Continuous fast Fourier transform admittance voltammetry as a new approach for studying the change in morphology of polyaniline for supercapacitors application

Abstract

In this study, continuous fast Fourier transform admittance voltammetry (CFFTAV) was used to study and characterize the surface morphology of polyaniline (PANI) on glassy carbon (GC) electrodes. Four polymer films with various thicknesses (0.5 μm to 11 μm) were synthesized by an electrochemical method. A new modified square wave voltammetry (SWV) method based on application of a discrete Fast Fourier Formation (FFT) method, background subtraction and two-dimensional integration of the electrode response over a selected potential range and time window was used. Moreover, the electrode response could be calculated by measuring the changes in SW voltammogram (or admittance). Results showed that by using the CFFTAV method, changes in the porosity of PANI and the behavior of PANI formation in H₂SO₄ solution were investigated more quantitatively than when using scanning electron microscopy (SEM), cyclic voltammetry (CV), charge–discharge (CD) and impedance spectroscopy (EIS) methods. By monitoring the electrode response, the kinetics for reaching steady state condition was studied. It was found that with the increasing thickness of polymer film from 0.5 μm to 11 μm, the accessible porosity decreased by up to three times. Furthermore, dimensions of nanochannels in the polymer film decreased with increasing the thickness. Moreover, maximum potential for ion insertion increased from 324 mV to 365 mV. Capability of electrodes for use as supercapacitor materials was tested by CV, CD and EIS, and the calculated capacity of electrodes was equal to 620 F g⁻¹ and 247 F g⁻¹ for thinnest and thickest polymer films respectively.

---

1. Introduction

Supercapacitors (SCs) are modern energy storage systems, which nowadays attract huge scientific interest due to their ability to deliver very high impulses of energy in a very short time, good cyclic efficiency, long cyclability and lack of environmental toxicity. SCs find applications in electric and hybrid cars and buses, wind turbines, and UPC systems.^1–5 Conducting polymers (CPs) also called synthetic metals such as PANI, polypyrrole or polythiophene, are promising electrode materials for SCs due to their good conductivity, good thermal and chemical stability, high doping effect and fast electrochemical switching.^6–8

PANI is one of the popular electrode materials, which has been used for SCs because of its diverse structure, thermal and radiation stability, relatively low cost and easy synthesis.^9,10 It can exhibit very high capacitance values, which can tend to reach even 1000 F g⁻¹ per one electrode, when electrochemical research is conducted in a three-electrode cell using a thin film of PANI.¹¹

The main parameter of CPs that indicates a possibility of their application as electrode material in SCs is porosity or the number of active sites, due to its critical influence on the capacitance of the CP electrode. Porosity or active sites of various materials can be determined using the following techniques: physical adsorption of gases; mercury porosimetry; small angle scattering (SAS), including small angle neutrons scattering (SANS) or small angle X-ray scattering (SAXS); transmission electron microscopy (TEM); scanning electron microscopy (SEM); scanning tunnel microscopy (STM); and immersion calorimetry.¹² However, in earlier publications for specifying the porosity of CPs, there is lack of studies that report a method for identifying the porosity type of the polymer film. The major techniques for the study of porosity cannot be used for thin polymer film such as adsorption of gases or mercury porosimetry. In addition, some other methods, such as TEM, SEM and other microscopy techniques, require film drying, which changes the structure of the film.¹³ As a result, investigation of CPs' porosity is a major challenge; therefore, we suggest a CFFTAV technique, which is a new modified SWV method, for the study of the surface of PANI film. Compared to some ordinary method, including CV and EIS, this electrochemical technique has various applications, such as a sensing and bio sensing material, which span various fields due to the unique feature of SWV such as selectivity and sensitivity. Electrochemical sensor and biosensors,^14–17 environmental pollution,^18,19 food science^20,21 and enzyme kinetics^22,23 are some examples that illustrate the applications of SWV. In this study, continuous fast Fourier transform admittance voltammetry (CFFTAV) was used to study and characterize the formed polyaniline film on glassy carbon (GC) electrodes. First, PANI films were synthesized by cyclic voltammetry technique on the surface of GC electrodes; furthermore, they were studied in H₂SO₄ solution using the CFFTAV technique. After film formation, porosity and the types PANI were determined. To show the capability of these electrodes for use in supercapacitor devices, the properties of electrodes were tested by CV, CD and EIS techniques.

