Results from Screening Polyurethane Foam Based ... - ACS Publications

Aug 23, 2016 - program in collaboration with Duke University's Superfund. Research Center that allowed members of the general US population to submit ...
0 downloads 13 Views 652KB Size
Subscriber access provided by Northern Illinois University

Article

Results from Screening Polyurethane Foam Based Consumer Products for Flame Retardant Chemicals: Assessing Impacts on the Change in the Furniture Flammability Standards Ellen M. Cooper, Gretchen Kroeger, Katherine Davis Warnell, Charlotte Clark, Patrick Lee Ferguson, and Heather M Stapleton Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.6b01602 • Publication Date (Web): 23 Aug 2016 Downloaded from http://pubs.acs.org on August 24, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Environmental Science & Technology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 26

1 2 3 4 5 6 7 8 9 10

Environmental Science & Technology

Results from Screening Polyurethane Foam Based Consumer Products for Flame Retardant Chemicals: Assessing Impacts on the Change in the Furniture Flammability Standards Ellen M. Cooper*, Gretchen Kroeger*, Katherine Davis, Charlotte R. Clark, P. Lee Ferguson and Heather M. Stapleton *- Co-First Authors Nicholas School of the Environment, Duke University, Durham, North Carolina USA

11 12 13

*Corresponding Author: Heather Stapleton Nicholas School of the Environment Duke University A220 LSRC Box 90328 Durham, NC 27708 USA phone: 919-613-8717 fax: 919-684-8741 [email protected]

14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

Keywords: Flame Retardants, foam, PUF, TB-117, furniture, PentaBDE, TDCIPP

ACS Paragon Plus Environment

1

Environmental Science & Technology

Page 2 of 26

37 38 39

Abstract

40

required state and federal flammability standards. However, some FRs (e.g. PBDEs and

41

TDCIPP) are associated with health hazards and are now restricted from use in some regions. In

42

addition, California’s residential furniture flammability standard (TB-117) has undergone

43

significant amendments over the last few years, and TDCIPP has been added to California’s

44

Proposition 65 list. These events have likely led to shifts in the types of FRs used, and the

45

products to which they are applied. To provide more information on the use of FRs in products

46

containing polyurethane foam (PUF), we established a screening service for the general public.

47

Participants residing in the US were allowed to submit up to 5 samples from their household for

48

analysis, free of charge, and supplied information on the product category, labeling, and year and

49

state of purchase. Between February 2014 and June 2016, we received 1,141 PUF samples for

50

analysis from various products including sofas, chairs, mattresses, car seats and pillows. Of

51

these samples tested, 52% contained a FR at levels greater than 1% by weight. Tris (1,3-

52

dichloroisopropyl) phosphate (TDCIPP) was the most common FR detected in PUF samples, and

53

was the most common FR detected in all product categories. Analysis of the data by purchasing

54

date suggests that the use of TDCIPP decreased in recent years, paralleled with an increase in the

55

use of TCIPP and a non-halogenated aryl phosphate mixture we call “TBPP.” In addition, we

56

observed significant decreases in FR applications in furniture products and child car seats,

57

suggesting the use of additive FRs in PUF may be declining, perhaps as a reflection of recent

58

changes to TB-117 and Proposition 65. More studies are needed to determine how these changes

59

in FR use in products relate to changes in exposure among the general population.

Flame retardant (FR) chemicals have often been added to polyurethane foam to meet

60 61

ACS Paragon Plus Environment

2

Page 3 of 26

62

Environmental Science & Technology

Introduction

63

Due to increasing numbers of home fires in which furniture was often an item first

64

ignited, the California Bureau of Electronic and Appliance Repair, Home Furnishings and

65

Thermal Insulation implemented a residential flammability standard known as Technical Bulletin

66

117 (TB-117) in 19721. This standard required that resilient filling materials used in upholstered

67

furniture meet a 12-second, open-flame test designed to reduce the ignition time and

68

combustibility of the material. In order to meet this standard, furniture manufacturers have

69

historically used additive chemical treatments, which primarily include applications of

70

brominated and/or organophosphate flame retardants (FRs) to the polyurethane foam used as

71

filling materials 2, 3. Brominated FRs work by releasing bromine radicals into the gas phase

72

which help to quench the reactions that propagate fire. In contrast, organophosphate FRs

73

typically work in the solid phase by increasing a char layer which reduces oxygen and

74

diminishes the spread of the flame4, 5.

