Optimization of porous polymer electrolyte for quasi-solid-state electrical double layer supercapacitor

Accepted Manuscript Title: Optimization of porous polymer electrolyte for quasi-solid-state electrical double layer supercapacitor Authors: Nitish Yadav, Kuldeep Mishra, S.A. Hashmi PII: DOI: Reference:

S0013-4686(17)30557-1 http://dx.doi.org/doi:10.1016/j.electacta.2017.03.101 EA 29135

To appear in:

Electrochimica Acta

Received date: Revised date: Accepted date:

7-1-2017 4-3-2017 12-3-2017

Please cite this article as: Nitish Yadav, Kuldeep Mishra, S.A.Hashmi, Optimization of porous polymer electrolyte for quasi-solid-state electrical double layer supercapacitor, Electrochimica Actahttp://dx.doi.org/10.1016/j.electacta.2017.03.101 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Optimization of porous polymer electrolyte for quasi-solid-state electrical double layer supercapacitor Nitish Yadav,a Kuldeep Mishrab and S. A. Hashmia,* a Department of Physics and Astrophysics, University of Delhi, Delhi-110007, Delhi b Department of Physics and Material Science and Engineering, Jaypee University Anoopshahr, Anoopshahr-203390, Uttar Pradesh

*Corresponding Author, E-mail: [email protected]

Graphical Abstract

Research Highlights    

A PVdF-HFP and EC-PC-NaClO4 based porous polymer electrolyte (PPE) is prepared using a simple phase-inversion method. A mechanism is proposed for pore-formation using TGA and DSC. PPE shows excellent flexibility and high ionic-conductivity, suitable as electrolyte for EDLCs. Activated carbon based EDLC with PPE offers capacitance ~130-150 F g-1, efficient enough to glow LED.

1

Abstract We report the poly(vinylidene fluoride-co-hexfluoropropylene) (PVdF-HFP) based porous polymer electrolyte membranes, prepared via phase-inversion/solvent-nonsolvent methods, activated with an organic liquid electrolyte ethylene carbonate (EC):propylene carbonate (PC)-NaClO4 for the application in electric double layer capacitor (EDLC). The simple, quick and environment-friendly phase-inversion method, involving condensing steam as non-solvent, has been taken as the optimized process to obtain the porous PVdF-HFP film. The films of porous PVdF-HFP and the electrolyte (after soaking with liquid electrolyte) have been characterized for their morphological/structural aspects, porosity, liquid electrolyte retention, interaction with electrolyte, thermal properties, electrochemical stability and ionic conductivity. A pore-formation mechanism during phase-inversion at 100°C has been proposed on the basis of thermal studies. The electrolyte film has been found to have excellent mechanical flexibility, porosity (~80%), electrolyte retention (~400%), ionic conductivity (~2 mS cm-1 at room temperature), and electrochemical stability window (ESW) of ~4.35 V. The EDLC, fabricated with activated carbon electrodes and porous polymer electrolyte, exhibits excellent performance characteristics in terms of the specific capacitance (~150 F g-1, evaluated from EIS), specific energy (~17.7 Wh kg-1) and specific power (14.3 kW kg-1). The device shows stable specific capacitance (after ~17% initial fading) and high Coulombic efficiency (over 99%) for ~10,000 charge-discharge cycles. Keywords:

Porous polymer electrolyte, poly(vinylidene fluoride-co-hexfluoropropylene,

Phase-inversion, Ionic conductivity, Electric double layer capacitor

2

1. Introduction Polymer-based electrolytes are widely used in energy storage devices like ion-batteries and supercapacitors due to their various advantageous properties including leakage prevention, flexibility, mechanical/dimensional stability, low dendrite formation tendency, etc. [1-4]. They are generally categorized in two broad classes namely: (i) solvent-free solid polymer electrolytes (SPEs), and (ii) gel polymer electrolytes (GPEs) [1-4]. While SPEs increase device safety by preventing direct contact between the electrodes, they have lower ionic conductivity than the bulk liquid electrolytes and also suffer from problems like poor interfacial contacts at electrolyteelectrode interfaces [3-6]. On the other hand, GPEs are a class of polymer electrolytes comprising aqueous, organic or ionic liquid based liquid electrolytes immobilized in host polymers e.g. poly vinylidene fluoride (PVdF), poly vinylidene fluoride-co-hexafluoropropylene (PVdF-HFP), poly methyl methacrylate (PMMA), poly ethylene oxide (PEO), etc. [7-16]. Such electrolytes are of quasi-solid state nature offering liquid-like ionic conductivity of the order of 10-4 – 10-2 S cm-1 at room or ambient temperatures. Although these electrolytes are widely used these days in various energy storage applications e.g. Li/Na-ion batteries, supercapacitors etc. [7, 13-20], they suffer from few drawbacks including poor mechanical and thermal stabilities. Another important class of polymer-based electrolytes, referred as porous polymer electrolytes (PPEs) or microporous gel polymer electrolytes (MPEs), was introduced by Tarascon and coworkers [21]. The first PPE was prepared by the extraction/activation process in which di-butyl phthalate (DBP) was extracted from the solvent-cast film of PVdF-HFP and DBP mixture by solvent-extraction process [21]. This process was followed by soaking of liquid electrolyte (ethylene carbonate (EC)-dimethyl carbonate (DMC)-LiPF6), referred as activation process. However, the extraction method to prepare porous membranes has been found to be 3

tedious and expensive due to the large consumption of organic solvents to extract DBP like components [22, 23]. Thereafter, the phase-inversion technique has become a popular method to prepare porous films of polymers, introduced by Du Pasquier et al. [24]. In this method, the solvent/non-solvent mixture is allowed to evaporate simultaneously from the polymer solution, which is cast to prepare the film. Thereafter, various PPEs have been reported based on porous polymer membranes prepared from different host polymers e.g. PVdF, PVdF-HFP, PEO, PMMA, polyacrylonitrile (PAN), polyvinyl carbonate (PVC), etc. [25-38]. The porous membranes have been activated by different organic liquid electrolytes viz: LiPF6/EC-PC [21], LiPF6/EC-DMC [25, 31, 38], LiClO4/EC-PC [26, 28, 34], LiClO4/EC-DMC [27], LiBF4/ECDMC [27], LiPF6/EC-DEC [29, 36], LiClO4/EC-DEC [30], LiPF6/EC-DMC-EMC [32, 33, 37], LiTFSI/EC-DMC [37], LiBOB/EC-DMC [37], LiClO4-BMIMBF4 [36], etc., to develop potential PPEs. The room temperature ionic conductivity values reported for such porous electrolytes are of the order of 10-3-10-2 S cm-1. Such electrolytes have been reported for use as promising separators/ electrolytes in lithium-ion systems. Few porous systems have been reported for sodium ion batteries also e.g. a porous PVdF-HFP film has been reported offering ionic conductivity of 0.60×10-3 S cm-1 at room temperature after activation with 1M NaClO4 in EC:DC:DEC and shows better electrochemical performance compared to commercial (Celegrad 2730) membrane [23]. Electrical double layer capacitors (EDLCs), also referred as supercapacitors, are a class of energy storage devices, in which high surface area porous carbon electrodes are employed to form electrochemical cells, separated by liquid or polymer-based electrolytes [39-43]. Various forms of carbon are used in EDLC electrodes including, primarily, activated carbon (AC; powder, fibre or fabric) [39-45], carbon nanotubes (CNTs) [16, 41, 42, 46, 47], carbon 4

nanofibres (CNFs) [48-50], carbon aerogels [51] and graphene [17, 40, 52, 53]. Most of the reported EDLCs are based on liquid electrolytes, which are soaked with different types of commercially available separators [41, 42, 51-55]. In recent years, the gel polymer electrolytes have been widely reported as separators/electrolytes in EDLCs due to their advantageous properties including their high ionic conductivity and flexible nature [16, 17, 45-47, 49]. Except a few studies [57-59], the PPEs have been hardly used as electrolytes in EDLCs/supercapacitors, despite their excellent electrochemical and mechanical properties. In the present study, we report the porous polymer electrolytes based on PVdF-HFP porous films, prepared by phase-inversion technique using two different combinations of solvents and non-solvents namely DMF/glycerol and DMF/water (steam), activated by the liquid electrolyte EC-PC-NaClO4. Morphological, structural, porosity and electrochemical studies have been performed to test the suitability of PPEs as EDLC-electrolyte. The optimized PPE film has been employed to fabricate EDLCs using commercial activated carbon electrodes. The performance of the EDLC cells has been evaluated by a.c. impedance spectroscopy, cyclic voltammetry and galvanostatic charge– discharge tests. 2. Experimental 2.1 Preparation of porous films PVdF-HPF (average molecular weight ~400,000) was procured from Sigma-Aldrich and used as received. DMF (AR grade) and glycerol (GR grade) were obtained from Spectrochem (India) and Merck, respectively. Porous PVdF-HFP films were prepared by phase-inversion (PI) and solvent-nonsolvent (SNS) methods. In the phase-inversion process, steam at the atmospheric pressure was used as the non-solvent, and DMF was employed as the solvent. 1g of P(VdF-HFP) was dissolved in 20 ml of DMF by continuous stirring for 12 hours at ~80 °C using a magnetic 5

