New Electromagnetic Shielding Materials Based on [611772]

New Electromagnetic Shielding Materials Based on
Viscose-Carbon Nanotubes Composites
Madalina Elena Culica,1Gabriela Biliuta,1Razvan Rotaru,1Gabriela Lisa,2Raluca Ioana Baron,1Sergiu Coseri1
1Petru Poni Institute of Macromolecular Chemistry of Romanian Academy, 41A, Grigore Ghica Voda Alley, 700487 Iasi,
Romania
2Faculty of Chemical Engineering and Environmental Protection, Gheorghe Asachi Technical University, 73 Prof.
dr. docent Dimitrie Mangeron Street, 700050, Iasi, Romania
A convenient approach for the preparation of cellulose –
carbon nanotubes (CNT) hybrid materials owning electro-
magnetic shielding properties, based on viscose (V) and
TEMPO-oxidized viscose fibers (VO) is proposed. Viscose –
carbon nanotubes (V-CNT) and TEMPO-oxidized viscose -carbon nanotubes (VO-CNT) composites were prepared by
embedding carbon nanotubes on the surface of two types
of cellulose fibers, that is, viscose and its C
6-oxidized deriv-
ative. The chemical composition, morphology, and thermal
stability of the prepared hybrid materials were thoroughlyinvestigated by Fourier transform infrared spectroscopy
(FTIR), scanning electron microscopy (SEM), and ther-
mogravimetric analyses. Moreover, electrical properties of
the original and composite fibers were assessed.
POLYM.
ENG. SCI., 00:000 –000, 2019. © 2019 Society of Plastics Engineers
INTRODUCTION
The hybrid materials based on natural resources, that is, cellulose
have gain a lot of interest, because these materials can be successfullyintroduced in sustainable, biodegradable, and eco-friendly products.The properties of these composit es depend on the properties of the
constituents, their composition ratio, as well as on the preparation
method. In the recent years, several studies have focused on compos-ites made of polysaccharides loaded with carbon nanotubes (CNT)[1, 2]. Among polysaccharides, ce llulose plays a privilege role due to
its largest availability, renewability , and good mechanical properties.
Viscose fibers, obtained from regenerated cellulose, are attractive,
readily available, nontoxic, biodegradable, biocompatible, and an
exceptional candidate for a wide range of applications [3, 4]. In manycases, the functionalization of hydroxyl groups of cellulose chainusing different chemical protocols such as esteri fication, etheri fication,
or oxidation is considered, in order to increase the versatility of thecellulose to achieving desired p roperties. The conversion of the
hydroxyl to carboxyl groups is often employed to gain new function-
alities in the cellulose chain, especi ally by using the water-soluble sta-
ble nitroxyl radical 2,2,6,6-tetramethylpyperidine-1-oxyl (TEMPO)[5–8], but also involving the nonpers istent nitroxyl radicals, as we
have recently reported [6, 9, 10]. Semiconductive viscose fibers can
be prepared by layer by layer assembly, chemical plating, electrolessplating, in situ polymerization, atomic laye r deposition, and the dip-
ping and drying method, and so forth. [11 –14]. However, thesemethods use rather complicated pro cedures, causing severe pollution
issues due to utilization of heavy metals.
CNT, discovered in 1991 [15], attracted extensive attention
to researchers, due to their thermal, electrical, and mechanicalproperties, their unique structure, and narrow size distribution,in nanometer range, highly accessible surface area, low resistiv-
ity, and high stability [16, 17]. These exceptional properties of
CNT have been investigated for devices such as field-emission
displays, scanning probe micros copy tips, and microelectronic
devices [18 –21]. Viscose fibers – CNT composites combine the
advantages of the high conductivity performance of CNT withrelatively low-cost and easily processed organic polymers. Thecomposites of cellulose and multiwalled CNT have a good elec-
trical conductivity, as has been recently reported [22]. These
composites are useful in fabrication devices acting as protectivetools at electromagnetic interferences. Conducting polymerscomposites with their unique combination of electrical, dielec-tric, magnetic, thermal, and mechanical properties are bene ficial
for suppression of electromagnetic noises (electromagnetic
shielding). Electromagnetic shielding refers to the re flection
and/or adsorption of electromagnetic radiation by a material,which acts as a shield against the penetration of the radiationthrough the shield.
When an electromagnetic wave encounters an obstacle, radia-
tion can be re flected and/or refracted and/or absorbed (Fig. 1). E
and H represents the power (P) of electromagnetic field (electric
and magnetic field intensity), σ,μ,εrepresents constants of the
shield material (electric conductivity, electric permittivity, andmagnetic permeability) [23].
