Applied Surface Science 474 (2019) 66–77

Contents lists available at ScienceDirect

Applied Surface Science

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

Full Length Article

Electrical properties of epoxy nanocomposites containing Fe3O4

nanoparticles and Fe3O4 nanoparticles deposited on the surface of
electrochemically exfoliated and oxidized graphite
Adrian Radoń a, Patryk Włodarczyk b, Aleksandra Drygała a, Dariusz Łukowiec a,⇑
a
Faculty of Mechanical Engineering, Silesian University of Technology, Konarskiego 18 a St., 44-100 Gliwice, Poland
b
Institute of Non-Ferrous Metals, ul. Sowinskiego 5, 44-100 Gliwice, Poland

a r t i c l e i n f o a b s t r a c t

Article history: Electrical conductivity, electromagnetic interference shielding and microwave absorption properties of
Received 4 November 2017 epoxy composites containing magnetite nanoparticles and modified electrochemically exfoliated gra-
Revised 2 May 2018 phite (mEEG) were described in this work. Different nanoparticles were prepared by co-precipitation
Accepted 7 May 2018
method and electrochemical exfoliation. Structure of synthesized Fe3O4 NPs and graphite based nanos-
Available online 12 November 2018
tructures was described in detail. The X-ray diffractometry, Fourier transform infrared spectroscopy
(FTIR), Raman spectroscopy as well as scanning and transmission electron microscopy were used in order
Keywords:
to characterize samples. Fe3O4 nanoparticles with average crystallite size equal to 8.45 nm and 8.30 nm
Electrical conductivity
Jonsher’s universal power law
were obtained in highly agglomerated structure and on the surface of mEEG. Reconstruction of graphite
Microwave absorption structure during precipitation of Fe3O4 on the surface of oxidized, exfoliated graphite was observed. In
Fe3O4 nanoparticles order to describe dispersion of AC electrical conductivity, correlated barrier hopping model was used.
Electrochemical exfoliation It was noticed, that addition of nanoparticles changes the values of hopping energy and activation energy
as well as electromagnetic interference shielding and microwave absorption properties. Epoxy composite
containing Fe3O4 NPs had the highest value of activation energy of electrical conductivity. It was shown,
that the addition of magnetite nanoparticles has also improved electromagnetic interference shielding
and microwave absorption properties of epoxy resin.
Ó 2018 Elsevier B.V. All rights reserved.

1. Introduction ites. To this end, measurements were performed with different fil-
lers concentration and constant temperature (for DC electrical
Until now, the influence of nanoparticles (NPs) on electrical conductivity) and at different temperatures and in wide frequency
properties (permittivity, conductivity) of nanocomposites based range but in constant fillers fraction. Based on the percolation the-
on various polymers e.g. epoxy, PMMA, polyurethane, polyaniline ory critical filler concentration (percolation threshold), in which a
or poly(ethylene oxide) was studied [1–5]. Different nanomaterials jump in DC electrical conductivity is observed, can be determined
as clays (for example montmorillonite), graphite oxide, carbon [9,10]. Frequency dependent electrical conductivity of nanocom-
nanotubes, TiO2 NPs, Al2O3 NPs, Fe3O4 NPs were incorporated in posites can provide information about conduction mechanism.
polymer matrix [1,4–6]. Recently, the electrical conductivity of Jonscher’s power law (universal power law) can be applied to
different polymer based nanocomposites containing carbon based determine not only DC electrical conductivity, but also type of con-
fillers (as carbon nanotubes, carbon black, graphene nanoplates) duction mechanism [8,11,12]. Different conduction mechanism
was studied in detail [7,8]. Addition of nanoparticles changes not models as non-overlapping small polaron tunneling (NSPT), over-
only value of DC electrical conductivity, but also the value of AC lapping large polaron tunneling (OLPT), quantum mechanical tun-
electrical conductivity. To determine these changes, two different neling (QMT) and correlated barrier hopping (CBH) were proposed
theories were applied to analyze electrical conductivity of compos- for different materials [13–16]. What is more, one material can be
characterized by two or more conduction mechanisms in different
temperature range [17]. Additionally, interactions between indi-
⇑ Corresponding author. vidual nanoparticles and polymer matrix change the mechanism
E-mail addresses: adrian-radon@wp.pl (A. Radoń), patrykw@imn.gliwice.pl (P. of composite electrical conductivity, what was presented earlier
Włodarczyk), aleksandra.drygala@polsl.pl (A. Drygała), dariusz.lukowiec@polsl.pl by Siddabattuni et al. [18]. The authors have reported, that an
(D. Łukowiec).

https://doi.org/10.1016/j.apsusc.2018.05.045
0169-4332/Ó 2018 Elsevier B.V. All rights reserved.
 A. Radoń et al. / Applied Surface Science 474 (2019) 66–77 67

