Transmission electron microscopy-energy dispersive X-ray analysis (TEM-EDX) represents an effective tool for determining the stoichiometric composition of clay minerals, but the methodology is often hampered by analytical difficulties. Studies of beam-sensitive minerals, such as smectites, are associated with low count intensities and dynamic loss of cations (e.g. K+, Na+, and Al3+), which can lead to erroneous quantifications of composition. After exploring how to minimize cation migration by reducing the beam current density to <5 pA cm–2, the most reliable and consistent compositions were determined using 1 μm2 area measurements of particles acquired in normal TEM mode where the electron beam was parallel, the degree of specimen damage was at its minimum and good acquisition intensities (>10,000 cps) were acquired. Based on 528 TEM-EDX area analyses, the composition of Wyoming montmorillonites (SWy-1, SWy-2, and SWy-3) was studied in their natural and Ca-saturated states from thin (<50 nm thick) particle aggregates lying on lacey carbon films. Overall, the TEM-EDX results confirmed the heterogeneous charge distributions of montmorillonite at the particle and sample levels. The average composition per formula unit of SWy-1 to -3 was determined as: (Na0.12Ca0.04Mg0.03K0.02)(Si3.91Al0.09)(Al1.57Mg0.27Fe0.19)2.03 O10(OH)2 · nH2O, where the tetrahedral and octahedral layer charges are –0.09 and –0.19 per O10(OH)2, respectively, and the total layer charge ranges from –0.25 to –0.30 per O10(OH)2 (mean of –0.28). This study demonstrates how TEM-EDX can provide new insight into the natural heterogeneities of smectite chemistry as long as adequate calibration and specimen damage control procedures are implemented.

The charge density distribution among different classes of a series of reduced charge mont-morillonites is heterogeneous as in the parent Camp Berteau clay. In addition, charge reduction proceeds inhomogeneously. Up to 20% differences in charge density can be accounted for by aikyl chains extending at the edges of the clay particle. A realistic charge density-cation-exchange capacity relationship for hec-torite, Otay montmorillonite, and a series of reduced charge montmorillonites of Camp Berteau is obtained by accounting for the influence of particle radius and for the extent of alkyl chains lying outside the clay layers in the charge density calculations.

The formation of an interstratified structure in dioctahedral smectite was assumed to be influenced by (1) the overall layer charge density and its distribution in the structure, (2) the solvation energy of the cation, and (3) the nature of the solvation agent. By holding factors (2) and (3) constant it was possible to calculate the average local charge densities $\overline {{\rm{QA}}}$, $\overline {{\rm{QC}}} $, and $\overline {{\rm{QE}}} $ which are necessary for formation of 10-, 14-, and 16.8-Å mixed-layer phases in potassium-treated and ethylene glycol (EG) saturated smectites. The values of $\overline {{\rm{QA}}} $, $\overline {{\rm{QC}}} $, and $\overline {{\rm{QE}}} $ were 1.99, 1.2, and 0.56 esu/unit cell, respectively. Ammonium-treated smectites saturated with EG gave corresponding mean local charge densities of 2.7, 1.6, and 0.72 esu/unit cell. Calculations were made under the limiting condition QA > QC > QE > 0.

For K-smectites saturated with EG, Qtot = 1.99pA + 1.2pC + 0.56pE, where Qtot is the total charge (esu/unit cell), and pA, pC, and pE are probability coefficients for 10-, 14-, and 16.8-Å phases in the interstratified structure. The above equation calculated with the aid of least squares and without the limiting condition yields

$${{\rm{Q}}_{{\rm{tot}}}} = {\rm{2}}{\rm{.05pA + 1}}{\rm{.29pC + 0}}{\rm{.33pE}}{\rm{.}}$$

Monoionic K- and NH4-smectites saturated with ethylene glycol form mixed-layer structures which usually consist of three kinds of layers: a 10-Å, non-expanded layer; a 14-Å, partly expanded layer; and a 16.8-Å, completely expanded layer. In some samples, the 14-Å layers formed 60–70% of all layers present. When saturated with water vapor the smectites commonly consisted of three kinds of layers (10-, 12.6-, and 15.5-Å). Generally these samples contained fewer expanded layers than those saturated with ethylene glycol. This result is attributed to the smaller dipole moment of water compared with that of ethylene glycol. The greater solvation energy of NH4+ in comparison with that of K+ causes the expansion of a part of layers which did not expand in the K forms. This result indicates that there is an inhomogeneous distribution of layer charge in the smectite structure. The prevalent type of mixed layering in the studied samples is that of random distribution of layers.

