betaFeedback

WorksPhysics and Astronomy

Reduction of the Twisted Bilayer Graphene Chiral Hamiltonian into a 2×22\times2 matrix operator and physical origin of flat-bands at magic angles

Abstract

The chiral Hamiltonian for twisted graphene bilayers is written as a 2×22\times2 matrix operator by a renormalization of the Hamiltonian that takes into account the particle-hole symmetry. This results in an effective Hamiltonian with an average field plus and effective non-Abelian gauge potential. The action of the proposed renormalization maps the zero-mode region into the ground state. Modes near zero energy have an antibonding nature in a triangular lattice. This leads to a phase-frustration effect associated with massive degeneration, and makes flat-bands modes similar to confined modes observed in other bipartite lattices. Suprisingly, the proposed Hamiltonian renormalization suggests that flat-bands at magic angles are akin to floppy-mode bands in flexible crystals or glasses, making an unexpected connection between rigidity topological theory and magic angle twisted two-dimensional heterostructures physics.

Introduction

Superconducting states are difficult to reach as they require very strict laboratory parameters [1]. High TcT_c superconductors use cuprates which are well-ordered structures of atoms combined in three-dimensional arrangements [2, 3, 4, 5, 6, 7]. For these materials, the mechanism that counteracts the Colombian repulsion force between electrons is not exactly known, and for this reason, these unconventional states are referred to as strongly correlated [8, 9]. Recently, it has been discovered that twisted bilayer graphene exhibits superconducting states at certain rotation angles [10, 11] where the electron interactions are maximized [12]. This rotated graphene bilayer model generates a Moiré pattern as a function of the rotation angle, defining a Moiré Brillouin zone (mBZ) in reciprocal space. These special angles are called ”magic” and were predicted as a possible consequence of flat-bands observed in previous theoretical work [13]. In the work of Cao et.al. [10], a Mott insulating state appears in the middle of these superconducting phases. The study of the electronic properties of rotated graphene over graphene started before the discovery of superconductivity at magic angles. In the work of J. Santos [14] and A. Macdonald [13], a continuous Hamiltonian model was presented; however, due to the presence of an interlayer amplitude AA coupling, this model did not present chiral symmetry. In a recent work by G. Tarnopolsky et. al. [15], a chiral continuum model was studied, and only the AB and BA inter-layer couplings are different from zero. Perhaps, so far, it is the simplest model that best captures the nature of magic angles; at these angles the dispersion energy becomes flat and has a recurrence behavior. At these magic angles the Fermi velocity also goes to zero. Due to its chiral symmetry, the Hamiltonian of this model also produces an intra-valley inversion symmetry [16], so the energy dispersion is

inversion symmetric at all twist angles. Also, we can distinguish between different magic angles when zero modes occurs, and thus the inter-valley inversion classifies the topology of the twist angle. The zero-mode flat-band solutions have some resemblance to the ground state of a quantum Hall effect wave function on a torus [15, 17], and, therefore, the solution is of the harmonic oscillator type, where Landau levels arise [17, 18, 19]. The mechanism that causes the appearance of these magic angles is still not known, however, many investigations suggest that it is a topological aspect of the band structure [20, 21, 22, 23, 24, 25, 26]. The aim of this work its to clarify the physical behavior of this model and to develop an effective equation for the Hamiltonian matrix in order to study all other non flat band states. In particular, here we renormalize the Hamiltonian to take into account the particle-hole symmetry which results in a folding of the spectrum around zero-energy. Then we discuss the physical picture that arises from the renormalization

Twisted Bilayer Graphene Effective Model

Tarnopolsky et. al. derived a chiral Hamiltonian for electrons-holes in twisted bilayer graphene. It captures the ”true magic” of the magic angle physics [15].

H=(0D∗(−r)D(r)0)(1)\mathcal{H}=\begin{pmatrix} 0 & D^{*}(-\mathbf{r}) \\ D(\mathbf{r}) & 0 \end{pmatrix} \tag*{(1)}

where the zero-mode operator is,

D(r)=(−i∂ˉαU(r)αU(−r)−i∂ˉ)(2)D(\mathbf{r})=\begin{pmatrix} -i\bar{\partial} & \alpha U(\mathbf{r}) \\ \alpha U(-\mathbf{r}) & -i\bar{\partial} \end{pmatrix} \tag*{(2)}

and,

D∗(−r)=(−i∂ˉαU∗(−r)αU∗(r)−i∂ˉ)(3)D^{*}(-\mathbf{r})=\begin{pmatrix} -i\bar{\partial} & \alpha U^{*}(-\mathbf{r}) \\ \alpha U^{*}(\mathbf{r}) & -i\bar{\partial} \end{pmatrix} \tag*{(3)}

with ∂ˉ=∂x+i∂y\bar{\partial}=\partial_x+i\partial_y, ∂=∂x−i∂y\partial=\partial_x-i\partial_y. The potential is,