stirrer. After full dissolution, the solution was poured in a glass petridish. The solution in the petridish was then kept in the steam atmosphere at 100 °C at normal pressure (~1 atm). The humid atmosphere initiates a diffusive injection of water (non-solvent) in the solution [60], whereas the heat from the injected water helps in the evaporation of DMF. After the slow cooling, the phase separation started quickly and film like texture was obtained within the first 15 minutes. The complete removal of DMF happened in about 1 hour after which the lightbrownish and smooth porous film was peeled-off from the dish. The film was washed with double distilled water once or twice and kept to dry in a vacuum oven at ~60 °C overnight. In the solvent-nonsolvent process, the glycerol was used as the non-solvent and pore inducer, and DMF as the solvent. The PVdF-HFP solution was stirred thoroughly for 12 hours at ~80 °C and then kept for vacuum drying in a glass petri-dish at ~80 °C until the complete evaporation of DMF. The glycerol was then washed away with double-distilled water. A porous and smooth film was obtained which was further dried at 60 °C to remove the traces of water. Five porous films were prepared using the solvent/non-solvent method by varying the amount of glycerol (non-solvent), i.e. 1, 2, 3, 4 and 5 ml of glycerol in 20 ml DMF. The film made by phase-inversion process is coded as PI-film while the films made by solvent/non-solvent process are given the code SNS-x (x = 1, 2, 3, 4, 5 ml of glycerol). Table 1 summarizes the method of preparation, codes, etc. 2.2 Physical characterization of porous polymer electrolytes The morphology and structure of the porous films was characterized using field-emission scanning electron microscopy (FESEM) and high resolution X-ray diffraction (HR-XRD). Fourier-transform infrared (FTIR) spectroscopy was also used to identify different possible interactions of the salt-solvent-polymer. FESEM was performed using a TESCAN MIRA3 LMH

6

(Czechoslovakia) instrument, set at a potential difference of 20 kV. Samples for FESEM were sputtered with gold to prepare conductive surfaces. The HR-XRD was performed on an X-ray diffractometer (D8 Discover, Bruker AXS Analytical Instruments, Germany) using CuKα radiation (λ = 1.5406 Å) with 2θ varying from 5 to 60° at a scan rate of 2o min-1. FTIR measurements were performed on a Spectrum RX I FT-IR Spectrophotometer (Perkin-Elmer, USA) in the wave number range from 400 to 4000 cm-1. The spectra were recorded by averaging 32 scans per sample with an optical resolution of 4 cm−1. Thermo-gravimetric analysis (TGA) was done on a Perkin Elmer TGA7 system from room temperature to 600 °C at a heating rate of 10 °C min-1 in dry N2 atmosphere. The modulated differential scanning calorimetry (mDSC) was performed using Differential Scanning Calorimeter (model Q100, TA Instruments) for the temperature range from -80°C to 180°C at a heating rate of 3°C min-1 in static N2 atmosphere. The samples were kept in sealed aluminum pans for the DSC analysis. Linear sweep voltammetry (LSV), electrochemical impedance spectroscopy (EIS), d.c. polarisation and cyclic voltammetry (CV) measurements were performed on the electrochemical analyzer (CHI660E, CH Instruments, USA). Electrochemical stability window (ESW) of the electrolyte was measured by LSV with a stainless steel (SS) foil as working electrode and a silver (Ag) foil as a combined reference and counter electrode. Ionic conductivity of the electrolyte was measured by EIS in the frequency range from 105 to 1 Hz with 10 mV signal amplitude applied at SS|GPE|SS cell configuration. Temperature dependent conductivity measurements were done on a Broadband Dielectric/Impedance Analyzer (Novocontrol C50 Alpha A, Germany) from -50 to 100 °C.

7

2.3

Preparation of AC electrodes and characterization of EDLCs Activated carbon (AC, YEC-8) was procured from Fuzhou Carbon Co., China. The

specific surface area and porosity analysis were carried out using a surface area and pore size analyser (Gemini-V, Micromeritics, USA). The N2-adsorption-desorption isotherm was recorded and found close to the Type-I isotherm, showing predominantly micropores, in addition to the mesoporous interiors with average pore size of ~36 nm. The specific surface area was observed to be ~1123 m2g-1. The capacitor electrodes were prepared by mixing AC powder, acetylene black (AB) and PVdF-HFP properly in the weight ratio of 70:10:20 in a mortar-pestle. The mixture was then drop-cast onto graphite sheets of 1 cm × 1 cm dimension. These electrodes were then vacuum dried overnight before use. The EDLC cells were fabricated by sandwiching porous polymer electrolyte film between two symmetric and planar electrodes. EIS, CV and galvanostatic charge-discharge studies were carried out to characterize the EDLCs. EIS was performed from 105 to 10-2 Hz at a signal level of 10 mV. Cyclic voltammetry (CV) was done at different scan rates. EIS and CV were performed using the electrochemical analyzer, mentioned above. Galvanostatic charge-discharge studies were carried out on a BT2000 (Arbin Instruments, USA) system. 3. Results and Discussion 3.1 Electrolyte Characterization 3.1.1 Electrolyte retention of porous polymer films Porosity of the polymer films was estimated by the measurement of absorption of nbutanol [34]. Small pieces of films were weighed before being placed in n-butanol and at different time intervals until saturation was achieved. The mass-uptake of n-butanol was determined using equation (1) [34] and the results are shown in Fig. 1(a). 8

The plots show a rapid increase in n-butanol uptake in the first hour, after which the curves get flattened as all the pores get saturated with the liquid. Porosity of the film corresponds to the maximum uptake of n-butanol, which is almost same (~80 %) for both the porous PVdF-HFP (PI and SNS-2) films. The leakage/retention of the liquid electrolyte 1M NaClO4 in EC:PC (v/v) binary solvent in porous polymer films was estimated. All the films were kept in the electrolyte solution for a period of three hours, taken out of the solution and kept between two pieces of ordinary paper to remove the excess solution from the surface. The films were then placed under a constant pressure of 80 kPa and the weight of the films were measured at different time intervals. The absorbed/retained electrolyte in the films was quantitatively estimated using the following equation (2) [34]: Electrolyt e absorbed / retained 

(Wt - W0 ) 100 % W0

......(2)

The plot of electrolyte retained/absorbed vs. time is shown in Fig. 1(b). The least tendency of leakage of electrolyte has been found for the films prepared by solvent/non-solvent method, particularly SNS-2 and SNS-3. The PI film also has a slightly lower, but comparable retention tendency. Considering the porosity and electrolyte retention characteristics, along with the simpler and less chemical-consuming (thus more environment friendly) synthesis process involved, we choose the PI film as our template to make porous electrolyte films. 3.1.2 Morphology The polymer films are visually observed to be opaque, flexible and free-standing with a rough texture due to the presence of pores. To demonstrate the mechanical flexibility, the PI 9

films (dry and saturated with liquid electrolyte) have been twisted in the spiral form and bent in the form of closed loops, as shown in Fig.2 (a-d). It has been observed that after releasing the external force, the films recover their original shapes. It has also been noted that the porous film maintains its integrity even after getting saturated with the liquid electrolyte (EC:PC/NaClO4), as depicted in Fig.2(c and d). FESEM micrographs of the films are shown in Fig. 3(a-d). At about 50 μm scale, the micrographs clearly show different pore geometries in the porous polymer films prepared from the two different methods. The pores in the PI film are observed to be elongated while those in SNS-2 film are spherical. In the latter case, the pores have relatively more volume fraction. However, the electrolyte uptake (as determined above) is same in both the cases, which indicates the higher pore density in the PI-film at an even smaller scale of 500 nm. 3.1.3 X-ray diffraction and FTIR studies Fig.4 shows the XRD patterns of PVdF-HFP films, prepared using "solution-cast" method, solvent-nonsolvent method (SNS-2), phase-inversion technique (PI) and the PI film absorbed with liquid electrolyte 1M NaClO4 in EC:PC. The PI and SNS-2 films show exactly overlapping XRD patterns, indicating the presence of similar crystallographic features, independent of the methods of synthesis. The strong peak at 18.6° and a low intensity peak at 38.9°, corresponding to (100) + (020) and (021) planes, respectively, show the presence of nonpolar crystalline α-phase (trans-gauche-trans-gauche’) of PVdF in pure PVdF-HFP films [61]. The peak at 20.3° of (110) plane reflects the presence of polar crystalline γ-phase (transtrans-trans-gauche-trans-trans-trans-gauche’) in PVdF-HFP films [62]. In the porous PVdFHFP films (PI and SNS-2), the crystalline peaks at 18.6° and 20.3° are found to be slightly broader and the intensity of the peak at 20.3° and 38.9° is suppressed as compared to the solution-cast film. This shows the partial disruption in the crystalline domains during the

