Chemosphere 151 (2016) 1e8
Contents lists available at ScienceDirect
Chemosphere journal homepage: www.elsevier.com/locate/chemosphere
Occurrence of bisphenols, bisphenol A diglycidyl ethers (BADGEs), and novolac glycidyl ethers (NOGEs) in indoor air from Albany, New York, USA, and its implications for inhalation exposure Jingchuan Xue a, Yanjian Wan b, Kurunthachalam Kannan a, c, * a
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, United States Center for Disease Control and Prevention of Yangtze River Administration and Navigational Affairs, General Hospital of the Yangtze River Shipping, Wuhan 430019, China c Biochemistry Department, Faculty of Science and Experimental Biochemistry Unit, King Fahd Medical Research Center, King Abdulaziz University, Jeddah, Saudi Arabia b
h i g h l i g h t s Eight bisphenols and 11 BADGE derivatives were determined in 83 indoor air samples. BPA, BPF, BPS and BADGE$2H2O were found at mean GM concentrations of 0.43, 0.69, 0.09 and 0.28 ng m3, respectively. Highest BADGE$2H20 and bisphenols concentrations were found in auto repair shops and cars, respectively. GM inhalation intake of bisphenols in adults was 0.37 ng kg bw1 day1 and that for BADGEs was 0.26 ng kg bw1$day1.
a r t i c l e i n f o
a b s t r a c t
Article history: Received 24 December 2015 Received in revised form 6 February 2016 Accepted 8 February 2016 Available online xxx
Bisphenols, bisphenol A diglycidyl ethers (BADGEs), and novolac glycidyl ethers (NOGEs) are used in the production of epoxy resins and polycarbonate plastics. Despite the widespread application of these chemicals in household products, studies on their occurrence in indoor air are limited. In this study, 83 indoor air samples were collected in 2014 from various locations in Albany, New York, USA, to determine the concentrations of bisphenols, BADGEs (refer to BADGE and its derivatives), and NOGEs (refer to NOGE and its derivatives) and to calculate inhalation exposure to these compounds. Among eight bisphenols measured, BPA, BPF, and BPS were found in bulk air (i.e., vapor plus particulate phases), at geometric mean (GM) concentrations of 0.43, 0.69 and 0.09 ng m3, respectively. Among 11 BADGEs and NOGEs determined, BADGE$2H2O was the predominant compound found in indoor air (detection rate [DR]: 85.5%), at concentrations as high as 6.71 ng m3. Estimation of inhalation exposure to these chemicals for various age groups showed that teenagers had the highest exposure doses to BPA, BPF, BPS, and BADGE$2H2O at 5.91, 9.48, 1.24, and 3.84 ng day1, respectively. The body weight-normalized estimates of exposure were the highest for infants, with values at 0.24, 0.39, 0.05, and 0.16 ng kg bw1 day1 for BPA, BPF, BPS, and BADGE$2H2O, respectively. This is the first survey to report inhalation exposure to bisphenols, BADGEs, and NOGEs. © 2016 Elsevier Ltd. All rights reserved.
Handling Editor: R Ebinghaus Keywords: Bisphenol A Bisphenol F Bisphenol S BADGE Indoor air Inhalation
1. Introduction Bisphenol analogs (hereafter ‘bisphenols’) are a group of
* Corresponding author. Wadsworth Center Empire State Plaza, P.O. Box 509 Albany, NY 12201-0509, United States. E-mail address: [email protected] (K. Kannan). http://dx.doi.org/10.1016/j.chemosphere.2016.02.038 0045-6535/© 2016 Elsevier Ltd. All rights reserved.
chemicals that share the basic structure of two phenol groups, with hydroxy moieties at the para positions, and joined by a carbon or sulfur bridge. 2,2-bis(4-hydroxyphenyl)propane, commonly known as bisphenol A (BPA), is the major compound in this chemical class. Being an intermediate in the production of epoxy resins and polycarbonate plastics, BPA is used in a broad range of applications, including coatings for beverage and food cans (Allan and Eddo, 2011; Xie et al., 2014), polycarbonate baby-feeding bottles (Allan
2
J. Xue et al. / Chemosphere 151 (2016) 1e8
and Eddo, 2011), and thermal receipt papers (Liao and Kannan, 2011). Produced in quantities of more than 4 million metric tons per year globally, bisphenols are categorized as “high production volume (HPV) chemicals” by the U.S. Environmental Protection Agency (EPA) (Bailin et al., 2008). As one of the well-studied endocrine-disrupting chemicals, BPA has been associated with a variety of human health effects, including obesity, diabetes, breast and prostate cancers, and neurological and reproductive problems (Huang et al., 2012; Kang et al., 2006). As the concern over the safety of BPA grows, the industry has been developing substitutes, such as BPF and BPS, for some applications (Liao et al., 2012a). These substitutes, however, have chemical structures similar to BPA and are expected to possess similar toxic effects. Further, estrogenic effects of BPF and BPS have been reported in certain studies (Chen et al., 2002; Kitamura et al., 2005). Similar to bisphenols, bisphenol A diglycidyl ethers (BADGEs) and novolac glycidyl ethers (NOGEs) are “HPV chemicals,” and the global BADGE production in 2003 was 957,000 metric tons (Nakazawa et al., 2002; Satoh et al., 2004; DCC, 2006; Terasaki et al., 2006). BADGE- and NOGE-based epoxy resins are the major types of polymers used in the inner coatings of food and beverage cans (Berger and Oehme, 2000). BADGEs and NOGEs also are used in the construction, automotive, electrical, and electronics