Solid-phase extraction combined with dispersive liquid-liquid microextraction for the determination for polybrominated diphenyl ethers in different environmental matrices.

Abstract

An analytical method, solid-phase extraction combined with dispersive liquid-liquid microextration (SPE-DLLME), was established to determine polybrominated diphenyl ethers (PBDEs) in water and plant samples. After concentration and purification of the samples in LC-C18 column, 1.0-mL elution sample containing 22.0 microL 1,1,2,2-tetrachloroethane was injected rapidly into the 5.0-mL pure water. After extraction and centrifuging, the sedimented phase was injected rapidly into gas chromatography with electron-capture detection (GC-ECD). For water samples, enrichment factors (EFs) are in the range of 6838-9405 under the optimum conditions. The calibration curves are linear in the range of 0.1-100 ng L(-1) (BDEs 28, 47) and 0.5-500 ng L(-1) (BDEs 100, 99, 85, 154, 153). The relative standard deviations (RSDs) and the limits of detection (LODs) are in the range of 4.2-7.9% (n=5) and 0.03-0.15 ng L(-1), respectively. For plant samples, RSDs and LODs are in the range of 5.9-11.3% and 0.04-0.16 microg kg(-1), respectively. The relative recoveries of well, river, sea, leachate, and clover samples, spiked with different levels of PBDEs, are 66.8-94.1%, 72.2-100.5%, 74.5-110.4%, 62.1-105.1%, 66.1-91.7%, 62.4-88.9%, and 64.5-83.2%, respectively. The results show that SPE-DLLME is a suitable method for the determination of PBDEs in water and plant samples.

Full text

Page 1

Journal of Chromatography A, 1216 (2009) 2220–2226
Contents lists available at ScienceDirect
Journal of Chromatography A
journal homepage: www.elsevier.com/locate/chroma
Solid-phase extraction combined with dispersive liquid–liquid microextraction
for the determination for polybrominated diphenyl ethers in different
environmental matrices
Xiujuan Liua,b, Jianwang Lib, Zhixu Zhaob, Wei Zhangc, Kuangfei Linc,
Changjiang Huangb,∗, Xuedong Wangb,∗
aKey Laboratory of Pesticide & Chemical Biology of Ministry of Education, Central China Normal University, Wuhan 430079, China
bSchool of Environmental Science and Public Health, Wenzhou Medical College, Wenzhou 325035, China
cState Environmental Protection Key Laboratory of Environmental Risk Assessment and Control on Chemical Process,
East China University of Science and Technology, Shanghai 200237, China
a r t i c l e i n f o
Article history:
Received 17 October 2008
Received in revised form 16 December 2008
Accepted 31 December 2008
Available online 9 January 2009
Keywords:
Solid-phase extraction combined with
dispersive liquid–liquid microextration
(SPE–DLLME)
Water sample
Plant sample
Gas chromatography–electron capture
detection (GC–ECD)
Polybrominated diphenyl ethers (PBDEs)
a b s t r a c t
An analytical method, solid-phase extraction combined with dispersive liquid–liquid microextration
(SPE–DLLME), was established to determine polybrominated diphenyl ethers (PBDEs) in water and plant
samples. After concentration and purification of the samples in LC-C18 column, 1.0-mL elution sam-
ple containing 22.0?L 1,1,2,2-tetrachloroethane was injected rapidly into the 5.0-mL pure water. After
extraction and centrifuging, the sedimented phase was injected rapidly into gas chromatography with
electron-capture detection (GC–ECD). For water samples, enrichment factors (EFs) are in the range of
6838–9405undertheoptimumconditions.Thecalibrationcurvesarelinearintherangeof0.1–100ngL−1
(BDEs 28, 47) and 0.5–500ngL−1(BDEs 100, 99, 85, 154, 153). The relative standard deviations (RSDs) and
the limits of detection (LODs) are in the range of 4.2–7.9% (n=5) and 0.03–0.15ngL−1, respectively. For
plant samples, RSDs and LODs are in the range of 5.9–11.3% and 0.04–0.16?gkg−1, respectively. The rela-
tive recoveries of well, river, sea, leachate, and clover samples, spiked with different levels of PBDEs, are
66.8–94.1%,72.2–100.5%,74.5–110.4%,62.1–105.1%,66.1–91.7%,62.4–88.9%,and64.5–83.2%,respectively.
The results show that SPE–DLLME is a suitable method for the determination of PBDEs in water and plant
samples.
© 2009 Elsevier B.V. All rights reserved.
1. Introduction
Polybrominated diphenyl ethers (PBDEs) are flame retardant
chemicalsthatareoftendissolvedinormixedwithdifferentindus-
trial products such as paints, plastic, and textiles. They can escape
from the surface of manufactured products and release to the
environment. Therefore, the major PBDE congeners 47 (2,2?,4,4?-
tetra-BDE), 99 (2,2?,4,4?,5-tetra-BDE), 100 (2,2?,4,4?,6-tetra-BDE)
have been often detected in house dust [1,2], water [3], soil,
sediment [4,5], marine organisms [6,7], food [8,9], and so on. Fur-
thermore, toxicological tests indicate that PBDE congeners may
causeliverandthyroidtoxicityinwildlifeandinhumanssincethey
are proved to have endocrine disrupting properties [10,11].
As the worldwide environmental contaminants, different envi-
ronmental monitoring programs have been established to find
the traces of PBDEs. Known methods for the detection of PBDEs
∗Corresponding author. Tel.: +86 577 86689733; fax: +86 577 86699135.
E-mail address: zjuwxd@yahoo.com.cn (X. Wang).
in different matrices include stir bar sorptive extraction [12,13],
matrix solid-phase dispersion (MSPD) [14], microwave-assisted
extraction (MAE) [15], solid-phase microextraction (SPME) [16],
and dispersive liquid–liquid microextraction (DLLME) [17]. Among
them, DLLME is a novel microextraction technique, which has been
successfully used for the extraction and determination of poly-
chlorinatedbiphenyl[18],polycyclicaromatichydrocarbons(PAHs)
[19], organophosphorus pesticides (OPPs) [20], and some kinds of
metal and non-metal in water [21,22,23]. Simplicity of operation,
rapidity, low sample volume, low cost, and high enrichment factor
(EF) are some advantages of DLLME. In this method, the optimized
mixture of extraction solvent and disperser solvent is injected into
theconstantvolumeofaqueoussamplerapidlybysyringe,andthen
a cloudy solution is formed. The analyte in the sample is extracted
into the extraction solvent which following separated by centrifu-
gation. The sedimented phase is determined by chromatography or
spectrometry.
Solid-phase extraction (SPE), a traditional sample preparation
technique, has the advantages of reducing organic solvent con-
sumption and processing time, and high reproducibility relative to
liquid–liquid extraction (LLE). It is widely used for the isolation and
0021-9673/$ – see front matter © 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.chroma.2008.12.092

