Kirkendall Growth and Ostwald Ripening Induced Hierarchical

Subscriber access provided by Washington University | Libraries

Energy, Environmental, and Catalysis Applications

Kirkendall growth and Ostwald ripening induced hierarchical morphology of Ni-Co LDH/MMoSx (M = Co, Ni, and Zn) heteronanostructures as advanced electrode materials for asymmetric solid-state supercapacitors Syam Kandula, Khem Raj Shrestha, Gaddam Rajeshkhanna, Nam Hoon Kim, and Joong-Hee Lee ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.9b02978 • Publication Date (Web): 06 Mar 2019 Downloaded from http://pubs.acs.org on March 6, 2019

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 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 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.

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 41 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

ACS Applied Materials & Interfaces

Kirkendall growth and Ostwald ripening induced hierarchical morphology of NiCo LDH/MMoSx (M = Co, Ni, and Zn) heteronanostructures

as

advanced

electrode

materials

for

asymmetric solid-state supercapacitors Syam Kandula†, Khem Raj Shrestha†, Gaddam Rajeshkhanna†, Nam Hoon Kim*,†, Joong Hee Lee**,†, ‡ †Advanced

Materials Institute for BIN Convergence Technology (BK21 plus Global Program),

Department of BIN Convergence Technology, Chonbuk National University, 567, Baekje-daero, Deokjin-gu, Jeonju-si, Jeollabuk-do, 54896, Republic of Korea. ‡Carbon

Composite Research Centre, Department of PolymerNano Science and Technology

Chonbuk National University, Jeonju, Jeonbuk 54896, Republic of Korea. Corresponding authors:

* Prof. N. H. Kim (E-mail: [email protected])

** Prof. J. H. Lee (E-mail: [email protected]) KEYWORDS: Kirkendall growth, Ostwald ripening, Ni-Co LDH/MMoSx heteronanostructures, solid-state supercapacitors, specific energy.

ABSTRACT

ACS Paragon Plus Environment

1

ACS Applied Materials & Interfaces 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 2 of 41

By changing the mixed metal sulfide composition, morphology tuning of the active electrode material can be possible, which can show a huge impact on its electrochemical performance. Here, an effective morphology tuning of the NiCo LDH/MMoSx (M = Co, Ni, and Zn) heteronanostructures is demonstrated by varying the composition of the MMoSx. Taking advantage of the benefits associated with Kirkendall growth and Ostwald ripening, tunable morphologies were successfully achieved. Among the NiCo LDH/MMoSx (M = Co, Ni, and Zn) heteronanostructures, the NiCo LDH/NiMoSx core-shell structured electrode delivers high specific capacity of 404 mA h g at 3 mA cm, as well as extraordinary cycling stability (after 10,000 cycles) of 93.2 % at 50 mA cm. In addition, ASC device coupled with NiCo LDH/NiMoSx as a cathode and Fe2O3/rGO as an anode exhibits excellent cell capacity and extraordinary cycling stability. Moreover, ASC device provides very high specific energy of 72.6 Wh kg1 at a specific power of 522.7 W kg1 and maintained the specific energy of 23.5 Wh kg at 5357.6 W kg demonstrating high applicability to energy storage devices.


2

Page 3 of 41 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


INTRODUCTION Presently, the demand for green and sustainable energy continues to rapidly increase due to energy inadequacy and environmental consequences, which cause natural fossil fuel combustion. Currently, many renewable resources have already been explored, but the intermittent features of these resources limit their real-time practical applications.1–3 Therefore, the development of economical, sustainable, and high-competence energy storage maneuvers is urgently essential to secure future defensible energy source. Among energy storage devices, supercapacitors are of particular attentiveness because of their greater specific power, quick charge-discharge ability, high lifespan, and eco-friendliness.4–6 These features allow for the applicability of supercapacitors in many potential areas such as hybrid electric vehicles, portable electronics, power grids, etc.7,8 Pseudocapacitors (PSCs) are a class of supercapacitors, which exhibit rapid multi-electron reversible Faradaic redox reactions on the surfaces of electrodes, leading to extraordinary specific capacities with a high specific energy, related to electrochemical double-layer capacitors (EDLCs). Metal oxides/hydroxides/sulfides/ phosphides has been explored as active electrode materials for PSCs, and controlling the sizes, shapes, spatial orientations, compositions, and crystalline structures of these materials enables the fine-tuning of physicochemical properties, which allows for designing effective electrode materials for practical applications.9–12 However, at a high scan rate, most PSC electrodes display drastic volume changes due to low electrical conductivity. Consequently, these materials exhibit deprived rate capability, low cyclic stability, and low specific energy, which limit their practical applications. In order to overcome these limitations, two methods have typically been employed for PSCs. One of these methods involves developing


3


Page 4 of 41

an asymmetric type supercapacitor through the use of PSC cathode and EDLC anode materials, and the other method is the proper design of electrode materials with various 3D architectures. To date, various supercapacitor materials consisting of transition metal oxides have been explored, and they are usually suffer from low intrinsic conductivity, which inhibits the large-scale real-time practical applications of PSCs.13,14 Recently, electrochemically active transition binary metal layered doubled hydroxides (LDHs), especially NiCo LDH, and binary transition metal sulfides (TMS), specifically NiCoSx and CoMoSx, have been stated to be effective pseudocapactive electrodes because of their superior electrical conductivity as well as electrochemical performance than those of the binary metal oxide counterparts.15–21 Furthermore, the binary metal LDHs and TMS provide richer redox reactions than their single components, resulting in high electrochemical activity. For instance, Liu et al. have demonstrated the hierarchical Ni–Co@Ni–Co LDH arrays on carbon cloth via template approach, which has displayed a specific capacitance (SC) of 2,200 F g at 5 A g with a high specific capacitance retention (SCR) of 98.8 % after 5,000 cycles at 5 A g.15 Song et al. have derived the Ni0.7Co0.3S2 on 3D graphene aerogel by sulfurization of NiCo LDH, which have demonstrated a SC of 2,165 F g at 1 A g with a SCR of 78.5 % after 1,000 cycles at 10 A g.16 Tang et al. have prepared FeCo2S4-tube arrays via the Kirkendall approach and investigated their applicability to PSCs, and they were found to exhibit a SC of 2,411 F g at a 5 mA cm with a SCR of 92.2 % after 5,000 cycles at the same current density.17 Wu et al. have demonstrated the effect of cation substitution on the pseudocapacitance of MCo2S4 (M = Ni, Fe, Zn), and among them, NiCo2S4 has exhibited the highest SC of 1,780 F g at 1 A g with a SCR of 92.4 % after 10,000 cycles at 10 A g.18 Mohamed et al. have prepared the hollow CNiCo2S4 nanostructures via solvothermal synthesis using the Kirkendall effect, which has displayed a SC of 1,722 F g at 1 A g with high SCR of 98.8 % after 10,000 cycles at 25 A g but the


