Oxygen functionalization-assisted anionic exchange toward unique construction of flower-like transition metal chalcogenide embedded carbon fabric for ultra-long life flexible energy storage and conversion

Abstract

The metal-organic framework (MOF) derived Ni–Co–C–N composite alloys (NiCCZ) were “embedded” inside the carbon cloth (CC) strands as opposed to the popular idea of growing them upward to realize ultrastable energy storage and conversion application. The NiCCZ was then oxygen functionalized, facilitating the next step of stoichiometric sulfur anion diffusion during hydrothermal sulfurization, generating a flower-like metal hydroxysulfide structure (NiCCZOS) with strong partial implantation inside CC. Thus obtained NiCCZOS shows an excellent capacity when tested as a supercapacitor electrode in a three-electrode configuration. Moreover, when paired with the biomass-derived nitrogen-rich activated carbon, the asymmetric supercapacitor device shows almost 100% capacity retention even after 45,000 charge–discharge cycles with remarkable energy density (59.4 Wh kg^–1/263.8 µWh cm^–2) owing to a uniquely designed cathode. Furthermore, the same electrode performed as an excellent bifunctional water-splitting electrocatalyst with an overpotential of 271 mV for oxygen evolution reaction (OER) and 168.4 mV for hydrogen evolution reaction (HER) at 10 mA cm⁻² current density along with 30 h of unhinged chronopotentiometric stability performance for both HER and OER. Hence, a unique metal chalcogenide composite electrode/substrate configuration has been proposed as a highly stable electrode material for flexible energy storage and conversion applications.

INTRODUCTION

The Faradic type supercapacitors with proper design optimization, and their inherent superiority such as fast charging–discharging ability, low maintenance cost, high-power density (PD), and high cyclic stability are well in the path of a safer and more powerful alternative to conventional lithium-ion batteries.^1,2 Furthermore, compared with conventional batteries, supercapacitors give more options for the flexible power source design owing to their reaction kinetics due to relatively safer electrode material and electrolytes.³ Conductive carbon cloth (CC) is the most used electrode substrate for flexible energy storage devices and supercapacitors, mainly due to its inherent advantages such as flexibility, lightweight, strength, cost-effectiveness, and environment-friendly nature.⁴ Despite several advantages of CC as an energy storage substrate material, the attachment of active material on its surface is not as strong as needed, mainly due to its hydrophobic nature coupled with surface incompatibility between the CC and the metal-based electrode material.^5,6 This material incompatibility always risks active material falling out even in rigid charge–discharge conditions, let alone flexible operation. Hence, fabricating the strongly attached binder-less self-grown active material over a CC substrate that can give an ultra-stable performance for several charge–discharge cycles has always been a challenge for supercapacitor applications.

On the other hand, the multivalence transition metal oxides/hydroxides are the prominent contenders for Faradic-type supercapacitors. In that, the nickel–cobalt duo is the most studied material owing to their well-known richer redox reaction kinetics generating high pseudocapacitance.^7,8 Furthermore, their chalcogenide form, the nickel–cobalt sulfides, or hydroxysulfides are found to have even better electrochemical properties than the oxide counterpart such as better electrical conductivity and stability owing to the smaller band gap during anionic exchange reaction, facilitating faster electron transport.^9–11 Hence, conjugating nickel–cobalt sulfide-based active material with CC seems to be the most prolific solution for high-performing flexible supercapacitor electrodes. Justifiably, numerous research has been performed on the formulation of CC-grown nickel–cobalt sulfide/hydroxysulfide for supercapacitor application. Zhao et al.¹² prepared hollow and ultrathin Ni–Co–S nanosheets on electrochemically activated carbon cloth (ACC) via the etching/ion exchange method employing a metal-organic framework (MOF). A flexible hybrid supercapacitor device is fabricated with the Ni–Co–S/ACC cathode and activated carbon (AC)/ACC anode, and a high energy density (ED) of 30.1 Wh kg^–1 at a PD of 800.2 W kg^–1 along with capacitance retention of 82% after 10,000 cycles is reported. Likewise, Liu et al.¹³ also prepared NiCoS-coated CC incorporating cobalt MOF and hydrothermal absorption of nickel and sulfur ions. The supercapacitor device is fabricated with the NiCoS/CC cathode and AC/CC anode, which shows an ED of 40 Wh kg^–1 at a PD of 379 W kg^–1 along with 84% capacity retention after 7000 charge–discharge cycles. Along with these representative works, almost every single work reporting self-grown nickel–cobalt sulfide, or any other metal oxide/sulfides for that matter has a common shortcoming in the form of poor long-term stability. Contrarily, a noteworthy work is reported by Chen et al.,¹⁴ where nickel–cobalt sulfide nanosheets over a CC are prepared via a one-step electrodeposition technique. The resulting positive electrode when coupled with a graphene film negative electrode gave some excellent device performance with an ED of 60 Wh kg^–1 at a PD of 1.8 kW kg^–1 and most importantly the capacitance retention of 90.1% for 10,000, 82.2% for 20,000, and 63.2% for 50,000 charge–discharge cycles. Authors pointed out probable reasoning for rapid capacity fading after 20,000 cycles as delamination of active material from the CC surface, active material evolution to less electroactive phases, and morphology degradation of the active material. Hence, eliminating these issues can prolong the stability of such active material current collector conjugate. Building on the necessity created by the poor stability performance of self-grown transition metal sulfide composite and its possible resolutions pointed out by some great work performed on the same, we devised a novel CC/active material composite with an ultra-stable attachment by embedding the MOF-promoted metal seeds composite (Ni–Co–C–N) inside the CC, contrary to the popular idea of just growing them upward. Thus embedded metal nanoseed alloys were oxygen functionalized as a crucial step to ensure better electrochemical performance and then sulfurized in aqueous media producing flower-like nickel–cobalt hydroxysulfide composite partially embedded inside CC. Hence, we report an excellent supercapacitor performance of a fabricated supercapacitor device with virtually no capacity fading even after 45,000 charge–discharge cycles without any compromise in energy and PD.

Furthermore, the supercapacitor electrode materials and water-splitting catalysts share fundamental requirements such as high stability, high electrical conductivity, high energy conversion efficiency, environmental benignity, cost-efficiency, and so forth.^15–17 Hence, motivated by excellent supercapacitive electrochemical results, as-fabricated electrodes were used in oxygen evolution reaction (OER), and hydrogen evolution reaction (HER) as a novel bifunctional water electrocatalyst. The hydrogen produced from water splitting has a multitude of green energy benefits. Firstly, doing so can eradicate the need for conventional fossil-fuel-derived hydrogen production. Secondly, the obtained hydrogen can be utilized to further minimize fossil fuel consumption such as by developing hydrogen-driven vehicles. Furthermore, the huge demand for pure hydrogen and oxygen for industrial applications necessitates a greener way of producing them.^18–20 While the metal embedding/etching inside the CC has been reported on a couple of occasions, the design optimization toward ultra-stable supercapacitor and water-splitting applications with MOF-incorporated bimetallic electrocatalyst has never been reported before as per the best of the author's knowledge.^21–23 Hence the applicability of the electrode we synthesized in water splitting work has made it even more versatile and worth pursuing further as a viable alternative to conventional electrocatalysts for energy storage and conversion applications.

EXPERIMENTAL DETAILS

Materials and accessories used in this work are presented in Section S1. Likewise, several electrode preparation techniques are presented in Section S2.

RESULTS AND DISCUSSIONS

The instruments and methodology implemented for physical and electrochemical characterizations are reported in Sections S3 and S4.

Physicochemical results

The process of synthesizing different types of electrodes, including NiCC, NiCCZIF, NiCCZ, and NiCCZOS, is illustrated in Figure 1. The NiCC electrode was prepared using the electrochemical deposition (ECD) technique, and ZIF-67 micrograss was grown on it using the direct precipitation technique under ambient conditions. The resulting sample was calcined at a high temperature under an Ar + H₂ reduction environment to produce Ni/Co alloys embedded in the CC. An oxygen functionalization step was carried out on the sample before sulfurization to ensure optimized oxygen-assisted sulfur loading. The sample was then sulfurized under hydrothermal conditions.

