Total Fluorine Measurements in Food Packaging: How Do Current Methods Perform?

Abstract

Per- and polyfluoroalkyl substances (PFASs) represent a class of more than 4000 compounds. Their large number and structural diversity pose a considerable challenge to analytical chemists. Measurement of total fluorine in environmental samples and consumer products is therefore critical for rapidly screening for PFASs and for assessing the fraction of unexplained fluorine(i.e., fluorine mass balance). Here we compare three emerging analytical techniques for total fluorine determination: combustion ion chromatography (CIC), particle-induced Î-ray emission spectroscopy (PIGE), and instrumental neutron activation analysis (INAA). Application of each method to a certified reference material (CRM), spiked filters, and representative food packaging samples revealed good accuracy and precision. INAA and PIGE had the advantage of being nondestructive, while CIC displayed the lowest detection limits. Inconsistencies between the methods arose due to the high aluminum content in the CRM, which precluded its analysis by INAA, and sample heterogeneity (i.e., coating on the surface of the material), which resulted in higher values from the surface measurement technique PIGE compared to the values from the bulk volume techniques INAA and CIC. Comparing CIC-based extractable organic fluorine to target PFAS measurements of food packaging samples by liquid chromatography-tandem mass spectrometry revealed large amounts of unidentified organic fluorine not captured by compound-specific analysis.

Full text

Total Fluorine Measurements in Food Packaging: How Do Current
Methods Perform?
Lara Schultes,*
,†
Graham F. Peaslee,
‡
John D. Brockman,
§
Ashabari Majumdar,
‡
Sean R. McGuinness,
‡
John T. Wilkinson,
‡
Oskar Sandblom,
†
Ruth A. Ngwenyama,
§
and Jonathan P. Benskin
†
†
Department of Environmental Science and Analytical Chemistry (ACES), Stockholm University, Svante Arrhenius väg 8, SE-10691
Stockholm, Sweden
‡
Department of Physics, University of Notre Dame, Notre Dame, Indiana 46556, United States
§
Department of Chemistry, University of Missouri, Columbia, Missouri 65211, United States
*
SSupporting Information
ABSTRACT: Per- and polyfluoroalkyl substances (PFASs) represent a class of
more than 4000 compounds. Their large number and structural diversity pose a
considerable challenge to analytical chemists. Measurement of total fluorine in
environmental samples and consumer products is therefore critical for rapidly
screening for PFASs and for assessing the fraction of unexplained fluorine(i.e.,
fluorine mass balance). Here we compare three emerging analytical techniques
for total fluorine determination: combustion ion chromatography (CIC),
particle-induced γ-ray emission spectroscopy (PIGE), and instrumental
neutron activation analysis (INAA). Application of each method to a certified
reference material (CRM), spiked filters, and representative food packaging
samples revealed good accuracy and precision. INAA and PIGE had the
advantage of being nondestructive, while CIC displayed the lowest detection
limits. Inconsistencies between the methods arose due to the high aluminum content in the CRM, which precluded its analysis
by INAA, and sample heterogeneity (i.e., coating on the surface of the material), which resulted in higher values from the
surface measurement technique PIGE compared to the values from the bulk volume techniques INAA and CIC. Comparing
CIC-based extractable organic fluorine to target PFAS measurements of food packaging samples by liquid chromatography−
tandem mass spectrometry revealed large amounts of unidentified organic fluorine not captured by compound-specific analysis.
1. INTRODUCTION
Per- and polyfluoroalkyl substances (PFASs) are a class of
ubiquitous chemicals that have found innumerable industrial
and consumer applications over the past seven decades.
1
PFASs can be categorized as polymeric or nonpolymeric,
2
collectively amounting to more than 4700 CAS-registered
substances according to the OECD.
3
Environmental concerns
pertaining to PFASs are centered primarily on the perfluoro-
alkyl acids (PFAA), a subclass of PFAS which display extreme
persistence and chain-length-dependent bioaccumulation and
adverse effects in biota.
2
The water- and grease-repellent properties of PFASs have
led to their extensive use in food contact paper and packaging.
Concentrations in the range of 1.0−1.5% per fiber dry weight
are typical in most finished products.
4
Historically, most
fluorinated coatings for paper and board were based on
perfluorooctanesulfonate (PFOS) precursors, such as N-ethyl
perfluorooctane sulfonamido alcohol-based phosphate diesters
(SamPAPs).
5,6
However, as a consequence of the phase-out of
PFOS and its precursors by 3M in 2001, most contemporary
formulations are now based on acrylate polymers with
fluorotelomer or sulfonamido alcohol side chains or perfluoro-
polyether-based polymers (PFPEs).
4,7
Several peer-reviewed
studies have reported the occurrence of PFAAs, polyfluoroalkyl
phosphates (PAPs), fluorotelomer alcohols, and saturated and
unsaturated fluorotelomer acids in various types of food
packaging materials.
5,8−13
While the importance of PFASs in
food packaging as a human exposure source remains unclear,
some PFASs have been shown to migrate from food packaging
into food.
