Advanced Techniques for PFAS Analysis in Soil

CUSTOMER APPLICATION NOTE 73937 Extraction and analysis of poly- and perfluoroalkyl substances (PFAS) from soil The soil samples were extracted using the Thermo Scientific™ Dionex™ ASE™ 350 Accelerated Solvent Extractor, which produced 70–130% recovery of all PFAS target compounds. Accelerated solvent extraction has outperformed commonly used, manual “shaking” extraction methods under the same conditions. After ASE extraction, the solution from the ASE sample collection vials underwent clean-up using solid-phase extraction (SPE) and were analyzed on an LC/MS/MS in a 15-min run. Blanks contained no significant amounts of PFAS. Accelerated solvent extraction is demonstrated to be acceptable for the extraction of short- and long-chain PFAS, with a variety of polarities and head-groups, from soil in the range of 1 ng/g to 400 ng/g. Authors: Matthew S. MacLennan1,2, Daniel Ng1 , David Hope1,3 1 Pacific Rim Laboratories, #103, 19575-55A Avenue, Surrey, B. C. Canada V3S 8P8 2 Matthew@pacificrimlabs.com 3Dave@pacificrimlabs.com Introduction Recent studies suggest that toxic and highly persistent polyand perfluorinated alkyl substances (PFAS) are much more prevalent in tissue and soil than in water. The increasing length of perfluoroalkyl chain in PFAS is correlated strongly to lower water solubility/higher adsorption behavior of a particular PFAS molecule in the environment (i.e., migration of PFAS at soil/water/air interfaces) and in remediation/ filtration (i.e., choice of filtration media or sorbents). There are over 6,000 PFAS commercially available, many of which have high environmental persistence and have been found in water and soils globally. This poses a significant challenge to developing analytical methods, especially for the extraction of a variety of PFAS from solid matrices such as soil. Previously, we reported unsatisfactory (0–50%) recovery of long-chain PFAS from soil using vortex/sonication.1 In the present study, soil was spiked with 24 PFAS (C4-C14 acids, C4-C10 sulfonates, 4:2, 6:2 and 8:2 fluorotelomers, C8 sulfonamide) at 1 ng/g, which were allowed to absorb overnight into the soil samples. Experimental Sample information A 10 g soil sample was in 250 mL polypropylene or polyethylene bottles with no PTFE or other fluorinated polymers. Two grams were taken for analysis. Spiking Sample preparation: Weigh out 2 g of soil sample in a 250 mL glass beaker. • Add 10 g diatomaceous earth to the beaker and mix with the soil sample. • Transfer the sample mixture into a 100 mL stainless steel ASE cell with cellulose filter at the bottom of the cell. • Top up the cell with diatomaceous earth. – Spike extraction surrogate and native standard. – Place a 250 mL bottle with septa cap in the bottle tray of the Dionex ASE 350 system. Accelerated solvent extraction (Dionex ASE 350 setup) Rinse settings Solvent 100% acetone Volume 13 mL Cycles 3 Extraction settings Cell type Stainless steel Oven temperature 100 °C Static cycle Time 300 s Solvent 80:20 methanol/ acetonitrile Volume 50 mL Cycles 3 Purge time 120 s Clean-up Sample clean-up used a styrene-divinylbenzene (SDVB) polymer SPE cartridge (500 mg, 6 mL), on a vacuumcontrolled manifold, under the following sequential conditions: • Condition cartridges with 15 mL methanol. • Run 20 mL of reagent water