Carbon and Silica Supports Enhance the Durability and Catalytic Performance of Cobalt Oxides Derived from Cobalt Benzene-1,3,5-Tricarboxylate Complex

Abstract

Addressing the urgent need for robust and sustainable catalysts to detoxify nitroaromatic pollutants, this study introduces a novel approach for synthesizing cobalt oxide nanocomposites via pyrolysis of cobalt benzene-1,3,5-tricarboxylate. By integrating porous carbon (PC) and nano silica (NS) supports with Co₃O₄ to form (Co₃O₄/PC) and (Co₃O₄/NS), we achieved precise morphological control, as evidenced by SEM and TEM analysis. SEM revealed 80–500 nm Co₃O₄ microspheres, 300 nm Co₃O₄/PC microfibers, and 2–5 µm Co₃O₄/NS spheres composed of 100 nm nanospheres. TEM further confirmed the presence of ~15 nm nanoparticles. Additionally, FTIR spectra exhibited characteristic Co–O bands at 550 and 650 cm⁻¹, while UV–Vis absorption bands appeared in the range of 450–550 nm, confirming the formation of cobalt oxide structures. Catalytic assays toward p-nitrophenol reduction revealed exceptional kinetics (k = 0.459, 0.405, and 0.384 min⁻¹) and high turnover numbers (TON = 5.1, 6.7, and 6.3 mg 4-NP reduced per mg of catalyst), outperforming most of the recently reported systems. Notably, both supported catalysts retained over 95% activity after two regeneration cycles. These findings not only fill a gap in the development of efficient, regenerable cobalt-based catalysts, but also pave the way for practical applications in environmental remediation.

1. Introduction

The primary source of water pollution is organic pollutants from industries. These toxins pose a serious threat to human health and the environment [1]. Aquatic plants and animals are poisoned by nitroaromatic chemicals. Nitrophenols not only cause stench, but they also partially alter the color of water bodies, which makes them sources of pollution by blocking sunlight and posing serious threats to aquatic ecosystems [2]. One of the most refractory contaminants, which can appear in industrial wastewater from the dye, explosive, pesticide, plasticizer, and herbicide industries as well as in agricultural wastewater, is para-nitrophenol (PNP). PNP is a very poisonous substance that poses a serious risk to humans. Exposure to PNP can harm the kidneys, liver, central nervous system, and blood cells [3,4].

In addition to being harmful to the environment and carcinogenic to individuals, nitrophenols are very difficult to biodegrade. As a result, nitrophenol and its derivatives have been classified as priority contaminants by the U.S. Environmental Protection Agency (EPA), which suggests that their concentration in natural water bodies should not exceed 10 mg/mL Nitrophenols induce skin necrosis and eye irritation. They damage the muscles, kidneys, and liver as well. According to some reports, an adult can die from exposure to as little as one gram of nitrophenols [5]. Numerous methods, including polyphenylsulfone-based blend membranes [6], microbial degradation [7], photocatalytic reduction using a reduced graphene oxide and ZnO composite [8], the electro-Fenton method [9], electrocoagulation [10], adsorption with 2-Phenylimidazole-Modified ZIF-8 [11], and electrochemical treatment [12], have been widely used for eliminating nitrophenols from contaminated water.

Physical methods such as membrane filtration and adsorption are widely employed for removing suspended solids, particulates, and a broad spectrum of dissolved contaminants from water. Adsorption onto activated carbon or other porous media is simple to implement, and is effective against organic pollutants, yet it generates spent adsorbent waste and oftentimes requires frequent regeneration or replacement to maintain performance. Membrane processes—including microfiltration, ultrafiltration, nanofiltration, and reverse osmosis—offer high removal efficiency and modular operation but suffer from fouling, high maintenance costs, and significant energy demands for operation and cleaning cycles [13]. Coagulation–flocculation can rapidly aggregate colloidal impurities but yields large volumes of sludge that require disposal [14,15]. Advanced oxidation processes (AOPs), ozonation, or photocatalysis achieve mineralization of recalcitrant organics with high efficiency, yet they demand precise pH control and elevated energy input (ozone generators, UV lamps), and can form by-products that necessitate further treatment [16,17].

