Synthesis and characterization of superparamagnetic iron nanocomposites by hydrazine reduction

Article abstract of DOI:10.1016/j.jallcom.2004.03.137

Superparamagnetic iron nanocomposites were synthesized via a modified oxidation-reduction process. Magnetic iron nanoparticles were prepared first with a moderate reducing agent such as hydrazine. Then, iron nanoparticles obtained were modified by nickel ions in aqueous solution to form core-shell structure nanocomposites. The morphology and magnetic properties of iron nanocomposites were examined by transmission electron microscopy (TEM), VSM and SQUID magnetometer. Temperature dependent magnetization and susceptibility data indicate that the nanocomposites are Superparamagnetic above 110 K. The diameter of the iron nanocomposites is determined to be about 6 nm from TEM. showing good agreement with the value acquired by Langevin function fit. X-ray diffraction (XRD) was used to investigate the crystal structure of the iron nanocomposites.

Full text of DOI:10.1016/j.jallcom.2004.03.137

Journal of Alloys and Compounds 385 (2004) 283–287
Synthesis and characterization of superparamagnetic
iron nanocomposites by hydrazine reduction
∗
Chengli Yang, Jianmin Xing, Yueping Guan, Junguo Liu, Huizhou Liu
Laboratory of Separation Science and Engineering and State Key Laboratory of Biochemical Engineering, Institute of Process Engineering,
Chinese Academy of Sciences, P.O. Box 353, Beijing 100080, PR China
Received 7 October 2003; received in revised form 3 March 2004; accepted 16 March 2004
Abstract
Superparamagnetic iron nanocomposites were synthesized via a modified oxidation–reduction process. Magnetic iron nanoparticles were
prepared first with a moderate reducing agent such as hydrazine. Then, iron nanoparticles obtained were modified by nickel ions in aqueous
solution to form core–shell structure nanocomposites. The morphology and magnetic properties of iron nanocomposites were examined by
transmission electron microscopy (TEM), VSM and SQUID magnetometer. Temperature dependent magnetization and susceptibility data
indicate that the nanocomposites are superparamagnetic above 110 K. The diameter of the iron nanocomposites is determined to be about
6
nm from TEM, showing good agreement with the value acquired by Langevin function fit. X-ray diffraction (XRD) was used to investigate
the crystal structure of the iron nanocomposites.
2004 Elsevier B.V. All rights reserved.
©
Keywords: Synthesis; Characterization; Iron nanocomposites; Superparamagnetic
1
. Introduction
than metallic iron [9], which will reduce their magnetism
largely. Thus, the iron nanoparticles produced have no super-
paramagnetic characteristics and their applications are also
limited.
Superparamagnetic nanoparticles have been widely used
in the fields of ferrofluid technology [1], magnetic catalyst
supports [2], contrast enhancement of magnetic resonance
imaging (MRI) [3] and magnetically guided site-specific
drugs delivery [4] due to their special magnetic proper-
ties, i.e. the magnetization curve does not show hystere-
sis. Superparamagnetic iron nanoparticles are also impor-
tant materials for their strong magnetism in comparison
with any other magnetic particles, such as magnetite and
ferrite. In the past decade, there have been quite a num-
ber of researches on the reduction of iron ions in aque-
ous solution to form iron nanoparticles by strong reduc-
ing agent such as KBH4 [5], NaBH4 [6,7] and LiAlH4 [8].
However, as the reducing agents mentioned above are so
strong that the nanoparticles produced are not uniform and
their sizes are larger than their critical sizes. At the same
time, the reduction of iron ions by KBH4 in aqueous so-
lution would rather produce various compounds like Fe2B
To make full use of advantages of magnetic nanopar-
ticles, magnetic nanoparticles with good physicochemical
properties are necessary and desirable. The ideal magnetic
nanoparticles are expected to be of high magnetic properties,
small size with narrow size distribution, and superparamag-
netic characteristics. However, the magnetic iron nanopar-
ticles prepared by conventional methods were of wide size
distribution and have no superparamagnetic characteristics
[10]. In the present work, the synthesis of iron nanoparticles
by reduction of ferrous ion using hydrazine as a reducing
reagent in the presence of surfactant was successfully per-
formed. The surface of iron nanoparticles coated by the so-
lution containing nickel ions produced the stable iron com-
posites of nanostructure, which shows superparamagnetic
characteristics. The magnetism and structure of the prod-
ucts were characterized with VSM, SQUID magnetometer,
transmission electron microscopy (TEM) and X-ray diffrac-
tion (XRD). The formation of iron nanocomposites was also
discussed as well as the results obtained from magnetic sus-
ceptibility experiments.
