You are on page 1of 120

Journal of Magnetism and Magnetic Materials 363 (2014) 2125

Contents lists available at ScienceDirect

Journal of Magnetism and Magnetic Materials


journal homepage: www.elsevier.com/locate/jmmm

Structural evolution and magnetic properties of nanocrystalline


magnesiumzinc soft ferrites synthesized by glycinenitrate
combustion process
S. Hajarpour a, A. Honarbakhsh Raouf a,n, Kh. Gheisari b
a
b

Department of Materials, Faculty of Engineering, Semnan University, Semnan, Iran


Department of Materials Science and Engineering, Faculty of Engineering, Shahid Chamran University, Ahvaz, Iran

art ic l e i nf o

a b s t r a c t

Article history:
Received 12 August 2013
Received in revised form
21 February 2014
Available online 26 March 2014

In this study, nanocrystalline Mg1  xZnxFe2O4 soft magnetic ferrites are synthesized by varying x from
0.0 to 0.6 with a step size of 0.1. A new combustion synthesis approach is taken using glycine as fuel and
metal (Fe, Mg and Zn) nitrates as reactants. The effect of varying chemical composition, i.e. changing the
parameter x, on the structural and magnetic properties is evaluated. X-ray diffraction results conrm
that all samples crystallize in a spinel-type structure. Lattice parameter (a) is found to increase with the
substitution of Zn2 ions. The eld emission scanning electron microscopy (FESEM) is used for
morphological investigations. Magnetic properties of Mg1  xZnxFe2O4 ferrites are also evaluated by a
vibrating sample magnetometer (VSM). It is found that the saturation magnetization increases as the Zn
content goes up to x 0.4 and decreases afterwards. The change in saturation magnetization with Zn
content is attributed to the variation of cation distribution in the spinel structure as chemical
composition is modied.
& 2014 Elsevier B.V. All rights reserved.

Keywords:
Glycinenitrate process
Combustion synthesis
MgZn ferrite
Magnetic property
Nanocrystalline

1. Introduction
Cubic spinel ferrites are ferrimagnetic materials, made by a
close-packed array of oxygen atoms with tetrahedral and octahedral cavities, containing iron oxide as the main constituent with
various metal oxides [1,2]. They constitute an important class of
magnetic materials, which have attracted much interest in many
elds. This can be attributed to the unique and novel properties
and also wide variety of potential applications of these materials
in high-density magnetic recording, color imaging, microwave
devices, bio-processing, magnetic refrigeration, magneto-optical
devices, hyperthermia, information storage systems, gas sensors,
magnetic recording media, electronic industries, humidity sensors
and green anode materials, to name a few [28]. They have
recently been the subject of much research with regard to both
their theoretical foundations and the possibility of deriving new
materials based on ferrites for further applications [18].
Ferrites have the general formula (MFe1  )[M1  Fe1 ]O4,
where M is a divalent metal ion (e.g. Zn, Mg, Mn, Fe, Co, Ni or a
mixture of them). The magnetic cations in the cubic spinel lattice
can occupy A-type (divalent and tetrahedrally coordinated with
oxygen) or B-type (trivalent and octahedrally coordinated with

Corresponding author.
E-mail address: ahonarbakhsh@semnan.ac.ir (A. Honarbakhsh Raouf).

http://dx.doi.org/10.1016/j.jmmm.2014.03.027
0304-8853/& 2014 Elsevier B.V. All rights reserved.

oxygen) crystallographic positions, as denoted by round and


square brackets, respectively. For example, Mg2 and Zn2 cations
exclusively prefer octahedral and tetrahedral sites, respectively
[1,2,9].
The MgZn ferrites are well known for their remarkable soft
magnetic properties, including electrical resistivity, low coercivity
and inappreciable eddy current loss at high frequencies. They are
widely used in power transformers, microwave devices and telecommunications [6,9].
Recent studies reveal that nanoparticle ferrites show an unusual magnetic behavior. This unusual behavior is caused by
redistribution from inverse to mixed spinel type, which occurs at
nanolevel, resulting in an enhancement of magnetization as
compared to the bulk. This underlines why the research on neparticle ferrites on a nanolevel scale is of particular signicance
and attraction. It is well known that the change in temperature
and/or method of sample preparation can strongly affect the
particle size and cation distribution [2,9]. In order to form the
ne spinel-type ferrite particles, various wet and dry methods
have been devised. Wet methods include co-precipitation, reverse
micelle synthesis and solgel. Mechanical alloying and thermal
plasma are examples of dry methods [1012]. Several attempts
have also been made using auto-ignited combustion reactions to
synthesize such particles, an example of which is glycinenitrate
process (GNP). GNP is a simple, rapid, inexpensive and self-sustaining
combustion synthesis technique, which allows for preparation of

22

S. Hajarpour et al. / Journal of Magnetism and Magnetic Materials 363 (2014) 2125

ceramic materials with high surface area and compositionally homogeneous powders in the nanometer range [6,1315]. This method
often provides ferrites at relatively low temperatures (o600 1C) and
no further processing of the combustion products is required [16,17].
As mentioned earlier, the new synthesis method bears a considerable
effect on the cation distribution, and consequently, modies the
magnetic properties.
This paper is focused on the preparation of Mg1  xZnxFe2O4
ferrites (with x 0, 0.1, 0.2, 0.3, 0.4, 0.5 and 0.6) by the GNP
method. The structure and magnetic properties of the assynthesized ferrite powders are discussed as dependent on the
zinc content.

3. Results and discussion


3.1. Nature of combustion synthesis
In our previous work, we showed that the nature of combustion depends on the compositional ratio of glycine to nitrate [6].
Redox reactions are usually exothermic by nature and often lead to
explosion, if not properly controlled. The combustion of metal
nitrateglycine mixture appears to undergo a self-propagating and
non-explosive exothermic reaction [10].
To explain the changes in adiabatic ame temperature (Tad)
with respect to Zn content in the case of a combustion reaction,
the following equation is employed:
T ad T 0

H0r  H0p
cp

2. Materials and methods


2.1. Materials and preparation techniques
Nanocrystalline Mg1  xZnxFe2O4 (with x 0, 0.1, 0.2, 0.3,
0.4, 0.5 and 0.6) ferrite powders are prepared using a wet chemical
synthesis approach with magnesium nitrate (Mg(NO3)2  6H2O),
zinc nitrate (Zn(NO3)2  6H2O), iron nitrate (Fe(NO3)3  9H2O) and
glycine (H2NCH2COOH) (all 499%, Merck, Germany) as precursors. The rationale for choosing glycine as the fuel is explained in
our previous work [6].
At an appropriate ratio, accurately weighed powders of precursor materials are dissolved in 100 cm3 of distilled water to
prepare the precursor solution. The solution is prepared with
different ratios of Mg and Zn cations. Based on conclusions from
our previous work [6], a glycine-to-nitrate ratio of 2 is used, as it
will result in the maximized saturation magnetization. The total
concentration of the metal ions is set to approximately 0.1 mol/l.
The precursor is poured in a round glass and the glass is placed on
a hot plate to be heated under atmospheric conditions. The
solution is then simply allowed to boil and ignite. The ignition is
accompanied by evolution of large volumes of gases, as described
by the following chemical equation:

where T0 is 25 1C, Hr and Hp are respectively the enthalpies of


formation of the reactants and products and Cp is the heat capacity
of the products at constant pressure. Using thermodynamic data
for various reactants and products [6], the combustion enthalpy
and the theoretical adiabatic ame temperature can be calculated
as a function of Zn content. Fig. 1 shows the adiabatic ame
temperature versus Zn content. Temperature varies according to a
direct relationship with Zn content. This can be explained by the
higher enthalpy of Zn formation compared to its counterpart, i.e.
Mg. However, it should be noted that the actual ame temperatures are much lower than the theoretically calculated ones owing
to radioactive losses, incomplete combustion and heating of air
[19,20].

(1 x)Mg(NO3)2  6H2O xZn(NO3)2  6H2O 2Fe


(NO3)3  9H2O 6NH2CH2COOH 3.5O2Mg1  xZnxFe2O4 7N2 12CO2 15H2O
where x is 0, 0.1, 0.2, 0.3, 0.4, 0.5 and 0.6. In less than 40 min, the
combustion reaction produces brown-to-black, porous products
that are quite easily ground to powder.

Fig. 1. Adiabatic ame temperature as a function of the Zn content.

2.2. Characterization
Structural studies on the as-prepared powders are conducted
using X-ray diffraction (XRD, Bruker D8) with Cuk radiation.
The lattice parameter of the spinel compounds is estimated from
the position of the (440) plane. The average crystallite size (D) and
the lattice strain () of all samples are evaluated based on the Xray diffraction line broadening by employing Scherrer's equation
and the tangent formula respectively. All structural parameters are
estimated by means of a commercial software package [18]. The
microstructure evaluation and morphological features of the
product powders are determined by eld emission scanning
electron microscopy (FESEM; Hitachi S-4160) of the gold-coated
samples. The room-temperature magnetic measurements are done
with a vibrating sample magnetometer (VSM, Meghnatis Daghigh
Kavir, Iran) in an operating range of 78.5 kOe.

x = 0.6
x = 0.5
x = 0.4
x = 0.3
x = 0.2
x = 0.1
x=0

Fig. 2. XRD patterns of Mg1  xZnxFe2O4 series.

S. Hajarpour et al. / Journal of Magnetism and Magnetic Materials 363 (2014) 2125

3.2. Structural evaluation


The X-ray diffraction patterns of the as-synthesized Mg1  xZnx
Fe2O4 ferrite powders are shown in Fig. 2. The phase analysis of
XRD patterns conrms the formation of cubic spinel structures.
The reections from (220), (311), (222), (400), (422), (511) and
(440) planes, corresponding to the spinel phase, indicate the
formation of single-phase spinel structures. The lattice parameter
for all samples is determined using the classical formula presented
by the following equation:
h
i0:5
2
2
2
h k l
a
2
2
sin
where is the wavelength of Cuk radiation, h, k, and l are the
Miller indices, and is the diffraction angle corresponding to the
(hkl) plane. The lattice parameter obtained from the XRD data
(Fig. 3) is in the range 8.3838.430 . Lattice parameter increases
with an increase in the level of zinc content (x), which indicates
that the present system conforms to Vegard's law. This can be
attributed to the ionic size differences, where Mg2 ions (0.66 )
are substituted by Zn2 ions with a larger ionic size (0.82 ) [4].
To accommodate these larger cations, the lattice parameter is
increased. For high concentration of zinc (x 0.6), the slight
decrease in the lattice parameter may be attributed to the
variation of cation distribution. Similar results have been obtained
by other researchers [9,21].
The X-ray density for each composition is calculated from the
following equation:

8M
Na3

where M is the molecular weight, N is Avogadro's number and a is


the lattice parameter. Variation of with respect to zinc content is
shown in Table 1. The atomic weight and density of Mg are 24.31
and 1.74 g/cm3, respectively, while the corresponding values for Zn
are 65.37 and 7.14 g/cm3, respectively. Having a higher density,
zinc is known to cause densication, grain growth and anisotropy

Fig. 3. Variation of lattice parameter with zinc content.

Table 1
Data of crystallite size (D), lattice strain (), lattice parameter (a) and X-ray density
(dx) of Mg1  xZnxFe2O4.
X

D (nm)

a ()

dx(g/cm3)

0
0.1
0.2
0.3
0.4
0.5
0.6

46.7
34.4
49.4
49
39.7
34.3
46.1

0.00278
0.0033
0.00327
0.00273
0.00291
0.00331
0.00251

8.38
8.38
8.40
8.40
8.42
8.43
8.42

4.51
4.60
4.66
4.75
4.83
4.89
4.99

23

reduction. Consequently, the X-ray density increases signicantly


as Zn content is raised.
The average crystallite size (D) and the lattice strain () are
computed through Scherrer's formula and the tangent formula,
respectively. As listed in Table 1, crystallite size and lattice strain
range from 34.3 to 49.4 nm and from 0.0025 to 0.0032, respectively. According to the results, there is no distinguishable trend
between crystallite size and the composition, and also between
lattice strain and the composition. Therefore, the results cannot be
explained by substitution of Mg2 ions (ionic radius 0.66 ) with
slightly bigger Zn2 ions (ionic radius 0.82 ).
The morphology and particle size of MgZn ferrite powders are
evaluated by FESEM microphotographs (Fig. 4(a) and (b)). It can be
clearly observed that the particle sizes are slightly larger than
those derived from Scherrer's equation. In fact, the crystallites
(termed as primary particles) are aggregated to form relatively
larger particles (termed as secondary particles). The secondary
particles are not single-crystalline in nature. Therefore, the particle
sizes, determined by FESEM as shown in Fig. 4(b) (high magnication microphotographs corresponding to the secondary particles),
are larger than those estimated from the X-ray diffraction data [9].
Voids and holes are observable in Fig. 4(a) (low magnication).
These result from the release of gases during combustion. Porosity
enlargement caused by heightened gas evolution in GNP is also
reported by other authors [10].
3.3. Magnetic properties
Magnetization measurements for Mg1  xZnxFe2O4 series are
performed using the VSM technique. MH curves of magnesium
zinc ferrites at room temperature are shown in Fig. 5. The variation
of saturation magnetization as a function of Zn content is shown in
Fig. 6. As this gure suggests, saturation magnetization is in the
range of 3647 emu/g. This is much higher than reported values of
MgZn ferrites, synthesized by different chemical production
methods [1,3,22,23]. Ms rises as more of Zn2 is introduced, up
to x 0.4, after which Ms starts to decrease.
The saturation magnetization of the soft magnetic ferrites
directly depends on the net magnetization of the spinel lattice.
On the other hand, the net magnetization of the spinel lattice
can be explained by the interaction of the octahedral-site (B-type)
and tetrahedral-site (A-type) sublattice magnetic moments,
i.e., M MB  MA. Therefore, Ms of each component indirectly
depends on the distribution of Fe3 ions between the two
sublattices A and B, where Mg2 and Zn2 are nonmagnetic
[21,24]. The equilibrium cation distribution in MgFe2O4 is given by
the formula Fe30:9 Mg 20:1 A Mg 20:9 Fe31:1 B [25]. Since Zn2 ions
prefer to occupy A-sites, the replacement of Mg2 ions by Zn2
ions leads to the migration of Fe3 ions from A-sites to B-sites.
This substitution results in a higher quantity of Fe3 ions on
B-sites, and hence a heightened magnetization in the B-sites. At
the same time the magnetization of A-site decreases owing to the
drop in the amount of Fe3 ions. Thus, the net magnetization
of the sample increases up to x 0.4. Beyond x 0.4, the
magnetization of A-sublattice is reduced and the magnetic
moments of Fe3 ions on the A-sites become so weak that they
are unable to maintain antiparallel alignment with those on the B
sites (spin canting). As a result, AB interaction is weakened,
leading to a decrease in the saturation magnetization. A similar
behavior of canted spins has been reported elsewhere [21,23].
In order to calculate the theoretical value of saturation magnetization and compare it against the observed one, the sample
containing low concentration of zinc (i.e., x 0.1) is investigated.
For such a sample, the magnetic moment can be interpreted well
in terms of Neel's two-sublattice theory [26] and a canted
structure should not appear. Saturation magnetization in units

24

S. Hajarpour et al. / Journal of Magnetism and Magnetic Materials 363 (2014) 2125

Fig. 4. FESEM photomicrographs of MgZn ferrite: (a) low magnication and (b) high magnication.

Fig. 7. Variation of coercive force with zinc content.


Fig. 5. Room-temperature hysteresis loops of as-synthesized product with different Zn contents.

measured magnetization can be explained by several factors,


including the following [26]:
i. the effect of thermal agitation at room temperature may be
signicant on the magnetic moments;
ii. the ion distribution on various sites may not be as perfect as
assumed;
iii. the orbital magnetic contribution may not be zero as assumed;
iv. the directions of the spins may not be antiparallel in the
interactions.

Variation of coercive force (Hc) with Zn content is shown in


Fig. 7. The gradual decrease in coercive force may be indicative of
decrease in anisotropy constant of the produced powder with Zn
content.

Fig. 6. Saturation magnetization versus Zn content.

4. Conclusion
of emu/g can be theoretically calculated by the following
equation [27]:
Ms

8nB
a3

where a is the lattice parameter, is the density, n is the


theoretical number of magnetic moments per formula unit, and
mB is the magnitude of the Bohr magneton, where 1 Bohr magneton is equal to 9.274  10  24 A m2. The theoretical value of
saturation magnetization for Mg0.9Zn0.1Fe2O4 is 59 emu/g, while
the observed one is about 41 emu/g. The lower value of the

Nanocrystalline Mg1  xZnxFe2O4 ferrite powders are produced


through the glycinenitrate process. Analysis of the X-ray diffraction spectrum shows that the prepared samples have single-phase
spinel structures. Crystallite sizes, calculated from XRD data, are in
the range 3647 nm. Magnetic measurements show that antiparallel spin coupling exists when Zn content is less than 0.4. For
higher Zn contents, spin canting occurs and saturation magnetization decreases. It is concluded from the conducted work that
MgZn ferrite can be successfully produced by glycinenitrate
process with higher values of saturation magnetization, in less
time and with easier steps, when compared to other techniques.

S. Hajarpour et al. / Journal of Magnetism and Magnetic Materials 363 (2014) 2125

Acknowledgment
The authors would like to thank the Semnan University and
Shahid Chamran University for supporting this research.
References
[1] S.D. Shenoy, P.A. Joy, M.R. Anantharaman, J. Magn. Magn. Mater. 269 (2004)
217226.
[2] H. Dutta, M. Sinha, Y.C. Lee, S.K. Pradhan, Mater. Chem. Phys. 105 (2007)
3137.
[3] N. Kikukawa, M. Takemori, Y. Nagano, M. Sugasawa, S. Kobayashi, J. Magn.
Magn. Mater. 284 (2004) 206214.
[4] K.A. Mohammed, A.D. Al-Rawas, A.M. Gismelseed, A. Sellai, H.M. Widatallah,
A. Yousif, M.E. Elzain, M. Shongwe, Physica B: Condens. Matter 407 (2012)
795804.
[5] M. Al-Haj, J. Magn. Magn. Mater. 299 (2006) 435439.
[6] S. Hajarpour, Kh. Gheisari, A. Honarbakhsh Raouf, J. Magn. Magn. Mater. 329
(2013) 165169.
[7] Z. Pdzich, M.M. Buko, M. Krlikowski, M. Bakalarska, J. Babiarz, J. Eur. Ceram.
Soc. 24 (2004) 10531056.
[8] M. Sinha, H. Dutta, S.K. Pradhan, Physica E: Low-Dimens. Syst. Nanostruct. 33
(2006) 367369.
[9] D.C. Bharti, K. Mukherjee, S.B. Majumder, Mater. Chem. Phys. 120 (2010)
509517.
[10] J.C. Toniolo, M.D. Lima, A.S. Takimi, C.P. Bergmann, Mater. Res. Bull. 40 (2005)
561571.
[11] I. Shari, H. Shokrollahi, M.M. Doroodmand, R. Sa, J. Magn. Magn. Mater. 324
(2012) 18541861.

25

[12] V. Vasanthi, A. Shanmugavani, C. Sanjeeviraja, R. Kalai Selvan, J. Magn. Magn.


Mater. 324 (2012) 21002107.
[13] L.A. Chick, L.R. Pederson, G.D. Maupin, J.L. Bates, L.E. Thomas, G.J. Exarhos,
Mater.i Lett. 10 (1990) 612.
[14] H. Mohseni, H. Shokrollahi, I. Shari, Kh. Gheisari, J. Magn. Magn. Mater. 324
(2012) 37413747.
[15] K. Gheisari, S.D. Bhame, J.T. Oh, S. Javadpour, J. Supercond. Nov. Magn. 26
(2013) 477483.
[16] R.S. Ningthoujam, S.S. Umare, S.J. Sharma, R. Shukla, Sajith Kurian, R.K. Vatsa,
A.K. Tyagi, R. Tewari, G.K. Dey, N.S. Gajbhiye, Hyperne Interact. 184 (2008)
227233.
[17] K. Vasundhara, S.N. Achary, S.K Deshpande, P.D. Babu, S.S. Meena, A.K. Tyagi, J.
Appl. Phys. 113 (194101) (2013) 19.
[18] XPert HighScore Plus v1.0d/2003, PANalytical B.V., Almelo, The Netherlands.
[19] C.C. Hwang, J.S. Tsai, T.H. Huang, C.H. Peng, S.Y. Chen, J. Solid State Chem. 178
(2005) 382389.
[20] R.D. Purohit, B.P. Sharma, K.T. Pillai, A.K. Tyagi, Mater. Res. Bull. 36 (2001)
27112721.
[21] M. Manjurul Haque, M. Huq, M.A. Hakim, Physica B: Condens. Matter 404
(2009) 39153921.
[22] V.D. Kassabova-Zhetcheva, Cent. Eur. J. Chem. 7 (2009) 415422.
[23] S.S. Khot, N.S. Shinde, B.P. Ladgaonkar, B.B. Kale, S.C. Watawe, Adv. Appl. Sci.
Res. 2 (2011) 460471.
[24] A.I. Nandapure, S.B. Kondawar, P.S. Sawadh, B.I. Nandapure, Physica B:
Condens. Matter 407 (2012) 11041107.
[25] R.A. McCurrie, Ferromagnetic Materials: Structure and Properties, Academic
Press, San Diego, 1994.
[26] A. Goldman, Modern Ferrite Technology, 2nd Ed., Springer, New York, 2006.
[27] R. Valenzuela, Magnetic Ceramics, Cambridge University Press, Cambridge,
2005.

ARTICLE IN PRESS

Journal of Magnetism and Magnetic Materials 269 (2004) 217226

Effect of mechanical milling on the structural, magnetic and


dielectric properties of coprecipitated ultrane zinc ferrite
S.D. Shenoya, P.A. Joyb, M.R. Anantharamana,*
a

Department of Physics, Cochin University of Science and Technology, Thrikkakara, Cochin 682 022, India
b
Physical Chemistry Division, National Chemical Laboratory, Pune 411 008, India
Received 15 March 2003; received in revised form 23 June 2003

Abstract
Nanosized ZnFe2O4 particles containing traces of a-Fe2O3 by intent were produced by low temperature chemical
coprecipitation methods. These particles were subjected to high-energy ball milling. These were then characterised using
X-ray diffraction, magnetisation and dielectric studies. The effect of milling on zinc ferrite particles have been studied
with a view to ascertaining the anomalous behaviour of these materials in the nanoregime. X-ray diffraction and
magnetisation studies carried out show that these particles are associated with strains and it is the surface effects that
contribute to the magnetisation. Hematite percentage, probably due to decomposition of zinc ferrite, increases with
milling. Dielectric behaviour of these particles is due to interfacial polarisation as proposed by Koops. Also the defects
caused by the milling produce traps in the surface layer contributes to dielectric permittivity via spin polarised electron
tunnelling between grains. The ionic mechanism is enhanced in dielectrics with the rise in temperature which results in
the increase of dielectric permittivity with temperature.
r 2003 Published by Elsevier B.V.
PACS: 75.50.K; 75.50.G; 75.30.P; 77.22.E; 81.20.W
Keywords: Magnetic nanoparticles; Zinc ferrite; High-energy ball milling; Superparamagnetism; Spinels; Strain parameter; Dielectric;
Interfacial polarisation; Spin polarised tunnelling; Lab VIEW

1. Introduction
Nanoparticles have been generating extensive
interest in recent years among researchers worldwide owing to their interesting physical and
chemical properties with respect to their coarser
sized cousins in the bulk [15]. Nanoparticles of
many ferrites have been successfully obtained by
*Corresponding author. Tel.: +91-484-2577404; fax: +91484-2577595.
E-mail address: mra@cusat.ac.in (M.R. Anantharaman).
0304-8853/$ - see front matter r 2003 Published by Elsevier B.V.
doi:10.1016/S0304-8853(03)00596-1

numerous chemical routes, such as reverse micelle


synthesis, coprecipitation, thermal decomposition,
solgel and aerogel process [112]. High-energy ball
milling (HEBM) is being extensively employed as
an alternate route to obtain novel materials
through solid state reactions [5,6,1321].
Ferrites are ferrimagnetic materials containing
iron oxide as the main constituent with various
metal oxides. They have been extensively studied
both for understanding these systems from a
theoretical point of view and in deriving newer
materials based on ferrites for possible applications.

ARTICLE IN PRESS
218

S.D. Shenoy et al. / Journal of Magnetism and Magnetic Materials 269 (2004) 217226

Ferrites and materials derived from ferrites nd


extensive application in devices like transformers,
TV yokes, loud speakers and in a horde of other
devices. They are also one of the largely used
material medium for audio/video applications and
computer memories [6,12]. They have the general
formula {(M)d(Fe)1d}[(M)1d(Fe)1+d]O4. The
divalent metal ion M (e.g. Zn, Mg, Mn, Fe,
Co, Ni or a mixture of them) can occupy either
tetrahedral (A) or octahedral (B) sites as depicted
by curled and square brackets, respectively
[1,310,22,23]. In the above formula when d 1;
it is called normal spinel and when d 0; it is
called inverse spinel. When d 1=3; it is called a
random spinel. C 1  d represents the inversion
parameter.
In the coarser regime, the magnetic and
structural properties of these ferrites are determined by a variety of factors like madelung
energy, cationic size, charge and site preference
energy. For example cation like Ni2+ and Zn2+
have exclusive octahedral and tetrahedral site
preferences, respectively [25,7,8,10,12,24]. So in
a typical Nickel ferrite, Nickel occupies only the
octahedral (B) sites and this results in an inverse
spinel while in zinc ferrite, zinc occupies tetrahedral (A) sites exclusively and results in a normal
spinel.
There are numerous reports where in anomaly
in structural and magnetic properties of these
ferrites in the ultrane regime have been reported.
For instance, zinc ferrite which is purported to be
a normal spinel, is a classic example of an
antiferromagnetic material with a Neel temperature of 10 K. Ultrane ZnFe2O4 have been
reported to be exhibiting a net magnetic moment
at room temperature [1,3,510,14,22,25,26]. From
the application point of view, zinc ferrites are
widely used catalysts [5] and in the nano regime,
the modication of surface properties will have a
profound bearing in determining the catalytic
properties of these materials. Moreover, since they
exhibit useful magnetisation at room temperature
in the nano regime they can be potential materials
for other applications too.
Zinc ferrite if prepared in the nanoregime,
exhibits inversion and the percentage of occupancy
of Zn2+ ion depends on the method of prepara-

tion, vis-a" -vis cold preparation technique or ball


milling etc [1,3,4,7,8,15]. Results by Hamdeh et al
[1,3,8] show a relatively high inversion parameter
when coprecipitation methods are employed.
There are also reports indicating that ball milling
facilitates inversion and induces magnetic ordering
[3,8,15]. In inverse spinels like nickel ferrite,
redistribution from inverse to mixed spinel type
occurs at nanolevel resulting in an enhancement in
magnetisation as compared to the bulk. Chinnasamy et al. [22,24] reported an 8% increase in the
value of Ms for 1 h milled nickel ferrite. But
prolonged milling reduced the Ms value because of
spin-glass like surface disorder exhibited by the
particles in the ultrane regime. The theoretical
and technological lowest size of magnetically
ordered systems are still an open question of
major relevance to applications of nano structured
systems in high density magnetic recording devices
[6]. So research on ne particle zinc ferrite is of
utmost interest in understanding the behavior at
the nanolevel.
Preliminary studies carried out by Anantharaman et al. [10] on ultrane zinc ferrite by employing techniques like Low energy ion scattering,
.
Mossbauer
spectroscopy and Vibration sample
magnetometry revealed that in the nano regime,
zinc ferrite exhibits an altogether different magnetic structure with a net magnetisation at room
temperature. From the theoretical point of view
the origin of magnetism in systems similar to
ultrane ZnFe2O4 is an important question, which
is under debate now, and still no conclusive theory
has been formulated. The relevance of dead layer
theory also is to be examined on systems belonging
to the family of inverse spinels like nickel ferrite in
the nano regime where a dead layer formation in
the surface have been invoked to account for the
reduced magnetisation exhibited by the ne
particles. This aspect and the unusual behaviour
of ZnFe2O4 nanoparticles coupled with the unusual claim of zinc ions occupying a substantial
number of octahedral sites motivated this work
[11]. It has also been reported that zinc ferrite
decomposes to a-Fe2O3 and magnetic ZnFe2O4
during continued milling. That was on particles
prepared by ceramic technique and subjected to
HEBM subsequently. In the present study

ARTICLE IN PRESS
S.D. Shenoy et al. / Journal of Magnetism and Magnetic Materials 269 (2004) 217226

ZnFe2O4 is synthesized by coprecipitation methods with traces of a-Fe2O3 and these are subjected
to HEBM.
The dielectric properties of nano particle ferrite
materials are inuenced mainly by the method of
preparation, cation distribution, grain size, sintering temperature, oxygen parameter, the ratio of
Fe2+/Fe3+ ions and oxygen anion vacancies in
lattices [27]. So far, there exists no convincing
reports on dielectric properties of nano particle
ferrites which throws light on the conduction
mechanism of these materials in the ultrane
regime.
The effect of ball milling on the structural,
magnetic and dielectric properties of the ne
particles is investigated. This is intended to check
the claim by various researchers that ZnFe2O4
shows anomalous behaviour when prepared in the
nanoregime by employing coprecipitation methods. Moreover, this is a sequel to our investigations on ne particle normal spinel ferrites.

219

different milling times30 and 300 min. The


samples were labelled as ZF30 and ZF300,
respectively.
2.3. Structural studies
The structural studies on these samples
were carried out using the method of X-ray
powder diffraction. The powder diffractogram
was recorded on a Rigaku D-MaxC model
X-ray diffractometer using Cu Ka radiation
( in the 2y range 1080 with a step
(l 1:5418 A)

size of 0.05 and a speed of 2 /min. The lattice
constants of the spinel compounds were estimated
from the X-ray diffraction peak positions. The
average crystallite size D was obtained by employing Scherrers formula from the line broadening of
the XRD peak corresponding to the (3 1 1) plane
of the spinel structure and also by strain calculations. The phases present were identied by
comparing the peak positions and intensities with
those listed in the JCPDS le.

2. Experimental

2.4. Magnetisation studies

2.1. Preparation of ZnFe2O4 by coprecipitation

The room temperature and low temperature


(liquid nitrogen) magnetic characterization were
carried out using a vibrating sample magnetometer
(Model:EG&G PAR 4500). Magnetization curves
at 300 K and 83 K were recorded for ZFP, ZF30
and ZF300. Saturation magnetisation (Ms ) and
Coercivity (Hc ) were obtained from these measurements.

Zinc ferrite was prepared by low temperature


preparative techniques as described by Sato et al
[9,28]. Aqueous solution of zinc nitrate of 0.1 m
and ferric nitrate of 0.2 m were prepared separately. Thousand millilitres of each solution was
mixed together. While stirring the mixture, 25%
ammonia solution was added until pH attains 11
at 373 K. The precipitate was dried at 373 K and
annealed in air at 573 K which yielded zinc ferrite
[10]. This sample is labelled as ZFP.
2.2. Mechanical treatment using HEBM
The milling of original ZnFe2O4 prepared as
above was carried out in a closed container using a
planetary ball mill (Fritsch pulverisette 7). A
portion of the powder was placed in a tungsten
carbide vial with 8:1 ball to powder weight ratio
keeping milling intensity at 360 rpm. Tungsten
carbide balls and toluene as liquid carrier was
employed for HEBM. Powders were milled at two

2.5. Dielectric studies


The dielectric studies were carried out using
a homebuilt dielectric cell and an HP 4285A
impedance analyser in the frequency range
100 KHz to 10 MHz. The details of the set up are
cited elsewhere [27]. The temperature could be
varied from room temperature to 393 K by a
digital temperature controller. Samples in the form
of pellets were employed for the evaluation of
dieletric permittivity (er ). If c is the capacitance of
the sample, d its thickness and A the surface area,
then, er cd=e0 A where e0 is the dielectric
constant of the air. All the measurements were

ARTICLE IN PRESS
S.D. Shenoy et al. / Journal of Magnetism and Magnetic Materials 269 (2004) 217226

3. Results and discussion


3.1. Structural
XRD pattern of the samples is shown in the
Fig. 1. XRD pattern indicates the presence of
small amounts of a-Fe2O3 in all the three samples
whose percentage rst decreases and then increases
with milling time.
Fig. 2 shows the relative intensities (I=I0  100)
of (1 2 1), (2 3 1) and (2 1 1) planes which are the
three strongest lines of a-Fe2O3 with respect to the
intensity of most prominent (3 1 1) plane of
ZnFe2O4. During the initial stage of milling, some
unreacted ferric and zinc salts and the a-Fe2O3
present may react to form ZnFe2O4 and thereby aFe2O3 line gets less intense during the rst 30 min
of milling. But during prolonged milling, due to
nanometer sized grains, high local pressures and
temperatures during collisions with balls and vial,
and the subsequent kinetic energy produced will
lead to a decomposition of ZnFe2O4 in the milling
process [13,21,29]. There are reports of production
of Fe3O4 (by product of a-Fe2O3) as a result of

60
Relative Intensity (I/I0*100)

carried out under high vacuum. The impedance


analyser was interfaced via a GPIB card to a PC
and the data acquisition was completely automated by employing the LabVIEW software
(National Instruments).

(121)

50

(231)
(-211)

40

30

20
0

100
200
Milling time (minutes)

300

Fig. 2. Intensity variation of prominent a-Fe2O3 lines with


milling time.

8.46

a (A)

220

8.44

8.42

8.40
0

100
200
Milling time (minutes)

300

440(ZF)
-211( )

231( )
511(ZF)

Intensity (arb.unit)

400(ZF)

220(ZF)
121( )
311(ZF)

Fig. 3. Lattice parameter variation with milling time.

ZF300

ZF30

ZFP

20

60

40

80

100

Fig. 1. XRD pattern of ZFP ZF30 ZF300 showing a-Fe2O3(a)


impurity lines and ZnFe2O4(ZF) lines.

prolonged milling [30]. But in our present study no


traces of Fe3O4 were detected by XRD.
Lattice parameter of coprecipitated unmilled
( which is higher than
sample is found to be 8.46 A
(
the standard value reported in JCPDS of 8.44 A
(Fig. 3). This is in agreement with the results
obtained by other researchers [3,8,9,13]. Lattice
parameter is found to be decreasing with milling.
The probable reason for this lattice parameter
reduction may be the inversion of cations taking
place with milling. There are also reports suggesting that as inversion increases lattice parameter
decreases [28]. Since Zn2+ has a large ionic radius
than Fe3+, this site reversal will cause a decrease

ARTICLE IN PRESS
S.D. Shenoy et al. / Journal of Magnetism and Magnetic Materials 269 (2004) 217226

10.0

Particle size (nm)

of lattice constant [29]. This effect has not been


observed in sintered (ceramic) ferrite. The other
probable reason for this unique property of
coprecipitated sample may be the presence of
lattice defects and its inuence on ultrane
particles, especially on the surface. In ultrane
particles surface energy and surface tension of
particles are high. This results in a tendency to
shrink the lattice which causes reduced lattice
constant [28]. It is also possible that the cation
distributions do differ between those present near
the particles surface and those of non-surface
atoms. This situation is highly probable for the
materials with a large number of defects in which
O2 ions are missing from normally occupied sites
and/or in which Fe3+ occupies normally unoccupied sites. In such instances, however, it is clear
that the crystal chemical structure is not those
appropriate to spinel ferrites [11]. This disordered
structure can also be attributed to the strong
internal strains introduced during mechanical
treatment [4,31,32].
It is known that the inversion if any, in these
systems can be detected by analysing the intensities
ratios of 220/400 lines. The intensity ratio corresponding to these planes have been calculated. If
the samples produced have an inverse component
in it (i.e. that is the reversal of cations) then 220/
400 lines must show a decreasing intensity ratio
from that of normal one. For normal spinels the
intensity ratio should be 35/17. Our result is in
agreement with cation distributions corresponding
to the normal spinel and also exhibits no change
with milling. So inversion, if any, cannot be
deduced from the XRD results. However XRD is
not an advisable tool to deal with cation occupancy, hence neutron scattering is recommended.
Size of the particle decreases as the milling time
increases because the kinetic energy generated by
the series of collisions among balls is transferred to
the system [33] but there is no sudden decrease as
in the case of ceramic powders (Fig. 4). This is
because the starting powder (coprecipitated) itself
is of nanometer sized. Also there are reports that
within 24 min of milling, size reduction may attain
its maximum [13]. The other reason for not
obtaining a great reduction in size is due to the
high local temperature and pressure generated

221

9.5

9.0

100
200
Milling time (minutes)

300

Fig. 4. Particle size reduction with milling.

during the combustion as a result of the highenergy ball milling [34].


The grain size and the strain developed during
the milling process can be calculated from the
linewidth (FWHM in radian) of XRD lines as
follows [35]:
FWHMb l=dg cos y 4e tan y;

where y is the bragg angle, e is the strain developed


and dg the grain size.
Rewriting the above equation
b cos y e4 sin y l=dg :

This equation represents a straight line between


b cos y (Y-axis) and 4sin y (X-axis). The slope of
this line gives the strain and dg can be calculated
from the intercept of this line on the Y-axis. The
particle size when calculated by employing
Scherrers formula at different y values was at
variance. This is an indication of the presence of
strain.
Fig. 5 shows 4sin y vs. b cos y lines for ZFP,
ZF30 and ZF300. The grain sizes calculated from
this graphs almost match with the one calculated
by Scherer formula for the most intense line.
Strain values are also calculated and found to
be increasing with milling time and are plotted in
Fig. 6. Strain continues to increase with an
increase in the milling time.

ARTICLE IN PRESS
S.D. Shenoy et al. / Journal of Magnetism and Magnetic Materials 269 (2004) 217226

222

effects, which include surface energy and surface


tension as well as the change in position of anions
and cations, will favour an increase of magnetisation, although hematite percentage may increase.
Furthermore, hematite may be acting as a facilitator in producing magnetic zinc ferrite, which is
magnetic due to the strain and surface effects. This
can be written in the following form:

0.03
ZF300

cos

0.02
ZF30
ZFP

0.01

a-Fe2 O3 ZnFe2 O4
-a-Fe2 O3 magnetic ZnFe2 O4 :
1.0

1.5
4 sin

2.0

Fig. 5. 4 sin yb cos y graphs indicating the presence of strain.

0.015

Strain

0.010

0.005

0.000
0

100
200
Milling time (minutes)

300

Fig. 6. Milling time dependence on the strain which increases


with milling time.

3.2. Magnetisation
Fig. 7 shows typical results for magnetisation
curves taken at 300 K and 83 K for ZFP, ZF30,
ZF300, respectively. Room temperature curves
show typical superparamagnetic behaviour with
no hysteresis. That is, both retentivity and
coercivity are zero, which is in tune with the
results obtained by other researchers [9,29,3638].
Fig. 8 depicts the variation of saturation
magnetisation with milling time. The strain associated with ne particles, as a result of milling,
produces a disorder by displacement of ions, which
is one cause for an increase in magnetisation. In
ultrane particles large fractions of atoms are
located on the surface and therefore the surface

From Fig. 6 it is evident that the strain increases


tetra fold for ZF300 with respect to ZF30. But
from Fig. 8, one does not observe a corresponding
increase in magnetisation for 300 min milled
sample. Though strain produces disorder and
thereby contributes to overall magnetisation
(could be because of surface spins) there is another
possibility. Since we are seeing Ms in emu/gm, the
overall increase in magnetisation may be masked
by the increased amount of a-Fe2O3 which
increases with milling as evidenced by XRD.
The hysteresis behaviour completely vanishes at
room temperature. Short-range magnetic ordering
of Fe spins at 83 K contributes to magnetic
hysteresis loops. Long range ordering is expected
at liquid helium temperature. These powders
exhibit a net magnetisation and are magnetically
ordered at much higher temperature than that of
normal zinc ferrite. It may be noted here that for
mechanically milled sample, TN has been reported
to be B115 K [14]. The magnetic ordering starts
with the formation of superparamagnetic clusters
whose size gradually increases with lowering
temperature, superparamagnetic clusters present
are heterogeneous at room temperature but get
ordered at lower temperature to contribute to
magnetisation [1].
Superparamagnetic relaxation time t; which is
the average time required for the particle to jump
from one axis of magnetisation to another is given
by the relation


t to exp Keff V =kB TB ;
3
where kB TB is the thermal energy, Keff is the
effective anisotropy constant and V the particle
volume. t increases with decreasing temperature
resulting in an enhanced magnetisation. Also

ARTICLE IN PRESS
S.D. Shenoy et al. / Journal of Magnetism and Magnetic Materials 269 (2004) 217226

223

Fig. 7. Magnetisation curves of ZFP ZF30 ZF300 at KRT and 83 K (LT).

25
83K

M s (emu/gm)

20
15
10

300K

5
0
0

100
200
Milling time (minutes)

300

Fig. 8. Dependence of milling time on saturation magnetisation.

the saturation magnetisation (Ms ) is governed by


the following equation [39]:
i
2K h
1  25kT =KV 1=2 :
Ms
Hc

According to the above relation Ms increases


with decrease of temperature and this is in
agreement with our results.
Coprecipitation results in large structural and
chemical disorder. Ball milling will produce greater atomic disorder due to mechanical displacement
of ions under high stress and shear. This order may
contribute to magnetisation [1,6,8].
As the particle size (hence the volume) decreases
the probability of thermally activated magnetisation increases. Surface effects become increasingly
important as the particle volume decreases. So
surface effects could not be ruled out in the ne
particle systems [4042].

As the inversion increases, the material makes


the transition to ferromagnetism through complex
magnetic states [1]. One complex state is the
distribution of Fe3+ and Fe2+ ions and the
oxygen vacancies resulting from milling which
can create the super exchange paths and induce
spin disorder [14]. Cation distribution may also
differ between those present near the particles
surface and those of non-surface atoms [11,43].
This FM ordering coupled with the antiferromagnetically (AFM) ordered a-Fe2O3 may produce
FMAFM exchange interaction that can modify
magnetic properties as suggested by various
researchers [1819]. But the dispersion of FM
clusters in an AFM matrix is ruled out here since
a-Fe2O3 percentage is small [21].
3.3. Dielectric
The dielectric properties of ferrites are inuenced by various factors out of which the grain size
plays a key role. The samples which assumed
heterogeneous nature due to the individual grains
inuences the dielectric variation with frequency.
Thus the dielectric behaviour of the ferrites is
attributed primarily to interfacial polarisation
resulting from their heterogeneous structure comprising low resistivity grains separated by high
resistivity grain boundaries as suggested by Koops
phenomenological dispersion theory [4447]. The
Fe2+ and Fe3+ ions present in ferrites contribute
effectively to produce interfacial polarisation. This
is supported by the inverse proportionality between dielectric permittivity and resistivity as
observed by various researchers.
Fig. 9 shows the dielectric variation with
frequency. The oxygen ion vacancies produced

ARTICLE IN PRESS
S.D. Shenoy et al. / Journal of Magnetism and Magnetic Materials 269 (2004) 217226

224

Fig. 9. Variation of dielectric permitivity with log (frequency) different temperatures.

10

100 KHz

8
6

5 MHz

10 MHz

14

Temperature=353K

100 KHz

8
7
6

5 MHz

5
10 MHz

Dielectric permittivity

9
Temperature=300K

Dielectric permittivity

Dielectric permittivity

12

2
0

100
200
300
Milling time (minutes)

Temperature = 393K

12
100 KHz

10
8
5 MHz

6
4

10 MHz

2
0

100
200
300
Milling time (minutes)

100
200
300
Milling time (minutes)

Fig. 10. Variation of dielectric constant with milling time at different frequencies (at xed temperatures).

during ball milling may also contribute to dielectric permittivity which is predominant at lower
frequencies [48]. The non-uniformity of the structure increases with reduction in size according to
our XRD results. Hence this should increase the
dielectric permittivity of ball milled samples and
our results are in accordance with this (Fig. 10).
The traps formed at the grain boundaries are
not of uniform shape and may become the cause
for deviation from parallel alignment of moments
of individual grains. Electrons are then transferred
from one grain to another by a process called spin
polarised tunnelling in which electron spin is
conserved. Tunnelling probability of spin polarised electrons is larger when the magnetisation
of the particles is aligned [43]. This will increase
the dielectric permittivity with milling.
Fig. 10 clearly shows the enhancement of
dielectric permittivity at low frequencies. However, dielectric permittivity decreases at higher
frequencies since the mobility of charge carriers is
low at higher frequencies and cannot follow the
alternation of applied AC electric eld [49].

The temperature dependence of dielectric


permittivity at different frequencies is shown in
Fig. 11. In dielectrics, the ionic mechanism of
polarisation increases dielectric permittivity when
the temperature increases. When the temperature
rises, the orientation of dipoles gets facilitated
resulting in an increased permittivity. But at very
high temperatures the chaotic thermal oscillations
of molecules are intensied and the degree of
orderliness of their orientation is diminished, thus
the permittivity passes through a maximum value.
We observe that the variation of dielectric
permittivity with temperature at low frequencies
(100 KHz) is much more pronounced than at
higher frequencies.

4. Conclusions
XRD results indicate that the particles formed
are ultrane and associated with strains. The
structural and magnetic studies on the milled
ZnFe2O4 samples indicate that zinc ferrite in the

ARTICLE IN PRESS
S.D. Shenoy et al. / Journal of Magnetism and Magnetic Materials 269 (2004) 217226

ZF30
100 KHz
500 KHz
1 MHz

5 MHz

9 MHz
10 MHz

300

350
400
Temp in K

10
500 KHz

1 MHz

5 MHz

9 MHz
10 MHz

4
300

12

100 KHz

350
400
Temp in K

Dielectric permittivity

ZFP

Dielectric permittivity

Dielectric permittivity

ZF300

10

225

100 KHz

500 KHz
1 MHz

8
5 MHz

6
9 MHz
10 MHz

4
300

350
400
Temp in K

Fig. 11. Variation of dielectric permitivity with temperature (at different frequencies).

ultrane regime exhibit net magnetisation. Strain


increases with milling producing an atomic disorder due to displacement of the position of both
anions and cations. Particle size measured show
that they are around 10 nm and no much decrease
with milling. This is because of the nanosized
starting precursor and the high local temperature
produced while milling. The prolonged milling
does not increase magnetisation much appreciably,
which may be due to the fact that the overall
increase in magnetisation may be masked by the
increased amount of a-Fe2O3, which increases with
milling. Dielectric behaviour of these particles
is due to interfacial polarisation as proposed
by Koops. The Fe2+ ions produced causes
Fe2+#Fe3+ hopping and the oxygen ion vacancy
in the lattice contribute to dielectric permittivity.
The ionic mechanism is enhanced in dielectrics
with the rise in temperature, which results in the
increase of permittivity with temperature. A clear
understanding of this anomalous behaviour of zinc
ferrite in the nano regime warrants further studies
using neutron scattering and/or ESCA.

Acknowledgements
One of the authors, MRA thanks Third World
Academy of Sciences (TWAS), Trieste, Italy for
nancial assistance received through project (Ref.
No. 00-118/RG/PHYS/AS). SDS and MRA thank
Inter University Consortium for Department of
Atomic Energy Facilities (IUC-DAEF), Govt. of
India for nancial support through project (Ref.
No. IUC/MUM/CRS-M-60).

References
[1] J.C. Ho, H.H. Hamdeh, Y.S. Chen, S.H. Lin, Y.D. Yao,
R.J. Willey, S.A. Oliver, Phys. Rev. B 52 (1995)
10122.
[2] W.C. Kim, S.J. Kim, Y.R. Uhm, C.S. Kim, IEEE Trans.
Magn. 37 (2001) 2362.
[3] H.H. Hamdeh, J.C. Ho, S.A. Oliver, R.J. Willey, G.
Oliveri, G. Busca, J. Appl. Phys. 81 (1997) 1851.
[4] B. Jeyadevan, K. Tohji, K. Nakatsuka, J. Appl. Phys. 76
(1994) 6325.
[5] H. Ehrhardt, S.J. Campbell, M. Hofmann, J. Alloys
Compounds 339 (2002) 255.
[6] G.F. Goya, H.R. Richenberg, M. Chen, W.B. Yelon,
J. Appl. Phys. 87 (2000) 8005.
[7] T. Kamiyama, K. Haneda, T. Sato, S. Ikeda, H. Asano,
Sol. State Commun. 81 (1992) 563.
[8] H.H. Hamdeh, J.C. Ho, S.A. Oliver, R.J. Willey,
J. Kramer, Y.Y. Chen, S.H. Lin, Y.D. Yao, M. Daturi,
G. Busca, IEEE Trans. Magn. 31 (1995) 3808.
[9] T. Sato, K. Haneda, M. Seki, T. Iijima, Appl. Phys. A 50
(1990) 13.
[10] M.R. Anantharaman, S. Jagatheesan, K.A. Malini, S.
Sindhu, A. Narayanasamy, C.N. Chinnasamy, J.P. Jacobs,
S. Reijne, K. Seshan, R.H.H. Smits, H.H. Brongersma,
J. Magn. Magn. Mater. 189 (1998) 83.
[11] M.T. Clerk, B.J. Evans, IEEE Trans. Magn. 33 (1997)
3745.
[12] T. Pannaparayil, S. Komarneni, R. Marande, M. Zadarko,
J. Appl. Phys. 67 (1990) 5509.
[13] V. Sepelak, K. Tkacova, V.V. Boldyrev, S. Wibmann,
K.D. Becker, Physica B 234236 (1997) 617.
[14] G.F. Goya, H.R. Rechenberg, J. Magn. Magn. Mater.
196197 (1999) 191.
[15] V. Sepelak, P. Druska, U. Sleinike, Mater. Struct. 6 (1999)
100.
[16] B.S. Murty, S. Ranganathan, Int. Mater. Rev. 43 (1998)
101.
[17] K. Haneda, H. Kojima, J. Am. Ceram. Soc. 57 (1974) 68.
[18] J. Sort, J. Nogues, X. Amils, S. Surinach, J.S. Munoz,
M.D. Baro, Appl. Phys. Lett. 75 (1999) 3177.

ARTICLE IN PRESS
226

S.D. Shenoy et al. / Journal of Magnetism and Magnetic Materials 269 (2004) 217226

[19] J. Sort, J. Nogues, S. Surinach, J.S. Munoz, M.D. Baro, E.


Chappel, F. Dupont, G. Chouteau, Appl. Phys. Lett. 79
(2001) 1142.
[20] Z. Jin, W. Tang, J. Zhang, H. Lin, Y. Du, J. Magn. Magn.
Mater. 182 (1998) 231.
[21] W.A. Kaczmarek, B. Idzikowski, K.H. Muller, J. Magn.
Magn. Mater. 177181 (1998) 921.
[22] C.N. Chinnasamy, A. Narayanasamy, N. Ponpandian, K.
Chattopadhyay, K. Shinoda, B. Jeyadevan, K. Tohji, K.
Nakatsuka, T. Furubayashi, I. Nakatani, Phys. Rev. B63
(2001) 184108.
[23] J.M. Hastings, L.M. Corliss, Phys. Rev. 102 (1956) 1460.
[24] C.N. Chinnasamy, A. Narayanasamy, N. Ponpandian,
R.J. Justin, B. Jeyadevan, K. Tohji, K. Chattopadhyay,
J. Magn. Magn. Mater. 238 (2002) 281.
[25] W. Schiessel, W. Potzel, H. Karzel, M. Steiner, G.M.
Kalvius, A. Martin, M.K. Krause, I. Halevy, J. Gal, W.
Schafer, G. Will, M. Hillberg, R. Wappling, Phys. Rev. B
53 (1996) 9143.
[26] F.K. Lotgering, J. Phys. Chem. Solids 27 (1996) 139.
[27] E.M. Mohammed, K.A. Malini, P. Kurian, M.R. Anantharaman, Mater. Res. Bull. 37 (2002) 753.
[28] T. Sato, IEEE Trans. Magn. Magn. 6 (1970) 795.
[29] Q. Chen, A.J. Rondinone, B.C. Chakoumakos, J.Z.
Zhang, J. Magn. Magn. Mater. 194 (1999) 1.
[30] G.F. Goya, H.R. Rechenberg, J.Z. Jiang, J. Appl. Phys. 84
(1998) 1101.
[31] D. Arcos, R. Valenzuela, M. Vazquez, M. ValletRegi,
J. Solid State Chem. 141 (1998) 10.
[32] M.I. Rosales, O.A. Valenzuela, R. Valenzuela, IEEE
Trans. Magn. 37 (2001) 2373.

[33] C. Suryanarayana, G.H. Chen, F.H. Samfroes, Scr.


Metall. 26 (1992) 1727.
[34] J. Ding, T. Tsuzuki, P.G. McCormick, R. Street, J. Magn.
Magn. Mater. 162 (1996) 271.
[35] S. Manojkumar, S. Sekhon, J. Phys. D 34 (2001) 2995.
[36] M. Zheng, X.C. Wu, B.S. Zou, Y.J. Wang, J. Magn.
Magn. Mater. 183 (1998) 152.
[37] V. Masheva, M. Grigorova, N. Valkov, H.J. Blythe, T.
Midlarz, V. Blaskov, J. Geshev, M. Mikhov, J. Magn.
Magn. Mater. 196197 (1999) 128.
[38] Q. Chen, J.Z. Zhang, Appl. Phys. Lett. 73 (1998) 3156.
[39] B.D. Cullity, Introduction to Magnetic Materials. Addison-Wesley, London, 1972, p. 415.
[40] R.V. Upadhyay, Pramana 49 (1997) 309.
[41] K. Haneda, Can. J. Phys. 65 (1987) 1233.
[42] A.H. Morrish, K. Haneda, X.Z. Zhou, Proceedings of
the NATO Advanced Study Institute on Nanophase
Materials-Syntheses, Properties and Applications, Kluwer
Academic, Netherlands, 1994, p. 515.
[43] X. Battle, A. Laborta, J. Phys. D 35 (2002) R15.
[44] C.G. Koops, Phys. Rev. 83 (1951) 121.
[45] D. Ravinder, A.V.R. Reddy, G. Rangamohan, Mater.
Lett. 52 (2002) 259.
[46] V.P. Miroshkin, Y.A.I. Panova, V.V. Passynkov, Phys.
Stat. Sol. (A) 66 (1981) 179.
[47] V.R.K. Murthy, B. Viswanathan, Ferrite Materials Science
and Technolgy, Narosa Publishing House, New Delhi,
1990, 15, 60.
[48] N. Ponpandian, A. Narayanasamy, J. Appl. Phys. 92 (5)
(2002) 2770.
[49] A.M. Abdeen, J. Magn. Magn. Mater. 192 (1999) 121.

International Scholarly Research Network


ISRN Ceramics
Volume 2011, Article ID 194575, 8 pages
doi:10.5402/2011/194575

Research Article
Microstructure Characterization of Nanocrystalline Magnesium
Ferrite Annealed at Elevated Temperatures by Rietveld Method
Swapan Kumar Pradhan,1 Sumanta Sain,1 and Hema Dutta2
1 Department
2 Department

of Physics, The University of Burdwan, Golapbag, West Bengal, Burdwan 713104, India
of Physics, Vivekananda College, West Bengal, Burdwan 713103, India

Correspondence should be addressed to Swapan Kumar Pradhan, skp bu@yahoo.com


Received 17 August 2011; Accepted 20 September 2011
Academic Editors: H. I. Hsiang, W.-C. Oh, S. Shannigrahi, and S. Wongkasemjit
Copyright 2011 Swapan Kumar Pradhan et al. This is an open access article distributed under the Creative Commons
Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is
properly cited.
Nanocrystalline magnesium ferrite is synthesized at room temperature by high energy ball milling the MgO and -Fe2 O3 (1 : 1 mol
fraction) powders. The Rietveld structure and microstructure refinements of X-ray powder diraction data of 9 h milled sample
reveal the presence of mixed and nearly inverse spinel nanocrystalline Mg-ferrite phases. Postannealing of nanocrystalline powder
within 8731473 K reveals continuous change in cation distribution among the tetrahedral and octahedral sites of mixed spinel
lattice leading to nearly inverse spinel structure with increasing temperature. Mixed spinel structure finally transformed into
inverse spinel structure after 1 h of annealing at 1073 K. The ferrite phase becomes completely stoichiometric by solid-state
diusion of unreacted (0.3 mol fraction) -Fe2 O3 into spinel lattice after 1 h of annealing at 1273 K. Interestingly, particle sizes
of ferrite phases do not increase considerably up to 1073 K and increase suddenly after transformation of mixed spinel into nearly
inverse spinel structure.

1. Introduction
The synthesis of nanocrystalline spinel ferrites has been
investigated intensively in recent years due to their potential
applications in nonresonant devices, radio frequency circuits, high-quality filters, rod antennas, transformer cores,
read/write heads for high-speed digital tapes, and operating
devices [13]. Nowadays, these materials are largely synthesized in nanometric scale for new and improved properties
[46]. Magnesium ferrite, MgFe2 O4 (Figure 1), is a soft
magnetic n-type semiconductor material [7], which finds a
number of applications in heterogeneous catalysis, adsorption, sensors, and in magnetic technologies. Among the other
methods of preparation of nanocrystalline ferrite powders,
high-energy ball milling is a very suitable solid-state processing technique for the preparation of nanocrystalline ferrite
powders exhibiting new and unusual properties [811].
Ferrites have the general formula (M1x Fex)[Mx Fe2x ]O4 .
The divalent metal element M (Mg, Zn, Mn, Fe, Co, Ni,
or mixture of them) can occupy either tetrahedral eight (A)
or sixteen octahedral [B] sites of a cubic mineral spinel
(MgAl2 O4) structure as depicted by the parentheses or brack-

ets, respectively. The structural formula of Mg-ferrite is


usually written as (Mg1x Fex )[Mgx Fe2x ]O4 , where x represents the degree of inversion (defined as the fraction of (A)
sites occupied by Fe3+ cations). Depending on distribution
of cations in (A) and [B] sites, ferrites may exist in two
extreme statesnormal (x = 0) and inverse (x = 1) or in
an intermediate mixed state. The magnetic properties of
spinel ferrites are strongly dependent on the distribution of
the dierent cations among (A) and [B] sites. The cation
distribution of slowly cooled Mg-ferrite (from 1773 K to
room temperature) was reported [12, 13] as nearly inverse
spinel lattice with x = 0.9, but it is also experimentally verified
that distribution of cations among these sites depends on
material preparation. The complexity of the above formula and the nature of preparation techniques strongly suggest a variation in unit cell composition and possibly a variation in the arrangement of Mg2+ and Fe3+ ions over the
available tetrahedral and octahedral sites. These types of
disorder should influence the integrated intensity and peak
broadening of diraction lines. In view of the large scattering
power of Fe3+ relative to the other ions in the structure, integrated intensities of the diraction lines of the crystalline

ISRN Ceramics
study for precise determination of several microstructural
parameters (change in lattice parameters, particle size, and
lattice strain), cation redistribution in terms of occupancy
factors of cations at dierent sites as well as relative phase
abundances of this multiphase nanocrystalline material containing a large number of overlapping reflections. However,
the selected area electron diraction pattern (SAED) of 9 h
milled sample could not detect the presence of any secondary
phase, and the pattern seems to be composed of single Mgferrite phase.

2. Experimental

Figure 1: Fragments from the bulk Mg-ferrite normal spinel: MgO tetrahedra, Fe-O octahedral; positions of Fe (blue) and oxygen
(red) atoms are shown.

