Separate montmorillonite suspensions were treated with CrCl3 to give seven Cr(III) concentrations. Each suspension was then titrated with 0.1 N NaOH to give a NaOH/Cr3+ molar ratio of 2.5. Montmorillonite was an effective sorbent for hydroxy-Cr species up to 1200 cmol(+)/kg; above that concentration, sorption continued, though less efficiently. However, N2-BET specific surfaces and cation exchange capacity measurements indicated that the montmorillonite could sorb significantly more than 1200 cmol(+)/kg. There was virtually no exchangeable Cr in any of the clays, suggesting that this element was covalently bonded to the siloxane surface. Infrared spectroscopy revealed a vibration at 1015–1020 cm–1 in the Cr clays which was not present in the control. This new absorption band was attributed to an attenuation of the Si–O a11 vibration caused by inner-sphere complexation of the interlayer Cr with the siloxane oxygen. Because Cr was strongly held and efficiently sorbed, montmorillonite was shown to be an effective sorbent for hydroxy-Cr polymers.

Upper Triassic sandy horizons in the Paris Basin were sampled at depths ranging from outcrop in the northeast to 2,700 m in the centre of the basin. The smallest clay sub-fractions (<0.2 μm) from the deepest central samples consist mainly of illite and chlorite whose K-Ar age is ∼ 190 Ma. These minerals formed at a relatively high temperature of 220-250°C, as determined from the oxygen isotope fractionation between authigenic illite and associated quartz overgrowths, but at a burial depth of only 500 m. In a nearby drillhole that crosscuts a fault zone reactivated during the 190 Ma event, illite is less well-crystallized and has a higher δ18O value suggesting different physical and chemical conditions of formation during the same hydrothermal episode. Two other generations of illite-smectite mixed-layer clays formed in the same Triassic horizon: one at ∼150 Ma and the other at ∼80 Ma. These younger clays have higher δ18O values and thus may have formed at somewhat lower temperatures. The δ18O values of the fluids from which the different illite-smectite mixed-layer minerals form range from +9 to +13.5 per mil (SMOW).

An Ar diffusion code was used to estimate, on the basis of the Ar loss of the clay-type material, the duration of these events. The results suggest that the duration of the hydrothermal events at 190 Ma and 150 Ma were rather short, <1 Ma, whereas the youngest event was protracted over a much longer period of ∼37 Ma. Comparison between K-Ar ages of the different mixed-layer minerals and sedimentation rates of the sediments since the Palaeozoic shows significant accelerations of the rates at ∼ 190-200 Ma and 150 Ma, and a less important one at ∼80 Ma. These observations provide additional evidence that the first two events were promoted by basement-related tectono-thermal activities. The third event is considered to be of diagenetic origin.

Dehydration and rehydration of a phlogopitic Ca-vermiculite were studied by thermoanalysis (TG, DSC) and by in situ X-ray powder diffractometry. A detailed analysis of the relationship between temperature and state of hydration of the vermiculite close to equilibrium at atmospheric pressure was rendered possible by use of low heating rates in combination with a fast recording position-sensitive detector. The results reveal the existence of a number of distinct states of hydration characterized by basal spacings of 1.515, 1.493, 1.473, 1.188, 1.176 and 0.95 nm. In confirmation of results obtained earlier with Mg-vermiculite, regular interstratification occurred at dehydration during the transition between two- to one- and between one- to zero-layer hydrates showing integral series of 00l reflections with basal spacings of 2.67 and 2.11 nm, respectively.

Isotherms were obtained for nitrogen adsorption and desorption on seven halloysite-rich samples from New Zealand and Western Australia. Calculations from these isotherms indicate that halloysites with mainly small particles (< c. 0.08 μm in width) had abundant cylindrical pores with narrow size distributions in the 5-15 nm range. They also indicate that halloysites with mainly large particles (> c. 0.1 μm in width) had few if any pores in the mesopore range (2–50 nm). Transmission electron microscopy (TEM) shows that cylindrical pores originate from the central holes in tubular particles. The TEM also suggests that slit-shaped pores can originate from the shrinkage of blocks of layers upon dehydration of halloysite.

Electron microscopy was used to study swollen samples of n-butylammonium vermiculite prepared in dilute aqueous solutions of n-butylammonium chloride. Three electrolyte concentrations were studied, 0.1 M, 0.01 M and 0.001 M. The 0.1 M samples studied were in a taetoid phase, exhibiting crystalline legions of about ten layers separated by large irregular aqueous regions. The dilute samples were studied in their gel phase, exhibiting regular spacings of 85 nm and 55 nm in the 0.001 M and 0.01 M solutions, respectively. These spacings are much larger than those measured previously by neutron scattering, probably due to an additional expansion of the interlayers during the freezing process employed in preparing the replicas. In all examples, the structures were found to be lamellar over regions of several microns.