10

preparation of the porous films, which is possibly due to pore-formation in PVdF-HFP at high temperature. As the phase separation occurs, the polymeric chains which are at a temperature of around 80-100°C do not get time to form crystalline domains (especially the polar domains). After absorption of the liquid electrolyte (1M NaClO4 in EC-PC) in the porous film, the crystallinity of the film has been found to be further disturbed. The peaks at 18.6° (α-phase) almost disappear under the presence of a broad amorphous halo from ~10° to 33°. The intensity of the peaks at 20.3° (γ-phase) and 38.9° (α-phase) has also been suppressed in the presence of liquid electrolyte in the porous film, indicating the substantial amorphous nature. FTIR spectra of solution-cast PVdF-HFP film (Po), porous PI-film (dry), porous PI film saturated with liquid electrolyte, EC-PC binary mixture (1:1 v/v) and liquid electrolyte (1M NaClO4 in EC-PC) are shown in Fig. 5. The pure solution-cast PVdF-HFP has three major regions, a non-polar crystalline α-phase with bands at 764, 978, 1229 and 872 cm-1, a polar crystalline γ-phase with bands at 774, 812 and 836 cm-1, and an amorphous region with bands at 880 and 840 cm-1 [63, 64]. We expect to see changes in the crystalline content upon (i) pore generation via phase inversion, and (ii) incorporation of the liquid electrolyte inside the pores films. Indeed, the dry porous film has has lesser crystalline bands than a solution-casted PVdFHFP film. The bands at 764, 774, 812, 978 cm-1 of α-phase are not present in the porous dry film. The band at 796 cm-1 of CF3 stretching of α-phase is also not present in the porous PI film, while the band at 1229 cm-1 for the α-phase is weakly present in the porous film. However, the bands at 836 cm-1 (mixed mode of CH2 rocking and CF2 asymmetric stretching vibration) and 872 cm-1 (combined CF2 and C-C symmetric stretch) corresponding to the polar γ-phase are clearly seen in PI PVdF-HFP film. These observations indicate the disruption of the α-phase of PVdF in PI PVdF-HFP film during the synthesis process, as evident from the XRD study. To understand the 11

reduction in α-phase peaks we must know the conditions for formation of the α- and γ- phases. The α-phase is dominant in usual evaporation processes while the γ-phase appears, for PVdF, at annealing temperatures between 140 °C and 168 °C or for long annealing times [64, 65]. If we assume the same for PVdF-HFP then the high temperature (100 °C) and annealing period lead to destruction of the α-phase and retention of the γ-phase. In the liquid-filled polymer electrolyte films, the ion conduction is attributed to the liquid like motion of the ion through the interconnected pores of the polymer film. In the present case, Na+ and ClO4- ions present in the liquid electrolyte contribute to the ionic conductivity. However, the plasticizing effect of EC:PC may also disturb the crystalline structure of the polymer. But, the crystalline γ-phase bands at 836 cm1

and 872 cm-1 remain unchanged upon introduction of the liquid electrolyte in the porous matrix

implying lack of any interaction between these two phases. The slightly visible α-phase band at 1228 cm-1 for the dry PI-film is also almost unchanged in wet film, again confirming very weak interaction, rather lack of interaction, of the polymer with the electrolyte. 3.1.4 Thermal studies Thermal properties of dry PI film and the PI film, soaked with the liquid electrolyte (EC:PC/NaClO4) have been studied using TGA and DSC. The TGA and DSC have also been performed for solution-cast film for comparison. Fig. 6(A) shows the TG curves for solution-cast film, PI (dry) and PI (soaked with electrolyte) films. The solution-cast film shows the usual TG pattern indicating the thermal stability up to 430 °C, after which a drastic mass loss begins (Fig. 6A-a). Almost 100% of the material gets decomposed up to the temperature ~565 °C. On the other hand for the PI (dry) film, the mass loss becomes very slow at ~481 °C after the decomposition of about three-fourth mass of the material and shows a plateau-like region (Fig.6A-b). One possible explanation for this plateau at non-zero value is the separation of 12

polymer chains into two spatially distinct regions during the porous film formation (Fig. 6B). The polymer chains segregate themselves according to their chain lengths (masses) during pore formation. While the large chains reside in the interior of the walls enclosing the pores, the smaller/lighter chains form the inner lining of the pores [66]. The sudden mass loss in the PI (dry) film is due to the smaller chains around the pore walls, and the final plateau-like region is possibly due to the remaining larger chains. In the case of solution-cast film we observe an increase in the slope of the steep drop at a temperature of 481 °C (which is the same temperature at which the final plateau of PI (dry) film begins). This may be due to the larger chains, along with the smaller ones, decomposing in a cooperative manner until the solvent-cast, homogenous polymer has fully decomposed. We observe a similar indication of the mechanism, discussed above, in the DSC result as discussed below. The PI film (soaked with EC:PC/NaClO4) shows stability up to ~80 °C with a small mass loss of ~5% possibly due to the desorption of moisture (Fig.6A-c). Thereafter, it shows a sharp decline in mass till ~232 °C followed by a plateau of constant remaining mass of ~35 wt.% till ~387 °C. It again looses mass rapidly up to ~440 °C with residual mass of ~13 wt.% till the end of the measurement. The first fast decline is attributed to the loss of liquid solvent EC:PC and the second to the evaporation of the polymer PVdF-HFP. Beyond ~440 oC, a very slow mass loss is observed which is attributed the decomposition of the salt NaClO4 following the reaction: NaClO4  NaCl  2O2 [67].

Fig. 6(C) shows the mDSC curves for solution-cast film, PI (dry) film and PI film saturated with the electrolyte. The solution-cast film shows its usual characteristic, indicating asymmetric melting peak at ~140 °C, whereas the PI (dry) film shows two distinct melting peaks, one at ~140 °C and the other at ~112 °C (Fig. 6C-b). After the de-convolution of the 13

asymmetric melting peak for solution-cast film, the two peaks are found at ~140 °C and ~124 °C, as shown in the Fig. 6(D). The PI film, saturated with the electrolyte, also shows two such melting peaks but at lower temperatures (a strong peak at ~87.5 oC and a weak peak at 70.5 oC). The resolution of two melting peaks, distinctly observed for PI films, confirms the separation of low and high mass chains of PVdF-HFP during the pore formation (Fig. 6B), as discussed above in TGA, i.e. the larger chains tend to stay in the polymer rich phase, whereas low mass (smaller) chains move towards the polymer poor phase [66]. The peak at higher temperature (e.g. ~140 °C for PI films) is attributed to the melting of the larger chains, while the peak at lower temperature is due to the smaller chains. An additional peak has been observed at ~-3.4 °C which is attributed to the melting temperature of liquid electrolyte (EC:PC-NaClO4). 3.1.5 Ionic conductivity and electrochemical stability The room temperature ionic conductivity of the liquid electrolyte (1M NaClO4 in EC:PC) is reported to be ~8 mS cm-1 [68]. When liquid electrolyte is soaked in the porous PI film, the ionic conductivity reduces to ~2 mS cm-1. However, the order of conductivity remains same with respect to the liquid electrolyte and good enough for various electrochemical applications. The value of conductivity is comparable to the reported values e.g. porous polymer films soaked with 1M LiPF6 in EC:DMC (1:1 w/w) ( = 1.9 mS cm-1) [25] and 1M NaClO4 in EC:DMC:DEC (1:1:1 w/w/w) ( = 0.60 mS cm-1) [23]. Recently, Janakiraman et al. presented an electrospun PVdF fiber film adsorbed with 1M NaClO4 in EC:DEC binary solvent offering ionic conductivity of ~0.74 mS cm-1 at 29 °C [69]. The temperature-dependent ionic conductivity variation is shown in Fig.7 showing a typical curved plot which is fitted with VTF (Vogel- Tamman -Fulcher) equation:

