sigma$ of the Bethe-ansatz integral equations at the Fermi cutoff $\Lambda$. The proof uses the difference form essentially: it decouples the even and odd sectors of the associated integral operators, and for a kernel that is merely antisymmetric, $\Theta(p,q)=-\Theta(q,p)$, the required factorization is not provable by these methods. Difference form is defined above for a single two-body kernel over one rapidity species; in a nested (multi-component) model, whose coupled equations carry a matrix of kernels over several rapidity species, the model is of difference form only when one shared odd reparametrization reduces all its kernels jointly -- not blockwise, one reparametrization per species -- to functions of the reparametrized rapidity difference. The complementary non-difference class therefore contains any coupled system with two or more independent kernels or with kernels coupling distinct rapidity variables (the two-kernel spinful systems), together with the single-kernel Bose-Fermi mixture and general antisymmetric kernels. The note therefore left the status of the relation for non-difference kernels open, together with the extension to models with internal degrees of freedom and the explicit evaluation of the Luttinger parameters of the Hubbard chain and the sine-Gordon/massive Thirring model.

Much of this program was completed later. For models with internal degrees of freedom the Bethe ansatz is nested -- spin is encoded in auxiliary rapidities obeying their own coupled quantization equations, with two coupled kernels -- and the finite-size spectrum is organized by the dressed charge matrix, a matrix-valued solution of the adjoint integral equations whose boundary values fix all critical exponents (Bogoliubov, Izergin and Korepin; Frahm and Korepin; Schulz). This yields the exact charge and spin velocities and Luttinger parameters of the Hubbard chain, realizing spin-charge separation, the propagation of charge and spin density waves with different velocities. For the sine-Gordon/massive Thirring model, the incommensurate phase at finite soliton density is a Luttinger liquid with exactly known exponents (Papa and Tsvelik). In modern language the multi-component Luttinger relation reads: the squares of the gapless-mode velocities are the eigenvalues of the product of the compressibility matrix and the Drude-weight matrix; exact derivations from the Bethe ansatz exist only for two-component systems whose kernels are of difference form or reduce to a single kernel -- the two-component Sutherland model with hyperbolic interactions (Römer and Sutherland), and the Bose-Fermi mixture (bosons plus spinless fermions), whose nested equations contain a single kernel (Chandak, Petković and Ristivojevic). The unresolved question of principle is whether the relation holds for Bethe-ansatz kernels that do not admit the difference-form reduction -- the paradigm being the two-kernel spinful systems -- which is precisely the item the 2026 literature leaves open.

Problem Statement

The overarching objective is to determine the exact status of the Luttinger relation for Bethe-ansatz-soluble one-dimensional quantum fluids whose two-body phase-shift kernel(s) are antisymmetric, $\Theta(p,q)=-\Theta(q,p)$, but cannot be brought to difference form $\Theta(\alpha(p)-\alpha(q))$ by any reparametrization of the pseudomomentum (for a coupled multi-kernel system, in the joint sense fixed in Background), with every quantity computed independently and microscopically from the Bethe ansatz. Concretely:

(1) Single-component models: prove or refute $v_S=(v_Nv_J)^{1/2}$, where $v_S$ is the velocity of the gapless mode obtained from the Bethe-ansatz excitation spectrum (the dressed-energy equations, which determine the excitation energy as a function of rapidity), $v_N$ is determined by the compressibility ($v_N$ proportional to $\partial\mu/\partial n$), and $v_J$ by the current stiffness, i.e. the Drude weight obtained from the response to twisted boundary conditions.

(2) Multi-component (nested Bethe ansatz) models, the paradigm being two-kernel systems such as spin-1/2 fermion models (the Yang-Gaudin gas of spin-1/2 fermions with contact repulsion, and the Hubbard chain) and boson-spin-1/2-fermion mixtures: derive the compressibility matrix and the full Drude-weight matrix, including off-diagonal elements coupling the components, from the coupled Bethe-ansatz integral equations, and establish whether the squares of the gapless-mode velocities are exactly the eigenvalues of the product of the two matrices -- the matrix form already established for the single-kernel Bose-Fermi mixture.

(3) General criterion: give a necessary and sufficient condition on the kernel(s) of the Bethe-ansatz integral equations for the Luttinger relation to hold, thereby delimiting the class of Bethe-ansatz-soluble models that are Luttinger liquids in Haldane's strict sense; a Bethe-ansatz-solvable model with an antisymmetric non-difference kernel for which the relation fails would refute the universal form of the relation; the class-level criterion must account for any such counterexample.

Items (1)-(3) are jointly posed parts of this single objective: the objective covers the full admissible class, including the two-kernel nested systems (the Yang-Gaudin gas, the Hubbard chain, and boson-spin-1/2-fermion mixtures) that the 2026 literature explicitly leaves open, and the criterion of item (3) is a necessary and sufficient class-level condition, valid for every admissible kernel. In particular, a demonstration that the single-component non-difference subclass is empty, or that the relation holds vacuously on it, does not by itself resolve the problem.

The relation must be consistent with the known special cases (difference-form kernels including the Lieb-Liniger gas; the Galilean one-component relation $v^2=(n/m)\,\partial\mu/\partial n$; the single-kernel Bose-Fermi mixture) and must not be made true by definition: the velocities on the two sides are to be obtained from independent microscopic determinations, spectral versus thermodynamic/stiffness.