Figure 1. Collective synthesis techniques of different samples.

The physical morphology of the electrodes was observed using a field emission scanning electron microscope (FE-SEM) and the resulting micrographs were shown in Figure 2. The images in Figure 2A–C depict nickel nanolayers that were uniformly coated over a CC substrate at varying magnifications. It is believed that the nickel cations present in the precursor solution start attaching to the CC microstrands when energy is applied through an electrochemical workstation.²⁴ This technique has several benefits, such as an ultrafast deposition rate, eco-friendliness, and the absence of surface activating agents and unwanted by-products. Additionally, the nickel nanolayer deposited using the ECD technique has highly improved adhesion with the CC substrate.²⁵ The FE-SEM image of NiCC in contrast to bare CC can be found in Figure S2A,B. Our previous work comprehensively studied and reported the necessary parameters for an optimal coating of Ni nanolayer over CC.²⁶ Next, the ZIF-67 micrograss was grown over NiCC to attain the collective benefit of bimetallic transition metals that helps improve the electrochemical results due to their multiple oxidation states. Furthermore, like an electrochemical deposition, the synthesis of ZIF-67 is a time-efficient approach to nanomaterial synthesis along with a multitude of benefits such as highly tunable material yield, porous morphology, large surface area, abundant active sites, and so forth.²⁷ Likewise, the imidazole anion linkers can be carbonized by calcination at high temperatures resulting in the metal-incorporated nitrogen-doped porous carbon matrix with enhanced electrical conductivity.^28–30

Figure 2. FE-SEM images of (A–C) NiCC, (D–F) NiCCZIF, (G–I) NiCCZO, and (J–L) NiCCZOS-4 samples over different magnifications. (M) Artistic representation of stability and corresponding ion/electron mobility depending on the surface grown and embedded active material.

Figure 2D–F shows uniformly and densely grown ZIF-67 2D micrograss over NiCC substrate over a reaction period of 1 h under different magnifications. Figure 2G–I shows FE-SEM images of NiCCZ samples, in which strawberry-like morphology can be seen with nanoseeds of NiCo metallic alloy embedded inside the CC. Owing to the ambiance provided by the hydrogen/argon mixture, the following probable reaction is proposed that generates the metal alloy nanoseeds at high temperatures. [Image Omitted. See PDF] [Image Omitted. See PDF]

The high-temperature NiCo alloy is then embedded inside the CC producing a rather pleasing strawberry seed-like homogenous morphology. The metal pitting inside the CC is mostly governed by the gasification of CC to form CO owing to the high-temperature metal etching.²³ The gaseous species from the said reaction are identified through the in situ Fourier-transform infrared (FTIR) spectroscopy. The FTIR spectrum is shown in Figure S3. As speculated, the CO is found to be the most prominent species that is generated consistently throughout the calcination, starting from 500°C, peaking at 800°C, and maintaining a slightly reduced intensity at 800°C for the remainder of calcination holding time. Whereas other gaseous byproducts observed were CH₄ and CO₂ at 500°C whose intensity was diminished to almost negligible toward higher temperatures. Careful observation of FE-SEM images in Figure 2H suggests that the NiCo alloys are integrated very well within CC. Also as seen in the FE-SEM image in Figure 2I, the NiCC alloys are very well placed inside the CC cavities making them firmly settled when compared with the vertically grown nanomaterials. Further cross-sectional view of metal seed embedding and the holes created by them in CC can be seen in Figure S4.

Figure 2J–L shows the FE-SEM images of NiCCZOS samples under different magnifications. The NiCCZ nanoalloys from earlier seem to have taken a flower-like structure owing to the hydrothermal sulfurization process. No obvious cavities are noticed here due to the slightly increased nanoalloys size due to the formation of metal hydroxysulfide heterostructure.³¹ The increase in the volume of the metal hydroxysulfide would ensure the proper attachment of active materials inside the CC cavities if they were not affixed already. The high magnification FE-SEM image in Figure 2L further enlightens that the exfoliated CC layers are also covered with hydroxysulfide which in turn will greatly enhance the electrochemical results of the electrodes. Furthermore, the FE-SEM images of sample NiCCZS (sulfurization without oxygen functionalization) are presented in Figure S5 along with that of NiCCZOS-4 for better contrast. Interestingly, different morphology is observed for NiCCZS with no discernible nanoseeds embedded inside the CC. Most probably, the sulfurization occurs on the surface of CC owing to the presence of oxygen functional groups attached from the atmospheric exposure. Furthermore, Figure S6 shows the energy-dispersive spectroscopy (EDS) elemental color mapping and EDS spectrum of NiCCZOS-4 and NiCCZS samples suggesting the presence of Ni, Co, S, and C elements. The prominently accumulated sulfur clusters in the case of NiCCZOS-4 suggest the sulfurization of metal nanoseeds rather than the superficial sulfurization of CC. Whereas, in the case of NiCCZS the sulfurization seems to occur all over the surface rather than the metal seeds, contrary to the case of NiCCZOS-4. This preliminary morphological characterization of different samples hints toward the positive impact of oxygen functionalization as only with the proper functionalization morphologically accurate samples obtained. Figure S7 shows actual electronic images of NiCC, NiCCZIF, NiCCZ, and NiCCZOS, respectively. Likewise, Figure 2M shows an artistic representation of ion/electron mobility and stability of active material/current collector interface depending on whether the active material is surface grown or embedded inside the current collector, highlighting the possible positive impact of sulfurization toward overall better electrochemical results. Furthermore, the FE-SEM images of electrodes from optimization processes such as the MOF incorporation time and calcination temperature optimization can be seen in Figure S8. These FE-SEM images highlighted the importance of MOF incorporation time and calcination temperature as the desired morphology is obtained only at precisely maintained conditions with optimal metal loading that is well adhered to CC.

Figure 3A represents the high-resolution transmission electron microscope (HRTEM) image of NiCCZOS-4 samples showing the distinct lattice planes (400), (311), and (220) matching most prominently with that of NiCo₂S₄.³² Besides the obvious lattice planes, the presence of several nondiscernible planes with mismatch morphology along with amorphous boundary layers strongly suggests the existence of a hydroxysulfide concoction. The lattice disorientation and defects created by the aqueous sulfurization process are believed to positively impact the active materials' catalytic properties.³³ Figure 3B shows the transmission electron microscope (TEM) image corresponding to Figure 3A. The fast Fourier transform (FFT) shown in the inset of the same figure represents the yellow circled region and supports the polycrystalline/amorphous assortment of the material as discussed earlier. Figure 3C shows the high-angle annular darkfield energy-dispersive spectroscopy (HAADF-EDS) color mapping images corresponding to the nanobulk of Figure 3B, highlighting the presence of nickel, cobalt, sulfur, oxygen, and nitrogen elements. The observed nitrogen is believed to be originated from the organic linker (2-methyl imidazole) used in the preparation of ZIF-67 nanograss. Thus present nitrogen has a multitude of benefits in energy storage and conversion such as improved conductivity, increased defect density, reduced diffusion barrier, and so forth.³⁴ Additional TEM analysis of the NiCCZOS-4 sample shed more light on the atomic structure of nanomaterials on CC. Figure S9A shows the crystalline NiCo₂S₄ phase and metal hydroxysulfide flower-like phases with yellow and red colors, respectively, along with corresponding color mapping in Figure S9B and map sum spectrum table showing the weight percentage of different elements in Figure S9C. Hence it can be concluded that the crystal structure is mostly embedded inside the CC and a flower-like structure is on its top owing to the extended hydrothermal reaction environment. Hence, the TEM images show the whole structure with metal sulfide crystals as well as a metal hydroxysulfide flower-like structure. However, the SEM image only shows the top view, that is, the flower-like structure. Analyzing these results the atomic structure of nanoparticles on fiber is proposed as shown in Figure S9D.

