EPA 600/R-12/703 July 2012 | www.epa.gov/ord
United States
Environmental Protection
Agency
Removal of Perfluorocarboxylic Acids (PFCAs) from Carpets Treated with
Stain-protection Products by Using Carpet Cleaning Machines
Office of Research and Development
-------
EPA/600/R-12/703
July 2012
Removal of Perfluorocarboxylic Acids (PFCAs) from
Carpets Treated with Stain-protection Products by
Using Carpet Cleaning Machines
Heidi F. Hubbard1, Zhishi Guo2, and Kenneth A. Krebs
U.S. Environmental Protection Agency
Office of Research and Development
National Risk Management Research Laboratory
Air Pollution Prevention and Control Division
Research Triangle Park, NC 27711
and
Sara K. Metzger, Corey A. Mocka, Robert H. Pope, and Nancy F. Roache
ARCADIS US, Inc.
4915 Prospectus Dr., Suite F
Durham, NC 27713
1 Current address: ICF International, 2222 East NC-54, Suite 480, Durham, NC 27713
2 Corresponding author (guo.zhishi@epa.gov)
-------
Notice
This document has been reviewed in accordance with the U.S. Environmental Protection Agency
policy and approved for publication. Mention of trade names or commercial products does not
constitute endorsement or recommendation for use.
-------
Executive Summary
E.I Background and Objective
Perfluorinated carboxylic acids (PFCAs) are fully fluorinated organofluorine compounds with a
carboxylic acid functional group (-COOH). As a member of the PFCA family, perfluorooctanoic
acid (PFOA) and its salts were once used as a surfactant in the manufacturing of perfluorinated
polymers such as polytetrafluoroethylene (PTFE). Thus, PFOA and its salts may exist as residuals
in PFTE or other fluoropolymer products. PFOA and other PFCAs may also exist in fluorotelomer
products (such as stain-repellants) as unwanted reaction by-products. The U.S. Environmental
Protection Agency (EPA) began investigating PFOA and its related chemicals in the 1990s and
found that it is very persistent in the environment, is found at low levels both in the environment
and in the blood of the general U.S. population, and causes developmental and other adverse
effects in laboratory animals (U.S. EPA, 2012). In 2006, EPA and the eight major companies in the
fluoropolymer and fluorotelomer industry launched the PFOA Stewardship Program, in which
companies committed to reduce global facility emissions and product content of PFOA and related
chemicals by 95 percent by 2010 and to work toward eliminating emissions and product content by
2015 (U.S. EPA, 2012).
Previous studies have shown that consumer articles that are made from or treated with
fluoropolymers and fluorotelomers products may contain low levels of PFCAs. PFCAs are found
in a variety of consumer products, including, but not limited to, treated clothing and textiles, floor
care products, paper containers for food, and carpets. Among the consumer articles examined by
Washburn et al. (2005) and Guo et al. (2009), carpet that was pretreated with stain-repellents and
carpet that was treated with after-market stain-resistant formulations were the largest PFOA
sources in homes. Once PFCAs are brought into the indoor environment, they are expected to stay
for a long period of time because PFCAs are persistent in the environment and because most
PFCAs are semi-volatile compounds. PFCAs can also be absorbed by household dust, which may
serve as a source for inhalation or digestive exposure. Therefore, it is important to understand the
feasibility of in-situ removal of PFCAs from treated carpet. To our knowledge, this issue has not
been addressed by any publications in the existing literature.
The main goal of this study was to quantify the efficiency of common carpet cleaning methods —
steam cleaning and hot water extraction — in the removal of PFCAs from residential and
commercial carpet that was manually treated with stain-protection solutions. The objective was to
determine if these cleaning techniques are viable methods for reducing indoor exposure to PFCAs
associated with carpet that was previously treated with PFCA-containing products.
-------
E.2 Test Method
E.2.1 Test Facility
The carpet cleaning experiments were performed under close-to-realistic conditions in the U.S.
EPA research house located in Gary, NC. The research house is a three-bedroom, ranch-style house
with a crawl space, a central, forced-air heating system that uses natural gas, and an electric air-
conditioning system. The total floor area is 126 m2. The master bedroom (MB) and front-corner
bedroom (FCBR) were used for the carpet cleaning tests.
E.2.2 Test Materials
By the time this project started, mill-treated carpet containing high levels of PFCAs were no longer
available in local stores, an indication that the manufacturers had taken actions to reduce or
eliminated PFCAs from their products. In this study, two types of carpet with low-levels of
background PFCAs were treated with after-market carpet treatment solutions (see below). The two
types of carpet were: 1) residential carpet that was Green-Label Certified by the Carpet and Rug
Institute (CRI) with a pile yarn content of 100% polytrimethylene terephthalate (PTT) - a product
from recycled plastic bottles - with a textured cut pile and a woven polypropylene backing, and 2)
commercial carpet made with Antron® fiber, a nylon 6,6, hollow-filament fiber with a soil-
resistant treatment incorporated into the fiber. This type of commercial carpet is commonly used in
schools and offices. The PFCA content in these carpets were below 8 ng/g. These background
PFCAs may have come from recycled old carpet.
Carpet Treatment Solutions
Two commercial carpet stain-protection treatments (CT-1 and CT-2) were used to treat the carpet.
The total PFCA contents in these treatment solutions are presented in Section E.3.1 below.
Carpet Cleaning Machines
Two carpet cleaning machines were chosen to clean the carpets after treatment with a carpet stain-
protection solution: 1) a residential cleaning machine (CM-1), the Rug Doctor Mighty Pro model
MP C-20, and 2) a portable professional steam cleaner (CM-2), the Century 400 Ninj a Warrior.
The residential unit uses hot tap water for the extraction with no additional heating during the
cleaning process; the professional steam cleaner has an 1850-watt, in-line heater that produces
steam. Figure E. 1 shows the Rug Doctor (front) and the Ninj a Warrior (rear).
-------
E.I. Residential and commercial carpet cleaning machines used for the tests
Carpet Cleaning Detergent
Two carpet cleaning agents were selected to test the removal of PFCAs from carpets, i.e., a
residential carpet cleaning detergent (CD-2) recommended for use with the residential machine
(CM-1) and a commercial carpet cleaning detergent (CD-I) chosen for use with the commercial
machine (CM-2). According to the material safety data sheets (MSDSs), CD-I contains 3 to 6% of
dipropylene glycol methyl ether (CAS# 34590-94-8) while CD-2 contains 1 to 5% of sodium 2-
ethylhexyl sulfate (CAS# 126-92-1) and <1% of branched tridecylalcohol (CAS# 69011-36-5).
Prior to use, each clean agent was screened for PFCAs. The residential detergent measured most
PFCAs below the quantification limits of the instrument, with perfluorotridecanoic acid (Co)
having the highest concentration at 3 ng/mL. The commercial detergent had only trace levels of Cg
through CIQ, and all were below the practical quantification limits.
E.2.3 Test Procedure
The test carpet was installed on a solid urethane carpet pad on the bedroom floors. Prior to the
application of the carpet treatment solution, carpet samples were taken to determine the PFCA
content in untreated carpet. The carpet was then treated with a carpet treatment solution.
Following the application of the treatment and a subsequent 48-hour drying period, a series of three
carpet cleanings was performed. A drying period of at least 48 hours followed each cleaning before
carpet samples were collected. The first set of tests used only hot water or steam for the cleaning
process to determine the efficiency of each carpet cleaning machine's method of removing the
applied PFCAs. Additionally, four experiments, one of which was a duplicate test, were conducted
using cleaning detergents (CD-I or CD-2) in conjunction with a carpet cleaning machine to assess
any additional PFCA removal. The complete experimental test matrix is summarized in Table E.I.
-------
Table E.I. Experimental matrix for testing of carpet care treatments
Experiment
1
2
3a
4
5
6
7
8b
9
Carpet Type
Commercial
Commercial
Residential
Residential
Residential
Commercial
Residential
Residential
Residential
Carpet Treatment
(CT)
CT-1
CT-2
CT-1
CT-1
CT-2
CT-2
CT-1
CT-1
CT-2
Cleaning
Machine (CM)
CM-2
CM-2
CM-1
CM-2
CM-2
CM-2
CM-1
CM-1
CM-2
Carpet
Detergent (CD)
None
None
None
None
None
CD-I
CD-2
CD-2
CD-I
Initial scouting test; Duplicate test.
E.3 Findings
E.3.1 Extractable PFCAs in Carpet Treatment Solutions
The total extractable PFCAs (i.e., the sum of all PFCAs quantified) in the two carpet treatment
solutions were, respectively, 6360 and 7500 ng/g. The distributions of individual PFCAs are shown
in Figure E.2. The error bars represent ±1 standard deviation (n = 4 for CT-1 and n = 6 for CT-2).
-------
-4^
o
u
2500
2000
1500
1000
500
A
X* X
*
I
t
X *
Figure E.2 PFCA contents in the two carpet treatment solutions (CT-1 and CT-2)
E.3.2 PFCA Content in Carpet Before and After Cleaning
As an example, Figure E.3 shows the general patterns of the carpet cleaning tests. The application
of the carpet treatment solution increased the PFCA content in the carpet and, following each
carpet cleaning event, there was reduction of PFCA content in the carpet. In this experiment, the
total reduction after three rounds of cleaning was approximately 50%.
-------
3000
D
O.
3
to
PLH
Ml
^
O
O
U
2000 -
1500 -
i I i
1— 1-1
1
— ±—
Pre-Application Post-Application Post-Cleaning Post-Cleaning Post-Cleaning
Round 1 Round 2 Round 3
Figure E.3 Average total PFCAs in composite carpet fiber samples for Experiment 4
[Residential carpet treated with carpet treatment solution CT-1; hot-water extraction with the residential
cleaning machine (CM-1); no detergent]
E. 3.3 Percent Removal of PFCAs by Carpet Cleaning
For each experiment, the percent removal of total PFCAs after the final cleaning was calculated by
comparison of the total PFCA concentration after application of the carpet treatment to the total
PFCA concentration after the third round of cleaning. As shown in Figure E.4, the percent removal
of the total PFCA ranged from 26% to 76% with an average of 55%.
-------
1UU/0
OAO/
3
s
W /TAG/
^ oU/o
•g
g
£
^ /I no/
Q, 4U/o
__
«
-w
P
H
ino/
zU/o
no/, .
123456789
Experiment ID
Figure E.4. Percent removal of total PFCAs following three rounds of cleaning for each experiment
E. 3.4 Effect of Cleaning Machine
The two types of cleaning machines yielded similar removal efficiency for total PFCAs. The
difference shown in Table E.2 is not statistically significant.
Table E.2 Comparison of the removal efficiency of the two cleaning machines for total PFCAs
following three rounds of cleaning
ID
CM-1
CM-2
Machine Type
Residential
Commercial
Cleaning Method
Hot-water extraction
Steam cleaning
Mean
52.2%
57.4%
SD
6.2%
18.5%
n
3
4
E. 3.5 Effect of Detergent
Using detergent during carpet cleaning showed modest increase in removal efficiency (Figure E.5).
However, comparison of the test with detergent CD-I with the test without using any detergent
yielded the two-tailed P value of 0.1063. By conventional criteria, this difference is considered to
be not statistically significant.
-------
100%
80%
•§ 60%
o
01
^, 40%
20%
0%
CD-I
CD-2
Detergent
None
Figure E.5 Effect of using carpet detergent on the average removal efficiency for total PFCAs
following three rounds of cleaning
E.4 Research Implications
r-M-, 9
The average American home has about 1000 square feet (93 m ) of carpet. Previous research has
shown that treated carpets represent one of the largest sources of PFCAs in homes. Additionally,
PFCAs are highly stable, with the potential for a long residence time indoors, and can also bind to
house dust, making it difficult to remove the PFCAs from the indoor environment. The results of
this research indicated that cleaning carpets can reduce the amount of PFCAs they contain, but the
chemicals cannot be totally eliminated by the cleaning process. Carpet cleaning by hot-water or
steam cleaning is only modestly effective in removing PFCAs. On average, the removal efficiency
for each round of cleaning is approximately 20%. At this removal efficiency, it requires three,
seven, and ten rounds of cleaning to remove, respectively, 50%, 80%, and 90% of total PFCAs in
treated carpet.
