Gas chromatography-Orbitrap mass spectrometry screening of organic chemicals in fly ash samples from industrial sources and implications for understanding the formation mechanisms of unintentional persistent organic pollutants

Science of the Total Environment 664 (2019) 107–115

Contents lists available at ScienceDirect

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

Gas chromatography-Orbitrap mass spectrometry screening of organic chemicals in fly ash samples from industrial sources and implications for understanding the formation mechanisms of unintentional persistent organic pollutants Lili Yang a,b, Shen Wang c, Xing Peng c, Minghui Zheng a,b, Yuanping Yang a,b, Ke Xiao a,b, Guorui Liu a,b,⁎ a b c

State Key Laboratory of Environmental Chemistry and Ecotoxicology, Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences, P.O. Box 2871, Beijing 100085, China University of Chinese Academy of Sciences, Beijing 100049, China Thermo Fisher Scientific, Shanghai 200136, China

H I G H L I G H T S

G R A P H I C A L

• A GC-Orbitrap/MS method was developed for screening organic pollutants in fly ash • GC-Orbitrap/MS screening of organic chemicals provides an overview of industrial pollution • Quite different chemical components were found in fly ash from different industries • Halogenated phenols and other precursors for unintentional persistent organic pollutants were identified • Possible formation pathways of chemicals from different industries were deduced

a r t i c l e

i n f o

Article history: Received 15 December 2018 Received in revised form 29 January 2019 Accepted 1 February 2019 Available online 01 February 2019 Editor: Jay Gan Keywords: Gas chromatography-Orbitrap/mass spectrometry Organic chemical Industrial source Polycyclic aromatic hydrocarbon

A B S T R A C T

ω

=

a b s t r a c t Clarifying the occurrences of organic chemicals in fly ash produced during industrial thermal processes is important for improving our understanding of the formation mechanisms of toxic pollutants such as polycyclic aromatic hydrocarbons (PAHs), halogenated PAHs, dioxins, and other unintentional persistent organic pollutants. We developed a highly sensitive gas chromatography-Orbitrap mass spectrometry (GC-Orbitrap/MS) method and applied it to screening of organic pollutants in fly ash samples from multiple industrial thermal processes. The GC-Orbitrap/MS method could detect and quantify organic pollutants at part per billion (ppb) levels. In total, 96 organic chemicals, including alkanes, benzene derivatives, phenols, polycyclic aromatic hydrocarbons, and biphenyl derivatives were identified in the fly ash samples. Several organic chemicals with chlorine or bromine substituents were abundant in secondary copper smelter fly ash, and these might act as precursors for formation of dioxins, brominated dioxins, and other dioxin-like compounds. Several chlorinated and brominated PAH compounds were also found in the secondary copper smelter fly ash. PAHs were dominant chemicals in the secondary aluminum smelter fly ash samples, and were present in much higher concentrations than in the samples from other industries. This indicates that there are different chemical formation pathways in different industries. Possible formation pathways of PAHs and dioxins were investigated and deduced in this study.

⁎ Corresponding author at: State Key Laboratory of Environmental Chemistry and Ecotoxicology, Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences, P.O. Box 2871, Beijing 100085, China. E-mail address: [email protected] (G. Liu).

https://doi.org/10.1016/j.scitotenv.2019.02.001 0048-9697/© 2019 Elsevier B.V. All rights reserved.

108

L. Yang et al. / Science of the Total Environment 664 (2019) 107–115

These results improve our understanding of the formation mechanisms of toxic unintentional persistent organic pollutants and could be useful for reducing their source emissions. © 2019 Elsevier B.V. All rights reserved.

1. Introduction The release of toxic pollutants from industrial activities is an obstacle to green and sustainable development of various industries, especially in developing countries. Fly ash produced and released during multiple industrial thermal processes is an important carrier of toxic pollutants. Many industrial activities, including iron ore sintering, secondary nonferrous smelting, cement production, and waste incineration, can produce large quantities of fly ash. Secondary non-ferrous metal smelting plants and iron ore sintering plants have been identified as major sources of unintentionally produced persistent organic pollutants (POPs) in China (Liu et al., 2013a; Liu et al., 2014b). Recycling of copper or aluminum scrap in secondary non-ferrous metal smelting processes may introduce massive quantities of POP precursors, such as chlorinated and brominated phenols and polybrominated diphenyl ethers. Therefore, a number of toxic organic pollutants such as carcinogenic dioxins, polycyclic aromatic hydrocarbons (PAHs), halogenated PAHs, polychlorinated biphenyls (PCBs), and polychlorinated naphthalenes (PCNs) have been detected in fly ash from multiple industrial thermal processes (Kosnar et al., 2016; Liu et al., 2013a; Liu et al., 2014b; Liu et al., 2015a; Liu et al., 2015b). The fly ash produced during many industrial thermal activities has been classified as hazardous waste in China (Ling and Hou, 1998). Moreover, fly ash is considered to be an important matrix catalyzing heterogeneous formation reactions of unintentional persistent organic pollutants (POPs) (Tuppurainen et al., 1998; Tuppurainen et al., 2003). It has been estimated that the amount of fly ash produced during municipal solid waste incineration in China increased from 431,000–718,000 tons in 2007 to 1,390,000–2,317,000 tons in 2013 (Liu et al., 2015c). Consequently, knowledge of the compositions of fly ash produced from multiple industrial thermal processes is essential for proper disposal and reduction of potential environmental risks. Many studies have characterized the heavy metal contents in various fly ash samples and reported the occurrences of unintentional POPs with high toxicity at trace levels. For example, the levels of dioxins in fly ash from coking industries and metallurgical sources have been reported (Liu et al., 2010; Liu et al., 2013a; Liu et al., 2014b). The occurrences of dioxin-like compounds, including brominated dioxins, halogenated PAHs, PCNs, and PCBs, in fly ash have also been reported for secondary nonferrous smelting and cement production (Hu et al., 2014; Jiang et al., 2015). Although many studies have reported on specific organic pollutants in fly ash samples, investigations on the occurrences of numerous organic chemicals in fly ash samples from multiple industrial processes still need to be conducted. Analyses of unintentional POPs from industrial sources have been conducted by gas chromatography/mass spectrometry (GC/MS) or high resolution gas chromatography combined with high resolution mass spectrometry (HRGC/HRMS) in many studies (Domeno et al., 2004; Liu et al., 2013b; Megson et al., 2016). The thermal extractionGC/MS method has been established for analysis of semi-volatile organic compounds, such as PCDD/Fs, PCNs, and PAHs, in solid matrices without pretreatment (Tsytsik et al., 2008; Tsytsik et al., 2010). Analysis of target unintentional POPs by GC/MS or HRGC/HRMS is useful for determining emission levels and congener profiles. However, target analysis cannot provide a comprehensive overview of organic pollutants from multiple industrial sources. Non-target screening by advanced GC/MS techniques could be used to achieve an overview of organic pollutants from industrial sources. This type of screening of numerous organic chemicals in fly ash from multiple industrial processes would have the following benefits. First, suspect screening is beneficial in many aspects such as time and cost and is suggested for rapid

