Distribution and Source of Polycyclic Aromatic Hydrocarbons (PAHs) in Water Dissolved Phase, Suspended Particulate Matter and Sediment from Weihe River in Northwest China

Weihe River is a typical river located in the arid and semi-arid regions of Northwest China. In this study, the distribution and sources of 16 polycyclic aromatic hydrocarbons (PAHs) in Weihe River were investigated. The concentrations of ∑PAHs ranged from 351 to 4427 ng/L with a mean value of 835.4 ng/L in water dissolved phase (WDP), from 3557 ng/L to 147,907 ng/L with a mean value of 20,780 ng /L in suspended particulate matter (SPM), and from 362 to 15,667 ng/g dry weight (dw) with a mean value of 2000 ng/g dw in sediment, respectively. The concentrations of PAHs in Weihe River were higher compared with other rivers in the world. In both WDP and sediment, the highest concentrations of ∑PAHs were observed in the middle reach, while the lowest concentrations of ∑PAHs were found in the lower reach. For SPM, however, the PAHs concentrations in the lower reach were highest and the PAHs concentrations in the upper reach were lowest. The ratios of anthracene/(anthracene + phenanthrene) and fluoranthene/(fluoranthene + pyrene) reflected a pattern of both pyrolytic and petrogenic input of PAHs in Weihe River. The potential ecosystem risk assessment indicated that harmful biological impairments occur frequently in Weihe River.


Introduction
Polycyclic aromatic hydrocarbons (PAHs) are a class of diverse organic compounds containing two or more fused aromatic rings of carbon and hydrogen atoms. PAHs are ubiquitous contaminants in the environment. These compounds are generally generated by natural and anthropogenic processes and can be introduced into the environment through various routes. PAHs are of environmental concern due to their toxic, mutagenic and carcinogenic potential [1]. Due to their environmental concern, 16 PAHs were designated as priority pollutants by the United States Environmental protection Agency (US EPA) and seven PAHs were designated as potentially carcinogenic pollutants by the US EPA. For these reasons, PAHs' behavior, transport, fate and environmental risk to ecological systems have become an advanced research hotspot in environmental fields all over the world [2][3][4][5][6][7][8][9].
A lot of papers have reported on PAHs in rivers in China [3,4,[6][7][8][9][10][11][12][13]. However, almost all of these rivers were located in the humid and semi-humid regions in North, East, South or Northeast China. There are also many rivers located in the arid and semi-arid region of Northwest China, but no study about PAHs in these rivers has been published. Weihe River is a typical river located in the arid and semi-arid region in Northwest China. This basin is the most developed area in Northwest China and is under rapid industrialization and urbanization. China is now planning to carry out its "One Belt and One Road" program. The Weihe River basin is in an important position in the Silk Road economic belt. Xi'an City, the national cooperation platform of "One Belt and One Road", locates in the Weihe River basin. Only the car ownership of Xi'an City is more than 1,300,000. More than 2.9 million people were living in the Weihe River basin in 2007. A lot of energy and chemical enterprises are located in this area. A large volume of raw sewage drains off into Weihe River or its branches, which inevitably results in bad water quality. With the rapid increase in population and economic wealth, the insufficient water supply and degradation of water environment have become the most restrictive factors for the basin's economic development. The immense pressure on the area's ecosystem makes the study of Weihe River necessary to safeguard its ecosystem health, so it is very important to study the characterization and distribution of PAHs in Weihe River.
The objectives of the present work were to determine the levels and spatial distribution of PAHs in Weihe River, and to identify the major sources of PAHs in this area. The results would be used to provide data for comparison with other rivers and to assess potential ecotoxicological effect.  The PAHs were analyzed using an Agilent 6890A gas chromatography−5975C ion trap mass spectrometry (GC/MS) system equipped with a DB-( EUPAH 20 m × 0.18 mm × 0.14 μm) capillary column (Agilent Technologies, Palo Alto, CA, USA). The carrier gas was high-purity helium (He). The injection port temperature was maintained at 270 °C. The GC-MS temperature program started at 60 °C for 1 min and then increased at a rate of 11 °C per minute to 270 °C. The temperature was then increased to 300 °C with a rate of 1.5 °C per minute and held at 300 °C for 2 min. The analysis was performed using the selected ion monitoring (SIM) mode with a splitless injection. The determination and quantification of 16 PAHs in the samples was achieved based on matching the compound ionization and retention time with those of the standard mixture of PAHs. The RSDs of retention time and peak area for the injection of 16 PAHs were all below than 0.5% and 2%, respectively.

