Magnetic Fe3O4@chitosan nanoparticle: Synthesis, characterization and application as catalyst carrier

Article abstract of DOI:10.1166/jnn.2010.2549

A novel method was developed to prepare Fe₃O₄@CS beads with core-shell structure using a double-crosslinking process. Before the coating process, an unique crosslinking agent, glutaraldehyde (GA), was adsorbed onto the surface of Fe₃O₄ in advance, so the subsequent CS can uniformly coat around the magnetic core processed from the strong interaction between GA and CS, forming a perfect core-shell structure. The obtained Fe₃O₄@CS beads were followed by the Pd deposition through in-situ reduction method, and the prepared composite catalyst was applied exemplarily in synthesizing nabumetone to check its reusing property. The nanoparticles were characterized by transmission electron microscopy (TEM), X-ray powder diffraction (XRD), Fourier transformation infrared spectroscopy (FTIR), X-ray photoelectron spectroscopy (XPS), and the magnetic hysteresis loop determination method. This novel composite catalyst showed admirable potential in reusable catalysis. Copyright

Full text of DOI:10.1166/jnn.2010.2549

Copyright © 2010 American Scientific Publishers
All rights reserved
Printed in the United States of America
Journal of
Nanoscience and Nanotechnology
Vol. 10, 6348–6355, 2010
Magnetic Fe O @Chitosan Nanoparticle: Synthesis,
3
4
Characterization and Application as Catalyst Carrier
1
1
1
2
1ꢀ2ꢀ∗
3ꢀ∗
Linghao He , Lu Yao , Fujun Liu , Bing Qin , Rui Song
, and Wei Huang
1
Key Lab of Surface and Interface Sciences of Henan Provincial, Zhengzhou University of Light Industry, Zhengzhou, China 450005
2
College of Chemistry and Chemical Engineering, Graduate University of Chinese Academy of Sciences, Beijing, China 100049
3
Laboratory of Materials of High Technology, Institute of Chemistry, Chinese Academy of Sciences, Beijing, China 100080
A novel method was developed to prepare Fe O @CS beads with core–shell structure using a
3
4
double-crosslinking process. Before the coating process, an unique crosslinking agent, glutaralde-
hyde (GA), was adsorbed onto the surface of Fe O in advance, so the subsequent CS can uniformly
3
4
coat around the magnetic core processed from the strong interaction between GA and CS, forming
a perfect core–shell structure. The obtained Fe O @CS beads were followed by the Pd deposi-
3
4
tion through in-situ reduction method, and the prepared composite catalyst was applied exemplarily
in synthesizing nabumetone to check its reusing property. The nanoparticles were characterized
by transmission electron microscopy (TEM), X-ray powder diffraction (XRD), Fourier transformation
infrared spectroscopy (FTIR), X-ray photoelectron spectroscopy (XPS), and the magnetic hysteresis
loop determination method. This novel composite catalyst showed admirable potential in reusable
catalysis.
Keywords: Superparamagnetic Nanoparticles, Double-Crosslinking, Core–Shell Structure, Pd
Deposition, Reusable Catalyst.
1
. INTRODUCTION
coprecipitation of an aqueous solution of ferrous and fer-
1
1
ric ions by a base, oxidation of the ferrous hydroxide
Superparamagnetic nanoparticles are attracting increasing
attentions for their advanced properties and wide applica-
tions such as magnetic resonance imaging (MRI) contrast
enhancement, tissue repair, immunoassay, detoxification
of biological fluids, hyperthermia, drug delivery and
12
13
gels using KNO3 or (NH ) S O , ꢀ-ray irradiation,
4 2
14
2
8
15
microwave plasma synthesis, polyol-mediated sol–gel,
sonochemistry, oil-in-water emulsion route using a small
amount of cyclohexane as the oil phase, hydrothermal
16
2
3
4
17
5
6
7
18
19
reaction, nonaqueous route, etc. However, the method
of chemical coprecipitation, in the view of economy and
environmental benignancy, will be the most ideal among
the above methods because water which is nontoxic is
used as the reaction medium and it doesn’t require too
much heat to maintain a high temperature as in the case
of hydrothermal reaction.
