An Electrically Rechargeable Zinc/Air Cell with an Aqueous Choline Acetate Electrolyte

materials

Article An Electrically Rechargeable Zinc/Air Cell with an Aqueous Choline Acetate Electrolyte

Mariappan Sakthivel , Sai Praneet Batchu, Abbas Ali Shah, Kwangmin Kim, Willi Peters and Jean-Francois Drillet * DECHEMA-Forschungsinstitut, Theodor-Heuss-Allee 25, 60486 Frankfurt am Main, Germany; [email protected] (M.S.); [email protected] (S.P.B.); [email protected] (A.A.S.); [email protected] (K.K.); [email protected] (W.P.) * Correspondence: [email protected]; Tel.: + 49-69-7564-476

Received: 20 May 2020; Accepted: 30 June 2020; Published: 3 July 2020

Abstract: Due to the feasibility of an electrically rechargeable zinc/air cell made of a zinc foil as active material, an aqueous choline acetate (ChAcO) mixture as an electrolyte and a spinel MnCo2O4 (MCO) and NiCo2O4 (NCO) as a bi-functional oxygen catalyst was investigated in this work. The 30 wt.% water-containing aqueous ChAcO solution showed high contact angles close to those of KOH favoring triple-phase boundary formation in the gas diffusion electrode. Conductivity and pH 1 value of 30 wt.% H2O/ChAcO amounted to 5.9 mS cm− and 10.8, respectively. Best results in terms of reversible capacity and longevity of zinc/air cell were yielded during 100 h charge/discharge 2 1 with the MnCo2O4 (MCO) air electrode polarization procedure at 100 µA cm− (2.8 mA g− zinc). The corresponding reversible capacity amounted to 25.4 mAh (28% depth of discharge (DOD)) and the energy efficiency ranged from 29–54% during the first and seventh cycle within a 1500 h polarization 2 period. Maximum active material utilization of zinc foil at 100 µA cm− was determined to 38.1 mAh (42% DOD) whereas a further charging step was not possible due to irreversible passivation of the zinc foil surface. A special side-by-side optical cell was used to identify reaction products of the zinc/air system during a single discharge step in aqueous ChAcO that were identified as Zn(OH)2 and ZnO by Raman analysis while no carbonate was detected.

Keywords: zinc/air battery; choline acetate; Bi-functional spinel catalyst; Oxygen reduction reaction (ORR) – oxygen evaluation reaction (OER) activity

1. Introduction The demand for efficient, low-cost, and energy storage systems for portable, mobile, and stationary applications is increasing dramatically. Currently, a traditional system such as lead-acid and Ni-Cd are dominating the market for electricity supply of back-up units, a pallets transporter, energy storage of solar collectors, and brake energy recuperation in trains. Especially in Germany where old nuclear plants will be successively replaced by a biomass/wind/photovoltaic mix, new battery technologies are required for storing surplus electricity produced during windy and/or sunny days and by the way buffer acts for seasonal variations. In past decades, many efforts aimed to 1 develop alkaline zinc/air battery whose theoretical capacity amounts to 820 mAh g− zinc. Battery compounds are relatively cheap, non-toxic, and not prone to thermal runaway reactions as in the case of Li-ion [1]. However, with the exception of the primary zinc/air battery for hearing aids devices, wide introduction of electrically rechargeable systems into the consumer market is hindered by numerous inherent technical drawbacks such as passive layer formation and dendrite growing at the metal electrode [2], electrolyte evaporation and contamination, pore plugging at the gas diffusion electrode (GDE) due to flooding and carbonate formation [3], and insufficient kinetics and stability

Materials 2020, 13, 2975; doi:10.3390/ma13132975 www.mdpi.com/journal/materials Materials 2020, 13, 2975 2 of 21 of the oxygen catalyst for both oxygen reduction (ORR) and evolution (OER) reactions that affect overall energy efficiency (<60%). In Zn-air cell with alkaline electrolyte, the electrode reaction and, during the discharge process and corresponding theoretical potential values, are shown below. Zinc electrode: Zn + 4OH [Zn(OH) ]2 + 2e (E = 1.26 V) − → 4 − − 0 − [Zn(OH) ]2 ZnO + H O + 2OH 4 − → 2 − Air electrode: O + 2H O + 4e 4OH (E = +0.4 V) 2 2 − → − 0 Overall reaction: 2Zn + O 2ZnO (E = +1.66 V) 2 → 0 During the charging step, reverse reactions occur. While deficits linked to zinc electrode and electrolyte appear to be technically solvable, problems related to the air electrode are rather inherent to thermodynamic issues of ORR (strong O=O bond) and OER reactions including hydrogen peroxide side reaction and poor chemical stability in a highly alkaline environment. In order to increase both activity and stability of the reversible air cathode, there are three main strategies that rely on: (i) mixture of ORR and OER catalysts (e.g., Ag + W2C) [4], (ii) hybrid bi-electrodes (e.g., MnO2/NiCo) [5], and (iii) bi-functional catalysts (e.g., La0.6Ca0.4CoO3 (LCCO) perovskite). According to Sintivich et al. [6], the transition-metal oxide in perovskite should have a surface-cation eg occupancy close to unity and high B-site oxygen covalency similarly to Ba0.5Sr0.5Co0.8Fe0.2O3 (BSCF) structure for yielding high OER activity. Bogolowski et al. reported 2 higher activity of BSCF/C65 over 3000 h at 10 mA cm− for both OER and ORR in comparison with LCCO/C65 GDE [7]. Metal oxides with a spinel structure are promising bifunctional oxygen catalysts as well [8]. Pletcher et al. reported on Ni foam-based GDE with a NiCo2O4/ Polytetrafluorethylen (PTFE) active 2 layer that was able to deliver high current densities up to 100 mA cm− [9]. Gebremariam et al. investigated a Zn/air battery based on manganese cobalt oxide nanoparticles anchored on carbon nanofibers and wrapped in a nitrogen-doped carbon shell (MnCo2O4/CNFs@NC) catalyst [10]. They 1 1 yielded a specific capacity of 695 mAh g− Zn and an energy density of 778 Wh kg− Zn at a current density 2 of 20 mA cm− . Some other Co-based systems such as CoMn2O4/N-rGO [11], MnCo mixed oxide [12], and Co(II)1 xCo(0)x/3Mn(III)2x/3S nanoparticles supported on B/N-co-doped mesoporous nanocarbon − NiCo2O4 [13–15], NiCo2O4/NCNT [16], and CoSx@N or S co-doped graphene nanosheets [17] were also tested in the zinc/air cell. Ishihara et al. developed a mesoporous MnCo2O4 with a high surface 2 1 area of 108 m g− that substantially lowered ORR/OER overpotential in 4 M KOH and allowed high reversible capacity values that are more than 250 cycles with a zinc foil [18]. Wang et al. demonstrated high throughput production of Mn (IV) and Co (II)-rich surface MnCo2O4 bifunctional electrode (low ∆E = 0.83 V) by an ultrasonic humidifier aerosol-route at 480 ◦C[19]. To counterpart battery dry-out, substitution of the alkaline solution by a protic ionic liquid (IL) with low vapor pressure seems to be an appealing approach [20]. The other key advantages of customized ILs for electrochemical applications are (i) wide electrochemical potential window up to 2–5 V, (ii) high solubility of metal salts (e.g., chloride), (iii) water mix ability of protic species [21], 1 and (iv) high conductivity compared to non-aqueous solvents (up to ~10 mS cm− )[22]. It should be emphasised that only a minority of ILs possesses all these properties. An extensive survey of alternative electrolytes for zinc-based batteries can be consulted in reference [23]. Moreover, they have to favour both Zn and oxygen reactions and exhibit sufficient surface tension to enable triple-phase boundary formation. In a preliminary screening step, some acetate-based ILs such as EMIMAcO and choline acetate (ChAcO) showed adequate surface tension values and higher contact angle values than 110◦ on Pt/C + 20 wt.% PTFE GDE compared to, e.g., DEMAOTf [24]. Additionally, their good affinity to H2O makes them even more interesting. This work focuses more specifically on the test of aqueous choline acetate (2-hydroxyethyl- trimethylammonium acetate) in combination with bifunctional manganese or nickel cobalt spinel oxygen catalysts that already exhibited easy tunable catalytic properties [25] and high activity in alkaline fuel cell [26]. Choline acetate is promising because of its relatively high ionic conductivity [27], bio-degradability, low-cost production [28], and high affinity to water. Thermophysical investigations of aqueous choline acetate binary solution revealed that strong solute-solvent interactions get stronger Materials 2020, 13, 2975 3 of 21 withMaterials an increase 2020, 13 in, x FOR temperature PEER REVIEW while computational studies confirmed a decrease in solute-solute3 of 21 interactions with an increase in aqueous solvent concentration [29]. The chemical structure of choline acetateconfirmed is shown a decrease in Figure in1. solute-solute interactions with an increase in aqueous solvent concentration [29]. The chemical structure of choline acetate is shown in Figure 1.

Figure 1. Chemical structure of 2-hydroxyethyl-trimethylammonium acetate [(C2H4OH)-(CH3)3N] [acetate] (choline acetate). Figure 1. Chemical structure of 2-hydroxyethyl-trimethylammonium acetate [(C2H4OH)-(CH3)3N] [acetate] (choline acetate). In this work, the behavior of 30 wt.% water-containing choline acetate [(C2H4OH)-(CH3)3N] [Acetate] in terms of contact angle, activity for zinc and air electrode reactions, and its interaction with ambientIn air this will work, be investigated. the behavior Furthermore, of 30 wt.% the water-containing nature of reaction choline products acetate during [(C the2H4 zincOH)-(CH/air cell3)3N] discharge[Acetate] step in will terms be identiﬁedof contact byangle, Raman activity analysis. for zinc and air electrode reactions, and its interaction with ambient air will be investigated. Furthermore, the nature of reaction products during the 2. Materialszinc/air cell and discharge Methods step will be identified by Raman analysis.

2.1.2. Hydrothermal Materials and Synthesis Methods of MnCo 2O4 (MCO) and NiCo2O4 (NCO)

A total of 25 mL of 50 mM manganese(II) nitrate hydrate (Mn(NO3)2 4H2O, 99.99%, Sigma-Aldrich, 2.1. Hydrothermal Synthesis of MnCo2O4 (MCO) and NiCo2O4 (NCO) · Taufkirchen, Germany) or nickel(II) nitrate hexahydrate (Ni(NO ) 6H O, 99.999% Sigma-Aldrich, 3 2· 2 Taufkirchen,A total Germany) of 25 mL were of mixed50 mM with manganese(II) 25 mL of 100nitrate mM hydrate cobalt(II) (Mn(NO nitrate3)2 hexahydrate∙4H2O, 99.99%, (Co(NOSigma-Aldrich,) 6H O, 99.999%, Taufkirchen, Sigma-Aldrich, Germany) Taufkirchen, or nickel(II) Germany) nitrate hexahydrate at 250 rpm for(Ni(NO 30 min,3)2∙6H as2 reportedO, 99.999% 3 2· 2 elsewhereSigma-Aldrich, [25]. Subsequently, Taufkirchen, 5.2 mLGermany) of 1.25 Mwere NaOH mixed solution with was 25 addedmL of dropwise 100 mM into cobalt(II) the reaction nitrate mixturehexahydrate under vigorous (Co(NO3 stirring)2∙6H2O, at 99.999%, 500 rpm Sigma-Aldrich, for 30 min until Taufkirchen, the color of theGermany) solution at turned 250 rpm brown for 30 or darkmin, greenas reported in case elsewhere of Mn(NO [25]) 4H. Subsequently,O and Ni(NO 5.2) 6HmL Oof precursor, 1.25 M NaOH respectively. solution Afterward, was added 3 2· 2 3 2· 2 thedropwise mixture was into transferred the reaction into mixture 60 mL under vials and vigorous the hydrothermal stirring at 500 process rpm wasfor 30 carried min until out atthe 160 color◦C of forthe 10 hsolution in a drying turned oven brown (FD 23,or dark Binder, green Tuttlingen, in case of Germany). Mn(NO3)2∙ The4H2O black-colored and Ni(NO3 hydroxide)2∙6H2O precursor, was collectedrespectively. by centrifugation Afterward, (3K30, the mixture Sigma, was Germany) transferr ated 10,000 into 60 rpm mL for vials 3 min, and washed the hydrothermal with deionized process waterwas until carried attaining out at a160 pH °C value for of10 7,h andin a storeddrying overnightoven (FD in23, a Binder, freeze-drying Tuttlingen, unit (L06,Germany). Dieter The Piatkowski-Forschungsgeräte,black-colored hydroxide was Petershausen, collected by centrifugation Germany) at (3K30,50 C Sigma, and 50 Germany) mbar. Then, at 10,000 the powder rpm for 3 − ◦ wasmin, grounded washed manually with deionized in a mortar water and calcinateduntil attainin in ag tubulara pH value oven (RHTCof 7, and 80-710 stored/15, Nabertherm,overnight in a Lilienthal,freeze-drying Germany) unit at(L06, 400 Dieter◦C for Piatkowski-F 1 h. orschungsgeräte, Petershausen, Germany) at −50 °C and 50 mbar. Then, the powder was grounded manually in a mortar and calcinated in a tubular oven 2.2.(RHTC Fabrication 80-710/15, of Gas DiNabertherm,ﬀusion Electrode Lilienthal, (GDE) Germany) at 400 °C for 1 h. A 36 cm2 sized 20 wt.% PTFE-coated Toray carbon paper (TP-060T, QuinTech, Göppingen, Germany)2.2. Fabrication was used of Gas as aDiffusion gas diﬀ usionElectrode layer (GDE) (GDL). The catalyst ink was prepared by successively dispersingA 36 as-prepared cm² sized powder 20 wt.% catalyst, PTFE-coated C65 carbon Toray (C-NERGY, carbon paper TIMCAL, (TP-060T, Bironico, QuinTech, Switzerland), Göppingen, and PTFEGermany) binder (60:20:20 was used wt.%) as a in gas 2 mL diffusion of a 1:2 layer weight (GDL). ratio water:isoproponalThe catalyst ink was mixture, prepared ultrasonicated by successively for 2 min,dispersing and then as-prepared sprayed on powder GDL. The catalyst, coated C electrodes65 carbon (C-NERGY, were sintered TIMCAL, at 80 ◦C Bironico, in air between Switzerland), each 2 sprayingand PTFE step forbinder 3 min (60:20:20 and for 1wt.%) h after in reaching 2 mL theof a overall 1:2 weight catalyst ratio loading water:isoproponal of 1 mg cm− . Amixture, disk 2 electrodeultrasonicated of 2.54 cm forgeometrical 2 min, and areathen wassprayed coined on from GDL. a GDEThe coated sheet for electrodes electrochemical were sintered measurements at 80 °C in in Plexiglasair between and each El-Cell spraying cells. step for 3 min and for 1 h after reaching the overall catalyst loading of 1 mg cm−2. A disk electrode of 2.54 cm2 geometrical area was coined from a GDE sheet for 2.3. Preparation and Characterization of Aqueous Ionic Liquid Electrolyte electrochemical measurements in Plexiglas and El-Cell cells. For ZnAcO-containing electrolyte preparation, 1.97 g ChAcO (C7H17NO3, 98%, IoLiTec, Heilbronn, Germany),2.3. Preparation 30 wt.% and H2O Characterization related to choline of Aqueous acetate massIonic Liquid and 3.32 Electrolyte mg Zn(AcO) 2 (99.99%, Sigma-Aldrich,