The rheology of aqueous suspensions of Na halloysites with different particle shape has been investigated using a Haake rotary viscometer. Three halloysites from New Zealand were used: Matauri Bay (MB), Te Akatea (TA) and Opotiki (Op) which are mainly composed of thick long tubules, short thin laths, and spherules, respectively. Suspensions of the materials subsequently prepared from the MB and TA samples show a pseudoplastic consistency under shear, characterized by a steep initial rise in shear stress as shear rate increases, followed by a linear increase in stress when a certain shear rate is exceeded. The Op halloysite, on the other hand, shows Newtonian flow behaviour for which shear stress increases linearly with shear rate throughout. For MB and TA, the shear stress developed during rotor acceleration is larger than that produced during deceleration, and the corresponding flow curves enclose a hysteresis loop. As suspension pH increases, the pseudoplastic character decreases as does the amount of hysteresis. Above pH 7.5, flow approaches Newtonian and hysteresis is absent. Plots of Bingham yield value against pH at different ionic strengths (0.003, 0.03 and 0.3 M NaCl) intersect at pH 6.0 for MB and at pH 7.1 for TA. These values are identified with the point of zero charge (PZC) of the particle edge surface. The flow characteristics of halloysites may be explained in terms of the influence of particle shape, pH, electrolyte concentrations, and layer composition on particle interactions.

Chemical analyses of the <0.2 μm illite-smectite (I-S) fraction from K-bentonites in Silurian age sediments sampled from three different localities in the Southern Uplands of Scotland and Northern Ireland indicate that the I-S from the three localities shows differences in both K+ content and the degree of Tschermak substitution of octahedral Al3+ for Mg2+. The K+ content correlates with percentage expandability (%E) of I-S as determined by X-ray diffraction (XRD), but both these parameters show poor correlation with palaeotemperature as estimated by organic reflectance methods. Octahedral Mg2+ content shows an inverse correlation with Al3+, and both these parameters show a good correlation with estimated palaeotemperature. Although K-substitution has been retarded at Dob's Linn, the region subjected to the highest temperature, Tschermak substitution of Al3+ for Mg2+ has apparently continued. In these K-bentonites, Tschermak substitution appears to be a better indicator of palaeotemperature than K-substitution.

Caesium aluminium silicates, caesium aluminium fluoride and aluminium fluoride were identified in a mixture obtained through the evaporation of kaolinite aqueous suspension with CsF. The mixture was studied by X-ray powder diffraction and IR spectroscopy. Two types of intercalation complexes of kaolinite were identified (A and B) which were characterized by c-spacings of ∼0.99 and 0.83 nm, respectively. Complex A seems to be an hydrated CsF intercalation complex, whereas complex B is a water intercalation complex, probably containing H3O+ and F−. It showed an IR absorption band at 3518 cm−1. The c-spacings of similar CsCl and CsBr complexes depend on the ionic radii. The intercalated ions are keyed into the ditrigonal cavities of the oxygen plane. A straight line plot of the keying values versus the ionic size indicates that keying is high for F−and decreases with increasing ionic size.

Une nacrite est intercalée par deux composés organiques polaires: le diméthylsulfoxide (DMSO) et le n-méthylacétamide (NMA). Les deux solvants ont des moments dipolaires très voisins (4 debyes) mais des constantes diélectriques différentes (49 pour le DMSO et 179 pour le NMA). L'intercalation du NMA est plus rapide que celle du DMSO. Les deux complexes homogènes obtenus sont étudiés par diffraction des rayons X, spectroscopie infrarouge et ATD. L'étude par spectroscopie IR a montré que la nacrite expansée par du DMSO se comporte comme la kaolinite expansée par le même solvant. Dans le cas de la nacrite intercalée par du NMA, trois nouvelles bandes d'absorption dues aux vibrations de valence des OH liés par pont hydrogène avec le groupement C=O et situées respectivement à 3500, 3543 et 3589 cm−1 apparaissent. La fréquence v(N-H) du NMA est intermédiaire entre celles du liquide et d'une solution diluée, indiquant une liaison par pont hydrogène probablement avec les oxygènes de la couche tétraédrique du silicate. La diffraction des rayons X sur des échantillons orientés nous a permis d'obtenir 13 réflexions 00l pour chaque complexe. Une étude quantitative, par transformée de Fourier monodimensionnelle dans la direction perpendiculaire au plan du feuillet, a permis de déterminer le nombre de molécules organiques intercalées (une molécule par Si2Al2O5(OH)4) et leur orientation dans l'espace interlamellaire. L'ATD a montré par ailleurs que le complexe Nac.DMSO est plus stable que le complexe Nac.NMA.

