Trace Amount of NiP2 Cooperative CoMoP Nanosheets Inducing Efficient Hydrogen Evolution.

As a very attractive clean energy, hydrogen has a high energy density and great potential to achieve zero pollution emission. Therefore, the preparation of hydrogen evolution electrocatalysts with excellent performance is an urgent task to ameliorate the global energy shortage and environmental pollution. Here, a trace amount of NiP2 coupled with CoMoP nanosheets (NCMP) was synthesized by the one-step hydrothermal method and low-temperature phosphidation. Studies have found that although the dosage of NiP2 is very low, its appearance has been efficient to improve the hydrogen evolution reaction (HER) performance of CoMoP, which may be induced by the synergistic effect of the two different components NiP2 and CoMoP. To find the superior catalyst, the effect of Ni content on the catalyst performance is also studied, and it is found that when the dosage of Ni is 0.02 mM, NCMP-2 (2 means 0.02 mM) displays the most outstanding overpotential (10 mA cm-2) of 46 mV.

Introduction

As society develops, the existing fossil energy stored on the earth has been consumed in large quantities and environmental pollution has deteriorated rapidly, prompting researchers to develop inexpensive, high-efficiency, and clean fungible power sources.[1−4] Hydrogen as a kind of new energy comes from various sources with high calorific value and pollution-free combustion, which can be used as an effective alternative to traditional energy.[5,6] Currently, the electrochemical hydrogen evolution reaction (HER) by water splitting, a practical and simple way to produce hydrogen, is currently considered as an effective and important method worth studying.[7,8] However, one of the important problems is the slow kinetic speed of the HER, which needs to be carried out using efficient catalysts. The noble metal Pt is known as the furthest ideal catalytic material utilized in the HER at present because of its excellent overpotential, fast kinetic process, and the best binding force with hydrogen.[9,10] However, Pt is a precious metal that is expensive and scarce and hence cannot be applied in large-scale applications.[11−14] Therefore, designing efficient and cheap catalysts is a goal of researchers. At present, various transition-metal materials, including transition-metal phosphides (TMPs),[15−20] transition-metal sulfides (TMSs),[21−25] transition-metal selenides (TMSes),[26−29] transition-metal carbides (TMCs),[30−32] transition-metal nitrides (TMNs),[33,34] transition-metal borides (TMBs),[35] etc. as outstanding catalyst candidates, have been extensively studied and reported.

Especially, TMPs with high activity and corrosion resistance have been explored to improve the HER catalytic efficiency.[36−39] In particular, CoMoP as one of bimetallic phosphides has various structural and chemical advantages in the application of the HER, and is extensively reported as a boundless prospect alternative of precious metals.[40−42] In the previous work, Huang et al. synthesized Co5Mo1.0P nanosheets with a three-dimensional structure, which due to its three-dimensional structure was conducive to electron and mass transfer, and its overpotential (10 mA cm–2) reached 173 mV under acidic conditions.[43] In another work, by ameliorating the ratio of Mo to P, Wu et al. synthesized the CoMoP catalyst with the prominent performance (Mo/P = 1:2, initial potential of 85 mV). This was because the mediation ratio could control the morphology of CoMoP (Mo/P = 1:2) to obtain larger specific surface areas and also could accelerate the electron transfer.[44] Although researchers have adopted a variety of methods to make CoMoP show good HER activity, the performance of CoMoP still has a large space to be improved compared with Pt.

A typical strategy to achieve high-performance HER activity is to combine another component with CoMoP. Chen et al. synthesized a CoP3/CoMoP heterostructure catalyst composed of two phases that effectively improved the conductivity and the number of active sites, making its overpotential only 125 mV (10 mA cm–2).[45] Besides, Hu et al. also reported the structure of a Co5.0Mo1P nanoflower and NiFe LDH coupling, whose overpotential was only 98.9 mV at 10 mA cm–2. This was because the synergistic effect between the two phases effectively slowed down the activity decline and exposed many active sites.[46] Moreover, Wang et al. prepared carbon-coated CoMoP nanosheets and the carbon not only promoted the dispersion of active particles but also provided more active sites for the HER through interaction with CoMoP, thus improving the conductivity and HER performance significantly (98.9 mV at 10 mA cm–2).[47] All of the above three catalysts have one characteristic in common: they have two components, and the synergistic effect between them can be exploited to improve the performance of the catalysts.[48,49] Based on the above references, it can also be found that adjusting the ratio of materials and morphology can be utilized to improve the catalytic performance. Therefore, we can combine the control of materials’ proportion and synergistic effect as a feasible strategy to enhance the HER performance of CoMoP.