For ZnAcO-containing electrolyte preparation, 1.97 g ChAcO (C7H17NO3, 98%, IoLiTec, Heilbronn, Germany), 30 wt.% H2O related to choline acetate mass and 3.32 mg Zn(AcO)2 (99.99%, Sigma-Aldrich, Taufkirchen, Germany) were mixed in a glass vessel under ambient atmosphere and stirred for 30 min. In order to determine conductivity and pH value of different H2O:ChAcO mass

Materials 2020, 13, 2975 4 of 21

Taufkirchen, Germany) were mixed in a glass vessel under ambient atmosphere and stirred for 30 min. In order to determine conductivity and pH value of diﬀerent H2O:ChAcO mass ratios, water was added stepwise to as-received choline acetate. A large electrolyte sample of at least 20 g was used for each measurement to reduce the weighing error. Due to the highly hygroscopic nature of ChAcO, the samples were purged with argon and kept closed during pH (GMH 5340/GE117 Greisinger, Regenstauf, Germany) and conductivity (LF40 in combination with LTC0,35/23 electrode, Meinsberg, Waldheim, Germany) measurements. The solution density was determined by using a simple narrow necked volumetric ﬂask (20 mL).

2.4. Structural Characterization Techniques The X-ray diffraction patterns were collected with a diffractometer (D8 Advance, Bruker, Karlsruhe, Germany) using nickel filtered Cu K-alpha source with λ = 0.154 nm. Catalyst morphology and composition were investigated by scanning electron microscopy (SEM)/energy dispersive x-ray spectroscopy (EDX) (XL 40, Philips, Eindhoven, The Netherlands) and transmission electron microscopy (TEM) (EM 420, Philips, Eindhoven, The Netherlands). The GDE/electrolyte drop interface was studied by contact angle goniometer (OCA 15Pro, Dataphysics, Filderstadt, Germany) in ambient air at 23 ◦C. In order to study the reaction products during the zinc/air discharge step, a side-by-side optical cell (ECC-opto-SBS, El-Cell, Hamburg, Germany) and a confocal Raman microscope (InVia Reflex, Renishaw, Kingswood, UK) were employed. The laser beam (Ar, 532 nm, 2 mW) was focused through a 50 objective lens (DM 2500, Leica, Mannheim, Germany) with ~1 mm spot size. A high-resolution × 1 diffraction grating density of 1800 grooves mm− was employed. A single spectrum consisted of 20 accumulated scans with an acquisition time of 20 s each. The background signal was subtracted according to the baseline correction mode.

2.5. Half-Cell Measurements

2.5.1. Catalyst Activity for ORR/OER Reactions All half-cell experiments were carried out with a potentiostat/galvanostat (IM6eX, Zahner, Kronach, Germany). Tests of bifunctional GDE (2.54 cm2) for OER/ORR were performed in a Poly ether ether ketone (PEEK) cell (FlexCell, Gaskatel, Kassel, Germany) equipped with a Pt-coated Ti sheet as counter and a reversible hydrogen electrode (RHE) as a reference electrode. Two experimental protocols were applied with respect to electrolyte used in order to avoid any deterioration of the catalyst structure in 7 M KOH. Especially during a cathodic scan where molecular oxygen is produced, a 1 2 maximal potential scanning rate and current density value was set to 5 mV s− and 100 mA cm− , respectively. For investigation in ionic liquid, the maximum current density was fixed to 10 mA 2 cm− . All experiments were carried out with synthetic dry air (79.5% N2 + 20.5% O2, Linde GmbH, Offenbach, Germany) at 10 mbar overpressure. Residual moisture was removed from the synthetic air feed by connecting a gas drying unit (Drierite™) filled with CaSO4 pellets (Sigma-Aldrich, Taufkirchen, Germany). A long-term stability protocol consisted of 2.0 h for OER (charging) and ORR (discharging) 2 step at 100 µA cm− each with a 30 min interruption at open circuit voltage (OCV) in-between.

2.5.2. Zn Redox Reaction The zinc redox reactions were investigated in an El-Cell (ECCair, El-Cell, Hamburg, Germany) laboratory cell. A zinc foil (Ø = 18 mm, 50 µm in thickness, 90 mg, 99.999%, Chempur, Karlsruhe, Germany) was used as a substrate, a Pt/C GDE (EC-20-10-7P, Ø = 18 mm, QuinTech, Göppingen, Germany) was used as a counter, a zinc wire (Ø = 1 mm, 99.999%, Chempur, Karlsruhe, Germany) was used as a pseudo-reference electrode, and a glass ﬁber separator (Ø = 18 mm,1.55 mm in thickness, El-Cell, Hamburg, Germany) was used as a separator. In 7 M KOH, the potential window was set between 400 and +400 mV while, in ChAcO, it was extended from 400 to +900 mV. The potential − − loss (IR-drop) was determined by electrochemical impedance spectroscopy at open circuit voltage Materials 2020, 13, 2975 5 of 21

(OCV) and corrected in half cell electrochemical data analysis. All potential values were normalized to the normal hydrogen electrode (NHE).

2.6. Full-Cell Measurements

2.6.1. Current-Voltage and Charge/Discharge Characteristics in the El-Cell The zinc/air battery test measurements were performed with the help of a battery cycler (BCS 10, Biologic Scientific Instruments/Gamec Analysentechnik, Illingen, Germany). A schematic diagram of the three-electrode El-Cell design is provided in Figure S1a. The El-Cell was assembled by successive pile-up of zinc foil, borosilicate glass fiber separator filled with 300 µL of alkaline or IL electrolyte, and GDE. The use of a zinc foil in spite of state-of-the-art zinc powder was motivated by easily handling during cell fabrication. A commercial 40 wt.% Pt/C benchmark GDE (EC-20-10-7P, QuinTech, Göppingen Germany) was tested for comparison. Before the test, the cell was purged with synthetic dry air for 6 h. The cell performance was evaluated first with the voltage/current (U/I) protocol 2 by applying constant current densities in the range of 1–100 µA cm− for 15 min charge/discharge duration and 15 min OCV at each step (1 h/cycle) under either dry or ambient air conditions. Due to the poor conductivity of ionic liquid compared to that of alkaline electrolyte, the OCV step was introduced between charge/discharge steps in full cell experiments to observe the transition period to an equilibrium state. After initial characterization, the cell was subjected to a cyclic stability procedure 2 at 100 µA cm− charge/discharge step for 8 h over one month (26 h/cycle) under ambient air condition (30% relative humidity (RH) at 23 ◦C). Lastly, best performing catalyst systems were investigated at 2 100 µA cm− for 100 h charge/discharge step (10 days/cycle) in an ambient air condition (30% RH at 23 ◦C).

2.6.2. Charge/Discharge Characteristic in the Coin Cell The coin cell (CR-2032) was fabricated by assembling a zinc foil (Ø = 16 mm, 0.05 mm thickness, 70 mg, 99.99%, Goodfellow, Friedberg, Germany), 2 glass fiber separators (Ø = 18 mm, 1 mm thickness, Whatman) filled with 250 µL of ChAcO + 30 wt.% H2O + 0.01 M ZnAcO electrolyte, and an as-prepared GDE (Ø = 18 mm) with a crimper (MSK-PN110-S, MTI, Richmond, VA, USA) system at an air pressure of 6 bar for 5 s. The GDE surface was ventilated through four pre-drilled holes (Ø = 1 mm) in the cover case (see Figure S1b). The polarization profile consisted of 5 h charge/discharge steps and 2 h OCV (12 h/cycle) under ambient air conditions (30% RH at 23 ◦C).

2.7. In-Situ Raman Investigation During Zn/Air Discharge Step in IL First, a zinc electrode with a well-defined state-of-charge (SOC) was designed by electrodeposition of 15 mg zinc from ChAcO + 30 wt.% H2O + 0.1 M ZnAcO solution at 1 mA for 12 h (12 mAh at 100% current efficiency) on a PTFE-free gas diffusion layer (TGPH90, Toray, QuinTech, Göppingen, Germany). For in-situ Raman experiment, a commercially available side-by-side (SBS) optical cell (El-Cell) was used. Hereby, both electrodes (10 mm2) and the glass fibre separator (1 mm) are aligned in the same plane (see Section 3.4.). Then, 50 µL of ChAcO + 30 wt.% H2O solution was filled in a manner that no air bubbles arise between the glass and electrode surface. Air circulation was provided by small holes in the cell body.

3. Results and Discussion

3.1. Physicochemical Characterization of Catalyst Powder

Figure2 shows X-ray di ffraction (XRD) patterns of as-prepared MnCo2O4 (MCO) and NiCo2O4 (NCO) catalyst powder and their respective standard. The diffraction spectra of the two as-prepared powder materials match perfectly the cubic structure of AB2O4 (A = Mn, Ni and B = Co) with the Fd3m space group (JCPDS 00-023-1237 (MnCo2O4) and JCPDS 00-073-1702 (NiCo2O4), where A-site atoms are Materials 2020, 13, x FOR PEER REVIEW 6 of 21 from hydrothermal [30] and MCO from sol-gel and flame spray pyrolysis methods [31]. Since the XRD pattern of pure Ni is also located at 44.5° [JCPDS 04-0850], we cannot definitely exclude the presence of a residual Ni phase in an as-prepared NCO sample. Figure 3 shows TEM images of MCO and NCO powder particles synthesized by the Materials 2020, 13, 2975 6 of 21 hydrothermal route after a calcination step at 400 °C (see SEM image and EDX pattern in Figure S2). While MCO particles exhibit a relatively small spherical shape (30–50 nm) [32], NCO material is madelocated of in a octahedralrather large and 2D B hexagonal atoms occupy [33] both (200 octahedralnm) flake andstructure tetrahedral that consists sites. No of additional smaller primary foreign particlespeaks were in the detected range inof the8 to XRD 26 nm. pattern These of latter MCO ar ande in NCO a similar indicating range theas carbon-supported high purity of the catalystPt ones (10preparation nm). BET-specific by a hydrothermal surface area route, and whichaverage is pore followed size byof as-synthesized a calcination step spinel at 400 powders◦C. The is collectedlisted in TableXRD spectra1 and their are inrespective good agreement isotherm with curves those and from pore literature size distribution such as NCO plotsfrom are provided hydrothermal in Figure [30] S3.and Especially, MCO from the sol-gel BET andvalue flame of MCO spray (120 pyrolysis m² g−1) methodsis substantially [31]. Since higher the than XRD those pattern reported of pure in Ni the is literaturealso located obtained at 44.5◦ from[JCPDS pyro 04-0850], and sol-gel we cannot (14 and definitely 3 m² g−1) exclude [31] routes the presenceas well as of from a residual hydrothermal Ni phase −1 −1 (59in an m²Materials as-prepared g ) [34]2020, 13and, x NCOFOR sono-chemical PEER sample. REVIEW (59 m² g ) [35] methods. 6 of 21

from hydrothermal [30] and MCO from sol-gel and flame spray pyrolysis methods [31]. Since the XRD pattern of pure Ni is also located at 44.5° [JCPDS 04-0850], we cannot definitely exclude the presence of a residual Ni phase in an as-prepared NCO sample. Figure 3 shows TEM images of MCO and NCO powder particles synthesized by the hydrothermal route after a calcination step at 400 °C (see SEM image and EDX pattern in Figure S2). While MCO particles exhibit a relatively small spherical shape (30–50 nm) [32], NCO material is made of a rather large 2D hexagonal [33] (200 nm) flake structure that consists of smaller primary particles in the range of 8 to 26 nm. These latter are in a similar range as carbon-supported Pt ones (10 nm). BET-specific surface area and average pore size of as-synthesized spinel powders is listed in Table 1 and their respective isotherm curves and pore size distribution plots are provided in Figure S3. Especially, the BET value of MCO (120 m² g−1) is substantially higher than those reported in the literature obtained from pyro and sol-gel (14 and 3 m² g−1) [31] routes as well as from hydrothermal (59 m² g−1) [34] and sono-chemical (59 m² g−1) [35] methods.

