An understanding of the behaviour of contaminants in compacted clays is important in assessing the effectiveness of clay-based barrier materials used in many waste containment strategies. Here the diffusion and sorption behaviour of I− with selected compacted clays is examined (129I is a relatively long-lived radioisotope present in high-level nuclear fuel waste.) Diffusion coefficients, D, and distribution coefficients, Kd , were measured for I− with four clays: bentonite, Lake Agassiz clay (a glacial lake clay composed mainly of smectite, illite, kaolinite, and quartz), interstratified illite-smectite, and kaolinite. For the diffusion experiments the clays were compacted to a dry bulk density, ρ, of ∼1.2 Mg m−3. The mean measured D values, D m, were as follows: bentonite, 310 µm2 s−1; Lake Agassiz clay, 0.62 µm2 s−1; illitesmectite, 190µm2s−1; and kaolinite, 74µm2s−1. The measured values were compared with those calculated, D c, from the following model: D=D oτa n/(n + ρKd ), where Do is the diffusion coefficient in pure bulk water, τ a the apparent tortuosity factor, and n the solution-filled porosity of the clay. Except for the Lake Agassiz clay, where D m was about an order of magnitude lower than D c, the D m and D c values agreed within a factor of about two. The Lake Agassiz clay has a markedly higher organic carbon content than the other three clays, and this could affect τ a, which may be overestimated in the model.

The pillaring process of montmorillonite and beidellite with Al and Ga polymers has been studied using XRD, IR, 27Al, 71Ga, and 29Si MAS NMR, TGA, TEM, N2 adsorption and chemical analyses. The Al adsorption maximum for montmorillonite is close to 5.5 mEq Al/g clay, whereas the maximum for Ga is higher. Basal spacings of both Ga- and Al-pillared clays vary between 16.7 and 18.8 Å. Freeze-drying of pillared products followed by calcination yielded more regular pillared structures. Pillaring montmorillonite increased the BET surface area from 35 m2/g to 350 m2/g mainly by the creation of micropores <20 Å in diameter. The Al-pillared clays are thermally stable to ∼700°C. Calcination of pillared montmorillonite liberates protons from the pillar, which diffuse into the clay sheet, lowering the thermal stability. In pillared beidellite, mainly silanol groups are formed by breaking Si-O-Al bonds. No reaction is observed between pillars and montmorillonite upon calcination, whereas in pillared beidellite a structural transformation links the pillar to inverted tetrahedra of the tetrahedral sheet. The basal spacing of Ga-pillared montmorillonite collapses to 9.5 Å at 350°C due to the Ga polymer decomposing to Ga3+ cations.

‘Epoxyphilic’ montmorillonites were prepared containing various functional groups which can react with one or both of the components in amine-cured epoxy formulations. The preparation involves the ion exchange of six organic ammonium compounds containing either carboxylic acid, anhydride, phenolic hydroxyl, alcoholic hydroxyl, amide or amine functional groups. X-ray diffraction, elemental analysis and FTIR studies confirm the intercalation of the organic cations to form epoxyphilic montmorillonites. The adsorption behaviour of the modified montmorillonites are compared with the unmodified montmorillonite in an epoxy resin before and after the curing process.

Methylene blue (MB) has been used as a probe molecule to examine how the uptake dynamics and the equilibria between this dye and the surface of Na+- and H+- montmorillonite were affected by the presence of a second dye. To prevent spectral interference, the yellow dyes thioflavin T, TFT, proflavine, PFH, and acridine yellow, ACY, were chosen to compete with MB for the exchange sites. The MB was initially adsorbed as trimer (MB+)3 and then redistributed via collisions between clay particles until equilibrium was reached. At equilibrium in the clay/MB systems, the protonated species (MBH2+) predominated at low loadings (1–5% CEC), whereas at higher loadings the trimer (MB+)3 was the major species. The presence of the second, competing dye slowed the approach to equilibrium, significantly reduced the amount of MBH2+ formed and provided evidence for the monomeric MB+, dimeric (MB+)2, and trimeric (MB+)3 forms of MB. Moreover, the presence of PFH and ACY, which are structurally similar to MB, resulted in more dimeric character in the aggregated form of MB compared to the aggregate formed in the presence of the structurally dissimilar TFT.