The basal spacing of a set of smectites, with layer charges between 0.74 and 1.14 electrons per unit cell, have been measured while the smectites were in equilibrium with NaCl solutions having concentrations of 0.25 to 2.5 molal. Except for the most highly charged smectites, an expansion from ~15.5 to ~18.5 Å occurred as the NaCl concentrations were reduced. This expansion (or “crystalline swelling”) corresponds to the transition from two to three sheets of water between the silicate layers. A random interstratification of 15.5 and 18.5 Å structure units was present during the change and gave rise to broad diffraction peaks for the 001 reflections. A balance between cation hydration forces, interlamellar electrostatic forces, and van der Waals forces was apparently the basis of the relationship between surface charge and swelling. The results can be expressed in terms of the relative vapor pressures at which the transitions were half complete; these pressures increased with surface density of charge, and over the range of surface charge studied, the P/P0 values ranged from 0.943 to 0.974.

Mixed-layer clays composed of randomly interstratified kerolite/stevensite occur as lake and/or spring deposits of probable Pliocene and Pleistocene age in the Amargosa Desert of southern Nevada, U.S.A. The percentage of expandable layers of these clays, determined from computer-simulated X-ray diffractograms, ranges from almost 0 to about 80%. This range in expandabilities most likely results from differences in solution chemistry and/or temperature at the time of formation. An average structural formula for the purest clay (sample P-7), a clay with about 70% expandable layers, is:

$${\left[ {\left( {M{g_{2,72}}A{l_{0,07}}F{e_{0.03}}L{i_{0.09}}} \right)\left( {S{i_{3.96}}A{l_{0.04}}} \right){O_{10}}{{\left( {OH} \right)}_2}} \right]^{ - 0.21}}{\left[ {X_{0.21}^ + } \right]^{ + 0.21}}.$$

The data suggest that talc, kerolite, and stevensite form a continuous structural series based on layer charge.

A mixed-layer kaolinite/smectite (K/S) containing trace amounts of quartz, discrete kaolinite, goethite-hematite, and calcite was hydrothermally reacted with deionized water at 150°, 200°, and 250°C for 1 to 12 months. The starting K/S contained 50% smectite consisting of 15% low-charge and 35% high-charge layers. The X-ray powder diffraction and chemical analyses of the reacted products indicated a progressive reaction from high-charge to low-charge smectite as a function of time and temperature. The reaction reached completion after 4 months at 250°C, at which point high-charge smectite layers entirely reacted to low-charge smectite layers, the latter maintaining a constant proportion of about 90% for longer run durations. For long reaction times, discrete kaolinite totally reacted, whereas quartz showed only partial dissolution and iron oxides remained stable. Thus, the reaction of high-charge to low-charge smectite layers may be expressed as: high-charge smectite + kaolinite (both interstratified and discrete component) + quartz → low-charge smectite.

The broad charge heterogeneity typical of nearly all smectites is not necessarily characteristic of vermiculites. In addition to vermiculites with pronounced heterogeneity, minerals with no or only limited charge heterogeneities are known. Layer charge and charge heterogeneity of 25 vermiculites were determined by alkylammonium ion exchange. The comparison of experimental basal spacings with dL/n-plots provided a simple determination of the average charge density. The spacings of high-charged vermiculites (≥0.8 eq/(Si,Al)4O10) with paraffin-type interlayers follow a straight line in the dL/n-plots. Lower-charged vermiculites were recognized by stepwise increasing spacings due to mono-, two-, or three-layer chain packings. Charge heterogeneity produced a superposition of the dL/n-curves for different charges, and the basal reflections of some of the alkylammonium derivatives became nonintegral.