Figure 2. XRD patterns of as-prepared NiCo O and MnCo O as well as Pt/C powder from a Figure 2. XRD patterns of as-prepared NiCo22O44 and MnCo2O2 4 4as well as Pt/C powder from a commercialcommercial GDE. Figure3 shows TEM images of MCO and NCO powder particles synthesized by the hydrothermal route after a calcination step at 400 ◦C (see SEM image and EDX pattern in Figure S2). While MCO particles exhibit a relatively small spherical shape (30–50 nm) [32], NCO material is made of a rather large 2D hexagonal [33] (200 nm) ﬂake structure that consists of smaller primary particles in the range of 8 to 26 nm. These latter are in a similar range as carbon-supported Pt ones (10 nm). BET-speciﬁc surface area and average pore size of as-synthesized spinel powders is listed in Table1 and their respective isotherm curves and pore size distribution plots are provided in Figure S3. Especially, 2 1 the BET value of MCO (120 m g− ) is substantially higher than those reported in the literature obtained 2 1 2 1 from pyro and sol-gel (14 and 3 m g− )[31] routes as well as from hydrothermal (59 m g− )[34] and Figure 2. XRD patterns of as-prepared NiCo2O4 and MnCo2O4 as well as Pt/C powder from a (a) 2 1 (b) (c) sono-chemicalcommercial (59 GDE. m g − )[35] methods. Figure 3. TEM images of as-prepared (a) MnCo2O4 (MCO), (b) NiCo2O4 (NCO), and (c) as-received Pt/C from commercial EC20 GDE.

Table 1. Physical properties of bifunctional catalysts.

BET Pore Size + D # CA * CA ** Catalyst m²∙g−1 nm nm ° °

MCO/C65 126 17 26 120 130

(a) (b) (c)

Figure 3. TEM images of as-prepared (a) MnCo2O4 (MCO), (b) NiCo2O4 (NCO), and (c) as-received Figure 3. TEM images of as-prepared (a) MnCo2O4 (MCO), (b) NiCo2O4 (NCO), and (c) as-received Pt/C from commercial EC20 GDE. Pt/C from commercial EC20 GDE.

Table 1. Physical properties of bifunctional catalysts.

BET Pore Size + D # CA * CA ** Catalyst m²∙g−1 nm nm ° °

MCO/C65 126 17 26 120 130

Materials 2020, 13, 2975 7 of 21

Table 1. Physical properties of bifunctional catalysts. Materials 2020, 13, x FOR PEER REVIEW 7 of 21 BET Pore Size + D # CA * CA ** Catalyst 2 1 m g− nm nm ◦ ◦ NCO/C65 108 3.8 100 111 117 MCO/C65 126 17 26 120 130 Pt/C NCO55 /C65 108 36 3.8 10 100 111141 117 149 + average pore diameter,Pt/C # average particle 55 diameter 36, * contact 10 angle 141of IL on GDE 149 surface measured at+ averageroom temperature pore diameter, (RT)# average for 30 particlemin, ** diameter,contact an *gle contact of 7 angleM KOH of IL on on GDE GDE surface measuredmeasured at at room RT temperature (RT) for 30 min, ** contact angle of 7 M KOH on GDE surface measured at RT for 30 min. All contact for 30 min. All contact angle (CA) values are given with an error bar of ± 1°. angle (CA) values are given with an error bar of 1 . ± ◦ Figure 4 shows SEM images of the spinel GDE surface structures fabricated by the spray coating method.Figure All4 samples shows SEM contain images an identical of the spinel catalyst GDE loading surface of structures 1 mg cm− fabricated2 that was bymixed the spraywith 20 coating wt.% 2 carbonmethod. black All samplesC65 and contain20 wt.% an PTFE identical binder. catalyst For comparison, loading of 1 mga commercial cm− that was Pt GDE mixed with with identical 20 wt.% compositioncarbon black was C65 andconsidered 20 wt.% (Figure PTFE binder. 4). While For th comparison,e carbon fiber a commercial network from Pt GDE the withToray identical paper structurecomposition is still was visible considered on MCO (Figure and4). NCO While spinel the carbon GDEs, fiber the gas network diffusion from layer the Toray (GDL) paper surface structure of Pt GDEis still is, visible with onexception MCO and of NCOcrevasse spinel regions, GDEs, completely the gas diff coveredusion layer by (GDL)the reaction surface layer. of Pt GDEThis is,is principallywith exception due ofto crevassethe smaller regions, size and completely lower density covered of by Pt the particles reaction compared layer. This to isspinel principally systems due that to werethe smaller mixed sizewith and a smaller lower density amount of of Pt carbon particles particles compared and to consequently spinel systems led that to werea thinner mixed reaction with a layer.smaller To amount some extent, of carbon the particlesdifference and in density consequently of carbon led tomaterials a thinner may reaction also play layer. a Torole. some However, extent, informationthe difference about in density density of carbonof the materialscommercial may carbon also play product a role. as However, well as GDL information manufacturer about density is not mentionedof the commercial on its data carbon sheet. product The cracks as well visible as GDL in the manufacturer commercial Pt/C is not reaction mentioned layer on are its typical data sheet.for 1 −2 2 mgThePt crackscm loaded visible GDE. in the commercial Pt/C reaction layer are typical for 1 mgPt cm− loaded GDE.

1 Figure 4. SEM images ofof didifferentfferent 11 mg mgcatalyst gg−1 GDEGDE surfaces surfaces at at (a (a)) 50× 50 andand (b (b) )1000× 1000 magnification.magnification. catalyst − × × TheThe contact angle measurementsmeasurements werewere carried carried out out in in order order to to investigate investigate the the wetting wetting behaviour behaviour of ofcholine choline acetate acetate and and KOH KOH on on di ffdifferenterent GDE GDE structures structures that that strongly strongly depends depends on on the the surface surface tension tension of ofboth both liquid liquid and and solid solid phase phase [36] as[36] well as aswell GDE as morphology,GDE morphology, porosity, porosity, and hydrophobicity. and hydrophobicity. Figure5 Figureshows the5 shows shape the of KOHshape andof KOH water-containing and water-contai ChAcOning droplets ChAcO afterdroplets 30 min after on 30 Pt /minC and on spinel Pt/C /andC65 spinel/Celectrode65 surfaces.electrode While surfaces. both While KOH both and ILKOH droplets and IL form droplets a contact form angle a contact higher angle than higher 140◦ with than a 140° Pt/C withsurface, a Pt/C some surface, discrepancies some discrepancies in terms of in droplet terms formof droplet and contactform and angle contact value angle are visiblevalue are on visible spinel onmaterials. spinel materials. It appears It thatappears MCO-based that MCO-based GDE is a-prioriGDE is a-priori more adequate more adequate for forming for forming slightly slightly higher highercontact contact angle values.angle values. However, However, large large pores pores of about of about 10–20 10–20µm in µm spinel in spinel GDL GDL in comparison in comparison to a tomaximum a maximum of 5 µofm 5 in µm case in of case commercial of commercial GDE observed GDE observed on TEM on images TEM appearsimages toappears be a handicap. to be a handicap.Some additional Some additional experiments experiments (not shown (not here) show weren here) performed were performed with choline with acetate choline for upacetate to 100 for h upin whichto 100 noh in substantial which no changesubstantial in droplet change shape in droplet and anglesshape and was angles observed. was Thisobserved. was a This promising was a promisingprerequisite prerequisite for forming for a triple-phaseforming a triple-phase boundary [boundary37]. [37].

Materials 2020, 13, x 2975 FOR PEER REVIEW 8 of 21 Materials 2020, 13, x FOR PEER REVIEW 8 of 21

. . µ Figure 5. PhotosPhotos of contact angle (CA) of 5 µLL dropletsdroplets ofof ((aa)) ChAcOChAcO+ + 3030 wt.%wt.% HH22O + 0.010.01 M ZnAcO Figure 5. Photos of contact angle (CA) of 5 µL droplets of (a) ChAcO + 30 wt.% H2O + 0.01 M ZnAcO and ( b) 7 M KOH after 30 min on didifferentﬀerent GDE substrates. and (b) 7 M KOH after 30 min on different GDE substrates. Choline acetate is a highly hygroscopic salt that forms with an appropriate amount of water and Choline acetate is a highly hygroscopic salt that forms with an appropriate amount of water and an aqueous ionicionic liquidliquid even even in in humid humid air. air. Below Below 10 wt.%10 wt.% water, water, however, however, the mixture the mixture is rather is rather a milky a an aqueous ionic liquid even in humid air. Below 10 wt.% water, however, the mixture is rather a milkymelt with melt high with viscosity high viscosity and low and ionic low conductivity. ionic conductivity. In the range In the of 20–70range wt.%, of 20–70 a linear wt.%, correlation a linear milky melt with high viscosity and low ionic conductivity. In the range of 20–70 wt.%, a linear1 correlationbetween water between content water and content conductivity and conductivity can be seen incan Figure be seen6, which in Figure culminates 6, which at culminates 24.9 mS cm at− 24.9for correlation between water content and conductivity can be seen in Figure 6, which culminates at 24.9 mSthe cm 88%−1 waterfor the to 88% a ChAcO water to mass a ChAcO ratio (47% mass water ratio to(47% electrolyte water to mass electrolyte ratio). mass ratio). mS cm−1 for the 88% water to a ChAcO mass ratio (47% water to electrolyte mass ratio).

Figure 6. Ionic conductivity and pH values in dependency of water/ChAcO mass ratio at 20 C. Figure 6. Ionic conductivity and pH values in dependency of water/ChAcO mass ratio at 20 °C. The◦ FigureThe corresponding 6. Ionic conductivity conductivity and andpH valu pH ases wellin dependency as density valuesof water/ areChAcO listed in mass Table ratio S1. at 20 °C. The corresponding conductivity and pH as well as density values are listed in Table S1. corresponding conductivity and pH as well as density values are listed in Table S1. The dependence of pH on water content shows a reverse trend. For 10 wt.% ratio, a pH value The dependence of pH on water content shows a reverse trend. For 10 wt.% ratio, a pH value closeThe to 12 dependence was measured. of pH Between on water 20 andcontent 50 wt.%, shows a linear a reverse decline trend. of pH For can 10 be wt.% observed, ratio, a whereupon pH value close to 12 was measured. Between 20 and 50 wt.%, a linear decline of pH can be observed, closea plateau to 12 evolves was measured. in proximity Between to a pH of20 8. and Thus, 50 thewt aqueous.%, a linear IL electrolyte decline of exhibits pH can alkaline be observed, behavior whereupon a plateau evolves in proximity to a pH of 8. Thus, the aqueous IL electrolyte exhibits whereuponwithin a broad a plateau range ofevolves water/ ChAcOin proximity ratios. to The a pH conductivity of 8. Thus, and the pH aqueous value ofIL theelectrolyte 30 wt.% exhibits mixture alkaline behavior within a broad range of water/ChAcO1 ratios. The conductivity and pH value of the alkalinewas determined behavior by within slight a extrapolationbroad range of to water/Ch 5.9 mS cmAcO− and ratios. 10.8, The respectively. conductivity and pH value of the 30 wt.% mixture was determined by slight extrapolation to 5.9 mS cm−1 and 10.8, respectively. 30 wt.% mixture was determined by slight extrapolation to 5.9 mS cm−1 and 10.8, respectively. 3.2. Half-Cell Investigations 3.2. Half-Cell Investigations 3.2. Half-Cell Investigations 3.2.1. Activity of MCO and NCO Spinel GDEs for ORR/OER in 7 M KOH with Synthetic Dry Air 3.2.1. Activity of MCO and NCO Spinel GDEs for ORR/OER in 7 M KOH with Synthetic Dry Air 3.2.1.The Activity activity of MCO of diﬀ anderent NCO spinel Spinel for ORR GDEs/OER for reactions ORR/OER was in investigated7 M KOH with ﬁrst Synthetic in 7 M KOH Dry alkalineAir electrolyteThe activity by cyclic of voltammogramdifferent spinel (CV)for ORR/OER and compared reactions with thatwas ofinvestigated Pt (see Figure first7). in While 7 M spinel KOH The activity of different spinel for ORR/OER reactions was investigated first in 7 M KOH alkalinecatalysts electrolyte exhibited by similar cyclic activity voltammogram for OER reactions(CV) and thatcompared are at leastwith aboutthat of 300Pt (see mV Figure lower than7). While that alkaline electrolyte by cyclic voltammogram (CV) and compared with that of Pt (see Figure 7). While spinelobserved catalysts at Pt at exhibited 50 mA cm similar2, an activity inverse for trend OER is observedreactions duringthat are ORR. at least At about the same 300 currentmV lower density, than spinel catalysts exhibited similar− activity for OER reactions that are at least about 300 mV lower than thata huge observed potential at shiftPt at of50 about mA cm to− 6002, an mV inverse to more trend negative is observed values during occurred ORR. with At spinel the same GDEs current when that observed at Pt at 50 mA cm−2, an inverse trend is observed during ORR. At the same current density, a huge potential shift of about to 600 mV to more negative values occurred with spinel density, a huge potential shift of about to 600 mV to more negative values occurred with spinel

Materials 2020, 13, x FOR PEER REVIEW 9 of 21

GDEs when compared to the Pt one. Since particle size amounts to about 10 nm for commercial Pt and 10–100 nm for as-prepared NCO, a definitive conclusion about the comparative intrinsic activity Materialsof both2020 systems, 13, 2975 is not straightforward. Emerging of additional peaks in voltammogram of Ni-based9 of 21 spinel within the potential range between 1.2 and 1.4 V (inlet in Figure 7a) is attributed to Ni2+/Ni3+ comparedredox couple to the and Pt one.an indication Since particle for sizeNi segregat amountsion to aboutfrom the 10 nmNCO for spinel commercial structure. Pt and A 10–100similar nm phenomenon was observed at LaNiO3 perovskite and is reported in reference [38]. for as-prepared NCO, a deﬁnitive conclusion about the comparative intrinsic activity of both systems In Figure 7b, potential/current behaviour of MCO/C65 and NCO/C65 containing GDEs is plotted is not straightforward. Emerging of additional peaks in voltammogram of Ni-based spinel within the in function of mass-normalized current density for OER and ORR reactions in a semi-logarithmic potential range between 1.2 and 1.4 V (inlet in Figure7a) is attributed to Ni 2+/Ni3+ redox couple and diagram and compared to that of Pt/C. In the Tafel plot of spinel systems, one can distinguish two an indication for Ni segregation from the NCO spinel structure. A similar phenomenon was observed main domains with a transition region close to 10 mA cm−2 for both OER and ORR reactions. All atspinel LaNiO samples3 perovskite exhibit and very is similar reported characteristics. in reference [38].

