Environ Monit Assess (2015) 187:232 DOI 10.1007/s10661-015-4465-y
Distribution and source apportionment of polycyclic aromatic hydrocarbons in the surface soil of Baise, China Bingfang Shi & Qilin Wu & Huixiang Ouyang & Xixing Liu & Jinlei Zhang & Weiyuan Zuo
Received: 6 May 2014 / Accepted: 18 March 2015 # Springer International Publishing Switzerland 2015
Abstract To estimate the distribution and sources of polycyclic aromatic hydrocarbons (PAHs) in the soils of Baise, in southwest China, soil sampling sites were selected from industry, traffic, rubbish, gas station, residential, and suburban areas for analysis of PAHs. The average concentrations of ∑16PAHs in the present study varied significantly, depending on the sampling location, and ranged from 16.8 to 6,437.0 μg/kg (dry weight basis), with a mean value of 565.8 μg/kg. PAH concentrations decreased significantly along the industry–traffic–rubbish–gas station–residential–suburban transect. The PAH profiles in the surface soil of the different areas imply that either source proximity to the sampling sites, or transport and deposition effects influenced PAH distributions. Two diagnostic ratios were selected and used to apportion PAH sources in the surface soil, and bivariate plots show general trends of covariation. Principal component analysis and multivariate linear regression were used to determine the primary sources and their contributions of PAHs to the soils. The model showed that factors 1 (coal and wood combustion) and 2 (petroleum combustion) contributed over 52.1 and 32.5 % of the total source of soil PAHs, respectively. The remaining 15.4 % came from evaporative and uncombusted petroleum. Keywords Distribution . Diagnostic ratio . Principal components analysis . Polycyclic aromatic hydrocarbons . Surface soil B. Shi (*) : Q. Wu : H. Ouyang : X. Liu : J. Zhang : W. Zuo Department of Chemistry and Life Sciences, Baise University, 21 2nd Zhongshan Road, Baise City, Guangxi 533000, China e-mail: [email protected]
Introduction Polycyclic aromatic hydrocarbons (PAHs) are known to be widely distributed throughout environmental media, such as air, water, sediment, and soil (Lin et al. 2015; Zhi et al. 2015; Mandalakis et al. 2014; Yu et al. 2014a). Most PAHs, particularly the four-ring and six-ring types, have strong carcinogenic properties, mutagenicity, and distortion (Yang et al. 2008). Environmental PAHs are formed during pyrolysis or through incomplete combustion of organic materials, and can also be derived from the spillage of petroleum products (Barhoumi et al. 2014; Wang et al. 2007). Consequently, their sources include burning of coal and wood for domestic uses, fossil and biomass fuel burning by power plants, industrial processes, and road transport. PAHs released into the atmosphere are later deposited onto surface waters and soils, where they have a long lifetime but, subsequently, reevaporate to the atmosphere. Thus, soils are a primary reservoir and sink for semivolatile PAHs in the terrestrial environment, and soil PAH concentrations are generally considered to be good indicators of local pollution (Yu et al. 2014b; Wild and Jones 1995). Soils are one of the most important natural resources for human beings. However, the accumulation of PAHs in soils can lead to various negative consequences for agricultural ecosystems and the human food chain, due to direct and indirect human exposure. In addition, PAH contamination of soil can directly affect public health because soil can be easily transferred to humans via ingestion, inhalation, or dermal contact. Studies have found that the amount of human exposure to PAHs
232
Page 2 of 14
through soils was higher than via contact through air or water (Yu et al. 2014a; Menzie et al. 1992). Therefore, research on PAH soil contamination can be used to significantly minimize the risks of human exposure (Ping et al. 2007). Since PAHs were first detected in surface soils (Blumer 1961), many reports have been written about the concentrations, sources, and risk assessment of PAHs in soils from different regions and countries (Morillo et al. 2007; Peng et al. 2011; Essumang et al. 2011; Arienzo et al. 2015). In addition, some research has indicated that anthropogenic sources, such as the incomplete combustion of fossil fuels or vehicle exhausts, are major sources of soil PAHs (Barhoumi et al. 2014; Fernandes et al. 1999). Characteristic PAH ratios have been commonly used as a means of estimating the dominant PAH sources in soils (Wang et al. 2013; Tobiszewski and Namiesnik 2012; Gschwend and Hites 1981). However, the use of these diagnostic ratios in qualitative analyses is limited due to reliability issues. Recently, multivariate analysis has been used in quantitative source apportionment of PAHs in natural environments. Principal component analysis (PCA), along with multivariate linear regression, is one such method used by many researchers for the identification and apportionment of soil PAH sources (Chow and Watson 2002; Zuo et al. 2007). In China, studies on soil PAHs have been conducted continually since the 1980s. Although several previous investigations on PAH concentrations and sources in soil have been conducted (Tang et al. 2005; Zhang et al. 2006; Cai et al. 2007; Li et al. 2008; Jiang et al. 2009), these studies have mainly focused on sites with known specific PAH pollution sources. These included industrial zones in Liaoning Province (Li et al. 2001), urban soils in Dalian (Wang et al. 2009) and Beijing (Peng et al. 2011), vegetable soils of Guangzhou (Chen et al. 2005), surface soils from outskirts of Beijing (Ma et al. 2005), surficial sediments of Xiamen Harbour and Yuan Dan Lake (Ou et al. 2004), residential coal combustion (Chen et al. 2004), and the Lalu wetlands of Lhasa, Tibet (Qi et al. 2003). However, most PAHs reach soils through atmospheric deposition, so diffuse PAHs may have a wide occurrence in the soils of regions where a number of PAH emission sources, concentrations, and profiles are present (Ping et al. 2007). The interplay between long-range/regionalscale atmospheric transport and PAH concentration trends in different areas remains unclear.