In this work, we use a facile method introducing NiP2 to optimize the HER properties of CoMoP. The Ni(OH)2/CoMoO4 (NH/CMO) composite structure nanosheets were grown on the surface of nickel foam by a one-step hydrothermal method instead of the universal two-step hydrothermal method. Then, NiP2/CoMoP (NCMP) nanosheets were obtained by a simple phosphidation process. It is found that the trace amount of NiP2 has exhibited the efficient enhancement of CoMoP performance. To obtain the optimized HER performance, the effect of different contents of Ni on the catalyst performance is studied, from which it was found that NCMP-2 (2 means Ni 0.02 mM) shows the most outstanding HER performance with the overpotential of only 46 mV at 10 mA cm–2. Furthermore, NCMP-2 demonstrates excellent stability under alkaline conditions. The results show that the efficient improvement of the HER performance of transition-metal phosphides might just by a little another kind of materials.

Results and Discussion

The synthetic process of NCMP nanosheets is shown in Scheme . First, Ni(OH)2/CoMoO4 (NH/CMO) nanosheets were successfully synthesized by the one-step hydrothermal reaction using nickel foam as a supporting material. The X-ray diffraction (XRD) pattern of the NH/CMO-2 is displayed in Figure a, which can be indexed to Ni(OH)2 (PDF# 73-1520) and CoMoO4 (PDF# 73-1331). For comparison, XRD patterns of other precursors with different compositions are also displayed in Figure S1. The comparison shows that NH/CMO-x (x = 1, 3, and 4) is also composed of Ni(OH)2 and CoMoO4, and another contrast sample CoMoO4 precursor has been successfully synthesized. Then, NCMP-x catalysts were synthesized by the phosphidation process. The diffraction peaks of NCMP-2 (Figure b) are consistent with those of NiP2 (PDF# 73-0436) and CoMoP (PDF# 71-0478), except for the peaks of glass and Ni foam substrate. Combining the XRD patterns of the materials obtained after processing CoMoO4 and NH/CMO-x (x = 1, 3, and 4) with the same method, it is found that the phosphidation product of CoMoO4 is CoMoP (Figure S2a), while the phosphidation products of NH/CMO-x (x = 1, 3, and 4, Figure S2b–d) are NiP2 and CoMoP that are the same as NCMP-2. The preliminary specification is that catalysts combining NiP2 and CoMoP are successfully prepared.

Scheme 1

Schematic Illustration of the Formation Process of NCMP

Figure 1

XRD patterns of (a) NH/CMO-2 and (b) NCMP-2.

The morphology of the samples is observed by scanning electron microscopy (SEM). The results show that NH/CMO-2 presents a uniform nanosheet structure (Figure a,b). After phosphidation, the morphology of NCMP-2 is also nanosheets (Figure c,d). Similarly, SEM images of CoMoP, NCMP-x (x = 1, 3, and 4), and their respective precursors are shown in Figures S3–S6, respectively. It can be found that the morphologies of CoMoP and NCMP-x (x = 1, 3, and 4) are similar to that of NCMP-2 nanosheets, and their precursors are also nanosheets. In addition, the relevant energy-dispersive X-ray (EDX) mapping of NCMP-2 illustrates that the corresponding elements (Ni, Co, Mo, P) are evenly distributed in NCMP-2 (Figure e,f) and also in NCMP-x (x = 1, 3, and 4; Figures S4–S6), and Co, Mo, and P are uniformly distributed in CoMoP (Figure S3).

Figure 2

(a, b) SEM images of NH/CMO-2. (c, d) SEM images of NCMP-2. (e) SEM image of NCMP-2 and (f) the corresponding element mapping images of Ni, Co, Mo, P.

The transmission electron microscopy (TEM) images of NCMP-2 have also shown further characterized morphology (Figure a), which displays the same nanosheets as shown in Figure d. As a contrast, TEM images of CoMoP, NCMP-1, NCMP-3, and NCMP-4 (Figures S7–S10) show similar nanosheets structures with their SEM images (Figures S3–S6). The high-resolution TEM (HRTEM) results (Figure b) further confirm the existence of both NiP2 and CoMoP in NCMP-2. The d-spacings of 0.221 and 0.183 nm are in good agreement with the (211) and (221) crystal planes of NiP2, respectively, while those of 0.149 and 0.267 nm are keeping well with the (114) and (201) crystal planes of CoMoP, respectively. In addition, the HRTEM images of CoMoP (Figure S7) and NCMP-x (x = 1, 3, and 4; Figures S8–S10) correspond to CoMoP, NiP2 and CoMoP, which is consistent with the XRD results. NCMP-2 element mapping images in Figure c,d show that Ni, Co, Mo, and P elements are evenly distributed, which is consistent with SEM results (Figure e,f). Meanwhile, EDX analysis of CoMoP and NCMP-x (x = 1, 3, and 4) also confirms the uniform distribution of Co, Mo, and P (Figure S7c) and Ni, Co, Mo, and P (Figures S8–S10). EDX results demonstrate the successful phosphidation of the precursors. More importantly, EDX results show that the content of Ni is very low (atomic percentages ca. 5–9%, Figure S11). The results of EDX, XRD, and HRTEM show that nanosheet catalysts with uniform element distribution and different Ni contents are successfully prepared.

