# Fileset

[Jiang_2023_2D_Mater._10_045027.pdf](https://mdr.nims.go.jp/filesets/da644217-6970-4663-b089-786bc2a71388/download)

## Creator

Zhihao Jiang, Kimberly Hsieh, Alfred J H Jones, Paulina Majchrzak, Chakradhar Sahoo, [Kenji Watanabe](https://orcid.org/0000-0003-3701-8119), [Takashi Taniguchi](https://orcid.org/0000-0002-1467-3105), Jill A Miwa, Yong P Chen, Søren Ulstrup

## Rights

[Creative Commons BY Attribution 4.0 International](https://creativecommons.org/licenses/by/4.0/)

## Other metadata

[Revealing flat bands and hybridization gaps in a twisted bilayer graphene device with microARPES](https://mdr.nims.go.jp/datasets/7b11a652-68bb-4240-ad81-4fbfd055fe7f)

## Fulltext

Revealing flat bands and hybridization gaps in a twisted bilayer graphene device with microARPES2D MaterialsPAPER • OPEN ACCESSRevealing flat bands and hybridization gaps in atwisted bilayer graphene device with microARPESTo cite this article: Zhihao Jiang et al 2023 2D Mater. 10 045027 View the article online for updates and enhancements.You may also likeMoiré engineering of spin–orbit coupling intwisted platinum diselenideLennart Klebl, Qiaoling Xu, AmmonFischer et al.-Electronic properties of twisted multilayergrapheneV Hung Nguyen, Trinh X Hoang and J-CCharlier-A primer on twistronics: a massless Diracfermion’s journey to moiré patterns and flatbands in twisted bilayer grapheneDeepanshu Aggarwal, Rohit Narula andSankalpa Ghosh-This content was downloaded from IP address 144.213.253.16 on 22/10/2023 at 04:10https://doi.org/10.1088/2053-1583/acf775/article/10.1088/2516-1075/ac49f5/article/10.1088/2516-1075/ac49f5/article/10.1088/2515-7639/ac6c4a/article/10.1088/2515-7639/ac6c4a/article/10.1088/1361-648X/acb984/article/10.1088/1361-648X/acb984/article/10.1088/1361-648X/acb9842D Mater. 10 (2023) 045027 https://doi.org/10.1088/2053-1583/acf775OPEN ACCESSRECEIVED29 June 2023REVISED29 August 2023ACCEPTED FOR PUBLICATION6 September 2023PUBLISHED18 September 2023Original Content fromthis work may be usedunder the terms of theCreative CommonsAttribution 4.0 licence.Any further distributionof this work mustmaintain attribution tothe author(s) and the titleof the work, journalcitation and DOI.PAPERRevealing flat bands and hybridization gaps in a twisted bilayergraphene device with microARPESZhihao Jiang1,5, Kimberly Hsieh1,5, Alfred J H Jones1, Paulina Majchrzak1, Chakradhar Sahoo1,Kenji Watanabe2, Takashi Taniguchi3, Jill A Miwa1, Yong P Chen1,4 and Søren Ulstrup1,∗1 Department of Physics and Astronomy, Interdisciplinary Nanoscience Center, Aarhus University, 8000 Aarhus C, Denmark2 Research Center for Electronic and Optical Materials, National Institute for Materials Science, 1-1 Namiki, Tsukuba 305-0044, Japan3 Research Center for Materials Nanoarchitectonics, National Institute for Materials Science, 1-1 Namiki, Tsukuba 305-0044, Japan4 Department of Physics and Astronomy and School of Electrical and Computer Engineering and Purdue Quantum Science andEngineering Institute, Purdue University, West Lafayette, IN 47907, United States of America5 These authors contributed equally.∗ Author to whom any correspondence should be addressed.E-mail: ulstrup@phys.au.dkKeywords:moiré superlattice, magic angle bilayer graphene, electronic structure, microARPES, in situ gatingAbstractControlling the electronic structure of two-dimensional materials using the combination of twistangle and electrostatic doping is an effective means to induce emergent phenomena. In bilayergraphene with an interlayer twist angle near the magic angle, the electronic dispersion is stronglymodified by a manifold of hybridizing moiré Dirac cones leading to flat band segments with strongelectronic correlations. Numerous technical challenges arising from spatial inhomogeneity ofinterlayer interactions, twist angle and device functionality have so far limited momentum-resolvedelectronic structure measurements of these systems to static conditions. Here, we present a detailedcharacterization of the electronic structure exhibiting miniband dispersions for twisted bilayergraphene, near the magic angle, integrated in a functional device architecture using micro-focusedangle-resolved photoemission spectroscopy. The optimum conditions for visualizing the minibanddispersion are determined by exploiting the spatial resolution and photon energy tunability of thelight source and applied to extract a hybridization gap size of (0.14± 0.03) eV and flat bandsegments extending across a moiré mini Brillouin zone. In situ electrostatic gating of the sampleenables significant electron-doping, causing the conduction band states to shift below the Fermienergy. Our work emphasizes key challenges in probing the electronic structure of magic anglebilayer graphene devices and outlines conditions for exploring the doping-dependent evolution ofthe dispersion that underpins the ability to control many-body interactions in the material.1. IntroductionHeterostructures composed of two graphene layersstacked with a minute interlayer twist angle exhibitlong-range moiré superlattices with strong hybridiz-ation effects between the linear Dirac bands of theconstituent layers, giving rise to van Hove singular-ities near the Fermi energy, EF [1–5]. Around themagic angle twist of 1.1◦, a flat band develops as aresult of hybridization between the Dirac bands at EF,leading to the emergence of several strongly correl-ated electronic phases [6–13]. The similarity of theassociated temperature- and doping-dependent elec-tronic