(a) (b)

−2 2 FigureFigure 7. 7. ((a)) CV of of 1 1 mg mgcatalystcatalyst cmcm as-prepared− as-prepared MCO MCO and NCO and based NCO basedGDEs and GDEs commercial and commercial Pt/C −1 1 PtGDE/C GDE at 5 at mV 5 mV s sin− Oin2-saturated O2-saturated 7 M 7 MKOH. KOH. (b) (bCorresponding) Corresponding semi-logarithmic semi-logarithmic Tafel Tafel plots plots in in ORRORR/OER/OER regions. regions.

InThe Figure Tafel7b, slopes potential are listed/current in Table behaviour 2 where of the MCO first/C value65 and is NCOassigned/C65 tocontaining 1 to 10 and GDEs second is plottedone infrom function 10 to of100 mass-normalized mA cm−2 current currentdensity regions, density respectively. for OER and These ORR results reactions are inin agood semi-logarithmic agreement diagramwith Wang and comparedet al. [39] work to that in ofwhich Pt/C. binder-free In the Tafel bifunctional plot of spinel NiCo systems,2O4@NiO one and can distinguishNiFe/NiCo2O two4/NF main were tested in 1.0 M KOH electrolyte [40]. Most relevant2 data related to the activity of different domains with a transition region close to 10 mA cm− for both OER and ORR reactions. All spinel samplescatalysts exhibit for ORR/OER very similar such characteristics.as mass activity (MA), Tafel slope, and ΔE are summarized in Table 2. The Tafel slopes are listed in Table2 where the ﬁrst value is assigned to 1 to 10 and second one Table 2. Summary2 of GDE electrochemical properties from experiments shown in Figure 7. from 10 to 100 mA cm− current density regions, respectively. These results are in good agreement with

WangCatalyst et al. [39] workMA inORR which binder-freeMAOER bifunctional Tafel NiCoORR 2O4@NiOTafel andOER NiFe /NiCo2OΔ4E/ NF were tested in 1.0 M KOH electrolyte [40]. Most relevant data related to the activity of diﬀerent catalysts for at 0.6 V at 1.65 V mV dec−1 mV dec−1 at 10 mA cm−² ORR/OER such as mass activity (MA), Tafel slope, and ∆E are summarized in Table2. mA mg−2 mA mg−2 V

MCO/C65Table 2. Summary15 of GDE electrochemical 47 properties70–185 from # experiments77–113 shown # in Figure0.97.

NCO/C65 16 74 84–186 # 30–70 # 0.86 MAORR MAOER ∆E TafelORR + TafelOER + 2 Pt/C Catalyst 98 *at 0.6 V at 1.651.5 V 60–1501 1178–330at 10 mA cm− 0.86 2 2 mV dec− mV dec− * measured at 0.95 V.mA # Tafel mg slope− formA ORR mg and− OER was calculated within 0.75–0.9 V andV 1.55–1.65 V + # # potentialMCO window,/C65 respectively.15 Tafel slope 47 for ORR70–185 and OER was77–113 calculated within 0.95–1.050.9 V and # # 1.8–2.0 VNCO potential/C65 window,16 respectively. 74 84–186 30–70 0.86 Pt/C 98 * 1.5 60–150 + 178–330 + 0.86 3.2.2.* measured Activity at of 0.95 MCO V. # Tafelfor OER/ORR slope for ORR in ChAcO and OER and was KOH calculated in Dry within Air 0.75–0.9 V and 1.55–1.65 V potential window, respectively. + Tafel slope for ORR and OER was calculated within 0.95–1.05 V and 1.8–2.0 V potential window,Since respectively.half-cell measurements of spinel in oxygen-saturated choline acetate were not easy to interpret because of the apparition of numerous peaks and NCO catalyst was not stable enough, we 3.2.2. Activity of MCO for OER/ORR in ChAcO and KOH in Dry Air Since half-cell measurements of spinel in oxygen-saturated choline acetate were not easy to interpret because of the apparition of numerous peaks and NCO catalyst was not stable enough, we decided to show only those related to the MCO system for better clarity. Figure8a shows Materials 2020 13 Materials 2020,, 13,, x 2975 FOR PEER REVIEW 1010 of of 21 21 decided to show only those related to the MCO system for better clarity. Figure 8a shows activity of activity of an MCO gas diﬀusion electrode (GDE) for oxygen evolution (OER) and reduction (ORR) in an MCO gas diffusion electrode (GDE) for oxygen evolution (OER) and reduction (ORR) in water-containing choline acetate with and without zinc acetate salt. water-containing choline acetate with and without zinc acetate salt.

(a) (b)

−1 1 FigureFigure 8. 8. (a(a) )OER/ORR OER/ORR activity activity of of MCO MCO in in ChAcO ChAcO + 30+ 30wt.% wt.% H2O H 2electrolyteO electrolyte at 10 at mV 10 mVs with s− with dry syntheticdry synthetic air. CVs air. CVsin 7 in M 7 MKOH KOH electrolyte electrolyte are are shown shown for for comparison. comparison. (b (b) )Overlay Overlay of of OER/ORR OER/ORR measurementsmeasurements at MCO and Pt in ChAcO + 3030 wt.% wt.% H H22OO ++ 0.010.01 M M ZnAcO. ZnAcO.

TheThe onset onset potential potential of of ORR ORR reaction reaction at at MCO MCO was was evaluated evaluated from from the the difference difference in in charge charge and and peakpeak shape betweenbetween voltammogramsvoltammograms recorded recorded in in air-saturated air-saturated electrolyte electrolyte and and those those in nitrogen in nitrogen and, and,accordingly, accordingly, located located at about at +about200 and +200+750 and mV +750 for CVmV in for ChAcO CV in and ChAcO KOH electrolyte,and KOH respectively.electrolyte, respectively.The ORR reaction The ORR is strongly reaction mass-transport is strongly mass-t limitedransport once reaching limited currentonce reaching density current close to density 2.5 mA 2 closecm− tovalue. 2.5 ThemA OERcm−2 reactionvalue. The kinetics OER relies reaction on OH kinetics− species’ relies concentration on OH- species’ and transport concentration in electrolyte and transportas well and in ariseselectrolyte at about as +well1200 and mV inarises ChAcO at about and + +12001500 mV mV in in KOH ChAcO solution. and Thus,+1500 amV diff erencein KOH in solution.onset potential Thus, betweena difference OER in and onset ORR potential (∆E) in ChAcObetween amounts OER and to 1ORR V compared (ΔE) in ChAcO to 750 mV amounts in KOH. to By 1 Vfurther compared comparison to 750 of MCOmV onsetin KOH. potentials By infurther water-containing comparison ChAcO of MCO with thoseonset of Ptpotentials (∆E = 1.4 in V) water-containingin Figure8b, it appears ChAcO that with the spinelthose of catalyst Pt (ΔE is = more 1.4 eV)ffi incient. Figure The 8b, onset it appears potential that (OP) the values spinel of catalystdifferent is MCO more and efficient. Pt catalyst The for onset OER potential/ORR as well(OP) as values corresponding of different∆E valuesMCO areand listed Pt catalyst in Table for3. OER/ORR as well as corresponding ΔE values are listed in Table 3. Table 3. List of onset potential values of different GDEs for OER/ORR in either 7 M KOH or ChAcO + 30 wt.% H O (IL) solution with dry synthetic air from experiments shown in Figures7 and8. Table 3. List2 of onset potential values of different GDEs for OER/ORR in either 7 M KOH or ChAcO + 30 wt.% H2O (IL) solutionGDE with Catalyst dry syntheticMCO air fromMCO experimentsPt shownPt in Figures 7 and 8. Electrolyte IL KOH IL KOH GDE Catalyst MCO # MCO Pt Pt OPOER/mV 1200 1500 1725 1800 # Electrolyte OPORRIL/mV 200KOH 750 35IL 1000 KOH ∆E/mV 1000 750 1690 800 OPOER/mV # 1200 1500 1725 1800 #: onset potential (OP) vs. RHE OPORR/mV # 200 750 35 1000 Δ 2 TheE/mV high increase in current1000 density of up to 10750 mA cm− observed1690 in both vertex potential800 regions can be associated to superposition of O#:2 onsetevolution potential with (OP) ChAcO vs. RHE oxidation in the anodic scan and H2 evolution with cathodic ChAcO decomposition in a cathodic scan. Apparently, the presence of 0.01 M ZnAcOThe in high choline increase acetate in doescurrent not density significantly of up a fftoect 10 activity mA cm of−2 MCO observed for ORR in both/OER vertex reactions. potential regions can be associated to superposition of O2 evolution with ChAcO oxidation in the anodic scan 3.2.3. Long-Term Activity of MCO for OER/ORR in ChAcO in Dry Air and H2 evolution with cathodic ChAcO decomposition in a cathodic scan. Apparently, the presence of 0.01Electrochemical M ZnAcO in stability choline of acetate as-prepared does MCOnot significantly/C65 GDE for affect OER andactivity ORR reactionsof MCO wasfor ORR/OER alternately reactions.tested for 2 h in ChAcO + 30 wt.% H2O + 0.01 M ZnAcO with synthetic air at room temperature. 2 At constant polarization value of 100 µA cm− , the GDE was able to perform about 2000 h of operation 3.2.3.with 100%Long-Term coulombic Activity efficiency of MCO (see for Figure OER/ORR9a). MCO in ChAcO/C65 GDE in Dry potential Air remained stable over 1400 h

(280Electrochemical cycles) with an energystability e ﬃofciency as-prepared of 52% whenMCO/C whole65 GDE charge for OER during and ORR ORR and reactions OER steps was is alternately tested for 2 h in ChAcO + 30 wt.% H2O + 0.01 M ZnAcO with synthetic air at room

Materials 2020, 13, x FOR PEER REVIEW 11 of 21

temperature. At constant polarization value of 100 µA cm‒2, the GDE was able to perform about 2000 Materialsh of operation2020, 13, 2975 with 100% coulombic efficiency (see Figure 9a). MCO/C65 GDE potential remained11 of 21 stable over 1400 h (280 cycles) with an energy efficiency of 52% when whole charge during ORR and OER steps is considered. At higher cycle number, however, overpotential of MCO for both OER and considered.ORR reactions At higher continuously cycle number, increased however, and overpotential decreased, respectively, of MCO for bothwhich OER confirms and ORR catalyst reactions continuouslydegradation increasedas reported and elsewhere decreased, [41]. respectively, This is, to whichour knowledge, conﬁrms catalystthe best degradationresult reported as reportedin the elsewhereliterature[ 41so ].far. This However, is, to our as can knowledge, been observed the best in a result single reported cycle (Figure in the 9b), literature it took a sowhile far. to However, reach assteady-state can been observed values of in1.4 a and single 0.6 V cycle for OER (Figure and9 b),ORR, it tookrespectively, a while which to reach is a steady-stateclear indication values for a of 1.4transport and 0.6 Vlimitation. for OER Moreover, and ORR, the respectively, potential gap which after is 30 a clearmin OCV indication between for OER a transport and ORR limitation. steps Moreover,amounts theup to potential 400 mV gap that after emphasizes 30 min OCVsluggish between transition OER from and ORRboth OER steps and amounts ORR mode up to to 400 an mV thatequilibrium emphasizes state. sluggish transition from both OER and ORR mode to an equilibrium state.

(a) (b) (c)

FigureFigure 9. 9.(a ()a Stationary) Stationary current current polarization polarization characteristics characteristics ofof MCOMCO GDEGDE forfor ORRORR/OER/OER as a function of −2 timeof time in ChAcO in ChAcO+ 30 + wt.%30 wt.% H2 HO2O electrolyte electrolyte at at 100 100µ µAA cm cm− withwith dry dry synthetic synthetic air. air. (b ()b 100th) 100th cycle cycle and and (c)(c corresponding) corresponding energy energy and and coulombic coulombic eefficiency.ﬃciency.