8
cell separation, etc. In the last decade, several types of
iron oxides have been successively carried out in the field
of superparamagnetic nanoparticles (mostly maghemite or
magnetite, with the diameter of about 5–20 nm), among
which the magnetite, Fe O , is a very promising candidate
for its outstanding biocompatibility. Magnetite, Fe O , is a
3
4
3
4
common paramagnetic iron oxide that has a cubic inverse
spinel structure, where oxygen forms an fcc closed pack-
ing and Fe cations occupy the interstitial tetrahedral sites
and octahedral sites, with a stoichiometric ratio of 2:1
In spite of its outstanding biocompatibility, Fe O
3
4
nanoparticles cannot be directly used in the practical appli-
cations with its uncovered surface exposed to the wild
environment, owing to the high surface energy for its
3+
2+
9
20
(Fe /Fe ) in a non-oxidizing oxygen environment. As
large ratio of surface area to volume. In oxidizing envi-
an important half-metallic material, the electrons of mag-
ronment magnetite is not very stable and is readily to
3
+
2+
netite can hop between Fe and Fe irons at room tem-
perature. So various synthetic methods have been reported
for the preparation of nanoscale Fe O particles, includ-
be oxidized into maghemite (ꢀ-Fe O ). Not only non-
2
3
resistant to the oxygen, magnetite is also not stable in
acidic and anaerobic medium, which is always applied
for biomedical applications, since various electron or ion
transfers will occur depending upon the pH of the sus-
pension, which involves an oxidation–reduction reaction of
3
4
1
0
ing the reduction of hematite Fe O by H2 or CO,
2
3
∗
21
Authors to whom correspondence should be addressed.
J. Nanosci. Nanotechnol. 2010, Vol. 10, No. 10
the surface of magnetite. The oxidation–reduction process
6
348
1533-4880/2010/10/6348/008
doi:10.1166/jnn.2010.2549

He et al.
Magnetic Fe O @Chitosan Nanoparticle: Synthesis, Characterization and Application as Catalyst Carrier
3 4
of ferrous ions is mainly made up of the migration of
cations through the lattice framework and the forming of
cationic vacancies to keep the charge, forming the struc-
ture of maghemit. In maghemite, iron ions are distributed
in the octahedral (Oh) and tetrahedral (Td) sites of the
3
+
3+
spinel structure (0.75[Fe ] [Fe
V
] O ), which dif-
Td
5/3 1/3 Oh
3+ 2+
4
4
3
+
fers from magnetite ([Fe ] [Fe Fe ] O ) by the pres-
Td
Oh
ence of cationic vacancies within the octahedral site. The
vacancies ordering scheme is closely related to the sample
preparation method and results in lowering symmetry and
possibly superstructures. The vacancies can be completely
random or partially or totally ordered.
Scheme 1. Schematic drawing of the coating process of Fe O @CS
3
4
nanoparticles, (a) with conventional once-crosslinking; (b) with novel
double-crosslinking. The yellow surroundings denote the original CS
without crosslinking, while the blue ones denote CS crosslinked by GA.
To overcome the instability of magnetite, the surfaces
of these particles must be modified through the creation of
few atomic layers of organic polymer or inorganic metal-
lic (e.g., gold) or oxide surfaces (e.g., silica or aluminum),
suitable for further function by the attachment of various
bioactive molecules.22 One the other hand, the coating of
suitable layer can also prevent the aggregation of mag-
netite nanoparticles caused mainly by van der Waals forces
and magnetic dipolar forces, which specially exist only
for magnetic particles. For magnetic suspensions, magnetic
dipolar forces between two particles introduce anisotropic
interactions, which will be very important when if the
anisotropic interparticle potential is integrated over all
directions. The adding of the coating layer, accomplished
by physical or chemical adsorption, has been proved to be
effective in the prevention of the unexpected aggregation
and improving the chemical stability of composite system.
Chitosan (noted as CS), a cationic linear polysaccharide
composed essentially of ꢁ(1→4) linked glucosamine units
together with some proportion of N-acetylglucosamine
units, is of interest as a potential coating component
because of its excellent biocompatibility and biodegra-
dation property.23 Generally obtained from natural chitin
after deacetylation by alkaline treatments, CS is a nat-
ural, nontoxic and biodegradable polymer that is com-
mercially available and can be produced from shellfish
as crosslinking agent to be absorbed onto the surface of
Fe O before CS loading, to make sure CS can be uni-
3
4
formly deposited on the surrounding of Fe O . After CS
3
4
loading, GA was used again to crosslink the adsorbed CS,
so that the solvent resistance of as-obtained nanosphere
would be enhanced. A schematic drawing of the coat-
ing process is given in Scheme 1 to show how CS can
be adsorbed to form the core–shell structure with this
double crosslinking method. To study the potential of
the obtained Fe O @CS as one catalyst carrier, palla-
3
4
dium was deposited, by a in-situ reduction method on
the surface of magnetic beads, to prepare composite cata-
lyst Fe O @CS/Pd. The role of this catalyst was signified
3
4
in the following synthesizing of nabumetone to investigate
the catalytic efficiency. And the experimental data shown
the catalyst had a high catalytic efficiency as the yield can
reach more than 90; in addition, its recycle property was
checked.
