betaFeedback

WorksPhysics and Astronomy

Moiré trapping of quadrupolar excitons in van der Waals trilayers

Abstract

Quadrupolar excitons in van der Waals heterostructures - quantum superpositions of anti-aligned dipolar excitons - offer a novel platform to explore exotic many-body physics, with applications to quantum sensing and photonic devices. Yet their internal structure, symmetry, and real-space localisation remain largely unknown. Here, we reveal the atomic-scale structure of quadrupolar excitons in twisted WSe2/WS2/WSe2 trilayers by solving the Bethe-Salpeter equation within a large-scale atomistic framework. We discover that large atomic relaxations at small twist angles give rise to two distinct quadrupolar excitons trapped at moiré lattice sites, differing in the in-plane symmetry of the electron density about the hole: one azimuthally symmetric, with the density maximal at the hole, and one threefold symmetric, with a node at the hole. Moiré trapping, neglected in commonly used models of quadrupolar exciton formation, is critical to their many-exciton phases. Without moiré trapping, quadrupolar excitons transition into anti-parallel dipolar excitons on a bipartite square lattice, while with trapping, the same dipoles are confined to a triangular lattice and experience geometric frustration. Our study uncovers the highly non-trivial nature of quadrupolar excitons, with direct implications for simulating frustrated quantum magnetism in a fully tunable excitonic platform.

Introduction

Introduction. Multipole expansions encode the symmetry of a source through its moments, with applications spanning molecular electrostatics[1], gravitational radiation[2], and many-body physics[3, 4]. For example, in electrostatics, dipoles arise from separated positive and negative charges, while quadrupoles appear in more symmetric arrangements where both the total charge and dipole moment vanish. Recently, this framework has found a striking realization in van der Waals heterostructures[5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18]. By stacking and twisting ultrathin two-dimensional (2D) materials, experimentalists have demonstrated tunable dipole moments of excitons, which are bound electron–hole pairs. For example, excitons in symmetric trilayer transition metal dichalcogenides such as WSe2/WS2/WSe2\mathrm{WSe}_{2}/\mathrm{WS}_{2}/\mathrm{WSe}_{2}[5, 6] and WS2/WSe2/WS2\mathrm{WS}_{2}/\mathrm{WSe}_{2}/\mathrm{WS}_{2}[7] show a transition from quadrupolar to dipolar character when an out-of-plane electric field is applied. The absence of a dipole moment leads to qualitatively distinct interaction physics, making these systems an attractive platform for exploring exotic many-body phases[19, 20, 21] and optoelectronic applications[22, 23]. To date, the dipolar and quadrupolar characters have been inferred almost exclusively from the electric-field dependence of photoluminescence (PL), which shifts from linear to non-linear behavior[24]. Direct real-space imaging of their structure, symmetry, and twist-angle dependence remains elusive.

Accurate theory is essential to close this gap, provide atomistic insight, and guide future experiments. Yet, theoretical modeling remains scarce. This scarcity is mainly due to the prohibitive computational cost of state-of-the-art many-body perturbation theory methods, such as the GW-plus-Bethe-Salpeter equation (GW-BSE) approach, for simulating multilayer 2D materials with hundreds to thousands of atoms[25, 26, 27, 28]. As a result, GW–BSE calculations have been limited to lattice-matched trilayer systems containing only a few atoms[13, 29, 30, 31, 32]. However, stacking and twisting 2D materials drive pronounced atomic rearrangements that qualitatively alter exciton behavior. Such effects are entirely absent in the lattice-matched models. For example, in WS2/WSe2\mathrm{WS}_{2}/\mathrm{WSe}_{2} bilayers, atomic rearrangements give rise to localized intralayer excitons in WSe2\mathrm{WSe}_{2} with strong electron-hole overlap and charge-transfer character[33, 34, 35, 36, 37, 38]. Structural relaxations are therefore essential for any predictive description of multipolar excitons and their field-tunable moments. Furthermore, accurate parameters characterizing individual quadrupolar and dipolar excitons are essential input for modeling the many-exciton phase diagram[30, 39]. Their importance is evident from a striking anomaly[5]: in the exciton density regime where experiments observe quadrupolar excitons, models that neglect structural relaxations instead predict a staggered dipolar phase.

Here, we overcome these challenges and provide the first atomistic description of the electric-field-driven quadrupolar-to-dipolar transition in twisted trilayer WSe2/WS2/WSe2\mathrm{WSe}_2/\mathrm{WS}_2/\mathrm{WSe}_2, fully accounting for structural relaxations. To this end, we develop a transfer-matrix formalism for electron–hole Coulomb interactions in arbitrarily twisted multilayer 2D materials and solve the fully atomistic BSE using ab initio electronic structure as input to determine the exciton spectrum and its real-space structure. We demonstrate that quadrupolar excitons form in WSe2/WS2/WSe2\mathrm{WSe}_2/\mathrm{WS}_2/\mathrm{WSe}_2 trilayers with parallel outer WSe2\mathrm{WSe}_2 layers, regardless of the twist angle with WS2\mathrm{WS}_2. At small twist angles, large atomic rearrangements trap quadrupolar excitons at moiré sites, splitting them into two species that differ in the in-plane symmetry of the electron density about the hole: one azimuthally symmetric with maximal density at the hole, and one threefold symmetric with a node at the hole. Moiré trapping resolves the anomaly[5]. By confining excitons to a triangular lattice, it induces geometric frustration that stabilises the quadrupolar phase across the experimentally observed density range.

Bethe-Salpeter equation in multilayers. We construct a fully atomistic BSE Hamiltonian using Wannier functions obtained from large-scale density functional theory (DFT) calculations of arbitrarily twisted multilayers[40].

⟨cvk∣HBSE∣c′v′k′⟩=(ϵck−ϵvk)δcc′δvv′δkk′−1N∑R,n1,n2(Cn1ck)∗Cn1c′kCn2vk(Cn2v′k′)∗W(R+(tn2−tn1))ei(k−k′)⋅R+1N∑R,n1,n2(Cn1ck)∗Cn1vkCn2c′k′(Cn2v′k′)∗V(R+(tn2−tn1)).(1)\begin{aligned} \langle cvk|\mathcal{H}_{\mathrm{BSE}}|c'v'k'\rangle ={}&(\epsilon_{ck}-\epsilon_{vk})\delta_{cc'}\delta_{vv'}\delta_{kk'}\\ &-\frac{1}{N}\sum_{\mathbf{R},n_1,n_2} (C_{n_1c}^{k})^{*}C_{n_1c'}^{k}C_{n_2v}^{k}(C_{n_2v'}^{k'})^{*} W\left(\mathbf{R}+(t_{n_2}-t_{n_1})\right)e^{i(\mathbf{k}-\mathbf{k}')\cdot\mathbf{R}}\\ &+\frac{1}{N}\sum_{\mathbf{R},n_1,n_2} (C_{n_1c}^{k})^{*}C_{n_1v}^{k}C_{n_2c'}^{k'}(C_{n_2v'}^{k'})^{*} V\left(\mathbf{R}+(t_{n_2}-t_{n_1})\right). \tag*{(1)} \end{aligned}

Here, ϵc(v)k\epsilon_{c(v)k} are the quasiparticle energies of conduction (valence) states, CnmkC_{nm}^{k} are the Wannier function coefficients, tnt_n is the position of the nn-th Wannier function in the home unit cell, R\mathbf{R} are lattice vectors, NN is the number of kk-points sampled in the Brillouin zone, and WW and VV are the screened and bare Coulomb interactions. Details of the derivation of Eqn. (1) are detailed elsewhere[40]. The key innovation is that WW is obtained from the multilayer electrostatic problem via a transfer-matrix formalism. This contrasts with previous approaches to excitons in multilayers that also use classical electrostatics for screening but were limited to dipolar interlayer excitons in lattice-matched bilayers and treated within effective-mass frameworks[41, 42, 43, 44, 45].

To illustrate the transfer-matrix method, we consider two 2D layers with arbitrary stacking (Figure 1(a)). We first construct the screened Coulomb potential in momentum space. ϕij(q)\phi_{ij}(q) in layer jj due to an electron in layer ii. Along the out-of-plane direction zz, we express the potential in each region as a sum of exponentials e±qze^{\pm qz} and enforce boundary conditions at each interface by combining propagation (PmP_m), layer (LmL_m), and source (SS) matrices. Here, Pm(d)P_m(d) propagates the potential across the spacing dd between the mm-th and (m+1)(m+1)-th layers, Lm(χm,ϵ1,ϵ2)L_m(\chi_m,\epsilon_1,\epsilon_2) describes the response of the mm-th layer with polarizability χm\chi_m between dielectrics ϵ1\epsilon_1 and ϵ2\epsilon_2, and SS is the electron source (see Figure 1(a)). The transfer matrix de-

Image not extracted

Figure 1. Structure and out-of-plane electric-field response of dipolar and quadrupolar excitons. (a), (b): Schematics of dipolar and quadrupolar excitons in WSe2_2/WS2_2 and WSe2_2/WS2_2/WSe2_2 heterostructures sandwiched between substrates. The top and bottom electrostatic potentials, ϕt\phi_t and ϕb\phi_b, are related by the layer (LiL_i), propagation (PiP_i), and electron source matrices. (c),(e): Dipolar excitons exhibit a linear Stark effect, whereas (d),(f) quadrupolar excitons transition from a quadratic to a linear Stark response as the electric field strength increases. The colorbar indicates both optical conductivity and photoluminescence. Optical conductivities are from lattice-matched structures. Experimental data adapted with permission from Yu et al.``{sup}5` (Copyright 2023 Springer Nature).

termines the top potential from the bottom one, ϕi=1,t(q)=L2P1L1[ϕi=1,b+S](q)\phi_{i=1,t}(q) = L_2P_1L_1\left[\phi_{i=1,b} + S\right](q). From this we propagate through the layers to find ϕi=1,j(q)\phi_{i=1,j}(q). We repeat this for the second layer, determining ϕi=2,t(q)=[L2P1L1ϕi=2,b+S](q)\phi_{i=2,t}(q) = \left[L_2P_1L_1\phi_{i=2,b} + S\right](q) and propagating to find ϕi=2,j(q)\phi_{i=2,j}(q). The same approach extends naturally to the three-layer system (Figure 1(b)),

ϕi=1,t(q)=L3P2L2P1[L1ϕi=1,b+S](q),(2)\phi_{i=1,t}(q) = L_3P_2L_2P_1\left[L_1\phi_{i=1,b} + S\right](q), \tag*{(2)}
ϕi=2,t(q)=L3P2[L2P1L1ϕi=2,b+S](q),(3)\phi_{i=2,t}(q) = L_3P_2\left[L_2P_1L_1\phi_{i=2,b} + S\right](q), \tag*{(3)}
ϕi=3,t(q)=[L3P2L2P1L1ϕi=3,b+S](q).(4)\phi_{i=3,t}(q) = \left[L_3P_2L_2P_1L_1\phi_{i=3,b} + S\right](q). \tag*{(4)}

See Sec. A of the Supplementary Information (SI) for details and generalization to arbitrary multilayers. We use χWS2=75.73\chi_{\mathrm{WS}_2} = 75.73 Å and χWSe2=90.18\chi_{\mathrm{WSe}_2} = 90.18 Å obtained from DFT[46]. Finally, we apply a Hankel transform to map ϕij(q)\phi_{ij}(q) to the real-space WW entering the BSE Hamiltonian, Eq. (1).

