Probing miniband structure and Hofstadter butterfly in gated graphene superlattices via magnetotransport

The presence of periodic modulation in graphene leads to a reconstruction of the band structure and formation of minibands. In an external uniform magnetic field, a fractal energy spectrum called Hof-stadter butterfly is formed. Particularly interesting in this regard are superlattices with tunable modulation strength, such as electrostatically induced ones in graphene. We perform quantum transport modeling in gate-induced square two-dimensional superlattice in graphene and investigate the relation to the details of the band structure. At low magnetic field the dynamics of carriers reflects the semi-classical orbits which depend on the mini band structure. We theoretically model transverse magnetic focusing, a ballistic transport technique by means of which we investigate the minibands, their extent and carrier type. We find a good agreement between the focusing spectra and the mini band structures obtained from the continuum model, proving usefulness of this technique. At high magnetic field the calculated four-probe resistance fit the Hofstadter butterfly spectrum obtained for our superlattice. Our quantum transport modeling provides an insight into the mini band structures, and can be applied to other superlattice geometries.


Introduction
Graphene, a 2D material characterized by a linear low-energy dispersion relation, hosts charge carriers named Dirac fermions due to the resemblance of relativistic (massless) particles described by the Dirac equation.Modifying the underlying graphene lattice by a smooth periodic potential can affect the band structure through folding of the pristine graphene Dirac cone into mini bands [1], formation of the secondary Dirac points, and anisotropic renormalization of velocity [2][3][4][5][6].Periodic modulation has been obtained in graphene through chemical functionalization [7], placing graphene on selfassembled nanostructures [8], and by stacking graphene together with aligned or slightly misoriented hexagonal boron nitride (hBN), resulting in periodic moiré modulation which generates hexagonal superlattices (SLs) [9][10][11][12][13][14][15].Moiré SLs were also created in low-angle twisted graphene bilayers , followed by van der Waals structures made up of few layers of graphene [16][17][18][19][20][21] and other 2D materials [22,23].SLs are suitable for the observation of the self-similar energy spectrum called Hofstadter butterfly [13,14] and Brown-Zak oscillations [24,25] that occur when the magnetic flux through the superlattice unit cell is of the order of the magnetic flux quantum, ϕ 0 = h/e, and in pristine 2D crystals require unattainable magnetic fields.Also worth mentioning are the many-body phenomena present in moiré SLs [26,27].
Despite their potential, artificial lattices tailored by chemical methods or moiré SLs suffer from inability to tune the strength of the periodic potential.In moiré SLs the period can be tuned to some extent via the rotation angle between the stacked layers, but they are inherent of a hexagonal symmetry.Precise control over the SL geometry, period, and strength is vital for the band structure engineering.The above limitations can be circumvented in electrostatic gate-induced SLs that allow an arbitrary design via the gates geometry, with the gate voltage being a knob for the potential strength.The experimental attempts to create gated SLs in graphene included 1D arrays of metal gates [28][29][30], followed by patterned dielectric substrates [31,32], and few-layer graphene patterned bottom gates for 1D [33,34] and 2D SLs [35,36] with down to sub-20 nm periods [34] manifesting the flexibility of this approach.With the recent advance in the fabrication techniques, gated SLs with the period of a few tens of nanometers are achievable with good device quality and long electron mean free path [32].This enabled observation of commensurability oscillations [32,33,36], Hofstadter butterfly [34,35] and Brown-Zak oscillations [36] at magnetic fields of the order of a few tesla.This is more affordable compared to about 25 T required for graphene/hBN SL, where, due to small lattice constant mismatch of 1.8%, the SL period can reach up to 14 nm for aligned lattices.
While transport in quantizing magnetic field in gated SLs has been thoroughly studied, the intermediate magnetic field regime remained mostly unexplored.In the semiclassical treatment, at low magnetic field fermions undergo cyclotron motion that can be probed in transport measurements via transverse magnetic focusing (TMF).This technique has been used to experimentally probe the band structure in pristine mono-, bi-, and tri-layer graphene [37], graphene/WSe 2 heterostructures [38], and moiré superlattices [39,40], and theoretically considered in graphene pn junctions [41,42] and 1D gated SLs [6].
In this work, we perform a theoretical study of the TMF in 2D gate-induced square SLs, and analyze the relation between the observed TMF spectra and the miniband structure.This is complemented by the investigation of magnetotransport in the quantum Hall regime, where we observe signatures of Hofstadter butterfly, matching the numerical Hofstadter spectrum calculated for our gated SL.Our study is performed using quantum transport calculations for multiterminal structures, considering realistic experimental conditions.To model a realistic geometry, we consider SL induced by a patterned dielectric substrate [31,32] with a uniform global backgate underneath the dielectric layer, and the graphene sheet sandwiched between two hBN layers lying on top [Figure 1 top gate.The voltage applied to the back gate V bg controls the strength of the periodic modulation, while the top gate voltage V tg is used to tune the carrier density across the SL.We follow the design of Ref. [31], with a square lattice etched in SiO 2 substrate, with a lattice period λ = 35 nm [Figure 1(b)].For the ease of the calculation, we use a model function C bg (x, y) [see Figure 1(b)] that approximates the electrostatically simulated capacitance, obtained previously by some of us [43].Previous studies [35,44] and modeling of previous experiment on magnetic focusing in monolayer graphene [37] show good agreement between experiment and simulation for graphene superlattice devices (see Methods), and the present purely theoretical work can be regarded as a guide for further experimental magneto-transport studies.

