The boundary conditions of saponite formation are generally considered to be well known, but significant gaps in our knowledge persist in respect to the influence of solution chemistry, temperature, and reaction time on the mineralogy, structure, stability, and chemical composition of laboratory-grown ferrous saponite. In the present study, ferrous saponite and Mgsaponite were synthesized in Teflon-lined, stainless steel autoclaves at 60, 120 and 180°C, alkaline pH, reducing conditions, and initial solutions with molar Si:Fe:Mg ratios of 4:0:2, 4:1:1, 4:1.5:0.5, 4:1.75:0.25, and 4:1.82:0.18. The experimental solutions were prepared by dissolution of sodium orthosilicate (Na4SiO4), iron(II)sulfate (FeSO4·6H2O) and magnesium chloride salts (MgCl2·6H2O with ≤ 0.005 mass% of K and Ca) in 50 mL ultrapure water that contained 0.05% sodium dithionite as the reducing agent. The precipitates obtained at two, five and seven days of reaction time were investigated by X-ray diffraction techniques, transmission electron microscopy analysis, infra-red spectroscopy, and thermo-analytical methods.

The precipitates were composed mainly of trioctahedral ferrous saponite, with small admixtures of co-precipitated brucite, opal-CT, and 2-line ferrihydrite, and nontronite as the probable alteration product of ferrous saponite. The compositions of the obtained ferrous saponites were highly variable, (Na0.44−0.59 K0.00−0.05 Ca0.00−0.02) (Fe2+0.37−2.41 Mg0.24−2.44 Fe3+0.00−0.28)Σ2.65−2.85 [(Fe3+0.00−0.37 Si3.63−4.00)O10](OH)2, but show similarities with naturally occurring trioctahedral Fe and Mg end members, except for the Al content. This suggests that a complete solid solution may exist in the Fe-Mg-saponite series.

A conceptual reaction sequence for the formation of ferrous saponite is developed based on the experimental solution and solid compositions. Initially, at pH ≥ 10.4, brucite-type octahedral template sheets are formed, where dissolved Si-O tetrahedra are condensed. Subsequent reorganization of the octahedra and tetrahedra via multiple dissolution-precipitation processes finally results in the formation of saponite structures, together with brucite and partly amorphous silica. The extent of Fe2+ incorporation in the octahedral template sheets via isomorphic substitution is suggested to stabilize the saponite structure, explaining (i) the abundance of saponite enriched in VIFe2+ at elevated Fe supply and (ii) the effect of structural Fe on controlling the net formation rates of ferrous saponite.

A geochemical and mineralogical study has allowed us to address the factors controlling distribution pattern, residence and behaviour of rare earth elements (REE) during kaolinization of Variscan granitoids in NW Spain. Mineral composition of the deeply weathered samples is dominated by kaolinite, with minor amounts of quartz, muscovite-illite, alkaline feldspar and traces of resistant minerals (rutile, ilmenite, zircon and monazite). Variable amounts of Si, Na, Ca, K, Rb, Cs, Ba, U and P were lost from the weathering profile, as a result of feldspars, mica and apatite breakdown, whereas Al, Fe, Ti, Zr, Th, Hf and REE were concentrated in the residual kaolin. Chondrite-normalized REE patterns of the kaolins show an overall enrichment of light REE (LaN/SmN = 1.22–2.53), heavy REE depletion (GdN/YbN = 2.42–15.10) and a strong negative Eu anomaly (Eu/Eu* = 0.11–0.25), probably inherited from the parent granite. Nevertheless, the normalization to the parent granite reveals some REE fractionation and increasing positive Eu anomalies with advancing weathering, in response to the breakdown of feldspars. Different grain-size fractions show similar REE distribution patterns, but differ in concentration levels. Although the fine fractions are the most important REE reservoir, there is no positive correlation with clay mineralogy. The correlative behaviour among P2O5, Th and REE in the <2 mm fraction suggests that monazite plays a dominant role controlling the REE budget in the weathering profile.

Under certain circumstances, both glycerol and ethylene glycol form interlayer complexes with vermiculites resembling those formed with the smectites. The rate and degree of complex formation are related to the surface charge on the silicate layers, to the type of interlayer cation present and to crystal size. With glycerol, but not with ethylene glycol, the hydration state of the mineral immediately prior to complex formation is of some importance.

