Landscape of Charge Puddles in Graphene Nanoribbons on Hexagonal Boron Nitride (original) (raw)

2020, physica status solidi (b)

Graphene on hexagonal boron nitride (Gr/h-BN) exhibits interesting physics stemming from its Moiré superlattice, such as the fractional quantum Hall effect and superconductivity. [1-4] In research on this topic, devices based on graphene nanoribbons (GNRs) on h-BN (GNR/h-BN) with a 1D Moiré superlattice have been intensively studied for use in logic gates and spintronic devices. [5-9] In these systems, the confinement-gap energy of the GNRs, which enables a logic-gate operation, is inversely proportional to the width of the GNRs, W, [10] and is also affected by their chirality. [5] Gani et al. predicted that h-BN could induce very strong spin splitting at the edge states of a GNR on an h-BN substrate. [11] To realize oriented GNR/h-BN devices, Chen et al. used chemical vapor deposition to grow GNRs embedded in h-BN trenches created using a Ni-particleassisted etching method. [12] Epitaxial growth of GNRs along the atomic steps of the h-BN structure has also been demonstrated, where W was successfully controlled in the range 15-150 nm through modulation of the growth time. [13] Importantly, compared with the electron-hole charge fluctuations due to charge puddles in Gr/ SiO 2 , those in Gr/h-BN are suppressed by as much as two orders of magnitude, [14] allowing the development of a high-mobility Gr/h-BN device. [15] Given these properties of Gr/h-BN, a GNR/h-BN device is expected to show enhanced mobility compared with the corresponding GNR/SiO 2 device. [13] The origin of charge puddles has been mainly studied in Gr/SiO 2 systems. Using a scanning single-electron transistor, [16] Martin et al. showed that the charge density fluctuation due to charge puddles does not originate from trapped oxide charges in SiO 2 .I n another work, Zhang et al. showed that charge puddles with an average length scale of 20 nm result from charge-donating impurities between Gr and SiO 2 , not from topographical corrugations related to the surface of a substrate. [17] They suggested that, because the samples were prepared in ambient conditions, air molecules such as N 2 and H 2 O trapped between graphene and the SiO 2 substrate may be the origin of the charge puddles. Given this, it is necessary to know if the puddle landscape in a