2. Experimental

2.1. Materials and morphological investigations and electrochemical evaluation

Aniline was vacuum-distilled at 120 °C and stored in a refrigerator before use. Analytical-grade reagents were used as received without any pretreatment. Double-distilled water was used for the preparation of solutions. Morphological investigations of the polymer films were carried out using SEM (Philips XL 30).

Electrochemical experiments (CV, CD and EIS) were carried out by an Autolab General purpose System PGSTAT 30 (Eco Chemie, Netherlands). A conventional three-electrode cell containing a glassy carbon electrode with an area of 0.03 cm² as the working electrode, a platinum wire as the counter electrode and an Ag/AgCl reference electrode (Argental, 3 M KCl) were used. The EIS experiments were conducted in the frequency range between 100 kHz and 10 MHz with a perturbation amplitude of 5 mV.

2.2. Synthesis of PANI

Electrochemical polymerization of PANI was done using the CV technique on a GC electrode. Four different thicknesses of PANI were formed by applying 3, 5, 7 and 12 cycles, which were named PANI1, PANI2, PANI3 and PANI4, respectively, in potential range from −0.2 to 1.2 V, at the scan rate of 50 mV s⁻¹ in a solution of 0.03 M aniline in 1.0 M H₂SO₄. The mass of the PANI films was approximated assuming a current efficiency for the electropolymerization process of 100% using Faraday's law of electrolysis.²⁴

2.3. Continuous FFT admittance voltammetric measurements

For the electrochemical CFFTAV measurements, a homemade potentiostat that was connected to a PC was used. The potentiostat was connected to an analog-to-digital (A/D) data acquisition board (PCL-818H, Advantech Co.). The data acquisition board was also used for generating the analog waveform and acquiring current. The potential waveform was repeatedly applied to the working electrode and then the data were acquired and stored by the software. In the measurements, the data acquisition requirements electrochemical software was developed using Delphi 6.0. Moreover, in this electrochemical setup, the data could be processed and plotted in real time or the stored data could be loaded and reanalyzed with the voltammograms. To improve the detector sensitivity, the SWV technique was modified in the potential excitation waveform and current sampling and data processing. Fig. 1 shows the potential waveform used for the measurements. In the modified technique, the currents were sampled four times per each SW polarization cycle, and the potential waveform contained one additional potential step, E_s, for conditioning the electrode. As is shown in Fig. 1, the measurement part of the waveform contains multiple SW pulse cycles with an amplitude of E_sw and a frequency of f₀, superimposed on a staircase potential function, which was changed by a small potential step of ΔE. The potential pulse values of SW (E_SW) and ΔE were within a narrow range of few mV (10 to 50 mV). In the computer program, the number of SW cycles, N_c, in each staircase potential step was calculated based on the SW frequency,

N_c = f₀/1400 Hz for f₀ > 1400 Hz (1)

and

N_c = 1 for f₀ ≤ 1400 Hz (2)

Fig. 1 Applied potential waveform during FFT measurement.

The values of N_c, f₀, E_sw, E_initial and E_final were the variable parameters in the measurements. It should be noted that in this technique, all the electrochemical processes involve insertion of negative ions, and hence both charging and faradic currents may potentially carry useful analytical information. To obtain such information, it was important to sample current at a frequency at least two times higher than the current transducer bandwidth. To fulfill this requirement, the sampling frequency was set always between 40 and 70 kHz (depending on the applied SW frequency). In addition, a second order low pass filter with a 20 kHz cutoff frequency was placed between the current output of the potentiostat and the data acquisition board. In this technique, a modified SW voltammetry method was based on application of a discrete FFT method, background subtraction and two-dimensional integration of the electrode response over a selected potential range and time window. Moreover, the electrode response could be calculated by measuring the changes in SW voltammogram (or admittance).^25–28 First, during the FFT measurements, a positive ramp voltage from 0.2 to 0.6 V was applied to the electrode by the frequency of 2500 Hz and amplitude of 0.05 V.

