This Ph.D. thesis addresses the recrystallization of barite, BaSO4(s), in carbonate bearing aqueous solution into witherite, BaCO3(s), and the influence of carbonate on radium incorporation into both minerals. The uptake of Ra by barite has been investigated for many decades in the context of various environmental and industrial settings, such as Ra retention in nuclear waste repositories, Ra accumulation due to scaling processes in geothermal energy plants and in pipelines of petroleum production fields, removal of Ra from brackish groundwater in desalination plants, as well as other settings of Naturally Occurring Radioactive Materials.
Radium uptake by barite occurs when dissolved Ra2+(aq) cations react with barite, leading to Ra retention by formation of a (Ba,Ra)SO4 solid-solution. In studies under ambient geochemical conditions, it has been shown that barite crystals can recrystallize and spontaneously adapt their composition to the solution conditions through dissolution and re-precipitation processes, and incorporate Ra and other trace elements into the crystal structure of the barite host mineral. The database for describing (Ra,Ba)SO4(s) mixture thermodynamics has improved considerably in recent years. ... mehrIn addition, quantum chemical methods for determining the equilibrium position of Ra in such solid solutions have become established. However, experimental work also shows that equilibration times can vary between one year and extrapolated time spans of about ten thousands of years depending on the specific sample characteristics and pre-treatment of the initial barite mineral. This demonstrates that a reliable consideration of the Ra immobilization potential of a solid solution requires not only a sound thermodynamic description but also a fundamental quantitative understanding of the kinetics of the reactions involved.
In contrast to the knowledge on Ra uptake by barite via formation of a (Ra,Ba)SO4(s) solid solution, little is known about the uptake of Ra by witherite. The starting hypothesis for this work was that the reaction of barite with dissolved CO32-(aq) anions at elevated pH can lead to the recrystallization of barite into witherite via dissolution and consecutive precipitation processes, resulting in the formation of a (Ra,Ba)CO3(s) solid solution. Besides a quantitative description of the (Ra,Ba)CO3(s) mixture thermodynamics, the kinetics of the potential (Ra,Ba)SO4(s) to (Ra,Ba)CO3(s) recrystallization process is of interest. The presence of carbonate likely alters the chemical behavior of barite surfaces, via surface mixing or by witherite layer formation through dissolution-precipitation which in turn has an effect on the uptake process.
The final aim of this work is to explore the fate of Ra bearing barite during the recrystallization process of barite in the presence of carbonate. By integrating experimental, analytical and computational approaches, the influence of carbonate concentration in solution, thereby the influence of degree of oversaturation, and the influence of barite mineral properties on the recrystallization of BaSO4(s) to BaCO3(s) is intensively studied. Based on the achieved knowledge about the recrystallization of the barite to witherite, incorporation of radium into the two host minerals is investigated. Since strontium occurs as trace element in natural (Sr,Ba)SO4(s) and (Sr,Ba)CO3(s) solid solutions, the fate of Sr during the recrystallisation of a natural Sr-bearing barite to witherite is studied as an analogy for the fate of Ra to provide further insights into the transformation of (Ra,Ba)SO4(s) to (Ra,Ba)CO3(s) in the presence of carbonate.
The influence of carbonate on the Ra uptake by barite and witherite is studied in batch type recrystallisation experiments with large barite cubes and microcrystalline barite powders as well as in coprecipitation batch type coprecipitation experiments with Ba2+(aq) + SO42-(aq) + Ra2+(aq) and Ba2+(aq) + CO32-(aq) + Ra2+(aq) bearing solutions, respectively. In the recrystallisation experiments, coarse grained natural barite samples from Androvo (Bulgaria) and Iberg (Germany), freshly precipitated barite and commercial synthetic high purity barite powder (Sachtleben Chemie GmbH, Germany) are used as starting materials. Besides ultra-pure Sachtleben barite powder used in barite and witherite recrystallization experiments, powder samples of Sachtleben barite, which had been equilibrated with 226Ra2+(aq) bearing and carbonate-free solutions for seven years is used in (Ra,Ba)SO4(s) to (Ra,Ba)CO3(s) recrystallization experiments.
