Xuemei
Zhou
a,
Sijia
Guo
a,
Qiran
Cai
b and
Shaoming
Huang
*a
aSchool of Material and Energy, Guangzhou Key Laboratory of Low Dimensional Materials and Energy Devices, Guangdong University of Technology, Guangzhou, 51006, P. R. China. E-mail: smhuang@gdut.edu.cn
bInstitute for Frontier Materials, Deakin University, Geelong Waurn Ponds Campus, Victoria 3216, Australia
First published on 19th July 2019
Oxygen evolution reaction (OER) catalysts are of importance for electrochemical water splitting and fuel generation. Despite enormous efforts, the design and development of OER catalysts with high catalytic activities under neutral conditions are highly desired but still remain a great challenge. Herein, we report a room temperature chemical route to prepare xceria/cobalt borate (xCeO2/Co–Bi) hybrids as efficient OER catalysts by tuning the molar ratio of Ce/Co (x represents the amount of CeO2). The optimised catalyst (20CeO2/Co–Bi hybrid) was found to exhibit remarkable OER catalytic activity with an overpotential of 453 mV at a current density of 10 mA cm−2, Tafel slope of 120 mV dec−1 and long-term stability in neutral medium due to its good conductivity, mass transportation and strong synergetic coupling effects, showing the potential of Co-based electrochemical materials for practical application in energy storage devices.
Oxygen-evolving complexes composed of manganese and calcium are well known as natural water-splitting complexes.17,18 Inspired by this thought, a new category of artificial water oxidation electrocatalysts with amorphous features consisting of metals (such as nickel and cobalt), oxygen, and inorganic borate, namely nickel–borate (Ni–Bi) or cobalt–borate (Co–Bi), has attracted significant attention due to their low cost, high intrinsic activity, and superior stability.13,14,19–22 For example, Nocera et al. reported that a nickel–borate (Ni–Bi) film, electro-deposited on indium tin oxide (ITO), had high catalytic activity towards electrochemical water oxidation under near-neutral conditions, achieving a current density of 1 mA cm−2 in 0.1 M borate buffer at an overpotential of 425 mV.19 A nickel–borate nanoarray supported by carbon cloth (Ni–Bi/CC) yielded a geometrical catalytic current density of 10 mA cm−2 at an overpotential of 470 mV.14 Recently, amorphous Co–Bi ultrathin nanosheets were designed for electrochemical water oxidation under neutral conditions.13,23 However, the application of Co–Bi as a catalyst is limited by its poor OER kinetics and mass transfer ability, which can be solved by adding OER co-catalysts.
As one of the most important rare earth oxides, ceria (CeO2) has been usually used in the electrochemical field due to the following advantages:7,24–27 (i) flexible transition between the Ce3+ and Ce4+ oxidation states, which can provide the opportunity to generate strong electronic interactions with other matrices and therefore probably improves the catalytic performance; (ii) good electronic/ionic conductivity, which can improve charge transfer and promote the reaction process; (iii) reversible surface oxygen ion exchange, which will absorb O2 produced during the OER and enhance the OER activity; (iv) large oxygen-storage capacity, which is helpful to form and repair oxygen vacancies on the CeO2 surface, binding adsorbates much more strongly than normal oxide sites.24 Therefore, the introduction of CeO2 onto Co–Bi nanosheets will be beneficial for the formation of hydroperoxy species (OOHad) on the surface of the hybrids owing to the high mobility of oxygen vacancies. Furthermore, the chemical synergistic effect between Co–Bi nanosheets and CeO2 would also improve the OER catalytic performance. Thus, the CeO2/Co–Bi hybrids should be efficient candidates to improve the electrocatalytic activity for OER in neutral medium.
In this work, we report a room temperature chemical route to prepare xCeO2/Co–Bi hybrids (10CeO2/Co–Bi, 20CeO2/Co–Bi, 30CeO2/Co–Bi; x represents the amount of CeO2) through tuning the molar ratio of Ce/Co, and their OER catalytic activities were compared. It was found that 20CeO2/Co–Bi exhibited the highest OER catalytic activity with an overpotential of 453 mV at a current density of 10 mA cm−2, Tafel slope of around 120 mV dec−1 and long-term stability in phosphate-buffered saline (PBS) solution because of the good conductivity, mass transportation and strong synergetic coupling effects.