No external magnetic field
We first simulate the four-point longitudinal resistance R xx .We consider a four terminal device shown in the right inset of Figure 1(c), where the system length L = 1152 nm, width W = 385 nm, and the top lead width w = 245 nm.
With the four leads labeled in the right inset of Figure 1(c), we calculate R xx = R 14, 23 (for details see Methods) and show its dependence on the top and back gate voltages in the main panel of Figure 1(c).As can be seen from the map, the strength of the backgate mostly influences the superlattice modulation.Figure 1(d) presents the linecuts of R xx marked in Figure 1(c) with the respective colors.Whereas at V bg ≈ 0 only a single Dirac peak is visible, for increasing |V bg | second and higher order satellite Dirac peaks start to appear, as the periodic modulation gets stronger.At V bg = ±40 V [linecuts in Figure 1(d)], a few higher-order Dirac points are resolved.On the other hand, changing the top gate voltage mostly tunes the carrier density in the device.The left inset of Figure 1(c) shows a close-up of the boxed region with 30 V ≤ V bg ≤ 45 V and −3.5 V ≤ V tg ≤ −1.6 V, where two sharp lines are visible at both sides of the main Dirac peak, corresponding to the secondary Dirac points, and several fainter lines, corresponding to higher order Dirac points.Similar results based on the same capacitance model function have been reported in [43], where two-terminal transport simulations were performed.The character of the bands can be verified in magnetotransport, as shown in the following subsections.

