PCCP

View Article Online

PAPER View Journal | View Issue

Dielectric properties of cobalt ferrite nanoparticles in

ultrathin nanocomposite films
Published on 10 October 2013. Downloaded by Laurentian University on 01/11/2013 10:11:11.

Cite this: Phys. Chem. Chem. Phys., 2013,

15, 19853
Gustavo B. Alcantara,a Leonardo G. Paterno,*b Fernando J. Fonseca,c
Marcelo A. Pereira-da-Silva,de Paulo C. Moraisaf and Maria A. G. Solera

Received 25th August 2013,
Accepted 9th October 2013
DOI: 10.1039/c3cp53602e
www.rsc.org/pccp
Multilayered nanocomposite films (thickness 50–90 nm) of cobalt ferrite nanoparticles (np-CoFe2O4,
18 nm) were deposited on top of interdigitated microelectrodes by the layer-by-layer technique in order
to study their dielectric properties. For that purpose, two different types of nanocomposite films were
prepared by assembling np-CoFe2O4 either with poly(3,4-ethylenedioxy thiophene):poly(styrene sulfonic
acid) or with polyaniline and sulfonated lignin. Despite the different film architectures, the morphology
of both was dominated by densely-packed layers of nanoparticles surrounded by polyelectrolytes. The
dominant effect of np-CoFe2O4 was also observed after impedance spectroscopy measurements, which
revealed that dielectric behavior of the nanocomposites was largely influenced by the charge transport
across nanoparticle–polyelectrolyte interfaces. For example, nanocomposites containing np-CoFe2O4
exhibited a single low-frequency relaxation process, with time constants exceeding 15 ms. At 1 kHz, the
dielectric constant and the dissipation factor (tan d) of these nanocomposites were 15 and 0.15, respec-
tively. These values are substantially inferior to those reported for pressed pellets made exclusively of
similar nanoparticles. Impedance data were further fitted with equivalent circuit models from which
individual contributions of particle’s bulk and interfaces to the charge transport within the nanocomposites
could be evaluated. The present study evidences that such nanocomposites display a dielectric behavior
dissimilar from that exhibited by their individual counterparts much likely due to enlarged nanoparticle–
polyelectrolyte interfaces.

Introduction uncontrolled effects may appear, for example the transition

The unique magnetic behavior of iron oxide nanoparticles1 has
been recently implemented in different technological applications,
especially in the biomedical field.2–8 In most of those, nanoparticles
are provided as a colloidal magnetic fluid in which agglomeration
is kinetically prevented by intense repulsive forces, e.g. electrostatic,
steric or electrosteric.9 Despite that, manufacturing of organized 2D
and 3D solid arrays of iron oxide nanoparticles still needs to be
developed further. One of the major concerns in this subject is to
develop strategies to minimize or even completely inhibit the
strong dipolar interactions that arise when magnetic nanoparticles
are brought close together. In fact, their magnetic signature as
individual entities is invariably lost in solid assemblies and
a
Universidade de Brası́lia, Instituto de Fı́sica, Brası́lia DF 70910-900, Brazil
b
Universidade de Brası́lia, Instituto de Quı́mica, Brası́lia DF 70904-970, Brazil.
E-mail: lpaterno@unb.br
c
Universidade de São Paulo, PSI-EPUSP, 05508-900, São Paulo-SP, Brazil
d
Instituto de Fı́sica de São Carlos, USP, 13560-9700, São Carlos-SP, Brazil
e
Centro Universitário Central Paulista – UNICEP, 13563-470, São Carlos-SP, Brazil
f
Department of Control Science and Engineering, Huazhong University of Science
and Technology, Wuhan 430074, China
from the superparamagnetic to the ferrimagnetic state.10,11
Besides the magnetic behavior, the dielectric properties of
iron oxide nanoparticles will also have a significant impact on
more advanced technologies, for example supercapacitors,12
electromagnetic interference shielding,13 MOS capacitors,14
and chemical sensors.15,16 The dielectric behavior of bulk iron
oxides is due to electric dipole moments that are created upon
the charge transfer between di- and trivalent metal cations
within the spinel structure. As a consequence of that, spinel
iron oxides exhibit high electrical resistivities, low eddy currents
and large dielectric losses, especially in the low frequency range.
The mechanisms of charge polarization become even more
complex when Fe(II) is replaced by other divalent metal cations
in mixed spinel ferrites, MFe2O4, where M is a divalent transition-
metal such as Co, Ni, Mn, Cu, Zn, and Cd. In cobalt ferrite, hole
and electron hopping are found to respond in low and high
conductivity regions, respectively.17 Moreover, the cation distribu-
tion within the spinel structure is highly dependent on preparation
conditions.18,19 When the particle size is reduced to the nanoscale,
the M2+ to Fe3+ ratio in a preferred site (generally in the octahedral)
changes and the material’s polarizability may increase or decrease.