Intercalation of macrocyclic compounds (crown-ethers and cryptands) in 2:1 charged phyllosilicates is a topotactic reaction, the lamellar structure of the pristine material being preserved. X-ray diffraction and spectroscopic (IR, NMR) results provide information on the main structural characteristics of the intercalates. Guest species, which form 1 : 1 or 2 : 1 ligand/cation intracrystalline complexes, lie flat or tilting between the host layers as can be deduced from the increase in the basal space of the silicate and the dichroic effect observed on the CH and NH IR vibration bands of the guest species, after intercalation. The chemical shift observed in 23Na NMR spectra of the solids is associated with the different environment of the cation before and after the intercalation process.

To investigate the long-term stability of bentonite under final disposal conditions of highly radioactive waste, K-bentonites from Kinnekulle (Sweden) and from the Marias River Formation in the Montana disturbed belt (USA) were studied. After separating the mixed-layer illite-smectite (I-S) from the K-bentonite samples, the interlayer charge was calculated from the cation exchange capacity (CEC) and the amount of fixed interlayer K+ ions (Kfix). The interlayer charge was also determined by the alkylammonium method. According to both methods the interlayer charge was in the range for smectite. The results show that the amount of exchangeable cations increased linearly with decreasing Kfix. A small increase in the interlayer charge with increasing Kfix was observed as was a linear correlation between the intracrystalline swelling up to the second water layer, the CEC and the content of Kfix. Divalent exchangeable cations were then found to be surrounded by approximately 24 water molecules per cation. Fixed interlayer K+ ions were unhydrated. Forming the third and fourth water layer, swelling was presumably limited by free silica formed by the vitrification of the volcanic ash.

The uptake of Ba2+ and K+ by a synthetic Na-phyllomanganate has been studied by chemical analysis, X-ray diffractometry and X-ray photoelectron spectroscopy. The changes in basal spacing arising from cation exchange have been used to monitor the progress of the exchange reaction and confirm the selectivity measured by bulk chemical analysis. However, the selective uptake of Ba2+ over Ca2+ is much greater than that of K+ over Na+ and it is suggested that charge distribution in the interlayer is important. The results are discussed in the light of recent advances in our understanding of the phyllomanganate structure.

The <2 µm clay fractions from low-grade metabasalts of eastern North Greenland are mixtures of mafic phyllosilicates and celadonite that show complex X-ray diffraction (XRD) patterns. Interpretation of such patterns is difficult and subjective using only visual examination of peak positions and shapes. Deconvolution analysis offers a less subjective means of identifying peak positions for overlapping peaks and is applied to several composite peaks in a pattern; the peaks identified are then rationalized in terms of specific mixed-layer phases. Comparison against NEWMOD patterns of the identified clay mixture provides an additional constraint on the phases identified. Three mafic phyllosilicates, discrete chlorite and two chlorite-smectites with 60:40 and 80:20 proportions are demonstrated in the same XRD pattern. Modelling using NEWMOD of a mechanical mixture between these mineral shows an excellent match to the observed XRD pattern. The recognition in low-grade metabasalts of chlorite and random and different varieties of regular mixed-layer chlorite-smectite offers support for the classical model of a tri-smectite to chlorite transition. This is in contrast to an alternative model proposing smectite and chlorite end-members with corrensite as a discrete phase crystallizing directly without intervening mixed-layer chlorite-smectite.

