Deformation micromechanics of model regenerated cellulose fibre-epoxy/polyester composites

The deformation micromechanics of model cellulose-epoxy and cellulose-polyester composites are presented. Two model systems are used; namely droplet-fibre and a thin flat film-fibre geometries. Each system consists of a high performance single fibre filament of cellulose travelling through a resin. The mechanical properties of these single filaments are reported and their potential as reinforcements in composite materials is highlighted. The molecular deformation, with stress, of the fibres is analysed using Raman spectroscopy. Shifts in the mean position of a band associated with the backbone deformation of the polymer are shown to be a useful calibration of local fibre stress. This is subsequently used to map stress along the interface between the fibres and the resin systems. The interfacial shear stress is then determined using fits to the data and an analysis that returns this parameter as a function of distance along the fibre. It is shown for the epoxy droplet system that the interface between epoxy and the cellulose remains intact. However, a very large maximum interfacial shear stress between the fibre and resin is returned from the analysis (94.2 +/- 18.3 MPa) which suggests that there may be a constraint at the point at which the fibre enters the resin. The stress is seen to decay quite rapidly and this gives rise to such a high value. The presence of both optical distortion and a residual compressive stress are suggested as reasons for a 'hump' in the centre of the stress profiles. However, the flat film specimens, which lack the curvature of the previous system, reveal a much reduced interfacial shear stress (17.9 +/- 1.9 MPa). Again, the polyester droplet system reveals a very large maximum interfacial shear stress (102.2 +/- 4.3 MPa), but although analysis of the flat film specimens returns a much reduced value of this parameter (8.8 +/- 1.9 MPa) these samples are reported to readily debond. The problems associated with the droplet method are discussed, and given more consistency and better agreement with a constitutive model for cellulose-resin systems, the flat film specimens are suggested as a possible and more reliable alternative. (c) 2006 Elsevier Ltd. All rights reserved.