Figure 3

(a) TEM image and (b) high-resolution TEM image of NCMP-2. (c) TEM image of NCMP-2 and (d) the corresponding EDX element mapping images of Ni, Co, Mo, P.

To further elucidate the elemental composition of CoMoP and NCMP-2, X-ray photoelectron spectroscopy (XPS) tests are used for characterization. The XPS survey spectra show that Co, Mo, and P coexist in CoMoP (Figure S12a) and Ni peaks are detected in NCMP-x (Figure S12b). In Figure a, it can be observed that obvious peaks 779.62 and 782.08 eV are identified as Co 2p3/2, 794.59 and 798.45 eV as Co 2p1/2, and then two satellite peaks. Obviously, these peaks are all negatively shifted compared to the Co 2p spectrum of CoMoP, the peaks of Co 2P3/2 are located at 782.01 and 784.02 eV, while those of Co2p1/2 are located at 796.89 and 798.76 eV.[50] The Mo 3d spectrum peaks in Figure b show that the two binding energies located at 231.33 and 234.46 eV matched the Mo4+ species. Beyond that, the second pair of peaks at 233.21 and 236.34 eV can be matched with Mo6+. The Mo 3d spectrum of CoMoP also involves Mo4+ and Mo6+ regions. There are two obvious peaks in the Mo4+ region around 232.16 and 235.28 eV, and the peaks at 233.22 and 236.41 eV are due to the Mo6+ species.[51] These results confirm that the negative shifts also exist in the Mo 3d spectrum peaks of NCMP-2 relative to CoMoP. The Ni 2p spectrum of NCMP-2 can be divided into six peaks (Figure c), of which 854.31 and 871.58 eV are attributed to the Ni–P bond of Ni 2p3/2 and Ni 2p1/2. In addition, the major locations at 857.26 and 874.53 eV are Ni 2p3/2 and Ni 2P1/2 of nickel oxide, and the remaining locations at 862.59 and 879.89 eV are satellite peaks.[52] The P 2p spectra of both CoMoP and NCMP-2 show one double peak (P 2p1/2 and P 2p3/2) and oxidized P species, while the three major peaks display a similar negative shift with the spectra of Co 2p and Mo 3d (Figure d). The two peaks at 130.21 and 131.05 eV can be attributed to P 2p3/2 and P 2p1/2 of NCMP-2, while the peak at 134.42 eV is attributed to the oxidized P species. The locations of P 2p3/2, P 2p1/2 and oxidized P species of CoMoP are 131.01, 131.63, and 134.64 eV, respectively. Among them, the emergence of oxidation peaks may be related to exposure to air.[53] Similarly, in the spectra of NCMP-1, NCMP-3, and NCMP-4, it can be found that the peak positions of Co, Mo, and P also have negative shifts compared with CoMoP (Figures S14–S16). On the contrary, the Ni 2p peaks of the NCMP-x are positive shifts with the increase of the dosage of Ni as shown in Figure S13. The relative reverse shifts of binding energies of Co, Mo, P, and Ni prove an interaction between NiP2 and CoMoP, which resulted in the transfer of electrons from Co, Mo, and P to Ni.[54,55] And this is also conducive to improving the HER performance of NCMP-2.[51,56,57]

Figure 4

High-resolution XPS spectra of (a) Co 2p, (b) Mo 3d, (c) Ni 2p, and (d) P 2p for NCMP-2 and CoMoP.

