Journal of Chromatography A, 1245 (2012) 8–16

Contents lists available at SciVerse ScienceDirect

Journal of Chromatography A
journal homepage: www.elsevier.com/locate/chroma

Synthesis and characterization of the core–shell magnetic molecularly imprinted

polymers (Fe3O4@MIPs) adsorbents for effective extraction and determination of
sulfonamides in the poultry feed
Xuan Kong a,b , Ruixia Gao a,b , Xiwen He a,b , Langxing Chen a,b,∗ , Yukui Zhang a,b,c
a
Department of Chemistry, Nankai University, Tianjin 300071, China
b
The State Key Laboratory of Medical Chemical Biology, Nankai University, Tianjin 300071, China
c
Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Dalian 116011, China

artic l e i nf o abstract

Article history: In this study, we present a general method to prepare the core–shell magnetic molecularly imprinted
Received 6 March 2012 polymers (MIPs) nanoparticles (NPs) for sulfamethazine (SMZ). The resulting Fe3O4@MIPs NPs possess
Received in revised form 19 April 2012 a highly improved imprinting effect, fast adsorption kinetics and high adsorption capacity, and can be
Accepted 23 April 2012
applied to extract sulfonamide in the poultry feed. In this protocol, the magnetite NPs were synthesized
Available online 18 May 2012
by co-precipitating Fe2+ and Fe3+ in an ammonia solution first. Silica was then coated on the Fe3O4 NPs
using a sol–gel method to obtain silica shell magnetic NPs. Subsequently, the vinyl groups were grated
Keywords:
onto silica-modified Fe3O4 surface by 3-methacryloyloxypropyltrimethoxysilane. Finally, the MIPs films
Magnetic nanoparticles
Molecularly imprinted polymers (MIPs) were formed on the surface of Fe3O4@SiO2 by the copolymerization of vinyl end groups with functional
Sulfonamide monomer, methacrylic acid, cross-linking agent, ethylene glycol dimethacrylate, the initiator azo-bis-
Poultry feed isobutyronitrile and template molecule, sulfamethazine. The morphology, magnetic, adsorption and
recognition properties of Fe3O4@MIPs NPs were characterized using transmission electron microscope
(TEM), scanning electron microscope (SEM), Fourier transform infrared (FT-IR) spectrometer, vibrating
sample magnetometer (VSM) and re-binding experiments. The results showed that the binding sites
of Fe3O4@MIPs were good accessibility, fast adsorption rate and the maximum adsorption capacity of
Fe3O4 @MIPs to SMZ was 344.8 µg g−1. The selectivity of the obtained Fe3O4@MIPs NPs were elucidated
by the different rebinding capability of SMZ and structural related sulfonamides in the mixed solution.
The results indicated that the Fe3O4@MIPs had high imprinting factor 9.5 and significant selectivity. A
method was developed for enrichment and determination of SMZ in the poultry feed samples with recov-
eries of duck and chicken feed ranging from 63.3 to 76.5% and 68.7 to 74.7%, respectively and the relative
standard deviations (RSD) (<6.7%).
© 2012 Elsevier B.V. All rights reserved.

1. Introduction [3]. Therefore, there is a constant requirement for good analytical

The antibiotic sulfonamides are commonly used as antibacterial
drugs in both human and veterinary medicine to fight infectious
diseases, and in animal feed to promote livestock growth [1,2]. The
continual overuse of sulfonamides, much attention has been paid to
their potential carcinogenesis and the worry of the development of
antibacterial resistance. Because of these health concerns, the Euro-
pean Commission (EC), America and other countries, for example
China, have adopted a maximum acceptable limit of residual sul-
fonamide in foods of animal origin, for example meat, milk and eggs
∗ Corresponding author at: Department of Chemistry, Nankai University, Tianjin
300071, China. Tel.: +86 22 2350 5091; fax: +86 22 2350 2458.
E-mail addresses: lxchen@nankai.edu.cn (L. Chen), ykzhang@dicp.ac.cn
(Y. Zhang).
protocols based on efficient analytical processes (sensitive, selec-
tive, fast and inexpensive) by legislation, health authorities and
companies operating in the food market. Many methods have been
described for determination of sulfonamides in biological sam-
ples including high performance liquid chromatography (HPLC)
[4,5], gas chromatography (GC) [6,7], liquid chromatography–mass
spectrometry (LC–MS) [8], and capillary electrophoresis (CE) [9].
Because of the complexity of food matrices and contaminants
presented in food at low concentration (ng µg g−1) levels, all sul-
fonamide analysis requires extensive sample pretreatment, for
example liquid–liquid extraction and solid-phase extraction (SPE)
[10]. Accordingly, there is a considerable interest in developing
an effective extraction method for extracting and isolating com-
ponents from complex food matrices prior to the measurement.
As economical, rapid and selective clean-up methods are needed,
the application of SPE procedures involving the use of synthetic

0021-9673/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.chroma.2012.04.061
 X. Kong et al. / J. Chromatogr. A 1245 (2012) 8–16 9

