J. Cosmet. Sci., 63, 103–117 (March/April 2012) 103 Simultaneous analysis and monitoring of 16 UV fi lters in cosmetics by high-performance liquid chromatography DOJUNG KIM, SANGSEOP KIM, SEOL-A KIM, MYOENGSIN CHOI, KYOUNG-JIN KWON, MIJEONG KIM, DONG-SUP KIM, SEUNG-HEE KIM, and BO-KYUNG CHOI, National Institute of Food and Drug Safety Evaluation, Korea Food and Drug Administration, Osong Health Technology Administration Complex, 187 Osongsaenmyeong2(i)-ro, Gangoe-myeon, Cheongwon-gun, Chungcheongbuk-do, Korea 363-951. Accepted for publication September 21, 2011. Synopsis Sixteen UV fi lters were simultaneously analyzed using the high-performance liquid chromatographic method. They were drometrizole (USAN Drometrizole), 4-methylbenzylidene camphor (USAN Enzacamene), men- thyl anthranilate (USAN Menthyl anthranilate), benzophenone-3 (USAN Oxybenzone), benzophenone-8 (USAN Dioxybenzone), butyl methoxydibenzoylmethane (USAN Avobenzone), ethylhexyl triazone (USAN Octyl triazone), octocrylene (USAN Octocrylene), ethylhexyl dimethyl p-aminobenzoic acid (USAN Padi- mate O), ethylhexyl methoxycinnamate (USAN Octinoxate), p-aminobenzoic acid (USAN Aminobenzoic acid), 2-phenylbenzimidazole-5-sulfonic acid (USAN Ensulizole), isoamyl p-methoxycinnamate (USAN Amiloxate), and recent UV fi lters such as diethylhexyl butamidotriazone (USAN Iscotrizinol), methylene bis-benzotriazolyl tetramethylbutylphenol (USAN Bisoctrizole), and terephthalylidene dicamphor sulfonic acid (USAN Ecamsule). Separation of the UV fi lters was carried out in a C18 column with a gradient of methanol-phosphate buffer, and the UV detection was at 300, 320, or 360 nm without any interference. The limits of detection were between 0.08 and 1.94 μg/ml, and the limits of quantitation were between 0.24 and 5.89 μg/ml. The extracting solvent for the UV fi lters was methanol, except for ethylhexyl triazone and methylene bis-benzotriazolyl tetramethylbutylphenol, which were prepared with tetrahydrofuran. The re- coveries from spiked samples were between 94.90% and 116.54%, depending on the matrixes used. The de- veloped method was applied to 23 sunscreens obtained from local markets, and the results were acceptable to their own criteria and to maximum authorized concentrations. Consequently, these results would provide a simple extracting method and a simultaneous determination for various UV fi lters, which can improve the quality control process as well as the environmental monitoring of sunscreens. INTRODUCTION Exposure to UV radiation can harmfully affect the public health by inducing conditions such as skin redness or skin cancer (1,2). As the ozone layers are depleted, protection from hazardous UV radiation, UVA (320 – 400 nm) and UVB (290 – 320 nm), has been a major concern for people (3,4). UVA has a long wavelength that reaches the corium of the skin to initiate skin aging, DNA damage, and immune alteration. UVB is known to

JOURNAL OF COSMETIC SCIENCE 104 cause sunburn and to induce malignant melanoma (2). UV fi lters have been used as an ingredient in sunscreens to protect the skin from UVA and UVB. Most organic UV fi lters have effective protection against UVB but little effect against UVA, whereas butyl methoxydibenzoylmethane (USAN Avobenzone), terephthalylidene dicamphor sulfonic acid (USAN Ecamsule), and methylene bis-benzotriazolyl tetramethylbutylphenol (USAN Bisoctrizole) are utilized to protect UVA. Inorganic UV fi lters, such as TiO2 (USAN tita- nium dioxide) and ZnO (USAN zinc oxide) were combined in sunscreens to protect UVA. To improve protective effects from UV radiation, many cosmetic companies have tried to synthesize new UV fi lters or to develop certain formulation containing more than one UV fi lter (5–7). Although UV fi lters protect skin from UV radiation and help prevent skin damage including skin cancer, they are also concerned about the possibility of causing an allergic reaction to human skin (8). Studies reported that UV fi lters such as benzophenone-3 (Oxybenzone) and 4-methylbenxylidene camphor (USAN Enzacamene) exert the effects of endocrine disruptors (9,10). To prevent adverse effects of UV fi lters, many countries have regulated the amount of UV fi lters that will be used in sunscreens the maximum autho- rized concentration is 3–10% for organic UV fi lters (Table I) and 25% for inorganic UV fi lters. An easy and accurate analytical method for quantitation of UV fi lters is required to assure the label claims and maximum authorized concentrations in sunscreens. In some studies, UV fi lters were investigated using GC-MS or LC-MS, due to their trace levels in environmental or biological samples (11,12). However, in cosmetics, they were generally measured using a HPLC system connected to a UV detector. Although several studies de- scribed analytical methods of 11 or 16 organic UV fi lters using HPLC connected to a diode array detector, they employed different extraction solvents and several steps of extraction according to the cosmetic formulations (13–16). Sunscreens include UV fi lters as well as other ingredients such as emulsifi ers, solubility enhancers, and stability enhancers that are necessary for a cosmetic formulation and that might interfere with reliable analysis. Various chemical properties and physical properties of UV fi lters also make it diffi cult for them to be analyzed in sunscreens (17,18). Regarding various formulations of sunscreen such as cream, lotion, or powder, practical methods with a simple preparation are still needed for industry to develop a new product and for administrative bodies to monitor the quality for marketed sunscreens. The purpose of the present study was to develop a simple extraction method and a simultaneous analysis for 16 UV fi lters, including recent UV fi lters such as di- ethylhexyl butamidotriazone (USAN Iscotrizinol), methylene bis-benzotriazolyl tetramethyl- butylphenol (USAN Bisoctrizole), and terephthalylidene dicamphor sulfonic acid (USAN Ecamsule) (Figure 1). They were determined using HPLC connected to a C18 column and diode array detector (DAD) at 300, 320, and 360 nm. Method validation was achieved through specifi city, linearity, and recovery. Several solvents were examined for extracting UV fi lters from sunscreens, regardless of the matrixes used. The validated method was ap- plied to various types of sunscreens obtained from local markets and evaluated with regard to the label claims and compliance to their maximum authorized concentrations. EXPERIMENTAL MATERIALS Drometrizole (2440-22-4), menthyl anthranilate (134-09-8), benzophenone-3 (205-031-5), benzophenone-8 (131-53-3), octocrylene (6197-30-4), ethylhexyl dimethyl p-aminobenzoic