∗
Corresponding author. Tel.: +86-10-62555005;
fax: +86-10-62554264.
E-mail address: hzliu@home.ipe.ac.cn (H. Liu).
0
925-8388/$ – see front matter © 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.jallcom.2004.03.137

2
84
C. Yang et al. / Journal of Alloys and Compounds 385 (2004) 283–287
2
. Experimental
3. Results and discussion
2
.1. Materials
3.1. Morphology of magnetic iron nanocomposites
Ferrous chloride tetrahydrate (FeCl2·4H2O), ammonium
The iron nanoparticles were prepared by the oxidation–
reduction reaction in ferrous ion solution with hydrazine
as a moderate reducing agent. At the same time, the iron
nanoparticles were also modified from nickel ions solution
in the presence of accessory ingredient such as ammonium
hydroxide to form iron nanocomposites. Nitrogen gas was
injected into the solution to prevent the oxidation of iron
nanoparticels in the aqueous solution and also to control the
particle size. The surfactant, SDS, was added to make the
iron nanocomposites disperse uniformly in the aqueous so-
lution. Fig. 1 shows the TEM photograph of iron nanocom-
posites with average diameter of about 6 nm by statistical
measurement of 500 particles from different regions of the
TEM pictures. From the image, we found the produced
nanocomposites were connected to each other through mag-
netostatic reaction and surface tension. Due to the equilib-
rium of self-repulsion and attracting, self-assembly and ag-
glomeration were occurred when magnetic iron nanocom-
posites were dispersed on a carbon-coated copper grid.
hydroxide, sodium dodecylbenzene sulfonate (SDS), nickel
sulfate hexahydrate (NiSO4·6H2O), sodium acetate and hy-
drazine (N2H4·H2O) were of reagent grade and were used
without any further purification. Water was deionized and
deoxygenated prior to use.
2
.2. Preparation of iron nanoparticles
The preparation of iron nanoparticles was carried out as
follows: 3.5 g FeCl2·4H2O and 0.5 g SDS were dissolved
in 100 ml deionized water under nitrogen gas with vigorous
◦
stirring at 70 C, 50 ml N2H4·H2O of 80% concentration
was then added dropwise into the solution in 10 min. After
several minutes, iron nanoparticles were obtained.
2
.3. Surface modification of iron nanoparticles
Ammonium hydroxide (4 ml) was added into the 100 ml
3
.2. Crystal structure of iron nanocomposites
water solution containing 0.8 g NiSO4·6H2O and 0.02 g
sodium acetate to form solution with pH 11–12, which was
transferred into solution with iron nanoparticles prepared
as given above to let the reaction proceed for 30 min. The
magnetic particles obtained were then isolated from the liq-
uid and solid mixture by magnetic decantation and washed
with deionized water and ethanol for several times. The
Fig. 2 shows the X-ray diffraction patterns of iron
nanocomposites. It was apparent from the X-ray diffraction
data that the standard iron crystals with cubic structure have
three diffraction peaks: {1 1 0}, {2 0 0}, {2 1 1}. Fig. 2 also
shows that the cubic structure of nickel oxidized have five
diffraction peaks:{1 1 1}, {2 0 0}, {2 2 0}, {4 2 0}, {4 2 2}.
As the shape and positions were far away from those of
Ni2O3 and metallic Ni, a conclusion can be drawn that the
nickel oxide obtained was NiO. Because the metallic nickel
deposited on the surface of iron nanoparticles was in con-
tact with air, the metallic nickel was oxidized to the nickel
oxide, which is probably NiO.
◦
solid products were dried in vacuum at 60 C for 12 h.
