Verification Analysis for PFAS in Pesticide Products (ACB Project B23-05b)

sr^ UNITED STATES ENVIRONMENTAL PROTECTION AGENCY

g Tb WASHINGTON D.C., 20460

I uj Analytical Chemistry Branch

701Mapes Road

PRO^&<0 Ft Meade> Maryland 20755-5350

OFFICE OF
CHEMICAL SAFETY AND
POLLUTION PREVENTION

5/18/2023

MEMORANDUM

SUBJECT: Verification Analysis for PFAS in Pesticide Products (ACB Project B23-05b)

FROM: Yaorong Qian, Senior Chemist
David French, Chemist
Analytical Chemistry Branch (ACB)

Biological and Economic Analysis Division (BEAD) 5/18/23

Office of Pesticide Programs (OPP)

THROUGH: Thuy Nguyen, Branch Chief v__r_y
-------
Table 1. Pesticide products tested by Lasee et al. (2022) and reported PFOS concentration 1

Pesticide Product

Manufacturer

Active ingredient(s)

PFOS found
(mg/Kg, or ppm)

AVID 0.15 EC*

Syngenta

Abamectin

3.92± 0.51

Pedestal*

Chemtura

Novaluron

9.18± 0.34

Ultra-Pure Oil

BASF

Mineral oil

8.64 ±0.67

Marathon 1%*

OHP

Imidacloprid

13.3± 1.4

Oberon*

Bayer

Spiromesifen

19.2± 1.2

Malathion 5 EC

Drexel

Malathion

17.8± 0.7

BotaniGard 22WP

LAM International

Beauveria bassiana

Corp

Overture 35WP

Valent

Pyridalyl

Conserve

Dow AgroSciences

Spinosad

XXpire

Dow AgroSciences

Spinetoram, Sulfoxaflor

* ACB also purchased these four products from open market and tested for the presence of PFAS, particularly PFOS:
AVID 0.15 EC, Pedestal, Marathon 1%, and Oberon.

ND - Not Detected
OBJECTIVES

The primary objectives of this study were:

• To screen for and quantify the potential presence of twenty-nine (29) PFAS compounds (see
Table 2 for the targeted analytes list) that might be present in these products

• To verify the presence of PFOS as reported by Lasee et al., in the aforementioned pesticide
products

Table 2 - List of twenty-nine (29) PFAS analytes screened in this study utilizing both the Lasee et
al. method and ACB's method with the exception of those noted with an *

PFBA

PFOS

PFTeDA

4:2 FTS*

PFBS

PFNA

PFHxDA

6:2 FTS*

PFPeA

PFNS

FOSAA

8:2 FTS*

PFPeS

PFDA

N-MeFOSAA

PFHxA

PFDS

N-EtFOSAA

PFHxS

PFUdA

9C1-PF30NS

PFHpA

PFDoA

llC-PF30UdS

PFHpS

PFDoS

NaDONA

PFOA

PFTrDA

*These three compounds were not analyzed with the dilution method but analyzed in the ACB's pesticide extraction
method.

-------
STUDY RESULTS

As mentioned above, all samples in this study were analyzed by two different methods,
Lasee's method and ACB's method, for presence of PFAS, especially PFOS. The main difference
between the two methods is in the sample preparation step. The sample preparation step in Lasee's
method is a simple dilution in a solvent/water solution to dilute the matrix using a single instrument
for analysis. ACB's method involves a more intense extraction and clean up procedure to isolate
PFAS compounds from the sample matrix before instrumental analysis, thus reducing matrix
interference which results in better/more accurate detection limits. Instrumental analysis for both
methods is based on the EPA SW 846 method 8327.pdf (epa.gov)2 for detection of PFAS, which
calls for using isotopically labeled (mass labeled) surrogates (standards added during sample
preparation step) and isotopically labeled internal standards (standards added prior to instrument
analysis). A mass labeled compound contains one or more carbon (12C) atom(s) which is replaced
by 13C isotope atom(s). Since their molecular masses are slightly different, the mass spectrometer
can differentiate the mass labeled from the non-labeled PFAS during sample analysis. Use of mass
labeled PFAS is to monitor the performance of the method and to accurately quantify the recovery
of non-labeled PFAS. Instrument response of an identified non-labeled PFAS compound is
compared to the response of its isotopically labeled analog for quantification. Finally, ACB utilized
two instruments to identify and quantify targeted analytes using liquid chromatography / tandem
mass spectrometry (LC/MS/MS) and liquid chromatography/high resolution accurate mass
spectrometry (LC/HRAMS) techniques.

