DEQ-CFW_00002962

is
0
I'
n
Io
0
0
0
N
co
0)
N
Perfluoroalkyl ether carboxylic
acids: Occurrence in the Cape Fear
river watershed and fate in drinking
water treatment processes
Mei Sun, Elisa Arevalo, Leigh -Ann Dudley,
Andrew Lindstrom, Mark Strynar, Detlef Knappe
NC DEQ
Raleigh, June 16, 2017
r J�tOD S Tq r�S
."Wolrz;
PR

Y
t
Perfluoroalkyl acids are organic
compounds in which all C=H bonds are
replaced with C=F bonds.
F
a
F F 0 H
F S—C?
F F O
n
Long -chain PFASs:
PFCAS: CnF2n+lCOOHI n>_7
PFSAS: CnF2n+1S03H7 I7>_6
F F F F F F 0
F, �OH
F F F F F F F F
F F F F F\ F F F
j S03H
F �
F F F F F F F F

0
To protect the public from adverse health
effects, health based guidelines have been
established
EPA Health Advisory
(chronic exposure)
New Jersey
recommended
MCLs
PFOS + C8:
70 ng/L
C8: 14 ng/L
C9: 13 ng/L

At first glance, UCIVIR3 data suggest
low PFAS detection frequency
UCMR3 requires monitoring for six PFASs in US drinking water.
Monitoring began in 2013, and latest data release was January 2017.
PFALS
C7
..
•/L)
10
•
0.64
•
(ng/L)
410
.n Locations'high
concentrations
Saipan, PA, NY, DE, CO
C8
20
1.03
349
PA, MN, Saipan, DE,
WV
C9
20
0.05
56
NJ, DE, PA, MA, NY
PFBS
90
0.05
370
GA, Saipan, CO, AL,
PA
PFHxS
30
0.56
11600
Saipan, AZ, DE, CO,
PA
PFOS
40
0.79
71000
Saipan, DE, CO, PA,
WA
36,972 samples from 4,920 PWSs
PFAS detects: 599 samples (1.6%) from 198 PWSs (4.0%)
Of samples with PFAS detects: 23.4% derived from surface water
Some drinking water samples had PFOA+PFOS levels well above the HAL 4

UCMR3 Data for North Carolina: PFAS detection
frequency higher than for entire US
Compound-
Detects
(ng/L)
Perfluoroheptanoic acid (PFHpA, C7)
10
29 (max. 60
ng/L)
Perfluorooctanoic acid (PFOA, C8)
20
10 (max. 30
ng/L)
Perfluorononanoic acid (PFNA, C9)
20
0
Perfluorobutanesulfonic acid (PFBS)
90
0
Perfluorohexanesulfonic acid (PFHxS)
30
5 (max. 110
ng/L)
Perfluorooctanesulfonic acid (PFOS)
40
8 (max. 90
ng/L)
1,320 samples from 151 PWSs in NC
PFAS detects: 43 samples (3.3%) from 20 PWSs (13.2%)
Of samples with PFAS detects: 79% derived from surface water

Elevated PFAS levels affect a
sizeable number of US residents
Hydrological units with
detectable PFASs
Detected
Not detected
No data
PFOS+PFOA levels
estimated to exceed the
70 ng/L HAL in the
drinking water of
6 million US residents
Hu et al. ES&T Letters (2016)

...but are we
seeing the
complete picture
?

