In spite of decades of research on the subject, the crystal structure of illite is still poorly understood. The purpose of this study was to address this problem by investigating the nature of the interlayer content in illite IMt-2 from Silver Hill, Montana, using analytical transmission electron microscopy (ATEM), thermogravimetry (TG), and X-ray powder diffraction (XRPD) analyses. The ATEM data, together with literature and TG results, yielded the formula K0.70a0.01(H2O)0.42 (Al1.53Fe2+0.06Fe3+0.19Mg0.28)Σ−2.06(Si3.44Al0.56)O10(OH)2 or, assuming the presence of H3O+, K0.69Na0.01(H3O)+0.28(Al1.47Fe2+0.06Fe3+0.19Mg0.28)Σ−1.99(Si3.40Al0.60)O10(OH)2. The first formula indicates surplus interlayer and octahedral species, whereas the second shows no excess. The XRPD data were refined by Rietveld techniques, down to an Rp factor of 10.48–13.8%. The mineral composition consists largely of illite-2M1, illite-1M, and minor quartz. Although the refinement accuracy is limited by the intrinsic poor quality diffraction of the illites, the partially refined model is consistent with the chemical composition; in particular, attempts to introduce octahedral cations in excess of 2 were fruitless. All the results support the simple structural model, by which the illite structure strictly corresponds to a dioctahedral mica with H3O+ replacing K. As a consequence, the crystalchemical formula of illites should be calculated on the basis of six tetrahedral plus octahedral cations.

The purpose of this study was to investigate bonding mechanisms of representative inorganic anions and citrate with imogolite and allophane using electrophoresis. The electrophoretic mobility (EM) of synthetic imogolite and allophanes with Al/Si molar ratios of 2.02, 1.64, and 1.26 was determined in 0.001 and 0.01 M sodium solutions. The highest point of zero mobility (PZM) values for imogolite and the highest point of zero charge (PZC) values for allophane occurred in the presence of ClO4, NO3, Br, I, and Cl. Below the PZM and PZC, Cl and I lowered the EM relative to the other anions but did not shift the PZM and PZC significantly. This indicates that Cl and I formed more outer-sphere complexes than the other ions. The EM of imogolite and allophane was negative at pH < 6 in 0.001 and 0.01 M NaF probably due to a phase change. We observed the formation of cryolite (Na3AlF6) with transmission electron microscopy (TEM) and X-ray diffraction (XRD) in the NaF systems at low pH. Conversely, phosphate at 0.001 and 0.01 M concentrations lowered both the PZM and the EM in imogolite and both the PZC and the EM in allophane compared with ClO4. Phosphate-treated allophane had the same PZC as a synthetic amorphous aluminum phosphate. The PZM values of imogolite and allophane with 2:1 Al/Si in 0.0001 M Na-citrate were 10.9 and 5.9, respectively. At pH 7.3, Na-citrate lowered allophane EM more than it lowered imogolite EM relative to ClO4.

The EM in NaClO4 and Na2SO4 was reversible by forward- and back-titration with NaOH and HClO4, indicated that ClO4 and SO4 were not specifically adsorbed. Chloride likely formed more outer-sphere complexes than ClO4. Imogolite EM and allophane EM in dilute NaF and NaH2PO4 solutions were not reversible, indicating either surface inner-sphere complexes or surface precipitates of aluminum fluoride and amorphous aluminum phosphate-like materials on these minerals. Sulfate gave a lower EM than the monovalent anions, implying a greater tendency to form outer-sphere complexes. Citrate appeared to form inner-sphere complexes on both imogolite and allophane, but formation was concentration-dependent. The tendency of anions to form surface complexes with imogolite and allophane is consistent with the tendency of anions to form soluble aluminum complexes.

