Remanence Properties Of Barium Hexaferrite, Volume 369

Journal of Alloys and Compounds 369 (2004) 185–189

Remanence properties of barium hexaferrite
P.E. Garcia-Casillas a,d,∗ , A.M. Beesley b , D. Bueno c , J.A. Matutes-Aquino c , C.A. Martinez d
a

Depto. de Inv. Grupo Cementos de Chihuahua S.A. de C.V., Chihuahua, Chih., Mexico
Department of Physics, University of Liverpool, Oxford Street, Liverpool L69 7ZE, UK
c Centro de Investigacion en Materiales Avanzados, Miguel de Cervantes 120, Complejo Industrial Chihuahua, C.P. 31109, Chihuahua, Chih., Mexico
d Instituto de Ingenier´ıa y Tecnolog´ıa Universidad Autónoma de Ciudad Juárez, Ave. del Charro # 610 Norte, Cd. Juárez, Chihuahua, Chih., Mexico
b

Abstract
Barium hexaferrites have been widely used as permanent magnets, however one of the most important parameters that characterizes these
ferrites is the switching field distribution (SFD). This work calculates the SFD on a small particle barium ferrite (BaFe12 O19 ) obtained by
coprecipitation from chlorides in an alkaline medium using the irreversible component of the magnetization. Forward and reverse switching
field distribution curves were obtained by differentiation of isothermal remanent magnetization (IRM) and dc demagnetisation (DCD) curves.
It was found that both values differ by a factor of 3.5, quite away from the value of non-interacting systems. The Henkel plot was built from
these data sets, indicating a predominant region with demagnetising interaction between particles, and a small region in which the particles
interact constructively to the magnetization, according to the Preisach model framework.
© 2003 Elsevier B.V. All rights reserved.
Keywords: Chemical synthesis; Scanning and transmission electron microscopy; Magnetic measurements

1. Introduction
Barium ferrite (BaFe12 O19 ) has been commonly used as
permanent magnet since its development in the beginning
of the 1950s by Phillips researches. In fact, this material
presents high saturation magnetization, high Curie temperature, high intrinsic coercivity and rather large crystal magnetic field anisotropy [1,2]. Due to these properties, many
methods of synthesis have been developed to obtain a low
production cost of fine particles of barium ferrite with a
good chemical homogeneity, in this sense coprecipitation

seems to be the most suitable at the moment [2–5]. However it is of great importance the level of stacking of Ba
ferrite particles which is believed to determine the magnetic interparticle interactions, affecting essential properties
such as thermal duplication, media noise and resolution in
longitudinal recording media [1,6]. The deviation from the
Stoner–Wohlfarth model is a technique to estimate interparticle interactions but it is restricted to uniaxial single domain
particles. Barium ferrite obtained by coprecipitation consists
of hexagonal plates and can be considered as a uniaxial system, however the size of these particles determines whether
∗ Corresponding author. Tel.: +52-614-442-3100;
fax: +52-614-442-3288/3275.
E-mail address: (P.E. Garcia-Casillas).

0925-8388/$ – see front matter © 2003 Elsevier B.V. All rights reserved.
doi:10.1016/j.jallcom.2003.09.100

or not they are single- or multi-domains. Therefore a more
general theory is needed to describe these interactions. The
Preisach hysteresis model provides the framework to make
the Wohlfarth relation valid for multi-domain systems. This
model considers the magnetic material as a distribution of
small hysteresis loops or “hysterons”, which are determined
by (hc , hu ) units in which hc represents its coercive field and
hu is the displacement of the hysteron from origin (0, 0) in
the magnetization versus field plane [7]. These hysterons can
be considered as Stoner–Wohlfarth particles only if they represent the reversible component of the magnetization. The
values hc and hu of each hysteron determine the Preisach
plane (hc versus hu ). The three remanent states, magnetising
(Mr ), demagnetising (Md ) and saturation (M∞ ) are associated with three partitions of the Preisach plane [7,8]. Considering that the Preisach distribution is concentrated along the
hc -axis (hu = 0), the upper limit of the Henkel plot region is
determined by the relation Md = M∞ − 2Mr , as the Wohlfarth model. In that case if Md ≥ (<)M∞ − 2Mr , the particles present a magnetizing (demagnetizing) nature. On the
other hand if the Preisach distribution is focused around a local coercivity h0 in which all the units present the same coercivity value hc = h0 , it leads to the lower limit of the Henkel

plot region as Md = −Mr [7,8]. According to this, the
known expression that determines the Henkel plot regions,
are deduced from the following inequalities −Mr ≤ Md ≤


186

P.E. Garcia-Casillas et al. / Journal of Alloys and Compounds 369 (2004) 185–189

M∞ − 2Mr [7,8]. The interparticle interaction process consists mainly on two contributions, the mean field effects and
the local disorder of the particles, which compete with each
other giving rise to magnetizing and demagnetizing effects.
In this work the type of interaction that governs our system
is deduced using the Henkel plot and the δm curves. Isothermal remanent magnetization (IRM) and dc demagnetization
(DCD) data were taken and analysed in this framework. The
first differential of these curves gives a direct measurement
of irreversible changes in the magnetization states, from
which the switching field distribution was calculated.

2. Experimental
Barium ferrite powders were produced by coprecipitation
of Fe(III) and Ba(II) nitrate solutions in an alkaline aqueous
medium. The ratio Fe:Ba was kept at 11:1. The alkaline solution was a mixture of NaOH (50 g) and Na2 CO3 (12.5 g),
with pH 13. The precipitate appeared after adding both nitrate solutions into the alkaline medium under vigorous stirring, after which it was washed and dried at 40 ◦ C for 24 h.
Further calcination at different temperatures (600, 700, 750,
800, 925, 950, 1000 and 1100 ◦ C) for 2 h was carried out in
order to study the influence of the calcination temperature on
the magnetic properties. In each case the mean particle size
was measured using an isothermal adsorption technique.
The magnetic properties of the samples were measured

at room temperature using a Vibrational Sample Magnetometer LDJ Electronics 9600, with a maximum applied
field of 1.5 T. The morphology of the samples were examined by transmission electron microscopy using a Phillips
CM-200 microscope. Associated to this, the mean parti-

Fig. 2. Transmission electron micrograph of barium ferrite powder, calcinated at 925 ◦ C.

cle size was measured using the Brunauer–Emmett–Teller
method (BET).
The IRM curve was obtained after ac demagnetization
followed by a positive applied field from where the remanent magnetization was then recorded. This procedure
was repeated while gradually increasing the field strength
until positive saturation remanence was obtained. The DCD

Fig. 1. Magnetic properties of barium ferrite powder calcinated at different temperatures.


P.E. Garcia-Casillas et al. / Journal of Alloys and Compounds 369 (2004) 185–189

curve was obtained in a similar manner. Initially the sample was saturated with a positive field of 1.5 T. A negative field was applied and the remanent magnetization was
recorded. This procedure was repeated while gradually increasing the field strength until negative saturation remanence was obtained. The δm values were obtained from the
relation Md − [M∞ − 2Mr ] and the switching field distribution from the irreversible contribution was calculated as
(M∞ /Hc )/χr |Hc .

3. Results and discussion
Fig. 1 shows the specific magnetization, intrinsic coercive field and mean particle size value as function of the

187

calcination temperature. The particle size obtained at different temperatures, determines the magnetic properties of the
barium ferrite, as expected [9]. The intrinsic coercivity increases with the reduction of particle size up to 700 ◦ C after

which tends to decrease. On the other hand the specific magnetization is dependant of the amount of crystalline barium
ferrite obtained at different calcination temperatures. X-ray
diffraction measurements indicate that the crystalline phase
appears when the precipitate was calcinated at 700 ◦ C [10],
and in fact the particle size shown in Fig. 1 starts to increase
at this temperature. At lower temperatures (<600 ◦ C) the
samples present a mean specific magnetization of 12 emu/g,
rising to 53 emu/g between 700 and 1100 ◦ C. These results are comparable with those found in the literature
[9,11].

Fig. 3. (a) Remanent magnetization curves IRM and DCD; (b) IRM and DCD magnetic susceptibilities as function of the applied field, for a Ba ferrite
calcinated at 925 ◦ C.


188

P.E. Garcia-Casillas et al. / Journal of Alloys and Compounds 369 (2004) 185–189

Fig. 4. Behaviour of the δm curve with the magnetic applied field, for a Ba ferrite calcinated at 925 ◦ C.

The morphology of the sample treated at 925 ◦ C is shown
in Fig. 2. The hexagonal plates observed are typical of well
crystalline barium ferrite. The particle size is in agreement
with those found by BET. The thickness of the particles,
varies around of 80–300 nm. According to the relation obtained by Pfranger et al. [11], the dependence of the domain
wall width D on the thickness L of barium ferrite crystals
along the magnetically preferred direction is given by D =
1.53L0.32 for L < 500 ␮m. Using this relation the domain
wall width in our particles lie around 6.2–9.5 nm. In this
sense it is possible to study the interparticle interactions us-

ing the Wohlfarth relation if the system is treated under the
Preisach framework.
The IRM and DCD remanence curves are shown in
Fig. 3a. The remanence at saturation M∞ , corresponds to
the Mr value at 15 kOe, equal to 0.92 emu/cm3 . The first
differential of these curves gives a direct measure of irreversible changes in the magnetization states χr and χd ,
shown in Fig. 3b. By differentiation of the Wohlfarth relation, one obtains the remanence susceptibility ratio as
χd /χr = 2, for non-interactive systems. In this case the amplitude of χd and χr peaks, differ by a factor of 3.5 with no

Fig. 5. Henkel plot for a Ba ferrite calcinated at 925 ◦ C, showing the upper and lower limits in dashed lines.