Environ Monit Assess (2015) 187:232
Baise is located in the northwest region of Guangxi, in southwest China. The industries in this city are flourishing due to the presence of rich aluminum and petroleum deposits in the area. However, the rapid growth of industrial production and increases in traffic density has caused Baise to be faced with serious environmental problems over the last two decades. The local government has recently taken strong measures to reduce environmental pollution. However, data related to changes in PAH soil contamination in Baise remains sparse, and research is yet to be conducted on the levels, sources, and the risk assessment of PAHs in soils of this location. To alleviate this deficiency of information, this study attempts to investigate the distribution and migration of PAHs in soils of Baise. This work also aims to identify major PAH sources in surface soils by application of diagnostic ratios and PCA.
Materials and methods Sampling site description Baise is one of the newest industrial bases and tourist cities in China. The city covers a total land area of 36,000 km2 and holds a population of 3,780,000. Baise has the largest aluminum production and processing facility in China as well as a petrochemical complex and has a high population density with heavy vehicular traffic. Thirty two surface soil samples from different locations in Baise were collected between June and November 2013. The samples were divided into six groups: those from industrial sites (including a new power plant, an old power plant, refinery, lubricant oil plant, cement plant, and a base for aluminum production and processing), traffic areas, rubbish areas near open burning sites (within 500 m of the E-waste open burning site), gas station fields, residential sites (in the city), and suburban areas (more remote sites up to 7 km from the city center). The locations of the sampling sites are shown in Fig. 1. Baise has a warm and humid subtropical climate with an average annual temperature of 21.5 °C, a relative humidity of 75 %, and a mean annual rainfall of 1,200 mm. The prevailing wind is from east to west. Youjiang River, which is a source of the Pearl River, and a large, mountainous area with distributed woodland can be found in Baise.
Environ Monit Assess (2015) 187:232
Page 3 of 14 232
Fig. 1 Map of the soil sampling sites in Baise, China
Chemicals and materials Standard PAH mixture were obtained from Supelco and included the following compounds: acenaphthene, fluoranthene, naphthalene, benzo (a )a nth ra cene, benzo(a)pyrene, benzo(b)fluoranthene, benzo(k)fluoranthene, acenaphthylene, chrysene, anthracene, benzo(ghi)perylene, fluorene, phenanthrene, dibenzo(ah)anthracene, pyrene, indeno(1,2,3cd)pyrene, with a concentration of 1,000 g/ml (acetonitrile as solvent) for all the PAHs. Silica gel (100 to 200 mesh) purchased from Qingdao Haiyang Chemical (Shandong, China) was activated for approximately 16 h at 130 °C and then stored in a sealed desiccator. Anhydrous sodium sulfate was baked at 450 °C for 5 h. All utilized solvents (Nanning Jingmi, Nanning, China) were of analytical grade. Purified water was taken from a Milli-Q system supplied by Millipore (Bedford, USA). Glassware was cleaned before use by repeated washing with hot methanolic potassium hydroxide, a hot chromic, and concentrated sulfuric acid mixture, and purified water before being dried at 150 °C. Samples procedure Sampling points were arrayed on the diagonal line of a rectangular sampling section based on topography. The sampling sections covered an area of approximately
1,000 m2. Soil samples were collected with a stainless steel scoop from 0 to 30 cm below the surface layer. Each sample was the mixture of five soil samples collected at the four corners and the center in an area of about 1,000 m2 from every six soil sampling sites. The samples were prepared by shaking the soil from roots of grass, and then the soil samples were evenly mixed and dried for 5 days at room temperature in a clean, shady room. The samples were then ground in a mortar and sieved through a 1-mm sieve and then preserved at a low temperature before assaying. Soxhlet extractions was performed on 5 g of soil sample using a 25 mm o.d. glass thimble with a coarse frit containing a 1 cm layer of 80 to 100 mesh magnesium sulfate pretreated at 400 °C for 4 h. The soil sample was extracted twice with acetone and n-hexane (100 ml, v/v 1:1) for 24 h at 60 °C. After extraction, the extract was concentrated in a rotary evaporator and then cleaned using a multilayer silica gel cleanup column (from top to bottom: 0.8 g silica gel, 3 g 2 % KOH– silica gel, 0.8 g silica gel, 4 g 44 % H2SO4–silica gel, 4 g 22 % H2SO4–silica gel, 0.8 g silica gel, 8 g 10 % AgNO3–silica gel, 5 g anhydrous sodium sulfate). The cleaned extract was eluted with 200 ml n-hexane and concentrated to dryness under a nitrogen stream. The residue was redissolved by acetonitrile and then transferred and diluted to a 5 ml volumetric flask. After treatment by the above procedure, all sample solutions
232