Chlorite is a layered silicate mineral group of importance in geology, agriculture, and in the processing of mineral resources. A more detailed analysis of the surface charge of chlorite minerals is important in order to improve our fundamental understanding of such particle structures and their behavior in suspension. In this study, the anisotropic surface charging of chlorite has been established using Atomic Force Microscopy surface-force measurements with a silicon nitride tip. The surface-charge densities and surface potentials at the chlorite basal-plane surfaces and edge surface were obtained by fitting force curves with the Derjaguin-Landau-Verwey-Overbeek theoretical model. The results show that at pH 5.6, 8.0, and 9.0 the chlorite mica-like face is negatively charged with the isoelectric point (IEP) less than pH 5.6. In contrast, the chlorite brucite-like face is positively charged in this pH range and the IEP is greater than pH 9.0. The surface charging of the chlorite edge surface was found to be pH-dependent with the IEP occurring at pH 8.5, which is slightly greater than the edge surfaces of talc and muscovite due to the larger content of magnesium hydroxide at the chlorite edge surface. Findings from the present research are expected to provide a fundamental foundation for the analysis of industrial requirements, e.g. collector adsorption, slime coating, and particle interactions in the area of mineral-processing technology.

The size and shape of clay particles in aqueous suspension have been determined by means of five experimental techniques—electro-optical birefringence, ultracentrifugation, viscometry, optical transmission and electron microscopy. Five clay minerals—montmorillonite, illite, hectorite, nontronite and attapulgite—each subdivided into five size fractions, were studied. The average particle dimensions in each fraction are tabulated.

An evaluation of the experimental techniques is presented. It is concluded that for a preliminary survey of the particle size and shape of clay minerals the technique of electron microscopy is best. However, once a model for the particles is obtained, a combination of the techniques of electro-optical birefringence and viscometry becomes more useful in characterizing the size and shape of the particles in suspension.

Three samples of volcanic tuff were hydrothermally altered at ∼82°C in a soxhlet apparatus for periods from 745 to 2706 h. The samples correspond to partially altered specimens of volcanic tuff with 6 wt. % (T3a) and 9 wt. % (T3b) smectite and to the calcination product of the latter (T3c). The calcination treatment melted the smectite in the sample. Untreated samples and alteration products were studied by X-ray diffraction (XRD), differential thermal analysis (DTA) and thermogravimetry (TG), scanning electron microscopy (SEM) with energy dispersive X-ray (EDX) analysis, and oxygen isotope analysis. DTA-TG of the <2-µm size fractions showed that there was a small increase of smectite relative to glass after alteration for samples T3a and T3b, with the amount of smectite increasing exponentially with time. No smectite formed in sample T3c. These results suggest that smectite acts as a nucleation site for the precipitation of new smectite. The amount of glass in the <2-µm size fraction increased, although slightly decreased relative to smectite. SEM-EDX analysis showed smectite with the approximate structural formula of Na0.22K0.08Mg0.12Ca0.03VI(Al1.47Fe0.05Mg0.48)IV(Si3.97Al0.03)O10(OH)2.

Oxygen isotope composition of the <2-µm size fraction became enriched in 18O by alteration, the >2-µm size fraction of T3b did not vary, and that of T3c was depleted in 18O. Our results are consistent with three processes during alteration: 1) oxygen isotope exchange between volcanic glass and water, 2) neoformation of smectite, and 3) hydration and consequent hydroxylation of the calcined glass.