75

The high use of these types of FRs in furniture has been suggested to contribute to higher

76

exposure among the general population, particularly for children, and to increased risks for

77

adverse health outcomes6, 7. In 2001, Noren and Meiryonte published a paper in which they

78

examined the historical trends in persistent organic pollutant levels in breast milk collected from

79

women in Sweden as a means of evaluating temporal changes in exposure8. While they observed

80

decreases in dichlorodiphenyl trichloroethane (DDT), dioxins, and polychlorinated biphenyls

81

with time (due to bans and restrictions placed on DDT use), the levels of polybrominated

82

diphenyl ethers (PBDEs) were observed to be increasing rapidly. Several successive studies also

83

observed similar trends for PBDEs in human and animal tissues9-11, suggesting that exposure was

84

increasing with doubling times ranging from 2-7 years. Within the last ten years alone, hundreds

ACS Paragon Plus Environment

3

Environmental Science & Technology

Page 4 of 26

85

of studies have been conducted examining the environmental distribution, fate and toxicity of

86

PBDEs. Due to mounting evidence suggesting that PBDEs were persistent, highly

87

bioaccumulative, and potentially toxic, the three primary PBDE commercial mixtures

88

(PentaBDE, OctaBDE, and DecaBDE) were subsequently either banned from use or voluntarily

89

phased-out of manufacturing, depending on the region 12. This led to a significant shift away

90

from the use of PBDEs (i.e. PentaBDE) and towards other types of brominated and

91

organophosphate FRs to meet TB-117. Based on our previous studies3, 13, our data suggest that

92

manufacturers primarily moved to chlorinated organophosphate flame retardants (e.g. Tris (1,3-

93

dichloro-isopropyl) phosphate, TDCIPP), or a mixture known as Firemaster®550 [FM550],

94

which consists of a mixture of brominated aryl esters and triaryl phosphate esters14. New studies

95

are now demonstrating that exposure to TDCIPP and FM550 is also becoming ubiquitous among

96

the general US population 15-18. The other chlorinated organophosphate FRs (e.g. Tris

97

(chloropropyl) phosphate, TCIPP) are now also detected ubiquitously in Australian and

98

European populations as well 19-21. Human exposure to additive FRs may occur through

99

inhalation, ingestion, and dermal absorption22-24. Given that recent research indicates

100

considerable potential for dermal absorption of the chlorinated FRs commonly found in furniture

101

(TDCIPP, TCIPP)23, concern for human exposure from contact with furniture and other foam-

102

containing consumer products is warranted. Recent animal studies suggest that exposure to

103

TDCIPP and components present in FM550 can lead to reproductive and developmental effects

104

including reduced fecundity, problems with heart development, reductions in bone development

105

25-30

106

to increased risks for adverse health outcomes.

, and even cancer 31. Therefore, the use of current FRs in residential furniture may still lead

ACS Paragon Plus Environment

4

Page 5 of 26

107

Environmental Science & Technology

Due to these concerns, and to the fact that a majority of home fires involving furniture are

108

ignited via cigarettes and not via an open flame, the state of California recently revised TB-117

109

(now known as TB117-2013) from an open flame test to a smolder-ignition standard. In 2014,

110

the state of California added a new law (SB1019) requiring all furniture be sold with a label

111

indicating whether or not the furniture is chemically treated with a FR. However, it’s currently

112

unclear how these changes to California standards will affect the use of FRs in furniture, and

113

how these changes will relate to exposures among the general population.

114

In order to prioritize FRs for human exposure assessments and biomonitoring programs,

115

scientists and policy-makers must understand which FRs are used most frequently, and in what

116

products. Currently, this information is not publicly available, and chemical treatments can be

117

identified only through material analysis. In 2011 and 2012, we reported on the levels of additive

118

FRs in infant products and residential sofas3, 13; however, other types of products containing

119

polyurethane foam were not tested (e.g. mattress pads, chairs, pillows etc). Furthermore, TB-117

120

was significantly altered in 2013, which may alter the way FRs are used in furniture. Therefore,

121

to better understand the use of and potential exposure to FRs in products containing polyurethane

122

foam (PUF), we initiated a program in collaboration with Duke University’s Superfund Research

123

Center that allowed members of the general US population to submit samples for testing free of

