Aluminosilicate inorganic polymers (AIPs) are network heteropolymers consisting of Si04 and AlO4 tetrahedra linked by a shared oxygen. The use of these materials as a cementing agent, toxic waste storage and fibre reinforced material, amongst a multitude of prospective applications, has grown in recent years. The utilisation of AIPs is hampered by a lack of knowledge about their formation and structure. In order to allow the materials to achieve their full potential, the way in which the material behaves and forms under different conditions must be elucidated. The basic questions that this study aimed to answer were: 1) How does the structure of these AIPs change with composition? and 2) Can this change in structure explain the material properties of the AIP? The AIPs investigated in the study covered the molar composition ranges Si:Al ratio = 1 - 3 and Na:Al ratio = 0.5 - 2. They were made by the sodium hydroxide activation of metakaolinite, derived from the dehydroxylation of kaolinite. The Si content of the AIP was altered by the addition of amorphous silica fume via the activation solution. The study considered the structural nature of the AIPs at the macro, micro and nanoscales, and found that the structure changed at all scales and with all compositions. The nature of the AIP structure was studied at the macroscale utilising compressive strength testing. The results from this work showed that the compressive strength of the AIPs varied systematically with the chemical composition. The strengths recorded ranged from 0.4 ± 0.2 MPa for a sample with Si:Al:Na molar ratios = 1.08:1:0.5, to 64 ± 3 MPa for a sample with Si:Al:Na molar ratios = 2.5:1:1.3. The higher strengths measured exceed those exhibited by Portland cement pastes. The microstructure of the AIPs was investigated by scanning electron microscopy and energy dispersive spectroscopy. / Microscopy showed that the microstructure variations correlated with the compressive strength. In general, AIPs with low compressive strengths exhibited an inhomogeneous two-phase microstructure; grain and matrix. The grain phase consisted of undissolved metakaolinite, whilst the matrix was the fully formed inorganic polymer. AIPs with high compressive strengths exhibited a microstructure that was more homogeneous than the samples with low compressive strength. The compressive strength of the AIPs depended on both the chemical composition and the level of residual MK present in the microstructure. EDS microanalysis showed that the composition of the two phases was significantly different, and that the differences depended on the overall composition of the AIP. EDS results also demonstrated that the impurity elements present in the metakaolinite were affected by the polymerisation process. Soluble elements such as Ca and Mg were found primarily in the matrix, indicating that they had leached out of the metakaolinite grains, whereas insoluble elements such as Fe and Ti were found primarily in the grains. The nanoscale structure of the AIPs was examined by solid-state nuclear magnetic resonance (NMR) and x-ray scattering (XRS). The NMR measurements revealed that the average coordination of Si varied according to the composition of the AIP, whereas the coordination of Al was constant. Na is present in the network in both hydrated and non-hydrated forms. It is postulated that the variation in the Si coordination can be explained by the formation of Si-O-Na bonds with Na forming an ionic bond with 0 in the polymer network. Radial distribution function (RDF) analysis of the XRS patterns revealed little difference in the structure of the different AIPs beyond ~2.5 Å. / Unfortunately, the data were of insufficient resolution to allow for a full evaluation of the differences in the Si-O and Al-O bonds between different AIPs. However, the trends present in the shape and position of the RDF peak corresponding to the Si-O and Al-O bonds do follow the composition of the AIP. It has been shown that a variety of experimental techniques can be used in concert to obtain information on the structural nature of AIPs. To this end, it has been found that the compressive strength of AIPs can be optimised, and that the microstructure of the AIPs changes systematically with variations in the compressive strength. An improved model for the structure of AIPs has also been proposed.
| Identifer | oai:union.ndltd.org:ADTP/222842 |
| Date | January 2004 |
| Creators | Rowles, Matthew Ryan |
| Publisher | Curtin University of Technology, Department of Applied Physics. |
| Source Sets | Australiasian Digital Theses Program |
| Language | English |
| Detected Language | English |
| Rights | unrestricted |
Page generated in 0.002 seconds