Nanophase formation of strontium hexaferrite fine powder by the sonochemical method using Fe(CO)5

Article abstract of DOI:10.1016/S0304-8853(03)00479-7

A fine strontium hexaferrite powder has been successfully synthesized by a sonochemical method employing Fe(CO)₅ and SrCO₃. SrCO ₃ was first synthesized using strontium nitrate and urea, and it was found that applying ultrasound radiation during this process results in the uniform formation of SrCO₃ hexagonal rods. These rods were then dispersed with in situ generated amorphous Fe₂O₃, using Fe(CO)₅ as the source. The resultant precursor was then calcined at 900°C, which is lower than the conventional solid-state reaction of applying 1300°C to get the strontium hexaferrite fine powder. This sonochemically derived ferrite exhibited an intrinsic coercivity of ～4600 Oe and a saturation magnetization of ～60 emu/g at 20 K and ～32 emu/g at 300 K.

Full text of DOI:10.1016/S0304-8853(03)00479-7

ARTICLE IN PRESS
Journal of Magnetism and Magnetic Materials 268 (2004) 95–104
Nanophase formation of strontium hexaferrite fine powder
by the sonochemical method using Fe(CO)5
a
M. Sivakumar , A. Gedanken *, W. Zhong , Y.W. Du , D. Bhattacharya ,
a,
b
b
c
c
Y. Yeshurun , I. Felner
d
a
Department of Chemistry, Bar-Ilan University, Ramat-Gan 52900, Israel
National Laboratory of Solid State Microstructures, Nanjing University, Nanjing 210093, People’s Republic of China
b
c
Department of Physics, Institute of Superconductivity, Bar-Ilan University, Ramat-Gan, Israel, 52900
d
The Racah Institute of Physics, The Hebrew University, Jerusalem 91904, Israel
Received 9 March 2003; received in revised form 7 May 2003
Abstract
A fine strontium hexaferrite powder has been successfully synthesized by a sonochemical method employing Fe(CO)
and SrCO . SrCO was first synthesized using strontium nitrate and urea, and it was found that applying ultrasound
radiation during this process results in the uniform formation of SrCO hexagonal rods. These rods were then dispersed
5
3
3
3
with in situ generated amorphous Fe
ꢀ
2
O
00 C, which is lower than the conventional solid-state reaction of applying 1300 C to get the strontium hexaferrite fine
3
, using Fe(CO)
5
as the source. The resultant precursor was then calcined at
ꢀ
9
powder. This sonochemically derived ferrite exhibited an intrinsic coercivity of B4600 Oe and a saturation
magnetization of B60 emu/g at 20 K and B32 emu/g at 300 K.
r 2003 Elsevier B.V. All rights reserved.
PACS: 75.50.Tt; 81.07.Bc; 75.75.+a
Keywords: Sonochemistry; Cavitation; Magnetic; Nanomaterial; Hexaferrite
1
. Introduction
crystallographic sites are coupled antiferromagne-
tically. This ferrimagnet exhibits a high saturation
magnetization and high coercivity because of the
relatively high magnetocrystalline anisotrophy
field [1]. This ferrite has been explored for many
technologically demanding and challenging appli-
cations, such as for permanent magnets and high-
density perpendicular recording media, with prop-
er doping [2]. The properties of this ferrite are
largely dependent on the processing routes used
for its fabrication [1,3]. Traditionally, this ferrite
powder is synthesized by a mixed oxide ceramic
The hexagonal M-type hard hexaferrites have
attracted much attention as the most widely used
permanent magnets, which account for about
90 wt% of the annual production of permanent
magnets. Strontium hexaferrite, SrFe O , is a
1
2
19
ferrimagnet in which iron ions in five different
*Corresponding author. Tel.: +97235318315; fax:
+97235351250.
E-mail address: gedanken@mail.biu.ac.il (A. Gedanken).
0304-8853/03/$ - see front matter r 2003 Elsevier B.V. All rights reserved.
doi:10.1016/S0304-8853(03)00479-7

ARTICLE IN PRESS
96
M. Sivakumar et al. / Journal of Magnetism and Magnetic Materials 268 (2004) 95–104
method, which involves the solid-state reaction
between SrCO and Fe O at a high calcination
and implosive collapse of the cavities. This
produces extreme conditions of temperature
3
2
3
ꢀ
ꢀ
temperature (B1300 C) [4]. However, this techni-
que has the disadvantage of resulting in larger
particle size and agglomerates and uncontrolled
particle morphology. Also, the subsequent milling
of the calcined ferrite powder undoubtedly intro-
duces a level of contamination and, therefore,
degrades the magnetic properties [5].