4

Page 5 of 41 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


CNiCo2S4//AC ASC device has displayed a low specific energy of 38.3 Wh kg.19 Yang et al. have synthesized the Co3S4/CoMo2S4 ultrathin nanosheets on rGO, which has exhibited a SC of 1,457.8 F g at 1 A g with a SCR of 97.0 % after 2,000 cycles at 10 A g.20 Patil et al. have recently demonstrated the synthesis of CoMoS@Co(OH)2 core@shell structures on carbon cloth and investigated their applicability to supercapacitors, which have exhibited a SC of 1,711 F g at 20 mA cm with a SCR of 90.3 % after 5,000 cycles.21 However, these reported binary LDH and TMS based materials have exhibited low SC and very poor durabilities, as well as low specific energy. This discrepancy is due to the improper usage of electroactive materials surface for effective Faradaic redox reactions as well as poor reaction kinetics at a higher charge-discharge rate. Therefore, rationally designing electroactive materials for PSCs not only improves the utilization of the surfaces of these electroactive materials and also shortens electron or ion diffusivity length. Moreover, direct growing of materials on conductive substrate will facilitate fast charge transfer and enhance the electrical conductivity in PSCs. Therefore, growing suitable functional nanomaterials on a conductive substrate with fine morphology tuning and rational designing enormously advances electrochemical performance of active materials. In this work, for the first time, we have successfully tuned the morphology of NiCo LDH/MMoSx (M = Co, Ni, and Zn) heteronanostructures by employing the concepts of ion exchange reaction, the Kirkendall effect, and Ostwald’s ripening. The NiCo LDH/MMoSx (M = Co, Ni, and Zn) heteronanostructured electrode materials have several advantages, which include: (i) all of these materials are straightly growing on 3D porous Ni foam, which improves electrical conductivity by minimizing the internal resistance and also helps to reduce electron/ion diffusivity length at the electrode and electrolyte junction, (ii) fine tuning and rational designing of the morphology of electroactive materials can enable the usage of a greater percentage of active materials surface for


5


Page 6 of 41

quicker Faradaic redox reactions by reducing “dead volume” and thus improving the chargetransfer kinetics, (iii) these heteronanostructured electrode materials are capable to host the buffer changes during cyclic stability at higher charge-discharge rates, and (iv) the PVA/KOH gel electrolyte effectively reduces the electrolyte leakage and improves both the electrochemical activity and the cyclic stability in the asymmetric supercapacitor (ASC) devices. Fine morphology tuning and rational design results in a NiCo LDH/NiMoSx heteronanostructured electrode that exhibits a high specific capacity of 404.3 mA h g at 3 mA cm with an extraordinary specific capacity retention of 92.8 % at a high current density of 50 mA cm. Moreover, asymmetric NiCo LDH/NiMoSx//Fe2O3/rGO ASC device carries a very high specific energy of about 72.6 Wh kg at a specific power of about 522.7 W kg. RESULTS AND DISCUSSION

Scheme 1. The established plausible mechanism for the NiCo LDH/MMoSx (M = Co, Ni, and Zn) heteronanostructures formation on Ni foam. The NiCo LDH/MMoSx (M = Co, Ni, and Zn) hierarchical heteronanostructures formation on Ni foam with differently tuned morphologies is represented in Scheme 1. At first, the precursor salts


6

Page 7 of 41 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


(cobalt and nickel nitrates) ionize to respective ions (Co2+ and Ni2+) in the solution. At an optimum temperature of 120 °C, urea slowly decomposes to produces ammonia (NH3) and carbon dioxide (CO2) simultaneously.5,22 During hydrothermal reaction, under high pressure, ammonia undertakes hydrolysis and produce hydroxide (OH) ions slowly into the solution containing precursor ions, resulting to the construction of NiCo LDH nanowires on Ni foam.23,24 In the second step, under hydrothermal conditions, sodium molybdate dihydrate, thioacetamide, and cobalt nitrate react with NiCo LDH nanowires coated on Ni foam, owing to the formation of NiCo LDH/CoMoSx heteronanoflakes (S2) composed by nanowires. Replacing cobalt nitrate with nickel nitrate and zinc nitrate, results in the formation of NiCo LDH/NiMoSx core/shell nanostructures (S3) and NiCo LDH/ZnMoSx flaky-spherical heteronanostructures (S4), respectively. Differently tuned morphologies of NiCo LDH/MMoSx (M = Co, Ni, and Zn) heteronanostructures can be explained based on ion exchange, the Kirkendall effect, and Ostwald ripening.17,23–25 The Kirkendall effect was discovered by Kirkendall and co-workers in 1940, and it is known for the exchange of inner cations to the outer surface and vice versa.17,25 Ostwald ripening was discovered by Wilhelm Ostwald in 1896, and it is generally observed in either solid or liquid solutions. This phenomenon described the evolution of inhomogeneous structures over time by dissolving the inner smaller crystallites and redepositing them over the outer surface of larger crystallites.23,24 In the present case, after reaching to 120 °C, thioacetamide releases S2 ions slowly into the solution via hydrolysis, and these ions are readily exchanged with surface OH groups of NiCo LDH, resulting into the formation of a few layers of NiCo2S4 on the surface of NiCo LDH.19,26 At the optimum temperature, the Kirkendall effect favors the exchange of some of the inner nickel and cobalt ions of NiCo LDH with surface adsorbed outer cobalt, nickel, zinc and molybdenum ions leading to the formation of CoSx, NiSx, ZnSx, and MoSx hetero nucleation sites at the interface. At


7


Page 8 of 41

this stage, the morphology tuning primarily depends on the rate of dissolution and recrystallization of the newly formed metal sulfide nucleation sites via Ostwald ripening, and it can be determined based on the solubility product (Ksp) of these metal sulfides. In all cases, Ksp of MoSx remains constant and only that of another metal sulfide varies, which decides the final morphology of the product. Among the MMoSx (M = Co, Ni, and Zn) materials, the rate of dissolution and recrystallization of NiMoSx is higher due to the higher Ksp of NiS (4×10) than that of CoS (2×10) and ZnS (65×10), leading to the formation of NiCo LDH/NiMoSx core/shell nanostructures (S3) followed by NiCo LDH/CoMoSx heteronanoflakes (S2), and NiCo LDH/ZnMoSx flaky-spherical heteronanostructures (S4). In order to better understand the morphology changes, we have chosen a NiCo LDH/NiMoSx core/shell nanostructures system (S3), since it shows better electrochemical properties. Figure. S1 represents time-dependent FESEM images of NiCo LDH/NiMoSx core/shell nanostructures. At 30 min (Figure. S1a), a rough surface is clearly visible, indicating the nucleation of NiMoSx particles on the surface of NiCo LDH, with no morphology changes noticed. At the reaction time of 60 min (Figure. S1b), the decoration of NiCo LDH core surface with some of the NiMoSx particles is observed, indicating the growth of NiMoSx particles on the backbone of NiCo LDH. At 120 min (Figure. S1c), the formation of uniform NiMoSx shell with few extra particles on the surface of NiCo LDH is noticed, which is beneficial for better electrochemical properties because the network is properly connected, and they act as directing channels for the ease of the transfer of electrons and electrolyte ions. By prolonging the reaction time further to 180 min (Figure. S1d), the overgrowth of NiMoSx shell on the surface of NiCo LDH is observed. To understand the effect of concentration of sulfur on the formation of NiCo LDH/MMoSx (M = Co, Ni, and Zn) heteronanostructures, we have carried out the reactions at 120 °C for 1 h by increasing the concentration of thioacetamide (1.5


