Abstract

The versatile cobalt-phosphorus (Co-P) precursor was synthesized on Ni foam (NF) with an electrodeposition method. Simple room-temperature treatment of the precursor with 1 M NaOH and 0.5 M H2SO4 allows for the production of cobalt oxide nanorods (CoxOy/NF) and cobalt phosphate/phosphide (Co-Pi/CoP/NF), respectively. The resulting CoxOy/NF shows a low overpotential () of 80 mV at −10 mA/cm2 in an alkaline electrolyte (pH = 14) for a hydrogen evolution reaction (HER) toward electrocatalytic water splitting. The Co-Pi/CoP/NF exhibits a low of 112 mV in an acidic electrolyte (pH = 0), in which the synergy between Co (+2) and Co (+3) may play an important role in the reaction.

1. Introduction

Hydrogen is widely considered as a clean and renewable fuel which is a promising replacement for fossil fuels [13]. Electrochemical and photoelectrochemical processes for water splitting are favourable strategies benefiting from abundant water resources and giving high-purity H2 production [47]. Overall, water splitting involves two half reactions: hydrogen evolution reaction (HER) and oxygen evolution reaction (OER). The mechanism of the HER process in an alkaline electrolyte can be considered as a combination of the following three basic steps: (1) the electroreduction of water molecules and absorption of hydrogen (Volmer reaction): , (2) electrochemical hydrogen absorption (Heyrovsky reaction): , and (3) chemical desorption (Tafel reaction): . Similarly, the process in an acidic electrolyte can be regarded as the following three basic steps: (1) the electroreduction of water molecules and absorption of hydrogen (Volmer reaction): , (2) electrochemical hydrogen absorption (Heyrovsky reaction): , and (3) chemical desorption (Tafel reaction): , where M represents a metal atom and stands for a H atom absorbed at an active site of the catalyst [813].

To achieve sustainable production of H2 fuel through water splitting, much effort has been made to design highly efficient electrocatalysts that are vital for lowering the dynamic overpotential during the HER. Currently, Pt is the most efficient catalyst for HER [14, 15]. However, the low natural abundance and high cost of Pt greatly limit their application in large-scale commercial production. Therefore, the synthesis of noble metal-free electrocatalysts with high efficiency for HER has drawn much attention. Over the past few years, researchers report various kinds of electrocatalysts such as metal chalcogenides [16, 17], metal carbides [14, 18], metal nitrides [19, 20], metal phosphides [21, 22], metal oxide hydroxides [23, 24], and metal alloys [25] with different morphologies and microstructures for HER. Transition metal phosphides and phosphate have drawn intense attention recently because of their remarkable catalytic activity for HER [2628]. However, there are few noble metal-free electrocatalysts with excellent activity and stability in both acidic and alkaline media [29].

In this study, we report the room-temperature fabrication of Co-based electrocatalysts with excellent catalytic activity and stability in both alkaline and acidic electrolytes through simply treating a Co-P precursor with either an alkaline or acidic solution. The related Co-P precursor was synthesized on Ni foam (NF) with an electrodeposition method. After treatment in an alkaline and acidic solution, CoxOy/NF and Co-Pi/CoP/NF electrodes were obtained, respectively. The resulting CoxOy/NF electrode shows a low of 80 mV for HER in an alkaline medium, and the Co-Pi/CoP/NF electrode exhibits a of 112 mV in an acidic medium for HER.

2. Experimental Section

2.1. Chemicals and Reagents

Sodium hypophosphite monohydrate (NaH2PO2⋅H2O), cobaltous chloride hexahydrate (CoCl2⋅6H2O), sodium chloride (NaCl), boric acid (H3BO3), ethanol (CH3CH2OH), sodium hydroxide (NaOH), hydrochloric acid (HCl, 36–38 wt%), and sulfuric acid (H2SO4, 98 wt%) were all purchased from Alfa Aesar with an analytical reagent (AR) and were used without further purification.

