Author's Accepted Manuscript

Size dependent Electrical and Magnetic Properties

of Mechanically-Activated MnFe2O4 Nanoferrite
C. Behera, R.N.P. Choudhary, Piyush R. Das

www.elsevier.com/locate/ceramint

PII: S0272-8842(15)01292-4
DOI: http://dx.doi.org/10.1016/j.ceramint.2015.07.006
Reference: CERI10890

To appear in: Ceramics International

Received date: 7 April 2015

Revised date: 29 June 2015
Accepted date: 1 July 2015

Cite this article as: C. Behera, R.N.P. Choudhary, Piyush R. Das, Size dependent
Electrical and Magnetic Properties of Mechanically-Activated MnFe2O4 Nanoferrite,
Ceramics International, http://dx.doi.org/10.1016/j.ceramint.2015.07.006

This is a PDF file of an unedited manuscript that has been accepted for publication. As a
service to our customers we are providing this early version of the manuscript. The
manuscript will undergo copyediting, typesetting, and review of the resulting galley proof
before it is published in its final citable form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply
to the journal pertain.
 Size dependent Electrical and Magnetic Properties of
Mechanically-Activated MnFe2O4 Nanoferrite

C. Behera*, R N P Choudhary, Piyush R. Das#

Multifunctional Materials Research Laboratory, Department of Physics, Institute of Technical Education

and Research, Siksha ‘O’ Anusandhan University, Bhubaneswar-30, India

#
Department of Physics, VSS University of Technology, Burla, Sambalpur (Odisha)

Abstract

The manganese nanoferrite (MnFe2O4) with an average particle diameter of 15nm was synthesized
by mechanical activation followed by low-temperature thermal treatment. The milling energy
induced during activation of precursors for the synthesis of MnFe2O4 nano-powders has been
calculated using the basic principles of the collision model. X-ray structural analysis using
Rietveld refinement confirmed the formation of spinel phase of the material in the Fd3m space
group. The diameter of the nano-powder of the material was confirmed from high-resolution
transmission electron microscope (HRTEM). From the HRTEM micrographs study reveals that
the size of the nano particles can be controlled by the application of required milling energy or
time. Dependence of relative dielectric constant and tangent loss of the material on the crystallite
size has been thoroughly studied and correlated in our present study. Variation of dielectric
parameters with frequency and temperature has been explained using different mechanism. The
resistive characteristics, dielectric relaxation and negative-temperature-co-efficient of resistance
(NTCR) of the material obtained from complex impedance measurements has been correlated with the
particle size. The change in activation energy with particle size suggests the presence of a typical
conduction mechanism in the material. The saturation magnetization is found to be particle size
dependent. The synthesized manganese nanoferrite has potential applications in magnetic storage
device, site specific drug delivery systems (particularly in magnetic hypothermia for cancer
treatment), magnetic imaging etc.
*Corresponding author

E-mail:- [email protected]
Phone: +91-9438651271

Keywords: Manganese ferrite, XRD ,HRTEM, nanoferrites and impedance spectroscopy

1 Introduction
Ferrites with spinel structure having a general formula MFe2O4 (M=bivalent metal ions) have
attracted much attention of current material science researchers because of their wide ranging
applications including magnetic storage device, magnetic resonance imaging etc. [1-10]. The
tailoring of physical properties of the spinel ferrites is generally governed by the choice of the
cations and their distribution between tetrahedral and octahedral sites of the structure. The
electrical and magnetic properties of spinel ferrites depend on the cations-distributions at
different atomic sites of the structure [2-4]. The spinel structure of MFe2O4 (MFO) with space
group Fd3m is generally characterized by its cubic unit cell containing eight formula units. Each
cubic unit cell (with close packing of thirty two oxygen anions) has twenty four cations located
in eight of the sixty four tetrahedral sites (M) and sixteen of the thirty two are located at the
octahedral sites (B-site) [5]. Manganese ferrite has 80% normal spinel structure (i,e., 20% of
Mn2+ ion occupy octahederal site along with Fe3+ ions). Neel temperature (TN) of the material is
reported to increase with decrease in particle size and the prediction is consistent with the finite
size scaling model [6-8, 10]. TN for bulk MFO has been reported to be 300 °C .With decrease
in particle size to nano range the highest TN is reported to be 400°C for an average particle size
of 50nm.With further decrease in particle size TN further decreases gradually to 320°C for an
average particle size of 4nm. The variation in the Neel temperature with particle size is reported
to be the consequence of cation distribution among the tetrahedral and octahedral sites of the
spinel lattices [9]. Such type of effect has also been observed in the case of mechanically
activated MnFe2O4.
In the recent years, nano-ferrites have drawn a considerable scientific and technological attention
because of their applications in high-density magnetic recording media, magnetic fluids [11],
catalysts, gas-sensor [12-14], high-performance electromagnetic and spintronic devices [9-11]
and in magnetic hypothermia and drug delivery system [13]. Among all the spinel ferrites
known today, manganese ferrite is of particular interest because of its high saturation
magnetization, excellent chemical stability, high mechanical hardness, and high cubic magneto-
crystalline anisotropy [2-4].
Detailed literature survey reveals that synthesis of nano-ferrites by the mechanical activation
route (high-energy ball milling) [6-8] induces high coercivity in the ferrite powders [14-15].
Though some works on the effect of activation (milling) time on structural and magnetic
properties of MFO have been reported [16-20], systematic studies on structural, electrical and
magnetic properties of MFO has not been reported so far. In view of the importance of the nano-
sized MFO for applications, we have synthesized nano-manganese ferrite using planetary ball
milling with different milling time, and studied the effect of milling time on structural, electrical
and magnetic properties.
2. Experimental Techniques
2.1 Sample Preparation
The MFO nano-ferrite was fabricated by the mechanical activation (using a high-energy ball-
mill) followed by relatively low-temperature (750oC) heat treatment. The high–purity (analytical
grade) ingredients; MnCO3 (M/s Emerck) and Fe2O3 (M/s LOBA Chemie Co Ltd) were mixed
thoroughly in an agate-mortar and pestle in dry (air) as well as wet (methanol) media. The mixed
powder was mechanically activated in a zirconium grinding jar of 125 ml capacity with 30
zirconium balls of 10 mm diameter. The milling of the mixed powder was carried out for
different duration (i.e., for 30, 60, 90 h) by a planetary ball mill (M/s Retsch PM 100, Germany)
with 450 rpm in a toluene medium. After each 30 h of milling, specific amount ( i.e 5 g )of the
mixed powder was taken out each time to calculate the amount of mechanical energy transferred
during milling of different duration. All the different time milled samples were calcined at a
relatively low temperature (750oC) as compared to that of compound prepared by mixed oxide
method (1200oC) [20-22]. The phase-purity, nature and size of the calcined powder were
determined by using X-ray diffraction technique and high-resolution transmission electron
microscope technique.The disc-shaped pellet samples of the milled powder were prepared using
a KBR hydraulic press at an isostatic pressure of 5×105 N/cm2.The pellets were sintered at a
temperature 770oC for microscopy and electrical characterization.
2.2 XRD pattern and HRTEM micrographs
The crystal system, unit cell dimension and related parameters of the milled powders were
determined using x-ray diffraction pattern of MFO recorded with CuKα radiation (λ=1.5406 Å)
in a wide range of Bragg angles at room temperature. In order to understand the structural
modification if any, we have carried out structural refinement by Rietveld method using MAUD
computer program. The crystallite size (P) of the different duration milled powder samples was
0.89.
calculated using Scherer’s formula; P = where ß is the broadening of the diffraction line
ßcos
(measured at half height of maximum intensity (in radian)), λ = 1.5406 Å (wavelength of Cu-Kα)
and θ is the Bragg’s angle. The surface morphology of the sintered pellets of the milled samples
was recorded by scanning electron microscopy (ZEISS-SEM). The highly-resolved transmission
electron micrograph of 90 h milled MnFe2O4 sample along with their corresponding SAED
pattern of the calcined powder was recorded using a copper grid with the help of the high-
resolution transmission microscope (HRTEM, model JEM-2100, JEOL, Japan).