This journal is c the Owner Societies 2013 Phys. Chem. Chem. Phys., 2013, 15, 19853--19861 19853

Paper
In fact, for cobalt ferrite, it has been observed that the dielectric
permittivity increases as the particle size decreases while the
dielectric loss has no straightforward correlation with the size
variation.20–22
In order to assess the magnetic and dielectric properties,
iron oxides are preferred provided as solid specimens, such as
pressed pellets and/or films, as in the investigation previously
mentioned.20–22 When the study deals with nanoparticles, the
interfacial contact plays a major role because the surface to volume
ratio is enhanced and inter-particle interactions significantly
Published on 10 October 2013. Downloaded by Laurentian University on 01/11/2013 10:11:11.
increase. Furthermore, most of the methodologies for sample
preparation are based on energy consuming procedures such as
ball milling, sintering, and so on. In face of these difficulties, the
combination of colloidal synthesis and layer-by-layer (LbL) film
assembly23 has recently provided a very simple methodology for
processing iron oxide nanoparticles into ultrathin solid films with
an unprecedented control over magnetic dipolar interactions at
very inexpensive and low energy consuming procedures.24–33 In the
LbL technique, nanoparticles and polyelectrolytes are sequentially
assembled from their liquid solutions onto a chosen solid surface
to form ultrathin films whose main features, including thickness
and morphology, can be fully modulated by external deposition
parameters such as physicochemical properties of deposition
suspensions (pH, ionic strength, nanoparticle and polyelectrolyte
concentrations) and film architecture (number of deposition
cycles, layer spacers, etc.).25–33 By setting-up these parameters
under some defined conditions, the volume fraction of nano-
particles and inter-particle distances within the produced film
can be adjusted and, consequently, magnetic dipolar inter-
actions can be tuned in a wide range of energies.32,33
The magnetic properties of LbL films made of iron oxide
nanoparticles have been reported in some detail but their
dielectric properties remain unvisited. In the present contribution,
we report on the dielectric behavior of ultrathin nanocomposite
films (thickness 50–90 nm) of cobalt ferrite nanoparticles
(np-CoFe2O4) assembled layer-by-layer with polyelectrolytes,
named poly(3,4-ethylenedioxy thiophene):polystyrene sulfonic
acid (PEDOT:PSS), doped polyaniline (PANI), and sulfonated
lignin (SL). For that purpose, impedance spectra of nanocomposites
were registered in a wide frequency range (1 MHz to 1 Hz) and
experimental data were fitted by equivalent circuit models from
which a charge transport mechanism is proposed. In addition, the
film growth and morphology were assessed by UV-vis spectroscopy
and atomic force microscopy, respectively, in order to give
additional support for the charge transport mechanism. The
present study is meant to give a realistic picture of the dielectric
behavior of np-CoFe2O4 while determining individual contributions
of particle’s bulk and interfaces to the charge transport. Moreover,
the study aims at evaluating the potential use of such nano-
composites in energy storage.
Experimental section
Materials
Cobalt(II) nitrate hexahydrate, iron(III) chloride hexahydrate, iron(III)
nitrate nonahydrate, sodium 3-mercapto propanesulfonate
19854 Phys. Chem. Chem. Phys., 2013, 15, 19853--19861
View Article Online
PCCP
(3-MPS), and polyaniline emeraldine base (PANI, Mw 10 000 g mol1)
were purchased from Aldrich (USA) and used as received.
PEDOT:PSS complex dispersion (Cleviost AIP 4083) was purchased
from H. C. Stark and used without additional purification. Sulfo-
nated lignin (SL, Mw 3000 g mol1) was provided by Melbar (Brazil)
in the form of sodium lignosulfonate (Vixilex SD) and used as
received. Perchloric acid, hydrochloric acid, sulfuric acid, sodium
hydroxide, hydrogen peroxide, and ammonium peroxydisulfate
were all analytical grade or better and purchased from Merck
(Germany) and Synth (Brazil). All water used for nanoparticle’s
synthesis, solution preparations and substrate cleaning was
provided by a purification system Milli-Q (water resistivity:
18 MO cm1). Optical glass slides (25  10  1 mm) used as
substrates were cleaned in a piranha solution followed by RCA
solution as described elsewhere.26 Interdigitated microelectrodes
(IMEs) used for electrical measurements were of the same type of
those employed by Wiziack et al.34 They were made of an optical
glass slide (10 mm  25 mm  1 mm) stamped with 50 pairs of
gold digits (50 mm wide, 0.2 mm height, 5 mm long, 10 mm interdigit
spacing). Prior to the nanocomposites’ assembly IMEs were soaked
overnight into an aqueous 3-MPS solution (20 mmol L1) in order
to render their gold-plated surface negatively-charged.
Preparation of CoFe2O4 nanoparticles
Aqueous solutions of Co(NO3)26H2O (0.1 mol L1, 250 mL),
FeCl36H2O (0.2 mol L1, 250 mL) and 40 mL of concentrated
nitric acid were mixed together under magnetic stirring and
heated until boiling (B95 1C).18 The solution’s pH was then
raised to 12 with sodium hydroxide solution (3.0 mol L1)
that immediately led to the formation of a black precipitate
of np-CoFe2O4. The reaction mixture was cooled down to room
temperature and nanoparticles were isolated from the super-
natant by centrifugation. Nanoparticles were then washed
several times with ultrapure water until the resulting super-
natant reached neutral pH. Nanoparticles were then passivated
after suspending them into Fe(NO3)3 (0.5 mol L1) aqueous
solution for 30 min, at boiling temperature.35 The reaction
mixture was cooled down to room temperature and the pH was
raised to 6 by adding sodium hydroxide solution (0.1 mol L1).
The as-passivated np-CoFe2O4 were washed several times with
ultrapure water until the supernatant’s pH was neutral. For the
LbL film deposition, nanoparticles were provided as an ionic
colloidal suspension after dispersing them into aqueous per-
chloric acid solution (0.25 mol L1) followed by subsequent
dialysis against ultra-pure water (several days) and centrifuga-
tion (5000 rpm, 5 minutes). The acidic medium protonates
nanoparticles’ surface sites (mainly the oxyhydroxy groups),
which then establishes an electrostatic repulsion between
nanoparticles. This mechanism ensures the colloidal stability
for several months. The zeta potential of these nanoparticles as
suspended in acidic medium was +37 mV.
Nanoparticle morphology and composition
The morphology and size distribution of np-CoFe2O4 were
accessed by transmission electron microscopy (TEM) employing
a JEOL 1011 microscope operated at 100 keV. Fig. 1 displays the
This journal is c the Owner Societies 2013
 View Article Online

PCCP Paper
Published on 10 October 2013. Downloaded by Laurentian University on 01/11/2013 10:11:11.

