Hot New Papers from LES

The new CASH+ core sublattice solid solution model of calcium-silicate-hydrate (C-S-H) can describe calcium uptake, solubility, water content, and mean silicate chain length up to 90 °C. In this study, the model has been consistently extended to describe the equilibrium uptake of alkalis (Li, Na, K, Rb, Cs) and alkaline earths (Mg, Sr, Ba, Ra). The new endmembers for each cation were defined, and their properties, along with binary interaction parameters, were fitted against known aqueous and solid phase compositions. The CASH+ model with its extensions can be directly used in GEMS codes or discretized and exported for use with other geochemical speciation computer programs. In agreement with the experimental data, the model indicates that the uptake of alkalis and alkaline earth cations occurs mainly as a competitive (interlayer) ion exchange, uptake is favored at low C/S ratios, silicate chains get shorter at higher pH, and the uptake of bivalent cations (in the order Mg²⁺ < Ca²⁺ < Sr²⁺ < Ba²⁺ < Ra²⁺) is preferential over that of monovalent cations (Li⁺ ≈ Na⁺ < K⁺ ≈ Rb⁺ ≈ Cs⁺).

A new thermodynamic model, CASH+, is proposed, aimed at accurately describing equilibrium composition, stability, solubility, and density of C-S-H gel-like phases at varying chemical conditions. Taking advantage from recent atomistic and spectroscopic studies, this sublattice solid solution model allows incremental extensions to accommodate alkali, aluminum and other cations. This incrementality, achieved first time for a C-S-H solid solution model, means that all thermodynamic properties of endmembers and interaction parameters can be kept fixed in further extensions. This paper describes principles of how endmembers of CASH+ solid solution model can be constructed by permutating moieties assigned to different sublattices, and how the structural consistency of the model can be established. Initial standard thermodynamic properties of endmembers were estimated using predictive methods and PSI/Nagra and Cemdata18 chemical thermodynamic databases. The parameterized core CASH+ sub-model in Ca-Si-H₂O system is shown to perform well in presence of liquid water at temperatures up to 90 °C.

Surface diffusion may enhance the diffusive fluxes of cationic species in argillaceous materials and lead to an underestimation of radioactive doses emanating from deep geological repositories for radioactive waste to the biosphere. The conceptual understanding of surface diffusion implies that competition effects between different cations would also affect diffusion and not only sorption. The present work provides for the first time diffusion data and simulations supporting this view. The effect of Ca²⁺ on the diffusion and sorption of a ⁵⁷Co²⁺ tracer was measured in compacted illite. The results could be quantitatively explained using an electrical double layer or a cation-exchange model for the uptake of cationic species, with both models involving mobile species at the planar surfaces. The observed effects were related to the presence of different sites for Co²⁺ uptake, exhibiting different surface mobilities and different affinities for Ca²⁺. The strong sorption sites, which dominate sorption at the conditions of the experiments, do not exhibit noticeable surface mobility and are not prone to competition with Ca²⁺. The effective diffusion coefficients are rather dominated by the cation species at the planar surfaces, which are affected by competition with Ca²⁺. The relevance of the illite model system for intact clay rock was demonstrated by a successful application of the illite source data to ⁵⁷Co²⁺ diffusion measurements in Opalinus Clay, in which the equilibrium pore water contained substantial amounts of alkaline earth cations. Our present findings enhance the conceptual understanding of sorption and diffusion of trace elements in compacted clay minerals and clay rocks.

Thermodynamic equilibrium calculations for cementitious materials enable predictions of stable phases and solution composition. In the last two decades, thermodynamic modelling has been increasingly used to understand the impact of factors such as cement composition, hydration, leaching, or temperature on the phases and properties of a hydrated cementitious system. General thermodynamic modelling codes such as GEM-Selektor have versatile but complex user interfaces requiring a considerable learning and training time. Hence there is a need for a dedicated tool, easy to learn and to use, with little to no maintenance efforts. CemGEMS (https://cemgems.app) is a free-to-use web app developed to meet this need, i.e. to assist cement chemists, students and industrial engineers in easily performing and visualizing thermodynamic simulations of hydration of cementitious materials at temperatures 0-99 °C and pressures 1-100 bar. At the server side, CemGEMS runs the GEMS code (https://gems.web.psi.ch) using the PSI/Nagra and Cemdata18 chemical thermodynamic data-bases (https://www.empa.ch/cemdata).

The present paper summarizes the concepts of CemGEMS and its template data, highlights unique features of value for cement chemists that are not available in other tools, presents several calculated examples related to hydration and durability of cementitious materials, and compares the results with thermodynamic modelling using the desktop GEM-Selektor code.

Two approaches to simulations of phonon properties of solids beyond the harmonic approximation, the self-consistent ab initio lattice dynamics (SCAILD) and decoupled anharmonic mode approximation (DAMA) are critically benchmarked against each other and molecular dynamics simulations using a density-functional-theory description of electronic states, and compared to experimental data for fcc aluminium. The temperature-dependence of phonon dispersion and the phonon density-of-states, heat capacity, and the mean atomic displacement for fcc aluminium are examined with these approaches at ambient pressure. A comparison of results obtained with the harmonic approximation to the ones predicted by SCAILD and DAMA reveal a negligible anharmonic contribution to phonon frequencies, a small, but significant influence on heat capacity, and a strong effect on atomic mean-square displacement. The phase space accessed with SCAILD and DAMA is reduced relative to molecular and harmonic lattice dynamics simulations. In particular the DAMA results are in good agreement with displacement amplitudes determined by the Debye–Waller factor in x-ray diffraction experiments.

Small-pore zeolites are successfully employed as catalysts, sorbents and molecular sieves. Their physiochemical properties can be tuned by modifying their extraframework cation (EF) composition via ion exchange. In this study, we investigate the crystal structure of a Cd-exchanged levyne (LEV) intergrown with erionite (ERI) by combining Single Crystal X-ray Diffraction (SCXRD), Molecular Dynamic simulations (MD) and Extended X-ray Absorption Fine-Structure spectroscopy (EXAFS). Data obtained from the different techniques consistently indicate that Cd²⁺ in LEV is arranged in a nearly ordered fashion. In contrast, strong disorder of the EF species (Cd²⁺ and H₂O) is observed in the ERI cavities. Here, Cd²⁺ forms aqueous complexes that are more mobile in comparison to LEV, where Cd²⁺ bonds to both H₂O and framework-oxygen atoms. The formation of Cd-clusters is excluded based on EXAFS analysis. Finally, to discriminate between thermal and static disorder, we propose a new approach based on combined MD and geometry optimization.

