**Optical and Electrochemical Applications of Li-Doped NiO Nanostructures Synthesized via Facile Microwave Technique**

**Aarti S. Bhatt <sup>1</sup> , R. Ranjitha <sup>2</sup> , M. S. Santosh 3,\*, C. R. Ravikumar 4,\*, S. C. Prashantha <sup>4</sup> , Rapela R. Maphanga 5,6 and Guilherme F. B. Lenz e Silva <sup>7</sup>**


Received: 21 January 2020; Accepted: 3 April 2020; Published: 2 July 2020

**Abstract:** Nanostructured NiO and Li-ion doped NiO have been synthesized via a facile microwave technique and simulated using the first principle method. The effects of microwaves on the morphology of the nanostructures have been studied by Field Emission Spectroscopy. X-ray diffraction studies confirm the nanosize of the particles and favoured orientations along the (111), (200) and (220) planes revealing the cubic structure. The optical band gap decreases from 3.3 eV (pure NiO) to 3.17 eV (NiO doped with 1% Li). Further, computational simulations have been performed to understand the optical behaviour of the synthesized nanoparticles. The optical properties of the doped materials exhibit violet, blue and green emissions, as evaluated using photoluminescence (PL) spectroscopy. In the presence of Li-ions, NiO nanoparticles exhibit enhanced electrical capacities and better cyclability. Cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS) results show that with 1% Li as dopant, there is a marked improvement in the reversibility and the conductance value of NiO. The results are encouraging as the synthesized nanoparticles stand a better chance of being used as an active material for electrochromic, electro-optic and supercapacitor applications.

**Keywords:** Li-doped NiO; microwave synthesis; computational simulation; electrochemical measurements; photoluminescence

#### **1. Introduction**

Nano metal oxides are attracting several researchers due to their varied applications. The reduction in their size contributes tremendously to expanding the surface area, thereby making their optical and electrical properties highly sensitive to surface morphology. Among the various nanometal oxides, nickel oxide (NiO) nanoparticles have been extensively investigated because of their excellent chemical stability and favourable opto-electrical properties. NiO is an antiferromagnetic material and possesses

a cubic structure. It has a band gap in the range of 3.6–3.8 eV [1] which can be tailored by reducing the size and/or by doping. It is therefore not surprising that NiO finds applications as antiferromagnetic materials [2], p-type transparent conducting films [3], electrochromic display materials [4] and chemical sensors [5].

For electro-optic applications, the stoichiometry of the crystal and optimization of the band structure are crucial. The chemical composition and crystal structure of the nanoparticles can be manipulated by introducing impurity atoms. The presence of dopants alters the energy configuration of the crystal lattice, thereby influencing the optical and transport properties. Generally, dopants from I to V group elements are introduced to obtain a stable p-type NiO semiconductor [6]. It has been observed that the presence of Li in the metal oxide lattice increases the electrical resistivity, thereby making them suitable for manufacturing transparent conducting oxides, piezoelectric devices and memory devices [7,8]. However, it is also claimed [9–11] that the presence of Li as a dopant results in a decrease in the resistivity of NiO leading to an improvement in its electrochromic properties. Hence, it becomes necessary to understand the effect of dopant concentration on the electrochemical properties.

The optical properties of the materials also play a crucial role in several applications, like optoelectronics, integrated optics, solar power engineering and optical sensor technology. Generally, compounds like C2S, Ca2SiO<sup>4</sup> and CaAl2O<sup>4</sup> are considered as standard phosphor materials [12–15]. It is reported that doping of P ions in the green emitting C2S:Eu2<sup>+</sup> effectively enhances the PL intensities [13]. Doping these with rare earth metals like Eu and Nd is known to enhance their emission characteristics. Nakano et al. have reported the effect of annealing temperature on the photoluminescence of Eu doped Ca2SiO4. They observed a red emission at 1773 K and green emission at 1473 K [14]. Recently, a significant work on Eu/Nd-doped Ca12Al14O<sup>33</sup> and CaAl2O<sup>4</sup> phosphors as long-lasting blue emitters has also been carried out [15]. Though binary metal oxides are well known for their superior luminous efficiency, related works on simple oxides are rare. Wide gap semiconductors like ZnO have been used extensively as nanophosphors [16,17]. It was found that the doping of these nanophosphors with dopants like Li [18], Al [19], Mg [20], V [21], Eu [22] and Er [23] significantly alters their visible emission spectra. It is interesting to note that the number of research papers reporting on the electrochemical properties of NiO materials are much greater in comparison to the reports on their optical properties. NiO, being a p-type semi-conductor, stands a fair chance of being employed as a light-emitting diode or a photodetector in photoelectronic devices. To the best of our knowledge, there are very few studies [18] exploring the effect of Li ion on the optical properties of NiO. The present work is an attempt to underline the effect of Li in tuning the optical behavior of NiO along with its electrochemical characteristics.

In this background, the present work is focused on understanding the influence of Li doping on the structural, optical and electrochemical properties of NiO. The nanoparticles were synthesized by microwave-assisted synthesis. Since here, high-frequency microwaves are used for heating, this method has an advantage over conventional heating of being faster and consuming low energy. Although there is literature available on the microwave synthesis of NiO [24–26] and doped NiO [27], to the best of our knowledge, the present work is the first attempt to synthesize Li-doped NiO by microwave radiations. A detailed mechanism has also been proposed to understand the decrease in the band gap and resistivity of the as-synthesized nanomaterials on inclusion of the dopant.

#### **2. Materials and Methods**

#### *2.1. Synthesis of NiO Nanoparticles*

Nickel acetate tetrahydrate, lithium carbonate, polyethylene glycol 200 (PEG 200) and sodium hydroxide pellets were obtained from Sigma Aldrich, Bangalore, India. All chemicals used in this work are of analytical grade and were used without any further purification. An aqueous solution of 0.2 M nickel acetate was prepared. To this, 0.3 M NaOH solution was added dropwise with continuous stirring. The mixture was stirred for an hour to achieve concentration homogeneity. The resulting green-colored solution was heated in a microwave oven for 5 min at 210 W and 95 ◦C. The green-colored

*2.2. Characterization* 

*2.4. Computational Details* 

carried out using Fluor Essence™, Version 3.9.

*2.3. Preparation of Carbon Paste Electrode for Electrochemical Studies* 

product thus obtained was alternately washed with water and ethanol till a pH of 7.0 was attained. The precipitate was separated by centrifugation and the solid precursor obtained was annealed in a muffle furnace at 300 ◦C for 2 h. resulting green-colored solution was heated in a microwave oven for 5 min at 210 W and 95 °C. The green-colored product thus obtained was alternately washed with water and ethanol till a pH of 7.0 was attained. The precipitate was separated by centrifugation and the solid precursor obtained was

*Materials* **2020**, *13*, x FOR PEER REVIEW 3 of 18

In an alternate procedure, PEG 200 which acts as a surfactant, was dissolved in NaOH solution before adding it dropwise to nickel solution in a similar manner described earlier. The surfactant to Ni(CH3COO)<sup>2</sup> ratio was fixed to 2:1. It can be inferred that the surfactant acts on the surface of NiO nuclei, thereby inhibiting its growth (Figure 1). In a similar manner, doped samples were prepared by adding lithium carbonate solutions of different concentrations (1, 2 and 5%) to nickel acetate solutions, followed by alkali addition. The overall idea of the synthetic method adapted is depicted in the flow chart represented in Figure 2. annealed in a muffle furnace at 300 °C for 2 h. In an alternate procedure, PEG 200 which acts as a surfactant, was dissolved in NaOH solution before adding it dropwise to nickel solution in a similar manner described earlier. The surfactant to Ni(CH3COO)2 ratio was fixed to 2:1. It can be inferred that the surfactant acts on the surface of NiO nuclei, thereby inhibiting its growth (Figure 1). In a similar manner, doped samples were prepared by adding lithium carbonate solutions of different concentrations (1, 2 and 5%) to nickel acetate solutions, followed by alkali addition. The overall idea of the synthetic method adapted is depicted

**Figure 1. Figure 1.** Structural impact of nickel acetate Structural impact of nickel acetate and polyethylene glycol(PEG) reaction. and polyethylene glycol(PEG) reaction.

**Figure 2.** Flow chart for the synthesis of different NiO nanoparticles. **Figure 2.** Flow chart for the synthesis of different NiO nanoparticles.

examined using the Shimadzu UV-VIS spectrophotometer (UV-2600, Kyoto City, Japan), in the range of 200–800 nm. Photoluminescence studies were carried out on a Horiba (model Fluorolog-3) spectrofluorimeter (Kyoto City, Japan) with 450 W Xenon excitation source. The spectral analysis was

Cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS) were performed on Model CHI604E potentiostat (CH Instruments, TX, USA). The experiment was carried out in acell comprised of a three-electrode system—a carbon paste working electrode (3.0 mm in diameter), a platinum wire counter electrode and a saturated Ag/AgCl reference electrode. The carbon paste electrode was prepared by grinding 70% graphite powder (particle size 50 µm and density 20 mg/100 mL), 15% of the prepared sample (A to E in Figure 2) and 15% silicone oil. The resulting homogeneous product was packed into the cavity of a customized polymer tube and smoothened at the ends.

The total energies were calculated by the projected augmented planewave (PAW) of density function-al theory, as implemented in the CASTEP code embedded in the Materials Studio software, Version 5.5.2 [28]. The core electrons were described using the projector augmented method. The exchange–correlation energy of the electrons was treated using the generalized gradient

X-ray diffraction studies were carried out on Philips X'pert PRO X-ray diffractometer, Malvern,

#### *2.2. Characterization*

X-ray diffraction studies were carried out on Philips X'pert PRO X-ray diffractometer, Malvern, UK, with graphite monochromatized CuK<sup>α</sup> (λ = 1.5418 Å) radiation, at a scan rate of 10◦ min−<sup>1</sup> . Surface morphology was studied using a Field Emission Scanning Electron Microscopy (FESEM; Neon 40 Crossbeam, Carl Zeiss, Jena, Germany) with a resolution of 1.1 nm. Diffuse reflectance spectra were examined using the Shimadzu UV-VIS spectrophotometer (UV-2600, Kyoto City, Japan), in the range of 200–800 nm. Photoluminescence studies were carried out on a Horiba (model Fluorolog-3) spectrofluorimeter (Kyoto City, Japan) with 450 W Xenon excitation source. The spectral analysis was carried out using Fluor Essence™, Version 3.9.

#### *2.3. Preparation of Carbon Paste Electrode for Electrochemical Studies*

Cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS) were performed on Model CHI604E potentiostat (CH Instruments, TX, USA). The experiment was carried out in acell comprised of a three-electrode system—a carbon paste working electrode (3.0 mm in diameter), a platinum wire counter electrode and a saturated Ag/AgCl reference electrode. The carbon paste electrode was prepared by grinding 70% graphite powder (particle size 50µm and density 20 mg/100 mL), 15% of the prepared sample (A to E in Figure 2) and 15% silicone oil. The resulting homogeneous product was packed into the cavity of a customized polymer tube and smoothened at the ends.

#### *2.4. Computational Details*

The total energies were calculated by the projected augmented planewave (PAW) of density function-al theory, as implemented in the CASTEP code embedded in the Materials Studio software, Version 5.5.2 [28]. The core electrons were described using the projector augmented method. The exchange–correlation energy of the electrons was treated using the generalized gradient approximation within the Perdew–Burke–Ernzerhof functional [29]. For geometry optimization, the Monkhorst–Pack scheme was used. Ground state geometries were calculated by minimizing stresses and Hellman–Feynman forces using the conjugate gradient algorithm with the convergence parameters set as follows: total energy tolerance 2 <sup>×</sup> <sup>10</sup>−<sup>6</sup> eV/atom, maximum force tolerance 0.05 eV/nm and maximum stress component 0.1 GPa. From the convergence test calculations, the basis set kinetic energy cutoff of 600 eV was sufficient to ensure optimum accuracy in the computed results. The dimensions for nanoclusters were set 20.00 × 20.00 × 20.00 Å, large enough to ensure that there were no interactions between the system and its self-image along all the axes within the periodic boundary conditions. Different sizes of the nanoparticles were constructed from the optimized bulk NiO system using supercells of varying sizes. The NiO nanoparticle was constructed by cleaving the bulk system along the (110) plane. Afterward, the Li ions were inserted into the nanoparticles using two different doping mechanisms, i.e., interstitial and substitutional doping. The k-points were generated using the Monkhorst-Pack method with a grid size of 1 × 1 × 1 for structural optimization for nanoparticle. The stoichiometry was maintained throughout nanoparticle construction and the vacuum was included in the x, y and z directions. The vacuum thickness was considered wide enough to prevent nanoparticle-to-nanoparticle interactions and 20 Å was sufficient to ensure that the energy converged to less than 1 meV/atom.

