Varsha
K.
a,
Kiran
G. K.
b,
Sutar Rani
Ananda
a,
Lokesh Koodlur
Sannegowda
*c and
Shambhulinga
Aralekallu
*a
aCentre for Research in Functional Materials (CRFM), JAIN (Deemed-to-be University), Jain Global Campus, Bengaluru 562112, Karnataka, India. E-mail: shambulinga.a@jainuniversity.ac.in
bDepartment of Chemical Engineering, Quantum Nano Centre, University of Waterloo, N2L 3G1 Ontario, Canada
cDepartment of Studies in Chemistry, Vijayanagara Sri Krishnadevaraya University, Vinayakanagara, Ballari-583105, India. E-mail: kslokesh@vskub.ac.in
First published on 17th March 2025
The oxygen evolution reaction (OER) is a critical and bottleneck process in electrochemical energy applications. This study presents a straightforward hydrothermal method for preparing a NiCo bimetallic organic framework (NiCo-MOF) with three unique ligands. The NiCo-trimesic acid-based MOF on carbon cloth (NiCo-t-MOF/CC) can sustain the industrially relevant current density of 100 mA cm−2 for over 62 hours despite the observed gradual increase in potential in 1 M KOH without replacing the electrolyte. The NiCo-t-MOF/CC electrocatalyst achieved a significantly lower overpotential of 440 mV to reach a current density of 100 mA cm−2, outperforming the benchmark RuO2 catalyst, which required 581 mV. A Tafel slope of 83 mV dec−1 at NiCo-t-MOF/CC indicates faster oxygen evolution kinetics than at RuO2/CC (97 mV dec−1). Interestingly, NiCo-t-MOF/CC||Pt–C/CC exhibited a relatively diminished cell voltage of 1.54 V to deliver a current density of 10 mA cm−2, which is close to the thermodynamic water splitting energy of 1.23 V. The performance of NiCo-t-MOF/CC is promising at higher current densities for industrial applications.
Metal–organic frameworks (MOFs) are crystalline, highly porous materials constructed from metal ions or clusters in coordination with organic linkers.12–14 Due to their tunable porosity and readily accessible active sites, MOFs have received tremendous attention in several electrochemical applications. In recent years, first-row transition metals such as Ni, Co, and Fe are considered as promising electrocatalysts owing to their superior electrochemical activity, low cost and stability towards the OER.12,15–17 Sun et al.18 reported that NiFe-based MOF nanosheets directly grown on a Ni foam substrate act as a robust electrocatalyst for the electrochemical OER. Duan et al.19 demonstrated that a bimetallic MOF nanosheet-based electrocatalyst with highly exposed molecular active sites enhances the catalytic activity for overall water-splitting reactions. Here, this study brings three unique ligands, i.e., 2-methyl imidazole, trimesic acid, and terephthalic acid, to prepare NiCo bimetallic MOFs (named as NiCo-im-MOF, NiCo-t-MOF and NiCo-tp-MOF respectively) with different morphologies and electronic structures via one-pot hydrothermal synthesis. We investigated the influence of ligands on the OER activity of NiCo bimetallic MOF electrocatalysts for three different MOFs under industrially relevant conditions (i.e., higher current density and longer stability), Fig. S1.† Surprisingly, the NiCo-t-MOF electrocatalyst showed extraordinary catalytic activity. Without changing the electrolyte, the NiCo-t-MOF catalyst maintained a current density of 100 mA cm−2 for more than 62 hours in 1 M KOH. The superior OER activity of the NiCo-t-MOF likely results from the robust coordination environment, higher surface area, and better hydrophilicity, which improves the electron transfer, active site availability, and interaction with electrolyte leading to enhanced electrocatalytic performance over NiCo-im-MOF and NiCo-tp-MOF. The nitrogen atoms in the imidazole ring can coordinate to metal centers in catalytic materials, forming strong metal–ligand interactions. This can enhance the material's ability to stabilize reaction intermediates or facilitate electron transfer. The symmetrical and rigid structure of terephthalic acid can lead to the formation of highly ordered catalytic sites in metal–organic frameworks, enhancing the material's structural integrity and stability under reaction conditions, whereas the carboxyl groups in trimesic acid can coordinate with metal ions, creating a stable porous structure that enhances the material surface area and porosity, facilitating access to active sites and improving the overall catalytic efficiency. As shown in Fig. 1, trimesic acid has a tricarboxylate structure, whereas terephthalic acid is a dicarboxylate ligand, and 2-methyl imidazole is a nitrogen-based ligand. The additional carboxyl groups in trimesic acid facilitate better interaction with metal sites, improving the electrocatalytic activity by offering more active coordination environments for OER activity. A more interconnected and denser coordination network with Ni and Co metal centres creates stronger metal–ligand interactions and, increases the overall stability of the MOF and expand the electron transfer process.20–22
0.2 M each solution of 2-methyl imidazole, trimesic acid, and terephthalic acid was prepared in DMF, followed by ultrasonication for 10 minutes. Then, 0.05 M nickel nitrate hexahydrate and 0.1 M cobalt nitrate hexahydrate were added to the same solution and ultrasonicated for an additional 30 minutes. The thoroughly mixed solutions were transferred to Teflon-lined autoclaves and subjected to heat treatment at 180 °C for 12 hours. After the heat treatment, the autoclaves were allowed to cool naturally, and the resulting products were collected by centrifugation. The solids obtained were washed with deionized water and ethanol and then dried at 70 °C for 10 hours. The final solid materials were categorised as NiCo-im-MOF, NiCo-t-MOF, and NiCo-tp-MOF based on their coordinated ligands. The synthetic scheme of the catalysts is illustrated in Fig. 2a.
