Recently, electrocatalysts have become of great interest in electrochemical reactions owing to their role in renewable energy conversion and storage systems, such as water splitting, metal-air batteries, and ion storage [1-3]. Oxygen evolution reaction (OER), oxygen reduction reaction (ORR) and hydrogen evolution reaction (HER) are the most significant and basest electrochemical reactions in the renewable energy system [4-6]. In addition, the OER and ORR performances indicate a reversible oxygen catalytic activity, which demonstrates the properties of metal-air batteries, such as Zn-air, Li-air, and Na-air batteries [7-9]. Traditionally, the best electrocatalysts for OER are Ir, Ru, and their oxides, and Pt is the most efficient one for ORR and HER. However, they are not widely used in industry, which can be attributed to high cost, scarcity, and poor durability [10]. Therefore, the design of ideal electrocatalysts based on inexpensive elements with a high current density at a low overpotential and long-term stability is important for the development of renewable energy systems [11]. To address these challenging issues, various transition-metal based catalysts and methods have been designed for the energy reactions, such as perovskite, metallic oxide, and their derivatives [12-20]. However, exploring efficient strategies to develop more active and stable electrocatalysts in a more economical way is still a scientific task in the development of renewable energy systems.

Transition metal dichalcogenides (TMDs), especially ultrathin nanosheets, such as MoS₂, TiS₂, TaS₂, WS₂, MoSe₂, WSe₂, etc., have received increasing research interest in the HER in recent years for their ideal atomic arrangements and high electrical conductivities [21-25]. In addition, many reports have stated that only the surface atoms of the TMD electrocatalysts can act as the active sites for the HER [26, 27]. Nevertheless, the reversible oxygen electrocatalytic performance of these TMDs is limited because it is difficult to develop effective strategies to tune their electronic structures. Recently, researchers have found that fabricating the nanointerface is an efficient way to optimize the electronic structure of various materials [15, 16]. The interface nanomaterials show optimized adsorption of the oxygen species (OH^-, O₂) and more catalytic sites for the different domains between the interface, which is crucial for electrocatalytic performance.

In this study, an efficient liquid exfoliation strategy was developed using formamide as the stripping reagent under continuous ultrasonic treatment (Scheme 1). The obtained FeS₂/NiS₂ hybrid nanocrystals showed two phase compositions and silk-like morphologies, which were confirmed using X-ray diffraction (XRD), transmission electron microscopy (TEM), and high-resolution transmission electron microscopy (HRTEM). The silk-like interface FeS₂/NiS₂ hybrid nanocrystals exhibited enhanced reversible oxygen catalytic performance with a low overpotential of 233 mV at j = 10 mA cm^-2 for the OER. This was an improvement over the recently reported relevant catalysts in 1.0 mol/L KOH with an onset potential of 911 mV and a half-wave potential of 640 mV for the ORR. Further, according to the potential of the OER and ORR, the reversible oxygen activity of the silk-like FeS₂/NiS₂ hybrid nanocrystals was calculated to be 0.823 V. Based on the excellent reversible oxygen performance, the homemade rechargeable Zn-air batteries using FeS₂/NiS₂ hybrid nanocrystals as the air-cathode displayed a high open-circuit voltage (OCV) of 1.25 V for more than 17 h and an excellent rechargeable performance for 25 h. In addition, the solid Zn-air batteries were successfully fabricated with an excellent rechargeable performance of approximately 15 h. This study provided a new method for designing interface nanocrystals with a unique morphology as efficient electrocatalysts for renewable energy systems.

The NiFe₂O₄ nanocrystals were obtained according to a previous study [13].

In a typical procedure, the NiFe₂O₄ nanocrystals were placed in a tube and heated to 500 ℃ with a rate of 10 ℃ min^-1 under a flowing N₂ atmosphere. After reacting for 2 h at 500 ℃, the system was cooled under a flowing N₂ atmosphere to room temperature naturally [15, 16].

The FeS₂/NiS₂ bulk powder (10 mg) was dispersed in formamide (30 mL) and ultrasonically treated in water with ice for more than 3 h. The silk-like FeS₂/NiS₂ nanocrystals and un-exfoliated product were obtained by centrifuging the solution at different rotational speeds, and then the solution was dried in a vacuum at 50 ℃ [17].

XRD experiments were conducted using an X'Pert Pro X-ray diffractometer with Cu K_α radiation (λ = 0.1542 nm) under a voltage of 40 kV from 20° to 90°. TEM and HRTEM observations were performed under an acceleration voltage of 200 kV with a JEOL JEM 2100 TEM. (XPS) analyses were performed with a VG ESCALAB 220I-XL device and corrected with a C 1s line at 284.6 eV.

