Dynamic Article LinksC<Journal ofEnvironmentalMonitoringCite this: J. Environ. Monit., 2012, 14, 1256

www.rsc.org/jem PAPER

Publ

ishe

d on

08

Febr

uary

201

2. D

ownl

oade

d by

Uni

vers

ity o

f C

alif

orni

a -

Sant

a C

ruz

on 3

1/10

/201

4 07

:27:

02.

View Article Online / Journal Homepage / Table of Contents for this issue

Source and risk assessment of PCBs in sediments of Fenhe reservoir andwatershed, China

Wei-Hong Li,a Ying-Ze Tian,b Guo-Liang Shi,*b Chang-Sheng Guo,c Yin-Chang Fengb and Xiu-Ping Yue*d

Received 8th December 2011, Accepted 7th February 2012

DOI: 10.1039/c2em10983b

The concentrations of polychlorinated biphenyls (PCBs) in sediments from the Fenhe reservoir and

watershed were detected at 28 sites in wet and dry seasons. TheP

123PCBs ranged from n.d. to 126.49

ng g�1 dw. The dominated congeners were tri-PCBs (34.29%) and tetra-PCBs (24.05%). In the Fenhe

reservoir,P

123PCBs presented a decreasing trend, while percentages of low chlorinated congeners

showed an increasing trend. For the temporal variations, PCBs homologues profiles of sediment

samples and spatial distribution ofP

123PCBs for the two periods were similar (with CD¼ 0.021 and r2

¼ 0.999 respectively), although PCBs concentrations in the wet season were significantly higher than in

the dry season. PCA was applied to analyze the possible sources for PCBs, suggesting that PCBs might

be mainly influenced by Aroclor 1016 and 1242. Compared with 3 established sediment quality

guidelines, levels of PCBs in sediments of the investigated watershed might have a potential biological

impact, especially in the wet season.

1. Introduction

Polychlorinated biphenyls (PCBs), which are listed as persistent

organic pollutants (POPs),1,2 have been the focus of great

attention by a number of governments and scientific communi-

ties because of their adverse effects on human health and the

environment.3,4 These compounds have drawn considerable

concern due to their toxicity, health risk as potential carcinogens

and mutagens and property of bioaccumulation through the

aCollege of Resources and Environment, Shanxi Agricultural University,Taigu, Shanxi, 030801, ChinabState Environmental Protection Key Laboratory of Urban Ambient AirParticulate Matter Pollution Prevention and Control, College ofEnvironmental Science and Engineering, Nankai University, Tianjin,300071, China. E-mail: nksgl@hotmail.comcState Key Laboratory of Environmental Criteria and Risk Assessment, andLaboratory of Riverine Ecological Conservation and Technology, ChineseResearch Academy of Environmental Sciences, Beijing 100012, ChinadCollege of Environmental Science and Engineering, Taiyuan University ofTechnology, Taiyuan, Shanxi, 030024, China. E-mail: yuexiuping1990@126.com; Fax: +86 0351-6010214; Tel: +86 0351-6010214

Environmental impact

PCBs have drawn considerable concern due to their toxicity, health r

Fenhe River, supplying water for irrigation and drinking, is one o

pollution status and the importance of the Fenhe River, as well as

significance to understand the pollution status of PCBs in sedime

characteristics, possible sources and potential biological effects of PC

sources and assess the risk of PCBs in sediments of the Fenhe Rive

1256 | J. Environ. Monit., 2012, 14, 1256–1263

food chain.5–8Due to their chemical stability and heat resistance,9

PCBs were extensively used in a variety of electrical and

hydraulic applications, such as dielectric fluids in transformers

and large capacitors, heat transfer fluids and hydraulic fluids in

hydraulic systems.1,10–12 Unfortunately, these properties also

contribute to the persistence and long-range transport capability

of PBCs after being released into the environment. With the

increasing concern of their persistent nature and harmful impacts

on humans, the production, use, and importation of PCBs have

been phased out since the 1970s.4

Furthermore, in order to understand the contamination status

of aquatic systems which are sensitive to pollution, it is important

to study PCBs in the sediments, as PCBs are hydrophobic

compounds that tend to be adsorbed in sediments in the aquatic

system, hence the contaminated sediments can be considered as

a pollution reservoir and act as an internal source of PCBs

because of their resuspension into the water.13 There have been

numerous studies focusing on the levels, distribution and sources

of PCBs in air, water, soil, animal species and humans,12,14,15 as

well as in sediments all over the world.13,16–18

isks and bioaccumulation through the food chain. Furthermore,

f the most important rivers in Shanxi, China. Considering the

the bioaccumulation nature and toxicity of PCBs, it is of great

nts. This work aims to study the levels, spatial and temporal

Bs. To our knowledge, this is the first effort to identify possible

r in China.