![]() | ||
| Fig. 2 (a) Schematic illustration of preparation, (b) XRD pattern, (c) SEM image, and (d) EDS spectrum of NiCo-t-MOF. | ||
XRD measurement was performed to explore the crystalline structure of the synthesized NiCo-t-MOF, and the pattern is displayed in Fig. 2b. The XRD pattern showed distinct, well-defined peaks revealing excellent crystallinity. Three characteristic peaks at 17.55°, 18.74°, and 26.52° can be attributed to the (220), (111), and (311) crystallographic planes, respectively23 of NiCo-t-MOF. The crystal structure of NiCo-t-MOF matches well with the JCPDS file no. 96-7.2-6689, which shows a monoclinic crystal system with lattice constant values of a, b, c, α, β, and γ as 17.41 Å, 12.95 Å, 6.50 Å, 90.00°, 111.90°, and 90.00° respectively. Along with the mentioned crystalline peaks, there are other significant peaks at 7.29°, 10.9°, and 21.96° which are ascribed to (100), (010), and (101) lattice planes. Comparison of the simulated PXRD pattern from the Cambridge Crystallographic Data Centre (CCDC) with the experimental pattern of the NiCo t-MOF showed comparable peaks with the simulated pattern of Ni-MOF with CCDC no. 802889 as shown in Fig. 2b.24,25 The SEM micrographs of NiCo-t-MOF in Fig. 2c reveal an elongated or rod-like structure with sizes of approximately 1.9 μm to 2.6 μm, as marked by the measurement labels. The images indicate the formation of crystallites, which potentially influence the material surface area and catalytic or electronic properties. The elemental composition by Energy Dispersive X-ray spectroscopy (EDS) shown in Fig. 2d infers the presence of all elements of the NiCo-t-MOF material.
The wide-scan XPS spectrum in Fig. 3 represents distinctive Ni 2p, Co 2p, O 1s, and C 1s peaks. The Ni 2p spectrum showed two significant peaks at 873.28 and 855.48 eV, corresponding to Ni 2p1/2 and Ni 2p3/2 states, respectively, with a spin-energy difference of 17.8 eV. In addition, satellite peaks were noticed at 878.68 and 860.18 eV due to the existence of high-spin Ni2+. Similarly, the Co 2p spectrum displayed peaks at 796.39 and 780.63 eV, inferring the +2 oxidation state of Co with two satellite peaks at 801.37 and 783.27 eV. The XPS data depict a +2 oxidation state for both Co and Ni in NiCo-t-MOF.26,27 The deconvoluted C 1s HR-XPS spectrum showed binding energy maxima at 284.08, 285.58, and 287.87 eV, which are attributed to C–C, C–O, and O–C–O environments, respectively.28 Further, the O 1s deconvoluted spectrum displayed peaks at 530.9 and 532.5 eV, responsible for oxygen bound to metal and –OH species.29
![]() | ||
| Fig. 3 Wide range XPS spectrum of NiCo-t-MOF and high resolution deconvoluted spectra of Ni 2p, Co 2p, C 1s and O 1s. | ||
The N2 adsorption–desorption isotherms were recorded at 77 K to study the porous structure of the NiCo-t-MOF. The adsorption–desorption hysteresis loop presented in Fig. 4 illustrates mesoporous properties of the NiCo-t-MOF material. The synthesized NiCo-t-MOF material had a surface area of 9.6 m2 g−1 and an average pore diameter of about 1.33 nm, which suggests that the micro–mesoporous structure is interlaced. The micro–mesoporous structure of NiCo-t-MOF would facilitate the quick transport of ions in aqueous electrolytes and is advantageous for enhancing the OER activity.
A three-electrode assembly was designed to examine the electrocatalytic OER activity of the NiCo-t-MOF electrocatalyst in 1 M KOH solution.
