The basic concept of the project is shown in Fig. 2, where single-layer graphene is formed on a substrate using a standard exfoliation process (Fig. 2(a)). Next, SiO2 is deposited, patterned by electronbeam lithography (EBL) and etched to form guiding structures with dimensions in the order of ~ 20 nm. Next, nanoparticles can be formed on the surface. Instead of the typical random dispersion method, top-down patterning methods (e.g. EBL or block copolymer lithography) can be utilized to pattern Ni particles as shown in Fig. 2(b). If the guiding structures are aligned along the directions favorable for crystallographic etching, then upon heating in a H-containing ambient, the nanoparticles should travel between the guiding structures, directionally etching the graphene in a continuous straight line until exiting at the ends of the trench (Fig. 2(c)). The etch process has previously been described as being due to a carbon dissolution mechanism, whereby carbon atoms dissolve into the catalyst at the graphene edges and then diffuse and react with hydrogen on the catalyst surface . Directional etching is favored since turning comes at an adhesive energy cost as the catalyst must partially detach from the graphene interface. By the same argument, the guiding structures should also suppress lateral motion of the nanoparticles during etching. Furthermore, since the etch process only takes place at the boundary between the catalyst and the graphene, no undercutting of the SiO2 features should occur. Therefore, after etching, aligned graphene nanoribbons with well-controlled edge orientation should result as shown in Fig. 2(d).
The majority of the seed effort will be aimed at addressing the primary material science questions that will determine the feasibility of the nanoribbon fabrication process just described. An explanation of these questions and the approaches by which they will be addressed are described in the section below.
Nanoparticle position: In previous studies, the correlation between the initial position of a nanoparticle and the resulting etch track was not rigorously characterized. In this work, because the initial nanoparticle position is known, it should be possible to study this dependence very precisely, and such studies will be performed using both guided and unguided geometries.
Role of defects: It has previously been noted that deviations from linear nanoparticle etching are primarily caused by defects in the graphene . Since fabrication of the guiding structures will require dry etching, a process that could introduce defects into the underlying graphene, experiments to study the impact of defects on nanoparticle graphene etching will be critical. A series of simple studies will therefore be planned whereby ion damage is intentionally induced onto a blanket graphene surface, and then the effect of the modified surface on the nanoparticle motion and etching can be characterized.
Barrier properties of guiding structures: The ability of the guiding structures to act as barriers to the catalytic etching will also require investigation, since the process is unlikely to yield oriented nanoribbons if nanoparticles are not confined within the boundaries of the patterned guiding structures. To study the blocking effect of the guiding structures on nanoparticle motion, an isolated SiO2 boundary feature can be patterned such that it separates nanoparticles dispersed onto the graphene into two regions. Etch tracks can then be monitored to determine how particles react to the presence of the barrier (i.e., do they cross, turn around, or run parallel to the barrier, etc.) and these properties can be studied as a function of particle size, guiding structure dimensions, and annealing conditions.
Orientation dependence of guiding structures: Investigating the dependence of the nanoparticle etching on the guiding structure orientation will be of critical importance in understanding the dynamics of the etch process. For instance, it has been shown that the bond breaking process for armchair termination is 10x more energetically favorable than for zigzag edge bonds, suggesting that the formation of armchair ribbons could be difficult. Studying the orientation dependence of the guiding structures could help to determine if the system can be forced into a configuration where the formation of zigzag edges is suppressed, thereby allowing ribbons with armchair edges to be energetically favorable.
435 Amundson Hall, 421 Washington Ave. SE, Minneapolis, MN, 55455
P: 612-626-0713 | F: 612-626-7805