Models of cation exchange mechanisms and driving forces have proven effective predictors of clay behavior and chemistry, but are largely theoretical, particularly in complex systems involving high ionic strength brines or systems where hydration is controlled by relative humidity. In arid and cold environments, such as Mars, cyclical relative humidity variations may play a role in chemical alteration, particularly if clay minerals such as smectite are in the presence of salts. This study examines the effects of relative humidity on smectite-salt mixtures using environmental scanning electron microscopy (ESEM) to observe the physiochemical effects of salt deliquescence and desiccation on smectite textures and elemental distributions. Results demonstrate that even reaction periods as short as a few minutes allow ample time for relative humidity to affect the smectite-salt mixtures. In addition to smectite swelling and salt deliquescence, we also observed rapid changes in element distributions within the smectite and new crystal growth in the presence of high relative humidity. Even in the absence of bulk liquid water, exchangeable cations migrated out of the smectite and formed new crystals at the smectite-salt interface. The observed microscopic changes in elemental distributions indicate that the migration of cations driven by cation exchange led to secondary mineral precipitation, likely a CaSO4 mineral, within a sub-micrometer-thick layer of water on the smectite grains. The results of this study demonstrate that during periods of elevated relative humidity, active smectite mineral alteration and secondary mineral precipitation may be possible on present-day Mars where salts and smectites are in direct physical contact.

Adsorption of Co2+ and Cd2+ on Wyoming montmorillonite was studied by the batch equilibration technique, as a function of salt concentration (0.01–4 M NaCl and NaNO3), pH (5.0–6.5), adsorbate concentration (trace-10−2 moles/liter), and presence of complexing ions. Comparison was made with the adsorbability of Sr2+, known to follow simple ion-exchange equations. The distribution coefficients for Co and Cd in noncomplexing media varied with salt concentration (from ∼500 liters/kg in 0.01 M Na+ to ∼10 liters/kg in 1 M Na+; pH = 5), but to a lesser extent than that of Sr. Adsorbability varied also with pH (∼1 order of magnitude/pH unit), especially at high ionic strength, compared to a negligible pH effect on Sr. The distribution coefficients of Cd and Co decreased with increasing loading on the clay at a very low percentage (0.2%) of the ion-exchange capacity compared to Sr (20%). These data suggest two classes of sites participating in the adsorption of Cd and Co.

The adsorbability of Cd in highly concentrated chloride solution (>1 M) was less than 1 liter/kg, presumably because of the chloride complex formation. This effect increased with increasing pH. The low adsorbability of Cd on montmorillonite from concentrated NaCl solution is promising with respect to its use as a tracer for monitoring flow through formations containing montmorillonite.

Zn- and Ca-adsorption equilibria of five Ca-saturated halloysite samples were measured at equilibrium Zn concentrations up to 10−2 M in 10−4 to 10−2 M CaCl2. The results were interpreted on KCaZn vs. [Zn]/CEC (%) plots, where KCaZn is the selectivity coefficient (= [Zn][Ca]/[Ca][Zn]), Zn and Ca represent the adsorbed species, and CEC is the cation-exchange capacity. All Zn adsorption occurred at cation-exchange sites, and 0.77 to 36.0 meq Zn/100 g clay, which constitutes 9 to 83% of the CEC, was adsorbed with “high selectivities” (KCaZn > 10). The higher values were found for two spherical and one “filmy” halloysites, whereas the lower values were found for two tubular halloysites. The magnitude of their 001 intensity, hydration in interlayer space, CEC, and “free” iron oxide content did not correlate with the selective Zn adsorption, but a good correlation was found between the proportion of “high selectivity” sites for Zn and proportion of “high affinity” sites for H+. The adsorption of Zn at the “high selectivity” sites was not completely reversible, and KCaZn values >1000 were recorded in 0.5 M CaCl2 for Zn which occupied 10–40% of the exchange sites. Selective Zn adsorption decreased with decreasing pH, and all adsorbed Zn was extracted with 0.1 M HCl.

The cation-exchange capacities (CEC) of a previously well-characterized beidellite have been estimated nondestructively from X-ray photoelectron spectroscopy (XPS) data following exchange by various cations. The CEC obtained as an atomic ratio to silicon from Na- and Ca-clays are close to that derived chemically, i.e., 0.10(2), and 0.10(1), respectively, compared with 0.12(1). However, the apparent CEC obtained following K-, Pb-, and Ba-exchange are all about 50% greater. These measurements are complemented by XPS studies of clay exchanged in a solution containing both Ca and Ba ions where XPS gave a total apparent exchange capacity relative to Si of 0.16, with 0.10 from Ca and 0.06 from Ba. Bulk chemical analysis has, however, shown that the true Ba value is only 0.01. It is concluded that the cation excess detected by XPS—a technique with a total sampling depth of only about 100 Å—is concentrated at the surfaces of the clay particles.

