HomeMy WebLinkAbout2021.04.06_CCO.p11_AnalysisOfBranched&LinearIsomers
The Chemours Company Fayetteville Works 22828 NC Highway 87 W Fayetteville, NC 28306
ANALYSIS OF THE BRANCHED AND LINEAR ISOMERS PMPA/PFMOPrA AND
PEPA/PFMOBA IN STANDARDS AND FIELD SAMPLES
Prepared by
The Chemours Company FC, LLC
Fayetteville Works
22828 NC Highway 87 W
Fayetteville, NC 28306
April 6, 2021
The Chemours Company Fayetteville Works 22828 NC Highway 87 W Fayetteville, NC 28306
ii 6-Apr-2021
TABLE OF CONTENTS
1 INTRODUCTION ................................................................................................ 1
2 CHROMATOGRAPHY AND MASS SPECTROMETRY ................................. 4
2.1 Standards ...................................................................................................... 4
2.2 Chromatography .......................................................................................... 4
2.3 Mass Transitions .......................................................................................... 4
3 ANALYSIS OF STANDARDS ............................................................................ 6
3.1 Determination of Purity of Standards .......................................................... 6
3.2 Analysis of Standards by Method 533 ......................................................... 7
4 ANALYSIS OF GROUNDWATER AND SURFACE WATER SAMPLES ..... 8
5 ANALYSIS OF DRINKING WATER SAMPLES ............................................ 10
6 SUMMARY AND CONCLUSIONS ................................................................. 11
7 REFERENCES ................................................................................................... 13
The Chemours Company Fayetteville Works 22828 NC Highway 87 W Fayetteville, NC 28306
iii 6-Apr-2021
LIST OF TABLES
Table 1: Mass Transitions
Table 2: Purity of Standards
Table 3: Results of Analysis of PMPA and PEPA Standards by EPA Method 533 at TestAmerica and
Lancaster
Table 4: Results of Analysis of PMPA Standards at Chemours
Table 5: Groundwater and Surface Water Sample Information
Table 6: Results of Analysis of Groundwater and Surface Water Samples by EPA Method 533
Table 7: Results of Analysis of Drinking Water Samples by EPA Method 537.1 Mod
LIST OF FIGURES
Figure 1: Structures of PMPA and PFMOPrA
Figure 2: Structures of PEPA and PFMOBA
Figure 3: Chromatographic Separation of Isomer Pairs
Figure 4: Mass Transitions for PMPA
Figure 5: Mass Transition for PFMOPrA
Figure 6: Mass Transitions for PEPA
Figure 7: Mass Transition for PFMOBA
Figure 8: Analysis of a Standard Containing Only PMPA
Figure 9: Analysis of a Standard Containing Only PFMOPrA
Figure 10: Analysis of a Standard Containing Only PEPA
Figure 11: Analysis of a Standard Containing Only PFMOBA
The Chemours Company Fayetteville Works 22828 NC Highway 87 W Fayetteville, NC 28306
iv 6-Apr-2021
LIST OF ATTACHMENTS
Attachment A: Letter from Chemours to DEQ proposing additional work analyzing PMPA and PEPA
standards and five archived field samples by Method 533 to look for the presence of the
linear isomers (August 26, 2020).
Attachment B: Letter from DEQ to Chemours with requests for experiments to further investigate the
presence of branched and linear isomers in standards and field samples (October 2,
2020).
The Chemours Company Fayetteville Works 22828 NC Highway 87 W Fayetteville, NC 28306
v 6-Apr-2021
ACRONYMS AND ABBREVIATIONS
CASN Chemical Abstract Services Number
Chemours The Chemours Company FC, LLC
Eurofins Eurofins Lancaster Laboratories Environmental (Lancaster, PA)
GEL GEL Laboratories (Charleston, SC)
min minutes
Method 533 EPA Method 533
Method 537.1M modified EPA Method 537.1
MRM multiple reaction monitoring
MS/MS tandem mass spectrometry
m/z mass-to-charge ratio
NCDEQ North Carolina Department of Environmental Quality
ng/L nanograms per liter
PEPA perfluoro-2-ethoxypropionic acid
PFAS per- and polyfluorinated alkyl substances
PFMOBA perfluoro-4-methoxybutanoic acid
PFMOPrA perfluoromethoxypropionic acid
PMPA perfluoro-2-methoxypropionic acid
T3 Method the Table 3 analytical method
T6 Method the Table 6 analytical method
TestAmerica Eurofins TestAmerica-Sacramento (Sacramento, CA)
TIC total ion count
USEPA United States Environmental Protection Agency
The Chemours Company Fayetteville Works 22828 NC Highway 87 W Fayetteville, NC 28306
1 6-Apr-2021
1 INTRODUCTION
This report has been prepared by The Chemours Company FC, LLC (Chemours) to summarize information
regarding the identification of the individual compounds in the isomer pairs PMPA/PFMOPrA (molecular
formula C4HF7O3) and PEPA/PFMOBA (molecular formula C5HF9O3). Each isomer pair represents one
branched and one linear arrangement of the constituent atoms (Figures 1 and 2). Production pathway
chemistries at the Chemours Fayetteville Works, North Carolina (the Facility) are based on branched isomer
chemistry and are unlikely to generate the linear isomers PFMOPrA and PFMOBA (Chemours, 2020).
However, PFMOPrA and PFMOBA were identified as potentially present in Cape Fear River water via
nontargeted analysis conducted by USEPA (Strynar et al, 2015; McCord and Strynar, 2019). Pending the
resolution of the identification of the isomers present (branched or linear) in environmental samples, the
Consent Order listed both isomers in each pair in the same row in its lists identifying specific per- and
polyfluorinated alkyl substances (PFAS) requiring action (see Attachments B and C of the Consent Order
entered on February 25, 2019).
Attachment B of the Consent Order identifies the compounds to be included in toxicity studies. An asterisk
applied to the PMPA/PFMOPrA and PEPA/PFMOBA isomer pairs states, “For clarification, compounds
identified with two common names in Attachment B or C shall be tested using a single CASN, to be
proposed by Chemours and approved by DEQ.” Attachment C of the Consent Order identifies the
compounds to be included under “PFAS” for the purposes of Paragraphs 19 (Provision of Public Water
Supplies of Whole Building Filtration Systems), 20 (Provision of Reverse Osmosis Drinking Water
Systems), 21 (Private Well Testing) and 24 (Drinking Water Compliance Plan) of the Consent Order.
Chemours’ interpretation of the asterisk, that there is uncertainty as to whether it is the linear or the branched
isomers which are present, as well as its understanding of the branched isomer production pathways at the
Facility has resulted in the branched isomers only being analyzed in samples taken to satisfy Consent Order
Paragraphs 19, 20, 21 and 24.
The branched isomers (PMPA and PEPA) have been incorporated into the Table 3 analytical method (T3
Method) as performed by Chemours and by two commercial analytical laboratories (Eurofins TestAmerica-
Sacramento (TestAmerica) and Eurofins Lancaster Laboratories Environmental (Eurofins)) on
environmental samples related to the Facility. PFMOPrA and PFMOBA have not been incorporated into
the T3 Method, nor into an alternative analytical method, the Table 6 analytical method (T6 Method), which
is designed to achieve lower reporting limits for PMPA and PEPA than can be achieved by the T3 Method.
The North Carolina Department of Environmental Quality (NCDEQ) and their commercial laboratory, GEL
Laboratories (GEL), conducts analysis of environmental samples related to the Facility by a modified EPA
Method 537.1 (Method 537.1M). These results report PFMOPrA and PFMOBA and do not report PMPA
and PEPA. Chemours has produced standards for PMPA and PEPA and has provided the standards to
NCDEQ and to commercial analytical laboratories, including TestAmerica, Eurofins and GEL. Chemours
has not produced PFMOPrA and PFMOBA standards; however, these standards are commercially available
(e.g., from SynQuest Laboratories (Alachua, FL) and Wellington Laboratories (Guelph, Canada)).
In a letter dated July 1, 2020, NCDEQ noted that Chemours’ contract laboratories have concluded that only
PMPA and PEPA, and not PFMOPrA and PFMOBA, are present in environmental samples. However,
The Chemours Company Fayetteville Works 22828 NC Highway 87 W Fayetteville, NC 28306
2 6-Apr-2021
given that a newly available method, EPA Method 533 (Method 533), included PFMOPrA and PFMOBA
as target analytes, NCDEQ requested that additional work be done to show that PFMOPrA and PFMOBA
are not present in environmental samples.