identification of priority compounds (Park et al., 2018). Full screening of organic pollutants in fly ash is important for comprehensive characterization of the compositions and potential risks of fly ash samples from multiple industrial thermal processes. Second, fly ash is an important matrix for catalyzing the formation of unintentional POPs from potential precursors, and the screening of numerous organic chemicals in fly ash from multiple industrial processes will improve our understanding of the presence of potential precursors for unintentional POP formation. Third, although iron ore sintering, secondary nonferrous smelting, and cement production are important industries in developing countries, little is known about the overall differences among these processes for the chemicals that are formed. Knowledge of composition differences among fly ash samples from various industries will be useful for understanding the formation mechanisms of toxic chemicals during different processes. Screening of numerous organic chemicals in fly ash from these industries will provide helpful information for improving process sustainability and controlling emissions. GC-Orbitrap/MS is an advanced technique for screening numerous organic chemicals (Casado et al., 2018; Mol et al., 2016). Ions injected into the Orbitrap are trapped in an electrostatic field and each ion oscillates axially with a frequency that is proportional to its mass. An image current of these oscillations is measured using a split outer electrode and this image is converted to a mass spectrum using Fourier transform, which is an analysis method commonly used to transform the current signal to a mass spectrum. The resolution in full scan mode by GC-Orbitrap/MS can reach or even better that of selected reaction monitoring with a triple quadruple MS. Moreover, high sensitivity can be achieved even in full scan mode. The high mass resolution of Orbitrap/MS can improve the selectivity for complex sample matrices (Kaufmann et al., 2010). Therefore, to identify as many substances and unknown components as possible for priority target selection, we applied GC-Orbitrap/ MS, which can reach a resolving power in excess of 60,000 (full width half maximum (FWHM) at m/z 200), to non-target analysis of fly ash samples collected from various industries. The chemical components in fly ash samples from iron ore sintering, secondary nonferrous smelting and cement production were compared. The formation mechanisms of chemicals in the various industries were preliminary explored according to the different distribution characteristics of organic chemicals. Possible transformation pathways were deduced and illustrated according to dominant chemicals with high peak areas. This study improves understanding of priority chemical species in fly ash samples from various industrial sources. Variations in the distribution patterns of compounds in fly ash from different sources indicate that toxic pollutants, such as dioxins and dioxin-like compounds, form by different pathways in different thermal processes. This knowledge of chemical formation and transformation mechanisms will provide important information for future control of pollutant formation. 2. Materials and methods 2.1. Fly ash sample information and sample extraction The fly ash samples from iron ore sinter plants, secondary copper, aluminum, and lead smelters, and cement kilns were collected. All the industrial plants used bag filters as air pollution control devices to reduce the release of fly ash. Fly ash samples were collected from the bag filters over 24 h and homogenized. These samples are representative of the fly ash produced over a 24 h period of industrial operation. The collected samples were wrapped with aluminum foil and placed in polyethylene valve bags in the dark. Basic information about the

L. Yang et al. / Science of the Total Environment 664 (2019) 107–115 Table 1 Basic information for the fly ash samples collected from various industries for this study. Source category

Abbreviation

Air pollution control device

Secondary aluminum smelting plant Secondary copper smelting plant Secondary lead smelting plant Cement kiln co-processing solid waste plant Iron ore sintering plant

Al Cu Pb CK

Bag filters Bag filters Bag filters Bag filters

Iron

Electrostatic precipitator

109

industrial plants is given in Table 1. The fly ash samples were Soxhlet extracted with 250 mL of toluene for about 24 h to ensure high extraction rates. Before analysis, the extracts were dehydrated with anhydrous sodium sulfate and concentrated to about 1 mL using a rotatory evaporator.