Our new, structurally consistent CASH+ sublattice solid solution model of calcium-silicate-hydrate (C-SH) has been successfully parameterized for the uptake of alkalis and aluminium. The model, developed by considering recent structural and atomistic views on C-S-H, can describe cation uptake, solubility, water content, and mean silicate chain length (MCL). End members for each cation have been defined by permutation of chemical moieties existing in two sublattices, and their properties along with the site interaction parameters were incrementally fitted against aqueous and solid phase composition and NMR-derived MCL data for a wide range of conditions (also considering incongruent phenomena). Good fits show that the model is incremental and can be extended stepwise (i.e. keeping previously fitted parameters fixed). Preliminary results show that the CASH+ model can be further extended for the uptake of iron, heavy metals, actinides, and fission products used in a broad range of cement chemistry and waste geochemistry applications.

The thermodynamic properties of carbonate minerals suggest a possibility for the use of the abundant materials (e.g., magnesite) for removing harmful divalent heavy metals (e.g., Pb²⁺). Despite the favourable thermodynamic condition for such transformation, batch experiments performed in this work indicate that the kinetic of the magnesite dissolution at room temperature is very slow. Another set of co-precipitation experiments from homogenous solution in the Mg-Pb^II-CO₂-H₂O system reveal that the solids formed can be grouped into two categories depending on the Pb/Mg ratio. The atomic ratio Pb/Mg is about 1 and 10 in the Mg-rich and Pb-rich phases, respectively. Both phases show a significant enrichment in Pb if compared with the initial stoichiometry of the aqueous solutions (Pb/Mg initial = 1 × 10 ^{− 2}–1 × 10⁻⁴). Finally, the growth of {10.4} magnesite surfaces in the absence and in the presence of Pb²⁺ was studied by in situ atomic force microscopy (AFM) measurements. In the presence of the foreign ion, a ten-fold increase in the spreading rate of the obtuse steps was observed. Further, the effect of solution ageing was also tested. We observed the nucleation of a secondary phase that quickly grows on the {10.4} surfaces of magnesite. The preferential incorporation of Pb²⁺ into the solid phase observed during precipitation and the catalytic effect of Pb²⁺ on magnesite growth are promising results for the development of environmental remediation processes. These processes, different from the transformation of magnesite into cerussite, are not limited by the slow dissolution rate of magnesite. Precipitation and growth require an external carbon source, thus they could be combined with carbon sequestration techniques.

The ASR products in concrete have various chemical compositions. It is yet unclear whether and how these products develop micro-expansion upon moisture ingress. This paper presents a 3D in-situ observation of the crystallography and volume change of an ASR-product-filled vein under varying relative-humidity (R.H.). The vein was observed to contain two layered nano-crystalline phases with distinct basal spacings, and was distributed heterogeneously in space. When R.H. changed from 10% to >38%, the basal spacing increased from 7.43 Å to 8.89 Å for one phase, whereas it remained constant (~10.9 Å) for the other. This is the first time that an ASR product is observed in-situ to exhibit crystal structural expansion during wetting process. However, the product-filled vein exhibited no noticeable swelling when R.H. varied from 10% to 97%. Our findings provide the first direct evidence that the moisturization-induced crystal structural change of ASR product may not be a plausible explanation to the macroscale concrete expansion.

Alkali-silica reaction (ASR) is one of the most important concrete durability issues worldwide. It was shown recently that ASR products have a structure similar to the natural mineral shlykovite. We carried out grandcanonical Monte Carlo simulations of ions sorption on the surfaces of C-S-H and shlykovite to better understand and compare the sorption of monovalent (K⁺, Cl^-) and divalent (Ca²⁺) ions by ASR and C-S-H at different pH (10.0 to 13.0) and pore water chemistry. Our results indicate that divalent ions tend to overcompensate for the negative surface charge, which leads to a co-adsorption of negative species at the surface of both shlykovite and C-S-H phases. At high K/Ca ratios in the pore solution, monovalent K⁺ will compete with divalent Ca²⁺ ions, resulting in the desorption of adsorbed calcium from the shlykovite surface. For C-S-H, this effect cannot be observed in the considered pH range. C-S-H has a higher Ca selectivity compared to ASR in the entire pH range studied.

Cation diffusion coefficients in clayey materials partly appear to be greater than diffusion coefficients of water tracers. The measured values vary between experiments performed at different salinities or different tracer concentrations. This effect is especially pronounced for cations that sorb strongly on the clay surfaces, such as Cs. The observations illustrate the difficulties in applying Fick’s law to cation diffusion in clays and demonstrate the need to find a consistent description of cation diffusion in clays that can be used to predict experiments performed at different conditions. In order to consistently describe Cs diffusion in Opalinus Clay, a multi-site surface diffusion model was implemented in the continuum-scale reactive transport code Flotran. The model combines pore and surface diffusion in one single diffusion coefficient, which accounts for the diffusion of sorbed cations along the clay surfaces.. The contribution from surface diffusion to the diffusion coefficient is directly coupled to the sorption behavior via the derivative of the sorption isotherm. The model parameters include the surface mobilities, which are specific for each cation and sorption site. To derive surface mobilities for Cs, in-diffusion experiments were conducted at eight different stable Cs background concentrations. A set of surface mobilities for Cs on three sorption sites in Opalinus Clay was estimated by fitting the surface diffusion model simultaneously to these experimental data. Moreover, the sensitivity of the model to sorption parameters and surface mobilities was evaluated. The surface diffusion model with the estimated surface mobilities was then successfully tested against independent experimental data for Cs in Opalinus Clay, illustrating the model’s predictive capabilities.