3. Results and discussion

3.1. Electrochemical measurements

Fig. 2 demonstrates CVs for electropolymerization of the PANI4 electrode. As mentioned, the procedure for synthesis of all electrodes is the same and the number of cycles for electropolymerization is the only difference between the electrodes. As can be seen, the CVs for electropolymerization show three redox peaks. The first redox peaks at 0.3 and −0.1 V are related to the formation of free radicals in the polymer chains, and the second peaks at 0.5 and 0.4 V are attributed to oxidation and reduction of an intermediate produced by degradation of products, and the third redox pair was related to the oxidation of PANI.²⁹

Fig. 2 Cyclic voltammograms for electropolymerization of polyaniline on the surface of GC electrode.

The thickness of the polymer (d) was estimated from the charge, Q_i, necessary to switch from leucoemeraldine (LE) to emeraldine (EM²⁺) form of PANI according to the equation:

where Q_i is the charge under the first cyclic voltammogram peak; M_W is the molecular weight of aniline; Z = 0.5 (number of electrons/aniline unit); A is the area of the electrode (0.0314 cm²); ρ is the specific density of aniline (1.02 g cm⁻³); and F is Faraday's constant. The method calculates the total quantity of PANI without taking into account the porosity factor and counter-ion volume.¹¹ Table 1 summaries the PANI electrodes used in this study.

Table 1 Charge (Q_i) and thickness (d) of PANI films

Electrode	Qi (mC)	d (μm)	Number of polymerization cycles
PANI1	0.86	0.5	3
PANI2	3.8	2.5	5
PANI3	7.5	4.5	7
PANI4	18.3	11	12

---

Fig. 3 presents the SEM photomicrographs of four PANI electrodes. As can be seen, by increasing the thickness of the polymer film, the surface morphology of the electrodes changed and the free spaces between polymer filaments decreased. Entangling the polymer fibers together, increasing the surface of electrodes for electropolymerization of PANI, caused compaction of the polymer fibers and resulted in decreasing the useful space for insertion/deinsertion of ions on PANI active sites. Although SEM electromicrographs are beneficial data for studying the surface morphology of conductive polymers, these results are qualitative, and for quantitative results, we used a CFFTAV method.

Fig. 3 SEM photomicrographs of PANI electrodes.

Fig. 4 shows (a) the CVs of PANI electrodes at a sweep rate of 25 mV s⁻¹ and (b) the CD plots of four polymer films at current density of 1.8 A g⁻¹ in 1.0 M H₂SO₄ media. As shown in the figure, in all CV curves, a pair of redox peaks related to the transition from emeraldine to pernigraniline form of PANI appears. With the increasing thickness of the polymer film that formed on the surface of GC electrode, the charging current of the CV curves (or charge Q) increased, which causes enhanced capacity of electrodes; however, due to the mass of the polymer, the specific capacitance of the electrode diminished significantly. The capacity of electrodes for PANI1, PANI2, PANI3, and PANI4 was calculated as 620, 332, 288, 247 F g⁻¹, respectively. According to the formula,

where V is the potential window and m is the mass of the PANI.³⁰ Galvanostatic charge–discharge (GCD) measurement is a comprehensive technique to confirm the CV data of PANI electrodes. Fig. 4b presents the CD plots of PANI electrodes at current density of 1.8 A g⁻¹ in 1 M H₂SO₄ solution. Increasing the time needed to reach to cutoff potential by increasing the thickness of the polymer film shows that the active sites of the polymer films increased, but these active sites were related to the whole film body. It is well known that in low sweep rates, the ions in the solution have enough time for the doping process to the polymer matrix.

Fig. 4 (a) CVs of PANI electrodes at sweep rate of 25 mV s⁻¹, (b) CD plots of four polymer films at current density of 1.8 A g⁻¹ in 1.0 M H₂SO₄ solution.