Three different mechanisms, namely re flection loss (R), absor-
bance loss (A), and internal re flections loss (or multiple internal
reflection-MR), may contribute to overall attenuation of the elec-
tromagnetic field. Mathematically shielding effectiveness (SE) can
be expressed in logarithmic scale (Eq. (1)) [24, 25]:
SE dB ½/C138=S E
R+S E A+S E MR= 10 log10Pt=Pi ðȚ
= 20 log10Et=Ei ðȚ = 20 log10Ht=Hi ðȚð1Ț
where SERis the shielding effectiveness for re flection, SEAis the
shielding effectiveness for absorption, SEMRis the shielding effec-
tiveness for multiple re flection, Pt,Piis the power of wave ( i=
incident wave, t= transmitted wave), Etori,Htoriis the intensity
of electric, magnetic field.
If the shield is thin or the material is made by nano fibers, nan-
otubes, nanometric pyramidal structures, the re flected wave from
the second boundary is re-re flected from the first boundary and
returns to the second boundary to be re flected again and again.
This means internal re flections or multiple internal re flections andAdditional Supporting Information may be found in the online version of this
article.
Correspondence to : S. Coseri; e-mail: coseris@icmpp.ro
DOI 10.1002/pen.25149
Published online in Wiley Online Library (wileyonlinelibrary.com).© 2019 Society of Plastics Engineers
POLYMER ENGINEERING AND SCIENCE —2019

the attenuation ( SEMR) can be mathematically expressed as in
Eq. (2) [26]:
SEMR= 20 log101−e−2t=δ/C16/C17
ð2Ț
For dielectric material [24, 25]:
δ=2=ωμrσ ðȚ1=2ð3Ț
Where ωis angular frequency ( ω=2πυ),υis electric field fre-
quency, μis the magnetic relative permeability of the shield mate-
rial,σis the conductivity of shield material (measured in S/m),
andtis the shield thickness.
For shields made of dielectric materials absorption loss occurs
because currents induced in the medium produce ohmic lossesand heating of the material so in this case E
tand Htcan be
expressed as below [26]:
Et=Eie−t=δð4Ț
Ht=Hie−t=δð5Ț
With Eqs. (1), (4) and (5), we can write the magnitude of
absorption term SEA:
SEAdB½/C138= 20 log10Ht=Hi ðȚ =−20 t=δðȚ log10e=−8,68 t =δðȚ
=−8,68 t σωμr=2 ðȚ1=2ð6Ț
σ=ωε0ε00ð7Ț
Where ε0is free space (vacuum) permittivity: 8,854 /C110−12F/m,
ε00is absorption permittivity.
These parameters of shield material (t, σ,μr) directly affect the
obtained value of electromagnetic shielding ( SEAcomponent of
total SE).
The re flection loss ( SER) is related to the relative impedance
mismatch between the shield ’s surface and propagating wave and
is speci fic for conductive materials with approximately 1 Ω/cmconductivity. The magnitude of this re flection loss under plane
wave can be expressed as below (Eq. 8) [24 –26]:
SERdB½/C138=−10 log10σ=16ωε0μr ðȚ ð 8Ț
A literature survey indicated a large variety of composites used
for electromagnetic shielding. Thus Melvin et al. obtained a hybrid
nanocomposites barium titanate/carbon nanotube with calculatedreflexion loss of −29.6 dB (for a shield thickness t=1m m )[ 2 7 ] .
Abbas et al. prepared polyaniline/barium titanate/carbon-based com-posites with maximum loss of −25 dB (for a 2.5-mm thick sample)
at 11.2 GHz and band width of 2.7 GHz [28]. Rotaru et al. preparedin various compositions through an ultrasound/microwave-assisted
procedure viscose/barium titanate composites with maximum
shielding obtained for frequencies between 10
4and 105Hz from
−18 dB to −22 dB [29].
Herein, we purpose to produce new flexible and biobased com-
posites by incorporation of multiwalled CNT in viscose fibers
(V) or its C 6-oxidized analogue (VO), as electromagnetic
shielding materials, in particular for electric field within industrial
frequency (50 –55 Hz).
EXPERIMENTAL
Materials
Viscose fibers (V), (Lenzing AG, Austria; linear density, 1.3
dtex; fiber length, 39 mm), multi-walled CNT (Timesnano, China)
with diameters of less than 9 nm, lengths of 5 μm, 95% carbon
purity. The other used chemicals: 2,6,6-tetramethylpiperidine-
1-yl)-oxyl radical (TEMPO), sodium periodate, sodium bromide,
9% (wt) sodium hypochlorite, were purchased from Sigma-Aldrich and used without further puri fication.