3.2.4.3.2.4. Activity Activity of of Redox Redox Zinc Zinc ReactionsReactions in ChAcO and and KOH KOH Electrolytes Electrolytes Similarly,Similarly, activity activity ofof zinczinc oxidationoxidation and zinc oxide oxide reduction reduction was was investigated investigated in in30 30wt.% wt.% water-containingwater-containing choline choline acetateacetate and and compared compared with with that that in in7 M 7 KOH M KOH (see (seeFigure Figure 10). Due10). to Due the to therelatively relatively strong strong alkaline alkaline nature nature of aqueous of aqueous choline choline acetate acetate (pH~11), (pH~11), we assume we assume that, first, that, zinc ﬁrst, zincdissolves dissolves into into zincate zincate that that further further precipitates precipitates into into zinc zinc oxide oxide on onzinc zinc foil foil after after reaching reaching saturation saturation atat an an electrode electrode/electrolyte/electrolyte interface interface similarly similarly to the to reactionthe reaction mechanisms mechanisms in 7 M in KOH. 7 M ZnOKOH. formation ZnO isformation supported is bysupported the Pourbaix by the Pourbaix diagram diagram of 0.01 of M 0.01 zinc M inzinc aqueous in aqueous medium medium shown shown in in Figure Figure S4. ThisS4. assumptionThis assumption will will be conﬁrmedbe confirmed by by Raman Raman investigation investigation reportedreported in Section Section 3.4. 3.4 Sluggishness. Sluggishness ofof zinc zinc reactions reactions in in aqueous aqueous ionicionic liquidliquid electrolyte and and especially especially that that of of zinc zinc oxide oxide reduction reduction is is clear.clear. The The reaction reaction kinetics kinetics depends depends onon availableavailable “f “free”ree” water water molecules. molecules. In In Liu’s Liu’s work, work, addition addition of of 10–50 wt.% H2O led to substantial enhancement of Zn deposition/stripping activity on a gold 10–50 wt.% H2O led to substantial enhancement of Zn deposition/stripping activity on a gold substrate substrate from a [EMim]TfO + 0.2 M Zn(TfO)2 mixture [42]. In the present study, addition of ZnAcO from a [EMim]TfO + 0.2 M Zn(TfO) mixture [42]. In the present study, addition of ZnAcO did not did not enhance zinc reaction kinetics2 (not shown here). It clearly appears that not only reaction enhance zinc reaction kinetics (not shown here). It clearly appears that not only reaction kinetics of both kinetics of both zinc reactions are faster in 7 M KOH, but also potential gap between both oxidation zinc reactions are faster in 7 M KOH, but also potential gap between both oxidation and reduction peak and reduction peak maxima is considerably narrower in KOH (170 mV) than that in choline acetate maxima(500 mV). is considerably Moreover, peak narrower intensit iny KOHof zinc (170 reactions mV) than in aqueous that in choline choline acetateacetate (500is about mV). 10 Moreover, times peaklower intensity than those of zinc in reactions 7 M KOH. inaqueous This should choline be acetaterelated isto about the low 10 times concentration lower than of thoseavailable in 7 M KOH.hydroxide This should ions as be well related as poor to the ionic low concentrationconductivity of of choline available acetate hydroxide solution ions by as an well approximate as poor ionic conductivityfactor of 100 of compared choline acetate to that solutionof the 7 M by KOH an approximate solution. factor of 100 compared to that of the 7 M KOH solution.

Materials 2020, 13, 2975 12 of 21 Materials 2020, 13, x FOR PEER REVIEW 12 of 21

Figure 10. CVCV of of a a zinc zinc foil foil in in (black) (black) 30 30 wt.% wt.% water- water-containingcontaining ChAcO ChAcO + +0.010.01 M M ZnAcO ZnAcO and and (blue) (blue) 7 7 M KOH electrolyte at 10 mV s 1. Inlet: zoom out of Zn oxidation/ZnO reduction voltammogram in M KOH electrolyte at 10 mV s−1.− Inlet: zoom out of Zn oxidation/ZnO reduction voltammogram in 7 7 M KOH. M KOH. 3.3. Zn/air Full-Cell Measurements 3.3. Zn/air Full-Cell Measurements 3.3.1. Zn/Air U/I Tests in ChAcO and KOH with Synthetic Dry Air in El-Cell 3.3.1. Zn/Air U/I Tests in ChAcO and KOH with Synthetic Dry Air in El-Cell Tests under dry air condition in El-Cell laboratory cells addressed the determination of the intrinsic activityTests of zincunder/air dry compounds air condition in 30 in wt.% El-Cell water-containing laboratory cells IL electrolyteaddressed whenthe determination compared to thatof the in intrinsicKOH solution. activity However,of zinc/air used compounds El-Cell designin 30 wt was.% water-containing not really adapted IL forelectrolyte long-term when tests compared with the toKOH that electrolyte. in KOH solution. Therefore, However, a fast chargeused El-Cell/discharge design procedure was not ofre oneally houradapted per for cycle long-term was applied tests within order the toKOH reduce electrolyte. the experiment’s Therefore, durationa fast char andge/discharge minimize diprocedureffusion and of one evaporation hour per ofcycle alkaline was appliedsolution. in Aorder Pt/C-based to reduce GDE the served experiment’s as an air dura referencetion and electrode minimize in thisdiffusion work. and In theevaporation present U of/I 2 alkalinetest protocol, solution. maximal A Pt/C-based reversible GDE capacity served yielded as an air per reference cycle was electrode about 0.063in this mAh work. at In 100 theµ Apresent cm− . U/IThis test represents protocol, about maximal 0.1% reversible depth of dischargecapacity yi (DOD)elded per of the cycle 73.8 was mAh about theoretical 0.063 mAh value at by100 assuming µA cm−2. Thisa complete represents dissolution about 0.1% of the depth zinc of foil discharge (90 mg) (DOD and neglecting) of the 73.8 the mAh contribution theoretical of value dissolved by assuming 0.01 M aZnAcO. complete The dissolution photographs of ofthe the zinc Zn foil anodes (90 beforemg) an andd neglecting after operation the contributi in 7 M KOHon of and dissolved choline 0.01 acetate M ZnAcO.are shown The in Figurephotographs S5. of the Zn anodes before and after operation in 7 M KOH and choline acetateIn are Figure shown 11, cellsin Figure with PtS5. GDE exhibits the highest OCV value in both electrolytes of about 1.33 V in ChAcOIn Figure and 11, 1.44 cells V inwith KOH. Pt GDE All the exhibits OCV the values high areest listedOCV value in Table in both S2. While electrolytes in KOH of electrolyte,about 1.33 Vall in cells ChAcO showed and a1.44 comparable V in KOH. charge All the and OCV discharge values are profiles, listed cellsin Table with S2. MCO While or in NCO KOH gas electrolyte, diffusion allelectrode cells showed exhibited a comparable poor activity charge for both and chargedischarge and profiles, discharge cells steps with in MCO choline or acetate. NCO gas Best diffusion overall electrodecell performance exhibited was poor yielded activity with for a both Pt-equipped charge and cell discharge in aqueous steps choline in choline acetate acetate. whereas Best an inverseoverall celltrend performance was observed was in yielded the cell with with a Pt Pt-equipped in KOH. In cell that in case, aqueous overpotential choline acetate values whereas were clearly an inverse lower trendthan those was observed in the KOH in the solution. cell with Figure Pt in 11 KOH.d shows In that the polarizationcase, overpotential and power values density were clearly curves lower of the thandifferent those zn in/air the cells KOH during solution. the discharge Figure 11d step shows in both the ChAcO polarization and KOH and electrolyte power density with drycurves air. of While the different zn/air cells during the discharge step in both ChAcO and KOH electrolyte2 with dry air.2 maximum peak power density of Zn/air cell with Pt/C amounts to about 122 µW cm− at 100 mA cm− While maximum peak power density of Zn/air cell with Pt/C amounts to about2 122 µW cm−2 at 100 and is very close to the values of all Zn/air cells in KOH (125–135 µW cm− ), Zn/air cells with spinel mAcatalysts cm−2 exhibitand is very about close 50% to lower the performancevalues of all inZn/air an aqueous cells in ILKOH electrolyte. (125–135 µW cm−2), Zn/air cells with spinel catalysts exhibit about 50% lower performance in an aqueous IL electrolyte.

MaterialsMaterials2020 2020,,13 13,, 2975x FOR PEER REVIEW 1313 ofof 2121

(a) (b)

−22 FigureFigure 11.11. UU/I/I curvescurves ofof didifferentﬀerent ZnZn/air/air cellscells betweenbetween 11 andand 100100µ µAA cmcm− forfor 11 hh perper cyclecycle (15(15 minmin chargecharge/discharge/discharge each) each) with with dry dry air air in (ain) 30(a) wt.% 30 wt.% ChAcO ChAcO+ 0.01 + M 0.01 ZnAcO M ZnAcO and (b) 7and M KOH(b) 7 solutions.M KOH 2 −2 (solutions.c) Selection (c of) Selection a single cycleof a single at 100 cycleµA cm at− 100in (solidµA cm lines)in (solid choline lines) acetate choline and (dottedacetate lines)and (dotted KOH. (lines)d) Comparison KOH. (d) ofComparison polarization of (solidpolarization lines) and(solid power lines) densityand power (dotted density lines) (dotted behavior lines) during behavior the dischargeduring the protocol discharge in ChAoC protocol from in FigureChAoC 11 afrom and KOHFigure from 11a Figure and KOH 11b at from current Figure increment 11b at from current 1 to 2 100incrementµA cm− fromin the 1 to forward 100 µA direction.cm−2 in the forward direction.

3.3.2.3.3.2. ZnZn/Air/Air I-UI-U TestsTests inin ChAcOChAcO withwith AmbientAmbient AirAir inin El-CellEl-Cell SimilarSimilar protocolprotocol asas inin SectionSection 3.3.13.3.1 waswas usedused forfor thethe UU/I/I experimentexperiment inin ambientambient air.air. ForFor betterbetter 2 1 clarity, however, only cycles carried out at 100 µA cm− (2.8 mA g− zinc) are shown in Figure 12a–c clarity, however, only cycles carried out at 100 µA cm−2 (2.8 mA g−1zinc) are shown in Figure 12a–c wherewhere evidenceevidence e effectsffects the the moisture moisture in in air air on on short shor charget charge/discharge/discharge (15 (15 min min each) each) polarization polarization time time of diofff differenterent Zn /Zn/airair cells. cells. TheThe followingfollowing conclusionsconclusions cancan bebe drawn:drawn: (i)(i) ChargeCharge//dischargedischarge characteristicscharacteristics ofof cellcell withwith PtPt asas aa catalystcatalyst airair electrodeelectrode surpassedsurpassed thosethose withwith MCOMCO andand NCONCO byby far,far, (ii)(ii) bestbest performancesperformances withwith PtPt cellscells werewere yieldedyielded with with dry dry air, air, it appearsit appears that that the the presence presence of humidityof humidity in air in aairffected affected kinetics kinetics of Pt of for Pt both for chargeboth charge (OER) (OER) and discharge and discharge (ORR) (ORR) reaction reaction step, (iii) step, performance (iii) performance of cell withof cell NCO with decayed NCO decayed under dryunder air feed,dry air and feed, (iv) and substantially (iv) substant positiveially epositiveffect ofmoisture effect of inmoisture air feed in on air discharge feed on(ORR) discharge step (ORR) of cell withstep MCOof cell whilewith MCO cell voltage while cell profiles voltage during profiles charging during steps charging remained steps almostremained unchanged almost unchanged and were apparentlyand were apparently insensitive insensitive to water. to water. In this preliminary survey, cells with the NCO air electrode demonstrated poor cycling ability, especially during charging steps. Therefore, this catalyst was not considered in further investigations.

Materials 2020, 13, x FOR PEER REVIEW 14 of 21

Materials 2020, 13, 2975 14 of 21 Materials 2020, 13, x FOR PEER REVIEW 14 of 21

(a) (b) (c)

Figure 12. Comparison of 1-h charge/discharge profiles of different Zn/air cells at 100 µA cm−2 during U/I test procedure in (a) 30 wt.% water-containing choline acetate electrolyte—dry air (WD) and (b) 30 wt.% water-containing choline acetate electrolyte—ambient air feed (30% RH at 23°C) (WW). (c) Overlay(a) of Figure 12a,b. (b) (c)

FigureFigure 12. 12.Comparison Comparison of of 1-h 1-h charge charge/discharge/discharge proﬁles profiles of of di ﬀdifferenterent Zn /Zn/airair cells cells at 100 at µ100A cmµA 2cmduring−2 In this preliminary survey, cells with the NCO air electrode demonstrated poor −cycling ability, Uduring/I test procedureU/I test procedure in (a) 30 in wt.%(a) 30 water-containingwt.% water-containing choline choline acetate acetate electrolyte—dry electrolyte—dry air air (WD) (WD) and especially during charging steps. Therefore, this catalyst was not considered in further (band) 30 ( wt.%b) 30 water-containingwt.% water-containing choline choline acetate acetate electrolyte—ambient electrolyte—ambient air feedair feed (30% (30% RH RH at 23 at◦ C)23°C) (WW). investigations. (c(WW).) Overlay (c) Overlay of Figure of 12 Figurea,b. 12a,b.

