DETERMINATION OF O-PHENYLPHENOL IN SKIN LOTION 133 NBD-CO-IS derivatives were 16.2 and 22.2 min, respectively. The running time was 25 min. METHOD VALIDATION Linearity. A standard curve was constructed by plotting integrated peak area ratio versus con- centration of OPP. The plot was linear (y = 6.297×–0.0217) in the range of 0.01–0.2 μg/ml with an r2 value of 0.9960. Sensitivity. The values of the lower limits of quantifi cation and detection were 10 ng/ml (absolute amount of 40 pg/20 μl injection, signal-to-noise ratio of 10:1) and 3 ng/ml (absolute amount of 12 pg/20 μl injection, signal-to-noise ratio of 3:1), respectively. As shown in Table I, the sensitivity of our method (3 ng/ml, absolute amount of 12 pg) was no less than those of various previously reported methods (range of 3.4–350 pg as absolute amount) (6,7,9,10,19–21). PRECISION AND ACCURACY Precision and accuracy for intra-day and inter-day assays of OPP are shown in Table II. In the intra-day assay, the range of standard deviation was within 4.5–6.4% of the mean, and recoveries were within the range of 91.4–96.0%. In the inter-day assay, the range of standard deviation was within 5.3–8.8% of the mean, and recoveries were within the range of 89.4–94.5%. Figure 3. Typical chromatograms of blank spiked with IS (A) and standard sample spiked with IS (B) after derivatization with NBD-COCl. Samples were reacted with NBD-COCl for 2 min at pH 8.5 at room temperature. Compounds: OPP, 0.4 ng/20 μl injection volume IS, 0.4 ng/20 μl injection volume. Retention times: 16.2 min, NBD-CO-OPP (1) 22.2 min, NBD-CO-IS (2). Chromatogram (A) shows no peak of NBD-CO-OPP (1).

JOURNAL OF COSMETIC SCIENCE 134 INTERFERENCE Cosmetics generally contain parabens as a preservative, because of their antimicrobial activi- ties, relatively low toxicity to human, nonvolatility and other properties (22–24). Therefore, interference with the detection of NBD-CO-OPP or NBD-CO-IS derivatives by seven para- bens (methyl 4-hydroxybenzoate, ethyl 4-hydroxybenzoate, propyl 4-hydroxybenzoate, iso- propyl 4-hydroxybenzoate, butyl 4-hydroxybenzoate, isobutyl 4-hydroxybenzoate, benzyl 4-hydroxybenzoate, each 1 μg/ml in 1% acetonitrile) was investigated. Derivatization was performed as described in Procedures. The relative retention times of seven NBD-CO-paraben derivatives were within 0.34–1.00 (Table III). Only NBD-CO-butyl 4-hydroxybenzoate de- rivative overlapped with NBD-CO-IS derivative in the chromatogram. Therefore, the present assay is not appropriate for analysis of cosmetics containing butyl 4-hydroxybenzoate. For cosmetics that include butyl 4-hydroxybenzoate as an ingredient, it would be necessary to se- lect a different IS or to conduct the assay without IS. Further studies are needed on this point. ANALYSIS OF OPP IN THREE SKIN LOTIONS We selected three skin lotions (A, B, and C) that were all labeled on the bottles as paraben-free. In preliminary tests, no peak that would overlap with NBD-CO-IS was observed in Table I Sensitivity of various methods for determination of OPP Method Limit of detection Reference Concentration (ng/ml) Absolute amount (pg) μHPLC with electrochemical detection 0.68 3.4 (7) HPLC-UV 17.6 350 (6) HPLC-UV 24 300 (9) LC-MS 10 200 (10) LC-MS/MS 0.1 Not described (20) GC-MS 9 Not described (21) HPLC-fl uorescence detection 10 100 (19) HPLC-fl uorescence detection 3 12 This paper Table II Intra- and inter-day assay reproducibility for determination of OPP Concentration (μg/ml) Measured (μg/ml, Mean ± S.D., n=5) C.V. (%) Recovery(%) Intra-day OPP 0.01 0.00914 ± 0.00062 6.8 91.4 0.05 0.0462 ± 0.0026 5.6 92.4 0.2 0.192 ± 0.008 4.2 96.0 Inter-day OPP 0.01 0.00894 ± 0.00079 8.8 89.4 0.05 0.047 ± 0.0032 6.8 94.2 0.2 0.189 ± 0.010 5.3 94.5