Synthesis of the Oxidized Viscose
Viscose fibers (5 g) was suspended in 700 mL of distilled
water under vigorous stirring, following by the addition ofTEMPO (0.08 g, 0.5 mmol) and NaBr (0.5 g, 5 mmol). 9%
NaClO solution (1.89 g, 25 mmol) was added to the viscose fibers
slurry under continuous stirring for 1 h at room temperature. Dur-ing this time, the pH value was carefully maintained at about10 by adding 2 m NaOH solution. The oxidation reaction wasstopped by adding a 5 mL of ethanol. The oxidized viscose fibers
separated by filtration were vacuum drying at 40
/C14C for 24 h.
Preparation of the Viscose/Oxidized Viscose-Embedded CNT
The device involved in the processes of preparation was ultra-
sound generator Sonics Vibracell (750 W nominal electric power,20 kHz ultrasound frequencies, display for information of theenergy delivered to the end of the probe, sensor for temperature).The CNT powder (0.2519 g) was dispersed in 200 mL Milli-Qultrapure water by sonication for 60 min. The CNT dispersion liq-
uid and the viscose/oxidized viscose (3 g) were mixed for 60 min
at room temperature. Then, the viscose/oxidized viscose was putinto the coagulation bath containing 5% Na
2SO4and 5% H 2SO4
(100 mL) for 10 min. The viscose/oxidized viscose was washedwith distilled water several times, divided into two fractions, andfinally one fraction is dried in air at room temperature for 24 h
(V-CNT1, VO-CNT1), and the second fraction was dried in vac-
uum at 120
/C14C for 20 min (V-CNT2 and VO-CNT2).
FIG. 1. Schematic representation of shielding mechanism (i: Incident, r: Re flected,
t: Transmitted).
2 POLYMER ENGINEERING AND SCIENCE —2019 DOI 10.1002/pen

Characterization Methods
X-Ray Photoelectron Spectroscopy (XPS). The compositional
analysis of the viscose (unoxidized and oxidized) samples was
carried out by X-ray photoelectron spectroscopy (XPS) using aPHI-5000 VersaProbe photoelectron spectrometer ( ΦULVAC-
PHI, INC.) with a hemispherical energy analyzer (0.85 eV bind-ing energy resolution for organic materials). A monochromatic AlKαX-ray radiation (h ν= 1,486.7 eV) was used as excitation
source. The standard take-off angle used for analysis was 45
/C14,
producing a maximum analysis depth in the range of 3 –5n m .
Spectra were recorded from at least three different locations oneach sample, with a 1 mm ×1 mm area of analysis. Low-
resolution survey spectra were recorded in 0.5 eV steps with117.4 eV analyzer pass energy. In addition, high-resolution car-bon (1 s) spectra were recorded in 0.1 eV steps with 58.7 eV ana-
lyzer pass energy.
FT-IR Analysis. Infrared absorption spectra of original viscose,
oxidized viscose, CNT, and oxidized viscose treated with CNT
were recorded using a Bruker Vertex 70 spectrometer at a scanrange from 4,000 to 650 cm
−1, at a resolution of 2 cm−1and
32 scans. Samples were measured as a KBr pellet.
Raman Spectroscopy. Raman spectra of the vibrational were
recorded at a temperature of 22/C14C, using a Renishaw InVia
Reflex spectrometer with a radiation source formed by a GaAlAs
diode laser, having wavelength 785 nm and energy 10 mW on the
sample. The analyses were done at 180/C14backscattering geometry
with respect to the incident beam.
Scanning Electron Spectroscopy (SEM)
Scanning electron spectroscopy (SEM) was performed by
means of colloidal silver on copper supports. All samples were
covered with a thin layer of gold by sputtering (EMITECH K550×). The coated surface was further studied using an Environ-
mental Scanning 200 instrument, working at 5 kV with secondaryelectrons in high vacuum mode.
Thermogravimetric Analysis
The thermal analysis was performed using TGA-SDTA 851e
Mettler Toledo system. The analyses were done by using of3–5 mg of sample. The samples were heated in temperature range
from 25 to 700
/C14C. The heating rate was 10/C14C/min. Analyses were
made in inert atmosphere, acquired by a continuous nitrogen flow
of 20 mL/min.
Dielectrically Properties and Shielding Calculation
The dielectric constants of the samples were measured on a
Concept 40 Novocontrol Dielectric Spectrometer in a frequencyrange of 10
6–100Hz, at room temperature, with silver electrodes.
The samples were prepared as pellets with the thickness of
1–1.5 mm. We used in the calculation of SE formulae (2) and (3)
for multiple re flection loss ( SEMR) and formula (6) and (7) for the
magnitude of absorption term ( SEA). We did not take into account
reflection term ( SER), because the precursors and the composites
are insulated materials and not conductive. In Eq. (3), the mag-netic relative permeability is equal to one (material without mag-
netic properties). Graphs were processed with the Origin Pro8 program based on data obtained from analysis of dielectric spec-
troscopy (dielectric constant, ε’and dielectric loss, ε”).