2. EXPERIMENTAL DETAILS
2.1. Materials
All the chemicals were of reagent grade and purified
before used. Deionized water was re-deionized (electrical
ꢀ
resistivity ≥18.2 Mꢂ · cm, 25 C) and deoxygenated by
boiling for 1 h before used. Ferric sulphate (Fe (SO ꢃ ),
2
4
2
4 3
waste. Recently materials scientists started to be inter-
ested in the utilization of CS as a coating material to syn-
thesize composites. Ding et al. synthesized Fe O /CS-PAA
ferrous sulphate heptahydrate (FeSO ·7H O, >99%) were
4
2
obtained from Kemel chemical reagent Co. Ltd., Tian-
3
4
jin, China. Chitosan (deacetylation degree ≥92%, M =
n
hollow spheres using glutaraldehyde (GA) as crosslink-
−
1
−1
7
9 500 g mol , M = 88 700 g mol , PDI = 1.12,
ing agent.25 From the TEM results, the as-obtained hybrid
w
as determined by GPC measurements) used for the
particle coating process was purchased from Huoshuo
fine chemical Co. Ltd., Shanghai, China. Glutaraldehyde
hollow spheres presented an anomalous shape which was
different from uncovered Fe O nanoparticles with an inte-
3
4
grate sphere shape. The reason of the discrepancy between
shapes of Fe O /CS-PAA and Fe O would be the asym-
(GA) 50% water solution was purchased from Kemel
3
4
3
4
chemical reagent Co. Ltd., Tianjin, China. Palladium
metric loading of CS in the surface of Fe O . The asym-
3
4
chloride (PdCl > 99%), 2-bromo-6-methoxynaphthalene
2
metric loading will lead to the incomplete coating and the
partly bareness of Fe O , so the faces with high surface
(
C H BrO, 98%) and 3-buten-2-ol (C H O, 70%) were
11 9 4 8
3
4
purchased from Alfa Aesar.
energy will be exposed to wild environment, which will
weaken the effect of coating.
In this case, we described a novel method, called
2.2. Synthesis of Magnetite
‘
double-crosslinking’ to synthesize Fe O @CS hybrid
Fe O was prepared without any additional stabilizer
3 4
3
4
beads with regularly spherical shape. GA was adopted
according to a controlled coprecipitation approach that
J. Nanosci. Nanotechnol. 10, 6348–6355, 2010
6349

Magnetic Fe O @Chitosan Nanoparticle: Synthesis, Characterization and Application as Catalyst Carrier
He et al.
3
4
could be found elsewhere.11 In a typical synthesis, an
aqueous mixture of ferric and ferrous solution (20.6 g)
was dropped into and alkalinized by 10% NaOH solu-
tion (75 ml) at room temperature under mechanical stirring
the remaining solution was treated through distillation, to
obtain the crystalline powder.
2.5. Characterization
(
500 rpm) in a three necked flask. N gas was inflated
2
during synthesis in a closed system to prevent the oxidiza-
X-ray powder diffraction (XRD) measurement was per-
formed using a Bruker D8 diffractometer with monochro-
matized Cu Kꢅ radiation (ꢆ = 1ꢇ5418 Å), 40 kV, 30 mA.
A continuous scan mode was used to collect 2-theta data
2
+
tion of Fe . When a light black sedimentation appeared,
the crystal growth was allowed to proceed for 1 hour with
constant stirring to produce a stable, water-based suspen-
sion. The expectant products were isolated from the sol-
vent by an external magnetic field followed by redispersed
in deionized water. This washing-redispersion procedure
was repeated several times until this colloid system was
neutrality.
ꢀ
ꢀ
−1
from 10 to 90 with a 0.04 sample pitch and a 4 min
scan rate. Fourier transform infrared spectroscopy (FT-IR,
Tensor 27, Bruker Inc.) was used to conform the chem-
ical abstraction between the starch layer and the surface
of magnetite particles. A JEM 200 CX transmission elec-
tron microscope (TEM) equipped with a CCD was used
for high-resolution images of suspension samples, oper-
ated at an accelerating voltage of 200 kV. The specimen
for TEM imaging was prepared from the Fe O @CS sus-
2
.3. Coating of Magnetite by CS
As the as-prepared Fe O nanoparticles (500 mg) were
3
4
3
4
completely washed and redispersed into a transparent sus-
pension, several drops of aqueous solution of GA, ca.
pension in deionized water under sonication for 5 min.
After sonication, a drop of well-dispersed suspension was
placed on a carbon-coated copper grid, followed by dry-
ing at an oxygen-free environment before it was laid
on the sample holder of the microscope. X-ray photo-
electron spectroscopy (XPS) was performed using a VG
Scientific ESCALab220i-XL photoelectron spectrometer
equipped with a twin anode, providing both unchromatized
AlKꢅ radiations (1486.6 eV). The spectrometer, which
was equipped with a multichannel detector, operated at
300 W and the C1s (284.8 eV) was choose as the reference
line. Magnetic measurements were done in a vibrating
sample magnetometer (VSM, PPMS-9, Quantum Design).