All the electrochemical measurements were carried out in a three-electrode system consisting of a working electrode made by the corresponding catalysts, a Pt auxiliary electrode, and an Hg/HgO reference electrode connected to a CHI 760 E electrochemical workstation (CHI Instruments, Shanghai Chenhua Instrument Corp., China) at a scan rate of 2 mV s^-1 in 1.0 mol/L KOH solution. The potentials were referenced to RHE (E(RHE) = E(Hg/HgO) + 0.0951pH + 0.098). In addition, a resistance test was performed, and the iR compensation was applied using CHI software. In this study, the electrochemical experiments were conducted at 20 ± 0.2 ℃.

The electrochemical measurements for the ORR were conducted on a rotating ring-disk electrode (RRDE) at 1600 rpm in an oxygen-saturated KOH solution (0.1 mol/L). The ring-disk electrode (RDE) was measured at the rotating rates of 400–2400 rpm at the scan rate of 2 mV s^-1. The relationship between the measured currents (j) with various rotating speeds (ω) under fixed potentials was expressed on the basis of the K-L equation [15, 16].

(1)

The parameter j_k is the kinetic current, and ω is the electrode rotating rate. The parameter B is determined from the slope of the K-L plots based on the Levich equation below.

(2)

The Zn-air battery was assembled by the zinc plate anode. 25-30 mL of 6.0 mol/L KOH solution were used as the electrolyte without a separator. All the Zn-air batteries were tested under an ambient atmosphere. The polarization curve measurements were performed by LSV (2 mV s^-1) at 25 ℃ with a CHI 760 E electrochemical working station (CH Instrument). The current and power densities were normalized to the effective surface area of the air-cathode electrode [15, 16].

Ultrasonic assisted liquid phase exfoliation was successfully developed to prepare the silk-like interface FeS₂/NiS₂ hybrid nanocrystals (Scheme 1) [15-17]. Fig. 1(a) shows the XRD pattern of the obtained FeS₂/NiS₂ hybrid nanocrystals. The peaks of FeS₂/NiS₂ were attributed to FeS₂ and NiS₂, which belonged to the space groups of P1 (JCPDS card no. 71-1680; a = b = c = 5.417 Å) with a triclinic structure and Pa3 (JCPDS card no. 11-99; a = b = c = 5.670 Å) with a cubic structure, respectively. The XRD pattern showed the hybrid structure for the two different phases. In addition, the XRD patterns of the single FeS₂ and NiS₂ are shown in Figs. S1 and S2. The TEM images were measured to conform the morphology of the FeS₂/NiS₂ hybrid nanocrystals. Figs. 1(b) and (c) show the TEM images of the FeS₂/NiS₂ hybrid nanocrystals, illustrating the silk-like structure. The HRTEM image of the FeS₂/NiS₂ hybrid nanocrystals (Fig. 1(d)) showed two different domains of FeS₂ and NiS₂, in which the clearly identified lattice fringe spaces of 2.71 and 2.83 Åcorresponded to the same (200) plane of the triclinic FeS₂ and cubic NiS₂.

The surface properties of the silk-like interface FeS₂/NiS₂ hybrid nanocrystals were investigated using the X-ray photoelectron spectra (XPS) measurements. Fig. 2(a) shows the Fe 2p spectra of the FeS₂/NiS₂ and FeS₂ nanocrystals. The 2p_3/2 shows an obvious shift of approximately 0.49 eV from 706.89 eV for FeS₂ to 707.38 eV for FeS₂/NiS₂, indicating the enhanced oxidizability of the interface FeS₂/NiS₂ hybrid nanocrystals [15]. The same results were found in the Ni 2p spectra (Fig. 2(b)). The Ni also showed a higher oxidation state in the interface FeS₂/NiS₂ hybrid nanocrystals for the 2p_3/2 shift of approximately 0.34 eV from 853.43 eV for NiS₂ to 853.77 eV for FeS₂/NiS₂. The S 2p spectra of those catalysts also showed the same changes. As shown in Fig. 2(c), the S 2p spectra of FeS₂ and NiS₂ contained the same peak position at 162.40 eV; however, after the fabrication of the nanointerface, the S 2p spectra show obvious about 0.48 eV from the single phase to the interface FeS₂/NiS₂ hybrid nanocrystals (162.87 eV). Based on the above analysis and experiment, the fabrication of the nanointerface enhanced the coupling between the different domains, which could further improve the electronic transfer ability.

