J. Cosmet. Sci., 65, 125–145 (May/June 2014) 125 Survey of cosmetics for arsenic, cadmium, chromium, cobalt, lead, mercury, and nickel content NANCY M. HEPP, WILLIAM R. MINDAK, JOHN W. GASPER, CHRISTOPHER B. THOMPSON, and JULIE N. BARROWS, U.S. Food and Drug Administration, College Park, MD 20740. Accepted for publication April 20, 2014. Synopsis As part of efforts to assess amounts of inorganic element contamination in cosmetics, the U.S. Food and Drug Administration contracted a private laboratory to determine the total content of seven potentially toxic or allergenic elements in 150 cosmetic products of 12 types (eye shadows, blushes, lipsticks, three types of lo- tions, mascaras, foundations, body powders, compact powders, shaving creams, and face paints). Samples were analyzed for arsenic, cadmium, chromium, cobalt, lead, and nickel by inductively coupled plasma-mass spectrometry and for mercury by cold vapor atomic fl uorescence spectrometry. The methods used to deter- mine the elements were tested for validity by using standard reference materials with matrices similar to the cosmetic types. The cosmetic products were found to contain median values of 0.21 mg/kg arsenic, 3.1 mg/kg chromium, 0.91 mg/kg cobalt, 0.85 mg/kg lead, and 2.7 mg/kg nickel. The median values for cadmium and mercury were below the limits of detection of the methods. The contract requirements, testing procedures, and fi ndings from the survey are described. INTRODUCTION Many inorganic elements and their compounds are known to be toxic or allergenic to humans (1). Cosmetics are applied directly to the human body and their use provides potential routes of exposure to elemental contaminants. Products that are partially in- gested (e.g., lipsticks) may provide direct oral exposure to the contaminants, and cosmet- ics applied to the surface of the skin (e.g., lotions) may provide indirect percutaneous exposure. The U.S. regulations specify limits for elemental contaminants in color addi- tives used in cosmetics (2), but there are no limiting specifi cations for these contaminants in cosmetics themselves, with one exception (mercury). We determined the amounts of seven elements present in cosmetics sold on the U.S. market to assess concentration levels of these contaminants. Many studies of elemental contamination of cosmetics have not focused on products marketed in the United States. Some have been limited to specifi c cosmetic types, such Address all correspondence to Nancy Hepp at Nancy.Hepp@fda.hhs.gov.

JOURNAL OF COSMETIC SCIENCE 126 as traditional eyeliners of the Middle East (3–5), lipsticks (6–10), and sunscreens (11– 13). Cosmetics have been analyzed throughout the world for various metals and arsenic (14–33) using a variety of sample preparations and analysis methods. Some of the methods may not completely account for contaminants, especially when acid solution techniques do not completely dissolve the samples before analysis by a solution tech- nique. For example, in 1997, Sainio et al. (29) surveyed eye shadows in 49 products purchased in Finland for lead, cobalt, nickel, chromium, and arsenic. The products were heated with equal parts of nitric and hydrochloric acids at atmospheric pressure until production of acid fumes ceased, and the residues were dissolved in 65% nitric acid and fi ltered when necessary. The authors analyzed the fi ltrates by electrothermal atomization atomic absorption with Zeeman background correction. In 2004, El-Shazly et al. (31) reported values for arsenic, chromium, and cobalt in 27 eye products that had been imported into Egypt. These samples were analyzed by neutron activation analysis. Verifi cation of accuracy was made by comparison with certifi ed reference ma- terials of lake and estuarine sediments. In 2009, Atz et al. (23) published results for arsenic, cadmium, cobalt, chromium, copper, nickel, lead, and mercury in 10 eye shad- ows and 9 lipsticks purchased in Brazil. Samples were completely dissolved by micro- wave heating at elevated pressures in sealed containers using mixtures of nitric and hydrofl uoric acids, followed in some cases by addition of hydrogen peroxide. The pre- cipitate-free solutions were analyzed by electrothermal atomization atomic absorption with Zeeman background correction. The authors verifi ed accuracy by measuring re- coveries of fortifi ed samples, as well as by comparing results with those obtained by Table I. ICP-MS Calibration Standards Calibration standard Final concentration (μg/kg) As Cd Cr Co Pb Ni 0 0 0 0 0 0 0 1 0.0020 0.001 0.0050 0.0050 0.0020 0.0050 2 0.025 0.002 0.010 0.010 0.0040 0.010 3 0.050 0.025 0.25 0.125 0.0625 0.3125 4 0.10 0.050 0.50 0.25 0.125 0.625 5 0.25 0.10 1.0 0.50 0.25 1.25 6 0.25 0.25 2.5 1.25 0.625 3.125 7 0.50 0.50 5.0 2.5 1.25 6.25 8 1.0 1.0 10. 5.0 2.5 12.5 Table II. Limits of Detection and Quantitation in the Analytical Portion Arsenic Cadmium Chromium Cobalt Lead Mercury Nickel mg/kg LOD 0.048 0.018 0.052 0.0066 0.0084 0.0010 0.032 LOQ 0.16 0.058 0.18 0.022 0.028 0.0032 0.10