TABLE OF CONTENTS Whichever type of nanotube is selected, it will be necessary to develop new synthesis routes. At present, catalytic MWCNTs and CNFs appear as the optimum choice, given that such materials can be most readily obtained in large quantities with a high purity. However, these materials are intrinsically defective and wavy, both of which are expected to be highly detrimental to the mechanical performance [223]. Somehow nanotubes with a crystalline quality closer to arc-grown nanotubes need to be obtained at a cost similar, or indeed below, current CVD-grown products.
As the absolute size of the reinforcement decreases towards the size of the polymer molecules, interactions between filler and matrix become more important, both during processing as well as in the solid-state. Significant progress has been made in under-standing the interactions between polymers and flat surfaces, but the interactions of polymers with highly curved surfaces at the molecular scale, are still largely unknown. Variations in crystallinity are important but the effects of constraint and other changes in polymer morphology in the vicinity of highly curved surfaces may also have significant effects on the deformation behaviour of the composite. As in biological nanocomposites, a high strength might not be exclusively linked to the intrinsic strength of the filler but might reflect increased yield stresses in the vicinity of the filler, as recently observed during pull-out experiments [224]. Given these issues and the change in scale towards molecular dimensions, it is not surprising that concepts of traditional fibre-reinforced composites are often not...