Fe(II) interaction with cement phases was studied by means of co-precipitation and sorption experiments in combination with X-ray absorption fine structure (XAFS) spectroscopy. Oxidation of Fe(II) was fast in alkaline conditions and therefore, a methodology was developed which allowed Fe(II) to be stabilised in the sorption experiments and to prepare samples for spectroscopy. X-ray diffraction (XRD) of the co-precipitation samples showed uptake of a small portion of Fe(II) by calcium-silicate-hydrates (C-S-H) in the interlayer indicated by an increase in the interlayer spacing. Fe(II) incorporation by AFm phases was not indicated. Wet chemical experiments using ⁵⁵Fe radiotracer revealed linear sorption of Fe(II) irrespective of the Ca/Si ratio of C-S-H and equilibrium pH. The K_d values for Fe(II) sorption on C-S-H are more than three orders of magnitude lower as compared to Fe(III), while they are comparable to those of other bivalent metal cations. XAFS spectroscopy showed Fe(II) binding by C-S-H in an octahedral coordination environment. The large number of neighbouring atoms rules out the formation of a single surface-bound Fe(II) species. Instead the data suggest presence of Fe(II) in a structurally bound entity. The data from XRD and XAFS spectroscopy suggests the presence of both surface- and interlayer-bound Fe(II) species.

Slag-containing pastes and concretes were analysed by element-specific synchrotron-based techniques to determine the speciation of iron on crushed materials through spatially resolved micro-spectroscopic studies. The investigated cement samples were hydrated either in the laboratory, or exposed to river or sea water. Metallic iron, along with minor proportions of iron sulphide and magnetite was detected in the laboratory sample. Iron sulphide, goethite, and siliceous hydrogarnet were discovered in the blended slag cements hydrated in contact with river water for up to 7 years. In contrast, no Fe(0) was observed in blended concretes exposed to sea water. Instead, iron sulphide, iron(II)-hydroxide and -oxide, hematite, magnetite, siliceous hydrogarnet, and goethite were detected as well as ilmenite (FeTiO₃) in the aggregates. The strong acceleration of Fe oxidation in samples exposed to sea water and the long-term passivation observed in the other samples indicate comparable processes as those occurring on steel bars.

Thallium (Tl) is a highly toxic trace metal. It occurs mostly as soluble monovalent Tl(I) and less frequently as poorly soluble trivalent Tl(III). Laboratory studies have shown that vacancy-containing hexagonal birnessites can sorb Tl with a very high affinity via a mechanism that involves the oxidation of Tl(I) to Tl(III) and strong complexation of Tl(III), whereas other manganese (Mn) oxides bind Tl(I) non-oxidatively and with lower sorption affinity. Information on the mode of Tl uptake by natural Mn oxides in soils, on the other hand, is still limited. In this study, we characterized the association of Tl with Mn oxides and Tl (redox) speciation in a naturally Tl-rich soil using micro-focused synchrotron X-ray absorption near edge structure (XANES) spectroscopy and X-ray fluorescence (XRF) chemical imaging. The results show that most soil Tl was Tl(I) associated with micaceous clay minerals in the soil matrix. High levels of Tl in soil Mn concretions, on the other hand, were mostly identified as Tl(III), suggesting that oxidative Tl uptake by vacancy-containing hexagonal birnessite was the main process of Tl accumulation in soil Mn concretions. The spectroscopic results in combination with chemical extractions and published sorption isotherms for Tl on synthetic Mn oxides suggest that the formation and transformation of natural Mn oxides in soils and sorption competition of Tl with major and trace metal cations determine the extent and mode of Tl uptake by soil Mn oxides. Methodologically, this study compares classical micro-XRF element mapping combined with point XANES analyses for spatially-resolved element speciation with high-resolution chemical imaging of entire sample areas, which is of great interest for the geochemical community in light of diffraction-limited storage ring upgrades to many synchrotron lightsources.

Cementitious materials in underground constructions are exposed to CO₂ rich ground waters which leads to combined carbonation and calcium leaching. Complex interplay between leaching, carbonation reaction and changes in transport pathways presents difficulty in parameterizing continuum scale models in a consistent way. Therefore, a novel multi-level pore-scale reactive transport model is presented to capture microstructure changes under combined carbonation and leaching. Model explicitly resolves capillary pores and phases with unresolved porosity as a porous media. Governing equations are solved using a lattice Boltzmann method based reactive transport solver with chemical reaction under thermodynamic equilibrium. The two-dimensional parametric study on idealized microstructures revealed that carbon content and pH of boundary solution strongly affects degradation rates, location and thickness of precipitated calcite layer. Furthermore, reactive surface area plays dominant role and tortuosity of media rather a secondary role. The three-dimensional simulations using virtual cement paste microstructure show that degradation rate exhibit non-linear behaviour with square root of time and time. This implies that simple empirical relations for prediction of progression of reaction fronts are not applicable and use of numerical reactive transport models is inevitable.. The developed model qualitatively captures the development of carbonation, leaching fronts and zonation as observed in experiments. Good quantitative agreement between modelling and experiments is obtained for initial stages. At later times, the modelling result and experimental observations diverge significantly. This discrepancy is likely due to lack of consideration of kinetics of C-S-H dissolution and calcite precipitation.

Large-scale ab initio Meta-Dynamics simulations were applied to elucidate the molecular mechanism and reaction free energies of pyrophyllite dissolution at the (1 1 0) edge surface in pure water at close to neutral pH under far from equilibrium conditions. The simulation setup allows realistic representation of the clay mineral surface and explicit consideration of solvent dynamics at finite temperature. The simulation reveals that dissolution of a single tetrahedral or an octahedral unit from the clay mineral edge is a complex multi-step process with several reaction intermediates. Typically, each reaction step changes denticity of the reacting site in a step-by-step manner and leads, eventually, to the leaching of ions forming octahedral and tetrahedral sheets of the phyllosilicate. The solvent rearrangement and the proton transfer reactions in the first and the second coordination shell of the dissolving unit play a critical role in the stabilization of reaction intermediates and the net progress of the dissolution reactions. The overall reaction mechanism can be rationalized as sequence of concurrent and reversible elementary reaction events, which are:

(1) the nucleophilic attack of H₂O molecules or OH groups on the dissolving surface site.
(2) ligand exchange reactions in the first coordination shell of the reacting sites leading to changes of its conformation and denticity at the mineral surface.
(3) collective proton transfer reactions between the acidic and basic oxygen sites mediated via a chain of the hydrogen bonded molecules in the first and second coordination shell of the reacting site.
The results obtained in this study are general and applicable to the group of 2:1 phyllosilicates in a wide range of chemical and thermodynamic conditions.

