DEQ-CFW_00000333

BVIROnMEnTHL
dog9 if TTf �R pubs.acs.org/journal/esticu
Legacy and Emerging Perfluoroalkyl Substances Are Important
Drinking Water Contaminants in the Cape Fear River Watershed of
North Carolina
Mei Sun,*'t'*G Ehsa Arevalo,$ Mark S ar,§ Andrew Lindstrom,§ Michael Richardson," Ben Kearns,"
Adam Pickett,l Chris Smith,# and Detlef R.U. Knappe
tDepartment of Civil and Environmental Engineering, University of North Carolina at Charlotte, Charlotte, North Carolina 28223,
United States
*Department of Civil, Construction, and Environmental Engineering, North Carolina State University, Raleigh, North Carolina
27695, United States
§National Exposure Research Laboratory, U.S. Environmental Protection Agency Research, Triangle Park, North Carolina 27711,
United States
"Cape Fear Public Utility Authority, Wilmington, North Carolina 28403, United States
1Town of Pittsboro, Pittsboro, North Carolina 27312, United States
#Fayetteville Public Works Commission, Fayetteville, North Carolina 28301, United States
© Supporting Information
ABSTRACT: Long -chain per- and polyfluoroalkyl substances point & A leg,:cy PFAS,
(PFASs) are being replaced by short -chain HASS and non -point y A
B
fluorinated alternatives. For ten legacy PFASs and seven sources PFPrOPrA ("GenX")
recently discovered perfluoroalkyl ether carboxylic acids c
(PFECAs), we report (1) their occurrence in the Cape Fear 0 200 400 600 800
River (CFR) watershed, (2) their fate in water treatment F F PFAS Concentration (ng/L)
processes, and (3) their adsorbability on powdered activated F F F FF �B
carbon (PAC). In the headwater region of the CFR basin, I I I O
PFECAs were not detected in raw water of a drinking water F I I OH Fluorochemical
treatment plant (DWTP), but concentrations of legacy PFASs F F F Manufacturer C•
were high. The U.S. Environmental Protection Agency's PFPrOPrA (`GenX")
lifetime health advisory level (70 ng/L) for perfluorooctane-
sulfonic acid and perfluorooctanoic acid (PFOA) was exceeded on 57 of 127 sampling days. In raw water of a DWTP
downstream of a PFAS manufacturer, the mean concentration of perfluoro-2-propoxypropanoic acid (PFPrOPrA), a replacement
for PFOA, was 631 ng/L (n = 37). Six other PFECAs were detected, with three exhibiting chromatographic peak areas up to 15
times that of PFPrOPrA. At this DWTP, PFECA removal by coagulation, ozonation, biofiltration, and disinfection was negligible.
The adsorbability of PFASs on PAC increased with increasing chain length. Replacing one CF2 group with an ether oxygen
decreased the affinity of PFASs for PAC, while replacing additional CF2 groups did not lead to further affinity changes.
■ INTRODUCTION
Per- and polyfluoroalkyl substances (PFASs) are extensively
used in the production of plastics, water/stain repellents,
firefighting foams, and food -contact paper coatings. The
widespread occurrence of PFASs in drinking water sources is
closely related to the presence of sources such as industrial
sites, military fire training areas, civilian_ airports, and waste-
water treatment plants. Until 2000, long -chain perfluoroalkyl
sulfonic acids [C„F2n+1S03H; n >_ 6 (PFSAs)] and perfluoro-
alkyl carboxylic acids [U2n+1COOH; n > 7 (PFCAs)] were
predominantly used? Accumulating evidence about the
ecological persistence and human health effects associated
with exposure to long -chain PFASs3'4 has led to an increased
level of regulatory attention. Recently, the U.S. Environmental
Protection Agency (USEPA) established a lifetime health
® ACS Publications @ XXXX American Chemical Society
advisory level (HAL) of 70 ng/L for the sum of
perfluorooctanoic acid (PFOA) and perfluorooctanesulfonic
acid (PFOS) concentrations in drinking water.5'6 Over the past
decade, production of long -chain PFASs has declined in Europe
and North America, and manufacturer's are moving toward
short -chain PFASs and fluorinated alternatives. 7-11 Some
fluorinated alternatives were recently identified,8,11 but others
remain unknown12-14 because they are either proprietary or
manufacturing byproducts.
Received: October 13, 2016
Revised: November 8, 2016
Accepted: November 10, 2016
Published: November 10, 2016
A kkt, 70N91L pros
i►� qt
DOI: 10.1021 /acs.e stlett.6b00398
Environ. Scl. TechnoL Lett. XXXX, XXX, XXX—XXX
DEQ-CFW 00000333