124

charge. To our knowledge, this is the first and only free service available to the public for testing

125

FRs in foam. Our research objectives were to analyze products containing PUF for 7

126

common additive FRs to determine which FRs are more commonly used across various

127

products. A secondary objective was to determine if the use of additive FRs has changed

128

over time. When submitting samples for testing, participants were asked to provide information

129

on the type of product (e.g. sofa, chair, mattress, etc), year and state in which the product was

ACS Paragon Plus Environment

5

Environmental Science & Technology

Page 6 of 26

130

purchased, and whether or not the product contains a TB-117 label. Between February 2014 and

131

June 2016, we received 1,141 samples for testing. Here we summarize our findings from these

132

analyses, and report on trends by product category and changes in flame retardant use over time.

133 134

MATERIALS AND METHODS

135

Participant recruitment and Sample Submission

136

Participant enrollment and sample submission is handled through the project website

137

(http://foam.pratt.duke.edu). Email advertisements were used to assist recruitment, which was

138

bolstered by news and media reports on flame retardant chemicals, blogs mentioning our project

139

(~31% of traffic to our website), social media (~5% website traffic), academic partners and

140

relevant interest groups. Search engines and direct traffic account for over 50% of website

141

traffic. Participants were United States residents age 18 or older. For each sample, participants

142

completed a questionnaire regarding contact information and product details including the

143

product type, whether the product was intended for a child or adult, what kind of flammability

144

standard label was present (e.g., CA TB-117 or CFR 1632), and year and state in which the

145

product was purchased. Product types included Sofas and Loveseats (i.e., two-seat furniture

146

pieces), Chairs, Rocking Chairs and Recliners, Other furniture, Mattresses, Mattress pads (i.e.,

147

cushions that sit on top of and are separate from mattresses), Child Mattresses, Child Car Seats,

148

Pillows, Other Child Products, Pit Cubes/Gymnastics Equipment, Other (miscellaneous).

149

Following completion of the questionnaire, participants were given instructions for sampling,

150

packaging, and mailing the PUF sample to Duke University for testing. Each participant was

151

asked to cut out a small marble-sized piece of PUF from their product, wrap it tightly in

152

aluminum foil, seal it in a plastic bag, and mail it to the address provided. Approximately 6 to 8

ACS Paragon Plus Environment

6

Page 7 of 26

Environmental Science & Technology

153

weeks later, the participants were mailed a letter providing their results. All study protocols were

154

approved by the institutional review board at Duke University.

155 156

Sample analyses

157

Materials

158

Standards of the PentaBDE technical mixture, 2,2-bis(chloromethyl)propane-1,3-diyl-tetrakis(2-

159

chloroethyl)bis(phosphate) (V6), 2-Ethylhexyl 2,3,4,5-tetrabromobenzoate (TBB, aka EHTBB)

160

and (2-ethylhexyl) tetrabromophthalate (TBPH, aka BEHTBP) were purchased from Wellington

161

Laboratories. Firemaster 550 (FM550) was a gift provided by Dr. Susan Klosterhaus who had

162

previously received the mixture from Great Lakes Chemical (West Lafayette, IN). H. Stapleton

163

was provided a sample of polyurethane foam from a foam manufacturer in North Carolina that

164

used Firemaster 600 (FM600) in the foam, from which a sample of FM600 was extracted and

165

used as a standard in this project. Triphenyl phosphate (TPHP) was purchased from Sigma-

166

Aldrich (St. Louis, MO). Tris(2-chloroethyl) phosphate (TCEP), tris(1-chloro-2-propyl)

167

phosphate (TCIPP), and tris(1,3-dichloroisopropyl) phosphate (TDCIPP) were purchased from

168

Sigma-Aldrich (St. Louis, MI), Pfaltz & Bauer (Waterbury, CT), and Fluka (St. Louis, MO),

169

respectively. Dichloromethane (DCM), Acetonitrile (ACN), water, and methanol were purchased

170

from Honeywell (Muskegon, MI)

171 172

Sample preparation

173

Preparation of foam samples was adapted from Stapleton et al. 20123. Briefly, 5 mL DCM was

174

added to 50 mg foam sample in a glass test tube and sonicated for 10 min. A 100µL aliquot of

175

the extract was diluted to 1 mL for analysis by gas chromatography mass spectrometry (GC/MS).

