High levels of medium-chain chlorinated paraffins and polybrominated diphenyl ethers on the inside of several household baking oven doors

Science of the Total Environment 615 (2018) 1019–1027

Contents lists available at ScienceDirect

Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv

High levels of medium-chain chlorinated paraffins and polybrominated diphenyl ethers on the inside of several household baking oven doors☆ Christoph Gallistl, Jannik Sprengel, Walter Vetter ⁎ University of Hohenheim, Institute of Food Chemistry (170b), Garbenstr. 28, D-70599 Stuttgart, Germany

H I G H L I G H T S

G R A P H I C A L

A B S T R A C T

• Fat residues on the inner baking oven surface revealed high HFR levels. • MCCPs were predominant in ~ 50% of the samples with levels in the mg/g fat range. • High MCCP concentrations indicated they originated from the baking oven. • Further compounds (PBDEs, PCBs, DBDPE, DP) were also detected at lower levels.

a r t i c l e

i n f o

Article history: Received 23 June 2017 Received in revised form 15 August 2017 Accepted 12 September 2017 Available online xxxx Editor: Adrian Covaci Keywords: Brominated flame retardants Polybrominated diphenyl ethers (PBDEs) Chlorinated paraffins Polychlorinated n-alkanes Indoor contamination Kitchen

a b s t r a c t Fat obtained by wipe tests on the inner surface of 21 baking ovens from Stuttgart (Germany) were analyzed for halogenated flame retardants (HFRs), namely polybrominated diphenyl ethers (PBDEs), decabromodiphenyl ethane (DBDPE), dechlorane plus (DP), short- and medium-chain chlorinated paraffins (SCCPs, MCCPs), as well as polychlorinated biphenyls (PCBs). In ~50% of the samples chlorinated paraffins (CPs) were present in the mg/g fat range, i.e. three to four orders of magnitude higher concentrated than the sum of all other target compounds. In contrast the remaining ~50% of the samples were free of CPs, while the other HFRs were comparable in CP-positive and CP-negative samples. The exceptionally high concentrations and exclusive presence of CPs in half of the samples produced strong evidence that these compounds were released from the baking oven itself. This hypothesis was supported by detection of MCCPs at even higher concentrations in the inner components of one dismantled baking oven. The release of substantial amounts of HFRs from the oven casing during its use may contribute to human exposure to these compounds, especially MCCPs and SCCPs. © 2017 Elsevier B.V. All rights reserved.

1. Introduction

☆ This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. ⁎ Corresponding author. E-mail address: [email protected] (W. Vetter).

https://doi.org/10.1016/j.scitotenv.2017.09.112 0048-9697/© 2017 Elsevier B.V. All rights reserved.

The growing use and importance of electronic equipment in daily life has been accompanied by improved safety standards regarding flammability mitigation; most consumer products are required to contain some sort of physical or chemical mechanism to decrease flammability. In terms of fire protection, halogenated flame retardants (HFRs) provide several benefits compared to other non-halogenated FRs,

1020

C. Gallistl et al. / Science of the Total Environment 615 (2018) 1019–1027

including low production costs, effective flame retardation particularly in the early stages of a growing fire, and easy application without negative effects on the material functionality of the treated material (Beard, 2013; Wang, 2008). However, these benefits do not outweigh the potential adverse effects of HFRs for consumers and animals, as well as their negative long-term effects on the environment (Shaw et al., 2010). For instance, harmful effects on human health due to HFR exposure raised concerns about their widespread application in consumer goods and building materials (Birnbaum and Staskal, 2004). In addition, polybrominated diphenyl ethers (PBDEs) and other HFRs are applied as additives, i.e. they are not chemically bound to the flame-retarded material, and can leach out of the treated products followed by their accumulation in indoor dust and other indoor matrices (de Boer et al., 2016; Eljarrat and Barceló, 2011; Hale et al., 2006; Schecter et al., 2009). Between the 1970s and the 1990s, PBDEs were the predominant group of HFRs with an estimated global production of 67,400 metric tons per year (Alaee et al., 2003; Birnbaum and Staskal, 2004). Their detrimental environmental properties (persistence, bioaccumulation, toxicity, and long-range-transport) eventually caused a voluntary phaseout (e.g. in the U.S.) or an outright ban on PBDE production and use (Alaee et al., 2003; Dodson et al., 2012). Discontinuation of these PBDE products in turn necessitated the introduction of alternatives. Chlorinated paraffins (CPs) are a currently used type of HFR only regulated in few national directives (Glüge et al., 2016). Their ubiquitous use also makes CPs the most relevant HFR class in terms of global production volume. Between 1935 and 2012 N 13 million metric tons of CPs were produced, and the production rates are still on the rise (Glüge et al., 2016). For instance, CP production is approximately ten times the total production volume of polychlorinated biphenyls (PCBs) (Breivik et al., 2002). In May 2017, short-chain chlorinated paraffins (SCCPs), a sub-group of CPs with chain lengths between ten and thirteen carbon atoms, were classified as persistent organic pollutants (POPs) in Annex A of the Stockholm Convention, i.e. their production and use is strongly restricted, although several exemptions were made, including their application in plastics (IISD Reporting Services, 2017; www.chm.pops.int). These restrictions may also result in an increased use of the unregulated CP sub-groups, namely the mediumchain (MCCPs) and long-chain chlorinated paraffins (LCCPs). Previous studies involving kitchen hoods, dishcloths and kitchen floor dust have indicated significant concentrations of CPs and other HFRs in the kitchen environment (Bendig et al., 2013b; Gallistl et al., 2017; Kuang et al., 2016). Furthermore, electric appliances like deep fryers, microwaves and hand blenders, as well as plastic consumer products (partly made of recycled materials), which are typically located within the kitchen, have been shown to contain HFRs and might therefore also represent potential HFR sources (Gallen et al., 2014; Samsonek and Puype, 2013; Yuan et al., 2016). Hence, the kitchen may be a potential HFR hotspot within the typical family home, due to the high density of HFR-containing appliances and products and the generation of heat by kitchen appliances (Kuang et al., 2016). Baking ovens (in this study defined as in any kitchenette integrated electrical appliances, which are used for heating, baking or roasting of foodstuff by electrically produced heat) may represent particular culprits of HFR introduction to the kitchen environment, as they are at least in Germany frequently present in the household kitchen. For instance, in 2011 93% of German households contained a baking oven (Forsa survey, 2010, access via Statista (https://de.statista.com/)). In addition, they are explicitly in place to heat food to high temperatures, thus they require commensurate flame retarding measures. However, to the best of our knowledge, baking ovens have not been investigated as a possible HFR source within the kitchen. Therefore, the aim of this study was to quantify HFRs on the inner oven door surfaces of baking ovens from German household kitchens. Twenty-one baking ovens were sampled (one oven twice, before and after pyrolysis program) and samples were collected by means of standardized wipe tests, followed by analysis for PBDEs and PCBs (major

