Featured Article
4,4’-Methylenedianiline (MDA) is an organic contaminant and irritant used in the production of 4,4’-methylenedianiline diisocyanate (MDI) for the manufacture of polyurethanes.1 MDA is considered a potential carcinogen by the International Agency for Research on Cancer.2 Although there are known risks to health, particularly in animal studies, MDA is not currently regulated. Despite this, the U.S. EPA has developed a method to test for contaminants in water, including MDA, using solid-phase extraction (SPE) and liquid chromatography/tandem mass spectrometry, indicating the importance of a method for MDA.3
MDA has low solubility in water and is typically prepared in organic solvents. However, when analyzing trace levels of MDA, an organic solvent is not required. This article presents a direct aqueous injection method for the detection of trace-level MDA using liquid chromatography/mass spectrometry (LC/MS).
Samples considered in this work were fitness tracker bands made by third-party manufacturers. Also analyzed were pouch-type water bottles marketed to runners and outdoor enthusiasts. Both products can be made of thermoplastic polyurethanes (TPUs), commonly manufactured using MDI, and thus are at risk for containing MDA. These products are also prone to being heated to body temperature, or simply from the heat of a warm summer day. This heat could lead to increased risk of migration of MDA into the water (or other beverage) or onto a person’s skin in the case of the fitness bands.
Experimental
MDA standards were prepared in water from a stock solution, which was heated slightly to promote solubility, and were serially diluted. Samples tested included two different-style pouch water bottles, one polyethylene (PE)-based and the other TPU-based. Multiple colors of one style of fitness tracker band were tested; these bands were listed as TPU-based. All sample migrations were performed in duplicate. All samples and standards were analyzed in duplicate.
For the LC/MS analyses, an Acquity UPLC and Xevo G2-XS Q-TOF-MS (Waters, Billerica, MA) were used with positive-mode electrospray ionization. The reversed-phase LC gradient (Table 1) utilized water with 25 mM ammonium acetate and acetonitrile mobile phases. The column was an Acquity BEH C8 (50 × 2.1 mm, 1.7 µm). Injection volume was 2 µL.
Migrations were performed using three different fluids: 100% water, water with 3% formic acid, and 50:50 water:ethanol. Water was used to simulate sweat, while formic acid and ethanol were used to simulate various beverages that may have been used in the pouch water bottles. The temperature used for all migrations was 37 °C (98.6 °F) and migration times were 1, 24, and 48 hours. A sample aliquot was taken at each time point, and the remainder of the fluid was left for continuation of the migration study. Additional 1-hour migrations were performed on the fitness tracker bands with fresh migration fluids (water only) for 1 hour to compare the rate of MDA migration over time. Only the inside surfaces of the bottles were exposed to the migration solvents. Fitness tracker bands were submerged in each migration fluid, as shown in Figure 1.
Figure 1 – Fitness tracker band migration setup.Infrared spectra were obtained for the TPU bottle and a fitness tracker band using a Nicolet Nexus 670 FTIR with a SmartORBIT attenuated total reflectance (ATR) accessory and diamond ATR crystal (Thermo Fisher Scientific, Waltham, MA). Each spectrum was corrected for optical dispersion and variation in effective optical pathlength using the Thermo/Nicolet Advanced ATR Correction function.
Results
MDA was detected in both the 100% water and 50:50 water:ethanol migration fluids at all times tested for the TPU-based fitness bands. The bands resulted in about 6–35 ppb MDA migrating into water, and 40 to greater than 400 ppb MDA migrating into water:ethanol over 1–48 hours. The 50:50 water:ethanol migration at 48 hours resulted in a peak area outside the linearity of the curve; thus the upper limit of quantitation was reported. A freshwater preparation showed that MDA continued to migrate at a similar rate after an additional 1-hour migration. The quantitation data is summarized in Table 2. All results are concentrations in the migration solutions, not in the polymers; actual amounts in the polymer may be higher.
