Spectroscopic characterisation of SWNT polymer composites
This item is available under a Creative Commons License for non-commercial use only
In this study hybrid systems of the conjugated organic polymer poly(p-phenylene vinylene-co-2,5-dioctyloxy-m-phenylene vinylene) (PmPV) with single-walled carbon nanotubes synthesised by the Arc discharge method and by gas-phase catalytic decomposition of carbon monoxide at high pressure (HiPco process) are explored using a wide variety of spectroscopic, microscopic and thermal techniques. Diameter dependent solubilisation has been previously shown in solutions of such composites. Firstly the selective interaction is investigated using UV/Vis/NIR absorption and Raman spectroscopy at an excitation wavelength 514.5nm. Examination of the radial breathing modes (RBMs) shows similar tube diameters of 1.2nm and 1.35nm were selected for both the Arc discharge and HiPco composites. The corresponding G-lines of both composites shows no specific type of tubes being selected. At 514.5nm the G-line of the HiPco composite (1%mass fraction) shows contributions from semiconducting and metallic tubes, while the Arc discharge composite (1% mass fraction) is dominated by semiconducting nanotubes. At 632.8nm the G-line of the HiPco composite (1%mass fraction) is dominated by semiconducting tubes, while the Arc discharge composite (1% mass fraction) shows strong contributions from metallic tubes. This finding is a strong indication that the selection process is dependent on tube diameter rather than backbone structure. The solubility limits of both composites are determined by investigating the G-lines of both composites and have been found to be greater than 1% mass fraction by weight for the Arc discharge composite and greater than 0.1% mass fraction by weight for the HiPco system. For the HiPco composite the greatest G-line downshift (~12 cm -1) occurred between 1% and 0.1% mass fraction. This downshift in the G-line, which also experience a decrease in full width at half maximum, results from the polymer intercalating into the nanotube ropes thus causing them to debundle. For this reason temperature dependant measurements were carried out at 0.1%mass fraction to further investigate this interaction. An endothermic transition is observed in the differential scanning calorimetry (DSC) studies of both the polymer and 0.1% HiPco composite in the region of 50°C. Also observed in the DSC of the composite is a double peaked endotherm at -39°C and -49°C which does not appear in the polymer. The Raman spectrum of the polymer upon increasing the temperature to 60°C shows a diminished cis-vinylene mode at 1575cm-1., with an increase in relative intensity of the trans-vinylene mode at 1630cm-1. Partially irreversible changes in isomerisation suggest increased order in the polymer. This change in the polymer is also manifest in the Raman spectrum of the composite upon increase to 60°C, where the spectrum becomes abruptly dominated by nanotubes. Raman spectra of the composite show no change at -35°C, however infrared absorption measurements suggest that the transition at -35° shows a change in the absorbance of the polymer side chain aryl-oxide linkage at 1250cm-1 and alkyl-oxide stretch at 1050cm-1. Infrared spectra thus suggest that the transitions in the lower temperature region ~-35°C are side chain induced, while Raman spectra suggest that the transition at 60°is backbone induced. Furthermore temperature cycling induces an irreversible decrease in the mean fluorescence intensity of the composite. This result indicates that an increase in cyrstallization of the composite is supported and enhanced by an increase in the ordering of the polymer. While carrying out temperature dependent Raman measurements on the G-line, changes in the RBM spectral distribution were also observed. At ambient termperatures size selective solubilisation occurred with tubes of ~1.2-1.4nm diameters dominating the spectrum. However, upon heating above the glass transition of the polymer (50°C) the contributions from smaller tubes around ~0.9nm increased significantly in relative intensity This suggests that below the glass transition of the Polymer (50°C) RBMs within the composite are damped and spectral changes cannot be interpreted as diameter selective solubilisation. The observed RBM damping at room temperature only occurred up to a concentratin of ~2X10-4 g/l and below this no damping was observed. Fluorescence intensity measurements were taken for a range of PmPV concentrations, in which HiPco sigle walled carbon nanotubes (SWNTS) were added. Fitting of the concentration dependence to a dynamic absorption/desorption model indicates that the nanotubes exist as bundles until a critical concentration of 2X10-4 g/l is reached, below which the nanotubes are isolated. This particular study shows that polymer and/or solvent has a significant effect on the debundling and aggregation within these systems. Bundling at higher concentrations can mask RBMs in the composite at ambient temperatures, providing an incomplete representation of the selection of diameters present within composites at a particular wavelength. The temperature coefficient of Raman frequency shift for the raw and composite materials was investigated. This coefficient is specific to the material under investigation and it has been shown that impurities, defects and disorder induce elevated coefficients in SWNTs. The composite exhibits a different coefficient to that of the raw materials and therefore has its own intrinsic thermal properties. This further highlights that the composite is not just a superposition of both the tube and the polymer but a unique material with its own physical and thermal properties.
Keogh, Sinead (Thesis), "Spectroscopic characterisation of SWNT polymer composites" (2006). Doctoral. Paper 20.