interaction between polymer matrix and particles has a critical black powder was next filtered using qualitative filter paper and
role in altering the dielectric properties of a composite. Therefore, washed by distilled water, acetone and ethanol. Dried powder
modification of particles surface can significantly change the elec- has been further modified. For this purpose, 0.7 g of EEG was added
trical conductivity and dielectric properties of polymer based into 100 ml solution containing 3 g of K2CO3 and sonication pro-
nanocomposites. Based on the measurements of electric permittiv- cess was carried out by 30 min. Afterwards, 3 g of dextrin was
ity and magnetic permeability, electromagnetic interference (EMI) added to solution and solution was sonicated by 30 min. 10 ml of
shielding properties and microwave absorption (MA) properties of 98% H2SO4 was added dropwise to this solution and 30 ml of
nanocomposites can be determined [19–21]. EMI shielding proper- H2O2 (30%) was finally added to the suspension. After 30 min of
ties of metals, conductive polymers, carbides and carbon based sonication black powder was received by centrifugation, washed
nanomaterials are generally associated with two different mecha- by distilled water, acetone and ethanol.
nisms of shielding. For conductive materials such as intrinsically
conducting polymers (ICP) and ICP based nanocomposites [22–
2.3. Nanocomposite mEEG/Fe3O4 synthesis
24], metals and composites containing metal nanowires [25,26]
and conduction thin layers [27] shielding mechanism is based on
Nanocomposite mEEG/Fe3O4 was prepared by co-precipitation
the reflection. The secondary mechanism is associated with
of Fe3O4 nanoparticles in solution containing of mEEG. For this pur-
absorption of electromagnetic waves by materials with high
pose, solution containing 200 ml of deionized water, 0.25 g of
dielectric permittivity or high magnetic permeability [28]. There-
mEEG, 0.406 g of FeCl3, 0.350 g of FeSO4
7H2O and 2 ml of oleic
fore, in order to improve EMI and MA properties, magnetic
acid was sonicated by 30 min in ultrasonic bath and heated up to
nanoparticles as Fe3O4 NPs, different type of carbon fillers (carbon
55 °C. Afterwards, 20 ml of 2.5 M NaOH was added dropwise to this
black, multiwall carbon nanotubes, graphene nanoplatelets, gra-
solution. The sonication process was carried out for 10 min after
phene oxide) and ZnO nanoparticles are embedded in dielectric
the addition of NaOH. Obtained nanocomposite was next filtered
polymer matrix [20,28–32]. Composites containing both Fe3O4
and washed by distilled water, acetone and ethanol.
NPs and carbon based nanomaterials can be prepared by different
methods e.g. chemical vapor deposition and electrophoretic depo-
sition [33]. In this study Fe3O4 nanoparticles have been prepared 2.4. Epoxy composites preparation
by co-precipitation method and additionally deposited on the sur-
face of prepared modified electrochemically exfoliated graphite Four different composites containing constant percentage of fil-
(mEEG). The main aim of this study is to check influence of differ- ler (Fe3O4 NPS, mEEG, mEEG/Fe3O4 and mixture of mEEG and
ent nanoparticles, as well as the modification of the surface of Fe3O4) were prepared. Based on the experience, 3 wt% of filler
mEEG on the electrical conductivity, electromagnetic interference
(EMI) shielding properties and microwave absorption properties.

2. Materials and methods

In this study co-precipitation method and electrochemical exfo-

liation were used to synthesize three different nanomaterials:
magnetite nanoparticles (Fe3O4 NPs), modified electrochemically
exfoliated graphite (mEEG) and nanocomposite mEEG/Fe3O4. All
the used chemicals were of analytical reagent-grade.

2.1. Fe3O4 nanoparticles synthesis

Fe3O4 nanoparticles were synthesized by co-precipitation

method according to the procedure reported before by us [34]
using oleic acid as surfactant, stoichiometric amount of iron salts
and excess of sodium hydroxide. For this purpose solution contain-
ing 100 ml of deionized water, 3.244 g FeCl3, 2.780 g FeSO4
7H2O,
16 ml of oleic acid and 2 ml of HCl (36%) was prepared and heated
up to 55 °C in ultrasonic bath. Afterwards, 50 ml of 1.6 M NaOH
solution was added dropwise to this solution. Finally, 20 ml of
2.5 M NaOH solution was added dropwise to complete precipita-
tion. The sonication process was carried out for 10 min after the
addition of NaOH. The black powder obtained in the process was
collected by vacuum filtration using qualitative filter paper and
washed by acetone, distilled water and ethanol.

2.2. mEEG synthesis

Electrochemical exfoliation of polycrystalline graphite was con-

ducted in solution containing 200 ml of deionized water, 5 g of
H2SO4 (98%) and 1 g of KClO3. Graphite rods were used as cathode
and anode. Electrochemically exfoliated graphite were prepared by
applying current with voltage equal to 6.5 V and intensity equal to
1.8 A. Process was carried out by 2 h. After that, the product was Fig. 1. XRD patterns of synthesized materials: (a) mEEG, (b) Fe3O4 NPs and mEEG/
sonicated in post-reaction solution by 1 h in ultrasonic bath. The Fe3O4.
68 A. Radoń et al. / Applied Surface Science 474 (2019) 66–77

was selected as the optimal. As a polymer matrix, pure epoxy resin 2.5. Materials characterization
(EPIDIAN 5) was chosen. To prepare composites 0.099 g of different
nanoparticles was added to the 3 g of epoxy resin and sonication Structure and morphology of synthesized nanoparticles (Fe3O4,
process were carried out by 10 min. Weight ratio of mEEG to the mEEG and mEEG/Fe3O4) were determined based on X-ray diffrac-
Fe3O4 (mEEG:Fe3O4) was calculated basing on the theoretical effi- tion (XRD) patterns, Raman and Fourier transform infrared (FTIR)
ciency of Fe3O4 synthesis on the surface of mEEG and it was spectra, transmission electron microscope (TEM) images, scanning
equaled to 0.86. Based on this proportion, epoxy composite con- electron microscope (SEM) images and Energy-dispersive X-ray
taining 3% of the mixture of mEEG and Fe3O4 was prepared using spectra (EDS). TEM images and selected area (electron) diffraction
0.0459 g of mEEG and 0.0531 g of Fe3O4. After sonication, 0.33 g patterns (SAED) were obtained using Transmission Electron Micro-
of triethylenetetramine was added to the samples with continuous scope S/TEM TITAN 80–300. Suspension of nanoparticles in ethanol
stirring. Two different types of composites were prepared to deter- was prepared for this purpose and samples were prepared on cop-
mine electrical and magnetic properties in the form of discs and per grid with carbon film. SEM images and EDS spectra were
toroids. obtained using Scanning Electron Microscope Supra 35. The images

Table 1
The results of analysis of XRD patterns.