This journal is ª The Royal Society of Chemistry 2012

Publ

ishe

d on

08

Febr

uary

201

2. D

ownl

oade

d by

Uni

vers

ity o

f C

alif

orni

a -

Sant

a C

ruz

on 3

1/10

/201

4 07

:27:

02.

View Article Online

China began to produce PCBs in 1965 and ceased in 1974, with

a total of about 10 000 tons of PCBs (including 9000 tons of tri-

PCBs and 1000 tons of penta-PCBs) produced.19However, due to

the fact that the widespread production and use of PCBs before

their legal restriction has led to serious contamination and PCBs

continue tobe released fromold equipment andwaste sites,20,21 the

pollution of PCBs inChina should be concentrated on. In order to

provide theoretical underpinnings for PCBs control strategies,

understanding the levels and sources of PCBs in China is very

important. In China, some studies on PCBs have already been

reported,1,12,17,19 however, most of the studies only focused on the

eastern coastal part of the country and studies on PCBs in sedi-

ments of inland reservoirs andwatersheds are limited, especially in

Shanxi Province, a province in northern China.

The parent river of Shanxi, Fenhe River, is the largest river in

this province. Supplying water for irrigation and drinking, Fenhe

River is one of the most important rivers in Shanxi. Shanxi

Province is a leading producer of coal in China with about one

third of China’s known coal deposits, and is one of the best-

known heavy industrial areas in China. Due to the rapid indus-

trial development and energy structure, Fenhe River and its

watershed are seriously polluted, with a long history of industrial

activities around the watershed, including iron and steel industry,

coal-fired power plants, and several coking and chemical plants.

Our prior work has reported the presence of polycyclic aromatic

hydrocarbons (PAHs) in the sediments of Fenhe reservoir and

upstream watershed.22 However, considering the pollution status

and the importance of the Fenhe River, as well as the bio-

accumulation nature and toxicity of PCBs, it’s still of great

significance to understand the pollution status of PCBs in sedi-

ments. This work aims to study the levels, spatial and temporal

distribution, possible sources and potential biological effects of

PCBs. To our knowledge, this is the first effort to identify

possible sources and assess the risk of PCBs in sediments of the

Fenhe River in Shanxi.

Therefore, in this work, PCBs homologues were measured in

surface sediments at 28 sampling sites throughout Fenhe reser-

voir and upstream watershed in wet and dry seasons. Firstly,

levels and composition patterns of PCBs in the investigated

watershed were discussed and compared with those in other

studies. Then, spatial and temporal distributions were explored

bymethods such as Pearson correlation coefficient and coefficient

of divergence (CD). Principal component analysis (PCA) was

applied to analyze the possible sources for PCBs. Finally,

potential biological effects were assessed by 3 useful sediment

quality guidelines: the effect range low (ERL),17,23,24 the threshold

effects level (TEL)17,25 and the lowest effect level (LEL).26

2. Materials and methods

2.1 Sample collection

As described in ref. 22, the investigated watershed is located in

Ningwu county, Jinle county of Xinzhou District, Lang county

of Lvliang District, and Loufan county of Taiyuan, with an area

of 5268 km2 and 184 industrial and domestic pollution sources

around the area.27 In recent years, with the rapid industrializa-

tion and urbanization in the surrounding region, Fenhe reservoir

and upstream watershed has been exposed to severe pollution.

This journal is ª The Royal Society of Chemistry 2012

According to Xu et al.,27 about 552.86 million tons of industrial

and municipal wastewater is discharged annually into the

investigated watershed. Industrial wastewater, municipal sewage,

roadway runoff and agricultural non-point sources are the major

primary dischargers to the watershed.

In present work, 56 samples of surface sediments at 28 sites in

two periods were collected (28 for March 2010 and 28 for August

2010) from the investigated watershed, including upstream

Fenhe principal stream, estuaries of main branch streams and

Fenhe reservoir in Shanxi Province. The two sampling periods

are defined as the wet season (when rains were heavy in August)

and the dry season (when there was little rain in March). Details

of the sampling sites are shown in Fig. 1. The 28 sampling sites

included 13 on the Fenhe principal stream (S1–S4, S6, S7, S9–

S11, S13, S14, S16, S18), 5 on the main branch streams estuaries

(S5 for Honghe River, S8 for Minghe River, S12 for Dongnianhe

River, S15 for Langhe River, S17 for Jianhe River), and 10 on the

Fenhe reservoir (S19–S28).

The sediments were collected using a stainless steel grab

sampler. Approximately 5 cm of sediment were taken from the

top of the river beds and placed in a precleaned aluminum box

using a stainless-steel spoon. All of the sediment samples were

freeze-dried, ground, homogenized and stored at�20 �C prior to

analysis.