Study Area and Sample Collection
The sampling sites are illustrated in Figure 1. A total of 37 sampling sites along Weihe River and its tributaries were selected. Among the 37 sampling sites, 16 sites locate in its 16 main tributaries and 21 sites locate in the main stream of Weihe River. Fourteen, fourteen and nine sites were located in the upper (sites 1-14), middle (sites 15-28) and lower (sites 29-37) reaches of Weihe River, respectively. A total of 73 samples, including 37 water samples and 36 sediment samples were collected from the 29-31 August, 2014. On 31 August, the Weihe river flow at site 35 was 245 m 3 /s. The suspended particulate matter (SPM) samples are derived from water dissolved phase (WDP) in the laboratory. Water samples were collected from 0.5 m below the water surface using 2.5 L glass jars. The surface sediment samples were collected using grab sampler. During the whole sampling process global position system (GPS) was used to locate the sampling stations. In each sampling point, 2.5 L of water (one amber bottle) were collected and transported to the laboratory. The separation of WDP and SPM was achieved by filtration according to the method of Montuori and Triassi [5]. Water samples were filtered through a previously kiln-fired (400 °C overnight) GF/F glass fiber filter (47 mm × 0.7 μm; Whatman, Maidstone, UK). Filters (SPM) were kept in the dark at −20 °C until analysis. Dissolved phase refers to the fraction of contaminants passing through the filter. This includes the compounds that are both truly dissolved as well as those associated with colloidal organic matter. These filtrates were kept in the dark at 4 °C and extracted within 24 h. Surface sediment (0-20 cm) samples were collected by using a grab sampler (Van Veen Bodemhappe 2 L, Kiel, Germany) and put in aluminum containers. An adequate quantity of surface sediment was collected in each sampling point. All samples were transported to the laboratory directly after sampling and kept at 4 °C before analysis.

Water Dissolved Phase
Solid phase extraction (SPE) cartridges system from Supelco (Sigma-Aldrich Corp., Saint Louis, MO, USA) was used to extract water dissolved phase samples according to the method of Chen et al. [9]. Before the extraction, the C18-bonded phase containing 500 mg of reversed phase octadecyl (Supelclean ENVI-18, Sigma-Aldrich Corp.) was first washed with 5 mL of dichloromethane, 5 mL methanol and 5 mL ultra-pure water, respectively. About 50 mL methanol was added to the water sample (1 L) in order to improve the recovery. Then the solution was percolated through the cartridges with a flow rate of 3 mL·min −1 under vacuum pump. After extraction, the PAHs trapped were eluted to a glass tube by 5 mL dichloromethane. Two g anhydrous Na2SO4 was used to remove trace amount of water. The solvent fractions were then evaporated on a rotary evaporator, and exchanged by hexane to a final volume of 1 mL.

Suspended Particulate Phase
The SPM content was determined by gravimetry, after drying the glass fiber filter in an air-heated oven (55 °C until constant weight) and equilibrated at room temperature in a desiccator. Suspended particulate phase containing glass fiber filters were cut into pieces and spiked with the surrogate (10 ng of anthracene-d10, pyrene-d10 and perylene-d12), and then extracted three times by ultrasonic-assisted solvent extraction in 20 mL of dichloromethane for 1 h followed by centrifugation. Then 10 mL of supernatant was filtered through a silica gel column (3 g) with 11 mL 1:1 (v/v) elution of hexane and dichloromethane. Anhydrous Na2SO4 (0.5 g) was used to remove moisture. The solvent fractions were then evaporated on a rotary evaporator, and exchanged by hexane with a final volume of 0.5 mL.

