Article

pubs.acs.org/ac

Microwave Accelerated Selective Soxhlet Extraction for the

Determination of Organophosphorus and Carbamate Pesticides in
Ginseng with Gas Chromatography/Mass Spectrometry
Ting Zhou, Xiaohua Xiao,* and Gongke Li*
School of Chemistry and Chemical Engineering, Sun Yat-sen University, Guangzhou, 510275, China
*
S Supporting Information

ABSTRACT: Microwave accelerated selective Soxhlet extrac-

tion (MA-SSE), a novel selective extraction technique, was
investigated in this study. A Soxhlet extraction system
containing a glass filter was designed as an extractor. During
the procedure of MA-SSE, both the target analytes and the
interfering components were extracted from the sample into
the extraction solvent enhanced by microwave irradiation.
After the solvent flowed though the sorbent, the interfering
components were adsorbed by the sorbent, and the target
analytes remaining in the solvent were collected in the
extraction bottle. No cleanup or filtration was required after
extraction. The efficiency of the MA-SSE approach was
demonstrated in the determination of organophosphorus and carbamate pesticide residues in ginseng by gas
chromatography/mass spectrometry (GC/MS). Under the optimized conditions, low limits of detection (0.050−0.50 μg/kg)
were obtained. The recoveries were in the range of 72.0−110.1% with relative standard deviations less than 7.1%. Because of the
effect of microwave irradiation, MA-SSE showed significant advantage compared with other extraction techniques. The sorbent
used in this study showed good cleanup ability. The mechanism of MA-SSE was demonstrated to be based on the rupture of the
cell walls according to the structural changes of ginseng samples. On the basis of the results, MA-SSE as a simple and effective
sample preparation technique for the analysis of pesticide residues in complex matrixes shows great promise.

S ample preparation is the bottleneck of analytical chemistry

because it constitutes the principal source of error and
remains one of the most time-consuming steps, particularly
with solid samples.1 Some modern extraction techniques, such
as microwave-assisted extraction (MAE), ultrasound-assisted
extraction (UAE), and pressurized liquid extraction (PLE),
have been proved to be rapid and efficient;2,3 however, poor
selectivity is one of their main drawbacks, which leads to
extracts rich in interfering components.4 Further cleanup steps
are needed to remove the interfering components before
instrumental analysis, making the processes tedious and costly.
In addition, the multiple operation steps severely hamper the
speed of analysis and have negative influences on the analysis
quality.5 Thus, some selective extraction techniques, such as
selective PLE6 and hybrid field-assisted solid−liquid−solid
dispersive extraction (HF-SLSDE),7 have been developed.
Selective PLE is a strategy for performing simultaneous
extraction and cleanup by loading the sorbents in the extraction
cells to selectively retain analytes or matrix.8 The system can be
adjusted to retain particular constituents by choosing an
appropriate sorbent and a suitable elution solvent in addition to
tuning the PLE parameters. Selective PLE has been widely
applied to the analysis of various analytes, such as
polybrominated biphenyls,9 polycyclic aromatic hydrocarbons,6
and dioxins10 in complex samples. The common sorbents used
in selective PLE are silica gel,11 modified silica,12 Florisil,13,14
diatomaceous earth,15 alumina,16 carbon,10 and so on. Selective
PLE offers a number of advantages, such as reduced analysis
cost, short operation time, low solvent consumption, minimized
manual labor, and increased quality of analysis. However, there
are some obvious disadvantages, such as high costs of
investment and unusual operating conditions.17 Therefore,
the design of a simple selective extraction technique is of
considerable interest.
HF-SLSDE, a new concept for the selective extraction of
solid samples, was proposed by Zhou et al.7 During the
procedure of HF-SLSDE, the analytes were extracted into the
extraction solvent while the interfering components were
retained by the dispersing sorbent. The whole procedure was
enhanced by a hybrid field formed by microwave and ultrasonic
irradiation. This technique integrated extraction and cleanup in
one step, dramatically cutting time and costs in the sample
preparation. However, there were several limitations of the HF-
SLSDE: a further filtration step was needed before the
chromatography analysis, and the recoveries of some analytes
were still poor.
