Magnetic and structural studies on CoFe

2
O
4
nanoparticles synthesized by
co-precipitation, normal micelles and reverse micelles methods
Ibrahim Shari
a
, H. Shokrollahi
a,n
, Mohammad Mahdi Doroodmand
b,c,1
, R. Sa
a
a
Electroceramics Group, Materials Science and Engineering Department, Shiraz University of Technology, 71555-313, Shiraz, Iran
b
Chemistry Department, School of Sciences, Shiraz University, Shiraz, Iran
c
Nanotechnology Research Center, Shiraz University, Shiraz, Iran.
a r t i c l e i n f o
Article history:
Received 1 June 2011
Received in revised form
31 December 2011
Available online 20 January 2012
Keywords:
Cobalt ferrite
Magnetic nanoparticle
X-ray diffraction
a b s t r a c t
Cobalt ferrite nanoparticles were synthesized by the chemical co-precipitation, normal micelles and
reverse micelles methods of iron and cobalt chlorides. X-ray diffraction analysis, Fourier Transform
Infrared (FTIR) and Vibrating Sample Magnetometer were carried out at room temperature to study the
structural and magnetic properties. X-ray patterns revealed the production of a broad single cubic
phase with the average particle sizes of 12 nm, 5 nm and 8 nm for co-precipitation, normal micelles
and reverse micelles methods, respectively. The FTIR measurements between 400 and 4000 cm
1
conrmed the intrinsic cation vibrations of spinel structure for each one of the three methods.
Moreover, the average particle sizes were lower than the single domain size (128 nm) and higher than
the super-paramagnetic size (23 nm) at room temperature. The results revealed that the magnetic
properties depend on the particle size and cation distribution, whereas the role of particle size is more
signicant.
& 2012 Elsevier B.V. All rights reserved.
1. Introduction
Traditionally, magnetic materials have played a great number
of crucial roles in the daily life [1]. Among magnetic materials,
magnetic ceramics have been paid special attention due to
chemical stability as well as high electrical resistivity. One of
the most magnetic ceramics is spinel. According to the magnetic
theory, the most important spinels are oxides 2, 3 or MFe
2
O
4
[2,3].
Among spinel ferrites, CoFe
2
O
4
has received special attention
because of its large magneto-crystalline anisotropy, high coerciv-
ity, moderate saturation magnetization, large magneto-strictive
coefcient, chemical stability and mechanical hardness [4].
CoFe
2
O
4
has an inverse spinel structure, where oxygen atoms
constitute an face center cubic (FCC) lattice and where half of the
Fe(III) ions occupies the tetrahedral A, site and where the other
half, together with Co(II) ions is located on the octahedral B site.
The different processing methods [57] inuence the particle size
and cation distributions between the A and B sites, as well as
the magnetic properties. The diameter of the nanoparticles is
much smaller than the critical single-domain diameter, thereby
avoiding the formation of the magnetic domain walls that
decrease the magnetization [8]. Cobalt ferrite (CoFe
2
O
4
) is one
of the most promising candidates for biological applications,
including magnetic resonance imaging (MRI), magnetic uid
hyperthermia (MFH), magnetic separations, biosensors, targeted
and controlled drug delivery [511]. Although the chemical
precipitation route is the most widely used process for the
synthesis of the magnetic nanocrystals with high simplicity and
good grain size control, it leads to the precipitation of the
nanocrystals with a relatively broad size distribution [12]. In
contrast, the nanocrystals, which are prepared by the micro
emulsion method, are generally very ne, shape-controlled and
highly crystalline, as compared to those which are synthesized by
other processes with narrow-sized distribution [1315].
This paper investigates the effect of cation distribution and
particle size in three widely used chemical methods, including co-
precipitation and two types of micro emulsion: normal micelles
and reverse micelles methods on the magnetic and structural
properties of cobalt ferrite.
2. Experimental method
2.1. Materials
All chemical reagentssodium dodecyl sulfate (SDS), ferric
chloride (FeCl
3
.6H
2
O), cobalt(II) chloride (CoCl
2
.6H
2
O), sodium
hydroxide (NaOH), petroleum oil and pyridine were acquired
Contents lists available at SciVerse ScienceDirect
journal homepage: www.elsevier.com/locate/jmmm
Journal of Magnetism and Magnetic Materials
0304-8853/$ - see front matter & 2012 Elsevier B.V. All rights reserved.