industries (Lee and Neville, 1982). Apart from their endocrine-disrupting potential, BADGEs and NOGEs have been reported to elicit adverse reproductive and developmental effects in in vivo bioassays (Hyoung et al., 2007; Kang et al., 2006). Although the carcinogenic effects of BADGEs and NOGEs are still unknown, a variety of in vitro bioassays have confirmed their genotoxic potentials (Cabaton et al., 2009; Suarez et al., 2000; Sueiro et al., 2006). BADGE and NOGE are reactive molecules, and several derivatives of them including BADGE$H2O, BADGE$HCl, BADGE$2H2O, BADGE$2HCl, BADGE$H2O$HCl, BFDGE$2HCl, and BFDGE$2H2O have been reported to occur in the environment. Studies have reported the occurrence of bisphenols, BADGEs, and NOGEs in human specimens and tissues, such as urine (Asimakopoulos et al., 2014; Xue et al., 2015b), plasma (Wang et al., 2015a), and adipose fat (Wang et al., 2015a). Reports on the occurrence of BADGEs and NOGEs in environmental matrices are limited to indoor dust (Wang et al., 2012) and sewage sludge (Xue et al., 2015a). Bisphenols have been reported to occur in various environmental matrices, including surface water (Yamazaki et al., mez et al., 2007), sewage 2015), wastewater (Ballesteros-Go sludge (Lee et al., 2015; Song et al., 2014), sediments (Liao et al., 2012b), and indoor dust (Wang et al., 2015b). Studies on the occurrence of these chemicals in indoor air, however, are limited. Among the chemicals noted above, BPA is the only compound for which concentrations in indoor air have been reported, and those studies monitored ambient air or indoor air in certain places, such as daycare centers (Fu and Kawamura, 2010; Inoue et al., 2006; Wilson et al., 2007). This study was designed to evaluate the inhalation exposure to bisphenols, BADGEs, and NOGEs and to investigate the distribution of these substances in particulate and vapor phases of indoor air in various microenvironments in Albany, New York, USA.
hexylidenebisphenol (BPZ, ~98%), BADGE (97%), bisphenol A (2, 3dihydroxypropyl) glycidyl ether (BADGE$H2O, 97%), bisphenol A (3-chloro-2-hydroxypropyl) glycidyl ether (BADGE$HCl, ~95%), bisphenol A bis(2,3-dihydroxypropyl) glycidyl ether (BADGE$2H2O, 97%), bisphenol A bis(3-chloro-2-hydroxypropyl) glycidyl ether (BADGE$2HCl, 99%), bisphenol A (3-chloro-2-hydroxypropyl) (2,3-dihydroxypropyl) glycidyl ether (BADGE$H2O$HCl, 98%), bisphenol F diglycidyl ether (BFDGE, ~97%), bisphenol F bis(3chloro-2-hydroxypropyl) glycidyl ether (BFDGE$2HCl, ~95%), bisphenol F bis(2,3-dihydroxypropyl) glycidyl ether (BFDGE$2H2O, 97%), novolac glycidyl ether 3-ring (3R-NOGE, 90%), and novolac glycidyl ether 4-ring (4R-NOGE, 90%) were purchased from SigmaeAldrich (St. Louis, MO, USA). Analytical standard of 2, 20 -bis(4hydroxyphenyl)butane (BPB, ~98%) was obtained from TCI America (Portland, OR, USA). 13C-isotopically labeled 13C12-BPA (99%) was purchased from Cambridge Isotope Laboratories (Andover, MA, USA). D6-BADGE was obtained from Toronto Research Chemicals Inc (Toronto, Ontario, Canada). The stock solutions of target analytes and internal standards were prepared at 1 mg/mL in methanol and stored at 20 C. Methanol (HPLC grade), ethyl acetate (ACS grade), and other solvents used in the experiments were purchased from Mallinckrodt Baker (Phillipsburg, NJ, USA). Milli-Q water was purified by an ultrapure water system (Barnstead International, Dubuque, IA, USA). 2.2. Sample collection
2. Materials and methods
The method for the collection of indoor air has been described elsewhere (Wan et al., 2015). Briefly, two polyurethane foam (PUF) plugs (ORBO-1000 PUF dimensions: 2.2 cm O.D 7.6 cm length) obtained from Supelco (Bellefonte, PA, USA) and a quartz fiber filter (Whatman, grade QM-A, pore size: 2.2 mm, 32 mm diameter) were assembled in tandem to collect vapor and particulate phases in indoor air, respectively. To determine the background levels of bisphenols, BADGEs, and NOGEs, newly purchased PUF plugs were extracted twice with 100 mL ethyl acetate and analyzed. The results showed that each of the newly purchased PUF plugs contained 0.25, 0.11, and 0.15 ng BPA, BPF, and BADGE$2H2O, respectively (n ¼ 3). Therefore, all PUF plugs were washed twice by shaking with 100 mL of ethyl acetate for 30 min and then stored in a glass jar that was kept in an oven at 100 C until further use. The quartz fiber filters were heated at 450 C for 20 h and held in an oven at 100 C until further use. The quartz fiber filters were weighed before and after the collection of air samples to determine the mass of particle content in air. All glassware used for sampling and analysis was heated at 450 C for 10 h prior to use. Indoor air samples were collected for 3e24 h by a low-volume air sampler (LP-20; A.P. Buck Inc., Orlando, FL, USA) at a flow rate of 5 L/min. The samples were collected from September to December 2014 at several locations in Albany, New York, USA. The sampling locations were grouped into eight categories: parking garages (n ¼ 3), auto repair shops (n ¼ 4), cars (n ¼ 7), barber shops (n ¼ 5), public places (n ¼ 13), homes (n ¼ 26), laboratories (n ¼ 12), and offices (n ¼ 13). In homes, indoor air samples were collected during day and night time, separately, to compare if there were any diurnal differences in the concentrations of target analytes.