The cyclic voltammogram (CV), linear sweep voltammetry (LSV) curves and the electrochemical active surface area (ECSA) were recorded with NiCo-t-MOF coated on glassy carbon (GC)/carbon cloth (CC) as the working electrode, the Hg/HgO reference electrode and a graphite rod as the counter electrode. The details of the experimental section are provided in the ESI.† The electrochemical properties of the prepared samples were studied by CV in 1 M KOH by selecting GC and CC as the substrate, Fig. S2.† The CV experiments were performed in the potential window of 1.2 to −1.2 V vs. Hg/HgO at a scan rate of 50 mV s−1. Fig. S2a and b† show clear redox peaks for corresponding metal ions in the NiCo-t-MOF sample.30Fig. 5a compares the OER polarization curves of bare CC, NiCo-t-MOF/CC, and RuO2/CC in 1 M KOH solution. LSV curves were collected at a scan rate of 5 mV s−1 after activating the electrode surface with 20 cycles of cyclic voltammograms (CVs) to minimize the capacitive current and to reach a relatively stable surface. The LSV curve of NiCo-t-MOF/CC showed a clear oxidation peak, caused by the oxidation of low valence metal ions to Ni3+ and Co3+ before the water oxidation. The NiCo-t-MOF/CC exhibited an overpotential of 440 mV to attain a current density of 100 mV cm−2, which is much lower than that of the benchmark OER catalyst RuO2 (581 mV), demonstrating that the fabricated electrocatalyst outperformed the noble metal catalyst even under industrially relevant conditions of higher current density. The low overpotential of NiCo-t-MOF/CC for the OER may be due to the higher catalytic activity sites and larger specific surface area of NiCo-t-MOF, indicating a greater area available for interaction between the active sites and the electrolyte. The formation of oxidised species of Ni3+ (NiOOH) and Co3+ (CoOOH) before the onset of the OER may act as the real active species for the OER activity.31 Also, additional species are involved during the OER as the electrochemical reaction occurs at a greater negative potential than Co2+/Co3+ oxidation and as evidenced by XPS, the smaller overpotential is attained because of the oxidation of both Co2+/Co3+ and Ni2+/Ni3+, which suggests that both Co and Ni ions are involved in facilitating the OER.32 However, the probable detailed OER mechanism is discussed in the ESI.† The overpotentials of commercial RuO2/CC and bare CC at any current density are much higher than that of the NiCo-t-MOF/CC, indicating superior electrocatalytic activity of the fabricated electrode for the OER process.
![]() | ||
| Fig. 5 Electrochemical activity of OER studies (a) polarization curves, (b) Tafel slopes, (c) Cdl values and (d) stability test of the NiCo-t-MOF catalyst. | ||
Furthermore, Fig. 5b represents the Tafel slopes calculated from the linear fits of the potential (V) as a function of the logarithm of current density (mA cm−2) extracted from the LSV measurements, Fig. 5a. The Tafel slope value represents the charge transfer rate at the catalyst surface during the electrochemical OER process. The drop in the Tafel slope for NiCo-t-MOF/CC (81 mV dec−1) indicates faster oxygen evolution kinetics than at RuO2/CC (96 mV dec−1) and bare CC (287 mV dec−1).33 In addition, the faradaic efficiency measured by the drainage method was found to be 92.7% for the OER. The detailed calculation for finding the faradaic efficiency is presented in ESI.† The electrochemical double-layer capacitance (Cdl) is directly proportional to the electrochemical active area (ECSA) of an electrocatalyst. CV was used to study the ECSA in the non-faradaic region, Fig. S3 and S4 (ESI).† The CVs were recorded at various scan rates of 10 to 100 mV s−1. The current density curves were plotted, (ja − jc)/2 versus scan rate, where ja and jc are the anode and cathode current densities, and the linear fitting is used to obtain the Cdl value from CVs.34 As shown in Fig. 5c, NiCo-t-MOF/CC exhibited a Cdl value of 8.64 mFcm−2, indicating that the fabricated electrocatalyst possesses more active sites than RuO2/CC (3.31 mF cm−2). To assess the charge transfer resistance during the electrocatalytic reaction, electrochemical impedance spectroscopy (EIS) was performed. The Nyquist plots of bare CC, NiCo-t-MOF/CC, and RuO2/CC at respective OER overpotentials are shown in Fig. S5 (ESI).† The plot of NiCo-t-MOF/CC suggests a smaller charge transfer resistance (∼1.32 Ω) than those of bare CC (17.45 Ω) and RuO2/CC (1.71 Ω), inferring favourable OER charge transfer kinetics at NiCo-t-MOF/CC. The NiCo-t-MOF/CC catalyst showed excellent stability for more than 62 hours at a current density of 100 mA cm−2 without changing the electrolyte in the electrochemical cell. In comparison with the recently reported transition metal-based alkaline OER electrocatalysts (Table S2†), the NiCo-t-MOF/CC exhibited surprisingly lower overpotential and outstanding stability at a current density of 100 mA cm−2, which is of high relevance for industrial applications.