2.3 Measurements of Electrical Parameters

Some electrical parameters (capacitance, dissipation factor, impedance, etc) of the disc-shaped
pellet samples (prepared from the milled powder of different duration) were obtained as a
function of temperature (30–5250C) over a wide range of frequencies (1 kHz –1 MHz) using a
computer-interfaced phase sensitive meter (N4L PSM LCR, Model 1735, UK) and a laboratory-
fabricated sample holder. The relative dielectric constant (εr) of the samples was calculated using
the relation; εr = Cd/ε0A where C is the measured capacitance, A is the surface area and d is the
thickness of the samples. To establish correlation between electrical properties with
microstructure, complex impedance spectroscopy (CIS) has been used and already been
discussed in our earlier report with others [23-26].

2.4 Magnetic study

A quantum design (Ever Cool SQUID VSM DC magnetometer Quantam Design, USA) SQUID
magnetometer was used for the magnetic characterization of the synthesized nanoparticles. The
magnetization was measured with magnetic field cycling between +8 tesla and-8 tesla at room
temperature.
3. Results and Discussion
3.1 Energy calculation based on collision model

Based on the mechanical activation approach of collision model, [27-31] energy of the ball is considered
to be reduced by a factor when there is more than one ball in the vial. The energy of each ball (Es) can be
written as Es=1/2 ms.Vs2=30416.73 erg, where ms is the mass of each ball and Vs is the linear velocity of
the ball. For 30 balls used in the present study (in the vial), the energy of each ball could be
written as En= ФbEs and Фb=1- nvε, where nv is the packing fraction of vial in a simple cubic
arrangement, and thus nv=nb/nbv=0.1935 where nbv=155 (the maximum number of ball that can be
accommodated in the vial). In this situation, the value of Фb=1- nvε=0.8035, and thus En=
ФbEs==24439.84 erg. The total energy (Et ) transferred during milling for one gram(g) of

material in one hour can be calculated as (Et)= ν En ´nb×t/ wp, where wp is the quantity of powder
taken. Thus Et=7×24439.84 erg×30×3600 s/20g=92.38 joule/g. For 30 h of milling with 450
rpm; energy transferred per gram of material= 30× Et=30×92.38 joule/g= 2771.4 joule/g=0.7698
kwhr/g.

After each 30 h interval of milling, 5g of material was taken out and milled the remaining one for
another 30 h (i.e., for 60 h of milling) keeping all parameters same except by changing the
weight of powder reduced to 15g. Again, after 30 h of interval, 5g of material were taken out
and milled for remaining one for 30 h (i.e., for 60 h of milled sample is 15 g and 10g of material
remaining for 90 h of milling). The total calculated mechanical energy for 30, 60 and 90 h of
milled samples is found to be 0.77, 2.05, 4.62 kwhr/g respectively. However, gradual reaction
process needs much longer milling time; this means that heat loss is much serious. Based on the
above analysis, a conclusion can easily be drawn on the formation of manganese ferrite of the
smaller particle size.