phase diagram to that of the high-temperaturecuprate superconductors has inspired the notion ofapplying twisted two-dimensional (2D) materials assolid state quantum simulators with the ability totailor superlattices, band structures and many-bodyinteractions [14, 15]. Angle-resolved photoemissionspectroscopy (ARPES) has been an essential tech-nique for uncovering the energy- and momentum-dependent electronic excitations that underpin thephase diagram of high-temperature superconductors[16, 17]. Employing this methodology to access thequasiparticle spectra of emerging correlated phases intwisted 2D materials is an important step to guidethe design of interactions in superlattices. Indeed,the approach has led to the direct visualization of a© 2023 The Author(s). Published by IOP Publishing Ltdhttps://doi.org/10.1088/2053-1583/acf775https://crossmark.crossref.org/dialog/?doi=10.1088/2053-1583/acf775&domain=pdf&date_stamp=2023-9-18https://creativecommons.org/licenses/by/4.0/https://creativecommons.org/licenses/by/4.0/https://orcid.org/0000-0001-7229-9636https://orcid.org/0000-0003-3701-8119https://orcid.org/0000-0001-5922-4488mailto:ulstrup@phys.au.dk2D Mater. 10 (2023) 045027 Z Jiang et alflat band in twisted bilayer graphene (tBLG) near themagic angle [18–20]. However, in order to gain accessto energy- and momentum-dependent quasiparticleinteractions that are relevant in transport experi-ments it is necessary to probe the electronic struc-ture beyond the static conditions by applying voltagesin situ while performing the ARPES experiment.Such experiments on tBLG are extremely challen-ging for the following reasons: firstly, samples are typ-ically produced using the dry-transfer method [21,22], which is prone to hydrocarbon contaminationfrom the fabrication process that ultimately lead tobroadening of the ARPES spectra. Hence, anneal-ing of the assembled stack at elevated temperaturesis necessary to remove any adsorbates. These pro-cessing steps naturally lead to inhomogeneities intwist angle, strain and interlayer interactions thatsuppress the salient correlation effects [23, 24]. It istherefore imperative to employ ARPES with nano-or micrometer spatial resolution (nanoARPES andmicroARPES). Secondly, in situ control of charge car-rier concentration ismost elegantly achieved by integ-rating the twisted 2D material in a functional devicearchitecture that permits electrostatic gating. Suchin operando micro- and nanoARPES experimentshave recently been demonstrated, but they remainhighly challenging as the additional electrostatic fieldmay impose detrimental energy- and momentum-broadening [25–33]. Moreover, the induced photo-current from the probing synchrotron light appearsto strongly affect the device components, includingthe hexagonal boron nitride (hBN) dielectric that iscustomarily used, causing gate leakage currents anddegradation of device performance after prolongedexposure [34]. So far, back gated ARPES measure-ments have been reported for single-layer graphene[26, 27, 30], Bernal-stacked bilayer graphene [25],tBLG with a twist angle of 12.2◦ [28], tBLG with atwist angle around 3◦ [33] and twisted monolayer-on-bilayer graphene with a twist angle of 3.4◦ [32],but there are no such gated ARPES measurementsreported for magic angle tBLG.Here, we present a detailed microARPES studyof a tBLG device with a twist angle around themagic angle performed at the SGM4 beamline of theASTRID2 light source at Aarhus University, Denmark[35, 36]. We characterize the spatially-dependentelectronic structure of the tBLG device and examinespectral intensity variations stemming from photonenergy-dependent photoelectron interference effects.The optimal combinations of sample spatial posi-tion and photon energy are identified and applied toextract hybridization gaps and determine the (E, k)-dependence of flat band segments in near-magicangle tBLG. Finally, we present measurements dur-ing in situ gating and discuss the impact of elec-trostatic doping on the electronic structure of near-magic angle tBLG.2. Results and discussionOur device is composed of tBLG with a nominal twistangle of 1.2◦ targeted during the fabrication. Thestack is supported on hBN, with a thickness of 30 nm,that makes a twist angle of (2.2± 0.4)◦ with respectto the top graphene layer of the tBLG, as determ-ined from the orientation of the Brillouin zone (BZ)in the ARPES measurements [37]. A graphite backgate below the hBN partially overlaps the area of thetBLG. This configuration is chosen to allow for elec-trostatic gating of the tBLG. The device is placedon 285 nm SiO2/p-doped Si. A three-terminal con-figuration is used with Au/Cr source and drain con-tacts to the bottom layer of the tBLG stack and acontact to the graphite flake. An optical micrographof the device is shown in figure 1(a), and a schem-atic of the architecture and photoemission process ispresented in figure 1(b). Photons with tunable energyhν are focused to a spot with minimum lateral dia-meter of 4.4µm in the present experiment by usingan achromatic capillary mirror [36, 38, 39]. The kin-etic energy and angular distributions of photoemittedelectrons are measured using a SPECS Phoibos 150SAL hemispherical analyzer. In order to verify that theelectrodes are functional, a sweep of the source-drainresistance, RSD, versus gate voltage, VG, is performed,as shown in figure 1(c). The resistance reaches a max-imum of 7.8 kΩ at the gate voltage where the carrierdensity has a minimum. The shape of the RSD(VG)-curve is