Interlayer sodium ions of montmorillonite were exchanged with hydroxy-bismuth polycations which were prepared from bismuth perchlorate solutions by the addition of NaOH. Assuming the charge density of the silicate layer to be unchanged, the compositions of the polycations involved in the exchange can be estimated from the amount of bismuth taken up by the montmorillonite and from the ignition loss between 110° and 800°C. The derived compositions are near [Bi6(OH)16]2+ irrespective of the ratio of OH:Bi in the perchlorate solution. The basal spacing of the hydroxy-bismuth montmorillonites is about 16 Å at 110°C, which corresponds to that of hydroxy-chromium montmorillonite having a high surface area of about 250 m2/g. The surface areas of the hydroxy-bismuth montmorillonites, however, are less than 80 m2/g.

Wyoming montmorillonite, <2-µm particle size, saturated with Na, K, Mg, Ca, and Ba ions, was reacted with uranyl nitrate solutions in the concentration range 1–300 ppm uranium. With constant amounts of clay and solution volume, the adsorption isotherms of uranyl ions on the clay followed Langmuir-type curves with increasing concentration of uranium. The maximum adsorption derived from linear Langmuir plots corresponds to the exchange capacity of the clay. Experiments with solutions of constant volume and constant ionicity, but with variable proportions of uranyl and other cations, showed that uranyl ions were strongly preferred by the clay to Na+ and K+, but less strongly than Mg2+, Ca2+, and Ba2+.

Chemical analyses of uranyl montmorillonites prepared with nitrate solutions, 0.05 M and pH ≃ 2.0, gave interlayer cations (UO2)0,094H0,12. Addition of NaOH to the uranyl nitrate solution to increase the pH to 4.0 gave a montmorillonite with interlayer cations (UO2)0,083Na0,12H0,04. A fully exchanged uranyl montmorillonite was prepared with 0.05 M uranyl acetate solution, pH ∼ 4.0, followed by further treatment with 1.25 × 10−5 M solution, pH ∼ 4.5. The resulting interlayer composition was (UO2)0,19(H2O)1.15 which corresponds with the hexahydrate ion, [(UO2)−6H2O]. X-ray powder diffraction and thermal data are recorded for the fully exchanged uranyl montmorillonite.

The decreasing preference of montmorillonite for K+ relative to Na+ as the clay adsorbs increasing amounts of K+ is shown to be the general rule for the exchange of strongly hydrating ions by weakly hydrating ions. Variability in the mass-action selectivity coefficient is interpreted in terms of a composition-dependent surface entropy, which is a function of the chemical properties of the exchanging ion as well as the nature of the adsorption sites. The generally used mass-action form of exchange equation may only be applicable to exchange systems in which both ions have solution-like mobility at the exchanger surface. It is suggested that experimental variables such as ionic strength can greatly influence the degree of fit of data to a given ion-exchange equation.

The conditions under which an exchanger phase will behave as an ideal mixture are established from thermodynamic principles. It is shown that, if a stoichiometric cation-exchange reaction is reversible, the exchanger phase will exhibit ideal behavior if the Vanselow selectivity coefficient is independent of the exchanger composition. This criterion is applied to some recently published data for Na+-trace metal cation exchange on Camp Berteau montmorillonite. An analysis of the data suggests that, so long as the exchange process is reversible, Na+-trace metal cation exchange produces an exchanger phase that behaves as an ideal mixture.

Two variables must be considered when calculating exchange free energies (ΔG°ex) for 2:1 clays: (1) anionic field strength, as expressed by equivalent anionic radius (ra), and (2) interlayer water content, as expressed by interlayer molality. For smectites that are in a state of high hydration, interlayer molality is determined by the cations undergoing exchange. Thus ΔG°ex for an exchanging cation pair can be calculated solely from measurements of ra. ra is related to layer charge per half unit cell (C) and ab unit cell area (A) by: ra = (-A/8πC)1/2. The layer charge necessary for cation fixation can be predicted by calculating the ra at which cation exchange with an illite structure expresses a AG°ex equal to that of exchange with a smectite structure. The theory can also be applied qualitatively to understand the high selectivity of illite for Cs+, the fixation of K+ rather than Na+ in shales during diagenesis, the stability of illite over muscovite in the weathering environment, and cation segregation in smectite.