Sediment
The PAHs in sediment were extracted by ultrasonic-assisted solvent extraction according to the method by Chen et al. [9]. The dry sediments were carefully collected, homogenized and passed through 250 μm standard sieve. Sample preparation included homogeneous mixing of 2 g of sediment sample with 0.5 g anhydrous Na2SO4 to remove moisture and ultrasonication in 20 mL of dichloromethane for 1 h followed by centrifugation. Then 10 mL of supernatant was filtered through 3 g of silica gel column with 11 mL 1:1 (v/v) elution of hexane and dichloromethane. The solvent fractions were then evaporated on a rotary evaporator, and exchanged with hexane with a final volume of 1 mL.

Parameters Determination
The total organic carbon (TOC) of sediment was determined by TOC analyzer (TOC-VCPH; Shimadzu Corp., Shimadzu, Japan). Chemical oxygen demands (COD) of water samples were determined by COD determinator (Shanghai; China). Electric conductivity (EC); total dissolved solids (TDS) and salinity were determined by a HQ30D water quality analyzer (Shanghai; China). The pH value was determined by a pH meter (Mettler Toledo, Columbus, OH, USA).

Quality Assurance
For every set of samples, a procedural blank and spike sample consisting of all reagents was run to check for interference and cross contamination. The method detection limits (MDLs) of PAHs were calculated as three times the standard deviation of the PAHs level in procedural blanks. A strict regime of quality control was employed before the onset of the sampling and analysis program. The surrogate averaged recoveries in the dissolved phase were 90.2% ± 3.9% for anthracene-d10, 94.6% ± 5.9% for pyrene-d10 and 95.1% ± 6.6% for perylene-d12. In the SPM samples, recoveries were 85.4% ± 6.4% for anthracene-d10, 92.1% ± 5.6% for pyrene-d10 and 97.8% ± 6.1% for perylene-d12. Finally, in the sediment samples the averaged recoveries were the following: 87.4% ± 6.8% for anthracene-d10, 92.2% ± 7.6% for pyrene-d10 and 101% ± 6.4% for perylene-d12.

PAHs Concentrations in WDP
The concentration of 16 PAHs, summed as ∑PAHs, determined in WDP, are shown in Table 1. The concentrations of ∑PAHs ranged from 351 to 4427 ng/L with a mean value of 835 ng/L. Samples with measured low concentration for one compound may have high concentrations for other compounds. The average concentrations were 189, 140, 70.1, 89.9 and 118 ng/L for 2-6-ring PAHs, respectively. The compositional profiles of PAH in the dissolved phase are illustrated in Figure 2, which indicates that low molecular weight PAHs (2-and 3-ring PAHs) were abundant in samples, representing on average over 54% of all PAHs. For individual PAH, naphthalene and phenanthrene were dominant in water (Table 1). Their average concentrations were 189 and 57.8 ng/L, accounting for 31.1% and 9.5% of the average concentrations of ∑PAHs, respectively. In addition, the total concentration of potentially carcinogenic PAHs (∑PAH7) (BaA, CHR, BbF, BkF, BaP, IcP, DhA) ranged from 58.1 ng/L to 1344 ng/L with a mean value of 192 ng/L, accounting for 31.6% of ∑PAHs. The concentration of BaP was in the range of 5.02 to 240 ng/L with the mean value of 36.9 ng/L. The concentration of BaP in each water dissolved phase samples from the 37 sites was much higher than 2.8 ng/L (Environmental Quality Standard for surface Water of China, GB 3838-2002), suggesting water from Weihe River has a high health risk for drinking.  Compared with other polluted rivers in the world (Table 2), the mean concentrations of ∑PAHs in WDP from Weihe River were much higher than those found in Jinjiang River in China (53.2 ng/L) by Yang et al. [11], in the Susquehanna River in USA (67 ng/L) by Ko et al. [14], in the Yellow River Delta in China (121 ng/L) by Wang et al. [13] and in Henan Reach of Yellow River in China (662 ng/L) by Sun et al. [12]. It was even higher than those found in Sarno River (739 ng/L) (Italy) by Montuori and Triassi [5], which was defined as "the most polluted river in Europe". It was obvious that PAHs concentrations in WDP from Weihe River were at a very high level compared with other rivers in the world.