through the cartridge. • Transfer the extracted sample solution from the Dionex ASE 350 system to the SPE cartridge using a largevolume sampler. • Rinse the sample bottle with 10 mL reagent water and transfer to the cartridge. • Vacuum dry the SPE cartridge. Elution phase • Rinse the sample bottle with 10 mL methanol and transfer to the SPE cartridge. • Elute all methanol extracts into a polypropylene centrifuge tube. LC parameters A Thermo Scientific™ Vanquish™ LC, with all Teflon™ lines replaced by PEEK tubing, coupled to a Thermo Scientific™ TSQ Quantis™ triple quadrupole mass spectrometer, was used for sample analysis. Solvent B 10 mM ammonium acetate in 19% v/v acetonitrile/81% methanol Solvent A 10 mM ammonium acetate in 19% v/v acetonitrile in water Column temperature 25 °C Gradient Solvent ramps from 40% Solvent B to 90% Solvent B over 15 min LC flow rate 0.300 mL/min 2 3 MS parameters The H-ESI source was used in the negative ionization mode and the optimized MS parameters were as follows: Q1 resolution 0.7 Da Q3 resolution 1.2 Da Use cycle time True Cycle time 0.5 s CID gas 2 mTorr Source fragmentation 0 V Chromatographic peak width 6 s ESI negative voltage 1,500 V Sheath gas 57.6 arb Aux gas 2.4 arb Sweep gas 0.4 arb Ion transfer tube temperature 325 °C Vaporizer temperature 350 °C Table 1. Monitored SRM transitions Compound Retention time (min) Precursor (m/z) Product (m/z) Collision energy (V) RF lens (V) PFBA 1.38 212.979 168.97 10.23 56 13C4-PFBA 1.38 216.979 171.97 12.81 71 PFPeA 1.89 262.976 219.042 10.23 79 13C5-PFPeA 1.89 268 223 10.23 64 PFHxA 3.00 312.9 268.8 10.23 73 13C5-PFHxA 3.00 318 273 10.23 76 4:2 FTS 2.70 326.974 307.042 19 129 13C2 -4:2 FTS 2.70 328.974 81.040 19 147 PFPeS 3.76 307.042 80.042 35 151 PFBS 2.33 298.912 79.946 32 152 13C2 -PFBS 2.33 302 79.946 32 152 PFHpA 4.74 362.97 319 10.23 81 13C4-PFHpA 4.74 367 322 10.23 82 PFHxS 5.56 398.912 79.929 37.2 154 13C3-PFHxS 5.56 401.947 79.957 37.3 171 PFOA 6.75 412.966 369 10.23 91 13C8-PFOA 6.75 421 376 10.23 92 6:2 FTS 6.36 426.968 406.988 22.43 164 13C2 -6:2 FTS 6.36 428.968 81.040 23.6 164 PFHpS 7.43 448.933 80.012 39.67 184 PFNA 8.68 462.963 418.946 10.23 99 13C9-PFNA 8.68 472 427 10.23 97 PFOS 9.19 498.862 79.946 40.78 164 13C8-PFOS 9.19 507 79.917 40.63 169 PFDA 10.39 512.96 469 10.23 108 13C6-PFDA 10.39 519 474 10.23 109 8:2 FTS 10.17 526.962 506.97 25.73 190 13C2 -8:2FTS 10.17 528.962 81.010 25.27 190 PFNS 10.73 548.927 80.071 44.4 177 PFUdA 11.87 562.957 519 10.23 117 13C7 -PFUdA 11.87 570 525 10.23 117 NMeFOSAA 11.13 569.925 418.926 19.36 200 4 Compound Retention time (min) Precursor (m/z) Product (m/z) Collision energy (V) RF lens (V) d3-N-MeFOSAA 11.13 572.912 418.958 20.39 167 PFOSA 9.47 497.946 78.071 30.24 175 13C8-PFOSA 9.47 505.946 78.071 31.57 186 NEtFOSAA 11.89 583.983 526 19.02 187 d5-N-EtFOSAA 11.89 588.962 530.97 18.96 144 PFDS 12.06 598.924 80.042 44.92 192 PFDoA 13.11 612.9 569 10.23 126 13C2 -PFDoA 13.11 615 570 10.23 122 PFTrDA 14.16 662.95 619 10.23 134 PFTeDA 15.06 712.95 668.97 11.782 143 13C2 -PFTeDA 15.06 714.95 670 10.23 142 PFOA and PFOS Alternatives HFPO-DA 3.52 284.96 168.887 17.35 68 13C3-HFPO-DA 3.52 286.912 168.833 10.23 67 ADONA 5.09 376.912 250.988 10.23 91 9Cl-PF3ONS 10.21 530.825 350.917 24.67 167 11Cl-PF3OUdS 12.80 630.825 450.887 27.17 254 All PFAS compounds analyzed in this