Low magnetic field
For a general band dispersion ϵ(k), the semiclassical equations of motion for an electron are given by [45]  where −e is the electron charge, E is the external electric field, and B the magnetic field.In the presence of constant out-of-plane magnetic field B = (0, 0, B) only, one can obtain the relation between the shape of the Fermi contour in the momentum space ∆k(t) and the carrier trajectory in real space ∆r(t) [45] ∆r meaning that the cyclotron orbit is obtained by rotating the orbit in the momentum space by 90 • clockwise, as illustrated in Figure 2 In a typical device designed for transverse magnetic focusing measurement, an emitter and collector [in Figure 2(a) marked as 1 and 2, respectively] are located on the same edge at a center to center distance ℓ = d + w, and other contacts act as absorbers.The current injected from an emitter flows along cyclotron orbits with a radius R c depending on the magnetic field strength, and at the boundary propagates along skipping orbits.The current can end up in the collector, when the diameter or its multiples match ℓ [Figure 2(b)], or otherwise in absorber contacts.In the nonlocal resistance measurement (f) [Figure 2(a)], this results in maximum or minimum, respectively, of the resistance R f = R 14,23 (see Methods).For electron-like (hole-like) orbits, the resistance maximum condition can be obtained for positive (negative) B.
In pristine graphene, a typical TMF spectrum as a function of magnetic field and voltage contains two fans of focusing peaks, one for the electron band and the other for the hole band [37].In a superlattice, the emerging replicas of the Dirac cone cause a substantial modification of the TMF signal.In the following discussion, we choose V bg = 40 V, such that a few higher-order Dirac peaks are present next to the main Dirac point as seen in Figure 1(c).
For transverse magnetic focusing, we choose the system geometry shown in Figure 2(a), with the distance between the bottom leads d = 1200 nm, their widths w = 100 nm, the side leads width W = 1636 nm, and the top lead width L = 1792 nm.
Figure 3(a) shows the R f map as a function of B and V tg .One can see several series of fans, with the focusing peaks appearing at B > 0 or B < 0. The map is put together with the R xx calculated at B = 0, shown in Figure 3(b), where the main and secondary Dirac peaks are seen.The sign change of the focusing peaks in Figure 3(a) occurs either at the Dirac points, or van Hove singularities, and for the former, it coincides with the R xx peaks.This confirms the sign change of the carriers when tuning V tg , occurring as an effect of the band reconstruction due to the superlattice potential.To understand the result in detail, in Figure 3(c)-Figure 3(f) we plot the miniband structures calculated at B = 0, as described in Ref. [43], and the Fermi contours at E = 0, at selected values of V tg marked by arrows in Figure 3(a).Additionally, in Figure 3(g)-Figure 3(k) we show zoomed regions of the R f map marked with the colored rectangles in Figure 3(a).
In Figure 3(c), at V tg = −2.75V, the Fermi level is located at the C 1 subband, and the Fermi contour has a rounded shape [Figure 2(b) and Figure 3(c)].In the semiclassical description, electrons injected from lead 1 [Figure 2(a)] with an initial velocity v = (0, v) and k = (0, k), in a moderate magnetic field travel along a rounded trajectory, and after half a period, t = T /2 encircle half of the closed orbit.From Eq. ( 3), this corresponds to traveling a distance equal to the diameter, 2R c = 2ℏk/eB along the x direction [Figure 2(b)].For ℓ = 2R c , it leads to the first maximum of R f .For smaller R c , the beam is reflected at the edge, and can flow to the collector when ℓ = 4R c , 6R c , . . ., giving rise to higher order R f peaks.In general, we can evaluate the field at which the jth maximum occurs as B j = 2ℏjk/eℓ, j = 1, 2, . . . .We find k(V tg ) numerically and plot B j with dashed lines in Figure 3(g).The C 1 subband cone is within −2.9 V ≲ V tg ≲ −2.5 V. We find a very good agreement with the R f signal for up to j ≈ 7. Higher j are not resolved as the system enters the quantum Hall regime, and semiclassical description of the skipping orbits at the edge no longer applies.At −3.1 V ≲ V tg ≲ −2.9 V, for the V 1 subband, R f is noisy due to scattering of low-energy carriers by the periodic potential.
When the Fermi level is tuned to the van Hove singularity (at V tg ≈ −2.5 for the electron subband, and V tg ≈ −3.1 for the hole subband), the focusing signal vanishes, and smaller fans reappear.Based on the miniband structures [Figure 3(d)], we interpret them as due to focusing of the secondary Dirac cones fermions.For example, at V tg = −3.12V [Figure 3(d)] at E = 0 there are tiny Dirac cones around the M and X points of the Brillouin zone.
For V tg ≲ −3.2 V and V tg ≳ −2.4 V, the miniband structures around the Fermi level get more complex, with many overlapping subbands.Nevertheless, we find ranges of V tg where a single isolated higher-order Dirac cone is present, giving clear focusing signal [see Figure 3(e) for the V 2 subband, the corresponding R f zoom in Figure 3(h), and the zoom in Figure 3(i) for the C 2 subband].When there are more overlapping subbands, the signal gets very faint.Nevertheless, one can spot fans that fit well to the hole-like orbit within the V 4 subband around the M point, see Figure 3(f) at V tg = −3.85V, and zoomed R f in Figure 3(j).A similar feature is resolved in Figure 3(k) for the electron-like orbits.
To further illustrate the relation of the real-space orbits to the subbands, we calculate the current density maps at selected focusing peaks.Figure 4 These semiclassical orbits show similarity with the current density, but the current density is obtained from quantum calculation so they are expected to be similar but not strictly identical, in particular, counter-intuitive patterns may occur at certain resonant conditions (e.g. a vertical blob on top of the rhombus-like pattern in Figure 4(h)).Note that in Figure 4(f)-Figure 4(g) the current density is non-zero in the area between the vertical segments, as it contains contributions from multiple initial k x for which v x is low.The noisy background visible in Figure 4(a)-Figure 4(c) originates from scattering and resonant states due to SL which are irregular and complex due to the wave-like nature of the carriers.
Although the focusing spectrum is not symmetric with respect to the main Dirac point, it is qualitatively similar in minibands above and below the main Dirac point, except for the noisy signal for low-energy valence subbands.Let us note that the modulation induced by electrostatic gates is a complex function of V bg and V tg , and the band structure is significantly modified merely by changing V tg [Figure 3(c)-Figure 3(f)].This is in contrast to the SLs induced via low-angle twisted hBN substrate or twisted graphene layers, where the top gate only sweeps the carrier density without affecting the periodic modulation, and the band structure remains unchanged.This leads to the shape of the fans in R f closer to straight lines, unlike Refs.[39,40] that observed ones close to parabolic.