The sample powder was placed in a Teflon-coated sample
holder and the mass was accurately measured. The mass
1
00 ꢄL were dropped to the reaction system with contin-
ual stirring of 500 rpm for 1 hour to form a adsorption
layer round Fe O . Then the above procedures of wash-
3
4
ing and redispersion were adopted again to remove the
no-adsorbed GA. After the redispersion procedure, a CS
solution (1% w/v) prepared by dissolving of 0.2 g of CS
in 20 mg acetic acid was added to the reaction system fol-
lowed by the coating process for 2 hours. To crosslink the
exterior of loaded CS layer, GA was dropped into the flask
again. 2 h later, the crosslinking reaction was completed
and the final products was gathered with a permanence
ꢀ
magnet and dried in an oven at 50 C under vacuum for
2
4 hours.
2
−1
magnetization, M (Am kg ), was measured for different
applied magnetic fields, at 300 K.
2
.4. Deposition of Palladium and Evaluation of
Catalytic Properties
3. RESULTS AND DISCUSSION
To prepare Fe O @CS/Pd, the obtained Fe O @CS dis-
3
4
3
4
perse system (ca. 1.176 g) was mixed with PdCl solution
2
3
.1. Synthesis of Fe O @CS
3
4
(
0.52% w/v) by stirring overnight under N atmosphere to
2
2
+
absorb Pd ion to the amino of CS out of co-ordination.
Then the mixture was heated under reflux at 140 C for
Preparation of the crystal particles in the corresponding
solution under optimum reaction conditions can be divided
into two stages: the formation of tiny crystalline nuclei
occurs when the concentration of the species reaches criti-
cal supersaturation, i.e., nucleation, and then the growth of
the nuclei by diffusion of the solutes to the surface of the
crystal. Here, among all of the particles with slightly dif-
ferent sizes, the large particles will grow larger at the cost
of that with small size. This crystallization mechanism is
called Ostwald ripening and has been generally accepted
ꢀ
4
h, which process was also under N gas inflating. After
2
separated from the solution by a magnet, the orange pro-
duction was dispersed in deionized water with another por-
tion of PdCl solution, followed by the reflux operation.
2
Then the above process was repeated times to obtain the
most palladium deposition.
The catalytic effect of the obtained composite catalyst
was determined via the synthetic experiment of nabume-
tone, an important drug intermediate. The composite cat-
alyst, approximately 210 mg was dispersed in 100 mL
2
6
to be the main path of the crystal growth.
As the comparison given in the scheme of synthesiz-
ing Fe O @CS nanoparticles in Scheme 1, CS is directly
ethanol with 2-bromo-6-methoxynaphthalene (6.7 g) and
3
4
ꢀ
3
-buten-2-ol (4.4 g), refluxed at 140 C for 24 h. Then
absorbed to the surface of Fe O particles by electro-
3 4
the magnetic catalyst was separated by a magnet, and
static force and then crosslinked by subsequently added
6
350
J. Nanosci. Nanotechnol. 10, 6348–6355, 2010

He et al.
Magnetic Fe O @Chitosan Nanoparticle: Synthesis, Characterization and Application as Catalyst Carrier
3 4
crosslinking agent, which will lead to the incomplete coat-
ing and some aggregate (showed in Scheme 1(a)). To
get ideal core–shell structure, the crosslinking agent was
added into the synthesis system in advance, so a layer
of crosslinking agent will be uniformly absorbed to the
surface of magnetic core, to induce a strong electro-
static force between the iron oxide nanoparticles and the
organic crosslinking agent (as in Scheme 1(b)). Thus, via
the double-crosslinking, the biocompatible CS can form a
robust coating layer around the surface of magnetic core,
which leads to a perfect core–shell structure.
As visually manifested in the TEM pictures (Fig. 1).
The size of particles in Figure 1(a and b) are about
60 nm and 17 nm, respectively. The serious aggregation
is distinctly shown in Figure 1(a) for particles synthesized
by the traditional once-crosslinking method, while a well
dispersion of the nanoparticles with the core–shell struc-
ture could be found, implying that the novel double-
crosslinking method will be helpful for the development
of magnetic materials with core–shell structure (Fig. 1(b)).
For comparison, the TEM image of the as prepared pure
Fe O particle was given in Figure 1(c).