Sample analyzed diffraction peak d-spacing [nm] average crystallite size [nm] number of layers [a.u.]
mEEG (0 0 2) 0.336 12.8 38
Fe3O4 NPs (3 1 1) 0.252 8.45 –
mEEG/Fe3O4 (0 0 2)/(3 1 1) 0.336/0.251 28.5/8.30 84

Fig. 2. STEM and HRTEM images and SAED patterns of Fe3O4 NPs (a–d), mEEG (e–h) and mEEG/Fe3O4 (i–h); (a) STEM image of highly agglomerated Fe3O4 NPs, (b) HRTEM
image of Fe3O4 NPs with marked area, which was enlarged in (c), (c) observed structure of Fe3O4 along to the [1 1 0] zone axis with model of unit cell, (d) SAED pattern of
Fe3O4 with marked Miller indices, (e) STEM image of mEEG, (f) HRTEM image of mEEG with marked area, which was enlarged in (g), (g) multilayered structure of mEEG with
marked interlayered distance between graphene layers, (h) SAED pattern of mEEG with marked Miller indices, (i) STEM image of mEEG/Fe3O4, (j) HRTEM image of mEEG/
Fe3O4 with marked areas, which were enlarged in (k) – blue marked area and in (l) – red marked area, (k) multilayered structure of mEEG with marked distance between
graphene layers, (l) distance between (3 1 1) lattice planes characteristic for Fe3O4, (m) SAED pattern of mEEG/Fe3O4. (For interpretation of the references to color in this
figure legend, the reader is referred to the web version of this article.)
 A. Radoń et al. / Applied Surface Science 474 (2019) 66–77 69

were taken at 20.00 kV voltage, 12.9 working distance and SE2 type identified on XRD patterns of mEEG and mEEG/Fe3O4. Characteris-
of signal (type II secondary electrons). FTIR spectra were recorded tic for magnetite (Fd-3m space group) diffraction peaks were
for each material at room temperature using spectrophotometer observed on XRD patterns of Fe3O4 NPs and nanocomposite.
FTIR Nicolet 6700/8700. Measurements were conducted in the Absence of additional peaks confirm phase purity of synthesized
transmission mode (KBr pellet method) in the mid infrared range material.
of 4000–400 cm1. Raman spectra were recorded using Renishaw’s The Bragg’s equation was applied to (0 0 2) diffraction peak for
Raman in Via Reflex spectrometer, equipped with the Leica evaluating distance d between graphene layers. Based on Scherrer’s
Research Grade confocal microscope. Excitations were made by equation (shape factor equals 0.9) number of layers in synthesized
ion-argon laser with a beam with a wavelength k = 514 nm and mEEG was calculated as well as crystallite size of Fe3O4 NPs (shape
with plasma filter for 514 nm. The phase purity and the crystal factor equals 0.94). Obtained results have been presented in
structures were determined by powder X-ray diffraction using X- Table 1. It can be noticed, that in both synthesis methods, Fe3O4
ray diffractometer Rigaku MiniFlex 600 with a copper tube Cu Ka nanoparticles have similar average crystallite sizes and that the
(k = 0.15406 nm). Concept 81 dielectric spectrometer equipped number of graphene layers in mEEG increases after precipitation
with Alpha and Keysight 4991E analyzers and Novocool tempera- of Fe3O4 on the surface. This effect can be associated with restoring
ture control system was used to determine dielectric, magnetic, of graphite structure, what was also observed in Raman spectra
EMI shielding and MA properties of epoxy composites. Measure- and discussed later.
ment were carried out in the frequency range 100 mHz–1 MHz Presence of Fe3O4 and graphite was additionally confirmed by
and temperatures from 223 K to 343 K. Additionally, to determine analysis of SAED patterns presented on Fig. 2d, h and m. The STEM
EMI shielding and MA properties measurements were performed image of Fe3O4 NPs (Fig. 2a) clearly show highly agglomerated
at ambient temperature in the frequency range 10 MHz–2.2 GHz. structure obtained by simple co-precipitation method. Addition-
ally, on HRTEM image (Fig. 2b) structure of Fe3O4 along to the
[1 1 0] zone axis has been identified. Layered structure of mEEG
3. Results and discussion
was observed both on STEM image (Fig. 2e) and HRTEM image
(Fig. 2f). Distance between layers identified in Fig. 2g is typical
3.1. Structure analysis
for interlayered distance in graphite. Morphology of synthesized
nanocomposite – mEEG/Fe3O4 was presented in Fig. 2i. Small, dis-
XRD patterns of Fe3O4 NPs, mEEG and mEEG/Fe3O4 were pre-
persive magnetite nanoparticles can be observed on the surface of
sented on Fig. 1a and b. The broadening of diffraction peaks
mEEG. Additionally, presence of Fe3O4 NPs was confirmed by anal-
observed for nanoparticles and nanocomposite is characteristic
ysis of HRTEM image (Fig. 2j). Two different regions were shown
for nanostructures. Graphite structure (P 63m c space group) was

Fig. 3. (a) SEM image of mEEG, (b) SEM image of mEEG/Fe3O4 and corresponding results of SEM-EDS elemental mapping analysis performed for area visible on (b) and
presented on (c–e).
70 A. Radoń et al. / Applied Surface Science 474 (2019) 66–77