2.2. Sample Preparation

The Co-P precursor electrodeposited on a Ni foam was prepared by galvanostatic electrodeposition in a standard three-electrode system. A platinum (Pt) sheet, a silver chloride electrode (Ag/AgCl, saturated potassium chloride solution), and a piece of Ni foam were used as the counter, reference, and working electrode, respectively. Before electrodeposition, the Ni foam was cleaned with deionized water and ethanol in an ultrasonic cleaner for 10 minutes each to remove the contaminants on its surface. Then, it was ultrasonically cleaned with 1 M HCl for about 10 minutes to get rid of the oxide. After that, the Ni foam was washed with deionized water and dried for further use. The electrolyte in the electroplating bath consists of 0.3 M NaH2PO2⋅H2O, 0.2 M CoCl2⋅6H2O, 0.15 M H3BO3, and 0.1 M NaCl. The electrodeposition was carried out at a potential of −1.5 V vs. a Ag/AgCl reference electrode for 10 minutes. To get CoxOy/NF (CoxOy on Ni foam) and Co-Pi/CoP/NF (Co-Pi/CoP thin film on Ni foam) electrodes, the as-prepared Co-P precursor was washed with deionized water and subsequently immersed in 1 M NaOH (pH = 14) or 0.5 M H2SO4 (pH = 0) for 12 hours at room temperature.

2.3. Characterization

The morphologies of the synthesized samples were recorded by scanning electron microscopy (SEM, S-3400N II, Hitachi Co., Japan) with an acceleration voltage of 20 kV. Transmission electron microscopy (TEM) images, high-resolution transmission electron microscopy (HRTEM) images, electron diffraction pattern, and element-mapping images were obtained on a JEOL JEM-2100 microscope with a LaB6 filament and an accelerating voltage of 200 kV. XPS was measured on a PHI 5000 VersaProbe X-ray photoelectron spectroscope (ULVAC-PHI Inc., Japan).

2.4. Electrochemical Measurements

All catalytic performances were conducted on a CHI660E electrochemical analyzer/workstation (Chenhua Instruments, Shanghai) using a typical three-electrode system at room temperature. The as-prepared electrodes (CoxOy/NF, Co-Pi/CoP/NF), Ni foam, and Pt sheet were used as the working electrodes. The Pt sheet and Ag/AgCl electrode were used as the counter and reference electrode, respectively. Linear sweep voltammetry (LSV) and Tafel plot were conducted, and LSV was carried out at a scan rate of 2 mV/s. For HER in alkaline and acidic media, 1 M NaOH (pH = 14) and 0.5 M H2SO4 (pH = 0) were used as the electrolyte. During these tests, the working surface area of the as-prepared samples was adjusted by changing the length of the Ni foam immersed in the electrolyte and the area was estimated to be 1 cm2. All the overpotentials measured in this work were calibrated with the reversible hydrogen electrode (RHE) according to

3. Results and Discussion

The Co-P precursor was synthesized with a typical three-electrode system by electrodeposition. NaH2PO2H2O and CoCl26H2O serve as the source of the P and Co atoms. The precursor was then immersed in 1 M NaOH or 0.5 M H2SO4 to obtain the CoxOy/NF and Co-Pi/CoP/NF electrodes, respectively. Figure 1 shows the morphology of the Co-P precursor and those of Co-Pi/CoP/NF and CoxOy/NF. The surface of the Ni foam was covered by a solid sphere-like Co-P precursor after the completion of electrodeposition. Co-Pi/CoP/NF exhibits a thin-film morphology and CoxOy/NF consists of lots of nanorods. Figures 2(a)2(c) present the TEM images of the as-synthesized products. The Co-P precursor shows an irregular microstructure. The Co-Pi/CoP features large amounts of interlacing thin wires, which can explain the film structure presented in the SEM image in Figure 1. Since the catalysts seem to be amorphous, the characteristic peaks of our catalysts can hardly be found in the XRD pattern, especially for Co-Pi. A similar situation has been reported earlier [30, 31]. For CoxOy, the existence of oxides in different valence states makes its morphology not single. Figure 2(c) presents the structure of a hexagonal nanoplate, and Figure 3 shows the nanorod structure which is consistent with the nanorods observed in the SEM images. We can also find the nanoplate structure although the nanoplate structure seems to be broken. Figures 2(d)2(f) show the element mapping, from which the existence of Co, O, and P can be easily detected in all products.