Environmental Science & Technology Letters
One group of fluorinated alternatives, perfluoroalkyl ether
carboxylic acids (PFECAs), was recently discovered in the Cape
Fear River (CFR) downstream of a PFAS manufacturing
facility.11 Identified PFECAs included perfluoro-2-methoxy-
acetic acid (PFMOAA), perfluoro-3-methoxypropanoic acid
(PFMOPrA), perfluoro-4-methoxybutanoic acid (PFMOBA),
perfluoro-2-propoxypropanoic acid (PFPrOPrA), perfluoro-
(3,5-dioxahexanoic) acid (PF02HxA), perfluoro(3,5,7-trioxa-
octanoic) acid (PF030A), and perfluoro(3,5,7,9-tetraoxadeca-
noic) acid (PF04DA) (Table S1 and Figure S1). The
ammonium salt of PFPrOPrA is a known PFOA alternative 15
that has been produced since 2010 with the trade name
"GenX". To the best of our knowledge, the only other
published PFECA occurrence data are for PFPrOPrA in Europe
and China,15 and no published data about the fate of PFECAs
during water treatment are available. Except for a few studies
(most by the manufacturer),16-20 little is known about the
toxicity, pharmacokinetic behavior, or environmental fate and
transport of PFECAs.
The strong C—F bond makes PFASs refractory to abiotic and
biotic degradation,21 and most water treatment processes are
ineffective for legacy PFAS removal.22-27 Processes capable of
removin PFCAs and PFSAs include nanofiltration,28 reverse
osmosis, 5 tion,2a,z9 ion exchange,zs,z9 and activated carbon adsorp-
with activated carbon adsorption being the most
widely employed treatment option.
The objectives of this research were (1) to identify and
quantify the presence of legacy PFASs and emerging PFECAs
in drinking water sources, (2) to assess PFAS removal by
conventional and advanced processes in a full-scale drinking
water treatment plant (DWTP), and (3) to evaluate the
adsorbability of PFASs on powdered activated carbon (PAC).
■ MATERIALS AND METHODS
Water Samples. Source water of three DWTPs treating
surface water in the CFR watershed was sampled between June
14 and December 2, 2013 (Figure S2). Samples were collected
from the raw water tap at each DWTP daily as either 8 h
composites (DWTP A, 127 samples) or 24 h composites
(DWTP B, 73 samples; DWTP C, 34 samples). Samples were
collected in 250 mL HDPE bottles and picked up (DWTPs A
and B) or shipped overnight (DWTP C) on a weekly basis. All
samples were stored at room temperature until they were
analyzed (within 1 week of receipt). PFAS losses during storage
were negligible on the basis of results of a 70 day holding study
at room temperature. On August 18, 2014, grab samples were
collected at DWTP C after each unit process in the treatment
train [raw water ozonation, coagulation/flocculation/sedimen-
tation, settled water ozonation, biological activated carbon
(BAC) filtration, and disinfection by medium -pressure UV
lamps and free chlorine]. Operational conditions of DWTP C
on the sampling day are listed in Table S2. Samples were
collected in 1 L HDPE bottles and stored at room temperature
until they were analyzed. On the same day, grab samples of
CFR water were collected in six 20 L HDPE carboys at William
0. Huske Lock and Dam downstream of a PFAS manufacturing
site and stored at 4 °C until use in PAC adsorption experiments
(background water matrix characteristics listed in Table S3).
Adsorption Experiments. Adsorption of PFASs by PAC
was studied in batch reactors (amber glass bottles, 0.45 L of
CFR water). PFECA adsorption was studied at ambient
concentrations (-1000 ng/L PFPrOPrA, chromatographic
peak areas of other PFECAs being approximately 10-800%
of the PFPrOPrA area). Legacy PFASs were present at low
concentrations (00 ng/L) and spiked into CFR water at
—1000 ng/L each. Data from spiked and nonspiked experi-
ments showed that the added legacy PFASs and methanol (1
ppm) from the primary stock solution did not affect native
PFECA removal. A thermally activated, wood -based PAC
(PicaHydro MP23, PICA USA, Columbus, OH; mean diameter
of 12 µm, BET surface area of 1460 mZ IS) 30 proven to be
effective for PFAS removal in a prior studyz99 was used at doses
of 30, 60, and 100 mg/L. These doses represent the upper
feasible end for drinking water treatment. Samples were taken
prior to and periodically after PAC addition for PFAS analysis.
PFAS losses in PAC -free blanks were negligible.
PFAS Analysis. Information about analytical standards and
liquid chromatography —tandem niass spectrometry (LC —MS/
MS) methods for PFAS quantification is provided in the
Supporting Information.
■ RESULTS AND DISCUSSION
Occurrence of PFASs in Drinking Water Sources. Mean
PFAS concentrations in source water of three DWTPs treating
surface water from the CPR watershed are shown in Figure 1.
PFBA
■ PFPeA ■ PFHxA ■ PFHpA w PFOA m PFNA
® PFDA
■ PFBS ■ PFHxS ■ PFOS ■ PFPrOPrA
Community A
n=127
Community B
n-73
Community C
n=34
0 200 400 600 800
Average concentration in drinking water source (ng/L)
Figure 1. Occurrence of PFASs at drinking water intakes in the CFR
watershed. Concentrations represent averages of samples collected
between June and December 2013. Individual samples with
concentrations below the quantitation limits (QLs) were considered
as 0 when calculating averages, and average concentrations below the
QLs were not plotted.
In communities A and B, only legacy PFASs were detected
(mean 2]PFAS of 355 ng/L in community A and 62 ng/L in
community B). Detailed concentration data are shown in Table
S6 and Figure S3. In community A, PFCAs with four to eight
total carbons, perflucrohexanesulfonic acid (PFHxS), and
PFOS were detected at mean concentrations above the
quantitation limits (QLs). During the 127 day sampling
campaign, the sum concentration of PFOA and PFOS exceeded
the USEPA HAL of 70 ng/L on 57 days. The mean sum
concentration of PFOA and PFOS over the entire study period
was 90 ng/L, with approximately equal contributions from
PFOS (44 ng/L) and PFOA (46 ng/L). Maximum PFOS and
PFOA concentrations were 346 and 137 ng/L, respectively.
Similar PFOS and PFOA concentrations were observed in the
same area in 2006,31 suggesting that PFAS source(s) upstream
of commurity A have continued negative impacts on drinking
water quality. Also, our data show that legacy PFASs remain as
surface water contaminants of concern even though their
production was recently phased out in the United States. It is
important to note, however, that among the PFCAs that were
measured in both 2006 and 2013 (PFHxA to PFDA), the
PFCA speciation shifted from long -chain (-80-85%
C„F2n+1COOH; n = 7-9) in 2006 to short -chain (76%
CnF2n+1COOH; n = 5-6) in 2013. In contrast, the PFSA
speciation was dominated by PFOS in both 2006 and 2013.
DOI: 10.1021 /acs.estlett.6b00398
Environ. ScL Technol. Lett. XXXX, XXX, XXX—XXX
o 10-7
'P,g`
DEQ-CFW 00000334