To understand the decay in the electrocatalytic performance after 62 h of stability studies, the catalyst was withdrawn after 62 h and then characterized by powder XRD, XPS and N2 adsorption–desorption isotherms. As shown in Fig. S6a,† no significant change in the sample morphology was noticed, but the careful observation of the images depicted that the crystalline structure, i.e. rod-type structure (Fig. 2c) of NiCo-t-MOF is slightly reformed as shown in the magnified portion of the SEM image. The magnified SEM image displayed both crystalline and non-crystalline portions. The powder XRD pattern of NiCo-t-MOF after 62 h stability is shown in Fig. S6b,† which revealed that the catalyst has retained some crystallinity along with a broad amorphous peak around 25°. In addition, peaks at 18° and 25.8° were noticed corresponding to nickel oxalate (JCPDS no. 98-019-0774). Furthermore, there were crystalline peaks at 43.7° and 50.9° due to the CC substrate (JCPDS card no. 04-0850)35 which was used to deposit the electrocatalyst for OER studies.
To assess the changes in the composition of the surface elements, XPS analysis of the NiCo-t-MOF sample after 62 hours of stability was also performed, Fig. S6c.† The sample showed peaks corresponding to Ni 2p, Co 2p, O 1s, and C 1s, along with a distinctive peak at 689 eV, which corresponds to F 1s. This can be attributed to the fluoride content of the Nafion binder used to produce the electrode for electrochemical investigations.35 The BET surface area analysis of the NiCo t-MOF sample after 62 hours of stability (Fig. S6d† showed a surface area of 1.24 m2 g−1 and an average pore diameter of about 1.49 nm, which suggested a slight change in the surface area with no significant changes in the average pore diameter.
To investigate the practical applicability of the designed NiCo-t-MOF, an alkaline two-electrode water electrolyser was constructed using commercial 20 wt% Pt–C/CC as the cathode and the as-synthesized catalyst as the anode, shown in Fig. 6a. Interestingly, NiCo-t-MOF/CC||Pt–C/CC exhibited a relatively small cell voltage of 1.54 V to deliver a current density of 10 mA cm−2, which is close to the thermodynamic water splitting energy of 1.23 V and lower than that of the various reported catalyst materials in the literature, shown in Table S3, ESI.† Table S3† presents a few OER electrocatalysts in the full-cell water electrolyser configuration where commercial Pt-based catalysts are used as the cathodic HER electrocatalysts and the designed catalysts are used as the anodic OER electrocatalysts in the two-electrode configuration. Additionally, the NiCo-t-MOF/CC||Pt/C/CC net-constructed water electrolyzer achieved a working voltage of 1.93 V at a higher current density of 100 mA cm−2 and demonstrated reliable usability without any degradation behaviour in the chronopotentiometric curve even after water electrolysis for 32 h (Fig. 6b). Both sides of the electrolytic cell generated visible bubbles vigorously (Fig. 6b) suggesting the viability of the fabricated electrolyser for hydrogen generation in practical applications.
It should be noted that the onset potential observed at less than 1.23 V in Fig. 5a and 6a is not the actual onset potential for the OER. Instead, it reflects the influence of the oxidation peaks associated with Ni and Co species. In alkaline media, it is well documented that transition metal-based electrocatalysts, particularly those involving Ni and Co, undergo surface redox transitions before the OER begins. These redox processes, such as the oxidation of Ni2+ to Ni3+ and Co2+ to Co3+, typically occur in the potential range of 1.2 to 1.4 V vs. RHE and may overlap with the region where the OER starts. In our study, the onset potential observed below 1.23 V is primarily attributed to the Ni/Co oxidation processes and does not represent the actual onset of the OER. These pre-OER redox activities are inherent of the electrocatalyst surface chemistry and significantly influence the voltammetric curves. Because of this interference, accurate reporting of the onset potential or calculating the overpotential at a standard specific current density (e.g., 10 mA cm−2) has become challenging. Therefore, given this overlap, it is not possible to report overpotential values at a current density of 10 mA cm−2 with high accuracy for this system. Instead, we focused on benchmarking the catalyst performance at higher and more distinguishable current density (e.g., 100 mA cm−2) to avoid the ambiguities.
All the results, especially the lower overpotential and long-term stability at higher current density, suggest that the NiCo-t-MOF/CC demonstrates promising potential for practical applications.
Footnote |
| † Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4se01656d |
| This journal is © The Royal Society of Chemistry 2025 |