Storage of Hydrogen in Multilayer Graphene

This project has investigated electrochemical hydrogen storage in multilayer graphene (MLG) electrodes with a proton battery setup. The proton battery combines both electrolyser (charging) and fuel cell (discharging) modes in one compact unit. In the design developed at RMIT University, it is effectively a reversible proton exchange membrane (PEM) fuel cell, with an integrated storage electrode to store neutralised protons, that is, atomic hydrogen (H). It is thus a very different approach to that of the usual hydrogen storage system with multiple components. Previous work at RMIT University on the proton battery concept has focused mainly on activated carbon (AC) electrodes. An electrochemical hydrogen storage (H-storage) of 0.8 wt% using unheated activated carbon from phenolic resin electrodes has been achieved, and 2.2 wt% with heated discharges (~ 70 °C). The present project has focused on the potential to increase H-storage both per unit mass and per unit volume of the electrode by employing MLG electrodes, considering storage by electric double layer capacitance, pseudocapacitance and chemisorption mechanisms. Research was undertaken on a number of MLG materials including chemically converted graphene oxide, graphene nanoplatelets (from Sigma Aldrich), graphene oxide (GO) (from Sigma Aldrich), reduced graphene oxide (rGO) and GO/rGO/single wall carbon nanotube (SWCNT) composites. These materials have been characterised by X-ray diffraction, scanning electron microscope, transmission electron microscope, X-ray photoelectron spectroscopy, Raman spectroscopy, Fourier transform infrared spectroscopy, Brunauer-Emmett-Teller surface area, and pore size distributions, as well as electrical conductivity measurements. The electrochemical behaviour of the various MLG electrodes has been measured using cyclic voltammetry, galvanostatic charge & discharge, and electrochemical H-storage in a proton battery. The H-storage results were in the following ranges: 0.19 – 0.51 wt% for chemically converted graphene oxide, 0.30 – 1.80 wt% for graphene nanoplatelets, 0.61 – 1.00 wt% for graphene oxide, 0.37 – 0.49 wt% for reduced graphene oxide, and 0.13 – 3.41 wt% for GO/rGO/SWCNT composites. It was concluded that a number of key factors significantly affected the gravimetric H wt% recovered from a charged electrode. These included the graphene morphology and inter-layer separation, functional group presence(s), and certain features of the testing method employed. The latter included, in particular, the soak time of the electrode in the electrolyte prior to charging and discharging, the faradic charge level delivered to the electrode and the electrode discharge temperature. The research has led to definition of a set of criteria for designing high performing MLG electrodes for the future. These criteria include an interlayer separation within the MLG of between 0.8 and 1 nm on average, holes in or gaps between layers to allow hydronium transport between layers, a preference for H-storage by chemisorption over pseudocapacitance over double layer capacitance, high electron conductivities both in-plane and between planes, and high proton conductivity supported by the Grotthus mechanism when acid is absorbed between layers. Some opportunities to increase the H-storage of MLG electrodes are identified such as preparing MLG electrodes with much higher specific surface areas thereby exposing more carbon surface for electrochemical hydrogen (EC) H-storage, increasing the MLG interlayer electron conductivity by preparing composites containing a conductive polymer component (such as polyaniline or polypyrrole) or synthesising targeted MLG/SWCNT composites based on hydrogen weight percent (H wt%) results achieved thus far.