GC-MS/MS analysis for source identification of emerging POPs in PM2.5
Yi-Jie Chena,1, Yanhao Zhangb,1, Yanyan Chena,b, Yan Lua, Ruijin Lic, Chuan Dongc, Zenghua Qia, Guoguang
Liua, Zhi-Feng Chena,∗, Zongwei Caia,b,∗∗
a Guangzhou Key Laboratory of Environmental Catalysis and Pollution Control, Guangdong Key Laboratory of Environmental Catalysis and Health Risk Control, School of
Environmental Science and Engineering, Institute of Environmental Health and Pollution Control, Guangdong University of Technology, Guangzhou, 510006, China b State
Key Laboratory of Environmental and Biological Analysis, Department of Chemistry, Hong Kong Baptist University, Hong Kong Special Administrative Region, China c
Institute of Environmental Science, Shanxi University, Taiyuan, China A R T I C L E I N F O A B S T R A C T Keywords:
Emerging POPs have received increasing attention due to their potential persistence and toxicity, but thus far the report GC-MS
regarding the occurrence and distribution of these POPs in PM2.5 is limited. In this study, an extremely sensitive and reliable Hexachlorobutadiene
method, using ultrasonic solvent extraction and silica gel purification followed by gas chromatography coupled with electron Pentachloroanisole
ionization triple quadrupole mass spectrometry, was developed and used for the trace analysis of hexachlorobutadiene Chlorobenzene
(HCBD), pentachloroanisole (PCA) and its analogs chlorobenzenes (CBs) in PM2.5 from Taiyuan within a whole year. The limits PM2.5
of detection and limits of quantitation of analytes were 1.14 × 10−4 Contamination profile
‒2.74 × 10−4 pg m−3 and 3.80 × 10−4‒9.14 × 10−4 pg m−3.
HCBD and PCA were detected at the mean concentrations of 3.69 and 1.84 pg m−3 in PM2.5, which is reported for the first time.
Based on the results of statistical analysis, HCBD may come from the unintentional emission of manufacture or incineration of
chlorinate-contained products but not coal combustion, while O3-induced photoreaction was the potential source of PCA in
PM2.5. The temporal distributions of CBs in PM2.5 were closely related to coal-driven or agricultural activities. Accordingly, our
study reveals the contamination profiles of emerging POPs in PM2.5 from Taiyuan. 1. Introduction
the reversible microbial conversion of pentachlorophenol in soil and sediment
(Vodicnik et al., 1980), and the degradation of structure-like chlorinated
In recent years, an additional set of persistent organic pollutants, including
hydrocarbons such as hexachlorobenzene (Kylin et al., 2017), which is one of
hexachlorobutadiene (HCBD) and pentachloroanisole (PCA), were listed in
the chlorobenzenes (CBs). CBs are widely used in dyeing, medicine, plastics,
Annex C and A of the Stockholm Convention for the elimination of production
and daily chemical products for a long period. The high usages of CBs cause
and use in 2017 (Rauert et al., 2018). HCBD is mainly used as a solvent in the
their residues in the aquatic and terrestrial environment and adverse effects
manufacture of organic chemicals such as rubbers, waterproof plastics, and on environmental organisms. Hexachlorobenzene (HCB) and
pesticides (Wang et al., 2018). It is not allowed to produce HCBD in the
pentachlorobenzene (PeCB) were therefore listed in the Stockholm Convention
European Union and the United States (Juang et al., 2010). In China, however,
in sequence, while their analogs 1,2,4-trichlorobenzene and 1,2,4,5-
there is evidence of HCBD emissions that may come from unintentional
tetrachlorobenzene were regulated as the priority pollutants in the European
production during the manufacture of perchloroethylene, trichloroethylene,
Union (Fliedner et al., 2016; Ma et al., 2014). These restricted chemicals were
and carbon tetrachloride (Wang et al., 2018). Previous reports indicated the
still found in the environmental samples such as soils/sediments from China
detection of HCBD in arctic fish, polar bear fat and water (Fernandes et al.,
(Zhang et al., 2011) and flue gas from Vietnam (Thi et al., 2019). Although they
2019), and the nephrotoxicity induced by HCBD at low levels after repeated
are not on the recent listing of the Stockholm Convention, CBs are chemically
and chronic exposure (Lock and Ishmael, 1979). For PCA, it was reported to be
similar to PCA, which allowed simultaneous extraction and detection in this
highly toxic with a slow biodegradation rate (Rauert et al., 2018). Although PCA study.
is not used as a commercial chemical or pesticide, it was detected in the
The target polychlorinated pollutants are lipophilic compounds with high
atmospheric samples from the Group of Latin America and Caribbean
vapor pressures. They have received increasing concerns because of their
(GRULAC) region (Rauert et al., 2018). PCA can be inadvertently generated by
potential persistence and toxicity (Weber et al., 2008). In general, these
∗ Corresponding author. Guangdong University of Technology, Guangzhou, China.
