betaFeedback

WorksPhysics and Astronomy

Quantum simulation of honeycomb lattice model by high-order moiré pattern

Abstract

Moiré superlattices have become an emergent solid-state platform for simulating quantum lattice models. However, in single moiré device, Hamiltonians parameters like lattice constant, hopping and interaction terms can hardly be manipulated, limiting the controllability and accessibility of moire quantum simulator. Here, by combining angle-resolved photoemission spectroscopy and theoretical analysis, we demonstrate that high-order moiré patterns in graphene-monolayered xenon/krypton heterostructures can simulate honeycomb model in mesoscale, with in-situ tunable Hamiltonians parameters. The length scale of simulated lattice constant can be tuned by annealing processes, which in-situ adjusts intervalley interaction and hopping parameters in the simulated honeycomb lattice. The sign of the lattice constant can be switched by choosing xenon or krypton monolayer deposited on graphene, which controls sublattice degree of freedom and valley arrangment of Dirac fermions. Our work establishes a novel path for experimentally simulating the honeycomb model with tunable parameters by high-order moiré patterns.

§ There authors contributed equally.

* E-mail: [email protected]; [email protected]

INTRODUCTION

Moiré superlattices provide not only a rich playground for the exploration of novel quantum phases [1, 2, 3, 4, 5, 6], but also a new quantum simulator of solid state lattice models [7, 8, 9]. The flat bands in magic-angle twisted bilayer graphene possess a fragile topology, which obstructs deriving a simple tight-binding model [10, 11]. However, recent theoretical progress revealed that twisted bilayer graphene effectively realizes a topological heavy fermion model with coexisting localized and itinerant electrons [12].

Moiré superlattices formed by transition metal dichalcogenides (TMD) were theoretically proposed to simulate quantum models on triangular lattices [7, 8] or honeycomb lattices [13, 14], depending on the exact material combinations. Experimental discoveries of Mott insulators [15, 16], generalized Wigner crystal states [17, 18], and topological states [19] in TMD-based moiré superlattices strongly support the feasibility of quantum simulation.

Although electron filling can be continuously tuned by electric gating, Hamiltonian parameters like lattice constant, single-particle hopping, and many-body interaction strength cannot be easily manipulated in moiré devices formed between two layers with similar lattice constants and small twist angles [20, 21, 22, 23, 24], limiting the controllability and accessibility of moiré quantum simulator.

Recently, high-order moiré patterns were realized in graphene-based heterostructures with large lattice constant mismatch Δa\Delta a, including graphene on SiC and graphene-monolayered xenon heterostructure (mXe/G) [25, 26, 27, 28]. Strikingly, the high-order moiré pattern in mXe/G can be in situ tuned by controlling the lattice constant of the xenon monolayer through annealing, which further generates movable Dirac cone replicas in the electronic structure as observed by angle-resolved photoemission spectroscopy (ARPES) [26]. An open question is whether high-order moiré pattern can provide a new route to realize a quantum simulator of model Hamiltonians.

Here, by combining ARPES experiment and theoretical study, we unveil that the high-order moiré pattern in graphene-noble gas monolayer heterostructure can be a good experimental platform to simulate mesoscale honeycomb lattice, with not only tunable magnitude but also a switchable sign of lattice constant. The magnitude of the honeycomb lattice constant, equivalent to the moiré period am′a'_m, can be in situ tuned by annealing processes, which determines the hopping term and intervalley interaction strength in the simulated mesoscale honeycomb lattice. By choosing Xe or Kr monolayer deposited on graphene, the sign of am′a'_m can also be changed with opposite sublattice arrangement, which is directly evidenced by the different locations of moiré Brillouin zone (BZ) near the K point and the opposite valley arrangement of the Dirac cones in mXe/G and graphene-Kr monolayer heterostructure (mKr/G) as observed in ARPES measurements. These experimental observations are consistent with our theoretical analysis, which indicates different sublattice arrangements for the two systems. By varying annealing temperature, the magnitude of am′a'_m can be in situ tuned, which controls the intervalley coupling strength and the hopping parameters in the tight-binding model on the honeycomb lattice. Our work demonstrates that the high-order moiré pattern in mXe/G and mKr/G is a promising experimental platform to simulate a mesoscale honeycomb model with in situ tunable parameters.

Experimental Details

Monolayer graphene was grown by a high-temperature annealing process [29] of n-type 6H-SiC(0001) from PrMat. Kr/Xe monolayers were grown on monolayer graphene at T=30T = 30 K with Kr/Xe pressure of 1×10−91 \times10^{-9} Torr, for a duration of one hour to deposit rare gases. Subsequently, the samples were gradually heated at a rate of 0.20.2 K/min to reach the corresponding annealing temperatures. The ARPES measurements were performed at lab-based ARPES facility utilizing He lamp as the photon source (He Iα: 21.221.2 eV). We use the same graphene sample and measurement geometry in this work to avoid the difference from effect of matrix element for the direct comparison between results of mXe/G and mKr/G.