2.1. Standards and reagents
2.3. Sample preparation
Analytical standards of BPA (97%), 4, 40 -(hexafluoroisopropylidene)-diphenol (BPAF, ~97%), 4, 40 -(10 phenylethylidene)bisphenol (BPAP, ~99%), 4, 4 -sulfonyldiphenol (BPS, ~98%), 4, 40 -dihydroxydiphenylmethane (BPF, ~98%), 4, 40 (1,4-phenylenediisopropylidene)bisphenol (BPP, ~99%), 4, 40 -cyclo-
The PUF plugs were extracted by shaking in an orbital shaker (Eberbach Corporation, Ann Arbor, MI, USA) with ethyl acetate for 30 min. The extraction was performed twice with 100 mL solvent each time. The particulate samples were extracted by shaking glass fiber filters with ethyl acetate (20 mL) three times, each time for
J. Xue et al. / Chemosphere 151 (2016) 1e8
5 min. For both PUF plugs and filters, the extracts were combined and concentrated in a rotary evaporator at 55 C to 5 mL. Then, the solution was transferred to a 12-mL glass tube, concentrated under a gentle stream of nitrogen to near dryness, reconstituted with 1 mL of methanol, and transferred into a vial for high performance liquid chromatography-tandem mass spectrometry (HPLC-MS/MS) analysis. Twenty nanograms of 13C12-BPA and d6-BADGE were spiked (as internal standards) into all of the samples and blanks before extraction. 2.4. LC-ESI()MS/MS analysis of bisphenols The chromatographic separation was carried out using a Shimadzu Prominence Modular HPLC system (Shimadzu Corporation, Kyoto, Japan), consisting of a system controller, binary pump, and auto sampler. Identification and quantification of target analytes were performed with an Applied Biosystems API 3200 electrospray triple quadrupole-mass spectrometer (ESI-MS/MS) (Applied Biosystems, Foster City, CA, USA). A Betasil C18 column (2.1 mm 100 mm, 5 mm) (Thermo Electron Corporation Waltham, MA, USA) serially connected to a Javelin guard column (Betasil C18, 2.1 mm 20 mm, 5 mm) (Thermo Electron Corporation) was used. The injection volume was 10 mL, and the mobile phase comprised methanol (A) and Milli-Q water that contained 1% (v/v) ammonium hydroxide (B). The target compounds were separated by gradient elution of mobile phase at a flow rate of 300 mL/min starting at 15% (v/v) A, held for 2 min; increased to 75% A within 3 min, held for 2 min; then further increased to 99% A within 3 min, held for 4 min; and reverted to 15% A at the 14.5th min and held for 5.5 min, for a total run time of 20 min. The MS/MS system was operated in multiple reaction monitoring (MRM) negative ion mode. The compound specific MS/MS parameters are shown in Table S1. Nitrogen was used as both a curtain and a collision gas. The electrospray ionization voltage was set at 4.5 kV. The curtain and collision gas flow rates were set at 25 and 2 psi, respectively, and the source heater was set at 650 C. The nebulizer gas (ion source gas 1) was set at 20 psi, and the heater gas (ion source gas 2) was set at 70 psi. The data acquisition was set at 80 ms for scan speed and 0.70 full width at half maximum (FWHM) for resolving power. 2.5. LC-ESI(þ)MS/MS analysis of BADGEs and NOGEs For the analysis of BADGE, BFDGE, 3R-NOGE, 4R-NOGE, and their derivatives, the chromatographic separation was carried out using an Agilent 1100 Series HPLC system (Agilent Technologies Inc., Santa Clara, CA, USA). Identification and quantification of target analytes were performed with an Applied Biosystems API 2000 ESIMS/MS. The chromatographic columns were similar to that reported above for the analysis of bisphenols. The injection volume was 10 mL, and the mobile phase comprised methanol (A) and MilliQ water/methanol (90:10, % v/v) that contained 2 mM ammonium acetate (B). The target compounds were separated by gradient elution of mobile phase at a flow rate of 300 mL/min starting at 20% (v/v) A, held for 2 min; increased to 75% A within 1 min; further increased to 95% A within 2 min, held for 10 min; and reverted to 20% A at the 16th min and held for 4 min, for a total run time of 20 min. The MS/MS system was operated in MRM positive ion mode. The compound specific MS/MS parameters for BADGEs and select NOGEs are shown in Table S2. Nitrogen was used as both a curtain and a collision gas. The electrospray ionization voltage was set at þ4.5 kV. The curtain and collision gas flow rates were set at 10 psi, and the source heater was set at 450 C. The nebulizer gas (ion source gas 1) was set at 45 psi, and the heater gas (ion source gas 2) was set at 75 psi. The data acquisition was performed at a scan speed of 40 ms and a resolving power of 0.70 FWHM.