Chemours evaluated compound identification, purity and concentration on the PMPA and PEPA standards
that were produced by Chemours (Chemours, 2020). Additionally, laboratory tests conducted by
TestAmerica assessed the ability of the T3 and T6 Methods to distinguish PMPA from PFMOPrA and
PEPA from PFMOBA. TestAmerica also reviewed the results of a specific field sample in which PMPA
and PEPA had been detected to assess the potential that PFMOPrA or PFMOBA were present. Results
indicated that:
• the PMPA and PEPA standards prepared by Chemours were properly identified and were of high
purity;
• both the lack of peak broadening or shoulder peaks for PMPA and PEPA and the lack of spectral
response characteristic of PFMOPrA and PFMOBA under the T6 Method conditions indicate that
Chemours’s PMPA and PEPA standards consist primarily or entirely of the branched isomer;
• both the T3 and the T6 Methods are likely to be able to resolve the PMPA/PFMOPrA and
PEPA/PFMOBA isomer pairs when chromatographic and spectral data are considered together;
and
• review of a Site sample containing PMPA and PEPA and analyzed by the T6 Method shows
PFMOPrA and PFMOBA are unlikely to be present, as indicated by the lack of peak broadening
or peak shoulders. This is consistent with Chemours’s assessment that production pathway
chemistries at the Facility are not likely to generate PFMOPrA and PFMOBA.
Chemours and NCDEQ held a technical call on August 10, 2020, after which Chemours submitted a letter
proposing additional work analyzing PMPA and PEPA standards and five archived field samples by
Method 533 to look for the presence of the linear isomers (Attachment A). This proposal was discussed in
a technical call between Chemours and NCDEQ on September 22, 2020. NCDEQ followed up with a letter
dated October 2, 2020 with requests for experiments to further investigate the presence of branched and
linear isomers in standards and field samples (Attachment B).
NCDEQ requested the use of improved analytical conditions (chromatography and mass spectrometry) to
assess the presence of linear isomers in the branched standards and linear isomers in environmental samples.
Specifically, NCDEQ requested:
• The use of four separate solutions for the four analytical standards rather than mixtures.
• The use of sufficient concentrations of standards for strong peaks (minimum instrument response
of 106) to determine any impurities in each of the four standards.
• The optimization of the mass spectrometer settings for the linear isomers when analyzing for them.
• The chromatographic resolution (separation) of each isomer pair, and the monitoring of multiple
MRM transitions for each peak to determine individual isomer presence/absence.
The Chemours Company Fayetteville Works 22828 NC Highway 87 W Fayetteville, NC 28306
3 6-Apr-2021
Work to address these requests was conducted by Chemours, TestAmerica, Eurofins-Lancaster (Lancaster)
and the United States Environmental Protection Agency (USEPA).
The remainder of this report consists of:
• Section 2: Chromatography and Mass Spectrometry
• Section 3: Analysis of Standards
• Section 4: Analysis of Groundwater and Surface Water Samples
• Section 5: Analysis of Drinking Water Samples
• Section 6: Summary and Conclusions
• Section 7: References
The Chemours Company Fayetteville Works 22828 NC Highway 87 W Fayetteville, NC 28306
4 6-Apr-2021
2 CHROMATOGRAPHY AND MASS SPECTROMETRY
2.1 Standards
The branched isomers (PMPA and PEPA) were synthesized by Chemours and provided to USEPA (and to
TestAmerica and Lancaster). The linear isomers (PFMOPrA and PFMOBA) were purchased from
SynQuest. Standard solutions containing 250,000 nanograms per liter (ng/L) of each component were
prepared.
2.2 Chromatography
The branched isomers (PMPA and PEPA) are typically analyzed by the T3 Method, and PFMOPrA and
PFMOBA are not on the standard T3 Method analyte list. Previous work has shown that, when PFMOPrA
and PFMOBA are present, they are not well-resolved chromatographically from their respective branched
isomers during analysis by the T3 Method (Chemours, 2020).
The linear isomers (PFMOPrA and PFMOBA) are typically analyzed by Method 533, which is a newly
developed method for the analysis of per- and polyfluorinated alkyl substances (PFAS) in drinking water
samples. PMPA and PEPA are not on the standard Method 533 analyte list. When PMPA is present, it is
not fully resolved chromatographically from PFMOPrA during analysis by Method 533 at TestAmerica and
it is fully resolved from PFMOPrA at Lancaster (personal communication, TestAmerica and Lancaster).
Chromatographic conditions to resolve PMPA/PFMOPrA and PEPA/PFMOBA were developed by Dr.
Mark Strynar of the USEPA. Conditions were:
• isothermal run at 50°C;
• isocratic hold 75:25 water:methanol (1 minutes (min));
• gradient ramp to 60:40 water:methanol (3 min); and
• gradient to 90% methanol to clean column; then back to isocratic equilibration (75:25).
The column used in the liquid chromatograph was a 50 mm Acquity BEH column. The mass spectral
conditions were -4.5 kv, with a source temperature of 150°C and a dryer gas temperature of 350°C.
Under these conditions, analysis of a solution containing all four compounds at approximately 250,000
ng/L indicated that the individual analytes in the isomer pairs were well-resolved, with sharp, non-
overlapping peaks (Figure 3).
2.3 Mass Transitions
The solution containing all four compounds was analyzed under the chromatographic conditions described
above, but using tandem mass spectrometry (MS/MS) to conduct multiple reaction monitoring (MRM) to
determine which mass transitions are unique for each compound. Results are summarized in Table 1 and
described below. This work was conducted by Dr. Mark Strynar of the USEPA.
The Chemours Company Fayetteville Works 22828 NC Highway 87 W Fayetteville, NC 28306
5 6-Apr-2021
PMPA and PFMOPrA
Results show that PMPA decarboxylated (i.e., loses a CO2 moiety) in the mass spectrometer’s ionization
source. PRMOPrA, however, did not decarboxylate in the ionization source. Therefore, the transition of
229→185 (these numbers represent the mass-to-charge ratio (m/z) of the detected ion), which represents
the molecular mass of the precursor (parent) compound transitioning to the molecular mass of the
decarboxylated precursor, is unique for PMPA, and is the major mass transition for PMPA.
For PMPA, further fragmentation that occurs in the mass spectrometer’s collision cell will result in mass
transitions from the 185 m/z product (daughter) ion that are unique to PMPA (because the 185 m/z product
ion is unique to PMPA). The mass transitions observed were 185→119, 185→85 and 185→69 (Figure 4).
For PFMOPrA, the only mass transition observed was 229→85 (Figure 5). However, this mass transition
was also observed for PMPA, although this was a much smaller peak for PMPA than for PFMOPrA. Likely
this is because not all the PMPA was decarboxylated in the ionization source, meaning some precursor
PMPA entered the collision cell and was fragmented to 85. Consequently, there is no unique mass transition
for PFMOPrA.
PEPA and PFMOBA
Results show that PEPA, like PMPA, decarboxylated readily in the ionization source. PFMOBA, however,
did not decarboxylate in the ionization source. Therefore, the transition of 279→235, which represents the
molecular mass of the precursor (parent) compound transitioning to the molecular mass of the
decarboxylated precursor, is unique for PEPA.
For PEPA, further fragmentation that occurs in the mass spectrometer’s collision cell will result in mass
transitions from the 235 m/z product (daughter) ion that are unique to PEPA (because the 235 m/z product
ion is unique to PEPA). The mass transitions observed were 235→135, 235→119 and 235→69 (Figure 6).
For PFMOBA, the only mass transition observed was 279→85 (Figure 7). This mass transition was not
observed for PEPA, likely because all the PEPA was decarboxylated in the ionization source.
Consequently, the 279→85 mass transition is unique to PFMOBA.
The Chemours Company Fayetteville Works 22828 NC Highway 87 W Fayetteville, NC 28306
6 6-Apr-2021
3 ANALYSIS OF STANDARDS
3.1 Determination of Purity of Standards
Standard solutions containing only a single compound were prepared at a concentration of approximately
250,000 ng/L, separated under the chromatographic conditions described in Section 2.2 and detected using
an MS/MS. When analyzing the PMPA and PFMOPrA standards, the mass transitions 229→185, 229→85,
185→119, 185→85 and 185→69 were monitored. When analyzing the PEPA and PFMOBA standards, the
mass transitions 279→235, 279→85, 235→135, 235→119 and 235→69 were monitored. Results are
shown in Table 2 and in Figures 8 through 11 and described below. Work described in this section was
conducted by Dr. Mark Strynar of the USEPA.