Page 2

X. Liu et al. / J. Chromatogr. A 1216 (2009) 2220–2226
2221
concentration of target analytes from matrices, such as soil [24],
sediment [25], lipid [26], human serum [27], and water [28].
To the best of our knowledge, the previous literatures on the
application of DLLME mainly focused on the determination of
organic compounds in liquid samples. However, the determination
insolidsamplesbyDLLMEhasreceivedonlylimitedattention.Zhao
et al. [29] successfully applied DLLME to detect OPPs in cucum-
ber and watermelon samples. Our research group also investigated
the determination of three pesticides: malathion, cypermethrin,
andlambda-cyhalothrininsoilandsedimentsamplesusingDLLME
technique[30].Itisworthytonotethattheenrichmentfactorusing
DLLME is often in the range of 50–1000, which still cannot be sat-
isfied for the requirement of the ultra trace residue analysis. In
principle, SPE combined with DLLME can provide a solution to this
problem.Notonlydoesthecombinationresultinaveryhighenrich-
ment factor (up to about 5000–10,000), but it can be also used
in complex matrices. Unfortunately, the application of SPE–DLLME
is only reported to date in aqueous samples [31], while no data
are available in solid matrices. Therefore, the aim of this study is
to establish a preconcentration method in water and plant sam-
ples, which can not only isolate the analytes from environmental
matrices,butalsoreducethematrixeffectseffectively.Thepractical
applicability of the method was investigated for PBDEs extraction
and determination in well, river, sea, leachate water samples, and
hyacinth and clover plant samples around solid waste site. As far as
our information goes, these studies may be the first report describ-
ing the application of SPE–DLLME as a preconcentration technique
for the analysis of PBDEs congeners from water and plant sam-
ples.
2. Experimental
2.1. Reagent and standards
2,4,4?-Tribrominated diphenyl (BDE 28), 2,2?,4,4?-tetrabro-
minated diphenyl ether (BDE 47), 2,2?,3,4,4?-pentabrominated
diphenyl ether (BDE 85), 2,2?,4,4?,5-pentabrominated diphenyl
ether (BDE 99),2,2?,4,4?,6-pentabrominated diphenyl ether (BDE
100), 2,2?,4,4?,5,5?-hexbrominated diphenyl ether (BDE 153),
2,2?,4,4?,5,6?-hexbrominated diphenyl ether (BDE 154) were pur-
chased from Accustandard (New Haven, CT, USA). Each compound
was dissolved in acetonitrile to make a 50mgL−1stock solution.
The working standard solutions were prepared by serial dilutions
of the stock solution with ultra Milli-Q water (Millipore, Molsheim,
France) freshly prior to analysis. The HPLC-grade acetonitrile and
methanol were obtained from Merck Company (Darmstadt, Ger-
many). The other solvents in this experiment (acetone, ethanol,
dichloromethane (DCM), chloroform, carbon tetrachloride, 1,1,2,2-
tetrachloroethane, n-hexane, tetrahydrofuran (THF)) were all of
analytical grade and redistilled prior to use. The plants (hyacinth
and clover) and leachate samples were collected from solid waste
sitelocatedatTaizhouCity,China.Thewatersampleswerecollected
from the region far away from solid waste site (Taizhou, China).
2.2. Instrumentation
Solid-phase extraction equipment and column (SUPEL LC-C18)
were purchased from Supelco (Bellefonte, PA, USA).
PBDEs analysis was carried out by an Agilent 6890 gas chro-
matograph (Agilent Technologies, Wilmington, DE, USA) with ECD
system. The GC system was coupled to a HP-5 capillary column
(30 m×0.25mm I.D., 0.25?m film thickness, Agilent). The injector
temperature was held at 290◦C, and the injection was performed
in the splitless mode. The GC oven temperature was programmed
from 110◦C (4min) to 220◦C at 30◦C/min; from 220◦C (4min) to
280◦C(5min)attherateof3◦C/min.Nitrogen(purity99.999%)was
employedascarriergasatconstantcolumnflow(1.6mLmin−1).The
split flow was set at 60mLmin−1.
2.3. Sample preparation
Water samples were filtered through the 0.45-?m membranes
and stored in amber bottles at 4◦C until analysis.
Afterbeingfreeze-dried,eachdriedplantwashomogenizedina
stainlesssteelblender.1gofdriedplantwasmixedwith4gofanhy-
drous sodium sulfate, respectively. The sample was extracted with
10mL hexane/acetone (1:1, v/v) for 30min by shaking on a rotary
shaker (EYELA, Tokyo, Japan). The extract was collected, concen-
trated to dryness under gentle nitrogen evaporation apparatus and
then redissolved in 2mL acetonitrile. The final extract dissolved in
acetonitrile was diluted with Milli-Q water to 100mL and further
subjected to clean up with LC-C18 column.
2.4. SPE–DLLME procedure
SPEofPBDEsfromwaterandplantsampleswascarriedoutusing
6mL (1g) of the Supelclean LC-C18 SPE columns. 2% acetonitrile
(v/v) was added to the spiked samples before performing SPE pro-
cedures. The SPE column bed was conditioned with 2.0mL DCM,
5mL methanol and water. The water samples passed through the
column at a flow rate of 10mLmin−1. Then the columns were dried
under vacuum in a manifold system (Supelco) for 10min. The tar-
get analytes were subsequently eluted with 2.0mL n-hexane and
the eluent solution was collected and concentrated under a gentle
nitrogen flow. The residue was redissolved in 1.0mL acetonitrile,
which was used as disperser solvent in the subsequent DLLME pro-
cedures.
5.0-mL aqueous solution was placed in a 10-mL screw cap glass
test tube with conical bottomed. 1.0-mL acetonitrile (disperser
solvent) containing 22.0?L 1,1,2,2-tetrachloroethane (extraction
solvent) was injected rapidly into the aqueous solution with a 50.0-
?L microsyinge (Agilent Technologies, Wilmington, DE, USA). A
cloudy solution, resulting from the dispersion of the fine 1,1,2,2-
tetrachloroethane droplets in the aqueous solution, was formed in
the test tube. After extracting for a few seconds, the mixture was
then centrifuged for 3min at 3000rpm. The dispersed fine 1,1,2,2-
tetrachloroethane droplets were deposited at the bottom of the
conical test tube. The sedimented phase was removed by 25.0-?L
microsyinge and concentrated under a gentle nitrogen flow. The
solvent was redissolved in 20.0-?L n-hexane. 1.0?L of n-hexane
solution was injected into GC for analysis.
3. Results and discussion
3.1. Optimized SPE conditions
SPE column is a key factor to the isolation and purification effi-
ciency of the target analytes. In this investigation, SUPEL LC-C18
column was chosen to isolate and concentrate PBDEs because of its
widely adapting range from non-polarity to moderate polarity in
aqueous solution. Additionally, other SPE conditions including the
flowrateofsamplesolution,thebreakthroughvolume,thesaltcon-
centration, organic solvent percentage, and eluting solvent, were
also tested in order to achieve acceptable recoveries upon extrac-
tion of large volumes (200mL) of Milli-Q water samples spiked
with the PBDEs analytes. As a pretreatment step, water sample,
containing 0.10?gL−1of PBDEs, was loaded.
3.1.1. Effect of the flow rate of the sample solution
The flow rate of the sample solution through LC-C18 column
controls the analytical time and affects the effective retention of