Results

Firstly, we compare the high-order moiré patterns in mXe/G and mKr/G, as sketched in Figure 1. The lattice constants of Xe monolayer (aXea_{Xe}) and Kr monolayer (aKra_{Kr}) are slightly larger [Figure 1(b)-I] and smaller [Figure 1(e)-I] than that of graphene 3×3\sqrt{3} \times\sqrt{3} supercell 3aGra\sqrt{3}a_{Gra}, respectively. The moiré period of mXe/G and mKr/G can be defined in a unified manner as

Image not extracted

Figure 1. High-order moiré patterns in mXe/G and mKr/G. (a) Large-scale atom arrangements showing moiré pattern in mXe/G, with aXe<3aGraa_{Xe} < \sqrt{3}a_{Gra}. (b) Zoom-in areas at the center (I) and corners (II–III) of moiré pattern in mXe/G. (c) Schematic high-order moiré effect on band structure modulation in mXe/G. (d) Same as (a), but for mKr/G with aKr>3aGraa_{Kr} > \sqrt{3}a_{Gra}. (e) Same as (b), but for mKr/G. (f) Same as (c), but for mKr/G. (g, h) Fermi surface maps of mXe/G with Ta_a = 67 K and mKr/G at Ta_a = 37 K, respectively. (i, j) Constant energy intensity plots at different EB_B of mXe/G and mKr/G, respectively. The dashed lines of different colors indicate Dirac cone replicas from different valleys.

am′=aGraaXe(Kr)aXe(Kr)−3aGra(1)a_{m}^{\prime} = \frac{a_{Gra}a_{Xe(Kr)}}{a_{Xe(Kr)}-\sqrt{3}a_{Gra}} \tag*{(1)}

With this definition, there is a sign difference in the moiré period am′a_{m}^{\prime} between mXe/G and mKr/G, which leads to different stacking configurations in mXe/G and mKr/G. The distinction is particularly pronounced at AM_{\mathrm{M}} and BM_{\mathrm{M}} sites, which are two inequivalent high-symmetry positions in the moiré pattern. At the AM_{\mathrm{M}} sites, the Xe atom overlaps with the B-sublattice of graphene in mXe/G [Figure 1(b)-II], but the Kr atom overlaps with the A-sublattice in mKr/G [Figure 1(e)-II]. On the other hand, at the BM_{\mathrm{M}} sites, the Xe atom overlaps with the A-sublattice of graphene in mXe/G [Figure 1(b)-III], but the Kr atom overlaps with the B-sublattice in mKr/G [Figure 1(e)-III]. The stacking patterns between graphene and noble gas monolayers are experimentally deteremined by Low Energy Electron Diffraction (LEED) measurements (Fig. S1 in the Supplemental Material [30]).

The sign difference of am′a_{m}^{\prime} also has a significant impact on the electronic structure modulation in mXe/G and mKr/G. In mXe/G, the Dirac replicas in the moiré BZ have the same valley arrangement as that of graphene [Figure 1(c)], as the reciprocal lattice vector of Xe monolayer is less in magnitude than the Γ\Gamma-K vector of graphene (∣G⃗Xe∣≤∣K⃗Gra∣|\vec{G}_{Xe}| \leq|\vec{K}_{Gra}|). For the negative am′a_{m}^{\prime} moiré pattern in mKr/G, the Dirac replicas have an opposite valley arrangement [Figure 1(f)] compared to that of both bare graphene and mXe/G, because of ∣G⃗Kr∣≥∣K⃗Gra∣|\vec{G}_{Kr}| \geq|\vec{K}_{Gra}|. For the same reason, the replicas near the K point locate outside the graphene first BZ in mKr/G [Fig. 1(f)], in contrast to the case of mXe/G [Fig. 1(c)].

The effect of the am′a'_{m} sign difference is directly observed by ARPES measurements, as both mXe/G and mKr/G possess an identical graphene layer and maintain a consistent geometric relationship with the helium lamp. Figure 1(g)-(h) present the Fermi surface (FS) maps of mXe/G and mKr/G, respectively, with aXe∼4.50a_{Xe} \sim4.50 Å and aKr∼4.04a_{Kr} \sim4.04 Å, exhibiting similar ∣am′∣∼4.600\lvert a'_{m} \rvert\sim4.600 nm. The most pronounced difference is that Dirac cone replicas near the K point sit inside the graphene BZ in mXe/G, but appear outside the graphene BZ in mKr/G, fully consistent with the discussion in Figure 1(c) and 1(f).

The different valley arrangement induced by the am′a'_{m} sign difference in mXe/G and mKr/G is also demonstrated by the ARPES results on the Dirac cone replicas near the Γ\Gamma point. Away from the Dirac point, the original graphene Dirac cone shows a valley-contrast trigonal warping. Therefore, the different valley arrangements of Dirac cone replicas in the moiré BZ in mXe/G and mKr/G induce obvious differences in the dispersions at high EBE_{\mathrm{B}}, as clearly seen from constant energy plots in ARPES experiments [Fig. 1(i)] and 1(j). For mXe/G, the constant energy plots show a flowerlike pattern at high EBE_{\mathrm{B}} in Figure 1(i). In contrast, mKr/G shows a wheel-like pattern for constant energy plots with the same EBE_{\mathrm{B}} in Figure 1(j). The differences are consistent with the different valley arrangement of the warping Dirac cones as indicated by guiding dashed lines in Figure 1(i) and 1(j).