8

Page 9 of 41 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


mmol). The FE-SEM images (Figure. S2) indicates drastic changes in morphology, which are ascribed to a faster rate of dissolution and recrystallization of the corresponding metal sulfides due to the availability of a higher amount of S2 in the reaction solution, leading to the destruction of morphology of NiCo LDH/MMoSx (M = Co, Ni, and Zn) heteronanostructures. Based on these results, the concentration of thioacetamide (0.75 mmol) and the reaction period of 120 min are determined to be the optimum condition for the formation of NiCo LDH/MMoSx (M = Co, Ni, and Zn) heteronanostructures with differently tuned morphologies.

Figure 1. (a) X-ray diffraction of NiCo LDH/MMoSx (M = Co, Ni, and Zn) powder samples. (b) Raman spectra of sample S1 to S5. (c) Barrett–Joyner–Halenda (BJH) plot for samples S1 to S5. Initially, the structural analysis and phase purity of S1 to S5 samples grown on Ni foam were inspected using powder XRD (Figure. S3 and 1a). All of these samples show two strong peaks at 2θ values of 44.02 and 52.06°, which are attributed to diffraction peaks of nickel coming from the nickel foam. The X-ray diffraction pattern of sample S1exhibits three major and two minor peaks at 2θ value of 19.21, 33.19, 38.63, 59.31, and 62.95°, which are ascribed to the (001), (100), (011), (110), and (111) planes, respectively, of nickel hydroxide (JCPDS file no. 731520) and cobalt hydroxide (JCPDS file no. 300443).23,24 XRD patterns of samples S3 to S5 on Ni foam show XRD reflections similar to those of sample S1, with no apparent XRD reflections due to MMoSx (M = Co, Ni, and Zn) was noticed. It is clear that the amount of sample on Ni foam lies at about


9


Page 10 of 41

3-4 milligrams, which could be below the XRD detection limit of NiCo LDH/MMoSx (M = Co, Ni, and Zn) samples. So, we accomplished the XRD analysis for NiCo LDH/MMoSx (M = Co, Ni, and Zn) heteronanostructure powder samples. All these samples show XRD reflections similar to sample S1indicating the retention of NiCo LDH phase after the formation of MMoSx shell. It is noteworthy to mention that the (002) peak of MoS2 at 2 value of 14.40° is absent in all these samples and show two new peaks at 2 values of 9.7 and 17.6°, which are ascribed to first and second order reflections of (002) plane of layered MMoSx in the NiCo LDH/MMoSx (M = Co, Ni, and Zn) heteronanostructures.27 The observed results are similar to NiMoS nanosheets with enlarged interlayer spacing reported by Miao et al.27 Another sharp peak at 2 value of 39.46° is noticed, which is corresponding to (103) plane of MoS2 (JCPDS no. 371492) in these samples. The sample S2 shows three major peaks of CoS2 (JCPDS no. 411471) indicating coexistence of mixed phase of CoS2, and MoS2. The sample S3 shows three peaks of NiS2 (JCPDS no. 730574) along with peak of MoS2 while the sample S4 shows three broad peaks of ZnS (JCPDS no. 651691) and the peak of MoS2. These results demonstrate the formation of mixed phase of MMoSx (M = Co, Ni and Zn) in the NiCo LDH/MMoSx (M = Co, Ni, and Zn) heteronanostructures. The X-ray diffraction pattern of sample S5 shows two set of reflections at 31.22, 38.27, 50.31, and 55.17°, corresponding to the (311), (400), (511), and (440) planes of NiCo2S4 (JCPDS file no. 431477), respectively,28 and at 2θ = 15.45, 29.87, and 47.55°, related to the (111), (311), and (511) planes, respectively of Co9S8 (JCPDS file no. 731442).5 Furthermore, in order to prove the presence of metaloxygen and metalsulfur bonds in the NiCo LDH/MMoSx (M = Co, Ni, and Zn) heteronanostructures, non-destructive Raman spectral analysis has been carried out at room temperature (Figure. 1b). The Raman spectrum of sample S1shows


10

Page 11 of 41 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


broad Raman bands centered at about 510 and 655 cm1 corresponding to the CoOH stretching mode of Co(OH)2, while the other band at about 472 cm1 attributed to NiOH stretching mode of Ni(OH)2 in the NiCo LDH.29,30 In the case of NiCo LDH/MMoSx (M = Co, Ni, and Zn) heteronanostructures, Raman bands are observed at about 285, 337, 368, 820, 883, 940, and 993 cm1. The sharp band at about 940 cm1 belongs to a symmetric stretching mode of MoS in the MMoSx and the other two strong Raman bands at about 820, and 883 cm1 are associated with asymmetric stretching modes of sulfur in SMoS bond.5,31 The broad Raman band between 368 to 375 cm1 attributed to symmetric stretching mode of MSMo (M = Co, Ni, and Zn) in the NiCo LDH/MMoSx (M = Co, Ni, and Zn) heteronanostructures.31 In addition, three other bands are noticed at about 285, 337, and 993 cm–1 relating to MoO3, this phenomenon is related to the slight surface oxidation of MMoSx.5 The Raman spectrum of sample S5 shows the metal-sulfur bands at about 517 and 665 cm1 consistent to F2g and A1g modes of NiCo2S4, respectively.32 Raman spectral analysis provides evidence for the presence of metaloxygen and metalsulfur bonds in the NiCo LDH/MMoSx (M = Co, Ni, and Zn) heteronanostructures. BET measurements and pore size distribution analysis provide important information about the specific surface area (SSA) and distribution of pores on surfaces of electroactive materials, which are the important governing factors that determine the activity of the electrode materials. In order to evaluate these properties, we have performed BET surface area measurements and Figures S4a-e show the associated BET figures. All of the samples exhibit type IV BET curves, indicating mesoporous character of the samples.5,33 The SSAs for the samples S1 to S5 were estimated to be 35.2, 66.5, 74.3, 94.6, and 82.7 m2 g. All of the samples other than sample S4 exhibits a slight hysteresis loop before 0.5 P/Po, indicating a higher degree of open porous structure. These results suggest the faster capillary evaporation of nitrogen with no significant delay.33,34 The BJH method is