-------
Table of Contents
Notice 1
Executive Summary 2
Table of Contents 10
List of Figures 12
List of Tables 13
Acronyms and Abbreviations 15
1. Introduction 16
1.1 Background 16
1.2 Goal and Objective 17
2. Materials and Methods 18
2.1 Test Facility 18
2.2 Test Materials 19
2.2.1 Carpet Selection for Research House Experiments 19
2.2.2 Professional Carpet Treatment Solutions 19
2.2.3 Carpet Cleaning Machines 20
2.2.4 Carpet Cleaning Detergents 21
2.3 Experimental Design 21
2.4 Test Procedure 22
2.4.1 Layout of Carpet Sampling Sections 22
2.4.2 Dilution and Application of the Carpet Treatment 23
2.4.3 Cleaning of Carpets 25
2.5 Sampling Methods 26
2.5.1 Collection of Carpet Samples 26
2.5.2 Wipe Sampling 27
2.6 Sample Analysis 28
2.6.1 Extraction of Residential Carpet Samples 28
2.6.2 Extraction of Commercial Carpet Samples 28
2.6.3 Extraction of Liquid Samples 29
2.6.4 QC Sample Preparation 29
2.6.5 Sample Analysis 29
2.7 Quality Assurance and Quality Control 30
10
-------
2.7.1 Uniformity of Carpet Protector Treatment Solution 31
2.7.2 Calibration of LC/MS/MS 32
2.7.3 Daily Calibration Checks 33
2.7.4 Contamination Checks 33
2.7.5 Weight Measurements 33
3. Results 34
3.1 Summary of Experimental Conditions 34
3.2 Extractable PFCA Content in Carpet Treatment Solutions 34
3.3 Extractable PFCA Content in Wipe Samples Taken from the Walls 35
3.4 Extractable PFCA Content in Carpet Samples 36
3.5 Percent Removal of PFCAs by Cleaning 41
4. Discussion 43
4.1 Removal Efficiency of Speciated Extractable PFCAs in Carpet Samples 43
4.2 Overall PFCA Removal Efficiency 43
4.3 Comparison of Duplicate Experiments 44
4.4 Factors Affecting the Efficiency of PFCA Removal 44
4.5 Composite vs. Individual Samples 46
4.6 Extractable PFCA Content Collected from Walls 46
5. Conclusions and Recommendations 47
Acknowledgments 48
References 49
Appendix A: Data Al
11
-------
List of Figures
Figure 2.1. Floor plan for the research house in Gary, NC 18
Figure 2.2. Residential and commercial carpet cleaning machines 20
Figure 2.3. Diagram of generic carpet sampling quadrant 23
Figure 2.4. Application cart 24
Figure 2.5. Diagram of first application process 24
Figure 2.6. Diagram of second application process 25
Figure 2.7. Diagram of carpet cleaning using commercial cleaner CM-2 (not to scale) 26
Figure 2.8. Residential carpet after all sampling stages 27
Figure 2.9. Average of individual quadrant samples for post-application PFCA concentrations in
each experiment and associated standard deviations 32
Figure 3.1. Extractable PFCA content in carpet treatment solutions CT-1 and CT-2 35
Figure 3.2. Average total PFCAs in composite carpet fiber samples for Experiment 1 (C-l-2-0) 37
Figure 3.3. Average total PFCAs in composite carpet fiber samples for Experiment 2 (C-2-1-0) 37
Figure 3.4. Average total PFCAs in composite carpet fiber samples for Experiment 3 (R-l-1-0) 38
Figure 3.5. Average total PFCAs in composite carpet fiber samples for Experiment 4 (R-l-2-0) 38
Figure 3.6. Average total PFCAs in composite carpet fiber samples for Experiment 5 (R-2-2-0) 39
Figure 3.7. Average total PFCAs in composite carpet fiber samples for Experiment 6 (C-2-2-1) 39
Figure 3.8. Average total PFCAs in composite carpet fiber samples for Experiment 7 (R-l-1-2) 40
Figure 3.9. Average total PFCAs in composite carpet fiber samples for Experiment 8 (R-l-1-2),
duplicate of Experiment 7 40
Figure 3.10. Average total PFCAs in composite carpet fiber samples for Experiment 9 (R-2-2-1) 41
Figure 3.11. Calculated percent removal of total PFCAs for each experiment 42
Figure 4.1. Average percent reduction in PFCAs by experimental variable for composite samples 45
12
-------
List of Tables
Table 2.1. Conditions in the research house 19
Table 2.2. Components of carpet treatments CT-1 and CT-2 20
Table 2.3. Experimental matrix for testing of carpet care treatments 22
Table 2.4. Carpet fiber collection strategy 27
Table 2.5. Analyte names, abbreviations, chemical formulas, molecular weights (g/mol), and
Chemical Abstracts Service registration numbers (CAS #) 30
Table 2.6. Measurement quality objectives 31
Table 3.1. Environmental parameters recorded during the cleaning experiments 34
Table 3.2. Results of wall wipe samples collected during Experiment 3 36
Table 4.1. Percent removal efficiency of speciated compounds 43
Table 4.2. Composite carpet sample data summarizing average percent reduction in PFCAs by
experiment 44
Table 4.3. Average percent reduction in PFCAs by experimental variable for composite samples 45
Table A.I. Average extractable PFCAs (ng/g) and percent of original amount removed in composite
carpet fiber samples at each experimental stage for Experiment 1 (C-1 -2-0) A2
Table A.2. Average extractable PFCAs (ng/g) and percent of original amount removed in composite
carpet fiber samples at each experimental stage for Experiment 2 (C-2-2-0) A3
Table A.3. Average extractable PFCAs (ng/g) and percent of original amount removed in composite
carpet fiber samples at each experimental stage for Experiment 3 (R-l-1-0) A4
Table A.4. Average extractable PFCAs (ng/g) and percent of original amount removed in composite
carpet fiber samples at each experimental stage for Experiment 4 (R-1-2-0) A5
Table A.5. Average extractable PFCAs (ng/g) and percent of original amount removed in composite
carpet fiber samples at each experimental stage for Experiment 5 (R-2-2-0) A6
Table A.6. Average extractable PFCAs (ng/g) and percent of original amount removed in composite
carpet fiber samples at each experimental stage for Experiment 6 (C-2-2-1). Re-use of
carpet from Experiment 2 A7
Table A.7. Average extractable PFCAs (ng/g) and percent of original amount removed in composite
carpet fiber samples at each experimental stage for Experiment 7 (R-l-1-2).
Experiments 7 and 8 considered replicate tests A8
Table A. 8. Average extractable PFCAs (ng/g) and percent of original amount removed in composite
carpet fiber samples at each experimental stage for Experiment 8 (R-l-1-2).
Experiments 7 and 8 considered replicate tests A9
Table A.9. Average extractable PFCAs (ng/g) and percent of original amount removed in composite
carpet fiber samples at each experimental stage for Experiment 9 (R-2-2-1). Re-use of
carpet from Experiment 5 A10
Table A. 10. Average extractable PFCAs (ng/g) post-application in individual carpet fiber samples
for Experiment 1 (C-1-2-0) All
13
-------
Table A. 11. Average extractable PFCAs (ng/g) post-application in individual carpet fiber samples
for Experiment 2 (C-2-2-0) All
Table A. 12. Average extractable PFCAs (ng/g) post-application in individual carpet fiber samples
for Experiment 3 (R-l-1-0) A12
Table A. 13. Average extractable PFCAs (ng/g) post-application in individual carpet fiber samples
for Experiment 4 (R-l-2-0) A12
Table A. 14. Average extractable PFCAs (ng/g) post-application in individual carpet fiber samples
for Experiment 5 (R-2-2-0) A13
Table A. 15. Average extractable PFCAs (ng/g) post-application in individual carpet fiber samples
for Experiment 6 (C-2-2-1) A13
Table A. 16. Average extractable PFCAs (ng/g) post-application in individual carpet fiber samples
for Experiment 7 (R-l-1-2). Experiments 7 and 8 considered replicate tests A14
Table A. 17. Average extractable PFCAs (ng/g) post-application in individual carpet fiber samples
for Experiment 8 (R-l-1-2). Experiments 7 and 8 considered replicate tests A14
Table A. 18. Average extractable PFCAs (ng/g) post-application in individual carpet fiber samples
for Experiment 9 (R-2-2-1) A15
14
-------
Acronyms and Abbreviations
AOC Articles of Commerce
APPCD Air Pollution Prevention and Control Division
BDL below detection limit
CAS# Chemical Abstract Service Registry Number
CD carpet detergent
CFR Code of Federal Regulations
CM carpet cleaning machine
CRT Carpet and Rug Institute
CT carpet treatment
DCC daily calibration check
FCBR front corner bedroom
FtPLC high performance liquid chromatography
TAP internal audit program
DDL instrument detection limit
IS internal standard
LC/MS/MS liquid chromatography/tandem mass spectrometry
LOQ limit of quantification
MBR master bedroom
MDL method detection limit
MQO measurement quality objectve(s)
MSDS material safety data sheet
NR not reported
NRMRL National Risk Management Research Laboratory
PFAA perfluoroalkyl acid
PFBA perfluorobutyric acid
PFC perfluorochemical
PFCA perfluorocarboxylic acid
PFOA perfluorooctanoic acid
PFOS perfluorooctanesulfonic acid
PFTeDA perfluorotetradecanoic acid
PFTrDA perfluorotridecanoic acid
PQL practical quantification limit
PTFE polytetrafluoroethylene
PIT polytrimethylene terephthalate
QAPP Quality Assurance Proj ect Plan
QC quality control
RCS recovery check standard
RIS recovery internal standard
RSD relative standard deviation
TPFCA total perfluorocarboxylic acids (the sum of all monitored PFCAs)
U.S. EPA United States Environmental Protection Agency
15
-------
1. Introduction
1.1 Background
The potential impacts of perfluorocarboxylic acids (PFCAs) on human health and the global
environment did not draw much attention until the turn of this century, when evidence of their
widespread presence in various environmental compartments appeared (Renner, 2001; Giesy and
Kannan, 2002). PFCAs have been detected in air, water, and soil (Boulanger et al., 2004; Stock et
al., 2007) and in a variety of wildlife around the world, including the United States (Kannan et al.,
2002), Arctic Canada (Martin et al., 2004), Europe, and the Mediterranean Sea (Giesy and Kannan,
2002). More recently, low levels of PFCAs have been found in various consumer products that are
either made from or treated with perfluorinated chemicals (Washburn et al., 2005; Guo et al.,
2009).
PFCAs persist in the environment due to their high stability, and there is evidence that they
bioaccumulate (Moody et al., 2002), a significant factor in their potential toxicity.
Perfluorooctanesulfonic acid (PFOS) and perfluorooctanoic acid (PFOA), two types of
perfluoroalkyl acids (PFAAs) that have been the most extensively studied PFCs to date, are
essentially ubiquitous in humans; they have been found in human blood and breast milk in the
United States (Tao et al., 2008; Calafat et al., 2006), China (So et al., 2006), and other countries
across Europe, South America, and Asia (Kannan et al., 2004). Toxicological studies indicate that
PFCAs cause developmental and systemic toxicity in laboratory animals (Kennedy et al., 2004;
Lau et al., 2004; U.S. EPA, 2005). In particular, PFOS has been shown to act as an endocrine
disrupter (Austin et al., 2003). The potential health risks associated with PFCAs have inspired
extensive research on the sources, transport, transformation, and distribution of these chemicals
and their precursors in environmental media, along with research into ways to reduce potential
health risks.
Despite significant progress thus far, researchers have yet to reach a consensus on the most
important routes by which the general population is exposed to these chemicals. In particular,
opinions differ on whether consumer products containing PFCAs are significant contributors to
overall exposure. For instance, a study in 2005 concluded that exposures to PFOA during
consumer use of the articles evaluated were not expected to cause adverse health effects in infants,
children, adolescents, or adults, or result in quantifiable levels of PFOA in human serum
(Washburn et al., 2005). In a study conducted in 2009 by Fromme et al., data from indoor
measurements in Canada and Norway were used to estimate the average daily intake of PFOA for
the general population in Western countries, and it was found that the inhalation of house dust
contributed only 0.6% of the average daily intake of PFOA and a maximum of only 8.2% of the
highest daily intake levels. By contrast, however, Tittlemier et al. (2007) identified treated
carpeting as an important source of PFOA exposure, second only to ingestion with food. Trudel et
al. (2008) agreed that the consumption of contaminated food is the most significant exposure
pathway for PFOA and that the ingestion of dust and inhalation of air containing PFOA is the
second most likely route of exposure in low- and intermediate-exposure scenarios. Their study also
found that direct, product-related exposure dominates in high-exposure scenarios in which
consumers have treated carpets in their homes or regularly use PFCA-containing products, such as
stain-protection sprays. It is apparent, then, that the scarcity of data related to indoor PFCA sources
16
-------
and exposures contributes to significant uncertainty and differences of opinion about the most
prevalent exposure routes for these compounds.
Elevated levels of PFCAs have been detected in house dust in Japan (Moriwaki et al., 2003),
Canada (Kubwabo et al., 2005), and the United States (Strynar and Lindstrom, 2008), strongly
suggesting the presence of significant indoor sources. Kubwabo et al. correlated the percentage of
indoor carpet, in particular, to increased PFCA levels in dust. In 2005, Washburn and colleagues
reported the PFOA content in 14 groups of articles based on theoretical calculations and analytical
measurements. Of these groups, it was found that pre-treated carpet and carpet treated with carpet-
care solution had the highest PFOA loadings, i.e., 0.2 to 0.6 mg and 0.2 to 2 mg of PFOA per kg,
respectively (Washburn et al., 2005). Previous research by Guo et al. (2009) examined 14 classes
of consumer products and also found pre-treated carpet and carpet treated with PFCA-containing
carpet-care solutions to have the largest PFCA source strength, i.e., approximately 70 mg of
PFCAs (C5 through C12) in a typical home. One study found that PFCA contamination was higher
on the inside than the outside of sampling films placed in windows (Gewurtz et al., 2009).