U(r)=e−iq1⋅r+eiφe−iq2⋅r+e−iφe−iq3⋅r(4)U(\mathbf{r})=e^{-i\mathbf{q}_1\cdot\mathbf{r}}+e^{i\varphi}e^{-i\mathbf{q}_2\cdot\mathbf{r}}+e^{-i\varphi}e^{-i\mathbf{q}_3\cdot\mathbf{r}} \tag*{(4)}
Real space Moiré unit cell and mBZ in reciprocal space

Figure 1. a) Real space Moiré unit cell, a1,2\mathbf{a}_{1,2} are two Moiré lattice vectors. Point r0=(a1−a2)/3\mathbf{r}_0=(\mathbf{a}_1-\mathbf{a}_2)/3 is the BA stacking point where all components of the wave function vanish at magic α\alpha. b) mBZ in reciprocal space, b1,2\mathbf{b}_{1,2} are the base vectors.

For this Hamiltonian, the parameters are ϕ=2π3\phi=\frac{2\pi}{3} and q1=kθ(0,−1)\mathbf{q}_1=k_\theta(0,-1), q2=kθ(32,12)\mathbf{q}_2=k_\theta\left(\frac{\sqrt{3}}{2},\frac{1}{2}\right) and q3=kθ(−32,12)\mathbf{q}_3=k_\theta\left(-\frac{\sqrt{3}}{2},\frac{1}{2}\right), the Moiré modulation vector is kθ=2kDsin⁡θ2k_\theta=2k_D\sin\frac{\theta}{2} with kD=4π3a0k_D=\frac{4\pi}{3a_0} is the magnitude of the Dirac wave vector and a0a_0 is the lattice constant of monolayer graphene. The physics of this model is captured by the parameter α\alpha, defined as α=w1v0kθ\alpha=\frac{w_1}{v_0k_\theta}. Here w1w_1 is the interlayer coupling of stacking AB and BA, take the value w1=110meVw_1=110\mathrm{meV} and v0v_0 is the Fermi velocity, with value v0=19.81eV2kDv_0=\frac{19.81\mathrm{eV}}{2k_D}. At magic angles α=0.586,2.221,3.751,5.276,6.795,8.313,9.829,11.345,…\alpha=0.586, 2.221, 3.751, 5.276, 6.795, 8.313, 9.829, 11.345,\ldots, flat-bands appear. Magic α\alpha’s follow a remarkable 3/2 quantization rule [15] for α>0.586\alpha>0.586.

This Hamiltonian is difficult to tackle and in fact most of the studies have been restricted to the zero-mode operator solutions at energy zero [16, 17]. Here, instead of solving the Schrödinger equation with H\mathcal{H} we first propose to reduce the dimensionality of the problem. Starting with the Schrödinger equation HΦ=EΦ\mathcal{H}\Phi=E\Phi, where Φ(r)=(ψ1(r),ψ2(r),χ1(r),χ2(r))T\Phi(r)=(\psi_1(r),\psi_2(r),\chi_1(r),\chi_2(r))^{\mathrm{T}} are the four components of the twisted graphene bilayer, and the index 1, 2 represent each graphene layer, we consider the squared Hamiltonian H2\mathcal{H}^2,

H2=(D∗(−r)D(r)00D(r)D∗(−r))(5)\mathcal{H}^2=\begin{pmatrix}D^*(-r)D(r)&0\\0&D(r)D^*(-r)\end{pmatrix} \tag*{(5)}

This transforms the off-diagonal blocks into zeros. We can understand such transformation as a removal of the particle-hole symmetry that is an anti-unitary anticommuting symmetry. Therefore, we obtain a decoupled equation H2Φ(r)=E2Φ(r)\mathcal{H}^2\Phi(r)=E^2\Phi(r), where the eigenvalues are the squares of the original energies. Therefore, the states at E=0E=0 are ground states of H2Φ(r)\mathcal{H}^2\Phi(r). For arbitrary α\alpha there are always two zero-mode solutions in the KK and K′K' points [15], the signature of a magic angle in H2\mathcal{H}^2 is that the Fermi velocity at k=K,K′k=K,K' points of the Moiré Brillouin (mBrillouin) zone approaches zero and a massive degenerate ground state. We now define a 2×22\times2 effective Hamiltonian H2=D∗(−r)D(r)\mathcal{H}^2=D^*(-r)D(r). As detailed in the supplementary material, the resulting effective Hamiltonian is,