2
.4. Characterization
Transmission electron microscopy (JEM100GX-II) was
used to measure the morphology and size of samples ob-
tained. The samples for TEM analyses were obtained by di-
luting the dispersed solution with ethanol and then placing a
drop of the diluted solution onto a covered copper grid and
evaporated in air at room temperature. Before the samples
were withdrawn, the nanocomposites dispersed ethanol so-
lutions were sonicated for 5 min to obtain the better particles
dispersion on the copper grid. Vibrating sample magnetome-
ter (VSM, EG & G Princeton Applied Research Vibration
Sample Magnetometer, Model 155) was used to investigate
the magnetic properties of nanocomposites by measuring the
magnetization as a function of magnetic field intensity. The
temperature dependence of magnetization for iron nanocom-
posites was investigated by a superconducting quantum in-
terference device (SQUID) magnetometer (MPMS5, Quan-
tum Design) at a temperature ranging from 5 to 300 K under
different applied magnetic fields. Powder X-ray diffraction
It is well known that the nanometer sized iron particles
are very easy to be oxidized in air. However, from the XRD
pattern (Fig. 2), except pure cubic iron, no other iron oxide
(Fe O or Fe O ) phase was detected, which might be re-
2
3
3
4
sulted from the protection of NiO on the surface of nanopar-
(MAC, MXP21VAHF, Cu K␣) was used to investigate the
crystal structure of samples.
Fig. 1. TEM of magnetic iron nanocomposites.

C. Yang et al. / Journal of Alloys and Compounds 385 (2004) 283–287
285
50
40
30
20
10
0
T =110K
B
0
100
200
300
T(K)
Fig. 4. Temperature dependence of magnetization for iron nanocomposites
at applied field of 800 Oe, the blocking temperature is about 110 K.
Fig. 2. X-ray diffraction patterns of magnetic iron nanocomposites.
ticles. On the other hand, a thin oxide layer on the surface
of iron particles might be useful to protect metals from fur-
ther oxidation. Because no iron oxide was detected by XRD
from the nanocomposites, only a very small fraction in crys-
talline form exists.
characteristics of superparamagnetism, showing the block-
ing temperature (TB) at which the magnetization curve with
the zero-field-cooled (ZFC) exhibits a cusp [11]. Fig. 4
shows the temperature dependence of MZFC for the iron
nanocomposites at an applied field of 800 Oe. While the
temperature is below about 110 K, MZFC decreases rapidly
with decreasing temperature revealing TB = 110 K.
3
.3. Magnetic properties of iron nanocomposites
In the theory of superparamagnetism, the magnetization
of magnetic nanoparticles is free to be aligned to the applied
field above the blocking temperature [12]. In consequence,
magnetization curves versus applied fields should exhibit no
hysteresis in superparamagnetic state. Then, the magnetiza-
tion M is described by the standard Langevin function de-
pending on the following relation:
The magnetic properties of iron nanocomposites were
tested by VSM and SQUID magnetometer at temperature
ranging from 5 to 300 K under different applied magnetic
fields. Fig. 3 shows the magnetization versus field plots for
the iron nanocomposites at temperature of 5 K (a) and 300 K
(b). Fig. 3(b) shows that hysteresis disappeared with a lit-
tle remanence and coercivity, indicating the absence of a
long-range magnetic dipole–dipole interaction among the
superparamagnetic iron nanocomposites. When the temper-
ature is decreased to 5 K, the magnetization of the sam-
ple increases with a symmetric hysteresis loop as shown in
Fig. 3(a), showing a transition from superparamagnetic to
ferromagnetic behavior.
M = M0[coth(µH/kBT) − 1/(µH/kBT)]
where M = Nµ (M , saturation magnetization of the sam-
0
0
ple at a given temperature and N, total number of particles)
[13]. The particle moment µ is given by the M ꢀVꢁ where
s
M is the saturation magnetization of the bulk phase and ꢀVꢁ
s
In magnetism, the temperature dependence of the mag-
netization for nano-sized magnetic particles reveals typical
is the average volume of iron particles. Fig. 5 shows a good
100
1
00
Experimental data
Langevin fit
T=5K (a)
T=300K (b)
80
8
6
4
2
0
0
0
0
0
60
40
20
0
-20
-40
-60
-80
-20
-40
-60
-80
-100
-100
-10
-8
-6
-4
-2
0
2
4
6
8
10
-10
-8
- 6
-4
- 2
0
2
4
6
8
10
Applied Magnetic Field (KOe)
Applied Magnetic Field (KOe)
Fig. 3. Magnetization curve of magnetic iron nanocomposites at (a) 5 K
and (b) 300 K.