In a study of “rana” soils of central Spain, hydroxy-interlayered 2:1 clay minerals were identified in the upper horizons and high-charge smectite in the lower horizons, with kaolinite the most abundant phyllosilicate. The high-charge smectite showed a basal spacing of 18 Å on Mg-saturation and glycerol solvation and 10 Å on K-saturation and air-drying. It is concentrated in the coarse fractions and appears to be basically a beidellite in the fine-silt fractions and a montmorillonite in the clay fractions. Pre-existing illite was probably transformed into hydroxy-interlayered 2:1 clays in the surface horizons, whereas high-charge smectite formed in the deeper horizons. These minerals, which were apparently not stable in the existing soil conditions, were the first stage in the degradation process of illite. The stable mineral in these soil profiles was kaolinite.

Saponites and vermiculites may assume at least 11 ordered or semi-ordered layer stacking sequences. For a given relative humidity, the layer stacking type assumed is a function of the nature of the interlayer cation, the layer charge density, the mean size of the particles, and the di- or trioctahedral character of the sheets. For each interlayer cation, a succession of layer stacking types can be observed as relative humidity increases. For high relative humidity, some particular layer stacking types exist, but only for low-charge minerals. No other differences have been found for saponites and vermiculites in each successive layer stacking type. The degree of order that these layer stacking types imply is probably due to the existence of electrostatic bonds between hydrated interlayer cations and surface oxygens of the substituted tetrahedra. For octahedrally substituted 2:1 phyllosilicates, however, the disorder of the layer stacking sequences is related to a highly delocalized distribution of negative charges on the surface oxygens of the layers.

A study of the superstructures detected in saponites and vermiculites indicates that the interlayer cations tend to be located as far as possible from one another. The superstructures exist only with some cations and some layer stacking types and if the layer charge density is compatible with the charge produced by the cation distribution in this kind of superstructure.

The layer charge of five smectites, one vermiculitic material, and five reduced-charge clays was determined by the alkylammonium and structural formula methods. The two sets of results were found to be linearly correlated (r =.961); however, the values that were determined by the alkylammonium method were 20 to 30% lower than those determined by the structural formula method, and the regression slope for their linear relationship was 1.67. The fact that the structural formula method includes the effects of cations on the lateral edges of the clay particles probably accounted for some of the differences in the magnitude of the results but should not have caused the regression slope to deviate substantially from 1.00. Therefore, inaccurate estimates of the packing density of alkylammonium cations in the interlayer space of 2:1 phyllosilicates were deemed responsible for the systematic divergence of the results of the two methods. To satisfy the need for a relationship between the two methods of determining layer charge, an empirical means of adjusting the alkylammonium values has been proposed and shown to yield values of layer charge that are comparable to those determined by the structural formula method.

Chemical analysis by X-ray fluorescence (XRF) and calculated structural formulae of clay-size fractions of smectites from Cretaceous bentonites and illitic clays from Cretaceous, Devonian, and Ordovician bentonites and Jurassic and Permian sandstones indicate the nature and extent of various types of ionic substitution. The determination of tetrahedral (Al, Si) and octahedral (Al, Mg, Fe) composition shows the variable chemistry of these materials. Structural formulae of the illitic clays show that they have tetrahedral charges between 0.4 and 0.8 per half unit cell, and can be divided into phengitic types having octahedral charges of 0.2-0.4 and muscovitic types having octahedral charges <0.2. Evaluation of the formulae in the light of X-ray powder diffraction (XRD) and transmission electron microscopy (TEM) data shows that the occupancy of non-exchangeable interlayer sites (predominantly K) varies from 47% to 90% of that of ideal muscovite. In some minerals as much as 20% of these sites is occupied by ammonium ions (determined independently). The amount of surface silicate charge balanced by non-exchangeable cations versus that balanced by exchangeable cations has been examined in conjunction with TEM data and suggests that in most samples the charges are about equal. The octahedral composition of smectites in Cretaceous bentonites precludes their having served as transformation precursors for most of the Cretaceous illitic bentonites. The results suggest that these illitic clays originated by neoformation.