References

1. F. D. M. Haldane, 'Demonstration of the "Luttinger liquid" character of Bethe-ansatz-soluble models of 1-D quantum fluids', Physics Letters A 81 (2-3), 153-155 (1981), DOI 10.1016/0375-9601(81)90049-9, https://doi.org/10.1016/0375-9601(81)90049-9
2. F. D. M. Haldane, ''Luttinger liquid theory' of one-dimensional quantum fluids. I. Properties of the Luttinger model and their extension to the general 1D interacting spinless Fermi gas', Journal of Physics C: Solid State Physics 14, 2585-2609 (1981), DOI 10.1088/0022-3719/14/19/010, https://doi.org/10.1088/0022-3719/14/19/010
3. F. D. M. Haldane, 'Quantum field ground state of the sine-Gordon model with finite soliton density: exact results', Journal of Physics A: Mathematical and General 15, 507-525 (1982), DOI 10.1088/0305-4470/15/2/021, https://doi.org/10.1088/0305-4470/15/2/021
4. Elliott H. Lieb, 'Exact Analysis of an Interacting Bose Gas. II. The Excitation Spectrum', Physical Review 130, 1616-1624 (1963), DOI 10.1103/PhysRev.130.1616, https://doi.org/10.1103/PhysRev.130.1616
5. Yukio Okwamoto, 'The Thirring Model at a Finite Fermion Density', Journal of the Physical Society of Japan 52, 1081-1083 (1983), DOI 10.1143/JPSJ.52.1081, https://doi.org/10.1143/JPSJ.52.1081
6. N. M. Bogoliubov, A. G. Izergin, V. E. Korepin, 'Critical exponents for integrable models', Nuclear Physics B 275, 687-705 (1986), DOI 10.1016/0550-3213(86)90579-1, https://doi.org/10.1016/0550-3213(86)90579-1
7. B. Sriram Shastry, Bill Sutherland, 'Twisted boundary conditions and effective mass in Heisenberg-Ising and Hubbard rings', Physical Review Letters 65, 243-246 (1990), DOI 10.1103/PhysRevLett.65.243, https://doi.org/10.1103/PhysRevLett.65.243
8. Holger Frahm, V. E. Korepin, 'Critical exponents for the one-dimensional Hubbard model', Physical Review B 42, 10553-10565 (1990), DOI 10.1103/PhysRevB.42.10553, https://doi.org/10.1103/PhysRevB.42.10553
9. H. J. Schulz, 'Correlation exponents and the metal-insulator transition in the one-dimensional Hubbard model', Physical Review Letters 64, 2831-2834 (1990), DOI 10.1103/PhysRevLett.64.2831, https://doi.org/10.1103/PhysRevLett.64.2831
10. Rudolf A. Römer, Bill Sutherland, 'Exact derivation of Luttinger liquid relation in a one-dimensional two-component quantum system with hyperbolic interactions', Physics Letters A 190, 295-300 (1994), DOI 10.1016/0375-9601(94)90758-7, preprint arXiv:cond-mat/9402072, https://doi.org/10.1016/0375-9601(94)90758-7
11. J. Voit, 'One-dimensional Fermi liquids', Reports on Progress in Physics 58, 977-1116 (1995), DOI 10.1088/0034-4885/58/9/002, https://doi.org/10.1088/0034-4885/58/9/002
12. Satoshi Fujimoto, Norio Kawakami, 'Exact Drude weight for the one-dimensional Hubbard model at finite temperatures', Journal of Physics A: Mathematical and General 31, 465-474 (1998), DOI 10.1088/0305-4470/31/2/008, https://doi.org/10.1088/0305-4470/31/2/008
13. Emiliano Papa, Alexei M. Tsvelik, 'Scaling exponents in the incommensurate phase of the sine-Gordon and U(1) Thirring models', Physical Review B 63, 085109 (2001), DOI 10.1103/PhysRevB.63.085109, https://doi.org/10.1103/PhysRevB.63.085109
14. Fabian H. L. Essler, Holger Frahm, Frank Göhmann, Andreas Klümper, Vladimir E. Korepin, 'The One-Dimensional Hubbard Model', Cambridge University Press (2005), DOI 10.1017/CBO9780511534843, https://doi.org/10.1017/CBO9780511534843
15. Rogier Vlijm, Sebas Eliens, Jean-Sébastien Caux, 'Correlations of zero-entropy critical states in the XXZ model: integrability and Luttinger theory far from the ground state', SciPost Physics 1, 008 (2016), DOI 10.21468/SciPostPhys.1.1.008, https://doi.org/10.21468/SciPostPhys.1.1.008
16. Frank Göhmann, Andreas Klümper, Karol K. Kozlowski, 'Ballistic Particle Transport and Drude Weight in Gases', Physical Review Letters 135, 263401 (2025), DOI 10.1103/395w-v9fq, preprint arXiv:2506.00686, https://doi.org/10.1103/395w-v9fq
17. Soham Chandak, Aleksandra Petković, Zoran Ristivojevic, 'Bethe-ansatz study of the Bose-Fermi mixture', Physical Review B 113, 195141 (2026), DOI 10.1103/5kqv-f1vt, preprint arXiv:2512.21732, https://doi.org/10.1103/5kqv-f1vt
18. Zi-yang Liu, Xiangguo Yin, Yunbo Zhang, Shizhong Zhang, Xi-Wen Guan, 'Universal Transport Properties of Continuous Quantum Gases', arXiv:2602.22009 (2026), https://arxiv.org/abs/2602.22009