PRODUCT STABILITY--PART II 323 It is possible that dilatometry can serve to monitor changes such as hidden phase transitions (formation of solid solutions or immiscible phases) and polymorphic transformations in semisolids such as pastes and ointments. Such techniques were employed by Ravin and Higuchi (58) in part of a series of studies on the melting behaviors of some fats and waxes of pharmaceutical importance. The procedures are reproducible and entail determining the specific volume of a system as a function of temperature. Although it was not done by these investigators, it is possible that samples would be stressed in the usual accelerated storage tests, after which dilatometric studies would be performed. Possibly, a plot of specific volume at a certain temperature rs. time of storage would give data from which predictions could be made. The utility of a differ- ent type of dilatometry, constant temperature dilatometry, has been pointed out by Mahler (59). In this technique the aim is to avoid the effect of temperature on polymorphic changes. Thus, the material is kept at room temperature in something akin to a mercury densitometer. The volume change, either an increase or a decrease, is then followed by recording the displacement of mercury in a connected capillary tube as a function of time. When the per cent change in volume is plotted against time, the curves plateau out, either upward or downward depending on whether a volume increase or decrease takes place. From the early part of these curves, which are essentially hyperbolas, one can calculate the time required to complete the variations in volume which are due to polymorphic transformations. Thus, the time to make observations for surface defects which are caused by such transformations can be pre- dicted for such products as lipsticks, eyebrow pencils, suppositories, etc. Reese, Chong, and Swintosky (60) studied lipid raw materials via photomicrography. They were able to demonstrate that some changes (crystal transformations, growth, cracks) occurred during aging at room and accelerated temperatures. Certainly, the integrity of lipid mate- rials affects the quality of creams and ointments, and it is possible that, even though it may be difficult to quantirate, this camera-microscope technique could be used to detect incipient graininess in ointments. Also, it may be possible to observe the solid phase of suspensions this way. Maclay, Shepherd, and Lotzkar (61), in a study of pectin in medicated pastes and ointments, measured the viscosity and pH of such products stored at various temperatures. Although the pH gradually decreased, the viscosity was relatively stable. However, they could demonstrate by analytical techniques a slow demethoxylation of the pectin.

324 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS Szepesy (62) studied the consistency of ointment bases by obtaining penetrometer readings at 6 ø, 25 ø, 35 ø, and 45øC initially and after storage of the ointments. An interesting property which permits evaluation of lotions, ereams, and ointments is that of tack. Wood and Lapham (63) described an instrument, the tackmeter, which permits the evaluation of tack of such products, both after they are dried on plates or after application to the skin. The tackmeter described is essentially a balance assembly which permits determination of a withdrawal weight and time of break to separate a fiat surface of the device from the skin. Wood, Giles, and Catacalos (64) also described a revised tackmeter in which forces were recorded by a strain gauge instead of by weights. Borehardt and Daniels (65) studied the application of differential thermal analysis to reaction kinetics. Although they used solutions, one could visualize the application of the technique to semisolids such as pastes, ointments, and lipsticks. The method detects the net heat of reaction as the temperature of the system is raised. The difference in temperature of the reacting solution and a reference solution obtained while both are heated is plotted against time (temperature). The equa- tions enable one to calculate the heat of reaction from the area under the curve and the reaction rate constant at any particular temperature from the slope and height of the curve at the particular temperature. The heat of activation and the order of reaction are then obtained from plots of log k rs. 1/7'. Pohle, Gregory, and Taylor (66) described work on the comparison of analytical techniques for predicting the relative stability of fats and oils to oxidation. Most of these stability testing methods attempt to force the oxidation of the fat or oil, after which either oxygen absorption is meas- ured or organoleptie or chemical methods are used to monitor the degra- dation of the product. As part of a series of studies on the oxidation of emulsified and solubilized oils, Carless and Nixon (67) studied the oxida- tion of methyl linoleate. They measured oxygen uptake manometri- eally in a Warburg apparatus, in addition to the more usual checking of peroxide values. Somewhat along this line, Ravin, Kennon, and Swin- tosky (68) studied an oxidative reaction by measuring oxygen uptake in a Warburg respirometer the oxygen uptake was monitored while the system was stressed with a sun lamp. They showed that oxygen uptake was linear with time and that it stopped when the light was turned off. Briefly continuing these comments on oxidation, we note that Reese and Guth (69) irradiated ealamine lotions with UV, sunlight, and both