To further understand the electronic regulation mechanism between NiP2 and CoMoP, the work functions of catalysts and other related information are obtained through ultraviolet photoemission spectroscopy (UPS). Work function is an important physical quantity reflecting the electron transfer ability, which can be calculated by the formula: W = hν (21.22 eV) – |Ef – Ecutoff|. As shown in Figure , the work functions of NCMP-1 (4.46 eV), NCMP-2 (4.44 eV), NCMP-3 (4.47 eV), and NCMP-4 (4.49 eV) are far less than that of CoMoP (4.53 eV, Figure S17), indicating that the addition of NiP2 can improve the electron transfer efficiency of CoMoP.[58] More importantly, NCMP-2 possesses the smallest work function, which means the best conductivity for NCMP-2.[59] Thus, it is supported to propel the materialization of the HER process. In addition, the ionization potential (E = Wwork function + Ev) can be obtained by UPS that reflects the power of the catalysts to supply electrons. NCMP-2 has the minimum ionization potential (8.45 eV), indicating that NCMP-2 can provide an electron-rich environment. This is also consistent with XPS results, where the introduction of NiP2 causes Co, Mo, and P to transfer electrons to Ni, which is also beneficial to the HER process.[60]

Figure 5

UPS spectra of (a) NCMP-1, (b) NCMP-2, (c) NCMP-3, and (d) NCMP-4.

The electrocatalytic performance of NCMP-x and CoMoP is first investigated through linear sweep voltammetry (LSV) in a basic environment. As shown in Figure a, with the introduction of a trace amount of NiP2, NCMP-x displays outstanding catalytic activity. To illustrate this point more clearly, in the partial enlargement of Figure a (inset of Figure a), it can be found that when the current density is 10 mA cm–2, the overpotential of NCMP-2 is only 46 mV in 1 M KOH compared to NCMP-1 (50 mV), NCMP-3 (55 mV), and NCMP-4 (61 mV). And the overpotential of NCMP-x (x = 1, 2, 3, and 4) are all better than CoMoP (76 mV), which may benefit from the synergistic effect between the cooperation of NiP2 and CoMoP. The Tafel slope of NCMP-2 is 106 mV dec–1 shown in Figure b that is smaller than NCMP-1 (110 mV dec–1), NCMP-3 (111 mV dec–1), NCMP-4 (118 mV dec–1), and CoMoP (119 mV dec–1). It can be noticed that NCMP-2 also has the lowest Tafel slope compared with other catalysts, indicating that NCMP-2 has a faster electrochemical process. Table S1 shows the overpotential of the HER and Tafel slope of other Co-, Mo-, and Ni-based phosphides in a basic environment. By comparison, it can be found that the HER performance of NCMP-2 in the basic environment is better than that of most previously reported TMPs electrocatalysts. Based on the above discussion, the combination of NiP2 and CoMoP provides excellent performance, and more importantly, only a trace amount of NiP2 can achieve this effect.

Figure 6

(a) HER polarization curves of NCMP-x (x = 1, 2, 3, and 4) and CoMoP in 1 M KOH; inset: partial enlargement of (a). (b) Corresponding Tafel plots. (c) Nyquist plots of NCMP-x (x = 1, 2, 3, and 4) and CoMoP. (d) Corresponding Cdl of CoMoP and NCMP-x (x = 1, 2, 3, and 4) at 0.11 V vs reversible hydrogen electrode (RHE). (e) Polarization curves of NCMP-2 before and after 10 h test. (f) Long-term stability of NCMP-2 at a constant overpotential for 10 h in 1 M KOH.

The EIS studies have investigated the kinetics process of electrode. The Nyquist plots in Figure c show that the charge transfer resistance of NCMP-2 (1.59 Ω) is smaller than those of NCMP-1 (2.18 Ω), NCMP-3 (2.21 Ω), NCMP-4 (3.09 Ω), and CoMoP (3.19 Ω), indicating that NCMP-2 has a favorable charge transport efficiency in the basic solution for the HER process, which is consistent with the result obtained by UPS that NCMP-2 has the best conductivity and results in the final excellent HER activity.[61,62] This can be due to the existence of two different components, NiP2 and CoMoP, which can induce the synergistic effect to be beneficial to HER performance. Another possible reason is that NCMP directly grown on Ni foam endows it with excellent electrical conductivity.[63,64] It has to be said that electrochemical active surface area (ECSA) is also an important factor affecting the catalytic performance. Cdl is employed to compare the ECSAs of NCMP-x (x = 1, 2, 3, and 4) and CoMoP because ECSA is directly related to Cdl. Based on the cyclic voltammograms of CoMoP and NCMP-x under different scan rates (Figures S18 and S19), Figure d shows that the Cdl values of NCMP-x (x = 1, 2, 3, and 4) and CoMoP are 73.38, 73.43, 59.44, 59.35, and 54.04 mF cm–2, respectively. Through Cdl comparison, we can find that NCMP-2 has a larger ECSA than others, suggesting that it can provide more active sites, which may be an important reason for the excellent performance of NCMP-2. Furthermore, the durability of NCMP-2 under basic conditions (1 M KOH) is studied through a chronoamperometry test at −0.075 V (vs RHE, corresponding overpotential from LSV), which is a necessary factor to judge the HER catalytic performance. Figure e shows that the initial and final LSV curves during the 10 h test are negligibly different and the current density is kept for 10 h at least 20 mV cm–2 in the test (Figure f), revealing the superior stability of the catalyst.

