For several decades clays have been used for decolorising mineral oils. A simple method is percolation of the oil through a layer of clay. Some twenty times the amount of oil is usually treated with such a layer. The first oil filtered through the fresh clay is much more changed than the rest of the oil. It has a very light colour and its low refractive index indicates a high content of paraffinic and naphthenic hydrocarbons.

Most of the aromatic hydrocarbons are left behind in the clay. This suggests adsorption analysis of oils using clay in the same large quantities that have been successful with other adsorbents (Mair and White, 1935; Willingham, 1939; Allibone, 1941). Quantities of the order of 100 gm. alumina per gm. hydrocarbon were used by Winterstein and associates (1934) when separating a number of aromatic hydrocarbons.

The absorption of dyestuffs by montmorillonite can be attributed to two different processes, namely (1) base exchange and (2) van der Waals adsorption. There are some mixed cases, but usually one of the two effects is predominant and can be easily distinguished by several of the following features (see table 1.).

This paper deals with the base exchange type of adsorption only and in particular with the quantitative relations between the adsorbed dye and the released exchangeable cations. The simplest assumption, which is very often taken for granted, is that the two quantities are equivalent and correspond to the base exchange capacity of the montmorillonite.

When moist clay bodies are heated on one side it is often observed that moisture moves towards the cooler side. This phenomenon has been studied using clay-water mixtures of initially uniform moisture content packed into sealed, vertical tubes, along which a constant temperature gradient is maintained, the tubes having a reservoir of water at constant temperature at their lower, cool ends. Figure 1 shows the moisture distributions obtained after various times in columns of a commercial kaolin (Supreme Kaolin) which contained initially 16-18 per cent. moisture. A “stepped” type of distribution results from the movement of the moisture, the step moving down the column with increasing time.

The use of thermal analysis for the identification of clay minerals differs from its use for the determination of phase transitions in that the latter are, in theory at least, reversible and occur at definite temperatures while the dehydration reactions utilized in the identification of clays are monotropic, irreversible decompositions proceeding at a velocity which increases continuously as the temperature is raised. There is, therefore, no definite transformation temperature and the temperature of the arrest registered on a heating curve depends on the heating rate employed.

In view of the increasing importance of thermal analysis it was decided that an attempt should be made to study the implications of this. An investigation of the kinetics of the dehydration reaction seemed the most obvious approach.

This work is part of a more general study of interactions between montmorillonite and organic substances, carried out in connection with the fundamental results of MacEwan and Mackenzie. The object of this work was the study of adsorption of acetone in the vapor phase.

Samples of sodium and calcium montmorillonites were placed in pyrex tubes provided each with a ground glass stopper and a tap. Each tube was joined to a vacuum pump through a desiccating column and heated for several hours to 280°C so as to desorb the water contained in the clay and on the glass walls. At the conclusion of the evacuation, the tube was opened while hot, so as to introduce a recipient containing acetone in solution in castor oil (which had been previously degassed in vacuum). The tube was again closed, re-evacuated, cooled, then isolated from the vacuum pump and put into a thermostat (40°C).

Among the lithological types making up the northern Appennines, an important part is played by the (occasionally schistose) clays which make up the main bulk of an enormous nappe (“liguride,” according to Stille) which has slipped from the Ligurian region towards the Po valley. These clays are particularly well developed in the mid-mountain region, and can, because of their plasticity, insinuate into the more recent formations.

Within the argillaceous nappe, frequent nuclei of basic and ultrabasic rocks (gabbros, diabases, serpentines) are observed, of size ranging from that of a pebble to masses of many thousands of cubic meters (Monte Penna, Sasso Tignoso), which appear immersed in the clay mass without having produced in it the well-known contact phenomena.

This mineral is found in an ahnost pure state, forming a layer 1–2 m. thick at an average depth of about 50 m. beneath ground level, on the northern slope of the Monteti near to Capelbio (Orbetello, Tuscany). Above the layer of micaceous mineral there is developed a series of clays containing halloysite, metahalloysite, montmorillonite and kaolinite, locally rich in iron and manganese hydroxides. The series is quaternary, and lies on top of phyllites which have been attributed to the Permian. The whole series of clayey and ochreous minerals is of low-temperature hydrothermal origin, as I have shown in a recent communication (1948).