antibody mimics, such as MIPs as adsorbents, called MISPE, offering
the advanced specificity in comparison with traditional SPE adsor-
bents, has received increasing attention over the past decade as an
attractive alternative for the analysis of complex samples [11–16]
where analyte selectivity in the presence of very complex samples
represents the main problem.
MIPs are tailor-made materials that can exhibit high affinity
and selectivity towards a given target or group of target molecules.
The synthesis of imprinted polymers is based on polymerization of
functional monomers and a cross-linker around a template. Extrac-
tion of the template leaves behind recognition sites of functional
and shape complementarity to the template [17,18]. In comparison
with antibodies, enzymes or biological receptors, MIPs show inher-
ent advantages: their production is quite simple and economical,
moreover the polymers show good physical and chemical stability
and applicability in harsh chemical media without loss of bind-
ing properties. In recent years, several MIPs for sulfamethoxazole
[19–23] and sulfamethazine [22–28] have been prepared as the sta-
tionary phase for HPLC and as solid-phase extracting agents, but
few published articles has focused on analysis of sulfonamides in
fish, pork, chicken and milk samples using methods based on MIP
technology. Most MIPs for sulfonamides have been synthesized in
bulk polymerization, followed by a grinding and sieving process to
acquire the desired particles; these can suffer from drawbacks in
some applications, for example, heterogeneous distribution of the
binding sites, poor site accessibility to the target molecules because
the template and functionality are totally embedded in the poly-
mer matrices. Therefore, the kinetics of the sorption/desorption
process is unfavorable and the mass transfer become slow. To
solve these problems, surface imprinting on solid support [29–32]
and surface-imprinted core–shell NPs [33–40] were exploited. We
recently developed the synthesis of silica-coated MIPs nanospheres
for selective extraction of sulfonamides from milk and eggs samples
with highly imprinting effect and adsorption capacity [22].
Magnetic field-based separations using magnetic nano- and
microparticles have received great attention for their superior char-
acteristics such as good dispersion, the fast and effective binding
of targets, reversible and controllable flocculation. The magnetic
separation process can be performed directly in crude samples
containing suspended solid particles or other biological particu-
lates in a rapid and easy way in which several stages of sample
pretreatment (especially centrifugation, filtration and membrane
separation) can be ignored, which greatly facilitate the lab
work. Moreover, the power and efficiency of magnetic separation
procedures are especially useful in large-scale operations [41]. For
many of these applications, surface modification of magnetic NPs is
a key challenge. When modified with a specific functional polymer,
for example, the MIPs, which provide functional materials able to
recognize and in some cases respond to biological and chemical
agents of interest, these magnetic NPs coated MIPs could be used
to separate and concentrate chemicals more conveniently with the
help of an external magnetic field [37–40]. Therefore, the combin-
ing magnetic separation and molecular imprinting would ideally
provide a powerful analytical tool with characteristics of simplicity,
flexibility, and selectivity.
In this work, we present a general method to prepare a
MIPs layer on magnetic Fe3O4 NPs with uniform core–shell
structure by combining surface imprinting and nanotech-
niques. Compared to traditional MIPs, the MIPs prepared onto
the surface of iron oxide would be an ideal candidate for
the multifunctional nanomaterials toward bioseparation. Nano-
sized Fe3O4 NPs were first prepared using a coprecipitation
method, and then silica shell deposited. Subsequently, the vinyl
groups were grated onto silica-modified Fe3O4 surface by 3-
methacryloyloxypropyltrimethoxysilane (MPS). The MIPs shells
were coated on the silica surface by the copolymerization of vinyl
end groups with functional monomer, methacrylic acid (MAA),
cross-linking agent, ethylene glycol dimethacrylate (EGDMA), the
initiator, azo-bis-isobutyronitrile (AIBN) and template molecule,
SMZ. The morphology, adsorption and recognition properties of
Fe3O4@MIPs nanomaterials were characterized using transmis-
sion electron microscope (TEM), scanning electron microscope
(SEM) and Fourier transform infrared (FT-IR) spectrometer. The
adsorption properties were demonstrated by equilibrium rebind-
ing experiments and Langmuir analysis. The selectivity of the
obtained Fe3O4@MIPs NPs was elucidated by the different rebind-
ing capability of SMZ and the related different representatives of
sulfonamides. The Fe3O4@MIPs were able to adsorb them specifi-
cally and applied to separate and enrich SMZ from the poultry feed
samples.
2. Experimental
2.1. Chemicals
Tetraethoxysilane (TEOS), MPS, MAA, EGDMA, SMZ, sulfadi-
azine (SDZ), sulfamerazine (SMR), sulfamethoxazole (SMO) and
sulfadimethoxine (SDM) (Fig. 1) were purchased from Alfa Aesar

O O
O
H2N S NH H2N S NH H2N S NH
O O O
N N N N
O
H3C CH3 H3C O N O CH3 H3C
Sulfamethazine(SMZ) Sulfadimethoxine(SDM) Sulfamethoxazole(SMO)

O O
H2N S NH H2N S NH
O O
N N N N

H3C
Sulfamerazine(SMR) Sulfadiazine(SDZ)

Fig. 1. Molecular structure of the sulfonamides used in the study.