The moiré pattern period in mXe/G and mKr/G can be in situ tuned, which covers a wide range in both positive and negative directions. As pointed out by our previous work [26], the moiré BZ in mXe/G shrinks by decreasing the annealing temperature TaT_{\mathrm{a}} [Fig. 2(a)], indicating an increase of moiré period am′a'_{m} in mesoscale. The mechanism is that the lattice constant of the Xe monolayer aXea_{Xe} becomes smaller and approaches the value of 3aGra\sqrt{3}a_{\mathrm{Gra}} by decreasing TaT_{\mathrm{a}}. According to Eq. (1), the moiré period am′a'_{m} is positive and increases with lowering TaT_{\mathrm{a}}. Correspondingly, the moiré BZ shrinks [Fig. 2(a)] and the Dirac replicas located at the opposite moiré BZ corners κm\kappa_m and κm′\kappa'_m move closer to each other in mXe/G at lower TaT_{\mathrm{a}}, as clearly presented by the band dispersion evolution along the momentum path #1 [Fig. 2(b)]. When aXea_{Xe} is equal to 3aGra\sqrt{3}a_{Gra} at Ta=51.5T_{\mathrm{a}} = 51.5 K, the moiré pattern evolves into the commensurate Kekulé distortion [Fig. 2(a)-IV and 2(b)-IV].

On the other hand, in the case of mKr/G, the Kr monolayer also has smaller lattice constants aKra_{Kr} with decreasing TaT_{\mathrm{a}}. However, in the range of aKr≤3aGraa_{Kr} \leq\sqrt{3}a_{Gra}, the smaller aKra_{Kr} results in a negative am′a'_m with a smaller absolute value ∣am′∣|a'_m|, according to Eq. (1). Correspondingly, the Dirac replicas at κm\kappa_m and κm′\kappa'_m move away from each other in mKr/G with lowering TaT_{\mathrm{a}} [Fig. 2(b)], resulting in an expanding moiré BZ with the opposite valley arrangements compared to that in mXe/G and original graphene [Fig. 2(a)].

The distinction between moiré patterns in mXe/G and mKr/G can be better visualized in TaT_{\mathrm{a}} dependent photoemission data at the Γ\Gamma point. While the double peaks become close to each other with increasing TaT_{\mathrm{a}} in mXe/G [Fig. 2(c)], mKr/G shows the opposite behavior [Fig. 2(d)]. For Kekulé distortion as Dirac replicas overlap with each other, the spectra weight at the Dirac point is suppressed and a double-peaks line shape with an energy separation of 0.48 eV is observed at the Dirac point in both mXe/G and mKr/G, in contrast to the single-peak profile in original Dirac cone observed in pristine graphene [Fig. 2(e)]. Although the gap size at Dirac point can be overestimated in ARPES measurements [28], our results clearly moiré modulation effect on Dirac band structure in mXe/G and mKr/G.

Moiré period am′a'_m can also be tuned by different annealing pressure of noble gas. By increasing the Kr pressure from 1.33×10−91.33 \times10^{-9} torr to 6.65×10−86.65 \times10^{-8} torr at Ta=46T_{\mathrm{a}} = 46 K [Fig. 2(f)], mKr/G evolves from Kekulé distortion (∣am′∣→∞|a'_m| \to\infty) to ∣am′∣=4.81|a'_m| = 4.81 nm [Fig. 2(g)]. The moiré pattern with a specific value am′a^{\prime}_m can be achieved at high temperature, by increasing Kr pressure. Therefore, ultra-low temperature and ultra-high vacuum conditions are in fact not necessary for achieving moiré pattern in mXe/G and mKr/G.

After the annealing processes, we can keep the moiré pattern at low temperature by cooling the system in an environment without noble gas partial pressure. At the lowest temperature we can achieve in our measurements (T = 6 K), we manage to achieve moiré pattern with ∣am′∣|a^{\prime}_m| from 4.32 nm to ∞\infty by different annealing temperatures [Figure 2(h)-(Figure 2i)]. Therefore, this route also provides a feasible way for the low-temperature transport and spectral measurement of mKr/G to resolve the possible flat moiré band and related interaction-driven quantum phases.

