Impact of MWCNTs on the structural, electrical, and optical characteristics of PVA/PANi blends

Abstract

Polyvinyl alcohol (PVA)/0.25% polyaniline (PANi)/x wt % MWCNTs (x = 0, 0.1, 0.2, 0.5, 0.8, 1) blends were formed. The X-ray diffraction technique was used to investigate the formation of the different blends. The effects of the MWCNTs amounts on the optical parameters and emitted colours of the different blends were investigated. The optical band gaps for PVA/0.25% PANi/x wt % MWCNTs blends are 3.24, 3.13, 3.05, 2.99, 2.84, and 2.95 eV when the MWCNTs amounts in the blends become 0.0, 0.1, 0.2, 0.5, 0.8, and 1 wt %, respectively. The modifications in optical band gaps upon loading PVA/ PANi with MWCNTs indicated that the investigated blends are promising candidates for solar cell and photocatalysis applications. The DC conductivity exhibited a maximum value in the blend with 0.8 wt% MWCNTs. All blends have two activation energies. The influence of MWCNTs amounts on the dielectric constant, AC conductivity, impedance, electric modulus, characteristic frequency, and Nyquist plot of the different blends was examined.

KEYWORDS:

• PVA
• PANi
• MWCNTs blends
• structure
• optical
• electric

1. Introduction

Poly(vinyl alcohol) (PVA) polymer has a semicrystalline structure. It has a high dielectric strength, good thermal and mechanical stability, and a high charge storage capacity [1,2], and it can be dissolved in water and is safe for the environment [2]. Polyaniline (PANi) exhibited semiconductor or metallic features that have gotten a lot of attention from scientists because they are easy to make, have distinctive chemical and electrical attributes that can be governed, and are very stable against changes in temperature and the environment [3]. Carbon nanotubes (CNTs) have distinguished characteristics such as thermal stability, good charge transport, and high mechanical strength. Adding a small amount of CNTs to polymer strongly enhanced the properties of the host matrix [4]. Carbon atoms’ p electrons form a wide variety of delocalized bonds, and the conjugation effect is significant, so CNTs conduct electricity well [5].

Sharna et al. found that the optical bandgap of PANi was reduced as the amount of MWCNTs increased in the xMWCNTs/PANi matrix [6]. With only 0.02 weight percent of MWCNT nanofiller added to the polymer matrix, Alam et al. observed a fourfold increase in dielectric constant. Nanocomposite dielectric strength, ac conductivity, and relaxation time were all improved by MWCNT incorporation into the polymer matrix. Because of its high dielectric constant and low loss tangent, the biopolymer-based nanocomposite material has potential for use in fully biocompatible electrostatic capacitors and energy storage devices [7].

Improved electrochemical stability window (4.0 V), transference number (0.968), and ionic conductivity of MWCNTs-based poly(vinyl alcohol) (PVA)-poly(ethylene glycol) (PEG), sodium nitrate (NaNO₃) polymer blend composite electrolytes make them suitable for use in energy storage devices [8]. As the amount of MWCNTs added increased, Rasheed et al. [9] showed that the optical band gap of PANi thin films could be tuned from 2.38–1.78 eV, and the conductivity was also significantly improved. Increased MWCNT content in the PVA/PVP blend polymeric matrix, as discovered by Abdelrazek et al. [10], causes the formation of a percolating network throughout the composite. High thermal conductivity and excellent electrical insulation flexible PVA/chitosan-MWCNT composite films were fabricated by Luo et al. [11] using a sandwich structure. PANi/MWCNTs nanocomposite with the highest MWCNT level absorbed more than 83% of the incident electromagnetic waves [12]. PVA's flexibility, CNT's conductivity, and PANi's pseudo-capacitance helped improve performance [13]. High performance PVA/MWCNTs/PANi hydrogel based dye sensitized solar cells were achieved [14]. The 3 wt% single-walled carbon nanotubes (SWCNTs) loaded into PVA/PANi film illustrated the highest AC conductivity and maximum shielding effectiveness of 22.73 dB in the X-band (8.2–12.4 GHz) range, among all other samples [15]. In terms of photovoltaic performance, thermal conductivity, and electrical conductivity, PANi/poly(methyl methacrylate) (PMMA) (1%), – MWCNTs (5%) showed marked improvements [16]. The best sensor performances for detecting NH₃ are shown by PANi/PMMA/polystyrene MWCNTs [17]. Alharbi et al. used graphene nanoplates to change the physical, optical, dielectric, and electrical properties of a PEO/PVA blend for use in energy storage devices [18]. Increasing the Fe₂O₃/TiO₂, Ag/TiO₂ or CoFe 2O4/MWCNTs loading in the chitosan/polyvinyl alcohol, chitosan/ polyethylene oxide or PVA/CMC blend significantly enhanced its AC conductivity and dielectric properties [19–21]. Maximum ionic conductivity and dielectric properties are found in the PEO/PVA nanocomposites film containing 0.9 wt% of MWCNTs/ZnO [22]. In our previous study, we found that as the amounts of PANi doped PVA became 0.25 wt %, the dielectric properties exhibited the maximum values [23]. In this work, the percolation threshold ratio of MWCNT content in PVA/0.25 wt % PANi blends and its effect on the optical and electric properties of the blends were explored in details.