10 X. Kong et al. / J. Chromatogr. A 1245 (2012) 8–16
Chemical Company. Iron chloride, ferric chloride, ammonium
hydroxide (25%), 2-propanol, methanol, ethanol, acetonitrile,
toluene and acetic acid (HAc) were purchased from Tianjin Chem-
icals Ltd. AIBN was purchased form Shanghai Fourth Reagent
Ltd. Highly purified water (18 MK cm−1) obtained from a Water-
Pro Water System (Aquapro Corporation, AFZ-6000-U, China) was
used in the experiments. Al2O3 based columella for solid phase
extraction (SPE) (1000 mg) was purchased from Tianjin Scienhome
Science Instrument Co. Ltd. The poultry feed samples were pur-
chased from Beijing Eastern Hoping Feed Ltd. All reagents used
were of analytical or HPLC grade and used without further treat-
ment.
2.2. Instrumentation and HPLC analysis
The morphology of Fe3O4, silica-coated Fe3O4 (Fe3O4@SiO2) and
Fe3O4@MIPs NPs were examined by Tecnai G2 T2 S-TWIN TEM
and a SS-550 SEM. The infrared spectra of Fe3O4, Fe3O4@SiO2,
Fe3O4@MIPs and Fe3O4@NIPs were recorded with Nicolet AVATAR-
360 FT-IR spectrometer. The identification of crystalline phase of
Fe3O4, Fe3O4@SiO2, vinyl-modified Fe3O4@SiO2 and Fe3O4@MIPs
NPs were performed by a Rigaku D/max/2500v/pc (Japan) X-ray
diffractometer with Cu Ka source. The 2& angles probed were from
10◦ to 80◦ at a rate 4◦ min−1. The magnetic properties were ana-
lyzed with a vibrating sample magnetometer (VSM) (LDJ 9600-1,
USA).
The HPLC analyses were performed on a Shimadzu LC-20A HPLC
system including a variable wavelength UV detector (Shimadzu,
Kyoto, Japan). The instrument control and data processing were
carried out by the LC solution software. A Shimadzu VP-ODS C18
(5 µm particle size, 150 mm × 4.6 mm) analytical column was used
for analyte separation. The mobile phase was methanol–H2O–HAc
(30/70/0.1, v/v/v) delivered at a flow rate of 1.0 mL min−1. The injec-
tion volume was 20 µL, and the column effluent was monitored at
270 nm.
2.3. Preparation of vinyl-modified magnetic NPs
2.3.1. Synthesis of Fe3O4 NPs
FeCl2· 4H2O (1.72 g) and FeCl3 6H2O (4.72 g) was dissolved in
80 mL of deaerated highly purified water contained in a three
· 2.5.2. Specific recognition of Fe3O4@MIPs and Fe3O4@NIPs
neckflask with vigorous stirring (800 rpm) under nitrogen. As the
temperature was elevated to 80 ◦ C, 10 mL of ammonium hydroxide
was added drop by drop, and the reaction was refluxed for 30 min.
The black product was separated by putting the vessel on a Nd–Fe–B
permanent magnet and the supernatant was decanted. The black
precipitate was washed for six times with highly purified water to
remove the unreacted chemicals, then the black product Fe3O4 was
dried in the vacuum.
2.3.2. Synthesis of the Fe3O4@SiO2 NPs
300 mg superparamagnetic NPs were dissolved in 50 mL 2-
propanol and 4 mL of highly purified water by sonication for 15 min,
followed by the addition of 5 mL ammonium hydroxide and 2 mL
TEOS sequentially. The mixture was reacted for 12 h at the room
temperature under a continuous stirring. The resultant product
was collected by an external magnetic field, and rinsed with highly
purified water for six times thoroughly, and dried in the vacuum.
2.3.3. Synthesis of vinyl-modified Fe3O4 NPs
200 mg Fe3O4@SiO2 was dissolved in 50 mL methanol by sonica-
tion for 10 min, then the mixture was stirred vigorously (550 rpm),
which 3 mL of MPS was added drop by drop. The mixture was
reacted for 24 h at the room temperature under a continuous stir-
ring. The resultant product was collected by an external magnetic
field, and rinsed with methanol for six times thoroughly, and dried
in the vacuum.
2.4. Preparation of the core–shell Fe3O4@MIPs and Fe3O4@NIPs
NPs
The MIPs were prepared by following a non-covalent imprinting
approach. SMZ (147 mg) and MAA (112 µL) was mixed in 30 mL sol-
vent (acetonitrile–toluene (3/1, v/v)), and the mixture was put in
refrigerator for 12 h to form template–monomer complex. Vinyl-
modified Fe3O4 (100 mg), EGDMA (1 mL) and AIBN(40 mg) were
added to 30 mL solvent by ultrasonic vibration for 15 min, then
the mixture was stirred under nitrogen following by adding the
template–monomer complex. The mixed solution was heated to
reflux, with first prepolymerized at 50 ◦C for 6 h, polymerized
at 60◦C for 24 h, and further aged at 85 ◦ C for 6 h. The reac- tion
was purged with nitrogen and stirring continued at 300 rpm. The
Fe3O4@MIPs were rinsed with acetonitrile until the super- natant
was clear, and the original SMZ temples in the product were
extracted with methanol–HAc (9:1, v/v). Finally, the product was
repeatedly washed with methanol then dried under vacuum. Non-
imprinted Fe3O4@NIPs were prepared following the same pro-
cedure in the absence of the template SMZ.
2.5. Binding experiments
2.5.1. Binding experiments of Fe3O4@ MIPs and Fe3O4@NIPs
In kinetic adsorption experiments, 10 mg of Fe 3O4@MIPs or
Fe3O4@NIPs were added to 3 mL of 10 µg mL−1 of SMZ solution,
and incubated at regular time intervals from 20 s till to 1200 s.
After the supernatants and polymers were separated by an exter-
nal magnetic field, the concentration of SMZ in the supernatants
was measured by HPLC analysis. The amount of SMZ bound to the
Fe3O4@MIPs or Fe3O4@NIPs was determined by a HPLC method.
In the isothermal binding experiment, 10 mg of Fe3O4@MIPs or
Fe3O4@NIPs were added to 3 mL acetonitrile solution of SMZ of
various concentrations from 0.10 to 20 µg mL−1 and incubated for
10 min at room temperature respectively. The supernatants and
polymers were separated by an external magnetic field, and the
concentration of SMZ in the supernatants was measured by HPLC.
10 mg of Fe3O4@MIPs or Fe3O4@NIPs was added to the 3 mL of
acetonitrile solution containing the mixture of SDZ, SMR, SMZ, SMO
and SDM with a concentration of 10 µg mL−1 for each compound.
After incubation for 10 min at room temperature, the supernatants
and polymers were separated by an external magnetic field, the
amount of SMZ and its analogues in the supernatants was measured
by HPLC.
2.5.3. Separation enrichment and determination of SMZ in
poultry feed samples
The purchased poultry feed samples were selected for spiked
sample analysis. The spiking concentrations for SMZ were 0.1,
0.5 and 1 mg kg−1. We added the 20 µL of standard acetonitrile
solution of SMZ with the concentration of 100 µg mL−1 into 2 g
feed to get the spiked level with 1.0 mg kg−1. The spiked level of
0.1 and 0.5 mg kg−1 were obtained by the same procedure. Prior
to Fe3O4@MIPs NPs extraction, the pre-enrichment procedure of
spiked feed sample was performed according to National Stan-
dard of the People’s Republic of China [42]. Typically, 2 g feed was
placed in a 80 mL polypropylene tube. Then, the spiking SMZ stan-
dard solution and 25 mL acetonitrile was added. The mixed sample
was shaken for 30 min and ultrasonic for 10 min to extract SMZ.
After centrifugation at 3000 rpm for 10 min, the supernatant solu-
tion was filtered through a 0.22 µm filter. Then 25 mL acetonitrile
 X. Kong et al. / J. Chromatogr. A 1245 (2012) 8–16 11

Fig. 2. Overview of synthesis of Fe3O4@MIPs NPs and their application for extraction of sulfonamide with the help of an applied magnetic field.

