Open problems of the R-matrix bootstrap for the 2d O(N) bosonic model with a boundary

The numerical S-matrix bootstrap — mapping the space of two-to-two amplitudes allowed by analyticity, unitarity and crossing — was extended by Kruczenski and Murali to reflection matrices (R-matrices) of 1+1-dimensional O(N)-symmetric quantum field theories on a half-line: for a fixed bulk S-matrix, the allowed boundary R-matrices form an infinite-dimensional convex space whose two-dimensional sections display vertices at which integrable reflection matrices appear. The paper itself acknowledges five unresolved issues that jointly define what remains to be done for this boundary bootstrap: (i) the bulk S-matrix and boundary R-matrix have only been determined sequentially, the simultaneous (coupled) bootstrap being blocked by non-linear constraints; (ii) some vertices of the allowed domains match no known integrable boundary theory; (iii) no local quantum field theory is known to realize the periodic Yang-Baxter (pYB) bulk S-matrix and its reflection matrices; (iv) the dual of the extended-analyticity problem requires an ad hoc regularization bound and the conditions for unitarity saturation are not understood; (v) boundary bound states and supersymmetric extensions are untreated.

Background

The analytic S-matrix bootstrap determines or constrains scattering amplitudes nonperturbatively, using only general principles — analyticity, unitarity, crossing symmetry and global symmetries — rather than a Lagrangian. In 1+1 dimensions its modern numerical form carves out the convex space of allowed two-to-two amplitudes by maximizing linear functionals, and integrable theories (theories whose multi-particle scattering factorizes into two-particle amplitudes) frequently appear at vertices — corner points — of the allowed regions. For the two-dimensional $O(N)$ bosonic model, whose exact factorized S-matrix was found by Zamolodchikov and Zamolodchikov, bootstrap studies by He, Irrgang and Kruczenski, by Córdova and Vieira, and by Córdova, He, Kruczenski and Vieira showed that the $O(N)$ nonlinear sigma model (NLSM) sits at a vertex of the space of allowed amplitudes, and rediscovered a second integrable solution: the periodic Yang-Baxter (pYB) S-matrix, a unitary, crossing-symmetric S-matrix periodic in the rapidity variable $\theta$ (defined by $s=4m^2\cosh^2(\theta/2)$, with $s$ the Mandelstam invariant), which first appeared in work of Hortaçsu, Schroer and Thun on a two-dimensional sigma model with particle production. Whether any local quantum field theory realizes this S-matrix has been unknown since.

A quantum field theory on a half-line is described, in its elastic sector, by a reflection matrix (R-matrix) $R_a^b(\theta)$: the amplitude for an incoming particle of rapidity $\theta$ and flavor $a$ to bounce off the boundary and emerge as flavor $b$. Beyond the unitarity-type bounds $|R_a^b(\theta)|\le1$ on the real rapidity axis (which allow particle production at the boundary) and analyticity in the physical strip $0\le\Im\theta\le\pi/2$, a double Wick rotation implies the boundary crossing relation of Ghoshal and Zamolodchikov, $R_a^b(i\pi/2-\theta)=S_{cd}^{ab}(2\theta)\,R_d^c(i\pi/2+\theta)$, which ties the R-matrix to the bulk S-matrix. Integrability of the boundary theory requires in addition the reflection (boundary) Yang-Baxter equation, a consistency condition on the three-particle reflection process. Exact integrable R-matrices for the $O(N)$ sigma model, corresponding to Dirichlet and Neumann boundary conditions, were found by Ghoshal.

Kruczenski and Murali combined these ingredients into the R-matrix bootstrap: fixing a bulk S-matrix (NLSM or pYB) and imposing only analyticity, boundary unitarity and boundary crossing — no integrability assumed — they mapped two-dimensional sections of the infinite-dimensional convex space of allowed R-matrices, for boundaries breaking $O(N)$ to $O(k)\times O(N-k)$. Integrable R-matrices without free parameters appeared at vertices of the carved domains; one-parameter families of integrable R-matrices occupied entire boundary segments. Requiring extended analyticity, i.e. analyticity in a region larger than the physical strip, which forbids resonance poles there, drastically shrank the allowed space and produced new vertices, again at integrable theories. They also derived the dual minimization problem; for extended analyticity the dual is well defined only after imposing a regularization bound $RR^\dagger\preceq M$ along interior curves of the enlarged analyticity domain, and saturation of boundary unitarity, $R(\theta)R(-\theta)=\mathbf{1}$, no longer follows automatically. Finally, solving the boundary Yang-Baxter equation for the pYB bulk they derived a previously unknown analytic reflection matrix.

The paper explicitly lists what remains undone: the simultaneous (coupled) bootstrap of bulk S-matrix and boundary R-matrix; the interpretation of vertices that match no known integrable boundary theory; the construction of a physical model for the pYB theory; the control of the regularization parameter $M$ and of unitarity saturation in the extended dual; and the extension to boundary bound states and to supersymmetric or other internal structures.

Problem Statement

The problem is to complete the boundary R-matrix bootstrap program that Kruczenski and Murali initiated for the 2d $O(N)$ bosonic model with a boundary: to make it a self-contained method that determines bulk and boundary scattering together, rests on a well-understood dual formulation, covers theories with boundary bound states and richer internal structures, and attaches physical interpretations to its outputs. The source paper's acknowledged open issues, which jointly define this completion, are:

(1) Coupled bootstrap: formulate and solve the simultaneous determination of the bulk S-matrix and the boundary R-matrix from the combined constraints — bulk and boundary analyticity, bulk and boundary unitarity, bulk crossing and boundary crossing — instead of fixing the bulk S-matrix beforehand; the joint constraints are non-linear and no method is currently known.