RESULTS AND DISCUSSION
The introduction of carboxyl groups at the fibers ’surface could
be evidenced by using XPS technique. XPS technique allows the
study of elemental composition study of a large variety of poly-
saccharides materials. The first XPS experiment was done to per-
form a low-resolution scan of the original and TEMPO-oxidizedfibers in order to determine the percentages of carbon and oxygen
(Fig. 2 and Table 1). It can be seen that the O/C ratio increasesfrom 0.66 in the original sample to 0.74 in the oxidized sample,as a result of introduction of one more oxygen atom originating
from carboxylic groups. Supplemental information can be
acquired from a high-resolution scan, performed on the C 1sregion. The high-resolution C 1s peak provide details about(a) types and (b) amounts of carbon –oxygen bond which are pre-
sent, see Fig. 2 and Table 1. According with the experimentalresults, the chemical shift of carbon (C 1s) in XPS could be
divided as C1, unoxidized carbon (C
C); C2, carbon with one
oxygen bond (C O); C3, carbon with two oxygen bonds
(OCOo rC O); and C4, carbon with three oxygen bonds
(OCO). For the four categories, we have found the peaks at
284.6, 286.4, 287.7, and 288.3 eV, respectively.
After the characterization of the oxidized fibers, the prepara-
tion of the composites has been performed. Figure 3 shows sche-
matically the typical process for fabrication of the fibers-CNT
composites. Due to the incorporation of the CNT at the fibers sur-
face, the original color of the fibers is changed, becoming darker.
The inhomogeneous appearance of the prepared samples isexplainable due to the lack of uniformity of the fiber’s surface in
the case of unoxidized viscose and also the randomly introduction
of COOH groups at the super ficial level in the case of oxidized
viscose fibers.
Characterization of the Composites
FTIR Analysis. The FTIR technique can be used as a straight-
forward method to evaluate the structural changes, which have
occurred in the viscose after oxidation and embedding of CNT. In
these conditions, the carboxylic group formed from TEMPO oxi-dation appears in the FTIR spectrum, between 1,700 and1,750 cm
−1. Figure 4 shows the FTIR spectra of the viscose (V),
oxidized viscose (VO), CNT, and the composites V-CNT1, V-CNT2, VO-CNT1, and VO-CNT2. In the FTIR spectrum ofV/VO, the broadband vibration of OH groups are found in
3445 cm
−1, and the stretching for aliphatic C-H bonds in
2892 cm−1. The symmetric bending of CH 2and C-O groups of
the pyranose ring of V/VO are found, respectively, at 1441 and1,377 cm
−1. In the range of 1,200 –1,000 cm−1, the symmetric and
asymmetric stretching of ether bond (C OC) are assigned. The
absorption peak at 1016 cm−1corresponds to the C O ether
groups. After oxidation, for the sample VO can be observed, the
presence of a new peak at 1735 cm−1, absorption attributed to the
CO stretching frequency of carboxyl groups in their acidic form.
The band is absent in the initial samples (V) and V-CNT compos-ites (V-CNT1 and V-CNT2) but can be detected in an oxidizedsample (VO) and VO-CNT composites (VO-CNT1 and VO-CNT2). Thus, as we can see the CNT composite curves does not
generate new groups, the infrared activity of the CNT being too
DOI 10.1002/pen POLYMER ENGINEERING AND SCIENCE —2019 3

low. Nevertheless, at 2892 and 1,643 cm−1, the peaks of the CNT
composites (V-CNT1 and VO-CNT1) is a very slow increasecompared with that the V and VO. This is due to the increasingwater bound in the micro-dissolution process (the sample V-CNT1 and VO-CNT1 were dried in air at room temperatures for24 h compared with the sample V-CNT2 and VO-CNT2 that were
dried in vacuum at 120
/C14C, for 20 min). Also, the change in inten-
sity of the peak at 3445 cm−1may be due to the formation of
intermolecular hydrogen bond between CNT and V/VO.
Raman Spectroscopy. Raman spectroscopy is a successfully, non-
destructive technique, largely used in the last years to characterize
carbon-based materials, including CNT. For CNT-based composites,Raman spectroscopy is a helpful analysis performed to assess the
level of CNT dispersion, as well as the degree of interactionbetween the carbonaceous material and organic matrix, based on
peaks shifting or width changes [30, 31]. The Raman spectrum of
the CNT presents two main bands: one located around 1,593 cm
−1
(Gband) coresponding to the C C band in-plane vibration, and
another one at 1301 cm−1(Dband) proieminent in disordered car-
bon systems (see Fig. 5 and Supporting Information section). Thereis also preset a weak band at 2594 cm
−1(called G’band), which is
the overtone of the Dband.