Page 3

2222
X. Liu et al. / J. Chromatogr. A 1216 (2009) 2220–2226
Fig. 1. The effect of the flow rate on the PBDEs recovery from SPE. Extraction con-
ditions: water sample volume, 100mL; eluent solvent, n-hexane; eluent solvent
volume, 2mL; no salt; concentration of PBDEs, 0.10?gL−1.
the analytes. The flow rate must be high enough to shorten the
analytical time. On the other hand, it also must be slow enough to
performaneffectiveretentiontoanalytes.Therefore,theoptimized
flow rate should consider the previous two factors. Under the fol-
lowing constant conditions (water sample volume, 100mL; elution
solvent, n-hexane; eluent solvent volume, 2mL; no salt), the effect
of flow rate on recoveries of seven PBDE congeners was investi-
gated in the flow rate range of 5–15mLmin−1. As can be seen from
Fig. 1, at the flow rate of 10 and 12mLmin−1, the PBDE congeners
recoverieswereslightlyhigherthanthatofothers.Additionally,the
control of the flow rate of 12mLmin−1was more difficult than that
of 10mLmin−1. As a result, 10mLmin−1was used as the optimized
flow rate.
3.1.2. Effect of the breakthrough volume
Under thefollowing
10mLmin−1; elution solvent, n-hexane; eluent solvent vol-
ume, 2mL; no salt), different volumes (100, 200, 300, 400 and
500mL) of the standard solution of the analyte mixture passed
through the LC-C18 column. The breakthrough volume influence
of the sample solution from the LC-C18 column on the PBDEs
recovery is shown in Fig. 2. The result showed that the different
sample volumes did not affect the recoveries of analytes obviously.
constant conditions(flow rate,
Fig. 2. The effect of the breakthrough volume on the PBDEs recovery from SPE.
Extraction conditions: flow rate, 10mLmin−1; eluent solvent, n-hexane; eluent sol-
vent volume, 2mL; no salt; concentration of PBDEs, 0.10?gL−1.
Fig. 3. The effect of the elution solvent type on the PBDEs recovery from SPE.
Extractionconditions:watersamplevolume,200mL;flowrate,10mLmin−1;eluent
solvent volume, 2mL; no salt; concentration of PBDEs, 0.10?gL−1.
Considering the analytical time and trace level of PBDEs in natural
environment, 200mL was used as the optimized breakthrough
volume.
3.1.3. Effect of ionic strength
Under the following constant conditions (water sample volume,
200mL; flow rate, 10mLmin−1; elution solvent, n-hexane; eluent
solvent volume, 2mL), the different NaCl concentrations (0–10%,
w/v) of the sample solution were investigated in order to study the
ionic effect. The results showed the different salt concentrations
had not significant effect on the extraction recoveries of PBDEs.
Therefore, NaCl was not added in this method.
3.1.4. Effect of the elution solvent type
The elution solvent plays an important role in the recovery of
analytes. Under the following constant conditions (water sample
volume, 200mL; flow rate, 10mLmin−1; eluent solvent volume,
2mL; no salt), n-hexane, n-hexane/DCM (3:2, v/v), DCM, THF, ace-
tone,acetonitrile,werestudiedforthepurpose.Theresultsshowed
that the recoveries for THF, acetone, acetonitrile, were lower than
those of other elution solvents, such as n-hexane, DCM and their
mixture (in Fig. 3). In addition, the mean recovery of seven kinds of
PBDE congeners for n-hexane was higher than those for DCM and
n-hexane/DCM. As the result, n-hexane was selected as the elution
solvent in the subsequent experiment.
3.1.5. Effect of the concentration of acetonitrile
Someresearcheshavereportedthattheconcentrationoforganic
solvent can affect the efficient recovery of PAHs [32,33,34]. In our
study, the concentration of the organic solvent was also considered
because the PBDEs residue extracted from plant sample was first
dissolved in consolute solvent (acetonitrile), and then made up to
200mL by Milli-Q water. Under the following constant conditions
(water sample volume, 200mL; flow rate, 10mLmin−1; eluent sol-
vent, n-hexane; eluent solvent volume, 2mL; no salt), the effect
of the concentration of acetonitrile on the recoveries is shown in
Fig. 4. It was obvious that the concentration of acetonitrile (2%, v/v)
increased the recovery of penta-BDEs (BDE 85, 99, 100) and hex-
BDEs (BDE 153, 154). This might be due to the higher solubility of
PBDEs in water at higher concentration of acetonitrile. Therefore,
the appropriate concentration of acetonitrile was 2% (v/v).
In total, the optimized SPE conditions were as follows: water
sample volume, 200mL; flow rate, 10mLmin−1; elution solvent,

