**Giovanni Emanuele Spina** †**, Federico Poli** †**, Alessandro Brilloni, Daniele Marchese and Francesca Soavi \***

Department of Chemistry "Giacomo Ciamician", Alma Mater Studiorum Universita' di Bologna, Via Selmi 2, 40126 Bologna, Italy; giovanni.spina3@unibo.it (G.E.S.); federico.poli8@unibo.it (F.P.);

alessandro.brilloni2@unibo.it (A.B.); daniele.marchese@studio.unibo.it (D.M.)

**\*** Correspondence: francesca.soavi@unibo.it

† These authors contributed equally.

Received: 11 May 2020; Accepted: 10 June 2020; Published: 16 June 2020

**Abstract:** Water-processable natural polymers represent a valuable alternative for the sustainable manufacturing of electrical double layer capacitors (EDLCs). Here, we demonstrate for the first time the feasibility of the use of pullulan to produce high mass loading electrodes (>10 mg cm−2) at low binder content (10%) for ionic-liquid based EDLCs. Pullulan has also been processed as a porous separator by electrospinning. Its ionic resistance and thermal stability have been evaluated in different electrolytes and were found to be superior compared to those of a cellulose triacetate electrospun separator. Pullulan-ionic liquid EDLCs were, thus, assembled and charged up to 3.2 V. The EDLCs delivered specific energy and power of 7.2 Wh kg−<sup>1</sup> and 3.7 kW kg−<sup>1</sup> and featured good cycling stability over 5000 cycles.

**Keywords:** green supercapacitor; water processable polymer; pullulan; ionic liquid; electrospinning

## **1. Introduction**

Today one of the biggest challenges our society is facing is how to replace the use of fossil energy sources (coal, oil, gas) with renewable ones (solar and wind). The inherent intermittence of the latter sources requires the development of efficient energy storage systems. Among all the possibilities, electrochemical energy storage by secondary batteries and electrical double layer capacitors (EDLCs) is one of the most efficient approach [1–4]. EDLCs are receiving great attention for their unique characteristics of outstanding power and cycle life, that are related to their electrostatic operating mechanism. However specific energies of EDLCs are one order of magnitude lower than that of batteries.

Commercial EDLCs feature activated carbon (AC) electrodes, a porous polymer separator, and an organic electrolyte, typically a solution of alkylammonium salts in acetonitrile or propylene carbonate. The use of the organic electrolyte enables cell voltages as high as 2.5 V [5].

The energy density of EDLCs can be improved by increasing: (i) the operating voltage window, (ii) electrodes specific capacitance, and (iii) the mass loading of the electrodes.

High operating voltage can be achieved by using an electrolyte with a wide electrochemical stability window, like ionic liquids (ILs) or highly concentrated aqueous electrolytes [6–9]. Electrode specific capacitance can be improved by tailoring carbon porosity to the electrolyte, in order to enhance ion access to the carbon surface. An alternative strategy is represented by the use of redox (pseudocapacitive) electrode materials, like metal oxides or electronically conductive polymers, in asymmetric or hybrid supercapacitors. Regarding the third approach, literature provides a very limited number of publications. Achieving mass loading higher than 5–10 mg cm−<sup>2</sup> is considered a great challenge. Indeed, thick electrodes might delaminate from the current collector that is detrimental

for cycling stability. Furthermore, high mass loading may lead to worse ionic and electronic connection between the carbon particles, leading to higher internal resistance [10–12].

Ionic liquids, thanks to their low flammability, represent an even safer alternative to the more volatile acetonitrile solutions. In addition, ILs are known for their high thermal stability, good conductivity and wide electrochemical stability window (>3 V). Despite these interesting properties, they cannot be considered as totally green and strategies to recover them after use are needed. The most investigated ILs for EDLCs are based on the bis(trifluoromethanesulfonyl)imide (TFSI) anion [13–15]. ILs feature bulky ions, therefore in order to promote a high and efficient exploitation of the electron carbon surface of the double layer, the porosity of the carbon has to be properly designed [16,17]. Furthermore, it has been demonstrated that the chemistry of ILs affects the double layer thickness and permittivity, and hence, the electrode capacitance. Indeed, in Ref. [1,2], the capacitive response of different carbon electrodes in *N*-butyl-*N*-methylpyrrolidinium bis(trifluoromethane-sulfonyl)imide (PYR14TFSI), 1-ethyl-3-methylimidazolium bis(trifluoromethanesulfonyl)imide (EmimTFSI) and PYR(2O1)TFSI was compared. In EmimTFSI, all the tested electrodes featured a specific capacitance that was double than that exhibited in the other ILs.