PMPA (Chemours standard)
Analysis of the standard solution containing only PMPA showed that the PMPA peak represented 99.16%
of the total ion count (TIC). The instrument response for PMPA at the concentration analyzed (250,000
ng/L) was 53,035 area counts (5.3x104). Two other peaks were present and were identified as PEPA and
PFMOBA, which are likely impurities in the PMPA standard. These peaks represented 0.59% and 0.24%
of the TIC, respectively (Figure 8).
There is no evidence of the linear isomer (PFMOPrA) in the Chemours PMPA standard.
PEPA (Chemours standard)
Analysis of the standard solution containing only PEPA showed that the PEPA peak represented 97.93%
of the TIC. The instrument response for PEPA at the concentration analyzed (250,000 ng/L) was 32,163
area counts (3.2x104). Two other peaks were present and were identified as PMPA and PFMOPrA, which
are likely impurities in the PEPA standard. These peaks represented 0.96% and 1.11% of the TIC,
respectively (Figure 10).
There is no evidence of the linear isomer (PFMOBA) in the Chemours PEPA standard.
PFMOPrA (SynQuest standard)
Analysis of the standard solution containing only PFMOPrA showed that the PFMOPrA peak represented
97.72% of the TIC. The instrument response for PFMOPrA at the concentration analyzed (250,000 ng/L)
was 43,369 area counts (4.3x104). One other peak was present and was identified as PMPA, which is likely
an impurity in the PFMOPrA standard. This peak represented 2.28% of the TIC (Figure 9).
PFMOBA (SynQuest standard)
Analysis of the standard solution containing only PFMOBA showed that the PFMOBA peak represented
93.30% of the TIC. The instrument response for PFMOBA at the concentration analyzed (250,000 ng/L)
was 14,465 area counts (1.4x104). Three other peaks were present and were identified as PMPA, PFMOPrA
and PEPA, which are likely impurities in the PFMOBA standard. These peaks represented 1.42%, 1.84%
and 3.44% of the TIC, respectively (Figure 11).
The Chemours Company Fayetteville Works 22828 NC Highway 87 W Fayetteville, NC 28306
7 6-Apr-2021
3.2 Analysis of Standards by Method 533
Work described in this section was conducted by TestAmerica, Lancaster and Chemours.
Dual-component standards containing PMPA and PEPA only at concentrations of 50, 1,000 and 20,000
ng/L were prepared at TestAmerica and at Lancaster. (Note: the 50 ng/L solution at TestAmerica was
created from a multi-component intermediate solution due a spiking error in the preparation of the double-
prepared solution at 50 ng/L.)
Single-component standards containing PMPA only at concentrations ranging from 50 to 100,000 ng/L
were prepared at Chemours.
Analysis of Branched Isomer Standards for the Presence of PFMOPrA
At TestAmerica, the dual-component standards were analyzed for the presence of PFMOPrA by Method
533 (Table 3). A small amount of what appeared to be PFMOPrA was reported in the 1,000 and 20,000
ng/L standards. Because the PMPA standard has already been shown to contain no PFMOPrA (Section
3.1) and PMPA and PFMOPrA are not completely chromatographically resolved by Method 533 at
TestAmerica, this result is a false positive from monitoring the 229→85 transition, for which PMPA will
generate a signal. The ratio of PMPA to “PFMOPrA” in the Method 533 analysis at TestAmerica was
between 590:1 and 290:1, with an average ratio of 440:1.
At Lancaster, the dual-component standards were analyzed for the presence of PFMOPrA by Method 533
(Table 3). PFMOPrA was not detected (with a reporting limit of 2.0 ng/L). This is because the Method
533 conditions at Lancaster result in chromatographic resolution of PMPA and PFMOPrA; therefore, the
229→85 transition common to PMPA and PFMOPrA is resolved.
At Chemours, the single-component standards were analyzed for PMPA and PFMOPrA under
chromatographic and spectral conditions that closely followed Method 533, with the addition that both
transitions that can be attributed to PMPA (229→185 and 229→85) were monitored. (The 229→185
transition is not monitored under standard Method 533 conditions.) Each standard was analyzed 3 times,
and the average area for PMPA and PFMOPrA for the 3 analyses was calculated. Results show that there
is a concentration-independent ratio of PMPA to PFMOPrA of about 7.4:1 (Table 4). As described above,
this result is likely a false positive from monitoring the 229→85 transition, which is common to both PMPA
and PFMOPrA, at a chromatographic retention time where PMPA and PFMOPrA coelute. These
experiments demonstrate that for standard solutions containing PMPA only, one can potentially reach an
incorrect conclusion about the presence of PFMOPrA. Additionally, any detections of PFMOPrA can be
corrected using this PMPA:PFMOPrA ratio and the PMPA results (from the 229→185 transition) to find
the true concentration of PFMOPrA.
Analysis of Branched Isomer Standards for the Presence of PFMOBA At both TestAmerica and Lancaster, the dual-component standards were analyzed for the presence of
PFMOBA by Method 533 (Table 3). PFMOBA was not detected by either laboratory (to a reporting limit
of 2.0 ng/L).
The Chemours Company Fayetteville Works 22828 NC Highway 87 W Fayetteville, NC 28306
8 6-Apr-2021
4 ANALYSIS OF GROUNDWATER AND SURFACE WATER SAMPLES
Five groundwater and surface water samples from the Facility, or its vicinity, with known concentrations
of PMPA and PEPA (Table 5) were analyzed for the presence of PFMOPrA and PFMOBA by Method 533
(Table 6). The Method 533 analyses were conducted by TestAmerica and Lancaster. Concentrations of
PMPA and PEPA in these 5 samples had been previously determined by the T3 Method.
PMPA and PFMOPrA
PMPA concentrations in the groundwater and surface water samples ranged from 66 to 6,800 ng/L.
At TestAmerica, the perceived PFMOPrA concentrations ranged from nondetect (with a reporting limit of
2.0 ng/L) to 36 ng/L The ratio of PMPA to PFMOPrA was roughly on the order of 200:1, which is slightly
higher than the ratio in the standard (in which the ratio averaged 400:1), indicating that there is potentially
a very small true presence of PFMOPrA in the environmental samples. The PFMOPrA results were
corrected for the false positive resulting from the presence of PMPA using the average ratio of
PMPA:PFMOPrA of 440:1 determined for the TestAmerica analytical procedure in Section 3.2. As an
example, at PW-06 the PMPA concentration was 1,600 ng/L and the uncorrected PFMOPrA result was 7.4
ng/L. The corrected PFMOPrA concentration was 7.4 - (1,600/440) = 3.8 ng/L.
At Lancaster, the corresponding PFMOPrA concentrations ranged from nondetect (with a reporting limit
of 2.0 ng/L) to 25 ng/L. Since PFMOPrA was not detected in the standard, there is potentially a very small
true presence of PFMOPrA in the groundwater and surface water samples, in the range of 200 to 530 times
less than the concentration of PMPA.
The corrected PFMOPrA concentrations measured by TestAmerica are very close to the PFMOPrA
concentrations measured by Lancaster.
PEPA and PFMOBA
PEPA concentrations in the groundwater and surface water samples ranged from 25 to 2,400 ng/L.
At TestAmerica, the corresponding PFMOBA concentrations ranged from nondetect (with a reporting limit
of 2.0 ng/L) to 5.2 ng/L. Since PFMOBA was not detected in the standard, there is potentially a very small
true presence of PFMOBA in the groundwater and surface water samples, likely around 500 times less than
the concentration of PEPA.
At Lancaster, the corresponding PFMOBA concentrations ranged from nondetect (with a reporting limit of
2.0 ng/L) to 2.3 ng/L. Since PFMOBA was not detected in the standard, there is potentially a very small
true presence of PFMOBA in the groundwater and surface water samples, likely around 1,000 times less
than the concentration of PEPA.
The Chemours Company Fayetteville Works 22828 NC Highway 87 W Fayetteville, NC 28306
9 6-Apr-2021
The PFMOBA concentrations measured by TestAmerica are approximately twice the PFMOBA
concentrations measured by Lancaster.
The Chemours Company Fayetteville Works 22828 NC Highway 87 W Fayetteville, NC 28306
10 6-Apr-2021
5 ANALYSIS OF DRINKING WATER SAMPLES
In January 2021, GEL began reporting all four isomers (PMPA, PEPA, PFMOPrA and PFMOBA) in
drinking water samples from residences in the vicinity of the Facility via EPA Method 537.1 Mod. Results
for 34 samples are provided in Table 7.