Many PFASs are
used in commerce
Sub -classes of PFASs
PFCSs
(Cri ,ri+l—COOH)
PFSAs
perfluoroalkyl adds (CnF2n+l--SOzH)
(PFAAs) PFPAs
(CnF2n+,--PO3H2;
PFPiAs
(CnF,,,—PO,H-C,„F
Examples of
Individual compounds*
o PFBA(n=4)
o PFPeA (n=5)
u PFHxA (n=6)
v PFHpA (n=7)
n PFOA (r)=B)
o PF-V.A (n=g)
n PFC•,=.(n-lo;.
PFLJM(n=ri)
r, PFDoA(n=r2)
o PFTrA
o PF Te;A (n=i4)
o PFBS (n=4)
o PFHxS i;n=6)
o PFC)S (r,=8)
C, PFDS (n=rc,)
PFBPA (n=4)
PFHxPA (n=6)
o PFiOPA (n=8)
n PFDPA (n-io)
o C41124 PFPiA (n.m=4)
C6tC6 PFPiA (n,m=6)
CSfCS PFPiA (n,m=8)
_6?C3 PMA (n 6 m> S)
Number of peer -reviewed
articles since 2oo2**
F6.-it-C3lFi
PFECA.s & PFESAs
€enx(C,F,-cr(cF,J—COt�H)
(C F_
t r �y
—O—C I —R )
.,+I
=' EEA!,C,F,-O-C_Ft-O-CF_-C.00H;
r e .
F-.3B (C1-i -G'-C,Fy- �t +i
m 2rr,+;
,F,
fvtc BSA (n=4, =N H jH
(n=8.R=N(CH'.)H)
CtFBSA (n=4,R.N(C,H6jH)
PFASs
PASF-based
: rtra;•.;n=a,�=N(e.,fl ,ll)
M,PBSE (n=4Y N(r_H' C,HaQH)
substances
(CnF2n+1— R)
(C n F2n+l— SO2 — R)
, r;•,,•=;; ; in=Fs,R=V(CH IC HyOHj
Et F BSE (n=4.R=N(C,.H,)CIH4OH)
EtFC;SE (n=.3.P=N(C;-i.,}C,HrOHj
> over 3000
D SAm:PAP III ,.'
PFASs may
PFAA
loos Vothers
have been
precursors
4:2 F70H (n=4,R=C1H)
on the global
0 6:2 FTOH (n•6,R»OH)
market
fluorotelomer-based
o S:2 F70:-1 tin=s,R=CH
substances
j io:2 FTOH (n-io,R=OH)
c7 12:2 FTOH ;n=,2,R=OH)
(CnF2n+i—C2H4—R)
v 6:2diiPAP[(C6F,,C,,HQO),-PO=HI
o 3:2 diPAP ';C:,F..C,H 1011-130,14;
loos oothers
o polytetrafluoroelhylene (PTFE)
fluoropolymers
Q polyAnylidenefluoride (PVDF)
others-
„ fluorinated ethylene propylene (FEP)
Wang et al. ES&T (2017)
-' perfluorcalkexyl _alymer(PFA)
Perfiuoropolyethers
(PFPEs)
928
698
1081
1186
4066
1496
1407
1069
1016
426
587
654
1081
3507
340
3
33
31
35
4
12
12
8
4
26
1L
25
134
7
259
24
116
4
146
8
106
375
412
165
42
23
25

Two series o
discovered
X,,Os I-ESI EIC(178.9T3) Scan Frag=80.Uv `;rorklistGa a3.d
O
x'V -ESI EIC(22E 9?�T� Scan Fra^,-?Q G`J'rJorklictCata3 d
1 y It
a5 T II�
f\ f
0
xT0' -E51 EI--[278.97091 Scan Frag-30.G4 1'1orklistData3.d
2
1
x1 4 I-ESI EICs320 HM Scan Frac=BOCi','lorklistCaa3d
0
x10 a -ESI EIC; 379.%45) Scan Frag4O.OV 1'tor$distbata3.d
x1�71 �-E£I E1=4�:E3 `:C13) Scan Frsg-3J Q'i 1:+srkl�atDn7e3 r1
2
Strynar et al. ES&T (2015)
f PFECAs were
in the Cape Fea
F�_"-//
F F
b
F
F
3
F
F OH
F ty40
F
F
F F
F
F
F �
F
F
F
F F
F F F
O
F F '
F HA
6 7 - T-
Counts vs Acquisition Titre (min)
recently
r River
"GenV
F
F F C F
F F
F\ // H
F F x 4t
F F O
F
F F F F
D� �F
F Jam, itl
P! F
9.5 10 10.5 11 11.5 12 12.5 13 135