Page 4

X. Liu et al. / J. Chromatogr. A 1216 (2009) 2220–2226
2223
Fig. 4. The effect of the concentration of acetonitrile on the PBDEs recovery from
SPE. Extraction conditions: water sample volume, 200mL; flow rate, 10mLmin−1;
eluent solvent, n-hexane; eluent solvent volume, 2mL; no salt; concentration of
PBDEs, 0.10?gL−1.
n-hexane; elution solvent volume, 2mL; the concentration of ace-
tonitrile, 2% (v/v); without addition of salt. The sample solution
passingthroughtheLC-C18columnundertheoptimizedconditions
was subsequently concentrated by DLLME.
3.2. Optimized DLLME conditions
Theparameters,affectingtheDLLMEprocedure,suchasthetype
of extraction and disperser solvent as well as their volume and
the salt addition, were optimized. For this purpose, the aqueous
solution containing 1.0?gL−1of PBDEs was used.
3.2.1. Effect of type and volume of the extraction solvent
Careful attention should be paid to the selection of the extrac-
tion solvent. The extraction solvent must have some properties,
such as higher density than water, high extraction capability of
the analytes, and low solubility in water. Considering the previ-
ous factors, carbon tetrachloride (CCl4), 1,1,2,2-tetrachloroethane
(C2H2Cl4), dichloromethane (CH2Cl2), chloroform (CHCl3), was
examined in this study. A series of sample solutions were tested
using 1.0mL acetonitrile, containing different volumes of the
extraction solvents to achieve a 20?L volume of the sedimented
phase. It had been reported that dichloromethane could not
form the two-phase with methanol, acetonitrile, ethanol and
acetone and chloroform could form the two-phase only with
acetonitrile [35]. In this investigation, we found that carbon
tetrachloride (40?L) and 1,1,2,2-tetrachloroethane (22?L) were
suitable for the extraction solvent. As can be seen from Table 1,
1,1,2,2-tetrachloroethane had higher extraction recovery than
Table 1
Efficiency of different extraction solvents evaluated for extraction of PBDEs by
DLLME.a.
Compounds Recovery (%)
1,1,2,2-tetrachloroethane
mean±SD (n=3)
78.6 ± 1.7
89.5 ± 1.8
71.6 ± 1.4
70.4 ± 3.6
76.6 ± 3.2
86.7 ± 1.8
81.2 ± 4.3
Carbon tetrachloride
mean±SD (n=3)
46.5 ± 1.5
46.7 ± 4.5
43.2 ± 3.5
41.3 ± 0.9
38.3 ± 3.8
40.1 ± 3.7
37.3 ± 3.1
BDE 28
BDE 47
BDE 100
BDE 99
BDE 85
BDE 154
BDE 153
aExtraction conditions: water sample volume, 5.00mL; disperser solvent
(acetonitrile)volume,1.00mL; extraction
tetrachloroethane,40?L carbontetrachloride;
20±0.5?L; room temperature; concentration of each PBDEs, 1.0?gL−1.
solventvolume,22?L
phase
1,1,2,2-
volume,sedimented
Fig. 5. Effect of the volume extraction solvent (C2H2Cl4) on the enrichment factor of
PBDEs obtained from DLLME. Extraction conditions: water sample volume, 5.00mL;
disperser solvent (acetonitrile) volume, 1.00mL; room temperature; concentration
of each PBDEs, 1.0?gL−1.
that of carbon tetrachloride. Therefore, 1,1,2,2-tetrachloroethane
(C2H2Cl4) was selected as the extraction solvent in the subsequent
experiment.
In order to examine the effect of the extraction solvent
volume, 1mL acetonitrile containing different volumes of 1,1,2,2-
tetrachloroethane (22, 32, 42 and 52?L) was subjected to the
same DLLME procedures. By increasing the volume of 1,1,2,2-
tetrachloroethane from 22 to 52?L, the volume of sedimented-
phase increased from 20 to 90?L. As a result, EF decreased with
increasing the volume of 1,1,2,2-tetrachloroethane (Fig. 5). For the
seven PBDE congeners, the higher EF values were all obtained at
low volume (22?L) of the extraction solvent. Therefore, 22?L of
Table 2
Efficiency of different disperser solvents evaluated for extraction of PBDEs by DLLME.a.
Compounds Recovery (%)
CH3CN mean±SD (n=3)
78.6 ± 1.7
89.5 ± 1.8
71.6 ± 1.4
70.4 ± 3.6
76.6 ± 3.2
86.7 ± 1.8
81.2 ± 4.3
CH3CH2OH mean±SD (n=3)
51.5 ± 5.5
62.2 ± 4.8
64.6 ± 4.9
54.5 ± 4.2
50.6 ± 2.2
54.0 ± 3.8
49.1 ± 3.9
CH3OH mean±SD (n=3)
64.6 ± 7.3
77.7 ± 7.7
75.1 ± 6.4
62.0 ± 6.1
58.6 ± 4.0
60.2 ± 3.5
53.9 ± 5.2
CH3COCH3mean±SD (n=3)
55.7 ± 6.9
69.8 ± 5.4
54.9 ± 3.2
66.2 ± 1.8
54.3 ± 3.5
58.3 ± 1.9
47.2 ± 3.2
BDE 28
BDE 47
BDE 100
BDE 99
BDE 85
BDE 154
BDE 153
aExtraction conditions: water sample volume, 5.00mL; disperser solvent volume (acetonitrile, ethanol, methanol, acetone), 1.00mL; extraction solvent volume (1,1,2,2-
tetrachloroethane), 22, 41, 41, 26?L; sedimented phase volume, 20±0.5?L; room temperature; concentration of each PBDEs, 1.0?gL−1.