2.2 PCBs extraction and cleanup

In this work, all sediment samples (freeze-dried) were ground and

homogenized in order to prepare for analyzing. Firstly, for each

sediment sample (weighing 10 g), it was treated with 70 mL of

hexane–dichloromethane (1 : 1, v/v, chromatographic-grade,

Fluka Co., USA) for three hours in an ultrasonic bath. Next, all

samples were sonicated for 45 min. Then, we decanted the

extracts and re-sonicated the remainder with 40 mL hexane for

45 min. After re-sonicating, the two extracts were mixed together

and treated with activated Cu, in order to remove sulfates. The

mixture was concentrated to 5.0 mL by a rotary evaporator.

Then a separating funnel was employed to remove the extract.

The concentrated sulfuric acid (10 mL, 98%, AR) was added

three times to remove the impurities. A 5% NaCl solution

(60 mL) was applied to wash the organic phase, and the washed

organic phase was then concentrated to about 1–2 mL. A glass

column (10 mm i.d.) loaded with 20 g of activated Florisil (100–

200 mesh, Tianjin Daimao Chemical Regents Factory, China),

which was activated in an oven at 600 �C for 10 h and then

deactivated withMilli-Q water at a ratio of 5%, was used to carry

out further purification. Then, PCBs were eluted with 100 mL of

hexane and the hexane eluate was reduced under a gentle stream

of nitrogen to 0.5 mL. An internal standard (penta-

chloronitrobenzene, Sigma Co., USA) was added for GC-MS

(Gas Chromatography-Mass Spectrometry) analysis. The

internal standard method relative to a multilevel calibration for

all compounds was applied to undertake quantification.17

2.3 GC-MS analysis

PCB congeners (PCB-1, 2, 3, 4/10, 5/8, 6, 7/9, 12/13, 14, 15, 17,

18, 19, 20/33, 21, 22, 23, 24/27, 25, 26, 29, 16/32, 30, 28, 34, 35, 39,

38, 51, 52/73, 53, 54, 45, 46, 47/48/75, 49, 59/42, 40, 67, 63, 74, 66,

J. Environ. Monit., 2012, 14, 1256–1263 | 1257

Fig. 1 Map of Shanxi Province in China and sampling sites in Fenhe reservoir and upstream watershed.

Publ

ishe

d on

08

Febr

uary

201

2. D

ownl

oade

d by

Uni

vers

ity o

f C

alif

orni

a -

Sant

a C

ruz

on 3

1/10

/201

4 07

:27:

02.

View Article Online

70, 56/60, 81,77, 76, 90/101, 91, 92, 93, 94, 95, 107/109, 115/117,

110, 82, 124, 114, 122, 105, 118, 134, 151, 136, 154, 135, 144, 147,

149, 146, 142, 137, 130, 158, 129, 168, 167, 164/163, 178, 176, 175,

187, 174, 183, 177, 171, 180/193, 170/190, 172, 201, 202, 200, 197,

199, 196/203, 195, 194, 205, 198, 207, 208, 206, 209) were iden-

tified and quantified by using a GC-MS (Varian 4000, USA) with

a DB-5 capillary column (30 m � 0.25 mm � 0.25 mm, J&W,

USA). The GC-MS conditions for sample analysis were as

follows: the oven began at a temperature of 150 �C for 3 min, and

then increased at a rate of 5 �Cmin�1 to 300 �C and was held for 3

min; the injection was at 270 �C and in splitless mode; the carrier

gas was nitrogen, with a flow rate of 1.0 mL min�1; the temper-

ature of transfer line was 205 �C; the electron energy was 70 eV,

and the detector voltage was 500 V.

2.4 Quality assurance and quality control

Along with each batch of six samples, a procedural blank and

a matrix sample spiked with standards were run to make sure the

analytical procedure was operating correctly. All results were

1258 | J. Environ. Monit., 2012, 14, 1256–1263

blank corrected. Duplicate (n ¼ 3) samples were studied for

quality assurance and control experiments. For duplicate

samples, the relative standard deviations were all below 15%.

Then the matrix sample (pre-extracted sediment) was performed

with spike recoveries. The matrix sample (10 g) spiked with

a mixture of 123 PCBs was equilibrated at 25 �C for 24 h. For

matrix samples, the same chemical analysis method mentioned

above was employed. The spiked recovery for PCB congeners in

the sediments ranged from 68% to 117%; the MDLs (method

detection limit) ranged from 0.01 to 0.05 ng g�1. All the results

were corrected with the recovery ratios and reported in ng g�1 dw

(dry weight).

The methods for extraction, chemical analysis, quality assur-

ance and quality control were carried out according to methods

reported in ref. 17.

2.5 Statistical methods

In this work, some statistical methods were used to analyse the

PCBs concentrations and profiles. A T-test was used to

This journal is ª The Royal Society of Chemistry 2012

Publ

ishe

d on

08

Febr

uary

201

2. D

ownl

oade

d by

Uni

vers

ity o

f C

alif

orni

a -

Sant

a C

ruz

on 3

1/10

/201

4 07

:27:

02.