Received: May 10, 2012
Accepted: June 11, 2012
Published: June 11, 2012

© 2012 American Chemical Society 5816 dx.doi.org/10.1021/ac301274r | Anal. Chem. 2012, 84, 5816−5822
Analytical Chemistry
Soxhlet extraction, developed in 1879, has been the most
widely used extraction technique for isolating target analytes
from solid samples.18 To improve Soxhlet performance, a
variety of modifications have been made, and the most success
was the use of microwaves as an auxiliary energy.19,20
Microwave-assisted Soxhlet extraction (MA-SE), developed by
Luque De Castro et al.,21 is an approach based on the same
principles as conventional Soxhlet extraction but using
microwaves to accelerate the process. Since 1998, three
different prototypes of MA-SE have been designed and
constructed. The first prototype was constructed by Prolabo
(Paris, France). Solvent distillation in this extractor is achieved
by electrical heating, which is independent of the extractant
polarity.22 The main drawback of the first extractor is the
difficulty of using high-boiling-point solvents.23 The second
prototype was constructed by SEV (Puebla, Mexico), called
MIC II. By shortening the distillation glassware, this prototype
has the availability to work with high-boiling-point solvents.24
Finally, a fully automated MA-SE extractor was constructed by
SEV. This extractor named, MIC V, operates with two
extraction units, allowing the simultaneous processing of two
samples.25 In addition to the extraction of lipids from food and
other products, MA-SE has been widely applied to the analysis
of organochlorine pesticides,26 polycyclic aromatic hydro-
carbons,25 and polychlorinated biphenyls27 in natural products
and environmental samples. Although MA-SE has been proved
to be rapid and efficient, its low selectivity often leads to further
cleanup steps.26,28 The processes are relatively laborious and
time-consuming. Therefore, a selective MA-SE technique is
needed.
In view of this, a novel one-step sample preparation
technique called microwave accelerated selective Soxhlet
extraction (MA-SSE) was developed in this study. During the
extraction procedure, the analytes and interfering compounds
of samples were extracted simultaneously into the extraction
solvent. When the solvent flowed though the sorbent, the
interfering components were retained by the sorbent while the
analytes remained in the extraction solvent. The whole
procedure was accelerated by microwave irradiation, and no
cleanup or filtration preceded chromatographic analysis. To
demonstrate the feasibility of the approach, MA-SSE was
applied to analysis of organophosphorus pesticide (OPP) and
carbamate pesticide residues in ginseng with gas chromatog-
raphy/mass spectrometry (GC/MS). The mechanism of MA-
SSE was investigated systematically, including the effect of
microwave irradiation, the cleanup ability of Florisil, and the
microstructure analysis. The mass transfer procedure was finally
proposed.
■
EXPERIMENTAL SECTION
Reagents and Samples. Standards of OPP and carbamate
pesticides were purchased from The Agro-Environment
Protection Institute (Tianjin, China). All standard solutions
were prepared in acetone and were stored at −20 °C. Florisil
with 60−100 mesh was purchased from Supelco (Bellefonte,
PA, USA). Neutral alumina, silicone, and diatomaceous earth,
all of which were the same size of 60−100 mesh, were
purchased from Sinopharm Chemical Reagent Co., Ltd.
(Shanghai, China). Ethyl acetate, acetone, dichloromethane,
and acetonitrile were of analytical grade and were purchased
from Guangzhou Chemical Reagents Factory (Guangzhou,
China). Distilled water was used throughout the study.
5817
Article
Ginsengs were collected from America (Wisconsin), Canada
(Ontario), and Northeast China (Jilin). After being dried in a
hot air cabinet with ventilation at 40 °C for 5 h, the ginsengs
were triturated, passed through a stainless steel sieve with 0.45
mm mesh sizes and stored in closed desiccators. The same
batch of samples was used here in the experiments.
Apparatus. MA-SSE experiments were performed on an
MAS-II microwave extraction apparatus (Sineo Microwave
Chemistry Technology Company, Shanghai, China). Micro-
wave irradiation, with a maximum delivered power of 1000 W
at a frequency of 2450 MHz, can be dynamically adjusted by
temperature and power feedbacks, which is able to provide
continuous nonpulse microwave heating. Temperature is
monitored by an infrared sensor installed on the inner wall.
GC/MS analysis was carried out on a Shimadzu GC-MS
QP2010 system (Shimadzu, Kyoto, Japan). A DB-5MS
(Agilent, Palo Alto, USA) capillary column (30 m long ×
0.25 mm i.d. × 0.25 μm film thickness) was used for the
chromatographic separation. The chromatographic data were
recorded and processed with GC/MS solution software
(version 2.10) supplied by Shimadzu.
SEM images were conducted by a Quanta 600 scanning
electron microscopy (SEM) instrument (FEI, Hillsboro, OR,
USA).
Sorbent Conditioning. Florisil was heated in a muffle
furnace at 550 °C for 6 h and cooled in a desiccator, then was
added with purified water (equivalent to 5% w/w) and
homogenized by rotation for 2 h. After that, the Florisil was
left in a closed container to equilibrate for 48 h before use.