Image not extracted

Figure 2. In situ tunable moiré pattern. (a) Fermi surface maps of mXe/G (I-IV) and mKr/G (IV-VII) at different am′a'_{m}. (b) Correspondingly band structures along the path #1 in (a)-I of mXe/G (I-IV) and mKr/G (IV-VII) at different am′a'_{m}. (c, d) The temperature dependent photoemission data at the Γ\Gamma point of mXe/G and mKr/G. The dashed lines represent the photoemission data at the Γ\Gamma point of mXe/G (I-IV) and mKr/G (IV-VII) in (b), respectively. (e) The energy distribution curves (EDCs) at the Γ\Gamma point for the Kekulé distortion of mXe/G and mKr/G. The EDC at K point of pristine graphene is also plotted as black line. (f) Band structure along the Γ\Gamma-K direction measured at T=46T = 46 K, in Kr pressure PKr=1.33×10−9P_{\mathrm{Kr}} = 1.33 \times10^{-9} torr (I), 1.33×10−81.33 \times10^{-8} torr (II) and 6.65×10−86.65 \times10^{-8} torr (III), respectively. (g) Corresponding momentum distribution curves (MDC) at Dirac point energy (EB=0.36E_{\mathrm{B}} = 0.36 eV). The derived moiré period am′a'_{m} is also labeled. (h) Band structure along the Γ\Gamma-K direction measured at T=6T = 6 K, as mKr/G annealed at Ta=48T_{\mathrm{a}} = 48 K (I), 42 K (II) and 36 K (III), respectively. (i) Similar to (g) but for different anneal temperature. (j) The moiré period as a function of aXea_{\mathrm{Xe}} (purple line) and aKra_{\mathrm{Kr}} (brown line). The data points I-III refer to (a)-I to (a)-III, and A-C refer to PXe=6.6×10−10P_{\mathrm{Xe}} = 6.6 \times10^{-10} torr, 6.6×10−96.6 \times10^{-9} torr and 4.0×10−84.0 \times10^{-8} torr at Ta=60T_{\mathrm{a}} = 60 K [26] in mXe/G. The data points VI-VIII refer to (a)-VI to (a)-VIII, and D-E refer to (f)-II to (f)-III.

The results in Fig. 2 indicate that, by tunning annealing temperature and pressure, the high-order moiré periodicity in mXe/G and mKr/G complementarily covers almost the full range of am′a^{\prime}_m in both positive and negative directions, as summarized in Fig. 2(j).

To gain an in-depth understanding of the experimental results, we perform theoretical analysis, which indicates that the high-order moiré pattern in mXe/G and mKr/G are good platforms to simulate the honeycomb lattice with tunable sublattice degree of freedom. We theoretically construct the low-energy continuum moiré Hamiltonians for mXe/G and mKr/G,

H=(h+(k−sκ+)Ts(r)Ts†(r)h−(k−sκ−)),H = \begin{pmatrix} h_{+}(\mathbf{k}-s\boldsymbol{\kappa}_{+}) & T_s(\mathbf{r}) \\ T_s^\dagger(\mathbf{r}) & h_{-}(\mathbf{k}-s\boldsymbol{\kappa}_{-}) \end{pmatrix},

where the diagonal terms h±(k)=ℏvF(±kxσx+kyσy)h_{\pm}(\mathbf{k})=\hbar v_F(\pm k_x\sigma_x+k_y\sigma_y) are the Dirac Hamiltonians for the two valleys of monolayer graphene with σx,y\sigma_{x,y} being Pauli matrices in the A and B sublattice space, the index ss is +1 (−1) for mXe/G (mKr/G), and κ±=4π/(3∣am′∣)(∓1/2,3/2)\boldsymbol{\kappa}_{\pm}=4\pi/(3|a^{\prime}_m|)(\mp1/2,\sqrt{3}/2) are located at the corners of the moiré Brillouin zone. The off-diagonal term Ts(r)T_s(\mathbf{r}) is the intervalley coupling induced by the moiré superlattices and is parametrized as follows:

Ts(r)=T0+T+1eig2⋅r+T−1eig3⋅r,T_s(\mathbf{r}) = T_0 + T_{+1}e^{i\mathbf{g}_2\cdot\mathbf{r}} + T_{-1}e^{i\mathbf{g}_3\cdot\mathbf{r}},
Tj=(w0e−i2πj/3w1w1w0ei2πj/3),T_j = \begin{pmatrix} w_0e^{-i2\pi j/3} & w_1 \\ w_1 & w_0e^{i2\pi j/3} \end{pmatrix},

where g2=4π/(3∣am′∣)(−3/2,1/2)\mathbf{g}_2 = 4\pi/(\sqrt{3}|a'_m|)(-\sqrt{3}/2,1/2) and g3=4π/(3∣am′∣)(−3/2,−1/2)\mathbf{g}_3 = 4\pi/(\sqrt{3}|a'_m|)(-\sqrt{3}/2,-1/2) are moiré reciprocal lattice vectors. By fitting to Kekulé gap of the experiment data, we have w1=67w_1 = 67 meV for mXe/G and w1=73w_1 = 73 meV for mKr/G, respectively. By fitting to first-principles band structures (see Supplemental Material [30]), we estimated that w0=1.33w_0 = 1.33 meV for mXe/G and w0=0.58w_0 = 0.58 meV for mKr/G, respectively. Here w0w_0 is an order of magnitude smaller than w1w_1.

We display the theoretical moiré band structure of mKr/G with am′=5a'_m = 5 nm in Figure 3(a). Our model clearly indicates two middle moiré bands connected by the Dirac points at the charge neutrality point. The symmetrized ARPES band structure near the K point of graphene [black line in Figure 3(b)-I)] are shown in Figure 3(b), with am′=5a'_m = 5 nm, 10 nm and 20 nm. To directly compare the experimental results, the theoretical ARPES results based on the calculated band structure are also appended. The sudden intensity changes in ARPES results support the hybridization gaps at the crossing points between the main Dirac cone and replicas as indicated by the simulations, and reveal the formation of moiré bands.