#### **3. Results and Discussion**

#### *3.1. XRD Analysis*

X-ray diffraction analysis was carried out to investigate the structures of pure NiO and Li doped NiO samples. Figure 3 shows the XRD peaks of pure NiO and samples prepared with (Sample B) and without (Sample A) surfactant, along with Li doped NiO (Samples C, D and E). The XRD patterns reveal the polycrystalline nature of the samples. The observed 2θ values are in good agreement with the standard JCPDS data (Card ID 75-0197). All the peaks are well indexed, with the favoured orientations being (111), (200) and (220) planes [23]. From the analysis of the peak positions and comparative intensities of the diffracted peaks, it was confirmed that the samples were monophasic, Fm-3m cubic NiO. Since the sample is in powder form, Scherrer equation (Equation (1)) was used to relate the size of the crystallites to the broadening of the peak. The crystallite sizes, evaluated using the (200) peak as reference, are tabulated in Table 1.

$$\pi = \text{K}\lambda / \beta \text{os}\theta \tag{1}$$

where τ is the mean size, *K* is a dimensionless shape factor with a value of 0.94, λ is X-ray wavelength, β is the FWHM value and θ is the Bragg angle. From Table 1, it can be inferred that the presence of PEG during the synthesis of nanoparticles inhibits grain growth, as manifested by the peak broadening. PEG plays a crucial role in modifying the surface properties, thereby arresting grain growth. It is also observed that the Li doping of NiO also modified its crystallite size. This may be mainly due to the slightly larger size and lower charge of Li<sup>+</sup> compared to that of the Ni2+; ionic radii of Li<sup>+</sup> and Ni2+, which are 0.74 Å and 0.69 Å, respectively [30]. This suggests that, upon doping with Li, the dopant ions enter the rock salt crystal lattice of NiO substitutionally, and the surface stress may retard the growth of NiO nanomaterials. However, what is curious is that, at larger dopant concentrations (at 5%), this behaviour is not so pronounced. As the concentration of dopant increases, the peak intensity increases, revealing the higher crystallinity of the samples. This is reflected in the gradual increase in the crystallite size of the doped samples. Because of the low activation energy and high ionic mobility, the Li ions enter the nucleation sites, leading to increased strain and grain size [31]. A similar trend was observed by Matsubara et al. while doping NiO with Li up to 15 wt.%, who attributed this to the small amount of lattice contraction of NiO matrix on efficient substitution of Li<sup>+</sup> for Ni2<sup>+</sup> [32]. *Materials* **2020**, *13*, x FOR PEER REVIEW 6 of 18 **Sample Size (nm)**  A 3.08 B 2.47 C 2.36 D 2.43 E 2.80

**Figure 3.** XRD pattern of A–E samples. **Figure 3.** XRD pattern of A–E samples.

*3.2. Surface Morphology*  **Table 1.** Crystallite size calculated using X-ray diffractometer.


concentration increases, the particles display a more prominent spherical shape.

compared to the undoped samples, the doped samples exhibit a more uniform micrograph. For samples C, D and E, the small crystallite size leads to agglomeration. However, as the dopant

#### *3.2. Surface Morphology*

*3.3. Computational Studies* 

size.

Figure 4 shows the FESEM images of pure NiO and Li-doped samples with different concentrations of Li. It is clear that the NiO particles are of spherical shape with a size distribution in the 50–100 nm range. It is evident that the obtained NiO particles are porous in nature. This suggests that since, during the synthesis, a large amount of gases evolve, this results in the disintegration of agglomerates and non-uniform dissipation of heat within the system. This can hinder the grain growth and eventually lead to an increase in the surface area and porosity of the material. When compared to the undoped samples, the doped samples exhibit a more uniform micrograph. For samples C, D and E, the small crystallite size leads to agglomeration. However, as the dopant concentration increases, the particles display a more prominent spherical shape. *Materials* **2020**, *13*, x FOR PEER REVIEW 7 of 18

**Figure 4.** SEM images of (**A**) pristine NiO, (**B**) pristine NiO with PEG, (**C**) NiO with PEG and 1 wt% Li, (**D**) NiO with PEG and 2 wt% Li, (**E**) NiO with PEG and 5 wt% Li. **Figure 4.** SEM images of (**A**) pristine NiO, (**B**) pristine NiO with PEG, (**C**) NiO with PEG and 1 wt% Li, (**D**) NiO with PEG and 2 wt% Li, (**E**) NiO with PEG and 5 wt% Li.

(110) and (111). The (100) is the most stable surface, with the surface energy of 1.15 J/m2 [34]. The constructed nanoparticles for NiO and Li-inserted NiO are depicted in Figure 5 and have an octahedron geometry with a diameter of 3 nm. All nanoparticles possess a high symmetry, and their surfaces are characterized by (100) Miller index. A recent study investigating the effects of NiO nanoparticle surface energies on catalytic efficiency showed that (110) and (111) dominate the nanoparticle surface [35]. The Li doping of NiO has an impact on its crystal structure and also crystal

#### *3.3. Computational Studies*

Geometrically, NiO is a stacked rocksalt structure with a space group Fm-3m and lattice parameter 4.17 Å [33]. Three possible surface orientations for NiO can be obtained, which are (100), (110) and (111). The (100) is the most stable surface, with the surface energy of 1.15 J/m<sup>2</sup> [34]. The constructed nanoparticles for NiO and Li-inserted NiO are depicted in Figure 5 and have an octahedron geometry with a diameter of 3 nm. All nanoparticles possess a high symmetry, and their surfaces are characterized by (100) Miller index. A recent study investigating the effects of NiO nanoparticle surface energies on catalytic efficiency showed that (110) and (111) dominate the nanoparticle surface [35]. The Li doping of NiO has an impact on its crystal structure and also crystal size. *Materials* **2020**, *13*, x FOR PEER REVIEW 8 of 18

**Figure 5.** Constructed nanoparticles for NiO and Li-inserted NiO. **Figure 5.** Constructed nanoparticles for NiO and Li-inserted NiO.

Computational modelling of optical functions for solids gives insights into understanding the optical properties of different materials. Due to the limitations and failure of GGA to capture the properties of strongly correlated electronic systems such as antiferromagnetic NiO, the simulated energy band gaps for the nanoparticles were significantly underestimated. The SCF energy is sensitive to the pseudopotential as well as the exchange and correlation functional used in the DFT calculation, hence the relative energies of simulated NiO nanoparticles are dependent on ground Computational modelling of optical functions for solids gives insights into understanding the optical properties of different materials. Due to the limitations and failure of GGA to capture theproperties of strongly correlated electronic systems such as antiferromagnetic NiO, the simulated energy band gaps for the nanoparticles were significantly underestimated. The SCF energy is sensitive to the pseudopotential as well as the exchange and correlation functional used in the DFT calculation, hence the relative energies of simulated NiO nanoparticles are dependent on ground state energy.

state energy. The frequency-dependent dielectric function is given by:

> π

ε ω

conductivity (real part), are derived from

*3.4. Diffuse Reflectance Spectral (DRS) Analysis* 

$$
\varepsilon\left(\omega\right) = \varepsilon\_1(\omega) + i\varepsilon\_2(\omega)\tag{2}
$$

() () 1 2 εω εω εω ) = + *i* ( (2) The dielectric function is closely related to the electronic band structure and it fully describes the optical properties of any homogeneous medium at all photon energies. The imaginary part of the The dielectric function is closely related to the electronic band structure and it fully describes the optical properties of any homogeneous medium at all photon energies. The imaginary part of the complex dielectric function is obtained from the momentum matrix elements between the occupied and unoccupied electronic states. The imaginary part is calculated using the analytical expression

complex dielectric function is obtained from the momentum matrix elements between the occupied

$$\varepsilon\_{2}(\omega) = \frac{2e^{2}\pi}{\Omega\varepsilon\_{0}} \sum\_{k,p,\varepsilon} \left| \left< \psi\_{k}^{\varepsilon} \right| \hat{\boldsymbol{\mu}} \bullet \boldsymbol{\overline{r}} \right| \left< \psi\_{k}^{p} \right|^{2} \delta \left( E\_{k}^{\varepsilon} - E\_{k}^{\upsilon} - E \right) \tag{3}$$

( ) ( ) 2 0 , , <sup>2</sup> <sup>ˆ</sup> *c v cv k k kk kvc <sup>e</sup> ur E E E* εω ψ ψ δ ε = • −− <sup>Ω</sup> (3) where ω is the frequency of light, *e* is the electronic charge, *u*ˆ is the vector defining the polarization of the incident electric field, *<sup>c</sup>* ψ *<sup>k</sup>* and *<sup>v</sup>* ψ *<sup>k</sup>* are the conduction and valence band wave where ω is the frequency of light, *e* is the electronic charge, *u*ˆ is the vector defining the polarization of the incident electric field, ψ *c k* and ψ *v k* are the conduction and valence band wave functions at *k*, respectively. The real part of the dielectric function ε<sup>1</sup> (ω) is derived from the imaginary part of the dielectric function ε<sup>2</sup> (ω) through the Kramers–Kronig relationship. Other optical properties, such as refractive index, absorption spectrum, loss function, reflectivity, and conductivity (real part), are derived from

$$\varepsilon\_1(\omega) = 1 + \frac{2}{\pi} P \int\_0^\infty \frac{\omega' \varepsilon\_1(\omega') d\omega'}{(\omega'^2 - \omega^2)}\tag{4}$$

ε ω

π

0

The reflectance, transmittance and scattering of the synthesized materials were determined using diffuse reflectance spectra. Figure 6 illustrates the diffuse reflectance spectra of different NiO samples. It can be seen from Figure 6a that each sample exhibits an absorption edge and the reflectance decreases with the addition of Li. The spectra were further used to plot the optical energy gap spectra of NiO samples as indicated in Figure 6b. The band gap decreases from 3.3 eV (undoped

1

ωε ω ω

> ωω

samples.

can be given as

#### *3.4. Di*ff*use Reflectance Spectral (DRS) Analysis*

The reflectance, transmittance and scattering of the synthesized materials were determined using diffuse reflectance spectra. Figure 6 illustrates the diffuse reflectance spectra of different NiO samples. It can be seen from Figure 6a that each sample exhibits an absorption edge and the reflectance decreases with the addition of Li. The spectra were further used to plot the optical energy gap spectra of NiO samples as indicated in Figure 6b. The band gap decreases from 3.3 eV (undoped NiO) to a minimum of 3.17 eV (NiO doped with 1% Li). The variation in the band gap cannot be explained plainly on the basis of Quantum Size Effect. This is because, as seen from the XRD data, the particle size obtained is beyond quantum size confinement. *Materials* **2020**, *13*, x FOR PEER REVIEW 9 of 18 NiO) to a minimum of 3.17 eV (NiO doped with 1% Li). The variation in the band gap cannot be explained plainly on the basis of Quantum Size Effect. This is because, as seen from the XRD data, the particle size obtained is beyond quantum size confinement.

**Figure 6.** (**a**) Diffuse reflectance spectra of NiO samples. (**b**) Optical energy gap spectra of NiO **Figure 6.** (**a**) Diffuse reflectance spectra of NiO samples. (**b**) Optical energy gap spectra of NiO samples.

With an octahedral symmetry, the divalent nickel ions possess a 3d8 electronic configurations and the spin-allowed d–d transitions of octahedral Ni2+ ions are because of the single broad absorption band visible at 410 nm. As the powder sample diffuses the light in large quantities with a greater thickness, the absorption spectra becomes more complicated to interpret. In order to minimize this difficulty, DRS along with Schuster–Kubelka–Munk (SKM) relation has been used [36,37] which With an octahedral symmetry, the divalent nickel ions possess a 3d<sup>8</sup> electronic configurations and the spin-allowed d–d transitions of octahedral Ni2<sup>+</sup> ions are because of the single broad absorption band visible at 410 nm. As the powder sample diffuses the light in large quantities with a greater thickness, the absorption spectra becomes more complicated to interpret. In order to minimize this difficulty, DRS along with Schuster–Kubelka–Munk (SKM) relation has been used [36,37] which can be given as

$$F(R) = \frac{\left(1 - R\right)^2}{2R} \tag{5}$$

ଶ)ܴ − 1) <sup>=</sup> (ܴ(ܨ 2ܴ (5) where *R* is the absolute reflectance of the sample and *F(R)* is the Kubelka–Munk function.

where *R* is the absolute reflectance of the sample and *F(R)* is the Kubelka–Munk function. Electron excitation from the valance band to conduction band gives a measure of the optical band gap which can be estimated using the relation *(F(R) hυ)n = A(hυ-Eg),* where *n* = 2 and ½ for directly allowed and indirectly allowed transition respectively, *A* is the constant, and *hυ* is the photon energy [38]. Extrapolation of the linear part of the curve in Figure 6b to *(F(R)hυ)2 =* 0 leads to the direct band gap energy. The decrease in band gap can be attributed to the concentration of free carriers as well as impurity effect when NiO is doped. Normally, NiO and Li doped NiO are p-type semiconductors containing an excess of oxygen. On doping, some of the Ni2+ are replaced with Li+. This results in an increase in the concentration of Ni3+ and holes too. According to the Moss–Burstein effect [39], this should have caused a shift in the reflectance edge towards a higher photon energy. However, in the present case, the band gap decreases due to the following two Coulomb forces: (i) Electron excitation from the valance band to conduction band gives a measure of the optical band gap which can be estimated using the relation *(F(R) h*υ*) <sup>n</sup>* = *A(h*υ*-Eg),* where *n* = 2 and <sup>1</sup>/<sup>2</sup> for directly allowed and indirectly allowed transition respectively, *A* is the constant, and *h*υ is the photon energy [38]. Extrapolation of the linear part of the curve in Figure 6b to *(F(R)h*υ*) <sup>2</sup>* = 0 leads to the direct band gap energy. The decrease in band gap can be attributed to the concentration of free carriers as well as impurity effect when NiO is doped. Normally, NiO and Li doped NiO are p-type semiconductors containing an excess of oxygen. On doping, some of the Ni2<sup>+</sup> are replaced with Li+. This results in an increase in the concentration of Ni3<sup>+</sup> and holes too. According to the Moss–Burstein effect [39], this should have caused a shift in the reflectance edge towards a higher photon energy. However, in the present case, the band gap decreases due to the following two Coulomb forces: (i) the exchange and correlation energy between the holes and the electrons in the valence and conduction band, respectively; and (ii) interaction between holes and impurity ions.

