Subscriber access provided by Queen Mary, University of London
Article
Brominated flame retardants and organophosphate esters in preschool dust and children’s hand wipes Kristin Larsson, Cynthia A. de Wit, Ulla Sellström, Leena Sahlström, Christian H Lindh, and Marika Berglund Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.8b00184 • Publication Date (Web): 23 Mar 2018 Downloaded from http://pubs.acs.org on March 25, 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 30
Environmental Science & Technology
1
Brominated flame retardants and organophosphate esters in
2
preschool dust and children’s hand wipes
3 4
Kristin Larsson*,†, Cynthia A. de Wit‡, Ulla Sellström‡, Leena Sahlström‡,#, Christian H. Lindh§,
5
Marika Berglund†
6
†
Institute of Environmental Medicine, Karolinska Institutet, Box 210, 171 77 Stockholm, Sweden.
7
‡
Department of Environmental Science and Analytical Chemistry (ACES), Stockholm University, 106 91
8
Stockholm, Sweden
9
§
Division of Occupational and Environmental Medicine, Lund University, 221 85 Lund, Sweden.
10 11
*
12
Kristin Larsson
13
Institute of Environmental Medicine, Karolinska Institutet
14
Box 210
15
171 77 Stockholm
16
Sweden
17
Phone: +46 707206094
18
E-mail:
[email protected] Corresponding author:
19 20
#
21
Recipharm Pharmaceutical Development AB
22
Gårdsvägen 10A
23
169 70 Solna
24
Sweden
Current address:
1 ACS Paragon Plus Environment
Environmental Science & Technology
Page 2 of 30
25
ABSTRACT: Children spend a considerable part of their day in preschool where they may be exposed
26
to hazardous chemicals in indoor dust. In this study, brominated flame retardants (BFRs) and
27
organophosphate esters (OPEs) were analyzed in preschool dust (n=100) and children’s hand wipe
28
samples (n=100) and diphenyl phosphate (DPHP) was analyzed in urine (n=113). Here, we assessed
29
children’s exposure via dust, identified predictors for chemicals in dust and studied correlations
30
between different exposure measures. The most abundant BFRs in dust were decabromodiphenyl ether
31
(BDE-209) and decabromodiphenyl ethane (DBDPE) found at median levels of 270 and 110 ng/g
32
dust, respectively. Tris(2-butoxyethyl) phosphate (TBOEP) was the most abundant OPE, found at a
33
median level of 79000 ng/g dust. For all OPEs and some BFRs, there were significant correlations
34
between the levels in dust and hand wipes. In addition, triphenyl phosphate (TPHP) in preschool dust
35
was significantly correlated with the corresponding metabolite DPHP in children’s urine. The levels of
36
pentaBDEs in dust were higher in older preschools compared to newer, whereas levels of DBDPE
37
were higher in newer preschools. Children’s estimated intakes of individual BFRs and OPEs via
38
preschool dust were below available health based reference values. However, there are uncertainties
39
about potential health effects of some emerging BFRs and OPEs.
40
2 ACS Paragon Plus Environment
Page 3 of 30
Environmental Science & Technology
41
INTRODUCTION
42
Brominated flame retardants (BFRs) are used to reduce the flammability of combustible materials,
43
such as plastics and textiles. Polybrominated diphenyl ethers (PBDEs), hexabromocyclododecane
44
(HBCDD) and tetrabromobisphenol A (TBBPA) are historically the most commonly used BFRs.