Image not extracted

Figure 3. Simulation of honeycomb lattice in mesoscale by high-order moiré patterns in mXe/G and mKr/G. (a) Calculated moiré band structure of mKr/G along high-symmetry path in the moiré BZ. The red dashed lines show the tight-binding-model fit to the two middle bands. ∣am′∣|a'_{m}| is 5 nm in the calculation. The Dirac points are set to be at zero energy in this plot. (b) Experimentally determined band structures of mKr/G near the K point along the black line in top left corner of b-I at am′=5a'_{m}=5 nm, 10 nm and 20 nm, achieved by annealing at Ta=37T_{\mathrm{a}}=37 K, 42 K and 43 K, respectively. Corresponding theoretical simulation are appended in the insets for the direct comparison. (c) Schematic simulation of honeycomb lattice formed by AM_{\mathrm{M}} and BM_{\mathrm{M}} sites. tAMBM(1)t_{\mathrm{A}_{\mathrm{M}}\mathrm{B}_{\mathrm{M}}}^{(1)} denotes the nearest-neighbour hopping. (d, e) The averaged sublattice polarization PP for the two middle bands in mXe/G and mKr/G, respectively. ∣am′∣|a'_{m}| is 5 nm in the calculation. (f) The calculated magnitude of the nearest-neighbour hopping parameter tAMBM(1)t_{\mathrm{A}_{\mathrm{M}}\mathrm{B}_{\mathrm{M}}}^{(1)} as a function of ∣am′∣|a'_{m}| in mXe/G (purple dashed line) and mKr/G (brown dashed line).

Symmetry analysis of these two moiré bands indicates that they can be effectively described by a tight-binding model defined on the honeycomb lattice formed by AM_{\mathrm{M}} and BM_{\mathrm{M}} sites [Figure 3(c)], with the lattice constant equivalent to am′a'_m. We construct the two Wannier states centered, respectively, at AM_{\mathrm{M}} and BM_{\mathrm{M}} sites. Using the obtained Wannier states, we further calculate the hopping parameters in the tight-binding model, which accurately reproduces the energy dispersion of the two middle bands [Figure 3(a)]. For mXe/G, a similar conclusion can also be achieved. Namely, the two middle moiré bands in mXe/G also match with the results of the tight-binding model on the honeycomb lattice formed by AM_{\mathrm{M}} and BM_{\mathrm{M}} sites. However, there is a crucial difference regarding the sublattice degree of freedom between mXe/G and mKr/G. The Wannier state centered at AM_{\mathrm{M}} (BM_{\mathrm{M}}) site is polarized to A(B) sublattice in mXe/G, but B(A) sublattice in mKr/G. This difference is revealed by the distinct real-space patterns of sublattice polarization for the two middle moiré bands in mXe/G and mKr/G, as shown in Figure 3(d)-(Figure 3e). We note that the different sublattice patterns in real space are counterparts of different valley arrangements in momentum space for the two systems, which could be experimentally detected by atomic scanning tunneling spectroscopy. In Figure 3(f), we present the envlution of ∣tAMBM(1)∣\left|t_{\mathrm{A}_{\mathrm{M}}\mathrm{B}_{\mathrm{M}}}^{(1)}\right| (the nearest-neighbor hopping parameter) as a function of am′a'_{\mathrm{m}}, which decreases rapidly accounting for the decreasing of bandwidth with the increasing of am′a'_{\mathrm{m}}.

For completeness, we finally present the topological phase diagram of the moiré Hamiltonians as a function of model parameters. Up to an energy constant, the Hamiltonian HH is fully determined by two dimensionless parameters w/ℏvF∣κ+∣w/\hbar v_{\mathrm{F}}|\boldsymbol{\kappa}_{+}| and β\beta, where ww is equal to w02+w12\sqrt{w_{0}^{2}+w_{1}^{2}} and β\beta is defined through (w0,w1)=w(sin⁡β,cos⁡β)(w_{0},w_{1})=w(\sin\beta,\cos\beta). In Figure 4, we present the topological phase diagram for the two middle moiré bands as a function of w/ℏvF∣κ+∣w/\hbar v_{\mathrm{F}}|\boldsymbol{\kappa}_{+}| and β\beta. The color shaded regions represent gapped phases, where the two middle bands are separated by an energy gap from other bands. Based on symmetry eigenvalues and Wilson loop spectra, the gapped phases can be further classified into fragile topological phases protected by C^2zT^\hat{C}_{2z}\hat{T} symmetry and topologically trivial phases described by a honeycomb lattice model. Here C^2z\hat{C}_{2z} is the twofold rotation symmetry around the z axis and T^\hat{T} is time-reversal symmetry. In the limit of w1=0w_{1}=0 and β=π2\beta=\frac{\pi}{2}, there is an exact mapping of HXe/KrH_{\mathrm{Xe/Kr}} to the chiral limit of twisted bilayer graphene [31, 32], which is consistent with the fact that the fragile topological phases appear around β=π/2\beta=\pi/2 in our phase diagram. Our systems mXe/G and mKr/G are in the opposite limit of w0≪w1w_{0}\ll w_{1}, where the two middle bands are topologically trivial and belong to phase (4) in Figure 4. We note that parameters w0w_{0} and w1w_1 are system dependent; the phase diagram in Figure 4 shows the possibility of realizing fragile topological phases in other graphene superlattice systems.