∗∗ Corresponding author. Hong Kong Baptist University, Hong Kong Special Administrative Region, China.
E-mail addresses: chenzhf@gdut.edu.cn (Z.-F. Chen), zwcai@hkbu.edu.hk (Z. Cai). 1 These
authors contributed equally to this work. https://doi.org/10.1016/j.ecoenv.2020.110368
Received 4 November 2019; Received in revised form 18 February 2020; Accepted 21 February 20200147-6513/ © 2020 Elsevier Inc. All rights reserved. Y.-J. Chen, et al.
Ecotoxicology and Environmental Safety 193 (2020) 110368
polychlorinated pollutants were detected in environmental samples (e.g., air,
the detailed procedures of target analyte pretreatment in environmental solid
soil, and fish muscle) (Zhang et al., 2019a) by gas chromatography coupled with samples are given in Text S2.
electron capture detector (GC-ECD) (Tölgyessy et al., 2016) and gas
GC-EI-MS, GC-EI-MS/MS, GC-PCI-MS/MS and GC–NCI–MS/MS were
chromatography coupled with mass spectrometry (GC-MS) (Zhang et al.,
performed on a Thermo TRACE™ 1300 gas chromatography coupled to a TSQ
2018a). Other than GC-ECD, GCMS is preferable with excellent identification
8000 Evo mass spectrometer (Thermo, USA), while GC-APCI-MS/MS was
and quantification capacities. There are various types of GC-MS with different
conducted on an Agilent 7890 B gas chromatography (Agilent, USA) equipped
ionization sources and mass analyzers. Electron ionization (EI) is commonly
with a Waters Xevo TQ-S triple-quadrupole mass spectrometer (Waters, USA).
used to ionize most analytes, while chemical ionization (CI) could yield a quasi-
The TG-5MS (30 m × 0.25 mm × 0.25 μm) column was used to separate target
molecular ion with high response. For example, negative chemical ionization
analytes. Chromatographic conditions were optimized based on the peak
(NCI) provides lower instrumental detection limits (IDLs) of electron-rich
shape and signal-to-noise ratio of analytes using GC-EI-MS (Table S1), after
chemicals such as halogenated hydrocarbon than EI (Ackerman et al., 2005;
which the mass spectrometric parameters of each type of GCMS, including ion
Fulara and Czaplicka, 2012; Gonzalez-Gago et al., 2015). In recent years, as an
source temperatures, quantifier/qualifier ions, and collision energies, were
alternative to EI and NCI, atmospheric pressure chemical ionization (APCI) is a
optimized and confirmed. The optimum GC and MS parameters are listed in
low-energy ionization technique using a corona needle, exhibiting better Tables S2 and S3.
performance for the detection of analytes (Geng et al., 2016; Tokumura et al.,
2018), such as nitro polycyclic aromatic hydrocarbons (PAHs) and their nitro
2.4. Statistical analysis
derivatives (NPAHs) (Zhang et al., 2019b), and brominated/chlorinated
compounds (Cherta et al., 2013; Portolés et al., 2015). In addition to ionization
Prior to statistical analysis, the concentration value below the limit of
sources, single quadrupole mass spectrometry (MS) is commonly used, but
detection (LOD) was calculated as zero. An independent t-test was used to
triple quadrupole mass spectrometry (MS/MS) shows better instrumental
compare the concentrations of PM2.5 between heating and nonheating
sensitivity by removing matrix interferences (Fulara and Czaplicka, 2012;
seasons. Pearson correlation analysis with a two-tailed test was employed to
Losada et al., 2010a, 2010b; Luo et al., 2007). However, the report about the
explore the pollution source of target analytes. Principal component analysis
comprehensive evaluation of instrumental methods on the detection of target
was conducted to determine the contamination profile of analytes in PM2.5
analytes is limited. It is necessary to develop an ultra-sensitive method for the
samples. Prior to principal component analysis, the data x were log (x+1)
trace analysis of analytes in environmental samples.
transformed. The significant level was set to p < 0.05. All data analysis was
The pretreatment process of target polychlorinated pollutants in fly ash
performed by SPSS 22.0, and Origin 2017 for windows.