H2=(−∇2+α2∣U(−r)∣2αA†(r)αA(r)−∇2+α2∣U(r)∣2)(6)\mathcal{H}^2=\begin{pmatrix}-\nabla^2+\alpha^2|U(-r)|^2&\alpha A^\dagger(r)\\\alpha A(r)&-\nabla^2+\alpha^2|U(r)|^2\end{pmatrix} \tag*{(6)}

The norm of the potential is,

∣U(r)∣2=3+2cos⁡(b1⋅r−ϕ)+2cos⁡(b2⋅r+ϕ)+2cos⁡(b3⋅r+2ϕ)(7)|U(r)|^2=3+2\cos(\mathbf{b}_1\cdot\mathbf{r}-\phi)+2\cos(\mathbf{b}_2\cdot\mathbf{r}+\phi)+2\cos(\mathbf{b}_3\cdot\mathbf{r}+2\phi) \tag*{(7)}

where b1,2=q2,3−q1\mathbf{b}_{1,2}=\mathbf{q}_{2,3}-\mathbf{q}_1 are the mBrillouin zone Moiré vectors and b3=q3−q2\mathbf{b}_3=\mathbf{q}_3-\mathbf{q}_2. The off-diagonal terms are,

A(r)=−i∑μ=13eiqμ⋅r(2q^μ⊥⋅∇−kθ)(8)A(r)=-i\sum_{\mu=1}^{3}e^{i\mathbf{q}_\mu\cdot\mathbf{r}}\left(2\hat{\mathbf{q}}_\mu^\perp\cdot\nabla-k_\theta\right) \tag*{(8)}

and,

A†(r)=−i∑μ=13e−iqμ⋅r(2q^μ⊥⋅∇+kθ)(9)A^\dagger(r)=-i\sum_{\mu=1}^{3}e^{-i\mathbf{q}_\mu\cdot\mathbf{r}}\left(2\hat{\mathbf{q}}_\mu^\perp\cdot\nabla+k_\theta\right) \tag*{(9)}

where ∇†=−∇\nabla^\dagger=-\nabla with ∇=(∂x,∂y)\nabla=(\partial_x,\partial_y) and μ=1,2,3\mu=1,2,3. This is an essential point as eigenvalues must be reals (notice that −A†(−r)=A(r)-A^\dagger(-r)=A(r)). Also, q^μ⊥\hat{\mathbf{q}}_\mu^\perp is a set of unitary vectors perpendicular to the set qμ\mathbf{q}_\mu (see supplementary). The term $\hat{\mathbf{q}}_\mu^\perp\cdot\nabla$ is the directional gradient along the triangle defined by the Moiré vectors and has an interpretation in terms of frustration (see supplementary).

Any eigenfunction of the original Hamiltonian is an eigenfunction of H2\mathcal{H}^2, therefore, the spinor Ψ(r)=(ψ1(r),ψ2(r))\Psi(r)=(\psi_1(r),\psi_2(r)) is a solution of H2\mathcal{H}^2. In a similar way, we can proceed to use D(r)D∗(−r)D(r)D^*(-r) with solutions (χ1(r),χ2(r))(\chi_1(r),\chi_2(r)). These solutions are easily obtained from (ψ1(r),ψ2(r))(\psi_1(r),\psi_2(r)) using symmetry operations, and thus here we use only study H2\mathcal{H}^2. H2\mathcal{H}^2 eigenfunctions are made from a superposition of pseudo-spin polarized states of H\mathcal{H}.

Physical interpretation of the effective model and flat-bands The renormalization into a 2×22\times2 matrix not only simplifies the mathematics but has a profound physical meaning. What we achieved is the decoupling of the AA and BB bipartite lattices for each graphene layer. In graphene, such squared Hamiltonian is equivalent to renormalize the honeycomb lattice into a triangular lattice with renormalized interactions and a self-energy [27, 28], here Eq. (6) shows that this is also the case. As sketched in Fig. 2 for monolayer graphene, the transformation of EE to E2E^2 produces a fold in the spectrum such that bonding states with the lowest energy are mapped into higher energies. In other words, the transformation deletes the alternating sign of one of the bipartite sub-lattices as explained in Figure 2.