Two series of PFECAs were recently
discovered in the Cape Fear River
x10
2
1
O
t
EIC(754 9123) Scan Frao-SO OV %-IMkhetD8t86 d
6 580 0
� F F
/�'Of `,O-',©H
F FF F F
•F-9r EIC(672 9289) Scan Frao-SO pV t•hvkh-lD-ta6 d
S 6a_
� 0
a
F F F F
'OH
1 F F F F
t
)
,
a 10 tt ,? 13 is
L cwn1-[-.rv. +gcOurAgnor. Irma ,nvr.
+ F*t FIC(490 9455) ' -n Frao-PA) OV 1Vcwktr.,D-t-, d
S
F F F�
OH
s �\
' F K F
5
C our,1�I-.)vim CicCureif._+n Irma rrun}
-ESI EICI358 %21) 'x a� Fraq-Sa OV `.•:'orkt.etData6 tl
8 1262 O
6 F F
s
2 F�O� OH
6 F F
s
2
10 11 12 13 14
Cour,ta I'•:1 vs ..lcn —h— T— (mrnl
Strynar et al. ES&T (2015)
Molecular Formula: cnH1711Or
Monoisotopic Mass 377.9596 Da
[M•H]•: 376.9525 Da
Molecular Formula C,HF.O.
Monoisotopic Masts 311.9680 Da
[M-H]-: 310.9608 Da
Molecular Formula: C4HF,O,
Monolsotoplc Mass: 245.9763 Da
[M.H]-: 244.9690 Da
Molecular Formula: C,HF,;O,
Monoisotopic Mass: 179.9846 Da
[M-H]-: 178.9773 Do

GenX
F� �F
F F F F F CF3
Commercially produced polymer
processing aid (ammonium salt) to replace
PFOA
• By-product of "vinyl ether process" hexafluoropropylene oxide (REPO) gas
can form a stable dimer (GenX)

Serum Elimination Half -Lives
Table 4
Serum elimination halfwthes (%2) ofGenX ADONA, PFBA,PFHx& and PFOA in male (M) and female (F) rats, mice and humans. In so me cases, halt -Ile isexpressed in the form of"arith-
metic mean ± standard deviation",while in other cases when the standard deviation is less than 15%o fthe arith metic me an on ly the arithmetic mean is p ro%ided. Noteson stud ies on rats
and mice and monkeys provide information on dosing method (e g. single oral done or single intravenous (IV) dose) and dosage fin ppm_ m8 substancerkg bww); notes on studies on
humans provide sample members (n) of humans involved. "-" means no data .available.
Rats. Mice Humans
ti,•: Notes t,;_ Notes t,;%-, Notes ti,.-.l Motes tl; Notes
GenX (F)
<=12 h
Oral' 30 ppm
- - >12 h. <7 d
Orar 3 ppm
- - -
-
GenX (M)
<12 h
- _ >12 h.<7d
- - -
-
ADONA (M)
44 h
5-,. oral
- - -
-
- - 23 ± lid
3'x'n
PFBA (F)
1 h
P41c 30 ppm
2 h Orai` 30 ppm 3 h
Orals 10 ppm
3 h Oralc3O ppm 87 f 31 h
2cm
PFBA(M)
6h
9h 13±5h
16 7h 68#35h
7c�m
PFHxA (F)
0.4 h
r\ d 10 ppm
12 h W 15 ppm <72 h
Gastric 50 ppm
-- - -
-
PFHxA (M)
1 h
2A h <72 h
- - <:28 days
8&m
PFOA (F)
2 h
W 20 ppm
- - 17 days
- - 3.3 years
2im
PFOA (M)
6 days
- - 19 days
- -- 3.8 $ 13 years
24im
PFOA (all)
-
•-
- - -
-
- - 3.26 years
138I.n
PFOA (all)
-
-
- - -
-
- - 2.3 years
200kn
PFOA (all)
-
-
- - -
-
- - 2.9 years
643tn
PFOA (all)
-
-
- - -
-
- - 8.5 years
10294n
a ECHA (2014).b EFSA (2011a).' Chang et al (20081"Chengelis et A (2009)! Oh mo ri et A (2003) f Iwvai (2011) Nilsson eta1. (2010) hLau eta1.(2007)' Olsen et al. (2007)*! Brede et al.
(2010)' Bartell et al. (2010).1 Seals et al. (2011 'Ibese studies foasson samples from people who were occupationally exposed to these substance sa nd the levels in serum were h igh.'Thes e
studies focus on samples from people who were exposed to PFOA maim through highly wntaminate d drinking water.