45
PBDEs are mainly used as flame retardants in electronics and upholstered furniture, whereas HBCDD
46
is used in polystyrene foams in e.g. building insulation and TBBPA is predominantly used in epoxy
47
resins found in circuit boards.1, 2
48
PBDEs and HBCDD are known to be toxic, mainly targeting the endocrine, reproductive and
49
nervous systems.3 Due to their toxic, persistent and bioaccumulative properties, the use of technical
50
pentaBDE (containing primarily BDE-47, -99, -100) and octaBDE (containing primarily BDE-183)
51
has been restricted to 0.1% by mass in preparations and articles put on the European market after
52
2004.4 After 2019, this restriction will also be applied to the use of technical decaBDE (containing
53
primarily BDE-209).5 HBCDD is listed in Annex XIV of the European regulation on registration,
54
evaluation, authorization and restriction of chemicals (REACH) and is thereby only allowed for
55
authorized use within the EU. TBBPA is still imported in large amounts (1000-10 000 tonnes/year) in
56
the EU.6, 7
57
The restricted use of PBDEs and HBCDD has led to their replacement with alternative BFRs by
58
industry. For example, decabromodiphenyl ethane (DBDPE) is used instead of decaBDE in
59
electronics, 2-ethylhexyl-2,3,4,5-tetrabromobenzoate (EH-TBB) and bis(2-
60
ethylhexyl)tetrabromophthalate (BEH-TEBP) substitute pentaBDE in polyurethane foam, and 1,2-
61
dibromo-4-(1,2-dibromoethyl)cyclohexane (DBE-DBCH) is used instead of HBCDD in polystyrene
62
insulation. However, these alternative or emerging BFRs have similar physiochemical properties as
63
the banned or restricted legacy BFRs and sufficient exposure and toxicity data are often lacking.1, 8, 9
64
The legacy BFRs have also been replaced by organophosphate esters (OPEs). Halogenated OPEs,
65
such as tris(1,3-dichloroisopropyl) phosphate (TDCIPP), tris(2-chloroethyl) phosphate (TCEP) and
66
tris(2-chloroisopropyl) phosphate (TCIPP), are mainly used as flame retardants in e.g. plastics, textiles
67
and polyurethane foam. Non-halogenated OPEs, such as tris(2-butoxyethyl) phosphate (TBOEP) and
3 ACS Paragon Plus Environment
Environmental Science & Technology
Page 4 of 30
68
triphenyl phosphate (TPHP), are also used for other applications, for example in plastics, hydraulic
69
fluids and floor polish.10 Some OPEs are suspected to be neurotoxic and/or reprotoxic.10-12 In addition,
70
TCEP and TDCIPP are suspected to be carcinogenic.11, 13-15 Because of the aforementioned toxicity,
71
TCEP has been phased out since the 1980s and is no longer produced within the EU.11, 16 The other
72
OPEs are still used, but TCIPP and TDCIPP are not allowed in toys produced in the EU.16
73
The major exposure source for BFRs is fatty foods, primarily fish.17 Humans are also exposed to
74
BFRs via indoor dust, which is considered to be an important exposure source especially for young
75
children and for populations living in areas where the environmental levels of some BFRs are high,
76
e.g. pentaBDE in North America.1, 17 There is less information about exposure to OPEs, but ingestion
77
of dust and food as well as inhalation of air have been suggested to be important exposure sources to
78
OPEs.18-21
79
Biomonitoring studies have shown that children have a higher exposure to PBDEs and OPEs than
80
adults,22-24 which makes them a particularly relevant group to include in exposure assessments.
81
Although children spend the majority of their time at home, small children spend up to a third of their
82
weekdays in the preschool, which makes it an important microenvironment to consider when
83
characterizing children’s exposure to both legacy and emerging chemicals. Although PBDEs and
84
HBCDD are no longer used, old consumer products and building materials which contain these
85
compounds may still be present in preschools. Therefore, to achieve the Swedish governmental goal of
86
a “non-toxic environment” for children, preschools have been advised to remove old products which
87
may contain hazardous chemicals. However, there is little knowledge about the importance of such
88
products and other factors in the preschool environment for children’s exposure to chemicals. During
89
the last 15 years, BFRs and/or OPEs have been measured in dust from European preschools in six
90
individual studies.20, 25-32 However, most of these studies included an insufficient number of preschools
91
to reliably identify which factors and sources that are important for the levels of chemicals in indoor
92
dust.
93
The objectives of this study were to evaluate the levels of legacy and emerging BFRs as well as
94
OPEs in preschool dust and identify important factors for these chemicals in the preschool
95
environment. Based on the measured levels, children’s intakes via preschool dust were calculated and 4 ACS Paragon Plus Environment
Page 5 of 30
Environmental Science & Technology
96
related to health based reference values. Also, correlations between BFRs and OPEs in different
97
exposure measures (dust, hand wipes, urine) were studied.