Impedance spectroscopy is a suitable method for studying the electrochemical features of CPs.^31,32

Fig. 5 shows the Nyquist plots of the electrodes in 1.0 M H₂SO₄ solution. As shown in the figure, a semicircular loop in the high frequency followed by a nearly vertical straight line in the low frequency is observed. The relationship between the imaginary impedance and capacitance of the model capacitor could be utilized to calculate capacitance (at 10 mHz frequency),

where f is the frequency and Z′′ is the imaginary part of the Nyquist diagram. The specific capacitances of the four electrodes were calculated as 570, 290, 251, and 190 F g⁻¹ for PANI1, PANI2, PANI3 and PANI4, respectively. The value of calculated capacitance was slightly smaller than that obtained from cyclic voltammograms, but it shows that the specific capacitance calculated from cyclic voltammograms is reliable. Another feature of CPs for use as supercapacitors is their stability during continuous cycles. The stability of the four electrodes was tested using continuous CVs and the results shows that by increasing the thickness of PANI electrodes from 0.5 μm to 11 μm, the stability of PANI electrodes was enhanced from 30% to 50% after 800 continuous cycles. The obtained CV, CD and EIS results confirmed that thicker films have more sites or porosity but do not provide information about the surface or beneath the surface of the polymer electrode. To study the reactions at the surface of polymer films, we must use higher scan rates. With increasing sweep rate, the sensitivity of the cyclic voltammetry method decreased and due to this phenomenon, a CFFTAV method was used.

Fig. 5 Nyquist plots of the four electrodes in 1.0 M H₂SO₄ solution.

Fig. 6 shows a 3D plot of FFTA voltammograms for the PANI1 electrode as a function of time in 1.0 M H₂SO₄ at a SW frequency of 2500 Hz and an amplitude 50 mV. At the potential of 0.2 V, the polymer oxidizes to the emeraldine form, which results in creation of positive charges on the electrode surface that help the insertion of negative ions into the polymer matrix; the overall process is neutralization of the positive charges on the surface of the polymer. Therefore, by continuous application of potential waveform (in that time window), a large number of insertions and deinsertions of the ions in the polymer matrix occurred. Consequently, this ion exchange between solution and polymer film produced a current or changed the admittance of the electrode.^33–35

Fig. 6 3D FFTAV graph of PANI1 electrode in 1.0 M H₂SO₄ at the frequency of 2500 Hz and amplitude 50 mV.

It can be seen that during the insertion of ions to the polymer matrix, initially the admittance grows due to the increasing amount of movement and accumulation of the ions, but when the whole polymer matrix is full of ions, admittance starts to decrease rapidly, to very low level. Furthermore, the deinsertion process with a strong impulse of negative voltage occurs (E_s) and then the next cycle is conducted. Using SW caused ions to rearrange during the doping process and all the active sites were fully used.^2,26,36 However, at high SW frequency, ions absorbed only on the surface of the polymer film and the inner section of the polymer has no ion absorption due to insufficient time for ions to reach that part. Therefore, at a high scan rate, only the surface of PANI film and beneath the surface can be examined, which are the boundary active sites or porosity of the electrode surface.

To observe the details of changes in the admittance of electrodes at each potential, differential admittance plots were used for various PANI samples, which are shown in Fig. 7. The background admittance was subtracted using this equation,

ΔA_E = A_E − A_r (6)

where A_E is the value of admittance in the voltammogram and A_r is the value of admittance in the reference voltammogram, which was recorded at the beginning of the experiment. It can be seen that by increasing the number of voltammograms, the admittance of the electrode was enhanced. This can be attributed to the fact that in the early cycles, solvent molecules physically adsorbed and occupied some of the porosities of the PANI films and by conducting the cycles, these molecules were removed from the polymer matrix and the porosities can be useable. As shown in the 3D diagrams, attainment of a steady state for the polymer films occurs with different time scales. The value of admittance reaches the optimum value after 7 seconds for PANI1 and after 20 seconds for PANI4.