It is generally acknowledged that chemical reduction by NaBH₄ with metallic nanoparticles (NPs) acting as catalysts is an environmentally friendly and effective method of removing 4-nitrophenol [18]. Additionally, because para-aminophenol (PAP) is a valuable chemical for making medications (such as analgesics and antipyretics) and has been utilized in various applications, including corrosion inhibitors, hair dyes, and photographic developers, the conversion from PNP to p-aminophenol PAP is of significant commercial value [19]. Catalytic reduction of nitroaromatic pollutants using supported metal-oxide nanocomposites has emerged as a promising alternative that addresses the shortcomings of traditional methods. Recent studies on cobalt oxide nanoparticles anchored on carbon frameworks demonstrate rapid kinetics, nearly complete conversion of p-nitrophenol, and easy catalyst recovery via magnetic or filtration techniques, thereby minimizing secondary waste and energy consumption [20]. In [21], cobalt oxide nanocomposites were produced by heating the cobalt acetylacetonate complex to 300 °C and subsequently used to catalytically degrade PNP to PAP. In [22], it was found that the reduction rate of PNP decreased as the crystallite size of the mesoporous Co₃O₄ catalysts prepared via the sol–gel method increased. In another study, Murraya koenigii plant leaves in an aqueous solution were successfully used to prepare CuO NPs via the green synthesis method. The biosynthesized CuO NPs facilitated the reduction of 4-NP to 4-AP. The catalyst’s activity remained constant for up to four recycles and excellent product yields were obtained [23].

In this work, we synthesize Co₃O₄, Co₃O₄/porous carbon (PC), and Co₃O₄/nano silica (NS) composites through pyrolysis of the metal–organic framework (MOF) complex (cobalt benzene-1,3,5-tricarboxylate) in the presence of porous carbon and nano silica supports, achieving precise control over morphology and active-site dispersion. Compared to membrane filtration, our catalysts eliminate fouling concerns and membrane replacements; relative to AOPs, they operate under ambient conditions without auxiliary oxidants or UV irradiation; and in contrast to biological systems, they deliver orders-of-magnitude-faster degradation rates and withstand multiple regeneration cycles with >95% retained activity. By integrating high catalytic efficiency, low operational energy, and robust recyclability, these cobalt oxide nanocomposites offer a complementary, sustainable strategy for water remediation that addresses the shortcomings of conventional treatments [24].

The reduction of PNP to p-aminophenol (PAP) has become a benchmark reaction in heterogeneous catalysis for several compelling reasons. It follows well-defined pseudo-first-order kinetics under excess NaBH₄, enabling direct comparison of rate constants. The reactant and product exhibit distinct UV–Vis absorption peaks for PNP and for PAP, facilitating in situ monitoring of conversion without complex sampling or derivatization. In a great number of studies published in the literature, this standardized assay is employed to benchmark catalytic activity, selectivity, and recyclability across diverse nanomaterials [25].

The as-synthesized cobalt oxides were characterized by various spectroscopic techniques. The cobalt oxide nanocomposites exhibited high catalytic activities for the reduction of p-nitrophenol (PNP) under mild reaction conditions. Notably, the supported catalysts maintained performance over numerous cycles and achieved turnover numbers that surpassed those of the conventional nano catalysts reported in the literature. These results highlight the promise of pyrolyzed cobalt MOFs—especially when anchored on customized supports—as robust, high-performance catalysts for environmental remediation.

2. Results and Discussion

2.1. Synthesis of Metal-Oxide Nanocomposite Catalysts

The starting material Co(BTC) was prepared by the reaction of cobalt chloride and benzene tricarboxylic acid in DMF at 150 °C for 2 days. Co(BTC)/PC and Co(BTC)/NS were prepared similarly but with the addition of porous carbon and nano silica, respectively. FTIR spectra of the three complexes are provided in Figure S1. The metal oxide Co₃O₄ and the Co₃O₄/PC and Co₃O₄/NS nanocomposites were prepared from the starting complexes Co(BTC), Co(BTC)/PC, and Co(BTC)/NS, respectively, by pyrolysis in air at 425 °C for 1 h.