Figure 3. (A) HR-TEM micrograph of NiCCZOS-4 sample. (B) TEM image of NiCCZOS-4 with corresponding FFT pattern (inset). (C) HAADF-EDS color mapping of NiCCZOS-4 showing Ni, Co, S, O, and N elements. (D) XRD spectrums of NiCCZ, NiCCZS, and NiCCZOS-4, XPS spectra of (E) NiCCZ and NiCCZOS-4 survey, and (F) high-resolution XPS spectra of Ni 2p, (G) Co 2p, (H) S 2p, and (I) N 1s.

Likewise, Figure S10A,B shows the TEM images of the NiCCZ sample under lower and higher magnifications. Interestingly carbon nanotube structure is also observed but in negligible amounts, hence, their electrochemical contribution can be neglected. Figure S10A,B along with the HAADF-EDS color mapping as shown in Figure S10C suggests the NiCCZ is mainly composed of Ni, Co, O, C, and N elements.

X-ray diffraction (XRD) analysis was performed on several samples to analyze the crystal phase and composition. As shown in Figure 3D, the distinctly visible diffraction peaks of sample NiCCZ at 44.43, 51.83, and 76.25 can be, respectively, assigned to (111), (200), and (220) planes of face-centered cubic structured Ni₅₀Co₅₀ alloy.^35,36 This suggests the reduction of nickel and cobalt to pure metallic form from their respective oxide/hydroxide forms owing to the argon/hydrogen atmosphere implemented during calcination. Interestingly the diffraction peaks of the NiCCZS sample matched consistently with that of Co(OH)₂ and Ni(OH)₂. Hence, the Ni(OH)₂ shared the crystal planes (001) and (100) at 2θ degrees of 19.11 and 32.55 with that of Co(OH)₂ with additional crystal planes (101), (102), and (200) at 2θ degrees 38.61, 52.23, and 69.45, respectively.³⁷ Likewise, the Co(OH)₂ crystal planes beside that matched with Ni(OH)₂ are indexed as (011), (100), (110), and (111) at 2θ degrees 37.95, 51.39, 57.91, and 61.83, respectively.³⁸ Whereas the diffraction peaks of the NiCCZOS-4 sample perfectly matched with that of NiCo₂S₄ with distinctly visible diffraction peaks at 31.57, 38.19, 50.21, and 55.25 2θ degrees corresponding to (311), (400), (511), and (440) crystal planes, respectively. These crystal planes validate those seen in HR-TEM analysis and carry earlier mentioned significance.

Reportedly, it has been observed that the Gibbs free energy for sulfurization of pure metal is considerably high when compared to that of a corresponding metal oxide. Moreover, even after applying the much higher energy, the final composition of metal sulfide from pure metal is observed to be nonstoichiometric.³⁹ The XRD results of NiCCZS imply a poor degree of sulfurization owing to the lack of sufficiently homogenized oxygen functional groups in NiCCZ before the hydrothermal sulfurization. The hydrothermal sulfurization technique employed here is insufficient to produce electrochemically viable nickel–cobalt sulfide from Ni/Co alloy and mainly produced nickel and cobalt hydroxides, along with partial sulfurization owing to the aqueous environment provided. The oxygen functionalized sample (NiCCZO-4) showed a stoichiometrically accurate final product as NiCCZOS-4. This is mainly due to the gradual replacement of oxygen by a sulfur element that eventually transforms metal oxide/hydroxide to metal sulfide/hydroxysulfide.³¹ This highlighted the exceptionally high significance of the partial oxidation step before the sulfurization.

X-ray photoelectron spectroscopy (XPS) was used to analyze all samples' surface elemental composition and valance state. Figure 3E shows the XPS survey spectra of NiCCZ and NiCCZOS-4 samples showing the absence and presence of a sulfur group in respective spectra. XPS survey spectra of the NiCCZOS-4 sample reveal the presence of nickel, cobalt, sulfur, oxygen, and nitrogen elements supporting the HAADF-EDS elemental mapping results. Figure 3F shows the comparative Ni 2p core-level spectrum of the NiCCZOS-4 sample having Ni 2$${p}_{3/2}$$ and Ni 2$${p}_{1/2}$$ doublets at binding energy (BE) of 856.85 and 874.98 eV, respectively and the NiCCZS sample has the same at BE of 857.52 and 875.48 eV, respectively. The negative peak shift for the oxygen-functionalized NiCCZOS-4 sample suggests that the sulfurization after partial oxidation greatly enhances the electron density around nickel species. Additionally, it has been observed from peak intensity analysis that the ratio of Ni³⁺ and Ni²⁺ has slightly increased for the NiCCZOS-4 sample suggesting an increase in Ni²⁺ density.⁴⁰ The negative peak shift and increased Ni²⁺ number suggest the decrease in d-orbitals density further assisting the H_adsorbed and OH group desorption and catalytic property enhancement of the prepared material.⁴¹ Each doublet is accompanied by corresponding satellite peaks for both samples, as shown in the figure. These satellite peaks around 870–860 eV are attributed to nickel oxide and hydroxide phases.⁴² Likewise, the peaks located around 853.74 and 870.59 eV are attributed to the 2$${p}_{3/2}$$ and 2$${p}_{1/2}$$ phases of metallic nickel (Ni°) owing to the high-temperature reduction of the sample.⁴³ Likewise, Figure 3G shows the Co 2p core-level spectrum of the NiCCZOS-4 and NiCCZS samples with two major deconvoluted peaks at BE of 778.66, 798.43 and 778.63, 798.97 eV corresponding to Co 2$${p}_{3/2}$$ and Co 2$${p}_{1/2}$$, respectively.⁴⁴ Both peaks are accompanied by the shakeup satellite peaks as shown in the figure. Furthermore, extra peaks around BE of 776.57 and 793.72 eV are attributed to the metallic Co (Co⁰), owing to the reduction ambience provided during high-temperature calcination. The negative peak shift is observed in the case of Co 2p core level spectrum as well for oxygen functionalized sample and holds the same meaning as discussed earlier in the case of Ni 2p core level spectrum analysis. Hence justifying the NiCCZOS-4 electrode as a superior electrocatalyst having multiple oxidation states.

Meanwhile, the comparative S 2p core level spectrum of NiCCZOS-4 and NiCCZS can be seen in Figure 3H. The characteristic peaks of the S 2p core level spectrum of the NiCCZOS-4 sample can be seen at 161.46 and 162.61 eV and are assigned to the S 2$${p}_{3/2}$$ and S 2$${p}_{1/2}$$, respectively suggesting the metal (Ni/Co)–sulfur bond.⁴⁵ A peak at 164.8 eV is attributed to the formation of the C─S bond, suggesting the infiltration of sulfur into the exfoliated CC matrix as shown in the inset of the same figure.⁴⁶ Additionally, S 2p doublets at 170.37 and 169.16 eV correspond to superficial sulfur oxidation with a high valence state of Ni–S or Co–S resulting in SO₄^2– species and ─SO₃─C─ bonds, respectively, suggesting the formation of cotton-like metal hydroxysulfide composite.^47,48 Likewise, similar peaks for the NiCCZS sample can be seen in the same image with a slight positive shift for all the peaks. Hence proportionate negative peak shift of S 2p core level spectra for NiCCZOS-4 in comparison with NiCCZS suggests the increased electron cloud over the sulfur species in the prior sample. The incorporation of the O atom owing to the extra partial oxidation step ensures the formation of more active Ni/Co─S dangling bonds, which will improve the OH–/H₂O* adsorption for enhanced electrocatalytic activity.³³ To reiterate, the oxygen functionalization step ensures more sulfurization sites in otherwise poorly sulfurized Ni/Co alloy by facilitating the partial removal of well-homogenized O by S, ensuring the enhanced electrocatalytic active sites in the active material, beneficial for energy storage and conversion applications both. Figure 3I shows N 1s core level spectrum of NiCCZOS-4 samples showing the distinct pyrrolic, and pyridinic nitrogen peaks at 399.95 and 394.57 eV of BE, respectively. Thus present nitrogen has a multitude of benefits in both the energy storage and conversion applications such as improved electron transport ability and catalyst/electrolyte Schottky barrier reduction, improved reaction kinetics by electronic structure regulation of reaction intermediates, an increased proportion of covalent bonds owing to the lower electronegativity and hence improved chemical stability of the electrodes and so on.⁴⁹

The O 1s XPS spectra of several samples are plotted and presented in Figure 4. As seen in Figure 4A, the NiCCZ sample is mainly composed of defects-rich oxygen peak at BE of 531.28 eV, owing to the oxygen deficiency provided by reducing ambiance during calcination. Besides the defect peak, a prominent lattice oxygen peak at BE of 529.47 and the resultant peak from C─O bonds and/or superficially adsorbed moisture can be seen at BE of 532.48 eV.⁵⁰

Figure 4. O 1s XPS spectra of different samples (A) before and (B) after sulfurization. (C) The atomic weight percentage of S and O elements of different samples, and (D) schematic illustration of oxygen functionalization-assisted anion exchange toward improved sulfurization.