High magnetic field
Now, we turn our attention to the high-field regime, where we expect the Hofstadter spectrum to emerge when the magnetic flux per SL unit cell of area A, ϕ = AB, is of the order of the flux quantum ϕ 0 = h/e.To simulate longitudinal resistance R xx and Hall resistance R xy at the same time, while keeping calculations for the four-point resistances minimized, we consider a 5-terminal Hall bar sketched in Figure 5(a) with the actually considered geometric dimensions shown.We compute all the 5 × 4 = 20 transmission functions between all pairs among the five leads, and process the data to obtain longitudinal resistance R xx = R 14,23 and Hall resistance R xy = R 14,52 following the Büttiker formalism [48,49], according to the lead labels shown in Figure 5(a).For all the following discussions, we convert our V tg axis to the numerically obtained average carrier density n over the entire lattice, in order for a more transparent presentation of our high-field transport simulations.xy as a function of n and magnetic field B, with the back gate voltage fixed at V bg = 0 and V bg = 40 V, respectively.To better highlight the Landau fans arising from the quantum Hall effect of graphene, we limit the color range to −40 ≤ G H /G 0 ≤ +40 in Figure 5(b), where G 0 = e 2 /h is the conductance quantum.The same limit of the color range is applied to Figure 5(c) in order for a direct comparison.Line cuts at B = 2 T marked by the black line on Figure 5(b) and orange line on Figure 5(c) are shown and compared in Figure 5(d).The lowest few quantum Hall conductance plateaus G H = ±2G 0 , ±6G 0 , ±10G 0 , • • • can be clearly seen in the V bg = 0 case free of superlattice potential (black line), while the combined strong magnetic field (B = 2 T) and strong superlattice modulation (V bg = 40 V) result in a more complex transport feature, in accordance with the predictions [50,51] for a periodic 2D modulation, which can be better understood by checking R xx , instead of R xy (or G H ), to be discussed soon below.
Without taking the inverse of R xy , Figure 5(e) shows the Hall resistance at B = 0.2 T with V bg = 0 [red, corresponding to the n range marked on Figure 5(b)] and V bg = 40 V [cyan, corresponding to the n range marked on Figure 5(c)].The former (without superlattice) shows a single sign change at n = 0, typical for graphene, while the latter (with superlattice) shows multiple sign changes at positive and negative n, in addition to the main Dirac point at n = 0, consistent with our previous low-field magnetotransport simulations discussed above.
Considering the same range of n and B as Figure 5(b) and Figure 5(c), Figure 5(f) shows the longitudinal resistance R xx with the back gate voltage fixed at V bg = 40 V. Since the period of our gate-controlled superlattice is λ = 35 nm, and hence the square SL unit cell area A = λ 2 = 1225 nm 2 , the condition ϕ/ϕ 0 = AB/(h/e) = 1 is reached at B ≈ 3.376 T ≡ B 1 , which is in sharp contrast with the graphene/hBN moiré superlattice that requires B ≈ 24 T in order to reach ϕ/ϕ 0 = 1, because of its periodicity limited to λ ≲ 14 nm and hence the area A = √ 3λ 2 /2 ≲ 170 nm 2 .As is visible on Figure 5(f), the seemingly complicated R xx map does exhibit certain features at B 1 and B 1 /2, and vaguely at B 1 /3 (marked by black triangles), corresponding to ϕ/ϕ 0 = 1, 1/2, 1/3, respectively, consistent with our calculation of the magnetic energy subbands shown in Figure 5(g), where the vertical axis of ϕ/ϕ 0 corresponds to exactly the same B range as Figure 5(f), as well as the average density range.Good consistency between the R xx map of Figure 5(f) and the Hofstadter butterfly shown in Figure 5(g) can be seen.For methods adopted to calculate Figure 5(g), see Methods.

