3.1. Material characterization
The sea urchin-like S-Ni(OH)2 catalyst was fabricated through a two-step process, as shown in Scheme 1. First, sea urchin-like Ni(OH)2 was uniformly grown on the surface of NF via a simple hydrothermal route. Then, the as-synthesized Ni(OH)2 was soaked in Na2S solution with different times (5 s, 5 min, 30 min, 1.5 h, 12 h) to form S-Ni(OH)2 with sea urchin-like structure.
The morphology and structure of the catalysts were elucidated by Scanning electron microscopy (SEM) and (TEM), as shown in Fig. 1. SEM images of Ni(OH)2 (Fig. 1) exhibit that a mass of nanowires assembled and generated sea urchin-like structure, covering the surface of the NF. After soaking in a Na2S solution, the obtained S-Ni(OH)2-30 min still kept sea urchin-like morphology. Energy dispersive spectroscopy (EDS) demonstrates the presence of Ni, C, O and S elements, which verified the atomic ratio of Ni and S is estimated to 36.59:7.38 (Fig. S1, Supporting Information). Transmission electron microscopy (TEM) was employed to characterize the detailed structure of S-Ni(OH)2-30 min scrapped off from Ni foam by ultrasonication. From TEM images (Fig. 1g), we find that S-Ni(OH)2-30 min still retains the sea urchin-like morphology, suggesting the structural stability of S-Ni(OH)2-30 min. HRTEM images (Fig. 1h) exhibits the lattice fringes of 0.219 nm, which match well with the (103) plan of α-Ni(OH)2, respectively. The EDS mapping exhibits homogeneous distribution of S, Ni, and O elements in the S-Ni(OH)2-30 min (Fig. 1i).
The crystal structures of the as-fabricated S-Ni(OH)2-30 min and Ni(OH)2 were studied by X-ray diffraction (XRD) (Fig. 2a). The diffraction peaks located around the 44.5o, 51.9o and 76.4o correspond to the (111), (200), and (220) crystal faces of metallic Ni (JCPDS No. 04-0850). Apart from the diffraction peaks of NF, the other peaks of Ni(OH)2 and S-Ni(OH)2-30 min are attributed to α-Ni(OH)2 (JCPDS No. 22–0444) and β-Ni(OH)2 (JCPDS No. 14–0117). However, relative to pristine α-Ni(OH)2, the (003) and (006) plans of Ni(OH)2 and S-Ni(OH)2-30 min both shift to lower angle, due to the insertion of anions into the lattice of Ni(OH)2 [10].
XPS was performed to study the surface composition and chemical state of Ni(OH)2 and S-Ni(OH)2-30 min. The XPS survey spectrum verifies the presence of Ni, C and O elements in Ni(OH)2, Ni, C, O and S elements in S-Ni(OH)2-30 min (Fig. S2, Supporting Information). As for the Ni 2p spectra of S-Ni(OH)2-30 min (Fig. 2b), the two main peaks located at the binding energies of 856.7 and 874.4 eV can be assigned to Ni 2p3/2 and Ni 2p1/2, respectively, and those at 862.5 and 880.5 eV corresponding to their satellite peaks [11]. The binding energies of Ni 2p3/2 and Ni 2p1/2 in Ni(OH)2 are negatively shifted to 856.5 and 874.2 eV, relative to those in S-Ni(OH)2-30 min [6]. The O 1s spectra for S-Ni(OH)2-30min (Fig. 2c) shows three peaks located at 530, 531.5 and 533 eV, which can be ascribed to Ni-O bonds [12], Ni-OH [12] and adsorbed molecular water [13], respectively. The S 2p spectrum of S-Ni(OH)2-30 min (Fig. 2d) presents two main characteristic peaks located at 162.6 and 164.4 eV, which can be indexed to S 2p3/2 and S 2p1/2 of Ni-S bonds [14], whereas the peak at 168.6 eV are ascribed to the oxidized sulfur species from surface oxidation [15]. Compared wth Ni(OH)2-S, the main peaks of O 1s in Ni(OH)2 are positively shifted about 0.2 eV. All the above XPS results demonstrate that the electron transfer from Ni to O and S, and strong electronic interaction between Ni and S in S-Ni(OH)2-30 min, which could effectively modify the OER performance of S-Ni(OH)2-30 min.