For the PMPA/PFMOPrA pair, PMPA was detected in 31 of 34 samples (91%), with a concentration range
of 0.324 to 1,080 ng/L, and PFMOPrA was detected in 7 of 34 samples (21%), with a concentration range
of 0.306 to 2.78 ng/L. The ratio of PMPA to PFMOPrA when both compounds were detected ranges from
190:1 to 1470:1. This range of ratio PMPA to PFMOPrA is similar to the ratio seen at TestAmerica for
false positive PFMOPrA detection (466 to 1). It is also similar to the range seen in groundwater and surface
water samples from TestAmerica (corrected for false positive) and Lancaster (range of 190:1 to 530:1).
Without additional evidence, the degree to which GEL’s results are influenced by false positives cannot be
determined; however, given the PFMOPrA detects in the groundwater and surface water samples, the
PFMOPrA detects in the drinking water samples are potentially real.
For the PEPA/PFMOBA pair, PEPA was detected in 26 of 34 samples (77%), with a concentration range
of 0.391 to 355 ng/L, and PFMOBA was detected in 1 of 34 samples (3%), at a concentration of 0.466 ng/L.
The ratio of PMPA to PFMOBA when both compounds were detected was 762:1. This is within the range
seen in groundwater and surface water samples from TestAmerica and Lancaster (range of 460:1 to 1000:1).
The Chemours Company Fayetteville Works 22828 NC Highway 87 W Fayetteville, NC 28306
11 6-Apr-2021
6 SUMMARY AND CONCLUSIONS
Work conducted for this report is summarized below:
• Specialized chromatographic conditions were developed to clearly separate the branched and linear
isomer pairs PMPA/PFMOPrA and PEPA/PFMOBA into well-resolved, sharp, non-overlapping
peaks using a standard containing all 4 compounds. Mass transitions for each of the 4 compounds
were evaluated; all compounds had at least one unique mass transition with the exception of
PFMOPrA. Under these specialized chromatographic conditions, therefore, PMPA, PEPA and
PFMOBA can be uniquely identified by retention time (i.e., chromatographically) and by mass
transition (i.e., spectrally). PFMOPrA can be uniquely identified by retention time;
• Single-component standards at a concentration of 250,000 ng/L were analyzed under the above
chromatographic and mass spectral conditions. Chemours’ PMPA and PEPA standards were shown
to be 99.16% and 97.93% pure, respectively, and to have no detectable concentration of the
corresponding linear isomer (PFMOPrA and PFMOBA, respectively). SynQuest’s PFMOPrA and
PFMOBA standards were shown to be 97.72% and 93.30% pure, respectively, and to contain 2.28%
and 3.44% of the corresponding branched isomer (PMPA and PEPA, respectively). The instrument
response for the 4 compounds was strong, ranging from 1.4x104 (PFMOBA) to 5.3x104 (PMPA);
• Dual-component standards containing PMPA and PEPA only at concentrations of 50, 1,000 and
20,000 ng/L were prepared at TestAmerica and at Lancaster and analyzed by Method 533 for the
presence of PFMOPrA and PFMOBA. A small false positive signal for PFMOPrA was detected by
TestAmerica due to incomplete chromatographic separation of PMPA and PFMOPrA; the 229→85
transition monitored during Method 533 is common to PMPA and PMOPrA. The ratio of true PMPA
to false PFMOPrA was approximately 440:1. No false positive for PFMOPrA was detected at
Lancaster because their chromatographic conditions result in better separation of PMPA and
PFMOPrA during analysis by Method 533. No PFMOBA was detected at either TestAmerica or
Lancaster during analysis of the PEPA standards;
• A series of standard solutions containing PMPA only was analyzed at Chemours under conditions
approximating EPA Method 533, with the mass transition of 229→185 added to quantify PMPA. A
false positive signal for PFMOPrA was detected. The ratio of true PMPA to false PFMOPrA was
approximately 7.4:1. This ratio is different than the ratio observed at TestAmerica, likely because the
analytical instrumentation is different between the two laboratories;
• Groundwater and surface water samples with known PMPA and PEPA concentrations (previously
determined by the T3 Method) were analyzed by Method 533 for the presence of PFMOPrA and
PFMOBA. Results indicated that there may be a very small true presence of PFMOPrA and
PFMOBA in the environmental samples. Concentrations of PFMOPrA measured by TestAmerica
and Lancaster were very similar (if the TestAmerica results were corrected for the false positive signal
from PMPA in the samples). Concentrations of PFMOBA measured by TestAmerica were
approximately 2 times higher than those measured by Lancaster; and
• Drinking water samples from residences in the vicinity of the Facility were analyzed by GEL for all
four isomers using Method 537.1 Mod. The linear isomers were detected at ratios of 190:1 to 1470:1
(PMPA to PFMOPrA) and 762:1 (PEPA to PFMOBA).
The Chemours Company Fayetteville Works 22828 NC Highway 87 W Fayetteville, NC 28306
12 6-Apr-2021
The linear isomers PFMOPrA and PFMOBA were identified as potentially present in Cape Fear River water
via nontargeted analysis conducted by USEPA (Strynar et al, 2015; McCord and Strynar, 2019). (Note:
Concentrations were not determined by the nontargeted analysis. Given the branched isomer nature of the
production pathways at the Facility, Chemours questioned the presence of the linear compounds.
Consequently, pending the resolution of the identification of the isomers present in the environmental
samples, the Consent Order listed both isomers in each pair in the same row in its lists identifying specific
per- and polyfluorinated alkyl substances (PFAS) requiring action (see Attachments B and C of the Consent
Order). Implied was that resolution of the identification of the isomers present, linear or branched, would
mean that the other isomers would not be significant.
There is much evidence for the presence of the branched isomers, PMPA and PEPA, in environmental
samples related to the Facility. The work provided in this report shows that the linear isomers PFMOPrA
and PFMOBA are also potentially present in some groundwater, surface water and drinking water samples
associated with the Facility. However, the concentration of the linear isomers is 2 to 3 orders of magnitude
lower than the corresponding branched isomer. (Note: The non-targeted analysis conducted by Strynar is
by definition not capable of determining concentration; therefore, these are the first concentration data
collected for PFMOPrA and PFMOBA.) Consequently, the linear isomers PFMOPrA and PFMOBA are
of much lower significance than the branched isomers PMPA and PEPA. Chemours may, therefore,
continue to analyze only for the branched isomers PMPA and PEPA in environmental samples related to
the Facility.
The Chemours Company Fayetteville Works 22828 NC Highway 87 W Fayetteville, NC 28306
13 6-Apr-2021
7 REFERENCES
Chemours, 2020. Evaluation of the Analysis of the Branched and Linear Isomers PMPA/PFMOPrA and
PEPA/PFMOBA. July 17, 2020.
McCord, J. and M. Strynar, 2019. Identification of Per- and Polyfluoroalkyl Substances in the Cape Fear
River by High Resolution Mass Spectrometry and NonTargeted Screening. Environ. Sci. Tech., 53
(9) pp. 4717-4727.
Strynar, M., S. Dagnino, R. McMahen, S. Liang, A. Lindstrom, E. Andersen, L. McMillan, I. Ferrer and
C. Ball, 2015. Identification of Novel Perfluoroalkyl Ether Carboxylic Acids (PFECAs) and Sulfonic
Acids (PFESAs) in Natural Waters Using Accurate Mass Time-of-Flight Mass Spectrometry
(TOFMS). Environ. Sci. Tech., 49 (19), pp. 11622-11630.