(3) 14

where, B is pseudo-activation energy (K), indicating the rate of change of conductivity with temperature, To is ideal glass transition temperature (K) and A (in S cm-1) is the pre-exponential factor which indicates the conductivity at infinitely high temperature. It quantifies the temperature dependent visco-elastic behavior and diffusion coefficients of a glassy system. The VTF equation explains the intermolecular cooperative relaxation in the gel polymer electrolyte films [70]. After non-linear regression fitting of the VTF equation, the ln T1/2 versus 1/(T-To) plot (as shown in the inset of Fig.7) is observed to be linear and the fitting parameters are found to be A = 0.79 S cm-1 K1/2, B = 444 K and To = 161 K. The excellent VTF fit of the porous polymer electrolyte film, in the present studies, explains the direct connectivity of the ionic conductivity of the electrolyte with the viscosity and ion diffusivity of the system. Also, the To value for this system is close to 160 K, about 20 °C lower than the To (180 K) for a GPE film incorporating 1-butyl-1-methylpyrrolidinium bis(trifluormethane sulfonyl) imide ionic liquid. This indicates an earlier onset of glassy behavior and so better ionic diffusivity [47]. Another important parameter that can be found from the VTF fit of the temperature dependence of ionic conductivity is the “fragility index” of the polymer film [71]. Quantitatively, fragility index (F) is the ratio of To (K) and B (K), which varies between 0 and 1. The reciprocal the fragility index is referred as the “strength parameter (D)” [70]. For the PI film saturated with liquid electrolyte, the value of F is found to be 0.36. This value is comparable to that of mechanically stable solutioncast ionic liquid based GPE (PVdF-HFP/EMImTf) reported by us earlier [72]. The electrochemical stability window (ESW i.e. working voltage range) of the PI film (soaked with liquid electrolyte) has been estimated by LSV on the cell: SS| porous polymer electrolyte |Ag using SS as working electrode and Ag as combined counter and reference electrodes. The ESW has been found to be between -2V and 2.35 V versus Ag (i.e. ~4.35 V) as 15

noted from Fig.8, which is in good agreement with the value reported earlier in literature [68, 69]. This is a sufficient working voltage range for supercapacitor-like applications. 3.3 Characterisation of EDLC cells Symmetric EDLC cells, fabricated with activated carbon electrodes and porous polymer electrolyte film (i.e. AC|PI-film soaked with liquid electrolyte|AC), were tested using EIS, CV and galvanostatic charge-discharge tests for prolonged cycles. The detailed descriptions are given in the following sections. 3.3.1 EIS and CV studies The EIS plot for the EDLC cell is shown in Fig.9(a), which indicates a steeply rising capacitive pattern in the lower frequency range, recorded up to 10 mHz. A small semicircular spur has been obtained in the high frequency region followed by a small portion of linear pattern making ~45o angle with respect Z’-axis. These features show bulk properties of the activated carbon electrode, porous polymer electrolyte and electrode-electrolyte interfacial properties of the EDLC cell. Various performance indicator parameters like charge-transfer resistance (Rct), bulk resistance (Rb), specific capacitance (Csp) evaluated at the frequency 10 mHz, and knee frequency (fk) have been evaluated. The resistive parameters Rb and Rct values are evaluated from the semicircular part of the impedance plot, marked by arrows (Fig.9a). The specific capacitance for the cell has been evaluated by the expression: Csp = 2/(m×Z”), where m is mass of the AC powder used in single electrode,  is angular frequency and Z” is the imaginary impedance recorded at 10 mHz. The knee frequency (which determines the rate performance) of the cell is also marked by an arrow on the impedance plot (Fig.9a) below which the cell shows steep rising capacitive behaviour. The values of Rb, Rct, Csp and fk are observed to be ~17 Ω cm2, ~27 Ω cm2, ~150 F g-1 and ~82 mHz, respectively. The resistive values are comparable to the 16

values generally reported for the solid state EDLCs incorporated with polymer-based electrolytes [16, 17, 46, 47, 58]. A substantially high value of specific capacitance, leading to a high specific energy of the device, indicates the excellent compatibility of the activated carbon electrodes with the porous polymer electrolyte in the present studies. The EIS plot of the present EDLC can be represented by an equivalent circuit (modified Randle’s circuit) [73-75], as depicted in the inset of Fig.9(a). The high frequency suppressed semicircular spur (Fig.9a) gives the combined effect of the bulk resistance of the electrolyte and the resistance of the two electrodes, represented by Rb [73-75]. The mid-frequency region represents the parallel combination of charge transfer resistance Rct and the double layer capacitance Cdl. The linear region of the 45o slope is attributed to the Warburg impedance element (W) owing to the ion diffusion via porous activated carbon electrodes [75]. The capacitance at the electrode-electrolyte interface, observed at the low frequency region, is represented by the capacitive element, CL. In addition to the knee frequency fk, evaluated above, the rate performance of the EDLC cell and hence its pulse power performance has also been estimated from the Bode plots (i.e. Z’ and Z” variations as a function of frequency, log f), proposed by Miller [76], and the plots, proposed by Taberna et al. [77], i.e. the variation of C’ and C” versus frequency, as shown in Fig.9(b and c). In the Miller’s Bode plots (Fig.9b), the Z’ and Z” plots against frequency cross each other at a resonant frequency fo (at phase angle of 45o) indicated by arrow. The value of fo is found to be 0.024 Hz and its reciprocal is ~42 s, which is a characteristic response time of the cell. The pulse power density (Po) of the cell can be estimated from the expression: Po=Eo/o, where Eo is the specific energy evaluated at the frequency fo. The value of Eo has been evaluated from Eo = ½ CoV2, where Co is the capacitance evaluated at fo and V is rated voltage range of the

17

device (2 V in the present case). The values of Co, Eo and Po are found to be ~65 mF cm-2, ~17.4 Wh kg-1 and 1.5 kW kg-1, respectively. The response time of the cell has also been evaluated from the approach of Taberna et al. [77]. Accordingly, the C” versus log f (Fig.9c) shows a peak shape. The peak frequency fo and its reciprocal (o, referred as response time) has been observed in conformity with the values, obtained from Miller’s Bode plots, mentioned above. Further, The behaviour of C' at low frequencies is dependent on the electrolyte-electrode interface [77] and the steep rise at low frequencies may indicate good electrode-electrolyte interface which means the porous electrolyte is able to connect properly with the electrode material. The EDLCs cell has been electrochemically tested by the cyclic voltammetry (CV) also, in two electrode configuration. Fig. 9(d) depicts the CV patterns of the cell with a gradual increase in the voltage range from 0 to 2.0 V, recorded at a scan rate of 10 mV s-1. The shapes of the voltammograms are found to be almost rectangular up to +2.0 V. This indicates the use of porous polymer electrolyte safely up to 2.0 V with activated electrodes. Fig. 9(e) shows the CV patterns of the cell (recorded for the voltage range from 0 to +2V) at different scan rates. The values of specific capacitance (Csp) at different scan rates have been evaluated by Csp = 2i/(m×s), where i is average voltammetric current, s is scan rate and m is the mass of activated carbon used in the single electrode. The values of the specific capacitance (Csp) have been found to be 136, 114, 85 and 58 F g-1 at the scan rates of 10, 20, 50, and 100 mV s-1, respectively. The value of Csp at 10 mV s-1 is comparable to the value observed by EIS at the low frequency 10 mHz, mentioned above. The capacitance value of the cell with the present porous polymer electrolyte (PI film saturated with with 1M NaClO4 in EC:PC) and AC-electrodes at the scan rate of 10 mV s-1 is consistent with the values of the capacitance for the cells with liquid electrolyte (1M NaClO4 solutions) with similar electrodes [78].

