Occurrence and Distribution of Synthetic Musks in Surface Sediments of Liangtan River,West China

Study area and sample collection

The Liangtan River is 88 km long, originates from the Liaojiagou Lake of Jiulongpo district, Chongqing, west China, and flows to the northeast, passing through 3 main districts of Chongqing (Jiulongpo, Shapingba, and Beibei, including 15 towns, for example, Baishiyi, Hangu, Xiyong, etc) into the Jialing River at the Maobeituo (Fig. 1). Liangtan River is the largest sub-river in the Chongqing area. As the city grows (industry and residential population increase), the water quality has deteriorated and cannot meet the standard for drinking water now. So far, there is only one sewage treatment plant (Fig. 1) with a daily capacity 25,000 m³, and the remaining sewage (about 40,000 m³) is discharged into Liangtan River directly. Most of the previous studies of the Liangtan River focused on detecting the water quality (chemical oxygen demand, biological oxygen demand, etc) and investigating pollutant sources along the river (Wei and Li, 2003; Liu et al., 2009), and there are no studies of emerging pollutants, such as SMs.

OPEN IN VIEWER

FIG. 1.

Sediment sampling locations in Liangtan River, Chongqing (•: sampling site; •: sewage treatment plant [STP]).

Sixteen sampling sites were selected along the Liangtan River. Three sites (LT1–LT3) located in the area above Baishiyi town, six sites (LT4–LT9) were along the area from downtown Baishiyi to the Xiajia Bridge. Details of the sampling sites are listed in Table 1. Surface sediment samples from Liangtan River were collected in July 2009. A grab sampler was used to bring undisturbed sediment from the river bottom to the surface, about top 10 cm thickness. The samples were kept in glass jars, transported at 4°C from the sampling point to the laboratory, where they were freeze-dried, ground, homogenized by sieving through a stainless-steel (40-mesh, pore size=0.45 mm), mixed thoroughly, and then stored in a sealed plastic bottle at +4°C until extraction. All of the surface sediment samples in this study have similar basic properties, such as porosity, particle size (fine matrix), color, and so on. So, the absorption of different sediments was not considered.

Test substances

The test substances and their physicochemical properties are given in Table 2. HHCB, AHTN, MX, MK, and Phenathrene-d10 (Phe-D10) were purchased from Dr. Ehrenstorfer GmbH (Augsburg, Germany), and Hexamethyl benzene from Aldrich (Milwaukee, WI). They are in the highest available purity (Table 2). Standard solutions of the test substances with concentrations of 10–200 and 200–1,000 μg/L were prepared in n-hexane (HEX), stored at +4°C, and renewed weekly. ENVI-C18 cartridge (a terminated silica-based bonded octadecyl cartridge, 500 mg/3 mL) was purchased from Supelco. Cellulose extraction thimbles (33×80 mm, thickness=1.5 mm) were purchased from Whatman. The organic solvents used, HEX (GC grade) and dichloromethane (DCM, GC grade), were purchased from Merck (Darmstadt, Germany). Methanol (HPLC grade) was obtained from Aldrich.

Extraction and clean-up

For extraction, an automated Soxtec Avanti 2050 from FOSS Analytical AB (Foss Tecator) was used. Approximately 0.5 g sediment (premixed with 0.5 g anhydrous sodium sulfate, which was used to dry sludge samples and spiked with 200 μL surrogate standard phenanthrene D10 of 2 mg/L) was placed in an extraction thimble. After 12 h of adsorption, the mixture was extracted with 60 mL of 1:1 (v/v) HEX/DCM. The extraction was carried out at 160°C for 1 h in the boiling extraction solvent, followed by extraction for 1 h in the rinse position. The final extracts were concentrated with a rotary evaporator to ∼2 mL, solvent-exchanged to 20 mL HEX, and, finally, reduced to ∼1 to 2 mL. The concentrated extracts were quantitatively transferred (rinsed with ∼2 mL HEX in 2–3 portions) to a 500 mL volumetric flask and filled/shaken with 400 mL DI water. The samples were then purified with solid phase extraction procedure. Before the sample loading, the cartridges were preconditioned with 5 mL of HEX, 5 mL of DCM, and 5 mL of methanol followed by 5 mL of de-ionized water (DI water). Samples were passed through the cartridges at a flow rate of 5 mL/min. After percolation, the cartridges were washed with 5 mL of 5% methanol and then dried under vacuum for about 30 min. Subsequently, SMs were eluted with 5 mL HEX, 4 mL HEX-DCM (3:1, v:v), and 3 mL DCM at a flow rate of 0.5 mL/min. The eluents were evaporated under a gentle flow of high purity nitrogen and were then dissolved to 0.9 mL with HEX and stored at 4°C until analysis. Finally, 100 μL of 1 mg/L Hexamethyl benzene was added as internal standard before gas chromatography-mass spectrometry (GC-MS).

