Xavier Cartoixà, Dept. d'Enginyeria Electrònica, Universitat Autònoma de Barcelona
In order for graphene and other two-dimensional materials to become competitive players amongst the myriads of electronic devices, one of the main technological hurdles that must be overcome is achieving a low contact resistance (Rc) so that high frequency performance is not compromised [1]. For graphene, it is often quoted that an Rc value lower than 100 Ω•μm is desirable, while larger values are thought to be a limiting factor on the graphene field effect transistor performance [2,3]. Recent experiments have achieved this landmark value with a top contact geometry [4]. On the other hand, an edge contact has been shown experimentally to achieve contact resistances with similar or lower values than most top contacts [5], challenging the conventional wisdom that having a large contact area will result in a decreased value of the contact resistance.
We will argue that ballistic electron injection into graphene (or any other 2D material) is basically a perimeter-dependent phenomenon, dependent only on the atomistic details of the graphene-metal configuration at the edge of the metal, and pretty much independent on the amount of metal-2DM overlap.
Our arguments are supported by first principles calculations of the conductance of a model {Ni,Al,Pd}(111)/Graphene contact, where the type of binding (chemi- vs physisorbed) relates to the size of the fluctuations of the transmission curves, but does not show any clear dependence on the amount of overlap. In fact, having a large overlapping region between the metal and the 2D material may be detrimental to the goal of a low contact resistance.
References
[1] J. S. Moon and D. K. Gaskill, IEEE Trans. Microwave Theory Tech. 59, (2011) 2702.
[2] A. Venugopal, L. Colombo and E. M. Vogel, Appl. Phys. Lett. 96, (2010) 013512.
[3] Bo-Chao Huang, Ming Zhang, Yanjie Wang and Jason Woo, Appl. Phys. Lett. 99, (2011) 032107.
[4] J. S. Moon, M. Antcliffe, H. C. Seo, D. Curtis, S. Lin et al., Appl. Phys. Lett. 100, (2012) 203512.
[5] L. Wang, I. Meric, P. Y. Huang, Q. Gao, Y. Gao et al., Science 342, (2013) 614.