Different analytical and spectroscopic methods like scanning electron microscopy and energy / wavelength dispersive X-ray spectroscopy (SEM-EDS / WDS), Raman spectroscopy, X-ray photoelectron spectroscopy (XPS), X-ray powder diffraction (XRD) as well as inductively coupled plasma-optical emission spectrometry (ICP-OES) and -spectroscopy are applied for the analysis of the composition of crystals and solutions studied in the batch experiments. A SEM equipped with a focused ion beam (FIB) cutting device is used for removing ultra-thin layers from surfaces of selected samples to reveal interior structures of altered barite. Using a combined FIB-SEM approach, precise imaging with nanometer resolution and simultaneous chemical analysis of the barite/witherite reaction fronts is achieved. In addition to the analytical and experimental methods, numerical simulations based on the Density Functional Theory (DFT) is used as a quantum-mechanical atomistic simulation tool to obtain electronic energies and atomic structures of the barite and witherite host minerals and the solid solutions (Ra,Ba)SO4(s) and (Ra,Ba)CO3(s). In case of the solid solution simulations, DFT-based electronic energies are employed in the Single Defect Method (SDM) approach to calculate the extent of solid solution non-ideality. Moreover geochemical modelling by means of the PHREEQC software package is used to calculate saturation levels, solid solutions mixing, precipitation of solid phases and diffusion processes at the barite / witherite reaction front.
In series of recrystallisation experiments under various temperatures, degrees of carbonate concentration and pH in solution it is observed that recrystallization of barite to witherite is a rather slow process in case of macroscopic single crystal cubes of Androvo and Iberg barite. It turned out that only under quite extreme conditions (60°C, 0.1 M Na2CO3, pH 11) the replacement (coupled recrystallization) of barite into witherite takes place to a measurable extent within the studied period of five weeks. Microscopic analyses of the initial and replaced barite cubes demonstrate that for the progress of the reaction, in particular the development of the porosity of the growing witherite layer is of importance. With increasing reaction time, the dissolution of barite and the formation of witherite slows down. The experimental results indicate that passivation of the surface by the growing witherite is to be expected over longer periods of time. Compared to results of PHREEQC calculations for full equilibration of the solid-solution system, the theoretical possible formation of approximately 40 wt.% witherite is not achieved or is only achieved over extremely long periods of time even in replacement experiments with barite cubes in 0.1 M Na2CO3 at pH 11 and 60°C. In crystal cubes of pure Androvo barite, a sharp interface between barite and witherite, associated with a sharp decrease in sulfur and an increase in carbon is observed. According to the thermodynamically expected distribution coefficients in (Sr,Ba)SO4(s) to (Sr,Ba)CO3(s) solid solutions, a significant change in the Sr/Ba ratio from the initial Sr-bearing Iberg barite to precipitated witherite would be expected. However, in experiments with Iberg barite the Sr/Ba ratio across the interface between barite and witherite does not change significantly. This indicates that at the barite-witherite interface where the replacement reaction proceeds, the cations released by barite dissolution are incorporated into witherite as they come, and that the thermodynamic affinities for incorporation into witherite under these transport-controlled conditions play no / or only a very minor role.