18

The effect of the temperature variation on the performance of the EDLC cell has been tested by recording EIS plots at varying temperatures from 0 to 80 oC, as shown in Fig.10(a). The variation of specific capacitance (Csp), evaluated from EIS plots, as a function of temperature is shown in Fig.10(b). It may be noted that no significant change in the impedance pattern has been observed, indicating the thermal stability of the EDLC cell between 0 and 80 oC. As observed from Fig.10(b), a slight (almost 25%) decrease in Csp has been noted when temperature decreases from 80 to 0 oC. The present result on the temperature dependent variation of specific capacitance for the cell with porous polymer electrolyte is comparable, rather superior than many reported results on the cells with liquid electrolytes or gel polymer electrolytes. We earlier reported the temperature dependent behaviour of an EDLC incorporated with a humidity sensitive proton conducting polymeric electrolyte PVA-H3PO4 film [79], which shows a fast decreasing trend of specific capacitance with decreasing temperature. This trend was explained on the basis of the decrease in humidity on increasing the temperature [79]. In our another report, the EDLC cell with ionic liquid incorporated gel polymer electrolyte film shows the stable performance in a wider range of temperature compared to pure ionic liquids [46]. Arulepp et al. [80] have shown the effect of solvents on temperature dependent performance of EDLCs. Accordingly, the acetonitrile shows almost no fading in capacitance from 60 to -30 oC,

whereas, the propylene carbonate shows poor performance. Recently, Pandey et al. [81] reported the EDLCs with succinonitrile and ionic liquid incorporated gel polymer electrolyte offering wider electrochemical working temperature range. Ionogel-based solid state supercapacitor has recently been reported showing temperature dependent performance, comparable to the performance of the present EDLC [82].

19

3.3.2 Galvanostatic Charge-Discharge Studies of AC symmetric EDLCs Fig. 11(a) shows the charge-discharge profiles of the EDLC cell at a constant current of 1 mA cm-2 (0.95 A g-1) for higher voltage limits varying from 1.0 to 2.0 V. The profiles are almost triangular for all the voltage ranges indicating the capacitive nature of the EDLC up to 2.0 V. The discharge specific capacitance ‘Cd’ has been calculated from the linear region of the discharge curve using the following equation: Cd = 2it/(m×V), where i is the discharge current, ∆t is the discharge time interval corresponding to the voltage range ∆V, m is the mass of active material in single electrode. The value of internal resistance of the cell (also referred as ESR) has been estimated from the IR drop observed at the beginning of the discharge curve. The typical value of Cd is found to be ~130 F g-1 when the cell is charged up to 2.0 V for a current of 1 mA cm-2 (0.95 A g-1). Fig.11(b) shows the variation of discharge specific capacitance (Cd) and ESR with respect to the applied currents. A gradual decrease in Cd has been noted with increasing current loads. About ~40% decrease in Cd has been observed when applied current increases from 1 to 10 A g-1, which indicates a good rate capability of the cell. Further, almost no change in ESR (45-47  cm2) has been observed for applied current range from 1 to 10 A g-1 (Fig.11b), which affects the specific power values of the cell, discussed below. The low ESR value represents good contact between the AC-electrodes and the porous polymer electrolyte film. The specific energy (E) and specific power (P) of the cell has been evaluated from the following expressions:

E

CdV 2 8

(4)

P

V2 8m  ESR

(5)

20

where V is the maximum voltage for charging. The typical values of E and P are 17.7 Wh kg-1 and 14.3 kW kg-1, respectively, for the applied current of 0.95 (A g-1), which are comparable values with various reported values in literature for the solid-like capacitors [45, 47, 55, 58]. Fig. 11(c) shows the variation of specific power against specific energy (i.e. Ragone plot) for the cell. It may be noted that the specific power remains almost constant for the varying specific energy values. This is due to the invariance in the ESR values at different applied currents. The charged capacitors have enough stored energy capable to glow the LED for substantially longer time, as typically shown in Fig.11(e). The prolonged cyclic test has been performed on the EDLC cell. Fig.11(d) shows the variation of discharge specific capacitance (Cd) as a function of charge-discharge cycles for the potential window between 0.01 and 1V. The value of Cd is found to be ~101 F g-1 for the second discharge and increases to a maximum of ~120 F g-1 after about 3000 cycles.

Thereafter, a

slight decay, i.e. a fading in Cd of ~17%, has been observed eventually reaching to ~100 F g-1 by 10,000 cycles. This small amount of fading and high retention of Cd indicates the stable performance of the cell for 10,000 cycles or even more. The first charge-discharge cycle has been excluded in the measurements to avoid the undesired value of Cd, etc. due to the irreversible charge consumption, etc. The increase in the capacitance for initial cycles is generally not observed, however in the present case, the increase in Cd is observed possibly due to the improvement in the interfacial contact during the initial charge-discharge. The stability of the cell has also been tested by recording the EIS plots at the regular interval of charge-discharge cycles, as shown in the inset of Fig.11(d). Almost no deviation in the shape of the EIS plots has been observed for different cycles, which confirms the stability of the cell.

21

The Coulombic efficiency ‘η’ of the cell was evaluated using the expression: η = tD/tC×100%, when the same current was used for charging and discharging. The tD and tC are the times for discharge and charge, respectively. The values were found to be between 99 and 100% for the 10,000 charge-discharge cycles, which further confirms the stability of the cell for large number of cycles. Such higher values of Coulombic efficiency are generally observed for the electrode and liquid electrolyte interfaces [83]. This suggests the liquid-like behavior of the porous polymer electrolyte, when interfaced with capacitor electrodes. 4. Conclusions Porous PVdF-HFP film has been prepared using simple phase-inversion method involving DMF as solvent and steam under normal atmospheric pressure as non-solvent. The film shows excellent retention (~400%) of liquid electrolyte (EC-PC-NaClO4), which offers ionic-conductivity of ~2 mS cm-1 at room temperature. XRD and FTIR studies confirm the reduction in crystalline content of the film and lack of interaction between the polymer matrix and trapped liquid electrolyte. The thermal studies (TGA and DSC) hint towards a mechanism of pore formation via polymer chain distribution depending upon length of the chains. The porous polymer electrolyte membrane has been found to be a suitable electrolyte/separator for EDLC cell (fabricated using AC powder electrodes), which offers high specific capacitance, energy and power (~130 F g-1, ~17.7 Wh kg-1 and 14.3 kW kg-1, respectively, at the current of 0.95 A g-1). The prolonged cyclic tests indicate the stable performance of the EDLC cell with only ~17% of initial fading in specific capacitance and high Coulombic efficiency between 99 and 100% up to ~10,000 charge-discharge cycles.

22

Acknowledgements The authors are grateful to the University of Delhi for the “Research and Development grant” as financial support to complete this work. One of us (NY) is also thankful to University Grants Commission, New Delhi for providing “Senior Research Fellowship.” References [1] [2]

[3] [4] [5] [6] [7] [8] [9]

[10]

[11]

[12]

[13]

C. Sequeira, D. Santos, Polymer electrolytes - Fundamentals and applications, Woodhead Publishing Limited (Cambridge) 2010. R.C. Agrawal, G.P. Pandey, Solid polymer electrolytes: materials designing and all-solidstate battery applications: an overview, J. Phys. D: Appl. Phys. 41 (2008) 223001 – 2230017. D.T. Hallinan Jr, N.P. Balsara, Polymer Electrolytes, Annu. Rev. Mater. Res. 43 (2013) 503 – 525. Editorial, Polymer electrolytes: Present, past and future, Electrochim. Acta 57 (2011) 4– 13. J.Y. Song, Y.Y. Wang, C.C. Wan, Review of gel-type polymer electrolytes for lithium-ion batteries, J. Power Sources 77 (1999) 183–197 W.H. Meyer, Polymer electrolytes for lithium-ion batteries, Adv. Mater. 10 (1998) 439 – 448. M. Stephan, Review on gel polymer electrolytes for lithium batteries, Eur. Polym. J. 42 (2006) 21–42 G.B. Appetecchi, F. Croce, A. De Paolis, B. Scrosati, Poly(vinylidene fluoride)-based gel electrolyte membrane for lithium batteries, J. Electroanal. Chem. 463 (1999) 248–252. Y. Kumar, S.A. Hashmi, G.P. Pandey, Lithium ion transport and ion–polymer interaction in PEO based polymer electrolyte plasticized with ionic liquid, Solid State Ionics 201 (2011) 73–80. H. Cheng, C. Zhu, B. Huang, M. Lu, Y. Yang, Synthesis and electrochemical characterization of PEO-based polymer electrolytes with room temperature ionic liquids, Electrochim. Acta 52 (2007) 5789–5794. Z. Osman, N.H. Zainol, S.M. Samin, W.G. Chong, K.B. Md Isa, L. Othman, I. Supa’at, F. Sonsudin, Electrochemical Impedance Spectroscopy Studies of Magnesium-Based Polymethylmethacrylate Gel Polymer Electroytes, Electrochim. Acta 131 (2014) 148–153. M. Suleman, Y. Kumar, S.A. Hashmi, Structural and electrochemical properties of succinonitrile-based gel polymer electrolytes: role of ionic liquid addition, J. Phys. Chem. B 117 (2013) 7436−7443. W. Li, Y. Xing, X. Xing, Y. Li, G. Yang, L. Xu, PVDF-based composite microporous gel polymer electrolytescontaining a novelsingle ionic conductor SiO2(Li+), Electrochim. Acta 112 (2013) 183– 190. 23