was optimized by a previous study (Zhang et al., 2018a) and therefore used in
this study. There is no report regarding the occurrence of HCBD and PCA in 3. Results and discussion
PM2.5, which is a serious pollution issue in China. Therefore, we aimed to
compare the performance of five types of GCMS, including GC-EI-MS, GC-EI-
3.1. Optimization of instrumental parameters
MS/MS, GC-PCI-MS/MS, GC–NCI–MS/ MS and GC-APCI-MS/MS, and to
develop a sensitive and precise approach for the trace analysis of emerging
In order to sharpen the peak of target analytes and increase the signal-to-
POPs (e.g., HCBD, and PCA) and priority pollutants such as CBs. The developed
noise ratio (S/N), instrumental conditions, including gas chromatography and
method was finally attempted to identify the pollution source of emerging
mass spectrometry parameters, were optimized. The mixed standard solution
POPs in PM2.5 samples from Taiyuan, China. 2. Materials and methods
at a concentration of 100 ng L−1 of each analyte was used for the optimization
of instrumental parameters. All target analytes were separated successfully
2.1. Chemicals and reagents
within 20 min (Fig. S1). For the parameters of gas chromatography, splitless
and glass liner tube with quartz wool were selected (Fig. S1) due to better peak
In this study, six target polychlorinated pollutants, including HCBD, PCA,
shapes. Glass liner tube with quartz wool could help the vaporization of
HCB, PeCB, 1,2,4,5-tetrachlorobenzene (TeCB), and 1,2,4-trichlorobenzene
analytes (Grob and Neukom, 1984). Higher injection volume led to the higher
(TCB), were investigated. The information of target analytes is listed in Table 1.
peak area of analytes, but there was no apparent change of the signal-to-noise
Supplier sources of chemicals and reagents are provided in Text S1.
ratio when the injection volumes were set to 1, 2, and 5 μL (Fig. S2). Transfer 2.2. Sampling
line temperature is a key factor that ensures the transportation of gaseous
analytes from chromatographic column to ion source. A peak value of each The sample collection of PM
analyte can be seen under the transfer line temperature of 290 °C (Fig. S3). The
2.5 was carried out in the city of Taiyuan (China)
within a whole year from November 2017 to October 2018. There were two
peak area of analyte showed an increasing trend in the range of 0.8–1.5 mL
samples collected per month, except for February with only one sample. The
min−1 for the flow rate of carrier gas or 250–300 °C for injection temperatures
sampling site was located at Shanxi University, where there is about 300 m
(Fig. S3). High flow rate of carrier gas can improve the peak width and shorten
away from an arterial road. Each PM
the run time, while injection temperature affects the vaporization of liquid
2.5 sample was collected over 24 h on a
preheated (550 °C for 5 h) Whatman quartz microfiber filter (90 mm in
samples. Therefore, 1.5 mL min−1 and 300 °C were chosen as the optimum flow
diameter) using mediumvolume air samplers (AMAE Co., Ltd, China) at a flow
rate and injection temperature. In summary, the optimum gas chromatography
rate of 100 L min−1. After collection, the filter was wrapped with aluminum foil
parameters are listed in Table S2.
and stored at −20 °C before extraction. The concentration of PM
For the parameters of mass spectrometry, the rising peak area of 2.5 was
calculated by the difference value of filters before and after sampling. In total,
23 of PM2.5 samples were analyzed in this study.
2.3. Sample pretreatment and instrumental optimization
Zhang et al. presented a simple pretreatment method for the extraction
and purification of HCBD, PeCB, and HCB in waste incineration fly ash samples
(Zhang et al., 2018a). Based on the previous method with a little modification, 2 Y.-J. Chen, et al.
Ecotoxicology and Environmental Safety 193 (2020) 110368 Table 1
Basic information of target analytes in this study.
The log Kow and vapor pressure values were estimated by the EPI suite model (USEPA, 2019), but experimental values were preferred to calculated values. Table 2
Quantifier/qualifier ions (m/z) and collision energies (V) of target analytes by GC-EI-MS/MS. Compound HCBD PCA TCB TeCB PeCB HCB Quantifier ion 227 > 192 263 > 235 182 > 147 216 > 181 250 > 215 286 > 251 Collison energy 14 10 14 14 16 18 Qualifier ion 225 > 190 265 > 237 180 > 145 214 > 143 215 > 179 284 > 249 Collison energy 14 10 14 26 8 18
Fig. 1. Radar plots of instrumental performance of target
polychlorinated compound standard solution by different
GC-MS methods. IDL, IQL and R2 represent instrumental
detection limit (pg mL−1), instrumental qua ntification limit
(pg mL−1), and the determination coefficient of calibration
Fig. 2. Extracted ion chromatograms (EICs) and signal-to-noise ratio (S/N) of target polychlorinated compounds in PM2.5 samples by different GC-MS methods.
curves. The units of intra-day repeatability and inter-day reproducibility are %.