View Article Online

investigate the significance between concentrations in wet and

dry seasons. Pearson’s correlation coefficient was calculated to

study the relationship of theP

123PCBs concentrations at 28 sites

for the two periods. Statistical analyses were conducted with

SPSS 16.0.

Coefficient of divergence (CD)28,29 was used to study the

temporal difference of PCBs homologues profiles in sediment

samples. CD can be calculated as follows:

CDfj ¼ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi1

p

Xp

i¼1

�xif � xij

xif þ xij

�2

vuut

where xif is the fraction of the ith PCBs homologue inP

123PCBs

in the fth period (averaged over 28 sites). f and j represent the two

sampling periods, and p is the number of PCBs homologues.

When the PCBs homologue profiles in the two sampling periods

are similar to each other, the CD values would approach 0; on

the other hand, as the two homologue profiles diverge, the CD

value would approach 1.29

Principal component analysis (PCA)30 is an important

receptor model which has been proved to be a useful tool for

source apportionment studies. In this work, PCA was carried out

for all the available samples and commercial mixtures (Aroclor

1254, 1221, 1043, 1016 and 1242) to identify the possible source

categories of PCBs in sediments. As described in our prior

work,30–32 the score or loading plots obtained by PCA can reflect

similarities or dissimilarities between ambient and sources’ PCBs

homologues profiles. On the score plot, the data points with

different PCBs homologue profiles are located apart, while those

having convergent patterns are located closely. That is to say, if

the point of a certain source is located near the points of ambient

samples, the source might give a greater contribution to the

ambient samples. While if a point of source is located further

apart from the points of ambient samples, the samples are under

little influence of the source.

3. Results and discussion

3.1 Levels and composition of PCBs

PCBs were detected in all the sediment samples from 28 different

locations in two periods. The results were based on dry sediment

weight. As shown in Table 1, concentrations ofP

123PCBs (the

sum of all the 123 measured PCBs concentrations) in the sedi-

ments ranged from n.d. (non-detectable) to 126.49 ng g�1 dw,

with a average value of 27.67 ng g�1 dw. The highestP

123PCBs

concentration occurred at S16 in wet season, and the lowestP123PCBs occurred at S1 with all the PCBs n.d. for both periods.

In order to understand the pollution status of the investigated

watershed, some published studies on PCBs levels in sediments

from other areas were used to make a comparison (also listed in

Table 1). Compared with reported data in other countries, the

concentrations ofP

123PCBs in sediments of Fenhe reservoir and

upstream watershed were relatively higher than those reported in

coastal areas of Spain33 and the Han River of Korea.34 However,

the total concentrations of PCBs were lower than those in other

areas, such as Naples harbour in Southern Italy,35 Lake Michi-

gan in the USA,36 and water reservoirs in Slovakia.37 A

comparison was also carried out with several rivers and lakes in

This journal is ª The Royal Society of Chemistry 2012

China: theP

123PCBs in this work was higher than those in the

Wuhan reach of the Yangtze River17 and Dianchi Lake,38 while

lower than those in Fu River39 and Haihe River.23 Although the

numbers of congeners in these studies were different and the

levels were not compared like with like, to some extent,

the comparison can reflect the pollution status, as has been done

in some other work.12,17 In summary, the concentrations ofP123PCBs in the sediments of the investigated watershed were at

a medium level compared with those of other watersheds in other

countries and China.

As for the composition of PCBs, we discussed it based on

mono- to deca-PCBs, as usually done in other studies.12,13,17 The

percentage of each congener (averaged over the 28 sites and two

sampling periods) are shown in Fig. 2. The PCBs homologues

profiles in sediments vary depending on PCBs sources, age of the

contamination, and environmental conditions. The dominated

congeners in sediments of Fenhe reservoir and upstream water-

shed were lowly chlorinated congeners: tri-PCBs (34.29%), fol-

lowed by tetra-PCBs (24.05%) and penta-PCBs (12.41%),

accounting for more than 70% of theP

123PCBs concentration.

This might due to the fact that the major PCBs congeners

produced and used in China were tri-PCBs followed by tetra-

PCBs17 and it is easier for less chlorinated congeners to be

transported over long distances.12 The other congeners were in

the order of hexa-PCBs (9.98%) > hepta-PCBs (7.63%) > bi-

PCBs (6.41%) > octa-PCBs (4.69%) > mono-PCBs (0.55%) >

nona-PCBs (n.d.) and deca-PCBs (n.d.).

