Interfaces with Epitaxial Graphene

Graphene is a single atomic layer of graphite that can have very high carrier mobilities, chemical and mechanical robustness, and novel 2D properties [1]. Since it is entirely carbon, it can also have a very long spin diffusion length due to minimal spin-orbit scattering [2-5]. We grow graphene directly on a SiC(0001) surface by thermally desorbing Si at high temperatures (Figure 1). This creates "epitaxial" graphene [6,7] that has the advantage of being more industrially scalable than the "exfoliation" process used in the recent Nobel-prize winning work of Novoselov and Geim (link to Nobel lectures).

Figure 1. Growth of epitaxial graphene in UHV.

Any device application involving graphene requires the creation of interfaces with other materials. Minimally, electrical contacts are required to inject carriers and often one or more electrical "gates" are needed to control carrier flow in the graphene. Our research goal is to create and characterize interfaces between epitaxial graphene on SiC(0001) and materials relevant to a graphene Spin Field Effect Transistor (spin-FET, Figure 2) [5] . This device represents the prototype application for graphene-based spintronics, but there are difficult materials science challenges to be overcome. Chief among these is the question of what material and processing conditions should be used to create the ferromagnetic gate dielectric needed for a spin FET. While some plausible material candidates exist for this application, it is likely that a practically-suitable material has not yet been invented.

Figure 2. Schematic of a graphene spin-FET proposed in Ref. [5].

We have started our interfacial studies for the graphene spin-FET by characterizing the growth of the 3d transition metals Fe and Co on epitaxial graphene since these might be common magnetic source/drain electrodes [4,5]. Direct growth of these metals results in three-dimensional islands (Volmer-Weber growth mode) and a correspondingly rough initial surface as shown in Figure 3.

Figure 3. STM image showing small Fe islands grown directly on epitaxial graphene on SiC(0001). The inset shows the line profile marked in blue in the image.

Our approach to improving film growth is to use the starting SiC(0001) substrate to modify the metal film growth mode prior to graphene creation (Figure 4). We have observed that Fe(110) grows epitaxially on the root 3 reconstructed surface of SiC(0001). Subsequent processing steps could lead to better contacts for graphene spintronics. This possibility shows the versatility of the SiC substrate for graphene-based electronics and may lead to new strategies for device creation.

Figure 4. STM image of an epitaxial Fe(110) film grown on the root-3 surface of SiC(0001).

References

[1]Das Sarma et al., Rev. Mod. Phys. 83, 407 (2011)

[2]Tombros et al., Nature 448, 571 (2007).

[3]Cho et al., Appl. Phys. Lett. 91, 123105 (2007).

[4]Han et al., Phys. Rev. Lett. 105, 167202 (2010).

[5]Semenov et al., Appl. Phys. Lett. 91, 153105 (2007).

[6]Berger et al., J. Phys. Chem B 108, 19912 (2004).

[7]de Heer et al., J. Phys. D: Appl. Phys. 43, 374007 (2010).

[8]Bostwick et al., Prog. Surf. Sci. 84, 380 (2009).