0
Comparing the potency in vivo of PFAS alternatives and their predecessors
Gomis Ferreira, Melissa Ines; Vestergren, Robin; Borg, Daniel; Cousins, Ian T.
Abstract
Since the year 2000, a number of per- and polyfluoroalkyl substances (PFASs) have
been introduced onto the market to replace long -chain perfluoroalkyl acids (e.g.
perfluoroctane sulfonic acid (PFOS) and perfluorooctanoic acid (PFOA)) and their
respective precursors. The main rationale for this industrial transition is that the
PFAS alternatives are less bioaccumulative and toxic than their predecessors. Here,
we evaluated to what extent differences in toxicological effect thresholds for PFASs,
expressed as an administered dose, were confounded by differences in their
distribution and elimination kinetics. Increased liver weight was selected as the
investigated endpoint based on the availability of sufficient toxicological and
toxicokinetic data to enable a comparison of sub -chronic effects. Converting
administered doses into equivalent serum and liver concentrations significantly
reduced the variability in the dose -response curves for perfluorobutanoic acid
(PFBA), perfluorohexanoic acid (PFHxA), perfluorooctanoic acid (PFOA),
perfluorononaoic acid (PFNA) and ammonium 2,3,3,3-tetrafluoro-2-
(heptafluoropropoxy)-propanoate (GenX). The toxicity ranking using serum
(PFNA>GenX> PFOA> PFHxA> PFBA) and liver (GenX> PFNAN PFOAN PFHxAN PFBA)
concentrations also indicated that some PFAS alternatives may have a higher toxic
potency than their predecessors when correcting for differences in toxicokinetics.
For PFOS and perfluorobutane sulfonic acid (PFBS) the conversion from
administered dose to serum concentration equivalents did not change the toxicity
ranking which, however, could be due to the internal dose of PFBS being too low to
allow a correct comparison. This study illustrates the importance of taking
toxicokinetics/internal dose into account in substitution of hazardous chemicals for
independent evaluation of bioaccumulation and toxicity criteria.

Cl)
DEQ-CFW 00002975

Haw
River
ELI
Cape Fear River Basin
Deep River
Flow direction
Community B
A
Cape Fear River
"E'
I•pA 7•C Ai
— IL
PEAS <' Surface water sampling site
CO f"
K8
NJ vey a�D
�,N �`M VA
manufacturing
for PAC test
AZ NM_ __
I .'. Ms' AL 00,
plant
.Fl.
�?
North Carolina
Comrunity C
Cape Fear river basin
0
I I
25 50
1 I I 1 I I
100 km �' {
I
�.
• Largest
watershed in
NC
• Supplies
~1.5M people
with drinking
water

Sampling Protocol
• Samples collected in 1-L HDPE bottles
• Two sampling approaches
— Daily composite samples of source water at three
drinking water treatment plants
— Grab samples to track PFAS fate in drinking water
treatment plant
• No preservative
• Storage at room temperature
• Analysis within 7 days of sample collection
• 70-day holding study showed no changes in
PFAS concentrations

PFAS Analytical Method
• PFAS concentrations measured by LC-MS/MS
• Large -volume direct injection (900 pL)
• Sample and standard preparation:
— filtration with a 0.45-pm glass fiber filter
— addition of mass -labeled internal standards
— addition of formic acid
• Calibration curves ranged from 10 - 750 ng/L
• Limit of quantitation was 10 ng/L for all PFASs
except C10 and PFOS (25 ng/L)
• Acceptance criterion: QCs within 30%