In EDLCs the biggest share of the cost is related to electrodes (28%) and electrolytes (27%). Electrodes are processed by casting slurries made of AC, conductive carbon, binder, and suitable solvents on metal current collectors. The binder material itself does not contribute significantly to the overall cost. However, its chemistry drives the selection of the solvent used for electrode processing, that has a great economic and environmental impact on EDLCs manufacturing [18]. Indeed, nowadays, commercial AC electrodes are mostly fabricated with F-based polymers as binders, such as poly (vinylidene difluoride) (PVdF) which needs N-methyl-2-pyrrolidone (NMP) as solvent/dispersant, both very toxic for humans and environment. This process requires expensive atmosphere-controlled environments [19].

In light of that and to meet the requirements of sustainable and cheaper production processes, much effort is being devoted to the substitution of F-based components with alternative ones. Transition to aqueous electrode preparation by non-toxic binders is expected to provide a great step forward towards an ideally sustainable and environmentally friendly technology for energy storage systems [20,21].

Carboxymethyl cellulose (CMC) represents the state of the art of water-soluble binders [10,21–24]. One of the first attempts of substituting F-based compounds with CMC, was reported by Bonnefoi et al. in 1999 [25]. Two of the major drawbacks in the use of CMC, are: (i) the relatively low achievable electrode mass loading, and (ii) the brittleness shown after the drying step. Winter et al. [23], first proposed Natural Cellulose (NC). While NC is cheaper (0.5–1.5 EUR kg−<sup>1</sup> vs. 1–2 EUR kg<sup>−</sup>1) and more abundant than CMC, it cannot be dissolved in water nor in almost all organic solvents while being soluble in certain ionic liquids [26–29]. Varzi et al. dissolved NC in 1-ethyl-3-methylimidazolium acetate (EmimAc) and demonstrated that NC has enhanced stability at high voltages. An EDLC assembled with electrodes featuring 10% NC binder and a mass loading of ca. 3 mg cm−<sup>2</sup> and PYR14TFSI ionic liquid electrolyte, exhibited a specific capacitance of ca. 13 F g<sup>−</sup><sup>1</sup> at 10 mA cm<sup>−</sup>2, and a capacitance retention of 52%, after cycling for 750 h at 3.7 V [30]. Pursuing the research of even more eco-friendly binders, potato starch, a highly abundant polysaccharide that can be extracted from non-edible potatoes, was also proposed. By the use of this polysaccharide, the production of thick electrodes (240 μm, 9.3 mg cm−2) was demonstrated [18]. These electrodes were used to assemble a 2.5 V-EDLC with 1 M Et4NBF4 in propylene carbonate (PC) electrolyte, that delivered 0.36 F cm−<sup>2</sup> at 10 mA cm<sup>−</sup>2.

Recently, we demonstrated the use of the biodegradable biopolymer pullulan (Pu) as a water processable separator and binder for EDLCs. Specifically, the separator was obtained by electrospinning and the EDLCs featured EmimTFSI electrolyte and pepper seed derived biochar carbon. The EDLC was able to operate at 3.2 V and delivered up to 5 kW kg−<sup>1</sup> specific power and 27.8 Wh kg−<sup>1</sup> specific energy. Its performances were compared with that of conventional electrical double-layer capacitor, with the added value of being eco-friendly and cheap.

Furthermore, the smart combination of the water-soluble, biodegradable Pu with the hydrophobic ionic liquid EmimTFSI, enabled a novel and easy approach for the recovery of EDLC components at the end-of-life. Indeed, the IL and carbon easily separate when immersed in water. The expensive IL can therefore be recollected for a second use [31].

Following these preliminary results, here we report the challenging study that aims to demonstrate the feasibility of the use of Pu to process electrodes at low binder content (10%) and high mass loading (>10 mg cm−2). In the first part of our work we compare the electrochemical response and thermal stability of Pu and cellulose triacetate (CTA) in different electrolytes, namely EmimTFSI, 0.5 m LiTFSI TEGDME, PYR14TFSI. The two natural polymers have been processed by electrospinning and their contribution to ionic resistance of the electrolyte has been investigated by Electrochemical Impedance Spectroscopy (EIS) at different temperatures. On the basis of this investigation, Pu and EmimTFSI were selected to assemble two different EDLCs, one with low electrode mass loading and high binder content (HBLME) and a second one with high mass loading and low binder content (LBHME). The EDLCs have been tested by cyclic voltammetry, EIS and galvanostatic charge/discharge cycles. A deep analysis of the EDLCs performance is reported and discussed to demonstrate that natural polymers and, specifically Pu, may pave the way towards a new approach for a green manufacturing of EDLCs.