(B5000 C), pressure (B20 Mpa), and cooling
1
0
rates (>10 K/s), which lead to many unusual
and unique physical and chemical effects. Re-
cently, the cavitational collapse effects have been
effectively used to prepare nanoscale metals, metal
oxides, and nanocomposites [24,25].
In the current research, we report on a new
method for the synthesis of single phase, ultrafine,
SrFe O particles. The method makes use of
Thus, the preparation of strontium ferrite
powder with refined particle size, narrowed parti-
cle size distribution, and minimal particle agglom-
eration has received considerable attention [6]. In
order to improve the material properties, numer-
ous nonconventional soft synthetic routes have
been devised, including co-precipitation [7–10],
hydrolysis of metalorganic complexes [11], sol–gel
synthesis [12–14], spray pyrolysis [15], the citrates
method [2,12,14], glass crystallization [16–19] and
hydrothermal reaction [20–22]. Ultrafine stron-
tium ferrite powder with particle sizes in the range
of B50 nm have been prepared successfully via
some of these processing routes, but the magnetic
properties of the resulting ferrite powders vary
considerably from one material to another. For
example, Choy et al. [12] reported a coercive force
of 6500 Oe for one of the strontium ferrite powders
prepared via a sol–gel processing route, although
the powder exhibits a relatively low magnetization
1
2
19
ultrasound irradiation in two stages: (i) to
synthesize uniform rods of SrCO , (ii) decompos-
3
ing Fe(CO)5 and dispersing the in situ formed
amorphous Fe O on the prepared SrCO3 ob-
2
3
tained in the first stage to get the precursor.
Calcinating this precursor then results in the
formation of a fine hexaferrite powder.
2. Experimental section
2.1. Starting materials
The starting materials used in the present work
included strontium nitrate hexahydrate (Aldrich,
99%), urea (Aldrich, 98%), Fe(CO)5 (Strem
chemicals, USA, 99.5%), decalin (Acros organics,
98%) and pentane (96%, Bio-lab, Israel).
(
50 emu/g).
One of the major difficulties encountered in
2.2. Preparation procedures
many of these wet-chemistry-based processing
routes for strontium ferrites, such as in co-
precipitation [8,23] and in hydrothermal reaction,
2.2.1. Stage 1: preparation of SrCO3
The synthesis of strontium carbonate particles
was carried out with the aid of ultrasound
radiation. About 5 g of urea was added to an
aqueous solution (50 ml) of 1.0 g of strontium
nitrate hexahydrate in a sonication flask. The
solution was then irradiated for 3 h with high
intensity ultrasound radiation by employing a
direct immersion titanium horn (Sonics and
Materials, 20 kHz, 600 W). Calorimetry was used
to estimate the electroacoustic or energy transfer
efficiency of the transducer to the solution. Based
on the calorimetry experiments, the transfer
efficiency of the ultrasonic energy from the
transducer to the reactor solution was estimated
as 54% (at 65% amplitude), with the assumption
is the relatively low solubility of Sr(OH) in an
2
aqueous solution. This low solubility poses pro-
blems in maintaining the stoichiometry of the
strontium ferrite, although excess strontium can be
introduced into the starting composition. For
+
2
+3
example, a Sr :Fe
ratio of 1:11 was used by
Kulkarni et al. [8] in their co-precipitation route, in
comparison with the stoichiometry ratio of 1:12 in
SrFe O .
1
2
19
Application of ultrasound for the synthesis of
newer materials has become an attractive area in
recent times. Cavitation is the phenomenon that
underlines the effects observed during the passage
of ultrasound. This includes the formation, growth

ARTICLE IN PRESS
M. Sivakumar et al. / Journal of Magnetism and Magnetic Materials 268 (2004) 95–104
97
that all ultrasonic energy was converted to heat in
the reactor. Based on these calorimetry measure-
ments, the power intensity of the system was
on a copper grid coated with carbon (400 mesh,
electron microscopy sciences) and allowing them
to dry in air. FT-IR spectra were recorded on a
Nicolet (Impact 410) infrared spectrophotometer
with KBr pellets over the range of
2
calculated as 29.7 W/cm . The titanium horn tip
was inserted into the solution to a depth of 1 cm.
The temperature during the experiment in-
ꢁ
1
40073000 cm . The thermogravimetric analysis
was performed in an oxygen atmosphere using a
Mettler Toledo TGA/SDTA851 instrument at-
tached to a mass spectrometer (Balzers Instru-
ꢀ
creased to a maximum of 85 C due to ultrasound
passage, as measured by an iron Constantan
thermocouple. After the precipitation process
was complete, the precipitates were separated from
the solution by centrifugation. The recovered
precipitate was then washed several times with
doubly distilled water and dried under vacuum at
room temperature.