was added to the subsidence again and repeated the extraction pro-
cess. The totally 50 mL of filtrate was cleaned by SPE method (Al2O3
based columella, 10 mL acetonitrile for activation, 5 mL sample for
loading, 5 mL acetonitrile–water (1:3, v/v) for elution). The resulted
sample was collected, evaporated to dryness under a stream of
nitrogen, then dissolved in the same volume of acetonitrile for
the later adsorption. For the cleanup step by Fe3O4@MIPs, 30 mg
of Fe3O4@MIPs were added in 5 mL of the sample solution, then
incubated for 10 min at room temperature. The supernatants and
polymers were separated by an external magnetic field. Finally,
after discarding the supernatant solution, the Fe3O4@MIPs which
had absorbed the target molecule were eluted with methanol–HAc
(9:1, v/v), and the elution was collected, evaporated to dryness
under a stream of nitrogen, and the residue was condensed in
0.5 mL of methanol–water (3:7, v/v) and measured by HPLC.
3. Result and discussion
3.1. Preparation of imprinted magnetic nanoparticles
and template molecule (SMZ). Finally, after the templates were
removed, the Fe3O4@MIPs NPs were obtained. Therefore, mag-
netic Fe3O4@MIPs NPs were expected to provide accessibility of the
recognition sites to the targets and to be easily removed in the pres-
ence of an external magnetic field. For comparison, the Fe3O4@NIPs
were also prepared using an identical procedure, but without the
addition of template SMZ.
3.2. Characterization of imprinted magnetic nanoparticles
The products of Fe3O4, Fe3O4@SiO2, Fe3O4@MIPs and
Fe3O4@NIPs were investigated by FT-IR spectroscopy (Fig. 3).
The peak at 574 cm−1 is attributed to the stretch of Fe O (Fig. 3A).
In comparison with the infrared data of pure Fe 3O4, the character-
istic peaks of Si O Si group at about 1093 cm−1 and OH group at
about 1628 cm−1 and 3405 cm−1(Fig. 3B–D) indicate the formation
of silica coating on the surface of Fe3O4. As can be seen from
Fig. 3C and D, the peaks of C O group at 1254 cm−1 and C O group

The synthesis of Fe3O4@MIPs via a multistep procedure is illus-

trated in Fig. 2. In this protocol, Fe3O4 NPs were synthesized (A)
by a co-precipitation method [43]. The surface of the Fe3O4 NPs
was then transformed to a silica shell using TEOS [44] to afford
Fe3O4@SiO2. Silica was selected to encapsulate the magnetite NPs (B)
1628 574
because it could not only screen the magnetic dipolar attraction
between magnetite NPs, favor the dispersion of magnetite NPs in 3405
liquid media, but also protect them from leaching in an acidic envi- 1093
ronment. Furthermore, a silica coating on the surface of magnetic
NPs may make them biocompatible and provide them with a silica- (C)
2987
like surface easily modified with various groups for bioconjugation 1254
1731
purpose [45].
The surface hydroxyl groups of Fe3O4@SiO2 were further (D)
reacted with MPS to introduce vinyl groups, which subsequently 2989
reacted with the functional monomer MAA. In addition, the
1259
1731
Fe3O4@SiO2 NPs had a small diameter with an extremely high 4000 3500 3000 2500 2000 1500 1000 500
surface-to-volume ratio, most of the templates were easily situ-
Wavenumber (cm-1)
ated at the surface of the magnetic NPs. The MIPs shells were coated
on the surface of Fe3O4@SiO2 by the copolymerization of functional Fig. 3. FT-IR spectra of Fe3O4 (A), Fe3O4@SiO2 (B), Fe3O4@MIPs (C) and Fe3O4@NIPs
monomer (MAA), cross-linking agent (EGDMA), the initiator (AIBN) (D).
12
(311)
(220)
(400)
Intensity[a.u.]
0 10 20 30 40
2 (degree)
Fig. 4. XRD patterns of Fe3O4 (A), Fe3O4@SiO2 (B), vinyl-modified Fe3O4 (C) and
Fe3O4@MIPs (D).
at 1731 cm−1 and C H group of methyl at 2987 cm −1 indicating
the methacrylic acid layer had been successfully formed on the
surface of Fe3O4@SiO2. In addition, the curves of Fe3O4@MIPs and
Fe3O4@NIPs have almost the same characteristic peaks, which
show the complete removal of templates. These results demon-
strate the coatings of MIPs and NIPs on the surface Fe3O4@SiO2 are
very stable and high-productivity.
The X-ray diffraction (XRD) patterns for the synthesized Fe3O4,
Fe3O4@SiO2, vinyl-modified Fe3O4 and Fe3O4@MIPs are displayed
in Fig. 4. In the 2& range of 10–80◦, six characteristic peaks for
Fe3O4 (2& = 30.38◦, 35.58◦, 43.14◦, 53.48◦, 57.08◦, and 62.66◦) were
observed for the four samples, and the peak positions at the
corresponding 2& value were indexed as (2 2 0), (3 1 1), (4 0 0),
(4 2 2), (5 1 1) and (4 4 0), respectively, which matched well with
the database of magnetite in the JCPDS-International Center for
Diffraction Data (JCPDS Card: 19-629) file. The XRD patterns show
the presence of the specific diffraction peaks of the synthesized
particles, which are highly crystalline materials. However, it is
insufficient to exclude the possibility of μ-Fe2O3, there are proba-
bly two types of iron oxide particles in this dispersion: maghemite
Fig. 5. TEM images of Fe3O4 (A), Fe 3O4@SiO2 (B) and Fe3O4@MIPs(C); SEM images of Fe3O4@MIPs (D, E) with different magnification.
X. Kong et al. / J. Chromatogr. A 1245 (2012) 8–16
and magnetite [46]. The trace amounts of maghemite could be
attributed to the oxidation of Fe3O4 to μ-Fe2O3 during the copre-
cipitation and silanization process [47]. Because they have similar
(440)
magnetic properties, the identification is not important in the
(511)
(422)
present study.
(A) The size and shape of Fe3O4, Fe3O4@SiO2 and Fe3O4@MIPs were
examined by TEM and SEM technique. It can be observed that the
mean diameter of Fe3O4 was about 8 nm (Fig. 5A) and the core–shell
(B)
structure of Fe3O4@SiO2 NPs with silica coatings were success-
fully prepared (Fig. 5B). After coating with SiO2, the diameter of
(C) the Fe3O4@SiO2 increased to approximately 13 nm, corresponding
to a 5 nm thickness of SiO2 layers were uniformly coated on the
Fe3O4 for all samples, and there were hardly any free Fe 3O4 NPs
(D)
in the TEM view. With SMZ as the template, the average diameter
of Fe3O4@MIPs increased to about 200 nm (Fig. 5C–E) with a rela-
50 60 70 80 tively narrow size distribution. This uniformly core–shell structure
will be effective to the mass transport between solution and the
shell surface of Fe3O4@MIPs.
Vibrating sample magnetometry (VSM) was employed to study
the magnetic properties of the synthesized magnetic NPs, and
the magnetic hysteresis loop of the dried samples at room tem-
perature is illustrated in Fig. 6. It is obvious that there is no
hysteresis, both remanence and coercivity are zero, suggesting
that the samples are superparamagnetism. The saturation magne-
tization values obtained at room temperature were 38.4 emu g−1,
22.2 emu g−1, 12.8 emu g−1 and 4.5 emu g−1 for Fe3O4, Fe3O4@SiO2,
vinyl-modified Fe3O4@SiO2 and Fe3O4@MIPs (Fig. 6) respectively.
The theoretical value of saturation magnetization for bulk mag-
netite is reported to be 92 emu g −1 [48,49]. The decrease in
magnetization value can be attributed to the small particle sur-
face effect such as magnetically inactive layer containing spins
that are not collinear with the magnetic field [50]. The satura-
tion magnetization of Fe3O4@MIPs were reduced to 4.48 emu g−1
in comparison with the pure Fe3O4, but remained strongly mag-
netic at room temperature and allowed for as effective magnetic
separation carrier. Fig. 7A shows the separation and redispersion
process of Fe3O4@MIPs. In the absence of an external magnetic field,
a dark homogeneous dispersion exists. When an external magnetic
field was applied, the black particles were attracted to the wall of
vial and the dispersion became clear and transparent. The super-
paramagnetism of Fe3O4@MIPs prevents MIPs from aggregating
 X. Kong et al. / J. Chromatogr. A 1245 (2012) 8–16 13