2. Methods and materials

PVA/0.25% PANi/x wt % MWCNTs (x = 0, 0.1, 0.2, 0.5, 0.8, 1) blends were prepared using polyvinyl alcohol (PVA, MW = 50,000 g/mol, Acros organics, 98%), polyaniline (PANi, sigma Aldrich), and multi-walled carbon nanotubes (MWCNT, Sigma Aldrich) as in the next steps:

1. 1.8 g PVA was dissolved in 60 ml dimethyl sulfoxide (DMSO, AppliChem) at 50 ^oC for 50 min using a magnetic stirrer.

2. 0.25 wt% PANi/ x wt % MWCNT (x = 0, 0.1, 0.2, 0.5, 0.8, 1) were dissolved separately in 20 ml N,N-Dimethylformamide (DMF, Thermo-Fisher) using a magnetic stirrer for 1 h at 60 ^oC, then the solutions were put in an ultrasonic bath at 50 ^oC for 140 min.

3. For 5 h at 50 ^oC, the aforementioned solutions were thoroughly combined with the help of a magnet stirrer.

4. The solutions were then transferred to Petri dishes and baked in a 70 ^oC electric oven for five to six days.

5. PVA/PANi blends were formed with a thickness of 0.16-0.25 mm (measured using a digital micrometer).

X-ray diffraction data was obtained using PANalytical diffractometers (X'pert MPD, Philips, copper source, step size: 0.02°, counting time of 10 s /step, 45 kV, 40 mA). The spectral lines of fluorescence were measured by a spectrofluorometer (FP-8200 JASCO) under excitation wavelength of 380 nm. Diffuse reflectance (R), absorbance (A), and transmittance (T) were measured for all blends using a spectrophotometer (JASCO-V-670) equipped with an integrating sphere assembly in the wavelength range of 200-800 nm.

The following procedures were used to adjust the R values before continuing with the calculations [24]: (1) $\begin{array}{rl}{R}_i& =\frac{\begin{array}{c}[2+T^2-{(1-R)}^2]-\{{[2+T^2-{(1-R)}^2]}^2\\ -4(2-R)R{\}}^{0.5}\end{array}}{2(2-R)}\end{array}$(1) (2) $\begin{array}{rl}{R}_F& =(2+T^2-{(1-R_i)}^2-\{[2+T^2-{(1-R_i)}^2{]}^2\\ & -4R_i(2-R_i){\}}^{0.5})(2(2-R_i){)}^{-1}\end{array}$(2) where R_i and R_F are the interface reflectance and the reflection from one face, respectively.

From [24], one can calculate the blended materials’ corrected absorbance (A_corrected), refractive index (n), and extinction coefficient (k): (3) $\begin{array}{rl}{A}_{\mathrm{corrected}}& =\text{ln}\left(\frac{R_{F}T}{R_i-R_F}\right)\end{array}$(3) (4) $\begin{array}{rl}k(R_i,T)& =\frac{\lambda }{4\mathrm{\pi d}}\ln \left[\frac{R_F(R_i,T)T}{R_i-R_F(R_i,T)}\right]\end{array}$(4) (5) $\begin{array}{rl}n& =\frac{1+R_F(R_i,T)}{1-R_F(R_i,T)}+\{\phantom{{\left(\frac{\lambda }{4\mathrm{\pi d}}\right)}^2}\frac{4R_F(R_i,T)}{{(1-R_F(R_i,T))}^2}\\ & -{{\left(\frac{\lambda }{4\mathrm{\pi d}}\right)}^{2}l{n}^2\left[\frac{R_F(R_i,T)T}{R_i-R_F(R_i,T)}\right]\}}^{0.5}\end{array}$(5)