ꢀ
ments), over the temperature range of 30–1000 C
ꢀ
(heating rate B10 C/min). Complexpermeability
spectra were measured with a coaxial transmission
line connected to an impedance analyzer (Agilen-
t4284A from 20 Hz to 1 MHz), or an impedance/
material analyzer (Agilent4191B from 1 MHz to
1.8 GHz). The magnetic measurements were per-
formed over a wide temperature range of 20–
900 K. The magnetization vs. temperature, as well
as hysteresis loops at different temperatures,
within the temperature range 20–300 K, were
measured in a SQUID magnetometer coupled
2
obtained from Fe(CO) on SrCO
.2.2. Stage 2: dispersion of amorphous Fe O3
2
5
3
This step utilizes the principle of obtaining pure
amorphous materials from suitable precursors by
means of sonication [24,26]. The preparation of
amorphous Fe O was reported elsewhere [26].
2
3
Following the same principle, the present process
involves taking stoichiometric amounts of SrCO3
and Fe(CO)5 in decalin and irradiating the
solution with ultrasound (the method followed is
similar to stage 1). The reaction was carried out in
with a superconducting magnet (MPMS-5S,
Quantum Design). The high-temperature magnetic
properties were measured in a vibrating sample
magnetometer coupled with a furnace attachment
(Oxford).
ꢀ
an air atmosphere at 0 C for 4 h. After 4 h, the
obtained product was thoroughly washed with
pentane, centrifuged, and dried in vacuum at room
temperature. The product thus obtained will
hereafter be referred to as the precursor. This
precursor was calcined at 900 C for 14 h in air
atmosphere to obtain strontium hexaferrite fine
powder.
3. Results and discussion
3.1. XRD
ꢀ
The SrCO sample synthesized by ultrasound
3
irradiation in stage 1 shows a crystalline X-ray
pattern (Fig. 1a), which matches that reported for
standard SrCO . Fig. 1c illustrates the XRD
2
.3. Characterization
3
pattern of the precursor product obtained by
sonicating the decalin solution of Fe(CO)5 and
SrCO . The XRD pattern of this precursor is very
The phase constitution of the hexaferrite pow-
der was recorded by employing a Rigaku X-ray
diffractometer (Model-2028, Co Ka). XRD mea-
surements were taken in the scanning range: 2y
3
similar to the pattern of Fig. 1a except that it has
an amorphous background in addition to the
SrCO peaks. This is due to the application of
ꢀ
ꢀ
from 15 to 70 . The microstructures of the
products were determined by transmission electron
microscopy (TEM) (JEOL-JEM 100SX micro-
scope). The powder morphology was also
examined by a scanning electron microscope
3
ultrasound on the Fe(CO) , which generates
5
amorphous Fe O . The amorphous Fe O was
3
2
3
2
then dispersed or coated on SrCO again with the
3
assistance of ultrasound. In order to confirm this,
a solution of Fe(CO) in decalin alone was
(JEOL-JSM-840). Samples for TEM were pre-
pared by placing a drop of the sample suspension
5
ꢀ
sonicated at 0 C for 4 h. Fig. 1b shows the XRD

ARTICLE IN PRESS
98
M. Sivakumar et al. / Journal of Magnetism and Magnetic Materials 268 (2004) 95–104
1800
1500
1200
phase of SrFe O . Thus, the formation tempera-
19
1
2
(a)
SrCO3
ture of SrFe O by this process compares
19
1
2
9
00
00
00
0
reasonably well with many other chemistry-based
processing routes.
6
3
1
000
3
.2. TGA
8
00
00
00
00
0
(b)
(c)
amorphous Fe O3
2
6
4
2
TGA results for the as-synthesized strontium
carbonate in stage 1 are shown in Fig. 2a. The only
ꢀ
weight loss that can be observed is between 900 C
ꢀ
and 1000 C. This weight loss may be ascribed to
6
4
2
00
00
00
amorphous Fe O on Sr CO3
2
3
the carbonate decomposition. Fig. 2b depicts the
TGA curve for the as-synthesized precursor in
stage 2. The TGA shows three distinct weight loss
steps. All these three steps are confined to the
weight loss stages of amorphous Fe O as from
0
00
00
00
00
00
00
00
00
0
2
3,
8
7
6
5
4
3
2
1
(d)
SrFe12O19
Fig. 2a it can be seen that SrCO gives a weight
3
ꢀ
ꢀ
loss only in between 900 C and 1000 C. It can also
be noted from Fig. 2b that due to the deposition of
amorphous Fe O on SrCO , the temperature of
2
3
3
the final weight loss for the precursor completes
ꢀ
2
0
30
40
50
60
70
around 900 C. For comparison, Fig. 2c shows the
Cobalt-2-theta
TGA curve for the sonochemical as-synthesized
amorphous Fe O obtained by the decomposition
of Fe(CO)5.