3
4
7
0
0
0
0
0
0
0
0
(
(
(
a)
b)
C)
6
5
4
3
2
1
20
40
60
80
100
Particles diameter (nm)
60
5
4
3
2
1
0
0
0
0
0
0
20
40
60
80
100
Particles diameter (nm)
8
0
0
0
0
0
6
4
2
100 nm
20
40
60
80
100
Particles diameter (nm)
Fig. 1. Representative TEM images for of Fe O : (a) Fe O @CS crosslinked once by the traditional method; (b) Fe O @CS crosslinked twice by
3
4
3
4
3
4
novel double-crosslinking method, the insert is the magnefied picture of the area in white square. For comparison, the as prepared Fe O nanoparticles
3
4
were given in (c). The average diameters of the nanoparticles were obtained by measuring about 100 clusters for each sample.
J. Nanosci. Nanotechnol. 10, 6348–6355, 2010
6351

Magnetic Fe O @Chitosan Nanoparticle: Synthesis, Characterization and Application as Catalyst Carrier
He et al.
3
4
Fe O @CS
3
4
1
0
5
0
5
Fe O
3
4
a
–10000
–5000
0
5000
10000
Applied magnetic field (Oe)
–
b
–10
20
40
60
80
2Theta/deg.
Fig. 4. Field dependence of magnetization for the Fe O and
3
4
Fe O @CS nanoparticles.
3
4
Fig. 2. XRD patterns: (a) for Fe O and (b) for Fe O @CS particles.
3
4
3
4
the Fe O nanocrystals have a very small grain size
3
4
3
.2. XRD Results
(being accurately 11.5 nm by means of Debye-Scherrer
formula). The Fe O @CS particles obtained through the
Magnetite (Fe O ) has an inverse spinel structure where
3
4
3
4
2
−
2+
novel double-crosslinking method was also characterized
by XRD (see Fig. 2(b)). Since CS is an amorphous mate-
rial, there is nearly no obvious difference between the
O
ions are cubic close packed (ccp), the Fe ions are
3
+
in octahedral sites, and the Fe ions are half in octahedral
and half in tetrahedral sites. The typical XRD pattern
for as-synthesized pure Fe O and Fe O @CS particles
2
7
XRD patterns of Fe O and that of Fe O @CS, which
3
4
3
4
3
4
3
4
means that the coating of CS layer would not affect the
is given in Figure 2. The grain size can be estimated by
means of the Debye-Scherrer formula based on the full D
width at half-maximum (FWHM) of main diffraction peak:
D = Kꢆ/F WꢈSꢃ·cosꢉ, where D/nm is the crystal size,
K is a constant (usually K = 1), ꢆ/nm is wavelength of
the X ray, FW(S) is the full width at half-maximum of
crystalline structure of Fe O .
3
4
3.3. FTIR Results
To confirm the forming of Fe O @CS core–shell struc-
3
4
2
8
ture generated from the double crosslinking reaction,
FTIR measurements of the Fe O , Fe O adsorbing GA
diffraction peaks, and ꢉ is the diffraction angle.
All the diffraction peaks in Figure 2(a), for as synthe-
sized pure Fe O can be indexed as the inverse spinel
3
4
3
4
(
noted as Fe O /GA) and Fe O @CS particles crosslinked
3 4 3 4
3
4
before and after loading (noted as Fe O /GA/CS and
structure magnetite (JCPDS file number 65-3107). The
phenomenon of peak broadening displayed indicates that
3
4
Fe O /GA/CS/GA, respectively) were carried out. The
3
4
FTIR spectra of different samples are presented in
Figure 3. The characteristic absorption bands of pure
−
Fe O mainly include those at 570 and 375 cm , which
3 4
1
Fe O
3
4
Fe O /GA
3
4
Fe O /GA/CS
3
4
Fe O /GA/CS/GA
3
4
a
b
c
d
(
a)
(b)
(c)
4000
3500
3000
2500
2000
1500
1000
500
Wavenumber/cm–1
Fig. 5. Optical photograph of the as-obtained Fe O @CS nanoparticles
3
4
Fig. 3. IR spectra of Fe O , Fe O /GA, Fe O /GA/CS and
dispersed in a glass battle: (a) without magnetic field, (b) with magnetic
field, (c) after the applied magnetic field was removed.
3
4
3
4
3
4
Fe O /GA/CS/GA.
3
4
6
352
J. Nanosci. Nanotechnol. 10, 6348–6355, 2010

He et al.
Magnetic Fe O @Chitosan Nanoparticle: Synthesis, Characterization and Application as Catalyst Carrier
3 4
the crosslinking chain.32 (iii) When the second crosslink-
ing was done, the outer CS shell was crosslinked ulteriorly,
thus spectra (a) and (b) are very similar besides a small
increase in intensity of N C bond at 1562 cm , which
shows that the second crosslinking had successfully treated
the un-crosslinked CS at the foregone step.