Most of the available thermodynamic data concerning radioactive waste disposal are restricted to values of reaction equilibrium constants (logK^o₂₉₈) at 25 ^oC and 1 bar. Simple estimation methods such as isocoulombic reactions can be used for extrapolating the properties of reactions involving aqueous species and minerals to elevated temperatures. The aim of this study was to validate the applicability of various alternative isocoulombic reactions to estimate logK^o_T values of aqueouscomplexation reactions for lanthanides and actinides to elevated temperatures while taking advantage of new additional literature data, and to identify criteria for choosing the ‘‘best” reactions. For each chemical species of interest, a systematic approach using dedicated software and database allowed us to identify the isocoulombic reactions and types of extrapolation that yield the best estimates of standard thermodynamic properties at elevated temperatures, when very limited or no experimental data are available. We have tested aqueous complexation reactions for selected lanthanides and actinides of different valences with chloride, fluoride, sulfate, carbonate, nitrate, phosphate and silicate ligands. ‘‘Model” complexation reactions, having known temperature trends, were systematically combined with complex formation reactions of interest whose temperature trends are unknown, into many alternative isocoulombic reactions. For each ion, we investigated which of the generated isocoulombic reactions provide the best estimates for log^eK^o_T of the reaction of interest at elevated temperatures in order to compile the guidelines for choosing the optimal ones, then applying these guidelines to ‘‘prediction” subsets. In most cases, knowing only log^eK^o_T at 25 ^oC (for the reaction of interest), it was possible to obtain rather accurate estimates of log^eK^o_T values at elevated temperatures using isocoulombic reactions that exchange ions with similar charge and hydration properties (hydrated ionic radius and structure of the hydration shell) and known log^mK^o_T of model reactions. These ions and their complexes interact with the solvent in comparable ways, so that their similar heat capacity and entropy effects largely cancel out on both sides of an ‘‘optimal” isocoulombic reaction.

Wet chemistry and spectroscopic investigations were conducted to study Fe(III) uptake by calcium silicate hydrate (C–S–H) at different Ca/Si ratios. Wet chemistry experiments were carried out by using a ⁵⁵Fe radiotracer while ²⁹Si NMR and XAS spectroscopy were performed on C–S–H phases loaded with different Fe(III) concentrations. Sorption kinetics experiments indicate that equilibrium was attained within 30 days. Over the studied concentration range, Fe(III) sorption was linear, irrespective of the difference in pH of the suspension and the Ca/Si of C–S–H. In addition, Fe(III) sorption on C–S–H phases was significantly stronger than Al(III) sorption. The total Fe(III) uptake by C–S–H phases, however, was limited by the lower solubility of Fe(OH)₃ compared to Al(OH)₃; up to 1 mol Fe/kg C–S–H (molar Fe(III)/Si ≈ 0.001) could be taken up. ²⁹Si NMR and EXAFS data suggest Fe(III) uptake in octahedral coordination into the interlayer of the C–S–H phases with Ca/Si ratios 1.2 and 1.5. However, such an uptake mechanism appears unlikely in the case of the C–S–H phase with Ca/Si 0.8 because structural parameters of Fe(III) deduced from EXAFS are different.

The Swiss disposal concept foresees that carbon-14 (¹⁴C) is predominantly released from irradiated steel disposed of in a cement-based repository for low- and intermediate-level radioactive waste. To predict how ¹⁴C migrates in the cementitious environment of the repository near field and subsequently in the host rock, knowledge about the carbon speciation during anoxic steel corrosion in alkaline conditions is therefore essential. To this end, batch-type corrosion experiments with carbon-containing zero-valent iron (ZVI) powders subject to oxidative pre-treatments were carried out in NaOH solution at pH 11 and 12.5. Alkanes and alkenes (C₁–C₇) were identified in the gas phase and produced on the iron surface by a Fischer-Tropsch type mechanism. The kind of oxidative pre-treatment has an effect on the production rate of hydrocarbons (HCs). In the liquid phase, carboxylic acids were identified and produced during the oxidative pre-treatment of the ZVI powders. They are released instantaneously from the oxide layer upon contact with the alkaline solution. The kind of oxidative treatment and the exposure time to oxic conditions directly influence the amount of carboxylic acids accommodated in the oxide layer.

The conceptual setup and the parametrisation of surface diffusion models for different types of cations in negatively charged clay minerals has still not been fully developed. In particular, the contribution of different types of surface-associated cationic species to the overall mass transfer rates is an open question. Further, the transferability of sorption data gained on dispersed clay suspensions to the compacted clay minerals or consolidated clay rocks remains also unanswered. This contribution presents experimental results for the diffusion of ¹³⁴Cs⁺ and ¹⁵²Eu³⁺ in and sorption on compacted illite (bulk-dry densities of 1900 and 1700 kg m⁻³, respectively) and their interpretation using thermodynamic and transport modelling. Different solution parameters, such as pH, ionic strength and the concentration of stable isotope background were varied. The results give information on both diffusional processes of the two radionuclides and their sorption behaviour at relevant solid-to-liquid ratios. The existing two site protolysis non electrostatic surface complexation and cation exchange (2SPNE-SC/CE) model turned out to be fully valid for the description of the equilibrium distribution of the test cations between the solution and the clay phase in the compacted state. The model had to be extended by an electrical double layer description of cationic species bound to the planar surfaces, involving mobile species in the diffuse layer, in order to successfully model the observed diffusion profiles. Despite the notable differences in sorption behaviour of Cs⁺ and Eu³⁺ on illite, their diffusional behaviour could thus be satisfyingly described using the same model approach.