XPS spectra further confirm the atom ratio, element composition, and chemical bonding of as-synthesized samples. Table 1 shows that the atom concentration makes a difference to the original precursor after treatment with an acidic or alkaline solution. It reveals that the atom ratio of Co : P is about 1 : 1, 2 : 3, and 10 : 1 in the Co-P precursor, Co-Pi/CoP, and CoxOy respectively. It is noteworthy that P can hardly be found after being treated with an alkaline solution. It means that the P moiety was rinsed from the Ni foam. Figure 4(a) shows the total XPS survey spectra for the Co-P precursor and the CoxOy and Co-Pi/CoP electrocatalysts. It is consistent with element mapping that three elements (Co, O, and P) exist, and no obvious peaks for other elements are observed. Furthermore, the high-resolution XPS spectra for Co2p3/2, O1s, and P2p are presented in Figures 4(b)4(d). The binding energy of Co2p appears in two peaks around 780 eV and 796 eV which are ascribed to Co2p3/2 and Co2p1/2 with their satellites beside them, respectively. The peaks for Co2p3/2 show that Co exists not only in the +2 but also in the +3 state. The representative peaks for Co2+ locate at 781.2 eV and 781.9 eV, and the peaks for Co3+ locate at 778.4 eV [32, 33]. The presence of Co3+ indicates the potential existence of CoP. The peak between 129.3 eV and 129.6 eV corresponding to P3− exactly confirms the presence of CoP [26, 33, 34]. Furthermore, the binding energy of 131.9 eV and 133.2 eV corresponds to phosphate, which suggests the existence of Co-Pi [29]. For the O1s spectra, the peaks appearing at 530.9 eV and 531.8 eV can be assigned to the P-O bonding, which also confirms the existence of Co-Pi [26, 29]. The binding energy between 529.0 eV and 529.5 eV is related to the Co-O bonding [33, 35], which may be caused by oxidation in air. The peak located at 530.8 eV and 531.5 eV can be attributed to Co(OH)2 [36, 37]. In addition, when the precursor is treated with 0.5 M H2SO4, obvious bubbles were found on the surface. It is reasonable to believe that Co2+ is reduced into Co metal during the electrodeposition process.

CoxOy and Co-Pi/CoP electrocatalysts with a unique composition were used as free-standing electrodes for electrocatalytic HER in 1 M NaOH and 0.5 M H2SO4. The linear sweep voltammetry (LSV) curves, corresponding Tafel slopes, and chronoamperometry - curves were measured on CoxOy/NF and Co-Pi/CoP/NF to estimate their HER activity and long-term stability. Control experiments were also performed on a bare Ni foam and Pt sheet for comparison. Production of alkaline-efficient electrocatalysts is of prime importance because alkaline water electrolysis is most widely used in the industry. The CoxOy/NF electrode is proven to have good HER performance in an alkaline solution. The LSV curves of CoxOy/NF in 1 M NaOH exhibit an overpotential of 80 mV at −10 mA/cm2 () (Figure 5(a)), much smaller than those of the bare Ni foam (232 mV) and Pt sheet (114 mV). It indicates excellent HER activity for CoxOy/NF benefitting from its relatively larger surface area and more active sites exposed. Figure 5(b) shows the Tafel plots of the bare Ni foam, CoxOy/NF, and Pt sheet in 1 M NaOH. Tafel plots fit well with the Tafel equation: , where is the current density and is the Tafel slope. The Tafel slopes of the three electrodes were calculated as 199.8, 110.1, and 104.2 mV/dec, respectively. Obviously, the bare Ni foam shows a higher slope than the CoxOy/NF electrode and Pt sheet, which is consistent with the HER activity evaluated with LSV curves. The CoxOy/NF electrode also shows superior activity to similar electrocatalysts reported (Table 2).