Adsorption of Co2+ nitrilotriacetic acid (NTA) and equal-molar Co2+ and NTA by a low surface area (LSA) commercial gibbsite (3.5 m2 g−1) was investigated in batch as a function of pH (4.5 to 10.5), adsorbate (0.5 to 10 µM) and adsorbent (0.5 to 75 g L−1) concentrations and ionic strength (0.01 to 1 M NaClO4). The adsorption of Co2+ (Co-only) and the acid form of NTA (NTA-only) by gibbsite in 0.01 M NaClO4 exhibit cation-like and anion-like adsorption edges, respectively. For the equal-molar CoNTA chelate, Co and NTA adsorption edges were similar but not identical to the Co-only and NTA-only edges. Differences suggest the existence of a ternary CoNTA surface complex with the Co in the intact chelate coordinated to surface hydroxyls. NTA-only adsorption was insensitive to ionic strength variation, indicating weak electrostatic contributions to surface coordination reactions. This is consistent with the formation of inner-sphere surface NTA complexes and ligand exchange reactions in which monodentate, bidentate and binuclear NTA surface complexes form. Cobalt adsorption increases (edge shifts to lower pH by 1 pH unit) on LSA gibbsite as ionic strength increases from 0.01 to 1 M NaClO4. For the same ionic strength change, a similar shift in the Co-only edge was observed for another commercial gibbsite (16.8 m2 g−1); however, no change was observed for δ-Al2O3. Ionic strength shifts in Co2+ adsorption by gibbsite were described as an outer-sphere CoOH+ surface complex using the triple-layer model. Results suggest that, at waste disposal sites where 60Co and NTA have been co-disposed, NTA will not promote ligand-like adsorption of Co for acid conditions, but will reduce cation-like adsorption for basic conditions. Thus, where gibbsite is the dominant mineral sorbent, NTA will not alter 60Co mobility in acidic pore waters and groundwaters; however, NTA could enhance 60Co mobility where alkaline conditions prevail, unless microbial degradation of the NTA occurs.

Diffractograms of lepidocrocites formed from the oxidation of Fe(II) chloride solutions at pH 7 and 25°C showed differential line broadening with (0k0) reflections being much sharper than (hk0) or (hkl). As the initial [Cl]/[Fe] ratio was increased from 2, crystallinity increased in all axial directions, as defined by the width at half height of diffraction peaks. At [Cl]/[Fe] ratios greater than ~8, the crystallinity did not increase further, and for reflections involving the a-axis direction, it appeared to decrease. Increased [Cl]/[Fe] ratios changed the rate of the oxidation/hydrolysis reactions. These variations reflect changes in the stability and/or composition of the intermediate green rust phase, (Fe(II)Fe(III) hydroxychloride) in the high [Cl] environment. The influence of the [Cl]/[Fe] ratio was related to the experimental conditions employed, especially where they influenced the rate of oxidation.

The catalytic activity of sepiolite from Amboseli, Tanzania, for the dehydration and dehydrogenation of ethanol at 150°–300°C has been studied using a flow reactor. Both reactions occur, but the catalyst activity decreases with use. The products include water, carbon dioxide, ethene, ethanal (acetaldehyde), diethyl ether, but-1,3-diene, but-2-enal (crotonaldehyde), and an unidentified aromatic compound. The proportions change with temperature, the dehydrogenation reaction being favored at the higher temperatures. The BET surface areas of the sepiolite are 316 m2/g (nitrogen adsorption at — 197°C) and 212 m2/g (ethanol vapor adsorption at 25°C, assuming a molecular cross-sectional area of 24.6 Å2), indicating a possible greater penetration of pores and channels by nitrogen compared with ethanol vapor under these conditions. The pore-size distribution reveals that approximately 55% of the surface area measured by nitrogen adsorption is contributed by micropores.

Sorption and desorption of two neutral, nonpolar organic compounds, 1,2-dibromoethane (EDB, a soil fumigant) and trichloroethene (TCE, an industrial solvent and common ground-water contaminant), by pyrophyllite, kaolinite, illite, and smectite were investigated. For sorption, vapors of the compounds in a stream of dry N2 gas were passed through columns of the powdered clay minerals for different periods of time. The compounds retained by the clays were extracted with methanol and analyzed by gas chromatography (GC). For desorption, N2 was passed through the treated samples, and the desorbed compounds were collected in hexane traps and analyzed by GC.