3.3.3.3.3.3. ZnIn Zn/Air /thisAir preliminary Charge Charge/Discharge/Discharge survey, Test cells Test in with in ChAcO ChAcO the NCO for for 26 air26 h /electrodeh/CycleCycle in in thedemonstrated the El-Cell El-Cell poor cycling ability, especially during charging steps. Therefore, this catalyst was not considered in further ForFor the the majority majority of of battery battery applications, applications, either either high-power high-power density density for for short short duration duration or or high high investigations. capacitycapacity over over a longa long period period in ambientin ambient air isair required. is required. Since Since zinc/ airzinc/air battery battery with cholinewith choline acetate acetate ionic liquidionic3.3.3. electrolyte liquidZn/Air electrolyteCharge/Discharge can clearly can not clearly deliverTest notin ChAcO high deliver power for high 26 densities, h/Cycle power indensities, the the next El-Cell experiments the next experiments shown in Figure shown 13 in focusFigure on 13 an focus extension on an ofextension cycle duration of cycle up duration to 8 h ofup the to charge8 h of the/discharge charge/discharge step (5 h OCV))step (5 ath OCV)) 100 µA at 2 For the−2 majority1 of−1 battery applications, either high-power density for short duration or high cm100− µA(2.8 cm mA (2.8 g− zincmA) withg zinc either) with Pteither or MCO Pt or as MCO a bifunctional as a bifunctional air catalyst air catalyst in ambient in ambient air. air. capacity over a long period in ambient air is required. Since zinc/air battery with choline acetate ionic liquid electrolyte can clearly not deliver high power densities, the next experiments shown in Figure 13 focus on an extension of cycle duration up to 8 h of the charge/discharge step (5 h OCV)) at 100 µA cm−2 (2.8 mA g−1zinc) with either Pt or MCO as a bifunctional air catalyst in ambient air.

(a) (b)

FigureFigure 13. 13.(a ()a Average) Average values values of of zinc zinc/air/air cell cell performance performance in in 30 30 wt.% wt.% H 2HO2/O/ChAcOChAcO + 0.01+ 0.01 M M ZnAcO ZnAcO 2 2 2 1 electrolyte during cycling at 100 µA cm −(2.822 mA g −1zinc) in ambient air for 26 h per cycle (8 h electrolyte during (cyclinga) at 100 µA cm− (2.82 mA −g zinc ) in ambient(b air) for 26 h per cycle (8 h chargecharge/discharge/discharge each) each) in in El-Cell. El-Cell. ( b(b)) Representation Representation of of the the fifth fifth cycle. cycle. Voltage Voltage cut-o cut-offff limits limits were were fixedfixedFigure to to 13. 2.5 2.5 ( anda and) Average 0.10.1 V.V. values TheThe absoluteabsolute of zinc/air charge cell performance and di dischargescharge in 30capacity capacity wt.% H value2 valueO/ChAcO wa wass limited+ limited0.01 M to ZnAcO to 2.2 2.2 and2 and 2.0 −2 −1 2.0mAh,electrolyte mAh, respectively. respectively. during cycling at 100 µA cm (2.82 mA g zinc) in ambient air for 26 h per cycle (8 h charge/discharge each) in El-Cell. (b) Representation of the fifth cycle. Voltage cut-off limits were OneOnefixed can canto 2.5 divide divide and 0.1 26 26 V. h /h/cycle cycleThe absolute experiment experiment charge depicted and depicted discharge in in Figure Figurecapacity 13 13a avalue into into wa two twos limited main main to regions. regions. 2.2 and Within2.0 Within the the firstfirst 15 15mAh, cycles, cycles, respectively. cells cells with with a a Pt Pt electrode electrode clearly clearly over-performed over-performed those those with with an an MCO MCO GDE, GDE, especially especially duringduring the the discharge discharge step. step. TheThe reason reason why why the the mean mean discharge discharge potential potential of of the the Pt Pt cells cells suddenly suddenly One can divide 26 h/cycle experiment depicted in Figure 13a into two main regions. Within the dropped from 0.75 V down to 0.15 V is not explicable yet. Cells with an MCO catalyst were relatively droppedfirst 15 cycles, from cells0.75 Vwith down a Pt to electrode 0.15 V is clearly not explicable over-performed yet. Cells those with with an MCOan MCO catalyst GDE, were especially relatively stable over 34 days. stableduring over the discharge34 days. step. The reason why the mean discharge potential of the Pt cells suddenly Figure 13b shows overlay of the fifth cycle of cells that can be commented as follows. (i) Even dropped from 0.75 V down to 0.15 V is not explicable yet. Cells with an MCO catalyst were relatively after 8 h of charge, no single cell reached steady-state potential. (ii) In case of the discharge step, stable over 34 days. achievement of steady-state is faster in the MCO cell than in the Pt one. (iii) OCV of both systems tended to approach 1 V value. (iv) Superior performance of Pt cell during discharge and of MCO cells

Materials 2020, 13, x FOR PEER REVIEW 15 of 21

Figure 13b shows overlay of the fifth cycle of cells that can be commented as follows. (i) Even after 8 h of charge, no single cell reached steady-state potential. (ii) In case of the discharge step, Materialsachievement2020, 13, of 2975 steady-state is faster in the MCO cell than in the Pt one. (iii) OCV of both systems15 of 21 tended to approach 1 V value. (iv) Superior performance of Pt cell during discharge and of MCO cells during charge is in good accordance with ORR/OER half-cell measurements. (v) Clear during charge is in good accordance with ORR/OER half-cell measurements. (v) Clear discrepancy discrepancy with respect to charging voltage of cell with a Pt catalyst in comparison with short U/I with respect to charging voltage of cell with a Pt catalyst in comparison with short U/I polarization polarization time (2.2 V in average vs. 1.5 V in Figure 11b). No explanation for this behaviour has time (2.2 V in average vs. 1.5 V in Figure 11b). No explanation for this behaviour has been found yet. been found yet. 3.3.4. Zinc/Air Charge/Discharge Tests in ChAcO for 10 Days/Cycle in the EL-Cell 3.3.4. Zinc/Air Charge/Discharge Tests in ChAcO for 10 Days/Cycle in the EL-Cell The best performing MCO and commercial Pt/C GDEs were further tested for a very large period The best performing MCO and commercial Pt/C GDEs were further tested for a very large of 100 h for a charge and discharge step each (about 10 days per cycle) in Zn/air cell under ambient air period of 100 h for a charge and discharge step each (about 10 days per cycle) in Zn/air cell under conditionsambient air (Figure conditions 14). Even(Figure though 14). Even polarization though po curveslarization are not curves very are ﬂat, not Zn very/air cellflat,with Zn/air MCO cell air electrodewith MCO exhibited air electrode clearly exhibited better performance clearly better than performance the cell with than Ptthe GDE. cell with Pt GDE.

(a) (b) (c)

2 FigureFigure 14. 14.( (aa)) Long-termLong-term stability stability cycling cycling test test of of MCO MCO and and Pt cells Pt cells in ChAcO in ChAcO + 0.01+ M0.01 ZnAcO M ZnAcO + 30 2 + wt.% H2O cell at 100 µA cm−2 for2 about 10 days/cycle in ambient air (30% RH at 23 °C). (b, c) 30 wt.% H2O cell at 100 µA cm− for about 10 days/cycle in ambient air (30% RH at 23 ◦C). (b,c) ComparisonComparison of of all all charge charge/discharge/discharge stepssteps of co correspondingrresponding GDE. GDE. Cut-off Cut-oﬀ limitslimits 2.5 2.5 and and 0.1 0.1 V. V.

Moreover,Moreover, reversible reversible capacity capacity of of zinc zinc/air/air battery battery amounted amounted to to 25.4 25.4 mAh mAh that that corresponds corresponds to aboutto 28%about DOD 28% at 100%DOD columbicat 100% columbic eﬃciency. efficiency. This excellent This excellent behaviour behaviour was observed was observed until the until ﬁrst 7ththe (1500first h) and7th 6th (1500 (1300 h) and h) cycles 6th (1300 for MCOh) cycles and for Pt MCO cells, and respectively. Pt cells, respectively.

3.3.5.3.3.5. Zn Zn/Air/Air Single Single Discharge Discharge StepStep in ChAcO for for 12 12 h/Cycle h/Cycle in in El-Cell El-Cell Next,Next, the the investigation investigation aimed aimed at at determining determining maximum maximum cell cell discharge discharge capacity capacity of of a a pristine pristine zinc zinc foil at 100 µA2 cm−2 under ambient air condition. foil at 100 µA cm− under ambient air condition. In 30% RH air atmosphere, capacities of Zn/air battery using Pt/C and MCO/C as GDE In 30% RH air atmosphere, capacities of Zn/air battery using Pt/C and MCO/C as GDE amounted amounted to 34.3 and 38.1 mAh that represents 38% and 42% DOD values, respectively (see to 34.3 and 38.1 mAh that represents 38% and 42% DOD values, respectively (see polarization polarization profile in Figure 15a). After these discharge tests, however, it was not possible to proﬁle in Figure 15a). After these discharge tests, however, it was not possible to recharge the cells recharge the cells again, presumably due to irreversible formation of dense zinc oxide/hydroxide again, presumably due to irreversible formation of dense zinc oxide/hydroxide layer and irreversible layer and irreversible passivation of the electrode surface. This is confirmed by SEM-EDX elemental passivation of the electrode surface. This is conﬁrmed by SEM-EDX elemental mapping and line scan of mapping and line scan of post-mortem analysis of zinc foil shown in Figure 14b where evidence of post-mortemhigh oxygen analysis concentration of zinc foilon the shown zinc insurface Figure an 14db near-surface where evidence layer of in high a depth oxygen of concentrationabout 10 µm on the(yellow-marked zinc surface and area near-surface in Figure 15b). layer in a depth of about 10 µm (yellow-marked area in Figure 15b).

Materials 2020, 13, x FOR PEER REVIEW 16 of 21 MaterialsMaterials2020 2020, 13,, 13 2975, x FOR PEER REVIEW 16 of16 21of 21

(a) (b) (a) (b) Figure 15. (a) Discharge profile of zinc/air cell with MCO and Pt as GDE at 100 µA cm−2 and (b) SEM 2 −2 FigureFigure 15. 15.(a) ( Dischargea) Discharge profile profile of zincof zinc/air/air cell cell with with MCO MCO and and Pt asPt GDEas GDE at 100 at 100µA µA cm −cmand and (b) ( SEMb) SEM image of a cross-sectional Zn electrode after compete discharge of ChAcO + 0.01 M ZnAcO2 + 30% imageimage of aof cross-sectional a cross-sectional Zn Zn electrode electrode after afte competer compete discharge discharge of ChAcOof ChAcO+ 0.01 + 0.01 M ZnAcOM ZnAcO2 +230% + 30% H2O cell containing MCO GDE at 100 µA cm−2 in ambient air with an EDX signal of specific elemental 2 H HO2 cellO cell containing containing MCO MCO GDE GDE at 100at 100µA µA cm cmin−2 in ambient ambient air air with with an an EDX EDX signal signal of of specific specific elemental elemental count2 rate profile in line scan for oxygen (red)− and zinc (green) along and zinc surface in contact with countcount rate rate profile profile in linein line scan scan for for oxygen oxygen (red) (red) and and zinc zinc (green) (green) along along and and zinc zinc surface surface in contactin contact with with an electrolyte, zinc foil cross section, and zinc foil in contact with the current collector (bright area). anan electrolyte, electrolyte, zinc zinc foil foil cross cross section, section, and and zinc zinc foil foil in contactin contact with with the the current current collector collector (bright (bright area). area). An enlarged SEM and line scan image are provided in Figure S6 for better readability. AnAn enlarged enlarged SEM SEM and and line line scan scan image image are are provided provided in Figurein Figure S6 forS6 for better better readability. readability. 3.3.6. Zn/Air Charge/Discharge Steps for 12 h/Cycle in CR2032 Coin Cell 3.3.6.3.3.6. Zn Zn/Air/Air Charge Charge/Discharge/Discharge Steps Steps for for 12 12 h/Cycle h/Cycle in CR2032in CR2032 Coin Coin Cell Cell Figure 16 illustrates the behaviour of zinc/air cell with either MCO or Pt as an air catalyst in FigureFigure 16 16illustrates illustrates the the behaviour behaviour of of zinc zinc/air/air cell cell with with either either MCO MCO or or Pt Pt as as an an air air catalyst catalyst in in aqueous choline acetate for 5-h charge/discharge step in coin cell design. The reversible capacity was aqueousaqueous choline choline acetate acetate for for 5-h 5-h charge charge/discharge/discharge step step in coinin coin cell cell design. design. The The reversible reversible capacity capacity was was 1.01 mAh (1.4% DOD). Although current density amounted to the same value of 100µ µA cm2 −2 (2.80 1.011.01 mAh mAh (1.4% (1.4% DOD). DOD). Although Although current current density density amounted amounted to the to same the valuesame ofvalue 100 ofA 100 cm −µA(2.80 cm− mA2 (2.80 mAg 1 g−1),zinc steady-state), steady-state operation operation seems seems to be to reached be reached at 1.80 at and 1.80 0.94 and V for0.94 the V chargefor the and charge discharge and −mAzinc g−1zinc), steady-state operation seems to be reached at 1.80 and 0.94 V for the charge and discharge step, repectively, in case of the cell with an MCO air electrode, which represents an energy step,discharge repectively, step, in repectively, case of the cellin case with of an the MCO cell with air electrode, an MCOwhich air electrode, represents which an energy represents efficiency an energy of efficiency of 52% for about 300 h. 52%efficiency for about of 300 52% h. for about 300 h.