each analyte was found with the increasing ion temperature value (Fig. S3),
Proposed fragmentation pathways of target polychlorinated compounds in
which was set to 300 °C finally. Before the comparison of the performance of
different types of GC-MS are shown in Fig. S4. Loss of 1–3
five instrumental approaches, including EI-MS, EI-MS/ MS, PCI-MS/MS, NCI-
MS/MS, and APCI-MS/MS, we need to confirm the quantifier and qualifier ion
chlorine atoms from CBs occurred in the EI source and the collision cell of
of each analyte. The quantifier and qualifier ion in the MS or MS/MS condition
MS/MS. For HCBD, the most intense parent ions were found to be [M-Cl]+ at
were optimized by analyzing standard solution in full scan mode or product ion
225 m/z in EI-MS/MS and PCI-MS/MS, and [M]+ at 260 m/ z in EI-MS and APCI-
scan mode to obtain MS or MS2 spectra(Geng et al., 2016; Tokumura et al.,
MS/MS. The difference of parent ions between EIMS and EI-MS/MS may be
2018). TCB, TeCB, and HCB cannot be detected in NCI-MS/MS, which was
explained by the more readily generation of product ions from 225 m/z than
therefore not discussed in the following text. The optimum mass spectrometry
260 m/z. For PCA, the fragmentation pathway was different from other
parameters of each instrumental approach are given in Table 2 and S3.
analytes. This analyte was favorite to lose [CH3] (M = 15) or [CO] (M = 28) rather
than chlorine atoms in MS. The parent ion and the corresponding product ion 3 Y.-J. Chen, et al.
Ecotoxicology and Environmental Safety 193 (2020) 110368
was [M-CH3]+ at 265 m/z and [M–CH3–CO]+ at 237 m/z in EI-MS/MS, while loss
of [CH3] (M = 15) and [CH3–CO] (M = 43) were observed in the collision cell for
PCI-MS/MS and APCI-MS/MS. The ion transition of PCA in MS implied the more
stable of [C6H5–Cl] bond than [O–CH3] bond. Fragmentation analysis of target
analytes in MS could help us predict and identify their environmental behavior,
such as the transformation intermediates during the degradation process.
3.2. Comparison and validation of different GC-MS methods
Good linearity of each analyte was obtained in all instrumental
approaches, with coefficients of determination (R2) > 0.99, especially in GC-EI-
MS/MS (R2 > 0.999) (Fig. 1 and Table S4). Intra-day repeatability and inter-day
reproducibility are expressed as the relative standard deviation (RSD) of at
least three replicates of spiked samples on the same day and three different
days. Three concentrations (low-, middle-, and high-level) including 1, 10, and
100 ng mL−1 were set (Tables S4 and S5). However, only low concentration
levels can be set for GC-APCI-MS/MS since it is not allowed for the detection
of samples with high concentrations of analytes. The results showed that the
intraand inter-day precision of most analytes in APCI-MS/MS were higher than
10%. The high variation may originate from the unstable condition that the
ionization is produced by the pulsed plasma (Geng et al., 2016; Tokumura et
al., 2018). For PCI-MS/MS, the weak intra-day repeatability and inter-day
reproducibility of HCB were found in the low concentration level. For GC-EI-MS
and GC-EI-MS/MS, the intra- and inter-day RSDs were satisfactory within 10%.
The instrumental detection limit (IDL) and instrumental quantitation limit
(IQL) were calculated as three and ten times ratios of signalto-noise under the
lowest detectable solvent-based standard solution, respectively (Table S4). The
IDLs and IQLs of analytes in GC-PCI-MS/ MS were approximately 1–268 times
higher than those of the rest instrumental approaches, indicating the weak
proton affinity of target analytes and weak sensitivity for analyte detection
(Wilson and McCloskey, 1975). HCBD analysis by GC-APCI-MS/MS was not
sensitive due to its high IDL and IQL. The strong electronegativity of chlorine
atoms will lead to the difficulty of loss of electrons from HCBD, affecting the
Fig. 3. The trends of relative abundances of target polychlorinated compounds and
ionization of HCBD in GC-APCI-MS/MS. When compared to GC-APCI-MS/MS,
environmental variables in PM2.5 samples from Taiyuan in heating (November to March)
most analytes exhibited better IDLs and IQLs in GCEI-MS or GC-EI-MS/MS. The
and non-heating (April to October) seasons.
detection of PCA and HCBD was more sensitive in GC-EI-MS/MS than GC-EI-
MS, which was in accordance with a previous report regarding the comparison
purification (Zhang et al., 2018a), the recoveries of analytes were in the range
of performance between MS and MS/MS (Fulara and Czaplicka, 2012). The IDLs
of 87.4–143% (Table S6), while the limits of detection and limits of quantitation
or IQLs of HCBD, HCB and PeCB in GC-EI-MS/MS were 488–1154 times lower
of analytes were 1.14 × 10−4
in comparison to those of GC-EI-MS in previous reports (Zhang et al., 2014,
‒2.74 × 10−4 pg m−3 and 3.80 × 10−4‒9.14 × 10−4 pg 2018a). m−3 for PM2.5 (Table S7).