Figure 2. For each non-interacting monolayer graphene case, we present the density of states ρ(E)\rho(E) and of ρ(E2)\rho(E^{2}) corresponding to the squared Hamiltonian, equivalent to a folding of the spectrum. The signs are a sketch of the relative signs and amplitudes on bipartite sublattices A and B (shown in red and blue). Arrows indicate how the states are mapped under such transformation. Zero modes and close-by states are in the antibonding limit on a triangular lattice (indicated by the triangle inside the hexagon). The frustration is seen here in the bond that joins the + and + sites in the triangle (in reality, the minimal phase difference is ϕ\phi). For the twisted bilayer with interlayer interaction, H2\mathcal{H}^{2} produces a similar DOS folding and relative signs for the superlattice.

As the zero-modes are ground states of H2\mathcal{H}^2, we end up having a clear picture of nearby states around zero-energy modes, they correspond to the antibonding limit in two coupled triangular lattices. Anti-bonding states in non-bipartite lattices are frustrated as they cannot achieve a phase difference of π\pi between sites as odd-rings are present [29, 30]. For disordered systems, states are localized in regions of lower frustration and a kind of Lifshitz tail appears [30]. Moreover, frustration is always associated with massive degeneration [29, 30]; it leads to Van Hove singularities or if possible, in a condensation of confined states. These confined states appear in chiral models of the Penrose lattice [31, 32, 33], where they form beautiful fractal patterns [32], in random binary alloys [34, 35] and in graphene with defects [27, 36]. Strictly confined states are degenerate. However, there is a basis in which the amplitude in one of the bipartite sublattices is zero while in the other, the sum of all neighbors amplitudes is always zero for any site [32, 34].

To understand how zero modes are related with confined states and frustration as in Figure 2, let us made the following remarks. From the Hamiltonian Eq. (1) we confirm that for E=0E = 0 there are always solutions of the form ΦkA(r)=(ψk,1(r),ψk,2(r),0,0)\Phi^{A}_{k}(r) = (\psi_{k,1}(r),\psi_{k,2}(r),0,0) and ΦkB(r)=(0,0,χk,1(r),χk,2(r))\Phi^{B}_{k}(r) = (0,0,\chi_{k,1}(r),\chi_{k,2}(r)), where the labels A and B are used to denote zero amplitude in the opposite bipartite sublattice. We remark that linear combinations,

Φk(r)=12(ΦkA(r)+eiγΦkB(r))(10)\Phi_{k}(r) = \frac{1}{\sqrt{2}}\left(\Phi^{A}_{k}(r) + e^{i\gamma}\Phi^{B}_{k}(r)\right) \tag*{(10)}

with γ\gamma a phase, result in a different basis which do not show zeros in one sublattice, as for example with the symmetrized/antisymmetrized cases γ=0,π\gamma= 0,\pi. As the potential does not brake the C3C_{3} symmetry, the states k=K,K′k = K,K' are always a E=0E = 0 solution for any α\alpha. As a conclusion, for α\alpha not a magical angle there are four linearly independent wavefunctions, as confirmed from a Wronskian analysis [17] and therefore, at any angle there are ”confined states” in the sense of Figure 2. At magic angles, the Wronskian of the solutions is zero and there are E=0E = 0 solutions at any kk, resulting in the flat band. Still, ΦkA(r)\Phi^{A}_{k}(r) and ΦkB(r)\Phi^{B}_{k}(r) are solutions meaning that now all states are ”confined”. Any linear combination using different sets of kk is a solution. As explained in the supplementary, this is similar to the Van-Hove singularity in monolayer graphene, where dimers are disconnected from the lattice and thus is a kind of highly-degenerated, confined state.

The anti-bonding or bonding nature and therefore frustration is obtained from all bonds energy contribution, the latter one obtained from the product of the wave-function in a site with the conjugated wavefunction of a neighboring site [35]. For the present analysis, this requires to take into account three factors: 1) the system has two layers, 2) we are dealing with a low-energy continuous version of the TBH and 3) the system has a superlattice. Concerning point 1), we look at the intralayer frustration to see how the interlayer interaction tunes such contribution. Points 2) and 3) are more delicate as we need to understand what kk is a moment that departs from KK and K′K'. In a two-layer bipartite continuous lattice, for a given state kk such procedure is equivalent to consider bonds joining A and B sublattices sites. Although this can be made using any basis, it is easier to use in a symmetrized one γ=0\gamma= 0. The frustration can be measured from the function (see supplementary),

gk(r)=ψk,1(r)χk,1∗(r)+ψk,2(r)χk,2∗(r)(11)g_{k}(r) = \psi_{k,1}(r)\chi^{*}_{k,1}(r) + \psi_{k,2}(r)\chi^{*}_{k,2}(r) \tag*{(11)}