Quadrupolar excitons in lattice-matched trilayers. WS2_2 and WSe2_2 have a lattice mismatch of 4%, producing a moiré pattern with about 8 nm lattice constant even in the absence of twist[5, 21, 34, 37]. To disentangle the moiré-related effects, we first construct lattice-matched WSe2_2/WS2_2 bilayers and WSe2_2/WS2_2/WSe2_2 trilayers with a lattice constant of 0.325 nm. In both systems, WSe2_2 and WS2_2 are antiparallel (θ=60∘\theta= 60^\circ). In the trilayer, we keep the two outer WSe2_2 layers parallel (AA stacking, θ=0∘\theta= 0^\circ), as DFT calculations showed this alignment to be necessary for quadrupolar exciton formation[7]. Both systems have type-II band alignment, with the valence band maximum in WSe2\mathrm{WSe}_{2} and the conduction band minimum in WS2\mathrm{WS}_{2}, and a direct band gap at the KK point. Figure 1(c),(Figure 1d) compare the calculated optical conductivity of the two systems under an applied out-of-plane electric field, EzE_z. To place all results on a common energy reference, the lowest-energy bright excitons are aligned to 1.4 eV at zero field using a rigid scissor shift applied throughout the paper. The bilayer 1s1s interlayer exciton shows a linear Stark effect over the entire field range, consistent with a permanent dipole. In contrast, the trilayer 1s1s interlayer exciton shows a quadratic response up to Ez=±50E_z = \pm50 mV/nm, crossing over to a linear regime at higher fields. This quadratic-to-linear crossover is the hallmark of a field-induced transition from quadrupolar to dipolar excitons[24]. Figure 1(c),(Figure 1d) also show higher-energy exciton Rydberg states up to the 2s2s level. At zero field, the 2s2s–1s1s energy splitting decreases from 82 meV in the bilayer to 68 meV in the trilayer, reflecting enhanced screening in the trilayer. We observe Rydberg excitons beyond the 2s2s state, but the calculations used here are not sufficiently converged

to accurately determine their energies and oscillator strengths (see SI, Sec. B). Figure 1(e),(Figure 1f) show measured PL spectra for structures where WSe2\mathrm{WSe}_{2} and WS2\mathrm{WS}_{2} form a 8 nm moiré pattern and the outer WSe2\mathrm{WSe}_{2} layers are nearly aligned[5]. Our calculations reproduce the experimentally observed lowest bright excitons in PL, validating our approach. We now turn to moiré systems accounting for the ≈4%\approx4\% lattice mismatch between WS2\mathrm{WS}_{2} and WSe2\mathrm{WSe}_{2}.

Twist-angle dependence of quadrupolar excitons. Figure 2 shows the twist-angle dependence of the lowest-energy excitons in WSe2/WS2/WSe2\mathrm{WSe}_{2}/\mathrm{WS}_{2}/\mathrm{WSe}_{2} under different out-of-plane electric fields. The two WSe2\mathrm{WSe}_{2} layers remain aligned, while the middle WS2\mathrm{WS}_{2} layer is twisted by an angle θ\theta relative to the bottom WSe2\mathrm{WSe}_{2} layer. Regardless of twist angle, the lowest bright exciton exhibits a quadratic-to-linear Stark response as ∣Ez∣\lvert E_z\rvert increases from zero. A zoom-in of the lowest excitons near zero field highlighting the quadratic region is also shown in Figure 4(b). At large twist angles (see Figure 2(a)), the peak intensities are weak, but they increase as the twist angle decreases. At small twist angles (see Figure 2(c),(Figure 2d)), we find two quadrupolar excitons, labeled I and II, with exciton II lying within 50 meV of exciton I. These peaks are fundamentally different from the 1s state in the lattice-matched system without a moiré pattern. They correspond to moiré-trapped quadrupolar excitons and are the main result of this work.

Image not extracted

Figure 2. Out-of-plane electric-field response of excitons in twisted WSe2_2/WS2_2/WSe2_2. (a–d) As the electric field increases from zero, the lowest bright exciton changes from a quadratic to a linear response, indicating a transition from quadrupolar to dipolar behaviour. Peak I and II in (c) and (d) are due to moiré trapping of quadrupolar excitons. The colorbar indicates the optical conductivities.

Image not extracted

Figure 4. Tunable dipole moments and avoided crossings in twisted WSe2_2/WS2_2/WSe2_2. Field-dependent (a) dipole moments of the lower branch and (b) avoided-crossing gaps between the upper and lower branches for several twist angles. The center of the gap is aligned across all twist angles. Hexagon size indicates the oscillator strength of the excitons in (b). Experimental data adapted with permission from Yu et al[5] (Copyright 2023 Springer Nature).

To demonstrate moiré trapping, we analyze the exciton wavefunctions associated with excitons I and II at zero electric field. Figure 3 shows the electron distributions at a 2.2° twist with the hole fixed at different stacking regions. For both excitons, the electron is located in WS2\mathrm{WS}_{2} and the holes are in the WSe2\mathrm{WSe}_{2} layers. As shown in Figure 3(a), exciton I has the electron and hole vertically aligned with the electron showing circular symmetry. The electron density is strongest in the BSe/W/Se\mathrm{B}^{\mathrm{Se}/\mathrm{W}/\mathrm{Se}} stacking region, corresponding to Bernal stacking with Se atoms of both WSe2\mathrm{WSe}_{2} layers directly above W atoms of WS2\mathrm{WS}_{2}. In contrast, Figure 3(b) shows exciton II has the electron distributed across three lobes at the vertices of a triangle, displaced laterally from the hole position. The electron density is strongest in the AAA stacking configuration, where the W and Se atoms in WSe2\mathrm{WSe}_{2} are directly on top of the W and S atoms in WS2\mathrm{WS}_{2}, respectively. Crucially, exciton II has reduced rotational symmetry, C3C_3, generating an in-plane octupolar moment absent in exciton I (see SI, Sec. C for analysis). The exciton trapping in different stacking regions, absent in lattice-matched trilayers and at large twist angles, arises from moiré-induced atomic rearrangements. In the energetically unfavorable AAA stacking region, atoms move anticlockwise in WSe2\mathrm{WSe}_{2} and clockwise in WS2_2 to reach a more stable configuration (see SI, Sec. D for more details). The symmetric WSe2_2 layers on either side of WS2_2 amplify these rearrangements in WS2_2. This produces a much stronger confining potential for electrons than in a twisted bilayer system, consistent with previous experiments[23]. As a result, quadrupolar excitons become trapped at approximately below 3∘3^\circ twist, corresponding to moiré lengths ≳5\gtrsim5 nm.

Image not extracted

Figure 3. Moiré quadrupolar exciton wavefunctions in a 2.2∘2.2^\circ twisted WSe2/WS2/WSe2\mathrm{WSe}_{2}/\mathrm{WS}_{2}/\mathrm{WSe}_{2}. (a) Exciton I shows azimuthally symmetric electron density and is trapped at the BSe/W/Se\mathrm{B}^{\mathrm{Se/W/Se}} stacking region. (b) Exciton II shows threefold symmetric electron density, laterally displaced from the hole location and largely trapped at the AAA stacking region. Blue dots indicate the hole positions, while solid lines show the moiré unit cell. The colorbar indicates the electron densities.

Two-dipole avoided crossing model. To quantify the electric field-driven quadrupole-to-dipole transition, we employ an avoided crossing model. In this model, two degenerate dipolar excitons with energy Ed\mathcal{E}_d and dipole moments ±p0\pm p_0 are coupled via a quantum tunneling strength tt under an applied electric field EzE_z. The system is described by the Hamiltonian[5, 6, 7, 9, 29]

\begin{pmatrix} \mathcal{E}_d+p_0E_z & -t \\ -t & \mathcal{E}_d-p_0E_z \end{pmatrix}. \tag*{(5)}\end{pmatrix}

Eqn. (5) has two eigenvalues, Eq±=Ed±p02Ez2+t2\mathcal{E}_q^{\pm}=\mathcal{E}_d\pm\sqrt{p_0^2E_z^2+t^2}, which correspond to two quadrupolar exciton branches. Since experiments have primarily observed the lower-energy branch[7], we extract p0p_0 and tt by fitting the field-dependent dipole moment p(Ez)p(E_z) from our BSE calculations. This branch corresponds to the highlighted 1s1s state at large twist angles and lattice-matched systems, and to peak I at small twist angles (Figure 2). We fit p(Ez)p(E_z) using

p(Ez)=−dEq−dEz=p02Ez(p0Ez)2+t2.(6)p(E_z)=-\frac{d\mathcal{E}_q^-}{dE_z}=\frac{p_0^2E_z}{\sqrt{(p_0E_z)^2+t^2}}. \tag*{(6)}

At strong fields (p0Ez≫tp_0E_z\gg t), the dipole saturates to ±p0\pm p_0, while at weak fields (p0Ez≪tp_0E_z\ll t), it vanishes, recovering the quadrupolar character. Figure 4(a) compares our calculated p(Ez)p(E_z) for all moiré structures and the lattice-matched trilayer with experiment[5], showing qualitative agreement. Table II lists the extracted parameters. The tunneling strength ∣t∣\lvert t\rvert increases monotonically as the twist angle decreases from 16.1∘16.1^\circ to 2.2∘2.2^\circ. The splitting (Eq+−Eq−)(\mathcal{E}_q^+-\mathcal{E}_q^-) in Table II, obtained from the BSE calculations, is larger than the model prediction of 2∣t∣2\lvert t\rvert, pointing to corrections beyond the simple two-level model.

Lattice constant (nm)Number of atomsNumber of Wannier functions
0.325 (0∘0^\circ)933
1.15 (16.1∘16.1^\circ)111407
3.49 (4.7∘4.7^\circ)10293773
4.89 (2.8∘2.8^\circ)20137381
5.59 (2.2∘2.2^\circ)26259625
0.325 (60∘60^\circ)933

Table II. Summary of lattice constants, number of atoms, and number of Wannier functions used in the electronic structure calculations for the BSE.

Another key distinguishing factor between quadrupolar and dipolar excitons is the field-dependent PL radiative rate, γrad\gamma^{\mathrm{rad}}[5, 7]. WWe capture this behavior using Fermi’s golden rule[47] together with the computed oscillator strengths, γdiprad/γquadrad≈PEz=20 meV/nm/PEz=0 meV/nm\gamma_{\mathrm{dip}}^{\mathrm{rad}}/\gamma_{\mathrm{quad}}^{\mathrm{rad}}\approx\mathcal{P}_{E_z=20\ \mathrm{meV/nm}}/\mathcal{P}_{E_z=0\ \mathrm{meV/nm}}. Figure 4(b) shows the oscillator strengths of both exciton branches, defined as P=∣∑cvkAcvks⟨vk∣px∣ck⟩∣2\mathcal{P}=\left|\sum_{cvk}A_{cvk}^{s}\langle vk|p_x|ck\rangle\right|^2. At 60∘60^\circ, P\mathcal{P} is highest for the lower branch, while at 0∘0^\circ it is highest for the upper branch. The bright quadrupolar exciton near zero EzE_z has a radiative rate γrad\gamma^{\mathrm{rad}} 1.5–1.8 times higher than the dipolar branch at high EzE_z, while the dark branch is 4–8×103\times10^{3} times lower, depending on twist angle. Our results agree well with the experimentally reported 1.7-fold faster radiative decay of the bright quadrupolar exciton branch[7].

Many-exciton phases at zero TT. To demonstrate the impact of moiré trapping, we map the zero-temperature phase diagram of interacting quadrupolar excitons, both free and trapped. We model excitons as point-like bosons and neglect their kinetic energy[30]. Building on the avoided crossing model, the phases are governed by three competing energy scales: electrostatic interactions, the energy cost (Eq+−Eq−)/2(\mathcal{E}_{q}^{+}-\mathcal{E}_{q}^{-})/2 of occupying the dipolar instead of the quadrupolar state, and the moiré trapping energy VMV_{\mathrm{M}}. Each dipolar exciton carries an Ising degree of freedom σi=±1\sigma_i=\pm1 (up or down), which selects the branch of the pair potential,

Vij(r)=f{2r−2r2+d2,σiσj=+1,1r+1r2+4d2−2r2+d2,σiσj=−1,(7)V_{ij}(r)=f\begin{cases} \frac{2}{r}-\frac{2}{\sqrt{r^{2}+d^{2}}}, & \sigma_i\sigma_j=+1,\\[6pt] \frac{1}{r}+\frac{1}{\sqrt{r^{2}+4d^{2}}}-\frac{2}{\sqrt{r^{2}+d^{2}}}, & \sigma_i\sigma_j=-1, \end{cases} \tag*{(7)}

where f=e24πϵ0ϵefff=\frac{e^{2}}{4\pi\epsilon_{0}\epsilon_{\mathrm{eff}}} is a prefactor, ϵeff\epsilon_{\mathrm{eff}} is the effective dielectric constant of the hexagonal boron nitride substrate, rr is the planar separation of i,ji,j-th dipoles, and dd is the interlayer distance. The interaction becomes a symmetric average VQ(r)=[Vσσ+Vσ,−σ]/2V_{\mathrm{Q}}(r) = [V_{\sigma\sigma} + V_{\sigma,-\sigma}]/2 for the lower quadrupolar exciton branch. The Hamiltonian is

Lattice constant (nm)p0p_0 (e.nm)∣t∣|t| (meV)[Eq+−Eq−][\mathcal{E}_q^+ - \mathcal{E}_q^-] (meV)
0.325 (60°)0.61 ± 0.023.8 ± 0.814.5
0.325 (0°)0.63 ± 0.021.3 ± 0.43.8
1.15 (16.1°)0.59 ± 0.020.7 ± 0.31.8
3.49 (4.7°)0.66 ± 0.020.7 ± 0.42.0
4.89 (2.8°)0.64 ± 0.021.7 ± 0.56.5
5.59 (2.2°)0.64 ± 0.021.9 ± 0.56.9

Table 1. Parameters of the avoided crossing model. We obtain Eq+\mathcal{E}_q^+ from the Bethe-Salpeter equation and extract p0p_0 and tt by fitting dEq−dEz\frac{d\mathcal{E}_q^-}{d\mathcal{E}_z} to the avoided crossing model. The twist angle is defined between neighboring WSe2_2 and WS2_2 layers.