Environmental Science & Technology Letters
Relating total PFAS concentration to average daily streamflow
(Figure S4) illustrated a general trend of low PFAS
concentrations at high flow, and high concentrations at low
flow, consistent with the hypothesis of one or more upstream
poin�ces.
In community B, perfluorobutanoic acid (PFBA) and
perfluoropentanoic acid (PFPeA) were most frequently
detected with mean concentrations of 12 and 19 ng/L,
respectively. Mean PFOA and PFOS concentrations were
below the QLs, and the maximum sum concentration of PFOA
and PFOS was 59 ng/L. Lower PFAS concentrations in
community B relative to community A can be explained by the
absence of substantive PFAS sources between the two
communities, dilution by tributaries, and the buffering effect
of Jordan Lake, a large reservoir located between communities
A�anc B
In community C (downstream of a PFAS manufacturing
site), only mean concentrations of PFBA and PFPeA were
above the QLs. The relatively low concentrations of legacy
PFASs in the finished drinking water of community C are
consistent with results from the USEPXs third unregulated
contaminant monitoring rule for this DWTP. 32 However, high
concentrations of PFPrOPrA were detected (up to 4� 500 nQ/
L). The average PFPrOPrA concentration (631 ng/L) was
approximately 8 times the average summed PFCA and PFSA
concentrations (79 ng/L). Other PFECAs had not yet been
identified at the time of analysis. Similar to communities A and
B, the highest PFAS concentrations for community C were also
observed at low flow (Figure S4). Stream flow data were used
in conjunction with PFPrOPrA concentration data to
determine PFPrOPrA mass fluxes at the intake of DWTP C.
Daily PFPrOPrA mass fluxes ranged from 0.6 to 24 kg/day with
a mean of 5.9 kg/day.
Fate of PFASs in Conventional and Advanced Water
Treatment Processes. To investigate whether PFASs can be
removed from impacted source water, samples from DWTP C
were collected at the intake and after each treatment step.
Results in Figure 2 suggest conventional and advanced
treatment processes (coagulation/flocculation/sedimentation,
raw and settled water ozonation, BAC filtration, and
disinfection by medium -pressure UV lamps and free chlorine)
did not remove legacy PFASs, consistent with previous
studies. 2-26 The data further illustrate that no measurable
PFECA removal occurred in this DWTP. Concentrations of
some PFCAs, PFSAs, PFMOPrA, PFPrOPrA, and PFMOAA
may have increased after ozonation, possibly because of the
oxidation of precursor compounds. 25 Disinfection with
medium -pressure UV lamps and free chlorine (located between
the BAC effluent and the finished water) may have decreased
concentrations of PFMOAA, PFMOPrA, PFMOBA, and
PFPrOPrA, but only to a limited extent. Small concentration
changes between treatment processes may also be related to
temporal changes in source water PFAS concentrations that
occurred in the time frame corresponding to the hydraulic
residence time of the DWTP.
Results in Figure 2 further illustrate that the PFAS signature
of the August 2014 samples was similar to the mean PFAS
signature observed during the 2013 sampling campaigns shown
in Figure 1; i.e., PFPrOPrA concentrations (400-500 ng/L)
greatly exceeded legacy PFAS concentrations. Moreover, three
PFECAs (PFMOAA, PF02HxA, and PF030A) exhibited peak
areas 2-113 times greater than that of PFPrOPrA (Figure 2b).
(a) Raw water
Pre -ozone effluent
Settled water
Settled -ozone effluent
BAC effluent
Finished water
0 100 200 300 400 500 600 700 800
Concentration of traditional PFASs
at a WTP in Community C (ng(L)
■ PFPrOPrA PFBA ■ PFPeA a PFHxA ■ PFHpA ■ PFOA
* PFNA to PFDA ■ PFBS ■ PFHS ■ PFOS
(b) Raw water
Pre -ozone effluent
Settled water
Settled -ozone effluent
BAC effluent
Finished water
0
M
0 50,000 100,000 130,000 200,000 250,000 300,001)
Peak area counts of emerging PFASs
at a WTP in Community C
■ PFPrOPrA ® PFMOAA E, PFMOPrA ■ PFMOBA
PF02HXA uPF030A �1`1`04DA
Figure 2. Fate of (a) legacy PFASs and PFPrOPrA and (b) PFECAs
through a full-scale water treatment plant. Because authentic standards
were not available for PFECAs other than PFPrOPrA, chromato-
graphic peak area counts are shown in panel b. PFPrOPrA data are
shown in both panels and highlighted with dashed ovals for reference.
Compounds with concentrations below the QLs were not plotted.
The existence of high levels of emerging PFASs suggests a need
for their incorporation into routine monitoring.
Adsorption of PFASs by PAC. PAC can effectively remove
long -chain PFCAs and PFSAs, but its effectiveness decreases
with decreasing PFAS chain length. 24,25,29 It is unclear,
however, how the presence of ether group(s) in PFECAs
impacts adsorbability. After a contact time of 1 It, a PAC dose
of 100 mg/L achieved >80% removal of legacy PFCAs with
total carbon chain lengths of >7. At the same PAC dose,
removals were 95% for PF04DA and 54% for PF030A, but
<40% for other PFECAs. Detailed removal percentage data as a
function of PAC contact time are shown in Figure S5. There
was no meaningful removal of PFMOBA or PFMOPrA, and the
variability shown in Figure S5 is most likely associated with
analytical variability. PFMOAA could not be quantified by the
analytical method used for these experiments; however, on the
basis of the observations that PFAS adsorption decreases with
decreasing carbon chain length and that PFECAs with one or
two more carbon atoms than PFMOAA (i.e., PFMOPrA and
PFMOBA) exhibited negligible removal (Figure 3), it is
expected that PFMOAA adsorption is also negligible under
the tested conditions.
To compare the affinity of different PFASs for PAC, PFAS
removal percentages were plotted as a function of PFAS chain
length [the sum of carbon (including branched), ether oxygen,
and sulfur atoms] (Figure 3b). The adsorbability of both legacy
and emerging PFASs increased with increasing chain length.
PFSAs were more readily removed than PFCAs of matching
chain length, a result that agrees with those of previous
DOI: 10.1021 /acs.estlett.6b00398
Environ. Sci. Technol. Lett. XXXX, XXX, XXX—XXX
DEQ-CFW 00000335

