FeCo_66.7Ni_33.3B Nanoparticles Integrated on Vertically Aligned Boron-Doped Graphene Array as Efficient Electrocatalyst for Overall Water Splitting in Wide pH Range

NiSO₄·6H₂O, CoSO₄·7H₂O, FeSO₄·7H₂O, C₆H₅Na₃O₇·2H₂O and C₂H₁₀BN (10% aqueous solution) were purchased from Shanghai Aladdin Biochemical Technology Co., Ltd Graphite powder (45 μm, 99.99% purity), H₂SO₄ (98%), HCl (37.5%), KOH, NaNO₃, KMnO₄, H₂O₂ (30%), ethanol, isopropanol and H₃BO₃ were purchased from Sinopharm Group Chemical Reagent Co., Ltd. Pt/C (20%) and IrO₂ (99.0%) were purchased from Shanghai Macklin Biochemical Co., Ltd. Polytetrafluoroethylene (PTFE) pipes with inner diameter of 40 mm and thickness of 2 mm were purchased from Yangzhong Hongda Fluoroplastics Co., Ltd Copper plates with inner diameter of 44 mm and depth of 10 mm were purchased from Zhuji Yishun Building Materials Co., Ltd The PTFE pipe and the copper plate composed the mold of directional freezing. All the reagents were of analytical grade and used as received. Deionized water was used in all experiments.

GO was prepared by a modified Hummers method. ¹¹ GO (120 mg) was dispersed in the mixture of 400 mg boric acid (the source of boron), 400 μl ethanol (antifreeze) and 10 ml deionized water by ultrasonicating for 1 h. Then the mixture was put in a mold assembled by polytetrafluoroethylene cylinder and copper plate. During the directional freezing process by liquid nitrogen, the ice crystals were influenced by gradients from bottom to the top, resulting in their growth along bottom to the top, and finally caused the formation of vertical aligned structure. ⁹ The mold was then put in a freeze-drying machine for 48 h. The freeze-dried samples were placed in a tubular furnace to get an annealing process, heated to 800 °C at a heating rate of 10 °C min⁻¹ for 3 h in nitrogen atmosphere. Finally, the tubular furnace was cooled to room temperature, and the preparation of B-VG was completed.

The electroless plating baths were prepared according to Table I. The prepared B-VG was immerse into the bath and then heated to 110 °C and held for 5 min, following by adding 0.03 M C₂H₁₀BN. After the reaction last for 20 min. The products were washed with deionized water to remove the residual solution, and then put into a vacuum freeze-drying machine for 12 h. Finally, a series of composite electrocatalysts with different cationic ratio of Co and Ni were prepared. The theoretical composition and metal element ratio are shown in Table II.

Field emission scanning electron microscopy (SEM, ZEISS Gemini 500) and field emission transmission electron microscope (HRTEM, Tecnai G2 F20) were utilized to examine the morphologies of the samples. Energy-dispersive X-ray spectrum (EDS) was used to analyze the proportion of the element in the samples. Elemental mapping was used to analyze the dispersion of different elements. X-ray diffraction (XRD) patterns using the Cu Kα radiation in Bragg-Brentano (θ − 2θ) configuration performed phase characterization of the FeCo_xNi_yB@B-VG. X-ray photoelectron spectra (XPS) was carried out on an X-ray photoelectron spectrometer system (ESCALAB Xi⁺).

The electrochemical performance of the catalyst was tested by CHI 660E electrochemical workstation (Shanghai Chenhua Instrument Co. Ltd.). In a typical three electrode system, the saturated calomel electrode (SCE) was used as the reference electrode and the graphite electrode was used as the counter electrode. The prepared electrocatalyst (effective area is 1 × 1 cm²) material was directly used as the working electrode. The electrochemical double-layer capacitance (C_dl) of the catalyst was measured by cyclic voltammetry (CV) in both 1.0 M KOH and 0.5 M H₂SO₄ electrolyte. The electrocatalytic properties of HER and OER were investigated by linear sweep voltammetry (LSV) and the scanning speed was 5 mV s⁻¹ in 0.5 M H₂SO₄ and 1.0 M KOH electrolyte, respectively. The charge transfer resistance (Rct) and solution resistance (Rs) of the electrocatalyst was measured by electrochemical impedance (EIS) in 1.0 M KOH electrolyte. The stability of the electrocatalyst material was studied in 0.5 M H₂SO₄ and 1.0 M KOH electrolyte by chronopotentiometry (CP). The test time was 10 h. To evaluate the application potential of the FeCo_66.7Ni_33.3B@B-VG electrocatalyst, the faraday efficiency was tested. All the test potential was transformed into the reversible hydrogen electrode (RHE) potential by the equation: E(vs RHE) = E(vs SCE) + 0.241 + 0.05916 pH.