Figure 7, along with experimental measurements. As observed experimentally, the nanoparticle depicts an absorption edge in the ultraviolet region. Optical absorption curve showed that the nanoparticles absorbance is attributed by peaks ranging from 100 to 400 nm. This suggests that the Li doped NiO particles can be stable at wavelengths below 400 nm. Furthermore, simulations showed

the exchange and correlation energy between the holes and the electrons in the valence and

conduction band, respectively; and (ii) interaction between holes and impurity ions.

a long wavelength activity in the visible light region.

Optical properties for NiO and Li doped systems were calculated using the first-principle density functional theory method. Simulated reflectivity as a function of wavelength is presented in Figure 7, along with experimental measurements. As observed experimentally, the nanoparticle depicts an absorption edge in the ultraviolet region. Optical absorption curve showed that the nanoparticles absorbance is attributed by peaks ranging from 100 to 400 nm. This suggests that the Li doped NiO particles can be stable at wavelengths below 400 nm. Furthermore, simulations showed a long *Materials*  wavelength activity in the visible light region. **2020**, *13*, x FOR PEER REVIEW 10 of 18

**Figure 7.** Simulated reflectivity as a function of wavelength. **Figure 7.** Simulated reflectivity as a function of wavelength.

#### *3.5. Photoluminescence Studies (PL)*

confirmed by CV and EIS analysis.

*3.5. Photoluminescence Studies (PL)*  The photoluminescence excitation spectra (Figure 8) of the prepared samples at emission wavelength of 410 nm exhibit a prominent broad band at 308 nm along with an intense peak at 371 nm, corresponding, respectively, to spin-allowed 3T1g(F) ← 3A2gand3T1g(P) ← 3A2g transitions of Ni2+ ions [40]. The intensity and position of these bands are characteristic of octahedral Ni2+. The photoluminescence excitation spectra (Figure 8) of the prepared samples at emission wavelength of 410 nm exhibit a prominent broad band at 308 nm along with an intense peak at 371 nm, corresponding, respectively, to spin-allowed 3T1g(F) ← 3A2gand3T1g(P) ← 3A2g transitions of Ni2<sup>+</sup> ions [40]. The intensity and position of these bands are characteristic of octahedral Ni2+. *Materials* **2020**, *13*, x FOR PEER REVIEW 11 of 18

**Figure 8.** Excitation spectrum of NiO emission wavelength monitored at 410 nm. **Figure 8.** Excitation spectrum of NiO emission wavelength monitored at 410 nm.

PL studies on NiO nanoparticles [41,42].The origin of photoluminescence peaks can be attributed to electronic transitions involving 3d8 electrons of the Ni2+ ions [43].The presence of Li+ resulted in the formation of several shoulder peaks: at 402 and 422 nm (violet emission), 452 nm (blue emission) and 508 nm (green emission).The violet emission peaks are probably due to the transition of trapped electrons at interstitial Ni to the valence band. The blue emissions are due to the recombination of electrons from the Ni2+ vacancy to the holes in the valence band [41]. The cause of the green emission peak is not yet clear, as some authors cite it to be due to increases in Ni vacancies [44] whereas others relate it to the oxygen vacancies [45]. The addition of lithium influences the PL spectra profoundly; samples D and E have a lower emission intensity, whereas sample C has a higher emission intensity, than pristine NiO. The UV and visible emission peak intensity is dependent on radiative recombination. At a lower concentration of dopant, the radiative recombination process is higher, emitting more energy and thereby intensifying the peak. A small amount of dopant results in the replacement of Ni2+ with Li+, generating one hole in the valence band to maintain charge neutrality [46]. However, as the dopant concentration increases, it induces higher defects. This probably leads to non-radiative recombination, which subsequently reduces the peak intensity [47]. Interestingly, the sample B (NiO–surfactant) exhibits a high-intensity broad peak from 400–550 nm with maxima at 440 nm. A similar result has been reported by Wang and his group for NiO synthesized using dodecylamine as surfactant. The authors have attributed the broad nature of PL emission to the multilayer structure formed by layered NiO–surfactant superlattices. This kind of structure is expected to influence the chemical and physical properties of NiO to a large extent [48]. The oxygen vacancies may interact with interfacial capping surfactants, forming a series of metastable energy levels within the band gap. The long lifetime and dipole allowed for by transitions in these metastable energy levels induces the interfacial effect between NiO and the surfactant. Such unusual room temperature photoluminescences have also been previously observed by Zou and his group for nanoparticles coated with stearic acid [49] and by Bai and co-workers for mesolamellar TiO2 structures [50]. The intensity emission peak indicates enhanced photoluminescence intensity with high charge transfer resistance and, hence, a decrease in the electrochemical behavior further

The PL emission spectra (Figure 9) of the samples, upon excitation at 308 nm, shows two PL

The PL emission spectra (Figure 9) of the samples, upon excitation at 308 nm, shows two PL peaks that are obvious at 345 and 466 nm, corresponding to ultraviolet emission (340–400 nm) and blue emission (450–495 nm), respectively. Similar observations have been made in the literature citing PL studies on NiO nanoparticles [41,42].The origin of photoluminescence peaks can be attributed to electronic transitions involving 3d<sup>8</sup> electrons of the Ni2<sup>+</sup> ions [43].The presence of Li<sup>+</sup> resulted in the formation of several shoulder peaks: at 402 and 422 nm (violet emission), 452 nm (blue emission) and 508 nm (green emission).The violet emission peaks are probably due to the transition of trapped electrons at interstitial Ni to the valence band. The blue emissions are due to the recombination of electrons from the Ni2<sup>+</sup> vacancy to the holes in the valence band [41]. The cause of the green emission peak is not yet clear, as some authors cite it to be due to increases in Ni vacancies [44] whereas others relate it to the oxygen vacancies [45]. The addition of lithium influences the PL spectra profoundly; samples D and E have a lower emission intensity, whereas sample C has a higher emission intensity, than pristine NiO. The UV and visible emission peak intensity is dependent on radiative recombination. At a lower concentration of dopant, the radiative recombination process is higher, emitting more energy and thereby intensifying the peak. A small amount of dopant results in the replacement of Ni2<sup>+</sup> with Li+, generating one hole in the valence band to maintain charge neutrality [46]. However, as the dopant concentration increases, it induces higher defects. This probably leads to non-radiative recombination, which subsequently reduces the peak intensity [47]. Interestingly, the sample B (NiO–surfactant) exhibits a high-intensity broad peak from 400–550 nm with maxima at 440 nm. A similar result has been reported by Wang and his group for NiO synthesized using dodecylamine as surfactant. The authors have attributed the broad nature of PL emission to the multilayer structure formed by layered NiO–surfactant superlattices. This kind of structure is expected to influence the chemical and physical properties of NiO to a large extent [48]. The oxygen vacancies may interact with interfacial capping surfactants, forming a series of metastable energy levels within the band gap. The long lifetime and dipole allowed for by transitions in these metastable energy levels induces the interfacial effect between NiO and the surfactant. Such unusual room temperature photoluminescences have also been previously observed by Zou and his group for nanoparticles coated with stearic acid [49] and by Bai and co-workers for mesolamellar TiO<sup>2</sup> structures [50]. The intensity emission peak indicates enhanced photoluminescence intensity with high charge transfer resistance and, hence, a decrease in the electrochemical behavior further confirmed by CV and EIS analysis.

#### *3.6. Cyclic VoltammetryAnalysis*

The cyclic voltammetric studies of the synthesized samples are shown in Figure 10. The CV curve has definite symmetric anodic and cathodic peaks, indicating the good reversibility of the redox reactions. However, by increasing the scan rates, no significant change was observed within the material. As evidenced by CV studies, the electrochemical process of NiO electrodes is limited by the proton diffusion through the lattice [51–53]. By taking into account the difference between the oxidation potential (*EO*) and the reduction potential (*ER*) at a scan rate of 10 mV/s, the reversibility of the electrode reaction was measured. The reversibility increases as the *E<sup>O</sup>* − *E<sup>R</sup>* value decreases. From Table 2, it is evident that the reversibility of the electrode reaction was maximum for NiO sample prepared with 1% Li dopant. The electrode reactions in NiO proceed according to the following reaction:

$$\text{NiO} \xrightarrow[\text{Reduction}]{\text{Oxidation}} \text{Ni}^{2+} + \text{O}\_2^{2-} \tag{6}$$

The anodic and cathodic peaks indicate the oxidation of Ni<sup>0</sup> into Ni2<sup>+</sup> and reduction of Ni2<sup>+</sup> into Ni<sup>0</sup> , respectively. The quasi-reversible electron transfer process seen in the CV curve indicates the measured capacitance based on the redox mechanism [54]. The peak heights or the area of the CV curve for doped samples are larger than pristine samples, denoting a high amount of stored charge. This could be related to the smaller size of the doped samples resulting in a higher surface

reaction:

different NiO electrodes.

area. However, NiO with 2% Li exhibits a smaller CV curve when compared to the undoped samples. The reason for this deviation is unclear.

The quantity of current generated by electrode C is comparatively higher, and is least for electrode B. This suggests that 1% Li plays a crucial role during lattice formation which is manifested by generation of higher current when compared to the rest of the doped oxides. On the other hand, the presence of PEG as a surfactant during the synthesis has a great impact on the surface morphology. This is explicit by a higher PL intensity, as well as a decrease in the electrode reversibility. Probably, the surfactant enclosing NiO diminishes the development of the current. *Materials* **2020**, *13*, x FOR PEER REVIEW 12 of 18

**Figure 9.** Emission spectra of different NiO samples excited at 308 nm. **Figure 9.** Emission spectra of different NiO samples excited at 308 nm.



Table 2, it is evident that the reversibility of the electrode reaction was maximum for NiO sample prepared with 1% Li dopant. The electrode reactions in NiO proceed according to the following

ଶି (6)

NiO Niଶା + Oଶ

C 0.392 0.238 0.154 D 0.419 0.253 0.166 E 0.428 0.254 0.174

**Table 2.** Oxidation potential (*EO*), reduction potential (*ER*) and the difference between *EO* and *ER* of

*Materials* **2020**, *13*, x FOR PEER REVIEW 13 of 18

**Figure 10.** Cyclic voltammogram of (**A**) pristine NiO, (**B**) pristine NiO with PEG, (**C**) NiO with PEG and 1 wt% Li, (**D**) NiO with PEG and 2 wt% Li, (**E**) NiO with PEG and 5 wt% Li at various scan rates. **Figure 10.** Cyclic voltammogram of (**A**) pristine NiO, (**B**) pristine NiO with PEG, (**C**) NiO with PEG and 1 wt% Li, (**D**) NiO with PEG and 2 wt% Li, (**E**) NiO with PEG and 5 wt% Li at various scan rates.

#### The anodic and cathodic peaks indicate the oxidation of Ni0 into Ni2+ and reduction of Ni2+ into *3.7. Impedance Studies*

Ni0, respectively. The quasi-reversible electron transfer process seen in the CV curve indicates the measured capacitance based on the redox mechanism [54]. The peak heights or the area of the CV curve for doped samples are larger than pristine samples, denoting a high amount of stored charge. This could be related to the smaller size of the doped samples resulting in a higher surface area. However, NiO with 2% Li exhibits a smaller CV curve when compared to the undoped samples. The reason for this deviation is unclear. The quantity of current generated by electrode C is comparatively higher, and is least for electrode B. This suggests that 1% Li plays a crucial role during lattice formation which is manifested by generation of higher current when compared to the rest of the doped oxides. On the other hand, The electrochemical impedance measurements were carried out in the frequency range of 1 Hz to 1 MHz at 5 mV steady state amplitude. The corresponding Nyquist plots are shown in Figure 11. The plots suggest a larger impedance for electrode A (semicircle with a bigger diameter) while the impedance of the electrode C was found to be smaller (Table 3), as is evident by the shifting of the imaginary line towards Y-axis. Consequently, electrode C manifests higher discharge rates and capacitance. The increase in the capacitance can also be attributed to the combined effect of the electric double-layer capacitance on the high surface area of AC and pseudocapacitance via the intercalation/extraction of Li ions in NiO lattice.

the presence of PEG as a surfactant during the synthesis has a great impact on the surface

intercalation/extraction of Li ions in NiO lattice.

*3.7. Impedance Studies* 

morphology. This is explicit by a higher PL intensity, as well as a decrease in the electrode reversibility. Probably, the surfactant enclosing NiO diminishes the development of the current.

The electrochemical impedance measurements were carried out in the frequency range of 1 Hz to 1 MHz at 5 mV steady state amplitude. The corresponding Nyquist plots are shown in Figure 11. The plots suggest a larger impedance for electrode A (semicircle with a bigger diameter) while the impedance of the electrode C was found to be smaller (Table 3), as is evident by the shifting of the imaginary line towards Y-axis. Consequently, electrode C manifests higher discharge rates and capacitance. The increase in the capacitance can also be attributed to the combined effect of the electric

**Figure 11.** Nyquist plots of samples A, B, C, D and E. **Figure 11.** Nyquist plots of samples A, B, C, D and E.

**Table 3.** The EIS fitted circuit parameters of *RCt* and *Cdl* values. **Table 3.** The EIS fitted circuit parameters of *RCt* and *Cdl* values.