3
5000
4000
3000
Pd 3d5/2
Pd 3d3/2
3
3
Fe 2p3
−1
O 1s
32000
1000
3
3
50 345 340 335 330 325
Binding energy (eV)
Pd 3p
C 1s
3.4. Analysis of Magnetic Properties
a
b
c
The as-obtained Fe @CS nanoparticles are superpara-
O
3 4
magnetic, which can be proved by the zero coerciv-
ity and remanence on a hysteresis loop observed in
the magnetization curves recorded at room temperature
d
(
shown in Fig. 4). The saturation magnetization (Ms) of
1200
1000
800
600
400
200
0
3
3ꢊ34
Binding energy/eV
buck magnetite is 87 emu/g.
However as the par-
ticle size decreased to the nanometer scale, the value
of Ms will remarkably reduce (here the Ms of Fe O
Fig. 6. XPS spectra of samples, (c), (b) and (a) curves for samples
refluxed once, two and three times, respectively; (d) curve for the
Fe O @CS without Pd deposition (The insert for Pd 3d).
3
4
is only 9.8 emu/g), which is due to the particle size
effects. Meanwhile, the magnetization should be normal-
ized by thermogravimetirc analysis with respect to mass
in the Fe O sample, considering the fact that the mag-
3
4
attributed to Fe–O band. For the as-synthesized Fe O
3
4
−
1
particles, the two bands at 634 and 394 cm are the
consequence of the split of the characteristic absorption
3
4
netic hybrid nanoparticles have far less content of mag-
netic substance per gram than iron oxides. The normalized
result shows the saturation magnetization of the CS coated
sample (∼58 emu/g), comparable to results obtained by
−
1
of Fe O at 570 cm that shifts to a higher wavenum-
3
4
−1
ber. Similarly, the band at 446 cm comes from another
absorption band of the Fe–O bond at 375 cm of bulk
Fe O . It should be noted that so far the reason behinds
−
1
3
5
others for iron oxide nanoparticles. In this case, a dead
magnetic layer, mainly made of double-crosslinked CS, is
formed over the surface layer due to the strong interaction
between the magnetite surface and the CS with the help
of the crosslinking agent GA, which prevents the coupled
dipoles of Fe O particles from aligning along the mag-
3
4
the shift of the absorption band of Fe–O bonds of nanopar-
2
9
ticles to higher wavenumber is unclear.
With the followed crosslinking process, the IR spectra
in Figure 3 show three sets of significant changes (i) The
new peaks appearing in curve (c) at 1415, 1720, 2768
3
4
−1
and 2854 cm (attributed to stretching of C–H of –CH3,
O, C–H of –CHO and C–H of –CH , respectively)
netic orientation. Thus there is a tendency toward lower
3
6
C
magnetization values.
2
arising from bonds in GA, which approved the adsorption
of GA onto the surface of Fe O particles by electrostatic
The dispersion and magnetic responsive properties of
as-obtained Fe O @CS were vividly given in the opti-
3
4
3
4
3
0
force. (ii) From curve (b) of Fe O /GA/CS, by use of
cal photographs in Figure 5. Here, Fe O @CS nanopar-
3
4
3 4
−
−1
1
a combination of the O–H stretching band at 3442 cm ,
amide I bands at 1618 cm , amide II band at 1527 cm ,
the bridge oxygen stretching band at 1153 cm , and
the C–O stretching bands at 1062 and 1035 cm , the
percent deacetylation can be determined. The peak at
655 cm can be attributed to an imine bond (N C)
and that at 1562 cm is associated with an ethylenic
bond (C–C), which can be attributed to the increase of
the contribution of the glutaraldehyde molecule in the
chitosan–glutaraldehyde reaction to cause an increase in
ticles has a high solubility and dispersibility in water
(see Fig. 5(a)) and will not aggregate or precipitate for
at least one month, which is highly satisfactory for the
practical application. When subjected to a strong magnetic
field, the particles can be completely separated from the
solution within seconds (see Fig. 5(b)). If the magnetic
field is removed and the resulted was vortexed again, the
Fe O @CS particles will re-disperse back into the orig-
−1
−
−
1
1
3
1
−
1
1
−1
3
4
inal solution to form a uniformly dispersed system (see
Fig. 5(c)).
O
OH
Br
Fe O @CS/Pd
3
4
140°C
O
O
H C
H C
3
3
Scheme 2. Schematic drawing of the synthesis of nabumetone.
J. Nanosci. Nanotechnol. 10, 6348–6355, 2010
6353

Magnetic Fe O @Chitosan Nanoparticle: Synthesis, Characterization and Application as Catalyst Carrier
He et al.
