J. Soc. Cosmet. Chem., 43, 207-213 (July/August 1992) Assay of 1,4-dioxane in cosmetic products by solid-phase extraction and GC-MS SANTO SCALIA, FABIO TESTONI, GIUSEPPE FRISINA, and MARIO GUARNERI, Dipartimento di Scienze Farmaceutiche, Universit• di Ferrara, Via Fossato di Mortara i7/i9, 44100 Ferrara (S.S., M.G. ), and Himont Research Centre, Ferrara (F.T., G.F.), Italy. Received January 27, 1992. Synopsis A rapid procedure has been developed for the assay of 1,4-dioxane in cosmetic products by gas chroma- tography selected-ion monitoring mass spectrometry. After solid-phase extraction using Bakerbond silica- and Bakerbond Cx8-disposable cartridges, samples were injected directly into a Poraplot Q capillary col- umn, employing toluene as an internal standard. Recovery of !,4-dioxane from different cosmetic matrices was between 9 !. ! and 93.2%, and the reproducibility of the method was better than 4.3 % relative standard deviation. The minimum quantifiable amount was 3 mg/kg. Of the total commercial cosmetics investigated (n = 25), 56% were found to contain 3.4-!08.4 mg/kg of !,4-dioxane. INTRODUCTION Polyethoxylated surfactants are widely used in shampoo and bath preparations (1) and are generally contained in skin cleansing products and lotion formulations (2). During the polymerization of ethylene oxide to produce the polyoxyethylene moiety of the emulsifiers, 1,4-dioxane may be formed (3-5). Hence, cosmetics containing ethoxylated surfactants may be contaminated by 1,4-dioxane (6-10), which has been shown to be carcinogenic in rats and mice (! 1, 12) and to be absorbed through the intact skin of animals (13). Furthermore, this compound has been classified as a possible carcinogen to humans (14,15). According to the European Economic Community directive on cosmetics (16), !,4-dioxane must not be present in commercial products. Consequently, the assay of this substance in marketed cosmetics is of direct concern to consumers. Published methods for the quantitative determination of 1,4-dioxane in finished cos- metic products are based on gas chromatography (GC) (6) or headspace GC (7,8). These techniques, however, have distinct drawbacks, such as complex and time-consuming sample pre-treatment (6), extremely long equilibrium times (7,8), the need for extensive calibrations (7,8), and unsatisfactory reproducibility and recoveries (6). More recently, an improved GC procedure that requires minimal sample preparation has been reported 207

208 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS (10) yet the lack of specificity of the flame ionization detector (FID) used and the fact that application of the method is restricted to shampoo products are disadvantages. In a previous paper (9), we described the first reversed-phase high-performance liquid chromatographic (RP-HPLC) method with UV spectrophotometry for the rapid assay of 1,4-dioxane in different cosmetic preparations. However, the selection of short UV wavelengths (9) for the detection of this compound lacking a strong chromophore generally results in low selectivity, due to increased interference from matrix constitu- ents. Routine analyses of 1,4-dioxane in cosmetics necessitated a simple method possessing a high degree of specificity and the required sensitivity. This study describes a procedure for the assay of 1,4-dioxane in cosmetic products by GC-mass spectrometry (GC-MS) using selected-ion monitoring (SIM). Prior to GC-MS analysis, rapid and efficient purification of the complex cosmetic matrices is achieved with combined silica- and octadecylsilica-disposable cartridges. The application of the method to the determina- tion of 1,4-dioxane in a wide range of commercial cosmetics is also reported. EXPERIMENTAL MATERIALS HPLC-grade 1,4-dioxane, hexane, dichloromethane, toluene, and acetonitrile were sup- plied by J. T. Baker (Phillipsburg, NJ). Bakerbond C•8 (BB-C•8), Bakerbond CN (BB-CN), and Bakerbond silica (BB-SiOH) cartridges were obtained from J. T. Baker. Commercial cosmetics containing ethoxylated surfactants were from retail stores or from manufacturers or importers of these products. GAS CHROMATOGRAPHY-MASS SPECTROMETRY GC-MS analyses were performed with an HP 5890 gas chromatograph (Hewlett Pack- ard, Palo Alto, CA) equipped with an HP 5970 mass-selective detector (Hewlett Packard) with transfer line held at 280øC. Samples (1 }xl) were introduced using on- column injection. A deactivated fused-silica pre-column (5 m x 0.32 mm i.d.) and a Poraplot Q analytical column (25 m x 0.32 mm i.d. Chrompack Italia, Milan, Italy) were used. The operating conditions were: initial temperature, 40øC ramp 40-220øC, rate 40øC/min carrier gas, helium inlet pressure, 100 kPa. The MS, connected directly to the capillary column outlet, was operated in the selected-ion monitoring mode scanning m/z 31, 58, and 88 for 1,4-dioxane and m/z 91 and 92 for toluene (internal standard) with dwell times of 350 ms. The GC-MS system was controlled by an HP 5970 MS Chemstation Rev. 3.2 data station. Quantification was on the basis of peak area for the ratio 1,4-dioxane/toluene. STANDARD SOLUTIONS A 1,4-dioxane stock solution was prepared at a level of 20 mg/ml in acetonitrile, and an aliquot of it was diluted to make standard solutions in the concentration range 0.3-20 }xg/ml. An internal standard solution (5.9 }xg/ml) was prepared by quantitative