2.2. GC-Orbitrap/MS parameters for screening of organic chemicals The screening of organic chemicals in the fly ash extracts was conducted by Trace 1310 GC (Thermo Fisher, Waltham, US). This was

b

Relative abundance (%)

Relative abundance (%)

Ralative abundance (%)

a

Fig. 1. (a) The fragments of 2-chloro-5-methylphenyl acetate retained at 21.69 min and (b) its refined isotope information.

110

L. Yang et al. / Science of the Total Environment 664 (2019) 107–115

connected to an Orbitrap/MS spectrometer (Thermo Fisher Scientific, Waltham, US) equipped with a Triplus RSH autosampler. A TG-5 MS column (60 m × 0.25 mm, 0.25 μm) (Thermo Fisher Scientific, Waltham, US) was used for separation of organic chemicals in the fly ash extracts. The injection volume was 1.0 μL, and we used a 50:1 split injection. The oven temperature was initially held at 50 °C for 1 min, increased to 280 °C at 5 °C min−1 and held for 5 min, and then increased to 310 °C at 20 °C min−1 and held for 20 min. The injector temperature was 280 °C. Helium (purity of 99.9995%) was used as the carrier gas with a constant flow rate of 1.2 mL min−1. We used an electron impact ion source with an ion source temperature of 250 °C. Organic chemicals in the fly ash extracts were screened in full scan mode. The full MS scan range was m/z 50 to 650. The resolution was around 60,000 FWHM (at m/z 200). The automatic gain control target was set at 1.0 × 106 to control the number of ions entering into the ion trap and reduce the space charge effect.

2.3. Data interpretation

3. Results and discussion 3.1. Method for structural identification of organic chemicals in the fly ash extracts by GC-Orbitrap/MS The high resolution of the Orbitrap/MS effectively isolated the chemicals and excluded interferences from the matrix or other chemicals. For non-target screening, the mass error tolerance filter was set at 3 ppm to eliminate matrix interferences. Toluene was analyzed by the same method as a solvent blank to exclude interferences from systematic errors. Compared with the theoretical values, the mass errors for operation at 60,000 FWHM (m/z 200) were below 1 ppm for extraction of the ion chromatograms of all analytes of interest. Taking phenol as an example, the mass error of every scanning point was around ±0.0002 Da (Fig. S1). The phenol standard (Dr. Ehrenstorfer GmbH, Augsburg, Germany) was injected into the GC-Orbitrap/MS to investigate the sensitivity of the MS. Our results showed that high sensitivity could be achieved with the GC-Orbitrap/MS even at an ultra-trace concentration (2.5 μg L−1) (Fig. S2). The original GC–MS scan spectrum may have interferences from instrumental noise, column bleeding, and the matrix. In addition, peak

Relative abundance (%)

Full scan mode was used to identify organic chemicals in the fly ash extracts. The data were acquired and processed using Trace Finder 4.1 (Thermo Fisher Scientific, Waltham, US). The full scan results were retrieved and matched with the MS spectrum database configured in the GC-Orbitrap/MS and the NIST Mass Spectral Library 2014. Some chemicals could not be retrieved from the mass spectral library. For these unknown chemicals, high accuracy mass numbers from the full

scan result by GC-Orbitrap/MS are useful for structure identification. Furthermore, isotope ratios could be used to further determine the elemental compositions and structures.

Relative abundance (%)

m/z

m/z Fig. 2. Identification parameters of 3-chlorophenol and perylene.

L. Yang et al. / Science of the Total Environment 664 (2019) 107–115

overlaps may occur between different compounds with similar retention times in the samples. Therefore, to ensure accurate identification of the peaks in the MS spectrum, the deconvolution function of Trace Finder 4.1 was used first to deal with overlapping peaks. Then, matching between the deconvoluted MS spectrum and the spectrum library (NIST 2014 in this study) was conducted to preliminary qualify the unknown compounds in the samples. All the matches were ranked using a retrieval score (SI) and a high resolution filtration (HRF) score. The HRF score was set using the high resolution and high mass accuracy of the Orbitrap/MS, which can precisely distinguish compounds with similar mass-to-charge ratios (m/z) and fragment ions. The equation for calculation of the HRF score is shown below. The intensity in the equation is the relative abundance of the fragment ion, and this was multiplied by the mass-to-charge ratio (m/z) of the fragment ion. P m z

HRF ¼ P m z

 Intensity

111

compound was compared with the RI in the database. In this way, even though the SI and HRF scores of isomers will be quite similar, the RI can be used to differentiate isomers without the use of standards. The representative characteristic of high resolution of Orbitrap/MS could be used to distinguish tiny differences in the molecular weights of the isotope-refined structure, which provides high accuracy for chemical identification. For example, for the compound eluted at 21.69 min, accurate fragments were obtained at the resolution of 60,000. From the accurate m/z 142.0180, this compound was identified as 2-chloro-5-methylphenyl acetate. The structure was verified using the refined isotope results, which showed it contained C, H, Cl, and O and had an elemental composition of C7H7ClO (Fig. 1).

