The surface areas obtained by application of the B.E.T. theory to adsorption isotherms of nitrogen and carbon dioxide gases at 77°K and 195°K respectively on homoionic samples of illite and montmorillonite clays have been examined. The isotherms were obtained using a standard volumetric adsorption system and the results are compared with those obtained by Thomas and Bohor (1968) using a dynamic sorption system.

Small amounts of residual water have been shown to have a marked influence on the accessibility of the internal surfaces of the montmorillonite clays to nitrogen and carbon dioxide adsorption, in this respect the standard outgassing procedure under high vacuum seems more efficient than that used in dynamic systems. The present data indicate that provided the sample has been satisfactorily outgassed there is little penetration of nitrogen or carbon dioxide gases into the quasi-crystalline regions of montmorillonite clays. With the exception of the caesium saturated montmorillonites the surfaces of the clays are more accessible to the smaller nitrogen molecules than to carbon dioxide assuming the values used for molecular area are correct.

Release of potassium by extraction with solutions containing sodium tetraphenyl boron or by leaching with 0·1 N barium chloride seemed to be related to the nature of the mica layers in the interstratified minerals studied. The rate of potassium release was lower when the calculated Si/AlIV ratio of the mica component layers resembled muscovite and higher when this ratio was intermediate between muscovite and pyrophyllite. This supported a recent hypothesis that the composition and structure of the mica component layers may vary in different interstratified minerals of similar total chemical composition.

Prompted by Foster’s observation that free swelling is related to octahedral substitution, the authors determined the free swelling of six Na-montmorillonites with different amounts of octahedral and tetrahedral substitution. They found that the montmorillonites exhibited marked differences in free swelling. These differences were not related to differences in cation exchange capacity. Nor were they related to differences in ζ potential, which is a criterion of cation dissociation. Further, calculations indicated that they could not be accounted for by differences in double-layer repulsion or van der Waals attraction. Therefore, to see if dimensional changes produced by isomorphous substitution were responsible, free swelling was plotted against the b-dimension of the clay structure, which was calculated from its mineralogical composition. The result was a straight line with a negative slope. A similar result was obtained with Foster’s data. In addition, free swelling was plotted against the degree of tetrahedral rotation in the clay structure, which was also calculated from its mineralogical composition. The result was a family of nearly parallel straight lines that were distinguished from each other by the amount of tetrahedral Al3+ in the clays identified with them. These results led to the proposal that the clay surface acts as a template for the structure of the adjacent water and that, as the configuration of the surface changes, the water structure changes accordingly. This causes a change in the free energy of the water and, hence, in the swelling of the clay.

The 0.2–5μ particle size fraction of montmorillonite from three sources was equilibrated with various solutions at room temperature. After 3–4 yr, kaolinite was found in some of the samples that were supersaturated with respect to kaolinite, but not in any of the undersaturated samples or in the original montmorillonite. X-ray diffraction analysis of the precipitated kaolinite showed no inter-layer expansion of glycerated, oriented samples. Random powder samples indicated a poor crystal-Unity. The thermal stability of the precipitated material was indistinguishable from that of crystalline kaolinite. The electron microscope did not reveal any distinctive sizes or shapes.

Equilibration behavior of several samples defined a single kaolinite solubility line at or above which kaolinite apparently begins to precipitate. The solubility line is equivalent to a standard free energy of formation (ΔG) of -904.2 kcal per mole of kaolinite. This represents highly crystalline kaolinite. The stability of kaolinite actually precipitated at room temperature probably depends upon precipitation conditions. Thus kaolinite stability could range from poorly crystalline up to the equivalent of the kaolinite solubility line at which initial precipitation begins.

Adsorption studies have been performed on Georgia kaolins having a broad range of crystallinity and particle size distributions (from 0·1μ to 44μ) using N2 (78°K.), H2O (273°K.), and BuNH2 (298°K.). Using both vapor and liquid phase adsorption techniques, surface affinities of the adsorbates were determined.

Modified Frenkel-Halsey-Hill plots were used to compute the preferential adsorptivity of H2O vapor over N2 (hydrophilicity index, H.I.) as a function of crystallinity index, C.I., and particle size. For amine adsorptivity, non-aqueous adsorption isotherms were obtained.