Mg-ferrite (cubic, a = 0.83998 nm, space group: Fd3m,


Z = 8; ICDD PDF #88-1943) change to a large extent due
to occupancy of cations in (A) and [B] sites. As the ionic
radii of the cations are quite dierent, random occupancy
of cations in any site will also lead to produce lattice strain
due to mismatch in ionic size, which in turn influences
the coherently diracting domain (particle size). To account
for all these eects on X-ray powder diraction patterns of
ball-milled and postannealed Mg-ferrites samples, a diraction profile analysis based on both structure and microstructure simulation is essential.
An important aspect, which was studied only to a limited extent until now [4, 14] is related to the thermal stability of mechanically induced structural defects in these
solids and the mechanism of their relaxation on heating
at elevated temperatures. Since in the process of relaxation
of mechanically induced metastable states the advantageous
properties of mechanically treated solids are mostly lost, it is
necessary to know the lifetime of the activated surface and
the time dependence of the relaxation. This understanding
would be of principal importance from the scientific point
of view in the development of an atomistic and microscopic
theory of the mechanochemical processes, as well as from
the practical point of view in the application of mechanically
treated solids at elevated temperatures.
Nanocrystalline Mg-ferrite powders were produced by
ball milling the stoichiometric mixture (1 : 1 mol%) of MgO
and -Fe2 O3 . It was found that the prepared ferrite powders
are nonstoichiometric one (3 wt%-Fe2 O3 remains unreacted), and the distribution of cations among the tetrahedral and octahedral sites is somewhat dierent (mixture of
mixed and nearly inverse spinels) from the reported value
(only nearly inverse spinel) [15]. In the present study, we
have examined the 9 h ball-milled sample for phase transformation kinetics, thermal stability of nanocrystalline ferrite
phase, and cation redistribution among (A) and [B] sites at
elevated temperatures.
The Rietveld analysis based on structure and microstructure refinement [1618] has been adopted in the present

Accurately weighed powders of 20.15 wt.% MgO (M/s


Merck, 99% purity) and 79.85 wt.% of -Fe2 O3 (M/s Glaxo,
99% purity) were hand-ground in an agate mortar under
doubly distilled acetone for more than 5 h for a homogeneous mixture of the powders. High-energy ball milling
of a part of the dried homogeneous powder mixture was
conducted in a planetary ball mill (Model P5, M/s Fritsch,
GmbH, Germany). The rotation speed of the disk was
325 rpm and that of the vials was 475 rpm. Milling of
powder samples was done at room temperature in hardened
chrome steel (Fe-1 wt.% Cr) vial (volume 80 mL) using 30
hardened chrome steel balls of 10 mm diameter at BPMR =
40 : 1. The time of milling varies from 1 h to 11 h depending
upon the rate of formation of Mg-ferrite phase. The 9 h ballmilled powder was postannealed at 873, 973, 1073, 1273,
and 1473 K successively each for 1 h duration in air in a
programmable furnace with 5 K/min and 20 K/min heating
and cooling rates, respectively.
The X-ray powder diraction profiles of unmilled homogeneous powder mixture, ball-milled and postannealed samples were recorded using Ni-filtered Cu K radiation from
Philips XPert powder diractometer operated at 40 KV and
40 mA. The background noise and fluorescent radiation were
reduced significantly employing a diracted beam graphite
monochromator. For this experiment, 0.5 divergence slit
and 0.5 receiving slit were used. The step-scan data (of step
size 2 =0.02 and counting time 10 s, in some cases 60 s) for
the entire angular range (2 = 15100 ) were recorded, and
crystalline phases were identified by search-match software.
Microstructure characterization of 9 h ball-milled Mgferrite powder has also been made using high-resolution
transmission electron microscopes (Model HR-TEM 2100F,
JEOL). For TEM study, a pinch of sample was dispersed
in ethyl alcohol medium and placed under ultrasonic bath
for breaking agglomeration of nanoparticles under vibration.
A small drop of alcohol with dispersed/suspended sample
particle was placed on the carbon-coated 3 mm Cu-grid, and
the grid was kept overnight under vacuum to get completely
dry and well distributed fine particles over the entire area of
the carbon-coated grid. HR-TEM was operated at 200 kV.

3. Method of Analysis
In the present study, we have adopted the Rietvelds powder
structure refinement analysis [1622] of X-ray powder
diraction data to obtain the refined structural parameters,

ISRN Ceramics
such as atomic coordinates, occupancies, lattice parameters,
and thermal parameters, and microstructural parameters,
such as particle size and r.m.s. lattice strain. For instrumental
broadening correction, a specially processed Si standard [23]
was used. The Rietvelds software MAUDWEB 2.24 (Java
web start version) [18] is specially designed to refine simultaneously both the structural and microstructure parameters through a least-squares method. The peak shape was
assumed to be a pseudo-Voigt (pV) function with asymmetry. The background of each pattern was fitted by a
polynomial function of degree 4. To simulate the theoretical
X-ray powder diraction pattern containing all the MgO,
-Fe2 O3 , and Mg-ferrite (normal spinel) and Mg-ferrite
(inverse spinel) phases in a single pattern, the following
considerations were made:
(i) for MgO (cubic, space group: Fm3m (225), a =
0.42 nm (ICDD PDF no. 87-0653)), Mg2+ , and O2
in special Wycko positions 4a and 4b, respectively,
(ii) for -Fe2 O3 (rhombohedral, space group: R3c (167),
a = 0.5032 and c = 1.3733 nm (ICDD PDF no. 890599)) with Fe and O atoms in special Wycko
positions 12c and 18e, respectively,
(iii) for Mg-ferrite (cubic, normal spinel, space group:
Fd3m (227), a = 0.83998 nm (ICDD PDF no. 881943)) with Mg2+ (A-site), Fe3+ [B-site], and O2 in
the 8a, 16d, and 32e Wycko positions, respectively,
(iv) for Mg-ferrite (cubic, inverse spinel, space group:
Fd3m (227), a = 0.83998 nm (ICDD PDF no. 881943)) with 0.1 Mg2+ + 0.9 Fe3+ (A-site), 0.5 Mg2+
+ 0.5 Fe3+ [B-site], and O2 in the 8a, 16d, and 32e
Wycko positions, respectively.
3.1. Crystalline Structure Characterization by X-Ray Powder
Diraction. For microstructure characterization, the experimental profiles are fitted with the most suitable pV analytical
function because it takes individual care for both the particle
size and strain broadening of the experimental profiles. A
detailed mathematical description of the Rietveld analyses
has been reported elsewhere [1622]. Refinement continues
till convergence is reached with the value of the quality
factor, GoF very close to 1 (varies between 1.1 and 1.3),
which confirms the goodness of refinement. The Rietvelds
method was successfully applied for determination of the
quantitative phase abundances in multiphase materials [19
22]. The structure refinement along with size-strain broadening analysis is carried out simultaneously by adopting the
standard procedure [17, 18].
3.2. Size-Strain Analysis. The basic consideration of this
method is the modeling of the diraction profiles by an analytical function, which is a combination of Cauchy, Gaussian,
and asymmetry functions as well. It is well established
that the observed peak broadenings of the sample profiles
are mainly due to the presence of small particle size and
r.m.s. strain inside the particles. The particle size and strain
broadening can be approximated better with Cauchy and
Gaussian type functions, respectively [1923]. Being a linear

3
combination of a Cauchy and Gaussian functions, the pV
function is the most reliable peak-shape function and is
being widely used in the Rietvelds structure refinement
software. The process of successive profile refinements modulated dierent structural and microstructural parameters
of the simulated pattern to fit the experimental diraction
pattern. Profile refinement continues until convergence is
reached in each case, with the value of the quality factor
(GoF) approaches very close to 1.

4. Results and Discussion


Figure 2(a) represents the X-ray powder diraction patterns
of 9 h ball-milled and postannealed powders heat treated at
dierent elevated temperatures each for 1 h duration. It is
evident from the plots that the powder pattern of 9 h ballmilled sample contains only reflections of Mg-ferrite and Fe2 O3 phases. Reflections of -Fe2 O3 phase are very weak
when compared to those of Mg-ferrite phase. Reflections of
Mg-ferrite phase are significantly broad and asymmetric in
nature as expected from nanocrystalline material containing
significant amount of lattice imperfections. Due to the broad
nature of the peaks, some of the reflections of Mg-ferrite
phase are either partially or completely overlapped with
some reflections of the -Fe2 O3 phase. In the selected area
electron diraction (SAED) pattern (Figure 2(b)), reflections
of these two phases are not well resolved, and it seems that
the 9 h ball-milled sample is composed of only the Mgferrite reflections. It is also evident in the SAED that these
reflections are quite broad like a nanocrystalline powder.
Since the content of -Fe2 O3 phase is insignificant in the
milled powder, some of the isolated -Fe2 O3 reflections
(Figure 2(a)) did not appear in the SAED pattern. It is
therefore impossible to detect the presence of trace amount
of -Fe2 O3 phase even from the transmission electron
microscopy image. The Rietvelds powder structure refinement method [1622] is being utilized for microstructure
evolution in several cases of nanocrystalline ball-milled and
annealed samples having very broad, asymmetric, partially
or even completely overlapping reflections. In the present
case, as some of the reflections of Mg-ferrite phase are
overlapped with very broadened -Fe2 O3 reflections, both
particle size and lattice strain of individual phases cannot be
estimated accurately considering all reflections of respective
phases without employing any Rietvelds software based on
both structure and microstructure refinements. To show the
accuracy in fitting, fitting residuals (Io -Ic ) are also plotted at
the bottom of respective patterns. From the plots of residue,
it is evident that except at the peak positions, the residual
plots are almost linear and indicate that all the reflections
of all three phases are fitted reasonably well. Results revealed
from microstructure analysis are shown in Figures 36.
A critical comparison between the 9 h ball-milled nanocrystalline ferrite and ICDD-reported (say, bulk) Mg-ferrite
phase reveals that there are anomalies in intensity distribution of some nanocrystalline ferrite peaks. Considering
the fact that bulk Mg-ferrite is a nearly inverse spinel,
this anomaly in intensity distribution may arise for the
following facts: (i) cation distribution among the (A) and [B]

ISRN Ceramics
6000

Intensity (a.u.)

4000
2000

1473 K

0
3000
2000
1000
0
1000

1273 K

500

1073 K

0
600
400
200
0

873 K

400
9h

200
0
MgFe2 O4
-Fe2 O3
20

30

40

50
2 (deg)

60

70

80

(a)

(440)
(511)
(400)
(311)
(220)

0.5 1/nm
(b)

Figure 2: (a) X-ray powder diraction patterns of 9 h ball-milled


MgO--Fe2 O3 mixture annealed at dierent temperatures for 1 h
duration. The peak positions of dierent phases are shown as
small solid bar markers at the bottom of the figure. (b) Indexed
selected area electron diraction pattern of Mg-ferrite powder after
annealing the 9 h milled sample at 1473 K for 1 h duration.

sites is somewhat dierent from the bulk Mg-ferrite and/or


(ii) the nanocrystalline ferrite may be an inhomogeneous
mixture of both the normal and inverse ferrite phases. The
dierence in intensity distribution has almost been resolved
by considering an additional ferrite phase with a relatively
larger lattice parameter and dierent cation distribution.
It has been found from the refinement result of the ballmilled sample that the major nanocrystalline ferrite phase is a

mixed spinel, while the minor nanocrystalline phase is nearly


an inverse spinel [15] with larger lattice parameter. In the
present study, progressive evaluation in the microstructure
of the annealed sample has been reported considering both
the mixed and inverse spinel phases in the refinement process
with equal probabilities of occupancy of both the Mg2+
and Fe3+ cations to be in both the (A) and [B] sites. The
occupancy factors of both the cations have been refined
during the refinement to fit the intensity distributions of
reflections.
As such there is no significant change in the XRD
patterns (Figure 2) of ball-milled and the annealed sample
at 873 K, but a close observation reveals a change in the
intensity ratio of the overlapping reflections of the two phases
which indicates the slow change in relative contents of both
phases in the annealed samples. However, the content of Mgferrite phase increases gradually, but that of -Fe2 O3 phase
decreases slowly with increasing annealing temperature in
the temperature range 8731073 K. A remarkable change
in peak broadening, peak asymmetry, peak intensity, and
intensity ratio in XRD patterns has been noticed when the
postannealed sample was annealed after 1073 K for 1 h of
duration. Sudden decrease in the content of -Fe2 O3 phase
is noticed from the reduction in the peak intensity values of
the reflections of -Fe2 O3 phase. Peak broadening and peak
asymmetry of Mg-ferrite phase decreases significantly with
increase in the peak intensity values. Comparing the intensity
and line position of individual reflection, it is clearly evident
that increase in annealing temperature after 1073 K results in
release in lattice strain and increase in particle size values of
Mg-ferrite phase to a large extent. Change in the intensity
ratio particularly in the higher angle side may be attributed
to the cation redistribution with the solid-state diusion of
-Fe2 O3 in the lattice of Mg-ferrite. Intensity distribution
of Mg-ferrite reflections approaches slowly to the ICDDreported value of the Mg-ferrite with nearly inverse spinel
lattice. Further annealing (from 1273 K to 1473 K) results
in complete absence of -Fe2 O3 phase, and samples contain
only the Mg-ferrite phase which is nearly an inverse spinel
lattice as identified from intensity distribution.
Figure 3 shows the dependence of relative phase abundances of dierent phases with increasing annealing temperature. The content (volume fraction) of -Fe2 O3 phase
decreases very slowly with annealing temperature and
becomes nil at 1473 K. It indicates that the -Fe2 O3 phase diffuses slowly into the Mg-ferrite lattice. Phase content of
mixed Mg-ferrite phase increases, and that of inverse Mgferrite phase decreases initially with increasing annealing
temperature in the temperature range 873 K1073 K. This
nature of variation depicts the solid-state diusion of Fe2 O3 phase into mixed spinel lattice, and the inverse spinel
phase has been partially transformed to its mixed counterpart. The mixed spinel phase disappears completely after
annealing the ball-milled sample at 1073 K followed by a
rapid increase in the phase content of inverse spinel phase.
It signifies that structure of the mixed spinel phase changes
suddenly after 1073 K of annealing temperature. It is interesting to note that after annealing the sample at 1073 K for
1 h duration, rate of diusion of -Fe2 O3 in Mg-ferrite lattice

ISRN Ceramics

5
1

1.1
1

0.9

0.9

0.8
Occupancy factor

Volume fraction

0.8
0.7
0.6
0.5
0.4
0.3

0.7
0.6
0.5

0.2
0.4

0.1
0

0.3
900

1000

1100
1200
1300
Annealing temperature (K)

1400

1500

Mixed spinel
Inverse spinel
-Fe2 O3

900

1000

1100

1200

1300

1400

1500

Annealing temperature (K)


Mixed spinel
Fe in (A) site
Mg in [B] site

Inverse spinel
Fe in (A) site
Mg in [B] site

Figure 3: Nature of variations of volume fractions of dierent


phases obtained by annealing 9 h ball-milled MgO--Fe2 O3 mixture
powders with increasing temperature.

Figure 4: Nature of variations of occupancy factors of cations in


dierent sites of Mg-ferrite with increasing temperature.

increases and that may result in decrease in Fe3+ cation deficiency nonstoichiometric Mg-ferrite phase matrix. Redistribution of cations inside mixed spinel leads to the complete
transformation of mixed spinel to inverse spinel in the
annealing temperature range 1073 K1273 K which results
the rapid increase in phase content of Mg-ferrite phase
with nearly inverse spinel lattice. Further annealing of the
sample after 1273 K for 1 h duration results complete solidstate diusion of unreacted -Fe2 O3 phase into the inverse
spinel lattice, and the composition of the Mg-ferrite becomes
completely stoichiometric.
In the present work, we found that the nanostructured
MgFe2 O4 prepared by high-energy milling is metastable
with respect to structural changes at elevated temperatures.
Sepelak et al. [24] also found that the thermal stability
of the mechanosynthesized product extends up to 623 K.
Above this temperature, a gradual crystallization of the
nanoscale MgFe2 O4 powders takes place. In the present case,
we found the thermal stability up to 1073 K. During the
annealing process, the nonequilibrium cation distribution
relaxes toward their equilibrium configuration.
The cation distribution in MgFe2 O4 , upon which many
physical and chemical properties depend, is a complex function of processing parameters and depends on the preparation method of the material [14]. The cation distribution in
mechanosynthesized spinel ferrites from Mossbauer spectra
and/or X-ray diraction patterns is not as straightforward
as is frequently claimed in the literature. Sepelak et al.
[24] reported for the first time the microstructure characterization and cation distribution in nanosized MgFe2 O4
synthesized in a one-step mechano-chemical process. They
also confirmed that the nonuniform cation distribution
within MgFe2 O4 nanoparticles by means of Mossbauer
spectroscopy is consistent with the results of Rietveld analysis

of the XRD data [15] showing that the mechanosynthesized


material consists of two spinel phases with dierent cation
distributions. In another paper, Sepelak et al. reported
that the enhanced coercivity in nanosized MgFe2 O4 can be
attributed to the high volume of the grain boundaries with a
nonequilibrium cation distribution and spin canting [25].
In the present study, mixed spinel phase is a major one
instead of inverse spinel in the ball-milled sample due to
Fe3+ cation deficiency inside the mixed spinel matrix at room
temperature. In the annealing temperature range 873 K
1073 K, no change is observed in the occupancy factors
of cations (Figure 4) in the inverse spinel phase. However,
the occupancy of Mg2+ cation on [B] site increases and
then decreases, and Fe3+ cation on (A) site increases with
annealing temperature in mixed spinel phase. It is interesting
to note that the occupancy of Mg2+ cation on [B] site
drops suddenly from 0.85 to 0.35 and that of Fe3+ cation
on (A) site increases inside inverse spinel matrix followed
by the transformation of mixed spinel phase to its inverse
counterpart at 1073 K. These observations suggest that at
higher annealing temperature with increasing solid-state
diusion of -Fe2 O3 , random distribution of cations in the
mixed spinel proceeds towards the inverse spinel lattice.
In the annealing temperature range 1073 K1273 K, rapid
increase in the occupancy of Fe3+ cation on (A) site in the
inverse spinel leads to a significant growth of inverse spinel
phase in the cost of excess -Fe2 O3 phase. Further, annealing
results in a slight increment in the increase of Fe3+ cation on
(A) site and decrease of Mg2+ cation on [B] site in inverse
spinel phase in favour of almost stoichiometric composition.
The variations of lattice parameters of dierent phases
with annealing temperature are shown in Figure 5. As
the cation compositions of inverse and mixed phases are
dierent, they also have dierent lattice parameter value at

ISRN Ceramics
0.865
0.86
0.855
0.85
0.845
0.84
0.835
1.38

120

MgFe2 O4

100

900

1000

1100
1200
1300
1400
Annealing temperature (K)

1500

Inverse Spinel
Mixed Spinel

Particle size (nm)

Lattice parameters (nm)

80
60
40
20

-Fe2 O3
1.375

26
24
22
20
18
16
14
12
850 900 950 1000 1050 1100

c
900

1000

a
0.53

1000

1100

1200

1300

1400

1500

1500

1400

1500

(a)

Annealing temperature (K)

Figure 5: Nature of variations of lattice parameters of dierent


phases obtained by annealing 9 h ball-milled MgO--Fe2 O3 mixture
powders with increasing temperature.

40

r.m.s. strain (L 2 )1/2 103

room temperature. The cation distribution in mixed spinel


proceeds towards the inverse spinel with increasing annealing
temperature due to solid-state diusion of nanocrystalline
-Fe2 O3 into mixed spinel matrix. Lattice parameter of
cubic mixed spinel phase does not change significantly
with annealing temperature but that of inverse spinel phase
decreases initially very rapidly and proceeds towards that
of mixed spinel phase. After annealing the sample for
1073 K for 1 h duration, lattice parameter values of both
the phases become nearly equal. Lattice parameter value of
inverse spinel phase then approaches slowly to mixed spinel
lattice parameter with increasing annealing temperature.
This nature of variation of inverse spinel lattice parameter
may be attributed to the transformation of mixed spinel
phase having major phase content to the inverse phase. A
steady-state nature of lattice parameter variation of inverse
spinel phase at higher annealing temperature indicates that
the cation distribution among the (A) and [B] sites becomes
saturated at the last stage of annealing. Lattice parameters
of rhombohedral -Fe2 O3 phase do not change significantly
with annealing temperature.
It is evident from Figure 2 that the peak broadening
decreases and peak intensity increases with increasing annealing temperature. The peak broadening in the ball-milled
sample may arise primarily due to small crystallite size and
lattice strain. These lattice imperfections can be reduced
to a large extent by postannealing the ball-milled powder
at elevated temperatures. The noticeable change in XRD
patterns with increasing annealing temperature indicates the
continuous growth of Mg-ferrite particles due to continuous
release in lattice strain followed by rearrangement of ions in
respective proper sites. The particle size (coherently diracting domain) and r.m.s. lattice strain values for individual

1400

Inverse spinel
-Fe2 O3
Mixed spinel

0.525
900

1100
1200
1300
Annealing temperature (K)

30

20

3
2

10

1
0

800

900

900

1000 1100 1200 1300

1000

1100
1200
1300
Annealing temperature (K)

Inverse spinel
Mixed Spinel
-Fe2 O3
(b)

Figure 6: Nature of variations of (a) particle sizes, (b) r.m.s strains


of dierent phases obtained by annealing 9 h ball-milled MgO-Fe2 O3 mixture powders with increasing temperature.

phases in postannealed samples have been estimated from


the Rietvelds refinement and plotted in Figures 6(a) and
6(b), respectively. Observation from Figure 6(a) indicates
that the particle size of mixed spinel phase increases from
12 nm to 25 nm within the temperature range 873 K
1073 K. There is no considerable change in particle sizes of
inverse spinel phase up to 1073 K. A significant increase in
the particle size value of inverse spinel phase can be noticed
from the nature of the plot after annealing the sample at
1473 K for 1 h duration. It is interesting to note that -Fe2 O3
has been completely diused into inverse spinel lattice and
results in formation of a stoichiometric inverse spinel lattice.

ISRN Ceramics
The significant grain growth may be attributed to the complete ordering of inverse spinel lattice with proper placement
of cations at their respective tetrahedral and octahedral positions. Particle size of -Fe2 O3 phase slowly increases with
annealing temperature in a usual manner.
The r.m.s. lattice strain values of both -Fe2 O3 and
mixed spinel phases are in general decreasing in nature with
annealing temperature in a usual manner. The inverse spinel
phase grows with a large lattice strain. Initially, the inverse
spinel lattice could not release strain, rather strain increases
slightly with increasing temperature upto 1073 K. It suggests
that up to that temperature there was no significant change
in cation distribution in the lattice, and the small increment
in lattice strain may be attributed to the solid-state diusion
of -Fe2 O3 into the lattice.
The r.m.s. strain value drops suddenly to almost nil after
annealing the sample at 1273 K for 1 h duration. It is interesting to note that the mixed spinel is completely absent after
that temperature of annealing. It seems that the annealing
temperature favours distribution cations to their preferred
sites. As a result, lattice strain reduces to nil when samples
are annealed at higher range of temperature 1273 K1473 K.
It is also evident from the figure that in the same temperature
range particle size of inverse phase increases abruptly from
10 nm to120 nm which may be considered as particle size
growth of inverse spinel phase in a less constraint environment.

5. Conclusions
Nanocrystalline Mg-ferrite powders have been prepared by
high-energy ball milling of stoichiometric mixture of MgO
and -Fe2 O3 powders. Analysis of 9 h ball-milled samples
by Rietvelds method reveals the presence of mixed and
nearly inverse spinel nanocrystalline Mg-ferrite as major and
minor phases, respectively, with a small amount (3 wt%)
of unreacted - Fe2 O3 . Nanocrystalline ball-milled powder
was postannealed within 8731473 K temperature range.
Structural changes and microstructure characterization of
-Fe2 O3 and both the mixed and nearly inverse spinel Mgferrite phases in the postannealed samples have been investigated by Rietvelds analysis of X-ray powder diraction data
for the first time, to best of our knowledge. The following
important observations may be summarized as follows.
Distribution of cations among (A) and [B] sites in mixed
spinel matrix changes continuously, and occupancy of Fe3+
in (A) site (inverse spinel) increases with increasing annealing temperature. Mixed spinel structure changes to inverse
spinel after 1 h annealing at 1073 K. The composition of Mgferrite becomes completely stoichiometric one by solid-state
diusion of unreacted -Fe2 O3 into nearly inverse spinel
lattice after 1 h of annealing at 1273 K. Particle sizes of inverse
spinel do not increase considerably up to 1073 K and increase
suddenly with a large release in r.m.s. lattice strain after transformation of mixed spinel to a nearly inverse spinel structure.
The present study reveals that the nanocrystalline stoichiometric nearly inverse spinel Mg-ferrite can be formed
completely by postannealing the 9 h ball-milled powder
sample.

Acknowledgments
One of the authors, S. K. Pradhan, wishes to thank UGC
authority, India for providing grants for instruments under
DSA III and COSIST programs. The authors are also thankful to Dr. V. Petkov, Department of Physics, Central Michigan
University, USA for providing X-ray diraction facility.

References
[1] P. Ravindranathan and K. C. Patil, Novel solid solution precursor method for the preparation of ultrafine Ni-Zn ferrites,
Journal of Materials Science, vol. 22, no. 9, pp. 32613264,
1987.
[2] H. Igarashi and K. Okazaki, Eects of porosity and grain
size on the magnetic properties of nizn ferrite, Journal of the
American Ceramic Society, vol. 60, no. 1-2, pp. 5154, 1977.
[3] S. M. Antao, I. Hassan, and J. B. Parise, Cation ordering in
magnesioferrite, MgFe2 O4 , to 982 C using in situ synchrotron
X-ray powder diraction, American Mineralogist, vol. 90, no.
1, pp. 219228, 2005.
[4] V. Sepelak, D. Baabe, D. Mienert et al., Evolution of structure and magnetic properties with annealing temperature in
nanoscale high-energy-milled nickel ferrite, Journal of Magnetism and Magnetic Materials, vol. 257, no. 2-3, pp. 377386,
2003.
[5] M. Pavlovic, C. Jovalekic, A. S. Nikolic, D. Manojlovic,
and N. Sojic, Mechanochemical synthesis of stoichiometric
MgFe2 O4 spinel, Journal of Materials Science, vol. 20, no. 8,
pp. 782787, 2009.
[6] P. Holec, J. Plocek, D. Niznansky, and J. P. Vejpravova, Preparation of MgFe2 O4 nanoparticles by microemulsion method
and their characterization, Journal of Sol-Gel Science and
Technology, vol. 51, no. 3, pp. 301305, 2009.
[7] R. J. Willey, P. Noirclerc, and G. Busca, Preparation and characterization of magnesium chromite and magnesium ferrite
aerogels, Chemical Engineering Communications, vol. 123, pp.
116, 1993.
[8] P. Druska, U. Steinike, and V. Sepelak, Surface structure of
mechanically activated and of mechanosynthesized zinc ferrite, Journal of Solid State Chemistry, vol. 146, no. 1, pp. 13
21, 1999.
[9] V. Sepelak, A. Y. Rogachev, U. Steinike, D. C. Uccker, S. Wibmann, and K. D. Becker, Structure of nanocrystalline spinelferrite produced by high-energy ball-milling method, Acta
Crystallographica A, vol. 52, supplement, p. 367, 1996.
[10] I. P. Muthuselvam and R. N. Bhowmik, Mechanical alloyed
Ho3+ doping in CoFe2 O4 spinel ferrite and understanding of
magnetic nanodomains, Journal of Magnetism and Magnetic
Materials, vol. 322, no. 7, pp. 767776, 2010.
[11] M. Sinha, H. Dutta, and S. K. Pradhan, X-ray characterization and phase transformation kinetics of ball-mill prepared nanocrystalline Mg-Zn-ferrite at elevated temperatures, Physica E, vol. 33, no. 2, pp. 367369, 2006.
[12] N. W. Grimes, R. J. Hilleard, J. Waters, and J. Yerkess, An
X-ray diraction study of disorder in MgFe2 O4 and Mg(Cr,
Al)O4 , Journal of Physics C, vol. 1, no. 3, pp. 663672, 1968.
[13] G. E. Bacon and F. F. Roberts, Neutron-diraction studies of
magnesium ferrite-aluminate powders, Acta Crystallographica, vol. 6, pp. 5762, 1953.

8
[14] V. Sepelak, D. Schultze, F. Krumeich, U. Steinike, and K. D.
Becker, Mechanically induced cation redistribution in magnesium ferrite and its thermal stability, Solid State Ionics, vol.
141-142, pp. 677682, 2001.
[15] S. K. Pradhan, S. Bid, M. Gateshki, and V. Petkov, Microstructure characterization and cation distribution of nanocrystalline magnesium ferrite prepared by ball milling, Materials
Chemistry and Physics, vol. 93, no. 1, pp. 224230, 2005.
[16] H. M. Rietveld, Line profiles of neutron powder-diraction
peaks for structure refinement, Acta Crystallographica, vol. 22,
pp. 151152, 1967.
[17] H. M. Rietveld, A profile refinement method for nuclear and
magnetic structures, Journal of Applied Crystallography, vol.
2, pp. 6571, 1969.
[18] L. Lutterotti, MAUD version 2.24 , 2010, http://www.ing
.unitn.it/maud.
[19] U. K. Bhaskar, S. Bid, and S. K. Pradhan, Mechanochemical solid state synthesis of (Cd0.8 Zn0.2 )S quantum dots:
microstructure and optical characterizations, Journal of Alloys
and Compounds, vol. 509, no. 10, pp. 41764184, 2011.
[20] H. Dutta, M. Sinha, Y. C. Lee, and S. K. Pradhan, Microstructure characterization and phase transformation kinetics of
ball-mill prepared nanocrystalline Mg-Zn-ferrite by Rietvelds
analysis and electron microscopy, Materials Chemistry and
Physics, vol. 105, no. 1, pp. 3137, 2007.
[21] M. Sinha, H. Dutta, and S. K. Pradhan, Phase stability of
nanocrystalline Mg-Zn ferrite at elevated temperatures, Japanese Journal of Applied Physics, vol. 47, no. 11, pp. 86678672,
2008.
[22] S. Sain, S. Patra, and S. K. Pradhan, Microstructure and optical band-gap of mechano-synthesized Cdx Zn1 -x S quantum
dots, Journal of Physics D, vol. 44, no. 7, Article ID 075101,
8 pages, 2011.
[23] J.G.M.V. Berkum, Strain fields in crystalline of materials, Ph.D.
thesis, Delft University of Technology, Delft, The Netherlands,
1994.
[24] V. Sepelak, A. Feldho, P. Heitjans et al., Nonequilibrium
cation distribution, canted spin arrangement, and enhanced
magnetization in nanosized MgFe2 O4 prepared by a one-step
mechanochemical route, Chemistry of Materials, vol. 18, no.
13, pp. 30573067, 2006.
[25] V. Sepelak, D. Baabe, D. Mienert, F. J. Litterst, and K. D. Becker, Enhanced magnetisation in nanocrystalline high-energy
milled MgFe2 O4 , Scripta Materialia, vol. 48, no. 7, pp. 961
966, 2003.

ISRN Ceramics

Smart Materials
Research

BioMed Research
International

International Journal of

Corrosion

Journal of

Nanotechnology
Hindawi Publishing Corporation
http://www.hindawi.com

Volume 2014

Hindawi Publishing Corporation


http://www.hindawi.com

Volume 2014

Hindawi Publishing Corporation


http://www.hindawi.com

Volume 2014

Journal of

Composites
Hindawi Publishing Corporation
http://www.hindawi.com

Volume 2014

Volume 2014

Journal of

International Journal of

Metallurgy

Polymer Science

Hindawi Publishing Corporation


http://www.hindawi.com

Volume 2014

Nanomaterials

Hindawi Publishing Corporation


http://www.hindawi.com

Hindawi Publishing Corporation


http://www.hindawi.com

Volume 2014

Submit your manuscripts at


http://www.hindawi.com
Journal of

Textiles
Hindawi Publishing Corporation
http://www.hindawi.com

Volume 2014

Journal of

Nanoparticles
Hindawi Publishing Corporation
http://www.hindawi.com

Volume 2014

Nanomaterials
Journal of

Advances in

Materials Science and Engineering


Hindawi Publishing Corporation
http://www.hindawi.com

Volume 2014

Hindawi Publishing Corporation


http://www.hindawi.com

Volume 2014

Journal of

Journal of

Volume 2014

Hindawi Publishing Corporation


http://www.hindawi.com

Hindawi Publishing Corporation


http://www.hindawi.com

Volume 2014

Journal of

Materials

Coatings
Hindawi Publishing Corporation
http://www.hindawi.com

Scientifica

Crystallography
Volume 2014

Hindawi Publishing Corporation


http://www.hindawi.com

Volume 2014

The Scientific
World Journal
Hindawi Publishing Corporation
http://www.hindawi.com

Volume 2014

Biomaterials
HindawiPublishingCorporation
http://www.hindawi.com

Volume2014

Journal of

Journal of

NaNoscieNce

Ceramics
Hindawi Publishing Corporation
http://www.hindawi.com

InternationalJournalof

Volume 2014

Hindawi Publishing Corporation


http://www.hindawi.com

Volume 2014

ARTICLE IN PRESS

Journal of Magnetism and Magnetic Materials 284 (2004) 206214


www.elsevier.com/locate/jmmm

Synthesis and magnetic properties of nanostructured spinel


ferrites using a glycinenitrate process
Nobuyuki Kikukawa, Makoto Takemori, Yoshinobu Nagano,
Masami Sugasawa, Satoru Kobayashi
National Institute of Advanced Industrial Science and Technology (AIST), Research Institute for Green Technology,
AIST Tsukuba West, 16-1 Onogawa, Tsukuba 305-8569, Japan
Received 17 February 2004; received in revised form 18 June 2004
Available online 23 July 2004

Abstract
To prepare zinc-substituted spinel-type ferrite ne particles of M1xZnxFe2O4 (M=Mg, Mn, Co, Ni, Cu, (Li, Fe)
x=01) with good crystallinity and stoichiometry, the authors investigated a glycinenitrate process. The product
powder was an agglomerate of ne particles whose typical diameter was several tens of nanometers. X-ray diffraction
patterns revealed that the produced particles were mono-phase in almost all reaction systems. Energy-dispersive X-ray
spectroscopy microanalysis of the product particles (MnZnFeO) revealed that the distributions of Mn/Fe ratio and
Zn/Fe ratio were highly sharp both within the agglomerate and between agglomerates.
r 2004 Elsevier B.V. All rights reserved.
PACS: 75.50.y; 75.50.Gg; 81.20.Ka
Keywords: Combustion synthesis; Fine particles; Magnetism; Spinel ferrite; Crystallinity; Stoichiometry

1. Introduction
Magnetic ne particles have attracted much
interest in many elds [15]. One application of
magnetic materials utilizes their heating effects, i.e.
magnetic hysteresis energy losses. For example,
there are a considerable number of works on
Corresponding author. Tel: +81-298618490; fax: +81-

298618459.
E-mail address: kikukawa-n@aist.go.jp (N. Kikukawa).

hyperthermia [6,7], which is one of the promising


approaches in cancer therapy. In this application,
the magnetic materials used are ne particles with
coercive forces suitable for use with applied
alternating magnetic elds.
We are planning to apply the heating effects of
magnetic materials to a new environmental application, which will be described in the near future.
In this study, we attempted to prepare zincsubstituted spinel-type ferrite ne particles with a
coercive force and Curie point that will work with

0304-8853/$ - see front matter r 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.jmmm.2004.06.039

ARTICLE IN PRESS
N. Kikukawa et al. / Journal of Magnetism and Magnetic Materials 284 (2004) 206214

the new application. Such particles would have to


be several tens of nanometers in diameter, and
their zinc contents must be controlled. These
characteristics are necessary because superparamagnetic particles (typically smaller than 10 nm)
do not show magnetic hysteresis around room
temperature. In general, the coercive forces of
magnetic particles have their maximum
when the diameter is between 20 and 50 nm [8].
Moreover,
since
zinc
content
strongly
affects the Curie point in spinel-type ferrite,
the stoichiometry in each ne particle is
important.
Various wet-methods [919] have been reported
for preparing spinel-type ferrite ne particles, such
as coprecipitation methods [9,10], hydrothermal
synthesis [11,12], microemulsion synthesis [1315],
and solgel methods [16]. In addition, various drymethods [2025] including grinding [21], mechanical alloying [22], and thermal plasma methods
[23,24] have been employed. Beside these
methods, there have been several attempts
using auto-ignited combustion reactions [2629].
Among them, a glycinenitrate process (GNP)
[2931] is also applicable, which is a very
simple, low-cost, fast method for preparing
well-crystallized double oxides in about 30 min
from preparing the precursor solution to
obtaining nal products. The production
speed of the combustion synthesis seems to be
as high as that of microwave-hydrothermal
synthesis [12]. However, the apparatus needed
for the former is far less expensive than that of the
latter.
Since, there have been only a few studies on the
preparation of spinel-type ferrites by GNP [29], we
investigated the synthesis and the properties of
various zinc-substituted spinel-type ferrites
(M1xZnxFe2O4, M=Mg, Mn, Co, Ni, Cu, (Li,
Fe) x=01) using this method.
In the GNP, the amino acid, glycine, prevents
selective precipitation from the precursor solution,
and it serves as fuel for the combustion
reaction, being oxidized by the nitrate ions [30e].
The main controllable processing variable is the
glycine-to-nitrate ion (G/N) ratio, which affects
the ame temperature, the combustion velocity,
and the product morphology and composition.

207

The peak ame temperature is said to be typically


obtained at a G/N ratio that corresponds to
complete combustion, producing H2O, CO2, and
N2 as the waste gases, with no atmospheric oxygen
required [30e]. The G/N ratio that corresponds to
complete combustion is called the stoichiometric
ratio [30e]. For the spinel-type ferrite synthesis
from tri-valent iron nitrate (Fe(NO3)3) and divalent metal nitrate (M(NO3)2), the stoichiometric
G/N molar ratio is 59; as shown in the following
equation:
18FeNO3 3 9MNO3 2 40NH2 CH2 COOH !
9MFe2 O4 56N2 80CO2 100H2 O
The optimum powder product is not always
obtained at the stoichiometric G/N ratio [30e]. In
this study, we changed the G/N ratio from 0.17 to
0.83 and examined the effects of altering the G/N
ratio on the morphology, crystal structure, and
magnetic properties.
2. Experimental
In an appropriate ratio, analytical-grade nitrates
Fe(NO3)3  9H2O, Zn(NO3)2  6H2O, Mg(NO3)2 
6H2O, Mn(NO3)2  6H2O, Co(NO3)2  6H2O, Ni(NO3)2  6H2O, Cu(NO3)2  3H2O, LiNO3 and
glycine (H2NCH2COOH) were dissolved in distilled water to obtain the precursor solution.
The total concentration of the metal ions was
approximately 1 mol/l. About 50 ml of the
precursor solution was put into a two-liter
stainless-steel beaker on a hot plate within a
draft chamber. The beaker was covered with a
stainless-steel mesh screen. The hot plate was set
on high, and the solution was simply allowed to
boil until it ignited. Within several tens seconds,
the combustion reaction produced brownto-black, porous products that were easily
ground to powder. The product powder was
characterized by an X-ray diffractometer (XRD,
Rigaku RU-300), a transmission electron microscope (TEM, Philips CM-30) equipped with an
energy-dispersive X-ray spectrometer (EDX), and
a vibrating sample magnetometer (VSM, Riken
Denshi BHV-55).

ARTICLE IN PRESS
208

N. Kikukawa et al. / Journal of Magnetism and Magnetic Materials 284 (2004) 206214

3. Results and discussion


3.1. Morphology and uniformity of composition

BET specific area [m2/g] and


crystallite size [nm]

The typical TEM images of the product powder


of Mn0.5Zn0.5 Fe2O4 are shown in Figs. 1(a) (low
magnication) and (b) (high magnication). The
G/N ratio of the precursor was 0.44. From the
gures, one can see that the product powder
consists of agglomerates of ne primary particles.
The size distribution of primary particles in Fig. 1b
showed lognormal distribution and the average
particle size was 51 nm. However, it should be
noted that, under current experimental conditions,
there were some variations in the crystallinity of
primary particles; sometimes highly grown crystal-

lites were observed and sometimes amorphous-like


particles were also observed. Nevertheless, Figs. 1a
and b are the most frequently observed makeup of
the products.
The crystallite size deduced from XRD peak
width is considered to show the average gure. As
shown in Fig. 2, the crystallite size and the BET
specic surface area (SBET) of the powder produced from the precursor of Co:Fe=1:2 were
dependent on the G/N ratio. The crystallite size
was the maximum and SBET was the minimum at a
G/N ratio of 59; which corresponds to the stoichiometric ratio. The crystallite sizes show fairly good
agreement with the TEM diameters. However,
when we calculated the average diameter from
SBET with the assumption of perfect dispersion of
primary particles, the average diameters were
several times larger than the corresponding crystallite sizes, which implies that the primary
particles were somewhat necked together during
the GNP reaction. Another example of a TEM
image is shown in Fig. 3. From the gure, it seems
that the agglomerate was generated through the
process where the primary particles gathered and
necked to each other, rather than through crystallization from liquid melt.
Since the zinc contents of the zinc-substituted
spinel-type ferrite strongly affect the Curie point,
the stoichiometry of each particle is important.
Therefore, we measured the Mn/Fe and Zn/Fe
50
40
30

BET
Crystallite
size

20
10
0
0.1

0.2

0.3

0.4

0.5

0.6

0.7

G/N ratio [-]

Fig. 1. Typical TEM images of product powder of


Mn0.5Zn0.5Fe2O4. (a) low magnication, (b) high magnication.
Run No. 16, G/N ratio: 0.44.

Fig. 2. Dependency of the BET specic surface areas and the


crystallite sizes of the powder produced from the precursor of
Co:Fe=1:2 upon glycine/nitrate ratio. circle: BET specic
surface area, square: crystallite size deduced from the peak
width of the XRD pattern.

ARTICLE IN PRESS
N. Kikukawa et al. / Journal of Magnetism and Magnetic Materials 284 (2004) 206214

209

Fig. 3. TEM image of the powder produced from the precursor


of Mn:Zn:Fe=1:1:4 with a G/N ratio of 0.67 (run no. 51).

ratios of individual particles corresponding to Fig.


1 by EDX microanalysis. Fig. 4. shows the results
of the distributions between agglomerates (a) and
between primary particles within an agglomerate
(b). The results of that of coprecipitated powder
are shown in (c) for comparison. As the standard
deviations are small enough compared with those
of coprecipitated powder, we can conclude that the
GNP yielded stoichiometric primary particles and
agglomerates.
3.2. Effects of the G/N ratio on crystal structure
All precursors with various G/N ratios examined were found to auto-ignite. The yields were
almost 100%. A higher G/N ratio often resulted in
a mixed phase with ferrite and a small portion of
low-valent phases such as FeO and/or metals or
alloys (in the case of Ni, Cu, Co and their zincsubstituted systems), and a lower G/N ratio
resulted in a higher oxidized phase of Mn3O4 (in
the case of Mn and MnZn systems with G/N
ratio less than 0.3). Generally, the precursors with
a G/N ratio less than about 0.5 generated monophase spinel-type ferrites. The XRD patterns of
some examples are shown in Fig. 5.
In the case of copper, the products were always
a mixture of cubic type spinel phase and various
Cu-containing phases such as CuFeO2, Cu2O,
CuO, and Cu metal, as shown in Fig. 6a. We

Fig. 4. Histograms of Mn/Fe and Zn/Fe ratios (a) between


agglomerates, (b) within agglomerates, and (c) of co-precipitated particles calcined at 873 K, measured through EDX. The
initial metal composition ratio was Mn:Zn:Fe=1:1:4.

added ammonium nitrate, which acts as both fuel


and oxidant and is said to raise the ame
temperature [30], to the precursor solution.
Although a small fraction of CuO remained, the
product became nearly the mono-phase of cuprospinel, as shown in Fig. 6b, when the molar ratio of
ammonium nitrate to nitrate ion and G/N ratio
were 0.5 and 0.33, respectively.
3.3. Effects of the G/N ratio on magnetic properties
Figs. 7a and b show the dependency of magnetic
properties on the G/N ratio in the cases of

ARTICLE IN PRESS
210

N. Kikukawa et al. / Journal of Magnetism and Magnetic Materials 284 (2004) 206214

Fig. 5. Examples of the XRD patterns of synthesized


M1xZnxFe2O4 (M=Mg, Mn, Co, Ni, x=0, 0.3, 0.5) particles.
The marks indicate peaks of spinel-type ferrites.

CoFeO, MnZnFeO and CuZnFeO. The


saturation magnetizations (Fig. 7a) show their
maximum at around a G/N ratio of 0.40.5, just
below the stoichiometric ratio of 59: The initial
increase of the saturation magnetization can be
attributed to the improved crystallinity, because
the crystallite size gradually increases with the G/
N ratio (see Fig. 2). When the G/N ratio is greater
than about 0.5, as the product powder becomes the
mixture of magnetic spinel phase and other phases
such as FeO and alloys, the saturation magnetization tends to decrease gradually other than in the
cases of CoFeO, in which coexisting cubic Co
phase is ferromagnetic.

Fig. 6. XRD patterns in the case of the CuZnFeO system.


(a) The product powder with a different G/N ratio. (b) The
product powder synthesized with ammonium nitrate.

On the other hand, the coercive forces (Fig. 7b)


show rather complicated dependency on the G/N
ratio. The dependency of the crystallite size of
spinel-type ferrite phases on the G/N ratio (Fig. 2)
is considered to affect mainly. The low coercive
forces at the lower G/N ratio should be attributed
to superparamagnetism. CoFe2O4 is known to
have very small superparamagnetic critical size of
14 nm [32] and the rst two points in Fig. 7b were
less than the critical size (see Fig. 2). Since another
ferrites have larger critical sizes [32], the coercive

ARTICLE IN PRESS
N. Kikukawa et al. / Journal of Magnetism and Magnetic Materials 284 (2004) 206214

forces of Mn0.5Zn0.5Fe2O4 (Fig. 7b), and others


such as MgFe2O4 (not shown) gradually increased
with the G/N ratio. However, CuZnFeO
system showed more complicated dependency.
As mentioned above, the GNP products of

211

CuZnFeO system always consisted of mixed


Cu-containing crystal phases and we do not have
any quantitative phase composition data yet.
Although, we think the coexisting magnetic and/
or nonmagnetic phases may affect this dependency, the detail of the CuZnFeO system is a
future subject to be solved.
Table 1 summarizes the magnetic properties of
produced MFe2O4 (M=Mg, Mn, Ni, Co, Cu, (Li,
Fe)) with a G/N ratio of around 0.4. The literature
values [33] are also included in the table for
comparison. Since the Curie points and the lattice
constants show fairly good agreement, we believe
the purities of magnetic crystal phase are high.
Regarding saturation magnetization, GNP powder
had slightly lower values than the literature ones,
probably because the crystallites have not fully
grown during the GNP.
Coercive force will be important in our planned
application using heating effect of magnetic
materials. Fig. 8a shows the magnetic hysteresis
curves in a magnetic eld of 100 Oe of MnZn
ferrite synthesized by the GNP (run no.G31; solid
line) and by the conventional ceramic method
(commercial powder; broken line). We preliminarily investigated the heating behaviors of these two
MnZn ferrite powders under a 56-kHz AC
magnetic eld of 100 Oe. Fig. 8b shows a
preliminary result. The GNP-synthesized powder
showed a far steeper rise than that of the

Fig. 7. Dependency of magnetic properties on G/N ratio at the


magnetic eld of 5 kOe: (a) saturation magnetization, (b)
coercive force. Circle: CoFeO (Co:Fe=1:2), square:
MnZnFeO (Mn:Zn:Fe=1:1:4), triangle: CuZnFeO
(Cu:Zn:Fe=0.7:0.3:2).

Table 1
Magnetic properties of synthesized spinel-type ferrite particles
Run
No.

LiFe5O8
MgFe2O4
MnFe2O4
CoFe2O4
NiFe2O4
CuFe2O4
ZnFe2O4

G66
G34
G24
G53
G08
G25
G49

Experimental results

Literature

Lattice
constant
(nm)

Saturation
magnetization
(emu/g)

Coercive
force (Oe)

Curie point Lattice constant


(K)
(nm)a

Saturation
magnetization
(emu/g)b

Curie point
(K)b

0.8339
0.8391
0.8499
0.8386
0.8349
0.8382
0.8435

57
32
72
79
42
49
7.3

155
72
89
885
183
149
19

903
696
625
797
868
749

69
31
112
94
56
30

903
713
573
793
858
728

0.8333
0.83873
0.8499
0.83919
0.8339
0.8369
0.84411

(17-114)
(36-398)
(10-319)
(22-1086)
(10-325)
(25-283)
(22-1012)

Saturation magnetization and coercive force were measured at the magnetic eld of 15 kOe and at room temperature.
a
Powder diffraction le (PDF).
b
Ref. [33].

ARTICLE IN PRESS
212

N. Kikukawa et al. / Journal of Magnetism and Magnetic Materials 284 (2004) 206214

glycinenitrate process (GNP) and drew the


following conclusions:
(1) We successfully synthesized mono-phase
spinel-type ferrites when the G/N ratio was
less than about 0.5 for the systems including
Mg, MgZn, Co, CoZn, Ni, NiZn, (Li, Fe),
and (Li, Fe)Zn, and the ratio between 0.3
and 0.5 for Mn, and MnZn systems.
(2) TEM observation showed that the product
powder consisted of agglomerates of primary
particles with a typical diameter was about
50 nm for a G/N ratioof around 0.4.
(3) EDX microanalysis of the product particles
revealed that the distributions of the elemental ratios were very sharp both within an
agglomerate and between agglomerates.

Fig. 8. Comparison of synthesized MnZn ferrite powder with


a commercial powder: (a) magnetic hysteresis curves in a
magnetic eld of 100 Oe at room temperature, (b) preliminary
measurements of temperature vs. time curves under an AC
magnetic eld (56 kHz, 100 Oe). Solid line: synthesized
Mn0.5Zn0.5Fe2O4 powder (run no. G31, Curie point: 501 K).
Broken line: commercial MnZn ferrite powder (Curie point:
523 K).

commercial powder, which reects the larger


hysteresis-loop area (see Fig. 8a). We are now
conducting a study on the heating behavior of
various GNP-synthesized ferrites under an extensive AC magnetic eld

4. Conclusion
In order to prepare zinc-substituted spinel-type
ferrite ne particles of M1xZnxFe2O4 (M=Mg,
Mn, Co, Ni, Cu, (Li, Fe) x=01) with good
crystallinity and stoichiometry, we investigated a

The GNP was found to be a rapid, low-cost


method for synthesizing spinel-type ferrites.
Although the nanometer-sized primary crystallites
were always somewhat necked together, the fact
will not affect our planning application using
heating effects of magnetic hysteresis energy losses,
since the coercive force does not depend on the
agglomerate size but on the crystallite size. Therefore, we conclude that the GNP is one of the
suitable methods for high-speed preparation of
various spinel-type ferrites.

References
[1] K. Raj, B. Moskowitz, R. Casciari, J. Magn. Magn.
Mater. 149 (1995) 174.
[2] D.L. Leslie-Pelecky, R.D. Rieke, Chem. Mater 8 (1996)
1770.
[3] S.A. Majetich, J.H. Scott, E.M. Kirkpatrick, K. Chowdary, K. Gallagher, M.E. McHenry, Nanostruct. Mater 9
(1997) 291.
[4] R.H. Kodama, J. Magn. Magn. Mater. 200 (1999) 359.
[5] M. Sugimoto, J. Am. Ceram. Soc. 82 (1999) 269.
[6] P. Moroz, S.K. Jones, B.N. Gray, Int. J. Hyperthermia 18
(2002) 267 and references therein.
[7] (a) R. Hergt, W. Andrae, C.G. dAmbly, I. Hilger, W.A.
Kaiser, U. Richter, H.-G. Schmidt, IEEE Trans.
Magn. 34 (1998) 3745;
(b) A. Jordan, R. Scholz, K. Maier-Hauff, M. Johannsen,
P. Wust, J. Nadobny, H. Schirra, H. Schmidt, S.
Deger, S. Loening, W. Lanksch, R. Felix, J. Magn.
Magn. Mater. 225 (2001) 118;

ARTICLE IN PRESS
N. Kikukawa et al. / Journal of Magnetism and Magnetic Materials 284 (2004) 206214

[8]

[9]
[10]

[11]

[12]

[13]

[14]
[15]

[16]

(c) R.E. Rosensweig, J. Magn. Magn. Mater. 252 (2002)


370.
S. Chikazumi, K. Ohta, K. Adachi, N. Tsuya, Y. Ishikawa.
(Eds.), JISEITAI HANDOBUKKU, Asakura, 1975, p.
1164, (in Japanese).
P.E.D. Morgan, J. Am. Ceram. Soc. 57 (1974) 499 and
references therein.
(a) M. Kiyama, Bull. Chem. Soc. Japan. 51 (1978) 134;
(b) R. Massart, IEEE Trans. Magn. 17 (1981) 1247;
(c) H. Tamura, E. Matijevic, J. Colloid Interface Sci. 90
(1982) 100;
(d) Z.X. Tang, C.M. Sorensen, K.J. Klabunde, G.C.
Hadjipanayis, J. Colloid Interface Sci. 146 (1991) 38;
(e) R.V. Upadhyay, K.J. Davies, S. Wells, S.W. Charles,
J.Magn. Magn. Mater. 132 (1994) 249;
(f) J.P. Chen, C.M. Sorensen, K.J. Klabunde, G.C.
Hadjipanayis, E. Devlin, A. Kostikas, Phys. Rev. B
54 (1996) 9288;
(g) Q. Chen, A.J. Rondinone, B.C. Chakoumakos, Z.J.
Zhang, J. Magn. Magn. Mater. 194 (1999) 1.
(a) S. Komarneni, E. Fregeau, E. Breval, R. Roy, J. Am.
Ceram. Soc. 71 (1988) 6;
(b) C. Rath, K.K. Sahu, S. Anand, S.K. Date, N.C.
Mishra, R.P. Das, J. Magn. Magn. Mater. 202 (1999)
77;
(c) A. Caban, M. Poliakoff, J. Mater. Chem. 11 (2001)
1408;
(d) J.-S. Kim, J.-R. Ahn, C.W. Lee, Y. Murakami, D.
Shindo, J. Mater. Chem. 11 (2001) 3373.
(a) C.K. Kim, J.-H. Lee, S. Katoh, R. Murakami, M.
Yoshimura, Mater. Res. Bull. 36 (2001) 2241;
(b) S. Komarneni, M.C. DArrigo, C. Leonelli, G.C.
Pellacani, H. Katsuki, J. Am. Ceram. Soc. 81 (2002)
3041;
(c) S. Verma, P.A. Joy, Y.B. Khollam, H.S. Potdar, S.B.
Deshpande, Mater. Lett. 58 (2004) 1092.
(a) M.P. Pileni, N. Moumen, P. Veillet, J. Magn. Magn.
Mater. 149 (1995) 67;
(b) N. Moumen, M.P. Pileni, J. Phys. Chem. B 100 (1996)
1867;
(c) A.T. Ngo, P. Bonville, M.P. Pileni, Eur. Phys. J. B 9
(1999) 583;
(d) C. Liu, A.J. Rondinone, Z.J. Zhang, Pure Appl. Chem.
72 (2000) 37;
(e) J.F. Hochepied, M.P. Pileni, J. Appl. Phys. 87 (2000)
2472;
(f) M.L. Kahn, Z.J. Zhang, Appl. Phys. Lett. 78 (2001) 23.
C.R. Vestal, Z.J. Zhang, Chem. Mater. 14 (2002) 3817.
(a) E.E. Carpenter, C.J. OConnor, V.G. Harris, J. Appl.
Phys. 85 (1999) 5175;
(b) M. Zheng, X.C. Wu, B.S. Zou, Y.J. Wang, J. Magn.
Magn. Mater. 183 (1998) 152;
(c) C. Liu, B. Zou, A.J. Rondinone, Z.J. Zhang, J. Phys.
Chem. B 104 (2000) 1141;
(d) C. Liu, Z.J. Zhang, Chem. Mater. 13 (2001) 2092.
(a) S.K. Saha, A. Pathak, P. Pramanik, J. Mater. Sci. Lett.
14 (1995) 35;

[17]

[18]

[19]
[20]
[21]

[22]

[23]

[24]
[25]
[26]
[27]

213

(b) X. Li, G. Lu, S. Li, J. Alloys Compounds 235 (1996)


150;
(c) A. Verma, T.C. Goel, R.G. Mendiratta, R.G. Gupta,
J. Magn. Magn. Mater. 192 (1999) 271;
(d) G. Xiong, Z. Mai, M. Xu, S. Cui, Chem. Mater. 13
(2001) 1943;
(e) S.W. Lee, Y.G. Ryu, K.J. Yang, K.-D. Jung, S.Y. An,
C.S. Kim, J. Appl. Phys. 91 (2002) 7610;
(f) R.N. Panda, J.C. Shih, T.S. Chin, J. Magn. Magn.
Mater. 257 (2003) 79.
S. Li, V.T. John, S.H. Rachakonda, G.C. Irvin, G.L.
McPherson, C.J. OConnor, J. Appl. Phys. 85 (1999)
5178.
S. Thimmaiah, M. Rajamathi, N. Singh, P. Bera, F.
Meldrum, N. Chandrasekhard R. Seshadri, J. Mater.
Chem. 11 (2001) 3215.
T. Hyeon, Y. Chung, J. Park, S.S. Lee, Y.-W. Kim, B.H.
Park, J. Phys. Chem. B 106 (2002) 6831.
R. Muller, W. Schuppel, J. Magn. Magn. Mater. 155
(1996) 110.
(a) R.H. Kodama, A.E. Berkowitz, E.J. McNiff Jr., S.
Foner, Phys. Rev. Lett. 77 (1996) 394;
(b) G.F. Goya, H.R. Rechenberg, J.Z. Jiang, J. Magn.
Magn. Mater. 218 (2000) 221;
(c) C.N. Chinnasamy, A. Narayanasamy, N. Ponpandian,
R.J. Joseyphus, K. Chattopadhyay, K. Shinoda, B.
Jeyadevan, K. Tohji, K. Nakatsuka, J.-M. Greneche, J.
Appl. Phys. 90 (2001) 527;
(d) C.N. Chinnasamy, A. Narayanasamy, N. Ponpandian,
R.J. Joseyphus, B. Jeyadevan, K. Tohji, K. Chattopadhyay, J. Magn. Magn. Mater. 238 (2002) 281.
(a) D. Arcos, R. Valenzuela, M. Vazquez, M. Vallet-Regi,
J. Solid State Chem. 141 (1998) 16;
(b) S.F. Moustafa, M.B. Morsi, Mater. Lett. 34 (1998)
241;
(c) J.S. Jiang, L. Gao, X.L. Yang, J.K. Guo, H.L. Shen, J.
Mater. Sci. Lett. 18 (1999) 1781;
(d) D.J. Fatemi, V.G. Harris, M.X. Chen, S.K. Malik,
W.B. Yelon, G.J. Long, A. Mohan, J. Appl. Phys. 85
(1999) 5172;
(e) G.F. Goya, H.R. Rechenberg, J. Magn. Magn. Mater.
203 (1999) 141;
(f) S.H. Gee, Y.K. Hong, M.H. Park, D.W. Erickson, P.J.
Lamb, J.C. Sur, J. Appl. Phys. 91 (2002) 7586.
(a) T.W. Or, P.C. Kong, E. Pfender, Plasma Chem.
Plasma Process. 12 (1992) 177;
(b) S. Son, M. Taheri, E. Carpenter, V.G. Harris, M.E.
McHenry, J. Appl. Phys. 91 (2002) 7589;
(c) Y. Kinemuchi, K. Ishizaka, H. Suematsu, W. Jiang, K.
Yatsui, Thin Solid Films 407 (2002) 109.
N. Kikukawa, M. Sugasawa, S. Kobayashi, Jpn. J. Appl.
Phys. 41 (2002) 5991.
X. Zhao, B. Zheng, H. Gu, C. Li, S.C. Zhang, P.D.
Ownby, J. Mater. Res. 14 (1999) 3073.
K. Suresh, K.C. Patil, J. Mater. Sci. Lett. 13 (1994) 1712.
(a) Z. Gao, T. Wu, S. Peng, J. Mater. Sci. Lett. 13 (1994)
1715;

ARTICLE IN PRESS
214

N. Kikukawa et al. / Journal of Magnetism and Magnetic Materials 284 (2004) 206214

(b) Y.-P. Fu, C.-H. Lin, J. Magn. Magn. Mater. 251


(2002) 4;
(c) A.C.F.M. Costaa, E. Tortellab, M.R. Morellib,
R.H.G.A. Kiminami, J. Magn. Magn. Mater. 256
(2003) 174.
[28] (a) Z. Yue, L. Li, J. Zhou, H. Zhang, Z. Gui, Mater. Sci.
Engineer. B 64 (1999) 68;
(b) Z. Yue, J. Zhou, L. Li, H. Zhang, Z. Gui, J. Magn.
Magn. Mater. 208 (2000) 55;
(c) H. Zhang, Z. Ma, J. Zhou, Z. Yue, L. Li, Z. Gui, J.
Magn. Magn. Mater. 213 (2000) 304.
[29] C.-H. Yan, Z.-G. Xu, F.-X. Cheng, Z.-M. Wang, L.-D.
Sun, C.-S. Liao, J.-T. Jia, Solid State Commun. 111 (1999)
287.
[30] (a) L.A. Chick, L.R. Pederson, G.D. Maupin, J.L. Bates,
L.E. Thomas, G.J. Exarhos, Mater. Lett. 10 (1990) 6;

(b) L.R. Pederson, G.D. Maupin, W.J. Weber, D.E.


McCready, R.W. Stephens, Mater. Lett. 10 (1991) 437;
(c) J.J. Kingsley, L.R. Pederson, Mater. Lett. 18 (1993) 89;
(d) N.J. Hess, G.D. Maupin, L.A. Chick, D.S. Sunberg,
D.E. McCready, T.R. Armstrong, J. Mater. Sci. 29
(1994) 1873;
(e) L.A. Chick, G.D. Maupin, L.R. Pederson, Nanostruct.
Mater. 4 (1994) 603.
[31] M.R. Murphy, T.R. Armstrong, P.A. Smith, J. Am.
Ceram. Soc. 80 (1997) 165.
[32] T. Sato, T. Iijima, M. Seki, N. Inagaki, J. Magn. Magn.
Mater. 65 (7) 252.
[33] S. Chikazumi, K. Ohta, K. Adachi, N. Tsuya, Y. Ishikawa
(Eds.), JISEITAI HANDOBUKKU, Asakura, 1975,
p. 612, (in Japanese).

