Determination of N-nitrosamines in Baby Bottle Rubber Teats by Liquid Chromatography-Atmospheric Pressure Chemical Ionization Mass Spectrometry
This application note illustrates an improved method to quantitate Nnitrosamines in baby bottle teats using APCI-MS. Six rubber nipple formulations were analyzed. Two formulations were composed of natural rubber, two were composed of a synthetic silicone polymer and the remaining two were composed of a synthetic polypropylene. The natural rubber formulations were found to have a significantly higher concentration of NDMA and NMOR compared to both the silicone and the polypropylene polymer formulations.
The presence of N-nitrosamines in baby bottle teats occurs due to the use of various dialkyl amines that are used as accelerators and stabilizers in the vulcanization process. These compounds have been shown to possess significant health risks upon infant ingestion. Current methodology for the analysis of nitrosamines in baby teats involves detection by GC/TEA.1 This method suffers from several drawbacks, the most significant being the inability to identify the peak of interest if another peak is coeluting or masking the analyte. A method is described for the determination of several N-nitrosamines that have been reportedly found in rubber baby bottle teats. They are: N-nitrosomethylethylamine (NMEA), N-nitrosodimethylamine (NDMA), N-nitrosodiethylamine (NDEA), N-nitrosomorpholine (NMOR), N-nitrosopiperidine (NPIP), and N-nitrosopyrrolidine (NPYR). Two samples from each of three different formulations were used to evaluate the level of nitrosamines in each. The first formulation is a synthetic polymer composed of polypropylene and a block copoloymer. The second formulation is a silicone polymer composed of vinyl, dimethyl, and methylvinyl siloxanes. The final sample is a natural rubber/isoprene polymer formulation.
The instrument used for analysis was the API 3000™ LC/MS/MS System equipped with the Atmospheric Pressure Chemical Ionisation (APCI) source. The rubber teats from each of the different formulations were cut up and extracted with methylene chloride. The extract solution was made basic, and the methylene chloride was distilled off. An additional volume of the aqueous phase was distilled into a separatory funnel and extracted with methanol. The extract was concentrated to a volume of 1 mL.
The HPLC system was a Perkin-Elmer Series 200 autosampler and micro pumps. A Keystone Scientific Inertsil ODS2 column (150 X 4.6 mm) was used with a water-methanol gradient increasing from 80% methanol to 90% methanol over 5 minutes at a flow rate of 1.0 mL/min. Both standards and samples were prepared in methanol. Injection volumes were 20 μL. The MRM optimization program of the Analyst® software was utilized to automatically determine optimal product ions, as well as correct settings for lens voltages and collision energies for each MRM transition. These values are shown in Table 1.
Figure 1 shows the XIC of the 1 ppb nitrosamine standard mixture. The signal to noise (S/N) ratio varies according to the analyte. The NDMA had the lowest S/N with a value of 3.0. The highest S/N was obtained by the NMEA with a value of 37.8. Other S/N values observed were NPYR (6.0), NDEA (8.6), NPIP (13.8), and NMOR (13.6). The S/N calculation within the Analyst software requires two regions to be selected in the chromatogram; a region containing the noise and a region containing the peak of interest. The algorithm first subtracts the average backgroundsignal intensity from the background and then divides the subtracted signal by 3 times the standard deviation of the noise region.
signal intensity from the background and then divides the subtracted signal by 3 times the standard deviation of the noise region. Standards containing 1, 5, 10, 50, 100, 500, 1000, and 2000 ppb (ng/mL) of the nitrosoamine standard mixture were prepared. Calibration curves without the use of internal standards produced correlation coefficients for all standards > 0.998. Figure 2 shows the calibration curve for NMEA. Samples were obtained in methanol after extraction from the rubber teats. A 100 ppb spike of NDEA was added to each sample prior to extraction. Representative samples from the relatively “clean” synthetic polymer formulation (Figure 3) and the more contaminated natural rubber formulation (Figure 4) are shown below. Table 2 provides the nitrosamine concentrations observed for each of the polymer formulations.
A method for analyzing nitrosamines in baby bottle rubber teats is presented. This method offers an improvement over the current AOAC method in that LC/MS/MS offers more specificity without sacrificing sensitivity. The data from the analysis of the six samples indicates that the naturalrubber formulation had a higher concentration of NDMA and NMOR then either of the other two formulations. Also, it can be observed that the synthetic polymer formulation had on average less total nitrosamines then the silicone or the natural rubber samples. The calibration data for all samples indicates linearity down to 1 ppb. Replicate analysis of the NDEA spiked samples indicates that there are inconsistant recoveries of this nitrosamine. This is probably due to the formation of artifact nitrosamines during the extraction process.