As listed in Table 1, the profile of PCBs homologues in present

study was generally consistent with the previous study results in

the Fu River of China,39 Haihe River of China23 and sea lots of

Trinidad,40 where tri- and tetra-PCBs were dominant, as well as

the PCBs profile of Beijing soils which was dominated by tetra-,

di- and tri-PCBs.12 However, it was a little different from that in

Wuhan reach of the Yangtze River in China where tri-PCBs

accounted for the smaller proportion17 and Alexandria Harbor in

Egypt which was dominated by tetra- to hepta-PCBs.41

3.2 Spatial distribution of PCBs

The concentrations of the individual PCBs homologues in the

wet and dry season for 28 sites are shown in Fig. 3. According

to Fig. 3, we can see that the lowestP

123PCBs concentrations

were detected at the upstream S1 station and the highest were

at the downstream S16 station, in both wet and dry seasons. S1

was at the headstream of the investigated watershed around

which there was no direct pollution source, which could explain

why theP

123PCBs concentrations were not detected there,

whilst S16 was located downstream of Langhe River, around

which the industrial and residential activities were heavy. In

addition, in the Fenhe reservoir,P

123PCBs concentrations

presented an obvious decreasing trend from S18 to S28 in both

periods. This might be due to the fact that more PCBs related

particles settled and accumulated in the upper parts of the

reservoir while less migrated to the middle and lower parts.

Furthermore, dilution, degradation and desorption of lower

homologues during transport and deposition may also be

influencing factors.

In addition, according to Fig. 3, it can easily be seen that the

two most dominant congeners in sediments at all 28 sites were

J. Environ. Monit., 2012, 14, 1256–1263 | 1259

Table 1 Comparison of sedimentP

PCBs levels (ng g�1 dw) in Fenhe reservoir and upstream watershed with those in other areas around the worlda

Location N Range (ng g�1 dw)Average (ngg�1 dw) Dominated homologues Reference

Fenhe reservoir and watershed, China 123 n.d.–126.49 27.67 tri- and tetra-PCBs This studyWuhan reach of the Yangtze River, China 39 1.2–45.1 9.2 tetra- and penta-PCBs 17Fu River, China 12 4.2–197.8 46.3 tri-, tetra-, and penta-CBs 39Haihe River, China 32 n.d.–253 66.8 tetra-, penta- and tri-PCBs 23Dianchi Lake, China 6 0.6–2.4 1.2 — 38Sea Lots, Trinidad 136 62–601 — tri- and tetra-PCBs 40Coastal area of Barcelona, Spain 12 2.33–44 — — 33Alexandria Harbor, Egypt 96 0.9–1211 — tetra- to hepta-PCBs 41Naples harbour, Southern Italy 38 10–899 — tetra- and penta- PCBs 35Han River, Korea 12 0.042–4.53 0.548 — 34Lake Michigan, USA 163 53–35000 7400 — 36Water reservoirs, Slovakia 12 20.4–2325 — tri- and hexa- PCBs 37

a N: Number of congeners measured.

Fig. 2 Composition of PCBs in sediments of Fenhe reservoir and

upstream watershed (%).

Publ

ishe

d on

08

Febr

uary

201

2. D

ownl

oade

d by

Uni

vers

ity o

f C

alif

orni

a -

Sant

a C

ruz

on 3

1/10

/201

4 07

:27:

02.

View Article Online

tri-PCBs and tetra-PCBs. The percentage of tri-PCBs (averaged

over the two periods) ranged from 30.03% (at S13) to 100% (at

S26, S27 and S28) and the percentage of tetra-PCBs ranged

from 0 (at S26, S27 and S28) to 27.85% (at S3). The sum of tri-

PCBs and tetra-PCBs accounted for more than 50% of theP123PCBs at all sites, which might be due to the fact that

Aroclor mixtures with lowly chlorinated congeners were the

most widely used PCBs mixtures in the investigated watershed,

as well as tri- and tetra-PCBs being the major PCB congeners

produced in China, as mentioned above. Noticeably, in the

Fenhe reservoir, it showed an increasing trend for the

percentages of lowly chlorinated congeners from S18 to S28.

This trend is likely due to the differences in the fate and

transport of the congeners; the highly chlorinated congeners

with selective retention water solubility and stability settle more

easily and so accumulate at the upper sites.40,42

3.3 Temporal variations of PCBs

From Fig. 3 it can also be found that the concentrations of

PCBs show obvious differences for the two periods. In the wet

season, the average concentrations ranged from n.d. to 126.49

ng g�1 dw, with an average value of 33.02 ng g�1 dw, while in

the dry season, the average concentrations ranged from n.d. to

1260 | J. Environ. Monit., 2012, 14, 1256–1263

87.41 ng g�1 dw, with an average value of 22.32 ng g�1 dw.

The T values of all of the PCB homologues andP

123PCBs

between the two periods were below 0.01, which suggests that

the differences are significant. Generally speaking, the PCBs

concentrations in the wet season were significantly higher than

those in the dry season. Precipitation and pollution sources are

considered to be the main factors causing fluctuations in water

quality. The temporal variations of PCBs in the two periods

might be mainly caused by the difference of precipitation in

these two periods. In the wet season, PCBs previously buried

in the surface soil of heavily contaminated sites and accumu-

lated in dry season were flushed into the river through surface

runoff due to the floods and heavy rains.4,16 While in the dry

season, with less precipitation and surface runoff, the

contaminants might selectively be accumulated in surface soil

instead of being washed into the river and depositing in the

sediments.