Besides, total ion chromatograms and extracted ion chromatograms of analytes in spiking PM 3.3. Contamination pro
2.5 samples revealed that GC-EI-MS/MS combined with
files of emerging POPs in PM2.5
the validated pretreatment method (Zhang et al., 2014, 2018a), including
ultrasonic solvent extraction and silica gel purification, produced the lowest
In order to test the practicability of the developed method for the analysis
matrix-related peaks than other approaches (Fig. 2). When compared to MS
of PM2.5 samples, a whole year monitoring was conducted in Taiyuan from
with potential interference originating from chlorinated compounds, MS/MS
November 2017 to October 2018. The mean concentration of PM2.5 was 165
is reported to provide low background noise because of the background
μg m−3 in heating season (Table S8), which was significantly higher (p < 0.05)
removal in the second stage of MS/MS analysis (Losada et al., 2010a, 2010b;
than that of non-heating season (65.9
Luo et al., 2007). Based on different fragmentation pathways of parent ions of
μg m−3). An urban centralized heating
analytes, using MS/MS could elevate the selectivity by eliminating the risk of
system will be open in heating season, indicating that coal combustion was the
misinterpretation of the chromatogram, in which two peaks with the same
predominant pollution source of PM2.5 in Taiyuan (He et al., 2017; Meng et al.,
parent ion cannot be separated (Losada et al., 2010a, 2010b; Luo et al., 2007). 2007). 13
Considering all of these results, GC-EI-MS/MS was selected for the analysis
C labeled surrogate standards were spiked before sample preparation for
of target analytes in this study. By the combined use of GC-EIMS/MS with the
the evaluation of extraction efficiency. The results suggested that good
validated ultrasonic solvent extraction and silica gel
extraction efficiency of analytes for each sample was obtained, with most
surrogate recoveries ranging from 70% to 120% (Table S8). The trends and
concentrations of target analytes in PM2.5 are given in Fig. 3 and Table S8. For
the emerging POPs, the concentrations of HCBD ranged from 0.74 to 5.37 pg
m−3 except for an outlier up to 25.8 pg m−3, while those of PCA were 0.01–11.6
pg m−3, with a peak value in non-heating season. For CBs, TCB, TeCB, PeCB, and 4 Y.-J. Chen, et al.
Ecotoxicology and Environmental Safety 193 (2020) 110368
HCB were detected at the mean concentrations of 0.91, 0.30, 1.37, and 6.09 pg m−3, respectively.
A component matrix followed by principal component (PC) analysis was used
to determine the contamination pattern of analytes in PM2.5
samples. Two principal components PC1 and PC2 that accounted for 40.2%
and 35.6% of the total variance were identified (Fig. 4a). There were three
clusters that were obviously separated from each other. Cluster 1 (PeCB, HCB
and HCBD) and cluster 2 (TCB and TeCB) had high loadings on PC1 and PC2,
implying their source-related characteristics. The mean concentrations of
PeCB and HCB in heating season were higher than those of non-heating
season (Fig. 3), indicating the correlation between their emissions and coal
combustion (He et al., 2017; Meng et al., 2007). There was no obvious
difference of HCBD concentrations between heating and non-heating seasons
so that coaldriven activities were not the primary pollution source of HCBD in
PM2.5 from Taiyuan. However, HCBD may come from the unintentional
emission of manufacture or incineration of chlorinate-contained products
(Wang et al., 2018), which was also the potential source of PeCB and HCB
because of the result of principal component analysis. As we know, the crop-
planting campaigns in Taiyuan are carried out from March to May and from
August to October every year. The higher concentrations of TCB and TeCB in
these two periods indicated that their temporal distributions were associated
with agricultural activities. Although TCB and TeCB as commonly used
pesticides were restricted as the priority pollutants in the European Union
(Fliedner et al., 2016; Ma et al., 2014), our results demonstrate these two
pollutants were still used in China, which was in accordance with previous
studies (Zhang et al., 2011; Zhou et al., 2017).
To further explore the pollution source of target analytes, Pearson
correlation analysis was conducted by the concentrations of analytes and
environmental variables (Fig. 4b). Based on the above-mentioned results,
statistically significant correlation between PeCB, HCB, and PM2.5 (R = 0.439–
0.717, p < 0.01) and TCB, and TeCB (R = 0.707, p < 0.01) confirmed that PeCB
and HCB originated from coal-driven activities, while agricultural activities will
influence the temporal distribution of TCB and TeCB in PM2.5. PCA was reported
to be formed from the transformation of structure-like chlorinated
hydrocarbons such as pentachlorophenol and HCB in certain conditions (Kylin
et al., 2017; Vodicnik et al., 1980). Under ultraviolet irradiation, free radical can
be generated from O3 (Arey et al., 1989; George and Abbatt, 2010; Simone and
Bertram, 2008; Zhang et al., 2018b). In this study, the concentration of PCA
was positively related to the level of O3 (R = 0.467, p < 0.05) or temperature (R
= 0.457, p < 0.05), suggesting that higher O3 concentration and temperature
were a benefit to the photo-generation of PCA. Since it was clearly separated
from other analytes in Fig. 4a, PCA in PM2.5 from Taiyuan probably originated
from O3-induced photoreaction. In addition, there was almost no correlation
between target analytes and other air pollutants such as SO2, CO, and NO2 (Fig. 4b).