enlarged on Fig. 2k and l and are associated with presence of graphene surface and with removal of functional groups during
graphite-like structure (interlayer distance equals 0.340 nm) and co-precipitation of Fe3O4 on the surface of mEEG [41]. Addition-
with presence of Fe3O4 on the surface of mEEG (distance between ally, presented in Fig. 4b HRTEM image of mEEG confirm exis-
(3 1 1) lattice planes in Fe3O4 equals 0.252 nm). Additionally, in tence of disordered structure and honeycomb lattice structure
Fig. 3 SEM images of mEEG and mEEG/Fe3O4 have been presented. characteristic for graphene.
The fragmented structure of mEEG was presented in Fig. 3a. To The FTIR spectra of synthesized nanomaterials have been pre-
confirm existence of Fe3O4 on the surface of mEEG, SEM-EDS ele- sented in Fig. 4c and d. It can be noticed, that the electrochemical
mental mapping analysis was performed for nanocomposite and exfoliation and modification of polycrystalline graphite allowed to
the results corresponding to the structure presented in Fig. 3b were obtain nanostructure with oxygen containing functional groups. In
shown in Fig. 3c–e. FTIR spectrum of mEEG, the vibrations associated with existence
Highly disordered carbon structure in mEEG and mEEG/Fe3O4 CAO bonds (800–1300 cm1 and at 1399 cm1), C@O bonds (at
was confirmed by Raman spectra analysis, which was presented 1399 cm1, 1640 cm1 and at 1725 cm1) have been identified.
in Fig. 4. Four characteristic for disordered structure in Presence of water in sample was confirmed by the existence of
graphite-like systems bands (D, G, 2D and D + G) were observed broad peak in wavenumber region 3000–3700 cm1 and with
for both materials. The D (1351 cm1) and D + G (2935 cm1) the presence of sharp peak at 1640 cm1. Additionally, this broad
bands are related to the disordered structure and can be peak can be associated with existence of alkoxy and carboxyl
observed among others in Raman spectra of graphene oxide groups [42]. The presence of Fe3O4 was confirmed by analysis of
and reduced graphene oxide. The G band (1578 cm1 for mEEG, FTIR spectra of Fe3O4 NPs and mEEG/Fe3O4 (Fig. 4d). Identified
1584 cm1 for mEEG/Fe3O4) is associated with stretching of sp2 peaks at 578 cm1 and at 580 cm1 are characteristic for FeAO
carbon atoms (in rings and in chains) and appears for all bond. For a bulk magnetite this peak appears at 570 cm1, but
graphite-base materials. The observed shift of this band can be for nanoparticles, shifts to higher wavenumbers can be observed,
associated with higher concentration of defects in mEEG/Fe3O4. which is associated with size of NPs and is consistent with the
Sensitive to the p band in the electronic structure of graphite results of the XRD patterns analysis. The peak which is observed
2D mode also appears for both materials at 2713 cm1 [35–37]. at 452 cm1 is also associated with FeAO bond [34]. The sharp
Additionally, Raman spectrum of mEEG/Fe3O4 confirms presence peaks identified at 1640 cm1, 1410 cm1, 2850 cm1 and at
of Fe3O4 (A1g mode at 673 cm1) [38–40]. The change of value 2920 cm1 are associated with existence of oleic acid, which was
of ID/IG ratio (0.75 for mEEG and 1.01 for mEEG/Fe3O4) can be used as surfactant and are related to the COO and CH2 stretching
associated with interaction between Fe3O4 nanoparticles and vibrations.

Fig. 4. (a) Raman spectra of mEEG and mEEG/Fe3O4, (b) HRTEM image of mEEG with visible disordered structure and honeycomb lattice structure characteristic for graphene,
(c) FTIR spectrum of mEEG, (d) FTIR spectra of Fe3O4 and mEEG/Fe3O4.
 A. Radoń et al. / Applied Surface Science 474 (2019) 66–77 71

3.2. Electrical conductivity and relaxation processes in epoxy Additionally, AC electrical conductivity, which occurs at higher
composites temperature range and which is associated with glass transition
follows the Jonscher’s power law given by [44]:
AC electrical conductivity of prepared composites containing 3
wt% of different nanoparticles in wide temperature range 223–333 rAC ¼ Að2pf Þs þ rDC ð1Þ
K has been presented in Fig. 5. It can be noticed, that rAC is both,
temperature and frequency dependent. Two different processes – where A is the characteristic parameter, f – frequency and s is fre-
a and b relaxation can be observed. These processes are associated quency exponent dependent on temperature and frequency with
with relaxation of the glyceryl groups (b relaxation) and with glass values in the range from 0 for ideal ionic-type crystals to 1 for ideal
transition (a relaxation) [43]. Debye dielectric dipolar-type crystals. The experimental rAC values

Fig. 5. AC electrical conductivity of pure epoxy resin (a) and different epoxy composites (insets show range associated with b relaxation) with: (b) Fe3O4 NPs, (c) mEEG, (d)
mEEG/Fe3O4, (e) Fe3O4 + mEEG.
72 A. Radoń et al. / Applied Surface Science 474 (2019) 66–77

Fig. 6. The experimental rac values fit well Jonscher’s power law in high temperature region for pure epoxy resin (a) and epoxy composites with: (b) Fe3O4 NPs, (c) mEEG, (d)
mEEG/Fe3O4, (e) Fe3O4 + mEEG; (f) s(T) behavior.

were fitted to Jonscher’s power law which was shown in Fig. 6a–e. in which exponent s is temperature dependent and decreases with
Basing on the s(T) behavior, conduction mechanism in disordered increasing temperature, was chosen to describe electrical conduc-
solids can be identified and described. Four different models were tivity in prepared epoxy composites. According to this model, fre-
successfully applied to describe electrical conductivity in disor- quency exponent s can be defined as:
dered solids: non-overlapping small polaron tunneling (NSPT)
model, overlapping large polaron tunneling (OLPT) model, quantum 6kT
s¼1
 ð2Þ
mechanical tunneling (QMT) model and correlated barrier hopping W H  kTln x1s0
(CBH) model [45]. Basing on the s(T) (Fig. 6f) behavior, CBH model,
 A. Radoń et al. / Applied Surface Science 474 (2019) 66–77 73

Table 2 structure during co-precipitation of Fe3O4 and was confirmed ear-

The values of hopping energy and activation energy of electrical conductivity lier by analysis of Raman spectra and XRD patterns. The addition of
determined for pure epoxy resin and different epoxy composites.
Fe3O4 NPs significantly increases value of activation energy. The
Sample WH [eV] at 341 K Ea [eV] lowest value of Ea was observed for composite containing mixture
Pure epoxy 0.48 1.09 eV of mEEG and Fe3O4. What is more, at 321 K value of rDC is constant
Epoxy + Fe3O4 0.52 1.67 eV for all composites and probably is associated with glass transition
Epoxy + mEEG 0.59 1.41 eV of epoxy resin – above this temperature mobility of different
Epoxy + Fe3O4 + mEEG 0.70 0.83 eV
Epoxy + mEEG/Fe3O4 0.45 1.07 eV
nanoparticles in matrix increases and the value of DC electrical
conductivity increases significantly. Presented in Fig. 8a–e dielec-
tric losses (e00 ) also confirm existence of a and b relaxation in ana-
lyzed frequency range. Moreover, addition of different
nanoparticles changes the value of e00 what was presented in
Fig. 8f. The highest changes can be observed approximately 20 K
below glass transition. It can be noticed, that the addition of syn-
thesized nanomaterials changes the values of electrical conductiv-
ity and dielectric properties. As has been shown in the literature,
addition of different carbon-based nanomaterials (such as gra-
phene oxide, reduced graphene oxide and carbon nanotubes) sig-
nificantly increases electrical conductivity [48,49]. However,
presented results indicate that this phenomenon does not occur
for mEEG, mEEG/Fe3O4 and mixture of mEEG and Fe3O4.