If PCBs are influenced dominantly by the same sources, they

would show similar homologues profiles. Thus, PCBs homo-

logues profiles of sediment samples could reflect the sources of

PCBs. In order to investigate whether the sources were similar in

wet and dry seasons, CD28,29 was used to study the temporal

difference of PCBs homologues profiles in sediment samples. In

this work, the value of CD is equal to 0.021, which could suggest

that the profiles of PCBs homologues for the two periods were

similar to each other, although the concentrations showed rela-

tively large variations.

As for the spatial distribution ofP

123PCBs in wet and dry

seasons, it can be seen in Fig. 3 that the patterns ofP

123PCBs’

spatial distribution were similar. Furthermore, a high value for

the correlation coefficient (0.999) was obtained, indicating that

the PCBs exhibited very similar spatial distribution in the two

periods.

3.4 Source analysis

In order to understand and control the PCBs pollution in Fenhe

reservoir and upstream watershed, it is of great significance to

identify the possible sources for the PCBs. In this study, PCAwas

performed on a normalized original matrix made up of 8 columns

(number of PCBs homologues, except for nona- and deca-PCBs

which were n.d.) and 61 rows (sum number of ambient samples

This journal is ª The Royal Society of Chemistry 2012

Fig. 3 Concentrations of the individual PCBs homologues in wet and dry seasons.

Fig. 4 Principal component plot of PCBs in Fenhe reservoir and

upstream watershed.

Publ

ishe

d on

08

Febr

uary

201

2. D

ownl

oade

d by

Uni

vers

ity o

f C

alif

orni

a -

Sant

a C

ruz

on 3

1/10

/201

4 07

:27:

02.

View Article Online

for 28 sites in two periods and 5 sources profiles). The first two

principal components (PCs) were extracted by PCA, explaining

50.6% and 24.9% of the total variance, respectively. The results

of the PCA are shown in Fig. 4.

From the score plot, it can be found that PCBs in most

sediment samples for both the periods were closely located,

indicating that they might originate from similar sources. It

agreed with the result analyzed by CD values. In addition, the

Aroclor 1016 and 1242 were also located near the sediment

samples, suggesting that the PCBs in Fenhe reservoir and

upstream watershed were mainly influenced by these two

potential sources. In contrast, Aroclor 1254, 1221 and 1043

were separated from the sample points, suggesting that their

PCBs homologues profiles were very different with those in the

investigated watershed and they presented little effect on the

PCBs in the sediment samples. Aroclor 1016 and 1242 are

dominated by tri- and tetra-PCBs. Therefore, the results of

PCA agree with actual situation that tri- and tetra-PCBs were

the dominant PCB congeners in the investigated watershed, as

mentioned above.

This journal is ª The Royal Society of Chemistry 2012 J. Environ. Monit., 2012, 14, 1256–1263 | 1261

Publ

ishe

d on

08

Febr

uary

201

2. D

ownl

oade

d by

Uni

vers

ity o

f C

alif

orni

a -

Sant

a C

ruz

on 3

1/10

/201

4 07

:27:

02.

View Article Online

3.5 Ecotoxicological concerns

Considering the toxicity and bioaccumulation property of PCBs,

as well as Fenhe River being a major source of water for irri-

gation and drinking, it is of great significance to evaluate the

potential risk of PCBs in sediments of Fenhe reservoir and

upstream watershed. So far, there is still no uniform standard

available to assess the biological effects of PCBs, but several

studies have been carried out and some useful indicators have

been provided. Among them, as shown in Fig. 3, the effect range

low (ERL),17,23,24 the threshold effects level (TEL) for freshwater

sediment,17,25 as well as lowest effect level (LEL),26 are the most

extensively applied guidelines in related studies. These indicators

would provide helpful guides to the management of PCBs in the

absence of environmental assessment criteria for PCBs in China.

The ERL (22.7 ng g�1 dw) value is intended to defineP

PCBs

concentration ranges that are rarely (P

PCBs < ERL) or occa-

sionally (P

PCBs > ERL) linked with adverse biological effects.24

For concentrations below the TEL (34.1 ng g�1 dw), the adverse

effects are negligible.25 LEL (70 ng g�1 dw) indicates a level of

sediment contamination that can be tolerated by the majority of

benthic organisms.26

For PCBs in sediments of Fenhe reservoir and upstream

watershed, the averageP

123PCBs in the wet season was higher

than the ERL but lower than the TEL and LEL; while the

averageP

123PCBs in the dry season were lower than all three

guidelines. This suggests that PCBs might cause adverse bio-

logical effects occasionally in wet season, but in dry season the

effects should be negligible. As for theP

123PCBs at each site in

the two periods, the comparisons of theP

123PCBs concentra-

tions with ERL, TEL and LEL can be found in Fig. 3. In the wet

season, 42.9%, 28.6% and 17.9% of the total 28 sites hadP123PCBs with concentrations greater than ERL, TEL and

LEL, respectively. In the dry season, 28.6%, 25.0% and 10.7% of

the 28 sites showedP

123PCBs concentrations above the ERL,

TEL and LEL, respectively. These findings could indicate that

PCBs in the sediments of the investigated watershed might lead

to occasional biological impacts, especially in the wet season.

