Positive Boltzmann weights and three-dimensional solvable lattice models from Clifford-algebra tetrahedron-equation solutions

Clifford-algebra methods recently produced an entire family of solutions of the tetrahedron equation — the three-dimensional generalization of the Yang–Baxter equation — depending on arbitrary complex coefficients, and the authors observe that several of these solutions carry positive Boltzmann weights and 'can find potential application in the construction of three dimensional statistical physics models'. No systematic classification of the parameter choices and Clifford realizations guaranteeing positivity exists, and no concrete three-dimensional lattice model has been built from any member of this family. This problem asks for (i) the positivity classification across the solution family and (ii) the explicit construction of a three-dimensional solvable lattice model — local states, weights, partition function, commuting transfer matrices — from a positive-weight member, with a basic physical characterization, the analogue for the Clifford family of what the 2013 q-oscillator model achieved for its own family.

Background

The Yang–Baxter equation is the consistency condition behind exactly solvable models in 1+1 dimensions: an operator $R$ acting on two copies of a local Hilbert space $V$ satisfies it precisely when the associated two-dimensional vertex model has commuting transfer matrices, equivalently when the scattering of particles in the plane factorizes. Its three-dimensional analogue is the tetrahedron equation, formulated by Zamolodchikov in 1981 from the factorized scattering of infinitely long straight strings (lines) in 2+1-dimensional spacetime. In its vertex (particle, or edge-labeled) form the equation reads

$$R_{123}R_{145}R_{246}R_{356} = R_{356}R_{246}R_{145}R_{123},$$

where the tetrahedron operator $R$ acts on three copies of $V$; the two orderings encode the two ways of resolving a three-string collision. Hietarinta's labeling schemes — the consistent ways of labeling the three-string scattering process by particles (edges), strings (faces), or vacua (cells) — give equivalent forms: the string (face-labeled) Frenkel–Moore form $R_{ijk}R_{ijl}R_{ikl}R_{jkl} = R_{jkl}R_{ikl}R_{ijl}R_{ijk}$, the vacua (cell-labeled) form, and algebraic variants from three-dimensional integrability — the quantized Yang–Baxter (RLLL) and MMLL equations, which are consistency relations among $L$- and $M$-operators whose associativity conditions reproduce the tetrahedron equation. The forms are related by Wu–Kadanoff-type dualities (three-dimensional analogues of the duality between vertex and face models in two dimensions), and each admits a spectral-parameter-dependent version, the spectral parameters being complex labels carried by the scattering strings.

Solutions of the tetrahedron equation are rare. The classical ones — Zamolodchikov's original solution, Baxter's treatment of it, and the Bazhanov–Stroganov construction — were long deemed insufficient for physical applications, largely because their local Boltzmann weights are not all positive: a genuine statistical-mechanical model requires weights of the form $e^{-E/T} > 0$, where $E$ is a local energy and $T$ the temperature. The first solvable three-dimensional lattice model with non-negative Boltzmann weights was constructed only in 2013 by Mangazeev, Bazhanov and Sergeev from $q$-oscillator $L$-operators; its edge variables take infinitely many discrete states, and its layer-to-layer transfer matrices form a commuting two-parameter family.

Recently Padmanabhan and Korepin introduced a universal algebraic method based on Clifford algebras — associative algebras generated by elements whose pairwise anticommutators equal $\pm 1$, denoted $CL(p,q)$ for signature $(p,q)$. Take two operators $A$ and $B$ that anticommute, $\{A,B\} = 0$, with copies $A_i, B_i$ on each factor that mutually commute between different sites. Monomials in the $A$'s and $B$'s of fixed letter content, and arbitrary linear combinations within a content class, then solve all of the above forms of the tetrahedron equation, the higher simplex equations, and their variants; the solutions form a linear space. In particular

$$R_{ijk} = \mu_0 B_i B_j B_k + \mu_1 A_i A_j B_k + \mu_2 A_i B_j A_k + \mu_3 B_i A_j A_k$$

solves the vertex form (reference [8] normalizes $\mu_0 = 1$ and writes the remaining arbitrary complex coefficients as $\alpha, \beta, \gamma$), with the coefficients identifiable as non-additive spectral parameters. The analytical proofs, explicit finite-dimensional matrix realizations (for example qubit, i.e. two-state, representations), and the invertibility analysis appear in the companion paper (Nuclear Physics B 1007 (2024) 116664); a companion 'lifting' construction (Nuclear Physics B 1014 (2025) 116865) builds further tetrahedron and higher-simplex solutions from Yang–Baxter operators and contains the only existing partial analysis of positivity.

In the short proceedings note [8], the authors observe that several of these Clifford solutions have positive Boltzmann weights and can therefore 'find potential application in the construction of three dimensional statistical physics models.' What is missing — and what the companion literature also does not supply — is, first, a systematic classification of which parameter choices and which Clifford realizations actually produce positive weights, and second, an explicit three-dimensional solvable lattice model built from any of them, analogous to what exists for the $q$-oscillator family. In the segment-colored vertex construction the local weight at the intersection point of three lattice planes is a matrix element of $R$ in the chosen local basis, so positivity is a property of the chosen realization and labeling, not of the algebraic solution alone — which is precisely what makes the classification nontrivial.

Problem Statement

Determine which members of the Clifford-algebra family of tetrahedron-equation solutions can serve as physically meaningful three-dimensional statistical models, and construct at least one such model explicitly.