LASEE'S METHOD:

A. Methodology

The analytical procedures described in the published paper (Lasee et al., 2022) followed a
simple solvent/water dilution technique for sample preparation, and the SW846 method
8327 for instrumental analysis of the prepared samples.

The ACB tested the samples using the same procedures, except for the final product
solution, which was made at 100 |ig/ml in methanol. The 100 |ig/ml solution is lOx more
concentrated than that of Lasee et al. (2022) and would ensure that the PFAS compounds, if
present as reported, would be detected.

Three different sets of samples were prepared, and mass labeled PFAS (surrogates),
including mass labeled PFOS, were fortified in each sample to measure the recovery of
PFAS. In addition, ACB spiked both mass labeled and non-labeled PFAS (including PFOS)
in two sets of samples as additional measurements for the detection and recovery of PFAS
by the method.

While Lasee's paper only discussed the use of an LC/HRAMS instrument for their samples,
as noted above, ACB used two different analytical instruments, an LC/HRAMS and an
LC/MSMS for confirmation of results, and mass labeled internal standards for
quantification.

-------
B. Results and Comments

None of the 29 PFAS compounds (Table 2), including PFOS, was detected in any of the
samples above the instrument's background levels, either in those obtained from Lasee or in
those purchased on the open market by ACB, by either LC/MSMS or LC/HRAMS. The
method's background level of each PFAS is 10 parts per trillion (ppt) or less (not taking the
dilution factor into consideration).

As part of our quality control, two sets of QC samples were fortified/spiked with PFAS at
known concentrations (1 and 9 ppm equivalent in the products), either with mass labeled
PFAS standards (a total of 12, including two differently labeled PFOS) or non-labeled
PFAS (a total of 26, including PFOS). Recoveries of PFAS in samples were greater than
60% for the 9 ppm spiking level, and greater than 40% for the 1 ppm level, using both
analytical instruments (LC/MSMS and LC/HRAMS). Presence of the matrix in the diluted
samples did not affect the detection of the spiked PFAS. The techniques used by ACB
would have detected the PFAS if any of the pesticide products contained reported PFAS.

The method detection limits ranged from 0.2-1 ppm (0.5 ppm for PFOS, based on sample
weight) for different PFAS in these pesticide products, taking into consideration of the
dilution factor.

The reported PFOS levels by Lasee et al. ranged from 3.9 ppm to 19.2 ppm in the tested
products (Table 1). These levels are well above the estimated method detection limit of 0.5
ppm for PFOS, and the spiking levels of our QC samples. If present, PFOS would have
been detected in these products.

ACB'S METHOD:

A. Method

All the pesticide products listed in Table 1 were processed and analyzed with a pesticide
extraction method recently developed and validated recently at ACB for PFAS. This
method is specific to these products, which are formulated in non-volatile oil and contain
non-ionic surfactants. Aliquots of purchased pesticide products were also spiked at about
0.5 ppb by ACB with PFAS to monitor the performance of the method. All sample extracts
were analyzed using the SW846 method 8327 and the same LC/MS/MS and LC/HRAMS as
with Lasee's dilution method. This pesticide extraction method has a detection limit of
approximately 0.2 ppb, which is more than lOOOx lower than that of the dilution method.
Both mass labeled surrogates and internal standards were used.