3.3. EMI shielding properties and microwave absorption properties

Electromagnetic interference shielding properties and micro-

wave absorption properties were determined for all nanocompos-
ites in wide frequency region 10 MHz–2.2 GHz. For this purpose,
dielectric loss tangent tande (expressed as ratio of imaginary part
of complex permittivity to real part of complex permittivity) and
magnetic loss tangent tandl (expressed analogous as imaginary
Fig. 7. Arrhenius plots for pure epoxy resin and epoxy composites with marked part of complex permeability to real part of complex permeability)
glassy transition, in which DC conductivity is constant for all materials. were measured and presented in Fig. 9a and b. It can be noticed,
that addition of different nanoparticles changes the value of dielec-
tric loss tangent, but increase of tande in high frequency region can
And for large values of WH/kT can be expressed as [46]: be observed only for composites containing mEEG and mEEG/
Fe3O4. More visible changes of magnetic loss tangent were
6kT observed for all composites. Addition of not only magnetic
s¼1 ð3Þ
WH nanoparticles (as Fe3O4, mEEG/Fe3O4 and mEEG + Fe3O4) but also
pure mEEG increases the value of tandl. Additionally, according
where: WH is the hopping energy, k is the Boltzmann constant, T is
to the shielding theory, shielding effectiveness (SE) and reflection
the temperature, x is the circular frequency (x = 2pf) and s0 is the
loss (RL) were calculated for all nanocomposites. Shielding effec-
characteristic relaxation time equaled to 1013 s. Basing on Eqs. (2)
tiveness is defined as the ratio of transmitted power to incident
and (3) hopping energy for all composites at T = 341 K was calcu-
power, while reflection loss expresses the ratio of reflected power
lated and presented in Table 2.
to incident power. These parameters can be expressed as [19,50]:
Additionally, activation energy of electrical conductivity was
calculated according to the Arrhenius relation [47] in high temper- RL ¼ 20 log jCj ð5Þ
ature region and also presented in Table 2 and in Fig. 7.

 
   1  C2 T 
Ea  
rDC ¼ r0 exp  ð4Þ SE ¼ 20 log  
2 2 
ð6Þ
kT  1T C 

where r0 is the pre-exponential factor associated with the charge qffiffiffiffiffiffiffiffiffiffiffi

pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ðl jl Þ
carrier mobility and density of states, Ea is the activation energy, ðe jeÞ
tanh j 2pcfd
ðl  jl Þðe  je Þ  1
k is the Boltzmann constant and T is the temperature. C ¼ qffiffiffiffiffiffiffiffiffiffiffi
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ð7Þ
ðl jl Þ
It can be noticed, that the addition of nanoparticles changes the ðe jeÞ
tanh j 2pcfd ðl  jl Þðe  je Þ þ 1
value of WH, rDC and activation energy, what was presented on
 2pfdpffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Arrhenius plots in Fig. 7 and in Table 2. What is more, addition
T ¼ e j c ðl jl Þðe je Þ
ð8Þ
of Fe3O4 and mEEG increases slightly the value of WH and increases
significantly the value of Ea. Contrary to this result, simultaneous where C is the reflection coefficient, T is the transmission coeffi-
addition of both Fe3O4 and mEEG increases the value of WH but cient, e0 and e00 are the real and imaginary part of complex permit-
decreases activation energy. The lowest value of WH and relatively tivity, l0 and l00 are the real and imaginary part of complex
low activation energy (lower than pure epoxy resin) was observed permeability, f is the frequency of electromagnetic radiation, c is
for nanocomposite containing mEEG/Fe3O4. The most visible the speed of light and d is the absorber thickness.
changes in the value of rDC can be observed for nanocomposite The results of analysis were presented in Fig. 9c–f. It can be
containing mEEG/Fe3O4 (in this case rDC is even higher than rDC noticed, that the addition of different nanoparticles results in more
observed for composite with pure mEEG). This phenomena is asso- negative values of RL and SE, but the highest changes were
ciated with reduction of mEEG and reconstruction of graphite observed for composite containing 3 wt% of magnetic Fe3O4
74 A. Radoń et al. / Applied Surface Science 474 (2019) 66–77

Fig. 8. Imaginary part of complex permittivity for pure epoxy resin (a) and epoxy composites with: (b) Fe3O4 NPs, (c) mEEG, (d) mEEG/Fe3O4, (e) Fe3O4 + mEEG; (f) influence
of temperature and addition of different nanoparticles on the value of e00 .

nanoparticles. It is concluded that the dielectric and magnetic sented in Fig. 9e and f, which were obtained for epoxy + Fe3O4
losses are enhanced when the mass fraction of the magnetic composite clearly show, that the values of RL and SE are related
nanoparticles increases, which may result in the improvement of to the absorber thickness – the most negative value of RL and SE
the microwave absorption properties. However, it was previously can be observed for incredibly thick absorber (d = 1 m), but also
reported, that the maximum of absorption of microwave radiation absorber with thickness equaled to 10 cm can find application.
for composites with Fe3O4 NPs and carbon-based nanostructures For 10 cm absorber, there are 3 different maxima observed in RL
occurs in higher frequencies and the preparation of absorbers for plot (at 375 MHz, 1.3 GHz and 2.13 GHz) and SE is maintained
the lower frequencies is difficult [51–53]. Additionally, results pre- for such system below 10 dB.
 A. Radoń et al. / Applied Surface Science 474 (2019) 66–77 75

Fig. 9. (a) Dielectric loss tangent, (b) magnetic loss tangent, (c) RL and (d) SE for pure epoxy resin and epoxy composites calculated for 1 cm absorber; (e) RL for epoxy + Fe3O4
NPs composite with different thickness, (f) SE for epoxy + Fe3O4 NPs composite with different thickness.