consistent with earlier transport measure-ments of single-, bi- andmulti-layer graphene devices[40–42], indicating that our device is operational.Note that the resistance sweep was done after expos-ure to the synchrotron beam,whichmay influence theresistance peak position and width [27].Spatially-resolved measurements of the (E, k)-dependent photoemission intensity over the areashown in the optical micrograph reveal that the tBLGpart exhibits three distinct types of dispersions thatare found in the regions delineated by dashed linesin figure 1(a). Example ARPES spectra from thesethree regions are presented in figures 1(d) and (e) andcolor-coded via circles that refer to the spots markedin the optical image. The spectra represent cuts per-pendicular and parallel to theΓ−KT direction,whereKT is the Dirac point of the top graphene layer.Corresponding (kx,ky)-dependent constant energycuts at −0.8 eV are shown in figure 1(f). This energyis chosen as it is simpler to distinguish the disper-sion emerging from two cones than closer to EF. Theintensity variation in these contours mainly derivesfrom sublattice interference effects between photo-electrons emitted from the graphene layers. Thiscauses a k-dependent modulation of the photoemis-sion matrix elements known as the dark corridorin graphene [43], which is characterized by a highintensity towards the center of the first BZ and a22D Mater. 10 (2023) 045027 Z Jiang et alFigure 1. Spatially-dependent electronic structure of tBLG device: (a) optical micrograph of device with top (bottom) graphenelayer demarcated by blue (red) outlines. Dashed lines indicate regions of tBLG with different ARPES dispersions. Source (S), drain(D) and gate (G) electrodes are indicated. The contacts visible in the left side of the image are disconnected. (b) Schematic ofmicro-focused photoemission process and device architecture. (c) Source-drain resistance as a function of gate voltage measuredin situ after exposure to synchrotron light. (d), (e) ARPES spectra measured (d) perpendicular and (e) parallel to the Γ−KTdirection, as indicated by the dashed lines in the blue BZ diagrams corresponding to the top graphene layer. The spectra areobtained from the areas demarcated by correspondingly colored circles in (a). The blue and red dashed lines are outlines of topand bottom graphene Dirac cones, respectively. The red arrows indicate the flat band expected around the magic angle. The moirélattice vector length, Gm, is demarcated by a bracket for a twist angle of 1.2◦. (f) Constant energy cuts extracted at−0.8 eV fromthe same areas. The dashed lines separated by double-headed blue arrows indicate rigid k-shifts of the energy contours. The greenarrows demarcate dim lobes from interacting graphene layers. The spectra were acquired at a photon energy hν of 47 eV.suppression in the direction towards the neighbour-ing BZs. Notably, this causes a strong suppression ofintensity from half of the dispersion in the cuts alongΓ−KT as seen in figure 1(e).Amore detailed inspection of the spectra from thethree regions reveals several key differences betweenthem. The region marked by a purple circle displaystwo sets of Dirac cones that cross without any signof interaction between their bands. The cone out-lined by blue dashed lines is approximately twiceas intense as the cone outlined by red dashed lines,which enables us to assign the cone outlined by bluedashed lines to the top graphene layer [28]. Analysisof the constant energy contour reveals a momentumshift∆k of (0.06± 0.01)Å−1 between the two cones,which is demarcated by a blue double-headed arrowin figure 1(f). The twist angle θ is then calculatedto be (2.0± 0.2)◦ using sin(θ/2) = ∆k/2|K|, where|K|= 1.7Å−1 is the distance fromΓ to K of graphene.The region marked by a green circle also displaystwo non-interacting Dirac cones that are shifted evenless in momentum with respect to each other. Thisis most clearly seen via the two closely spaced linearbranches in figure 1(e) and slightly shifted contoursin figure 1(f). The shift is determined to be ∆k=(0.04± 0.01)Å−1, which corresponds to a twist angleof (1.4± 0.2)◦. In the region marked by an orangecircle an immediately striking feature resembling aflat band is noticeable, as seen via the red arrows infigures 1(d) and (e). The constant energy contourexhibits additional segments of intensity, demarc-ated by green arrows, that cannot be described as32D Mater. 10 (2023) 045027 Z Jiang et alFigure 2. Photon energy dependence of tBLG ARPES intensity around the magic angle: (a) constant energy cuts at−0.4 eV at thegiven values of hν. (b), (c) ARPES spectra extracted along (b) horizontal and (c) vertical dotted lines in (a). The dashed lines in(b) and (c) indicate the energy where the cuts in (a) are obtained. The photon energy is the same as stated in (a) along the panelrows. The location of KB- and KT-points are marked by red and blue dots, respectively. Green and orange arrows link side lobesand arcs in the constant energy cuts to dispersing branches in the E(k)-cuts. Black arrows indicate minibands along the darkcorridor. The red arrow demarcates a hybridization gap around−0.2 eV.just two shifted single-layer graphene contours, asin the other regions. These spatially-resolved ARPESspectra establish the region of the device where thetwo graphene layers interact. The extent of the flatband segment is comparable to the moiré lattice vec-tor around the magic angle twist, which is givenby Gm = 0.06Å−1, as indicated by the brackets