E 8.091 × 10−4 1.201 × 10−<sup>7</sup> Since the Nyquist plots reveal the presence of depressed semicircles with a centre below the real axis at higher frequencies, it becomes necessary to use a constant phase element (*Q*1) to fit the data Since the Nyquist plots reveal the presence of depressed semicircles with a centre below the real axis at higher frequencies, it becomes necessary to use a constant phase element (*Q*1) to fit the data into an equivalent circuit. The impedance of *Q*<sup>1</sup> can be described [55,56] as

$$Z\_{\rm CPE} = \frac{1}{Y(j\omega)^n} \tag{7}$$

ܻ(݆߱) (6) where ω is the angular frequency in rad s−<sup>1</sup> , *Y* and *n* are adjustable parameters of constant phase element (*Q*1). For double layer capacitance, the value of *n* = 1, for resistance and Warburg diffusion *n* = 0 and *n* = 0.5, respectively.

The equivalent circuit for the Nyquist plots of impedance measurements of NiO electrodes A–E is shown in Figure 12. In the given circuit, the high frequency region corresponds to solution resistance (*Rs*) at the electrode–electrolyte interface. The semicircles are attributed to an interfacial charge transfer resistance (*Rct*) or polarization resistance (*Rp*) to which the double-layer capacitance (*C*) is connected in parallel. The low frequency region straight line is represented by the Warburg element (W) in series with *Rct*. Warburg element is an estimation of the redox reactions occurring in the system, i.e., the diffusion of electrons from the working electrode and deposition of nickel ions from the electrolyte into the pores on the electrode surface [57], the electrolytic diffusion of ions takes place during the transition from the high-frequency semicircle to the mid-frequency point [58]. The constant phase element (*Q*1) lies parallel to the charge-transfer resistance (*Rct*), as does the low frequency capacitance (*Q*2) to the leakage resistance (*R<sup>l</sup>* ).

frequency capacitance (*Q*2) to the leakage resistance (*Rl*).

= 0 and *n* = 0.5, respectively.

where *ω* is the angular frequency in rad s−1, *Y* and *n* are adjustable parameters of constant phase element (*Q*1). For double layer capacitance, the value of *n* = 1, for resistance and Warburg diffusion *n* 

The equivalent circuit for the Nyquist plots of impedance measurements of NiO electrodes A–E is shown in Figure 12. In the given circuit, the high frequency region corresponds to solution resistance (*Rs*) at the electrode–electrolyte interface. The semicircles are attributed to an interfacial charge transfer resistance (*Rct*) or polarization resistance (*Rp*) to which the double-layer capacitance (*C*) is connected in parallel. The low frequency region straight line is represented by the Warburg element (W) in series with *Rct*. Warburg element is an estimation of the redox reactions occurring in the system, i.e., the diffusion of electrons from the working electrode and deposition of nickel ions from the electrolyte into the pores on the electrode surface [57], the electrolytic diffusion of ions takes place during the transition from the high-frequency semicircle to the mid-frequency point [58]. The

**Figure 12.** Equivalent circuit for Nyquist plot of samples A, B, C, D and E. **Figure 12.** Equivalent circuit for Nyquist plot of samples A, B, C, D and E.

EIS-fitted circuit parameters are tabulated in Table 3. The data were obtained by fitting the experimental data as per the equivalent circuit. A decrease in *Rct* and an increase in *Cdl* indicates an enhancement in the electrochemical activity of the electrode. It is evident from Table 3 that the electrochemical activity of the electrode *C* (1% Li dopant) was higher, which may be attributed to the Li grains' effectiveness in current collection and further improves the charge transfer process on interface between electrolyte and electrode. However, this effect subsides with increasing Li concentration. On the other hand, due to the surfactant effect there is an increase in the *Rct* value and a decrease in *Cdl* value. This is consistent with the PL emission results. EIS-fitted circuit parameters are tabulated in Table 3. The data were obtained by fitting the experimental data as per the equivalent circuit. A decrease in *Rct* and an increase in *Cdl* indicates an enhancement in the electrochemical activity of the electrode. It is evident from Table 3 that the electrochemical activity of the electrode *C* (1% Li dopant) was higher, which may be attributed to the Li grains' effectiveness in current collection and further improves the charge transfer process on interface between electrolyte and electrode. However, this effect subsides with increasing Li concentration. On the other hand, due to the surfactant effect there is an increase in the *Rct* value and a decrease in *Cdl* value. This is consistent with the PL emission results.

#### **4. Conclusions 4. Conclusions**

In summary, undoped and Li doped NiO nanoparticles have been successfully prepared by microwave technique. The nanostructures were found to be crystalline in nature, with a cubic structure. The crystallite size decreased after the introduction of an Li dopant as revealed from X-ray diffraction studies. A decrease in band gap and an ultraviolet-blue emission along with small amount of green emission suggests that the photoluminescence of NiO nanomaterials can be tuned by doping. The optical results were further confirmed by computational modelling. From CV measurements, it was observed that the reversibility of NiO sample was maximum with 1% Li dopant. The same electrode also had enhanced electrochemical property with the charge transfer resistance reducing to 5.592 × 10−<sup>8</sup> Ω. Through the present work, the optical and electrochemical behavior of Li doped NiO have been successfully demonstrated. We believe that these nanostructures can be applied to nanoscale electrochromic and electro-optical devices for various consumer and industrial In summary, undoped and Li doped NiO nanoparticles have been successfully prepared bymicrowave technique. The nanostructures were found to be crystalline in nature, with a cubic structure. The crystallite size decreased after the introduction of an Li dopant as revealed from X-ray diffraction studies. A decrease in band gap and an ultraviolet-blue emission along with small amount of green emission suggests that the photoluminescence of NiO nanomaterials can be tuned by doping. The optical results were further confirmed by computational modelling. From CV measurements, it was observed that the reversibility of NiO sample was maximum with 1% Li dopant. The same electrode also had enhanced electrochemical property with the charge transfer resistance reducing to 5.592 <sup>×</sup> <sup>10</sup>−<sup>8</sup> <sup>Ω</sup>. Through the present work, the optical and electrochemical behavior of Li doped NiO have been successfully demonstrated. We believe that these nanostructures can be applied to nanoscale electrochromic and electro-optical devices for various consumer and industrial applications.

applications. **Author Contributions:** Conceptualization, A.S.B.; methodology, R.R.; software, R.R.M.; validation, A.S.B., R.R., C.R.R. and S.C.P.; formal analysis, A.S.B. and M.S.S.; investigation, R.R.; resources, C.R.R., S.C.P., M.S.S. and **Author Contributions:** Conceptualization, A.S.B.; methodology, R.R.; software, R.R.M.; validation, A.S.B., R.R., C.R.R. and S.C.P.; formal analysis, A.S.B. and M.S.S.; investigation, R.R.; resources, C.R.R., S.C.P., M.S.S. and G.F.B.L.eS.; data curation, A.S.B.; writing—original draft preparation, A.S.B. and C.R.R.; writing—review and editing, M.S.S., G.F.B.L.eS.; visualization, R.R.; supervision, A.S.B. and M.S.S.; project administration, A.S.B.; funding acquisition, A.S.B. All authors have read and agreed to the published version of the manuscript.

**Funding:** Aarti S. Bhatt thanks the VGST, Govt. of Karnataka, India, (VGST/SMYSR/GRD-437/2014-15) for extending financial helps to carry out this research work. All the authors thank the University of Sao Paulo, Brazil for funding the APC.

**Conflicts of Interest:** The Authors declares that they have no Conflict of Interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

#### *Article*

## **A Study of Methylcellulose Based Polymer Electrolyte Impregnated with Potassium Ion Conducting Carrier: Impedance, EEC Modeling, FTIR, Dielectric, and Device Characteristics**

**Muaffaq M. Nofal <sup>1</sup> , Jihad M. Hadi <sup>2</sup> , Shujahadeen B. Aziz 3,4,\* , Mohamad A. Brza <sup>3</sup> , Ahmad S. F. M. Asnawi <sup>5</sup> , Elham M. A. Dannoun <sup>6</sup> , Aziz M. Abdullah <sup>7</sup> and Mohd F. Z. Kadir <sup>8</sup>**


**Abstract:** In this research, a biopolymer-based electrolyte system involving methylcellulose (MC) as a host polymeric material and potassium iodide (KI) salt as the ionic source was prepared by solution cast technique. The electrolyte with the highest conductivity was used for device application of electrochemical double-layer capacitor (EDLC) with high specific capacitance. The electrical, structural, and electrochemical characteristics of the electrolyte systems were investigated using various techniques. According to electrochemical impedance spectroscopy (EIS), the bulk resistance (*R<sup>b</sup>* ) decreased from 3.3 <sup>×</sup> <sup>10</sup><sup>5</sup> to 8 <sup>×</sup> <sup>10</sup><sup>2</sup> <sup>Ω</sup> with the increase of salt concentration from 10 wt % to 40 wt % and the ionic conductivity was found to be 1.93 <sup>×</sup>10−<sup>5</sup> S/cm. The dielectric analysis further verified the conductivity trends. Low-frequency regions showed high dielectric constant, *ε* 0 and loss, *ε*" values. The polymer-salt complexation between (MC) and (KI) was shown through a Fourier transformed infrared spectroscopy (FTIR) studies. The analysis of transference number measurement (TNM) supported ions were predominantly responsible for the transport process in the MC-KI electrolyte. The highest conducting sample was observed to be electrochemically constant as the potential was swept linearly up to 1.8 V using linear sweep voltammetry (LSV). The cyclic voltammetry (CV) profile reveals the absence of a redox peak, indicating the presence of a charge double-layer between the surface of activated carbon electrodes and electrolytes. The maximum specific capacitance, *Cs* value was obtained as 118.4 F/g at the sweep rate of 10 mV/s.

**Keywords:** MC polymer electrolyte; impedance study; ion transport; ftir analysis; TNM; LSV; CV analyses

#### **1. Introduction**

Besides improving energy and power efficiency, one of the remaining challenges in the development of energy storage systems, including smart grids, portable electronic

**Citation:** Nofal, M.M.; Hadi, J.M.; Aziz, S.B.; Brza, M.A.; Asnawi, A.S.F.M.; Dannoun, E.M.A.; Abdullah, A.M.; Kadir, M.F.Z. A Study of Methylcellulose Based Polymer Electrolyte Impregnated with Potassium Ion Conducting Carrier: Impedance, EEC Modeling, FTIR, Dielectric, and Device Characteristics. *Materials* **2021**, *14*, 4859. https:// doi.org/10.3390/ma14174859

Academic Editors: Diogo Miguel Franco dos Santos and Biljana Šljuki´c

Received: 2 June 2021 Accepted: 20 August 2021 Published: 26 August 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

devices, and hybrid vehicles, is to minimize manufacturing costs and reduce environmental pollution [1]. A more recent emphasis has been focused on solid polymer electrolytes (SPEs) as an alternative conventional organic sol–gel electrolyte. Dimensional stability, durability, a comparatively wide potential window above 1.5 V, and eco-friendliness are all properties of these materials [2]. In the technology area, natural polymers for fabricating SPEs have gained interest for application in electrochemical devices such as electrical double-layer capacitors (EDLCs) and proton batteries. Due to their exceptional chemical and mechanical performances, many studies have shown that natural SPEs exhibit a good potential for device applications [3,4]. Natural polymers are defined as materials that extensively happen in nature or are obtained from animals or plants. Natural polymers are vital to way of life as our human forms are based on them. Some of the examples of natural polymers are nucleic acid and proteins that happen in human body, natural rubber, silk, and methylcellulose (MC). MC is known to be competitively marketed and is environmentally safe. It has suitable film-forming characteristics with good mechanical and electrical properties. Through dative bonds, cations can interact with oxygen atoms of MC. As a consequence, MC comprises functional groups, such as alcohol (R-OH), ether (R-O-R), and ester (RCOOR) groups which are promising as an ion conduction mechanism due to their single pair of electrons. MC is also considered an amorphous polymer with its comparatively high glass transition temperature [5,6].

Supercapacitors consist of two porous electrodes separated by an ionically conducting electrolyte. The electrodes could be made of substances including polymers, carbon and metal oxides. Supercapacitors can be a favorable energy conversion device for a wide range of applications, where significant amounts of energy must be stored or released in a short period. A supercapacitor is classified into three major types, namely pseudocapacitors, EDLCs, and hybrid capacitors. Pseudo-capacitors undergo a fast Faradaic mechanism [7], some examples of which include under potential deposition, intercalation, and reduction-oxidation reactions using metal oxide-based electrodes or electroactive conducting polymer. However, EDLCs do not involve any Faradaic mechanisms. EDLCs only require the accumulation of ions induced by the adsorption of charge carrier at the electrode/electrolyte interfaces. Owing to the storage process, EDLC is the non-Faradaic mechanism [8]. The main features of EDLCs, such as reliability, high energy capacity, reversibility, and safety improvements have drawn considerable interest, and making it a strong choice for various applications [9].

Activated carbon electrodes play a crucial role in the fabricating of EDLC due to their good chemical and physical properties such as low cost and easy availability, and high conductivity above 10−<sup>4</sup> S/cm, which can be manufactured from a diversity of precursors [10,11]. As a result, coal is the most common supply of activated carbon production due to its availability, high content of carbons from 60% to 80%, and costeffectiveness [12,13].

