FIXED AND BULK EMULSION WATER 49 EXPERIMENTAL In order to manufacture the O/W cream, the lipophilic components were melted at 70øC using a water bath. Water heated to the same temperature was added (with stirring) and the creams were allowed to cool with continuous stirring. Evaporated water was not replaced in order not to shift the ratio of the interlamellarly fixed water fraction and bulk water fraction adjusted during manufacturing of the creams. THERMOGRAVIMETRIC ANALYSIS (TG) TGA-system TA 500 S (Heraeus, Hanau, FRG) Heating rate: 2øK/min weight of specimens: 2-5 mg temperature range: 20-105øC surface of the Pt-sample holder: 66.5 ram2 purge rate of the thermobalance with dried nitrogen: 25 ml/min. The thermogravimetric results are influenced by the geometry of the sample holder. Preferably a shallow sample holder should be used. In order to get reproducible results, a thin layer with homogeneous thickness is a prerequisite. RESULTS AND DISCUSSION A prior condition for differentiation of interlamellarly fixed water and bulk water is that there exists a sufficiently high difference in free energy and/or structural hindrance for these two types of water to evaporate. Differential thermal analysis (DTA) of the water-containing hydrophilic ointment DAB 8 shows higher melting points for the hydrophilic gel phases than for the water-free systems (2). The same is true for stearate creams (3,4). For monoglycerol systems (5) and O/W creams with non-ionic surfactants (6,7) such a rise of the melting temperature of the hydrophilic gel phase could not be found. It appeared that the most effective separation of the different water types is achieved at melting temperatures of the hydrophilic gel phases between 60øC and 72øC. Through the use of an X-ray low angle technique a dynamic equilibrium between interlamellarly fixed water and bulk water was previously postulated (2). Thus, if bulk water evaporates, according to the equilibrium rate constant for interlamellarly fixed water and bulk water an equivalent amount of interlamellarly fixed water is simulta- neously transformed into bulk water. The thermogravimetric results demonstrate that a heating rate of 2øK/rain gives the best separation effects with the O/W creams investigated. A thermogravimetric curve (TG-curve) and its derivative curve (DTG-curve) typical for these O/W creams are shown in Figure 4. The following formulation was used: Polyoxyethylen-400-stearate 5 parts Cetyl alcohol 5 parts Stearyl alcohol 5 parts Liquid paraffin 7.5 parts White petrolatum 17.5 parts Water 60 parts The TG-curve represents the weight loss of the sample as a function of temperature. The scale for weight loss is on the left side of Figure 4. The DTG-curve is the first derivative of the thermogravimetric curve with respect to temperature. The DTG-curve

50 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS 3.0 WEIGHT [m• ........... WEIGHT LOSS BULK WATER TOTAL WATER LOSS OF THE CREAM WEIGHT LOSS OF INTERLAMELLARLY FIXED WATER AFTER MELTING OF THE HYDROPHILI½ GEL PHASE INCREASE 1.5 1,0 I I I 0.5 , i 2% 3% 40 5•0 6 • 0 7•0 TEMPERATURE •C] Figure 4. Example of a typical TG-curve (I) and DTG-curve (II) for further information see text. •.g/s] -ODO II d•___• dt 0.04 DECR908mO EASE LO, represents the rate of evaporation of incorporated water. The scale for evaporation rate is on the right side of Figure 4. The thermogravimetric measurements are started at 20øC. Following the DTG-curve (II) in Figure 4, the evaporation rate of the bulk water phase increases until 36øC is reached. Then the evaporation rate decreases. At a temperature of 43øC the TG-curve (I) shows a point of inflection corresponding with a minimum in the DTG-curve. Then the evaporation rate according to the DTG-curve increases very strongly until at 48øC a maximum of evaporation is reached. After this the evaporation rate decreases to zero. The strong increase of the evaporation rate is due to the melting of the hydrophilic gel phase at 43øC. The melt of the hydrophilic gel phase at 43øC can be controlled either by hot stage microscopy or by DSC. The ratio of interlamellarly fixed water and bulk water can be calculated by using the areas under curve (AUC) of the DTG-curve (hatched areas) or by weight loss using the point of inflection of the TG-curve (Figure 4). This calculation, however, is not precise due to the above-mentioned dynamic equilibrium between interlamellarly fixed water and bulk water. A quantitative differentiation is possible by fitting the first parts of the DTG-curve of the cream with the measured DTG-curve of pure water and then dividing the DTG-curve of the cream into two completely separated curves (dotted lines a and b of the DTG-curve, Figure 4). The ratio of interlamellarly fixed water and bulk water can now be calculated from the peak areas A and B (Figure 4). Figure 5 shows the TG- and DTG-curves of an emulsion used in cosmetics (Skin Life Emulsion ©, Helena Rubinstein, FRG) with a weakly pseudo-plastic flow behavior. The TG-evaporation pattern of this liquid emulsion corresponds with that of pure water. Figure 6 represents the TG- and DTG-curves of water-containing hydrophilic ointment DAB 8 with a total water amount of 70 percent. Mainly bulk water evaporates up to 56øC. At 56øC the cetostearyl alcohol monohydrate melts. In the temperature range of 56øC to 72øC the evaporation of bulk water is superceded by the evaporation of the