infigures 1(d) and (e). The features are thus consist-ent with those expected from tBLG near the magicangle [18–20]. The non-interacting parts likely con-tain graphene layers that are not sufficiently mech-anically pressed together during stacking. It is alsopossible that residual strain, impurities between lay-ers and pits in the underlying hBN spoil the contactbetween the flakes. In the following, we focus solelyon the dispersion from the interacting region of thedevice.A moiré superlattice arises from the twisting ofthe graphene layers, which appears as minibands inthe electronic structure measurements. The interac-tions between the minibands give rise to hybridiza-tion gaps that are appreciable in the photoemissionintensity. To further substantiate the previous ana-lysis, we proceed to search for the optimum photonenergy where the hybridization effects are most vis-ible. Such a dependence on photon energy is expecteddue to interference effects involving photoelectronsfrom the moiré sites, in addition to the two sublatticesites in each graphene layer [19, 32]. By performing ascan of hν from 40 to 76 eV we are able to determinesubstantial redistributions of intensity between themanifold of bands in tBLG near the magic angle, asshown in figure 2. Representative spectra have beenselected at photon energies of 40, 50, 60 and 76 eV, as42D Mater. 10 (2023) 045027 Z Jiang et alFigure 3. Tracking hybridization effects across mini Brillouin zones: (a) constant energy cuts extracted at the given energies. ThemBZs corresponding to a moiré lattice vector of 0.06 Å−1 have been overlaid. Red and blue dots indicate KB and KT, respectively.(b) ARPES spectra extracted along mBZ high symmetry directions as indicated by dashed lines in the corresponding mBZdiagrams. (c) Energy distribution curves (markers) integrated over a k-range of 0.05 Å−1. The integration region is indicated by abracket in the corresponding ARPES spectrum in the same column in (b). Peak-to-peak energy separations extracted from fits(black curves) are indicated by vertical bars separated by double-headed arrows and values are stated in units of eV. The error baron the extracted energies is±0.03 eV. The fitted peak positions are shown via ticks in (b). The ARPES spectra were obtained athν = 40 eV.these highlight the most dramatic intensity changeswithin the probed range.Figure 2(a) presents constant energy contours at−0.4 eV around the KB- and KT-points, correspond-ing to the main BZ K-points of bottom and topgraphene layers, respectively. In this energy cut, thefeatures resulting from the superlattice are clearly dis-tinguishable, whereas at lower energies the hybridiz-ation effects and flat band lead to a larger degree ofcomplexity. At a photon energy of 40 eV, two arcs ofnearly equal intensity occur (see orange arrows) alongthe vertical dotted line through KT, in addition to aset of dim lobes (see green arrows) on the sides ofKB and KT along the horizontal dotted line. From 50to 76 eV, the intensity switches from the outer arc tothe inner arc centred on the KB–KT line, while theside lobes remain relatively dim. E(k)-spectra result-ing from horizontal and vertical cuts along the dot-ted lines are displayed in figures 2(b) and (c). Thetwo side lobes are formed by two linearly dispers-ing bands (see green arrows in figure 2(b)) that dis-perse from lower energy and hybridize to form a flatband around EF. Cuts along the vertical directionsimilarly exhibit two linearly dispersing bands (seeorange arrows figure 2(c)) that form the arcs seenin the constant energy contours. The linear parts areinterrupted by the formation of hybridization gapsaround −0.2 eV (see red arrow in figure 2(c)). Therelative intensity levels of these branches vary signi-ficantly. At 50 eV, the intensity of the outer branchis high while it is almost completely suppressed forthe inner branch. At 76 eV, this behavior is reversed.The important consequence of these intensity vari-ations is that faint minibands that form an extendedflat band at EF are resolvable in the dark corridor sideof the vertical cut at 50 and 60 eV (see black arrowsin figure 2(c)). Additionally, hybridization effects aremost pronounced at 40 eV where the relative intens-ity levels are more similar between the branches, asexemplified by the gap highlighted by a red arrow infigure 2(c). We therefore choose a photon energy of40 eV for more in-depth analysis of the hybridizationeffects and the extent of the flat band dispersion, asdiscussed in the following.The evolution of constant energy contoursfrom EF to −0.4 eV with overlaid mini Brillouinzones (mBZs) corresponding to a near-magic anglemoiré lattice vector Gm of 0.06 Å−1 is presentedin figure 3(a). Around EF, the intensity is primar-ily concentrated in the four mBZs surrounding KBand KT, consistent with previous reports [18, 19].Horizontal high symmetry E(k)-cuts along thesemBZs are shown in figure 3(b). We observe flatbands that are separated by gaps at the energiesnoted by tick marks. The features are analyzed infigure 3(c) using energy distribution curves (EDCs)52D Mater. 10 (2023) 045027 Z Jiang et alFigure 4. Characterization of flat band dispersion: (a)ARPES intensity (left panel) and corresponding secondderivative map of the intensity (right panel) measuredalong the cut marked by a dashed line in the mBZ diagram.