Interaction between positively charged Fe-oxyhydroxide (β-FeOOH) particles and negatively charged particles of montmorillonite was studied as a function of particle concentration, pH and time. The results showed that the clay particles appear to bridge between FeOOH particles leading to associations in which the ratio of positive to negative particles varied between 1.4 and 2.56 depending on pH and particle concentrations. The data could be represented by isotherms of interaction of the high affinity type and, together with the interaction kinetics, could be described by Langmuir type formulations.

Current methods of quantitative whole-rock clay mineral analysis of sandstones often provide little more than an estimate of clay mineral abundances, especially where the total clay mineral content is <10 wt% of the sandstone. More accurate determinations of clay mineral abundance in the whole rock can be made by combining thermogravimetry/evolved water analysis (TG/EWA) and X-ray diffraction (XRD) data. The TGA/EWA system incorporates a purpose built thermobalance linked to a water specific infrared detector which is used to measure quantitatively the clay mineral dehydroxylation water evolved from the whole rock when heated from 250°C to 900°C. This gives a measure of the total hydroxyl content of the clay minerals in the whole rock which, when combined with XRD analysis of a separated clay size-fraction, enables individual clay mineral abundances in the whole-rock sample to be determined. Results on artificial sand/clay mineral mixtures prepared with known amounts of different clay minerals (chlorite, illite and kaolinite) show that the accuracy of the combined method is most influenced by the accuracy of the XRD data. Errors associated with TG/EWA were found to be negligible by comparison. A case study is included in which the technique has been used to determine accurately the illite abundance in the Magnus Sandstone Reservoir, Northern North Sea.

The comparison between measured and predicted smectite percent (%Sm) in an illite-smectite (I-S) mixed-layer series was used to estimate the thermal conditions of diagenesis in the Ardèche sedimentary series. Measurements were performed using the decomposition procedure (Lanson & Besson, 1992) of XRD patterns obtained from oriented preparations (<0.2 μm fractions). Three particle populations were identified: I-S, poorly crystallized illite (PCI), and well crystallized illite (WCI); the %Sm was measured in the I-S population by comparison with calculated XRD patterns using the NEWMOD software (Reynolds, 1985). The theoretical evolution of %Sm with depth obtained with the two first-order kinetic reaction processes described by Velde & Vasseur (1992) was compared to the measured one at Balazuc. The best fit was obtained for the following conditions: 35°C km–1 and between 1500 and 2000 m for the thermal gradient and the eroded piles, respectively. The predicted values for I-S compositions in the 10–0% Sm range are overestimated by the model because the reaction controlling the composition is no longer the dissolution of I-S but the ripening of illitic particles (PCI and WCI).

Two zones can be distinguished in the sedimentary series: (1) down to 55 m, the %Sm of I-S particles is highly variable and seems to be controlled by local variations in permeability and chemical composition; (2) down to 550 m, the %Sm of I-S particles is more strictly dependent on the general time-temperature conditions. An anomalously high %Sm value is located in the vicinity of faults at 1669 m due to local change of the water/rock ratio.

The development of the infrared (IR) absorption band at 1640 cm−1 (δH2O) is employed to monitor the gain in molecular water of progressively mortar-ground kaolinite. The planimetred area of this band shows a linear correlation with weight loss at 105°C (free moisture) between 2 and 18 h grinding, indicating a steady increase of molecular water in this range. Heating of ground products to 105°C causes a decrease of about ⅓ in the 1640 cm−1 peak area for all ground samples. This area decrease corresponds to a 34% (average) loss of free moisture as determined by gravimetry. The remaining water is held up to 280°C and is more tightly held in the kaolinite ground for shorter (2–5 h) intervals than in further ground kaolinites. The 1640 cm−1 peak is not detected in any ground kaolinite after heating to 600°C the temperature at which kaolinite dehydroxylates completely. The differentiation of the energetically different OH groups present in ground kaolinite and the mechanism of water gain are compared and discussed from IR spectroscopy and weight loss results.

Pisolites in the bauxitic laterites of Darling Range, Western Australia may contain up to 50% AI2O3, although no major Al-containing phases can be identified using standard X-ray diffraction techniques. These pisolites have been investigated using a range of techniques, and it has been demonstrated that Al is present in the form of x-alumina, a poorly crystalline form of A12O3. The x-alumina exhibits an indurating morphology and appears to have formed from a hydrated precursor.