Initially sorption was rapid for several hours but then proceeded at a slower rate for many hours. Surprisingly, the clays sorbed large quantities of these neutral compounds; for example, the pyrophyllite, kaolinite, illite, and smectite sorbed about 3, 5, 6, and 9% EDB, respectively, by weight. The amounts of TCE sorbed, which is more volatile than EDB, were somewhat less. Only a portion of the compounds sorbed over a period of time were desorbed in the same period of time. Desorption was rapid initially but then proceeded at a slow rate; slow desorption continued for over 100 hr from samples which had undergone sorption for only 1–2 hr. A two-compartment efflux model was used to describe the sorption behavior. A rapidly desorbing component was considered to be present on the outer surfaces of the clay aggregates and a slowly desorbing component, in the interior pores of the aggregates.

The relative growth rates of the three joint chains of silica tetrahedra and metal octahedra in the [100], [1̅10] and [1̅1̅0] directions within the mica layer (referring to the 1M unit-cell) seem to control the morphology of mica crystallites. Laths and fibers are the products of relatively fast growth along the [100] direction compared to growth along the [1̅10] and [1̅1̅0] directions. The (010) growth front in 1M micas with trans-octahedral vacancies exposes a pair of reactive OH ions that can form organic or inorganic complexes and ‘poison’ the growth on the (010) face.

Authigenic illite fibers in two sandstones with contrasting lithologies are found to have grown on mica or kaolinite cores. Illite fibers appear in single sets or in multiple sets, 120° apart. This texture seems to be related to the stacking sequence of the layers in mica or kaolinite in the core of these fibers.

Birnessite can be used as a precursor in the preparation of manganese oxides. Synthesis of pure birnessite is difficult because of a side reaction, which yields hausmannite. This study aimed to develop a modified oxidation-deprotonation reaction (ODPR) method to eliminate the formation of hausmannite, and to investigate the influence of alkalinity on the synthetic products. In contrast to the conventional synthesis of birnessite through oxygen or permanganate oxidation, the ODPR method can produce birnessite without any impurities, within 5 h, and in a reproducible fashion. The distinctive feature of the ODPR method is the bubbling of N2 gas into NaOH and Mn2+ solutions before mixing the NaOH with Mn2+, in order to keep oxygen away from each solution. As soon as white pyrochroite was formed, oxygen gas was forced in as an oxidant to initiate the oxidation reaction. A black suspension with a blue tint appeared after 5 h of reaction. These precipitates were collected and examined by X-ray diffraction (XRD), scanning electron microscopy (SEM), infrared (IR) and Raman spectroscopy. The average oxidation state of the Mn oxides was also determined. The results showed that pure birnessite with good crystallinity was formed. Oxidation of 1 M NaOH mixed with Mn2+ solution formed random-stacked birnessite. However, the oxidation of 4 M NaOH mixed with Mn2+ formed birnessite. Random-stacked birnessite can be transformed into birnessite by ageing suspensions at 313 to 373 K.

The hydrolysis of 2-(2-trichlorosilylethyl)pyridine (TCSEP) or 3-aminopropyltrimethoxysilane (APTMS) yields a complex mixture of oligomers, the composition of which is a function of time. 29Si NMR measurements show that the same oligosilsesquioxanes can be obtained from both compounds after prolonged hydrolysis. Ion exchange by montmorillonite with the pillaring solution obtained from TCSEP results in partial intercalation, with a noticeable fraction of the clay exchange sites occupied by protons not by the Si cationic species. Calcination of this pillared clay results in a mixture of phases showing 1.86 or 1.56 nm basal spacings. Pillaring of the same montmorillonite by the solution obtained by partial hydrolysis of APTMS results in a homogeneous solid with a basal spacing of 2 nm, which reduces to 1.65 nm upon calcination above 773 K. The microporous volume (pores < 1 nm) of these Si-clays, determined after calcination at 773 K, 0.15 ml/g for the PILC obtained from TCSEP, and 0.186 ml/g for the PILC obtained from APTMS is comparable to that measured for Al or Zr pillared clays. A better thermal stability is observed for the clay pillared by APTMS, which retains a large microporosity (0.167 ml/g) up to 973 K in air. These Si-pillared clays show a small number of weakly acidic sites and some strongly acidic sites retaining ammonia up to 723 K, which are most likely localized on the clay layers.

