Mixed-layer illite-smectites (I-S) have a stable charge on the illite interlayer equal to 0·89/O10(OH)2, as shown by electron microscope measurements of mean fundamental particle thickness, using both Pt-shadowing and high-resolution techniques. This has been verified by independent measurements of total surface area and CEC. Typically, the smectite interlayer charge is close to 0·4/O10(OH)2 but clays evolving in K-deficient environments may exhibit higher values. In the course of illitization, the silicate layer composition becomes more restricted than the smectite composition, AlIV and AlVI increasing while Fe and Mg decrease. These substitutions lead to the composition of non-expandable illite differing distinctly from that of muscovite and phengite, close to FIX0·89Al1·85Fe0·05Mg0·10Si3·20Al0·80. HRTEM data show that XRD systematically underestimates %S layers in I-S due to the small size of coherent scattering domains. An experimental curve is proposed for correcting XRD data. The above results were obtained for I-S formed from volcanic materials in different diagenetic and hydrothermal environments. The illite interlayer charge value close to 0·9 reconciles well the available data on I-S produced by wetting and drying of K-smectites.

The weathering of montmorillonite in the acid clay deposit at Itoigawa, Japan has been investigated by X-ray diffraction (XRD), infrared spectroscopy (IR) and scanning electron microscopy (SEM). In the lower, least weathered part of the deposit, weathering has occurred by the formation of minor amounts of halloysite with concomitant precipitation of a silica phase identified as opal C-T. A relatively rapid transition occurs to the most weathered, upper part of the deposit where the mineralogy is dominated by interstratified halloysite-smectite. Computer simulations of XRD profiles indicate that the interstratified arrangement is random with a probable composition of 70% halloysite and 30% smectite, a composition supported by the 75–85% and 15–25% of the phases, respectively, calculated from the IR spectra.

The weathering of biotite in Quaternary lahars from Monti Ernici (Central Italy) via interstratification stages has been studied by X-ray diffraction of both the interstratified minerals and the initial (biotite) and final (vermiculite) end-members. The structure of the biotite and vermiculite is described by means of monodimensional Fourier series. The interstratified phases have been studied using the INTER program which permits an analysis of two interstratified components by Fourier transform methods. It is shown that in the alteration range for biotite to vermiculite even small samples have different spatial concentrations of biotite, vermiculite and an intermediate biotite-vermiculite-type phase. Hydrobiotite and a rare biotite-hydrobiotite with a tendency to regularity have also been documented. Chemically, the process involves considerable oxidation of Fe2+ and the removal of most of the elements, corresponding to a 34% loss in matter. The sequence of the element losses is: K+ > Fe2+ > F− > Mn2+ > Si4+ > Mg2+ > Al3+ > Ti4+. There are also some gains in the order: H+ > Ca2+ > Fe3+ > Na+.

The expandability of K-depleted biotite and natural vermiculite was studied using transmission electron microscopy (TEM) and X-ray diffraction (XRD). K-depletion in layer-silicates was achieved by treating ultrathin sections with 0·1 m CaCl2 and BaCl2 solutions. The natural sample of biotite, which by XRD revealed no expandability on ethylene glycol or glycerol solvation, displayed 10–15% expanded layers when viewed by TEM after alkylammonium intercalation. The proportion of expanded layers increased after CaCl2 treatment. XRD of vermiculite samples revealed two sets of expandable interlayers after alkylammonium treatment, corresponding to two types of particles with different layer structures observed by TEM. Identification of vermiculite by TEM based on basal spacings is reliable with alkylammonium treatment. Intercalation of alkylammonium ions into the interlayers of vermiculite improved the degree of stacking order of the 2:1 layers.

Understanding the chemical variability of clay minerals depends on the analytical reliability of the techniques used. Uncertainties in clay mineral compositions were computed for common sources of analytical error such as those that arise from contaminant phases present in clay size-fractions in concentrations below the detection limit of routine XRD screening techniques, and the analytical limits of precision for TEM/AEM analysis. When plotted on standard diagrams used to represent clay mineral compositions, the calculated error envelopes were found to be of significant size, such that analyses of pure illites and pure smectites straddled the field for mixed-layered illite-smectites. In addition, on some diagrams random analytical errors resulted in linear trends identical to those caused by mixtures of two or more clay minerals. It follows that variations in the compositions of clay minerals reported in the literaure do not always necessarily represent the actual chemical variation of the minerals. As a result, current interpretations of clay compositional variations may not be as definitive as hoped.

Rhaetian (Upper Triassic) sandy horizons were sampled from the Paris Basin from SW to NE, crosscutting the Rhaetian at different depths from outcrop in the NE to 2700 m in the centre of the basin. The smallest day sub-fractions (<0·2 μm) from the deepest samples consist mainly of illite and chlorite having a K-Ar age of 〜190 Ma. Both minerals probably formed under specific hydrothermal conditions at high temperature, but at a burial depth of only 500 m. This thermal event could represent an echo of the "crustal" breakdown of the Northwestern European craton during the opening of the North Atlantic Ocean. Two other generations of illite-smectite mixed-layers formed in the same Rhaetian horizons at somewhat lower temperatures about 150 and 80 Ma ago. The three generations of clay minerals could be characterized and dated because of combined mineralogical, crystallographical and morphological data supporting the dating attempts.

Iron-rich dioctahedral 2:1 clays were sampled in metalliferous sediments from Atlantis II Deep (Red Sea), and studied by XRD, TEM, FTIR and Mössbauer spectroscopy. One sample with the structural formula (Si3·94Fe3+0·06)(Fe3+1·90Mg0·10)K0·1(Mg, Ca)0·03 is closer to the theoretical end-member ferripyrophyllite Si4Fe3+2O10(OH)2 than any other Fe-rich clay previously described. Other samples appear as mixed-layered clays: ferripyrophyllite-hydroferripyrophyllite or hydroferripyrophyllite-celadonite.