Pseudocapacitive Charge Storage in Single ... - ACS Publications

Subscriber access provided by UNIVERSITY OF THE SUNSHINE COAST

Article

Pseudocapacitive Charge Storage in Single-Step Synthesized CoOMnO- MnCoO Hybrid Nanowires in Aqueous Alkaline Electrolytes 2

2

4

Midhun Harilal, Syam G Krishnan, Asfand Yar, Izan Izwan Misnon, Mogalahalli Venkatashamy Reddy, Mashitah Mohd Yusoff, John Ojur Dennis, and Rajan Jose J. Phys. Chem. C, Just Accepted Manuscript • DOI: 10.1021/acs.jpcc.7b06630 • Publication Date (Web): 15 Sep 2017 Downloaded from http://pubs.acs.org on September 19, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

The Journal of Physical Chemistry C is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Pseudocapacitive Charge Storage in Single-Step Synthesized CoOMnO2-MnCo2O4 Hybrid Nanowires in Aqueous Alkaline Electrolytes Midhun Harilal,1 Syam G. Krishnan,1 Asfand Yar,2 Izan Izwan Misnon,1 M. Venkatashamy Reddy,3 Mashitah M. Yusoff,1 John Ojur Dennis,2 Rajan Jose1* 1

Nanostructured Renewable Energy Materials Laboratory, Faculty of Industrial Science &Technology, Universiti Malaysia Pahang, Kuantan, 26300 Pahang, Malaysia; 2Department of fundamental and applied sciences, Universiti Teknologi Petronas, Malaysia; 3Department of Materials Science & Engineering, National University of Singapore, Singapore *Corresponding author: [email protected] (R. Jose)

Abstract A new pseudocapacitive combination, viz. CoO-MnO2-MnCo2O4 hybrid nanowires (HNWs) is synthesized using a facile single step hydrothermal process and its properties are benchmarked with conventional battery-type flower shaped MnCo2O4 obtained by similar processing. The HNWs showed high electrical conductivity and specific capacitance (Cs) (1650 F g-1 or 184 mA h g-1 at 1 A g-1) with high capacity retention whereas MnCo2O4 nanoflower electrode showed only one third conductivity and one-half of its capacitance (872 F g-1 or 96 mA h g-1 at 1 A g-1) when used as a supercapacitor electrode in 6 M KOH electrolyte. The structure property relationship of the materials are deeply investigated and reported herewith. Using the HNWs as a pseudocapacitive electrode and commercial activated carbon as a supercapacitive electrode we achieved battery-like specific energy (Es) and supercapacitor-like specific power (Ps) in aqueous alkaline asymmetric supercapacitors (ASCs). The HNWs ASCs have shown high Es (90 Wh kg-1) (volumetric energy density Ev ~ 0.52 Wh cm-3) with Ps up to ~104 W kg-1 (volumetric power density Pv ~ 5 W cm-3) in 6 M KOH electrolyte, allowing the device to store an order of magnitude more energy than conventional supercapacitors.

1 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1. Introduction Advances in smart-electronics and electric vehicles call for energy storage devices capable of storing large amount of electric charges (termed as specific energy, Es), in miniaturized sizes, with high current (i) rating (termed as specific power, Ps). Most popular energy storage devices are lithium ion batteries (Es ~100 – 300 Wh kg-1; Ps ~200 W kg-1),1, 2 lead-acid batteries (Es ~60 Wh kg-1; Ps ~180 W kg-1),3 nickel-metal hydride batteries (Es ~80 – 100 Wh kg-1; Ps ~10 W kg-1),4 and supercapacitors (Es ~10 Wh kg-1; Ps ~500 – 1000 W kg-1).5 Recently, research on supercapacitor electrodes are getting increased attention due to their high Ps,6 however, their Es is much inferior to the above mentioned batteries. Supercapacitors are classified into electric double layer capacitors (EDLCs) and pseudocapacitors (PCs) based on charge storage mechanism. In the EDLCs, accumulation of charges occur at the electrode – electrolyte interface (non-faradic), similar to the conventional electrolytic capacitors, whereas the charges move across the interface (faradic) in PCs. Because of the non-faradic charge accumulation, EDLCs offer orders of magnitude higher Ps than PCs. On the other hand, the Es of PCs are much higher than that of EDLCs because they involve ~2.5 – 10 electrons per surface atom of active material in the charge storage mechanism compared to ~0.17 – 0.20 electrons per surface atom in the EDLCs. This limits the practically achieved capacitance in EDLC (~20 – 50 µF cm-2) whereas in pseudocapacitors it is orders of magnitudes higher (~1 F cm-2).7,