3.2 Structural studies

Fig. 1 shows the XRD pattern of the milled and later calcined powders of MFO. Fig 1(a-c)
compares the Rietveld refined pattern with that of experimental one for different milled MFO
sample. The best agreement between refined profiles with that of experimental one in Fd3m
space group clearly suggests the validity of the refined model and calculated parameters. The
refined lattice parameters and unit cell volume of different milled samples are compared in Table
I. The broadening of the reflection peaks indicates that the samples are composed of small
particles. On increasing the milling time, the broadening of the reflection peaks goes on
increasing and the intensity goes on decreasing. Thus crystallite/particle size of the powders
decreases on increasing milling time. The broadening of experimental peaks suggests that a large
number of defects can be introduced into the samples on increasing the milling time. The
broadening of the reflection profiles is not only caused by the small particle or crystallite size but
also due to the lattice strain, instrumental errors, dislocation, impurities etc. [26]. Broadening of
the peaks caused by the strain and mechanical defects has been ignored in our calculation, as the
powder samples were used to record XRD patterns. There is good agreement between
experimental XRD pattern of the sample and reported one (JCPDS Card No-10-0319).
Fig.2 shows the scanning electron micrograph and also confirms that on increasing milling time,
the grain size goes on decreasing. Based on SEM micrograph of the different milled sample the
calculated average grain size is found to be 0.29µm (for 30h), 0.18 µm (for 60h) and 0.13 µm
(for 90h).
Fig.3 shows the HRTEM micrograph along with its SAED pattern of 90 h milled MnFe2O4 nano-
ceramics. The calcined fine powders were spread over a copper grid via sonication for recording
of HRTEM images. The HRTEM imaging was used for the investigation of the morphology and
crystallographic feature of the single phase of the prepared material. The bright field image of
the samples shows nearly spherical nanoprticles with more or less uniform size distribution. The
calculated particle size matches with the coherently diffracted domain size calculated from the
XRD pattern. The particle size of 90 h milled sample is found to be 15nm. This observation
confirms the formation of nano-sized powder or sample of the material. The distinct but diffused
rings of the SAED pattern are a clear indication of the formation of polycrystalline samples at the
nano-scale. Representative SAED patterns are indexed in the cubic spinel structure of the
MnFe2O4 phase, and also support the single crystalline nature of the sample.
3.3 Dielectric studies
Fig. 4 shows the temperature dependence of relative dielectric constant and tangent loss at some
selected frequencies. The value of εr increases gradually on increasing temperature with a same
trend as that of tanδ. The rate of increase in tan δ in the material at low temperatures is slow,
whereas at higher temperatures the increase is sharper. This sharp increase in the value of tan δ
at higher temperatures is due to scattering of thermally activated charge carriers and presence of
some unknown defects including oxygen vacancies in the material. At higher temperatures, the
conductivity begins to dominate, which in turn, is responsible for rise in tanδ. On increasing
milling time, the dielectric constant increases in the low temperature-frequency region. The same
trend is observed in the high temperature and frequency region. As the composition, milling
configuration and annealing temperature for all the three samples are same but with difference in
milling time; the decrease in particle size is responsible for the enhancement of dielectric
constant. It is also observed that milling time changes the loss spectrum of the samples
significantly. There is an anomaly in both εr and tanδ at about 300 oC. The dielectric peak shifts
towards low-temperature side on increasing frequency, and thus peak shifts toward higher
temperature side on increasing milling time in all the samples.

Fig.5 shows the variation of relative dielectric constant (εr) of the different duration milled
samples of MFO with frequency at selected temperature. The dielectric parameters (εr) decrease
on increasing frequency, which is a general trend of variation for oxide materials [32]. At low
frequency, all the different types of polarizations (interfacial, dipole, atomic, ionic, electronic,
etc) exist in the material. That is the main reason for the higher value of dielectric constant in the
low-frequency region. In the low-frequency region, dipolar/interfacial or surface polarization
plays a dominant role in the higher value of dielectric constant as compared to that of ionic or
electronic polarization. The existence of these polarizations vanishes slowly on increasing
frequency. As a result, dielectric constant decreases. Beside this, there are some other reasons for
higher value of dielectric constant at lower frequency. The higher value of dielectric constant at
lower frequency can be understood using the Maxwell–Wagner interfacial polarization and
Koop’s phenomenology theory [33]. The low-frequency dispersion may be due to
inhomogeneous nature of studied dielectric compound having two layers [34,35]. The first layer
is of larger ferrite grains of highly conducting elements that is separated by the second thin layer
of relatively poor conducting substance (grain boundary). The grain boundaries and grains are
more effective at low and high frequencies respectively. As electrons are the majority charge
carriers in the spinel ferrites, the hopping mechanism of electrons takes place between Fe3+ ions
and Fe2+ ions present at different crystallographic sites. The smaller value of εr at higher
frequencies may be due to the electron hopping between Fe3+ ions and Fe2+ ions situated at the
octahedral and tetrahedral-sites. Therefore, electrons have to pass through the well conducting
grains and the poorly conducting grain boundaries. As grain boundaries offer high resistance, the
electrons get crowded there, and thus producing an enhanced space charge polarization. As a
result, the material has larger εr value in the low-frequency range. On increasing frequency, the
direction of motion of electrons changes rapidly. This hinders the movement of electrons inside
the dielectric material, and thereby, reducing accumulation of charges at grain boundaries.

