Quantum phase diagrams for bosons in hexagonal optical potentials: A continuous-space quantum Monte Carlo study
Danilo Nascimento Guimaraes, Laurent Sanchez-Palencia · 2026
Read Quantum phase diagrams for bosons in hexagonal optical potentials: A continuous-space quantum Monte Carlo study by Danilo Nascimento Guimaraes, Laurent Sanchez-Palencia (2026) 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-2605-06171.txt.
Introduction
Danilo Nascimento Guimaraes and Laurent Sanchez-Palencia map quantum phase diagrams for bosons in hexagonal optical lattices. Such lattices mimic graphene and hexagonal boron nitride, offering a clean platform for strongly correlated quantum matter. Text book models like Bose-Hubbard may miss real effects in these geometries, which is what the authors set out to check with exact methods.
Using continuous space exact diagonalization and quantum Monte Carlo, they study honeycomb and boron nitride type lattices. Even at strong lattice depth, the honeycomb case deviates from the standard model, with suppressed Mott lobes and no higher order insulating phases. Density assisted tunneling explains the difference. The boron nitride case shows a rich diagram with several Mott lobes at varied sublattice fillings.
Catalog record
| CREATOR | Danilo Nascimento Guimaraes, Laurent Sanchez-Palencia |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 60 KB |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.06171
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Quantum phase diagrams for bosons in hexagonal optical potentials: A continuous-space quantum Monte Carlo study Danilo Nascimento Guimaraes, Laurent Sanchez-Palencia Abstract Hexagonal optical lattices, emulating graphene and hexagonal boron nitride (h-BN) structures, provide a versatile platform for exploring strongly correlated quantum matter. Using continuous-space exact diagonalization and quantum Monte Carlo simulations, we investigate the phase diagrams of ultracold bosons in honeycomb and h-BN lattices. For the honeycomb lattice, we find significant deviations from the standard Bose-Hubbard model even for strong lattice amplitudes. We observe suppressed Mott insulator lobes and the absence of higher-order insulating phases, attributed to strong density-assisted tunneling effects. In the h-BN case, a rich phase diagram emerges, featuring multiple Mott lobes with various sublattice occupations, driven by the interplay of lattice asymmetry, interactions, and particle filling. Our results highlight the necessity of continuous-space treatments for capturing the full complexity of bosonic quantum phases in hexagonal geometries, paving the way for experimental realizations with ultracold atoms and further theoretical work. I Introduction Solid-state materials exhibit a remarkable diversity of geometric structures, arising from the interplay of atomic electronic configurations and complex interactions. Beyond conventional cubic lattices, these include triangular, checkerboard, and honeycomb lattices, as well as quasicrystals, and multilayer systems 33 ; 27 ; 11 ; 6 ; 2 . Such structures give rise to a wealth of nontrivial phenomena ranging from exotic superconductivity and anomalous quantum Hall effects to fractal behavior, localization, and topological phases, which are currently attracting considerable attention 21 ; 22 ; 31 ; 63 ; 49 ; 15 ; 24 . The challenge lies in unraveling the intricate interplay of matter-wave interference, electronic interactions, and geometric effects that underpin these phenomena. In parallel, ultracold atomic systems have emerged as a versatile platform for quantum simulation, offering unprecedented control over interaction strengths, sign, and range, as well as, lattice geometries and dynamical parameters 37 ; 8 ; 7 ; 20 ; 14 ; 60 ; *lsp2018 ; 40 ; 50 . By engineering laser configurations, exotic lattice structure, such as triangular, honeycomb, Lieb, and Kagome lattices, as well as super-lattices and quasiperiodic lattices, can be emulated 26 ; 67 ; 68 ; 4 ; 57 ; 59 ; 17 ; 58 ; 52 ; 29 ; 1 ; 62 ; 51 ; 25 ; 64 . These systems allows us to not only revisit the physics of strongly correlated electronic systems but also explore new physics of bosonic quantum matter in such structures. Moreover, recent advances, including multi-frequency lattices and dynamical phase control, further expand the range of accessible geometries 34 ; 10 ; 53 ; 41 . Hexagonal lattices, exemplified by graphene and certain carbon nanotubes 43 ; 11 , are of particular interest due to their unique band structures, which feature Dirac cones, flat bands, and exaggerated correlation effects, as well as their rich topological properties 19 ; 39 ; 23 ; 11 . Up to date only a few experiments have realized hexagonal lattices but they have already facilitated studies of exotic effects, including multi-orbital superconductivity 54 , antiferromagnetic Néel ordering 55 , and Dirac point physics 59 . It is also worth mentioning cavity polariton systems with nanopillars arrays arranged in honeycomb geometry 56 ; 48 . Theoretically, tight-binding models have