congeners in technical mixtures), as well as for current-use HFRs (SCCPs, MCCPs, decabromodiphenyl ethane (DBDPE) and dechlorane plus (DP)). In addition, investigation on potential HFR sources was performed using components obtained from a dismantled baking oven. 2. Materials and methods 2.1. Chemicals and standards Cyclohexane (≥99.5%) and ethyl acetate (≥99.5%), both from SigmaAldrich (Seelze, Germany), were aceotropically distilled (C/E mixture, 46:54, w/w). Silica gel 60 (for column chromatography), 2,2,4trimethylpentane (isooctane, for pesticide residue analysis) and anhydrous sodium sulfate (p.a., ≥99%) were also from Sigma-Aldrich, whereas acetone (p.a., ≥ 99%) and n-hexane (for pesticide residue analysis, ≥99%) were obtained from Th. Geyer (Renningen, Germany). Concentrated sulfuric acid (N98%) was from BASF (Ludwigshafen, Germany) and demineralized water was produced in-house by means of an ELGA purelab classic ultrapure water system (Celle, Germany). Polybrominated diphenyl ether (PBDE) congeners BDE 153, BDE 154, BDE 183 (Great Lakes Chemical, Indianapolis, USA) and BDE 209 (Cambridge Isotope Laboratories, Twerksbury, MA, USA) were combined and diluted in isooctane (BFR-mix). The commercial technical PBDE mixtures DE-71 (c-PentaBDE), DE-79 (c-OctaBDE) and DE-83 (cDecaBDE) were from Great Lakes Chemical. A mix of 39 PBDEs (BDEAAP-A), decabromodiphenyl ethane (DBDPE), as well as technical dechlorane plus (syn- and anti-DP) were obtained from AccuStandard (New Haven, USA). Perdeuterated α-1,2,3,4,5,6-hexachlorocyclohexane (α-PDHCH, recovery standard, purity: N 98%) and 6′-MeO BDE 66 (BCIS, injection standard, purity: N 99%) were synthesized in our laboratory (Vetter et al., 2011; Vetter and Luckas, 1995; von der Recke and Vetter, 2007). Analytical standard solutions of the individual polychlorinated biphenyl (PCB) congeners PCB 28, 52, 99, 101, 114, 118, 123, 138, 141, 144, 151, 153, 170, 180, 183, 187, 194, 196 and 201 (all from Dr. Ehrenstorfer, Augsburg, Germany) were combined (PCB mix). Technical mixtures of short-chain chlorinated paraffins, SCCPs (51.5%, 55.5% and 63% chlorine content) and medium-chain chlorinated paraffins, MCCPs (42%, 52% and 57% chlorine content), all from Dr. Ehrenstorfer, were mixed to give additional SCCP mixtures with 53.5% and 59.3% and MCCP mixtures with 47% and 55% chlorine content, respectively. 2.2. Study design Twenty-one baking ovens from households in Stuttgart (South of Germany) were sampled by volunteers between February–April 2016 by means of wipe tests. This sampling technique is a non-destructive sampling method, which has also been successfully applied in previous studies to investigate HFRs on surfaces (Gallen et al., 2014; Watkins et al., 2011). Participants were not informed beforehand of the aims of his study and sampling was conducted using ovens in the condition in which they were found. Information about the baking oven (manufacturer, model, age and availability of a pyrolysis program, are included in Table S1, Supporting information) were collected by means of a questionnaire, but no information was available about cooking and cleaning frequencies. 2.3. Sampling The wipe tests were performed with cotton pads (diameter: 6 cm, obtained from a retail store in Stuttgart and previously tested to be free of any targeted HFRs), which were stored in aluminum foil until use. Directly before sampling, the cotton pads were wet with 5 mL nhexane and the entire surface area of the oven door was sampled by wiping the wet cotton pad in a continuous circular pattern across the entire area. The sample cotton pads were then placed in 20 mL screw cap brown glass vessels and were stored at −20 °C in darkness until

C. Gallistl et al. / Science of the Total Environment 615 (2018) 1019–1027

1021

sample preparation. In one case (baking oven #8) an additional sample was facilitated after performance of a pyrolysis cleaning program of the baking oven. Only one half of the surface of the inside of the door was wiped prior (#8A) and the second half was wiped after (#8B) the cleaning program. The baking oven was not used for food preparation between the two sample collections. Components (inside of the oven door, surfaces of the cable insulation, circuit board housing, exhaust shaft, and thermal insulation, Fig. S1, Supporting information) of a cast-off, ten year old baking oven, which was stored unused for N1 year within a household basement, were sampled in the same way by wipe tests with cotton pads. Optical inspection of the circuit board housing and the cable insulation did not indicate the presence of adherent dust, while low amounts of dust could not be excluded on the exhaust shaft and the thermal insulation. The surface of the baking oven door was similar to those of the other sampled baking ovens. The wipe samples of the components were stored as described above.

temperature vaporizing (PTV) injector (Mülheim, Germany) (Gallistl and Vetter, 2016). The applied GC oven program and further GC measurement parameters were identical with those reported previously by Gallistl and Vetter (2016) and are also briefly mentioned in the Supporting information.

2.4. Sample clean-up

3. Results and discussion

After sampling, cotton pads were cold-extracted (5 min in ultrasonic bath) three times with 15 mL n-hexane in the brown glass screw caps vessels. Extracts were combined in a pre-weighed 100 mL amber glass pear-shaped flask and the solvent was evaporated by means of a rotary evaporator (30 °C, 320 mbar) to ~1 mL, followed by further evaporation to complete dryness by a gentle stream of nitrogen. The residue was weighed, re-dissolved in 3 mL n-hexane, transferred into a 20 mL test tube with a screw cap and supplemented with 10 μL α-PDHCH (10.7 ng/μL, recovery standard, diluted in isooctane) and 5 mL concentrated sulfuric acid. Vials were shaken and left to react for at least 1 h. In some cases, phases did not separate after 1 h. In these cases, phase separation was accelerated by an additional centrifugation step (5 min at 4000 rpm). Then the upper organic layer was removed and the sulfuric acid phase was re-extracted twice with 5 mL n-hexane. The combined organic extracts were evaporated to b1 mL and subjected to adsorption chromatography (1 cm i.d. glass column, filled with 3 g silica gel deactivated with 30 wt% water, topped with ~ 0.5 cm water-free Na2SO4) according to Vetter et al. (1998). Target compounds were eluted with 60 mL n-hexane. The solution was first concentrated by a rotary evaporator (30 °C, 320 mbar) to a volume of ~2 mL before it was further evaporated by a gentle stream of nitrogen (30 °C) to a final volume of 1 mL. An 800 μL aliquot of this sample was concentrated to 80 μL (nitrogen stream, 35 °C) and spiked with 10 μL BCIS (1.0 ng/μL, injection standard, diluted in isooctane) before instrumental analysis (Section 2.6), whereas the remaining 200 μL was kept as reserve sample. Quantification of SCCPs and MCCPs was performed by means of an aliquot of the unconcentrated solutions, which were also spiked with 10 μL BCIS.

3.1. Characterization of the samples

2.5. Quality assurance/quality control (QA/QC) Before use, all glassware was thoroughly pre-cleaned with hot water, detergent, demineralized water, acetone and C/E mixture (46:54, w/w). Amber or clear glassware covered with aluminum foil was used to minimize exposure to light. At least one procedural reagent blank was processed alongside with each sample batch consisting of 4–6 samples to verify that contamination did not occur during the clean-up. Field blank and procedural reagent blanks were free of any of the target compounds (blimit of detection, LOD) and recovery tests verified a sufficient extraction of target compounds (further information presented in the Supporting information and Fig. S2, Supporting information). 2.6. Gas chromatography coupled to mass spectrometry (GC/MS) Analyses were performed with an Agilent 7890/5975c system (Waldbronn, Germany) equipped with a Gerstel CIS-4 programmed

2.7. Data analysis All statistical evaluations were conducted by IBM SPSS Statistics 23. Normal distribution was tested by the Shapiro-Wilk test and was verified for sample weight (p = 0.142). Log-normal distribution was only observed for ΣPCBs and sum level of all brominated compounds (ΣBr), but not for the other compounds and compound classes (Table S5, Supporting information). Therefore, correlations between the individual compounds and compound classes were conducted by non-parametric test, i.e. using Spearman correlation coefficients. The level of significance was generally set at a value of α = 0.05.

