Multiscale design of hierarchical self-assembled graphene network for efficient thermal dissipation.

Network structures with branches have been shown to be an effective way to dissipate heat from a small interior region to the periphery. The branching junction in these structures is a key factor in the total thermal conductivity of the entire structure. Using molecular dynamics simulation, we explore this network geometry using rectangular graphene flakes as the building blocks to study thermal dissipation. The network is created by allowing each graphene flake to branch at the junction at a preferred angle. The position of the junction at each flake can be restrained by allowing hydrogen bond using a functional group. We use Lennard Jones potential to model the bilayer graphene interaction at the junction. We study the binding energy profile due to overlap with rotation between two flakes at the branching junction. We see that there is a transition of increased probability of flakes branching at from 0 degrees to 90 degrees with increasing width of flakes. We see a similar opposite transition occurring as we increase the junction position of the flakes from the periphery to the center. We also see that there is increased probability of the allowed angles as the temperature increases. We use this binding energy at the junction to create an ensemble using rejection sampling to design a hierarchical self-assembled network. We compare the thermal conductivity of these different network structures with a branching probability at each junction.