The vertically aligned structure of the side part and porous structure of the top part of FeCo_66.7Ni_33.3B@B-VG electrocatalyst resulting from the directional freezing are showed in the SEM image (Fig. 1a). The well-maintained structure demonstrates that B-VG has a remarkable mechanical strength, which allows no damage during the electroless plating process. In the gaps between the parallel graphene sheets, whose width is approximately 20–30 μm, some curly graphene bridges can be observed. The graphene bridges are beneficial for the conductivity of the electrocatalyst, and the bridges may come from the annealing process. ¹⁰ The even-distributed FeCo_66.7Ni_33.3B nanoparticles exhibited in Fig. 1b prove the excellent effect of B-VG on avoiding the nanoparticles from agglomeration. HRTEM (Fig. 1c) exhibits that the average size of the FeCo_66.7Ni_33.3B nanoparticles is around 100 nm. Figure 1d demonstrates the high crystallinity of the FeCo_66.7Ni_33.3B nanoparticles with the lattice fringes of (211), (110) and (110) attribute to BFe₂, Fe₇Ni₃ and Co₃Fe₇ phases, respectively. EDS spectrum and column of detailed data (Fig. 1e) display that the atomic percentage of Co and Ni is 15.81% and 7.79%, and the relative contents of Co and Ni are x = Co/(Co + Ni)% = 67.0% and y = Ni/(Co + Ni)% = 33.0%, which proves tuning cationic ratio of FeCo_66.7Ni_33.3B nanoparticles by electroless plating is successful. Elemental mapping pictures (Figs. 1f–1k) exhibit the dispersion situation of element C, B, O, Fe, Co and Ni, respectively. The dispersion situation of C and B (Figs. 1f and 1g) demonstrates that the B element may successfully doped into the graphene sheets, and the B element concentrates more in the center of the Fig. 1g means that B element exists in the nanoparticle. Figure 1h displays the distribution of O element, the even-distributed O element may come from the oxygen-containing groups on the surface of the graphene. And in the center of Fig. 1h, the more concentrated O element may come from the oxidized surface of the FeCo_66.7Ni_33.3B nanoparticle, which is attributed to the inevitable contact with the air. Figures 1i–1k show the well-distributed element of Fe, Co and Ni in single nanoparticle, thus proving the successful synthesis of FeCo_66.7Ni_33.3B.

Zoom In Zoom Out Reset image size

Catalytic property of the prepared electrocatalysts for HER

According to Fig. 2a, FeCo_66.7Ni_33.3B@B-VG electrocatalyst achieves a current density of 10 mA cm⁻² at a low overpotential of 157 mV in 0.5 M H₂SO₄ electrolyte, which is comparable to that of Pt/C@B-VG (146 mV), and much better than FeCo_33.3Ni_66.7B@B-VG, FeCo_83.3Ni_16.7B@B-VG and FeCo_16.7Ni_83.3B@B-VG. Furthermore, the Tafel slope of FeCo_66.7Ni_33.3B@B-VG is given as merely as 47 mV dec⁻¹ (Fig. 2b), which is comparable to that of Pt/C@B-VG (57 mV dec⁻¹), and significantly smaller than the others. As shown in Fig. 2c, the FeCo_66.7Ni_33.3B@B-VG electrocatalyst delivers the best HER catalytic activity in 1.0 M KOH solution by displaying a low overpotential of 125 mV at a current density of 10 mA cm⁻², indicating its outstanding HER electrocatalytic performance, which is even better than Pt/C@B-VG (150 mV). The Tafel slope of FeCo_66.7Ni_33.3B@B-VG is 52 mV dec⁻¹ in 1.0 M KOH solution (Fig. 2d), also lower than that of Pt/C@B-VG (55 mV dec⁻¹) and other prepared electrocatalysts. The different electrocatalytic performance of FeCo_xNi_yB means that the appropriate increase of Co content in FeCo_xNi_yB@B-VG can effectively reduce the overpotential of water splitting and improve the electrocatalytic activity of the materials. Excessive content of Co will decrease the electrocatalytic performance instead. Because different metal has its own function for water splitting. For example, Co element is beneficial to decrease the overpotential, Ni element is helping to boost the current. ¹² Excessive content of Co will cause the reduction of the relative content of Ni and result a weakened synergistic effect between Co and Ni, thus reducing the electrocatalytic activity of the materials.

