General Framework for Twisted Bilayer Photonic Crystal with Interlayer Coupling and Far-Field Response
Shupeng Xu, Dun Wang, Ritesh Agarwal · 2025
Read General Framework for Twisted Bilayer Photonic Crystal with Interlayer Coupling and Far-Field Response by Shupeng Xu, Dun Wang, Ritesh Agarwal (2025) 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-2509-23952.txt.
Introduction
Twisted bilayer photonic crystals promise flat bands and sharp resonances, but they need a theory that covers both near-field coupling and far-field response. Shupeng Xu and 2 co-authors build that general theory on a perturbation framework for non-Hermitian Hamiltonians. A universal form for interlayer coupling emerges, tied to Fourier transforms of the single-layer Wannier functions. At the K point of hexagonal lattices, the coupling reduces to the Bistritzer-MacDonald form known from graphene.
For a honeycomb slab with air holes, the theory predicts a fourfold band splitting in the far field, confirmed by simulation. The authors also propose a wide-angle, high-Q flat band cavity design.
Catalog record
| CREATOR | Shupeng Xu, Dun Wang, Ritesh Agarwal |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 28 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2509-23952
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
General Framework for Twisted Bilayer Photonic Crystal with Interlayer Coupling and Far-Field Response Preprint: APS/123-QED Shupeng Xu, Dun Wang, Ritesh Agarwal Abstract We develop a general theory for twisted bilayer photonic crystals that takes into account both far-field response and near-field coupling. The theory is based on the framework of a generalized Rayleigh-Schrödinger perturbation theory for non-Hermitian Hamiltonians. A universal form for interlayer coupling is derived, which relates the hopping strength to the Fourier transforms of the Wannier functions in the single layer photonic crystal. For low energy states at the K point in hexagonal lattices, the interlayer coupling reduces to that in the Bistritzer-MacDonald model for graphene. As an example, we study a twisted bilayer photonic crystal slab with air holes arranged in a honeycomb lattice in each layer. The first order solution of our model predicts a four-fold band splitting in the far-field spectrum compared to the single-layer case, which is confirmed by numerical simulations. Moreover, our theory reveals that for low energy states at K points, scattering towards the Γ \Gamma point via the moiré potential is suppressed. Based on our theory, we propose a wide-angle, high-Q tunable flat band cavity by combining the bilayer at a large twist angle with a Brillouin-zone-folding perturbation within each layer. The cavity behaves like a collection of quasi-bound states in the continuum with a divergent density of states, with potential applications in nonlinear optics, lasing and quantum optics. In recent years, twisted moiré systems have garnered intense interest across a diverse range of research communities. When two layers with similar lattice structures are stacked with a twist angle, they form a moiré superlattice with enlarged periodicity. This large length scale modulation dramatically alters the band structure, enabling the emergence of flat bands. In twisted van der Waals materials, the kinetic energy of electrons is quenched, and many-body interactions dominate their physical properties. This leads to a plethora of exotic states such as correlated insulators, superconductors, topological insulators and observation of novel phenomena such as fractional quantum anomalous Hall and nonlinear Hall effects Cao et al. 2018a ; Cao et al. 2018b ; Park et al. 2023 ; Zeng et al. 2023 ; Kang et al. 2024 ; Ji et al. 2024a ; Ji et al. 2024b . In photonic systems, twisted bilayer photonic crystals (TBPCs) enable localized optical modes and tunable resonances, leading to moiré lasers Ouyang et al. 2024 ; Luan et al. 2023 ; Mao et al. 2021 , quasi-bound states in the continuum (quasi-BICs) Huang et al. 2022 , sensors Tang et al. 2025 and other novel devices Lou et al. 2024 ; Lou and Fan 2022 . Theoretical studies have been conducted to understand the properties of TBPCs, which can be broadly classified into two categories. The first category includes photonic analogs of the Bistritzer-MacDonald model Bistritzer and MacDonald 2011 originally developed for twisted bilayer graphene (TBG). These models Dong et al. 2021 along with experiments Huang et al. 2022 ; Mao et al. 2021 , can solve for band structures in the moiré Brillouin zone; however, they offer no insights into the far-field optical response. Moreover, these models are difficult to generalize to photonic structures that lack the analogy of tightly bound atomic orbitals, such as dielectric slabs with air hole arrays, where the hopping integral of orbitals is not defined. The second category, exemplified by Ref. Lou et al. 2021 , focuses on solving the far-field response, but neglects the near-field coupling between the two layers. For moiré photonic crystals, both far-field response and near-field coupling are critical. The former determines the optical functionalities relevant for many device applications, while the latter enables the flat band physics. A theoretical framework that integrates both effects