The Nriagu polymer model of 2:1 layer type clay minerals develops from the premise that clay minerals are condensation copolymers of solid hydroxides. In the Mattigod-Sposito formulation of the model, standard state chemical potentials (standard Gibbs energies of formation from the elements) of 2:1 clay minerals are predicted quantitatively with a linear correlation equation relating the standard Gibbs energy of the polymerization reaction (ΔGor) to the half-cell layer charge of the clay mineral and to the valence and ionic radius of the exchangeable cation. It is now shown that this correlation equation can be derived from two basic assumptions: (1) that the standard Gibbs energy change for the transfer of a cation in a pure hydroxide solid to a hydroxide component in the tetrahedral or octahedral sheet of a 2:1 clay mineral is independent of the nature of the cation and (2) that the difference between ΔGor for the polymerization reaction to form a 2:1 clay mineral and ΔGor for the same reaction to form the zero layer-charge analog of the clay mineral is proportional to the number of interlayer exchangeable cations per unit cell of the clay mineral and to the radius of its exchangeable cation. Both of these assumptions can be tested experimentally, independent of the polymer model.

Layer charge and cation-exchange characteristics of weathered biotite isolated from a Gray Luvisol in Saskatchewan showed that this product was a high-charge vermiculite. Cation-exchange capacity (CEC) of the sand-size (100–250 μm) weathered biotite particles from different horizons, as determined by a 30-min Ca/Mg exchange, was low (23–71 meq/100 g). Increasing the exchange period to 24 hr increased the CEC to 33–95 meq/100 g for samples from the solum horizons and to 155–163 meq/100 g for samples from C horizons. The free oxide coatings on the mineral particles in the solum horizon samples apparently prevented the exchange cations from entering the interlayers. Removal of free oxides followed by 24-hr exchange sharply increased the CEC of these samples to 105–155 meq/100 g. Oxidation and loss of structural iron resulted in lower octahedral-cation occupancy (2.4–2.6 per half unit cell), suggesting a transition of the trioctahedral biotite to dioctahedral vermiculite. X-ray powder diffraction and high-resolution transmission electron microscopic analyses of alkylammonium-exchanged samples from the sand fractions of all horizons and magnetic separates of the coarse clays showed a linear increase in the d-value of the vermiculite with increasing chain length of the cations. The linear relationships suggest a paraffin-type arrangement having a uniform and high layer charge (0.70–0.76 per half unit cell in the sand-size particles; 0.80–0.84 in the clay fraction). The empirical relationship between tilt angle (α) and the layer charge density for paraffin-type structures, as suggested by Lagaly and Weiss, cannot be used for accurate layer-charge determination. Based on experimental evidence, a straight line relationship between the tilt angle and layer charge is suggested. The layer charge values show no marked variation within the profile and agree well with those calculated from the chemical composition. The repotassified layers of vermiculite in the sand-size weathered biotite particles did not respond to the Ca/Mg exchange, but were expanded by the alkylammonium cations; in contrast, a pure biotite standard (Bancroft, Ontario) of same particle size was unaffected by either treatment.

Sodium-saturated Wyoming bentonite was hydrothermally reacted at 150° and 250°C for 30 to 180 days to determine smectite alteration rates that might be applied to nuclear-waste repository design. Na-Ca solutions deficient in K were used to determine the role of interlayer cations in the creation of high layer-charge in the smectites. The results provide insight into the mechanism and timing of various steps in the diagenetic alteration of smectite to illite. X-ray powder diffraction (XRD) analyses of the reacted clay showed little effect on the character of the 17-Å reflection even after 180 days at 250°C. Potassium saturation of these reacted clays and re-examination by XRD indicated collapse of some smectite layers, leaving at most only 60% expandable layers. The development of layer charge sufficient to cause collapse on saturation with a low hydration energy exchange-cation does not require K in the reacting fluid. Rate constants for the illitization reactions as determined by K-saturated collapse are between 1.0 × 10−3 and 2.8 × 10−3/day with activation energies <3.5 kcal/mole. Ca in a Na-silicate-bicarbonate solution slightly reduced the illitization rate constants. These rate constants are higher than expected from extrapolation of studies of beidellite-composition glasses at higher temperatures, but lower than values obtained in studies of natural clays in artificial sea water. The release of Si, Al, and Mg in the 150°C experiments suggests congruent dissolution of the smectite. In contrast, at 250°C the release of Al was not stoichiometric with Si; as little as one half of the relative available Si was released. Rather than different mechanisms for dissolution at the two temperatures, the conclusion is that noncrystalline Al-rich phases formed at greater rates at higher temperatures. The cation-exchange capacities for several of these reacted smectites were significantly less than expected, suggesting a clogging of interlayer sites, perhaps by Al-complexes.