Diatomaceous earth was boiled in a 5% HCL aqueous solution
for 1 h and then was filtered and washed with water until the
pH value was 6.0. After that, diatomaceous earth was heated in
a muffle furnace at 500 °C for 2 h. The silicone was heated at
110 °C for 1 h and cooled in a desiccator. Neutral alumina was
heated at 450 °C for 4 h and cooled in a desiccator.
Extraction Procedures. Microwave Accelerated Selective
Soxhlet Extraction. A 1.0 g portion of ginseng sample was
accurately weighed and put into the Soxhlet thimble after the
addition of 2.0 g of anhydrous sodium sulfate and a certain
amount of sorbent. The entire filler was then compressed to
form a column packing. A certain volume of extraction solvent
and a stirrer were added into the 50 mL round-bottom flask.
MA-SSE was performed at a certain temperature with a certain
power of microwave irradiation. The extraction conditions were
optimized by a monofactor test. The extract obtained was
concentrated to 1.0 mL under a nitrogen stream in a 40 °C
water bath.
Selective Soxhlet Extraction. A 1.0 g portion of ginseng
sample was accurately weighed and put into the Soxhlet thimble
after the addition of 2.0 g of anhydrous sodium sulfate and 3.0 g
of Florisil. The entire filler was then compressed to form a
column packing, then 30 mL of ethyl acetate and a stirrer were
added into the 50 mL round-bottom flask. The experiment was
performed at 85 °C in a thermostatic water bath for some time,
and the stirring speed was 1000 rpm. The extract obtained was
concentrated to 1.0 mL under a nitrogen stream in a 40 °C
water bath.
Microwave-Assisted Soxhlet Extraction. A 1.0 g portion of
ginseng sample was accurately weighed and put into the Soxhlet
thimble after the addition of 2.0 g of anhydrous sodium sulfate.
The entire filler was then compressed to form a column
packing, then 30 mL of ethyl acetate and a stirrer were added
into the 50 mL round-bottom flask. The experiment was
dx.doi.org/10.1021/ac301274r | Anal. Chem. 2012, 84, 5816−5822
Analytical Chemistry
performed at 75 °C with 900 W microwave irradiation power
for 15 min, and the stirring speed was set at 1000 rpm. The
extract obtained was concentrated to 1.0 mL under a nitrogen
stream in a 40 °C water bath.
Ultrasound-Assisted Extraction Coupled with Solid-Phase
Extraction. A 1.0 g portion of ginseng sample was accurately
weighed and put into the conical flask. After the addition of 20
mL of acetonitrile and 2.0 g of anhydrous sodium sulfate, the
flask was sonicated for 20 min in an ultrasonic bath under 120
W ultrasonic irradiation power with a frequence of 40 kHz. The
extraction solution obtained was filtered through a G4 glass
filter with 4−7 μm pore size. The solution was evaporated to
dryness under a gentle stream of nitrogen in a 45 °C water
bath. The residue was reconstituted in 1 mL of acetone.
After that, the reconstituted solution was transferred into a
column filled with 2.0 g of anhydrous sodium sulfate, 1.0 g of
Florisil, and 1.0 g of neutral alumina as a solid-phase extraction
(SPE) column, which was employed to clean up the sample.
Elution was made by gravity flow with 20 mL of acetone, and
the eluent was collected into a graduated conical tube. The
solution was evaporated to dryness under a gentle stream of
nitrogen in 45 °C a water bath. The residue was reconstituted
in 1 mL of ethyl acetate.
Analysis. A 1.0 μL portion of the extract was injected
directly into the GC/MS system for analysis. The GC
conditions were as follows: injector temperature was 220 °C.
The column temperature was maintained at 50 °C for 1 min,
programmed at 20 °C/min to 190 °C, then increased to 210 at
2 °C/min, and finally increased to 250 at 10 °C/min and held
for 10 min. The injector was operated in the splitless mode (1
min) at 250 °C. The total analysis time was 32 min, and the
equilibration time was 2 min. Helium was used as the carrier
gas at a constant flow rate of 1.58 mL/min.
The mass spectrometric detector was operated in the
electron ionization (EI) mode with an ionizing energy of 70
eV, an ion source temperature of 230 °C, and a quadrupole
temperature of 150 °C. Date acquisition was carried out in the
selected ion monitoring (SIM) mode with a solvent delay of 4
min. Each compound was confirmed by its retention time and
the identification of three selected ions and was quantified on
the basis of the peak area using the quantitative ions (Table S-1
of the Supporting Information).