method are shown in Figure 1. Figure 1. Chromatogram of PFAS compounds analyzed Table 1. Monitored SRM transitions (cont.) 100 95 90 85 80 75 70 65 60 55 50 45 40 35 30 25 20 15 10 5 0 Relative abundance 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 1.38 Time (min) 2.00 2.72 3.16 3.98 4.95 5.76 6.18 4.31 6.35 7.03 8.03 8.73 9.09 10.20 10.69 11.19 11.59 12.46 11.75 12.59 13.71 14.68 15.55 15.98 16.40 15.34 14.49 13.48 12.30 10.91 9.27 7.41 5.40 3.52 2.17 1.49 9.72 C4 acid (PFBA) C14 acid (PFTeDA) PFOA PFOS Acids Sulfonates Fluorotelomer Sulfonamides 5 Figure 2. Native analytes found at appreciable levels in a soil sample Chromatogram of a soil sample Only native analytes found at appreciable levels in the soil sample are shown in Figure 2. These levels are between 1 and 5 ng/g, with the exception of PFOS at 50 ng/g. From top to bottom, the analytes are PFPeA, PFHxA, PFHpA, PFOA, PFNA, PFHxS, and PFOS. The other 21 PFAS analytes were at or below detection limit. Internal standard recoveries (spiked in at 10 ng/g or 5 ng/g) are as listed in Table 2. Table 2. Recovery of the isotopically labeled PFAS compounds Compound Recovery (%) 13C4-PFBA 71 13C5-PFPeA 93 13C5-PFHxA 97 13C4-PFHpA 96 13C8-PFOA 94 13C9-PFNA 104 13C6-PFDA 99 13C7 -PFUdA 95 13C2 -PFDoA 97 13C2 -PFTeDA 108 Compound Recovery (%) 13C3-PFBS 98 13C3-PFHxS 95 13C8-PFOS 91 13C3-HFPODA 56 2 H3-NMEFOSAA 93 2 H3-NETFOSAA 90 13C8-FOSA 92 13C2 -4:2FTS 110 13C2 -6:2FTS 93 13C2 -8:2FTS 98 PFAS background levels The Dionex ASE 350 system, which was used to develop this method, contains Teflon lines. Due to the lines being exposed to a variety of solvents over several years, the PFAS background is at a minimum. Table 3 shows the levels of PFAS measured in the blanks processed within different spike batches (i.e., in a batch of soil samples spiked with PFAS, these are the PFAS levels in the blanks processed with that batch). These blanks contain diatomaceous earth and isotopically labeled PFAS internal standards (surrogates), but no spiked native PFAS and no soil. Table 4 shows the linearity for PFAS in soil. 6 Table 3. Measured PFAS levels in blanks Spike level batch 1 ng/g spike 20 ng/g spike 400 ng/g spike Units ng/g ng/g ng/g ng/g PFBA 0.01 0.01 0.05 0.05 PFPeA n.d. n.d. n.d. n.d. PFHxA 0.01 0.01 0.03 0.04 PFHpA n.d. n.d. 0.02 0.01 PFOA 0.01 0.01 0.04 0.04 PFNA n.d. n.d. n.d. n.d. PFDA n.d. n.d. n.d. n.d. PFUdA n.d. n.d. n.d. n.d. PFDoA n.d. n.d. n.d. n.d. PFTRDA n.d. n.d. n.d. n.d. PFTEDA n.d. n.d. n.d. n.d. NMEFOSAA n.d. n.d. n.d. n.d. Table 4. PFAS in soil linearity Spike level 1 ng/g 5 ng/g 20 ng/g 100 ng/g 400 ng/g Slope r2 PFBA 0.979 5.05 21.7 101.0 408 1.020 1.000 PFPeA 1.035 5.22 22.0 101.4 423 1.058 1.000 PFHxA 1.024 5.15 22.1 102.4 429 1.073 1.000 PFHpA 0.985 5.28 22.4 99.2 423 1.056 1.000 PFOA 1.02 5.02 22.1 100.7 425 1.062 1.000 PFNA 1.032 5.21 21.8 102.6 426 1.064 1.000 PFDA 1.000 5.06 21.5 100.5 428 1.071 1.000 PFUdA 0.982 5.01 22.6 96.8 418 1.044 1.000 PFDoA 1.05 5.43 23.5 77.3 339 0.841 0.999 PFTRDA 0.567 3.65 15.5 45.4 200 0.496 0.998 PFTEDA 1.076 5.57 23.9 76.6 317 0.786 0.999 NMEFOSAA 1.13 4.86 22.8 97.2 368 0.915 1.000 NETFOSAA 1.097 5.18 19.5 117.0 424 1.061 0.999 FOSA 0.991 5.16 21.7 93.0 438 1.097 0.998 PFBS 0.966 5.14 22.4 104.6 453 1.133 1.000 PFPeS 0.915 4.93 21.1 99.6 433 1.084 1.000 PFHxS 0.945 4.98 21.6 110.6 