Optimized nanodevice fabrication using clean transfer of graphene by polymer mixture: Experiments and Neural Network based simulations

Abstract

In this study, we investigate both experimentally and computationally the molecular interactions of two distinct polymers with graphene. Our experimental findings indicate that the use of a polymer mixture reduces the transfer induced doping and strain in fabricated graphene devices as compared to conventional single polymer wet transfer. We found that such reduction is related to the decreased affinity of mixture of polymethyl methacrylate and angelica lactone polymer for graphene. We investigated changes in binding energy (BE) of polymer mixture and graphene by considering energy decomposition analysis using a pre-trained potential neural network. It was found that numerical simulations accurately predicted two-fold reduction of BE and order of magnitude reduction of electrostatic interaction between polymers.