Regenerated cellulose fibres like Cordenka and Lyocell have been studied for their potential use as reinforcement in polymer composites. These fibres are attractive candidates for improving the mechanical and environmental characteristics of various polymer materials. In our research group we have devolved the idea of manufacturing ‘all-cellulose’ composites from a single cellulosic source. The idea is to create the ‘matrix’ of all cellulose composite by selectively dissolving the surface of each fibre or filament, which on coagulation forms the matrix. Being also cellulose, this should give excellent compatibility/adhesion between the phases.
This thesis has studied the dissolution of two commercial regenerated cellulose yarns, namely Cordenka™ and Lyocell™. Optical microscopy, Wide-angle X-ray diffraction (WAXS) and mechanical testing techniques have been used to track the dissolution of these multifilament bundles in the ionic liquid 1-ethyl-3-methyl-imidazolium acetate [C2mim]+ [OAc]- for different times and temperatures. This allowed both the speed of the dissolution to be determined at different temperatures as well as the dissolution activation energy Ea from time-temperature superposition.
The different nature of the multifilament bundles (Cordenka™, which was untwisted and Lyocell™ where the bundles were twisted) resulted in different techniques being most suitable for their study. For Cordenka, WAXS and mechanical measurements on partially dissolved composite filaments proved most successful. In the dissolution process, the oriented cellulose II crystals in the regenerated cellulose fibres dissolve and then reform into randomly oriented crystals to form a matrix phase. This change in orientation enabled the dissolution process to be followed and hence determine the growth of the dissolved matrix fraction of v_m with time and the dissolution activation energy. On the other hand, optical microscopy was found to work very well with the Lyocell multifilament bundles to directly determine the dissolved matrix volume fraction v_m. Mechanical measurements of Young’s Modulus and ultimate tensile strength on partially dissolved composites proved successful for both Cordenka and Lyocell multifilament yarns.
The change in the average molecular orientation P_2 determined from an azimuthal (α) X-ray scan, allowed the growth of the matrix volume fraction v_mto be calculated with time and temperature. This is an indirect measurement and relies on using a rule of mixtures approach.
The optical microscopic method offered a direct method to measure the growing area of the dissolved and coagulated fraction for the Lyocell multifilament bundle with increasing time and temperature. The twisted fibres meant that the dissolved fraction formed a ring on the outside of the multifilament, allowing a measurement of the decrease of the inner core (the undissolved original fibre fraction) and the increase in the area and thickness of the dissolved and coagulated outer ring. The decrease of the inner core and the growth of coagulation fraction C.F. and the thickness and area of the dissolved and coagulated outer ring was found to follow time temperature superposition, with an Arrhenius behaviour, giving consistent values for the activation energy of Ea= 141 ± 15, Ea= 141 ± 16 and Ea= 127 ± 14 respectively.
Young’s modulus and ultimate tensile stringth was measured on all the resulting processed composites for Cordenka and Lyocell multifilament bundles. The fall of Young’s modulus and ultimate tensile strength with dissolution time and temperature was found to follow time-temperature superposition for the Cordenka multifilament bundle, with an Arrhenius behaviour giving a value for Ea= 198± 29 kJ/mol. The Young’s Modulus and ultimate tensile strength results were plotted against v_m determined from the WAXS measurements and were found to agree well to the Voigt upper bound parallel Rule of Mixtures. This suggests that the resulting composites are well bonded and that the dissolved Cordenka material (which has a higher molecular weight compared to the Lyocell material) is a suitable matrix material for to make all a cellulose composite.
For the Lyocell multifilament bundle, the Young’s modulus of the processed composites was found to be quite scattered and so it could not be ascertained if this followed time-temperature superposition. However, the fall of the ultimate tensile strength of the composites with dissolution time and temperature was found to follow time-temperature superposition, with an Arrhenius behaviour giving a value for Ea= 144± 27 kJ/mol. The ultimate tensile strength results plotted against v_m determined from the optical microscopic method was found to lie significantly below the Voigt rule of mixtures. This suggests that either the dissolved Lyocell material is less successful as a matrix, or that the twisted nature of the Lyocell multifilaments does not allow dissolution to happen in the interior of the bundle as the ionic liquid cannot penetrate.
In terms of the difference between the Cordenka and Lyocell multifilament bundles, it was found from the Optical microscopic results, that the geometry of the Cordenka multifilament bundle is untwisted with a few hundred individual multifilaments, which appeared as a loose microstructure with significant inner spaces in between. On the other hand, the geometry of Lyocell multifilament bundle is twisted with few hundred individual fibres that are close to each other without significant inner spaces. The Cordenka multifilament bundle has higher average orientation, and a higher Young’s modulus, ultimate tensile strength, and activation energy compared to the Lyocell multifilament bundle, which we attribute to the fibres being untwisted. The Lyocell bundle has lower average orientation, which was shown to be due to the significant twist of the bundle.
These findings, especially the geometry and molecular weight lead to the Cordenka multifilament bundle having a faster dissolution rate than the Lyocell multifilament bundle. The comparative geometry (untwisted fibres), the speed of dissolution and the higher molecular weight, lead to the important result that the Cordenka multifilament bundle would make an excellent basis for an all- cellulose regenerated fibre composite (ACC). However, it is appreciated that if woven cloth is to be used to manufacture all-cellulose composites (ACC) then some degree of twist will be required to stop the individual fibres from breaking during the weaving process, so there is maybe an optimum bundle twist to be discovered in any future work.
