The conditions for the formation of various heteropolytypic or mixed-layer phases were determined in a hydrothermal study of synthetic mixtures along the join talc—Na-saponite—Na-phlogopite composition. Heteropolytypes of talc and saponite from compositions between talc and a saponite solid solution and of saponite and a 10 Å sodium mica (Na-phlogopite?) from compositions lying between saponite solid solutions and Na-phlogopite have been prepared reproducibly.

At 1 kb in the temperature range 450-575° C, randomly stacked heteropolytypic phases were formed and persisted. At higher pressures the regular heteropolytypes (regularly interstratified or mixed-layer minerals), generally 1:1 phases, were formed. The genesis of heteropolytypes in nature is discussed on the basis of these results.

The Upper Eocene Jackson Group is the chief host rock of the uranium deposits in the Karnes area of the southeast Texas Coastal Plain. It is highly tuffaceous and the uranium deposits are within approximately 100 ft of the unconformity with the overlying Catahoula Tuff (Miocene?). Glass shards, fragments of sanidine and plagioclase, grains of finegrained volcanic rocks, and biotite and other minerals in these sediments were highly reactive chemically, causing complex diagenetic alteration and the development of alkaline carbonate pore water. Extensive caliche development and silica induration associated with a recent hotter, drier climate favored the concentration of uranium. The origin of these shallow uranium deposits is believed to have been controlled by the complex diagenesis of the highly reactive volcanic detritus, by development of a “builtin” solvent for uranium (the alkaline carbonate pore water), and by climatic, structural, and permeability conditions that allowed concentration and deposition of uranium rather than dilution and dispersal. The tuffaceous rocks are considered to have been the source of the uranium and associated molybdenum, phosphorus, and arsenic.

The imbibition of liquid by argillaceous rocks containing predominating kaolinite, illite, “open” illite, and montmorillonite with the use of absorption apparatus is described. Results of the following studies are given: (1) imbibition by pulverized and freely poured samples; (2) imbibition by specimens pressed (30 atm) into the form of small cylinders having a base of 100 mm2; (3) the course of imbibition by artificial aggregates of kaolin from Sedlec, Bohemia, pressed under different pressures (20, 30, 40 atm); and (4) imbibition by natural, oriented, ground, flat sections of the same samples which were prepared in the form of rectangular prisms, with a base of 100 mm2.

The constant k in the sorption equation, which is dependent on the form of clay particles and the tortuosity of pores in argillaceous aggregates under particular working conditions, has been computed. This constant has been calculated for water (polar liquid) and tetraline (nonpolar liquid).

A new quantitative diagram for identification of clay minerals by the imbibometric method has been constructed. In it, not only the time of imbibition and the qualitative features expressing themselves on the ground, flat section—such as swelling and nature of contours (Konta, 1961)—are taken into consideration, but also the area occupied by a drop of water and a drop of ethylene glycol and the wettability ratio Pegl/aq.

Vermiculite occurrences can be divided into two major categories, those having macro-scopic and those having microscopic or clay vermiculite. This paper is concerned with the former. The macroscopic type of vermiculite deposits can be subdivided into four categories based on the host rock: (1) ultramafic and mafic; (2) gneiss and schist; (3) carbonate rocks; and (4) granitic rocks. Field and laboratory evidence strongly suggests that most and possibly all macroscopic-type vermiculite deposits formed by the action of supergene solutions on biotite and phlogopite. Vermiculite in mixed-layer association with biotite, phlogopite, and chlorite is likewise believed to be of supergene origin.

Formulas calculated from modern analyses of vermiculites and hydrobiotites show that these minerals resemble phlogopites and magnesian biotites in tetrahedral and octahedral layer composition, but usually have a higher proportion of trivalent octahedral cations. None of the formulas studied resembles those of ferroan biotites in composition.