Gabbroic saprolite at Mont Mégantic, Quebec, was studied in detail mineralogically to gain a better understanding of the origin of ferruginous smectite reported previously from these rocks. The parent rock is a ferrogabbro composed of plagioclase, augite, calc-alkalic amphiboles, biotite, olivine, magnetite, ilmenite, and apatite. Extensive weathering has decomposed most of the mafic minerals and magnetite to goethite, lepidocrocite, and iron-rich clay minerals, which occur in numerous microcracks distributed irregularly in the outer shells of boulders and in cracks and fissures in the bedrock. Some of the felsic minerals have altered to kaolinite. The secondary minerals and the more resistant primary minerals, such as plagioclase, ilmenite, and apatite, have subsequently moved to the lower part of the saprolite. The major ferruginous clay minerals present are smectite and vermiculite, which are compositionally similar, except that the smectite is slightly richer in SiO2 and MgO and poorer in Fe2O3 than the vermiculite.

To establish a structural formula for the ferruginous smectite, the oxidation state of Fe in a sample treated with dithionite-citrate-bicarbonate was examined by Mössbauer spectroscopy. All structural iron was found to be ferric. The calculated structural formula of a Na-saturated sample is:

$$N{a_{0.61}}\left( {M{g_{1.39}} + Fe_{0.85}^{3 + }A{l_{0.17}}M{n_{0.03}}} \right)\left( {S{i_{3.49}}A{l_{0.51}}} \right){O_{10}}{\left( {OH} \right)_2},$$

X-ray photoelectron spectroscopy (XPS) has been used to characterize the bonding state of Cu2+, Si4+, Al3+, and O2− ions in structural (octahedral and interlamellar) or adsorbed position in phyllosilicates. Five smectites, 5 kaolinites, and 1 chrysocolla with Cu(II) in known positions (octahedral, interlamellar, or surface adsorbed) have been investigated. Their spectra were compared with those of pure Cu metal and of pure Cu(I) and Cu(II) oxides.

The line for Cu 2p3/2 (binding energy of 935.4 eV) and well-defined shake-up lines (binding energy of about 943 eV) observed after 1 hr of X-ray irradiation are characteristic of Cu(II) in phyllosilicate octahedral sites. But due to the photoreduction effect, they show Cu(I) oxidation states (Cu 2p3/2, binding energy of 933.2 eV and near absence of shake-up lines) for the phyllosilicates with adsorbed Cu or in interlamellar positions. The kinetics of photoreduction distinguishes octahedral from interlamellar positions, and the latter from a surface adsorbed position. The enlargement of the FWHM (full width at half maximum) of XPS lines has been used to describe crystallochemical parameters linked to local ordering around the probe cations. Crystallization produces decreasing O 1 s and Cu 2p (octahedral cation) line widths but has no effect on the Si 2p (tetrahedral cation) line width. The enlargement of FWHM for all ion lines of the lattice is linked to the nature (Cu > Mg > Al) and the number and amount of structural cations in the phyllosilicates.

Clays are used in various cosmetic formulations, such as sunscreens, toothpastes, deodorants, creams, hair cosmetics, makeups, nail polish, facial masks, and shampoos, among others, to improve the organoleptic and physicochemical characteristics, to increase the stability, or to facilitate elaboration. Together with their technological functionalities, clays are cosmetologically active ingredients with cleaning, anti-aging, anti-wrinkling, and sun-care functionalities. Talc, kaolinite, mica, and some smectites are the clay minerals used most frequently in cosmetic products, but several other phyllosilicates as well as modified and synthetic clays are also used. Sometimes, clays are useful in the design of cosmetics just because they are made of rigid, small, and anisometric particles. Kaolinite and mica are made of hard prismatic particles which are lightly abrasive over the skin, teeth, or hair. Electric charges in smectites result in ion-exchange capacities useful in the loading of active cosmetics but also adsorbing and cleaning waste substances. Intermediate net negative charges of smectites result in layer expansion in polar media and specific rheological properties that are very useful in cosmetic formulations. The absence of charged particles in talc and kaolin make them flow easily resulting in lubricant effects. Protection against radiation from the sun by clay particles and decorative effects complete the possibilities of clays in cosmetics. The nomenclature for clays used as ingredients in cosmetics follows historical use and the names of commercial products, rather than following strict compositional principles. In this sense, an effort was made here to correlate the names of the minerals that make up each of the clay-based cosmetic ingredients.