and ∇gk(r)\nabla g_{k}(r). As we are interested in states near E=0E = 0, we set k=Kk = K in Eq. (11). Using the symmetry of the problem, we can show that χK∗(r)=ψK(−r)\chi^{*}_{K}(r) = \psi_{K}(-r) and,

gK(r)=ψK,1(r)ψK,1(−r)+ψK,2(r)ψK,2(−r)∼vF(α)(12)g_{K}(r) = \psi_{K,1}(r)\psi_{K,1}(-r) + \psi_{K,2}(r)\psi_{K,2}(-r) \sim v_{F}(\alpha) \tag*{(12)}

where this last step is obtained from the fact that gKg_{K} turns out to be an invariant (see supplementary) which can be identified with the Fermi velocity at a given angle [15], here denoted by vF(α)v_{F}(\alpha). For any angle ∇gK(r)=0\nabla g_{K}(r) = 0 indicating that K=0K = 0 is minimally frustrated. However, at magic angles vF(α)=0v_{F}(\alpha) = 0 and therefore also gK=0g_{K} = 0, i.e., frustration is zero, making a flat band by pushing all states towards E=0E = 0. This leads to a special condition for the wave function Fourier components. The explicit form of the wavefunction is,

ψk(r)=∑m,n(am,nbm,neiq1⋅r)ei(Kmn+k)⋅r(13)\psi_{k}(r) = \sum_{m,n}\left(\frac{a_{m,n}}{b_{m,n}}e^{i\mathbf{q}_{1}\cdot r}\right)e^{i(\mathbf{K}_{mn}+k)\cdot r} \tag*{(13)}

where Kmn=mb1+nb2\mathbf{K}_{mn} = m\mathbf{b}_{1} + n\mathbf{b}_{2}. am,na_{m,n} and bm,nb_{m,n} are the Fourier coefficients. At the BABA stacking point r0=(a1−a2)/3r_{0} = (a_{1}-a_{2})/3, and for any α\alpha and due to symmetry reasons ψK,2(r0)ψK,2(−r0)=0\psi_{K,2}(r_{0})\psi_{K,2}(-r_{0}) = 0, while for magic α\alpha we have that [15] ψK,1(r0)=0\psi_{K,1}(r_{0}) = 0. As shown in the supplementary, when translated into Eq. (13) results in,

∑m,s(am,3s+am+1,3seiϕ+am,3s+1e−iϕ)=0(14)\sum_{m,s}\left(a_{m,3s} + a_{m+1,3s}e^{i\phi} + a_{m,3s+1}e^{-i\phi}\right) = 0 \tag*{(14)}

and the same equation holds for bmnb_{mn}, a fact to be expected as the Wronskian is zero [17]. The previous equation shows a precise tuning of components and hints how states develop a sharp peak at the AA stacking point as α→∞\alpha\to\infty (see supplementary).

Quite surprisingly, our renormalization is akin to a phonon problem (see supplementary). Thus, flat-band states seem to have a resemblance to floppy modes in an equivalent rigidity-phonon problem, i.e., the flat-band can be interpreted as a massive zero frequency vibrational band, since E2E^2 is analogous to a frequency [37]. These floppy modes are well known in the Phillips rigidity theory of glasses [38, 39, 40] and are reminiscent of the protected electronic boundary modes that occur in the quantum Hall effect and in topological insulators[37].

Effective fields and Hamiltonian in triangular coordinates

It is useful to write the renormalized twisted graphene bilayer Hamiltonian in terms of the Pauli matrix vector σ^\hat{\boldsymbol{\sigma}} as follows.

H2=h0(r)σ0+h(r)⋅σ^(15)H^2 = h_0(r)\sigma_0 + \boldsymbol{h}(r) \cdot\hat{\boldsymbol{\sigma}} \tag*{(15)}

where the vector h(r)\boldsymbol{h}(r) is,

h(r)=(αhx(r),αhy(r),α2hz(r))(16)\boldsymbol{h}(r) = \left(\alpha h_x(r), \alpha h_y(r), \alpha^2 h_z(r)\right) \tag*{(16)}

with,

hx(r)=−∑μ[kθsin⁡(qμ⋅r)+2icos⁡(qμ⋅r)q^μ⊥⋅∇]h_x(r) = -\sum_{\mu}\left[k_\theta\sin\left(\boldsymbol{q}_\mu\cdot r\right) + 2i\cos\left(\boldsymbol{q}_\mu\cdot r\right)\hat{\boldsymbol{q}}_\mu^{\perp} \cdot\nabla\right]
hy(r)=∑μ[kθcos⁡(qμ⋅r)−2isin⁡(qμ⋅r)q^μ⊥⋅∇]h_y(r) = \sum_{\mu}\left[k_\theta\cos\left(\boldsymbol{q}_\mu\cdot r\right) - 2i\sin\left(\boldsymbol{q}_\mu\cdot r\right)\hat{\boldsymbol{q}}_\mu^{\perp} \cdot\nabla\right]
hz(r)=∣U(−r)∣2−∣U(r)∣22(17)h_z(r) = \frac{\left|U(-r)\right|^2-\left|U(r)\right|^2}{2} \tag*{(17)}