In a few formulas the negative layer unit charge and the neutralizing positive interlayer charge are as high as those found in phlogopite; about 1.00 in the half-cell formula, but in most vermiculite formulas the charge on the layer unit and the interlayer charge are considerably lower than in phlogopite. Such a decrease in the amount of these charges may be due to oxidation of Fe2+ to Fe3+, which would increase the positive octahedral charge, and, consequently, decrease the layer unit charge. Oxidation of Fe2+ to Fe3+ also explains the higher proportion of trivalent octahedral cations in vermiculites and hydrobiotites.

Few papers furnish both cation exchange data and analyses of the samples studied; consequently it is difficult to correlate interlayer charge and cation exchange capacity, but empirically the interlayer charge multiplied by 200 yields a value that approximates the cation exchange capacity, provided the Mg has been allotted correctly between octahedral and interlayer positions.

On the basis of composition and charge relations, vermiculites may be interpreted as formed from phlogopites or magnesian biotites by replacement of K by Mg.

Analysis of the argillaceous fraction of some podzolic soils and podzols of France has shown a double evolution of illite and chlorite through mixed layering into vermiculite. $ilite \to \left( {illite - vermiculite) \to vermiculite} \right) \\ chlorite \to \left( {chlorite - vermiculite} \right) \to vermiculite \\$

Vermiculite is more concentrated in the upper zones of the profiles where it constitutes 50 to 90 percent of the argillaceous fraction.

The weathering of micas to form the dioctahedral analogue of vermiculite is a common process in acid soils. Laboratory weathering of the micaceous 5–2μ fraction of parent materials of certain Virginia soils resulted in a product similar in character (whether mixtures of vermiculite and regularly and randomly interstratified mica—vermiculite, or vermiculite only) to those formed by natural weathering processes. The formation of vermiculite rather than (or, in addition to) kaolinite is attributed to substitution of Na and other ions for K, other structural imperfections, and the small size of the original mica flakes.

The low cation exchange capacity and stability, in contraction and expansion, of the 14Å product is attributed partly to retention of hydroxy-Al and Fe groups in the interlayer spaces. The removal of these interlayers by natural or laboratory processes results in a product that contracts easily but may or may not expand to 18 Å on glycerol solvation. This may be associated with the original charge of the mica as evidenced by its K and Na content.

The clay mineralogy of a soil profile developed in a New England glacial till has been studied to a depth of 55 in. Degraded soil micas, probably produced by weathering, have been chloritized by both iron and aluminum hydroxide complexes. Aluminum and iron determinations and X-ray analyses indicate that iron chloritization has played the greater role. Citrate-extractable chlorite exists to a depth of 55 in. X-ray and heat stability data indicate that the chlorite is better developed or crystallized near the surface. X-ray and glycol retention data suggest that intense weathering at the surface has reduced the surface charge density of some of the degraded mica.

A complex of vermiculite with ethylene glycol was prepared by repeated boiling of a sodium-saturated cleavage flake in dry ethylene glycol. A stable product with a c period of 12.9 Å afforded 15 terms for a one-dimensional Fourier synthesis. A model composed of glycol molecules with the plane of the zigzag normal to the silicate face is found to be consistent with the observed diffraction data. The complex apparently conforms to a stoichiometric ratio of two molecules per unit cell per layer of vermiculite.

Various workers have studied the mechanism of swelling of Na-montmorillonite and have demonstrated that its ability to swell is due to water penetrating between the individual silicate sheets. In the present study intraerystalline swelling was followed by diffraction techniques which enabled the movement of the silicate sheets with respect to one another to be measured as a function of electrolyte concentration, with or without an externally applied load. The observations have been made on oriented flakes of Na-montmorillonite and single crystals of Li-vermiculite.

Qualitatively the minerals behave similarly when swollen in salt solutions. They both show an initial stage of crystalline swelling, after which there is an explosive increase to the gel state, and then the distance apart of the silicate sheets increases linearly with C-½, where C is electrolyte concentration. Quantitatively, however, their swelling is very different, particularly after the “explosion”, where vermiculite generally gives higher spacings than montmorillonite. There are two other important differences in the swelling of montmorillonite and vermiculite. Firstly, while the swelling of vermiculite appears to be reversible both with respect to electrolyte concentration and applied pressure, the swelling of montmorillonite shows a marked hysteresis. Secondly, in montmorillonite, swelling depends very strongly on pH or chemical treatment.