11


Page 12 of 41

employed in order to determine pore size distribution and Figure. 1c displays the corresponding results. The estimated pore sizes for the samples S1 to S5 are centered at 2.16, 2.01, 2.06, 1.92, and 2.10 nm, respectively. The pore-size distributions in these samples lies below 8 nm, indicating the presence of distinct mesopores, which can lead to faster electron and electrolyte ion moving at the junction, thereby results in higher electrochemical performance of the electrodes. The morphological features of NiCo LDH/MMoSx (M = Co, Ni, and Zn) heteronanostructures display significant part in determining electrochemical activity of the electrode materials. For this purpose, we have performed the FE-SEM analysis first, and Figures S5a-b displays FE-SEM images of samples S1 and S5 while Figures 2a-c show those of NiCo LDH/MMoSx (M = Co, Ni, and Zn) heteronanostructures (S2 to S4). The Figure. S5a confirms the formation of 1D nanowire-like architecture for sample S1 with a smooth surface and sharp edge tips.5,22 The length and the diameter were estimated to be about 3-4 m and about 50 ± 6 nm. Following sulfurization, pure NiCo LDH nanowires are converted to irregular spherical NiCo2S4 nanoparticles (Figure. S5b). The size of nanoparticles was assessed to be about 87 ± 16 nm. The conversion of 1D nanowires morphology to irregular spherical nanoparticles can be due to the presence of a higher amount of sulfur ion availability during the sulfurization reaction. The FE-SEM image of sample S2 exhibits porous nanoflakes-like morphology which is composed by nanowires, and the thickness of each nanoflake was estimated to be about 13 nm (Figure. 2a). In the case of sample S3, 1D morphology of the NiCo LDH core is retained by the formation of core-shell structures (Figure. 2b), and the shell thickness of NiMoSx was measured to be about 4 nm. The combination of the 1D core with 2D shell architectures favors the rapid transfer of electrolyte ions and electrons via effective connecting channels at the interface of electrode-electrolyte, thereby resulting in faster Faradaic redox reactions and thus improving the electrochemical activity of the material.


12

Page 13 of 41 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 FE-SEM image of sample S4 shows flaky-spherical heteronanostructures (Figure. 2c), and the thickness of each nanoflake was determined to be about 7 nm. Among the NiCo LDH/MMoSx (M = Co, Ni, and Zn) heteronanostructures, sample S3 is expected to show better electrochemical properties due to their 1D-2D hierarchical core-shell structures, as well as a thinner NiMoSx outer shell on conductive 1D NiCo LDH core.

Figure 2. (a-c) FE-SEM, (d-f) TEM, (g-i) HRTEM images, and (j-l) SAED patterns for the samples S2, S3 and S4, respectively. (m-q) EDS color mapping for the sample S3.


13


Page 14 of 41

To get more details on the morphological features, we have performed TEM, HRTEM, SAED, and EDAX studies. Figure. S5c shows TEM image of sample S1 confirming the formation of 1D nanowire structures; the diameter of the each nanowire was measured to be about 50±4 nm and its length lies in several micrometers. The TEM image of sample S5 shows aggregated spherical particle morphology (Figure. S5d) and the diameter of particles were measured to be about 90±10 nm. The TEM image of sample S2 displays nanobelt-like morphology with rough mesoporous surface (Figure. 2d) and thickness of nanobelt lies between 8-10 nm. The TEM image of sample S3 exhibits core-shell structures with partial transparency in the core region (Figure. 2e). The diameter was found to be about 50 nm and NiMoSx shell thickness were measured to be about 20 nm. It can clearly be seen that the diameter of nanowire core is reduced to about 30 nm from 50 nm, which can be due to the exchange of inner ions with outer ions via Kirkendall growth followed by Ostwald ripening during synthesis. The TEM image of sample S4 demonstrates thick sheet-like morphology (Figure. 2f) and the thicknesses of the nanosheet are estimated to be about 10 nm. The HRTEM image of the sample S1 exhibits lattice spacing (0.275 nm) correlating to the (100) plane of the NiCo LDH (Figure. S5e). HRTEM image of sample S5 shows the lattice spacing of 0.28 and 0.33 nm, attributed to (311) and (220) reflections of NiCo2S4 (Figure. S5f). The sample S2 show the lattice spacing (0.465 nm) in the darker region corresponding to the (001) plane of NiCo LDH and that of 0.227 nm in the brighter region related to the (103) plane of MoS2 (Figure. 2g). The HRTEM image of sample S3 displays interface between NiCo LDH nanowire core and NiMoSx shell (Figure. 2h). This configuration favors fast electron transfer and also helps increase the cyclic stability of the overall electroactive material. The lattice spacing (0.465 nm) in the inner region associated with (001) reflection of NiCo LDH, and that in the outer region (0.615 nm), attributed to (002) reflection of the MoS2. The sample S4 exhibits the lattice spacing of 0.465,


14

Page 15 of 41 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


0.159, and 0.227 nm ascribed to (001), (110) reflections of NiCo LDH, and (103) plane of MoS2, respectively (Figure. 2i). The SAED pattern of samples S1 displays a weak ring along through a spot pattern, indicating polycrystalline character (Figure. S5g), and the diffraction rings consistent to (100), (011), (012), and (110) planes of the Ni(OH)2 and the (101), (102) and (111) planes of the Co(OH)2. The SAED pattern of sample S5 exhibits bright spots with a ring pattern, demonstrating the polycrystalline character of NiCo2S4 (Figure. S5h) and diffraction rings associated with the (220), (311), (440), (400), and (444) planes of NiCo2S4 and the (111) plane of Co9S8. The SAED pattern of sample S2 shows bright spots indicating the crystalline nature of the material (Figure. 2j) and which exhibits three set of diffractions corresponding to the (100), (011), (110), (022), (001), (111), and (113) planes of the Ni(OH)2, the (100) plane of the Co(OH)2, and the (100) and (103) planes of the MoS2. The SAED pattern of sample S3 shows similar pattern like sample S1 demonstrating polycrystalline nature (Figure. 2k) and which also exhibits three set of diffractions similar to sample S2 with the absence of the (113) plane of the Ni(OH)2, and presence of (101) plane of the Co(OH)2. SAED pattern of sample S4 exhibits a ring pattern representing the crystalline nature of the material (Figure. 2l) and the diffraction rings attributing to (110), (201), and (011) planes of the Ni(OH)2, and the (100) and (103) planes of the MoS2. Figure S5. i and j show the EDAX spectra of the samples S1 and S5. The EDAX spectrum of sample S1 displayed the existence of nickel, cobalt, and oxygen elements in the sample, and that of sample S5 confirms the existence of all the required elements. Figure S6. a-c displays the EDAX spectra of the NiCo LDH/MMoSx (M = Co, Ni, and Zn) heteronanostructures, which confirm the presence of all of the elements in its samples. The atomic ratios of Co:Mo:S are found to be about 1.00:0.99:1.05 for sample S2, Ni:Mo:S is about 1.00:1.15:1.22 for sample S3, and Zn:Mo:S is about 1.00:1.13:1.22 for sample S4. To study the distribution of elements in these samples, we


15


Page 16 of 41

have carried out the elemental color mapping studies. Figure. 2m-q shows the color mapping of sample S3, which also confirming the formation NiCo LDH/NiMoSx core-shell structures with uniform distribution of the elements in the core as well as shell regions. Figure. S6d-h shows color mapping of the sample S2 and Figure. S6 i-n shows color mapping of the sample S4. These results indicate the distribution of all of the elements throughout the samples. All of these results support the formation of NiCo LDH/MMoSx (M = Co, Ni, and Zn) heteronanostructures with tunable morphology, sufficient crystalline nature, and uniform elemental distribution.