Fig. 7 3D characterizations of differential admittance voltammograms to potential and time for PANI electrodes in 1.0 M H₂SO₄ at the frequency of 2500 Hz and amplitude 50 mV.

It can also be seen that when the polymer film is thicker, obtaining to the optimal value of admittance takes longer and admittance of the polymers gets smaller.³⁷ In addition, in the admittance-potential-time curves, for the PANI3 electrode, some little noises appeared, which have higher values for the PANI4 electrode. The appearance of noise could be attributed to the fast replacement of the ions between active sites because of the proximity of the polymer chains that entangled together.

By comparing PANI1 and PANI4, it can be seen that by increasing the thickness of the polymer film, the value of admittance of the electrodes decreased from 19877 μA V⁻¹ for the PANI1 electrode to 11 018 μA V⁻¹ for the PANI4 electrode. These phenomena occurred at high frequency and therefore only accessible pores were used. Therefore, by increasing the thickness of PANI, the number of accessible pores at the surface of PANI electrodes decreased. Thicker polymers also were checked and their behavior was the same and the magnitude of admittance decreased for thicker films. Furthermore, the quantity of noise for thicker films increased. 2D plots of differential admittance–potential are presented in Fig. 8. From 2D differential admittance–potential curves, the maximum potential for the insertion of ions into the polymer matrix can be obtained. The horizontal plateau line is related to the first voltammogram that was saved as reference, and subsequent voltammograms were subtracted from the reference voltammogram. All these plots have three parts; the first part has a positive sign in potential range between 0.2 V to 0.3 V corresponding to oxidation of PANI and transformation to emeraldine form.³⁷ As polymer film is oxidized and electrons are removed from the electrode, the sign of admittance is positive.

Fig. 8 2D characterizations of differential admittance to potential for PANI electrodes in 1.0 M H₂SO₄ at the frequency of 2500 Hz.

The second part (0.3–0.4 V) relates to the insertion of ions to accessible pores of the polymer surface. As shown in Fig. 8, the potential of ion diffusion for the PANI4 electrode is more than the PANI1 film near 41 mV. Thus, it could be suggested that for the PANI4 electrode, the amount of ion insertion in the polymer matrix is lower than in thinner films, therefore it is more difficult for ions to reach deeper pores of the polymer matrix. This could be proof that PANI4 electrodes possess higher amounts of non-accessible porosities than other electrodes, in its surface or beneath the surface. Decreasing the size of the micro-channel caused a higher peak potential, which is related to diffusion of the ions into channels for the doping process.^38,39 This occurs because the ions are very close together and the repulsive forces between them cause use of more potential for the doping process. The third part of the plots appears after the peak in 400 to 600 mV. This could be attributed to surface adsorption of the ions on the polymer film. The positive sign of admittance was related to surface adsorption of the ions on the polymer film surface. In this section, after insertion of the ions into the film polymer on the surface, an electrical double layer formed. Table 2 shows the obtained results for all electrodes.

Table 2 Results obtained for all of the electrodes

Electrode	Maximum potential for insertion (mV)	Time of optimization (s)	Magnitude of admittance, μA V^−1
PANI1	324	7	19 877
PANI2	334	11	12 940
PANI3	353	16	11 167
PANI4	365	20	11 018

---

3.2. Effect of dilution

The calculation for the electrode response is based on integration of admittance in a selected potential range after background subtraction. The result of such calculation is the charge changes, ΔQ, to obtain the signal of the polymer film oxidation and reduction. This can be obtained using the following equation:where A(S, E) is the measured admittance of the PANI response at potential E, during the s-th sweep and A(S_r, E) is the reference admittance data of the PANI at potential E. τ is the time period between subsequent sweeps, Δt is the time difference between two subsequent admittance points at the voltammogram, E_i and E_f are initial and final of the potential integration, respectively. The reference data was obtained by saving the first PANI CFFTA voltammogram in 1.0 M H₂SO₄ at the frequency of 2500 Hz and amplitude 50 mV. This was used for background subtraction calculations.

