A series of reduced-charge (RC) hectorites were prepared by multiple heat (250°C) treatments of Mg-saturated hectorites (SHCa-1 ). Cation exchange capacity (CEC) measurements and alkylammonium exchange indicated that a decrease in layer charge occurred with each Mg-250 treatment. Chemical analyses showed that decreases in structural Li and increases in structural Mg contents coincided with charge reduction. Fluorescence measurements of adsorbed quinoline indicated that the hectorite surface was acidified during charge reduction; hydroxyl group deprotonation is a possible source for the acidity. Fourier transform infrared spectra (FTIR) indicated that the Mg-250 treatment induced the loss of structural Li and shifted the SiO stretch band to a position similar to that in talc. The relative intensities of the OH and SiO stretch bands in FTIR spectra suggest that some of the hydroxyl groups in hectorite were lost, possibly by deprotonation. However, thermogravimetric data (TG) reveal no significant difference in the hydroxyl contents of the hectorites.

The FTIR spectra, CEC, layer charge, chemical, and TG data all supported the view that Mg substitution for octahedral Li occurred which resulted in a more “talc-like” structure. Charge reduction in smectites is evidently a general phenomenon and can be induced by heat treatment with the proper exchangeable cation. The ability to reduce the charge of hectorites makes it possible to prepare a series of clays which vary in charge but lack structural Fe. Such RC smectites should be suitable for expandable clay mineral studies which utilize spectroscopic techniques that are sensitive to Fe content.

An environmental infrared microbalance (EIRM) cell was used to study H2O sorption on two montmorillonite samples as a function of water content and type of exchangeable cation. The vibrational spectra showed that H2O sorbed to the clay at low-water content was strongly influenced by the exchangeable cation and by the close proximity to the clay surface. At water contents <6 H20 molecules per exchangeable cation, the H-O-H bending mode of H2O (v2 mode) shifts to a lower frequency and is characterized by an increase in molar absorptivity. In contrast, the positions of the asymmetric and symmetric OH-stretching modes of sorbed water (v1 and v3 modes) shift to higher energies. These observations indicate that H2O molecules sorbed to the clay surface at low-water content are less hydrogen bonded than in bulk H2O. In addition, the vibrational-stretching and bending bands of the structural OH groups of the 2:1 layer are also strongly influenced by H2O content and type of exchangeable cation. By using the EIRM cell, the molar absorptivities of the structural OH-bending vibrations were measured as a function of H2O content. The position and molar absorptivity of the structural OH-bending bands at 920, 883, and 840 cm-1 are strongly influenced by H2O content and type of exchangeable cation. The molar absorptivity of the 920-cm-1 band, which is assigned to the AlAlOH group, decreased strongly at low-H2O content. This reduction in intensity is assigned to a dehydration-induced change in orientation of the structural OH groups resulting from the penetration of H2O molecules into siloxane ditrigonal cavities that are not associated with a net negative charge from isomorphous substitutions.

Low-temperature FTIR spectroscopy was used to characterize the v(OH) region of kaolin-group minerals including well ordered to poorly ordered kaolins from Georgia, Brazil, and England, along with samples of discrete dickite and nacrite. Low-temperature FTIR spectra were useful in resolving dickite- and nacrite-like features present in the spectra of kaolins when cooled to <30 K. These features were not resolved at room temperature and only partially resolved at liquid N2 temperature (77 K). The room-temperature and low-temperature positions of the ν(OH) bands of kaolinite, dickite, and nacrite were linearly correlated with the interatomic OH⋯O distances and this relationship served as the basis for polytype/disorder identification. Dickite or dickite-like disorder was found in high Hinckley-Index kaolinite from Keokuk, Iowa, and from Cornwall, England. Dickite- and nacrite-like features were observed in both high- and low-Hinckley-index kaolinite and the amounts of these stacking sequences generally increased with decreasing Hinckley Index.