3.2. Organic contaminants in fly ash samples from different industrial sources



explained  100%  Intensity observed

ð1Þ

Furthermore, to distinguish compounds with the same fragment ions, such as isomers, the calculated retention index (RI) of the

In this study, we used the following screening and identification strategy. First, the Orbitrap/MS data and low resolution data from the NIST 2014 library were used to identify compounds in fly ash samples from different industries. Then, the SI and HRF scores were used to differentiate

Table 2 Information of the 96 organic chemicals detected in fly ash samples by GC-Orbitrap/MS screening. No Component name

Formula

Rt M/Z (min)

Error No Component name (ppm)

Formula

Rt M/Z (min)

Error (ppm)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48

C8H16 C8H16 C6H14O C7H16 C8H16 C8H18 C8H16 C8H18 C7H12O2 C9H20 C9H18O C9H20 C8H16 C6H14 C8H16 C9H20 C9H18 C8H10 C7H6O C6H6O C7H7Cl C7H8O C7H7Br C7H7Br C7H8O C7H7Br C7H8O C7H8O C8H8O2 C7H7ClO C10H12O C11H7N C8H9Br C9H10O2 C7H6O2 C9H9BrO2 C7H7BrO C6H5ClO C9H9ClO2 C6H5BrO C7H8O2 C7H7BrO C7H7BrO C13H12 C14H14 C11H16O2 C14H20O2 C14H14

7.60 7.86 7.88 7.91 8.07 8.15 8.21 8.28 8.37 8.44 8.61 8.79 8.86 8.88 8.93 9.10 9.45 9.57 12.56 12.78 14.16 14.68 14.97 15.07 15.15 15.21 15.77 16.06 16.64 16.76 18.24 28.08 18.38 18.62 18.63 18.70 19.27 19.50 21.69 22.13 23.60 24.25 24.75 25.51 26.07 26.89 27.09 27.90

0.0 0.2 0.4 0.2 0.0 0.6 0.8 0.4 0.0 0.2 0.0 0.4 0.2 0.0 0.8 0.4 0.4 0.9 0.3 0.6 0.9 1.1 0.2 0.2 0.4 0.2 1.1 0.4 0.3 0.3 1.2 0.5 0.5 0.4 0.3 0.4 0.4 0.5 0.3 0.2 0.0 0.4 0.4 0.7 0.2 0.3 0.3 0.2

C14H14 C14H14 C14H14 C14H14 C14H14 C14H14 C14H14 C15H16 C15H16 C15H16 C15H16 C14H14O C15H16 C14H12 C14H12O2 C14H14O C14H10 C16H12 C14H8O2 C14H9Cl C12H6O3 C15H8O C16H10 C16H10 C16H10 C13H6Cl2O C14H8Cl2 C16H9Cl C18H12 C18H10 C18H12 C18H12 C20H14 C20H14 C20H14 C20H12 C20H12 C20H12 C16H8Br2 C16H8Br2 C20H12 C20H12 C20H12 C22H12 C22H12 C22H12 C22H14 C22H12

28.28 28.69 28.77 28.88 29.13 30.34 30.61 30.95 31.23 31.33 31.43 31.72 31.72 33.19 34.35 34.63 34.97 38.80 38.83 39.25 39.92 40.31 40.68 41.13 41.74 43.07 43.19 44.53 46.49 46.54 47.48 47.66 50.26 50.65 50.95 53.35 53.46 53.78 54.04 54.12 54.51 54.73 55.10 59.62 60.13 60.27 60.81 61.57

0.7 0.0 0.7 0.2 0.7 0.2 0.2 0.3 0.1 0.1 0.1 0.9 0.9 0.0 0.2 0.0 0.0 0.6 1.0 0.2 0.1 1.1 1.0 0.1 0.1 0.2 0.2 0.0 1.2 1.2 1.2 1.2 0.1 0.1 0.1 0.6 0.6 0.6 0.4 0.4 0.6 0.6 0.6 0.5 0.5 0.0 0.0 0.5

Diisobutylene Cyclopentane, 1-ethyl-1-methyl2-Hexanol Hexane, 2-methylCyclohexane, 1,3-dimethylHexane, 2,4,4-trimethyl2-Pentene, 3,4,4-trimethyl-, cis Hexane, 3-ethylAllyl isobutyrate Hexane, 2,3,5-trimethylEther, 3-butenyl pentyl Heptane, 3,5-dimethyl1,2-Dimethyl cyclohexane, cis 2,4,4-Trimethyl hexane Cyclohexane, ethyl2-Methyl-3-ethylhexane 7-Methyl-1-octene Ethyl benzene Benzadehyde Phenol Benzyl chloride Benzyl alcohol Benzene, 1-bromo-3-methylBenzene, 1-bromo-2-methylPhenol, 2-methylBenzyl bromide Phenol, 3-methylBenzyl formate Methyl benzoate Phenol, 2-chloro-5-methylBenzyl isopropenyl ether 2-Naphthalenenitrile Benzene, 1-bromo-2,4-dimethyl4-Methylphenyl acetate Benzoic acid Phenol, 2-bromo-4-methyl-, acetate Phenol, 2-bromo-4-methylPhenol, 3-chloroPhenol, 2-chloro-5-methyl-, acetate Phenol, 4-bromo1,2-Benzenediol, 4-methylPhenol, 4-bromo-3-methylPhenol, 2-bromo-5-methyl1,1′-Biphenyl, 4-methyl1,1′-Biphenyl, 2,3′-dimethylPhenol, 3-(1,1-dimethylethyl)-4-methoxy2,6-Di-tert-butylquinone 1,1′-Biphenyl, 2-ethyl-