3.4 Impedance studies

Fig.6 shows the variation of real and imaginary part of impedance ( Z ' and Z '' ) with frequency at
different temperature of all the three samples of the MFO. The magnitude of Z ' decreases on
increasing temperature in the low-frequency range that merges in the high-frequency region
irrespective of temperature. This nature of the plot may be due to the release of space charge
[35]. The reduction in barrier characteristics of the material with rise in temperature may be a
responsible factor for the enhancement of ac conductivity of the material at higher frequencies
[36, 37]. Further, in the low-frequency region, a decrease in magnitude of Z ' with rise in
temperature shows negative temperature coefficient of resistance (NTCR) behavior of the
material. This behavior is changed drastically in the high-frequency region showing complete
merger of Z ' plots above a certain frequency. At higher frequency, the value of Z ' of different
temperatures coincides implying the possible release of space charge [38]. A peak at a particular
frequency (at which Z ' becomes independent of frequency) was observed which shifts towards
the higher frequency side on increasing temperature. The shift in Z ' plateau indicates the
existence of frequency relaxation process in the material. The nature of the curves displaying
single relaxation process shows the increase in ac conductivity with rise in temperature and
frequency [26]. The monotonous decrease in Z '' in the low-temperature region, appearance of
peaks in the loss spectrum at high temperatures and significant increase in peak broadening with
increase in temperature are the special features of the loss spectrum. The absence of peaks in the
low-temperature range for all the milled samples (in the loss spectrum) suggests the lack of
current dissipation in the mentioned temperature region. The different peaks of the different
samples (appeared at particular frequency) describe the type and strength of electrical relaxation
phenomenon in the material [24, 25]. The Z '' max shifts to higher frequency side on increasing
temperature indicating an increase of tangent loss in the samples. A significant increase in the
broadening of the peaks with increase in temperature suggests the existence of a temperature
dependence of electrical relaxation phenomenon in the material. The relaxation process may be
due to the presence of electrons and/or immobile species at low temperatures and defects or
vacancies at higher temperatures. The asymmetric broadening of the peaks suggests a spread of
relaxation time with two equilibrium positions. The peak height is proportional to the bulk
resistance (Rb), that can be explained using the equation, Z ''[ Rb{ / (1   2 2 )}] . Further, the
magnitude of Z '' max decreases gradually with shift of peak position towards high-frequency side.
Finally, they merge in the high-frequency region. This merger is an indication of the
accumulation of space charge in the samples [32]. Finally, it is observed that on increasing
milling time, both Z′ and Z′′ first increase with rise in temperature, and then remain constant
above 400oC. These parameters are also independent of frequency. The shift of Z '' max towards
higher frequency side (on increasing milling time of the materials) shows decrease of tangent
loss. A significant increase in the broadening of the peaks with increase in milling time suggests
the enhancement in electrical relaxation phenomenon in the materials.
Fig.7 shows Nyquist plots (complex impedance spectrum) of the material at different
temperatures (up to 3000C) over a wide range of frequency range (100 Hz–1MHz). The micro-
structural characteristics of the materials are characterized by the formation of temperature
dependence of semicircular arcs in the impedance spectrum. The extent of intercept of
semicircles on the real axis (x-axis) and its number in the spectrum provide information on the
kind of electrical processes occurring within the material. The correlation between the arcs and
microstructure of the samples is established by drawing an equivalent electrical circuit. The
experimental and theoretically fitted data were compared using commercially available software
(ZSIMP WIN version 2.0). A proposed equivalent circuit of the fitting curves consists of a
parallel combination of CR and CQR where Q is known as constant phase element (CPE) that is
used for the checking the validity of experimental data with that of the proposed model. Using
the fitted curves, the values of bulk resistance (Rb), grain boundary resistance (Rgb), bulk
capacitance (Cb) and grain boundary capacitance (Cgb) at different temperatures were calculated,
and compared at 300oC (Table II). The semicircular arcs of the impedance pattern are mainly
attributed to a parallel combination of resistance and capacitance. As temperature increases from
room temperature, an arc progressively becomes semicircular with a shift of the centre towards
origin of the complex plane plot. On further increase in temperature, the slope of the line
decreases, and bend towards Z ' - axis (above 3000C), and thus a semicircle could be traced
indicating the increase in conductivity of the material [32]. The presence of semicircular arcs
above 3000C suggests that the electrical processes in the samples arise basically due to the
contributions from bulk material (grain interior) and grain boundary. With increase in milling
time, the radius of semicircular arc generally increases. A comparative study at 300oC of
different capacitive and resistive parameters of the milled samples (30 h, 60 h and 90 h) are
given in Table.II which give some fundamental ideas on disordering of the manganese nano-
ferrite. The Nyquist plots of milled samples clearly shows depressed semi-circles suggesting the
presence of non–Debye type of relaxation in the material. Therefore, a distribution of relaxation
time is expected in the material instead of a single relaxation time [26, 32]. The intercept of each
semi-circle on real Z' axis gives the value of bulk and grain boundary resistance and capacitance.
The impedance spectra have characteristic peaks occurring at a unique relaxation frequency
usually referred to as resonance frequency (fr) (r = 2fr). The relaxation time is calculated
using a simple equation; RC = r = 1, or fr = 1/ 2RC, where  is the relaxation time. The
relaxation time due to bulk effect (b) is calculated using the equation rb = 1 or, b = 1 / 2fr.