Raman spectroscopy was applied to further confirm the presence of CeO2 in the as-prepared composites because the F2g mode of CeO2 of the fluorite structure is Raman active at 465 cm−1.32,33Fig. 2f shows the Raman spectra collected from the composites (black line indicates pure Co–Bi nanosheets), and the intensity increases with the increase in the amount of CeO2. The ∼10 cm−1 shift of Raman frequency between our samples and literature reports33–35 is attributed to the size effects, which is consistent with a previous report.36
The specific surface area of the as-grown samples was measured from the standard BET procedure (Fig. S3†). The BET surface areas of 10CeO2/Co–Bi, 20CeO2/Co–Bi and 30CeO2/Co–Bi were 249.8 m2 g−1, 290.2 m2 g−1 and 330.1 m2 g−1, respectively, much higher than that of pure Co–Bi nanosheets (134.0 m2 g−1). It is generally accepted that a larger surface area means better catalytic activity.
To evaluate the OER activities of the CeO2/Co–Bi composites, the catalysts were loaded onto the CFP electrode with a density of 0.56 mg cm−2 by drop casting. The potentials reported in this work were calibrated to the RHE using the following equation: E(RHE) = E(Ag/AgCl) + (0.197 + 0.0591 × pH) V.13Fig. 3a shows the LSV curves of a bare CFP substrate, Co–Bi, 10CeO2/Co–Bi, 20CeO2/Co–Bi, 30CeO2/Co–Bi and commercial RuO2/C electrodes in 0.1 M PBS. Compared to the strong catalytic current from electrodes with catalysts, almost no current density from the bare CFP substrate was detected at 1.8 V (vs. RHE), indicating that CFP is inactive towards O2 evolution and a background correction for the CFP support is unnecessary for all catalytic electrodes. The overpotentials at the specified current density of 5 mA cm−2, the current densities at 1.8 V (vs. RHE) and the Tafel slopes of all catalysts, obtained in 0.1 M PBS, are summarized in Table 1. The 20CeO2/Co–Bi electrode demands an overpotential of only 346 mV to deliver a catalytic current density of 5 mA cm−2, much lower than those of 10CeO2/Co–Bi (446 mV) and 30CeO2/Co–Bi (525 mV) under the same catalytic current density, respectively. For comparison, the electrocatalytic OER activity of Co–Bi nanosheets was measured as well, obtaining an overpotential of 560 mV at a current density of 5 mA cm−2, much higher than those of CeO2/Co–Bi composite electrodes. In addition, we obtained an overpotential of 453 mV from the 20CeO2/Co–Bi electrode at a current density of 10 mA cm−2, which is close to that of RuO2/C (430 mV) and also comparable to the values reported previously under neutral conditions (Table S1†).13,14,23,37–44 The current density of the 20CeO2/Co–Bi electrode at 1.8 V (vs. RHE) was 19.2 mA cm−2, not only much larger than 11.4, 7.2, and 5.7 mA cm−2 obtained from 10CeO2/Co–Bi, 30CeO2/Co–Bi and Co–Bi electrodes, respectively, but also superior to those of A-CoS4.6O0.6 PNCs (4.59 mA cm−2) and Co–Bi NS/G (14.4 mA cm−2) measured at the same potential (1.8 V vs. RHE).13,45 All these results indicate the high OER catalytic activity of the 20CeO2/Co–Bi hybrid under neutral conditions.