■
RESULTS AND DISCUSSION
Actualization of Microwave Accelerated Selective
Soxhlet Extraction. MA-SSE was performed with an
apparatus shown in Figure 1. It consists of a microwave
extraction device, a reflux condenser, and an improved Soxhlet
extractor. The improved Soxhlet extractor comprises a Soxhlet
thimble and a round-bottom flask. The Soxhlet thimble
containing a glass filter with 10−15 μm pore size is used to
hold the entire sample and sorbent. The extraction solvent is
added into the round-bottom flask. The cooling system fitted
on top of the microwave cavity refrigerates the solvent
continuously, and the solvent is refluxed into the Soxhlet
extractor.
Before the extraction, the sorbent and sample are added into
the Soxhlet thimble in order, and the extraction solvent is
added into the round-bottom flask. During the extraction
procedure, both the extraction solvent and sample are heated
by the microwave irradiation. When the solvent begins to reflux,
both the target analytes and the interfering compounds in the
sample are extracted into the extraction solvent. After the
5818
Article
Figure 1. The schematic diagram of MA-SSE apparatus.
solvent flows though the sorbent, the interfering compounds
are adsorbed by the sorbent while the target analytes remain in
the extraction solvent, which is finally collected in the round-
bottom flask. By tuning the solvent composition and choosing
appropriate sorbent as well as other parameters, MA-SSE can
be adjusted to extract particular compounds. The sorbents used
in the experiments, of which the quality is a critical feature, have
to be pretreated as described in the Experimental Section.29
The sorbents need to be active enough to adsorb interfering
compounds while allowing the target analytes to be eluted.
Optimization of the MA-SSE Process. The influences of
the main variables potentially affecting the MA-SSE process
were evaluated to obtain an efficient extraction. All the
experiments were performed at 200 μg/kg of fortified level
according to the Experimental Section.
The sorbents tested in this study were Florisil, neutral
alumina, underivatized silica, and diatomaceous earth. The
results showed that the cleanup abilities of the sorbents
decreased in the following order: Florisil > diatomaceous earth
> underivatized silica > neutral alumina. The cleanup abilities of
underivatized silica and neutral alumina were so poor that the
analytes were severely interfered during GC/MS analysis. The
adsorption of diatomaceous earth is nonselective, which
resulted in low recoveries. Therefore, Florisil was chosen as
the sorbent for further application on the basis of the best
compromise between cleanup ability and extraction efficiency.
The extraction solvents tested in this study were ethyl
acetate, acetone, dichloromethane, and acetonitrile. Figure S-1A
of the Supporting Information shows the influence of extraction
solvent on the extraction efficiency. The results demonstrated
that ethyl acetate and acetone obtained higher extraction yields
than dichloromethane and acetonitrile. That is because OPP
and carbamate pesticides are polar molecules; thus, the
extraction effect of less polar solvents, such as dichloromethane,
is poor based on the principle of the dissolution in the similar
material structure. In this case, ethyl acetate, which is less toxic
than acetone, was chosen as the extraction solvent for further
application.
dx.doi.org/10.1021/ac301274r | Anal. Chem. 2012, 84, 5816−5822
Analytical Chemistry
Figure S-1B of the Supporting Information shows the
influence of the proportion of the sorbent and sample on the
extraction efficiency. The extraction yields of the pesticides
increased as the proportion of the sorbent and sample
decreased. That is because although the sorbent has been
pretreated before the experiment, it would still adsorb an
amount of target analytes. So the less sorbent that is used, the
higher extraction yields that will be obtained. However, some
interference occurred in GC/MS analysis when the proportion
decreased to 1:1. Consequently, the optimized proportion of
the sorbent and sample was 2:1.
Figure S-1C of the Supporting Information shows the
influence of microwave irradiation power on the extraction
efficiency. The results revealed that the extraction yields of the
pesticides increased along with the increment of microwave
irradiation power from 300 to 900 W. The probable reason was
that a large microwave irradiation power obtained good effects
of heating. The more times the solvent refluxes, the more
sufficiently the target analytes are eluted. Thus, the optimized
microwave irradiation power was 900 W.
Figure S-1D of the Supporting Information shows the
influence of extraction temperature on the extraction efficiency.
The results demonstrated that the extraction yields of the
pesticides increased as the extraction temperature increased.
That was probably because the high temperature led to fast
reflux of the solvent. Moreover, the high temperature was
beneficial for diffusion and mass transfer during extraction, and
it increased the dissolving capacity of the extraction solvent. In
addition, the high temperature achieved by microwave
irradiation would cause the rupture of the sample cell and
the release of the target compounds. However, it was
discovered that the burst like popcorn occurred easily when
the temperature reached 80 °C, which was probably because
ginseng contains a lot of starch. Finally, 75 °C was chosen as
the extraction temperature.
The effect of the extraction time was examined in the range
of 5−20 min. The results showed that after 15 min, the
extraction yields of the pesticides did not increase significantly.
Thus, 15 min was chosen as the extraction time in this study.