There is little doubt that the development of diffuse “double layers” gives rise to repulsive forces, which cause the silicate sheets to move apart. Van der Waals’ forces have been regarded as providing attraction, but it is found that, at the observed interlayer separations, the magnitude of these attractive forces is inadequate to balance repulsion, both in montmorillonite and vermiculite.

The present results demonstrate that in vermiculite swelling proceeds until the interlayer separation causes the repulsion between sheets to drop to a value of ~ 2.5 × 104 dyn/cm2. This attractive force, which limits swelling, appears to be independent of sheet separation and electrolyte concentration. In montmorillonite swelling is opposed by edge-to-face bonds between sheets, whose number and strength can be controlled by chemical treatment. These bonds act to resist any displacement of the sheets, either swelling or contraction. This mechanism is compatible with the observed swelling behavior of montmorillonite, and explains the differences between the swelling of this mineral and that of vermiculite where, because the sheets are considerably larger, the force arising from edge-to-face bonds is small or absent.

An analytical method is described whereby the exchange of various organic cations for other organic cations on montmorillonite is determined. The exchange is followed by means of a Beckman DK-1 ultraviolet spectrometer.

It has been observed that organic cations from compounds such as primary, secondary, and tertiary amine salts and quaternary ammonium salts can be exchanged for one another on the natural cation exchange substance, montmorillonite. The exchange is carried out in isooctane to which a small amount of isopropyl alcohol has been added to aid in dispersion of the organic-clay complex. By choosing pairs of organic substances, of which one has strong absorbance in the ultraviolet and the other does not, the exchange can be followed quantitatively with the u.v. spectrometer.

Basal spacing measurements were made for a montmorillonite saturated with a series of small aliphatic primary, secondary, and tertiary amine and quaternary ammonium cations, using wet, oven-dry, and glycerol-treated specimens. The spacings ranged from about 12 to 14Å, depending on the cation, indicating that in each case a monolayer of cations is interleaved between adjacent montmorillonite layers. The observed spacings are consistent with the concept that the cations are oriented with their minimum thickness in the c-axis direction, but it is possible that the smallest cations have a long axis in this direction and are partially embedded in the clay surfaces.

Montmorillonite saturated with cations containing up to four carbon atoms retained some sensitivity to water, as indicated by a slight contraction of the lattice on oven drying and rapid re-expansion on exposure to the humidity of the laboratory atmosphere.

For cations having a layer thickness less than glycerol, limited lattice expansion occurred on glycerol treatment, usually to a spacing sufficient to accommodate a single layer of glycerol molecules. Apparently the cations function as pillars separating the montmorillonite layers, and the glycerol molecules may expand the lattice as necessary to enter the spaces between the pillars. However, exceptions occurred with methylamine-saturated montmorillonite, in which both one- and two-layer complexes of glycerol were present simultaneously in the same specimen, and with trimethylamine-saturated clay, in which two different one-layer complexes were present. It is suggested that the glycerol in one-layer montmorillonite complexes may exist in two orientations, one leading to a layer thickness of about 4.1 Å, the other to a thickness of about 4.6 Å.

Portions of Wyoming bentonite were saturated with a series of short-chain aliphatic amine and quaternary ammonium cations, and the amine content remaining after washing was determined by combustion analysis. For those amine cations containing one or two carbon atoms, the amount of amine found was greatly in excess of the cation-exchange capacity. It is thought that this excess consists of amine cations or uncharged amine molecules which have become partly embedded in the holes associated with the oxygen rings of the silica sheet surfaces. For the larger tertiary amine cations, slightly less than the exchange capacity was found, presumably owing to blocking of some exchange sites by steric interference.

Upon treatment of the amine-saturated clays with glycerol and quantitative determination of the glycerol retained as a monolayer, it was found that these clays retain considerably less glycerol than the original bentonite. The glycerol retention values were employed to provide quantitative estimates of the interlayer clay surface left unoccupied between cation pillars. The percentage of original interlayer surface left unoccupied ranged from about 75 percent for methylamine to essentially zero for the larger tertiary amine cations.