Figure 3. Survey spectra a) and deconvoluted XPS spectrum of b) cobalt 2p, c) nickel 2p, d) molybdenum 3d, e) oxygen 1s, and f) sulfur 2p of the sample S3. XPS is a great tool for better understanding of the surface oxidation states and elemental composition of the NiCo LDH/MMoSx (M = Co, Ni, and Zn) heteronanostructures. The XPS spectral analysis of the sample S3 is shown in Figure. 3 and the XPS spectral analysis of the samples S2, S4, and S5 samples are displayed in Figures S7, S8, and S9, respectively. XPS survey


16

Page 17 of 41 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


spectrum (SS) of sample S3 shows the presence of all the represented elements (Figure. 3a). The high-resolution (HR) spectra of Co 2p and Ni 2p (Figure. 3b and 3c) split into two spin-orbit doublets as well as two shakeup satellites.5,22 Co 2p shows the first doublet at about 778.9 and 794.1 eV and the second doublet at about 781.5 and 797.5 eV associated to Co 2p3/2 and Co 2p1/2 peaks, respectively.21,35–37 In addition, two satellite peaks at about 786.6 and 803.6 eV associated to characteristics of Co3+ and Co2+ ions, respectively.21 The spectrum of Ni 2p displays first doublet at about 856.2 and 874.2 eV and the second doublet at about 857.7 and 876.5 eV related with Ni 2p3/2 and Ni 2p1/2 peaks, respectively.6,38,39 The another two peaks at about 862.5 and 880.8 eV were associated to the shake-up satellite peaks of the Ni2+ ion. These results proves co-existence of Co2+, Co3+, and Ni2+, Ni3+ ions in sample S3. The deconvoluted Mo 3d spectrum displays three peaks, and the peak centered at about 226.2 eV attributed to MoS of NiMoSx (Figure. 3d).5,21 The peaks centered at about 231.8 and 235.2 eV accredited to Mo(V) 3d5/2 and Mo(V) 3d3⁄2 peaks, while two other peaks at about 232.8, and 233.8 eV corresponding to Mo(IV) 3d3⁄2, and Mo(VI) 3d5/2, respectively, indicating co-existence of Mo+4, Mo+5, and Mo+6 ions.5,40 The O 1s spectrum deconvoluted into three main peaks (Figure. 3e) and the peak centered at 530.7 eV is associated with lattice oxygen of NiCo LDH.21 The other two peaks centered at about 532.7 and 533.8 eV corresponding to shoulder peak of O 1s in the NiCo LDH and MoMO (M = Ni or Co) in O 1s ions, respectively.21 The S 2p spectrum splits into two peaks positioned at about 161.4 corresponds to S 2p3⁄2 and 162.6 eV consistent to S 2p1⁄2 metal-sulfur (S2) binding energies of NiMoSx (Figure. 3f).35,41,42 The additional peak centered at about 168.5 eV is attributed to residual sulfate groups on NiMoSx surface.5,41 These findings confirm the composition of sample S3 is composed with Co2+, Co3+, Ni2+, Ni3+, Mo+4, Mo+5, Mo6+, O2, and S2. Figure. S7 a-f shows the SS and HR spectrums of sample S2. The observed binding energies of these elements are similar to that of the


17


Page 18 of 41

sample S3, indicating the presence of required ion components in the sample. Figure. S8 a-g shows the SS and HR spectrums of sample S4. This sample shows an additional peak due to Zn 2p, and the HR spectrum shows peaks centered at about 1022.1 related to Zn 2p3/2 and 1045.2 eV accredited to Zn 2p1/2.43,44 The remaining elements show binding energies similar to that of the sample S2, demonstrating the presence of all the attributing ions in the sample. The SS and HR spectrums of sample S5 are shown in Figure. S9 a-d, which confirms the presence of all respective ions in the NiCo2S4 sample.The amount of O2 on the surface of the samples was found to be 6.9 %, 12.9 %, and 8.3 % for the samples S2, S3 and S4, respectively.

Figure 4. (a) Cyclic voltammogram of the sample S1 to S5 at 5 mV s1 and (b) cyclic voltammogram curves of sample S3 at 5 to 100 mV s. (c) GCD plot of the sample S3 at 3 to 50 mA cm. (d) specific capacity vs. current density plot of the samples S1 to S5. (e) EIS spectra of the samples S1 to S5 at open circuit voltage. (the inset displays enlarged view of the highfrequency region.) (f) Coulombic efficiency and cyclic stability and of the sample S3 at 50 mA cm2. (the inset represents initial and final ten cycles of GCD curves). To better understand electrochemical activity of the NiCo LDH/MMoSx (M = Co, Ni, and Zn) heteronanostructures, we performed comparative studies in three-electrode assembly along with


18

Page 19 of 41 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


samples S1 and S5 (Figure. 4). To begin, typical CV analysis were conceded in a 3 M KOH electrolyte solution in the window of  0.2 to 0.7 V at a scan rate of 5 mV s1 (Figure. 4a). During the both anodic and cathodic sweeps, all of the samples reveal a pair of divergent redox peaks, suggesting happening of the Faradaic redox reactions, indicating pseudocapacitive character.5 In the case of NiCo LDH/MMoSx (M = Co, Ni, and Zn) heteronanostructures, the Faradaic redox reactions are mainly governed by both NiCo LDH nanowires as well as MMoSx. For this reason, NiCo LDH/MMoSx (M = Co, Ni, and Zn) heteronanostructures exhibit higher activity than sample S1 and S5. Among the NiCo LDH/MMoSx (M = Co, Ni, and Zn) heteronanostructures, the CV curves of sample S3 shows larger integrated area and specific current, indicating higher specific capacity than that of other samples, which is mainly ascribed to better electrical conductivity because of the synergistic interaction between 1D NiCo LDH core to 2D NiMoSx ultrathin shell. Moreover, sample S3 shows a greater surface area and smaller pore diameter, which also favors higher electroactive surface for easy penetration of ions and electrons at the junction of electrolyte and electrode. The sample S1 exhibits redox peaks at about 0.32 and 0.17 V vs. Ag/AgCl. In the case of sample S5, the oxidation peak is perceived at about 0.36 V and the reduction peak is noticed at about 0.16 V. The redox peaks in these samples are mainly attributed to multi-electron transfer reactions of Co2+↔Co3+↔Co4+ and Ni2+↔Ni3+.45–47 The reactions 1-3 possibly occurred in sample S1 while 4-5 in sample S5, which are as follows: 𝑁𝑖(𝑂𝐻)2 + 𝑂𝐻 ― ⇄ 𝑁𝑖𝑂𝑂𝐻 + 𝐻2𝑂 + 𝑒 ―