(a) (b) (a) (b) FigureFigure 16. 16. (a(a) )Long-term Long-term stability stability cycle cycle tests tests in in zinc/air zinc/air coin coin cell cell with with MCO MCO or or Pt Pt as as an an air air electrode electrode in in Figure 16. (a) Long-term stability cycle tests in zinc/air coin cell with MCO or Pt as an air electrode2 in ChAcOChAcO + 0.01 MM ZnAcOZnAcO22+ +30% 30% H H2O2O electrolyte electrolyte for for 12 h12 per h cycleper cycle (5 h charge (5 h charge/discharge)/discharge) at 100 µ atAcm 100− ChAcO + 0.01 M ZnAcO2 + 30% H2O electrolyte for 12 h per cycle (5 h charge/discharge) at 100 µAcmin ambient−2 in ambient air (30% air RH (30% at 23 ◦RHC). (bat) Comparison23°C). (b) Comparison of charge/discharge of charge/discharge proﬁle during profile 18th cycle.during Voltage 18th µAcm−2 in ambient air (30% RH at 23°C). (b) Comparison of charge/discharge profile during 18th cycle.cut-o ﬀVoltagelimits ofcut-off 2.5 and limits 0.1 V.of 2.5 and 0.1 V. cycle. Voltage cut-off limits of 2.5 and 0.1 V. TheThe CR2032 CR2032 coin coin cell cell design design with with 16 16 mm mm in in diameter diameter zinc zinc foil foil appears appears to to be be more more adapted adapted for for The CR2032 coin cell design with 16 mm in diameter zinc foil appears to be more adapted for long-termlong-term experiments experiments than than that that of of El-Cell. El-Cell. So Someme relevant relevant OCV OCV values values are are listed listed in in Table Table S2. S2. long-term experiments than that of El-Cell. Some relevant OCV values are listed in Table S2.

Materials 2020, 13, x FOR PEER REVIEW 17 of 21

Table 4 summarized all results from Zn/air charge/discharge experiments of this work. Best performance in terms of test duration, reversible capacity, and energy efficiency were yielded in Materials 2020, 13, 2975 17 of 21 El-cell design with an MCO air catalyst during a 100 h charge/discharge protocol.

TableTable4 4.summarized Summary of relevant all results results from of Zn/air Zn / aircharge/discharge charge/discharge experiments experiments at 100 µA of cm this−2 (2.8 work. Best performancemA∙g−1zinc) in ChAcO in terms + 30 of wt.% test duration,H0.01 M ZnAcO reversible2 and capacity,ambient air. and energy eﬃciency were yielded in El-cell designDesign with an MCO air catalyst during aEl-Cell 100 h charge /discharge protocol. Coin Cell

Current/µA 254 200 2 Table 4. Summary of relevant results of Zn/air charge/discharge experiments at 100 µA cm− 1 (2.8Time mA perg −cycle/hzinc) in ChAcO + 30 wt.% H 262 O + 0.01 M ZnAcO2 and 240 ambient air. 12 · Rev. capacity/mAhDesign 2.03 El-Cell 25.40 Coin Cell 1.01 DOD/%Current /µA2.2 25428.0 200 1.4 Air catalystTime per cycle/h Pt MCO 26 Pt 240 MCO Pt 12 MCO Rev. capacity/mAh 2.03 25.40 1.01 CyclesDOD number//% *- 1–17 1–33 2.2 1–6 28.0 1–7 1–30 1.4 1–33 Air catalyst Pt MCO Pt MCO Pt MCO Duration/h * 440 850 1300 1500 360 396 Cycles number/ *- 1–17 1–33 1–6 1–7 1–30 1–33 CoulombicDuration efficiency/%/h * * 100 440 100 850 100 1300 1001500 360 100 396 100 Coulombic efficiency/% * 100 100 100 100 100 100 Energy efficiency/%* 28–35 15–24 37–51 29–54 39–48 33–56 Energy efficiency/%* 28–35 15–24 37–51 29–54 39–48 33–56 * Values related to the region in which relatively stable cell voltage behavior was observed. * Values related to the region in which relatively stable cell voltage behavior was observed. 3.4. In-Situ Raman Investigation in the Side-by-Side Zn/Air Cell 3.4. In-Situ Raman Investigation in the Side-by-Side Zn/Air Cell In order to identify reaction products during zinc/air cell discharge process in 30 wt.% In order to identify reaction products during zinc/air cell discharge process in 30 wt.% water-containing choline acetate, an in-situ Raman spectroscopy was carried out at different DODs water-containing choline acetate, an in-situ Raman spectroscopy was carried out at different DODs in in an optical SBS cell displayed in Figure 17. an optical SBS cell displayed in Figure 17.

(a) (b)

FigureFigure 17.17. ((aa)) SBSSBS cellcell designdesign (1:GDE,(1:GDE, 2:separator,2:separator, 3:Zn3:Zn/carbon/carbon paper)paper) andand ((bb)) ZnZn/air/air cellcell dischargedischarge 2 proﬁleprofile in in ChAcO ChAcO+ +30 30 wt.%wt.% HH22OO atat 55 µµAA cmcm−−2..

ForFor thisthis specialspecial cellcell design,design, however,however, itit waswas necessarynecessary toto useuse highlyhighly porousporous zinczinc particlesparticles forfor appropriatedappropriated wetting.wetting. Accordingly,Accordingly, aa zinczinc layerlayer waswas previouslypreviously electrodepositedelectrodeposited onon aa carboncarbon paperpaper substrate.substrate. AA line-scanline-scan consistingconsisting ofof 1010 RamanRaman spectraspectra waswas performedperformed atat equidistantequidistant locationslocations everyevery 10%10% DOD. DOD. The The cell cell delivered delivered an an initial initial OCV OCV of 1.4 of V 1. and4 V aand constant a constant current current discharge discharge step was step carried was 2 outcarried at 5 µoutA cmat 5− µAfor cm 340−2 hfor that 340 represents h that represents a capacity a capacity of 1.75 mAhof 1.75 (~9.4% mAh DOD).(~9.4% ForDOD). higher For clarity,higher however,clarity, however, only spectra only before spectra and before after completeand after discharge complete are discharge shown inare Figure shown 18a. in The Figure formation 18a. The of theformation reaction of product the reaction in a form product of white in a precipitateform of whit layere precipitate was observed layer at was the observed zinc anode at andthe zinc glass anode ﬁbre separatorand glass interface, fibre separator as can interface, be seen in as a redcan windowbe seen in displayed a red window in Figure displayed 18a. However, in Figure the 18a. GDE However, surface wasthe GDE nearly surface unchanged. was nearly unchanged. FigureFigure 18 18aa shows shows superposition superposition of Ramanof Raman spectra spectr of thea of Zn the/air cellZn/air surface cell atsurface 10 identical at 10 locationsidentical (a-jlocations position) (a-j before position) and before after cell and discharge. after cell After discha 340rge. h ofAfter discharge, 340 h of two discharge, strong bands two strong at ca. 1040 bands and at 1 1270ca. 1040 cm− andwithin 1270 the cm electrolyte−1 within the separator electrolyte (position separator e in Figure(position 18a) e indicatein Figure zincate 18a) indicate formation zincate [43]. 1 Two other strong bands at 720 and 1603 cm− are assigned to OH rotational modes [44]. The presence of the zinc oxide compound is also shown by SEM-EDX mapping, as shown in Figure 18b. The peak at 1 2 1583 cm− (G peak) in Raman spectra is assigned to sp hybridized carbon from Toray carbon ﬁbers. Materials 2020, 13, x FOR PEER REVIEW 18 of 21

formation [43]. Two other strong bands at 720 and 1603 cm−1 are assigned to OH rotational modes [44]. The presence of the zinc oxide compound is also shown by SEM-EDX mapping, as shown in MaterialsFigure2020 18b., 13 ,The 2975 peak at 1583 cm−1 (G peak) in Raman spectra is assigned to sp2 hybridized carbon18 of 21 from Toray carbon fibers.

(a)

(b)

FigureFigure 18. 18.(a ()a Raman) Raman line line scanning scanning spectraspectra atat 1010 identicalidentical positions positions of of electrodes, electrodes, separators, separators, and and electrolyteselectrolytes with with a a photophoto of system set-up set-up (black (black)) before before and and (red) (red) after after the discharge the discharge procedure. procedure. (b) (b)SEM SEM and and EDX EDX mapping mapping images images of discharge of discharge product product in glass in fiber glass separator ﬁber separator in a red marked in a red square marked squareregion. region.

4. Conclusions4. Conclusions

InIn this this work, work, feasibility feasibility of a of zinc a /airzinc/air cell operating cell operating with awith 30 wt.% a 30 H wt.%2O-containing H2O-containing choline choline acetate as 2 electrolyteacetate as and electrolyte MnCo2O and4 as aMnCo bifunctional2O4 as a airbifunctional electrode atair 100 electrodeµA cm− atfor 100 up µA to 100cm−2 h for charge up to/discharge 100 h stepcharge/discharge for about 1500 hstep under for ambientabout 1500 air h condition under am wasbient successfully air condition demonstrated was successfully in El-Cell. demonstrated Best results in termsin El-Cell. of energy Best results efficiency in terms up to 52%of energy for 300 efficiency h (25 cycles) up to were 52% obtained for 300 h during (25 cycles) a 5 h chargewere obtained/discharge stepduring in the a CR20325 h charge/discharge coin cell. Due step to in a relativelythe CR2032 high coin pH cell. value Due to of a about relatively 11, reaction high pH productsvalue of about such as zinc11, oxide reaction/hydroxide products of such zinc as/air zinc cell oxide/hydroxide with aqueous ionicof zinc/air liquid cell were with as aqueous expected, ionic which liquid is were similar as to thoseexpected, with highly which concentrated is similar to potassiumthose with hydroxide.highly concentrated In contrast potassium to experiments hydroxide. in KOH, In contrast neither to zinc dendriteexperiments nor carbonate in KOH, formation neither zinc was dendrite detected inno choliner carbonate acetate-based formation electrolyte. was detected The in use choline of a zinc foilacetate-based as a negative electrolyte. electrode wasThe use principally of a zinc motivated foil as a negative by its facile electrode integration was principally into El-Cell motivated and CR2032 by coinits cellfacile designs. integration Relatively into El-Cell high reversible and CR2032 capacities coin ce upll designs. to 25.4 mAhRelatively (28% high DOD) reversible were yielded capacities despite up to 25.4 mAh (28% DOD) were yielded despite a smooth and dense surface of zinc foil material. a smooth and dense surface of zinc foil material. Further work aims to improve the electrode design Further work aims to improve the electrode design and cell configuration in order to achieve higher and cell configuration in order to achieve higher reversible capacity and current density values by reversible capacity and current density values by using zinc powder and a thinner separator for using zinc powder and a thinner separator for rechargeable zinc/air applications. The amount of cobalt rechargeable zinc/air applications. The amount of cobalt in the bifunctional air electrode has to be in the bifunctional air electrode has to be reduced as well. reduced as well. Supplementary Materials: The following are available online at http://www.mdpi.com/1996-1944/13/13/2975/s1. FigureSupplementary S1: (a) Schematic Materials: representation The following and are picture available of El-Cell online configuration at www.mdpi.com/xxx/s1. (SS–stainless steel).Figure(b) S1: Picture (a) of coinSchematic cell type representation CR2032 with and holes picture provided of El-Cell for configuration air access at (SS–stainless the cathode cap.steel). Electrode (b) Picture arrangement of coin cell type in the coin cell is identical as in El-Cell except the number of separators, which was two 1-mm thick, Figure S2: SEM images and EDX pattern of as-prepared (a) MCO and (b) NCO powder sample, Figure S3: BET isotherm curve of (a) MCO, (b) NCO, and (c) Pt/C. (d–f). Corresponding pore size distribution plots, Figure S4: Pourbaix diagram of 0.01 M zinc in aqueous solution from materials project.org, Figure S5: Photographs of Zn/air El-Cell components (top) before experiments, (middle) after 20 days test in choline acetate, and (bottom) 1-day test in 7 M KOH. (GF = glass fiber separator), Figure S6: SEM image of cross-sectional Zn electrode after compete discharge of 2 ChAcO + 0.01 M ZnAcO2 + 30% H2O cell containing MCO GDE at 100 µA cm− in ambient air with an EDX Materials 2020, 13, 2975 19 of 21 signal of specific elemental count rate profile in line scan for oxygen (red) and zinc (green) along and zinc surface in contact with electrolyte, zinc foil cross section, and zinc foil in contact with the current collector (bright area). Table S1: Physicochemical properties of different water/ChAcO mixtures at 20 ◦C, Table S2: List of OCV values from different Zn/air cells. Author Contributions: M.S. performed half-cell and El-Cell experiments. S.P.B. synthesized and characterized catalyst materials by XRD and SEM analysis. A.A.S. fabricated GDEs, coin cell, and performed tests. K.K. performed in situ Raman experiments. W.P. determined pH and conductivity of IL mixture. M.S., S.P.B., A.A.S., W.P. and J.-F.D. wrote the article. J.-F.D. reviewed the article. All authors have read and agreed to the published version of the manuscript. Funding: The “Bundesministerium für Bildung und Forschung (BMBF)” (6.Energieforschungsprogram, LuZi, grant no.: 03SF0499B) funded this research. Acknowledgments: We express our sincere thanks to LuZi project partners for an excellent cooperation and especially Frank Endres´s group from TU Clausthal for sharing their knowledge on choline acetate electrolyte development. Sai Praneet Batchu expresses his gratitude to the Alumni association of IIT-Madras for financial support to complete his summer internship at DECHEMA-Forschungsinstitut. Conflicts of Interest: The authors declare no conflict of interest. The founding sponsors had no role in the design of the study, in the collection, analyses, or interpretation of data, in the writing of the manuscript, or in the decision to publish the results.