Image not extracted

Figure 4. Topological phase diagram of the two middle bands of moiré Hamiltonian HH as a function of parameters w/ℏvF∣κ+∣w/\hbar v_F|\kappa_+| and β\beta. In the white regions, the middle bands are not separated by an overall energy gap from other bands. In the color shaded regions, the gapped phases (1)–(3) possess fragile topology, while the gapped phases (4)–(7) are topologically trivial and described by a tight-binding model defined on a honeycomb lattice.

Conclusions

In summary, the in situ tunable high-order moiré patterns in mXe/G and mKr/G provide good platforms to simulate honeycomb lattices in mesoscale. As summarized in Figure 2(j), by tuning the temperature and partial pressure in the annealing process, the honeycomb lattice constant am′a'_m in mXe/G and mKr/G covers the range in mesoscale along positive and negative directions, respectively. In the meantime, the hopping term in the tight-binding model between the nearest-neighbor AM_{\mathrm{M}} and BM_{\mathrm{M}} decreases rapidly with increasing am′a'_m [Figure 3(f)]. This is expected since the bandwidth of the two middle bands decreases when am′a'_m increases, as indicated in Figure 3(b). A honeycomb lattice model with narrow or nearly flat bands can host strong many-body correlation effects and will serve as a testbed for exotic phases that are highly sought-after theoretically [33, 34, 35]. Our work will stimulate further theoretical and experimental studies to reveal the novel properties of mesoscale graphene.

Appendix

Excluding photoelectron diffraction effect

The observation of replica bands can not be explained by photoelectron diffraction effect [36]. In the photoelectron diffraction scenario as illustrated in Figure 5(a), photoelectrons from graphene Dirac cone are diffracted by Kr monolayer, resulting in replica band with the same position as κm′\kappa'_m [Figure 5(b)]. However, the original photoelectrons have a limitation of momentum parallel to the sample surface, K∥=2meEKinℏK_{\parallel} = \frac{\sqrt{2m_eE_{K\mathrm{in}}}}{\hbar}, where EKinE_{K\mathrm{in}} and mem_e are photoelectrons’ kinetic energy and mass. Correspondingly, the diffraction induced replica bands are also expected to have a cutoff [Figure 5(a)]. In Figure 5(b), we clearly observe that κm′\kappa'_m replica exceed the limitation, which contradicts to photoelectron diffraction scenario. Therefore, our results confirm that the observed replicas are induced by moiré potential modulated initial state, rather than photoelectron diffraction.

Image not extracted

Figure 5. The photoelectron diffraction scenario. (a) Schematic diagram of final state diffraction picture in mKr/G. The black dashed line indicates the maximum measurement range of graphene in ARPES with a photon energy of 21.2eV. (b) The curvature of Dirac replicas km′k'_m along black solid line in a, the dashed line corresponds momentum cutoff limit.