3.2. Electrocatalytic OER
To evaluate the OER electrocatalytic activities of S-Ni(OH)2-5 s, S-Ni(OH)2-5 min, S-Ni(OH)2-30 min, S-Ni(OH)2-1.5 h, S-Ni(OH)2-12 h and NF, we performed linear-sweep voltammetry (LSV) measurements in 1 M KOH electrolyte. As shown in Fig. 3a, the S-Ni(OH)2-30 min exhibits the best OER performance, requiring a lower overpotential of 306 mV to deliver density of 10 mA cm− 2, while S-Ni(OH)2-5s, S-Ni(OH)2-5 min, S-Ni(OH)2-1.5 h, S-Ni(OH)2-12h, Ni(OH)2, RuO2 and NF require larger overpotentials of 304, 306, 295, 312, 318, 329 and 521 mV to reach the same current density, respectively. Moreover, in comparison of S-Ni(OH)2-5 s (415 mV), S-Ni(OH)2-5 min (418 mV), S-Ni(OH)2-1.5 h (399 mV), S-Ni(OH)2-12 h (424 mV), Ni(OH)2 (455 mV) and RuO2 (452 mV), S-Ni(OH)2-30 min reach the current density of 100 mA cm− 2 at the lower overpotential of 392 mV. To gain insight into the OER kinetics, Tafel slopes were analyzed (Fig. 3b). S-Ni(OH)2-30 min shows the Tafel slope of 89.2 mV dec− 1, which is smaller than those of S-Ni(OH)2-5 s (93 mV dec− 1), S-Ni(OH)2-5 min (94.5 mV dec− 1), S-Ni(OH)2-1.5 h (90.8 mV dec− 1), S-Ni(OH)2-12 h (96.1 mV dec− 1), and NF (138.2 mV dec− 1). The Tafel slope of Ni(OH)2 is slightly smaller than that of S-Ni(OH)2-30 min. The electrochemical impedance spectroscopy (EIS) was performed to explore the electrode kinetics during the OER process. All EIS data (Fig. 3c) were fitted well with the equivalent circuit model, which was comprised of the resistance of the electrolyte solution (Rs), constant phase element (CPE) and charge-transfer impedance (Rct). Based on Nyquist plots, S-Ni(OH)2-30 min has the smallest semicircle diameter than S-Ni(OH)2-5 s, S-Ni(OH)2-5 min, S-Ni(OH)2-1.5 h, S-Ni(OH)2-12 h, Ni(OH)2 and NF, indicating the fastest charge transfer kinetics for OER.
To further probe the origin of excellent performance of S-Ni(OH)2-30 min, we performed Cdl measurements to estimate ECSA via CV method (Fig. S3). As shown in Fig. 3d, the Cdl value of S-Ni(OH)2-30 min is 1.35 mF cm− 2, which is close to that of S-Ni(OH)2-5 s (1.33 mF cm− 2), and larger than those of Ni(OH)2 (0.99 mF cm− 2), S-Ni(OH)2-1.5 h (1.11 mF cm− 2), S-Ni(OH)2-12 h (1.23 mF cm− 2), but smaller than that of S-Ni(OH)2-5 min (1.57 mF cm− 2), implying that ECSA is not the only factor for the outstanding OER activity of S-Ni(OH)2–30 min. Moreover, the long-term stability is another critical factor for broader applications. The stability test of S-Ni(OH)2-30 min was conducted by chronopotentiometric measurements at a current density of 20 mA cm− 2. S-Ni(OH)2-30 min could maintain its catalytic performance for at least 12 h with slight attenuation, as shown in Fig. 3e. Subsequently, the polarization curves (Fig. 3f) show negligible change after stability test, indicating that S-Ni(OH)2-30 min has high stability.
To explore the morphology and structure changes of S-Ni(OH)2-30 min after the CP test, we performed a series of structural characterizations. As shown in Fig. S4, S-Ni(OH)2-30 min maintains sea urchin-like morphology after the long-term stability test, indicating its structural stability. Meanwhile, as shown in Fig. S5, the lattice fringe of 0.209 nm and 0.219 nm are assigned to the (105) and (104) crystal plane of NiOOH (PDF No. 06–0075), respectively [16]. After OER test, the XRD pattern shows the existence of α-Ni(OH)2 (JCPDS No. 22–0444) and β-Ni(OH)2 (Fig. S6a). Moreover, XPS characterization was applied to investigate the chemical states information of S-Ni(OH)2-30 min after OER test. As for Ni 2p XPS spectra (Fig. 6b), the peaks located at 855.8 and 873.5 eV as well as two satellite peaks at 861.9 and 880.1 eV correspond to Ni2+ 2p3/2 and Ni2+ 2p1/2, respectively [17]. In addition, the spin-energy separation was calculated to be 17.7 eV, indicating the presence of Ni(OH)2 species. The other two peaks (857.5 and 875.4 eV) are assigned to Ni3+ 2p3/2 and Ni3+ 2p1/2, which proves the formation of NiOOH species [18]. In addition, the O 1s spectra (Fig. S6c) shows the coexistence of three peaks corresponding to adsorbed water (532.6 eV), hydroxyl groups bonded with metal cations (531.3 eV), lattice oxygen (529.9 eV), implying the generation of Ni–Fe(oxy) hydroxide during water oxidation process [19, 20]. As for S 2p XPS spectra (Fig. S6d), the two major peaks ascribed to S 2p3/2 and S 2p1/2 of Ni-S bonds at 162.6 and 164.4 eV disappeared after the OER test, which may be due to the surface oxidation of S element. Raman characterization was carried out to further investigate the structure evolution after the OER stability test (Fig. S7). The broad bands at 474 and 554 cm− 1 matched well with the spectral features of NiOOH [21]. Based on all the above results, we speculate that the formed nickel oxide/hydroxide converted from S-Ni(OH)2-30 min at the surface during OER. The generated nickel oxide/hydroxide are known as the real active species, which can optimize adsorption/desorption ability of intermediates, thereby enhancing intrinsic catalytic activity.
According to the above structural analysis and electrochemical results, the improved catalytic activity of S-Ni(OH)2-30 min could be attributed to the following points: (1) The optimal catalyst presents sea urchin-like structure can provide enough active sites, and facilitate the penetration of reactive species. (2) The self-supported S-Ni(OH)2-30 min directly grown on conductive Ni foam eliminates using polymer binder, which not only improves the stability of catalysts due to the strong mechanical connection between the catalysts and conductive substrates, but also provides more exposed active sites and facilitates the efficient electron transport. (3) S doping can effectively adjust the electronic structure of catalysts and thus boosting intrinsic activity, which is consistent with the previous reports [7, 22].