40 (A) 350
Fe3O4@MIPs
30 300
(B) Fe3O4@NIPs
20
(C) 250
10
(D)
M(emu/g)

200

Q(g/g)
0

-10 150

-20
100
-30
50
-40
0
-6000 -4000 -2000 0 2000 4000 6000
(A) (B) (C) (D) (E)
H(Oe)
Fig. 8. Adsorption of SMZ onto Fe3O4@MIPs and Fe3O4@NIPs in acetonitrile (A),
Fig. 6. Magnetization curve at 298 K with Fe3O4 (A), Fe3O4@SiO2 (B), vinyl-modified acetonitrile/water = 9/1 (v/v) (B), methanol (C), methanol/water = 9/1 (v/v) (D) and
Fe3O4 (C) and Fe3O4@MIPs (D). acetonitrile/toluene = 3/1 (v/v) (E).

and enables them to redisperse rapidly after the magnetic field is

removed (Fig. 7B).
3.3. Binding properties of imprinted magnetic NPs
3.3.1. Adsorption of Fe3O4@MIPs and Fe3O4@NIPs NPs
In order to find a suitable solvent for the rebinding of templates,
we tried several kinds of solvents in the adsorption test. The results
were presented in Fig. 8. As can be seen, the addition of water
can disturb the specific binding effect of the Fe3O4@MIPs, perhaps
because the strong polarity of water destroyed the non-covalent
bonding (hydrogen bond) between Fe3O4@MIPs and template
molecules, so did the methanol. Furthermore, we tried the synthe-
sis solvent as showed, but the result was not as good as acetonitrile,
so the acetonitrile was chosen as the adsorption solvent in the later
work.
The recognition ability of Fe3O4@MIPs NPs towards SMZ was
investigated by an adsorption kinetics test, a steady-state binding
method and subsequently Langmuir analysis. The adsorption kinet-
ics of 10 µg mL−1 SMZ solution onto Fe3O4@MIPs and Fe3O4@NIPs
was presented in Fig. 9. It can be seen that the adsorption capacity
increased with time and the imprinted NPs had a fast adsorp-
tion rate. The adsorption capacity increased rapidly in the first
90 s and almost reached equilibrium after 200 s. The equilibrium
time of the Fe3O4@MIP in this work is also shorter than that of
the other surface imprinting technology for sulfonamides [27,51].
The imprinted Fe3O4@MIPs need only 90–200 s to reach adsorp-
tion equilibrium for templates, which make SMZ molecules reach
the surface imprinting cavities of Fe3O4@MIPs easily. This demon-
strated the resulting Fe3O4@MIPs have the property of good mass
transport and thus overcome some drawbacks of traditional pack-
ing imprinted materials.
The binding isotherms of SMZ onto Fe3O4@MIPs and
Fe3O4@NIPs were determined in the concentration range of
0.10–20 µg mL−1 (initial concentration) and the results were
shown in Fig. 9B, from which we can see that the amount of SMZ
bound to the Fe3O4@MIPs increased quickly along with increasing
the initial concentration when it was below 10 µg mL−1. However,
when the initial concentration was above 10 µg mL−1, the adsorp-
tion curve became relatively flat and reached saturation at high
SMZ concentration. The amount of SMZ bound to Fe3O4@MIPs was
dramatically higher than that of Fe3O4@NIPs at the same initial
concentration.
The saturation binding data were further processed with
Langmuir equation to estimate the binding properties of the
Fe3O4@MIPs. The Langmuir equation is as follows:
[SMZ] 1 [SMZ]
= +
Q (QmaxKD) Qmax
where Q is the amount of SMZ bound to Fe3O4@MIPs at equilib-
rium, Qmax is the apparent maximum adsorption capacity, [SMZ] is
the free analytical concentration at equilibrium and KD is the dis-
sociation constant. The values of KD and the Qmax can be calculated
from the slope and intercept of the linear plotted in [SMZ]/Q versus
[SMZ].

Fig. 7. Separation of Fe3O4@MIPs by a magnet (A); redispersion of Fe3O4@MIPs after removing the magnet (B).
14 X. Kong et al. / J. Chromatogr. A 1245 (2012) 8–16

350 (A)

150 Fe3O4@MIPs
300 Fe 3O4 @MIPs
Fe3O4@NIPs
250 Fe3O4@NIPs

200
Q(g/g)

100

Q(g/g)
150

100
50
50

0
0 200 400 600 800 1000 1200 0
t(s) SDZ SMR SMZ SMO SDM

350 Fig. 10. Adsorption capacity of Fe3O4@MIPs and Fe3O4@NIPs towards five sulfon-
(B) amides in the mixture in which SDZ, SMR, SMZ, SMO and SDM with a concentration
300 of 10 µg mL−1 for each compound.