B. Results and Comments

None of the 29 PFAS compounds, including PFOS, was detected in any of the samples
above the method detection limits, either in those obtained from Lasee or in those purchased
by ACB, by either LC/MSMS or LC/HRAMS.

As part of our quality control, samples were fortified/spiked with known concentration of
PFAS (2 ppb), either with mass labeled PFAS standards (a total of 12, including two
differently labeled PFOS) or with non-labeled PFAS (including PFOS), then processed and

-------
analyzed using the ACB's method. All spiked compounds were successfully recovered
(greater than 50% of the fortification level) from the extracts of the pesticide products in the
analyses with both analytical instruments (LC/MSMS and LC/HRAMS). These techniques
used by ACB would have detected the PFAS if any of the pesticide products contained
reported PFAS at or above 0.2 parts per billion (ppb) levels.

Detailed information on ACB's method is in Attachment I

CONCLUSION

BEAD'S Analytical Chemistry Branch could not confirm the presence of PFOS as reported in
Lasee's publication (3.9 ppm to 19.2 ppm), nor detect any PFAS above the method detection limits
(0.2 ppb) in those pesticide products. Some background levels of PFAS were seen at less than 10
ppt (based on instrument response only, and not taking into consideration any dilution factor or
sample preparation factor).

Although the SW846 Test Method 8327 is applicable for analyzing PFAS in samples that have been
previously prepared using solvent dilution or extraction, due to the complex nature of pesticide
products, preparation by solvent dilution is not an appropriate method. A more robust preparation
method is necessary. Furthermore, since low amounts of PFAS are readily observed in the
environment, incorrectly interpreted background data could be multiplied by a large dilution factor
(if dilution was used as sample preparation), resulting in reporting of an overexaggerated
concentration of a background PFAS or a false-positive identification. These large dilution factors
utilized by Lasee et al. could have contributed to the high results obtained in that study.

REFERENCES

1. Steven Lasee, Kaylin McDermett, Naveen Kumar, Jennifer Guelfo, Paxton Payton, Zhao Yang,
Todd A. Anderson, Targeted analysis and Total Oxidizable Precursor assay of several insecticides
for PFAS - ScienceDirect. Journal of Hazardous Materials Letters, 2022, 3, 100067

2. EPA Method 8327. Per- and polyfluoroalkyl substances (PFAS) by Liquid Chromatography/
Tandem Mass Spectrometry (LC/MS/MS). https://www.epa.gov/svstem/files/documents/2021-
07Z8327.pdf

ATTACHMENT

ATTACHMENT I - ACB Method for Pesticide Formulation Containing Non-ionic Surfactants
and Non-volatile Oils

Scope of Method and Application

This method is for the analysis of poly- and per-fluorinated alkyl substances (PFAS) in pesticide
formulations containing non-ionic surfactants and oil. It is based on a QuEChERS (Quick. Easy,

-------
Cheap, Effective, Rugged and Safe) extraction approach, followed by Solid Phase Extraction (SPE)
cleanup to remove excess oily substances, and analysis using Liquid Chromatography-Tandem
Mass Spectrometry (LC-MS/MS). This method is not applicable if formulations contain ionic
surfactants (such as sodium lauryl sulfate, quaternary ammonium compounds, etc.) or only organic
solvents/liquids (petroleum distillates, mineral oil, etc.). A different method Analysis of PFAS in
Oily Matrix (epa.gov) can be used for pesticide products formulated in organic solvents/oils.

Note: Due to the wide occurrence of PFAS in the environment, it is highly recommended to verify
that all supplies and equipment are free of PFAS above the limit of detection. Certain PFAS
compounds have been found in SPE cartridges, SPE manifold, and filters during the method
development.