P.E. Garcia-Casillas et al. / Journal of Alloys and Compounds 369 (2004) 185–189

coincidence of widths or maximum positions. This indicates
a strong interparticle interaction that could be due to local
disorder of the system or to the mean magnetic field distribution associated to each particle, producing magnetizing
or demagnetizing-like effects, at certain values of magnetic field. The switching field distribution was calculated
considering only irreversible events, i.e. (M∞ /Hc ) /χe |Hc ,
leading to a value of 0.61 [12].
Fig. 4 shows the δm plot as a function of the applied field.
For a non-interactive system, this plot would show a horizontal line through the origin. In this case this plot presents
a positive region between zero applied field and 500 Oe,
associated to interparticle interactions that contribute constructively to the magnetization following the applied field
(magnetizing-like effect). The opposite occurs at higher applied fields; from 500 Oe to saturation the particles interact
against the field producing a demagnetizing-like effect. The
values of δm in the positive region are comparable to those
found in hard disks and tapes [6,12]. On the other hand the
values of δm in the negative region are comparable to those

found in floppy disks [6,12].
The Henkel plot built from the IRM and DCD data is
shown in Fig. 5, where the upper and lower limits are also
represented. The positive region of the Henkel plot and its
shape is associated according to the literature to a system in
which the particles are interacting with each other governed
by the mean field distribution of the system up to a value
of field around 500 Oe, at that point the competition starts
between the mean field and the local disorder of the particles and the latter finally governs the system as the curve
approaches to the lower limit of the Henkel region.

189

4. Conclusion
It has been studied the remanence properties of barium ferrite using the Wohlfarth relation in the Preisach framework.
The system shows a magnetizing interparticle interaction
where the mean field effects are dominating between zero
applied field and 500 Oe. Above this field the system shows
a demagnetizing-like interparticle interaction with government of local disorder. A further investigation will clarify
the influence of the particle size on the remanent properties.

References
[1] P. Campbell, Permanent Magnet Materials and Their Applications,
Cambridge University Press, Cambridge, 1994.
[2] S.R. Janasi, M. Emura, F.J.G. Landgraf, D. Rodrigues, J. Magn.
Magn. Mater. 238 (2002) 168–172.
[3] T. González-Carreoo, M.P. Morales, C.J. Serna, Mater. Lett. 43 (2000)
97.
[4] W. Zhong, W.P. Ding, N. Zhang, J.M. Hong, Q.J. Yan, Y.W. Du, J.
Magn. Magn. Mater. 168 (1997) 196.

[5] M.L. Wang, Z.W. Shih, J. Cryst. Growth 114 (1991) 435.
[6] K. Yamanaka, Y. Uesaka, T. Okuwaki, J. Magn. Magn. Mater. 127
(1993) 233–240.
[7] G. Bertotti, Hysteresis in Magnetism, Academic Press, New York,
1998.
[8] E.D. Torre, Magnetic Hysteresis, IEEE Press, New York, 1999.
[9] J. Huang, H. Zhuang, W. Li, J. Magn. Magn. Mater. 256 (2003)
390–395.
[10] P. Castillas, Ph.D. Thesis, CIMAV, Chihuahua, Mexico.
[11] R. Pfranger, D. Plusa, S. Szymura, B. Wyslocki, Physica 101B (1980)
239–242.
[12] N. Kodama, J. Magn. Magn. Mater. 224 (2001) 113–123.


ARTICLE IN PRESS

Journal of Magnetism and Magnetic Materials 270 (2004) 77–83

Magnetic studies of NiSn-substituted barium hexaferrites
processed by attrition milling
! c,
A. Gonza! lez-Angelesa, G. Mendoza-Sua! reza,*, A. Gruskova! b, I. Toth
V. JanWa! rikd, M. Papa! nova! b, J.I. Escalante-Garc!ıaa
a

Cinvestav-IPN Unidad Saltillo Ceramics, Carr. Saltillo-Mty. Km. 13, P.O. Box 663, 25000 Saltillo,
Coahuila, Mexico
b
Department of Electrotechnology, Faculty of Electrical Engineering and Information Technology (FEEIT),
Slovak University of Technology, Ilkovic˘ova 3, 812 19 Bratislava, Slovak Republic

c
Department of Nuclear Physics and Technology, Faculty of Electrical Engineering and Information Technology (FEEIT),
Slovak University of Technology, Ilkovic˘ova 3, 812 19 Bratislava, Slovak Republic
d
Department of Electromagnetic Theory, Faculty of Electrical Engineering and Information Technology (FEEIT),
Slovak University of Technology, Ilkovic˘ova 3, 812 19 Bratislava, Slovak Republic
Received 23 January 2003; received in revised form 4 August 2003

Abstract
BaFe12À2x(NiSn)xO19 compounds were synthesized via attrition milling and characterized. The substitution, x was
changed from 0 to 0.3. The magnetic properties were evaluated by vibrating sample magnetometry and thermomagnetic
.
analysis. Mossbauer
spectroscopy and X-ray diffraction techniques were used to investigate the magnetocrystalline
structure. Our experimental results suggest that Ni–Sn cationic mixtures produce similar effects on the magnetic
properties, to those observed for Ir–Co and Ir–Zn substitutions. The Ni2+–Sn4+ substitution reduced the uniaxial
.
magnetic anisotropy field, Ha rather fast, becoming planar at low rates. Mossbauer
studies revealed that Sn4+ ions
3+
2+
replace Fe
ions on 2b and 4f1 sites, while Ni
ions preferred 4f2 and 2a sites. A large variation of the intrinsic
coercivity, Hci ; (381.9–93.9 kA/m) was obtained as a function of the substitution.
r 2003 Elsevier B.V. All rights reserved.
PACS: 71.20.Ps,75.50,74.25.Ha,75.50.Ss,33.45.+x
.
Keywords: Hexagonal ferrites; Substitution effects; Magnetic properties; Mossbauer
spectra

1. Introduction
*Corresponding author. Cinvestav-IPN Unidad Saltillo
Ceramics, Carr. Saltillo-Mty. Km. 13, P.O. Box 663, 25900
Ramos Arizpe Coahuila, Mexico. Tel.: +52-844-4389600;
fax: +52-844-4389610.
E-mail addresses:
(A. Gonz!alez-Angeles),
(G. Mendoza-Su!arez).

Nowadays, the demand for the development of
new magnetic materials has increased extraordinarily, owing to their applications in new emerging
technologies. This relates to the big research effort
carried out on barium hexaferrites (BaM), BaFe12O19, used mainly for the permanent magnets.

0304-8853/$ - see front matter r 2003 Elsevier B.V. All rights reserved.
doi:10.1016/j.jmmm.2003.08.001


ARTICLE IN PRESS
78

!
A. Gonzalez-Angeles
et al. / Journal of Magnetism and Magnetic Materials 270 (2004) 77–83

However, by substituting other cations to Fe3+
and/or Ba2+, it is possible to obtain new properties leading to a variety of applications. In
addition, substituted hexaferrites are very promising materials as electronic wave absorbers, especially at high frequencies (1–50 GHz), and compete
with the spinel compounds, at least in the same

frequency range [1]. The replacement of Fe3+ ions
by other metallic cations reduces the high uniaxial
magnetic anisotropy field, Ha ; of the BaFe12O19
compound. At certain substitution rate, a change
from uniaxial to planar anisotropy, Hp ; occurs,
which makes the material suitable for microwave
absorbers, as long as an important saturation
magnetization, Ms is maintained.
The magnetic properties of substituted hexaferrites, depend directly on the electronic configuration of the dopant cations and on their
preference to occupy the different Fe sublattices
of the magnetoplumbite structure. For instance,
it is known that Ni2+ reduces the temperature
coefficient of coercivity, dHci =dT; which is an important parameter for the stability of the recording
data. Ideally, this should be zero for magnetic
applications [2]. In addition, it has been
shown that doping with Ni+2 ions, both the
saturation magnetization (Ms ) and the intrinsic
coercivity (Hci ) can be easily controlled [3]. A
narrow switching field distribution (SFD) [4],
which is another important parameter for
high-density magnetic recording, can also be
controlled by Ni+2 substitution. Moreover, it has
been reported that Ni2+ ions are more effective
than Co2+ ions to reduce the uniaxial anisotropy
field of BaM [5]. Regarding Sn4+ or mixtures
of this and other cations, Fang et al. [6] found
that Zn–Sn substitutions diminish both the
anisotropy constants and the particle size, mainly
due to the effect of Sn4+. The properties also vary
with the processing parameters and conditions

[2,7].
As part of the constant search for new magnetic
materials, we report the effect of Ni2+–Sn4+
cationic mixtures, on the magnetic properties
of barium hexaferrites, synthesized by mechanical alloying, aiming at possible high-density
magnetic recording and microwave absorbers
applications.

2. Experimental procedure
BaFe12À2x(NiSn)xO19 powders with x ¼ 0; 0.1,
0.2 and 0.3 were prepared by mechanical alloying.
Metallic oxides (Fe2O3, NiO, and SnO2) and
barium carbonate (BaCO3), (ACS reagent B98%
purity, Aldrich Co.) were used as raw materials.
Mechanical alloying was performed in a Segvary
Attritor using a ball/powder ratio of 15. Milling
was carried out for 30 h in air with an angular
frequency of 400 rpm and a Fe/Ba ratio of 10. A
liquid medium (250 ml of benzene) was used to
avoid agglomeration of powders at the bottom of
the mill, and to assure active participation of
powders in the milling process. After mechanical
milling, the samples were annealed at 1050 C for
1.5 h.
The magnetization was measured with a Lake
Shore 7300 vibrating sample magnetometer (VSM)
with an applied field up to 1.2 T. To determine the
distribution of Ni2+ and Sn4+ in the hexagonal
.
structure, Mossbauer

spectroscopy studies were
performed using a Spectrometer operated at a
constant acceleration mode. The g-ray source
.
was 57Co in a rhodium matrix. The Mossbauer
spectra were fitted using the NORMOS software
package.
The temperature dependence of the magnetic
susceptibility, wðyÞ was measured employing the
bridge method in an alternating magnetic field
of 360 A/m and 1 kHz. The sample was heated
up to 730 C at a constant rate of 4 C/min.
The measurement of wðTÞ is, in many cases, an
appropriate and powerful method to analyze
several kinds of magnetic materials. It can
reveal
information
concerning
crystalline
phases, magnetic and chemical stability, that other
techniques cannot. This technique is more sensitive
to low contents of impurities than X-ray diffrac.
tion and Mossbauer
spectroscopy. The Curie
temperature, Tc ; was determined from the Hopkinson’s effect in the susceptibility curves. The
susceptibility is given in arbitrary units and is
related to the same amount of sample (1 mg) in all
cases.
The identification of crystalline phases was
carried out by X-ray diffraction in an X 0 Pert

Philips diffractometer using Cu-Ka radiation.