Nineteen baking oven models (age: 1–20 years) from eight manufacturers (“A”-“H”) were present in the 21 different household kitchens (Table S1). The weight of fat (i.e. the hexane-soluble residue on the surface, which most likely consisted predominantly of lipid matrix from prepared food and if at all only minor amounts of dust, since it was apparently free of any particles) collected from the inner surface of the baking oven doors ranged from 1.0–200 mg (geometric mean: 54 mg) with an average wiped surface area of 2400 cm2. In the following discussion, levels of polyhalogenated compounds will generally be based on the sample weight and not on the wiped area (surface area based concentrations additionally presented in Table S6, Supporting information), because in a previous study on kitchen hood fat deposits it was shown that concentration of HFRs correlated with the amount of deposited fat (Bendig et al., 2013b). Similarly, a significant positive correlation between the absolute sum levels of polybrominated (∑ Br), polychlorinated (∑Cl) and polyhalogenated (∑Hal, i.e. sum of ∑Br and ∑Cl) compounds and the sample weight (Spearman correlation coefficients: ∑Br: ρ = 0.465, ∑Cl: ρ = 0.430 and ∑Hal: ρ = 0.531 each with p b 0.05) was observed in the present study. Stronger correlations between ∑ Hal and the sample weight compared to those of ∑Br or ∑Cl can be explained by cumulative effects due to summation of ∑Br and ∑Cl levels. Initial non-targeted measurements indicated that SCCPs, MCCPs, PBDEs, PCBs, DBDPE and DP were the most relevant acid-stable HFRs. All samples contained at least one of these polyhalogenated compounds, but the concentrations varied by several orders of magnitude (0.01–93,200 μg/g fat residue, Fig. 1a). Without consideration of chlorinated paraffins, polybrominated compounds (n.d.-246,200 ng/g) were more abundant than polychlorinated compounds (n.d.-2300 ng/g). In contrast, when CPs were included, their proportion amounted 99.8– 100% of the sum of all targeted HFRs in the CP-positive samples (Fig. 1a). However, concentrations of PCBs, only ranged from n.d.-2300 ng/g (geometric mean: 21 ng/g, Table 1); PCBs are usually found in the highest concentration in food compared to other polyhalogenated compounds (Darnerud et al., 2006). Yet, the age of the sampled ovens likely explains this apparent anomaly. The production of PCBs was phased-out in the 1970s, therefore their application in the baking ovens (e.g. as HFRs) can be excluded, because the oldest sampled baking oven was produced in 1992 (Table S1, Supporting information). Due to the sampling design, it is very likely that detected HFRs directly released from the oven materials would result in exceptionally high observed concentrations, presumably exceeding those of PCBs. The detected PCBs in the samples were therefore most likely from an external source and not the baking oven itself, with release from prepared food as a plausible

1022

C. Gallistl et al. / Science of the Total Environment 615 (2018) 1019–1027

Fig. 1. Detected levels (in logarithmic scale) of a) all polychlorinated and polybrominated target compounds (in decreasing order, the relative proportion of the sum of all polychlorinated compounds [%] is illustrated by red values within the bars), of b) of short- (SCCPs) and medium-chain chlorinated paraffins (MCCPs) and of c) non-209 PBDEs and BDE 209. Levels of the compounds/compound classes were classified in high moderate levels (further information in the text).

introduction pathway. For this reason, a “critical” concentration slightly above the maximum level of PCBs, i.e. 2500 ng/g fat, was selected to assess and differentiate between internal release (HFR levels N 2500 ng/g fat, in the following defined as “high level”) from external sources, like prepared food, indoor air and dust (HFR levels b2500 ng/g fat, in the following defined as “moderate level”). In agreement with this hypothesis, CPs were, if present, three to four orders of magnitude higher in concentration than PCBs. In addition to that, the sample extracts contained no visible particles, hence adherent dust might play only a minor role in contamination of the surface, although it cannot generally be excluded that low amounts of dust were also sampled via the wipe tests. Considering a wide range of compounds, we noted that CPs (i.e. medium- and short-chain chlorinated paraffins, MCCPs and SCCPs) and major congeners of technical polybrominated diphenyl ethers (PBDEs) mixtures were particularly abundant in the collected wipe samples. These two groups were evaluated individually while further compounds were discussed together in one subchapter. 3.2. Chlorinated paraffins (CPs) When detected, CPs (sum of MCCPs and SCCPs) generally represented the highest proportion (on average of all other HFRs were also exclusively in the range of “high levels” (range: 2200–93,200 μg/g, Fig. 1b). Noteworthy, the highest four reported MCCP levels were above the calibration range and were calculated by extrapolation. The fact, that these compounds were either detected in exceedingly high concentrations in ~50% of the samples or otherwise were not detected, strongly pointed towards their origination from the baking oven itself. In these ten

samples, CPs typically represented the highest proportion of all HFRs, whereas the sum concentration of all other target compounds was three to four orders of magnitude lower abundant and therefore in the same range as those of the CP-negative samples (Fig. 2). Furthermore, the mean and maximum CP levels were one order of magnitude higher than the highest reported CP levels in house dust so far (Wong et al., 2017). At this point it cannot be excluded that CPs, emitted from the baking oven when it is open, may contribute to the CPs frequently detected in fat deposits in kitchen hoods (Bendig et al., 2013b) and also in dishcloths after their common use in household kitchens (mean CP level: 4900 ng/dishcloth, Gallistl et al., 2017). CPs might be collected also from the surface of the baking oven, for example for cleaning purposes of the baking oven door or rubber seal, which might be flame-retarded with CPs (Glüge et al., 2016; Takasuga et al., 2012). However, further research is required to investigate this point. For instance, one further known CP-source within the kitchen environment could be release from hand blenders (Yuan et al., 2016). Release and/or extraction from painted components like the oven door frame might play also an important role since CPs are also used as additives in paint (Zeng et al., 2013; Reth and Oehme, 2004). In addition, data from China suggest that cooking oils may be contaminated with CPs (Cao et al., 2015): However, the reported SCCP concentrations were at least 40 times lower than in the CP-positive baking oven samples and MCCPs, which were predominant in our samples, were not detected in cooking oils. Therefore, the contribution from release via cooking oil was considered to be low. Noteworthy as well, Schinkel et al. (2017) recently reported formation of chlorinated alkenes after heating of CPs at 220 °C, which might also occur in the baking oven, since these electric appliances

C. Gallistl et al. / Science of the Total Environment 615 (2018) 1019–1027

1023

Table 1 Levels of polyhalogenated compounds [ng/g fat] detected in the baking oven samples (∑Hal: sum of all targeted compounds; MDL/MQL: method detection/quantification limit in [ng/g]; DF: detection frequency [%]; n.d., not detected). Sample no.