This method is intended for use by analysts skilled in the performance of solid phase extractions,
the operation of LC-MS/MS instrumentation, and the interpretation of the associated data. EPA has
validated this method through the Analytical Chemistry Branch (ACB) of the Biological and
Economic Analysis Division, Office of Pesticide Programs.

Sample Preparation

Solvents:

• Milli-Q water

• Ethyl acetate

• Hexane

• Methanol

Materials:

• QuEChERS salt mix (6 g MgSOVl .5 g NaCl)

• Ammonium acetate

• Solid Phase Extraction cartridge —Florisil 1 g/6 mL column

• Polypropylene test tubes 15 and 50 mL

Solutions:

• Mobile phase A: Aqueous 20 mM ammonium acetate

• Methanol/water (99/1, v/v)

• Hexanes/ethyl acetate (9/1, v/v)

Standards:

Equipment:

Extraction Standard: Mixture of isotopically labeled PFAS standards,
different from injection standards

Injection Standard: Mixture of isotopically labeled PFAS standards, different
from Extraction standards

Native PFAS standard: Mixture of all the target PFAS compounds.

Geno/Grinder or equivalent
Centrifuge

N-Evap or equivalent
Sonicator

Liquid chromatography/tandem mass spectrometry (LC-MS/MS)

-------
Extraction Procedure:

1. Weigh approximately 4 grams of pesticide products into 50 mL polypropylene centrifuge
tubes.

2. For the procedural blank, transfer approximately 4 grams of Milli-Q water into a 50 mL
tube.

3. For blank spikes and matrix spikes, weigh approximately 4 grams of Milli-Q water and
pesticide product, respectively, into 50 mL tubes.

4. Add appropriate amount of "Extraction Standard" into each sample.

5. Add appropriate amount of spiking solution containing PFAS to spike samples.

6. Mix by vortexing or shaking and then let samples equilibrate after addition of PFAS
standards for 15 minutes.

7. Add 5 mL of Milli-Q water and 25 mL of ethyl acetate to each sample.

8. Shake each sample on Geno/Grinder for 20 minutes at 1000 rpm.

9. Add QuEChERS salt mix (6 g MgS04/l .5 g NaCl) to each sample, shaking by hand to
break all salt clumps.

10. Shake all samples on Geno/Grinder for 20 minutes at 1000 rpm, followed by centrifugation
for 10 minutes at 4000 rpm.

11. Transfer 20 mL of organic supernatant to a new 50 mL centrifuge tube and concentrate to
dryness under N2 flow at 50°C-60°C. Note: Some oil may remain after concentration
depending on product formulation.

12. Add 20 mL of hexane/ethyl acetate (9/1, v/v) to the dried extracts and sonicate for 30
minutes, followed by a round of brief hand-shaking and then centrifugation at 4000 rpm for
10 minutes.

13. For solid precipitates: Decant entire supernatant into a new 50 mL tube.

14. For biphasic layers: Carefully transfer 20 mL of organic supernatant to a new 50 mL tube.

15. Concentrate samples as much as possible as in Step 11. Then combine with 5 mL of
hexane/ethyl acetate (9/1, v/v) and proceed to SPE cleanup.

16. Attach Florisil SPEs to manifold and condition with 10 mL of methanol, followed by 10 mL
of hexane/ethyl acetate (9/1, v/v).

17. Load sample onto SPE, and wash with 10 mL of hexane/ethyl acetate (9/1, v/v). Do not let
the column run dry.

18. Place collection tubes under the manifold and elute samples with 10 mL of methanol.

19. For all samples: Add appropriate amounts of "Injection Standard" mixture to all solutions.

20. Concentrate all samples to dryness. Reconstitute with 1 mL of methanol/water (99/1,
v/v). Note: If precipitate is visible in tube, centrifuge the tubes.

21. Transfer the solutions to LC vials for instrument analysis with LC-MS/MS.

Sample Analysis and Procedure

Calibration:

• Prepare a calibration curve of at least 5 levels in the range of 0.02 - 20 ng/mL of
"Native" compounds.