Gerwurtz's study also found higher PFCA contamination in houses with new carpet and in carpet
stores, suggesting that emissions from new carpets contribute to increased indoor PFCA levels.
Due to the highly stable nature of PFCAs, few mechanisms exist, short of eliminating sources
completely, by which PFCAs can be removed from the indoor environment. Given its high source
strength and duration, its potential contribution to indoor dust, and its close proximity to humans,
treated carpeting may contribute to human exposure to PFCAs directly, e.g., dermal contact and
hand-to-mouth transfer, and indirectly via the inhalation of suspended particles from treated carpet.
1.2 Goal and Objective
The main goal of this study was to quantify the efficiency of common carpet cleaning methods —
steam cleaning and hot water extraction — in the removal of PFCAs from residential and
commercial carpet that was manually treated with stain-protection solutions. Our aim in the study
was to determine if these cleaning techniques are viable methods for reducing indoor exposure to
PFCAs. These results will be valuable to policy makers and manufacturers for risk management
purposes and may be of particular interest to people who wish to reduce the levels of PFCA in their
household environments.
17
-------
2. Materials and Methods
2.1 Test Facility
Carpet cleaning tests were conducted in a research house located in Gary, NC. The research house
is a three-bedroom, ranch-style house with a crawl space, a central, forced-air heating system that
uses natural gas, and an electric air-conditioning system. The total floor area is 126 m2, and the
house has a total volume of approximately 300 m3. Additional information on the test house was
provided by Tichenor et al. (1990) and Sparks et al. (1991). The two rooms of the house employed
for this work were the front corner bedroom (FCBR) and the master bedroom (MBR), as shown in
Figure 2.1. For this study, the carpeted areas of the MBR and FCBR measured 13.4 m and 12.2
m , respectively.
Master
Master
Bedroom
Bath
Clos
Clos | Clos | Return
Air
Den
Clos
Corner
Bedroom
-,*
cios
Utility
Middle
Bedroom
Kitchen
I I
Living
Room
Instruments
Garage
mm = Registers
Figure 2.1. Floor plan for the research house in Gary, NC
The air exchange rate inside the house was monitored but not reported for these experiments. An
OPTO 22 data acquisition system (OPTO 22, Temecula, CA) continuously logged temperature and
humidity data in the FCBR and den. Air exchange rate, temperature, and humidity were not
controlled as critical parameters, but temperature and relative humidity were recorded and
reported. Additional conditions in the research house are listed in Table 2.1.
18
-------
Table 2.1. Conditions in the research house
Parameter
Exterior doors and windows
Interior doors
Ceiling fans
Heating/air conditioning fan
Heating/air conditioning registers
Status
Closed
Open
On low
On
Open
2.2 Test Materials
2.2.1 Carpet Selection for Research House Experiments
The following two carpet types were chosen for the research house experiments: 1) residential
carpet that was Green-Label Certified by the Carpet and Rug Institute (CRI) with a pile yarn
content of 100% polytrimethylene terephthalate (PTT) - a product from recycled plastic bottles -
with a textured cut pile and a woven polypropylene backing and 2) commercial carpet made with
Antron® fiber, a nylon 6,6, hollow-filament fiber with a soil-resistant treatment incorporated into
the fiber. The residential carpet fiber, also called Triexta®, shares the chemical structure of
polyester; however, unlike traditional polyester carpet, Triexta® has extreme durability, high stain
resistance, and a much softer texture. The fiber used for the commercial carpet is considered to be
one of the most durable in the industry, and it has been utilized extensively in schools and offices.
A solid urethane carpet pad was selected as the underpad for each type of carpet. Each of these
products was screened to determine the content of PFCAs prior to installation. A 1-g sample of
fiber from each type of carpet and 1 g of the solid memory foam backing carpet pad were analyzed
to determine their PFCA content before the experiments were initiated. All of the samples from all
of the products were found to contain less than 8 ng/g of PFCAs.
2.2.2 Professional Carpet Treatment Solutions
Two carpet stain-protection treatment solutions (CT-1 and CT-2) were selected using the following
criteria: (1) shown to contain high levels of PFCAs (greater than 10 times the PQL); and (2)
analytical data with good recovery for the recovery internal standards.
Table 2.2 lists the components of the selected treatment products as documented in each product's
material safety data sheet (MSDS). The documentation indicated that both treatments are
dispersible in water. The PFCA content in these treatment solutions are reported in Section 3.2
below.
19
-------
Table 2.2. Components of carpet treatments CT-1 and CT-2
Chemical
Polymethylmethacrylate (CAS#
9011-14-7)
NJ Trade Secret Registry #00850201001-5155?
NJ Trade Secret Registry #00850201001-5259?
Citric Acid (CAS# 77-92-9)
Water (CAS# 7732-18-5)
Content (%)
CT-1
1.3-2.5
1-1.5
0-1.3
0.5-1
90-97
CT-2
4.7
-
-
-
95
2.2.3 Carpet Cleaning Machines
Two carpet cleaning machines were chosen to clean the carpets after treatment with a carpet stain-
protection treatment: 1) a residential cleaning machine (CM-1), the Rug Doctor Mighty Pro model
MP C-20 and 2) a portable professional steam cleaner (CM-2), the Century 400 Ninj a Warrior.
Figure 2.2 shows the Rug Doctor (front) and the Ninj a Warrior (rear). The residential machines,
which can be rented at most grocery and home improvement stores, use hot tap water for the
extraction process with no additional heating during the cleaning process. A 28-psi (1.93 x 105 Pa)
vacuum pump is used to extract the applied water, and a brush vibrates during the process. The
professional steam cleaner has an 1850-watt, in-line heater. Steam at a pressure of approximately
150 psi (1.03><106 Pa) soaks the carpet fibers, and dual 2-stage vacuum motors extract the residue
from the carpet. The professional cleaning machine does not agitate the carpet fibers.
Figure 2.2. Residential and commercial carpet cleaning machines
20
-------
2.2.4 Carpet Cleaning Detergents
Two detergents were selected to test the removal of PFCAs from carpets, i.e., a residential carpet
cleaning detergent (CD-2) recommended for use with the residential machine (CM-1) and a
commercial carpet cleaning detergent (CD-I) chosen for use with the commercial machine (CM-2).
The residential detergent was purchased at a local grocery store, and the commercial detergent was
purchased from a local distributor of janitorial supplies. According to the material safety data
sheets (MSDSs), CD-I contains 3 to 6% of dipropylene glycol methyl ether (CAS# 34590-94-8)
while CD-2 contains 1 to 5% of sodium 2-ethylhexyl sulfate (CAS# 126-92-1) and <1% of
branched tridecylalcohol (CAS# 69011-36-5). Prior to use, each cleaner was screened for PFCAs.
The residential detergent measured most PFCAs below the quantification limits of the instrument,
with perfluorotridecanoic acid (Co) having the highest concentration at 3 ng/mL. The commercial
detergent had only trace levels of Cg through Cio, and all were below the practical quantification
limits.
2.3 Experimental Design
All experiments were conducted at the research house in Gary, NC. The FCBR was used primarily
for the residential carpet, and the MBR was used for both residential and commercial carpet. The
residential carpet was tested with both types of cleaning machines and detergents to represent what
might be used in a typical household, while the commercial carpet was tested with only the
commercial cleaning machine and the commercial detergent. An initial scouting test (Experiment
3) was performed with the residential carpet and CM-1 to finalize testing procedures before
beginning the planned experiments. To ensure the presence of detectable PFCAs, all experiments
involved the application of a commercial carpet stain-protection treatment solution containing
PFCAs (CT-1 or CT-2). Following the application of the treatment and a subsequent 48-hour
drying period, a series of three carpet cleanings was performed. A drying period of at least 48
hours followed each cleaning before carpet samples were collected. The first set of tests used only
hot water or steam for the cleaning process to determine the efficiency of each carpet cleaning
machine's method of removing the applied PFCAs. Additionally, four experiments, one of which
was a duplicate test, were conducted using cleaning detergents (CD-I or CD-2) in conjunction with
a carpet cleaning machine to assess any additional PFCA removal. The complete experimental test
matrix is summarized in Table 2.3.
21
-------
Table 2.3. Experimental matrix for testing of carpet care treatments
Experiment
1
2
3a
4
5
6
7
8b
9
Research House
Location
MBR
MBR
FCBR
FCBR
FCBR
MBR
FCBR
MBR
FCBR
Carpet
Typec
C
C
R
R
R
C
R
R
R
Treatment
Type
CT-1
CT-2
CT-1
CT-1
CT-2
CT-2
CT-1
CT-1
CT-2
Cleaning
Machine
CM-2
CM-2
CM-1
CM-2
CM-2
CM-2
CM-1
CM-1
CM-2
Detergent
None
None
None
None
None
CD-I
CD-2
CD-2
CD-I
a Initial scouting test.
b Duplicate of Experiment 7.
0 C = Commercial; R = Residential.
2.4 Test Procedure
2.4.1 Layout of Carpet Sampling Sections
The carpet in the MBR measured 3.66 x 3.66 meters, and the carpet in the FCBR measured 3.66 x
3.20 meters. For each room, a grid was created in order to evenly gauge five sampling sections, A-
E, as shown in Figure 2.3. Once each section was located, a small piece of tape the size of a dime
was placed in its center. This was the only indicator placed on the carpet itself and acted as a
reference point for sampling (Section 2.5.1).
22
-------
AT
BT
ET
CT
DT
Figure 2.3. Diagram of generic carpet sampling quadrant
2.4.2 Dilution and Application of the Carpet Treatment
The dilution of the carpet treatment varied slightly for each set of tests. The treatment (CT-1)
dilution for the scouting test (Experiment 3) was 1:1 distilled water to treatment solution to ensure
measurable concentrations of PFCA through three cleanings. The manufacturer's recommendation
was 4:1, and the resulting concentrations of PFCAs from the 1:1 dilution on the carpet indicated
that a lower concentration of application treatment would be more applicable for this study;
therefore, a dilution of 4:1 was used for all subsequent applications with both products. An exact
measurement for the dilution was not critical because the dried product on the carpet fiber was
collected and extracted for analysis before each test.
To start the application, the diluted carpet stain-protection treatment solution was placed in a
commercial sprayer recommended in the product use guide for application of the treatment. The
sprayer was modified by adding a pressure gauge to better control the application rate and
uniformity of the application. Then, the sprayer was inserted into the "application cart," an
apparatus manufactured in-house (Figure 2.4). The application cart was on wheels in order to
provide a more uniform application of the solution, and it held the nozzle of the sprayer
approximately 0.6 m from the ground.
23
-------
Figure 2.4. Application cart
First, the carpet treatment was applied according to Figure 2.5. For each path, the applicator was
pumped to a pressure of 10 psi (6.89* 104 Pa), the spray nozzle was turned on, and the cart was
pulled in the direction of the arrows for approximately 24 s. At the end of each path, the spray
nozzle was turned off. A grid on the wall allowed for proper path widths based on the swath of the
sprayer; the MBR had seven paths, while the FCBR had eight.
1
1
1
1
1
1
1
End
Poor)
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Start
1
1
1
1
1
1
1
1
Figure 2.5. Diagram of first application process
24
-------
Once the application shown in Figure 2.5 was completed, a second, S-shaped application was
performed, as shown in Figure 2.6. During this application, the process was not stopped at the end
of each path. Instead, the application was done in a continuous motion, only stopping to pump the
applicator to 8-10 psi (5.52x 104 - 6.89x 104 Pa) when necessary. After application the carpet was
allowed to dry for at least 48 hour before carpet samples were collected.
Start
End
Figure 2.6. Diagram of second application process
2.4.3 Cleaning of Carpets
The manufacturer's instructions were followed for each carpet cleaning machine. When using the
commercial cleaner (CM-2), the carpets were cleaned in paths that were approximately 3-3.5 ft
(0.91-1.07 m) wide with about 1 in (2.54 cm) of overlap. Figure 2.7 demonstrates this process,
where each section of the schematic represents a path, and the arrows indicate the direction of
cleaning. For each path, there were four passes with the carpet cleaning apparatus with the cleaning
solution being expelled and vacuumed up in the first pass and any remaining solution being
vacuumed up in the final three passes, which, in turn, dried the carpet. Each 3-ft pass was timed for
approximately 10 seconds. The residential cleaner (CM-1) followed a similar cleaning process,
except that 6-ft-wide paths were used.