Molecular diffusion is the dominant solute transport process in clays and claystones that are considered as sealing materials in the deep geological disposal of radioactive waste. These materials are typically water saturated, but during construction and later, at elevated temperatures or when gas may be produced, unsaturated conditions prevail. Investigating the clay's water retention properties as well as solute transport under unsaturated conditions is therefore mandatory. Here, functional dependencies of these properties were derived from atomistic and pore-scale simulations. In the absence of tomographic maps that resolve all pores in clays, model clay structure maps with different pore size distributions were generated using a previously developed algorithm. Upscaled water retention functions and upscaled diffusion coefficients of unsaturated samples were derived from these maps based on the shifted Young-Laplace equation that considers film adsorption and capillary condensation. Pore-scale parameters (water film thickness, diffusion coefficients) used for the upscaling were taken from Grand Canonical Monte Carlo (GCMC) and molecular dynamics (MD) simulations, thus connecting molecular and pore-scale simulations. We focused on effects of the pore size distribution and of the adsorbed water film on upscaled parameters. Sample-scale diffusion coefficients were clearly reduced in unsaturated samples compared to the saturated state, with less reduction when including adsorbed water films. The reduction was stronger in samples with a narrow size distribution of the interparticle pores as compared to those with a wide distribution (but equal mean size). The results follow the trends of the experimental data, even though the scale of the simulations is still clearly smaller than that of typical experiments.

Baryte is of interest to nuclear waste disposal as the main scavenger of ²²⁶Ra, a long-lived nuclide playing a major role in the safety assessment of planned disposal sites. In specific repository setups, Ba and Ra released from the nuclear waste will react with sulphate-rich pore water, potentially leading to formation of Ra-bearing baryte. Baryte has a complex kinetic behaviour and its precipitation may strongly be inhibited. Because highly supersaturated solutions may persist metastably, it can be anticipated that the migration of Ra through the repository near-field will strongly depend on parameters related to nucleation and precipitation kinetics, so that thermodynamic equilibrium calculations will not be sufficient for a reliable prediction of ²²⁶Ra mobility. In this study, we implement Classical Nucleation Theory (CNT) and a saturation-state dependent precipitation rate equation into a Lattice-Boltzmann (LB) reactive transport code to model Ra-bearing baryte precipitation within a porous medium analogous to fragmented nuclear waste glass. In the simulations, baryte precipitation is induced by counter-diffusion of BaCl₂ and Na₂SO₄ solutions. Radium co-precipitation is taken into account by assuming a fixed partition coefficient and constant Ra concentration at the BaCl₂ injection boundary. Both homogeneous and heterogeneous growth were considered. Critical CNT parameters, particularly supersaturation-dependent induction times, were calibrated against independent turbidity and X-ray absorption experiments. The model allows exploring the influence of baryte nucleation/precipitation kinetics on the partitioning of Ra between aqueous phase and solid at the pore (micrometre) scale. Our results indicate that quantitative knowledge of kinetic and nucleation parameters is essential to predict radionuclide transport towards the geosphere in nuclear waste repository systems.

Speciation of carbon during the anoxic corrosion of steel is poorly known, whereas its knowledge would be of great importance in connection with assessments of the safe disposal of ¹⁴C-containing irradiated steel in repositories for radioactive waste. The chemical form of the ¹⁴C-bearing organic compounds determines routes of migration from engineered barrier systems and their reactivity at disposal sites. Batch-type corrosion experiments with unirradiated iron powders reported in this study show for the first time that both reduced and oxidized carbon species are present in corroding iron-water systems in anoxic conditions. Methane and volatile C2–C4 alkanes and alkenes were produced during the course of corrosion whereas formate, acetate, and oxalate were released to solution in the early stage of the corrosion process. Evidence is provided that reduced and oxidized hydrocarbons were produced by two different processes. Formation of reduced hydrocarbons occurred at the surface of iron particles by a Fischer-Tropsch-type mechanism, whereas oxidized hydrocarbons were produced in the course of oxidative pretreatment of iron particles and released instantaneously from the surface in contact with alkaline solution. Results from this study have implications for safety assessments of radioactive waste disposal sites as they suggest predominant formation of alkanes and alkenes during anoxic steel corrosion and instantaneous release of only a small fraction of carbide carbon as oxidized hydrocarbons.

Low- and intermediate-level (L/ILW) radioactive waste produced in Switzerland consists of large amounts of ¹⁴C-containing irradiated steel. ¹⁴C will be released during the anoxic corrosion of the steel in the cementitious near field of an L/ILW repository. In this study, a corrosion experiment with irradiated steel was carried out to determine the speciation of ¹⁴C released during the corrosion process in conditions similar to those anticipated in the near field of a cement-based repository. The development of the experimental setup, including installation of the reactor and development of suitable analytical methods based on compound-specific ¹⁴C analysis with accelerator mass spectrometry (CSRA AMS) is reported. Time-dependent increase in the total content of 14C-bearing organic compounds in solution (TO¹⁴C) was determined by AMS and the main organic corrosion products that are ¹⁴C bearing formate, acetate and lactate were identified by CSRA AMS after a pre-concentration step. The concentration of the ¹⁴C-bearing organic compounds was found to be very low (fmol to pmol ¹⁴C/L). Stable carbon compounds were identified and quantified while the source of stable carbon in the system has not yet been identified and the temporal evolution of the concentration of these carbon species is presently not understood.

The combination of ion chromatography (IC) with accelerator mass spectrometry (AMS) was developed to determine the speciation of ¹⁴C-(radiocarbon) bearing organic compounds in the femto to pico molar concentration range. The development of this compound-specific radiocarbon analysis (CSRA) of carboxylic acids is reported and the application of the method on a leaching solution from neutron-irradiated steel is demonstrated. The background and the dynamic range of the AMS-based method were quantified. On using ¹⁴C-labelled standards, the measurements demonstrate the repeatability of the analytical method and the reproducible recovery of the main target carboxylic acids (i.e., acetate, formate, malonate, and oxalate). The detection limit was determined to be in the mid fmol ¹⁴C per L level while the dynamic range of the analytical method covers three orders of magnitude from the low fmol to the mid pmol ¹⁴C per L level. Cross contamination was found to be negligible during IC fractionation and was accounted for during eluate processing and ¹⁴C detection by AMS. The ¹⁴C-bearing carboxylates released from an irradiated steel nut into an alkaline leaching solution were analysed using the CSRA-based analytical method with the aim to check the applicability of the approach and develop appropriate sample preparation. The concentrations of ¹⁴C-bearing formate and acetate, the main organic corrosion products, were at a low pmol ¹⁴C per L level for convenient dimensions of the alkaline leaching experiment which demonstrates that compound-specific ¹⁴C AMS is an extremely sensitive analytical method for analyzing ¹⁴C-bearing compounds. The content of total organic ¹⁴C in solution (TO¹⁴C) determined by the direct measurement of an aliquot of the leaching solution agrees well with the sum of the ¹⁴C concentrations of the individual carboxylates within the uncertainty of the data. Furthermore, the TO¹⁴C content is in good agreement with the calculated value using the corrosion rate determined from the ⁶⁰Co release and the ¹⁴C inventory of the irradiated steel specimen.