H=∑i<jV(rij)+(Eq+−Eq−2)Nd+VMNM,(8)\mathcal{H} = \sum_{i<j} V(\mathbf{r}_{ij}) + \left(\frac{\mathcal{E}_q^+ - \mathcal{E}_q^-}{2}\right)N_d + V_{\mathrm{M}}N_{\mathrm{M}}, \tag*{(8)}

where the first term sums electrostatic interactions, NdN_d counts excitons in the dipolar state, and NMN_{\mathrm{M}} counts excitons not trapped in a moiré.

Figure 5(a) shows the resulting phase diagram for free quadrupolar excitons, computed with ϵeff=4\epsilon_{\mathrm{eff}} = 4, Eq+−Eq−≤20\mathcal{E}_q^+ - \mathcal{E}_q^- \le20 meV, d=0.61d = 0.61 nm, and VM=0V_{\mathrm{M}} = 0. We use the dimensionless parameters nd2nd^2, and R=(Eq+−Eq−)4πϵ0ϵeffd2e2R = \frac{(\mathcal{E}_q^+ - \mathcal{E}_q^-)4\pi\epsilon_0\epsilon_{\mathrm{eff}}d}{2e^2} to define the phase diagram, where nn is the exciton density. Three phases emerge: a triangular lattice of quadrupolar excitons [Figure 5(b)], a square lattice of staggered dipolar excitons [Figure 5(c)], and an inhomogeneous “droplet” phase of dense dipolar clusters without long-range order [Figure 5(d)][30]. Quadrupolar excitons repel via V(r)∼1/r5V(r) \sim1/r^5, while antiparallel dipoles attract via V(r)∼−1/r3V(r) \sim-1/r^3. At low nn, the electrostatic gain from conversion to staggered dipoles is outweighed by RR, stabilising the triangular quadrupolar lattice. At high nn, the gain exceeds RR and the system transitions to the staggered dipolar phase, selecting a square bipartite geometry to avoid the frustration inherent to the triangular lattice [Figure 5(e),(Figure 5g)]. Moiré trapping fundamentally alters this competition by confining excitons to the triangular lattice, expanding the quadrupolar phase boundaries Figure 5(e) with VM=20V_{\mathrm{M}} = 20 meV]. This resolves an anomaly reported in Ref. 5, where quadrupolar excitons were observed at densities below n≲3.3×1013 cm−2n \lesssim3.3 \times10^{13}\ \mathrm{cm}^{-2} (nd2≲0.12nd^{2} \lesssim0.12).

Image not extracted

Figure 5. Many-exciton phase diagram for free and trapped quadrupolar excitons. (a)–(d): Free quadrupolar exciton phases include (b) a triangular lattice of quadrupoles, (c) a square lattice of antiparallel dipoles, and (d) droplet phases lacking long-range order. (e)–(g): Trapped quadrupolar exciton phases include (e) a triangular lattice of quadrupoles, (g) frustrated triangular lattice of anti-parallel dipoles, (f) droplet.

Below such nn with R=0.02R = 0.02, untrapped excitons are expected to favour the dipolar droplet phase (Figure 5(a)), making the observation of quadrupolar excitons surprising. Moiré trapping stabilises the quadrupolar phase (Figure 5(e)). While evidence for staggered dipolar excitons has been found at higher densities[5], the in-plane lattice ordering (for instance, triangular versus square) has yet to be resolved experimentally.

In summary, we develop a transfer-matrix formalism coupled with a fully atomistic BSE framework, revealing the non-trivial real-space structure of quadrupolar excitons in van der Waals trilayers. Moiré trapping further establishes these excitons as a platform for exotic many-body physics[48], from frustrated quantum magnetism to SU(4)SU(4) spin liquids.

Author contributions

I. M. and J. L. discussed the initial project ideas. I. M. led the project, developed the algorithms, carried out the simulations, and wrote the first draft of the manuscript. A. A. M. and J. L. provided computing resources. All authors discussed the results and contributed to the manuscript.

Ethics declaration

The authors declare no competing interests.

Data availability

All data used to produce the plots in the main text of the paper will be made publicly available at https://github.com/imaitygit/quadrupolar-excitons.

Code availability

Structure construction, atomic relaxations, electronic structure, and exciton calculations presented in this work were performed using publicly available codes. An updated version of the PyMEX code, including the new capabilities described here, is planned for release by the end of 2026. Code for computing many-exciton phases will be made publicly available at https://github.com/imaitygit/quadrupolar-excitons.

Acknowledgements

This work used the ARCHER2 UK National Supercomputing Service (https://www.archer2.ac.uk)[49] through our membership of the UKCP consortium (EP/X035891/1) and the HEC Materials Chemistry Consortium (EP/X035859), both funded by EPSRC. We thank Sufei Shi for valuable comments on the manuscript.

Methods

Atomic structure. Twisted bilayer WSe2_2/WS2_2 structures were generated using the Twister package[50]. In the trilayer system, the top and bottom WSe2_2 layers are aligned in parallel to form a symmetric moiré superlattice. Atomic relaxations were performed using classical interatomic potentials in Lammps[51, 52]. Interactions within each layer were modeled using the Stillinger–Weber potential[53]. Interactions between neighboring layers were modeled using the Kolmogorov–Crespi potential[54], as these dominate over contributions from next-nearest-neighbor layers[55].

Electronic structure. The Siesta package was employed for electronic structure calculations using localized atomic orbitals as basis functions[56]. Norm-conserving Troullier–Martins pseudopotentials[57] and the local density approximation were used for the exchange-correlation energy[58]. Wavefunctions were expanded in a double-ζ\zeta plus polarization basis. All moiré calculations used Γ\Gamma-point sampling, while lattice-matched system calculations employed a 12×12×112 \times12 \times1 kk-point grid for the charge density, a plane-wave energy cutoff of 100 Ry, an energy shift of 0.02 Ry, and a vacuum spacing of 20 Å in the out-of-plane direction. Spin-orbit coupling[59] was excluded during Wannier function generation but was included to extract the Γ\Gamma–KK ordering of bands.

Wannier functions were generated via the one-shot projection method[60] as implemented in the Wannier90 code[61], using dd and pp orbitals as initial projections. Specifically, Kohn-Sham wavefunctions were projected onto atom-centered dd (dxyd_{xy}, dyzd_{yz}, dzxd_{zx}, dx2−y2d_{x^2-y^2}, dz2d_{z^2}) orbitals on every W atom and pp (pxp_x, pyp_y, pzp_z) orbitals on every S and Se atom, followed by orthogonalization and disentanglement[62]. We used minimal-distance replica selection for Fourier interpolation of the Wannier Hamiltonian[35, 61].

Spin-orbit coupling (SOC) was excluded from Wannier function generation, which caused Γ\Gamma-derived bands of the pristine unit cell to fold above KK-derived bands in the moiré Brillouin zone, incorrectly ordering the Γv→Kc\Gamma_v \to K_c and Kv→KcK_v \to K_c transition energies. We corrected this by applying a rigid shift to KK-derived valence bands identified at each kk-point via the orbital-weight score

snk=∑α(∑m∈{dx2−y2,dxy}∣Cmnkα∣2−∑m∈{dz2,dxz,dyz}∣Cmnkα∣2),(9)s_{nk}=\sum_{\alpha}\left(\sum_{m\in\{d_{x^2-y^2},d_{xy}\}}\left|C_{mnk}^{\alpha}\right|^2-\sum_{m\in\{d_{z^2},d_{xz},d_{yz}\}}\left|C_{mnk}^{\alpha}\right|^2\right), \tag*{(9)}

where CmnkαC_{mnk}^{\alpha} are the Wannier projection coefficients for orbital mm on atom α\alpha, summed over all atoms per layer. Bands with snk>0.1s_{nk}>0.1 were shifted by a rigid ΔE\Delta E determined by explicit comparison to SOC-included calculations, restoring the correct Γ\Gamma–KK ordering. This is physically motivated by the well-known orbital composition of 2D transition metal dichalcogenides valence bands: KK states are dominated by dx2−y2d_{x^2-y^2} and dxyd_{xy}, while Γ\Gamma states are dominated by dz2d_{z^2}[63]. We implemented this procedure in PyMEX.

Excitonic structure. We implemented the transfer matrix approach described above in PyMEX[40] for arbitrarily stacked multi-layer structures. We employed PyMEX to construct and diagonalise the BSE Hamiltonian and compute absorption spectra using Wannier functions as basis, with diagonalisation accelerated by the Eigenvalue soLvers for Petaflop Applications (ELPA) library[64, 65]. We applied a vertical electric field EzE_z by adding the term −eEz∑izi∣i⟩⟨i∣-eE_z\sum_i z_i|i\rangle\langle i| to the electronic Hamiltonian, where ziz_i is the zz-coordinate of the ii-th Wannier function and ee is the electron charge. At each twist angle, we performed BSE calculations at 25 electric-field values and interpolated the results to construct a heatmap of the field-dependent absorption spectra. We used kk-point grids of 78×78×178\times78\times1, 18×18×118\times18\times1, 6×6×16\times6\times1, 6×6×16\times6\times1, and 3×3×13\times3\times1, and included 3, 6, 20, 24, and 40 valence bands and 3, 6, 10, 8, and 16 conduction bands in the BSE Hamiltonian for the lattice-matched, 16.1∘16.1^\circ, 4.7∘4.7^\circ, 2.8∘2.8^\circ, and 2.2∘2.2^\circ WSe2_2/WS2_2/WSe2_2 trilayers, respectively. A full description of PyMEX, including implementation details, documentation, and an open-source release, is beyond the scope of this work and will appear in a subsequent paper. Spin-orbit coupling is omitted from our BSE Hamiltonian but can be included perturbatively[40]. It splits the quadrupolar 1s-like exciton into singlet and triplet states separated by ∼40\sim40 meV, with the singlet being optically bright[29]. All results presented in the main text apply to both branches independently.

Exciton phase diagram. Lattice sums were evaluated over all integer pairs (n1,n2)(n_1,n_2) with ∣ni∣≤65|n_i|\le65, excluding the origin, on both square and triangular lattices, with lattice constants set by the site density nn via a2=1/na^2=1/n and 32a2=1/n\frac{\sqrt{3}}{2}a^2=1/n respectively. See SI, Sec. E for additional details. The dipolar phases carry an additional energy cost RR per exciton (the quadrupole-to-dipole gap), and the dipolar square lattice incurs a further moiré penalty VMV_{\mathrm{M}} per site for being incommensurate with the triangular moiré lattice. Below their respective optimal densities n∗n^{*}, each dipolar phase undergoes a droplet construction: excitons phase-separate into dense droplets at n∗n^{*} surrounded by empty regions, with the interaction energy fixed to its minimum value. Since moiré potentials can exceed 100 meV[66, 67], the value VM=20V_{\mathrm{M}} = 20 meV used here is conservative, and our conclusions remain robust.