97.1012 83.0855 69.0699 85.1012 97.1012 71.0855 97.1012 85.1012 71.0855 85.1012 71.0855 70.0777 97.1012 71.0855 83.0855 69.0699 69.0699 91.0541 105.0335 94.0413 91.0541 107.0493 169.9725 169.9725 108.0570 169.9725 107.0493 108.0570 105.0334 142.0180 91.0543 153.0572 183.9881 105.0335 122.0362 185.9675 185.9675 128.0024 142.0180 171.9519 124.0519 185.9675 185.9675 167.0854 182.1089 165.0910 177.1273 182.1090

49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96

2,4′-Dimethyl biphenyl Bibenzyl 2,2′-Dimethylbiphenyl 1,1′-Diphenylethane 4-Ethylbiphenyl 4,4′-Dimethylbiphenyl 4-Benzyltoluene 2,4′-Dimethyldiphenylmethane 2,2′-Dimethyldiphenylmethane Methane, di-p-tolyl2,3′-Dimethyldiphenylmethane Benzyl ether 3-Isopropyl-1,1′-biphenyl 1-Methylfluorene Benzyl Benzoate 2-Benzyl-p-cresol Phenanthrene Naphthalene, 2-phenylCorbit Benzene, 1-chloro-3-(phenylethynyl)Naphthalic anhydride Cyclopenta(def)phenanthrenone Fluoranthene Diphenylbiacetylene Pyrene 3,6-Dichlorofluorenone 9,10-Dichloroanthracene 1-Chloropyrene Chrysene Cyclopenta[cd]pyrene Benz[a]anthracene Triphenylene Heptafulvadienediyne 2,2′-Binaphthalene Phenanthrene, 2-phenylPerylene Benzo[b]fluoranthene Benzo[e]pyrene 2,7-Dibromopyrene 1,3-Butadiyne, 1,4-di(4-bromophenyl)benzo[e]pyran Benzo[e]pyrene Benzo[k]fluoranthene Anthanthren Benzo[ghi]perylene Picene Benzo[b]chrysene Indeno[1,2,3-cd]pyrene

167.0854 182.1090 167.0854 182.1089 167.0854 182.1090 182.1090 196.1247 181.1012 181.1012 181.1012 91.0541 91.0541 178.0777 194.0726 198.1039 178.0777 204.0932 208.0516 212.0387 154.0413 204.0572 202.0779 202.0777 202.0777 247.9791 245.9998 236.0387 228.0931 226.0774 228.0930 228.0930 254.1090 254.1090 254.1090 252.0932 252.0932 252.0932 357.8989 357.8989 252.0932 252.0932 252.0932 276.0932 276.0932 278.1090 278.1090 276.0932

112

L. Yang et al. / Science of the Total Environment 664 (2019) 107–115

between compounds with very small differences in their accurate mass numbers. Finally, the identities of the analytes were verified using their RIs. Using these three parameters, 96 organic chemicals were identified unambiguously in the fly ash extracts from different industrial sources. The identification parameters of 3-chloro-phenol and perylene were shown as examples in Fig. 2. Chain alkanes, benzene derivatives, ethers, PAHs, halogenated PAHs, and biphenyls and their derivatives were identified in the fly ash extracts (Table 2). The alkanes eluted first from the chromatographic column, followed by the benzene derivatives, biphenyl derivatives, and high molecular weight PAHs. The total ion chromatograms of organic chemicals in fly ash extracts from multiple industrial sources are shown in Fig. 3. The mass concentrations of the compounds in the fly ash samples were compared using their peak areas (Table S1). The masses of organic compounds in fly ash samples from secondary copper smelting were notably higher than those in samples from the other four industries, followed in order by secondary lead smelters, secondary aluminum smelters, iron ore sinter plants, and cement kilns. Among the industries investigated, fly ash samples from cement kilns had the lowest mass concentrations of pollutants and lowest number of species. This could occur because these plants are operated under alkaline conditions at high temperatures, which may result in high destruction efficiencies for organic chemicals (Zhao et al., 2017). Therefore, cement kilns will show relatively low pollution levels compared with other industries. The peak areas of benzene derivatives were much higher than those of alkanes or PAHs. Organic chemicals substituted with chlorine or bromine were detected in the fly ash samples from secondary copper smelters. The chlorinated benzene derivatives such as benzyl chloride, 2-chloro-5methylphenol, 3-chlorophenol, 2-chloro-5-methylphenyl-acetate, and 1-chloro-3-(phenylethynyl)-benzene had much higher peak areas in the secondary copper smelter fly ash samples than in the samples from the four other industries. The concentrations of the brominated benzene derivatives such as 1-bromo-3-methylbenzene, benzyl bromide, 1bromo-2,4-dimethylbenzene, 2-bromo-4-methylphenyl-acetate, 2bromo-4-methylphenol, 4-bromophenol, and 4-bromo-3-methylphenol were also much higher in the secondary copper smelter fly ash samples