Muscovite-kaolinite intergrowths found in Albian sandstones of the Basque Cantabrian basin (northern Spain) were studied by optical, scanning and electron microscopy and electron microprobe analysis. Kaolinitization begins at grain edges, forming the characteristic fanned-out textures, and propagates toward the interior along the cleavages of muscovite. Kaolinite and muscovite occur as thick packets, being free of interlayering. Phase boundaries between both minerals show bidimensional crystallographic continuity, and no intermediate phases have been identified. The data obtained suggest that muscovite only supplied a template suitable for the epitactic crystallization of kaolinite, while Al was available in sufficient amounts due to the dissolution of detrital K-feldspar. Very small packets of magnetite or maghemite showing a coherent orientation with the kaolinite crystals have been recognized, and could be responsible for the small Fe contents usually detected in electron microprobe analyses of kaolinite.

Textural relationships between authigenic kaolinite and deformation microstructures in the intergrowths, combined with previous information about burial conditions, show that alteration proceeded during a late stage of the diagenetic history, related to the uplift of the studied materials as a result of the Alpine orogeny.

The spatial distribution of the dominant matrix minerals present in the middle-Proterozoic Athabasca Group sandstone (kaolin, illite, sudoite, dravite, hematite) was studied at a regional scale in the Shea Creek region (Saskatchewan, Canada), in which two epigenetic unconformity-type uranium deposits have been discovered. 3D models of matrix mineral distribution were derived from normative mineral calculations and 3D interpolation using whole-rock geochemical analyses of sandstone samples collected from both mineralized and barren areas. The calculations were constrained by information obtained from petrographic and crystal-chemical clay mineralogical studies on representative samples. The 3D mineral distribution models were compared to the lithostratigraphy and structural features of the Athabasca Group sandstone to ascertain the source and mobility of the main elements involved in the sandstone host-rock alteration processes related to the U mineralization. The distribution of Al is conformable with the lithostratigraphy throughout the studied area, regardless of proximity to basement-rooted structures and U ore bodies. The distribution of illite displays similar features, but the intensity of the illitization of kaolin decreases with increasing distance from the structures and U ore bodies. Hematite bleaching and neoformation of sudoite and dravite were restricted to the vicinity of the fault zones above the U ore bodies. The spatial configurations of the mineral anomalies show that syn-ore fluids flowed from the basement towards the sandstone cover via the fault zones, as described in current metallogenic models. Although Al remained immobile (mass transfer), the anomalous K, B and Mg present in the host-rock alteration haloes were probably imported from the basement rocks (mass transport). Unlike B and Mg, K migrated laterally at least several kilometers from the basement-rooted faults. The mineral distribution models were used to quantify the volume of altered sandstone (10−2−10−1 km3) and the amounts of K, Mg and B which were imported to the alteration haloes above the Shea Creek U ore bodies: 186,000 t of K, 66,000 t of Mg, and 11,000 t of B above the Anne ore body, and 24,000 t of K, 185,000 t of Mg, and a similar 11,000 t of B above the Colette ore body.

The K-Ar dating of glauconite has been used as an important stratigraphic tool for many decades. The application of this technique is limited to pure glauconites, free of detrital contamination by K-bearing phases, often not easy to detect. This study extends the application of isotope dating to the contaminated glauconites and offers a precise technique for detecting the detrital contamination of glauconites.