doi:10.1016/j.jmmm.2012.01.015
n
Corresponding author. Tel.: 98 711 7353509; fax: 98 711 7354520.
E-mail addresses: shokrollahi@sutech.ac.ir (H. Shokrollahi),
doroodmand@shirazu.ac.ir (M.M. Doroodmand).
1
Tel.: 98 711 6137363; fax: 98 711 2286008.
Journal of Magnetism and Magnetic Materials 324 (2012) 18541861
from E. Merck Co., Germany and used as they received without
further treatment.
2.2. Synthesis
2.2.1. Co-precipitation
The synthesis of the cobalt ferrite nanoparticles was prepared
by the co-precipitating aqueous solutions of CoCl
2
and FeCl
3
mixtures in alkaline medium. In addition, the mixed solutions of
CoCl
2
6H
2
O (100 ml, 1.0 M) and FeCl
3
6H
2
O (100 ml, 2.0 M) were
prepared and kept at 60 1C. This mixture was also added abruptly
to the boiling solution of NaOH (1200 ml, 0.63 M). Then the
solution was stirred with constant velocity. After that the solu-
tions were maintained at 85 1C for 1 h. The time duration to form
the spinel ferrite was sufcient. A sufcient amount of ne
particles was collected at this stage by using magnetic separation.
These nanoparticles were washed several times with distilled
water, after that, with acetone and nally dried them at room
temperature [16].
2.2.2. Normal micelles
Cobalt(II) chloride (6.9 mmol) and iron(II) chloride
(14.9 mmol) were mixed in an aqueous solution (500 ml). An
aqueous surfactant solution of sodium dodecyl sulfate (SDS)
(38.3 mmol) in deionized water (500 ml) was also added to the
mixture. Then the mixture was stirred at room temperature for
30 min to form a mixed micellar solution of Co (DS)
2
and Fe (DS)
2
.
Afterward, it was heated to 5565 1C. A solution of methylamine
(300 ml, 40%, w/w, aqueous solution) in deionized water (700 ml)
was heated to the same temperature and rapidly added to the
surfactant mixture. Black nanoparticles were precipitated. After
the reaction mixture was stirred vigorously for 3 h, the nanopar-
ticles were isolated by centrifugation and washed with copious
amounts of deionized water, ethanol and hexanes through mag-
netic decantation. The nal product was dried in air at 50 1C
overnight to yield cobalt spinel ferrite nanoparticles [17].
2.2.3. Reverse micelles
In this study, the micro emulsion was used as a template to
control the size of magnetic nanoparticles. The micro emulsion
consists of three independent phases, including petroleum oil,
pyridine and partially polar species such as water or alcohol. To
synthesize micro emulsion briey, pyridine was mixed with
petroleum oil with a volume ratio of 1/10. After that this
mixture was shaken vigorously for an hour to obtain the partial
emulsion mixture. This step lasted 1 h. Then the emulsion was
reuxed at temperature to 70 1C in an inert atmosphere of
nitrogen for 2 h. Afterward the emulsion was annealed until the
temperature of the emulation reached the room temperature.
In this reverse micelle, a triplet system of emulsion/iso-
butanol/H
2
O was selected. The CoFe
2
O
4
ferrite, aqueous solution
was prepared by mixing stoichiometric amounts of 0.5 M of FeCl
3
and 0.25 MCoCl
2
6 H
2
O. The two reverse micro-emulsions, ME1
and ME2, were also prepared. Then 33.33 wt% of aqueous
solution containing the precursor salts and 11.11 wt% of isobuta-
nol were added to the oil under magnetic stirring. The stirring
continued for 1 h, resulting in a stable reverse micro-emulsion
ME1. Moreover, the reverse micro-emulsion (ME2) was prepared
with a 4.0 M aqueous solution of NaOH as the water phase under
similar conditions.
The reverse micro-emulsion ME2 was then heated to 80 1C and
added to reverse micro-emulsion ME1 dropwise under constant
magnetic stirring. After few minutes, the appearance of dark
brown color reveals the completion of the reaction and the
formation of the desired ferrite colloidal solution. Then the
reaction mixture was further stirred for 4 h on magnetic stirrer
at 80 1C. Finally, the synthesized nanoparticles were isolated by
centrifugation and washed several times with distilled water and
then, with ethanol, hexane and acetone through magnetic decan-
tation. The nal product was dried in air at 50 1C overnight to
yield cobalt spinel ferrite nanoparticles. In this research work, the
abbreviations of CNP, NNP and RNP are used for co-precipitation,
normal micelles and reverse micelles micro-emulsion methods,
respectively.
2.3. Characterization
The X-ray diffraction (XRD) patterns of the samples were
recorded on a BRUKER X-ray powder diffractometer by using
Cu-K
a1
(1.54060

