Per- and Polyfluoroalkyl Substances (PFASs) in Indoor Air and Dust

Subscriber access provided by READING UNIV

Article

Per- and polyfluoroalkyl substances (PFASs) in indoor air and dust from homes and various microenvironments in China: Implications for human exposure Yiming Yao, Yangyang Zhao, Hongwen Sun, Shuai Chang, Lingyan Zhu, Alfredo Carlos Alder, and Kurunthachalam Kannan Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.7b04971 • Publication Date (Web): 08 Feb 2018 Downloaded from http://pubs.acs.org on February 15, 2018

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 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 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 31

Environmental Science & Technology

1

Per- and polyfluoroalkyl substances (PFASs)

2

in indoor air and dust from homes and

3

various

4

Implications for human exposure

microenvironments

in

China:

5 6 7 8 9 10 11 12 13 14 15 16

Yiming Yao,a,1 Yangyang Zhao,a,1 Hongwen Sun,*,a Shuai Chang,a Lingyan Zhu,a Alfredo C. Alder,a,b Kurunthachalam Kannanc a MOE Key Laboratory of Pollution Processes and Environmental Criteria, College of Environmental Science and Engineering, Nankai University, Tianjin 300071, China. 38 Tongyan Road, Jinnan District, Tianjin 300350, China TEL: 86-22-23509241, Email: [email protected] b Eawag, Swiss Federal Institute of Environmental Science and Technology, 8600 Dübendorf, Switzerland c Wadsworth Center, New York State Department of Health, and Department of Environmental Health Sciences, School of Public Health, State University of New York at Albany, Albany, NY 12201, USA 1 Equally contribution

17

ABSTRACT

18

A newly developed solid-phase extraction cartridge composed of mixed sorbents

19

was optimized for collection of both neutral and ionizable per- and polyfluoroalkyl

20

substances (PFASs) in indoor air. Eighty-one indoor air samples and 29 indoor dust

21

samples were collected from rooms of homes and hotels, textile shops, and cinemas in

22

Tianjin, China. Fluorotelomer alcohols (FTOHs) were the predominant PFASs found

23

in air (250-82,300 pg/m3) and hotel dust (24.8-678 ng/g). Polyfluoroalkyl phosphoric

24

acid diesters were found at lower levels of n.d.-125 pg/m3 in air and 0.32-183 ng/g in

25

dust. Perfluoroalkyl carboxylic acids (PFCAs) were dominant ionizable PFASs in air

26

samples (121-20,600 pg/m3) with C4-C7 PFCAs contributing to 54%±17% of the

27

profiles, suggesting an ongoing shift to short-chain PFASs. Long-chain PFCAs (C>7)

28

were strongly correlated and the intermediate metabolite of FTOHs, fluorotelomer

29

unsaturated carboxylic acids, occurred in all the air samples at concentrations up to 1

ACS Paragon Plus Environment


30

413 pg/m3, suggesting the transformation of precursors such as FTOHs in indoor

31

environment. Daily intake of ΣPFASs via air inhalation and dust ingestion was

32

estimated at 1.04-14.1 ng/kg bw/d and 0.10-8.17 ng/kg bw/d, respectively,

33

demonstrating that inhalation of air with fine suspended particles was a more

34

important direct exposure pathway than dust ingestion for PFASs to adults.

35

ABSTRACT ART

36

37 38

2


Page 2 of 31

Page 3 of 31


39

1. INTRODUCTION

40

Per- and polyfluoroalkyl substances (PFASs) are a family of synthetic

41

compounds extensively applied in industrial processes and commercial products as

42

protective coatings for fabrics and metals, and as additives in fire-fighting foams, due

43

to their surface activity, and thermal and chemical stability.1,

44

perfluoroalkane sulfonic acids (PFSAs, n ≥ 6) and perfluoroalkyl carboxylic acids

