Thickness dependent band gap and effective mass of BN/graphene/BN and graphene/BN/graphene heterostructures Dongyoo Kim a , Arqum Hashmi a , Chanyong Hwang b , Jisang Hong a, ⁎ a Department of Physics, Pukyong National University, Busan 608-737, Republic of Korea b Center for Nano-imaging Technology, Korea Research Institute of Standards and Science, Daejeon 305-340, Republic of Korea abstract article info Article history: Received 7 November 2012 Accepted 31 December 2012 Available online 10 January 2013 Keywords: Band gap BN Graphene Effective mass Using the full potential based WIEN 2K method, we have explored thickness dependent energy band gaps and effective masses of BN layer sandwiched by graphene (G/BN/G) and graphene sandwiched by BN (BN/ G/BN) systems. Here, capping and substrate layer coverages are changed from 1 monolayer (ML) to 2 ML and 1 to 4 ML, respectively. In G/BN/G systems, we find rather small energy gaps and the energy dispersions become quadratic near the K-point. Unlike in G/BN/G, a trilayer BN/G/BN shows large gap of 117 meV and it decreases as the number of BN layer increases, but we still find a gap about 90 meV in BN (2 ML)/G/BN (4 ML) system. The thickness dependent suppression of band gap in BN/G/BN can be nicely interpreted in terms of interlayer distance from BN substrate to graphene. Furthermore, very interestingly, the energy dis- persion is nearly linear near the K-point and this linearity is still preserved even in all BN/G/BN systems. Sur- prisingly, the effective mass decreases as the number of BN layer increases. For instance, the smallest effective mass of 0.00235 m e is estimated in conduction band along K -Γ direction. Overall, our calculations may sug- gest that the BN/G/BN system has potential application for fast radio frequency (RF) device or on-off switching transistor because the linearity and small effective mass is kept without any external factors such as electric field, strain, and doping. © 2013 Elsevier B.V. All rights reserved. 1. Introduction Graphene (G) displays extraordinary physical properties such as peculiar band structure at K-point, high mobility, and fractional quan- tum hall effect [1–5]. Extensive studies have been done to utilize these properties for novel next generation electronic devices. None- theless, the graphene has a zero band gap at K-point and this has be- come a major obstacle to make a graphene transistor because it is hard to completely control on–off switching without a band gap. Therefore, opening a band gap is one the most important issues in the graphene study. Various approaches have been used to open the band gap of graphene. For instance, it is reported that the band gap can be induced by applying electric field, doping, applying strain effect, growing in bi- layer or trilayer structures, substrate effect, etc. [6–10]. For practical de- vice application purposes, two issues should be addressed; (i) size of band gap at K-point for on–off switching ratio and (ii) linearity of ener- gy dispersion relation for high mobility. It is believed that the band gap should be larger than approximately 300 meV according to Si based electronic device experiences for high on–off switching ratio. Besides, a linear dispersion in conduction band is desirable because this linearity may result in high mobility and will be good for fast RF applications. Indeed, the free standing graphene shows remarkable transport proper- ties, however it is limited to take advantage of these outstanding phys- ical properties of free standing graphene layer because of several issues found in device fabrication process. As an alternative way to open the band gap, one can consider a graphene supported on suitable substrate under the condition that the essential properties of graphene are mini- mally affected by a substrate material. Thus, it is favorable to grow the graphene on insulating substrate. In this respect, a hexagonal boron nitride (BN) is receiving great research interests because BN is a wide band gap material and the lattice constant is quite close to that of graphene. Due to this fact, BN is considered as an ideal substrate for graphene device applications. Experimentally, the graphene/BN heterostructure can be formed [11–13] and one can find many re- ports of physical properties found in graphene/BN heterostructures. For instance, it is shown that the band gap can be opened or tuned by ap- plying electric field in bilayer graphene/BN heterostructure [14–16]. The transport properties are also measured in graphene/BN sandwiched structure [17–19]. In graphene/BN heterostructures, mostly BN/graphene bilayer systems or BN/graphene/BN trilayer structures are widely investi- gated and it is shown that the gap can be induced due to the substrate ef- fect or stacking order [20–22]. However, it is rare to find systematic studies of the influence on the band gap and effective mass according to substrate or adlayer thickness change beyond simple trilayer graphene/BN/graphene and BN/graphene/BN configurations using state-of-the-art density functional method. It will be an interesting Surface Science 610 (2013) 27–32 ⁎ Corresponding author. Tel.: +82 051 629 5573. E-mail address: hongj@pknu.ac.kr (J. Hong). 0039-6028/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.susc.2012.12.017 Contents lists available at SciVerse ScienceDirect Surface Science journal homepage: www.elsevier.com/locate/susc