Surface chemistry of cellulose : from natural fibres to model surfaces

The theme of the thesis was to link together the research aspects of cellulose occurring in nature (in natural wood fibres) and model surfaces of cellulose. Fundamental changes in cellulose (or fibre) during recycling of paper was a pragmatic aspect which was retained throughout the thesis with varying emphasis. The applicability of the surface analytical methods of X-ray Photoelectron Spectroscopy (XPS), Attenuated Total Reflectance Infrared Spectroscopy (ATR-IR), and Scanning Electron Microscopy (SEM) to natural fibres is assessed in Chapter 3 and 4. The visual link by SEM suggested that changes are, indeed, taking place, for instance, during recycling of the fibre. Samples of purely an industrial origin, however, provide great obstacles for scientific interpretation (Chapter 3) also due to the lack of correlation in the treatments of the samples. Chapter 4 is a more pronounced example of the contradictions raised by the unclear chemical and morphological definitions of natural fibres. A selected species of natural fibres (bleached kraft) are given defined treatments in controlled conditions. This laboratory recycling procedure shows that the strength properties of the fibres are reduced by only drying and wetting the fibre. However, sophisticated surface analysis, like XPS or ATR-IR, fails to provide answers to what is really happening within the fibres during wetting and drying. The complex chemistry of natural fibres obscures the interpretation. The vagueness of chemical and morphological qualities in natural fibres pointed the way towards a model surface approach, which eventually spawned three individual methods to create three different surfaces. Because of the feeble literature on cellulose model surfaces, a novel method for preparation was devised. The method, described in Chapter 5, served as a basis for two additional cellulose surfaces: open films of nanosized cellulose (Chapter 6) and cellulose domains on cellulose (Chapter 7). Both of the latter cellulose films are unprecedented in literature. The novel method to prepare conventional, closed films of cellulose (Chapter 5) involved first derivatizing the otherwise immiscible cellulose into a hydrophobic derivative, trimethylsilyl cellulose (TMSC). TMSC dissolves easily into hydrophobic solvents and the solution can be spin coated onto a smooth substrate (silicon or gold). The TMSC is then hydrolysed to cellulose by a vapour phase acid hydrolysis which leaves the smooth morphology from spin coating intact. This method is fast and reproducible, resulting in ca. 20 nm closed films of cellulose with a roughness variation of 2-3 nm. The chemical properties of the films were analysed by XPS and ATR-IR. The morphology was investigated by Atomic Force Microscopy (AFM). As the concentration of the spin coating solution was decreased by a factor of 500 to that used in Chapter 5, the cellulose formed open films upon its hydrolysis from TMSC (Chapter 6). In these open films, the cellulose formed nanosized structures (100x50x1 nm) on a flat silicon substrate. A method for determining the volume of the cellulose patches by AFM data was established. Open films and the volume analysis were tested on a wetting/drying experiment which suggested supramolecular rearrangements of cellulose during drying - a link to the recycling application. A truly novel morphology in cellulose model surfaces was established in Chapter 7. TMSC was blended with polystyrene and the blend was spin coated, resulting in laterally segregated films. The TMSC was then hydrolysed to cellulose and the polystyrene removed by a hydrophobic solvent, leaving conspicuous cellulose domains on a sublayer (3-7 nm) of cellulose. The cellulose domains were micrometers in lateral dimensions, 10-15 nm in height. The cellulose on cellulose films overcome the traditional limitation of organic model surfaces on inorganic substrates: the incompatibility of the two, which may result in difficulties upon treatments of the model surfaces. The thesis provides insight to the difficulties of drawing reliable conclusions on samples from natural fibres. It also illustrates the challenges of preparing cellulose model surfaces. These challenges were the main reason why the latter part turned out to be methodology oriented. The thesis establishes two completely new cellulose model surfaces and provides a firm basis for their applications to come.