Understanding the mechanism of ion diffusion in hardened cement paste is of great importance for predicting long-term durability of concrete structures. Gel pores in calcium silicate hydrate (CeSeH) phase forms dominant pathway for transport in cement paste with low w/c ratios where the electrical double layer effects play an important role. Experimental results suggest that the effective diffusivity of chloride ions is similar as that of tritiated water (HTO) and higher than the sodium ions. This difference can be attributed to the electrical double layer near the charged CeSeH surfaces. In order to understand species transport processes in CeSeH and to quantify its effective diffusivity, a multiscale modeling technique has been proposed to combine atomic-scale and pore-scale modeling. At the pore scale, the lattice Boltzmann method is used to solve a modified Nernst Planck equation to model transport of ions in gel pores. The modified Nernst Planck equation accounts for steric and ion-ion correlation effects by using correction term for excess chemical potential computed through the results from the grand canonical Monte Carlo scheme at atomic scale and in turn bridges atomic scale model with pore scale model. Quantitative analysis of pore size influence on effective diffusivity carried out by this multiscale model shows that the contribution of the Stern layer to ion transport is not negligible for pores with diameter less than 10 nm. The developed model is able to reproduce qualitatively the trends of the diffusivity of different ions reported in literature.

Atomistic simulations provide insight into the crystal structure of minerals, surfaces, and mineral-fluid interaction mechanisms. Such modelling has been successfully applied to better understand structural, thermodynamic, and transport properties of clay minerals, the thermodynamics of ions adsorption, and clay mineral surface reactivity at an atomic scale. In principle, quantum mechanics- based modelling allows system description without use of any empirical system-dependent parameters. In practice, however, the complete quantum mechanical description of the condensed matter is only feasible for small systems containing few atoms, due to the limitations of currently available computational resources. Therefore, the simulations of complex reactive processes rely on a number of approximations at different levels of theory. These approximations are chosen as a compromise between the computational accuracy and the ability to include the relevant chemical processes.
This chapter starts with a short overview of the methods of quantum chemistry currently applied to the simulations of clay minerals. Theoretical equations are intentionally excluded. The focus is on the physical rationale behind the methods, assumptions applied, and their consequences for the interpretation of the results. For the theoretical details of the methods, the reader is directed to the specialized text books and review articles provided as references. The second part of the chapter deals with the applications. The chapter starts with the description of the bulk crystal structure, continues with the structural properties of the surfaces and surface-fluid interfaces, and concludes with thermodynamic and structural aspects of adsorption.

Clay minerals are important adsorbents in soils and sediments for hazardous contaminants in the environment. A whole variety of adsorption models for clay minerals has been developed over the past few decades. In the first part of this chapter, a brief overview of existing models describing cation exchange and surface complexation is presented. The second part presents for a large number of heavy metals and radionuclides exhibiting oxidation states from + I to + VI experimental adsorption data onto montmorillonite (Mt) and illite. A quasimechanistic nonelectrostatic adsorption model is applied to describe quantitatively the uptake of these elements over a broad pH, background electrolyte, and adsorbate concentration range. In the third part, special focus is put on the adsorption behaviour of iron on Mt. Wherever possible, a multidisciplinary approach is followed, whereby the wet chemistry studies are complemented by modelling and spectroscopic investigations with the aim of validating the underlying assumptions in the adsorption model.

Dissolution of carbonate minerals is a complex multistep process, characterized by the particular sequence of steps dependent on pH and background electrolyte concentration. Currently, available dissolution models for carbonates do not consider dependence of the surface speciation on the local surface topography. We have developed a new approach combining grand canonical Monte Carlo (GCMC) and kinetic Monte Carlo (KMC) methods to investigate the influence of water pH and electrolyte concentration onto processes of surface charging and dissolution of carbonates. GCMC simulations of the calcite−electrolyte system are used to calculate populations of protonated sites. We consider two basic speciation models characterized by different spatial charge distributions at the surface: “ionic”, where surface >CO₃²⁻ sites are represented by “−2” charges at the corresponding lattice positions; and “oxygen”, where surface >CO₃²⁻ sites are represented by triplets of “−2/3” charges at the positions of oxygen atoms. The speciation of carbonate ion protonation probabilities is found to be controlled by local charge densities and the presence of electrolyte species. In all simulation results, protonation affinity of the surface >CO₃²⁻ sites followed the trend kink (most acidic) > step > terrace (least acidic), with the same trend observed with respect to adsorption probabilities of Cl− ions. The influence of protonated site concentrations obtained in GCMC simulations was investigated in KMC simulations. The direct comparison of simulated and experimental data showed that the oxygen model, with an assumption of congruent dissolution, reproduces both the pH dependence of the calcite dissolution rate and the morphology of the calcite surface. On the basis of the considered model, we could identify four key factors that define pH-dependent dissolution mechanisms of calcite: (1) increase of the kink site propagation rate at pH < 10; (2) increase of kink site generation frequency at pH 4−7; (3) increase of monolayer pit generation frequency at pH = 2−4; and (4) acceleration of kink site propagation and generation at pH 2−4 due to the second protonation step. The combined GCMC + KMC approach shows great potential in resolving surface speciation of carbonates as functions of solvent composition and surface geometry and their influence on the dissolution mechanisms and rates. Generally, this approach could potentially be applied to any other mineral−fluid system.