The operator in front of the identity σ0\sigma_0 is,

h0(r)=−∇2+α2V‾(r)(18)h_0(r) = -\nabla^2 + \alpha^2\overline{V}(r) \tag*{(18)}
V‾(r)=∣U(r)∣2+∣U(−r)∣22(19)\overline{V}(r) = \frac{\left|U(r)\right|^2+\left|U(-r)\right|^2}{2} \tag*{(19)}

We observe that h0(r)h_0(r) corresponds to a Hamiltonian with an average potential. The Hamiltonian structure is akin to the one found in Ref. [41]. Observe that the term h(r)\boldsymbol{h}(r) contains all the topological properties of the operator. Let us further simplify this Hamiltonian. Define ψ±(r)=ψ1(r)±ψ2(r)\psi_{\pm}(r)=\psi_1(r)\pm\psi_2(r). The Schrödinger equation is transformed into Heff(ψ+(r),ψ−(r))T=E2(ψ+(r),ψ−(r))TH_{\mathrm{eff}}(\psi_+(r),\psi_-(r))^T=E^2(\psi_+(r),\psi_-(r))^T. The stated effective Hamiltonian is,

Heff=(−∇2+Veff(r)Aeff†(r)Aeff(r)−∇2+Veff(r))(20)H_{\mathrm{eff}} = \begin{pmatrix} -\nabla^2+V_{\mathrm{eff}}(r) & A_{\mathrm{eff}}^{\dagger}(r) \\ A_{\mathrm{eff}}(r) & -\nabla^2+V_{\mathrm{eff}}(r) \end{pmatrix} \tag*{(20)}

Here we defined the effective potentials as,

Veff(r)=α2V‾(r)+αhx(r)(21)V_{\mathrm{eff}}(r)=\alpha^2\overline{V}(r)+\alpha h_x(r) \tag*{(21)}

and,

Aeff(r)=α2hz(r)+iαhy(r)(22)A_{\mathrm{eff}}(r)=\alpha^2 h_z(r)+i\alpha h_y(r) \tag*{(22)}

In the supplementary section, we show how this system can be simplified using triangular coordinates to map the problem into a friendly rectangular domain.

Conclusions

We showed that the chiral Hamiltonian for twisted graphene bilayers can be written into a 2×22 \times2 matrix operator. The action of the proposed renormalization maps the zero-mode region into the ground state. Modes next to zero energy have an antibonding nature in a triangular lattice and at zero energy are similar to confined modes observed in many other bipartite systems [33, 34, 35]. A surprising result is that our renormalization suggests that flat-bands are somehow analogous to floppy modes in rigidity phonon models [35, 37, 42].

We thank UNAM-DGAPA project IN102620 and CONACyT project 1564464.