Physica B 407 (2012) 795804

Contents lists available at SciVerse ScienceDirect

Physica B
journal homepage: www.elsevier.com/locate/physb

Infrared and structural studies of Mg1xZnxFe2O4 ferrites


K.A. Mohammed a,n, A.D. Al-Rawas b, A.M. Gismelseed b, A. Sellai b, H.M. Widatallah b, A. Yousif b,
M.E. Elzain b, M. Shongwe c
a

Department of Mathematical and Physical Sciences, College of Arts & Sciences, University of Nizwa, P.O. Box 33, Code 616, Oman
Department of Physics, College of Science, Sultan Qaboos University, P.O. Box 36, Code 123, Al-khoud, Oman
c
Department of Chemistry, College of Science, Sultan Qaboos University, P.O. Box 36, Code 123, Al-khoud, Oman
b

a r t i c l e i n f o

abstract

Article history:
Received 23 August 2011
Received in revised form
10 December 2011
Accepted 13 December 2011
Available online 21 December 2011

Compositions of polycrystalline MgZn mixed ferrites with the general formula Mg1  xZnxFe2O4
(0 r x r 1) were prepared by the standard double sintering ceramic method. The structural properties
of these ferrites have been investigated using X-ray diffraction and infrared absorption spectroscopy.
The lattice parameter, particle size, bonds length, force constants, density, porosity, shrinkage and
cation distribution of these samples have been estimated and compared with those predicted
theoretically. Most of these values were found to increase with increasing Zn content. The energy
dispersive (EDS) analysis conrmed the proposed sample composition. The scanning electron microscope (SEM) and transmission electron microscope (TEM) micrographs showed aggregates of stacked
crystallites of about 200800 nm in diameter. Far infrared absorption spectra showed two signicant
absorption bands. The wave number of the rst band, n1, decreases with increasing Zn content, while
the band, n2 shifts linearly towards higher wave numbers with Zn contents, over the whole composition
range. The room temperature electrical resistivity was found to decrease as Zn-content increases.
Values of the vacancy model parameters showed that the packing factors Pa and Pb decrease, the
fulllment coefcient, a, remains almost constant and the vacancy parameter, b, strongly increases
with increasing Zn content in the sample. The small values of Pa, Pb, a and the strong increase of the
vacancy parameter, b, indicate the presence of cation or anion vacancies and the partial participation of
the Zn2 vacancies in the improvement of the electrical conductivity in the MgZn ferrites.
& 2011 Elsevier B.V. All rights reserved.

Keywords:
MgZn Ferrites
Spinel
IR
XRD
Electrical resistivity
Force constants
Vacancy model parameters

1. Introduction
Ferrites have been the subject of extensive investigation
because of their wide range applications such as information
storage systems, gas sensors, microwave devices and magnetic
recording and electronic industries. These materials are characterized by high electrical resistivity and unique magnetic and
electrical properties, which result in low eddy currents and
dielectric losses [13]. The crystalline structure of the spinel
ferrites has two sites, the tetragonal A-sites and octahedral
B-sites. These materials have the general chemical formula

3
2
3
2
(TM21 
. The factor d represents the fracdFed ) [TMd Fe2  d]O
3
tion of Fe
ions at the A-sites. Magnesium ferrite, MgFe2O4, is
predominantly an inverse spinel and the degree of inversion
depends upon heat treatment [4]. The zinc ferrite, ZnFe2O4, has
the normal spinel structure, in which the Zn2 ions occupy the
tetrahedral sites where the octahedral B-sites are occupied by the
Fe3 ions. The IR spectra are important tools to investigate the

Corresponding author. Tel.: 968 25446434; fax: 968 225446289.


E-mail address: Kadhimm@unizwa.edu.om (K.A. Mohammed).

0921-4526/$ - see front matter & 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.physb.2011.12.097

structural properties of the ferrite materials. They give information about the positions of the ions in the spinel lattice, the
vibration modes, the valance state of the ions and their occupation in the spinel lattice crystal. Moreover, the IR spectra scans
can be used to detect the completion of the solid state reaction,
cations distribution and the deformation of spinel structure. The
IR spectra absorption bands mainly appear due to vibrations of
the oxygen ions with the cations producing various frequencies in
the unit cell. The vibration frequency depends on the cations
mass, cationoxygen distance and the bonding force [5].
The vacancy model [6] describes the inuence of vacancies and
mixed valence on the transport processes in solid solutions with
the spinel structure. This model deals with four quantities; the
coefcients of ion packing at the tetrahedral and octahedral sites,
the fulllment coefcient of the unit cell, which determines the
degree of the ionic packing of the spinel structure, the difference
of Pauling electronegativities, which give an evaluation of the
ionic bonds strength in the tetrahedral and octahedral sites and
the vacancy parameter, which is a measure of the total vacancy
concentration in the sample.
Moreover, distinct relations were reported [7] to exist between
the ionic packing factors at the tetrahedral and octahedral sites

796

K.A. Mohammed et al. / Physica B 407 (2012) 795804

and the electrical conductivity and the magnetic ordering at these


sites, respectively.
The aims of this study are to investigate the effect of Zn
substitution on X-ray parameters, IR absorption bands, room
temperature electrical resistivity and to test the validity of the
vacancy model on the relevant parameters of the mixed
Mg1  xZnxFe2O4 ferrites.

2. Experimental
Samples of the spinel series Mg1  xZnxFe2O4 (x 01 step 0.1)
were prepared using the conventional double sintering technique,
in which powders of high purity MgO, ZnO and Fe2O3 oxides were
mixed in required proportions in agate mortar. The mixture was
rst sintered at 1000 1C for 24 h in air medium followed by
cooling to room temperature. The powder was then remixed and
ground once more to promote homogeneity. Pellets of 13 mm and
23 mm thick are prepared using hydraulic press of 10 tons/cm2.
These pellets and the rest of the powder were sintered at 1050 1C
for 24 h followed by natural cooling to room temperature. The
nal total losses in sample masses were negligible (less than
0.1%). Moreover, the samples compositions were conrmed by
Mossbauer and the transmission electron microscope analysis.
The single-phase spinel structure was conrmed by the X-ray
diffraction spectrum of these samples. The X-ray diffraction
spectrum of these samples were examined using Phillips
The
PW1820 diffractometer with CuKa radiation (l 1.5404 A).
scans range was kept the same for all samples 2y 101001 using
a step size of 0.021 with sample time of 2 s.
Samples for recording IR spectra were prepared by mixing
small quantity of the powder of the samples with solid KBr. The
mixed powder of samples was then placed in a cylindrical disc and
pressed at 10 tons/cm2 by a hydraulic press. The IR measurements
of the prepared samples were recorded at room temperature in
the range from 400 up to 1000 cm  1 using PERKIN-ELMER-1430
infrared Spectrophotometer.
Samples for the DC resistance measurements at room temperature were prepared by rst grinding the surfaces to remove
any oxide, such as Fe2 , which might be caused by the distillation
off the sample surface, such as the Zn2 , during sintering at high
temperatures and to remove any contamination to the pellet
surfaces during the pressing process [8]. The pellets surfaces were
coated with a silver paste conductor to ensure a good electrical
contact between the sample and the two copper terminals for the
electrical resistance measurements.
The microstructure and sample morphology were examined
with analytical scanning electron microscope (ASEM) model
JSM-6510LA-JEOL and transmission electron microscope model
JEM-1400-JEOL.

3. Results and discussions


The X-ray diffraction patterns for the Mg1  xZnxFe2O4 samples
are shown in Fig. 1. The analysis of XRD patterns conrms the
formation of spinel cubic structure. The reections from the
plains (1 1 1), (2 2 0), (3 1 1), (2 2 2), (4 4 0), (4 2 2), (5 1 1),
(6 2 0), (5 3 3), (6 2 2), (4 4 4) and (5 3 3) are shown by all
samples. This is an indication of the spinel cubic single-phase
formation. The lattice parameters aexp for all samples have been
determined from different diffraction lines of the XRD patterns
employing the Wint-XRD software program. Values of aexp are
tabulated in Table 1(Fig. 3). as a function of the lattice parameter
with Zn content. It is clear that the lattice constant increases
smoothly with increase in Zn content. The increase in lattice

Fig. 1. The X-ray diffraction pattern for the Mg1  xZnxFe2O4 samples.

Table 1
Values of rA, rB, aexp, ath and cation distribution of Mg1  xZnxFe2O4 ferrites.
x

aexp (A)

ra (A)

rb (A)

ath (A)

Cation distributions (Ref. [15])

0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0

8.378
8.393
8.405
8.409
8.417
8.421
8.430
8.439
8.442
8.448
8.450

0.6697
0.6840
0.6992
0.7142
0.7293
0.7445
0.7595
0.7746
0.7897
0.8047
0.8200

0.6652
0.6660
0.6664
0.6669
0.6674
0.6678
0.6683
0.6687
0.6692
0.6697
0.6700

8.483
8.508
8.532
8.556
8.581
8.605
8.630
8.654
8.679
8.703
8.728

(Mg.03Fe.97)A [Mg.97Fe1.03]B
(Zn.1Mg.1Fe.8)A [Mg.8Fe1.2]B
(Zn.2Mg.08Fe.72)A [Mg.72Fe1.28]B
(Zn.3Mg.08Fe.62)A [Mg.62Fe1.38]B
(Zn.4Mg.07Fe.53)A [Mg.53Fe1.47]B
(Zn.5Mg.05Fe.45)A [Mg.45Fe1.55]B
(Zn.6Mg.05Fe.35)A [Mg.35Fe1.65]B
(Zn.7Mg.04Fe.26)A [Mg.26Fe1.74]B
(Zn.8Mg.035Fe.165)A [Mg.165Fe1.835]B
(Zn.9Mg.03Fe.07)A [Mg.07Fe1.93]B
(Zn1)A [Fe2]B

parameters is attributed to the replacement of smaller Fe3


ions with a larger ionic radius of the Zn2 (0.82 A),
at
(0.67 A)

the tetrahedral sites, and the replacement of Mg2 ions (0.66 A)


with a larger ionic radius Fe3 ion at the octahedral sites. The
nonlinear behavior of lattice parameter aexp with Zn content (i.e.
violating Vegards law) was reported for other ferrite systems,
which deviate from the complete normal or complete inverse
spinel type structure [9]. Values of the lattice parameter aexp for
MgFe2O4 and ZnFe2O4 samples have been calculated as 8.378 and
respectively, which agree quite well with literature
8.450 A,
values of 8.387 [10], 8.384 [11] and 8.377 A [12] for MgFe2O4
and 8.437 [13] and 8.33 A [14] for ZnFe2O4. The larger value of the
lattice parameter for ZnFe2O4 compared with that of MgFe2O4 is
due to the larger ionic radius of Zn2 than both Fe3 and Mg2 .

K.A. Mohammed et al. / Physica B 407 (2012) 795804

8.70
8.65

r A C AMg2 r Mg2 C AZn2 r Zn2 C AFe3 r Fe3

8.50
8.45
8.40

8.40

8.35

8.25
0.0

0.2

0.4

0.6

0.8

1.0

Zn (x)

Fig. 3. Variation of the lattice parameter, aexp and ath with x. (The inset: shows
variation of aexp with rA.)

C BMg2 r Mg2 C BZn2 r Zn2 C BFe3 r Fe3

2
B

where rA and rB are the cations radii of element at the tetrahedral


and octahedral sites, respectively and Ro is the oxygen ion radius
[20]. Table 1 and Fig. 3 show the calculated values of
( 0.138 A)
the lattice parameter, aexp and ath. It is clear that ath values
are higher than aexp values. The ath values vary linearly with

0.672

0.64
0.62

Table 2
Values of u, d, aexp, Ra and Rb as a function x.
x

aexp (A)

( 70.001)
RA (A)

(7 0.001)
RB (A)

0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0

0.3810
0.3814
0.3818
0.3822
0.3826
0.3830
0.3834
0.3838
0.3842
0.3846
0.3850

0.0060
0.0064
0.0068
0.0072
0.0076
0.0080
0.0084
0.0088
0.0092
0.0096
0.0100

8.378
8.393
8.405
8.409
8.417
8.421
8.430
8.439
8.442
8.448
8.450

1.901
1.910
1.919
1.925
1.933
1.940
1.948
1.956
1.962
1.970
1.976

2.045
2.046
2.046
2.043
2.042
2.040
2.039
2.038
2.036
2.034
2.032

2.050
1.98
2.045

Bond Length (A)

where C and C are the ionic concentration at the tetragonal and


octahedral sites, respectively, while r Mg2 , r Zn2 and r Fe2 are the
ionic radii of Mg2 , Zn2 and Fe3 ions, respectively. The cation
distributions have been obtained from the Mossbauer study [17].
Table 1 shows results for the cation distributions, aexp, ath, rA and
rB calculations. Fig. 2 shows clearly that values of both rA and rB
increase with increasing the Zn2 ions content; this can be
assigned to the replacement of Fe3 ions with a larger ionic
radius of the Zn2 , at the tetrahedral sites, and the replacement of
Mg2 ions with a larger ionic radius Fe3 ions at the octahedral
sites. Similar results have been reported for the Co substituted
NiZn ferrites [18].
The lattice parameter can be calculated using the following
relation, which relates the cation radius of the constituent
elements at the different lattice sites to the lattice parameter
[19]:
p
r A Ro 3r B Ro
p
ath 8
3
3 3

Cations Radii (A)

8.45

8.55

8.30

The average cation radius at the tetrahedral and octahedral


sites, rA and rB, can be estimated for all samples employing the
cation distributions and the relations [16]:

aexp

ath

8.60

8.35

A Mg21xy
Fe31
Zn2x Mg2y Fe31xy
x y B

rB

8.50

8.75

Lattice constants (A)

Values of the experimental, aexp, and theoretical, ath, lattice


constants are shown in Table 1 (Fig. 3). It is clear that the ath
values are greater than the aexp values. This is because the ath
calculations assume a perfectly lled close packed spinel structure and the anions and cations are rigid spheres [6]. Similar
behaviors have been shown by the CoCd Ferrites [15]. In the
Mg1  xZnxFe2O4 ferrites the Zn2 ions exclusively occupy the
tetrahedral sites, while most of the Mg2 ions prefer to occupy
the octahedral sites. The Fe3 ions are distributed among the
tetrahedral and octahedral sites. Thus the cation distribution for
the Mg1  xZnxFe2O4 ferrites can be written as

797

1.96

1.94

2.040
RB

RA

1.92

2.035
1.90

0.670

0.60

2.030
0.0

0.58

0.4

0.6

0.8

1.0

Zn (x)

rA

0.668

rB

0.56
0.54

0.666

0.52
0.50

0.664
0.0

0.2

0.2

0.4
0.6
Zn (x)

0.8

1.0

Fig. 2. Variation of the ionic radii, rA and rB, with x.

Fig. 4. Variation of the bond length, RA and RB, with x.

the increased Zn concentration in the sample. The inset in Fig. 3


shows the variation of the lattice constant aexp as a function of
the mean ionic radius, rA, at the tetrahedral site. It can be seen
that aexp values increase linearly with rA, verifying the suggested
linear relation between the two parameters [16]. These results
agree with results of the NiZn and NiCd ferrites [16].
The inter-ionic cationanion distances (the bond lengths) at
the A-sites, RA, and B-sites, RB, can be evaluated using the

798

K.A. Mohammed et al. / Physica B 407 (2012) 795804

3.3

Table 3
Variation of Rx, Rx0 , Rx00 , dexp, dxrd, DWH and DXRD with x.
x

Rx

(A)

Rx 0

(A)

Rx00

(A)

dexp
(g/cm3)

dXRD
(g/cm3)

DWH
(nm)

DXRD
(nm)

0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0

3.104
3.119
3.133
3.144
3.157
3.168
3.181
3.194
3.204
3.216
3.227

2.820
2.815
2.810
2.802
2.795
2.787
2.780
2.774
2.765
2.757
2.749

2.964
2.969
2.974
2.975
2.979
2.980
2.984
2.987
2.989
2.991
2.992

2.80
2.85
2.83
2.91
2.96
2.97
3.03
3.14
3.10
3.26
3.45

4.16
4.21
4.26
4.31
4.36
4.40
4.45
4.49
4.53
4.57
4.61

33
43
57
32
48
70
52
46
58
53
33

58
76
101
56
83
123
92
81
102
92
57

relations [21]
p
RA a 3d 18
RB a

r
d
1
2
3d 
2 16

d u0:375

Edge Length (A)

3.1

r
11
4u2 3u
16

Rx00 a

R x''

3.0
2.9

R x'

2.8
2.7
2.6
0.0

0.2

0.6

0.4

0.8

1.0

Zn (x)

Fig. 5. Variation of the bond lengths Rx, Rx0 and Rx00 with x.

5
6

where d represents the deviation from the oxygen parameter uideal


( 0.375). Values of u for Mg1  xZnxFe2O4 samples were found by
interpolation between those reported earlier for pure magnesium
respecand zinc ferrites, which are equal to 0.381 and 0.385 A,
tively [21]. Values of RA (RB) increase (decrease) with increasing
zinc content (x). The changes in RA and RB values with x might be
caused by the substitution process. Similar results have been
reported for other ferrite systems, such as the MnAl ferrites [22].
The increase in Zn 2 content leads to the A-site expansion and to
a relative displacement of the oxygen anions cause shrinkage of
the B-sites. Values of d, RA and RB are shown in Table 2 and Fig. 4.
It is clear that the bond lengths at tetrahedral sites, RA, are smaller
than those of octahedral sites, RB. This can be interpreted as more
covalent bonding of Fe3 ions at the B-sites than A-sites. These
results support the interpretation that correlates the decrease in
the bond length to the increased covalent bonding characteristics
[23]. The deviation of u from the standard value can be taken to
some extent as a measure of the triagonal distortion of the oxygen
coordinates at the octahedral sites. This increased deviation
reects the increasing effect of the triagonal distortion at
the B-sites, which increases as x increased. Similar effect has
been reported for other systems such as Co substituted NiZn
ferrites [18].
The increased values of d with Zn content (x) implies the
existence of a progressive expansion of the tetrahedral interstices
to accommodate the Zn2 ions which replace the Fe3 ions at the
A-sites. The anions move away from the nearest tetrahedral
cations without changing the structure symmetry. The increasing
values of the mean ionic tetrahedral radii rA, the bond length RA
and the tetrahedral edge Rx are all a consequence of this expansion [24].
The tetrahedral edge length Rx, the shared octahedral length Rx0
and the unshared octahedral edge length Rx00 of these cubic mixed
spinel oxides have been calculated using the following equations
[24]:
p
Rx a 22u0:5
7
p
Rx0 a 212u

Rx

3.2

Values of Rx, Rx0 and Rx00 are shown in Table 3 and Fig. 5. It is clear
that values of Rx, Rx0 and Rx00 increase, decrease and increase
slowly, respectively, with Zn content. Similar results have been
reported for the CdZn [24] and CdNiZn [25] ferrite systems.
The changes in u cause a distortion of the octahedral sites
symmetry. This distortion results in shortening the octahedral
shared edge bringing the anions into close contact along those
edges. The calculated results for the octahedral shared edges, Rx0 ,
and the bond length, RB, for these samples support such explana corresponds to about
tions. The average value of Rx0 ( 2.79 A)

twice that of the O2  ions ( 2.70 A).


The interionic distances between the cationcation (MeMe)
(b, c, d, e and f) and cationanion (MeO) (p, q, r and s) and the
angles between these distances in spinels are shown in Fig. 6 [26].
These distances are calculated for the Mg1  xZnxFe2O4 ferrites
using the experimental values of lattice constant, aexp, and oxygen
positional parameter u using the following relations [27]:
Cation-anion distances

Cation-cation distances



p a 58 u
p


q a u 14 3
p


r a u 14 11

p
s a 13 u 18 3

ap
2
a4p
c 8 11
 p
d 4a 3
3ap
e 8
3
 p
f 4a 6
b

The calculated values are shown in Tables 4 and 5. It can be


seen that all these distances (except p) increase slowly with
increasing Zn content (x). The increases in the interionic distances
are in accordance with increase in the unit cell volume. The bond
angles y1, y2, y3, y4 and y5 are calculated applying simple
trigonometry principles using the values of the interionic distances [27]. The increase (decrease) of the angle values suggests
strengthening (weakening) between the concerned cationcation
or cationanion interactions. The distances effects are opposite to
that of the angles. It is seen that angles y1, y2 and y5 decrease
while angles y3 and y4 increase with Zn content (x). Accordingly
the interactions between cationanion (AB) strengthen, while
the interactions between cations (BB) weaken. But the increases
in the distances values suggest weakening of these interactions.
The overall resultant strength of the different magnetic interactions decreases as the Zn content increases. These results are
consistent with the magnetic behaviors of these MgZn ferrites
[11,17].

K.A. Mohammed et al. / Physica B 407 (2012) 795804

799

Fig. 6. The interionic distances and angles between cationscations and anionscations at the tetrahedral and octahedral sites in spinel ferrites.
Table 4
as a function of Zn(x).
Values of the interionic distances in (A)
x

0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0

2.0442
2.0445
2.0441
2.0417
2.0403
2.0379
2.0367
2.0355
2.0328
2.0309
2.0280

1.9010
1.9102
1.9187
1.9255
1.9331
1.9399
1.9478
1.9557
1.9623
1.9695
1.9758

3.6400
3.6577
3.6741
3.6870
3.7017
3.7146
3.7298
3.7449
3.7575
3.7713
3.7834

3.6568
3.6653
3.6725
3.6762
3.6816
3.6853
3.6912
3.6971
3.7003
3.7049
3.7077

2.9621
2.9674
2.9716
2.9730
2.9759
2.9773
2.9805
2.9836
2.9847
2.9868
2.9875

3.4733
3.4796
3.4845
3.4862
3.4895
3.4912
3.4949
3.4986
3.4999
3.5024
3.5032

3.6278
3.6343
3.6395
3.6412
3.6447
3.6464
3.6503
3.6542
3.6555
3.6581
3.6590

5.4417
5.4514
5.4592
5.4618
5.4670
5.4696
5.4754
5.4813
5.4832
5.4871
5.4884

5.1305
5.1396
5.1470
5.1494
5.1543
5.1568
5.1623
5.1678
5.1696
5.1733
5.1745

4.7

Table 5
Values of the interionic angles (in degrees) as a function of Zn(x).

3.6
3.5

4.6
x

h1

h2

h3

h4

h5

0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0

123.34
123.22
123.08
122.96
122.83
122.71
122.58
122.45
122.33
122.20
122.08

144.95
144.38
143.83
143.29
142.75
142.22
141.69
141.17
140.66
140.15
139.65

92.85
93.05
93.25
93.45
93.65
93.85
94.06
94.26
94.47
94.67
94.88

125.92
125.96
126.01
126.05
126.09
126.14
126.18
126.23
126.27
126.31
126.36

74.48
74.14
73.80
73.46
73.12
72.79
72.45
72.12
71.80
71.47
71.15

Density (g/cm3)

3.4
4.5
3.3
4.4

3.2

d xrd
4.3

3.1

d exp

4.2

3.0
2.9

4.1

2.8
4.0
0.0

The X-ray density dXRD for each sample is calculated according


to the relation [28]:
dXRD

ZM
Na3

dexp

m
V

11

Values of the calculated dXRD and dexp are shown in Table 3 and
plotted in Fig. 7 as a function of Zn content in the sample. Both
dexp and dXRD densities increase with increasing Zn content. The
true density, dXRD, is higher than the bulk density, dexp, for all

0.4

0.6

0.8

1.0

Zn (x)
Fig. 7. Variation of the dexp and dXRD densities with x.

10

where Z, M, N and a3 represent number of molecules per unit cell


( 8), molecular weight, Avogadros number and volume of the
unit cell, respectively. The bulk dexp densities of the specimens
were evaluated using the measured mass, m, and volume, V, of the
samples using the following relation:

0.2

samples. This might be caused by the existence of pores in these


samples. The nature of variation of dexp supports the variation in
the lattice parameter. The difference between values of dexp and
dXRD data may be attributed to the porosity of the prepared
samples.
The porosity, p, was calculated according to the following
equation:
dexp
p 1
dXRD

12

The effect of Zn substitution on the porosity is shown in Fig. 8.


Zinc substitution increases the porosity (up to x0.6) thus

800

K.A. Mohammed et al. / Physica B 407 (2012) 795804

0.34

Porosity

0.32

0.30

0.28

0.26
Fig. 10. EDS pattern of the surface of a solid piece of x 0.5.

0.24
0.0

0.2

0.4

0.6

0.8

1.0

Zn (x)
Fig. 8. Variation of the porosity, p, with x (the inset shows 2r vs. x).

0.010

B Cos (0)

0.008

0.006

0.004
Fig. 11. SEM micrograph for the surface of a solid piece of x 0.5.

0.303

0.304

0.305
0.306
Sin (0)

0.307

0.308

Fig. 9. Variation of b cos(y) vs. sin(y) for Mg1  xZnxFe2O4 ferrites. (the inset:
b cos(y) vs. sin(y) for one sample).

increase the sample density. These data reveal clearly that with
increasing Zn concentration in the ferrite apparent porosity of the
samples increase because of increase in lattice parameter. Meanwhile the sample volume shrinkage increases as Zn content increase.
The inset in Fig. 8 shows the sample diameter as a function of Zn
content. The increased sample porosity and volume shrinkage with
increasing Zn content might be attributed to the difference in specic
gravity and hence the lattice parameter of the ferrite components.
Similar results have been reported for the CdNiZn ferrites [25].
The average crystallite size, DXRD, of all samples were evaluated from the reected X-ray diffraction peaks using Scherrers
equation [28]:
DXRD

kl
b cos y

13

where the constant k0.89, l is the wavelength of the X-ray


y is the diffraction angle and b is the full
radiation ( 1.5404 A),
width at half maximum (FWHM) of the most intense peak (3 1 1)
in radian. Values of the crystallite size are given in Table 5. The
average value for DXRD was observed to be about 48 nm. The
average crystallite size might be higher than this because the
width, b, is not corrected for the instrumental broadening.
Comparable results have been reported for the MgFe2O4 with
high sintering temperature [29], NiCuZn [30], CoMg [31] and

Fig. 12. TEM micrograph for crashed small solid piece of x 0.5.

CdNiZn [25] ferrites. Moreover, the average particle sizes have


been estimated for each sample using the WilliamsonHall plot
[32] in which values of b cos(y) vs. sin(y) have been plotted. The
intercept of the tted straight line equals kl/DWH:

b cos y

kl
2e sin y
DWH

14

where e is the strain introduced inside the sample. A typical


WilliamsonHall plot for one of the samples is shown in the inset

K.A. Mohammed et al. / Physica B 407 (2012) 795804

Table 6
Values of n1, n2, Kt, Ko and r as a function of Zn(x).
x

0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0

n1 (cm  1)
571
572
571
570
568
566
562
562
562
564
564

n2 (cm  1)
432
432
434
436
438
436
439
438
440
446
447

Kt  105
(dynes/cm)

Ko  105
(dynes/cm)

q  108 (X-cm)

2.372
2.381
2.372
2.364
2.347
2.331
2.298
2.298
2.298
2.314
2.314

1.358
1.358
1.370
1.383
1.396
1.383
1.402
1.396
1.409
1.447
1.454

34.5
13.6
4.30
1.91
3.00
2.85
1.10
0.67
0.35
0.28
1.97

7 105

Fig. 13. IR spectra for the Mg1  xZnxFe2O4 ferrites.

of Fig. 9. Values of DWH are shown in Table 3. The average value


for DWH was found to be about 84 nm.
Values of b cos(y) vs. sin(y) for the main X-ray (3 1 1) plane
have been plotted for all samples are shown in Fig. 9. The inset
shows b cos(y) vs. sin(y) for one sample. The general trend of data
variation in this gure supports the increase of the crystallite size
as Zn content increase in the sample.
The results of the analytical scanning electron microscope
(ASEM) conrm the purity of samples, the lack of any impurities
and the expected stoichiometry. It also shows a well-packed and
continuous grain structure with porosity and small holes at the
crystallites boundaries. A representative energy dispersive analysis (EDS) graph is shown in Fig. 10 for x 0.5. The SEM micrograph

801

of the surface of a solid pellet is shown in Fig. 11. Aggregates of


nearly rounded to cubic stacked crystallites of about 200800 nm
can be identied. Fig. 12 shows the TEM micrographs of a crashed
small solid piece conrm the SEM results. The crystallites sizes
obtained from SEM and TEM micrographs are larger than those
obtained from the XRD scans using Scherrers equation and
WilliamsonHall plot. These differences might be caused by
ignoring the crystal strain or defects contributions to the line
broadening of the FWHM values. Grain sizes as large as 0.335,
1.44 and 41.9 mm for MgZn ferrites were reported [33].
The infrared spectra measurements of the Mg1  xZnxFe2O4 in
the range 4001000 cm  1 indicate the presence of two strong
absorption bands n1 (571564 cm  1) and n2 (432447 cm  1). The
band positions n1 and n2 are listed in Table 6 as a function of Zn
content. These bands are common features of all the spinel
ferrites [5]. The absorption band n1 corresponds to stretching
vibration mode of Fe3 O2  in the tetrahedral sites, and n2
corresponds to the metaloxygen vibration in octahedral sites,
Fe3 O2  . Fig. 13 shows the typical IR spectra of the
Mg1  xZnxFe2O4 ferrites. The difference in the position of the
two strong bands can be related to the differences in the Fe3
O2  distances for A-sites and B-sites. It is clear (Table 2) that the
FeO bond lengths at the A-sites (1.9011.976 A) is shorter than
that of the B-sites (2.0452.032 A). This effect has been interpreted on the base of more covalent bonding of the Fe3 ions on
the A-sites than that on the B-sites [34]. The wave number of the
rst band, n1, decreases with increasing Zn content. This variation
in the band position can be attributed to the increase in the
cationoxygen (AO) bond length. This indicates the weakening
of the metaloxygen bonds at the tetrahedral sites due to the
replacement of the Fe3 ions with Zn2 ions at the tetrahedral
sites. The residence of Zn2 ions on the tetrahedral (A) site causes
an equivalent amount of Fe3 ions to migrate from A-site to
B-site, which results in an increase of the (AO) bond length of
the tetrahedral site of spinel structure (i.e. decrease in n1 band).
On substitution of Zn2 ions the position of n2 band shifts almost
linearly towards higher wave numbers over the whole composition range. This indicates the strengthening of the metaloxygen
bonds at the octahedral sites due to the shifts of more Fe3 ions
from the tetrahedral sites to the octahedral sites. These results
agree to some extent with results of the CuZn mixed ferrites [35]
and the Cu1  xZnxFeCrO4 ferrites [36]. The presence of Fe2 ions
causes local deformation in the spinel lattices, which has been
attributed to the JahnTeller distortion effect. This effect causes
splitting or shoulders on the absorption bands. A third band, n3, of
lower intensity, closer to n2 (less than 400 cm  1) has been
reported by many authors, for the Zn rich compositions [37]. It
is clear that the wave numbers of n3 and n4 lie outside the range of
these measurements. A relatively weak shoulder appears in the
range 650 up to 700 cm  1. The intensity of this shoulder
increases with increasing Zn2 ions; this leads to narrowing of
the band n1. This shoulder might be caused by the presence of two
covalent states of the ions such as the Fe2 O2  and Zn 2O2 
bonds at the tetrahedral sites. Similar results have been reported
for the alumina-doped zinc ferrites [38]. The broadening is
commonly observed in the inverse spinel ferrites caused by the
statistical distribution of Fe3 ions on A- and B-sites. Therefore,
the bands are sharpened as one goes from the inverse spinel
sample, MgFe2O4, to the normal spinel sample, ZnFe2O4. Similar
results have been reported for the mixed spinel Ni1  xZnxFe2O4
ferrites [39,40], which agree quite well with these results. It is
clear from the XRD calculations that the bond lengths of the ions
present in the tetrahedral sites, Ra, increases while the bond
lengths at the octahedral sites, Rb, decreases with increase in Zn
contents, x. It is well known that the bond length is inversely
proportional to the vibration frequency. This explains the increase

802

K.A. Mohammed et al. / Physica B 407 (2012) 795804

Rb (A)
2.02

2.03

2.04

Kt (dyne/cm) x 105

1.44

2.36

1.42
2.34
1.40
2.32

1.38

2.30

Ko (dyne/cm) x 105

1.46

2.38

1.36

1.90

1.95

2.00

2.05

Ra (A)
Fig. 14. The force constants Kt and Ko vs. bond lengths, Ra and Rb.

10.0
Log p (ohm-cm) x 10

35

p (ohm-cm) x 108

30
25
20
15

9.5
9.0
8.5
8.0
7.5
7.0
0.0

10

0.2

0.4 0.6
Zn (x)

0.8

1.0

5
0
0.0

0.2

0.4
0.6
Zn (x)

0.8

1.0

Fig. 15. Room temperature electrical resistivity, r as a function of x. (The inset:


shows log r as a function of x).

(decrease) in n1 (n2) values. Similar results have been reported for


the alumina doped zinc ferrite [38].
The force constants of the ions at the tetrahedral site (Kt) and
octahedral site (Ko) have been calculated by substituting the IR
band frequencies n1 and n2 in the following standard formulae
[35,40]:
K 4p2 c2 n2 m

stretching in the tetrahedral sites leads to higher force constant


than the octahedral site. Variations of Kt and Ko with RA and RB are
shown in Fig. 14. The increased (decreased) values of Ra (Rb) with
Zn content support the decrease (increase) in the force constants
at the A-sites (B-sites), respectively. Similar results have been
reported for the mixed CuZn ferrites [35].
Values of the measured room temperature DC-electrical resistivity for the Mg1  xZnxFe2O4 ferrites are shown in Table 4 and in
Fig. 15 as a function of Zn content. All samples showed high
resistivity of order of (108109) O-cm at room temperature.
Similar results have been reported for the Mg-ferrites [42,31],
Mn-ferrites [43] and NiZn ferrites [44]. It is clear that the
resistivity of these ferrites decreases with increase in Zn-content.
The addition of larger ionic sized Zn2 ions distorts the ferrite
lattice by rstly, increasing the lattice parameter, a and secondly, increasing the number of Fe2 and Zn ions and forcing x
Fe3 ions to migrate from A-site to B-site to replace x Mg2 ions.
These lead to the increase in the hopping processes for both
electrons and holes and to reduce the indirect interaction
between the Fe3 ions on the A- and B-sites. The motion of
charges will be affected by this distortion causing the resistivity
of these samples to decrease. There is an increase in the resistivity
for the sample Mg0.5Zn0.5Fe2O4 (x0.5). The structure of this
sample is probably neither inverse nor normal. The sample
ZnFe2O4 (x 1) exhibits noticeable increase in its resistivity. The
transition to the complete normal structure might be the reason
for this increase. The resistivity values for these ferrites are
considerably higher than those reported for NiZnMe (Me Cu,
Cd, Co, Ca and Mn) ferrites [45], CoNiMn ferrites [40] and
CdNiZn ferrites [25]. It is well known that Zn ions prefer the
occupation of tetrahedral A-sites while Mg and Fe ions partially
occupy A- and B-sites. As Zn content increases at the A-sites, the
Mg ions concentration at the B-sites will decrease. As a result
some of the Fe3 ions will be forced to migrate from A-sites to
B-sites. This leads to the increase in the electron hopping process
(Verwey mechanism) between Fe2 and Fe3 ions at the B-sites,
which causes the decrease in the electrical resistivity (increase in
conductivity) as the Zn content increases. The decrease in resistivity value with increasing Zn content has been shown by other
ferrites such as MnZn [43], NiZn [8,44] and ZnMg [46] ferrites,
the presence of Zn ions enhanced the increase in the conduction
process. The inset in Fig. 15 shows values of log (r) vs. Zn content.
The two distinct regions might suggest that there are two well
dened system structures inverse and normal spinel separated by
a narrow mixed region.
The ionic packing coefcients Pa and Pb at the tetrahedral and
octahedral sites, respectively, can be estimated using the following equations [7]:
p
16
r xt u0:25a 3Ro

15
10

where c is the light speed (  2.99  10 cm/s), n is the vibration


frequency of the A- and B- sites and m is the reduced mass of the
Fe3 and O2  ions (  2.601  10  23 g). Table 6 shows variation of
Kt and Ko with Zn content (x). It is clear that values of the force
constants Kt (Ko) decrease (increase) as Zn content (x) increases.
This behavior has been attributed to the variation in cation
oxygen bond lengths at the A- (B-)sites. The decrease (increase) in
Kt (Ko) values can be related to the increase (decrease) in bond
length, Ra (Rb) at the tetrahedral (octahedral) sites. Longer bond
length requires less energy to break the bond. The increase in Ko
values as x increases can be attributed to the more transfer of
Fe3 ions from A-sites to the B-sites, which causes some kind of
charge imbalance at the B-sites, which is likely to shift the oxygen
ions towards the Fe3 ions causing the force constant, Ko (bond
length, RB) to increase (decrease) as x increases [41]. The

r xo 0:625uaRo

17

Pa

r xt
RA

18

Pb

r xo
RB

19

where rxt and rxo are the interstitial radii and RA and RB are the
average values of the ionic radii at the tetrahedral and octahedral
sites, respectively, u is the anion parameter, a is the lattice
parameter and Ro is the anion radius. It is claimed that [6] the
small values of the packing factors, Pa and Pb ( o1), testify to the
smaller ion distances and larger overlapping of the cation and
anion orbital, suggesting the existence of cation or anion vacancies while the opposite holds for larger values of Pa and Pb ( 1).

K.A. Mohammed et al. / Physica B 407 (2012) 795804

Table 7
Values of the ionic packing, fulllment, vacancy and Pauling electronegativity
coefcients of Mg1  xZnxFe2O4 ferrites.
x

rxt (A)

rxo (A)

Pa

Pb

b (%)

DvA

DvB

0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0

0.5210
0.5302
0.5387
0.5455
0.5531
0.5599
0.5678
0.5757
0.5823
0.5895
0.5958

0.6642
0.6645
0.6641
0.6617
0.6603
0.6579
0.6567
0.6555
0.6528
0.6509
0.6480

0.778
0.775
0.770
0.764
0.758
0.752
0.748
0.743
0.737
0.733
0.727

0.998
0.998
0.997
0.992
0.989
0.985
0.983
0.980
0.975
0.972
0.967

0.649
0.647
0.646
0.646
0.646
0.646
0.646
0.645
0.646
0.646
0.648

3.668
4.000
4.399
5.066
5.625
6.279
6.793
7.270
7.971
8.535
9.254

2.986
3.000
3.002
3.006
3.010
3.012
3.016
3.019
3.023
3.027
3.028

3.177
3.170
3.166
3.162
3.158
3.155
3.150
3.146
3.142
3.138
3.135

The fulllment coefcient of the unit cell, a, determines the


degree of the ionic packing of the spinel structure. This coefcient
can be calculated using the following equation [6]:

32pR3A 2R3B 4R3o


3V

20

where V is the volume of the unit cell ( a3exp ).


The vacancy parameter, b, is dened as a normalized volume
of the missing ions at the nodal points of the spinel structure. It is
a measure of the total vacancy concentration existing in the
sample. It can be calculated using the following equation [6]:

a3th a3exp
a3th

n100%

21

The difference in the Pauling electronegativities, Dw is an


evaluation of the ionic bonds strength in the tetrahedral and
octahedral sites. These differences in the electronegativity per
cation at the A- and B-sites can be estimated (wmg 1.31,
wZn 1.65, wFe 1.83 and wO 3.44) using the following equations
[6]:

wA

DwA wo 

22

wB

DwB wo 

23

where wA and wB represent the Pauling electronegativities of the


cations at the A- and B-sites, respectively, and wo represent the
oxygen electronegativity. The wA and wB values depend on the
kind of cations and anions at the A- and B-sites [6]:

wA C AMg2 wMg2 C AZn2 wZn2 C AFe3 wFe3


wB

C BMg2 wMg 2 C BZn2 wZn 2 C BFe3 wFe3


2

24

25

where wmg 1.31, wZn 1.65, wFe 1.83 and wO 3.44. Calculations
of Pa, Pb, a, b, DwA and DwB coefcients for the Mg1  xZnxFe2O4
ferrites using Eqs. (16)(25) are shown in Table 7.
It is clear that values of the ion packing factor Pa decrease more
strongly than Pb and the fulllment coefcient a values remain
almost constant with increasing Zn content. The small values
( o1) of Pa, Pb, a and the strong increase of the vacancy parameter,
b indicate the presence of cation or anion vacancies and the
domination of the Zn2 vacancies at the tetrahedral sites. The
electrical conductivity in ferrites was explained on the basis of the
Verwey mechanism, i.e. exchange of electrons between the
adjacent Fe2 and Fe3 ions that are distributed randomly over
the octahedral sites. Since the hopping mechanism is the most
probable conduction mechanism in the MgZn ferrites the presence of the Zn vacancies in the MgZn ferrites might be partially
responsible for the improvement of the electrical conductivity

803

(decrease in r values) in these mixed ferrites. These results agree


with results of the thermoelectric power of the MgZn ferrites
[4649] in which the presence of both negative and positive
Seebeck coefcients were reported, suggesting that both n-type
and p-type carriers are responsible for the conduction process in
these materials as follows:
Fe3 e 3Fe2
Zn2 3Zn e
Similar role has been claimed for Zn vacancies in the MgZn
[50], ZnCr2  xNixSe4 [50] and ACr2  4 (A Zn, Cd, Mn, XO, S, Se)
[6] spinels.
It has been reported [7] that a distinct relation exists between
the ionic packing factor of the metals at the tetrahedral site in
normal ferrites and their electrical conductivity type. The samples, Mg1  xZnxFe2O4 (except ZnFe2O4), are mixed spinel. The
packing factor, Pa and Pb, for the normal spinel sample; ZnFe2O4
was calculated and found to be equal 0.727 and 0.967, respectively. The Pa value lies approximately at the border between the
insulatorsemiconductor conducting regions in the proposed
classication. The measured electrical resistivity of the ZnFe2O4
sample (conductivity 5.08  10  7 O  1-cm  1) supports this
classication.

4. Conclusions
The analysis of the X-ray diffraction spectrum showed that the
Mg1  xZnxFe2O4 system, which has been prepared by the conventional solid state reaction technique with double sintering at
temperatures around 1000 1C is pure single-phase ferrite system.
Values of the theoretically and experimentally calculated lattice
parameter, a, increase with Zn content in the sample. The
variation was nonlinear for the experimental values of a. With
the exception of the volume shrinkage, some bonds length and
grain sizes the other lattice parameters increase with increasing
Zn concentration. The estimated values agree quite well with
those predicted theoretically. The EDS, SEM and TEM analysis of
the surface of a solid piece showed the existence of aggregates
of stacked nearly rounded to cubic crystallites of about
(200800) nm in diameter. The infrared spectra of these ferrites
give rise to two most prominent absorption envelopes. The high
frequency band, n1, lies in the range between 564 and 571 cm  1
was assigned to the Fe3 O2  and Zn2 O2  stretching vibrations at the tetrahedral sites. The second main absorption band,
n2, is present in the range between 432 and 447 cm  1, which is
assigned to the Fe3 O2  stretching vibrations at the octahedral
sites. The absorption bands n1 and n2 revealed the formation of
single-phase spinel structure with two sublattices: the tetrahedral (A) and octahedral (B) sites. These results showed that the
normal mode of vibration of tetrahedral clusters is higher (shorter
bond length) than that of octahedral clusters (longer bond
length). Calculated values of the bond lengths RA and RB and ionic
radii rA and rB support this interpretation. The force constants of
the tetrahedral site, Kt, decrease and the octahedral site, Ko,
increase with increase in Zn content x. This behavior has been
attributed to the variation in cationoxygen bond length and the
charge imbalance at the concerned sites. We can conclude that
the gradual increase of the Zn content, x, in the studied
Mg1  xZnxFe2O4 system leads to gradual transformation from the
inverse to normal spinel structure. The system transfer from an
almost inverse spinel MgFe2O4 (x0) to normal spinel ZnFe2O4
(x 1.0) leads to sharpening of the absorption bands. The room
temperature electrical resistivity is of order of (108109)O-cm.
The increase (decrease) of Zn (Mg) content lowers (raises) the

804

K.A. Mohammed et al. / Physica B 407 (2012) 795804

resistivity in these ferrites. The ionic packing factors Pa and Pb


decrease, the fulllment coefcient, a, remains almost constant
and the vacancy parameter, b, strongly increases as the Zn
content increase in the sample. The small values ( o1) of Pa, Pb,
a and the strong increase of the vacancy parameter, b indicate the
presence of cation or anion vacancies and the domination of the
Zn2 vacancies at the tetrahedral sites. The presence of the Zn2
vacancies in the MgZn ferrites might be partially responsible for
the improvement of the electrical conductivity in these mixed
ferrites.

Acknowledgment
One of us (Kadhim Ahmed) would like to thank Dr. C.B. Kolekar
for useful comments.
References
[1] E.C. Snelling (Ed.), Soft Ferrites, Properties and Applications, 2nd ed., Butterworth Publishing, London, 1989. (and references therein).
[2] J.G. Lee, J.Y. Park, Y.-J. Oh, C.S. Kim, J. Appl. Phys. 84 (1988) 2801.
[3] B.P. Ladgonkar, P.N. Vasambekar, A.S. Vaingankar, J. Magn. Magn. Mater.
210 (2000) 289.
[4] H. Knoch, H. Dannheim, Phys. Status. Solidi A 37 (1976) K135.
[5] R.D. Waldron, Phys. Rev. 99 (1955) 1727.
[6] T. Gron, Philos. Mag. B 70 (1994) 121.
[7] T. Satoh, T. Tsushima, K. Kudo, Mater. Res. Bull. 9 (1974) 1297.
[8] U. Ghazanfar, S.A. Siddiqi, G. Abbas, J. Mater. Sci. Eng. B 118 (2003) 132.
[9] S.T. Alone, Sagar E. Shirsath, R.H. Kadam, K.M. Jadhav, J. Alloys Compds.
509 (2011) 5055.
[10] V.K. Mittal, P. Chandramohan, S. Bera, M.P. Srinivasan, S. Velmurugan,
S.V. Narasimhan, Solid State Commun. 137 (2006) 6.
[11] M.M. Haque, M. Huq, M.A. Hakim, Physica B 404 (2009) 3915.
[12] S.M. Kadam, S.I. Patil, S.H. Patil, B.K. Chougule, Bull. Mater. Sci. 15 (1992) 127.
[13] C. Upadhyay, H.C. Verma, S. Anand, J. Appl. Phys. 95 (2004) 5746.
[14] M. Chakrabarti, D. Sanyal, A. Chakrabarti, J. Phys.:Condens. Matter 19 (2007) 1.
[15] O.M. Hemeda, M.M. Barakat, J. Magn. Magn. Mater. 223 (2001) 127.
[16] A. Globus, H. Pascard, V. Cagan, J. Physique Colloq. 38 (1977). C1-163-168.
[17] A.M. Gismelseed, K.A. Mohammed, H.M. Widatallah, A.D. Al-Rawas,
M.E. Elzain, A.A. Yousif, J. Phys.: Conf. Ser. 217 (012138) (2010) 1.
[18] M.A. Amer, A. Tawk, A.G. Mostafa, A.F. El-Shora, S.M. Zaki, J. Magn. Magn.
Mater. 323 (2011) 1445.

[19] S.A. Mazen, M.H. Abdallah, B.A. Sabrah, H.A.M. Hasham, Phys. Status Solidi
A 134 (1992) 263.
[20] R.D. Shannon, Acta Crystallogr. A 32 (1976) 751.
[21] K.J. Standley, Oxide Magnetic Materials, Claredon Press, Oxford, 1990.
[22] R.L. Dhiman, S.P. Taneja, V.R. Reddy, Adv. Condens. Matter Phys. 2008 (2008)
1. Article ID 703479.
[23] B.F. Levine, Phys. Rev. B 87 (1973) 2591.
[24] C. Otero Arean, E. Garcia Diaz, J.M. Rubio Gonzalez, M.A. Villa Garcia, J. Solid
State Chem. 77 (1988) 275.
[25] M. Siva Ram Prasad, B.B.V.S.V. Prasad, B. Rajesh, K.H. Rao, K.V. Ramesh,
J. Magn. Magn. Mater. 323 (2011) 2115.
[26] A. Goldman, Modern Ferrite Technology, 2nd ed., Springer Science Business
Media, Inc., New York, 2006.
[27] V.K. Lakhani, T.K. Pathak, N.H. Vasoya, K.B. Modi, Solid State Sci. 13 (2011)
539. (and references there in).
[28] B.D. Cullity, Elements of X-ray Diffraction, Addison Wesley, 1959.
[29] Y. Huang, Y. Tang, J. Wang, Q. Chen, Mater. Chem. Phys. 97 (2006) 394.
[30] W.-C. Hsu, S.C. Chen, P.C. Kuo, C.T. Lie, W.S. Tasi, J. Mater. Sci. Eng. B 111
(2004) 142.
[31] M.A. Ahmed, A.A. EL-Khawlani, J. Magn. Magn. Mater. 321 (2009) 1959.
[32] G.K. Williamson, W.H. Hall, Acta Metall. 1 (1953) 22.
[33] S.S. Khot, N.S. Shinde, B.P. Ladgaonkar, B.B. Kale, S.C. Watawe, J. Adv. Appl. Sci.
Res. 2 (2011) 460. (and references there in).
[34] B. Evans, S. Hanfner, J. Phys. Chem. Solids 29 (1968) 1573.
[35] H.M. Zaki, H.A. Dawoud, Physica B 405 (2010) 4476.
[36] M.C. Chhantbar, U.N. Trivedi, P.V. Tanna, H.J. Shah, R.P. Vara, H.H. Joshi,
K.B. Modi, Indian J. Phys. A 78 (2004) 321.
[37] H.A. Dawood, S.K. Shaat, J. Al-Aqsa Uni. 10 (2006) 247.
[38] N.M. Deraz, J. Anal. Appl. Pyrol. 91 (2011) 48.
[39] A.T. Raghavender, N. Biliskov, Z. Skoko, Mater. Lett. 65 (2011) 677.
[40] P.A. Shaikh, R.C. Kambale, A.V. Rao, Y.D. Kolekar, J. Alloys Compd. 492 (2010)
590.
[41] K.B. Modi, U.N. Trivedi, P.U. Sharma, V.K. Lakhani, M.C. Chhantbar, H.H. Joshi,
Indian J. Pure. Appl. Phys. 44 (2006) 165.
[42] K S Rane, V.M.S. Verennkar, P.Y. Sawant, Bull. Mater. Sci. 24 (2001) 323.
[43] D. Ravinder, K. Latha, Matter. Lett. 41 (1999) 247.
[44] A.M. Abdeen, J. Magn. Magn. Mater. 185 (1998) 199.
[45] E. Rezlescu, L. Sachelarie, P.D. Popa, N. Rezlescu, IEEE Trans. Magn. 36 (2000)
3962.
[46] H.M. Zaki, Physica B 404 (2009) 3356.
[47] B.P. Ladgaonkar, P.N. Vasambekar, A.S. Vaingankar, Bull. Mater. Sci 23 (2000)
87.
[48] M.U. Islam, A.Y. Abbasi, T. Abbas, M.A. Chaudhry, A.Z. Chaudhry, J. Res. Sci.
14 (2003) 103.
[49] P.P. Hankare, V.T. Vader, U.B. Sankpal, L.V. Gavali, R. Sasikala, I.S. Mulla, Solid
State Sci. 11 (2009) 2075.
[50] H. Duda, I. Jendrzejewska, T. Gron, S. Mazur, P. Zajdel, A. Kita, J. Phys. Chem.
Solids 68 (2007) 80.

Journal of Magnetism and Magnetic Materials 329 (2013) 165169

Contents lists available at SciVerse ScienceDirect

Journal of Magnetism and Magnetic Materials


journal homepage: www.elsevier.com/locate/jmmm

Characterization of nanocrystalline Mg0.6Zn0.4Fe2O4 soft ferrites synthesized


by glycine-nitrate combustion process
S. Hajarpour a,b,n, Kh. Gheisari b, A. Honarbakhsh Raouf a
a
b

Department of Materials, Faculty of Engineering, Semnan University, Semnan, Iran


Department of Materials Science and Engineering, Faculty of Engineering, Shahid Chamran University, Ahvaz, Iran

a r t i c l e i n f o

abstract

Article history:
Received 25 April 2012
Received in revised form
11 October 2012
Available online 26 October 2012

In this study, MgZn ferrite with the chemical formula of Mg0.6Zn0.4Fe2O4 is synthesized through a
modied combustion synthesis using glycine as fuel and metal (Mg, Zn and Fe) nitrates as reactants.
The technique, known as glycine-nitrate process, involves exothermic decomposition of a viscous
liquid, prepared by thermal dehydration of an aqueous solution containing metal nitrates and glycine.
The product powders produced at seven different molar ratios of glycine to nitrate (G/N ratio), varying
from 0.37 to 0.75, are agglomerates of ne particles whose typical diameter are several tens of
nanometers. Thermodynamic modeling of the combustion reaction indicates that as the fuel-to-oxidant
ratio increases, the amount of gases produced and the adiabatic ame temperature rise. X-ray
diffraction shows that samples crystallize in a spinel-type structure in all reactions. The morphology
of the powders is examined using eld emission scanning electron microscopy and transmission
electron microscopy. Through magnetic measurements conducted by a vibrating sample magnetometer, the maximum saturation magnetization (46 emu/g) is found to occur at the highest G/N ratio.
& 2012 Elsevier B.V. All rights reserved.

Keywords:
Glycine-nitrate process
Combustion synthesis
MgZn ferrite
Magnetic properties
Nanocrystalline

1. Introduction
Soft ferrites constitute an important class of magnetic materials which has attracted much interest in many elds [1,2]. They
have been the subject of extensive research owing to their wide
range of applications such as those in hyperthermia, information
storage systems, gas sensors, microwave devices, magnetic
recording media, electronic industries, humidity sensors and
green anode materials [16].
Zn-substituted ferrites exhibit very noticeable magnetic properties that have especially captured the attention of specialists for
high-frequency applications [7]. Mg1  xZnxFe2O4 series, commercially known as Ferrocube-Z, is customarily used as ferrite core
where effective coupling between an electrical current and
magnetic ux is required [4]. MgZn ferrite, with its wide usage
in power transformers, microwave devices and telecommunication is usually considered a favorable choice in the industry. High
resistivity, low coercivity and inappreciable eddy current loss
contribute to this favorability and are the reasons why MgZn
ferrite is capable of being utilized as a soft-magnetic material [2].
In order to form these ne spinel-type ferrite particles, various
wet methods have been devised including freeze-drying, spray-

n
Corresponding author at: Department of Materials, Faculty of Engineering,
Semnan University, Semnan, Iran. Tel.: 98 9365281698; fax: 98 21 88521951.
E-mail address: sadeghhajarpour@hotmail.com (S. Hajarpour).

0304-8853/$ - see front matter & 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.jmmm.2012.10.023

drying, co-precipitation and solgel as well as dry methods such


as mechanical alloying and thermal plasma [810]. In addition,
combustion synthesis has recently been developed to synthesize
ultra-ne powders of ceramic oxides [11]. This method is very
simple, cost-effective and fast. In brief, in this technique, an
aqueous solution, containing an organic compound as fuel (for
example glycine, citric acid, urea, etc.) and suitable metal salts as
oxidizer (nitrates are generally preferred since they have good
solubility in water), is introduced to a furnace or microwave oven
[8,10,12]. The solution then starts boiling. The heating and
evaporation of the nitrate solution results in an exothermal,
self-sustaining and fast chemical reaction [1214].
The choice of fuel is important. In general, a good fuel should
react non-violently, produce non-toxic gases, and act as a complexant for metal cations [8]. If glycine and metal nitrate are
chosen as fuel and oxidizer respectively, the process is called
glycine nitrate process (GNP). To be more precise about the
advantages of this method, it could be stated that the method is
easy, offers less energy-intensive steps, needs inexpensive raw
materials and has a fast production rate. The favorability of
combustion synthesis over solid-state synthesis in terms of
compositional homogeneity and purity of the nal product should
also not be neglected [12,14,15].
The most signicant parameters in this kind of reaction are
type of fuel, fuel-to-oxidizer ratio, whether or not excess oxidizer
has been used, ignition temperature, and water content of the
precursor mixture. The primary controllable processing variable is

S. Hajarpour et al. / Journal of Magnetism and Magnetic Materials 329 (2013) 165169

the glycine-to-nitrate ion (G/N) ratio, which affects ame temperature, combustion rate, and product morphology and composition. The peak ame temperature is reported to be typically
obtained at a G/N ratio that corresponds to complete combustion,
producing H2O, CO2, and N2 as waste gases, with no atmospheric
oxygen required. The G/N ratio that corresponds to complete
combustion is referred to as the stoichiometric ratio [1,13,16].
There have been a limited number of reports on preparation
of spinel-type ferrites by GNP. This paper is a study of the
synthesis and properties of zinc-substituted spinel-type ferrite
Mg0.6Zn0.4Fe2O4, produced by GNP, at different G/N molar ratios
varying from 0.37 to 0.75. The effect of G/N ratio on morphology,
crystal structure, and magnetic properties is also examined.

due to insufcient oxidizer quantity [5]. Redox reactions are


usually exothermic in nature and often lead to explosion, if not
properly controlled. The combustion of metal nitrate-glycine
mixture appears to undergo a self-propagating and nonexplosive exothermic reaction [5]. According to the principle of
propellant chemistry, for stoichiometric redox reaction between a
fuel and an oxidizer, the ratio of the net oxidizing valency of the
metal nitrate to the net reducing valency of the fuel should be
unity [16]. Examples of fuel-decient, stoichiometric and fuel-rich
combustion reactions are presented hereunder.
Fuel-decient, G/N 0.375:
0.6Mg(NO3)2  6H2O0.4Zn(NO3)2  6H2O2Fe(NO3)3  9H2O
3NH2CH2COOH-Mg0.6Zn0.4Fe2O4 5.5N2 6CO2 7.5H2O
3.25O2

2. Experimental procedure

0:6MgNO3 2  6H2 O 0:4ZnNO3 2  6H2 O 2FeNO3 3  9H2 O


xNH2 CH2 COOH-Mg0:6 Zn0:4 Fe2 O4



x
5x
9x
H2 O 10
O2
4 N2 2xCO2
2
2
4
where x is 3.5, 4, 4.44 (stoichiometric ratio), 5, 5.5 and 6. Within
20 to 30 min, the combustion reaction yields brown to black,
porous products that are quite easily ground to powder.
X-ray diffraction (XRD) is performed on powders for phase
characterization and estimation of crystallite size, with a step
size of 0.05 and sample time of 1 s, using CuKa radiation on a
(XRD; Bruker D8) X-ray diffractometer. The morphological features of the product powders are determined by Transmission
Electron Microscopy (TEM; LEO-906E) and Field Emission Scanning Electronic Microscopy (FESEM; Hitachi S-4160) after coating
the samples with gold. Magnetic measurements are carried out
at room temperature with a Vibrating Sample Magnetometer
(VSM, Meghnatis Daghigh Kavir, Iran) in an operating range of
78.5 kOe.

3. Results and discussion


3.1. Nature of combustion synthesis
One of the conclusions of the conducted research is that the
nature of combustion reaction and characteristics of the synthesized powder strongly depend on the compositional ratio of fuel
to nitrate (here glycine to nitrate or the G/N ratio). The color of
the powder is also found to change with the G/N ratio, from light
brown to black as G/N ratio varied from 0.375 to 0.75. This is
attributed to the carbonaceous residue remaining from glycine

0.6Mg(NO3)2  6H2O0.4Zn(NO3)2  6H2O2Fe(NO3)3  9H2O


4.44NH2CH2COOH-Mg0.6Zn0.4Fe2O4 6.22N2 8.88CO2
11.11H2O

DHf1  3033.94 kJ/mol


Fuel-rich, G/N0.75:
0.6Mg(NO3)2  6H2O0.4Zn(NO3)2  6H2O2Fe(NO3)3  9H2O
6NH2CH2COOH3.5O2-Mg0.6Zn0.4Fe2O4 7N2 12CO2
15H2O

DHf1  4386.88 kJ/mol


As observed in Fig. 1, as G/N ratio is raised, the amount of
produced gas slightly increases until stoichiometric ratio is
reached, after which it continues to increase at a higher rate.
To explain the variation of adiabatic ame temperature (Tad)
with respect to the G/N ratio in the case of a combustion reaction,
the following equation is employed:
T ad T 0

DH0r DH0p

Cp

where T0 is 25 1C, DHr and DHp are respectively the enthalpies of


formation of the reactants and products and Cp is the heat
capacity of the products at constant pressure. Using thermodynamic data for various reactants and products, as given in Table 1
[13], the combustion enthalpy and the theoretical adiabatic ame
temperature can be calculated as a function of G/N ratio. Fig. 1
shows the adiabatic ame temperature vs. the G/N ratio. As
expected, the temperature is substantially elevated according to
a direct relationship with the amount of fuel used in combustion.