TABLES
TABLE 1MASS TRANSITIONS Chemours Fayetteville Works, North CarolinaPMPA 229 → 185 229 → 85 185 → 119 185 → 85 185 → 69PFMOPrA -- 229 → 85 -- -- --PEPA 279 → 235 -- 235 → 135 235 → 119 235 → 69PFMOBA -- 279 → 85 -- -- --Notes:--- mass transition is not observedm/z- mass to charge ratioPFMOBA - perfluoro-4-methoxybutanoic acidPFMOPrA - perfluoromethoxypropionic acidPEPA - perfluoro-2-ethoxypropionic acidPMPA - perfluoro-2-methoxypropionic acid- major transition for PMPA- minor transition for PMPAAnalyteMass Transitions (m/z)Precursor Ion Transitions Product Ion TransitionsApril 2021
TABLE 2 PURITY OF STANDARDSChemours Fayetteville Works, North CarolinaPMPA PFMOPrA PEPA PFMOBAPMPA Chemours 99.16% 0.00% 0.59% 0.24%PEPA Chemours 0.96% 1.11% 97.93% 0.00%PFMOPrA SynQuest 2.28% 97.72% 0.00% 0.00%PFMOBA SynQuest 1.42% 1.84% 3.44% 93.30%Notes:--- not applicableChemours- The Chemours Company FC, LLC PFMOBA - perfluoro-4-methoxybutanoic acidPFMOPrA - perfluoromethoxypropionic acidPEPA - perfluoro-2-ethoxypropionic acidPMPA - perfluoro-2-methoxypropionic acidSynQuest - SynQuest Laboratories (Alachua, FL)% - percent% of Total Ion CountStandard SupplierApril 2021
TABLE 3 RESULTS OF ANALYSIS OF PMPA AND PEPA STANDARDS BY EPA METHOD 533 AT TESTAMERICA AND LANCASTERChemours Fayetteville Works, North CarolinaPFMOPrA(ng/L)Ratio ofPMPA to PFMOPrAPFMOPrA(ng/L)Ratio ofPMPA to PFMOPrAPMPA 50 ng/L < 2.0 > 25:1 < 2.0 > 25:11,000 ng/L 3.5 290:1 < 2.0 > 500:120,000 ng/L 34 590:1 < 2.0 > 10,000:1Average Ratio = 440:1Compound ConcentrationPFMOBA(ng/L)Ratio ofPEPA to PFMOBAPFMOBA(ng/L)Ratio ofPEPA to PFMOBAPEPA 50 ng/L < 2.0 > 25:1 < 2.0 > 25:11,000 ng/L < 2.0 > 500:1 < 2.0 > 500:120,000 ng/L < 2.0 > 10,000:1 < 2.0 > 10,000:1Notes:Analysis of PFMOPrA and PFMOBA was by EPA Method 533Lancaster - Eurofins Lancaster Laboratories (Lancaster, PA)ng/L - nanograms per literPFMOBA - perfluoro-4-methoxybutanoic acidPFMOPrA - perfluoromethoxypropionic acidPEPA - perfluoro-2-ethoxypropionic acidPMPA - perfluoro-2-methoxypropionic acidTestAmerica - Eurofins TestAmerica (Sacramento, CA)< - less than associated value> - greater than associated valueTestAmerica LancasterCompound ConcentrationApril 2021
TABLE 4 RESULTS OF ANALYSIS OF PMPA STANDARDS AT CHEMOURSChemours Fayetteville Works, North Carolina50 100 250 500 1,000 5,000 10,000 25,000 50,000 100,000PMPAArea of 1st analysis 89.18 202.66 488.43 937.58 1565.57 8265.26 17457.26 44700.20 89985.54 177467.00Area of 2nd analysis 103.40 178.31 505.07 884.77 1721.38 8470.33 16393.65 45577.77 88289.41 179122.20Area of 3rd analysis 140.89 231.95 462.98 902.57 1717.33 8362.25 17543.34 44889.03 89470.79 176852.40Average of 3 analyses 111.16 204.31 485.49 908.31 1668.09 8365.95 17131.42 45055.67 89248.58 177813.90PFMOPrAArea of 1st analysis 21.09 29.00 76.44 107.13 227.53 1230.05 2296.96 6141.74 12143.26 23948.82Area of 2nd analysis 20.50 26.28 56.88 113.58 220.41 1191.29 2340.08 6281.24 11900.82 24009.33Area of 3rd analysis 13.49 53.02 65.12 119.15 227.85 1121.70 2327.67 6110.88 12213.04 24097.97Average of 3 analyses 18.36 36.10 66.15 113.29 225.26 1181.01 2321.57 6177.95 12085.71 24018.71Ratio of Average Areas 6.1 5.7 7.3 8.0 7.4 7.1 7.4 7.3 7.4 7.4Average Ratio of PMPA:PFMOPrA*Notes:Analysis of of PMPA and PFMOPrA was under conditions approximating EPA Method 533, with the mass transition of 229→185 added to quantify PMPAng/L - nanograms per literPFMOPrA - perfluoromethoxypropionic acidPMPA - perfluoro-2-methoxypropionic acid* - average ratio of PMPA:PFMOPrA excludes the results from 50 and 100 ng/L as they had poor accuracy with respect to their expected concentrations Concentration of Compound in Single-Component PMPA Standard (ng/L)7.4April 2021
TABLE 5 GROUNDWATER AND SURFACE WATER SAMPLE INFORMATIONChemours Fayetteville Works, North CarolinaLocation Sample ID Sample Date NotesOutfall 002 66 25 O00210 2/10/2020 On-site outfall. Grab sampleBladen-1D 450 120 CAP2Q20-BLADEN-1D-050620 5/6/2020 Off-site well in Black Creek Aquifer. Grab samplePW-06 1,600 580 CAP1Q20-PW-06-020620 6/2/2020 On-site well in perched water zone. Grab sampleOld Outfall 4,900 1,400 CAP1Q20-OLDOF-1-24-040320 3/4/2020 On-site outfall. Composite sampleLTW 6,800 2,400 CAP1Q20-LTW-02-022420 2/24/2020 On-site well in Black Creek Aquifer. Grab sampleNotes:PMPA and PEPA were analyzed by the Table 3+ Methodng/L - nanograms per literPEPA - perfluoro-2-ethoxypropionic acidPMPA - perfluoro-2-methoxypropionic acidSample InformationPMPA(ng/L)PEPA(ng/L)April 2021
TABLE 6 RESULTS OF ANALYSIS OF GROUNDWATER AND SURFACE WATER SAMPLES BY EPA METHOD 533Chemours Fayetteville Works, North CarolinaPFMOPrA(ng/L)Ratio ofPMPA to PFMOPrAPFMOPrA(ng/L)Ratio ofPMPA to PFMOPrATestAmerica Outfall 002 66 < 2.0-- --> 33:1 25 < 2.0 > 12:1Bladen-1D 450 < 2.0-- --> 230:1 120 < 2.0 > 60:1PW-06 1,600 7.4--3.8 430:1 580 < 2.0 > 290:1Old Outfall 4,900 37--26 190:1 1,400 2.9 480:1LTW 6,800 36--21 330:1 2,400 5.2 460:1Lancaster Outfall 002 66 < 1.6 > 41:1-- --25 < 1.6 > 16:1Bladen-1D 450 < 1.7 > 270:1-- --120 < 1.7 > 71:1PW-06 1,600 3.0 530:1-- --580 < 1.8 > 320:1Old Outfall 4,900 25 200:1-- --1,400 < 1.9 > 740:1LTW 6,800 23 300:1-- --2,400 2.3 1,000:1Notes:Analysis of PMPA and PEPA was by the Table 3+ MethodAnalysis of PFMOPrA and PFMOBA was by EPA Method 533Lancaster - Eurofins Lancaster Laboratories (Lancaster, PA)ng/L - nanograms per literPFMOBA - perfluoro-4-methoxybutanoic acidPFMOPrA - perfluoromethoxypropionic acidPEPA - perfluoro-2-ethoxypropionic acidPMPA - perfluoro-2-methoxypropionic acidTestAmerica - Eurofins TestAmerica (Sacramento, CA)< - less than associated value> - greater than associated value* - the PFMOPrA results from TestAmerica were corrected for the false positive resulting from the presence of PMPA using the average ratio of PMPA:PFMOPrA of 440: 1 (see Table 3). For example, at PW-06 the PMPA concentration was 1,600 ng/L and the uncorrected PFMOPrA result was 7.4 ng/L. The corrected PFMOPrA concentration was 7.4 - (1,600/440) = 3.8 ng/LUncorrectedPMPA / PFMOPrA PEPA / PFMOBALaboratorySample LocationPMPA(ng/L)Corrected for False Positive*PEPA(ng/L)PFMOBA(ng/L)Ratio ofPEPA to PFMOBAApril 2021
TABLE 7
RESULTS OF ANALYSIS OF DRINKING WATER SAMPLES BY EPA METHOD 537.1 MOD
Chemours Fayetteville Works, North Carolina
PMPA PFMOPrA Ratio PEPA PFMOBA Ratio
0072-W1-092320 i 207 0.306 676 39.4 <0.201 >196
0094-W1-091620 i 350 0.323 1084 72.2 <0.192 >376
0147-W1-120920-RAW i 226 0.460 491 59.9 <0.189 >317