3
2.6. Quality assurance and quality control Quantification of bisphenols, BADGEs, and NOGEs was performed by the isotope dilution method, with responses relative to 13 C12-BPA and d6-BADGE, respectively. A 9- to 11-point standard calibration curve, with concentrations ranging from 0.05 to 200 ng/ mL, was used for the quantification of target analytes. The calibration curves were prepared by plotting a concentration-response factor for each target analyte (peak area of analyte divided by peak area of the internal standard) versus the response-dependent concentration factor (concentration of analyte divided by concentration of internal standard). The regression coefficients (r) were 0.99 for all calibration curves. The limits of quantification (LOQs) were determined based on the lowest point of the calibration standard with a signal-to-noise ratio of >10. Method LOQs (MLOQs) were determined in the same manner, with the post-matrix spiked 9- to 11-point calibration curves. MLOQs in the bulk air equaled to the higher MLOQs obtained in either the vapor or the particulate phase. Chemical concentrations in the bulk air were calculated by dividing the sum of the mass of the chemical in the vapor and particulate phases by the volume of air collected. As a check for instrumentation drift in response factors, a midpoint calibration standard was injected after every 10 samples. A pure solvent (methanol) was injected after every 10 samples as a check for carryover of target analytes. Several procedural blanks were analyzed with each batch of samples to determine the contamination arising from laboratory materials and solvents. Efforts were taken to minimize background levels of target analytes. Throughout the analysis, six sets of PUF and glass fiber filter were selected for pre-extraction matrix spike (MS) (two for every 25 samples) by spiking 40 ng of target analytes and passing them through the entire analytical procedure. 2.7. Data analysis Data were acquired with Analyst software version 1.4.1 (Applied Biosystems, Foster City, CA, USA). Statistical analyses were performed with statistics software package R v.3.1.0 and Microsoft Excel 2007. For the calculation of arithmetic mean, median and GM, concentrations below the MLOQ were substituted with a value equal to half the MLOQ. To examine the relationship between chemicals, Spearman (when data did not follow a normal distribution after logarithmic transformation) or Pearson (when data followed a normal distribution after logarithmic transformation) correlation analysis was used. Only those samples with detectable concentrations of target analytes were used when performing correlation analysis. To assess the difference between means, a Student's t-test (when data followed a normal distribution after logarithmic transformation) or ManneWhitney U test (when data did not follow a normal distribution after logarithmic transformation) was used. A ShapiroeWilk test and quantileequantile (QeQ) plot were used to determine the normality of the data. Statistical significance was set at p < 0.05. Daily exposure doses were calculated using Eq. (1).
EDIinh ¼ C AIR IEF=BW
(1)
where C is the concentration of analytes in (bulk) indoor air (ng$m3), AIR is the air inhalation rate (m3$day1), IEF is the indoor exposure factor (represents the fraction of time spent indoors per day), and BW is the body weight (kg). 2.8. Method performance MLOQs of the target analytes ranged from 0.05 to 1.29 ng m3 in
4
J. Xue et al. / Chemosphere 151 (2016) 1e8
the vapor phase and from 0.71 to 35.7 mg g1 in the particulate phase (Table S3). Trace levels of BPA (0.57 ng), BPF (0.88 ng), and BADGE$2H2O (0.2 ng) were found in procedural blanks that contained two PUFs (Table S4). In procedural blanks that contained a quartz filter, BPF and BADGE$2H2O were found at average concentrations of 0.82 and 0.26 ng, respectively. The concentrations found in procedural blanks were subtracted from the concentrations measured in samples. Absolute and relative recoveries were used in this study. Calculation of absolute recovery was based on the instrumental response of analyte. The ratio of the signal for the analyte to that of the internal standard was used for the computation of relative recovery. In the vapor phase, relative recoveries of bisphenols ranged from 86.6% to 103%, and those of BADGEs and NOGEs ranged from 87.5% to 113%. In the particulate phase, relative recoveries of bisphenols, BADGEs, and NOGEs ranged from 98.5% to 132%. 3. Results and discussion 3.1. Concentrations of bisphenols in particulate phase, vapor phase, and bulk air Among the eight bisphenols analyzed, BPA and BPS were the most frequently detected compounds in the vapor phase, with detection rates (DRs) of 48.2% and 26.5%, respectively (Table 1). BPA concentrations in the vapor phase ranged from below MLOQ to 1.85 ng m3 (GM: 0.15 ng m3; median: 0.07 ng m3), and BPS concentrations ranged from below MLOQ to 0.94 ng m3 (GM: 0.04 ng m3; median: 0.03 ng m3) (Table 1). The highest concentration of BPA was found in automobile repair shops (GM: 0.4 ng m3; median: 0.39 ng m3) (Table 1) and BPS in barbershops (GM: 0.08 ng m3; median: 0.03 ng m3) (Table 1). In the particulate phase, BPA and BPF were the predominant compounds found in 75.9% and 63.9% of the samples, respectively (Table 1). The concentrations of BPA in the particulate phase ranged from below MLOQ to 153 mg g1 (GM: 7.19 mg g1; median: 6.75 mg g1) (Table 1). BPF concentrations in the particulate phase ranged from below MLOQ to 2220 mg g1 (GM: 18.1 mg g1; median: 13.9 mg g1) (Table 1). The highest concentration of BPA in the particulate phase was found in barbershops (GM: 11.2 mg g1; median: 9.07 mg g1) (Table 1), and the highest BPF concentration was in cars (GM: 33.8 mg g1; median: 58.8 mg g1) (Table 1). BPA is increasingly replaced with BPF in epoxy resin production, which is the predominant type of epoxy resins used in the automotive industry, and high concentrations of BPF in cars can be explained by this usage. BPS, BPAF, and BPB were less frequently detected in the particulate phase, with DRs ranging from 1.2% to 4.82% (Tables 1 and S5). The GM concentrations of BPA and BPF in airborne particles measured in this study were one to three orders of magnitude higher than those reported in indoor dust collected from homes in Albany, New York, USA (Wang et al., 2015b). The differences in particle sizes between airborne particles (
Contents lists available at ScienceDirect
Chemosphere journal homepage: www.elsevier.com/locate/chemosphere
Occurrence of bisphenols, bisphenol A diglycidyl ethers (BADGEs), and novolac glycidyl ethers (NOGEs) in indoor air from Albany, New York, USA, and its implications for inhalation exposure Jingchuan Xue a, Yanjian Wan b, Kurunthachalam Kannan a, c, * a
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, United States Center for Disease Control and Prevention of Yangtze River Administration and Navigational Affairs, General Hospital of the Yangtze River Shipping, Wuhan 430019, China c Biochemistry Department, Faculty of Science and Experimental Biochemistry Unit, King Fahd Medical Research Center, King Abdulaziz University, Jeddah, Saudi Arabia b
h i g h l i g h t s Eight bisphenols and 11 BADGE derivatives were determined in 83 indoor air samples. BPA, BPF, BPS and BADGE$2H2O were found at mean GM concentrations of 0.43, 0.69, 0.09 and 0.28 ng m3, respectively. Highest BADGE$2H20 and bisphenols concentrations were found in auto repair shops and cars, respectively. GM inhalation intake of bisphenols in adults was 0.37 ng kg bw1 day1 and that for BADGEs was 0.26 ng kg bw1$day1.