PAHs Concentrations in SPM
As shown in Table 1 Compared with other polluted rivers in the world (Table 2), the ∑PAHs in SPM from Weihe River were much higher than most of other rivers. It was close to the Daliao River watershed (China) [10]. The level of PAHs in SPM of Weihe River was in the highest range. Table 1 illustrates the concentrations of PAHs in sediment samples in the study river. The concentrations of PAHs ranged from 362 to 15,667 ng/g dry weight (dw) with a mean value of 2000 ng/g. In terms of individual PAH composition, all of the 16 PAHs were detected at all sampling sites. The detected concentrations were 98.2-1605 ng/g dw for 2-ring PAHs, 201-5094 ng/g dw for 3-ring PAHs, 45.1-5137 ng/g dw for 4-ring PAHs, 4.92-1645 ng/g dw for 5-ring PAHs, and 16.1-2187 ng/g dw for 6-ring PAHs, respectively ( Table 2). Two-six ring PAHs accounted for 18.0%, 45.9%, 20.3%, 6.4% and 9.4% of the concentration of PAHs, respectively ( Figure 2). The concentration of potentially carcinogenic PAHs (∑PAH7) ranged from 34.4 to 5490 ng/g dw with the mean concentration of 406 ng/g dw, accounting for 20.2% of the total PAHs concentration. The concentration of BaP ranged from 6.80 to 1051 ng/g dw with a mean value of 85.7 ng/g dw.

PAHs Concentrations in Sediment
Compared with other polluted rivers in the world (Table 2), the concentrations of ∑PAHs in sediment from Weihe River were much higher than those found in Guan River Estuary (China) by He et al. [17], Qiantang River (China) by Chen et al. [9] and Malacca River (Malaysia) by Keshavarzifard et al. [6]. Values higher than those found in Weihe River were found in Tianjing River (China) by Shi et al. [4], in Pearl River (China) by Mai et al. [18] and in Prai River (Malaysia) by Keshavarzifard et al. [6]. The concentrations of PAHs in sediment from Weihe River were in moderate range.
PAHs would rapidly become associated with SPMs and then aggregated in sediment after them into the aquatic environment. Therefore, relatively high concentration of PAHs in water and SPMs but moderate in sediment indicated that the contamination of PAHs in Weihe River might be caused by fresh input of PAHs.

The Spatial Distribution of PAHs
The concentrations of ∑PAHs in WDPs, SPMs and sediments in the 37 sampling sites from Weihe River were illustrated in Figure 2. To WDP samples, the highest ∑PAHs were found in sites 22-28. The possible reason was that these sites were located in the mainstream and its branches of Weihe River in Xi'an district. Xi'an city is the most developed area in Weihe River catchment and a large quantity of PAHs entered the river in various ways. Especially the ∑PAHs of sample from site 24 was the highest among all the WDP samples. The possible cause was that site 24 located in Zaohe River, a branch of Weihe River, through which most of sewage from Xi'an area was discharged. The highest ∑PAHs in sediment samples was also found in this site.
The spatial distribution of PAHs in WDP, SPM and sediments were studied by comparing the concentrations of 2-6 rings PAHs among upper (sites [1][2][3][4][5][6][7][8][9][10][11][12][13][14], middle (sites [15][16][17][18][19][20][21][22][23][24][25][26][27][28] and lower (sites 29-37) reaches. The results were summarized in Figure 3. It could be observed from Figure 3a that both in WDP and sediment, the highest concentrations of ∑PAHs were observed in middle reach, while lowest concentrations of ∑PAHs were found in lower reach. For SPM, however, the highest concentrations were found in the lower reach and lowest concentrations were found in the upper reach. These results may be due to the following reasons. First, large quantities of waste water contained PAHs was discharged into middle reach of Weihe River, which caused the ∑PAHs in WDP were highest in middle reach. Second, the water in middle reach was flat and the SPM adsorbed PAHs from WDP, then sink to the bottom of the river, which caused the ∑PAHs in sediment increased. Third, the ∑PAHs in SPM were closely related to SPM content in water. The SPM contents in lower reach of Weihe River were very high ( Table 3). The highest SPM contents were found in site 31, site 33 and site 34, which were high even to 24.4 g dw/L, 17.7 g dw/L and 31.6 g dw/L, respectively. The ∑PAHs in SPM from these three sites were also highest, which were 75,936 ng/L, 70,056 ng/L and 147,907 ng/L, respectively ( Table 3). The lower reaches of Weihe River were characterized by torrential water flow with the high silt content, which caused that the PAHs in sediment were desorbed into water, and then be absorbed by SPM. Thus, ∑PAHs in sediments from lower reaches were even lower than those from upper reaches.