Environmental Science & Technology Letters
b°
0
W
4
w
a
100%
(a)
80%
60%
40%
20% i r
■ 30 mg/L
■ 60 mg/L
■ 100 mg/L
01/0
p T O0
c a c O O p a
l 1
PFCAs Mono-etherPFECAs Multi-etherPFECAs PFSAs
100%
80%
u6 �1 60%
A 40%
0
N 20%
c 0%
< -20%
(b)
aw 3 5 7 9 11
Chain length
tPFCAs—Mono-ether PFECAs -a-Multi-ether PFECAs-rPFSAs
Figure 3. PFAS adsorption on PAC (a) at carbon doses of 30, 60, and 100 mg/L and (b) as a function of PFAS chain length. The PAC contact time
in CFK water was 1 h. Legacy PFASs were spiked at -1000 ng/L, and the emerging PFASs were at ambient concentrations. Figures show average
PFAS removal percentages, and error bars show one standard deviation of replicate experiments.
studies.24,21,29 PFECAs exhibited adsorbabilities lower than
those of PFCAs of the same chain length (e.g., PFMOBA <
PFHxA), suggesting that the replacement of a CF2 group with
an ether oxygen atom decreases the affinity of PFASs for PAC.
However, the replacement of additional CF2 groups with ether
groups resulted in small or negligible affinity changes among
the studied PFECAs (e.g., PFMOBA — PF02HxA, PFPrOPrA
— PF030A). Alternatively, if only the number of perfluorinated
carbons were considered as a basis of comparing adsorbability,
the interpretation would be different. In that case, with the
same number of perfluorinated carbons, PFCAs have an affinity
for PAC higher than that of monoether PFECAs (e.g., PFPeA >
PFMOBA) but an affinity lower than that of multi -ether
PFECAs (e.g., PFPeA < PF030A).
To the best of our knowledge, this is the first paper reporting
the behavior of recently identified PFECAs in water treatment
processes. We show that PFECAs dominated the PFAS
signature in a drinking water source downstream of a
fluorochemical manufacturer and that PFECA removal by
many conventional and advanced treatment processes was
negligible. Our adsorption data further show that PFPrOPrA
("GenX") is less adsorbable than PFOA, which it is replacing.
Thus, PFPrOPrA presents a greater drinking water treatment
challenge than PFOA does. The detection of potentially high
levels of PFECAs, the continued presence of high levels of
legacy PFASs, and the difficulty of effectively removing legacy
PFASs and PFECAs with many water treatment processes
suggest the need for broader discharge control and contaminant
monitoring.
■ ASSOCIATED CONTENT
Q Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.estlett.6b00398.
Six tables, five figures, information about PFASs,
analytical methods, and detailed results (PDF)
■ AUTHOR INFORMATION
Corresponding Author
*E-mail: [e-mail]. Phone: [phone].
ORCID 0
Mei Sun: 0000.0001-5854-9862
Notes
The views expressed in this article are those of the authors and
do not necessarily represent the views or policies of the
USEPA.
The authors declare no competing financial interest.
■ ACKNOWLEDGMENTS
This research was supported by the National Science
Foundation (Grant 15S0222), the Water Research Foundation
(Project 4344), and the North Carolina Urban Water
Consortium.
■ REFERENCES
(1) Hu, X. C.; Andrews, D. Q; Lindstrom, A. B.; Bruton, T. A.;
Schaider, L. A.; Grandjean, P.; Lohmann, R; Carignan, C. C.; Blum,
A ; Balan, S. A.; et al. Detection of poly- and perfluoroalkyl substances
(PFASs) in U.S. drinkutg water linked to industrial sites, military fare
training areas, and wastewater treatment plants. Environ. Sci. Technol.
Lett. 2016, 3 (10), 344-350.
D DOI: 10.1021/acs.estlett.6b00398
Environ. 5d. Technol. Lett, XXXX, XXX, XXX—XXX
DEQ-CFW 00000336

