Synthese und Charakterisierung von Geopolymeren für die Entsorgung der Spaltprodukte $^{137}$Cs und $^{90}$Sr

Since the 1990s, geopolymers have been investigated as potential substitutes for classical cement materials for certain issues in nuclear waste management. They are intended to overcome limitations arising, for example, from the structurally relevant role of water in classic cements and its radiolysis risk, since water is not an integral component of the geopolymer structure according to the current state of knowledge. In addition, previous investigations of the leaching resistance of various geopolymers with regard to Cs and Sr proved promising. In order to guarantee consistently good properties of the geopolymers in the case of their use as nuclear disposal matrix and to determine possible influencing factors (e.g. chemical composition, temperature, time), comprehensive knowledge of the geopolymer structure is required. However, due to the predominant amorphous nature of aluminosilicate geopolymers, structure elucidation is only possible within certain limits, so that the geopolymer structure is only incompletely known. For this reason, the focus of this work was the structural characterization of various metakaolin-based and thus aluminosilicate geopolymers using microscopic, X-ray diffractometric and various spectroscopic methods. In particular, the integration of Cs$^{+}$ and Sr$^{2+}$ ions of non-radioactive isotopes as surrogates of the fission products $^{137}$Cs and $^{90}$Sr as well as the structural differences between these and other cations containing geopolymers were investigated. It was also determined how variations of the reactants and synthesis parameters influence the reaction product. The main aim was a deeper understanding of the geopolymer structure ranging from the molecular building blocks and their interconnections to the microstructure. Another issue was to gain new insights into cation integration and the role of water and hydroxyl groups in geopolymers. In addition, structural differences between various metakaolins were identified. According to the investigation results presented here, the geopolymer structure appears more complex and diverse than often assumed. For example, different microstructures have been observed, which on the one hand depend on the choice of starting materials (metakaolin and silicate source) and on the other hand on whether the outer surface or the inside (fracture surfaces) of a geopolymer sample was observed. With regard to the short-range order, several geopolymers not only showed Al in fourfold coordination by O (short: Al(IV)) but also Al(V) and Al(VI). Although these species are usually attributed to an incomplete geopolymerization reaction and thus to residues of the starting material metakaolin, other investigation results do not confirm this assumption here. For example, Raman or IR signals of AlO$_6$ octahedra that are found in the spectra of kaolinite were generally not observed for the geopolymers examined here. Irrespective of this, the Raman spectra of those geopolymers without interfering anatase admixtures from the metakaolin revealed differences whose structural origins, however, could not be satisfactorily clarified yet. In addition to first Raman spectra of metakaolin-based geopolymers in general, the present work provides for the first time a direct PDF, Raman, IR and MAS NMR comparison of geopolymers with four different alkali metals and two alkaline earth metals within an in other respects constant system. Samples of comparable stoichiometry and age were examined, which were prepared with the same educts (except for the alkaline and alkaline earth source) and under identical synthesis conditions. Measurement conditions and the procedure for evaluating the data were also constant. The majority of geopolymers cured at room temperature showed an amorphous structure, the order of which was limited to less than 10Å and thus to the region of short-range order. In contrast, the geopolymers containing Sr revealed crystalline SrCO$_3$ as a secondary phase. Thermal treatments in most cases caused a crystallization of the material to pollucite and leucite as well as a feldspathoid (in smaller quantities). Since the formation of pollucite is already observed at temperatures < 100 °C, metakaolin-based Cs geopolymers are apparently also suitable as low-temperature precursors for a ceramic nuclear waste form for Cs (a pollucite ceramic). The obvious structural relationship between the amorphous geopolymers and the crystal structures of pollucite and leucite was utilized by fitting the pair distribution functions of the amorphous geopolymers cured at room temperature based on the crystal structure model for tetragonal leucite. Subsequent analyses of the partial pair distribution functions indicated narrowed channels in the aluminosilicate framework and shortened Cs–O distances compared to the mean value of typical other inorganic Cs-compounds. This may also explain the generally low leaching rates of Cs from geopolymers. Evidence of hydration of the alkali cations in geopolymers was only found for Na. Nevertheless, all geopolymers cured at room temperature seemed to contain certain amounts of water and hydroxyl groups. Although numerous new insights into the structure of geopolymers have been gained through the present work, the suitability of geopolymers as nuclear waste form cannot be assessed conclusively. Especially microstructural, but also spectroscopic differences between single or groups of geopolymers were found. Variations in the choice of starting materials (metakaolin, silicate source and cation) led to partly comparable, partly also very different results. To what extent the obvious variability of the geopolymer structure has an influence on properties important in nuclear waste management, such as the leaching resistance with regard to the radionuclides $^{137}$Cs and $^{90}$Sr, and how adverse influences can be minimized, has to be clarified in future investigations. In this context, also possibilities arising from the use of Cs geopolymers as pollucite precursors should be examined