The remaining XPS spectrums of different oxygen functionalized samples clearly show that the oxygen deficiency is decreasing and the lattice oxygen peak and O–C/O–H peak are increasing with increasing the degree of oxygen functionalization as anticipated. The 200°C functionalized sample shows a sudden increase in O–H/O–C peak intensity most probably arising from oxidation of the CC substrate. Whereas, as the temperature increased to 200°C and 30°C, the lattice oxygen peak intensity increased along with a gradual reduction of O–H/O–C peak intensity suggesting proper oxidation of metal nanoseeds. Figure 4B shows O 1s XPS spectrums of different samples after sulfurization. In this case, only two discernible peaks are observed attributed to the metal hydroxyl group and adsorbed moisture.⁵¹ Since sulfur anions have a higher polarization degree compared to oxygen, they can afford more electrons to the metal cations promoting better electronic modulation.⁵² Hence, the gradual negative peak shift concerning the degree of oxygen functionalization suggests the increased surface electron density owing to the higher degree of sulfurization. The O 1s XPS peaks and their corresponding shifts agree with the XPS data presented earlier for metals and sulfur species, suggesting better sulfurization with oxygen functionalization ensuing the optimized electronic structure, contributing toward the conception of a superior electrocatalyst. Figure 4C shows the O and S atomic weight percentages of various electrodes with respect to the degree of oxygen functionalization. As hoped, the sulfur concentration increases with the degree of oxidation of metal alloys. In the meantime, contrary to some reports, the oxygen concentration also increases along with sulfur.^31,53 This could be due to the presence of the superficial passivation layer formed over metal oxide due to spontaneous surface oxidation while keeping at ambient conditions. Nonetheless, the linear increase in sulfur concentration holds useful meaning as verified from XPS analysis as well and is expected to have a crucial effect in energy storage and conversion application. Hence, Figure 4D schematically summarizes the rather significant role of the oxygen functionalization steps as oxygen greatly facilitates the anionic exchange process to form stoichiometric metal chalcogenide heterostructure.

Electrochemical results Supercapacitor

The electrodes prepared as discussed in the experimental part were characterized electrochemically in three electrodes and two electrode configurations for supercapacitor application. Series of cyclic voltammetry (CV), galvanostatic charge–discharge (GCD), and electrochemical impedance spectroscopy (EIS) were conducted for each electrode to find out the electrochemically best possible electrode for the asymmetric supercapacitor device (ASCD). Figure 5A shows the CV curves of different electrodes namely, NiCCZ, NiCCZO, NiCCZS, and NiCCZOS; ran within the potential window of 0–0.7 V versus Hg/HgO counter electrode at a scan rate of 5 mV s^–1. Whereas the inset of the same figure shows the zoomed-in CV curves of NiCCZ and NiCCZO. Likewise, Figure 5B shows discharge curves of the same electrodes obtained from GCD analysis. The CV and GCD analyses of said electrodes highlight the importance of metal oxide sulfurization and the importance of partial oxidation in that. As expected, the lack of electrochemically active sites in the case of NiCCZ, basically a Ni/Co alloy, renders it unsuitable for supercapacitor application as evidenced by the negligible redox peaks in CV analysis and poor discharge time in GCD. Even the partial oxidation at 400°C (NiCCZO) could not provide satisfactory performance as seen in the CV and GCD results.

Figure 5. (A) CV curves of different electrodes under a potential window of 0–0.7 V at 5 mV s–1 scan rate. (B) Discharge curves of different electrodes from GCD analysis at 1 mA cm–2 current density. (C) CV curves of different oxygen functionalization samples at 5 mV s–1 scan rate. (D) Discharge curves of different oxygen functionalized samples at 1 mA cm–2 current density. (E) CV curves of NiCCZOS-4 electrode at different scan rates. (F) GCD curves of NiCCZOS-4 electrode at different current densities. (G) Linear fitting of logarithmic oxidation peak current for different electrodes with line parameters in the inset table. (H) b-Value bar diagram of different electrodes along with oxidation peak current linearity. (I) Area-specific capacity of different electrodes over different current densities.

The metal oxide in itself is believed to have poor electrochemical abilities, particularly in the case of supercapacitor application, such as poor electrical conductivity, and sluggish ion diffusion in the bulk phase, hence the inadequate performance of the NiCCZO electrode. On the contrary, the hydrothermal sulfurization of the NiCCZ sample seems to provide substantially improved electrochemical performance as seen in CV and GCD discharge curves of the NiCCZS electrode. The main reasoning behind this is the formation of electrochemically rich metal hydroxide composite and partial sulfurization of passivating oxidation layer of metal oxide. The formation of metal hydroxide and partial sulfurization was evident in several physical characterizations such as XRD and XPS in the physical characterization section. Likewise, the NiCCZOS electrode fabricated using sulfurization after simple oxygen functionalization (400°C at air) step produced the best result with substantial electrochemical improvements as evidenced by the CV and GCD analysis. Both the CV redox peaks and discharge curve of NiCCZOS are almost twice the magnitude when compared with NiCCZS. The superior performance of the NiCCZOS electrode is facilitated by the homogeneous sulfurization of Ni/Co owing to the optimum oxidation, also backed by the physical characterization results. Once the superiority of NiCCZOS is established, all the electrochemical analysis hereafter is conducted only in oxygen-functionalized/sulfurized electrodes.

Different temperatures were employed to synthesize different electrodes in an ambient condition to further illustrate the impact of partial oxidation on electrochemical performance. Then obtained electrodes were sulfurized with a hydrothermal sulfurization process to get an optimized NiCCZOS electrode. Figure 5C shows CV curves of such electrodes under a scan rate of 5 mV s^–1 scan rate within the potential window of 0–0.7 V versus the Hg/HgO counter electrode. As seen in the figure, NiCCZOS-4 shows a maximum CV area suggesting its superior charge storage ability. Whereas NiCCZOS-3 and NiCCZOS-2 have, respectively, inferior CV areas suggesting the straightforward notion that with reduced oxidation temperature, less homogenous metal oxide sites and subsequently lesser sulfur-enriched electrochemical active sites are formed. This notion is quantitatively verified by XPS analysis as discussed earlier. The superiority of NiCCZOS-4 can also be seen in the discharge curves of GCD analysis as shown in Figure 5D. Furthermore, the CV of NiCCZOS-4 over different scan rates can be seen in Figure 5E. The ideal pseudocapacitive nature of charge storage can be established as evident by the presence of prominent redox peaks over different scans governed by the well-known faradic redox reactions.¹⁰ [Image Omitted. See PDF] [Image Omitted. See PDF] [Image Omitted. See PDF] [Image Omitted. See PDF]

As the redox reaction suggests, the metal sulfide and metal hydroxysulfide cites are the actively participating entities and in our case, a meticulously optimized electrode with rich active cites is delivering superior charge storage performance as evident from comparative CV analysis as well. Furthermore, the increase in anodic and cathodic peak intensity with respective scan rates suggests the low resistance and faster charge transfer kinetics in the electrode–electrolyte interface. Whereas the positive shift of the anodic peak and negative shift of the cathodic peak with respect to the increasing scan rates hint toward the insufficient time for ion intercalation into dense electrode nanocluster.⁵⁴ Likewise, the GCD was performed in the NiCCZOS-4 electrode and respective charge–discharge plots from 1 mA cm^─2 can be seen in Figure 5F. GCD as well followed the pseudocapacitive nonlinear discharge curves analogous to the CV analysis supporting the pseudocapacitive nature of the material.

