LIQUID CHROMATOGRAPHY OF BACTERIOSTATS 367 RI ISOPROPANOL SOLVENT IRGASAN TCC 0.50•g--• DP $00 0.25 0.004 ABS UNITS H•HLOROPH2.5/-/..9 TBS 2.5/../. g uv TIME IN MINUTES Figure 4. Gradient elution from 4.0% to 12.0% isopropanol in heptane over 6.5 min, non- linear. Refractive index detector used to monitor gradient in tandem with uv photometer at 254 nm. Column: Corasil II, 1 m x 3.0 mm, at room temperature pressure: 150 psi injection volume: 5/xl Gradient elution (Fig. 4) made it possible to separate all four classes: Irgasan DP 300, carbanilide, hexachlorophene, and salicylanilide within 15 min, obtaining sharp and well-separated peaks. The mobile phase composi- tion at the exit of the column is shown by the refractive index detector re- sponse. The gradient was begun immediately, and it took 4 min for the solvent change to reach the column exit. The final composition was reached after 11 min, and this composition was held till the end of the analysis. The use of gradient to solve this separation problem is elegant but is ex- pensive in terms of equipment. It was found possible (Fig. 5) to achieve a similar result by the more simple technique of switching the mobile phase composition from 4.4% to 12.0% isopropanol in heptane, which is easy to arrange with the small displacement piston pumps used on many inexpensive chromatographs. Therefore, in practical terms, this chromatogram is the most significant, since it gives separation of all four classes without requiring a gradient elution device.

368 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS TCC 0.50/•g-• IRGASAN DP$00 0.25•cj--• t 0.004 ABS UNITS UV I RI OF CONC. AT DETECTOR /-- HEXACHLOROPHE2,5 / TBS 2.5/.c , ,½_ TIME IN MINUTES Figure 5. Step elution. 4.4% initial and 12.0% final concentration of isopropanol in hep- tane. Column: Corasil II, I m x 3.0 mm, at room temperature pressure: 150 psi iniection volume: Finally, this technique was tested on an extract from two soaps. One soap contained two bacteriostats. The other soap was identical, except that it con- tained no bacteriostat. One gram of grated soap was extracted with acetone, and the extract was filtered and evaporated to dryness. The residue was dried under vacuum at 60øC and taken up in 1.0 ml methylene chloride. Using the conditions of Fig. 4, the presence of both Irgasan DP 300 and of TBS was shown, but the former was only partially resolved from inter- ferences extracted from the soap. Figure 6 shows the chromatograms ob- tained by changing the chromatographic condition so that the interfering substances were completely resolved from the Irgasan DP 300 peak. Thus the feasibility of qualitative analysis of the bacteriostats in soap has been established. The quantitative aspects are now being examined. SUMMARY High-pressure liquid chromatography has been applied in providing a fast, convenient method for separating and identifying bacteriostats in cosmetic