ARTICLE IN PRESS
!
A. Gonzalez-Angeles
et al. / Journal of Magnetism and Magnetic Materials 270 (2004) 77–83
480

3. Results and discussions
x= 0.3

400

320

Intensity (a.u.)

79

x= 0.2

240

x= 0.1

160

80

x= 0

0
22

32

42

52

62

2θ

Fig. 1. X-ray diffraction patterns for BaFe12À2x(NiSn)xO19
with x ¼ 020:3:

I [%]
↑
100

4f2
2a
12k 4f1
2b

98

The analysis of the X-ray diffraction patterns

revealed that only the hexaferrite phase (BaFe12À2x(NiSn)xO19) was present in the samples
after milling; no secondary phases were detected,
at least, within the errors inherent to the technique
(Fig. 1).
.
The Mossbauer
spectra at room temperature are
presented in Fig. 2; these were fitted according to
Ref. [8]. The spectra were analyzed in terms of five
Zeeman sextets [9] for x ¼ 0; which correspond to
the five different Fe3+ sites (4f2, 2a, 4f1, 12k and
2b) in the magnetoplumbite structure. The intensity of each sextet is directly proportional to the
number of iron atoms in that site, so that, it gives
an estimate of the occupancy rate on that site.

I [%]
↑

4f2
2a
4f1
12k
12k'
2b

100
98

96
96

94

94

92

(a)

-10

-8

-6

-4

-2

0

2

4

6

I [%]
↑

8 10
+ v [mm/s]
4f2
2a
4f
12k 1
12k'
2b

100

(b)

96

98

94

97

-8

-6

-4

-2

0

2

4

6

8 10
+ v [mm/s]

-6

-4

-2

0

2

4

6

12k
12k

2b

-10

(d)

8 10
+ v [mm/s]
4f
2a
4f-A

100
99

-10

-8

I [%]
↑

98

(c)

-10

-8

-6

-4

-2

0

2

4

6

8 10
+ v [mm/s]

.
Fig. 2. 57Fe Mossbauer
spectra at room temperature. (a) A high purity barium hexaferrite sample with all five sublattices shown (+7
to +10 mm/s) interval, (b) 0.1, (c) 0.2, and (d) 0.3 substitution.


ARTICLE IN PRESS
!
A. Gonzalez-Angeles
et al. / Journal of Magnetism and Magnetic Materials 270 (2004) 77–83
4f2

52
2a
48

Hef ( T )

The substitution of part of the Fe3+ ions by
other magnetic or diamagnetic cations leads to
both changes in the exchange interactions between
the magnetic sublattices, and to the appearance of
new positions of the iron ions. Therefore, as x
increases, the 12k site can split into two subpatterns, 12k and 12k0 . This may be the result of
changes in the environments or neighbors of the
Fe3+ ions on the 12k site, when substitution takes
place at R blocks. The 12k0 site increases at the
expense of the disturbed 12k sites.
According to ligand field theory [2], ions with
d1, d2, d3 and d4 orbitals prefer tetrahedral
coordination and ions with d6, d7, d8 and d9
orbitals occupy octahedral positions. Ions with d0,
d5 and d10 orbitals have no site preference. It is
also known that less electronegative ions prefer
tetrahedral coordination [7]. However, the tendency to occupy a particular site depends also of
the partner cation. Under these considerations,
and taken into account that Ni2+ and Sn4+ ions
possess nearly the same electronegativity (1.91 and
1.96, respectively), we believe that Ni2+ (d8) ions
would go to octahedral (12k, 2a, 4f2) sites, and
Sn4+ (d10) ions would to the tetrahedral (4f1) and
bipyramidal (2b) ones.
Figs. 3 and 4 show the variation of the relative
areas (S %) and hyperfine fields (Hef ), respectively,
with increasing doping rate. From Fig. 3, it is
observed that the reduction of S (%) for all sites,

4fi

44
12k

40

2b
12k1

36
0

0.1

0.2

0.3

x

(a)

0.3

12k
12k 1
2a

4fi

Hef ( T )

80

4f2

-0.5

2b
-1.3
0

(b)

0.1

0.2

0.3

x

Fig. 4. Variation of Hef ðTÞ vs. substitutions x: (a) Variation of
DHef ðTÞ with substitution.

50
12k

S (%)

40

30
4f1
20

12k1

4f2
10

2a
2b

0
0

0.1

0.2

0.3

x

Fig. 3. Variation of the relative areas S (%) with increasing
doping rate.

owes to the replacement of iron ions by Ni2+ and
Sn4+, with the exception of 4f1. The observed
increase in 4f1 at low substitutions might indicate
the nucleation of a secondary phase, with a
hyperfine field near that of BaM at 4f1 position,
which can presumably be the nickel spinel
(NiFe2O4) compound. This possesses a Hef of
B50 T at the tetrahedral site and of B45 T in the
octahedral one. The Hef fall for 4f1 at higher
substitutions was related to the fitting of the
spectra with seven subspectra (Fig. 2d), where
those for NiFe2O4 (tetraHef B48.8 T and octaHef B45.9 T) can be observed.
As can be seen from Fig. 4, Hef decreased for all
sites as x increased. In addition, a negative DHef


ARTICLE IN PRESS
!
A. Gonzalez-Angeles
et al. / Journal of Magnetism and Magnetic Materials 270 (2004) 77–83

slope from x ¼ 0 to 0.2 and a positive from x ¼
0:2 to 0.3 were observed. These changes of slope
can be related to the variation in the lattice
volume, taking into account the differences of
ionic radii between Ni2+, Sn4+ and Fe3+ ions.
Rane [2] has reported that a fall in Hef is due to
lattice expansion, substitutional effects and grain
growth. From Fig. 4b, it is observed that at low
substitutions, DHef was maximum for 2b followed

by 4f2 and 4f1 sites, indicating the strong
preference of Ni2+ and Sn4+ ions for these sites.
The least affected were 12k and 2a sites. At higher
substitutions, DHef behavior indicates that 2a site
was also substituted, probably by Ni2+ ions. The
steep fall of the hyperfine field at 2b positions
could be related to the substitution of Ni2+ ions
on 4f2 site, since 2b is coupled to both 4f2 and 12k.
The slight drop in Hef for 2a may be explained,
taking into account that 4f1 couples to both 2a and
12k, so an additional decrease can be expected.
This reinforces the assumption that Sn4+ replaced
Fe3+ on 2b and 4f1, and Ni2+ did on 4f2 and 2a
sites, respectively.

6
x= 0.2

5

χ (a. u.)

4

x= 0.3

3
x= 0.1

NiFe2O4

2
x= 0
1
0
0

200

400

600

800

θ (°C)

Fig. 5. Temperature dependence of the magnetic susceptibility,
w (a.u.), of BaFe12À2x(NiSn)xO19 with x ¼ 020:3:

81

Measurements of the temperature dependence
of the magnetic susceptibility, wðyÞ (Fig. 5) showed
that at room temperature, w increased with the Ni–
Sn substitution. w depends partly on the contribution of noncollinear spins in the magnetic structure. Consequently, it increases due to the
disruption of the collinear uniaxial magnetic
structure for the replacement of Fe3+ ions, by
either nonmagnetic or less magnetic cations. In
Ni–Sn substituted BaM, the iron ions are replaced

by nonmagnetic (Sn4+) and less magnetic
(Ni2+B2 mB) cations, provoking partial disappearance of the superexchange interactions between
Fe3+ ions via O2À, leading to weakening of the
collinear uniaxial magnetic structure.
Table 1 summarizes the Curie temperatures, Tc
determined from the Hopkinson’s effect, showing
in the wðyÞ curves (Fig. 5). The increase of the
substitution level caused a slight Tc decrease. This
could be explained by the reduction of both, the
strength of superexchange interactions and the
magnetic interactions among iron ions. Also from
Fig. 5, the nucleation of a secondary phase was
evident, which was not detected by XRD analyses.
As mentioned earlier, this phase is most likely to
be the NiFe2O4 compound, showing Ms of
B50 A–m2/kg, Hci B1 kA/m and Tc of 585 C.
This assumption could be supported by the
.
Mossbauer
measurements and the results obtained
by Turilli et al. [5], who found difficult to avoid the
formation of NiFe2O4 as a secondary phase in Ni–
Ti substituted SrM. Since, NiFe2O4 was not
identified by XRD; we assume that the volume
fraction of this spinel is low.
The magnetization (M) vs. external magnetic
field (H) measurements (Fig. 6), taken up to a
maximum applied field of 1.2 T, showed that M
was not saturated. Table 1 and Fig. 6 show the
effect produced by Ni–Sn mixtures on both the

Table 1
The effect of NiSn-substitution on magnetic properties and Curie temperature
x

Ms (A-m2/kg)

Hci (kA/m)

Mr (A-m2/kg)

Mr =Ms

Tc ( C)

0
0.1
0.2
0.3

61.0
61.6
59.6
60.6

381.9
213.2
151.2
93.9

34.8
33.3
31.0
27.8

0.57
0.54
0.52
0.46

442
443
438
434


ARTICLE IN PRESS
!
A. Gonzalez-Angeles
et al. / Journal of Magnetism and Magnetic Materials 270 (2004) 77–83

82
60

Ms (A-m2/kg)

40
20

x= 0

0

x = 0.3

-20
-40
-60
-1.2

-0.8

-0.4

0

0.4

0.8

1.2

H (T)

Fig. 6. Typical plots of M vs. H for samples with limit values
of x (0 and 0.3) at room temperature.