The diffusion-controlled or surface adsorption-controlled nature of charge storage was determined by employing the methods described in Section S8. As governed by Equation (S13), the slope of a linear line drawn in the plot of log(i) versus log(v), as shown in Figure 5G, will give the b-value. Here, the log of oxidation peak current for electrodes NiCCZOS-2, NiCCZOS-3, and NiCCZOS-4 was plotted against the log of respective scan rates from 1 to 5 mV s^–1. The linear fit of the respective points gave a slope of the line (b-value) as shown in the inset of Figure 5G. As obtained b-values are plotted in Figure 5H inset for better perusal. Whereas the oxidation half of CV for the NiCCZOS-4 electrode in Figure 5H supports the proper linear fitting of peak currents as they seem to have linear increments. The b-values are inversely proportional to the degree of partial oxidation with 0.76 for NiCCZOS-2, 0.69 for NiCCZOS-3, and 0.67 for NiCCZOS-4, suggesting the dominant pseudocapacitive (diffusion-controlled) contribution with increasing degree of oxidation supporting the characteristic nonlinear CV and GCD curves. This could be due to a couple of reasons such as the oxygen functional group itself boosting the pseudocapacitive charge storage and/or the increased oxygen concentration facilitating the better sulfurization of metal oxide and hence increasing the pseudocapacitance with a multitude of reasonings as discussed earlier in previous sections. The latter seems to be the logical explanation as the pseudocapacitive contribution from oxygen functional groups after the sulfurization should be negligible if compared with metal sulfide active sites. This also explains the small increase in surface adsorption-controlled contribution in NiCCZOS-2 and NiCCZOS-3 electrodes owing to the slightly poor sulfurization. Hence with the b-value analysis, it can be concluded that only through the proper oxygen functionalization, the pseudocapacitive charge storage, which is obviously the dominating charge storage phenomenon in metal oxide/chalcogenides, can be increased, improving the overall charge storage capability of the electrode.

The specific capacity of different electrodes is calculated using the Equations (S2–S4) and the result is plotted as shown in Figure 5I along with gravimetric capacity and faradic capacitance of an optimized electrode in Figure S11. As evidenced by the CV and GCD results of the respective electrodes, the specific capacity is in increasing order with the increased degree of oxygen functionalization. The NiCCZOS-4 electrode has registered the maximum area-specific capacity of 555.35 µAh cm^–2 at a current density of 1 mA cm^–2 and still maintains it to 390.88 µAh cm^–2 even at a high current density of 10 mA cm^–2. Whereas the NiCCZOS-3 and NiCCZOS-2 have 438.84 and 317.98 µAh cm^–2, respectively, at 1 mA cm^–2. Despite the best electrochemical performance of NiCCZOS-4s, a slight decline in its performance toward the higher current density could be due to the insufficient time available for proper ion diffusion in its electrochemically rich sites, when the current increases.⁵⁴ Furthermore, metal sulfide nanoflowers uniquely embedded inside the CC ensure a higher substrate/active material contact area and hence facilitate the enhanced electron flow as depicted in the inset scheme of Figure 5I. In addition, the exfoliated outer layer of CC owing to the high-temperature treatment generated more active sites and better electrolyte dispersion resulting in enhanced charge storage. Hence, in addition to the optimum sulfurization of metal clusters owing to meticulous oxygen functionalization, the synergistic effect of flower-like metal chalcogenide nanobuds, their unique embedding inside the CC, and exfoliated CC substrate contribute to the enhanced charge storage of NiCCZOS-4 electrode.

The superiority of the NiCCZOS-4 electrode was further explored in light of the electrochemical charge transfer property with EIS analysis as shown in the resulting Nyquist plot (Figure S12). The high-frequency region of all electrodes is composed of a semicircle (inset of the same figure) led by a straight line toward the low-frequency region. All the EIS Nyquist plots were fitted with equivalent electrical elements and the fitting circuit can be seen in the inset of the same figure. Among the circuit elements, the equivalent series resistance (R_s) and the charge transfer resistance (R_ct) are of particular interest as they represent the equivalent series resistance and rate of redox reaction at the electrode–electrolyte interface, respectively.⁵⁵ The smallest R_s and R_ct values for NiCCZOS-4 verify its superior charge transfer ability and better conductivity resulting in the best electrochemical performance.

Owing to the superior performance of NiCCZOS-4 in a three-electrode configuration, it is implemented as a cathode half to realize the ASCD and will be mentioned as NiCCZOS hereafter, unless stated otherwise. The anodic half of the ASCD is composed of a cherry flower biowaste-derived activated carbon (CFAC) coated over a CC. The CV, GCD, and specific capacitance of CFAC are shown in Figure S13A–C in the ESI file for better perusal. Furthermore, the supercapacitive performance along with all the relevant physical characterization analysis of CFAC is extensively optimized and reported in our earlier work.³⁴ The CFAC electrode preparation method is explained in detail in Section S6 along with the relevant anode–cathode charge balance computations in Section S7.

The CVs of CFAC and NiCCZOS-4 at 5 mV s^–1 are shown in Figure S13D to verify their respective contribution toward determining ASCD working potential window. Figure 6A represents the artistic representation of the most probable charge storage and electron flow mechanism in ASCD configuration. On the cathodic side, the flower-like metal sulfide composite with its cottony hydroxysulfide nanoexterior and exfoliated CC outer layer provided tremendous amounts of active sites for ion diffusions. In the meantime, the metal sulfide composite, well-lodged inside CC facilitates superior electron transfer pathways owing to the better active material/current collector bonding. On the anodic side, the porous CFAC electrode with its ultra-large surface area (∼2500 m² g^–1) provided a profusion of active sites for charge counterbalancing. Figure 6B shows the CV curves of NiCCZOS//CFAC ASCD over different potential windows all conducted at 10 mV s^–1 of scan rate to determine the working potential window of the device. It is evident from the CV curve that 1.7 V is the reliable potential window to further work on; as a small polarization peak, most probably from oxygen evolution, is observed at 1.8 V. Hence the CV curves of ASCD at different scan rates were measured within the potential window of 0–1.7 V and plotted for 5–30 mV s^–1 as shown in Figure 6C. The quasi-rectangular CV curves with a hint of redox peaks demonstrate a well-balanced CFAC and NiCCZOS-4 contribution of electric double layer capacitor (EDLC) and pseudocapacitive nature, respectively. Furthermore, the gradual increment of CV area toward a higher scan rate with no change in its nature suggests an excellent reversible charge storage ability of ASCD. Likewise, the GCD analysis performed on the same ASCD is presented in Figure 6D, showing its quasi-linear charge–discharge nature supporting the CV analysis results. The area-specific and mass-specific capacity of the ASCD were calculated from the galvanostatic discharge data employing the formulae presented in Equations (S5) and (S6) and the result is presented in Figure 6E. Promisingly, the ASCD shows an area-specific capacity of 318.8 µAh cm^–2 at the current density of 2 mA cm^–2, which is analogous to the gravimetric capacity of 71.7 mAh g^–1 at the same current density. Even at higher current density (10 mA cm^–2), more than 68% of initial capacity remained with 217.8 µAh cm^–2 (49 mAh g^–1) showing an excellent rate capability of as-fabricated ASCD. The inferior electronegativity of sulfur, when compared with oxygen, renders the optimally sulfurized NiCCZOS electrode highly electroactive and hence facilitates superior electrochemical performance. Furthermore, the synergistic contribution of two metal ions in redox reactions ensures enhanced ionic mobility owing to the multiple oxidation states such as Co²⁺ to Co³⁺ and Ni²⁺ to Ni³⁺ oxidation half during charging cycles and its counter-reaction during subsequent reduction half. In the anodic half, the CFAC electrode with its large surface area can accommodate a substantial amount of ions to counterbalance the charge, mainly via surface-controlled adsorption–desorption phenomena also known as EDLC charge storage. The contribution of surface-controlled (capacitive) charge storage and diffusion-controlled charge storage in the total capacity of ASCD is calculated using the technique mentioned in Section S8.