250 3.3.2. Binding specificity of Fe3O4@MIPs and Fe3O4@NIPs

In order to verify that the Fe3O4@MIPs is selective for SMZ,
200
four sulfonamides (SDZ, SMR, SMO and SDM) in Fig. 1 were
Q(g/g)

selected as structure analogues. The adsorption of Fe3O4@MIPs and

150
Fe3O4@NIPs to mixed solutions of SMZ, SDZ, SMR, SMO and SDM in
Fe3O4@MIPs
100 3 mL of acetonitrile with a concentration of 10 µg mL−1 respective
Fe3O4@NIPs
was listed in Fig. 10 and Table 1.
50 The specificity of Fe3O4@MIPs was estimated by the partition
coefficients of the selected sulfonamide between polymers and
0 solution. The partition coefficient K was determined according to
0 5 10 15 20 the following formula [52]:
Concentration(g/mL) Cp
K=
Cs
(C)
0.06 where Cp is the amount of test analyte bound by Fe3O4@MIPs or
Fe3O4@NIPs NPs and Cs is the concentration of test analyte remain-
0.05 ing in solution.
Additionally, the imprinting factor (IF) and selectivity coefficient
0.04 (SC) were generally used to evaluate the selectivity properties of
C/Q(g/mL)

Fe3O4@MIPs and Fe3O4@NIPs towards SMZ and structurally related

0.03
sulfonamides SDZ, SMR, SMO and SDM. The IF and SC was calculated
by the following formulas [53]:
0.02
Ki
Imprinting factor (IF) =
0.01 Kc

0.00
0 5 10 15
Concentration(g/mL)
Fig. 9. Adsorption kinetics of Fe3O4@MIPs and Fe3O4@NIPs (A); Adsorption
isotherm of SMZ onto Fe3O4@MIPs and Fe3O4@NIPs (B); Langmuir plot to estimate
the binding mechanism of Fe3O4@MIPs towards SMZ (C).
The Langmuir analysis for Fe3O4@MIPs was performed. Accord-
ing to Langmuir model, adsorption was occurred uniformly on
the active sites of the adsorbent. Once a template molecule occu-
pied the site, no further adsorption could take place at this site.
It was observed that the experimental data was fitted to Lang-
muir adsorption isotherm model (Fig. 9C). The adsorption process
of SMZ onto Fe3O4@MIPs could be considered the monolayer
adsorption. The linear regression equation for the linear region is
[SMZ]/Q = 0.00308 + 0.0029[SMZ] (r = 0.9958). From the slope and
the intercept of the straight line obtained, the values of KD and
Qmax were 0.94 mL µg−1 and 344.8 µg g−1 respectively.
IFSMZ
Selectivity coefficient (SC) =
20 IFi
where Ki and Kc represent the partition coefficients of analyte for
the Fe3O4@MIPs and Fe3O4@NIPs, IFSMZ and IFi are the imprinting
factor for SMZ and four other sulfonamides respectively.
As shown in Fig. 10 and Table 1, the binding amount of SMZ
for Fe O @MIPs is much higher than those of four other sulfon-
3 4
amides, meaning that the template molecule has a relatively higher
affinity for the imprinted polymer than its analogues. Moreover,
the IF of SMZ is also much higher than those of four other sulfon-
amides. As shown in Fig. 1, the structures of the five sulfonamides
are analogous with the same sulfa group (NH 2C6H4NH ), but the
difference is that SMZ, SMR, SDM, SDZ all have two N atoms in the
pyrimidine ring whereas SMO has one N atom and one O atom in
the oxazole ring. Because of the steric hindrance effect, the SMZ
fits the cavities in the imprinted polymers best. At the same time,
SMZ was imprinted on the silica layers during the preparation of
imprinted polymers, after removal of SMZ, the complementarity
of cavities in imprinted polymers to the template both in size and
 X. Kong et al. / J. Chromatogr. A 1245 (2012) 8–16 15

Table 1
The adsorption capacity, partition coefficients, imprinting factors and selectivity coefficients of SMZ and its analogues SDZ, SMR, SMO and SDM for the imprinted Fe3O4@MIPs
and control Fe3O4@NIPs NPs.a

Analyte QMIPb (µg g−1) QNIPb (µg g−1) KMIP (mL g−1) KNIP (mL g−1) IF SC

SMZ 163.8 18.1 17.3 1.8 9.5 –

SDZ 35.1 8.6 3.6 0.9 4.1 2.3
SMR 40.8 20.5 4.1 2.1 2.0 4.8
SMO 39.3 11.2 4.0 1.1 3.6 2.7
SDM 56.9 35.7 5.8 3.6 1.6 5.9
a
In this experiment, 10 mg of Fe3O4@MIPs or Fe3O4@NIPs NPs, were added to the 3 mL of acetonitrile solution containing the mixture of SDZ, SMR, SMZ, SMO and SDM
with a concentration of 10 µg mL−1 for each compound.
b
Subscripts “MIP” and “NIP” represent imprinted Fe 3O4@MIPs and control Fe3O4@NIPs NPs, respectively.

Table 2
Recoveries of SMZ form spiked poultry feed samples (n = 5).

Samples SMZ

0.1 mg kg−1 0.5 mg kg−1 1 mg kg−1

Recovery (%) RSD(%) Recovery (%) RSD(%) Recovery (%) RSD (%)
Duck feed 63.3 5.0 71.6 4.8 76.5 2.7
Chicken feed 68.7 2.7 72.3 6.3 74.7 6.7