been extensively employed to describe the single-particle band structure of honeycomb and gapped-honeycomb systems 65 ; 11 ; 3 ; 36 , while Bose–Hubbard models have been used to study superfluid–Mott transitions and superlattice-induced insulating phases 61 ; 38 ; 66 . Beyond, dynamical control of optical lattices has enabled the realization of hexagonal boron nitride (h-BN, aka gapped graphene) models 34 , with recent theoretical work revealing transitions between Mott insulating, checkerboard and superfluid phases at unit filling, tunable by the lattice gap 66 . In this work, we investigate the dynamics of ultracold bosons in hexagonal optical potentials, including honeycomb and h-BN lattices. Using continuous-space calculations, we determine the band structure of these models, establishing the validity limits of tight-binding approximations and extracting their parameters. Moreover, quantum Monte Carlo (QMC) simulations yield exact phase diagrams at vanishingly small temperatures and arbitrary fillings. For the honeycomb lattice, we find standard superfluid and Mott insulator phases. However, we observe significant deviations from Bose-Hubbard models, even at strong lattice amplitudes where band structures are very well reproduced. For instance, for a lattice depth V_{0}=15\,E_{\textrm{r}} , the lowest band is reproduced by the simplest tight-binding model with an accuracy of about 1% but the Bose–Hubbard approximation fails to describe but the first Mott lobe. These deviations are attributed to strong density-assisted tunneling effects beyond the Bose-Hubbard framework. For the h-BN lattice, our results unveil a rich phase diagram characterized by multiple Mott lobes with various fillings, hence extending the findings of Ref. 66 and highlighting the interplay of interactions, filling, and lattice geometry. Our work paves the way for experiments with ultracold atoms in hexagonal optical lattices and motivates further exploration of multilayer systems. II Model The dynamics of a 2D gas of identical spinless bosons with mass m and repulsive two-body interactions is governed by the Hamiltonian \hat{H}=\int d\mathbf{r}\,\hat{\Psi}^{\dagger}\left[\hat{H}_{0}+\frac{1}{2}\int d\mathbf{r}^{\prime}\,\hat{\Psi}^{\prime}\,{}^{\dagger}U(\mathbf{r}-\mathbf{r}^{\prime})\hat{\Psi}^{\prime}\right]\hat{\Psi}, (1) where \hat{\Psi} and \hat{\Psi}^{\prime} are, respectively, the field operators at positions \mathbf{r} and \mathbf{r}^{\prime} , \hat{H}_{0}=-{\hbar^{2}\nabla^{2}}/{2m}+V(\mathbf{r}) is the one-body Hamiltonian with potential V(\mathbf{r}) , and U(\mathbf{r}-\mathbf{r}^{\prime}) is the two-body interaction potential. We consider the honeycomb-like lattice potential V(\mathbf{r})=2V_{0}\sum_{j=1}^{3}\cos(\mathbf{G}_{j}\cdot\mathbf{r}+\phi_{j})\,, (2) where V_{0} is the potential amplitude, the \phi_{j} ’s are phases, and the \mathbf{G}_{j} ’s are vectors with size \pi/a and arranged to form angles of 2\pi/3 with each other. A possible implementation of such a potential with controlled parameters involves using multifrequency settings with three pairs of lasers 34 , see Fig. 1 (a). We then have \mathbf{G}_{j}=\mathbf{k}_{j}-\mathbf{k}^{\prime}_{j} , where \mathbf{k}_{j} and \mathbf{k}^{\prime}_{j} are the laser wavevectors with |\mathbf{k}_{j}|=|\mathbf{k}^{\prime}_{j}|=k , and \phi_{j} is the phase difference for beams in pair j . Since |\mathbf{G}_{j}|=\sqrt{3}\,k , the choice |\mathbf{G}_{j}|=\pi/a implies k=\pi/(\sqrt{3}a) . In what follows, we use the a=\pi/\sqrt{3}k as the length unit and recoil energy E_{\textrm{r}}=\frac{\hbar^{2}k^{2}}{2m}=\frac{\pi^{2}\hbar^{2}}{6ma^{2}}, (3) as the energy unit. Unless stated otherwise, the zero of energy is taken to be the ground-state energy of the corresponding single-particle Hamiltonian \hat{H}_{0} for the parameters (V_{0},\phi_{\textrm{g}}) under consideration. Note that we can restrict ourselves to V_{0}\geq 0 since changing the sign of V_{0} is equivalent to shifting each \phi_{j} into \phi_{j}+\pi . The choice of the spatial origin provides two independent degrees of freedom, allowing two of the three phases to be absorbed by a coordinate shift. As a result, the geometry of the potential is determined by a single phase. Without loss of generality, we may choose the so-called geometric phase \phi_{\textrm{g}}=\sum_{j}\phi_{j} 34 and take \phi_{1}=\phi_{2}=\phi_{3}=\phi_{\textrm{g}}/3 . Irrespective of the value of \phi_{\textrm{g}} , the potential V(\mathbf{r}) is periodic with Bravais vectors \mathbf{a}_{1}=(4a/\sqrt{3})\mathbf{e}_{x} and \mathbf{a}_{2}=(2a/\sqrt{3})(\mathbf{e}_{x}+\sqrt{3}\mathbf{e}_{y}) , with \mathbf{e}_{x},\mathbf{e}_{y} the canonical basis of \mathbb{R}^{2} , see Fig. 1 (b). For \phi_{\textrm{g}}=0 , the potential minima form a honeycomb lattice with identical A and B sites, see Fig. 1 (b). An increasing value of \phi_{\textrm{g}} progressively lifts the degeneracy between A and B sites, hence forming an hBN-like potential. For |\phi_{\textrm{g}}|>\pi/2 , the B points are no longer local minima while the A sites are still minima, hence forming a triangular