Scheme 1 Schematic illustration of different nanocomposite architectures and

Fig. 1 Particle size histogram (vertical bars) of the CoFe2O4 sample. The solid materials employed in depositions.
line represents the best curve fitting using the log-normal distribution function.
The inset shows its typical TEM micrograph.
the properties of LbL films of SL were extensively investigated.36
Secondly, because of its availability and much lower cost than
particle size histogram (vertical bars) of the np-CoFe2O4 and the
conventional polyanions such as sulfonated polystyrene.
inset provides a typical TEM image. The size distribution was
fitted by a log-normal function that provided an average Film thickness, morphology and electrical measurements
particle diameter (DTEM) equal to 18.2  0.2 nm and standard
The nanocomposite’s surface morphology and thicknesses
diameter deviation (s) equal to 0.26  0.02. In agreement with
were assessed by atomic force microscopy (AFM) using a
the JCPDS 22-1086, X-ray diffraction patterns of the powdered
Nanoscope Multimode IIIa atomic force microscope (Veeco
sample (not shown) confirmed the CoFe2O4 phase. Nano-
Instruments Inc.). The film thickness was determined with
particles’ chemical composition was further determined by
the procedure developed by Lobo et al.37 In that, the AFM tip
atomic absorption spectrometry using a Perkin-Elmer 5000
was used to scratch and peel out a small spot of the film
spectrometer from which a 3 : 1 [Fe]/[Co] ratio was found. In
deposited onto a glass substrate. The as-created step was
addition, the saturation magnetization of powdered nano-
further imaged under the tapping mode (Si tip, spring constant
particles was 55 emu g1 (at 20 K).
of 70 N m1, resonance frequency of 325 kHz). Electrical
Layer-by-layer nanocomposite’s deposition measurements were carried out with the experimental setup
described elsewhere,38 which consists of a 1260 Solartron
Nanocomposite’s deposition was carried out by hand with glass
Impedance Analyser controlled by a computer with the aid of
slides and IMEs being alternately dipped into cationic and
a virtual instrument created in LabView. Impedance spectra in
anionic dispersions. The entire deposition process was per-
the range of 10 MHz to 1 Hz were collected while the film
formed as follows: (i) substrates were firstly dipped for 3 min into
sample was interrogated with a 50 mV amplitude signal. During
the cationic np-CoFe2O4 dispersion, 1  106 mol L1, pH 3 or
measurements, samples were conditioned in a glass vessel
PANI solution, 0.1 g L1, pH 3, followed by (ii) rinsing for 30 s with
(100 mL) open to air and whose temperature (25.0  0.1 1C)
an aqueous acidic solution (HCl, pH 3) kept under magnetic
was controlled by an automatic thermal bath. Spectra were
stirring and (iii) drying with N2 flow. The substrate-cationic layer
collected after a conditioning period of 5 minutes. Fitting of
was then dipped for 3 min into the polyanion solution (PEDOT:PSS
raw data was performed with the Z-view software based on
complex or SL solution, 0.1 g L1, pH 3), rinsed and dried as in the
appropriate equivalent circuits which will be presented in the
deposition of the cationic layer. These cycles were sequentially
Results and discussion section.
repeated until the desired number of layers is deposited.
All depositions were conducted at room temperature (25 1C).
The deposition of nanocomposites onto plain glass slides was Results and discussion
monitored by UV-vis spectroscopy (UV 1600 PC Shimadzu).
Three types of nanocomposite architectures were assembled on Nanocomposite’s deposition, morphology, and thickness
top of IMEs: film-I (control): (PANI/SL)10, film-II (np-CoFe2O4/ The layer-by-layer technique provides great control of film
PEDOT:PSS)10, film-III: (PANI/SL/np-CoFe2O4/SL)5. The subscript thickness and morphology, especially when electrolyte materials
refers to the number of adsorbed bi- or tetra-layers, which were are used in the film deposition.23,24 This feature is driven by a
set in order to produce films with comparable thicknesses. charge overcompensation process, under which the upcoming
Film-I, without np-CoFe2O4, was used as a control. Film-III adsorbate species overcompensates the electrostatic charges of
displays a tetra-layer architecture after alternating layers of the underneath substrate.39 As such, the adsorption of a particular
PANI, SL and np-CoFe2O4. Their internal architectures are layer ends at its own because the initial substrate signal is then
schematically depicted in Scheme 1. The choice of SL as the reversed and repels the adsorbate species in the supernatant
polyanion was twofold. Firstly, because in previous contributions, solution. This dynamics ensures that each layer contains a same

This journal is c the Owner Societies 2013 Phys. Chem. Chem. Phys., 2013, 15, 19853--19861 19855