Conclusions

We use a simple method by introducing a trace amount of NiP2 to improve the HER performance of CoMoP. NH/CMO precursors were first synthesized by a one-step hydrothermal method, and then NCMP-x were synthesized by the phosphidation process. Benefiting from the synergistic effect of trace amounts of NiP2 and CoMoP, NCMP-x catalysts optimize the HER activity in alkaline solution. Especially, when the dosage of Ni is 0.02 mM, the catalyst shows the best HER efficiency under basic conditions, a low overpotential of 46 mV was obtained at 10 mA cm–2. Meanwhile, NCMP-2 shows good stability for 10 h. After all, our strategy that introduces a trace amount of another component in this work can contribute to further development of electrocatalysts based on TMPs.

Experimental Section

Synthesis of NCMP-x

First, the precursor Ni(OH)2/CoMoO4-x (x = 1, 2, 3, and 4, meaning 0.01, 0.02, 0.03, and 0.04 mM Ni(NO3)2·6H2O, respectively) was synthesized through the one-step hydrothermal process. 0.5 mM Na2MoO4, y (y = 0.01, 0.02, 0.03, and 0.04) mM Ni(NO3)2·6H2O, and 0.5-y mM Co(NO3)2·6H2O were added to 30 mL of deionized (DI) water and then stirred. After 5 h, the mixed solution was transferred to Teflon autoclaves with a treated piece of Ni foam (NF) (3 × 4 cm2) reacting at 160 °C for 16 h. After that, Ni(OH)2/CoMoO4 (NH/CMO) was rinsed with DI water repeatedly and then dried. The synthesized substances were named NH/CMO-1, 2, 3, and 4 according to the used dosages of Ni. Then, the precursors were placed at the downstream side in a porcelain boat, while 1 g of the NaH2PO2·H2O was at the upstream side. Oxygen was carefully removed by three times purging (with Ar gas) and pumping process. The final products were obtained by annealing at 350 °C for 2 h under Ar, denoted as NiP2/CoMoP-1 (NCMP-1), NiP2/CoMoP-2 (NCMP-2), NiP2/CoMoP-3 (NCMP-3), and NiP2/CoMoP-4 (NCMP-4). For comparison, pure CoMoP was synthesized similarly to NCMP-x, by only adding Co(NO3)2·6H2O and Na2MoO4 without changing other experimental parameters.

Characterization

Transmission electron microscopy (TEM) was carried out by a JEM-2200FS, 200 kV from JEOL, while scanning electron microscopy (SEM) was performed by a SUPRA 55, 5 kV from Zeiss, equipped with an energy-dispersive X-ray spectroscopy (EDX) setup. X-ray diffraction was recorded by a SmartLab (3) X-ray diffraction system. X-ray photoelectron spectroscopy (XPS) with an Al Kα X-ray source and ultraviolet photoemission spectroscopy (UPS) with He I ultraviolet light were performed on a PHI 5000 VersaProbe III.

Electrochemical Measurements

All HER tests were performed by the CHI 760E electrochemical workstation (Chenhua, China) at a constant temperature. The reference electrode was Hg/HgO, and the carbon rod was the counter electrode. All tests were carried out in a basic environment (1 M KOH). The electrolyte was ventilated with high-purity Ar for 0.5 h before each HER test. The electrocatalytic performances toward the HER were measured by linear sweep voltammetry (LSV) at a scan rate of 2 mV s–1 without iR correction. Cyclic voltammetric (CV) curves with a scan rate of 10–100 mV s–1 were plotted over a voltage range of 0.06–0.16 V (vs RHE) without Faradaic currents to compute double-layer capacitance (Cdl) and thus estimated electrochemical active surface area (ECSA). Electrochemical impedance spectra (EIS) were recorded under identical conditions (100 kHz to 10 mHz, −0.075 V vs RHE). Chronoamperometry (CP) at constant overpotentials was conducted to evaluate durability. All potentials were corrected by reversible hydrogen electrode (RHE) potentials.