Page 4 of 14
were stored in a refrigerator (4 °C) before analysis (Yang et al. 2008). PAHs analysis PAHs were detected by use of high-performance liquid chromatography (HPLC, LC20A) with a fluorescence detector (RF-20Axl) and ultraviolet detection, equipped with an Agilent ZORBAX Eclipse PAH column (4.6× 100 mm, 3.5 μm). The injection volume was 5.0 μl, and the column temperature was 30 °C. The utilized gradient elution program consisted of 50 % water and 50 % acetonitrile for 5 min and programmed to 100 % acetonitrile for 19 min and then reprogrammed to 50 % water and 50 % acetonitrile for 10 min at a flow rate of 1.5 ml/ min. Source apportionment PCA and multivariate linear regression are common types of source-receptor models that have been successfully used for PAH apportionment. Both PAH diagnostic ratios and PCA were chosen in this study to identify the major sources of PAHs in the surface soils of Baise. PCA and multivariate linear regression were also employed to estimate source contributions quantitatively. Source apportionment using PAH diagnostic ratios and PCA was first applied to the dataset, with the PCA being performed after varimax rotation. For a quantitative analysis of relative source contributions, multivariate linear regressions were conducted using PCA factor scores and 15 PAH concentration values [except for dibenzo(ah)anthracene] as independent and dependent variables, respectively. The regression was run using a forward stepwise method. The standardized regression coefficients were used to calculate the relative contributions of various PAH sources. PCA and multivariate linear regression were performed using SPSS11.5 software. Quality control All soil samples were analyzed in triplicate. The variation in PAH concentrations of the replicated samples was less than 10 %. PAHs were quantified by an internal calibration curve with a linear relationship of r2 >0.996. The laboratory blanks were generally low and posed no problems for analytical quantification. The limits of detection (LODs) calculated with a signal-to-noise ratio
Environ Monit Assess (2015) 187:232
of 3:1 in the blank sample (n=5) ranged from 0.50 to 0.91 ng/g. In addition, 10 μl of 16 PAHs with concentrations of 10 μg/g were added to the standard soil. The spiked soil was kept at room temperature in darkness overnight before the extraction was performed. Spiked soil samples were extracted and cleaned up as described previously. The average recoveries of 16 PAH standards from soil samples ranged from 78 to 104 %. Results presented for the soil samples were not corrected according to the spike recoveries.
Results and discussion Total PAH concentrations and distributions in surface soils The individual PAH concentrations of most samples were above the limits of detection (LODs) with the exception that dibenzo(ah)anthracene was detected in only 8 of the 32 stations. Table 1 presents the data for individual PAHs and the total concentrations (∑16PAHs) for the six sampling groups from Baise. Average concentrations of 15 PAHs (except for Daa) in most soil samples from Baise was clearly higher than the endogenous concentration (1–10 ng/g for individual PAHs) resulting from biosynthesis and forest fires (Edwards 1983). The average concentrations of ∑16PAHs in the present study varied significantly depending on the sampling location and ranged from 16.8 to 6,437.0 μg/kg (dry weight basis), with a mean value of 565.8 μg/kg. The five-ring PAH concentrations ranged from undetectable to 1,348.1 μg/kg, with an average value of 82.7 μg/kg from the 32 surface soil samples. PAH concentrations appeared to be strongly linked to regional land use. The data showed that soils in industrial areas were heavily contaminated (1,176.4 ± 978.0 μg/kg) by PAHs. PAH concentrations in sampling soils collected from traffic (507.3±383.0 μg/kg), gas stations (301.2 ± 182.3 μg/kg), and rubbish areas (482.1±287.0 μg/kg) were generally higher than those found in suburban soil (5–7 km from the center). As expected, PAH concentrations showed a strong industry–traffic–rubbish–gas station–residential–suburban gradient. The highest concentrations of Σ16PAHs were observed in industrial site 1, followed by industrial sites 2, 3, and 5. Industrial sites 1 and 2 were near the old power plant (above 160 m from the old power plant) and the new power plant (above 360 m from the new power
81.0
40.1
51.5
2
3
3
3
4
4
4
4
4
5
Fluorene
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Benzo(a)anthracene
Chrysene
Benzo(b)fluoranthene
Benzo(k)fluoranthene
Benzo(a)pyrene
Dibenzo(a,h)anthracene 5
Indeno(1,2,3-cd)pyrene 5
Benzo(ghi)perylene
51.5
1,176.4± 978.0 716.9± 316.5
Six-ring PAH
∑PAHs
a
123.1
12.0
37.6
24.6
65.4
38.4
62.4
37.6
5.1
15.4
4.1
3.5
42.6
3.0
14.1
2.2
1.5
34.5
2.4
3.5
25.6
21.3
5.6
84.6
187.4
161.1
48.3
25.9
84.6
99.3
ND
88.1
66.9
25.3
34.7
12.6
21.6
17.7
7.4
23.2
6.0
2.8
11.5
195.8
407.1
366.5
73.3
47.1
195.8
221.2
ND
185.9
199.1
82.3
42.1
14.1
28.9
29.4
12.9
31.0
12.9
13.1
12.0
9.1
22.2
83.6
28.6
10.5
9
22.2
70.5
ND
13.1
11.9
10.4
2.6
0.6
3.1
3.8
0.6