The Raman spectrum of viscose fibers (see Supporting Infor-
mation) is composed by two main regions: 1500 –900 cm−1and
600–300 cm−1.T h e first region is located with bands of skele-
tal, symmetric and asymmetric glycosidic ring (1,092, and1,121 cm
−1), methylene bending, rocking, and wagging (bands
at 1469, 1374, 1,267, and 900 cm−1). In the second region,
mainly CCC and COC ring deformation bands could be
d e t e c t e d ,a t4 5 7 ,4 2 1 ,3 7 1 ,3 4 6 ,a n d3 0 3c m−1[32]. The Raman
spectra of the CNT composite samples (V-CNT2 and VO-CNT2) are barely dominated by the characteristic bands of theCNT, Fig. 5 and Supporting Information, but these bands areslightly shifted to the higher wavenumbers, from 1,301, and
1,593 cm
−1(in CNT sample) to 1,311, and ~1,600 cm−1,
respectively, (in V-CNT2, and VO-CNT2 samples), which indi-cates a lower intertube interaction.1200 1000 800 600 400 200 001000200030004000500060007000
C4C3C2
C1XPS signal (counts)
Binding energy (eV)292 290 288 286 284 282 292 290 288 286 284 282010002000300040005000XPS signal (counts)
Binding energy (eV)C1C2
C3
C4
Binding Energy (eV)- O KLL
– O1s
– C1s
– O2soriginal fibersoxidized fibersoriginal fibers oxidized fibers
FIG. 2. XPS survey spectra and scan of C 1s region of the original and TEMPO-oxidized fibers. [Color figure can be
viewed at wileyonlinelibrary.com]
TABLE 1. XPS analysis of the original and TEMPO-oxidized fibers.
Binding energy (eV)
284.6 286.4 287.7 288.3
Fiber O/C C1 (at. %) C2 (at. %) C3 (at. %) C4 (at. %)
Original 0.66 24.92 52.04 20.30 2.77
Oxidized 0.74 30.66 50.12 12.33 6.91
4 POLYMER ENGINEERING AND SCIENCE —2019 DOI 10.1002/pen

Termogravimetric Analysis. The thermostability of the CNT-
composites was evaluated by means of thermogravimetric experi-
ments carried out at a heating rate of 10/C14C/min under inert
atmosphere, acquired by a continuous nitrogen flow of 20 mL/min.
The values of initial weight loss, the maximum degradation tem-peratures, and total weight loss at 700
/C14C are listed in Table 2.
Figure 6 shows the weight loss rate curves obtained from TGA
experiments.
The initial weight loss of all samples around 100/C14C was
between 6 and 10%, which relates to residual moisture present inall samples. For V, pyrolysis started at 281
/C14C and for V-CNT
composites pyrolysis started at 207/C14C and for all continued until
around 330/C14C leading to depolymerization of viscose fibers to
various anhydro-monosaccharides, dehydrated species, carbon
oxide, and char. Regarding of the VO and VO-CNT composites,pyrolysis started around 210
/C14C and continued until around 350/C14C.In Fig. 5, the thermal degradation of VO show two peaks around
241/C14C (correspond to the degradation of anhydro-glucuronate
units) and 319/C14C (correspond to the degradation of viscose chains
containing more unstable anhydroglucuronate units in the crystal
FIG. 3. Schematization of the V/VO –CNT composites fabrication, emphasizing the color changes of the fibers. [Color
figure can be viewed at wileyonlinelibrary.com]
4000 3500 3000 2500 2000 1500 1000 500
Wavenumber (cm-1)344528921735
1643
1441 1377VV-CNT1V-CNT2VOVO-CNT1VO-CNT2CNT
1200-1000
FIG. 4. The FTIR curves of CNT, V, VO and the composites V-CNT1, V-
CNT2, VO-CNT1, VO-CNT2. The dashed rectangle area delineate the speci fic
absorption band of the >C O moiety in carboxyl group. [Color figure can be
viewed at wileyonlinelibrary.com]500 1000 1500 2000 2500 3000 3500V-CNT2
VO-CNT2Raman Intensity
Raman shift (cm-1)CNT1301
1593
2594
FIG. 5. Raman spectra of CNT, V-CNT2, and VO-CNT2 samples. [Color
figure can be viewed at wileyonlinelibrary.com]
TABLE 2. The initial weight loss, maximum degradation, and total weight loss
of V, VO, and composites samples.