25
-------
1
1
1
1
1
1
1
1
1
End
(Door)
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Start
1
4
i
4-
I
4
4
4
4
4
4
4
Figure 2.7. Diagram of carpet cleaning using commercial cleaner CM-2 (not to scale)
2.5 Sampling Methods
2.5.1 Collection of Carpet Samples
A carpet sample from each of the five sampling locations (Figure 2.3) was collected according to
Table 2.4, using a 82-mm-diameter carpet cutter (05-174 Instant Repair Tool, Grain Cutter
Company, Milpitas, CA). Methanol was used to clean the tool between samples to prevent cross-
contamination. Each sample was placed in an individually-labeled plastic bag and transferred to the
EPA campus for processing and analysis. Each 82-mm-diameter carpet sample was processed by
removing the fibers from the backing and placing them in an aluminum foil dish. Each set of carpet
samples was placed in a separate dessicator. After a minimum of 16 hours, the fiber sample was
removed and mixed. A 1-g sub-sample was transferred to a 50-mL polypropylene centrifuge tube
(Evergreen Scientific, Los Angeles, CA), and a second 1-g sub-sample was weighed and placed in
a plastic bag. Once all individual samples from the stage had been processed, the plastic bag
containing the remaining 5 g of carpet from quadrants A-E were used to compile five composite
fiber samples. The bag was shaken to ensure that all fibers were distributed evenly, and five
individual 1-g sub-samples of composite carpet fiber were removed from the plastic bag and
placed in separate centrifuge tubes. In summary, each stage of the experiment produced a total of
10 samples, each weighing 1 g, i.e., five individual fiber samples (A-E) and five composite fiber
samples composed of fibers from all five sampling locations. The individual sample data were used
only to evaluate the uniformity of the application of the carpet treatment, while the composite
samples were used to evaluate cleaning efficiency.
26
-------
Table 2.4. Carpet fiber collection strategy
Stage of Experiment
Prior to Treatment
Application (Initial
Sampling)
Post-Treatment
Application
(Application Sampling)
Post-cleaning, Round 1
(Clean 1 Sampling)
Post-cleaning, Round 2
(Clean 2 Sampling)
Post-cleaning, Round 3
(Clean 3 Sampling)
Number of Samples
Collected from Carpet
5
5
5
5
5
Number of Fiber
Samples Processed for
Analysis a
5
5
5
5
5
Number of Composite
Fiber Samples
Processed for Analysis b
5
5
5
5
5
One ~l-g sample per sampling location.
' Five ~l-g samples taken from an equal mixture of fiber samples from all five locations.
Figure 2.8 shows a residential carpet after all sampling stages were completed.
Figure 2.8. Residential carpet after all sampling stages
2.5.2 Wipe Sampling
Wipe samples were collected only during Experiment 3, the scouting test, to evaluate over-spray
during the application process. These data illustrate any possible exposure from wall surfaces in a
typical home. Using ASTM Method D 6661, Field Collection of Organic Compounds from
Surfaces Using Wipe Sampling (ASTM, 2010), as a general guideline, a 10 cm x 10 cm template
27
-------
was used to outline the wipe area using painters tape. Four wipe areas were sampled on each of the
four walls in the center of the carpet length, four inches above the baseboard molding. For each
wipe sample, approximately 2 mL of methanol were applied to a piece of cotton gauze. The gauze
was pressed firmly to the wall, and the sample area was wiped vertically with minimal overlap
between strokes. Then, the sample area was wiped horizontally. The gauze used for the wipe was
stored in a 50-mL polypropylene centrifuge tube and later extracted using the same procedure that
was used for the residential carpet samples, which is described in Section 2.6.1. Background wipe
samples were collected at each location before application of the carpet treatment.
2.6 Sample Analysis
2.6.1 Extraction of Residential Carpet Samples
To each of the residential fiber samples prepared in Section 2.5, 45 mL of HPLC-grade methanol
(pre-screened for PFCAs) and 100 uL of recovery check standard (2 ng/uL each of perfluoro-«-[l,
2-13C2] decanoic acid Wellington Laboratoies, Canada) were added. The samples were extracted
for 24 ± 2 hr using a Nutating Mixer (Model VSN-5, PRO Scientific, Inc., Oxford, CT). The
extract was transferred to a 170-mL borosilicate glass blow-down tube. The original sample vial
was rinsed three times with ~3 mL of methanol, and each of the rinse liquids was transferred to a
170-mL concentration tube. A spatula was used to agitate the samples during rinsing. Next, the
sample was concentrated to approximately 1.5 mL in a heated (50 °C) nitrogen atmosphere by
using a RapidVap N2 Evaporation System (Model 791000, LabConco, Kansas City, MO), which
was previously modified at the factory to remove all PTFE parts and coatings.
The blow-down sample was transferred to a 10-mL volumetric flask through a 0.1-um Anotop
syringe filter (Whatman International, Madestone, England). The tube was rinsed five times with a
solution consisting of 60% (v/v) methanol and 40% (v/v) 2 mM ammonium acetate aqueous
solution (hereafter referred to as 60:40 solution). The rinse liquids were filtered and combined with
the blow-down sample in the volumetric flask. After adding 100 uL of the internal standard
solution (0.5 ng/uL each of perfluoro-w-[l, 2, 3, 4-13C/i] octanoic acid), the sample was brought
to 10 mL with 60:40 solution and sonicated for 10 min. The sample solution was transferred to a
15-mL polypropylene tube and stored at 4 °C until LC/MS/MS analysis.
2.6.2 Extraction of Commercial Carpet Samples
After extraction, due to clogging of the 0.1-um Anotop filter, the sample first had to be first filtered
through a Corning 0.22-um cellulose acetate, low-binding filter and into a polystyrene tube. The
170-mL concentration tube was rinsed five times, and the extract was passed through the 0.22-um
Corning filter. Then, the sample extract was transferred through a 0.1-um Anotop syringe filter into
a 10-mL volumetric flask. The filter tube was rinsed five times with 60:40 solution into the flask.
After adding 100 uL of the internal standard solution, the sample was brought to volume with
60:40 solution and sonicated for 10 min. The sample solution was transferred to a 15-mL
polypropylene tube and stored at 4 °C until LC/MS/MS analysis.
28
-------
2.6.3 Extraction of Liquid Samples
Approximately 1 mL of the liquid sample was weighed, spiked with 1 mL of a 2 ng/|iL recovery
check standard solution, and diluted to 10 mL with 60:40 solution. The diluted samples were
sonicated for 10 min and then filtered with a 0.1 -um Anotop syringe filter. After filtration, 1 mL of
the filtrate was transferred into a 10-mL volumetric flask and spiked with 100 jiL of the internal
standard. The resulting solution was sonicated for 10 min, transferred to a polypropylene tube, and
stored at 4 °C until LC/MS/MS analysis. Note: for samples with levels of PFCAs above the
calibration range, a second dilution of 1:10 was needed before adding the recovery check standard.
2.6.4 QC Sample Preparation
A set of five quality control (QC) samples was prepared for every batch of carpet extractions, i.e., a
field blank, a solvent blank, and three recovery internal standard blanks. The field blank consisted
of 45 mL of methanol and 100 uL of recovery internal standard, and it went through the same
extraction process as the residential carpet samples described in Section 4.6.1. The solvent blank
consisted of 60:40 solution with 100 uL of internal standard, brought to volume in a 10-mL
volumetric flask. Each recovery internal standard blank contained 60:40 solution with 100 uL
internal standard and 100 uL recovery internal standard, brought to volume in a 10-mL volumetric
flask. All samples including field blank samples were sonicated, transferred to 15-mL
polypropylene tubes, and stored at 4 °C with the corresponding batch of extraction samples until
LC/MS/MS analysis. The analyte content in the solvent blank was subtracted from all samples and
field blanks if it exceeded the practical quantification limit (PQL). The analyte content in the field
blanks was required to be below the PQL.
2.6.5 Sample Analysis
Sample quantification was conducted using an Agilent 1100 HPLC equipped with an Applied
Biosystems API 3200 Triple Quadrupole Mass Spectrometer with a Turbo V ion-spray interface. A
CIS reversed-phase guard column and analytical CIS reversed phase column were used for analyte
separation. Samples were injected at a flow rate of 0.250 mL/min and maintained at 50 °C. The
initial gradient mobile-phase composition was 25% mobile phase B, where mobile phase A was
95% 2 mM aqueous ammonium acetate-5% methanol, and mobile phase B was 95% methanol-5%
2 mM aqueous ammonium acetate, held for 0.5 min. A linear gradient was used from 25% to 85%
B over 4.5 min, then increased to 100% B over 0.10 min and held for 2 min. Then, a linear gradient
decreased mobile phase B to 25% over 2 min, where it was held for 3 min. PFCAs were observed
in the negative ion mode, and both primary and secondary ion transitions were collected for each
analyte. The instrument was calibrated for 11 PFCA homologues (Table 2.5) plus the recovery
check standards at eight concentration levels in the concentration range of 0.3 to 100 ng/mL with
triplicate injections. This procedure followed methods detailed in Liu et al. (2009).
29
-------
Table 2.5. Analyte names, abbreviations, chemical formulas, molecular weights (g/mol), and
Chemical Abstracts Service registration numbers (CAS #)
Analyte
Perfluorobutyric acid
Perfluoropentanoic acid
Perfluorohexanoic acid
Perfluoroheptanoic acid
Perfluorooctanoic acid
Perfluorononanoic acid
Perfluorodecanoic acid
Perfluoroundecanoic acid
Perfluorododecanoic acid
Perfluorotridecanoic acid
Perfluorotetradecanoic acid
Perfluoro-«-[l,2-13C2]
hexanoic acid a
Perfluoro-«-[l,2,3,4-13C4]
octanoic acid b
Perfluoro-«-[l,2-13C2]
decanoic acid a
Abbreviation
PFBA - C4
PFPeA - C5
PFHxA - C6
PFHpA - C7
PFOA - C8
PFNA - C9
PFDA-C10
PFUdA-Cll
PFDoA-C12
PFTrDA-C13
PFTeDA-C14
13C-PFHxA
13C-PFOA
13C-PFDA
Formula
C4HF7O2
C5HF9O2
C6HFn02
C7HF1302
C8HF1502
C9HF1702
CioFfFi9O2
CnHF2102
C12HF23O2
C13HF2502
Ci4HF27O2
C2 C4HF]]O2
13C412C4HF15O2
13C212C8HF15O2
MW
214.04
264.04
314.05
364.05
414.06
464.07
514.07
564.08
614.09
664.11
714.12
316.04
418.03
516.07
CAS#
375-22-4
2706-90-3
307-24-4
375-85-9
335-67-1
375-95-1
335-76-2
2058-94-8
307-55-1
72629-94-8
376-06-7
n/a
n/a
n/a
a Recovery check standard.
b Internal standard.
2.7 Quality Assurance and Quality Control
A Category II quality assurance project plan (QAPP) was developed and approved before the start
of the project and detailed the Measurement Quality Objectives (MQOs) for the project, which are
summarized in Table 2.6. For this study conducted at the EPA research house, the data quality
objective for the precision of solvent extraction was relaxed to ± 30% to incorporate more
information to better interpret the desired goal of this project. All data that fell outside the original
MQOs listed below are highlighted in the data tables presented in Appendix A.
30
-------
Table 2.6. Measurement quality objectives
Measurement
Calibration of
LC/MS/MS
PFCA quantification
by LC/MS/MS
Solvent extraction
Weight of AOC samples
Parameter
Coefficient of
determination (r2) for
calibration curve
Instrument
detection limit
Accuracy
Precision
Agreement of primary
and secondary ions
Precision
Accuracy
System blank
Accuracy
Objective
>0.99
<0.2 ng/mL
85-115%
20%
25%
20% a
80-120%
-------
1OUUU -
-?r IzUUU •
M)
=
K
<
^T cnnn
u. oUUU
o-
H
4*
M)
«
aj innn
SJ 4UUU
•^
fi -
]
1
-
1
1
I
-L _L_ ~T
i -±-
123456789
Experiment
Figure 2.9. Average of individual quadrant samples for post-application PFCA concentrations in each
experiment and associated standard deviations
2.7.2 Calibration ofLC/MS/MS
The instrument was calibrated for 11 PFCA homologs (Table 2.5) plus the recovery check
standards at eight concentration levels in the concentration range of 0.3 to 100 ng/mL with
triplicate injections. The instrument detection limits (DDLs) for individual PFCAs in the injected
sample were < 0.2 ng/mL. The practical quantification limit (PQL) for the injection sample was < 1
ng/mL, which is equivalent to 1 ng/g for carpet extracts and 10 ng/g for liquid samples.
Initially, standards were prepared from individual analytes (Liu et al., 2009). However, in March
2010 calibrations were performed using a composite standard mix distributed by the manufacturer
that contained the all of the analytes of interest. This change was due to the availability of a
composite standard and the resulting simplification of the standard preparation process.