45

(PFCAs, n ≥ 7) are persistent and bioaccumulative and have been ubiquitously

46

detected in food,

47

associated with hepatotoxicity, reproductive toxicity, developmental toxicity, and

48

immunotoxicity.9, 10

3, 4

2

Long-chain

drinking water,5, 6 and human specimens.7, 8 PFASs exposure is

49

The phase-out of C8 PFASs has been implemented in most developed

50

countries,11, 12 where shifting to shorter-chain PFASs and other alternatives has led to

51

their increasing levels in environmental matrices and progressive human exposure

52

risk.13 Meanwhile, the production and application of legacy long-chain perfluoroalkyl

53

acids (PFAAs) and their precursor fluorotelomer alcohols (FTOHs) rehabilitated in

54

mainland China.2 In China, perfluorooctanoic acid (PFOA) was found dominant in

55

both indoor and outdoor dust around a fluorochemical industrial park.14 Dominating

56

air emission of long-chain PFASs was found at a textile manufacturing although

57

short-chain PFASs were claimed to be used.15 Moreover, the product profiles of

58

short-chain substitutes may vary between countries. The emission profiles of PFASs

59

in China may be unique and more complicated due to a co-occurrence of PFASs with

60

different chain lengths in final consumer products. Therefore, exposure to PFASs in

61

residential places is of primary interest. Besides, the occurrence of PFASs in public

62

places where PFAS-containing products are stored, sold or used may also raise

63

concerns for occupational exposure risk and is yet to be clarified in China.

64

In addition to ingestion through diet and drinking water,16-18 indoor air inhalation

65

and dust ingestion are major pathways of human exposure to PFASs.19 As for PFAAs, 3



66

which are most concerned PFASs due to their persistence and toxicity, apart from

67

direct uptake of from the environment,20 inhalation of volatile precursors including

68

FTOHs and perfluorooctane sulfonamidoethanols/sulfonamides (FOSE/FOSAs) and

69

subsequent internal biotransformation also contributes to human exposure to PFAAs

70

(called as indirect uptake). FTOHs and polyfluoroalkyl phosphoric acid diesters

71

(diPAPs) of newly concern have been shown to be metabolized in human and rodents

72

to produce PFCAs,21-24 whereas biotransformation of FOSE/FOSAs in vivo can yield

73

perfluorooctane sulfonic acid (PFOS).25, 26 Till now, the direct uptake of PFAAs in

74

indoor dust has been well documented, however, a direct exposure to PFAAs via air

75

inhalation has been less emphasized on. The ionizable PFASs (i.e. PFAAs) can be

76

enriched in fine particles (< 5 µm)27 that may as well undergo pulmonary uptake.

77

Even so, limited studies have simultaneously investigated the occurrence of both

78

neutral and ionizable PFASs in indoor air,28, 29 whereas ionizable PFASs have been

79

frequently detected in indoor dust.20, 30, 31 This knowledge gap presents considerable

80

uncertainty in exposure and risk assessment of human exposure to PFAAs. Therefore,

81

simultaneous investigation on both neutral and ionizable PFASs in air and paired dust

82

samples is indispensable for a comprehensive and accurate exposure assessment in

83

indoor environment.

84

A suitable sampling technique is crucial for simultaneous and accurate collection

85

and measurement of both neutral and ionizable PFASs in air. Both passive (XAD-4

86

impregnated polyurethane foam (PUF) disks in steel housings)32 and active samplers

87

(glass column containing XAD-2 resin sandwiched between two PUF plugs) have

88

been used for air monitoring of PFASs.33 However, sampling using these two types of

89

samplers either requires much space or takes a long sampling time, which are primary

90

drawbacks when sampling indoor air. Besides, these two techniques require tedious

91

pre-sampling and extraction processes, which might incur contamination of target

92

chemicals. Apart from these two types of samplers, solid-phase extraction (SPE)

93

cartridge is portable, flexible, and feasible for indoor air collection. A low-volume

94

SPE sampler (Isolute ENV+ SPE cartridges) was initially developed for volatile 4


Page 4 of 31

Page 5 of 31


95

neutral PFASs,34 but its use for the collection of ionizable PFASs was not validated.35

96

Therefore, it is necessary to develop a new type of SPE sampler that would enrich

97

both neutral and ionizable PFASs including diPAPs in ambient air from indoor

98

environments.