14,15
In response to concerns surrounding PFASs,
several states in the United States have implemented bans on
the use of these chemicals in food packaging,
16
and the Danish
Ministry of the Environment and Food has established a
total fluorine indicator value of 0.1 μg/cm2in food
packaging.
17
Given their vast number and structural diversity, a
comprehensive characterization of PFASs in consumer
products represents a considerable analytical challenge.
18
Typically, a limited number of PFASs are quantified using a
Received: December 19, 2018
Revised: January 26, 2019
Accepted: January 28, 2019
Letter
pubs.acs.org/journal/estlcu
Cite This: Environ. Sci. Technol. Lett. XXXX, XXX, XXX−XXX
© XXXX American Chemical Society ADOI: 10.1021/acs.estlett.8b00700
Environ. Sci. Technol. Lett. XXXX, XXX, XXX−XXX
Downloaded via STOCKHOLM UNIV on February 5, 2019 at 11:21:28 (UTC).
See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

combination of solvent extraction and either liquid chromatog-
raphy−tandem mass spectrometry (LC−MS/MS; ionic
PFASs) or gas chromatography−mass spectrometry (GC−
MS; semivolatile PFASs). These methods cannot capture all
PFASs, and even when high-resolution MS is employed (i.e.,
suspect or nontarget screening), a lack of authentic standards
precludes unequivocal identification and quantification. As a
result, questions about how many PFASs are missed during
routine analysis remain. Several analytical approaches have
emerged for quantifying total fluorine (TF) regardless of
chemical structure or molecular weight (reviewed by
McDonough et al.
19
). These approaches can be used for
rapid screening of PFASs or in combination with targeted
analyses (i.e., LC−MS/MS or GC−MS) to assess the fluorine
mass balance in a sample. Combustion ion chromatography
(CIC) is the most common method and was first used for
fluorine mass balance experiments in 2007 by Miyake et al.
20
Since then, there have been several applications of CIC for
determination of TF in environmental samples and consumer
products.
21−24
Another method, particle-induced γ-ray emis-
sion (PIGE) spectroscopy, is a long-established ion beam
technique used for analysis of solid materials.
25−28
The
approach was recently applied to papers and textiles by
Robel et al., who showed that ΣPFAS concentrations
accounted for a mere 0.2−14% of the TF content.
29
Lastly,
instrumental neutron activation analysis (INAA) is another
nuclear technique with widespread uses,
30−32
with one recent
application involving measurement of fluorine in biological and
environmental matrices.
33
The current paper reports the first
application of INAA to consumer products. Other approaches
for measuring TF exist, such as inductively coupled plasma
(ICP) MS,
34
molecular absorption spectroscopy,
35
and X-ray
photoelectron spectroscopy;
36
only the latter has recently been
applied to fluorine mass balance experiments in consumer
products. The data produced by these approaches have not yet
been compared. This is clearly needed, given the growing
interest in fluorine mass balance experiments and the use of TF
for regulation of PFASs in consumer products.
The objectives of this study were (1) to compare the
accuracy, precision, linearity, and detection limits of TF meas-
urements by CIC, PIGE, and INAA, (2) to assess the
limitations of each approach, and (3) to assess the fluorine
mass balance in several food packaging materials from the
Swedish market using LC−MS/MS and CIC.
2. MATERIALS AND METHODS
2.1. Sample Collection and Preparation. Total fluorine
measurements from three different laboratories were compared
using (a) a certified reference material (CRM), (b) PFOA-
spiked cellulose filters, and (c) a variety of food packaging
materials (Table S1). CIC analysis was carried out at
Stockholm University (SU); PIGE analysis was carried out at
the University of Notre Dame (UND), and INAA was carried
out at the University of Missouri Research Reactor (MURR).
The CRM (BCR-461, fluorine in clay) was purchased from
Sigma-Aldrich and subsampled at SU into 13 mL poly-
propylene tubes prior to being shipped to UND and MURR
for direct analysis. Cellulose filter papers (Whatman) were
prepared by UND by spiking perfluorooctanoic acid (PFOA)
solutions (prepared in methanol with food coloring to
determine the area with which the standard was applied)
with total masses of 0, 1.9, 3.8, 9.5, 19.1, and 38.2 μg of PFOA.
The methanol was evaporated to dryness prior to analysis.
Triplicates of each PFOA-spiked filter were sealed in individual
zip-lock bags and shipped to SU and MURR for analysis.
Finally, three french-fry bags (FF1−FF3) and six microwave
popcorn bags (MP1−MP6) purchased in Sweden in 2012 were
selected for analysis in this study. Pieces (2 cm ×2 cm) of each
sample were cut and sealed in individual zip-lock bags and
shipped to UND and MURR for analysis.
2.2. Extraction of Food Packaging. For targeted PFAS
analysis, food packaging samples were extracted according to a
Gebbink et al.
37
In short, samples (5 cm ×5 cm) were cut into
small pieces, fortified with internal standards (0.5 ng each),
and stirred in 40 mL of methanol at room temperature for 8 h.