Supplementary Information to “Moiré trapping of quadrupolar excitons in van der Waals trilayers”

Methods

Atomic structure and Wannier functions

Multilayer screening using the transfer matrix method

  • Two stacked 2D materials

Outside the stacked 2D materials, there are three regions: I (z>dz > d), II (0<z<d0 < z < d), and III (z<0z < 0). We take d=d12d = d_{12}. The electrostatic potential admits the following solutions in each regions

reg. I:ϕI(q,z)=Ae−qzreg. II:ϕII(q,z)=Beqz+Ce−qzreg. III:ϕIII(q,z)=Deqz(10)\begin{aligned} \text{reg. I:}\quad&\phi_I(q,z) = A e^{-qz}\\ \text{reg. II:}\quad&\phi_{II}(q,z) = B e^{qz} + C e^{-qz}\\ \text{reg. III:}\quad&\phi_{III}(q,z) = D e^{qz} \tag*{(10)} \end{aligned}

We define a coefficient vector \psi_z = \begin{pmatrix} \text{coeff. of }e^{-qz}\text{ evaluated at }z \ \text{coeff. of }e^{qz}\text{ evaluated at }z \end{pmatrix}.

Propagation matrix: Inside a homogeneous dielectric, the coefficients of the exponential basis evolve independently. Starting from

ψ(z)=(UbVb),(11)\psi(z) = \begin{pmatrix} U_b \\ V_b \end{pmatrix}, \tag*{(11)}

an out-of-plane shift by a distance dd gives

ψ(z+d)=(UtVt)=(Ube−qdVbeqd).(12)\psi(z+d) = \begin{pmatrix} U_t \\ V_t \end{pmatrix} = \begin{pmatrix} U_b e^{-qd} \\ V_b e^{qd} \end{pmatrix}. \tag*{(12)}

This evolution can be captured compactly by a propagation matrix,

ψ(z+d)=P(d,ϵ) ψ(z),(13)\psi(z+d) = P(d,\epsilon)\,\psi(z), \tag*{(13)}

where

P(d)=(e−qd00eqd)(14)P(d) = \begin{pmatrix} e^{-qd} & 0 \\ 0 & e^{qd} \end{pmatrix} \tag*{(14)}

Layer matrix: The layer matrix is 2D polarizable, and the boundary conditions on ϕ(q,z)\phi(q,z) and ∂ϕ∂z\frac{\partial\phi}{\partial z} need to be adjusted suitably. The setup consists of a 2D material with polarizability χ\chi, and dielectric constants ϵt\epsilon_t and ϵb\epsilon_b above and below it, respectively. Potenital continuity condition (i.e., ϕ(q,z→0−)=ϕ(q,z→0+)\phi(q,z\to0^-) = \phi(q,z\to0^+)) leads to

Ub+Vb=Ut+Vt(15)U_b + V_b = U_t + V_t \tag*{(15)}

On the other hand, the discontinuous jump in the normal component of the electric field (i.e., Dz→0+−Dz→0−=σb+σfD_{z\to0^+} - D_{z\to0^-} = \sigma_b + \sigma_f) leads to

−ϵ0ϵt∂ϕ(q,z)∂z∣0++ϵ0ϵb∂ϕ(q,z)∂z∣0−=−ϵ0χq2ϕ(q,0)(16)-\epsilon_0\epsilon_t\left.\frac{\partial\phi(q,z)}{\partial z}\right|_{0+} + \epsilon_0\epsilon_b\left.\frac{\partial\phi(q,z)}{\partial z}\right|_{0-} = -\epsilon_0\chi q^2\phi(q,0) \tag*{(16)}

In the above, we did not include source charge (i.e., σf=0\sigma_f = 0). After a bit of algebra, this simplifies to

ϵt(Ut−Vt)−ϵb(Ub−Vb)=−χq(Ut+Vt)(17)\epsilon_t(U_t - V_t) - \epsilon_b(U_b - V_b) = -\chi q(U_t + V_t) \tag*{(17)}

Both Equations 13 and 15 must be satisfied simultaneously. Using the matrix formulation, both equations can be solved efficiently.

(11ϵt+χq−ϵt+χq)(UtVt)=(11ϵb−ϵb)(UbVb)(UtVt)=(11ϵt+χq−ϵt+χq)−1(11ϵb−ϵb)(UbVb)(18)\begin{aligned} \begin{pmatrix} 1 & 1 \\ \epsilon_t + \chi q & -\epsilon_t + \chi q \end{pmatrix} \begin{pmatrix} U_t \\ V_t \end{pmatrix} &= \begin{pmatrix} 1 & 1 \\ \epsilon_b & -\epsilon_b \end{pmatrix} \begin{pmatrix} U_b \\ V_b \end{pmatrix} \\ \begin{pmatrix} U_t \\ V_t \end{pmatrix} &= \begin{pmatrix} 1 & 1 \\ \epsilon_t + \chi q & -\epsilon_t + \chi q \end{pmatrix}^{-1} \begin{pmatrix} 1 & 1 \\ \epsilon_b & -\epsilon_b \end{pmatrix} \begin{pmatrix} U_b \\ V_b \end{pmatrix} \tag*{(18)} \end{aligned}

We can extract (UtVt)=L(χ,ϵb,ϵt)(UbVb)\begin{pmatrix} U_t \\ V_t \end{pmatrix} = L(\chi,\epsilon_b,\epsilon_t)\begin{pmatrix} U_b \\ V_b \end{pmatrix}, where

L(χ,ϵb,ϵt)=12ϵt(ϵt+ϵb−χqϵt−ϵb−χqϵt−ϵb+χqϵt+ϵb+χq)(19)L(\chi,\epsilon_b,\epsilon_t) = \frac{1}{2\epsilon_t} \begin{pmatrix} \epsilon_t + \epsilon_b - \chi q & \epsilon_t - \epsilon_b - \chi q \\ \epsilon_t - \epsilon_b + \chi q & \epsilon_t + \epsilon_b + \chi q \end{pmatrix} \tag*{(19)}

Source matrix: Note that the source charge is not included in the above. The inclusion of source charge modifies Eqn. (17)

ϵt(Ut−Vt)−ϵb(Ub−Vb)=−eϵ0q−χq(Ut+Vt)(20)\epsilon_t(U_t - V_t) - \epsilon_b(U_b - V_b) = -\frac{e}{\epsilon_0 q} - \chi q(U_t + V_t) \tag*{(20)}

This leads to different set of equations to solve

(11ϵt+χq−ϵt+χq)(UtVt)=(11ϵb−ϵb)(UbVb)+(0−eϵ0q)(UtVt)=(11ϵt+χq−ϵt+χq)−1(11ϵb−ϵb)(UbVb)+(11ϵt+χq−ϵt+χq)−1(0−eϵ0q)(UtVt)=12ϵt(ϵt+ϵb−χqϵt−ϵb−χqϵt−ϵb+χqϵt+ϵb+χq)(UbVb)+(−e2ϵ0ϵtqe2ϵ0ϵtq)(UtVt)=L(χ,ϵb,ϵt)(UbVb)+S(21)\begin{aligned} \begin{pmatrix} 1 & 1 \\ \epsilon_t + \chi q & -\epsilon_t + \chi q \end{pmatrix} \begin{pmatrix} U_t \\ V_t \end{pmatrix} &= \begin{pmatrix} 1 & 1 \\ \epsilon_b & -\epsilon_b \end{pmatrix} \begin{pmatrix} U_b \\ V_b \end{pmatrix} + \begin{pmatrix} 0 \\ -\frac{e}{\epsilon_0 q} \end{pmatrix} \\ \begin{pmatrix} U_t \\ V_t \end{pmatrix} &= \begin{pmatrix} 1 & 1 \\ \epsilon_t + \chi q & -\epsilon_t + \chi q \end{pmatrix}^{-1} \begin{pmatrix} 1 & 1 \\ \epsilon_b & -\epsilon_b \end{pmatrix} \begin{pmatrix} U_b \\ V_b \end{pmatrix} + \begin{pmatrix} 1 & 1 \\ \epsilon_t + \chi q & -\epsilon_t + \chi q \end{pmatrix}^{-1} \begin{pmatrix} 0 \\ -\frac{e}{\epsilon_0 q} \end{pmatrix} \\ \begin{pmatrix} U_t \\ V_t \end{pmatrix} &= \frac{1}{2\epsilon_t} \begin{pmatrix} \epsilon_t + \epsilon_b - \chi q & \epsilon_t - \epsilon_b - \chi q \\ \epsilon_t - \epsilon_b + \chi q & \epsilon_t + \epsilon_b + \chi q \end{pmatrix} \begin{pmatrix} U_b \\ V_b \end{pmatrix} + \begin{pmatrix} -\frac{e}{2\epsilon_0\epsilon_t q} \\ \frac{e}{2\epsilon_0\epsilon_t q} \end{pmatrix} \\ \begin{pmatrix} U_t \\ V_t \end{pmatrix} &= L(\chi,\epsilon_b,\epsilon_t) \begin{pmatrix} U_b \\ V_b \end{pmatrix} + S \tag*{(21)} \end{aligned}

The source matrix is given below, applies only at the source layer, and determines the propagated solutions thereafter.

S=(−e2ε0εtqe2ε0εtq)(22)\mathcal{S}=\begin{pmatrix}-\frac{e}{2\varepsilon_{0}\varepsilon_{t}q}\\[2pt]\frac{e}{2\varepsilon_{0}\varepsilon_{t}q}\end{pmatrix} \tag*{(22)}

We use the transfer matrix method to calculate the response of a multilayer system. Consider the bilayer 2D material setup shown in Figure 6. In the bottom region (z<0z<0), there is a dielectric with constant εb\varepsilon_{\mathrm{b}}. Layer 1 is modeled as an ideal 2D layer at z=0z=0 with polarizability χ1\chi_{1}. Between Layers 1 and 2 (0<z<d0<z<d), there is vacuum with ε=1\varepsilon=1. Layer 2 is an ideal 2D layer at z=dz=d with polarizability χ2\chi_{2}. In the top region (z>dz>d), a dielectric with constant εt\varepsilon_{\mathrm{t}} is present.

Image not extracted

Figure 6. Schematics of dipolar and quadrupolar excitons.

Case I: Source charge in layer 1

The transfer matrix for the electrostatic potential, expressed in terms of the state vector ψz\psi_{z}, can be written as

ψz→∞=Mψz→−∞=L(χ2,1,εt)P(d,1)[L(χ1,εb,1)ψz→−∞+S]\begin{aligned} \psi_{z\to\infty}=M\psi_{z\to-\infty} \\ &=L(\chi_{2},1,\varepsilon_{\mathrm{t}})P(d,1)\left[L(\chi_{1},\varepsilon_{\mathrm{b}},1)\psi_{z\to-\infty}+\mathcal{S}\right] \end{aligned}
(A0)=12εt(εt+1−χ2qεt−1−χ2qεt−1+χ2qεt+1+χ2q)(e−qd00eqd)12(1+εb−χ1q1−εb−χ1q1−εb+χ1q1+εb+χ1q)(0D)+12εt(εt+1−χ2qεt−1−χ2qεt−1+χ2qεt+1+χ2q)(e−qd00eqd)e2ε0q(−11)(23)\begin{aligned} \begin{pmatrix}A\\0\end{pmatrix} = \frac{1}{2\varepsilon_{\mathrm{t}}} \begin{pmatrix} \varepsilon_{\mathrm{t}}+1-\chi_{2}q & \varepsilon_{\mathrm{t}}-1-\chi_{2}q\\ \varepsilon_{\mathrm{t}}-1+\chi_{2}q & \varepsilon_{\mathrm{t}}+1+\chi_{2}q \end{pmatrix} \begin{pmatrix} e^{-qd} & 0\\ 0 & e^{qd} \end{pmatrix} \frac{1}{2} \begin{pmatrix} 1+\varepsilon_{\mathrm{b}}-\chi_{1}q & 1-\varepsilon_{\mathrm{b}}-\chi_{1}q\\ 1-\varepsilon_{\mathrm{b}}+\chi_{1}q & 1+\varepsilon_{\mathrm{b}}+\chi_{1}q \end{pmatrix} \begin{pmatrix}0\\D\end{pmatrix} \\ &+\frac{1}{2\varepsilon_{\mathrm{t}}} \begin{pmatrix} \varepsilon_{\mathrm{t}}+1-\chi_{2}q & \varepsilon_{\mathrm{t}}-1-\chi_{2}q\\ \varepsilon_{\mathrm{t}}-1+\chi_{2}q & \varepsilon_{\mathrm{t}}+1+\chi_{2}q \end{pmatrix} \begin{pmatrix} e^{-qd} & 0\\ 0 & e^{qd} \end{pmatrix} \frac{e}{2\varepsilon_{0}q} \begin{pmatrix}-1\\1\end{pmatrix} \tag*{(23)} \end{aligned}

We obtain DD and propagate the solutions to extract AA, BB, CC.