Page 5

2224
X. Liu et al. / J. Chromatogr. A 1216 (2009) 2220–2226
Table 3
Quantitative results of SPE–DLLME and GC–ECD of PBDEs from samples.
Compounds RSDa(%) n=5RSDb(%) n=5EFa
LRa(ngL−1)LRb(?gkg−1)
r2a
r2b
LODa(ngL−1) LODb(?gkg−1)
PBDE 28
PBDE 47
PBDE 100
PBDE 99
PBDE 85
PBDE 154
PBDE 153
4.2
4.3
6.2
7.9
6.8
5.6
5.4
9.9
5.9
7.8
9.1
8.6
11.3
6.0
8047
7936
7462
8015
9405
6838
7164
0.1–100
0.1–100
0.5–500
0.5–500
0.5–500
0.5–500
0.5–500
0.1–20
0.1–20
0.2–20
0.2–20
0.2–20
0.2–20
0.2–20
0.9980
0.9978
0.9994
0.9997
0.9992
0.9997
0.9995
0.9978
0.9982
0.9974
0.9989
0.9979
0.9984
0.9996
0.03
0.06
0.09
0.12
0.15
0.12
0.13
0.05
0.04
0.08
0.16
0.11
0.08
0.16
aWater sample.
bPlant sample.
1,1,2,2-tetrachloroethane was selected as the volume of extraction
solvent.
3.2.2. Effect of type and volume of disperser solvent
The main criterion for selecting the disperser solvent is its
miscibility with the extraction solvent and the aqueous sample.
For this purpose, different solvents such as acetonitrile, ethanol,
methanol and acetone were examined. A series of sample solutions
were tested using 1.0mL disperser solvent, containing different
volumes of the extraction solvents to achieve a 20-?L volume
of the sedimented phase. Therefore, 22, 41, 41 and 26?L 1,1,2,2-
tetrachloroethane was added to 1.0mL of acetonitrile, methanol,
ethanol and acetone, respectively. The result listed in Table 2 indi-
cated that acetonitrile exhibited the higher extraction efficiency.
Thus, acetonitrile was chosen as the dispersive solvent for subse-
quent experiment.
In order to examine the effect of disperser solvent volume, the
volume of the sedimented phase was kept constant (20?L) and the
volume of acetonitrile and 1,1,2,2-tetrachloroethane was changed,
simultaneously. The different volumes of acetonitrile (0.5, 1.0, 1.5
and 2.0mL) were in concomitant with the corresponding volumes
of 25, 22, 23, 27?L of 1,1,2,2-tetrachloroethane, respectively. It was
obviousfromFig.6that1.0mLacetonitrilehasslightlyhigherrecov-
erythanthatofothers.Therefore,1.0mLwasselectedasthevolume
of acetonitrile.
3.2.3. Effect of salt
The influence of ionic strength was evaluated at 0–6% (w/v)
NaCl levels while other parameters were kept constant. The exper-
Fig. 6. Effect of the volume disperser solvent (CH3CN) on the recovery of PBDEs
obtained from DLLME. Extraction conditions: water sample volume, 5.00mL; dis-
perser solvent (acetonitrile) volume, 0.5, 1.00, 1.25, 1.50mL; extraction solvent
(C2H2Cl4) volume, 25, 22, 23, 27?L; sedimented phase volume, 20±0.5?L; room
temperature; concentration of each PBDEs, 1.0?gL−1.
imental result showed that salt addition had no significant effect
on extraction recovery, which was in comparable with the con-
clusions by other researches [35,36]. Therefore, all the following
experiments were carried out without adding salt.
3.3. Quantitative aspects
As summarized in Table 3, the calibration curve was obtained
under the optimized SPE–DLLME–GC–ECD conditions. For water
samples,thelinearityofcalibrationcurvewasobservedintherange
of 0.1–100ngL−1for BDE 28 and 47, and 0.5–500ngL−1for BDE
100, 99, 85, 154 and 153. The coefficients or correlation (r2) ranged
from 0.9978 to 0.9997. The enrichment factors of PBDEs were high
and ranged from 6838 to 9405. The precision of the method was
evaluated by carrying out five independent measurements of the
studiedcompoundsat1.0ngL−1.Theresultshowedthattherelative
deviations (RSDs) ranged from 4.2% to 7.9%. The limits of detec-
tion (LODs), based on signal-to-noise ratio (S/N) of 3, ranged from
0.030 to 0.15ngL−1. However, for plant samples, linearity of plant
sample evaluated by spiking PBDEs was observed in the range of
0.1–20?gkg−1for BDE 28 and 47, and 0.2–20?gkg−1for BDE 100,
99, 85, 154 and 153. The r2ranged from 0.9974–0.9996 RSDs and
LODs were in the range of 5.9–11.3% and 0.04–0.16?gkg−1, respec-
tively.
3.4. Analysis of natural water and plant samples
The proposed SPE–DLLME technique was applied for the deter-
mination of PBDEs in several water and plant samples to elucidate
theapplicabilityandreliabilityofthismethod.Thedetectionresults
showed the seven PBDE congeners all at below detectable level
(<0.03ngL−1and0.04?gkg−1)inrealwaterandplantsamples.The
spiking recoveries of the target PBDEs in the real samples at the
different concentration levels are summarized in Tables 4 and 5.
In the fortification level of 1ngL−1, the relative recoveries were
between 66.8% and 94.1% in water samples. In the fortification level
of 0.2?gkg−1, the relative recoveries ranged from 66.1% to 91.7%
in plant samples. The previous results demonstrated that the pro-
posed SPE–DLLME method could be used in trace PBDEs analysis
of the environmental water and plant samples. Fig. 7 depicts the
attained chromatograms from the sea water (A), leachate (B), leaf
of clover (C) and the spiked water samples (D) at the concentration
level of 100ngL−1for BDE 28, 47, 100, 85, 153,154, 200ngL−1for
BDE 99.
3.5. Comparison of SPE–DLLME with other methods
This proposed SPE–DLLME technique was compared with other
published methods such as SPME–GC–ECD, DLLME–HPLC–VWD,
SDME–GC–ECD, etc. The respective LOD, EF value and RSD of each
method are summarized in Table 6. In terms of LOD, SPE–DLLME
has the one order of magnitude lower than other methods. Addi-
tionally, the EF value of SPE–DLLME is about 10–30 times higher