(b), (c) Similar plots for spectra obtained along thedirections indicated by dashed lines in the mBZ diagrams.The estimated extent of the flat band in k along theextracted cuts is indicated by red double-headed arrows onthe second derivative maps. The size of the moiré latticevector Gm is provided via a bracket for a twist angle of 1.2◦.The measurements were performed at hν = 40 eV.integrated over a k-range of 0.05 Å−1 centred onthe flat band segments, as shown via brackets infigure 3(b). The dispersive bands give rise to broadpeaks in the EDCs from −0.4 eV and towards lowerenergies. Step-like features marked by vertical barsbetween EF and−0.4 eV correspond to flat band seg-ments separated by hybridization gaps. The EDCsare modelled by Lorentzian peaks multiplied by aFermi–Dirac cut-off. Fitting to this model revealsa peak pinned at (−0.04± 0.03) eV in all of themBZ cuts, consistent with the flat band expectedfor tBLG near the magic angle [18, 19]. The averagelinewidth of the flat band determined from the fullwidth at half maximum (FWHM) of the Lorentzianfits is (0.07± 0.02) eV. This energy broadening islikely attributed to sub-micrometer inhomogen-eities of the twist angle arising from strain fluctu-ations and stacking faults [23, 24]. The average valueof the associated hybridization gaps is determinedto be (0.14± 0.03) eV, which reflects the interlayerinteraction strength and moiré potential establishedbetween the twisted graphene layers. The observedseparation of the flat state from dispersive features bya hybridization gap is consistent with the theoreticalresults corroborated by experimental data in [18, 19].The previously calculated values of the gap around40meV [18] and 110meV [19] appear to underestim-ate the interaction strength we observe. Indeed, themagnitude of the gap depends sensitively on latticerelaxation effects and is therefore difficult to model[18].The intensity of minibands rapidly drops towardshigher order mBZs due to the incommensurabilityof a general twist angle. In other words, the bandstructure of the superlattice is not a well-definedquantity such that the flat band dispersion acquiresa finite momentum range in the measured ARPESspectral weight. This range is visualized and extrac-ted using the analysis presented in figure 4. Threecuts are extracted along directions marked in the cor-responding mBZ diagrams. The second derivative ofthe ARPES intensity corresponding to these cuts isthen determined [44]. The resulting second deriv-ative images seen in figure 4 emphasize flat bandsegments that are separated from the dispersive fea-tures by hybridization gaps. The average extent of theflat band in momentum obtained from these cuts is(0.07± 0.03)Å−1, which is in line with the moiré lat-tice vector of Gm = 0.06Å−1, indicating coherenceacross roughly one moiré unit cell.Finally, we utilize the above analysis in order toinvestigate the impact on the ARPES dispersion ofapplying an electrostatic gate voltage of 14V whilekeeping the tBLG grounded. This is the maximumvoltage we could achieve while avoiding a large gateleakage current. ARPES spectra obtained along KT −KB are compared at 0 and 14V in figure 5(a). Themain effect of the applied electric field is a substan-tial momentum broadening, which precluded moredetailed measurements at intermediate gate voltagesas the gradual changes in the spectra were not possibleto discern. Consequently, we are limited to estim-ating the doping-induced energy shift of the tBLGbands, which is done via the second derivative intens-ity in figure 5(b). The broadened manifold of valencebands shifts down by (0.22± 0.05) eV, revealing anapparent gap of (0.15± 0.05) eV and a distributionof states around EF that likely corresponds to the con-duction band of tBLG. Such a gap may develop dueto the Coulomb potential energy difference betweenthe two graphene layers induced by the perpendicu-lar electric field, as seen in potassium-doped Bernal-stacked bilayer graphene [45]. Note that our data doesnot enable us to preclude the presence of faint in-gapstates. Furthermore, the electric field leads to differ-ent charge carrier concentrations in the top and bot-tom graphene layers (see [28, 32, 33] and section 4),which results in an increasingly asymmetric overlapof the top and bottom Dirac cones with voltage andthereby substantially affect the hybridization energiesand possible flat band dispersion [1]. These detailsare difficult to resolve due to the broadening of ourspectra. This is caused by the imperfect overlap ofgraphite back gate and tBLG, which is evident from62D Mater. 10 (2023) 045027 Z Jiang et alFigure 5. Effect of in situ electrostatic gating: (a) ARPESspectra extracted along the KT −KB direction at 0 V (leftpanel) and 14V (right panel). (b) Corresponding plots ofthe second derivative of the intensity. Red lines indicate ashift of the valence band manifold. Dashed red linesdemarcate an apparent gap in the dispersion measured at14V. The corresponding energies are marked bydouble-headed red arrows and stated in units of eV. Theerror bars are±0.05 eV. The spectra were measured athν = 40 eV.the optical micrograph in figure 1(a). The result-ing inhomogeneous electric field between back gateand the tBLG region marked by an orange circle infigure 1(a) causes broadening of the photoelectronmomentum distributions, spoiling the linewidths ofthe acquired spectra. Nevertheless, our data impliesa significant modification of the dispersion over theachievable range of gating. Detailed (E, k)-dependentmeasurements and analysis of such effects as a func-tion of gate voltage are critical to access and tailorthe electronic structure of tBLG at small