98
MATERIALS AND METHODS
99
Ethical permission for this study was granted by the regional ethical review board in Stockholm (Dnr
100 101
2015/128-31/1). Dust sample collection. Dust samples were collected from a total of 100 preschools visited
102
between February and April 2015 or between September and November 2015. Characteristics of the
103
participating preschools are described in the supporting information and in Larsson et al. 2017.33 Six
104
of the preschools were Waldorf preschools (based on the Steiner education philosophy) having no
105
plastics, electronics or foam mattresses in the indoor environment. From each participating preschool,
106
one settled dust sample was collected in a play room where 4-year-old children usually played. The
107
dust sample was collected on a pre-weighted cellulose filter (7 cm diameter) fixed in a styrene-
108
acrylonitrile holder (Krim.Teknisk Materiel AB, Bålsta, Sweden). A sieve was placed on top of the
109
filter before the filter holder was inserted in a nozzle made of polypropylene (Krim.Teknisk Materiel
110
AB, Bålsta, Sweden) and mounted on the intake nozzle of a vacuum cleaner. Settled dust was
111
collected from surfaces 50-250 cm above the floor, until there was a sufficient amount of dust on the
112
filter. After sampling, the filter holder lid was replaced, the holder was wrapped in aluminum foil and
113
then sealed in a polyethylene plastic bag. The samples were stored at -20°C until analysis. Field blank
114
samples were collected in one third of the preschools. The field blank samples were collected
115
following standard procedure, but instead of vacuuming surfaces, the vacuum nozzle was held up in
116
the air for a second while the vacuum cleaner was turned on.
117
At the time of the dust sampling, a field worker gathered information about the preschool building,
118
routines, presence of certain products, etc. The field worker also asked the preschool personnel to
119
estimate the age of certain items, such as mattresses and furniture.
120
Urine sample collection. Four-year-old children, attending any of the 30 preschools visited in the
121
first sampling round in spring of 2015, were recruited via a written invitation. Children were eligible
5 ACS Paragon Plus Environment
Environmental Science & Technology
Page 6 of 30
122
to participate if they were born between June 2010 and November 2011 and attended preschool more
123
than 20 hours/week. An informed consent form was signed by the parents before the sample
124
collection. Urine samples were collected between March and May 2015, within a month after the
125
collection of dust at the respective preschool. A median number of 4 (range 1-13) children per
126
preschool from a total of 28 preschools participated, resulting in a total of 113 children. The parents
127
collected the child’s first morning urine in a polypropylene tube (Sarstedt, Numbrecht, Germany) on a
128
Thursday morning. The urine samples were stored at -20°C until analysis. Details about the
129
recruitment and sampling procedure are described in the supporting information.
130
Hand wipe sample collection. Among the children who provided urine samples, a total of 100
131
children from 27 preschools also provided hand wipe samples. The number of participating children
132
per preschool ranged between 1 and 7 children, with a median of 3 children per preschool.
133
Hand wipe samples were collected at the preschool, normally at mid-day or afternoon. Prior to
134
sampling, the children had been engaged in indoor activities and had not washed their hands for at
135
least 30 minutes. A sterile 5x5 cm gauze compress soaked in 3 mL >99.5% isopropanol (Sigma-
136
Aldrich) was used to wipe the palm, back of the hand and between the fingers on both hands of the
137
child. The compress was enclosed in a glass jar and stored at -20°C until analysis. Field blank samples
138
were collected from one third of the preschools by soaking a gauze compress in isopropanol and
139
placing it directly into the glass jar.
140
Chemical analysis of dust and hand wipes. The dust and hand wipe samples were analyzed at the
141
Department of Environmental Science and Analytical Chemistry (ACES) at Stockholm University,
142
Sweden. Details of the chemicals and materials used, sample treatment, extraction, clean up,
143
instrumental analysis, quality control, recovery and analysis of standard reference material (SRM
144
2585) can be found in the supporting information.