Fig. 1. Powder X-ray diffraction patterns of the (a) as-
synthesized SrCO , (b) amorphous Fe (for comparison),
c) amorphous Fe dispersed on SrCO , and (d) SrFe12
obtained by calcining the precursor at 900 C in air for 14 h.
2
3
3
2 3
O
(
2
O
3
3
ꢀ
O
19
This curve matches that of the precursor
Fig. 2b) for amorphous Fe O weight loss stages.
Also, it has already been found that by heating the
(
2
3
ꢀ
pattern of the resultant product obtained during
this process. It can be seen from this pattern that
the Fe O obtained in this process is amorphous.
amorphous Fe O around 300 C, it loses its
2
3
amorphous character and converts into crystalline
material [26].
2
3
Fig. 1d shows the XRD pattern of SrFe O
19
1
2
obtained by calcining the precursor sample at
ꢀ
9
00 C in air atmosphere for 14 h. The diffraction
1
00
(a) SrCO3
peaks match those reported for standard
SrFe O . Heating the precursor at a rate of
8
ferrite phase. This was indicated by the appearance
of the most intense reflections (1 1 4), (1 0 7) and
1
2
19
95
ꢀ
ꢀ
C/min to 900 C resulted in a well-crystallized
(b)
9
0
amorphous Fe O3 on SrCO3
2
(c)
amorphous Fe O3
2
85
80
75
ꢀ
ꢀ
(
1 1 0) peaks at 2y angles of 40.86 , 38.63 and
ꢀ
3
formation of SrFe O is completed at 900 C.
6.28 respectively, as shown in Fig. 1d. Thus, the
ꢀ
12
19
Also, the formation temperature for SrFe O by
1
2
19
7
0
the sonochemically prepared precursor is appar-
ently lower than that observed in the conventional
solid-state reaction utilizing SrCO and Fe O
3
200
400
600
800
1000
o
Temperature ( C)
3
2
Fig. 2. TGA results of the as-synthesized SrCO
Fe
dispersed on SrCO
3
, amorphous
ꢀ
(
B1300 C) and in the precursors obtained via
2
3 2 3
O (for comparison) and precursor (amorphous Fe O
ꢀ
spray pyrolysis (B1200 C) for forming a single
3
)

ARTICLE IN PRESS
M. Sivakumar et al. / Journal of Magnetism and Magnetic Materials 268 (2004) 95–104
99
3
.3. M o. ssbauer investigations
damping mechanism of domain wall displace-
ments, can be checked. In general, the dispersions
caused by wall displacements occur in radio
frequency range (below 10 MHz), and those
caused by domain rotations often appear in the
microwave range (100 MHz to 10 GHz). In parti-
cular, this methodology has been extensively used
in ferrites [28,29]. Thus, a similar study has been
carried out in the SrFe O powder prepared by
As a confirmation of the phase purity, Fig. 3
5
shows the Fe M o. ssbauer spectrum at 300 K for
7
ꢀ
the SrFe O powder calcined at 900 C for 14 h in
1
2
19
air atmosphere. Dots in this figure represent the
experimental data points, while the solid lines are
the least-squares fitting for these data. Generally
speaking, this complicated spectrum is composed
of five sub-spectra corresponding to the five
inequivalent crystallographic sites of Fe in this
hexagonal ferrite material. Unfortunately, there is
considerable controversy regarding the five hyper-
fine fields in this material. We adopted the
interpretation and the values suggested by Evans
et al. [27]. The isomer shifts and the hyperfine
fields obtained are in excellent agreement with the
data already obtained [27] and for the sake of
brevity they have not been listed and discussed
here. Thus, this spectrum fits reasonably well with
the spectrum of the standard SrFe O phase and
1
2
19
the present ultrasonic method and calcined at
ꢀ
900 C for 14 h in air atmosphere.
The frequency dependence of the real part m
0
0
0
and imaginary part m of complexpermeability
spectra is shown in Fig. 4. For permeability
measurements, the nanosized SrFe O powder
1
2
19
sample was pressed into a ring with a typical size
of 13 mm OD, 7 mm ID and 1 mm thickness. The
pressed ring has a lower density than the sintered
sample. According to the theory contemplation, in
single-domain grains, there are no domain wall
motions. Therefore, the m (static permeability, i.e.