Ms

65

400

4. Conclusions

H ci
200
35

Hci (kA/m)

Ms (A-m2/kg)

300
50

100

Mr
20

0
0

0.1

0.2

2b positions have a strong influence on Ha ; which
may become planar at small substitution rates.

Vincent et al. [10] have shown that the uniaxial
magnetic anisotropy for Ir–Co and Ir–Zn mixtures
becomes planar at smaller substitutions (xB0:4)
than for Co–Ti mixtures (xB1:3). Regarding Mr ;
this registered a fall of B20%, as x increased,
presumably, owing to the increase of the volume
fraction of NiFe2O4 in the structure. On the other
hand, Ms remained practically constant. Our results
suggest that the effects of Ni–Sn substitutions on the
magnetic properties are similar to those produced
by Ir–Co and Ir–Zn mixtures. However, from the
economical point of view, Ni and Sn-bearing oxides
are cheaper than rare earth oxides or salts.

0.3

x

Fig. 7. Magnetic properties vs. NiSn-substitution.

hysteresis loop and the magnetic properties. The
slope of the initial magnetization curve gives one
idea of the behavior of the magnetic susceptibility.
It can be seen that w increased with the Ni–Sn
substitution. This observation is in agreement with
the temperature dependence of the magnetic
susceptibility, wðyÞ measurements (Fig. 5).
The behavior of the magnetic properties of
NiSn-substituted BaM is shown in Fig. 7. Both,
the intrinsic coercivity, Hci ; and the remanent

magnetization, Mr ; decreased as the substitution
took place, while the saturation magnetization,
Ms ; remained almost constant. Hci ; decreased from
381.98 up to 93.9 kA/m, which represents a 75%
drop. The fast reduction of Hci is related to the
reduction of the magnetocrystalline anisotropy
field, Ha ; which has been attributed to the
preference of the Sn4+ ions for the bipyramidal
site, 2b. This site has the largest contribution to Ha
in BaM. Therefore, the substitutions of Fe3+ ions

It was possible to synthesize NiSn-substituted
barium hexaferrites by mechanical alloying. These
possess the required microstructure and magnetic
properties for high-density magnetic recording
applications. The magnetic ordering was reduced
somewhat (B2%) for small levels of substitution.
Changing the substitution rate x; the coercivity
could be easily controlled without a significant
reduction of Ms : The magnetic susceptibility
measurements showed that with the increase in x;
w increased, which is believed to relate to the
disappearance of some superexchange interactions. The Sn4+ cations substituted Fe3+ mainly
on the bipyramidal (2b) and slightly on tetrahedral
(4f1) site, while the Ni2+ ions preferred the
octahedral (4f2 and 2a) sites at low and high
substitution rates, respectively. Our investigation
suggests that the Ni-Sn mixtures produce similar
effects on the magnetic properties than those
reported for Ir–Co and Ir–Zn mixtures.

Acknowledgements
The authors to thank CONACYT-Me! xico and
VEGA Scientific Agency of Ministry of Education
of the Slovak Republic, for the support given to
carry out this work under the projects J28283U
and No. G-1/7610/20, respectively. Also, we


ARTICLE IN PRESS
!
A. Gonzalez-Angeles
et al. / Journal of Magnetism and Magnetic Materials 270 (2004) 77–83

acknowledge Prof. Lipka and Prof. Sla! ma for their
contributions to this work, along with Darina
Kevicka! for her help in the preparation of the
samples.
References
[1] J. Kreisel, H. Vincent, F. Tasset, M. Pat!e, J.P. Ganne,
J. Magn. Magn. Mater. 224 (2001) 17.
[2] M.V. Rane, D. Bahadur, A.K. Nigam, C.M. Srivastava,
J. Magn. Magn. Mater. 192 (1999) 288.
[3] Z. Yang, C.S. Wang, X.H. Li, X.H. Zeng, Mater. Sci. Eng.
B90 (2002) 142.

83

[4] N. Nagai, N. Sugita, M. Maekawa, J. Magn. Magn.
Mater. 120 (1993) 33.

[5] G. Turilli, F. Licci, A. Paoluzi, T. Besagni, IEEE Trans.
Magn. 24 (1988) 2146.
[6] H.C. Fang, Z. Yang, C.K. Ong, Y. Li, C.S. Wang,
J. Magn. Magn. Mater. 187 (1998) 129.
[7] M.V. Rane, D. Bahadur, S.D. Kulkarni, S.K. Date,
J. Magn. Magn. Mater. 195 (1999) L256.
[8] J. Lipka, M. Miglierini, J. Elec. Eng. 45 (1994) 12.
.
[9] J. Gary, I. Long, Mossbauer
Spectroscopy Applied to
Inorganic Chemistry, Vol. 1, Plenum Press, New York,
1984.
[10] H. Vincent, B. Sugg, V. Letez, B. Bochu, D. Boursier,
P. Chaudovet, J. Magn. Magn. Mater. 101 (1991)
170.


Materials Letters 58 (2004) 1147 – 1153
www.elsevier.com/locate/matlet

X-ray diffraction studies on aluminum-substituted barium hexaferrite
D. Mishra a, S. Anand a,*, R.K. Panda b, R.P. Das a
a

Regional Research Laboratory (CSIR), Hydro and Electro Metallurgy Division, Bhubaneswar, 751 013, Orissa, India
b
Material Science Division, Department of Chemistry, Berhampur University, Berhampur 760 007, Orissa, India
Received 16 January 2003; received in revised form 15 August 2003; accepted 21 August 2003

Abstract

Effect of aluminum substitution in barium hexaferrite was studied following the hydrothermal precipitation – calcination techniques. It was
attempted to prepare aluminum-substituted barium hexaferrites with compositions BaAlxFe12 À xO19 having x = 2,4, 6, 8 and 10. The
precursors were prepared by using stoichiometric amounts of Ba, Al and Fe3 + nitrate solutions with urea as the precipitating agent. The
hydrothermally prepared precursors were calcined at temperatures in the range of 800 – 1200 jC. The detailed work carried out on
identification of crystalline phases through XRD revealed that, after the hydrothermal treatment, the samples showed barium carbonate,
hematite and boehmite as the crystalline phases (except that boehmite was not identified for Ba/Al/Fe ratio as 1:2:10). Irrespective of the Al
content, none of the 1000 and 1200 jC calcined samples showed any crystalline phase of Al. The 1200 jC calcined samples showed that Alsubstituted barium hexaferrite is formed only with the Ba/Al/Fe atomic ratio of 1:2:10. With increase in the Al content, formation of
hexaferrite was not observed. BaCO3 was found be present even at 1200 jC in all the samples except for the one having Fe/Al ratio 5. The
normal decomposition temperature of BaCO3 is between 950 and 1050 jC. It is difficult to explain the increased stability of BaCO3, which is
perhaps responsible for hindering the formation of aluminum-substituted barium ferrite having Fe/Al ratio V 2. The Al substitution in barium
hexaferrite was confirmed through magnetic measurements.
D 2003 Elsevier B.V. All rights reserved.
Keywords: Ferrites; Aluminum; Hydrothermal; Urea; Calcination

1. Introduction

2. Experimental

Studies on the substitution of various cations into the
lattice of barium or strontium ferrites have been reported in
order to vary their magnetic behavior [1 – 4]. Al3 + in
principle can be substituted for Fe3 + in the lattice of
BaFe12O19, and depending on the amount and site occupancy of Al the magnetic properties would change [5,6].
Preparation of Al-substituted hexaferrites by ceramic route
has been reported by Suzuki et al. [7] and Hameda and
Kojima [8]. In an earlier work, nanosize barium hexaferrite
was prepared following hydrothermal precipitation –calcination technique [9]. Since there is no literature available on
the effect of Al substitution on the structural properties of
barium hexaferrite, the present work has been taken up to
study the effect of aluminum substitution on the phase

formations of barium ferrite following a hydrothermal
precipitation – calcination technique.

In order to substitute aluminum in barium hexaferrite to
obtain compositions as BaAlxFe12 À xO19 (x = 2, 4, 6, 8 and
10) through the hydrothermal precipitation – calcination
route stoichiometric amounts of Ba, Al and Fe3 + nitrate
solutions with urea as the precipitating agent were taken
[stock solutions of metal nitrates (AnaLR grade) were first
prepared and required amounts were added]. Urea to total
metal ion molar ratio was kept as 2. The precipitation was
carried out in a 2-l Parr autoclave at 180 jC for 2 h. The
products obtained after hydrothermal treatment were filtered, washed till free of anions and dried in an air oven at
100 jC for 24 h. Analysis of filtrate for Al, Ba and Fe
showed presence of only ppm level of Ba confirming near
complete precipitation of metal ions. The dried samples
were calcined in a muffle furnace for 2 h at different
temperatures. The X-ray diffractograms of the powder
samples were taken using Philips PW 1710 X-ray diffractometer. The diffracted X-ray intensities were recorded as a
function of 2h using nickel-filtered copper target (Cu-Ka

* Corresponding author. Tel.: +91-674-581750; fax: +91-674-581750.
E-mail address: (S. Anand).
0167-577X/$ - see front matter D 2003 Elsevier B.V. All rights reserved.
doi:10.1016/j.matlet.2003.08.027


1148

D. Mishra et al. / Materials Letters 58 (2004) 1147–1153

radiation with k = 1.5404jA). The magnetic measurements
were carried out at room temperature with a Vibrating
Sample Magnetometer (VSM, PARR, Model 4500). The
field at the actual position was calibrated using a solid
nickel cylindrical sample. The maximum applied field of 15
kOe has been used to evaluate the magnetic parameters.