The OER activity of the catalysts was evaluated using a linear scan voltammogram (LSV) in 1.0 mol/L KOH at the scan rate of 2 mV s^-1 [15, 16]. The silk-like interface FeS₂/NiS₂ hybrid nanocrystals exhibited greater OER activity than that with the FeS₂ and NiS₂ nanocrystals. In Fig. 3(a), the OER current increased under the silk-like interface FeS₂/NiS₂ hybrid nanocrystals catalyzed and more quickly than other prepared catalysts. The onset potential of the silk-like interface FeS₂/NiS₂ hybrid nanocrystals for the OER was 92 mV, which was better than those of FeS₂ (221 mV) and NiS₂ (227 mV), and the overpotential at the OER current density of 10 mA cm^-2 for the interface FeS₂/NiS₂ hybrid nanocrystals was 233 mV lower than those of FeS₂ (285 mV) and NiS₂ (346 mV). The silk-like interface FeS₂/NiS₂ hybrid nanocrystals could maintain stability for more than 30 h (Fig. S3). The Tafel slopes were further confirmed to determine the reaction kinetics for the OER. As shown in Fig. 3(b), the silk-like interface FeS₂/NiS₂ hybrid nanocrystals displayed the lowest Tafel slope of 64 mV dec^-1 for the OER. This was smaller than those of FeS₂ (137 mV dec^-1) and NiS₂ (196 mV dec^-1). In addition, the silk-like interface FeS₂/NiS₂ hybrid nanocrystals showed excellent OER activity, more than those of the FeNi based electrocatalysts (Fig. 3(c)), such as NiFe LDH, Fe₇S₈ bulk, NiFe alloy/NiFe₂O₄, etc. [28-34]. Fig. 3(d) shows several important parameters for the OER, suggesting a superior activity of the silk-like interface FeS₂/NiS₂ hybrid nanocrystals compared to those of the other obtained catalysts.

The ORR activity of the silk-like interface FeS₂/NiS₂ hybrid nanocrystals, FeS_2, and NiS₂ were further explored in 0.1 mol/L KOH [15]. As shown in Fig. 4(a), the silk-like interface FeS₂/NiS₂ hybrid nanocrystals showed a more repaid catalytic speed and lower potential. The onset potential (E_onset: 0.91 V) and half-wave potential (E_1/2: 0.64 V) of the silk-like interface FeS₂/NiS₂ hybrid nanocrystals were similar to those of the benchmarked Pt/C (onset potential: 0.95 V, halfwave potential: 0.78 V) [15] and superior to those of FeS₂ (E_onset: 0.804 V and E_1/2: 0.553 V) and NiS₂ (E_onset: 0.853 V and E_1/2: 0.632 V). In addition, it had good stability (Fig. S4). Then, the electron transfer number (n) was investigated. In Fig. 4(b), the Koutechy-Levich (K-L) plots were calculated to confirm the electron transfer number of the obtained catalysts for the ORR, according to the LSV curves at different rotating rates (Fig. S5). The electron transfer number of the silk-like interface FeS₂/NiS₂ hybrid nanocrystals was approximately 4.15, and the kinetic current density (j_k) of 19.6 mA cm^-2 (inset of Fig. 4(b)) was better than those of FeS₂ (n: 3.9 and j_k: 16.7 mA cm^-2) and NiS₂ (n: 3.66 and j_k: 15.6 mA cm^-2), based on the K-L equation (experimental part). These results indicated the superior ORR catalytic activity through a four-electron pathway of the silk-like interface FeS₂/NiS₂ hybrid nanocrystals. Further, the Tafel slopes were calculated to confirm the reaction kinetics. As shown in Fig. 4(c), the silk-like interface FeS₂/NiS₂ hybrid nanocrystals displayed a lower Tafel slope of 73 mV dec^-1 for the OER, which was smaller than those of FeS₂ (121 mV dec^-1) and NiS₂ (152 mV dec^-1) and was approaching that of Pt/C [35, 36]. Fig. 4(d) shows several important parameters for the ORR. The silk-like interface FeS₂/NiS₂ hybrid nanocrystals exhibited excellent ORR performance in a basic media.

Based on the above analysis and experimental results, the silk-like interface FeS₂/NiS₂ hybrid nanocrystals exhibited excellent bifunction for the OER and ORR (Fig. 5). Furthermore, the reversible oxygen electrode activity of the silk-like interface FeS₂/NiS₂ hybrid nanocrystals was confirmed according to the different OER and ORR metrics, which could be calculated by the equation below [15].

(3)

The silk-like interface FeS₂/NiS₂ hybrid nanocrystals exhibited a ΔE value of 0.823 V (insert in Fig. 5), which was less than those of Pt/C (1.13 V) [15], FeS₂ (0.944 V), and NiS₂ (0.962 V), indicating better reversible oxygen activity of the silk-like interface FeS₂/NiS₂ hybrid nanocrystals and a potential application in metal-air batteries.