The extracted ion chromatograms for MDA in a 7.5-ppb standard, a 1-hour 100% water migration of a fitness band, and a 24-hour 100% water migration of a fitness band are provided in Figure 2. All chromatograms are normalized on the y-axis so as to visually compare the intensities. The chromatographic peak for MDA elutes at 1.21 minutes in all samples. Mass spectra are given in Figure 3 for two MDA standards, and 1- and 24-hour fitness band migrations in 100% and 50:50 water:ethanol, respectively. The accurate mass of the [M+H]+ ion for MDA is 199.1235, which is very close to the measured mass (0.5 ppm error). This high mass accuracy allows for a narrower mass window when extracting the MDA ion for quantitation.
Figure 2 – Extracted ion chromatograms for MDA in a 7.5-ppb MDA standard, fitness band in 100% water at 1 hour, and fitness tracker band in 100% water at 24-hour migration times (top to bottom).
Figure 3 – Mass spectra at 1.21 minutes for two MDA standards of 7.5 and 75 ppb, 100% water 1-hour fitness band migration, and 50:50 water:ethanol 24-hour fitness tracker band migration (top to bottom). Accurate mass of MDA is 199.1235 Da.Water with 3% formic acid resulted in complex data that could not be extracted or integrated for MDA; this may be due to other leachables and additional salt content from the formic acid. It is expected that the acid solution would result in more migration of MDA than water alone.
MDA was not detected in any of the migrations of the PE- or TPU-based pouch water bottles. It is unclear why the TPU-based water bottle did not result in detection of MDA. Analyses by FTIR indicated differences in polyester versus polyether functionality between the TPU fitness band and bottle. FTIR spectra of the inside of the TPU-based pouch water bottle and the outside surface of the fitness band are provided in Figure 4. Based on the FTIR data, both are TPU with MDI, but the water bottle is polyether-based and the bands are polyester-based. Additionally, residues left on the FTIR crystal indicated a stearate salt on the surface of the bottle and a waxy amide on the bands (data not shown). Polyether-based TPU may result in less migration of MDA, or the stearate detected on the TPU water bottle inner surface may help prevent migration.
Figure 4 – FTIR spectra of inside surface of the TPU-based pouch water bottle (top) and the outside surface of the fitness tracker band (bottom). Both are TPU with MDI, but the water bottle is polyether-based and the fitness tracker bands are polyester-based. A stearate salt was observed on the surface of the bottle and a waxy amide was observed on the fitness tracker bands.Conclusion
An LC/MS method for trace-level detection of MDA in water-based migrations was successfully applied to TPU fitness tracker bands. Concentrations detected in the migrations were above the American Conference of Governmental Industrial Hygienists (ACGIH) threshold limit value of 0.1 ppm, or the time-weighted average of 0.81 ppb for air. Future studies using the fitness tracker bands may include the use of simulated sweat and elevated temperatures.
References
- https://www.epa.gov/sites/production/files/2016-09/documents/4-4-methylenedianiline.pdf
- https://www.atsdr.cdc.gov/toxfaqs/tfacts122.pdf
- https://cfpub.epa.gov/si/si_public_record_report.cfm?dirEntryId=309413&simpleSearch=1&searchAll=method+540
Noelle M. Elliott, Ph.D., is a scientist; Marshall Henry is an analyst; Sharon Gardner is a research chemist; and Kate Willis is a chemist, Intertek Allentown, 7201 Hamilton Blvd., RD1, Dock #5, Allentown, PA 18195, U.S.A.; tel.: 610-295-0384; e-mail: [email protected]; www.intertek.com. Portions of this article were presented in a poster presentation at the 66th ASMS Conference on Mass Spectrometry and Allied Topics. The poster was entitled, “Analysis of 4,4’-Methylenedianiline in Water Extracts Without Sample Preparation Using Liquid Chromatography Mass Spectrometry (LCMS).”