Zoom In Zoom Out Reset image size

Catalytic property of the prepared electrocatalysts for OER

According to Fig. 3, FeCo_66.7Ni_33.3B@B-VG possesses the best OER activity among as-prepared samples, with overpotential of 569 mV and 200 mV at current density of 10 mA cm⁻² and Tafel slope of 225 mV dec⁻¹ and 34 mV dec⁻¹ in 0.5 M H₂SO₄ and 1.0 M KOH, respectively. FeCo_66.7Ni_33.3B@B-VG even shows a better catalytic performance than IrO₂@B-VG (overpotential of 318 mV and Tafel slope of 90 mV dec⁻¹) in 1.0 M KOH solution. In Fig. 3c, the slightly raised peak at 1.38 V may be attributed to the oxidation of metal. The higher performance of OER is usually related to the conversion from metal to oxide/hydroxide, which serves as active sites during OER, on the surface of the borides, and the oxide/hydroxide layer can also improve the stability of OER in alkaline solution. ^13–15 All the as-prepared samples demonstrate a worse catalytic performance in 0.5 M H₂SO₄ solution compared with IrO₂ (overpotential of 256 mV and Tafel slope of 245 mV dec⁻¹), because the corrosion charge of transition metal compounds at anode is fast in acidic solution. ¹⁴ From these phenomena, it can be found that with the relative content of Co increasing from 16.7% to 66.7%, HER and OER electrocatalytic activities of the catalyst also increase correspondingly. However, when the relative content of Co is further increased to 83.3% and the content of Ni is reduced to 16.7%, the electrocatalytic activity of the catalyst decreases greatly. In a word, when the relative contents of Co and Ni are 66.7% and 33.3% respectively, a best catalytic performance of FeCo_xNi_yB is achieved.

Zoom In Zoom Out Reset image size

The XRD patterns of FeCo_66.7Ni_33.3B@B-VG and B-VG as comparison are exhibited in Fig. 4a. The peak of B-VG located at 2θ = 24° disappears in the XRD pattern of FeCo_66.7Ni_33.3B@B-VG for the intensities of peaks ascribe to FeCo_66.7Ni_33.3B are too strong. The peak centered at 2θ = 44.84° in the XRD pattern of FeCo_66.7Ni_33.3B@B-VG can be ascribed to the (211) plane of BFe₂ (JCPDS card no.65-2693) and (110) planes of Co₃Fe₇ and Fe₇Ni₃ (JCPDS card no.48-1816, no.65-7251). The peak centered at 2θ = 65.1° can be ascribed to the (110) planes of Co₃Fe₇ and Fe₇Ni₃ (JCPDS card no. 48-1816, no. 65-7251). The planes of FeCo_66.7Ni_33.3B@B-VG can match well with the results achieved by the HRTEM analysis, and the shape of the peaks means that FeCo_66.7Ni_33.3B@B-VG electrocatalyst is well crystalline. Figure 4b shows the XPS spectra about FeCo_66.7Ni_33.3B@B-VG and there are six characteristic peaks of B 1s, C 1s, O 1s, Fe 2p, Co 2p and Ni 2p, illustrating the FeCo_66.7Ni_33.3B@B-VG catalyst material is composed of B, C, O, Fe, Co and Ni. The Fe 2p spectrum of FeCo_66.7Ni_33.3B@B-VG (Fig. 4c) observes three peaks, the fitted peak at 715.35 eV can be attributed to Fe²⁺ 2p3/2. The peaks at 709.7 and 721.9 eV are due to the zero-valence state of Fe 2p3/2 and Fe 2p1/2, respectively. ^16,17 The Co 2p spectrum of FeCo_66.7Ni_33.3B@B-VG (Fig. 4d) exhibits four peaks. The peaks located around 778.5 and 794.15 eV are assigned to metal Co 2p3/2 and Co 2p1/2. The peaks around 783.15 eV and 799.95 eV are also detected, which are assigned to the Co²⁺ 2p3/2 and Co²⁺ 2p1/2. ^16,17 In the Ni 2p spectrum (Fig. 4e), the peaks at 825.90 eV and 870.60 eV are assigned to Ni 2p3/2 and Ni 2p3/2, respectively. And the peak at 849.5 eV is assigned to satellite peak. The other peaks at 858.4 and 876.85 eV are attributed to Ni²⁺ 2p3/2 and Ni²⁺ 2p1/2. ^16,17 From the B1s spectrum in Fig. 4f, two peaks at 188.48 and 195.73 eV for FeCo_66.7Ni_33.3B@B-VG are attributed to the alloying structure of B, which confirms the existence of FeCoNi-B, B₃C and the oxidation species of B, including CB₂O, CBO₂ and B₂O₃. ^18–20 B₂O₃ can be attributed to the oxidation of the electrocatalyst. ²¹ In the C 1s spectrum (Fig. 4g), the peaks at 208.24, 281.2, 281.85, 282.65 and 285.35 eV are related to the C-B, sp²C, sp³C, C−O and O−C=O, respectively. ^22–24 C-B indicates that B has successfully substituted for C atom and doped into graphene lattice, which is correspond to the B₃C peak observed in Fig. 4f and the hypothesis from the EDS mapping (Fig. 1g). ^25–28 C−O and O−C=O structures with higher binding energy originate from incomplete reduction of graphene. For the high-resolution O 1 s spectrum in Fig. 4h, three types of O configurations are ascribed to adsorbed oxygen (528.34 eV), OH (526.56 eV), and OOH (529.19 eV). ^29–31 The specific content calculated by fitting the peak area of each element is shown in Fig. 4i, it was found that the atomic ratio of Co and Ni in the sample is 3.98% and 1.89%, respectively. The relative content of Co is x = Co/(Co + Ni)% = 67.8%, and the relative content of Ni is y = Ni/(Co + Ni)% = 32.2%, which is very close to the data got from EDS analysis, indicating that the adjustment of the relative contents of Co and Ni in this sample is very successful.

