JOURNAL OF COSMETIC SCIENCE 240 trace metals content of facial cosmetics commonly used in Nigeria were examined. Re- sults indicated that the concentration of Pb in the candy lipstick product is 2.8 times higher than the limit value regulated by American Food and Drug Administration (FDA) (5). In another study, it was found that more than half of the 33 tested lipsticks from top local brands contained detectable Pb (6). In a similar study, Al-Saleh et al. assessed the levels of Pb, Cd, Ni, chromium (Cr), mercury (Hg), arsenic (As), and antimony (Sb) in 28 samples of 14 lipstick brands in Saudi Arabia (7). While this study found that the levels of all metals were lower than the FDA limits, As level (3 μg/g) was higher than the limit value in one brand of lipstick. Also, the levels of Ni in 10 and Cr in 4 of the 28 lipsticks were found to be above the trigger limit of 1 μg/g for allergic contact dermatitis. Ayen- imo et al. demonstrated that personal care products, including soaps, cosmetic, and hair creams, are notable sources of Cd and Cr in the human body (8). In another study, Sin et al. revealed high levels of Hg in cosmetic creams (9). Although some information have been reported on the metal content of different cos- metic products, few data exist on the hair care products in literature. There is also a lim- ited amount of information on this fi eld in Turkey. This study examined the levels of Cd, Pb, and Ni in commonly used hair care products commercially sold in Turkey. Data gen- erated through this study will procreate awareness about risks associated with the use of these products and also provide baseline information for further studies. MATERIALS AND METHODS All chemicals and metal stock standard solutions were obtained from Sigma-Aldrich (St. Louis, MO) and Merck (Darmstadt, Germany). One hundred and fi ve samples from 51 different brands commonly sold in Turkey were randomly purchased from different markets in Istanbul and classifi ed as shampoos, hair conditioners, hairstyling agents (hair sprays, setting lotions and setting mousses, hair creams and hair gels), and pediatric bath cleaning products. Samples (0.5 g) were fi rst transferred to Tefl on acid digestion cups. Three milliliters of 65% nitric acid (HNO3) and 1 ml of 44% hydrofl uoric acid were then added to samples and digested in a microwave acid digestion unit (Milestone Mega 1200, Bergamo, Italy) according to the manufacturer’s application notes. The digested samples were then quan- titatively transferred and diluted to 5.0 ml with triply distilled water (Simplicity Water Purifi cation System, Millipore, Molsheim, France). Cd, Pb, and Ni determination in digested samples were done by using graphite furnace atomic absorption spectrometry (AAS) with Zeeman background correction (Analytic Jena, Zeenit 700, Jena, Germany). The calibration curve was plotted with standard solutions of Cd (0.2–1 μg/l), Pb (6–34 μg/l), and Ni (6–30 μg/l). All standards were prepared in 0.05% HNO3. Suitable dilutions were prepared from digested samples with triply dis- tilled water. Peak Signal area mode was used for quantifi cations and the manufacturer’s application notes were used for metal measurements checking validation parameters. Table I shows AAS furnace parameters. All of the results are expressed as the mean ± S.D. The differences between the groups were evaluated with Kruskal–Wallis analysis of variance and comparisons between two

CADMIUM, LEAD, AND NICKEL IN HAIR CARE PRODUCTS IN TURKEY 241 independent groups were made with the Mann–Whitney U-test. p 0.05 was considered statistically signifi cant. RESULTS AND DISCUSSION The mean Cd, Pb, and Ni levels of our samples were found to be 6.09 ± 4.26 ng/ml, 0.20 ± 0.09 μg/ml, and 1.08 ± 2.78 μg/ml, respectively (Tables II—IV). The samples with the concentrations below the limit of detection (LOD) were excluded from mean, median, and range calculations. LOD of Cd, Pb, and Ni were 0.03, 2.25, and 1.40 ng/ml, respectively. Cd was detected in 40% of the total samples. The results with detectable Cd levels varied from 1.01 to 25.08 (ng/g). Maximum levels of Cd were detected in two shampoo samples (25.08 and 17.14 ng/g). Both of them were herbal-based formulations. The order of con- tent of Cd in the hair care products was determined as shampoos hairstyling agents pediatric products hair conditioners. No signifi cant differences were found between different classes of hair care products. Pb was determined in 23 of the 105 total samples (21.91%). The highest Pb level (0.42 μg/g) was observed in hair conditioner followed by hairstyling samples (0.37 μg/g), while the lowest Pb value was detected in shampoo samples (0.07 μg/g). The highest values were found in coconut and wheat protein-based formulations. The mean amount of Pb in Table I Atomic Absorption Spectrometer and Graphite Furnace Parameters for Metal Analysis Element Wavelength (nm) Maximum ash temperature (°C) Atomization temperature (°C) Modifi er Zeeman modea Cd 228.8 300 950 - 2-fi eldb Ni 232 1100 2350 - 2-fi eld Pb 283.3 500 1650 5 μl Pd (NO3)2 (0.1 %) 3-fi eld dynamicc a The measurement mode for the background correction. b The measurement with magnetic fi eld strength 0 and maximum fi eld strength. c The measurement of the absorbance values at various magnetic fi eld strengths. Table II Cd Content (ng/g) in Hair Care Products Classes of analyzed hair care product Tested samples (n) Mean S.D. C.V. (%) Range Median Samples with detectable Cd levels n % Shampoos 30 6.89 5.73 83.16 1.01–25.08 6.51 20 66.67 Hair conditioners 39 4.05 2.62 64.69 1.26–9.23 3.53 9 23.08 Hairstyling agents 21 6.38 0.79 12.38 5.23–7.72 6.24 7 33.33 Pediatric 15 6.17 1.47 23.82 4.47–8.70 6.00 6 40 Total 105 6.09 4.26 69.95 1.01–25.08 6.05 42 40