Extrapolating the properties of individual carbon nanotubes (CNTs) into macro-scale CNT materials using a continuous and cost effective process offers enormous potential for a variety of applications. The floating catalyst chemical vapour deposition (FCCVD) method discussed in this dissertation bridges the gap between generating nano- and macro-scale CNT material and has already been adopted by industry for exploitation. A deep understanding of the phenomena that occur within the FCCVD reactor and how to control the formation of the catalyst nanoparticles is, therefore, essential to producing a desired CNT product and successfully scaling up the FCCVD process. This dissertation connects information on the decomposition of reactants, axial catalyst nanoparticle dynamics and the morphology of the resultant CNTs and demonstrates how these factors are strongly related to the temperature and chemical availability of reactants within the reactor. For the first time, in-situ measurements of catalyst particle size distributions paired with reactant decomposition profiles and detailed axial SEM studies of formed CNT materials revealed specific temperature domains that have important implications for scaling up the FCCVD process. A novel observation was that the evaporation and re-condensation of catalyst nanoparticles results in the formation, disappearance and reformation of the nanoparticles along the reactor axis. The combined influences of pyrolytic carbon species and catalytic nanoparticles are shown to influence CNT aerogel formation. This work also examines the source of carbon in the formed CNTs and the location of aerogel formation. Axial measurements using isotopically-labelled methane (C13H4) demonstrate that carbon within all CNTs is primarily derived from CH4 rather than some of the early-forming CNTs being predominantly supplied with carbon from decomposed catalytic precursor components. Quantification of CNT production along the axis of the reactor dispels the notion that injection parameters influence CNT formation and shows that bulk CNT formation occurs near the reactor exit regardless of the carbon source (CH4, toluene or ethanol). By supplying carbon to different reactor locations, it was discovered that CNT aerogel formation will occur even when carbon is delivered near the exit of the reactor provided the carbon source reaches a temperature sufficient to induce pyrolysis (>1000°C). Furthermore, experimental studies that identify a new role of sulphur (S) in the CNT formation process are discussed in this work. Analogous to effects observed in other aerosol systems containing S, in the FCCVD reactor, S lowers the nucleation barrier of the catalyst nanoparticles and enhances not only CNT growth but catalyst particle formation itself. The new concept of critical catalyst mass concentration for CNT aerogel formation was identified by implementing the novel approach of completely decoupling catalyst particle formation from CNT aerogel production. Rather than aerogel formation being dependent on a critical particle number concentration and ideal sized catalyst nanoparticles at the entrance of the reaction furnace, it was identified that the important metric is instead a minimum critical catalyst mass concentration. Application of the principle using other catalyst precursors such as cobaltocene, with continuous CNT aerogel formation from cobalt based catalyst nanoparticles being reported for the first time, and iron-based nanoparticles from a spark generator, provides proof of the new principle’s robustness and ubiquity. In addition to the experimental studies above, theoretical studies have been carried out to understand the agglomeration occurring in a CNT aerosol. The agglomeration eventually leads to a gas phase synthesized CNT aerogel at the end of the reactor, which can be collected and spun continuously.
The results of this work are not only scientifically interesting, they also provide a strong foundation for further research aimed at optimizing and controlling large-scale CNT reactors by modifying downstream dynamics.