2500
Adiabatic flame temperature (C)

This study focuses on synthesizing Mg0.6Zn0.4Fe2O4. Glycine


(NH2CH2COOH) is selected as the fuel since it is more costeffective and its combustion heat ( 3.24 kcal g  1) is greater than
that of urea ( 2.98 kcal g  1) or citric acid (  2.76 kcal g  1)
[12,13]. Metal (Mg, Zn and Fe) nitrates are utilized since
they can act as the oxidant and at the same time as the metal
source. At an appropriate ratio, analytical-grade nitrates of
Fe(NO3)3  9H2O, Zn(NO3)2  6H2O, Mg(NO3)2  6H2O and glycine
(all 499%, Merck, Germany) are dissolved in 100 cc of distilled
water to obtain the precursor solution. The solution is prepared
with varying glycine proportions, i.e., different G/N ratios. The
total concentration of the metal ions is set to approximately
0.1 mol/l. The precursor is put in a round glass and placed on a hot
plate under atmospheric conditions for heating. The solution is
then allowed to boil and ignite. The ignition is accompanied by
evolution of large volumes of gases, according to the fuel-tooxidant ratio, as described by the following chemical equation:

DHf1  1778.38 kJ/mol


Stoichiometric, G/N0.55:

40
35

2000

30
25

1500

20
1000
500

15
Adiabatic flame temperature
Mole of gases evolved

0
0.35

10

mol of gases produced

166

5
0

0.45
0.55
0.65
0.75
Glycine to nitrate molar ratio

Fig. 1. Variation of adiabatic ame temperature and mole of gases evolved as a


function of G/N ratio.

S. Hajarpour et al. / Journal of Magnetism and Magnetic Materials 329 (2013) 165169

Table 1
Enthalpy of formation and specic heat for the combustion synthesis of MgZn
ferrite.
Compound

DHf1 (kJ/mol)

Cp (kJ/mol K)

Mg(NO3)2
Zn(NO3)2
Fe(NO3)3
NH2CH2COOH
ZnFe2O4
CO2
H2O
N2
O2

 790.1
 482
 671
 528
 1179.1
 395
 243

0.148
0.061
0.051
0.024
0.039

Intensity (a.u.)

(222)

(400)

(422) (511)

G/N ratio

D (nm)

a (A)

dx (g/cm3)

0.37
0.43
0.5
0.55
0.62
0.68
0.75

12.86
30.36
46.70
39.18
46.52
31.25
39.75

8.412
8.424
8.430
8.430
8.424
8.424
8.413

4.829
4.808
4.798
4.798
4.808
4.808
4.827

Lattice parameter (A)

(220)

(111)

Table 2
Crystallite size (D), lattice parameter (a) and X-ray density (dx) of Mg0.6Zn0.4Fe2O4.

8.435

(311)

G/N = 0.75

(440)

G/N = 0.68
G/N = 0.62
G/N = 0.55

8.43
8.425
8.42
8.415

G/N = 0.5

8.41
0.35

G/N = 0.43
G/N= 0.37

15

25

35
45
2 Theta (degree)

55

167

65

0.45
0.55
0.65
Glycine to nitrate molar ratio

0.75

Fig. 3. Variation of the lattice parameter as a function of G/N ratio.

However, the actual ame temperatures are much lower than the
theoretically calculated values owing to radioactive losses, incomplete combustion, and heating of air [12,16].

give rise to a moderate crystallite size in samples with excess fuel


compared to those with decient (ne crystallites) and stoichiometric (coarser than the other two cases) quantities. The same
trends are observed in crystallite sizes, calculated from XRD data.
The lattice constant of the as-synthesized ferrite, a, shown in
Fig. 3, is determined via the classical formula presented as Eq. (3):

3.2. Characterization

Fig. 2 shows XRD patterns of the as-synthesized powders


prepared at seven different G/N ratios. The characteristic peaks
of the spinel phase are observed in all cases. There is an overall
increase in signal/noise ratio with increasing G/N ratio. Additionally, fuel-decient samples present diffraction peaks of Fe2O3 as
well as an unknown peak, indicating the as-synthesized powders
contain impurities.
The average crystallite size, DXRD, of all samples is evaluated
based on the reected X-ray diffraction peaks using Scherers
Eq. [11]:

where l is the wavelength of CuKa; h, k, and l are the Miller


indices; and y is the diffraction angle corresponding to the (h k l)
plane. The value of a ranges from lattice constant of MgFe2O4
(a 8.391 nm) to the lattice constant of ZnFe2O4 (a 8.441 nm).
Judging from Ref. [2], it can be inferred that the spinel phase is
MgZn ferrite. As seen in Fig. 3, the highest attained value of the
lattice parameter pertains to the stoichiometric composition. This
value drops, moving towards right or left starting at the stoichiometric point. This behavior can be attributed to the fact that in a
stoichiometric sample, crystal structure is well formed without
any strain.
The X-ray density was calculated using Eq. (4):

Fig. 2. XRD patterns of Mg0.6Zn0.4Fe2O4 at various G/N ratios.

l h2 k2 l2 2
2
sin y
1

DXRD

kl
bcos y

where k is a constant equal to 0.89, l is the wavelength of the


y is the diffraction angle and b
X-ray radiation (Cuka 1.5404 A),
is the full width at half maximum (FWHM) of the selected peak in
radian. In this work, crystallite sizes are calculated for three
different peaks between 30 and 50 degrees (220, 311 and 400
planes) whose averages are given in Table 2.
The rst sample with the lowest G/N ratio has the smallest
crystallite size. This is resulted from insufcient quantity of fuel,
and thus the lower combustion temperature, which hinders grain
growth [2]. Faster cooling, which is the consequence of the higher
gas evolution, results in smaller grain size. Accordingly, samples
with higher G/N ratios than stoichiometric should have ner
crystallite sizes. On the other hand, excess fuel, results in higher
reaction temperature, counteracting the effect of heightened gas
production. The combined effect of these two factors will ultimately

dx

ZM
Na3

where Z is number of molecules per unit cell (eight, in this case),


M is the molecular weight of the ferrite, a is the lattice parameter
and N is Avogadros number. The variation of dx with respect to
G/N ratio is documented in Table 2. This variation obviously has a
reverse trend compared to that of the lattice parameter. Characteristics of the synthesized powders are summarized in Table 2
as well.
Fig. 4 presents the typical FESEM and TEM photomicrographs
of the product powder of Mg0.6Zn0.4Fe2O4 with a G/N ratio of 0.75.
Voids and holes are observable in Fig. 4(a) (low magnication).
These result from release of gases during combustion. Porosity
swelling caused by heightened gas evolution in higher G/N ratios
is also reported by others [2]. Particle sizes can be estimated from

168

S. Hajarpour et al. / Journal of Magnetism and Magnetic Materials 329 (2013) 165169

Fig. 4. FESEM and TEM photomicrographs of Mg0.6Zn0.4Fe2O4 produced by fuel-rich reaction (G/N 0.75): (a) FESEM, low magnication and (b) TEM, high magnication.

Table 3
Hysteresis loss, saturation magnetization (Ms), coercive force (Hc) and remanent
magnetization (Mr) of Mg0.6Zn0.4Fe2O4.
G/N ratio

Hysteresis loss (emu Oe/g)

Ms (emu/g)

Hc (Oe)

Mr (emu/g)

0.37
0.43
0.5
0.55
0.62
0.68
0.75

1523
3778
6834
11861
9310
7250
9013

24.91
40.50
42.49
44.44
40.63
44.05
46.61

18.56
45.06
68.50
74.34
82.06
81.47
77.54

1.28
5.02
6.79
8.29
7.37
7.79
8.69

90

Fig. 5. Room temperature hysteresis loops of as-synthesized product at G/N ratios


of 0.375, 0.555 and 0.75.

Saturation magnetization
(emu/g)

60
50

Coercive force (Oe)

80
70
60
50
40
30
20
10

40

0
0.35

30
20

0.45
0.55
0.65
Glycine to nitrate molar ratio

0.75

Fig. 7. Coercive force vs. G/N ratio.

10
0
0.35

0.45
0.55
0.65
Glycine to nitrate molar ratio

0.75

Fig. 6. Saturation magnetization vs. G/N ratio.

Fig. 4(b) to be approximately 100 nm, slightly more than those


calculated from XRD results, since the observed particles in TEM
in actuality consist of many crystallites.
3.3. Magnetic properties
Fig. 5 shows the MH loops of the produced MgZn ferrite at
different G/N ratios. Hysteresis loop of the as-synthesized ferrite
for the rst sample (G/N0.375) is characterized with a fairly low
magnetization value even at 8.5 kOe. As apparent in Fig. 2, lack of
sufcient fuel for combustion has led to more noises and less
intense peaks than is the case with other samples. This is an
indication that the spinel structure is not well established.

Fig. 6 is a plot of saturation magnetization (Ms) vs. G/N ratio.


Moving from lower to higher G/N ratios, there is a generally
increasing trend in Ms values, although stoichiometric ratio seems
to have yielded the second largest saturation magnetization. The
spinel structure appears to be well established at G/N ratios of
0.55 (stoichiometric) and 0.75, as the highest Ms and XRD signal/
noise values are observed at these points.
Coercivity (Hc), remanent magnetization (Mr) and hysteresis
loss (quantied in Table 3) constitute other important properties
of soft ferrites. It is well known that in soft ferrites these
parameters should be at their smallest possible values.
As the plot of coercivity versus G/N ratio (Fig. 7) indicates,
coercivity gradually increases in the rst ve samples until a
maximum of 82 Oe is reached at G/N ratio of 0.625. Applying larger
quantities of fuel causes coercivity to drop to about 77 Oe thereafter.
The coercivity (Hc) for nanocrystalline soft-magnetic materials
can be calculated by the following equation:
Hc

P c /KS
Ms

S. Hajarpour et al. / Journal of Magnetism and Magnetic Materials 329 (2013) 165169

where /KS is the average anisotropy constant, covering all


sources of anisotropy energy, for instance the intrinsic magnetocrystalline and the shape anisotropies. Ms is the saturation
magnetization and Pc is the pre-factor that reects the symmetry
of /KS and is close to unity.
For grain sizes smaller than the magnetic exchange length the
averaged anisotropy can be expressed as:
/KS

v2 D6 K 41
A3

4. Conclusions
A simple chemical process has been devised for producing
nanocrystalline MgZn ferrite powders using glycine as fuel and
metal nitrates as oxidant with different glycine to nitrate molar
ratios. It has been observed that the fuel-to-oxidant ratio bears a
signicant inuence on characteristics of the produced powder.
Thermodynamic modeling of the combustion reaction shows that
the amount of gas production and adiabatic ame temperature
are in direct relationship with G/N ratio. The main conclusions of
the work are listed below.

 GNP is found to be a rapid, cost-effective method for synthesizing spinel-type ferrites.

 The characteristic peaks of spinel phase are observed in


all cases.

 TEM observations indicate that the product powder consists of

 Considering the dependency of coercivity on crystallite size, it




can be concluded that the exchange length of Mg0.6Zn0.4Fe2O4


is about 32 nm.
In conclusion, it appears using the stoichiometric ratio of fuel
and oxidizer yields the best overall results.

Acknowledgements

where v is the volume fraction of the grains, K1 is the anisotropy


of individual grains, D is grain size and A is the exchange constant.
Accordingly, in the nanocrystalline soft magnetic materials,
whenever grain size is smaller than the magnetic exchange
length, Hc follows a D6-power law. On other hand, when a grain
becomes large enough to contain a domain wall, domains can be
formed within the grains and magnetization is dominated by
domain wall movement, causing Hc to decrease according to the
well-known 1/D law. Hence, with increasing grain size, Hc should
rise to a maximum and then decrease [17,18].
The coercivity is at its maximum when the crystallite size is
about 32 nm. According to Herzers diagram [18], it can be
concluded that the exchange length of Mg0.6Zn0.4Fe2O4 is about
32 nm. This is in good agreement with ndings of others suggesting that the coercive forces of magnetic particles reach their
maximum when the grain size is between 20 and 50 nm [1] and
that the random-anisotropy model apparently provides a good
description of the magnetic properties for grain sizes below
4050 nm [18].
While remanent magnetization shows quite similar behavior
to saturation magnetization, the trend of hysteresis loss versus
G/N ratio is initially upwards with a maximum at stoichiometric
ratio and then downwards. Magnetic properties of synthesized
powders prepared at seven different G/N ratios, derived from
VSM, are tabulated in Table 3.

169

agglomerates of primary particles with a diameter of about


100 nm for a G/N ratio of 0.75.
The decrease in the lattice parameter when moving to larger or
smaller G/N ratios than the stoichiometric one might indicate
presence of strain in the spinel lattice.
The maximum saturation magnetization is obtained at the
highest G/N ratio.

We would like to extend our gratitude to Semnan University


and Shahid Chamran University for their nancial support. We
also feel indebted to Mr. S.R. Asadolahpour for his contribution to
the paper.
References
[1] N. Kikukawa, M. Takemori, Y. Nagano, M. Sugasawa, S. Kobayashi, Synthesis
and magnetic properties of nanostructured spinel ferrites using a glycine
nitrate process, Journal of Magnetism and Magnetic Materials 284 (2004)
206214.
[2] K.A. Mohammed, A.D. Al-Rawas, A.M. Gismelseed, A. Sellai, H.M. Widatallah,
A. Yousif, M.E. Elzain, M. Shongwe, Infrared and structural studies of
Mg1  xZnxFe2O4 ferrites, Physica B 407 (2012) 795804.
[3] M.
Al-Haj,
Structural
characterization
and
magnetization
of
Mg0.7Zn0.3SmxFe2  xO4 ferrites, Journal of Magnetism and Magnetic Materials
299 (2006) 435439.
[4] H. Spiers, I.P. Parkin, Q.A. Pankhurst, L. Afeck, M. Green, D.J. Caruana,
M.V. Kuznetsov, J. Yao, G. Vaughan, A. Terry, A. Kvick, Self propagating high
temperature synthesis of magnesium zinc ferrites (MgxZn1  xFe2O3): thermal
imaging and time resolved X-ray diffraction experiments, Journal of Materials
Chemistry 14 (2004) 11041111.
[5] Z. Pedzich, M.M. Bucko, M. Krolikowski, M. Bakalarska, J. Babiarz, Microstructure and properties of MgZn ferrite as a result of sintering temperature,
Journal of the European Ceramic Society 24 (2004) 10531056.
[6] P. Poddar, H. Srikanth, S.A. Morrison, E.E. Carpenter, Inter-particle interactions and magnetism in manganesezinc ferrite nanoparticles, Journal of
Magnetism and Magnetic Materials 288 (2005) 443451.
[7] V.D. Kassabova-Zhetcheva, Characterization of the citrate precursor, used for
synthesis of nanosized MgZn ferrites, Central European Journal of Chemistry
7 (2009) 415422.
[8] J.C. Toniolo, M.D. Lima, A.S. Takimi, C.P. Bergmann, Synthesis of alumina
powders by the glycinenitrate combustion process, Materials Research
Bulletin 40 (2005) 561571.
[9] I. Shari, H. Shokrollahi, M.M. Doroodmand, R. Sa, Magnetic and structural
studies on CoFe2O4 nanoparticles synthesized by co-precipitation, normal
micelles and reverse micelles methods, Journal of Magnetism and Magnetic
Materials 324 (2012) 18541861.
[10] V. Vasanthi, A. Shanmugavani, C. Sanjeeviraja, R. Kalai Selvan, Microwave
assisted combustion synthesis of CdFe2O4: magnetic and electrical properties, Journal of Magnetism and Magnetic Materials 324 (2012) 21002107.
[11] L.A. Chick, L.R. Pederson, G.D. Maupin, J.L. Bates, L.E. Thomas, G.J. Exarhos,
Glycinenitrate combustion synthesis of oxide ceramic powders, Materials
Letters 10 (1990) 612.
[12] C.-C. Hwang, J.-S. Tsai, T.-H. Huang, Combustion synthesis of NiZn ferrite by
using glycine and metal nitratesinvestigations of precursor homogeneity,
product reproducibility, and reaction mechanism, Materials Chemistry and
Physics 93 (2005) 330336.
[13] C.-C. Hwang, J.-S. Tsai, T.-H. Huang, C.-H. Peng, S.-Y. Chen, Combustion
synthesis of NiZn ferrite powderinuence of oxygen balance value,
Journal of Solid State Chemistry 178 (2005) 382389.
[14] R.D. Purohit, S. Saha, A.K. Tyagi, Nanocrystalline thoria powders via glycinenitrate combustion, Journal of Nuclear Materials 288 (2001) 710.
[15] A.C.F.M. Costa, V.J. Silva, C.C. Xin, D.A. Vieira, D.R. Cornejo, R.H.G.A. Kiminami,
Effect of urea and glycine fuels on the combustion reaction synthesis of
MnZn ferrites: evaluation of morphology and magnetic properties, Journal
of Alloys and Compounds 495 (2010) 503505.
[16] R.D. Purohit, B.P. Sharma, K.T. Pillai, A.K. Tyagi, Ultrane ceria powders via
glycine-nitrate combustion, Materials Research Bulletin 36 (2001)
27112721.
[17] Y. Shen, H.H. Hng, J.T. Oh, Synthesis and characterization of high-energy ball
milled Ni15%Fe5%Mo, Journal of Alloys and Compounds 379 (2004)
266271.
[18] K.H.J. Buschow, F.R. de Boer, Physics of magnetism and magnetic materials,
Kluwer Academic Publishers, New York, 2004.

Journal of the European Ceramic Society 24 (2004) 10531056


www.elsevier.com/locate/jeurceramsoc

Microstructure and properties of MgZn ferrite as a result of


sintering temperature
Zbigniew Pedzicha,*, Mirosaw M. Buckoa, Micha Krolikowskia,
Magorzata Bakalarskab, Joanna Babiarzb
a

University of Mining and Metallurgy, Faculty of Materials Science and Ceramics, al. Mickiewicza 30, 30-059 Cracow, Poland
b
LG.Philips Displays Poland, ul. Zwierzyniecka 2, 96-100 Skierniewice, Poland

Abstract
This paper presents results of investigations on property changes of MgZn ferrite used commercially for deection yoke cores
sintered at temperatures ranging from 900 to 1400  C. Physical and mechanical properties were determined: apparent density,
strength, elastic properties. The most important magnetic properties as initial permeability, coercivity, saturation magnetic ux,
retentivity and Curie point temperature were determined. Microstructure evolution during sintering was observed and compared
with changes of mechanical and magnetic properties as well as electrical conductivity. Presented results could be helpful in
optimisation of ferrite heat-treatment conditions.
# 2003 Elsevier Ltd. All rights reserved.
Keywords: Ferrites

1. Introduction
From the early days of ferrites commercial application as a TV deection yoke MnZn ferrites were the
material of choice. These materials connected good
magnetic properties with low production costs. Introduction of High Denition Television (HDTV) changed
the situation. Signicant increase in horizontal grids
meant that ferrite material had to be eective at a higher
frequency. It means that ferrite used in HDTV application should have higher resistivity than the MnZn
one (  102 .cm).1 Mentioned problem could be solved
with the utilization of MgZn ferrite. Although, the
most of magnetic parameters are better for MnZn
ferrites, the decisive parameter is resistivity much
higher for MgZn materials (  106107 .cm). Processing of MgZn ferrites should be provided with care on
their densication, microstructure and mechanical
properties to assure possible high values of magnetic
properties.
The present work shows the results of investigation
on inuence of sintering conditions of commercially
utilised MgZn ferrite powder on product properties.
* Corresponding author. Tel.: + 48-12617-2397; fax: +48-126334630.
E-mail address: pedzich@uci.agh.edu.pl (Z. Pedzich).
0955-2219/03/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0955-2219(03)00386-8

The aim of the research was to optimise manufacturing of ferrite with suitable magnetic properties
accompanied with as good as possible mechanical
properties.

2. Experimental
All investigation, conducted in the presented work,
based on commercially utilized by LG.Philips Displays
Poland MgZn ferrite powder (containing a small
amount of Mn). This powder mixed with plastiers was
uniaxial pressed under 50 MPa. Samples in O-ring
shape were sintered in a furnace with Superkanthal 1900
heating elements in air atmosphere. The maximum sintering temperature ranged from 900 to 1400  C with
100  C steps. The heating rate was 10  C/min. The
soaking time at maximum temperature was 2 h in each
case.
Apparent density () and water wetability (W) of sintered samples were determined by the Archimedian
method. The strength of sintered bodies was estimated
by diagonal compression tests of O-rings. The destructive compression stresses ( u) were calculated utilizing
Frochts formulas:2
u P  K=D-2dh

Z. Pedzich et al. / Journal of the European Ceramic Society 24 (2004) 10531056

1054

Table 1
Results of densication parameters (, W), ultrasonic wave propagation velocity measurements (vL, vT), strength estimation of O-ring samples ( u)
and grain size changes (M,sM) for all sintered samples
Sintering
conditions

Apparent
density
(, g/cm3)

Wetability,
W(%)

Longitudinal
ultrasonic wave
velocity, vL (m/s)

Transverse
ultrasonic wave
velocity, vT (m/s)

Destructive
compressive
stress,  u (MPa)

Mean grain
size, M (mm)

Mean grain
size deviation,
sM (mm)

Starting powder
900  C
1000  C
1100  C
1200  C
1300  C
1400  C

2.570.2a
3.100.2
4.070.2
4.460.1
4.460.1
4.480.1

17.2 0.3
10.5 0.3
2.7 0.3
0
0
0

2260.819.7
3885.418.0
6297.642.3
7127.8115.6
7181.9112.9
7250.292.6

1344.237.8
2375.730.8
3572.710.5
3864.832.5
3882.020.2
3914.825.9

0.750.07
3.180.46
9.480.82
12.082.10
10.941.31
6.790.39

0.5
0.5
1.0
1.4
4.1
10.6
16.9

0.21.0b
0.52.5b
0.6
2.1
5.5
8.3

a
b

, denotes standard deviation of mean value.


Denotes range of grain size distribution.

where: P is a load at destruction moment; K the coecient of stress concentration (assumed in our calculations as (1); D, d and h are the external diameter,
internal diameter and height of the ring, respectively.
Tests were performed using ZWICK 1435 apparatus.
The microstructural analyses of sintered bodies
encompass SEM (Philips XL30) and optical microscopy
(Nikon Ephipot 300) observations of polished sample
surfaces after chemical etching. The microstructural
parameters as a mean grain size (M) and mean grain
size deviation (sM) were determined by Saltykovs
method.3
The X-ray diraction patterns analysis (Seifert
XRD7) was helpful in phase identication and calculation of unit cell parameter values for ferrite powder and
sintered materials.
Electrical conductivity of samples was measured using
Hewlett Packard HP33501A multimeter at temperatures
ranged from 200 to 700  C.
Longitudinal (vL) and transverse (vT) ultrasonic waves
propagation velocity measurements were performed
with INCO (Poland) equipment.
The basic magnetic parameters as Curie temperature
(TC), initial permeability (i), coercivity (Hc), saturation
magnetic ux (Bsat) and retentivity (Br) were determined
using dynamic methods.

Strength of the samples estimated by destructive


compressive stress S is low for these with open porosity.
Maximum strength is reached for dense ferrite body
with the smallest grain size. Increase of heat treatment
temperature causes signicant grain growth and
strength decrease.
Ultrasonic measurements show that for samples with
stable, high density wave propagation velocity slightly
increases with grain size increase. It means that grain
boundaries inuence ultrasonic wave propagation. In
the material with small grains (i.e. a large amount of
boundaries) ultrasonic waves propagate slower than in
ferrites with big grains. It suggests that grain boundaries
are probably not homogenous.
Fig. 1 demonstrates an X-ray diraction pattern
achieved for a sample sintered at 1300  C. Such an
image is typical for all investigated samples. Only one
non-stoichiometric phase with Fd3m structure (corresponding to franklinite type for the ZnFeO system4)
could be detected in each pattern. Fig. 2 illustrates
changes of unit cell parameter a due to sintering temperature. Up to sintering temperature of 1100  C parameter a has the same value, above this temperature
signicant increase of a value is observed. These

3. Results
Table 1 summarized densication parameters (,W),
ultrasonic wave propagation velocity measurements
results (vL, vT), strength estimation of O-ring samples
( u) and grain size parameters (M,sM) for all sintered
bodies. Samples sintered at temperatures ranging from
1200  C and higher shows no wetability. Their apparent density is practically stable. In samples sintered at
lower temperatures signicant porosity can strongly
inuence measured properties (mechanical, elastic and
magnetic).

Fig. 1. The X-ray diraction pattern of the sample sintered at 1300  C.

Z. Pedzich et al. / Journal of the European Ceramic Society 24 (2004) 10531056

changes are the most probably due to the various occupation of cation positions in the spinel structure. The
iron and manganium cations valency changes
(Fe+2$Fe+3 and Mn+2$Mn+3) may cause the
cations to occupy dierent Wycko positions according
to their charge, a and d respectively.5
Description of microstructure evolution could be
completed with data included in Fig. 3, where the
cumulative curves of grain size distribution for materials
sintered at 11001400  C were plotted. These data show
that in samples sintered at 1300 and 1400  C an exaggerated grain growth occurred. Microstructures of the
mentioned ferrites are given in Fig. 4.
The basic magnetic property values are collected in
Table 2. The most suitable properties show materials
sintered at 1200 and 1300  C. They have the highest
values of initial permeability, saturation magnetic ux
and retentivity. The coercivity of these materials is the
lowest among investigated materials.
The measurement of electrical resistance as a function
of temperature allows us to calculate conductivity of the

1055

investigated material. The data ploted in Ahrenius


coordinates make it possible to estimate activation
energy of ferrite conductivity. Fig. 5 compares results of
measurements for materials sintered at 1100 and
1400  C. Coecients aii placed in the gure are slopes of
estimation lines correlated with activation energy of
conductivity. Both calculated curves have two ranges of
linearity (200440  C and 440700  C). At higher temperatures, where the structure decides on conductivity,
values of activation energy are similar. It suggest that
the structure of the grain interiors are almost the same
in both ferrites. At lower temperatures some additional
factors such as: distribution of respective cations, concentration of impurities or/and dierences in microstructure inuence the conductivity. Conductivity plots
compared in Fig. 5 concern two relatively dense ferrites
with signicantly dierent microstructure (grain size and
amount of grain boundaries) and unit cell parameters.
Generally, the electrical resistance measurements show
that materials sintered at higher temperatures have
lower resistance.

Fig. 2. The unit cell parameter a changes vs sintering temperature.

Fig. 3. Cumulative curves of grain size distribution in ferrites sintered


at indicated temperatures.

Fig. 4. Microstructures of polished and eatched surfaces of ferrite


bodies sintered at 1300 (a) and 1400  C (b).

Z. Pedzich et al. / Journal of the European Ceramic Society 24 (2004) 10531056

1056

Table 2
Basic magnetic properties of investigated samples
Sintering
conditions

Initial
permeability,
I, (mT)

Coercivity,
Hc, (Am)

Saturation
magnetic ux,
Bsat, (mT)

Retentivity,
Br, (mT)

Curie
temperature,
TC ( C)

900  C
1000  C
1100  C
1200  C
1300  C
1400  C

13
43
215
467
569
549

252
104
36
23
21

25
88
203
234
248
246

6.3
59
163
175
173
148

136
141
141
142
145
151

Fig. 5. Arrhenius plot of conductivity of ferrites sintered at indicated


temperatures. The aii parameters are the measure of activation energy
of conductivity at dierent temperature ranges.

Maximisation of magnetic parameter values demands


high sintering temperature (1300  C). However, such
heat-treatment conditions cause an exaggerated grain
growth and signicantly a decrease of mechanical
strength.
The measurements of ultrasonic wave propagation
velocity indicate that the grain growth is accompanied
with wave propagation velocity increase. Because the
densities of samples sintered at 12001400  C are practically the same, mentioned increase could be ascribed
to microstructure changesamount of grain boundaries
decrease and probably more uniform distribution of
impurities. It is conrmed by the decrease of ferrite
resistivity with sintering temperature increase. This fact
can be attributed to lower amount of grain boundaries.

References
4. Summary
A concise characteristic of dierent ferrite properties
as a function of sintering temperature given above
shows that denition of the optimal method of ferrite
heat-treatment requires the reconciling of opposing
tendencies in temperature dependence of individual
properties.

1. Goldman, A., Handbook of Modern Ferromagnetic Materials.


Kluwer Academic Publishers, Boston-Dordrecht-London, 1999.
2. Frocht, M. M., Photoelasticity. Wiley, New York, 1949.
3. Saltykov, S., Stereometric Metallurgy. Metallurgia Publ.,
Moscov, 1950.
4. JCPD-ICDD card No. 22-1012.
5. Lucchesi, S., Russo, U. and Della Giusta, A., Crystal chemistry
and cation distribution in some Mn-rich natural and synthetic
spinels. European Journal of Mineralogy, 1997, 9, 3142.

ARTICLE IN PRESS

Physica E 33 (2006) 367369


www.elsevier.com/locate/physe

X-ray characterization and phase transformation kinetics of ball-mill


prepared nanocrystalline MgZn-ferrite at elevated temperatures
M. Sinha, H. Dutta, S.K. Pradhan
Department of Physics, The University of Burdwan, Golapbag, Burdwan 713104, West Bengal, India
Received 2 March 2005; accepted 11 April 2005
Available online 12 June 2006

Abstract
Nanocrystalline MgZn-ferrite is prepared by ball milling the stoichiometric powder mixture of MgO, ZnO and a-Fe2O3. A nonstoichiometric ferrite phase is noticed to form after 3 h of milling when particles of starting materials became nano-sized. After 25 h of
milling, stoichiometric ferrite phase is formed with 9 nm particle size. Post annealing study of ball-milled sample reveals that the
nanocrystalline ferrite phase is stable up to 873 K and then starts to decompose into individual starting phases. However, heat treatment
of unmilled stoichiometric powder mixture even at 1473 K for 1 h duration does not result in formation of stoichiometric MgZn-ferrite
phase.
r 2006 Published by Elsevier B.V.
PACS: 61.72.y; 75.50.Gg; 75.50.Tt; 81.40.Lm
Keywords: (Mg,Zn) nanoferrite; XRD; Ball milling; Post-annealing

Ferrites are a group of technologically important


materials that can be used in magnetic, electronic and
microwave elds. Magnetic nanocrystalline materials hold
great promise for atomic engineering of materials with
functional magnetic properties [13]. Many magnetic
nanocrystals show superparamagnetism [4] in single
domain particles below a certain critical size. Magnetic
nanocrystals have been extensively applied in magnetic
recording medium, information storage, bio-processing
and magneto-optical devices [5,6]. Many synthetic roots
have been employed to prepare magnetic nanocrystals
[79]. High-energy ball milling is a very suitable solid state
processing technique for the preparation of nanocrystalline
ferrite powder exhibiting new and unusual properties
[1013]. To the best of our knowledge, so far the phase
transformation kinetics of ball mill prepared nanocrystalline MgZn-ferrite have not yet been studied in detail at
elevated temperatures by X-ray powder diffraction (XRD)
method.

Corresponding author. Tel.: +91 342 2557282; fax: +91 342 2530452.

E-mail address: skp_bu@yahoo.com (S.K. Pradhan).


1386-9477/$ - see front matter r 2006 Published by Elsevier B.V.
doi:10.1016/j.physe.2005.04.016

Spinels are characterized by a very compact oxygen


array with cations in tetrahedral (T) and octahedral (M)
coordination and may be described by the IV(A1i Bi)
VI
(B2i Ai)O4 structural formula, where IV and VI
represent tetrahedrally and octahedrally coordinated sites,
A and B are cations with variable valency and i the
inversion parameter. Normal spinels are those with i 0,
inverse spinels those with i 1, as MgFe2O4.
Zinc ferrite prepared by conventional ceramic method
forms the structure of a normal spinel with Zn in the
tetrahedral [13] and iron in the octahedral [14] sites of a
cubic close packing of oxygen atoms IVZnVIFeO4 [15,16]. It
is well known that the change in temperature may result in
the change in the inverse degree of zinc ferrite. A new
mechanochemical method of zinc ferrite synthesis from a
mixture of zinc oxide and iron oxide has received increased
attention in recent years and offers the possibility of
forming zinc ferrite inverse spinel structural state. The
objectives of the present work are (i) to prepare the
nanocrystalline MgZn-ferrite by ball milling the stoichiometric mixture of MgO, ZnO, and a-Fe2O3 and (ii) to study
the phase transformation kinetics of ball-milled and postannealed samples at elevated temperatures.

ARTICLE IN PRESS
M. Sinha et al. / Physica E 33 (2006) 367369

368

(101)

High-energy ball milling of MgO (M/S Merck, 98%


purity), ZnO (M/S Merck, 99% purity) and a-Fe2O3 (M/S
Glaxo, 99% purity) in 0.5:0.5:1 mol% was conducted in a
planetary ball mill (Model P5, M/S Fritsch, GmbH,
Germany). The rotation speed of the disk was 300 rpm
and that of vials was about 450 rpm. Milling was done at
room temperature in hardened chrome steel vial of volume
80 ml using 30 hardened chrome steel ball of 10 mm dia. at
BPMR 40:1. The time of milling varies from 15 min to 25 h,
depending upon the rate of phase transformation. The 8,
20 and 25 h ball-milled samples are post annealed at 773,
873, 973, 1073, 1273 and 1473 K, each for 1 h duration in a
programmable furnace.
The XRD proles of the unmilled, ball-milled and postannealed samples were recorded using Ni-ltered Cu Ka
radiation from a highly stabilized and automated Philips
generator (PW1830) operated at 35 kV and 25 mA. The
generator is coupled with a Philips X-ray powder
diffractometer consisting of a PW 3710 mpd controller,
PW1050/37 goniometer and a proportional counter. The
step scan data (of step size 0.021 2y and counting time 5 s)
for the entire angular range (15801 2y) of the experimental
samples were recorded. The XRD powder patterns
recorded from unmilled mixture and ball-milled samples
are shown in Fig. 1. The powder pattern of unmilled
mixture contains only the individual reections of MgO
(ICDD PDF # 43-1022), ZnO (ICDD PDF #36-1451) and
a-Fe2O3 (ICDD PDF #33-0664) phases only. The intensity

(220)
(202)

(111)

14000

(102)
(024)

(012)

16000

(116)
(110)
(018)
(214) (220)
(103)
(300)
(200)
(112)
(201)
(1 0 10)

ZnO
MgO
(Mg,Zn) Fe2O4

(113)
(200)

18000

Intensity (arb. unit)

Fe2O3

(100)
(104)
(002) (110)

20000

(PURE)

12000
3H

10000

8H
12H

8000

20H

20

30

40

50

60

(533)

(440)

(511)

(400)

(222)

(311)

(220)

6000

(111)

25H

70

80

2 (degree)
Fig. 1. X-ray powder diffraction patterns of unmilled and ball-milled
MgOZnOa-Fe2O3 mixture powders. The peak positions of different
phases are marked in the gure.

ratio of individual reections is in accordance with the


stoichiometric composition of the mixture. It is evident
from the gure that in course of milling, the formation of
MgZn-ferrite phase has been noticed after 3 h milling.
From the indexed pattern, it is clear that the XRD pattern
also contains some intense reections of MgO and a-Fe2O3
phases but ZnO reections seemed to be absent. All the
reections are broadened sufciently due to ball milling.
With increasing milling time the intensities of both the
MgO and a-Fe2O3 reections diminish gradually and the
XRD pattern of 25 h ball-milled powder seems to contain
only the MgZn-ferrite reections. All the reections are
sufciently broadened and the particle size of ferrite phase
has been estimated from Scherrer formula and to be
8.95 nm. It seems that nanocrystalline MgZn-ferrite can
be obtained by ball milling the stoichiometric composition
of MgO, ZnO and a-Fe2O3.
To study the phase transformation kinetics and phase
stability of nanocrystalline ferrite at elevated temperatures
8, 20 and 25 h ball-milled powders were post annealed
successively at different temperatures, each for 1 h duration. The unmilled stoichiometric mixture powder were
heat treated simultaneously with the ball-milled samples
for comparative phase transition studies. The post annealing and heat treatment studies reveal that upto 873 K, fullwidth at half-maxima (FWHM) of all the reections
irrespective of milling time reduced signicantly and one
of the intense reections of a-Fe2O3 phase was found to be
present in both the 8 and 20 h ball-milled samples and only
one reection of ZnO in 8 h sample only. There is no
indication of formation of ferrite phase in unmilled
powder. After post annealing at 973 K, the a-Fe2O3
reection seemed to be absent but the presence of ZnO
reection is noticed in all the ball-milled samples. It is
interesting to note that the formation of ferrite phase is
clearly evident in the XRD pattern of unmilled sample
(Fig. 2). In the course of increase in post annealing
temperatures FWHM value of all the reections reduces
continuously which signify the continuous increase in
particle size with increasing post annealing temperatures.
After post annealing at 1073 K, it is observed that the XRD
patterns of all the ball-milled samples have been started to
decompose into individual oxides phases but ferrite phase
is still present as the major phase. However, the amount of
ferrite in the unmilled sample is increased considerably in
the expense of oxide phases. It is interesting to note that
after post annealing at 1273 K, rate of decomposition in
ball-milled samples increased to a large extend and at the
same time, more amount of ferrite phase is formed in
unmilled powder and as a consequence, all the XRD
patterns irrespective of milling become almost equal. This
trend in pattern decomposition of ball-milled samples
remains unaltered after post annealing at 1473 K (Fig. 3)
though the reections of starting materials resolved
clearly into individual components due to further increase
in particle size (79.34 nm) followed by reduction in
FWHM values.

ARTICLE IN PRESS
(101)

(300)
(103)
(200)
(112)
(201)
(1 0 10)

(116)

(102)
(024)

(220)

(110)

ZnO
MgO
(Mg,Zn) Fe2O4

(018)

(111)

Intensity (arb.unit)

12000

PURE

10000
8H

8000

20H

References
25H

20

30

40
50
2 (degree)

(533)

(440)

(511)

(422)

(400)

(311)
(222)

(111)

(220)

6000

60

70

80

Fig. 2. X-ray powder diffraction patterns of unmilled and ball-milled


MgOZnOa-Fe2O3 mixture milled for different times and annealed at
973 K temperature for 1 h duration.

(311)

28000

(Mg,Zn)Fe2O4

26000

MgO
ZnO

24000

(220)

(620)

(533)
(622)
(444)

(440)

(422)

(200)

(400)
PURE

(102)

14000

(100)
(002)
(101)
(111)
(222)

18000
(111)

Intensity (arb.unit)

20000

(511)

(220)

22000

16000

369

The above observations suggest that MgZn-Ferrite


phase can be obtained by ball milling the stoichiometric
composition of MgO, ZnO and a-Fe2O3 when the particles
of all the starting phases reduce to nanometric stage. The
ferrite phase remains stable in nanocrystalline form till the
873 K post annealing temperature and when particle size of
ferrite phase exceeds at critical limit (14.73 nm), it starts
to decompose. However, a complete ferrite phase cannot be
obtained by heat treating the unmilled stoichiometric
mixture even at 1473 K. A detail microstructure characterization and phase transformation kinetic study employing
Rietvelds powder structure renement method would be
reported soon.

Fe2O3

(113)
(200)

(012)

(100)

14000

(002) (110)

(104)

M. Sinha et al. / Physica E 33 (2006) 367369

12000
8H

10000
20H

8000
25H

6000
4000
20

30

40

50

60

70

80

2 (degree)
Fig. 3. X-ray powder diffraction patterns of unmilled and ball-milled
MgOZnOa-Fe2O3 mixture milled for different times and annealed at
1473 K temperature for 1 h duration.

[1] T. Hirai, J. Kobayashi, I. Koasawa, Langmuir 15 (1999) 6291.


[2] R.H. Kodama, J. Magn. Magn. Mater. 200 (1999) 359.
[3] K.V.P.M. Sha, Y. Koltypin, A. Gedanken, R. Prozorov, J. Balogh,
J. Lendvai, I. Felner, J. Phys. Chem. 101B (1997) 6409.
[4] M. Elbschutz, S. Shtrikman, J. Appl. Phys. 39 (1968) 997.
[5] I. Anton, I.D. Dabata, L. Vekas, J. Magn. Magn. Mater. 85 (1990)
219.
[6] R.D. McMickael, R.D. Shull, L.J. Swartzendruber, L.H. Bennett,
R.E. Watson, J. Magn. Magn. Mater. 111 (1992) 29.
[7] D. Niznansky, N. Viart, J.L. Renspinger, IEEE Trans. Magn. 30
(1994) 821.
[8] J.M. Yang, W.J. Tsuo, F.S. Yen, J. Sol. State Chem. 145 (1999)
50.
[9] Y. Shi, J. Ding, X. Liu, J. Wang, J. Magn. Magn. Mater. 205 (1999)
249.
[10] P. Druska, U. Steinike, V. Sepelak, J. Sol. State Chem. 146 (1990) 13.
[11] V. Sepelak, K. Tkacova, V.V. Boldyrev, U. Steinike, Mater. Sci.
Forum. 783 (1996) 228.
[12] V. Sepelak, A.Yu. Rogachev, U. Steinike, D.Chr. Uecker,
S. Wibmann, K.D. Becker, Acta Crystallogr. Suppl. A 52 (1996)
C367.
[13] V. Sepelak, A.Yu. Rogachev, U. Steinike, D.Chr. Uecker,
F. Krumcich, S. Wibmann, K.D. Becker, Mater. Sci. Forum 139
(1997) 235.
[14] V. Sepelak, U. Steinike, D.Chr. Uecker, S. Wibmann, K.D. Becker,
J. Sol. State Chem. 135 (1998) 52.
[15] R.J. Hill, J.R. Craig, G.V. Gibs, Phys. Chem. Miner. 4 (1979)
317.
[16] C.P. Marshall, W.A. Dollase, Am. Miner. 69 (1984) 928.

Materials Chemistry and Physics 120 (2010) 509517

Contents lists available at ScienceDirect

Materials Chemistry and Physics


journal homepage: www.elsevier.com/locate/matchemphys

Wet chemical synthesis and gas sensing properties of magnesium zinc ferrite
nano-particles
D.C. Bharti, K. Mukherjee, S.B. Majumder
Materials Science Center, Indian Institute of Technology, Kharagpur 721302, West Bengal, India

a r t i c l e

i n f o

Article history:
Received 22 June 2009
Received in revised form
25 November 2009
Accepted 28 November 2009
Keywords:
Semiconducting oxide
Solgel growth
X-ray diffraction
Adsorption

a b s t r a c t
In the present work we have synthesized, magnesium ferrite, zinc ferrite, and magnesium zinc ferrite solid
solutions using an economic wet chemical synthesis route. To understand the phase formation behavior of
the synthesized powders, infra-red spectroscopy in conjunction with X-ray Rietveld renement analyses
has been performed. The structural characteristics of these ferrite powders are correlated with their
room temperature magnetic properties. Phase pure, low temperature synthesized magnesium zinc ferrite
nano-particles are investigated in terms of carbon monoxide and hydrogen gas sensing properties. The
response and recovery transients of conductance are modeled using Langmuir adsorption kinetics with
two active sites in the sensing elements used as CO sensors. For these two adsorption sites the activation
energies for response and recovery behavior, estimated from the temperature dependence of respective
time constants are found to be different. The difference in respective activation energies for response as
well as recovery is thought to be due to different chemi-adsorbed oxygen species in these two sites.
2009 Elsevier B.V. All rights reserved.

1. Introduction
Cubic spinel ferrites are one of the most attractive materials for
magnetic, catalysis, micro-electronic, as well as gas sensing applications [15]. The magnetic cations in cubic spinel lattice occupy A
(divalent, tetrahedrally coordinated with oxygen) and B (trivalent,
octahedrally coordinated with oxygen) types of crystallographic
positions. In this type of materials, ferrimagnetisms may result
from stronger AB antiferromagnetic interaction than the BB antiferromagnetic interaction. Magnesium ferrite (MgFe2 O4 (MFO)),
zinc ferrite (ZnFe2 O4 (ZFO)) and the solid solutions between these
two crystallize into cubic spinel structure and have several important technological applications. The cation distribution in MFO is
predominantly inverse whereas ZFO crystallizes in normal spinel
structure. The reported magnetic properties of these materials are
scattered and strongly depend on their synthesis routes. As for
example, for MFO powders prepared by high energy ball mill, secondary phase formation (e.g. MgO, Fe2 O3 ) as well as cation mixing
(inverse or mixed) is reported to be related to the milling time
[6]. In mechano-chemical synthesis too the co-existence of Fe2 O3
with MFO is related to milling time [7]. MFO prepared using double hydroxide precursor have secondary MgO co-existed with MFO
[8]. Efforts have been made to correlate magnetic properties with
secondary phase contents; microstructure and compositional heterogeneity.

Corresponding author. Tel.: +91 3222 283986.


E-mail address: subhasish@matsc.iitkgp.ernet.in (S.B. Majumder).
0254-0584/$ see front matter 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.matchemphys.2009.11.050

From several recent literatures it is apparent that in nanocrystalline form, the properties of these ferrite materials are very
different from their bulk counterpart. As for example, below a critical particle size MFO exhibits super-paramagnetic behavior at room
temperature [9]. Since Mg2+ is diamagnetic, a weaker coupling
between Fe3+ cations reduces the anisotropic energy in MFO which
also yield super-paramagnetism at room temperature. These ultrane super-paramagnetic particles may have several applications
viz. as magnetic resonance imaging contrast agents, in ferro-uid
based technology, also in high density information storage device,
etc. Nano-crystalline ferrite particles have also reported to exhibit
superior gas sensing properties than their bulk counterpart [10].
Interestingly, some of these ferrite particles also exhibit distinct
selectivity for a specic gas in a gas mixture [11].
In the present work we have synthesized (Mg1x Znx )Fe2 O4
(0.0 x 1.0) solid solution particles using a simple, economic wet
chemical synthesis route. These powders are calcined in a temperature range between 300 and 900 C. The phase formation behavior
of MgFe2 O4 , ZnFe2 O4 and (Mg1x Znx )Fe2 O4 solid solution powders calcined at different temperatures are studied systematically
using Rietveld renement of the X-ray diffractograms. The morphology as well as room temperature magnetic properties of these
calcined powders are characterized using transmission electron
microscope (TEM) and a vibrating sample magnetometer vibrating
sample magnetometer (VSM).
We have also investigated the hydrogen and carbon monoxide
sensing performance of the low temperature synthesized phase
pure (Mg0.5 Zn0.5 )Fe2 O4 particles. The synthesized particles are
lightly pressed and sintered at low temperature in order to achieve

510

D.C. Bharti et al. / Materials Chemistry and Physics 120 (2010) 509517

a microstructure that contains both macro-pores and meso-pores


to yield molecular as well as Knudsen gas diffusion behavior [12].
The gas sensing performance of the ferrite is evaluated in a specially designed home made dynamic gas sensing measurement
system. The kinetics of the response as well as recovery behavior is investigated through the analyses of respective conductance
transients which are found to t reasonably well assuming Langmuir adsorption kinetics model [13]. The variation of relaxation
times during response and recovery as a function of gas concentration and temperature has been studied and the underlying
mechanism/(s) controlling the carbon monoxide sensing behavior
of Mg0.5 Zn0.5 Fe2 O4 is discussed.

2. Experimental
Nano-crystalline magnesium zinc ferrite powders are prepared using a wet
chemical synthesis route using hydrated magnesium acetate (Mg(CH3 COO)2 2H2 O),
hydrated zinc acetate (Zn(CH3 COO)2 2H2 O) and iron nitrate (Fe(NO3 )3 9H2 O) as precursor materials. For the powder synthesis, stoichiometric precursor materials of
Mg, Zn and Fe cations are dissolved in hot acetic acid through continuous stirring.
Poly-vinyl alcohol (PVA) solution (2%) is added into the mixed precursor solution
(in 1:0.25 volume ratio) and the resultant precursor is repeatedly heated 80 C and
cooled in an ice bath to yield a gelled mass. The gel is dried in a vacuum oven kept
about 80 C overnight to form powder. The resultant powder, termed as dried powder, is crushed in a mortar pestle and calcined in a two step heat treatment schedule.
In the rst step, the powders are red at 400 C for 1 h for the removal associated
organics followed by a second step calcination in a temperature range between 600
and 900 C for 2 h. The powders prepared below 400 C are calcined in a one step
heat treatment process in respective temperature for 2 h for crystallization.
The phase formation behavior of the calcined powders are studied using Fourier
transformed infra-red (FTIR) spectroscopy and X-ray diffraction analyses. A transmission electron microscope (TEM) is used to study the microstructure evolution.
The room temperature measured magnetic properties of the synthesized powder is
characterized using a VSM. The synthesized powders are mixed with few drops of
10% PVA solution as binder and pressed in the form of thin circular discs (12 mm
diameter 0.1 mm thick) using a hydraulic press. The pellets are heated at 600 C
for 2 h in air. The above heat treatment ensures minimal particle growth, necking
between particles, sufcient handling strength and porous nature of the sensing element. The sintered pellets are polished on a velvet cloth and cleaned ultrasonically
for 2 min each in acetone followed by water and absolute alcohol. For the electrical
measurements, one of the surfaces of the sensing elements are sputter coated with
planar gold stripes of 7 mm length, 3 mm wide and separated from each other by
4 mm. The sensing element is placed in a reactor equipped with pressure contact
probes and heater assembly with precise temperature control. For gas sensing studies hydrogen and carbon monoxide gas diluted in N2 (10,000 ppm) and compressed
air (95%+ purity) is used as test and carrier gas, respectively. The ow rates of test
(dVtest /dt) as well as carrier gas (dVcarrier /dt) are controlled by mass ow controllers
interfaced with a PC. Both test and carrier gas with controlled ow rate is fed to
the reactor through a home made mixing unit. During the sensing measurements
continuous ow of test as well as carrier gas is maintained through the reactor to
the exhaust line of it. The concentration of the mixed gas (Cmixed gas ) in the reactor is
calculated using the following relation:

Cmixed gas =

Ctest gas (dVtest /dt)


(dVtest /dt) + (dVcarrier /dt)

(1)

The probes are connected to an electrometer (6517A, Keithley Instruments)


which acts as a source-measure unit for resistance measurements. The mass ow
controllers (MFC) (PR 4000, MKS Technology and Products), electrometer and temperature controller (6400 West Instruments.) are all interfaced with a PC equipped
with a general purpose interface bus (GPIB) card (National Instruments) and Labview 8.5 (National Instruments) software. For the resistance measurements a xed
voltage of 1 V is applied to one of the planar electrodes and the surface current is
measured. In the voltage range between 1 and 5 V an Ohmic behavior is observed
and from the measured current, resistance is calculated. To evaluate the sensing
characteristics of (Mg,Zn)Fe2 O4 , its planar resistance is measured as a function of
time varying either the gas concentration at constant temperature or varying the
temperature keeping the gas concentration constant. From the measured value of
equilibrium resistance in air (Ra ) and gas (Rg ), the sensitivity (S) was calculated using
the relation


S=

Ra Rg
Ra


100

(2)

3. Results and discussion


3.1. Structural analyses of (Mg1x Znx ) Fe2 O4 (0.0 x 1.0)
(MZFO) powders
Fig. 1 shows the X-ray diffractograms of (a) x = 0.0, (b) x = 0.25, (c)
x = 0.50, and (d) x = 0.75 and (e) x = 1.0 as dried and calcined MZFO
powders in the temperature range of 300900 C for 2 h in air. Note
that for all these compositions, powders calcined at temperature
as low as 300 C, crystallize into cubic spinel structure. At lower
calcination temperature, minute quantity of unreacted MgO and
ZnO phase co-exist in Mg and Zn rich composition, respectively. At
higher calcination temperature (600 C) in some cases (especially
in Mg rich compositions) Fe2 O3 co-exists as impurity phase. It
is interesting to note that phase pure Mg0.5 Zn0.5 Fe2 O4 crystallizes
at 300 C (Fig. 1(c)). Comparing the FWHM of the XRD pattern of
Mg0.5 Zn0.5 Fe2 O4 calcined at different temperatures it is apparent
that powder calcined at 300 C may have ultra-ne crystallite size.
The space group of spinel ferrite is Fd3m. The cations are located
in tetrahedral and octahedral voids. The two most prominent infrared vibration modes (1 and 2 , respectively) for MgFe2 O4 (MFO)
are reported to be at 565 and 406 cm1 and for ZnFe2 O4 (ZFO)
are 555 and 393 cm1 , respectively [14]. From the detailed lattice
dynamics analyses it has been suggested that the tetrahedral bond
have the effect of substantially increasing the frequency for vibration along (tetrahedral) cation oxygen (CT O) axis. This would lead
to the IR absorption mode at 1 . The other absorption at relatively
lower frequency (2 < 1 ) is associated with the vibration of oxygen
ions in the direction perpendicular to the CT O axis. The potential
energy (V) of the system is dened as
2V = Kt

rt2 + K0

r02

(3)

where Kt and Ko are the force constants associated with unit displacement of the CT O and Co O bonds, and rt and r0 represents
the corresponding components of displacement from equilibrium
in the direction of the appropriate bond. From the measured 1 and
2 frequencies the force constants for the tetrahedral and octahedral cations can be determined [15]. If more than one type of cation
is present in the respective sites, the effective mass (summation
of the product of respective masses and their weight fraction) will
inuence the force constant which eventually leads to the shift of
the tetrahedral and octahedral frequencies. Therefore, the shift of
the frequencies is related to the cation distribution in the tetrahedral and octahedral sites. Finally, depending on the distribution of
the cations the cationoxygen bond lengths in the tetrahedral and
octahedral sites will be changed.
We have recorded the FTIR spectra of as prepared as well as calcined MFO and ZFO powders at different temperatures (not shown).
Two important features are apparent from these gures, rst: the
mode frequencies for tetrahedral cation/(s) are invariant to the calcination temperature and second the width of the absorption mode
decreases with the increase in calcination temperature. The atomic
weight of Mg, Zn, and Fe are 24.31, 65.37, and 55.84. Therefore, any
change of their fractional contents in A (tetrahedral) or B (octahedral) sites would change the values of 1 and 2 appreciably.
However, both for MFO and ZFO it is observed that the mode frequencies for tetrahedral sites (1 ) are invariant to the calcination
temperature; therefore, it seems that the cation distribution does
not alter with calcination temperature in both powders. The second
observation indicates that the crystallinity of MFO/ZFO powders
improves with the increase in calcination temperature. Similar to
that for the end members MFO and ZFO, for all their solid solution
compositions, we have observed that the mode frequency due to
tetrahedral bond does not shift appreciably with calcination temperature. As observed also in case of the end members MFO and

D.C. Bharti et al. / Materials Chemistry and Physics 120 (2010) 509517

511

Fig. 1. X-ray diffractograms of Mg1x Znx Fe2.0 O4 (0.0 x 1.0) powders calcined in the temperature range of 300900 C for 2 h in air.

ZFO, this indicates that cation occupancy in tetrahedral and octahedral site is invariant to the calcination temperature. However,
interesting variation of the mode frequency is observed when we
compare FTIR spectra of the solid solutions calcined at identical
temperatures. Thus, Fig. 2 shows the FTIR spectra of MZFO powders calcined at 900 C. Note that as the Zn contents in the solid
solution are increased the tetrahedral mode frequency is shifted
to lower wave number. This is expected as Zn has higher atomic
mass as compared to Mg ion in a specic crystallographic lattice
site.

In order to quantify structural and microstructural parameters,


we have performed Rietveld renement of the X-ray diffractograms
of the MZFO powders. The Rietveld renement is done for all the
MZFO powders calcined at 300900 C and a typical renement
plot is shown in Fig. 3 for MZFO powder calcined at 300 C for 2 h.
The rened composition, structural and microstructural parameters (i.e. phase, volume fraction, lattice parameter, rms microstrain,
and crystallite size) as well as the statistical tting parameters for
MFO, MZFO (50:50) and ZFO powders calcined at 300 C for 2 h
is tabulated in Tables 13, respectively. The continuous range of

512

D.C. Bharti et al. / Materials Chemistry and Physics 120 (2010) 509517

Table 1
Rened structural and microstructural parameters of magnesium Ferrite powders
calcined at 300 C for 2 h.
Magnesium ferrite MgFe2 O4 (MFO)
Phase/(s) present (SG)
MFO cubic spinel (Fd-3m) 82 wt%
MgO cubic (Fm-3m) 18 wt%
Calcination temperature 300 C
Rened parameters (MFO)
Lattice parameter () 8.38 (0.0014)
Rms microstrain 0.0017%
Crystallite size (nm) 81.7 (2.09)
Density (gm cm3 ) 4.49 gm cm3
Rened parameters (MgO)
Lattice parameter () 4.23 (0.0011)
Crystallite size (nm) 145 (9.9)

Rened composition
(Mg0.14 Fe0.86 )T (Mg0.86 Fe1.14 )O O4
Statistical parameters
RWp 5.6%
RP 4.87%
Rwbp 4.47%
Rpb 4.23%

Table 3
Rened structural and microstructural parameters of ZFO powders calcined at 300 C
for 2 h.
Zinc ferrite ZnFe2 O4 (ZFO)
Rened composition
Phase/(s) present (SG)
(Zn0.92 Fe0.08 )T (Zn0.08 Fe1.92 )O O4
ZFO cubic spinel (Fd-3m) 93 wt%
Statistical parameters
ZnO cubic (P63mc) 7 wt%
RWp 6.93%
RP 5.15%
Calcination temperature 300 C
Rened parameters (ZFO)
Rwbp 8.55%
Lattice parameter () 8.43 (0.0014)
Rpb 6.54%
Rms microstrain 0.0085%
Crystallite size (nm) 38.0 (0.69)
Density (gm cm3 ) 5.37 gm cm3
Rened parameters (ZnO)
Lattice parameter () a = 3.25 (0.0011), c = 5.20
Crystallite size (nm) 100 (9.9)

Atom

(2 )

Atomic coordinates

MgT
FeT
MgO
FeO
O

0.125
0.125
0.5
0.5
0.255

0.125
0.125
0.5
0.5
0.255

0.125
0.125
0.5
0.5
0.255

1.21
1.21
0.37
0.37
0.713

Atom

(2 )

ZnT
FeT
ZnO
FeO
O

0.125
0.125
0.5
0.5
0.255

0.125
0.125
0.5
0.5
0.255

0.125
0.125
0.5
0.5
0.255

0.68
0.68
0.68
0.68
0.67

the solid solution is conrmed by systematic change in the lattice


parameter as a function of Zn contents in MFO lattice. Fig. 4 shows
the variation of the lattice parameter of MZFO powders calcined
at 300 and 900 C. As shown in the gure, the lattice parameter is
increased with the increase in calcination temperature as well as Zn
contents in MFO lattice. Similar to the end members MFO and ZFO,
the crystallite size of MZFO solid solution was found to be increased
with the increase in calcination temperature. For Mg0.5 Zn0.5 Fe2.0 O4
(MZFO 50:50) composition calcined at 300 and 900 C the crystallite
size was found to be 11.1 and 151 nm, respectively. The particle size
of MZFO 50:50 powders calcined at 300 and 900 C was also investigated using a transmission electron microscope (not shown). It is
interesting to note that the particles are spherical and their average
size matches quite well with the X-ray derived crystallite size only
in case of powder calcined at elevated temperature. The crystallite size corresponds to the regions of coherent X-ray diffraction
and these regions are single crystalline in nature. In the particles calcined at lower temperature (300 C), the crystallite size
as determined by XRD is smaller in size (11 nm). These crystallites (termed as primary particle) are aggregated to form relatively
bigger particles (termed as secondary particles). The secondary particles are not single crystalline in nature and the particle size as
determined by TEM studies (corresponding to the secondary particles) are therefore bigger than that has been estimated from the
X-ray diffractograms. When these particles are calcined at higher

temperature the individual crystallites in the secondary particles


grow and each individual particle becomes single crystalline to
yield coherent X-ray diffraction. Therefore, the particle calcined at
higher temperature the crystallite size matches quite well with the
particle size estimated by TEM measurements.
3.2. Magnetic properties of MZFO calcined powder
We have measured the room temperature hysteresis loops of
MgFe2 O4 (MFO) powders calcined at 300 and 900 C for 2 h. For both
the cases a well saturated magnetic hysteresis loops are obtained.
For powders calcined at 300 and 900 C the remnant, saturation
magnetization and coercive eld values are 7.01, 38.45 emu g1
and 84.5 Oe and 5.22, 22.76 emu g1 , 80.52 Oe, respectively. The
saturation magnetization of the MFO powder calcined at lower
temperature (300 C) is close to that reported in bulk MgFe2 O4
(30 emu g1 ) by others in the literature [16]. The reduction in the
magnetization values of MFO powder calcined at higher temperature (900 C) is thought to be due to the presence of Fe2 O3
impurity phase as emerged in X-ray diffractograms presented
earlier (Fig. 1(a)). It is interesting to note that the measured saturation magnetization is very close to the calculated magnetic
moment (33 emu g1 ) assuming the mixed spinel rened com-

Table 2
Rened structural and microstructural parameters of MZFO (50:50) calcined at 300 C for 2 h.
Magnesium zinc ferrite (Mg0.5 Zn0.5 )Fe2 O4
Phase/(s) present (SG)
Cubic spinel (Fd-3m) 100 wt%
Calcination temperature 300 C
Rened parameters
Lattice parameter () 8.42 (0.0064)
Rms microstrain 0.0007%
Crystallite size (nm) 11.1 (0.143)
Density (gm cm3 ) 4.89 gm cm3

Statistical parameters
RW 5.04%
Rwnb 4.81%
R 4.05%
Rnb 4.23%
Rened composition
(Mg0.04 Zn0.46 Fe0.50 )T (Mg0.46 Zn0.04 Fe1.50 )O O4

Atom

(2 )

MgT
ZnT
FeT
MgO
ZnO
FeO
O

0.125
0.125
0.125
0.5
0.5
0.5
0.2555

0.125
0.125
0.125
0.5
0.5
0.5
0.2555

0.125
0.125
0.125
0.5
0.5
0.5
0.2555

1.28
1.28
1.28
1.12
1.12
1.12
1.21

D.C. Bharti et al. / Materials Chemistry and Physics 120 (2010) 509517

513

Fig. 3. X-ray Rietveld renement plot of Mg0.5 Zn0.5 Fe2.0 O4 powders calcined at
300 C for 2 h in air.

materials the zero eld cooled magnetization curve deviates from


the eld cooled magnetization curve at a characteristic blocking
temperature (TB ). Below the blocking temperature one observes
magnetic hysteresis in super-paramagnetic materials. In line to
this, further temperature dependent magnetization measurements
are required to conrm the super-paramagnetic behavior in the
measured composition as well as in other solid solution compositions.
3.3. Gas sensing performance of porous (Mg0.5 Zn0.5 )Fe2 O4 pellet

Fig. 2. FTIR spectra of Mg1x Znx Fe2.0 O4 (0.0 x 1.0) powders calcined at 900 C for
2 h in air.

position (Mg0.14 Fe0.86 ) (Mg0.86 Fe1.14 ) and density (4.49 g cm3 ) as


presented in Table 1. As emerged through the Rietveld structural
renement studies presented above, ZnFe2 O4 powders crystallize
into a mixed spinel structure with a very low degree of cation
mixing ( 0.08). In that respect, ZFO can be considered to have
a normal spinel structure. ZFO powder exhibits a typical paramagnetic behavior similar to that reported in the literature [17].
Fig. 5 shows the room temperature hysteresis loops of MZFO
50:50 powders calcined at 300 and 900 C for 2 h in air. As envisaged from the hysteresis loop, both the powders calcined at 300
and 900 C, exhibit S type hysteresis loops and the magnetization increases with the increase in calcination temperature. As
mentioned earlier, ZFO exhibits typical paramagnetic ordering at
room temperature. In magnesium zinc ferrite replacing part of Mg
cation by Zn would reduce the strength of the exchange interaction
between the tetrahedral (A) and octahedral (B) cations. This may
induce the super-paramagnetic behavior in MZFO solid solutions.
Although in several recent literature report super-paramagnetic
behavior has been concluded merely from room temperature hysteresis loop measurements [15], ideally one should get it conrmed
from the zero eld cooled and eld cooled magnetization measurement. In zero eld cooled measurement, the system is cooled
fast in zero magnetic elds. Then the magnetization is measured as
temperature T is increased. A eld cooled magnetization curve is
obtained by measuring magnetization at step wise decreasing temperature applying a weak magnetic eld. In super-paramagnetic

As envisaged from the Rietveld structural renement analyses,


only the (Mg0.5 Zn0.5 )Fe2 O4 powders calcined at 300 C are phase
pure and crystallized in cubic spinel structure. These powders are
pressed in the form of circular pellet, sintered at 600 C for 2 h and
studied in terms of their H2 and CO sensing properties. In the presence of hydrogen gas in the concentration range of 1001660 ppm,
the resistance transient of Mg0.5 Zn0.5 Fe2 O4 porous pellet (measured at an operating temperature 380 C) is shown in Fig. 6. The
gas on and off state at each of this gas concentration is denoted
by down and up arrow, respectively. When the gas ow is turned
off, the sensor recovery is achieved by owing 500 ppm air into
the measurement unit. Minimal base-line drift is observed when

Fig. 4. Variation of the lattice parameter of Mg1x Znx Fe2.0 O4 (0.0 x 1.0) powders
calcined at 300 and 900 C for 2 h in air.