Case II: Source charge in layer 2

ψz→∞=Mψz→−∞=[L(χ2,1,εt)P(d,1)L(χ1,εb,1)ψz→−∞+S]\begin{aligned} \psi_{z\to\infty}=M\psi_{z\to-\infty} \\ &=\left[L(\chi_{2},1,\varepsilon_{\mathrm{t}})P(d,1)L(\chi_{1},\varepsilon_{\mathrm{b}},1)\psi_{z\to-\infty}+\mathcal{S}\right] \end{aligned}
\begin{pmatrix}A\\0\end{pmatrix} = \left[L(\chi_{2},1,\varepsilon_{\mathrm{t}})P(d,1)L(\chi_{1},\varepsilon_{\mathrm{b}},1) \begin{pmatrix}0\\D\end{pmatrix} +\mathcal{S}\right] \tag*{(24)}\end{pmatrix}

Again, DD is computed first, and the solutions are then propagated to determine the potentials in the other layer.

ϕ11=C(χ2q+ϵt+1+(−χ2q−ϵt+1)e−2d12q)q[(χ1q+ϵb−1)(χ2q+ϵt−1)e−2d12q−(χ1q+ϵb+1)(χ2q+ϵt+1)](25)\phi_{11}=\frac{C\left(\chi_{2q}+\epsilon_t+1+\left(-\chi_{2q}-\epsilon_t+1\right)e^{-2d_{12}q}\right)}{q\left[(\chi_{1q}+\epsilon_b-1)(\chi_{2q}+\epsilon_t-1)e^{-2d_{12}q}-(\chi_{1q}+\epsilon_b+1)(\chi_{2q}+\epsilon_t+1)\right]} \tag*{(25)}
ϕ12=2Ce−d12qq[(χ1q+ϵb−1)(χ2q+ϵt−1)e−2d12q−(χ1q+ϵb+1)(χ2q+ϵt+1)](26)\phi_{12}=\frac{2Ce^{-d_{12}q}}{q\left[(\chi_{1q}+\epsilon_b-1)(\chi_{2q}+\epsilon_t-1)e^{-2d_{12}q}-(\chi_{1q}+\epsilon_b+1)(\chi_{2q}+\epsilon_t+1)\right]} \tag*{(26)}
ϕ21=2Ce−d12qq[(χ1q+ϵb−1)(χ2q+ϵt−1)e−2d12q−(χ1q+ϵb+1)(χ2q+ϵt+1)](27)\phi_{21}=\frac{2Ce^{-d_{12}q}}{q\left[(\chi_{1q}+\epsilon_b-1)(\chi_{2q}+\epsilon_t-1)e^{-2d_{12}q}-(\chi_{1q}+\epsilon_b+1)(\chi_{2q}+\epsilon_t+1)\right]} \tag*{(27)}
ϕ22=C[(χ1q+ϵb+1)−(χ1q+ϵb−1)e−2d12q]q[(χ1q+ϵb−1)(χ2q+ϵt−1)e−2d12q−(χ1q+ϵb+1)(χ2q+ϵt+1)](28)\phi_{22}=\frac{C\left[(\chi_{1q}+\epsilon_b+1)-(\chi_{1q}+\epsilon_b-1)e^{-2d_{12}q}\right]}{q\left[(\chi_{1q}+\epsilon_b-1)(\chi_{2q}+\epsilon_t-1)e^{-2d_{12}q}-(\chi_{1q}+\epsilon_b+1)(\chi_{2q}+\epsilon_t+1)\right]} \tag*{(28)}

Sanity check and comparison with other works:

ϕ11=C[(r2q+1)−r2qe−2d12q]2q[r1r2q2e−2d12q−(r1q+1)(r2q+1)]ϕ21=Ce−d12q2q[r1r2q2e−2d12q−(r1q+1)(r2q+1)]ϕ21=Ce−d12q2q[r1r2q2e−2d12q−(r1q+1)(r2q+1)]ϕ22=C[(r1q+1)−r1qe−2d12q]2q[r1r2q2e−2d12q−(r1q+1)(r2q+1)](29)\begin{aligned} \phi_{11}&=\frac{C\left[(r_{2q}+1)-r_{2q}e^{-2d_{12}q}\right]}{2q\left[r_1r_{2q}^{2}e^{-2d_{12}q}-(r_{1q}+1)(r_{2q}+1)\right]}\\ \phi_{21}&=\frac{Ce^{-d_{12}q}}{2q\left[r_1r_{2q}^{2}e^{-2d_{12}q}-(r_{1q}+1)(r_{2q}+1)\right]}\\ \phi_{21}&=\frac{Ce^{-d_{12}q}}{2q\left[r_1r_{2q}^{2}e^{-2d_{12}q}-(r_{1q}+1)(r_{2q}+1)\right]}\\ \phi_{22}&=\frac{C\left[(r_{1q}+1)-r_{1q}e^{-2d_{12}q}\right]}{2q\left[r_1r_{2q}^{2}e^{-2d_{12}q}-(r_{1q}+1)(r_{2q}+1)\right]} \tag*{(29)} \end{aligned}

In the limit where both sides are vacuum (i.e., ϵt=ϵb=1\epsilon_t=\epsilon_b=1), our results are consistent with those of Danovich and co-workers[42]. We have used χi=2ri\chi_i=2r_i.

Three stacked 2D materials

There are four regions of interest: region I (z>d12+d23z>d_{12}+d_{23}), region II (d12<z<d12+d23d_{12}<z<d_{12}+d_{23}), region III (0<z<d120<z<d_{12}), and region IV (z<0z<0). Figure 6 shows the setup. The electrostatic potential in these regions admits solutions of the form

reg. I:ϕI(q,z)=Ae−qzreg. II:ϕII(q,z)=Beqz+Ce−qzreg. III:ϕIII(q,z)=Deqz+Ee−qzreg. IV:ϕIV(q,z)=Feqz(30)\begin{aligned} \text{reg. I:}\quad\phi_I(q,z)&=Ae^{-qz}\\ \text{reg. II:}\quad\phi_{II}(q,z)&=Be^{qz}+Ce^{-qz}\\ \text{reg. III:}\quad\phi_{III}(q,z)&=De^{qz}+Ee^{-qz}\\ \text{reg. IV:}\quad\phi_{IV}(q,z)&=Fe^{qz} \tag*{(30)} \end{aligned}

Three 2D materials have polarizabilities χ1\chi_1, χ2\chi_2, and χ3\chi_3. Using the same approach as for the bilayer, we construct the electrostatic potential with the transfer matrix method. The source can be located in any of the three layers.

Case I: Electron in layer 1

ψz→∞=Mψz→−∞=L(χ3,1,ϵt)P(d23,1)L(χ2,1,1)P(d12,1)[L(χ1,ϵb,1)ψz→−∞+S]=(M11M12M21M22)ψz→−∞+[L(χ3,1,ϵt)P(d23,1)L(χ2,1,1)P(d12,1)S]=(M11M12M21M22)ψz→−∞+S~(A0)=(M11M12M21M22)(0F)+(S~1S~2)(31)\begin{aligned} \psi_{z\to\infty} &= \mathcal{M}\psi_{z\to-\infty}\\ &= L(\chi_{3},1,\epsilon_{t})P(d_{23},1)L(\chi_{2},1,1)P(d_{12},1)\left[L(\chi_{1},\epsilon_{b},1)\psi_{z\to-\infty}+S\right]\\ &= \begin{pmatrix} M_{11} & M_{12}\\ M_{21} & M_{22} \end{pmatrix}\psi_{z\to-\infty}+\left[L(\chi_{3},1,\epsilon_{t})P(d_{23},1)L(\chi_{2},1,1)P(d_{12},1)S\right]\\ &= \begin{pmatrix} M_{11} & M_{12}\\ M_{21} & M_{22} \end{pmatrix}\psi_{z\to-\infty}+\widetilde{S}\\ \begin{pmatrix} A\\ 0 \end{pmatrix} &= \begin{pmatrix} M_{11} & M_{12}\\ M_{21} & M_{22} \end{pmatrix}\begin{pmatrix} 0\\ F \end{pmatrix}+\begin{pmatrix} \widetilde{S}_{1}\\ \widetilde{S}_{2} \end{pmatrix} \tag*{(31)} \end{aligned}

We extract F=−S~2M22F=-\frac{\widetilde{S}_{2}}{M_{22}}. As a sanity check, we verify that the left-hand side and right-hand side match after obtaining the solution. We propagate the solution, FF to obtain ϕ11\phi_{11}, ϕ12\phi_{12}, ϕ13\phi_{13}.

(ϕ111ϕ112)=[L(χ1,ϵb,1)(0F)+S](ϕ121ϕ122)=L(χ2,1,1)P(d12,1)[L(χ1,ϵb,1)(0F)+S](ϕ131ϕ132)=L(χ3,1,ϵt)P(d23,1)L(χ2,1,1)P(d12,1)[L(χ1,ϵb,1)(0F)+S](32)\begin{aligned} \begin{pmatrix} \phi_{11}^{1}\\ \phi_{11}^{2} \end{pmatrix} &= \left[L(\chi_{1},\epsilon_{b},1)\begin{pmatrix} 0\\ F \end{pmatrix}+S\right]\\ \begin{pmatrix} \phi_{12}^{1}\\ \phi_{12}^{2} \end{pmatrix} &= L(\chi_{2},1,1)P(d_{12},1)\left[L(\chi_{1},\epsilon_{b},1)\begin{pmatrix} 0\\ F \end{pmatrix}+S\right]\\ \begin{pmatrix} \phi_{13}^{1}\\ \phi_{13}^{2} \end{pmatrix} &= L(\chi_{3},1,\epsilon_{t})P(d_{23},1)L(\chi_{2},1,1)P(d_{12},1)\left[L(\chi_{1},\epsilon_{b},1)\begin{pmatrix} 0\\ F \end{pmatrix}+S\right] \tag*{(32)} \end{aligned}

Case II: Electron in layer 2

ψz→∞=Mψz→−∞=L(χ3,1,ϵt)P(d23,1)[L(χ2,1,1)P(d12,1)L(χ1,ϵb,1)ψz→−∞+S]=(M11M12M21M22)ψz→−∞+[L(χ3,1,ϵt)P(d23,1)S]=(M11M12M21M22)ψz→−∞+S~(A0)=(M11M12M21M22)(0F)+(S~1S~2)(33)\begin{aligned} \psi_{z\to\infty} &= \mathcal{M}\psi_{z\to-\infty}\\ &= L(\chi_{3},1,\epsilon_{t})P(d_{23},1)\left[L(\chi_{2},1,1)P(d_{12},1)L(\chi_{1},\epsilon_{b},1)\psi_{z\to-\infty}+S\right]\\ &= \begin{pmatrix} M_{11} & M_{12}\\ M_{21} & M_{22} \end{pmatrix}\psi_{z\to-\infty}+\left[L(\chi_{3},1,\epsilon_{t})P(d_{23},1)S\right]\\ &= \begin{pmatrix} M_{11} & M_{12}\\ M_{21} & M_{22} \end{pmatrix}\psi_{z\to-\infty}+\widetilde{S}\\ \begin{pmatrix} A\\ 0 \end{pmatrix} &= \begin{pmatrix} M_{11} & M_{12}\\ M_{21} & M_{22} \end{pmatrix}\begin{pmatrix} 0\\ F \end{pmatrix}+\begin{pmatrix} \widetilde{S}_{1}\\ \widetilde{S}_{2} \end{pmatrix} \tag*{(33)} \end{aligned}

We extract F=−S~2M22F=-\frac{\widetilde{S}_{2}}{M_{22}}. As a sanity check, we verify that the left-hand side and right-hand side match after obtaining the solution. Very similar to above, we propagate the solutions.