Compound
MS/MS
Transition
Internal
standard
PFBA
212.8 --* 168.8
13C4-PFBA
PFPeA
262.9 ---).218.8
13C2- PFHxA
PFHxA
313.6 --.),268.8
13C2- PFHxA
PFHpA
362.9 --),318.8
13C4- PFOA
PFOA
413.0 -.368.8
13C4- PFOA
Legacy PFASs
PFNA
463.0 418.8
13C4- PFOA
PFDA
513.1 i 68.8
13C2-PFDA
PFBS
299.1-98.8
1802-PFHxS
PFHxS
399.1-98.8
1802-PFHxS
PFOS
498.9 98.8
13C4-PFOS
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
PFECAs
PF02HxA
245.1 85.0
N/A
PF030A
311. 84.9
N/A
PF04DA
3777.1 --OS5.0
N/A
Perfluoro-n-[1,2,3,4-13C4]butanoic acid
(13C4-PFBA)
217.0 ---).172
Perfluoro-n-[1,2-13C2]hexanoic acid
(13C2-PFHxA)
315.1 269.8
Perfluoro-i-i-[1,2,3,4-13C2]octanoic acid
(13C4,-PFOA)
417.0 372.0
Internal standards
Not applicable
Perfluoro-n-[1,2-13C2]decanoic acid
(13C2-PFDA)
515.1--+ 469.8
Sodium perfluoro-l-
hexane[1802]-,Ulfonate (1802-PFHxS)
403.1 83.8
Sodium perfluorc)-1 -[ 1,2,3,4-'3C4] octane
sulfonate (13C4-PFOS)
502.9 79.9

W
M
1.4
M
1.2
Cn
1.0
_
L
_
0.8
L 0.6
a
m 0.4
0.2
IM
Representative
calibration curves
0 200 400 600
Concentration, ng/L
O
HE
a 1.2
L
= 1.0
Cn
0.8
L
0.6
a
aL 0.4
0 500
Concentration, ng/L

0
rn
p
n
n
0
0
0
0
N
Back -calculated standard
concentrations and QC results
GenX
C4
C5
C6
C7
Sample Name
Mean
Accuracy
Mean
Accuracy
Mean
Accuracy
Mean
Accuracy
Mean
Accuracy
10 ng/L
7.4
73.9
8.3
81.7
10.1
99.5
8.3
80.5
7.6
70.7
25 ng/L
28.5
114.1
27.7
110.5
23.3
91.2
29.7
115.2
29.6
110.4
50 ng/L
56.5
112.9
54.4
108.2
52.2
102.2
54.0
104.6
64.5
120.6
100 ng/L
104.1
104.1
103.1
102.1
112.5
110.3
107.9
104.7
110.6
103.4
250 ng/L
240.0
96.0
242.3
96.6
252.6
99.0
245.6
95.2
260.5
97.2
500 ng/L
488.0
97.6
503.1
100.0
496.0
97.1
517.6
100.1
511.0
95.5
750 ng/L
760.6
101.4
755.8
100.2
773.1
100.9
777.8
100.4
819.9
102.1
r
0.9975
0.9991
0.9927
0.9978
0.9975
Sample Name GenX C4 C5 C6 C7
100QC 98.1 90 101 104 105 111 98 96.9 113 101
500QC 462 441 497 511 526 540 512 478 412 485
C8
C9
C10
PFBS
PFHS
PFOS
Mean
Accuracy
Mean
Accuracy
Mean
Accuracy
Mean
Accuracy
Mean
Accuracy
Mean
Accuracy
8.2
80.3
7.4
73.7
7.8
76.1
7.44
73.63
7.2
71.1
8.7
86.5
28.5
111.7
28.1
111.2
27.8
109.0
28.25
111.66
29.0
115.0
25.5
101.6
56.1
109.8
60.5
119.9
56.7
111.3
57.76
114.16
57.3
113.4
58.1
115.7
105.3
103.2
99.4
98.4
110.1
108.0
108.95
107.88
107.0
105.9
99.9
99.9
243.1
95.3
246.9
97.6
250.7
98.3
237.15
93.74
241.8
96.0
244.3
97.3
511.3
100.1
494.4
97.9
482.1
94.7
497.59
98.34
491.4
97.3
496.5
98.9
769.4
100.5
765.6
101.0
781.6
102.3
768.91
101.31
767.4
101.4
758.8
100.8
0.9987
0.9982
0.998
0.9972
0.9981
0.9985
C8
C9
C10
PFBS
PFHS
PFOS
117
103
109
98.5
110
102
105
105
106
101
98.5
85.8
528
509
415
422
501
434
507
493
463
449
462
426