GC-MS analyses

SMs were determined by GC-MS, using a Finnigan Voyager quadrupole MS under electron impact ionization (EI, 70 eV, 250°C) and full scan (m/z 35–400 amu, 0.08 s/scan, nominal mass resolution) or selected-ion monitoring conditions. Analytes were separated with a 30-m HP-5MS column (0.25 mm i.d. and 0.25 μm film thickness) with helium as the carrier gas (1.0 mL/min). The oven temperature program was 60°C for 5 min, 10°C/min to 250°C, 20°C/min to 280°C, and then 280°C for 5 min. The sample (1 μL) was injected into GC-MS by the splitless mode for 0.75 min and an inlet temperature of 280°C. The following masses were used for quantification and confirmation (Table 2): m/z=243, 258 for HHCB; 243, 258 for AHTN; 282, 283 for MX; and 279, 294 for MK. Concentrations of SMs were obtained by the internal standard calibration method based on a five-point calibration curve. All samples were analyzed in triplicates. The limits of detection (LODs) and limits of quantification (LOQs) were based on a signal-to-noise ratio of 5 and 10, respectively. Calculation of SMs concentrations were determined using Xcalibur software.

Quality assurance and quality control

SMs can be found in many personal products (soaps, detergents, hand lotions, etc). Therefore, it is necessary to avoid possible contamination from laboratory personnel during sample treatment and analyses (Kupper et al., 2004; Horii et al., 2007; Lee et al., 2010), and multiple sequential extractions of the same sediment sample were conducted to assure quantitative extraction. In addition, method blanks, performed with the same procedures as the pretreatment of solid matrices or aqueous samples but without samples, were used to monitor potential contamination during sample extraction, cleanup, and analysis. None of the target SMs were detected, and surrogate standard Phe-D10 did not occur in the method blanks.

The LOQs were 10 μg/kg for HHCB, AHTN, MK, and 30 μg/kg for MX, which were determined as the quantity of compounds corresponding at the ratio signal/noise 10 on the linear range of calibration curve using spiked matrix (Table 3). The values of LOQs were low enough for environmental monitoring of the target SMs. Recovery studies were performed by spiking a volume of 100 μL standard solution with a concentration of 1 mg/L into different respective matrices, including DI water (300 mL) and sediment (0.5 g pre-extracted sediment). Recoveries of SMs were 58%–88% for spiked sediment matrix, 83%–122% for DI water, depending on the analytes. The precisions of the entire method, represented by the relative standard deviation, were in a range of 3%–11%, 7%–16% in DI water and sediment, respectively. These results indicated good quality control and consistencies achieved in different batch analyses.

Occurrence of SMs in sediment samples

The concentrations of SMs in sediments are listed in Table 4. The total concentrations of SMs ranged from <LOQ to 364.48 μg/kg (dry weight, dw), with an average of 123.54 μg/kg dw. The sampling location of highest total concentration of SMs was close to Jiulongpo Hospital in Baishiyi town. Polycyclic musks were also more frequently detected in sediment samples than nitro musks. HHCB was detected in 14 sediment samples, AHTN was found in12, whereas MK was found in 3 out of 16 samples. MX was below the quantification limit in all samples. Of the four target SM compounds, concentrations of the polycyclic musk HHCB were highest, <LOQ∼268.49 μg/kg dw, followed by AHTN (0–99.75 μg/kg dw) in accord with previous studies in the literature, which supports the hypothesis that nitro musk compounds are being phased out in favor of the polycyclic musk fragrances.

Spatial distribution of SMs

The composition profiles of SMs in sediments from different sampling locations were compared and are shown in Fig. 2. Hierarchical cluster analysis was performed with SMs concentrations in sludge samples to investigate their distribution patterns. Generally, HHCB was dominant in all sediment samples, and it accounted for 30%–100% of the four SMs, followed by AHTN (16%–69%) and MK (0%–8%). These results coincided with the reported values of SMs in the United States and EU countries (Rimkus, 1999; Heim et al., 2004; Peck et al., 2006).

OPEN IN VIEWER

FIG. 2.

Distribution pattern of four synthetic musks in sediments from Liangtan River.

The distribution of SMs in sediments may reflect the relevant material production and usage patterns. In the process of this investigation, a clear relationship was found between the content of SMs in sediment samples and the proportion of sewage water in the area concerned. SM concentrations in sediments showed a certain degree of geographical distribution, and from upstream to downstream, the concentration of SMs in sediments studied increase first, then decrease gradually after downtown Baishiyi. Relatively high concentrations of SMs were found in samples (LT4-LT9) from the Baishiyi downtown region, ranging from 56.16 to 364.48 μg/kg dw. These sampling sites are located in a typical urban area (Baishiyi downtown) of Chongqing, with a higher population density. One previous study showed that domestic wastewater along the Liangtan River was either untreated or inadequately treated and directly discharged into the aquatic environment (Liu et al., 2009), and the inadequately treated and untreated municipal sewage from domestic discharges was the main source of SMs to Liangtan River, reflecting the influence of anthropogenic activities on this aquatic environment. Therefore, SMs in sediments can be used as molecular tracers to indicate the impacts of anthropogenic activities on the aquatic environment (Buerge et al., 2003; Kronimus et al., 2004; Zeng et al., 2008a; Zhang et al., 2008).