2.2. UV–Visible Absorption Spectra

Figure 1 shows UV–Vis spectra of cobalt oxide and nanocomposites. The UV–Vis spectra of the prepared samples were measured in the range of 200 to 600 nm. The spectra indicated the existence of absorption peaks at 273 and 490 nm for Co₃O₄, 268 nm and 513 nm for Co₃O₄/PC, and 230 nm and 434 nm for Co₃O₄/NS. The peaks can be attributed to the presence of cobalt oxide in the samples.

Figure 1 shows two sets of absorption bands for Co₃O₄ and its composites. A high-energy UV band (230–273 nm) arises from ligand-to-metal charge transfer (LMCT: O²⁻(2p) → Co³⁺(3d)). Its exact position can shift depending on particle size, surface defects, and composite–support interactions [26]. A lower-energy visible band (434–513 nm) is due to spin-allowed d–d transitions of cobalt ions in the spinel lattice. The band at ∼490 nm in bare Co₃O₄ may be due to the Co²⁺(Td) 4A₂(F) → 4T₁(P) transition. A shoulder or secondary peak near 430–440 nm may arise from lower-intensity Co³⁺(Oh) (1A₁g → 1T₁g) transitions [27]. In Co₃O₄/PC and Co₃O₄/NS, these bands shift to 513 nm and 434 nm, respectively, reflecting modifications in the crystal-field splitting (Δ) caused by interactions with the porous carbon (PC) or nano silica (NS) matrices [28]. Embedding Co₃O₄ into conductive carbon (PC) or silica (NS) alters local symmetry and bond covalency around Co centers, tuning the crystal-field splitting. This leads to red-shifts (Co₃O₄/PC) or blue-shifts (Co₃O₄/NS) in the d–d bands relative to bare Co₃O₄ [28].

2.3. FTIR Spectra

The FTIR spectra (cm⁻¹) of Co₃O₄, Co₃O₄/PC, and Co₃O₄/NS all revealed twin peaks, at (654.2, 543.4); (655.4, 550.4), and (657.4, 555.6), respectively, that could be attributed to the Co-O stretching vibration mode and the bridging vibration of Co-O bonds of the spinel structure of Co₃O₄. The band at approximately 655 cm⁻¹ is assigned to the Co²⁺–O stretching vibration in tetrahedrally coordinated sites, while the feature at 550 cm⁻¹ corresponds to the Co³⁺–O stretch in octahedrally coordinated environments [28]. Also, a weak peak was observed for Co₃O₄/PC at 1654.5 due to >C=C< stretching. Additionally, the Co₃O₄/NS exhibited a weak peak at 3411.4 due to OH stretch, O-H bending at 1651.7, a broad weak peak at 1112.6 which could be attributed to Si-O-Si asymmetric stretch, Si-OH stretch at 917, and Si-O symmetric stretch at 835.6 (Figure 2).

2.4. SEM and EDX Analysis

The SEM image of Co₃O₄ microspheres in Figure 3a and Figure S2 reveals significant aggregation, with particle sizes ranging from 200 nm to 500 nm, indicating a broad size distribution. The Co₃O₄/PC displays microfibers that range from around 100 nm to about 1000 nm (Figure 4a and Figure S3). Co₃O₄/NS presents nanoparticles in larger scale due to agglomeration of 2–5 µm Co₃O₄/NS spheres built from nanospheres of about 100 nm (Figure 5a and Figure S4). This analysis highlights the diverse morphology of Co₃O₄ without and with support material, which is crucial for understanding their potential applications in various fields such as catalysis, capacitance, and sensing. Using energy-dispersive X-ray spectroscopy (EDX), the Co₃O₄ composite was found to contain Co and O elements with the following percentage composition (weight %): 80.92% for Co and 19.08% for O (Figure 3b). The Co₃O₄/PC composite was found to contain Co and O elements with the following percentage composition: 73.13% for Co and 26.87% for O (Figure 4b). The Co₃O₄/NS sample was found to have percentage compositions of 76.07% and 19.84% for Co and O, respectively (Figure 5b). An additional peak was observed in Co₃O₄/NS for Si with a composition percentage of 4.09%, as illustrated in Figure 5b.

