Graphene (EG), an atomic monolayer of graphite, is an exciting new material for a wide variety of applications, including high-frequency electronics, opto-electronic devices, and micro-electromechanical systems. While EG can be made using various methods, only two—metal-catalyzed growth on copper films and epitaxial growth on silicon carbide substrates—are currently amenable to the large-area EG fabrication needed for commercialization. However, both use substrates that are not ideal for most applications. Thus, it is desirable to develop a method for removing and transferring such large EG films onto arbitrary substrates that are amenable to specific applications.
Efforts to transfer EG from copper and nickel foils have been successfully developed, but no technique exists for epitaxial EG on silicon carbide. These efforts require a chemical wet-etching process that dissolves the metal-catalyst film and involve nonprecise mounting onto the chosen substrate. We have developed a process that could be used more generally, so it would not require consumption of the growth substrate and still allows transfer of large areas of EG onto arbitrary substrates in a simple, inexpensive way.
The EG-transfer method we developed1 is a dry, general-use process capable of transferring large-area, voidless epitaxial and metal-catalyzed EG films onto arbitrary substrates including silicon dioxide, gallium nitride, silicon carbide, and aluminum oxide (see Figure 1). The process uses a thermal-release adhesive tape to remove the EG layers from the growth substrate, after which the removed EG is pressed onto the substrate of interest and the adhesion strength of the tape is eliminated through a thermal process, leaving the EG bound to the substrate. We further improved this process using a thin polymethyl methacrylate (PMMA) film between the EG surface and the transfer tape to help eliminate voids within the transferred films.2
(a) Schematic of the graphene (EG)-removal process from silicon carbide (SiC) substrate, (b) graphene transferred onto an arbitrary substrate, and (c) Nomarski optical microscopic image of a graphene film transferred onto silicon dioxide (SiO2
). (Images created by M. Kraus: http://www.mammalcreative.com
. Accessed 13 October 2010.)
The EG films transferred by this process have enabled a wide array of materials-characterization efforts. For instance, epitaxial EG grown on the carbon-terminated surface of silicon carbide was successfully transferred to silicon dioxide and aluminum oxide surfaces and van der Pauw devices. The structural and electrical properties of the devices were probed through Raman spectroscopy and Hall-effect measurements, respectively, both illustrating that the transfer process does not degrade the structural integrity or electrical properties of the EG layers. This technique also enables processing of devices on transparent substrates for optical measurements, such as transient absorption or optical transmission.
The electrical properties of the transferred films show that the carrier density, or the number of electrons/holes participating in current conduction, was reduced by a factor of two to three following the transfer of epitaxial EG from silicon carbide. No reductions were seen in transferred, metal-catalyzed EG. As the transfer process leaves a residual one- to five-monolayer-thick EG film behind on the silicon carbide surface, it is presumed that this reduction is caused by the highest-carrier-density layers remaining on the surface. Examination of epitaxial EG films attached to the transfer tape (using x-ray photo-emission spectroscopy) determined that a few tenths of a percent of atomic silicon is present, presumably intercalated between the EG sheets. Annealing such films at temperatures of 1200°C in an ultrahigh vacuum for 30min led to reductions in both carrier density and silicon concentration. Thus, it is possible that this atomic silicon is the native dopant within epitaxial EG.
The ability to transfer EG films from the growth substrate onto substrates amenable to specific applications is necessary to fully realize the extraordinary advancements that EG promises for thin-film electronics, opto-electronics, and micro-electromechanical systems. Our technique is an enabling technology for these advancements. We have been able to explore various materials properties of the isolated EG films, for instance, observation of the presence of atomic silicon within epitaxial EG. We are currently exploring the influence of different substrates with varying dielectric constants and surface-phonon-scattering energies3and the optical-damage threshold of monolayer graphene films.
Looking forward, since this technique has now been shown successful in transferring both epitaxial and metal-catalyzed EG, we plan to use it to investigate the optical and electrical properties of various forms of EG, including their optical-damage threshold, the influence of high-dielectric-constant gate dielectrics and isolation layers, and transferring thinner films of epitaxial EG.
Joshua D. Caldwell
US Naval Research Laboratory (NRL)
Joshua Caldwell graduated from Virginia Tech in May 2000 with a BS in chemistry (history minor). His graduate work was in physical chemistry at the University of Florida, from which he graduated in December 2004. He then began a postdoctoral appointment at the NRL, where he is now a staff scientist.