The extract was concentrated under a stream of nitrogen to
approximately 1 mL, cleaned up using EnviCarb, and fortified
with a recovery standard (0.5 ng). For analysis of total
extractable organic fluorine (EOF), the extraction procedure
was modified slightly by omitting the addition of internal and
recovery standards and cleanup by EnviCarb.
2.3. Instrumental Analysis. Detailed descriptions of
instrumental analysis and quantification for CIC, PIGE, and
INAA can be found in the Supporting Information. A brief
overview is provided here.
2.3.1. TF and EOF Analysis by CIC. TF and EOF were
analyzed according to the method described by Schultes et
al.
24
Briefly, samples (neat samples and extracts) were placed
directly onto a ceramic boat that was introduced into a
combustion oven (HF-210, Mitsubishi) heated to 1100 °C
under an atmosphere of argon (carrier gas) and oxygen
(combustion gas) for ∼5 min. All gases were collected in Milli-
Q water (GA-210, Mitsubishi). Ions were separated on an ion
exchange column and measured by conductivity detection.
2.3.2. TF Analysis by PIGE. TF was analyzed according to
the method described by Ritter et al.
28
Briefly, samples were
mounted across a stainless steel target frame and bombarded
with a 3.4 MeV beam of protons (∼50 nA for 180 s) to
produce γ-rays, which were measured using a high-purity
germanium detector (HPGe, Canberra, 20%) located at
approximately 75°to the beam. The combined number of
counts of two γ-rays characteristic of the decay of the 19F
nucleus at 110 and 197 keV/μC of beam delivered is
proportional to the TF. The beam intensity was measured in
a suppressed Faraday cup before and after each 3 min run and
normalized to a current measured in a tantalum collimator near
the beam exit window. For the powdered CRM material,
replicate targets were prepared by hydraulically compressing
the powder into a self-supporting pellet at approximately 350
bar for 30 s and then taped onto target frames.
2.3.3. TF Analysis by INAA. Samples were analyzed
according to the method described by Spate et al.
32
Briefly,
samples were weighed into 0.5 mL high-density polyethylene
(HDPE) vials. The vials were encapsulated in HDPE “rabbits”
for irradiation in the pneumatic tube irradiation position of the
MURR at a neutron flux of 5.5 ×1013 ncm
−2s−1. Samples
were irradiated for 7 s, decayed for 11 s, and were counted for
30 s using an HPGe detector (Canberra, 20%). The 20F decays
(11.03 s half-life) by β-particle emission with a characteristic γ-
ray at 1633.6 keV. A correction was made for the fast neutron
reaction 23Na(n,α)20F using a single-element Na standard
irradiated and counted under the same conditions. The
neutron activation product 24Na emits a characteristic γ-ray
at 1368.6 keV. The measured ratio of the 1633.6 keV/1368.6
keV γ-ray in the single-element standards is used to correct
interference in the samples based on the measured 1368.6 keV
Environmental Science & Technology Letters Letter
DOI: 10.1021/acs.estlett.8b00700
Environ. Sci. Technol. Lett. XXXX, XXX, XXX−XXX
B

γ-ray. The correction in samples analyzed in this study was
<1%.
2.3.4. Target PFAS Analysis by UHPLC−MS/MS. Target
PFASs, including perfluoroalkyl carboxylic acids (PFCAs, C4−
C15), perfluoroalkyl sulfonic acids (PFSAs, C4, C6, C8, and
C10), perfluorooctane sulfonamide (FOSA), perfluoroalkane
sulfonamidoacetic acids, fluorotelomer sulfonates (4:2, 6:2,
and 8:2 FTSAs), fluorotelomer carboxylic acids (5:3, 7:3, and
9:3), ADONA, F53-B, and polyfluoroalkyl phosphoric acid
mono- and diesters (mono- and diPAPs, respectively), were
analyzed using a Waters Acquity UHPLC instrument coupled
to a Xevo TQ-S triple-quadrupole mass spectrometer
according to the methods described by Vestergren et al.
38
and Gebbink et al.
37
Instrumental parameters are listed in
Table S2. Limits of detection (LODs) for individual PFASs are
based on the average concentration in the extraction blank plus
3 times the standard deviation. In the absence of a blank signal,
the LOD was based on the concentration of the lowest
calibration standard at a minimum signal-to-noise ratio of 3.
Individual LODs are listed in Table S8.
2.4. Fluorine Mass Balance Calculations. To compare
PFAS concentrations (CPFAS, nanograms of PFAS per gram)
derived from UHPLC−MS/MS analysis to EOF and TF
(CF_EOF and CF_TF, respectively; nanograms of F per gram)
measured by CIC, molecular PFAS concentrations are
converted to fluorine equivalents using the following equation:
CnA C/MW
F PFAS F F PFAS PFAS
=× ×
_(1)
where CF_PFAS (nanograms of F per gram) is the corresponding
fluorine concentration of a given PFAS, nFis the number of
fluorine atoms on the molecule, MWPFAS is the molecular
weight of the PFAS, and AFis the atomic weight of fluorine.