To calculate the optical energy gaps (E_g) both directly and indirectly, the following formula was used [25]: (6) $\begin{array}{r}\mathrm{h\nu }=D(\mathrm{h\nu }-E_g{)}^m\end{array}$(6) Planck's constant, incident light frequency, the so-called disorder parameter, blended polymer thickness, and absorption coefficient are represented by the numbers h, D, d, and α (A/d), respectively. For a direct or indirect transition, m could be 0.5 or 2.

The AC electrical measurements were conducted at room temperature using an LCR metre (Instek LCR-8105G). The measurements were performed over a frequency range of 100 Hz −1 MHz using a sample holder consisting of two circular stainless-steel discs with a 13 mm diameter. The thickness of the obtained samples varied from 0.16–0.25 mm. Under AC voltage of 2 volts, the capacitance (C) and dissipation factor(tan δ) were measured. The real (ε′), imaginary (ε′′) dielectric constant parts, and ac conductivity (σ_ac) of the different blends were calculated using the following formula [26]: (7) $\begin{array}{r}{\epsilon }^{\mathrm{\prime }}=\text{dC}/{\epsilon }_0\text{A}\end{array}$(7) where d and A are the thickness and surface area of the sample, respectively. ε₀ is the permittivity of free space. (8) $\begin{array}{rl}{\sigma }_{\mathrm{ac}}& =2\mathrm{\pi f}{\epsilon }_0{\epsilon }^{\mathrm{\prime }}\tan \delta \end{array}$(8) (9) $\begin{array}{rl}\tan \delta & ={\epsilon }^{\mathrm{\prime \prime }}/{\epsilon }^{\mathrm{\prime }}\end{array}$(9) The complex impedance (Z* = Z′+iZ′′), the complex dielectric constant (ε* = ε′+i ε′′), the complex electric modulus (M* = M′+iM′′) and electrical conduction were obtained using the following formula [26]: (10) $\begin{array}{rl}{Z}^{\mathrm{\prime }}& =\frac{1}{2\mathrm{\pi f}{C}_0}\left[\frac{{\epsilon }^{\mathrm{\prime \prime }}}{{\epsilon }^{\mathrm{\prime }2}+{\epsilon }^{\mathrm{\prime \prime }2}}\right]\end{array}$(10) (11) $\begin{array}{rl}{Z}^{\mathrm{\prime \prime }}& =\frac{1}{2\mathrm{\pi f}{C}_0}\left[\frac{{\epsilon }^{\mathrm{\prime }}}{{\epsilon }^{\mathrm{\prime }2}+{\epsilon }^{\mathrm{\prime \prime }2}}\right]\end{array}$(11) (12) $\begin{array}{rl}{M}^{\mathrm{\prime }}& =\frac{{\epsilon }^{\mathrm{\prime }}}{{\epsilon }^{\mathrm{\prime }2}+{\epsilon }^{\mathrm{\prime \prime }2}}\end{array}$(12) (13) $\begin{array}{rl}{M}^{\mathrm{\prime \prime }}& =\frac{{\epsilon }^{\mathrm{\prime \prime }}}{{\epsilon }^{\mathrm{\prime }2}+{\epsilon }^{\mathrm{\prime \prime }2}}\end{array}$(13) where (Z′ and Z′′), (M′ and M′′), C₀, and f are the real and imaginary parts of (impedance, electric modulus), empty capacitance, and frequency, respectively.

DC conductivity measurements were conducted at different temperatures using a setup consisting of an evacuated cryostat operating with liquid nitrogen, a copper sample holder, a temperature controller, and a Keithley 6517-B electrometer.

The electrical conductivity ($\sigma$) is given by the equation: (14) $\begin{array}{r}\sigma =\frac{{\mathrm{d}}^{\ast }\mathrm{I}}{\text{VA}}\end{array}$(14) where d, A, I, and V are the sample thickness, cross sectional area (m²) of the sample, measured electric current, and applied voltage, respectively.