Sample weight [mg]

∑non-209 PBDEs

BDE 209

∑PCBs

MCCPs

SCCPs

DBDPE

∑DP

∑Hal

#1 #2 #3 #4 #5 #6 #7 #8A #8B #9 #10 #11 #12 #13 #14 #15 #16 #17 #18 #19 #20 #21 Min Max Mean Geometric mean Median

1.0 13.3 58.7 22.9 135 52.1 26.3 179 45.6 94.9 147 63.2 44.1 109 132 87.2 195 119 155 50.0 20.8 23.1 1.0 195 79.3 51.5 61.0 MDL MQL DF

25 1200 320 n.d. bLOQ 240,000 10 n.d. 24 76 bLOQ 110 n.d. 16 bLOQ 72 bLOQ 2500 bLOQ 9400 220 n.d. b1.8 240,000 11,500 34 20 1.8 6.2 82%

420 bLOQ 39 n.d. n.d. 6200 180 bLOQ n.d. 65 15 8.1 n.d. bLOQ 6.9 1200 30 36 n.d. 490 bLOQ n.d. b3.0 6200 400 20 8.1 3.0 10 73%

bLOQ 60 35 40 120 n.d. 500 58 24 93 10 15 n.d. 11 2300 2.7 n.d. n.d. 1400 72 400 4.6 b1.1 2300 230 21 30 1.1 3.6 82%

13,000,000 4,500,000 n.d. 4,500,000 21,600,000 20,900,000 n.d. n.d. n.d. n.d. n.d. n.d. 93,200,000 n.d. 35,800,000 n.d. 19,800,000 1,900,000 61,900,000 n.d. n.d. n.d. b3.2 93,200,000 12,600,000 100,000 1100 3.2 11 45%

n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. 1,900,000 n.d. n.d. 320,000 1,800,000 n.d. n.d. n.d. b5.0 1,900,000 180,000 5600 1700 5.0 17 14%

n.d. n.d. 9.4 n.d. n.d. n.d. 30 bLOQ n.d. 17 16 n.d. n.d. n.d. n.d. 10 190 n.d. n.d. n.d. n.d. n.d. b2.9 190 13 3.2 0.95 2.9 9.5 32%

800 n.d. 20 n.d. n.d. n.d. 16 n.d. n.d. n.d. 4.1 n.d. n.d. n.d. n.d. 18 13 n.d. n.d. n.d. n.d. n.d. b0.60 800 40 1.0 0.2 0.60 2.0 27%

13,000,000 4,500,000 420 4,500,000 21,600,000 21,100,000 740 58 48 250 45 140 93,200,000 30 37,700,000 1300 19,800,000 2,200,000 63,700,000 10,000 620 4.6 4.6 93,200,000 12,800,000 34,000 5600

often work at temperatures above 200 °C. Hence, presence of olefinic CP-degradation products could not be excluded in the present study due to mass overlaps with CP fragment ions formed in GC/ECNI-MS (Schinkel et al., 2017). All CP-positive samples were dominated by medium-chain chlorinated paraffins (MCCPs) (Fig. 1b) with a geometric mean level of 2.4 μg/g (range: 1900–93,200 μg/g fat). MCCPs were generally more abundant and also more frequently detected than SCCPs (only samples

Fig. 2. Box-plot of the sum of polyhalogenated compounds without consideration of CPs (∑non-CPs in CP-positive samples) and CPs (∑non-CPs in CP-positive samples) in the CP-positive samples, as well as sum of polyhalogenated compounds in the CP-negative samples (∑non-CPs in CP-negative samples).

#14, #17 and #18 were SCCP-positive, range: 320–1900 μg/g fat). Six of the ten MCCP-positive samples (from manufacturers “A”, “D”, and “H”) showed a very similar MCCP pattern which was dominated by hexachlorinated homologs (MCCP type 1 pattern, Cl6-isomers: 48– 52%, Fig. 3). Slight fluctuations in the MCCP congener group composition were observed for example within the MCCP type 1 samples which may be the result of different stability or volatility of selected MCCP congeners. No relationship was observed between the pattern and the age of the baking ovens, thus, other influences like the usage frequency or different applied temperatures of the baking oven may explain these deviations. In three further samples (#14, #18 and #2), the MCCP pattern was slightly shifted towards equally high or higher proportions of heptachlorinated homologs (MCCP type 2 pattern, Cl7-isomers: 29– 38%, Cl6-isomers: 29–31%, Fig. 3). In terms of the carbon chain length pattern, only little deviations were observed for samples with MCCP type 1 and type 2 pattern (Fig. S3, Supporting information). Both types were dominated by isomers with C14– (59–70%), followed by C15-chain lengths (26–33%) and low proportions of isomers with C16– and C17-chain lengths (≤10%). Predominance of C14-isomers in indoor matrices was also reported in previous studies, e.g. in indoor dust samples (Hilger et al., 2013; Huang et al., 2017; Shi et al., 2017; Wong et al., 2017). In contrast, sample #17 was unique in that it only showed C14– and C15-chain lengths (Fig. S3, Supporting information) with equal proportions of hexa- and heptachlorinated homologs (MCCP type 3 pattern, Fig. 3). Four samples with type 1 MCCP pattern were collected in baking ovens (1–25 years old) from the same manufacturer (manufacturer “A”). However, six further samples from baking ovens of manufacturer “A” were free of MCCPs. Noteworthy as well, there were eight different models among the ten baking ovens of manufacturer “A”. It is possible that some of the models were produced in different plants, possibly even in different countries. Moreover, manufacturers may be delivered by different component suppliers, which on their part may ensure flame-retardancy of their components by different HFRs. This would

1024

C. Gallistl et al. / Science of the Total Environment 615 (2018) 1019–1027

Fig. 3. MCCP pattern of congener groups with identical chlorination degree detected in the individual baking oven samples. Similar patterns are classified in type 1, type 2 and type 3. Manufacturers and age [years] of the sampled baking ovens are illustrated above the bars.

also explain why only one of three baking oven models from manufacturer “D” was MCCP-positive (#6), although its age was in between the two MCCP-negative samples (#11 and #15). Hence, this might also explain why no relationship was found between age and presence of MCCPs for the other two MCCP type 1 baking oven samples (#6 and #16) of manufacturers “D” and “H”. 3.3. CPs in baking oven components of a dismantled device One cast-off baking oven of manufacturer “H” was also analyzed (Section 2.3). The oven door wipe test sample of this baking oven showed MCCP type 1 pattern very similar to those found in another model from this manufacturer (sample #16). Following the initial wipe test, the discarded baking oven was dismantled and further wipe tests were performed on four further components of this appliance (Fig. S1, Supporting information). MCCPs were only detected in samples taken on the thermal insulation, the exhaust shaft, and the inside of the baking oven door (Fig. S4, Supporting information). Due to very low sample lipid weights within samples taken beyond the oven door, MCCP concentrations were calculated in relation to the wiped area to enable comparison of concentrations detected across the different oven components. The highest MCCP concentration (96,000 pg/cm2) was measured on the circuit board housing, whereas levels on the exhaust shaft and on the inner oven door surface (which were located close to the circuit board housing, Fig. S1, Supporting information) were only ~1/3rd and 1/100th of this value, respectively. In terms of chlorination degree and carbon chain length, all three components showed similar MCCP type 1 patterns (Fig. S5, Supporting information). Decreasing MCCP concentrations from the circuit board to the oven door indicated that MCCPs were likely released from the circuit board or from adjacent components, most likely due to high temperatures during the heating/baking process. Liberated MCCPs from heated components may then re-condense depending on the distance of the individual components from the circuit board housing (Fig. S1), i.e. in highest amounts on the exhaust shaft, followed by the more remote oven door. In contrast, no MCCPs were detected (b LOD) on the more remote cable insulation and the thermal insulation on the backside of the baking oven, which might be explained by decreased transfer of CPs to locations more distant from the circuit board or a better insulation of these components. The MCCP concentration in the wipe test on the inside of the door (~375 μg/g fat) was in the low range of CP-positive baking ovens