3. Results and discussion
The various samples prepared hydrothermally using
different Ba/Al/Fe ratio as well as the calcined samples
have been coded as given in Table 1.
3.1. XRD studies
3.1.1. XRD analysis of hydrothermally prepared precursors
The XRD patterns of three typical samples obtained
hydrothermally by initially taking Fe/Al ratio as: 5, 2 and
0.2 (corresponding to the desired compositions of
BaFe10Al2O19, BaFe8Al4O19 and BaFe2Al10O19 coded as
S-1, S-2, S-5) while maintaining urea to metal ratio as 2 are
shown in Fig. 1 and the corresponding d-values along with
identified phases are listed in Table 2a. It is observed that all
the samples contain barium carbonate [10a] and a-Fe2O3
[10b]. The sample having Fe/Al ratio 5 (S-1) does not show
any reflections (Fig. 1a) for aluminum oxide-hydroxides
(boehmite). When initial Fe/Al ratio decreased to 2.0 (S-2),
broad and weak peaks corresponding to boehmite appeared
at 2h = 14.48, 28.39 and 38.47j (Fig. 1b) having relative
intensities of 11%, 27% and 20.8%, respectively [10c].
Besides these three peaks, no other peaks for boehmite are
noticed in the XRD of this sample. Though the d-values

match with those of boehmite but the relative intensity are
different. The sample with initial Fe/Al ratio as 0.2 (S-5)
showed formation of well-crystalline boehmite with major
peaks at 2h = 14.80, 28.11, 38.515 and 49.130. The XRD
pattern suggests (Fig. 1c) that in this hydrothermally prepared precursor boehmite is the major crystalline phase
together with cubic BaCO3 and a-Fe2O3. Thus, the nonappearance of any peaks for boehmite in the sample S-1
may be due to the small weight fraction of Al in that sample,
or the presence of amorphous aluminum oxide-hydroxide
particles.

Table 1
Code numbers for various samples
Hydrothermally prepared samples
Ba/Al/Fe 1:2:10
Fe/Al ratio 5
S-1
Ba/Al/Fe 1:4:8
Fe/Al ratio 2
S-2
Ba/Al/Fe 1:6:6
Fe/Al ratio 1
S-3
Ba/Al/Fe 1:8:4
Fe/Al ratio 0.5
S-4
Ba/Al/Fe 1:10:2
Fe/Al ratio 0.2
S-5
Calcined at 800 jC for 2 h S-1800, S-2800, S-3800, S-4800, S-5800
Calcined at 1000 jC for 2 h S-11000 , S-21000, S-31000, S-41000, S-51000

Calcined at 1200 jC for 2 h S-11200, S-21200, S-31200, S-41200, S-51200

Fig. 1. X-ray diffraction patterns: (a) S-1, (b) S-2, (c) S-5. D: BaCO3, q: aFe2O3, o: g-AlOOH. (For sample identification, see Table 1.)

3.1.2. XRD analysis of 800 jC calcined samples
The XRD patterns of the calcined samples S-1800, S-2800,
S-3800, S-4800 and S-5800 (calcined for 2 h at 800 jC) are
shown in Fig. 2a – e. The d-values are given in Table 2b. All
the samples showed barium carbonate and hematite to be the
major crystalline phases. No crystalline phase of any metastable aluminum oxides were found in these samples. In a
previous study on barium mono-aluminate, similar amorphous forms of Al-oxide/hydroxides were observed [11]
after calcination at 800 jC. In all the XRD patterns, main
peak of maximum intensity for BaCO3 was observed for 2h
varying between 23.55j and 23.96j, while the other major
peak at 24.22j was not observed even when the main peak
had 100% relative intensity. In general, slight shift was
observed for the crystalline phases.
3.1.3. XRD analysis of the 1000 jC calcined samples
The XRD patterns of the samples calcined at 1000 jC for
2 h are shown in Fig. 3. Sample S-11000 (Fig. 3a, where Fe/
Al ratio is 5) shows peaks for a-Fe2O3, BaFe12O19, cubic
BaCO3 and some peaks of cubic BaO [10d]. Sample S-21000
to S-51000 (Fig. 3b –e, Table 2c) show major phases as cubic
BaCO3 and a-Fe2O3. The relative intensities of BaCO3


D. Mishra et al. / Materials Letters 58 (2004) 1147–1153

1149

Table 2a
XRD data of hydrothermally precipitated precursor samples obtained at
180 jC (data corresponding to Fig. 1)
Fe/Al ratio 5
sample S-1 (Fig. 1a)

Fe/Al ratio 1
sample S-3 (Fig. 1b)

Fe/Al ratio 0.2
sample S-5 (Fig. 1c)

d-value

Phase

d-value

Phase

d-value

Phase

4.578
4.456
3.730
3.680
3.256
2.910

2.690
2.650
2.509
2.280
2.202
2.150
2.020
1.860
1.833
1.730
1.693
1.647
1.598
1.481
1.450
1.315
1.281
1.243
1.225

D
D
D
D
D
D
j
D
j
D

j
D
D
D
j
D
j
Dj
j
j
j
Dj
D
j
j

6.110
4.450
3.715
3.680
3.630
3.140
2.680
2.645
2.497
2.344
2.194
2.142
2.095
2.041

2.010
1.930
1.830
1.690
1.640
1.592
1.478
1.450
1.364
1.347
1.328
1.307
1.280
1.220
1.180

o
D
D
D
j
o
j
D
j
o
j
D
D
D

D
D
j
j
Dj
j
j
j
D
Dj
D
Dj
D
j
j

5.995
4.487
3.689
3.180
2.680
2.651
2.580
2.496
2.341
2.273
2.147
2.101
2.014
1.940

1.857
1.678
1.648
1.628
1.520
1.450
1.370
1.328
1.308
1.280
1.270
1.251
1.233
1.210
1.180

o
D
D
o
j
D
D
j
o
D
D
D
Do
D

jo
jo
Dj
j
o
jo
Do
D
jo
D
j
D
jo
jo
jo

j: a-Fe2O3 (File No. 13-534), D: BaCO3 (File No. 5-378), o: g-AlOOH
(File No. 21-1307).

peaks increase while those of a-Fe2O3 decrease with the
decrease of Fe/Al ratio from 1 to 0.2. It is interesting to note
that no crystalline phases of Al are detected even in the
samples having very high Al concentration such as in S41000 and S-51000 samples (Fig. 3d and e).
3.1.4. XRD analysis of the 1200 jC calcined samples
All the precursors (S-1 to S-5) were calcined at 1200 jC
for 2 h and XRD patterns of the calcined sample are shown
in Fig. 4 and the d-values along with phases are given in
Table 2d. The XRD pattern of the S-11200 sample shows the
reflections for BaFe12O19 and a-Fe2O3. The major peaks of
barium hexaferrite showed negative shift of about 0.02 –

˚ in their d-values. With further decrease in the Fe/Al
0.04 A
ratio (S-21200, S-31200, S-41200 and S-51200) formation of
barium hexaferrite was not observed. The samples S-21200,
S-31200, S-41200 and S-51200 (Fig. 4b –e and Table 2d) show
a-Fe2O3 and BaCO3 as the only crystalline phases present.
The two main observations in the calcined samples are (a)
non-appearance of any crystalline phase of aluminum and
(b) the increased stability of BaCO3 as the normal decomposition temperature of BaCO3 is between 950 and 1050
jC. It is difficult to explain these observations. This study

Fig. 2. X-ray diffraction patterns of samples: (a) S-1800, (b) S-2800, (c)
S-3800, (d) S-4800, (e) S-5800. D: BaCO3, q: a-Fe2O3. (For sample
identification, see Table 1.)


1150

D. Mishra et al. / Materials Letters 58 (2004) 1147–1153

Table 2b
XRD data on samples calcined at 800 jC (data corresponding to Fig. 2)
Fe/Al
ratio 5
sample
S-1800
(Fig. 2a)

Fe/Al
ratio 2

sample
S-2800
(Fig. 2b)

Fe/Al
ratio 1
sample
S-3800
(Fig. 2c)

Fe/Al
ratio 0.5
sample
S-4800
(Fig. 2d)

Fe/Al
ratio 0.2
sample
S-5800
(Fig. 2e)

d-value Phase d-value Phase d-value Phase d-value Phase d-value Phase
3.720
3.20
2.698
2.629
2.52
2.20
2.155

2.060
2.038
1.936
1.834
1.691
1.604
1.484
1.452
1.308
1.261
1.228

D
D
j
D
j
j
D
D
D
D
j
j
j
j
j
Dj
j
j

3.785
3.243
2.717
2.644
2.530
2.218
2.167
2.05
1.946
1.847
1.70
1.604
1.489
1.458
1.315
1.261
1.231

D
D
j
D
j
j
D
D
D
j
j

j
j
j
Dj
j
j

3.768
3.250
2.713
2.648
2.535
2.217
2.167
2.049
1.951
1.847
1.695
1.604
1.489
1.457
1.314
1.250
1.229

D
D
j
D
j

j
D
D
D
j
j
j
j
j
Dj
j
j

3.750
3.246
2.701
2.638
2.520
2.209
2.157
2.110
2.050
1.942
1.840
1.696
1.598
1.485
1.453
1.311
1.258

1.229

D
D
j
D
j
j
D
D
D
D
j
j
j
j
j
Dj
j
j

3.780
3.263
2.721
2.644
2.537
2.217
2.169
2.127
2.046

1.956
1.843
1.701
1.606
1.486
1.455
1.376
1.312
1.243

D
D
j
D
j
j
D
D
D?
D?
j
j
j
j
j
D
Dj
j

j: a-Fe2O3 (File No. 13-534), D: BaCO3 (File No. 5-378), ?: relative

intensity much higher than reported for BaCO3.