A) radiation. The scans of the selected diffraction
peaks were carried out in the step mode (step size 0.021,
measurement time 2 s, measurement temperature 25 1C and
standard: Si powder).The lattice parameters, the oxygen position
and the cation distribution were determined by means of Riet-
veld renement [18] and the reex program [19]. In order to
calculate the crystallite size, the Scherrers method was
applied [9]. After that the crystallite size was rened by Rietveld
renement. The magnetic measurements of the prepared powder
were determined at room temperature by using the vibrating
sample magnetometer (VSM, Dexing, Model: 250). The Fourier
Transform Infrared Spectroscopy (FTIR) spectrum was recorded as
(KBr) disks in the range 4004000 cm
1
by using the (FTIR
Shimadzu-8000) spectrophotometer.
3. Results and discussion
3.1. XRD analysis and cation distribution
Fig. 1 shows the single-phase spinel nature of the powders,
which were produced by three different methods and conrmed
by the X-ray diffraction pattern. The nal product is CoFe
2
O
4
with
the expected inverse spinel structure and without any trace of
impurity peaks. The unit cell symmetries can be described by the
space groups fd3

m. The reections are comparatively broader,

revealing the nanosize of the crystals with the highest intensity
and crystallinity in the co-precipitation method [20] Fig. 2.
The degree of crystallinity of ferrite in the different synthesis
routes was estimated by the background subtraction method
based on their XRD patterns. Table 1 lists the crystallite size,
degree of crystallinity and Rietveld renement parameters. The
quality of the renement was quantied by the corresponding
gures of merit: prole residual R
p
, weighted prole residual R
wp
and goodness of t w
2
[19]. The low intensity of the MNPs
obtained by microemulsion suggests poor crystallinity, and the
sharpness of the peaks reects the large grain size (Table 1). In
other words, the peak sharpness indicates a greater degree of
crystallinity.
The degree of crystallinity can be controlled by adjusting the
supersaturation during the nucleation and crystal growth pro-
cesses. The main factors affecting the supersaturation degree are
temperature, pH value and the reactant concentration. In the
coprecipitation method the high crystallinity can be related to the
crystal growth at a faster rate than the nucleation growth. The
crystallinity of the sample synthesized by the copericipitation
method is greater than that of the samples prepared by micro-
emulsion routes [21,22]. The size distribution of microemulsion
method is not as broad as those produced using the coprecipita-
tion, but they are still polydisperse.
To calculate the average grain size from the broadening of the
XRD peaks of CoFe
2
O
4
, the Rietveld renement of XRD pattern
I. Shari et al. / Journal of Magnetism and Magnetic Materials 324 (2012) 18541861 1855
was applied as shown in Fig. 1. Table 2 lists the cation distribution
Formula, experimental lattice parameter (a
exp
), theoretical lattice
parameter (a
th
), oxygen positional parameter (1/4,1/4,1/4) (U
3m
)
and oxygen positional parameter (3/8,3/8,3/8) (u). The difference
in the crystallite size is potentially due to the different prepara-
tion conditions which rise to the different rates of ferrite
formation, thereby favoring the variation in crystallite size. The
micro-reactors in micro-emulsion not only control shape and size
but also decrease the aggregation process of crystals [15].
The distribution of cations over A and B sites is determined
mainly by the whole energy of the crystal that is dependent on
their ionic radius, the Coulomb energy, the crystal eld effects
and the ordering of the cations. The distribution of the cations on
these two sites can be expressed as:Co
2
x
Fe
3
1x

A
Co
2
1x
Fe
3
1x

B
O
4
,
where the square practices contain the ions on the octahedral
Fig. 1. (a) crystal structure of cobalt ferrite and XRD patterns of (b) co-precipitation, (c) normal micelles and (d) reverse micelles.
Fig. 2. Conguration of the ion pairs in spinel ferrites with favorable distances and
angles.
Table 1
Crystallite size, degree of crystallinity and Rietveld renement parameters.
Synthetic
method
Degree of
crystallinity
Crystallite
size (nm)
R
wp
R
p
w
2
CNP 79.93 11.70 9.53 6.81 1.38
NNP 70.57 5.58 9.92 7.60 1.65
RNP 71.50 7.63 8.50 6.51 1.12
Table 2
Experimental lattice parameter (a
exp
), theoretical lattice parameter (a
th
), XRD
crystallite size (D), oxygen positional parameter (1/4,1/4,1/4) (U
3m
), oxygen
positional parameter (3/8,3/8,3/8) (u).
Synthetic
method
Formula a
exp
(nm)
a
th
(nm)
U
3m
(1/4,
1/4,1/4)
u (3/8,
3/8,3/8)
CNP
Co
2
0:16
Fe
3
0:84

A
Co
2
0:84
Fe
3
1:16

B
O
4
0.8403 0.8402 0.2614 0.3830
NNP
Co
2
0:01
Fe
3
0:99

A
Co
2
0:99
Fe
3
1:01

B
O
4
0.8391 0.8399 0.2603 0.3832
RNP
Co
2
0:12
Fe
3
0:88

A
Co
2
0:88
Fe
3
1:12

B
O
4
0.8401 0.8401 0.2611 0.3831
I. Shari et al. / Journal of Magnetism and Magnetic Materials 324 (2012) 18541861 1856
sites, and x is a constant, which is rened by the Rietveld
method as it can be used to determine the cation distribution.
r
A
x r
2
Co
1xr
3
Fe
1
r
B

1
2
1xr
2
Co
1xr
3
Fe
h i
2
Lattice parameters were calculated by the Rietveld method.
In the co-precipitation method, the crystal structure tends to
behave like mixed spinel (Table 2), and achieve a high value of the
lattice parameter. The relation recommended by Mazen [23]
can calculate the lattice parameter theoretically
a
th