1
2
19
0
0
clearly indicates that the material measured is
indeed SrFe O .
m with frequency limit to zero) is only caused by
reversible domain rotations and follows the rela-
1
2
19
tion m ¼ 1 þ 8pMs=3Ha; where Ha is the effective
0
magnetic anisotropy field. Because the Ha of
SrFe O is very large (8100 Oe), the complex
3
.4. Permeability spectra
1
2
19
permeability is small. From Fig. 4, one can see that
0
A useful route to investigate the mechanism of
the m is about 1.2, smaller than the theory pro-
gnosis. We think that the nonmagnetic impurities
domain wall motions and domain rotations is to
measure the complexpermeability as a function of
frequency, i.e., the so-called magnetic spectrum. In
addition, by studying such a magnetic spectrum,
the effective magnetic anisotropy field, as well as
2.0
1.5
1.0
0.5
0.0
SrFe O
19
1
2
µ′
µ′′
6
7
8
9
10
10
10
Frequency (Hz)
Fig. 4. The plots of complexpermeability as a function of
10
Fig. 3. M o. ssbauer spectra of the SrFe12
ꢀ
O
19 obtained by
calcining the precursor at 900 C in air atmosphere for 14 h.
(ꢂ) indicates the experimental points and the (—) is the
theoretical fitting.
frequency for the polycrystalline SrFe12O19 sample at room
temperature.

ARTICLE IN PRESS
100
M. Sivakumar et al. / Journal of Magnetism and Magnetic Materials 268 (2004) 95–104
such as boundaries, and cavities resulting from the
low-density structure, may cause the permeability
to decline further. In the corresponding bulk
sample, there are domain wall motions, whereas
in nanosized sample, there are only reversible
phous Fe O synthesized by ultrasound alone (for
2 3
comparison) shows absorption bands at 473, 1424
ꢁ
1
and 1550 cm in Fig. 5b.
Well-established absorption bands in Fig. 5d at
ꢁ
1
443, 552 and 600 cm in the powder calcined at
ꢀ
domain rotations. The m contributed by domain
900 C indicate the formation of strontium ferrite.
Meanwhile, the characteristic bands of SrCO and
0
rotations is generally smaller than that by domain
wall displacements.
3
iron oxides vanish at this calcination temperature.
The disappearance of the carbonate peaks pro-
vides unequivocal proof for the lowering of the
3
.5. IR spectra
ꢀ
ꢀ
sintering temperature from 1200 C to 900 C. The
above interpretation of the IR spectral results is
based on the assumption that the synthesized
The IR spectra of Figs. 5a and c provide
evidence for the presence of carbonate ions in
SrCO3 and in the precursor, respectively. The
carbonate, Fe O , precursor and hexaferrite are
2 3
essentially pure materials. These results agree with
the XRD phase-analysis findings.
ꢁ
1
strong absorption band at 1470 cm in the case of
ꢁ
1
as-synthesized SrCO and 1469 cm in the case
3
of precursor had been attributed to the n3 mode of
2
ꢁ
the CO ion of SrCO3.
3.6. TEM
3
ꢁ
1
The other bands around 850 and 700 cm are
assigned to the n2 and n4 modes of the carbonate
ion, respectively. The IR spectrum of the amor-
Fig. 6a shows the results of TEM investigation
of the synthesized strontium carbonate using
ultrasound. It can be clearly seen from this picture
that ultrasound promotes the homogeneous for-
mation of rods and their average diameter and
length are 1.2 and 8 mm, respectively. Fig. 6b
depicts the TEM picture of the amorphous Fe O
0
0
0
0
0
0
0
0
.35
.30
.25
.20
.15
.10
.05
.00
SrCO3 (a)
2
3
powder dispersed on the hexagonal SrCO rods. It
3
-0.05
is clearly seen that there is no crystalline formation
of Fe O in this powder. It can also be observed
2 3
0.6
0.4
0.2
0.0
amorphous Fe O
(b)
2
3
that this amorphous Fe O is an agglomerate of
2
3
small particles aggregated in a spongelike form.
Fig. 6c micrograph indicates the morphology of
the ferrite sample typical of the M-ferrites
obtained after treating it at 900 C. It can be
observed from this micrograph that after calcina-
tion, the ferrite particles have an average diameter
0
0
0
0
0
0
0
0
0
.8
.7
.6
.5
.4
.3
.2
.1
.0
(c)
ꢀ
amorphous Fe O on SrCO
2
3
3
of about 50 nm. No remaining SrCO and Fe O
3
3
2
-
0.1
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
could be observed in the micrograph in spite of the
low heat treatment temperature. This result agrees
with the results of XRD and IR.