has indicated that the substitution of Al into the barium
hexaferrite can be done only up to Fe/Al ratio as 5.
3.2. Reactions during hydrothermal treatment and
calcination
Formation of barium carbonate, boehmite and hematite
during hydrothermal treatment while using metal to urea
ratio as 2 can be expressed by the following equations:
(i) Formation of barium carbonate:
BaðNO3 Þ2 þ 2NH2 À CO À NH2 þ 3H2 O
! BaCO3 ðcÞ þ 2NH4 NO3 þ 2NH3 þ CO2

ð1Þ

(ii) Formation of boehmite
2AlðNO3 Þ3 þ 4NH2 À CO À NH2 þ ð7 þ X ÞH2 O
! 2AlOOH þ ðX À 1ÞH2 O þ 6NH4 NO3
þ 2NH3 þ 4CO2

ð2Þ

(iii) Formation of a-Fe2O3
2FeðNO3 Þ3 þ 4NH2 À CO À NH2 þ ð7 þ X ÞH2 O
! Fe2 O3 þ 6NH4 NO3 þ 2NH3 þ 4CO2 ðgÞ
þ xH2 O

ð3Þ

Fig. 3. X-ray diffraction patterns of samples: (a) S-11000, (b) S-21000, (c)

S-31000, (d) S-41000, (e) S-51000. D: BaCO3, q: a-Fe2O3, E: BaFe12O19,
z: both for a-Fe2O3 and BaFe12O19, Â : BaO. (For sample identification,
see Table 1.)


D. Mishra et al. / Materials Letters 58 (2004) 1147–1153
Table 2c
XRD data on samples calcined at 1000 jC (data corresponding to Fig. 3)
Fe/Al
ratio 5
sample
S-11000
(Fig. 3a)

Fe/Al
ratio 2
sample
S-21000
(Fig. 3b)

Fe/Al
ratio 1
sample
S-31000
(Fig. 3c)

Fe/Al
ratio 0.5
sample
S-41000

(Fig. 3d)

Fe/Al
ratio 0.2
sample
S-51000
(Fig. 3e)

d-value Phase d-value Phase d-value Phase d-value Phase d-value Phase
3.673
3.193
2.930
2.766
2.686
2.617
2.512
2.410
2.227
2.200
2.040
1.920
1.835
1.689
1.663
1.643
1.617
1.593
1.482
1.457
1.306

1.250
1.230

jD
D Â
E
E Â
z
z
z
E
E
z
D?
D Â
j
z
E Â
E
E
E Â
z
j
zD
j Â
Â

3.697
3.221
2.710

2.619
2.527
2.292
2.208
2.159
1.997
1.950
1.842
1.696
1.603
1.486
1.455
1.311
1.259

D
D
j
D
j
D
j
D
D
D
j
j
j
j
j

Dj
j

3.750
3.250
2.700
2.519
2.300
2.207
2.159
2.047
1.953
1.842
1.691
1.598
1.483
1.452
1.307
1.257
1.245

D
D
j
j
D
j
D
D
D

j
j
j
j
j
Dj
j
j

3.749
3.246
2.699
2.591
2.522
2.291
2.209
2.161
2.109
2.048
1.949
1.842
1.760
1.694
1.597
1.525
1.484
1.455
1.308

D

D
j
D
j
D
j
D
D
D
D
j
D
j
j
D
j
j
Dj

3.758
3.253
2.711
2.648
2.589
2.527
2.290
2.167
2.111
2.045
1.953

1.843
1.763
1.695
1.590
1.527
1.389
1.249

D
D
j
D
D
j
D
D
D
D
D
j
D
j
j
D
D
j

j: a-Fe2O3 (File No. 13-534), D: BaCO3 (File No. 5-378), E: BaFe6O19
(File No. 7-276), z: both a-Fe2O3 and BaFe6O19, Â : BaO (File No. 1746).

Depending on the Ba/Al/Fe ratio, formation of different
compounds take place during calcination at temperatures in
the range of 800 –1200 jC. The decomposition of BaCO3
will give BaO and CO2 and de-hydroxylation of AlOOH
will give Al2O3. These oxides would stay as such or form
aluminum-substituted barium hexaferrite as discussed in the
previous sections.
3.3. Comparison of Al-substituted barium ferrites with pure
barium hexaferrite prepared through hydrothermal precipitation – calcination technique
The major phases observed for hydrothermally prepared
precursor obtained during preparation of pure barium hexaferrite and the samples obtained after calcination at 800,
1000 and 1200 jC [9] showed similar crystalline phases as
obtained for Ba/Fe/Al ratio as 1:2:10. Formation of barium
hexaferrite at 1200 jC was observed for both these samples.
Substitution of aluminum resulted in negative shift of about
˚ in the d-values for BaFe12O19 phase. With
0.02 –0.04 A
further increase in aluminum content, formation of cubic
Fig. 4. X-ray diffraction patterns of samples: (a) S-11200, (b) S-21200, (c)
S-31200, (d) S-41200, (e) S-51200. D: BaCO3, q: a-Fe2O3, E: BaFe12O19,
z: both for a-Fe2O3 and BaFe12O19, Â : BaO. (For sample identification,
see Table 1.)

1151


1152

D. Mishra et al. / Materials Letters 58 (2004) 1147–1153

Table 2d
XRD data for samples calcined at 1200 jC (data corresponding to Fig. 4)
Fe/Al
ratio 5
sample
S-11200
(Fig. 4a)

Fe/Al
ratio 2
sample
S-21200
(Fig. 4b)

Fe/Al
ratio 1
sample
S-31200
(Fig. 4c)

Fe/Al
ratio 0.5
sample
S-41200
(Fig. 4d)

Fe/Al
ratio 0.2
sample
S-51200

(Fig. 4e)

d-value Phase d-value Phase d-value Phase d-value Phase d-value Phase
3.807
3.642
2.910
2.870
2.746
2.667
2.595
2.495
2.400
2.210
2.100
1.940
1.829
1.801
1.681
1.657
1.476
1.463
1.446
1.380
1.292
1.292

E
j
E
E

E
j
E
z
E
E
E
E
j
E
z
E
j
E
j
E
z
z

3.737
3.222
2.689
2.627
2.514
2.289
2.196
2.049
1.837
1.692
1.594

1.486
1.452
1.372
1.309
1.259

D
D
j
D
j
D
j
D
j
j
j
j
j
D
Dj
j

3.720
3.208
2.696
2.596
2.512
2.290
2.260

2.200
2.051
2.017
1.919
1.840
1.690
1.599
1.484
1.451
1.351
1.310
1.259
1.224

D
D
j
D
j
D
j
j
D
D
D
j
j
j
j
j

D
Dj
j
j

3.773
3.256
2.774
2.714
2.596
2.526
2.290
2.260
2.212
2.168
2.117
1.959
1.843
1.767
1.697
1.654
1.621
1.529
1.489
1.450
1.380
1.255
1.225

D

D
Â
j
D
j
D
j
j
D
D
D
j
D
j
D
j
D
j
j
D
j
j

3.759
3.259
2.708
2.654
2.529
2.290
2.199

2.167
2.053
1.952
1.847
1.690
1.650
1.620
1.529
1.480
1.455
1.380

D
D
j
D
j
D
j
D
D
D
j
j
D
j
D
j
j
D

j: a-Fe2O3 (File No. 13-534), D: orthorhombic BaCO3 (File No. 5-378),
E: BaFe6O19 (File No. 7-276), z: both a-Fe2O3 and BaFe6O19, Â : BaO
(File No. 1-746).

BaCO3 and a-Fe2O3 at 1200 jC. This study indicates that
the substitution of Al into the barium hexaferrite can be
done only up to Fe/Al as 5. However, the substituted
hexaferrite also had hematite as the crystalline phase. Due
to the stability of BaCO3 with higher aluminum content in
the presence of iron, formation of aluminum-substituted
barium ferrite is hindered.
3.4. VSM measurements
The XRD patterns for the BaFe12O19 and its iso-phasic
Al-substituted barium ferrite will be similar but the magnetic properties are affected by substitution of Al in the
barium hexaferrite matrix. In order to confirm the substitution of Al in barium hexaferrite matrix, VSM studies were
carried out for the sample S-11200. An increase in the
coercivity to 4.28 kOe from 2.87 kOe was observed for
the Al-substituted barium ferrite and decrease in the saturation magnetization and remenance magnetization from 40.0
to 16.69 emu/g and from 21.0 to 9.45 emu/g, respectively,
were observed. The increase in coercivity and decrease in
the magnetizations with Al substitution is in agreement with
the observation made for Al-substituted barium hexaferrite
prepared by glass-ceramic method [7]. This observation

confirms the formation of Al substitution in barium ferrite
in the present study.

4. Conclusions
The observations from the XRD and VSM studies are

summarized below:
(i) Hydrothermally prepared precursors obtained for different Ba/Al/Fe ratios showed barium carbonate,
hematite and boehmite as the crystalline phases with
the exception of the compound obtained for Ba/Al/Fe
ratio as 1:2:10 for which boehmite was not identified.
(ii) Irrespective of the Al content none of the 1000 and
1200 jC calcined samples showed any crystalline
phase of Al. The 1200 jC calcined samples showed
that Al-substituted barium ferrite is formed only with
the Ba/Al/Fe atomic ratio of 1:2:10. Increasing the Al
content either decomposed the ferrite matrix or did not
allow its formation.
(iii) Intermediate crystalline compounds of Ba, Al and Fe
such as barium aluminates, spinel ferrite of Ba
(BaFe2O4) or aluminum ferrites (Al2Fe2O6) were not
observed.
(iv) BaCO3 was found to be present even at 1200 jC in all
the samples except for the one having Ba/Al/Fe ratio as
1:2:10 The normal decomposition temperature of
BaCO3 is between 950 and 1050 jC. It is difficult to
explain the increased stability of BaCO3, which is
perhaps responsible in hindering the formation of
aluminum-substituted barium ferrites having Fe/Al
ratio V 2.
(v) Substitution of Al in barium hexaferrite was indirectly
confirmed from the magnetic measurements.