Sample | η @5 (mA cm−2) | J @1.8 V (vs. RHE) | Tafel slope (mV dec−1) | BET surface area (m2 g−1) | Capacitance (mF cm−2) |
---|---|---|---|---|---|
a η refers to the overpotential (mV) of the samples at a current density of 5 mA cm−2. b J refers to the current densities of the catalysts at a specific overpotential. | |||||
Co–Bi | 560 | 5.7 | 210 | 134.0 | 0.98 |
10CeO2/Co–Bi | 446 | 11.4 | 160 | 249.8 | 1.33 |
20CeO2/Co–Bi | 346 | 19.2 | 120 | 290.2 | 1.58 |
30CeO2/Co–Bi | 525 | 7.2 | 198 | 330.1 | 1.09 |
We also fitted the polarization curves (Fig. 3a) using the Tafel equation η = blog(j/j0), where η represents the overpotential, b is the Tafel slope, j refers to the current density and j0 is the exchange current density.9 Accordingly, the derived Tafel slope of 120 mV dec−1 from 20CeO2/Co–Bi was much lower than the slope of 210 mV dec−1 from Co–Bi (Fig. 3b) and close to or even lower than the values measured from other reported catalysts, including 3D Ni–Bi nanowire arrays (107 mV dec−1),14 Co–Bi/graphene composites (160 mV dec−1),13 Ni–Bi/RGO (176 mV dec−1)46 and 3D Co–Pi nanowires (187 mV dec−1),12 suggesting the high OER activity of 20CeO2/Co–Bi. Additionally, Fig. 3c shows a multi-step chronopotentiometric curve for the 20CeO2/Co–Bi composite electrode at current densities from 2 to 9 mA cm−2 (ca. 1 mA cm−2 per 500 s). The potential immediately levels off at 1.13 V (vs. Ag/AgCl) at the initial current value and remains unchanged for 500 s, and the other steps also show similar results, implying good conductivity, mass transportation, and mechanical robustness of this electrode.47,48 These results further demonstrate that 20CeO2/Co–Bi composite electrodes have excellent OER catalytic performance.
The reasons for the enhancement of electrochemical catalytic OER activity of CeO2/Co–Bi compared to that of Co–Bi are summarised as follows. First, CeO2/Co–Bi composites have a much larger surface area than Co–Bi and thus more catalytically active sites, which can significantly improve the OER process (Table 1). Although the surface area of the 30CeO2/Co–Bi composite (330.1 m2 g−1) is larger than that of the 20CeO2/Co–Bi composite (290.2 m2 g−1), 30CeO2/Co–Bi shows worse OER performance than 20CeO2/Co–Bi because the nanosheet structure of 30CeO2/Co–Bi has been destroyed (Fig. 2c). Electrochemical impedance spectroscopy (EIS) was used to further investigate the detailed characteristics of Co–Bi, 10CeO2/Co–Bi, 20CeO2/Co–Bi, and 30CeO2/Co–Bi as capacitive electrodes (see ESI Fig. S4†). The semicircular characteristic of EIS curves suggests the smallest charge transfer and the best charge conductivity of 20CeO2/Co–Bi compared to those of other catalysts (Co–Bi, 10CeO2/Co–Bi and 30CeO2/Co–Bi) and thus the highest OER catalytic activity among the samples.
Second, CeO2/Co–Bi composites have a larger electrochemical surface area (ECSA) compared to Co–Bi. A larger ECSA means more accessible surface permeation and thus higher catalytic activity. According to the CV curves of Co–Bi, 10CeO2/Co–Bi, 20CeO2/Co–Bi, and 30CeO2/Co–Bi (see ESI Fig. S5†), capacitive currents as a function of the scan rate of the samples are plotted in Fig. 3c. The double-layer capacitance (Cdl) of each sample was estimated from the slope of the corresponding linear fitting, and therefore, one can obtain the value of ECSA, which is double the Cdl value.45 Obviously, 20CeO2/Co–Bi has the highest Cdl among the four samples (Fig. 3d), and thus, the largest ECSA contributes to the best catalytic performance.
Lastly, in order to investigate the strong electronic interactions between Co–Bi and CeO2, XPS was applied to characterize pure CeO2, Co–Bi, 20CeO2/Co–Bi, 20CeO2/Co–Bi and 30CeO2/Co–Bi hybrids, and the results are plotted in Figs 4, S6 and S7.† All the spectra were referenced to the aliphatic carbon at a binding energy (BE) of 284.5 eV. The survey spectra of 20CeO2/Co–Bi showed the presence of Co, B, O and Ce, but no Ce was detected from Co–Bi nanosheets (Fig. 4c), consistent with EDS and EELS results (see ESI Fig. S1 and S2†).