ACS Paragon Plus Environment

7

Environmental Science & Technology

Page 8 of 26

176

A laboratory blank (test tube containing DCM but no foam) was included with each batch of

177

samples processed. Foam extracts were first screened using gas chromatography mass

178

spectrometry (GC/MS). Based on GC analyses, some samples required further analysis by liquid

179

chromatography tandem mass spectrometry (UPLC/MS/MS). For these samples, 100 µL of

180

extract was exchanged to 200 µL methanol and diluted to 1 mL with water. Due to the generally

181

small sample size, we were not able to conduct thorough testing to characterize physical

182

properties of the foam (e.g., density) or classify the type of polyurethane foam.

183 184

Mass Spectrometry Analyses

185

A suite of mass spectrometric analyses were used to identify individual flame retardant (FR)

186

chemicals and evaluate commercial mixtures. FRs in extracts were analyzed by GC/MS in scan

187

mode with both electron ionization (Agilent 7890 GC, Agilent 5795C MS, Wilmington, DE) and

188

electron capture negative ionization (Agilent 6890 GC, Agilent 5975N MS). Separation was

189

conducted on a DB-5 column (Agilent) using a thermal gradient (40°C for 1 min, 18°C /min to

190

250°C, 1.5°C /min to 260°C, 25°C/min to 300°C, 300°C for 20 min). Initial inlet temperature

191

was 80°C for 0.3 min and then ramped 600°C/min to 275°C. The transfer line was held at 280°

192

C. Chromatograms were screened against a custom spectral library for known flame retardants.

193

To assist in detecting FRs that were likely intentionally applied to the foam to meet flame

194

retardant standards, FRs were semi-quantitatively analyzed using authentic standards (listed

195

above) representing a FR concentration equivalent to 1% by weight of foam. This concentration

196

was selected based on the amounts of FRs typically applied to foam to meet flame retardant

197

standards to limit and avoid false positives. All blanks were screened alongside samples. FRs

198

were rarely detected in blanks, and whenever detected were well below a concentration

ACS Paragon Plus Environment

8

Page 9 of 26

Environmental Science & Technology

199

equivalent to 0.1% by weight of foam. UPLC/MS/MS analysis for V6 was conducted on an

200

Accela UPLC system (ThermoFisher Scientific, Bremen, Germany) with ThermoFisher TSQ

201

Vantage triple quadrupole mass spectrometer. UPLC separation was conducted on Hypersil

202

Gold, C18 column (ThermoFisher, 100 × 2.1 mm, 1.9 µm particle size) using an ACN: water

203

gradient (20% ACN for 2 min to 99% ACN at 13 min, held at 99% ACN for 2 min). Injection

204

volume was 20 µL, and flow rate was 300 µL/min. Ion transition 582.9 m/z to 360.8 m/z was

205

used for quantification, and transitions 582.9 m/z to 296.8 m/z and 582.9 m/z to 98.9 m/z were

206

used for confirmation. In total, seven FR commercial mixtures were evaluated in this screen

207

using authentic standards: FM550 (TPHP, isopropylated TPHP isomers, TBB and TBPH),

208

FM600®, TDCIPP, TCIPP, V6, PentaBDE (containing TPHP, isopropylated TPHP isomers, and

209

BDE congeners 47, 85, 99, 100, 153, and 154), “TBPP” (containing TPHP and mono-di- and tri-

210

butylated TPHP). Structures for FR compounds evaluated in this screen are included in Table S1.

211 212 213 214

Results and Discussion

215

Between January 2014 and June 2016, 1141 foam samples were received for testing and which

216

were collected from a wide variety of products including upholstered furniture, mattresses,

217

mattress toppers, a suite of child products and others (Table 1). Foam samples were submitted

218

from 567 different households and products varied substantially by year of purchase (1960 to

219

2016) and state of purchase. Participant households represent 44 states, with most participants

220

living in the following states: California (24%), North Carolina (16%), Massachusetts (9%), New

221

York (6%), Washington (5%), and New Jersey (5%). Similarly, samples were purchased from

222

the following states: California (22%), North Carolina (10%), New York (7%), Massachusetts

Samples Received for Screening

ACS Paragon Plus Environment

9

Environmental Science & Technology

Page 10 of 26

223

(5%), online purchase (5%), Michigan (4%), and Pennsylvania (3%). Products were grouped into

224

12 categories, of which the most abundant category was Sofas and Loveseats (n=451, 40%,

225

Table 1; Figure 1). Child products (e.g., child car seats, child mattresses, sleep positioners)

226

comprised 16% of all samples. Some product categories (e.g., Pillows) contained very few

227

samples; therefore, this paper reports primarily on the 7 major non-miscellaneous product

228

categories: Sofas and Loveseats, Chairs, Rocking Chairs and Recliners, Mattress Pads, Child Car

229

Seats, Child Mattresses, and Other Child Products. Results from testing other product categories

230

are provided in Supporting Information (Figure S1).