Acknowledgements
The authors are thankful to Dr. Vibhuti N. Misra,
Director, Regional Research Laboratory for his kind

permission to publish this paper. The authors wish to thank
Dr. S.C. Das, Head, Hydrometallurgy Department. The
support provided by Dr. S.K. Date and Ms. S.D. Kulkarni,
National Chemical Laboratory, Pune during XRD and VSM
measurements is gratefully acknowledged. We also express
our thanks to Ms. Mamta Mohapatra, SRF, RRL(B) for her
help during preparation of final manuscript.

References
[1] Q.A. Pankhurst, D.H. Jones, A.H. Morrish, X.Z. Zhou, A.R. Corradi, in: C.M. Srivastva, H.J. Patni (Eds.), Proceedings of the International Conference on Ferrites, Held at Bombay, India, 10 – 13
January 1989, Trans. Tech. Publishers, Aerdermann sdorf, Switzerland, 1990, p. 303.


D. Mishra et al. / Materials Letters 58 (2004) 1147–1153
[2] R. Carey, P.A. Engo Sandoval, D.M. Newman, B.W.J. Thomas, J.
Appl. Phys. 75 (1994) 6789.
[3] A. Gruskaova, IEEE Trans. Magn. 75 (1994) 639.
[4] H. Kohma, in: E.P. Wohlfarth (Ed.), Ferromagnetic Materials: A
Handbook on the Properties of Magnetically Ordered Substances,
vol. 1, North Holland, Amsterdam, 1982, p. 305, Chapter 5.
[5] G.F. Dianne, J. Appl. Phys. 40 (1969) 431.
[6] Y.M. Yokoviev, E.W. Rubalkoya, N. Lapovak, Sov. Phys., Solid State
10 (1969) 2302.
[7] T. Suzuki, K. Kani, K. Warri, S. Kawakami, Y. Torii, J. Mater. Sci.
Lett. 11 (1992) 83 – 895.

1153

[8] K. Hameda, H. Kojima, Jpn. J. Appl. Phys. 12 (1973) 355.
[9] D. Mishra, PhD thesis, Berhampur University, Berhampur, India,

2003.
[10] Joint Committee On Powder Diffraction Standard (JCPDS), International Center For Diffraction Data, Swathmore, PA, File Nos. (a) 5 –
378, (b) 13 – 534 and (c) 21 – 1307 (d), 1 – 746 and (e) 7 – 276.
[11] D. Mishra, S. Anand, R.K. Panda, R.P. Das, J. Am. Ceram. Soc. 85
(2) (2002) 437 – 443.


Materials Science and Engineering B99 (2003) 270 Á/273
www.elsevier.com/locate/mseb

High-frequency magnetic properties of low-temperature sintered
Co à Ti substituted barium ferrites
/

Changsheng Wang *, Xiwei Qi, Longtu Li, Ji Zhou, Xiaohui Wang, Zhenxing Yue
State Key Lab of New Ceramics and Fine Processing, Department of Materials Science and Engineering, Tsinghua University, Beijing 100084,
People’s Republic of China
Received 14 June 2002; received in revised form 2 September 2002

Abstract
Co Ã/Ti substituted barium ferrites BaFe12(2x Cox Tix O19 were prepared by a modified chemical coprecipitation method and lowtemperature sintering with Bi2O3 doping. Effects of Co Ã/Ti substitution amount and the barium ferrite particles fabrication
temperature on high-frequency magnetic properties were investigated. The initial permeability of sintered barium ferrites could be
promoted effectively with Bi2O3 doping. Higher temperature for barium ferrite particles fabrication resulted in higher permeability
for sintered ferrites. Barium ferrites with initial permeability m ? /10, magnetic resonant frequency /1 GHz were obtained when
they were sintered at 900 8C with 2 wt.% Bi2O3 doping in the CoÃ/Ti substitution range of 1.20 Á/1.30. The effect of Bi2O3 doping on
the magnetizing process of sintered ferrites was also discussed.
# 2002 Elsevier Science B.V. All rights reserved.
Keywords: CoÃ/Ti substituted barium ferrites; Modified chemical coprecipitation; Low-temperature sintering; High-frequency magnetic properties

1. Introduction

With the development of surface-mounting devices
(SMDs) in the higher frequency range, materials with
high permeability are needed for application in multilayer chip inductors (MLCIs) or multi-layer chip beads
(MLCBs) in frequency region of 300 Á/1000 MHz. Many
works have been done on hexaferrites that could be used
in higher frequency region other than spinel ferrites [1 Á/
6]. Hexaferrites are classified into five main types (M,
W, Y, Z and X) depending on the chemical formula and
crystal structure. Y and Z types have been investigated
widely for high frequency applications [4,5]. However,
few investigations have been made on M-type, especially
those sintered at temperatures below the melting point
(961 8C) of silver. Silver is usually used as the internal
electrode material due to its high conductivity and low
cost, so the ferrite powder needs to be sintered below
950 8C [7]. M-type barium ferrite denoted by BaFe12O19

* Corresponding author. Tel.: '/86-10-6278-4579; fax: '/86-106278-1346.
E-mail address: (C. Wang).

has strong uniaxial magnetocrystalline anisotropy.
However, the strong uniaxial magnetocrystalline anisotropy can be reduced by substitution of Fe3' cations,
which distribute on five distinct crystallographic sites.
The mostly studied example is the substitution of Fe3'
by Co2' and Ti4', which undergoes a magnetic
anisotropy change from uniaxial to planar when the
Co Ã/Ti substitution amount is 0.9 Á/1.20 [8]. Complex
permeability of Co Ã/Ti substituted barium ferrites have
been investigated by Autissier et al. using classical
ceramics method with sintering temperature 1230Á/

1290 8C [5]. In this paper, high-frequency magnetic
properties of Co Ã/Ti substituted barium ferrites prepared by modified chemical coprecipitation and lowtemperature sintering with Bi2O3 doping are presented.

2. Experiment
Co Ã/Ti substituted barium ferrite BaFe12(2x Cox Tix O19 particles were prepared by a modified
chemical coprecipitation method which combines the
chemical coprecipitation process and the synthesis from
salt melts [9]. An aqueous solution of the metal

0921-5107/02/$ - see front matter # 2002 Elsevier Science B.V. All rights reserved.
PII: S 0 9 2 1 - 5 1 0 7 ( 0 2 ) 0 0 5 2 1 - 4


C. Wang et al. / Materials Science and Engineering B99 (2003) 270 Á/273

chlorides containing Ba2', Fe3', Co2' and Ti4' in the
ratio required for the ferrite was stirred into an aqueous
solution of NaOH and Na2CO3. A suspension containing intermediate precipitates was formed during mixing.
The product of coprecipitation was filtered off, washed
thoroughly, dried and mixed with NaCl. When heated at
810 Á/950 8C, ferrite particles crystallized from the NaCl
matrix. When the salts were dissolved in water and
eliminated
by
washing
several
times,
BaFe12(2x Cox Tix O19 particles were obtained. The ferrite
particles were then mixed with appropriate amount of
Bi2O3. For electromagnetic properties measurement,

pellets (10 mm diameter, 0.7 Á/1.0 mm thickness) and
toroidal samples (20 mm outside diameter, 10 mm inside
diameter, about 1Á/3 mm thickness) were pressed and
then sintered at 900 8C in air.
The identification of the crystalline phases for ferrite
particles was carried out by XRD. Microstructures of
sintered ferrites were observed by SEM. The highfrequency magnetic properties were measured using a
HP4291B RF impedance/materials analyzer from 10
MHz to 1.8 GHz. The electrical resistivity of all ferrite
samples was determined with a HP4140B meter.

3. Results and discussion
The XRD patterns of particle samples with Co Ã/Ti
substitution amount x 0/1.20 obtained by chemical
coprecipitation process and synthesis from NaCl melts
at different temperatures are shown in Fig. 1. As can be
seen, the only crystalline phase that can be detected by
XRD is M-type hexaferrite for all particle samples, no
other phases were apparently detectable. The temperature of 810 8C, which is slightly higher than the melting
point (801 8C) of NaCl, is much lower than (above

Fig. 1. The XRD patterns of BaFe9.6Co1.2Ti1.2O19 particles fabricated
at different temperatures.