Figure 6. (A) Schematic representation of most probable charge storage and electron flow mechanism in ASCD. (B) CV curves of ASCD over different potential windows, all conducted at 10 mV s–1. (C) CV curves of ASCD at different scan rates within the potential window of 0–1.7 V. (D) GCD curves of ASCD over different current densities. (E) Area-specific and mass-specific capacity of ASCD over different scan rates. (F) Linear fitting of the square root of anodic and cathodic peak current with the square root of scan rates to determine capacitive and diffusion-controlled contribution, with corresponding CV curves of ASCD in the inset. (G) Diffusion-controlled contribution CV of ASCD as compared to total current at 6 mV s–1 scan rate. (H) Bar diagram of diffusion-controlled and capacitive contribution in total charge storage of ASCD with respect to the CV scan rates. (I) Ragone plots showing area-specific and mass-specific energy density and power density of as-fabricated ASCD while comparing with similar reported results, and (J) capacity retention and Coulombic efficiency stability test of ASCD over 45,000 charge–discharge cycles with several intermediate GCD cycles in the inset.

Figure 6F represents the linear fitting of different anodic and cathodic scans as depicted in the inset of the same figure. The diffusion-controlled contribution $$({a}_{2}{v}^{\frac{1}{2}})$$ plotted with the resultant intercept of linear fitting from Figure 6F is plotted in contrast with the total current at 6 mV s^–1 of scan rate in Figure 6G. Subsequently, the capacitive and diffusion-controlled contribution for the NiCCZOS//CFAC electrode in relation to the increasing scan rate is presented in Figure 6H. Interestingly, the ∼60–30 contribution ratio seems to be consistent from 6 to 10 mV s^–1 with diffusion-controlled being the dominant one. The dominance of the diffusion process contrary to the anticipated equal capacitive contribution from EDLC-type CFAC anode can be attributed to the presence of oxygen and nitrogen functional groups presented in the biomass-derived CFAC. The nitrogen and oxygen functional groups both contributed to the pseudocapacitive nature of charge storage and hence the superior diffusion-controlled contribution. Nonetheless, the pseudocapacitive contribution from the anode has also contributed to the higher charge storage of ASCD, which would have been inferior, had there been an ideal capacitive nature of the anode. The ED and PD of ASCD were calculated and plotted in a Ragone plot as shown in Figure 6I. For better comparability with past and possibly future research, both gravimetric and areal ED and PD were calculated using Equations (S7–S10). The gravimetric ED of NiCCZOS//CFAC ASCD is obtained as 59.4 Wh kg^–1 at a PD of 292.5 W kg^–1, which remained at 25.4 Wh kg^–1 even at a high PD of 3 kW kg^–1. As obtained ED and PD are higher and comparable to the reported results as mentioned in Table S1. Likewise, The ASCD device shows high areal ED of 263.8 µWh cm^–2 at the PD of 1.3 mW cm^–2 and maintains an ED of 112.1 µWh cm^–2 even at a high PD of 13.5 mW cm^–2. This is considerably higher and comparable to the previously reported similar electrode configurations, which are also plotted in the corresponding figure and tabulated in Table S1 for better perusal. The superior performance of the NiCCZOS//CFAC electrode is attributed to the flower-like nanostructures having an electrochemically rich amorphous hydroxysulfide exterior facilitating enhanced ion diffusion. Furthermore, the embedding of such active sites inside the current collector ensures superior electron transfer, synergistically contributing to better device performance from the cathodic site. Whereas, on the anodic side the ultra-large specific surface area along with abundantly present nitrogen and oxygen functional groups, the biomass-derived CFAC provides supporting charge storage mainly via ionic adsorption–desorption.

The long-term stability test of ASCD was performed under a GCD protocol at a high current density of 60 mA cm^–2. The NiCCZOS//CFAC ASCD shows astounding stability with virtually no capacity fading (108% capacity retention and 92% Coulombic efficiency) even after 45,000 repetitive charge–discharge cycles. The higher than 100% capacity retention is attributed to the activation of more electrochemically active sites after several cycles mainly due to proper electrolyte diffusion over all surfaces.^34,56 It is evident from Figure 6J that, full activation of active materials is achieved around 10,000 cycles, after which capacity seems to be going down slightly. Capacity drop in the first few cycles can mainly be ascribed to poor wettability and incomplete utilization of active material. The subsequent increase in capacitance can be related to an improvement in the surface wetting of the electrode by the electrolyte during extended cycling. Interestingly several similar results have been reported mainly on nickel- and cobalt-based active material works.^57–59 Nonetheless, even if the capacity retention is determined with the 10,000th cycle as a starting point, owing to the complete active site activation, 72% of capacity retention is obtained at the end of the 45,000th cycle, which still is a great result. This ultra-long life stability performance of CC-supported self-grown binderless metal chalcogenide is one of the best among reported results for ASCD performance up to date. It should be noteworthy that this remarkable feat is achieved just by uniquely designed morphology of metal oxide embedded CC without any additional binding agent. This highlights the ASCDs practicability and possible commercial producibility. The extraordinary stability of ASCD is attributed to the firm embedding of flower-like metal chalcogenide nanoseeds inside the CC substrate. Unlike the surface-grown nanostructures, embedding them inside the substrate makes them much less susceptible to the dislodgment caused by the restructuring of nanostructures owing to the repetitive redox reactions. The sulfurization performed after oxygen functionalization ensured a more robust fitting of metal chalcogenide seeds inside the conductive substrate by slightly increasing the size of already embedded metal alloy nanoseeds. This can also be verified by the FE-SEM analysis, where, well grown flower-like structure with no vacancy between metal chalcogenide and CC substrate was observed after sulfurization, and in TEM analysis a cotton-like amorphous hydroxysulfide outer layer was observed, contributing toward ultra-stable cyclic performance. The performance of as devised NiCCZOS//CFAC ASCD is tabulated (Table S1) alongside similar works to further highlight its superiority.

The physical morphology of the NiCCZOS electrode after the 45,000 cyclic stability test was observed with FE-SEM analysis and corresponding images can be seen in Figure S14A–C. The metal chalcogenide nanoseeds seem to have gone under a major restructuring owing to the repetitive redox reactions as the homogenous flower-like structure is no longer noticeable. This is mainly due to the restructuring of the Ni/Co hydroxysulfide outer layer toward a more stable and active phase owing to the redox activities.³¹ Nonetheless, a copious amount of active material can be seen still attached to the CC, which compliments the unhinged cyclic stability test result. Furthermore, the EDS color mapping and elemental spectrum of the same sample in Figure S14D show the presence of all the relevant elements such as Ni, Co, S, O, and C. The sulfur concentration seems to have reduced slightly, which can be attributed to the phase change of hydroxysulfide to hydroxide owing to the repetitive charge–discharge induced catalyst restructuring.⁶⁰ The HR-TEM analysis of the same sample as shown in Figure S14E–G confirms the FE-SEM findings, as no flower-like structures are observed. Instead, the analysis reveals the presence of nano-sized NiCo₂S₄ crystal phases in what appears to be a porous NiCCZOS nanocomposite. Furthermore, Figure S15 shows the EIS spectrum of the NiCCZOS//CFAC ASCD before and after the stability test. The smaller semicircle in the high-frequency region after the long-term stability test clearly suggests the reduced solution resistance, as well as the charge transfer resistance as compared to the EIS performed before cyclic stability. This is mainly due to the activation of more electrochemically active sites after the cyclic stability test. Furthermore, the less steep straight line toward the low-frequency region in the case of EIS after cyclic stability suggests the easier diffusion of ionic species inside the active material which again is facilitated by the enhanced active material activation.