twist angles,requiring further optimization of device designs formicro- and nanoARPES experiments [46].3. ConclusionsWe have applied microARPES to map the energy-and momentum-resolved dispersion of a tBLGdevice, allowing us to spatially disentangle regionsof the device where the graphene layers are inter-acting. Variations in photoemission intensity areobserved between minibands, which arise fromphotoelectron interference effects, enabling us toselect an optimum photon energy to visualizethe dispersion and analyse hybridization effects.The dispersion in the interacting regions of thedevice is thereby found to be consistent with near-magic angle tBLG, exhibiting hybridization gaps of(0.14± 0.03) eV and flat band segments extend-ing across a moiré mBZ. Electrostatic doping viaan applied gate voltage causes a downwards shiftof the manifold of interacting minibands, whilea narrow distribution of states appears aroundEF which is separated from the manifold by(0.15± 0.05) eV. Our detailed microARPES analysisof a tBLG device highlights key challenges that areimportant to address in order to access energy- andmomentum-resolved quasiparticle dynamics in thesalient correlated phases of twisted 2D materials.4. Methods4.1. Device fabricationHigh-quality flakes of single-layer graphene, hBN andgraphite were exfoliated onto SiO2/Si substrates andidentified using an optical microscope. The flakeswere assembled into a heterostructure using the dry-transfer technique [21, 22]. We used a polycarbonate(PC)/polydimethylsiloxane (PDMS) stamp on a glassslide to pick up all flakes. Single-layer graphene wascut in two pieces using a tungsten micro-needle tipwith a diameter of 0.001mm. One of the pieces waspicked up at 130 ◦C. The other piece was rotated by1.2◦ and then picked up using the other piece. ThehBN flake was subsequently picked up at 120 ◦C fol-lowed by the graphite flake at 130 ◦C. The entire stackwas then released onto a SiO2/Si substrate with a pre-defined Au/Cr pattern with a thickness of 50/5 nm at180 ◦C. The device was finally annealed in a H2/Arenvironment at 250 ◦C for 4 h and then wire-bondedto a chip carrier.4.2. PhotoemissionmeasurementsThe microARPES measurements were performed atthe SGM4 beamline of the ASTRID2 synchroton atAarhus University, Denmark. The chip carrier withthe device was annealed in the ultrahigh vacuumsystem at a temperature of 200 ◦C for several hoursbefore exposure to the beam. The device was keptat room temperature at a base pressure better than3 · 10−10mbar during data acquisition.The synchrotron beam was focused to a spot sizeof 4.4µm using an elliptical capillary mirror [SigrayInc.]. The focus and position of the spot was stableover the probed photon energy range from 40 to76 eV. The ARPES spectra were obtained using aSPECS Phoibos 150 SAL analyser. The energy- andangular resolution were better than 20 meV and 0.1◦,respectively. The scans of (E,kx,ky)-dependent pho-toemission intensity were acquired with the scan-ning angle lens mode of the analyser while keepingthe sample position fixed. Photon energy depend-ent scans were done by collecting (E,kx,ky)-spectraat each photon energy in order to track the posi-tions of the Dirac points. The sample was alignedsuch that the analyzer cut was along the directionof momentum space that is perpendicular to theΓ−KT high-symmetry direction of the top graphenelayer.Electrostatic gating and source-drain resistancemeasurements were achieved using two Keithley 245072D Mater. 10 (2023) 045027 Z Jiang et alsourcemeters. It is not possible to directly determinethe induced carrier concentration n in the two layersfrom the data, however, one may obtain a qualitativeestimate using the n(VG) dependence measured forthe tBLG device with a twist angle of 12.2◦ investig-ated in [28]. The larger twist angle prevents any inter-action between the linearly dispersing Dirac bandsclose to EF, enabling a direct extraction of n= kF 2/π,where kF is the Fermi wavevector obtained from theARPES spectra. For tBLG near the magic angle thismethod does not work, as the dispersion is stronglymodified via hybridization and flat band formation.Based on the 12.2◦ tBLGdevice, the doping in the bot-tom layer at 14V would amount to 7 · 1012 cm−2 andin the top layer it would be 3 · 1012 cm−2 [28].Data availability statementThe data used in this study is available on the Zenodoplatform at https://doi.org/10.5281/zenodo.8087181.AcknowledgmentsThe authors acknowledge funding from the DanishCouncil for Independent Research, Natural Sciencesunder the Sapere Aude program (Grant Nos. DFF-9064-00057B and DFF-6108-00409), the AarhusUniversity Research Foundation, the Novo NordiskFoundation (Project Grant NNF22OC0079960)and from VILLUM FONDEN under the VillumInvestigator Program (GrantNo. 