Intervalley-Coupled Twisted Bilayer Graphene from Substrate Commensuration
Bo-Ting Chen, Michael G. Scheer, Biao Lian · 2025
Read Intervalley-Coupled Twisted Bilayer Graphene from Substrate Commensuration by Bo-Ting Chen, Michael G. Scheer, Biao Lian (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2503-03869.txt.
Introduction
Bo-Ting Chen and co-authors coupled graphene's valleys through a substrate in two thousand twenty-five. Twisted bilayer graphene hosts famously flat energy bands. Its two valleys usually stay separate worlds. Aligning the lower sheet with a matching crystal changes that. The substrate's pattern forces the valleys to merge. Flat bands then hybridize into a new four-band system.
The merged model mirrors a honeycomb orbital lattice. Some bands touch at points and can flatten further. Substrate spin-orbit effects open gaps between bands. Those gaps carry strong topological character. Best flatness still sits near the magic twist angle. Two real substrate materials are named as candidates. The route could yield new quantum phases on demand.
Catalog record
| CREATOR | Bo-Ting Chen, Michael G. Scheer, Biao Lian |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 72 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2503-03869
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Intervalley-Coupled Twisted Bilayer Graphene from Substrate Commensuration Bo-Ting Chen, Michael G. Scheer, Biao Lian Abstract We show that intervalley coupling can be induced in twisted bilayer graphene (TBG) by aligning the bottom graphene layer with either of two types of commensurate insulating triangular Bravais lattice substrate. The intervalley coupling folds the ± K \pm K valleys of TBG to the Γ \Gamma -point and hybridizes the original TBG flat bands into a four-band model equivalent to the p x p_{x} - p y p_{y} orbital honeycomb lattice model, in which the second conduction and valence bands have quadratic band touchings and can become flat due to geometric frustration. The spin-orbit coupling from the substrate opens gaps between the bands, yielding topological bands with spin Chern numbers 𝒞 \mathcal{C} up to ± 4 \pm 4 . For realistic substrate potential strengths, the minimal bandwidths of the hybridized flat bands are still achieved around the TBG magic angle θ M = 1.05 ∘ \theta_{M}=1.05^{\circ} , and their quantum metrics are nearly ideal. We identify two candidate substrate materials Sb 2 Te 3 and GeSb 2 Te 4 , which nearly perfectly realize the commensurate lattice constant ratio of 3 \sqrt{3} with graphene. These systems provide a promising platform for exploring strongly correlated topological states driven by geometric frustration. Introduction. —Twisted bilayer graphene (TBG) at the magic angle \theta_{M}\approx 1.05^{\circ} [ 13 ] has attracted extensive interest in recent years. This system hosts topological flat electron bands with strong interactions [ 63 , 3 , 76 , 75 ] and possesses a rich phase diagram with superconductivity, correlated insulator, and Chern insulator phases [ 18 , 19 , 95 , 22 , 52 , 24 , 73 , 64 , 26 , 50 , 74 , 23 , 5 , 72 , 58 , 90 , 46 , 14 , 42 , 57 ] . The exploration of flat bands in various other 2D moiré systems has also led to fruitful discoveries of novel correlated and topological states such as the fractional Chern insulators (FCIs) at zero magnetic field recently observed in twisted MoTe 2 [ 66 , 15 , 100 , 61 , 92 , 36 , 40 , 91 , 60 ] and pentalayer rhombohedral graphene [ 53 ] . Generically, the topological and geometrical properties of flat bands are crucial for realizing strongly correlated and topological states such as FCIs [ 81 , 83 , 25 , 62 , 89 ] , and moiré systems are an ideal platform for tuning these properties. In this letter, we are interested in designing graphene-based moiré flat bands with a geometric frustration origin, such as the flat bands of the kagome lattice and p_{x} - p_{y} orbital honeycomb lattice tight-binding models [ 48 , 55 , 32 , 85 , 11 ] , which are promising systems for spin liquids and FCIs [ 94 , 96 , 80 ] . While such moiré flat bands were previously predicted in \Gamma -valley twisted transition metal dichalcogenides (TMDs) [ 6 , 87 , 51 , 82 ] , these predicted bands lie far from charge neutrality and may suffer from inaccuracies of TMD model parameters. Recently, it was shown that twisted Kekulé ordered graphene systems can give kagome lattice and p_{x} - p_{y} orbital honeycomb lattice flat bands [ 71 , 70 , 69 ] due to the intervalley coupling induced by Kekulé order. Kekulé ordered graphene can be synthesized by lithium deposition [ 78 , 39 , 10 , 21 , 65 ] , but this method introduces