6.1
1.0
ND
6.4
1.6
Min
507.3± 301.2± 482.1± 176.7± 76.5± 383.0 182.3 287.0 148.8 66.3 450.8± 125.2± 306.0± 116.7± 23.9± 412.2 103.6 201.3 67.7 16.1
311.2
1,348.1
2,871
1,944.3
1,079
311.2
274.8
497.1
576.2
642.7
447.2
674.2
637.2
469.7
377.6
787.4
779.3
771.4
92.1
92.4
Max
34.6
49.3
ND
38.1
36.9
45.3
34.7
12.6
31.6
11.1
16.6
57.3
63.7
72.2
75.9
85
16.7
33
34.6
87.4
68.8 161.1
50.2
132.5 312.9
16.7
13.5
ND
19.5
16.6
27.4
2.5
3.7
18.6
6.7
4.2
39.3
14.9
35.3
35.9
46.4 101.1
Mean Max
Gas station
44.5
68.3
42.0
61.0
6.4
35.0
46.4
61.0
57.1
9.1
8.7
19.8
43.1
23.8
9.1
14.4 ND
21.9 ND
8.7
9.5 ND 20.8
6.5 209.1 361.2 4.3
9.8
3.0
7.0
18.6
5.1
19.4
26.3 ND
81.8 192
35.0
10.1
18.9
17.4
6.5
23.2
49.0
98.8 173.1 29.4
12.6
4.6
6.6
Min
59.2 ND
97.9
19.6
15.3
51.2 103.3
35.2
37.6
37.0
47.6
11.3
12.4
10.5
Mean Max
13.3 109.8 154.8
31
6.4
2.3
ND
2.0
2.2
1.8
0.6
0.8
1.1
2.3
1.6
9.4
6.7
7.8
7.9
8.6
Min
Rubbish
17.7
2.7
1.2
ND
0.9
2.8
6.1
10.2
0.4
5.6
1.0
0.6
1.5
1.4
81.6
2.1
25.1 6.9 ND
26.4
3.1
7.8
6.9 ND
13.2
ND
13.2
7.1
33.2
36.2
9.8
73.9
85.4
11.4
42.5
38.0
2.6 ND
73.9 160.2 11.3
6.4
Min
23.3 ND
54.1 139.3
34.7
2.7
6.3
ND
5.0
3.7
19.8
19.1
3.9
27.4
28.8
3.8
21.5
12.1
1.6
9.7
11.3
Mean Max
Residential
1.3
2.1
14
38.2
20.9
1.3
0.8
ND
1.3
0.5
1.2
6.7
0.8
4.8
6.5
8.4
23.3
3.0
1.5
5.3
11.1
Min
2.1
1.0
1.2 1.3 1.1
ND
9.7
9.3
37.8
62.5
59
0
0
2.4
3.3
1
9.7 ND
4.7 ND
ND
4.6 ND
3.2 ND
8.1 ND
14.4
3.8 ND
8.3
11.6
10.7 ND
40.2
12.5
9.2 ND
17.0 ND
20.3 ND
Mean Max
Suburban
∑COMB: fluoranthene, pyrene, benzo(a)anthracene, benzo(b)fluoranthene, benzo(k)fluoranthene, benzo(a)pyrene, indeno(1,2,3-cd)pyrene, and benzo(ghi)perylene
ND not detected
∑COMBa
460.9
216.3
Five-ring PAH
278.8
Three-ring PAH
Four-ring PAH
168.9
109.9
100.2
90.6
79.8
80.4
69.2
104.8
104.8
86.3
36.2
Two-ring PAH
6
95.2
2
Acenaphthylene
21.5
24.9
2
2
Acenaphthene
Mean
Min
Mean
Max
Traffic
Rings Industrial areas
Naphthalene
PAHs
Table 1 PAH Concentration in the soil in the surface soils of different areas in Baise (μg/kg)
Environ Monit Assess (2015) 187:232 Page 5 of 14 232
232
Environ Monit Assess (2015) 187:232
Page 6 of 14
plant), respectively. Industrial sites 4 and 6 were located in the vicinity of lubricant and cement plants and were associated with pyrogenic combustion and heavy traffic. The lowest PAH concentrations occurred in suburban sites 1, 2, 3, 4, and 5, all of which were dominated by agricultural soils and vegetable plots. These low values (76.5±66.3 μg/kg) were probably due to the location of these sites, as they were far from areas of heavy traffic and extensive anthropogenic activities. The percentages of total PAH concentrations are presented in Fig. 2. These values indicate that the concentrations of total PAHs in three fourths of the samples were higher than 200 μg/kg, and that about one half were within the range of 200 to 600 μg/kg. Total PAH concentrations above 1,000 μg/kg and five-ring PAH concentrations higher than 200 μg/kg were only observed in the industrial sites. According to the criteria established by Maliszewska-Kordybach (1996), which were based on an investigation of PAHs in European agricultural soils, values of soil PAH contamination can be divided into four categories: unpolluted soil (1,000 μg/kg). Based on this classification, three fourths of the sampling sites in Baise would be considered contaminated with PAHs (>200 μg/kg), with the overall soil PAH level from Baise classified as Bweakly contaminated.^ However, the environmental and human health risks posed by the higher PAH concentrations found in the industrial sites deserve further attention. Overall, total PAH concentrations found within industrial soils from Baise are comparable to those
50
Percentage (%)
40
30
20
10
0 >1000
600
1000
200
600
Concentration ( g/kg)
Fig. 2 Frequency of the total PAHs concentrations
0.1 indicates a combustion source, whereas a ratio of 0.5 is suggestive of coal, grass, and wood combustion, whereas values in between 0.4 and 0.5 suggests petroleum (i.e., liquid fossil fuel, vehicle, and crude oil) combustion, while those
Distribution and source apportionment of polycyclic aromatic hydrocarbons in the surface soil of Baise, China Bingfang Shi & Qilin Wu & Huixiang Ouyang & Xixing Liu & Jinlei Zhang & Weiyuan Zuo
Received: 6 May 2014 / Accepted: 18 March 2015 # Springer International Publishing Switzerland 2015
