Houssem
Lakhali
a,
Ömer
Şahin
b and
Ayhan Abdullah
Ceyhan
*a
aDepartment of Chemical Engineering, Engineering and Natural Sciences Faculty, Konya Technical University, Konya, Turkey. E-mail: lakhalihoussem91@gmail.com; aaceyhan@ktun.edu.tr
bDepartment of Chemical Engineering, Faculty of Chemical and Metallurgical Engineering, Istanbul Technical University, Istanbul, Turkey. E-mail: omersahin@itu.edu.tr
First published on 13th June 2025
This study presents a novel core–shell magnetic catalyst, Fe3O4@SiO2/Co–Cr–B, engineered for efficient and reusable hydrogen generation from NaBH4 hydrolysis, offering significant advancement in sustainable hydrogen production technologies. The innovation lies in the synergistic integration of a magnetic Fe3O4@SiO2 core with a bimetallic Co–Cr–B shell, which enhances catalytic activity, structural stability, and facile magnetic recovery. Field emission scanning electron microscopy (FE-SEM) revealed a distinctive grape-like morphology resulting from nanoparticle agglomeration, which increased the surface area and active site accessibility. Transmission electron microscopy (TEM) confirmed a well-defined core–shell architecture with a uniform Co–Cr–B shell thickness of 40–50 nm and a consistent particle distribution. These structural features directly contribute to the catalyst's high hydrogen generation rate of 22.2 L gmetal−1 min−1 at 30 °C with a turnover frequency (TOF) of 2110.61 molH2 molcat−1 h−1. The catalyst demonstrated remarkable stability and maintained >90% of its initial activity after six consecutive reusability tests. These findings highlight the potential of this catalyst for large-scale hydrogen production and offer a promising route for industrial applications with improved efficiency and durability.
The combustion of fossil fuels releases hazardous pollutants, including sulfur dioxide, nitrogen dioxide, carbon monoxide, black carbon, polycyclic aromatic hydrocarbons (PAHs), and volatile chemicals, which contribute to issues such as the greenhouse effects and ground-level ozone formation (O3).3 In response to these environmental impacts, global efforts, as highlighted in the 2015 Paris Agreement, seek to limit global temperature rise to well below 2 °C, with an aspirational target of 1.5 °C above pre-industrial levels.4,5
Among various renewable energy alternatives, hydrogen is particularly noteworthy because of its clean combustion characteristics. Its oxidation produces only water as a by-product, and in the specific case of NaBH4 hydrolysis, sodium metaborate (NaBO2), without emitting carbon dioxide, sulfur compounds, or hydrocarbons. Moreover, hydrogen is a promising substitute for fossil fuels, as it can be sustainably produced from diverse renewable sources, such as solar, wind, and hydropower.6,7 Furthermore, hydrogen has a high calorific value and high energy density (120–142 kJ kg−1), nearly double that of gasoline, which allows more energy to be produced than conventional fuels. Its high octane number also enables operation at high compression ratios.8,9 Despite these advantages, its low density is a barrier to efficient storage and transportation, necessitating advanced technological solutions. Hydrogen storage can be categorized into two main types: material-based storage and physical storage.
Material-based storage includes liquid organic hydrogen carriers (e.g., BN-methyl cyclopentane), adsorbents (e.g., MOF-5, zeolite, and carbon nanotube nanostructures), interstitial hydrides (e.g., LaNi5H6), complex hydrides (e.g., NaAlH4), and chemical hydrides (e.g., NH3BH3).
Physical storage includes cryogenic liquid, pressurized gas and solid fuel as chemical/physical combination with material such as metal hydrides.10
Sodium borohydride (NaBH4) has been identified as a promising chemical hydride for hydrogen storage and generation because of its high storage capacity (10.8 wt%), which can release hydrogen through hydrolysis at relatively low temperatures especially in the presence of a catalyst. Additionally, compared with other hydrogen carriers, NaBH4 is stable in dry air and alkaline solutions, making it safe and manageable. NaBH4 hydrolysis can be performed under ambient conditions as shown in eqn (1):
![]() | (1) |
Catalysts play a crucial role in enhancing catalytic activity and hydrogen generation via NaBH4 hydrolysis. Various catalysts have been developed, including metal oxides,11 metal–organic frameworks,12 noble metals,13etc. However, the noble metal catalysts used for NaBH4 hydrolysis are scarce, expensive, and have limited abundance,14 and the development of cost-effective non-noble metal catalysts is a priority in the industry.