514

D.C. Bharti et al. / Materials Chemistry and Physics 120 (2010) 509517

Fig. 6. The resistance transient of Mg0.5 Zn0.5 Fe2 O4 porous pellet in the presence
of different concentration of hydrogen gas, measured at an operating temperature
380 C.

behavior can also be explained using the depleted layer width (LD )
concept. For the semiconducting material the charge carrier concentration (n0 ) (and therefore the conductance) increases with the
increase in temperature. The relation between the depletion layer
width (LD ) with carrier concentration (n0 ) and response (S) is given
in Eqs. (4) and (5), respectively [18]
LD =
S=

Fig. 5. Room temperature magnetic hysteresis loops of Mg0.5 Zn0.5 Fe2.0 O4 powders
calcined at 300 and 900 C for 2 h in air.

the sensing element is switched back and forth between air and
test gas environment. The sensitivity is found to increase from
10% to 62% when the H2 concentration was increased from 100 to
1660 ppm. For a xed gas concentration (H2 /CO 1660 ppm), Fig. 7
shows the temperature dependent sensitivity of Mg0.5 Zn0.5 Fe2 O4
sensors. As shown in the gure, in case of CO sensing a maximum sensitivity (44%) is achieved at operating temperature about
320 C. However, in case of H2 sensing the sensitivity increases with
the operating temperature of the sensing element. The slope of S
vs. T, however, starts to decrease 350 C and as envisaged from
the gure, probably the sensitivity (S) would have a tendency to
decline beyond 380 C. Due to the limitation of the heating unit
we could not measure the sensitivity beyond 380 C. Several physical phenomena could be invoked to explain the observed behavior.
For example, such reduction of sensitivity at higher temperature
may be explained by the adsorptiondesorption principle. Thus the
enhancement of desorption rates at the sensor surface may lead to
the decreased sensitivity at higher temperature. Additionally, the

 kT 1/2
0
n0 e2

n
LD
n0

(4)
(5)

where 0 is the static dielectric constant, n0 is the total carrier concentration, e is the carrier charge, k is the Boltzmann
constant, T is the absolute temperature, n is the change in
the carrier concentration. From these equations it is clear that
beyond a critical temperature, the sensitivity is reduced due to the
decrease in depletion layer width. Table 4 compares the H2 and
CO sensing characteristics of various cubic spinel ferrite reported
in recent literatures. Comparing these data, it is ascertained that
Mg0.5 Zn0.5 Fe2 O4 prepared by a simple economical wet chemical
synthesis route exhibit reasonably good CO and H2 sensing behavior.
The sensing mechanism for the reducing gases such as H2 (or
CO) involves the formation of electron accepting adsorbates on the
surface of ceramic oxides as described in several recent research

Fig. 7. Temperature dependent H2 and CO sensitivity of Mg0.5 Zn0.5 Fe2 O4 sensor.

D.C. Bharti et al. / Materials Chemistry and Physics 120 (2010) 509517

515

Table 4
H2 and CO sensing characteristics of various cubic spinel ferrites.
Composition

Synthesis route

Morphology

Measurement system

Sensor characteristics

Ref.

Topt ( C)

H2 (ppm)

CO (ppm)

ZnFe2 O4
CuFe2 O4
CuFe2 O4
ZnFe2 O4
CoFe2 O4
NiFe2 O4
CuFe2 O4
ZnFe2 O4
CdFe2 O4
MgFe2 O4
NiFe2 O4
BiFe0.6 Mn0.4 O3
CdFe2 O4

Molten salt
Solid state reaction

Thick lm
Powder

Dynamic
Static

250
332

45
20

200
1000

Chemical route

Thick lm

Dynamic

250400

40 (H2 )
40(CO)

NA

NA

Chemical route

Thick lm

Static

250400
40 (CO)

40 (H2 )

6000
6000

Chemical route
Solgel method
Coprecipitation

Pellet
Thick lm
Thick lm

Dynamic
Static
Static

400
300
350

Mg0.5 Zn0.5 Fe2 O4

Wet chemical synthesis

Porous pellet

Dynamic

380
320

20 (H2 )
22 (CO)
25 (H2 )
25 (CO)
62 (H2 )
44 (CO)

200

1000

1660
1660

1000

1000

[19]
[20]
[21]

[22]

[23]
[24]
[25]
This work

Table 5
Fitted parameters for the response kinetics as a function of temperature at xed (1660 ppm) CO gas concentration.
response

Temperature ( C)

1

250
300
350
380

267.25
193.57
102.46
77.54

(s)

response

2

(s)

270.25
207.05
142.24
93.05

reports [26,27]. The rate of the increase of conductance (in response


transients) is in fact proportional to the steady state adsorbate
(gas molecules) concentration on the sensor surface. The rate of
adsorbate concentration () could be estimated from the following
equation of rst order kinetics [9]
d
= Ka (1 )[Cgas ] Kd []
dt

(6)

where Ka and Kd are the adsorption and desorption rate constants


and Cgas is the concentration of the adsorbate gas. The steady state
adsorbate coverage [] of the gas molecule, which is proportional
to the conductance (G) can be estimated by the following relation

=

Ka Cgas
Ka Cgas + Kd

G0 (mho)

G1 (mho)

G2 (mho)

8.44e9
2.59e8
8.60e8
1.49e7

2.80e9
1.12e8
2.72e8
2.43e8

2.86e9
1.31e8
2.73e8
2.57e8

Accordingly, the conductance recovery is given by

G(t)recovery = G0 + G1 exp

(7)

For single adsorption site the conductance has the following


form

 t 

(8)

1

where G = G0 , when t = 0, G1 is a constant and  1 = (Ka Cgas + Kd )1 is


the response time.
For recovery kinetics, Cgas = 0 and from Eq. (6) one can write
d
= Kd 
dt

(9)

integrating Eq. (9) with the boundary condition at t = 0,  = 1 one


gets
 = 0 exp(Kd t)

(10)

(11)

recovery

1

G(t)response = G0 + G1 1 exp

[1 exp((Ka Cgas + Kd ) t)]

Attempts are made to t the (response) conductance transients


with a single site Langmuir isotherm (Eq. (8)). The ts were poor
with typical R2 0.70. On the other hand the curve ts well with R2
value >0.95 when two adsorption sites were assumed with modied G(t)response given by the following equation

G(t)response = G0 + G1 1 exp

+ G2 1 exp
response

t
response

1

(12)

response

2

response

and 2
are the relaxation times for two differwhere 1
ent energetically different adsorption sites 1 and 2, respectively.
Similar to response the recovery is also found to t well assuming two energetically different adsorption sites using the following
equation

recovery

G(t)

G0

+ G1

exp

t
recovery
1

+ G2 exp

t
2 recovery

(13)

Fig. 8 shows the conductance transient of Mg0.5 Zn0.5 Fe2 O4 used


as CO gas sensor at 500 ppm (measured at T 300 C) and best
tted curve using Eq. (12) for response and Eq. (13) for recovery. Similar tting is also performed for the response and recovery
transients of Mg0.5 Zn0.5 Fe2 O4 sensors measured at other operating

Table 6
Fitted parameters for the recovery kinetics as a function of temperature at xed (1660 ppm) CO gas concentration.
recovery

Temperature ( C)

1

250
300
350
380

124.92
115.69
103.25
94.35

(s)

recovery

2

124.92
120.68
112.14
105.65

(s)

G0 (mho)

G1 (mho)

G2 (mho)

9.26e9
2.66e8
8.76e8
1.66e7

2.17e9
1.35e8
2.43e8
1.82e8

2.18e9
1.33e8
2.68e8
1.88e8

516

D.C. Bharti et al. / Materials Chemistry and Physics 120 (2010) 509517

Fig. 8. Conductance transient for response and recovery of Mg0.5 Zn0.5 Fe2 O4 sensors
for 500 ppm carbon monoxide measured at 300 C. The solid lines are calculated
using Eq. (12) (for response) and Eq. (13) (for recovery) (see text).
recovery

recovery

Fig. 10. Plots of ln 1


and ln 2
as a function of 1000/T for Mg0.5 Zn0.5 Fe2 O4
gas sensor in presence of 1660 ppm CO gas.

temperatures. The tted parameters of the conductance transient


of response as well as recovery for CO 1660 ppm measured at
different operating temperatures are tabulated in Tables 5 and 6,
respectively. When the kinetics of sensor response is controlled
by adsorption/desorption process the response time constant ()
usually follow the following temperature dependence [28]
 = 0 exp

E + Q 
A

(14)

2kT

where EA + Q is the activation energies for the adsorption of CO gas.


response
response
Fig. 9 shows the plots of ln 1
and ln 2
as a function
of 1000/T. From the slopes of these curves, activation energies for
adsorption process at site 1 and 2 were found to be 0.57 and 0.46 eV,
respectively. As expected the two different adsorption sites have
different activation energies for CO adsorption. The recovery time
recovery
recovery
constants for site 1 and 2 1
, 2
are assumed to follow
following relation
 = 0 exp

E 
D

(15)

2kT

where ED is the desorption energy for RO gas. Accordingly, as shown


in Fig. 10 ln  varies linearly with 1000/T and from the linear t the
activation energies of desorption for site 1 and 2 were estimated to
be 0.13 and 0.07 eV, respectively.

response

response

Fig. 9. Plots of ln 1
and ln 2
as a function of 1000/T for Mg0.5 Zn0.5 Fe2 O4
gas sensor in presence of 1660 ppm CO gas.

In light of the above analyses, we are making an attempt to comment on the nature of site 1 and site 2 as the different adsorption
sites in the magnesium zinc ferrite sensing elements. Note that as
compared to site 2, site 1 always exhibits faster response as well
response
recovery
response
recovery
as recovery kinetics (1
, 1
< 2
, 2
). This in
turn indicates faster CO adsorption and CO2 molecule desorption
from site 1 as compared to that from site 2. Probably, the sensor
block has macro-porous region (between secondary particles) and
meso-porous regions (between primary particles i.e. crystallites)
[29], then gas diffusion in macro-porous region (molecular) will be
faster (lower ) as compared to that in meso-porous regions (Knudsen diffusion). Finally, the difference in the respective activation
energies for site 1 and 2 (as derived from response and recovery
kinetics) probably indicates different types of oxygen are chemiadsorbed in these two sites. Further research is required to clarify
these issues.
4. Conclusions
In the present work we have synthesized multifunctional magnesium ferrite, zinc ferrite and magnesium zinc ferrite solid
solution nano-powders using chemical solution synthesis route.
Through FTIR analyses in conjunction with XRD Rietveld renement it is found that at a temperature as low as 300 C, MgFe2 O4
crystallizes into a mixed spinel structure with the degree of inversion () 0.89. The degree of inversion is invariant to calcination
temperature. However, it is difcult to obtain phase pure MgFe2 O4
powders. The MgFe2 O4 powder calcined at a temperature as low
as 300 C exhibits ferrimagnetic behavior with remnant and saturation magnetization 7.01 and 38.45 emu g1 , respectively. Zinc
ferrite also crystallized into a mixed spinel structure, however, with
the degree of inversion () only 0.16 it can be considered as a
normal spinel. Similar to that of magnesium ferrite, the degree of
inversion is also found to be invariant to calcination temperature.
The measurement of magnetization vs. magnetic eld supports
paramagnetic ordering of zinc ferrite at room temperature. From
the FTIR analyses it is observed that the tetrahedral mode frequency is shifted to lower wave number with the increase of Zn
contents in Mg0.5 Zn0.5 Fe2 O4 solid solutions. This is expected as Zn
has higher atomic mass as compared to Mg ion in a specic crystallographic lattice site. The continuous range of the solid solution is
conrmed by systematic change in the lattice parameter as a function of Zn contents in MFO lattice. The Mg0.5 Zn0.5 Fe2 O4 powder
yields saturation magnetization 26 emu g1 .

D.C. Bharti et al. / Materials Chemistry and Physics 120 (2010) 509517

The gas sensing performance of magnesium zinc ferrite is studied varying the carbon monoxide and hydrogen concentration in
the test gas (1001660 ppm) for various operating temperature (in
the range of 250380 C) of the sensing elements. It is argued that
for magnesium zinc ferrite sensors the chemi-adsorption of the
reducing gas and desorption of the reaction product (carbon dioxide
or water molecules) are the rate limiting processes for response and
recovery mechanisms, respectively. Through the analyses of conductance transients it is demonstrated that for the sensing of carbon
monoxide gas, both the sensor response and recovery behavior can
be modeled assuming a two site Langmuir adsorption kinetics. It
is found that for both the adsorption sites, at constant temperature T, Langmuir adsorption isotherm behavior is obeyed. For a
xed carbon monoxide concentration from the linear variance of
ln  response vs. 1/T, the activation energies for the carbon monoxide
gas adsorption at two different adsorption sites are estimated to
be 0.57 eV (for site 1) and 0.46 eV (for site 2). Similar to response
kinetics ln  recovery also follows a linear relationship with 1/T. The
activation energies for the desorption from site 1 and 2 are estimated to be 0.13 and 0.07 eV, respectively. Both for response and
recovery, the time constant for site 1 are faster than that for site
2. Similar to the response kinetics, for a xed temperature T, site 1
yields faster recovery than site 2. At lower temperatures, probably
the recovery trends for these two sites are reversed. Based on the
above ndings, nally it is argued that macro-porous regions of the
sensing elements are responsible to yield faster response as well as
recovery kinetics as compared to regions with meso-porous structures. The results are also indicative to the fact that the natures of
the chemi-absorbed oxygen species are also different in these two
different adsorption sites.
Acknowledgments
The above research work is partially supported by the research
grant from Department of Information Technology vide grant No.
20(7)2006-NANO. One of the authors (Mr. Kalisadhan Mukhejee)
also wishes to acknowledge the partial nancial support through

517

a JRF fellowship from the Council of Scientic and Industrial


Research. We would like to thank Mr. Rajasekhar for the magnetic
measurements of the samples.
References
[1] Y. Li, Q. Li, M. Wen, Y. Jhang, Y. Jhai, Z. Xie, F. Xu, S. Wei, J. Electron. Spectrom.
Relat. Phenom. 160 (2007) 1.
[2] P.Y. Lee, K. Ishizaka, H. Suematsu, W. Jiang, K. Yatsui, J. Nanopart. Res. 8 (2006)
29.
[3] Y. Zhihao, Z. Lide, Mater. Res. Bull. 33 (1998) 1587.
[4] G.R. Dube, V.S. Darshane, J. Mol. Catal. 79 (1993) 285.
[5] G. Zhang, C. Li, F. Cheng, J. Chen, Sens. Actuators B 120 (2007) 403.
[6] V. Sepelak, D. Baabe, D. Mienert, F.J. Litterst, K.D. Becker, Scr. Mater. 48 (2003)
961.
[7] J.G. Paik, M.J. Lee, S.H. Hyun, Thermochim. Acta 425 (2005) 131.
[8] S.K. Pradhan, S. Bid, M. Gateshki, V. Petkov, Mater. Chem. Phys. 93 (2005) 224.
[9] C. Liu, B. Zou, A.J. Rondinone, Z.J. Zhang, J. Am. Chem. Soc. 122 (2000) 6263.
[10] O.K. Varghese, C.A. Grimes, J. Nanosci. Nanotechnol. 3 (2003) 277.
[11] R.B. Kamble, V.L. Mathe, Sens. Actuators B 131 (2008) 205.
[12] N. Matsunaga, G. Sakai, K. Shimanoe, N. Yamazoe, Sens. Actuators B 83 (2002)
216.
[13] S. Aygun, D. Cann, J. Phys. Chem. B 109 (2005) 7878.
[14] R.D. Waldron, Phys. Rev. 99 (1955) 1727.
[15] C.P. Liu, M.W. Li, Z. Cui, J.R. Huang, Y.L. Tian, T. Lin, W.B. Mi, J. Mater. Sci. 42
(2007) 6133.
[16] S. Krupica, P. Novak, in: E.P. Wohfarth (Ed.), Ferromagnetic Material, vol. 3,
1982, p. 291.
[17] F. Li, H. Wang, L. Wang, J. Wang, J. Magn. Magn. Mater. 309 (2007) 295.
[18] S.K. Biswas, P. Pramanik, Sens. Actuators B 133 (2008) 449.
[19] S.L. Darshane, R.G. Deshmukh, S.S. Suryavanshi, I.S. Mullah, J. Am. Ceram. Soc.
91 (2008) 2724.
[20] Z. Sun, L. Liu, D.Z. Jia, W. Pan, Sens. Actuators B 125 (2007) 144.
[21] C.V.G. Reddy, S.V. Manorama, V.J. Rao, J. Mater. Sci. Lett. 19 (2000) 775.
[22] N.S. Chen, X.-J. Yang, E.-S. Liu, J.-L. Huang, Sens. Actuators B 66 (2000) 178.
[23] S.L. Darshane, S.S. Suryavanshi, I.S. Mulla, Ceram. Int. (2008), doi:10.1016/J.
Ceram. Int.2008.10.013.
[24] A.B. Bodade, A.V. Kadu, G.N. Chaudhari, J. SolGel Sci. Technol. 45 (2008)
27.
[25] Z. Tianshua, P. Hinga, Z. Jianchengb, K. Lingbing, Mater. Chem. Phys. 61 (1999)
192.
[26] N. Yamazoe, G. Sakai, K. Shimanoe, Catal. Surv. Asia 7 (2003) 63.
[27] M. Tiemann, Chem. Eur. J. 13 (2007) 8376.
[28] G. Korotcenkov, M. Ivanov, I. Blinov, J.R. Stetter, Thin Solid Films 515 (2007)
3987.
[29] G. Sakai, N. Matsunaga, K. Shimanoe, N. Yamazoe, Sens. Actuators B 80 (2001)
125.

Journal of Magnetism and Magnetic Materials 324 (2012) 18541861

Contents lists available at SciVerse ScienceDirect

Journal of Magnetism and Magnetic Materials


journal homepage: www.elsevier.com/locate/jmmm

Magnetic and structural studies on CoFe2O4 nanoparticles synthesized by


co-precipitation, normal micelles and reverse micelles methods
Ibrahim Shari a, H. Shokrollahi a,n, Mohammad Mahdi Doroodmand b,c,1, R. Sa a
a

Electroceramics Group, Materials Science and Engineering Department, Shiraz University of Technology, 71555-313, Shiraz, Iran
Chemistry Department, School of Sciences, Shiraz University, Shiraz, Iran
c
Nanotechnology Research Center, Shiraz University, Shiraz, Iran.
b

a r t i c l e i n f o

a b s t r a c t

Article history:
Received 1 June 2011
Received in revised form
31 December 2011
Available online 20 January 2012

Cobalt ferrite nanoparticles were synthesized by the chemical co-precipitation, normal micelles and
reverse micelles methods of iron and cobalt chlorides. X-ray diffraction analysis, Fourier Transform
Infrared (FTIR) and Vibrating Sample Magnetometer were carried out at room temperature to study the
structural and magnetic properties. X-ray patterns revealed the production of a broad single cubic
phase with the average particle sizes of  12 nm, 5 nm and 8 nm for co-precipitation, normal micelles
and reverse micelles methods, respectively. The FTIR measurements between 400 and 4000 cm  1
conrmed the intrinsic cation vibrations of spinel structure for each one of the three methods.
Moreover, the average particle sizes were lower than the single domain size (128 nm) and higher than
the super-paramagnetic size (23 nm) at room temperature. The results revealed that the magnetic
properties depend on the particle size and cation distribution, whereas the role of particle size is more
signicant.
& 2012 Elsevier B.V. All rights reserved.

Keywords:
Cobalt ferrite
Magnetic nanoparticle
X-ray diffraction

1. Introduction
Traditionally, magnetic materials have played a great number
of crucial roles in the daily life [1]. Among magnetic materials,
magnetic ceramics have been paid special attention due to
chemical stability as well as high electrical resistivity. One of
the most magnetic ceramics is spinel. According to the magnetic
theory, the most important spinels are oxides 2, 3 or MFe2O4 [2,3].
Among spinel ferrites, CoFe2O4 has received special attention
because of its large magneto-crystalline anisotropy, high coercivity, moderate saturation magnetization, large magneto-strictive
coefcient, chemical stability and mechanical hardness [4].
CoFe2O4 has an inverse spinel structure, where oxygen atoms
constitute an face center cubic (FCC) lattice and where half of the
Fe(III) ions occupies the tetrahedral A, site and where the other
half, together with Co(II) ions is located on the octahedral B site.
The different processing methods [57] inuence the particle size
and cation distributions between the A and B sites, as well as
the magnetic properties. The diameter of the nanoparticles is
much smaller than the critical single-domain diameter, thereby
avoiding the formation of the magnetic domain walls that

decrease the magnetization [8]. Cobalt ferrite (CoFe2O4) is one


of the most promising candidates for biological applications,
including magnetic resonance imaging (MRI), magnetic uid
hyperthermia (MFH), magnetic separations, biosensors, targeted
and controlled drug delivery [511]. Although the chemical
precipitation route is the most widely used process for the
synthesis of the magnetic nanocrystals with high simplicity and
good grain size control, it leads to the precipitation of the
nanocrystals with a relatively broad size distribution [12]. In
contrast, the nanocrystals, which are prepared by the micro
emulsion method, are generally very ne, shape-controlled and
highly crystalline, as compared to those which are synthesized by
other processes with narrow-sized distribution [1315].
This paper investigates the effect of cation distribution and
particle size in three widely used chemical methods, including coprecipitation and two types of micro emulsion: normal micelles
and reverse micelles methods on the magnetic and structural
properties of cobalt ferrite.

2. Experimental method
2.1. Materials

Corresponding author. Tel.: 98 711 7353509; fax: 98 711 7354520.


E-mail addresses: shokrollahi@sutech.ac.ir (H. Shokrollahi),
doroodmand@shirazu.ac.ir (M.M. Doroodmand).
1
Tel.: 98 711 6137363; fax: 98 711 2286008.
0304-8853/$ - see front matter & 2012 Elsevier B.V. All rights reserved.
doi:10.1016/j.jmmm.2012.01.015

All chemical reagentssodium dodecyl sulfate (SDS), ferric


chloride (FeCl3.6H2O), cobalt(II) chloride (CoCl2.6H2O), sodium
hydroxide (NaOH), petroleum oil and pyridine were acquired

I. Shari et al. / Journal of Magnetism and Magnetic Materials 324 (2012) 18541861

from E. Merck Co., Germany and used as they received without


further treatment.
2.2. Synthesis
2.2.1. Co-precipitation
The synthesis of the cobalt ferrite nanoparticles was prepared
by the co-precipitating aqueous solutions of CoCl2 and FeCl3
mixtures in alkaline medium. In addition, the mixed solutions of
CoCl2  6H2O (100 ml, 1.0 M) and FeCl3  6H2O (100 ml, 2.0 M) were
prepared and kept at 60 1C. This mixture was also added abruptly
to the boiling solution of NaOH (1200 ml, 0.63 M). Then the
solution was stirred with constant velocity. After that the solutions were maintained at 85 1C for 1 h. The time duration to form
the spinel ferrite was sufcient. A sufcient amount of ne
particles was collected at this stage by using magnetic separation.
These nanoparticles were washed several times with distilled
water, after that, with acetone and nally dried them at room
temperature [16].
2.2.2. Normal micelles
Cobalt(II) chloride (6.9 mmol) and iron(II) chloride
(14.9 mmol) were mixed in an aqueous solution (500 ml). An
aqueous surfactant solution of sodium dodecyl sulfate (SDS)
(38.3 mmol) in deionized water (500 ml) was also added to the
mixture. Then the mixture was stirred at room temperature for
30 min to form a mixed micellar solution of Co (DS)2 and Fe (DS)2.
Afterward, it was heated to 5565 1C. A solution of methylamine
(300 ml, 40%, w/w, aqueous solution) in deionized water (700 ml)
was heated to the same temperature and rapidly added to the
surfactant mixture. Black nanoparticles were precipitated. After
the reaction mixture was stirred vigorously for 3 h, the nanoparticles were isolated by centrifugation and washed with copious
amounts of deionized water, ethanol and hexanes through magnetic decantation. The nal product was dried in air at 50 1C
overnight to yield cobalt spinel ferrite nanoparticles [17].
2.2.3. Reverse micelles
In this study, the micro emulsion was used as a template to
control the size of magnetic nanoparticles. The micro emulsion
consists of three independent phases, including petroleum oil,
pyridine and partially polar species such as water or alcohol. To
synthesize micro emulsion briey, pyridine was mixed with
petroleum oil with a volume ratio of 1/10. After that this
mixture was shaken vigorously for an hour to obtain the partial
emulsion mixture. This step lasted 1 h. Then the emulsion was
reuxed at temperature to  70 1C in an inert atmosphere of
nitrogen for 2 h. Afterward the emulsion was annealed until the
temperature of the emulation reached the room temperature.
In this reverse micelle, a triplet system of emulsion/isobutanol/H2O was selected. The CoFe2O4 ferrite, aqueous solution
was prepared by mixing stoichiometric amounts of 0.5 M of FeCl3
and 0.25 MCoCl2  6 H2O. The two reverse micro-emulsions, ME1
and ME2, were also prepared. Then 33.33 wt% of aqueous
solution containing the precursor salts and 11.11 wt% of isobutanol were added to the oil under magnetic stirring. The stirring
continued for 1 h, resulting in a stable reverse micro-emulsion
ME1. Moreover, the reverse micro-emulsion (ME2) was prepared
with a 4.0 M aqueous solution of NaOH as the water phase under
similar conditions.
The reverse micro-emulsion ME2 was then heated to 80 1C and
added to reverse micro-emulsion ME1 dropwise under constant
magnetic stirring. After few minutes, the appearance of dark
brown color reveals the completion of the reaction and the
formation of the desired ferrite colloidal solution. Then the

1855

reaction mixture was further stirred for 4 h on magnetic stirrer


at 80 1C. Finally, the synthesized nanoparticles were isolated by
centrifugation and washed several times with distilled water and
then, with ethanol, hexane and acetone through magnetic decantation. The nal product was dried in air at 50 1C overnight to
yield cobalt spinel ferrite nanoparticles. In this research work, the
abbreviations of CNP, NNP and RNP are used for co-precipitation,
normal micelles and reverse micelles micro-emulsion methods,
respectively.
2.3. Characterization
The X-ray diffraction (XRD) patterns of the samples were
recorded on a BRUKER X-ray powder diffractometer by using
radiation. The scans of the selected diffraction
Cu-Ka1 (1.54060 A)
peaks were carried out in the step mode (step size 0.021,
measurement time 2 s, measurement temperature 25 1C and
standard: Si powder).The lattice parameters, the oxygen position
and the cation distribution were determined by means of Rietveld renement [18] and the reex program [19]. In order to
calculate the crystallite size, the Scherrers method was
applied [9]. After that the crystallite size was rened by Rietveld
renement. The magnetic measurements of the prepared powder
were determined at room temperature by using the vibrating
sample magnetometer (VSM, Dexing, Model: 250). The Fourier
Transform Infrared Spectroscopy (FTIR) spectrum was recorded as
(KBr) disks in the range 4004000 cm  1 by using the (FTIR
Shimadzu-8000) spectrophotometer.

3. Results and discussion


3.1. XRD analysis and cation distribution
Fig. 1 shows the single-phase spinel nature of the powders,
which were produced by three different methods and conrmed
by the X-ray diffraction pattern. The nal product is CoFe2O4 with
the expected inverse spinel structure and without any trace of
impurity peaks. The unit cell symmetries can be described by the
space groups fd3m. The reections are comparatively broader,
revealing the nanosize of the crystals with the highest intensity
and crystallinity in the co-precipitation method [20] Fig. 2.
The degree of crystallinity of ferrite in the different synthesis
routes was estimated by the background subtraction method
based on their XRD patterns. Table 1 lists the crystallite size,
degree of crystallinity and Rietveld renement parameters. The
quality of the renement was quantied by the corresponding
gures of merit: prole residual Rp, weighted prole residual Rwp
and goodness of t w2 [19]. The low intensity of the MNPs
obtained by microemulsion suggests poor crystallinity, and the
sharpness of the peaks reects the large grain size (Table 1). In
other words, the peak sharpness indicates a greater degree of
crystallinity.
The degree of crystallinity can be controlled by adjusting the
supersaturation during the nucleation and crystal growth processes. The main factors affecting the supersaturation degree are
temperature, pH value and the reactant concentration. In the
coprecipitation method the high crystallinity can be related to the
crystal growth at a faster rate than the nucleation growth. The
crystallinity of the sample synthesized by the copericipitation
method is greater than that of the samples prepared by microemulsion routes [21,22]. The size distribution of microemulsion
method is not as broad as those produced using the coprecipitation, but they are still polydisperse.
To calculate the average grain size from the broadening of the
XRD peaks of CoFe2O4, the Rietveld renement of XRD pattern

1856

I. Shari et al. / Journal of Magnetism and Magnetic Materials 324 (2012) 18541861

Fig. 1. (a) crystal structure of cobalt ferrite and XRD patterns of (b) co-precipitation, (c) normal micelles and (d) reverse micelles.

Table 1
Crystallite size, degree of crystallinity and Rietveld renement parameters.
Synthetic
method

Degree of
crystallinity

Crystallite
size (nm)

Rwp

Rp

w2

CNP
NNP
RNP

79.93
70.57
71.50

11.70
5.58
7.63

9.53
9.92
8.50

6.81
7.60
6.51

1.38
1.65
1.12

Table 2
Experimental lattice parameter (aexp), theoretical lattice parameter (ath), XRD
crystallite size (D), oxygen positional parameter (1/4,1/4,1/4) (U3m), oxygen
positional parameter (3/8,3/8,3/8) (u).
Synthetic Formula
method
CNP
NNP
RNP

Fig. 2. Conguration of the ion pairs in spinel ferrites with favorable distances and
angles.

was applied as shown in Fig. 1. Table 2 lists the cation distribution


Formula, experimental lattice parameter (aexp), theoretical lattice
parameter (ath), oxygen positional parameter (1/4,1/4,1/4) (U3m)
and oxygen positional parameter (3/8,3/8,3/8) (u). The difference
in the crystallite size is potentially due to the different preparation conditions which rise to the different rates of ferrite

aexp
(nm)

ath
(nm)

U3m (1/4, u (3/8,


1/4,1/4) 3/8,3/8)

A
B
Fe30:84
 Co20:84
Fe31:16
 O4 0.8403 0.8402 0.2614
Co20:16

A
B
Fe30:99
 Co20:99
Fe31:01
 O4 0.8391 0.8399 0.2603
Co20:01
Co2 Fe3 A Co2 Fe3 B O4 0.8401 0.8401 0.2611
0:12

0:88

0:88

1:12

0.3830
0.3832
0.3831

formation, thereby favoring the variation in crystallite size. The


micro-reactors in micro-emulsion not only control shape and size
but also decrease the aggregation process of crystals [15].
The distribution of cations over A and B sites is determined
mainly by the whole energy of the crystal that is dependent on
their ionic radius, the Coulomb energy, the crystal eld effects
and the ordering of the cations. The distribution of the cations on

A
B
these two sites can be expressed as:Co2x Fe31x
 Co21x
Fe31
x  O4 ,
where the square practices contain the ions on the octahedral

I. Shari et al. / Journal of Magnetism and Magnetic Materials 324 (2012) 18541861

sites, and x is a constant, which is rened by the Rietveld


method as it can be used to determine the cation distribution.
rA x

r B 12

r 2Co
1xr 3Fe

Lattice parameters were calculated by the Rietveld method.


In the co-precipitation method, the crystal structure tends to
behave like mixed spinel (Table 2), and achieve a high value of the
lattice parameter. The relation recommended by Mazen [23]
can calculate the lattice parameter theoretically
i
p
8 h
ath p r A RO 3r B RO
3 3

M e O

M e Me
b a=421=2

p a5=8u
q au1=43

c a=8111=2

r au1=8111=2

d a=431=2

s a1=3u 1=831=2

e 3a=831=2

f a=461=2
p

RA a 3 u14 RO


1=2
 RO
RB a 3u2 2:75u 43
64

The
where RO is the radius of the oxygen ions (R1.32 A).
agreement between ath and aexp (Table 2) achieved from X-ray
data indirectly conrms the cation distribution deduced from
X-ray intensity calculations. The small deviation between ath and
aexp may be due to the presence of some ferrous ions Fe(II)
(r2Fe 0.078 nm) on octahedral sites with larger radii than Fe(III)
(r3Fe 0.0645 nm) [24]. The oxygen, positional parameter and
anion parameter (u) for each composition were calculated using
the formula:
3m

hyperne eld of the A site is greater than that of the B site


(HA 4HB)

1=2

h
i

1xr 2Co
1 xr 3Fe

1857

1=4 R2 2=3 11=48R2 1=181=2

2R2 2

where R BO=AO.
Table 3 shows the average ionic radius per molecule of the
octahedral and tetrahedral sites (rA and rB), tetrahedral site radius
(Rt), octahedral site radius (Ro), ionic radii(r), tetrahedral edge
(Edt), shared octahedral edge (Edso), unshared octahedral edge
(Eduo) and average bond lengths, (AO and BO), /rMe r(O  2)S.
The highest value of BO length in the normal micelles methods
can be explained by the highest Co(II) ( 0.0745 nm ) occupancy
with a higher ionic radius, as compared to Fe(III) (0.0645 nm) [25].
Both Tables 4 and 5 list the inter-ionic distances and bond
angles [25,26] for cobalt ferrite, which were calculated by use of
the experimental values of crystallographic parameters based on
the Eq. (5). These parameters are necessary to give full description
of the crystallographic structure, and they are also in connection
with the magnetic properties. In order to explain this connection
in the normal micelles method the Fe(III) (5 mB) concentration of
the A site is higher than that of the B site, or the Co(II) (3 mB)
of the A site is lower than that of the B site, the magnetic

3.2. FTIR analysis


IR is used to collect to obtain information about the structure of
a compound. It is also utilized as an analytical tool for assessing the
purity of a compound. Fig. 3 and Table 6 show the FTIR absorption
bands of CoFe2O4 produced by co-precipitation, micro-emulsion
Table 5
Bond angles (degree) for CoFeO system.
Synthesis method

y1

y2

y3

y4

y5

CNP
NNP
RNP

122.690
122.636
122.658

142.150
142.348
142.181

93.879
93.965
93.930

126.144
126.175
126.155

72.744
72.603
72.661

Fig. 3. FT-IR spectra of the CoFe2O4 system.

Table 3
Tetrahedral site radius (Rt), octahedral site radius (Ro), ionic radii(r), tetrahedral edge (Edt), shared octahedral edge (Edso), unshared octahedral edge (Eduo), average bond
lengths(AO and BO).
Synthetic method

RA (nm)

RB (nm)

rA (nm)

rB(nm)

Edt (nm)

CNP
NNP
RNP

0.0616
0.0616
0.0617

0.0715
0.0711
0.0714

0.0662
0.0646
0.0657

7 0.0002 nm
0.0687
0.3162
0.0694
0.3163
0.0689
0.3164

Edso (nm)

Eduo (nm)

AO (nm)

BO (nm)

0.2776
0.2772
0.2776

0.2974
0.2970
0.2973

0.1982
0.1966
0.1977

0.2008
0.2014
0.2009

Table 4
Inter-ionic distances for CoFeO system.
Synthetic method

b (nm)

c (nm)

d (nm)

e (nm)

f (nm)

p (nm)

q (nm)

r (nm)

s (nm)

CNP
NNP
MNP

0.2971
0.2967
0.2970

0.3484
0.3479
0.3483

0.3638
0.3634
0.3638

0.5458
0.5450
0.5457

0.5146
0.5139
0.5144

0.2033
0.2029
0.2032

0.1936
0.1936
0.1937

0.3708
0.3708
0.3710

0.3678
0.3673
0.3677

1858

I. Shari et al. / Journal of Magnetism and Magnetic Materials 324 (2012) 18541861

Table 6
Tetrahedral molecular weights (M1), band frequency (u1) and force constant (kt).
Synthesis method

M1 (kg)  10  3

u1  10  2 m  1

kt  102 (N/m)

CNP
NNP
RNP

56.36
55.88
56.23

594
609
594

1.515
1.579
1.512

methods at room temperature in the wave number range of 400


4000 cm  1. It is obvious that the higher frequency band (u1) is
600 cm  1 and the lower frequency band (u2) is 400 cm  1. The
bands u1 and u2 are related to the intrinsic vibration of tetrahedral
and octahedral complexes [27]. The absorption bands observed
within this limit reveal the formation of the spinel structure. This
difference in the band positions is observed because of the difference
in the Fe3 O2 distance and (1) the perturbation occurring, thereby
(2) introducing Co(II) ions for tetrahedral and octahedral sites.
The splitting of the octahedral (B site) absorption band near
u2 is due to the presence of different kinds of cations, including
Co(II), Fe(III) and Fe(II) on the B site: [28]. Based on the small
effect of the production method on the octahedral peak shift and
the splitting effect, the exact calculation of the octahedral force
constant is difcult in the current study. This is attributed to the
JanTellar distortion produced by Fe(II) ions, which produces the
local deformation in the crystal eld potential, thereby splitting
the absorption band. The force constant for the tetrahedral site
(KT) was calculated by employing the method suggested by
Woldron [27]. The force constant is the second derivative of
the potential energy with respect to the site radius [25].
N
kt 7:62  M 1  u21  107 m

M1 is the molecular weights of cations on the A site. As the


bond length, AO, increases by increasing cobalt ion concentration on the tetrahedral site, (Tables 2, 3) the required energy for
breaking the longer bonds will decrease. Therefore, the force
constant in the normal micelles method on the A site is slightly
higher than the two other methods. The decrease in the bond
lengths is attributed to the decrease in the lattice parameter and
indicates the increase in the covalent characters. Srivastav and
Srinivasan [29] have stated that the band stretching for the
tetrahedral sites would lead to a higher force constant than that
evaluated for the octahedral site.
3.3. Magnetic properties
The magnetic properties of the materials which originate from
the quantum couplings at the atomic level, including the coupling
between electron spins (SS coupling) and the coupling between
the electron spin and the angular momentum of the electron
orbital (LS coupling) [30]. Each of the magnetic nanoparticles
tends to possess a single magnetic domain. Therefore, the nanoparticles provide excellent opportunities for the fundamental
studies on the relationship between magnetic behavior and the
magnetic couplings at the atomic level. The magnetization curves
of the cobalt ferrite nanoparticles measured at 27 1C are shown in
Fig. 4. It is clear that the synthesis method can inuence the
magnetization curves as a result of a change in the particle size
and distribution of cations.
3.3.1. Saturation magnetization
Table 6 illustrates the dependency of the saturation magnetization on the synthesis method. The MS value obtained for the
samples varies between 12.6 and 58.4 emu/g and is different from
the bulk value of 80.8 emu/g at room temperature for CoFe2O4 [30].

Fig. 4. MH curves of different prepared samples.

Table 7
Saturation magnetization, coercivity, remanence, maximum magnetic eld and
canted layer thickness measured at 27 1C.
Synthetic method MS (emu/g) HC (Oe) Mr (emu/g) Hmax (kOe) tshell (nm)
CNP
NNP
RNP

58.4
12.6
29.4

286.0
23.7
25.2

12.45
0.17
0.84

18.0
18.4
17.6

0.74
0.80
0.87

The high MS in the co-precipitation method is mainly due to large


particle sizes and is slightly due to the high cation difference
between the two sub-lattices (Table 2). These two effects are
discussed here in detail.

3.3.1.1. Size effect. When the diameter of the nanoparticles is


much smaller than the critical single-domain diameter, avoiding
the formation of magnetic domain walls decreases the
magnetization.
Even though the particle size becomes even smaller, the
thermal stability of the magnetization orientation decreases.
The limited superparamagnetic, as it is known, refers to the
particle size because the thermal energy at room temperature
causes uctuations in the magnetization orientation of a bit.
Magnetic anisotropies such as the magnetocrystalline anisotropy
energy (Ku), which aligns the easy magnetization axis of a
ferromagnet along a preferred crystallographic axis, stabilize the
magnetization against thermal uctuations. The anisotropy determines how strongly a particle can hold onto its magnetic
information [8]. The existence of some degree of the spin canting
in the whole volume of the nanoparticles, the disordered surface/
dead layer and the spin-glass properties at the surface can explain
the decrease of the saturation magnetization [31]. The dead
magnetic layer originates from the demagnetization of the surface
spin, causing the surface spins to be disordered or misaligned. It
also leads to the super-exchange interaction between the FeO
Fe bonds, caused by the termination of the bonds. This phenomenon weakens the total magnetization of the nanoparticles [31].
(Eq. (9)) [30] can calculate the dead layer thickness (t) of spherical
nanoparticles. Where, D is the particle size, Msat is the saturation magnetization and Msat-0 is the saturation magnetization at
0 K. The saturated magnetizations are about 80.8 emu/g and
93.9 emu/g at room temperature and 5 K, respectively [30]. It is
clear from Table 7 the dead layers depend not only on the particle

I. Shari et al. / Journal of Magnetism and Magnetic Materials 324 (2012) 18541861

1859

Table 8
Theoretical magnetization per unit chemical formula 9MA  MB9 and predicted
magnetization for bulk status were compared with reference [33] (MS-b) and
measured magnetization for nanosized synthesized powders (MS-n).
Synthesis
method

MA
(mB)

MB
(mB)

9MA  MB9
(mB)

MS-b
(emu/g)

MS-n
(emu/g)

CNP
NNP
RNP

4.670
4.980
4.752

8.330
8.020
8.248

3.660
3.040
3.496

87.2
72.5
83.3

58.4
12.6
29.4

size, but also on the saturation magnetization to the size ratio.


M sat M sat0 16t=D

3.3.1.2. Cation distribution. The magnetic moments on the A and


B sites are antiparallel oriented in ferrites [32], whereas those
on the B sites are parallel to each other. In the present study, by
assuming 5 mB for Fe(III) and 3 mB for Co(II) ions [33], the net
magnetic moment m/molecule9MA  MB9 is theoretically
calculated and the results are presented in Table 8.
It is obvious that the magnetization increases for the prepared
sample by the co-precipitation method as compared to the two
others. In addition, the larger particles in the co-precipitation
method can be attributed to the slightly higher concentration
difference of Fe3 with the 5 mB magnetic moment between A
and B sites (Table 2) by assuming the collinear ferrimagnetic
spin structure. It should be mentioned that the magnetic moment
of 3 mB per unit chemical formula for partially inversed Co ferrite
is equals 71.5 emu/g [33]. Table 8 lists the saturation magnetizations for bulk conditions. The difference in saturation magnetizations which is observed in this table conrms the role of particle
sizes, as compared to the cation distribution.
3.3.1.3. Exchange interactions. Intrinsic magnetic properties
depend on the exchange interactions, and the exchange
interactions strongly depend on the inter-ionic length and bond
angles. In our study, because of the main effect of the particle size,
the discussion about the exchange length and exchange angle is
slightly difcult. The high value of saturation magnetization in
nanoparticles has been obtained from co-precipitation method
that can be explained by the general decrease in the bond angles
for each site (AA and BB) and the increase between the two
sites (AB) (Tables 4, 5)
3.3.1.4. Degree of crystallinity. It is known that saturation
magnetization gradually increases with the crystalline size and
coercivity and is dened by decreased domain walls displacement
as the crystalline size increases in the multi-domain range
[21,22]. The energy of a magnetic particle in the external eld is
proportional to its crystallinity (Table 1) and particle sizes via the
number of molecules in a single magnetic domain. Therefore, the
decrease of the Ms value with the decrease of particle sizes and
crystallinity can be attributed to the surface effects that are the
result of nite-size scaling of nanocrystallites obtained by the
micro-emulsion method.
3.3.2. Magnetic coercivity (HC)
The magnetic coercivity of the nanoparticles depends signicantly on their magneto-crystalline anisotropy, micro-strain,
inter-particle interaction, temperature, size and shape [34]. In
the CoFe2O4 spinel, the Co(II) (3d7, 4F9/2, L3, S3/2, J 9/2)
cations have high spin ligand elds and possess seven d electrons
three of which are unpaired (Fig. 5). Evidently, the large magnetocrystalline anisotropy of EA of CoFe2O4 nanoparticles is due to the
strong LS couplings on the Co(II) cation sites. Energy barriers in

Fig. 5. Spinorbit (d7, Co2 ) stabilizations.

the individual nanoparticles [35] is considered and introduced by


the magnetocrystalline anisotropy energy of cobalt ferrite. This
effect results in the nanoparticles with multi-axial anisotropy,
and the zero coercivity can be achieved at very low particle sizes
(23 nm) but not at room temperature [36]. Obviously, by
reducing temperature, the anisotropy energy barriers in the
particles become sufciently high to prevent the reorientation
of the dipole moments by the stray eld. Upon cooling the
assembly of nanoparticles, the short-range order is preserved,
but with a signicant increase in the amount of disorder [35].
Table 6 lists the magnetic coercivity and magnetic remanence for
the nanoparticles which are obtained from three different synthesis methods with a uniform composition. It is clear that the
coercivity decreases from 286 Oe for co-precipitation to 23.7 Oe
for the normal micelle methods at room temperature. The
following table shows the obtained results:

3.3.2.1. Size effect. Below the critical single domain the particle
size of cobalt-ferrite (DC 128 nm) for three samples the
coercivity decreases as a result of a reduction in the particle
volume, strain size and magnetic anisotropy barrier to block the
ips of the magnetic moments. Based on the following
discussions, the coercivity is at the highest level for the
particles obtained from the co-precipitation route as result of
reducing in the particle volume, grain size and magnetic
anisotropy. According to the StonerWohlfarth theory [37], by
decreasing the particle size in the single domain region, KV
decreases based on Eq. (10)
EA KV sin2 y,

10

where EA is the anisotropic energy (shape or crystalline), K is the


magnetic anisotropy for a particle and V is the particle volume
and y is the angle between the magnetization direction and the
easy axis of the nanoparticle. The EA is reduced when the size of
the nanoparticles V and/or the anisotropy constant K (strength of
the LS couplings) is reduced. Furthermore, HC follows the D6
power law (Eq. (11)) [34], where A is the exchange stiffness
constant, PC is the unity order constant, MS is the saturation
magnetization, K1 is the magneto-crystalline constant and D is

1860

I. Shari et al. / Journal of Magnetism and Magnetic Materials 324 (2012) 18541861

dened as the grain size.


HC

PC K 41 D6
3
0 MS A

11

3.3.2.2. Cation distribution. There are some reasons that show the
difference in coercivity for nanoparticles obtained by the normal
and reverse micelle methods. In addition, the particle size can be
explained by the cation distributions. While quantum couplings
may not be the same on A and B sites with tetrahedral and
octahedral symmetries, respectively, the contribution to the
magnetic anisotropy from the LS couplings in Fe(III) and/or
Co(II) cations may not be the same when the cation distribution
changes.
3.3.2.3. Some particles below the super-paramagnetic limit. The
presence of further super-paramagnetic particles (zero coercivity
and zero remanence at room temperature) in the smaller distributed
particles leads to a decrease in HC. The average size of all the
particles is smaller than the critical single domain size (128 nm),
(Table 2) [38] and larger than the critical super-paramagnetic size at
room temperature (23 nm) for cobalt ferrite [36]. However, the
probability of nding further particles below the superparamagnetic limit at room temperature for the fewer particles
obtained from micro-emulsion methods is higher, as compared to
the co-precipitation method.

4. Conclusions
In this paper, CoFe2O4 powders were synthesized by three
widely used wet chemical methods, that is, co-precipitation,
normal micelle and reverse micelle methods with one starting
material/uniform chemical composition. The data obtained from
XRD, FTIR and VSM revealed that:
1. The crystallite size has a stronger effect on the magnetic
properties than the degree of crystallinity and the cation
distribution, respectively. Furthermore, the size effect covers
the role of slight difference of cation distribution.
2. The higher crystallinity in the coprecipitation sample might be
an additional contributing factor in the higher saturation
magnetization.
3. The nanocrystalline CoFe2O4 shows the absorption bands
 600 and 400 cm  1, respectively. The high frequency band
u1 around 600 cm  1 is attributed to the tetrahedral complexes
and the band u2  400 cm  1 corresponds to the octahedral
complexes.
4. The magnetic properties of the sample strongly depend on the
size of the nanoparticles. In our study, the magnetic properties
of Co ferrite are slightly dependent on the distribution of the
cations between A and B sites and depends considerably
on the particle size.

Acknowledgments
The authors wish to thank the Department of Chemistry of Shiraz
University and the Department of Chemistry at Shiraz University of
Technology, especially Dr. R. Khalifeh for his kindly constructive
scientic advice and Dr M. N. Soltani-Rad for his support.
References
[1] U. Jeong, X. Teng, Y. Wang, H. Yang, Y. Xia, Superparamagnetic colloids:
controlled synthesis and niche applications, Advanced Materials 19 (2007)
3360.

[2] H. Shokrollahi, K. Janghorban, Inuence of additives on the magnetic properties, microstructure and densication of MnZn soft ferrites, Materials
Science and Engineering, B 141 (2007) 91107.
[3] H. Shokrollahi, Magnetic properties and densication of manganesezinc soft
ferrites (Mn1  xZnxFe2O4) doped with low melting point oxides, Journal of
Magnetism and Magnetic Materials 320 (2008) 463474.
[4] G.A. El-Shobaky, A.M. Turky, N.Y. Mostafa, S.K. Mohamed, Effect of preparation conditions on physicochemical, surface and catalytic properties of cobalt
ferrite prepared by coprecipitation, Journal of Alloys and Compounds 493
(2010) 415422.
[5] M.A. Ahmed, A.A. EL-Khawlani, Enhancement of the crystal size and magnetic
properties of Mg-substituted Co ferrite, Journal of Magnetism and Magnetic
Materials 321 (2009) 19591963.
[6] S. Rana, J. Philip, B. Raj, Micelles based synthesis of cobalt ferrite nanoparticles and its characterization using Fourier transform infrared transmission
spectrometry and thermogravimetry, Materials Chemistry and Physics 124
(2010) 264269.
[7] S. Ayyappan, S. Mahadevan, P. Chandramohan, M.P. Srinivasan, John Philip,
Baldev Raj, Inuence of Co2 Ion concentration on the size, magnetic
properties, and purity of CoFe2O4 spinel ferrite nanoparticles, Journal of
Physical Chemistry C 114 (2010) 63346341.
[8] S.B. Darling, S.D. Bader, A materials chemistry perspective on nanomagnetism, Journal of Materials Chemistry 15 (2005) 4189.
[9] M.H. Youse, S. Manouchehri, A. Arab, M. Mozaffari, Gh.R. Amiri, J. Amighian,
Preparation of cobalt-zinc ferrite (Co0.8Zn0.2Fe 2O4) nanopowder via combustion method and investigation of its magnetic properties, Materials Research
Bulletin 45 (2010) 17921795.
[10] S.R. Dave, X. Gao, Monodisperse Magnetic Nanoparticles for Biodetection,
Imaging, and Drug Delivery: A Versatile and Evolving Technology, vol. 1, John
Wiley & Sons, Inc., 2009. 583609.
[11] R. Arulmurugan, B. Jeyadevan, G. Vaidyanathan, S. Sendhilnathan, Effect of zinc
substitution on CoZn and MnZn ferrite nanoparticles prepared by co-precipitation, Journal of Magnetism and Magnetic Materials 288 (2005) 470477.
[12] Zh Zi, Y. Sun, X. Zhu, Zh. Yang, J. Dai, W. Song, Synthesis and magnetic
properties of CoFe2O4 ferrite nanoparticles, Journal of Magnetism and
Magnetic Materials. 321 (2009) 12511255.
[13] M.P. Pileni, Reverse micelles as microreactors, Journal of Physical Chemistry
97 (1993) 69616973.
[14] M. Mahmoudi, Sh. Sant, B. Wang, S. Laurent, T. Sen, Superparamagnetic iron
oxide nanoparticles (SPIONs): development, surface modication and applications in chemotherapy, Advanced Drug Delivery Review 63 (2011) 2446.
[15] M.A. Lopez-Quintela, C. Tojo, M.C. Blanco, L. Garca Rio, J.R. Leis, Microemulsion dynamics and reactions in microemulsions, Current Opinion in Colloid
and Interface Science 9 (2004) 264278.
[16] G. Vaidyanathan, S. Sendhilnathan, R. Arulmurugan, Structural and magnetic
properties of Co1  xZnxFe2O4 nanoparticles by co-precipitation method,
Journal of Magnetism and Magnetic Materials. 313 (2007) 293299.
[17] Nam T.S. Phan, Ha V. Le, Superparamagnetic nanoparticles-supported phosphine-free palladium catalyst for the Sonogashira coupling reaction, Journal
of Molecular Catalysis A: Chemical 334 (2011) 130138.
[18] H. Rietveld, A prole renement method for nuclear and magnetic structures,
Journal of Applied Crystallography 2 (1969) 6571.
[19] R.A. Young, The Rietveld Method, Oxford University Press, 1993.
[20] S. Kumar, V. Singh, S. Aggarwal, U.K. Mandal, R.K. Kotnala, Inuence of
processing methodology on magnetic behavior of multicomponent ferrite
nanocrystals, Journal of Physical Chemistry C 114 (2010) 62726280.
[21] L. Zhao, H. Yang, L. Yu, Y. Cui, X. Zhao, B. Zou, S. Feng, Structure and magnetic
properties of nanocrystalline CoLa0.08Fe1.92O4 ferrite, Journal of Magnetism
and Magnetic Materials 301 (2006) 445451.
[22] N.M. Deraza Alaria, S. Shabanc, Structural, morphological and magnetic
properties of NiFe2O4 nano-particles, Journal of Alloys and Compounds 486
(2009) 501506.
[23] S. Mazen, M.H. Abbdallah, B.A. Sabrah, H.A.M. Hasham, The effect of titanium
on some physical properties of CuFe2O4, Physica Status Solidi A 134 (1992)
263271.
[24] M.T. Sebastian, Dielectric Materials for Wireless Communication, 1st edn.,
Elsevier, UK, 2008.
[25] S.S. Bhatu, V.K. Lakhani, A.R. Tanna, N.H. Vasoya, J.U. Buch, P.U. Sharma,
U.N. Trivedi, H.H. Joshi, K.B. Modi, Effect of nickel substitution on structural,
infrared and elastic properties of lithium ferrite, Indian Journal of Pure and
Applied Physics 45 (2007) 596608.
[26] A. Goldman, Modern Ferrite Technology, 2nd edn, Pittsburgh, PA, USA,
Springer.
[27] R.D. Waldron, Infrared spectra of ferrite, Physical Review 99 (1955)
17271735.
[28] U.N. Trivedi, K.H. Jani, K.B. Modi, H.H. Joshi, Study of cation distribution in
lithium doped nickel ferrite, Journal of Materials Science Letters 19 (2000)
12711273.
[29] C.M. Srivastav, T.T. Srinivasan, Journal of Applied Physics 53 (1982)
81488150.
[30] L.D. Tung, V. Kolesnichenko, D. Caruntu, N.H. Chou, C.J. OConnor, L. Spinu,
Magnetic properties of ultrane cobalt ferrite particles, Journal of Applied
Physics 93 (2003) 74867488.
[31] C. Caizer, M. Stefanescu, Magnetic characterization of nanocrystalline NiZn
ferrite powder prepared by the glyoxylate precursor method, Journal of
Physics D: Applied Physics 35 (2002) 30353040.

I. Shari et al. / Journal of Magnetism and Magnetic Materials 324 (2012) 18541861

[32] J. Samuel Smart, The Neel theory of ferrimagnetism, American Journal of


Physics 23 (1955) 356370.
[33] K. Hanedas, A.H. Morrish, Noncollinear magnetic structure of CoFe2O4 small
particles, Journal of Applied Physics 63 (1988) 42584260.
[34] Q. Zeng, I. Baker, V. McCreary, Zh. Yan, Soft ferromagnetism in nanostructured mechanical alloying FeCo-based powders, Journal of Magnetism and
Magnetic Materials 318 (2007) 2838.
[35] M. Georgescu, J. Viota, M. Klokkenburg, B. Erne, D. Vanmaekelbergh,
P. Zeijlmans van Emmichoven, Short-range magnetic order in two-dimensional cobalt-ferrite nanoparticle assemblies, Physical Review B 77 (2008).