8

One of the most successful ways to develop a

supercapacitor offering high Es and Ps is making an asymmetric configuration with one of the electrode to be EDLC and the other to be PC; they are called asymmetric supercapacitors (ASCs). The enhanced Cs of the PC electrode and larger potential window of ASCs help to improve their Es than symmetric supercapacitors employing EDLCs as both electrodes. Preliminary criteria for selection of a pseudocapacitive electrode for ASCs are (i) electrochemical reversibility with a chosen electrolyte; (ii) a range of oxidation states to accommodate many electrons; (iii) a high specific surface area enabling larger electrode – 2 ACS Paragon Plus Environment

Page 2 of 37

Page 3 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


electrolyte interface; and (iv) high electrical conductivity for improved charge–discharge rate.9 The conditions (ii) and (iii) could be related to Es, (iv) to Ps, and (i) to cycling stability.10 High surface area and improved electrical conductivity nominates nanowires as the preferred energy storage medium; many ceramic nanowires show improved Es and Ps than their particulate analogues.9, 11 The most common pseudocapacitive materials are RuO212 and MnO213 in neutral electrolytes owing to the condition (ii) as well as surface related charge storage. The cyclic voltammogram (CV) of these materials do not show an oxidation and reduction events in the chosen electrolytes. On the other hand, materials such CuO, Co3O4 and NiO have been investigated for supercapacitor energy storage applications; however, the oxidation peaks in CV lead to characterize as battery-type.12,

14

Recently, we

have reported composite nanowires in the CuO/Co3O4 and NiO/Co3O4 systems to combine the condition (ii) and (iv) with superior charge storage properties than their constituent binary oxides; however, resulting materials are still battery-type due to the presence of redox peaks in the CV.15, 16 Many attention has been focused on spinel type MCo2O4 (M = Mn, Ni, Cu, Zn, Mg) materials due to cobalt economy and high theoretical charge storability (See Supporting Information, Fig. S1) than binary metal oxides.17-22 However, as has been discussed recently they are battery-type due to oxidation and reduction events in CV.

14, 15

Although MnCo2O4 has high theoretical charge storability its practically achieved value is less impressive than other materials in this group (See Table 1). One of the possible reasons for the battery-type behavior and inferior charge storability is low electrical conductivity of MnCo2O4 (i.e., the condition (iv)); therefore, increasing electrical conductivity could be a remedy. We note that CoO has higher conductivity (See Supporting Information, Table S1) than MnCo2O4; therefore, a composite nanowire of MnCo2O4 and CoO would increase the conductivity and overcome the bottlenecks. There have been several reports on MnCo2O4based supercapacitor electrodes synthesized through various methods; a brief summary of which is given in Table 1. In general, hydrothermal synthesis has been versatile to produce diverse morphologies by controlling the precursors or reaction conditions. Besides, the 3 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 37

electrodes materials made using this method showed high CS. However, so far, the process employed stoichiometric precursor mixture and produced stoichiometric compounds (See Supporting Information, Table S2). Growth of MnCo2O4 from nonstoichiometric precursors and their capacitive charge storage would likely to result in materials with unprecedented properties and mechanisms. We have now synthesized CoO-MnO2-MnCo2O4 hybrid nanowires (HNWs) through a single step hydrothermal process using 1:1 molar ratio of precursors, which contain MnCo2O4 nanowires as the major phase (>97%) and CoO (>1%) and MnO2 (>1%) nanoparticles as minor phases. Structural and electrochemical properties of these ternary–binary HNWs are studied in detail and compared with that of pure MnCo2O4 in this article. ASCs fabricated using the HNWs as anode and commercial activated carbon as cathode in 6 M KOH showed nearly an order of magnitude higher specific energy (90 W h kg-1) than that of conventional supercapacitors (< 10 W h kg-1) while maintaining supercapacitive specific power (~104 W kg-1). The results show the superiority of the HNWs over the existing ones and this procedure of developing many components in a single nanowire could be the way forward for high performing and safe energy storage devices as well as other functional devices.

2. Experimental details 2.1. Material Synthesis: A facile hydrothermal method (HTM) was adopted for synthesizing both MnCo2O4 and HNWs. We used MnCl2.4H2O, CoCl2.6H2O, hexadecyl trimethyl ammonium bromide (HTAB) and urea as starting chemicals (all received from Merck & Co.). For synthesis of HNWs, a solution of 2 mM HTAB and 9 mM urea in 100 ml de-ionized water was prepared by stirring for 30 minutes. 1:1 molar ratio of MnCl2.4H2O and CoCl2.6H2O (5 mM of CoCl2.6H2O and 5 mM of MnCl2.4H2O) were then added into the above solution and stirred vigorously for 2 hours until a clear solution was obtained. The solution was then transferred 4 ACS Paragon Plus Environment