[14] M. Rao, J. S. Liu, W. S. Li, Y. Liang, D.Y. Zhou, Preparation and performance analysis of PE-supported P(AN-co-MMA) gel polymer electrolyte for lithium ion battery application, J. Membr. Sci. 322 (2008) 314–319. [15] Y. Zhang, Y. Zhao, Z. Bakenov, D. Gosselink, P. Chen, Poly(vinylidene fluoride-cohexafluoropropylene)/poly(methylmethacrylate)/nanoclay composite gel polymer electrolyte for lithium/sulfur batteries, Solid State Electrochem. 18 (2014) 1111–1116. [16] G.P. Pandey, S.A. Hashmi, Ionic liquid 1-ethyl-3-methylimidazolium tetracyanoboratebased gel polymer electrolyte for electrochemical capacitors, Mater. Chem. A 1 (2013) 3372 – 3378. [17] M. Suleman, Y. Kumar, S.A. Hashmi, High-rate supercapacitive performance of GO/r-GO electrodes interfaced with plastic-crystal-based flexible gel polymer electrolyte Electrochim. Acta 182 (2015) 995–1007 [18] D. Kumar, M. Suleman, S.A. Hashmi, Studies on poly(vinylidene fluoride-co-hexafluoropropylene) based gel electrolyte nanocomposite for sodium–sulfur batteries, Solid State Ionics 202 (2011) 45–53. [19] V. Palomares, P. Serras, I. Villaluenga, K. B. Hueso, J. Carretero-Gonzalez, T. Rojo, Naion batteries, recent advances and present challenges to become low cost energy storage systems, Energy Environ. Sci. 5 (2012) 5884 – 5901. [20] A.S. Lee, J.H. Lee, S.S. Hwang, S. M. Hong, E. Kim, C. M. Koo, W. Na, Lithium ion capacitors fabricated with polyethylene oxide-functionalized polysilsesquioxane hybrid ionogel electrolytes, Electrochim. Acta, 188 (2016) 582–588. [21] J.-M. Tarascon, A.S. Gozdz, C. Schmutz, F. Shokoohi, P.C. Warren, Performance of Bellcore's plastic rechargeable Li-ion batteries, Solid State Ionics 86–88 (1996) 49-54. [22] G.C. Li, P. Zhang, H. Zhang, Y. Wu, A porous polymer electrolyte based on P(VDF-HFP) prepared by a simple phase separation process, Electrochem. Commun. 10 (2008) 18831885. [23] Y.Q. Yang, Z. Chang, M.X. Li, X.W. Wang, Y.P. Wu, A sodium ion conducting gel polymer electrolyte, Solid State Ionics 269 (2015) 1–7. [24] A. Du Pasquier, P.C. Warren, D. Culver, A.S. Gozdz, G.G. Amatucci, J.-M. Tarascon, Plastic PVDF-HFP electrolyte laminates prepared by a phase-inversion process, Solid State Ionics 135 (2000) 249-257. [25] D.-W. Kim, Y.-K. Sun, Electrochemical characterization of gel polymer electrolytes prepared with porous membranes, J. Power Sources 102 (2001) 41–45. [26] C.-G. Wu, M.-I Lu, H.-J. Chuang, PVdF-HFP/P123 hybrid with mesopores: a new matrix for high-conducting, low-leakage porous polymer electrolyte, Polymer 46 (2005) 5929– 5938. [27] F.A. Amaral, R.M. Sousa, L.C.T. Morais, R.G. Rocha, I.O. Campos, W.S. Fagundes, C.N.P. Fonseca, S.C. Canobre, Preparation and characterization of the porous solid polymer electrolyte of PAN/PVA by phase inversion, J. Appl. Electrochem. 45 (2015) 809-820.

24

[28] Q. Xiao, X. Wang, W. Li, Z. Li, T. Zhang, H. Zhang, Macroporous polymer electrolytes based on PVDF/PEO-b-PMMA block copolymer blends for rechargeable lithium ion battery, J. Membr. Sci. 334 (2009) 117-122. [29] L. Liu, Z. Wang, Z. Zhao, Y. Zhao, F. Li, L. Yang, PVDF/PAN/SiO2 polymer electrolyte membrane prepared by combination of phase inversion and chemical reaction method for lithium ion batteries, J. Solid State Electrochem. 20 (2016) 699-712. [30] N.T. Kalyana Sundaram, A. Subramania, Microstructure of PVdF-co-HFP based electrolyte prepared by preferential polymer dissolution process, J. Membr. Sci. 289 (2007) 1-6. [31] R. Miao, B. Liu, Z. Zhu, Y. Liu, J. Li, X. Wang, Q. Li, PVDF-HFP-based porous polymer electrolyte membranes for lithium-ion batteries, J. Power Sources 184 (2008) 420-426. [32] J.-H. Cao, B.-K. Zhu, Y.-Y. Xu, Structure and ionic conductivity of porous polymer electrolytes based on PVDF-HFP copolymer membranes, J. Membr. Sci. 281 (2006) 446– 453 [33] G.C. Li, P. Zhang, b, H.P. Zhang, L.C. Yang, Y.P. Wu, A porous polymer electrolyte based on P(VDF-HFP) prepared by a simple phase separation process, Electrochem. Commun. 10 (2008) 1883–1885. [34] F. Deng, X. Wang, D. He, J. Hua, C. Gong, Y. S. Ye, X. Xie, Z. Xue, Microporous polymer electrolyte based on PVDF/PEO star polymer blends for lithium ion batteries, J. Membr. Sci. 491 (2015) 82–89. [35] P. Yan, Z. Huang, Y. Lin, X. Wu, Y. Yang, D. Wang, F. Chen, C. Zhang, D. He, Composite-porous polymer membrane with reduced crystalline for lithium–ion battery via non-solvent evaporate method, Ionics 21 (2015) 593–599. [36] W. Zhai, H. Zhu, L. Wang, X. Liu, H. Yang, Study of PVDF-HFP/PMMA blended microporous gel polymer electrolyte incorporating ionic liquid [BMIM]BF4 for Lithium ion batteries, Electrochim. Acta 133 (2014) 623–630. [37] S.M. Seidel, S. Jeschke, P. Vettikuzha , H.-D. Wiemhöfer, PVDF-HFP/ether-modified polysiloxane membranes obtained via airbrush spraying as active separators for application in lithium ion batteries, Chem. Commun. 51 (2015) 12048-12051. [38] N.H. Idris, M.M. Rahman, J.-Z. Wang, H.-K. Liu, Microporous gel polymer electrolytes for lithium rechargeable battery application, J. Power Sources 201 (2012) 294–300. [39] B.E. Conway, Electrochemical Supercapacitor, Kluwer Academic Publishers/Plenum Press, New York, 1999. [40] T. Brousse, D. Bélanger, K. Chiba, M. Egashira, F. Favier, J. Long, J.R. Miller, M. Morita, K. Naoi, P. Simon, W. Sugimoto, Materials for electrochemical capacitors, in: Springer Handbook of Electrochemical Energy, 1st ed., Cornelia Breitkopf, Karen SwiderLyons (Eds.), Springer-Verlag Berlin Heidelberg, pp. 495-561 (2017). [41] P. Sharma, T.S. Bhatti, A review on electrochemical double-layer capacitors, Energy Conv. and Manage. 51 (2010) 2901–2912.

