44 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS RESULTS AND DISCUSSION Under the conditions used for denitrosation, nitrite represented a positive interference. For samples of unknown purity, the aqueous suspension must be analyzed for nitrite before the analysis of N-nitroso content is initiated. As a result the methodology presented was evaluated for nitrite in cosmetics. To evaluate the applicability of the previously described method to the determination of nitrite in cosmetics, the recovery of nitrite added to cosmetics was examined. To two grams of sample, nitrite was added in the range of 0.010 to 0.100/xg NO2--N/gm cosmetic before suspension of the sample in water. Recoveries of nitrite, presented in Table I, were greater than 80% for levels of 0.010/xg/gm NO 2-- N and 0.024/xg/gm NO 2-- N and greater than 95% for NO 2-- N concentrations greater than 0.049/xg/gm NO 2 -- N. Table I Recovery of Nitrite Added to Cosmetics NO2-N Added NO2-N Recovered Recovery Cosmetic Sample Sample Type ttg ttg (% CV^D) % (% CV^D) A cream 0.021 0.018 + 0.001(6) 86 _+ 4(6) B cream O.049 O.040 + O.002(5) 83 + 3(4) C lotion 0.098 0.095 _+ 0.003(3) 98 + 2(2) D lotion 0.182 0.180 + 0.019(10) 100 + 10(10) Once the nitrite level was known, the necessity of further clean-up was determined. Ion exchange was required to remove nitrite when the ion was present in excess of 1.0 ppm NO2--N in solution. The effect of this clean-up procedure on the recovery of NDELA from the ion exchange columns is presented in Table II. N-nitrosodiethanolamine was not retained by the ion exchange resin and 96 + 7 percent was recovered. Table II Recovery of NDELA Applied to Ion Exchange Column NDELA Applied NDELA Recovered Recovery ,u,g NNO-N* ,u,g NNO-N (% CV^D) % (% CV^D) 6.95 6.34 + 0.15(2) 91 + 2(2) 1.51 1.45 _ 0.01(1) 96 _ 1(1) 0.666 0.709 _ 0.003(0) 106 _+ frO) 0.110 0.098 __ 0.012(12) 90 -+ 10(12) *weight of nitroso nitrogen When the sample is mixed with water to extract NDELA many other water soluble components are also extracted. The presence of these materials was considered to be either a potential interference in the denitrosation reaction or would affect the degassing efficiency of NO from the reaction mixture. Standard addition to the aqueous solution after centrifugation was used to examine any effect. The results presented in Table III indicate that no effect occurred. Recovery studies were run using samples of different types which were analytically free of nitrosamine. Samples were spiked in the range of 0.07 to 1.10 /xg/gm NDELA.

NDELA DETERMINATION 45 Table III Effect of Sample Matrix on Denitrosation Reaction NDELA Added Signal Recovery Sample Type /•g Counts (% CV) % (% CV) 0.036 2294 + 74(3) lotion 0.036 2172 +_ 2(0) 95 + 0(0) cream 0.036 2398 + 77(3) 104 + 3(3) cream 0.036 2372 +_ 153(6) 103 +_ 7(7) 0.109 9358 +_ 892(10) lotion 0.109 9340 _ 120(1) 100 _ 1(1) cream 0.109 9374 + 86(1) 100 + 1(1) cream 0.109 10300 + 200(2) 110 + 2(2) Recoveries of NDELA, presented in Table IV, were greater than or equal to 82% for the spiking levels examined. The mean recovery for all spiking levels was 89 + 6%. Table IV Recovery of N-nitrosodiethanolamine (NDELA) Added to Cosmetics NDELA Added NDELA Recovered Recovery Cosmetic Sample Sample Type /•g /•g (% CV^•)) % (% CV^•)) A cream 1.46 1.29 + 0.01(1) 88 +_ 1(1) B cream 0.146 0.120 +_ 0.007(6) 82 +_ 5(6) C lotion 0.730 0.642 + 0.007(1) 88 + 1(1) D lotion 0.511 0.427 _+ 0.008(2) 84 +_ 2(2) E lotion 0.365 0.332 + 0.011(3) 91 + 3(3) F lotion 2.19 2.17 + 0.00(0) 99 + 0(0) Results of the analysis of a variety of cosmetics for nitrite and N-nitroso content are presented in Table V. Data is reported as the weight of nitrite nitrogen or nitroso-nitrogen per gram of cosmetic sample. Nitrite levels ranged from 0.012 to 13.41 /xg NO2--N/gm, while nitrosamine levels ranged from less than 0.002 to 1.10 /xg NNO-N/gm. The precision expressed as the percent coefficient of variation (%CV) was less than or equal to 13% for nitrite and less than or equal to 8% for nitrosamine in the samples presented. Table V Analysis of Several Cosmetic Preparations Nitrite Conc. N-Nitroso Content Cosmetic /•g NO2-N/gm /•g NNO-N/gm # of Sample Sample Type (% CV) (% CV) Determinations A cream 0.012 +_ 0.001(8) 0.002 4 B cream 0.014 + 0.001(7) 0.002 2 C lotion 0.015 +_ 0.002(13) 0.002 2 D lotion 0.027 + 0.003(11) 0.002 4 E lotion 0.050 + 0.006(12) 0.002 3 F lotion 13.41 + 1.12(8) 0.26 + 0.01(4) 2 G cream 1.36 + 1.12(8) 1.10 + 0.09(8) 2 H shampoo 0.189 + 0.17(12) 0.26 + 0.01(4) 2 *weight of nitroso nitrogen per gram of cosmetics