The most common K-bearing detrital contaminants have smaller (K-feldspars, Al-rich dioctahedral micas) or greater (trioctahedral micas) dissolution rates than glauconite in extremely low pH solutions. The differences in the dissolution rates can be applied to evaluate the purity of the glauconite and its crystallization age.

The interlaboratory GLO glauconite standard and grain-size fractions separated from glauconitic sandstones of the Paleogene (sample GL) and Jurassic (sample GW8) ages were treated with acid (3M HCl, at 99±2°C) for different reaction times (0.5–7 h) and measured for their apparent isotopic ages.

Microporous amorphous silica with large specific surface area is the solid product of the reaction and its content increases with reaction time. The K-Ar dates (apparent ages) of the solid residues increase significantly with reaction time: from 44.6 to 107 Ma for the GL sample and from 125.7 to 394.7 Ma for GW8. The increase is negligible in the case of the GLO standard. The Rb-Sr data of the GL sample were modeled using initial 87Sr/Sr ratios of 0.707–0.709, which resulted in a 29.9–35.8 Ma date for the untreated portions of GL, and ∼42.6 Ma after 7 h of treatment.

The increase of isotopic K-Ar date with increasing time of dissolution is interpreted to be a result of increasing concentration of detrital, acid-resistant, K-bearing minerals, observed also with the electron microscope and X-ray diffraction. Probabilistic modeling based on single (K-Ar) or double (K-Ar and Rb-Sr) isotopic systems evaluated the isotopic ages of the detrital and authigenic minerals, and their K2O and Rb concentrations. The crystallization ages computed using these two methods are: 24.0, 26.5, and 32.3 Ma for the GL material, and 117.3–121.8 Ma for the GW8 series.

The proposed method based on partial dissolution is a potential tool for evaluating the reliability of glauconite dating.

The dissolution of two Ca2+-exchanged nontronite samples has been studied in 10% HCl. Early acid-dissolution studies (Osthaus, 1954) have indicated that after two hours of dissolution most of the octahedral Fe3+ (VIFe3+) would be removed leaving mainly tetrahedral Fe3+ (IVFe3+) in the nontronite structure. In the present study, 57Fe Mössbauer spectra of acid-treated samples were recorded and fitted with two octahedral Fe3+ (2 × VIFe3+) and two octahedral and one tetrahedral (2 × VIFe3+, 1 × IVFe3+) doublet models. The Mössbauer spectra of acid-treated Garfield nontronite samples could be adequately fitted with two-doublet models but acid-treated Hohen Hagen nontronite samples could not. Isomer shift and quadrupole splitting values obtained from the two-doublet models corresponded to VIFe3+ and not IVFe3+, as was suggested by the Osthaus (1954) experiment. When an IVFe3+ doublet was included in the model used to fit the Mössbauer spectra of acid-treated Garfield nontronite samples, a slight increase in the intensity of the IVFe3+ doublet occurred with increasing dissolution, but this was much lower than indicated by Osthaus (1954). No trend in the intensity of the IVFe3+ doublet was observed for acid-treated Hohen Hagen nontronite. Therefore, acid treatment appears to remove VIFe3+ and IVFe3+ from the nontronite structure at about the same rate. Mössbauer spectroscopy, infrared spectroscopy and X-ray powder diffraction data indicate that the nontronite that remains undissolved following acid treatment is structurally similar to the untreated nontronite.

The clay fraction of an Attic soil has been fired at temperatures up to 1100°C and the transformations induced by firing were studied by Mössbauer spectroscopy. The unfired clay contained iron in the form of small particles of α-Fe2O3 and β-FeOOH and as a substitutional ion in the clay mineral structure. A transformation of β-FeOOH to α-Fe2O3 is observed in the region of 200–400°C followed by dehydroxylation of the clay mineral between 500 and 800°C. A disintegration of the clay mineral structure is inferred above 800°C, followed by recrystallization in an iron-rich phase, most probably in the form of very fine particles. The room temperature spectra of the fired samples indicate that an increase in particle size of the oxides occurs upon firing.