Spatial and temporal trends of polyhalogenated carbazoles in sediments of the Yangtze river: insights into the origin

Polyhalogenated carbazoles (PHCZs) have attracted increasing attention due to their mobility, toxicity, and semi-volatile nature that have raised environmental concerns globally (Riddell et al. 2015). Several recent studies (Wu et al. 2017b; Ji et al. 2019; Yue et al. 2020) have revealed that PHCZs exert bioaccumulation, biomagnification, and dioxin-like effects. As early as in the 1980s, the PHCZ 1,3,6,8-tetrachlorocarbazole was first discovered by Kuehl et al. (1984) in sediments from the Buffalo River of the USA, which suggested the existence of PHCZs in the aquatic environment. In recent decades, various studies have indicated the presence of PHCZs in sediments, soil, and air in different areas worldwide (Kronimus et al. 2004; Altarawneh & Dlugogorski 2015; Wu et al. 2016; Zhou et al. 2021). For example, 1,3,6,8-tetrabromocarbazole (1368-BCZ) and other halogenated carbazoles (including ClBr₃, ClBr₄, ClBr₃I, Br₄I, and Br₃I2-carbazole) have been detected in aquatic sediments from the Lippe River of Germany (Zhu et al. 2019), the North Sea Estuary of Germany (Zhou et al. 2021), Lake Michigan (Guo et al. 2014), the Saginaw River of the USA (Zhou et al. 2021), Ontario of Canada (Kronimus et al. 2004; Guo et al. 2014), Lake Tai of China (Eljarrat et al. 2005), the northern South China Sea (Zhou et al. 2021), and the Jiulong River of China (Zhou et al. 2019). The potential sources of these PHCZ congeners are not thoroughly documented (Pena-Abaurrea et al. 2014). Halogenated indigo dyes, optoelectronic materials, and p-chloroaniline structure herbicides have been suggested as the anthropogenic sources of some PHCZs. For example, 1368-BCZ may form as an impurity in the manufacturing process of indigo dyes (Mumbo et al. 2015). In addition, some PHCZs may be produced from natural sources; for example, the marine fungus Caldariomyces fumago can synthesize BCZ in water with the help of enzymes (Kronimus et al. 2004; Xu et al. 2017). The discharge of saline wastewater containing PHCZs (e.g., brine) degrades water quality and thus water cannot be directly used for potable water (via desalination) and industrial applications (Ji et al. 2019). However, studies on the sources of PHCZs and the effect of the input volume of human activities on PHCZs have been limited yet. For instance, PHCZs were detected in unpolluted soil from Germany and the sediment cores formed before 1900. Chen et al. (2018) suggested that PHCZs are widespread in the depositional environment. The sediments have long been considered reservoirs of various organic pollutants. Although more and more studies have been conducted on the environmental sources of PHCZs, knowledge of their historical contamination in the aquatic environment and evaluation of toxic equivalency to organisms remains inadequate (Guo et al. 2017; Ji et al. 2019). Therefore, PHCZs' traceability and migration behavior in environmental media must be further investigated. Better understanding of aquatic systems is necessary to determine the occurrence and distribution characteristics of PHCZs as well as their correlations with anthropogenic factors in a region.

The Yangtze River (YR) with a length of 6,387 km runs across central China from west to east. It flows through eight provincial regions in China and into the east China Sea to the east of Chongming Island. Most cities on both sides of the YR are China's industrial centers and crucial economic driving forces. Therefore, the YR is polluted by the industrial development of coastal cities (Da et al. 2020). Because of human activities, it is one of the most affected rivers in the world. The present study targets the section of the YR extending from the Anhui Province to Chongming Island; the districts in this section represent regions with the highest degree of industrialization and urbanization in China (Sun et al. 2016). To our knowledge, the presence of PHCZs in sediments of the YR has not been reported yet. This study aims to (1) study both the residual level and composition profiles of PHCZs in sediments of the YR, (2) analyze the historical variation tendency of PHCZs and anthropogenic factors in the sediment cores, and (3) determine the toxic effects of PHCZs in sediments.