T.-Y. Chen et al. [14] electrodeposited NiSe nanoparticles on a carbon nanotube (CNT) forest to prepare a porous and intertwined network (denoted as CNT@NiSe/stainless steel (SS)). They then used the CNT@NiSe/SS as a free-standing and multifunctional electrode for supercapacitor (SC) application. The CNT@NiSe/SS composite electrode showed excellent capacity retention of 85%, and higher specific capacity of 126 mA h g−<sup>1</sup> (1007 F g−<sup>1</sup> ) in comparison with individual CNTs and NiSe. Lien et al. [15] developed a co-solvent-in-deep eutectic solvent (DES) system by mixing acetonitrile and water with a typical DES electrolyte composed of lithium perchlorate and acetamide. They have also used hydrogel composed of reduced graphene oxide (rGO) and 1T(trigonal)-MoS2 as the electrode materials for SC application. The authors fabricated high voltage symmetric supercapacitors using hydrogel and hybrid DES as the electrode and electrolyte materials, respectively. The SC at an operating voltage of 2.3 V achieved the maximum energy density of 31.2 Wh/kg at a power density of 1164 W/kg. The fabricated SC also showed 91% capacitance retention after 20,000 cycles. Hsiang et al. [16] presented rationally materials design of an optimum NiCo2S<sup>4</sup> nanoparticle in a rGO matrix as a NiCo2S4/rGO nanocomposite. The

authors reported the enhancements in the materials technology, showing the NiCo2S4/rGO nanocomposite electrode material with a very good specific capacitance of 963–700 F/g at 1–15 A/g, long cycle life of 3000 cycles, and high capacitance retention of 70%. NiCo2S4/rGO nanocomposite electrode material with a very good specific capacitance of 963–700 F/g at 1–15 A/g, long cycle life of 3000 cycles, and high capacitance retention of 70%. Adding inorganic salt to a polymer provides ion mobility and the polymer host chain

capacitance retention after 20,000 cycles. Hsiang et al. [16] presented rationally materials design of an optimum NiCo2S4 nanoparticle in a rGO matrix as a NiCo2S4/rGO nanocomposite. The authors reported the enhancements in the materials technology, showing the

*Materials* **2021**, *14*, x FOR PEER REVIEW 3 of 21

Adding inorganic salt to a polymer provides ion mobility and the polymer host chain plays a crucial role in the ion transport mechanism of the polymer electrolytes. Consequently, ion motion arises across the amorphous area, which is aided by the segmental motion of the polymer chains [17]. The use of potassium complexed electrolyte films has been discovered to have some benefits over their lithium counterparts. Nadimicherla et al. [18] reported that the smaller ions such as (Li<sup>+</sup> and Mg2+) possess lower mobility compared to the larger cations of (K<sup>+</sup> and Zn2+) in polymer-based electrolytes. The smaller cations are entrenched or captured by the polymeric network. Furthermore, lithium–ion interactions with the polar polymer chains are stronger than potassium ions, and thus lithium–ion transport involves higher activation energy of 97.4 kJ mol−<sup>1</sup> [19]. The aim of this study is to prepare an SPE film using a biopolymer of MC doped with various concentrations of potassium iodide (KI) as the ionic source for application in EDLC device. We have investigated the effect of different KI concentration has on the conductivity of MC. Also, the electrolyte with the highest conductivity was employed in the EDLC and its decomposition potential and specific capacitance were investigated. Figure 1 depicts the schematic diagram of an EDLC cell. As seen in Figure 1, the electrolyte is inserted between two activated carbon (AC) electrodes and then packed in coin cells of CR2032 to fabricate the EDLC. The prepared EDLC device was sandwiched in a Teflon case holder with two stainless steel electrodes to investigate the capacitive behavior of the device. While measuring the impedance data of the films, the arrangement of the cell was stainless steel electrolyte film stainless steel. plays a crucial role in the ion transport mechanism of the polymer electrolytes. Consequently, ion motion arises across the amorphous area, which is aided by the segmental motion of the polymer chains [17]. The use of potassium complexed electrolyte films has been discovered to have some benefits over their lithium counterparts. Nadimicherla et al. [18] reported that the smaller ions such as (Li+ and Mg2+) possess lower mobility compared to the larger cations of (K+ and Zn2+) in polymer-based electrolytes. The smaller cations are entrenched or captured by the polymeric network. Furthermore, lithium–ion interactions with the polar polymer chains are stronger than potassium ions, and thus lithium–ion transport involves higher activation energy of 97.4 kJ mol−1 [19]. The aim of this study is to prepare an SPE film using a biopolymer of MC doped with various concentrations of potassium iodide (KI) as the ionic source for application in EDLC device. We have investigated the effect of different KI concentration has on the conductivity of MC. Also, the electrolyte with the highest conductivity was employed in the EDLC and its decomposition potential and specific capacitance were investigated. Figure 1 depicts the schematic diagram of an EDLC cell. As seen in Figure 1, the electrolyte is inserted between two activated carbon (AC) electrodes and then packed in coin cells of CR2032 to fabricate the EDLC. The prepared EDLC device was sandwiched in a Teflon case holder with two stainless steel electrodes to investigate the capacitive behavior of the device. While measuring the impedance data of the films, the arrangement of the cell was stainless steel electrolyte film stainless steel.

**Figure 1.** The schematic diagram of an EDLC cell. Adapted from reference [4]. **Figure 1.** The schematic diagram of an EDLC cell. Adapted from reference [4].

#### **2. Experimental Details 2. Experimental Details**

#### *2.1. Materials and Electrolyte Preparation 2.1. Materials and Electrolyte Preparation*

MC powder was used as a host polymeric raw material and KI salt was used as the ionic source. Both reagents were purchased from Sigma-Aldrich (Kuala Lumpur, Malaysia). The electrolytes were prepared using a solution casting technique by dissolving 1 g MC in 50 mL distilling water, with constant stirring, at room temperature for ~3 h. Subsequently, various amounts of KI salt were added to the MC solutions separately. The solutions were stirred continuously until a homogenous polymer–salt complex was obtained. The quantity of salt was varied from 10 to 40 weight percent (wt %) in steps of ten to obtain MC-KI electrolytes. The electrolyte samples were correspondingly specified as MCKI0, MCKI1, MCKI2, MCKI3, and MCK4 for MC incorporated with 0, 10, 20, 30,and40 wt % of KI. The choice of KI concentrations is based on the ability of the MC to accommodate and dissolve the salt. Eventually, the solutions were cast on four individual categorized glass MC powder was used as a host polymeric raw material and KI salt was used as the ionic source. Both reagents were purchased from Sigma-Aldrich (Kuala Lumpur, Malaysia). The electrolytes were prepared using a solution casting technique by dissolving 1 g MC in 50 mL distilling water, with constant stirring, at room temperature for ~3 h. Subsequently, various amounts of KI salt were added to the MC solutions separately. The solutions were stirred continuously until a homogenous polymer–salt complex was obtained. The quantity of salt was varied from 10 to 40 weight percent (wt %) in steps of ten to obtain MC-KI electrolytes. The electrolyte samples were correspondingly specified as MCKI0, MCKI1, MCKI2, MCKI3, and MCK4 for MC incorporated with 0, 10, 20, 30, and 40 wt % of KI. The choice of KI concentrations is based on the ability of the MC to accommodate and dissolve the salt. Eventually, the solutions were cast on four individual categorized glass Petri dishes and left at room temperature to slowly evaporate the solvent. The films were further dried by transferring the prepared films to a desiccator.

#### *2.2. Impedance Spectroscopy and FTIR Study*

Electrical impedance spectroscopy (EIS) at the SPE was conducted using a Z HI-tester (Nagano, Japan) at a DC potential was 0.04 V, onto which an Ac voltage of peak-to-peak amplitude 10 mV was superimposed, over a frequency range of 5 MH and 50 HZ.

The inductance-capacitance-resistance (LCR) meter (Z HI-tester) was used to study the solid polymer electrolyte's electrical impedance spectroscopy (EIS) in the frequency range of (50 Hz ≤ *f* ≤ 5 MHz). The DC potential was 0.04 V. An SPE film of geometric area of 2.01 cm<sup>2</sup> was kept between two stainless-steel electrodes by applying a spring pressure which is used to press the electrolyte films. The stainless-steel electrode was used as the working, reference, and counter electrodes while the reference and counter electrodes were combined together. The EIS data were fitted with the electric equivalent circuit (EEC) model. The common electrical elements such as resistors and capacitors are used in this model. The EEC model is simple method and provides the entire picture of the system [5].

A spotlight 400 Perkin-Elmer spectrometer (Malvern Panalytical Ltd., Malvern, UK) was employed to perform the Fourier Transforms Infrared (FTIR) spectroscopy measurements. The transmitting range was performed between 940 and 4000 cm−<sup>1</sup> with a resolution of 2 cm−<sup>1</sup> .

It is vital to use Equation (1) to measure the DC ionic conductivity (*σdc*) of the MCKI samples based on the bulk resistance (*R<sup>b</sup>* ) value [20,21]

$$
\sigma\_{dc} = \left(\frac{1}{R\_b}\right) \times \left(\frac{t}{A}\right) \tag{1}
$$

where *t* and *A* denote the sample thickness and electrode area, respectively. The dielectric constant (*ε* 0 ) and dielectric loss (*ε* 00) are obtained using Equations (2) and (3) [20,21].

$$\varepsilon' = \frac{Z''}{(Z'^2 + Z''^2)\mathbb{C}\_{\vartheta}\omega} \tag{2}$$

$$
\varepsilon'' = \frac{Z\prime}{(Z'^2 + Z''^2)\mathbb{C}\_o\omega} \tag{3}
$$

where, *ω* and *C<sup>o</sup>* denote the angular frequency and capacitance, which are given by (*ω =* 2*πf*) and *εoA*/*t*, respectively, where *ε<sup>o</sup>* stands for the free space permittivity, *A* the electrode area and *t* the thickness of the film [22].

The real and imaginary (*M<sup>i</sup>* and *Mr*) parts of complex electric modulus (*M*\*) were calculated using Equations (4) and (5) [23,24].

$$M' = \left[\frac{\varepsilon'}{\left(\varepsilon'^2 + \varepsilon'^2\right)}\right] = Z'' \mathbb{C}\_{\boldsymbol{\theta}} \boldsymbol{\omega} \tag{4}$$

$$M'' = \begin{bmatrix} \frac{\varepsilon''}{\left(\varepsilon'^2 + \varepsilon''^2\right)} \end{bmatrix} = Z' \mathbb{C}\_o \omega \tag{5}$$

#### *2.3. Study of Transference Number Measurement (TNM) and Linear Sweep Voltammetry (LSV)*

In TNM, two types of ionic transport, *tion* and electron transport *tel* for the most conducting sample (MCKI4) were studied. A DP3003 digital DC power supply (V & A instrument, Shanghai, China) was employed to polarize the cell against time at room temperature by applying a working voltage of 0.2 V. Linear sweep voltammetry (LSV) was used to determine the maximum potential window for the (MCKI4) film using a Digi-IVY DY2300 potentiostat (Neware, Shenzhen, China). The scan rate was fixed at 10 mV/s, and then the sample was sandwiched between two stainless steel electrodes with Teflon holders. Equations (6) and (7) were used to measure the transport ions (*tion*) and transport

electrons (*tel*) of the MCKI4 film, as the film was positioned between two stainless-steel electrodes [25].

$$t\_{\rm ion} = \frac{I\_{\rm i} - I\_{\rm ss}}{I\_{\rm i}} \tag{6}$$

$$t\_{el} = 1 - t\_{ion} \tag{7}$$

where *I<sup>i</sup>* refers to the initial current, containing ions and electrons and *Iss* stands for the current of the steady-state that contains only electrons.

#### *2.4. EDLC Fabrication*

Typically, the ingredients used to prepare electrodes include solvent and carbonaceous materials. In preparing the EDLC electrodes, 0.25 g of carbon black, 3.25 g of activated carbon, and 0.5 g of polyvinylidene fluoride (PVdF) were dry mixed in a planetary ball miller (XQM-0.4, Fujian, China) at 500 rpm for ~20 min. Then, all powders were dissolved and stirred continuously in 20 mL of N-methyl pyrrolidone until it became a dark black solution. In the next step, the black solution was covered by an aluminum foil using a doctor blade technique. Subsequently, an oven was used to dry the coated aluminum foil for a specific time at ~60 ◦C. To eliminate any excess moisture, the electrodes were placed in a silica gel desiccator. The relatively uppermost conducting sample was located between a pair of activated carbon electrodes and packaged in coin cells of CR2032. Eventually, in order to perform cyclic voltammetry (CV) of the assembled EDLC, the Digi-IVY DY2300 potentiostat has been employed at various scan rates of 10, 20, 50, and 100 mV/s and charged from 0 to 0.9 V. The specific capacitance, *C<sup>s</sup>* for the assembled EDLC has been determined using Equation (8) [25].

$$\mathbf{C}\_{\mathbf{cv}} = \int\_{V\_i}^{V\_f} \frac{I(V)dV}{2mv\left(V\_f - V\_i\right)}\tag{8}$$

where *V<sup>i</sup>* is the initial potential (i.e., 0 V), and *V<sup>f</sup>* is the final potential (i.e., 0.9 V), *m* and *υ* are the mass of active material and the potential sweep rates (mV/s), respectively. *I*(*V*)*dV* denotes the area under a cyclic voltammetric trace.