Figures

References

  1. [1]Y. Cao, V. Fatemi, A. Demir, S. A. Fang, S. L. Tomarken, J. Y. Luo, J. D. Sanchez-Yamagishi, K. Watanabe, T. Taniguchi, E. Kaxiras, R. C. Ashoori, and P. Jarillo-Herrero, Correlated insulator behaviour at half-filling in magic-angle graphene superlattices. Nature 556, 80 (2018).
  2. [2]Y. Cao, V. Fatemi, S. A. Fang, K. J. Watanabe, T. Taniguchi, E. Kaxiras, and P. Jarillo-Herrero, Unconventional superconductivity in magic-angle graphene superlattices. Nature 556, 43 (2018).DOI
  3. [3]G. R. Chen, L. L. Jiang, S. Wu, B. Lyu, H. Y. Li, B. L. Chittari, K.Watanabe, T. Taniguchi, Z. W. Shi, J. Jung, Y. B. Zhang, and F. Wang, Evidence of a gate-tunable Mott insulator in trilayer graphene moiré superlattice. Nat. Phys. 15, 237 (2019).DOI
  4. [4]G. R. Chen, A. L. Sharpe, P. Gallagher, I. T. Rosen, E. J. Fox, L. L. Jiang, B. Lyu, H. Y. Li, K. Watanabe, T taniguchi, J. Jung, Z. W. Shi, D. Goldhaber-Gordon, Y. B. Zhang, and F. Wang, Signatures of tunable superconductivity in a trilayer graphene moiré superlattice. Nature 572, 215 (2019).DOI
  5. [5]X. B. Lu, P. Stepanov, W. Yang, M. Xie, M. A. Aamir, I. Das, C. Urgell, K. Watanabe, T. Taniguchi, G. Y. Zhang, A. Bachtold, A. H. MacDonald, and D. K. Efetov, Superconductors, orbital magnets and correlated states in magic angle bilayer graphene. Nature 574, 653 (2019).DOI
  6. [6]P. Stepanov, I. Das, X. B. Lu, A. Fahminiya, K. Watanabe, T. Taniguchi, F. H. L. Koppens, J. Lischner, L. Levitov, and D. K. Efetov, Untying the insulating and superconducting orders in magic-angle graphene. Nature 583, 375 (2020).
  7. [7]F. C. Wu, T. Lovorn, E. Tutuc, and A. H. Macdonald, Hubbard model physics in transition metal dichalcogenide moiré bands. Phys. Rev. Lett. 121, 026402 (2018).DOI
  8. [8]M. H. Naik, and M. Jain, Ultraflatbands and shear solitons in moiré patterns of twisted bilayer transition metal dichalcogenides. Phys. Rev. Lett. 121, 266401 (2018).DOI
  9. [9]D. M. Kennes, M. Claassen, L. Xian, A. Georges, A. J. Millis, J. Hone, C. R. Dean, D. N. Basov, A. N. Pasupathy, and A. Rubio, Moiré heterostructures as a condensed-matter quantum simulator. Nat. Phys. 17, 155 (2021).DOI
  10. [10]H. C. Po, L. J. Zhou, A. Vishwanath, and T. Senthil, Origin of Mott insulating behavior and superconductivity in twisted bilayer graphene. Phys. Rev. X 8, 031089 (2018).
  11. [11]Z. D. Song, Z. J. Wang, W. J. Shi, G. Li, C. Fang, and B. A. Bernevig, All magic angles in twist bilayer graphene are topological. Phys. Rev. Lett. 123, 036401 (2019).DOI
  12. [12]Z. D. Song, and B. A. Bernevig, Magic-angle twisted bilayer graphene as a topological heavy fermion problem. Phys. Rev. Lett. 129, 047601 (2022).DOI
  13. [13]F. C. Wu, T. Lovorn, E. Tutuc, I. Martin, and A. H. MacDonald, Topological insulators in twisted transition metal dichalcogenide homobilayers. Phys. Rev. Lett. 122, 086402 (2019).DOI
  14. [14]M. Angeli and A. H. Macdonald, Γ-valley transition-metal-dichalcogenide moiré bands. PNAS 10, 118 (2021).DOI
  15. [15]Y. H. Tang, L. Z. Li, T. X. Li, Y. Xu, S. Liu, K. Barmak, K. Watanabe, T. Taniguchi, A. H. Macdonald, J. Shan and K. F. Mak, Simulation of Hubbard model physics in WSe₂/WS₂ moiré superlattices. Nature 579, 353 (2020).DOI
  16. [16]E. C. Regan, D. Q. Wang, C. H. Jin, M. B. Utama, B. N. Gao, X. Wei, S. H. Zhao, W. Y. Zhao, Z. C. Zhang, K. Yumigeta, M. Blei, J. D. Calström, K. J. Watanabe, T. Taniguchi, S. Tongay, M. Crommie, A. Zettl, and F. Wang, Mott and generalized Wigner crystal states in WSe₂/WS₂ moiré superlattices. Nature 579, 359 (2020).DOI
  17. [17]Y. Xu, S. Liu, D. A. Rhode, K. Watanabe. T. Taniguchi, J. Hone, V. Elser, K. F. Mak, and J. Shan, Correlated insulating states at fractional fillings of moiré superlattices. Nature 587, 214 (2020).DOI
  18. [18]H. Y. Li, S. W. Li, E. C. Regan, D. Q. Wang, W. Y. Zhao, S. Kahn, K. Yumigeta, M. Blei, T. Taniguchi, K. Watanabe, S. Tongay, A. Zettl, M. F. Crommie, and F. Wang, Imaging two-dimensional generalized Wigner crystals. Nature 597, 650 (2021).DOI
  19. [19]T. X. Li, S. W. Jiang, B. Shen, Y. Zhang, L. Z. Li, Z. Tao, T. Devakul, K. Watanabe, T. Taniguchi, L. Fu, J. Shan, and K. F. Mak, Quantum anomalous Hall effect from intertwined moiré bands. Nature 600, 641 (2021).DOI