451 1.129 1.000 PFHpS 0.976 4.55 22.8 111.4 467 1.169 1.000 PFOS 1.076 6.14 20.3 108.2 468 1.172 1.000 PFNS 0.893 5.51 21.4 107.1 462 1.156 1.000 PFDS 0.999 5.54 20.8 104.4 447 1.119 1.000 4:2FTS 1.129 5.89 22.2 57.8 272 0.672 0.997 6:2FTS 1.128 6.19 21.8 89.2 430 1.074 0.998 8:2FTS 1.149 5.43 21.0 79.8 384 0.958 0.998 Spike level batch 1 ng/g spike 20 ng/g spike 400 ng/g spike Units ng/g ng/g ng/g ng/g NETFOSAA n.d. n.d. n.d. n.d. FOSA n.d. n.d. n.d. n.d. PFBS n.d. n.d. 0.01 n.d. PFPeS n.d. n.d. n.d. n.d. PFHxS n.d. n.d. n.d. n.d. PFHpS n.d. n.d. n.d. n.d. PFOS n.d. n.d. n.d. n.d. PFNS n.d. n.d. n.d. n.d. PFDS n.d. n.d. n.d. n.d. 4:2FTS n.d. n.d. n.d. n.d. 6:2FTS n.d. n.d. n.d. n.d. 8:2FTS n.d. n.d. n.d. n.d. For Research Use Only. © 2021 Thermo Fisher Scientific Inc. All rights reserved. Teflon is a trademark of the Chemours Company FC, LLC. All other trademarks are the property of Thermo Fisher Scientific and its subsidiaries. This information is presented as an example of the capabilities of Thermo Fisher Scientific Inc. products. It is not intended to encourage use of these products in any manners that might infringe the intellectual property rights of others. Specifications, terms and pricing are subject to change. Not all products are available in all countries. Please consult your local sales representative for details. CAN73937-EN 0121S Find out more at thermofisher.com Conclusion Accelerated solvent extraction can extract a variety of PFAS from soil including acids, sulfonates, fluorotelomer sulfonates, and sulfonamide compounds. Although the Dionex ASE 350 system utilized in this method contained Teflon lines, the lines had been used under a variety of different solvent conditions, effectively reducing the PFAS background contamination to a minimum. Isotopic dilution quantification was utilized for most analytes and demonstrated linearity for all PFAS studied in soil over the range of 1 ppb to 400 ppb. PFTrDA linearity is biased low which is partially due to PFTrDA being quantified via the internal standard method against 13C2 -PFTeDA and not an isotopically labeled analog of PFTrDA, but more strongly due to the effects of naturally abundant 13C2 -PFTeDA (from the native spiked levels) artificially increasing the recoveries of internal standard, thus underestimating the native levels of PFTrDA. Previously experienced difficulties recovering long-chain PFAS from soil using sonication/vortex methods were surmounted by using ASE, demonstrating that the absence of long-chain PFAS is a true absence and not simply low recovery. The approach described here was applied to a soil sample and it was found to contain between 1 and 50 ng/g of several PFAS. Accelerated solvent extraction is an acceptable method for extracting a wide selection of PFAS, from 4-carbon to 14-carbon fluoroalkyl chain lengths and five different polar head-groups, from soil over a wide range of concentrations. Accelerated solvent extraction shows high potential for effective extraction of the growing list of PFAS from solid samples. Reference 1. MacLennan, Matthew S.; Ng, Daniel; Hope, David (2019): Extraction of poly-and perfluorinatedalkyl substances (PFAS) from solid matrices. Society of Ecotoxicology and Chemistry (SETAC) North America, 40th Annual Meeting, Toronto, Poster. https://doi. org/10.6084/m9.figshare.13557185.v1