Discussion
In summary, we theoretically investigated transport in gated superlattices based on monolayer graphene.Our zero and low magnetic field transport calculations remain in a good agreement with the continuum model band structure calculated in presence of periodic modulation.We explored the potential of TMF for probing the intricate band structure of graphene with periodic modulation.It offers possibilities to study a plethora of phenomena in superlattices, and opens the door for studies of strongly correlated systems in twisted bilayer graphene [52] or in bilayer graphene superlattices [53].By exploring the reconstructed band structure via magnetotransport calculations it is possible to engineer devices relying on directed electron flow due to the distortion of Fermi contour, as well as for other applications based on mini band electron optics.We also obtained high-magnetic-field response consistent with the Hofstadter spectrum calculated for a gated superlattice as a function of the gate voltage.Our modeling can be generalized to other superlattice geometries, and is promising for the investigations of future band structure engineered devices working in a broad range of magnetic fields.

Gated superlattice model
To model a realistic geometry, we consider SL induced by a patterned dielectric substrate [31,32] with a uniform global backgate underneath the dielectric layer, and the graphene sheet sandwiched between two hBN layers lying on top [Figure 1(a) of the main text].The hBN/graphene/hBN sandwich is covered by a global top gate.The voltage applied to the back gate V bg controls the strength of the periodic modulation, while the top gate voltage V tg is used to tune the carrier density across the SL.We follow the design of Ref. [31], with a square lattice etched in SiO 2 substrate, with a lattice period λ = 35 nm.We use a model function C bg (x, y) where d smooth = 7.5 nm is the smoothness of the modulation, C out /e = 0.77 × 10 11 cm −2 V −1 and C in /e = 0.22 × 10 11 cm −2 V −1 , and a%b means the remainder after dividing a by b.The top gate capacitance is assumed to be C tg /e = 1 × 10 12 cm −2 V −1 .Using the parallel capacitor model, this roughly corresponds to the top hBN thickness d t ≈ 16.6 nm, from C tg /e = ε 0 ε hBN /ed t , where ε 0 is the vacuum permittivity, ε hBN = 3 is the dielectric constant of hBN, and −e is the electron charge.Figure 1(b) of the main text shows the profile of the above model function (5).
For the dual-gated graphene sample free from intrinsic doping, the carrier density is given by Assuming that the carrier energy in graphene is given by E = ±ℏv F k, where ℏ is the reduced Planck constant, v F ≈ 10 6 m/s is the Fermi velocity of graphene, and using ℏv F ≈ 3 √ 3/8 eVnm, the onsite potential energy can be obtained from in order to set the global Fermi energy at E = 0 where transport occurs.