1861

[36] Ch. Liu, B. Zou, A.J. Rondinone, Z.J. Zhang, Chemical control of superparamagnetic properties of magnesium and cobalt spinel ferrite nanoparticles
through atomic level magnetic couplings, Journal of the American Chemical
Society 122 (2000) 62636267.
[37] E.C. Stoner, E.P. Wohlfarth, A mechanism of magnetic hysteresis in heterogeneous alloys, The Philosophical Transactions of the Royal Society A 240
(1948) 599642.
[38] O. Philippova, A. Barabanova, V. Molchanov, A. Khokhlov, Magnetic polymer
beads: recent trends and developments in synthetic design and applications,
European Polymer Journal 47 (2011) 542559.

Journal of Magnetism and Magnetic Materials 324 (2012) 21002107

Contents lists available at SciVerse ScienceDirect

Journal of Magnetism and Magnetic Materials


journal homepage: www.elsevier.com/locate/jmmm

Microwave assisted combustion synthesis of CdFe2O4: Magnetic and


electrical properties
V. Vasanthi a, A. Shanmugavani a, C. Sanjeeviraja b, R. Kalai Selvan a,n
a
b

Solid State Ionics and Energy Devices Laboratory, Department of Physics, Bharathiar University, Coimbatore 641046, India
School of Physics, Alagappa University, Karaikudi 630003, India

a r t i c l e i n f o

a b s t r a c t

Article history:
Received 27 September 2011
Received in revised form
24 January 2012
Available online 22 February 2012

CdFe2O4 particles were synthesized by the microwave assisted combustion method using two different
fuelsglycine and urea. Microwave heating provides higher chemical yield within a minute. The
synthesized particles were characterized by X-ray diffraction (XRD), Fourier transform infrared
spectroscopy (FTIR), scanning electron microscope (SEM), ac impedance spectroscopy, vibrating sample
magnetometry (VSM) and electron spin resonance (ESR) methods. XRD analysis shows the cubic
structure of CdFe2O4. The high and low frequency absorption bands of CdFe2O4 were found using FTIR
analysis. Spherical morphology was revealed from the SEM images. ESR and VSM measurements reveal
the antiferromagnetic behavior of CdFe2O4. The electrical conductivities of CdFe2O4 synthesized using
glycine and urea are 6.5  10  7 S cm  1 and 4.7  10  8 S cm  1 respectively at 240 oC. At elevated
temperatures an occurrence of increase in conductivity was observed, which indicates the semiconducting behavior of CdFe2O4. The dielectric spectral analysis reveals that dielectric constant of CdFe2O4
decreases with frequency and increases with temperature.
& 2012 Elsevier B.V. All rights reserved.

Keywords:
Cadmium ferrite
Microwave-assisted combustion synthesis
Reaction mechanism
Structural
Magnetic and electric property

1. Introduction
Nanomaterials exhibit different physical and chemical properties compared to their bulk materials. It is well known that by
controlling the size of the particles it is possible to tune their
properties [1]. Recently magnetic nanoparticles of spinel ferrites
were focussed by many researchers because of their interesting
electrical and magnetic properties. Spinel type oxides with the
general formula AB2O4 have considerable application in information storage devices, magnetic bulk cores, magnetic uids, microwave absorbers, catalysts and medical diagnostics. Among them,
CdFe2O4 has normal spinel structure where Cd2 ions occupy the
tetrahedral (A) sites and Fe3 cations occupy octahedral sites.
Several synthetic approaches have been used to prepare CdFe2O4
particles including the combustion method [2], co-precipitation [3,4],
ball milling [5,6], sol gel [7], pulsed laser deposition (PLD) [8] and so
on. Detailed studies on their structural [2,5], and magnetic properties
[3,5,6] and electrical conductivity [8] have been investigated and
reported. CdFe2O4 is an important complex oxide in gas-sensing
materials. Many works have been carried out with regard to their
high sensitivity to volatile suldes such as CH3SH, H2S, and (CH3)2S,
and selectivity to ethanol was analyzed [4,7,8].
In the recent years, microwaves are used to synthesize many
organic and inorganic compounds with novel structures due its

Corresponding author.
E-mail address: selvankram@buc.edu.in (R. Kalai Selvan).

0304-8853/$ - see front matter & 2012 Elsevier B.V. All rights reserved.
doi:10.1016/j.jmmm.2012.02.018

attractive advantages such as rapid and uniform heating, high


reaction rate, high chemical yield, very less reaction time and
pure homogeneous product [9]. Bensebaa et al. [10] have prepared CoFe2O4 nanoparticles by co-precipitation using microwave
heating. Nanocrystalline photocatalysts of spinel MgFe2O4,
ZnFe2O4 and orthorhombic CaFe2O4 have been synthesized using
microwave sintering methods and they show that synthesis time
is reduced up to 60 times less than the required time in other
conventional methods [11]. Sertkol et al. [12] have synthesized
NiZn ferrite and Koseoglu et al. [13] have synthesized Co doped
ZnFe2O4 using microwave assisted combustion method. The
nanocrystalline core shell Zn0.7Ni0.3Fe2O4 has been prepared by
the microwave assisted synthesis by Sertkol et al. and a superparamagnetic behavior at room temperature was reported [14].
Moreover there is no previous literature regarding microwave
assisted combustion synthesis of CdFe2O4 particles and its
detailed magnetic and electrical properties at elevated temperatures were reported. This tempts us to employ the microwave
method for CdFe2O4 preparation. With this interest, cadmium
ferrite (CdFe2O4) particles are synthesized and characterized
using various techniques.

2. Experimental procedure
CdFe2O4 was prepared by the microwave assisted combustion
method using a domestic LG microwave oven operating at a

V. Vasanthi et al. / Journal of Magnetism and Magnetic Materials 324 (2012) 21002107

frequency of 2.45GHz with the output power of 800 W. For a


typical synthesis, to obtain 5 g of CdFe2O4, the stoichiometric amounts of Cd(NO3)2  4H2O (5.353 g), Fe(NO3)3
 9H2O (14.022 g) and CH2NH2COOH (1.302 g) are mixed thoroughly by grinding in an agate mortar for 20 min to get a
homogeneous mixture. The mixture was then transferred
to a quartz crucible and kept in a microwave oven with 80%
of output power. This mixture undergoes dehydration followed
by decomposition and spontaneous combustion with the
evolution of voluminous gases. The entire reaction was
completed within 60 s. The yield of the product is ca. 3.755 g
(ca. 75%). The obtained nal product is in the form of a
foamy powder and is calcined at 800 1C for 5 h. Similarly
the CdFe2O4 particles are also prepared using CO(NH2)2
(1.042 g). The representative reaction mechanism for the formation of CdFe2O4 is given in Scheme 1. It shows that the metal
nitrates are reacted with glycine to form an unstable organometallic compound (Steps 1 and 2) which further decomposed
into CdFe2O4 particles (Step 3).

Step 1:

+ NH2CH2COOH

Cd(NO3)2.4H2O

2101

Subsequently, the prepared CdFe2O4 particles were characterized by various techniques in order to analyze their structural,
morphological, magnetic and electrical properties. The compound
formation was studied by x-ray diffraction technique using
BRUKER Germany, D8 Advanced model with 2.2 kW Cu anode
as a source. Fourier transform infrared spectroscopy (FTIR) spectral analysis was done using Shimadzu/Nicolet instruments in the
range of 4000500 cm  1 and the morphological analysis was
performed through scanning electron microscope (SEM). The
magnetic properties were studied by vibrational sample magnetometer (VSM) Lakeshore 7410 and electron spin resonance
spectroscopy (ESR) JEOL JES-FA200. The electrical properties of
the material are studied using a computer controlled impedance
analyzer HIOKI 3532 LCR HITESTER at the frequency ranging from
50 Hz to 10 kHz. For conductivity measurements, the powder was
pressed into a pellet having diameter of 1 cm and a silver paste
was applied on both surfaces of the pellet before being sandwiched between two electrodes of the sample holder to get a
better ohmic contact. Hereafter, the synthesized CdFe2O4 particles

Basic
condition

2NH3+CH2COO- + Cd2+

H 2O
O

O
+

C- CH2NH3+

Cd2+

H3NH2C -C
O

O
H2O
unstable

Step 2:

2NH3+CH2COO- + Fe3+

Fe(NO3)3.9H2O + NH2CH2COOH
H2O
O3N

O
Fe

3+

C - CH2NH2
O

O3N
H2O
unstable
Step 3:

H2O
O

O
+

H2O

C- CH2NH3+

Cd2+

H3NH2C -C
O

O
H2O

O3N

O
3+

Fe

C-CH2NH2
O

O3N
H2O

decomposition

CdFe2O4
Scheme 1. Reaction mechanism of formation of CdFe2O4.

2102

V. Vasanthi et al. / Journal of Magnetism and Magnetic Materials 324 (2012) 21002107

using fuelsglycine and urea are termed as CdFe2O4-A and


CdFe2O4-B respectively.

3. Results and discussion


3.1. Structural and morphological properties
XRD patterns of as-prepared and 800 1C calcined CdFe2O4-A
and CdFe2O4-B are shown in Fig. 1. A broad peak at 33.061
observed in the as-prepared CdFe2O4-A (Fig. 1a) indicates the
formation of amorphous nature CdFe2O4 along with the presence
of secondary phases like CdO and Fe2O3. The 800 1C calcined
CdFe2O4-A (Fig. 1b) and CdFe2O4-B (Fig. 1c) show sharp and well
dened peaks. This shows that the secondary phases are decomposed into pure crystalline CdFe2O4 phase. Moreover the obtained
planes in the diffraction pattern conrm the formation of cubic
spinel structure. The crystallite size is calculated using Scherrers
formula [8] and the value is around 45 nm and 43 nm for both the

(620)
(533)

(440)

(422)
(511)

(220)

(311)

Intensity (a.u)

(c)

samples. The lattice constant value can be calculated and reported


to be a 8.809 A and a 8.795 A for CdFe2O4-A and CdFe2O4-B
respectively. The calculated lattice density value of CdFe2O4-A is
5.597 g/cm3 and that of CdFe2O4-B is 5.623 g/cm3. All these calculated values agree well with the standards (PDF No. 22-1063). The
tetrahedral and octahedral ionic radii (rA and rB) and the bond
lengths on tetrahedral (CdO) and octahedral (FeO) are calculated
[15] from the XRD data and presented in Table 1. These values also
agree well with those of bulk CdFe2O4 reported in literature by
Gadkari et al. [16]. From the table, it is evident that the bond length
of octahedral structure is greater than that of tetrahedral structure,
which is a normal behavior of ferrites [16].
The FTIR spectrum is recorded at room temperature in the
wave number region of 4000  400 cm  1 and the representative
CdFe2O4-A spectrum is shown in Fig. 2. The absorption bands
around 600 cm  1 belong to the stretching vibration of Cd2 in
tetrahedral site and the absorption bands around 400 cm  1
belongs to the stretching vibration of Fe3 in the octahedral site
of spinel structure [17]. The high frequency (n*1) and low
frequency (n*2) absorption bands are at 578.66 cm  1 and
416.64 cm  1, respectively, for CdFe2O4-A. For CdFe2O4-B, the high
frequency band is at 586.38 cm  1 and low frequency band is
observed at 416.64 cm  1. The force constant for tetrahedral and
octahedral sites is determined using the method of Wilson Matrix
formalism [18] and the values are given in Table 1. It shows that
the force constant is inversely proportional to the bond length
[19]. This trend is well acceptable for the force constant and their
respective bond lengths in CdFe2O4-A and CdFe2O4-B materials.
The SEM images and relative particle size histogram of 800 1C
calcined CdFe2O4-A and CdFe2O4-B are shown in Fig. 3. Various
sizes of spherical particles of CdFe2O4-A and CdFe2O4-B are
observed. It is measured that the maximum number of particles
is in the size range of 0.30.4 mm and 0.40.5 mm for CdFe2O4-A
and CdFe2O4-B respectively. This result is contrary to the crystallite size calculated according to Scherrers equation whereas the
bigger particles are aggregates of small crystallites of CdFe2O4.

(b)
60

Transmittance %

50

(a)

40
30
20
10

10

20

30

40
50
2 (degree)

60

70

0
4000

80

Fig. 1. XRD pattern of (a) as-prepared CdFe2O4-A, (b) 800 1C calcined CdFe2O4-A
and (c) 800 1C calcined CdFe2O4-B.

3500

3000

2500
2000
1500
Wavenumber (cm-1)

1000

500

Fig. 2. FTIR spectrum of CdFe2O4-A.

Table 1
XRD and FTIR parameters.
Sample

Grain
size D
(nm)

Lattice
constant

a (A)

Cell
volume
3
V (A)

Dx
(g/cm3)

ra

rb

CdO

FeO

Tetrahedral
site v1*

Octahedral
site m 2*

kt  102
(N/m)

ko  102
(N/m)

Avg.
k  102

CdFe2O4 -A
CdFe2O4-B

45
43

8.809
8.795

683.680
680.470

5.597
5.623

0.77
0.76

0.72
0.72

2.07
2.07

2.12
2.12

578.66
586.38

416.64
416.64

2.80
2.88

1.02
1.02

1.91
1.95

V. Vasanthi et al. / Journal of Magnetism and Magnetic Materials 324 (2012) 21002107

2103

1400

60

1200
1000

50

600

40

(b)

400
Intensity

Number of Particles

800

30
20

200
(a)

0
-200
-400

10

-600

-1000

-800

0.2 0.4 0.6 0.8 1.0 1.2 1.4


Particle size (m)

-1200
-1400
0

200

400
600
Field (mT)

800

1000

Fig. 4. ESR spectrum of (a) CdFe2O4-A and (b) CdFe2O4-B.

50

1.2

40

0.8
Magnetization (M), emu/g

Number of Particles

(b)

30

20

10

(a)
0.4

0.0

-0.4

-0.8

0
0.2 0.4 0.6 0.8 1.0 1.2 1.4
Particle size (m)

-1.2
-20000

-10000
0
10000
Applied Field (H), Oe

20000

Fig. 5. Magnetic hysteresis loop of (a) CdFe2O4-A and (b) CdFe2O4-B.


Fig. 3. SEM images of (a) CdFe2O4-A and (b) CdFe2O4-B; (inset) corresponding
particle size histogram.

This agglomeration may be due to the magnetic dipole interaction


between ferrite particles. [20].
3.2. Magnetic properties
The magnetic properties of CdFe2O4-A and CdFe2O4-B are
studied using ESR and VSM methods. Fig. 4 shows the ESR
spectrum of both the samples. It has been observed that the
g-value for CdFe2O4-A and CdFe2O4-B is 1.99242 and 2.00746,
respectively. Both the observed values are closely equal to the
free electron g-value of 2.0023. Further the magnetization behaviors of CdFe2O4-A and CdFe2O4-B are shown in Fig. 5. The
magnetic eld (H) versus magnetization (M) curve at room
temperature is recorded with the eld strength of  20 kOe to
20 kOe. It can be seen that there is no saturation magnetization
observed up to the eld strength of 20 kOe. Rucha et al. [21]
reported that this non-saturating behavior indicates the presence
of two components: (i) ferro/ferrimagnetic and (ii) super-

paramagnetic or paramagnetic. It is also reported that the bulk


CdFe2O4 is considered to have a normal spinel structure, where
Cd2 ions occupy the tetrahedral sites and are categorized as
antiferromagnets [21]. In the present case, the prepared CdFe2O4
also exhibits antiferromagnetic behavior due to the larger particle
size of 0.30.5 mm. Further, the zero magnetization values reveal
the zero magnetic moment and this is the characteristic property
of antiferromagnets. The measured coercivity and the magnetization at 20 kOe for CdFe2O4-A are 255 Oe and 0.39 emug  1 and for
CdFe2O4-B, 299 Oe and 0.99 emug  1 respectively. These results
substantiate the antiferromagnetic property of CdFe2O4.
3.3. Electrical and dielectric properties
The complex impedance spectroscopy technique is used to
study the electrical property of the materials, mainly polycrystalline solids, by correlating the sample electrical behavior to its
microstructure. Fig. 6a and b shows the complex impedance plots
(ColeCole plot) of CdFe2O4-A and CdFe2O4-B for the temperature
range of 240360 1C. From the ColeCole plot a single semicircle

2104

V. Vasanthi et al. / Journal of Magnetism and Magnetic Materials 324 (2012) 21002107

250 C
o
270 C
o
290 C

240 C
o
260 C
o
280 C

3000
2500

250 C
o
270 C
o
290 C

Z Sin ,x10

Z Sin , x10

150

3500

240 C
o
260 C
o
280 C

200

100

50

2000
1500
1000
500

0
0

25

50

75

100

125

150

175

200

225

50
o

310 C
o
330 C
o
350 C

1500

2000

2500

3000

3500

20
10

300 C
o
320 C
o
340 C

400

Z Sin ,x10

Z Sin , x10

30

1000

300 C
o
320 C
o
340 C
o
360 C

40

500

500

310 C
o
330 C
o
350 C

300
200
100

0
0

10

20

30

40

50

100

Z Cos , x10

200

300

Z COS ,x10

400

500

Fig. 6. Complex impedance plots for (a) CdFe2O4-A and (b) CdFe2O4-B at various temperatures.
Table 2
Impedance and conductance parameters of CdFe2O4-A.
T (1C)

240
250
260
270
280
290
300
310
320
330
340
350
360

Cb

(pF)

38.960
53.060
62.145
41.000
45.080
61.520
59.570
59.970
67.630
67.840
76.700
74.890
77.006

Rb (O)  104

20.430
15.000
12.811
9.708
7.064
5.176
4.454
3.793
3.363
2.934
2.595
2.362
2.298

ac conductivity  10  6 (S cm  1)
ColeCole plot

Conductance

0.635
0.865
1.013
1.330
1.838
2.509
2.916
3.424
3.862
4.428
5.005
5.498
5.653

0.650
0.881
1.020
1.320
1.770
2.450
2.900
3.389
3.838
4.389
4.840
5.382
5.680

is observed for all the temperatures. Appearance of a semicircular


arc is a characteristic of impedance spectrum. The obtained single
semicircle for both CdFe2O4 materials shows that the samples are
pure and indicates a complete absence of grain boundary effects
and conduction through bulk of the material [22]. The radius of
semicircle decreases with increasing temperature such that its
center lies below the real axis. The intercepts of the semicircular
arcs with the real axis (Z0 ) give us an estimate of the bulk
resistance (Rb) of the material. Moreover the bulk resistance
values decrease with increasing temperature which shows a
typical semiconducting nature of the material i.e., negative
temperature coefcient of resistance behavior. In the semicircle
at a particular frequency (nmax) the real and imaginary parts of the
impedance are maximum. This is due to the parallel combination
of bulk resistance and capacitance. The value of Cb is determined
using the relation 2pvmaxRC 1. For both the samples the
observed capacitance values are in the order of pF. These results
also substantiate the occurrence of conduction mechanism
through the bulk of the materials. The ionic conductivity is

op (Hz)  104

N (S cm  1 kHz  1)  10  8

m (cm2/Vs)  1020

2.13
3.12
3.71
4.73
6.93
7.51
7.62
9.93
10.40
11.21
11.94
12.88
16.34

1.56
1.47
1.46
1.51
1.41
1.80
2.10
1.98
2.18
2.35
2.40
2.64
2.20

2.59
3.74
4.36
5.46
7.84
8.50
8.59
10.60
10.90
11.60
12.20
12.70
16.10

calculated from the bulk resistance using the following equation:

l
S cm1
ARb

where A is the area of the sample; l is the thickness of the sample.


The calculated values of conductivity and bulk capacitance for
both the samples are given in Tables 2 and 3.
Fig. 7 presents the real part of the impedance (Z0 ) as a function
of frequency at different temperatures for CdFe2O4-A. It shows
that there is a gradual decrease in the value of Z0 with rise in
temperature and the curves almost merge in the high frequency
region for all temperatures. This is due to the mobility of space
charge as a result of lowering the barrier properties of the
materials at higher temperature. This shows that the impedance
value is higher in the low frequency region and gradually
decreases with increase in temperature. This leads to the increase
in ac conductivity [23].
Fig. 8 presents the imaginary part of the impedance Zv as a
function of frequency at different temperatures for CdFe2O4-A.

V. Vasanthi et al. / Journal of Magnetism and Magnetic Materials 324 (2012) 21002107

2105

Table 3
Impedance and conductance parameters of CdFe2O4-B.
T (1C)

240
250
260
270
280
290
300
310
320
330
340
350
360

Cb

(pF)

26.060
21.310
27.520
23.590
42.450
37.747
44.057
53.028
31.760
45.660
34.530
46.440
45.023

ac conductivity  10  7 (S cm  1)

Rb (O)  104

30.540
24.900
19.287
13.490
7.501
6.026
4.517
3.336
2.506
1.743
1.152
0.857
0.707

ColeCole plot

Conductance

0.417
0.511
0.660
0.943
1.690
2.110
2.819
3.817
5.080
7.306
11.050
14.860
18.000

0.470
0.510
0.661
0.943
1.720
2.100
2.880
3.900
5.190
7.490
11.100
15.000
17.900

op (Hz)  104

N (S cm  1 kHz  1)  10  8

m (cm2/Vs)  1020

1.50
2.49
3.73
4.53
6.45
6.98
9.04
12.55
17.69
23.14
32.70
42.70
51.60

1.39
1.07
0.94
1.12
1.47
1.69
1.82
1.81
1.73
1.95
2.00
2.18
2.19

1.83
2.97
4.38
5.26
7.31
7.76
9.89
13.39
18.70
24.00
34.68
43.00
51.00

2.0
250C

260C

270C

280C

290C

300C

310C

320C

330C

340C

350C

360C

1.0

240C
260C
280C
300C
320C
340C

1E-4

ac, S.cm-1

Z' x 105 (Ohms)

1.5

240C

1E-5

250C
270C
290C
310C
330C

1E-6
0.5

0.0
0.1

1E-7
1

10
x

104

100

(Hz)

Fig. 7. Plot of real part of impedance with frequency of CdFe2O4-A at different


temperatures.

8
7

Z" x 104 (Ohms)

6
5

240C

250C

260C

270C

280C

290C

300C

310C

320C

330C

340C

350C
360C

4
3
2
1
0
0.1

101

10

100

x104 (Hz)
Fig. 8. Plot of imaginary part of impedance with frequency of CdFe2O4-A at
different temperatures.

102

103
104
Frequency (Hz)

105

106

Fig. 9. Plot of ac conductivity with frequency of CdFe2O4-A.

Appearance of peaks at a particular frequency is a characteristic of


this spectrum. It has been observed that there is a decrease in
height of the peak and peak shift towards higher frequency with
increasing temperature. Moreover the spectra merge at higher
frequencies irrespective of the temperatures. These results reveal
the presence of electrical relaxation in the material [23].
The representative plots of ac conductivity with frequency
of CdFe2O4-A are shown in Fig. 9. The conductivity observed at
low frequency is frequency independent and corresponds
to dc conductivity. This is due to the high resistance effect of
the grain boundary. In the high frequency region the increase
in conductivity with increasing frequency is because the
magnitude of electronic exchange depends upon the concentration of Fe3 /Fe2 ion pairs present on B-site and it corresponds to
ac conductivity [24]. A similar behavior was observed for
CdFe2O4-B (not presented here). The transition from frequency
independent region to frequency dependent is called as the
onset of conductivity relaxation [25]. This shifts towards
the high frequencies as temperature increases. Furthermore the
conductivity in the high frequency region also increases with
temperature. The obtained conductance spectra obey Jonschers
power law:

so sdc Aon

V. Vasanthi et al. / Journal of Magnetism and Magnetic Materials 324 (2012) 21002107

A nonlinear tting was carried out on the basis of Jonschers


power law for all the conductivity curves. From the non-linear
tting value, the dc conductivity (sdc), carrier concentration (N),
hopping frequency (op), and mobility (m) are calculated and are
given in Tables 2 and 3.
From the tables it is seen that the conductivity increases with
respect to temperature. The calculated frequency exponent n has
a value in the range of 1.411.47 and 1.411.51 for CdFe2O4-A and
CdFe2O4-B respectively. According to Jonschers power law, sdc is
directly proportional to on. The value of i plays an important role
in the hopping mechanism; if n r1, the hopping of the ion
involves a translation motion with a sudden hopping whereas if
n Z1, the hopping is localized [25]. It is observed that with
increase in temperature the carrier mobility and hopping frequency increase. Comparatively, the conductivity of CdFe2O4-A is
high when compared with CdFe2O4-B. This may be due to the high
yield of carbon as the byproduct in addition with single phase
CdFe2O4, when using glycine as the fuel. The activation energy Ea

is calculated using the Arrhenius equation: sdc s0 exp Ea =kT .
A graph is plotted for 1000/T versus logarithmic value of conductivity. From the slope obtained for linear tting, the activation
energy is calculated. The activation energy for CdFe2O4-A is
0.4556 eV and for CdFe2O4-B is 0.8010 eV. This also substantiates
the high conducting behavior of CdFe2O4-A due to the low
activation energy. This reveals that the ionic conductance in the
material is due to the hopping of ions between allowed sites [26].
The dielectric analysis is often used quantitatively to study the
response of a material to a eld induced perturbation such as
orientation polarization of dipoles and motion of ions. The
relation between the dielectric constant (e0 ) and frequency at
various temperatures for CdFe2O4-A is shown in Fig. 10. This
curve resembles the dynamic response of the constituents of the
solid [19]. The complex permeability is dened as

eer e0 je00 e0 j

s
oe0

where e0 is real the part of permeability and e00 is the imaginary


part. From Fig. 10 it is observed that the value of real part of the
dielectric constant e0 is high at low frequencies and it decreases
with increase in frequency. The reason for the observation of high
e0 value at low frequency is due to the space charge polarization
produced at grain boundary. The polarization mechanism

2500

240C
260C
280C
300C
320C
340C
360C

2000

'

1500

250C
270C
290C
310C
330C
350C

1000

106

240C
260C
280C
300C
320C
340C
360C

105

104

''

2106

250C
270C
290C
310C
330C
350C

103

102

101

100
101

102

103
104
Frequency (Hz)

105

106

Fig. 11. Relation between the dielectric loss (e00 ) and frequency at different
temperature for CdFe2O4-A.

involves the exchange of electrons between the ions of the same


element, which have different valance states and are distributed
randomly over the crystallographic equivalent sites. Here the
exchange of electrons takes place between Fe3 Fe2 ions present at octahedral sites (B-site). During this exchange, the
electrons have to pass through the grain boundary; the electrons
accumulate at the grain boundary and produce space charge
polarization [24]. When the frequency is increased, the dipoles
cannot rotate their direction rapidly. Thus the orientation of the
dipoles decreases and disappears due to the inertia of the ions.
Since the orientation of the dipole increases with increase in
temperature, the dielectric constant e0 also increases. Similarly
the dielectric constant e0 increases on increasing the temperature
at a low frequency and decreases when the frequency increases.
This is due to the orientation of the dipoles which causes the
permittivity to increase at high temperature [25].
Fig. 11 presents the relation between the dielectric loss (e00 )
and frequency at different temperatures for CdFe2O4-A. From the
plot it is seen that the log e00 value is maximum at low frequency
and decreases with increase in frequency. This decrease in e00 with
increasing frequency agrees well with the Debye type relaxation
process [27]. It is an established fact that the log e00 value
increases with increase in temperature at low frequency and
decreases with increase in frequency. Similarly CdFe2O4-B also
reveals it. The occurrence of loss peaks can be discussed by
correlating the conduction mechanism and the dielectric behavior
for the spinel ferrites [28]. The dielectric loss in ferrites is
generally reected in the conductivity measurements i.e., the
material that possesses high conductivity exhibits high dielectric
loss and vice versa. Thus the CdFe2O4-A and CdFe2O4-B samples
revealed dielectric dispersion according to MaxwellWagner
interfacial polarization [29,30]. This is in good agreement with
the Koops phenomenological theory.

500
4. Conclusion

0
101

102

103
104
Frequency (Hz)

105

106

Fig. 10. Relation between the dielectric constant (e0 ) and frequency at various
temperature for CdFe2O4-A.

CdFe2O4 particles were successfully synthesized by microwave


assisted combustion synthesis within a minute. XRD results
conrmed that the material possesses a cubic spinel structure
with high purity, as also substantiated by infrared spectroscopy.
Analysis of the SEM images conrms the spherical shape of the

V. Vasanthi et al. / Journal of Magnetism and Magnetic Materials 324 (2012) 21002107

particles. The magnetization measurements of ESR and VSM


techniques reveal the antiferromagnetic nature of the heat treated
samples. Further, high coercivity (255 Oe) and small magnetization (0.39 emu/g) values also substantiate the antiferromagnetic
nature of the particles. The conductivity measurement implies the
semiconducting nature of materials. The conductivity for CdFe2O4A is 5.68  10  6 S cm  1 and for CdFe2O4-B is 1.76  10  6 S cm  1
at 360 1C.
References
[1] D.S. Mathew, Ruey-Shin Juang, Chemical Engineering Journal 129 (2007) 51.
[2] P.K. Nayak, Materials Chemistry and Physics 112 (2008) 24.
[3] M. Yokoyama, E. Ohta, T. Sato, T. Sato, Journal of Magnetism and Magnetic
Materials 183 (1998) 173.
[4] Chu Xiangfeng, Zheng Chemou, Sensors and Actuators B 96 (2003) 504.
[5] C.N. Chinnasamy, A. Narayanasamy, N. Ponpandian, R.Justin Joseyphus,
Scripta Materialia 44 (2001) 1411.
[6] M.H. Mahmoud, A.M. Abdallas, H.H. Hamdeh, W.M. Hikal, S.M. Taher, J.C. Ho,
Journal of Magnetism and Magnetic Materials 263 (2003) 269.
[7] Xiangdong Lou, Shuping Liu, Dongyang Shi, Wenfei Chu, Materials Chemistry
and Physics 105 (2007) 67.
[8] Fengxiu Miao, Zanhong Deng, Xianshnu Lv, Guixin Gu, Sogming Wan,
Xiaodong Fang, Qingli Zhang, Shaotang Yin, Solid State Communications
150 (2010) 2036.
[9] Idalia Bilecka, Markus Niederberger, Nanoscale 2 (2010) 1358.
[10] F. Bensebaa, F. Zavaliche, P. LEcuyer, R.W. Cochrane, T. Veres, Journal of
Colloid and Interface Science 277 (2004) 104.

2107

[11] Rekha Dom, R. Subasri, K. Radha, Pramod H. Borse, Solid State Communications 151 (2011) 470.
[12] M. Sertkol, Y. Koseoglu, A. Baykal, H. Kavas, A. Bozkurt, M.S. Toprak, Journal of
Alloys and Compounds 486 (2009) 325.
[13] Y. Koseoglu, A. Baykal, F. Gozuak, H. Kavasa, Polyhedron 28 (2009) 2887.
[14] M. Sertkol, Y. Koseoglu, A. Baykal, H. Kavas, M.S. Toprak, Journal of Magnetism and Magnetic Materials 322 (2010) 866.
[15] K.J. Standley, Oxide Magnetic Materials, Clarendon, Oxford, UK, 1972.
[16] A.B. Gadkari, T.J. Shinde, P.N. Vasambekar, Materials Chemistry and Physics
114 (2009) 505.
[17] R. Kalai Selvan, C.O. Augustin, L. John Berchmans, R. Saraswathi, Materials
Research Bulletin 38 (2003) 41.
[18] R.D. Waldron, Physical Review 99 (6) (1955) 1727.
[19] J. Smit, H.P.J. Wijn, Ferrites, Cleaver-Hume Press, London, 1959.
[20] Yuping Wang, Liangchao Li, Jing Jiang, Hui liu, Haizhen Qiu, Feng Xu, Reactive
and Functional Polymers 68 (2008) 1587.
[21] Rucha Desai, R.V. Mehta, R.V. Upadhyay, Amita Gupta, A. Praneet, K.V. Rao,
Bulletin of Materials Science 30 (2007) 197.
[22] Archana Shukla, R.N.P. Choudhary, A.K. Thakur, D.K. Pradhan, Physica B 405
(2010) 99.
[23] Ashok Kumar, B.P. Singh, R.N.P. Choudhary, Awalendra K. Thakur, Journal of
Alloys and Compounds 394 (2005) 292.
[24] N. Singh, A. Agarwal, S. Sanghi, Current Applied Physics 11 (2011) 783.
[25] V.D. Nithya, R. Kalai Selvan, Physica B 406 (2011) 24.
[26] M. Prabu, S. Selvasekarapandian, A.R. Kulkarni, S. Karthikeyan, G. Hirankumar,
C. Sanjeeviraja, Journal of Rare Earths 28 (2010) 435.
[27] P. Debye, Polar Molecules, Chemical Catalogue Company, New York, 1929.
[28] K. Iwauchi, Journal of Applied Physics 10 (1971) 1520.
[29] J.C. Maxwell, Electricity and Magnetism, vol. 1, Oxford University Press,
section 1873, Section 328.
[30] K.W. Wagner, Annals of Physics 40 (1913) 817.

Journal of Magnetism and Magnetic Materials 324 (2012) 37413747

Contents lists available at SciVerse ScienceDirect

Journal of Magnetism and Magnetic Materials


journal homepage: www.elsevier.com/locate/jmmm

Magnetic and structural studies of the Mn-doped MgZn ferrite


nanoparticles synthesized by the glycine nitrate process
H. Mohseni a, H. Shokrollahi b,n, Ibrahim Shari b, Kh. Gheisari a
a
b

Department of Materials Science and Engineering, Faculty of Engineering, Shahid Chamran University, Ahvaz, Iran
Electroceramics Group, Materials Science and Engineering Department, Shiraz University of Technology, 71555-313, Shiraz, Iran

a r t i c l e i n f o

abstract

Article history:
Received 6 December 2011
Received in revised form
20 May 2012
Available online 28 June 2012

In this study, Nanocrystalline MnMgZn ferrite with the chemical formula MnxMg0.5  xZn0.5Fe2O4
(x 0, 0.1, 0.2, 0.3, 0.4, 0.5) was successfully synthesized by the glycine-nitrate autocombustion process
using glycine as a fuel and nitrates as oxidants. The as-synthesized powders were characterized by the
X-ray diffraction analysis, eld emission scanning electron microscopy, Fourier transform infrared
spectroscopy (FTIR) and vibrating sample magnetometer. The X-ray diffraction data was used to
determine the lattice constant, cation distribution and the oxygen position parameter. The results
reveal that the nanocrystalline MnMgZn ferrite has an average crystallite size of 3567 nm and
particle size of 40 nm. The lattice parameter increases linearly with an increase in the Mn content. The
FTIR analysis conrms the intrinsic vibrational frequencies of the tetrahedral and octahedral of the
spinel structure. The magnetic measurements indicate that the coercivity decreases, and the magnetization increases by increasing the Mn content.
& 2012 Elsevier B.V. All rights reserved.

Keywords:
Ferrite
Magnetic properties
X-ray diffraction
Cation distribution

1. Introduction
Ferrites are the magnetic ceramics commonly used in the
production of electronic components [14]. Among different
types of ferrites, MgZn ferrites represent an important class of
soft-magnetic materials, which are widely used in computer
memory and logic devices, cores of transformers, recording heads,
antenna rods, loading coils and microwave devices, and so forth.
As compared to other ferrites, they are preferable because of their
high resistivity and low eddy current losses [5].
Recently, nanoparticle ferrites with a high surface to volume
ratio have received much attention due to their useful, electrical
and magnetic properties used in magnetic uid, information
storage and medical diagnostics [6,7]. These types of nanoparticles can be produced by different wet chemical methods, such as
the co-precipitation [8], hydrothermal synthesis [9], micro-emulsion synthesis [7] and sol gel method [10]. In addition, various dry
methods, including grinding [11], mechanical alloying [12] and
thermal plasma methods [13] have been employed. Besides these
methods, several attempts have been made using auto-ignited
combustion reactions [14,15]. Among them, the glycinenitrate
process (GNP) is a simple, inexpensive and self-sustaining combustion synthesis technique, containing metal nitrates as oxidizers and glycine as fuel.

Corresponding author.
E-mail address: shokrollahi@sutech.ac.ir (H. Shokrollahi).

0304-8853/$ - see front matter & 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.jmmm.2012.06.009

In this method, the aqueous solution begins boiling, afterward


ignites, and an exothermal, self-sustaining and very rapid chemical reaction occurs [16]. The liberated heat can be utilized to
provide the energy for phase transformation in synthesizing the
desired products. In the combustion reaction the reaction temperature will not rise considerably. The reason can be attributed
to the gas release and convection of heat loss. As a result, ultrane powders could be achieved to prevent both the over-growth
of crystallites and the agglomeration of particles [16,17].
In addition, GNP has many other advantages, such as the ne
particle size, high energy efciency, fast heating rates, high
compositional homogeneity, low-cost and fast method for preparing well-crystallized double oxides in about 30 min from
preparing the precursor solution to obtain the nal products
[16,18]. The above-mentioned advantages enable GNP to be a
promising candidate method for preparing nanostructure oxides
[19]. In this area, several researches have investigated the
magnetic properties of Ni0.5Zn0.5-ferrite by solgel and GNP
methods. The results revealed that the saturation magnetizations
for solgel and GNP methods are 14 and 22 emu/g, respectively
[20,21]. The aim of this work is to study the structure and
magnetic properties of MgZn ferrite doped by Mn2 ions. The
replacement of nonmagnetic ion of Mg2 by magnetic ion of
Mn2 (L0, S5/2, J5/2) affects the magnetic properties. The
angular momentum for Mn2 is zero (L0) which can reduce the
magnetic coercivity. Mn2 cation in spinel is at its high spin state
(S 5/2). Its ve d electron congurations are e2t32 and t32ge2g at the
tetrahedral and octahedral lattice site, respectively which can

3742

H. Mohseni et al. / Journal of Magnetism and Magnetic Materials 324 (2012) 37413747

affect the saturation magnetization. It would be worth mentioning that as compared with the literature in this eld [2224], this
paper has paid special attention to the synthesis of MnMgZn
ferrite by the GNP method.

The Fourier Transform Infrared Spectroscopy (FTIR) spectrum was


recorded as (KBr) discs in the range 4004000 cm  1 using the
(FTIR-Shimadzu-8000) spectrophotometer.

3. Results and discussion


2. Experimental method

3.1. FESEM

Nanoparticles of Mnx Mg0.5  x Zn0.5 Fe2O4 (where x 0, 0.1, 0.2,


0.3, 0.4, 0.5 and named P1P6, respectively) were synthesized by
the auto-combustion synthesis method using glycine as a fuel
and nitrates as oxidants. In an appropriate ratio, reagent grade
Fe(NO3)39H2O, Zn(NO3)24H2O, Mg(NO3)26H2O, Mn(NO3)2
4H2O and glycine (H2NCH2COOH) were used as starting materials.
The metal nitrates dissolved in distilled water to obtain the
precursor solution. A specied amount of glycine was then added
into the nitrate solution at a molar ratio of 2:1 for fuel: oxidant
(stoichiometric combustion). The mixed solution was put into a
large beaker and heated on a hot plate to evaporate excess water
until an auto-ignited and self-sustaining combustion occurred.
The self-sustained combustion resulted in the ne MnMgZn
ferrite ash of brown in color.
X-ray diffraction (XRD) patterns of all samples were taken at
room temperature by an X-ray diffractometer (Philips Xpert
MPD), using a Co Ka radiation (l 1.789010 A1) at a voltage of
40 kV and a current of 30 mA. The XRD patterns of samples were
recorded in the range 2y 10901 using a step size of 0.021 and a
counting time of 2 s per step. The size and morphology of the
particles were determined by eld emission scanning electron
microscope (FESEM, Hitachi, Japan S4160). The lattice parameters,
the oxygen position and the cation distribution were determined
by means of Rietveld renement [25], using the reex program
[26]. For the calculation of the crystallite size, Scherrers method
was applied [27]. Afterward the crystallite size was rened by
Rietveld renement. FTIR study was used to indicate the vibrational modes in the samples. The magnetic measurements of the
prepared powder were determined at room temperature using
the vibrating sample magnetometer (VSM) (Dexing, Model 250).

FESEM produces clearer, less electro-statically distorted images


than conventional SEM. The shape, size and morphology of the
samples were investigated by the eld emission scanning electron
microscope (FESEM) (Fig. 1). FESEM images reveal that the particles
are approximately spherical in shape with the diameter between
35 and 45 nm and tend to agglomerate.
3.2. XRD analysis and cation distribution
The XRD patterns of the as-synthesized Mnx Mg0.5  x Zn0.5
Fe2O4 (where x0, 0.1, 0.2, 0.3, 0.4, 0.5) shown in Fig. 2 conrm
the formation of the single-phase spinel structure. To calculate
the average crystallite size from the broadening of the XRD peaks
of samples, the Rietveld renement of the XRD pattern was
applied. The goodness of t was tested by means of values of
Rwp, whose values were below 10%, indicating satisfactory t [28],
as shown in Fig. 3 for the P2 sample. Table 1 lists the rened
crystallographic variables including crystallite size (D), chemical
composition, lattice parameter (a), site occupancy, site coordination and oxygen positional parameter (3/8, 3/8, 3/8) (u).
The structure was rened in the space group Fd3m with
tetrahedral A cations in the spatial 8a position (1/8, 1/8, 1/8);
octahedral B cations, in the spatial 16d position (1/2, 1/2, 1/2);
and oxygen ions, in the spatial 32e position (u, u, u). The
positional parameter of oxygen is conventionally called u-parameter. In spinels, the oxygen anions are often displaced along a
[1 1 1] direction and thus u-parameter varies between 0.24 and
0.275 (origin at the center of symmetry), over which the octahedral site has site symmetry 3m, except for u0.25 when it has

Fig. 1. FESEM micrograph of (a) p1 (b) p4 (c) p4 and (d) p6.

H. Mohseni et al. / Journal of Magnetism and Magnetic Materials 324 (2012) 37413747

cubic site symmetry m3m [29]. The renement started with an


ideal value of 0.25 for u. The shapes of the Bragg peaks were
quantied by a pseudo-Voigt function. Each structure model was
rened to convergence and the best result was selected on the
basis of the various agreement factors of renement. The distribution of the cations on two sub-lattices can be expressed as:
h

Mg2a Mn2b Zn2g Fe31abg

iA h

2
2
Mg20:5x
a Mnbx Zn0:5g Fe1 a b g

iB

3743

where the square practices contain the ions on the octahedral


sites. In this formula, a, g and b are constants rened by the
Rietveld method and determine the cation distribution.
It is evident from Fig. 4 that with increasing substitution
different cations have shown different nature of variation. The
occupancy of Mn increases on both tetrahedral and octahedral
sites with increasing its concentration. The population of other
ions on the tetrahedral and octahedral sites shows different
behaviour.

O4

3.2.1. Particle size


The crystal growth in the solution depends on various parameters including extrinsic (molecular concentration of the material, pH and temperature) and intrinsic parameters (heat of
formation, site preferences, electronic conguration). Both the
above parameters can affect the growth process of the crystal. In
this study the results show that with increasing the Mn content
the particle size increase. This effect can be explained by the
following points.
(i) Site preferences: It may happen that the grain growth is
obstructed when the cationic preferences are not fully
satised [30]. The dependence of the particle size on Mn
concentration may also be related to the site preferences of
Mn, Zn, Mg and Fe in ferrite structure. It is interesting to note
that Zn2 ions in the spinel structure have a very strong
preference for tetrahedral sites, Mg2 ions have a preference
for octahedral site and Mn2 ions have a preference for both
octahedral and tetrahedral sites. Mn2 is uniformly distributed amongst the different sites and therefore has a higher
probability of being absorbed by a nucleus. Zn2 forces Fe3
to occupy octahedral site and it leads to the cationic preferences are not fully satised. Similarly, the restriction of
Fe3 to occupy only the octahedral sites as in the case of
ZnFe2O4 is no longer applied for MnFe2O4 [30]. This accounts
for higher particle sizes in Mn-doped ferrites, as compared to
Zn-doped ferrites.

Fig. 2. XRD patterns of samples.

Fig. 3. Rietveld renement of X-ray powder diffraction pattern for p2 sample.

Table 1
Lists crystallite size (D), chemical composition, lattice parameter (a), site occupancy, site coordinations and oxygen positional parameter (3/8, 3/8, 3/8) (u).
Sample

P1
P2
P3
P4
P5
P6

Chemical composition

Mg0.5 Zn0.5 Fe2 O4


Mg0.4 Mn0.1 Zn0.5 Fe2
Mg0.3 Mn0.2 Zn0.5 Fe2
Mg0.2 Mn0.3 Zn0.5 Fe2
Mg0.1 Mn0.4 Zn0.5 Fe2
Mn0.5 Zn0.5 Fe2 O4

O4
O4
O4
O4

D (nm)

35.5
42.1
48.1
54.9
67.8
65.1

a (nm)

8.388
8.403
8.423
8.437
8.449
8.465

Site occupancy

Atomic coordinates (x yz)

A-site

B-site

A-site

B-site

0.814
0.824
0.795
0.812
0.833
0.822

0.745
0.728
0.717
0.710
0.727
0.723

1.000
1.000
1.000
1.000
1.000
1.000

0.0000
0.0000
0.0000
0.0000
0.0000
0.0000

0.6250
0.6250
0.6250
0.6250
0.6250
0.6250

0.3832
0.3833
0.3795
0.3812
0.3818
0.3827

Rwp (%)

Rp (%)

8.60
8.14
7.05
7.09
6.53
7.36

5.91
5.80
5.22
5.10
4.82
5.51

3744

H. Mohseni et al. / Journal of Magnetism and Magnetic Materials 324 (2012) 37413747

Fig. 4. Variation of cation occupation factors with Mn concentration.

(ii) Enthalpy of formation: The presence of Mn promotes the


crystal growth. The entropy of formation will reect
any congurational disorder present in the spinel, and at
high temperature a considerable entropy stabilization may
result from this disorder. The free energies of formation will
be comparable to the enthalpy of formation for normal
spinels and somewhat more negative for spinels with intermediate or inverse cation distributions [31]. The surface
temperature affects the molecular concentration as an
important factor in growing the tiny crystal at the surface
of the crystal and hence the crystal growth. The formation of
Mn-ferrite is more endothermic, as compared to the formation of other ferrites like Zn-ferrite. Thus, it is expected that
should Mn be introduced in the system, less heat will be
liberated, increasing the molecular concentration at the
crystal surface and hence increasing the grain growth [31].
(iii) Electronic conguration and bonding energy: The increase
in the particle size by the increase in Mn content may be
explained by the electronic conguration of Mn2 (3d5), and
its more tendency to interact with ligands and oxygen
anions, as compared to Mg2 2S2 2P6, which has a complete
electronic conguration (noble gas). The lack of d electrons is
important because there are very little covalent interaction
and tendency toward extension between Mn2 and its
ligand. Furthermore, it is reported by several researchers
[32] that the smaller particle sizes of the samples doped
with Zn ions are due to the lower bond energy of Zn2
O2  (159 kJ/mol) as compared with that of Co2 O2 
(384 kJ/mol). In this line, the larger particle sizes of the
samples doped with Mn ions are due to the higher bond
energy of Mn2 O2  (946 kJ/mol) as compared with that of
Zn2 O2  (159 kJ/mol) [33,34].

3.2.2. Lattice constant


As can be seen in Table 1 the lattice parameter increases from
8.388 to 8.465 nm with Mn2 concentration.
The lattice constant was
p
2
2
2
calculated using the relation a h l k [35]. The d spacing
values were calculated for the recorded peaks using Braggs law, and
the lattice constant (a) was calculated for the highest peak plane.
From this table, it is clear that the experimental lattice constant and
the crystallite size increase by increasing the Mn concentration. The
Mn2 ion has a larger radius than most of the ions of the 3dn series,
0.080 nm instead of 0.064 nm (Mn2 0.080 nm, Mg2 0.072 nm,
Zn2 0.074 nm, Fe3 0.064 nm in six-fold coordination). The
noticeable increase in a up to XMn 0.3 relates to the higher tendency

of Mn2 to go to the A site. To accommodate this larger cation, the


lattice parameter is increased and hence some of the factors of the
covalent bonding are reduced, thereby increasing the NBB coefcient
(ferromagnetic double exchange).
3.3. FTIR analysis
Fig. 5 gives the IR absorption spectra of the investigated ferrite
(x 00.5). From this gure, it is found that the spectra consist of
two signicant absorption bands, rst at about 600 cm  1 and
second at about 425 cm  1. Absorption bands observed within this
limit reveal the formation of the single phase spinel structure
having two sub-lattices, tetrahedral (A) site and octahedral
(B) site [36]. The absorption band, v1, observed at about
600 cm  1 is attributed to the tetrahedral site, whereas that of
v2 observed at about 425 cm  1 is assigned to the octahedral
group complexes. Furthermore, the FTIR spectra of the samples
show bands at about 3415 and 1600 cm  1, which are ascribed to
the stretching modes and HOH bending vibration of the free or
absorbed water. This suggests that hydroxyl groups are retained
in the ferrites when they are prepared following the low temperature soft chemistry routes [37]. The bands at about 1384 and
570 cm  1 are attributed to the stretching vibration of the antisymmetric NO3 1 and tetrahedral complexes of ferrite [38].
3.4. Magnetic properties
In spinel oxides, the magnetic moments are mainly from the
parallel uncompensated electron spins of the individual ions. The
magnetization curves of the MnMgZn ferrite nano-particles
measured at 27 1C are shown in Fig. 6. It is clear that the chemical
composition can inuence the magnetization curves as a result of
a change in the particle size and distribution of cations. The
nanoparticles provide excellent opportunities for the fundamental studies on the relationship between the magnetic behavior
and the magnetic couplings at the atomic level.
Table 2 illustrates the dependence of the saturation magnetization on chemical composition. The MS value obtained for the
samples varies between 48.59 and 65.25 emu/g at room temperature as the Mn content increases. This variation can be understood as follows.
(i) Size effect: As the Mn content increases the particle size and
consequently the saturation magnetization increase. The
existence of some degree of the spin canting (lack of full
alignment of the spins in large applied elds) in the whole
volume of the nanoparticles, the disordered surface/dead

H. Mohseni et al. / Journal of Magnetism and Magnetic Materials 324 (2012) 37413747

3745

Fig. 5. FT-IR spectra of powders.

layer and the spin-glass properties at the surface can explain the
decrease of the saturation magnetization as the particle size
decreases [39]. The dead magnetic layer originates from the
demagnetization of the surface spin, causing the surface spins to
be disordered or misaligned. This phenomenon weakens the
total magnetization of the nanoparticles [39]. Basically, two
mechanisms have been suggested to explain the origin of spin
canting: one is the surface (or interface) effect, and the other is
the nite- size effect. Variations in coordination numbers and
distances of surface cations could result in a distribution of net
exchange elds. Also, organic molecules bonded to the particles
act as pinning centers for the spins at the surface.
(ii) Cation distribution: In the present work, by assuming 5 mB for
Fe(III) (L0, S5/2, J5/2), and 5 mB for Mn(II) (L0, S 5/2,
J5/2) [35], the net magnetic moment (m/molecule) 9MAMB9
is theoretically calculated, and the results are presented in Fig. 7.
It is obvious that the magnetization of tetrahedral (MA)
decreases; whereas the magnetization of octahedral (MB)
and the net magnetization increase for the samples with

higher Mn content. This can be attributed to the higher


concentration difference of magnetic ions (Fe3 and Mn3 )
in octahedral and tetrahedral sites by assuming the collinear
ferrimagnetic spin structure. Consequently, the FeFe and
MnMn interactions are increased due to an increase in the
concentration of Fe and Mn ions on the B-sites.

It appears that the substitution of magnetic Mn has strengthened the ferrimagnetic interactions in the system causing an
increase of magnetic order with increasing its concentration (x).
Since, because the occupancy of magnetic Mn increases with x on
both tetrahedral and octahedral sites, the net magnetization
increases relatively signicantly. The synthesis method can affect
strongly the particle size and cation distribution. These parameters change the magnetic properties like saturation magnetization and coercivity. For this reason, it is clear that, a magnetic
powder obtained by GNP method may be have different magnetic
properties as compared with other methods [20,21].

3746

H. Mohseni et al. / Journal of Magnetism and Magnetic Materials 324 (2012) 37413747

Fig. 6. MH curves of powders.

Table 2
Magnetic parameters as a function of Mn content in Mg-Mn-Zn Ferrite
Sample

MS(emu/g)7 0.01

HC(Oe)7 0.01

Mr(emu/g)7 0.01

Hmax(kOe)

P1
P2
P3
P4
P5
P6

48.59
53.98
55.14
58.27
59.74
65.25

67.71
64.89
63.47
60.13
57.35
52.39

7.46
7.9
8.08
8.45
8.5
8.86

13.6
13.5
13.9
13.7
13.7
13.6

Fig. 7. Variation of the sublattice and net magnetic moments at room temperature
with Mn concentration.

The magnetic coercivity of the particles depends signicantly


on their magneto-crystalline anisotropy, micro-strain, inter-particle interaction, temperature, size and shape [40]. Mg2 (2S2 2P6)
and Mn2 (3d5) do not have contribution in angular momentum

(l0), magneto-crystalline anisotropy and/or lattice-spin coupling. For this reason, the reduction in magnetic coercivity from
67.7 to 52.4 Oe Table 2 can be contributed only to the increase in
the particle size (Hc a D  1) and/or the decrease in the saturation
magnetization (Hc a MS 1) [40].

4. Conclusions
In this paper, MnMgZn ferrite powders were synthesized by
the glycinenitrate process. The data obtained from XRD, FESEM,
FTIR and VSM revealed that:
1. The nanocrystalline MnMgZn ferrite shows the absorption
bands near 600 and 400 cm  1, respectively. The high frequency band around 600 cm  1 is attributed to the tetrahedral
complexes and the band near 400 cm  1 corresponds to the
octahedral complexes.
2. To calculate the average crystallite size from the broadening of
the XRD peaks of samples and other micro-structural properties, the Rietveld renement of XRD pattern was applied.
3. As the Mn content increases, the lattice parameter due to its
high ionic radius and incomplete-symmetry electronic
conguration.
4. As the Mn content increases, the crystallite size increases as a
result of site preferences, enthalpy formation, electronic conguration and bond energy.
5. Cation site preferences have also caused a large variation in
the magnetic ion concentration over the two sublattices giving
rise to signicant change in the strength of sublattice magnetic
moments.
6. The saturation magnetization increases due to an increase in
the particle size and an increase in the difference in the
saturation magnetization of two sub-lattices.
7. As the Mn content increases, the magnetic coercivity decreases
as a result of an increase in the saturation magnetization and
the particle size. Furthermore, Mg2 (2S2 2P6) and Mn2 (3d5)
do not have contribution in angular momentum and magnetocrystalline anisotropy.

H. Mohseni et al. / Journal of Magnetism and Magnetic Materials 324 (2012) 37413747

Acknowledgements
The authors wish to thank the Department of Chemistry at
Shiraz University for FTIR measurements.
References
[1] S. Solyman, Transport properties of La-doped MnZn ferrite, Ceramics
International 32 (2006) 755760.
[2] I. Shari, H. Shokrollahi, S. Amiri, Ferrite-based magnetic nanouids used in
hyperthermia applications, Journal of Magnetism and Magnetic Materials 324
(2012) 903915.
[3] H. Shokrollahi, Magnetic properties and densication of ManganeseZinc soft
ferrites (Mn1xZnxFe2O4) doped with low melting point oxides, Journal of
Magnetism and Magnetic Materials 320 (2008) 463474.
[4] H. Shokrollahi, K. Janghorban, Inuence of additives on the magnetic properties, microstructure and densication of MnZn soft ferrites, Materials
Science and Engineering: B 141 (2007) 91107.
[5] R.B. Pujar, S.S. Bellad, S.C. Watawe, B.K. Chougule, Magnetic properties and
microstructure of Zr4 -substituted MgZn ferrites, Materials Chemistry and
Physics 57 (1999) 264267.
[6] R. Arulmurugan, B. Jeyadevan, G. Vaidyanathan, S. Sendhilnathan, Effect of
zinc substitution on CoZn and MnZn ferrite nanoparticles prepared by coprecipitation, Journal of Magnetism and Magnetic Materials 288 (2005)
470477.
[7] D.S. Mathew, R.-S. Juang, An overview of the structure and magnetism of
spinel ferrite nanoparticles and their synthesis in microemulsions, Chemical
Engineering Journal 129 (2007) 5165.
[8] C.F. Zhang, X.C. Zhong, H.Y. Yu, Z.W. Liu, D.C. Zeng, Effects of cobalt doping on
the microstructure and magnetic properties of MnZn ferrites prepared by
the co-precipitation method, Physica B: Condensed Matter 404 (2009)
23272331.
[9] L. Nalbandian, A. Delimitis, V.T. Zaspalis, E.A. Deliyanni, D.N. Bakoyannakis,
E.N. Peleka, Hydrothermally prepared nanocrystalline MnZn ferrites: synthesis and characterization, Microporous and Mesoporous Materials 114 (2008)
465473.
[10] P.P. Hankare, R.P. Patil, U.B. Sankpal, S.D. Jadhav, K.M. Garadkar, S.N. Achary,
Synthesis and morphological study of chromium substituted ZnMn ferrites
nanostructures via solgel method, Journal of Alloys and Compounds 509
(2011) 276280.
[11] B. Gillot, B. Domenichini, Effect of the preparation method and grinding time
of some mixed valency ferrite spinels on their cationic distribution and
thermal stability toward oxygen, Materials Chemistry and Physics 47 (1997)
217224.
[12] S. Dasgupta, K.B. Kim, J. Ellrich, J. Eckert, I. Manna, Mechano-chemical
synthesis and characterization of microstructure and magnetic properties
of nanocrystalline Mn1xZnxFe2O4, Journal of Alloys and Compounds 424
(2006) 1320.
[13] A.B. Nawale, N.S. Kanhe, K.R. Patil, S.V. Bhoraskar, V.L. Mathe, A.K. Das,
Magnetic properties of thermal plasma synthesized nanocrystalline nickel
ferrite (NiFe2O4), Journal of Alloys and Compounds 509 (2011) 44044413.
[14] A.C.F.M. Costa, E. Tortella, M.R. Morelli, R.H.G.A. Kiminami, Synthesis, microstructure and magnetic properties of NiZn ferrites, Journal of Magnetism
and Magnetic Materials 256 (2003) 174182.
[15] C.-C. Hwang, J.-S. Tsai, T.-H. Huang, C.-H. Peng, S.-Y. Chen, Combustion
synthesis of NiZn ferrite powderinuence of oxygen balance value,
Journal of Solid State Chemistry 178 (2005) 382389.
[16] J.-S.T.Chyi-Ching Hwang, Ting-Han Huang, Combustion synthesis of NiZn
ferrite by using glycine and metal nitratesinvestigations of precursor
homogeneity, product reproducibility, and reaction mechanism, Materials
Chemistry and Physics 93 (2005) 330336.
[17] L.A. Chick, L.R. Pederson, G.D. Maupin, J.L. Bates, L.E. Thomas, G.J. Exarhos,
Glycine-nitrate combustion synthesis of oxide ceramic powders, Materials
Letters 10 (1990) 612.