High concentrations of SMs were present in sediment collected from LT4, a location proximate to the largest hospital of Baishiyi town. According to previous researches, hospitals are known to be intensive consumers of water, thus discharging significantly higher wastewater flows than conventional households (400–1200 L/bed/d vs. 100 L/capita/d) (Gautam et al., 2007). Moreover, hospital effluents constitute a very complex water matrix, loaded with microorganisms, pharmaceuticals, heavy metals, personal care products, and radioactive elements (Suarez et al., 2009). Until we collected the sediment samples, a small amount of the hospital wastewater was discharged into the Liangtan River directly (not treated). Apparently, in addition to the main discharges from domestic wastewater, the effluent from the untreated medical wastewater is another potential source of SMs at the site of LT4. However, the major source of SMs along the Liangtan River is still domestic wastewater. Other factors that could influence the sediment concentrations are the total input of these compounds from wastewater to each site and differences in loss processes including outflow, photolysis, volatilization, and sedimentation (Peck et al., 2006).

Comparison with published data around the world

The concentration levels of four SMs found in this study and other reports were shown in Table 5. Most of the studies were in Germany, such as the Lippe river, Elbe river, Weser river, and so on.

Heim et al. (2004) analyzed SMs quantitatively based on geochemical analyses applied to a dated sediment core derived from a riparian wetland of the Lippe River. The first occurrences of HHCB and AHTN compounds were encountered at a depth of 54 cm, with concentration levels increasing toward the top. The maximum values were determined in the 14 cm layer, with 151 and 44 μg/kg dw, respectively (Heim et al., 2004).

The HHCB concentrations measured in Lake Ontario were higher than those measured in Lake Erie, but they were also similar to the lower concentrations measured in river sediments in Germany (Fromme et al., 2001; Dsikowitzky et al., 2002; Peck et al., 2006)

Levels of SMs measured in the sediments of Liangtan River are comparable with those measured in sediments collected from areas in Germany with low and moderate proportions of sewage effluents (HHCB, from less than the LOD to 520 μg/kg; AHTN, from less than the LOD to 610 μg/kg), but the results are higher than the HHCB and AHTN concentrations detected in sediments from Lake Erie and Lake Ontario (both in United States) (Peck et al., 2006).The relatively high levels of SMs in surface sediments of the Liangtan River reflected a substantial change in water quality as a result of the rapid economic growth and urbanization in the Baishiyi area during the last decade.

Sewage was discharged directly into the Liangtan River for a long-term period, which resulted in the accumulation of SMs in sediments gradually. So far, there are few studies of SMs in river sediments in China.

Six polycyclic musks were measured in surface sediments collected from the Pearl River Delta and Macao coastal region, South China, to investigate contamination from domestic sewage. The concentrations of total polycyclic musks ranged from 5.76 to 167 μg/kg dw. Generally, the concentrations of polycyclic musks followed the sequence Zhujiang River > Dongjiang River > Macao coast > Xijiang River. As expected, the sediments collected from Zhujiang River had the highest concentrations of polycyclic musks because of the large amounts of municipal sewage and industrial wastewater discharged from the city of Guangzhou (Zeng et al., 2008a).

In general, concentration of sedimentary SMs in the Liangtan River was relatively higher compared with those in other river sediments (Fromme et al., 2001; Dsikowitzky et al., 2002; Heim et al., 2004; Peck et al., 2006; Zeng et al., 2008a, 2008b; Zhang et al., 2008). It has become potential pollution source in the overlying water. Sewage treatment sectors and environmental protection agencies should pay more attention to this problem.

Implications of the HHCB to AHTN ratios

The ratios of HHCB/AHTN in sediments could be used to trace fragrance composition in the personal care products (Moldovan, 2006; Zhang et al., 2008). The ratios of HHCB/AHTN in the sediment samples collected from the Liangtan River ranged between 0.43 and 5.25, with an average of 1.83, except for four sediment samples (LT1, LT2, LT2, and LT16) with the concentrations of AHTN lower than LOQ (Table 4). Similar results, ratios of HHCB/AHTN, were reported from other studies, such as <LOQ-3.1 in Lippe River (Germany) (Dsikowitzky et al., 2002), 0.72–4.33 in the Pearl River Delta (PRD, China) (Zeng et al., 2008a), and 2.2–4.7 in the sediments of Suzhou Creek (China) (Zeng et al., 2008b). However, a higher ratio, up to 16.7, was found in the sediments of Ontario Lake (Peck et al., 2006). Therefore, the ratios of HHCB to AHTN among different countries are completely different. Differences recorded for ratios of HHCB/AHTN could probably relate to product usage, effectiveness of sewage treatment plant processes to remove the compounds, and selective losses through different environmental processes (e.g., volatilization, degradation, and sorption).

The sediments collected from locations near urban sources (e.g., LT 4, LT5, and LT6) generally showed relatively high concentrations of polycyclic musks and ratios of HHCB/AHTN, whereas the samples collected from sites far from sources (e.g., LT1, LT2, LT3, and LT 16) displayed lower values for both concentrations and ratios.