Fig. 5. The plot of magnetization as a function of applied field with
Langevin fit at 300 K.

2
86
C. Yang et al. / Journal of Alloys and Compounds 385 (2004) 283–287
+
agreement of experimental data with Langevin function fit
for the iron nanocomposites in superparamagnetic state at
2
Fe2 + N2H4 + 4OH− → 2Fe + N2 + 4H2O
(1)
(2)
Ni2
+ + 6NH3 · H2O → [Ni(NH3)6]
2+
+ 6H2O
3
00 K. The magnetization data of the iron nanocomposites
−
1
at 300 K reveals M0 = 87.2 emu g . Using the saturation
3
magnetization of the bulk iron (1707 emu/cm ) [14], we can
[Ni(NH3)6] + N2H4 + 4OH−
→ 2Ni + N + 12NH + 4H O
2+
2
estimate the average volume ꢀVꢁ of Fe particles. The av-
−
20
3
(3)
2
3
2
erage volume is determined to be 8.2 × 10
cm , which
corresponds to a spherical Fe particle with a diameter of
During the course of the reaction, the pH of solution
5
6
.4 nm. This is in good agreement with the diameter of about
nm determined by TEM micrographs of this sample. When
plays an important role in the formation of nanoparticles.
The reduction of iron and nickel ions will be effectively
achieved by N2H4, which is a reducing agent dependent on
the pH of solution. This can be deduced from the standard
electrode potentials in acidic and alkali solution, namely
the composite nanoparticle is assumed to have a core–shell
structure (Fe–NiO) with a total diameter of 6 nm determined
by TEM micrographs of the sample, the chemical composi-
tion obtained from calculation (Fe:NiO = 6:1 in molar ra-
tio) corresponds to 5.6 nm Fe core diameter and 0.4 nm shell
thickness of NiO. The values to be compared for examin-
ing the core–shell structure are now 5.4 and 5.6 nm, which
may be regarded as a good agreement. The critical size of
Fe particle is 18 nm [15], so it is also concluded that the iron
nanocomposite is enough smaller than its critical size, and
shows superparamagnetic characteristics.
−
0.23 and −1.16 V for the electrode reactions of N2H4 ⇔
+
−
N2 + 4H + 4e and N2H4 + 4OH ⇔ N2 + 4H2O + 4e,
respectively [16]. The latter is desirable for the reduction
2+
2+
since the redox potential for Fe
and [Ni(NH3)6]
is
−
0.44 and −0.49 V, respectively. Therefore, in the case
that the alkalinity of solution is enough, the iron and
nickel ions will be reduced by N2H4 to metallic iron and
nickel.
Compared to the conventional methods [5–8], because
hydrazine was a moderate reducing agent, the speed of the
oxidation–reduction reaction might be controlled by adjust-
ing the temperature during the reaction. The products pre-
pared were metallic irons, and their sizes were very small
and uniform, which were smaller than their critical sizes. At
the same time, after the nickels on the surface of the iron
nanoparticles were oxidized in air to nickel oxides, the iron
nanoparticles will be coated more firmly to form core–shell
structure iron nanocomposites.
Based on the studies, results of the characterization and
analysis given above, the formation of iron nanocomposites
might be described by the model as shown in Fig. 6.
We assumed that the surface of iron nanoparticles
(Fig. 6(a)) prepared were adsorbed uniformly by nickel
ions (Fig. 6(b)), nickel ion will be reduced by reducing
agent to the metallic nickel particles, which deposited uni-
formly on the surface of iron nanoparticles in the presence
of accessory ingredient (Fig. 6(c)). Metallic nickel was oxi-
dized under the air atmosphere to form core–shell structure
(
Fe–NiO) nanocomposites (Fig. 6(d)).
During the formation of nanocomposites, metallic nickel
3
.4. Mechanism of reactions involved
was not only from the in situ electrochemical reaction of
iron with nickel ions but also from the oxidation–reduction
of hydrazine with nickel ions, so the iron nanoparticles were
coated firmly by the metallic nickel. When metallic nickel
deposited on the surface of iron nanoparticles was oxidized
first to nickel oxides in the air atmosphere, iron nanoparti-
cles were coated more compactly by nickel oxides. Thus, it
is very difficult for oxygen in the atmosphere to approach
to the iron through the NiO thin coating to produce iron
oxide between the outer layer of NiO and the inner Fe
core. As a result, the iron nanocomposites were very stable
and showed superparamagnetic characteristics with strong
magnetism.