SamplesInitial weight
loss around
100/C14C (%)Maximum
degradation
temperature (/C14C)Total weight
loss at
700/C14C (%)
V 10 331 80
V-CNT1 8 339 76
V-CNT2 9 320 70
VO 9 319 91
VO-CNT1 6 335 86VO-CNT2 7 319 74
CNT 1 567 57
DOI 10.1002/pen POLYMER ENGINEERING AND SCIENCE —2019 5

surface), the two peaks being below the degradation point of the
V (331/C14C) thus con firming the formation of carboxylic groups at
the C6 primary hydroxyls groups (the carboxylic groups leads toa decrease in the thermal degradation point).
Scanning Electron Spectroscopy
Microphotographs of the V and the VO are shown in Fig. 7.
After oxidation for 1 h, the morphologically of VO was identical
to those of the V, only some fine particles formed by deoxidation
have been observed. The oxidized sample VO shows insigni ficant
regions of deterioration only at the surface level. Likewise, as wecan see in SEM imagines, the powder of CNT consists of agglom-erated particles characterized by a quite uniform shape. For theCNT composites, in all cases, the surface became rough and brit-
tle, and the CNT were lowly adsorbed only on the surface of theV/VO fibers without aggregations. The dark regions in the SEM
images were attributed to the CNT.
Dielectric and Shielding Properties
The real part of dielectric permittivity (dielectric constant, ε’)
and the imaginary part (dielectric loss, ε”) has been obtained from
dielectric measurements performed in the frequency range of 106–
100Hz. The dielectric constant and dielectric losses of insulating
materials are attributed to the polarizations of the electrons andmolecules, which includes five types of polarization: electronic,
vibrational (atomic), orientation (dipolar), ionic, and interfacial
polarizations. As one may see from Fig. 8, for all samples, as thefrequency increases, the real ( ε’) and imaginary ( ε”) permittivity
decrease and remain constant at higher frequencies, indicating theoccurrence of dielectric dispersion (ionic and orientation polariz-abilities) [29, 33]. This may be attributed to the dipoles resulting
from changes in valence states of cations and space-charge polari-
zation [27, 34, 35]. Also the interfacial polarizabilities (commonto composites with precursors who have different dielectric prop-erties) can contribute to this decrease in low-frequency region[36]. The chemical oxidation of the viscose increase the polariza-tion (high value of ε’andε”for VO comparing with pristine V).
Thermal treatment (dried in vacuum procedure at 120
/C14C)
increased electronic polarization ( ε’for CNT2 composites, shows
higher value than ε’for CNT1 composites), which can be attrib-
uted to an increased charge carriers, due to increased π-electrons
in the nanocomposites. Values for industrial frequency(50–55 Hz) are given in Table 3.
In terms of shielding effectiveness (Fig. 9), V and VO have no
shielding properties (SE V > 0 dB). The composites present a
maximum shielding for low frequency ( −100 to −75 dB at
0–10 Hz) and a minimum for high frequency (approximately
−25 dB for 10
5–106Hz). The thermal treatment at 120/C14C (for V-
CNT2 and VO-CNT2 composites) lead to a decrease of theshielding properties. Multiple re flection shielding is the high com-
ponent of the total shielding effectiveness while absorption
FIG. 7. SEM microphotographs of V, V-CNT1, V-CNT2, VO, VO-CNT1, VO-CNT2, and CNT.0 100 200 300 400 500 600 700-0,07-0,06-0,05-0,04-0,03-0,02-0,010,00DTG
Temperature (oC) (1) CNT
(2) V
(3) VO
(4) VO-CNT2
(5) VO-CNT1
(6) V-CNT2
(7) V-CNT1
FIG. 6. DTG curves of CNT, V, VO, VO-CNT2, VO-CNT1, V-CNT2, and
V-CNT1 samples. [Color figure can be viewed at wileyonlinelibrary.com]
6 POLYMER ENGINEERING AND SCIENCE —2019 DOI 10.1002/pen

shielding component is almost negligible at low frequencies
(approximately 0 dB for 1 –104Hz and −1.5-3 dB for 104–
106Hz). For industrial frequency (50 –55 Hz) shielding effective-
ness is −75-100 dB, provided a good potential for blocking elec-
tromagnetic interferences in electric devices. For comparison, inTable 4, we present some values for shielding effectiveness
FIG. 9. Multiple re flection shielding (SE MR), absorption shielding (SE A), and total shielding effectiveness (SE) with
frequency in logarithmic scale (base ten) for V, VO, and their composites (V-CNT1, V-CNT2, VO-CNT1, and VO-
CNT2). [Color figure can be viewed at wileyonlinelibrary.com]
TABLE 4. Shielding effectiveness of various cellulose fibers –CNT composites.
