Selective Mineralisation by Hydrothermal Methods

Abstract

The use of hydrothermal synthesis has led to the control of the size and shape of various materials of inorganic nature. It has also been highlighted as a green, cheap, and reliable synthetic method, by using water in superheated conditions as the solvent. Mineralisation processes in nature can be easily replicated via hydrothermal synthesis. Such natural hydrothermal mineralisation processes often involve the reaction of metals dissolved or partially dissolved in running water with the components of the ground (silicates, carbonates, phosphates, ...) to form minerals. These processes frequently occur in hydrothermal vents, but are also associated with biological processes (carbonate and phosphates) or might occur in natural repositories of Spent Nuclear Fuel (SNF), as for the silicates. The primary aim of this thesis is to examine the scope of hydrothermal methods for the controlled synthesis of important mineral species. To this end, three mineral classes have been targeted: silicates, phosphates, and carbonates. In particular the potential of the newly developed hydrothermal injection reactor will be assessed in detail. A secondary aim of this work is to establish the possibilities offered by the statistical method of Principal Component Analysis in facilitating the mapping of synthetic landscapes which is a necessary facet of both process optimisation and materials discovery. With regards to the silicates, the natural existence and formation of coffinite (uranium silicate, USiO4), reported for decades as occurring in geological environments has, to the best of our knowledge, never been synthesised pure in situ in the lab. Given its relevance in natural systems, and importantly, its potential implications for the storage of spent nuclear fuels, the synthesis of pure coffinite is enormously important in order to understand its role in these systems. The overarching aim of this part of the project is thus to simulate how nature yields this material by an easy mild-hydrothermal process. Different conditions must be tested as the synthetic stability of this material is fairly narrow. Due to the low availability of uranium to perform these reactions, various silicates will be attempted firstly, as non-radioactive analogues. Zircon (ZrSiO4) and stetindite (CeSiO4) have been reported to have a zircon-type structure, as coffinite. This means that the atoms arrange periodically to form a lattice in a similar way. Additionally, cerium has been proposed to have chemical similarities to Pu making it a great candidate to test the optimum conditions for the synthesis of other metal silicates. Zirconium, on the other hand, has been reported to yield mixed uranium silicates known as Chernobylite due to its appearance at chernobyl?s nuclear plant residual ?lava?. The presence of uranium in this structure supports zircon to be a promising silicate to store uranium waste in long-term repositories as a vitrification material. Therefore, once the optimum synthetic conditions are known for these materials, they may prove to be transferrable for coffinite synthesis. Thirdly, thorite (ThSiO4) is known to be the chemically closest silicate to coffinite. The optimum conditions obtained from the previous surrogates will be tested and optimised by the formation of this mineral. Once the conditions required for the synthesis of the pure form of three different structurally similar silicates and by reducing the synthetic parameters required for the stability of these materials the synthesis of coffinite is suggested to be achievable. The optimisation of the parameters required to obtain a certain material, as for the coffinite synthesis, is time consuming and large numbers of samples are usually required prior to grasping the most effective synthetic protocol. Nonetheless, the improvement of the mineralisation techniques accompanied by the use of statistical approaches to reduce the complexity of the data has led to an easy and fast mapping of the relationship between synthetic condition parameters and compound selectivity. For that reason, Principal Component Analysis (PCA) is proposed as a great candidate for the identification of the main variables that play a role in the mineralisation, reducing the dimensionality of the data to the simplest possible and only expressing the main ones when being correlated. To test the applicability of the PCA approach, two minerals have been aimed to be subjected to this analysis: calcium orthophosphates and calcium carbonate. They are an example of non-classic nucleation materials and show a great complexity of composition and polymorph selectivity closely related to the synthetic conditions (temperature, pH, pressure, etc.). Firstly, calcium orthophosphates are known to be a family of materials in which the crystallisation of one phosphate or another will be related to the conditions of synthesis. In particular, three calcium phosphates will be studied: brushite, monetite and hydroxyapatite. Monetite is known to be the dehydrated version of brushite. On the other hand, hydroxyapatite is reported to have a high biological significance. The selective formation of the three compounds will be tested at different temperatures, phosphate precursor, Ca/P precursor ratio and pH. The presence of the different phosphate phases will be estimated by XRD and FTIR. Thereafter, in order to plot the relationship of conditions and compound formed, the XRD patterns will be subjected to Principal Component Analysis. The most optimised parameters for the selective synthesis of each phosphate will be represented by this analysis. For the sake of controlling the different synthetic conditions effectively and processing a high number of samples in the most efficient way, the injection hydrothermal reactor will be employed. By the in situ tracking of the temperature of mineralisation, the specific parameters at which each compound is formed will be known. On the other hand, calcium carbonate, is known to be essential for different processes for life. The main difference with the phosphates is that although only calcium carbonate is yielded at different synthetic conditions, it exhibits different atomic arrangement in the structure when changing the external conditions. These differentiator atomic dispositions are known as polymorphs of a compound. Calcium carbonate is known to have three anhydrous polymorphs (calcite, aragonite and vaterite) with a different stability energy and properties. Calcium carbonate has been reported to follow a non-classical pathway, in which prenucleation clusters will lead to an amorphous phase that will influence the polymorphic selection. The conditions of mineralisation will drastically determine the polymorphic content in the samples. The way to recognise the presence of the different polymorphs is by Rietveld refinement of the structure against the XRD patterns. The presence of the different polymorphs will be corroborated by FTIR and Raman spectroscopy. The mapping of the conditions for the synthesis of each polymorph is aimed to be performed in a simple way by tuning the synthesis conditions in the hydrothermal injection reactor. A full tracking of the different precursors employed, their concentration and temperature of synthesis, among others, will be explored in this project. Nonetheless, due to the number of variables considered, the FTIR, Raman and XRD patterns will be subjected to PCA. It will correlate the polymorph present in the sample and the conditions of formation. As well, the effectiveness of this statistical analysis for the different analysis techniques will be tested aiming to point out which one is more effective to show the correlation of the polymorphs. A full mapping of the optimal conditions for the phosphates and the polymorphs of calcium carbonate is targeted. As well, it is expected to show the capability of the reactor to control the properties of the material by an easy and fast synthesis of the compounds at different conditions. Finally, we aim to show how PCA is capable of showing the optimal synthetic conditions to obtain from different compounds, to more essential, such as the atomic arrangement expressed as polymorphs.