One-atom thick crystalline layers and their vertical heterostructures carry the promise of designer electronic materials that are unattainable by standard growth techniques. In order to realize their potential it is necessary to isolate them from env
ironmental disturbances in particular those introduced by the substrate. But 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 is exquisitely sensitive to potential fluctuations on both local and global length scales. Harnessing this technique we demonstrate that the insertion of an intermediate graphene layer provides superior screening of substrate induced disturbances, more than doubling the electronic mean free path. Furthermore, we find that the proximity of hBN acts as a nano-scale vacuum cleaner, dramatically suppressing the global potential fluctuations. This makes it possible to fabricate high quality devices on standard SiO2 substrates.
One of the enduring challenges in graphene research and applications is the extreme sensitivity of its charge carriers to external perturbations, especially those introduced by the substrate. The best available substrates to date, graphite and hBN, s
till pose limitations: graphite being metallic does not allow gating, while both hBN and graphite having lattice structures closely matched to that of graphene, may cause significant band structure reconstruction. Here we show that the atomically smooth surface of exfoliated MoS2 provides access to the intrinsic electronic structure of graphene without these drawbacks. Using scanning tunneling microscopy and Landau-level spectroscopy in a device configuration which allows tuning the carrier concentration, we find that graphene on MoS2 is ultra-flat producing long mean free paths, while avoiding band structure reconstruction. Importantly, the screening of the MoS2 substrate can be tuned by changing the position of the Fermi energy with relatively low gate voltages. We show that shifting the Fermi energy from the gap to the edge of the conduction band gives rise to enhanced screening and to a substantial increase in the mean-free-path and quasiparticle lifetime. MoS2 substrates thus provide unique opportunities to access the intrinsic electronic properties of graphene and to study in situ the effects of screening on electron-electron interactions and transport.
The discovery of graphene has put the spotlight on other layered materials including transition metal dichalcogenites (TMD) as building blocks for novel heterostructures assembled from stacked atomic layers. Molybdenum disulfide, MoS2, a semiconducto
r in the TMD family, with its remarkable thermal and chemical stability and high mobility, has emerged as a promising candidate for post-silicon applications such as switching, photonics, and flexible electronics. Since these rely on controlling the position of the Fermi energy (EF), it is crucial to understand its dependence on doping and gating. Here we employed scanning tunneling microscopy (STM) and spectroscopy (STS) with gating capabilities to measure the bandgap and the position of EF in MoS2, and to track its evolution with gate voltage. For bulk samples, the measured bandgap (~1.3eV) is comparable to the value obtained by photoluminescence, and the position of EF (~0.35eV) below the conduction band, is consistent with n-doping reported in this material. Using topography together with spectroscopy we traced the source of the n-doping in bulk MoS2 samples to point defects, which we attribute to S vacancies. In contrast, for thin films deposited on SiO2, we found significantly higher levels of n-doping that cannot be attributed to S vacancies. By combining gated STS with transport measurements in a field effect transistor (FET) configuration, we demonstrate that the higher levels of n-doping in thin film samples is due to charge traps at the sample-substrate interface.