Composite Thickness [mm] Shielding typeShielding
value [dB] Frequency [Hz] Reference
Viscose/barium titanate 3 Absorption 18 –22 104–105[29]
Viscose/barium titanate 3 Absorption 4 –95 0 –55 [29]
Polyaniline coated nickel spheres/
carbon black/co poly(ethylene-propylene)- Refraction ≥20 104–108[37]
Polyaniline/silver, graphite, and carbon black – Refraction ≥20 108–109[38]
Barium titanate/carbon nanotube 1 Re flection 29.6 13.6 /C1109[27]
Barium titanate/carbon nanotube 1.1 Re flection 56.5 13.2 /C1109[27]
Cellulose/carbon nanotube 0.2 Re flection, multiple re flection 20 (15 –40) /C1109[35]
Viscose/carbon nanotube 1 –1.5 Multiple re flection 60 –110 50 –55 This work
Viscose/carbon nanotube 1 –1.5 Multiple re flection 25 106This work
Oxidized viscose/carbon nanotube 1 –1.5 Multiple re flection 70 –80 50 –55 This work
Oxidized viscose/carbon nanotube 1 –1.5 Multiple re flection 20 106This work
FIG. 8. Dispersion ( ε’=ε’(ν) )a n da b s o r p t i o ns p e c t r a( ε”=ε”(ν)) for V, VO, and their composites (V-CNT1, V-CNT2, VO-
CNT1, and VO-CNT2) with frequency in logarithmic scale (base ten). [Color figure can be viewed at wileyonlinelibrary.com]
TABLE 3. Value of ε’andε”for V, VO, and their composites, for industrial
frequency.
Sample V VO V-CNT1 V-CNT2 VO-CNT1 VO-CNT2
ε’ 18.9 46.5 18.8 80.8 23.2 69.7
ε” 43.1 249.3 15.7 541.2 21 12.7
DOI 10.1002/pen POLYMER ENGINEERING AND SCIENCE —2019 7

obtained for composites with cellulose fibers and/or graphite or
CNT, cited in the literature for 100–108Hz frequency or even for
GHz values (109Hz).
It is generally assumed that three mechanisms are involved in elec-
tromagnetic interference shielding: re flection, absorption, and
multiple-re flection [39]. When CNT are incorporated into polymer
matrix, absorption plays a primordial role, followed by shielding by
reflection. Some theoretical analysis revealed that multiple re flection
within the CNT internal surfaces might have an obstructive impact onthe overall electromagnetic interference shielding, because the diame-ter of the CNT is orders of magnitude smaller than the skin depth.Multiple re flection between external surfaces of CNT also diminish
the overall shielding, but this in fluence is lower than that between the
internal surfaces (according to the theoretical calculations) [39].
CONCLUSIONS
In this work, we have used both viscose and its TEMPO-
oxidized analogue as organic matrices for the incorporation of
CNT using ultrasonication process. The shielding values (dB) for
both viscose and oxidized viscose –CNT composites, have larger
values (60 –110 and 70 –80 dB, respectively) than those reported
in the literature at 50 –55 Hz. Therefore, the as prepared bio-based
composites, containing multiwalled CNT and viscose fibers or C 6-
oxidized analogue, can be easily adapted to various requirementsdue to their flexibility, being ef ficient alternatives to the existing
electromagnetic shielding materials, in particular for electric field
within industrial frequency (50 –55 Hz).
ACKNOWLEDGMENTS
This work was supported by a grant of Ministry of
Research and Innovation, CNCS – UEFISCDI, project number
PN-III-P4-ID-PCE-2016-0349, within PNCDI III.
REFERENCES
1. L.Q. Li, T. Fan, R.M. Hu, Y.P. Liu, and M. Lu, Cellulose ,24,
1121 (2017).
2. M. Khodaei, A.E. Pirbazari, and A. Talebizadeh, Cell. Chem.
Technol. ,51, 703 (2017).
3. G. Biliuta and S. Coseri, Cellulose ,23, 3407 (2016).
4. S. Coseri, Biotechnol. Adv. ,35, 251 (2017).
5. P.L. Bragd, H. van Bekkum, and A.C. Besemer, Top. Catal. ,27,
49 (2004).
6. S. Coseri, G. Biliuta, B.C. Simionescu, K. Stana-Kleinschek,
V. Ribitsch, and V. Harabagiu, Carbohydr. Polym. ,93, 207 (2013).
7. B. Poyraz, A. Tozluo /C21glu, Z. Candan, A. Demir, M. Yavuz,
Ü. Büyuksari, and H. _l. Ünal, Fiber. Polym. ,19, 195 (2018).
8. D. Zhang, M. Liu, Y. Liu, and H. Li, Fiber. Polym. ,17, 1330 (2016).
9. G. Biliuta, L. Fras, M. Drobota, Z. Persin, T. Kreze, K. Stana-
Kleinschek, V. Ribitsch, V. Harabagiu, and S. Coseri, Car-
bohydr. Polym. ,91, 502 (2013).