99

In the present study, a new type of SPE sampler composed of two kinds of

100

sorbents was initially developed to enable simultaneous collection of the totality of

101

neutral PFASs, ionizable PFASs, and diPAPs in indoor air. Sampling campaign was

102

conducted in different kinds of rooms of residential and hotel buildings, and

103

microenvironments of public places (cinemas/outdoor equipment/textile shops);

104

paired indoor dust samples of some sites were also collected and analyzed. The aims

105

of this study were to determine the occurrence and composition profiles of PFASs in

106

different indoor environments, to elucidate potential influencing factors on indoor

107

profiles of PFASs based on different room setups, as well as to evaluate human

108

exposure to PFASs via air inhalation and dust ingestion. Our results may provide

109

further information on exposure pathways and clarify exposure risks in various

110

microenvironments in China.

111

2. MATERIALS AND METHODS

112

2.1. Sample Collection.

113

Airborne PFASs were collected on a SPE cartridge, which was initially

114

developed and optimized for the target analytes (Table S1 in supporting

115

information(SI)). The SPE cartridge (6 mL) consisted of two layers of absorbents: the

116

upper layer of HC-C18 (250 mg) for trapping mainly neutral PFASs and the bottom

117

layer of WAX (250 mg) for trapping ionizable PFASs. A diaphragm vacuum pump

118

(Jinteng, GM-0.33A) was used for pumping air consecutively through the HC-C18

119

and WAX layers of absorbents. The validation of the sampling device including

120

extraction efficiency, breakthrough, and micro-chamber spiking tests was given in the

121

SI. 5



122

Indoor air sampling was performed in Tianjin, China, in the summer (June-September)

123

of 2015 in standard rooms of 13 hotels, living rooms of 19 homes, and

124

microenvironments of 13 public places. Nine living rooms of the homes were

125

resampled in the winter (December) for comparison of PFAS concentrations between

126

the two seasons. The public places included 5 outdoor equipment shops (OS) that sell

127

outdoor wear, 3 curtain shops (CurS), 2 carpet shops (CarS), and 3 cinemas, where the

128

goods sold or decorations used were thought to be potential sources of PFASs. The

129

details of indoor air sampling are given in Table S2. Prior to assembly, the SPE

130

cartridges were sequentially conditioned with methanol and ethyl acetate and wrapped

131

in aluminum foil sealed in polyethylene plastic (PP) bags. For sampling, two

132

cartridges were connected to the pump in parallel to reduce the loading pressure and

133

to increase sampling efficiency. Indoor air about 1 m above the floor was collected at

134

an initial sampling rate of 4.5 L/min for 8-12 h to reach a volume of 2.16-3.24 m3. In

135

some cases, sampling time was extended for convenience that gave a sampling

136

volume up to 8.33 m3. At some public places, due to logical constraints, sampling

137

time was less than 8 h but that volume was sufficient for the quantification of PFASs

138

(Table S2). After sampling, the cartridges were detached on site, wrapped and sealed

139

in aluminum foil and PP bags, transported to the laboratory, and kept at -20 °C before

140

extraction.

141

In summer, a total of 11 hotel and 18 home dust samples were collected within

142

the period of air sampling. The dust samples from hotels were collected from dust

143

bags in vacuum cleaners and those from homes were collected using a pre-cleaned

144

disposable bristle brush from multiple sites on surfaces of furniture and floor. The

145

sampled areas were all kept from cleaning at least for one week. All dust samples

146

were directly swept into PP tubes, sealed in PP bags, and kept at -20 °C until analysis.