Case III: Electron in layer 3

ψz→∞=Mψz→−∞=[L(x3,1,ϵt)P(d23,1)L(x2,1,1)P(d12,1)L(x1,ϵb,1)ψz→−∞+S]=(M11M12M21M22)ψz→−∞+IS=(M11M12M21M22)ψz→−∞+S~(A0)=(M11M12M21M22)(0F)+(S~1S~2)(34)\begin{aligned} \psi_{z\to\infty} &= \mathcal{M}\psi_{z\to-\infty}\\ &= \left[L(x_3,1,\epsilon_t)P(d_{23},1)L(x_2,1,1)P(d_{12},1)L(x_1,\epsilon_b,1)\psi_{z\to-\infty}+S\right]\\ &= \begin{pmatrix} M_{11} & M_{12} \\ M_{21} & M_{22} \end{pmatrix}\psi_{z\to-\infty}+\mathbb{I}S\\ &= \begin{pmatrix} M_{11} & M_{12} \\ M_{21} & M_{22} \end{pmatrix}\psi_{z\to-\infty}+\widetilde{S}\\ \begin{pmatrix} A \\ 0 \end{pmatrix} &= \begin{pmatrix} M_{11} & M_{12} \\ M_{21} & M_{22} \end{pmatrix}\begin{pmatrix} 0 \\ F \end{pmatrix}+\begin{pmatrix} \widetilde{S}_1 \\ \widetilde{S}_2 \end{pmatrix} \tag*{(34)} \end{aligned}

We extract F=−S~2M22F=-\frac{\widetilde{S}_2}{M_{22}}. Very similar to above, we propagate the solutions.

Explicit expressions are given below. Even with three layers the equations quickly become very complicated.

ϕ11=−C[A+Be−2d12q]Nϕ12=ϕ21=2CCe−d12qNϕ13=ϕ31=4Ce−q(d12+d23)Nϕ22=CCDNϕ23=ϕ32=2CDe−d23qNϕ33=−C[E+Fe−2d23q]N(35)\boxed{ \begin{aligned} \phi_{11} &= -\frac{C\left[A+B e^{-2d_{12}q}\right]}{\mathcal{N}}\\ \phi_{12} &= \phi_{21}=\frac{2C C e^{-d_{12}q}}{\mathcal{N}}\\ \phi_{13} &= \phi_{31}=\frac{4C e^{-q(d_{12}+d_{23})}}{\mathcal{N}}\\ \phi_{22} &= \frac{C C D}{\mathcal{N}}\\ \phi_{23} &= \phi_{32}=\frac{2C D e^{-d_{23}q}}{\mathcal{N}}\\ \phi_{33} &= -\frac{C\left[\mathcal{E}+F e^{-2d_{23}q}\right]}{\mathcal{N}} \end{aligned} } \tag*{(35)}

with

N=q[B(χ1q+ϵb−1)e−2d12q+A(χ1q+ϵb+1)]A=χ2q(χ3q+ϵt−1)e−2d23q−(χ2q+2)(χ3q+ϵt+1)B=χ2q(χ3q+ϵt+1)−(χ2q−2)(χ3q+ϵt−1)e−2d23qC=(χ3q+ϵt+1)−(χ3q+ϵt−1)e−2d23qD=(χ1q+ϵb+1)−(χ1q+ϵb−1)e−2d12qE=χ2q(χ1q+ϵb−1)e−2d12q−(χ2q+2)(χ1q+ϵb+1)F=χ2q(χ1q+ϵb+1)−(χ2q−2)(χ1q+ϵb−1)e−2d12q(36)\begin{aligned} \mathcal{N} &= q\left[B(\chi_{1q}+\epsilon_b-1)e^{-2d_{12}q}+\mathcal{A}(\chi_{1q}+\epsilon_b+1)\right] \\ \mathcal{A} &= \chi_{2q}(\chi_{3q}+\epsilon_t-1)e^{-2d_{23}q}-(\chi_{2q}+2)(\chi_{3q}+\epsilon_t+1) \\ \mathcal{B} &= \chi_{2q}(\chi_{3q}+\epsilon_t+1)-(\chi_{2q}-2)(\chi_{3q}+\epsilon_t-1)e^{-2d_{23}q} \\ \mathcal{C} &= (\chi_{3q}+\epsilon_t+1)-(\chi_{3q}+\epsilon_t-1)e^{-2d_{23}q} \\ \mathcal{D} &= (\chi_{1q}+\epsilon_b+1)-(\chi_{1q}+\epsilon_b-1)e^{-2d_{12}q} \\ \mathcal{E} &= \chi_{2q}(\chi_{1q}+\epsilon_b-1)e^{-2d_{12}q}-(\chi_{2q}+2)(\chi_{1q}+\epsilon_b+1) \\ \mathcal{F} &= \chi_{2q}(\chi_{1q}+\epsilon_b+1)-(\chi_{2q}-2)(\chi_{1q}+\epsilon_b-1)e^{-2d_{12}q} \tag*{(36)} \end{aligned}

Sanity checks:

Symmtery: ϕij=ϕji\phi_{ij}=\phi_{ji}. This is consistent with Green’s reciprocity theorem.

Symmetric trilayer limit: When χ1=χ3\chi_1=\chi_3, ϵb=ϵt\epsilon_b=\epsilon_t, d12=d23d_{12}=d_{23}, we find A=E\mathcal{A}=\mathcal{E}, B=F\mathcal{B}=\mathcal{F}, C=D\mathcal{C}=\mathcal{D} with ϕ11=ϕ33\phi_{11}=\phi_{33}, ϕ12=ϕ23\phi_{12}=\phi_{23}, ϕ22\phi_{22} unchanged.

Exponential decay of Off-diagonal elements: ϕ12\phi_{12}, ϕ21∼e−d12q\phi_{21}\sim e^{-d_{12}q}, ϕ23\phi_{23}, ϕ32∼e−d23q\phi_{32}\sim e^{-d_{23}q}, and ϕ13\phi_{13}, ϕ31∼e−(d12+d23)q\phi_{31}\sim e^{-(d_{12}+d_{23})q}.

Bilayer limit: i.e.,, (d23=0, χ3=0)(d_{23}=0,\ \chi_3=0) and examine the potentials to check consistency with bilayer calculations.

ϕ11∣d23=0,χ3=0=C[(χ2q+ϵt+1)−(χ2q+ϵt−1)e−2d12q]q[(χ2q+ϵt−1)(χ1q+ϵb−1)e−2d12q−(χ2q+ϵt+1)(χ1q+ϵb+1)](37)\left.\phi_{11}\right|_{d_{23}=0,\chi_3=0}= \frac{\mathcal{C}\left[(\chi_{2q}+\epsilon_t+1)-(\chi_{2q}+\epsilon_t-1)e^{-2d_{12}q}\right]} {q\left[(\chi_{2q}+\epsilon_t-1)(\chi_{1q}+\epsilon_b-1)e^{-2d_{12}q}-(\chi_{2q}+\epsilon_t+1)(\chi_{1q}+\epsilon_b+1)\right]} \tag*{(37)}
ϕ12∣d23=0,χ3=0=2Ce−d12qq[(χ2q+ϵt−1)(χ1q+ϵb−1)e−2d12q−(χ2q+ϵt+1)(χ1q+ϵb+1)](38)\left.\phi_{12}\right|_{d_{23}=0,\chi_3=0}= \frac{2\mathcal{C}e^{-d_{12}q}} {q\left[(\chi_{2q}+\epsilon_t-1)(\chi_{1q}+\epsilon_b-1)e^{-2d_{12}q}-(\chi_{2q}+\epsilon_t+1)(\chi_{1q}+\epsilon_b+1)\right]} \tag*{(38)}
ϕ13∣d23=0,χ3=0=ϕ12∣d23=0,χ3=0(39)\left.\phi_{13}\right|_{d_{23}=0,\chi_3=0}=\left.\phi_{12}\right|_{d_{23}=0,\chi_3=0} \tag*{(39)}
ϕ21∣d23=0,χ3=0=ϕ12∣d23=0,χ3=0(40)\left.\phi_{21}\right|_{d_{23}=0,\chi_3=0}=\left.\phi_{12}\right|_{d_{23}=0,\chi_3=0} \tag*{(40)}
ϕ22∣d23=0,χ3=0=C[(χ1q+ϵb+1)−(χ1q+ϵb−1)e−2d12q]q[(χ2q+ϵt−1)(χ1q+ϵb−1)e−2d12q−(χ2q+ϵt+1)(χ1q+ϵb+1)](41)\left.\phi_{22}\right|_{d_{23}=0,\chi_3=0}= \frac{\mathcal{C}\left[(\chi_{1q}+\epsilon_b+1)-(\chi_{1q}+\epsilon_b-1)e^{-2d_{12}q}\right]} {q\left[(\chi_{2q}+\epsilon_t-1)(\chi_{1q}+\epsilon_b-1)e^{-2d_{12}q}-(\chi_{2q}+\epsilon_t+1)(\chi_{1q}+\epsilon_b+1)\right]} \tag*{(41)}
ϕ23∣d23=0,χ3=0=ϕ22∣d23=0,χ3=0(42)\left.\phi_{23}\right|_{d_{23}=0,\chi_3=0}=\left.\phi_{22}\right|_{d_{23}=0,\chi_3=0} \tag*{(42)}
ϕ31∣d23=0,χ3=0=ϕ12∣d23=0,χ3=0(43)\left.\phi_{31}\right|_{d_{23}=0,\chi_3=0}=\left.\phi_{12}\right|_{d_{23}=0,\chi_3=0} \tag*{(43)}
ϕ32∣d23=0,χ3=0=ϕ22∣d23=0,χ3=0(44)\left.\phi_{32}\right|_{d_{23}=0,\chi_3=0}=\left.\phi_{22}\right|_{d_{23}=0,\chi_3=0} \tag*{(44)}
ϕ33∣d23=0,χ3=0=ϕ22∣d23=0,χ3=0(45)\left.\phi_{33}\right|_{d_{23}=0,\chi_3=0}=\left.\phi_{22}\right|_{d_{23}=0,\chi_3=0} \tag*{(45)}

These are consistent with the bilayer results.

  • n stacked 2D materials

We follow the same strategy to solve the n-layer system. For example, when the source charge is in the bottom layer, we solve

ψz→∞=Mψz→−∞=L(χn,1,ϵt)P(d(n−1)n,1)⋯[L(χ1,ϵb,1)ψz→−∞+S′](A0)=(M11M12M21M22)(0F)+(S~1S~2)(46)\begin{aligned} \psi_{z\to\infty} &= \mathcal{M}\psi_{z\to-\infty} \\ &= L(\chi_n,1,\epsilon_t)P(d_{(n-1)n},1)\cdots\left[L(\chi_1,\epsilon_b,1)\psi_{z\to-\infty}+S'\right] \\ \begin{pmatrix} A \\ 0 \end{pmatrix} &= \begin{pmatrix} M_{11} & M_{12} \\ M_{21} & M_{22} \end{pmatrix}\begin{pmatrix} 0 \\ F \end{pmatrix}+\begin{pmatrix} \widetilde{S}_1 \\ \widetilde{S}_2 \end{pmatrix} \tag*{(46)} \end{aligned}

We extract F=−S~2M22F=-\frac{\widetilde{S}_2}{M_{22}}. As a sanity check, we verify that the left-hand side and right-hand side match after obtaining the solution. We propagate the solution, FF to obtain ϕ11,⋯ϕ1n\phi_{11},\cdots\phi_{1n}. And very similar to bilayer and trilayer system, we compute ϕij\phi_{ij} where i,j=1,…,ni,j=1,\ldots,n.