Paper
Published on 10 October 2013. Downloaded by Laurentian University on 01/11/2013 10:11:11.
Fig. 2 UV-vis monitoring of deposition of CoFe2O4 nanoparticle based nano-
composites onto plain glass slides: (a) film-II (np-CoFe2O4/PEDOT:PSS) and
(b) film-III (PANI/SL/np-CoFe2O4/SL). The upward arrows indicate the increase
in number ‘‘n’’ of bi- or tetralayers.
amount of material and also that film thickness increases
linearly with the number of deposition cycles. In the present study
this assumption was verified while monitoring the deposition of
nanocomposites onto plain glass slides by UV-vis spectroscopy.
Fig. 2 provides UV-vis spectra of film-II (np-CoFe2O4/PEDOT:PSS)
and film-III (PANI/SL/np-CoFe2O4/SL) recorded at every deposited
bi- or tetra-layer, respectively. In film-II (Fig. 2a), the spectra of all
bilayers exhibit a subtle shoulder at 460 nm which is ascribed to
the charge transfer between Co2+ and Fe3+ ions in the spinel
structure.40 The spectrum of film-III (Fig. 2b) presents a broad
band with a maximum at 700 nm that is ascribed to the polaron
band of doped PANI.41 The inserted graphics display the absor-
bance increase regarding the electronic transition in np-CoFe2O4
(band at 460 nm). The linear increase of absorbance provides clear
evidence that both nanocomposites grow onto the substrate
following a stepwise regime, i.e., each bi- or tetra-layer contributes
with a same amount of material. Although not shown, plotting
film growth by using the absorption band of PANI at 700 nm in
film-III resulted in the same linear relationship.
Fig. 3 displays topography AFM images of nanocomposites
containing two (np-CoFe2O4/PEDOT:PSS) bilayers (Fig. 3a,
film-II) and two (PANI/SL/np-CoFe2O4/SL) tetra-layers (Fig. 3b,
film-III). In both samples, the presence of nanoparticles is
clearly noted. They spread over the entire film surface and
form a densely-packed coating with a few aggregates. This
feature is shared by different film architectures that include
19856 Phys. Chem. Chem. Phys., 2013, 15, 19853--19861
View Article Online
PCCP
Fig. 3 Topography AFM images of nanocomposite films containing CoFe2O4
nanoparticles. (a) Film-II (np-CoFe2O4/PEDOT:PSS)2 and (b) film-III (PANI/SL/
np-CoFe2O4/SL)2.
iron oxide nanoparticles and polyelectrolytes.25–29,42–43 While
being surface charged, these colloidal particles are readily
transferred from aqueous suspension onto the polyelectrolyte
modified substrate. The electrostatic-driven adsorption is spon-
taneous and reaches equilibrium in a few minutes when a
continuous layer of nanoparticles is formed.26,43 The root mean
square (rms) surface roughness as determined by averaging
of three different film regions was: 3.3  0.2 nm for film-II and
5.1  0.2 nm for film-III. Film-III is rougher because of a few
nanoparticle aggregates (see Fig. 3b). Since the same nanopar-
ticle suspension was employed for deposition of both types of
films and film-II does not exhibit aggregates, the aggregation of
nanoparticles in film-III is caused somehow by the presence of
two polyelectrolytes (PANI and SL).
Values of film thicknesses as measured by AFM were as
follows: film-I (PANI/SL)10: 90  2 nm; film-II (np-CoFe2O4/
PEDOT:PSS)10: 50  7 nm; and film-III (PANI/SL/np-CoFe2O4/
SL)5: 52  7 nm. If one considered that each np-CoFe2O4 is
about 18 nm in diameter, a top-to-top stacking in a 10-bilayer
film would result in a film thickness of roughly 180 nm.
However, as already discussed in our previous studies,26,32,42
upcoming nanoparticles occupy the voids left around nano-
particles of the previous layer and thus make the layer more
compact than that expected for a top-to-top arrangement. The
number of bi- or tetra-layers in each film was set in purpose
just to provide films with comparable thickness. This criter-
ion was chosen in order to permit a straightforward compar-
ison of dielectric properties independently of the film
thickness.
This journal is c the Owner Societies 2013

PCCP
Characterization of the dielectric behavior
As pointed out before, the main goal of this study was to assess
dielectric properties of thin nanocomposite films of np-CoFe2O4
and to evaluate their potential in energy storage. For that purpose,
the electrical properties were investigated by impedance spectro-
scopy. This choice was made because of the vast range of possible
frequencies allowed by impedance spectroscopy, under which
different electric/dielectric phenomena can be assessed and resistive
and capacitive behaviors can be separately detected and analyzed.44
Published on 10 October 2013. Downloaded by Laurentian University on 01/11/2013 10:11:11.
Furthermore, the impedance spectrum can be fitted by the response
of equivalent model circuits whose elements can be associated with
different charge transport mechanisms within the material under
study. The impedance study is extended to the PANI/SL that was
without np-CoFe2O4 and used as a control for comparison.
Fig. 4 displays real (Z 0 ) and imaginary (Z00 ) impedance
spectra of films-I, -II, and -III acquired in air. All spectra exhibit
a single drop in impedance as the frequency increases. This is
ascribed to the polarization relaxation, whose frequency range
depended on the type of nanocomposite film. In film-I the
relaxation is observed above 104 Hz, which is typical of con-
ducting PANI.45,46 For films II and III, the relaxation is seen at
Fig. 4 Impedance spectra of (a) film-I (PANI/SL), (b) film-II (np-CoFe2O4/PED-
OT:PSS), and (c) film-III (PANI/SL/np-CoFe2O4/SL).
This journal is c the Owner Societies 2013
View Article Online
Paper
much lower frequencies, 110 Hz and 70 Hz, respectively, which
is typical of more insulating materials. np-CoFe2O4 nano-
particles turned the films more resistive and in fact, at the
limit of the DC regime, when f - 0, films-II and -III were 2
orders of magnitude more resistive than the control film-I.
The influence of np-CoFe2O4 on the dielectric behavior of
nanocomposite films is clearly seen when impedance data are
displayed in Nyquist diagrams. As presented in Fig. 5, the
Nyquist diagrams of all films are formed by a single semicircle
whose diameter is larger in films containing np-CoFe2O4. In the
case of film-III, the presence of SL makes the film even more
resistive and the semicircle is larger. The experimental curves
were fitted with eqn (1) and (2), which represent the impedance
Fig. 5 Nyquist plots of (a) film-I (PANI/SL); (b) film-II (np-CoFe2O4/PEDOT:PSS),
and (c) film-III (PANI/SL/np-CoFe2O4/SL).
Phys. Chem. Chem. Phys., 2013, 15, 19853--19861 19857