Transition metal catalysts such as Co and Cr are appropriate and desirable because of their solid acids, low cost, high stability, and high catalytic activity.15 For example, Callis et al. used a Cr-based catalyst that significantly enhanced hydrogen generation, resulting in a significant increase in catalytic activity (∼59% v/v).16 Co-based catalysts, particularly Co–B, are recognized for their strong performance in NaBH4 hydrolysis.14,17
However, transition-metal-based catalysts such as cobalt and chromium tend to agglomerate during the reaction, and their poor stability at high temperatures limits their use and reduces their catalytic activity.18 To address these problems, core–shell structures have been explored as catalysts that can enhance catalytic performance and efficiently inhibit the migration and aggregation of nanoscale active sites owing to the protective shell.19 Furthermore, core–shell architectures, particularly those employing magnetic cores like Fe3O4, not only increase stability by avoiding particle aggregation but also provide the special benefit of magnetic separation, thereby facilitating simple recovery, reusability, and durability of the catalyst.20
Fe3O4, a well-known magnetic material, is particularly valued for its resistance to oxidation, acids and alkalis, low toxicity, and biocompatibility.21,22 The considerable surface area of Fe3O4 NPs allows for uniform dispersion of active catalytic sites, which enhances catalytic efficiency and boosts hydrogen generation rate.23 Fe3O4 NPs exhibit high magnetism, facilitating their efficient separation from reaction mixtures under an applied magnetic field and promoting catalyst recovery and reusability without significant performance degradation.24 In the context of NaBH4 hydrolysis, Fe3O4-based catalysts have proven to be highly effective. For instance, F. Baye et al. synthesized an Fe3O4@C–X–Y–Co (Fe3O4@C core–shell) composite that achieved a hydrogen generation rate of 1746 mL g−1 min−1 and an activation energy of 47.3 kJ mol−1. This composite demonstrated excellent structural integrity and recyclability, making it promising for real-life applications with a minimal loss of catalytic activity.25
Despite these properties, Fe3O4 tends to agglomerate during the reaction, which significantly decreases catalytic performance. To overcome these issues, support materials including Al2O3,26 TiO2,27 CeO2,28 and carbon nanotubes29 have been employed to immobilize the catalyst, prevent agglomeration, and improve catalytic performance.30
SiO2, in particular, is an excellent support material for Fe3O4 NPs, preventing agglomeration by reducing magnetic dipole interactions and maintaining a uniform distribution of nanoparticles. In addition, SiO2 provides a high surface-to-volume ratio, which enhances its ability to interact with other substances and improves the performance of chemical reactions.21 In recent studies, Fe3O4@SiO2 core–shell structures have demonstrated excellent performance in hydrogen generation, attributed to their high surface area and effective dispersion of active sites. For instance, Dou et al. synthesized a Ru/CoOx@NPC catalyst using a core–shell metal–organic framework (MOF) as a template, achieving a hydrogen generation rate of 8.01 L min−1 g−1 with a turnover frequency, TOF, of 1118.6 molH2 min−1 molRu−1.31
This work focuses on the synthesis and characterization of a new Fe3O4@SiO2/Co–Cr–B core–shell magnetically separable catalyst for hydrogen generation from NaBH4 hydrolysis, contributing to the broader context of the green energy transition. While previous studies have explored Fe3O4-based catalysts and core–shell structures for this reaction, this work distinguishes itself by the specific combination of a Fe3O4@SiO2 core–shell structure and a Co–Cr–B active catalytic layer. This unique combination is hypothesized to offer several advantages crucial for practical application in a sustainable hydrogen economy: (i) the Fe3O4 core provides magnetic separability for easy catalyst recovery and reuse, minimizing waste and resource consumption, directly supporting circular economy principles; (ii) the SiO2 shell acts as a support to prevent Fe3O4 agglomeration and enhance the dispersion of the active catalytic sites, maximizing their availability for the reaction and improving long-term catalyst performance, a key factor for cost-effective hydrogen production; and (iii) the Co–Cr–B ternary metal boride catalytic layer, instead of a single metal or binary metal boride, is expected to exhibit synergistic effects, potentially leading to enhanced catalytic activity and stability compared to Co–B or Cr–B catalysts alone. This combination is relatively unexplored in the context of NaBH4 hydrolysis, and offers a promising avenue for developing highly active, stable, and cost-effective catalysts. Cost effectiveness and durability are essential for the widespread adoption of hydrogen as a clean energy carrier.
Furthermore, this study systematically investigates the influence of key reaction parameters, including the catalyst amount, NaOH and NaBH4 concentrations, and reaction temperature, on the hydrogen generation rate. The reusability of the catalyst is also evaluated, which is crucial for real-world applications and reducing the environmental footprint of catalyst synthesis. A comprehensive characterization of the synthesized catalyst is performed using XRD, FE-SEM, EDX, BET, TEM, XPS, and FTIR techniques to establish structure–property relationships and provide insights into the catalytic mechanism. This detailed investigation aims to demonstrate the superior performance of the novel Fe3O4@SiO2/Co–Cr–B catalyst for efficient and sustainable hydrogen generation, a critical component of the transition away from fossil fuels and towards a cleaner hydrogen-based energy future.
The catalyst was initially placed in a reaction flask connected to a gas collection system consisting of an exit tube leading to an inverted water-filled gas burette for volume measurement. Preprepared solutions of NaBH4 and NaOH were added to the catalyst in the flask without agitation to maintain consistency in the reaction kinetics. The hydrogen gas produced was directed through the tubing, displacing the water in the burette. The volume of displaced water was recorded as a direct measure of hydrogen gas generation.
It was assumed that H2 was the sole gaseous product of NaBH4 hydrolysis under the studied conditions, as commonly reported in the literature. However, in the absence of direct gas-phase analysis (e.g., GC or MS), this assumption is acknowledged as a limitation of the present study.
The FTIR spectrum of Fe3O4 (Fig. 2-a) exhibited a prominent absorption peak at approximately 590 cm−1, which corresponds to the stretching vibration mode of Fe–O bonds in Fe3O4, occupying tetrahedral positions.33 The C–O stretching vibration of epoxide was observed at 1067 cm−1.34 The characteristic peaks observed at wavenumbers of 1393 cm−1 and 1587 cm−1 corresponding to the COO–Fe bond were likely formed due to the interaction between the carboxylate anion of sodium citrate and the hydroxide radical present on the Fe3O4 surface.35 A typical hydroxyl group (–OH) bond peak was observed at approximately 3378–3400 cm−1, indicating the successful binding of sodium citrate to the Fe3O4 surface.36
The FTIR spectrum of Fe3O4@SiO2 (Fig. 2-b) also exhibited a peak around 580 cm−1, which is again associated with the Fe–O bond vibrations of Fe3O4 at the Fe–O6 (octahedral) sites.37 Peaks observed at 758 cm−1 and 843 cm−1 indicate the presence of C–Si–C stretching vibrations,38 while an additional peak at 948 cm−1 corresponds to Si–OH stretching and bending vibrations.39 The prominent peaks at 805 cm−1 and 1077 cm−1 were attributed to the asymmetric (υas) and symmetric (υs) stretching vibrations of the Si–O–Si bonds, respectively.40 The CO and C–OH stretching vibrations of the carboxyl groups (–COOH) contributed to the peaks at 1743 cm−1 and 1384 cm−1 respectively, while the peak at 1639 cm−1 was due to the bending vibration of the hydroxyl (–OH) group.41,42 Finally, in this spectrum, the peak at 2920 cm−1 was attributed to the C–H stretching vibration.43
The FTIR spectrum of Fe3O4@SiO2/Co–Cr–B (Fig. 2-c) displayed peaks in the range of 563 cm−1 to 578 cm−1 corresponding to the Fe–O bond vibrations.44,45 However, the peaks at 798 cm−1 and 1082 cm−1 are attributed to the presence of Si–O–Si bonds.46,47 The peak at 976cm−1 was assigned to the Cr
O bending vibration, indicating the presence of Cr in its oxidation state. The peaks at 826 cm−1, 890 cm−1, and 942 cm−1 were associated with Cr(VI), resulting from Cr–O–Cr bond vibrations.48,49 The peak at 558 cm−1 is attributed to the Co–O vibration.50,51 The bending vibration of the hydroxyl (–OH) group was responsible for the peak at 1639 cm−1, while the carboxyl group (–COOH) C
O and C–OH stretching vibrations contributed to the peaks at 1743 cm−1 and 1384 cm−1, respectively.41,42
The surface morphologies of the synthesized catalysts were examined using field-emission scanning electron microscopy (FE-SEM) on a Zeiss Sigma 300 instrument. Prior to imaging, the samples were gently ground and uniformly dispersed onto conductive carbon tape mounted on an aluminum SEM stub. To prevent charging during the analysis, the samples were coated with a thin conductive layer of gold using a sputter coater. Imaging was carried out under high-vacuum conditions using an accelerating voltage of 5–10 kV and optimized to obtain high-resolution micrographs without damaging the sample structure.