The clay mineral distribution in the Early Cretaceous depositional sequences along the Cameros Basin has been established on the basis of XRD traces and TEM/EDAX analyses. Samples from the Latest Berriasian-Barremian depositional sequences are characteristic of epimetamorphic conditions. Illite ‘crystallinities’ are broader than expected (0.35–0.490°Δ2θ), due to the consistent presence of K-mica-paragonite mixed-layer and discrete paragonite associated with the illite 10 Å peak. The Late Barremian-Early Aptian depositional sequence, around the depocentral sector of the basin, represents a sudden change to anchimetamorphic conditions, marked by the presence of pyrophyllite and rectorite and by a significant decrease in the A1IV content of the chlorites. Towards the northwestern border of the basin, this sequence was affected by deep diagenetic conditions, as deduced from the clay mineral association and the ‘crystallinity’ values (0.57°Δ2θ). The changes in the clay mineral assemblages and ‘crystallinity’ data can hardly be explained in terms of the varying burial depth and are related to changes in the circulation of fluids associated with varying facies (modal composition, grain size). Permeability is regarded as the main factor controlling the circulation of these migratory fluids.

The surface/near-surface chemistry of allophane has been studied by X-ray photoelectron spectroscopy (ESCA) and the bulk material by 27A1 and 29Si solid-state NMR and other techniques. The surface/near-surface Si/Al ratio of allophane is c.1.0, similar to that for kaolinite, zeolite Na-A and sodalite. The core level binding energies for kaolinite and allophane are almost identical, but quite different from those for zeolite Na-A and sodalite, both framework aluminosilicates. The nature and size of these differences is consistent with the differences between the chemistry of sheet and framework silicates. The small variations in the Si(2p) spectra for kaolinite and allophane are discussed in terms of bonding of the tetrahedral units in the two materials.

Pecoraite and nepouite, Ni-serpentines, occur in the serpentinized ultramafic rocks in the Kwangcheon area, Korea, where the parent rock is classified as harzburgite and/or lherzolite. Pecoraite was precipitated twice from the solution; the early-formed pecoraite coexists with magnetite, millerite, and polydymite both in the Buk- and Nam-sites, while the late-formed pecoraite appears as well-grown colloform and opaque-free phase only in the Buk-site. The typical colloform texture of the late-formed pecoraite strongly indicates that it was precipitated from the solution in supergene conditions. Pecoraite is characterized by its extremely high Ni content and the difference in Fe content between the early- and late-formed pecoraite. Nepouite is distinguished from pecoraite by its prismatic morphology and the large degree of isomorphous substitution between Ni and Mg. The phase relations among coexisting magnetite-millerite-polydymite assemblage with the early-formed pecoraite suggest that the pecoraite might have precipitated in the extremely limited fO2 and fs2 environment from the highly Ni-concentrated solutions and is stable at 25°C and 1 bar.

: Smectite genesis in two granitic saprolites was studied, using qualitative and quantitative mineralogical methods and high resolution transmission electron microscopy (HRTEM). The primary minerals (plagioclase and mica) give rise to the formation of different kinds of smectites according to the weathering conditions (pH and drainage). Under a relatively basic, drained environment, a neoformed montmorillonite with a spheroidal organization exists in fillings of plagioclase cores. In the same conditions biotites are transformed into beidellites with a quasi-crystal organization. Under hydromorphic conditions, muscovites evolve incompletely towards smectites, whereas biotites lead to a montmorillonite with a tactoid organization.

Seven samples from a chronosequence of soils developed in historically created polders on the Atlantic coast (Marais Poitevin, Vendée, France) were investigated in order to illustrate the rate of mineralogical change in a clay-dominated system. The oldest polder was constructed in 1665, the last one in 1912; thus the time span of soil evolution is from 80 to 330 years. All the samples had more than 50% clay (<2 μm). The most reactive, fine clay sub-fraction (<0.1 μm) was investigated in detail by X-ray diffraction and chemical analysis. The observed mineralogical changes with increasing age followed the schematic reaction:

smectite + mica = illite + mixed-layer minerals.

The progress of reaction in time appears to be non-linear. This reaction seems to occur in a chemically constant system, and the mineralogical change can be seen as a readjustment of species to a given chemical composition.

Sepiolite from Vallecas (Spain) was treated with HCl solutions (1.25; 2.5; 5.0; 10.0 and 20.0 wt%) at 25°C for 2, 6, 24 and 48 h. The solutions obtained after washing the acid treated samples were analysed and a kinetic model was applied to the cation dissolution data. The development of porosity and number of acid centres was studied. The experimental conditions which yield products with the highest surface activity were established. The size of the silica fibres, obtained after the treatment, decreases when the intensity of the treatments is increased. These characteristics are important for the adsorptive properties and the catalytic potential of a solid.