Page 5 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


to a stainless steel autoclave and placed in a furnace at 100oC for 12 h. The furnace was subsequently cooled to room temperature at 1oC per minute ramping and the reaction product was removed from the reactor for washing. Careful washing process followed where the product was initially ultra-sonicated in ethanol for 20 minutes and filtered by centrifugation at 2000 rpm (repeated 2 times). Then the product is mixed with de-ionized water and the ultra-sonication and centrifugation process is repeated in similar way as before. The filtrate was subsequently dried at 60oC and calcined at 450oC in air for 3 hours at a ramping rate of 1oC per minute. To synthesize MnCo2O4 nanoflowers, stoichiometric amounts of MnCl2.4H2O and CoCl2.6H2O were used and all the other procedures were maintained similar to the synthesis of the HNWs. 2.2. Structural, Morphological and Electrochemical measurements: Crystal structure of the materials were studied by X-ray diffraction (XRD) (Rigaku Miniflex II X-ray diffractometer, CuKα radiation, λ=1.5406 Å), high resolution transmission electron microscopy (HRTEM), and selected area electron diffraction (SAED) (FEI, Titan 80–300 kV). X-ray photoelectron spectroscopy (XPS) (Thermo scientific, K-Alpha, USA: operating with an X-ray source of Al-Kα radiation) and Energy-dispersive X-ray spectrometry (EDX) (7800F, FESEM, JEOL, USA) analysis was carried out to determine composition of HNWs and MnCo2O4. The XPS survey spectra were recorded in the range 0 – 1000 eV at a pass energy of ~200 eV with a resolution of ~1 eV maintaining at a pressure of ~10-8 Torr; high-resolution spectra were recorded with smaller constant pass energy of ~50 eV with a resolution of ~0.1 eV. Charge referencing was carried out against adventitious carbon, assuming its binding energy at 284.8 eV. The spectra were analysed using Origin 9.0 by fitting the high resolution spectra into multiple Gaussian curves; the baseline was modelled by adjacent averaging. Morphology of the material was analysed using scanning electron microscopy (7800F, FESEM, JEOL, USA) and TEM. To analyse the unreacted precursor concentration in the left over hydrothermal solution, inductively coupled plasma 5 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 37

mass spectroscopy (ICP-MS) (Agilent, 7500 series, USA) was used. The porosity and specific surface area analyses of the samples were performed using Brunauer–Emmett–Teller (BET) method and Barrett–Joyner–Halenda (BJH) equation, respectively, using ASAP 2020 (Micromeritics, USA) surface area analyser. 2.3. Electrochemical Studies: The electrodes for electrochemical studies were prepared on nickel foam substrates, which were cleaned by etching in 1M HCl for 20 min, degreasing in acetone and washing in ethanol and de-ionized water. In a typical experiment, HNWs powder was mixed with carbon black (Super P conductive, Alfa Aesar, UK) and polyvinylideneﬂuoride (PVDF) (Sigma Aldrich, USA) in the ratio 75:15:10. N-methyl-2-pyrrolidinone (NMP), which works as a homogeniser, was added to the above mixture and sonicated for 24 h. The semi-fluidic slurry thus obtained was pasted on the pre-cleaned nickel substrate in an area of ~1 cm2 and dried at 60oC. The dried electrode was then pressed at ~5 ton using a hydraulic press. Mass-loading of the active material in the electrode was ~2.4 mg cm-2, which is higher than that used in majority of reports although electrodes showing up to ~4 times more mass-loading are reported (Table 1). The electrochemical properties of the electrodes were determined in 6 M KOH in three-electrode system configuration employing a saturated Ag/AgCl electrode and a platinum wire as the reference and the counter electrodes, respectively. Cyclic voltammetry (CV), charge-discharge cycling (CDC), electrochemical impedance spectroscopy (EIS) measurements were carried out using potentiostat-galvanostat (PGSTAT M101, Metrohm Autolab B.V., Netherlands) employing NOVA 1.9 software. The EIS measurements were carried out in the frequency range 100 kHz–0.01 Hz at the open circuit potential of the respective electrodes. Linear sweep voltammetry (LSV) at a sweep rate of 10 mV s-1 (potential range 0.1 to 0.35) was used to measure electrochemical resistance of the electrodes.

6 ACS Paragon Plus Environment

Page 7 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Asymmetric supercapacitors (ASC) were fabricated using the HNWs (and MnCo2O4) and activated carbon (AC) as anode and cathode, respectively, which were separated by a glass microfiber, filter (fioroni). The electrolyte was 6 M KOH. Charge storage properties of the ASCs were studied using the PGSTAT M101 potentiostat-galvanostat.