For elemental composition analysis, energy-dispersive X-ray spectroscopy (EDX) was conducted using the same Zeiss Sigma 300 system equipped with an Oxford Instruments EDX detector. EDX spectra were collected at an accelerating voltage of 15–20 kV to ensure sufficient X-ray generation and signal clarity. Elemental mapping and quantification were performed to confirm the presence and distribution of the key elements (Fe, Al, Co, Cs, O, C, etc.) in the catalyst structure.
As shown in the FE-SEM image in Fig. 3-a, the magnetic effect causes the particles to agglomerate, which is a common characteristic of Fe3O4.52 Based on the obtained particles, the catalyst predominantly consists of particles in the 100–200 nm range. The Fe3O4 NPs appear highly spherical and exhibit a rough surface texture with a proportional distribution.53
![]() | ||
Fig. 3 FESEM images/EDX spectrum/EDX map image of (a), (b) and (g) Fe3O4, (c), (d) and (h) Fe3O4@SiO2 and (e), (f) and (i) Fe3O4@SiO2/Co–Cr–B. |
The energy dispersive X-ray (EDX) results (Fig. 3-b) confirmed the efficient synthesis of Fe3O4, composed of 66.7 wt% Fe and 33.3 wt% O.
From the FE-SEM image of Fig. 3-c, it was observed that coating Fe3O4 with SiO2 decreased agglomeration, and the surface of the catalyst became smooth, with a particle size of approximately 200 nm. The SiO2-coated Fe3O4 nanoparticles maintained a spherical shape and were well-distributed, which suggests improved dispersion and stability.
In contrast, the EDX spectrum of Fe3O4@SiO2 (Fig. 3-d) shows that the percentages of Fe, O, C, and Si were 24.3 wt%, 30.3 wt%, 38.4 wt% and 7.1 wt%, respectively.
The adsorption of Co and Cr on the Fe3O4@SiO2 surface produced a catalytic surface (Fig. 3-e) characterized by a consistent size distribution and structure, with particle agglomeration forming a grape-like morphology. This morphology increased the active surface area of the catalysts, enhancing their interaction with NaBH4 for hydrogen generation.52
For Fe3O4@SiO2/Co–Cr–B (Fig. 2-f), the percentages of Co, Si, Fe, Cr, and C were 35.6 wt%, 7.7 wt%, 15.01 wt%, 0.6 wt%, and 13.0 wt%, respectively. Additionally, carbon (13.0 wt%) was detected, which is most likely attributed to the carbon coating applied to the FE-SEM sample holder.54
However, the “O” contents of Fe3O4, Fe3O4@SiO2, and Fe3O4@SiO2/Co–Cr–B are 33.3 wt%, 40.1 wt%, and 28.1 wt%, respectively. When Fe3O4 nanoparticles were coated with SiO2, the total oxygen concentration increased from 33.3 wt% (Fe3O4) to 40.1 wt% (Fe3O4@SiO2). The FTIR results suggest that the silica surface may contain hydroxyl groups (Si–OH) produced by water interactions, contributing to the overall oxygen concentration.38 In the Fe3O4@SiO2/Co–Cr–B catalyst, the high oxygen content is associated with characteristic bond vibrations corresponding to Cr–O–Cr and Co–O bonds, indicating an increased incorporation of oxygen within the material.48,51
The EDX map of Fe3O4 (Fig. 3-g) reveals the elemental composition confirming the presence of Fe and O, with a uniform distribution of Fe and O throughout the Fe3O4 surface supporting the high purity and uniformity of the synthesized Fe3O4 NPs.
The EDX map of Fe3O4@SiO2 in Fig. 3-h shows the presence of Fe, O, and Si, confirming the successful coating of Fe3O4 NPs with SiO2. The Fe signals were primarily concentrated in the core region, while Si and O were found surrounding Fe.
The EDX map image of Fe3O4@SiO2/Co–Cr–B in Fig. 2-i reveals a uniform and homogeneous distribution of the elements throughout the catalyst surface, with distinct signals for Fe, O, Si, Co, and Cr. This composite system is expected to exhibit enhanced magnetic and catalytic properties owing to the synergistic interaction between the Fe3O4 core, the SiO2 shell and the Co–Cr–B layer.
The TEM image (Fig. 4-a and b) of the synthesized Fe3O4@SiO2/Co–Cr–B clearly demonstrates the core–shell structure. The core is composed of the dark Fe3O4 which is known for its magnetic properties. This core is essential for applications requiring magnetic separation.55 The Fe3O4 cores are surrounded by a uniform gray SiO2 layer, which acts as a protective and stabilizing intermediate shell,56,57 followed by an outer Co–Cr–B layer with a reported thickness of approximately 40–50 nm which is clearly distinguishable in the TEM images.