25

[42] A.K. Shukla, A. Banerjee, M.K. Ravikumar, A. Jalajakshi, Electrochemical capacitors: Technical challenges and prognosis for future markets, Electrochim. Acta 84 (2012) 165– 173. [43] N.-S. Choi, Z. Chen, S.A. Freunberger, X. Ji, Y.-K. Sun, K. Amine, G. Yushin, L.F. Nazar, J. Cho, P.G. Bruce, Challenges facing lithium batteries and electrical double-layer capacitors, Angew. Chem. Int. Ed. 51 (2012) 9994 – 10024. [44] P. Simon, P.‐L. Taberna, F. Béguin, Electrical double‐layer capacitors and carbons for EDLCs, in: supercapacitors: materials, systems, and applications, Eds. F. Béguin, E. Frackowiak, Wiley-VCH Verlag GmbH (2013) pp. 131-165. [45] G.P. Pandey, S.A. Hashmi, Y. Kumar, Performance studies of activated charcoal based electrical double layer capacitors with ionic liquid gel polymer electrolytes, Energy Fuels 24 (2010) 6644–6652. [46] G.P. Pandey, S.A. Hashmi, Y. Kumar, Multiwalled carbon nanotube electrodes for electrical double layer capacitors with ionic liquid based gel polymer electrolytes, J. Electrochem. Soc. 157 (2010) A105-A114. [47] M.K. Singh, Y. Kumar, S.A. Hashmi, 'Bucky gel’ of multiwalled carbon nanotubes as electrodes for high performance, flexible electric double layer capacitors, Nanotechnology 24 (2013) 465704-465713. [48] D. Lee, J.-Y. Jung, M.-J. Jung, Y.-S. Lee, Hierarchical porous carbon fibers prepared using a SiO2 template for high-performance EDLCs, Chem. Engg. J. 263 (2015) 62–70 [49] J. Ren, W. Bai, G. Guan, Y. Zhang, H. Peng, Flexible and weaveable capacitor wire based on a carbon nanocomposite fiber, Adv. Mater. 25 (2013) 5965–5970. [50] A. Ghosh, Y. H. Lee, Carbon-based electrochemical capacitors, ChemSusChem 5 (2012) 480 – 499. [51] M. Oschatz, S. Boukhalfa, W. Nickel, J. P. Hofmann, C. Fischer, G. Yushin, S. Kaskel, Carbide-derived carbon aerogels with tunable pore structure as versatile electrode material in high power supercapacitors, Carbon 113 (2017) 283 – 291. [52] M. H. Ervin, C. M. Pereira, J. R. Miller, R. A. Outlaw, J. Rastegar, and R. T. Murray, Graphene-based and other electrochemical double layer capacitors for energy harvesting systems, ECS J. Solid State Sci. Tech. 2 (2013) M3135-M3139. [53] J.R. Miller, R.A. Outlaw, B.C. Holloway, Graphene electric double layer capacitor with ultra-high-power performance, Electrochim. Acta 56 (2011) 10443– 10449. [54] S. Ye, L. Zhu, I.-J. Kim, S. Yang, W.-C. Oh, Characterization of expanded graphene nanosheet as additional material and improved performances for electric double layer capacitors, J. Ind. and Engg. Chem. 43 (2016) 53–60. [55] L.J. Wang, M.F. El-Kady, S. Dubin, J.Y. Hwang, Y. Shao, K. Marsh, B. McVerry, M.D. Kowal, M.F. Mousavi, R.B. Kane, Flash converted graphene for ultra-high power supercapacitors, Adv. Energy Mater. 5 (2015) 1500786 – 1500793. [56] R. Raccichini, A. Varzi, S. Passerini, B. Scrosati, The role of graphene for electrochemical energy storage, Nature Mater. 14 (2015) 271 – 279. 26

[57] C.-C. Yang, G. M. Wu, Study of microporous PVA/PVC composite polymer membrane and it application to MnO2 capacitors, Mater. Chem. Phys. 114 (2009) 948–955. [58] R. Na, G. Huo, S. Zhang, P. Huo, Y. Du, J. Luan, K. Zhu, G. Wang, A novel poly(ethylene glycol)–grafted poly(arylene ether ketone) blend micro-porous polymer electrolyte for solid-state electric double layer capacitors formed by incorporating a chitosan based LiClO4 gel electrolyte, J. Mater. Chem. A 4 (2016) 18116 – 18127. [59] D. Karabelli, J.-C. Leprêtre, F. Alloin, J.-Y. Sanchez, Poly(vinylidene fluoride)-based macroporous separators for supercapacitors, Electrochim. Acta 57 (2011) 98–103. [60] I.M. Wienk, R.M Boom, M.A.M. Beerlage, A.M.W. Bulte, C.A. Smolders, H. Strathmann, Recent advances in the formation of phase inversion membranes made from amorphous or semi-crystalline polymers, J. Membr. Sci. 113 (1996) 361-371. [61] J.-H. Cao, B.-K. Zhu, Y.-Y. Xu, Structure and ionic conductivity of porous polymer electrolytes based on PVDF-HFP copolymer membranes, J. Membr. Sci. 281 (2006) 446– 453. [62] S. Abbrent, J. Plestil, D. Hlavata, J. Lindgren, J. Tegenfeldt, Å. Wendsjö, Crystallinity and morphology of PVdF–HFP-based gel electrolytes, Polymer 42 (2001) 1407–1416. [63] Shalu, S.K. Chaurasia, R.K. Singh, S. Chandra, Thermal Stability, Complexing Behavior, and Ionic Transport of Polymeric Gel Membranes Based on Polymer PVdF-HFP and Ionic Liquid, [BMIM][BF4], J. Phys. Chem. B 117 (2013) 897−906. [64] Y. Bormashenko, R. Pogreb, O. Stanevsky, E. Bormashenko, Vibrational spectrum of PVDF and its interpretation, Polym. Testing 23 (2004) 791–796. [65] K.M. Kim, W.S. Jeon, N.-G. Park, K.S. Ryu, S.H. Chang, Effect of evaporation temperature on the crystalline properties of solution-cast films of poly(vinylidene fluoride)s, Korean J. Chem. Eng. 20 (2003) 934-941. [66] A. Dobry, F. Boyer-Kawenoki, Phase separation in polymer solution, J. Polym. Sci. 2 (1947) 90-100. [67] S. Shimada, R. Furuchi, A study on the thermal decomposition of KClO4 and NaClO4 by acoustic emission thermal analysis, Thermochim. Acta, 163 (1990) 313-316. [68] A. Ponrouch, E. Marchant, M. Courty, J.-M. Tarascon, M.R. Palacín, In search of an optimized electrolyte for Na-ion batteries, Energy Environ. Sci. 5 (2012) 8572-8583. [69] S. Janakiraman, A. Surendran, S. Ghosh, S. Anandhan, A. Venimadhav, Electroactive poly(vinylidene fluoride) fluoride separator for sodium ion battery with high coulombic efficiency, Solid State Ionics 292 (2016) 130–135. [70] I.-S. Park, K. Saruta, S. Kojima, Dielectric relaxation and calorimetric measurements of glass transition in the glass-forming dyhydroxyl alcohols, J. Phys. Soc. Japan 67 (1998) 4131-4138. [71] A. Angell, Why C1 = 16–17 in the WLF equation is physical—and the fragility of polymers?, Polymer 38 (1997) 6261-6266.

27

[72] S.A. Hashmi, M.Y. Bhat, M.K. Singh, N.T.K. SundaramBala, P.C. Raghupathy, H. Tanaka, Ionic liquid-based sodium ion-conducting composite gel polymer electrolytes: effect of active and passive fillers, J. Solid State Electrochem. 20 (2016) 2817–2826. [73] U. Retter, A. Widmann, K. Siegler, H. Kahlert, On the impedance of potassium nickel(ii) hexacyanoferrate(ii) composite electrodes-the generalization of the Randle’s model referring to inhomogeneous electrode. J. Electroanal. Chem. 546 (2003) 87-96. [74] T.C. Girija, M.V. Sangaranarayanan, Polyaniline-based nickel electrodes for electrochemical supercapacitors-influence of Triton X-100, J. Power Sources 159 (2006) 1519-1526. [75] M.K. Singh, M. Suleman, Y. Kumar, S.A. Hashmi, A novel configuration of electrical double layer capacitor with plastic crystal based gel polymer electrolyte and graphene nano-platelets as electrodes: a high rate performance, Energy 80 (2015) 465-473. [76] J.R. Miller, in: 8th International Seminar on double layer capacitor and Similar Energy Storage devices, Deerfield Beach, Florida; December 1998. [77] P.-L. Taberna, P. Simon, J.F. Fauvarque, Electrochemical characteristics and impedance spectroscopy studies of carbon-carbon supercapacitors, J. Electrochem. Soc. 150 (2003) A292-A300. [78] A. Laheäär, A. Jänes, E. Lust, NaClO4 and NaPF6 as potential non-aqueous electrolyte salts for electrical double layer capacitor application, Electrochim. Acta 82 (2012) 309– 313. [79] S.A. Hashmi, R.J. Latham, R.G. Linford and W.S. Schlindwein, Studies on all solid state electric double layer capacitors using proton and lithium ion conducting polymer electrolytes, J. Chem. Soc., Faraday Trans. 93 (1997) 4177-4182. [80] M. Arulepp, L. Permann, J. Leis, A. Perkson, K. Rumma, A. Jänes, E. Lust, Influence of the solvent properties on the characteristics of a double layer capacitor, J. Power Sources 133 (2004) 320–328. [81] G.P. Pandey, T. Liu, C. Hancock, Y. Li, X.S. Sun, J. Li, Thermostable gel polymer electrolyte based on succinonitrile and ionic liquid for high-performance solid-state supercapacitors, J. Power Sources 328 (2016) 510–519. [82] L. Negre, B. Daffos, V. Turq, P.L. Taberna, P. Simon, Ionogel-based solid-state, supercapacitor operating over a wide range of temperature, Electrochim. Acta 206 (2016) 490–495. [83] C. Schütter, A.R. Neale, P. Wilde, P. Goodrich, C. Hardacre, S. Passerini, J. Jacquemin, A. Balducci, The use of binary mixtures of 1-butyl-1-methylpyrrolidinium bis{(trifluoromethyl)sulfonyl}imide and aliphatic nitrile solvents as electrolyte for supercapacitors, Electrochim. Acta 220 (2016) 146-155.