The soils over about half of South Africa are developed from sedimentary rocks of the Karoo Supergroup. Some units within the Karoo strata display variations in their mineralogical composition which correlate with fluctuations in the environment of deposition and with palaeotemperature. The former trends are reflected in the ratio of kaolinite to 2:1 layer-silicates, range from kaolinite-dominated to kaolinite-free, and are particularly evident in rocks of the Vryheid Formation. Geothermal history is preserved in various stages of smectite illitization ranging from R = 0 structures with 25% illite to 100% illite. Compositional trends may be significant on a regional, local or topographic scale down to a distance equal to the depth of a soil profile or even an horizon.

The use of electron microscopy to study clay microfabrics in thin-sections is discussed. A technique is described to isolate undisturbed microparts of pedofeatures from thin-sections, which are subsequently used for TEM analysis. Re-embedding with a polyester resin of undisturbed, in situ, neoformed clay microfabrics, obtained by microdrilling and preparation of ultrathin sections by microtoming with a diamond knife are emphasized; these steps enable micromorphology, clay mineralogy, microchemical and HRTEM analysis to be performed on one unique microsample of clay fabrics, with conserved micro-organization. Two examples on clay neoformation are presented to demonstrate that this technique can successfully be applied to unravel the impact of mineral alteration and clay neoformation in undisturbed soil samples on a micro- and a nanometer scale.

The rocks of the palygorskite deposit at Bercimuel (Segovia, Spain) have been studied by transmission electron microscopy and microanalysis. These rocks correspond to the zone of convergence of two alluvial fan systems that have filled the small basin of the River Riaza and would originally have been composed of illite and quartz silts. Among the accumulations of palygorskite it is possible to observe surrounded particles of micromicas that have undergone dissolution and opening thereby giving rise to disordered illite-smectite mixed-layer clays. This process continued up to the individualization of structural relics formed of units of 1–5 layers. At the same time, the chemical composition was modified with a loss of K and Al and a relative increase in Si and Mg, progressively evolving towards the composition of palygorskite. The paleogeographic position of the deposits, and the climatic conditions (arid environment), appear to be the dominant factors in the neoformation of palygorskite by alteration of the original sediments. The basic mineralogical process could be referred to as ‘early diagenesis’ in the formation of calcretes.

Corrugated mats of palygorskite occurring in regolith overlying a fissured limestone were examined by X-ray diffraction, infrared spectroscopy, and electron optical techniques after heating at 50°C intervals to 700°C. Palygorskite ‘anhydride’ formed at 400°C and sillimanite formed at 500°C. The palygorskite is believed to have formed in joints in limestone of Upper Oligocene age (Duntroonian) prior to uplift and subsequent weathering. It is highly crystalline, and appears not to have altered or weathered since precipitation.

Three duricrust types (calcretes, palycretes and silcretes) have been distinguished in the Lower Paleogene arkosic materials from the margins of the Duero and Tajo basins (central Spain). In the calcretes the cements are composed of calcite plus palygorskite, whereas palygorskite is the only cement in the palycretes. In the silcretes, the cement consists chiefly of opal and chalcedony, with minor quartz. The important duricrust development in the Paleogene detrital materials can be correlated with a drastic climatic change towards colder and more arid conditions during this time, as inferred from the sedimentological features and the clay mineral evolution. The origin of calcretes and palycretes is related to the movement of alkaline phreatic waters generated from dolomitic rocks of Cretaceous age cropping out in the near surroundings. Silcretes originated from the vertical transfer of silica from supersaturated groundwaters, from the Hercynian granitic rocks, up to the surface. Silcretes are developed either on unconsolidated arkosic materials or replace previous calcrete or palycrete levels.

Highly crystalline kaolinite has been studied by 27Al satellite transition magic angle-spinning (MAS) nuclear magnetic resonance (NMR) spectroscopy. This technique provides an additional narrowing of the MAS spectral lines by a factor of more than three compared with the width of the prominent central transition. Evidence for the presence of two crystallographically inequivalent hexacoordinated Al sites has been found. The results support the kaolinite crystal structure proposed by Bish & Von Dreele (1989).