Within any geographic deposit, crystallinity exhibits an inconsistent pattern with respect to particle size. A single generality is the tendency for crystallinity to increase toward the fine particle size range, D → 0·2μ. Adsorptivities of N2, H2O, and BuNH2 show no dependence upon crystallinity within a given particle size range. However, F.H.H. compensated slopes, describing the preferential adsorptivity over N2, show a definite decrease as crystallinity increases. A striking anomaly occurs in the vicinity of 0·2 > C.I. > 0·7 where H.I. increases briefly then returns to the original trend. The rate of decrease of H.I. vs. C.I. is consistently steeper with increasing particle size. Adsorption of water vapor most likely occurs as a 1:1 configuration on each silica-alumina edge group, 1:1 on each basal silica, and 1:2 (hindered configuration) on each basal alumina group.

The data suggest that amines adsorb preferentially and quantitatively on the edges, i.e. the Lewis and Bronsted acid sites, and follow a Langmuir pattern.

A hydromuscovite in association with gypsum and anhydrite was collected from the Shakanai mine, Akita Prefecture, Japan. Chemical composition: SiO2 47·14%; TiO2 0·34%; A12O3 37·09%; Fe2O3 0·49%; MgO 0·83%; CaO 0·57%; Na2O 0·35%; K2O 7·10%; H2O+ 5.18%; H20–0·99%; P2O5 0·01%; total 100·09%. Differential thermal and i.r. absorption analyses were similar to those of hydromuscovite. The X-ray diffraction pattern differed clearly from those of the 1M and/or 2M1 polymorphs and it was similar to that of the 2M2 polymorph, which is known to occur in lepido-lites.

Electron probe micro-analysis studies on individual particles (40–60 mesh) of weathered micas treated with solutions containing equivalent amounts of Rb and Sr showed partial segregation of these elements. Rb was concentrated at particle and step edges, at cracks, and, in the case of partially K-depleted biotite, at boundaries of vermiculite and mica zones (“wedge zones”). The scarcity of wedge zones in mica from which nearly all of the K had been removed reduced the overall selectivity for Rb. The restricted exchange of interlayer Mg ions from vermiculite-like zones by a mixed Rb-Sr solution was observed in earlier studies with these micas. The proposed explanation for these results was a closing down of the interlayer space at the edge of the particle due to Rb concentration in these positions. This explanation is confirmed by the present study.

The accurate optical observation of alteration fringes developing in phlogopite flakes, shows that the morphology of the fringes is specific for some cations, such as Na, Mg and Ca, replacing K.

A careful chemical study of the exchange kinetics of interlayer K shows that, beside the classical effect of blocking the exchange reaction with cations such as K, Rb, Cs, NH4, it is possible to induce an increase of the rate of K exchange when adding to a concentrated solution of one cation, a very small amount of other cation such as Na or H. The effects of mixtures such as Ca-Na and Ca-H are reported here in detail. Attention is drawn to the decisive part played by the impurities which may be contained in the reagents used.

Some authors have stated that anionic surfactants are not adsorbed by montmorillonite while others indicate very minor adsorption. Attempts to quantitatively determine the degree of adsorption have shown that certain problems exist which, unless recognized, will completely mask results. Some of these difficulties have been overcome but exact data have not yet been obtained. However, enough information has been gathered to prove that adsorption of anionic surfactants on montmorillonite is real and substantial. Reliable adsorption data for cationic and nonionic surfactants and one having both nonionic and anionic character have been obtained. It is indicated that 550 mg of these surfactants are adsorbed by 1 g of montmorillonite. X-ray diffraction data for complexes of all surfactants investigated confirm positive adsorption. However, the thickness of the adsorbed surfactant layers cannot always be quantitatively related to the amount of adsorbed surfactant.

A study was conducted to (1) determine the conditions of hydroxy-Mg interlayer formation with respect to type of clay mineral, acidity, and time; (2) evaluate the stability of this interlayer to dissolution treatments; and (3) ascertain the effects of such treatments upon the determination of clay minerals in soils and sediments. Hydroxy-Mg interlayers were formed in montmorillonite and vermiculite by adding MgCl2 and NaOH in amounts to give a wide range of pH. The resulting chloritic intergrades were examined after 10 days, 6 months, and 1 yr.