Considering the excellent bifunctional catalytic and reversible oxygen performance of the silk-like interface FeS₂/NiS₂ hybrid nanocrystals, a homemade Zn-air battery was assembled to further identify its performance under practical battery operational conditions. This battery used a zinc plate as the anode, silk-like interface FeS₂/NiS₂ hybrid nanocrystals as the air-cathode, and 6.0 mol/L KOH as the electrolyte (insert in Fig. 6(a)) [15, 16]. The fabricated battery operated stably with a high OCV of 1.25 V for more than 17 h (Fig. 6(a)). The maximum power density of the silk-like interface FeS₂/NiS₂ hybrid nanocrystals was 26 mW cm^-2 at 70.8 mA cm^-2 (Fig. 6(b)), which was greater than those of FeS₂ (16.7 mW cm^-2 at 35.89 mA cm^-2) and NiS₂ (6.8 mW cm^-2 at 41 mA cm^-2). As shown in Fig. 6(c), a lower charge-discharge voltage gap was observed for the Zn-air battery with the silk-like interface FeS₂/NiS₂ hybrid nanocrystals as the air-cathode than those of FeS₂ and NiS₂, indicating a better recharge ability and a good capacity (Fig. S6). Further, the battery cycling test was performed to confirm the recharge ability of the silk-like interface FeS₂/NiS₂ hybrid nanocrystals. The ideal reversible oxygen catalysts should possess a low charging voltage (E_charging), high discharging voltage (E_discharging), and minimal fluctuation of E_charging and E_discharging. As expected, the battery fabricated with the silk-like interface FeS₂/NiS₂ hybrid nanocrystals showed a stable charging voltage at different current densities for approximately 25 h (Fig. 6(d)), indicating an excellent recharge ability of the silk-like interface FeS₂/NiS₂ hybrid nanocrystals in the Zn-air battery.

Furthermore, solid Zn-air batteries based on the silk-like interface FeS₂/NiS₂ hybrid nanocrystals were fabricated. As shown in Fig. 7(a), the flexible solid Zn-air batteries were built using a zinc anode, hydrogel polymer electrolyte, catalyst cathode, and nickel foam (NF) air-collector [15, 16]. The recharge ability of the flexible solid Zn-air batteries was confirmed. Fig. 7(b) shows the OCV of the single battery (1.24 V and more than 15 h). The OCV for three-series solid batteries is 3.72 V and more than 7 h. As shown in Fig. 7(c), the discharge/charge cycles for the solid battery were maintained for 15 h at the current densities of 0.5 and 1 mA cm^-2 (10 min per cycle), indicating excellent performance of the silk-like interface FeS₂/NiS₂ hybrid nanocrystals fabricated flexible solid Zn-air batteries. In addition, as shown in Fig. 7(d), a green LED (approximately 3.0 V) was powered by three-series solid Zn-air batteries using the silk-like interface FeS₂/NiS₂ hybrid nanocrystals as the air-cathode.

In summary, silk-like interface FeS₂/NiS₂ hybrid nanocrystals were prepared by an efficient liquid exfoliation strategy using formamide as the stripping reagent and a continuous ultrasonic treatment. The obtained hybrid nanocrystals showed an interface structure for two domains of FeS₂ and NiS₂ in the HRTEM image and a unique silk-like morphology. The silk-like interface FeS₂/NiS₂ hybrid nanocrystals showed enhanced electrocatalytic performance with a low overpotential of 233 mV at j = 10 mA cm^-2 for the OER and an onset potential of 911 mV and half-wave potential of 640 mV for the ORR. Further, the silk-like FeS₂/NiS₂ hybrid nanocrystals also exhibited excellent reversible oxygen activity for the small ΔE of 0.823 V. In addition, based on the excellent reversible oxygen performance, the homemade rechargeable Zn-air batteries using FeS₂/NiS₂ hybrid nanocrystals as the air-cathode displayed a high OCV and excellent rechargeable performance. The method designed in this study could be used for designing other interface nanocrystals with a unique morphology for potential applications in renewable energy systems.

Nomenclature

j_k  kinetic current, mA cm^-2

ω  electrode rotating rate, rpm

B   slope of the K-L plots

n   transferred electron number per oxygen molecules

F   Faraday's constant, 96485.3 C mol^-1

D_O2 diffusion coefficient of O₂ in 0.1 mol/L KOH, D_O2 = 1.9 × 10^-5 cm² s^-1

C_O2  bulk concentration of O₂, C_O2 = 1.2 × 10^-6 mol cm^-3

v  kinetic viscosity, v = 0.01 cm² s^-1

E_j=10  OER potential at a current density of 10 mA cm^-2, V

E_1/2  ORR potential taken at the half wave, V