The nature and origin of authigenic clay minerals and silicate cements in the Jurassic and Cretaceous sediments of England and the North Sea are discussed in relation to penecontemporaneous volcanism in and around the North Sea Basin. Evidence, including new REE data, suggests that the authigenic clay minerals represent the argillization of volcanic ash under varying diagenetic conditions, and that volcanic ash is a likely source for at least the early silicate cements in many sandstones. The nature and origin of smectite-rich, glauconite-rich, berthierine-rich and kaolin-rich volcanogenic clay mineral deposits are discussed. Two patterns of volcanogenic clay minerals facies are described. Pattern A is related to ash argillization in the non-marine and marine environments. Pattern B is developed by the argillization of ash concentrated in the sand and silt facies belts in the seas bordering ash-covered islands and massifs. It is associated with regression/ transgression cycles which may be related to thermal doming and associated volcanism, including the submarine release of hydrothermal fluids rich in Fe. The apparent paucity of volcanogenic clay deposits in the Jurasssic and Early Cretaceous sediments of the North Sea is discussed.

The distribution of diagenetically-formed clay minerals in Rotliegend dune sandstones of the Southern North Sea Basin is closely related to the following factors: (i) the petrography of the sandstones, (ii) the paleoburial depth and tectonic setting of the area, (iii) the thickness of gas-generating Carboniferous strata underlying the Rotliegendes, and (iv) the facies distribution of the overlying Zechstein. The diagenetic clay minerals are mainly conversion products of feldspars and, to a lesser extent, of detrital clays and micaceous lithic fragments. Sandstones containing dominant drusy illitic and chloritic clay minerals have been buried to depths > 3000 m; if kaolinite is the dominant clay mineral, burial depths were less. Sandstones containing feldspars (detrital and authigenic) up to approximately 7% of bulk volume have permeabilities that are about four times higher than sandstones with similar amounts of kaolinite, and as much as 200 times higher than sandstones with similar amounts of illite and/or chlorite.

The surface free energy of Ca-montmorillonite has been determined from measurement of the contact angle of water on an oriented deposit immersed in n-alkanes; the oriented deposits were equilibrated at different relative humidities. These experiments provided the polar, γSP, and dispersive, γSD, components of the surface free energy γs. The results indicate that: (1) γs is mainly due to dispersive forces (γSD ≃ 145 mJ.m−2) and, to a lesser degree, polar forces (γSP ≃ 35 mJ.m−2). (2) The value of the apparent surface free energy of the clay, γα, tends towards that measured for water when the solid water content exceeds 50 wt%. (3) The sorption of water molecules substantially modifies the surface free energy of the solid; γs decreases from 180 mJ.m−2 to 60 mJ.m−2 when the water content increases from 0 to 50%. These results agree with other observations made on solids whose surface characteristics are similar to montmorillonite, i.e. micas, silicates and glasses. In addition, variations of γs with water content can be related to the hydration processes of smectites. Finally, these results show that the silicate no longer influences the surface properties when the montmorillonite water content exceeds 60%.

The smectites present in some bentonites from the Island of Milos, Greece, derived from acidic rocks, have been examined by XRD and electron microprobe. The possible influence of loss of alkalis during analysis and contamination by fine grained Si-phases on microanalyses has also been evaluated. Smectites are closely associated with opal-CT and their composition ranges from beidellite to Tatatilla-type montmorillonite. Their chemistry is dominated by negative relationships between Mg and AlVI and Si and total Al. However, it cannot be resolved whether there is a solid-solution between the smectites or whether the variation observed is due to mechanical mixing of domains with different composition.The role of Fe is of minor importance reflecting an inherited factor imposed by the parent rock, but the role of the fluid phase is also important. Beidellite occurs in the rims, while Tatatilla-type montmorillonite, often associated with free opal-CT, is formed in the interior of the altered shards, indicating a variable degree of Si release during glass dissolution. The formation of beidellite is not necessarily associated with Mg uptake.

Pillared smectites in which the pillars consist of iron oxide are expected to have interesting and unusual magnetic properties. Several possible routes by which such materials might be made have been investigated, namely intercalation of hydroxy-Fe(III) polycations, mixed hydroxy-Fe(III)/Al polycations, phenanthroline-Fe(II) cations, and trinuclear Fe(III) acetato cations into Na-montmorillonite. Only the last of these yielded a pillared clay (PILC) on calcination. The products have been characterized using X-ray powder diffraction and 57Fe Mössbauer spectroscopy. The precursor Fe-PILC has a d-spacing of 21 Å and expands to 23 Å on solvation with glycol. The calcined Fe-PILC has a d-spacing of 19 Å (gallery height 9·4 Å) and does not expand with glycol, confirming cross-linking of the layers. From Mössbauer spectra at 4·2 K it is estimated that there are of the order of some hundred Fe atoms per pillar.