The overall particle shape was mostly spherical, which is advantageous for achieving uniform coating and enhancing the surface area. This spherical morphology also facilitates better dispersion of the catalyst in reactive systems, promoting effective interactions with reactants such as NaBH4 for hydrogen generation. The observed core–shell structure and particle morphology are in line with expected outcomes of the synthesis strategy in the literature and further confirm the successful integration of Fe3O4, SiO2, and Co–Cr–B into a single composite system.58
Fig. 5-b displays the XRD pattern for Fe3O4@SiO2, which shows slight deviations from the Fe3O4 pattern with peaks at 2θ = 31.39°, 36.63°, 44.08°, 54.66°, 58.24°, and 63.88°. These shifts provide evidence for the presence of SiO2 in an amorphous state and show that its introduction does not change the crystalline structure of Fe3O4.60
In Fig. 5-c, the XRD pattern of the synthesized catalyst demonstrates diffraction peaks of cobalt and chromium oxides overlapping with those of Fe3O4, indicating the presence of this magnetic core, which is essential for the magnetic properties and ease of recovery of the catalyst.61 Peaks at approximately 2θ = 31.2°, 36.6°, 44.7°, 54.66°, 59.1°, and 65.2° corresponding to the (220), (311), (400), (511) and (440) planes suggest the formation of Co3O4 (JCPDS Card No. 96–900–5888).62 In addition, the presence of Cr2O3 is indicated by a peak at approximately 2θ = 35°.63 The peaks at approximately 2θ = 23.0°, 32.7°,45°, 53°, 58.2°, and 68.5° correspond to Cr3O4, according to the JCPDS File No. 12-559.64 The hydrolysis of NaBH4 was greatly improved by the combination of Fe3O4, Co3O4, and Cr2O3. Co and Cr oxides boost catalytic activity by increasing the surface area and electrical interactions, while the Fe3O4 core offers magnetic characteristics for simple separation and reuse.61,65,66
Textural properties | Fe3O4 | Fe3O4@SiO2 | Fe3O4@SiO2/Co–Cr–B |
---|---|---|---|
Surface area (m2 g−1) | 53.39 | 27.08 | 46.67 |
Pore diameter (nm) | 13.24 | 13.85 | 9.47 |
According to the BET data, the specific surface areas of the synthesized Fe3O4, Fe3O4@SiO2, and Fe3O4@SiO2/Co–Cr–B catalysts were 53.39, 27.08, and 46.67 m2 g−1, respectively. However, the results indicated a decrease in the surface area of Fe3O4@SiO2 compared to that of Fe3O4. This decrease is attributed to SiO2 coating, which leads to blockage of the surface pores through adsorption. Similar observations have been reported previously.21,67
Interestingly, upon the adsorption of metals onto the Fe3O4@SiO2 surface, the surface area increased significantly to 46.67 m2 g−1. For instance, Vishwanath et al. synthesized TiO2 as a support, while Madona et al. synthesized ZrO2 as a support for ZrO2 (M–ZrO2), in which M = Zr, Y, La, Ca, and Mg. The surface areas of TiO2 and Zr–ZrO2 have been reported to be 146 and 82 m2 g−1, respectively. Following metal adsorption, the surface area of Co–TiO2 increased to 309.8 m2 g−1, while the surface area of La–ZrO2 increased to 130 m2 g−1. The observed increase in the surface area may be attributed to the addition of a small amount of metal cations to the support, which improves the structural stability of the catalyst.68–71
To investigate the surface electronic states of the synthesized catalyst, XPS analysis was performed using a PHI 5000 Versa Probe spectrometer equipped with a monochromatic Al Kα X-ray source (hv = 1486.6 eV). The catalyst powders were gently ground and uniformly distributed onto copper (Cu) substrates, which were used as sample holders for analysis. To minimize the uncontrolled adsorption of atmospheric contaminants, such as water vapor, hydrocarbons, and oxygen, the samples were immediately transferred to a vacuum desiccator after mounting and stored under dry nitrogen conditions. All handling steps prior to the XPS analysis were carried out in a clean, controlled environment to limit exposure to ambient air. The sample transfer into the analysis chamber was performed rapidly to further reduce the possibility of contamination. The spectrometer was operated under ultra-high vacuum conditions (∼10−9 mbar) during the measurement. It is important to note that any Cu-related peaks in the spectra may originate from the underlying substrate and were carefully considered during the process of spectral deconvolution and interpretation.
This technique provided critical insights into the successful incorporation of key elements, as confirmed by the general XPS spectrum in Fig. 6-a, which revealed the presence of Si, O, Co, Cr, and B. To gain a detailed understanding of the chemical state of each element, the XPS peaks were deconvoluted using Gaussian functions.