8
3

3
p r
A
R
O

3
p
r
B
R
O

h i
3
where R
O
is the radius of the oxygen ions (R1.32

A). The
agreement between a
th
and a
exp
(Table 2) achieved from X-ray
data indirectly conrms the cation distribution deduced from
X-ray intensity calculations. The small deviation between a
th
and
a
exp
may be due to the presence of some ferrous ions Fe(II)
(r
Fe
2
0.078 nm) on octahedral sites with larger radii than Fe(III)
(r
Fe
3
0.0645 nm) [24]. The oxygen, positional parameter and
anion parameter (u) for each composition were calculated using
the formula:
U
3m
1=4 R
2
2=311=48R
2
1=181=2
2R
2
2
4
where R BO=AO.
Table 3 shows the average ionic radius per molecule of the
octahedral and tetrahedral sites (r
A
and r
B
), tetrahedral site radius
(R
t
), octahedral site radius (R
o
), ionic radii(r), tetrahedral edge
(Ed
t
), shared octahedral edge (Ed
so
), unshared octahedral edge
(Ed
uo
) and average bond lengths, (AO and BO), /r
Me
r(O
2
)S.
The highest value of BO length in the normal micelles methods
can be explained by the highest Co(II) ( 0.0745 nm ) occupancy
with a higher ionic radius, as compared to Fe(III) (0.0645 nm) [25].
Both Tables 4 and 5 list the inter-ionic distances and bond
angles [25,26] for cobalt ferrite, which were calculated by use of
the experimental values of crystallographic parameters based on
the Eq. (5). These parameters are necessary to give full description
of the crystallographic structure, and they are also in connection
with the magnetic properties. In order to explain this connection
in the normal micelles method the Fe(III) (5 m
B
) concentration of
the A site is higher than that of the B site, or the Co(II) (3 m
B
)
of the A site is lower than that of the B site, the magnetic
hyperne eld of the A site is greater than that of the B site
(H
A
4H
B
)
M
e
O M
e
M
e
p a5=8u b a=42
1=2
q au1=43
1=2
c a=811
1=2
r au1=811
1=2
d a=43
1=2
s a1=3u1=83
1=2
e 3a=83
1=2
f a=46
1=2
5
R
A
a