Exchange isotherms for the pairs Na-K and Na-Ca were measured by use of 0.1 N solutions at 5°, 35°, and 70°C in phillipsite from Tecopa, California (3.63 Al/32 oxygen unit cell), and Oki Islands, Shi-mane Prefecture, Japan (6.31 Al/32 oxygen unit cell). All isotherms except those for Na-Ca at 5°C were reversible. Free energy was evaluated for all reversible exchanges. The thermodynamic affinity sequences were K > Na > Ca in both phillipsites. The selectivity for K in competition with Na and that for Na competing with Ca became larger at the lower temperatures. The siliceous phillipsite preferred the larger cation more strongly for the Na-K system, and Na more strongly for the Na-Ca system than the aluminous phillipsite.

Ion-exchange experiments in expanding clay minerals conducted over a wide range of surface ionic compositions and ionic strength produce variable mass-action selectivity coefficients. When the exchanging ions are of unequal charge, tactoid structure appears to influence selectivity, although configurational entropy of adsorbed ions may also generate variable selectivity. The degree of deviation from ideal mass-action exchange is related to the dissimilarity of the ions undergoing exchange. Data involving trivalent ion adsorption on smectites suggest that mass-action is a poor approximation when the adsorbing and desorbing ions have different hydration energies and charge. No form of exchange equation is successful in describing ion exchange for a wide range of experimental conditions, although the fluctuation of the selectivity coefficient follows consistent trends with changing experimental conditions. The strong adsorption of high-charge ions on clays is not exothermic, but must be driven by the increasing disorder of ions and/or water.

An assemblage of clay minerals, Na-zeolites, Fe- and Sb-sulfides, and borates occurs in cavities in the basalt flow that underlies the main borax ore body exposed by the U.S. Borax open pit at Boron, Kern County, California. Analcime and saponite are widespread in cavities and fractures near the top of the basalt. Part of the basalt recently exposed, although restricted in distribution, contains a much more diverse assemblage of diagenetic minerals. Although only a few different minerals occur in each cavity, the composite order of deposition is: ferroan saponite, pyrrhotite, saponite, phillipsite, gmelinite, clinop-tilolite, herschelite, analcime, greigite, rhodochrosite, searlesite, borax, calcite, and colemanite. In some cavities early-formed zeolites appear to have dissolved as later ones crystallized.

Microprobe analyses of the zeolites yielded the following compositions: phillipsite, Na4.28K0.08(Al4.36-Si11.64O32)·xH2O (Si/Al = 2.67); gmelinite, Na6.64K0.10(Al6.80Si17.22O48· xH2O (Si/Al = 2.53); herschelite, Na2.96K0.24(Al3.21Si8.79O24)· xH2O (Si/Al = 2.75); clinoptilolite, Na6.48K0.56(Al7.15Si28.77O72) · xH2O (Si/Al = 4.02); analcime, Na14.02K0.09(Al13.33Si34.44O96)· xH2O (Si/Al = 2.58).

These zeolites formed by reactions between the basalt and evolving fluids, which were controlled by the development of an overlying, Na-borate lake. The wide distribution of saponite, analcime, and searlesite suggests that these minerals formed by the diagenetic reaction between basaltic glass and Na- borate water. The localized occurrence of the complex mineral assemblage, including some of the saponitic clay and the zeolites, phillipsite, gmelinite, and clinoptilolite, formed through a different process, perhaps deuteric or hydrothermal alteration of the basalt, before the lake developed. Because these zeolites are extraordinarily rich in Na, cation exchange may have taken place as the pore fluids became increasingly Na rich. In the latest stages of mineral growth, each of these zeolites was apparently partially dissolved and epitaxially overgrown by a second generation crystal, phillipsite on phillipsite, herschelite on gmelinite, and sodian heulandite on clinoptilolite.

Batch cation-exchange experiments were performed on chlorite, using synthetic ground water containing cesium in the concentration range 10−9–10−3 M. The sorption behavior was complex; with increasing Cs concentration, the distribution ratio of Cs decreased at first, but passed through a pronounced maximum at high Cs loadings. The desorption data differed from the sorption data only by a shift towards larger distribution ratios in the region of the maximum. The distribution ratios for isotope exchange were even larger, but varied less with Cs loadings. Similar results were obtained for KCl-treated chlorite; however, grinding or acid-treatment of the chlorite considerably altered its behavior, the absence of the maximum being the most striking change. Most of the effects at higher Cs loadings can be explained by postulating structural changes induced by the sorption of Cs.