shape were formed. In addition, we astonishingly found that the
IF of SMZ in the sulfonamides mixture is much higher than that in
the single SMZ solution from the data in Fig. 10 and Table 1. This
phenomenon is probably ascribed that the binding sites are fixed in
Fe3O4@MIPs and Fe3O4@NIPs NPs and five sulfonamides competed
with each other in the mixed solution, which led to the decrease
of adsorption of five sulfonamides in both imprinted and non-
imprinted polymers. Furthermore, because the specific sites existed
in the imprinted polymers are complementary in shape, size and
spatial distribution to template molecule, the template molecule
has advantage in occupying the binding sites over the other sul-
fonamides. The adsorption of five sulfonamides on Fe 3O4@NIPs
is non-specific, so every sulfonamide has the equal opportunity
to occupy the fixed binding sites. This resulted in the adsorption
capacity of template molecule on Fe3O4@NIPs in the mixed solution
is smaller than that of the Fe3O4@NIPs in the single SMZ solution,
but the adsorption capacity of template molecule on Fe3O4@MIPs
is in the other way. These results further verified the excellent
imprinting efficiency of the present method.
3.3.3. Selective enrichment and determination of SMZ in poultry
feed samples
The prepared Fe3O4@MIPs were applied to selective isolation
and enrichment of SMZ in poultry feed samples. The chro-
matograms of spiked poultry feed samples at a concentration of
1 mg kg−1, the elution of adsorbed Fe3O4@MIPs and Fe3O4@NIPs to
SMZ washed with a mixing methanol–HAc (1:1, v/v) after adsorb-
ing poultry feed samples are displayed in Fig. 11. Fig. 11A is
the chromatogram of feed sample spiked SMZ at a concentra-
tion of 1 mg kg−1 after pre-enrichment of feed sample spiked SMZ
of 1 mg kg−1 using method of National Standards of the People’s
Republic of China [42]. The detection limit of the standard method
is 2 mg kg−1 for sulfonamides in feeds and standard method also
lacks the selective enrichment, so the peak of SMZ could not be
seen from chromatograms of the poultry feed samples spiked with
SMZ at the concentration of 1 mg kg−1 (Fig. 11A). After the enrich-
ment of spiked poultry feed samples with Fe3O4@MIPs, and washed
by methanol–HAc (1:1, v/v), the peak of SMZ appeared distinctly
at 8.66 min and other irrelevant compounds in the feed samples
were nearly eliminated (Fig. 11B). Besides, the peak appeared
at 10.13 min was probably interference from the feed samples,
but the curves of the elution of Fe3O4@MIPs and Fe3O4@NIPs
were performed by condensed to 10 times versus the spiked
feed samples. The chromatograms confirmed that SMZ in spiked
poultry feed samples were selectively enriched by Fe3O4@MIPs and
could be recovered by the washing step. As shown in Fig. 11C,
the extraction selectivity of Fe3O4@NIPs was much lower than
those of Fe3O4@MIPs. The enrichment of Fe3O4@NIPs showed a
lack of selectivity, and some other unconcerned compounds were
enriched besides the SMZ. The developed Fe3O4@MIPs NPs method
is only more step using Fe3O4@MIPs as adsorbent for the same feed
sample in comparison with standard method. This demonstrated
that the developed method based on Fe3O4@MIPs NPs not only
has the lower detection limit, but also selectively enrichment for
sulfonamides in feeds.
To evaluate the accuracy and application of the developed
method, the feed samples spiked with three levels of SMZ (0.1, 0.5,
and 1 mg kg−1) were analyzed. At each concentration, five mea-
surements were performed (Table 2). The recoveries of duck and
chicken feed samples ranged from 63.3 to 76.5% and 68.7 to 74.7%,
respectively. The relative standard deviations (RSD) were less than
6.7%. The recoveries of feed samples of less than 80% are similar to
the recoveries of four tetracycline antibiotics [54] and þ2-agonists
spiked feed samples [55]. In our recent work [22], the recoveries
of SDZ in milk samples ranged from 69.8% to 80.2%, which were
(C)
(B)
(A)
0 2 4 6 8 10 12 14 16 18
t (min)
Fig. 11. Chromatograms of duck feed samples. (A) Samples spiked with SMZ at the
concentration of 1 mg kg−1, elution of (B) Fe3O4@MIPs and (C) Fe3O4@NIPs washed
with a mixture of methanol/HAc (1:1, v/v) after Fe 3O4@MIPs and Fe 3O4@NIPs
adsorbing feed spiked samples, respectively.
16 X. Kong et al. / J. Chromatogr. A 1245 (2012) 8–16
a little better than this work. This demonstrated the developed
Fe3O4@MIPs NPs method is reliable to selectively enrich SMZ in
contrast to Fe3O4@NIPs in the complicated samples.
The calibration curve ranging from 0.050 to 20 µg mL−1 with
the highly linear regression coefficients (r > 0.9998) by injecting
of 20 µL standard solution was obtained for SMZ. The detection
limit which was calculated as the concentration corresponding to a
signal-to-noise ratio of three was 14.6 µg L−1 for SMZ. These results
revealed that the Fe3O4@MIPs can be directly used for selective
adsorption and determination of SMZ in feed samples.
4. Conclusion
In this work, a general method was presented for the syn-
thesis of a MIPs layer on magnetic Fe3O4 NPs with uniform
core–shell structure by combining surface imprinting and nan-
otechnique. The resulting Fe3O4@MIPs NPs possess a highly
improved imprinting effect, fast adsorption kinetics with only
90–200 s to reach adsorption equilibrium and high adsorption
capacity (Qmax = 344.8 µg g−1), and can be applied to extract sul-