The Kämpf & Schwertmann (1982) procedure for concentrating iron oxides in soil clays by dissolution of kaolin and gibbsite by boiling for 1 h in 5 m NaOH may not dissolve all kaolin, and also results in the precipitation of sodalite. For the complete dissolution of kaolin in kaolin-rich soil clays a boiling time of 2 h in 5 m NaOH was required. The large amounts of sodalite produced were not removed by the prescribed single wash in 0·5 m HCl. Oxalate soluble Al contents of iron oxide concentrates were sometimes very high and dithionite Fe contents were very low both in concentrates containing sodalite, and in those for which sodalite was not detected by XRD, but where a previously unsuspected amorphous sodalite-like phase may have been present. Complete removal of precipitated sodalite was achieved by two extractions with 0·5 m HCl at 25°C for 20 min. This modified procedure does not alter the Al-substitution and crystal size of goethite, hematite and maghemite as determined by XRD measurements.

High-resolution gas adsorption techniques were used to analyse the evolution of the aspect ratio and adsorption energy distribution on synthetic saponite samples with increasing layer charge. Using Ar as a gaseous probe, the aspect ratio of the saponite particles can be determined easily by decomposing the derivative adsorption isotherms and taking into account high-energy sites which can be assigned to talc-like ditrigonal cavities. Changes in the shape of the elementary particles are observed for layer charges above 1.30, i.e. when all the ditrigonal cavities contain at least one Al atom substituting for Si. When N2 is used as a probe, high-energy sites that could be wrongly interpreted as micropores on the basis of classical t-plot treatments are observed whatever the layer charge. Using the information obtained from both Ar and N2, schemes for describing adsorption can be proposed for all layer charges and suggest complex adsorption mechanisms for charged clay minerals.

The clay mineral composition is one of the factors that affect the physical properties of soils, and a knowledge of it is required to promote a fuller understanding of the origin of these properties. The relationships between the clay content and the plastic and liquid limits of natural montmorillonitic and kaolinitic soils and of artificial mixtures have been examined and compared. Factors affecting the relationships are discussed, and illustrated by the effect of particle aggregation on the measurement of the liquid limit of tropical red clays and on the sedimentation analysis of the Keuper Marl. The effect of muscovite and of silt-sized material on the position of soils on the Casagrande classification chart has also been examined.

The effects of dry grinding on the structure of sepiolite were studied by XRD, surface area measurement, TG-DTA and IR spectroscopy techniques. The results show that the sepiolite framework is rather resistant to mechanical stress and its alteration by dry grinding occurs through three successive, overlapping processes: (i) thinning of the fibrous particles without structural alteration; (ii) distortion of the parallel units of sepiolite fibres, affecting essentially the particle surface which transforms into an amorphous protective coating; (iii) disruption of the inner structure and a prototropy effect, producing an amorphous phase with some residual order.

A method is described for plotting and comparing measurements made on kaolinite clays. In this the ratio of the two strongest bands in the OH-stretch region of the spectrum (band maxima at 3695–3700 cm−1 and 3620–3625 cm−1) of an oriented preparation at 45° incidence is plotted against the corresponding ratio from the spectrum of a randomly oriented preparation. Both ratios are affected by the shape of the kaolinite particles and methods for minimizing this are given. The position of a kaolinite upon a plot so prepared gives information on the following points: (a) the type of kaolinite, (b) the degree of crystallinity of the kaolinite, (c) the shape of the kaolinite particles.

Temperature-programmed desorption (TPD) and IR spectroscopy were used to characterize the number and strength of acid sites in Al3+-, Cr3+- and Fe3+-exchanged montmorillonite. The bases pyridine and n-butylamine occupied three different sites in the interlamellar space: (i) physisorbed base, (ii) base bound to Lewis acid sites, and (iii) protonated base. TPD profiles for pyridine were characterized by maxima at 40°, 150° and 340°C, whilst those for n-butylamine occurred at 30°, 200° and 410°C. The Al3+- and Cr3+-exchanged forms were stable up to pretreatment temperatures of 300°C, but the Fe3+-form required > 3 day exposure to base vapour to re-establish the high-temperature desorption peak. Variable-temperature IR studies showed that the number of Brönsted-bound pyridine molecules increased with increased outgassing temperature.

Silico-ferric coprecipitates, with chemical formula , were aged in suspension at 75°, 100° and 150°C and the structural evolution of solids with time studied by XRD, TEM, and IR, Mössbauer and EXAFS spectroscopy. The initial Si-Fe coprecipitate was found to be amorphous but showed local order similar to that of a smectite layer. At 75°C only a weak structural evolution of the silico-ferric product towards a smectite-like structure was observed. Experiments performed at 100° and 150°C led to synthesis of a ferric smectite with structural formula . During syntheses a highly soluble silico-ferric complex appeared; the Si/Fe atomic ratio of this complex was 3, and the apparent concentration of Fe3+ in solution reached 27 mm/l. These syntheses prove that the crystallization of a dioctahedral smectite, containing only Fe3+ atoms in the octahedral sheet, is possible under strictly oxidizing conditions. However, crystal growth of a ferric smectite under these conditions is slow and only syntheses carried out at sufficiently high temperatures give convincing results.