231 232

Flame Retardants Detected by Product Category

233

Identification of a flame retardant in PUF samples was based on comparison to authentic

234

standards and to a FR mass spectral response greater than 1% by weight of the foam. For all

235

samples analyzed, 598 (52%) were positively identified as containing a FR based on our

236

screening process and criteria (Table 1, Figure 1). Among the major product categories, the

237

portion of samples containing a FR ranged from 29% (n=106) of Mattress Pads to 74% of Child

238

Car Seats (n=98). Among Sofas and Loveseats, 45% contained no identifiable FR, which is

239

much higher than reported by Stapleton et al., (2012) for couches purchased between1985-2010,

240

for which only 12% contained no known FRs. The differences may be attributable to the fact

241

that the current study included many products purchased 2013 and later, for which time period

242

we observed a decrease in FR detection as described below. Of all samples tested, 90 (8%)

243

contained more than one FR, which is similar to results reported in our earlier studies 3, 13. There

244

was no clear trend relating multiple FR applications to product type, however, the most

245

commonly observed combination of FRs across all categories was TDCIPP with TCIPP. The

ACS Paragon Plus Environment

10

Page 11 of 26

Environmental Science & Technology

246

presence of more than on FR may be co-application, or possibly cross-contamination during

247

manufacturing.

248 249

Across all samples analyzed tris(1,3,-dichloro-2-propyl) phosphate (TDCIPP) was the most

250

frequently detected FR (24% of all samples tested), followed by FM550 (10%) > tris(2-

251

chloroisopropyl) phosphate (TCIPP) (9%) > PentaBDE (6.0%). This trend varied somewhat

252

among the major product categories, however TDCIPP was still the most frequently detected FR

253

in all categories. TDCIPP was also observed as the most frequently detected FR in US sofas

254

purchased between 1985-2010, followed by FM550 and a mixture of isobutylated triaryl

255

phosphates that we refer to as “TBPP”3. Although no FRs were detected in 48% of samples, this

256

may not necessarily indicate that these samples are not treated with flame retardants; the

257

screening process is not likely to catch reactive or polymeric FRs that are bound to the foam

258

structure or more strongly sorbed, and our method may overlook FRs for which the chemical

259

composition is unknown.

260 261

No general trend was observed based on whether or not products were purchased in the state of

262

CA, even for products that fall under the regulatory guidance of TB-117. For example, 28% of

263

the Sofas and Loveseats tested contained a FR and were purchased in CA; however, 15% of the

264

Sofas and Loveseats from CA that were tested did not contain a FR (Table 1). Interestingly,

265

however, a greater percentage of Child Car Seats (54%) and Child Mattresses (22%) with no FR

266

detected were purchased in CA than observed for FR-containing products in the 7 major

267

categories. This observation may partly reflect changes to FR use in response to the 2011

268

addition of TDCIPP to Prop 65, given that TDCIPP is the most frequently detected FR for these

ACS Paragon Plus Environment

11

Environmental Science & Technology

Page 12 of 26

269

products, and that 84% of Child Car Seats and 64% of Child Mattresses were purchased after

270

2011. Trends on FR use based on state of purchase must be considered with some caution given

271

that CA was the most represented state in the study with regard to where participants lived and

272

where products were purchased.

273 274

Temporal Trends in FR Use Over Time

275

To better understand potential changes in FR use over time and across products, temporal trends

276

were examined among the product categories with the highest sample submissions: Sofas and

277

Loveseats; Chairs, Rocking Chairs and Recliners; Mattress Pads; and Child Car Seats (Figure 2).