3. Results and discussion The FESEM images of the as-prepared stoichiometric and off stoichiometric powders are shown in the Supporting Information [Fig. S2 (a & b)]. It can be seen that the stoichiometric compound is formed as flowers while the reaction with precursors in the 1:1 molar ratio resulted in nanowire morphology. Role of increased precursor concentration in activating a wire-like morphology has been well discussed in literature.23, 24 Upon annealing them, flower-like pure MnCo2O4 (MnCo2O4) [Fig. 1 (a & b)] and HNWs [Fig. 1 (c & d)] are formed. Although, HNWs are mostly wire shaped with diameter in the ~100 –150 nm range, particle aggregates can be clearly seen from Fig. 1 (c). A higher magnification image of typical HNWs [Fig. 1 (d)] shows randomly oriented nanowires of smaller diameter in the 10 – 50 nm range and the nanoparticle aggregates formed from the excess precursor. TEM analysis was done to further characterize HNWs. The particle packing and morphology of HNWs are more obvious in the bright field TEM image shown in Fig. 1 (e). Particle packing appears to be tight, however, many bright regions composing small particles ( 97%), CoO (> 1%) and MnO2 (> 1%) (See Supporting Information, Section S1). As the scattering area (volume) of the XRD is ~0.25 mm2 (0.05 mm3), it is generally considered as a phase characterization tool for bulk material; therefore, high resolution TEM and electron diffraction are superior to determine phase distribution from a smaller segment. Selected area electron diffraction (SAED) patterns from two regions of the HNW specimen are in Fig. 2 (b & c), which shows a collection of bright and weak spots with random orientation (typical for polycrystalline and multiphase materials). The bright spots in SAED, which corresponds to the highly crystalline phases, could be indexed to MnCo2O4 as expected from XRD results. The weak spots, corresponding to minor peaks of XRD, were indexed to MnO2 and CoO phases. Similar observations could be drawn from the HRTEM images also. Fig. 3 (a) is the HRTEM image of a typical powder particle showing collection of larger particles with lesser active surface and a large distribution of smaller crystallites providing larger active surface. As the supercapacitive properties are surface related, presence of such smaller particles with large area is suggested to be ideal in achieving high Cs in supercapacitor electrodes.26 While the well-defined regions of HRTEM in Fig. 3 (a – d) could be indexed to MnCo2O4, the other phases could be found in the semi-amorphous regions. Interestingly, the crystalline boundaries also found to possess secondary phases such as highly conducting CoO, which would benefit the charge dynamics in the electrode material. Thus, calcined material is a combination of many phases, all of which are actively considered for high performance supercapacitor electrode with unique benefits. For example, the CoO offers high capacitance27 due to its high electrical conductivity whereas MnO2 is one of the widely investigated pseudocapacitive materials. The surface atomic structure of the HNWs and MnCo2O4 were analyzed using XPS. Fig. 4 (a) shows the survey scan spectrum of the HNWs compared with MnCo2O4, featuring peaks of Co 2p, Mn 2p, O 1s, C 1s, Co 3s and Co 3p. The detailed Co 2p spectrum [Fig.4 (b)] consists of peaks at ~780 eV of Co 2p3/2 and ~795 eV of Co 2p1/2, indicating Co3+ and Co2+ 8 ACS Paragon Plus Environment

Page 9 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


oxidation states for MnCo2O4.28 One satellite peak each near Co 2p3/2 and Co 2p1/2 were observed which corresponds to oxidation state for CoO.29 The Mn 2p spectrum [Fig.4 (c)] displays two distinct peaks at binding energies ~642 and ~653 eV, corresponding to Mn 2p3/2 and Mn 2p1/2, respectively, fitted to Mn2+ and Mn3+ cations.30 As shown in Fig. 4 (d) the Co 3p, Mn 3s and Co 3s peaks were at binding energies ~61, ~83 and ~103 eV, respectively. Co 3p and Co 3s along with a satellite peak at ~72 eV indicate Co2+ oxidation state of CoO,31, 32 whereas Mn 3s along with the satellite peak confirms presence of MnO2.33 A summary of all the observed elements and their binding energies are in the Table S3 (See Supporting Information, Table S3). Fig. S3 shows the XPS detailed spectrum of Co 2p phase in pure MnCo2O4, which consists of two broad peaks ~780 and ~800 eV (See Supporting Information, Fig. S3). These peaks correspond to Co 2p3/2 and Co 2p1/2 indicating Co3+ and Co2+ oxidation states. No peaks corresponding to CoO phase was observed in the spectrum of MnCo2O4. Moreover, peaks corresponding to Co 3p, Co 3s (CoO phase) and Mn 3s (MnO2 phase), which were present in the hybrid system, were also absent in the pure phase [Fig.4 (a)]. The area of the peaks corresponding to CoO and MnO2 phases in the hybrid material was ~3%, which is consistent with XRD analysis. EDX analysis was also conducted to determine the percentage of composition of elements in the synthesized materials. The results indicated marginal increase in elemental composition of Mn and Co in the hybrid material confirming presence of traces of CoO, and MnO2 (See Supporting Information, Fig. S4 and Table S4). Thus, all these characterizations leading us to conclude that we have a nanowire system containing MnCo2O4, CoO, and MnO2 in the hybrid system. Thus the XRD, XPS and EDX data indicate that a considerable amount of Mn precursor would have left unreacted in the reaction that lead the HNWs. To quantify the leftover precursor, ICP-MS analysis was conducted using the supernatant solution after the hydrothermal reaction (See Supporting Information, Section S2). Approximately 2.4 mM of 9 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 37