(1)

𝐶𝑜(𝑂𝐻)2 + 𝑂𝐻 ― ⇄ 𝐶𝑜𝑂𝑂𝐻 + 𝐻2𝑂 + 𝑒 ―

(2)

𝐶𝑜𝑂𝑂𝐻 + 𝑂𝐻 ― ⇄ 𝐶𝑜𝑂2 + 𝐻2𝑂 + 𝑒 ―

(3)


19


𝑁𝑖𝐶𝑜2𝑆4 + 𝑂𝐻 ― + 𝐻2𝑂 ⇄ 𝑁𝑖𝑆𝑂𝐻 + 2𝐶𝑜𝑆𝑂𝐻 + 2𝑒 ― 𝐶𝑜𝑆𝑂𝐻 + 𝑂𝐻 ― ⇄ 𝐶𝑜𝑆𝑂 + 𝐻2𝑂 + 𝑒 ―

Page 20 of 41

(4) (5)

The samples S2, S3 and S4 exhibit the oxidation peaks at about 0.38, 0.40, and 0.40 V, respectively, and the reduction peaks at about 0.16, 0.14, 0.12 V, respectively. The redox peaks in these samples are accredited similar to sample S1 and S5.45–47 In these samples, MoSx and ZnSx does not show any Faradaic redox reactions, as they only allow intercalation and surface adsorption of alkali metal cations by formation of electrochemical double layer at the junction of electrode and electrolyte.5,48 The Faradaic redox reactions are primarily accredited by NiCo LDH and MSx (M = Co, and Ni). The following reactions 6-9 can be expected along with reactions 1-3:5,20,48 𝐶𝑜𝑆𝑥 + 𝑂𝐻 ― ⇄ 𝐶𝑜𝑆𝑥𝑂𝐻 + 𝑒 ―

(6)

𝐶𝑜𝑆𝑥𝑂𝐻 + 𝑂𝐻 ― ⇄ 𝐶𝑜𝑆𝑥𝑂 + 𝐻2𝑂 + 𝑒 ―

(7)

𝑁𝑖𝑆𝑥 + 𝑂𝐻 ― ⇄ 𝑁𝑖𝑆𝑥𝑂𝐻 + 𝑒 ―

(8)

(𝑀𝑆𝑥)𝑠𝑢𝑟𝑓𝑎𝑐𝑒 + 𝐾 + + 𝑒 ― ↔(𝑀𝑆𝑥 ― 𝐾 + )𝑠𝑢𝑟𝑓𝑎𝑐𝑒 (𝑀 = 𝑍𝑛 𝑎𝑛𝑑 𝑀𝑜)

(9)

Afterwards, the CV analysis were exhibited at various scan rates for all the samples. Figure. 4b shows the CV curves of sample S3, while those of the samples S1, S2, S4 and S5 are shown in Figure. S10 a-d. All the samples exhibit increases in specific current with ascending the scan rate from 5 mV s1 to 100 mV s1, suggesting well-reversible Faradaic redox reactions with wide potential separation for rapid GCD process.5,22 The positions of redox peaks shift towards respective potential sides by increasing the scan rate, demonstrating exceptional electrochemical reversibility of the samples.49 However, at each scan rate, sample S3 exhibit superficially greater


20

Page 21 of 41 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


area under CV curves as well as a larger specific current than that of other samples, demonstrating higher specific capacity. Figure. S10e shows the plot of Ip vs. 1/2 at various scan rates, which provides information on the nature of the current (surface controlled or diffusion controlled).5 A linear relationship between Ip and 1/2 is noticed and it is also well-matched with a power-law relationship i.e. Ip = ab. All of the samples show a value of ‘b’ equal to 0.5, suggesting semiinfinite linear diffusion-controlled battery-type Faradaic process.5,50 To study further electrochemical performance of the samples, we have performed GCD analysis at different current densities. Figure. 4c shows the GCD curves of sample S3, while those of the samples S1, S2, S4, and S5 are presented in Figure. S10 f-i. All of the samples exhibits divergent plateau regions with battery type GCD curves, corroborating pseudocapactive character.5,45,49 The identical shapes of GCD curves in all these samples indicating excellent capacitive characteristics and better Faradaic reversible redox reactions of the active electrode materials. In order to determine superiority of the NiCo LDH/MMoSx (M = Co, Ni, and Zn) heteronanostructures over the samples S1 and S5, we have assessed the specific and areal capacities using GCD discharge curves, and the respective plots are displayed in Figures. 4d and S10j. The predicted specific and areal capacities for the sample S1 were 121, 116, 104, 96, 87, 75, 66, 59, 47, and 40 mAh g1 and 0.21, 0.20, 0.18, 0.17, 0.15, 0.13, 0.11, 0.10, 0.08, and 0.07 mAh cm2 at various current densities. The calculated specific and areal capacities values for the sample S2 were observed to be 251, 220, 200, 184, 168, 148, 131, 118, 100, and 83 mAh g1 and 0.72, 0.64, 0.58, 0.53, 0.48, 0.42, 0.38, 0.34, 0.29, and 0.24 mAh cm2. The measured specific and areal capacities values for the sample S3 were found to be 404, 382, 359, 338, 307, 272, 243, 219, 183, and 151 mAh g1 and 1.33, 1.26, 1.18, 1.11, 1.01, 0.89, 0.80, 0.72, 0.60, and 0.50 mAh cm2. The estimated specific and areal


21


Page 22 of 41

capacities values for the sample S4 were found to be 327, 299, 276, 257, 234, 196, 166, 147, 120, and 97 mAh g1 and 1.04, 0.95, 0.88, 0.82, 0.74, 0.62, 0.53, 0.47, 0.38, and 0.31 mAh cm2. In the case of sample S5, the estimated specific and areal capacities values were found to be 153, 144, 139, 131, 126, 114, 105, 97, 82, and 69 mAh g1 and 0.32, 0.31, 0.30, 0.30, 0.28, 0.27, 0.25, 0.24, 0.22, and 0.20 mAh cm2. Among all of these samples, the sample S3 shows a significantly higher specific as well as areal capacity, which is accredited to higher electrical conductivity of 1D backbone of NiCo LDH nanowire core with thinner NiMoSx shell. It could also because of the presence of higher electrochemical active surface area as well as the ability to transfer electrolyte ions and electrons rapidly at the junction of electrode and electrolyte. During the GCD measurements, all of the samples exhibited decreasing trends of specific and areal capacity by ascending the current density, which is attributed to the internal resistance.5,22,51 At 50 mA cm  15 times increment to initial current density), the specific and areal capacity retention values were calculated to be 33.3, 33.3, 37.6, 29.8, and 44.5 % for the samples S1, S2, S3, S4, and S5, respectively. Among all of these materials, sample S3 shows higher specific capacity retention, other than sample S5, indicating the higher electrochemical activity and practical applicability of the sample. EIS is an important technique for studying inherent electrochemical and kinetic mechanisms of the electroactive electrode materials. In the present study, EIS measurements were performed and the corresponding Nyquist plot is shown in Figure. 4e. In order to better understand the EIS studies, general Nyquist plot diagram and an equivalent circuit are shown in Figure. S10k. The Nyquist plot shows three important distinct regions.5,52 The internal resistance (Rs) of the electrode can be measured, where the intercept of the curve touching the Xaxis.49,52 The Rs values for the samples S1, S2, S3, S4, and S5 were estimated to be 0.71, 0.66, 0.58, 0.62, and 0.65 Ω, respectively,