4. Conclusion

The current status of PCBs in the Fenhe reservoir and upstream

watershed was investigated in this work. In conclusion, the

concentrations ofP

123PCBs in the sediments of the investigated

watershed were at a medium level compared with those of other

watersheds in China and other countries. The dominant conge-

ners were those that were less chlorinated: tri-PCBs, followed by

tetra-PCBs and penta-PCBs. For the spatial distribution, in the

Fenhe reservoir,P

123PCBs concentrations presented an obvious

decreasing trend, while it showed an increasing trend for the

percentages of less chlorinated congeners. As for the temporal

variation of PCBs, PCBs homologue profiles of sediment samples

and spatial distribution ofP

123PCBs for the two periods were

similar, although PCBs concentrations in the wet season were

significantly higher than those in the dry season. In addition, the

results of PCA suggest that PCBs in the Fenhe reservoir and

upstream watershed might be mainly influenced by Aroclor 1016

and 1242. Compared with ERL, TEL and LEL, PCBs in the

sediments of the investigated watershed may occasionally lead to

1262 | J. Environ. Monit., 2012, 14, 1256–1263

potential biological impacts, especially in the wet season.

Therefore, it is important to control the PCB contamination in

sediments by monitoring Aroclor 1016 and 1242 in Shanxi

Province.

Acknowledgements

This study is supported by the Youth Science Foundation of

Shanxi Province, China (NO. 2008021036-4) and the Funda-

mental Research Funds for the Central Universities.

References

1 S. Fu, H. X. Cheng, Y. H. Liu, Z. Z. Yang and X. B. Xu,Chemosphere, 2009, 74, 1477–1484.

2 L. Webster, M. Russell, P. Walsham, L. A. Phillips, I. Hussy,G. Packer, E. J. Dalgarno and C. F. Moffat, J. Environ. Monit.,2011, 13, 1288–1307.

3 K. Vorkamp, J. Strand, J. H. Christensen, T. C. Svendsen, P. Lassen,A. B. Hansen, M. M. Larsen and O. Andersen, J. Environ. Monit.,2010, 12, 1141–1152.

4 G. Salihoglu, K. Salihoglu, E. Aksoy and Y. Tasdemir, J. Environ.Manage., 2011, 92, 724–732.

5 UNEP, Preparation of an Internationally Legally Binding Instrumentfor Implementing International Action on Certain Persistent OrganicPollutants; UNEP/POPs/Inc.1/6, Nairobi, Kenya, 1998.

6 D. Lakshmanan, N. L. Howell, H. S. Rifai and L. Koenig,Chemosphere, 2010, 80, 100–112.

7 L. Linderholm, K. Jakobsson, T. Lundh, R. Zamir, M. Shoeb,N. Nahar and A. Bergman, J. Environ. Monit., 2011, 13, 2728–2734.

8 M. Raach, M. Lebeuf and E. Pelletier, J. Environ. Monit., 2011, 13,649–656.

9 J. Kobayashi, T. Sakurai and N. Suzuki, J. Environ. Monit., 2010, 12,647–653.

10 USEPA, PCBs: Cancer Dose-response Assessment and Application toEnvironmental Mixtures. U.S. Environmental Protection Agency,Washington, DC, 1996.

11 H. X. Liu, Q. F. Zhou, Y. W. Wang, Q. H. Zhang, Z. W. Cai andG. B. Jiang, Environ. Int., 2008, 34, 67–72.

12 S. Wu, X. Xia, L. Yang and H. Liu, Chemosphere, 2011, 82, 732–738.13 W. B. Ameur, S. Trabelsi, B. EL Bedoui and M. R. Driss, Bull.