Sorption and transformation of 1-naphthol by a K-smectite (K-SWy-2) were studied using batch sorption isotherms, Fourier-transform infrared spectroscopy (FTIR), and X-ray diffraction (XRD). The sorbents included three preparations of the reference smectite clay (SWy-2): (1) whole clay containing naturally occurring carbonate impurities, (2) SWy-2 with the removal of carbonate impurities, and (3) the carbonate-free SWy-2 fraction amended with calcite. For the whole clay and carbonate-free clay amended with calcite, >80% of added 1-naphthol disappeared from aqueous solution within 24 h, corresponding to a sorbed concentration of ≥2 mg/g of clay. In contrast, only 35% of the added 1-naphthol disappeared from solution in the carbonate-free clay after 24 h of exposure. For the clays from the three preparations in this study, <1% of sorbed 1-naphthol could be recovered by methanol extraction from the clays. The XRD data suggested that 1-naphthol was intercalated in the smectite, but was not conclusive because the 1-naphthol sorption range (1.5–2.8 mg/g of clay) in this study had relatively minor effects on the XRD patterns. The FTIR spectra of sorbed 1-naphthol-clay complexes demonstrated structural Fe3+ reduction. The spectra also showed evidence of the transformation of 1-naphthol. It suggests that reduction of structural Fe3+ may be coupled to oxidation/polymerization of 1-naphthol. Further transformation of oxidized 1-naphthol, such as by oxidative coupling reactions, is implicated by formation of a dark gray color on the clay and the inability to extract sorbed 1-naphthol.

The Brunauer-Emmett-Teller (BET) theory models the effective specific surface area and water content of solids as a function of the relative vapor pressure of water. A modified form of the BET equation has been used successfully to model water activity in concentrated electrolyte solutions as a function of electrolyte concentration. This modified form, referred to here as the Stokes-Robinson BET model, is based on the electrolyte molality rather than on the mass of solute sorbed. The present study evaluates the Stokes-Robinson form of the BET equation to model water-sorption data on two smectites with different layer charges. One smectite was saturated with Na+ and another with Na+, Ca2+, or Mg2+. These results are compared to the Stokes-Robinson BET results of aqueous electrolyte solutions. Given published data on cation exchange capacities and water-vapor sorption isotherms for various clays, the molality of the aqueous phase in contact with the clay surface is calculated and related to water activity. The Stokes-Robinson BET model was found to describe accurately the water activity as a function of cation molality below water activities of 0.5 for the smectites. Good relative agreement was obtained between the number of water binding sites predicted by the model and the experimental data reported in the literature for other smectites. Water molecules were found to have a significantly greater affinity for montmorillonite than electrolyte solutions with the same cation molality as the montmorillonite interlayer. This modified BET approach simplifies water-activity modeling in highly saline environments because the same equation can be used for both the liquid- and mineral-surface phases.

The adsorption of two dinitrophenol herbicides, 4,6-dinitro-o-cresol (DNOC) and 4,6-dinitro-o-sec-butyl phenol (dinoseb), by two reference smectite clays (SWy-2 and SAz-1) was evaluated using a combination of sorption isotherms, Fourier transformation infrared (FTIR) spectroscopy, X-ray diffraction (XRD) and molecular dynamic simulations. Clays were subject to saturation with various cations, and charge reduction. The DNOC adsorption decreased with increasing pH indicating that DNOC was primarily adsorbed as the neutral species. The FTIR spectra of DNOC-clay films showed that DNOC molecules are oriented parallel to the clay surface. Interlayer cations have a strong effect on adsorption depending largely on their hydration energies. Weakly hydrated cations, e.g. K+ and Cs+, resulted in greater sorption compared to more strongly hydrated cations such as Na+ or Ca2+. Lower hydration favors direct interactions of exchangeable cations with -NO2 groups of DNOC and manifests optimal interlayer spacings for adsorption. In the presence of sorbed DNOC, an interlayer spacing for K-SWy-2 of between 12 and 12.5 Å was maintained regardless of the presence of water. This d-spacing allowed DNOC molecules to interact simultaneously with the opposing clay layers thus minimizing contact of DNOC with water. The charge density of clays also affected sorption by controlling the size of adsorption domains. Accordingly, DNOC adsorption by low-charge clay (K-SWy-2) was much higher than by high-charge clay (K-SAz-1) and Li-charge reduction greatly enhanced dinoseb adsorption by K-SAz-1. Steric constraints were also evident from the observation that adsorption of DNOC, which contains a methyl substituent, was much greater than dinoseb, which contains a bulkier isobutyl group. Adsorption of DNOC by K-SAz-1 was not affected in the presence of dinoseb, whereas dinoseb adsorption was greatly reduced in the presence of DNOC.