Fig. 4 XPS spectra of Co–Bi and 20CeO2/Co–Bi composites (a), and high-resolution XPS spectra for B (b), Ce (c) and Co (d). |
As shown in Fig. 4b, the peak at 192.5 eV is assigned to the core level of B3+ in borate species. The O 1s spectra of pure CeO2, Co–Bi, 10CeO2/Co–Bi, 20CeO2/Co–Bi and 20CeO2/Co–Bi are shown in Fig. S8.† The peak at 529.4 eV is ascribed to the O 1s of Ce–O and that at 531.9 eV corresponds to O 1s of the central oxygen atoms in borate.25,49 For high-resolution Ce 3d spectra of pure CeO2 (Fig. 4c), the peaks located at 896–926 eV correspond to Ce 3d3/2, and the peaks located at 880–892 eV are consistent with Ce 3d5/2, which demonstrated the coexistence of Ce3+ and Ce4+ in CeO2. However, after introducing CeO2 NPs onto the Co–Bi nanosheets, the ratio of Ce3+:Ce4+ in the CeO2/Co–Bi hybrid changed compared with that of pure CeO2, suggesting that the valence states of Ce in the CeO2/Co–Bi hybrid have rearranged.25,26 As shown in Fig. 4d, the binding energy of Co 2p from 20CeO2/Co–Bi showed an ∼0.5 eV negative shift compared to that of pure Co–Bi nanosheets, presumably caused by the electron transfer from CeO2 to Co–Bi, and it is also found that there is a more negative shift of the CeO2/Co–Bi hybrid with the increase in the amount of CeO2 (Fig. S7†). This phenomenon has been demonstrated in previous reports.24,26,50 Moreover, this modification in the electronic structure makes CeO2 more acidic (Lewis acid) and thus facilitates the activation of H2O molecules (Lewis base), benefiting the formation of OOHad on the surface of the composites and facilitating the OER process and thus enhancement of the efficiency of water oxidation.24 Additionally, the increase in the charge conductivity of CeO2/Co–Bi after introducing CeO2 indicates faster charge transfer during the catalytic reaction, improving the catalytic efficiency of water oxidation (see ESI Fig. S6†). After introducing the small CeO2 nanoparticles, the hybrids display a larger surface area, smaller charge transfer, better charge conductivity and stronger synergistic effects. All these advantages of the hybrids together are responsible for the good OER catalytic activity of the new CeO2/Co–Bi catalysts.
Besides the catalytic activity, the stability of the catalyst, another major concern in designing cost-effective OER catalysts, was investigated as well. After 1000 CV cycles in 0.1 M PBS (Fig. S8†), there was almost no obvious loss in catalytic current (Fig. 5a), suggesting its superior stability. Additionally, the stability of the 20CeO2/Co–Bi composite was determined by measuring chronoamperometric responses (i–t) at 1.65 V (vs. RHE) in 0.1 M PBS. During the measurements, the working electrode was continuously rotating at 1600 rpm to remove the generated O2 bubbles. As shown in Fig. 5b, the stabilized current density indicates no obvious deactivation of the 20CeO2/Co–Bi composite over 19 h. The durability test of the 20CeO2/Co–Bi electrode under higher current density was also performed, as shown in Fig. S9.† The current density of the 20CeO2/Co–Bi composite over 30 h was only slightly attenuated. The results suggest that the 20CeO2/Co–Bi composite exhibits good catalytic stability under neutral conditions.
Footnote |
† Electronic supplementary information (ESI) available: SEM, TEM, XRD and SAED of Co–Bi nanosheets, EDS mapping images, EEIS spectrum of 20CeO2/Co–Bi, nitrogen adsorption/desorption isotherms of as-grown products (Co–Bi, 10CeO2/Co–Bi, 20CeO2/Co–Bi and 30CeO2/Co–Bi composites), CVs of all the samples (Co–Bi, 10CeO2/Co–Bi, 20CeO2/Co–Bi and 30CeO2/Co–Bi composites), impedance Nyquist plots of Co–Bi, 10CeO2/Co–Bi, 20CeO2/Co–Bi and 30CeO2/Co–Bi composite electrodes. CVs of the 20CeO2/Co–Bi composite electrode, comparison of the OER activities of cobalt-based catalysts in neutral medium listed in Table S1. See DOI: 10.1039/c9na00356h |
This journal is © The Royal Society of Chemistry 2019 |