22

Page 23 of 41 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


indicating lower internal resistance for the sample S3. The charge transfer resistance (Rct) was estimated to be 0.59, 0.42, 0.36, 0.44, and 0.45 Ω for the samples S1, S2, S3, S4, and S5, respectively. Rct is kinetically controlled, which depends on the applied frequency and it can be estimated by calculating the time constant.5 The estimated o values for the samples S1, S2, S3, S4, and S5 were found to be 1.31 × 105, 8.94 × 106, 6.09 × 106, 7.44 × 106, and 1.09 × 105, respectively. These results demonstrate that the sample S3 exhibits rapid charge transfer kinetics. The Warburg impedance provides information on the electrolyte ion diffusivity52 and it can be estimated from the slope of the Nyquist plot in the middle frequency region, and it was calculated to be 5.1, 11.5, 16.4, 15.3, and 3.4 for the samples S1, S2, S3, S4, and S5, respectively. A higher slope indicates a straighter line (nearer to 90o angle) in the middle frequency region, suggesting that sample S3 exhibits better electrolyte ion diffusion, resulting in faster Faradaic redox reactions and leading to higher specific capacity. The low-frequency region specifies capacitive nature of electroactive material, and in the present case, sample S3 shows an almost vertical straight line than other samples, indicating a higher capacitive nature of the sample.5,52 These results prove that sample S3 exhibits faster charge transfer kinetics, lower time constant, and higher capacitive nature than other samples. The high cyclic stability and good electrochemical efficiency of the electrode materials are two main important features for practical energy storage applications. With respect to this aspect, we have chosen sample S3 since it shows high electrochemical activity. We have carried out the cyclic stability studies in the same electrolyte at 50 mA cm for 10,000 GCD cycles (Figure. 4f). After 5,000 cycles, a minor loss in initial specific capacity is observed (2.7 %), and upon increasing to 10,000 cycles, only 6.8 % loss is noticed, suggesting excellent cyclic stability of the sample S3.


23


Page 24 of 41

The inset of Figure. 4f indicating retention of pseudocapacitive nature of the electroactive material after 10,000 cycles, which is credited to active Faradaic redox reactions at the electrolyte-electrode interface.5 Moreover, it exhibits nearly 99.8 % coulombic efficiency throughout the cyclic stability, indicating better electrochemical applicability. After the cyclic stability, EIS studies were performed to understand the electrode kinetics (Figure. S10l). After the cyclic stability, minor changes in Rs and Rct values are noticed, a negligible change is observed in the leaning of a straight line, indicating high feasibility of reversible Faradaic redox reactions with high cyclic stability. The sample S3 exhibits extraordinary electrochemical properties compared to earlier testified LDH and metal sulfide-based electrode materials (Table. S1). The high specific and areal capacity as well as exceptional cyclic stability of the NiCo LDH/NiMoSx core-shell structured sample S3 can be accredited to the following factors: (i) direct growth of electroactive material on a conductive substrate (e.g., 3D Ni foam) helps to overcome the “dead volume” affected by conductive additives and polymer binders and allows for complete operation of available electroactive surface area for quicker Faradaic redox reactions, (ii) the synergistic interaction of 3D hierarchical core-shell morphology composed by 1D NiCo LDH nanowire core and 2D ultrathin NiMoSx shell facilitates faster transfer of electrolyte ion and electrons via conducting interconnecting channels, (iii) with the help of the Kirkendall effect and Ostwald ripening, partially transparent void formation happened in the core region; these voids can effectively store the electrolyte ions and enhance the wettability of the electrode, (iv) 3D hierarchical core-shell morphology of the electroactive electrode material can competently house buffer changes during cycling stability and avoid maceration of the electrode. After the cyclic stability, we have check the morphology and elemental distribution of the sample S3, we have performed FE-SEM and XPS analysis (Figure. S11 a-h). XPS analysis confirms the retention of valence states and FE-


24

Page 25 of 41 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


SEM images show the retention of morphology with a slightly improved rough surface. These results indicate high practical applicability of the sample S3 as PSC electrodes to achieve a high specific and areal capacity with good rate capability and exceptional cyclic stability.

Figure 5. (a) Cyclic voltammogram of the NiCo LDH/NiMoSx//Fe2O3@rGO ASC device at various scan rates. (b) GCD plot of the ASC device at various current densities. (c) cell capacity vs. current density plot of the ASC device. (d) Coulombic efficiency and cyclic stability of the ASC device at 40 mA cm2. (inset displays initial and final ten GCD cycles). (e) the plot of energy density (specific energy) vs. power density (specific power) vs. charge time of ASC device. (f) Ragone plot of the ASC device with the reported literature. To check the high real-world applicability of the sample S3, fabricated the hybrid ASC device in a PVA/KOH gel electrolyte. One can increase the specific energy of the ASC device by extending the operational potential window and it will purely depends on the appropriate selection of the combination of the cathode and anode electrodes.50 Prior to the ASC device fabrication, we have chosen Fe2O3/rGO core-shell aerogel as an anode material and the NiCo LDH/NiMoSx (S3) as a cathode material and performed CV analysis in a three-electrode assembly at 20 mV s using same aqueous electrolyte (Figure. S12a). The Fe2O3/rGO, i.e. negative electrode displays deviated