3.5 Studies of AC Conductivity

Based on detailed studies of electrical conductivity of a number of ferrites, Jonker [37] suggested
two regions of conductivity in them; (i) low conductivity containing Mn2+ and Mn3+ ions and (ii)
high conductivity containing Fe2+ and Fe3+ ions. The presence of Mn at the octahedral-site of the
spinel structure favors predominantly conduction mechanism of Mn2+ +Fe3+↔Mn3++Fe2+ type.
The presence of impurities in such type of materials can influence the conductivity of the ferrite
[38-41]. As the electrons of the manganese are localized, the electrical conductivity in such type
of ferrites increases with increase in temperature(due to the hopping of the charge carrier). Thus
electrical conduction in MFO is due to electron hopping between the Fe2+ and Fe3+ ions
(displacement of electron in the direction of the applied field). These electrons determine the
polarization in ferrites. The polarization decreases with the increase in frequency, and then
reaches a constant value. It is because, above a certain frequency of the external field, the
electronic exchange cannot follow the alternating field. Therefore, all the studied samples show
an increasing trend of conductivity on increasing frequency, and follow the hole hopping process
of Mn3+ and Mn2+ ions [39] as per the equations;
Mn 3+ ↔ Mn2+ +e+ ( polaron hopping)
Fe2+ ↔ Fe3+ + e− (electron hopping)
Some dielectric parameters were used for the calculation of ac conductivity using the relation:
ζac = ωεrε0tanδ……….(1)
where ε0 = dielectric permittivity of free space and other parameters have their usual meanings.
Based on the thermal activation, the activation energy Ea of the samples can be calculated using
the conductivity relation:
ζac = ζ0exp (-Ea / kT)……….(2)
where k = Boltzmann constant and ζ0 = pre-exponential factor.
Fig.8 exhibits the temperature dependence of ac conductivity (Arrhenius plot) of milled samples
at different frequencies. The Arrhenius plots (log ζac vs. 1000/T) show that the value of log ζac
increases linearly with rise in temperature. The estimated (using linear fitting) value of activation
energy in the high-temperature region (300-500 OC) at different frequencies is compared in Table
III. The different slope of the plots (in different temperature regions and frequency) suggests the
presence of multiple conduction processes in the prepared samples with different activation
energy [40-43]. The conductivity decreases with increase in milling time, because of the
decrease in grain size and reduction in the number of Fe3+ - Fe2+ pairs in the octahedral site
arising from the partial displacement of Fe3+ ions from the octahedral site to the tetrahedral site
[40]. Beside this, on increasing milling time, the disorderness (i.e., amorphous nature) of the
material increases, as a result, conductivity decreases. The activation energy due to electron
hoping is in the order of 0.4eV. The medium value of activation energy (>0.4eV) is due to the
hopping of holes in addition to the electron hoping [41].
Fig.9 shows the variation of ζac with frequency at selected (high) temperatures. Our experimental
results can be explained with non-linear fitting model of the Jonscher’s power law. The value ac
is frequency dependent in the high-frequency region, but almost independent in the low-
frequency region. The extrapolation of low-frequency ζac gives dc conductivity ζdc. However,
the increasing trend of ζac with rise in frequency (in low-frequency region) may be due to the
disordering of cations of the neighboring sites and presence of space charge. In the high-
frequency region, the curves approach to each other. The curves exhibit low-frequency
dispersion phenomena obeying the Jonscher’s power law. The origin of the frequency
dependence of conductivity lies in the relaxation phenomena arising due to mobile charge
carriers. When a mobile charge carrier hops to a new site from its original position, it remains in
a state of displacement between two potential well energy minima. Also, the conduction
behavior of the material obeys the power law;
ζ(ω)α ωn …..(3)
with a slope change (governed by n in the low-temperature region). The small value of n (<1)
suggests involvement of a translational motion of the charge carriers with a sudden hopping
whereas larger value of n (>1) suggests involvement of localized hopping without the species
leaving the neighborhood. The frequency at which change in the slope takes place is known as
hopping frequency of the polarons (ωp), and is temperature dependent. From non-linear fitting it
is found that that motion of charge carriers is translational because of small value of n (<1) [41].
Gopalan et al [41] have proposed the polaron conduction mechanism in the nanostructure ferrite,
which reveals that the polaron formation takes place in the nanostructure ferrite due to the defect
levels present at nano-level. As per the OLPT (overlapping large polaron tunneling) model of
Long [44], the large polaron wells situated at two sites are reduced by the polaron hopping
energy(WH). The resultant excitation is called a large polaron. Due to long-range coulomb
interaction, the potential well of a site will be extended over many inter-atomic distances and
overlap with the well of other sites. The activation energy of the sites can be found out according
to the relation mentioned in our earlier report [26]. The value of n initially decreases with rise in
temperature, and then increases with increase in milling time. The formation polaron in the nano
ferrite system can be considered due to the defect formation at the nano scale. The multiple
valence states of Mn (Mn2+ and Mn3+) in the material point towards possible Jahn-Teller
distortion occurring in the creation of defects which can facilate polaron formation. Because of
these the dc conductivity is expected to increase with rise in temperature.
3.6 M-H Study
As per the Neel’s theoretical model for bulk MnFe2O4 [l] all the Mn ions are divalent and Fe ions are
trivalent, and hence magnetic moment at zero temperature is expected to be 5μB. However, taking both
Mn2+ and Fe3+ of d5 ions with a 5/2 spin and zero orbital momentum, the calculated value of magnetic
moment of 5μB would be independent of the inversion degree. Thus, the distribution of metal ions with
the same Bohr magnetron has no influence on the net magnetization. However, from magnetic scattering
measurements, the magnetic moment of MnFe2O4 was found to be ~4.6μB/f.u. This reduced magnetic
moment of MnFe2O4 has attracted much attention of researchers, and different models have been
proposed to explain this reduced value [1, 42-43]. The presence of Mn3+ ions is counterbalanced either by
Fe2+ ions or some vacancies in the spinel structure. In spinel ferrites, it is possible to calculate the net
magnetic moment per formula unit at zero degree absolute temperature (based on Néel’s two-sublattice
model of ferrimagnetism):
μ =MB - MA (4)
where MB and MA are the sub-lattice magnetic moments of the B and A sites ions.
In the present study, when Fe2+ ions are not present at the Oh sites, the chemical formula for manganese
ferrite may be written as
[Mn1-x2+ Fex3+]A(Mnx3+Fe2-x3+)BO4 (5)
As magnetic moments of Fe3+, Mn2+ and Mn3+ are 5, 5 and 4μB respectively, the net magnetic moment
(μ) per formula unit can be expressed as: μ= 5(2-x)+ 4x- 5x -5(1-x)= (5- x)μB/f:u (6)
In MnFe2O4 nanoparticles the formation of trivalent manganese ions should be as low as possible. As
Mn3+ shows a preference for octahederal sites and raising the inversion degrees, a decrease in the value
of μ can be derived using equation (6) rather than the reported μ=(5-2x) μB /f.u by others[43]
The moderate value of saturation magnetization (Ms) and coactivity (Hc) of the synthesized
materials shows the dependence of these parameters on milling energy and time. The moderate
value is due to the crystalline nature of the nano-powders despite of their finite size.
Fig.10 clearly shows that the value of Ms for 30 h milled sample is greater than that of 60 h and
90 h sample. It can be concluded that at a particular particle size the magnetization of the
material decreases rather than following the general rule of milling time or particle size
dependence of magnetization. It decreases to a considerable level up to ~15nm for 90 h milled
sample as evident from the HRTEM micrograph. In this sample, the size of the particles is too
small and non-collinear of magnetic moments on the surface. It is because of broken exchange
bonds at the external layers of the particles. Therefore, it is sufficiently important to reduce the
magnetization of the whole particles. When the surface area to volume ratio increases, the
particle size decreases, and hence the surface canting increases [45, 46]