a r t i c l e i n f o
a b s t r a c t
Article history: Received 24 December 2015 Received in revised form 6 February 2016 Accepted 8 February 2016 Available online xxx
Bisphenols, bisphenol A diglycidyl ethers (BADGEs), and novolac glycidyl ethers (NOGEs) are used in the production of epoxy resins and polycarbonate plastics. Despite the widespread application of these chemicals in household products, studies on their occurrence in indoor air are limited. In this study, 83 indoor air samples were collected in 2014 from various locations in Albany, New York, USA, to determine the concentrations of bisphenols, BADGEs (refer to BADGE and its derivatives), and NOGEs (refer to NOGE and its derivatives) and to calculate inhalation exposure to these compounds. Among eight bisphenols measured, BPA, BPF, and BPS were found in bulk air (i.e., vapor plus particulate phases), at geometric mean (GM) concentrations of 0.43, 0.69 and 0.09 ng m3, respectively. Among 11 BADGEs and NOGEs determined, BADGE$2H2O was the predominant compound found in indoor air (detection rate [DR]: 85.5%), at concentrations as high as 6.71 ng m3. Estimation of inhalation exposure to these chemicals for various age groups showed that teenagers had the highest exposure doses to BPA, BPF, BPS, and BADGE$2H2O at 5.91, 9.48, 1.24, and 3.84 ng day1, respectively. The body weight-normalized estimates of exposure were the highest for infants, with values at 0.24, 0.39, 0.05, and 0.16 ng kg bw1 day1 for BPA, BPF, BPS, and BADGE$2H2O, respectively. This is the first survey to report inhalation exposure to bisphenols, BADGEs, and NOGEs. © 2016 Elsevier Ltd. All rights reserved.
Handling Editor: R Ebinghaus Keywords: Bisphenol A Bisphenol F Bisphenol S BADGE Indoor air Inhalation
1. Introduction Bisphenol analogs (hereafter ‘bisphenols’) are a group of
* Corresponding author. Wadsworth Center Empire State Plaza, P.O. Box 509 Albany, NY 12201-0509, United States. E-mail address: [email protected] (K. Kannan). http://dx.doi.org/10.1016/j.chemosphere.2016.02.038 0045-6535/© 2016 Elsevier Ltd. All rights reserved.
chemicals that share the basic structure of two phenol groups, with hydroxy moieties at the para positions, and joined by a carbon or sulfur bridge. 2,2-bis(4-hydroxyphenyl)propane, commonly known as bisphenol A (BPA), is the major compound in this chemical class. Being an intermediate in the production of epoxy resins and polycarbonate plastics, BPA is used in a broad range of applications, including coatings for beverage and food cans (Allan and Eddo, 2011; Xie et al., 2014), polycarbonate baby-feeding bottles (Allan
2
J. Xue et al. / Chemosphere 151 (2016) 1e8
and Eddo, 2011), and thermal receipt papers (Liao and Kannan, 2011). Produced in quantities of more than 4 million metric tons per year globally, bisphenols are categorized as “high production volume (HPV) chemicals” by the U.S. Environmental Protection Agency (EPA) (Bailin et al., 2008). As one of the well-studied endocrine-disrupting chemicals, BPA has been associated with a variety of human health effects, including obesity, diabetes, breast and prostate cancers, and neurological and reproductive problems (Huang et al., 2012; Kang et al., 2006). As the concern over the safety of BPA grows, the industry has been developing substitutes, such as BPF and BPS, for some applications (Liao et al., 2012a). These substitutes, however, have chemical structures similar to BPA and are expected to possess similar toxic effects. Further, estrogenic effects of BPF and BPS have been reported in certain studies (Chen et al., 2002; Kitamura et al., 2005). Similar to bisphenols, bisphenol A diglycidyl ethers (BADGEs) and novolac glycidyl ethers (NOGEs) are “HPV chemicals,” and the global BADGE production in 2003 was 957,000 metric tons (Nakazawa et al., 2002; Satoh et al., 2004; DCC, 2006; Terasaki et al., 2006). BADGE- and NOGE-based epoxy resins are the major types of polymers used in the inner coatings of food and beverage cans (Berger and Oehme, 2000). BADGEs and NOGEs also are used in the construction, automotive, electrical, and electronics industries (Lee and Neville, 1982). Apart from their endocrine-disrupting potential, BADGEs and NOGEs have been reported to elicit adverse reproductive and developmental effects in in vivo bioassays (Hyoung et al., 2007; Kang et al., 2006). Although the carcinogenic effects of BADGEs and NOGEs are still unknown, a variety of in vitro bioassays have confirmed their genotoxic potentials (Cabaton et al., 2009; Suarez et al., 2000; Sueiro et al., 2006). BADGE and NOGE are reactive molecules, and several derivatives of them including BADGE$H2O, BADGE$HCl, BADGE$2H2O, BADGE$2HCl, BADGE$H2O$HCl, BFDGE$2HCl, and