2.5. TEM Analysis

The Co₃O₄ sample displayed particles ranging from 4.28 nm to 19.96 nm (Figure 6a). In comparison, the Co₃O₄/PC sample showed particles ranging from 11.30 nm to 21.73 nm, with a graphite layer observed due to the presence of porous carbon (Figure 6b). Contrarily, the Co₃O₄/NS sample exhibited particles in a range between 6.06 nm to 13.48 nm, with a grey-white background attributed to the silica support (Figure 6c).

The percentage of Co₃O₄ nanoparticles versus their size was displayed in a size distribution histogram (Figure 7). For the three samples Co₃O₄ (Figure 7a), Co₃O₄/PC (Figure 7b), and Co₃O₄/NS (Figure 7c), it was discovered that many of the particles ranged in size from 5 to 70 nm.

2.6. XRD Analysis

Diffraction peaks for the three cobalt oxide samples were observed at about 31.28°, 36.75°, 44.82°, 51.40°, 59.27° and 65.21°, as illustrated in Figure 8. These can be attributed to the formation of Co₃O₄, consistent with reports in the literature.

Meng et al. observed that Co₃O₄ nanorods exhibited distinct X-ray diffraction peaks at 2θ = 19.08°, 31.3°, 36.8°, 44.7°, 59.7°, and 65.3°, which corresponded to the (111), (220), (311), (400), (333), and (440) planes of pure spinel Co₃O₄, in agreement with JCPDS No. 43-1003 (space group Fd-3m) (Figure 8) [29]. Richa Bhargava et al. reported sharp and strong diffraction peaks at 2θ values 18.73, 31.19, 36.86, 38.56, 55.66, 59.35, and 65.23, corresponding to the (111), (220), (311), (222), (400), (422), (511), and (440) planes, respectively [30]. These values are in good agreement with the standard XRD data for the cubic phase of Co₃O₄ (JCPDS no. 42-1467) [31]. Similarly, Xinmeng Zhang et al. found that five reflection peaks of Co₃O₄ appeared at 2θ values of 19.07°, 31.38°, 36.98°, 44.97°, 59.58°, and 65.45°, which could be assigned, respectively, to the (111), (220), (311), (400), (511), and (440) crystalline planes of cubic Co₃O₄ (JCPDS Card 03-065-3103) [32]. Additionally, Loukas Belles et al. illustrated the XRD patterns for the #Co1, #Co2, and #Co3 materials. For Co sample 1 and Co sample 2, the characteristic diffraction peaks at angles 19.1°, 31.3°, 36.8°, 38.8°, 44.6°, 55.8°, 59.5°, and 65.3° could be perfectly indexed to the (111), (220), (311), (222), (400), (422), (511), and (440) planes of Co₃O₄ (JCPDS card no. 75-2480), respectively. For Co sample 3, additional characteristic peaks at 36.7°, 42.64°, 61.71°, 74.03°, and 77.98° could be indexed to the (111), (200), (220), (311), and (222) planes of CoO (JCPDS card no. 14-0133), respectively [33].

2.7. Catalytic Study

The study examined the process of catalytic reduction of PNP by the metal-oxide nanocomposites Co₃O₄, Co₃O₄/PC, and Co₃O₄/NS by tracking the UV–visible absorption spectra of organic compound PNP in water using a hydrogen source, sodium borohydride (NaBH₄), and the nano-catalysts at various times, as described in the experimental section, at room temperature.