Environmental Science & Technology Letters
(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) Kennedy, G. L.; Butenhoff, J. L.; Olsen, G. W.; O'Connor, J. C.;
Seacat, A M.; Perkins, R G.; Biegel, L. B.; Murphy, S. R.; Farrar, D. G.
The toxicology of perfluorooctanoate. Crit. Rev. Toxicol. 2004, 34 (4),
351-384.
(4) Borg, D.; Hakansson, H. Environmental and health risk
assessment of perfluoroallrylated and polyfluoroalkylated substances
(PFASs) in Sweden. Report 6513; The Swedish Environmental
Protention Agency: Stockholm, 2012.
(5) Drinking water health advisory for perfluorooctanoic acid
(PFOA). Report 822-R 16-005; U.S. Environmental Protection
Agency: Washington, DC, 2016.
(6) Drinking water health advisory for perfluorooctane sulfonate
(PFOS). Report 822-R-16-004; U.S. Environmental Protection
Agency: Washington, DC, 2016.
(7) Scheringer, M.; Trier, X.; Cousins, I. T.; de Voogt, P.; Pletcher,
T.; Wang, Z.; Webster, T. F. Helsingor Statement on poly- and
perfluorinated alkyl substances (PFASs). Chemosphere 2014, 114,
337-339.
(8) Wang, Z.; Cousins, I. T.; Scheringer, M.; Hungerbiihler, K.
Fluorinated alternatives to long -chain perfluoroalkyl carboxylic acids
(PFCAs), perfluoroalkane sulfonic acids (PPSAs) and their potential
precursors. Environ. Int. 2013, 60, 242-248.
(9) Wang, Z.; Cousins, I. T.; Scheringer, M.; Hungerbuehler, K.
Hazard assessment of fluorinated alternatives to long -chain perfluor-
oalkyl acids (PFAAs) and their precursors: Status quo, ongoing
challenges and possible solutions. Environ. Int. 2015, 75, 172-179.
(10) Barzen-Hanson, K. A.; Field, J. A. Discovery and implications of
C2 and C3 perfluoroalkyl sulfonates in aqueous film -forming foams
and groundwater. Environ. Sci. Technol. Lett. 2015, 2 (4), 95-99.
(11) Strynar, M.; Dagnino, S.; McMahen, R; Lig, S.; Lindstrom,
A.; Andersen, E.; McMillan, L.; Thurman, M.; Ferrer, I.; Ball, C.
Identification of novel perfluoroalkyl ether carboxylic acids (PFECAs)
and sulfonic acids (PPESAs) in natural waters using accurate mass
time -of -flight mass spectrometry (TOPMS). Environ. Sci. Technol
2015, 49 (19), 11622-11630.
(12) 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.
(13) Wagner, A.; Raue, B.; Branch, H.-J.; Worch, E.; Lange, F. T.
Determination of adsorbable organic fluorine from aqueous environ-
mental samples by adsorption to polystyrene-divinylbenzene based
activated carbon and combustion ion chromatography. J. Chromatogr.
A 2013, 1295, 82-89.
(14) Willach, S.; Branch, H. J.; Lange, F. T. Contribution of selected
perfluoroalkyl and polyfluoroalkyl substances to the adsorbable
organically bound fluorine in German rivers and in a highly
contaminated groundwater. Chemosphere 2016, 145, 342-350.
(15) Heydebreck, F.; Tang, J.; Xie, Z.; Ebinghaus, R. Alternative and
legacy perfluoroalkyl substances; Differences between european and
chinese river/estuary systems. Environ. Sci. Technol. 2015, 49 (14),
8386-8395.
(16) Gannon, S. A.; Pasano, W. J.; Mawn, M. P.; Nabb, D. L.; Buck,
R C.; Buxton, L. W.; Jepson, G. W.; Frame, S. R Absorption,
distribution, metabolism, excretion, and kinetics of 2,3,3,3-tetrafluoro-
2-(heptafluoropropoxy)propanoic acid ammonium salt following a
single dose in rat, mouse, and cynomolgus monkey. Toxicology 2016,
340, 1-9.
(17) Hoke, R A.; Ferrell, B. D.; Sloman, T. L.; Buck, R C.; Buxton,
L. W. Aquatic hazard, bioaccumulation and screening risk assessment
for ammonium 2,3,3,3-tetrafluoro-2-(heptafluoropropoxy)-propanoate.
Chemosphere 2616, 149, 336-342.
(18) Caverly Rae, J. M.; Craig, L.; Slone, T. W.; Frame, S. R; Buxton,
L. W.; Kennedy, G. L. Evaluation of chronic toxicity and
carcinogenicity of ammonium 2,3,3,3-tetrafluoro-2-(heptafluoropro-
poxy)-propanoate in Sprague—Dawley rats. Toxicol. Rep. 2015, 2,
939-949.
(19) Wang, J.; Wang, X.; Sheng, N.; Zhou, X.; Cui, R.; Zhang, H.;
Dai, J. RNA -sequencing analysis reveals the hepatotoxic mechanism of
perfluoroalkyl alternatives, HPPO2 and HFPO4, following exposure in
mice. J. Appl. Toxicol. 2016.
(20) Gomis, M. I.; Wang, Z.; Scheringer, M.; Cousins, I. T. A
modeling assessment of the physicochemical properties and environ-
mental fate of emerging and novel per- and polyfluoroalkyl substances.
Sci. Total Environ. 2015, SOS, 981-991.
(21) Rayne, S.; Forest, K. Perfiuoroalkyl sulfonic and carboxylic acids:
A critical review of physicochemical properties, levels and patterns in
waters and wastewaters, and treatment methods. J. Environ. Sci. Health,
Part A: Toxic/Hazard. Subst. Environ. Eng, 2009, 44 (12), 1145-1199.
(22) Quinones, 0.; Snyder, S. A. Occurrence of perfluoroalkyl
carboxylates and sulfonates in drinking water utilities and related
waters from the United States. Environ. Sci. Technol. 2009, 43 (24),
9089-9095.
(23) Shivakoti, B. R; Fujii, S.; Nozoe, M.; Tanaka, S.; Kunacheva, C.
Perfluorinated chemicals (PFCs) in water purification plants (WPPS)
with advanced treatment processes. Water Sci. Technol.: Water Supply
2010, 10 (1), 87-95.
(24) Eschauzier, C.; Beerendonk, E.; Scholte-Veenendaal, P.; De
Voogt, P. Impact of treatment processes on the removal of
perfluoroalkyl acids from the drinking water production chain. Environ.
Sci. Technol. 2012, 46 (3), 1708-1715.
(25) Ralunwi, M. F.; Peldszus, S.; Anderson, W. B. Behaviour and
fate of perfluoroalkyl and polyfluoroalkyl substances (PFASs) in
drinking water treatment: A review. Water Res. 2014, SO, 318-340.
(26) Appleman, T. D.; Higgins, C. P.; Quinones, O.; Vanderford, B.
J.; Kolstad, C.; Zeigler-Holady, J. C.; Dickenson, E. R V. Treatment of
poly- and perfluoroalkyl substances in U.S. full-scale water treatment
systems. Water Res. 2014, 51, 246-255.
(27) Merino, N.; Qu, Y.; Deeb, R, A.; Hawley, E. L.; Hoffmann, M.
R; Mahendra, S. Degradation and removal methods for perfluoroalkyl
and polyfluoroalkyl substances in water. Environ. Eng. Sci. 2016, 33 (9),
615-649.
(28) Appleman, T. D.; Dickenson, E. R. V.; Bellona, C., Higgins, C.
P. Nanofiltration and granular activated carbon treatment of
perfluoroalkyl acids. J. Hazard. Mater. 2013, 260, 740-746.
(29) Dudley, L. A.; Arevalo, E. C.; Knappe, D. R. U. Removal of
p4uoroalkyl substances by PAC adsorption and anion exchange; Water
Research Foundation: Denver, 2015.
(30) Dunn, S. E.; Knappe, D. R. U. Disinfection by-product precursor
and micropollutant removal by powdered activated carbon; Water
Research Foundation: Denver, 2013.
(31) Nakayama, S.; Strynar, M. J.; Helfant, L.; Egeghy, P.; Ye, X.;
Lindstrom, A. B. Perfluorinated compounds in the Cape Fear drainage
basin in North Carolina. Environ. Sci. Technol. 2007, 41 (15), 5271-
5276.
(32) Unregulated contaminant monitoring rule 3 (UCMR 3). http://
water.epa.gov/lawsregs/rulesregs/sdwa/ucmr/ucmr3/; U.S. Environ-
wental Protection Agency: Washington, DC (acceseed July 29, 2016).
D0i: 10.1021 /acs.estlettht,00398
Environ. Sch Technol. Lett. XXXX, XXX, XXX—XXX
DEQ-CFW 00000337