The Distribution of PAHs between WDP, SPM and Sediment
Many researchers and environmentalists considered that the environmental fate and behavior of hydrophobic organic compounds such as PAHs is ultimately determined by their physicochemical properties, such as organic content, size distribution, partition coefficient and salinity [19,20]. Physiochemical parameters such as COD (7.14-1056 mg/L), EC (546-1591 us/cm), TDS (154-1370 mg/L), salinity (0.27-0.63 ppt) and pH (7.46-9.96) in water samples from Weihe River were detected, but no significant correlations were observed between concentrations of PAH and these physiochemical parameters. The TOC of sediments in Weihe River were also detected, which ranged from 0.22% to 3.2% (Table 3). Although some papers reported positive linear relations between PAH concentrations and the total organic carbon contents (TOC) in sediment [21,22], no linear relation have been found between each individual PAH concentrations and TOC in Weihe River's sediment.
The sediments and WDP system, and the SPM and WDP system in Weihe River are undergoing dynamic sorption and desorption and possibly have not reached chemical equilibrium, but an evaluation of the distribution of PAHs between sediments and WDP, and the SPM and WDP, can still provide useful information to understand the transport and fates of PAHs. In order to reach this purpose, the apparent distribution coefficient (Kd) of total PAHs, which is defined as the ratio of total PAH in sediment or SPM to that in WDP, was calculated (Table 3). No correlations existed between Kd (sediment) and sediment TOC values for the 16 PAHs in Weihe River, which was similar to some other water bodies in the world [3,9,22]. The possible reason was that the sorption of PAHs to sediments will be affected by both organic matter content and the inorganic matrix (e.g., clay minerals) when sediment TOC values were low. In every site, the Kd (SPM) was much higher than Kd(sediment) ( Table 3), which probably indicated the fresh PAHs input in Weihe River.   Notes: ∑PAHs (ng/L) a = ∑PAHs in SPM per liter water; ∑PAHs (ng/g) b = ∑PAHs per gram SPM.

Source of PAHs in Weihe River
Pyrolytic and petrogenic sources are two major origins of anthropogenic PAHs in the environment. Understanding of the sources of PAHs is very important to study the transportation and fate of PAHs in environment. The molecular ratios of specific hydrocarbons were the most widely used method to distinguish the sources of PAHs in sediment environment. Ratios such as NAP/FLU, PHE/ANT, FLA/PYR, CHR/BaA, PYR/BaP, BaP/BeP, and MPHE/PHE have been developed for interpreting PAH composition and inferring the possible sources in many researches [13,[23][24][25][26]. Among these molecular ratios, ratios of mass 178 and 202, were more widely used. For mass 178, ANT and PHE are two structural isomers, PHE is more thermodynamically stable. An ANT to ANT plus PHE (ANT/178) ratio < 0.1 usually is taken as an indication of petroleum while a ratio > 0.1 indicates a dominance of combustion. As for mass 202, a FLA to FLA plus PYR (FLA/(FLA + PYR)) ratio of 0.50 is usually defined as the petroleum/combustion transition point. A ration of FLA/(FLA + PYR) < 0.5 indicates a petrogenic source, while FLA/(FLA + PYR) > 0.5 means a pyrolytic origin.
In this study, concentration ratios of FLA/(FLA + PYR) and ANT/(ANT + PHE) were used to distinguish the possible PAH origins in sediment in Weihe River. The ratios for ANT/(ANT + PHE) versus FLA/(FLA + PYR) were showed in Figure 4. It could be seen from Figure 4 that sediments from Weihe River originated from both pyrolytic and petrogenic sources. The pyrolytic source could be explained that coal burning was the main source for energy in this area. The petrogenic source was due to that many petroleum industry and chemical plant located in the drainage basin. Therefore, complex sources of PAHs contributed to the ∑PAHs burden in catchment of Weihe River.