Reactivity of minerals is controlled by chemical processes at mineral- fluid interfaces acting at different time- and length scales. Various modeling approaches are available to characterize scale-specific aspects of mineral-fluid interface chemistry. Most fundamental aspects of mineral reactivity are provided by atomic scale simulations. Several attempts have been made to interpret macroscopic observation based on atomic scale simulations alone. Many of them have failed however, because of neglecting the pore scale transport phenomena. Pore scale simulation, provide an elegant way to link idealized nanometer scale atomistic description of mineral reactivity with structural and compositional heterogeneities of natural systems. The main challenges are the spatial and temporal coupling of physical models and the upscaling of transport parameters for the macroscopic interpretation of the system behavior. This paper summarizes the current molecular-scale knowledge on mineral-fluid inter- face chemistry, obtained from complementary coarse-grain simulation approaches. Using the most recent developments in this field, we highlight the complexity and challenges of the pore-scale modeling and suggest a roadmap for the process-based description of mineral dissolution/precipitation across different scales.

Through-diffusion experiments with ³⁶Cl^-, ³⁵SO₄²⁻ and HTO in Opalinus Clay (OPA) samples from a deep borehole in North-East Switzerland (Benken; BE) have been performed. The effect of burial depth on the experimental results has been investigated. It could be shown that the effective diffusion coefficients decrease with sample depth for all three tracers. Moreover, there was a good correlation with the texture of the samples. The diffusion coefficients for HTO are the largest (D_e = 5.4–8.8 × 10⁻¹² m² s⁻¹), followed by those for ³⁶Cl^- (D_e = 0.7–1.9 × 10⁻¹² m² s⁻¹), and finally ³⁵SO₄²⁻ (D_e = 0.2–0.6 × 10⁻¹² m² s⁻¹). ³⁶Cl^- was partially excluded from the total porosity resulting in an accessible porosity smaller than the total porosity (ε_Cl = 0.041–0.064). ³⁵SO₄²⁻, on the other hand, showed interaction with OPA resulting in a capacity factor (α) larger than the total porosity (ε_tot = 0.13–0.16). Using extended Archie's law the accessible porosity for ³⁵SO₄²⁻ was estimated between 0.013 and 0.030. This enabled to evaluate the sorption coefficient of ³⁵SO₄²⁻ from the measured capacity factor, resulting in values of K_d between 6 × 10⁻⁵ and 9 × 10⁻⁵ m³ kg⁻¹.

Through-diffusion experiments with tritiated water (HTO) and ³⁶Cl^- as a function of pore water concentration (0.01–5 M) were performed on two Ordovician-age argillaceous rock samples from the Blue Mountain Fm and Queenston Fm shales of the Paleozoic intracratonic Michigan Basin in Canada. This study reveals that the effect of ionic strength on the anion-transport porosity is similar, and only the minimal anion excluded porosity is higher in the Blue Mountain Fm shale. The differences in rock sample mineralogy cannot explain this effect. It is hypothesized that the structure of the Blue Mountain Fm shale samples has led to pore space openings suffi- ciently small that they behave as interlayers. Such pores are defined as interlayer equivalent (ILE) pores. These ILE pores, as in the case of interlayer pores, can act to permanently limit the anion-accessible porosity. Pore-size distribution measurements provide further evidence of increased potential for ILE pores within the Blue Mountain Fm samples. A Donnan model, which includes consideration of both ILE and uncharged pores, is shown to describe the effect of molar concentration on the anion-accessible porosity in the argillaceous rocks investigated.

The pore size distribution of two natural argillaceous rock samples, Opalinus Clay (OPA) and Helvetic Marl (HM) was investigated with five different methods: NMR, NMR cryoporometry, mercury intrusion porosimetry and CO₂ adsorption, as well as N₂ adsorption. Due to different physical principles of these methods different ranges of pore width could be detected, from micropores (< 2 nm) to mesopores (2–50 nm) and macropores (> 50 nm). The aim was to shed light on the role of small pores on the transport properties of natural ar- gillaceous rocks, in particular to explain the differences of anion diffusion in the two argillaceous rock sam- ples. Knowing that Helvetic Marl exhibits a stronger anion exclusion than Opalinus Clay it was hypothesized that HM (with its smaller phyllosilicate and smectite content compared to OPA) has more interlayer equivalent (ILE) pores than OPA. ILE pores are defined as pores so narrow (< 0.5 nm) that diffuse double layers, formed at negatively charged surfaces, are overlapping. Accordingly, ILE pores behave similarly as interlayer pores and may block the anion diffusion. This study could not confirm the hypothesis that HM has more ILE pores. Similar pores size distributions were determined for both materials, even with a tendency of a larger fraction of small pores in OPA as compared to HM. However, all methods have limitations in the range of very small (nm) pores.

The effect of the pore water composition on the diffusive anion transport was studied for two different argillaceous, low permeability sedimentary rocks, Opalinus Clay (OPA) and Helvetic Marl (HM). The samples were saturated with different solutions with varying molar concentration and different main cations in the solution: NaCl based pore solutions and CaCl₂ based pore solutions. The total porosity was measured by through-diffusion experiments with the neutral tracer HTO. Experiments performed in NaCl solutions resulted in a porosity of 0.12 for OPA and 0.03 for HM, and are consistent with results of the experiments in CaCl₂ solutions. The total porosity was independent of the molar concentration, in contrast to the measured anion porosity, which increased with increasing molar concentration. It could further be observed that the pore solution based on the bivalent cation calcium shielded the negative surface charge stronger than the monovalent cation sodium, resulting in a larger measureable anion-accessible porosity in the case of CaCl₂ solutions. The data was modelled based on an adapted Donnan approach of Birgersson and Karnland (2009). The model had to be adjusted with a permanent free, uncharged porosity, as well as with structural information on the permanent anion exclusion because of so-called bottleneck pores. Both parameters can only be evaluated from experiments. Nevertheless, taking these two adaptions into account, the effect of varying pore water compositions on the anion-accessible porosity of the investigated argillaceous rocks could be satisfactorily described.