271

1000 8C) what is needed for complete reaction in
classical ceramics method.
Fig. 2 shows the SEM micrographs of ceramic
samples sintered at 900 8C for 5 h with 2 wt.% Bi2O3

doping (a) and without Bi2O3 (b), with same substitution x0/1.20. As can be seen, the microstructure of
sample sintered with Bi2O3 doping (a) is very different
from that without Bi2O3 (b), although both of the two
sintered samples were based on same ferrite particles
obtained at 950 8C for 2 h in NaCl matrix. When
sintered without Bi2O3, sample (b) is composed of small
grains with grain sizes of 100 Á/300 nm with little
porosity, indicating little change of particle size compared with the original barium ferrite particles. While
the sample (a) is composed of much larger grains with a
wide grain size distribution of 2 Á/7 mm. Large pores were
also formed due to the irregular alignment of larger
hexagonal platelets.
Fig. 3 shows the effect of particles fabrication
temperature on high-frequency magnetic properties of
Co Ã/Ti substituted barium ferrites sintered at 900 8C for
5 h with Co Ã/Ti substitution x0/1.20 and Bi2O3 content
of 2 wt.%. Obviously, the magnetic properties were
improved significantly by using the barium particles
fabricated at higher temperature. The value of initial
permeability m? has risen from 8 (for 810 8C) to 12.5
(for 950 8C). The frequency where m? has the maximum
value lowers when the particle fabrication temperature
increases. The permeability mƒ has little change in the
frequency region below 100 MHz, but has a larger value
corresponding to larger magnetic loss in the relatively
higher frequency region of 300 Á/1000 MHz with increasing particle fabricating temperature.
The Co Ã/Ti substitution dependence of low-temperature sintered samples is given in Fig. 4. Samples were
sintered at 900 8C for 5 h with a Bi2O3 content of 2
wt.%, and barium particles used were fabricated at
950 8C for 2 h. It shows that magnetic resonant

frequency has been greatly affected by the Co Ã/Ti
substitution. With the increase of Co Ã/Ti substitution
amount, the permeability m? decreases and the magnetic
resonant frequency increases. Barium ferrites with high
initial permeability m? /10 and magnetic resonant
frequency /1 GHz could be obtained when the Co Ã/
Ti substitution amount was x0/1.20 Á/1.30.
The permeability of polycrystalline ferrites can be
described as the superposition of two different magnetizing mechanisms: spin rotation and domain wall
motion [10]. In case of ferrites composed of small grains,
with the relatively larger volume fraction of grain
boundary and defects, where domain wall pinning could
occur, would decrease the contribution of domain wall
motion. On the contrary, since the melting temperature
of Bi2O3 is 825 8C, when samples were sintered with
Bi2O3 doping, higher density could be obtained through
particle re-arrangement and solution re-precipitation


272

C. Wang et al. / Materials Science and Engineering B99 (2003) 270 Á/273

Fig. 2. SEM micrographs of BaFe9.6Co1.2Ti1.2O19 ferrites sintered at 900 8C for 5 h with 2 wt.% Bi2O3 doping (a) and without Bi2O3 (b).

magnetic properties was mainly due to the change of
magnetocrystalline anisotropy field, which is in proportion to the magnetic resonant frequency. In addition, the
dc electrical resistivity for all sintered samples was
measured using silver contacts, and was found to be
above 108 V cm.

4. Conclusions

Fig. 3. Effect of magnetic particles fabrication temperature on the
complex permeability spectra of sintered samples (x0/1.20).

BaFe12(2x Cox Tix O19 ferrites can be prepared by a
modified chemical coprecipitation process and lowtemperature sintering with Bi2O3 doping. High-frequency magnetic properties could be improved obviously with Bi2O3 doping. Higher temperature for
barium ferrite particles fabrication resulted in higher
permeability for sintered barium ferrites. Barium ferrites
with permeability m ?/10, magnetic resonant frequency
/1 GHz were obtained when it was sintered at 900 8C
with 2 wt.% Bi2O3 doping in the Co Ã/Ti substitution
range of 1.20 Á/1.30. The high-frequency magnetic behaviors of Co Ã/Ti substituted barium ferrites has the
potential to be used for MLCBs in the high frequency
range.

Acknowledgements

Fig. 4. CoÃ/Ti substitution dependence of complex permeability of
barium ferrites sintered at 900 8C/5 h with 2 wt.% Bi2O3 doping.

processes by formation of the Bi2O3 liquid phase [11].
An increase in sintered density of ferrites not only causes
the reduction of demagnetizing field due to existence of
pores but also raises the spin rotational contribution,
which in turn increases the permeability. Also, as the
grain size increased with the addition of Bi2O3, the
multidomain grains appeared which could result in
higher permeability values due to domain wall motions.

The Co Ã/Ti substitution dependence of high-frequency

This work was supported by the High Technology
Research and Development Project of People’s Republic
of China (Grant No. 2001AA320502).

References
[1] H.M. Sung, C.J. Chen, W.S. Ko, H.C. Lin, IEEE Trans. Magn.
30 (1994) 4906.
[2] H.G. Zhang, L.T. Li, Z.W. Ma, J. Zhou, Z.X. Yue, Z.L. Gui, J.
Magn. Magn. Mater. 218 (2000) 67.
[3] D. Autissier, A. Podembski, C. Jacquiod, J. Phys. IV 7 (1997) 409.
[4] I.G. Chen, S.H. Hsu, Y.H. Chang, J. Appl. Phys. 87 (2000) 6247.
[5] P. Allegri, D. Autissier, T. Taffary, J. Phys. IV France 8 (1998)
Pr2-397.


C. Wang et al. / Materials Science and Engineering B99 (2003) 270 Á/273
[6] R.C. Pullar, S.G. Appleton, A.K. Bhattacharya, J. Mater. Sci.
Lett. 17 (1998) 973.
[7] K. Hirota, T. Aoyama, S. Enomoto, M. Yoshinaka, O. Yamaguchi, J. Magn. Magn. Mater. 205 (1999) 283.
[8] D.J. de Bitetto, J. Appl. Phys. 35 (1964) 3482.

273

[9] C.S. Wang, F.L. Wei, M. Lu, D.H. Han, Z. Yang, J. Magn.
Magn. Mater. 183 (1998) 241.
[10] J.P. Bouchaud, P.G. Zerah, J. Appl. Phys. 67 (1990) 5512.
[11] S.F. Wang, Y.R. Wang, T.C.K. Yang, C.F. Chen, C.A. Lu, C.Y.
Huang, J. Magn. Magn. Mater. 220 (2000) 129.

December 2002

Materials Letters 57 (2002) 868 – 872
www.elsevier.com/locate/matlet

Preparation and magnetic properties of Zn–Ti subtituted
Ba-ferrite powders
G. Mendoza-Sua´rez a,*, L.P. Rivas-Va´zquez a, A.F. Fuentes a, J.I. Escalante-Garcı´a a,
O.E. Ayala-Valenzuela b, E. Valde´z c
a
Cinvestav-Saltillo, Carr. Saltillo-Mty. Km. 13. Ramos Arizpe, P.O. Box 663, 25000 Saltillo, Coah. Mexico
CIMAV, Ceramics. Miguel de Cervantes #120, Complejo Industrial Chihuahua, Chihuahua Chih. 31109, Mexico
c
Institute Tec. de Saltillo, Departimento de Metal Meca´nica. Blvd. V. Carranza # 2400, Saltillo Coah. 25280, Mexico
b

Received 26 April 2002; accepted 29 April 2002

Abstract
Zn – Ti-substituted barium ferrite powders were prepared by the sol – gel method and characterized. The room temperature
magnetic properties were evaluated for different levels of dopant, Fe/Ba ratios, and heat-treatment temperatures. The results
showed that Ms was not very much influenced by the substitution level up to 0.6 at.%. Contrarily, Hci did show a marked decrease,
owing to the reduction of the magnetocrystalline anisotropy of the BaM phase. On the other hand, an excess of Ba led to the decrease
of Ms and the increase of Hci. Additionally, Zn – Ti substitutions were effective in decreasing crystallite sizes below 100 nm.
D 2002 Elsevier Science B.V. All rights reserved.
Keywords: Ba-ferrites; Substitutions; Zinc; Titanium; Magnetic properties; Powder technology

1. Introduction

Ever since their discovery as suitable materials for
applications in high density magnetic recording, Baferrites with magnetoplumbite structure (BaM) have
been intensively studied because of their potential use
in the form of thin-film or particulate recording media.
Many investigations have reported the magnetic properties and microstructure of pure and substituted Baferrites. The substitutions aim for the development of
new materials with improved properties and character*
Corresponding author. Tel.: +52-844-4389600; fax: +52-8444389610.
E-mail address:
(G. Mendoza-Sua´rez).

istics. The main issue concerning the substitution of
the Ba2 + and/or the Fe3 + ions with mixtures of
paramagnetic and diamagnetic ions in the magnetoplumbite structure is to reduce the high magnetocrystalline anisotropy of the material [1]. Besides the
prime importance of the anisotropy field, also significant are the decrease of the crystallite size below 100
nm, the increase of the aspect ratio (diameter/thickness), and the enhancement of the saturation magnetisation. Also important is to decrease the switching
field distribution and the temperature coefficient of
the coercivity [2].
The Zn – Ti substitution has been previously
studied and researchers [3,4] have agreed that this
cationic combination yields materials with suitable
characteristics for magnetic recording [5]. Particularly,

0167-577X/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved.
PII: S 0 1 6 7 - 5 7 7 X ( 0 2 ) 0 0 8 8 7 - X


G. Mendoza-Sua´rez et al. / Materials Letters 57 (2002) 868–872

869

Zn2 + ions have been reported to have a positive
influence on the saturation magnetisation [1]. As
pointed out by Rane et al. [6], the magnetic properties
of substituted BaM directly depend on the electronic
configuration of the substituting cations and on their
preference to occupy the different Fe sublattices of the
hexagonal structure. It has also been highlighted that
the magnetic properties of BaM are related to the
occupancy of the Fe3 + ions at the different sublattices
of the magnetoplumbite structure [7]. The influence of
the Fe/Ba ratio on the magnetic properties of pure
BaM has been investigated by Surig et al. [8]. They
found that a Fe/Ba value of 10.5 yielded the best

Fig. 2. Saturation magnetisation and coercivity of BaFen À 2xZnx
TixO19 samples heat-treated from 900 to 1000 jC for 4 h
(10 V n V 11.6).

magnetic properties due to the absence of intermediate
phases.
This paper reports on the room temperature magnetic properties of Zn – Ti-substituted barium ferrites
(BaMZn – Ti) prepared by sol – gel, emphasising the
influence of the Fe/Ba ratio, which has been little
studied in substituted ferrites.

2. Experimental

Fig. 1. Saturation magnetisation and coercivity of BaFe11.6 À 2xZnx
TixO19 samples heat-treated from 900 to 1000 jC for 4 h.

Ba-ferrite powders of composition BaFen À 2xZnxTixO19 (where 10 V n V 11.6 and = Fe/Ba) were prepared by a sol –gel method. The synthesis was carried
out using Fe(NO3)3Á9H2O, ZnCl2, TiCl4 and BaCO3
as starting materials. The details on the preparation