(2) Unexplained vertices: determine the physical or integrable interpretation of the vertices of the numerically carved allowed R-matrix domains that do not correspond to any known integrable boundary theory (for instance, the unidentified vertex in the diagonal NLSM case at $N=6$, $k=1$), or establish that no such interpretation exists; resolving any single such vertex counts as a partial answer.

(3) pYB realization: construct a local quantum field theory whose factorized scattering realizes the periodic Yang-Baxter bulk S-matrix together with its derived reflection matrices, or rule out the existence of such a model.

(4) Extended-analyticity dual: clarify the role of the regularization bound $RR^\dagger\preceq M$ imposed on interior curves of the enlarged analyticity domain — in particular whether and how the results depend on $M$ and whether the bound can be removed — and establish the precise conditions under which boundary unitarity is or is not saturated on the boundary of the allowed region in the extended problem.

(5) Extensions: extend the R-matrix bootstrap to boundary bound states (poles of the R-matrix on the imaginary rapidity axis) and to supersymmetric or other internal structures.

These components are scientifically linked: (1), (4) and (5) complete the bootstrap framework itself, while (2) and (3) demand the physical interpretation of the framework's outputs for the $O(N)$ models; a full solution addresses all five. Partial answers are admissible.

References

1. Alexander B. Zamolodchikov and Alexey B. Zamolodchikov, 'Relativistic Factorized S Matrix in Two-Dimensions Having O(N) Isotopic Symmetry', Nucl. Phys. B 133 (1978) 525, DOI 10.1016/0550-3213(78)90239-0, https://doi.org/10.1016/0550-3213(78)90239-0
2. Subir Ghoshal and Alexander B. Zamolodchikov, 'Boundary S Matrix and Boundary State in Two-Dimensional Integrable Quantum Field Theory', Int. J. Mod. Phys. A 9 (1994) 3841, arXiv:hep-th/9306002, DOI 10.1142/S0217751X94001552, https://doi.org/10.1142/S0217751X94001552
3. Subir Ghoshal, 'Boundary S-matrix of the O(N)-symmetric non-linear sigma model', Phys. Lett. B 334 (1994) 363, DOI 10.1016/0370-2693(94)90701-3, https://doi.org/10.1016/0370-2693(94)90701-3
4. M. Hortaçsu, B. Schroer and H. J. Thun, 'A Two-dimensional σ Model With Particle Production', Nucl. Phys. B 154 (1979) 120, DOI 10.1016/0550-3213(79)90374-2, https://doi.org/10.1016/0550-3213(79)90374-2
5. Yifei He, Andrew Irrgang and Martin Kruczenski, 'A note on the S-matrix bootstrap for the 2d O(N) bosonic model', JHEP 11 (2018) 093, arXiv:1805.02812, DOI 10.1007/JHEP11(2018)093, https://arxiv.org/abs/1805.02812
6. Lucía Córdova and Pedro Vieira, 'Adding flavour to the S-matrix bootstrap', JHEP 12 (2018) 063, arXiv:1805.11143, DOI 10.1007/JHEP12(2018)063, https://arxiv.org/abs/1805.11143
7. Lucía Córdova, Yifei He, Martin Kruczenski and Pedro Vieira, 'The O(N) S-matrix Monolith', JHEP 04 (2020) 142, arXiv:1909.06495, DOI 10.1007/JHEP04(2020)142, https://arxiv.org/abs/1909.06495
8. Martin Kruczenski and Harish Murali, 'The R-matrix bootstrap for the 2d O(N) bosonic model with a boundary', JHEP 04 (2021) 097, arXiv:2012.15576, DOI 10.1007/JHEP04(2021)097, https://arxiv.org/abs/2012.15576
9. Yifei He and Martin Kruczenski, 'S-matrix bootstrap in 3+1 dimensions: regularization and dual convex problem', JHEP 08 (2021) 125, arXiv:2103.11484, DOI 10.1007/JHEP08(2021)125, https://arxiv.org/abs/2103.11484
10. Christian Copetti, Lucía Córdova and Shota Komatsu, 'S-matrix bootstrap and non-invertible symmetries', JHEP 03 (2025) 204, arXiv:2408.13132, DOI 10.1007/JHEP03(2025)204, https://arxiv.org/abs/2408.13132
11. Lucía Córdova and Ricardo Rodrigues, 'On the space of U(N) scattering amplitudes', JHEP 09 (2025) 076, arXiv:2504.20239, DOI 10.1007/JHEP09(2025)076, https://arxiv.org/abs/2504.20239
12. Zhao Zhang, 'Bootstrapping the R-matrix', SciPost Phys. 20 (2026) 102, arXiv:2504.17773, DOI 10.21468/SciPostPhys.20.4.102, https://arxiv.org/abs/2504.17773
13. Soichiro Shimamori and Satoshi Yamaguchi, 'Boundary scattering and non-invertible symmetries in 1+1 dimensions', JHEP 02 (2026) 088, arXiv:2504.08375, DOI 10.1007/JHEP02(2026)088, https://arxiv.org/abs/2504.08375
14. Jan Albert and Alexandre Homrich, 'Imprints of Asymptotic Freedom on Confining Strings', Phys. Rev. Lett. 137 (2026) 041601, arXiv:2602.15097, DOI 10.1103/6399-gw3k, https://doi.org/10.1103/6399-gw3k