The total known extractable fluorine concentration
(ΣCF_PFAS, nanograms of F per gram), which is the sum of
all individual CF_PFAS values, can be related to CF_EOF by eq 2:
CCC
FEOF FPFAS Fextr.unknow
n
=Σ +
___ (2)
where CF_extr. unknown (nanograms of F per gram) is the total
concentration of unidentified, extractable organic fluorine.
Lastly, CF_EOF and CF_TF are related to each other via the
total nonextractable fluorine concentration (CF_non extr., nano-
grams of F per gram) according to eq 3:
CC C
FTFFEOFFnon extr
.
=+
__ _
(3)
2.5. Quality Assurance and Quality Control. Accuracy
and precision were assessed through (a) replicate (n=3)
spike−recovery experiments using printer paper fortified with
36 native PFASs (targeted analysis) and (b) comparison of
replicate (n= 8) measurements of the CRM to certified
concentrations (TF analysis). Extraction blanks were processed
in every batch to monitor for background contamination, while
solvent blanks were injected intermittently during UHPLC−
MS/MS and CIC analysis to monitor for carryover. Statistical
analysis was carried out at an α= 0.05 confidence level in all
instances.
3. RESULTS AND DISCUSSION
3.1. Total Fluorine Method Comparison. A comparison
of the measured (n= 8) versus certified concentrations (568 ±
60 μg/g) of CRM BCR-461 revealed no statistically significant
differences for CIC (p= 0.18) or PIGE (p= 0.84) [one-sample
ttests (Figure 1a and Table S3)], indicating good accuracy
for both methods. Precision was also reasonable for both
approaches but slightly better for CIC (2.5% CV) than for
PIGE (8.1%; p= 0.005; F-test). INAA was unable to measure
F in the CRM due to the high Al content. The 27Al captures a
neutron, yielding unstable 28Al, which decays by β-emission
with a characteristic γ-ray at 1779 keV. The high levels of Al in
geological materials result in detector dead times of >90% for
the F analysis. Thus, INAA was deemed unsuitable for this
matrix.
PFOA-fortified filters (six fortification levels, including a
blank, each prepared in triplicate) were measured by CIC,
PIGE, and INAA (Figure 1b and Table S4). The blank filters
were below the LOD for all methods and therefore excluded
from statistical analysis. Concentrations measured by CIC
were, on average, 2.1% higher than those from PIGE and 4.1%
higher than those from INAA, while those from PIGE were
1.9% higher than those from INAA. Repeated ttests with
Bonferroni correction revealed that these differences were not
statistically significant (for individual pvalues, see Table S6).
All methods displayed good linearity (r2> 0.99) and precision
Figure 1. Comparison of TF methods (CIC, PIGE, and INAA) by
means of (a) certified reference material (CRM) measurements (n=
8; circles) and means (dots) (the error bar represents the standard
deviation, and the gray line indicates the certified concentration), (b)
PFOA-spiked filter measurements vs theoretical concentrations (the
gray line indicates slope = 1 and intercept = 0), and (c) food
packaging samples (circles, data points; dots, means). Note the
discrepancy in FF2 by PIGE due to the thickness and heterogeneity of
this paperboard sample. *n.a., not analyzed due to interference.
Environmental Science & Technology Letters Letter
DOI: 10.1021/acs.estlett.8b00700
Environ. Sci. Technol. Lett. XXXX, XXX, XXX−XXX
C

[CV; 3−12% (CIC), 3−39% (PIGE), and 2−6% (INAA)]
over the range of concentrations on the filters.
Nine food packaging materials were analyzed in triplicate by
CIC, PIGE, and INAA (Figure 1c and Table S5). It is
important to note that while measurement replicates were
obtained for PIGE and INAA, the destructive nature of CIC
requires technical replicates to assess precision. Another
difference between methods arises from the fraction of the
sample that contributes to the fluorine signal. For example,
depending on the penetration depth of the PIGE particle
beam, only the surface F content is measured in thick samples;
in contrast, CIC and INAA measure the F content of the entire
sample independent of material thickness. As a result, TF
measurements determined by PIGE will be higher than those
determined by INAA and CIC for surface-coated products
when expressed on a weight basis. This difference is clearly
demonstrated through measurements of FF2, which was the
only thick paperboard material analyzed in this work and was
over-reported by PIGE relative to CIC and INAA. When FF2
was excluded, CIC produced TF concentrations that were, on
average, 2.4% higher than those produced by PIGE and 7.3%
higher than those produced by INAA, while PIGE measure-
ments were an average of 4.9% higher than INAA measure-
ments. These differences were not statistically significant [two-
way analysis of variance with replication (p= 0.39)]. All
methods showed good precision [CV; 2−8% (CIC), 1−8%
(PIGE), and 1−6% (INAA)]. Assuming a 10 mg sample size,
detection limits were lowest for CIC (0.8 μg/g), followed by
INAA (20 μg/g) and PIGE (38 μg/g).
3.2. Fluorine Mass Balance of Food Packaging
Samples. 3.2.1. Target PFAS Analysis. A total of 22 of 44
target PFASs were detected in the food packaging materials
investigated here with Σ44PFAS concentrations ranging from
23.9 to 2220 pg/cm2(Figure 2a and Table S8). Spike−
recovery experiments demonstrated good accuracy and
precision of the targeted PFAS analysis (see Table S7).