The objectives of the study were to determine the chemical composition and layer charge of smectite found in calcareous till of the Interior Plains region of western Canada and to examine the effects of acidification on alteration of the smectite. Samples of acidified and non-acidified (calcareous) late-Wisconsin till were obtained from four soil pits located immediately adjacent to an elemental sulfur block located in southern Alberta. Samples of the surface material (0–10 cm depth) had been subjected to extreme acidity for 25 years due to the oxidation of elemental sulfur and displayed pH values of about 2.0. Samples of the till obtained at depth (65–75 cm) remained calcareous with pH values between 7.3 and 7.6. A combination of analytical methods was used to determine the chemical composition of the smectite found in the samples. The layer charge of the smectite was determined independently using X-ray diffraction data for n-alkylammoniurn saturated specimens. Smectite found in the non-acidified calcareous material was characteristic of montmorillonite with a low content of Fe and very little substitution of Al for Si in the tetrahedral sheet. The smectite had a structural formula of M+0.40(Si3.96Al0.04)(Al1.56Fe3+0.10Mg0.33)O10(OH)2, which compared well with a mean value for layer charge of 0.399 mol(−)/O10(OH)2 determined using X-ray diffraction data for n-alkylammonium treated specimens. Smectite remaining in the till material subjected to extreme acidity underwent incongruent dissolution with a net loss of layer charge and preferential loss of octahedral Mg.

Recent pedological evidence of the widespread distribution of beidellites in soils indicates the need for a greater knowledge of the effect of charge location on the microstructural organization of Na+ smectite in gels. After equilibration at a suction pressure of 3.2 kPa before and after desiccation, TEM observations showed large differences between a beidellite and both a low and high charge montmorillonite. Monolayers were rare; individual layers were instead organized in particles with larger interparticle distances. This has implication for theories relating swelling pressures to interlayer distances or surface areas and implies the need for a geometrical approach to the study of swelling in smectites. Location of isomorphous substitution in the tetrahedral sheet of smectites results in an increased lateral extension of overlapping layers. This was reflected in a greater capacity to rehydrate after desiccation. Increased number of layers in particles were found with increasing surface charge density. The geometric organization of the particles is critical to the understanding of the ability of Na+ smectite to hold water against an applied suction.

A critical demand in environmental modeling and a desirable but elusive goal of research on the ion exchange properties of the charged solid surface has been to determine the selectivity coefficient from fundamental properties of the ions and surface. We developed a Hard and Soft Acid and Base (HSAB) Model to describe exchangeable cation selectivity on solid surfaces. Our previous work has shown that the model quantitatively describes alkali cation exchange on clay minerals in terms of the absolute electronegativity and softness of the exchangeable cations and two fitting parameters: α and β. This study was conducted to determine the relationship between α and β and surface charge characteristics of 2:1 clays. The layer charge and cation selectivity of seven smectites and one vermiculite were used. The regression of log Kvo against four combinations of charge properties was performed and the appropriate relationship between α, β, and surface charge was selected based on both statistical criteria (R2) and their consistency with the assumptions of the HSAB model. The selected model was then cross-validated using separate cation exchange data from the literature. It was found that α and β are linearly related to the amount of charge arising from mineral tetrahedral and octahedral sites, respectively. These results make it possible to predict the alkali cation selectivity of 2:1 clay minerals from their chemical composition data and the alkali cation properties.