Enhanced hydrogen generation from NaBH4 hydrolysis over green-synthesized Co₃O₄ catalysts through control of parsley ratio and calcination temperature

Characterization of Co 3 O 4 catalysts

Phase identification

The XRD patterns of the calcined catalysts at various ratios and calcination temperatures confirm the face-centered cubic structure and verify the crystalline nature of the resulting NPs. The XRD pattern of Co 3 O 4 NPs exhibits eight characteristic peaks, as shown in Fig. 1(a), at 2θ = 19.12°, 31.27°, 36.79°, 41.75°, 44.85°, 55.72°, 59.65°, and 65.21°. The observed diffraction peaks correspond to the (111), (220), (311), (222), (400), (422), (511), and (440) planes of cubic Co 3 O 4 NPs, which are well-matched with JCPDS card No. 03-065-3103 (a lattice constant of a = 8.056 Å and space group Fd-3 m). The diffractogram of the prepared catalysts highlights the absence of any phase devoid of other reflections from carbon species or cobalt compounds such as Co(OH) 2 or CoO. These findings confirm the high purity of the spinel Co 3 O 4 structure. Similarly, the same planes of Co 3 O 4 NPs are maintained in the synthesis procedures with varying ratios of Solanum lycopersicum leaf extract and cobalt acetate hexahydrate30, as reported by Tadesse Ayanie et al. The peak intensity became sharper and more intense as the calcination temperature increased, and the cubic phase of Co 3 O 4 NPs remained stable up to 900 °C, indicating thermal stability. In this context, Zhou et al. synthesized Co 3 O 4 NPs at various calcination temperatures from 600 to 1100 °C and demonstrated the superior thermal stability26. Notably, higher calcination temperatures resulted in superior crystallinity of the synthesized Co 3 O 4 NPs, as seen in the PWB600 and PWB900 catalysts compared to other catalysts. Obviously, the (311) plane in the XRD patterns of PWB600 and PWB900 samples showed a narrow peak, whereas catalysts calcined at lower temperatures exhibited broader peaks for the same plane. Li et al. reported that the calcination process up to 700 °C leads to the presence of mixed phases (CoO and Co 3 O 4 ) due to oxygen vacancies31.

Fig. 1 Full size image X-ray diffraction patterns (a) and FT-IR spectra of the various synthesized Co 3 O 4 NPs (b).

Specifically, the catalysts PWB600 and PWB900 displayed magnetic properties resulting from the in-situ reduction during the catalytic hydrolysis of NaBH 4 . From these observations, the XRD measurements were performed before and after the in-situ reduction of the PWB600 catalyst, as depicted in Fig. S1. Notably, the XRD pattern for PWB600 remained consistent before and after the reaction, although peak intensities decreased post-reaction without changing the chemical structure. The catalyst has a single detectable phase without any mixture of other phases. The Co x B phase was difficult to detect in the XRD diffraction because of its amorphous nature, as noted by Wang et al.6. Similarly, Simagina et al. discussed that the amorphous Co x B phase can interfere with the cubic crystalline structure of Co 3 O 4 NPs32. Obviously, the crystal structure of the catalyst before the surface reduction process is retained after the process. XRD patterns were performed to elucidate this difference with a range of 2θ scales from 40° to 50°; the presence of peaks corresponding to the boron element implies changes in the atomic arrangement in the crystal structure. The phase of Co x B is well-matched with JCPDS card No. 00-039-1107. Based on the XRD results, it can be concluded that varying ratios and pre-treatment temperatures do not significantly affect the crystal structure of Co 3 O 4 NPs. Therefore, catalytic testing is necessary to evaluate their impact on performance.

FT-IR spectra of the synthesized Co 3 O 4 catalysts were recorded over the wavelength range of 4000 –400 cm− 1. As depicted in Fig. 1(b), most absorption peaks for all catalysts are observed at similar wavelengths. Consistent with the XRD findings, the FT-IR data affirm the spinel structure of the prepared Co 3 O 4 catalysts, evidenced by two peaks in the fingerprint region corresponding to the metal ions. Notably, two sharp absorption peaks at 668 –664 cm− 1 and 570 − 564 cm− 1 indicate Co 3 O 4 spinel stretching vibrations. The peak at approximately 666 cm− 1 is attributable to Co2+ vibrations (Co-O) at tetrahedral sites, whereas the peak around 567 cm− 1 corresponds to Co3+ vibrations (O-Co-O) at octahedral sites within the Co 3 O 4 lattice21,22. The weak absorptions observed at 3824 –3732 cm− 1 may be associated with O-OH vibrations related to PWB400-150, PWB600, and PWB900 samples. This observation suggests the presence of surface cobalt hydroxycarbonate, which formed through interactions between CO 2 , H 2 O (from the atmosphere), and the Co 3 O 4 spinel. In this context, Soltani and Zabihi reported about the stronger hydroxyl groups corresponding to the active centers on the catalysts surface for the immobilization of cobalt33. Upon examining the FT-IR spectrum of the PWB400-100 catalyst, broad peaks at 3417 cm− 1 are indicative of hydroxyl (-OH) stretching vibrations, corresponding to adsorbed H 2 O molecules21,22,34,35,36. Additionally, two weak absorptions near 2925 cm− 1 and 2856 cm− 1 may correspond to C-H stretching (symmetric and asymmetric), respectively17,35. The peak at 1635 cm− 1 may be linked to the stretching mode of the -COO group or the bending vibration of H 2 O21,22,35. As reported in33, the existence of hydroxyl (OH) and carboxyl (COO) groups in the FT-IR spectra of the synthesized catalysts facilitates well-dispersed formation of NPs, enhancing of catalytic performance for hydrogen production. The peaks at 1384 cm− 1 and 1117 cm− 1 are associated with surface carbonate vibrations37. No detectable peaks related to functional groups are observed in the FT-IR spectra of the synthesized catalysts derived from different ratios. Following calcination treatment at up to 900 °C, additional carbonaceous peaks emerge in the spectra. The spectrum of the PWB300 catalyst shows three weak absorption peaks within the range of 1432 –1257 cm− 1, which diminish at higher calcination temperatures. Peaks observed approximately at 887 –835 cm− 1 in the pre-treated catalysts are characteristic of aromatic C-H out-of-plane bending vibrations. Comparison of the FT-IR spectra reveals two prominent peaks at 1050 cm− 1 and 1008 –1004 cm− 1 attributed to cochineal molecules (surface impurity)38 in the PWB600 and PWB900 catalysts. Overall, carbonaceous surface species are evident across all FT-IR spectra, originating from residual carbon compounds formed during the calcination of the green source, which were not detectable via XRD analysis; however, their presence was confirmed by XPS analysis (vide infra).

Morphology inspection

FE-SEM images reveal the morphology of the synthesized catalysts and the impact of varying ratios of parsley to metal source and the calcination treatment. Contents of plants have phytochemicals which play an essential role in stabilizing and controlling the size, agglomeration, and morphology of metal oxide nanoparticles. Inspection of the samples’ morphology from different ratios, the lower weight plant ratio (PWB400-15) results in larger spherical particle aggregates with fewer voids, attributable to inadequate capping and stabilization, as depicted in Fig. 2(a). Conversely, the optimal ratio (PWB400-100) yields capsule-shaped particles with increased aggregation and voids, as shown in Fig. 2(b), which correlates with higher catalytic activity (vide infra) due to the balanced presence of bioactive compounds such as flavonoids, polyphenols, and alkaloids. Morphological similarities are observed across samples derived from various ratios. For instance, the PWB400-150 catalyst exhibits small, uniform spherical nanoparticles owing to increased bioactive compounds that act as reducing and stabilizing agents, thereby resulting in lower catalytic performance. Flavonoids and polyphenols are crucial phytochemicals that cause the reduction of metal ions due to their ease of oxidation39. Liu et al.40 revealed that phytochemicals surrounding NPs could provide spatial resistance and electrostatic repulsion, enhancing NP stability. Moreover, increasing the concentration of these compounds raises electron donation. Hosseinzadeh et al.41 investigated how phytochemical compounds contribute to preventing aggregation during NP formation. Concerning the calcination effects, Fig. 2(d) illustrates that the PWB300 catalyst has uniform spherical nanoparticles with agglomeration, and their morphology is consistent across catalysts prepared from different ratios. Whereas the precursor is calcined up to 400 °C, the morphology changes of PWB600 represent increased voids on the surface and larger particles interconnected beyond those of others, as shown in Fig. 2(e). Inspection of their image at a scale of 100 nm reveals triangular-shaped nanoparticles, which represent a unique feature among all synthesized catalysts. Moreover, Fig. 2(f) illustrates that the features of the PWB900 catalyst display more aggregated, larger capsule-like particles. The dense clustering of PWB600 and PWB900 contributes to a reduced surface area, a consequence of the elevated calcination temperature. Overall, the FE-SEM images highlight the significant influence of optimal ratio selection and calcination temperature in controlling the morphological enhancement of the catalysts.

Fig. 2 Full size image FE-SEM images of Co 3 O 4 NPs in various ratios and calcination treatment degrees, PWB400-15 (a), PWB400-100 (b), PWB400-150 (c), PWB300 (d), PWB600 (e), and PWB900 (f).

EDX analysis of annealed Co 3 O 4 catalysts

For the annealed catalysts at various calcination temperatures (300, 600, and 900 °C), Energy Dispersive X-ray (EDX) spectroscopy was employed to investigate the elemental composition. As depicted in Fig. 3, the EDX spectra of the pre-treatment Co 3 O 4 catalysts distinctly revealed the presence of cobalt, oxygen, and carbon elements. According to Fig. 3(a) and (c), the cobalt content increased from 46.9% to 69.5%, while the oxygen content declined from 32.9% to 23.4% for PWB300 and PWB900, respectively. The observed reduction in oxygen content at higher calcination temperatures is attributable to the formation of oxygen vacancies within the spinel structure27. Meanwhile, the carbon content decreased from 20.2% in PWB300 to 5.1% in PWB900 as the annealing temperature increased. Consistent with FT-IR and XRD data, these findings indicate fewer carbon species at elevated temperatures and support the higher crystallinity observed in PWB600 and PWB900 catalysts. The atomic ratios of Co to O were ordered as 1.43%, 2.86%, and 2.97% with rising calcination temperatures, confirming the enhanced purity of the synthesized catalysts and aligning with their catalytic activity order. Prior studies have demonstrated that the higher purity of prepared Co 3 O 4 catalysts correlates with an increasing atomic ratio of Co to O42,43. PWB600 was selected after its use in the catalytic reaction to verify the in-situ reduction of Co 3 O 4 NPs to Co x B. This selection was performed owing to the magnetic properties observed during the hydrolysis reaction, indicating in-situ reduction. As shown in Fig. 3(d), the EDX spectrum of PWB600 confirmed this reduction through the presence of boron. The atomic percentages of the composed elements were 52.1% Co, 25.8% O, 52.1% C, and 5.6% B. Post in-situ reduction, the decrease in active centers (cobalt) on the catalyst surface was observed, due to their consumption during NaBH 4 hydrolysis, as reported previously44. The detection of boron in the EDX spectrum may be challenging via XRD analysis because the detection size limit is approximately 5 nm.

Fig. 3 Full size image EDX spectra of PWB300 (a), PWB600 (b), PWB900 (c), PWB600 after in-situ reduction (d).

N 2 adsorption-desorption analysis

The textural properties of the formed Co 3 O 4 NPs, such as surface area, pore size, and pore volume, were analyzed through N 2 physisorption measurements. According to IUPAC classification, the sorption isotherm curves of all annealed samples at different calcination temperatures display a type IV isotherm with a hysteresis loop of type H3. This indicates that they are mesoporous materials, with nanoparticle sizes ranging between 2 and 50 nm within the catalysts. As shown in Fig. S2(a), the adsorbed volume of all catalysts declined with higher calcination temperature from 94 to 14 cm3g− 1, from PWB300 to PWB900. Conversely, the samples (PWB300 and PWB400) annealed at a lower temperature exhibited higher adsorbed volumes, which implies a larger surface area among other catalysts. All isotherms revealed increased adsorption with rising relative pressure (P/P o ) until reached 1 and the monolayer formation occurred at the initial part with a lower P/P o , followed by multilayer formation at a higher P/P o . The desorption isotherm indicated capillary condensation around 0.1 < P/P o < 0.95, implying higher textural porosity of the Co 3 O 4 NPs, as noted in45. In this context, the hysteresis loop (H3) at P/P o = 0.1 signifies the presence of plate-like particles forming slit-like pores (Fig. S2(a)). The specific surface area (S BET ) was measured using the BET method, yielding values of 100, 88.2, 15.3, and 15.6 m2g− 1 for catalysts calcined at 300, 400, 600, and 900 °C, respectively. The relationship between surface area and pore volume is evident from FE-SEM images. The PWB300 and PWB400-100 catalysts exhibit a spherical shape with high agglomeration and fewer voids, resulting in higher S BET and pore volume. Conversely, PWB600 and PWB900 catalysts possess a larger shape with more voids, which leads to lower S BET and pore volume. Compared to previous studies, Dang et al. prepared hierarchical cobalt oxide with lower S BET values of 33.05 and 66.61 m2g− 1 for 2D-SCO and 3D-CCO, respectively46. At the calcination temperature (300 °C), Hu et al. synthesized two samples, Co-p as nanoplates and Co-f as nanoflowers, with S BET values of 51.52 and 57.09 m2g− 1, respectively47. Green-synthesized Co 3 O 4 /C nanocomposite using basil leaves extract by Abu-Zied et al., where the precursor was calcined at 400 °C with S BET of 19 m2g− 1, which is about four times lower than PWB400-10021. Additionally, all these abstract S BET values from the synthesized catalysts are significantly higher than those reported for the green-synthesized Co 3 O 4 NPs using jasmine flower extract via the combustion method and subsequent calcination at 300 °C (6 m2g− 1) and 500 °C (2 m2g− 1)22. Interestingly, these findings highlight the significant role of the green precursor (parsley extract) in controlling the texture of Co 3 O 4 NPs. As depicted in Fig. S2(b), the particle size distribution of Co 3 O 4 NPs was determined using the DFT method. Inspection of the obtained results from the DFT method of the formed catalysts reveals poly-disperse distribution. Notably, the plot exhibits similarities as demonstrated in Fig. S2(b), where the PWB300 and PWB900 samples are grouped, and the PWB400 and PWB600 samples are similarly associated. Obviously, there is a fixed sharp peak in all synthesized catalysts at approximately 30 nm and another around 13 nm, except PWB900, where all these peaks lie in the mesoporous range (2–50 nm). Additionally, a broader peak appears in the macroporous region, with significant differences among samples; notably, PWB400-100 shows a broader peak than the others. For PWB300 and PWB900, the pore size distributions show three maxima of pore widths at approximately 78, 116, and 185 nm, which is in the vicinity of the macroporous region (50 nm < NP size). All texture properties of the synthesized Co 3 O 4 catalysts are tabulated in Table 1, including the BET surface area, pore volume, and pore size.

Table 1 Textural data are abstracted from N 2 adsorption data for PWB300, PWB400, PWB600, and PWB900. Full size table

XPS analysis

The fundamental components of the green-synthesized Co 3 O 4 NPs were analyzed using X-ray photoelectron spectroscopy (XPS). The XPS survey spectra of Co 3 O 4 catalysts (Fig. 4(a), 5(a)) confirm the presence of cobalt (Co 2p), oxygen (O 1s), and carbon (C 1s) elements, indicative of Co 3 O 4 core formation. The detection of C 1s in the survey spectra can be attributed to phytochemicals derived from parsley leaf extract adhering to the surface of the Co 3 O 4 NPs, serving as stabilizing agents. Table 2 displays their binding energies and atomic percentages of the constituent elements observed in the survey spectra.

Fig. 4 Full size image XPS data of the prepared catalysts by different ratios of parsley extract: survey (a), Co2p (b), O1s (c), and C1s (d) fitted spectra.

Table 2 XPS surface binding energies and atomic percentages of the elements detected in the surface of the synthesized catalysts. Full size table

In the core-level analysis of Co 2p, the XPS spectra were deconvoluted into Co 2p 3/2 at lower BE and Co 2p 1/2 at higher BE, with a splitting of approximately 15.2 eV, as shown in Fig. 4(b) and 5(b). The spectrum reveals two valence states of cobalt in the synthesized catalysts for each spin–orbit splitting of Co 2p 3/2 and Co 2p 1/2 , with Co3+ located at lower BEs and Co2+ at higher BEs; the additional peaks correspond to shake-up satellites4,6,22,26. For instance, examining the four peaks of Co 2p 3/2 in the PWB400-15 catalyst, the peaks at 779.95 eV and 781.45 eV are attributed to Co3+ and Co2+ ions, respectively. The observed peaks at 785.92 eV and 789.82 eV are associated with satellite features for Co 2p 3/2 . The fitting of the Co 2p 1/2 spectral level exhibits similar characteristics, with peaks at 794.93 eV and 796.53 eV assigned to Co3+ and Co2+ ions. The final two peaks at higher BEs are probably satellite peaks related to cobalt ions in the Co 2p 1/2 level. These spectral features are observed consistently across all prepared catalysts. According to the summarized XPS data in Table S1, the PWB400-100 catalyst exhibits a slight shift of binding energies toward higher values, correlating with enhanced catalytic performance. Conversely, catalysts subjected to pre-treatment show a shift toward lower BEs, as detailed in Table S2. Furthermore, the Co2+/Co3+ ratio in the synthesized catalysts decreases from 1.46 to 0.78 with increasing amounts of green source. Nevertheless, this ratio of the annealed catalysts increases to 1.3, 2.42, and 3.41 at higher calcination temperatures. These findings verify the superior catalytic activity of the PWB900 catalyst, attributed to its higher Co2+ content, which enhances its potential to facilitate in-situ reduction reactions. Moreover, the Co2+ levels of PWB900 observed at higher BEs promote electron transfer from Co3+ to Co2+ during catalytic reactions or modify the cobalt’s electronic configuration because of oxygen defects6,26,46.

Fig. 5 Full size image XPS data of the prepared catalysts by various temperatures of calcination treatment: survey (a), Co2p (b), O1s (c), and C1s (d) fitted spectra.

Two peaks were identified in the O 1s XPS spectra, indicating the presence of lattice oxygen (O 2 ), oxygen defect (OH), and an additional peak corresponding to surface oxygen (H 2 O)4,46 observed in the PWB400-100 catalyst. The high-resolution spectrum of O 1s (Fig. 4(c)) displays two oxygen species on the surface of PWB400-15, which are located at BEs of 529.92 eV and 531.19 eV. These peaks are indexed to lattice oxygen (O 2 ) and oxygen defect (OH)6,26, respectively; similar peaks with shifted BEs are observed in other catalysts, as shown in Fig. 5(c). However, the O 1s spectrum of the PWB400-100 catalyst is composed of three surface species, as depicted in Fig. 4(c). The first two peaks at BEs of 531.01 eV and 532.46 eV can be ascribed to lattice oxygen (O lattice ) and surface hydroxyl (O defect ), respectively. The third peak at 534.37 eV is attributed to adsorbed oxygen from water molecules on the catalyst surface. The positive shift in O lattice and O defect of the PWB400-100 catalyst, compared to other catalysts (PWB400-15 and PWB400-150), is due to its higher efficiency resulting from the increased generation of surface oxygen vacancies46. As the proportion of green source increases, the ratio of O lattice to O defect exhibits an increase, and subsequently decreases to 0.553, 0.548, and 1.098, while the ratios for the annealed catalysts initially increase and then decline to 0.54, 0.66, and 0.60, corresponding to PWB300, PWB600, and PWB900, respectively.

The core fitting of the C 1s spectra in PWB400-15 (Fig. 4(d)) reveals three peaks at 284.57 eV (C-C/C-H)17, 286.1 eV (C-O/ C = O)17,22, and 288.38 eV (O-C = O)22,48; these peaks are similar across other catalysts. These findings are consistent with IR and EDX spectra that confirm the presence of carbonaceous species. The results also validate the high purity of the green-synthesized Co 3 O 4 NPs. As the calcination temperature rises, the atomic % of C = O peaks decrease to 22.99%, 13.19%, and 6.71%, corresponding to PWB300, PWB600, and PWB900. The atomic percentages and binding energies (BEs) of the synthesized catalysts are listed in Table S1, S2. The decline in atomic % of C = O may be due to higher annealing temperatures, as supported by XRD, FT-IR, and EDX analyses. Overall, the atomic percentage of carbon in all prepared catalysts correlates well with the observed orderliness of catalytic activity, as detailed in Table 2.

VSM analysis

As depicted in Fig. 6, the M-H hysteresis loops of synthesized Co 3 O 4 NPs at various calcination temperatures. Obviously, the variations in the shapes of the hysteresis loops among the prepared catalysts result in differences in their magnetic properties, as presented in Table 3. The PWB300 sample shows a narrow hysteresis loop, whereas the PWB600 and PWB900 samples display wider loops. These findings reveal the weak ferromagnetism of PWB300 and the ferromagnetic behavior of PWB600 and PWB900; however, the bulk Co 3 O 4 NPs are antiferromagnetic26,32,49. Magnetic parameters such as coercivity (H c ), saturation magnetization (M s ), and remanent magnetization (M r ) were measured utilizing VSM and are detailed in Table 3. According to the VSM results summarized in Table 3, the catalysts obtained in this study exhibit higher values for all measured magnetic properties (M s , M r, and H c ) compared to those synthesized by Goudarzi and Salavati-Niasari38. The selection of these samples (PWB600 and PWB900) is due to the magnetic response observed after the hydrolysis, indicating in-situ reduction of Co 3 O 4 to Co x B. This indication was verified by VSM analysis performed at room temperature for annealed catalysts. The PWB300 sample exhibits lower M s and M r values (0.58 emu.g− 1 and 0.01 emu.g− 1), confirming its weak ferromagnetic nature as shown during the catalytic reaction. This behavior may relate to finite-size effects or uncompensated surface spins. Conversely, PWB600 shows higher M s and M r values (2.96 emu.g− 1 and 0.65 emu.g− 1), while PWB900 exhibits 1.05 emu.g− 1 and 0.21 emu.g− 1, respectively. The reason is due to the higher ratio of Co3+ ions in PWB600, which are reduced during hydrolysis to Co2+ ions, contributing to the presence of magnetic properties in Co x B, with more final Co2+ ions leading to a higher ratio of M s /M r .

Fig. 6 Full size image VSM analysis revealed M-H hysteresis of ferromagnetic Co 3 O 4 catalysts after hydrolysis reaction at room temperature: PWB300, PWB600, and PWB900.

Table 3 Magnetic properties (saturation magnetization, remanence, and coercivity) of pre-treatment ferromagnetic Co 3 O 4 catalysts. Full size table

These higher values confirm the ferromagnetic properties of Co 3 O 4 catalysts. However, the bulk Co 3 O 4 displays zero net magnetization resulting from the ion exchange of cobalt ions, where occupying in tetrahedral and octahedral sites, leading to complete compensation of sublattice magnetizations50. In nanomaterials, magnetic properties depend on the crystallinity, size, and shape of nanoparticles.

Catalytic hydrolysis of NaBH 4

Evaluation of various synthesized Co 3 O 4 catalysts

This study investigated the impact of green-synthesized Co 3 O 4 catalysts by manipulating the ratios of the green source and calcination temperatures. These factors play a crucial role in accelerating the hydrolysis process of NaBH 4 . Experiments were conducted under consistent conditions during the hydrolysis reaction, involving 10 mg of catalyst and 150 mg of NaBH 4 (0.75%) in 20 ml of distilled water. Hydrogen gas evolved during hydrolysis was examined at reaction temperatures of 30, 35, 40, and 45 °C. As depicted in Fig. 7 , the reaction rate increases with elevated temperature due to higher energetic molecular collisions between hydrogen molecules and the catalyst’s active sites. Initially, the influence of the cobalt source-to-parsley leaves ratio on hydrogen generation rate (HGR) was assessed during hydrolysis reactions. Catalyst performance is enhanced with increasing ratio of parsley leaves until a very high concentration is reached, with PWB400-100 manifesting the superior activity. Subsequently, a decline in performance was observed for PWB400-15 and PWB400-150, as depicted in Fig. 7. This trend indicates that biocompounds in the green source act as stabilizers and capping agents, enhancing catalytic activity at optimal ratios but inhibiting it at higher concentrations. Moreover, the extract ratio affects surface area and active site availability, impacting catalytic efficiency. The moderate HGR in PWB400-15 is due to insufficient parsley plant content, while the optimal plant extract concentration (PWB400-100) enhances NP yield by facilitating the complete reduction of Co2+ ions. An optimized extract ratio results in the formation of well-dispersed, small NPs with high surface energy, thereby improving catalytic performance. However, the PWB400-150 catalyst, with its higher parsley leaves content, exhibited lower catalytic efficiency during hydrolysis. For instance, the reaction completion times were 33, 46, and 64 min, corresponding to hydrogen volumes of 294, 312, and 330 ml, with HGR values of 869, 660, and 497 ml.min− 1.g− 1 at 30 °C for PWB400-100, PWB400-15, and PWB400-150 catalysts, respectively, as shown in Fig. 7(a). Arrhenius plots revealed lower activation energy for PWB400-100 (79.15 kJ.mol− 1) compared to PWB400-15 and PWB400-150 (92.3 and 97.26 kJ.mol− 1). These results demonstrate a rapid and efficient hydrogen production process, highlighting the effectiveness of the PWB400-100 catalyst for hydrogen evolution from NaBH 4 hydrolysis.

Fig. 7 Full size image Effect of various ratios in the synthesized catalysts on hydrogen production during NaBH 4 hydrolysis at different reaction temperatures: 30 °C (a), 35 °C (b), 40 °C (c), and 45 °C (d).

The second part of the study evaluated HGRs from the hydrolysis process and optimized the calcination temperature of the most effective catalyst. Post-annealing was performed at various temperatures (300, 600, and 900 °C). An observable trend indicates that increasing the calcination temperature generally improved performance, as shown in Fig. 8. The Vt-plot conspicuously shows two behaviors: the synthesized catalysts (PWB300 and PWB400) at lower calcination temperatures maintained linear reaction progress, indicating catalyst stability without changing their chemical composition. Conversely, the prepared catalysts (PWB600 and PWB900) at higher temperatures exhibited linearity at the initial stage, followed by a curve pattern at the reaction’s end. This curvature is attributed to the transformation of Co 3 O 4 NPs into Co x B via in-situ reduction during the reaction, evidenced by ferromagnetic properties observed in the catalysts, although Co 3 O 4 NPs exhibited weaker ferromagnetism26,32,49. As depicted in Fig. 8, there were observable induction periods for PWB600 and PWB900 at lower reaction temperatures (30 and 35 °C) compared to PWB300 and PWB400-100. The catalytic measurements at 30 °C with 120 ml of hydrogen manifested HGRs of 748, 2454, and 2598 ml.min− 1.g− 1 for PWB300, PWB600, and PWB900, respectively. These data underscore the superior catalytic activity of the higher-temperature calcined catalysts, which is due to a balance between active surface sites and the degree of crystallinity7. As seen in Fig. 9(b), the HGR values of the optimal PWB900 catalyst are 2598, 4050, 6204, and 8308 ml.min− 1.g− 1 with 120 ml of H 2 evolved. The Arrhenius plot illustrates the relationship between ln(k) and 1/T for all catalysts. From the slopes of these lines, the activation energies (E a ) were calculated, with values of 70.8, 56.9, and 62.78 kJ.mol− 1 corresponding to PWB300, PWB600, and PWB900, respectively. The lower activation energies of the synthesized catalyst indicate higher HGR values compared to those reported in previous studies22,28.

Fig. 8 Full size image Calcination treatment effect of the prepared catalysts on hydrogen generation during NaBH 4 hydrolysis at different reaction temperatures: 30 °C (a), 35 °C (b), 40 °C (c), and (d).

Fig. 9 Full size image Effect of reaction temperatures on HGR values for the synthesized catalysts derived from: different ratios of parsley plant (a), and calcination temperatures (b).

Investigating the impact of different experimental parameters on HGR

Catalyst dosage effect

At ambient temperature, the self-hydrolysis of NaBH 4 was challenging in aqueous medium because of the slow generation of hydrogen gas. Consequently, several parameters affecting the efficiency of catalysts in evaluating hydrogen generation rates (HGRs) were examined; one of these parameters was the effect of catalyst weight. The assessment was conducted on PWB400-100 catalyst across all synthesized catalysts; this selection was justified by their superior catalytic performance and the combined effects observed in two study variables. Fixed conditions were employed in this reaction, including varying catalyst dosages (5, 10, 15, and 20 mg) and maintaining 0.75 wt% NaBH 4 in 20 ml of deionized water at 35 °C. As depicted in Fig. 10(a), increasing the catalyst weight resulted in enhanced catalytic activity, leading to a shorter completion time of the reaction. This is attributable to the higher availability of cobalt ions as active centers on the catalyst surface. Regarding the impact of catalyst weight on PWB 400, the reaction completion time declined from 14.17 to 5.3 min as the catalyst amount increased from 5 to 20 mg. Moreover, the HGR value (Fig. S3(a)) declined from 1804 to 1240 ml.min− 1.g− 1 with increasing catalyst weight from 5 to 20 mg. This decrease is due to the results from the reaction slope over catalyst weight. However, the evidence of increased activity is supported by higher rate constants and shorter reaction times at greater catalyst amounts, as seen in Fig. 10(a). As shown in Fig. S4(a), the linear relationship between ln (k) and ln (m cat ) was established for PWB400-100, with a linear slope of 0.74, which is a higher value compared to values reported by Hassan et al., Beheshti et al., and Wang et al.6,22,44. This finding demonstrates high fit accuracy, evidenced by an R2 value of 0.997 in these experiments. Based on this result, the kinetic study confirmed that hydrolysis follows first-order kinetics.

Fig. 10 Full size image Parameter effects: \(\:{\text{V}}_{{\text{H}}_{2}}\)-time curves (catalyst weight (a), NaBH 4 concentration (b), alkalinity effect (c), and catalyst recyclability (d)).

Effect of concentrations of NaBH 4

The effect of NaBH 4 concentrations was tested at a lower reaction temperature (35 °C) to minimize the self-hydrolysis of NaBH 4 . Measurements were performed with a constant catalyst weight (10 mg of PWB400-100) and varying concentrations of 0.25, 0.5, 0.75, and 1 wt% NaBH 4 . Lower concentrations were chosen to evaluate the change in HGR values with different NaBH 4 concentrations. The evolved volume of hydrogen at 132 ml was consumed with durations of 20, 14, 8.65, and 6.2 min, corresponding to increasing NaBH 4 concentrations, as depicted in Fig. 10(b). The data clearly demonstrate that increasing NaBH 4 concentration from 0.25 to 1 wt% causes the HGR value to increase from 658 to 2067 ml.min− 1.g− 1, as shown in Fig. S3(b). For instance, PWB400-100 exhibits higher HGR values at lower NaBH 4 concentrations. As depicted in Fig. S4 (b), the linear relationship between ln (k) and ln (NaBH 4 weight) was performed for PWB400-100, with a linear slope of 0.82 and a high fit accuracy (R2 value = 0.95). Combined with the catalyst weight effect, these results confirm that catalytic hydrolysis follows first-order kinetics. These findings underscore the significant influence of catalyst dosage and NaBH 4 concentrations on the catalytic activity of NaBH 4 hydrolysis, which is consistent with the reported data for this catalyzed by various cobalt-based catalysts18,21,22,44.