and agreed therefore with our hypothesis that CPs were released from the baking oven itself. However, other sources like MCCP-contaminated house dust or other external sources might also contribute to the detected MCCP concentration. With only one dismantled baking oven it is not possible to provide a general explanation describing the occurrence of the CPs detected on the other baking oven doors. Comparably high or even higher detection frequency (DF) of CPs (both SCCPs and MCCPs) has been observed in dishcloth samples commonly used in kitchens (DF of MCCPs: 58%, SCCPs: 21%, Gallistl et al., 2017), as well as in kitchen hood fat deposits (DF of CPs: 100%, Bendig et al., 2013b). The high frequency of contamination in these matrices may partly be explained by exposure to or contact with CP-treated baking oven components although other CPtreated devices might also contribute to their contamination, which is in support with the higher DF in these matrices. Ostensibly, CPs may be transferred to dishcloths through wiping of the oven surface or transferred to hoods via the outflow of contaminated hot air from baking oven. In addition to flame-retarded components a minor proportion of the detected CPs might also be released from kitchen utensils, which are typically placed in the baking oven during the food preparation, for example baking dishes made of silicon materials (Glüge et al., 2016; Takasuga et al., 2012), although it is unlikely that this would lead to a high concentration on the surface of the circuit board housing. 3.4. Polybrominated diphenyl ethers (PBDEs) PBDEs (Table S7, Supporting information) were detected in 82% of the samples with widely ranging concentrations (6.9–246,000 ng/g fat). However, only three (14%) samples showed ∑ PBDE levels N2500 ng/g fat (Fig. 1c). Two of the three above mentioned baking oven samples (#6 and #17) additionally showed high CP levels (Table S8, Supporting information), which might be explained by application of both HFRs (PBDEs and CPs) in one or more installed components of these 18 and 15 year old electrical appliances. The PBDE congener pattern in two of the “high level” samples (samples #6 and #19) was similar to technical c-OctaBDE (DE-79) while the third one (sample #17) displayed a pattern indicative of technical c-PentaBDE (DE-71) (La Guardia et al., 2006). Their presumed origin from the baking oven itself agreed with their production date and the phase-out timeline of these mixtures; all three baking ovens were produced before technical cPentaBDE and c-OctaBDE mixtures were phased out in 2004 (Alaee et

C. Gallistl et al. / Science of the Total Environment 615 (2018) 1019–1027

al., 2003). Ten further samples were in the range of “moderate” concentration. The mean PBDE concentration of 290 ng/g fat (geometric mean: 84 ng/g fat) in these ten samples was comparable with PBDE levels found in indoor dust samples within the literature (Table S9, Supporting information). Median proportion of BDE 209 (29%) to ∑ PBDE levels was lower than observed in kitchen hood fat deposit samples (median: 50%, Bendig et al., 2013b) and dishcloths commonly used in kitchens (median: 55%, Gallistl et al., 2017) in Germany, and was also significantly lower than the value reported for settled kitchen floor dust (median: ~ 95%, Kuang et al., 2016) from the UK. In six samples BDE 209 represented N 90% of the PBDE pattern (Fig. 4) and reflected therefore the pattern of technical c-DecaBDE (DE-83) mixture (La Guardia et al., 2006). The three samples with the highest BDE 209 levels (samples #6, #15 and #19) were taken from baking ovens manufactured before 2008, i.e. before the ban of technical c-DecaBDE in electric devices (including baking ovens) in the EU (directives 2002/95/EC and directive 2011/ 65/EU of the European Parliament). Sample #15 showed “moderate” PBDE levels, for which it was considered uncertain that BDE 209 originated from the baking oven itself. However, it seemed rather unlikely, that BDE 209 originated from foodstuff heated in the baking oven. Besides for the baking oven itself, BDE 209 contamination may have occurred by different sources, e.g. via dishware for backing, during usage of flame-retarded textiles like oven mitts or via introduction of indoor air particles while the baking oven door was open. Hence, BDE 209 may originate from other old kitchen devices, probably installed at the same time as the baking oven (installed in 2003). Further support of this hypothesis was derived from the dismantled baking oven (section 3.3). This oven showed PBDEs in all five components with clear dominance of BDE 209 (N 90% of ∑PBDE levels) in all tested components. The PBDE concentration decreased sequentially from thermal insulation (1700 pg/cm2 = 100%), the exhaust shaft (20%), the circuit board housing (17%), the baking oven door, and the cable insulation (1%). Notably, the concentration observed on the baking oven door of 3200 ng/g fat was in the “high level” range, which is in agreement with the detection of higher proportions of BDE 209 in the inner parts of the dismantled oven, as it was observed in the case of MCCPs. For further exploration of potential sources from inside/outside the oven we studied the ratio of BDE 47 and PCB 153 (usually the predominant congeners represented in foodstuff, Fattore et al., 2008; Hayward et al., 2007; Ryan and Patry, 2001) in the 21 baking oven samples. Currently, PCB 153 levels in foodstuff samples are generally higher than those of BDE 47 (PCB 153/BDE 47 ratio: 3.0–5.0, Table S10, Supporting Information), whereas BDE 47 is more abundant than PCB 153 in

1025

contemporary indoor samples (PCB 153/BDE 47 ratio: 0.01–0.40, Table S11, Supporting information). Since both, PCBs (Takaoka et al., 2000) and BDE 47 (Bendig et al., 2012; Bendig et al., 2013a) have been shown to be stable to decomposition by heat at temperatures, which are typical for food preparation or even higher, influence of the accelerated temperatures in the baking oven on the PCB 153/BDE 47 ratio is most likely neglectable. Therefore, the PCB 153/BDE 47 ratio may serve as an indicator to identify food-born PCB/PBDE sources in the baking oven samples. Eight samples contained both compounds with PCB 153/BDE 47 ratios ranging from 0.04 (25-fold excess of BDE 47) to 1.7 (excess of PCB 153) (Fig. S6, Supporting information). Importantly, the sample with the highest PCB 153/BDE 47 ratio belonged to those samples with “high” PBDE levels (N2500 ng/g fat), with a congener pattern similar to those of c-DecaBDE (see above), i.e. in this sample BDE 47 contributed only a negligible share to the ∑ PBDE concentration and the baking oven itself was most likely the origin of the PBDEs. Surprisingly, none of the samples showed a PCB 153/BDE 47 ratio in the typical range of foodstuff (blue area in Fig. S6, Supporting information). In fact, the remaining seven baking ovens were produced after the ban of technical c-Penta- and c-OctaBDE. For these samples, BDE 47 was very unlikely to originate from the oven itself. However, since use of recycled cPenta- or c-OctaBDE containing materials for production of the mentioned baking ovens might also represent a potential route of reintroduction of these HFRs (Li et al., 2013; Sindiku et al., 2015), origin of BDE 47 via recycled material cannot be completely excluded. Under the assumption that PCBs originated from prepared foodstuff, the detected low PCB 153/BDE 47 ratios indicated that release of BDE 47 from foodstuff was not the major contamination pathway. Hence, the bulk of BDE 47 most likely originated from outside the baking oven. In fact, the PCB 153/BDE 47 ratio of samples #11 and #2, both with “moderate” PBDE levels, mirrored typical ratios detected in indoor samples (green area in Fig. S6, Supporting information). The remaining five samples showed PCB 153/BDE 47 ratios between those of foodstuff and indoor contamination. In these cases, contamination with BDE 47 (and other non-BDE 209 congeners) and PCBs most likely originated in varied proportions from both foodstuff and from indoor air particles wafted into the baking oven, with minor contributions from the oven itself possible. This was in support with the fact, that all baking ovens were installed close to the ground (height of the bottom edge of the baking ovens ≤40 cm, Table S1, Supporting information), which might benefit the introduction of floor dust into the baking oven, when the oven door is opened. Contamination released from baking ovens with “moderate” levels could be due to recent cleaning procedures (see also below) or due to a more effective insulation of the oven chamber