The acceptance criterion for the calibration curve was that the coefficient of determination (r2) be
no less than 0.99. The average of the r values for 14 compounds over seven calibrations was 0.995
± 0.003, giving 100% completeness for the MQO for calibration.
An internal audit program (LAP) standard was analyzed for each of the seven calibrations. The LAP
standard contained at least three of the calibrated PFCAs using a different chemical source
(Oakwood Laboratories or Aldrich) and was prepared by someone other than the analyst who
prepared the calibration standards. The analyst who conducted the calibration received the TAP
standard without knowing the concentrations. LAP standards were analyzed after each calibration
as a measurement of calibration verification. The average of the percent accuracy for each analyte
32
-------
ranged from 76% to 115%. Only two data points out of 26 did not meet the MQO criterion for
acceptance of ± 15%, giving 92% completeness for this QA criterion.
2.7.3 Daily Calibration Checks
Daily calibration check (DCC) standards, approximately 10 ng/mL for each analyte, were analyzed
to evaluate the performance of the LC/MS/MS. The DCC was conducted at the beginning of each
analytical sequence and at the end of the sequence or, for extended sequences, after 24 hours of
instrumental analysis. Analytical results of a sample batch were considered acceptable only if the
percent recovery of the DCC was within 100 ± 15% and the percent relative standard deviation
(%RSD) of triplicate injections of the DCC was within ± 15%. The MQO for DCC recovery was
relaxed for this study to 100 ± 30%, and the percent relative standard deviation (%RSD) of
triplicate injections of the DCC was instead required to be within ± 30%. Data that fell outside the
original MQO for DCCs are highlighted in the data tables in Appendix A. Data that fell outside the
relaxed MQOs were not used in this report. We collected 952 data points for the DCC
measurements with only 55 points falling outside the relaxed MQO, giving 94% completeness
using the relaxed MQO.
2.7.4 Contamination Checks
All purchased and prepared solvents, glassware, and the HPLC system were checked routinely for
PFCA contamination. Also, a solvent blank was prepared with each set of standards and samples to
assess the solvent and the HPLC system. Solvents and blanks were rejected if they contained the
analytes of interest at concentrations higher than the DDL for individual PFCAs in the injection
sample of <0.2 ng/mL.
2.7.5 Weight Measurements
Two balances were employed for weight measurements during this project, i.e., 1) a five-place
microbalance and 2) a three-place pan balance. The microbalance was used early in the project for
weighing the calibration standards for the LC/MS/MS. Both balances are calibrated every 12
months. To monitor the daily performance of the balances, two Class A weights were weighed at
the beginning and end of each weigh session. The precision of each balance for this proj ect was
determined by averaging the measurements for each class A weight and determining the standard
deviation for those measurements. The accuracy was determined by comparison of the class A
weight measurement to the certified value of the weight. Table 2.7 presents these MQO
measurements.
Table 2.7. MQOs for weight measurements
Parameter
Precision (n > 10)
Accuracy %
Microbalance
10 mg: 9.997 ± 0.038 mg
20 g: 19.999 ± 0.00014 g
10 mg: 99.97%
20 g: 99.99%
Pan Balance
Ig: 1.000 ± 0.001 g
10 g: 10.002 ± 0.001 g
1 g: 100.05%
10 g: 100.01%
33
-------
3. Results
3.1 Summary of Experimental Conditions
The research house conditions for the nine experiments are summarized in Table 3.1 .The relative
humidity and air exchange rate were measured but not controlled during the research house
experiments. After each application and subsequent cleaning, the relative humidity was elevated
for several hours during the drying period for the carpets. The temperature was controlled by the
conventional heating and air conditioning system in the house. The HVAC fan was operated
continuously, and the ceiling fans in the FCBR, MBR, den, and living room were kept on low
speed, blowing upward. Measurements for relative humidity and temperature were monitored in
the FCBR and den by a Vaisala INTERCAP humidity and temperature transmitter recording to an
OPTO DAS. The air exchange rate inside the house was not reported for any of the experiments.
The only purpose of presenting the environmental data in Table 3.1 is to show that the carpet
cleaning experiments were conducted under typical indoor environmental conditions. The data was
not used anywhere else in this report.
Table 3.1. Environmental parameters recorded during the cleaning experiments
Experiment a
1 and 4
2 and 5
3
6 and 9
7 and 8
Temperature, °C b
Average c
22±ld
24 ± 0.5e
22±ld
24 ±r
21±ld
19 ±r
21±ld
23±0.5e
NAf
High
24
25
24
26
26
22
23
25
NAf
Low
20
22
20
22
19
18
19
21
NAf
Relative Humidity, %b
Average c
59 ±6
54 ±4
61±4
56 ±2
24 ±6
25 ±4
59 ±6
52 ±4
NAf
High
77
65
75
65
58
40
75
63
NAf
Low
48
44
55
51
14
18
47
42
NAf
a Tests are grouped together by the start date for each test (one test being conducted in the FCBR and
the other in the MBR). Test 3 was the initial scouting test conducted in the FCBR.
b Measurement locations: front corner bedroom (FCBR) and den.
0 Mean ± standard deviation for n > 2500 (Sample number varied with each experiment).
d Transmitter in the FCBR.
e Transmitter in the Den.
f Not available — Data lost due to power failure at the house.
3.2 Extractable PFCA Content in Carpet Treatment Solutions
The levels of PFCAs quantified in the undiluted carpet protector treatments CT-1 and CT-2 are
presented in Figure 3.1. The error bars show ±1 standard deviation with n = 4 for CT-1 and n = 6
34
-------
for CT-2. The treatments were analyzed for 11 PFCAs; however, the levels of PFTeDA (C14) in
the treatment solutions were below the limit of detection. The total PFCA content for CT-1 and
CT-2 was, respectively, 6360 and 7500 ng/g.
2500
o
U
1500
1000
500
t
I
r
t
-
Figure 3.1. Extractable PFCA content in carpet treatment solutions CT-1 and CT-2
(The error bars represent ±1 SD; n = 4 for CT-1 and n = 6 for CT-2)
3.3 Extractable PFCA Content in Wipe Samples Taken from the Walls
The results from the wipe samples collected during Experiment 3 to evaluate overspray during the
application process are presented in Table 3.2. Background wipe samples were collected from each
of the four locations prior to the application of the carpet stain-protection treatment. Three of the
four background samples showed no evidence of PFCAs on the walls prior to application. The
fourth background sample was lost during analysis. The concentration of PFCAs measured on the
wall varied significantly from location to location, ranging from 2.3 ng/cm2 to 47.4 ng/cm2. This
exposure source is minimal when compared to the post-application concentration on the carpet of
approximately 4000 ng/cm2 of carpet area, as measured from the protectant application rate on the
carpet.
35
-------
Table 3.2. Results of wall wipe samples collected during Experiment 3
Wipe Samples
PFBA-C4
PFPeA-C5
PFHxA-C6
PFHpA-C7
PFOA-C8
PFNA-C9
PFDA-C10
PFUnDA-Cll
PFDoDA-C12
PFTrDA-C13
PFTeDA-C14
Total PFCA, ng/cm2
Perfluoro-«-[l,2-13C2] decanoic acid
(recovery check standard)
Wipe A,
ng/cm2
0.8
0.5
2.5
6.9
3.8
4.5
2.9
2.4
1.8
1.2
«b
29.1
113%
Wipe B,
ng/cm2
1.0
0.7
3.3
8.6
4.7
5.9
3.9
3.2
2.2
1.4
«>b
36.8
117%
Wipe C,
ng/cm2
0^-b
0^b
0.2
0.4
0.2
0.3
0.2
0.1
&2-b
OJ-b
0^b
2.3
120%
Wipe D,
ng/cm2
1.4
1.0
5.3
10.9b
6.1
7.8
4.9
4.1
2.7
1.5
3£
47.7
121% a
Did not pass MQO of ± 20% but is presented with the relaxed MQO.
Hft = values below quantification limit; nn = values above calibration range.
3.4 Extractable PFCA Content in Carpet Samples
The concentrations of individual PFCAs and total PFCA in composite carpet fiber samples for
Experiments 1 through 9 are presented in Appendix A, Tables A.I through A.9. The results of total
PFCAs are presented graphically in this section in Figures 3.2 through 3.10. The composite
samples represent fibers collected from each of the five sampling locations compiled into one
sample and then divided into five replicate samples. The data in each graph represent the reportable
data using both the QAPP standard MQOs and the relaxed MQOs from each experiment. The
relaxed data are highlighted in the data tables in Appendix A. The error bars represent the
variability between replicate samples and are equal to ± one standard deviation.
In the following graphs, each experiment is coded with the parameters of the experiment as
presented in Table 4.3. The resulting code is presented as: Type of carpet: R = residential or C =
commercial; carpet stain-protection treatment: 1 or 2; carpet cleaning machine: 1 or 2; and carpet
detergent: 0 = no detergent, 1 or 2 [i.e., Experiment 1 (C-l-2-0) = Commercial carpet, CT1, CM1,
and no detergent].
It was noticed that the commercial carpet (see Figures 3.2, 3.3, and 3.7) retained more PFCAs after
treatment than the residential carpet did.
36
-------
ouuu -
"^ ^nnn
O.
•_
es
u
o* /mnn
VI
3
u
fe "?nnn
S/j
^G,
5 onnn
'-^
«
•_
-^
aj i nnn
a
o
U
n -
,
T
1
Pre-Application Post-Application Post-Cleaning Post-Cleaning Post-Cleaning
Round 1 Round 2 Round 3
Figure 3.2. Average total PFCAs in composite carpet fiber samples for Experiment 1 (C-l-2-0)
10000
Oil
OK
8000 -
•B 4000 -
O
c
<=> 2000 --
I
1
T
1
T T
~r 1
Pre-Application Post-Application Post-Cleaning Post-Cleaning Post-Cleaning
Round 1 Round 2 Round 3
Figure 3.3. Average total PFCAs in composite carpet fiber samples for Experiment 2 (C-2-1-0)
37
-------
tuuu -
-4^
0>
*• ^nnn
cS JUUU
u
S/j
(/3
«<
r innn
ft< zUUU
0.
S/j
o
• rt 1 A A A
I!oncentrai
c
o c
I
I
I.
,
Pre-Application Post- Post-Cleaning Post-Cleaning Post-Cleaning
Application Round 1 Round 2 Round 3
Figure 3.4. Average total PFCAs in composite carpet fiber samples for Experiment 3 (R-l-1-0)
JUUU
-4^
O.
O
zUUU
r i ^nn
U. 1DUU
0.
OK
a
s— ^ i fifin
(j 1UUU
o
e
u
c
^ n
^J U H
1
i I i
1
i— l-i
1
1
1
Pre-Application Post-Application Post-Cleaning Post-Cleaning Post-Cleaning
Round 1 Round 2 Round 3
Figure 3.5. Average total PFCAs in composite carpet fiber samples for Experiment 4 (R-l-2-0)
38
-------
tuuu -
o.
•_
Uoflflfl
S/j
0-H OflflfJ
c
c
o
•_
"g 1000 -
-------
tuuu
-fcj
nnn
OJ) ZUUU
o
?"! i nnn
•*J 1UUU
-------
U
Ml
Ml
-*j
u
u
o
u
5000
4000 -
3000 -
1000 -
Pre-Application Post-Application Post-Cleaning Post-Cleaning Post-Cleaning
Round 1 Round 2 Round 3
Figure 3.10. Average total PFCAs in composite carpet fiber samples for Experiment 9 (R-2-2-1)
[Re-use of carpet from Experiment 5]
3.5 Percent Removal of PFCAs by Cleaning
For each experiment, the percent removal of total PFCAs after the final cleaning was calculated by
comparison of the total PFCA concentration after application of the carpet treatment to the total
PFCA concentration after the third round of cleaning. These data are presented in Appendix A,
Tables Al through A9. Figure 3.11 is a graphic representation of these results.
41
-------
100%
«
o
S
S
80% -•-
60%
0%
123456789
Experiment
Figure 3.11. Calculated percent removal of total PFCAs for each experiment
42
-------
4. Discussion
4.1 Removal Efficiency of Speciated Extractable PFCAs in Carpet Samples
A relative comparison of the post-application concentration of the speciated carpet treatment to the
concentration after the third cleaning for each experiment as percent removal of individual
compounds PFPeA-C5 through PFDoDA-C12 is presented in Table 4.1. PFBA-C4 and PFTrDA-
C13 are not presented due to incomplete data sets. Three of the four experiments with detergent (7,
8, and 9) and two of the experiments without detergent (4 and 5) present a clear trend that the
higher removal efficiency was associated with the lower molecular weight compounds rather than
high molecular weight compounds. This trend is confirmed in the comparison of duplicate
experiments 7 and 8. However, in experiments 1, 2, 3, and 6, the trend indicated that the removal
efficiencies were more uniform. No clear conclusion about speciated removal efficiency could be
determined from these data.
