Local, global, and nonlinear screening in twisted double-layer graphene

Abstract

One-atom-thick crystalline layers and their vertical heterostructures carry the promise of designer electronic materials that are unattainable by standard growth techniques. To realize their potential it is necessary to isolate them from environmental disturbances, in particular those introduced by the substrate. However, finding and characterizing suitable substrates, and minimizing the random potential fluctuations they introduce, has been a persistent challenge in this emerging field. Here we show that Landau-level (LL) spectroscopy offers the unique capability to quantify both the reduction of the quasiparticles' lifetime and the long-range inhomogeneity due to random potential fluctuations. Harnessing this technique together with direct scanning tunneling microscopy and numerical simulations we demonstrate that the insertion of a graphene buffer layer with a large twist angle is a very effective method to shield a 2D system from substrate interference that has the additional desirable property of preserving the electronic structure of the system under study. We further show that owing to its remarkable nonlinear screening capability a single graphene buffer layer provides better shielding than either increasing the distance to the substrate or doubling the carrier density and reduces the amplitude of the potential fluctuations in graphene to values even lower than the ones in AB-stacked bilayer graphene.he recent realization of one-atom-thick layers and the fabrication of layered Van der Waals heterostructures revealed fascinating physical phenomena and novel devices based on interlayer interactions (1-10). Inherent to the 2D structure of these layers is an extreme vulnerability to disturbances introduced by the substrate (11)(12)(13)(14). Substrate interference can be eliminated by suspending the sample, an approach that led to the observation of ballistic transport (15, 16) and the fractional quantum Hall effect in graphene (17)(18)(19)(20), but this method only works for small (micrometer-sized) samples at relatively low doping. Another approach is to use atomically smooth metallic substrates (21-23) or graphite (24-28), which screen the random potential. However, these substrates short-circuit the 2D channel and prevent tuning the carrier density by gating, rendering them unsuitable for device applications. Among insulating substrates atomically flat hBN (29-31) and MoS 2 (6) have recently emerged as promising alternatives to SiO 2 substrates.Here we show that by inserting a graphene buffer layer between the 2D sample (in this case, also graphene) and the insulating substrate, the random potential fluctuations are screened without compromising the electronic structure of the 2D system under study and the gating capability. This capability relies on the fact that in van der Waals structures the stacking configuration in the third direction can be set arbitrarily and is not fixed by the chemistry of the elements forming the heterostructure. Consequently it is possible to electronically decouple two 2D cry...

Layer‐by‐Layer Decoupling of Twisted Graphene Sheets Epitaxially Grown on a Metal Substrate

“…By fitting the field‐dependent positions of the single LLs we obtain v F = (1.12 ± 0.07) × 10 6 m s −1 and the position the Dirac point with respect to E F , E D = +(127 ± 2) meV. The value for the Fermi velocity is in very good agreement with the values found for the completely decoupled single‐layer graphene on HOPG, SiC, and SiO 2 . The LL peaks measured on the trilayer are much sharper compared to the bilayer and can be fitted with a sum of Lorentzian peaks (Figure b) which suggests that the linewidth reflects the intrinsic quasiparticle lifetime rather than impurity broadening .…”

Electrostatic imaging of encapsulated graphene

“…We find that the contrast has inverted, indicating a change in sign of the charges in graphene (and also the gold backgate). Again, comparing to the potential measured at the surface of the bottom layer of hBN, we determine a charge density = −2.08 × 10 10 contrary to the expectation that thermal annealing would reduce the amount of contamination in our encapsulated samples. Instead, we find that the sign of dominant dopants reverses, the carrier concentration increases, and the charge fluctuations within the device reduce some.…”

High thermoelectricpower factor in graphene/hBN devices

“…In the high-temperature limit, this region is dominated by thermal excitations, whereas at low temperatures it is controlled by the energy scale of the RPF. Currently, most measurements of the EHP are carried out by scanning probe microscopy, which are typically performed at low temperatures and over a scanning range much smaller than the size of transport devices (28,29,39). Although the boundary of the EHP region can be estimated from the gate dependence of the resistivity (27), V p provides a more direct measure of the EHP region.…”