Abstract To estimate the distribution and sources of polycyclic aromatic hydrocarbons (PAHs) in the soils of Baise, in southwest China, soil sampling sites were selected from industry, traffic, rubbish, gas station, residential, and suburban areas for analysis of PAHs. The average concentrations of ∑16PAHs in the present study varied significantly, depending on the sampling location, and ranged from 16.8 to 6,437.0 μg/kg (dry weight basis), with a mean value of 565.8 μg/kg. PAH concentrations decreased significantly along the industry–traffic–rubbish–gas station–residential–suburban transect. The PAH profiles in the surface soil of the different areas imply that either source proximity to the sampling sites, or transport and deposition effects influenced PAH distributions. Two diagnostic ratios were selected and used to apportion PAH sources in the surface soil, and bivariate plots show general trends of covariation. Principal component analysis and multivariate linear regression were used to determine the primary sources and their contributions of PAHs to the soils. The model showed that factors 1 (coal and wood combustion) and 2 (petroleum combustion) contributed over 52.1 and 32.5 % of the total source of soil PAHs, respectively. The remaining 15.4 % came from evaporative and uncombusted petroleum. Keywords Distribution . Diagnostic ratio . Principal components analysis . Polycyclic aromatic hydrocarbons . Surface soil B. Shi (*) : Q. Wu : H. Ouyang : X. Liu : J. Zhang : W. Zuo Department of Chemistry and Life Sciences, Baise University, 21 2nd Zhongshan Road, Baise City, Guangxi 533000, China e-mail: [email protected]
Introduction Polycyclic aromatic hydrocarbons (PAHs) are known to be widely distributed throughout environmental media, such as air, water, sediment, and soil (Lin et al. 2015; Zhi et al. 2015; Mandalakis et al. 2014; Yu et al. 2014a). Most PAHs, particularly the four-ring and six-ring types, have strong carcinogenic properties, mutagenicity, and distortion (Yang et al. 2008). Environmental PAHs are formed during pyrolysis or through incomplete combustion of organic materials, and can also be derived from the spillage of petroleum products (Barhoumi et al. 2014; Wang et al. 2007). Consequently, their sources include burning of coal and wood for domestic uses, fossil and biomass fuel burning by power plants, industrial processes, and road transport. PAHs released into the atmosphere are later deposited onto surface waters and soils, where they have a long lifetime but, subsequently, reevaporate to the atmosphere. Thus, soils are a primary reservoir and sink for semivolatile PAHs in the terrestrial environment, and soil PAH concentrations are generally considered to be good indicators of local pollution (Yu et al. 2014b; Wild and Jones 1995). Soils are one of the most important natural resources for human beings. However, the accumulation of PAHs in soils can lead to various negative consequences for agricultural ecosystems and the human food chain, due to direct and indirect human exposure. In addition, PAH contamination of soil can directly affect public health because soil can be easily transferred to humans via ingestion, inhalation, or dermal contact. Studies have found that the amount of human exposure to PAHs
232
Page 2 of 14
through soils was higher than via contact through air or water (Yu et al. 2014a; Menzie et al. 1992). Therefore, research on PAH soil contamination can be used to significantly minimize the risks of human exposure (Ping et al. 2007). Since PAHs were first detected in surface soils (Blumer 1961), many reports have been written about the concentrations, sources, and risk assessment of PAHs in soils from different regions and countries (Morillo et al. 2007; Peng et al. 2011; Essumang et al. 2011; Arienzo et al. 2015). In addition, some research has indicated that anthropogenic sources, such as the incomplete combustion of fossil fuels or vehicle exhausts, are major sources of soil PAHs (Barhoumi et al. 2014; Fernandes et al. 1999). Characteristic PAH ratios have been commonly used as a means of estimating the dominant PAH sources in soils (Wang et al. 2013; Tobiszewski and Namiesnik 2012; Gschwend and Hites 1981). However, the use of these diagnostic ratios in qualitative analyses is limited due to reliability issues. Recently, multivariate analysis has been used in quantitative source apportionment of PAHs in natural environments. Principal component analysis (PCA), along with multivariate linear regression, is one such method used by many researchers for the identification and apportionment of soil PAH sources (Chow and Watson 2002; Zuo et al. 2007). In China, studies on soil PAHs have been conducted continually since the 1980s. Although several previous investigations on PAH concentrations and sources in soil have been conducted (Tang et al. 2005; Zhang et al. 2006; Cai et al. 2007; Li et al. 2008; Jiang et al. 2009), these studies have mainly focused on sites with known specific PAH pollution sources. These included industrial zones in Liaoning Province (Li et al. 2001), urban soils in Dalian (Wang et al. 2009) and Beijing (Peng et al. 2011), vegetable soils of Guangzhou (Chen et al. 2005), surface soils from outskirts of Beijing (Ma et al. 2005), surficial sediments of Xiamen Harbour and Yuan Dan Lake (Ou et al. 2004), residential coal combustion (Chen et al. 2004), and the Lalu wetlands of Lhasa, Tibet (Qi et al. 2003). However, most PAHs reach soils through atmospheric deposition, so diffuse PAHs may have a wide occurrence in the soils of regions where a number of PAH emission sources, concentrations, and profiles are present (Ping et al. 2007). The interplay between long-range/regionalscale atmospheric transport and PAH concentration trends in different areas remains unclear.