(i) Positivity classification. For the solution family of Padmanabhan–Korepin (Nucl. Phys. B 1007 (2024) 116664) and reference [8] — the vertex-form operators

$$R_{ijk} = \mu_0 B_i B_j B_k + \mu_1 A_i A_j B_k + \mu_2 A_i B_j A_k + \mu_3 B_i A_j A_k$$

with $\{A_i, B_i\} = 0$ and mutual commutation between different sites, in reference [8]'s normalization $\mu_0 = 1$ and $\mu_1, \mu_2, \mu_3$ (written there as $\alpha, \beta, \gamma$) arbitrary complex parameters, their $A \leftrightarrow B$ variants, the spectral-parameter-dependent versions (the Baxterized 'non-regular' solutions of the companion literature, in which site-dependent parameters replace the constants and the spectral parameters enter non-additively), and the Frenkel–Moore-form solutions of the same family — determine the parameter regions and Clifford realizations (signature $CL(p,q)$, representation, local dimension, basis/labeling) for which the local Boltzmann weights are real and positive, where in the segment-colored vertex construction a local Boltzmann weight is a matrix element of $R$ in the chosen local basis, and the face-labeled Frenkel–Moore members are treated in the corresponding face-colored construction. Positivity of all matrix elements of $R$ is a sufficient but not necessary condition for positive weights; if such a stronger sufficient condition is classified instead of the weights themselves, the classification must make explicit which condition it establishes and which labelings it covers. The pointwise-verified positive lifted classes $H_{1,4}$, $H_{2,3}$, $H_{3,1}$ of Nucl. Phys. B 1014 (2025) 116865 lie outside this classification scope; they serve as known positive members of the broader Clifford framework and are available for part (ii).

(ii) Model construction. From a positive-weight member of the Clifford family — for instance one of the pointwise-verified lifted classes $H_{1,4}$, $H_{2,3}$, $H_{3,1}$ of Nucl. Phys. B 1014 (2025) 116865, or any positive-weight member identified by the classification in part (i) — construct a concrete three-dimensional solvable lattice statistical model: specify the lattice, the local state variables and their dimension, the explicit Boltzmann weights as functions of the ansatz parameters, the partition function, and the transfer-matrix construction; establish the model's integrability (commuting transfer matrices following from the tetrahedron equation); and give a basic physical characterization of the resulting model.

The two parts jointly serve one scientific objective: converting the authors' observation that 'several of the solutions shown in this work have positive Boltzmann weights' into an actual exactly solvable three-dimensional statistical system.

References

1. A. B. Zamolodchikov, "Tetrahedron equations and the relativistic S-matrix of straight-strings in 2+1-Dimensions", Communications in Mathematical Physics 79 (1981) 489-505, DOI: 10.1007/BF01209309, https://doi.org/10.1007/BF01209309
2. R. J. Baxter, "On Zamolodchikov's solution of the tetrahedron equations", Communications in Mathematical Physics 88 (1983) 185-205, DOI: 10.1007/BF01209476, https://doi.org/10.1007/BF01209476
3. V. V. Bazhanov, Yu. G. Stroganov, "Conditions of commutativity of transfer matrices on a multidimensional lattice", Theoretical and Mathematical Physics 52 (1982) 685-691, DOI: 10.1007/BF01027789, https://doi.org/10.1007/BF01027789
4. I. Frenkel, G. Moore, "Simplex equations and their solutions", Communications in Mathematical Physics 138 (1991) 259-271, DOI: 10.1007/BF02099492, https://doi.org/10.1007/BF02099492
5. V. V. Mangazeev, V. V. Bazhanov, S. M. Sergeev, "An integrable 3D lattice model with positive Boltzmann weights", Journal of Physics A: Mathematical and Theoretical 46 (2013) 465206, arXiv:1308.4773, DOI: 10.1088/1751-8113-46-46-465206, https://arxiv.org/abs/1308.4773
6. Pramod Padmanabhan, Vladimir Korepin, "Solving the Yang-Baxter, tetrahedron and higher simplex equations using Clifford algebras", Nuclear Physics B 1007 (2024) 116664, arXiv:2404.11501, DOI: 10.1016/j.nuclphysb.2024.116664, https://arxiv.org/abs/2404.11501
7. Pramod Padmanabhan, Vladimir Korepin, "Majorana fermions solve the tetrahedron equations as well as higher simplex equations", Nuclear Physics B 1014 (2025) 116865, arXiv:2410.20328, DOI: 10.1016/j.nuclphysb.2025.116865, https://arxiv.org/abs/2410.20328
8. Pramod Padmanabhan, Vivek Kumar Singh, Vladimir Korepin, "Clifford Solver for the Tetrahedron Equation and its Variants", Bulgarian Journal of Physics 52 (2025) s1 126-131, arXiv:2510.23944, https://arxiv.org/abs/2510.23944
9. Vivek Kumar Singh, Akash Sinha, Pramod Padmanabhan, Vladimir Korepin, "Unitary tetrahedron quantum gates", arXiv:2407.10731, https://arxiv.org/abs/2407.10731
10. Shinsuke Iwao, Kohei Motegi, Ryo Ohkawa, "Tetrahedron equation and Schur functions", Journal of Physics A: Mathematical and Theoretical 58 (2025) 015201, arXiv:2405.10011, DOI: 10.1088/1751-8121/ad9675, https://arxiv.org/abs/2405.10011
11. Myungbo Shim, Xiaoyue Sun, Hao Ellery Wang, Junya Yagi, "Solving the tetrahedron equation by Teichmüller TQFT", Journal of High Energy Physics 06 (2026) 141, arXiv:2602.22060, DOI: 10.1007/JHEP06(2026)141, https://doi.org/10.1007/JHEP06(2026)141