3
p
u
1
4

R
O
6
R
B
a 3u
2
2:75u
43
64

1=2
R
O
7
3.2. FTIR analysis
IR is used to collect to obtain information about the structure of
a compound. It is also utilized as an analytical tool for assessing the
purity of a compound. Fig. 3 and Table 6 show the FTIR absorption
bands of CoFe
2
O
4
produced by co-precipitation, micro-emulsion
Table 3
Tetrahedral site radius (R
t
), octahedral site radius (R
o
), ionic radii(r), tetrahedral edge (Ed
t
), shared octahedral edge (Ed
so
), unshared octahedral edge (Ed
uo
), average bond
lengths(AO and BO).
Synthetic method R
A
(nm) R
B
(nm) r
A
(nm) r
B
(nm) Ed
t
(nm) Ed
so
(nm) Ed
uo
(nm) AO (nm) BO (nm)
70.0002 nm
CNP 0.0616 0.0715 0.0662 0.0687 0.3162 0.2776 0.2974 0.1982 0.2008
NNP 0.0616 0.0711 0.0646 0.0694 0.3163 0.2772 0.2970 0.1966 0.2014
RNP 0.0617 0.0714 0.0657 0.0689 0.3164 0.2776 0.2973 0.1977 0.2009
Table 4
Inter-ionic distances for CoFeO system.
Synthetic method b (nm) c (nm) d (nm) e (nm) f (nm) p (nm) q (nm) r (nm) s (nm)
CNP 0.2971 0.3484 0.3638 0.5458 0.5146 0.2033 0.1936 0.3708 0.3678
NNP 0.2967 0.3479 0.3634 0.5450 0.5139 0.2029 0.1936 0.3708 0.3673
MNP 0.2970 0.3483 0.3638 0.5457 0.5144 0.2032 0.1937 0.3710 0.3677
Fig. 3. FT-IR spectra of the CoFe
2
O
4
system.
Table 5
Bond angles (degree) for CoFeO system.
Synthesis method y
1
y
2
y
3
y
4
y
5
CNP 122.690 142.150 93.879 126.144 72.744
NNP 122.636 142.348 93.965 126.175 72.603
RNP 122.658 142.181 93.930 126.155 72.661
I. Shari et al. / Journal of Magnetism and Magnetic Materials 324 (2012) 18541861 1857
methods at room temperature in the wave number range of 400
4000 cm
1
. It is obvious that the higher frequency band (u
1
) is
600 cm
1
and the lower frequency band (u
2
) is 400 cm
1
. The
bands u
1
and u
2
are related to the intrinsic vibration of tetrahedral
and octahedral complexes [27]. The absorption bands observed
within this limit reveal the formation of the spinel structure. This
difference in the band positions is observed because of the difference
in the Fe
3
O
2
distance and (1) the perturbation occurring, thereby
(2) introducing Co(II) ions for tetrahedral and octahedral sites.
The splitting of the octahedral (B site) absorption band near
u
2
is due to the presence of different kinds of cations, including
Co(II), Fe(III) and Fe(II) on the B site: [28]. Based on the small
effect of the production method on the octahedral peak shift and
the splitting effect, the exact calculation of the octahedral force
constant is difcult in the current study. This is attributed to the
JanTellar distortion produced by Fe(II) ions, which produces the
local deformation in the crystal eld potential, thereby splitting
the absorption band. The force constant for the tetrahedral site
(K
T
) was calculated by employing the method suggested by
Woldron [27]. The force constant is the second derivative of
the potential energy with respect to the site radius [25].
k
t
7:62 M
1
u
2
1
10
7N
m
8
M
1
is the molecular weights of cations on the A site. As the
bond length, AO, increases by increasing cobalt ion concentra-
tion on the tetrahedral site, (Tables 2, 3) the required energy for
breaking the longer bonds will decrease. Therefore, the force
constant in the normal micelles method on the A site is slightly
higher than the two other methods. The decrease in the bond
lengths is attributed to the decrease in the lattice parameter and
indicates the increase in the covalent characters. Srivastav and
Srinivasan [29] have stated that the band stretching for the
tetrahedral sites would lead to a higher force constant than that
evaluated for the octahedral site.
3.3. Magnetic properties
The magnetic properties of the materials which originate from
the quantum couplings at the atomic level, including the coupling
between electron spins (SS coupling) and the coupling between
the electron spin and the angular momentum of the electron
orbital (LS coupling) [30]. Each of the magnetic nanoparticles
tends to possess a single magnetic domain. Therefore, the nano-
particles provide excellent opportunities for the fundamental
studies on the relationship between magnetic behavior and the
magnetic couplings at the atomic level. The magnetization curves
of the cobalt ferrite nanoparticles measured at 27 1C are shown in
Fig. 4. It is clear that the synthesis method can inuence the
magnetization curves as a result of a change in the particle size
and distribution of cations.
3.3.1. Saturation magnetization
Table 6 illustrates the dependency of the saturation magnetiza-
tion on the synthesis method. The M
S
value obtained for the
samples varies between 12.6 and 58.4 emu/g and is different from
the bulk value of 80.8 emu/g at room temperature for CoFe
2
O
4
[30].
The high M
S
in the co-precipitation method is mainly due to large
particle sizes and is slightly due to the high cation difference
between the two sub-lattices (Table 2). These two effects are
discussed here in detail.
3.3.1.1. Size effect. When the diameter of the nanoparticles is
much smaller than the critical single-domain diameter, avoiding
the formation of magnetic domain walls decreases the
magnetization.
Even though the particle size becomes even smaller, the
thermal stability of the magnetization orientation decreases.
The limited superparamagnetic, as it is known, refers to the
particle size because the thermal energy at room temperature
causes uctuations in the magnetization orientation of a bit.
Magnetic anisotropies such as the magnetocrystalline anisotropy
energy (Ku), which aligns the easy magnetization axis of a
ferromagnet along a preferred crystallographic axis, stabilize the
magnetization against thermal uctuations. The anisotropy deter-
mines how strongly a particle can hold onto its magnetic
information [8]. The existence of some degree of the spin canting
in the whole volume of the nanoparticles, the disordered surface/
dead layer and the spin-glass properties at the surface can explain
the decrease of the saturation magnetization [31]. The dead
magnetic layer originates from the demagnetization of the surface
spin, causing the surface spins to be disordered or misaligned. It
also leads to the super-exchange interaction between the FeO
Fe bonds, caused by the termination of the bonds. This phenom-
enon weakens the total magnetization of the nanoparticles [31].
(Eq. (9)) [30] can calculate the dead layer thickness (t) of spherical
nanoparticles. Where, D is the particle size, M
sat
is the satura-
tion magnetization and M
sat-0
is the saturation magnetization at
0 K. The saturated magnetizations are about 80.8 emu/g and