  • Monolayer 2D material

A special case of the transfer matrix method is the monolayer Keldysh potential. Using the same strategy as above, we set n=1n=1. The potential becomes

ϕ11=−Cq(2r1q+ϵb+ϵt)(47)\phi_{11}=-\frac{C}{q(2r_1q+\epsilon_b+\epsilon_t)} \tag*{(47)}

This is the well-known Rytova-Keldysh potential in momentum space[68, 69, 70].

K-point convergence of dipolar and quadrupolar excitons

Image not extracted

Figure 7. Convergence of the lowest-energy excitons in lattice-matched WSe2_2/WS2_2 and WSe2_2/WS2_2/WSe2_2 van der Waals heterostructures with a lattice constant of 0.325 nm. We find that the 1s dipolar and quadrupolar excitons converge with a 45 ×\times 45 kk-point grid, while the 2s excitons require a 78 ×\times 78 grid for convergence. We do not apply any rigid energy shifts. In constructing the BSE Hamiltonian, we include 2 valence and 2 conduction bands for the bilayer, and 3 valence and 3 conduction bands for the trilayer. We find that the oscillator strength of both dipolar and quadrupolar excitons is ∼102\sim10^2 times weaker than that of intralayer excitons in WSe2_2 or WS2_2. The inset shows a plot zoomed in by a factor of 500.

Electrostatic potential for excitons I and II

Exciton I is the collinear arrangement +q/2+q/2, −q-q, +q/2+q/2 along zz. Exciton II replaces the central −q-q by three charges −q/3-q/3 on an equilateral triangle of side bb in the z=0z=0 plane. For exciton I, with k=1/4πε0k=1/4\pi\varepsilon_0 and cylindrical coordinates (ρ,z)(\rho,z), superposing the Coulomb potentials gives

V(ρ,z)=k[q/2ρ2+(z−2d)2−qρ2+(z−d)2+q/2ρ2+z2].(48)V(\rho,z)=k\left[\frac{q/2}{\sqrt{\rho^2+(z-2d)^2}}-\frac{q}{\sqrt{\rho^2+(z-d)^2}}+\frac{q/2}{\sqrt{\rho^2+z^2}}\right]. \tag*{(48)}

We characterize both distributions through their multipole moments. The Cartesian quadrupole tensor, with aa labeling the point charges, is defined as

Qij=∑aqa(3xa,ixa,j−ra2δij),(49)Q_{ij}=\sum_a q_a\left(3x_{a,i}x_{a,j}-r_a^2\delta_{ij}\right), \tag*{(49)}

and is traceless by construction (Qii=0Q_{ii}=0). Equivalently, the moments are organized by the spherical multipole expansion,

qlm=∑aqaralYlm∗(θa,ϕa),(50)q_{lm}=\sum_a q_a r_a^l Y_{lm}^{*}(\theta_a,\phi_a), \tag*{(50)}

with ll the multipole order and mm the azimuthal index (the quadrupole is l=2l=2, the octupole l=3l=3), and Ylm(θ,ϕ)∝Plm(cos⁡θ)eimϕY_{lm}(\theta,\phi)\propto P_l^m(\cos\theta)e^{im\phi}, where PlmP_l^m are the associated Legendre polynomials.

Exciton I. Measuring positions from the central −q-q at z=dz=d (za′=za−dz'_a=z_a-d), the lower moments vanish:

∑aqa=q2−q+q2=0,(51)\sum_a q_a=\frac{q}{2}-q+\frac{q}{2}=0, \tag*{(51)}
∑aqaza′=q2(+d)−q(0)+q2(−d)=0.(52)\sum_a q_a z'_a=\frac{q}{2}(+d)-q(0)+\frac{q}{2}(-d)=0. \tag*{(52)}

The charges lie on the axis, so Qzz=∑aqa(3za′2−ra′2)=2∑aqaza′2=2qd2Q_{zz}=\sum_a q_a(3z_a^{\prime2}-r_a^{\prime2})=2\sum_a q_a z_a^{\prime2}=2qd^2, and tracelessness yields

Qxx=Qyy=−qd2,Qzz=2qd2,(53)Q_{xx}=Q_{yy}=-qd^2,\qquad Q_{zz}=2qd^2, \tag*{(53)}

so Qxx−Qyy=0Q_{xx}-Q_{yy}=0. Thus, q22∝(Qxx−2iQxy−Qyy)=0q_{22}\propto(Q_{xx}-2iQ_{xy}-Q_{yy})=0, and q21∝(Qxz−iQyz)=0q_{21}\propto(Q_{xz}-iQ_{yz})=0. Also, note that ql,−m=(−1)mqlm∗q_{l,-m}=(-1)^m q_{lm}^{*}.[71]

Exciton II. Centering at the origin (the midpoint of the hole charges and the centroid of the triangle), the in-plane charges sit at circumradius R=b/3R=b/\sqrt{3},

r1′=(0,R,0),r2′=(−32R,−12R,0),r3′=(32R,−12R,0).(54)\mathbf{r}'_1=(0,R,0),\qquad\mathbf{r}'_2=\left(-\frac{\sqrt{3}}{2}R,-\frac{1}{2}R,0\right),\qquad\mathbf{r}'_3=\left(\frac{\sqrt{3}}{2}R,-\frac{1}{2}R,0\right). \tag*{(54)}

The monopole and dipole again vanish:

∑aqa=q2+q2+3(−q3)=0,(55)\sum_a q_a = \frac{q}{2} + \frac{q}{2} + 3\left(-\frac{q}{3}\right) = 0, \tag*{(55)}
∑aqara′=q2(0,0,d)+q2(0,0,−d)−q3∑iri′=0.(56)\sum_a q_a \mathbf{r}'_a = \frac{q}{2}(0,0,d) + \frac{q}{2}(0,0,-d) - \frac{q}{3}\sum_i \mathbf{r}'_i = 0. \tag*{(56)}

The positive charges each contribute q2(2d2)=qd2\frac{q}{2}(2d^2) = qd^2 to QzzQ_{zz}, while each in-plane charge contributes −q3(−R2)=q3R2-\frac{q}{3}(-R^2) = \frac{q}{3}R^2, so

Qzz=2qd2+qR2=2qd2+13qb2,(57)Q_{zz} = 2qd^2 + qR^2 = 2qd^2 + \frac{1}{3}qb^2, \tag*{(57)}

giving Qxx=Qyy=−qd2−16qb2Q_{xx} = Q_{yy} = -qd^2 - \frac{1}{6}qb^2 and again Qxx−Qyy=0Q_{xx} - Q_{yy} = 0. As a result, ql,mq_{l,m} remains unchanged.

Octupole

The two distributions first differ at the octupole (l=3l = 3). For exciton I, all charges lie on the axis (θa=0,π\theta_a = 0,\pi), where Y3m=0Y_{3m} = 0 for m≠0m \ne0; the only possible component again vanishes,

q3,0∝∑aqaza3=q2d3−q(0)+q2(−d)3=0,(58)q_{3,0} \propto\sum_a q_a z_a^3 = \frac{q}{2}d^3 - q(0) + \frac{q}{2}(-d)^3 = 0, \tag*{(58)}

as enforced by the inversion center, which forbids all odd-ll moments. For exciton II, the positive charges again give no octupole, while the triangle (θa=π2\theta_a = \frac{\pi}{2}, ra=Rr_a = R, ϕa=ϕ0+2πn3\phi_a = \phi_0 + \frac{2\pi n}{3}) gives

q3,m∝−q3R3P3m(0)e−imϕ0∑n=02e−im2πn/3.(59)q_{3,m} \propto-\frac{q}{3}R^3 P^m_3(0)e^{-im\phi_0}\sum_{n=0}^{2}e^{-im2\pi n/3}. \tag*{(59)}

The azimuthal sum imposes a threefold selection rule, ∑n=02e−im2πn/3=3δm≡0  (mod  3)\sum_{n=0}^{2}e^{-im2\pi n/3} = 3\delta_{m\equiv0\;(\mathrm{mod}\;3)}, leaving m∈{0,±3}m \in\{0,\pm3\}; since P3(0)=0P_3(0) = 0 and P33(0)=−15≠0P^3_3(0) = -15 \ne0, only

q3,±3∝qR3∝qb3≠0(60)q_{3,\pm3} \propto qR^3 \propto qb^3 \ne0 \tag*{(60)}

remains. Exciton II thus carries a threefold in-plane octupole, absent in exciton I.

Structural relaxations in a 2.2∘2.2^\circ twisted WSe2/WS2/WSe2\mathrm{WSe}_2/\mathrm{WS}_2/\mathrm{WSe}_2 trilayer

Image not extracted

Figure 8. Large atomic relaxations in a 2.2∘2.2^\circ twisted WSe2/WS2/WSe2\mathrm{WSe}_2/\mathrm{WS}_2/\mathrm{WSe}_2 trilayer. Arrows show the displacement of W atoms in each layer relative to the rigid, unrelaxed structure. Displacements of W atoms in the middle WS2\mathrm{WS}_2 layer are amplified.

Electrostatic energies of dipolar and quadrupolar excitons on different lattice geometries

Electrostatic energies of dipolar and quadrupolar excitons on different lattice geometries

Figure 9. We consider three electrostatic energies: EsymQE_{\mathrm{sym}}^{Q}, quadrupolar excitons on a triangular lattice; EstgDE_{\mathrm{stg}}^{D} (square), staggered dipolar excitons on a square lattice without frustration; and EstgDE_{\mathrm{stg}}^{D} (triangular), staggered dipolar excitons on a triangular lattice with minimal frustration.