28

Figure Captions Fig. 1: (a) Measurements of n-butanol uptake, and (b) electrolyte retention measurements (under a constant weight) of the porous PVdF-HFP films prepared by solvent-nonsolvent method (SNS'x' films) and phase-inversion method (PI-film). Fig. 2: Photographs of flexible porous PVdF-HFP films prepared using phase-inversion (PI) process i.e. PI (dry) film (a) twisted in spiral form and (b) folded in closed loop, and the PI film (soaked with liquid electrolyte (c) twisted in spiral form and (d) folded in closed loop. Fig. 3: FESEM micrographs of the porous polymer films (a and b) prepared by with solventnonsolvent method (SNS-2 film), and (c and d) prepared by phase-inversion method (PI-films). Fig. 4: XRD patterns of solution-cast and porous PVdF-HFP films, and PI-film saturated with liquid electrolyte 1M NaClO4 in EC-PC. Fig. 5: FT-IR spectra of (a) PVdF-HFP film, prepared by solution-cast method, (b) porous PVdF-HFP film (PI film, prepared by phase-inversion method), (c) PI film saturated with liquid elecrolyte 1M NaClO4 in EC:PC, (d) 1M NaClO4 in EC:PC, and (e) EC:PC mixture (1:1 v/v). Fig. 6: (A) TGA curves for PVdF-HFP film: (a) prepared using solution-cast method (b) phaseinversion method i.e. PI film without any electrolyte absorbed, and (c) PI film saturated with 1M NaClO4 in EC-PC, (B) schematic drawing of the possible mechanism of separation of light and heavy chains of polymer around a pore, (C) DSC curves for for PVdF-HFP film: (a) prepared using solution-cast method, (b) Phase-inversion method i.e. PI film without any electrolyte absorbed, and (c) PI film saturated with 1M NaClO4 in EC-PC, and (D) de-convoluted DSC curves for PVdF-HFP film. Fig. 7: Temperature dependence of ionic conductivity of the PI film soaked in 1M NaClO4 in EC-PC (1:1 v/v) solution. The ln (T1/2) versus 1/(T-T0) plot after VTF fit is shown as inset. The film thickness is ~140 µm. Fig. 8: LSV pattern of the cell SS| porous polymer electrolyte |Ag (SS: stainless steel foil) to estimate electrochemical stability window of the electrolyte. Fig. 9: (a) EIS plot of symmetric EDLC cell AC|porous polymer electrolyte|AC. The equivalent circuit of the cell is shown as inset. (b) Bode plots, (c) real and imaginary capacitance variation as function of frequency of the EDLC cell, (d) CV patterns for gradually increasing voltages recorded at scan rate of 10 mV s-1, and (e) CV patterns at different scan rates. Fig.10: (a) EIS plots for EDLC cell at different operating temperatures, (b) specific capacitance, ESR and charge-transfer resistance values calculated from the EIS plots at different. temperatures. Fig. 11: (a) Galvanostatic charge-discharge curves of EDLC for different voltage limits, (b) specific discharge capacitance and ESR as a function of current density, (c) Ragone plot, (d) specific discharge capacitance and Coulombic efficiency as a function of charge-discharge cycles 1V (inset shows EIS plots for fresh cell and after every 3,000 cycles), and (d) a demonstration of the glow of LED powered by the present EDLC. 29

(a)

(b)

500

60

PI film SNS-1 SNS-2 SNS-3 SNS-4 SNS-5

40

20

Electrolyte Uptake/%

n-butanol uptake/%

80

400 300 200 PI film SNS-1gm SNS-2gm SNS-3gm SNS-4gm SNS-5gm

100 0

0 0

5

10

15

t/h

20

25

0

5

10

15

20

25

t/h

Fig. 1

30

Fig. 2

31

Fig. 3

32

(110) PI film saturated with electrolyte PI film (dry) SNS-2 film Solution-cast film

Intensity/a.u.

(100) + (020)

(021)

5

10

15

20

25

30

35

40

45

50

55

60

2deg. Fig. 4

33

(e)

Transmission/%

(d)

(c)

(b)

(a)

1600

1400

1200

1000

Wavenumber/cm

800

-1

Fig. 5

34

(A) 100

60 40 (b)

20 (c) (a)

0 100

200

300

400

500

600

o

T/ C

(C) -1

(D)

(b)

(a)

-80

-40

0

40 80 o T/ C

120

160

Fit Peak 2

Fit Peak 1

o 124 C

Endo

Reversible Heat Flow/W g

(c)

Endo

Reversible Heat Flow/W g-1

Wt%

80

100

o 140 C

110

120

130

140

150

o

T/ C

Fig. 6

35

-2.0

-3.0

-4.0

1/2 1/2

-1

-3.5

-2 ln(T /S cm K )

log/S cm-1)

-2.5

-4.5

-4 -6 -8

0.006

0.012 -1

-1

0.018

(T-To) /K

2.6

2.8

3.0

3.2

3.4

3.6

3.8

4.0

4.2

4.4

4.6

1000/T/K-1

Fig. 7

36

2

j/mA cm

-2

1 4.35 V

0

-1

-2 -2

-1

0

1

2

3

E vs Ag/V

Fig. 8

37

-250

(a)

0.01 Hz

(b)

200

Z', Z"/cm

Z"/cm



-2

-200

-150 Cdl

-100

CL

Rct

-50

fo

100

50

Zw

Rb

fk = 0.082 Hz

150

Rb + Rct

0

Rb

0

-2

0

50

100

150

200

10

250

-1

10

0

10

2

3

10

10

4

5

10

10

f/Hz



Z'/cm 0.06

1

10

(c)

(d)

1.2

-2

fo

0.04

j/mA cm

C', C"/F cm

-2

0.05

0.03 0.02

0.8 0.4 0.0

C"

C'

0.01

-0.4

0.00

-0.8 10

-2

10

-1

10

0

10

1

10 f/Hz

2

10

3

10

4

10

5

0.0

0.4

0.8

1.2

1.6

2.0

E/V

(e) 4

100 50

j/mA cm

-2

2

20 10

0

-2

-4 0.0

0.4

0.8

1.2

1.6

2.0

E/V

Fig. 9

38

(a)

-600

Z"/ cm

2

-500 -400 T (oC) 0 30 40 50 60 70 80

-300 -200 -100 0 0

100

200

300

400

500

600

700

2

Z"/ cm

(b)

140

Csp

120

100

80

Csp/F g-1

60

80 60

ESR

40

ESR, Rct/ cm2

100

40 20 20

Rct

0

0 0

20

40

60

80

T/oC

Fig.10

39

(b)

(a) 2.0

120

1.2 0.8

40 2

100

ESR/ cm

Cd/F g-1

1.6

V/V

50

140

30 80 60

20

40

0.4

10 20

0.0 0

100

200

300

0

400

0

2

4

10

0

(d) 140

100

120 80 100

-250

14

10

80 60

-150

40

8

60

-200

Z"/

12

/%

16

40

-100

Fresh cell at 3000 cycles

-50

at 6000 cycles

20

at 9000 cycles

20

6

8

(c)

24 22 20 18

-1 Cd/F g

Specific power/kW kg

-1

t/s

6 j/A g-1

0

0

50

100

150

200

250

Z'/

8

10

12

14

16

18 20 22 24 -1 Specific energy/Wh kg

0

0 0

2000

4000

6000

8000

10000

Cycle Index

(e)

Fig. 11

40

Table 1: Name and codes of films prepared by various methods Method of preparation Solvent-cast Phase Inversion Solvent Non-solvent

Solvent Acetone DMF (20 ml) DMF (20 ml)

Non-solvent Steam/Water Glycerol (x ml), 'x' = 1, 2, 3, 4, 5

Codes PI PSNS-'x'

41