4. Conclusions The restoration of graphite structure was observed by analysis of

XRD patterns, Raman spectra and electrical conductivity of epoxy
Three different types of nanoparticles were obtained by simple composites. Correlated barrier hopping model was chosen to
and low cost methods. Co-precipitation method and electrochem- describe electrical conductivity of epoxy composites with 3 wt%
ical exfoliation allowed to obtain Fe3O4 NPs, mEEG and mEEG/ of Fe3O4 NPs, mEEG, mEEG/Fe3O4 and mixture of Fe3O4 NPS with
Fe3O4. Modified electrochemically exfoliated graphite was charac- mEEG. It was noticed, that the addition of different nanoparticles
terized by highly disordered structure. The presence of oxygen changes not only value of hopping energy but also activation
containing functional groups on the surface of mEEG was con- energy calculated with use of Arrhenius equation. The highest acti-
firmed by analysis of FTIR spectra. These functional groups take a vation energy had composite with Fe3O4 NPs and the highest value
part in the precipitation of Fe3O4 NPs on the surface of mEEG. of hopping energy had composite with mixture of mEEG and Fe3O4.
76 A. Radoń et al. / Applied Surface Science 474 (2019) 66–77

Additionally, it was observed, that the addition of nanoparticles [14] A. Ladhar, M. Arous, H. Kaddami, M. Raihane, A. Kallel, M.P.F. Graça, L.C. Costa,
AC and DC electrical conductivity in natural rubber/nanofibrillated cellulose
improves the EMI shielding and MA properties of epoxy resin.
nanocomposites, J. Mol. Liq. 209 (2015) 272–279, https://doi.org/10.1016/
The composite containing only Fe3O4 NPs had the most negative j.molliq.2015.04.020.
values of RL and SE. Further research associated with optimization [15] M. Vellakkat, D. Hundekal, Electrical conductivity and supercapacitor
of the Fe3O4 NPs/epoxy resin weight ratio and modification of mor- properties of polyaniline/chitosan/nickel oxide honeycomb nanocomposite, J.
Appl. Polym. Sci. 134 (2017), https://doi.org/10.1002/app.44536.
phology and structure of these nanoparticles may allow to develop [16] O.G. Abdullah, Y.A.K. Salman, S.A. Saleem, Electrical conductivity and dielectric
methods for the production of composites that can be successfully characteristics of in situ prepared PVA/HgS nanocomposite films, J. Mater. Sci.
used as paints to protect electrical equipment. Moreover, pre- Mater. Electron. 27 (2016) 3591–3598, https://doi.org/10.1007/s10854-015-
4196-4.
sented low cost and simple method synthesis of magnetite [17] A. Kahouli, A. Sylvestre, F. Jomni, B. Yangui, J. Legrand, Experimental and
nanoparticles and nanocomposite can be used on an industrial theoretical study of AC electrical conduction mechanisms of semicrystalline
scale. parylene C thin films, J. Phys. Chem. A. 116 (2012) 1051–1058, https://doi.org/
10.1021/jp207114u.
[18] S. Siddabattuni, T.P. Schuman, F. Dogan, Dielectric properties of polymer-
Conflicts of interest particle nanocomposites influenced by electronic nature of filler surfaces, ACS
Appl. Mater. Interfaces 5 (2013) 1917–1927, https://doi.org/10.1021/
am3030239.
None. [19] F. Qin, C. Brosseau, A review and analysis of microwave absorption in polymer
composites filled with carbonaceous particles, J. Appl. Phys. 111 (2012),
Acknowledgments https://doi.org/10.1063/1.3688435.
[20] Y. Chen, Y. Wang, H. Bin Zhang, X. Li, C.X. Gui, Z.Z. Yu, Enhanced
electromagnetic interference shielding efficiency of polystyrene/graphene
A. Drygała and D. Łukowiec are acknowledged for a statutory composites with magnetic Fe3O4 nanoparticles, Carbon N.Y. 82 (2015) 67–
grant from the Faculty of Mechanical Engineering of the Silesian 76, https://doi.org/10.1016/j.carbon.2014.10.031.
[21] S.P. Pawar, D.A. Marathe, K. Pattabhi, S. Bose, Electromagnetic interference
University of Technology in 2017 with contract number 10/990/ shielding through MWNT grafted Fe3O4 nanoparticles in PC/SAN blends, J.
BK-17/0064. P. Wlodarczyk is acknowledged for a OPUS grant from Mater. Chem. A. 3 (2015) 656–669, https://doi.org/10.1039/C4TA04559A.
the National Science Center, Poland with contract number UMO- [22] H.R. Tantawy, D.E. Aston, J.R. Smith, J.L. Young, Comparison of electromagnetic
shielding with polyaniline nanopowders produced in solvent-limited
2016/23/B/ST8/03405. conditions, ACS Appl. Mater. Interfaces 5 (2013) 4648–4658, https://doi.org/
10.1021/am401695p.
References [23] M.S. Kim, H.K. Kim, S.W. Byun, S.H. Jeong, Y.K. Hong, J.S. Joo, K.T. Song, J.K. Kim,
C.J. Lee, J.Y. Lee, PET fabric/polypyrrole composite with high electrical
conductivity for EMI shielding, Synth. Met. 126 (2002) 233–239, https://doi.
[1] S. Singha, M. Thomas, Dielectric properties of epoxy nanocomposites, IEEE
org/10.1016/S0379-6779(01)00562-8.
Trans. Dielectr. Electr. Insul. 15 (2008) 12–23, https://doi.org/10.1109/T-
[24] P. Saini, V. Choudhary, B.P. Singh, R.B. Mathur, S.K. Dhawan, Polyaniline-
DEI.2008.4446732.
MWCNT nanocomposites for microwave absorption and EMI shielding, Mater.
[2] B. Sannakki, A. Anita, Dielectric properties of PMMA and its composites with
Chem. Phys. 113 (2009) 919–926, https://doi.org/10.1016/
ZrO2, Phys. Procedia (2013) 15–26, https://doi.org/10.1016/j.
j.matchemphys.2008.08.065.
phpro.2013.10.006.
[25] G.A. Gelves, M.H. Al-Saleh, U. Sundararaj, Highly electrically conductive and
[3] A. Plyushch, J. Macutkevic, P.P. Kuzhir, J. Banys, V. Fierro, A. Celzard, Dielectric
high performance EMI shielding nanowire/polymer nanocomposites by
properties and electrical conductivity of flat micronic graphite/polyurethane