Table 4.1. Percent removal efficiency of speciated compounds
Experiment
PFPeA-C5
PFHxA-C6
PFHpA-C7
PFOA-C8
PFNA-C9
PFDA-C10
PFUnDA-Cll
PFDoDA-C12
1 (C120)
29%
29%
28%
44%
25%
28%
3%
19%
2 (C220)
63%
70%
62%
67%
70%
64%
59%
59%
3 (R110)
57%
70%
58%
40%
38%
59%
44%
49%
4 (R120)
100%
100%
52%
27%
18%
22%
5%
14%
5(R220)
NR
89%
76%
38%
39%
84%
4%
NR
6 (C221)
60%
70%
64%
73%
81%
81%
NR
89%
7 (R112)
89%
94%
80%
47%
13%
73%
4%
43%
8 (R112)
79%
91%
73%
43%
30%
64%
23%
20%
9 (R221)
84%
91%
81%
61%
64%
58%
79%
70%
4.2 Overall PFCA Removal Efficiency
For the nine cleaning tests conducted, the overall PFCA removal efficiency after three rounds of
cleaning ranged from 26% to 76% (Table 4.2). Repeated cleanings with and without detergent did
not return either carpet to its pre-application concentration of < 8 ng/g of TPFCAs. On average,
each round of carpet cleaning removed approximately 20% of the total PFCAs. The overall
removal efficiency of the PFCAs could be attributed to the differences in the carpet fibers of each
of the products. Although there was high variability in the percent of PFCAs removed in each
round of cleaning (Appendix A), certain trends were evident. In general, the first round of carpet
cleaning resulted in the largest decrease in PFCAs. The decrease in PFCAs from the second to the
third cleaning continued to show removal of PFCAs, albeit with less efficiency, almost indicating a
leveling off of PFCA removal with each cleaning system.
43
-------
Table 4.2. Composite carpet sample data summarizing average percent reduction in
PFCAs by experiment
Experiment
1
2
3
4
5
6
r
8a
9
Experimental ID
C- 1-2-0
C-2-2-0
R-l-1-0, scouting
R- 1-2-0
R-2-2-0
C-2-2-1
R-l-1-2
R-l-1-2
R-2-2-1
Total Removal, %
26%
64%
52%
45%
63%
76%
58%
46%
70%
' Duplicate experiments.
4.3 Comparison of Duplicate Experiments
Experiments 7 and 8 were performed using the same residential carpet, treatment, cleaning
machine and detergent and represents the only duplicate experiment that was conducted.
Experiment 8 had slightly higher initial PFCA concentrations than Experiment 7. Analysis for
average PFCA removal following each round of cleaning showed less than 30% variance between
the duplicate experiments, which was within the ± 30% deviation indicated in the relaxed MQOs
for this investigation, suggesting that these data are reproducible.
4.4 Factors Affecting the Efficiency of PFCA Removal
To investigate the data more closely, average reductions in PFCAs were examined by experimental
variable. These data are presented in Table 4.3 and Figure 4.1. Although it is not possible to draw
statistically relevant conclusions due to the large number of variables relative to the number of
experiments and the cross-correlation of the variables, some preliminary conclusions can be drawn.
A detergent, either CD-I or CD-2, was used in Experiments 6 through 9. The detergent was used in
addition to the carpet cleaning machine potentially to enhance PFCA removal. As compared to
Experiments 1 through 4, the experiments that used detergent exhibited no statistically significant
increases in percent removals of PFCAs after three rounds of cleaning. As compared by detergent
(none vs. 1 or 2), CD-I does appear to have removed a greater amount of PFCAs, 73% as
compared to >50% for both CD-2 and no detergent, but percent removals for most individual
experiments ranged from 45% to 76% regardless of detergent use. A larger number of replicate
experiments would be needed to determine with greater certainty the effects of detergent
application.
44
-------
Table 4.3. Average percent reduction in PFCAs by experimental variable
for composite samples
Variable
Carpet
Treatment
Carpet Cleaning
Machine
Detergent
Type
Residential
Commercial
CT-1
CT-2
CM-1
CM-2
CD-I
CD-2
No Detergent
Total Removal
56%
55%
46%
68%
52%
57%
73%
52%
50%
I
u
to
CLH
o>
M
e«
l)
80% -
60% -
40% -
20% -
no/
U/o
...
Residential
O
B
>d
ft
r^-
Commercial
i
O
Prote
rfl
H
O
ctant
H
i
O
Carpet C
Mac
O
Cleaning
hine
1*1
_
Q
O
i
• —
•*•
(N
-------
The concentrations presented in these results represent PFCAs available via methanol extraction,
which may not equal the level of PFCAs available under actual normal use conditions for carpet.
For example, it has been shown that extraction of PFCAs from carpets using either water or sweat
(laboratory-prepared sweat simulant), which are more common household solvents than methanol,
recovered only about 10% to 65% of the PFCAs that were recovered with up to 24 hours of
methanol extraction (Mawn et al., 2005). As treated carpet ages, wear and fiber degradation may
also increase the availability of PFCAs through methods such as dust formation or transfer to other
surfaces.
4.5 Composite vs. Individual Samples
The analysis of individual quadrant samples was conducted only to show the uniformity of the
application of the carpet treatment, as presented in Section 2.7.1, Figure 2.9. Complete uniformity
of application is not possible with the manual application procedure that was recommended by the
manufacturer of the carpet treatment products. Thus, the discussion focused only on the data
obtained from composite carpet samples.
4.6 Extractable PFCA Content Collected from Walls
The collection of wipe samples from the wall surfaces was conducted during the initial scouting
experiment as a point of interest and does not concern the central discussion of PFCA removal
from carpets. However, these samples do demonstrate that PFCAs are, in fact, deposited on the
surface of the walls during the process of applying a carpet treatment (Table 3.2). Although the
reported concentration of PFCAs on the surfaces of the walls was several orders of magnitude
lower than that applied to the carpet, the presence of PFCAs on the walls after applying a carpet
treatment should still be noted.
46
-------
5. Conclusions and Recommendations
The efficiency of PFCA removal from treated carpet by hot water extraction and stream cleaning
was evaluated under close-to-realistic conditions in a research house. On average, each carpet
cleaning event removed approximately 20% of total PFCAs. At this removal efficiency, it takes
three, seven, and ten rounds of cleaning to remove, respectively, 50%, 80%, and 90% of total
PFCAs.
9
The average American home has about 1000 square feet (93 m ) of carpet. Previous research has
shown that treated carpets represent one of the largest sources of PFCAs in homes. Additionally,
PFCAs are highly stable, with the potential for a long residence time indoors, and can also bind to
house dust, making it difficult to remove the PFCAs from the indoor environment by any available
mechanisms. The use of a carpet cleaning machine may be a risk management option for people
who wish to reduce their indoor exposure to PFCAs. However, as mentioned above, this method is
only modestly effective in removing PFCAs from treated carpet.
Using the household hot water machine in four experiments with the residential carpet, reductions
of PFCA averaged between 45% and 58% after three rounds of cleaning. The removal efficiency
using the commercial cleaning machine on the commercial carpet had the highest variability. The
results from two experiments using the commercial cleaning machine with a steam extraction
indicated only a 26% reduction of PFCAs for carpet treatment 1 (CT-1), but there was a reduction
of 64% for carpet treatment 2 (CT-2). Differences in the carpet fibers of each product may have
also contributed to the overall removal efficiency of PFCAs. No significant difference in the
reduction of PFCAs was observed when detergent was added to the carpet cleaning machine. With
the commercial steam cleaner using a commercial detergent, a slight improvement was noted
during one experiment in which an effective removal of more than 70% of total PFCA was
obtained. However, the difference was not statistically significant.
Further work is needed in the following areas to better understand human exposure to PFCAs from
treated carpeting: (1) PFCA transfer mechanisms from carpets to indoor air, household surfaces,
and dust; (2) the relationship between PFCA-treated carpet and inhalation exposure with regard to
particle resuspension; (3) determination of the significance of dermal exposure; and (4) effective
risk management measures for reducing PFCA levels in houses with PFCA-treated carpet. It is also
recommended that market monitoring be continued to determine if all the carpet-treatment
products on today's market are virtually PFCA-free.
47
-------
Acknowledgments
We thank Andrew Lindstrom and Mark Strynar of the EPA National Exposure Research
Laboratory for technical consultation and assistance; Robert Wright and Joan Bursey of the EPA
National Risk Management Research Laboratory and Libby Nessley of ARCADIS for QA support.
48
-------
References
ASTM (2010). D6661-10 Standard practice for field collection of organic compounds from
surfaces using wipe sampling, ASTM International, West Conshohocken, PA.
Austin, M. E., Kasturi, B. S., Barber, M., Kannan, K, MohanKumar, P. S., and MohanKumar, S.
M. (2003). Neuroendocrine effects of perfluorooctane sulfonate in rats. Environmental Health
Perspectives, 111(12): 1485-1498.
Boulanger, B., Vargo, J., Schnoor, J. L., and Hornbuckle, K. C. (2004). Detection of
perfluorooctane surfactants in Great Lakes water. Environmental Science & Technology, 38(15):
4064-4070.
Calafat, A. M., Kuklenyik, Z., Caudill, S. P., Reidy, J. A., and Needham, L. L.(2006).
Perfluorochemicals in pooled serum samples from United States residents in 2001 and 2002.
Environmental Science & Technology, 40(7): 2128-2134.
Fromme, H., Tittlemier, S. A., Volkel, W., Wilhelm, M., and Twardella, D. (2009). Perfluorinated
compounds - exposure assessment for the general population in western countries. International
Journal of Hygiene and Environmental Health, 212: 239-70.
Gewurtz, S., Bhavsar, S., Crozier, P., Diamond, M., Helm, P., Marvin, C., and Reiner, E. (2009).
Perfluoroalkyl contaminants in window film: indoor/outdoor, urban/rural, and winter/summer
contamination and assessment of carpet as a possible source. Environmental Science &
Technology, 43: 7317-7323.
Giesy, J. P., and K. Kannan (2002). Perfluorochemical surfactants in the environment.
Environmental Science & Technology. 36: 147A-152A.
Guo, Z. Liu, X., Krebs, K. A., and Roache, F. N. (2009). Perfluorocarboxylic acid content in 116
articles of commerce, U.S. EPA, National Risk Management Research Laboratory, Research
Triangle Park, NC, Report No. EPA/600/R-09/033, 52 pp.
http://www.epa.gov/nrmrl/pubs/600r09033/600r09033.html
Kannan, K., Newsted, J., Halbrook, R.S., and Giesy, J.P. (2002). Perfluorooctanesulfonate and
related fluorinated hydrocarbons in mink and river otters from the United States. Environmental
Science & Technology, 36: 2566-2571.
Kannan, K., Corsolini, S., Falandysz, J., Fillmann, G., Kumar, K. S, Loganathan, B. G., Mohd,
M. A., Olivero, J., Van Wouwe, N., Yang, J. H., and Aldoust, K. M. (2004).
Perfluorooctanesulfonate and related fluorochemicals in human blood from several countries.
Environmental Science & Technology, 38(17): 4489-4495.
Kennedy Jr., J. L., Butenhoff, J. L., Olsen, G. W., O'Connor, J. C., Seacat, A. M., Perkins, R. G.,
Biegel, L. B., Murphy, S. R., and Farrar, D. G. (2004). The toxicology of perfluorooctanoate.
Critical Review of Toxicology, 34: 351-384.
49
-------
Kubwabo, C., Stewart, B., Zhu, J., andMarro, L. (2005). Occurrence of perfluorosulfonates and
other perfluorochemicals in dust from selected homes in the city of Ottawa. Journal of
Environmental Monitoring, 7: 1074-1078.
Lau, C., Butenhoff, J. L., and Rogers, J. M. (2004). The developmental toxicity of perfluoroalkyl
acids and their derivatives. Toxicology and Applied Pharmacology, 198: 231-241.
Liu, X., Krebs, K., Guo, Z., and Roache, N. (2009). Method development for liquid
chromatography/triple quadrupole mass spectrometer analysis of trace level perfluorocarboxylic
acids in articles of commerce. Journal of 'Chromatography A, 1216 (18): 3910-3918.
Martin, J. W., Smithwick, M. M., Braune, B. M., Hoekstra, P. F., Muir, D. C. G., and Mabury, S.
A. (2004). Identification of long-chain perfluorinated acids in biota from the Canadian Arctic.
Environmental Science & Technology, 38(2): 373-380.
Mawn, M., McKay, R., Ryan, T., Szostek, B., Powley, C., and Buck, R. (2005). Determination of
extractable perfluorooctanoic acid (PFOA) in water, sweat simulant, saliva simulant, and methanol
from textile and carpet samples by LC/MS/MS. The Analyst, 130: 670-678.
Moody, C. A., Martin, J. W., Kwan, W. C., Muir, D. C. G., and Mabury, S. A. (2002).