Transport calculation
For transport calculation, we use the tight-binding Hamiltonian where c i (c † i ) is an annihilation (creation) operator of an electron on site i located at r i = (x i , y i ).The first sum contains the nearest-neighbor hoppings with the hopping parameter t ij , and the second sum describes the onsite potential energy profile.In the presence of an external magnetic field B = (0, 0, B), the hopping integral is modified by t ij → t ij exp(iϕ), where the Peierls phase ϕ = (−e/ℏ) rj ri A • dr, with A being the vector potential that satisfies ∇ × A = B, and the integral going from the site at r i to the site at r j .For a feasible simulation of realistic devices, we use the scalable tightbinding model [54], where the hopping parameter becomes t ij = t 0 /s f , and the lattice spacing a = a 0 s f , s f is the scaling factor, and we use t 0 = −3 eV and a 0 = 1/4 √ 3 nm.Transport calculations based on Hamiltonian (8) are done within the wave-function matching for the TMF, and real-space Green's function method in other cases, at the global Fermi energy E = 0 and zero temperature.The conductance from lead i to lead j is obtained from the Landauer formula G ji = 2(e 2 /h)T ji , where the transmission probability T ji is evaluated as a sum over the propagating modes T ji = q T q ji , and Here, t pq ij is the probability amplitude for the transfer from the incoming mode p in lead i to the outgoing mode q in lead j.
In the multiterminal devices, we solve the transport problem for each lead as an input, and build the conductance matrix G [48,49] which relates the current I i fed to the system in lead i to the voltage at j-th lead V j through For an N -terminal system, the matrix elements are We set the voltage at l-th lead equal to zero, and eliminate the l-th row and column of the matrix.The reduced (N − 1) × (N − 1) matrix Ḡ can be inverted to get R = Ḡ−1 , where the R matrix satisfies With the elements of matrix R, one can evaluate the resistance with the current flowing from lead k to lead l, zero current in other terminals, and voltage drop measured between leads m and n.

Transverse magnetic focusing
As a numerical example of applying the above outlined Landauer-Büttiker formalism for computing the four-point resistance Eq. ( 13), we revisit the first TMF experiment on graphene [37], considering the same probe spacing (ℓ = 500 nm) and width (100 nm) but slightly different geometry of the scattering region (total length 700 nm and width 500 nm) for a 6-terminal Hall bar, made of a graphene lattice scaled by s f = 12.Choosing the same configuration of the leads for injector, ground, and voltage probes as the revisited experiment, the computed R 61,54 as a function of the external magnetic field B perpendicular to the plane of graphene and the carrier density n is reported in Figure 6(a), showing a map rather consistent with the experiment.Due to the isotropic low-energy dispersion of graphene, the resulting cyclotron trajector is a simple circle of radius R c = ℏk/eB, which is simplified from Eq. (3).By requiring the probe spacing to be equal to an integer times the cyclotron diameter, ℓ = j • 2R c , j = 1, 2, • • • , together with k = π|n| for graphene, one can solve for carrier density corresponding to the jth peak of the TMF on the B-n map of Figure 6(a): The dashed lines on Figure 6

Choosing the gauge
In the presence of an external magnetic field and semi-infinite leads, the vector potential must satisfy the translational invariance of the leads.For the magnetic field along the ẑ axis, the most common choice is the Landau gauge, A = (−yB, 0, 0) or A = (0, xB, 0) for the lead which is translationally invariant along the x or y direction, respectively.For other lead orientation, in general, the proper gauge is different.Therefore, in a multi-terminal device, the required vector potential is not uniform in the entire space.This is not a problem since adding an arbitrary curl-free component to the vector potential does not change the magnetic field.Here, we use the approach introduced in [55].
Assuming that the proper gauge in the 1st lead is A 1 (r), for another lead that is at an angle θ n with respect to lead 1, the gauge can be chosen as A n (r) = A 1 (r) + ∇f n (r), (15) with The addition of a gradient of a scalar function does not influence the requirement ∇ × A = B.As an illustration of the transformation, consider A 1 (r) = (−yB, 0, 0), and θ n = 90 • .Then, f n (r) = Bxy, ∇f n (r) = (By, Bx, 0), and A n (r) = (−yB, 0, 0) + (yB, xB, 0) = (0, xB, 0).Applying the transformation (15) so that it only affects lead n is possible by defining a smooth step function ζ n (r) which is only nonzero in the translationally invariant part of lead n r in lead n, 0, r in lead m ̸ = n, smooth interpolation r elsewhere.(17) edge states are pushed to k x = 0, which results in the appearance of the states in the gaps.To lower the computational burden, we use the scalable tightbinding model [54] with s f ∼ 8 to calculate E kx=0 as a function of magnetic field for an armchair ribbon with periodic length along the x axis equal to the superlattice period (λ).Here, in order to ensure superlattice period equal to a multiple of 3a, s f is not an integer, and the ribbon width is larger than 20λ to show the superlattice effect.
In transport, only the states at the Fermi level contribute to the conductance.Therefore, we calculate Hofstadter butterfly spectra for all V tg values and collect the Fermi states to construct the gate-dependent Hofstadter butterfly spectrum to compare with R xx and G H obtained from the transport calculation.Note that the spectrum in Figure 5(g) contains energy levels that appear across the gaps, which are an artifact of the method due to the finite width of the ribbon.As mentioned above, they appear since at some value of B the edge states are pushed to k x = 0.