• Each calibration point should also have "Extraction Standards" and "Injection
Standards" at, for example, 0.50 ng/mL.

Data Analysis Note:

• Quantitation calculations are based on the response ratio of "Native PFAS" signal
to "Extraction Standard" signal.

-------
• Matrix effects can be assessed by comparing responses of "Injection Standards" in
samples and calibration sets.

LC-MS/MS Specifications/Parameters

Equipment: Agilent 6470 LC-MS/MS or Equivalent

Mobile Phase A: Aqueous 20 mM Ammonium Acetate
Mobile Phase B: Methanol
Flow Rate: 0.400 mL/min

Solvent Gradient: 70% Mobile Phase A to 5% Mobile Phase A in 13 min.
Total Run Time: 26 minutes + 5 minutes Post Time Equilibration
MS Operation Mode: Electrospray Negative Ionization (ESI-) mode

List of Analyzed PFAS Compounds

Acronym

Chemical Name

Limits of Quantitation
(ppb)

Comments

PFBA

Perfluoro-n-butanoic acid

0.40

High background

PFPeA

Perfluoro-n-pentanoic acid

0.40

High background

PFHxA

Perfluoro-n-hexanoic acid

0.40

PFHpA

Perfluoro-n-heptanoic acid

0.40

PFOA

Perfluoro-n-octanoic acid

0.40

PFNA

Perfluoro-n-nonanoic acid

0.40

PFDA

Perfluoro-n-decanoic acid

0.40

PFUdA

Perfluoro-n-undecanoic acid

0.40

PFDoA

Perfluoro-n-dodecanoic acid

0.40

PFTrDA

Perfluoro-n-tridecanoic acid

0.40

PFTeDA

Perfluoro-n-tetradecanoic acid

0.40

PFHxDA

Perfluoro-n-hexadecanoic acid

0.40

PFODA

Perfluoro-n-octadecanoic acid

2.00

Low recovery

PFPeS

Perfluoro-1 -pentanesulfonate, Potassium
Salt

0.40

PFHxS

Perfluoro- 1-hexanesulfonate, Sodium Salt

0.40

PFHpS

Perfluoro-1-heptanesulfonate, Sodium Salt

0.40

PFOS

Perfluoro-1-octanesulfonate, Sodium Salt

0.40

PFNS

Perfluoro-1-nonanesulfonate, Sodium Salt

0.40

PFDS

Perfluoro-1-decanesulfonate, Sodium Salt

0.40

PFDoS

Perfluoro-1 -dodecanesulfonate, Sodium
Salt

0.40

FOSAA

Perfluoro-1 -octanesulfonamidioacetic acid

2.00

N-MeFOSAA

N-methylperfluoro-1 -
octanesulfonamidoacetic acid

0.40

N-EtFOSAA

N-ethylperfluoro -1 -octane sulfonamidoacetic
acid

2.00

11C1-
PF30UDS

11 -chloroeicosafluoro-3-oxaundecane-1 -
sulfonate, Potassium Salt

0.40

9-CL-PF30NS

9-chlorohexadecafluoro-3 -oxanonane-1 -
sulfonate, Potassium Salt

0.40

-------
Acronym

Chemical Name

Limits of Quantitation
(ppb)

Comments

4:2 FTS

1H, 1H, 2H, 2H-perfluorohexanesulfonate,
Sodium Salt

2.0

6:2 FTS

1H, 1H, 2H, 2H-perfluorooctanesulfonate,
Sodium Salt

2.0

High background

8:2 FTS

1H, 1H, 2H, 2H-perfluorodecanesulfonate,
Sodium Salt

0.40

ADONA

Dodecafluoro-3H-4,8-dioxanonanoate,
Sodium Salt

0.40

Vote: PFBA, PFPeA, and 6:2 FTS have high background leve
recovery by this extraction procedure.

s in this procedure. PFO

3 A have low

-------
10

-------