25931). C S acknow-ledges Marie Sklodowska-Curie PostdoctoralFellowship (Proposal Number 101059528). Growthof hexagonal boron nitride was supported by theJSPS KAKENHI (Grant Nos. 20H00354, 21H05233and 23H02052) and World Premier InternationalResearch Center Initiative (WPI), MEXT,Japan.ORCID iDsKimberly Hsieh https://orcid.org/0000-0001-7229-9636Kenji Watanabe https://orcid.org/0000-0003-3701-8119Søren Ulstrup https://orcid.org/0000-0001-5922-4488References[1] Lopes dos Santos J M B, Peres N M R and Castro Neto A H2007 Graphene bilayer with a twist: electronic structure Phys.Rev. Lett. 99 256802[2] Suárez Morell E, Correa J D, Vargas P, Pacheco M andBarticevic Z 2010 Flat bands in slightly twisted bilayergraphene: tight-binding calculations Phys. Rev. B82 121407[3] Li G, Luican A, Lopes dos Santos J M B, Castro Neto A H,Reina A, Kong J and Andrei E Y 2010 Observation of vanHove singularities in twisted graphene layers Nat. Phys.6 109–13[4] Bistritzer R and MacDonald A H 2011 Moiré bands intwisted double-layer graphene Proc. Natl Acad. Sci.108 12233–7[5] Lopes dos Santos J M B, Peres N M R and Castro Neto A H2012 Continuum model of the twisted graphene bilayerPhys. Rev. B 86 155449[6] Cao Y et al 2018 Correlated insulator behaviour at half-fillingin magic-angle graphene superlattices Nature 556 80–84[7] Cao Y, Fatemi V, Fang S, Watanabe K, Taniguchi T, Kaxiras Eand Jarillo-Herrero P 2018 Unconventionalsuperconductivity in magic-angle graphene superlatticesNature 556 43–50[8] Kerelsky A et al 2019 Maximized electron interactions at themagic angle in twisted bilayer graphene Nature 572 95–100[9] Xie Y, Lian B, Jäck B, Liu X, Chiu C-Li, Watanabe K,Taniguchi T, Bernevig B A and Yazdani A 2019 Spectroscopicsignatures of many-body correlations in magic-angle twistedbilayer graphene Nature 572 101–5[10] Jiang Y, Lai X, Watanabe K, Taniguchi T, Haule K, Mao J andAndrei E Y 2019 Charge order and broken rotationalsymmetry in magic-angle twisted bilayer graphene Nature573 91–95[11] Choi Y et al 2019 Electronic correlations in twisted bilayergraphene near the magic angle Nat. Phys. 15 1174–80[12] Yankowitz M, Chen S, Polshyn H, Zhang Y, Watanabe K,Taniguchi T, Graf D, Young A F and Dean C R 2019 Tuningsuperconductivity in twisted bilayer graphene Science363 1059–64[13] Lu X et al 2019 Superconductors, orbital magnets andcorrelated states in magic-angle bilayer graphene Nature574 653–7[14] Andrei E Y and MacDonald A H 2020 Graphene bilayerswith a twist Nat. Mater. 19 1265–75[15] Kennes D M, Claassen M, Xian L, Georges A, Millis A J,Hone J, Dean C R, Basov D N, Pasupathy A N and Rubio A2021 Moiré heterostructures as a condensed-matterquantum simulator Nat. Phys. 17 155–63[16] Damascelli A, Hussain Z and Shen Z-X 2003 Angle-resolvedphotoemission studies of the cuprate superconductors Rev.Mod. Phys. 75 473–541[17] Sobota J A, He Y and Shen Z-X 2021 Angle-resolvedphotoemission studies of quantum materials Rev. Mod. Phys.93 025006[18] Utama M I B et al 2021 Visualization of the flat electronicband in twisted bilayer graphene near the magic angle twistNat. Phys. 17 184–8[19] Lisi S et al 2021 Observation of flat bands in twisted bilayergraphene Nat. Phys. 17 189–93[20] Li Y et al 2022 Observation of coexisting Dirac bands andmoiré flat bands in magic-angle twisted trilayer grapheneAdv. Mater. 34 2205996[21] Kim K et al 2016 van der Waals heterostructures with highaccuracy rotational alignment Nano Lett. 16 1989–95[22] Kim K, DaSilva A, Huang S, Fallahazad B, Larentis S,Taniguchi T, Watanabe K, LeRoy B J, MacDonald A H andTutuc E 2017 Tunable moiré bands and strong correlationsin small-twist-angle bilayer graphene Proc. Natl Acad. Sci.114 3364–9[23] Uri A et al 2020 Mapping the twist-angle disorder andlandau levels in magic-angle graphene Nature581 47–52[24] Lau C N, Bockrath MW, Mak K F and Zhang F 2022Reproducibility in the fabrication and physics of moirématerials Nature 602 41–50[25] Joucken F et al 2019 Visualizing the effect of an electrostaticgate with angle-resolved photoemission spectroscopy NanoLett. 19 2682–7[26] Nguyen P V et al 2019 Visualizing electrostatic gating effectsin two-dimensional heterostructures Nature 572 220–38https://doi.org/10.5281/zenodo.8087181https://orcid.org/0000-0001-7229-9636https://orcid.org/0000-0001-7229-9636https://orcid.org/0000-0001-7229-9636https://orcid.org/0000-0003-3701-8119https://orcid.org/0000-0003-3701-8119https://orcid.org/0000-0003-3701-8119https://orcid.org/0000-0001-5922-4488https://orcid.org/0000-0001-5922-4488https://orcid.org/0000-0001-5922-4488https://doi.org/10.1103/PhysRevLett.99.256802https://doi.org/10.1103/PhysRevLett.99.256802https://doi.org/10.1103/PhysRevB.82.121407https://doi.org/10.1103/PhysRevB.82.121407https://doi.org/10.1038/nphys1463https://doi.org/10.1038/nphys1463https://doi.org/10.1073/pnas.1108174108https://doi.org/10.1073/pnas.1108174108https://doi.org/10.1103/PhysRevB.86.155449https://doi.org/10.1103/PhysRevB.86.155449https://doi.org/10.1038/nature26154https://doi.org/10.1038/nature26154https://doi.org/10.1038/nature26160https://doi.org/10.1038/nature26160https://doi.org/10.1038/s41586-019-1431-9https://doi.org/10.1038/s41586-019-1431-9https://doi.org/10.1038/s41586-019-1422-xhttps://doi.org/10.1038/s41586-019-1422-xhttps://doi.org/10.1038/s41586-019-1460-4https://doi.org/10.1038/s41586-019-1460-4https://doi.org/10.1038/s41567-019-0606-5https://doi.org/10.1038/s41567-019-0606-5https://doi.org/10.1126/science.aav1910https://doi.org/10.1126/science.aav1910https://doi.org/10.1038/s41586-019-1695-0https://doi.org/10.1038/s41586-019-1695-0https://doi.org/10.1038/s41563-020-00840-0https://doi.org/10.1038/s41563-020-00840-0https://doi.org/10.1038/s41567-020-01154-3https://doi.org/10.1038/s41567-020-01154-3https://doi.org/10.1103/RevModPhys.75.473https://doi.org/10.1103/RevModPhys.75.473https://doi.org/10.1103/RevModPhys.93.025006https://doi.org/10.1103/RevModPhys.93.025006https://doi.org/10.1038/s41567-020-0974-xhttps://doi.org/10.1038/s41567-020-0974-xhttps://doi.org/10.1038/s41567-020-01041-xhttps://doi.org/10.1038/s41567-020-01041-xhttps://doi.org/10.1002/adma.202205996https://doi.org/10.1002/adma.202205996https://doi.org/10.1021/acs.nanolett.5b05263https://doi.org/10.1021/acs.nanolett.5b05263https://doi.org/10.1073/pnas.1620140114https://doi.org/10.1073/pnas.1620140114https://doi.org/10.1038/s41586-020-2255-3https://doi.org/10.1038/s41586-020-2255-3https://doi.org/10.1038/s41586-021-04173-zhttps://doi.org/10.1038/s41586-021-04173-zhttps://doi.org/10.1021/acs.nanolett.9b00649https://doi.org/10.1021/acs.nanolett.9b00649https://doi.org/10.1038/s41586-019-1402-1https://doi.org/10.1038/s41586-019-1402-12D Mater. 