188 JOURNAL OF COSMETIC SCIENCE Table I Retention Times, Regression Equation, and Correlation Coefficients (r) of Triclocarban (TCC) and Triclosan Chromatographed on Nucleosil C 18 With Various Mobile Phases Using Differential Refractometer Retention time (min) Regression equation (0.1 %-0.8%) Flow rate Mobile phases (v/v) (ml/min) TCC Triclosan TCC Triclosan Methanol:water (80:20, 1.0 10.23 10.53 y = -0.47 + 77 .5x y = -1.1 + 76.lx 0.5 M NaClO4) (r = 0.9999) (r = 0.9999) Methanol:water (80:20, 1.0 8.48 9.10 y = -0.31 + 23.9x y = -1.1 + 18.3x acetate buffer pH 4.95) (r = 0.9990) (r = 0.9900) Acetonitrile:water (50:50, 1.0 7.20 17.34 y = 91 + 819x y = -1.6 + 27.2x phosphate, pH 3.05) (r = 0.9999) (r = 0.9999) Acetonitrile:water (60:40, 1.0 5.37 10.09 y = 74 + 768x y = 0.12 + 46.5x phosphate, pH 3.05) (r = 0.9958) (r = 0.9960) Acetonitrile:water (70:30, 1.0 4.49 9.21 y = -28 + 1076x y = 0.47 + 48.3x phosphate, pH 3.05) (r = 0.9900) (r = 0.9999) Acetonitrile:water (70:30, 1.5 2.99 5.81 y = -37 + 950x y = -2.1 + 42.5x phosphate, pH 3.05) (r = 0.9995) (r = 0.9990) Acetonitrile:water (70:30, 2.0 2.23 4.50 y = 48 + 832x y = -1.1 + 40 .4x phosphate, pH 3.05) (r = 0.9950) (r = 0.9999) eluded that 1.5 ml min- 1 is a suitable flow rate for separation, shown in Figure 3. The retention times were 2.99 min and 5.81 min for TCC and triclosan, respectively. The optimum composition of the mobile phase was acetonitrile and water (70:30, v/v, pH 3.05) containing 0.05 M potassium dihydrogen phosphate. PRECISION AND RECOVERY The calibration graphs of LC-RI obtained by plotting the peak height against the concentration of TCC and triclosan show good linearity over the range 0.1 - 0.8% and 0.1 - 1.2% for TCC and triclosan, the regression equations being y = -37 + 950x (correlation coefficient r = 0.9995) and y = -21 + 42.5x (r = 0.9990), respectively. Detection limits (three times the standard deviation) of TCC and triclosan on LC-RI were 34 mg 1- 1 and 23 mg 1- 1 , respectively. Recovery tests were carried out on de- odorants and dentifrices for evaluation of the reproducibility and accuracy of the pro- posed method. Four commercial products were spiked with the amounts of the agents reported in Table II and subjected to the full extraction procedure. As seen, excellent recoveries and precision were observed. (recoveries ranging from 99 ± 0.5% to 105 ± 5.6%). APPLICATION TO COSMETIC AND PHARMACEUTICAL PRODUCTS The proposed LC-RI method was applied to the determination of triclosan and TCC in fourteen samples of commercial toothpastes, deodorants, anti-acne washing cleansers, and dentifrices. Identification of triclosan and TCC was achieved after comparison with the relative retention times of standard dilutions of the respective bacteriostats in the acetonitrile:water (70:30) eluent. A representative LC-RI chromatogram of a commercial

CHLORINATED BACTERIOSTATS z -..., a) b) 0 2 4 6 8 IO 12 Time (min) 189 Figure 3. Separation of mixture of (a) triclocarbon and (b) triclosan. Stationary phase: Nucleosil C 18 (4.6 x 250 mm). Mobile phase: acetonitrile/water (70:30, v/v). Flow rate 1.5 ml/min, injection 200 µl. RI temperature: 39 ° C. Table II Recoveries of Triclosan and TCC From Spiked Samplesa Triclosan TCC Amount added Found Recovery Amount added Found Recovery Samples (weight %) (weight %) (%) (weight %) (weight %) (%) Toothpaste 0.024 0.024 101 (3.4%i' Deodorant stick 0.120 0.121 101 (5.6%) Anti-acne washing cleanser 0.120 0.126 105 (0.5%) Antibacterial body wash 1.00 X 10- 3 0.988 X 10- 3 99 (3.6%) a Number of determinations (n = 3). 6 Relative standard deviation. foam shower bath liquid is shown in Figure 4. Solid electrodes include various types of carbon and different noble metals. Glassy carbon electrode (GCE) is the most widely used carbon electrode material. It is inert to common organic solvents and has a wide potential range (from - 1.0 to + 1.3 V). Examination of the triclosan structure shows that