To further demonstrate the practicability of NiCCZOS//CFAC ASCD, the device is fabricated with vacuum sealing inside the plastic pouch, sandwiching a negative electrode, a positive electrode, and cellulose paper as a separator. Figure 7A shows a side view of such vacuum sealed pouch cell where two ASCDs are connected in parallel to increase the stored charge quantity. Figure 7B shows the top view of the same device emphasizing its petite thickness. Figure 7C,D demonstrates the flexibility of fabricated ASCD in the form of twisting and bending, respectively. CV was run for the as-fabricated ASCD in normal, twisted, and bent conditions to analyze its firmness in adverse conditions. The resulting CV presented in Figure 7E shows virtually no change in the characteristic CV curves highlighting the strength of the ASCD. Figure 7F–H shows the illumination of three 3 V light-emitting diodes (LEDs) connected in parallel, which were powered by two serially connected ASCD pouch cells fabricated earlier. Figure 7I shows the operation of a 3 V DC motor-operated fan, powered by the same ASCD configuration highlighting the versatility of fabricated NiCCZOS//CFAC ASCD. Hence, NiCCZOS has been presented as an excellent supercapacitor electrode having high electrochemical capabilities and outstanding stability, establishing itself as an excellent supercapacitor electrode for possible practical applications.

Figure 7. (A) Side view and (B) top view of NiCCZOS//CFAC ASCD fabricated inside the vacuum-sealed pouch cell. (C) Twisting and (D) bending of fabricated ASCD. (E) CV of ASCD under normal and flexible conditions. (F–I) Operation of LED and DC motor-operated fan powered by ASCD.

Water splitting

Inspired by the abundance in electrocatalytically active sites verified by physicochemical and supercapacitor results, the fabricated electrodes were subjected to OER as working electrodes in 1 M KOH solution. The polarization curves for all electrodes were collected from higher potential to lower potential to avoid the oxidation peak of metal chalcogenide nanocluster, Co²⁺ to Co³⁺ and Ni²⁺ to Ni³⁺ in particular, which was also evident in the CV analysis earlier.^31,33 As shown in Figure 8A, the NiCCZOS-4 electrode has the lowest overpotential (η) of 271 mV at 10 mA cm^–2 among all the fabricated electrodes used in this study, making it the second-best performer only after commercial IrO₂, which exhibits the η of 197 mV at the same current density. In the meantime, the NiCCZS, NiCCZO, and NiCCZ electrodes exhibited the overpotential of 299.3, 363.1, and 506.0 mV, respectively at 10 mA cm^–2. The η of all the electrodes over different current densities was plotted and presented in Figure 8B for better perusal. The NiCCZOS-4 has maintained the low η trend with 346 mV of η at a high current density of 50 mA cm^–2 showing its excellent electrocatalytic capability. The synergistic effect of S²⁻ and OH⁻ of NiCCZOS-4 facilitates the adsorption of OER intermediates over the electrocatalyst surface, resulting in superior performance. The reaction kinetics of as-fabricated electrodes and their efficiency in producing current per applied potential was further analyzed using Tafel slopes as shown in Figure 8C. The Tafel slope 71.1 mV dec^–1 of NiCCZOS-4 is lower than that of all other electrodes such as 100.9 mV dec^–1 for NiCCZS, 134.1 mV dec^–1 for NiCCZO, and 166.1 mV dec^–1 for NiCCZ. This explains the low overpotential required for NiCCZOS-4 to achieve higher current further demonstrating its rapid OER kinetics. The meticulously optimized oxygen/sulfur concentration and hence electrocatalytically rich active sites, flower-like active material nanoarchitecture, exfoliated current collector, and firm embedding of active material inside the current collector, all contribute to the superior performance of the NiCCZOS-4 electrode. As discussed earlier in the physical characterization section, the negative shift of Ni 2p and Co 2p BE after optimum sulfurization suggests a decrease in metal oxidation states hinting toward the anionic modulation and alteration of electronic properties of metals in a way that promotes the breaking of metal–oxygen bond and releases the oxygen, resulting in superior OER activity.^53,61 Furthermore, owing to the higher polarization ability and lower electronegativity of sulfur when compared with oxygen, the partial replacement of oxygen with sulfur facilitates efficient H₂O adsorption over the NiCCZOS surface improving the OH* adsorption and OOH* desorption and hence optimal OER activity.^19,31 The superiority of the NiCCZOS electrode as an OER electrocatalyst over similar electrocatalysts is compared and tabulated in Table S2.

Figure 8. OER performance of the as-prepared electrodes. (A) Polarization curves at a 10 mV s–1 scan rate, (B) comparative overpotential bar diagram over different current densities, and (C) Tafel slopes. HER performance of as-prepared electrodes: (D) polarization curves at a 10 mV s–1 scan rate, (E) Tafel slopes, and (F) comparative overpotential bar diagram over different current densities. (G) Chronopotentiometry stability test of both HER and OER application for 30 h.

EIS was carried out to further elucidate the OER kinetics and the resulting data is presented in Figure S16. The electrical circuit fitted with equivalent elements is presented in the inset of the same figure. As verified by the EIS data and supported by the equivalent electrical circuit, the EIS is composed of two semicircles which can be attributed to the charge transfer resistance mainly due to the oxidation of pseudocapacitive active material ($${R}_{\mathrm{CT}}^{\mathrm{Redox}}$$) and OER activity ($${R}_{\mathrm{CT}}^{\mathrm{OER}}$$).⁶² The $${R}_{\mathrm{CT}}^{\mathrm{OER}}$$, in this case, is of particular interest as it signifies the electron transfer process during OER. The NiCCZOS-4 electrode shows the lowest $${R}_{\mathrm{CT}}^{\mathrm{OER}}$$ among all electrodes as evident in the EIS plot and also presented in Table S3. As discussed earlier, along with the high polarizability and low electronegativity than oxygen, the sulfur incorporation additionally produces more amorphous hydroxysulfide active layers and strongly coupled heterointerfaces possibly altering the electronic structure, and synergistically contributing toward better charge transfer. Furthermore, the alloy core with metallic properties can expedite charge transportation to the active hydroxysulfide shell.⁶³

HER was conducted in a similar three-electrode configuration as in OER to determine the bifunctional ability of the as-fabricated materials. Figure 8D shows the polarization curves of different samples, from where it can be seen that the NiCCZOS-4 electrode has the best performance among the fabricated ones with 168.4 mV of η at 10 mA cm^–2 current density. Whereas the commercial Pt/C electrode has the lowest η of 69.9 mV at the same current density. On the other hand, the NiCCZ, NiCCZO, and NiCCZS all have inferior η with 182.0, 306.7, and 395.0 mV at a 10 mA cm^–2 current density highlighting the positive impact of oxygen functionalization on NiCCZOS-4 before sulfurization. Figure 8F depicts the bar diagram of η versus current density for different samples and a linear increment in η value can be seen for all samples for increasing current suggesting the need for larger activation energy to achieve HER. Nonetheless, the NiCCZOS-4 electrode still maintained the overpotential of 221.9 mV even at a very high current density of 50 mA cm^–2. Tafel slopes were plotted for all electrodes to determine the reaction kinetics of HER as shown in Figure 8E. Besides the Tafel slope of 36.4 mV dec^–1 for the commercial Pt/C electrode, the NiCCZOS-4 shows the lowest value among the fabricated electrodes with a Tafel slope of 77.8 mV dec^–1. This is much lower than NiCCZ (149.6 mV dec^–1), NiCCZO (131.4 mV dec^–1), and NiCCZS (119.4 mV dec^–1), verifying the excellent HER kinetics of NiCCZOS-4 electrode. The reaction pathways for HER reaction in alkaline media are proposed to be composed of Volmer–Heyrovsky process or Volmer–Tafel pathways.^64,65 In both reaction mechanisms, the crucial part is the adsorption of H₂O molecules over the catalyst surface followed by the reduction of that H₂O into OH⁻ and H, leading to the desorption of OH⁻ refreshing the catalyst surface and H adsorbed intermediate formation for H₂ generation.⁶⁶ In this regard, the firmly embedded flower-like metal hydroxysulfide structure with metal alloy core along with the exfoliated carbon substrate synergistically provides an abundance of active sites, contributing toward enhanced electron/ion mobility and hence outstanding HER activity for the NiCCZOS-4 electrode. Furthermore, as evidenced by HRTEM analysis, the defect and disorder-rich catalyst with cotton-like hydroxysulfide amorphous outer layer provide the hospitable environment for HER reaction to proceed in a nanoscale. The superiority of the NiCCZOS electrode as an HER electrocatalyst over similar electrocatalysts is compared and tabulated in Table S4. To further elucidate the charge transfer kinetics of as-fabricated electrodes, EIS was performed, and the resulting plots fitted with equivalent electrical elements are shown in Figure S17. The charge transfer resistance associated with HER reaction for NiCCZOS-4 electrode is the lowest one as evident from Table S5, suggesting its superior electrocatalytic property among all others, verifying the overpotential results mentioned earlier. To further establish the commercial viability of NiCCZOS-4 as a well-optimized bi-functional electrocatalyst, chronopotentiometry stability tests were conducted for 30 h at 50 mA cm^–2 for HER and OER individually, as shown in Figure 8G. Virtually no significant change was observed in the stability test for both reactions suggesting an outstanding resilience of a prepared electrocatalyst toward robust performance. Unique embedding of electrocatalyst nanoclusters inside the CC current collector aided in its excellent stability performance. The positioning of metal alloys embedded inside the CC was further well established by the oxygen functionalization-driven sulfurization process, synergistically providing stable electrocatalytic performance. Hence NiCCZOS is presented as a viable bifunctional electrocatalyst for water-splitting work with possible practicability.