Paper
Published on 10 October 2013. Downloaded by Laurentian University on 01/11/2013 10:11:11.
Scheme 2 Equivalent circuits for (a) eqn (1) and (b) eqn (2).
Table 1 Fitting parameters for the impedance response in air of films I, II, and III,
after eqn (1) and (2) and equivalent circuits of Scheme 2
Sample
Circuit element Film-I Film-II
RS (kO) 0.06 0.05
Cp (nF) 0.06 0.17
Rp (kO) 200 8.6  107
Rnp (kO) — 5.0  103
Cnp (nF) — 0.07
t (ms) 0.0016 16
of equivalent circuits depicted in Scheme 2. Table 1 collects the
respective fitting parameters.
As reported in our previous investigation,32 cross-sectional
TEM images of LbL films of maghemite nanoparticles and
polyaniline revealed that nanoparticles are individually
enclosed by a very thin (2–4 nm) polyelectrolyte layer. Indeed,
these films did not exhibit any kind of anisotropic stratification
despite the fact that they were assembled layer after layer. Thus,
polymer–nanoparticle interfaces are spread throughout the
film bulk. Once the internal structure of the present films
should be similar to that, impedance spectra of films containing
np-CoFe2O4 should reflect contributions from charge transport
across polyelectrolyte and nanoparticle individual phases plus the
charge transport along nanoparticle–polyelectrolyte interfaces.
Thus, the model circuits proposed in Scheme 2 were meant to
represent the charge transport mechanisms suggested above.
For the control film (film-I, PANI/SL), the equivalent circuit
is depicted in Scheme 2a. It contains a single resistor, Rs, which
accounts for the charge (polarons) transport along the polymeric
chain, in series with a RC parallel combination, which accounts for
the charge transfer resistance (Rp) and polarization capacitance
(Cp) of the charge transport across different polymer grains.
This model is in accordance with the conduction mechanism
based on intra and inter-chain charge transports, proposed for
19858 Phys. Chem. Chem. Phys., 2013, 15, 19853--19861
View Article Online
PCCP
conducting polymers.47,48 The overall impedance (Z) of this circuit
is expressed by eqn (1) as follows:
1
Z ¼ RS þ   (1)
joCp þ 1=Rp
pffiffiffiffiffiffiffi
where j ¼ 1 and o (=2pf) is the angular frequency. As to
film-II and film-III, the presence of np-CoFe2O4 is accounted for
by a second RC parallel element, as depicted in Scheme 2b. As
was proposed by Sivakumar et al.,20 the charge transport in
pressed pellets of np-CoFe2O4 is divided into two regimes. The
first, at the high frequency range, deals with the transport
in the nanoparticle’s bulk whereas the second, at the low
frequency regime, deals with the transport along interfaces
between neighbour nanoparticles. For such a model, it was
proposed an equivalent circuit with two elements in series,
where each element is composed of a RC parallel combination.
However, we have found that a single RC element, besides the
one regarded to the polymer, was sufficient to account for
the presence of np-CoFe2O4 in the nanocomposite films since
the fitting was satisfactory, with less than 10% error. This
second RC element expresses the charge transport along the
polyelectrolyte–nanoparticle interface which we shall see later,
and it dominates the impedance spectra of films containing
Film-III np-CoFe2O4. The overall impedance of this second equivalent
circuit is expressed by eqn (2) as follows:
0.06
0.06 1 1
7.0  104 Z ¼ RS þ þ (2)
1.1  102 joCp þ ð1=Rp Þ joCnp þ ð1=Rnp Þ
0.4
23 where Rnp and Cnp represent, respectively, the resistance and
capacitance for charge transport at the polyelectrolyte–nano-
particle interface.
According to Table 1, Rs is almost equal in the three films
with values of about 600 O. This is because the presence of
conducting PEDOT:PSS or PANI in all films. Besides that, film-I
displays Rp of 200 kO and Cp equal to 0.06 nF. All these values
are comparable to those found for thin films of conducting
polymers.49,50 When np-CoFe2O4 are added, there is a signifi-
cant increase in Rp, about five orders (8.6  107 and 7.0  104
for film II and III, respectively). This is because these nano-
particles are insulating and when placed between polymeric
chains block the conduction paths within the nanocomposite.
This is a general effect that has been already observed in LbL
films of PANI and maghemite nanoparticles.26,29 Similarly
to np-CoFe2O4, maghemite is also an electrical insulator. The
Rnp values found were 5 MO and 0.1 MO and Cnp values were
0.07 nF and 0.4 nF, for film-II and film-III, respectively. In
accordance with the literature, the resistance values found
herein are comparable to those found for np-CoFe2O4 (8 nm
diameter) tested in pressed pellets.20 Additionally, relaxation
times (t) associated with films-II and -III were 16 ms and 23 ms,
respectively, in contrast to much shorter 16 ms determined for
film-I. These values are comparable to those determined for
pressed pellets of np-CoFe2O4, as well as for pellets of NiFe2O4
nanoparticles (8–12 nm diameter) in the low frequency
regime.51 These observations allow one to conclude that as
the np-CoFe2O4 are introduced the charge transport in the
This journal is c the Owner Societies 2013
 View Article Online