145
Briefly, the samples were spiked with isotopic labelled standards of BDE-155, BDE-209, EH-TBB,
146
BEH-TEBP, α-, β-, and γ-HBCDD, TBBPA, TCEP, TPHP, and TBOEP and extracted repeatedly with
147
a mixture of n-hexane/acetone 1:1. The first step in the clean-up of all raw extracts was done by
148
fractionation on a silica column. The first set of 30 dust samples were fractionated into three fractions
149
according to Ionas and Covaci.34 The clean-up of each fraction is described in Figure SI-1. The silica 6 ACS Paragon Plus Environment
Page 7 of 30
Environmental Science & Technology
150
column in this clean-up method was found to behave unreliably as analytes sometimes eluted in other
151
fractions than expected, which resulted in some missing data for BEH-TEBP and HBCDD. Therefore,
152
for the remaining 70 dust samples and the hand wipes, the first fractionation column was changed to
153
the silica column described by Sahlström et al. 201235 with some modifications. The clean-up for each
154
fraction by this method is described in Figure SI-2.
155
After the clean-up, the sample volumes were reduced under a gentle stream of nitrogen and
156
transferred to autosampler vials containing a recovery standard. Fractions containing PBDEs, α- and β-
157
DBE-DBCH, DBDPE, EH-TBB and BEH-TEBP were analyzed using a gas chromatograph (GC)
158
(Trace GC Ultra) coupled to a mass spectrometer (MS) (DSQ II MS; both Thermo Scientific,
159
Waltham, USA). Fractions containing PFRs were analyzed with a Trace 1310 GC coupled to an ISQ
160
MS (both Thermo Scientific). Fractions containing HBCDDs and TBBPA were analyzed using an
161
ultra-performance LC (ACQUITY™ UPLC) coupled to a tandem-quadrupole MS (Xevo™ TQ-S).
162
The UPLC/MS instrument and columns used were from Waters (Milford, USA). Details on GC-MS
163
and UPLC/MS parameters are described in Table SI-3 and SI-4.
164
Six quality control samples (SRM 2585), 8 solvent blanks and 26 field blanks were processed
165
together with the dust samples, which were divided into 6 batches. Data were corrected for the mean
166
levels detected in the blanks (BDE-209, TCEP, TCIPP, TPHP and TBOEP; Table SI-5). The filter
167
holders that were used to collect the dust samples were found to contain TCEP and TCIPP, leading to
168
high levels of these in the field blanks. Therefore the mean field blank levels were used in the data
169
correction for TCEP and TCIPP. Together with the hand wipes, 10 solvent blanks and 10 field blanks
170
were processed, divided into 5 batches. The hand wipe field blanks contained TDCIPP that was not
171
present in the solvent blanks. The source for this is unknown. The hand wipe results were corrected for
172
the mean blank levels of BEH-TEBP, BDE-209, TCEP, TCIPP, TDCIPP, TPHP and TBOEP (Table
173
SI-5). Limits of quantification and detection (LOQ and LOD) for compounds present in the blanks
174
were set as 5 and 3 times the standard deviation (for LOQ and LOD values, see Table SI-6). If not
175
present in the blanks, signal to noise (s/n) ratios of 10 and 3 were used as LOQ and LOD, respectively.
176
The absolute recovery of the different surrogate standards were generally above 60% for dust and
177
hand wipes (Table SI-7). For MTBBPA in hand wipes, the average recovery was only 41%, caused by 7 ACS Paragon Plus Environment
Environmental Science & Technology
Page 8 of 30
178
low recoveries in two of the five sample batches. The reason for this is unknown. Also, the recovery of
179
Mα-HBCDD was lower than or just above 60% (54% and 62% for hand wipes and dust, respectively)
180
when using clean-up method B. This is due to the partial elution in fraction II from the silica column
181
used. For analytes lacking a labelled equivalent (α- and β-DBE-DBCH, DBDPE, TCIPP, TDCIPP),
182
the relative recovery versus the surrogate standard was established with spiked samples and used to
183
correct the final concentrations (Table SI-7).
184
The dust reference material SRM 2585 is certified for some of the analytes (BDE-47, -99, -100, -
185
153 and -209)36 and the concentrations obtained in this study were comparable with the certified
186
values (Table SI-8). For the other analytes in this study there are no certified values but for most of
187
them there exist values published by others and our data were in line with published values.