manganese precursor and 0.2 mM of cobalt precursor was found unreacted in the leftover solution after hydrothermal treatment, which means the reactant quantities were 2.6 mM of Mn and 4.8 mM of Co precursors. Initially, stoichiometric amounts of the Mn and Co precursors (2.2 mM manganese precursor and 4.4 mM of cobalt precursor) react to form the major phase in the product MnCo2O4 according to reactions (1) & (2) through formation of MnCo2(OH)6. Subsequently, the left over 0.4 mM of the cobalt precursor and 0.4 mM of the manganese precursor react to form CoO and MnO2 through formation of MnCo2(OH)4 as in reaction schemes (3) & (4). This mechanism has been speculated from the reaction scheme reported for MnCo2O4.34 The proposed mechanism is schematically shown in Fig. 5. 2 Co + Mn + 6 OH → MnCo OH

(1)

MnCo OH + O → MnCo O + 3H O

(2)

Co + Mn + 4OH → MnCoOH

(3)

MnCoOH + O → MnO + CoO + 2H O

(4)

The crucial role of the 1:1 precursor mixture for the formation of wire plus particle decorated HNWs was further experimented. Our experiments using 2.6 mM of Mn and 4.8 mM of Co precursors did not yield wire morphology; the resulting material neither showed a regular morphology nor gave impressive charge storage properties (See Supporting Information, Fig.S5). The materials were further characterized by nitrogen adsorption and desorption isotherms to probe the BET surface area and porosity [See Supporting Information, Fig. S6 (a & b)]. The isotherm of HNWs, classified as type IV, indicates the presence of a mesoporous structure. The HNWs shows a BET surface area of ~53 m2g-1, whereas that of MnCo2O4 is ~46 m2g-1. According to BJH pore size distribution data [See Supporting Information, inset Fig. S6 (a)], the average pore size of HNWs is ~20 nm and that of MnCo2O4 is ~15 nm. 10 ACS Paragon Plus Environment

Page 11 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Because the surface area and pore size of both samples do not appreciably differ, variation in electrochemical properties of HNWs is less likely related to these factors. Electrochemical measurements of HNWs and MnCo2O4 electrodes were carried out in 6 M KOH electrolyte. Fig. 6 shows the LSV curves of HNWs and MnCo2O4 in 6 M KOH electrolyte, from which their electrochemical resistances could be compared. The LSV of HNWs is a straight line in the entire scan range indicating high conductivity; and showed an electrochemical resistance of 2.1 Ω. On the other hand, MnCo2O4 is electrically resistive until a knee voltage of ~0.27 V after which it showed high conductivity. The resistance calculated at this voltage shows that HNWs is 3 times more conducting than MnCo2O4. This low electrochemical resistance is in line with the other characteristic resistances observed by CDC and EIS, which will be discussed later in this article. Fig. 7 (a), and Fig. S7 (a) displays the cyclic voltammetry (CV) curves of HNWs and MnCo2O4 electrodes at a scan rate of 30 mV s-1 and the CV curves of HNWs at different scan rates, respectively. While the CV curves of HNWs suggested a pseudocapacitive behaviour, redox peaks were observed in the CV of MnCo2O4 characterizing battery-type behaviour. Although both materials show similar voltage window, the current density and total area enclosed by the CV curve is much higher for HNWs than MnCo2O4 electrodes. The redox reactions (shown in schemes 5, 6) occurring in HNWs electrodes resulting in its chargedischarge properties could be written as 35: 2CoOOH + MnOOH + e– ↔ MnCo2O4 + H2O + OH–

(5)

MnO2 + H2O + e– ↔ MnOOH + OH–

(6)

In the above, we assumed that the minor phases of cobalt do not contribute much to the redox reaction. The specific capacitance, Cs (F g-1) of both electrodes was estimated from the cathodic or anodic part of the CV and its variation with scan rate is detailed in Fig. 7 (b). The Cs values calculated from CV at 2 mV s-1 were 845 and 1690 F g-1 for MnCo2O4 and HNWs respectively. i.e., the capacitance of the hybrid material was ~2 times higher than that of the 11 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 37

stoichiometric compound. Interestingly, while the Cs increased by a factor of ~2 in HNWs than the pure compound the rate-capability (observed from the variation in Cs with scan rate) is similar in the two electrodes. This is probably due to their similar pores sizes; however, the molecular mechanism for this discrepancy is not clear to us now. To address the source of the two-fold increase in capacitance, the quantity of electrochemically active sites in the electrodes accessed by the solvated ions (n) in the electrolyte, which varies with scan rates, was determined by modifying the reactions in (5) as MnCo O + 3nOH ↔ nMnOOH + 2nCoOH + 1 − nMnCo O + ne (7)