Page 6

X. Liu et al. / J. Chromatogr. A 1216 (2009) 2220–2226
2225
Table 4
Relative recoveries and standard deviations of PBDEs from water samples.
Compounds Well waterRiver waterSea waterleachate
Added
(?gL−1)
Found (RSD,
n=5)
(?gL−1)
Relative
recovery (%)
Added
(?gL−1)
Found (RSD,
n=5)
(?gL−1)
Relative
recovery (%)
Added
(?gL−1)
Found (RSD,
n=5)
(?gL−1)
Relative
recovery (%)
Added
(?gL−1)
Found (RSD,
n=5)
(?gL−1)
Relative
recovery (%)
BDE 28
BDE 100
BDE 99
BDE 85
BDE 154
BDE 153
0.0010
0.0010
0.0010
0.0010
0.0010
0.0010
0.000876 (0.075)
0.000838 (0.074)
0.000942 (0.048)
0.000803 (0.085)
0.000839 (0.101)
0.000876 (0.061)
87.6
83.8
94.1
66.8
70.4
87.6
0.0100
0.0100
0.0100
0.0100
0.0100
0.0100
0.00938 (0.074)
0.00878 (0.035)
0.0101 (0.049)
0.00891 (0.052)
0.00763 (0.049)
0.00722 (0.057)
93.8
87.7
100.5
89.1
76.3
72.2
0.1000
0.1000
0.1000
0.1000
0.1000
0.1000
0.0985 (0.042)
0.0887 (0.041)
0.0110 (0.082)
0.0814 (0.022)
0.0842 (0.069)
0.0745 (0.048)
98.5
88.7
110.4
81.4
84.2
74.5
0.0100
0.0100
0.0100
0.0100
0.0100
0.0100
0.00897 (0.077)
0.00979 (0.055)
0.0105 (0.067)
0.00771 (0.113)
0.00732 (0.0457)
0.00621 (0.103)
89.7
97.9
105.1
77.1
73.2
62.1
Table 5
Quantitative results of SPE–DLLME and GC–ECD of PBDEs from plant samples.
CompoundsLeaf of clover
Added (?gkg−1)Found (RSD n=5)
(?gkg−1)
Relative recovery
(%)
Added (?gkg−1)Found (RSD n=5)
(?gkg−1)
Relative recovery
(%)
Added (?gkg−1)Found (RSD n=5)
(?gkg−1)
Relative recovery
(%)
BDE 28
BDE 47
BDE 100
BDE 99
BDE 85
BDE 154
BDE 153
0.2000
0.2000
0.2000
0.2000
0.2000
0.2000
0.2000
0.1734 (0.037)
0.1834 (0.091)
0.1356 (0.075)
0.1572 (0.12)
0.1430 (0.093)
0.1378 (0.042)
0.1321 (0.026)
86.7
91.7
67.8
78.6
71.5
68.9
66.1
2.0000
2.0000
2.0000
2.0000
2.0000
2.0000
2.0000
1.6054 (0.042)
1.7780 (0.050)
1.2484 (0.032)
1.5942 (0.064)
1.4068 (0.051)
1.4448 (0.079)
1.2776 (0.041)
80.3
88.9
62.4
79.7
70.3
72.2
63.9
20.0000
20.0000
20.0000
20.0000
20.0000
20.0000
20.0000
14.2890 (0.073)
15.2920 (0.064)
13.6916 (0.10)
16.6302 (0.12)
14.0860 (0.057)
13.1260(0.092)
13.7110 (0.071)
71.4
76.5
68.5
83.2
70.4
64.5
68.6