than in the samples from the other industries. A previous study showed that high concentrations of toxic chlorinated and brominated dioxins could be formed and emitted to the atmosphere during secondary copper smelting (Wang et al., 2016), for the reason that Cu has stronger catalytic power for the polychlorinated dibenzo-p-dioxin and dibenzofuran (PCDD/F) formations than other metals such as Mg, Zn, Fe, and Pb (Chin et al., 2012). Chlorinated and brominated benzene derivatives, which were abundant during secondary copper smelting, were identified as dominant precursors for dioxins (Blumenstock et al., 1999; Lavric et al., 2005). Therefore, the occurrences of chlorinated and brominated benzene derivatives could lead to high concentrations of chlorinated and brominated dioxins and other similar toxic pollutants in fly ash from secondary copper smelters. For the industries investigated in this study, except for the secondary aluminum smelters, chain alkanes and benzene derivatives were present in higher proportions in the fly ash than other chemicals. PAHs were dominant in the fly ash samples from secondary aluminum smelters, quite different from the chemical compound distribution pattern of other sources. PAHs with high numbers of benzene rings, such as chrysene, triphenylene, perylene, benzo[e]pyrene, and anthanthrene, were mainly detected in the samples from secondary aluminum smelters (Fig. 4a). Meanwhile, PAHs with low numbers of benzene rings and chlorine and bromine substituents, such as 3,6dichlorofluorenone, 9,10-dichloroanthracene, 1-chloropyrene, and 2,7dibromopyrene, were mostly observed in secondary copper smelting fly ash. These results show that halogenated PAHs with low numbers of rings were mainly present in the secondary copper smelter fly ash, whereas nonhalogenated PAHs with high numbers of rings were mainly present in the aluminum smelter fly ash. These observations may be determined by the concentrations of dissociated Cl and Br in the samples in this study. Benzene chemicals with halogen substituents were present in high concentrations in the secondary copper smelter samples and barely detectable in the secondary aluminum smelter samples (Fig. 4b). The results can be explained by the different catalysis effects of metals. In a previous study, we investigated the catalysis effects of

Fig. 3. Total ion chromatograms of organic chemicals in fly ash extracts from multiple industrial sources. Al is short for secondary aluminum smelter; Cu is short for secondary copper smelter; Pb is short for secondary lead smelter; CK is short for cement kiln co-processing solid waste; and Iron is short for iron ore sintering plant.

L. Yang et al. / Science of the Total Environment 664 (2019) 107–115

different metals on dioxin formation from 2,4-dichloro-1-naphthol (Yang et al., 2017a). Among the metal catalysts, CuO provided the best catalysis of chlorinated radical intermediate formation. Then, higher chlorinated PCDD/Fs formed by dimerization of the chlorinated radical intermediates (Fig. S3) (Yang et al., 2017a). However, radical intermediates without halogen substituents were formed under Al2O3 catalysis when the same precursor was used, and PCDD/Fs with a low degree of chlorination (e.g., 2,3,7,8-tetrachlorinated dioxins) became dominant. Therefore, under CuO catalysis, chlorinated benzene derivatives such as chlorinated phenol were dominant in the products, and this agreed with the results shown in Fig. 4b. However, under the catalysis of Al2O3, benzene derivatives and active Cl or Br atoms could exist in the system. The high concentrations of active Cl atoms could accelerate the abstraction of aromatic H from stable PAH molecules as shown in Eq. 2, which would activate them for further growth (Frenklach, 1990). Aryl−H þ Cl→Aryl  þHCl

ð2Þ

Therefore, PAHs are dominant in the secondary aluminum smelter fly ash samples and halogenated PAHs are mainly present in the copper

113

smelter fly ash. We confirmed the occurrences of these chlorinated PAH congeners in secondary copper smelter fly ash in a previous study using HRGC/HRMS (DFS, Thermo Fisher Scientific, Waltham, US) (Jin et al., 2017). Because PAHs and halogenated PAHs have been identified as important dioxin precursors, dioxins with a low degree of halogenation will form easily during the secondary aluminum smelting processes. By contrast, highly halogenated dioxins will be more abundant in secondary copper smelter fly ash because of the high contents of halogenated PAHs and halogenated benzene derivatives. 3.3. Implications for the formation mechanisms and emission control There are thousands of metallurgical plants in operation (Yang et al., 2017b) and control of pollutants emitted from these plants is a huge obstacle to sustainable development of these industries in developing countries. To reduce emissions, the formation mechanisms of toxic pollutants such as PAHs, halogenated PAHs, PCDD/Fs, and PCBs need to be clarified, especially for different kinds of industries. The formation mechanisms of PCBs, PCNs, and PCDD/Fs during thermal processes were quite similar, and good correlation was found among these POPs

Fig. 4. Distribution patterns of (a) PAHs and halogenated PAHs and (b) benzene derivatives with chlorine or bromine substituents in fly ash samples from multiple industries.

114

L. Yang et al. / Science of the Total Environment 664 (2019) 107–115

Fig. 5. Possible relationships and formation pathways for chemicals observed in fly ash samples from various industries.