4. Conclusion
The MnFe2O4 nano-ferrite was synthesized by mechanical activation (high-energy ball milling)
followed by thermal treatment (calcinations and sintering at relatively low- temperature) routes.
The formation of face–centered cubic phase (inverse spinel structure) of MnFe2O4 was checked
by preliminary x-rays structural analysis. The crystallite size of the material was initially
calculated using x-rays data in Scherrer’s equation. On increasing milling or activation time, the
crystallite size of the samples was found decreasing. The formation of nano-structured materials
is finally confirmed by high-resolution transmission electron microscopic analysis. The
crystallite size thus obtained is very much consistent with that found from the x-rays data. The
variation in the value of dielectric permittivity/constant, tangent loss and ac conductivity of the
milled samples with frequency and temperature was explained on the basis of the Maxwell-
Wagnor theory of interfacial polarization and Koop’s phenomenological theory. The ac
conductivity of the samples is found increasing with rise in temperature but decreases with
increase in milling time. The impedance studies exhibit the presence of grain (bulk) and grain
boundary in all the samples milled for 30, 60 and 90 h. The existence of negative temperature
coefficient of resistance (NTCR) in the material is observed. An equivalent circuit was used to
demonstrate electrical phenomena occurring inside the material. The ac conductivity obeys
Jonscher’s universal power law. The impedance spectroscopy of the materials confirms the
origin of the relaxation mechanism in the system. The calculated activation energy is very much
in agreement with the type of conduction mechanism occurring in the material. The polaron
hopping in the material is in good agreement with that of experimental data. The high value of
saturation magnetization shows that material is a good ferri-magnetic. All the above
characterizations suggest that the properties of nano-structured MFO can be designed for suitable
applications.
Acknowledgement
The authors gratefully acknowledge the grant received from DRDO (Grant number:
ERIP/ER/1102202/M/01/1438 dated 25/07/2012) Government of India to carry out this work.
The authors are also grateful to CRF, IIT Kharagpur for providing some experimental facilities

References

1. S. Bullita, A. Casu, M. F. Casula, G. Concas, F. Congiu, A. Corrias, A. Falqui,D . Loche and

C. Marras, ZnFe2O4 nanoparticles dispersed in a highly porous silica aerogel matrix: a
magnetic study, Phys. Chem. Chem. Phys. 16 (2014) 4843-4852
2. J. M. Hastings, L.M. Corliss, Neutron Diffraction Study of Manganese Ferrite,
Phys.Rev.104 (1956) 328-331
3. A. Elfalaky, S.Soliman, Theoretical investigation of MnFe2O4, J. Alloys Compd. 580 (2013)
401-406
4. K. Vamvakidis, M. Katsikini, D. Sakellari, E. C. Paloura, O. Kalogirou and C. Dendrinou-
Samara, Reducing the inversion degree of MnFe2O4 nanoparticles through synthesis to
enhance magnetization: evaluation of their 1H NMR relaxation and heating efficiency, Dalton
Transaction 43 (2014)12754-12765
5. Dongen Zhang, Xiaojun Zhang, Xiaomin Ni, Ji Mei Song, Huagui Zheng, Low-
temperature fabrication of MnFe2O4 octahedrons: Magnetic and electrochemical
properties, Chemical Physics Letters 426 (2006)120–123
6. M.H. Mahmoud, H.H. Hamde, A.I. Abdel-Mageed, A.M. Abdallah, M.K.