miscible mixing and precipitation, J. Mater. Chem. 21 (2011) 829, https://
composites, J. Nanophotonics 10 (2015) 12511, https://doi.org/10.1117/1.
doi.org/10.1039/c0jm02546a.
JNP.10.012511.
[26] K. Ji, H. Zhao, J. Zhang, J. Chen, Z. Dai, Fabrication and electromagnetic
[4] I. Bekri-Abbes, E. Srasra, Solid phase mechanochemical synthesis of
interference shielding performance of open-cell foam of a Cu-Ni alloy
polyaniline-montmorillonite nanocomposite using grinded montmorillonite
integrated with CNTs, Appl. Surf. Sci. 311 (2014) 351–356, https://doi.org/
as oxidant, Mater. Sci. Semicond. Process. 56 (2016) 76–82, https://doi.org/
10.1016/j.apsusc.2014.05.067.
10.1016/j.mssp.2016.07.020.
[27] F. Shahzad, M. Alhabeb, C.B. Hatter, B. Anasori, S. Man Hong, C.M. Koo, Y.
[5] R.J. Sengwa, S. Choudhary, Dielectric and electrical properties of PEO-Al2O3
Gogotsi, Electromagnetic interference shielding with 2D transition metal
nanocomposites, J. Alloys Compd. 701 (2017) 652–659, https://doi.org/
carbides (MXenes), Science 353 (2016) 1137–1140, https://doi.org/
10.1016/j.jallcom.2017.01.155.
10.1126/science.aag2421.
[6] G.V. Pham, A.T. Trinh, T.X.H. To, T.D. Nguyen, T.T. Nguyen, X.H. Nguyen,
[28] S.U.D. Khan, M. Arora, M.A. Wahab, P. Saini, Permittivity and electromagnetic
Incorporation of Fe3O4/CNTs nanocomposite in an epoxy coating for corrosion
interference shielding investigations of activated charcoal loaded acrylic
protection of carbon steel, Adv. Nat. Sci. Nanosci. Nanotechnol. 5 (2014) 35016,
coating compositions, J. Polym. 2014 (2014) 1–7, https://doi.org/10.1155/
https://doi.org/10.1088/2043-6262/5/3/035016.
2014/193058.
[7] P.-C. Ma, M.-Y. Liu, H. Zhang, S.-Q. Wang, R. Wang, K. Wang, Y.-K. Wong, B.-Z.
[29] X. Liu, Z. Zhang, Y. Wu, Absorption properties of carbon black/silicon carbide
Tang, S.-H. Hong, K.-W. Paik, J.-K. Kim, Enhanced electrical conductivity of
microwave absorbers, Compos. Part B Eng. 42 (2011) 326–329, https://doi.org/
nanocomposites containing hybrid fillers of carbon nanotubes and carbon
10.1016/j.compositesb.2010.11.009.
black, ACS Appl. Mater. Interfaces 1 (2009) 1090–1096, https://doi.org/
[30] M. Giorcelli, P. Savi, M. Miscuglio, M.H. Yahya, A. Tagliaferro, Analysis of
10.1021/am9000503.
MWCNT/epoxy composites at microwave frequency: reproducibility
[8] A.A. Khurram, S.A. Rakha, P. Zhou, M. Shafi, A. Munir, Correlation of electrical
investigation, Nanoscale Res. Lett. 9 (2014) 168, https://doi.org/10.1186/
conductivity, dielectric properties, microwave absorption, and matrix
1556-276X-9-168.
properties of composites filled with graphene nanoplatelets and carbon
[31] S.C. Tjong, Polymer Composites with Carbonaceous Nanofillers: Properties and
nanotubes, J. Appl. Phys. 118 (2015) 44105, https://doi.org/10.1063/
Applications, 2012, doi: 10.1002/9783527648726.
1.4927617.
[32] A. Yakubu, Z. Abbas, N.A. Ibrahim, A. Fahad, Reduction of electromagnetic
[9] H. Pang, C. Chen, Y. Bao, J. Chen, X. Ji, J. Lei, Z.M. Li, Electrically conductive
interference using zno-pcl nanocomposites at microwave frequency, Adv.
carbon nanotube/ultrahigh molecular weight polyethylene composites with
Mater. Sci. Eng. 2015 (2015), https://doi.org/10.1155/2015/132509.
segregated and double percolated structure, Mater. Lett. 79 (2012) 96–99,
[33] M. Gholampoor, F. Movassagh-Alanagh, H. Salimkhani, Fabrication of nano-
https://doi.org/10.1016/j.matlet.2012.03.111.
Fe3O4 3D structure on carbon fibers as a microwave absorber and EMI
[10] W. Park, J. Hu, L.A. Jauregui, X. Ruan, Y.P. Chen, Electrical and thermal
shielding composite by modified EPD method, Solid State Sci. 64 (2017) 51–61,
conductivities of reduced graphene oxide/polystyrene composites, Appl. Phys.
https://doi.org/10.1016/j.solidstatesciences.2016.12.005.
Lett. 104 (2014), https://doi.org/10.1063/1.4869026.
[34] A. Radoń, A. Drygała, Ł. Hawełek, D. Łukowiec, Structure and optical properties
[11] D.K. Pradhan, R.N.P. Choudhary, B.K. Samantaray, Studies of dielectric
of Fe3O4 nanoparticles synthesized by co-precipitation method with different
relaxation and AC conductivity behavior of plasticized polymer
organic modifiers, Mater. Charact. 131 (2017), https://doi.org/10.1016/
nanocomposite electrolytes, Int. J. Electrochem. Sci. 3 (2008) 597–608,
j.matchar.2017.06.034.
https://doi.org/10.1016/j.matchemphys.2009.01.008.
[35] S. Some, Y. Kim, Y. Yoon, H. Yoo, S. Lee, Y. Park, H. Lee, High-quality reduced
[12] B.M. Greenhoe, M.K. Hassan, J.S. Wiggins, K.A. Mauritz, Universal power law
graphene oxide by a dual-function chemical reduction and healing process,
behavior of the AC conductivity versus frequency of agglomerate
Sci. Rep. 3 (2013) 1929, https://doi.org/10.1038/srep01929.
morphologies in conductive carbon nanotube-reinforced epoxy networks, J.
[36] R. Beams, L. Gustavo Cançado, L. Novotny, Raman characterization of defects
Polym. Sci. Part B Polym. Phys. 54 (2016) 1918–1923, https://doi.org/10.1002/
and dopants in graphene, J. Phys. Condens. Matter 27 (2015) 83002, https://
polb.24121.
doi.org/10.1088/0953-8984/27/8/083002.
[13] M. Megdiche, C. Perrin-Pellegrino, M. Gargouri, Conduction mechanism study
[37] G. Khurana, N. Kumar, R.K. Kotnala, T. Nautiyal, R.S. Katiyar, Temperature
by overlapping large-polaron tunnelling model in SrNiP2O7 ceramic
tuned defect induced magnetism in reduced graphene oxide, Nanoscale 5
compound, J. Alloys Compd. 584 (2014) 209–215, https://doi.org/10.1016/
(2013) 3346, https://doi.org/10.1039/c3nr34291c.
j.jallcom.2013.09.021.
 A. Radoń et al. / Applied Surface Science 474 (2019) 66–77 77