BFDGE$2H2O have been reported to occur in the environment. Studies have reported the occurrence of bisphenols, BADGEs, and NOGEs in human specimens and tissues, such as urine (Asimakopoulos et al., 2014; Xue et al., 2015b), plasma (Wang et al., 2015a), and adipose fat (Wang et al., 2015a). Reports on the occurrence of BADGEs and NOGEs in environmental matrices are limited to indoor dust (Wang et al., 2012) and sewage sludge (Xue et al., 2015a). Bisphenols have been reported to occur in various environmental matrices, including surface water (Yamazaki et al., mez et al., 2007), sewage 2015), wastewater (Ballesteros-Go sludge (Lee et al., 2015; Song et al., 2014), sediments (Liao et al., 2012b), and indoor dust (Wang et al., 2015b). Studies on the occurrence of these chemicals in indoor air, however, are limited. Among the chemicals noted above, BPA is the only compound for which concentrations in indoor air have been reported, and those studies monitored ambient air or indoor air in certain places, such as daycare centers (Fu and Kawamura, 2010; Inoue et al., 2006; Wilson et al., 2007). This study was designed to evaluate the inhalation exposure to bisphenols, BADGEs, and NOGEs and to investigate the distribution of these substances in particulate and vapor phases of indoor air in various microenvironments in Albany, New York, USA.
hexylidenebisphenol (BPZ, ~98%), BADGE (97%), bisphenol A (2, 3dihydroxypropyl) glycidyl ether (BADGE$H2O, 97%), bisphenol A (3-chloro-2-hydroxypropyl) glycidyl ether (BADGE$HCl, ~95%), bisphenol A bis(2,3-dihydroxypropyl) glycidyl ether (BADGE$2H2O, 97%), bisphenol A bis(3-chloro-2-hydroxypropyl) glycidyl ether (BADGE$2HCl, 99%), bisphenol A (3-chloro-2-hydroxypropyl) (2,3-dihydroxypropyl) glycidyl ether (BADGE$H2O$HCl, 98%), bisphenol F diglycidyl ether (BFDGE, ~97%), bisphenol F bis(3chloro-2-hydroxypropyl) glycidyl ether (BFDGE$2HCl, ~95%), bisphenol F bis(2,3-dihydroxypropyl) glycidyl ether (BFDGE$2H2O, 97%), novolac glycidyl ether 3-ring (3R-NOGE, 90%), and novolac glycidyl ether 4-ring (4R-NOGE, 90%) were purchased from SigmaeAldrich (St. Louis, MO, USA). Analytical standard of 2, 20 -bis(4hydroxyphenyl)butane (BPB, ~98%) was obtained from TCI America (Portland, OR, USA). 13C-isotopically labeled 13C12-BPA (99%) was purchased from Cambridge Isotope Laboratories (Andover, MA, USA). D6-BADGE was obtained from Toronto Research Chemicals Inc (Toronto, Ontario, Canada). The stock solutions of target analytes and internal standards were prepared at 1 mg/mL in methanol and stored at 20 C. Methanol (HPLC grade), ethyl acetate (ACS grade), and other solvents used in the experiments were purchased from Mallinckrodt Baker (Phillipsburg, NJ, USA). Milli-Q water was purified by an ultrapure water system (Barnstead International, Dubuque, IA, USA). 2.2. Sample collection
2. Materials and methods
The method for the collection of indoor air has been described elsewhere (Wan et al., 2015). Briefly, two polyurethane foam (PUF) plugs (ORBO-1000 PUF dimensions: 2.2 cm O.D 7.6 cm length) obtained from Supelco (Bellefonte, PA, USA) and a quartz fiber filter (Whatman, grade QM-A, pore size: 2.2 mm, 32 mm diameter) were assembled in tandem to collect vapor and particulate phases in indoor air, respectively. To determine the background levels of bisphenols, BADGEs, and NOGEs, newly purchased PUF plugs were extracted twice with 100 mL ethyl acetate and analyzed. The results showed that each of the newly purchased PUF plugs contained 0.25, 0.11, and 0.15 ng BPA, BPF, and BADGE$2H2O, respectively (n ¼ 3). Therefore, all PUF plugs were washed twice by shaking with 100 mL of ethyl acetate for 30 min and then stored in a glass jar that was kept in an oven at 100 C until further use. The quartz fiber filters were heated at 450 C for 20 h and held in an oven at 100 C until further use. The quartz fiber filters were weighed before and after the collection of air samples to determine the mass of particle content in air. All glassware used for sampling and analysis was heated at 450 C for 10 h prior to use. Indoor air samples were collected for 3e24 h by a low-volume air sampler (LP-20; A.P. Buck Inc., Orlando, FL, USA) at a flow rate of 5 L/min. The samples were collected from September to December 2014 at several locations in Albany, New York, USA. The sampling locations were grouped into eight categories: parking garages (n ¼ 3), auto repair shops (n ¼ 4), cars (n ¼ 7), barber shops (n ¼ 5), public places (n ¼ 13), homes (n ¼ 26), laboratories (n ¼ 12), and offices (n ¼ 13). In homes, indoor air samples were collected during day and night time, separately, to compare if there were any diurnal differences in the concentrations of target analytes.