PCCP Paper

nanocomposite is dominated by polarization/charge accumulation

at polyelectrolyte/nanoparticle interfaces. Since transport at these
interfaces occurs at much lower rates, the overall nanocomposite
impedance is much higher than that presented by the control film-I
which is without nanoparticles. This conclusion is supported by the
down-shift of the relaxation frequencies observed in Fig. 4 and also
by the conduction models proposed by other investigators for iron
oxide nanoparticles in pressed pellets.20,51
The dielectric response of LbL films incorporated with
np-CoFe2O4 was investigated further by inspecting the depen-
Published on 10 October 2013. Downloaded by Laurentian University on 01/11/2013 10:11:11.

dence of the complex dielectric constant (e*) on frequency.

According to eqn (3), the complex dielectric constant is
composed of real e 0 and imaginary e00 parts,

e* = e 0  je00 (3)

and each component is described by

Z 00
e0 ¼ (4a)
oC0 ðZ002 þ Z02 Þ

Z0
e00 ¼ (4b)
oC0 ðZ002 þ Z02 Þ

where C0 is the geometrical capacitance of the electrodes

(= 0.2 nF). The real part, e 0 , represents the energy stored during
the dielectric polarization whereas the imaginary one, e00 , is
referred to the energy spent by the dielectric to align its dipole
moments according to the oscillating electrical field. The ratio
between dissipated and stored energy is referred to as the
dissipation factor or loss tangent (tan d)
e00
tan d ¼ 0 (5)
e
For practical purposes, it is expected for a dielectric material to
exhibit the lowest possible value for tan d at the operation
frequency.
Fig. 6 shows real and imaginary parts of the dielectric
constant as well as the loss tangent of films-I, -II, and -III as a
function of frequency. For the three films, the real component
monotonically decreases with frequency and it seems that
nanoparticles have little effect on that. This behavior is expected
for such nanoparticles because their dielectric constant is
dependent on the hole hopping between cobalt ions which is
quite slow and is not capable of following the alternating electric
field at higher frequencies.17,20 The same seems to be true for
the imaginary counterpart of their dielectric constant and dis-
sipation factor. A similar behavior has been observed for both
nano and microcrystalline cobalt ferrite samples.20,51,52 Mean-
while, film-I, without nanoparticles, exhibits a maximum peak
(at 3 kHz) in the imaginary dielectric constant as well as in the
loss tangent. This is the polarization relaxation of dipoles in the
polyelectrolyte chains. In highly doped PANI, the relaxation
should occur at much higher frequencies, but in the present
system, the interaction with SL chains turns the process slower
and thus shifts relaxation to lower frequency.
Table 2 collects values for real dielectric constant and loss
tangent measured at 1 kHz for the three films. The real
Fig. 6 Dielectric properties of nanocomposites. (a) Real and (b) imaginary parts
of the dielectric constant, and (c) loss tangent.
Table 2 Dielectric parameters of films -I, -II, and -III, determined at 1 kHz
Sample e0 tan d
Film-I 15.7 3.5
Film-II 14.9 0.14
Film-III 14.4 0.12
dielectric constant of the control as well as for nanoparticle
containing nanocomposites is about 15. However, the presence
of nanoparticles in films II and III decreases the dissipation
factor to about two orders in comparison to the control film.
Firstly, the dielectric constant and the dissipation factor found
in nanocomposite films with np-CoFe2O4 are significantly lower
than those measured in pressed pellets (e 0 = 737; tan d = 1, as
found in ref. 15) and thick films (e00 = 130; tan d = 2, as found in
ref. 50). Furthermore, in comparison to the control film, the
presence of np-CoFe2O4 had no effect over e 0 , but it significantly
reduced e00 and, consequently, the dissipation factor as well.
Such a very low dissipation factor makes such nanocomposites
quite promising materials for capacitors. It seems that in the
nanocomposite with nanoparticles the energy stored during
polarization is greater than that needed to align dipoles. This is
probably due to the presence of polyelectrolyte chains around
nanoparticles that can assist the dipole alignment. Therefore,

This journal is c the Owner Societies 2013 Phys. Chem. Chem. Phys., 2013, 15, 19853--19861 19859