Fig. 9 (main figure) shows the admittance–time plots of PANI samples. Initially, the magnitude of polymer admittance plots as function of time in 1 M H₂SO₄; then, after 100 seconds, water was added to the electrochemical cell to obtain half the concentration of ions. As shown in the figure, adding water to the reactor during the admittance measurement of PANI1 caused destabilization of the system and a large decrease of admittance, because of the decreasing amount of the ions, which disables occupation of some pores. Nevertheless, when adding water to the PANI4 electrode system, a small effect is seen; therefore, it could be concluded that in the PANI4 electrode, the amount of accessible porosity or accessible active sites is low, and after adding water the proportion of the number of ions to the accessible porosities changes very little compared to the PANI1 electrode. Thus, in case of the PANI4 electrode, most of the porosity is not accessible for the ions and consequently the admittance is independent of the electrolyte concentration. Fig. 9 (inset figure) shows the plot of charge differences after adding water for all electrodes. As illustrated, by increasing the thickness of polymer film from 0.5 μm to 11 μm, the recorded charge decreased more than three times. It resulted that the magnitude of active sites for the PANI4 electrode is 3 times less than for the PANI1 electrode, and the micro channels of thin film have more active sites for absorption of the ions compared with thicker film.

Fig. 9 Response of the PANI electrodes to injection of water in 1.0 M H₂SO₄ at the frequency of 2500 Hz and amplitude 50, (inset figure) plot of charge differences after adding water for all electrodes.

4. Conclusions

The CFFTAV technique is a promising tool for studying changes in the porosity of PANI and the behavior of PANI formation in H₂SO₄ solution. Research revealed that using differential admittance analysis, it is possible to observe how the ions interact with the polymer film during measurements and compare the quantity of active sites that exist in the polymer film. Results obtained by this method showed that by increasing the thickness of PANI film, the size of micro channels and the number of active sites in PANI film decreases. By this method, some parameter of polymer electrodes, such as maximum potential of ion insertion and times needed for reaching steady state ion exchange for the PANI electrodes, was studied. It can provide an estimation of accessible porosity based on the film thickness of PANI, for example, non-accessible sites for the PANI4 electrode are more than three times that of thin film polymer such as in the PANI1 electrode.

Acknowledgements

The authors thank the research council of University of Tehran and Iranian Nano Council for financial support of this work.