According to the preliminary experiments, no iron parti-
cles were formed in aqueous solution by adding sufficient
◦
amount of hydrazine at 25 C, even after 2 days. But it was
found that the formation of iron nanoparticles might be com-
◦
pleted within 15 min at 70 C. Also, it was found that in-
creasing the pH of solution was necessary for the formation
of iron nanoparticles and an elevated reaction temperature
was quite helpful in accelerating the reaction rate. There-
◦
fore, the reaction temperature was fixed at 70 C and the pH
of the solution was controlled at more than 10, in this work.
A reaction scheme, in the present study, could be ex-
pressed as the following equations:
(a)
(b)
(c)
(d)
Fig. 6. Formation of iron nanocomposites.

C. Yang et al. / Journal of Alloys and Compounds 385 (2004) 283–287
287
4
. Conclusions
References
The results of the present study show that iron nanopar-
[1] K. Raj, R. Moskowitz, R. Casciari, J. Magn. Magn. Mater. 149
1995) 174.
2] G.D. Zhang, Y.P. Guan, G.B. Shan, H.Z. Liu, Chin. J. Proc. Eng. 4
2002) 319–324.
(
ticles could be synthesized with a moderate reducing agent
such as hydrazine, and iron nanoparticles could be modified
to form the nanocomposites structure, that is, the corona was
NiO, and the core was metal iron. The iron nanocomposites
were stable and showed superparamagnetic characteristics
[
(
[
[
3] D.G. Mitchell, J. Magn. Reson. Imag. 7 (1997) 1–5.
4] P.K. Cupta, C.T. Hung, Life Sci. 44 (1989) 175–186.
[5] C.Y. Wang, Z.Y. Chen, B. Cheng, Mater. Sci. Eng. B60 (1999) 223–
26.
6] L.J. Cain, S.R. Harrison, A. Jacqueline, J. Magn. Magn. Mater. 155
1996) 67–69.
7] M. Arturo, J. Rivas, J. Colloid Interface Sci. 158 (1993) 446–
51.
[8] M.Q. Li, J. Colloid Interface Sci. 186 (1997) 498.
9] A. Corrias, G. Ennas, Chem. Mater. 12 (1993) 1722.
10] C.P. Zhang, W. Deng, Z.C. Hu, L. Hu, Chin. J. Appl. Chem. 3 (2000)
48–251.
−
1
2
with its saturation magnetization of 87.2 emu g . The iron
nanocomposites were very useful for ferrofluid technology
and magnetic catalyst supports. They can also be modified
to form magnetic polymer microspheres for the immobiliza-
tion of proteins and affinity ligands.
[
[
(
4
[
[
[
[
[
[
[
[
2
Acknowledgements
11] M. Yoon, Y.M. Kim, Y. Kim, V. Volkov, H.J. Song, I.W. Park, J.
Magn. Magn. Mater. 265 (2003) 357–362.
12] C.P. Bean, J.D. Livingston, J. Appl. Phys. 30 (1959) 120–
This work was financially supported by the National High
Technology Research and Development Program of China
no. 2002AA302211), the National Natural Science Foun-
dation of China (no. 20206032) and the Outstanding Young
Scientist Foundation of China (no. 29925617). The authors
would like to express their gratitude to Prof. Jiayong Chen
and Dr. Fan Li at the Institute of Process Engineering, Chi-
nese Academy of Sciences, for their valuable help in exper-
iment and writing.
129.
(
13] I.S. Jacobs, C.P. Bean, G.T. Rado, H. Shul, Magnetism, Academic
Press, New York, 1963.
14] R.C. O’Handley, Modern Magnetic Materials, Wiley, New York,
2
000.
15] Y.C. Guo, Iron Magnetism, People’s Education Press, Beijing,
965.
1
16] K. Yu, D.J. Kim, H.S. Chung, H.Z. Liang, Mater. Lett. 57 (2003)
3992–3997.