Fayek, Effect of HEBM on the cation distribution of Mn-ferrite, Physica B 291

(2000) 49-53
7. E. Ranjith Kumar, R. Jayaprakash, T. Arun Kumar ,Sanjay Kumar, Effect of reaction
time on particle size and dielectric properties of manganese substituted CoFe2O4
nanoparticles, Journal of Physics and Chemistry of Solids 74, (2013)110–114
8. D. J. Fatemi, V. G. Harris, V. M. Browning and J. P. Kirkland, Processing and cation
redistribution of MnZn ferrites via high-energy ball milling, J. Appl. Phys 83 (1998)
6878
9. C. N. Chinnasamy, Aria Yang, S. D. Yoon, and Kailin Hsu, Size dependent magnetic
properties and cation inversion in chemically synthesized MnFe2O4 nanoparticles ,
 J. Appl. Phys, 101 (2007)09M509
10. Jianjun Li, Hongming Yuan, GuodongLi , Yanju Liu, Jinsong Leng, Cation
distribution dependence of magnetic properties of sol–gel prepared MnFe2O4 spinel
ferrite nanoparticles, J. Magn. Magn. Mater. 322 (2010)3396–3400
11. D. Makovec, A. Košak, A. Žnidrašič, M. Drofenik, The synthesis of spinel–ferrite
nanoparticles using precipitation in microemulsions for ferrofluid applications, J.
Magn. Magn. Mater., 289 (2005)32–35
12. A.T. Raghavender, Nguyen Hoa Hong, Dependence of Ne´el temperature on the particle size
of MnFe2O4 , J. Magn. Magn. Mater. 323 (2011)2145–2147
13. Saqlain A. Shah , A. Majeed , K. Rashid , Saif-Ullah Awan, PEG-coated folic acid-modified
superparamagnetic MnFe2O4 nanoparticles for hyperthermia therapy and drug
delivery, Materials Chemistry and Physics, 138 (2013)703-708
14. D. Carta, M. F. Casula, P. Floris, A. Falqui, G. Mountjoy, A. Boni, C. Sangregorio and A.
Corrias, Synthesis and microstructure of manganese ferrite colloidal nanocrystals, Phys.

Chem. Chem. Phys., 12 (2010) 5074-5083

15. I. Koh, L. Josephson, Magnetic nanoparticle sensors, Sensors, 9 , (2009) 8130–8145

16. Dalia S. Mathew, Ruey-Shin Juang, An overview of the structure and magnerism of
spinel ferrite nanoparticles and their synthesis in microemulsion, Chemical
engineering Journal, 129 (2007) 51-65
17. Sadan Ozcan, Serkan Akansel, Abdullah Ceylan , The influence of reactive milling on the
structure and magnetic properties of nanocrystalline MnFe2O4, Ceram. Int., 39 (2013)5335-
5341
18. Telem simsek,Serkan Akansel,sadan Ozcan,Abdullah Ceylan, Synthesis of MnFe 2O4
nanocrystals by wet-milling under atmospheric conditions, Ceram. Int., 40 (2014)7953-7956
19. M. Muroi R. Street P. G. McCormick J. Amighian, Magnetic properties of ultrafine
MnFe2O4 powders prepared by mechanochemical processing, Phys.Rev. B, Condensed
matter, 63 (2001)184414
20. B. Aslibeiki,P.Kameli and H.Salamati, The effect of grinding on magnetic properties of
agglomerated MnFe2O4 nanoparticles, J. Magn. Magn. Mater. 324 (2012) 154-160
21. J. Ding, P.G. McCormic, R. Street, Formation of spinel Mn-ferrite during mechanical
alloying, J. Magn. Magn. Mater.,171 (1997)309-314
22. N. Sivakumar and A. Narayansamy,K Shinoda ,C.N Chinnasamy, B.Jayadevan ,J.M.
Greneche,Electrical and magnetic properties of chemically derived nanocrystalline
nano crystalline cobalt ferrite, J. Appl. Phys,102 (2007)013916
23. J. Suchanicz , The low-frequency dielectric relaxation, Na0.5Bi0.5TiO3 ceramics Mater. Sci.
Eng. B, 55 (1998)114-118
24. J. R. Macdonald, Impedance spectroscopy and its use in analyzing the steady-state
AC response of solid and liquid electrolytes, J. Electroannal. Chem., 223 (1987)25-50
25. J.R. Macdonald, Impedance Spectroscopy Emphasizing Solid Materials and Systems
(Chapter-4), Wiley, New York (1987)
26. C. Behera, R.N.P Choudhary, Piyush R. Das, Size dependent electrical and magnetic
properties of mechanically alloyed CoFe2O4 nanoferrite. J. Mater. Sci: Mater.
Electron. ,26 ( 2015) 2343-2356
27. N. Burgio, A. Iasonna and M. Magini ,S.Martelli and F.Padella, Mechanical Alloying of
Fe-Zr System, Correlation between input energy and end product, 1L Nuovo Cemanto 13D
April (1991) N4
28. A. Iasonna and M. Magini, Power measurements during mechanical milling: an
experimental way to investigate the energy transfer phenomena, Acta. Mater. 44 (1996)
1109-1117
29. C. Suryanarayana, Mechanical alloying and milling, Progress in Materials Science 46,
(2001)1-184
30. B.S. Murty, M. Mohan Rao, S. Ranganathan, Milling map and amorphization during
mechanical alloying, Acta metall. mater. 43 (1995) 2443-2450.
31. M. Ahmadzadeh, A. Ataie , E. Mostafavi, The effects of mechanical activation energy on
the solid-state synthesis process of BiFeO3, J. Alloys Compd., 622 (2015) 548–556
32. A. K. Jonscher, The `universal' dielectric response, Nature 267 (1977) 673-679
33. C.G. Koops, On the Dispersion of Resistivity and Dielectric Constant of Some
Semiconductors at Audio frequencies, Phys. Rev., 83 (1951)121
34. H. Jain, C.H. Hsieh, Window’ effect in the analysis of frequency dependence of ionic
conductivity , J. Non-Cryst. Solids, 172 (1994) 1408-1412
35. Mathew George, Swapna S Nair, K A Malini1, P A Joy and and M R Anantharaman,
Finite size effects on the electrical properties of sol–gel synthesized CoFe2O4
 powders: deviation from Maxwell–Wagner theory and evidence of surface
polarization effects, J. Phys. D: Appl. Phys. 40 (2007)1593-1602
36. M.J. Akhtar, M. Younas, Structural and transport properties of nanocrystalline
MnFe2O4 synthesized by co-precipitation method, Solid State Sciences 14,
(2012)1536-1542
37. G .H. Jonker, Analysis of the semiconducting properties of cobalt ferrite, Journal of
Physics and Chemistry of Solids 9(1959) 165-175
38. K. Funke, Jump relaxation in solid state electrolyte, Solid State Chem., 22
(1993)111-195
39. N.Sivakumar, A.Narayanasamy, N.Ponpadian, J-M Greneche, K. Shinoda,
B.Jayadevan and K Tohji, Effect of mechanical milling on the electrical and magnetic
properties of nanostruructured Ni0.5 Zn0.5 Fe2O4, J. Phys. D:Appl Phys.39 (2006)4688-
4694
40. E. V. Gopalan, P.A. Joy, I.A. Al-Omari, D. S. Kumar, Y. Yoshida, M.R.
Anantharaman, On the structural, magnetic and electrical properties of sol–gel
derived nanosized cobalt ferrite, J. Alloys Compd., 485 (2009) 711-717
41. E. V. Gopalan, K. A. Malini, S Saravanan, D S Kumar,Y. Yoshida And M.R
Anantharaman, Evidence for polaron conduction in nanostructured manganese ferrite,
J.Phys.D:Appl.Phys,41(2011)185005
42. J.R. Huang and C. Cheng, Cation and magnetic orders in MnFe2O4 from density
functional calculations, J. Appl. Phys., 113( 2013)033912
43. Xu Zuoa,, Bernardo Barbiellinib, Calculation of exchange constants in manganese
ferrite (MnFe2O4), J. Magn. Magn. Mater. 272– 276 (2004) 306–311
44. A. R. Long, Frequency dependent loss in amorphous semiconductors, Adv. Phy, 31
(1982) 553-637
45. G. Balaji, N.S. Gajbhiye, G.Wilde, J. Weissmuller, Magnetic properties of MnFe2O4
nanoparticles, J. Magn. Magn. Mater. 242–245 (2002)617–620
46. T. E. Whall ,N. Salerno ,Y. G. Proykova ,V. A. M. Brabers, The electrical properties of
MnFe2O4 at the Néel temperature, Philosophical Magazine B 56 (1987) 99-105
 MnFe2O4 30h Experimental Data
Computed(Refined data)