Monitoring perfluorinated surfactants in biota and surface water samples following an accidental
release of fire-fighting foam into Etobicoke Creek. Environmental Science & Technology, 36(4):
545-551.
Moriwaki, H., Takatah, Y., and Arakawa, R. (2003). Concentrations of perfluorooctane sulfonate
(PFOS) and perfluorooctanoic acid (PFOA) in vacuum cleaner dust collected in Japanese homes.
Journal of Environmental Monitoring, 5: 753-757.
Renner, R. (2001). Growing concern over perfluorinated chemicals. Environmental Science &
Technology, 35: 154A-160A
So, M. K., Yamashitam N., Taniyasu, S., Jiang, Q., Giesy, J. P., Chen, K., and Lam, P. K.
(2006). Health risks in infants associated with exposure to perfluorinated compounds in human
breast milk from Zhoushan, China. Environmental Science & Technology, 40(9): 2924-2929.
Sparks, L. E., Tichenor, B. A., White, J. B., and Jackson, M. D. (1991). Comparison of data from
an IAQ test house with predictions of an IAQ computer model. Indoor Air, 4: 577-592.
Stock, N. L., Furdui, V. L, Muir, D. C. G., and Mabury, S. A. (2007). Perfluoroalkyl
contaminants in the Canadian Arctic: Evidence of atmospheric transport and local contamination.
Environmental Science & Technology, 41(10): 3529-3536.
Strynar, M. J., and Lindstrom, A. B. (2008). Perfluorinated compounds in house dust from Ohio
and North Carolina. Environmental Science & Technology, 42: 3751-3756.
50
-------
Tao, L., Kannan, K., Wong, C. M., Arcaro, K. F., and Butenhoff, J. L. (2008). Perfluorinated
compounds in human milk from Massachusetts, U.S.A. Environmental Science & Technology.,
42(8): 3096-3101.
Tichenor, B. A., Sparks, L. E., White, J. B., and Jackson, M. D. (1990). Evaluating sources of
indoor air pollution. Journal of Air and Waste Management Association., 40: 487-492.
Tittlemier, S. A., Pepper, K., Seymour, C., Moisey, J., Bronson, R., Cao, X. L., and Dabeka, R. W.
(2007). Dietary exposure of Canadians to perfluorinated carboxylates and perfluorooctane
sulfonate via consumption of meat, fish, fast foods, and food items prepared in their packaging.
Journal of Agricultural and Food Chemistry, 55: 3203-3210 .
Trudel, D., Horowitz, L., Wormuth, M., Scheringer, M., Cousins, I. T., and Hungerbiihler, K.
(2008). Estimating consumer exposure to PFOS and PFOA. Risk Analysis, 28: 251-269.
U.S. EPA (2005). Draft risk assessment of the potential human health effects associated with
exposure to perfluorooctanoic acid and salts, U.S. EPA, Office of Pollution Prevention and
Toxics, http://www.epa.gov/oppt/pfoa/pubs/pfoarisk.pdf
U.S. EPA (2012). Perfluorooctanoic Acid (PFOA) and Fluorinated Telomers — Basic Information.
http ://www. epa. gov/oppt/pfoa/pub s/pfoainfo.html
Washburn, S. T., Bingman, T. S., Braithwaite, S. K., Buck, R. C., Buxton, L. M., Clewell, H. J.,
Haroun, L. A., Kester, J. E., Rickard, R. W., and Shipp, A. M. (2005). Exposure assessment and
risk characterization for perfluorooctanoate in selected consumer articles. Environmental Sciences
& Technology, 39(11): 3904-3910.
51
-------
Appendix A: Data
Data for total PFCA content in carpet fibers for all the experiments are presented below.
Both individual carpet fiber samples and composite fiber samples, five per type, were
collected at five stages of each experiment. These stages included before the application of
a carpet treatment solution, after the application of a carpet treatment solution, and after
each of three rounds of cleaning with a carpet cleaning machine. The PFCA concentrations
(as mean ± standard deviation) in composite carpet fiber samples for Experiments 1
through 9 are presented in Tables Al- A9. Averaged PFCA concentrations at each step for
individual samples are not presented, but post-application data are given in Tables A10-
A18 to demonstrate the lower data quality for individual samples and to give an indication
of application uniformity, as discussed previously in Section 2.7.1.
In the following tables, each experiment is coded with the parameters of the experiment, as
presented in Table 2.3. The resulting code is presented as: Type of carpet: R = residential or
C = commercial; carpet stain-protection treatment: 1 or 2; carpet cleaning machine: 1 or 2;
and carpet detergent: 0 = no detergent, 1, or 2 [i.e., Experiment 1 (C-l-2-0) = Commercial
carpet, CT1, CM1, and no detergent]. BDL indicates that a result was below the instrument
detection limit of 0.2 ng/mL, while NR indicates that data could not be reported because
the results did not meet data quality requirements or were not obtained. Italicized values
indicate that the results are above the highest calibration concentration of 100 ng/mL.
PFTeDAwas not found above the practical quantification limit (PQL) and, therefore, was
not reported.
Data highlighted in bold font did not pass one or more of the QAPP-stated MQOs, but they
passed the relaxed MQOs of ± 30% established for this project.
A-l
-------
Table A.I. Average extractable PFCAs (ng/g) and percent of original amount
experimental stage for Experiment 1
removed in composite carpet fiber samples at each
(C-l-2-0).
PFBA-C4
PFPeA-C5
PFHxA-C6
PFHpA-C7
PFOA-C8
PFNA-C9
PFDA-C10
PFUnDA-Cll
PFDoDA-C12
PFTrDA-C13
Total PFCAs
Pre-Application
n=5
BDL
BDL
BDL
BDL
7.5 ±1.0
BDL
BDL
BDL
BDL
BDL
7.5 ±1.0
Post-Application
n=3
63.4 ±13
55.1 ±10
399 ±91
1150±201
809 ±126
934±118
514±88
382 ±94
235 ± 66
(NR)
4540 ±801
Post-Cleaning 1
n=2
55.6 ± 6.6
82.7 ± 9.2
349 ± 46
1530 ± 140
828 ± 89
1030 ± 155
363 ± 8.9
227 ±6.1
201 ± 10
71.8 ±1.1
4740 ± 437
Post-Cleaning 2
n=5
52.3 ±2.8
43.2 ±3.0
310±9.6
1000 ±57
560 ±17
864 ± 69
446 ± 37
413 ±25
195 ±8.5
91.6±6.6
3980 ±135
Post-Cleaning 3
n=4
49.5 ±1.8
39.3 ±1.1
24 ±17
819±77
450 ±39
703 ± 89
373 ± 63
373 ± 42
190±31
73.4 ±10
3350 ±364
Total PFCA
removal, %
22%
29%
29%
28%
44%
25%
28%
3%
19%
(NR)
26%
A-2
-------
Table A.2. Average extractable PFCAs (ng/g) and percent of original amount removed in composite carpet fiber samples at each
experimental stage for Experiment 2 (C-2-2-0).
PFBA-C4
PFPeA-C5
PFHxA-C6
PFHpA-C7
PFOA-C8
PFNA-C9
PFDA-C10
PFUnDA-Cll
PFDoDA-C12
PFTrDA-C13
Total PFCAs
Pre-Application
n=5a
BDL
BDL
BDL
BDL
BDL
BDL
BDL
BDL
BDL
BDL
BDL
Post-Application
n=5
(NR)
73 .2 ±4.8
411±41
2520 ± 90
1 480 ±140
1980 ±113
1090 ±81
549 ± 29
488 ± 60
(NR)
8580 ±471
Post-Cleaning 1
n=3
(NR)
35.4 ±4.0
158 ±4.4
1340 ±18
744 ± 68
1 060 ±86
610 ±47
287 ±7.7
271 ±6.7
154 ±14
4660 ± 234
Post-Cleaning 2
n=4
(NR)
26.8 ±2.4
115±15
916±99
492 ± 78
602 ± 85
377 ± 74
191±31
207 ± 45
98.8 ±8.6
3030 ± 424
Post-Cleaning 3
n=4
(NR)
27.0 ±2.3
122 ± 12
962 ± 68
495 ± 37
591±51
391 ±34
227 ± 28
198 ±21
105±11
3 120 ±252
Total PFCA
removal, %
(NR)
63%
70%
62%
67%
70%
64%
59%
59%
(NR)
64%
' Recovery standards for these samples exceeded the measurement quality objective (MQO) but data were all BDL and presented for completeness.
A-3
-------
Table A.3. Average extractable PFCAs (ng/g) and percent of original amount removed in composite carpet fiber samples at each
experimental stage for Experiment 3 (R-l-1-0).
PFBA-C4
PFPeA-C5
PFHxA-C6
PFHpA-C7
PFOA-C8
PFNA-C9
PFDA-C10
PFUnDA-Cll
PFDoDA-C12
PFTrDA-C13
Total PFCAs
Pre-Application
n=3
BDL
BDL
BDL
BDL
BDL
BDL
BDL
BDL
BDL
BDL
BDL
Post-Application
n=2
135 ±29
112±15
524 ±53
716±84
384 ±16
447 ± 23
257±11
279 ±10
197 ±26
136±15
3 190 ±236
Post-Cleaning 1
n=3
95.5 ±2.8
73.2 ±2.8
309 ±14
483 ± 46
290 ±18
315±15
187 ±15
225 ± 15
143 ±0.92
107±1.7
2230 ±6.1
Post-Cleaning 2
n=3a
83.6 ± 7.9
62.1 ±3.1
189 ± 9.1
342 ± 26
249 ± 14
283 ± 13
127 ± 5.4
159 ± 9.1
112 ±6.2
82.1 ± 5.9
1690 ± 81
Post-Cleaning 3
n=3a
59.0 ±11
48.1 ± 8.9
159 ± 27
298 ± 56
232 ± 36
278 ± 48
106 ± 19
155 ± 25
99.8 ± 18
82.2 ± 9.5
1520 ± 255
Total PFCA
removal, %
56%
57%
70%
58%
40%
38%
59%
44%
49%
39%
52%
' Recovery standards for these samples were between 45-70% but data included for completeness.
A-4
-------
Table A.4. Average extractable PFCAs (ng/g) and percent of original amount removed in composite carpet fiber samples at each
experimental stage for Experiment 4 (R-l-2-0).
PFBA-C4
PFPeA-C5
PFHxA-C6
PFHpA-C7
PFOA-C8
PFNA-C9
PFDA-C10
PFUnDA-Cll
PFDoDA-C12
PFTrDA-C13
Total PFCAs
Pre-Application
n=5
BDL
BDL
BDL
0.821 ±1.4
BDL
BDL
BDL
BDL
BDL
BDL
0.821 ±1.4
Post-Application
n=4
55.1±3.9
53.0±1.6
413 ±22
668 ± 24
349 ±28
367 ±5.3
177 ±14
195 ±15
85.8 ±1.9
73 .9 ±1.6
2440 ± 42
Post-Cleaning 1
n=5
35.7 ±4.0
41.0±5.5
334 ±43
564 ±36
292 ±16
303 ± 23
156±14
168 ±14
79 .2 ±8.1
66.6 ±9.5
2040 ±119
Post-Cleaning 2
n=5
BDL
BDL
BDL
381±21
273 ± 14
303 ± 15
139 ±13
165 ±12
75.9 ±6.3
56.4 ± 6.5
1470 ± 62
Post-Cleaning 3
n=3
BDL
BDL
BDL
319±9.7
256 ±3.6
302 ±8.9
139 ±14
185 ±15
73.7 ±3.2
56.7±3.1
1330 ±35
Total PFCA
removal, %
100%
100%
100%
52%
27%
18%
22%
5%
14%
23%
45%
A-5
-------
Table A.5. Average extractable PFCAs (ng/g) and percent of original amount removed in composite carpet fiber samples at each
experimental stage for Experiment 5 (R-2-2-0).
PFBA-C4
PFPeA-C5
PFHxA-C6
PFHpA-C7
PFOA-C8
PFNA-C9
PFDA-C10
PFUnDA-Cll
PFDoDA-C12
PFTrDA-C13
Total PFCAs
Pre-Application
n=5
BDL
BDL
BDL
BDL
BDL
BDL
BDL
BDL
BDL
BDL
BDL
Post-Application
n=5
10.8 ±0.88
59.2 ±4.7
252 ± 24
870 ± 64
389 ±6.9
583 ±40
312±48
229 ± 25
173 ±13
95.0 ±6.0
3 100 ±333
Post-Cleaning 1
n=4
BDL
29.8±1.1
143 ±5.5
480 ±29
261 ±15
367 ±18
180±11
228 ±11
171 ±9.8
(NR)
1940 ± 83
Post-Cleaning 2
n=5a
(NR)
(NR)
BDL
227 ± 8.1
168 ± 2.8
230 ±10
68.7 ± 9.7
177 ±10
124 ± 5.4
(NR)
1060 ± 21
Post-Cleaning 3
n=5
(NR)
(NR)
26.0 ± 1.7
213 ± 23
240 ±12
353 ± 22
49 ±10
219±19
(NR)
(NR)
1144 ±70
Total PFCA
removal, %
100%b
50%b
89%
76%
38%
39%
84%
4%
l%b
(NR)
63%
a Recovery standards for reported data only 64% but data included for completeness.
b Total PFCA % removal as calculated from the results from the first cleaning. All others calculated from the third cleaning.