93.9 emu/g at room temperature and 5 K, respectively [30]. It is
clear from Table 7 the dead layers depend not only on the particle
Table 6
Tetrahedral molecular weights (M
1
), band frequency (u
1
) and force constant (k
t
).
Synthesis method M
1
(kg) 10
3
u
1
10
2
m
1
k
t
10
2
(N/m)
CNP 56.36 594 1.515
NNP 55.88 609 1.579
RNP 56.23 594 1.512
Fig. 4. MH curves of different prepared samples.
Table 7
Saturation magnetization, coercivity, remanence, maximum magnetic eld and
canted layer thickness measured at 27 1C.
Synthetic method M
S
(emu/g) H
C
(Oe) M
r
(emu/g) H
max
(kOe) t
shell
(nm)
CNP 58.4 286.0 12.45 18.0 0.74
NNP 12.6 23.7 0.17 18.4 0.80
RNP 29.4 25.2 0.84 17.6 0.87
I. Shari et al. / Journal of Magnetism and Magnetic Materials 324 (2012) 18541861 1858
size, but also on the saturation magnetization to the size ratio.
M
sat
M
sat0
16t=D 9
3.3.1.2. Cation distribution. The magnetic moments on the A and
B sites are antiparallel oriented in ferrites [32], whereas those
on the B sites are parallel to each other. In the present study, by
assuming 5 m
B
for Fe(III) and 3 m
B
for Co(II) ions [33], the net
magnetic moment m/molecule9M
A
M
B
9 is theoretically
calculated and the results are presented in Table 8.
It is obvious that the magnetization increases for the prepared
sample by the co-precipitation method as compared to the two
others. In addition, the larger particles in the co-precipitation
method can be attributed to the slightly higher concentration
difference of Fe
3
with the 5 m
B
magnetic moment between A
and B sites (Table 2) by assuming the collinear ferrimagnetic
spin structure. It should be mentioned that the magnetic moment
of 3 m
B
per unit chemical formula for partially inversed Co ferrite
is equals 71.5 emu/g [33]. Table 8 lists the saturation magnetiza-
tions for bulk conditions. The difference in saturation magnetiza-
tions which is observed in this table conrms the role of particle
sizes, as compared to the cation distribution.
3.3.1.3. Exchange interactions. Intrinsic magnetic properties
depend on the exchange interactions, and the exchange
interactions strongly depend on the inter-ionic length and bond
angles. In our study, because of the main effect of the particle size,
the discussion about the exchange length and exchange angle is
slightly difcult. The high value of saturation magnetization in
nanoparticles has been obtained from co-precipitation method
that can be explained by the general decrease in the bond angles
for each site (AA and BB) and the increase between the two
sites (AB) (Tables 4, 5)
3.3.1.4. Degree of crystallinity. It is known that saturation
magnetization gradually increases with the crystalline size and
coercivity and is dened by decreased domain walls displacement
as the crystalline size increases in the multi-domain range
[21,22]. The energy of a magnetic particle in the external eld is
proportional to its crystallinity (Table 1) and particle sizes via the
number of molecules in a single magnetic domain. Therefore, the
decrease of the M
s
value with the decrease of particle sizes and
crystallinity can be attributed to the surface effects that are the
result of nite-size scaling of nanocrystallites obtained by the
micro-emulsion method.
3.3.2. Magnetic coercivity (H
C
)
The magnetic coercivity of the nanoparticles depends signi-
cantly on their magneto-crystalline anisotropy, micro-strain,
inter-particle interaction, temperature, size and shape [34]. In
the CoFe
2
O
4
spinel, the Co(II) (3d
7
, 4F
9/2
, L3, S3/2, J 9/2)
cations have high spin ligand elds and possess seven d electrons
three of which are unpaired (Fig. 5). Evidently, the large magneto-
crystalline anisotropy of E
A
of CoFe
2
O
4
nanoparticles is due to the
strong LS couplings on the Co(II) cation sites. Energy barriers in
the individual nanoparticles [35] is considered and introduced by
the magnetocrystalline anisotropy energy of cobalt ferrite. This
effect results in the nanoparticles with multi-axial anisotropy,
and the zero coercivity can be achieved at very low particle sizes
(23 nm) but not at room temperature [36]. Obviously, by
reducing temperature, the anisotropy energy barriers in the
particles become sufciently high to prevent the reorientation
of the dipole moments by the stray eld. Upon cooling the
assembly of nanoparticles, the short-range order is preserved,
but with a signicant increase in the amount of disorder [35].
Table 6 lists the magnetic coercivity and magnetic remanence for
the nanoparticles which are obtained from three different synth-
esis methods with a uniform composition. It is clear that the
coercivity decreases from 286 Oe for co-precipitation to 23.7 Oe
for the normal micelle methods at room temperature. The
following table shows the obtained results:
3.3.2.1. Size effect. Below the critical single domain the particle
size of cobalt-ferrite (D
C
128 nm) for three samples the
coercivity decreases as a result of a reduction in the particle
volume, strain size and magnetic anisotropy barrier to block the
ips of the magnetic moments. Based on the following
discussions, the coercivity is at the highest level for the
particles obtained from the co-precipitation route as result of
reducing in the particle volume, grain size and magnetic
anisotropy. According to the StonerWohlfarth theory [37], by
decreasing the particle size in the single domain region, KV
decreases based on Eq. (10)
E
A
KV sin
2
y, 10
where E
A
is the anisotropic energy (shape or crystalline), K is the
magnetic anisotropy for a particle and V is the particle volume
and y is the angle between the magnetization direction and the
easy axis of the nanoparticle. The E
A
is reduced when the size of
the nanoparticles V and/or the anisotropy constant K (strength of
the LS couplings) is reduced. Furthermore, H
C
follows the D
6
power law (Eq. (11)) [34], where A is the exchange stiffness
constant, P
C
is the unity order constant, M
S
is the saturation
magnetization, K
1
is the magneto-crystalline constant and D is
Table 8
Theoretical magnetization per unit chemical formula 9M
A
M
B
9 and predicted
magnetization for bulk status were compared with reference [33] (M
S
-b) and
measured magnetization for nanosized synthesized powders (M
S
-n).
Synthesis
method
M
A
(m
B
)
M
B
(m
B
)
9M
A
M
B
9
(m
B
)
M
S
-b
(emu/g)
M
S
-n
(emu/g)
CNP 4.670 8.330 3.660 87.2 58.4
NNP 4.980 8.020 3.040 72.5 12.6
RNP 4.752 8.248 3.496 83.3 29.4
Fig. 5. Spinorbit (d
7
, Co
2
) stabilizations.
I. Shari et al. / Journal of Magnetism and Magnetic Materials 324 (2012) 18541861 1859
dened as the grain size.
H
C