PFTeDA, PFDoDA, PFHpA, 6:2 diPAP, PFHxA, 6:2/8:2
diPAP, and 8:2 diPAP were detected in >50% of the samples.
10:2 monoPAP was detected at by far the highest
concentration (2100 pg/cm2) but only in a single sample
(MP5). In all other samples, the contribution of PFAA
precursors was minor. PFCAs were the major compound class
in 6 samples (FF2, FF3, MP2−MP4 and MP6), with PFHxA
(<0.37−160 pg/cm2) accounting for 53−90% of Σ44PFAS
concentrations in these samples. PFOA was detected in only
FF2 and MP5 (4.32 and 22.4 pg/cm2, respectively), possibly
reflecting the shift among industries from C8 to C6 chain
lengths.
3.2.2. Fluorine Mass Balance. CIC-based CF_TF concen-
trations in food packaging [2.05−17.8 μg/cm2(Figure 2b)]
were high relative to the Danish Ministry of the Environment
and Food indicator value of 0.1 μg/cm2. This value was
established as a means of differentiating between intentionally
added and background PFASs in food packaging.
39
The CF_EOF
was greater than the method detection limit (MDL) (0.04−
0.07 μg/cm2) in four samples, but low (0.22−0.49 μg/cm2)
compared to CF_TF (accounting only for 0−5.5%). No
significant correlations were observed among CF_TF,CF_EOF,
and ΣCF_PFAS. According to eq 3,CF_non extr. was high in all
samples, ranging from 94.5 to 99.9%. These findings affirm the
presence of polymeric coatings [e.g., perfluoropolyethers and
fluorotelomer (meth)acrylate-based side-chain fluorinated
polymers] on these papers and paperboards, as polymers
have low solubility in most nonfluorinated solvents.
40
Furthermore, the fractions of CF_TF and CF_EOF explained by
ΣCF_PFAS were negligible in all samples [means of 0.002 and
0.08%, respectively (Table S9)], leaving the majority of TF and
EOF unattributed. Therefore, we assume that our UHPLC−
MS/MS method does not capture the PFASs intentionally
used in these products. Here it is germane to note that possible
degradation products and/or unreacted monomers of the
aforementioned polymeric coatings [e.g., FTOHs, fluoro-
telomer olefins, or fluorotelomer (meth)acrylates] were not
included in our targeted analysis.
Previous reports on TF in food packaging align well with
concentrations reported in our study. Robel et al. measured
concentrations from below the LOD to 8.17 μg/cm2, and
Schaider et al. concentrations of ≤15.2 μg/cm2in U.S. fast
food packaging.
13,29
To the best of our knowledge, no prior
studies have performed direct quantification of EOF in food
packaging. While Robel et al. analyzed TF by PIGE in
packaging samples before and after extraction, thereby
indirectly measuring EOF, the differences were not statistically
significant.
29
This is a general problem arising when comparing
two large numbers, whereby a small difference easily lies within
the error bounds. In our study, that problem is avoided by
direct measurement of EOF. Similar to our results, Robel et al.
report that the sum of ionic PFASs accounted for only 0−
0.03% of the TF.
3.3. Implications. The results of the method comparison
revealed excellent agreement among all three total fluorine
methods. However, technical differences help determine their
Figure 2. Fluorine mass balance in food packaging samples comprising (a) PFAS concentrations displayed as the sum of each class (in picograms
per square centimeter) as measured by UHPLC−MS/MS and (b) total fluorine and extractable organic fluorine contents (error bars represent the
standard deviation of triplicate measurements) in micrograms per square centimeter as measured by CIC. The percentage of EOF identified by the
sum of target PFAS (ΣCF_PFAS) is indicated in red.
Environmental Science & Technology Letters Letter
DOI: 10.1021/acs.estlett.8b00700
Environ. Sci. Technol. Lett. XXXX, XXX, XXX−XXX
D

applicability domains. For example, the rapid and non-
destructive nature of PIGE and INAA allows for quick
screening applications, as for example for regulatory purposes.
CIC on the other hand excels at sensitivity and versatility, with
lower detection limits and the possibility for direct IC analysis
for determination of inorganic fluoride. All three methods can
be used to analyze solid and liquid samples, although
preconcentration methods are used to increase sensitivity for
PIGE, which are not required for INAA and CIC.
In the case of food packaging materials, all three methods
prove to be applicable. Because of the limited penetration
depth of the particle beam, PIGE can distinguish between
coated and uncoated surfaces. Most fluorine was persistent on
the paper and paperboards after methanol extractions, as
determined by comparably low EOF and target PFAS
concentrations. More broadly, the cross-validation of these
three TF methods means that they can be used as a
complement to high-specificity targeted analysis. The mass
balance measurements demonstrated in this work are critical in
fate and transport studies of PFASs in the environment, such as
those that can be found in the end-of-life options for paper
packaging. For example, regardless of whether PFAS-treated
paper decays in a landfill, is composted and used as fertilizer, or
is recycled directly into more paper, these TF methods can be
used to study the environmental release of all PFASs over a
broad range of samples, disposal conditions, and locations.