![]() | ||
Fig. 6 XPS spectrum of (a) the synthesized catalyst, (b) Si 2p), (c) B 1s, (d) C 1s, (e) O 1s, (f) Cr 2p and (g) Co 2p. |
The high-resolution XPS spectrum of Si, shown in Fig. 6-b, highlights a peak at a binding energy of 102.5 eV related to Si–O bonding, which splits into 2P1/2, consistent with the presence of SiO2 on the surface of the catalyst.72 Splitting into 2P1/2 is a characteristic feature of the SiO2 structure, confirming the chemical state of silicon in the form of silicon dioxide.73
Similarly, Fig. 6-c illustrates the high-resolution B 1s spectrum, which reveals two distinct peaks, with the first attributed to the B-O bond with a binding energy of 192.5 eV indicative of boron oxide species,74 and the second peak centered at 188 eV corresponding to elemental boron or boron in a lower oxidation state, which may be indicative of unoxidized boron or boron carbide structures (B-sub-C).75 The B 1s binding energy observed near 198 eV is characteristic of boron in tetrahedral coordination, typically associated with BO4 units.76
The analysis of the C 1s spectrum, shown in Fig. 6-d, further complements these results by exhibiting a strong peak at 284.0 eV, indicative of C–C bonding. This observation aligns well with the previously reported FTIR results, which also confirm the presence of such bonding structures.77
In the XPS spectrum of O 1s (Fig. 6-e), a prominent peak at a binding energy of 532.8 eV was detected, which corresponds to C–OH and CO bonding. This observation is consistent with the presence of hydroxyl groups on the surface, as reported in the FTIR analysis, where –OH groups were identified on the surfaces through XPS analysis.78
As shown in Fig. 6-f, the high-resolution Cr 2p spectrum exhibited two characteristic main peaks. The first peak with a binding energy of 576.0 eV corresponds to Cr 2p3/2, while the second at 586 eV corresponds to Cr 2p1/2.79
The high-resolution XPS spectrum of Co 2p (Fig. 6-g) displays two prominent peaks: one at 781.6 eV, corresponding to Co 2p3/2, and another at 797.1 eV, assigned to Co 2p1/2, both of which are characteristic of the Co3+ oxidation state. The presence of Co3+ is particularly significant, as it is commonly associated with enhanced catalytic activity due to its ability to facilitate electron transfer processes, which are essential for catalytic reactions such as NaBH4 hydrolysis.6
However, the reported binding energies can be attributed to various oxides such as Co3O4, as reported in the XRD analysis.64 There are further signature shakeup satellite peaks at 786.0 eV and 803.1 eV that are designated as “sat”, which further supports the presence of cobalt in the catalyst.80,81
Atomic percentages were calculated by numerically integrating the intensity over the binding energy using the trapezoidal method. Each elemental peak was approximated using raw intensity data extracted from the XPS spectra. The integrated areas were then normalized to the total signal to derive the atomic percentage of each element (Table 2).
Element | Binding energy (eV) | Assigned species | Atomic percentage (%) |
---|---|---|---|
Si 2p | 102.5 | Si–O (SiO2) | ∼6.4 |
B 1s | 192.5/188.0 | B–O/elemental B | ∼3.7 |
C 1s | 284.0 | C–C | ∼22.5 |
O 1s | 532.8 | C–OH, C![]() |
∼39.2 |
Cr 2p | 576.0/586.0 | Cr3+ (Cr 2p3/2, 2p1/2) | ∼4.9 |
Co 2p | 781.6/797.1 | Co3+ (Co 2p3/2, 2p1/2) | ∼7.8 |
Co 2p (sat) | 786.0/803.1 | Satellite peaks (Co3+) | ∼7.8 |
The optimized catalyst was synthesized with a targeted Cr:
Co atomic doping ratio of 4
:
96, corresponding to a molar ratio of approximately 0.042. This intended composition was confirmed by energy-dispersive X-ray spectroscopy (EDX), which measured Cr and Co contents of 0.74 wt% and 14.84 wt%, respectively. These values yielded a Cr
:
Co weight ratio of approximately 0.049, which aligns well with the designed atomic ratio when considering the relative atomic weights of Cr and Co.
As shown in Fig. 7, for 5% (Co–Cr) the HGR was 3.36 L gmetal−1 min−1, with the required hydrogen volume being achieved within 23 min. An increase in the Co–Cr content from 10 to 30% led to a significant enhancement in the hydrogen generation rate (HGR) from 12.21 L gmetal−1 min−1 to 22.2 L gmetal−1 min−1. However, at 40% (Co–Cr), the HGR decreased to 13.58 L gmetal−1 min−1.
This observed situation can be attributed to several factors. The addition of chromium (Cr) as a dopant in Co-based catalysts has been shown to significantly enhance hydrogen generation rates, with improvements reaching up to 3–4 times that of undoped catalysts.
Cr promotes the formation of chromium oxide on the catalyst surface, as evidenced by XRD analysis, which confirms the presence of Cr3O4 and Cr2O3 phases. These oxides play a critical role in improving the dispersion of Co-based particles, such as Co–B, by preventing particle agglomeration and increasing the active surface area, ensuring greater accessibility to catalytic sites.
Additionally, Cr introduces Lewis acid sites into the catalyst, enhancing the adsorption and activation of OH− ions – key intermediates in the NaBH4 hydrolysis reaction – thereby accelerating the reaction kinetics. Furthermore, Cr strengthens the electronic interactions with Co, optimizes the charge transfer processes, and boosts the catalytic efficiency. These effects are further amplified when Cr is combined with other transition metals, such as tungsten (W), molybdenum (Mo), or copper (Cu), which collectively contribute to an expanded active surface area and improved catalytic performance.
Collectively, these findings emphasize the pivotal role of Cr in enhancing the catalytic performance of Co-based systems for hydrogen generation. Moreover, for effective NaBH4 hydrolysis, the ideal Co–Cr ratio (30%) probably strikes a balance between sufficient active sites and the required electrical interactions. This is consistent with the general finding that by maximizing the surface shape and electrical characteristics, certain metal ratios can enhance the catalytic performance.63,82
Among the synthesized catalysts, the Fe3O4@SiO2/30%(Co–Cr) composition was identified as the optimal loading for the NaBH4 hydrolysis reaction.
NaOH is commonly used as a stabilizer to prevent the self-hydrolysis of NaBH4 when it is dissolved in distilled water.83,84 As shown in Fig. 8, increasing the NaOH concentration from 3% to 7% resulted in a significant increase in the HGR from 19.14 L gmetal−1 min−1 to 22.2 L gmetal−1 min−1.
This increase may be due to the addition of hydroxyl ions (OH−) to the reaction medium, which changes the electronic properties of the catalyst, thereby enhancing the substrate–catalyst interaction, resulting in the release of more hydrogen.52 However, with a further increase in NaOH concentration (10%), the HGR decreased to 21.07 L gmetal−1 min−1. This decrease can be explained by the increased viscosity of the solution caused by excess hydroxyl ions (OH−), resulting in a decrease in the diffusion rate, which inhibits hydrogen generation.83
OH− ions can be competitively adsorbed on the catalyst surface with respect to BH+ ions, ultimately decreasing the HGR.85 Furthermore, a higher NaOH concentration facilitates the formation of by-products, such as NaBO2·xH2O and NaB(OH)4, which are deposited on the surface of the catalyst, inhibiting hydrogen release.86
The presence of –OH groups on the catalyst support plays a pivotal role in enhancing the metal–support interaction, which is essential for the uniform dispersion of active metal sites such as Pt. This enhanced dispersion increases the availability of catalytic sites, resulting in a higher hydrogen generation rate (HGR) and reduced activation energy during the hydrolysis process.