3. Results and discussion

3.1. Structural analysis

The X-ray diffraction patterns obtained for PVA/0.25 wt% PANi pristine and doped with (x) wt% MWCNTs (x = 0.1, 0.2, 0.5, 0.8, and 1.0) are shown in Figure 1(a). The patterns disclosed a high background as a result of diffuse scattering, signifying the main amorphous nature of the blend. Only diffraction peaks typical of the main constituent polymer PVA are observed: a broad peak at 2θ = 20.0° (10$\overline{1}$)/(101), a weak broad peak at about 22.5° (200), and a small distinctive peak at about 41.0° (111)/(1$\overline{1}$1)/(210)/(2$\overline{1}$0) [27,28]. No diffraction peaks characteristic of PANi or MWCNTs could be detected owing to their small percentages. As the amount of MWCNTs increased the main peak of PVA/ 0.25% PANi (Figure 1(b)) changed slightly, due to the small doping amount.

Figure 1. (a, b) XRD diffraction data for PVA/0.25 wt%PANi/ x wt% MWCNTs blends.

3.2. Optical properties

Figure 2 reveals the absorbance, transmittance, and reflectance data for the PVA/0.25 wt PANi/ %/x wt% MWCNTs films. The polaron's evolution into the conducting PANi is revealed by the peaks at 324 and 633 nm, which correspond to the π-π* transition of the benzenoid ring and the n-π* transition of the quinoid ring, respectively [29]. Including MWCNTs shifts the UV band slightly toward shorter wavelengths, increasing the wavelength spread. As a result, the energy required for interactions between polyaniline and MWCNTs in the polymer matrix increases [9]. Furthermore, surface and grain boundary scattering were considered the most common scattering processes that are responsible for the lowest transmittance spectra for blends as they are doped with MWCNTs. It is clear that increasing the content of MWCNTs up to 0.8 wt % enhanced the absorbance and reflectance while decreasing the optical transmittance of the films but the situation is reverse with further doping. Absorption improved with increasing concentration of MWCNTs in PVA/PANi matrix, indicating generation of more polarons and bipolarons [9]. Similar results were observed as PVP, PVA/PVP, SWCNTs/PVA, graphene/PVA or PVA/PEO blends doped with Zn_1-xSn_xS, (tin sulfide, Co_0.9Cu_0.1S), Cd_0.9Co_0.1S, FeS or graphene nanoplates, respectively [18,30–33]. The reduction in absorbance with further doping beyond 0.8 wt % indicated the reduction in the created number of polarons and bipolarons within the polymer.

Figure 2. (a) Absorbance, (b) transmittance and (c) reflectance spectra for PVA/0.25 wt%PANi/ x wt% MWCNTs blends.

To calculate the direct optical energy gap (E_g) for PVA/0.25 wt% PANi/ wt% MWCNTs blends, a graph between (αhν)² vs. hν was plotted, as revealed in Figure 3. The value of E_g can be obtained by extrapolating the linear part to zero photon absorption, (αhν)² = 0. The obtained values of E_g are 3.24, 3.13, 3.05, 2.99, 2.84, and 2.95 eV when the MWCNT amounts become 0.0, 0.1, 0.2, 0.5, 0.8, and 1 wt %, respectively. When MWCNTs were added to a polymer, the structure of the polymer changed, resulting in a reduction in the optical band gap, which revealed an increase in the degree of disorder in the films [34]. Similar results were obtained as PVA/PEO doped with graphene nanoplates [18]. Amorphous semiconductors (hopping transport) are more likely to experience the carrier transport process in the conducting polymer than crystalline semiconductors (band transport) [35]. A dense localizing state between the lowest-unoccupied molecular orbital (LUMO) and the highest-occupied molecular orbital (HOMO) bands may exist in amorphous materials because of the lack of order. This is called the Urbach energy (Urbach tail) [3]. Reduced E_g values for polymer blend films are indicative of improved electrical conductivity and reduced Fermi energies [9].

Figure 3. Tauc relations for PVA/0.25 wt% PANi/ x wt% MWCNTs blends.