‘n’ can be calculated using equation (8),36, 37

n =

CS × m × ∆V F×Z

(8)

where ‘m’ is the molecular weight, ∆V is the redox potential, F is the Faraday’s constant and Z is the oxidation state of electrode material. The n of MnCo2O4 and HNWs at 2 mV s-1 were ~20 and ~36%, respectively. Nearly two-fold increase in n of HNWs is expected to arise from the presence of the high conductivity of CoO (~10-1 S m-1), presence of semi-amorphous regions,26 and morphological differences. The pure phase, due to its bulky morphology, is expected to have poorer electrode – electrolyte surface interface and could deteriorate the electrochemical performance.38, 39 The reversibility of the electrodes, rationalized in terms of coulombic efficiency, was evaluated from the CV data by calculating the ratio between the areas of the anodic and cathodic cycles. The coulombic efficiency was ~99% for both MnCo2O4 and HNWs; indicating that all electrode types have excellent reversibility over the selected number of scans. Now the significant question is whether there is a difference in charge storage mechanism in the two types of electrodes. In pseudocapacitive materials, the voltammetric current (i) and scan rate (v) dependence is analysed to determine whether the origin of 12 ACS Paragon Plus Environment

Page 13 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


capacitance is from bulk diffusion or from surface redox reaction.40 If the current arises from bulk intercalation process, it follows diffusion kinetics and the current i varies as v1/2 (batterytype). On the contrary, if the current arises from surface charge storage processes (pseudocapacitance), a linear relation with v can be observed. As can be observed from Fig. 7 (c), HNWs show a linear relation of current with scan rate ( ∝ ) while MnCo2O4 doesn’t, suggesting dominance of surface charge storage processes in HNWs. MnCo2O4 shows a linear relationship of current with square root of scan rate ( ∝ √ [See Supporting Information, Fig. S8 (a)], indicating diffusion kinetics’ dominance and is classified as a battery-type electrode.41 For further verification of surface electrochemical mechanism, scan rate dependence plots of Cs vs. v-1/2 and Cs-1 vs. v1/2 was drawn [See Supporting Information, Fig. S8 (b & c)]. The surface contribution to capacitance can be extracted at = ∞, i.e. at / = 0 whereas the total capacitance can be extracted at / = 0 using a linear fit. Comparing surface contribution with the total Cs, ~66% of capacitance in HNWs is from the surface (pseudocapacitance thereby) whereas the stoichiometric compound (MnCo2O4) shown only ~26% (battery-type). Brousse et al

12

further suggest that CV of the asymmetric

configuration is more accurate to see whether the electrode is pseudocapacitive or not. The CV of the ASCs [Fig. 8 (b), to be discussed later] confirms the surface dominance and pseudocapacitive nature of HNWs over the other electrode. We could identify only one source for this shift in charge storage mechanism from battery-type to pseudocapacitive, which is the improved electrical conductivity of the HNWs as shown in Figure 6 due to the presence of highly conductive CoO. Charge-discharge cycling (CDC) provides a method to measure the total resistance and practically achievable capacitances of the electrodes. Fig. 7 (d) compares the discharge curves of the electrodes at a galvanostatic current density of 1 A g-1 for the HNWs and MnCo2O4 electrodes. The discharge curves of pseudocapacitor electrodes are usually as a result of three processes: (i) an abrupt initial potential drop due to the surface contribution of 13 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 37

the PC followed by (ii) a slow decay of potential through deep intercalation, and (iii) a faster voltage drop corresponding to EDLC processes.9,

42

Thus, the non-linear shape of the

discharge curves reveal that a major contribution of the capacitance of the electrodes is from faradic reactions. The HNWs and MnCo2O4 electrodes show a maximum potential of ~0.4 V in the 6M KOH electrolyte. The discharge time of HNWs is >2 times that of MnCo2O4, indicating the capacitance enhancement. Another valuable parameter obtained from CDC is the equivalent electrical resistance (ESR), which could be measured from the potential drop (VIR) between the charge and discharge curves. Three factors contribute to ESR such as (i) intrinsic resistance of the electro-active material, (ii) electrolyte resistance, and (iii) the contact resistance between the active material and the current collector. The ESR of HNWs and MnCo2O4 calculated from the ratio of VIR to the corresponding discharge current (ID) are 1.4 and 3.6 Ω, respectively. The ESR of the HNWs electrodes is less than half of that of MnCo2O4, which is expected to arise from (i) the presence of highly conductive CoO and (ii) difference in morphology of the two materials. The morphological difference of the active materials would lead to a different percolation of the conductive carbon at the conductive carbon-active material interface thereby resulting in different conductivity of the electrode.43, 44

The charge discharge curves at 1 A g-1 and discharge curves of the HNW electrode at different current densities were also measured [See Supporting Information, Fig. S7 (b & c)]. The variation of capacitance with current density is plotted [See Supporting Information, Fig. S7 (d)]. A current density dependency of capacitance is seen with the HNW electrode for lower values, but there is little influence for current densities as it goes higher. This stable performance could be ascribed to the lower ESR of the HNWs. Table 2 summarises the electrochemical parameters of both materials from CV and CD analyses. The charge kinetics at the electrode – electrolyte interface studied using the EIS measurements confirmed fast ion transport responsible for superior capacitive performance in 14 ACS Paragon Plus Environment