Paper
the dielectric behavior of such nanocomposites results from a
synergistic combination of dielectric properties of np-CoFe2O4
and polyelectrolytes plus the enlarged nanoparticle–polyelectrolyte
interfaces, which make them very different from their individual
counterparts. Moreover, these nanocomposites where interfacial
effects are dominating, as a consequence of the film morphology,
can be used as sensing elements in chemical sensors operated
under the ac regime and at room temperature.16
Conclusions
Published on 10 October 2013. Downloaded by Laurentian University on 01/11/2013 10:11:11.
The charge transport across the polyelectrolyte/nanoparticle
interfaces dominates the dielectric behavior of multilayered
nanocomposite films based on cobalt ferrite nanoparticles
(np-CoFe2O4) assembled layer-by-layer with different poly-
electrolytes. The complex permittivity and the dissipation factor
of these films are much smaller than those measured in
pressed pellets made exclusively of np-CoFe2O4. The morphology
provided by the layer-by-layer processing ensures that nano-
particles remain isolated from each other and, consequently,
the respective nanocomposite films exhibit a synergistic dielectric
behavior far apart from their individual components’ properties.
This unique feature, rarely observed in nanocomposites obtained
by other physical or solution processing methods, enables applica-
tions where energy dissipation has to be minimized, for example in
capacitors. Moreover, the dominance of interfacial morphology and
dielectric behavior should find interesting use in chemical sensors.
Acknowledgements
The financial support from Brazilian agencies MCT-CNPq,
CAPES, FAP-DF and FAPESP is gratefully acknowledged.
Notes and references
1 S. Bedanta and W. J. Kleemann, J. Phys. D: Appl. Phys., 2009,
42, 013001.
2 F. Benyettou, Y. Lalatonne, I. Chebbi, M. Di Benedetto,
J. M. Serfaty, M. Lecouvey and L. Motte, Phys. Chem. Chem.
Phys., 2011, 13, 10020–10027.
3 R. Banerjee, Y. Katsenovich, L. Lagos, M. Mclintosh,
X. Zhang and C. Z. Li, Curr. Med. Chem., 2011, 17,
3120–3141.
4 W. R. Viali, G. B. Alcantara, P. P. C. Sartoratto, M. A. G. Soler,
E. Mosiniewicz-Szablewska, B. Andrzejewski and P. C. Morais,
J. Phys. Chem. C, 2010, 114, 179–188.
5 G. Unsoy, S. Yalcin, R. Khodadust, G. Gunduz and
U. Gunduz, J. Nanopart. Res., 2012, 14, 964.
6 J. Huang, X. Zhong, L. Wang, L. Yang and H. Mao, Theranostics,
2012, 2, 86–102.
7 J. Philip and J. M. Laskar, J. Nanofluids, 2012, 1, 3–20.
8 Y. P. Chen, M. Zou, C. Qi, M.-X. Xie, D.-N. Wang, Y.-F. Wang,
Q. Xue, J.-F. Li and Y. Chen, Biosens. Bioelectron., 2013, 39,
112–117.
9 D. Eberbeck, F. Wiekhorst, U. Steinhoff and L. Trahms,
J. Phys.: Condens. Matter, 2006, 18, S2829–S2846.
19860 Phys. Chem. Chem. Phys., 2013, 15, 19853--19861
View Article Online
PCCP
10 S. Morup and E. Tronc, Phys. Rev. Lett., 1994, 72, 3278–3281.
11 D. Fiorani, J. L. Dormann, R. Cherkaoui, E. Tronc, F. Lucari,
F. D’Orazio, L. Spinu, M. Nogues, A. Garcia and A. M. Test,
J. Magn. Magn. Mater., 1999, 196, 143–147.
12 W. Shi, J. Zhu, D. H. Sim, Y. Y. Tay, Z. Lu, X. Zhang,
Y. Sharma, M. Srinivasan, H. Zhang, H. H. Hng and
Q. Yan, J. Mater. Chem., 2011, 21, 3422–3427.
13 A. Caprile, M. Coı̈sson, F. Fiorillo, P. Kabos, O. M. Manu,
E. S. Olivetti, M. A. Olariu, M. Pasquale and
V. A. Scarlatache, IEEE Trans. Magn., 2012, 48, 3394–3397.
14 A. M. Yashchenok, D. A. Gorin, M. Badylevich, A. A.
Serdobintsev, M. Bedard, Y. G. Fedorenko, G. B. Khomutov,
D. O. Grigoriev and H. Mohwald, Phys. Chem. Chem. Phys., 2010,
12, 10469–10475.
15 C. J. Belle, A. Bonamin, U. Simon, J. Santoyo-Salazar,
M. Pauly, S. Begin-Colin and G. Pourroy, Sens. Actuators, B,
2011, 160, 942–950.
16 G. B. Alcantara, L. G. Paterno, F. J. Fonseca, M. A. Pereira-
da-Silva, P. C. Morais and M. A. G. Soler, J. Nanofluids, 2013,
2, 175–183.
17 G. H. Jonker, J. Phys. Chem. Solids, 1959, 9, 165–175.
18 M. A. G. Soler, E. C. D. Lima, S. W. da Silva, T. F. O. Melo,
A. C. M. Pimenta, J. P. Sinnecker, R. B. Azevedo, V. K. Garg,
A. C. Oliveira, M. A. Novak and P. C. Morais, Langmuir, 2007,
23, 9611–9617.
19 S. W. da Silva, F. Nakagomi, M. S. Silva, A. Franco,
V. K. Garg, A. C. Oliveira and P. C. Morais, J. Nanopart.
Res., 2012, 14, 798.
20 N. Sivakumar, A. Narayanasamy, C. N. Chinnasamy and
B. Jeyadevan, J. Phys.: Condens. Matter, 2007, 19, 386201.
21 X. H. Guo, Y. H. Deng, D. Gu, R. C. Che and D. Y. Zhao,
J. Mater. Chem., 2009, 19, 6706–6712.
22 G. B. Sun, B. X. Dong, M. H. Cao, B. Q. Wei and C. W. Hu,
Chem. Mater., 2011, 23, 1587–1593.
23 G. Decher, Science, 1997, 277, 1232–1237.
24 L. G. Paterno and M. A. G. Soler, JOM, 2013, 65, 1–11.
25 M. A. G. Soler, L. G. Paterno and P. C. Morais, J. Nanofluids,
2012, 1, 101–119.
26 L. G. Paterno, M. A. G. Soler, F. J. Fonseca, J. P. Sinnecker,
E. H. C. P. Sinnecker, E. C. D. Lima, M. A. Novak and
P. C. Morais, J. Phys. Chem. C, 2009, 113, 5087–5095.
27 L. G. Paterno, F. J. Fonseca, G. B. Alcantara, M. A. G. Soler,
P. C. Morais, J. P. Sinnecker, M. A. Novak, E. C. D. Lima,
F. L. Leite and L. H. C. Mattoso, Thin Solid Films, 2009, 517,
1753–1758.
28 A. Pereira, S. Alves, M. Casanova, V. Zucolotto and
I. H. Bechtold, J. Nanopart. Res., 2010, 12, 2779–2785.
29 L. G. Paterno, M. A. G. Soler, F. J. Fonseca, J. P. Sinnecker,
E. H. C. P. Sinnecker, E. C. D. Lima, S. N. Báo, M. A. Novak
and P. C. Morais, J. Nanosci. Nanotechnol., 2010, 10,
2679–2685.
30 B. P. Pichon, P. Louet, O. Felix, M. Drillon, S. Begin-Colin
and G. Decher, Chem. Mater., 2011, 23, 3668–3675.
31 V. S. Marangoni, M. V. A. Martins, J. A. Souza, O. N. Oliveira Jr.,
V. Zucolotto and F. N. Crespilho, J. Nanopart. Res., 2012,
14, 769.
This journal is c the Owner Societies 2013