The effects of exchangeable Na and Ca and the ionic strength of NaCl and CaCl2 solutions on the boron adsorption by the Na and Ca forms of montmorillonite and illite at pH 9 were studied. The boron adsorption by montmorillonite was greater for the Ca form than for the Na form at any boron activity in solution and at any ionic strength. For example, the amount of boron adsorbed by Na- and Ca-montmo-rillonite was 2.9 and 4.3 μmole/g, respectively, at an ionic strength of 0.36 and a boron activity of 0.875 mmole/liter in equilibrium solution. The boron adsorption by illite was much greater than by montmorillonite. For example, the amount of boron adsorbed by Na-montmorillonite and Na-illite was 1.5 and 8.7 μmole/g, respectively, at an ionic strength of 0.02 and a boron activity of 0.5 mmole/liter. However, the nature of the exchangeable cation had little effect on the boron adsorption by illite. The effect of the ionic strength on the boron adsorption by Na-montmorillonite was greater than that observed for either Camontmorillonite or illite. The data suggest that the negative electric field around the clay particles is one of the main factors controlling boron adsorption at alkaline pHs.

Synthetic sodium bimessite, having a cation-exchange capacity (CEC) of 240 meq/100 g (cmol/kg) was transformed into Li, K, Mg, Ca, Sr, Ni, and Mn2+ cationic forms by ion exchange in an aqueous medium. Competitive adsorption studies of Ni and Ba vs. Mg showed a strong preference for Ni and Ba by bimessite. The product of Mg2+-exchange was buserite, which showed a basal spacing of 9.6 Å (22°C, relative humidity (RH) = 54%), which on drying at 105°C under vacuum collapsed to 7 Å. Of the cation- saturated bimessites with 7-Å basal spacing, only Li-, Na-, Mg-, and Ca-bimessites showed cation exchange.

Heating bimessite saturated with cations other than K produced a disordered phase between 200° and 400°C, which transformed to well-crystallized phases at 600°C. K-exchanged bimessite did not transform to a disordered phase; rather a topotactic transformation to cryptomelane was observed. Generally the larger cations, K, Ba, and Sr, gave rise to hollandite-type structures. Mn- and Ni-bimessite transformed to bixbyite-type products, and Mg-bimessite (buserite) transformed to a hausmannite-type product. Li-bimessite transformed to cryptomelane and at higher temperature converted to hausmannite. The hollandite-type products retained the morphology of the parent bimessite. The mineralogy of final products were controlled by the saturating cation. Products obtained by heating natural bimessite were similar to those obtained by heating bimessite saturated with transition elements.

Electron spin resonance (ESR) spectra of Cu2+-hexadecylpyridinium (HDP) montmorillonites were investigated as a function of HDP+ content and the hydration state of Cu2+ at relative humidities of p/p0 = 0.52-8 × 10-7 at 298°K. The symmetry of the Cu2+ ESR spectra and the intensity of the ESR signal increased upon dehydration of the complex. The HDP+ cation caused an increase in the hydration state of Cu2+ at a given p/p0 and an increase in the covalency of the Cu-O bond.

To evaluate the importance of oxides to the surface chemistry of acid mineral soils, clay fractions were separated from a surface and subsurface horizon of an Inceptisol representative of many of the acid soils of the Southern Tier of New York state. Portions of the clays were treated to remove selectively noncrystalline and microcrystalline Fe and A1 oxides (acid ammonium oxalate extraction), total free iron oxides (dithionite reduction in buffered citrate solution), and organic matter (hypochlorite oxidation). Charge and ion-adsorption characteristics of the treated and untreated clays were investigated by means of Ca2+- and Cl−-exchange capacities, potentiometric titrations, and electrophoretic mobility (zeta potential) measurements of the CaCl2-treated clays.

Based upon surface area and anion- and cation-exchange measurements, the Fe and A1 oxides or oxideorganic matter complexes were found to contribute a large part of the surface area and pH-dependent charge of these clays. Oxide removal increased the cation-exchange capacity (CEC) and virtually eliminated the anion-exchange capacity (AEC) at pH 3 and 5.5 while shifting the positive zeta potential (ZPC) of the B-horizon clay toward negative values. Organic matter oxidation increased the AEC at pH 3 and the CEC at pH 5.5 and markedly shifted the ZPCs of both A- and B-horizon clays toward more positive values, probably by the removal of adsorbed organics from oxide surfaces. Estimates of the ZPCs of the clays varied among the three methods used, Ca2+- and Cl−-exchange capacities giving the lowest, and electrophoresis giving the highest values.