Environ Monit Assess (2015) 187:232
Baise is located in the northwest region of Guangxi, in southwest China. The industries in this city are flourishing due to the presence of rich aluminum and petroleum deposits in the area. However, the rapid growth of industrial production and increases in traffic density has caused Baise to be faced with serious environmental problems over the last two decades. The local government has recently taken strong measures to reduce environmental pollution. However, data related to changes in PAH soil contamination in Baise remains sparse, and research is yet to be conducted on the levels, sources, and the risk assessment of PAHs in soils of this location. To alleviate this deficiency of information, this study attempts to investigate the distribution and migration of PAHs in soils of Baise. This work also aims to identify major PAH sources in surface soils by application of diagnostic ratios and PCA.
Materials and methods Sampling site description Baise is one of the newest industrial bases and tourist cities in China. The city covers a total land area of 36,000 km2 and holds a population of 3,780,000. Baise has the largest aluminum production and processing facility in China as well as a petrochemical complex and has a high population density with heavy vehicular traffic. Thirty two surface soil samples from different locations in Baise were collected between June and November 2013. The samples were divided into six groups: those from industrial sites (including a new power plant, an old power plant, refinery, lubricant oil plant, cement plant, and a base for aluminum production and processing), traffic areas, rubbish areas near open burning sites (within 500 m of the E-waste open burning site), gas station fields, residential sites (in the city), and suburban areas (more remote sites up to 7 km from the city center). The locations of the sampling sites are shown in Fig. 1. Baise has a warm and humid subtropical climate with an average annual temperature of 21.5 °C, a relative humidity of 75 %, and a mean annual rainfall of 1,200 mm. The prevailing wind is from east to west. Youjiang River, which is a source of the Pearl River, and a large, mountainous area with distributed woodland can be found in Baise.
Environ Monit Assess (2015) 187:232
Page 3 of 14 232
Fig. 1 Map of the soil sampling sites in Baise, China
Chemicals and materials Standard PAH mixture were obtained from Supelco and included the following compounds: acenaphthene, fluoranthene, naphthalene, benzo (a )a nth ra cene, benzo(a)pyrene, benzo(b)fluoranthene, benzo(k)fluoranthene, acenaphthylene, chrysene, anthracene, benzo(ghi)perylene, fluorene, phenanthrene, dibenzo(ah)anthracene, pyrene, indeno(1,2,3cd)pyrene, with a concentration of 1,000 g/ml (acetonitrile as solvent) for all the PAHs. Silica gel (100 to 200 mesh) purchased from Qingdao Haiyang Chemical (Shandong, China) was activated for approximately 16 h at 130 °C and then stored in a sealed desiccator. Anhydrous sodium sulfate was baked at 450 °C for 5 h. All utilized solvents (Nanning Jingmi, Nanning, China) were of analytical grade. Purified water was taken from a Milli-Q system supplied by Millipore (Bedford, USA). Glassware was cleaned before use by repeated washing with hot methanolic potassium hydroxide, a hot chromic, and concentrated sulfuric acid mixture, and purified water before being dried at 150 °C. Samples procedure Sampling points were arrayed on the diagonal line of a rectangular sampling section based on topography. The sampling sections covered an area of approximately
1,000 m2. Soil samples were collected with a stainless steel scoop from 0 to 30 cm below the surface layer. Each sample was the mixture of five soil samples collected at the four corners and the center in an area of about 1,000 m2 from every six soil sampling sites. The samples were prepared by shaking the soil from roots of grass, and then the soil samples were evenly mixed and dried for 5 days at room temperature in a clean, shady room. The samples were then ground in a mortar and sieved through a 1-mm sieve and then preserved at a low temperature before assaying. Soxhlet extractions was performed on 5 g of soil sample using a 25 mm o.d. glass thimble with a coarse frit containing a 1 cm layer of 80 to 100 mesh magnesium sulfate pretreated at 400 °C for 4 h. The soil sample was extracted twice with acetone and n-hexane (100 ml, v/v 1:1) for 24 h at 60 °C. After extraction, the extract was concentrated in a rotary evaporator and then cleaned using a multilayer silica gel cleanup column (from top to bottom: 0.8 g silica gel, 3 g 2 % KOH– silica gel, 0.8 g silica gel, 4 g 44 % H2SO4–silica gel, 4 g 22 % H2SO4–silica gel, 0.8 g silica gel, 8 g 10 % AgNO3–silica gel, 5 g anhydrous sodium sulfate). The cleaned extract was eluted with 200 ml n-hexane and concentrated to dryness under a nitrogen stream. The residue was redissolved by acetonitrile and then transferred and diluted to a 5 ml volumetric flask. After treatment by the above procedure, all sample solutions