Such a broad question could not be answered in a timely
manner with compound-specific analysis, yet LC−MS/MS
identification of PFASs will remain an essential complement to
these robust TF methods.
■ASSOCIATED CONTENT
*
SSupporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.es-
tlett.8b00700.
Additional information about food packaging samples,
instrumental methods, target PFAS compounds, quality
assurance and quality control results, pvalues for
statistical analysis, and tabular overviews of CIC,
PIGE, INAA, and LC−MS/MS results (PDF)
■AUTHOR INFORMATION
Corresponding Author
*E-mail: lara.schultes@gmail.com.Telephone:
0046729440914.
ORCID
Lara Schultes: 0000-0002-3409-4389
Graham F. Peaslee: 0000-0001-6311-648X
Notes
The authors declare no competing financial interest.
■ACKNOWLEDGMENTS
The authors thank the Swedish Research Council FORMAS
(Grant 2013-00794) for funding this project. Jan-Olov Persson
is acknowledged for statistical support. Kristina von Dolwitz
(Testfakta) is thanked for providing food packaging materials.
■REFERENCES
(1) KEMI Swedish Chemical Agency. Occurrence and Use of Highly
Fluorinated Substances and Alternatives. 2015.
(2) Buck, R. C.; Franklin, J.; Berger, U.; Conder, J. M.; Cousins, I.
T.; De Voogt, P.; Jensen, A. A.; Kannan, K.; Mabury, S. A.; van
Leeuwen, S. P. J. Perfluoroalkyl and Polyfluoroalkyl Substances in the
Environment: Terminology, Classification, and Origins. Integr.
Environ. Assess. Manage. 2011,7(4), 513−541.
(3) Organisation for Economic Co-operation and Development.
Toward a New Comprehensive Global Database of Per- and
Polyfluoroalkyl Substances (PFASs): Summary Report on Updating
the OECD 2007 List of per- and Polyfluoroalkyl Substances (PFASs).
OECD Environment, Health and Safety Publications Series on Risk
Management No. 39; 2018.
(4) UNEP/POPS/POPRC.9/INF/11. http://chm.pops.int/
Convention/POPsReviewCommittee/LatestMeeting/POPRC9/
POPRC9Documents/tabid/3281/Default.aspx (accessed September
12, 2018).
(5) Trier, X.; Granby, K.; Christensen, J. H. Polyfluorinated
Surfactants (PFS) in Paper and Board Coatings for Food Packaging.
Environ. Sci. Pollut. Res. 2011,18 (7), 1108−1120.
(6) Trier, X.; Nielsen, N. J.; Christensen, J. H. Structural Isomers of
Polyfluorinated Di- and Tri-Alkylated Phosphate Ester Surfactants
Present in Industrial Blends and in Microwave Popcorn Bags. Environ.
Sci. Pollut. Res. 2011,18 (8), 1422−1432.
(7) Posner, S.; Roos, S.; Poulsen, P. B.; Jörundsdottir, H. Ó.;
Gunnlaugsdóttir, H.; Trier, X.; Jensen, A. A.; Katsogiannis, A. A.;
Herzke, D.; Bonefeld-Jörgensen, E. C.; et al. Per- and Polyfluorinated
Substances in the Nordic Countries; 2013.
(8) Liu, X.; Guo, Z.; Folk, E. E.; Roache, N. F. Determination of
Fluorotelomer Alcohols in Selected Consumer Products and
Preliminary Investigation of Their Fate in the Indoor Environment.
Chemosphere 2015,129,81−86.
(9) Wang, Z.; Cousins, I. T.; Scheringer, M.; Hungerbühler, K.
Fluorinated Alternatives to Long-Chain Perfluoroalkyl Carboxylic
Acids (PFCAs), Perfluoroalkane Sulfonic Acids (PFSAs) and Their
Potential Precursors. Environ. Int. 2013,60, 242−248.
(10) Zabaleta, I.; Bizkarguenaga, E.; Bilbao, D.; Etxebarria, N.;
Prieto, A.; Zuloaga, O. Fast and Simple Determination of
Perfluorinated Compounds and Their Potential Precursors in
Different Packaging Materials. Talanta 2016,152, 353−363.
(11) Yuan, G.; Peng, H.; Huang, C.; Hu, J. Ubiquitous Occurrence
of Fluorotelomer Alcohols in Eco-Friendly Paper-Made Food-Contact
Materials and Their Implication for Human Exposure. Environ. Sci.
Technol. 2016,50 (2), 942−950.
(12) Sinclair, E.; Kim, S. K.; Akinleye, H. B.; Kannan, K.
Quantitation of Gas-Phase Perfluoroalkyl Surfactants and Fluoro-
telomer Alcohols Released from Nonstick Cookware and Microwave
Popcorn Bags. Environ. Sci. Technol. 2007,41 (4), 1180−1185.