Alkaline conditions favored the formation of hydroxy magnesium interlayers in phyllosilicates. Hydroxy-Mg interlayered montmorillonite which resulted from 10 days equilibration at pH 10·4 did not expand upon solvation with ethylene glycol and exhibited practically no collapse after K-saturation and heating at 550°C. A small amount of interlayer was formed between pH 6·8 and 9·8 (10 days). In contrast, vermiculite exhibited no evidence of interlayer formation at pH values up to 9·7 (10 days). Chloritic intergrades formed at pH 10·7 did not collapse after K-saturation and heating at 300°C but did so at 550°C. Hydroxy-Mg interlayers were not formed in either mineral by using a drying method. This method apparently failed to provide the required alkaline conditions for interlayer formation.

The amount of magnesium interlayers present in the phyllosilicate systems decreased with time. The interlayers formed in vermiculite decreased more sharply than those in montmorillonite.

Sequential dissolution treatments included boiling 2 per cent Na2CO3, buffered sodium citrate-dithionite, a second citrate-dithionite treatment, and boiling NaOH. Hydroxy-Mg interlayers in montmorillonite exhibited a higher stability to sequential treatments than the interlayers formed in vermiculite. A stable 14 Å line was observed in interlayered montmorillonite after the dithionite-citrate and NaOH treatments.

The interlayers in montmorillonite showed a relatively high stability to HCl dissolution treatments. In contrast, most of the magnesium interlayer in vermiculite was removed by two HCl washings.

The reagents used in this study are sometimes used to remove coatings and cementing agents from soil surfaces prior to particle size and clay analysis. The present data show that these treatments also remove some hydroxy-Mg interlayers and produce changes in properties of clays. A proper interpretation of data for clay mineral identification and characterization must recognize these changes due to treatment.

Subsurface samples of shales ranging in age from Pleistocene to Eocene obtained from five Gulf Coast oil wells were examined mineralogically and chemically to determine the nature and extent of burial diagenesis in pelitic sediments. Illite/montmorillonite dominates the mineralogy and undergoes a monotonic decrease in expandability from about 80 to a limit of 20 per cent montmorillonite layers with increasing depth. The interstratification changes from random to ordered at about 35 per cent expanded layers. Discrete illite and kaolinite phases are ubiquitous and judged detrital. The discrete illite (mica) content of the whole rock decreases with depth, while the kaolinite shows no systematic variation. Chlorite occurs in appreciable amounts in only one well and there only in samples from the shallow water facies. This chlorite is also considered detrital.

Whole rock chemical analyses show no systematic variation with depth except for a decrease in calcium and magnesium caused by solution of carbonate. In contrast, potassium increases progressively with depth in the clay-size fraction, indicating a redistribution of potassium within the rock. Detrital illite (mica) seems to break down with increasing depth, thereby supplying potassium for interlayer fixation in illite/montmorillonite as the proportion of illite layers increases. The diagenetic reaction is independent of the geologic age of the sediments and of stratigraphic boundaries. Temperature seems to be more important than pressure in governing the reaction.

The microelectrophoretic and adsorption behaviour of lithium vermiculite has been studied as a function of lithium chloride concentration. This was done in an attempt to establish the applicability of such systems to the testing of theories of interaction of flat plates, and in so doing to throw further light on swelling measurements performed on such materials. The studied behaviour, while highly unusual, gave quite good agreement between adsorption and microelectrophoretic parameters and agreed, qualitatively, with some earlier measurements on similar materials.

The observed properties appear to be due to some rather specific structuring effects, either of the oxide surface or of the electrolyte ions. If this is so, these systems are far from ideal models for the testing of the theory of interaction of two uniform flat plates.

Two octyl-ammonium vermiculite complexes with different 001 periodicities have been studied by i.r. spectroscopy. In each case i.r. spectroscopy affords information on the orientation of the — NH3+ groups and the strength of the hydrogen bond between these groups and the silicate oxygen surfaces. Also a perturbation of the vibration of the OH groups of the silicate has been observed that seems to be related to the distance from the center of the layer at which the — NH3+ groups are situated.

The i.r. results are discussed in relation to the structural models deduced from X-ray analysis.

The thermal reaction sequence of a synthetic hectorite (Laponite CP) was studied by X-ray diffraction, i.r. spectroscopy and thermal analysis. Although most of the interlayer water is removed at 200°C, a smally steady weight loss occurs until dehydroxylation is complete at about 700°C, indicating that an anhydrous intermediate phase is not formed prior to dehydroxylation. Immediately after dehydroxylation, enstatite and cristobalite can be identified, but lithium silicates are formed only from lithium-saturated hectorite. Around 1200°C a glass is formed by reaction of the alkalis with cristobalite, and removal of silica from the enstatite produces some forsterite. An inhomogeneous mechanism of dehydroxylation is postulated by analogy with that proposed for talc.