P
C
K
4
1
D
6
m
0
M
S
A
3
11
3.3.2.2. Cation distribution. There are some reasons that show the
difference in coercivity for nanoparticles obtained by the normal
and reverse micelle methods. In addition, the particle size can be
explained by the cation distributions. While quantum couplings
may not be the same on A and B sites with tetrahedral and
octahedral symmetries, respectively, the contribution to the
magnetic anisotropy from the LS couplings in Fe(III) and/or
Co(II) cations may not be the same when the cation distribution
changes.
3.3.2.3. Some particles below the super-paramagnetic limit. The
presence of further super-paramagnetic particles (zero coercivity
and zero remanence at roomtemperature) in the smaller distributed
particles leads to a decrease in H
C
. The average size of all the
particles is smaller than the critical single domain size (128 nm),
(Table 2) [38] and larger than the critical super-paramagnetic size at
room temperature (23 nm) for cobalt ferrite [36]. However, the
probability of nding further particles below the super-
paramagnetic limit at room temperature for the fewer particles
obtained from micro-emulsion methods is higher, as compared to
the co-precipitation method.
4. Conclusions
In this paper, CoFe
2
O
4
powders were synthesized by three
widely used wet chemical methods, that is, co-precipitation,
normal micelle and reverse micelle methods with one starting
material/uniform chemical composition. The data obtained from
XRD, FTIR and VSM revealed that:
1. The crystallite size has a stronger effect on the magnetic
properties than the degree of crystallinity and the cation
distribution, respectively. Furthermore, the size effect covers
the role of slight difference of cation distribution.
2. The higher crystallinity in the coprecipitation sample might be
an additional contributing factor in the higher saturation
magnetization.
3. The nanocrystalline CoFe
2
O
4
shows the absorption bands
600 and 400 cm
1
, respectively. The high frequency band
u
1
around 600 cm
1
is attributed to the tetrahedral complexes
and the band u
2
400 cm
1
corresponds to the octahedral
complexes.
4. The magnetic properties of the sample strongly depend on the
size of the nanoparticles. In our study, the magnetic properties
of Co ferrite are slightly dependent on the distribution of the
cations between A and B sites and depends considerably
on the particle size.
Acknowledgments
The authors wish to thank the Department of Chemistry of Shiraz
University and the Department of Chemistry at Shiraz University of
Technology, especially Dr. R. Khalifeh for his kindly constructive
scientic advice and Dr M. N. Soltani-Rad for his support.
References
[1] U. Jeong, X. Teng, Y. Wang, H. Yang, Y. Xia, Superparamagnetic colloids:
controlled synthesis and niche applications, Advanced Materials 19 (2007)
3360.
[2] H. Shokrollahi, K. Janghorban, Inuence of additives on the magnetic proper-
ties, microstructure and densication of MnZn soft ferrites, Materials
Science and Engineering, B 141 (2007) 91107.
[3] H. Shokrollahi, Magnetic properties and densication of manganesezinc soft
ferrites (Mn
1x
Zn
x
Fe
2
O
4
) doped with low melting point oxides, Journal of
Magnetism and Magnetic Materials 320 (2008) 463474.
[4] G.A. El-Shobaky, A.M. Turky, N.Y. Mostafa, S.K. Mohamed, Effect of prepara-
tion conditions on physicochemical, surface and catalytic properties of cobalt
ferrite prepared by coprecipitation, Journal of Alloys and Compounds 493
(2010) 415422.
[5] M.A. Ahmed, A.A. EL-Khawlani, Enhancement of the crystal size and magnetic
properties of Mg-substituted Co ferrite, Journal of Magnetism and Magnetic
Materials 321 (2009) 19591963.
[6] S. Rana, J. Philip, B. Raj, Micelles based synthesis of cobalt ferrite nanopar-
ticles and its characterization using Fourier transform infrared transmission
spectrometry and thermogravimetry, Materials Chemistry and Physics 124
(2010) 264269.
[7] S. Ayyappan, S. Mahadevan, P. Chandramohan, M.P. Srinivasan, John Philip,
Baldev Raj, Inuence of Co
2
Ion concentration on the size, magnetic
properties, and purity of CoFe
2
O
4
spinel ferrite nanoparticles, Journal of
Physical Chemistry C 114 (2010) 63346341.
[8] S.B. Darling, S.D. Bader, A materials chemistry perspective on nanomagnet-
ism, Journal of Materials Chemistry 15 (2005) 4189.
[9] M.H. Youse, S. Manouchehri, A. Arab, M. Mozaffari, Gh.R. Amiri, J. Amighian,
Preparation of cobalt-zinc ferrite (Co
0.8
Zn
0.2
Fe
2
O
4
) nanopowder via combus-
tion method and investigation of its magnetic properties, Materials Research
Bulletin 45 (2010) 17921795.
[10] S.R. Dave, X. Gao, Monodisperse Magnetic Nanoparticles for Biodetection,
Imaging, and Drug Delivery: A Versatile and Evolving Technology, vol. 1, John
Wiley & Sons, Inc., 2009. 583609.
[11] R. Arulmurugan, B. Jeyadevan, G. Vaidyanathan, S. Sendhilnathan, Effect of zinc