Based on the previous studies (Wu et al. 2017a; Zhou et al. 2021), in this study, we selected 10 common PHCZs in the environment and purchased their standard mixtures from Cambridge Isotope Laboratories, Inc., USA. The selected PHCZs are as follows: 3-monochlorocarbazole (3-CCZ); 1,3,6,8-tetrabromocarbazole (1368-BCZ); 3,6-dichlorocarbazole (36-CCZ); 3,6-dibromocarbazole (36-BCZ); 2,3,6,7-tetrachlorocarbazole (2367-CCZ); 1,3,6-tribromocarbazole (136-BCZ); 1-bromo-3,6-dichlorocarbazole (1-B-36-CCZ); 1,3,6,8-tetrachlorocarbazole (1368-CCZ); 1,8-dibromo-3,6-dichlorocarbazole (18-B-36-CCZ); and 3-monobromocarbazole (3-BCZ). The chemical structures of the investigated PHCZ congeners are shown in Supplementary Figure S1. The toxicological information on PHCZs is listed in Supplementary Table S1. The surrogate standards (F-BDE69, F-BDE160, and PCB-204) and internal standards (F-BDE154 and DCDE) were purchased from Wellington Laboratories (Guelph, ON, Canada). Silica gel (100–200 mesh), sodium sulfate (Na₂SO₄), and neutral alumina (Al₂O₃) were purchased from Sigma-Aldrich (St. Louis, MO, USA), and they were baked at 400 °C for 16 h and stored at 200 °C before use.

The counting statistical errors were higher than ±2.9%. The layers corresponded to years 1959–2017 spanning 58 ± 3.42 years in C1, 1962–2018 spanning 56 ± 3.40 years in C2, and 1959–2017 spanning 58 ± 3.39 years in C3. The average deposition rates for C1, C2, and C3 were 0.75, 0.80, and 0.78 cm y⁻¹, respectively.

Referring to a previous study (Wu et al. 2016), analytical methods were used in the present study. The sediment samples were dehydrated through freeze drying at −200 °C for approximately 40 h in a freeze dryer (Jinteng, Tianjin, China), pulverized to <200 mesh, and then homogenized. To 8 g each of the homogenized sediment samples, 40 ng of the surrogate standard (F-BDE69) was added, and accelerated solvent extraction (Thermo Fisher Scientific, Inc.) was conducted with hexane and acetone mixture (1:1, v/v). A moderate amount of activated copper powder was used to remove sulfur in the concentrated sediment extract. After sulfur removal, the extract was characterized through high-performance liquid chromatography with a gel permeation chromatography column (250 mm × 18.9 mm, 4 um). The extraction solution was heated to 120 °C in four cycles, with heating for 10 min and then allowing the solution to stand for 10 min. The extracts were concentrated with a rotary evaporator and then dried and solvent-exchanged into l mL of hexane. A burette with a scale was filled from the top to the bottom with anhydrous Na₂SO₄ (length = 1.0 cm), alumina (length = 8.0 cm), silica gel (length = 4.0 cm), and anhydrous Na₂SO₄ (length = 4.0 cm). The hexane extract was added to the burette (baked at 240 °C before use) and eluted with 50 mL of hexane. It was then washed with a 100 mL mixture of hexane and dichloromethane (v/v = 4:1) and 100 mL of dichloromethane. The eluent was concentrated and placed in a 200 μL brown glass bottle for the subsequent analysis.

PHCZ concentrations were determined using an Agilent 1290 gas chromatography system coupled with an Agilent 5973 MS in the electron impact ionization mode. The target substances were analyzed on an HP-5MS column (100 mm i.d., 2.1 mm) and an injector protected in the pulsed-splitless mode at 250 °C. The starting temperature of the oven was kept at 30 °C for 5 min, which was increased to 100 °C at 5 °C min⁻¹, and then increased to 200 °C at 10 °C min⁻¹ and maintained for 12 min. Three PHCZs, namely 3-CCZ, 3-BCZ, and 36-CCZ, were measured in the splitless mode, whether the other PHCZs were detected in the negative ion mode. Supplementary Figure S3 shows a chromatogram of the target PHCZs by using the selected SIM ions. The qualitative and quantitative ionic characteristics of each target compound are presented in Supplementary Table S2. According to the description in previous studies (Wu et al. 2016; Zhou et al. 2019), the PHCZs were identified and quantified based on the sample peaks, retention times, peak areas, and signal-to-noise ratios.

The total organic carbon (TOC) concentration in the sediments was determined through combustion oxidation. For this analysis, 0.1 g of the sediment samples was weighed and placed in a beaker, to which phosphoric acid solution was added until all bubbles in the beaker disappeared. The samples were then placed in an EL Cube elemental analyzer (Elementar Analysensysteme GmbH, Langenselbold, Germany) to measure the values.

The particle sizes of the sediments were measured by sieving on meshes of different sizes and then categorized as follows: >0.3 mm, 0.1–0.3 mm, and <0.1 mm. The percentages of finer particles (particles size <0.1 mm) were calculated.

To avoid the interference of pollutants, all glassware were pre-cleaned with ultrapure water and hexane, followed by roasting at 180 °C for 10 h before use. A procedural blank sample was analyzed after every five samples. No studied compounds were found in the blank samples. The surrogate recovery rates were 91.1 ± 2.3%, 89.5 ± 2.1%, and 94.3 ± 5.2% for F-BDE69, F-BDE160, and PCB-204, respectively. The mixed standards with individual PHCZs (10 ng g⁻¹) were injected into the samples without analytes. The limits of detection (LOD) were set as a signal-to-noise ratio of 2. The relative standard deviation (RSD), LOD, and recoveries (R) of individual PHCZ congeners are shown in Supplementary Table S3.