Supporting information
Legacy and Emerging Perfluoroalkyl Substances Are Important
Drinking Water Contaminants in the Cape Fear River Watershed
of North Carolina
Supporting information includes analytical method description, 6 tables, and 5 figures.
Mei Sunl,2, % Elisa Arevalo2, Mark Strynar3, Andrew Lindstrom3, Michael Richardson4, Ben
Kearns4, Adam Pickett5, Chris Smith6, and Detlef R.U. Knappe2
'Department of Civil and Environmental Engineering
University of North Carolina at Charlotte
Charlotte, North Carolina 28223, USA
2 Department of Civil, Construction, and Environmental Engineering
North Carolina State University
Raleigh, North Carolina 27695, USA
3 National Exposure Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711, USA
4 Cape Fear Public Utility Authority
Wilmington, North Carolina 28403, USA
5 Town of Pittsboro
Pittsboro, North Carolina 27312, USA
6 Fayetteville Public Works Commission
Fayetteville, North Carolina 28301, USA
*Corresponding Author Email: [e-mail]; Phone: [phone]
Page 1 of 12
DEQ-CFW 00000338

Analytical standards: PFASs studied in this research are listed in Table S1. For legacy PFASs,
native and isotopically labeled standards were purchased from Wellington Laboratories
(Guelph, Ontario, Canada). Native PFPrOPrA was purchased from Thermo Fisher Scientific
(Waltham, MA). No analytical standards were available for other PFECAs.
PFAS quantification: PFAS concentrations in samples from DWTPs and adsorption tests were
determined by liquid chromatography tandem mass spectrometry (LC-MS/MS) using a large -
volume (0.9 mL) direct injection method. An Agilent 1100 Series LC pump and PE Sciex API
3000 LC-MS/MS system equipped with a 4.6 mm x 50 mm HPLC column (Kinetex C18 5µm
100A, Phenomenex Inc.) was used for PFAS analysis. The eluent gradient is shown in Table S4
in SI. All samples, calibration standards, and quality control samples were spiked with
isotopically labeled internal standards, filtered through 0.45-µm glass microfiber syringe filters,
and analyzed in duplicate. The MS transitions for PFAS analytes and internal standards are
shown in Table S5 in SI. The quantitation limit (QL) was 25 ng/L for PFOS and
perfluorodecanoic acid, and 10 ng/L for other legacy PFASs and PFPrOPrA. The QL was
defined as the first point of the standard curve, for which the regression equation yielded a
calculated value within ±30% error. For PFECAs without analytical standards, chromatographic
peak areas are reported.
PFAS concentrations along the treatment train of DWTP C were analyzed using a Waters
Acquity ultra performance liquid chromatograph interfaced with a Waters Quattro Premier XE
triple quadrupole mass spectrometer (Waters, Milford, MA, USA) after solid phase extraction.
Method details are described elsewhere.' The QL for all PFASs with analytical standards was
0.2 ng/L, and peak areas were recorded for PFECAs without standards.
Page 2 of 12
DEQ-CFW 00000339

Table S1. Perfluoroalkyl substances (PFASs) detected in the Cape Fear River (CFR) watershed
Molecular
Compound. .• -. (including all
carbonsweight
• and S)
Perfluorocarboxylic acids (PFCAs)
Perfluorobutanoic acid (PFBA)
214.0
CaHF702
375-22-4
3
4
Perfluoropentanoic acid (PFPeA)
264.0
C5HF9O2
2706-90-3
4
5
Perfluorohexanoic acid (PFHxA)
314.1
C6HF11O2
307-24-4
5
6
Perfluorohepanoic acid (PFHpA)
364.1
C7HF13O2
375-85-9
6
7
Perfluorooc=anoic acid (PFOA)
414.1
C8HF1502
335-67-1
7
8
Perfluorononanoic acid (PFNA)
464.1
C9HF17O2
375-95-1
8
9
Perfluorodecanoic acid (PFDA)
514.1
CloHF19O2
335-76-2
9
10
Perfluorosulfonic acids (PFSAs)
Perfluorobutane sulfonic acid (PFBS)
300.1
C4HF9SO3
375-73-5
4
5
Perfluorohexane sulfonic acid (PFHxS)
400.1
C6HF13SO3
355-46-4
6
7
Perfluorooctane sulfonic acid (PFOS)
500.1
C8HFi7SO3
1763-23-1
8
9
Perfluoroalkyl ether carboxylic acids with one ether group (mono -ether PFECAs)
Pertluoro-2-methoxyacetic acid (PF1VIOAA)
180.0
C3HF5O3
674-13-5
2
4
Perfluoro-3-methoxypropanoic acid (PFMOPrA)
230.0
C4HF703
377-73-1
3
5
Perfluoro-4-methoxybutanoic acid (PFMOBA)
280.0
C5HF9O3
863090-89-5
4
6
Perfluoro-2-propoxypropanoic acid (PFPrOPrA)
330.1
C6HF11O3
13252-13-6
5
7
Perfluoroalkyl ether carboxylic acids with multiple ether group (multi -ether PFECAs)
Perfluoro(3,5-dioxahexanoic) acid (PFO2HxA)
246.0
CaHF704
39492-88-1
3
6
Perfluoro(3,5,7-trioxaoctanoic) acid (PFO3OA)
312.0
C5HF9O5
39492-89-2
4
8
Perfluoro(3,5,7,9-tetraoxadecanoic) acid (PFO4DA)
378.1
C6HF1106
39492-90-5
5
10
Page 3 of 12