SrFe12O19
(d)
3.7. Measurement of the magnetic properties
-
0.1
3000
2000
1000
-
1
The magnetization vs. temperature pattern over
the entire temperature range is shown in Fig. 7.
From this plot, one extracts the Curie point to be
B800 K. Hysteresis loops at 20 and 300 K are
shown in the inset of Fig. 7. From the hysteresis
Wavenumbers (cm )
Fig. 5. IR spectra of the (a) as-synthesized SrCO
3
, (b)
amorphous Fe (for comparison), (c) amorphous Fe
2
O
3
2 3
O
dispersed on SrCO
3
, and (d) SrFe12
O
19 obtained by calcining
ꢀ
the precursor at 900 C in air for 14 h.

ARTICLE IN PRESS
M. Sivakumar et al. / Journal of Magnetism and Magnetic Materials 268 (2004) 95–104
101
(a)
(b)
(c)
Fig. 6. TEM of (a) as-synthesized SrCO
ꢀ
3
, (b) amorphous Fe
O
2 3
dispersed on SrCO
3
, and (c) SrFe12O19, obtained by calcinating the
precursor b at 900 C in air for 14 h.
loop, we extract the coercive field (H ) and the
c
temperature pattern exhibits a strong temperature
dependence over a range of 300–800 K. This could
be because of the presence of single-domain
nanoparticles and consequent superparamagnet-
ism over this temperature range.
saturation magnetization (M ). The H and M
s
s
c
values are found to be B4600 Oe and B60 emu/g,
respectively, at B20 K. While the Ms value drops
down with temperature and, at room temperature
(
remains more or less temperature independent
300 K), it becomes B32 emu/g, the Hc value
At a lower temperature, however, magnetization
tends towards saturation, possibly because of the
fact that the blocking temperature (TB) is crossed,
which helps in restoring the bulk magnetization
behavior. The TB could be around 300 K. The
over the 20–300 K range. Other pertinent data are
the remanence magnetization (M ) and the loop
r
squareness ratio (M =M ); they are found to be
r
s
B45 emu/g and B0.8, respectively, at 20 K and
B19 emu/g and B0.6, respectively, at 300 K. It is
interesting to note that the magnetization vs.
Curie point (T ) extracted from the high-tempera-
c
ture magnetization vs. temperature pattern
(B800 K) appears to be higher than the bulk

ARTICLE IN PRESS
102
M. Sivakumar et al. / Journal of Magnetism and Magnetic Materials 268 (2004) 95–104
60
50
40
30
20
10
0
Applied Field ~ 16000 Oe
~ 800 K
T
c
2
0 K
300 K
Tc
Field (Oe)
0
100 200 300 400 500 600 700 800 900
Temperature (K)
Fig. 7. The magnetization vs. temperature pattern over the temperature ranges 20–900 K for the ultrafine SrFe12
ꢀ
O
19 system prepared
by a sonochemical technique followed by calcination at 900 C for 14 h. Hysteresis loops measured at 20 and 300 K are shown in the
inset.
Curie point B750 K. This result corroborates the
already obtained observation [30], where a rise in
Tc as a function of drop in particle size is reported.
The magnetization (M) vs. field (H) curve,
especially at higher temperature, is not found to
be pure Langevin-type due to the distribution of
the randomly oriented anisotropy axes.
It is known that magnetic properties such as
saturation magnetization and the magnetic hyper-
fine field value of nanoparticles are much smaller
than those of the corresponding bulk materials.
The energy of a magnetic particle in an external
field is proportional to its size or volume via the
number of magnetic molecules in a single magnetic
domain. When this energy becomes comparable to
thermal energy, thermal fluctuations will signifi-
cantly reduce the total magnetic moment at a given
field [31]. Another reason for the reduced magne-
tization could be the large surface-to-volume ratio
for small particles. The magnetic molecules on the
surface lack complete co-ordination, and the spins
are also disordered. As a result of this, we observe

ARTICLE IN PRESS
M. Sivakumar et al. / Journal of Magnetism and Magnetic Materials 268 (2004) 95–104
103
a drop in saturation magnetization value compar-
able to the single crystal data (Ms ¼ 68 emu/gm).
This result, therefore, provides evidence of the
presence of ultrafine particles in the sample.
However, the coercive field value (B4600 Oe) too
exhibits a drop, in comparison with the theoretical
value predicted for a single crystal (B7500 Oe) [7].
Whereas strontium ferrite powders from an
ethanol-based microemulsion show a maximum
coercivity value of 6195 Oe [32], single-domain
strontium hexaferrite particles obtained by a
modified co-precipitation method also show a
high coercivity value of 5550 Oe [33]. Thus, the
low value of the coercive field obtained in the
present case can be due to the low crystalline
anisotropy, which arises from crystal imperfection
and a high degree of aggregation.