Fig. 4. Relative PBDE pattern detected in the baking oven samples as well as in the technical PBDE mixtures DE-83 (“c-DecaBDE”), DE-79 (“c-OctaBDE”) and DE-71 (“c-PentaBDE”).

1026

C. Gallistl et al. / Science of the Total Environment 615 (2018) 1019–1027

accompanied with lower leakage of HFRs into the inside compartment compared to “high level” samples. 3.5. Further halogenated flame retardants (HFRs) No further targeted compound reached “high” levels (N2500 ng/g fat) which would have indicated the release of relevant amounts from the inside of the baking ovens (Fig. S7, Supporting information). However, both DBDPE and DP were present in ~30% of the samples. DBDPE was detected in seven baking oven samples at 9.4–190 ng/g (geometric mean: 3.2 ng/g), which was approximately one order of magnitude lower than the levels of BDE 209. DBDPE was introduced in the early 1990s as a major alternative to technical c-DecaBDE (DE-83) (Covaci et al., 2011; Kierkegaard et al., 2004). Recent data suggest that the annual global production of DBDPE (23,000–46,000 t in 2012, with an estimated increasing production rates of 80%/year in China, Chen et al., 2010; Covaci et al., 2011; Hong et al., 2014) will likely soon exceed the highest annual production volume of c-DecaBDE (~ 56,000 t/year between 1999 and 2003, Guerra et al., 2011). The lower mean levels of DBDPE compared to BDE 209 in the baking oven samples indicated that DBDPE has not yet reached its use and concentration climax in this type of indoor samples. Interestingly, DBDPE-positive samples were collected from baking ovens installed during the past five years. Due to exclusive DBDPE concentrations in the “moderate level” range the origin of this HFR in the baking ovens was most likely caused by external sources. DP levels in the six positive samples were in the same range (4.1– 800 ng/g fat) as those of DBDPE. Five of these six samples were taken from baking ovens ≤5 years old. This indicated that DP is also an emerging HFR in kitchens. The syn/anti-DP ratio (range: 0.21–0.51, mean/median: 0.33/0.29) was similar to those of commercial formulations (0.2– 0.41, depending on the manufacturer, Li et al., 2015). However, published syn/anti-DP ratios in indoor dust samples were often higher, most likely due to the stability of syn-DP when exposed to UV light (Sverko et al., 2008; Zhu et al., 2007). 3.6. Effects of baking ovens with pyrolysis program on HFR levels Three of the sampled baking ovens were equipped with a pyrolysis program which cleans the oven by the combustion of fat and other food residues at temperatures N400 °C (Table S12, Supporting information). On one occasion this program was run one day after the regular sample (Section 2.3). After pyrolysis (#8B) the sample weight was only 1/4th of the value before the combustion cleaning (#8A). The total HFR concentration before and after using the pyrolysis program was 64 and 48 ng/g fat, respectively. BDE 209 and DBDPE were only detected (at levels bLOQ) before the pyrolysis program was started (Table 1) and non-209 PBDEs only were detected after pyrolysis. Similarly, the ∑PCB level dropped by ~50% with a more pronounced decrease of volatile congeners, in particular PCB 118 and PCB 138. Although it seemed that the use of the pyrolysis function not only cleaned the baking oven from filth but also decreased the concentration of volatile compounds on the surface of the sampled baking oven door, it was not possible to give a general statement on efficiency of the pyrolysis program in terms of HFR decrease. For example, differences in the HFR concentrations before and after the pyrolysis program might also be explained by unequal distribution of the compounds on the sampled surface. 3.7. Correlations of HFR levels in the baking oven samples Different sources likely contributed to the variable presence of the examined CPs, PBDEs and PCBs, since no significant correlations between these HFRs were observed (Spearman correlation coefficients and levels of significance in Table S13, Supporting information). These findings were in agreement with the observation that CPs - if present - were exclusively in the “high level” range, indicating their origin in

the baking ovens themselves, which was also supported by the results from the wipe tests on components from one dismantled baking oven. Noteworthy, a highly significant positive correlation (Table S13, Supporting information) was observed between DBDPE and ΣDP concentrations, which indicated a similar increase in the use of which both are considered as PBDE alternatives (Feo et al., 2012; Li et al., 2015; Wang et al., 2012). 4. Conclusions Detection of HFRs in residues collected from the surface of all 21 baking ovens analyzed in this study confirmed their presence and release in every tested household. The absence of significant correlations between PBDEs and CPs (as well as PCBs) supports our suggestion that different sources exist for these classic HFRs. Aside from three “high level” concentrations, PBDE sources were variable and originated most likely from external sources. The occurrence of HFRs on the oven door and the likely contamination by HFRs from outside the oven makes it also very likely that HFRs used in the baking ovens can be released into the kitchen, especially while the door is open. This is of special concern for the “high level” HFRs and thus mainly SCCPs and MCCPs. Hence, release of CPs from the baking ovens may be a relevant source contributing to the high levels previously detected in kitchen hoods and dishcloths (Bendig et al., 2013b; Gallistl et al., 2017). Likewise, HFRs released from components of the baking oven housing into the oven chamber may enter foodstuff (at least if it is not covered with a lid or by aluminum foil) baked within the oven. Therefore, further studies on the transfer of HFRs from the inner baking oven surface into prepared foodstuff are necessary, since this process might represent a secondary contamination of the raw foodstuff. In either case, release from the baking oven will most likely contribute to the human exposure to MCCPs (and other HFRs), both by inhalation of contaminated air and by consumption of food contaminated during the baking process. Especially for MCCPs, this problem seems to be of particular concern, as these compounds were elevated in a significant portion of sampled ovens. Our results suggest that the use of MCCPs and other HFRs in baking ovens should be considered critically, and HFR exposure evaluations and appliance safety standards should be revised accordingly. For safety reasons, product requirements should include tests which confirm that HFRs are not released during baking processes. Such standards should also be established for other kitchen appliances and equipment. Acknowledgements The authors want to thank all participants for their permission to collect samples in their household kitchens. Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi. org/10.1016/j.scitotenv.2017.09.112. References Alaee, M., Arias, P., Sjödin, A., Bergman, Å., 2003. An overview of commercially used brominated flame retardants, their applications, their use patterns in different countries/ regions and possible modes of release. Environ. Int. 29 (6), 683–689. Beard, A., 2013. European initiatives for selecting sustainable flame retardants. In: Hester, R., Harrison, R. (Eds.), Issues in Environmental Science and Technology, 36: Chemical Alternatives Assessments. The Royal Society of Chemistry, Cambridge, pp. 44–66. Bendig, P., Blumenstein, M., Vetter, W., 2012. Heating of BDE-209 and BDE-47 in plant oil in presence of o,p’-DDT or iron(III) chloride can produce monochloro-polybromo diphenyl ethers. Food Chem. Toxicol. 50 (5), 1697–1703. Bendig, P., Hägele, F., Blumenstein, M., Schmidt, J., Vetter, W., 2013a. Fate of polybrominated diphenyl ethers during cooking of fish in a new model cooking apparatus and a household microwave. J. Agric. Food Chem. 61 (27), 6728–6733. Bendig, P., Hägele, F., Vetter, W., 2013b. Widespread occurrence of polyhalogenated compounds in fat from kitchen hoods. Anal. Bioanal. Chem. 405, 7485–7496. Birnbaum, L.S., Staskal, D.F., 2004. Brominated flame retardants: cause for concern? Environ. Health Perspect. 112 (1), 9–17.