Additionally, –OH groups contribute to improved electronic conductivity by creating electron-enriched active sites, which facilitate charge transfer and accelerate reaction kinetics. These combined effects highlight the crucial role of –OH groups in enhancing the catalytic performance for hydrogen generation.87
However, for the next stages, 7% NaOH was tallied as a suitable concentration.
As shown in Fig. 9, increasing the catalyst amount from 25 mg to 100 mg resulted in a substantial decrease in reaction time and a significant increase in the hydrogen generation rate (HGR), which increased from 13.24 L gmetal−1 min−1 to 22.2 L gmetal−1 min−1.
This enhancement can be attributed to the increased number of active sites as the catalyst amount increased, which accelerated the reaction and boosted hydrogen generation efficiency.88–90
As highlighted in the FESEM image (Fig. 2-e), spherical particles, such as those observed in the NiFe2O4 and Co3O4 catalysts, have been shown to optimize catalytic performance by providing a uniform surface area that enhances reaction kinetics and stability.91,92
The spherical shape facilitates better contact with the reactants, thereby improving the efficiency of hydrogen generation. Moreover, spherical particles often have a higher surface-area-to-volume ratio, which increases the number of active sites available for the reaction. This is crucial for enhancing the hydrogen generation rate because more active sites can facilitate faster reaction rates.93,94 The uniformity of spherical particles aids in better heat and mass transfer during the reaction, which is essential for maintaining optimal reaction conditions and improving the overall hydrogen generation rate.95
The surface composition and chemical states of the Fe3O4@SiO2/Co–Cr–B catalyst, as revealed by XPS analysis, play a critical role in the effect of catalyst amount on NaBH4 hydrolysis.
The SiO2 support, evidenced by the Si 2p and O 1s states, prevents the aggregation of Co–B nanoclusters, thereby maintaining a high surface area and enhancing catalytic activity through improved reaction kinetics.65 Co 2p states indicate the presence of both elemental and oxidized forms of Co, which facilitate electron transfer and strengthen the electronic interactions with B and Cr, further boosting catalytic performance.63,65
Similarly, Cr 2p states reveals the dual role of Cr in the system: at optimal concentrations, Cr enhances the dispersion of Co–B particles and increases the surface area, significantly improving catalytic activity. However, excessive Cr leads to the formation of surface oxides that block active sites, thereby reducing efficiency.65,96 These findings highlight the importance of optimizing surface composition and chemical states to maximize the catalyst's performance in NaBH4 hydrolysis. In the following step of the study, parameter impacts were investigated using 100 mg of catalyst.
Fig. 9 also shows the change in ln(HGR) with ln(catalyst amount), which fluctuates linearly with ln(catalyst amount), in order to ascertain the relationship between the reaction rate and catalyst amount. This finding indicates that depending on the catalyst amount, the hydrolysis process was approximately first order (n = 0.982).
The hydrolysis reaction in the liquid phase proceeds via two critical steps.
1. Equilibrium adsorption of BH4− species on the catalyst surface.
2. Hydrogen generation from adsorbed species.97
As NaBH4 concentration increases, the viscosity of the solution also increases, leading to the precipitation of NaBO2 by-products on the catalyst surface. This precipitation impairs hydrogen generation by obstructing the active sites. Additionally, increased viscosity inhibits mass transfer, further decreasing the overall catalytic performance.98,99 Moreover, the interaction of functional groups with NaBH4 molecules, as observed in the FTIR analysis, is pivotal in determining the efficiency of hydrogen generation. The C–O and CO groups can facilitate the adsorption of NaBH4 onto the catalyst surface, promoting its decomposition and subsequent hydrogen release. This is because these groups can form hydrogen bonds or other interactions with NaBH4, stabilizing the transition state and lowering the activation energy required for the reaction.100 Additionally, the presence of C–OH groups can further enhance these interactions by providing additional sites for NaBH4 adsorption and activation.
As shown in Fig. 11, the HGR increases from 14.67 L gmetal−1 min−1 to 27.51 L gmetal−1 min−1 as the temperature increases from 20 °C to 50 °C. This resulted in a significant decrease in the time required for hydrogen generation. Increasing the temperature increased the mobility of the molecules, thereby increasing their collision probability. In addition, the interactions between the catalyst surface and molecules become more intense. This leads to a significant increase in HGR.101,102
Table 3 highlights the performance of the polynomial regression model used to analyze the data for all parameters (NaOH concentration, catalyst amount, NaBH4 concentration, and temperature). The key performance metrics include mean squared error (MSE), standard deviation (S), and R-squared (R2) values.
Parameter | Mean squared error (MSE) | Standard deviation (S) | R-squared (R2) |
---|---|---|---|
NaOH concentration (wt%) | 2.31 × 103 | 4.81 × 101 | 0.998 |
Catalyst amount (mg) | 3.39 × 105 | 5.82 × 102 | 0.976 |
NaBH4 concentration (wt%) | 2.48 × 105 | 4.98 × 102 | 0.987 |
Temperature (°C) | 2.24 × 105 | 4.73 × 102 | 0.991 |
For NaOH concentration, the model explained 99.8% of the variability in HGR (R2 = 0.998), indicating strong predictive capability with minimal error (MSE = 2.31 × 103; S = 48.1). In the case of catalyst amount, the model effectively captured nonlinear behavior, reflected in an R2 of 0.976, and a corresponding MSE of 3.39 × 105. Similarly, for NaBH4 concentration, the polynomial regression yielded robust performance (R2 = 0.987; MSE = 2.48 × 105). Regarding temperature, the model demonstrated the highest predictive accuracy, with an R2 of 0.991 and the lowest MSE (2.24 × 105).
However, it is important to note that the turnover frequency (TOF) was calculated based on a single experimental condition (7 wt% NaOH, 1 wt% NaBH4, and 30 °C). Therefore, no statistical averaging or error analysis can be applied to TOF in this context.