Intensity(Arb. unit)

20 30 40 50 60 70
Angle(2)

1-(a) Reitvial refinement of XRD of MFO 30h

MnFe2O4 60h
Experimental data
Computed /Refined data
Intensity(Arb.Unit)

20 30 40 50 60 70
Angle(2)

1-(b) Reitvial refinement of XRD of MFO 60h

 MnFe2O4 90h Experimental data
Computed Data(Refined)
Intensity(Arb. unit)

20 30 40 50 60 70
Angle(2)

1-(c)Reitvial refinement of XRD of MFO 90h

30 h
60 h
 90 h

Fig.2 SEM micrograph of different milled sample of MnFe2O4 ceramics.

Fig.3-HRTEM micrograph of 90h milled MnFe2O4 sample.
Fig.4 Temperature dependence dielectric and loss spectrum of different milled MnFe2O4sample
Fig-5 Variation of dielectric with frequency of 30h,60h and 90h of milled MnFe2O4 sample
Fig.6 Variation of Z’ and Z’’ with frequency of 30h,60h and 90h milled MnFe2O4 .
Fig.7 Nyquist plot of 30h,60h and 90h of milled MnFe2O4 nanoferrite.
Fig.8 Temperatute dependent a.c conductivity of different milled MnFe 2O4 sample
Fig.9 frequency dependent a.c conductivity of different milled sample MnFe2O4
 Fig.10- Size dependent magnetic properties of 30h,60h and 90h milled sample of MnFe 2O4

Milling χ2 Weighted Rnw(%) Rb(%) Rexp(%) Cell Crystallite

time(hour) profile parameter(AO) size(nm)
factor(Rw)
30 1.61 24.05 0.24 17.44 14.88 8.3648 89
60 0.79 12.05 0.12 6.82 8.24 8.3231 54
90 0.66 4.42 0.04 2.81 6.63 8.2590 24

Table-I Refined Structural parameter details (Refined with Space group Fd3m) of the
different milled sample.
Table II. Comparison of impedance fitting parameters ( Rb, Cb, Rgb, CPE, Cgb) of MnFe2O4 at
300oC of different milling time.
______________________________________________________________________________
Milling
time(h) Cb (nF) Rb(Ω) CPE Cgb(nF) Rgb(Ω)

30 7626 72.73 1.005E-8 1.25E-10 5180

60 6383 53.98 4.929E-8 9.506E-11 3565

90 4277 18.88 9.982E-8 2.035E-10 1512

_____________________________________________________________________________________
 Milling Frequency(Hz) Activation Activation
time(hr.) energy(eV)(60- Energy(eV)(227-
o
127 C) 441oC)
25K 0.3022 0.5877
100K 0.2027 0.7169
30 500K 0.1287 0.6515
1M 0.1106 0.6144
25K 0.2590 0.1137
100K 0.2149 0.0921
60 500K 0.1666 0.1162
1M 0.1676 0.0758
25K 0.2286 0.2506
90 100K 0.154 0.2725
500K 0.1145 0.1549
1M 0.1075 0.1963

Table-III Activation energy of different milled time samples of MnFe2O4 nanoferrite at different
temperature range.