Figure 4 represents the wavelength dependence of the extinction coefficient (k) and refractive index (n) for PVA/0.25 wt% PANi/ x wt% MWCNTs blends. The k values of PVA/0.25 wt% PANi fluctuated in the visible range and exhibited fixed values beyond this range. When PVA/0.25 wt% PANi was doped with different MWCNTs ratios, the k values increased as the wavelength increased. Furthermore, as the amount of MWCNT increased up to 0.8 wt%, the k values increased, and then decreased as the amount increased further. This variation indicated the successful alteration of the electronic structure and absorption coefficient (Figure 2(a)) of the host polymer matrix upon loading with PANi and MWCNTs [36]. In addition, the n values in general for all blends decreased with increasing wavelength, denoting a normal dispersion behaviour for all samples. The refractive index was increased as PVA/PANi loaded with MWCNTs as compared with PVA/PANi, Figure 2(c). Similar results were obtained as the amounts of iron oxide increased in polyvinyl alcohol/graphene nanocomposite films or non-stoichiometric SnS doped PVA/PVP [37,38]. On the contrary, as PVA was loaded alone with MWCNTs, the n values decreased as the amount of MWCNTs increased in the PVA matrix [39]. These variations in the behaviour of n values may be due to the modification of the host matrix microstructure. The variations in the degree of crystallinity and number of microvoids of the host matrix affected the density of the host matrix and thus the n values [40].

Figure 4. Wavelength dependence o of the (a) extinction coefficient, (b) refractive index and (c) variation of refractive index with the amount of MWCNTs for PVA/0.25 wt% PANi/ x wt% MWCNTs blends.

3.3. Fluorescence (FL) analysis

The obtained FL spectra for pure and doped PVA/PANi blends with wt.% (x = 0.1, 0.2, 0.5, 0.8, and 1.0%) MWCNTs are depicted in Figure 5. As revealed from the graph, three main peaks were detected around 403 nm (violet), 574 nm (yellow), and 711 nm (red). Also there are two weak peaks at 427 nm (violet) and 445 nm (blue). The emission peaks at 403, 427, and 445 nm are characteristic fluorescence peaks of PVA polymer, which exhibit luminescence emissions in the range 400-500 nm attributed to π*/n electronic transitions in free –OH groups in the PVA molecular chain. The emission peak positions were subjected to the spatial arrangement of the electrons within the oriented PVA molecules [41–43]. Chain separation and conformation distortions of the molecular segments confined within the mixing zones may be responsible for the other two peaks at 574 and 711 nm, which may have originated from PVA/PANi intermixing at the interface [44–46]. It is worth noting that the fluorescence is most intense for the blend loaded with 0.2 wt% MWCNTs content. Loading PVA/PANi blend with MWCNTs changed the fluorescence intensities of three main emission peaks nonmonotonically in different way. As shown also from the graph, the FL intensity was enhanced as the blend doped with 0.2 wt% MWCNTs and reduced for other doping values. The intensity of the characteristic emission peak at 403 nm reduced nonlinear with MWCNTs amounts: It decreased for x = 0.1 wt%, increased for 0.5 and 0.8%, and then decreased for 1.0 wt%. For the peak at 711 nm, the FL intensity increased above that of the unloaded blend for blends with x = 0.5 and 0.8% MWCNTs. In the FL phenomenon, the fluorescence intensity of a peak imitates the density of states for a specific transition corresponding to this peak, which is affected by many factors like LUMO and HOMO levels, the inter-band defect levels, cluster formations, cross- linking through bonding, etc. [47–51]. For our present samples, hydrogen bonding may be formed between the –OH functional groups from PVA and the functionalized MWCNTs, which changes the charge distribution among the molecules. Moreover, for high MWCNTs, clustering may occur, resulting in a reduction in intensity through absorption.

Figure 5. (a, b) Fluorescence spectra for PVA/0.25 wt% PANi/ x wt% MWCNTs blends.

3.4. Dielectric characteristics

The frequency (Log f) dependence of the real and imaginary parts of the dielectric constant (ε_r, ε_i) for all blends is represented in Figure 6. It is noticed that at the lower frequency ranges, the values of ε_r are high owing to the interfacial effect of the blend and electrode. Also, the values of ε_i are high at lower frequencies due to the mobile charges within the polymer backbone [52]. ε_r values were reduced with a further increase in frequency for all samples because of the orientation of dipoles within polymeric films in the direction of the applied field. At the high frequency range, ε_r has nearly constant values. This tendency could be attributed to the dipole orientation, which makes rotating the dipoles at high frequencies difficult. The values ε_r and ε_i were reduced as MWCNT amounts increased up to 0.5 wt% MWCNTs. On the other hand, ε_r and ε_i were increased to maximum values as the amounts MWCNTs became 0.8 wt%. Finally, ε_r and ε_i were reduced again as the contents of MWCNTs increased further. Alghunaim et al. found that the addition of single or multi-wall carbon nanotubes to PVA increased its dielectric constant and ac conductivity [4]. Also as the amounts of graphene nanoplates, nano Ag/TiO₂, nickel and zinc oxide nanoparticle or MWCNTs/Li-doped TiO2 nanoparticles increased in PVA/PEO, chitosan/ polyethylene oxide, sodium alginate/polyethylene oxide or poly (ethylene oxide)/ poly (methylmethacrylate) blend, respectively, the dielectric constant increased [18,20,53,54].