Page 7

2226
X. Liu et al. / J. Chromatogr. A 1216 (2009) 2220–2226
Fig. 7. The chromatograms of the sea water (A), leachate (B), the plant (C) and the
spiked water samples (D) at the concentration level of 100ngL−1for BDE 28, 47,
100, 85,153, 154, 200ngL−1for BDE 99, obtained using SPE–DLLME combined with
GC–ECD. Extraction conditions: water sample volume, 200mL; plant sample mass,
1g; sample flow rate, 10mLmin−1; eluent solvent, n-hexane; the concentration of
acetonitrile in water, 2%; disperser solvent (acetonitrile) volume, 1.0mL; extrac-
tionsolvent(C2H2Cl4),22?L;aqueoussolutionforDLLME,5.00mL;sedimentphase
volume, 20±0.5?L; room temperature.
Table 6
Comparison of SPE–DLLME with SDME, SPME and DLLME for the determination of
PBDEs in water samples.
MethodsLOD (ngL−1) RSD (%) EF References
SDME–HPLC–VWD
SPME–GC–ECD
DLLME–HPLC–VWD
SPE–DLLME–ECD
7004.4
6.9–8.8
3.8–6.3
4.2–7.9
10.6
616–1756
268–305
6838–9405
[37]
[16]
[17]
Represented
method
3.6–8.6
12.4–55.6
0.30–1.5
than those of other three kinds of published methods. In general,
this proposed method has a similar extraction performance as oth-
ers, while exhibits higher EF value and lower LOD than those of
previously published techniques.
4. Conclusions
WehavefoundthatSPE–DLLME–GC–ECDisanaccurateandreli-
able method for the PBDEs determination in environmental water
and plant samples. The analytical technology offered numerous
advantages such as ease of operation, ultra preconcentration factor,
and low detection limit. Accordingly, the proposed method pos-
sesses great potential in the analysis of ultra trace compounds in
real water and plant samples.
Acknowledgements
This work was jointly funded by Key Project of Natural Sci-
enceFoundationofZhejiangProvince(Z2080266),NationalNatural
Science Foundation of China (20677015) and International Cooper-
ation Project of Wenzhou City (H20080063).
References
[1] J. Regueiro, M. Llompart, C. Garcıa-Jares, R. Cela, J. Chromatogr. A 1137 (2006) 1.
[2] J. Tan, S.M. Cheng, A. Loganath, Y.S. Chong, J.P. Obbard, Chemosphere 66 (2007)
985.
[3] D.R. Oros, D. Hoover, F. Rodigari, D. Crane, J. Sericano, Environ. Sci. Technol. 39
(2005) 33.
[4] J. Pan, Y.L. Yang, Q. Xu, D.L. Xi, Chemosphere 66 (2007) 1971.
[5] Z.W. Cai, G.B. Jiang, Talanta 70 (2006) 88.
[6] O. Päpke, P. Fürst, T. Herrmann, Talanta 63 (2004) 1203.
[7] K. Akutsu, H. Obana, M. Okinhashi, M. Kitagawa, H. Nakazawa, Y. Matsuki, T.
Makino, H. Oda, S. Hori, Chemosphere 44 (2001) 1325.
[8] J.W. She, A. Holden, M. Sharp, M. Tanner, C. Williams-Derry, K. Hooper, Chemo-