Organic dyes such as methylene blue (MB) are often used in the characterization of clays and related minerals, but details of the adsorption mechanisms of such dyes are only partially understood from spectroscopic data, which indicate the presence of monomers, dimers, and higher aggregates for varying mineral surfaces. A combination of quantum (density functional theory) and classical molecular simulation methods was used to provide molecular detail of such adsorption processes, specifically the adsorption of MB onto kaolinite basal surfaces. Slab models with vacuum-terminated surfaces were used to obtain detailed structural properties and binding energies at both levels of theory, while classical molecular dynamics simulations of aqueous pores were used to characterize MB adsorption at infinite dilution and at higher concentration in which MB dimers and one-dimensional chains formed. Results for the neutral MB molecules are compared with those for the corresponding cation. Simulations of the aqueous pore indicate preferred adsorption on the hydrophobic siloxane surface, while charge-balancing chloride ions adsorb at the aluminol surface. At infinite dilution and in the gas-phase models, MB adsorbs with its primary molecular plane parallel to the siloxane surface to enhance hydrophobic interactions. Sandwiched dimers and chains are oriented perpendicular to the surface to facilitate the strong hydrophobic intermolecular interactions. Compared with quantum results, the hybrid force field predicts a weaker MB adsorption energy but a stronger dimerization energy. The structure and energetics of adsorbed MB at infinite dilution are consistent with the gas-phase binding results, which indicate that monomer adsorption is driven by strong interfacial forces rather than by the hydration properties of the dye. These results inform spectroscopic studies of MB adsorption on mineral surfaces while also revealing critical areas for development of improved hybrid force fields.

Due to the increased importance of bionanocomposites, protamine and papain proteins were adsorbed on Na+- and on Cs+-exchanged saponite from aqueous solution. Protein analysis of equilibrium solutions and thermogravimetric analyses of biocomposites were used to prepare adsorption isotherms. Based on the isotherm shape, and on the amounts of protein adsorbed and the amounts of Na+ and Cs+ released, the initial protein sorption apparently was due to ion exchange. Additional sorbed protein was weakly retained and could be removed by washing with water. From ion exchange, the average charge of the protamine adsorbed was estimated to be +13.1 to +13.5. Similar papain measurements could not be made due to partial decomposition. Quantitatively, protamine was adsorbed at levels up to 400 mg/g on Na+-saponite and 200 mg/g on Cs+-saponite. The maximum protamine adsorption was 650 to 700 mg/g for Na+-saponite and 350–400 mg/g for Cs+-saponite. Protamine was sorbed to edge surfaces and the basal spacing of the interlamellar region of saponite was 1.75 nm. Protamine displaced only 36% of the Cs+ in Cs+-saponite and expanded the interlamellar region by 36% for a basal spacing of 1.6 nm. Papain sorption to Na+-saponite occurred by a two-step process: (1) adsorption to saponite particle external surfaces followed, (2) by partial intercalation. Quantitatively, Papain was adsorbed up to 100 mg/g for Na+-and Cs+-saponite. Greater initial papain concentrations resulted in a 450 mg/g maximum for Na+-saponite, but no increase above 100 mg/g for Cs+-saponite. Papain apparently only sorbed to external Cs+-saponite surfaces that were estimated to be 33–40 m2/g. Step-wise thermal decomposition of the saponite-protein composites occurred between 300 and 800°C.

Although structural disorder in kaolinite has been investigated extensively, it is still not understood properly. To investigate the problem, a kaolinite specimen of sedimentary origin from Capim, Brazil, was examined, mainly by transmission electron microscopy (TEM). Selected-area electron diffraction (SAED) along the Xi ([100], [110], and []) directions shows various features, from completely discrete patterns to heavily streaked ones along the c* direction, suggesting that the degree of stacking disorder is variable among individual grains. High-resolution TEM images indicate that stacking faults are mainly caused by disorder of alternating t1 (~ −a/3) and t2 (−a/3 +b/3) layer displacements. Furthermore, stacking faults have been observed (1) as isolated stacking faults (e.g. insertion of an isolated t2 `fault' in an ordered sequence with t1 layer displacement) and (2) as interstratification of two kinds of multilayer blocks having regular t1 and t2 layer displacements. A mixture of grains with various degrees and modes of disorder with alternating t1 and t2 layer displacements may explain the experimental profile of the 02, 11 X-ray diffraction band.

Faults related to displacement of the octahedral vacancy and/or to layer rotation were also observed in HRTEM images. The SAED patterns along the Yi ([010], [310], and []) directions occasionally have extra spots and/or streaks, suggesting the presence of stacking sequences with (±60°, 180°) mutual layer rotation and/or with (0, ±b/3) layer displacements. The local dickite or nacrite-like fragments formed by these faults are in qualitative agreement with recent low-temperature FTIR results from this sample, where distinct ν(OH) absorption bands reflect multiple interlayer O-H⋯O environments that are possibly ascribed to dickite and nacrite.