0161-W1-120920 i 9.66 <0.176 >55 0.653 <0.176 >4
0178-W1-100720 i 92.1 <0.184 >501 14.0 <0.184 >76
0345-W1-091620 i 51.0 <0.207 >246 1.69 <0.207 >8
0347-W1-092320 i 15.9 <0.183 >87 1.08 <0.183 >6
0348-W1-092320 i 460 0.475 968 111 <0.174 >638
0350-W1-100720 i 44.4 <0.187 >237 1.99 <0.187 >11
0351-W1-100720 i 20.1 <0.200 >101 1.91 <0.200 >10
0353-W2-100720 i 5.07 <0.182 >28 0.730 <0.182 >4
0354-W1-100720 i 49.3 <0.198 >249 5.71 <0.198 >29
0355-W1-102120 i 30.5 <0.187 >163 8.42 <0.187 >45
0357-W1-102120-RAW i 1,080 2.78 388 355 0.466 762
0358-W1-102120-RAW i 969 2.36 411 301 <0.181 >1663
0362-W1-110420 i 1.62 <0.186 >9 <0.186 <0.186 --
0363-W1-110420 i 20.7 <0.186 >111 <0.186 <0.186 --
0365-W1-120920 i 18.8 <0.180 >104 2.08 <0.180 >12
0368-W1-121720-TR i 10.5 <0.179 >59 0.415 <0.179 >2
0371-W1-121720 i 14.0 <0.178 >79 1.58 <0.178 >9
0365-W1-011321 16.8 <0.178 >94 1.90 <0.178 >11
0372-W1-011321 <0.183 <0.183 -- <0.183 <0.183 --
0372-W2-011321 150 <0.174 >862 18.8 <0.174 >108
0373-W1-011321 1.04 <0.184 >6 0.391 <0.184 >2
0373-W2-011321 3.23 <0.180 >18 1.10 <0.180 >6
0374-W1-011321 0.324 <0.176 >2 <0.176 <0.176 --
0375-W1-011321 17.1 <0.187 >91 1.98 <0.187 >11
0095-W1-021021 <0.182 <0.182 -- <0.182 <0.182 --
0301-W1-021021 18.3 <0.177 >103 <0.177 <0.177 --
0384-W1-021021 2.02 <0.189 >11 <0.189 <0.189 --
0385-W1-021021 62.6 0.329 190 15.8 <0.187 >84
0386-W1-021021 12.3 <0.185 >66 0.566 <0.185 >3
0387-W1-021021 <0.179 <0.179 -- <0.179 <0.179 --
0388-W1-021021 540.0 0.368 1470 167 <0.185 >903
Notes:
PFMOBA - perfluoro-4-methoxybutanoic acid
PFMOPrA - perfluoromethoxypropionic acid
PEPA - perfluoro-2-ethoxypropionic acid
PMPA - perfluoro-2-methoxypropionic acid
< - not detected; associated value is the detection limit
PMPA/PFMOPrA PEPA/PFMOBA
Field Sample ID
April 2021
FIGURES
Notes:PFMOPrA - perfluoromethoxypropionic acidPMPA - perfluoro-2-methoxypropionic acidStructures of PMPA and PFMOPrAChemours Fayetteville Works, North CarolinaFigure1April 2021
Notes:PEPA - perfluoro-2-ethoxypropionic acidPFMOBA - perfluoro-4-methoxybutanoic acidStructures of PEPA and PFMOBA Chemours Fayetteville Works, North CarolinaFigure2April 2021
Notes:PEPA - perfluoro-2-ethoxypropionic acidPFMOPrA - perfluoromethoxypropionic acidPFMOBA - perfluoro-4-methoxybutanoic acidPMPA - perfluoro-2-methoxypropionic acidChromatographic Separation of Isomer PairsChemours Fayetteville Works, North CarolinaFigure3April 2021PMPAPEPAPFMOPrAPFMOBAIsocratic hold 75:25 water:methanol (1 min)Then gradient ramp to 60:40 water:methanol (3 min)Gradient to 90% methanol to clean column; then back to isocratic equilibration (75:25).50 mm column Acquity BEH column at 50°CMS conditions:‐4.5kvSource 150°C, dryer gas 350°CStandards: Made individual and mixed0.25 ng/ul solution 75:25 water/methanol5 ul injection for ~1.25 ng on column
Notes:PMPA - perfluoro-2-methoxypropionic acidPFMOPrA - perfluoromethoxypropionic acidMass Transitions for PMPAChemours Fayetteville Works, North CarolinaFigure4April 2021Mass TransiƟon 229→85PMPA only respondsPMPAPMPAPMPAPMPAMass TransiƟon 185→119PMPA only responds Mass TransiƟon 185→85PMPA only respondsMass TransiƟon 185→69PMPA only responds
Notes:PMPA - perfluoro-2-methoxypropionic acidPFMOPrA - perfluoromethoxypropionic acidMass Transition for PFMOPrAChemours Fayetteville Works, North CarolinaFigure5April 2021Mass TransiƟon 229→85PMPA and PFMOPrA both respondPMPAPFMOPrA
Notes:PEPA - perfluoro-2-ethoxypropionic acidPFMOBA - perfluoro-4-methoxybutanoic acidMass Transitions for PEPAChemours Fayetteville Works, North CarolinaFigure6April 2021PEPAMass TransiƟon 235→135PEPA only respondsMass TransiƟon 235→119PEPA only respondsMass TransiƟon 235→69PEPA only respondsPEPAPEPAPEPAMass TransiƟon 279→235PEPA only responds
Notes:PEPA - perfluoro-2-ethoxypropionic acidPFMOBA - perfluoro-4-methoxybutanoic acidMass Transition for PFMOBAChemours Fayetteville Works, North CarolinaFigure7April 2021Mass TransiƟon 279→85PFMOBA only respondsPFMOBA
Notes:PEPA - perfluoro-2-ethoxypropionic acidPFMOBA - perfluoro-4-methoxybutanoic acidPMPA - perfluoro-2-methoxypropionic acidAnalysis of a Standard Containing Only PMPAChemours Fayetteville Works, North CarolinaFigure8April 2021PMPAPEPA PFMOBAChemours Std in waterPercentageof TotalAnalyte Area Count Area CountPMPA 53,035 99.16%PEPA 317 0.59%PFMOBA 131 0.24%Total Area Count 53,483
Notes:PFMOPrA - perfluoromethoxypropionic acidPMPA - perfluoro-2-methoxypropionic acidAnalysis of a Standard Containing Only PFMOPrAChemours Fayetteville Works, North CarolinaFigure9April 2021PMPAPFMOPrASynQuest StdPercentageof TotalAnalyte Area Count Area CountPFMOPrA 32,163 97.72%PMPA 749 2.28%Total Area Count 32,912
Notes:PEPA - perfluoro-2-ethoxypropionic acidPFMOPrA - perfluoromethoxypropionic acidPMPA - perfluoro-2-methoxypropionic acidAnalysis of a Standard Containing Only PEPAChemours Fayetteville Works, North CarolinaFigure10April 2021Chemours Std in waterPMPAPEPAPFMOPrAPercentageof TotalAnalyte Area Count Area CountPEPA 43,369 97.93%PMPA 427 0.96%PFMOPrA 490 1.11%Total Area Count 44,286
Notes:PEPA - perfluoro-2-ethoxypropionic acidPFMOPrA - perfluoromethoxypropionic acidPFMOBA - perfluoro-4-methoxybutanoic acid PMPA - perfluoro-2-methoxypropionic acidAnalysis of a Standard Containing Only PFMOBAChemours Fayetteville Works, North CarolinaFigure11April 2021PMPAPEPAPFMOPrAPFMOBASynQuest StdPercentageof TotalAnalyte Area Count Area CountPFMOBA 14,465 93.30%PMPA 220 1.42%PFMOPrA 285 1.84%PEPA 534 3.44%Total Area Count 15,504
ATTACHMENT A
Letter from Chemours to DEQ proposing additional work
analyzing PMPA and PEPA standards and five
archived field samples by Method 533 to look for the
presence of the linear isomers (August 26, 2020).
ATTACHMENT B
Letter from DEQ to Chemours with requests for
experiments to further investigate the presence of
branched and linear isomers in standards and field
samples (October 2, 2020).
October 2, 2020
Ms. Christel Compton
The Chemours Company FC, LLC
22828 Hwy 87 W
Fayetteville, NC 28306
Re: Isomer Response
Dear Ms. Compton,
In response to the August 26, 2020 letter, the July 17, 2020 Isomer Report, and the
technical call on September 22, 2020, DEQ recommends additional LC/MS/MS experiments
be performed prior to running environmental samples. The purpose of these experiments
is to definitively show not only which isomers are present in the standards and samples but
also to confirm if other isomers are present, even at low levels, or if they are entirely
absent.