The initial reaction progress in recrystallization experiments with microcrystalline powder samples of natural Androvo barite, natural Iberg barite and of synthetic Sachtleben barite appears to be similar to that of the macroscopic Androvo and Iberg barite single crystal cubes. After a fast start, the transformation (uncoupled recrystallization) slows down and does not reach the theoretically achievable limit as calculated by means of PHREEQC for full equilibration of the solid-solution system. Still, the transformation of barite to witherite is considerably faster in the experiments with Androvo and Iberg barite powders compared to those with the respective single crystal cubes. Sachtleben and Iberg barite powders react at a closer rate. Similar to the kinetic trends in the replacement experiments with macroscopic single crystal cubes, the Androvo samples, which are ground from macroscopic single crystals, also shows in the powder transformation experiments a significantly lower reactivity compared to the reactivity of the Iberg barite powder. Experiments with Androvo and Iberg barite powder samples show that powders are by no means just small single crystals, but that an enormously increased complexity of the processes taking place can be observed here due to shifts in the reaction rates. SEM-EDX investigations demonstrate that in the powders, surface crystal rebuilding / transformation at a reactive barite-witherite interface plays a subordinate role. Instead, barite dissolves, while witherite crystals often form in idiomorphic shapes in various spatial arrangements with respect to the initial barite. The ratios between barite dissolution rate and witherite growth rate are decisive for this difference in the process sequence. If the barite dissolution is the slowest - rate-controlling process, a barite-witherite interface is formed, as in the single crystal experiments. If, on the other hand, witherite growth is slower than barite dissolution, and therefore rate-controlling, the witherite crystals grow increasingly independently of the initial barite.
In order to show what miscibility can be expected for radium incorporation in barite and witherite, both theoretical calculations using DFT and co-precipitation experiments are carried out. The DFT calculation results show that the incorporation of Ra in barite and witherite is almost ideal with rather small Guggenheim parameters ("non-ideality parameter") of 0.84 and 0.58, respectively. The value for Ra in barite agrees very well with calculated results of Vinograd et al., 2013. Using published solubility products of BaSO4(s) and RaSO4(s) at 60°C, a theoretical distribution coefficient of Dtheo = 0.42 is calculated. Coprecipitation experiments with Ba2+(aq) + SO42-(aq) + Ra2+(aq) at 60° result in a distribution coefficient of Dexp = 0.34 ± 0.14 which is also in excellent agreement with the thermodynamically predicted value for (Ra,Ba)SO4(s). Co-precipitation experiments for Ra-incorporation into witherite over a wide radium concentration range yield a distribution coefficient of Dexp = 0.15 ± 0.05 for (Ra,Ba)CO3(s). Only few literature data is available for the Ra-witherite system. An existing partition coefficient (D = 0.13 ± 0.07 of Yoshida et al., 2015) agrees exactly with the measured (Ra,Ba)CO3(s) composition. Together with the Guggenheim parameter of 0.58 for (Ra,Ba)CO3(s) calculated in this Ph.D. work, this results in partition coefficients of Dtheo = 0.06 , which closer to the lower limit of the value derived from the coprecipitation experiments with Ba2+(aq) + CO32-(aq) + Ra2+(aq), i.e. Dexp = 0.15 ± 0.05. The attempt to simulate the incorporation of sulfate into witherite by means of DFT calculation leads to a highly distorted structure indicating an extreme non-ideality of the solid solution. This corresponds to the experimental findings that sulfate could never be detected in witherite from the recrystallization experiments. Experiments on the recrystallization of Sachtleben barite powders, which had been equilibrated with 226Ra2+(aq) bearing and carbonate-free solutions for seven years, to Ra-bearing witherite shows that the various Ra-bearing barites investigated were quite inert. Interestingly, these long-time equilibrated (Ra,Ba)SO4(s) powders are less reactive compared to a Ra-free Sachtleben barite powder, which had been equilibrated in parallel with 133Ba2+(aq) bearing and carbonate-free solutions for seven years. Based on this observation, it is assumed that the Ra content in the barite host mineral actually has an inhibiting influence on the reactivity. Nevertheless, formation of microscopic (Ra,Ba)CO3(s) crystals is observed. But these do not contain enough Ra for a quantitative evaluation of Ra incorporation after recrystallization.
The comprehensive work undertaken within this Ph.D. represents an important step forward in the scientific understanding of the knowledge of radium geochemistry and contributes to a broader understanding of radium behavior in both natural and anthropogenic contexts. In addition to its contribution to the understanding of the fate of radium, the findings on barite recrystallization into witherite contributes to knowledge of the broader field of mineral dissolution and precipitation processes.