3
4
3
.5. Deposition of Palladium and Evaluation of
Catalytic Properties
could be found in Scheme 2). The product was approved
to be nabumetone by IR characterization (see SI-2 in the
supporting Information part). The yield was derived from
the UV-Vis measurement, and the variation of production
yield with the recycle times indicated that the activity
decreased slightly from 94.7% at the first time to 89.6%
after 20 times (Fig. 7), which suggested that the obtained
catalyst had a perfect recycle property. All these results
imply the high potential of this resultant to be applied in
the filed of magnetic separation.
Chitosan is a special kind of national polymer, having
amino group on C and hydroxyl group on C , which are
2
3
all equatorial bonds in terms of conformation. This spe-
cial structure endows this material with a perfect chelating
property with some heavy metal ions of appropriate ionic
radius, which makes it a hot spot in the field of polymer-
metal catalyst.
The treatment of mixing and stirring was to accelerate
the rate of Pd2 sorption on chitosan out of chelating func-
+
tion, and the modeling of this sorption isotherm and kinet-
4
. CONCLUSION
3
7
ics was reported by Guibal. By in-situ reduction reaction,
2
+
the reflux treatment transformed the absorbed Pd ion to
In summary, a novel double-crosslinking strategy was put
forward to prepare Fe O @CS hybrid nanoparticles with
0
Pd deposition. In order to understand the surface varia-
3
4
tion of catalyst with the increase in the times of reduction,
XPS characterization was employed. Figure 6 shows the
full XPS patterns, in which curves (c), (b) and (a) rep-
resent catalyst Fe O @CS/Pd after the first, second and
excellent core/shell structure, and magnetic responsive
properties. These superparamagnetic Fe O @CS nanoma-
3
4
terials are water soluble because of their surface CS layer,
so the nanoparticles are easily dispersed in water and thus
easily combined to a variety of biological macromolecules,
such as peptides, proteins, and nucleic acids. The sim-
ple dispersion–separation and re-dispersion experiment
results revealed that the as-obtained CS-coated magnetic
nanocrystals synthesized by this novel double-crosslinking
method possess very good magnetism and water solubility.
The deposition of palladium and subsequent synthesis of
nabumetone indicated that the obtained Fe O @CS nano-
3
4
third reduction, respectively. As the reference, curve (d)
represent Fe O @CS without palladium deposition. And
3
4
the insert in Figure 6 shows XPS patterns of Pd 3d of
composite catalyst Fe O @CS/Pd. For catalyst with dif-
3
4
ferent reduction times, the peak of Pd 3d at 336 eV (335
and 340 eV due to Pd 3d5/2 and Pd 3d3/2, respectively,
in the insert) increased with the reduction times. These
results indicate that after the first reduction, Pd atoms were
3
4
bound to the surface of Fe O @CS nanoparticles through
3
4
materials would function well as one carrier of recycle
catalyst.
the coordination of –NH ligand with Pd and then grown
2
into bigger clusters when the reduction was performed fur-
ther, for the later Pd crystallization was initiated by the
former Pd as crystal nucleus. The conclusion was also
approved by the increase in absorption peaks of UV-Vis
for samples after different reduction times (see the SI-1 in
the supporting Information part).
Acknowledgments: This work was supported by
Henan Innovation Project for University Prominent
Research Talents (HAIPURT) program, the President Fund
of GUCAS, and the KF2008-04, from the State Key Lab-
oratory of Environmental Chemistry and Ecotoxicology,
Research Center for Eco-Environmental Sciences, Chinese
Academy of Sciences.
Herein, the catalytic activity of Fe O @CS/Pd was
3
4
investigated by the experiment of synthesizing nabume-
tone (one modification of the Heck reactions, the detail
9
9
9
9
9
8
8
8
8
8
8
6
4
2
0
8
6
4
2
0
References and Notes
1
. J. Schoonman, Solid State Ionics 135, 5 (2000).
2. V. S. Kalambur, E. K. Longmire, and J. C. Bischof, Langmuir
3, 12329 (2007).
2
3. E. Alsberg, E. Feinstein, M. P. Joy, M. Prentiss, and D. E. Ingber,
Tissue Eng. 12, 324 (2006).
4. H. Y. Tsai, C. F. Hsu, I. W. Chiu, and C. B. Fuh, Anal. Chem.
79, 8416 (2007).
5. H. Chen, M. Kaminski, X. Liu, C. Mertz, Y. Xie, M. Torno, and
A. Rosengart, Med. Hypotheses 68, 1071 (2007).
6
7
8
. J. P. Fortin, C. Wilhelm, J. Servais, C. Menager, J. C. Bacri, and
F. Gazeau, J. Am. Chem. Soc. 129, 2628 (2007).
. J. Rubio-Retama, N. E. Zafeiropoulos, and C. Serafinelli, Langmuir
2
3, 10280 (2007).