fonamide in the poultry feed. The imprinting factor of template
molecule in the mixture is 9.5, which is much higher than that in
the single solution. Successful application in the selective separa-
tion and enrichment of SMZ from poultry feed samples and good
recovery after a reasonably mild elution suggested that the func-
tional MIPs coated magnetic NPs could be an alternative solution
for selectively capturing antibiotic residuals in a complex matrix.
Acknowledgments
This work was supported by the National Basic Research
Program of China (No. 2012CB910601), the National Natural
Science Foundation of China (Nos. 20935001 and 20875050)
and the National Natural Science Foundation of Tianjin (No.
10JCZDJC17600).
References
[1] X.L. Wang, K. Li, D.S. Shi, N. Xiong, X. Jin, J. Agric. Food Chem. 55 (2007) 2072.
[2] T. Korpimäki, V. Hagren, E.C. Brockmann, M. Tuomola, Anal. Chem. 76 (2004)
3091.
[3] N. Furusawa, Anal. Chim. Acta 481 (2003) 255.
[4] G. Stoev, A. Michailova, J. Chromatogr. A 871 (2000) 37.
[5] A. Preechaworapun, S. Chuanuwatanakul, Y. Einaga, K. Grudpan, S. Motomizu,
O. Chailapakul, Talanta 68 (2006) 1726.
[6] B. Chiavarino, M.E. Crestoni, A.D. Marzio, S. Fornarini, J. Chromatogr. B 706
(1998) 269.
[7] V.B. Reeves, J. Chromatogr. B 723 (1999) 127.
[8] S. Bogialli, R. Curini, A.D. Corcia, M. Nazzari, R. Samperi, Anal. Chem. 75 (2003)
1798.
[9] M.E.P. Hows, D. Perrett, J. Kay, J. Chromatogr. A 768 (1997) 97.
[10] B. Shao, D. Dong, Y.N. Wu, J.Y. Hu, J. Meng, X.M. Tu, S.K. Xu, Anal. Chim. Acta
546 (2005) 174.
[11] O. Ramstöm, K. Skudar, J. Haines, P. Patel, O. Brüggemann, J. Agric. Food Chem.
49 (2001) 2105.
[12] C. Baggiani, L. Anfossi, C. Giovannoli, Anal. Chim. Acta 591 (2007) 29.
[13] F. Puoci, M. Curcio, G. Cirillo, F. Iemma, U.G. Spizzirri, N. Picci, Food Chem. 106
(2008) 836.
[14] B. Boyd, H. Björk, J. Billing, O. Shimelis, S. Axelsson, M. Leonora, E. Yilmaz, J.
Chromatogr. A 1174 (2007) 63.
[15] F.G. Tamayo, E. Turiel, A. Martín-Esteban, J. Chromatogr. A 1152 (2007) 32.
[16] X.L. Sun, X.W. He, Y.K. Zhang, L.X. Chen, Talanta 79 (2009) 926.
[17] Y. Ge, A.P.F. Turner, Chem. Eur. J. 15 (2009) 8100.
[18] M. Yan, O. Ramström, Molecularly Imprinted Materials: Science and Technol-
ogy, Marcel Dekker, New York, 2005.
[19] N. Zheng, Y.Z. Li, M.J. Wen, J. Chromatogr. A 1033 (2004) 179.
[20] X.J. Liu, C.B. Ouyang, R. Zhao, D.H. Shangguan, Y. Chen, G.Q. Liu, Anal. Chim. Acta
571 (2006) 235.
[21] M. Valtchev, B.S. Palm, M. Schiller, U. Steinfeld, J. Hazard. Mater. 170 (2009)
722.
[22] R.X. Gao, J.J. Zhang, X.W. He, L.X. Chen, Y.K. Zhang, Anal. Bioanal. Chem. 398
(2010) 451.
[23] L.G. Chen, X.P. Zhang, L. Sun, Y. Xu, Q.L. Zeng, H. Wang, H.Y. Xu, A. Yu, H.Q.
Zhang, L. Ding, J. Agric. Food Chem. 57 (2009) 10073.
[24] Z.Y. Chen, R. Zhao, D.H. Shangguan, G.Q. Liu, Biomed. Chromatogr. 19 (2005)
533.
[25] M.P. Davies, V.D. Biasi, D. Perrett, Anal. Chim. Acta 504 (2004) 7.
[26] A. Guzmán-Vázquez de Prada, P. Martínez-Ruiz, A.J. Reviejo, J.M. Pingarrón,
Anal. Chim. Acta 539 (2005) 125.
[27] J.X. He, S. Wang, G.Z. Fang, H.P. Zhu, Y. Zhang, J. Agric. Food Chem. 56 (2008)
2919.
[28] S.F. Su, M. Zhang, B.L. Li, H.Y. Zhang, X.C. Dong, Talanta 76 (2008) 1141.
[29] E. Yilmaz, K. Haupt, K. Mosbach, Angew. Chem. Int. Ed. 39 (2000) 2115.
[30] A. Bossi, S.A. Piletsky, E.V. Piletska, P.G. Righetti, A.P.F. Turner, Anal. Chem. 73
(2001) 5281.
[31] Y. Li, H.H. Yang, Q.H. You, Z.X. Zhuang, X.R. Wang, Anal. Chem. 78 (2006) 317.
[32] R.X. Gao, X.Q. Su, X.W. He, L.X. Chen, Y.K. Zhang, Talanta 83 (2011) 757.
[33] R.X. Gao, X. Kong, F.H. Su, X.W. He, L.X. Chen, Y.K. Zhang, J. Chromatogr. A 1217
(2010) 8095.
[34] S.R. Carter, S. Rimmer, Adv. Funct. Mater. 14 (2004) 553.
[35] C.H. Lu, W.H. Zhou, B. Han, H.H. Yang, X. Chen, X.R. Wang, Anal. Chem. 79 (2007)
5457.
[36] D.M. Gao, Z.P. Zhang, M.H. Wu, C.G. Xie, G.J. Guan, D.P. Wang, J. Am. Chem. Soc.
129 (2007) 7859.
[37] X. Wang, L.Y. Wang, X.W. He, Y.K. Zhang, L.X. Chen, Talanta 78 (2009) 327.
[38] L. Li, X.W. He, L.X. Chen, Y.K. Zhang, Chem. Asian J. 4 (2009) 286.
[39] R.X. Gao, X. Kong, X. Wang, X.W. He, L.X. Chen, Y.K. Zhang, J. Mater. Chem. 21
(2011) 17863.
[40] L. Li, X.W. He, L.X. Chen, Y.K. Zhang, Sci. China Ser. B: Chem. 52 (2009) 1402.
[41] D. Horák, M. Babič, H. Macková, M.J. Beneš, J. Sep. Sci. 30 (2007) 1751.
[42] National Standard of the People’s Republic of China, GB/T 19542-2007, Beijing,
2007.
[43] J.R. Jeong, S.J. Lee, J.D. Kim, S.C. Shin, Phys. Stat. Sol. B 241 (7) (2004) 1593.
[44] M. Zhang, D. Cheng, X.W. He, L.X. Chen, Y.K. Zhang, Chem. Asian J. 5 (2010)
1332.
[45] D.K. Yi, S.S. Lee, J.Y. Ying, Chem. Mater. 18 (2006) 2459.
[46] Y. Lu, Y. Yin, B.T. Mayers, Y. Xia, Nano Lett. 2 (2002) 183.
[47] E. Tronc, P. Belleville, J.P. Jolivet, J. Livage, Langmuir 8 (1992) 313.
[48] V.S. Zaitsev, D.S. Filimonov, I.A. Presnyakov, R.J. Gambino, B. Chu, J. Colloid
Interface Sci. 212 (1999) 49.
[49] J. Popplewell, L. Sakhnini, J. Magn. Magn. Mater. 149 (1995) 72.
[50] R.H. Kodama, A.E. Berkowitz, E.J. McNiff-Jr., S. Foner, Phys. Rev. Lett. 77 (1996)
394.
[51] Y.Q. Li, X. Li, Y. Li, C.K. Dong, P.F. Jin, J.Y. Qi, Biomaterials 30 (2009) 3205.
[52] J. Zhou, X.W. He, Anal. Chim. Acta 381 (1999) 85.
[53] D.M. Han, G.Z. Fang, X.P. Yan, J. Chromatogr. A 1100 (2005) 131.
[54] X.G. Hu, J.L. Pan, Y.L. Hu, Y. Huo, G.K. Li, J. Chromatogr. A 1188 (2008) 97.
[55] Z.G. Xu, Y.F. Hu, Y.L. Hu, G.K. Li, J. Chromatogr. A 1217 (2010) 3612.