10 (2023) 045027 Z Jiang et al[27] Muzzio R et al 2020 Momentum-resolved view of highlytunable many-body effects in a graphene/hBN field-effectdevice Phys. Rev. B 101 201409[28] Jones A J H et al 2020 Observation of electrically tunable vanHove singularities in twisted bilayer graphene fromNanoARPES Adv. Mater. 32 2001656[29] Hofmann P 2021 Accessing the spectral function of inoperando devices by angle-resolved photoemissionspectroscopy AVS Quantum Sci. 3 021101[30] Dale N et al 2022 Correlation-driven electron-holeasymmetry in graphene field effect devices npj QuantumMater. 7 9[31] Jones A J H et al 2022 Nanoscale view of engineered massiveDirac quasiparticles in lithographic superstructures ACSNano 16 19354–62[32] Nunn J E, McEllistrim A, Weston A, Garcia-Ruiz A,Watson M D, Mucha-Kruczynski M, Cacho C,Gorbachev R V, Fal’ko V I and Wilson N R 2023 ARPESsignatures of few-layer twistronic graphenes Nano Lett.23 5201–8[33] Dale N et al 2023 Layer-dependent interaction effects in theelectronic structure of twisted bilayer graphene devices NanoLett. 23 6799–806[34] Ju L et al 2014 Photoinduced doping in heterostructures ofgraphene and boron nitride Nat. Nanotechnol. 9 348–52[35] Bianchi M et al 2023 Status and strategy at ISA, centre forstorage ring facilities, Aarhus University, Denmark Eur. Phys.J. Plus 138 132[36] Volckaert K et al 2023 Surface electronic structureengineering of manganese bismuth tellurides guided bymicro-focused angle-resolved photoemission Adv. Mater.35 2301907[37] Koch R J, Katoch J, Moser S, Schwarz D, Kawakami R K,Bostwick A, Rotenberg E, Jozwiak C and Ulstrup S 2018Electronic structure of exfoliated and epitaxial hexagonalboron nitride Phys. Rev. Mater. 2 074006[38] Koch R J, Jozwiak C, Bostwick A, Stripe B, Cordier M,Hussain Z, Yun W and Rotenberg E 2018 Nano focusing ofsoft x-rays by a new capillary mirror optic SynchrotronRadiat. News 31 50–52[39] Ulstrup S et al 2020 Direct observation of minibands in atwisted graphene/WS2 bilayer Sci. Adv. 6 eaay6104[40] Morozov S V, Novoselov K S, Katsnelson M I, Schedin F,Elias D C, Jaszczak J A and Geim A K 2008 Giant intrinsiccarrier mobilities in graphene and its bilayer Phys. Rev. Lett.100 016602[41] Yang T-Y et al 2011 Observation of long spin-relaxationtimes in bilayer graphene at room temperature Phys. Rev.Lett. 107 047206[42] Maassen T, Dejene F K, Guimarães M H D, Józsa C and vanWees B J 2011 Comparison between charge andspin transport in few-layer graphene Phys. Rev. B83 115410[43] Gierz I, Henk J, Höchst H, Ast C R and Kern K 2011Illuminating the dark corridor in graphene: polarizationdependence of angle-resolved photoemission spectroscopyon graphene Phys. Rev. B 83 121408[44] Zhang P, Richard P, Qian T, Xu Y-M, Dai X and Ding H 2011A precise method for visualizing dispersive features in imageplots Rev. Sci. Instrum. 82 043712[45] Ohta T, Bostwick A, Seyller T, Horn K and Rotenberg E 2006Controlling the electronic structure of bilayer grapheneScience 313 951–4[46] Jiang Z, K Hsieh, A J H Jones, P Majchrzak, C Sahoo, KWatanabe, T Taniguchi, J A Miwa, Y P Chen and S Ulstrup2023 Revealing flat bands and hybridization gaps in a gatedtwisted bilayer graphene device with microARPES Data setZenodo (https://doi.org/10.5281/zenodo.8087181)9https://doi.org/10.1103/PhysRevB.101.201409https://doi.org/10.1103/PhysRevB.101.201409https://doi.org/10.1002/adma.202001656https://doi.org/10.1002/adma.202001656https://doi.org/10.1116/5.0038637https://doi.org/10.1116/5.0038637https://doi.org/10.1038/s41535-021-00404-8https://doi.org/10.1038/s41535-021-00404-8https://doi.org/10.1021/acsnano.2c08929https://doi.org/10.1021/acsnano.2c08929https://doi.org/10.1021/acs.nanolett.3c01173https://doi.org/10.1021/acs.nanolett.3c01173https://doi.org/10.1021/acs.nanolett.3c00253https://doi.org/10.1021/acs.nanolett.3c00253https://doi.org/10.1038/nnano.2014.60https://doi.org/10.1038/nnano.2014.60https://doi.org/10.1140/epjp/s13360-023-03748-1https://doi.org/10.1140/epjp/s13360-023-03748-1https://doi.org/10.1002/adma.202301907https://doi.org/10.1002/adma.202301907https://doi.org/10.1103/PhysRevMaterials.2.074006https://doi.org/10.1103/PhysRevMaterials.2.074006https://doi.org/10.1080/08940886.2018.1483660https://doi.org/10.1080/08940886.2018.1483660https://doi.org/10.1126/sciadv.aay6104https://doi.org/10.1126/sciadv.aay6104https://doi.org/10.1103/PhysRevLett.100.016602https://doi.org/10.1103/PhysRevLett.100.016602https://doi.org/10.1103/PhysRevLett.107.047206https://doi.org/10.1103/PhysRevLett.107.047206https://doi.org/10.1103/PhysRevB.83.115410https://doi.org/10.1103/PhysRevB.83.115410https://doi.org/10.1103/PhysRevB.83.121408https://doi.org/10.1103/PhysRevB.83.121408https://doi.org/10.1063/1.3585113https://doi.org/10.1063/1.3585113https://doi.org/10.1126/science.1130681https://doi.org/10.1126/science.1130681https://doi.org/10.5281/zenodo.8087181 Revealing flat bands and hybridization gaps in a twisted bilayer graphene device with microARPES 1. Introduction 2. Results and discussion 3. Conclusions 4. Methods 4.1. Device fabrication 4.2. Photoemission measurements References