. Y. Jing, L. R. Moore, and P. S. Williams, Biotechnol. Bioeng.
6, 1139 (2007).
0
2
4
6
8
10
12
14
16
18
20
22
Recycle times
9
Fig. 7. The relation between yields of nabumetone and cycle index of
catalyst.
9. J. P. Jolivet, C. Chaneac, and E. Tronc, Chem. Commun. 5, 481
(2004).
6
354
J. Nanosci. Nanotechnol. 10, 6348–6355, 2010

He et al.
Magnetic Fe O @Chitosan Nanoparticle: Synthesis, Characterization and Application as Catalyst Carrier
3 4
1
0. A. Pineau, N. Kanari, and I. Gaballah, Thermochim. Acta. 447, 89
2006).
1. T. Fried, G. Shemer, and G. Markovich, Adv. Mater. 13, 1158 (2001).
2. A. B. Bourlinos, A. Bakandritsos, and V. Georgakilas, J. Mater. Sci.
25. Y. Ding, Y. Hu, and L. Zhang, Biomacromolecules 7, 1766 (2006).
26. P. W. Voorhees, Annu. Rev. Mater. Sci. 22, 197 (1992).
27. R. M. Cornell and U. Schwertmann, The Iron Oxides: Struc-
ture, Properties, Reactions, Occurrence and Uses, VCH, New York
(1996).
(
1
1
4
1, 5250 (2006).
1
1
3. S. Z. Wang, H. W. Xin, and Y. T. Qian, Mater. Lett. 33, 113 (1997).
4. D. Vollath, D. V. Szabó, R. D. Taylor, and J. O. Willis, J. Mater.
Res. 12, 2175 (1997).
28. D. R. Askeland and P. P. Phulé, Essentials of Materials Science and
Engineering, Thomson Learning, New York (2004).
29. (a) J. Zhang, R. S. Srivastava, and R. D. K. Misra, Langmuir
23, 6342 (2007); (b) M. Yamaura, R. L. Camilo, and L. C. Sampaio
J. Magn. Magn. Mater. 279, 210 (2004).
30. C. K. Yeom and K. H. Lee, J. Membrane Sci. 109, 257 (1996).
31. S. Yoshihiro, M. Hiroaki, and S. Hiroyuki, Int. J. Biol. Macromol.
18, 237 (1996).
1
1
1
1
5. C. Feldmann and H. O. Jungk, Angew. Chem. Int. Ed. 40, 359
(
2001).
6. R. V. R. Kumar, Y. Koltypin, and Y. S. Cohen, J. Mater. Chem.
0, 1125 (2000).
7. Z. Zhou, J. Wang, X. Liu, and H. Chan, J. Mater. Chem. 11, 1704
2001).
8. F. Vereda, J. de Vicente, and R. Hidalgo-Alvarez, Langmuir 23, 3581
2007).
1
(
32. (a) E. R. Blout, M.Fields, and R. Karplus, J. Am. Chem. Soc. 70, 194
(1948); (b) A. C. Oyrton, Monteiro, Jr. and A. Claudio, Int. J. Biol.
Macromol. 26, 119 (1999).
(
1
2
2
9. N. Pinna, Chem. Mater. 17, 3044 (2005).
0. G. Zhao, Chem. Mater. 17, 3154 (2005).
1. M. P. Morales, S. V. Verdaguer, and M. Montero, Chem. Mater.
33. A. E. Berkowitz, J. A. Lahut, and C. E. Vanburen, IEEE T. Magn.
16, 184 (1980).
34. J. M. D. Coey, A. H. Morrish, and G. A. Sawatzky, J. Phys. C: Solid
State Phys. 32, 2386 (1971).
35. (a) Y. C. Chang and D. H. Cheng, J. Colloid Interface Sci. 283, 446
(2005); (b) C. Barrera and C. Rinaldi, J. Mater. Chem. 19, 6870
(2009).
36. R. Kaiser and G. Midkolczy, J. Appl. Phys. 41, 1064 (1970).
37. M. Ruiz, A. M. Sastre, and E. Guibal React. Funct. Polym. 45, 155
(2000).
1
1, 3058 (1999).
2. C. C. Berry and A. S. G. Curtis, J. Phys. D: Appl. Phys. 36, R198
2003).
2
(
2
2
3. F. Quignard, A. Choplin, and A. Domard, Langmuir 16, 9106 (2000).
4. P. A. Sandford, Sources, Chemistry, Biochemistry, Physical Prop-
erties and Applications, edited by Chitin and Chitosan, Elsevier
Applied Science, New York (1989).
Received: 11 August 2009. Accepted: 19 October 2009.
J. Nanosci. Nanotechnol. 10, 6348–6355, 2010
6355