Figure 1
Figure 1 Zero magnetic field characterization of the superlattice.(a) Sketch of the gated superlattice system.The backgate (dark gray) lays below the patterned substrate (light gray), which induces modulated potential in graphene sandwiched between two hBN layers (blue).The global top gate is shown in yellow.(b) The backgate capacitance profile with the periodicity λ = 35 nm in the units of 10 12 cm −2 V −1 .(c) Longitudinal resistance Rxx as a function of the backgate and top gate voltages in a four-terminal device shown in the right inset.Left inset: zoom of Rxx in the boxed region.(d) Rxx line cuts as marked with the respective line colors in (c).

Figure 2
Figure 2 The TMF schematics.(a) Sketch of the geometry considered for magnetic focusing.(b)-(c) Fermi contours in the reciprocal space for a circular (b) and noncircular (c) orbit, and the corresponding focused trajectories in real space.
(b)-Figure2(c).Carriers encircle closed orbits of electron type or hole type in the counterclockwise or clockwise direction, respectively, as determined via the group velocity, v = ∇ k ϵ(k)/ℏ, and the equation ℏ k = −eBv × ẑ.In pristine graphene at low energy, cyclotron orbits exhibit a circular shape [Figure2(b)], but the bands formed in systems with SL modulation can be highly distorted from the original, conical shape, and thus, noncircular Fermi contours can be observed [e.g.Figure2(c)].

Figure 3
Figure 3 Magnetic focusing in the superlattice.(a) Nonlocal resistance R f as a function of magnetic field and top gate voltage at V bg = 40 V.(b) Rxx calculated for the same system at B = 0. (c)-(f) Top row: the miniband structures calculated at Vtg marked by the arrows in (a).In (f) C 1 , C 2 , and V 1 -V 4 label selected minibands.Bottom row: the corresponding Fermi contours at E = 0. (g)-(k) The zoomed map in (a) at the rectangles marked with the corresponding colors.The dashed lines mark the focusing condition ℓ/2n = ℏk/eB, n = 1, 2, . . . .

Figure 4
Figure 4 Non-circular motion of Dirac fermions.Line cuts at Vtg and B ranges as marked in Figure 3(g), Figure 3(h), and Figure 3(j) by the respective colors at (a) Vtg = −2.75V, (b) Vtg = −3.3V, and (c) Vtg = −3.85V. (d) -(i) are the current density maps in arbitrary units at B marked in (a) -(c) by the respective symbols.The red dashed lines show the expected semi-classical trajectory.

Figure 5
Figure 5 High-field transport.(a) The 5-terminal system used for the calculation.(b) and (c) are G H = R −1 xy as a function of average carrier density n and magnetic field B with V bg fixed at 0 and 40 V, respectively, where horizontal lines at B = 2 T and B = 0.2 T mark the line cuts shown in (d) for G H and (e) for Rxy.(f) Longitudinal resistance Rxx as a function of n and B. (g) Calculated Hofstadter spectrum corresponding to (c) and (f).

Figure 5 (
b) and Figure 5(c) show the Hall conductance G H = R −1

Figure 6
Figure 6 Revisiting the TMF experiment on graphene.(a) Computed four-point resistance R 61,54 as a function of magnetic field B and carrier density n revisiting the TMF experiment on graphene [37], and two exemplary transmission functions required for R 61,54 : (b) T 23 and (c) T 41 .The inset in (a) shows the orientation of the considered Hall bar and the configuration of the leads for ground, injector, and voltage probes.
(a)  show n 1 , n 2 , n 3 , n 4 , matching very well with the patterns of the simulated R 61,54 , which requires totally 6 × 5 = 30 transmission functions for such a 6-terminal device, as explained above.Figures6(b) and (c) show two exemplary maps of transmission functions, which can look generally very different from the resulting four-point resistance.