Page 15 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


HNWS. Fig. 7 (e) shows EIS Nyquist spectra of both electrodes fitted using a standard Randles circuit [inset of Fig. 7 (e)]. The equivalent circuits for both electrodes consists of a series and parallel combination of device bulk resistance (RS), charge transfer resistance (RCT), electric double layer capacitance (Cd), Warburg impedance (ZW), and a constant phase element (CPEPC). The HNWs electrodes showed lower resistances than MnCo2O4 corroborating results from CDC analysis. The charge kinetic parameters determined from the circuit for HNWs are RS = 1.4 Ω, RCT = 1.5 Ω, Cd = 1.7 mF, ZW = 225 mMho and ZCPE = 38.5 (mFs)1/n (n = 0.94). This lower value of RCT for HNWs electrode, which is approximately one third of that of MnCo2O4 electrode (Table 3), facilitates improved ionic conduction through their pores over the other material thereby providing higher capacitance. The variation in resistance value is consistent with the difference in morphologies of the two materials as shown before. Another important parameter obtained from EIS studies is the characteristic frequency (fo) at which the circuit is equally capacitive and resistive. The phase angle is 90o for an ideal capacitor and 45o when the capacitive and resistive impedances are equal. A phase angle of 90o is typically observed only in EDLCs using carbon materials whereas deviations from 90o are observed in pseudocapacitors.45 The phase angles of the HNWs and MnCo2O4 compared in Fig. 7 (f) show the large difference in capacitive behaviours of the two electrodes. The fo is ~3.9 Hz in HNWs, which is much higher than the conventional activated carbon (~0.05 Hz),46 whereas that of MnCo2O4 is ~0.04 Hz. Thus, the HNWs are capacitive over a wider range of frequencies, which usually observed in highly conductive carbon structures such as graphene (fo ~30 Hz).47 The time constant corresponding to fo, termed as relaxation time constant (τ), are ~0.25 and ~25 s for HNWs and MnCo2O4, respectively. This rapid frequency response of the HNWs electrode would provide superior cycling stability and specific power in practical devices.48 Similarly the electrochemical performance of commercially available activated carbon (AC) – based electrodes was also evaluated in a three-electrode configuration. Nearly 15 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 37

rectangular shaped CV curves were obtained for AC electrodes at different scan rates in 6 M KOH aqueous electrolyte indicating the mode of charge storage to be EDLC. The Cs of the AC is calculated to be 151 F g-1 at 1 A g-1 [See Supporting Information, Fig. S9 (a & b)]. To address the contribution of nickel foam substrates to the capacitance of electrode materials,49 CV study of pure nickel foam in 6M KOH was conducted (See Supporting Information, Section S3). The area of curve obtained was 100 times lesser than the area of HNWs CV, thereby ruling out the possibility of contribution of Ni foam to the actual device capacitance. In order to examine whether the electrode properties of the HNWs could be practically realized, ASCs were fabricated using HNWs and MnCo2O4 as anodes and activated carbon as cathode. For obtaining maximum effective capacitance, electrodes of both ASCs should have similar capacitance values.50 Similarity in capacitance is followed by a charge balance (as charge q = CV) q+ = q- ; where q+ refers to charge stored at the anode and q- to that at the cathode.51, 52 The accumulated charge over each electrode is given by # = C $ × ΔV × m , where ∆V is the potential window. From this equation (9) for the electrode mass for optimal performance follows: )*

)+ =

,- ./0,× 12+ ,- 3,× 12*

(9)

Mass ratio for ASC fabrication materials was optimized on the basis of the Cs values and potential windows of the PC electrodes and was calculated to be ~0.40 and ~0.30 for MnCo2O4//AC and HNWs//AC devices, respectively. Five ASCs were fabricated with cathode mass-loadings up to 4.2 mg and anode mass-loadings up to 2.6 mg with consistent results. Fig. 8 (a) shows the comparative CV curves of HNWs//AC and MnCo2O4//AC at 10 mV s-1. Although the CV curves of HNWs//AC [Fig. 8 (b)] device deviate considerably from a rectangular shape exhibited by EDLC they are much improved than those reported in literature for ASCs.9, 53 The maximum achieved voltage for both ASC devices were ~1.6 V, 16 ACS Paragon Plus Environment