It should be noted that, to our knowledge, the quantification and source
identification of HCBD and PCA in PM2.5 are reported for the first time. PM2.5
pollution is a hot environmental issue in China since it has a strong affinity with
toxic substances and causes enormous adverse effects, e.g., respiratory and
cardiovascular diseases, on human health after long term exposure to the
combination of PM2.5 and toxic contaminants (Santibañez et al., 2013; Weber
et al., 2016). Therefore, it is necessary to further study the fate and risk of
emerging POPs in the atmospheric environment. 5 Y.-J. Chen, et al.
Ecotoxicology and Environmental Safety 193 (2020) 110368 4. Conclusions
products but not coal combustion was the potential pollution source of HCBD,
while PCA originated from O3-induced photochemical reaction. In addition, the
Firstly, the parameters of chromatography and mass spectrometry,
temporal variations of CBs in PM2.5 were affected by coal-driven or agricultural
including injection mode, glass liner tube, injection volume, flow rate, injection
activities. Our results not only elucidate the contamination profiles of HCBD
temperature, transfer line temperature, and ion source temperature, were
and PCA in PM2.5 samples but also are a reminder to all researchers in further
optimized. Through the comparison of the performance of different GC-MS
risk assessment of HCBD and PCA in the atmospheric environment.
methods, the developed method obviously elevated the method sensitivity
CRediT authorship contribution statement
and selectivity for the trace determination of emerging POPs in complicated
environmental samples. Secondly, by the use of the developed method, HCBD
Yi-Jie Chen: Writing - original draft, Formal analysis, Investigation,
Fig. 4. Principal component analysis ordination plots (a) and Pearson correlation coefficients (b) based on the concentrations of target analytes or environmental variables in PM2.5
samples. The asterisks, including * and **, represent p < 0.05 and p < 0.01, respectively.
and PCA were found in PM2.5 from Taiyuan, China, which is reported for the
Methodology. Yanhao Zhang: Writing - original draft, Conceptualization,
first time. According to the results of principal component analysis and Pearson
Methodology. Yanyan Chen: Formal analysis, Investigation. Yan Lu: Formal
correlation analysis, the manufacture or incineration of chlorinate-contained
analysis, Investigation. Ruijin Li: Resources, Project administration. Chuan 6 Y.-J. Chen, et al.
Ecotoxicology and Environmental Safety 193 (2020) 110368
Dong: Resources, Conceptualization. Zenghua Qi: Resources, Project
Luo, Q., Wong, M., Cai, Z., 2007. Determination of polybrominated diphenyl ethers in freshwater
administration. Guoguang Liu: Resources, Supervision. Zhi-Feng Chen:
fishes from a river polluted by e-wastes. Talanta 72, 1644–1649.
Ma, H., Zhang, H., Wang, L., Wang, J., Chen, J., 2014. Comprehensive screening and priority
Writing - review & editing, Conceptualization, Methodology, Resources,
ranking of volatile organic compounds in Daliao River, China. Environ.
Formal analysis, Project administration. Zongwei Cai: Writing - review &
Monit. Assess. 186, 2813–2821.
editing, Conceptualization, Resources, Funding acquisition,
Meng, Z., Jiang, X., Yan, P., Lin, W., Zhang, H.Y., 2007. Characteristics and sources of PM2.5 and Supervision.
carbonaceous species during winter in Taiyuan, China. Atmos. Environ. 41, 6901–6908.
Declaration of competing interest
Portolés, T., Sales, C., Gómara, B., Sancho, J.V., Beltrán, J., Herrero, L., González, M.J.,
Hernández, F., 2015. Novel analytical approach for brominated flame retardants based on
The authors declare that they have no known competing financial interests
the use of gas chromatography-atmospheric pressure chemical ionizationtandem mass
spectrometry with emphasis in highly brominated congeners. Anal.
or personal relationships that could have appeared to influence the work Chem. 87, 9892–9899. reported in this paper.
Rauert, C., Harner, T., Schuster, J.K., Eng, A., Fillmann, G., Castillo, L.E., Fentanes, O., Ibarra, M.V.,
Miglioranza, K.S., Rivadeneira, I.M., 2018. Air monitoring of new and legacy POPs in the Acknowledgments
Group of Latin America and Caribbean (GRULAC) region. Environ. Pollut. 243, 1252–1262.
Santibañez, D.A., Ibarra, S., Matus, P., Seguel, R., 2013. A five-year study of particulate matter
We would be grateful for the financial support from the National Natural
(PM2.5) and cerebrovascular diseases. Environ. Pollut. 181, 1–6.