Environ. Contam. Toxicol., 2011, 86, 539–544.14 M. Bazzanti, S. Chiavarini, C. Cremisini and P. Soldati, Environ. Int.,

1997, 23, 799–813.15 L. Webster, M. Russell, P. Walsham, L. A. Phillips, I. Hussy,

G. Packer, E. J. Dalgarno and C. F. Moffat, J. Environ. Monit.,2011, 13, 1288–1307.

16 C. T. Fu and S. C. Wu, Chemosphere, 2006, 62, 1786–1794.17 Z. Yang, Z. Shen, F. Gao, Z. Tang and J. Niu,Chemosphere, 2009, 74,

1522–1530.18 I. J. Allan and S. B. Ranneklev, J. Environ. Monit., 2011, 13, 2420.19 Y. Xing, Y. Lu, R. W. Dawson, Y. Shi, H. Zhang, T. Wang, W. Liu

and H. Ren, Chemosphere, 2005, 60, 731–739.20 K. Breivik, A. Sweetman, J. M. Pacyna and K. C. Jones, Sci. Total

Environ., 2002, 290, 181–198.21 Y. F. Li, T. Harner, L. Liu, Z. Zhang, N. Q. Ren, H. Jia, J. Ma and

E. Sverko, Environ. Sci. Technol., 2010, 44, 2784–2790.22 W. H. Li, Y. Z. Tian, G. L. Shi, C. S. Guo, X. Li and Y. C. Feng,

Ecotoxicol. Environ. Saf., 2012, 75, 198–206.23 L. Zhao, H. Hou, Y. Y. Zhou, N. D. Xue, H. Y. Li and F. S. Li,

Chemosphere, 2010, 78, 1285–1293.24 E. R. Long, D. D.MacDonald, S. L. Smith and F. D. Calder,Environ.

Manage., 1995, 19, 81–97.25 CCME (Canadian Council of Ministers of the Environment),

Canadian Council of Ministers of the Environment Canadiansediment quality guidelines for the protection of aquatic life:polychlorinated biphenyls (PCBs). In: Canadian EnvironmentalQuality Guidelines, Canadian Council of Ministers of theEnvironment: Winnipeg, Manitoba, 1999.

26 OntarioMinistry of Environment and Energy (OMEE),Guidelines forthe protection and management of aquatic sediment quality in Ontario,Toronto: Standards Development Branch, 1993.

This journal is ª The Royal Society of Chemistry 2012

Publ

ishe

d on

08

Febr

uary

201

2. D

ownl

oade

d by

Uni

vers

ity o

f C

alif

orni

a -

Sant

a C

ruz

on 3

1/10

/201

4 07

:27:

02.

View Article Online

27 J. J. Xu, Q. Guo and Y. Yang, The report of water environmentalcapacity and total pollution discharge in Fenhe watershed, ShanxiResearch Academy of Environmental Sciences, 2009.

28 V. Wongphatarakul, S. K. Friedlander and J. P. Pinto, Environ. Sci.Technol., 1998, 32, 3926–3934.

29 I. Hwang, P. K. Hopke and J. P. Pinto, Environ. Sci. Technol., 2008,42, 3524–3530.

30 G. L. Shi, Y. C. Feng, J. H. Wu, X. Li, Y. Q. Wang, Y. H. Xue andT. Zhu, Aerosol Air Qual. Res., 2009, 9, 309–315.

31 Z. Cao, Y. Wang, Y. Ma, Z. Xu, G. Shi, Y. Zhuang and T. Zhu, J.Hazard. Mater., 2005, 122, 51–59.

32 Y. C. Feng, G. L. Shi, J. H. Wu, Y. Q.Wang, T. Zhu and S. G. Dai, J.Air & Waste Manage. Assoc., 2007, 57, 164–171.

33 P. Castells, J. Parera, F. J. Santos and M. T. Galceran, Chemosphere,2008, 70, 1552–1562.

34 K. S. Kim, S. C. Lee, K. H. Kim,W. J. Shim, S. H. Hong, K. H. Choi,J. H. Yoon and J. G. Kim, Chemosphere, 2009, 75, 580–587.

This journal is ª The Royal Society of Chemistry 2012

35 M. Sprovieri, M. L. Feo, L. Prevedello, D. S. Manta, S. Sammartino,S. Tamburrino and E. Marsella, Chemosphere, 2007, 67, 998–1009.

36 A. Martinez, K. Norstr€om, K. Wang and K. C. Hornbuckle, Environ.Int., 2010, 36, 849–854.

37 E. Hiller, L. Zemanov�a, M. Sirotiak and L. Jurkovic, Environ. Monit.Assess., 2011, 173, 883–897.

38 X. Wan, X. Pan, B. Wang, S. Zhao, P. Hu, F. Li and B. Boulanger,Chemosphere, 2011, in press.

39 G. Hu, X. Luo, F. Li, J. Dai, J. Guo, S. Chen, C. Hong, B. Mai andM. Xu, J. Environ. Sci., 2010, 22(2), 176–183.

40 A. Mohammed, P. Peterman, K. Echols, K. Feltz, G. Tegerdine,A. Manoo, D. Maraj, J. Agard and C. Orazio, Mar. Pollut. Bull.,2011, 62, 1324–1332.

41 A. O. Barakat, M. Kim, Y. Qian and T. L. Wade, Mar. Pollut. Bull.,2002, 44, 1421–1434.

42 B. Ar�aoz and A. A. Viale, Rev. Argent. Microbiol., 2004, 36, 47–51.

J. Environ. Monit., 2012, 14, 1256–1263 | 1263