References

1. K. Jurewicz, S. Delpeux, V. Bertagna, F. Beguin and E. Frackowiak, Chem. Phys. Lett., 2001, 347, 36–40.
2. R. Kötz and M. Carlen, Electrochim. Acta, 2000, 45, 2483–2498.
3. F.-J. Liu, J. Power Sources, 2008, 182, 383–388.
4. A. Ehsani, A. Vaziri-Rad, F. Babaei and H. M. Shiri, Electrochim. Acta, 2015, 159, 140–148.
5. A. Ehsani, Prog. Org. Coat., 2015, 78, 133–139.
6. Z. Niu, Z. Yang, Z. Hu, Y. Lu and C. C. Han, Adv. Funct. Mater., 2003, 13, 949–954.
7. R. Partch, S. Gangolli, D. Owen, C. Ljungqvist and E. Matijevic, Conducting polymer composites-polypyrrole metal-oxide latexes, in ACS Symposium Series, American Chemical Society, Washington DC, 1992.
8. M. E. Roberts, D. R. Wheeler, B. B. McKenzie and B. C. Bunker, J. Mater. Chem., 2009, 19, 6977–6979.
9. J. Shabani Shayeh, A. Ehsani, M. Ganjali, P. Norouzi and B. Jaleh, Appl. Surf. Sci., 2015, 353, 594–599.
10. J. Shabani Shayeh, P. Norouzi and M. R. Ganjali, RSC Adv., 2015, 5, 20446–20452.
11. L. Duić and S. Grigić, Electrochim. Acta, 2001, 46, 2795–2803.
12. E. Frackowiak, V. Khomenko, K. Jurewicz, K. Lota and F. Beguin, J. Power Sources, 2006, 153, 413–418.
13. M. Ates, Prog. Org. Coat., 2011, 71, 1–10.
14. R. N. Goyal, S. Chatterjee, A. R. S. Rana and H. Chasta, Sens. Actuators, B, 2011, 156, 198–203.
15. K. Omidfar, A. Dehdast, H. Zarei, B. K. Sourkohi and B. Larijani, Biosens. Bioelectron., 2011, 26, 4177–4183.
16. E. V. Suprun, A. A. Saveliev, G. A. Evtugyn, A. V. Lisitsa, T. V. Bulko, V. V. Shumyantseva and A. I. Archakov, Biosens. Bioelectron., 2012, 33, 158–164.
17. S. Viswanathan, C. Rani, A. V. Anand and J.-a. A. Ho, Biosens. Bioelectron., 2009, 24, 1984–1989.
18. R. O. Kadara, N. Jenkinson and C. E. Banks, Electroanalysis, 2009, 21, 2410–2414.
19. A. Chen and B. Shah, Anal. Methods, 2013, 5, 2158–2173.
20. S. M. Ghoreishi, M. Behpour and M. Golestaneh, Anal. Methods, 2011, 3, 2842–2847.
21. R. A. Medeiros, B. C. Lourencao, R. C. Rocha-Filho and O. Fatibello-Filho, Anal. Chem., 2010, 82, 8658–8663.
22. V. Shumyantseva, T. Bulko, A. Y. Misharin and A. Archakov, Biochemistry (Moscow) Supplement Series B: Biomedical Chemistry, 2011, vol. 5, pp. 55–59.
23. S. Martić, M. Labib and H.-B. Kraatz, Electrochim. Acta, 2011, 56, 10676–10682.
24. N. Plesu, A. Kellenberger, M. Mihali and N. Vaszilcsin, J. Non-Cryst. Solids, 2010, 356, 1081–1088.
25. B. Ebrahimi, S. A. Shojaosadati, P. Daneshgar, P. Norouzi and S. M. Mousavi, Anal. Chim. Acta, 2011, 687, 168–176.
26. P. Norouzi, V. K. Gupta, F. Faridbod, M. Pirali-Hamedani, B. Larijani and M. R. Ganjali, Anal. Chem., 2011, 83, 1564–1570.
27. P. Norouzi, B. Larijani, M. Ganjali and F. Faridbod, Int. J. Electrochem. Sci., 2012, 7, 10414–10426.
28. P. Norouzi, B. Larijani, M. Ganjali and F. Faridbod, Int. J. Electrochem. Sci., 2014, 9, 3130–3143.
29. D. E. Stilwell and S. M. Park, J. Electrochem. Soc., 1988, 135, 2254–2262.
30. J. F. Mike and J. L. Lutkenhaus, J. Polym. Sci., Part B: Polym. Phys., 2013, 51, 468–480.
31. A. Ehsani, M. G. Mahjani, R. Moshrefi, H. Mostaanzadeh and J. S. Shayeh, RSC Adv., 2014, 4, 20031–20037.
32. J. Shabani-Shayeh, A. Ehsani and M. Jafarian, J. Korean Electrochem. Soc., 2014, 17, 179–186.
33. A. Ehsani, M. G. Mahjani, F. Babaei and H. Mostaanzadeh, RSC Adv., 2015, 5, 30394–30404.
34. F. Fusalba, P. Gouérec, D. Villers and D. Bélanger, J. Electrochem. Soc., 2001, 148, A1–A6.
35. Z. H. Zhou, N. C. Cai, Y. Zeng and Y. H. Zhou, Chin. J. Chem., 2006, 24, 13–16.
36. B. Zheng, Y. Zhao, W. Xue and H. Liu, Surf. Coat. Technol., 2013, 216, 100–105.
37. V. K. Gupta, P. Norouzi, H. Ganjali, F. Faridbod and M. Ganjali, Electrochim. Acta, 2013, 100, 29–34.
38. G. A. Snook, P. Kao and A. S. Best, J. Power Sources, 2011, 196, 1–12.
39. A. Karatrantos, R. J. Composto, K. I. Winey and N. Clarke, Macromolecules, 2011, 44, 9830–9838.

---