Validation and Application of MA-SSE−GC/MS Meth-
od. All pesticides determined by GC/MS were identified by
their chromatogram and retention times compared with those
of the standards. Typical chromatograms obtained from a
standard solution of all pesticides and spiked ginsengs by MA-
SSE are shown in Figure 2. There was no significant
interference from the ginseng sample observed at the retention
times of the analytes.
Linearity was assessed using matrix matched calibration
solutions to compensate for the detector response enhance-
ment effect.30 The detector response showed good linearity
over the studied concentration range, which is given in Table 1.
Linear regression coeffcients for all pesticides were in the range
of 0.9993−0.9999. Limits of detection (LODs) were calculated
at 3 times the standard deviation above the blank ginseng
extracts, ranging from 0.050 to 0.50 μg/kg.
Recoveries of the pesticides from ginseng matrix are shown
in Table S-2 of the Supporting Information. The spiked ginseng
samples at three fortification levels (5, 50, 200 μg/kg) were
extracted by MA-SSE in three replicates. The recoveries of all
pesticides ranged from 72.0% to 110.1%, with relative standard
deviations (RSDs) of <7.1%. The results demonstrated that the
method was precise and accurate.
5819
Article
Figure 2. Typical chromatograms obtained by GC/MS of (A) a blank
sample spiked with OPP and carbamate pesticides at 200 μg/kg by
MA-SSE, (B) a standard solution of 200 μg/L OPP and carbamate
pesticides (for peak assignments, see Table 1).
The reproducibility was estimated by five repetitive samples
obtained by MA-SSE. The RSDs of the pesticides were in the
range of 2.5−6.2%, indicating that the repeatability of the
method was good. The stability was investigated by
determining the varieties of the analytes in the extraction
solutions on 6 separate days. All the RSDs of intraday and
interday were <1.7% and 3.9%, respectively. The results
indicated that the analytes were stable in the extraction
solutions for at least 6 days.
The present method was finally applied to the analysis of
OPP and carbamate pesticides in ginseng samples collected
from America, Canada, and Northeast China. No OPP or
carbamate pesticides were detected in the real samples above
the LODs. The recoveries of the pesticides in the real samples
were in the range of 77.0−106.3%, with RSDs of <6.8%.
Comparison of Different Methods. To evaluate the
predominance of MA-SSE, ultrasound -assisted extraction
coupled with solid-phase extraction (UAE−SPE) used in the
literature31 was tested in this paper. It was performed at the
optimum conditions in the Experimental Section. The
recoveries of all pesticides obtained by MA-SSE and UAE−
SPE at 200 μg/kg are shown in Figure 3. The recoveries
obtained by MA-SSE and UAE−SPE were in the range of
92.5−109.4% and 41.1−96.2%, with RSDs of <7.1% and 12.0%,
respectively. The results showed that MA-SSE achieved more
satisfactory recoveries than UAE−SPE, and the precision of
MA-SSE was better.
That is probably because in the MA-SSE procedure, the
analytes were extracted continuously by fresh solvent under
microwave irradiation, making the extraction complete. In
contrast, in the procedure of UAE−SPE, the extraction was
primarily related to the cavitation effects of ultrasound.32 The
bubbles generated by cavitation imploded and produced a
highly efficient pressure, enhancing the solubility of the analytes
and facilitating the mass transfer during the extraction.33 At the
late stage of UAE, there was an equilibrium of extraction for the
solvent that was unchanged. So it was hard for UAE to extract
the analytes completely. Moreover, the extraction and cleanup
were separated in UAE−SPE, which was much more laborious
and would introduce operation error.31 In addition, the
multiple operation steps led to the unsatisfactory recoveries
dx.doi.org/10.1021/ac301274r | Anal. Chem. 2012, 84, 5816−5822
Analytical Chemistry Article

Table 1. Analytical Performance of GC/MS for the Determination of OPP and Carbamate Pesticides
no. pesticides tR (min) equation linear range (μg/kg) R LOD (μg/kg)
1 isoprocarb 8.36 Y = 2.61 × 102 X − 2.35 × 10 0.50−400 0.9998 0.10
2 phorate 9.64 Y = 4.26 × 102 X − 9.24 × 102 0.50−400 0.9996 0.050
3 iprobenfos 11.25 Y = 1.92 × 102 X + 9.96 × 10 0.50−400 0.9998 0.10
4 chlorpyrifos methyl 11.81 Y = 2.04 × 102 X + 1.30 × 102 0.50−400 0.9996 0.10
5 parathion methyl 12.00 Y = 8.58 × 10 X + 5.13 × 10 1.0−400 0.9993 0.20
6 tolclofos methyl 12.02 Y = 3.15 × 102 X + 2.61 × 102 0.50−400 0.9996 0.10
7 malathion 13.06 Y = 1.18 × 102 X − 2.97 × 102 2.0−400 0.9996 0.50
8 chlorpyrifos 13.28 Y = 6.09 × 10 X + 1.57 × 102 0.50−400 0.9999 0.10

Figure 3. Comparison of MA-SSE and UAE−SPE for analyzing

ginseng samples spiked with OPP and carbamate pesticides at 200 μg/ Figure 4. Comparison of MA-SSE (15 min) and SSE (40 min and 2 h)
kg. A, phorate; B, parathion methyl; C, isoprocarb; D, malathion; E, for analyzing ginseng samples spiked with OPP and carbamate
chlorpyrifos methyl; F, chlorpyrifos; G, tolclofos methyl; H, pesticides at 200 μg/kg.