A-6
-------
Table A.6. Average extractable PFCAs (ng/g) and percent of original amount removed in composite carpet fiber samples at each
experimental stage for Experiment 6 (C-2-2-1). Re-use of carpet from Experiment 2.
Pre-Application
n=4a
Post-Application
n=5'
Post-Cleaning 1
n=5
Post-Cleaning 2
n=4
Post-Cleaning 3
n=2
Total PFCA
removal, %
PFBA-C4
(NR)
27.7 ±3.4
21.4±3.0
12.7 ±1.3
11.5 ±1.5
59%
PFPeA-C5
27.0 ±2.3
84.7 ±9.0
61.3 ±6.7
30.4 ± 2.1
33.5 ± 4.5
60%
PFHxA-C6
122 ±12
577 ±38
384 ±40
200 ± 3.2
171 ± 22
70%
PFHpA-C7
962 ± 68
2640 ±119
2010 ±116
1200 ±39
958 ±135
64%
PFOA-C8
495 ± 37
7700 ±116
1190 ±61
595 ± 30
454 ± 0.14
73%
PFNA-C9
591±51
2330± 49
1480 ±23
615 ±27
445 ± 15
81%
PFDA-C10
391 ±34
1460 ±106
1000 ±59
415 ±8.1
276 ± 5.9
81%
PFUnDA-Cll
227 ± 28
(NR)
438 ±16
181 ± 8.5
75.3 ± 6.0
(NR)
PFDoDA-C12
198 ±21
669 ± 99
377 ±12
123 ± 38
71.4 ± 0.09
89%
PFTrDA-C13
105±11
380 ± 16
243 ± 13
50.1 ± 5.4
40.7 ± 4.9
89%
Total PFCAs
3120 ±252
10450 ±445
7210 ±239
3420 ± 43
2540 ±130
76%
a Pre-Application data are the results of Post-Cleaning 3 for Experiment 2.
b Initial post-application results are higher than previous experiments due to the re-use of the carpet from Experiment 2.
A-7
-------
Table A.7. Average extractable PFCAs (ng/g) and percent of original amount removed in composite carpet fiber samples at each
experimental stage for Experiment 7 (R-l-1-2). Experiments 7 and 8 considered replicate tests.
PFBA-C4
PFPeA-C5
PFHxA-C6
PFHpA-C7
PFOA-C8
PFNA-C9
PFDA-C10
PFUnDA-Cll
PFDoDA-C12
PFTrDA-C13
Total PFCAs
Pre-Application
n=5
BDL
BDL
BDL
BDL
BDL
BDL
BDL
BDL
BDL
BDL
BDL
Post-Application
n=5
85 .4 ±4.1
79.9 ±5.4
537 ±29
809 ±36
417±27
495 ± 42
288 ±20
258 ± 17
83.0 ±6.4
(NR)
3110±136
Post-Cleaning 1
n=5
27.5 ±1.0
12.6 ±0.47
69.9 ±2.1
276 ±18
311±13
436 ±39
142 ± 29
214±32
47.5 ±3.1
(NR)
1580 ±123
Post-Cleaning 2
n=5
28.8 ±5.5
16.7±1.0
60.9 ±8.8
263 ±31
324 ± 26
516±53
159 ±23
281 ±39
60.7 ±7.2
62.3 ±11
1770 ±234
Post-Cleaning 3
n=5
19.0±1.9
9.11±1.4
29.8 ±1.6
163 ±22
220 ± 17
430 ±32
79.0 ±4.4
247 ± 20
47.6 ±3.5
53.2±5.1
1300 ±97
Total PFCA
removal, %
79%
91%
95%
81%
48%
13%
72%
4%
45%
(NR)
58%
A-8
-------
Table A.8. Average extractable PFCAs (ng/g) and percent of original amount
experimental stage for Experiment 8 (R-l-1-2). Experiments
removed in composite carpet fiber samples at each
7 and 8 considered replicate tests.
PFBA-C4
PFPeA-C5
PFHxA-C6
PFHpA-C7
PFOA-C8
PFNA-C9
PFDA-C10
PFUnDA-Cll
PFDoDA-C12
PFTrDA-C13
Total PFCAs
Pre-Application
n=5
BDL
BDL
BDL
BDL
BDL
BDL
BDL
BDL
BDL
BDL
BDL
Post-Application
n=5
78.1±3.5
63 ± 2.5
384 ±16
904 ± 28
548 ±10
482 ±32
234 ±4.9
1620 ±91
63.2 ±0.59
49.8 ±6.8
4430 ± 142
Post-Cleaning 1
n=5
30.6 ±1.6
22.9±1.1
76.4 ±9.3
343 ± 25
314±21
394 ± 22
148 ±15
1240 ±132
58.2 ±6.9
59.8 ±4.4
2680 ±218
Post-Cleaning 2
n=5
20.3 ±1.8
17.5 ±1.5
49.3 ±4.5
320 ±21
347 ±30
340 ± 25
84.8 ±6.8
1150 ±102
50.4 ±0.30
49.3 ±3.6
2370 ± 209
Post-Cleaning 3
n=4
15.5 ± 0.70
13.1 ±0.76
34.2 ±1.5
241 ± 10
310±15
340 ±21
84.2 ±3.8
1250 ± 73
49.9 ± 3.5
50.5 ± 3.6
2390±111
Total PFCA
removal, %
80%
79%
91%
73%
43%
30%
64%
23%
20%
-1%
46%
A-9
-------
Table A.9. Average extractable PFCAs (ng/g) and percent of original amount removed in composite carpet fiber samples at each
experimental stage for Experiment 9 (R-2-2-1). Re-use of carpet from Experiment 5.
PFBA-C4
PFPeA-C5
PFHxA-C6
PFHpA-C7
PFOA-C8
PFNA-C9
PFDA-C10
PFUnDA-Cll
PFDoDA-C12
PFTrDA-C13
Total PFCAs
Pre-Application
n=5a
(NR)
(NR)
26.0 ± 2.0
213 ± 23
240 ±12
353 ± 22
49.0 ±10
219±19
(NR)
(NR)
1 144 ± 70
Post-Application
n=5
9.29 ± 0.74
56.8 ±2.5
296 ±6.5
1 060 ±37
607 ± 26
835 ±33
328 ± 25
377 ±17
286±11
(NR)
3 840 ±25
Post-Cleaning 1
n=l
7
34
98
624
370
522
173
314
208
159
2510
Post-Cleaning 2
n=5
2.84 ± 0.58
20.5 ± 0.50
45.0 ± 0.96
415 ±11
263 ± 60
368 ± 18
113 ±24
241 ± 12
141 ± 22
88.9 ± 27
1700 ± 107
Post-Cleaning 3
n=3
3.24 ± 2.7
8.56 ± 0.55
27.7 ± 0.55
199 ±18
236 ±8.4
304 ± 32
137 ±37
193 ±12
85 ±26
72.9 ± 20
1190 ±123
Total PFCA
removal, %
65%
84%
91%
81%
61%
64%
58%
79%
70%
(NR)
69%
' Pre-Application data are the results of Post-Cleaning 3 for Experiment 5.
A-10
-------
Table A.10. Average extractable
sampl
PFCAs (ng/g) post-application in individual carpet fiber
es for Experiment 1 (C-l-2-0).
PFBA-C4
PFPeA-C5
PFHxA-C6
PFHpA-C7
PFOA-C8
PFNA-C9
PFDA-C10
PFUnDA-Cll
PFDoDA-C12
PFTrDA-C13
Total PFCAs
Post-Application
n=5
63.8
57.1
324
977
684
794
479
341
213
179
4110
Standard Deviation
n=5
67.5
61.9
287
818
544
629
398
270
166
152
3390
Table A.ll. Average extractable
sampl
PFCAs (ng/g) post-application in individual carpet fiber
es for Experiment 2 (C-2-2-0).
PFBA-C4
PFPeA-C5
PFHxA-C6
PFHpA-C7
PFOA-C8
PFNA-C9
PFDA-C10
PFUnDA-Cll
PFDoDA-C12
PFTrDA-C13
Total PFCAs
Post-Application
n=5
23.6
94.9
515
2960
1680
2310
1300
619
472
245
10220
Standard Deviation
n=5
4.3
31
117
472
274
344
190
109
60
44
1610
A-ll
-------
Table A.12. Average extractable
sampl
PFCAs (ng/g) post-application in individual carpet fiber
es for Experiment 3 (R-l-1-0).
PFBA-C4
PFPeA-C5
PFHxA-C6
PFHpA-C7
PFOA-C8
PFNA-C9
PFDA-C10
PFUnDA-Cll
PFDoDA-C12
PFTrDA-C13
Total PFCAs
Post-Application
n=5
112
98.9
490
634
355
445
240
245
169
126
2920
Standard Deviation
n=5
27
28
134
154
85
97
56
34
23
16
640
Table A.13. Average extractable
sampl
PFCAs (ng/g) post-application in individual carpet fiber
es for Experiment 4 (R-l-2-0).
PFBA-C4
PFPeA-C5
PFHxA-C6
PFHpA-C7
PFOA-C8
PFNA-C9
PFDA-C10
PFUnDA-Cll
PFDoDA-C12
PFTrDA-C13
Total PFCAs
Post-Application
n=4
63.9
56.9
413
730
407
384
203
200
108
85.0
2650
Standard Deviation
n=4
13
13
83
115
45
68
15
41
18
11
392
A-12
-------
Table A.14. Average extractable
sampl
PFCAs (ng/g) post-application in individual carpet fiber
es for Experiment 5 (R-2-2-0).
PFBA-C4
PFPeA-C5
PFHxA-C6
PFHpA-C7
PFOA-C8
PFNA-C9
PFDA-C10
PFUnDA-Cll
PFDoDA-C12
PFTrDA-C13
Total PFCAs
Post-Application
n=5
12.0
67.1
273
906
421
564
336
243
192
102
3120
Standard Deviation
n=5
1.7
9.8
59
182
104
147
56
53
41
22
663
Table A.15. Average extractable PFCAs (ng/g) post-application in individual carpet fiber
samples for Experiment 6 (C-2-2-1).8
PFBA-C4
PFPeA-C5
PFHxA-C6
PFHpA-C7
PFOA-C8
PFNA-C9
PFDA-C10
PFUnDA-Cll
PFDoDA-C12
PFTrDA-C13
Total PFCAs
Post- Application
n=5
37.2
158
672
3090
1920
2670
1660
721
681
493
12100
Standard Deviation
n=5
11
74
171
660
424
602
264
115
148
125
2580
Re-use of carpet from Experiment 2.
A-13
-------
Table A.16. Average extractable
samples for Experiment 7 (R-
PFCAs (ng/g) post-application in individual carpet fiber
1-1-2). Experiments 7 and 8 considered replicate tests.
PFBA-C4
PFPeA-C5
PFHxA-C6
PFHpA-C7
PFOA-C8
PFNA-C9
PFDA-C10
PFUnDA-Cll
PFDoDA-C12
PFTrDA-C13
Total PFCAs
Post-Application
n=4
113
95.2
524
753
365
460
272
293
69.5
44.2
2960
Standard Deviation
n=4
51
46
86
141
69
91
47
33
51
12
586
Table A. 17. Average extractable
samples for Experiment 8 (R-
PFCAs (ng/g) post-application in individual carpet fiber
1-1-2). Experiments 7 and 8 considered replicate tests.
PFBA-C4
PFPeA-C5
PFHxA-C6
PFHpA-C7
PFOA-C8
PFNA-C9
PFDA-C10
PFUnDA-Cll
PFDoDA-C12
PFTrDA-C13
Total PFCAs
Post-Application
n=5
79.9
63.3
342
674
416
471
253
243
61.0
49.9
2650
Standard Deviation
n=5
15
15
86
157
100
100
62
43
13
10
589
A-14
-------
Table A.18. Average extractable PFCAs (ng/g) post-application in individual carpet fiber
samples for Experiment 9 (R-2-2-1).a
PFBA-C4
PFPeA-C5
PFHxA-C6
PFHpA-C7
PFOA-C8
PFNA-C9
PFDA-C10
PFUnDA-Cll
PFDoDA-C12
PFTrDA-C13
Total PFCAs
Post- Application
n=4
7.52
51.7
271
965
585
801
355
369
282
136
3830
Standard Deviation
n=4
5.1
18
68
269
169
234
88
78
67
26
954
a Re-use of carpet from Experiment 5.
A-15
------- |