Understanding ion transport through clays and clay membranes is important for many geochemical and environmental applications. Ion transport is affected by electrostatic forces exerted by charged clay surfaces. Anions are partly excluded from pore water near these surfaces, whereas cations are enriched. Such effects can be modeled by the Donnan approach. Here we introduce a new, comparatively simple way to represent Donnan equilibria in transport simulations. We include charged surfaces as immobile ions in the balance equation and calculate coupled transport of all components, including the immobile charges, with the Nernst-Planck equation. This results in an additional diffusion potential that influences ion transport, leading to Donnan ion distributions while maintaining local charge balance. The validity of our new approach was demonstrated by comparing Nernst-Planck simulations using the reactive transport code Flotran with analytical solutions available for simple Donnan systems. Attention has to be paid to the numerical evaluation of the electrochemical migration term in the Nernst-Planck equation to obtain correct results for asymmetric electrolytes. Sensitivity simulations demonstrate the influence of various Donnan model parameters on simulated anion accessible porosities. It is furthermore shown that the salt diffusion coefficient in a Donnan pore depends on local concentrations, in contrast to the aqueous salt diffusion coefficient. Our approach can be easily implemented into other transport codes. It is versatile and facilitates, for instance, assessing the implications of different activity models for the Donnan porosity.

Safety assessment studies of future nuclear waste repositories carried out in many countries predict selenium-79 to be a critical radionuclide due to its presence as anions in three relevant oxidation states (VI, IV, -II) resulting in weak retardation by most common rock minerals. This assumption, however, ignores its potential uptake by AFm phases, positively charged anion exchangers, which are present in significant quantities in the cementitious materials used in artificial barriers. Here we report for the first time wet chemistry and spectroscopic data on the interaction of the most relevant selenium anion species under the expected strongly reducing conditions, i.e. HSe^-, with two AFm phases commonly found in cement, monocarbonate (AFm-MC) and hemicarbonate (AFm-HC). Batch sorption experiments showed that HSe^- is retained much more strongly by AFm-HC (solid-liquid distribution ratio, R_d, of 100±50 L kg-1) than by AFm-MC (R_d = 4±2 L kg^-1) at the equilibrium pH (~12). X-ray absorption fine-structure (XAFS) spectroscopy revealed that the larger d-spacing in AFm-HC (d-spacing = 8.2 Å) provides easy access for HSe- to the AFm interlayer space for sorption, whereas the smaller d-spacing of AFm-MC (d-spacing = 7.55 Å) hinders interlayer access and limits HSe^- sorption mostly to the outer planar surfaces and edges of the latter AFm phase. XAFS spectra further demonstrated that Se(-II) prevalently sorbed in the interlayers of AFm-HC, is better protected from oxidation than Se(-II) prevalently sorbed onto the outer surfaces of AFm-MC. The quantitative sorption data along with the molecular-scale process understanding obtained from this study provide crucial insight into the Se retention by the cementitious near-field of a radioactive waste repository under reducing conditions.

¹⁴C-containing dissolved organic compounds may significantly contribute to the calculated annual overall dose emanated from a deep geological repository for radioactive waste. To date, there is a general lack of knowledge concerning the transport behaviour of low molecular weight organic compounds in the geosphere. The present work is aiming at a generic approach to measure weak adsorption of such compounds onto selected clay minerals. Percolation experiments were employed to sensitively measure the retardation of low molecular weight carboxylates and alcohols in compacted illite and kaolinite as a function of the ionic strength. Detection limits of ~10^-5m³kg^-1 for the involved sorption distribution coefficients were attained thereby. The adsorption of alcohols on clays was near the detection limit and assumed to occur predominately via H-bonding. The adsorption of organic anions was influenced by several factors such as molecular structure, type of clay surfaces and the chemical composition of the aqueous phase. It was found that the relative position of neighbouring hydroxyl groups strongly influ- enced the retardation behaviour. Alpha-hydroxylated carboxylates, such as lactate, were found to be most retarded. Ligand exchange at the edge aluminol sites is the most probable explanation for the uptake of the negatively charged organic test compounds by the clay surface. The breakthrough behaviour of organic anions was additionally impacted by anion exclusion in illite. The demonstrated weak retardation of the test compounds can be robustly introduced in transport models, leading thus to a much lower contribution of ¹⁴C to the expected long-term overall dose.

An internally consistent thermodynamic dataset for aqueous species in the system Ca-Mg-Na-K-Al-Si-O-H-C-Cl was generated using the thermodynamic database for minerals of Holland and Powell (1998; updated Thermocalc dataset ds55). This dataset makes it possible to perform geochemical and reactive transport modeling with high levels of accuracy and reliability.
The stability of major aqueous complexes at elevated temperatures and pressures was constrained using selected reaction constant data (for example plot A and B). The Gibbs energy of formation of aqueous ions and complexes was simultaneously optimized with GEMSFITS against a large selection of solubility experiments over a wide range of conditions, taking the standard properties of minerals (unmodified) from the Holland and Powell internally consistent database (for example, plot C). The resulting thermodynamic dataset is consistent with the complex formation data, with the mineral solubility experiments, and with the standard properties of minerals from Holland and Powell database. The internally consistent dataset can be used to model natural fluid-rock interaction (for example, plot D).
Comparison of calculated and measured experimental data for: (A) the stability constant of HCO₃^- as function of pressure at temperatures of 55, 150 and 250 °C; (B) the association constant of CaCl⁺ as function of temperature at saturated water vapor pressure; (C) calcite solubility in NaCl solutions at 400 °C; (D) log(Ca/Mg) molar ratios from sedimentary fluids in equilibrium with calcite and disordered dolomite, at temperatures of 50 to 150 °C and at saturated water vapor pressure.

Mineral precipitation and dissolution in aqueous solutions has a significant effect on solute transport and structural properties of porous media. The understanding of the involved physical mechanisms, which cover a large range of spatial and temporal scales, plays a key role in several geochemical and industrial processes. Here, by coupling pore scale reactive transport simulations with classical nucleation theory, we demonstrate how the interplay between homogeneous and heterogeneous precipitation kinetics along with the non-linear dependence on solute concentration affects the evolution of the system. Such phenomena are usually neglected in pure macroscopic modelling. Comprehensive parametric analysis and comparison with laboratory experiments confirm that incorporation of detailed microscale physical processes in the models is compulsory. This sheds light on the inherent coupling mechanisms and bridges the gap between atomistic processes and macroscopic observations.