Figure 6. (a) Real, (b) imaginary parts of the dielectric constant for PVA/0.25 wt% PANi/ x wt% MWCNTs blends.

The enhancement/reduction in ε_r values may be due to the increase/decrease in the order distribution of PANi and MWCNTs over the PVA blend, which affects the interfacial polarization. Furthermore, with the addition of the MWCNTs, the mobile charges (polarons and/or bipolrons) that associate PANi and free charge increase, causing influence and lower values of ε_i at high frequencies [55]. The frequency dependence of AC conductivity (σ_ac) for the PVA/PANi/MWCNTs system is shown in Figure 7. Almost, all blends exhibited a similar trend, where at the lower frequency range, there is no change in the σ_ac up to 10³ Hz, and a rapid increase is noticed after this frequency. This rapid rise in σ_ac results from the increased hopping of the electrons at relatively high frequencies. The AC conductivity decreased as MWCNT concentration increased up to 0.5 wt%, then increased as MWCNT concentration increased further. The highest AC conductivity values were obtained as the percentage of MWCNTs reached 0.8. The irregular variation in the AC conductivity as the amount of MWCNTs changed may be due to the variation in the number of charge carriers and defects at the surface of the host polymer due to the interaction of PANi and MWCNTs inside the polymer matrix [56].

Figure 7. A C conductivity for PVA/0.25 wt% PANi/ x wt% MWCNTs blends.

Figure 8. (a) Real and (b) imaginary parts of impedance for PVA/0.25 wt% PANi/ x wt% MWCNTs blends.

Figure 9. Nyquist plot for PVA/0.25 wt% PANi/ x wt% MWCNTs blends.

Figure 10. (a) Real and (b) imaginary parts of electric modulus for PVA/0.25 wt% PANi/ x wt% MWCNTs blends.

Figure 8 displays the plots of Z’ and Z'’ for all samples. As can be seen from the graph, Z′ values decreased monotonically with the increase in frequency as a result of the weakening of the interface polarization. The Z′ curves superimpose at extremely high frequencies, suggesting that the interfacial polarization disappears [57]. Furthermore, the highest Z'’ values at low frequencies indicate the presence of dipolar relaxation in that frequency range. It is also noticed that both real and imaginary impedance components are frequency-independent after a certain cut-off frequency, called the characteristic frequency (f_c). Blends’ behaviour changed at the f_c from being resistive-like (frequency independent) to being capacitive-like (frequency dependent). The f_c increases with low MWCNT loading amounts of up to 0.5 wt%. MWCNT amount and PANi-MWCNT interaction increase the blend's interconnected pathways, increasing frequency-independent electrical conductivity. Below f_c, Z′ increased as the filler amount increased up to 0.5 wt%, then increased due to a change in blend conductivity, or the number of conductive pathways in the host blend [57]. Except for samples containing 0.5 and 0.8% MWCNTs, no loss peaks can be seen in Z'’ curves, demonstrating that there are no relaxation processes in this system. The Nyquist plot of the complex impedance (Z′ versus Z'’) at room temperature for all blends is revealed in Figure 9. All plots for all blends display a similar trend, like the ideal Nyquist impedance plot. Blend with 0.1 wt % MWCNTs has the larger semicircle, while blend with 0.8 wt % MWCNTs has the smaller one among the other blends, which represents higher/lower interfacial charge-transfer resistance and is attributed to the poor/good electrical conductivity of the blend, respectively.

Since electrode polarization, absorption, and interactions of impurities can be ignored in the electric modulus formalism, large deviations in the components of complex dielectric permittivity can be minimized. Figure 10 displays the frequency-dependent real and imaginary parts of the electric modulus (M′, M′′) for all blends. M′ has small values in the low-frequency range, as shown by the plot, except in blends containing 0.8 wt% MWCNTs.

