Processing graphene oxide and carbon nanotubes: routes to self- assembly of designed architectures for energy storage applications

The development of processing routes, methods and protocols to process graphene oxide and carbon nanotubes for Energy storage applications is presented in this thesis. The main objectives that are addressed in this thesis are: i) developing routes for solvophobic soft self-assembly of ultra-large liquid crystalline (LC) graphene oxide (GO) sheets in a wide range of organic solvents, many of which were not known to afford solvophobic self-assembly prior to this thesis, ii) developing a solid understanding in the required criteria and mechanisms through which the solvophobic self-assembly is enabled and providing new insights contributing to the fundamental understanding of the solvophobic effect and the parameters affecting the selfassembly process, iii) probing and quantifying dynamic rheological behaviour of LC GO dispersions to extend the general understanding of the mechanics involved in order to deliberately translate these intrinsic properties into the design process and device applications, iv) utilizing the fundamental knowledge gained through these investigations to create a platform to process these materials at industrially highly-scalable levels for energy storage applications. As such, a novel soft self-assembly process is first introduced to synthesize graphene oxide liquid crystals with an ultra large sheet size in a wide–range of solvents based on a solution-phase method involving pre-exfoliation of graphite flakes. Spontaneous formation of lyotropic nematic liquid crystals is identified upon the addition of the ultralarge graphene oxide sheets in these solvents above a critical concentration of about 0.025 wt%. It is the lowest filler content ever reported for the formation of liquid crystals from any colloid, arising mainly from the ultrahigh aspect ratio of the graphene oxide sheets of over 30000. It is then, demonstrated that the scalable liquid crystal route can be employed as a new method to prepare unique 3-D framework of graphene oxide layers with proper interlayer spacing as building blocks for cost-effective high-capacity energy storage media. The intercalation of MWCNTs as 1D spacers between graphene oxide framework results in a strong synergistic effect between the two materials consequently leading to a robust and superior hybrid material with higher capacitance compared to either graphene oxide or MWCNTs and unrivalled hydrogen storage capacity at ambient temperature. Based on rheological insights, an entirely new, scalable, and commercially viable wet-spinning strategy is demonstrated to fabricate unlimited lengths of highly porous, yet densely packed and mechanically robust and flexible all-around multifunctional graphene yarns from liquid crystals of ultra-large graphene oxide sheets, for the first time. The produced yarns, which are the only practical form of these architectures for real-life device applications, are found to be mechanically robust (Young’s modulus in excess of 29 GPa), with high native electrical conductivity (2508 ± 632 S m−1) and exceptionally high specific surface area (2605 m2 g-1 before reduction and 2210 m2 g-1 after reduction) and capacitances as high as 410 F g-1/electrode in a practical two electrode configuration set-up. At the next step, the feasibility of achieving biaxial liquid crystals employing hybrid dispersions of one-dimensional and two-dimensional colloidal dispersions is demonstrated. It is shown that the collective behaviour and coordination of individual components in such biaxial liquid crystals enables stigmergic ordering which is the first report on stigmergic-emergent intelligence in non-biological materials. As an example of many promising applications, it is demonstrated that the inherent electrochemical, electrical properties of the final self-constructed architectures are practical in alternative energy storage and conversion devices. The system is shown to exhibit excellent high-rate capability far exceeding the recent literature capacitance values reported for other comparable architectures including aligned/patterned SWNTs thin films supercapacitors and microdevices even at much lower current densities