(13) Schaider, L. A.; Balan, S. A.; Blum, A.; Andrews, D. Q.; Strynar,
M. J.; Dickinson, M. E.; Lunderberg, D. M.; Lang, J. R.; Peaslee, G. F.
Fluorinated Compounds in U.S. Fast Food Packaging. Environ. Sci.
Technol. Lett. 2017,4(3), 105−111.
(14) Begley, T. H.; White, K.; Honigfort, P.; Twaroski, M. L.;
Neches, R.; Walker, R. A. Perfluorochemicals: Potential Sources of
and Migration from Food Packaging. Food Addit. Contam. 2005,22
(10), 1023−1031.
(15) Begley, T. H.; Hsu, W.; Noonan, G.; Diachenko, G. Migration
of Fluorochemical Paper Additives from Food-Contact Paper into
Foods and Food Simulants. Food Addit. Contam., Part A 2008,25 (3),
384−390.
(16) Washington State Legislature. HB 2658 - 2017−18 Concerning
the use of perfluorinated chemicals in food packaging. http://apps.leg.
wa.gov/WAC/default.aspx?cite=388-78A-2180 (accessed December
9, 2018).
(17) Granby, K.; Pedersen, G. A. Vurdering Af Metode Til
Bestemmelse Af Total Organisk Fluor (TOF) i
Fødevarekontaktmaterialer Og Bestemmelse Af Baggrundsniveau for
TOF i Pap Og Papir.
(18) Martin, J. W.; Kannan, K.; Berger, U.; De Voogt, P.; Field, J.;
Franklin, J.; Giesy, J. P.; Harner, T.; Muir, D. C. G.; Scott, B.; et al.
Environmental Science & Technology Letters Letter
DOI: 10.1021/acs.estlett.8b00700
Environ. Sci. Technol. Lett. XXXX, XXX, XXX−XXX
E

Peer Reviewed: Analytical Challenges Hamper Perfluoroalkyl
Research. Environ. Sci. Technol. 2004,38 (13), 248A−255A.
(19) McDonough, C. A.; Guelfo, J. L.; Higgins, C. P. Measuring
Total PFASs in Water: The Tradeoff between Selectivity and
Inclusivity. Current Opinion in Environmental Science &Health 2018,
DOI: 10.1016/j.coesh.2018.08.005.
(20) Miyake, Y.; Yamashita, N.; Rostkowski, P.; So, M. K.; Taniyasu,
S.; Lam, P. K. S.; Kannan, K. Determination of Trace Levels of Total
Fluorine in Water Using Combustion Ion Chromatography for
Fluorine: A Mass Balance Approach to Determine Individual
Perfluorinated Chemicals in Water. J. Chromatogr. A 2007,1143
(1−2), 98−104.
(21) Miyake, Y.; Yamashita, N.; So, M. K.; Rostkowski, P.; Taniyasu,
S.; Lam, P. K. S.; Kannan, K. Trace Analysis of Total Fluorine in
Human Blood Using Combustion Ion Chromatography for Fluorine:
A Mass Balance Approach for the Determination of Known and
Unknown Organofluorine Compounds. J. Chromatogr. A 2007,1154
(1−2), 214−221.
(22) Yeung, L. W. Y.; Miyake, Y.; Li, P.; Taniyasu, S.; Kannan, K.;
Guruge, K. S.; Lam, P. K. S.; Yamashita, N. Comparison of Total
Fluorine, Extractable Organic Fluorine and Perfluorinated Com-
pounds in the Blood of Wild and Pefluorooctanoate (PFOA)-Exposed
Rats: Evidence for the Presence of Other Organofluorine Com-
pounds. Anal. Chim. Acta 2009,635 (1), 108−114.
(23) Yeung, L. W. Y.; Miyake, Y.; Wang, Y.; Taniyasu, S.; Yamashita,
N.; Lam, P. K. S. Total Fluorine, Extractable Organic Fluorine,
Perfluorooctane Sulfonate and Other Related Fluorochemicals in
Liver of Indo-Pacific Humpback Dolphins (Sousa Chinensis) and
Finless Porpoises (Neophocaena Phocaenoides) from South China.
Environ. Pollut. 2009,157 (1), 17−23.
(24) Schultes, L.; Vestergren, R.; Volkova, K.; Westberg, E.;
Jacobson, T.; Benskin, J. P. Per- and Polyfluoroalkyl Substances and
Fluorine Mass Balance in Cosmetic Products from the Swedish
Market: Implications for Environmental Emissions and Human
Exposure. Environ. Sci.: Processes Impacts 2018,20 (12), 1680−1690.
(25) Kiss, A. Z.; Koltay, E.; Nyakó, B.; Somorjai, E.; Anttila, A.;
Räisänen, J. Measurements of Relative Thick Target Yields for PIGE
Analysis on Light Elements in the Proton Energy Interval 2.4−4.2
MeV. J. Radioanal. Nucl. Chem. 1985,89 (1), 123−141.
(26) Tuurnala, T.; Hautojärvi, A.; Harva, K. Nondestructive Analysis
of Paintings by PIXE and PIGE. Stud. Conserv. 1985,30 (2), 93−99.
