A Chemical Study of the Dissolution of Hectorite with Special Reference to the Release

I N T R O D U C T I O N The study of the ease with which clay minerals are decomposed by acids can provide useful information about the properties of clays. Many of the earlier studies in this field, reported by Nutting (1943) and Grim (1953), were carried out under rather severe conditions when compared to the natural weathering environment. Investigations of the reactions of clays with strong acids have continued to the present but with greater emphasis on their use as a tool in the study of the structural and surface properties of clays (Brindley & Youcll, 1951; Osthaus, 1954, 1956; A 246 Kevin G. Tiller Mathers, Weed & Coleman, 1955; Grandquist & Sumner, 1959; Gastuche, 1961). Kerr et al. (t956) studied the kinetics of the decomposition of H+-hectorite by titrations over a period of several hours and suggested a mechanism for the decomposition of H+-hectorite. Kalovoulos (1961) also found that H+-hectorite converts quickly to Mg++-hectorite. Studies in silicate weathering are more appropriately carried out with dilute acid media rather than the fairly severe conditions of acid dissolution, such as the requirement of a large excess of acid, which are often necessary in the kinetic studies used in the structural investigations of clays. The treatment of suspensions of hectorite with dilute acids in the laboratory may be considered in some respects as a simple model of silicate weathering in an environment low in aluminium. With this aim, the decomposition of hectorite has been studied as a function of variables such as time of reaction, amount of acid used, the kind and amount of salts added and clay/solution ratio, with particular reference to the chemistry of silica. It was hoped that such data would provide a basis for the elucidation of the chemistry of the reactions involved. In these experiments no attempt was made to produce H+-hectorite as a starting point for weathering studies since its extremely labile character makes its existence in nature very unlikely as well as making it difficult to use experimentally. E X P E R I M E N T A L Preparation of hectorite suspensions Hectorite is a trioctahedral montmorillonite characterized by a partial replacement of Mg by Li in the octahedral layer, with the general composition, (OI-I),Sis(Mgs.s~Li0.eG)O20. The material used in these experiments was separated from the sample A.P.I. No. 34 from Hector, California, as supplied by Ward's Natural Science Establishment, Rochester, N.Y. This clay occurs naturally in a rather impure form and is often mixed with a large proportion of calcite. The coarser impurities were separated by hand-picking, followed by sedimentation and decantation after vigorous mechanical stirring in distilled water. To remove calcite without using strong acids, the finer material (<20/~) remaining was suspended in N NH,C1 solution and heated at 80 ~ C for several days, the solution being changed regularly at least three times daily. Reg-l ,r changing of this solution was necessary to prevent the precipitation of amorphous silica. The sample was dispersed in water and the < 2 ~ fraction separated by a centrifuge method. The purity of the final product, particularly with respect to calcite, was checked by X-ray diffraction and elemental analysis. Some batches required retreatment with NH,C1. X-ray diffraction patterns of the purified product corresponded to that of a pure trioctahedral magnesium montmoriUonite. Electron micrographs of the hectorite used showed that much of the clay consists of elongated crystals, sometimes in fibrous aggregates. The crystals were commonly of the order of 2 ~ long and 200 A wide. After removal of calcite, the clay was saturated with Ca ++, suspended in M/20 CaC12 and diluted to contain 5"0 mg hectorite (105 ~ C oven dry basis) per millilitre of solution. Dissolution of hectorite 247 Analytical methods Dilute solutions of silicic acid were prepared by passing potassium silicate solutions of the appropriate concentrations through a column of H+-saturated strongly acidic cation exchanger. Silicic acid in solution was determined by a spectrophotometric method based on the reduced silico-molybdate complex under conditions which allow measurement of monomeric and dimeric silicic acid. Mg was determined by atomic absorption spectroscopy in solutions containing M/lO0 CaClz. Preliminary tests showed that this concentration of CaClz sufficed to prevent a reduction in the atomic absorption of Mg for the levels of Si introduced into the flame. pH of the clear supernatant solutions was measured using a Radiometer pH meter.