SrFe O
1
hexaferrite at a lower calcination
ꢀ
2
19
temperature of 900 C, employed in the present
method.
The amorphous nature of Fe O synthesized by
ultrasound might also play a key role in increasing
its reactivity with SrCO , resulting in the forma-
2
3
3
tion of a hexaferrite phase at the lower calcination
temperature. The importance of amorphicity in
increasing the reactivity has already been discussed
[34]. Suslick et al. [34] have clearly shown that
sonochemically prepared amorphous Fe, Co and
Fe–CO nanomaterials are better catalysts than the
corresponding crystalline nanomaterials prepared
by other methods. In addition, Yee et al. [35] have
functionalized amorphous Fe O nanoparticles
with alkanesulfonic and octadecanephosphonic
acids. This work further supports the fact that it
is possible to form a coating of organic molecules
more easily on amorphous Fe O . In addition to
2
3
2
3
4
. Results and discussion
the above factors, the size of the reactants (in
specific, nanocrystalline amorphous Fe O ) also
2
3
The role of urea in the homogeneous precipita-
contributes to the lowering of calcination tem-
perature. It can be seen from the work of Ataie
and Heshmati-Manesh that by using conventional
strontium and iron nitrates, the mean particle size
of the ferrite obtained is about 225 nm, at a
tion of strontium carbonate particles (as synthe-
sized in stage 1) is that it decomposes by hydrolysis
homogeneously in the solution upon heating,
ꢁ
+
producing CO , OH and NH4 ions in the
2
ꢀ
solution. This increases the pH of the medium,
which is suitable for the formation of strontium
carbonate particles. Measurement of the pH
before and after the sonication experiment re-
vealed that the pH of the solution increased from
calcination temperature of 900 C [33]. In the
present case, however, by applying the same
ꢀ
calcination temperature of 900 C, the mean
particle size of the obtained ferrite is 50 nm, which
might be due to amorphous Fe O already present
in the reacting system.
2
3
5
.4 to 6.2. Applying ultrasound radiation during
the reaction might accelerate the homogeneous
precipitation process. In addition to this, it might
also be beneficial in controlling the particle size
and the shape.
5. Conclusions
In the conventional solid-state reaction of Fe O
2
To summarize, we have demonstrated a novel
sonochemical method for the synthesis of well-
crystallized monophasic SrFe O fine particles
3
with SrCO3 to form the hexaferrite, the major
limitation is the diffusion factor. In the present
sonochemical method, this diffusion resistance
may be overcome as there is an in situ generation
of amorphous Fe O from Fe(CO) by ultrasound,
1
2
19
by the decomposition of Fe(CO) , followed by its
5
reaction with SrCO , the synthesis of which was
3
again assisted by ultrasound. More importantly,
by this sonochemical method, formation of the
SrFe O single phase occurs at a temperature as
2
3
5
followed by its dispersion or deposition on SrCO3,
assisted again by ultrasound. This process might
facilitate the diffusion, as well as the reaction, of
1
2
19
ꢀ
low as 900 C. It should also be noted that similar
results, as reported here for the SrFe O synth-
esis, should be expected for BaFe O due to their
+
2
highly reactive strontium ion (Sr ) with the
amorphous Fe O , resulting in the completion of
the phase transition of the precursor to the
1
2
19
2
3
12 19
similar chemical behavior.

ARTICLE IN PRESS
104
M. Sivakumar et al. / Journal of Magnetism and Magnetic Materials 268 (2004) 95–104
Acknowledgements
[15] Y. Senzaki, J. Caruso, M.J. Hampden-Smith, T.T. Kodas,
L.-M. Wang, J. Am. Ceram. Soc. 78 (1995) 2973.
16] H. Sato, T. Umeda, Mater. Trans. JIM. 34 (1993) 76.
17] D. Bahadur, D. Chakravorty, J. Mater. Sci. 32 (1997) 189.
18] K. Oda, T. Yoshio, K.O. Oka, F. Kanamaru, J. Mater.
Sci. Lett. 3 (1984) 1007.
[
[
[
We thank The Ministry of Science, Sport and
Culture for a Materials Science grant through
the Sino-Israeli program in Materials Science. We
also thank Ms. Louise Braverman for editorial
assistance.
[19] O. Kubo, T. Ido, H. Yokoyama, IEEE Trans. Magn. 18
1982) 1122.
20] J.H. Lee, H.S. Kim, C.W. Won, J. Mater. Sci. Lett. 15
1996) 295.
21] A. Ataie, I.R. Harris, C.B. Ponton, J. Mater. Sci. 30 (1995)
429.