147

Neutral PFASs were analyzed using GC-MS, while ionizable PFASs were

148

analyzed using HPLC-MS/MS. Details of chemical information, sample pretreatment,

149

and instrumental analysis are provided in SI. 6


Page 6 of 31

Page 7 of 31


150

2.2. Calculation of Daily Intake.

151

Direct and indirect exposure pathways to PFASs via indoor air inhalation and

152

dust ingestion were estimated for toddlers (1-2 years) and adults (>20 years) using the

153

following equations: 36

154

Eair = (Cair × Vair × Fuptake × Ymeta) / (mbw × 1000)

(1)

155

Edust = (Cdust × Qdust × Fuptake × Ymeta) / (mbw × 1000)

(2)

156

where Eair (ng/kg bw/d) and Edust (ng/kg bw/d) are the estimated daily intakes of

157

PFASs through indoor air and dust samples, respectively; Cair (pg/m3) is PFAS

158

concentration in air samples; Cdust (ng/g) is PFAS concentration in dust samples; Vair

159

(m3/d) is daily inhalation volume; Qdust (mg/d) is dust ingestion rate; Fuptake is uptake

160

fraction (unitless); For estimation of indirect exposure, Ymeta was introduced as a

161

factor to account for metabolic yield (mass basis, unitless) of PFAAs from precursors

162

and mbw is body weight (kg).

163

For direct exposure assessment, two scenarios were considered: an average-case

164

scenario with measured median PFAS concentrations and a worst-case scenario with

165

95th percentile concentrations. Daily inhalation volume was kept constant between the

166

two scenarios (Vair, 8.0 m3/d and 15.7 m3/d for toddlers and adults, respectively);

167

median dust ingestion rates were used for average-case scenario (Qdust, 60 mg/d and

168

30 mg/d for toddlers and adults, respectively) and high rates for the worst-case

169

scenario (200 mg/d and 100 mg/d for toddlers and adults, respectively).35, 37-39 Due to

170

the lack of adequate pharmacokinetic data, pulmonary and gastrointestinal uptake

171

fractions (Fuptake unitless) in both scenarios were assumed to be 100% for all PFASs,

172

which was usually adopted in other studies.20,

173

recommended in Korean Exposure Factors Handbook were 11.7 kg and 62.8 kg for

174

toddlers and adults, respectively. These values adopted for Korean population were

175

more appropriate for the Chinese population than those from the western countries.43

40-42

7


The mean body weights


176

For

indirect

exposure,

metabolic

yield

(Ymeta)

Page 8 of 31

values 44, 45

that

represent

177

biotransformation of 8:2 FTOH to PFOA were proposed.22,

178

0.0027 (mass basis) was used for human studies, and this value was fitted in the

179

average-case exposure scenario.42 For the worst-case exposure scenario, instead of

180

assuming a 100% transformation yield, a value of 0.05 (Ymeta) was used as a

181

conservative estimation, which was so far the highest yield of those used in previous

182

studies.46 As for 8:2 diPAP, a value of 0.1, which was obtained from an in vivo rodent

183

study, was used for Ymeta in average-case exposure scenario,47 whereas that for the

184

worst-case scenario was 1.

185

2.3. Quality Control and Quality Assurance.

A Ymeta value of

186

Procedure blanks consisting of SPE cartridges fortified with internal standards

187

were extracted with each batch of samples to evaluate potential analytical

188

interferences or possible carryover of the target chemicals between samples. During

189

analysis, a calibration standard and an instrumental blank were injected between each

190

batch of 20 samples to check instrumental performance. Quantification was

191

performed using an eight-point calibration curve (200-50,000 pg/mL) spiked with 5

192

ng of each internal standard as same as those in air and dust samples. The regression

193

coefficients (R2) of the standard calibration curves were ≥ 0.99 for all target analytes.