The rate constant k and the activation energy can be calculated according to eqn (2):
![]() | (2) |
![]() | (3) |
![]() | (4) |
![]() | (5) |
![]() | ||
Fig. 12 (1/T) variation with ln(k) for NaBH4 hydrolysis for the (a) nth-order model and (b) Langmuir–Hinshelwood model. (c) (1/T) variation with ln![]() |
According to the results, the activation energy was determined as 20.75 kJ mol−1.
![]() | (6) |
![]() | (7) |
To enhance the interpretation of the temperature-dependent adsorption behavior, a thermodynamic analysis was conducted using the van’t Hoff equation.
However, eqn (8) can be used to compute the entropy change (ΔS°) and enthalpy change (ΔHads) based on the obtained Kads values. From this, it is also possible to calculate the Gibbs free energy change (ΔG°).
![]() | (8) |
A linear correlation was observed in the van’t Hoff plot of ln(Kads) versus 1/T, indicating that the adsorption process adheres to classical thermodynamic expectations. From the slope and intercept of the regression line in Fig. 12-c, the standard enthalpy change of adsorption (ΔH°) was calculated to be –5.44 kJ mol−1, confirming the exothermic nature of the adsorption process. The corresponding standard entropy change (ΔS°) was determined to be +4.54 J mol−1 K−1, suggesting a modest increase in system disorder upon adsorption.
Furthermore, the standard Gibbs free energy change (ΔG°) (eqn (9)) was computed over the investigated temperature range, yielding values from –6.76 kJ mol−1 at 293 K to –690 kJ mol−1 at 323 K:
ΔG° = ΔHads − T·ΔS° | (9) |
These negative ΔG° values indicate that the adsorption remains thermodynamically spontaneous across all studied temperatures. However, the diminishing magnitude of ΔG° with increasing temperature aligns with the observed decrease in the adsorption constant (Kads), attributable to enhanced desorption kinetics. This analysis reinforces the mechanistic conclusion that temperature-induced desorption is the primary factor governing the decline in adsorption capacity at elevated temperatures.
Nevertheless, the Gibbs free energy change (ΔG°) at each temperature can also be determined using eqn (9) and is summarized in Table 4.
T | K ads | ΔG° (kJ mol−1) | ΔS° (J mol−1 K−1) | ΔHads (kJ mol−1) |
---|---|---|---|---|
20 °C | 16 | −6.67 | 4.54 ± 0.01 | −5.44 ± 0.01 |
30 °C | 15 | −6.81 | ||
40 °C | 14 | −6.85 | ||
50 °C | 13 | −6.90 |
The fit of the experimental data to the Langmuir–Hinshelwood kinetic model was rigorously evaluated across a range of temperatures (20–50 °C) and corresponding concentration variations. The model exhibited excellent agreement with the experimental results at all temperatures, as indicated by the high coefficients of determination (R2) of 0.998 at 20 °C, 0.9994 at 30 °C, 0.9966 at 40 °C, and 0.9976 at 50 °C. These R2 values suggest a strong correlation between the model predictions and observed data, supporting the applicability of the Langmuir–Hinshelwood model under the studied conditions.
Further statistical validation was conducted by calculating the mean squared error (MSE) and the standard deviation (S) of the experimental concentrations. At 20 °C, MSE was notably low at 8.0 × 10−6, with a standard deviation of 0.07599. At 30 °C, MSE increased slightly to 0.00248, accompanied by a standard deviation of 0.07069. At 40 and 50 °C, the MSE values were 0.00339 and 0.00393, respectively, with the corresponding standard deviations of 0.07779 and 0.08571, respectively. These relatively low MSE values across all temperatures demonstrated a consistently strong fit of the model, whereas the standard deviations reflected reasonable variability within the experimental data, which is typical for catalytic kinetic systems.
In addition to the goodness-of-fit metrics, the kinetic parameters derived from model fitting, including the rate constant (k) and adsorption constant (K), also showed logical trends with temperature. The rate constant increased from 0.0077 at 20 °C to 0.0173 at 50 °C, indicating an acceleration of the reaction kinetics with temperature, as expected from Arrhenius-type behavior.
Conversely, the adsorption constant (K) decreased from 16 to 13 over the same temperature range, which is consistent with the reduced surface adsorption at elevated temperatures owing to enhanced desorption dynamics. However, the absorption equilibrium constant (Kads) measures the similarity between the adsorbate and adsorbent systems, and a higher Kads value often indicates greater adsorption effectiveness of inhibitor molecules on the metallic surface.
Collectively, the combination of high R2 values, low MSEs, and reasonable trends in the kinetic parameters confirmed that the Langmuir–Hinshelwood model not only fits the experimental data well across the investigated conditions but also yields statistically robust and physically meaningful parameters (Table 5).
Temperature (°C) | Rate constant (k) | Adsorption constant (K) | R-squared (R2) | Mean squared error (MSE) | Standard deviation (S) |
---|---|---|---|---|---|
20 | 0.0077 | 16 | 0.9980 | 0.000008 | 0.075986 |
30 | 0.0125 | 15 | 0.9994 | 0.002478 | 0.070691 |
40 | 0.0146 | 14 | 0.9966 | 0.003390 | 0.077790 |
50 | 0.0173 | 13 | 0.9976 | 0.003930 | 0.085710 |
The typical decrease in the adsorption constant (K) with increasing temperature is primarily due to enhanced desorption kinetics. As the temperature increased, the adsorbed species gained more thermal energy, making it easier for them to overcome the adsorption energy barrier and desorb from the surface. This leads to reduced surface coverage and a lower value of K. This behavior is characteristic of exothermic adsorption processes, where equilibrium shifts toward desorption at higher temperatures.103–105
In heterogeneous catalysis, turnover frequency (TOF) is a crucial kinetic parameter that shows how many reactant molecules are transformed per active site in a certain amount of time. It is commonly represented as molH2 molcat−1 h−1, as eqn (10) illustrates. TOF concentrates on the number of active sites engaged in the reaction, as opposed to the hydrogen generation rate (HGR), which counts the quantity of hydrogen generated in relation to the mass of the catalyst. Through this process of normalizing catalytic activity, TOF offers a clear indicator of intrinsic efficiency that is unaffected by changes in mass, surface area, or catalyst loading.106–108
![]() | (10) |
At 30 °C, the Fe3O4@SiO2/Co–Cr–B catalyst exhibited a remarkable turnover frequency (TOF) of 2110.61 molH2 molcat−1 h−1 during NaBH4 hydrolysis, corresponding to a hydrogen generation rate (HGR) of 22.2 L gmetal−1 min−1.