The good stability of the time dependence with the current density is also essential for the long-term application of electrocatalysts. Figure 5(c) demonstrates that the current density of the CoxOy/NF electrode in 1 M NaOH at an overpotential of 80 mV remains quite stable in a time range of 40,000 s, which can prove the excellent stability of the electrode. Figure 5(d) shows linear sweep voltammetry curves of CoxOy/NF before and after potential sweeps between −0.2 and +0.2 V vs. RHE for 3000 cycles. With almost no shifts between each curve, it further proves the long-term stability.

In contrast to the CoxOy/NF electrode, Figures 6(a)6(d) exhibit the catalytic activity of the bare Ni foam, Co-Pi/CoP/NF electrode, and Pt sheet showing that the Co-Pi/CoP/NF electrode has a lower of 112 mV than the bare Ni foam (over 300 mV), and just a little higher than that of the Pt sheet (84 mV) (Figure 6(a)). The Tafel slope of the Co-Pi/CoP/NF electrode (ca. 99.8 mV/dec) is lower than the bare Ni foam (ca. 282.3 mV/dec) and a little higher than that of the Pt sheet (ca. 37.3 mV/dec) (Figure 6(b)). It also ranks among the best of similar electrocatalysts reported (Table 2). At the same time, the time dependence of the current density test and polarization curves obtained over Co-Pi/CoP/NF before and after potential sweeps between −0.2 and +0.2 V vs. RHE (Figures 6(c)6(d)) also proves the good stability of the Co-Pi/CoP/NF electrode, the same as the CoxOy/NF electrode.

Considering the excellent catalytic performance of Co-Pi/CoP and CoxOy for HER, the synergetic effect of Co (+2) and Co (+3) may play an important role during the process. The proposed mechanism is illustrated in Scheme 1. As reported by a previous study [37], the adsorption and desorption of the H atom are the key procedures for HER. Gibbs’ free energy (ΔGH) of absorbed H atoms on catalysts can efficiently characterize how difficult the adsorption and desorption of the H atom are. Pt possesses a nearly zero ΔGH, and higher or lower ΔGH can increase the difficulty for the desorption or adsorption of the H atom [38]. CoP has a relatively strong bond with H, which means it can easily adsorb H but may suffer from the desorption of H [39, 40]. In contrast, Co-Pi has a weaker bond with H. The compounds of Co-Pi and CoP can enhance the adsorption and desorption at the same time. Through the synergy between Co-Pi and CoP, an outstanding catalytic activity is achieved. The synergetic effect of CoxOy along with Co (+2) and Co (+3) remains the same.

(a)
(b)

4. Conclusions

A Co-P precursor is successfully synthesized on Ni foam with a simple electrodeposition method. After being selectively treated in an alkaline or acidic solution, the obtained electrodes can be used for HER in an alkaline or acidic electrolyte. Both CoxOy/NF and Co-Pi/CoP/NF electrodes show excellent HER activity, comparable to the Pt sheet. The changed composition of the electrodes with the treating precursor in an acidic or alkaline solution is the main factor that leads to different catalytic characteristics. The synergy between Co (+2) and Co (+3) may also play an important role in the reaction. This work may give a new thinking to exploit new electrocatalysts for HER.

Data Availability

The data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Acknowledgments

This work was supported by the 973 Programs (no. 2014CB239302), the National Natural Science Foundation of China (nos. 21473091 and 21773114), and the Applied Basic Research Programs of the Science & Technology Department of Sichuan Province (2017JY0146).