Science Foundation of China (91543202), the Special research program from
Simone, G., Bertram, A.K., 2008. Reactive uptake of NO3, N2O5, NO2, HNO3, and O3 on three types
Ministry of Science and Technology of China (2017YFE0191000), and the Local
of polycyclic aromatic hydrocarbon surfaces. J. Phys. Chem. 112, 3104–3113.
Innovative and Research Teams Project of Guangdong Pearl River Talents
Thi, H.N., Thu, T.N.T., Hai, L.P., Thanh, H.N., Viet, H.C., Thi, H.C., Xuan, T.N., Tung, N.H., 2019. Program (2017BT01Z032).
Emission of unintentionally produced persistent organic pollutants from some industrial
processes in northern Vietnam. Bull. Environ. Contam. Toxicol. 102,
Appendix A. Supplementary data 287–296.
Tokumura, M., Miyake, Y., Wang, Q., Nakayama, H., Amagai, T., Ogo, S., Kume, K., Kobayashi, T.,
Takasu, S., Ogawa, K., 2018. Methods for the analysis of organophosphorus flame
Supplementary data to this article can be found online at https://
retardants—comparison of GC-EI-MS, GC-NCI-MS, LC-ESI-MS/MS, and LC-APCI-MS/MS. J.
doi.org/10.1016/j.ecoenv.2020.110368.
Environ. Sci. Heal. A. 53, 475–481.
Tölgyessy, P., Miháliková, Z., Matulová, M., 2016. Determination of selected chlorinated priority References
substances in fish using quechers method with dual dspe clean-up and gas chromatography.
Chromatographia 79, 1561–1568.
USEPA, 2019. United States Environmental Protection Agency, EPI (Estimation Programs
Ackerman, L.K., Wilson, G.R., Simonich, S.L., 2005. Quantitative analysis of 39 polybrominated
Interface) SuiteTM for Microsoft® Windows. Washington, DC, USA. https://www.
diphenyl ethers by isotope dilution GC/low-resolution MS. Anal. Chem.
epa.gov/tsca-screening-tools/epi-suitetm-estimation-program-interface, version 4.11, 77, 1979–1987. Accessed date: November 2019.
Arey, J., Atkinson, R., Zielinska, B., Mcelroy, P.A., 1989. Diurnal concentrations of volatile
Vodicnik, M.J., Glickman, A.H., Rickert, D.E., Lech, J.J., 1980. Studies on the disposition and
polycyclic aromatic hydrocarbons and nitroarenes during a photochemical air pollution
metabolism of pentachloroanisole in female mice. Toxicol. Appl. Pharmacol. 56,
episode in Glendora, California. Environ. Sci. Technol. 23 (3), 321–327. 311–316.
Cherta, L., Portolés, T., Beltran, J., Pitarch, E., Mol, J.G., Hernández, F., 2013. Application of gas
Wang, L., Bie, P., Zhang, J., 2018. Estimates of unintentional production and emission of
chromatography–(triple quadrupole) mass spectrometry with atmospheric pressure
hexachlorobutadiene from 1992 to 2016 in China. Environ. Pollut. 238, 204–212. Weber, R., Gaus,
chemical ionization for the determination of multiclass pesticides in fruits and vegetables. J.
C., Tysklind, M., Johnston, P., Forter, M., Hollert, H., Heinisch, E., Holoubek, I., Lloyd-Smith, M.,
Chromatogr., A 1314, 224–240.
Masunaga, S., 2008. Dioxin-and POP-contaminated sites—contemporary and future relevance
Fernandes, A., Mortimer, D., Rose, M., Smith, F., Steel, Z., Panton, S., 2019. Recently listed
and challenges. Environ. Sci. Pollut. Res. 15, 363.
Stockholm convention POPs: analytical methodology, occurrence in food and dietary
Weber, S.A., Insaf, T.Z., Hall, E.S., Talbot, T.O., Huff, A.K., 2016. Assessing the impact of fine
exposure. Sci. Total Environ. 678, 793–800. particulate matter (PM
Fliedner, A., Lohmann, N., Rüdel, H., Teubner, D., Wellmitz, J., Koschorreck, J., 2016. Current
2.5) on respiratory-cardiovascular chronic diseases in the New York
City Metropolitan area using Hierarchical Bayesian Model estimates.
levels and trends of selected EU Water Framework Directive priority substances in Environ. Res. 151, 399–409.
freshwater fish from the German environmental specimen bank. Environ.
Wilson, M., McCloskey, J.A., 1975. Chemical ionization mass spectrometry of nucleosides. Pollut. 216, 866–876.
Mechanisms of ion formation and estimations of proton affinity. J. Am. Chem.
Fulara, I., Czaplicka, M., 2012. Methods for determination of polybrominated diphenyl ethers in Soc. 97, 3436–3444.
environmental samples–review. J. Separ. Sci. 35, 2075–2087.