iprobenfos.
The Cleanup Ability of Florisil. Florisil added in MA-SSE
was used to purify the extracts from the ginseng sample, so the
cleanup ability is very important. The cleanup ability of Florisil
because of the sample loss. Therefore, MA-SSE has significant was investigated by comparing the colors of the extraction
advantage compared with UAE−SPE. solutions obtained by MA-SSE (with Florisil) and microwave-
Mechanism of MA-SSE. The Effect of Microwave assisted Soxhlet extraction (MA-SE, without Florisil), which is
Irradiation During MA-SSE. To study the effect of microwave shown in Figure 5. It is revealed that the use of Florisil helped
irradiation during MA-SSE, selective Soxhlet extraction (SSE) the adsorbing of pigment in ginseng. In addition, the cleanup
was tested in this study. The standard addition recoveries of ability of Florisil was further supported by comparing the
MA-SSE (15 min), SSE (40 min), and SSE (2 h) at 200 μg/kg chromatograms of extracts obtained by MA-SSE and MA-SE.
are shown in Figure 4. The recoveries obtained by MA-SSE,
SSE (40 min), and SSE (2 h) were in the range of 92.5−
109.4%, 68.5−94.5%, and 92.6−110.1%, with RSDs of <7.1%,
11.1%, and 8.7%, respectively.
On the basis of the results, the recoveries obtained by MA-
SSE were much more satisfactory than those of SSE (40 min)
and were close to those of SSE (2 h). The reason was probably
based on their different extraction mechanisms. The microwave
irradiation during MA-SSE could initiate the movement and
rotation of polar molecules, leading to a fast heating of the
sample and the solvent.34 The rapid increase of the temperature
in the cells caused a dramatic expansion with subsequent
rupture of the cell walls, releasing the internal compounds into
the extraction solvent. In the SSE process, the extraction was
based on the leaching of the solvent, and the target compounds
were extracted by solubilization; thus, prolonged extraction
time was needed, and the efficiency was low.18 Thus, the
microwave irradiation during MA-SSE accelerated the speed of Figure 5. Pictures of extraction solutions using Florisil (A) and
extraction significantly compared with SSE. without using Florisil (B).

5820 dx.doi.org/10.1021/ac301274r | Anal. Chem. 2012, 84, 5816−5822

Analytical Chemistry
Figure 6. Extraction process of the MA-SSE technique: (A) before extraction, (B) during extraction, and (C) after extraction.
There were many more interferences in the extract obtained by
MA-SE, which seriously interfered with the analysis of
malathion and chlorpyrifos.
As a normal phase sorbent, Florisil can adsorb strongly polar
compounds, such as pigments and other interfering compo-
nents from sample matrixes.35 The retention of the interfering
compounds by Florisil is primarily due to the interactions
between polar functional groups such as hydrogen bonding.36
Moreover, Florisil is a nice fat retainer and could give fat-free
extracts of ginseng.8 According to the results, Florisil can
adsorb the major interferences in ginseng effectively, and the
use of Florisil would prevent the GC system from serious
pollution.
Microstructure Analysis. To further investigate the mecha-
nism of MA-SSE, SEM was used to identify the microstructures
of samples over various conditions, which are shown in Figure
S-2 of the Supporting Information. Distinguishable physical
changes produced by different extraction methods were
observed.
Compared with the intact surface of the untreated ginseng
sample (Figure S-2A), the surface of the sample after MA-SSE
(Figure S-2B) was greatly destroyed, and the cell walls were
ruptured. The result was in accord with the hypothesis of MAE
proposed by Paré et al.37 The rapid increase in the temperature
generated by microwave heating caused the rupture of the cell
walls, leading to the quick migration of the compounds in the
cells into the extraction solvent.38 In contrast, the micro-
structures of the ginseng samples after SSE and UAE (Figure S-
2C and S-2D, respectively) were not significantly changed. The
5821
Article
surface of the ginseng shrank a little after SSE, probably because
of the high extraction temperature and the long extraction
time.18 The surface of the ginseng after UAE was still intact
without any ruptures. The results indicate that in the SSE and
UAE process, the solvent transferred into the matrix and
extracted the target compounds by solubilization. Therefore, a
long extraction time was needed in SSE and low recoveries
were obtained by UAE.