References

References

  1. [1]S. Nishijima, S. Eckroak, A. Marian, K. Choi, W. S. Kim, M. Terai, Z. Deng, J. Zheng, J. Wang, K. Uemoto, J. Du, P. Febvre, S. Keenan, O. Mukhanov, L. D. Cooley, C. P. Foley, W. V. Hassenzahl, and M. Izumi, Superconductivity and the environment: a Roadmap, Superconductor Science Technology 26, 113001 (2013).DOI
  2. [2]S. Kudo, T. Yoshida, S. Ideta, K. Takashima, H. Anzai, T. Fujita, Y. Nakashima, A. Ino, M. Arita, H. Namatame, M. Taniguchi, K. M. Kojima, S. Uchida, and A. Fujimori, Temperature evolution of correlation strength in the superconducting state of high-$T_c$ cuprates, Phys. Rev. B 92, 195135 (2015).DOI
  3. [3]B. Keimer, S. Kivelson, and M. e. a. Norman, From quantum matter to high-temperature superconductivity in copper oxides, Nature 518, 179–186 (2015).DOI
  4. [4]T. Wang, N. F. Q. Yuan, and L. Fu, Moiré surface states and enhanced superconductivity in topological insulators, arXiv e-prints, arXiv:2010.09753 (2020), arXiv:2010.09753 [cond-mat.supr-con].DOI
  5. [5]K. Yamazaki, M. Ochi, D. Ogura, K. Kuroki, H. Eisaki, S. Uchida, and H. Aoki, Superconducting mechanism for the cuprate Ba₂Cu₃O₃₊δ based on a multiorbital Lieb lattice model, Phys. Rev. Research 2, 033356 (2020).
  6. [6]M.-S. Nam and A. A. Oudovenko, Universal limiting transition temperature for the high $T_c$ superconductors, arXiv e-prints , arXiv:2010.00572 (2020), arXiv:2010.00572 [cond-mat.supr-con].arxiv.org/abs/2010.00572
  7. [7]S. Tan, Y. Liu, Y. Mou, and S. Feng, Anisotropic dressing of electrons in electron-doped cuprate superconductors, Phys. Rev. B 103, 014503 (2021).DOI
  8. [8]B. Kyung, D. Sénéchal, and A.-M. S. Tremblay, Pairing dynamics in strongly correlated superconductivity, Phys. Rev. B 80, 205109 (2009).
  9. [9]M. Capone, M. Fabrizio, C. Castellani, and E. Tosatti, Strongly Correlated Superconductivity, Science 296, 2364 (2002).DOI
  10. [10]Y. Cao, V. Fatemi, S. Fang, K. Watanabe, T. Taniguchi, E. Kaxiras, and P. Jarillo-Herrero, Unconventional superconductivity in magic-angle graphene superlattices, Nature 556, 43 (2018).DOI
  11. [11]M. Yankowitz, S. Chen, H. Polshyn, Y. Zhang, K. Watanabe, T. Taniguchi, D. Graf, A. F. Young, and C. R. Dean, Tuning superconductivity in twisted bilayer graphene, Science 363, 1059 (2019).DOI
  12. [12]A. Kerelsky, L. McGilly, and D. e. a. Kennes, Maximized electron interactions at the magic angle in twisted bilayer graphene, Nature 572, 95 (2019).DOI
  13. [13]R. Bistritzer and A. H. MacDonald, Moiré bands in twisted double-layer graphene, Proceedings of the National Academy of Sciences 108, 12233 (2011).DOI
  14. [14]J. M. B. L. dos Santos, N. M. R. Peres, and A. H. C. Neto, Graphene bilayer with a twist: Electronic structure, Phys. Rev. Lett. 99, 256802 (2007).
  15. [15]G. Tarnopolsky, A. J. Kruchkov, and A. Vishwanath, Origin of magic angles in twisted bilayer graphene, Phys. Rev. Lett. 122, 106405 (2019).DOI
  16. [16]J. Wang, Y. Zheng, A. J. Millis, and J. Cano, Chiral Approximation to Twisted Bilayer Graphene: Exact Intra-Valley Inversion Symmetry, Nodal Structure and Implications for Higher Magic Angles, arXiv e-prints , arXiv:2010.03589 (2020), arXiv:2010.03589 [cond-mat.mes-hall].DOI
  17. [17]F. K. Popov and A. Milekhin, Hidden wave function of twisted bilayer graphene: Flat band as a Landau level, arXiv e-prints , arXiv:2010.02915 (2020), arXiv:2010.02915v1 [cond-mat.mes-hall].DOI
  18. [18]K. Hejazi, C. Liu, and L. Balents, Landau levels in twisted bilayer graphene and semiclassical orbits, Phys. Rev. B 100, 035115 (2019).DOI
  19. [19]A. Uri, S. Grover, Y. Cao, J. Á. A. Crosse, K. Bagani, D. Rodan-Legrain, Y. Myasoedov, K. Watanabe, T. Taniguchi, P. Moon, M. Koshino, P. Jarillo-Herrero, and E. Zeldov, Mapping the twist-angle disorder and Landau levels in magic-angle graphene, Nature (London) 581, 47 (2020).
  20. [20]X.-L. Qi and S.-C. Zhang, Topological insulators and superconductors, Reviews of Modern Physics 83, 1057 (2011).DOI