3747

[18] M.T.Nobuyuki Kikukawa, Yoshinobu


Nagano, Masami
Sugasawa,
Satoru Kobayashi, Synthesis and magnetic properties of nanostructured
spinel ferrites using a glycinenitrate process, Journal of Magnetism and
Magnetic Materials 284 (2004) 206214.
[19] J.J. Kingsley, L.R. Pederson, Combustion synthesis of perovskite LnCrO3
powders using ammonium dichromate, Materials Letters 18 (1993) 8996.
[20] D.D.A Chatterjee, S.K Pradhan, D Chakravorty, Synthesis of nanocrystalline
nickle-zinc ferrite by the solgel method, Journal of Magnetism and Magnetic
Materials 127 (1993) 214218.
[21] A.P.P. Priyadharsinia, P. Sambasiva Raob, G. Chandrasekarana, Structural,
spectroscopic and magnetic study of nanocrystalline NiZn ferrites, Materials
Chemistry and Physics 116 (2009) 207213.
[22] F. Sale, J. Fan, C. Y-T, Effects of V 2 O 5 On The microstructure and magnetic
properties of gel-derived MgMnZn AND MnZn ferrites, Ceramic Transactions 47 (1994) 155167.
[23] S. Seyyed Ebrahimi, Z.Pishgahi Fard, An investigation on the optimum
conditions for preparation of pure MnMgZn ferrite powder, Key Engineering Materials 336 (2007) 699702.
[24] K. Ikegami, Y. Masuda, H. Takei, T. Maeda, Low loss of MgMnZn ferrite for
deection yoke, Le Journal de Physique IV 7 (1997) 1-1.
[25] H.M. Rietveld, A prole renement method for nuclear and magnetic
structures, Journal of Applied Crystallography 2 (1969) 6571.
[26] R.A. Young, The Rietveld Method, Oxford University Press, 1995.
[27] M.H. Youse, S. Manouchehri, A. Arab, M. Mozaffari, G.R. Amiri, J. Amighian,
Preparation of cobaltzinc ferrite (Co0.8Zn0.2Fe2O4) nanopowder via combustion method and investigation of its magnetic properties, Materials Research
Bulletin 45 (2010) 17921795.
[28] I.S.A. Farag, M. Ahmed, S. Hammad, A. Moustafa, Application of Rietveld
method to the structural characteristics of substituted copper ferrite compounds, Crystal Research and Technology 36 (2001) 8592.
[29] S.M. Yunus, H. Yamauchi, A.K.M. Zakaria, N. Igawa, A. Hoshikawa, Y. Ishii,
Cation distribution and crystallographic characterization of the quaternary
spinel system MgxCo1xCrxFe2xO4, Journal of Alloys and Compounds 454
(2008) 1015.
[30] S.A. Chandana Rath, R.P. Das, K.K. Sahu, S.D. Kulkarni, Dependence on cation
distribution of particle size, lattice parameter, and magnetic properties in
nanosize MnZn ferrite, journal of Applied Physics 91 (2002) 2211.
[31] O.J.K.A. Navrotsky, Thermodynamics of formation of simple spinels, Journal
of Inorganic and Nuclear Chemistry 30 (1968) 479498.
[32] D. Domide, O. Walter, S. Behrens, E. Kaifer, H.-J. Himmel, Synthesis of
heterobimetallic Zn/Co carbamates: single-source precursors of nanosized
magnetic oxides under mild conditions, European Journal of Inorganic
Chemistry 2011 (2011) 860867.
[33] Y. Wei, K.-B. Kim, G. Chen, Evolution of the local structure and electrochemical properties of spinel LiNixMn2xO4, Electrochimica Acta 51 (2006)
33653373.
[34] T. Inoue, T. Honma, V. Dimitrov, T. Komatsu, Approach to thermal properties
and electronic polarizability from average single bond strength in ZnO
Bi2O3B2O3 glasses, Journal of Solid State Chemistry 183 (2010) 30783085.
[35] B.D. Cullity, Elements of X Ray Diffraction, BiblioBazaar, 2011.
[36] R. Waldron, Infrared spectra of ferrites, Physical Review 99 (1955) 1727.
[37] K.H. Wu, T.H. Ting, M.C. Li, W.D. Ho, Solgel auto-combustion synthesis of
SiO2-doped NiZn ferrite by using various fuels, Journal of Magnetism and
Magnetic Materials 298 (2006) 2532.
[38] P. Priyadharsini, A. Pradeep, P.S. Rao, G. Chandrasekaran, Structural, spectroscopic and magnetic study of nanocrystalline NiZn ferrites, Materials
Chemistry and Physics 116 (2009) 207213.
[39] C. Caizer, M. Stefanescu, Magnetic characterization of nanocrystalline NiZn
ferrite powder prepared by the glyoxylate precursor method, Journal of
Physics D: Applied Physics 35 (2002) 3035.
[40] Q. Zeng, I. Baker, V. McCreary, Z. Yan, Soft ferromagnetism in nanostructured
mechanical alloying FeCo-based powders, Journal of Magnetism and Magnetic Materials 318 (2007) 2838.

J Supercond Nov Magn (2013) 26:477483


DOI 10.1007/s10948-012-1769-9

O R I G I N A L PA P E R

Comparative Studies on the Structure and Magnetic Properties


of NiZn Ferrite Powders Prepared by Glycine-Nitrate
Auto-combustion Process and Solid State Reaction Method
K. Gheisari S.D. Bhame J.T. Oh S. Javadpour

Received: 16 June 2012 / Accepted: 5 September 2012 / Published online: 18 October 2012
Springer Science+Business Media, LLC 2012

Abstract NiZn ferrite compositions (Ni1x Znx Fe2 O4 ) are


well known due to their remarkable soft magnetic properties, which potentially have a broad range of applications in
many areas. In this study, NiZn ferrite with the chemical
formula of Ni0.64 Zn0.36 Fe2 O4 was prepared by the glycinenitrate autocombustion process (GNP) and solid state reaction method (SSRM). In order to achieve a desirable particle size, the SSRM powders were milled for 3 h at a milling
rate of 200 rpm. The structure and magnetic properties of
the ferrite powders, which were synthesized by both methods, were characterized and their properties were compared.
The results indicate that a significant amount ( 90 wt.%) of
nanocrystalline Ni0.64 Zn0.36 Fe2 O4 ferrite with the average
crystallite size of 47 nm, particle size of 200 nm, saturation
magnetization of 73 emu/g and coercivity of 54 Oe has been
formed by means of the glycine-nitrate process. The results
also show that not only the saturation magnetization of the
GNP ferrite powder is relatively similar to that of the milled
K. Gheisari ()
Department of Materials Science and Engineering, Faculty
of Engineering, Shahid Chamran University, Ahvaz, Iran
e-mail: khalil.gheisari@yahoo.com
K. Gheisari
e-mail: khgheisari@scu.ac.ir
S.D. Bhame
Laboratorie CRISMAT, UMR 6508 CNRS-ENSICAEN,
6 Boulevard du Marechal Juin, 14050 CAEN Cedex, France
J.T. Oh
School of Materials Engineering, Nanyang Technological
University, Nanyang Avenue, Singapore 639798, Singapore
S. Javadpour
Department of Materials Science and Engineering, School
of Engineering, Shiraz University, Shiraz, 7134851154, Iran

SSRM powders, but also it is synthesized at a much shorter


duration than that of the solid state reaction method.
Keywords NiZn ferrite Nanocrystalline Magnetic
properties Glycine-nitrate process Solid state reaction
method

1 Introduction
NiZn ferrite (Ni1x Znx Fe2 O4 ) is the most versatile ferrite
from the viewpoint of its large number of applications, due
to its high value of magnetic permeability, high electrical
resistivity, high Curie temperature, and low power loss at
high frequencies [13]. Ni0.64 Zn0.36 Fe2 O4 ferrite is one of
the NiZn ferrite compositions which is well known for its
high electrical resistivity (
= 6200  cm) [4] and its high
saturation magnetization (Ms
= 78 emu/g) [5].
These ferrites are usually prepared by the solid state reaction method (SSRM) for large scale production because
of its easy manufacturing process [2]. Nevertheless, this
method has some serious limitations in the production of
fine powders [6]. Long heating schedules and high temperatures result in the large particle nature of the powders.
Thus, a subsequent grinding or milling operation must be
involved in the process to achieve a specific particle size
distribution, which leads to contamination [2, 7]. As a consequence, these powders usually present undesirable characteristics [7]. Also, in NiZn ferrite, the volatilization of
zinc at high temperatures results in the formation of Fe2+
ions, which increases electron hopping and reduces resistivity [8].

478

Several wet chemical methods, including coprecipitation


[9], hydrothermal [10], citrate precursor [11], solgel process [12], and reverse micelle process [13] have been used
to overcome the above mentioned limitations. Furthermore,
chemical synthesis techniques often provide the best method
for production of nanoparticles due to enhanced homogeneity from the molecular level design of the materials. In addition, solution chemical routes allow control of particle
size, size distribution, and morphology of production powders [14].
These methods often provide ferrites at relatively lower
temperatures (<600 C) with smaller particles (10100 nm)
[15]. However, relatively complex schedules, expensive precursors, and low production rate are the most common problems of the above mentioned methods. The glycine-nitrate
process (GNP) is a wet chemical process known as a rapid
and simple combustion synthesis route, which allows the
preparation of ceramic materials with a high-surface-area
and compositionally homogeneous powders in the nanometer range [16]. The GNP method has been also used to
prepare NiZn ferrite nanoparticles; nevertheless, the assynthesized powders did not exhibit high magnetization values (Ms 26 emu/g) because of their small particle size
(D 10 nm) [1719]. The finite size effect of the nanoparticles leads to a noncollinearity or canting of spins on their
surface, reducing the magnetization value [19].
This work aims to improve the magnetization of assynthesized NiZn ferrite by controlling particle size. From
this point of view, nanocrystalline Ni0.64 Zn0.36 Fe2 O4 powders with a desirable particle size were synthesized by the
GNP method, and for comparison, the same ferrite powder
was also prepared by the solid state reaction method, which
is also known as a conventional ceramic method. The structure and magnetic properties of the prepared powders are
discussed as a function of the synthesis process.

2 Experimental Method
Nanocrystalline Ni0.64 Zn0.36 Fe2 O4 ferrite was synthesized
by the auto-combustion synthesis method using glycine as
a fuel and nitrates as oxidants. The stoichiometric quantities of the water solutions of reagent grade Fe(NO3 )3 9H2 O,
Ni(NO3 )2 6H2 O and Zn(NO3 )2 6H2 O were mixed with two
moles of glycine for one mole of metal in a large beaker. The
mixed solution was then evaporated on a water bath. A thick
formed gel was kept on a hot plate to start combustion reaction at a temperature of 200 C. Ni0.64 Zn0.36 Fe2 O4 ferrite was also synthesized via solid state reaction method by
heating the stoichiometric amount of Fe2 O3 , ZnO and NiO
at 1250 C for 2 h. In order to reduce the synthesized ferrite particle size, the calcined powders were milled for 3 h
at a milling rate of 200 rpm with a ball-powder weight ratio
of 10:1.

J Supercond Nov Magn (2013) 26:477483


Table 1 Instrumental parameters used for the X-ray diffraction Rietveld refinement
Goniometer

Shimadzu XRD-6000

Radiation type, source

X-ray, Cu K

Secondary monochromator

Graphite

Instrument power

40 kV, 40 mA

Detector

Scintillation counter

Receiving slit

0.3 mm

Divergence slit

1 mm

Scattering slit

1 mm

Scan mode

/2

Step size

0.02 (2)

Time per step

1.2 s

Angular range

1090

Temperature

25 C

The phase analysis and the structural properties were


characterized by X-ray diffraction (XRD, Shimadzu XRD6000 diffractometer with Cu K1 radiation). Structural parameters, including crystallite size, lattice parameter, lattice strain, and the quantitative amount of NiZn ferrite
were derived from the Rietvelds powder structure refinement analysis of X-ray powder diffraction data. Instrumental parameters used for the X-ray diffraction Rietveld refinement are listed in Table 1. The Rietveld calculations
were performed by the TOPAS 3 software (from Bruker
AXS). The crystallographic model of NiZn ferrite (Fd3m,
a = 0.84025 nm) was used as the starting model for the
refinements. For each refinement, the following variations
were applied: the background parameter, scale factor, cell
parameter, zero point correction, Lorentzian crystal size,
and Gaussian lattice strain. In order to determine the quality of fitting on the structure model, the Bragg reliability factor (RBragg ) was used. The refinements resulted in
proper fits to the experimental data and average RBragg
value was less than 5 %. The size and morphology of
the particles were determined by scanning electron microscopy (SEM, JEOL JSM-5310) and transmission electron microscopy (TEM, JEOL 2010 at 200 kV). In addition, TEM images were studied to examine the powder microstructure. Magnetic measurements were obtained at room
temperature with a vibrating sample magnetometer (VSM,
Lakeshore 7404) with a saturating field of 10 kOe, field
increment of 404 Oe and a field ramp rate of 40.4 Oe/s.
The average value of the saturation magnetization and coercivity were determined after three times of measurement.

J Supercond Nov Magn (2013) 26:477483

479

Fig. 1 X-ray diffraction pattern


of the Ni0.64 Zn0.36 Fe2 O4
powders synthesized by the
GNP combustion technique.
The experimental data are
shown as a continuous black
line and refined simulated
patterns are shown as red dots.
The difference between the
experimental data and fitted
simulated pattern is shown as a
continuous gray line under the
diffraction pattern. Only the
NiZn ferrite Bragg reflections
are labeled by miller indices
(Color figure online)
Table 2 Structural parameters of the synthesized Ni0.64 Zn0.36 Fe2 O4 powders revealed from the Rietvelds X-ray powder structure refinement
method
Synthesis
method

Weight fraction
(wt.%)

Crystallite size
(nm)

Lattice strain
(%)

Lattice parameter
(nm)

GNP

89.63 0.44

47.29 0.64

0.838180 0.000046

SSRM

100

340 18

0.838887 0.000013

Milled SSRM

100

22.79 0.40

0.1575 0.0062

0.838389 0.000086

3 Results and Discussion


3.1 Structure
The XRD diffraction pattern of the Ni0.64 Zn0.36 Fe2 O4 ferrite powders synthesized by the GNP combustion technique
is shown in Fig. 1. The structural parameters, such as crystallite size, lattice parameter, lattice strain and the amount
of the ferrite phase were derived from the Rietvelds powder structure refinement analysis of the XRD data. In the
Rietveld method, the least squares refinements were carried out until the best fit is obtained between the observed
diffraction patterns and calculated patterns which are based
on the refined structure models [20, 21]. It is clear from this
figure that the observed diffraction pattern (I0 ; continuous
black line) fits well with the refined simulated pattern (Ic ;
small red dots) because the residue of fitting (I0 Ic ) is
negligible. As a result, the structural parameters (Table 2)
can be estimated from the Rietvelds powder structure refinement analysis with high accuracy. It is evident from the
figure that the NiZn ferrite phase has been formed during the autocombustion process. The formation of the Ni
Zn ferrite phase has been noticed clearly due to the appearance of at least (111) (isolated, 2 = 18.32 ) and (220) (isolated, 2 = 30.20 ) reflections and the strongest (311) (overlapped, 2 = 35.52 ) reflection in the respective XRD pattern. The Rietveld analysis of the diffraction pattern reveals
that 89.63 0.44 wt.% of Ni0.64 Zn0.36 Fe2 O4 ferrite phase
with the average crystallite size of 47.29 0.64 nm and the

lattice parameter of 0.838180 0.000046 nm is formed. In


addition to the NiZn ferrite, Fe2 O3 and ZnO phases also
appeared.
It is worth noting that the formation of the cubic spinel
structure of the NiZn ferrite obtained by the GNP combustion technique was confirmed by Verma et al. [15, 17], Deka
et al. [18, 22], and Peng et al. [23]. Verma et al. [15, 17]
and Deka et al. [18, 22] obtained an average crystallite size
less than 10 nm and lattice parameter about 0.840 nm, and
Peng et al. [23] achieved an average crystallite size value of
25 nm. It should be mentioned that they have used different molar ratio of raw materials to prepare the NiZn ferrite
with the composition of Ni0.50 Zn0.50 Fe2 O4 . They have also
used a dissimilar molar ratio of glycine and additional reactants, which may be responsible for their different crystallite
size.
Figure 2 presents a comparison of the X-ray diffraction
pattern of the GNP, SSRM, and the milled SSRM ferrite
powders. The structural parameters of the powders synthesized by the above mentioned processes are listed in Table 2.
As opposed to the glycine-nitrate process, the solid state reaction method leads to a single phase ferrite powders in such
a way the Hematite (Fe2 O3 ), Bunsenite (NiO), and Zincite
(ZnO) peaks are completely disappeared. Although the Xray diffraction pattern of the milled SSRM ferrite powders
also presents single phase ferrite powders, all the peaks are
broadened and their intensity is much weaker than that of
the as-synthesized powders obtained by the SSRM process.
The peak broadening is probably caused by a decrease in

480

crystalline size and an increase in lattice strain. From the


Rietvelds powder structure refinement analysis, the value
of 22.79 0.40 nm and 0.1575 0.0062 are estimated for
crystalline size and lattice strain, respectively. Therefore, the
milling process leads to the formation of nanocrystalline single phase ferrite powders with a high level of internal strain

Fig. 2 X-ray powder diffraction patterns of the Ni0.64 Zn0.36 Fe2 O4


powders synthesized by solid state reaction method (a), solid state reaction method with a subsequent milling process (b) and GNP combustion method (c). Only NiZn-ferrite reflections have been indexed

J Supercond Nov Magn (2013) 26:477483

in the crystal structure of the powders. Generally, internal


strain has a negative effect on the magnetic properties.
Figure 3 indicates the SEM and TEM micrographs of the
as-synthesized powders prepared by the GNP method, having a considerably large surface area. The mean particle size
which is roughly obtained by the SEM image is less than
1 m. The mean particle size, estimated with more accuracy
by the TEM micrographs, is found to be 200 nm. The particle size derived from the SEM and TEM micrographs is
several times greater than the crystallite size estimated from
the XRD data. This is indicative of the fact that every particle is formed by the aggregation of a large number of crystallites or grains. The nanocrystalline structure of the powders can be confirmed by the high resolution electron microscope (HRTEM) images. Figure 4 shows the HRTEM bright
field image and the corresponding selected area electron
diffraction (SAED) of the as-synthesized powder. The corresponding diffraction pattern and the lattice fringes indicate
an almost completely random orientation of the resulting
nanocrystalline structure with the crystallite size of 15 nm.
Although this value is less than that obtained from the XRD
results, it is relatively consistent with the value measured by
Deka et al. [18, 22]. The different values of crystallite size
are caused by the different measuring methods. In the Xray powder diffraction method, the average crystallite size
is determined from the X-ray powder diffraction data of the
powder samples, which may contain thousands of powder
particles. Nevertheless, the crystallite size is estimated from
few powder particles images in the HRTEM observations.

Fig. 3 SEM (a) and TEM micrograph (b) of the Ni0.64 Zn0.36 Fe2 O4 powders prepared by the GNP method

J Supercond Nov Magn (2013) 26:477483

481
Table 3 Magnetization (Ms ) and coercivity (Hc ) of the NiZn ferrite
powders obtained by the different synthesis methods

Fig. 4 HRTEM image and the corresponding electron diffraction pattern of a nanocrystalline Ni0.64 Zn0.36 Fe2 O4 powder particle prepared
by the GNP method

Fig. 5 Hysteresis loops of the NiZn ferrite powders obtained by the


glycine-nitrate process (a) and solid state reaction method with (b) and
without (c) a subsequent milling process

3.2 Magnetic Properties


Figure 5 shows a comparison of the hysteresis curves (magnetization vs. external field) of the Ni0.64 Zn0.36 Fe2 O4 synthesized powders obtained by the glycine-nitrate process
and the solid state reaction method with and without the
subsequent milling. The saturation magnetization and coercivity can be evaluated from these curves. The relevant data

Synthesis method

Ms (emu/g)

GNP

73.47

SSRM

80.54

Milled SSRM

73.88

Hc (Oe)
53.97
6.41
220.5

is listed in Table 3. Magnetic measurements show that the


as-synthesized powder prepared by the solid state reaction
method has relatively higher saturation magnetization and
lower coercivity than those obtained by the GNP method.
The saturation magnetization and coercivity values of the
as-synthesized SSRM and GNP powders are 80.54 emu/g,
6.41 Oe and 73.47 emu/g, 53.97 Oe, respectively. The lower
saturation magnetization and the higher coercivity of the
powder synthesized by the GNP combustion technique may
be due to the introduction of impurities and the formation
of the nonmagnetic phases during the synthesis process. It
is well known that a soft magnetic material must have a low
coercivity and a high saturation induction for most application, such as transformer cores [24, 25]. From this point of
view, the magnetic properties of the as-synthesized SSRM
powders are superior to those obtained by the GNP method.
Although the single phase ferrite powders prepared by
the solid state reaction method indicates the preferable magnetic properties, the subsequent milling process is unavoidable because of the undesirable particle size. As can be seen
in Table 3, the saturation magnetization of the milled SSRM
powders decreases to 73.88 emu/g and the coercivity increases to 220.5 Oe. One of the possible reasons for the
dramatic increase of coercivity is internal strain induced by
the subsequent milling process. In fact, some authors believe
that internal strain can be the dominating factor in the coercivity [26, 27]. According to this hypothesis, the increase of
the coercivity can be understood as being the consequence
of a considerable introduction of structure defects, such as
internal strain into the material, which is inevitably related
to the milling process. Internal strain and other structure
defects impose limitations on domain wall motion during
magnetization-reversal process, enhancing coercivity [28].
The magnetic measurement data given in Table 3 show that
the GNP method not only can be successfully applied to synthesize nanostructured Ni0.64 Zn0.36 Fe2 O4 composition with
a desirable particle size, but also it can present the magnetic
properties, which is comparable to the milled ferrite powder
prepared by the solid state reaction method. In addition, a
further milling process is not required to obtain a desirable
particle size.
Some structure and magnetic properties of the GNPnickelzinc ferrite powder in comparison with those of
as-prepared nickelzinc ferrites obtained by different wet
chemical synthesize methods are listed in Table 4. These

482

J Supercond Nov Magn (2013) 26:477483

Table 4 Comparison of the particle size and magnetic properties of the as-synthesized NiZn ferrite powders prepared by the different wet
chemical methods
Sample
Ni0.5 Zn0.5 Fe2 O4
Ni0.20 Zn0.44 Fe2.36 O4

Ms (emu/g)

Hc (Oe)

Particle size (nm)

Techniques

References

5.4

Reverse micelle method

[29]

25.9

Reverse micelle method

[13]

Ni0.5 Zn0.5 Fe2 O4

20

GNP method

[17]

Ni0.5 Zn0.5 Fe2 O4

26

50

Modified GNP method

[18]

Ni0.6 Zn0.4 Fe2 O4

34

35

Oxalate based precursor method

[30]

Ni0.5 Zn0.5 Fe2 O4

44.24

52.12

2025

Combustion method

[23]

Ni0.6 Zn0.4 Fe2 O4

53.5

100150

Solvothermal method

[31]

Ni0.64 Zn0.36 Fe2 O4

73.47

53.97

200

GNP method

This study

data have been previously reported by other authors. Referring to Table 4, the saturation magnetization of the ferrite
powders processed in this study is higher than that reported
for the as-prepared ferrite powders processed by other wet
chemical methods. This result can be attributed to the particle size effect. As shown in Table 4, the saturation magnetization decreases as the particle size is decreased in such
a way a lower saturation magnetization corresponds to a
smaller particle size.

4 Conclusion
Nanocrystalline Ni0.64 Zn0.36 Fe2 O4 ferrite powders with
high purity have been successfully synthesized by the
glycine-nitrate autocombustion, which was confirmed by the
XRD studies. The same ferrite powder also has been prepared by the solid state reaction method. The as-synthesized
ferrite powders prepared by this method indicate the preferable magnetic properties. However, due to the undesirable
particle size, the subsequent milling process is unavoidable,
which leads to a decrease in saturation magnetization (73.88
emu/g) and an increase in coercivity (220.5 Oe). On the
other hand, nanocrystalline Ni0.64 Zn0.36 Fe2 O4 ferrite powders with the average particle size of 200 nm, coercivity
of 53.97 Oe, and saturation magnetization of 73 emu/g is
obtained by the GNP method. In addition to a desirable
particle size, the magnetic properties of the as-synthesized
GNP powders are relatively comparable to those measured
for the same milled ferrite powders prepared by the solid
state reaction method and, in most cases, even better that
those reported for as-synthesize NiZn ferrite powders prepared by other wet chemical methods. Consequently, the
as-synthesized GNP powder without an additional process
can be used for practical applications.
The glycine-nitrate process reported in this paper is technically simple and cost effective because initial materials are
very cheap and the method does not require any complex
equipment. The method is versatile and can be utilized to
synthesize different ferrite powders.

Acknowledgement The authors would like to thank the Nanyang


Technological University (NTU) for providing support to this research.

References
1. Goldman, A.: Modern Ferrite Technology. Van NostrandReinhold, New York (1990)
2. Snelling, E.C.: Ferrites for Inductors and Transformers. Research
Studies Press, New York (1983)
3. Harris, V.G.: Modern microwave ferrites. IEEE Trans. Magn.
48(3), 10751104 (2012)
4. Akther Hossain, A.K.M., Mahmud, S.T., Seki, M., Kawai, T.,
Tabata, H.: Structural, electrical transport, and magnetic properties
of Ni1x Znx Fe2 O4 . J. Magn. Magn. Mater. 312, 210219 (2007)
5. Mirzaee, O., Golozar, M.A., Shafyei, A.: Influence of V2 O5 as an
effective dopant on the microstructure development and magnetic
properties of Ni0.64 Zn0.36 Fe2 O4 soft ferrites. Mater. Charact. 59,
638641 (2008)
6. Chandrasekaran, G.: Spectroscopic study of autocombustion process in the synthesis of nano particles of NiCu ferrite. Mater.
Manuf. Process. 22, 366369 (2007)
7. Costa, A.C.F.M., Tortella, E., Morelli, M.R., Kiminami,
R.H.G.A.: Synthesis, microstructure and magnetic properties
of NiZn ferrites. J. Magn. Magn. Mater. 256M, 174182 (2003)
8. Verma, A., Goel, T.C., Mendiratta, R.G., Gupta, R.G.: Highresistivity nickelzinc ferrites by the citrate precursor method.
J. Magn. Magn. Mater. 192, 271276 (1999)
9. Albuquerque, A.S., Ardisson, J.D., Macedo, W.A.A., Lopez, J.L.,
Paniago, R., Persiano, A.I.C.: Structure and magnetic properties of
nanostructured Ni-ferrite. J. Magn. Magn. Mater. 226230, 1379
1381 (2001)
10. Huo, J., Wei, M.: Characterization and magnetic properties
of nanocrystalline nickel ferrite synthesized by hydrothermal
method. Mater. Lett. 63, 11831184 (2009)
11. Verma, A., Goel, T.C., Mendiratta, R.G.: Frequency variation of
initial permeability of NiZn ferrites prepared by the citrate precursor method. J. Magn. Magn. Mater. 210, 274278 (2000)
12. Albuquerque, A.S., Ardisson, J.D., Macedo, W.A.A.: A study
of nanocrystalline NiZnferriteSiO2 synthesized by solgel.
J. Magn. Magn. Mater. 192, 277280 (1999)
13. Morrison, S.A., Cahill, C.L., Carpenter, E.E., Calvin, S., Swaminath, R., McHenry, M.E., Harris, V.G.: Magnetic and structural
properties of nickel zinc ferrite nanoparticles synthesized at room
temperature. J. Appl. Phys. 95, 63926395 (2004)
14. Willard, M.A., Kurihara, L.K., Carpenter, E.E., Calvin, S., Harris, V.G.: Chemically prepared magnetic nanoparticles. Int. Mater.
Rev. 49, 125170 (2004)

J Supercond Nov Magn (2013) 26:477483


15. Verma, S., Pradhan, S.D., Pasricha, R., Sainkar, S.R., Joy, P.A.:
A novel low-temperature synthesis of nanosized NiZn ferrite.
J. Am. Ceram. Soc. 88, 25972599 (2005)
16. Chick, L.A., Pederson, L.R., Maupin, G.D., Bates, J.L., Thomas,
L.E., Exarhos, G.J.: Glycine-nitrate combustion synthesis of oxide
ceramic powders. Mater. Lett. 10, 612 (1990)
17. Verma, S., Joy, P.A.: High Curie temperature of nanosized NiZn
ferrite particles synthesized by a combustion method. Int. J.
Nanomed. 7, 4349 (2008)
18. Deka, S., Joy, P.A.: Characterization of nanosized NiZn ferrite
powders synthesized by an autocombustion method. Mater. Chem.
Phys. 100, 98101 (2006)
19. Priyadharsini, P., Pradeep, A., Sambasiva Rao, P., Chandrasekaran, G.: Structural, spectroscopic and magnetic study of
nanocrystalline NiZn ferrites. Mater. Chem. Phys. 116, 207213
(2009)
20. Choo, K.S., Gheisari, Kh., Oh, J.T., Javadpour, S.: Structure and
magnetic properties of nanostructured Ni0.77 Fe0.16 Cu0.05 Cr0.02
(Mumetal) powders prepared by mechanical alloying. Mater. Sci.
Eng. B 157, 5357 (2009)
21. Pourghahramani, P., Altin, E., Mallembakam, M.R., Peukert, W.,
Forssberg, E.: Microstructural characterization of hematite during
wet and dry Millings using Rietveld and XRD line profile analyses. Powder Technol. 186, 921 (2008)
22. Deka, S., Joy, P.A.: Enhanced permeability and dielectric constant
of NiZn ferrite synthesized in nanocrystalline form by a combustion method. J. Am. Ceram. Soc. 90, 14941499 (2007)
23. Peng, C.H., Hwang, C.C., Wan, J., Tsai, J.S., Chen, S.Y.:
Microwave-absorbing characteristics for the composites of

483

24.
25.
26.

27.

28.
29.

30.

31.

thermal-plastic polyurethane (TPU)-bonded NiZn-ferrites prepared by combustion synthesis method. Mater. Sci. Eng. B 117,
2736 (2005)
Cullity, B.D., Graham, C.D.: Introduction to Magnetic Materials,
2nd edn. Wiley, Hoboken (2009)
Smith, W.F., Hashemi, J.: Foundations of Materials Science and
Engineering, 5th edn. McGraw-Hill, New York (2010)
Hamzaoui, R., Elkedim, O., Gaffet, E., Greneche, J.M.: Structure, magnetic and Mossbauer studies of mechanically alloyed Fe
20 wt.% Ni powders. J. Alloys Compd. 417, 3238 (2006)
Guittoum, A., Layadi, A., Bourzami, A., Tafat, H., Souami,
N., Boutarfaia, S., Lacour, D.: X-ray diffraction, microstructure,
Mossbauer and magnetization studies of nanostructured Fe Ni alloy prepared by mechanical alloying. J. Magn. Magn. Mater. 320,
13851392 (2008)
Tremolet de Lacheisserie, E.du., Gignoux, D., Schlenker, M.:
Magnetism: Materials and Application. Springer, Boston (2005)
Uskokovic, V., Drofenik, M., Ban, I.: The characterization of
nanosized nickelzinc ferrites synthesized within reverse micelles
of CTAB/1hexanol/water microemulsion. J. Magn. Magn. Mater.
284, 294302 (2004)
Sarangi, P.P., Vadera, S.R., Patra, M.K., Ghosh, N.N.: Synthesis
and characterization of pure single phase NiZn ferrite nanopowders by oxalate based precursor method. Powder Technol. 203,
348353 (2010)
Yan, W., Jiang, W., Zhang, Q., Lia, Y., Wang, H.: Structure and
magnetic properties of nickelzinc ferrite microspheres synthesized by solvothermal method. Mater. Sci. Eng. B 171, 144148
(2010)

Hyperfine Interact (2008) 184:227233


DOI 10.1007/s10751-008-9794-6

Magnetic and Mssbauer studies on nanocrystalline


Co1x Lix Fe2 O4 (x = 0, 0.2)
R. S. Ningthoujam S. S. Umare S. J. Sharma
R. Shukla Sajith Kurian R. K. Vatsa
A. K. Tyagi R. Tewari G. K. Dey N. S. Gajbhiye

Published online: 2 October 2008


Springer Science + Business Media B.V. 2008

Abstract Ultrafine particles of Co1x Lix Fe2 O4 (x = 0, 0.2) samples are prepared
by glycinenitrate combustion route. X-ray diffraction and transmission electron
microscopy studies show that the samples have cubic spinel structure and average
crystallite sizes of x = 0 and 0.2 are 36 and 44 nm respectively. Vibrating sample
magnetometer studies revealed the ferromagnetic nature of the samples. Li-doped
CoFe2 O4 sample showed higher values of coercive field, remanent magnetization
and saturation magnetization compared to pure CoF2 O4 indicating the enhancement
of magnetic interactions. Mssbauer spectra at 77 K exhibited two broad sextets
indicating that Fe3+ ions occupy both tetrahedral and octahedral sites. From these
studies, it is concluded that Co1x Lix Fe2 O4 (x = 0, 0.2) samples exhibit an inverse
spinel structure. At room temperature, two sextets are superimposed on a very broad

R. S. Ningthoujam (B) R. Shukla R. K. Vatsa A. K. Tyagi


Chemistry Division, Bhabha Atomic Research Centre,
Mumbai 400085, India
e-mail: rsn@barc.gov.in
R. K. Vatsa
e-mail: rkvatsa@barc.gov.in
S. S. Umare
Department of Chemistry,
Visvesvaraya National Institute of Technology, Nagpur, India
S. J. Sharma
Department of Electronics, S. K. Porwal College,
Kamptee, Nagpur 441002, India
S. Kurian N. S. Gajbhiye
Department of Chemistry, Indian Institute of Technology Kanpur,
Kanpur 208016, India
R. Tewari G. K. Dey
Material Science Division,
Bhabha Atomic Research Centre, Mumbai 400085, India

228

R.S. Ningthoujam et al.

non-Lorentzian background indicating the presence of superparamgnetic fraction in


agreement with the microscopic observations.
Keywords Mssbauer studies Ferrites CoFe2 O4 sample

1 Introduction
Ferrites are useful materials in many applications such as magnetic devices, microwave absorbers, chemical sensors, catalysts, and biomedical applications [16].
In order to control properties, size and shape of the particles and their distribution
are very critical. CoFe2 O4 is the cubic spinel with space group Fd3m . The cationic
distribution is described by the chemical formula: (Fe1x Cox )[Fe1+x Co1x ]O4 . i. e.
(A)[B/ B// ]O4 , ( ) and [ ] denote the tetrahedral A and octahedral B sites respectively.
When x = 0, structure is fully inverse, otherwise the degree of inversion changes
leading to the metastable properties. In order to understand the cationic distribution,
it is necessary to know the thermal history of the sample. When CoFe2 O4 is doped
with other elements such as Mg, Ni, Cu, Zn, Li etc, there is effect on cationic
distribution at tetrahedral and octahedral sites.
Bulk CoFe2 O4 is ferrimagnetic below the Curie temperature (Tc = 860 K) [5]. At
room temperature, it has saturation magnetization (Ms ) of 80 emu/g and coercivity
(Hc ) of (110) kOe [16]. It has easy magnetization axis along [100] with magnetic
saturation (3.95 B ) per formula unit at 5 K and 3.35 B at 300 K [7]. From a
molecular field model, there are six exchange parameters between tetrahedral Fe3+
2+
(A), octahedral Fe3+ (B/ ) and Co
(B// ) when x = 0. It leads to antiferromagnetic

exchange constants J = 20 K JAA , JAB , JAB / , JB / B / , JB / B // except for JB // B //
+40 K. When x = 0, there is no Co2+ ions at the A site. In view of this, expected value
of saturation magnetization is 3 B by considering all the A moments parallel to one
another and antiparallel to the B moments. This 3 B is less than the experimentally
measured value of 3.95 B . The higher measured value could be due to contribution
from orbital moments [5]. In this work, synthesis, magnetic and Mssbauer studies
of Li doped CoFe2 O4 ultrafine particles are presented.

2 Experimental section
Li-doped CoFe2 O4 samples (Co1 x Lix Fe2 O4 , x = 0, 0.2) are prepared by glycine
nitrate combustion route. According to the stoichiometric composition of
Co1 x Lix Fe2 O4 (x = 0, 0.2) specified amounts of lithium acetate, ferric nitrate, cobalt
acetate and glycine in equimolar ratio of metal ions and glycine are dissolved in
minimum quantity of dilute nitric acid. At this stage, the metal acetates are converted
to metal nitrates. The solution is heated on a hot plate at 250 C for 30 min. Initially it
undergoes dehydration at 80 C forming a viscous gel followed by decomposition of
glycine with the evolution of large amount of gases. The powder obtained is calcined
at 575 C for 30 min to remove all the traces of carbonaceous impurities.
X-ray diffraction measurements are carried on the combustion-synthesized powder for phase confirmation and crystallite size estimation using a Philips X-ray
diffractometer (Model-1718) with Ni filtered Cu-K radiation. The average crystallite

Magnetic and Mssbauer studies on nanocrystalline Co1 x Lix Fe2 O4


Fig. 1 Field dependent
magnetization (MH)
curves for Co1 x Lix Fe2 O4
(x = 0.0, 0.2) at room
temperature

229

80

x = 0.0
x = 0.2

60

M (emu/g)

40
20
0
-20
-40
-60
-80

-20

-15

-10

-5

10

15

20

H (kOe)

size has been calculated from the line broadening using the Scherrers relation.
Morphology of particles (size and shape) is determined from Scanning Electron
Microscope (SEM, model JSM-840A) on pellet sample as well as Transmission Electron Microscope (TEM, JEOL 2000 FX microscope) on powder sample. Magnetization measurements are carried out using Vibrating Sample Magnetometer (VSM).
Mssbauer spectra were recorded using a commercial spectrometer operating in
constant acceleration mode at room and liquid nitrogen temperatures. Magnetic
hyperfine fields (Hhf ), quadrupole splitting () and isomer shift () values are calculated with a WIN-NORMOS fit program.

3 Results and discussion


XRD patterns of Co1 x Lix Fe2 O4 (x = 0, 0.2) were recorded at room temperature
and all the diffraction lines are indexed for the cubic structure (not shown). The
measured lattice parameter a of pure CoFe2 O4 is 8.391(1) , which is in good
agreement with literature data (a = 8.391 , JCPDS 22-1086). For x = 0.2 sample,
lattice parameter a is found to be 8.360(1) indicating decrease in the cell
parameter on lithium doping. The ionic size of Li+ ion is 0.59 , which is less
than that of Co2+ (0.74 ). The reported value of lattice parameter of LiFe5 O8
is a = 8.331 (JCPDS 38-259) indicating that Li-ferrite has lower lattice parameter
compared to that of Co-ferrite. The average crystallite sizes of x = 0.0 (CoFe2 O4 )
and x = 0.2 samples calculated from Scherrers formula are found to be 36 and 44 nm,
respectively.
The smallest particle size of pure CoFe2 O4 measured from SEM is about 100 nm
(not shown). The presence of agglomerated particles is due to the magnetic interactions amongst the particles. TEM study and associated selected area diffraction pattern for pure CoFe2 O4 (not shown) indicating that CoFe2 O4 is highly crystalline.The
smallest particle size measured from TEM is found to be 40 nm.

230

R.S. Ningthoujam et al.

Counts (arb. units)

Fig. 2 Mssbauer spectra of


Co1 x Lix Fe2 O4 (x = 0.0, 0.2)
at liquid nitrogen temperature

(b) x = 0.2

(a) x = 0.0
-10

-5

0
Velocity (mm/s)

10

Table 1 The hyperfine values (Hhf ), isomer shift (), quadrupole splitting (), linewidth ( ) and
areas in percentage of tetrahedral and octahedral sites of Fe3+ ions at liquid nitrogen temperature
for Co1 x Lix Fe2 O4
Phases

Iron sites

Isomer shift
() mm/s

Quadrupole
splitting
(EQ ) mm/s

Hyperfine
splitting
(Hf )104 Oe

Width
( ) mm/s

Area (%)

CoFe2 O4

Sextet 1
Sextet 2
Sextet 1
Sextet 2

0.38
0.49
0.36
0.48

0.00
0.03
0.00
0.03

47.72
50.54
46.54
49.08

0.44
0.46
0.51
0.50

62
38
63.6
36.3

Co0.8 Li0.2 Fe2 O4

Figure 1 shows the field dependent magnetization (MH) curves for


Co1 x Lix Fe2 O4 (x = 0.0, 0.2) at room temperature indicating their ferromagnetic
nature at room temperature. The M values at 17 kOe for x = 0.0 and 0.2 are 65.7 and
67.5 emu/g respectively. Since magnetization value is not saturated up to 17 kOe,
saturation magnetization (Ms ) can be estimated by extrapolation of 1/H 0 in
M vs. 1/H curve [8], which yielded the Ms values for x = 0 and 0.2 as 71.3 and
72.7 emu/g respectively which are less than bulk CoFe2 O4 value of 80 emu/g [5] at
room temperature. Coercivity (Hc ) values for x = 0 and 0.2 are 950 and 1,000 Oe
respectively. It means that Li-doping enhances Hc . Enhancement in Hc is related

Magnetic and Mssbauer studies on nanocrystalline Co1 x Lix Fe2 O4

231

Counts (arb. units)

Fig. 3 Mssbauer spectra of


Co1 x Lix Fe2 O4 (x = 0.0, 0.2)
at room temperature

(b) x = 0.2

(a) x = 0.0
-10

-5

0
Velocity (mm/s)

10

Table 2 The hyperfine values (Hhf ), isomer shift (), quadrupole splitting (), linewidth ( ) and
areas in percentage of tetrahedral and octahedral sites of Fe3+ ions at room temperature for
Co1 x Lix Fe2 O4
Phases

Iron sites

Isomer shift
() mm/s

Quadrupole
splitting
(EQ ) mm/s

Hyperfine
splitting
(Hf )104 Oe

Width
( )mm/s

CoFe2 O4

Sextet 1
Sextet 2
Sextet 3
Doublet
Sextet 1
Sextet 2
Sextet 3
Doublet

0.30
0.40
0.51
1.08
0.40
0.29
0.55
0.94

0.00
0.08
0.00
0.60
0.04
0.00
0.02
0.76

49.36
52.00
46.01

51.69
49.45
45.77

0.44
0.30
0.64
0.42
0.43
0.43
0.83
0.58

Co0.8 Li0.2 Fe2 O4

to increase in anisotropy (Ka ). The enhancement in Ms and Hc on Li-doping may


be due to the presence of Co2+ in order to compensate charge balance in the
Li-doped sample. Remanence (Mr ) values for x = 0 and 0.2 are 26.2 and 30.9 emu/g
respectively. The ratio (Mr /Ms ) are 0.37 and 0.43 for x = 0 and 0.2 respectively.
For an assembly of non-interacting, randomly oriented single domain particles one
expects this ratio (Mr /Ms ) 0.5 [9]. Hadjipanayis et al. [10] have found that ratio
Mr /Ms > 0.5 for the strong ferromagnetic interactions among particles present in
the system and Mr /Ms < 0.5 for anti-ferromagnetic interactions. The Mr /Ms values
measured in our work are less than 0.5 indicating presence of anti-ferromagnetic
interactions.

232

R.S. Ningthoujam et al.

Figure 2 shows the Mssbauer spectra of Co1 x Lix Fe2 O4 (x = 0, 0.2) at liquid
nitrogen temperature and the derived hyperfine interaction parameters are given
in Table 1. The x = 0.0 sample (pure CoFe2 O4 ) has two sextets (Zeeman patterns)
indicating that CoFe2 O4 has inverse spinel structure. The values of these parameters
are different from those reported ones for the bulk samples indicating the ultrafine
nature of these particles. Linewidths ( ) for tetrahedral and octahedral sites of pure
CoFe2 O4 are 0.44 and 0.46 mm/s respectively and of Li-doped CoFe2 O4 are 0.51
and 0.50 mm/s respectively. These values are larger than the reported bulk value in
CoFe2 O4 (0.35 mm/s) [6]. Areas in relative percentage in tetrahedral and octahedral
sites are 62% and 38% respectively in CoFe2 O4 . It means that the formula of
composition will be (Co0.8 Fe1.2 )[Co0.2 Fe0.8 ]O4 . While, areas in relative percentage in
tetrahedral and octahedral sites are 64% and 36% respectively in Li (x = 0.2) doped
CoFe2 O4 . It corresponds to the formula of composition: (Co0.7 Fe1.3 )[Co0.3 Fe0.7 ]O4 .
Figure 3 shows the room temperature Mssbauer spectra of Co1 x Lix Fe2 O4 (x =
0.0, 0.2). The sextet peaks are more asymmetric compared to that in liquid nitrogen
temperature and also an additional unresolved broad contribution is seen clearly in
the spectrum and the parameters are given in Table 2. Due to the nano-nature of
some of the particles, it is possible to observe the partially collapsed spectrum due
to the shorter spin relaxation time of ferric ions. Similar complex spectra have been
reported for nano crystalline spinel ferrites in the literature [11].

4 Conclusions
Ultrafine particles of Co1 x Lix Fe2 O4 (x = 0, 0.2) are prepared by glycinenitrate
combustion method. The enhancement in Ms and Hc on Li-doping is due to the
presence of Co2+ in order to compensate charge balance in the Li-doped sample.
Room temperature Mssbauer studies have indicated the presence of nano particles
(spin relaxation effects) in addition to the larger particles exhibiting partially inverted
spinel structure. However, the spectra at 77 K, spin relaxation has sufficiently slowed
down to observe the characteristic spectra assigned to the iron ions occupying both
the tetrahedral and octahedral lattice sites in spinel ferrites.

References
1. Gajbhiye, N.S., Bhattacharyya, S., Balaji, G., Ningthoujam, R.S., Das, R.K., Basak, S.,
Weissmller, J.: Mssbauer and magnetic studies of MFe2 O4 (M = Co, Ni) nanoparticles.
Hyperfine Interact. 165, 153159 (2005)
2. Xu, Z.C.: Magnetic anisotropy and Mssbauer spectra in disordered lithiumzinc ferrites.
J. Appl. Phys. 93, 47464749 (2003)
3. Gajbhiye, N.S., Ningthoujam, R.S., Weissmller, J.: Mssbauer study of nanocrystalline
-Fe3x Cox N system. Hyperfine Interact. 156, 5156 (2004)
4. Kikukawa, N., Takemori, M., Nagano, Y., Sugasawa, M., Kobayashi, S.: Synthesis and magnetic
properties of nanostructured spinel ferrites using a glycinenitrate process. J. Magn. Magn.
Mater. 284, 206214 (2004)
5. Cullity, B.D.: Introduction to Magnetic Materials. Addison-Wesley, London (1972)
6. Ngo, A.T., Bonville, P., Pileni, M.P.: Spin canting and size effects in nanoparticles of nonstoichiometric cobalt ferrite. J. Appl. Phys. 89, 33703376 (2001)
7. Teillet, J., Bouree, F., Krishnan, R.: Magnetic structure of CoFe2 O4 . J. Magn. Magn. Mater. 123,
9396 (2003)

Magnetic and Mssbauer studies on nanocrystalline Co1 x Lix Fe2 O4

233

8. Gajbhiye, N.S., Ningthoujam, R.S., Bhattacharyya, S.: Magnetic properties of Co and Ni


substituted-Fe3 N nanoparticles. Hyperfine Interact. 164, 1726 (2005)
9. Vandenberghe, R.E., Grave de, E.: Mssbauer spectroscopy applied to inorganic chemistry,
vol. 3. Plenum, New York (1989)
10. Hadjipanayis, G., Sellmyer, D.J., Brandt, B.: Rare-earth-rich metallic glasses. I. Magnetic hysteresis. Phys. Rev. B. 23, 33493354 (1981)
11. Prasad, S.: Solid state reactivity, characterization and anomalous magnetic behaviour of
nanocrystalline spinel ferrite particles synthesized by citrate precursor technique. Dissertation,
Indian Institute of Technology, Kanpur, India (1997)

ARTICLE IN PRESS

Journal of Solid State Chemistry 178 (2005) 382389


www.elsevier.com/locate/jssc

Rapid communication

Combustion synthesis of NiZn ferrite powderinuence of


oxygen balance value
Chyi-Ching Hwanga,, Jih-Sheng Tsaia, Ting-Han Huanga, Cheng-Hsiung Pengb,
San-Yuan Chenb
a

Department of Applied Chemistry, Chung Cheng Institute of Technology, National Defence University, Number 190, Sanyuan First Street,
Tashi Jen, Taoyuan 33509, Taiwan
b
Department of Materials Science and Engineering, National Chao Tung University, Hsinchu 300, Taiwan
Received 30 August 2004; received in revised form 26 October 2004; accepted 28 October 2004

Abstract
In this study, Ni0.5Zn0.5Fe2O4 powder was synthesized via an exothermic reaction between nitrates [Ni(NO3)2  6H2O,
Zn(NO3)2  6H2O, Fe(NO3)3  9H2O, and NH4NO3] and glycine [NH2CH2COOH]. By adjusting the glycine-to-nitrates ratio, the
oxygen balance (OB) values of the reactant mixtures can be varied in which the combustion phenomena is altered and thereby the assynthesized products with different characteristics are obtained. An interpretation based on the measurement of maximum
combustion temperature (Tc) and the amounts of gas evolved during reaction for various OB values has been proposed regarding the
nature of combustion and its correlation with the characteristics of as-synthesized products. After instrumental analyses, it is shown
that the as-synthesized powders are nanoscale crystallites with a large specic surface area and they inherit a superparamagnetic
behavior.
r 2004 Elsevier Inc. All rights reserved.
Keywords: Combustion synthesis; Nanocrystalline; Oxygen balance; Nitrates; Glycine

1. Introduction
In the past decade, a solution combustion method has
been narrated and utilized to synthesize simple and
mixed metal oxides [19]. With this method, the heating
and evaporation of desired nitrate solution employing
an organic compound (usually glycine, urea, or citric
acid, etc.) can result in self-ring to generate heat by
exothermic reaction. This liberated heat is used to
synthesize the ceramic oxide powders. Together, this
method has advantages of applying inexpensive raw
materials, maintaining a relatively simple, quick and
straightforward preparation process, and achieving a
ne powder with high homogeneity. Recently, a wide
range of technological applications has made homoCorresponding author. Fax: +886 3 389 2494.

E-mail address: cchwang1@ccit.edu.tw (C.-C. Hwang).


0022-4596/$ - see front matter r 2004 Elsevier Inc. All rights reserved.
doi:10.1016/j.jssc.2004.10.045

geneous and nano-crystalline particles the necessary


materials in electronics industry. NiZn ferrite with its
commercial usage in power transformers, microwave
devices and telecommunication applications [10] made a
favorable choice due to its high resistivity, low coercivity
and inappreciable eddy current loss [11] arising from the
fact that they are capable of being used as a softmagnetic material.
In this study, to open the door to sources of other
preferences, Ni0.5Zn0.5Fe2O4 was synthesized using
metal (Ni, Zn, Fe) nitrates (acting as the dual roles of
oxidant and metal source), ammonium nitrate (oxidant)
and glycine (fuel) via combustion reaction. Glycine
was chosen as fuel owing to its inexpensive price.
Also, its melting point is higher and its heat of
combustion is more negative when compared with
urea and citric acid. When complete combustion is
assumed, the reaction equation can be expressed (greatly

ARTICLE IN PRESS
C.-C. Hwang et al. / Journal of Solid State Chemistry 178 (2005) 382389

simplied) as follows:
0:5NiNO3 2  6H2 Oc 0:5ZnNO3 2  6H2 Oc
2FeNO3 3  9H2 Oc cNH2 CH2 COOHc
kNH4 NO3c

mixing & dehydration

!

Ni0:5 Zn0:5 Fe2 C2c H5c4k N8c2k O242c3k


combustion

the general formula of mixture !




5c
2k H2 Og
Ni0:5 Zn0:5 Fe2 O4c 2cCO2g
2




c
9c k

4 k N2g 10 
O2g
2
4
2



11
7
1
14 c k moles of gas produced :
4
2
According to the denition of oxygen balance (OB)
employed in the eld of propellants and explosives
described in Ref. [12], it is appropriate to express OB
value in terms of the weight percent of excessive oxygen
required versus the formula weight of the mixture.


AWoxygen
9c
OB%
k  100%;
(2)
20 
2
FWmixture
where AWoxygen and FWmixture are the atomic weight of
oxygen and the formula weight of the mixture,
respectively. From Eqs. (1) and (2), OB value of the
reactant mixture can be adjusted by varying the values
of C and k; as illustrated below:
1. OBo0 (fuel-rich), when C4 40
9 4:44 and k 0:
Obtaining oxygen from atmosphere is required for
combustion between glycine and metal nitrates.
2. OB 0 (stoichiometric), when C 4.44 and k 0:
The oxygen content of metal nitrates can react
completely in oxidizing glycine equivalently.
3. OB40 (fuel-lean), when C 4.44 and k40: There is
an excess amount of oxygen in the reactant mixture
caused by the addition of NH4NO3.

383

economy and accuracy of the experiment, the OB value


was limited in the region of 15%+10%. At a desired
OB value, the starting materials Ni(NO3)2  6H2O,
Zn(NO3)2  6H2O, and Fe(NO3)3  9H2O (all X99.0%,
Merck, Germany) and glycine and NH4NO3 (both
X98.0%, Riedel-de Haen, Germany) were directly
mixed in a porcelain crucible without adding water,
resulting in a slurry substance due to the hygroscopicity
of nitrates. (It is our nding to omit the procedure of
dissolving the reactants in water to form a solution).
This reactant mixture was heated by using a hot-plate at

100 1C to a state of dehydration. In the meantime, the


reactant mixture was stirred vigorously by using a
magnetic agitator to homogenize, in that the dried
reactant mixture was ignited to start combustion
reaction by using mini gas burner in air at room
temperature. In Fig. 1, the combustion reaction of the
dried reactant mixture is illustrated in the case of
OB 0:
2.2. Characterization
Pt/Pt10%Rh thermocouple of diameter 0.127 mm
was used to measure the temperature variations during
combustion reaction and an alumina tube of 1.2 mm was
inserted into the reactant mixture to reduce the
measuring disturbance resulting from the violent
gas evolution occurred during reaction. Concurrently,
the signals from the thermocouple were stored and
processed using a data acquisition system (NOTEBOOK, Labtech, Wilmington, MA). Phase formation
of the product was identied by X-ray diffraction
(XRD; SIEMENS D5000) with CuKa radiation
(
(l 1:5418 A).
The morphological feature of the

Although there exist some literatures reporting that


nano-sized NiZn ferrites were synthesized by the
combustion method [4,8,9,11,1316], hardly is any
information available on the effects of reactant composition (i.e., the OB value), the nature of combustion
reaction, and the properties of as-synthesized powders.
In this work, the study of correlation among the OB
value, reaction phenomena and product characteristics
are undertaken.

2. Experimental
2.1. Sample synthesis
Ni0.5Zn0.5Fe2O4 powder was synthesized by an
amount of 25 g per batch. After considering both

Fig. 1. Combustion reaction of dried reactant mixture with OB 0:


Luminous ame and smoke gas arise as the dried reactant mixture is
ignited, causing the product left behind to be dry and uffy in nature.

ARTICLE IN PRESS
C.-C. Hwang et al. / Journal of Solid State Chemistry 178 (2005) 382389

384

product powder can be observed by a scanning


electronic microscope (SEM; Hitachi S-3000N) with an
accelerating voltage of 15 kV. The grain size and the
electron diffraction pattern of the product were imaged
by transmission electron microscope (TEM; Hitachi H7100). As far as TEM studies are concerned, assynthesized products were fully grounded and treated
with oscillation thoroughly, then the powders were
supported on carbon-coated copper TEM grids and
analyzed at an accelerating voltage of 120 kV. The BET
surface area measurement was made by nitrogen
absorption employing a Micromeritics ASAP 2010
instrument and then calculated using the ve point
BET theory. A Perkin-Elmer CHN elemental analyzer
(Model: 2400(II)) was employed to measure the content
of residual carbon. The magnetic measurements were
performed on a vibrating sample magnetometer (VSM;
Toei, VSM-5) at room temperature in an operating
range of 715 kOe.

3. Results and discussion


3.1. Thermodynamic analysis
To understand the variation of adiabatic ame
temperature (Tad) with respect to the OB value, the
following equation is used to calculate theoretically the
adiabatic Tad for a combustion reaction [6]:
T ad T 0

DH r  DH p
;
cp

(3)

where T 0 is 25 1C, DH r and DH p are the enthalpies of


formation of the reactants and products, respectively,
and cp is the heat capacity of the products at constant
pressure. Substituting the thermodynamic data from
Table 1 [9,17], Eqs. (1), and (2) into Eq. (3), the
variations of Tad together with the amount of gas

produced versus the OB value are shown in Fig. 2. As


can be seen, the theoretical Tad decreases substantially
with increasing OB value of the reactant mixture,
whereas the amount of produced gas is slightly
decreased with the increase of OB value up till OB 0
and begins to rapidly increase with further increase in
OB value.

3.2. Nature of combustion reaction


In this work, it was found that the nature of the
combustion reaction and the characteristics of assynthesized product are dependent on the OB value.
Typical temperature histories of the reactant mixtures
with OB 0; 8% and +5% during combustion
reaction are shown in Fig. 3. When OB 0 was used,
once the reactant mixture was ignited, the temperature
increased suddenly up to a maximum of
1250 1C and
then began to decrease (The maximum temperature
during combustion reaction is symbolized as T c ). In the
case of OB +5%, Tc is 995 1C. In addition, a
shoulder of round shape is observed in the temperature
prole between 240 and 280 1C, which could be caused
by the endothermic decomposition of NH4NO3. When
using OB 8%, the temperature increased slowly
and, after a period of time, it reached to a maximum of

700 1C and then began to decrease. It is interesting to


notice that the slope of the temperature versus time
curve for OB 0 is more abrupt than those of the other
two cases, signifying that its reaction rate is more rapid.
The plot of Tc as a function of OB value together with
its regimes having characteristic reaction is also shown
in Fig. 2. Tc is typically lower than the calculated value
of Tad due to heat loss [18] and/or incomplete reaction.
In addition, Tc increases with increasing OB value by
reaching to a maximum of
1300 1C (at OB 2%)
and then decreases with further increment in OB value.

Table 1
Entropy of formation and specic heat for the combustion synthesis of NiZn ferrite
Compound

DHf1 (kJ/mol)

Ni(NO3)2(c)
Zn(NO3)2(c)
Fe(NO3)3(aq)a
NH2CH2COOH(c)
NH4NO3(c)
Ni0.5Zn0.5Fe2O4(c)b
CO2(g)
H2O(g)
N2(g)
O2(g)

428
482
671
528
365
1041
395
243

cp (kJ/mol K)

Notes

OB 96%, Heat of combustion 973 kJ/mol


OB +20%, Heat of combustion 210 kJ/mol

0.148
0.061
0.051
0.024
0.039

The enthalpy of formation of Fe(NO3)3(aq) was used because that of Fe(NO3)3 (c) was not available.
Due to a lack of thermodynamic data for Ni0.5Zn0.5Fe2O4, the average enthalpy and specic heat of Fe3O4(c) (DH 0f 1117 kJ=mol;
cp 0:147 kJ=mol K) and CuFe2O4(c) (DH 0f 965 kJ=mol; cp 0:149 kJ=mol K) were used [9] because these compounds are isostructure and these
divalence metallic radii (Fe2+, Ni2+, Cu2+, and Zn2+) are nearly the same [19].
b

ARTICLE IN PRESS
C.-C. Hwang et al. / Journal of Solid State Chemistry 178 (2005) 382389

385

OB (%)
-16

-14

-12

-10

-8

-6

-4

-2

10

2800

Temperature (C)

2400

Un-ignitable

SCS

SHS

60

2000
50
1600
40
1200
30

Amount of produced gas (mole)

70

800
20
-16

-14

-12

-10

-8

-6

-4

-2

 = 6.92

6.44

6.00

5.66

5.33

5.00 4.73

 =0

10

OB (%)

4.44

4.44

4.44

0.66

2.21

4.18

4.44

4.44
12.55

6.76

 = 4.44
 =0

Fig. 2. Theoretical adiabatic ame temperature (Tad) and amount of produced gas, and maximum combustion temperature (Tc) achieved for
different reaction nature as a function of OB value. (&) Tad, (X) Tc, and (K) amount of produced gas (SCS smoldering combustion synthesis;
SHS self-propagating high-temperature synthesis).

Tc=1250C

1400

Temperature (C)

1200

1200

Tc=995C

1000

1000

(a) OB=0

Tc=700C

800

800

(b) OB=+5%
600

600

(c) OB=-8%
400

400

Shoulder

200

Temperature (C)

1400

200

0
0

10

15

20
0

10

15

0
20

10

15

20

25

Time (sec)

Fig. 3. Typical temperaturetime histories during combustion for dried reactant mixtures with (a) OB 0, (b) OB +5%, and (c) OB 8%:

Depending on the OB value, the reaction may


be classied into three different modes from our
observation:
1. Un-ignitable OBo10%.
2. Smoldering Combustion Synthesis (SCS), when
10%pOBo3%,
Tco950 1C,
and
when
OB4+4%, Tco1000 1C;
3. Self-propagating High-temperature Synthesis (SHS),
3%oOBo+4%, 1100 1CoTco1300 1C.
SCS mode is characterized by a slow, essentially
ameless reaction. When there is an excess fuel

10%pOBo  3%; the combustion reaction needs


oxygen to be supplied externally. Oxygen enters via
diffusion action into the reaction zone controlled by
kinetic factors that limit the reaction rate and, eventually, the SCS reaction mode. In an extreme case
OBo  10%; the combustion reaction may not be
ignited. The characteristic feature of the SHS mode is
that the reaction initiates locally and it propagates as a
combustion wave in a self-sustained manner through the
reaction volume. In this regime, it is to be emphasized
that the fuel-to-oxidant ratio is within a proper range
3%oOBo 4% and the oxygen content contained
in the reactant mixture is the main source of oxygen

ARTICLE IN PRESS
C.-C. Hwang et al. / Journal of Solid State Chemistry 178 (2005) 382389

386

required for combustion reaction. Meanwhile, oxygen


can react with glycine and oxidize/consume most of the
fuel as it originates from NO
3 , thus resulting in the
phenomenon of SHS reaction. With increased OB value
4 4%; since the fuel content is small, the heat
evolved is not enough and thus the temperature is lower.
This leads to slower reaction rates as manifested in the
smoldering combustion behavior.
3.3. Characterization of as-synthesized powders
In the following experiments, the OB values of 8%,
0 and +5% were selected to represent fuel-rich,
stoichiometric and fuel-lean compositions, respectively.
Fig. 4 shows the XRD patterns of the as-synthesized
powders prepared at the three different OB values. The
characteristic peaks of spinel phase were observed for all
cases. When OB 8% was used, the XRD pattern of
the as-synthesized product is different from the other
two cases due to its relatively low signal/noise ratio. In
addition, the diffraction peaks of Fe2O3 and some
unknown peaks were also observed, which delineates
that the as-synthesized powder consists of impurities.