disorder and electron doping and is challenging in the twistronics context. This motivates us to study the engineering of the TBG flat bands through intervalley coupling arising from a substrate material. We show that an intervalley coupling can be induced in TBG by aligning the bottom graphene layer with an insulating substrate of either of two types of commensurate lattice. This eliminates the valley degree of freedom and modifies the magic angle TBG flat bands into a p_{x} - p_{y} orbital honeycomb lattice model which hosts flat bands with topological quadratic band touchings [ 85 ] . With spin-orbit coupling (SOC) from the substrate, these flat bands develop into flat spin Chern bands with spin Chern number up to \pm 4 and close-to-ideal quantum metrics [ 83 , 25 ] . From ab initio calculations, we identify two candidate substrate materials Sb 2 Te 3 and GeSb 2 Te 4 . Intervalley-coupled TBG provides a new platform for studying interacting topological states in geometrically frustrated flat bands. The model setup. —We consider a TBG system on a commensurate non-magnetic substrate as follows. The top and bottom graphene layers (denoted by l=\pm ) are twisted relative to an aligned configuration by angles -l\theta/2 so that the two layers have a relative twist angle of \theta . The bottom graphene layer is aligned with a substrate with a triangular Bravais lattice commensurate with the graphene lattice. We require the substrate to be a gapped insulator with the graphene chemical potential lying in the gap. This way, the substrate contributes no electrons to graphene and simply induces a substrate potential at low energies. This gives a generic Hamiltonian for the TBG system of the form H=\begin{pmatrix}\mathcal{H}_{+}&\mathcal{H}_{\text{hop}}\\ \mathcal{H}_{\text{hop}}^{\dagger}&\mathcal{H}_{-}\end{pmatrix},\quad\begin{cases}&\mathcal{H}_{+}=\mathcal{H}_{+}^{(0)}\ ,\\ &\mathcal{H}_{-}=\mathcal{H}_{-}^{(0)}+\mathcal{H}_{-}^{\text{(sub)}}\ ,\end{cases} (1) where \mathcal{H}_{l}^{(0)} is the graphene Hamiltonian in layer l , \mathcal{H}_{-}^{\text{(sub)}} is the substrate potential acting on layer l=- , and \mathcal{H}_{\text{hop}} is the hopping between the two graphene layers. Each graphene layer l=\pm has Dirac electrons at two valleys \mathbb{K}_{l} and -\mathbb{K}_{l} in the graphene Brillouin zone (BZ). In this work, we are interested in substrates which couple the two graphene valleys of layer l=- . This constrains the substrate lattice constant, as we will show below. Figure 1: (a)-(b): Top view of the bottom graphene lattice (black) on a substrate lattice (pink and green), with (r_{s},\phi_{s})=(\sqrt{3},30^{\circ}) (type Y) in (a), and (r_{s},\phi_{s})=(3,0^{\circ}) (type X) in (b). (c) The BZs of the top/bottom layer graphene (blue/red solid hexagon) and the substrate in (a) (red dashed hexagon), and the moiré BZ (black hexagon). (d) Zoom-in of the moiré BZ. In the continuum limit, we adopt a real space basis \ket{\mathbb{r},l,\eta,\alpha,s} , in terms of position \mathbb{r} , layer l , graphene valley \eta=\pm (for \mathbb{K}_{l} and -\mathbb{K}_{l} ), sublattice \alpha=\pm (for A and B ), and z -direction spin s=\pm (for \uparrow and \downarrow ). We use \tau_{\mu} , \sigma_{\mu} , and s_{\mu} to denote the 2\times 2 identity ( \mu=0 ) and Pauli ( \mu=x,y,z ) matrices in the basis of valley \eta , sublattice \alpha , and spin s , respectively. The graphene Hamiltonians \mathcal{H}_{l}^{(0)} and \mathcal{H}_{\text{hop}} are spin independent because of the negligible spin-orbit coupling (SOC) in graphene, and they do not couple the two graphene valleys. Explicitly, these terms give the Bistritzer-MacDonald (BM) continuum model [ 13 ] of TBG without substrate: \begin{split}\mathcal{H}_{l}^{(0)}&=-i\hbar v_{l}\nabla\cdot\Big[\tau_{+}\bm{\sigma}_{l\theta/2}-\tau_{-}\bm{\sigma}_{l\theta/2}^{*}\Big]s_{0},\\ \mathcal{H}_{\text{hop}}&=\Big[\tau_{+}T(\mathbb{r})+\tau_{-}T^{*}(\mathbb{r})\Big]s_{0}\ ,\end{split} (2) where v_{l} is the layer l Fermi velocity, \tau_{\pm}=\frac{1}{2}(\tau_{0}\pm\tau_{z}) , and \bm{\sigma}_{\phi}=(\cos(\phi)\sigma_{x}+\sin(\phi)\sigma_{y},-\sin(\phi)\sigma_{x}+\cos(\phi)\sigma_{y}) . We take v_{+}=v_{-}=$611.4\text{\,}\mathrm{meV}\text{\,}\mathrm{nm}$ which is typical [ 88 ] , assuming that any substrate effects on v_{l} are negligible. The periodic moiré hopping T(\mathbb{r}) is \begin{split}T(\mathbb{r})&=\sum_{j=1}^{3}T_{\mathbb{q}_{j}}e^{i\mathbb{q}_{j}\cdot\mathbb{r}}\ ,\\ T_{\mathbb{q}_{j}}&=w_{0}\sigma_{0}+w_{1}(\sigma_{x}\cos\zeta_{j}+\sigma_{y}\sin\zeta_{j})\ ,\end{split} (3) where \mathbb{q}_{j}=R_{\zeta_{j}}(\mathbb{K}_{-}-\mathbb{K}_{+}) as illustrated