Optimizing PMMA solutions to suppress contamination in the transfer of CVD graphene for batch production.

Abstract

Mass production and commercial adoption of graphene-based devices are held back by a few crucial technical challenges related to quality control. In the case of graphene produced by chemical vapor deposition, the transfer process represents a delicate step that can compromise device performance and reliability, thus hindering industrial production. In this context, the impact of poly(methyl methacrylate) (PMMA), the most common support material for transferring graphene from the Cu substrate to any target surface, can be decisive in obtaining reproducible sample batches. Although effective in mechanically supporting graphene during the transfer, PMMA solutions needs to be efficiently designed, deposited, and post-treated to serve their purpose while minimizing potential contaminations. Here, we prepared and tested PMMA solutions with different average molecular weight (AMW) and weight concentration in anisole, to be deposited by spin coating. Optical microscopy and Raman spectroscopy showed that the amount of PMMA residues on transferred graphene is proportional to the AMW and concentration in the solvent. At the same time, the mechanical strength of the PMMA layer is proportional to the AMW. These tests served to design an optimized PMMA solution made of a mixture of 550,000 (550k) and 15,000 (15k) AMW PMMA in anisole at 3% concentration. In this design, PMMA-550k provided suitable mechanical strength against breakage during the transfer cycles, while PMMA-15k promoted depolymerization, which allowed for a complete removal of PMMA residues without the need for any post-treatment. An XPS analysis confirmed the cleanness of the optimized process. We validated the impact of the optimized PMMA solution on the mass fabrication of arrays of electrolyte-gated graphene field-effect transistors operating as biosensors. On average, the transistor channel resistance decreased from 1860 to 690 Ω when using the optimized PMMA. Even more importantly, the vast majority of these resistance values are distributed within a narrow range (only ca. 300 Ω wide), in evident contrast with the scattered values obtained in non-optimized devices (about 30% of which showed values above 1 MΩ). These results prove that the optimized PMMA solution unlock the production of reproducible electronic devices at the batch scale, which is the key to industrial production.