PFAS Occurrence in the CFR Watershed
PFBA
■ PFOA
■ PFHxS
Community A
Community B
Community C
111
■ PFPeA
PFNA
■ PFOS
■ PFHxA ■ PFHpA
7 PFDA ■ PFBS
■ PFPrOPrA = "GenX"
n=127
F
F
"GenX"
F F ���- F
n=76XFF
OH
F
FIF O
n=351
200
Average concentration in drinking water source (ng/L)
Sun et al. (2016) ES&T Letters

0
No measurable PFAS removal by
conventional and advanced treatment-,
Raw water
Pre -ozone effluent
Settled water
Settled -ozone effluent
BAC effluent
Finished water
Raw TOC: 6.0 mg/L, 03: 3.1 mg/L
M
Settled TOC: 1.9 mg/L, 03: 1.3 mg/L
u
MP UV: 25 mJ/Cm2, FAC: 1.3 mg/L, 17 h
200 400 600 800
PFAS Concentration (ng/L)
C4 IN C5 0 C6 0 C7 C8 C9 C10 0 PFBS 0 PFHS 0 PFOS 0 GenX

Recently discovered perfluoroalkyl ether carboxylic
acids occur at substantially higher concentrations
than traditional PFASs and GenX
Raw water
Pre -ozone effluent
Settled water
Settled -ozone effluent
BAC effluent
Finished water
F
i
F F F O
► L �
► o
► o
F O I F OF
I F
' 0 I 'F PF02HxA
I F
I
F OH
I
F F
PFMOAA
I
I
0 503000 100,000 150,000 200,000 250,000 300,000
Peak area counts of emerging PFASs
at a WTP in Community C
® PFPrOPrA PFMOAA PFMOPrA ■ PFMOBA PF02HxA PF030A m PF04DA
Sun et al. (2016) ES&T Letters

What about
activated carbon?.
PAC: thermally activated, wood -based
PAC Doses: 30, 605 100 mg/L
Contact time: 60 minutes
Water: Cape Fear River (TOC: 9.0 mg/L)
PFECAs: Native levels
PFCAs and PFSAs: Spiked at 1000 ng/L

��
a
Adsorbability of PFASs varies greatly. The PFECAs
that were present at the highest concentrations
were essentially non -adsorbable
100%
■ 30 mg jL
80% _
■ 60 mg/L
60% -
■ 100 mgy'L
40% -
20%
0%
-20%
U, Vi rj
PFCAs Mono -ether PFECAs Multi -ether PFECAs PFSAs
Sun et al. (2016) ES&T Letters

.
1000/0
05
.0
80%
It
60%
40%
a
E
o
200%
V l0
^E
W
20%
-o- PFCAs
3
PFAS adsorbability:
PFSA>PFCA>PFECA
�Y
GenX
Mono -ether PFECAs
Sun et al. (2016) ES&T Letters
am
7 9
Chain length
--♦-Multi-ether PFEC.As
11
y PFSAs

Conclusions
• Legacy PFASs dominant in upstream river reaches
• PFOA+PFOS levels exceeded 70 ng/L in community A on
57 of 127 sampling days
• PFECAs dominated PFAS signature downstream of a
fluorochemical manufacturer
• PFECA concentrations were not attenuated by:
- Conventional treatment
- Ozonation
- Biofiltration
- Disinfection by medium pressure UV lamps and free chlorine
• Activated carbon adsorption only effective for long -chain
PFASs

Acknowledgments
• National Science Foundation (Award #1550222)
• North Carolina Urban Water Consortium
• Adam Pickett, Chris Smith, Michael Richardson,
Ben Kearns at participating utilities
J� QED 5 Tq'f.
I _ " =\
r 4P�rq[ ano�E�sg

DEQ-CFW 00002990

Land application of biosolids in watershed of a NC drinking water reservoir
1< 43 2,010
ell
LFOS = ND
PFOA = ND,:
54ever
knimal
ctuary Ni"KS I;d
PFOS = 65 ng/L
4e PFOA = 109 ng/L
Cane Creek
0 Baptist Church
Oairyla'16'
�roFelI P(_1
CLaarry Rdca
PFOS 500 ng/L
PFOA ­9 qgiL
PFOS = 720 ng/C'
PFOA = 1020 ng/L
0
Data from A. Lindstrom, USEPA, RTF
C)
C)
C)
N) 7, CD
en 01