(c) OB=-8%

There exists a notably broadening phenomenon in terms


of diffraction peaks of the products prepared using
OB 8% and +5%, which may be attributed to the
fact that the products crystallite are relatively ne.
The lattice constant a was determined by using the
classical formula:
l h2 k2 l 2 1=2
;
(4)
2
sin y
where l is the wavelength of CuKa, (hkl) are the Miller
indices, and y is the diffraction angle corresponding to
the (hkl) plane. The a values of the as-synthesized
ferrites are 8.3453, 8.3782, and 8.4130 A for OB 8%,
0 and +5%, respectively. Though these values are
different from the value of a 8.38 A for Ni0.5Zn0.5
Fe2O4 in Ref. [20]. All of them fall in between the lattice
constant (a 8.3393 A) of NiFe2O4 [21] and the lattice
constant (a 8.4411 A) of ZnFe2O4 [22]. Judging from
the Ref. [23], it can be inferred that the spinel phase is
NiZn ferrites.
Table 2 presents the characteristics of the assynthesized powder prepared at the three different OB
values. The grain size of the as-synthesized products was
a

Table 2
Effect of OB value on characteristics of as-synthesized powders

spinel ferrite
Fe2O3

OB (%)

Crystallite
sizea (nm)

Surface area
(m2/g)

Carbon content
(wt%)

8 (fuel-rich)
0 (stoichiometric)
+5 (fuel-lean)

27.8
34.5
25.3

39.1
30.2
44.5

9.25
1.53
2.15

Relative Intensity (a. u.)

a
Crystallite size of the as-synthesized Ni0.5Zn0.5Fe2O4 powders
calculated from the line broadening of the (311) XRD peak by
Scherrer formula.

30

35

40

45
50
2 (Degree)

(440)

(422)

(222)

(400)

(a) OB=0%

(220)

25

(333)/(511)

(311)

(b) OB=+5%

55

60

65

Fig. 4. XRD patterns of as-synthesized product made with (a)


OB 0, (b) OB +5%, and (c) OB 8%:

Fig. 5. Typical SEM micrograph of as-synthesized product made with


OB 0.

ARTICLE IN PRESS
C.-C. Hwang et al. / Journal of Solid State Chemistry 178 (2005) 382389

estimated according to the Scherrer formula [24]. It was


found that all the products obtained in this work are
nanocrystallites with the sizes ranging between 25 and
35 nm, which are smaller than the critical size of monomagnetic-domain (40 nm for NiZn ferrite [25]) and thus
these as-synthesized samples may display a superparamagneic behavior. Also, their surface areas are large
(
3045 m2/g), and the as-synthesized powder obtained
using OB +5% has the largest specic surface area as
compared with that of using OB 0 and 8%. As more
gases are evolved during combustion, the reaction heat
could be carried away from the system by convection to
hinder grain growth, which may produce powder with a
high specic surface area. The carbon contents of the assynthesized product with OB 8% is relatively high
as compared with that of using OB 0 and +5%. This

387

result coupled with its XRD spectrum with impurity


contained (Fig. 2(a)) is attributed to the lesser amount of
oxygen available for combustion. Hence, the local
temperature of reaction zone remains low (
700 1C),
causing combustion reaction to be incomplete.
Fig. 5 shows the typical SEM photograph of the assynthesized product prepared with OB 0. A continuous network of powders is formed. Voids and holes can
be seen, which result from the escaping of gases during
combustion. By SEM observation, it was found that as
the OB value was increased, the porosity of the resultant
powders also increased due to the evolution of moreproduced amounts of gas. The morphology of assynthesized powder prepared by using OB 0 and
+5% with their corresponding electron diffraction
patterns are shown in Fig. 6. The size of crystallites

Fig. 6. TEM images of as-synthesized powders with corresponding diffraction patterns for (a) OB 0 and (b) OB 5% (Taken under the same
operation conditions for both specimens).

ARTICLE IN PRESS
C.-C. Hwang et al. / Journal of Solid State Chemistry 178 (2005) 382389

388

were estimated to be
2025 nm for the OB +5%
product and
3540 nm for OB 0 one, in terms of
which they are about the same sizes as the ones
estimated by using XRD method. In addition, diffraction patterns indicate that the OB 0 product has a
higher degree of crystallinity than do the OB +5%
one. (The TEM photograph of the as-synthesized
product with OB 8% is similar to the one using
OB +5%, but not shown here).
Fig. 7 shows the real trace of the hysteresis loops of
the as-synthesized powders (crystallite size: o40 nm)
and the sintered sample (grain size:
1 2 mm). The
hysteresis loops of the as-synthesized ferrites were
characterized to have quite low magnetization values
and they do not reach saturation value even at 15 kOe,
indicating their superparamagnetic nature. However, on
sintering the ferrite at 950 1C/2 h, saturation magnetization was attained (
72 emu/g) at 2.5 kOe. This result is
obviously caused by the dependence of grain size on Ms,
which can be explained through the transition from
mono- to multi-magnetic-domain behavior [25]. As
shown in Figs. 7(b) and (c), the maximum magnetization
attained for powders synthesized with different OB
value was in the order of OB 04OB +5%4OB
8%, whereas the order of coercive force was just
opposite. It is suggested that impurity content or poor
crystallization may affect the magnetic behavior of the
as-synthesized ferrite by reducing the maximum magnetization and increasing the coercive force. (It is worth
noting that the soft-magnetic properties of the sintered
sample are comparable with those obtained by other
80
60

20

4. Conclusions
1. A combustion synthesis method has been used to
prepare nanocrystalline Ni0.5Zn0.5Fe2O4 containing a
considerably large surface area. Utilization of glycine,
metal (Ni, Zn, and Fe) nitrates, ammonium nitrate
joined with mixing the reactants without adding
water and combining a step of thorough dehydration
are the key techniques of this method. Once ignited in
air at room temperature, the reactant mixtures
undergo a combustion process and directly transform
into porous ferrite powders.
2. The combustion reaction is between nitrates and
glycine, in that NO
3 is oxidant and the organic group
is fuel. The reaction mode and the combustion
temperature can be controlled by changing the
oxygen balance (OB) values of the reactant mixtures.
Thus, the properties (such as crystallite size, surface
area, carbon content, etc.) of as-synthesized product
can vary based on OB value.
3. The maximum magnetization and coercive force
of the resulting ferrites show a strong dependence
on OB value. In addition, the as-synthesized powders

(b)
6

0
0

12

16

Magnetic Field (kOe)

5.0

Magnetization (emu/g)

Magnetization (emu/g)

40

Magnetization (emu/g)

(a)

methods [2628].) The detailed insights of the reaction


mechanism (by means of thermal analysis technique)
and the as-synthesized powders characteristics (such as
sinterability, electromagnetic properties), however, require further investigation, which is currently being
undertaken.

-20
-40
-60

(c)
2.5

0.0

-2.5

-5.0
-1.0

-0.5

0.0

0.5

1.0

Magnetic Field (kOe)

-80
-15

-10

-5

0
5
Magnetic Field (kOe)

10

15

Fig. 7. (a) Room temperature hysteresis loops of as-synthesized powders made with various OB values and the sintered sample obtained by using the
product of OB 0: (b) A zoom plot showing the rst quadrant of Fig. 7(a). (c) A zoom plot showing the vicinity of the origin in Fig. 7(a). () assynthesized powder for OB 0; (n) as-synthesized powder for OB 5%, (J) as-synthesized powder for OB 8%; and (.) sintered sample.

ARTICLE IN PRESS
C.-C. Hwang et al. / Journal of Solid State Chemistry 178 (2005) 382389

contain a superparamagnetic nature due to their lowranking and unsaturated magnetization value as
compared with those of sintered sample.

Acknowledgment
Support for this research by the National Science
Council of the Republic of China under Grant No. NSC
93-2214-E-014-001 is gratefully acknowledged.
References
[1] L.A. Chick, L.R. Pederson, G.D. Maupin, J.L. Bates, L.E.
Thomas, G.J. Exarhos, Mater. Lett. 10 (1990) 612.
[2] K. Suresh, K.C. Patil, J. Solid State Chem. 99 (1992) 1217.
[3] J.J. Kingsley, L.R. Pederson, Mater. Res. Soc. Symp. Proc. 296
(1993) 361366.
[4] L.E. Shea, J. McKittrick, O.A. Lopez, E. Sluzky, J. Am. Ceram.
Soc. 79 (1996) 32573265.
[5] C.H. Yan, Z.G. Xu, F.X. Cheng, Z.M. Wang, L.D. Sun,
C.S. Liao, J.T. Jia, Solid State Commun. 111 (1999) 287291.
[6] R.D. Purohit, B.P. Sharma, K.T. Pillai, A.K. Tyagi, Mater. Res.
Bull. 36 (2001) 27112721.
[7] Y.P. Fu, C.H. Lin, J. Magn. Magn. Mater. 251 (2002) 7479.
[8] R. Kalai Selvan, C.O. Augustin, L. John Berchmans, R.
Saraswathi, Mater. Res. Bull. 38 (2003) 4154.
[9] R.V. Mangalaraja, S. Ananthakmar, P. Manohar, F.D. Gnanam,
M. Awano, Mater. Sci. Eng. A 367 (2004) 301305.
[10] T.Y. Tsay, K.S. Liu, I.N. Lin, J. Magn. Magn. Mater. 209 (2000)
189192.

389

[11] P.S.A. Kumar, J.J. Shrotri, S.D. Kulkarni, C.E. Deshpande,


S.K. Date, Mater. Lett. 27 (1996) 293296.
[12] P.W. Cooper, S.R. Kurowski, Technology of Explosives, Wiley,
New York, NY, 1996 (pp. 57).
[13] C. Caizer, Mater. Sci. Eng. B 100 (2003) 6368.
[14] A.F.C.M. Costa, A.P.A. Diniz, L. Gama, M.R. Morelli,
R.H.G.A. Kiminami, J. Metastable Nanocryst. Mater. 2021
(2003) 582587.
[15] C.C. Hwang, C.Y. Wu, J. Wan, J.S. Tsai, Mater. Sci. Eng. B 111
(2004) 4956.
[16] R.V. Mangalaraja, S. Ananthakmar, P. Manohar, F.D. Gnanam,
M. Awano, Mater. Lett. 58 (2004) 15931596.
[17] J.A. Dean, in: Langes Handbook of Chemistry, 15th ed,
McGraw-Hill, New York, 1998.
[18] Z.A. Munir, Am. Ceram. Soc. Bull. 67 (1988) 342349.
[19] Y.M. Chiang, D.P. Brinie III, W.D. Kingery, Physical Ceramics:
Principles for Ceramic Science and Engineering, Wiley, New
York, NY, 1997, p. 16.
[20] P.C. Fannin, S.W. Charles, J.L. Dormann, J. Magn. Magn.
Mater. 201 (1999) 98101.
[21] Joint Committee on Powder Diffraction Standards (JCPDS)
le number: 44-1485.
[22] Joint Committee on Powder Diffraction Standards (JCPDS) le
number: 22-1012.
[23] E.E. Sielo, R. Rotelo, S.E. Jacobo, Physica B 320 (2002) 257260.
[24] H.P. Klug, L.E. Alexander, X-ray Diffraction Procedures for
Polycrystalline and Amorphous Materials, Wiley, New York,
1997 (p. 637).
[25] A.S. Albuquerque, J.D. Ardisson, W.A.A. Macedo, J. Appl.
Phys. 87 (2000) 43524357.
[26] Y. Hayashi, T. Kimura, T. Yamaguchi, J. Mater. Sci. 21 (1986)
28762880.
[27] P.S.A. Kumar, J.J. Shrotri, C.E. Deshpande, S.K. Date, J. Appl.
Phys. 81 (1997) 47884790.
[28] Y. Li, J. Zhao, J. Han, Mater. Res. Bull. 37 (2002) 583592.

ARTICLE IN PRESS
Physica B 404 (2009) 39153921

Contents lists available at ScienceDirect

Physica B
journal homepage: www.elsevier.com/locate/physb

Effect of Zn2+ substitution on the magnetic properties of Mg1xZnxFe2O4


ferrites
M. Manjurul Haque a,, M. Huq b, M.A. Hakim c
a

Department of Applied Physics, Electronics and Communication Engineering, Islamic University, Kushtia 7003, Bangladesh
Department of Physics, Bangladesh University of Engineering and Technology, Dhaka 1000, Bangladesh
c
Materials Science Division, Atomic Energy Centre, Ramna, Dhaka 1000, Bangladesh
b

a r t i c l e in fo

abstract

Article history:
Received 7 December 2008
Accepted 6 July 2009

Polycrystalline Mg1xZnxFe2O4 (x 0.00.6) ferrites have been prepared using solid-state reaction
technique. The X-ray diffraction analysis revealed that the samples crystallize in a single-phase cubic
spinel structure. The lattice parameter increases linearly with increase in zinc content obeying Vegards
law. The continuous decrease in Curie temperature (Tc) with an increase in Zn content is attributed to
the weakening of AB exchange interaction. Saturation magnetization (Ms) and magnetic moment are
observed to increase up to x 0.4, and thereafter decrease due to the spin canting in B-sites. The initial
permeability is found to increase with the addition of Zn2+ ions but the resonance frequency shifts
towards the lower frequency.
& 2009 Elsevier B.V. All rights reserved.

PACS:
75.50.Gg
Keywords:
MgZn ferrites
Complex permeability
Spin canting

1. Introduction
MgFe2O4 is a partially inverse cubic spinel and it can be
considered as a collinear ferrimagnet whose degree of inversion is
sensitive to the sample preparation history. Magnetic properties
of ferrites strongly depend on their chemical compositions and
additives/substitutions. Small amount of foreign ions in the ferrite
can dramatically change the properties of ferrites. Therefore, most
of the latest ferrite products contain a small amount of additives/
substitutions. Nonmagnetic Zn is a very promising and interesting
candidate among them. The magnetic properties of Zn-substituted
ferrites have attracted considerable attention because of the
importance of these materials for high-frequency applications.
Zinc ferrite (ZnFe2O4) possesses a normal spinel structure, i.e.,
2+
2
ions reside on A-sites and Fe3+
(Zn2+)A [Fe3+
2 ]BO4 , where all Zn
ions on B-sites. Therefore, substitution of Mg by Zn in
Mg1xZnxFe2O4 is expected to increase the magnetic moment up
to a certain limit, thereafter it decreases for the canting of spins in
B-sites. The structural and magnetic characteristics of MgFe2O4
with nonmagnetic substances such as Zn2+ [1], Ca2+ [2,3], Ti4+ [4],
Zr4+ [5] and Al3+ [6] have been investigated. Zn- and Cdsubstituted mixed ferrites such as FeZn [7], CuZn [8], NiCd
[9], MgCd [10] and MgZn [11] show canted-spin arrangement
on the octahedral B-sites. In the present study, a series of
Mg1xZnxFe2O4 (x 0.0, 0.1, 0.2, 0.3, 0.4, 0.5 and 0.6) were
 Corresponding author. Tel.: +88 01715 157526; fax: +88 07154400.

E-mail address: manju_iu@yahoo.com (M. Manjurul Haque).


0921-4526/$ - see front matter & 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.physb.2009.07.124

prepared using the solid-state reaction method and their


magnetic properties were investigated.

2. Experimental
The MgZn ferrites of the composition Mg1xZnxFe2O4 (x 0.0,
0.1, 0.2, 0.3, 0.4, 0.5 and 0.6) have been prepared using the solidstate reaction technique at Materials Science Division, Atomic
Energy Centre, Dhaka, Bangladesh. The starting materials were in
the form of AR-grade oxides (Fe2O3, MgO and ZnO) of E. Merck of
Germany. The detailed procedure of sample preparation is
described elsewhere [12,13]. The pellet and toroid-shaped samples were sintered at 1275 1C for 12 h in air and cooled in the
furnace. All the samples were heated slowly in the Thermolyne
high-temperature furnace at the rate of 1 1C/min and kept at the
ring temperature for 12 h. The cooling was done in the furnace at
the same rate of heating. The samples were polished in order to
remove any oxide layer formed during the process of sintering.
The weight and dimensions of the pellets were measured to
determine bulk densities.
Phase analysis was done by X-ray diffraction using a Phillips
(PW3040) X0 Pert PRO X-ray diffractometer. The powder specimens were exposed to CuKa radiation with a primary beam power
of 40 kV and 30 mA with a sampling pitch of 0.021 and time for
each step data collection was 1.0 s. X-ray diffraction patterns
conrmed the spinel phase for all the samples. Frequency
dependence of complex permeability of the toroid-shaped

ARTICLE IN PRESS
3916

M. Manjurul Haque et al. / Physica B 404 (2009) 39153921

samples was measured with the Agilent precision impedance


analyzer (Agilent 4294A). For these measurements an applied
voltage of 5 mV was used with a 5 turn low-inductive coil. Initial
permeability as a function of temperature at 100 kHz was carried
out using a Hewlett Packart impedance analyzer (HP 4291A) in
conjunction with a laboratory-made furnace, which maintains the
desired temperature with the help of a temperature controller.
The Curie temperature Tc was determined from the temperature
dependence of permeability curves. Field dependence of magnetization at room temperature was measured using a vibrating
sample magnetometer (VSM 02, Hirstlab, England).

ZnMg ferrite system have been reported by Mazen et al. [18],


Joshi et al. [19] and Ladgaonkar et al. [20].
Fig. 3 shows the variation of density as a function of Zn
content. The bulk density, dB, was measured by usual mass and
dimensional consideration whereas X-ray density, dX, was
calculated for each sample using the expression dX ZM/Na3
[16], where M is the molecular weight of the corresponding
composition, N the Avogadro number, V a3 the volume of the
cubic unit cell and Z the number of molecules per unit cell, which

8.43

3. Result and discussion


3.1. XRD analysis

8.42

8.40

8.39

8.38

0.1

0.0

0.3
0.4
Zn content (x)

Mg1-xZnxFe2O4

(440)

(511)

(422)

x = 0.6

(400)

(222)

(220)
(620)

(440)

(511)

(422)

(400)

(311)

(311)

0.2

x = 0.2

x = 0.1

x = 0.4

x = 0.0

20

30

40

50

60

70

0.5

0.6

Fig. 2. Variation of lattice parameter (a) as a function of zinc content (x).

Mg1-xZnxFe2O4

(222)

(220)

Intensity (a.u)

8.41

a ()

Fig. 1 shows the X-ray diffraction (XRD) patterns of the samples


Mg1xZnxFe2O4. The XRD patterns exhibited that all the samples
were identied as a single phase of cubic spinel structure with no
extra lines corresponding to any other crystallographic phase or
unreacted ingredient. The peaks (2 2 0), (3 11), (2 2 2), (4 0 0),
(4 2 2), (5 11) and (4 4 0) correspond to the spinel phase. The
lattice parameter was determined using the NelsonRiley
extrapolation method. Values of the lattice parameter obtained
from each reected plane were plotted against NelsonRiley
function [14] Fy 12cos2 y=sin y cos2 y=y, where y is
Braggs angle and straight lines were obtained. Values of the
lattice parameter were estimated by extrapolating these lines to
F(y) 0 or y 901. Variation of the lattice parameter a as a
function of Zn content is presented in Fig. 2. An increase in lattice
parameter is observed with increase in zinc content (x) in the
lattice. Increase in the lattice parameter with zinc content
indicates that the present system obeys Vegards law [15]. This
increase can be attributed to the ionic size differences since the
unit cell has to expand when substituted by ions with large ionic
[16] than Mg2+ ions
size. Zn2+ ions have larger ionic radius (0.82 A)

(0.66 A) [17], which when substituted reside on A-site and


displaces small Fe3+ ions from A-site to B-site. Similar results for

x = 0.3

80 20

30

40

50
2

2
Fig. 1. XRD patterns of Mg1xZnxFe2O4 ferrites.

60

70

ARTICLE IN PRESS
M. Manjurul Haque et al. / Physica B 404 (2009) 39153921

3917

400
5.0

dB
dX

x = 0.0

350

x = 0.1
x = 0.2

300

4.8

x = 0.3

x = 0.5
x = 0.6

'

Density (g-cm-3)

x = 0.4

250

4.6

200

150

4.4

100

50
4.2
0.0

0.1

0.2

0.3

0.4

0.5

0.6

Zn content (x)
Fig. 3. Variation of density with Zn content (x) of Mg1xZnxFe2O4 ferrites.

200

300

400

500

T (C)
Fig. 4. Temperature dependence of real permeability of Mg1xZnxFe2O4 ferrites.

is 8 for the spinel cubic structure. The X-ray density and bulk
density increase signicantly with increase in Zn content. The
atomic weight and density of Mg are 24.31 and 1.74 g cm3,
respectively, while the corresponding values of Zn are 65.37 and
7.14 g cm3, respectively. The bulk density (dB) is less than the
X-ray density (dX). This may be due to the existence of pores,
which were formed and developed during sample preparation or
sintering process.

500

3.2. Temperature dependence of initial permeability

300

y = -667.86x + 473.64

400

Tc (C)

Fig. 4 shows the thermal variation of initial permeability (m0 )


for the toroid-shaped ferrite samples. Since the initial
permeability is directly related to the magnetization and to the
ionic structure, the thermal spectra of permeability can be taken
as a test of the formation and homogeneity of the ionic structure
of the samples. Generally, it is found that the initial permeability
increases with increase in temperature, while it falls abruptly
close to the Curie point. The sharpness of the permeability drop at
the Curie point can be used as a measure of the degree of
compositional homogeneity according to Globus et al. [21]. The
present ferrites show good homogeneity as shown in Fig. 4, where
an abrupt drop in permeability occurs near the Curie point. The
variation of m0 with temperature can be expressed as follows The
anisotropy constant (K1) and saturation magnetization (MS)
usually decreases with increase in temperature, which disturbs
the alignment of magnetic moments. But, decrease in K1 with
temperature is faster than that of MS. When the anisotropy
constant reaches zero, m0 attains its maximum value according to
the relation m0 pMsD/OK1 (where D is the average grain diameter)
and then drops off to minimum value near the Curie point. The
m0 T plot does not show any secondary maximum that occur due
to excess formation of Fe2+ ions, thereby indicating the singlephase formation of the samples. This observation is supported by

100

200

100

0
0.0

0.1

0.2

0.3
0.4
Zn content (x)

0.5

0.6

0.7

Fig. 5. Variation of Curie temperatures (Tc) with Zn content (x) of Mg1xZnxFe2O4


ferrites.

XRD patterns, which do not show any impurity peaks. It is clearly


seen that the curves show maximum m0 just below Tc and steep
decrease in m0 is followed as the temperature is increased beyond
Tc. m0 increases with increase in zinc content but the plateau in the

ARTICLE IN PRESS
M. Manjurul Haque et al. / Physica B 404 (2009) 39153921

80

x = 0.2

60

40

x = 0.1
'

20

0.2

x = 0.0

x=

0.1

x=

m0 T curves narrows. Similar phenomena were observed by


Reslescu et al. [22] for copper-substituted MgZn ferrite.
Fig. 5 represents the variation of Curie temperature Tc as a
function of Zn content of Mg1xZnxFe2O4 ferrites. Tc is the
transition temperature above which the ferrite material loses its
magnetic properties. It gives an idea of the amount of energy
required to break up the long-range ordering in the material. Tc of
the studied ferrite system was determined from the m0 T curves
where the Hopkinson type of effect at the Tc was observed with
the manifestation of sharp fall of permeability. The Curie
temperature mainly depends on the strength of AB exchange
interaction. As the AB exchange interactions increase with
density and magnetic moment of the magnetic ions, greater
amount of thermal energy is required to offset the effects of
exchange interaction. It is observed from Fig. 5 and Table 1 that Tc
decreases continuously with increase in Zn content. It is wellknown that whenever nonmagnetic ions are introduced in the
ferrite sublattice, the Curie temperature decreases. Zn-substituted
mixed spinel ferrites are examples of this [16]. Zn substitution
only weakens the AB exchange interaction effectively. Owing to
this reduced AB exchange interaction with increase in Zn2+
content, the Curie temperature is expected to drop. The addition
of nonmagnetic Zn2+ ions replaced the magnetic Fe3+ ions at the
A-sites; thus the number of Fe3+ ions decrease at the A-sites. This
tends to decrease the strength of AB exchange interactions of the
2
Fe3+
type Fe3+
A O
B . This decreases the number of bonds or
linkages between the magnetic ions that determine the
magnitude of the Curie temperature [23]. The sample with
x 0.6 shows the lowest Tc and beyond this limit the samples
become paramagnetic at room temperature because of weak
interaction. A similar trend in the variation of Tc has been
observed by some workers in Zn-substituted Mg-ferrites [1,18,24].
A linear tting of the Curie temperature with Zn content (x) is
given by an empirical relation for the samples as

', ''

3918

''

0.1

10

0.

100

f (MHz)
350

Tc x Tc 0  667:86x

300
x = 0.6

where Tc(0) is the Curie temperature of pure Mg-ferrite and Tc(x)


corresponds to the Curie temperature of any composition having
Zn concentration (x). From this empirical relation Curie temperature of pure Mg-ferrite is found to be 473 1C. Our experimental
value of the Mg-ferrite is 460 1C, which is in good agreement with
the literature value of 440 1C [16].

250

x = 0.5

200
', ''

3.3. Frequency dependence of complex permeability


The complex permeability spectrum is shown in Fig. 6. The
complex permeability is given by m* m0 m00 , where m0 and m00 are
the real and imaginary parts, respectively, of initial permeability.
The real part m0 describes the stored energy expressing the
component of magnetic induction B in phase with the alternating
magnetic eld H. The imaginary permeability m00 describes the
dissipation of energy expressing the component of B 901 out of

150

Table 1
Data of the lattice parameter (a), bulk density (dB), X-ray density (dX), Curie
temperature (Tc), initial permeability (m0 ) and resonance frequency (fr) of
Mg1xZnxFe2O4 ferrites sintered at 1275 1C.

50

x = 0.4

'

x = 0.3

100

0.6

Zn Content (x) a (A)

dB (g/cm3) dX (g/cm3) Tc (1C) m0 (100 kHz) fr (MHz)

0.0
0.1
0.2
0.3
0.4
0.5
0.6

4.253
4.378
4.484
4.531
4.625
4.665
4.755

8.384
8.391
8.396
8.401
8.411
8.417
8.420

4.508
4.588
4.676
4.759
4.826
4.911
4.996

460
399
356
285
222
149
49

24
37
68
122
164
224
335

4100
E100
56.234
31.623
14.125
10.000
7.499

0.

''

0.

0.

x=

0
0.1

10

100

f (MHz)
Fig. 6. Complex permeability spectra for Mg1xZnxFe2O4 ferrite system sintered at
1275 1C.

ARTICLE IN PRESS
M. Manjurul Haque et al. / Physica B 404 (2009) 39153921

phase with the alternating magnetic eld H. It is observed from


Fig. 6 that the real component of permeability m0 is fairly constant
with frequency up to certain low frequencies, and rises slightly to
reach a maximum and then falls rather rapidly to very low value
at a high frequency. The imaginary component m00 rst rises slowly
and then increases quite abruptly making a peak at a certain
frequency (called resonance frequency, fr), where the real
component m0 falls sharply. This phenomenon is attributed to
ferrimagnetic resonance [25]. The resonance frequency peaks are
the results of the absorption of energy due to matching of the
oscillation frequency of the magnetic dipoles and the applied
frequency. Resonance frequency (fr) was determined from the
maximum of imaginary permeability of the samples. It is
observed from Fig. 6 and Table 1 that the higher the
permeability of the material, the lower the frequency of the
onset of ferrimagnetic resonance. This is in conformity with
Snoeks limit, in which Snoek found a relation between the
resonance frequency and the initial permeability as
m0 fr constant [26]. This means there is an effective limit to the
product of resonance frequency and permeability. Since the onset
of resonance frequency determines the upper limit of the
operational frequency of any device, it infers that the
operational frequency range of the samples is greater than
10 MHz except for the sample with x 0.6. The resonance
frequencies along with permeabilities of the samples are listed
in Table 1. The initial permeability m0 increases with increase in Zn
content, which is closely correlated to the densication of the
samples with zinc content. An increase in the density of ferrites
not only results in the reduction of demagnetizing eld, due to the
presence of pores, but also increases the spin rotational
contribution, which in turn increases the permeability [27].

6000

3.4. Frequency dependence of relative quality factor (RQF)


Fig. 7 shows the frequency dependence of relative quality
factor (RQF) of the samples sintered at 1275 1C. Q-factor increases
with increase in frequency showing a peak, and then decreases
with further increase in frequency. It is seen that RQF deteriorates
beyond 10 MHz, i.e., the loss tangent is minimum up to 10 MHz
and then it rises rapidly. The loss is due to lag of domain-wall
motion with respect to the applied alternating magnetic eld and
is attributed to various domain defects [28], which include nonuniform and non-repetitive domain-wall motion, domain-wall
bowing, localized variation of ux density and nucleation and
annihilation of domain walls. The peak corresponding to maxima
in Q-factor shifts to lower frequency range as zinc content
increases. The sample with x 0.3 possesses the maximum
value of RQF.
3.5. Magnetization study
Fig. 8 shows the variation of saturation magnetization (Ms) of
Mg1xZnxFe2O4 ferrites. Using the values of saturation
magnetization obtained from the MH curve, the magnetic
moment per formula unit, nB (in Bohr magneton), has been
calculated using the relation [29]
nB

Molecular weight  Ms
mB N

where M is the molecular weight of the ferrite sample, ss the


saturation magnetization, N the Avogadro number and mB the
Bohr magneton. The values of nB for each of the samples measured
at room temperature (300 K) are given in Table 2 and the variation
of nB with Zn2+ content is shown in Fig. 9. It is observed that Ms
and nB are found to increase up to x 0.4 and then decrease. The
initial increase with increased zinc content is due to the increase
in resultant sublattice magnetic moment, which can be explained

x = 0.3

5000

3919

70

x = 0.4

60
x = 0.5

50

x = 0.2

3000

2000

Ms (emu/g)

Realative quality factor

4000

x = 0.6

40
Mg1-xZnxFe2O4

x = 0.1

30
1000
x=0.0

20
0

1E-3

0.01

0.1

10

100

f (MHz)
Fig. 7. Frequency dependence of relative quality factor (RQF) of Mg1xZnxFe2O4
ferrites sintered at 1275 1C.

10
0.0

0.1

0.2

0.3
0.4
Zinc content (x)

0.5

0.6

Fig. 8. Saturation magnetization (Ms) as a function of zinc content (x).

ARTICLE IN PRESS
3920

M. Manjurul Haque et al. / Physica B 404 (2009) 39153921

Table 2
Data of saturation magnetization (Ms), theoretical and experimental magnetic
moment (nB) and YafetKittel angle (aYK).
Zinc content (x)

0.0
0.1
0.2
0.3
0.4
0.5
0.6

Ms (emu/g)

27.9
45
57.5
60.8
63.9
54.5
33.8

nB (mB)

aYK

Theoretical

Experimental

1.0
1.9
2.8
3.7
4.6
5.5
6.4

0.985
1.645
2.144
2.311
2.476
2.152
1.359

0
0
0
0
0
32.241
53.141

7
nB(Theoretical)
nB(Experimental)

consequently the magnetization of the B-sites increases. At the


same time the magnetization of A-site decreases according to
decrease in the Fe3+ ions on A-sites. So the net magnetization of
the samples increases up to x 0.4. Decrease in nB after x40.4
indicates the possibility of a non-collinear spin canting effect in
the system [3133]. According to the random canted model,
substitution of diamagnetic cations in one sublattice of
ferrimagnet leads to spin canting in the other sublattice
resulting in decrease in total magnetization per formula unit.
The reason for the decrease in magnetization beyond x 0.4 is
that the magnetization of A-sublattice is so diluted that the AB
exchange interaction no longer remains stronger and thereby BB
sublattice interaction becomes strong, which in turn disturbs the
parallel arrangement of spin magnetic moments on the B-site and
hence canting of spin occurs. Neels two-sublattice collinear
ferrimagnetism is observed for the system up to xr0.4 and
beyond this limit three-sublattice non-collinear spin canting
model is predominant. The existence of canted spin gives rise to
the YafetKittel angle (aYK), which compares the strength of AB
and BB exchange interactions [31]. YK angles are calculated at
300 K using the following formula:
nCB MB x cos aYK  MA x;

Magnetic moment ()

0.0

0.2

0.4

0.6

Fig. 9. Variation of magnetic moment as a function of Zn content (x) of


Mg1xZnxFe2O4 ferrites.

on the basis of Neels two sublattice models. Neel [30] considered


three types of exchange interactions between unpaired electrons
of two ions lying in A- and B-sites. AB interaction heavily
predominates over AA and BB interactions. The AB interaction
aligns all the magnetic spins at A-sites in one direction and those
at B-site in the opposite direction. The net moment of the lattice is
therefore the difference between the magnetic moments of B and
A sublattices, i.e., M MBMA. The magnetic moment of each
composition depends on the distribution of Fe3+ ions between
the two sublattices A and B, where Zn2+ is nonmagnetic. It is
known that the Mg-ferrite is a spinel with inversion degrees of
about d 0.9 [16], with the following cation distribution:
(Mgd2+Fe1d3+)A(Mg1d2+Fe1+d3+)BO2
4 . The magnetic moment of
Mg2+ is zero and the magnetic moment of Fe3+ is 5mB.
The replacement of x amount of Zn2+ ions (where Zn2+ ions
prefer to occupy A-sites) gives (1dx)Fe3+ ions on A-sites and
(1+d+x)Fe3+ ions on B-sites. The cation distribution then takes the
following form: (Mgd2+Fe1xd3+)A[Mg1xd2+Fe1+x+d3+]BO2
4 . The
substitution leads to increased Fe3+ ions on B-sites and

where aYK is the canted angle. The condition for YK angles to


occur in Zn-containing NiZn ferrites was investigated by Satya
Murthy et al. [34] in molecular eld approximation using the noncollinear three-sublattice model. According to the similarities
between NiZn, MgZn and MgCd ferrites, the existence of YK
angles in the present MgZn system for x40.4 is assumed. The
values of aYK are presented in Table 2. It is seen from Table 2 that
aYK angles for the samples x 0.0, 0.1, 0.2, 0.3, 0.4 are zero. This
suggests that Neels two-sublattice model holds well in these
ferrites. The calculated values of the YK angles for the present
system are non-zero for x40.4 and the non-zero YK angles
suggest that the magnetization behavior cannot be explained by
Neels two-sublattice model due to the presence of spin canting
on B-sites, which increases the BB interaction and consequently
decreases the AB interaction. Increase in YK angles for the
samples with Zn content (x40.4) is attributed to the increased
favour of triangular spin arrangements on B-sites leading to the
reduction in the AB exchange interaction and subsequent
decrease in magnetization [32,35]. Almost all Zn2+- and Cd2+substituted ferrites have shown a similar type of canting behavior
above a certain limit of their content [3537]. Hence in the
present system of ferrites, frustration and randomness increases
as Zn content increases in the Mg-ferrites and shows signicant
departure from Neels collinear model.

4. Conclusions
The X-ray diffraction conrmed the single-phase cubic spinel
structure of the samples. The lattice parameter increases with
increase in Zn content obeying Vegards law. Curie temperatures
show a decreasing trend with successive addition of zinc ions. This
is due to the fact that the replacement of Fe3+ ions by Zn2+ ions in
the A-sites results in decrease in the strength of AB super
exchange interaction. Saturation magnetization and magnetic
moment are found to increase with Zn content up to xZ0.4. This
is because Zn substitution leads to increased Fe3+ ions on the
B-sites and decreases Fe3+ ions on the A-sites. Decrease in the
magnetic moment after x40.4 indicates the possibility of a noncollinear spin canting effect in the system. An increase in initial
permeability has been observed with increase in zinc content but
the resonance frequency shifts towards the lower frequency.

ARTICLE IN PRESS
M. Manjurul Haque et al. / Physica B 404 (2009) 39153921

References
[1] H.H. Joshi, R.G. Kulkarni, J. Mater. Sci. 21 (1986) 2138.
[2] R.V. Upadhyay, R.G. Kulkarni, Mater. Res. Bull. 19 (1984) 653.
[3] S.D. Chhaya, M.P. Pandya, M.C. Chantbar, K.B. Modi, G.J. Baldha, H.H. Joshi, J.
Alloys Compd. 377 (2004) 155.
[4] R.A. Brand, H. Georges-Gilber, J. Hubseh, J.A. Hellor, J. Phys. F: Mater. Phys. 15
(1985) 1987.
[5] N.N. Jani, B.S. Trivedi, H.H. Joshi, R.G. Kulkarni, Indian J. Pure Appl. Phys. 35
(1997) 419.
[6] K.B. Modi, H.H. Joshi, R.G. Kulkarni, J. Mater. Sci. 31 (1996) 1311.
[7] C.M. Srivastava, S.N. Shringi, R.G. Srivastava, N.G. Nandikar, Phys. Rev. B 14
(1976) 2032.
[8] R.G. Kulkarni, V.U. Patil, J. Mater. Sci. 17 (1982) 843.
[9] V.G. Panicker, R.V. Upadhyay, S.N. Rao, R.G. Kulkarni, J. Mater. Sci. Lett. 3
(1984) 385.
[10] R.V. Upadhyay, S.N. Rao, R.G. Kulkarni, Mater. Lett. 3 (1985) 273.
[11] R.B. Pujar, S.S. Bellad, S.C. Watawe, B.K. Chougule, Mater. Chem. Phys. 57
(1999) 264.
[12] M. Manjurul Haque, M. Huq, M.A. Hakim, J. Magn. Magn. Mater. 320 (2008)
2792.
[13] M. Manjurul Haque, M. Huq, M.A. Hakim, J. Phys. D: Appl. Phys. 41 (2008)
055007.
[14] J.B. Nelson, D.P. Riley, Proc. Phys. Soc. London 57 (1945) 160.
[15] L. Vegard, Z. Phys. 5 (1921) 17.
[16] J. Smit, H.P.J. Wijn, Ferrites, Wiley, New York, 1959, p. 143.

3921

[17] L. John Berchamans, R. Kalai Selvan, P.N. Selva Kumar, C.O. Augustin, J. Magn.
Magn. Mater. 279 (2004) 103.
[18] S.A. Mazen, S.F. Mansour, H.M. Zaki, Cryst. Res. Technol. 38 (6) (2003) 471.
[19] H.H. Joshi, R.G. Kulkarni, R.V. Upadhyay, Indian J. Phys. A 65 (4) (1991) 310.
[20] B.P. Ladgaonkar, P.N. Vasambekar, A.S. Vaingankar, J. Magn. Magn. Mater. 210
(2000) 289.
[21] A. Globus, H. Pascard, V.J. Cagan, J. Physique 38 (1977) C1163.
[22] N. Reslescu, E. Reslescu, L. Reslescu, P.D. Popa, C. Paniscu, M.L. Craus, J. Magn.
Magn. Mater. 182 (1998) 199.
[23] R. Mitra, R.K. Pure, R.G. Mendiratta, J. Mater. Sci. 27 (1992) 1275.
[24] D. Ravinder, T.S. Rao, Cryst. Res. Technol. 25 (1998) 8.
[25] F.G. Brockman, P.H. Dowling, W.G. Steneck, Phys. Rev. 77 (1950) 85.
[26] J.L. Snoek, Physica 14 (4) (1948) 207.
[27] J.J. Shrotri, S.D. Kulkarni, C.E. Deshpande, S.K. Date, Mater. Chem. Phys. 59
(1999) 1.
[28] K. Overshott, IEEE Trans. Magn. 17 (1981) 2698.
[29] J. Smit, Magnetic Properties of Materials, Mcgraw Hill Book Co., 1971, p. 89.
[30] I. Neel, Ann. Phys. 3 (1948) 137.
[31] Y. Yafet, C. Kittel, Phys. Rev. 87 (1952) 290.
[32] S. Geller, Phys. Rev. 181 (1969) 980.
[33] C.E. Patton, Y. Liu, J. Phys. C: Solid State Phys. 16 (1983) 5995.
[34] N.S. Satya Murthy, M.G. Natera, S. Youssef, R.J. Begum, C.M. Srivastava, Phys.
Rev. 181 (1969) 4412.
[35] S.S. Bellad, B.K. Chougule, Mater. Chem. Phys. 52 (1998) 166.
[36] S.S. Bellad, S.C. Watawe, B.K. Chougule, J. Magn. Magn. Mater. 195 (1999) 57.
[37] S.S. Bellad, B.K. Chougule, Mater. Res. Bull. 33 (8) (1996) 1165.

Physica B 407 (2012) 11041107

Contents lists available at SciVerse ScienceDirect

Physica B
journal homepage: www.elsevier.com/locate/physb

Effect of zinc substitution on magnetic and electrical properties of


nanocrystalline nickel ferrite synthesized by reuxing method
A.I. Nandapure a, S.B. Kondawar b,n, P.S. Sawadh a, B.I. Nandapure b
a
b

B.D. College of Engineering, Sevagram, Wardha 442001, India


Department of Physics, R.T.M. Nagpur University, Nagpur 440033, India

a r t i c l e i n f o

a b s t r a c t

Article history:
Received 28 September 2011
Received in revised form
4 January 2012
Accepted 5 January 2012
Available online 17 January 2012

Nanocrystalline Nickel ferrite (NiFe2O4) and Zn substituted nickel ferrite (NiZnFe2O4) have been
synthesized by the reuxing method. These ferrites were characterized by XRD, TEM, Mossbauer
spectroscopy and VSM in order to study the effect of zinc substitution in nickel ferrite. XRD diffraction
results conrm the spinel structure for the prepared nanocrystalline ferrites with an average crystallite
size of 1416 nm. Lattice parameter was found to increase with the substitution of Zn2 ions from
TEM images conrmed average particle size of about 20 nm and indicates nanocrystal8.40 A to 8.42 A.
line nature of the compounds. A shift in isomeric deviation with the doublet was observed due to the
inuence of Zn substitution in the nickel ferrite. The Zn content has a signicant inuence on the
magnetic behavior and electrical conductivity of NiFe2O4. Saturation magnetization drastically
increased whereas room temperature electrical conductivity decreased due to the addition of Zn
content in NiFe2O4, indicating super magnetic material with lesser coercivity.
& 2012 Elsevier B.V. All rights reserved.

Keywords:
Nickel ferrite
Nickel zinc ferrite
Reux method
Mossbauer spectroscopy
VSM

1. Introduction
Recently, nanocrystalline magnetic materials have been
receiving more attention due to their novel material properties,
which are signicantly different from those of their bulk counterparts. Spinel-type oxides (MFe2O4, where M is a divalent metal)
which include the magnetic ferrites, are often denoted by the
formula AB2O4, where A and B refer to tetrahedral and octahedral
sites, respectively, in the face centered cubic lattice. These
materials are technologically important and have been used in
many applications including magnetic recording media and
magnetic uids for the storage and/or retrieval of information,
magnetic resonance imaging (MRI) enhancement, catalysis, magnetically guided drug delivery, sensors and pigments [1]. Among
all ferrites, NiFe2O4 has been synthesized because it has promising applications like ferro-uids, gas sensor, catalyst, microwave
absorber, information storage devices, magnetic uids, drug
delivery [2]. Nickel ferrite (NiFe2O4) is basically an inverse spinel
ferrite in which the tetrahedral (A) sites are occupied by ferric
ions and the octahedral (B) sites by ferric and nickel ions. Thus,
the compound can be represented as (Fe3 )A[Ni2 Fe3 ]BO2 4. The
magnetic properties and cation distribution are found to be
different in nanocrystalline spinel ferrites, when compared to

Corresponding author. Tel.: 91 0712 2042086; fax: 91 0712 2500736.


E-mail address: sbkondawar@yahoo.co.in (S.B. Kondawar).

0921-4526/$ - see front matter & 2012 Elsevier B.V. All rights reserved.
doi:10.1016/j.physb.2012.01.081

the bulk counterpart [3]. Various methods have been developed


to synthesize nanocrystalline NiFe2O4 including the reuxing
method [4], citrate precursor techniques [5], radio-frequency
thermal plasma torch [2], self-combustion method [6], solid state
reaction [7], solgel [8], precipitation route [9], solgel auto
combustion [10], reactive milling [11], doping with niobium ions
[12], thermal plasma [13], pulsed wire discharge [14] and hydrothermal route [15]. Wan and Qu prepared soft magnetic spinel
ferrites of the form A2 B32 O4 such as NiFe2O4 and ZnFe2O4
materials with very low coercive force (Hc 0) and relatively
high saturation magnetization (Ms 72 emu/g) are widely used in
microwave devices due to their high saturation magnetization,
high permeability, high electrical resistivity and low eddy current
losses [1618]. Doping of NiFe2O4 ferrite with non-magnetic Zn2
ion is a simple route to drastically enhance electrical and
magnetic properties of NiFe2O4. Zinc substituted ferrites are
technologically important materials because of their high magnetic permeability and low core losses. The investigations of
various characteristic properties of NiZn ferrites such as their
structure, electrical conductivity and dielectric behavior showed
that the grain size and grain boundary plays a major role in the
conduction mechanism in these materials [19]. Resulting Zn2
substituted NiFe2O4 ferrites have spinel conguration based on a
face centered cubic lattice of the oxygen ions with the unit cell
consisting of 8 formula units of the type (Zn2 Fe3 )A
[Ni2 Fe3 ]BO2 4. Metallic cations in ( ) occupy the tetrahedral
sites (A) and the metallic cations in [ ] occupy the octahedral sites

A.I. Nandapure et al. / Physica B 407 (2012) 11041107

1105

(B). Whatever the method of synthesizing the nano-sized ferrites,


it is very important to consider their stability in terms of their size
because the various properties begin to change signicantly as we
go to smaller size. Amongst all the methods used for synthesizing
the nanocrystalline ferrites, reuxing method is a cheap and lowtemperature technique that allows the ne control of the products chemical composition as well as densication and high
productivity is often achieved at lower temperature. Kim et al.
studied Ag-coated NiZn ferrite microspheres prepared by electroless plating [20], Sharma et al. [21] prepared NiZnFe2O4 of
average particle size  45 nm, using citrate precursor method and
studied the inuence of Zn substitution on dielectric properties of
nanocrystalline nickel ferrites only. In the present study, we
report the synthesis of NiFe2O4 and Zn2 substituted NiFe2O4
ferrites of average particle size 20 nm, their characterizations
and the effect of Zn2 substitution on electrical and magnetic
properties of the nickel ferrite.

2. Experimental
Ferric nitrate [Fe(NO3)3  9H2O], nickel nitrate [Ni(NO3)2  6H2O]
and zinc nitrate [Zn(NO3)2  6H2O] with the purity 99.5% were
purchased from Hi-media and used as received. NiFe2O4 and Zn2
substituted NiFe2O4 ferrites have been synthesized by simple
approach of reuxing method. In a typical procedure, specic molar
concentration of nickel nitrate and ferric nitrate as precursors were
mixed in a starch solution and stirred for half an hour for the
synthesis of NiFe2O4. Under reux condition, NaOH was added drop
by drop for 4 h to provide a net negative surface charge to the nuclei
limiting their further growth and aggregation. After reuxing, the
solution was kept overnight and then ltered and dried in hot air
oven at 80 1C for 12 h. Dried sample was treated at different
temperatures 800 1C, 900 1C, 1000 1C, 1100 1C in order to maintain
the stability of compound giving stoichiometric ferrite structure
according to the following reaction,
2Fe3 M2 8OH -Fe2 MOH8 -MFe2 O4 4H2 O:
The nucleation rate is quite high at the beginning of the precipitation process whereas the excess of OH  ions provides a net negative
surface charge to the nuclei avoiding aggregation. For synthesis of
Zn2 substituted NiFe2O4 ferrite, specic molar concentration of
nickel nitrate, zinc nitrate and ferric nitrate as precursor were mixed
in a starch solution and stirred for half an hour. Under reux
condition, NaOH was added drop by drop for 4 h. After reuxing,
the solution was kept overnight and then ltered and dried in hot air
oven at 80 1C for 12 h. Finally sample was treated at different
temperatures 800 1C, 900 1C, 1000 1C, 1100 1C resulting into the
Zn2 substituted nanocrystalline NiFe2O4. These ferrites have been
abbreviated as NF for NiFe2O4 and NZF for Zn2 substituted NiFe2O4,
respectively. Ferrites have been characterized by XRD (Philips PW
1730 automatic X-ray diffractometer with Cu-Ka radiation of

l 1.5428 A),
TEM (Hitachi H-7000 operated at 100 kV and
30 mA), Mossbauer spectroscopy (4.2300 K, 5 T) using a commercial oxford cryostat. Magnetic measurement using EG & G PAR model
4500 vibrating sample magnetometer and electrical resistivity measurement using four point probe technique was done to study the
effect of zinc substitution in nickel ferrite.

3. Results and discussion


3.1. X-ray diffraction
Fig. 1 shows the XRD pattern of (a) NiFe2O4 and (b) Zn2
substituted NiFe2O4 treated at 1100 1C in air. Most of the XRD

Fig. 1. XRD pattern of (a) NiFe2O4 and (b) Zn2 substituted NiFe2O4.

characteristic peaks of NiFe2O4 belong to the spinel ferrite. No


other separate phase oxides could be identied by XRD. All (h k l)
reections are in good agreement with JCPDS le no. 10-325.
Average crystallite size from the broadening of the (3 1 1) peak
using Scherrer equation [22] and lattice parameter using the
formula [23] of as-synthesized ferrite found to be 14 nm and
respectively. NiFe2O4 belongs to the class of ferrites with
8.40 A,
inverse spinel structure having structural formula Fe3 [Ni2
Fe3 ]O4. The metal ions given in square bracket are called
octahedral (sites B) ions and that of outside square bracket called
tetrahedral (sites A) ions. The nickel ions Ni2 together with half
of iron ions Fe3 occupy B site and remaining half of iron ions
reside in A site. XRD pattern of Zn2 substituted NiFe2O4 showing
pure spinel phase as all the peaks matched well with JCPDS le
no. 8-234. Average crystallite size from the broadening of the
(3 1 1) peak and lattice constant of NiZnFe2O4 were found to be
respectively. Zn2 substituted NiFe2O4 is a soft
16 nm and 8.42 A,
magnetic material having structural formula [Zn2 Fe3 ]A[Ni2
Fe3 ]BO4, where the subscript A and B denote tetrahedral sites
and octahedral sites in spinel AB2O4. Ni2 is stabilized in the
octahedral crystal eld whereas Zn2 preferred tetrahedral sites
because of its facility to form covalent bonds involving sp3 hybrid
orbital. The average crystallite size and lattice constant were
found to increase with the substitution of Zn2 ions. This increase
can be attributed to the substitution of the larger Zn2 ions for
the smaller Ni2 ions [24,25].
3.2. Transmission electron microscopy
The morphology and structure of NiFe2O4 and Zn2 substituted NiFe2O4 ferrites were investigated by TEM. It is clearly seen
from the TEM bright-eld images Figs. 2(a) and (b) that the
particles of the synthesized ferrites are mostly in the form of
square and paralelloids as well as their truncated forms. Large
number of small scattered grains with the strongest spotty
patterns as observed in TEM, indicating a highly crystalline spinel
structure [1] with the particle size about 20 nm for both the
ferrites. By comparing TEM images of both the ferrites, it can be
seen that the particle size and morphology of NiFe2O4 have been

1106

A.I. Nandapure et al. / Physica B 407 (2012) 11041107

Fig. 2. (a) TEM image of NiFe2O4. (b) TEM image of Zn2 substituted NiFe2O4.

affected by Zn2 substitution which also conrmed from


their XRD.

3.3. Mossbauer spectroscopy


Figs. 3(a) and (b) shows Mossbauer spectrum of NiFe2O4 and
Zn2 substituted NiFe2O4, respectively, taken at room temperature. The spectrum of NiFe2O4 consists of two sextets corresponding to magnetic scattering. The isomeric shift (d) of 0.36 mm/s
and 0.25 mm/s were congruent with the presence of Fe(III) in two
different states [26]. This ferrite displays two different interaction
of super-exchange between Fe3 OFe3 and Fe3 ONi2 in
NiFe2O4 lattice, where Fe3 ions occupy the center of the
octahedral oxygen and tetrahedral oxygen in equal proportion
while Ni2 ions occupy mainly octahedral sites [27].
Spectra of Zn2 substituted NiFe2O4 reveals the presence of
iron atoms in tetrahedral (A) and octahedral (B) sites. The value of
quadrupole splitting close to zero indicates the cubic symmetry
attributed to spinel phase. The values of isomer shift (d) between
0.28 to 0.37 mm/s are consistent with iron ions in trivalent
state [8]. Spectrum shows a presence of doublet with two sextets
indicating presence of Zn2 ions in NiZnFe2O4. Mossbauer spectra
of Zn2 substituted NiFe2O4 shows a central doublet superimposed on two sextets indicates the increased saturation magnetization towards super paramagnetic, which also supported
from magnetic behavior of the materials using VSM.

Fig. 3. (a) Mossbauer spectra of NiFe2O4. (b) Mossbauer spectra of Zn2 substituted NiFe2O4.

3.4. Magnetization measurements


Magnetization measurements of NiFe2O4 and Zn2 substituted
NiFe2O4 were performed using VSM technique. MH curves of
NiFe2O4 and Zn2 substituted NiFe2O4 ferrites at room temperature are shown in Fig. 4. It can be seen from the gure that the
saturation magnetization (Ms), coercive eld (Hc) and remenant
magnetization (Mr) for NiFe2O4 were found to be 36.77 emu/g,
120.48 Oe and 5.64 emu/g, respectively. In comparison with
Fe3O4, these ferrites have low coercive eld because of the nonmagnetic Ni2 ion presence in the lattice. The nickel ions are
located between the grains and increase the effect of the grains on
each other. The value of Ms in the crystals is attributed to the
greater fraction of surface spins in these crystals that tend to be in
a canted or a spin glass-like state with a smaller net moment [28].
Saturation magnetization (Ms), coercive eld (Hc) and remenant
magnetization (Mr) for Zn2 substituted NiFe2O4 were found
to be 62.23 emu/g, 23.92 Oe and 1.42 emu/g, respectively.
Saturation magnetization was found to be increased due to the
addition of Zn2 content. The increase in saturation magnetization in NiZnFe2O4 is due to preference of non-magnetic Zn2 ion
in A site encourage the migration of Fe3 ion into B site which
gives rise antiparallel spin coupling and spin canting resulting in
the weakening of AB exchange interaction [29]. The effect of
Zn2 substitution on magnetic parameters in NiFe2O4 as
decreased in Hc, Mr and increased in Ms is an indication of
NiZnFe2O4 ferrite towards super magnetic nature due to central

A.I. Nandapure et al. / Physica B 407 (2012) 11041107

1107

4. Conclusions
The synthesized NiFe2O4 and Zn2 substituted NiFe2O4 ferrites
are crystalline, having average particle size of 20 nm as revealed
by TEM. Lattice parameter was found to increase with the
substitution of Zn2 ions in NiFe2O4. VSM results show that
saturation magnetization of NiFe2O4 drastically increased due to
the addition of Zn2 content indicating super magnetic material
with lesser remanance as supported from its Mossbauer spectroscopy. The substitution of Zn2 in NiFe2O4 lowered the electrical
conductivity due to widening of band gap as a result of size
dependent property.

Acknowledgment
The authors are grateful for the nancial support of University
Grants Commission, New Delhi, India (F No.39-540/2010 SR).
Fig. 4. MH curves for (a) NiFe2O4 and (b) Zn2 substituted NiFe2O4.

Fig. 5. Electrical conductivity of (a) NiFe2O4 and (b) Zn2 substituted NiFe2O4.

doublet superimposed on two sextets as also seen in its Mossbauer


spectrum.
3.5. Electrical conductivity measurements
The DC electrical conductivity measurements were carried out
on pressed pellets of the synthesized ferrites using four probe
technique. The pellets were prepared in hydraulic press by
applying pressure of 5 T. The pellets had diameter of 1.0 cm and
thickness 0.005 cm. DC electrical conductivity was determined
from resistance measurement. Fig. 5 shows the variation of DC
electrical conductivity of NiFe2O4 and Zn2 substituted NiFe2O4
with temperature in the range from 303 to 383 K. The room
temperature conductivity of NiFe2O4 and Zn2 substituted
NiFe2O4 is almost same but difference of conductivity at higher
temperature was noticed. The substitution of Zn2 ions in
NiFe2O4 may increase the band gap of NiFe2O4 which affects the
conductivity and is in good agreement of increased in average
crystallite size as determined from XRD and conrmed from TEM.

References
[1] S. Maensiri, C. Masingboon, B. Boonchom, S. SEraphin, Scripta Mater. 56
(2007) 797.
[2] S. Son, M. Taheri, E. Carpenter, V.G. Harris, M.E. McHenry, J. Appl. Phys. 91
(2002) 7589.
[3] A.T. Raghavender, R.G. Kulkarni, K.M. Jadhav, J. Phys. 46 (2008) 366.
[4] Z. Zhong, Q. Li, Y. Zhang, H. Zhong, M. Cheng, Y. Zhang, Powder Technol. 155
(2005) 193.
[5] A. Verma, O.P. Thakur, C. Prakash, T.C. Goel, R.G. Mendiratta, Mater. Sci. Eng. B
116 (2005) 1.
[6] N. Iftimie, E. Rezlescu, P.D. Popa, N. Rezlescu, J. Optoelect. Adv. Mater. 8
(2006) 1016.
[7] T.K. Kundu, S. Mishra, Bull. Mater. Sci. 31 (2008) 507.
[8] M. Popovici, C. Savii, D. Niznanskya, J. Subrta, J. Bohaceka, D. Becherescub,
C. Caizerc, C. Enache, C. Ionescu, J. Optoelect. Adv. Mater. 5 (2003) 251.
[9] S.K. Date, P.A. Joy, P.S. Anilkumar, B. Sahoo, W. Keune, Phys. Stat. Sol. (c) 1
(2004) 3495.
[10] Z. Yue, J. Zhou, H. Zhang, Z. Gui, L. Li, Chi. J. Mater. Res. 13 (1999) 483.
[11] T.F. Marinca, I. Chicinas-, O. Isnard, V. Pop, F. Popa, J. Alloy Comp. 509 (2011)
7931.
[12] S. Mishra, N. Karak, T.K. Kundu, D. Das, N. Maity, D. Chakravorty, Mater. Lett.
60 (2006) 1111.
[13] A.B. Nawale, N.S. Kanhe, K.R. Patil, S.V. Bhoraskar, V.L. Mathe, A.K. Das,
J. Alloy. Comp. 509 (2011) 4404.
[14] Y. Kinemuchi, K. Ishizaka, H. Suematsu, W. Jiang, K. Yatsui, Thin Solid Films
407 (2002) 109.
[15] A. Baykal, N. Kasapoglu, H. Kavas, M.S. Toprak, Y. Koseoglu, Turk. J. Chem. 33
(2009) 33.
[16] M. Wan, J. Fan, Polym. Sci. part A: Polym. Chem. 36 (1998) 2749.
[17] Y. Qu, H. Yang, N. Yang, Y. Fan, H. Zhu, G. Zou, Mater. Lett. 60 (2006) 3548.
[18] M.P. Gonzalez-Sandoval, A.M. Beesley, M.M. Yoshida, L.F. Cobas,
J.A.M. Aquino, J. Alloy Comp. 369 (2004) 190.
[19] J.J. Thomas, S. Krishnan, N. Kaldrikkal, Procesdings ICNM-2009, Appl. Sci.
Inno. 2009.
[20] J.H. Kim, S.S. Kim, J. Alloy. Comp. 509 (2011) 4399.
[21] S. Sharma, K. Verma, U. Chaubeya, V. Singh, B.R. Mehta, Mater. Sci. Eng. B 167
(2010) 187.
[22] H.P. Klug, L.E. Alexander, X-ray Diffraction Procedure, Second ed., Wiley, New
York, 1974.
[23] C.C. Hwang, J.S. Tsai, T.H. Huang, Mater. Chem. Phys. 93 (2005) 330.
[24] S.M. Haque, M.A. Choudhury, M.F. Islam, J. Magn. Magn. Mater. 251 (2002)
292.
[25] K.B. Modi, P.V. Tanna, S.S. Laghate, H.H. Joshi, J. Mater. Sci. Lett. 19 (2000)
1111.
[26] Y. Sui, W.H. Su, F.L. Zheng, D.P. Xu, Mater. Sci. Eng. A 286 (2000) 115.
[27] M.A.F. Ramalho, L Gama, S.G Antonio, J. Mater. Sci. 42 (2007) 3603.
[28] M.S. Niasari, F. Davar, T. Mahmoudi, Polyhedron 28 (2009) 1455.
[29] N. Yahya, A. Aripin, A.A. Aziz, H. Daud, H. Zaid, L.K. Pah, N. Maarof, Am. J. Eng.
Appl. Sci. 1 (2008) 53.

You might also like