232
Page 4 of 14
were stored in a refrigerator (4 °C) before analysis (Yang et al. 2008). PAHs analysis PAHs were detected by use of high-performance liquid chromatography (HPLC, LC20A) with a fluorescence detector (RF-20Axl) and ultraviolet detection, equipped with an Agilent ZORBAX Eclipse PAH column (4.6× 100 mm, 3.5 μm). The injection volume was 5.0 μl, and the column temperature was 30 °C. The utilized gradient elution program consisted of 50 % water and 50 % acetonitrile for 5 min and programmed to 100 % acetonitrile for 19 min and then reprogrammed to 50 % water and 50 % acetonitrile for 10 min at a flow rate of 1.5 ml/ min. Source apportionment PCA and multivariate linear regression are common types of source-receptor models that have been successfully used for PAH apportionment. Both PAH diagnostic ratios and PCA were chosen in this study to identify the major sources of PAHs in the surface soils of Baise. PCA and multivariate linear regression were also employed to estimate source contributions quantitatively. Source apportionment using PAH diagnostic ratios and PCA was first applied to the dataset, with the PCA being performed after varimax rotation. For a quantitative analysis of relative source contributions, multivariate linear regressions were conducted using PCA factor scores and 15 PAH concentration values [except for dibenzo(ah)anthracene] as independent and dependent variables, respectively. The regression was run using a forward stepwise method. The standardized regression coefficients were used to calculate the relative contributions of various PAH sources. PCA and multivariate linear regression were performed using SPSS11.5 software. Quality control All soil samples were analyzed in triplicate. The variation in PAH concentrations of the replicated samples was less than 10 %. PAHs were quantified by an internal calibration curve with a linear relationship of r2 >0.996. The laboratory blanks were generally low and posed no problems for analytical quantification. The limits of detection (LODs) calculated with a signal-to-noise ratio
Environ Monit Assess (2015) 187:232
of 3:1 in the blank sample (n=5) ranged from 0.50 to 0.91 ng/g. In addition, 10 μl of 16 PAHs with concentrations of 10 μg/g were added to the standard soil. The spiked soil was kept at room temperature in darkness overnight before the extraction was performed. Spiked soil samples were extracted and cleaned up as described previously. The average recoveries of 16 PAH standards from soil samples ranged from 78 to 104 %. Results presented for the soil samples were not corrected according to the spike recoveries.
Results and discussion Total PAH concentrations and distributions in surface soils The individual PAH concentrations of most samples were above the limits of detection (LODs) with the exception that dibenzo(ah)anthracene was detected in only 8 of the 32 stations. Table 1 presents the data for individual PAHs and the total concentrations (∑16PAHs) for the six sampling groups from Baise. Average concentrations of 15 PAHs (except for Daa) in most soil samples from Baise was clearly higher than the endogenous concentration (1–10 ng/g for individual PAHs) resulting from biosynthesis and forest fires (Edwards 1983). The average concentrations of ∑16PAHs in the present study varied significantly depending on the sampling location and ranged from 16.8 to 6,437.0 μg/kg (dry weight basis), with a mean value of 565.8 μg/kg. The five-ring PAH concentrations ranged from undetectable to 1,348.1 μg/kg, with an average value of 82.7 μg/kg from the 32 surface soil samples. PAH concentrations appeared to be strongly linked to regional land use. The data showed that soils in industrial areas were heavily contaminated (1,176.4 ± 978.0 μg/kg) by PAHs. PAH concentrations in sampling soils collected from traffic (507.3±383.0 μg/kg), gas stations (301.2 ± 182.3 μg/kg), and rubbish areas (482.1±287.0 μg/kg) were generally higher than those found in suburban soil (5–7 km from the center). As expected, PAH concentrations showed a strong industry–traffic–rubbish–gas station–residential–suburban gradient. The highest concentrations of Σ16PAHs were observed in industrial site 1, followed by industrial sites 2, 3, and 5. Industrial sites 1 and 2 were near the old power plant (above 160 m from the old power plant) and the new power plant (above 360 m from the new power