194

For samples with PFAS signals above the highest point of the calibration curve, the

195

extracts were re-analyzed after dilution. Method detection limits (MDLs) were set at

196

three times the standard deviation (SD) of values found in procedural blanks, if the

197

analyte was present in the blank; otherwise limits of quantification (LOQ) was

198

derived from the peak values with a signal-to-noise ratio (S/N) of 10. The MDL

199

values for each analyte are given in Table S3. Most of the target analytes were not

200

found in procedural and instrumental blanks. PFOA was detected in procedural blanks

201

but correction to concentrations in samples was not made because the background

202

level was far below the concentrations found in samples in this study (Table S3).

203

Recoveries of PFASs spiked into home and hotel dust ranged from 90% to 110%, with 8


Page 9 of 31


204

the exception of FTOHs in hotel dust, which were between 60% and 65% (Table S4).

205

For cartridge validation, extraction efficiency for neutral and ionizable PFASs

206

successively was achieved with ethyl acetate and 0.5% NH4OH/MeOH solution at

207

79-123% (Table S5) and no significant breakthrough was observed (Table S6-1). In

208

the micro-chamber test at the spiking level of 5 ng, the recoveries of neutral and

209

ionizable PFASs were 45-87% and 68-102%, respectively (Table S6-2).

210

2.4. Statistical analysis.

211

Non-detected analytes were assigned with a value of zero and incorporated into

212

statistical analysis. Concentrations below the MDLs were taken as left-censored data

213

and the dataset was treated with robust regression on order statistics for median

214

calculation. Spearman rank correlation analysis and Mann-Whitney Test were

215

conducted using IBM SPSS Statistics 22. Prior to analysis, concentrations were

216

natural log-transformed. Pearson’s correlation was only conducted for ventilation test.

217

The hierarchical cluster analysis and heatmap were conducted and created using R

218

3.4.3 with agglomeration method of Ward.D (Euclidean). All other illustrations

219

presented on this manuscript were created using OriginLab OriginPro 2016.

220

3. RESULTS AND DISCUSSION

221

3.1. PFASs in Air Samples

222

Hotel and Residential Buildings. For neutral PFASs, FTOHs were more frequently

223

detected than FOSE/FOSAs in hotels (100% vs. 68%) and homes (100% vs. 27%), at

224

similar median levels of 3,330 and 3,210 pg/m3, respectively (Fig. 1A and Table 1).

225

8:2 FTOH was the dominant FTOH homologue. The highest concentration of

226

ΣFTOHs was 62,100 pg/m3 found in a hotel room. In comparison, the concentrations

227

of ΣFOSE/FOSAs were 1 to 3 orders of magnitude lower and the highest

228

concentration found was 2,460 pg/m3 in a hotel room. DiPAPs were present in 53%

229

and 77% of the hotel and home samples, with much lower median levels at 1.08 9



230

Page 10 of 31

pg/m3 and 1.17 pg/m3, respectively.

231

PFCAs and PFSAs were ionizable and found occurring in most samples (Table

232

S7). Detection of ionizable PFASs was also reported in the literature in indoor air and

233

the atmosphere using low-volume or passive air samplers.28,

234

partially be due to that the finest suspended particles (< 5 µm) was sampled together

235

with air. Due to their particular affinity to glass fiber filters, it is not likely to

236

differentiate ionizable PFASs between gas and particle phases with them.49 Moreover,

237

to account for this total levels in the bulk air is important for health risk assessment.50

238

The medians of ΣPFCAs (C4-C12) in hotels and homes were 563 pg/m3 and 691

239

pg/m3, respectively (Fig. 1A). In most cases, PFOA was the dominant long-chain

240

PFCA (39.0-378 pg/m3), while in some homes, perfluorononanoic acid (PFNA) was

241

dominating at concentrations of 182-380 pg/m3. The concentrations of short-chain

242

PFCAs (C4-C7) were significantly higher in homes than in hotels (p

Per- and Polyfluoroalkyl Substances (PFASs) in Indoor Air and Dust

Recommend Documents