C. Gallistl et al. / Science of the Total Environment 615 (2018) 1019–1027 Breivik, K., Sweetman, A., Pacyna, J.M., Jones, K.C., 2002. Towards a global historical emission inventory for selected PCB congeners - A mass balance approach: 1. Global production and consumption. Sci. Total Environ. 290 (1-3), 181–198. Cao, Y., Harada, K.H., Liu, W., Yan, J., Zhao, C., Niisoe, T., et al., 2015. Short-chain chlorinated paraffins in cooking oil and related products from China. Chemosphere 138, 104–111. Chen, S.J., Ma, Y.J., Wang, J., Tian, M., Luo, X.J., Chen, D., et al., 2010. Measurement and human exposure assessment of brominated flame retardants in household products from South China. J. Hazard. Mater. 176 (1-3), 979–984. Covaci, A., Harrad, S., Abdallah, M.A.-E., Ali, N., Law, R.J., Herzke, D., et al., 2011. Novel brominated flame retardants: A review of their analysis, environmental fate and behaviour. Environ. Int. 37 (2), 532–556. Darnerud, P.O., Atuma, S., Aune, M., Bjerselius, R., Glynn, A., Grawé, K.P., et al., 2006. Dietary intake estimations of organohalogen contaminants (dioxins, PCB, PBDE and chlorinated pesticides, e.g. DDT) based on Swedish market basket data. Food Chem. Toxicol. 44 (9), 1597–1606. de Boer, J., Ballesteros-Gómez, A., Leslie, H.A., Brandsma, S.H., Leonards, P.E.G., 2016. Flame retardants: Dust - and not food - might be the risk. Chemosphere 150, 461–464. Directive 2011/65/EU of the European Parliament and of the Council of 8 June 2011 on the restriction of the use of certain hazardous substances in electrical and electronic equipment: Directive 2011/65/EU; 2011. Dodson, R.E., Perovich, L.J., Covaci, A., van den Eede, N., Ionas, A.C., Dirtu, A.C., et al., 2012. After the PBDE phase-out: A broad suite of flame retardants in repeat house dust samples from California. Environ. Sci. Technol. 46 (24), 13056–13066. Eljarrat, E., Barceló, D. (Eds.), 2011. Brominated Flame Retardants: Degradation of brominated flame retardants. Springer, Berlin, Germany. European Parliament, 2003. Directive 2002/95/EC of the European Parliament and of the Council of 27 January 2003 on the restriction of the use of certain hazardous substances in electrical and electronic equipment: Directive 2002/95/EC. Fattore, E., Fanelli, R., Dellatte, E., Turrini, A., Domenico, A., 2008. Assessment of the dietary exposure to non-dioxin-like PCBs of the Italian general population. Chemosphere 73, 278–283. Feo, M.L., Barón, E., Eljarrat, E., Barceló, D., 2012. Dechlorane Plus and related compounds in aquatic and terrestrial biota: a review. Anal. Bioanal. Chem. 404 (9), 2625–2637. Gallen, C., Banks, A., Brandsma, S., Baduel, C., Thai, P., Eaglesham, G., et al., 2014. Towards development of a rapid and effective non-destructive testing strategy to identify brominated flame retardants in the plastics of consumer products. Sci. Total Environ. 491-492, 255–265. Gallistl, C., Lok, B., Schlienz, A., Vetter, W., 2017. Polyhalogenated compounds (chlorinated paraffins, novel and classic flame retardants, POPs) in dishcloths after their regular use in households. Sci. Total Environ. 595, 303–314. Gallistl, C., Vetter, W., 2016. Synthesis, liquid chromatographic fractionation and partial characterization of polybrominated dibenzofuran congeners. J. Chromatogr. A 1442, 62–72. Glüge, J., Wang, Z., Bogdal, C., Scheringer, M., Hungerbühler, K., 2016. Global production, use, and emission volumes of short-chain chlorinated paraffins – a minimum scenario. Sci. Total Environ. 573, 1132–1146. Guerra, P., Alaee, M., Eljarrat, E., Barceló, D., 2011. Introduction to brominated flame retardants: Commercially products, applications, and physicochemical properties. In: Eljarrat, E., Barceló, D. (Eds.), The Handbook of environmental chemistry. Vol. 16: Brominated Flame Retardants. Springer, Heidelberg, Dordrecht, London, New York, pp. 1–17. Hale, R.C., La Guardia, M.J., Harvey, E., Gaylor, M.O., Mainor, T.M., 2006. Brominated flame retardant concentrations and trends in abiotic media. Chemosphere 64 (2), 181–186. Hayward, D., Wong, J., Krynitsky, A.J., 2007. Polybrominated diphenyl ethers and polychlorinated biphenyls in commercially wild caught and farm-raised fish fillets in the United States. Environ. Res. 103 (1), 46–54. Hilger, B., Fromme, H., Völkel, W., Coelhan, M., 2013. Occurrence of chlorinated paraffins in house dust samples from Bavaria, Germany. Environ. Pollut. 175, 16–21. Hong, B., Wu, T., Zhao, G., Sun, Y., Wang, X., Zhao, J., et al., 2014. Occurrence of decabromodiphenyl ethane in captive Chinese alligators (Alligator sinensis) from China. Bull. Environ. Contam. Toxicol. Huang, H., Gao, L., Xia, D., Qiao, L., Wang, R., Su, G., et al., 2017. Characterization of shortand medium-chain chlorinated paraffins in outdoor/indoor PM10/PM2.5/PM1.0 in Beijing, China. Environ. Pollut. 225, 674–680. IISD Reporting Services, 2017. Summary of the meetings of the conferences of the parties to the Basel, Rotterdam and Stockholm Conventions: 24. April–5. May 2017. Earth Negot. Bull. 15 (252):1–33 (accessed August 04, 2017). http://enb.iisd.org/download/pdf/enb15252e.pdf. Kierkegaard, A., Björklund, J., Fridén, U., 2004. Identification of the flame retardant decabromodiphenyl ethane in the environment. Environ. Sci. Technol. 38 (12), 3247–3253. Kuang, J., Ma, Y., Harrad, S., 2016. Concentrations of "legacy" and novel brominated flame retardants in matched samples of UK kitchen and living room/bedroom dust. Chemosphere 149, 224–230. La Guardia, M.J., Hale, R.C., Harvey, E., 2006. Detailed polybrominated diphenyl ether (PBDE) congener composition of the widely used penta-, octa-, and deca-PBDE technical flame-retardant mixtures. Environ. Sci. Technol. 40 (20), 6247–6254. Li, Y., Li, J., Wang, L., 2013. Recycling of PBDEs containing plastics from waste electrical and electronic equipment (WEEE): A review. Proceedings - 2013 IEEE 10th