25


Page 26 of 41

rectangular shape with slight broad redox peaks in the window of  1.0 to 0.0 V, suggesting a features of PSC properties, while that of the sample S3 displays distinctive redox peaks in the window of  0.2 to 0.7, indicating a typical PSC nature of the material.5 Therefore, the fabrication of NiCo LDH/NiMoSx//Fe2O3/rGO ASC device deliberately improves the working potential window, thus ensuring high specific energy without negotiating the specific power. In order to decide the working potential window of the ASC device, the CV measurements were performed in different windows at 10 mV s1 (Figure. S12b). An almost rectangular feature of the CV curves can clearly see below 1.4 V, indicating inadequate Faradaic process.5 Upon prolonging working window potential above 1.6 V, release of oxygen gas was noticed.5,22 Based on these results, the optimum working window potential of 0.0-1.6 V was fixed for the ASC device in further electrochemical analysis. Afterward, electrochemical activity of the ASC device was examined in a fixed window potential of 0.0-1.6 V (Figure. 5a). The CV feature displays prominent oxidation peak at about 1.32 V, and that of the reduction peak at about 1.05 V. By increasing the scan rate, redox peaks was slightly shifted, indicating better reversibility and quicker electrode kinetics for the Faradaic electrode reactions.5,22 To study the GCD window potential of the ASC device, we have accomplished GCD measurements in the potential windows ranging from 0.0-1.0 V to 0.01.6 V at 40 mA cm2 (Figure. S12c). The Figure. 5b displays GCD curves of ASC device in an applied operating potential window at various current densities, indicating distinctive battery type GCD curves.5,22 Moreover, symmetrical shapes of GCD curves at all current densities, implying better capacitive behavior. At 6 mA cm2, a negligible IR drop of 0.015 V was noticed in the GCD curves, which is mainly attributed to the ESR of the ASC device.5,44 The cell capacities (𝐶𝑐, mA h g) as a function of various current densities were plotted (Figure. 5c). The cell capacities of the ASC device were estimated to be 82.4, 70.2, 63.7, 57.3, 52.5, 46.1, 41.4, 37.3, 34.6, and 33.8 mA


26

Page 27 of 41 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


h g1 at various current densities. With increasing current densities, the cell capacities were gradually decreased, which can be credited to intrinsic resistance of electrode as well as incomplete Faradaic redox reactions at higher current densities.5,53 It is worth mentioning that ASC device exhibits an apparently better cell capacity of 82.4 mA h g1 at 6 mA cm2. In addition, by increasing current density to 60 mA cm2 (about 10 times), the ASC device still retains about 41.9 % of its original cell capacity, indicating better rate ability of ASC device. The cyclic stability of ASC device was carried out at a high current density of 40 mA cm up to 10,000 successive GCD cycles (Figure. 5d). It displays about 5.9 % loss of its original cell capacity up to 5,000 cycles, and later shows a minor loss of 2.6 % up to 10,000 cycles, and finally reaches to 91.5 % of its original cell capacity, demonstrating the extraordinary cyclic stability of the ASC device. The inset of Figure. 5d displays initial and final ten GCD cycles of the ASC device, indicating better reversibility of the Faradaic redox reactions of the ASC device electrodes. Moreover, ASC device shows 99.9 % coulombic efficiency throughout cyclic stability studies. To understand about electrode kinetics of the device, EIS analysis was carried out prior and after cyclic stability of the ASC device (Figure. S12d). In the present case, we couldn’t see the semicircle (high frequency region), which is because of fast transfer of electrons between cathode and anode ascribed to their high porosity as well as better electrical conductivity. Moreover, PVA/KOH gel electrolyte between Fe2O3/rGO and NiCo LDH/NiMoSx helps to shorten ion transportation and diffusion distance resulting in low charge transfer resistance. The intercept on the Xaxis provides the equivalent series resistance (ESR), which was determined to be 0.19 V vs. Ag/AgCl prior to and after cyclic stability indicating similar ESR of the ASC device. A slight deviation in inclination of straight line is noticed (low-frequency region), suggesting minor change in its capacitive behavior of the ASC device. All these results further demonstrated the worthy


27


Page 28 of 41

stability and enhanced contact ability on the current collector. The high cyclic stability of the ASC device is because of credibility of PVA/KOH gel electrolyte. The specific energy and specific power of the solid-state ASC device were assessed from the Ragone plots (Figure. 5e and 5f). The solid-state ASC device carries a very high specific energy of 72.6 Wh kg1 at a specific power of 522.7 W kg1. The ASC device quiet achieved an specific energy of 25.3 Wh kg1 at 5357.6 W kg1. The solid-state ASC device exhibits superior performance than recently reported ASC devices such as Co3S4/CoMo2S4// activated carbon (AC) (33.1 Wh kg1 at 850 W kg1),20 CoMoS@Co(OH)2//AC (58.1 Wh kg1 at 450 W kg1),21 MnCo2O4@Ni(OH)2//AC (48 Wh kg1 at 1400 W kg1),45 H-TiO2@Ni(OH)2//nitrogen doped carbon (N–C) (70.9 Wh kg1 at 102.9 W kg1),54 NiCo2S4@Co(OH)2//AC (35.89 Wh kg1 at 400 W kg1),55 ZnCo2O4@NixCo2x(OH)6x//AC (26.2 Wh kg1 at 511.8 W kg1),56 FeCo2S4–NiCo2S4//AC (46 Wh kg1 at 1070 W kg1),57 Ni1.77Co1.23S4//homemade activated carbon (HMC) (42.7 Wh kg1 at 190.8 W kg1)58, and many others that are reported in Table. S2. The excellent electrochemical activity of the ASC device is accredited to its very high cell capacity, as well as high working window potential of the ASC device. All of these results support high electrochemical activity of the NiCo LDH/NiMoSx as a positive electrode or a cathode material for the ASC device and its commercial applicability towards practical energy storage applications, especially for supercapacitors. CONCLUSION We have successfully tuned the morphology of the NiCo LDH/MMoSx (M = Co, Ni, and Zn) heteronanostructures by varying the composition of the MMoSx (M = Co, Ni, and Zn) via a versatile hydrothermal method. The morphology tuning of the NiCo LDH/MMoSx (M = Co, Ni, and Zn) heteronanostructures were demonstrated with the help of Kirkendall growth and Ostwald


28

Page 29 of 41 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


ripening. The phase analysis, morphological studies, valence states and elemental composition of the NiCo LDH/MMoSx (M = Co, Ni, and Zn) heteronanostructures were established using XRD, Raman, BET, and XPS. Among NiCo LDH/MMoSx (M = Co, Ni, and Zn) heteronanostructures, NiCo LDH/NiMoSx core-shell electrodes demonstrated a very high specific capacity of 404 mA h g at 3 mA cm, which can be credited to 3D hierarchical core-shell morphology, better electrical conductivity, and faster Faradaic redox reactions. Moreover, the same electrode exhibited an extraordinary cyclic stability of 93.2 % at 50 mA cm after 10,000 cycles. In addition, solid-state ASC device composed by NiCo LDH/NiMoSx as a cathode and Fe2O3/rGO as an anode supplies a very high specific energy of 72.6 Wh kg1 at 522.7 W kg1. Moreover, it is also demonstrated a favourable specific energy of 25.3 Wh kg1 at 5357.6 W kg1. ASC device also displays extraordinary cyclic stability (91.5 %) after 10,000 cycles at 40 mA cm. The demonstrated synthetic method for the tuning of the morphology of the electroactive electrode materials, as well as the device fabrication technique, are expected to boost up the designing and fabrication of the effective practical energy storage devices. EXPERIMENTAL SECTION Reagents: Cobalt(II) nitrate hexahydrate, nickel(II) nitrate hexahydrate, zinc nitrate hexahydrate, iron(III) chloride hexahydrate, ammonium fluoride, urea, sodium molybdate dihydrate, thioacetamide, graphite powder (

Kirkendall Growth and Ostwald Ripening Induced Hierarchical

Recommend Documents