The validity of the areas calculated from the glycerol retention results was indirectly confirmed by deriving estimates of the cross-sectional areas of the several cations from the observed reductions in amount of glycerol retained. These experimentally derived cross-sectional areas were found to be in satisfactory agreement with areas of the cations projected from atom models.

More than 8000 different derivatives of mica-type layer silicates (muscovite, biotite, illite, vermiculite, beidellite, montmorillonite) with alkylammonium ions and different organic molecules as swelling liquids were prepared and characterized analytically by X-ray and chemical methods. The following results have been discussed in some detail.

Layer-lattice compounds are formed in simple stoichiometric relations. In n-alkyl-ammonium compounds, for instance, the sum (n-Cx,H2x+1NH3++n-CxH2x+1Y) per (Si, Al)4O10 is equal to 2.0 (Y = OH-, Cl-, Br-, NH2-, COH-, CONH2-, CN-).

The layer distance of the pure n-alkylammonium derivatives enables one to determine the charge density in the silicate layers. But in soils the ability of humic acids to undergo two-dimensional intracrystalline swelling and to react with alkylammonium ions must also be considered.

The properties of n-alkylammonium layer silicates are similar to those of alkyl-ammonium-uranium micas and of swelling vanadates (e.g. hewettite and pascoite). The swelling properties of alkyl-(α, ω)-diammonium ions are markedly influenced by the opportunities for hydrogen bonding. Mica-type layer silicates may show catalytic properties like protease or oxidase with ammonium ions, which are of general interest.

A detailed investigation of the adsorption of phenol and m-cresol from dilute aqueous solution by dodecyl- and tetradecylammonium bentonites has been carried out. The Freundlich isotherms for the tetradecylammonium bentonite system indicate that during adsorption, the process proceeds in two stages, the partial molar heat of adsorption changing from a positive to a negative value with a relatively small change in concentration of adsorbate. In the light of these findings, the aspects of the adsorption are discussed in terms of thermodynamic quantities, spatial considerations, and interlamellar expansion of the clay mineral.

Studies of quaternary ammonium chlorides containing long unsaturated hydrocarbon chains have shown their effectiveness in stabilizing soils primarily by waterproofing the clay fraction. The nature of the interaction between organic salts and Wyoming bentonite was studied by means of X-ray diffraction.

Treatment of bentonite with a dialkyl dimethyl ammonium salt in an amount less than the exchange capacity results in the appearance of water-stable spacings of 10, 13 and 19 Å, but only if the sample has been dry cured. Moist curing a bentonite paste results in a marked expansion of the lattice despite treatment. The necessity of dry curing may relate to orientation of organic cations around bentonite aggregates; grinding prior to rewetting destroys the waterproofing action, and the clay lattice again expands.

Additions of the quaternary ammonium cations close to or in excess of the exchange capacity of the bentonite waterproofs regardless of dry curing or grinding, or both. Clays treated in this manner show spacings of the pure crystalline chemicals and other spacings which relate to interlayer penetration by water. There is no direct evidence for interlayer penetration by the organic cation, perhaps because of the unwieldy tetra-hedral arrangement of the organic molecule. The degree to which organic cations are held depends on the nature of the surface as the cations are readily leached from quartz powder but are held more strongly on kaolinite and bentonite.

Halloysite and endellite yield 10 Å basal spacings after grinding with potassium acetate (using a modification of Wada’s procedure) and washing with water and ethylene glycol, whereas the basal spacing of kaolinite, similarly treated, remains at 7 Å. The technique applies satisfactorily to sedimentary and hydrothermal halloysites, various commercial kaolins, ball clays, flint and plastic fire clays, and artificially prepared mixtures of halloysite and kaolinite. Similar responses to the treatment by both laboratory-dehydrated endellite (thereby collapsed to 7 Å) and naturally occurring halloysite add support to the concept that endellite is a progenitor of halloysite.