The July 17, 2020 report does not provide sufficient analytical assessment of the linear
isomers (PFMOPrA and PFMOBA). While the report does show that the dominant species
present are likely to be the branched isomers (PMPA and PEPA), additional work should be
conducted to show whether or not the linear isomers are also present in the analytical
standards and environmental samples. The LC/MS/MS experiments performed should be
repeated with method improvements that allow for better chromatographic separation and
instrument detection of the four individual analytes of interest.
Method improvements are recommended for this investigation only. Routine analysis of
the isomers would not require implementation of all suggested method changes for the
isomer evaluation. However, some of the method changes may help with future routine
analysis. Chemours’ August 26, 2020 letter states an intention to analyze a set of samples
to establish if there are any false positives. It is recommended that further method
development be conducted prior to analyzing these samples so it is clear that the linear
isomers are sufficiently assessed by these methods. Recommended method improvements
are listed below. More detailed information, including a proposed path forward, is
described in the attached supporting information. In summary:
1) Use four separate solutions for the four analytical standards rather than
mixtures.
2) Use sufficient concentrations of standards for strong peaks (minimum
instrument response of 106) to determine any impurities in each of the four
standards.
3) Optimize the mass spectrometer settings for the linear isomers when analyzing
for them.
4) Chromatographically resolve (separate) each isomer pair, and monitor multiple
MRM transitions for each peak to determine individual isomer
presence/absence.
Review Certificates of Analysis from Wellington and publications by Strynar et al. 2015 and
Song et al. 2018 for examples of LC/MS/MS methods for analysis of these isomers. Once
these methods are demonstrated for characterizing the 4 analytical standards, the
optimized methods for each compound should be applied to investigating the potential
presence of linear isomers in the environmental samples using linear calibration standards.
Please let us know of any questions.
Sincerely,
Michael E. Scott
Director, Division of Waste Management
NC DEQ
Review of Chemours Report Evaluation of the Analysis of the Branched and Linear Isomers
PMPA/PFMOPrA and PEPA/PFMOBA
Overall finding: The LC/MS/MS experiments performed for the evaluation of branched and linear
isomers PMPA/PFMOPrA and PEPA/PFMOBA should be repeated with method improvements that
better allow for better chromatographic separation and instrument detection of the four individual
analytes of interest. Method improvements suggested are for this investigation only. Routine analysis
of the isomers would not require implementation of all suggested method changes for the isomer
evaluation. However, some of the method changes may help with future routine analysis.
Monitoring of individual analytes: Two solutions containing two analytes were used in this experiment
(PMPA/PEPA and PFMOPrA/PRMOBA). One solution for each of the four analytes would provide
unambiguous data that any instrument response observed can only be coming from the single analyte in
solution.
Mass spectrometer signal for PMPA and PEPA: Several of the specific comments and recommendations
in this review are related to the amount of instrument response for each isomer. Figures 11 and 12 of
the report (below) show that under the instrument conditions used, a signal of 4 x 106 to 1 x 107 can be
achieved for PEPA and PMPA in a Site sample without saturating the instrument detector. This indicates
that global mass spectrometer settings and analyte specific parameters are providing good signal for
PMPA and PEPA in the field sample. However, similar instrument response was not achieved for the
analysis of standard material of PMPA and PEPA. In order to achieve an instrument response of about 1
x 106 for the most abundant MRM, a higher concentration stock solution for PMPA and PEPA should be
used.
Mass spectrometer signal for PFMOPrA and PFMOBA: The report indicates that there was difficulty
achieving a strong MRM signal in the mass spectrometer for PFMOPrA and PFMOBA. Supporting
information on the Wellington certificates of analysis for the two linear isomers show a good LC/MS
response for the parent ion and a strong signal for the most predominant MRM using LC/MS/MS (see
figures below). Sufficient signal is also achieved for monitoring a second MRM transition for each
analyte (see figures below). HPLC and mass spectrometer conditions under which the chromatograms in
the Wellington C of A were collected are also shown below.
The information from the Wellington Certificates of Analysis indicate that it is possible to get sufficient
response for two MRMs each for PFMOPrA and PFMOBA. The following procedures may allow for
increased signal for PFMOPrA and PFMOBA: ensuring complete solubilization of the analyte in solution
and optimizing analyte specific mass spectrometer settings (including collision energy and use of correct
masses of parent and product ion to the nearest tenth of a decimal place based on current mass
spectrometer calibration – ex. 279.0 vs 278.9 can make a difference in signal intensity).
If the above recommendations do not provide adequate signal for PFMOPrA and PFMOBA, then global
mass spectrometer settings (ex. ESI source conditions) may need to be optimized for the linear isomers.
If source parameters are different enough between the isomers such that ionization of one or the other
isomers is significantly inhibited, then two analytical methods may be needed for the purposes of this
investigation (one method with source conditions optimized for linear isomers and the second method
with source conditions optimized for branched isomers)—the same HPLC conditions would be used for
both methods.
MRM selection: Multiple MRM transitions can be monitored for each analyte once a signal of
approximately 1 x 106 is achieved for the predominant MRM. This level of instrument signal for the
predominant MRM should allow for increased signal and detection of analytes using additional MRM
transitions. Expected fragmentation patterns for each analyte is shown below:
Specifically, the following MRMs can be monitored for each analyte:
o PMPA: 229185 and 22985
o PFMOPrA: 229185 and 22985
o PEPA: 279235, 27985, and 279135
o PFMOBA: 279235, 27985, and 279135
PMPA and PFMoPrA have the same two MRM transitions. Therefore, chromatographic separation of
these two isomers is necessary for unambiguous identification. PEPA and PFMOBA have unique MRM
transitions that should provide spectral data for isomer identification even when the compounds
coelute. The unique transitions are 279135 for PEPA and 27985 for PFMOBA. The MRM transition
of 279235 is common to both isomers. When both isomers are present in solution, and particularly
when they coelute, both isomers will contribute to this signal. Chromatographic separation of PEPA and
PFMOBA, if possible, would also provide more definitive isomer identification.
Chromatographic separation: Retention times of isomers run by the Table 3 and Table 6 method in the
report are as follows:
PMPA: 6.3 min (Table 3), 6.5 min (Table 6)
PFMOPrA: 6.8 min (Table 3), 6.3 min (Table 6)
PEPA: 8 min (Table 3), 7.13 min (Table 6)
PFMOBA: 8 min (Table 3), 7.14 min (Table 6)
Once sufficient signal is obtained for each analyte in multiple MRM transitions, chromatographic
separation of the isomer pairs should be optimized, with baseline resolution of the isomers preferred.
Separation of PMPA and PFMoPrA is the most important because both share the same two MRM
transitions and therefore are not able to be distinguished spectrally. The Table 3 and Table 6 methods
do not provide baseline resolution between these isomers. Identification of this isomer pair cannot be
made using retention times alone because the isomers are not well enough resolved and retention
times can shift, particularly in different sample matrices. Three chromatographic methods that may
provide better separation (from Strynar et al 2015, EPA Method 533, and Song et al 2018) are shown
below along with representative chromatograms. The Strynar et al 2015 and Song et al 2018 HPLC
conditions in particular show chromatographic separation for both isomer pairs.
Strynar et al 2015 – the second chromatogram shows likely PFMOPrA and PMPA separation
EPA Method 533 – Peaks 4 and 9 are PFMOPrA and PFMOBA
Song et al 2018 – Second and third chromatograms in the surface water portion of figure below show
separation of PFMOPrA/PMPA and PFMOBA/PEPA using an Acquity HSS PFP HPLC Column
The above chromatographic methods are examples that may be used. Any chromatographic method
that provides baseline resolution of PMPA and PFMOPrA would be sufficient. Resolution of PEPA and
PFMOBA is desired, however this isomer pair may be more difficult to resolve chromatographically and
has unique MRMs that can be used for identification.
Surrogates: After MRM signal and chromatographic separation have been optimized and PMPA, PEPA,
PFMOPrA, and PFMOBA have each been run with the optimized method, a surrogate can be introduced
into the method. The surrogate serves as a reference point to monitor any retention time shifts that
occur in samples. Good surrogate candidates are perfluoropentanoic acid (PFPeA) or labeled
perfluoropentanoic acid (13C5‐ PFPeA) because these compounds elute at a retention time in between
that of the isomer pairs (see figure below‐ PFMOPrA, PFPeA, 13C5‐ PFPeA, and PFMOBA are analytes
4,5,6, and 9 in the chromatogram, respectively).