1027

International Conference on e-Business Engineering. KTH Royal Institute of Technology, Stockholm, Sweden: IEEE Computer Society. Li, W.-L., Qi, H., Ma, W.-L., Liu, L.-Y., Zhang, Z., Zhu, N.-Z., et al., 2015. Occurrence, behavior and human health risk assessment of dechlorane plus and related compounds in indoor dust of China. Chemosphere 134, 166–171. Reth, M., Oehme, M., 2004. Limitations of low resolution mass spectrometry in the electron capture negative ionization mode for the analysis of short- and medium-chain chlorinated paraffins. Anal. Bioanal. Chem. 378 (7), 1741–1747. Ryan, J.J., Patry, B., 2001. Body burdens and food exposure in Canada for polybrominated diphenyl ethers (BDPEs): 21st International Symposium on Halogenated Persistent Organic Pollutants (Dioxin 2001, Korea). Organohalogen Compd. (51), 226–229. Samsonek, J., Puype, F., 2013. Occurrence of brominated flame retardants in black thermo cups and selected kitchen utensils purchased on the European market. Food Addit. Contam. Part A Chem. Anal. Control Expo. Risk Assess. 30 (11), 1976–1986. Schecter, A., Shah, N., Colacino, J.A., Brummitt, S.I., Ramakrishnan, V., Harris, T.R., et al., 2009. PBDEs in US and German clothes dryer lint: A potential source of indoor contamination and exposure. Chemosphere 75 (5), 623–628. Schinkel, L., Lehner, S., Heeb, N.V., Lienemann, P., McNeill, K., Bogdal, C., 2017. Deconvolution of Mass Spectral Interferences of Chlorinated Alkanes and Their Thermal Degradation Products: Chlorinated Alkenes. Anal. Chem. 89 (11), 5923–5931. Shaw, S.D., Blum, A., Weber, R., Kannan, K., Rich, D., Lucas, D., et al., 2010. Halogenated flame retardants: do the fire safety benefits justify the risks? Rev. Environ. Health 25 (4), 261–305. Shi, L., Gao, Y., Zhang, H., Geng, N., Xu, J., Zhan, F., et al., 2017. Concentrations of short- and medium-chain chlorinated paraffins in indoor dusts from malls in China: implications for human exposure. Chemosphere 172, 103–110. Sindiku, O., Babayemi, J., Osibanjo, O., Schlummer, M., Schluep, M., Watson, A., et al., 2015. Polybrominated diphenyl ethers listed as Stockholm Convention POPs, other brominated flame retardants and heavy metals in e-waste polymers in Nigeria. Environ. Sci. Pollut. Res. Int. 22 (19), 14489–14501. Survey of the Association of Social Research and Statistical Analysis (Forsa), 2010. Furnishing of German household kitchens with kitchen appliences in 2011 (3,011 participants in the age 30-70 years). access via Statista. https://de.statista.com/. Sverko, E., Tomy, G.T., Marvin, C.H., Zaruk, D., Reiner, E., Helm, P.A., et al., 2008. Dechlorane plus levels in sediment of the lower great lakes. Environ. Sci. Technol. 42 (2), 361–366. Takaoka, M., Fukatsu, A., Takeda, N., 2000. Catalytic decomposition of polychlorinated biphenyls (PCBs). Toxicol. Environ. Chem. 76 (1-2), 95–109. Takasuga, T., Nakano, T., Shibata, Y., 2012. Unintentional POPs (PCBs, PCBz, PCNs) contamination in articles containing chlorinated paraffins and related impacted chlorinated paraffin products. Organohalogen Compd. 74, 1437–1440. Vetter, W., Kirres, J., Bendig, P., 2011. Bromination of 2-methoxydiphenyl ether to an average of tetrabrominated 2-methoxydiphenyl ethers. Chemosphere 84 (8), 1117–1124. Vetter, W., Luckas, B., 1995. Synthesis, isolation, and chromatography of perdeuterated α1,2,3,4,5,6-hexachlorocyclohexane. J. High Resolut. Chromatogr. 18 (10), 643–646. Vetter, W., Weichbrodt, M., Hummert, K., Glotz, D., Luckas, B., 1998. Combined microwave-assisted extraction and gel permeation chromatography for the determination of chlorinated hydrocarbons in seal blubber and cod livers. Chemosphere 37 (9-12), 2439–2449. von der Recke, R., Vetter, W., 2007. Synthesis and characterization of 2,3-dibromopropyl2,4,6-tribromophenyl ether (DPTE) and structurally related compounds evidenced in seal blubber and brain. Environ. Sci. Technol. 41 (5), 1590–1595. Wang, Y.-Z., 2008. Halogen-free flame retardants. In: Horrocks, A., Price, D. (Eds.), Advances in Fire Retardant Materials. Woodhead Publishing In Materials, CRC Press, Boca Raton, Boston, New York, Washington DC. Wang, J., Chen, S., Nie, X., Tian, M., Luo, X., An, T., et al., 2012. Photolytic degradation of decabromodiphenyl ethane (DBDPE). Chemosphere 89 (7), 844–849. Watkins, D.J., McClean, M.D., Fraser, A.J., Weinberg, J., Stapleton, H.M., Sjödin, A., et al., 2011. Exposure to PBDEs in the office environment: Evaluating the relationships between dust, handwipes, and serum. Environ. Health Perspect. 119 (9), 1247–1252. Website of the Stockholm Convention on Persistent Organic Pollutants, www.chm.pops.int, Accessed date: 4 August 2017. Wong, F., Suzuki, G., Michinaka, C., Yuan, B., Takigami, H., de Wit, C.A., 2017. Dioxin-like activities, halogenated flame retardants, organophosphate esters and chlorinated paraffins in dust from Australia, the United Kingdom, Canada, Sweden and China. Chemosphere 168, 1248–1256. Yuan, B., Nyström, J., de Wit, C.A., Bergman, Å., Strid, A., 2016. Leakage of chlorinated paraffins from hand blenders sold on the Swedish market: 36th International Symposium on Halogenated Persistent Organic Pollutants (Dioxin 2016, Firenze). Organohalogen Compd. (78), 456–460. Zeng, L., Chen, R., Zhao, Z., Wang, T., Gao, Y., Li, A., et al., 2013. Spatial distributions and deposition chronology of short chain chlorinated paraffins in marine sediments across the Chinese Bohai and Yellow Seas. Environ. Sci. Technol. 47 (20), 11449–11456. Zhu, J., Feng, Y.L., Shoeib, M., 2007. Detection of dechlorane plus in residential indoor dust in the City of Ottawa, Canada. Environ. Sci. Technol. 41 (22), 7694–7698.