Page 17 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


but the cathodic and anodic currents and total area of the CV curves were much higher for HNWs//AC, indicating the superiority over the other. Galvanostatic charge – discharge analysis were conducted at varying current densities for further electrochemical performance evaluation. The CD cycles of HNWs//AC and MnCo2O4//AC at 1 A g-1 were drawn [See Supporting Information, Fig. S10 (a & b)]. The potential window was similar to that observed from CV, while an increased discharge time was achieved in HNWs//AC [Fig. 8 (c)] compared to MnCo2O4//AC. Near linear shape of the discharge curves indicates the potential independent nature of the pseudocapacitive reactions.7 Fig. 8 (d) shows capacitance variation with current density of MnCo2O4//AC, HNWs//AC and activated carbon electrodes. HNWs//AC clearly yield much higher capacitance (more than twice as that of the nearest contender, MnCo2O4//AC) than others. The operational stability of the HNWs//AC was examined by CDC testing at (i) a constant current density (5 A g-1) and (ii) at varying current densities (2 – 10 A g-1). A capacitive retention of ~97 % and coulombic efficiency of ~98 % were exhibited by HNWs//AC at the end of the 5000 cycle test program [Fig. 8 (e)] at 5 A g-1. The rate capability of the device is shown in the inset of Fig. 8 (e), which was evaluated by CDC at current densities of 2, 5, and then 10 A g-1. As can be observed from the figure the device maintains a steady capacitance at different current densities. Cycling was continued for 3750 cycles at these different current densities and was brought back to 2 A g-1 for the final 1250 cycles and the device capacitance was found to decrease only by 3%. The operational stability of MnCo2O4//AC was also examined in a similar way and is shown in Fig. S11. At the end of 5000 cycle test program, the capacitance value of MnCo2O4//AC was found to decrease by 6% in comparison to the decrease of 3% for HNWs//AC. The superior operation stability of HNWs//AC was evident form this comparative study. Fig. 8 (f) shows the Nyquist plots of the ASCs determined by EIS in the frequency range 0.01 Hz–10 kHz at an open circuit potential in 6 M KOH electrolyte. Clearly the HNWs based devices show superior properties in all aspects. The 17 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 37

ultimate result of interest for an energy storage device is specific energy and power.54 The gravimetric (Es, Ps) and volumetric (Ev, Pv) densities were calculated using the equations shown in Supporting information section S4. The HNWs//AC system delivered an Es of 90, 64, 54 and 41 Wh kg-1 at Ps values of 797, 1634, 4418 and 8682 W kg-1, respectively. Comparatively, the performance of the MnCo2O4//AC and control carbon-based EDLC was much inferior: Es values of 32, 21, 14 and 12 Wh kg-1 at Ps values of 842, 1750, 4516 and 10000 W kg-1 ; Es values of 9.3, 8.8, 8.3 and 6.9 Wh kg-1 at Ps values of 357, 696, 1410 and 3520 W kg-1, respectively. These achieved Es and Ps values of HNWs ASC are compared with that of other metal oxides as shown in Fig. 9 (a). Fig. 9 (b) shows the Ragone plot for the two devices under study in terms of volumetric energy and power densities. As observed from the Ragone plot, the HNWs ASC has much higher specific energy than the MnCo2O4 ASC. However, we note that other studies have been reported with comparative and higher specific energy than the present one. But most of these studies utilises graphene, MWCNTs or ionic electrolytes to achieve this result.55-58 The high value of ES was achieved either by utilizing the voltage window achievable by the particular electrode – electrolyte system or by increasing conductivity and surface area through MWCNTs or graphene addition. High cost, toxicity and rigorous device manufacturing requirements in air free atmosphere hamper them from large scale fabrication. On the other hand, the present achievement is by tailoring the capacitive properties of the electrode; the HNWs ASCs are expected to perform even better than the currently achieved ES and PS by increasing practically achievable potential window. This unique property will allow the ASC to store more energy than the conventional supercapacitor while charging it at a faster rate.

Conclusions In conclusion, employing a simple synthetic strategy we engineered hybrid nanowire in the Mn–Co oxide system with desirable properties as a high performing pseudocapacitive electrode than a stoichiometric battery-type MnCo2O4 electrode. The hybrid nanowire 18 ACS Paragon Plus Environment

Page 19 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


contains MnCo2O4 as major phase (>97%) decorated by nanoparticles of highly conducting metal oxides (CoO >1%) and popular pseudocapacitive material (MnO2 >1%). Cyclic voltammetry, charge-discharge cycling, and electrochemical impedance spectroscopy measurements show that the hybrid nanowire electrode store more charges on account of its lower electrochemical series resistance, lower charge relaxation time, and lower charge recombination time than the stoichiometric compound. The hybrid nanowire electrodes showed high specific capacitance of 1650 F g-1 (at 1 A g-1) whereas the pure MnCo2O4 electrode showed only 50% of it (872 F g-1 at 1 A g-1). Also, ~70% of the capacitance in the hybrid system is surface related and could offer superior specific energy and power. On the other hand, the stoichiometric compound showed only ~26% capacitance to be surface related. A working supercapacitor fabricated using the hybrid nanowire (anode) and commercial activated carbon (cathode) in an aqueous alkaline electrolyte (6 M KOH) showed nearly an order of magnitude higher specific energy (90 Wh kg-1) than that of conventional supercapacitors (

Pseudocapacitive Charge Storage in Single ... - ACS Publications

Recommend Documents