sphere 67 (2007) S307.
[9] C. Pirard, E. DePauw, Chemosphere 66 (2007) 320.
[10] L.H. Tseng, M.H. Li, S.S. Tsai, C.W. Lee, M.H. Pan, W.J. Yao, P.C. Hsu, Chemosphere
70 (2008) 640.
[11] L.G. Costa, G. Giordano, NeuroToxicology 28 (2007) 1047.
[12] P. Serodio, M. Salome Cabral, J.M.F. Nogueira, J. Chromatogr. A 1141 (2007)
259.
[13] A. Prieto, O. Zuloaga, A. Usobiaga, N. Etxebarria, L.A. Fernández, Anal. Bioanal.
Chem. 390 (2008) 739.
[14] A. Martinez, M. Ramil, R. Montes, D. Hernanz, E. Rubi, I. Rodriguez, R. CelaTor-
rijos, J. Chromatogr. A 1072 (2004) 83.
[16] J.X. Wang, D.Q. Jiang, Z.Y. Gu, X.P. Yan, J. Chromatogr. A 1137 (2006) 8.
[15] S. Bayen, H.K. Lee, J.P. Obbard, J. Chromatogr. A 1035 (2004) 291.
[17] Y.Y. Li, G.H. Wei, J. Hu, X.J. Liu, X.N. Zhao, X.D. Wang, Anal. Chim. Acta 615 (2008)
96.
[18] F.Rezaei,A.Bidari,A.P.Birjandi,M.R.MilaniHosseini,Y.Assadi,J.Hazard.Mater.
158 (2008) 621.
[19] M. Rezaee, Y. Assadi, M.R. Milani Hosseini, E. Aghaee, F. Ahmadia, S. Berijani, J.
Chromatogr. A 1116 (2006) 1.
[20] S. Brijani, Y. Assadi, M. Anbia, M.R. Milani Hosseini, E. Aghaee, J. Chromatogr. A
1123 (2006) 1.
[21] M.T. Naseri, M.R. Milani Hosseini, Y. Assadi, A. Kiani, Talanta 75 (2007) 56.
[22] N. Shokoufia, F. Shemirani, Y. Assadi, Anal. Chim. Acta 597 (2007) 349.
[23] A. Bidari, E.Z. Jahromi, Y. Assadi, M.R. Milani Hosseini, Microchem. J. 87 (2007)
6.
[24] D.L. Wang, Z.W. Cai, G.B. Jiang, A. Leung, M.H. Wong, W.K. Wong, Chemosphere
60 (2005) 810.
[25] Q. Luo, Z.W. Cai, M.H. Wong, Sci. Total Environ. 383 (2007) 115.
[26] N. Basu, A.M. Scheuhammer, M.O. Brien, Environ. Pollut. 149 (2007) 25.
[27] E. Rogers, M. Petreas, J.S. Park, G.M. Zhao, M.J. Charles, J. Chromatogr. B 813
(2004) 269.
[28] F.Busetti,A.Heitz,M.Cuomo,S.Badoer,P.Traverso,J.Chromatogr.A1102(2006)
104.
[29] E.C. Zhao, W.T. Zhao, L.J. Han, S.R. Jiang, Z.Q. Zhou, J. Chromatogr. A 1175 (2007)
137.
[30] X.D. Wang, X.N. Zhao, X.J. Liu, Y.Y. Li, L.Y. Fu, J. Hu, C.J. Huang, Anal. Chim. Acta
620 (2008) 162.
[31] N. Fattahi, S. Samadi, Y. Assadi, M.R. Milani Hosseini, J. Chromatogr. A 1169
(2007) 63.
[32] G. Kiss, Z. Varga-Puchony, J. Hlavay, J. Chromatogr. A 725 (1996) 261.
[33] R. El Harrak, M. Calull, R.M. Marcé, F. Borrull, Int. J. Environ. Anal. Chem. 64
(1996) 47.
[34] I. Urbe, J. Ruana, J. Chromatogr. A 778 (1997) 337.
[35] P. Liang, J. Hu, Q. Li, Anal. Chim. Acta 609 (2008) 53.
[36] R.R. Kozani, Y. Assadi, F. Shemirani, M.R. Milani Hosseini, M.R. Jamali, Talanta
72 (2007) 387.
[37] Y.Y. Li, G.H. Wei, X.D. Wang, J. Sep. Sci. 30 (2007) 2698.