The Mass Transfer and Prospect of MA-SSE. The mass
transfer of MA-SSE was proposed on the basis of the above
results, which is shown in Figure 6. MA-SSE was divided into
three procedures. Before the extraction, the ginseng samples
were placed on the sorbents. During the extraction procedure
of MA-SSE, the extraction solvent was heated up to the boiling
point and the steam was generated. After the condensation, the
solvent drips flowed though the samples and the sorbents.
Meanwhile, the cell walls of the ginseng samples were ruptured
under the microwave irradiation, and the inside target analytes
and interfering compounds dissolved and dispersed into the
extraction solvent simultaneously. When the extraction solution
flowed though the sorbent, the target analytes and interfering
compounds distributed between the extraction solvent and the
sorbent on the basis of their interactions. The normal phase
sorbent Florisil can adsorb strongly polar interfering com-
pounds and most lipids in the samples.35 Ethyl acetate, the
extraction solvent used in this study, has enough polity to elute
the pesticides but could not elute more polar interfering
compounds. Thus, after the extraction, the interfering
compounds were retained by the sorbent, and the pesticides
dx.doi.org/10.1021/ac301274r | Anal. Chem. 2012, 84, 5816−5822
Analytical Chemistry
were collected in the flask. Nonpolar solvent could not be
heated under microwave irradiation, for it is incapable of
absorbing microwaves.39 Therefore, MA-SSE is particularly
suited to extract polar analytes from a complex matrix using a
polar extraction solvent.
■
CONCLUSION
MA-SSE, a novel selective extraction technique, was originally
developed in this study. The feasibility of MA-SSE was
demonstrated by analyzing OPP and carbamate pesticide
residues in ginseng with GC/MS. Low LODs, satisfactory
recoveries, and good repeatability were obtained. Compared
with UAE−SPE, MA-SSE provided more satisfactory accuracy
and precision. The microwave irradiation during MA-SSE
accelerated the speed of extraction significantly compared with
SSE. The sorbent used in this technique was proved to have
nice cleanup ability, which resulted from the interactions of the
polar functional groups. The SEM results showed that the
enhanced extraction of MA-SSE was based on the explosion of
sample cells. It is strongly believed that MA-SSE as a simple
and effective technique can be potentially extended to other
polar target analytes in complex matrixes. This technique will
be further explored for the possibility of analyzing less polar
compounds by separating the heating-up resources of
extraction solvent and sample, which is under investigation in
our laboratory.
■
*
ASSOCIATED CONTENT
S Supporting Information
Additional information as noted in text. This material is
available free of charge via the Internet at http://pubs.acs.org.
■
AUTHOR INFORMATION
Corresponding Author
*Phone: +86-20-84110922. Fax: +86-20-84115107. E-mail:
(G.L.) cesgkl@mail.sysu.edu.cn, (X.X.) xiaoxhua@mail.sysu.
edu.cn.
Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS
This work was supported by the Natural Science Foundation of
China (No. 21127008 and 20905080) and the Guangdong
Provincial Department of Science and Technology of China
(No. 2009B010900021) and the Fundamental Research Funds
for the Central Universities (No. 11lgpy80).
■
REFERENCES
(1) Chen, Y.; Guo, Z.; Wang, X.; Qiu, C. J. Chromatogr., A 2008,
1184, 191−219.
(2) Raynie, D. E. Anal. Chem. 2006, 78, 3997−4004.
(3) Tadeo, J. L.; Sánchez-Brunete, C.; Albero, B.; García-Valcárcel, A.
I. J. Chromatogr., A 2010, 1217, 2415−2440.
(4) Hyötyläinen, T. Anal. Bioanal. Chem. 2009, 394, 743−758.
(5) Ridgway, K.; Lalljie, S.; Smith, R. M. J. Chromatogr., A 2007,
1153, 36−53.
(6) Lundstedt, S.; Haglund, P.; Ö Berg, L. Anal. Chem. 2006, 78,
2993−3000.
(7) Zhou, T.; Xiao, X.; Li, G. Anal. Chem. 2011, 84, 420−427.
(8) Björklund, E.; Sporring, S.; Wiberg, K.; Haglund, P.; Holst, C. V.
Trends Anal. Chem. 2006, 25, 318−325.
(9) Björklund, E.; Müller, A.; von Holst, C. Anal. Chem. 2001, 73,
4050−4053.