Table 6 compares the activation energy and HGR of the Fe3O4@SiO2/Co–Cr–B catalyst with those of other Co, Cr, and Fe3O4@SiO2 catalysts reported in the literature. Notably, our catalyst exhibited a high hydrogen generation rate with very low activation energy.
Catalyst | Experimental conditions | HGR (L g−1 min−1) | E a (kJ mol−1) | Ref. |
---|---|---|---|---|
Ru–Co3O4 | 1 wt% NaOH, 10 wt% NaBH4 | 6.51 | 28.26 | 13 |
CNT-promoted Co–B | 5 wt% NaOH, 5 wt% NaBH4 | 12.00 | 23.5 | 109 |
Amorphous Co–Cr–B | 1 mol NaOH/L, 1 mol NaBH4/L | 0.11 | 44.51 | 96 |
Co–P–B | 2.5 M NaOH, 0.025 M NaBH4 | 4 | 32 | 110 |
Carbon-supported Co–B | 20 mM NaOH, 0.2 M NaBH4 | 2.07 | 57.8 | 111 |
Ni–B–Cr0.8@RH | 1.25 wt% NaOH, 2.5 wt% NaBH4 | 1.42 | 50.64 | 112 |
Co–Cr–B@NG | 1.0![]() |
2.23 | 38.41 | 113 |
Co–B/HPCM | 5 wt% NaOH, 2.25 wt% NaBH4 | 3.08 | 43.3 | 114 |
Fe3O4@SiO2/Co–Cr–B | 7 wt% NaOH, 1 wt% NaBH4 | 22.2 | 23.6 | This study |
To contextualize the catalytic performance of the synthesized Fe3O4@SiO2/Co–Cr–B catalyst, a benchmark comparison was performed with several state-of-the-art non-noble metal catalysts reported in the literature, as well as our previously developed IA-CNT@Co–Mo–B catalyst. The Fe3O4@SiO2/Co–Cr–B catalyst demonstrated a hydrogen generation rate (HGR) of 22.20 L min−1 g−1 and a low activation energy (Ea) of 23.60 kJ mol−1, both of which significantly exceed the performance metrics of many reported systems.
Compared to our previously synthesized IA-CNT@Co–Mo–B, which achieved an HGR of 5.3 L min−1 g−1 and an Ea of 24.58 kJ mol−1, the Fe3O4@SiO2/Co–Cr–B catalyst exhibits over a fourfold increase in HGR, while maintaining a slightly lower activation energy. These results suggest a more efficient catalytic pathway and superior hydrogen evolution kinetics under similar experimental conditions.115
This performance surpasses that of g-C3N4/Co–W–B/Ni foam, which exhibited an HGR of 7.3 L min−1 g−1 and an Ea of 31.5 kJ mol−1.116 Additionally, it outperforms the Ni–Co/r-GO catalyst, which achieved a hydrogen production rate of 1.28 L min−1 g−1 and a considerably higher Ea of 55.12 kJ mol−1.117
The superior HGR and notably lower activation energy of Fe3O4@SiO2/Co–Cr–B highlight its highly efficient catalytic behavior and favorable reaction kinetics. These attributes indicate that the catalyst can effectively initiate and sustain hydrogen generation reactions at lower energy thresholds, which is a key consideration for practical and scalable hydrogen production systems.
As depicted in Fig. 14, the hydrogen generation rate (HGR) decreased from 18.80 L gmetal−1 min−1 in the first cycle to 4.08 L gmetal−1 min−1 after six reaction cycles. This decline in performance can be attributed either to the loss of approximately 90% of the catalyst during its recovery from the hydrolysis process, or to the precipitation of NaBO2 on the catalyst surface, which blocks active sites and reduces hydrogen generation. Despite this, theoretical hydrogen volume was achieved in each reusability experiment, indicating that the catalyst maintained its hydrogen generation potential, even after repeated use.118
![]() | ||
Fig. 14 Hydrogen generation rate variation with the number of reusability (30 °C, 1% NaBH4, 100 mg catalyst, 7% NaOH). |
FE-SEM analysis revealed that the catalyst maintained its spherical morphology with no significant structural alterations after the reusability experiments, indicating good morphological stability under the applied conditions, which is consistent with previous reports (Fig. 15-a and -c). However, EDX analysis confirmed the presence of borate species on the catalyst surface, suggesting that surface interactions or deposition occurred during the reaction.119,120 Moreover, inductively coupled plasma (ICP) analysis revealed that the catalyst retained its elemental integrity over five consecutive hydrolysis cycles with no detectable leaching of Co or Cr into the reaction medium. This confirmed the excellent compositional stability and reusability of the catalyst under the applied conditions.
![]() | ||
Fig. 15 FESEM/EDX image of the synthesized Fe3O4@SiO2/Co–Cr–B catalyst (a) and (b) before the reusability test and (c) and (d) after the reusability test. |
The magnetic recovery feature of the Fe3O4@SiO2/Co–Cr–B catalyst simplifies the separation process, reduces the downtime, and enables continuous operation, making it highly practical for industrial applications. Additionally, the use of non-noble metals, such as Co and Cr, offers a cost-effective alternative to precious metal catalysts, enhancing the feasibility of large-scale production. Future work should explore the integration of this catalyst into continuous-flow reactors, which are widely employed in the chemical industry, to further optimize its application. Combining a catalyst with renewable hydrogen sources could also strengthen its role in achieving global sustainability goals, such as those outlined in the Paris Agreement.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2025 |