2.1. Standards and reagents
2.3. Sample preparation
Analytical standards of BPA (97%), 4, 40 -(hexafluoroisopropylidene)-diphenol (BPAF, ~97%), 4, 40 -(10 phenylethylidene)bisphenol (BPAP, ~99%), 4, 4 -sulfonyldiphenol (BPS, ~98%), 4, 40 -dihydroxydiphenylmethane (BPF, ~98%), 4, 40 (1,4-phenylenediisopropylidene)bisphenol (BPP, ~99%), 4, 40 -cyclo-
The PUF plugs were extracted by shaking in an orbital shaker (Eberbach Corporation, Ann Arbor, MI, USA) with ethyl acetate for 30 min. The extraction was performed twice with 100 mL solvent each time. The particulate samples were extracted by shaking glass fiber filters with ethyl acetate (20 mL) three times, each time for
J. Xue et al. / Chemosphere 151 (2016) 1e8
5 min. For both PUF plugs and filters, the extracts were combined and concentrated in a rotary evaporator at 55 C to 5 mL. Then, the solution was transferred to a 12-mL glass tube, concentrated under a gentle stream of nitrogen to near dryness, reconstituted with 1 mL of methanol, and transferred into a vial for high performance liquid chromatography-tandem mass spectrometry (HPLC-MS/MS) analysis. Twenty nanograms of 13C12-BPA and d6-BADGE were spiked (as internal standards) into all of the samples and blanks before extraction. 2.4. LC-ESI()MS/MS analysis of bisphenols The chromatographic separation was carried out using a Shimadzu Prominence Modular HPLC system (Shimadzu Corporation, Kyoto, Japan), consisting of a system controller, binary pump, and auto sampler. Identification and quantification of target analytes were performed with an Applied Biosystems API 3200 electrospray triple quadrupole-mass spectrometer (ESI-MS/MS) (Applied Biosystems, Foster City, CA, USA). A Betasil C18 column (2.1 mm 100 mm, 5 mm) (Thermo Electron Corporation Waltham, MA, USA) serially connected to a Javelin guard column (Betasil C18, 2.1 mm 20 mm, 5 mm) (Thermo Electron Corporation) was used. The injection volume was 10 mL, and the mobile phase comprised methanol (A) and Milli-Q water that contained 1% (v/v) ammonium hydroxide (B). The target compounds were separated by gradient elution of mobile phase at a flow rate of 300 mL/min starting at 15% (v/v) A, held for 2 min; increased to 75% A within 3 min, held for 2 min; then further increased to 99% A within 3 min, held for 4 min; and reverted to 15% A at the 14.5th min and held for 5.5 min, for a total run time of 20 min. The MS/MS system was operated in multiple reaction monitoring (MRM) negative ion mode. The compound specific MS/MS parameters are shown in Table S1. Nitrogen was used as both a curtain and a collision gas. The electrospray ionization voltage was set at 4.5 kV. The curtain and collision gas flow rates were set at 25 and 2 psi, respectively, and the source heater was set at 650 C. The nebulizer gas (ion source gas 1) was set at 20 psi, and the heater gas (ion source gas 2) was set at 70 psi. The data acquisition was set at 80 ms for scan speed and 0.70 full width at half maximum (FWHM) for resolving power. 2.5. LC-ESI(þ)MS/MS analysis of BADGEs and NOGEs For the analysis of BADGE, BFDGE, 3R-NOGE, 4R-NOGE, and their derivatives, the chromatographic separation was carried out using an Agilent 1100 Series HPLC system (Agilent Technologies Inc., Santa Clara, CA, USA). Identification and quantification of target analytes were performed with an Applied Biosystems API 2000 ESIMS/MS. The chromatographic columns were similar to that reported above for the analysis of bisphenols. The injection volume was 10 mL, and the mobile phase comprised methanol (A) and MilliQ water/methanol (90:10, % v/v) that contained 2 mM ammonium acetate (B). The target compounds were separated by gradient elution of mobile phase at a flow rate of 300 mL/min starting at 20% (v/v) A, held for 2 min; increased to 75% A within 1 min; further increased to 95% A within 2 min, held for 10 min; and reverted to 20% A at the 16th min and held for 4 min, for a total run time of 20 min. The MS/MS system was operated in MRM positive ion mode. The compound specific MS/MS parameters for BADGEs and select NOGEs are shown in Table S2. Nitrogen was used as both a curtain and a collision gas. The electrospray ionization voltage was set at þ4.5 kV. The curtain and collision gas flow rates were set at 10 psi, and the source heater was set at 450 C. The nebulizer gas (ion source gas 1) was set at 45 psi, and the heater gas (ion source gas 2) was set at 75 psi. The data acquisition was performed at a scan speed of 40 ms and a resolving power of 0.70 FWHM.