#### **3. Result and Discussion**

#### *3.1. Impedance Study*

Polymer electrolytes were commonly applied to devices as a part of an advanced material class. Impedance spectroscopy plays a crucial role in studying the electrical properties of a wide range of polymeric electrolyte materials. It is also a powerful technique for analyzing the ionic conductivity of new materials used in electrochemical energy systems, including EDLCs, charge transfer resistance, and diffusion layer. Plots of impedance spectra (*Z<sup>i</sup>* versus *Zr*) for the MCKI1, MCKI2, MCKI3, and MCKI4 systems are shown in Figure 2a–d. In general, the impedance responses are usually characterized by a semicircle in the high frequency region and a straight line in the low frequency region [26].

EIS data are commonly analyzed by fitting to an equivalent electrical circuit model (EEC). Most of the circuit elements in the model are common electrical elements such as resistors, capacitors, and inductors. To be useful, the elements in the model should have a basis in the physical electrochemistry of the system. The EEC method has been used to investigate the EIS because it is simple and shows the entire picture of the system [5,27]. The impedance diagrams in Figure 1 can generally be represented by an equivalent circuit consisting of a charge transfer resistance (*R<sup>b</sup>* ) in a parallel arrangement with constant phase element 1 (CPE1) in high frequency region and in a series arrangement with constant phase element 2 (CPE2) in the low frequency region, as shown in the inset of Figure 1. The impedance arising from CPE, ZCPE, is expressed by Equation (9) [5,27]

$$Z\_{\rm CPE} = \frac{1}{\mathcal{C}\omega^p} \left[ \cos\left(\frac{\pi p}{2}\right) - i\sin\left(\frac{\pi p}{2}\right) \right] \tag{9}$$

Here, *C* is the CPE capacitance, *ω* is the angular frequency and *p* is related to the EIS deviation from the imaginary axis. The *Z<sup>r</sup>* and *Z<sup>i</sup>* related to the EEC (insets of Figure 2a–d) are formulated by Equations (10) and (11) deviation from the imaginary axis. The *Zr* and *Zi* related to the EEC (insets of Figure 2a– d) are formulated by Equations (10) and (11) ଶଵଵ ( ଵ/2)+

*Materials* **2021**, *14*, x FOR PEER REVIEW 6 of 21

$$\begin{aligned} \text{My Equations (10) and (11)}\\ Z\_r &= \frac{R\_b^2 \mathbb{C}\_1 \omega^{p\_1} \cos(\pi p\_1/2) + R\_b}{2R\_b \mathbb{C}\_1 \omega^{p\_1} \cos(\pi p\_1/2) + R\_b^2 \mathbb{C}\_1^2 \omega^{2p\_1} + 1} + \frac{\cos(\pi p\_2/2)}{\mathbb{C}\_2 \omega^{p\_2}} \end{aligned} \tag{10}$$

Here, *C* is the CPE capacitance, *ω* is the angular frequency and *p* is related to the EIS

(10)

(11)

$$\mathbf{R}\_{i} = \frac{\mathbf{R}\_{b}\mathbf{C}\_{1}\omega\_{1}\omega\_{1} + \mathbf{R}\_{b}\omega\_{1}}{2R\_{b}\mathbf{C}\_{1}\omega\_{1}\omega\_{1}\cos(\pi p\_{1}/2)} + \frac{\mathbf{R}\_{b}\omega\_{1}}{R\_{b}\omega\_{1}\omega\_{1}\omega\_{1}} + \frac{\sin(\pi p\_{2}/2)}{\mathbf{C}\_{2}\omega\_{2}\omega\_{2}}\tag{11}$$

$$\mathbf{Z}\_{i} = \frac{R\_{b}^{2}\mathbf{C}\_{1}\omega\_{1}\omega\_{1}\cos(\pi p\_{1}/2) + R\_{p}^{2}\mathbf{C}\_{1}^{2}\omega\_{1}\omega\_{1} + 1}{2R\_{b}^{2}\omega\_{1}^{2}\omega\_{1}^{2} + 1} + \frac{\sin(\pi p\_{2}/2)}{\mathbf{C}\_{2}\omega\_{2}\omega\_{2}}\tag{12}$$

Here, *C*<sup>1</sup> is the capacitance of CPE1 at the bulk of the electrolyte; *C*<sup>2</sup> is the CPE2 capacitance at the electrode-electrolyte interface; *p*<sup>2</sup> is the offset from the real axis and *p*<sup>1</sup> is the offset of the semicircle from the imaginary axis. The fitting parameters in the EEC are listed in Table 1. As seen in Table 1, *C*<sup>1</sup> and *C*<sup>2</sup> increased with increasing salt concentration as the number density of ions increases and they transport from the bulk of the electrolyte to the surface of the electrodes. In addition, the conductivity is also increased with increasing salt amount due to the dissociation of more salts to ions and the decrease in the *R<sup>b</sup>* value as seen in Figure 2a–d. pacitance at the electrode-electrolyte interface; *p*2 is the offset from the real axis and *p*1 is the offset of the semicircle from the imaginary axis. The fitting parameters in the EEC are listed in Table 1. As seen in Table 1, *C*1 and *C*2 increased with increasing salt concentration as the number density of ions increases and they transport from the bulk of the electrolyte to the surface of the electrodes. In addition, the conductivity is also increased with increasing salt amount due to the dissociation of more salts to ions and the decrease in the *Rb* value as seen in Figure 2a–d.

**Figure 2.** Impedance plots for (**a**) MCKI1, (**b**) MCKI2, (**c**) MCKI3, and (**d**) MCKI 4 electrolyte films. **Figure 2.** Impedance plots for (**a**) MCKI1, (**b**) MCKI2, (**c**) MCKI3, and (**d**) MCKI 4 electrolyte films.

**Table 1.** The EEC fitting parameters for the systems fabricated

**Sample P1 (rad) P2 (rad) C1 (F) C2 (F)**  MCKI1 0.90 0.41 2 × 10−10 3.33 × 10−<sup>7</sup> MCKI2 0.87 0.42 4 × 10−10 1.43 × 10−<sup>6</sup> MCKI3 0.76 0.65 6.67 × 10−9 2.44 × 10−<sup>6</sup>


**Table 1.** The EEC fitting parameters for the systems fabricated.

The electrode polarization is responsible for appearing the spike in Figure 2a–d at the interfaces between electrodes and electrolytes owing to blockage of ions at the electrodeelectrolyte interfaces. Consequently, the electrode polarization outcome is caused by the formation of an electric double layer, resulting in free charge accumulation at the interfaces between electrodes and electrolytes. The linear increase in impedance in low frequency region in Figure 2 is expected to be a straight line (90 degree) parallel to the imaginary axis. However, there is an inclination by nearly 45◦ from the straight line due to the electrode polarization which causes to block of ions at the surface of the electrodes as seen in Figure 2a–d. Notably, the semicircular feature in the high frequency region has significantly diminished as KI was increased to 30 wt % and 40 wt %.

Equation (1) is used to compute the dc ionic conductivity by measuring the sample thickness and *R<sup>b</sup>* and the conductivity values are summarized in Table 2. As seen in Table 2, the dc conductivity increased when concentration of salt increased as more ions formed at higher salt concentration. From Equation (1), the lowest *R<sup>b</sup>* value shows the highest ionic conductivity [28]. It can be noted that the bulk resistance decreases with increasing the KI salt concentrations from 10 to 40 wt %. *µ* is related to the number density (*n*) and electrolyte conductivity (*σdc*) by Equation (12) [29]

$$
\sigma\_{\rm dc} = \mathfrak{ne}\mu \tag{12}
$$

where, *n* is the density of the charge carrier, *µ* denotes mobility of ions, and *e* denotes an electronic charge. It was established that the polymer electrolytes must have a dc ionic conductivity in the range between 10−<sup>3</sup> and 10−<sup>5</sup> S cm−<sup>1</sup> in order for it to be used in electrochemical devices [25,30,31]. Researchers have discovered that the conductivity value in this range is desirable for use in energy devices [25,30,31]. Shuhaimi et al. [32] were obtained the highest conductivity of 2.1 <sup>×</sup> <sup>10</sup>−<sup>6</sup> S cm−<sup>1</sup> for the system of MC-NH4NO<sup>3</sup> based biopolymer electrolyte.


**Table 2.** Numerical values of *σdc***,** *D*, *µ*, and *n* at ambient temperature.

As all the impedance data composed of a semicircular feature and a linear impedance, transport parameters including *D*, *µ* and *n* of ions are determined using the following equations [26,28]. The *D* of the ions is calculated using Equation (13),

$$D = \frac{\left(K\_2 \varepsilon\_o \varepsilon\_r A\right)^2}{\tau\_2} \tag{13}$$

where *ε<sup>r</sup>* is the dielectric constant, *τ<sup>2</sup>* is the reciprocal of angular frequency, which corresponds to the lowest value of *Z<sup>i</sup>* .

The *µ* of the ions is determined using Equation (14)

$$
\mu = \left[\frac{eD}{K\_B T}\right] \tag{14}
$$

where *T* is the absolute temperature and *K<sup>b</sup>* is the Boltzmann constant. **Table 2.** Numerical values of *σdc***,** *D*, *µ*, and *n* at ambient temperature

Since the *σdc* is given by Equation (12), the number density of ions (*n*) is calculated using Equation (15) **Sample** *σdc* **(S cm−1)**  *D*  **(cm2 s−1)**  *µ n*

$$n = \frac{\sigma\_{dc} \sigma\_{F}}{\left[ \left( e \mathbf{K}\_{2} \varepsilon\_{0} \varepsilon\_{F} A \right)^{2} \right]} \tag{15}$$

Table 2 lists the ion transport parameters for each electrolyte system. MCKI3 1.35 × 10−5 2.00 × 10−9 7.78 × 10−8 1.08 × 1021

Based on Table 2, the *D* increased as the KI concentration increased from 10 to 40 wt %. The identical tendency is seen by *µ* as listed in Table 2 where *µ* increased. The increase of *µ* and *D* is related to the increase of chain flexibility with the existence of slat [28]. Consequently, an improvement of conductivity is resulted. MCKI4 1.93 × 10−5 2.13 × 10−9 8.29 × 10−8 1.45 × 1021 Figure 3a,b show the Bode plot for each electrolyte film at room temperature. An

Figure 3a,b show the Bode plot for each electrolyte film at room temperature. An earlier study [33] indicated that the capacitive region is a plateau region between 10−<sup>2</sup> Hz and 100 Hz. However, this feature is not observed in Figure 3 because of the limitation of frequency of our measuring equipment. As described at the EIS plots, the semicircle is associated with ion transfer in the electrolyte and the linear feature arises from ions diffusion and therefore their accumulation at the interfaces between electrode and electrolyte [33] which leads to an electrical double-layer capacitances. It was shown that, by increasing the amount of salt from 10 wt % to 40 wt %, the linear feature increased and the resistance reduced from 3.3 <sup>×</sup> <sup>10</sup><sup>5</sup> to <sup>8</sup> <sup>×</sup> <sup>10</sup><sup>2</sup> <sup>Ω</sup>, because of the more carrier density. As seen in Figure 3a the electrolyte film has high charge transfer resistance (*Rct*) while with increasing salt the *Rct* decreased as shown in Figure 3b. The dispersion region between 40 Hz and 40,000 Hz is ascribed to the phenomena of ion diffusion and the high-frequency region is ascribed to the *Rct*. In Figures 2 and 3, it is seen that the sample loaded with 40 wt % of KI has the lowest *Rct* and hence a large conductivity resulted. Therefore, the Bode plot supports the result measured from the impedance study. earlier study [33] indicated that the capacitive region is a plateau region between 10−2 Hz and 100 Hz. However, this feature is not observed in Figure 3 because of the limitation of frequency of our measuring equipment. As described at the EIS plots, the semicircle is associated with ion transfer in the electrolyte and the linear feature arises from ions diffusion and therefore their accumulation at the interfaces between electrode and electrolyte [33] which leads to an electrical double-layer capacitances. It was shown that, by increasing the amount of salt from 10 wt % to 40 wt %, the linear feature increased and the resistance reduced from 3.3 × 105 to 8 × 102Ω, because of the more carrier density. As seen in Figure 3a the electrolyte film has high charge transfer resistance (*Rct*) while with increasing salt the *Rct* decreased as shown in Figure 3b. The dispersion region between 40 Hz and40,000 Hz is ascribed to the phenomena of ion diffusion and the high-frequency region is ascribed to the *Rct*. In Figures 2 and 3, it is seen that the sample loaded with 40 wt % of KI has the lowest *Rct* and hence a large conductivity resulted. Therefore, the Bode plot supports the result measured from the impedance study.

**Figure 3.** *Cont*.

**Figure 3.** Bode plots for (**a**) MCKI1 and MCKI2 and (**b**) MCKI3 and MCKI4 electrolyte samples. **Figure 3.** Bode plots for (**a**) MCKI1 and MCKI2 and (**b**) MCKI3 and MCKI4 electrolyte samples.