References

References

  1. [1]A. D. Buckingham, “Permanent and induced molecular moments and long-range intermolecular forces,” in Advances in Chemical Physics (John Wiley & Sons, Ltd, 1967) pp. 107–142.
  2. [2]K. S. Thorne, Rev. Mod. Phys. 52, 299 (1980).
  3. [3]M.-T. Suzuki, H. Ikeda, and P. M. Oppeneer, Journal of the Physical Society of Japan 87, 041008 (2018).
  4. [4]P. Santini, S. Carretta, G. Amoretti, R. Caciuffo, N. Magnani, and G. H. Lander, Rev. Mod. Phys. 81, 807 (2009).
  5. [5]L. Yu, K. Pistunova, J. Hu, K. Watanabe, T. Taniguchi, and T. F. Heinz, Nature Materials 22, 1485 (2023).
  6. [6]Z. Lian, D. Chen, L. Ma, Y. Meng, Y. Su, L. Yan, X. Huang, Q. Wu, X. Chen, M. Blei, T. Taniguchi, K. Watanabe, S. Tongay, C. Zhang, Y.-T. Cui, and S.-F. Shi, Nature Communications 14, 4604 (2023).
  7. [7]W. Li, Z. Hadjri, L. M. Devenica, J. Zhang, S. Liu, J. Hone, K. Watanabe, T. Taniguchi, A. Rubio, and A. Srivastava, Nature Materials 22, 1478 (2023).
  8. [8]Y. Bai, Y. Li, S. Liu, Y. Guo, J. Pack, J. Wang, C. R. Dean, J. Hone, and X. Zhu, Nano Letters 23, 11621 (2023).
  9. [9]Y. Xie, Y. Gao, F. Chen, Y. Wang, J. Mao, Q. Liu, S. Chu, H. Yang, Y. Ye, Q. Gong, J. Feng, and Y. Gao, Phys. Rev. Lett. 131, 186901 (2023).
  10. [10]M. Chen, R. Li, H. Wang, Y. Yang, Y. Lai, C. Hu, T. Taniguchi, K. Watanabe, J. Yan, J.-H. Chu, E. Henriksen, C. Zhang, L. Yang, and X. Wang, Nature Communications 16, 10359 (2025).
  11. [11]J. Jasiński, J. Hagel, S. Brem, E. Wietek, T. Taniguchi, K. Watanabe, A. Chernikov, N. Bruyant, M. Dyksik, A. Surrente, M. Baranowski, D. K. Maude, E. Malic, and P. Plochocka, Nature Communications 16, 1382 (2025).
  12. [12]Y. Xie, F. Chen, Y. Gao, Y. Wang, J. Mao, Q. Liu, S. Chu, H. Yang, Y. Ye, Q. Gong, J. Feng, and Y. Gao, Phys. Rev. B 110, L201402 (2024).
  13. [13]S. Feng, A. J. Campbell, B. M. Francis, H. Baek, T. Taniguchi, K. Watanabe, I. C. Gerber, B. D. Gerardot, and M. Brotons-Gisbert, Phys. Rev. Lett. 134, 166901 (2025).
  14. [14]Z. Zhang and S. Wang, The Journal of Physical Chemistry Letters 15, 10104 (2024).
  15. [15]J. Sung, Y. Zhou, G. Scuri, V. Zólyomi, T. I. Andersen, H. Yoo, D. S. Wild, A. Y. Joe, R. J. Gelly, H. Heo, S. J. Magorrian, D. Bérubé, A. M. M. Valdivia, T. Taniguchi, K. Watanabe, M. D. Lukin, P. Kim, V. I. Fal’ko, and H. Park, Nature Nanotechnology 15, 750 (2020).
  16. [16]Z. Sun, H. Sun, X. Jia, A. Li, N. J. Zhang, K. Seungmin Hong, J. DePinho, C. Y. Long, K. Watanabe, T. Taniguchi, O. Chen, J. Wang, J. Li, B. Rubenstein, and Y. Bai, arXiv e-prints , arXiv:2602.00419 (2026), arXiv:2602.00419 [cond-mat.mtrl-sci].arxiv.org/abs/2602.00419
  17. [17]S. Gim, H. Jo, S. Kim, S. M. Yang, K. Watanabe, T. Taniguchi, and H. Baek, ACS Photonics 12, 6957 (2025).
  18. [18]J. Zhu, T. Liang, F. Shen, Z. Chen, and J. Xu, Nature Communications 16, 10661 (2025).
  19. [19]B. Xu, J. Fu, L. Huang, C. Ge, Z. Xu, W. Zheng, Q. Deng, S. Xie, Q. Tong, D. Li, Y. Jiang, H. Yuan, and A. Pan, Nature Communications 16, 3938 (2025).
  20. [20]Z.-H. Guo, T. Yan, J.-Z. Zhao, Y.-J. Jin, and Q. Zhu, arXiv preprint arXiv:2504.10189 (2025).arxiv.org/abs/2504.10189
  21. [21]Y. Meng, L. Ma, L. Yan, A. Khalifa, D. Chen, S. Zhang, R. Banerjee, T. Taniguchi, K. Watanabe, S. A. Tongay, B. Hunt, S.-Z. Lin, W. Yao, Y.-T. Cui, S. Chatterjee, and S.-F. Shi, Nature Photonics 19, 1219 (2025).
  22. [22]H. Liu, S. Chen, H. Xu, R. Han, Z. Sun, S. Huang, X. Zhang, J. Luo, and D. Liu, Phys. Rev. Lett. 135, 206302 (2025).
  23. [23]H. Zheng, B. Wu, S. Li, J. Ding, J. He, Z. Liu, C.-T. Wang, J.-T. Wang, A. Pan, and Y. Liu, Light: Science & Applications 12, 117 (2023).
  24. [24]E. Barré, M. Dandu, and A. Raja, Nature Materials 22, 1437 (2023).
  25. [25]X. Blase, I. Duchemin, D. Jacquemin, and P.-F. Loos, The Journal of Physical Chemistry Letters 11, 7371 (2020).
  26. [26]M. Rohlfing and S. G. Louie, Phys. Rev. B 62, 4927 (2000).
  27. [27]G. Onida, L. Reining, and A. Rubio, Rev. Mod. Phys. 74, 601 (2002).
  28. [28]S. G. Louie, Y.-H. Chan, F. H. da Jornada, Z. Li, and D. Y. Qiu, Nature Materials 20, 728 (2021).
  29. [29]T. Deilmann and K. Sommer Thygesen, 2D Materials 11, 035032 (2024).
  30. [30]Y. Slobodkin, Y. Mazuz-Harpaz, S. Refaely-Abramson, S. Gazit, H. Steinberg, and R. Rapaport, Phys. Rev. Lett. 125, 255301 (2020).
  31. [31]Z. Zhang and S. Wang, The Journal of Physical Chemistry C 129, 15843 (2025).
  32. [32]L.-A. Qin, Y. Wang, Y. Xu, D. Sun, Y. Dai, B. Huang, and W. Wei, Nano Letters 25, 16123 (2025).
  33. [33]M. H. Naik, E. C. Regan, Z. Zhang, Y.-H. Chan, Z. Li, D. Wang, Y. Yoon, C. S. Ong, W. Zhao, S. Zhao, M. I. B. Utama, B. Gao, X. Wei, M. Sayyad, K. Yumigeta, K. Watanabe, T. Taniguchi, S. Tongay, F. H. da Jornada, F. Wang, and S. G. Louie, Nature 609, 52 (2022).
  34. [34]S. Susarla, M. H. Naik, D. D. Blach, J. Zipfel, T. Taniguchi, K. Watanabe, L. Huang, R. Ramesh, F. H. da Jornada, S. G. Louie, P. Ercius, and A. Raja, Science 378, 1235 (2022).
  35. [35]I. Maity, J. Lischner, A. A. Mostofi, and A. Rubio, Nano Letters 26, 1349 (2026).
  36. [36]R. Dalal, H. Singh, R. Dutta, H. Swaminathan, K. Watanabe, T. Taniguchi, M. H. Naik, M. Jain, and A. Singh, arXiv preprint arXiv:2601.03045 (2026).arxiv.org/abs/2601.03045
  37. [37]C. Jin, E. C. Regan, A. Yan, M. Iqbal Bakti Utama, D. Wang, S. Zhao, Y. Qin, S. Yang, Z. Zheng, S. Shi, K. Watanabe, T. Taniguchi, S. Tongay, A. Zettl, and F. Wang, Nature 567, 76 (2019).
  38. [38]R. Wang, K. Chang, W. Duan, Y. Xu, and P. Tang, Phys. Rev. Lett. 134, 026904 (2025).
  39. [39]G. E. Astrakharchik, I. L. Kurbakov, D. V. Sychev, A. K. Fedorov, and Y. E. Lozovik, Phys. Rev. B 103, L140101 (2021).
  40. [40]I. Maity, A. A. Mostofi, and J. Lischner, npj 2D Materials and Applications 9, 20 (2025).
  41. [41]L. S. R. Cavalcante, A. Chaves, B. Van Duppen, F. M. Peeters, and D. R. Reichman, Phys. Rev. B 97, 125427 (2018).
  42. [42]M. Danovich, D. A. Ruiz-Tijerina, R. J. Hunt, M. Szyniszewski, N. D. Drummond, and V. I. Fal’ko, Phys. Rev. B 97, 195452 (2018).
  43. [43]S. Ovesen, S. Brem, C. Linderälv, M. Kuisma, T. Korn, P. Erhart, M. Selig, and E. Malic, Communications Physics 2, 23 (2019).
  44. [44]M. N. Brunetti, O. L. Berman, and R. Y. Kezerashvili, Phys. Rev. B 98, 125406 (2018).
  45. [45]M. Van der Donck and F. M. Peeters, Phys. Rev. B 98, 115104 (2018).
  46. [46]T. C. Berkelbach, M. S. Hybertsen, and D. R. Reichman, Phys. Rev. B 88, 045318 (2013).
  47. [47]M. Palummo, M. Bernardi, and J. C. Grossman, Nano Letters 15, 2794 (2015).
  48. [48]R. Moessner and A. P. Ramirez, Physics Today 59, 24 (2006).
  49. [49]G. Beckett, J. Beech-Brandt, K. Leach, Z. Payne, A. Simpson, L. Smith, A. Turner, and A. Whiting, “ARCHER2 service description,” (2024).DOI
  50. [50]S. Naik, M. H. Naik, I. Maity, and M. Jain, Computer Physics Communications 271, 108184 (2022).
  51. [51]“Lammps,” Accessed: December 16, 2025.
  52. [52]A. P. Thompson, H. M. Aktulga, R. Berger, D. S. Bolintineanu, W. M. Brown, P. S. Crozier, P. J. in ’t Veld, A. Kohlmeyer, S. G. Moore, T. D. Nguyen, R. Shan, M. J. Stevens, J. Tranchida, C. Trott, and S. J. Plimpton, Comp. Phys. Comm. 271, 108171 (2022).
  53. [53]J.-W. Jiang and Y.-P. Zhou, Handbook of Stillinger-Weber Potential Parameters for Two-Dimensional Atomic Crystals (IntechOpen, Rijeka, 2017).DOI
  54. [54]M. H. Naik, I. Maity, P. K. Maiti, and M. Jain, The Journal of Physical Chemistry C 123, 9770 (2019).
  55. [55]I. Maity, P. K. Maiti, and M. Jain, Phys. Rev. B 97, 161406 (2018).
  56. [56]J. M. Soler, E. Artacho, J. D. Gale, A. García, J. Junquera, P. Ordejón, and D. Sánchez-Portal, Journal of Physics: Condensed Matter 14, 2745 (2002).
  57. [57]N. Troullier and J. L. Martins, Phys. Rev. B 43, 1993 (1991).
  58. [58]J. P. Perdew and A. Zunger, Phys. Rev. B 23, 5048 (1981).
  59. [59]Fernandez-Seivane, L. and Oliveira, M. A. and Sanvito, S. and Ferrer, J., Journal of Physics: Condensed Matter 18 (2006), 10.1088/0953-8984/19/19/489001.
  60. [60]N. Marzari, A. A. Mostofi, J. R. Yates, I. Souza, and D. Vanderbilt, Rev. Mod. Phys. 84, 1419 (2012).
  61. [61]G. Pizzi, V. Vitale, R. Arita, S. Blügel, F. Freimuth, G. Géranton, M. Gibertini, D. Gresch, C. Johnson, T. Koretsune, J. Ibañez-Azpiroz, H. Lee, J.-M. Lihm, D. Marchand, A. Marrazzo, Y. Mokrousov, J. I. Mustafa, Y. Nohara, Y. Nomura, L. Paulatto, S. Poncé, T. Ponweiser, J. Qiao, F. Thöle, S. S. Tsirkin, M. Wierzbowska, N. Marzari, D. Vanderbilt, I. Souza, A. A. Mostofi, and J. R. Yates, Journal of Physics: Condensed Matter 32, 165902 (2020).
  62. [62]I. Souza, N. Marzari, and D. Vanderbilt, Phys. Rev. B 65, 035109 (2001).
  63. [63]A. Kormányos, G. Burkard, M. Gmitra, J. Fabian, V. Zólyomi, N. D. Drummond, and V. Fal'ko, 2D Materials 2, 022001 (2015).
  64. [64]T. Auckenthaler, V. Blum, H.-J. Bungartz, T. Huckle, R. Johanni, L. Krämer, B. Lang, H. Lederer, and P. Willems, Parallel Computing 37, 783 (2011), 6th International Workshop on Parallel Matrix Algorithms and Applications (PMAA’10).
  65. [65]A. Marek, V. Blum, R. Johanni, V. Havu, B. Lang, T. Auckenthaler, A. Heinecke, H.-J. Bungartz, and H. Lederer, Journal of Physics: Condensed Matter 26, 213201 (2014).
  66. [66]I. Maity, A. A. Mostofi, and J. Lischner, Nano Letters 23, 4870 (2023).
  67. [67]A. Rossi, J. Zipfel, I. Maity, M. Lorenzon, M. Dandu, E. Barré, L. Francaviglia, E. C. Regan, Z. Zhang, J. H. Nie, E. S. Barnard, K. Watanabe, T. Taniguchi, E. Rotenberg, F. Wang, J. Lischner, A. Raja, and A. Weber-Bargioni, ACS Nano 18, 18202 (2024).
  68. [68]N. S. Rytova, arXiv preprint arXiv:1806.00976 (2018).arxiv.org/abs/1806.00976
  69. [69]L. V. Keldysh, Soviet Journal of Experimental and Theoretical Physics Letters 29, 658 (1979).
  70. [70]P. Cudazzo, I. V. Tokatly, and A. Rubio, Phys. Rev. B 84, 085406 (2011).
  71. [71]J. D. Jackson, Classical Electrodynamics (Wiley, 1998).

Paper details

vlnc:W_zcva56k2
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 8

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents