Introduction

Increasing environmental pollution and the rapid decrease of fossil fuels have diverted the attention of researchers towards the development of new alternate sources of energy [1]. Over the past few decades, many efforts have been made in this regard, and it has been found that hydrogen can be used as a clean, secure, non-polluting, and environment friendly source of energy [2]. Hydrogen can be produced from the hydrogen-containing compounds like fossil fuels and biomass, but they contaminate the environment [3]. Alternately, hydrogen can be effectively produced by photocatalytic water splitting which has no harmful effect upon the environment [4]. The production of hydrogen by water splitting has attracted great attention after first reported by Fujishima and Honda in 1972 [5].

During water splitting, hydrogen and oxygen are produced at the surface of catalyst due to redox reaction [6]. Currently, Ru/Ir-based OER-based catalysts have emerged as the best OER catalysts, but due to their low earth abundant and high cost, they cannot be used at commercial levels [7, 8]. So, the development of highly active OER catalyst based on low-cost earth abundant 3D-element has attracted a huge research attention. Therefore, up until now, more than a hundred different materials have been reported as catalyst for water splitting [9]. However, high electron-hole pair recombination lowers the solar energy conversion and reduces the efficiency of catalysts [10]. Many efforts have been made to solve this problem, but efficiency has not been increased up to satisfactory level [11].

In the recent years, metal organic frameworks (MOFs) have emerged as potential candidate for different applications such as storage [12], separation [13], catalysis [14], and biological imaging [15]. MOFs have unique properties such as high surface area and large number of cavities and channels [16]. In 2010, Garcia and co-workers reported Zr-containing MOFs known as UiO-66 and UiO-66 (NH2) as prominent catalysts for water splitting in the presence of visible light [17]. It has been observed that both UiO-66 and UiO-66 (NH2) have greater catalytic activity towards water splitting when Pt nanoparticles are incorporated within the pores [18]. Since then, many successful efforts have been made for the preparation of composites of MOFs by incorporation of different materials such as metal/metal oxide nanoparticles [19], quantum dots [20], graphene [21], dense and porous silica nanospheres [22], and magnetic beads [23] to get specific properties. In MOFs, the organic ligands serve as antenna to harvest light and activate the metal. Photogenerated electrons produced in MOFs are transferred to nanoparticles and increase the charge separation for efficient photocatalytic activity [24].

Manganese-based oxides (MnOx) nanomaterials have been emerged as efficient OER catalysts due to their unique properties such as well-controlled morphology and electronic state [25, 26]. Manganese oxide (MnO2) nanoparticles, due to their stability, large surface area, and small size [27], are used in redox reactions [28] and catalysis [29]. MOF-5 is an important metal organic framework that consists of [Zn4O]6+ clusters linked by octahedral arrangement of 1,4-benzenedicarboxylate groups to form a cubic porous framework of MOF-5 [30]. In the present work, nanoparticles of MnO2 have incorporated into MOF-5 to form MnO2@MOF-5 composite. Different experimental conditions are tested, and composite materials showed superior OER performance to individual MnO2 particles.

Experimental

Chemicals

The chemicals used for the synthesis of materials were zinc acetate dihydrate, Zn(CH3COO)2.2H2O, 1,4-benzenedicarboxylic acid (H2BDC), triethylamine (TEA), N,N-dimethylformamide (DMF), KMnO4, MnSO4.H2O, and H2O2. All these chemicals were purchased from Merck. These were of analytical grade and used as such without any further purification.

Synthesis of MnO2@MOF-5 composite

Hydrothermal method was used for the preparation of nanoparticles of MnO2 as reported in literature [31]. MnO2@MOF-5 composite was prepared by incorporation of pre-synthesized nanoparticles of MnO2 into MOF-5 during its synthesis. In a typical procedure, a suspension of 10 mg MnO2 nanoparticles was prepared in 20 mL DMF. The suspension was added into 100 mL DMF solution containing 2.7 g Zn(CH3COO)2.2H2O. Then a 50 mL DMF solution containing 0.8 g H2BDC and 2 mL TEA was added dropwise to the mixture under constant magnetic stirring. The whole mixture was stirred at room temperature for 24 h. The brownish precipitates were obtained and collected by centrifugation, washed with distilled water and DMF several times. The resulting product was dried at 50 °C in a vacuum oven for 3 h and activated at 120 °C for 6 h. The dried precipitates were ground and stored for further characterization and photoelectrochemical studies. A sample of pure MOF-5 was also prepared by the same procedure without adding MnO2 nanoparticles.

Oxygen evolution reaction studies

The oxygen evolution reaction studies of synthesized samples were studied in the presence of visible light as well as in the dark at room temperature by cyclic voltammetry (CV), linear sweep voltammetry (LSV), and chronoamperometry using the electrochemical workstation (Autolab PG station 204) in 1.0 M NaOH aqueous electrolyte at various scan rates. The electrochemical measurements were conducted using a three-electrode setup containing Ag/AgCl as reference electrode and Pt-wire as counter electrode. All the applied potentials were converted into reversible hydrogen electrode (RHE) by using following equation

$$ {\mathrm{E}}_{\mathrm{RHE}}={\mathrm{E}}_{\mathrm{Ag}/\mathrm{AgCl}/\mathrm{Sat}.\mathrm{KCl}}+0.059\ \mathrm{pH}+0.197 $$
(1)

LSV was measured at 1 mVs−1 scan rate, and it was used to evaluate Tafel plot according to following equation:

$$ \upeta =\mathrm{a}+\mathrm{b}\ \log\ \mathrm{j} $$
(2)

where ƞ is the overpotential, j is the current density, and b is the Tafel slope. For OER, the overpotential was calculated by the following equation:

$$ \upeta ={\mathrm{E}}_{\mathrm{RHE}}-1.23 $$
(3)

The working electrode was prepared on a nickel foam (NF). For the preparation of working electrode, a piece of NF (1 cm × 1 cm) was cleaned with ethanol and acetone by sonication for 30 min, respectively, washed with distilled water, and dried at room temperature. In total, 10 mg of the prepared sample was added in distilled water to make the slurry, which was then uniformly pasted on the NF and dried at 50 °C for overnight. CV, LSV, and chronoamperometric measurements used the dried NF containing sample as working electrode for the study of water splitting activity. It was observed from SEM images that the catalytic material remains deposited on the surface of Ni foam before and after photoelectrochemical reaction (Fig. 1). It indicated the significant stability of these working electrodes for PEC studies.

Fig. 1
figure 1

Schematic representation of coating on the Ni foam and its use for photoelectrochemical study; inset are the magnified images of blank Ni foam and working electrode before CV studies and after CV studies

Full size image

Characterization

Powder X-ray diffraction (XRD) patterns of pure MOF-5 and MnO2@MOF-5 were recorded on a Shimadzu XRD diffractometer with Cu-Kα radiation (λ = 0.15406 nm) in the range of 2θ between 10° and 80° at scan rate of 5° min−1. Raman spectrometry was carried out by using the Horiba Jobin-Yvon Lab RAM HR800 Raman spectrometer in the range from 100 to 2000 cm−1. The 532-nm solid-state laser was used to avoid florescence and decomposition of samples. The output power was reduced to 10%, and the acquisition time ranged from 5 to 10 min. Fourier transform infrared spectra were obtained by using Nicolet Nexus 870 in range from 4000 to 400 cm−1.The morphology and composition of samples were studied by scanning electron microscopy (SEM) by using the Philips XL30 Environmental SEM attached with Oxford Instrument Inca 500 energy-dispersive X-ray (EDX) spectrometer. The optical properties of the samples were studied by using the Shimadzu UV-2600 UV-visible spectrophotometer at room temperature in the range between 200 and 900 nm.

Results and discussion

Powder XRD patterns of MOF-5 and MnO2@MOF-5 are shown in Fig. 2; both the materials have grown well in crystalline form and intense peaks are observed. The P-XRD pattern of pure MOF-5 matches well with that reported in literature [32, 33]. In the P-XRD pattern of MnO2@MOF-5, the peaks that index to both MnO2 and MOF-5 are observed, and well-defined diffraction peaks at 2θ about 19.18°, 20.47°, and 30.52° correspond to MOF-5 and 13.55°, 18.61°, 28.45°, 38.65°, 41.11°, and 50.13° to MnO2. The diffraction peaks correspond to MnO2 matches with the standard XRD pattern ICSD 44-141. From the P-XRD patterns of MOF-5 and MnO2@MOF-5, it is observed that the host MOF-5 maintains its characteristic reflection pattern and crystallinity.

Fig. 2
figure 2

Powder XRD patterns of MOF-5 and MnO2@MOF-5 in comparison with simulated pattern of MOF-5

Full size image

The Raman spectra of MOF-5 and MnO2@MOF-5 are shown in Fig. S1 (Supplementary information). The Raman spectrum of MOF-5 consists of five strong Raman bands at 635 cm−1, 865 cm−1, 1139 cm−1, 1448 cm−1, and 1615 cm−1, and these are in accordance with the results reported in literature [34]. These Raman bands are due to vibrational modes of benzene rings and vibration modes of carboxylate groups [35]. Similarly, the Raman spectrum of MnO2@MOF-5 also consist of five Raman bands at 635 cm−1, 865 cm−1, 1141 cm−1, 1434 cm−1, and 1615 cm−1, which indicates that MOF-5 shows its dominance and maintains its crystalline structure. In the Raman spectrum of MnO2@MOF-5, slight Raman shifts can be observed at 1141 cm−1 and 1434 cm−1 which may be due to the interaction of MnO2 with [Zn4O]6+ clusters of MOF-5 and some displacement of organic ligand.

These five vibrational modes of MOF-5 and MnO2@MOF-5 are comparable with DFT quantum calculations of MOF-5 as reported previously [36]. The Raman spectra of MnO2@MOF-5 do not contain characteristic peaks of MnO2 because the dominance of host MOF, peaks of MnO2, is masked.

The FTIR spectra of both MOF-5 and MnO2@MOF-5 are shown in Fig. S2. The FTIR spectra of MOF-5 shows two sharp vibrational peaks at 1575 cm−1 and 1370 cm−1 due to symmetric and asymmetric stretching vibration of C-O, whereas MnO2@MOF-5 shows vibrational bands at 1586 cm−1 and 1377 cm−1 due to symmetric and asymmetric stretching vibration of C-O bonded to Zn, respectively. In the FTIR spectra of MOF-5, symmetric stretching vibration of Zn4O appears at 642 cm−1 [37]. A strong peak appears at 530 cm−1 in the FTIR spectrum of MnO2@MOF-5 which may be due to the overlapping of Mn-O vibration with secondary building unit of MOF-5, coordinated [Zn4O]6+ cluster.

The morphology and composition of MOF-5 and MnO2@MOF-5 was studied by scanning electron microscopy and SEM-based EDX, respectively. The SEM images are shown in Fig. 3, which shows that MOF-5 has grown in crystalline form having a rectangular shape (Fig. 3a–b). Figure 3d–e shows the SEM images of MnO2@MOF-5 composite which indicates that MnO2@MOF-5 also have morphologically smooth surface of the crystals. The EDX spectra of MOF-5 and MnO2@MOF-5 are shown in Fig. 3c and f, respectively. Figure 3c shows that the EDX spectrum of MOF-5 contains all the three elements (Zn, C, and O) of MOF-5. Similarly, Fig. 3f shows that the EDX spectrum of MnO2@MOF-5 composite contains all the four elements (Zn, Mn, C, and O) and indicated successful incorporation of MnO2 into MOF-5.

Fig. 3
figure 3

(a–b) SEM images, (c) EDX of MOF-5, and (d–e) SEM images; (f) EDX of MnO2@MOF-5

Full size image

The elemental mapping analysis of MOF-5, as shown in Fig. 4a–d, shows that there is a uniform and homogeneous distribution of elements, and it has grown in a rectangular shape. Figure 4a–c show elemental mapping of MOF-5 for Zn, C, and O respectively and Fig. 4d is the mix mapping of the metal organic framework, MOF-5, which indicated that all the basic elements are uniformly and homogenously distributed in the sample. Similarly, Fig. 4e–g show the elemental mapping of MnO2@MOF-5 for Mn, Zn, and O respectively and Fig. 4h is the mix mapping of MnO2@MOF-5, which indicated that nanoparticles of MnO2 has been successfully incorporated within MOF-5 as it can be observed that Mn is uniformly distributed in the composite.

Fig. 4
figure 4

Elemental mappings of MOF-5 and MnO2@MOF-5

Full size image

The optical properties of synthesized samples were studied by UV-visible absorption spectrophotometry. As shown in Fig. S3, MOF-5 has maximum absorption in UV-region at 265 nm due to π→π* electronic transition of 1,4-BDC linkers and with no significant absorption in visible light. After incorporation of MnO2 into MOF-5, its λmax shifts from 265 to 275 nm. Furthermore, a new absorption peak having λmax around 400 nm has introduced with increase in absorption in the visible region. Such a modification increases the charge separation because photogenerated electron efficiently transfer from organic ligand 1,4-BDC to [Zn4O]6+ clusters of MOF-5 and then to MnO2, similar to UIO-66(NH2) and Ti-MOF-NH2. The increase of charge separation has improved the photoelectrochemical activity. The band gap of a semiconductor can be calculated from Tauc plot by linear extrapolation of absorption edge by using the following equation:

$$ \upalpha \mathrm{h}\upnu \propto {\left(\mathrm{h}\upnu -{\mathrm{E}}_{\mathrm{g}}\right)}^{0.5} $$
(4)

where α is the absorption coefficient, h is the Planck’s constant, and ν is the wavenumber.

The band gap of MOF-5 is determined to be 3.70 eV, which is in accordance with the literature reported previously [38], while the band gap of MnO2@MOF-5 is 2.47 eV, which indicated that the incorporation of MnO2 nanoparticles brings the band gap of MOF-5 into the visible region. Hence, it increases the absorption of visible light and improves the photoelectrochemical activity of MnO2@MOF-5. The mechanism of increase of charge separations is proposed in a schematic representation as shown in Fig. 5.

Fig. 5
figure 5

Schematic representation of increased charge separation and water splitting by MnO2@MOF-5

Full size image

Photoelectrochemical OER analysis

Firstly, photoelectrochemical studies towards OER are determined by cyclic voltammetry. CV curves of these electrodes are recorded using 1.0 M NaOH electrolyte both in the presence of dark and under visible light at different scan rates (10, 20, 30, 40, 50, and 100mVs−1) for MOF-5/NF and (5, 10, 20, 30, 40, 50, 60, 70, 80, 90, and 100mVs−1) for MnO2/NF and MnO2@MOF-5/NF. It is found that all the working electrodes delivered almost zero current density in the dark due to the absence of OER activity. Figure 6a shows that MOF-5/NF generates no significant current density at various scan rates within RHE potential range of 0.2 to 1.2 V. Figure 6b and c showed CV curves for MnO2/NF and MnO2@MOF-5/NF electrodes under RHE potential range 1.0 to 1.6 V, in the dark as well as in visible light. It is investigated from the CV curves of MnO2/NF and MnO2@MOF-5/NF that in the presence of visible light, a prominent increase in current density is observed, due to OER activity. However, pre-OER oxidation-reduction peaks appeared in the CV curves of MnO2/NF and MnO2@MOF-5/NF due to redox reaction of Ni(II)/Ni(III) [39]. In MnO2/NF, the oxidation and reduction peaks appeared in the range between 1.44 and 1.50 V and 1.38 and 1.34 V, respectively. The CV curves of MnO2@MOF-5/NF electrode showed that there are two oxidation peaks at various scan rates (10 to 100mVs−1), first between 1.35 and 1.41V and then between 1.40 and 1.47 V, and only one reduction peak is observed between 1.28 and 1.20 V. It is revealed from the comparison of CV curves of MOF-5/NF, MnO2/NF, and MnO2@MOF-5/NF that MnO2@MOF-5/NF has higher current density as compared to pure MOF-5/NF and MnO2/NF due to the hetero-junction formation between MnO2/NF and the central metallic cluster of MOF-5. Thus, synergistic effect and hetero-junction formation played a significant role to enhance the OER activity of MnO2@MOF-5/NF as compared to all other synthesized samples. The inset of Fig. 6b and c represented the anodic and cathodic peak current density with respect to the scan rate of MnO2/NF and MnO2@MOF-5/NF, respectively.

Fig. 6
figure 6

CV curves of MOF-5/NF (a) and MnO2/NF (b); inset is peak current density vs scan rate and MnO2@MOF-5/NF (c). Inset is peak current density vs scan rate at different scan rates

Full size image

The catalytic activities of MOF-5/NF, MnO2/NF, and MnO2@MOF-5/NF electrodes are further evaluated by linear sweep voltammetry (LSV) in the presence of visible light in 1 M NaOH electrolyte, as shown in Fig. 7a. As in CV, the anodic oxidative peak is observed in LSV of both MnO2/NF and MnO2@MOF-5/NF due to oxidation of Ni2+→Ni3+ from the Ni foam. The LSV curve of MOF-5/NF electrode indicated that there is no significant generation of current density; even at very high overpotential, it generated just 0.46 mAcm−2 current density. For a comparison between MnO2/NF and MnO2@MOF-5/NF, the overpotential required to achieve 2 mAcm−2current density is considered. It is observed that MnO2@MOF-5/NF required a low overpotential of 284 mV to achieve 2 mAcm−2 current density, whereas MnO2/NF required 324 mV overpotential for 2 mAcm−2. It is observed that MnO2@MOF-5/NF delivered the benchmark of 10 mAcm−2 at an overpotential of 324 mV, which could be comparable and even less than some of reported Mn-based and other 3d transition metal–based catalyst for OER, represented in Table 1. Furthermore, to understand the kinetics of MnO2@MOF-5/NF towards OER, a Tafel plot is derived from LSV. The calculated Tafel slope value for MnO2@MOF-5/NF is just 71 mVdec−1, which is lower than MnO2/NF (157 mVdec−1) and previously reported Mn-based materials such as MnO2/Ni-Co carbonate precursor/NF (95 mVdec−1) and MnO2/NiCo2O4/NF (139 mVdec−1) [40]. The lower Tafel slope value for MnO2@MOF-5/NF indicated that it has more favorable electron transferred and improved catalytic activity as compared to MnO2/NF, and its rate-determining step for OER was the first electron transferred as shown below,

$$ \mathrm{M}+{\mathrm{OH}}^{-}\to \mathrm{M}\mathrm{OH}+{\mathrm{e}}^{-} $$
(5)

where M is the catalytic active site [37, 38].

Fig. 7
figure 7

a LSV curves of MOF-5/NF, MnO2/NF, and MnO2@MOF-5/NF. b Tafel plot for MnO2/NF and MnO2@MOF-5/NF

Full size image
Table 1 The comparison of OER activity of MnO2@MOF-5/NF with previously reported Mn-based and other 3D transition metal–based OER catalyst
Full size table

From the comparison of LSV curves, it can be found that as a result of the incorporation of MnO2 into MOF-5 and due to synergistic effect, MnO2@MOF-5/NF has improved OER catalytic activity over both pure MOF-5/NF and MnO2/NF.

The stability of the synthesized materials during OER activity is evaluated by chronoamperometric studies (Fig. 8). It is observed that MnO2@MOF-5/NF generated constant current density at constant applied voltage of 1.0 V for 6000 s in the presence of visible light. The stability of MnO2/NF was also constant during this time, but it produced less current density.

Fig. 8
figure 8

Chronoamperometric measurements of MnO2/NF and MnO2@MOF-5/NF at 1.0 V applied potential

Full size image

Furthermore, the stability of MnO2/NF and MnO2@MOF-5/NF is studied by continuous CV sweeps in 1.0 M NaOH at a scan rate of 100 mV/s for 100 cycles. The CV curves revealed negligible degradation after 100 cycles of CV scanning, which confirmed the stability and durability of these electrodes. Figure S4 (a and b) represented the 1st and 100th cycles of CV curves at 100 mV/s of these electrodes, and it revealed almost the same curves with negligible difference.

Conclusions

MnO2@MOF-5 composite has been synthesized successfully by in situ incorporation of pre-synthesized MnO2 nanoparticles into MOF-5 and used as efficient OER catalyst. The MnO2@MOF-5 composite has shown better OER activity as compared to MnO2 and MOF-5. MnO2@MOF-5 can expose active sites more effectively due to the incorporation of MnO2 nanoparticles and synergistic effect. The incorporation of MnO2 nanoparticles leads a strong electron interaction among 1,4-BDC (1,4-benzenedicarbxylate), [Zn4O]6+ clusters, and MnO2 and further optimizes the charge transfer; so charge separation increased and electron-hole pair recombination decreased. Furthermore, binder’s free formation of MnO2@MOF-5/NF enhanced the electrical conductivity. The CV, LSV, and chronoamperometric results show that MnO2@MOF-5 is an efficient OER catalyst and has more stability. This study will encourage for designing more versatile efficient materials by nanoparticle incorporation into MOFs for water splitting, fuel cells, supercapacitors, and batteries.