5822
Article
(10) Haglund, P.; Sporring, S.; Wiberg, K.; Björklund, E. Anal. Chem.
2007, 79, 2945−2951.
(11) Subedi, B.; Mottaleb, M. A.; Chambliss, C. K.; Usenko, S. J.
Chromatogr., A 2011, 1218, 6278−6284.
(12) Zhang, Z.; Rhind, S. M.; Kerr, C.; Osprey, M.; Kyle, C. E. Anal.
Chim. Acta 2011, 685, 29−35.
(13) Sanchez-Prado, L.; Lamas, J. P.; Lores, M.; Garcia-Jares, C.;
Llompart, M. Anal. Chem. 2010, 82, 9384−9392.
(14) Ghosh, R.; Hageman, K. J.; Björklund, E. J. Chromatogr., A 2011,
1218, 7242−7247.
(15) Racamonde, I.; González, P.; Lorenzo, R. A.; Carro, A. M. J.
Chromatogr., A 2011, 1218, 6878−6883.
(16) Chuang, J. C.; Van Emon, J. M.; Schrock, M. E. Chemosphere
2009, 77, 1217−1223.
(17) Mustafa, A.; Turner, C. Anal. Chim. Acta 2011, 703, 8−18.
(18) de Castro, M.; Priego-Capote, F. J. Chromatogr., A 2010, 1217,
2383−2389.
(19) Luque-García, J. L.; de Castro, M. Talanta 2004, 64, 571−577.
(20) Luque-García, J. L.; Luque De Castro, M. D. Trends Anal. Chem.
2003, 22, 90−98.
(21) García-Ayuso, L. E.; Sánchez, M.; Fernández De Alba, A.; Luque
De Castro, M. D. Anal. Chem. 1998, 70, 2426−2431.
(22) García-Ayuso, L. E.; Luque-García, J. L.; de Castro, M. Anal.
Chem. 2000, 72, 3627−3634.
(23) García-Ayuso, L. E.; Luque De Castro, M. D. Trends Anal. Chem.
2001, 20, 28−34.
(24) Luque-García, J. L.; de Castro, M. Anal. Chem. 2001, 73, 5903−
5908.
(25) Priego-Capote, F.; Luque-García, U.; de Castro, M. J.
Chromatogr., A 2003, 994, 159−167.
(26) Prados-Rosales, R. C.; García, J.; de Castro, M. J. Chromatogr., A
2003, 993, 121−129.
(27) Luque-García, J. L.; de Castro, M. J. Chromatogr., A 2003, 998,
21−29.
(28) Morales-Muñoz, S.; Luque-García, J. L.; Ramos, M. J.; Martínez-
Bueno, M. J.; Luque De Castro, M. D. Chromatographia 2005, 62, 69−
74.
(29) Kristenson, E. M.; Brinkman, U. A. T.; Ramos, L. Trends Anal.
Chem. 2006, 25, 96−111.
(30) Hajšlová, J.; Zrostlíková, J. J. Chromatogr., A 2003, 1000, 181−
197.
(31) González-Rodríguez, R. M.; Rial-Otero, R.; Cancho-Grande, B.;
Simal-Gándara, J. J. Chromatogr., A 2008, 1196−1197, 100−109.
(32) Luque-García, J. L.; Luque De Castro, M. D. Trends Anal. Chem.
2003, 22, 41−47.
(33) Pérez-Serradilla, J. A.; Priego-Capote, F.; Luque De Castro, M.
D. Anal. Chem. 2007, 79, 6767−6774.
(34) Sparr Eskilsson, C.; Björklund, E. J. Chromatogr., A 2000, 902,
227−250.
(35) Moreda-Piñeiro, J.; Alonso-Rodríguez, E.; López-Mahía, P.;
Muniategui-Lorenzo, S.; Prada-Rodríguez, D.; Romarís-Hortas, V.;
Míguez-Framil, M.; Moreda-Piñeiro, A.; Bermejo-Barrera, P. Trends
Anal. Chem. 2009, 28, 110−116.
(36) García-López, M.; Canosa, P.; Rodríguez, I. Anal. Bioanal. Chem.
2008, 391, 963−974.
(37) Paré, J. R. J.; Belanger, J. M. R.; Stafford, S. S. Trends Anal.
Chem. 1994, 13, 176−184.
(38) Valérie, C. Trends Anal. Chem. 2000, 19, 229−248.
(39) Chan, C.; Yusoff, R.; Ngoh, G.; Kung, F. W. J. Chromatogr., A
2011, 1218, 6213−6225.
dx.doi.org/10.1021/ac301274r | Anal. Chem. 2012, 84, 5816−5822