  20. [20]N. R. Wilson, K. V. Nguyen, K. Seyler, P. Rivera, A. J. Marsden, Z. L. Laker, G. C. Constantinescu, V. Kandyba, A. Barinov, N. D. M. Hine, X. D. Xu, and D.H. Cobden, Determination of band offsets, hybridization, and exciton binding in 2D semiconductor heterostructures. Sci. Adv. 3, 1601832 (2017).DOI
  21. [21]S. Lisi, X. B. Lu, T. Benschop, T. A. D. Jong, P. Stepanov, J. R. Duran, F. Margot, I. Cucchi, E. Cappelli, A. Hunter, A. Tamai, V. Kandyba, A. Giampietri, A. Barinov, J. Jobst, V. Stalman, M. Leeuwenhoek, K. Watanabe, T. Taniguchi, L. Rademaker, S. J. Molen, M. P. Allan, D. K. Efetov, and F. Baumberger, Observation of flat bands in twisted bilayer graphene. Nat. Phys. 17, 189 (2021).DOI
  22. [22]M. L. B. Utama, R. J. Koch, K. Lee, N. Leconte, H. Y. Li, S. H. Zhao, L. Jiang, J. Y. Zhu, K. Watanabe, T. Taniguchi, P. D. Ashby, A. W. Bargioni, A. Zettl, C. Jozwiak, J. Jung, E. Rotenberg, A. Bostwick, and F. Wang, Visualization of the flat electronic band in twisted bilayer graphene near the magic angle twist. Nat. Phys. 17, 184 (2021).DOI
  23. [23]Y. W. Li, S. H. Zhang, F. Q. Chen, L. Y. Wei, Z. L. Zhang, H. B. Xiao, H. Gao, M. Y. Chen, S. J. Liang, D. Pei, L. X. Xu, K. Watanabe, T. Taniguchi, L. X. Yang, F. Miao, J. P. Liu, B. Cheng, M. X. Wang, Y. L. Chen, and Z. K. Liu, Observation of Coexisting Dirac Bands and Moiré Flat Bands in Magic-Angle Twisted Trilayer Graphene. Adv. Mater. 10, 1002 (2022).DOI
  24. [24]D. Pei, B. B. Wang, Z.S. Zhou, Z. H. He, L. H. An, S. M. He, C. Chen, Y. W. Li, L. Y. Wei, A. J. Liang, J. Avila, P. Dudin, V. Kandyba, A. Giampietri, M. Cattelan, A. Barinov, and Z. K. Liu, Observation of Γ-Valley moiré bands and emergent hexagonal lattice in twisted transition metal dichalcogenides. Phys. Rev. X 12, 021065 (2022).DOI
  25. [25]C. L. Wu, Q. Wan, C. Peng, S. K. Mo, R. Z. Li, K. M. Zhao, Y. P. Guo, C. D. Zhang, and N. Xu, Observation of high-order moiré effect and multiple Dirac fermions replicas in graphene-SiC heterostructure. Phys. Rev. B 104, 235130 (2021).DOI
  26. [26]C. L. Wu, Q. Wan, C. Peng, S. K. Mo, R. Z. Li, K. M. Zhao, Y. P. Guo, S. J. Yuan, F. C. Wu, C. D. Zhang, and N. Xu, Tailoring dirac fermions by in-situ tunable high-order moiré pattern in graphene-monolayer xenon heterostructure. Phys. Rev. Lett. 129, 176402 (2022).DOI
  27. [27]S. Im, H. Im, K. Kim, J.-E. Lee, J. Hwang, S.-K. Mo, C. Hwang, Modified Dirac Fermions in the Crystalline Xenon and Graphene Moiré Heterostructure. Adv. Phys. Res. 2, 2200091 (2023).
  28. [28]Y. W. Li, Q. Wan, N. Xu, Recent Advances in Moiré Superlattice Systems by Angle-Resolved Photoemission Spectroscopy. Adv. Mater. 2305175 (2023).DOI
  29. [29]Q. Y. Wang, W. H. Zhang, L. L. Wang, K. He, X. C. Ma, and Q. K. Xue, Large-scale uniform bilayer graphene prepared by vacuum graphitization of 6H-SiC (0001) substrates. J Phys.: Condens. Mater. 25, 095002 (2013).DOI
  30. [30]See Supplemental Material for LEED patterns, moiré Hamiltonian, estimation of model parameters, bloch states, band symmetries for mXe/G and mKr/G, wannier states and tight-binding model constructions, and topological phase diagram, which includes Refs. [26, 37-38].
  31. [31]Q. Lu, C. Le, X. Zhang, J. Cook, X. He, M. Zarenia, M. Vaninger, P. F. Miceli, D. J. Singh, C. Liu, H. Qin, T.-C. Chiang, C.-K. Chiu, G. Vignale, and G. Bian, Dirac fermion cloning, moiré flat bands, and magic lattice constants in epitaxial monolayer graphene, Adv. Mater. 34, 2200625 (2022).DOI
  32. [32]V. Crépel, A. Dunbrack, D. Guerci, J. Bonini, and J. Cano, Chiral model of twisted bilayer graphene realized in a monolayer, arXiv:2305.14423 (2023).DOI
  33. [33]S. Raghu, X. L. Qi, C. Honerkamp, and S. C. Zhang, Topological Mott insulators. Phys. Rev. Lett. 100, 156401 (2008).pubmed.ncbi.nlm.nih.gov/18518132
  34. [34]T. Grover, and T. Senthil, Topological spin hall states, charged skyrmions, and superconductivity in two dimensions. Phys. Rev. Lett. 100, 156804 (2008).DOI
  35. [35]Z. Y. Meng, T. C. Lang, S. Wessel, F. F. Assaad, and A. Muramatsu, Quantum spin liquid emerging in two-dimensional correlated Dirac fermions. Nature 464, 847 (2010).DOI
  36. [36]C. M. Polley, L. I. Johansson, H. Fedderwitz, T. Balasubramanian, M. Leandersson, J. Adell, R. Yakimova, C. Jacobi, Origin of the π-band replicas in the electronic structure of graphene grown on 4H-SiC(0001). Phys. Rev. B 99, 115404 (2019).DOI

Paper details

vlnc:W_396qdpb3
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

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