Surrogate should be added to each individual analyte standard solution and analyzed by LC/MS/MS to
establish a relative retention time for each analyte. Environmental samples analyzed should have
surrogate added prior to analysis to provide additional confidence in the identification of isomers using
retention time information. Surrogates should be optimized in the mass spectrometer to determine two
suitable MRMs transitions (PFPeA has a transition of 263219 and 13C5‐ PFPeA has a transition of
268223 in EPA Method 533).
Environmental Samples: Environmental samples should be freshly collected, preserved, and stored
refrigerated within the holding time of the method rather than from archived samples. Analysis for all
analytes should occur at the same time using the following MRMs for each analyte:
o PMPA: 229185 and 22985
o PFMOPrA: 229185 and 22985
o PEPA: 279235, 27985, and 279135
o PFMOBA: 279235, 27985, and 279135
The chromatographic method should achieve as much separation as possible between PMPA and
PFMOPrA. A surrogate should be added to samples in order to better track shifts in retention times that
may occur in environmental samples analyzed by LC/MS/MS.
Report figures: This section contains specific comments about figures contained in the report.
Figure 3 has no explanation regarding the hump on the back side of the peak in the
229185 transition. The 22985 may have more signal with a more concentrated
standard.
Figure 4 PFMOPrA signal for both MRMs likely will be better with mass spectrometer
optimized conditions
Figure 5 has no explanation of peak at 7.336 minutes. More signal from a more
concentrated standard would allow for observations in MRM 22985 as well
Figure 6 signal likely will be better with optimized mass spectrometer conditions
Figure 7 zooming to a narrower scale on the y axis for the 27985 transition would
allow the viewer to see whether there is any signal in the 102 to 103 range. The
transition 279135 should also be monitored
Figure 8 expected retention time is at 8 minutes. Please expand the chromatogram in
this area to see any 102 to 103 intensity peaks present there. Signal is present at 8
minutes for the 27985. Please adjust the range of the y‐axis to better show this
response. The transition 279135 should also be monitored
Figure 9 PEPA data should be collected at the MRMs 279235, 27985, and 279135
in a higher concentration standard (to obtain a signal of 1 x 106 response for the MRM
shown)
Figure 10 PFMOBA data should be collected at the MRMs 279235, 27985, and
279135 with optimized mass spectrometer conditions and a more concentrated
standard solution
Figure 11 Demonstrates good signal for PMPA in 229185 transition. Should also
monitor 22985 transition. Chromatographic resolution of isomers needed.
Figure 12 Figure 11 Demonstrates good signal for PMPA in 279235 transition. Should
also monitor 27985 and 279135 transitions. Chromatographic resolution of
isomers needed.
Each analyte should be monitored using the Table 3 and Table 6 methods, even when weak
signal was observed previously because this is work is investigative in nature and the
analytes may show up unexpectedly.
Comments on Report Conclusions:
Conclusion stated in Isomer Report: Additionally, both the lack of peak broadening or
shoulder peaks for PMPA and PEPA and lack of spectral response characteristic of
PFMOPrA and PFMOBA under Table 6 conditions indicate that Chemours’s PMPA and
PEPA standards consist primarily or entirely of the branched isomer.
o Reviewer Responses:
o Figure 3 does show a shoulder on the peak for PMPA in the 229185 transition.
o Figure 5 additionally shows a peak at 7.336 minutes in the 229185 transition
for PMPA.
o There is not enough signal in the MRM transitions for either PFMOPrA or
PFMOBA to assess spectral response. Method optimization (mass spectrometer
conditions) for these compounds would likely allow for increased signal and
collection of data for these compounds, from which then conclusions can be
drawn. Two different mass spectrometer methods may need to be run, one
each with MS conditions optimized for linear and branched isomers.
o The branched isomer is likely to be the predominant form of the isomers;
however more data needs to be collected, especially regarding the linear
isomers to support this statement and the purity assessment.
Conclusion stated in Isomer Report: Both the Table 3 and Table 6 Methods are likely to
be able to resolve the PMPA/PFMoPrA and PEPA/PFMOBA isomer pairs when
chromatographic and spectral data are considered together.
Reviewer Responses:
o PEPA/PFMOBA cannot be resolved with the chromatographic and spectral data
collected in this report. The compounds are not chromatographically resolved
using either the Table 3 or Table 6 method. The MRM transitions that could
identify the compounds are either not monitored (279135 for PEPA) or
appear to not be optimized (27985, PFMOBA). The MRM transition
(279235) is used but is common to both PEPA and PFMOBA and cannot be
used to distinguish between the two isomers.
o PMPA/PFMoPrA cannot be resolved with the chromatographic and spectral data
collected in this report. These isomers share the same two MRM transitions
(229185 and 22985) and cannot be distinguished spectrally when the peaks
coelute. Section 3.2 of the report further states the following: The peak width
for PMPA in both methods is wide, and the peak for PFMOPrA overlaps,
indicating that PMPA and PFMOPrA will not likely be fully chromatographically
resolved in both are present in a sample.
Conclusion stated in Isomer Report: Review of a Site sample containing PMPA and PEPA
and analyzed by the Table 6 method shows PFMOPrA and PFMOBA are unlikely to be
present, as indicated by the lack of peak broadening or peak shoulders.
Reviewer Responses:
o PFMOBA elutes at the same retention time PEPA and would not likely be
present as a peak shoulder, but within the center of the PEPA peak.
o The MRM transition monitored (279235) is common to both PFMOBA and
PEPA and would therefore not distinguish between the two isomers.
Additionally, the signal for PFMOBA in this MRM does not appear to be
optimized. Additional MRMs unique to PEPA (279135) and PFMOBA
(27985) under optimized MS conditions would provide data for isomer
identification. Two analytical methods may have to be run – one each
containing optimized MS parameters for linear and branched isomers.
o PFMOPrA and PMPA may not be chromatographically resolved in the Table 6
method and have the same MRM transitions (22985 and 229185). Analysis
of a sample with baseline resolution of the isomer pair would provide more
definitive results.
Proposed Path Forward: Repeat the experiments as outlined below (additional information about each
aspect is covered in a previous section of this document):
o Step 1: Optimize the method
o Prepare four individual analyte standard solutions, one each for PFMOPrA, PMPA,
PFMOBA, and PEPA
o Ensure that PFMOPrA and PFMOBA standards are completely solubilized in the standard
solution
o Ensure analyte specific mass spectrometer settings are optimized for PFMOPrA and
PFMOBA.
o Optimize global mass spectrometer settings for PFMOPrA and PFMOBA (if these
parameters are significantly different between linear and branched isomers may need
one method each for linear and branched isomers‐one each with optimized ESI source
conditions)
o Run each analyte standard by LC/MS/MS using full scan mode over the range 50 to 300
m/z in order to confirm correct selection of MRM transitions for each analyte and
provide the full scan chromatograms
o Optimize MRM signal for predominant transition so that the response is approximately
1 x 106 using an existing chromatography method. Once this is achieved, add in
additional MRMs for each analyte
o Use a different chromatographic method to baseline resolve PFMOPrA and PMPA.
Ideally PFMOBA and PEPA would have better separation than in the Table 3 and Table 6
methods, but separation is less important for these isomers.
o Run each of the four individual analyte standard solutions using additional MRMs,
optimized mass spectrometer settings for the linear isomers, and a chromatography
method that baseline resolves PFMOPrA and PMPA
o Perform LC/MS/MS analysis of two solutions, one containing PFMOPrA and PMPA and
the other containing PFMOBA and PEPA to show resolution of isomer pairs together in a
solution
o Step 2: Add in surrogates
o Prepare a standard solution of surrogate standard
o Optimize two MRMs for the surrogate
o Add surrogate to an aliquot of each standard solution (one each PFMOPrA, PMPA,
PFMOBA, and PEPA)
o Run each of the four individual analyte standard solutions containing surrogate using
additional MRMs, optimized mass spectrometer settings for the linear isomers and
surrogate, and a chromatography method that baseline resolves PFMOPrA and PMPA
o Establish relative retention times for each analyte to the surrogate
o Step 3: Analyze Environmental/Site Samples
o Collect samples fresh, add appropriate preservative, store refrigerated
o Add surrogate to sample
o Analyze within refrigerated holding time for sample
o Perform analysis for all compounds at once using the optimized LC/MS/MS method
containing additional MRMs and chromatography method capable of baseline resolving
PFMOPrA and PMPA
o Any quantitation of linear isomers should be done with linear isomer calibration
standards
o Any quantitation of branched isomers should be done with branched calibration
standards