SPSS AMOS 21.0 was used for the statistical analysis. The principal component analysis (PCA) was performed with Kaiser normalization and a factor adjustment of 2. T-test was performed to determine significant differences in the level of PHCZs among different samples.

Table 1 shows 10 target PHCZ congeners in the surface sediments from the YR. The detection rates of individual PHCZs ranged from 60 to 100% in the surface sediments, reflecting the ubiquitous occurrence of PHCZs in different regions of the YR. All 10 target PHCZ congeners were detected in all samples. The detection rates for 3-CCZ, 36-CCZ, 136-BCZ, 1368-CCZ, and 1368-BCZ in all the samples were up to 100%, whereas that for 2367-CCZ was up to 93.33%, indicating that these compounds were the predominant contaminants in the surface sediments. The detection rate for 3-BCZ was much less than those for the other compounds, possibly because 3-BCZ was not stable and tended to be further halogenated into other PHCZ congeners (Zhou et al. 2021). Moreover, the total concentration of PHCZs ranged from 1.58 to 5.45 ng g⁻¹, with a mean value of 3.11 ng g⁻¹. The comparison of PHCZ distribution in various regions of the world (Supplementary Table S4) indicates that the mean concentrations (i.e., 3.11 ng g⁻¹) of PHCZs in the YR were higher than those in the surface sediments sampled from the northern South China Sea and Lake Tai of China but lower than those in the surface sediments sampled from the Jiaozhou Bay wetland of China, Jiulong Estuary of China, Saginaw River system of the USA, North Sea Estuary of Germany, Great Lakes of the USA, and San Francisco Bay of the USA. The occurrences of compounds in various regions suggest wide distributions of the compounds in global aquatic ecosystems.

Compared with other persistent organic pollutants in the YR, the mean levels of PHCZs were lower than those of polybrominated diphenyl ethers (PBDEs; 378 ng g⁻¹) (Da et al. 2020), organochlorine pesticides (OCPs; 45.8 ng g⁻¹) (Gong et al. 2020), and polycyclic aromatic hydrocarbons (PAHs; 4,002 ng g⁻¹) (Li et al. 2020) but higher than that of polychlorinated biphenyls (PCBs; 1.98 ng g⁻¹) (Guo et al. 2021).

In addition, Kuehl et al. (1984) discovered an elevated concentration of 1368-BCZ near the outfall of a former dye manufacturer. 136-BCZ was reported to be an intermediate product of photoelectric materials (Parette et al. 2015). Therefore, this study deduced that brominated carbazoles emerged from anthropogenic sources in the study area. This composition profile differs from that obtained from the investigation of sediments in Lake Tai, where 36-CCZ and 36-BCZ were the major congeners, while 1368-BCZ had relatively less contribution. Of the PHCZs in Lake Tai sediments, the contribution of 36-CCZ and 36-BCZ was 97.31%, whereas 3-BCZ and 1368-BCZ accounted for up to 1.91 and 1.92%, respectively. In addition, 36-CCZ was found to be predominant in the Saginaw Bay River and the coastal North Sea (Zhou et al. 2021). The dominance of 36-CCZ in various aquatic systems may be related to the electrophilic aromatic substitution mode, which is conducive to halogen substitution on carbazole relative to the ortho and para positions of nitrogen (Zhou et al. 2021).

To study the sedimentary characteristics of PHCZs, the variation trend of vertical concentrations of PHCZs during different years was analyzed in the three sediment cores from the YR. As shown in Table 2, the detection rates of individual PHCZs were 43.18–95.45% in C1, 46.34–88.80% in C2, and 42.50–90.00% in C3. The three sediment cores displayed high detection rates, which indicated the widespread use of PHCZs historically. Overall, all 10 target PHCZ congeners were detected in all the samples. 36-CCZ, 36-BCZ, and 1368-BCZ dominated the composition profile in the three cores and were also observed in the surface sediments collected from the YR. Moreover, the residual levels of ∑PHCZs were 0.01–3.07 ng g⁻¹ (average: 1.06 ng g⁻¹) for C1, 0.01–2.91 ng g⁻¹ (average: 0.83 ng g⁻¹) for C2, and 0.01–2.61 ng g⁻¹ (average: 0.84 ng g⁻¹) for C3. The mean and detection rates of the individual PHCZs in C1 were higher than those in C2 and C3. The difference in the residual levels of ∑PHCZs in the three cores was mainly attributed to the different geographic locations and surrounding human activities, which also account for their high concentration in surface sediments of site S3; the dye factory was approximately 10 km from C1, leading to the residual levels of ∑PHCZs in C1.