in fly ash samples in previous study (Iino et al., 2001). In addition to formation via precursor dimerization, in the presence of a rich carbon source (e.g., PAHs), chlorine source, and metal catalyst, PCBs, PCNs and PCDD/Fs can also be formed through de novo synthesis. We postulated potential formation pathways for toxic pollutants (Fig. 5), according to the obtained products with high response values analyzed by Orbitrap/MS and the studies that have been performed previously (Liu et al., 2014a). Halogenated or methyl-substituted benzene chemicals were important precursors and can be dimerized to biphenyls such as 4-ethylbiphenyl and dimethyl biphenyl (Fig. 5). Chlorination of these chemicals could also be occurred to form more toxic polychlorinated biphenyls. In addition, chlorinated or brominated phenols, which are dominant precursors of POPs such as PAHs, PCDD/Fs and PCBs (Kim et al., 2007), were also observed in the fly ash, especially that from secondary copper smelting. The large differences in chemical compositions of fly ash samples from secondary copper and secondary aluminum smelting suggest that chlorinated precursors are easily dechlorinated and active Cl can be formed to further accelerate the growth of PAHs under Al3+ catalysis. Therefore, PAHs with many benzene rings are the dominant products from the secondary aluminum smelting process. Conversely, under the catalysis of Cu2+, chlorinated precursors tend to be dehydrogenized and dimerized to PCDD/Fs and some other chlorinated PAHs with few benzene rings. In conclusion, according to the screening results, benzene chemicals such as phenols and benzadehyde were abundant in the samples, and acted as precursors for toxic PAHs, PCBs, PCNs, and Cl/Br-PAHs. However, the formation pathways of these toxic pollutants were quite different under catalysis with different metals, and this contributed to the differences in predominant chemicals in the fly ash samples from different industries. The catalysis effect and formation pathways in different industries need to be investigated further for better targeting and implementation of source emission control. 4. Conclusions A sensitive GC-Orbitrap/MS method was developed and applied to the screening of organic chemicals in fly ash from industrial sources including the secondary metal smelting plants, cement kilns, and iron ore sinter plants. With this method, totally 96 organic chemicals, including alkanes, benzene derivatives, PAHs, halogenated PAHs, and biphenyl derivatives, were identified in the fly ash samples. Chlorinated or brominated benzenes and PAHs were mostly observed in secondary copper smelting fly ash, and might act as precursors for formation of dioxins,

brominated dioxins, and other dioxin-like compounds. Unlike the chemical distribution pattern in secondary copper smelting fly ash, PAHs were dominant in samples from secondary aluminum smelters, where they present in much higher concentrations than in samples from other sources. These results indicate there are different chemical formation pathways in different industries. Further studies should be conducted to clarify the detailed molecular mechanisms of the unintentional POPs produced by various industrial processes. This could provide important information for POP control. Acknowledgments Funding: this work was supported by the National Natural Science Foundation of China [grant number 21777172, 91843301] Beijing Natural Science Foundation [grant number 8182052], the Strategic Priority Research Program of the Chinese Academy of Sciences [grant number XDB14020102], and the Youth Innovation Promotion Association of the Chinese Academy of Sciences [grant number 2016038]. Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi. org/10.1016/j.scitotenv.2019.02.001. References Blumenstock, M., Zimmermann, R., Schramm, K.W., Kaune, A., Nikolai, U., Lenoir, D., et al., 1999. Estimation of the dioxin emission (PCDD/FI-TEQ) from the concentration of low chlorinated aromatic compounds in the flue and stack gas of a hazardous waste incinerator. J. Anal. Appl. Pyrolysis 49, 179–190. Casado, J., Santillo, D., Johnston, P., 2018. Multi-residue analysis of pesticides in surface water by liquid chromatography quadrupole-Orbitrap high resolution tandem mass spectrometry. Anal. Chim. Acta 1024, 1–17. Chin, Y.-T., Lin, C., Chang-Chien, G.-P., Wang, Y.-M., 2012. PCDD/F formation catalyzed by the metal chlorides and chlorinated aromatic compounds in fly ash. Aerosol Air Qual. Res. 12, 228–236. Domeno, C., Martinez-Garcia, F., Camop, L., Nerin, C., 2004. Sampling and analysis of volatile organic pollutants emitted by an industrial stack. Anal. Chim. Acta 524, 51–62. Frenklach, M., 1990. Production of polycyclic aromatic hydrocarbons in chlorine containing environments*. Combust. Sci. Technol. 74, 283–296. Hu, J., Zheng, M., Liu, W., Nie, Z., Li, C., Liu, G., et al., 2014. Characterization of polychlorinated dibenzo-p-dioxins and dibenzofurans, dioxin-like polychlorinated biphenyls, and polychlorinated naphthalenes in the environment surrounding secondary copper and aluminum metallurgical facilities in China. Environ. Pollut. 193, 6–12. Iino, F., Tsuchiya, K., Imagawa, T., Gullett, B.K., 2001. An isomer prediction model for PCNs, PCDD/Fs, and PCBs from municipal waste incinerators. Environ. Sci. Technol. 35, 3175–3181.