References

1. Lee, J.S.; Kim, S.T.; Cao, R.; Choi, N.S.; Liu, M.; Lee, K.T.; Cho, J. Metal–air batteries with high energy density: Li–air versus Zn–air. Adv. Energy Mater. 2011, 1, 34–50. [CrossRef] 2. Rahman, A.; Wang, X.; Wen, C.; Rani, J.V.; Kanakaiah, V.; Dadmal, T.; Bhavanarushi, S.; Rao, M.S. High energy density metal-air batteries: A review. J. Electrochem. Soc. 2013, 160, A1759–A1771. [CrossRef]

3. Drillet, J.F.; Holzer, F.; Kallis, T.; Müller, S.; Schmidt, V.M. Influence of CO2 on the stability of bifunctional oxygen electrodes for rechargeable zinc/air batteries and study of different CO2 filter materials. Phys. Chem. Chem. Phys. 2001, 3, 368–371. [CrossRef] 4. Neburchilov, V.; Wang, H.; Martin, J.J.; Qu, W. A review on air cathodes for zinc-air fuel cells. J. Power Sources 2010, 195, 1271–1291. [CrossRef] 5. Toussaint, G.; Stevens, P.; Akrour, L.; Rouget, R.; Fourgeot, F. Development of a Rechargeable Zinc-Air Battery. ECS Trans. 2010, 28, 25–34. [CrossRef] 6. Suntivich, J.; May, K.J.; Gasteiger, H.A.; Goodenough, J.B.; Shao-Horn, Y. A perovskite oxide optimized for oxygen evolution catalysis from molecular orbital principles. Science 2011, 334, 1383–1385. [CrossRef]

7. Bogolowski, N.; Ngaleu, O.; Sakthivel, M.; Drillet, J.-F. Long-life bifunctional BaSrCoFeO3/C gas diﬀusion electrode. Carbon 2017, 119, 511–518. [CrossRef] 8. Jörissen, L. Bifunctional oxygen/air electrodes. J. Power Sources 2006, 155, 23–32. [CrossRef]

9. Pletcher, D.; Li, X.; Price, S.W.T.; Russell, A.E.; Sönmez, T.; Thompson, S.J. Comparison of the SpinelsCo3O4 and NiCo2O4 as Bifunctional Oxygen Catalysts in Alkaline Media. Electrochim. Acta 2016, 188, 286–293. [CrossRef] 10. Gebremariam, T.T.; Chen, F.; Wang, Q.; Wang, J.; Liu, Y.; Wang, X.; Qaseem, A. Bimetallic Mn–Co Oxide Nanoparticles Anchored on Carbon Nanoﬁbers Wrapped in Nitrogen-Doped Carbon for Application in Zn–Air Batteries and Supercapacitors. ACS Appl. Energy Mater. 2018, 1, 1612–1625. [CrossRef]

11. Prabu, M.; Ramakrishnan, P.; Shanmugam, S. CoMn2O4 nanoparticles anchored on nitrogen-doped graphene nanosheets as bifunctional electrocatalyst for rechargeable zinc–air battery. Electrochem. Commun. 2014, 41, 59–63. [CrossRef] 12. Davari, E.; Johnson, A.D.; Mittal, A.; Xiong, M.; Ivey, D.G. Manganese-cobalt mixed oxide ﬁlm as a bifunctional catalyst for rechargeable zinc-air batteries. Electrochim. Acta 2016, 211, 735–743. [CrossRef] 13. Wang, Z.; Xiao, S.; An, Y.; Long, X.; Zheng, X.; Lu, X.; Tong, Y.; Yang, S. Co(II)1-xCo(0)x/3Mn(III)2x/3S Nanoparticles Supported on B/N-Codoped Mesoporous Nanocarbon as a Bifunctional Electrocatalyst of Oxygen Reduction/Evolution for High-Performance Zinc-Air Batteries. ACS Appl. Mater. Interfaces 2016, 8, 13348–13359. [CrossRef][PubMed] 14. Zhang, J.; Zhao, Z.; Xia, Z.; Dai, L. A metal-free bifunctional electrocatalyst for oxygen reduction and oxygen evolution reactions. Nat. Nanotechnol. 2015, 10, 444–452. [CrossRef] Materials 2020, 13, 2975 20 of 21

15. Zhan, J.; Cai, M.; Zhang, C.; Wang, C. Synthesis of mesoporous NiCo2O4 ﬁbers and their electrocatalytic activity on direct oxidation of ethanol in alkaline media. Electrochim. Acta 2015, 154, 70–76. [CrossRef]

16. Ma, C.Y.; Xu, N.N.; Qiao, J.L.; Jian, S.S.; Zhang, J.J. Facile synthesis of NiCo2O4 nanosphere-carbon nanotubes hybrid as an eﬃcient bifunctional electrocatalyst for rechargeable Zn–air batteries. Int. J. Hydrogen Energy 2016, 41, 9211–9218. [CrossRef] 17. Geng, D.S.; Ding, N.N.; Hor, T.S.A.; Chien, S.W.; Liu, Z.L.; Zong, Y. Cobalt sulﬁde nanoparticles impregnated nitrogen and sulfur co-doped graphene as bifunctional catalyst for rechargeable Zn-air batteries. RSC Adv. 2015, 5, 7280–7284. [CrossRef]

18. Ishihara, T.; Ishihara, T.; Yokoe, K.; Miyano, T.; Kusaba, H. Mesoporous MnCo2O4 spinel oxide for a highly active and stable air electrode for Zn-air rechargeable battery. Electrochim. Acta 2019, 300, 455–460. [CrossRef] 19. Wang, W.; Kuai, L.; Cao, W.; Huttula, M.; Ollikkala, S.; Ahopelto, T.; Honkanen, A.P.; Huotari, S.; Yu, M.;

Geng, B. Mass-Production of Mesoporous MnCo2O4 Spinels with Manganese (IV)- and Cobalt (II)-Rich Surfaces for Superior Bifunctional Oxygen Electrocatalysis. Angew. Chem. Int. Ed. 2017, 56, 14977–14981. [CrossRef] 20. Walsh, D.A.; Ejigu, A.; Muhammad, S.; Licence, P. The formation and role of oxide layers on Pt during hydrazine oxidation in protic ionic liquids. ChemElectroChem 2014, 1, 281–288. [CrossRef] 21. Endres, F.; Zein El Abedin, S. Air and water stable ionic liquids in physical chemistry. Phys. Chem. Chem. Phys. 2006, 8, 2101–2116. [CrossRef][PubMed] 22. Haerens, K.; Matthijs, E.; Chmielarz, A.; Bruggen, B.V.D. The use of ionic liquids based on choline chloride for metaldeposition: A green alternative? J. Environ. Manag. 2009, 90, 3245–3252. [CrossRef][PubMed] 23. Mainar, A.R.; Iruin, E.; Colmenares, L.C.; Kvasha, A.; Meatza, I.; Bengoechea, M.; Leonet, O.; Boyano, I.; Zhang, Z.; Blazquez, J.A. An overview of progress in electrolytes for secondary zinc-air batteries and other storage systems based on zinc. J. Energy Storage 2018, 15, 304–328. [CrossRef] 24. Ingale, P.; Sakthivel, M.; Drillet, J.-F. Test of Diethylmethylammonium Triﬂuoromethanesulfonate Ionic Liquid as Electrolyte in Electrically Rechargeable Zn/Air Battery. J. Electrochem. Soc. 2017, 164, H5224–H5229. [CrossRef] 25. Ge, X.; Liu, Y.; Goh, F.T.; Hor, T.A.; Zong, Y.; Xiao, P.; Zhang, Z.; Lim, S.H.; Li, B.; Wang, X.; et al. Dual-Phase

Spinel MnCo2O4 and Spinel MnCo2O4/Nanocarbon Hybrids for Electrocatalytic Oxygen Reduction and Evolution. ACS Appl. Mater. Interfaces 2014, 6, 12684–12691. [CrossRef]

26. Liu, W.; Bao, J.; Xu, L.; Guan, M.; Wang, Z.; Qiu, J.; Huang, Y.; Xia, J.; Lei, Y.; Li, H. NiCo2O4 ultrathin nanosheets with oxygen vacancies as bifunctional electrocatalysts for Zn-air battery. Appl. Surf. Sci. 2019, 478, 552–559. [CrossRef] 27. Vreese, P.D.; Skoczylas, A.; Matthijs, E.; Fransaer, J.; Binnemans, K. Electrodeposition of copper–zinc alloys from an ionic liquid-like choline acetate electrolyte. Electrochim. Acta 2013, 108, 788–794. [CrossRef] 28. Liu, Z.; Pulletikurthi, G.; Endres, F. A Prussian Blue/Zinc Secondary Battery with a Bio-Ionic Liquid–Water Mixture as Electrolyte. ACS Appl. Mater. Interfaces 2016, 8, 12158–12164. [CrossRef] 29. Kaur, A.; Bansal, S.; Chauhan, D.; Bhasin, K.K.; Chaudhary, G.R. The study of molecular interactions of aqueous solutions of Choline Acetate at diﬀerent temperatures. J. Mol. Liq. 2019, 286, 110878–110884. [CrossRef] 30. Yang, X.; Yu, X.; Yang, Q.; Zhao, D.; Zhang, K.; Yao, J.; Li, G.; Zhou, H.; Zuo, X. Controllable synthesis and

magnetic properties of hydrothermally synthesized NiCo2O4 nano-sphere. Ceram. Int. 2017, 43, 8585–8589. [CrossRef]

31. Baji, D.S.; Jadhav, H.S.; Nair, S.V.; Rai, A.K. Porous MnCo2O4 as superior anode material over MnCo2O4 nanoparticles for rechargeable lithium ion batteries. J. Solid State Chem. 2018, 262, 191–198. [CrossRef] 32. Vadivel, S.; Balaji, G.; Rathinavel, S. High performance ethanol and acetone gas sensor based nanocrystalline

MnCo2O4 using clad-modiﬁed ﬁber optic gas sensor. Opt. Mater. 2018, 85, 267–274. [CrossRef] 33. Ding, R.; Qi, L.; Jia, M.; Wang, H. Simple hydrothermal synthesis of mesoporous spinel NiCo2O4 nanoparticles and their catalytic behavior in CH3OH electro-oxidation and H2O2 electro-reduction. Catal. Sci. Technol. 2013, 3, 3207–3215. [CrossRef]

34. Mehrez, J.A.A.; Owusu, K.A.; Chen, Q.; Li, L.; Hamwi, K.; Luo, W.; Mai, L. Hierarchical MnCo2O4@NiMoO4 as free-standing core–shell nanowire arrays with synergistic eﬀect for enhanced supercapacitor performance. Inorg. Chem. Front. 2019, 6, 857–865. [CrossRef] Materials 2020, 13, 2975 21 of 21

35. Karkera, G.; Chandrappa, S.G.; Prakash, A.S. Viable Synthesis of Porous MnCo2O4/Graphene Composite by Sonochemical Grafting: A High-Rate-Capable Oxygen Cathode for Li–O2 Batteries. Chem. Eur. J. 2018, 24, 17303–17310. [CrossRef][PubMed] 36. Liu, H.; Jiang, L. Wettability by ionic liquids. Small 2016, 12, 9–15. [CrossRef] 37. Michael, M.S.; Kawamura, J.; Kuwata, N.; Prabaharan, S.R.S. Nanoengineered Lithium–Air Secondary Batteries: Fundamental Understanding and the Current Status of Development; Prabaharan, S.R.S., Michael, M.S., Eds.; Taylor & Francis: New York, NY, USA, 2013; pp. 89–103.

38. Sakthivel, M.; Bhandari, S.; Drillet, J.F. On Activity and Stability of Rhombohedral LaNiO3 Catalyst towards ORR and OER in Alkaline Electrolyte. ECS Electrochem. Lett. 2015, 4, A56–A58. [CrossRef]

39. Wang, L.; Gu, C.; Ge, X.; Zhang, J.; Zhu, H.; Tu, J. Highly Eﬃcient Bifunctional Catalyst of NiCo2O4@NiO@Ni Core/Shell Nanocone Array for Stable Overall Water Splitting. Part. Part. Syst. Charact. 2017, 34, 1700228–1700244. [CrossRef]

40. Xiao, C.; Li, Y.; Lu, X.; Zhao, C. Bifunctional Porous NiFe/NiCo2O4/Ni Foam Electrodes with Triple Hierarchy and Double Synergies for Eﬃcient Whole Cell Water Splitting. Adv. Funct. Mater. 2016, 26, 3515–3523. [CrossRef] 41. Song, M.K.; Park, S.; Alamgir, F.M.; Choand, J.; Liu, M. Nanostructured electrodes for lithium-ion and lithium-air batteries: The latest developments, challenges, and perspectives. Mater. Sci. Eng. R 2011, 72, 203–252. [CrossRef] 42. Liu, Z.; Zein El Abedin, S.; Endres, F. Electrodeposition of zinc ﬁlms from ionic liquids and ionic liquid/water mixtures. Electrochim. Acta 2013, 89, 635–643. [CrossRef]

43. Cai, W.-B.; Shi, Q.; Mansuetto, M.F.; Scherson, D.A. In situ Raman spectroscopy on an operating AA Zn-MnO2 battery under high discharge currents. Electrochem. Solid-State Lett. 2000, 3, 319–320. [CrossRef] 44. Ferraro, J.R. Low Frequency Vibration of Inorganic and Coordination Compounds; Springer: New York, NY, USA, 1971.

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