(27) Roelandts, I.; Robaye, G.; Weber, G.; Delbrouck-Habaru, J. M.
The Application of Proton-Induced Gamma-Ray Emission (PIGE)
Analysis to the Rapid Determination of Fluorine in Geological
Materials. Chem. Geol. 1986,54 (1−2), 35−42.
(28) Ritter, E. E.; Dickinson, M. E.; Harron, J. P.; Lunderberg, D.
M.; DeYoung, P. A.; Robel, A. E.; Field, J. A.; Peaslee, G. F. PIGE as a
Screening Tool for Per- and Polyfluorinated Substances in Papers and
Textiles. Nucl. Instrum. Methods Phys. Res., Sect. B 2017,407,47−54.
(29) Robel, A. E.; Marshall, K.; Dickinson, M.; Lunderberg, D.; Butt,
C.; Peaslee, G.; Stapleton, H. M.; Field, J. A. Closing the Mass
Balance on Fluorine on Papers and Textiles. Environ. Sci. Technol.
2017,51 (16), 9022−9032.
(30) Carioni, V. M. O.; Brockman, J. D.; Morris, M. C.;
Ngwenyama, R. A.; Schell, L. A.; Spate, V. L.; Crane, S. Instrumental
Neutron Activation Analysis, a Technique for Measurement of Se, Hg,
Fe, Zn, K, Mn, Br, and the Hg:Se Ratio in Brain Tissue Samples with
Results from the Memory and Aging Project (MAP). J. Radioanal.
Nucl. Chem. 2018,318,43−48.
(31) Feskanich, D.; Owusu, W.; Hunter, D. J.; Willett, W.; Ascherio,
A.; Spiegelman, D.; Morris, S.; Spate, V. L.; Colditz, G. Use of Toenail
Fluoride Levels and an Indicator for the Risk of Hip and Forearm
Fractures in Women. Epidemiology 1998,9(4), 412−416.
(32) Spate, V. L.; Morris, J. S.; Baskett, C. K.; Mason, M. M.; Reams,
C. L.; Hunter, D.; Willett, W. C. Determination of Fluoride in Human
Nails via Cyclic Instrumental Neutron Activation Analysis. J.
Radioanal. Nucl. Chem. 1994,179 (1), 27−33.
(33) Havránek, V.; Kučera, J.; Řanda, Z.; Voseček, V. Comparison of
Fluorine Determination in Biological and Environmental Samples by
NAA, PAA and PIXE. J. Radioanal. Nucl. Chem. 2004,259 (2), 325−
329.
(34) Azua Jamari, N. L.; Behrens, A.; Raab, A.; Krupp, E. M.;
Feldmann, J. Plasma Processes to Detect Fluorine with ICPMS/MS as
[M-F]+: An Argument for Building a Negative Mode ICPMS/MS. J.
Anal. At. Spectrom. 2018,33 (8), 1304−1309.
(35) Qin, Z.; McNee, D.; Gleisner, H.; Raab, A.; Kyeremeh, K.;
Jaspars, M.; Krupp, E.; Deng, H.; Feldmann, J. Fluorine Speciation
Analysis Using Reverse Phase Liquid Chromatography Coupled Off-
Line to Continuum Source Molecular Absorption Spectrometry (CS-
MAS): Identification and Quantification of Novel Fluorinated
Organic Compounds in Environmental and Biologica. Anal. Chem.
2012,84 (14), 6213−6219.
(36) Tokranov, A. K.; Nishizawa, N.; Amadei, C. A.; Zenobio, J. E.;
Pickard, H. M.; Allen, J. G.; Vecitis, C. D.; Sunderland, E. M. How Do
We Measure Poly- and Perfluoroalkyl Substances (PFASs) at the
Surface of Consumer Products? Environ. Sci. Technol. Lett. 2019,6
(1), 38−43.
(37) Gebbink, W. A.; Ullah, S.; Sandblom, O.; Berger, U.
Polyfluoroalkyl Phosphate Esters and Perfluoroalkyl Carboxylic
Acids in Target Food Samples and Packaging-Method Development
and Screening. Environ. Sci. Pollut. Res. 2013,20 (11), 7949−7958.
(38) Vestergren, R.; Ullah, S.; Cousins, I. T.; Berger, U. A Matrix
Effect-Free Method for Reliable Quantification of Perfluoroalkyl
Carboxylic Acids and Perfluoroalkane Sulfonic Acids at Low Parts per
Trillion Levels in Dietary Samples. J. Chromatogr. A 2012,1237,64−
71.
(39) Fluorinated substances in paper and cardboard food contact
materials (FCM). www.fvst.dk (accessed October 28, 2018).
(40) Rankin, K.; Mabury, S. A. Influence of Fluorination on the
Characterization of Fluorotelomer-Based Acrylate Polymers by
Matrix-Assisted Laser Desorption/Ionization Time-of-Flight Mass
Spectrometry. Anal. Chim. Acta 2014,808, 115−123.
Environmental Science & Technology Letters Letter
DOI: 10.1021/acs.estlett.8b00700
Environ. Sci. Technol. Lett. XXXX, XXX, XXX−XXX
F