Nous proposons l'étude du mècanisme de gonflement d'une smectite, basée sur l'exploitation des diagrammes de diffusion aux petits angles obtenus avec la source de rayons X synchrotron de Lure (Orsay). Cette technique a permis d'étudier le gonflement, en fonction de la température, d'une montmorillonite Na, en suspension dans l'eau et concentrée à 20%.

L'analyse des diagrammes est éffectuée par une méhode basée sur la comparaison des courbes expérimentales à des courbes théoriques. Les modèles de structure permettant le calcul des diagrammes théoriques fait appel à la notion de particule (définie comme un empilement de feuillets parallèles) et de structure interne de la particule (définie par un ensemble de translations di affectées des probabilités pi qui déterminent la position mutuelle des feuillets).

Les résultats montrent qu'à basse température, les particules de très grande taille sont constituées par la superposition de sous empilements, comportant 4 à 5 feuillets, hydratés d'une manière homogène et se désorientant mutuellement pour former la structure du gel. On explique ainsi que la structure du gel formé est correlée avec celle du solide initial et que le phénomène de gonflement est quasi réversible.

During the reaction of solutions of the leuco dyes crystal violet lactone and N-benzoyl leuco methylene blue with acid-treated montmarillonite (as occurs with carbonless copying papers), most of the offered dye is intercalated within the interlayers of the montmorillonite structure remaining after acid attack. This phenomenon was demonstrated by XRD and also by high-resolution transmission electron microscopy. During acid treatment, octahedral cations are dissolved from exposed edges of the montmorillonite lamellae and amorphous silica builds up in these areas. As a result, the BET(N2) specific surface area increases from ∼60 m2/g to ∼300 m2/g and, consequently, the theoretical reactivity towards leuco dyes. Nevertheless, only small quantities of the dye molecules are adsorbed onto the amorphous silica. The excellent colour intensity and, in particular, the resistance to fading in the light of the montmorillonite dye complexes may be explained by the protected position of the dye molecules within the interlayers of the montmorillonite.

The pillaring of a Laponite by alumina pillars, iron pillars and mixed Al-Fe pillars was obtained by a simple in situ method starting from AlCl3 and/or FeCl3 solutions with different molar ratio Fe/(Fe + Al), and from NaOH solution added simultaneously to the clay suspension. The Al13 pillars were evidenced by XRD and NMR spectra. The density and the charge of these pillars were calculated from the chemical analysis. The Fe pillars (in samples with Fe content >26%) are larger than Al13 pillars and slowly reducible by hydrogen temperature programmed reduction (H2TPR), but we do not have any information on their structure. There is also an excess of non-pillar Fe retained by the pillared clays probably as oxyhydroxides. The isomorphic substitution in the mixed (Al-Fe)13 pillars (in samples with an Fe content >26%) was evidenced by H2TPR results compared to the chemical analyses and by the interpretation of the 27A1 peaks in the NMR spectra.

Aluminous hydroxy 2/1 clay minerals (‘intergrades’) can form in two ways: Al can either come externally from the solution, or internally from the mineral itself (octahedral or tetrahedral layers). This study is more concerned with the latter which seems the main source in acid soils.

The most common organic acids present under natural conditions were used on two trioctahedral micas (phlogopite and biotite), and their chemical and mineralogical weathering effects were determined.

Different kinds of mineralogical evolution are possible: mineral destruction by a few very complexing acids, possibly with intermediate stages (interstratified minerals); formation of ‘transformation’ smectite by acids derived from sugars; Al-hydroxy vermiculite formation (the 14 Å diffraction peak of which is stable under K saturation and vacuum). This evolution is the most frequent and is produced by the most numerous organic acids (even CO2), the acidity of which results in Al migration in the interlayer, but the complexing ability of which is not sufficient for Al removal.

An Fe-rich smectite from Ölberg (Germany) was characterized using X-ray diffraction, infrared spectroscopy and analytical electron microscopy. Progressive reduction of the octahedral charge was performed through the Hofmann & Klemen effect at increasing temperatures. The sample was heterogeneous, consisting of two smectite populations. One population, which comprises a minor portion of the sample, is an ‘Fe3+-montmorillonite’ with little or no tetrahedral charge and Fe3+ as the major octahedral cation. The other population, a major constituent of the sample, contains less Mg and more Al than the first one and exhibits some tetrahedral charge. This fraction may be considered as an inter-grade between nontronite (dominant tetrahedral charge) and Fe3+-montmorillonite (dominant octahedral charge). These two populations may occur as separate particles but also as interstratified layers.