DETERMINATION OF SUNSCREEN AGENTS BY UPLC AND HPLC 473 Table II HPLC and UPLC Operation Conditions for Determination of Six Sunscreen Agents in Suncare Products System Column Eluent Gradient method Flow rate Temperature Sample loop Column of injection Detector wavelength VALIDATION STUDIES HPLC (Waters Alliance) Waters symmetry C 18column ( 4.6 mm x 75 mm, 3.5 µm) Solvent A (water) Solvent B (ethanol) Time % A % B 25 75 2.0 25 75 3.5 10' 90 5.5 00 100 10.5 00 100 11.0 25 75 1.0 ml/min Ambient Partial loop 10 µl 310 nm UPLC (Waters Acquity) Waters UPLC BHS C18 column (2.1 mm x 50 mm, 1.7 µm). Solvent A (water) Solvent B (methanol) Time % A % B 15 85 1.0 15 85 2.0 20 80 2.5 20 80 3.0 00 100 8.5 00 100 9.0 15 85 0.4 ml/min Ambient Partial loop with needle overfill 2 µl 310nm Specificity could be tested as the difference between the test sample and the placebo sample. The difference between the placebo sample and the test sample of the specificity test is that the test sample has the additional ingredients of the six sunscreen agents. The linearity of the method was evaluated by the sum of squares errors (R 2 ) of the concen­ trations of sunscreen agents in the range 5-200 µg/ml. To check the repeatability of the method, multiple injections of a sample solution were measured. The relative standard deviations (RSDs) of the retention time and the peak area were used as the repeatability data. The intermediate precision could be estimated by the RSDs of the amount of sunscreen agents in the suncare product. The test values of assay were measured two times. Each measurement was independently performed. When the experiments were carried out, the date, the analyst, and the instruments were different from the ones used before. The system's stability was tested with the RSDs of the retention time and the peak area of the standard solution within the following time frames: the time in the initial stage, 24 h, and 48 h. The recovery value was the difference between the observed concentration and the theoretical concentration of the test sample. The theoretical concentration could be calculated by the quantity of the added amount of the six sunscreen agents in the suncare products. RESULTS AND DISCUSSION CHROMATOGRAPHIC SEPARATION The objective of the study was to determine the optimum condition for the analysis of six major sunscreen agents in suncare product using HPLC and UPLC. The chemical

474 JOURNAL OF COSMETIC SCIENCE behaviors of these six sunscreen agents were different from each other. As TM and TS were certified as UV absorbers by the regulatory authorities of the European Union (15), an analytical procedure for the determination of the compounds in suncare products became necessary. A simultaneous analysis of six sunscreen agents that include TM and TS has not been reported yet. As TM and TS are extremely hydrophobic compounds, the analytical method of TM and TS is also different from the method of OMC, IMC, OS, and OT. In the case of the non-aqueous reversed-phase condition, TM and TS could be analyzed with the most favorable test results of peak shape, retention type, and peak tailing. In the non-aqueous reversed-phase condition, OMC, IMC, OS, and OT will be weakly retained and eluted in the void volume region. The elution in this region should be avoided because of poor resolution in the region caused by interference from unretained matrix constituents. The columns of different lengths and different column manufac­ turers were tested under the non-aqueous reversed-phase condition. The separation of all component peaks could not be exactly obtained in this case. In the case of the column with less hydrophobic sorbent, the retention and the separation were satisfactory for the six sunscreen agents. The principal weak points of the column test with less hydrophobic sorbent were long-time analysis, peak broadness, and peak tailing. From the test, we can conclude that it is difficult to improve test results by variation of the column. The optimum conditions of the gradient and the organic-aqueous solvent mixture were chosen for the HPLC and UPLC methods, respectively. In these conditions, it was possible to determine OMC, IMC, OS, and OT in the aqueous reversed phase. The analysis of TM and TS, however, was possible in the non-aqueous reversed phase. It took less than 10 min to analyze the six sunscreen agents with satisfactory retention values and resolutions. We have validated this method with the suncare products of the Amore-Pacific Cosmetic Company. The method proposed in this paper using HPLC and UPLC enables the fast and efficient determination of the six sunscreen agents in a single chromatographic run. It is suitable for the quality control of sunscreen agents in cosmetic products. OPTIMIZATION CONDITION OF HPLC (HIGH-PERFORMANCE LIQUID CHROMATOGRAPHY) FOR THE ANALYSIS OF THE SIX SUNSCREEN AGENTS Several combinations of eluent mixtures were tested in order to achieve the simultaneous determination of TM and TS with OMC, IMC, OS, and OT. Methanol as an organic solvent was found out to be improper, as it needed too long a retention time for TM and TS. Also, in the methanol test, the peaks of TM and TS were broad and tailed, and the repeatability of the test was poor. Second, acetonitrile was inadequate as an organic solvent, as the retention time and the shape of peak were not suitable for the test. The test of ethanol made an acceptable result. This was due to the short retention time, high resolution, and good repeatability that were produced in the test (Figure 2). Therefore, ethanol was proven to be superior as an organic solvent in the determination. For the analysis of OMC, IMC, OS, and OT, the peak shape, retention time, and resolution of the test with 0.1 % phosphoric acid and 20 mM ammonium buffer were compared with the result of the test with water as an aqueous solvent. There was no significant difference between these three aqueous solvents and so the ethanol-water