Alkalinity effect

Sodium hydroxide (NaOH) was used as an alkaline medium to inhibit the self-hydrolysis of NaBH 4 and maintain the solution’s stability during storage51. To evaluate this effect, the experiments were conducted with three different concentrations of NaOH (1, 3, and 5 wt%) and 0.75 wt% NaBH 4 , using 10 mg of PWB400-100 as the catalyst at a reaction temperature of 35 °C. The results showed that the catalytic performance of PWB400-100 declined with increasing NaOH concentration, as shown in Fig. 10(c). The obtained HGR over the bare catalyst was 1620 ml.min− 1.g− 1; the addition of 1 wt% NaOH resulted in a decline to 399 ml.min− 1.g− 1. A significant decrease in activity was observed at 3 and 5 wt% NaOH, with a drop in HGR values to 81 and 66 ml.min− 1.g− 1, respectively (Fig. S3(c)). Consistent with these findings, Demirci et al. reported a decline in HGR values during the reaction attributable to higher concentrations of the basic medium.52. This decline in activity occurs because higher alkalinity increases solution density, and hydroxyl ions (OH−) can hinder hydrolysis by decreasing NaBO 2 solubility. This decrease in solubility causes NaBO 2 to occupy active sites on the catalyst surface, leading to a decline in the catalytic activity53.

Recyclability and stability of the catalyst

Catalyst recyclability is a crucial parameter for assessing the life cycle and stability of catalytic efficiency over multiple NaBH 4 hydrolysis cycles; this is essential for ensuring long-term hydrogen generation sustainability. Recycle measurements were conducted using one of the synthesized catalysts (PWB400-100) across five hydrogen production cycles at 35 °C, maintaining consistent conditions (10 mg catalyst, 0.75 wt% NaBH 4 ). The utilized catalyst accumulated with filter paper after the end of each cycle; hence, the catalyst was washed with distilled water to remove any buildups, and dried at room temperature. Although the catalytic activity slightly declined after each run, the catalyst remained effective after five cycles, as shown in Fig. 10(d). The hydrogen generation rate (HGR) decreased from 1505 to 1080 ml.min− 1.g− 1 (Fig. S3(d)), with an approximate catalytic efficiency of 72%, after completing the 5th cycle. The performance decline over time is mainly due to the blockage of active centers by reaction by-product (NaBO 2 ) that adsorb on the catalyst surface54. Additionally, the reduction in conversion rate was ascribed to catalyst loss and fouling during repeated reactions.

Comparative study

Table 4 presents previous investigations of Co 3 O 4 catalysts with varying hydrogen generation rates (HGR) and provides a comparative analysis between these and our synthesized catalysts. This table demonstrates the operational conditions employed in this research, including (1) the method of catalyst preparation; (2) experimental conditions such as wt% of NaBH 4 and temperature; (3) the hydrogen generation rate (HGR); and (4) the activation energy (E a ). Consequently, the synthesized Co 3 O 4 catalysts demonstrate promising catalytic performance, evidenced by higher HGRs and lower activation energies, attributable to the utilization of a lower concentration of NaBH 4 compared to the higher concentrations used in other catalysts, thus surpassing many previous results. In conclusion, green-synthesized Co 3 O 4 catalysts derived from parsley plants via a co-precipitation method are suitable catalysts for the hydrolysis of NaBH 4 to generate pure hydrogen, employing an environmentally friendly and sustainable approach.

Table 4 Comparison of HGR values of various catalysts utilized for NaBH 4 hydrolysis. Full size table

Hydrolysis reaction mechanism

The schematic mechanism proposed in Fig. 11 depicts an appropriate representation of the reaction pathway for Co 3 O 4 catalysts during the hydrolysis of NaBH 4 . The surface of Co 3 O 4 NPs comprises two types of active sites: electron-deficient sites (Co3+) and electron-rich sites (Co2+). The Langmuir-Hinshelwood mechanism is deemed suitable for this process, as suggested in references4,21,56, and involves four key steps: (i) chemisorption of \(\:{\text{B}\text{H}}_{4}^{-}\)ions on Co2+ sites, while Co3+ sites bind with H 2 O molecules; (ii) cleavage of the B-H bonds in \(\:{\text{B}\text{H}}_{4}^{-}\)ions into \(\:{\text{B}\text{H}}_{3}^{+}\) and H− ions; (iii) reaction of H− ions with H+ from H 2 O to produce H 2 gas, along with adsorbed \(\:{\text{B}\text{H}}_{3}^{+}\) and OH− ions; and (iv) transformation of OH− ions from adsorbed H 2 O to \(\:{\text{B}\text{H}}_{3}^{+}\) ions on the catalyst surface, resulting in the formation of an adsorbed species (BH 3 (OH)−). This species interacts again with active sites and continues the cycle repeatedly to form BH 2 (OH) 2 , which further reacts similarly to yield BH(OH) 3 . The process continues with the respective reactions and repetitions, culminating in the evolution of four molecules of H 2 and the formation of \(\:{\text{B}\left(\text{O}\text{H}\right)}_{4}^{-}\) as a byproduct.