278

Given that TB-117 is a California standard, we also included the percent of samples purchased in

279

the state of California within each category. Of the 828 samples analyzed in these product

280

categories, 97 had an unknown or uncertain year of purchase, and were thus excluded from this

281

analysis. In all product categories, we observed an increase in the frequency of FR use up until

282

2014. For these product categories, 35% of samples purchased between 1950 and 1999 (n=124),

283

61% of samples purchased between 2000 and 2004 (n=85), and 68% of samples purchased

284

between 2005 and 2013 (n=390) contained a FR. However, only 37% of the samples purchased

285

between 2014 and 2016 (n=132) contained FRs, representing a significant decrease in detection

286

frequency (p TBPP (4%). Most samples ( 1 flame retardant

49 15 contained > 1 flame retardant

15contained > 1 flame retardant

1 1 9 contained > 1 flame retardant

Sofas and Love seats

Chairs

Rocking chairs and Recliners

Mattress pads

n=451

n=130

n=44 1

n=106 3

41

9 14

47 8 201

48

1

9 7

8 4

17

9 1

21 14

6

4

1 113 3 31 23 contained > 1 flame retardant

PentaBDE

1

5 35

16 contained > 1 flame retardant

FM550

75 3

6

FM600

TCIPP

12 5 contained > 1 flame retardant

TDCIPP

TBPP

V6

2 contained > 1 flame retardant

none detected

Figure 1. Detection of FRs in major product categories.

ACS Paragon Plus Environment

22

Page 23 of 26

Environmental Science & Technology

569 Sofas and Loveseats n = 81

n = 61

n = 187

Chairs, Rocking Chairs and Recliners n = 62

n = 39

80

% ND

60

% FR 39.3%

40 21.9%

20

25.8%

14.8%

% from CA

n = 16

60 40

31.3% 19.0% 12.8%

7.7%

0 1950-1999

2000-2004

2005-2013

2014-2016

1950-1999

Purchase Year

n = 11

n = 71

2000-2004

2005-2013

2014-2016

Purchase Year

Mattress pads n=4

Child Car Seats n = 53

n = 15

100

n = 42

100

80 60 40 25.0% 18.2%

21.1%

20

16.7%

0

Percentage of Samples

Percentage of Samples

n = 79

80

20

0

80 60

47.6%

40

32.1%

20 n=0

n=0

1950-1999

2000-2004

0 1950-1999

2000-2004

2005-2013

2014-2016

Purchase Year

570 571

n = 13

100 Percentage of Samples

Percentage of Samples

100

2005-2013

2014-2016

Purchase Year

Figure 2. Detection of FRs in selected product categories by year.

ACS Paragon Plus Environment

23

Environmental Science & Technology

Page 24 of 26

572

% of Products Containing an FR

100 90 80

< 2005 (n = 75)

70

>= 2005 (n = 150)

60 50

44%

41%

44%

40 28%

30 20 10

13% 2% n = 33 3

PentaBDE 573 574 575 576 577 578

4%

0%

0 31 66

0

TDCIPP

17%

19

TCIPP

5%

3 25

4 42

TBPP

FM550

Figure 3. Presence of TDCIPP, TBPP, PentaBDE, TCIPP and FM550 in samples containing a FR purchased before and after 2005. Note that the sum of products containing a FR purchased after 2005 exceeds 100% because some products (18) contained more than one FR.

ACS Paragon Plus Environment

24

Page 25 of 26

Environmental Science & Technology

579

Sofas and Loveseats

% of Products Containing an FR

100 (a) 90 80

< 2014 (n = 215)

70

>= 2014 (n = 18)

60 50

47%

40 30 20

22% 11%

10

7%

17% 12%

20%17%

0 n = 101 2

TDCIPP

11 3

4

4

TCIPP

TBPP

26 5

FM550

580

Child Car Seats

% of Products Containing an FR

100 90

(b)

80

77%

70 60 50

48%

40

< 2014 (n = 48) >= 2014 (n = 21) 52%

29%

30 20 10

0% 0%

0 n = 32 10

TDCIPP 581 582 583 584 585 586

14 11

0

0

TCIPP

TBPP

4% 5% 2

1

FM550

Figure 4. Presence of TDCIPP, TBPP, TCIPP and FM550 in (a) Sofas and Loveseats and (b) Child Car Seats purchased before and after 2014. Note that the sum of Child Car Seats containing an FR purchased after 2014 exceeds 100% because some products (8) contained more than one FR.

ACS Paragon Plus Environment

25

Environmental Science & Technology

ACS Paragon Plus Environment

Page 26 of 26