PCCP
32 M. A. G. Soler, L. G. Paterno, J. P. Sinnecker, J. G. Wen,
E. H. C. P. Sinnecker, R. F. Neumann, M. Bahiana, M. A. Novak
and P. C. Morais, J. Nanopart. Res., 2012, 14, 1–10.
33 L. G. Paterno, E. H. C. P. Sinnecker, M. A. G. Soler,
J. P. Sinnecker, M. A. Novak and P. C. Morais, J. Nanosci.
Nanotechnol., 2012, 8, 6672–6678.
34 N. K. L. Wiziack, L. G. Paterno, F. J. Fonseca and L. H. C.
Mattoso, Sens. Actuators, B, 2007, 122, 484–492.
35 T. F. O. Melo, S. W. da Silva, M. A. G. Soler, E. C. D. Lima and
P. C. Morais, Surf. Sci., 2006, 600, 3642–3645.
Published on 10 October 2013. Downloaded by Laurentian University on 01/11/2013 10:11:11.
36 L. G. Paterno and L. H. C. Mattoso, Polymer, 2001, 42,
5239–5245.
37 R. F. M. Lobo, M. A. Pereira-da-Silva, M. Raposo, R. M. Faria
and O. N. Oliveira Jr., Nanotechnology, 1999, 10, 389–393.
38 G. S. Braga, L. G. Paterno and F. J. Fonseca, Sens. Actuators,
B, 2012, 171, 181–189.
39 J. B. Schlenoff, H. Ly and M. Li, J. Am. Chem. Soc., 1998, 120,
7626–7634.
40 U. Schwertmann and R. M. Cornell, in The Iron Oxides in the
Laboratory: Preparation and Characterization, VCH, Wein-
heim, 1991.
41 J. C. Chiang and A. G. MacDiarmid, Synth. Met., 1986, 13,
193–205.
42 G. B. Alcantara, L. G. Paterno, F. J. Fonseca, P. C. Morais and
M. A. G. Soler, J. Magn. Magn. Mater., 2011, 323, 1372–1377.
This journal is c the Owner Societies 2013
View Article Online
Paper
43 G. B. Alcantara, L. G. Paterno, A. S. Afonso, R. C. Faria,
M. A. Pereira-da-Silva, P. C. Morais and M. A. G. Soler, Phys.
Chem. Chem. Phys., 2011, 13, 21233–21242.
44 E. Barsoukov and J. R. Macdonald, in Impedance Spectro-
scopy: Theory, Experiment and Applications, VCH, New Jersey,
2005.
45 R. F. Bianchi, G. F. L. Ferreira, C. M. Lepienski and
R. M. Faria, J. Chem. Phys., 1999, 110, 4602–4607.
46 R. F. Bianchi, H. N. da Cunha, R. M. Faria, G. F. L. Ferreira
and J. M. G. Neto, J. Phys. D: Appl. Phys., 2005, 38, 1437–1443.
47 V. N. Prigodin and A. J. Epstein, J. Phys. B: Condens. Matter,
2003, 338, 310–317.
48 R. S. Kohlman, A. Zibold, D. B. Tanner, G. G. Ihas,
T. Ishiguro, Y. G. Min, A. G. MacDiarmid and
A. J. Epstein, Phys. Rev. Lett., 1997, 78, 3915–3918.
49 M. Irimia-Vladu and J. W. Fergus, Synth. Met., 2006, 156,
1396–1400.
50 D. Zhua, J. Zhang, C. Xu and M. Matsuo, Synth. Met., 2011,
161, 1820–1827.
51 (a) N. Ponpandian, P. Balaya and A. Narayanasamy, J. Phys.:
Condens. Matter, 2002, 14, 3221–3237; (b) G. B. Alcantara,
L. G. Paterno, F. J. Fonseca, M. A. Pereira-da-Silva,
P. C. Morais and M. A. G. Soler, J. Nanofluids, 2013, 2, 175–183.
52 W. Chen, W. Zhu, O. K. Tan and X. F. Chen, J. Appl. Phys.,
2010, 108, 034101.
Phys. Chem. Chem. Phys., 2013, 15, 19853--19861 19861