#### *3.2. Dielectric Properties 3.2. Dielectric Properties*

Complex electric modulus, defined as the inverse of complex relative permittivity, can be a significantly powerful tool for analyzing dielectric behavior of a polymeric insulating material, especially at relatively high temperatures, where complex permittivity usually becomes very high due to electrode polarization and carrier transport. The core of electrochemical devices are ions conducting solid electrolytes, and its electrical properties investigation such as *σdc*, ε\*, and electric modulus (M\*) are essential to understanding the ions transport process [21]. The real part (*ε′*) is related to ion storage efficiency or polarizing ability, while the imaginary part (*ε″*) is the necessary energy for dipole alignment [34]. Complex electric modulus, defined as the inverse of complex relative permittivity, can be a significantly powerful tool for analyzing dielectric behavior of a polymeric insulating material, especially at relatively high temperatures, where complex permittivity usually becomes very high due to electrode polarization and carrier transport. The core of electrochemical devices are ions conducting solid electrolytes, and its electrical properties investigation such as *σdc*, ε\*, and electric modulus (*M*\*) are essential to understanding the ions transport process [21]. The real part (*ε* 0 ) is related to ion storage efficiency or polarizing ability, while the imaginary part (*ε* 00) is the necessary energy for dipole alignment [34]. The *ε* 0 and *ε* 00 are determined using Equations (2) and (3).

The *ε′* and *ε′*' are determined using Equations (2) and (3). Figure 4a,b display the frequency dependency of the *ε′* and *ε″* for the MC polymer incorporated with various concentrations of KI salt. It can be noted that the system integrated with 40 wt % of KI has the highest dielectric constant at a low-frequency region. It might be owing to the electrode polarization and also space charge effects. The rise in dielectric constant can be explained by the high charge carrier concentration of the system and its amorphous composition [35]. It is seen that as the salt content (KI) increases, the *ε′* and *ε″* increase. This is in agreement with the increase in number density and mobility of ions when the KI content increased as shown in Table 2. Both of the *ε′* and *ε″* values are elevated at low frequencies and decreased as frequency rises, indicating polarization effect due to charge accumulations near electrodes at low frequency and dipoles do not obey the field variation at a high dispersion frequency region [36]. The dielectric values remain stable at high-frequency regions due to the interfaces of the electrode–electrolyte become marginal as the frequency increases. The decreased value of both *ε′* and *ε″* with increasing Figure 4a,b display the frequency dependency of the *ε* 0 and *ε* 00 for the MC polymer incorporated with various concentrations of KI salt. It can be noted that the system integrated with 40 wt % of KI has the highest dielectric constant at a low-frequency region. It might be owing to the electrode polarization and also space charge effects. The rise in dielectric constant can be explained by the high charge carrier concentration of the system and its amorphous composition [35]. It is seen that as the salt content (KI) increases, the *ε* 0 and *ε* 00 increase. This is in agreement with the increase in number density and mobility of ions when the KI content increased as shown in Table 2. Both of the *ε* 0 and *ε* 00 values are elevated at low frequencies and decreased as frequency rises, indicating polarization effect due to charge accumulations near electrodes at low frequency and dipoles do not obey the field variation at a high dispersion frequency region [36]. The dielectric values remain stable at high-frequency regions due to the interfaces of the electrode–electrolyte become marginal as the frequency increases. The decreased value of both *ε* 0 and *ε* 00 with increasing frequency means that the electrolyte films are non-Debye behavior [37].

frequency means that the electrolyte films are non-Debye behavior [37].

**Figure 4.** Dielectric plot for (**a**) *ε′* and (**b**) *ε″* variation against frequency for the MCKI samples. **Figure 4.** Dielectric plot for (**a**) *ε* 0 and (**b**) *ε* 00 variation against frequency for the MCKI samples.

The *Zr* and *Zi* data were achieved from the EIS data and then used to determine the *ɛ′* and *ɛ″* data. The *ɛ′* and *ɛ″* were used to find the tan *δ*. The tan *δ* is the ratio between energy disperse and energy stored in a periodical field which is also called dissipation factor [23] and it is determined using Equation (16). The *Z<sup>r</sup>* and *Z<sup>i</sup>* data were achieved from the EIS data and then used to determine the *ε* 0 and *ε* 00 data. The *ε* 0 and *ε* 00 were used to find the *tan δ*. The *tan δ* is the ratio between energy disperse and energy stored in a periodical field which is also called dissipation factor [23] and it is determined using Equation (16).

$$
\tan \delta = \frac{\varepsilon''}{\varepsilon'} \tag{16}
$$

Dielectric loss is the energy dissipation by the transfer of charges in an alternating electric field as polarization switches direction. When the electric field is applied, polarization happens and charges are moved relative to the electric field. Dielectric loss causes a decrease in the overall electric field. The total amount of polarization that can happen in a dielectric relies on the molecular symmetry of the insulator material and is known as dipole moment. The influence of the dipole moment in a dielectric material is called loss tangent. The ratio of *ɛ*″ to *ɛ′* is defined as tan *δ*, where *δ* denotes a loss angle. The tan *δ* is Dielectric loss is the energy dissipation by the transfer of charges in an alternating electric field as polarization switches direction. When the electric field is applied, polarization happens and charges are moved relative to the electric field. Dielectric loss causes a decrease in the overall electric field. The total amount of polarization that can happen in a dielectric relies on the molecular symmetry of the insulator material and is known as dipole moment. The influence of the dipole moment in a dielectric material is called loss tangent.

=

ᇳ

The ratio of *ε* 00 to *ε* 0 is defined as *tan δ*, where *δ* denotes a loss angle. The *tan δ* is determined using the relation below [23]. Loss tangent (*tan δ*) was further investigated for the MC polymer incorporated with various KI concentrations. Figure 5 shows the loss tangent (*tan δ*) spectra versus frequency at room temperature. The relation between loss tangent and frequency reveals some interesting behavior. Overall, the loss tangent increases with increasing the applied frequency due to the domination of the Ohmic components. It reaches a high value at a certain frequency, and followed by decreases at a high frequency, owing to the increasing nature of the reactive components [38]. Notably, MCKI4 displays the highest shift to the high frequency and the maximum value relative to the other samples due to the value of dielectric constant *ε* 0 for the MCKI4 as shown in Figure 4a [39]. The presence of the peaks at a characteristic frequency can be argued for indicating the presence of dipole relaxation in the electrolytes. It has been reported that improving the segmental motion of polymer chains decreases the relaxation time, allowing the transport process easier. This is expressed mathematically as *τ* = 1/2*π fmax*, where *τ* is the ionic charge carrier's relaxation time [40]. determined using the relation below [23]. Loss tangent (tanδ) was further investigated for the MC polymer incorporated with various KI concentrations. Figure 5 shows the loss tangent (tanδ) spectra versus frequency at room temperature. The relation between loss tangent and frequency reveals some interesting behavior. Overall, the loss tangent increases with increasing the applied frequency due to the domination of the Ohmic components. It reaches a high value at a certain frequency, and followed by decreases at a high frequency, owing to the increasing nature of the reactive components [38]. Notably, MCKI4 displays the highest shift to the high frequency and the maximum value relative to the other samples due to the value of dielectric constant *ε′* for the MCKI4 as shown in Figure 4a [39]. The presence of the peaks at a characteristic frequency can be argued for indicating the presence of dipole relaxation in the electrolytes. It has been reported that improving the segmental motion of polymer chains decreases the relaxation time, allowing the transport process easier. This is expressed mathematically as = 1/2௫, where *τ* is the ionic charge carrier's relaxation time [40].

*Materials* **2021**, *14*, x FOR PEER REVIEW 11 of 21

**Figure 5.** (tanδ) spectra versus frequency at room temperature for the MCKI electrolytes. **Figure 5.** (*tan δ*) spectra versus frequency at room temperature for the MCKI electrolytes.

The real, *Mr* and imaginary, *Mi* components of the electric modulus M\* against frequency for the MCKI based solid polymer electrolytes are shown in Figures 6 and 7, respectively. The *M′* and *M″* are determined using Equations (4) and (5). The real, *M<sup>r</sup>* and imaginary, *M<sup>i</sup>* components of the electric modulus *M*\* against frequency for the MCKI based solid polymer electrolytes are shown in Figures 6 and 7, respectively. The *M*0 and *M*00 are determined using Equations (4) and (5).

From the figures, *Mr* values are noted to decrease with decreasing frequencies until they reach zero, meaning that the polarization was eliminated. Therefore, the *Mr* values rise with increasing frequency and at the highest frequency, the maximum *Mr* was obtained. This could be attributed to the fact that the relaxation process occurs at various frequency values [41]. The observed dispersion is essentially as of conductivity relaxation covering several frequencies, indicating the presence of *τ* that has to occur with a loss peak in the figure of the imaginary part of the dielectric modulus versus frequency. As *Mi*has clearly a lower value at a low frequency, this may be attributed to the higher capacitance coupled with the polarization effect. No peak is present in Figure 6 along with its entire From the figures, *M<sup>r</sup>* values are noted to decrease with decreasing frequencies until they reach zero, meaning that the polarization was eliminated. Therefore, the *M<sup>r</sup>* values rise with increasing frequency and at the highest frequency, the maximum *M<sup>r</sup>* was obtained. This could be attributed to the fact that the relaxation process occurs at various frequency values [41]. The observed dispersion is essentially as of conductivity relaxation covering several frequencies, indicating the presence of *τ* that has to occur with a loss peak in the figure of the imaginary part of the dielectric modulus versus frequency. As *M<sup>i</sup>* has clearly a lower value at a low frequency, this may be attributed to the higher capacitance coupled with the polarization effect. No peak is present in Figure 6 along with its entire frequency range. It could be referring to the *M<sup>r</sup>* which is equivalent to the *ε* 0 in the *ε\** representation, which *M<sup>r</sup>* shows the material's potential for energy conversion [42].

frequency range. It could be referring to the *Mr* which is equivalent to the *ε′* in the *ε\** representation, which *Mr* shows the material's potential for energy conversion [42].

frequency range. It could be referring to the *Mr* which is equivalent to the *ε′* in the *ε\** representation, which *Mr* shows the material's potential for energy conversion [42].

*Materials* **2021**, *14*, x FOR PEER REVIEW 12 of 21

0.4

MCKI1

**Figure 6.** Electric modulus plot of *Mr* against log(f)for the MCKI samples. **Figure 6.** Electric modulus plot of *Mr* against log(f)for the MCKI samples. **Figure 6.** Electric modulus plot of *Mr* against log(f)for the MCKI samples.

**Figure 7.** Electric modulus plot of Mi against log(f) for the MCKI samples. **Figure 7.** Electric modulus plot of Mi against log(f) for the MCKI samples. **Figure 7.** Electric modulus plot of *M<sup>i</sup>* against log(f) for the MCKI samples.

#### *3.3. FTIR Study 3.3. FTIR Study 3.3. FTIR Study*

The technique of FTIR spectroscopy has been used to investigate the interactions between ions and atoms of the MCKI electrolytes. Also, such interactions can lead to the The technique of FTIR spectroscopy has been used to investigate the interactions between ions and atoms of the MCKI electrolytes. Also, such interactions can lead to the The technique of FTIR spectroscopy has been used to investigate the interactions between ions and atoms of the MCKI electrolytes. Also, such interactions can lead to the changes in the vibration modes of the polymer electrolyte. The FTIR spectra of the pure MC and MCKI based solid polymer electrolyte over the wavenumber range of 940–4000 cm−<sup>1</sup> are displayed in Figure 8a,b. The broad peak observed at around 1050 cm−<sup>1</sup> corresponds to the antisymmetric stretch of an asymmetric oxygen bridge in its cyclohexane ring of pure

MC. The water contamination from the KI salt causes a broad peak at 3400 cm−<sup>1</sup> of the O-H stretching band. The observed peak intensity changes as the weight percent of KI salt was increased from 0 to 40 % in the MC-KI electrolyte systems, as shown in Figure 8a,b [43,44]. A peak that appears in the wavenumber region of 2800–2950 cm−<sup>1</sup> is corresponding to the C-H stretching mode of methylcellulose. Through the inclusion of KI salt, the peak seems to shift slightly from 2850 cm−<sup>1</sup> to 2990 cm−<sup>1</sup> . This shift of the peak may be an indication of the complexation of K<sup>+</sup> cation and the MC host polymer. However, the slight change in the C-O ether bands indicates that the complexation did not considerably modify the molecular structure of the MC host polymer. Furthermore, the change in peak intensity with increasing KI concentrations supports that the presence of KI salt in the system has a significant impact on the conductivity of the MCKI electrolyte systems [45]. cent of KI salt was increased from 0 to 40 % in the MC-KI electrolyte systems, as shown in Figure 8a,b [43,44]. A peak that appears in the wavenumber region of 2800–2950 cm−1is corresponding to the C-H stretching mode of methylcellulose. Through the inclusion of KI salt, the peak seems to shift slightly from 2850 cm−1 to 2990 cm−1. This shift of the peak may be an indication of the complexation of K+ cation and the MC host polymer. However, the slight change in the C-O ether bands indicates that the complexation did not considerably modify the molecular structure of the MC host polymer. Furthermore, the change in peak intensity with increasing KI concentrations supports that the presence of KI salt in the system has a significant impact on the conductivity of the MCKI electrolyte systems [45].

changes in the vibration modes of the polymer electrolyte. The FTIR spectra of the pure MC and MCKI based solid polymer electrolyte over the wavenumber range of 940–4000 cm−1 are displayed in Figure 8a,b. The broad peak observed at around 1050 cm−1 corresponds to the antisymmetric stretch of an asymmetric oxygen bridge in its cyclohexane ring of pure MC. The water contamination from the KI salt causes a broad peak at 3400 cm−1of the O-H stretching band. The observed peak intensity changes as the weight per-

*Materials* **2021**, *14*, x FOR PEER REVIEW 13 of 21

**Figure 8.** FTIR spectra of the MCKI samples at a wavenumber of (**a**) 940–1200 cm<sup>−</sup><sup>1</sup> and (**b**) 2500–4000 cm<sup>−</sup><sup>1</sup> for (i) **Figure 8.** FTIR spectra of the MCKI samples at a wavenumber of (**a**) 940–1200 cm−<sup>1</sup> and (**b**) 2500–4000 cm−<sup>1</sup> for (i) MCKI1 (ii) MCKI2, (iii) MCKI3, and (iv) MCKI4 electrolyte samples.