(
[
[
[
(
References
1
22] C.H. Lin, Z.W. Shih, T.S. Chin, M.L. Wang, Y.C. Yu,
IEEE Trans. Magn. 26 (1990) 15.
[
1] H. Kojima, in: E.P. Wohlfarth (Ed.), Ferromagnetic
Materials: A Handbook on the Properties of Magnetically
Ordered Substances, Vol. 3, North-Holland, Amsterdam,
[23] A. Ataie, I.R. Harris, C.B. Ponton, in: J. David (Ed.),
Introduction to Magnetism and Magnetic Materials,
Chapman & Hall, London, 1991.
1982.
[
[
2] P. Hernandez, C. de Francisco, J.M. Munoz, J. Iniguez, L.
Torres, M. Zazo, J. Magn. Magn. Mater. 158 (1996) 123.
3] H. Stablein, in: E.P. Wohlfarth (Ed.), Ferromagnetic
Materials: A Handbook on the Properties of Magnetically
Ordered Substances, Vol. 3, North-Holland, Amsterdam,
[24] K.S. Suslick, S.B. Choe, A.A. Cichowlas, M.W. Grinstaff,
Nature 353 (1991) 414.
[25] K.S. Suslick, M. Fang, T. Hyeon, A.A. Cichowlas, in: K.E.
Gonslaves, G.M. Chow, T.O. Xiao, R.C. Cammarata
(Eds.), Molecularly Designed Ultrafine Nanostructured
Materials, Materials Research Society, Pittsburgh, PA,
1994.
1
982.
4] F. Haberey, A. Kockel, IEEE Trans. Magn. 12 (1976)
83.
[
[
[
[
[
[
9
[26] X. Cao, Yu. Koltypin, R. Prozorov, G. Kataby,
A. Gedanken, J. Mater. Chem. 12 (1997) 2447.
[27] B.J. Evans, F. Grandjen, A.P. Lilot, R.H. Vogel,
A. Gerard, J. Magn. Magn. Mater. 67 (1987) 123.
[28] G.T. Rado, R.W. Wright, W.H. Emerson, Phys. Rev. 80
(1950) 273.
5] S. Blackburn, T.P. Johnson, C.B. Ponton, M.H.L. Wise,
Trans. Inst. Chem. Eng. 73 (Part A) (1995) 71.
6] Z. Jin, W. Tang, J. Zhang, H. Lin, Y. Du, J. Magn. Magn.
Mater. 182 (1993) 231.
7] V.V. Pankov, M. Pernet, P. Germi, P. Mollard, J. Magn.
Magn. Mater. 120 (1993) 69.
[29] G.T. Rado, R.W. Wright, W.H. Emerson, A. Terris, Phys.
Rev. 88 (1952) 909.
8] S. Kulkarni, J. Shrotri, C.E. Deshpande, S.K. Kate,
J. Mater. Sci. 24 (1989) 3739.
[30] A. Vijayalakshmi, N.S. Gajbhiye, J. Appl. Phys. 83 (1998)
400.
9] S.E. Jacobo, C. Domingo-Pascual, R. Rudriguez-Clem-
ente, M.A. Blesa, J. Mater. Sci. 32 (1997) 1025.
10] W. Roos, J. Am. Ceram. Soc. 63 (1980) 601.
[31] K.V.P.M. Shafi, A. Gedanken, R. Prozorov, J. Balogh,
Chem. Mater. 10 (1998) 3445.
[
[
11] K. Haneda, C. Miyakawa, K. Goto, IEEE Trans. Magn.
[32] J. Fang, J. Wang, L-M. Gan, S-C. Ng, J. Ding, X. Liu,
J. Am. Ceram. Soc. 83 (2000) 1049.
2
3 (1987) 3134.
12] J.-H. Choy, Y.-S. Han, S.-W. Song, Mater. Lett. 19 (1994)
57.
13] C. Surig, K.A. Hempel, D. Bonnenberg, Appl. Phys. Lett.
3 (1993) 2836.
[
[
[
[33] A. Ataie, S. Heshmati-Manesh, J. Eur. Ceram. Soc. 21
(2001) 1951.
2
[34] K.S. Suslick, T. Hyeon, M. Fang, Chem. Mater. 8 (1996)
2172.
6
14] R. Ardiaca, R. Ramos, A. Isalgue, J. Rodriguez, X.
Obradors, M. Pernet, M. Vallet, IEEE Trans. Magn. 23
[35] C. Yee, G. Kataby, A. Ulman, T. Prozorov, H. White,
A. King, M. Rafailovich, J. Sokolov, A. Gedanken,
Langmuir 15 (1999) 7111.
(1987) 22.