Table S2.Operational conditions of DWTP C on sampling day (August 18, 2014)
Parameter
Raw water ozone dose
Value
3.1 m /L
Raw water total organic carbon concentration
6.0 m /L
Aluminum sulfate coagulant dose
43 m /L
Coagulation pH
5.70
Settled water ozone dose
1.3 m /L
Settled water total organic carbon concentration
1.90 m /L
Empty bed contact time in
biological activated carbon filters
9.4 minutes for granular activated
carbon layer
2.3 minutes for sand layer
Medium pressure UV dose
25 mJ/cm2
Free chlorine dose
1.26 m /L as C12
Free chlorine contact time
L17.2 hours
Table S3. Water quality characteristics of surface water used in adsorption tests
Table S4. LC gradient method for PFAS analysis
Time (min)
Mobile Phase A%, (v/v)
Mobile Phase W/O
Flow Rate (mL/min)
1 1
•�
.
1.
•
•1
•
Mobile phase A: 2 mM ammonium acetate in ultrapure water with 5% methanol
Mobile phase B: 2 mM ammonium acetate in acetonitrile with 5% ultrapure water
Page 4 of 12
DEQ-CFW 00000341

Table S5. MS transitions for PFAS Analysis
Legacy PFASs
Compound
PFBA
MS/MS
Transition
212.8 - 168.8
Internal
standard
13C4-PFBA
PFPeA
262.9 - 218.8
13C2- PFHxA
PFHxA
313.6 -> 268.8
13C2- PFHxA
PFHpA
362.9 - 318.8
13C4- PFOA
PFOA p
413.0 - 368.8
13C4- PFOA
PFNA
463.0 -> 418.8
13C4- PFOA
PFDA
513.1 - 68.8
13C2-PFDA
PFBS
299.1 - 98.8
1802-PFHxS
PFHxS
399.1-> 98.8
1802-PFHxS
PFOS s
498.9 - 98.8
13C4-PFOS
PFECAs
PFMOAA
180.0 - 85.0
N/A
PFMOPrA
229.1 -> 184.9
N/A
PFMOBA
279.0 -> 234.8
N/A
PFPrOPrA
329.0 --+ 284.7
13C2- PFHxA
PFO2HxA
245.1 -> 85.0
N/A
PFO3OA
311. - 84.9
N/A
PFO4DA
377.1 - 85.0
N/A
Internal standards
Perfluoro-n-[1,2,3,4-13C4]butanoic acid
(13C4-PFBA)
217.0 - 172
Not applicable
pp
Perfluoro-n-[1,2-13C2]hexanoic acid
(13C2-PFHxA)
315.1 - 269.8
Perfluoro-n-[1,2,3,4-13C2]octanoic acid
(13C4-PFOA)
417.0 -> 372.0
Perfluoro-n-[1,2-13C2]decanoic acid
(13C2-PFDA)
515.1 - 469.8
Sodium perfluoro-l-
hexane[18O2]sulfonate (1802-PFHxS)
403.1 -> 83.8
Sodium perfluoro-l-[1,2,3,4-13C4]octane
sulfonate (13C4-PFOS)
502.9 -> 79.9
Page 5 of 12
DEQ-CFW 00000342

Table S6. Maximum,
minimum, mean and median concentrations
(ng/L) of PFASs at three drinking
water intakes.
Community
min
median
mean
max
min
median
mean
max
medianmax
min
PFBA
99
<10
26
33
38
<10
12
12
104
<10
12
22
PFPeA
191
14
44
62
38
<10
19
19
116
<10
30
36
PFHxA
318
<10
48
78
42
<10
<10
11
24
<10
<10
<10
PFHpA
324
<10
39
67
85
<10
<10
11
24
<10
<10
<10
PFOA
137
<10
34
46
32
<10
<10
<10
17
<10
<10
<10
PFNA
38
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
PFDA
35
<25
<25
<25
<25
<25
<25
<25
<25
<25
<25
<25
PFBS
80
<10
<10
<10
11
<10
<10
<10
<10
<10
<10
<10
PFHxS
193
<10
10
14
14
<10
<10
<10
14
<10
<10
<10
PFOS
346
<25
29
44
43
<25
<25
<25
40
<25
<25
<25
PFPrOPrA
<10
QO
<10
<10
10
<10
<10
<10
4560
55
304
631
PFOA+PFOS
447
0
64
90
59
0
0
9
55
<10
<10
<10
PFASs**
1502
18
212
355
189
0
47
62
4696
55
345
1710
* Concentrations less than quantitation limits were considered as zero to calculate means and Y, PFASs.
** Other PFECAs were present in water samples from community C but could not be quantified and were therefore not included in
Y, PFASs
Page 6 of 12