Toxicity and Risk Assessment
Sediment Quality Guidelines (SQGs) on the basis of Biological Effects Database for Sediments (BEDS) have been suggested to evaluate the contamination effects in sediments of marine and estuarine ecosystems [27,28]. In this study, we applied four reliable sets of SQGs, including Effects Range-Low value (ERL), Effects Range-Median value (ERM), Probable Effects Level (PEL), and Threshold Effects Level (TEL), to evaluate the potential eco-toxicological risks of the individual and total PAH levels in the sediments of Weihe Rivers. Harmful biological impairments occur rarely (<ERL and <TEL), occasionally (≥ERL, ≥TEL, <ERM, and <PEL), and frequently (≥ERM and ≥PEL) based on the SQGs [27,28]. The concentrations of PAHs in Weihe River were compared with the ERL, ERM, PEL, and TEL values (Table 4). Notes: a ERL = effects range-low value [27]; b ERM = effects range-median value [27]; c TEL = threshold effect levels [28]; d PEL = probable effect levels [28].
The comparison of the mean values of individual PAH and total PAHs in the upper sediments of the Weihe River with the SQGs (Table 4) indicate that the concentrations of the NAP, ACE, FLU, PHE, FLA and total PAHs were higher than ERL values. The mean values of ACY and DhA were higher than TEL values. These results suggest the occurrence of occasional harmful biological impairments at upper reach of Weihe River. The mean values of ACE and FLU were even higher than corresponding PEL values, suggesting frequently harmful biological impairments.
The comparison of the mean values of individual PAH and total PAHs in the middle sediments of the Weihe River with the SQGs ( Table 4) indicate all of the concentrations of individual PAH and total PAHs were higher than corresponding ERL or TEL values. These results suggest the occurrence of occasional harmful biological impairments at middle reach of Weihe River. The mean values of NAP, ACE and FLU were even higher than corresponding PEL values, suggesting frequently harmful biological impairments in middle reach in Weihe River.
The concentrations of PAH in lower reach of Weihe River were lowest. However, the concentrations of the NAP, ACY, ACE, FLU and PHE, were higher than ERL or TEL values, suggesting the occurrence of occasional harmful biological impairments. The ACE and FLU average concentrations were higher than corresponding PEL values. These results indicated harmful biological impairments occur frequently.

Conclusions
Analyses of Weihe River have provided very useful information for the evaluation of trace PAHs and probable sources in rivers that are located in the arid and semi-arid regions in Northwest China. The results obtained in this study showed that concentrations of total and individual PAHs in WDP, SPM and sediment varied significantly among sampling locations. The concentrations of PAH in Weihe River were at a higher level in comparison with other rivers worldwide. In both WDP and sediment, the highest concentrations of ∑PAHs were observed in middle reach, while lowest concentrations of ∑PAHs were found in lower reach. For SPM, however, the highest concentrations were found in lower reach and lowest concentrations were found in upper reach. The calculation of molecular ratio of ANT/(ANT + PHE) and FLA/(FLA + PYR) suggested a mixture of pyrolytic and petrogenic contamination in this river. The ecotoxicological risk evaluation based on international SQGs (ERL, ERM, TEL and PEL) indicated that harmful biological impairments would be caused by PAH compounds occasionally or frequently. It could be concluded that government officials should not only emphasize GDP growth, but also attach importance to environmental protection, and for environmental engineers should develop theories and devices that could fight environmental problems.