This result indicated the small contribution of electrode polarization and electrode effects [58]. The M′ values increased as the frequency increased. The unsaturated values at higher frequencies indicated the presence of more effective interfacial polarization. Doped PVA with ZnBr₂ displayed saturation in M′ values at higher frequencies [59]. Also, the M′ values increased as the amount of MWCNTs reached 0.1 and 0.2 wt % in the blend, while they decreased with further loading of the filler into the blend.

Figure 11. (a-f) The change of Lnσ as function of temperature for PVA/0.25 wt% PANi x/ wt% MWCNTs blends.

In every blend, M′′ showed a peak whose location was influenced by the amount of MWCNTs present. As in the case of higher MWCNT doping concentrations, the shift of the peaks to the higher frequency indicated a reduction in the relaxation time, which directly supports the improvement in ionic conductivity as a result of an increase in the mobility of free ions [60]. Frequency ranges that were less than the peak maximum defined the range of mobile charge carriers. Additionally, the charge carriers are restricted to potential wells that are mobile over short distances in the frequency range above the maximum. As MWCNT content reached 0.8 and 1 wt% and increased with lower doping ratios, M′′ decreased. Due to the large number of charge carriers that accumulated at the electrode-electrolyte interface (electrode polarization effect) as the MWCNT content of the matrix increased, the capacitance increased [58].

3.5. DC electrical conductivity

The changes in DC conductivity (${\sigma }_{\mathrm{dc}}$) with the inverse of temperature (1000/T) are depicted in Figure 11(a–f). The graph clearly showed that the dc conductivity of the blends changes as the amounts of MWCNTs increased. The results suggested that DC conductivity was enhanced as the blend loaded with MWCNTs and reached its highest values when the amount of the filler became 0.08 wt. This enhancement may be due to the extent of polymerization as the blend is doped with MWCNTs, resulting in a higher number of delocalized charge carriers [61].

The DC conductivity for the different blends can be described by the following Arhenius relation: (15) $\begin{array}{r}{\sigma }_{\mathrm{dc}}={\sigma }_o\text{exp}(-{\text{E}}_{\text{A}}/{\text{k}}_{\text{B}}\text{T})\end{array}$(15) where E_A, ${\sigma }_o$, and k_B are the activation energy, pre-exponential factor, and Boltzman’s constant, respectively.

The activation energy E_A can be obtained from the slope of the plots as revealed on the graph. The E_A values varied irregularly as the amount of MWCNTs in the blend changed. Similar behaviour was detected for PVA/PVP loaded with MWCNTs [10].

4. Conclusion

The XRD technique confirmed the formation of the PVA/PANi blends. As the content of MWCNTs increased in blend, the absorbance and reflectance increased, while the optical transmittance of the films decreased. When the amount of MWCNTs in the blend reaches 0.8 wt%, the optical band gap value becomes the smallest. As the amount of MWCNTs increased up to 0.8 wt%, the k and n values increased. The blends emitted violet, yellow, and red colours. Loading the PVA/PANi blend with different x wt% of MWCNTs changed the fluorescence intensity monotonically and in different ways. The content of MWCNTs affects the DC conductivity of the different blends. The values of the activation energies varied irregularly with the change in the amounts of MWCNTs. The values of the dielectric constant and AC conductivity were increased to their maximum values as the amount of MWCNTs reached 0.8 wt%. The characteristic frequency increased as the amount of MWCNTs increased up to 0.5 wt %. The blend with 0.1 wt % MWCNTs has the larger semicircle, while the blend with 0.8 wt % MWCNTs has a smaller one in the Nyquist plot. The ionic conductivity and capacitance of the blend were enhanced as it loaded with MWCNTs.

Authors’ contribution

All authors have contributed, discussed the results and approved the final manuscript.

Acknowledgment

Acknowledgement: The authors extend their appreciation to the Deputyship for Research & Innovation, Ministry of Education in Saudi Arabia for funding this research. (IFKSURC-1-1006).

Disclosure statement

No potential conflict of interest was reported by the author(s).

Research data policy and data availability statements

The authors confirm that the data supporting the findings of this study are available within the article.

Additional information

Funding

This work was supported by the Deputyship for Research & Innovation, Ministry of Education in Saudi Arabia for funding this research. (IFKSURC-1-1006).