[38] A.M. Jubb, H.C. Allen, Vibrational spectroscopic characterization of hematite, diphosphate compound, Appl. Phys. A Mater. Sci. Process. 120 (2015) 1537–
maghemite, and magnetite thin films produced by vapor deposition, ACS Appl. 1543, https://doi.org/10.1007/s00339-015-9353-3.
Mater. Interfaces 2 (2010) 2804–2812, https://doi.org/10.1021/am1004943. [47] T. Arfin, F. Mohammad, DC electrical conductivity of nano-composite
[39] Q. Tian, Q. Wang, Q. Xie, J. Li, Aqueous solution preparation, structure, and polystyrene-titanium-arsenate membrane, J. Ind. Eng. Chem. 19 (2013)
magnetic properties of nano-granular ZnxFe3-xO4 ferrite films, Nanoscale Res. 2046–2051, https://doi.org/10.1016/j.jiec.2013.03.019.
Lett. 5 (2010) 1518–1523, https://doi.org/10.1007/s11671-010-9672-4. [48] R. Atif, I. Shyha, F. Inam, Mechanical, thermal, and electrical properties of
[40] M. Cai, H. Qian, Z. Wei, J. Chen, M. Zheng, Q. Dong, Polyvinyl pyrrolidone- graphene-epoxy nanocomposites–a review, Polymers (Basel) 8 (2016), https://
assisted synthesis of a Fe3O4/graphene composite with excellent lithium doi.org/10.3390/polym8080281.
storage properties, RSC Adv. 4 (2014) 6379, https://doi.org/10.1039/ [49] M. Monti, M. Rallini, D. Puglia, L. Peponi, L. Torre, J.M. Kenny, Morphology and
c3ra45993d. electrical properties of graphene-epoxy nanocomposites obtained by different
[41] T. Peik-See, A. Pandikumar, L.H. Ngee, H.N. Ming, C.C. Hua, Magnetically solvent assisted processing methods, Compos. Part A Appl. Sci. Manuf. 46
separable reduced graphene oxide/iron oxide nanocomposite materials for (2013) 166–172, https://doi.org/10.1016/j.compositesa.2012.11.005.
environmental remediation, Catal. Sci. Technol. 4 (2014) 4396–4405, https:// [50] W. Zhang, Y. Xu, L. Yuan, J. Cai, D. Zhang, Microwave absorption and shielding
doi.org/10.1039/C4CY00806E. property of composites with FeSiAl and carbonous materials as filler, J. Mater.
[42] M. Acik, G. Lee, C. Mattevi, A. Pirkle, R.M. Wallace, M. Chhowalla, K. Cho, Y. Sci. Technol. 28 (2012) 913–919, https://doi.org/10.1016/S1005-0302(12)
Chabal, The role of oxygen during thermal reduction of graphene oxide studied 60150-9.
by infrared absorption spectroscopy, J. Phys. Chem. C 115 (2011) 19761– [51] L. Yang, H. Cai, B. Zhang, S. Huo, X. Chen, Enhanced microwave absorption
19781, https://doi.org/10.1021/jp2052618. property of epoxy nanocomposites based on PANI@Fe3O4@CNFs nanoparticles
[43] M. Pomes-Hadda, M. Aufray, S. Diaham, Interval analysis: a new tool for the with three-phase heterostructure, Mater. Res. Exp. 5 (2018), https://doi.org/
characterization of an epoxy-amine/aluminum system, Polymers (Basel) 7 10.1088/2053-1591/aaab8d.
(2015) 644–654, https://doi.org/10.3390/polym7040644. [52] J. Guo, H. Song, H. Liu, C. Luo, Y. Ren, T. Ding, M.A. Khan, D.P. Young, X. Liu, X.
[44] J. Kolte, P.H. Salame, A.S. Daryapurkar, P. Gopalan, Impedance and AC Zhang, J. Kong, Z. Guo, Polypyrrole-interface-functionalized nano-magnetite
conductivity study of nano crystalline, fine grained multiferroic bismuth epoxy nanocomposites as electromagnetic wave absorbers with enhanced
ferrite (BiFeO3), synthesized by microwave sintering, AIP Adv. 5 (2015), flame retardancy, J. Mater. Chem. C 5 (2017) 5334–5344, https://doi.org/
https://doi.org/10.1063/1.4931818. 10.1039/C7TC01502J.
[45] R. Ben Said, B. Louati, K. Guidara, S. Kamoun, Thermodynamic properties and [53] L. Zhu, X. Zeng, M. Chen, R. Yu, Controllable permittivity in 3D Fe3O4/CNTs
application of CBH model in the ac conductivity of LiNi1.5P2O7 ceramic, Ionics network for remarkable microwave absorption performances, RSC Adv. 7
(Kiel) 20 (2014) 1071–1078, https://doi.org/10.1007/s11581-013-1060-5. (2017) 26801–26808, https://doi.org/10.1039/C7RA04456A.
[46] Y. Ben Taher, A. Oueslati, N.K. Maaloul, K. Khirouni, M. Gargouri, Conductivity
study and correlated barrier hopping (CBH) conduction mechanism in