L. Yang et al. / Science of the Total Environment 664 (2019) 107–115 Jiang, X., Liu, G., Wang, M., Zheng, M., 2015. Formation of polychlorinated biphenyls on secondary copper production fly ash: mechanistic aspects and correlation to other persistent organic pollutants. Sci. Rep. 5, 13903. Jin, R., Liu, G., Zheng, M., Jiang, X., Zhao, Y., Yang, L., et al., 2017. Secondary copper smelters as sources of chlorinated and brominated polycyclic aromatic hydrocarbons. Environ. Sci. Technol. 51, 7945–7953. Kaufmann, A., Butcher, P., Maden, K., Walker, S., Widmer, M., 2010. Comprehensive comparison of liquid chromatography selectivity as provided by two types of liquid chromatography detectors (high resolution mass spectrometry and tandem mass spectrometry): "where is the crossover point?". Anal. Chim. Acta 673, 60–72. Kim, D.H., Mulholland, J.A., Ryu, J.Y., 2007. Chlorinated naphthalene formation from the oxidation of dichlorophenols. Chemosphere 67, S135–S143. Kosnar, Z., Mercl, F., Perna, I., Tlustos, P., 2016. Investigation of polycyclic aromatic hydrocarbon content in fly ash and bottom ash of biomass incineration plants in relation to the operating temperature and unburned carbon content. Sci. Total Environ. 563-564, 53–61. Lavric, E., Konnov, A., Ruyck, J., 2005. Modeling the formation of precursors of dioxins during combustion of woody fuel volatiles. Fuel 84, 323–334. Ling, Y.C., Hou, P.C.C.A., 1998. Taiwanese study of 2,3,7,8-substituted PCDD/DFs and coplanar PCBs in fly ashes from incinerators. J. Hazard. Mater. 58, 83–91. Liu, G., Zheng, M., Lv, P., Liu, W., Wang, C., Zhang, B., et al., 2010. Estimation and characterization of polychlorinated naphthalene emission from coking industries. Environ. Sci. Technol. 44, 8156–8161. Liu, G., Zheng, M., Cai, M., Nie, Z., Zhang, B., Liu, W., et al., 2013a. Atmospheric emission of polychlorinated biphenyls from multiple industrial thermal processes. Chemosphere 90, 2453–2460. Liu, G.R., Zheng, M.H., Cai, Z.W., Wu, Y.N., Jiang, G.B., 2013b. Dioxin analysis in China. TrAC, Trends Anal. Chem. 46, 178–188. Liu, G., Cai, Z., Zheng, M., 2014a. Sources of unintentionally produced polychlorinated naphthalenes. Chemosphere 94, 1–12. Liu, G., Lv, P., Jiang, X., Nie, Z., Zheng, M., 2014b. Identifying iron foundries as a new source of unintentional polychlorinated naphthalenes and characterizing their emission profiles. Environ. Sci. Technol. 48, 13165–13172. Liu, G., Cai, Z., Zheng, M., Jiang, X., Nie, Z., Wang, M., 2015a. Identification of indicator congeners and evaluation of emission pattern of polychlorinated naphthalenes in industrial stack gas emissions by statistical analyses. Chemosphere 118, 194–200. Liu, G., Jiang, X., Wang, M., Dong, S., Zheng, M., 2015b. Comparison of PCDD/F levels and profiles in fly ash samples from multiple industrial thermal sources. Chemosphere 133, 68–74.

115

Liu, G., Zhan, J., Zheng, M., Li, L., Li, C., Jiang, X., et al., 2015c. Field pilot study on emissions, formations and distributions of PCDD/Fs from cement kiln co-processing fly ash from municipal solid waste incinerations. J. Hazard. Mater. 299, 471–478. Megson, D., Reiner, E.J., Jobst, K.J., Dorman, F.L., Robson, M., Focant, J.F., 2016. A review of the determination of persistent organic pollutants for environmental forensics investigations. Anal. Chim. Acta 941, 10–25. Mol, H.G.J., Tienstra, M., Zomer, P., 2016. Evaluation of gas chromatography - electron ionization - full scan high resolution Orbitrap mass spectrometry for pesticide residue analysis. Anal. Chim. Acta 935, 161–172. Park, N., Choi, Y., Kim, D., Kim, K., Jeon, J., 2018. Prioritization of highly exposable pharmaceuticals via a suspect/non-target screening approach: a case study for Yeongsan River, Korea. Sci. Total Environ. 639, 570–579. Tsytsik, T., Czech, J., Carleer, R., 2008. Thermal extraction coupled with gas chromatography–mass spectrometry as a tool for analysing dioxin surrogates and precursors in fly ash. J. Chromatogr. A 1210, 212–221. Tsytsik, P., Czech, J., Carleer, R., Bueken, A., 2010. Thermal extraction of dioxin surrogates and precursors from filter dust: effects of temperature, time, and matrix. Environ. Eng. Sci. 27 (8), 657–668. Tuppurainen, K., Halonen, I., Ruokojarvi, P., Tarhanen, J., Ruuskanen, J., 1998. Formation of PCDDs and PCDFs in municipal waste incineration and its inhibition mechanisms: a review. Chemosphere 36, 1493–1511. Tuppurainen, K., Asikainen, A., Ruokojarvi, P., Ruuskanen, J., 2003. Perspectives on the formation of polychlorinated dibenzo-p-dioxins and dibenzofurans during municipal solid waste (MSW) incineration and other combustion processes. Acc. Chem. Res. 36, 652–658. Wang, M., Liu, G., Jiang, X., Zheng, M., Yang, L., Zhao, Y., et al., 2016. Thermochemical formation of polybrominated dibenzo-p-dioxins and dibenzofurans mediated by secondary copper smelter fly ash, and implications for emission reduction. Environ. Sci. Technol. 50, 7470–7479. Yang, L., Liu, G., Zheng, M., Jin, R., Zhao, Y., Wu, X., et al., 2017a. Pivotal roles of metal oxides in the formation of environmentally persistent free radicals. Environ. Sci. Technol. 51, 12329–12336. Yang, L., Liu, G., Zheng, M., Jin, R., Zhu, Q., Zhao, Y., et al., 2017b. Atmospheric occurrence and health risks of PCDD/Fs, polychlorinated biphenyls, and polychlorinated naphthalenes by air inhalation in metallurgical plants. Sci. Total Environ. 580, 1146–1154. Zhao, Y., Zhan, J., Liu, G., Ren, Z., Zheng, M., Jin, R., et al., 2017. Field study and theoretical evidence for the profiles and underlying mechanisms of PCDD/F formation in cement kilns co-incinerating municipal solid waste and sewage sludge. Waste Manag. 61, 337–344.