OF the various thermal methods of investigation available, differential thermal analysis has found most application in mineralogy, and particularly in clay mineralogy. The method is now, however, receiving more wide-spread recognition and is being used not only in inorganic and analytical chemistry but also, with controlled-atmosphere techniques, in organic chemistry. It is our opinion that many possible fields for. application in analytical chemistry still exist.The method is fundamentally simple. A sample of the material being investigated is heated up side-by-side with a sample of a thermally inert material in a suitable specimen holder, and the difference in temperature between the two is recorded as they are heated. When no reaction occurs in the specimen there is no difference between the temperatures of the two samples, but as soon as any reaction begins the specimen becomes hotter or cooler than the inert material, and a peak develops on the curve for difference in temperature (AT) against time (t) or temperature (T). In Fig. 1, the peak is represented as BCD, and it should be noted that the reaction is not complete at C., but at some point, X, along the return portion of the peak, CD, whereas the maximum rate of reaction occurs at a point, E, along the line BC. The temperature of the specimen at point C, which is usually referred to as the "peak temperature," is therefore not indicative of any specific stage of reaction, but, being the most easily observable point of inflexion, is usually the criterion quoted. The distance BD is referred to as the "peak width," the perpendic3lar distance to C from a line joining BD as the "peak height'' or "amplitude," the angle :BCD as the "peak angle" and the area enclosed by BECXD as the "peak area."It will be appreciated from the above brief description that the curve is not a true diferential curve, but simply a straightfoward diference curve. I t will also be appreciated that such a curve gives information on reactions occurring in the specimen when it is heated, rather than on the ultimate composition of the specimen. The name is therefore somewhat of a misnomer. Nevertheless, the usefulness of the method lies in the fact that the curve obtained for a specific compound is reasonably characteristic of that compound and that the peak area is related to the amount of reacting materia1.lJJ Although the method is usually traced back to early work on clays by Le Chatelier? it was in fact first developed by Roberts-Austtm5 for metallurgical studies. Adoption of the Le Chatelier -Saladin6 and Kurnakov' recording devices in place of the rather cumbersome original arrangement of Roberts-Austen made the method more attractive for laboratory use, but up to the late 1930's differential thermal analysis was not extensively used, except by restricted schools, and by far the greatest amount of work in this period was carried out in Russia, where the first book devoted largely to the subject was published in 1944.8 From the late 1930's to the early 1950's, most differential therma...