Zhang, H., Jiang, L., Zhou, X., Zeng, T., He, Z., Huang, X., Chen, J., Song, S., 2018a.
Geng, D., Jogsten, I.E., Dunstan, J., Hagberg, J., Wang, T., Ruzzin, J., Rabasa-Lhoret, R., van Bavel,
Determination of hexachlorobutadiene, pentachlorobenzene, and hexachlorobenzene in
B., 2016. Gas chromatography/atmospheric pressure chemical ionization/ mass
waste incineration fly ash using ultrasonic extraction followed by column cleanup and GC-
spectrometry for the analysis of organochlorine pesticides and polychlorinated biphenyls in
MS analysis. Anal. Bioanal. Chem. 410, 1893–1902.
human serum. J. Chromatogr., A 1453, 88–98.
Zhang, H., Shen, Y., Liu, W., He, Z., Fu, J., Jiang, G., Cai, Z., 2019a. A review of sources,
George, I.J., Abbatt, J.P., 2010. Heterogeneous oxidation of atmospheric aerosol particles by gas-
environmental occurrences and human exposure risks of hexachlorobutadiene and its
phase radicals. Nat. Chem. 2, 713–722.
association with some other chlorinated organics. Environ. Pollut.
Gonzalez-Gago, A., Pröfrock, D., Prange, A., 2015. Comparison of GC–NCI MS, GC–ICPMS, and
Zhang, H., Wang, Y., Sun, C., Yu, M., Gao, Y., Wang, T., Liu, J., Jiang, G., 2014. Levels and
GC–EI MS–MS for the determination of PBDEs in water samples according to the Water
distributions of hexachlorobutadiene and three chlorobenzenes in biosolids from
Framework directive. Anal. Bioanal. Chem. 407, 8009–8018.
wastewater treatment plants and in soils within and surrounding a chemical plant in China.
Grob, K., Neukom, H., 1984. Glass wool in the injector insert for quantitative analysis in splitless
Environ. Sci. Technol. 48, 1525–1531.
injection. Chromatographia 18, 517–519.
Zhang, T., Fiedler, H., Yu, G., Ochoa, G.S., Carroll Jr., W.F., Gullett, B.K., Marklund, S., Touati, A.,
He, Q.S., Yan, Y.L., Guo, L.L., Zhang, Y.L., Zhang, G.X., Wang, X.M., 2017. Characterization and
2011. Emissions of unintentional persistent organic pollutants from open burning of
source analysis of water-soluble inorganic ionic species in PM2.5 in Taiyuan city, China.
municipal solid waste from developing countries. Chemosphere 84, Atmos. Res. 184, 48–55. 994–1001.
Juang, D.-F., Lee, C.-H., Chen, W.-C., Yuan, C.-S., 2010. Do the VOCs that evaporate from a heavily
Zhang, X., Yang, M., Sun, X., Wang, X., Wang, Y., 2018b. The experimental observation,
polluted river threaten the health of riparian residents? Sci. Total Environ. 408, 4524–4531.
mechanism and kinetic studies on the reaction of hexachloro-1,3-butadiene initiated by
Kylin, H., Svensson, T., Jensen, S., Strachan, W.M., Franich, R., Bouwman, H., 2017. The trans-
typical atmospheric oxidants. Sci. Total Environ. 627, 256–263.
continental distributions of pentachlorophenol and pentachloroanisole in pine needles
Zhang, Y., Chen, Y., Li, R., Chen, W., Song, Y., Hu, D., Cai, Z., 2019b. Determination of PM
indicate separate origins. Environ. Pollut. 229, 688–695. 2.5-
bound polyaromatic hydrocarbons and their hydroxylated derivatives by atmospheric
Lock, E.A., Ishmael, J., 1979. The acute toxic effects of hexachloro-1:3-butadiene on the rat
pressure gas chromatography-tandem mass spectrometry. Talanta 195,
kidney. Arch. Toxicol. 43, 47–57. 757–763.
Losada, S., Parera, J., Abalos, M., Abad, E., Santos, F.J., Galcerán, M.T., 2010a. Suitability of
Zhou, T., Shang, B., Dong, H., Tao, X., Liu, T., Wang, Y., 2017. Emission characteristics of volatile
selective pressurized liquid extraction combined with gas chromatography–iontrap tandem
organic compounds during pilot swine manure composting. Trans. Chin. Soc.
mass spectrometry for the analysis of polybrominated diphenyl ethers. Agric. Eng. 33, 192–198.
Anal. Chim. Acta 678, 73–81.
Losada, S., Santos, F., Covaci, A., Galceran, M., 2010b. Gas chromatography–ion trap tandem
mass spectrometry method for the analysis of methoxylated polybrominated diphenyl
ethers in fish. J. Chromatogr., A 1217, 5253–5260. 7