  21. [21]L. Zou, H. C. Po, A. Vishwanath, and T. Senthil, Band structure of twisted bilayer graphene: Emergent symmetries, commensurate approximants, and Wannier obstructions, Phys. Rev. B 98, 101103/Phys-RevB.98.085435 (2018).DOI
  22. [22]C. Xu and L. Balents, Topological superconductivity in twisted multilayer graphene, Phys. Rev. Lett. 121, 087001 (2018).DOI
  23. [23]M. Fidrysiak, M. Zegrodnik, and J. Spałek, Unconventional topological superconductivity and phase diagram for an effective two-orbital model as applied to twisted bilayer graphene, Phys. Rev. B 98, 085436 (2018).DOI
  24. [24]Y.-Z. You and A. Vishwanath, Superconductivity from valley fluctuations and approximate SO(4) symmetry in a weak coupling theory of twisted bilayer graphene, npj Quantum Materials 4, 16 (2019).DOI
  25. [25]E. Khalaf, S. Chatterjee, N. Bultinck, M. P. Zaletel, and A. Vishwanath, Charged Skyrmions and Topological Origin of Superconductivity in Magic Angle Graphene, arXiv e-prints , arXiv:2004.00638 (2020), arXiv:2004.00638 [cond-mat.str-el].DOI
  26. [26]M. Rodriguez-Vega, M. Vogl, and G. A. Fiete, Floquet engineering of twisted double bilayer graphene, Phys. Rev. Research 2, 033494 (2020).
  27. [27]G. G. Naumis, Internal mobility edge in doped graphene: Frustration in a renormalized lattice, Phys. Rev. B 76, 153403 (2007).
  28. [28]J. E. Barrios-Vargas and G. G. Naumis, Doped graphene: the interplay between localization and frustration due to the underlying triangular symmetry, Journal of Physics: Condensed Matter 23, 375501 (2011).DOI
  29. [29]M. C. Fumiko Yonezawa, Theory of electronic properties of amorphous semiconductors, in Fundamental Physics of Amorphous Semiconductors: Proceedings of the Kyoto Summer Institute Kyoto, Japan, September 8–11, 1980, Springer Series in Solid-State Sciences, Vol. 46, edited by F. Y. Fumiko Yonezawa (Springer-Verlag Berlin Heidelberg, 1981) 1st ed., pp. 119–144.DOI
  30. [30]M. Cohen, Topology, geometry, elementary excitations and physical properties of disordered materials, in Topological Disorder in Condensed Matter, Springer Series in Solid-State Sciences, Vol. 25, edited by F. Y. Hellmut Fritzsche (Springer-Verlag Berlin Heidelberg, 1983) 1st ed., pp. 122–141.DOI
  31. [31]M. Kohmoto and B. Sutherland, Electronic states on a penrose lattice, Phys. Rev. Lett. 56, 2740 (1986).pubmed.ncbi.nlm.nih.gov/10033078
  32. [32]M. Arai, T. Tokihiro, T. Fujiwara, and M. Kohmoto, Strictly localized states on a two-dimensional penrose lattice, Phys. Rev. B 38, 1621 (1988).pubmed.ncbi.nlm.nih.gov/9946442
  33. [33]G. G. Naumis, R. A. Barrio, and C. Wang, Effects of frustration and localization of states in the penrose lattice, Phys. Rev. B 50, 9834 (1994).pubmed.ncbi.nlm.nih.gov/9975063
  34. [34]S. Kirkpatrick and T. P. Eggarter, Localized states of a binary alloy, Phys. Rev. B 6, 3598 (1972).DOI
  35. [35]G. G. Naumis, C. Wang, and R. A. Barrio, Frustration effects on the electronic density of states of a random binary alloy, Phys. Rev. B 65, 134203 (2002).DOI
  36. [36]J. Barrios-Vargas and G. G. Naumis, Pseudo-gap opening and dirac point confined states in doped graphene, Solid State Communications 162, 23 (2013).DOI
  37. [37]C. L. Kane and T. C. Lubensky, Topological boundary modes in isostatic lattices, Nature Physics 10, 39 (2014).DOI
  38. [38]J. C. Phillips, Topology of covalent non-crystalline solids i: Short-range order in chalcogenide alloys, Journal of Non-Crystalline Solids 34, 153 (1979).DOI
  39. [39]A. Huerta and G. G. Naumis, Evidence of a glass transition induced by rigidity self-organization in a network-forming fluid, Phys. Rev. B 66, 184204 (2002).DOI
  40. [40]H. M. Flores-Ruiz, G. G. Naumis, and J. C. Phillips, Heating through the glass transition: A rigidity approach to the boson peak, Phys. Rev. B 82, 214201 (2010).
  41. [41]P. San-Jose, J. González, and F. Guinea, Non-abelian gauge potentials in graphene bilayers, Phys. Rev. Lett. 108, 216802 (2012).
  42. [42]S. Dasgupta and O. Tchernyshyov, Theory of spin waves in a hexagonal antiferromagnet, Phys. Rev. B 102, 144417 (2020).DOI

Paper details

vlnc:W_abpcja6q
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