Scalable Production of Graphene Oxide Using a 3D-Printed Packed-Bed Electrochemical Reactor with a Boron-Doped Diamond Electrode

Abstract

Although graphene oxide (GO) has shown enduring popularity in the research community, its synthesis remains cost prohibitive for many of its demonstrated applications. While significant progress has been made on developing an electrochemical route to GO, existing methods have key limitations regarding their cost and scalability. To overcome these challenges, we employ a combination of commercially available fused-deposition-modeling-based 3D printing and highly robust boron-doped diamond with a wide electrochemical potential window to fabricate a scalable packed-bed electrochemical reactor for GO production. The scalability of the reactor along the vertical and lateral dimensions was systematically demonstrated to facilitate its eventual industrial application. Our current reactor is cost-effective and capable of producing electrochemically derived GO (EGO) on a multiple-gram scale. By oxidizing flake graphite directly in an 11.6 M sulfuric acid electrolyte, the production of EGO was streamlined to a one-step electrochemical reaction, followed by a simple water-wash purification. Almost all of the converted graphite oxide can be recovered, and the final mass yield is typically 155% of the starting graphite material. The as-produced EGO is dispersible in water and other polar organic solvents (e.g., ethanol and dimethylformamide) and can be exfoliated down to predominantly single-layered GO. Through a detailed study of the product intermediates, the graphite was found to first form a stage III or higher graphite intercalation compound, followed by electrochemical oxidation proceeding from the top of the packed graphite bed down. The EGO can be easily deoxygenated with low-temperature thermal annealing (<200 °C) to produce thermally converted EGO with significantly enhanced conductivity, and its promising application as a conductive nanofiller in lithium-ion battery cathodes was demonstrated. The simplicity, cost-effectiveness, and unique EGO properties make our current method a viable contender for large-scale synthesis of GO.