Simultaneous X-ray difffraction-differential thermal analysis equipment for an X-ray diffractometer was evaluated. A brief description of the apparatus and of the method of preparing the sample was given.

Simultaneous XRD-DTA was used to study various clay minerals and inorganic hydrates. It was found in the study of clays such as hectorite, montmorillonite, kaolinite, and attapulgite that information concerning the rate of removal of water and hydroxyl units at various temperatures could be correlated to changes in the X-ray diffraction pattern. Various inorganic hydrates such as barium chloride dihydrate, nickel sulfate hexahydrate, and potassium ferrocyanide trihydrate were examined under dynamic and static conditions of analysis along with vacuum conditions. It was found that good correlation could be obtained between the X-ray diffraction pattern and the loss of water from the hydrates as indicated by the DTA curves.

Selectivity of a number of vermiculites, montmorillonites and micas for K and Cs ions was determined by sorption of these ions from equilibrium solutions of diverse concentrations. The selectivity coefficients were related to the layer charge density and the area of the frayed edges in layer silicates.

Montmorillonites had the smallest selectivity for the two ions, while biotite and illite had the greatest selectivity. Selectivity of biotite and illite was limited to small concentrations of K, however. At greater concentrations the selectivity of vermiculite for K exceeded the selectivity of the micas.

The greater selectivity of vermiculites than montmorillonites for K and Cs ions was attributed to the greater layer charge density in vermiculites. The greater selectivity of micas than montmorillonites and vermiculites was attributed to the frayed edges of micas in addition to their larger layer charge density. As the frayed edges in illite were increased in area by removal of the interlayer K, the selectivity of illite for K also increased; thus confirming the selectivity of frayed edges for the K ions.

Recurrent reports of analyses of X-ray diffraction data from regularly alternating and nearly regularly alternating structures like rectorite provide opportunity for a descriptive discussion of their nature. MacEwan-transform-based analyses have uniformly shown higher incidence of unlike adjacent layers than would be expected from random distribution.

The continuous Fourier transforms for pairs of 2: 1 layer silicate structures evaluated normal to the layers, provide a family of similar curves that are differently stretched out in reciprocal space for different assumed intervals between the two members of a pair.

For each exactly alternating example the relative amplitudes of the ordered reciprocal nodes distinguish the instances for which fixed interlayer populations are Na+ from those for which they are K+.

For sequences not exactly alternating scattering maxima are displaced from exact incremental positions in conformity with the Hendricks and Teller mixing functions, but relative intensities of adjacent observations still distinguish the alkali identities.

Goethite, lepidocrocite and hematite were reacted with glycerol and reaction products were studied at various steps of the reaction. After a treatment of 16 hr at 245°C the final form of the reaction product, a deep green soft solid, was obtained whatever the starting material. According to hydrolysis and chemical analysis, the reaction product can be identified as iron alkoxide. Hydrolysis of the solid by boiling water yielded glycerol and a strongly magnetic material characterized by a spinel structure like maghemite and magnetite.

The iron alkoxide is seen by electron microscope as large well-formed hexagonal platelets. The similarity between a and b parameters of this compound and those of brucite suggests that the iron alkoxide structure is based on an iron (II and III) octahedral layer. The glycerol chains should be located between the sheets. Two models were proposed in which iron and oxygen, linked to the chain of glycerol, could be accommodated in octahedral layers giving a basal distance of 8·1 Å.

At least two steps are involved in the formation of the iron alkoxide: (1) a rearrangement of the structure of the starting material with accomodation of carbon in the structure and (2) a reduction of ferric iron.

A semi-micro quantity (10-15 mg) sample of clay is pressed into a 3 mm diameter pellet for differential thermal analysis (DTA) in a STONE-TRACOR ring microsample holder. The pellet is transferred to the sample holder dish with a vacuum tweezers. Platinum sample dishes are also handled by vacuum tweezers to avoid denting the dishes or damaging the thermocouple wires. The DTA curves obtained by the pellet method are highly reproducible. The resolution and sensitivity of reaction peaks are equivalent to or better than those produced by a large sample (200 mg) packed into a nickel block holder.