F O
O
F
\/ OH
F/ `F
PVNfOA A
F F F O
FO
�
F F F OH
F
F
PFMOBA
F F O
O
F
r F P OH
P
PFiMOPrA
PFPrOPrA
PF02HxA PF030A
F F F F F
F \O�O�O�0 OH
F F F F F
PF04DA O
Figure S1. Molecular structures of PFECAs evaluated in this study
Page 7 of 12
DEQ-CFW 00000344

Cape Fear River watershed
�--
Haw
River
t Comm ' A N
DWTP
-�'IW t--., AA
Deep River
Flow direction
Community B DW T' P
North Carolina4 `~' Cape Fear River
PFAS
Cape Fear river basin manufacturing
plant
ce water sampling site
for PAC test
WA ND
MN me
QR K! SD YYI
WY -IM NY Vr
Ne U' .. W
UT IN QH PANJ`
CA CO Kt MO KY VAOK TN
No
D t
Al NM
�1•^ Mt AL CA
T% ._. . _ ✓
FL
100 km
I i I ly I i i i I
Community C
h �
Figure S2. Sampling sites in the Cape Fear River watershed, North Carolina. The scale is for the
Cape Fear River watershed map.
Page 8 of 12
DEQ-CFW 00000345

1200
RIIIII7
800
rn
c
c
0 600
m
c
c
400
0
U
200
0
PFQt�QtPPF� PFQe QF� QQ�QPPPoPPF�PPF�PPF� QF���PF�QF�PxQPo�
P
071
150
rn
c
0 100
c�
c
0
U
0 50
U
0
•
Community B •
•
QPQt�QtP PF0 ePQe QF,Z, QFQ,QP QFOP PF�P QFOP PF0 QFQ,+S PF� Q �OPxPFo�
P
Page 9 of 12
DEQ-CFW 00000346

5000
4000
J
3000
U 1000
I
P PtOP�PPFg PFPe PF� PF�PP- POP-- &- � - 0 \>,*' FO PFP� 0,0
P � PPOP.
Figure S3. PFAS concentration distributions in the CFR watershed at three drinking water
intakes. Concentrations less than quantitation limits were considered as zero. Upper and lower
edges of a box represent the 751h and 251h percentile, respectively; the middle line represents the
median; upper and lower bars represent the 901h and 10ffi percentile, respectively; and dots
represent outliers (>9011, or <101h percentile).
Page 10 of 12
DEQ-CFW 00000347

2.5 E+07
2.0E+07
1.5E+07
0
c: 1.0E+07
v
5.0E+06
I
Community A
• Mean flow
A PFASs
g
n
nn
n
n�
2500
r 11 0f
J
1500 4
N
1000 U-
w
500
0.0E+00 I ®'--Fo'°` ' 0
6/15/13 7/30/13 9/13/13 10/28/13 12/12/13
5. E+07
4. E+07
0
m
3. E+07
2. E+07
W
00,
1. E+07
a
W
400
J
300
N
200 U-
w
100
0
6/15/13 7/30/13 9/13/13 10/28/13 12/12/13
o.t+ui
5. E+07
4. E+07
0 3.E+07
C
w 2. E+07
1. E+07
0 E+00
Community B
j• Mean flow
a PFASs
d
n
0 a eA
AA a a
�sF 'r
A4 Ift 4^
6/1/13 7/1/13 7/31/13 8/30/13 9/29/13
4000
on
3000
d
LL
r 666
mme
0
Figure S4. Total PFAS concentrations in the source water and stream flow at the three studied
DWTPs. Stream flow data were acquired from US Geological Survey stream gage records
Page 11 of 12
DEQ-CFW 00000348

80%
a
a 600/
40SG
w
°C 20%
0%
•ao%
0 20 40 60 80 100 120 140
time (min)
-1-PFBA - 1-PFPeA -4-PFHxA-+4-PFHpA-3IE-PFOA
-*.PFNA -(>-PFDA - -PFBS -¢ PFHxS -k-PFOS
100% - ..__......._� _.
60%
40%
E
v
a 20%
O% C
-20%
0 20 40 60 80 100 120 140
time (min)
�- PFBA-a-PFPeA ,► PFIIxA-i't-PFHpA-*-PFOA
--o-PFNA-p-PFDA-m-PFBS-0--PFHS -A PFOS
100%�
80•�
e._
601A
40%
20%
0%
-20%
80%
60%
® 400.6
of
°C 20%
0%
-20%
100%
80%
60%
40%
E
w
°C 20%
0%
-20%
awm
80%
A 60%
40%
20%
0%
-20%
b
0 20 40 60 80 100 120 140
time (min)
-PFMOPrA-0-PFMOBA-P-PFPrOPrA
-0-PF02HxA-fl-PF030A-C>-PF04DA
d
0 20 40 GO 80 100 120 140
time (min)
-*-PFMOPrA -41-PFMOSA-1111-PFPrOPrA
-O-PF02HxA-3-PF030A-0-PF04DA
f
0 20 40 60 80 100 120 140 0 20 40 60 80 100 120 140
time (min) time (min)
-*-PFBA-a-PFPeA -A-PFHxA-*-PFHpA-E•PFOA-0-PFMOPrA i-PFMOBA •--PFPrOPrA
- -PFNA->-PFDA -o—PFBS-O-PFHxS-i-PFOS-0-PF02HxA -0-PF030A--o-PF04DA
Figure S5. PFAS adsorption at powdered activated carbon doses of (a, b) 30 mg/L, (c, d) 60 mg/L
and (e, f) 100 mg/L. Figures show average PFAS removal percentages of duplicate tests.
Reference
1. Nakayama, S.; Strynar, M. J.; Helfant, L.; Egeghy, P.; Ye, X.; Lindstrom, A. B.,
Perfluorinated compounds in the Cape Fear drainage basin in North Carolina. Environ. Sci.
Technol. 2007, 41, (15), 5271-5276.
Page 12 of 12
DEQ-CFW 00000349