Zoom In Zoom Out Reset image size

As shown in Fig. 5a, the FeCo_66.7Ni_33.3B@B-VG exhibits a C_dl value of 40.4 mF cm⁻² in acidic solution, which is larger than those of the FeCo_33.3Ni_66.7B@B-VG (30.9 mF cm⁻²), FeCo_83.3Ni_16.7B@B-VG (24.6 mF cm⁻²) and FeCo_16.7Ni_83.3B@B-VG (23.5 mF cm⁻²) electrocatalysts, indicating that more active sites and a larger ECSA exist in FeCo66.7Ni33.3B@B-VG. Similarly, the C_dl values of the prepared electrocatalysts in alkaline solution are also exhibited in Fig. 5b, FeCo_66.7Ni_33.3B@B-VG exhibits a C_dl value of 43.8 mF cm⁻², which is much bigger than those of FeCo_33.3Ni_66.7B@B-VG (32.2 mF cm⁻²), FeCo_83.3Ni_16.7B@B-VG (31.4 mF cm⁻²) and FeCo_16.7Ni_83.3B@B-VG (26.7 mF cm⁻²). As presented in Fig. 5c, the FeCo_66.7Ni_33.3B@B-VG shows lower Rct and Rs than others, suggesting that when the relative contents of Co and Ni are 66.7% and 33.3%, the FeCo_xNi_yB exhibits best intrinsic catalytic activity. ³² As exhibited in Fig. 5d, the catalytic activities towards HER in both 0.5 M H₂SO₄ and 1.0 M KOH solutions and catalytic activity towards OER in 1.0 M KOH solution can be maintained for at least 10 h. After 10 h, the potential of FeCo_66.7Ni_33.3B@B-VG decays very little, indicating that FeCo_66.7Ni_33.3B@B-VG electrocatalyst maintains good long-term stability towards HER in a wide pH range, even in acid condition. And FeCo_66.7Ni_33.3B@B-VG electrocatalyst exhibits remarkable long-term stability towards OER in alkaline solution as well. It can be observed from the picture that in the double-electrode system, when the FeCo_66.7Ni_33.3B@B-VG electrocatalyst is directly used as the electrodes in the water splitting reaction, dense H₂ and O₂ bubbles are formed at the cathode and anode respectively. The material does not deform or disintegrate, which means a good mechanical strength of FeCo_66.7Ni_33.3B@B-VG. In short, the long-term stability comes from the great mass transfer capacity and mechanical strength of the FeCo_66.7Ni_33.3B@B-VG electrocatalyst and the good intrinsic stability of FeCo_66.7Ni_33.3B nanoparticles. Furthermore, the oxide/hydroxide layer produced during OER on the surface of the FeCo_66.7Ni_33.3B nanoparticles helps to improve the stability towards OER in alkaline solution. ¹⁴

Zoom In Zoom Out Reset image size

Figure 6a shows the measured volume of H₂ and O₂ generated by FeCo_66.7Ni_33.3B@B-VG when the current density reaches 10 mA cm⁻² during water splitting. The amount of H₂ collected is about twice as much as the amount of O₂. Figure 6b exhibits that the corresponding Faraday efficiency of FeCo_66.7Ni_33.3B@B-VG is around 90% according to the measured gas volume, indicating that the FeCo_66.7Ni_33.3B@B-VG electrocatalyst has a high efficiency and great potential for application.

Zoom In Zoom Out Reset image size

The catalytic properties of different transition metal composite electrocatalysts for water splitting in different electrolytes are exhibited in Fig. 7. ^33–38 It can be concluded that FeCo_66.7Ni_33.3B@B-VG has higher electrocatalytic activity than other transition metal composite electrocatalysts.

Zoom In Zoom Out Reset image size