Supplementary Figure S4 indicates that slightly different from the congener composition profiles found in the surface sediments, the predominant PHCZs of the three cores in various sedimentary times were 36-CCZ, 36-BCZ, and 1368-BCZ, accounting for more than 50% of the total PHCZs. The ratios of the three compounds were 57–100%, 55–100%, and 31–100% in C1, C2, and C3, respectively. PHCZ concentrations in these three cores were less than those found in Lake Michigan and the Upper Great Lakes but greater than those found in the northern South China Sea (Guo et al. 2017; Zhou et al. 2021). Parette et al. (2015) reported that 1368-BCZ possibly originated from tetrabromodiiodine produced by the halogenated indigo dye factory. Additionally, in this study, 1368-BCZ was predominant among the 10 analyzed PHCZs in the three cores, suggesting that anthropogenic sources of PHCZs affected the historical compositions in the cores. The environmental source of 36-CCZ and 36-BCZ is not completely known (Zhu et al. 2019). According to some studies, they were formed from monuron or monolinuron, an intermediate product of optoelectronic materials and polymers that is biodegradable in nature (Mai et al. 2005; Mumbo et al. 2013). For example, PHCZs were believed to have originated from the natural biodegradation in the sediments from Lake Michigan and the northern South China Sea. Therefore, the relatively high concentrations of 36-CCZ and 36-BCZ suggest that the PHCZs in the study area originated likely from a natural source.

As shown in Supplementary Table S5, PCA was applied to show the rotated component matrix of anthropogenic parameters. The result indicated that 96.04% of the data variation could be explained by the first six components. The principal component factor of the total variance was distinguished by the high positive loadings of total population, urbanization, GDP, sewage, per capital GDP, the ratio of heavy industry, ratio of light industry, and industrial wastewater emission. The presence of these anthropogenic parameters suggests a strong correlation with PHCZs.

As shown in Supplementary Table S6, the evaluated TEQ values of PHCZs in surface sediments from the YR ranged from 25.6 × 10⁻⁵ to 107.2 × 10⁻⁵ ng g⁻¹ (mean = 60.57 × 10⁻⁵ ng g⁻¹), which was lower than the safe sediment value of 0.020 ng g⁻¹ (Zhu et al. 2019). The comparatively high TEQ values of PHCZs were detected in the sediments at sites S3 and S5, where the high residue levels of 1368-CCZ were observed. Comparison with the other regions indicates that the TEQ values of PHCZs in sediments from the YR were higher than those from Lake Tai of China (0–140 × 10⁻⁵ ng g⁻¹, mean = 10 × 10⁻⁵ ng g⁻¹) (Wu et al. 2017a), Jiaozhou Bay wetland of China (10 × 10⁻⁵–390 × 10⁻⁵ ng g⁻¹, mean = 110 × 10⁻⁵ ng g⁻¹) (Zhu et al. 2019), and San Francisco Bay of the USA (20 × 10⁻⁵–180 × 10⁻⁵ ng g⁻¹, mean = 120 × 10⁻⁵ ng g⁻¹) (Wu et al. 2017b). Although the TEQ values of PHCZs in sediments from the YR were relatively low, synergistic, or superposition effects caused by the coexistence of PHCZs and other persistent toxic pollutants such as dioxin PCB, PBDEs, and PAHs cannot be ruled out (Yan et al. 2017). Therefore, the existence of PHCZs in sediments and their harmful effects on aquatic organisms necessitate further research in the field.

This study revealed the spatial and temporal trends of the distribution of PHCZs in the cores and surface sediments obtained from the YR, China. The analysis results indicated a ubiquitous distribution of PHCZs in the sediments from the YR, and their residue levels were comparable to those of the other prominent organic pollutants such as OCPs, PCBs, and PBDEs. The compositions and distribution profiles suggested that a combination of natural and anthropogenic sources contributed possibly to the sediments of the YR. PHCZs and sediment physicochemical properties (including TOC and finer particles) were found to have a strong correlation. Moreover, human activities were found to directly or indirectly affect PHCZ concentrations. The evaluation of the toxicity of PHCZs suggested that the toxic equivalent of PHCZs was less than that of the safe sediment. The applicability of the findings will understand the pollution of YR aquatic systems. The harmful effects of PHCZs on aquatic organisms must be investigated in future research.

This work was supported by Anhui Provincial Natural Science Foundation (2008085MD119), Hefei Municipal Natural Science Foundation (2021011), the Project Supported by Anhui Postdoctoral Fund (2019b332), and the Key Project of Anhui University Scientific Research Project (KJ2019A0826).

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.