#### *3.4. EDLC Study 3.4. EDLC Study* 3.4.1. Study of the TNM

MCKI1 (ii) MCKI2, (iii) MCKI3, and (iv) MCKI4 electrolyte samples.

3.4.1. Study of the TNM Both ions and electrons in polymer electrolytes are generally responsible for their conductivity. Through this technique, the dominant charge carrier in the polymer electrolyte can be evaluated [46]. Figure 9 shows the current versus time plot, obtained by dc Both ions and electrons in polymer electrolytes are generally responsible for their conductivity. Through this technique, the dominant charge carrier in the polymer electrolyte can be evaluated [46]. Figure 9 shows the current versus time plot, obtained by dc polarization at 0.2 V, for the MCK1<sup>4</sup> film. Equations (6) and (7) were used to determine the *tion* and *tel* of the MCKI4 film.

polarization at 0.2 V, for the MCK14 film. Equations (6) and (7) were used to determine the *tion* and *tel* of the MCKI4 film. According to Figure 9, the initial total current was found to be 22 µA [47]. Therefore, a large drop is observed over time until being constant in a completely depleted case due to the transport of ionic species from the bulk of the MCKI4 electrolyte to the electrodeelectrolyte interfaces. When the cell reaches the steady state, it is polarized, and the residual current is only carried by electrons due to the stainless-steel electrodes block both cations and anions while allowing only electrons to move through it. In this analysis, the measured *tel* value was 0.12 and the *tion* was found to be 0.88, which is close to an ideal According to Figure 9, the initial total current was found to be 22 µA [47]. Therefore, a large drop is observed over time until being constant in a completely depleted case due to the transport of ionic species from the bulk of the MCKI4 electrolyte to the electrodeelectrolyte interfaces. When the cell reaches the steady state, it is polarized, and the residual current is only carried by electrons due to the stainless-steel electrodes block both cations and anions while allowing only electrons to move through it. In this analysis, the measured *tel* value was 0.12 and the *tion* was found to be 0.88, which is close to an ideal value of 1 [28], indicating that ions in the MCKI4 film is the majority charge carrier [48]. The finding obtained in this work is comparable with the *tion* value of 0.86 as reported by Aziz et al. for the polymer electrolyte system of chitosan: dextran: NH4Br [49].

value of 1 [28], indicating that ions in the MCKI4 film is the majority charge carrier [48]. The finding obtained in this work is comparable with the *tion* value of 0.86 as reported by

Aziz et al. for the polymer electrolyte system of chitosan: dextran: NH4Br [49].

**Figure 9.** DC polarization curve of current versus time for the MCKI4 sample. **Figure 9.** DC polarization curve of current versus time for the MCKI4 sample.

#### 3.4.2. LSV Study 3.4.2. LSV Study

The potential stability of the polymer electrolyte systems needs to be established for energy device research. The absolute potential limit of the electrolytes can be computed in terms of linear sweep voltammetry LSV examination [50]. The LSV for the most conducting sample MCKI4 at 10 mV/s is shown in Figure 10, in which the potential was scanned from 0 to 2.5 V. When potential approaches to 1.8 V, the electrolyte reaches decomposition voltage as revealed by a significant increase in current values. Also, there is no evidence of a redox reaction occurring within the potential window until 1.8 V. Based on the previous study, the electrolyte with the potential window of 1.8 V is sufficient to be used for application in proton energy devices [51]. Other research findings relating to MC-based biopolymer electrolytes are comparable to this work. According to Kadir et al. [52], MC-based electrolytes displayed a decomposition voltage of 1.53 V when NH4Br and glycerol were used as the ionic source and plasticizer, respectively. The breakdown potential of 1.9 V was reported for the biopolymeric system of starch-chitosan-NH4I with the existence of glycerol [53], which is similar to this study. The potential stability of the polymer electrolyte systems needs to be established for energy device research. The absolute potential limit of the electrolytes can be computed in terms of linear sweep voltammetry LSV examination [50]. The LSV for the most conducting sample MCKI4 at 10 mV/s is shown in Figure 10, in which the potential was scanned from 0 to 2.5 V. When potential approaches to 1.8 V, the electrolyte reaches decomposition voltage as revealed by a significant increase in current values. Also, there is no evidence of a redox reaction occurring within the potential window until 1.8 V. Based on the previous study, the electrolyte with the potential window of 1.8 V is sufficient to be used for application in proton energy devices [51]. Other research findings relating to MC-based biopolymer electrolytes are comparable to this work. According to Kadir et al. [52], MC-based electrolytes displayed a decomposition voltage of 1.53 V when NH4Br and glycerol were used as the ionic source and plasticizer, respectively. The breakdown potential of 1.9 V was reported for the biopolymeric system of starch-chitosan-NH4I with the existence of glycerol [53], which is similar to this study.

#### 3.4.3. Cyclic Voltammetry (CV) Study

CV as an insightful technique can be employed to examine the EDLCs in terms of both qualitative and quantitative features [54]. It is used to further evaluate the efficiency of the MCKI4 electrolyte in the construction of the EDLC. The CV responses of the MCKI4 electrolyte at various scan rates of 10, 20, 50, and 100 mV/s are shown in Figure 11 in the potential range of 0 to 0.9 V.

**Figure 10.** Current versus potential for the highest conducting (MCKI4) sample. **Figure 10.** Current versus potential for the highest conducting (MCKI4) sample.

**Figure 11.** Cyclic voltammograms for the assemble EDLC in the potential range of 0 to 0.9 V. **Figure 11.** Cyclic voltammograms for the assemble EDLC in the potential range of 0 to 0.9 V.

The specific capacitance (*Cs*)are determined using Equation (8) by measuring the area of the CV profile, mass of the activated carbon electrode, scan rate, and the initial and final values of applied voltage. The measured specific capacitance values, *Cs* using CV curves for the assembled EDLC at different scan rates are shown in Table 3 and Figure 12. The calculated *Cs* value of 113.39 F/g at the sweep rate of 10 mV/s decreased to 11.84 F/g at 100 mV/s. The low *Cs* value at high scan rates is attributed to the high energy loss caused by the decrease in the density of stored charges, which results in a lower *Cs* value [60]. Table 4 displays the measured *Cs* value of the EDLC for several systems based on solid biopoly-The CV response has a rectangular form, indicating that the current is independent of the potential. However, the shape of the cyclic voltammogram (CV) deviates from the rectangular shape when the scan rate increases [55]. The CV in Figure 11 showed that the EDLC exhibits a capacitive behavior, indicating that the system of the energy storage is a non-Faradaic mechanism. In this process, the charge stored in the EDLC system comes from ion accumulation at the electrode/electrolyte interfaces. As a consequence, ion accumulation and adsorption occur in the place of deintercalation and intercalation via a non-Faradaic mechanism. In addition, ions from the bulk of the electrolyte form a charge double-layer, which then saves potential energy [56,57]. Notably, the CV displays a

**Scan Rates (mV/s) Specific Capacitance,** 

10 113.39 20 69.16 50 27.48 100 11.84

*Cs* **F/g** 

**Table 3.** Specific capacitance (*Cs*) of the EDLCs using CV curves.

leaf-like shape with no redox peaks. The CV profile revealed a little divergence from its rectangular form at higher scan rates, which may be due to the porosity of the electrodes as well as internal resistance. The porosity of the carbon electrodes induces a relatively high internal resistance, which causes the CV to appear leaf-like in shape [58]. Since the CV possesses no redox peaks, it is reasonable to infer that a quick Faradaic reversible reaction has not occurred [59].

The specific capacitance (*Cs*)are determined using Equation (8) by measuring the area of the CV profile, mass of the activated carbon electrode, scan rate, and the initial and final values of applied voltage. The measured specific capacitance values, *C<sup>s</sup>* using CV curves for the assembled EDLC at different scan rates are shown in Table 3 and Figure 12. The calculated *C<sup>s</sup>* value of 113.39 F/g at the sweep rate of 10 mV/s decreased to 11.84 F/g at 100 mV/s. The low *C<sup>s</sup>* value at high scan rates is attributed to the high energy loss caused by the decrease in the density of stored charges, which results in a lower *C<sup>s</sup>* value [60]. Table 4 displays the measured *C<sup>s</sup>* value of the EDLC for several systems based on solid biopolymer electrolytes mentioned in the literature. Interestingly, the *C<sup>s</sup>* value obtained in this work is high and comparable to some of these results.

**Table 3.** Specific capacitance (*Cs*) of the EDLCs using CV curves.


**Figure 12.** The calculated specific capacitance, *Cs* for the assembled EDLC at a different scan rate. **Figure 12.** The calculated specific capacitance, *Cs* for the assembled EDLC at a different scan rate.

**Table 4.** Specific capacitance (*Cs*) of the EDLCs using different polymer electrolytes at room temper-

Mg(CF3SO3)2:glycerol 32.69 10 [3]

MC-chitosan-NH4I-glycerol 9.97 10 [62] Cellulose acetate-LiClO4 90 10 [63] Chitosan-NH4Br-glycerol 7.5 10 [64] MC-Starch-LiClO4-glycerol 45.8 10 [65]

Starch-LiClO4 8.7 10 [7] MC-NH4NO3-PEG 38 1 [45] MC-chitosan-NH4SCN 66.3 10 [33]

MC-KI 113.39 10 This work

In conclusion, a biopolymer-based electrolyte using methylcellulose (MC) incorporated with various content of potassium iodide (KI) salt is crucial for EDLC device applications. The EIS outcome shows that the resistance of the transfer of charge at the bulk of the electrolyte reduced from 3.3 × 105 Ω to 8 × 102 Ω with KI concentration increased from 10 wt % to 40 wt % due to an increase in the charge carrier density. The highest conductivity of 1.93 × 10−5 S/cm was obtained for the electrolyte doped with 40 wt % of KI. The dielectric analysis further verified the conductivity trends. The results from the FTIR spectra indicated that the complexation between (K+) cation and (MC) host polymer has occurred through intensity variations of bands. TNM measurements stated that the ions were the dominant charge carrier, as the (*tion*) was identified to be 0.88. LSV analysis showed that the most conducting sample has an electrochemical stability window up to 1.8 V, verifying the suitability of the electrolyte for EDLC application. The CV response

*Cs* **F/g Scan Rates (mV/s) Reference** 

1.8 Not stated [61]

Chitosan-PVA-

Carboxymethyl cellulose - NH4NO3

**4. Conclusions** 

ature


**Table 4.** Specific capacitance (*Cs*) of the EDLCs using different polymer electrolytes at room temperature.

#### **4. Conclusions**

In conclusion, a biopolymer-based electrolyte using methylcellulose (MC) incorporated with various content of potassium iodide (KI) salt is crucial for EDLC device applications. The EIS outcome shows that the resistance of the transfer of charge at the bulk of the electrolyte reduced from 3.3 <sup>×</sup> <sup>10</sup><sup>5</sup> <sup>Ω</sup> to 8 <sup>×</sup> <sup>10</sup><sup>2</sup> <sup>Ω</sup> with KI concentration increased from 10 wt % to 40 wt % due to an increase in the charge carrier density. The highest conductivity of 1.93 <sup>×</sup> <sup>10</sup>−<sup>5</sup> S/cm was obtained for the electrolyte doped with 40 wt % of KI. The dielectric analysis further verified the conductivity trends. The results from the FTIR spectra indicated that the complexation between (K<sup>+</sup> ) cation and (MC) host polymer has occurred through intensity variations of bands. TNM measurements stated that the ions were the dominant charge carrier, as the (*tion*) was identified to be 0.88. LSV analysis showed that the most conducting sample has an electrochemical stability window up to 1.8 V, verifying the suitability of the electrolyte for EDLC application. The CV response displayed its capacitance behavior, where no visible redox peak has appeared. A relatively high value of the specific capacitance *C<sup>s</sup>* (113.39 F/g) was obtained at the scan rate of 10 mV/s.

**Author Contributions:** Conceptualization, formal analysis, writing—original draft, methodology, supervision, S.B.A., M.M.N. and J.M.H.; Formal analysis, A.S.F.M.A. and M.A.B.; Investigation, M.A.B., M.M.N., A.M.A.; Methodology, S.B.A. and M.A.B.; Project administration, S.B.A. and M.F.Z.K.; Validation, E.M.A.D., M.A.B., A.M.A. and M.F.Z.K.; Writing—original draft, M.M.N., J.M.H., S.B.A. and A.S.F.M.A.; Writing—review & editing, S.B.A., M.A.B., M.F.Z.K. and E.M.A.D. All authors have read and agreed to the published version of the manuscript.

**Funding:** The authors would like to acknowledge the support of Prince Sultan University for paying the Article Processing Charges (APC) of this publication and for their financial support.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** The data presented in this study are available on request from the corresponding author.

**Acknowledgments:** We would like to acknowledge all support for this work by the University of Sulaimani, Charmo University, Prince Sultan University and Komar University of Science and Technology.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


*Article*