is therefore needed to fully understand these systems. Here we present a unified theoretical framework that incorporates both far-field response and near-field coupling in TBPCs. We start with a toy model to gain some intuition about the far-field response. Consider an effective non-Hermitian Hamiltonian for a photonic crystal subjected to a perturbation with moiré periodicity: \mathcal{\hat{H}=\hat{H}}_{0}+\mathcal{\hat{V}}, (1) where \mathcal{\hat{H}}_{0} has unperturbed solutions \mathcal{\hat{H}}_{0}|\psi_{0,\mathbf{k}}\rangle=E_{0,\mathbf{k}}|\psi_{0,\mathbf{k}}\rangle , and \langle\psi_{0,\mathbf{k}}|\mathcal{\hat{V}}|\psi_{0,\mathbf{p}}\rangle=V_{\mathbf{kp}}\delta_{\mathbf{k,p+G}_{m}} obeys momentum conservation, in which \mathbf{G}_{m} is the moiré reciprocal lattice vector, and \mathbf{k} and \mathbf{p} are momenta in K space. Since exceptional points are generally rare in dielectric photonic crystals at general momenta, this allows for the following perturbative solution in Rayleigh-Schrödinger series Chen et al. 2025 ; Buth et al. 2004 |\psi_{\mathbf{k}}\rangle=|\psi_{0,\mathbf{k}}\rangle+\sum_{\mathbf{G}_{m}}\frac{V_{\mathbf{k,k+G}_{m}}|\psi_{0,\mathbf{k+G}_{m}}\rangle}{E_{0,\mathbf{k+G}_{m}}-E_{0,\mathbf{k}}}+\cdots. (2) A solution in this form enables us to keep track of the far-field response instead of folding every state into the moiré Brillouin zone. In the far field, the perturbative corrections radiate at momenta \mathbf{k+G}_{m} , which show up as resonant features in experiments. The net result is the emergence of satellite peaks due to the reduced translational symmetry. We note that, by appropriately choosing the unperturbed Hamiltonian \mathcal{H}_{0} , this result is reminiscent of the corresponding relation derived in Ref. Lou et al. 2021 . However, as we will show later, instead of the satellite peaks, the most prominent feature of TBPCs is a band splitting which arises from degenerate perturbation theory. Having analyzed the toy model for the far-field response under moiré perturbation, we now turn to the actual TBPC system. The system is governed by the following Hamiltonian acting on the magnetic field \mathbf{H(r)} : \mathcal{\hat{H}}=\nabla\times\left(\frac{1}{\epsilon(\mathbf{r})}\nabla\times\right), (3) where \epsilon(\mathbf{r}) is the permittivity of the system. The Hamiltonian can be partitioned into three parts \mathcal{\hat{H}}=\mathcal{\hat{H}}_{0}+\mathcal{\hat{V}}_{1}+\mathcal{\hat{V}}_{2}, (4) such that \mathcal{\hat{H}}_{0} accounts for the vacuum background, and \mathcal{\hat{H}}_{0}+\mathcal{\hat{V}}_{i} equals to the Hamiltonian of the i th layer. This theory takes the form of a tight binding model for two 2D potentials separated in the z-direction, in which the dominant term to consider is the interlayer coupling (see Supplemental Material for details): \displaystyle\begin{split}\langle\psi_{1}|\mathcal{\hat{H}}|\psi_{2}\rangle&\approx\langle\psi_{1}|\mathcal{\hat{V}}_{2}|\psi_{2}\rangle\\ &\propto\int\mathbf{E}_{1}^{*}(\mathbf{r})\Delta\epsilon_{2}(\mathbf{r})\mathbf{E}_{2}(\mathbf{r})d\mathbf{r},\end{split} (5) where |\psi_{i}\rangle and \mathbf{E}_{i}(\mathbf{r}) denote the eigenstate and the corresponding electric field distribution for the i th layer, respectively; \Delta\epsilon_{2}(\mathbf{r}) is the change in permittivity induced by the second layer. Eq. ( 5 ) takes the same form as the hopping integral in coupled mode theory Yariv 2003 , where \Delta\epsilon(\mathbf{r}) can be interpreted as an optical potential. Since the Hamiltonians for individual layers are periodic, their solutions are Bloch waves. This allows us to write the electric fields of the eigenstates in the first and second layer as |\psi_{\mathbf{k}}\rangle\big|_{\mathbf{E}}=\frac{1}{\sqrt{N}}\mathbf{u}(\mathbf{r})e^{i\mathbf{kr}} (6) and |\psi_{\mathbf{p}}\rangle\big|_{\mathbf{E}}=\frac{1}{\sqrt{N}}\mathbf{v}(\mathbf{r})e^{i\mathbf{pr}}, (7) respectively, where \mathbf{u(r)} and \mathbf{v(r)} are the periodic parts of the Bloch functions, and \mathbf{k} and \mathbf{p} are the in-plane Bloch vectors for each layer, with zero out-of-plane components. Plugging them into Eq. ( 5 ) and performing Fourier transform in the x-y plane, we have: \displaystyle\begin{split}\langle\psi_{\mathbf{k}}|\mathcal{\hat{V}}_{2}|\psi_{\mathbf{p}}\rangle&=\frac{1}{N}\int e^{i(\mathbf{p-k})\mathbf{r}}\mathbf{u}(\mathbf{r})^{*}\Delta\epsilon_{2}(\mathbf{r})\mathbf{v}(\mathbf{r})d\mathbf{r}\\ &=\Omega\sum_{\mathbf{G,G^{\prime}}}\delta_{\mathbf{k+G,p+G^{\prime}}}\int\tilde{\mathbf{u}}^{*}(\mathbf{G})\tilde{\mathbf{V}}(\mathbf{G^{\prime}})dz,\end{split} (8) where \Omega is the unit cell area, \mathbf{G} ( \mathbf{G^{\prime}} ) is the reciprocal lattice vector of the first (second) layer, \tilde{\mathbf{u}}(\mathbf{G}) is the Fourier transform of \mathbf{u}(\mathbf{r}) , and \tilde{\mathbf{V}}(\mathbf{G^{\prime}}) is the Fourier transform of \Delta\epsilon_{2}(\mathbf{r})\mathbf{v}(\mathbf{r}) . Eq. ( 8 ) is interpreted as follows: a state with momentum \mathbf{k} in the first layer can couple to a state with momentum \mathbf{p=k+G-G^{\prime}} in the second layer, with the coupling strength determined by the product \mathbf{u^{*}(G)V(G^{\prime})} . This selection rule in momentum space is the same as that in typical continuum