10. S. Coseri and G. Biliuta, Carbohydr. Polym. ,90, 1415 (2012).
11. A. Clarke, A. A. Vasileiou, and M. Kontopoulou, Polym Eng. Sci. ,
59, 989 (2019).
12. G.N. Parsons, S.E. Atanasov, E.C. Dandley, C.K. Devine, B. Gong,
J.S. Jur, and P.S. Williams, Coord. Chem. Rev. ,257, 23 (2013).13. Y. Zhang, A.X. Dong, Q. Wang, X.R. Fan, A. Cavaco-Paulo,
and Y. Zhang, Appl. Biochem. Biotechnol. ,174, 820 (2014).
14. M. Pasta, F.L. Mantia, L. Hu, H.D. Deshazer, and Y. Cui, Nano.
Res.,6, 452 (2010).
15. S. Iijima, Nature ,354, 56 (1991).
16. P.J. Brigandi, J.M. Cogen, J.R. Reffner, C.A. Wolf, and R.
A. Pearson, Polym. Eng. Sci. ,57, 1329 (2017).
17. S. Thomas, S.C. George, and S. Thomas, Polym Eng. Sci. ,58,
961 (2017).
18. S. Fan, M.G. Chapline, N.R. Franklin, T.W. Tombler, A.
M. Cassell, and H. Dai, Science ,283, 512 (1999).
19. S.S. Wong, E. Joselevich, A.T. Woolley, C.L. Cheung, and C.
M. Lieber, Nature ,394, 52 (1998).
20. T. Rueckes, K. Kim, E. Joselevich, G.Y. Tseng, C.L. Cheung,
and C.M. Lieber, Science ,289, 94 (2000).
21. Y. Maimaiti, N. Dongmulati, S. Baikeri, A. Maimaiti,
R. Maitisidike, and X. Maimaitiyiming, Fiber. Polym. ,19, 927
(2018).
22. A. Hosseinpour, R. Nasseri, S. Ghiassinejad, M. Mehranpour, A.
A. Katbab, and H. Nazockdast, Polym Eng. Sci. ,59, 447 (2018).
https://doi.org/10.1002/pen.24942.
23. H.W. Ott, Electromagnetic Compatibility Engineering , John
Willey & Sons, New Jersey (2009).
24. R.P. Clayton, Introduction to Electromagnetic Compatibility . 2rd
ed., John Wiley & Sons, Inc, Hoboken, New Jersey (2006).
25. P. Saini, V. Choudhary, B.P. Singh, R.B. Mathur, and S.
K. Dhawan, Mater. Chem. Phys. ,113, 919 (2009).
26. S.A. Schelkunoff, Electromagnetic Waves , Van Nostrand, New
Jersey (1943).
27. G.J. Hong Melvin, Q.Q. Ni, and T. Natsuki, J. Alloys Compd. ,
615, 84 (2014).
28. S.M. Abbas, M. Chandra, A. Verma, R. Chatterjee, and T.
C. Goel, Compos. A ,37, 2148 (2006).
29. R. Rotaru, C. Peptu, and V. Harabagiu, Cell. Chem. Technol. ,50,
621 (2016).
30. A. Kumar and P.S. Alegaonkar, ACS Appl. Mater. Interfaces ,7,
14833 (2015).
31. K. Mishra, K.P. Bastola, R.P. Singh, and R. Vaidyanathan,
Polym Eng. Sci. accepted manuscript 2019, https://doi/epdf/10.
1002/pen.25100.
32. J. Was-Gubala and W. Machnowski, Spectrosc. Lett. ,47, 527
(2014).
33. Y. Feldman, A. Puzenko, and Y. Ryabov, Adv. Chem. Phys. ,
133A , 1 (2005).
34. F. Kremer and A. Schonhals, Broadband Dielectric Spectros-
copy, Springer Verlag, Berlin Heidelberg, 35 (2003).
35. M. Imai, K. Akiyama, T. Tanaka, and E. Sano, Compos. Sci.
Technol. ,70, 1564 (2010).
36. G.Z. Liu, C. Wang, C.C. Wang, J. Qiu, M. He, J. Xing, K.J. Jin,
H.B. Lu, and G.Z. Yang, Appl. Phys. Lett. ,92, 122903 (2008).
37. S. Geetha, K.K. Satheesh Kumar, and D.C. Trivedi, J. Composite
Mater ,39, 647 (2005).
38. D. C. Trivedi, and S. K. Dhawan, In Frontiers of Polymer
Research ; J. K. Nigam, P. N. Prasad, Eds., Plenum: New York,
pg. 419, (1992).
39. M.H. Al-Saleh and U. Sundararaj, Carbon ,47, 1738 (2009).
8 POLYMER ENGINEERING AND SCIENCE —2019 DOI 10.1002/pen