substitution on CoZn and MnZn ferrite nanoparticles prepared by co-precipita-
tion, Journal of Magnetism and Magnetic Materials 288 (2005) 470477.
[12] Zh Zi, Y. Sun, X. Zhu, Zh. Yang, J. Dai, W. Song, Synthesis and magnetic
properties of CoFe2O4 ferrite nanoparticles, Journal of Magnetism and
Magnetic Materials. 321 (2009) 12511255.
[13] M.P. Pileni, Reverse micelles as microreactors, Journal of Physical Chemistry
97 (1993) 69616973.
[14] M. Mahmoudi, Sh. Sant, B. Wang, S. Laurent, T. Sen, Superparamagnetic iron
oxide nanoparticles (SPIONs): development, surface modication and appli-
cations in chemotherapy, Advanced Drug Delivery Review 63 (2011) 2446.
[15] M.A. Lopez-Quintela, C. Tojo, M.C. Blanco, L. Garca Rio, J.R. Leis, Microemul-
sion dynamics and reactions in microemulsions, Current Opinion in Colloid
and Interface Science 9 (2004) 264278.
[16] G. Vaidyanathan, S. Sendhilnathan, R. Arulmurugan, Structural and magnetic
properties of Co
1x
Zn
x
Fe
2
O
4
nanoparticles by co-precipitation method,
Journal of Magnetism and Magnetic Materials. 313 (2007) 293299.
[17] Nam T.S. Phan, Ha V. Le, Superparamagnetic nanoparticles-supported phos-
phine-free palladium catalyst for the Sonogashira coupling reaction, Journal
of Molecular Catalysis A: Chemical 334 (2011) 130138.
[18] H. Rietveld, A prole renement method for nuclear and magnetic structures,
Journal of Applied Crystallography 2 (1969) 6571.
[19] R.A. Young, The Rietveld Method, Oxford University Press, 1993.
[20] S. Kumar, V. Singh, S. Aggarwal, U.K. Mandal, R.K. Kotnala, Inuence of
processing methodology on magnetic behavior of multicomponent ferrite
nanocrystals, Journal of Physical Chemistry C 114 (2010) 62726280.
[21] L. Zhao, H. Yang, L. Yu, Y. Cui, X. Zhao, B. Zou, S. Feng, Structure and magnetic
properties of nanocrystalline CoLa
0.08
Fe
1.92
O
4
ferrite, Journal of Magnetism
and Magnetic Materials 301 (2006) 445451.
[22] N.M. Deraza Alaria, S. Shabanc, Structural, morphological and magnetic
properties of NiFe
2
O
4
nano-particles, Journal of Alloys and Compounds 486
(2009) 501506.
[23] S. Mazen, M.H. Abbdallah, B.A. Sabrah, H.A.M. Hasham, The effect of titanium
on some physical properties of CuFe
2
O
4
, Physica Status Solidi A 134 (1992)
263271.
[24] M.T. Sebastian, Dielectric Materials for Wireless Communication, 1st edn.,
Elsevier, UK, 2008.
[25] S.S. Bhatu, V.K. Lakhani, A.R. Tanna, N.H. Vasoya, J.U. Buch, P.U. Sharma,
U.N. Trivedi, H.H. Joshi, K.B. Modi, Effect of nickel substitution on structural,
infrared and elastic properties of lithium ferrite, Indian Journal of Pure and
Applied Physics 45 (2007) 596608.
[26] A. Goldman, Modern Ferrite Technology, 2nd edn, Pittsburgh, PA, USA,
Springer.
[27] R.D. Waldron, Infrared spectra of ferrite, Physical Review 99 (1955)
17271735.
[28] U.N. Trivedi, K.H. Jani, K.B. Modi, H.H. Joshi, Study of cation distribution in
lithium doped nickel ferrite, Journal of Materials Science Letters 19 (2000)
12711273.
[29] C.M. Srivastav, T.T. Srinivasan, Journal of Applied Physics 53 (1982)
81488150.
[30] L.D. Tung, V. Kolesnichenko, D. Caruntu, N.H. Chou, C.J. OConnor, L. Spinu,
Magnetic properties of ultrane cobalt ferrite particles, Journal of Applied
Physics 93 (2003) 74867488.
[31] C. Caizer, M. Stefanescu, Magnetic characterization of nanocrystalline NiZn
ferrite powder prepared by the glyoxylate precursor method, Journal of
Physics D: Applied Physics 35 (2002) 30353040.
I. Shari et al. / Journal of Magnetism and Magnetic Materials 324 (2012) 18541861 1860
[32] J. Samuel Smart, The Ne el theory of ferrimagnetism, American Journal of
Physics 23 (1955) 356370.
[33] K. Hanedas, A.H. Morrish, Noncollinear magnetic structure of CoFe
2
O
4
small
particles, Journal of Applied Physics 63 (1988) 42584260.
[34] Q. Zeng, I. Baker, V. McCreary, Zh. Yan, Soft ferromagnetism in nanostruc-
tured mechanical alloying FeCo-based powders, Journal of Magnetism and
Magnetic Materials 318 (2007) 2838.
[35] M. Georgescu, J. Viota, M. Klokkenburg, B. Erne , D. Vanmaekelbergh,
P. Zeijlmans van Emmichoven, Short-range magnetic order in two-dimen-
sional cobalt-ferrite nanoparticle assemblies, Physical Review B 77 (2008).
[36] Ch. Liu, B. Zou, A.J. Rondinone, Z.J. Zhang, Chemical control of superpara-
magnetic properties of magnesium and cobalt spinel ferrite nanoparticles
through atomic level magnetic couplings, Journal of the American Chemical
Society 122 (2000) 62636267.
[37] E.C. Stoner, E.P. Wohlfarth, A mechanism of magnetic hysteresis in hetero-
geneous alloys, The Philosophical Transactions of the Royal Society A 240
(1948) 599642.
[38] O. Philippova, A. Barabanova, V. Molchanov, A. Khokhlov, Magnetic polymer
beads: recent trends and developments in synthetic design and applications,
European Polymer Journal 47 (2011) 542559.
I. Shari et al. / Journal of Magnetism and Magnetic Materials 324 (2012) 18541861 1861