81.0
40.1
51.5
2
3
3
3
4
4
4
4
4
5
Fluorene
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Benzo(a)anthracene
Chrysene
Benzo(b)fluoranthene
Benzo(k)fluoranthene
Benzo(a)pyrene
Dibenzo(a,h)anthracene 5
Indeno(1,2,3-cd)pyrene 5
Benzo(ghi)perylene
51.5
1,176.4± 978.0 716.9± 316.5
Six-ring PAH
∑PAHs
a
123.1
12.0
37.6
24.6
65.4
38.4
62.4
37.6
5.1
15.4
4.1
3.5
42.6
3.0
14.1
2.2
1.5
34.5
2.4
3.5
25.6
21.3
5.6
84.6
187.4
161.1
48.3
25.9
84.6
99.3
ND
88.1
66.9
25.3
34.7
12.6
21.6
17.7
7.4
23.2
6.0
2.8
11.5
195.8
407.1
366.5
73.3
47.1
195.8
221.2
ND
185.9
199.1
82.3
42.1
14.1
28.9
29.4
12.9
31.0
12.9
13.1
12.0
9.1
22.2
83.6
28.6
10.5
9
22.2
70.5
ND
13.1
11.9
10.4
2.6
0.6
3.1
3.8
0.6
6.1
1.0
ND
6.4
1.6
Min
507.3± 301.2± 482.1± 176.7± 76.5± 383.0 182.3 287.0 148.8 66.3 450.8± 125.2± 306.0± 116.7± 23.9± 412.2 103.6 201.3 67.7 16.1
311.2
1,348.1
2,871
1,944.3
1,079
311.2
274.8
497.1
576.2
642.7
447.2
674.2
637.2
469.7
377.6
787.4
779.3
771.4
92.1
92.4
Max
34.6
49.3
ND
38.1
36.9
45.3
34.7
12.6
31.6
11.1
16.6
57.3
63.7
72.2
75.9
85
16.7
33
34.6
87.4
68.8 161.1
50.2
132.5 312.9
16.7
13.5
ND
19.5
16.6
27.4
2.5
3.7
18.6
6.7
4.2
39.3
14.9
35.3
35.9
46.4 101.1
Mean Max
Gas station
44.5
68.3
42.0
61.0
6.4
35.0
46.4
61.0
57.1
9.1
8.7
19.8
43.1
23.8
9.1
14.4 ND
21.9 ND
8.7
9.5 ND 20.8
6.5 209.1 361.2 4.3
9.8
3.0
7.0
18.6
5.1
19.4
26.3 ND
81.8 192
35.0
10.1
18.9
17.4
6.5
23.2
49.0
98.8 173.1 29.4
12.6
4.6
6.6
Min
59.2 ND
97.9
19.6
15.3
51.2 103.3
35.2
37.6
37.0
47.6
11.3
12.4
10.5
Mean Max
13.3 109.8 154.8
31
6.4
2.3
ND
2.0
2.2
1.8
0.6
0.8
1.1
2.3
1.6
9.4
6.7
7.8
7.9
8.6
Min
Rubbish
17.7
2.7
1.2
ND
0.9
2.8
6.1
10.2
0.4
5.6
1.0
0.6
1.5
1.4
81.6
2.1
25.1 6.9 ND
26.4
3.1
7.8
6.9 ND
13.2
ND
13.2
7.1
33.2
36.2
9.8
73.9
85.4
11.4
42.5
38.0
2.6 ND
73.9 160.2 11.3
6.4
Min
23.3 ND
54.1 139.3
34.7
2.7
6.3
ND
5.0
3.7
19.8
19.1
3.9
27.4
28.8
3.8
21.5
12.1
1.6
9.7
11.3
Mean Max
Residential
1.3
2.1
14
38.2
20.9
1.3
0.8
ND
1.3
0.5
1.2
6.7
0.8
4.8
6.5
8.4
23.3
3.0
1.5
5.3
11.1
Min
2.1
1.0
1.2 1.3 1.1
ND
9.7
9.3
37.8
62.5
59
0
0
2.4
3.3
1
9.7 ND
4.7 ND
ND
4.6 ND
3.2 ND
8.1 ND
14.4
3.8 ND
8.3
11.6
10.7 ND
40.2
12.5
9.2 ND
17.0 ND
20.3 ND
Mean Max
Suburban
∑COMB: fluoranthene, pyrene, benzo(a)anthracene, benzo(b)fluoranthene, benzo(k)fluoranthene, benzo(a)pyrene, indeno(1,2,3-cd)pyrene, and benzo(ghi)perylene
ND not detected
∑COMBa
460.9
216.3
Five-ring PAH
278.8
Three-ring PAH
Four-ring PAH
168.9
109.9
100.2
90.6
79.8
80.4
69.2
104.8
104.8
86.3
36.2
Two-ring PAH
6
95.2
2
Acenaphthylene
21.5
24.9
2
2
Acenaphthene
Mean
Min
Mean
Max
Traffic
Rings Industrial areas
Naphthalene
PAHs
Table 1 PAH Concentration in the soil in the surface soils of different areas in Baise (μg/kg)
Environ Monit Assess (2015) 187:232 Page 5 of 14 232
232
Environ Monit Assess (2015) 187:232
Page 6 of 14
plant), respectively. Industrial sites 4 and 6 were located in the vicinity of lubricant and cement plants and were associated with pyrogenic combustion and heavy traffic. The lowest PAH concentrations occurred in suburban sites 1, 2, 3, 4, and 5, all of which were dominated by agricultural soils and vegetable plots. These low values (76.5±66.3 μg/kg) were probably due to the location of these sites, as they were far from areas of heavy traffic and extensive anthropogenic activities. The percentages of total PAH concentrations are presented in Fig. 2. These values indicate that the concentrations of total PAHs in three fourths of the samples were higher than 200 μg/kg, and that about one half were within the range of 200 to 600 μg/kg. Total PAH concentrations above 1,000 μg/kg and five-ring PAH concentrations higher than 200 μg/kg were only observed in the industrial sites. According to the criteria established by Maliszewska-Kordybach (1996), which were based on an investigation of PAHs in European agricultural soils, values of soil PAH contamination can be divided into four categories: unpolluted soil (1,000 μg/kg). Based on this classification, three fourths of the sampling sites in Baise would be considered contaminated with PAHs (>200 μg/kg), with the overall soil PAH level from Baise classified as Bweakly contaminated.^ However, the environmental and human health risks posed by the higher PAH concentrations found in the industrial sites deserve further attention. Overall, total PAH concentrations found within industrial soils from Baise are comparable to those
50
Percentage (%)
40
30
20
10
0 >1000
600
1000
200
600
Concentration ( g/kg)
Fig. 2 Frequency of the total PAHs concentrations
0.1 indicates a combustion source, whereas a ratio of 0.5 is suggestive of coal, grass, and wood combustion, whereas values in between 0.4 and 0.5 suggests petroleum (i.e., liquid fossil fuel, vehicle, and crude oil) combustion, while those