188
Chemical analysis of urine. Urine concentrations of the metabolite of TPHP, diphenyl phosphate
189
(DPHP), were analyzed at the laboratory of Occupational and Environmental Medicine in Lund,
190
Sweden, using liquid chromatography-tandem-mass-spectrometry (LC/MS/MS) by a modified method
191
described in Bornehag et al 2015.37 Briefly, urine was added with 3H10-DPHP as internal standard and
192
treated with glucoronidase. The samples were thereafter analyzed without any further work-up. A C18
193
column was used prior to the injector to reduce the interferences of contaminants in the mobile phase.
194
The DPHP metabolite was separated on a Genesis Lightn. C18 column (4µm, 50x2.1mm) using 0.1%
195
ammonium hydroxid in water and methanol as mobile phase. The samples were thereafter centrifuged
196
and 4 µl of the supernatant was analyzed using a LC (UFLCXR, Shimadzu Corporation, Kyoto, Japan)
197
connected to the MS/MS (QTRAP 5500, AB Sciex, Foster City, CA, USA). The samples were
198
analyzed in duplicate. The limit of detection, determined as the concentration corresponding to three
199
times the standard deviation of the response in chemical blanks, was 0.04 ng/ml. In the analysis, two
200
homemade quality control samples were used. Coefficient of variation of the samples were 12% at 0.8
201
ng/ml and 9% at 2 ng/ml. Urine density was determined using a hand refractometer.
202 203
Exposure assessment. The daily exposures doses (DED) of BFRs and OPEs from preschool dust ingestion in 4-year-old children were calculated using the following equation:38 =
∗ 8
ACS Paragon Plus Environment
Page 9 of 30
Environmental Science & Technology
204
Cdust is the concentration of chemicals in dust. Idust is the daily intake of dust from the preschool
205
environment (30 mg), assuming that the total daily dust intake during the waking hours is 60 mg and
206
that children spend half of that time in the preschool.39 BW is the mean body weight of the children in
207
our study (17.6 kg). We assumed the absorption to be 100%.
208 209
The daily intake of BFRs and OPEs in preschool dust via dermal absorption was calculated using the following equation:38
= 210
∗ ∗ ∗ ∗ ∗
BSA is the exposed body surface area (hands, arms, legs) of children, equal to 3380 cm2.39 DA is the
211
amount of dust adhered to the skin, weighted for the exposed body parts (0.03 mg/cm2).39 The
212
absorption factors (AF) were 0.03 for all BFRs, 38 0.28 for TCEP, 0.25 for TCIPP, 0.13 for TDCIPP
213
and 0.22 (average of the other OPEs) for TPHP and TBOEP.40 TF is the fraction of the day spent in
214
the preschool (8/24 hours).
215
Statistical analysis. The data were analyzed using the statistical software SPSS 22 (IBM Inc.) and
216
STATA 13 (Statacorp 220 TX, USA).Values below the LOD were replaced by LOD/√2 and values
217
between the LOD and the LOQ were replaced by LOQ/√2. Due to high LODs and LOQs for single
218
dust samples, we did not include one value for BDE-99, BDE-209, TBBPA and TPHP, three values
219
for BDE-100 and four values for DBDPE in the statistical analysis. Compounds with more than 50%
220
of the values below the LOQ were not included in the statistical analysis.
221
To correct for the degree of dilution in the urine samples, levels of DPHP were adjusted to
222
children’s mean density of 1.023 kg/L according to the formula Csample x (1.023-1)/(Dsample-1), where
223
Csample is the concentration of DPHP in the sample and Dsample is the density of the sample.
224
The data were not normally distributed and non-parametric tests were therefore used. Associations
225
between factors in the preschool environment and levels of flame retardants in preschool dust were
226
analyzed using the Mann Whitney U-test. Variables from this univariate analysis having significance
227
levels LOQ >LOD
N
GM
a
HAND WIPE SAMPLES
Median P95
Range
% % >LOQ >LOD
N
ng/g dust
GM
a
Median
P95
Range
pg/hand wipe
BDE-47
99%
100%
100
9.6
7.7
110
