NEHA November 2022 Journal of Environmental Health

10 Volume 85 • Number 4 A D VANC EME N T O F T H E SCIENCE phthalate-spiked PVC powders at levels below 0.5%. This finding suggests visual identification based on the six key peaks is at least as reliable in detecting lower-level phthalates as are purely software-based searches. Figure 3 shows total phthalate levels measured in 90 of 114 consumer products tested by FTIR and GC/MS. The 24 samples not shown had no phthalate detection by either technique. Product types are summarized in Table 1. Most samples had a PVC matrix; five samples were other polymers (see Supplemental Table 1 at mental). A log scale is used in Figure 2 to ensure subpercent levels are visible. GC/MS phthalate levels ranging from 1.36% to 50% (solid circles) were correctly identified by FTIR as containing phthalates. Phthalate levels of ≤0.45% (open circles) were not detectable by visual inspection or software searching. Thus, the e•ective LOD for product samples was approximately 1%. As expected, this LOD is higher than for the higher-purity reference materials. Some samples (Supplemental Table 1) required extraction-IR for confident plasticizer identification. Figure 4 illustrates how passive extraction removes the matrix and fillers from the spectrum. In Figure 4A, a cowmilking inflation liner spectrum (“intact” in figure) indicates synthetic rubber of polystyrene and polybutadiene with a curved baseline typical of samples containing carbon black. The baseline distortion is caused by similar infrared absorptivities of carbon black and the diamond ATR crystal (Thermo Scientific, 2013). The presence of phthalate (Figure 4A) became clear after extraction. Several nonphthalate plasticizers were identified by FTIR (Table 1). Di(ethylhexyl) terephthalate (DEHT) was the most commonly detected. DEHT, DINCH, acetyltributylcitrate (ATBC), and di(ethylhexyl)adipate (DEHA) were confirmed by GC/MS. GC/MS typically was carried out nonquantitatively due to cost constraints; thus, e•ective LODs could not be determined. Five plasticizers Structure of Ortho-Phthalate Note. R and R’ are alkyl groups. For example, R and R’ in diisononyl phthalate (DINP) are both isononyl groups with nine carbons each. FIGURE 1 Attenuated Total Reflectance–Fourier-Transform Infrared (ATR-FTIR) Spectra of Polyvinyl Chloride (PVC) Powder With Differing Levels of Total Phthalates Note. The six key peaks useful for identifying phthalates in PVC are labeled and marked with dotted lines. Peaks from PVC are not labeled; the most prominent of these is the C-Cl stretch at 610 cm-1. 1,600 1,580 1,124 1,073 1,040 743 Phthalate Level: 7.8% 1.0% 0.8% 0.5% 0.4% 0.3% 0.2% 0.1% 0% Absorbance Wavenumber (cm-1) 1,700 1,500 1,300 1,100 900 700 500 FIGURE 2 Total Phthalates Measured by Gas Chromatography/Mass Spectrometry (GC/MS) in 90 Consumer Product Samples Note. Fourier-transform infrared (FTIR) spectroscopy correctly identified the presence of phthalates in samples represented by solid circles. FTIR did not detect phthalates in samples represented by open circles, all of which had <1% total phthalate. Data were taken from Supplemental Table 1. 0.001 0.01 0.1 1 10 100 Total Phthalates (% by Mass, GC/MS) Consumer Product Samples Tested for Total Phthalates by GC/MS and FTIR FTIR Positive Detection FTIR No Detection FIGURE 3