3
2.6. Quality assurance and quality control Quantification of bisphenols, BADGEs, and NOGEs was performed by the isotope dilution method, with responses relative to 13 C12-BPA and d6-BADGE, respectively. A 9- to 11-point standard calibration curve, with concentrations ranging from 0.05 to 200 ng/ mL, was used for the quantification of target analytes. The calibration curves were prepared by plotting a concentration-response factor for each target analyte (peak area of analyte divided by peak area of the internal standard) versus the response-dependent concentration factor (concentration of analyte divided by concentration of internal standard). The regression coefficients (r) were 0.99 for all calibration curves. The limits of quantification (LOQs) were determined based on the lowest point of the calibration standard with a signal-to-noise ratio of >10. Method LOQs (MLOQs) were determined in the same manner, with the post-matrix spiked 9- to 11-point calibration curves. MLOQs in the bulk air equaled to the higher MLOQs obtained in either the vapor or the particulate phase. Chemical concentrations in the bulk air were calculated by dividing the sum of the mass of the chemical in the vapor and particulate phases by the volume of air collected. As a check for instrumentation drift in response factors, a midpoint calibration standard was injected after every 10 samples. A pure solvent (methanol) was injected after every 10 samples as a check for carryover of target analytes. Several procedural blanks were analyzed with each batch of samples to determine the contamination arising from laboratory materials and solvents. Efforts were taken to minimize background levels of target analytes. Throughout the analysis, six sets of PUF and glass fiber filter were selected for pre-extraction matrix spike (MS) (two for every 25 samples) by spiking 40 ng of target analytes and passing them through the entire analytical procedure. 2.7. Data analysis Data were acquired with Analyst software version 1.4.1 (Applied Biosystems, Foster City, CA, USA). Statistical analyses were performed with statistics software package R v.3.1.0 and Microsoft Excel 2007. For the calculation of arithmetic mean, median and GM, concentrations below the MLOQ were substituted with a value equal to half the MLOQ. To examine the relationship between chemicals, Spearman (when data did not follow a normal distribution after logarithmic transformation) or Pearson (when data followed a normal distribution after logarithmic transformation) correlation analysis was used. Only those samples with detectable concentrations of target analytes were used when performing correlation analysis. To assess the difference between means, a Student's t-test (when data followed a normal distribution after logarithmic transformation) or ManneWhitney U test (when data did not follow a normal distribution after logarithmic transformation) was used. A ShapiroeWilk test and quantileequantile (QeQ) plot were used to determine the normality of the data. Statistical significance was set at p < 0.05. Daily exposure doses were calculated using Eq. (1).
EDIinh ¼ C AIR IEF=BW
(1)
where C is the concentration of analytes in (bulk) indoor air (ng$m3), AIR is the air inhalation rate (m3$day1), IEF is the indoor exposure factor (represents the fraction of time spent indoors per day), and BW is the body weight (kg). 2.8. Method performance MLOQs of the target analytes ranged from 0.05 to 1.29 ng m3 in
4
J. Xue et al. / Chemosphere 151 (2016) 1e8
the vapor phase and from 0.71 to 35.7 mg g1 in the particulate phase (Table S3). Trace levels of BPA (0.57 ng), BPF (0.88 ng), and BADGE$2H2O (0.2 ng) were found in procedural blanks that contained two PUFs (Table S4). In procedural blanks that contained a quartz filter, BPF and BADGE$2H2O were found at average concentrations of 0.82 and 0.26 ng, respectively. The concentrations found in procedural blanks were subtracted from the concentrations measured in samples. Absolute and relative recoveries were used in this study. Calculation of absolute recovery was based on the instrumental response of analyte. The ratio of the signal for the analyte to that of the internal standard was used for the computation of relative recovery. In the vapor phase, relative recoveries of bisphenols ranged from 86.6% to 103%, and those of BADGEs and NOGEs ranged from 87.5% to 113%. In the particulate phase, relative recoveries of bisphenols, BADGEs, and NOGEs ranged from 98.5% to 132%. 3. Results and discussion 3.1. Concentrations of bisphenols in particulate phase, vapor phase, and bulk air Among the eight bisphenols analyzed, BPA and BPS were the most frequently detected compounds in the vapor phase, with detection rates (DRs) of 48.2% and 26.5%, respectively (Table 1). BPA concentrations in the vapor phase ranged from below MLOQ to 1.85 ng m3 (GM: 0.15 ng m3; median: 0.07 ng m3), and BPS concentrations ranged from below MLOQ to 0.94 ng m3 (GM: 0.04 ng m3; median: 0.03 ng m3) (Table 1). The highest concentration of BPA was found in automobile repair shops (GM: 0.4 ng m3; median: 0.39 ng m3) (Table 1) and BPS in barbershops (GM: 0.08 ng m3; median: 0.03 ng m3) (Table 1). In the particulate phase, BPA and BPF were the predominant compounds found in 75.9% and 63.9% of the samples, respectively (Table 1). The concentrations of BPA in the particulate phase ranged from below MLOQ to 153 mg g1 (GM: 7.19 mg g1; median: 6.75 mg g1) (Table 1). BPF concentrations in the particulate phase ranged from below MLOQ to 2220 mg g1 (GM: 18.1 mg g1; median: 13.9 mg g1) (Table 1). The highest concentration of BPA in the particulate phase was found in barbershops (GM: 11.2 mg g1; median: 9.07 mg g1) (Table 1), and the highest BPF concentration was in cars (GM: 33.8 mg g1; median: 58.8 mg g1) (Table 1). BPA is increasingly replaced with BPF in epoxy resin production, which is the predominant type of epoxy resins used in the automotive industry, and high concentrations of BPF in cars can be explained by this usage. BPS, BPAF, and BPB were less frequently detected in the particulate phase, with DRs ranging from 1.2% to 4.82% (Tables 1 and S5). The GM concentrations of BPA and BPF in airborne particles measured in this study were one to three orders of magnitude higher than those reported in indoor dust collected from homes in Albany, New York, USA (Wang et al., 2015b). The differences in particle sizes between airborne particles (