Furthermore, the electrochemical surface area (ECSA) of different catalysts was interpreted by determining the double-layer capacitance (C_dl) based on the non-Faradaic regions of the CV curves at the scan rates ranging from 10 to 100 mV s⁻¹ as demonstrated in Figure S18A–F. As governed by Equation (7), the ECSA is directly proportional to the C_dl hence the electrocatalytic activity of as-fabricated electrodes is interpreted with respect to the C_dl values to avoid any misinterpretation in ECSA possessed by the uncertainty in specific capacitance of the material (C_s).⁶⁷ [Image Omitted. See PDF]

As confirmed by the slopes of different samples from Figure S18G, the NiCCZOS has the largest C_dl of 3.15 mF cm⁻² compared to NiCCZO-4 (2.16 mF cm⁻²), NiCCZO-3 (1.98 mF cm⁻²), NiCCZO-2 (1.9 mF cm⁻²), NiCCZ (0.14 mF cm⁻²), and bare CC (0.09 mF cm⁻²). The highest C_dl and subsequently the highest ECSA for NiCCZOS are acquired by virtue of stoichiometrically formed flower-like nickel–cobalt hydroxysulfide structures. Likewise, almost 15 times increment in C_dl for the oxygen-functionalized sample, when compared to the NiCCZ sample (the sample just after electrocatalyst embedding and before oxygen functionalization), is mostly a result of the introduction of superficial oxygen functional group over CC. The simple air calcination of commercial CC has been found to increase its specific surface area by around 700-fold allowing improved charge storage.⁶⁸ Furthermore, the turnover frequency (TOF) of different electrocatalysts was calculated employing the methods explained in Section S9. As suggested by Figure S19B the TOF of NiCCZOS is two times and three times higher than the NiCCZS and NiCCZO, respectively, validating the highest electrocatalytic activity for the NiCCZOS electrode.

Theoretical modeling

The computational details are described in Section S10 of the ESI. First, the formation of NiCo₂S₄ from the CoNi alloy and the oxidation of H₂S to S, as well as the exchange of O²⁻ and S²⁻ anions in the formation of NiCo₂S₄, have been studied by density functional theory (DFT) calculation. The results, obtained through DFT calculation, are presented in the equations shown below. [Image Omitted. See PDF] [Image Omitted. See PDF] [Image Omitted. See PDF]

It was found that it is difficult to produce NiCo₂S₄ directly from the CoNi alloy, but it is easier to produce it through the exchange of anions process from NiCo₂O₄. The hypothesis is that in the formal process, the oxygen required to oxidize the CoNi alloy to Ni²⁺ and Co²⁺ in the final product competes with S²⁻. The thermodynamic analysis shows that oxidizing H₂S to elemental S (Reaction 2) is more favorable compared to the formation of NiCo₂S₄ (Reaction 1) where the ΔE values are −56.8 and −48.3 kcal mol^‒1 of O₂, respectively. On the other hand, the exchange anion process (Reaction 3) is very exergonic with a ΔE of −63.4 kcal mol^‒1 of NiCo₂O₄ and can avoid side reactions, making it a better synthetic path.

The density of state (DOS) and band gap of NiCo, NiCo₂O₄, and NiCo₂S₄ obtained by DFT calculation as shown in Figures S20 and S21 demonstrates that NiCo and NiCo₂S₄ are metal-like (no band gap) and NiCo₂O₄ is semiconductor-like (band gap = 0.1–0.2 eV). When exchanging a portion of sulfur with oxygen in the NiCo₂S₄ owing to the results suggested by experimental characterization of the optimized sample (NiCCZOS), the compound NiCo₂S₄ (40% S and 60% O) behaved as a semiconductor with a very small band gap of ∼0.03 eV highlighting its superior electrical conductivity among all the samples and hence the best electrocatalytic performance. Furthermore, Figure 9A shows the partial DOS of different samples with NiCCZOS having the highest DOS integral area around the Fermi level indicating that more charge is activated and transferred in its electrochemical interfaces.^69,70 Furthermore Figure 9B shows the H₂O and OH adsorption energy variations with respect to the increasing anionic replacement of oxygen by sulfur. The experimentally verified O:S concentration again prevailed in both H₂O and OH adsorption with the lowest adsorption energy facilitating the superior H₂O dissociation for the generation of hydrogen and oxygen as well as the ionic adsorption in the case of supercapacitor application accelerating the redox reaction kinetics.⁸ Furthermore, Figure 9C–E represents the charge density difference of the NiCO, NiCCZO, and NiCCZOS samples. The surface charge surrounding the metal atoms seems to increase greatly in the case of NiCCZOS when compared to NiCo and NiCo₂O₄ owing to the lower electronegativity of S relative to O facilitating easier ionic adsorption-desorption.^19,71

Figure 9. (A) Partial density of states of different samples. (B) H2O and OH adsorption energy variations with respect to the increasing anionic replacement of oxygen by sulfur, charge density difference plots of (C) NiCo, (D) NiCo2O4, and (E) NiCCZOS.

CONCLUSIONS

Flower-like transition metal chalcogenide “embedded” CC is prepared via oxygen functionalization enhanced anion exchange for ultra-long life flexible energy storage (supercapacitor) and conversion (water splitting) applications. As an ASCD, when conjugated with cherry flower biomass-derived nitrogen rich AC counter electrode, the NiCCZOS//CFAC device delivered 45,000 cycles of ultrastable charge–discharge performance with an admirable ED of 59.36 Wh kg^–1 (263.8 µWh cm^–2) at the PD of 292.46 W kg^–1 (1.3 mW cm^–2). Likewise, the NiCCZOS delivered a minimal overpotential of 271 mV for OER and 168.4 mV for HER application at a current density of 10 mA cm^–2 along with unhinged stability performance over 30 h of chronopotentiometry test for both applications. The superior performance of an optimized electrode is attributed to the flower-like nanostructures having an electrochemically rich amorphous hydroxysulfide exterior facilitating enhanced ion diffusion. Furthermore, embedding such active sites inside the current collector ensures superior electron transfer, synergistically contributing to better electrochemical and ultra-stable cyclic performance. Hence, an improved electrode fabrication technique resulting in superior electrochemical performance is presented highlighting its outstanding practicability and potential extension of ideas toward numerous other active material configurations contributing to advanced energy storage and conversion research.

ACKNOWLEDGMENTS

This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIT) (2021R1A4A2000934).

CONFLICT OF INTEREST STATEMENT

The authors declare that there are no conflicts of interests.