18 I3-GLYCYRRHETINIC ACID 81 I I I I I I 0 /4 8 12 16 20 {min) Figure 8. HPLC separation of the cream components extracted with 1% ammonium hydroxide solution. Chromarographic conditions as in Figure 7. 1 = methyl paraben 2 = ethyl paraben 3 = propyl paraben. the chloroform solution was evaporated to dryness. The residue, containing just GT and the p-hydroxybenzoates, was dissolved in methanolic potassium hydroxide, and the resulting solution was used for the second derivative determination.

82 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS Experiments carried out with hydrogels spiked with a known amount of GTP provided essentially quantitative recoveries by both the spectrophotometric and HPLC proce- dures. On the other hand, when cream 3 was analyzed, the recovery obtained by the spectrophotometric procedure was about 90%, owing to a minor percentage of the analyte remaining in the chloroformic solution. This recovery value, however, can be considered adequate to the purpose. Concerning samples 4 and 5, both the first and second derivative methods were applied, but the second derivative method proved to be more suitable because the blank presents a zero over a wider range of wavelengths. To obtain information about the determination limit for GT in the examined formu- lations by the described spectrophotometric method, laboratory-made cosmetic prepa- rations containing GT over the concentration range 0.015-1.5 % were analyzed. For the analysis of the formulations examined in this work, the results show the spectrophoto- metric method to be suitable with content of GT •0.3% for lower concentrations, the interference from the excipients made the method inaccurate. However, it is likely that for different formulations with lower excipient-to-GT ratios, a higher sensitivity for GT can be achieved. CONCLUSION Derivative UV spectrophotometry and liquid chromatography (HPLC) proved to be effective techniques suitable for the determination of GT and GTP in cosmetic formu- lations. The described derivative UV method showed good reproducibility and accuracy for GT contents •0.3% in the examined formulations. This technique can be consid- ered a useful alternative approach when HPLC instrumentation is not available, its application field being dependent on the formula of the GT preparation. HPLC was confirmed to be a versatile analytical tool useful as a reference method and able to selectively determine small amounts of GT in complex cosmetic matrices. ACKNOWLEDGMENTS We would like to thank Piana Cosmetici SPA (Bologna, Italy) for having kindly supplied some cosmetic samples for the purpose of this analysis. REFERENCES (1) G. Proserpio and S. Ferraris, Phytosomes: Cosmetic application, Cosmet. Toiletr. (Ital. Ed.), 9, 6-14 (1988). (2) CTFA Cosmetic Ingredient Dictionary, 3rd ed. (The Cosmetic, Toiletry and Fragrance Association, Inc., 1982). (3) Martindale, The Extra Pharmacopoeia, 29th ed. (The Pharmaceutical Press, London, 1989). (4) K. Norifumi, H. Itaru, S. Yoshihide, M. Yasuhiro, K. Kenzo, K. Hiroko, M. Seji, M. Takeyuki, and Y. Sukeo, Effect of glygyrrhizin and glycyrrhetinic acid on arachidonic acid metabolism in peritoneal exudate cells of rat, Ensho, 9(1), 29-32 (1989) Chem. Abst., 112, 36 (1990). (5) D. Xingying, W. Qihao, P. Yaqun, and L. Rongchun, Colorimetric determination of glycyrrhizic acid in licorice and its preparations with methylene blue, Yiyao Gongye 18(10), 451-455 (1987) Chem. Abst., 108, 408 (1988).