After adding NaBH₄, the p-nitrophenol (PNP) was rapidly converted to p-nitrophenolate ion, causing the peak at 317 nm to shift to 400 nm. As a result of the p-aminophenol (PAP) product forming, the absorbance peaks at 400 nm gradually diminished, and new peaks at 235 and 300 nm emerged. Figure 9, Figure 10 and Figure 11 shows that the nano-catalysts Co₃O₄, Co₃O₄/PC, and Co₃O₄/NS had durations of 4, 3, and 3 min, respectively, for the twentieth cycle. Plotting the ln A_t/A₀ data over time revealed that it represented the absorbance of the solution at time t (min), with A₀ representing the initial absorbance of the solution at the beginning. For instance, the slope of the straight line for several cycles was used to determine the first-order rate constant k (min⁻¹) for the metal-oxide nanocomposites Co₃O₄, Co₃O₄/PC, and Co₃O₄/NS, as illustrated in Figure 9, Figure 10, and Figure 11, respectively [34].

When the amount of PNP reactant had vanished, the cycle was finished. Then, after adding comparable numbers of organic molecules, a new cycle began.

Table 1. Catalytic activity: numbers of cycles, TON and TOF values, and kinetic rate constants of different nano-catalysts for reduction of PNP with NaBH₄ with (a) Co₃O₄, (b) Co₃O₄/PC, and (c) Co₃O₄/NS.

Catalyst	PNP mg	No. of Cycles	Duration of All Cycles (min)	First-Order Rate Constant k min−1 (R2)	TON mg PNP/mg Nano (mmol PNP/mg nano)	TOF (mg PNP/mg nano)/min (mmol PNP/mg nano)/min
Co3O4	1.31	46.70	152	0.4591 (0.9719)	3.27 (0.024)	0.0215 (1.5 × 10−4)
Co3O4/PC	1.48	52.67	178	0.4053 (0.9717)	3.67 (0.026)	0.0206 (1.4 × 10−4)
Co3O4/NS	1.54	54.22	218	0.3837 (0.9595)	3.79 (0.027)	0.0174 (1.2 × 10−4)

, Table S1, and Figure S5 show the outcomes of the catalytic reduction of PNP by the nano metal oxides Co₃O₄, Co₃O₄/PC, and Co₃O₄/NS. Each nano-catalyst’s first-order rate constant value was determined for each cycle. The average first-order rate constant was also computed in order to evaluate the representative catalytic efficiency. The average rate constant slightly dropped for the nano metal oxides Co₃O₄, Co₃O₄/PC, and Co₃O₄/NS in the catalytic reduction of PNP, with values of 0.4591, 0.4053, and 0.3837 min⁻¹, respectively, being obtained (Table 1 and Table S1). The extensive variation in particle diameters of catalysts observed in TEM did not track with changes in reaction rates or product yields, suggesting that particle size heterogeneity plays a secondary role compared to electronic properties and metal–support interactions.

Computed values for turnover number (TON) and turnover frequency (TOF) were also used to express the catalytic reduction efficiencies of nano metal oxides (Table 1 and Table S1). For the catalysts Co₃O₄, Co₃O₄/PC, and Co₃O₄/NS, the TOF for PNP reduction dropped to 0.0215, 0.0206, and 0.0174 (mg PNP/mg catalyst/min), while the TON for Co₃O₄, Co₃O₄/PC, and Co₃O₄/NS increased to 3.27, 3.67, and 3.79 (mg PNP/mg catalyst). Also, the numbers of cycles for reduction of PNP increased to 47, 53, and 55 cycles (Table S1, Figure 12). This indicated that the addition of PC and nano silica supports improved the durability of the catalyst, because the number of completed cycles increased compared to non-supported samples. In addition, the total amount of PNP converted to PAP increased in the catalysts with supports (TON), indicating a more profitable method.

2.8. Batch Chemical Catalytic Experiment and Regeneration

In a standard batch experiment, 0.5 mg of PNP was combined with 1 mg of catalyst (Co₃O₄, Co₃O₄/PC, or Co₃O₄/NS) in 5 mL of water (Figure 13a). The reaction’s development was tracked at various points in time (min). Then, with every subsequent cycle, 0.5 mg PNP was added and 15 mg of NaBH₄ was consumed in total. For Co₃O₄, Co₃O₄/PC, and Co₃O₄/NS, the total numbers of cycles attained for PNP reduction were 11, 14, and 13, respectively. TONs were 5.1, 6.7, and 6.3 mg of 4-NP/mg catalyst; total duration was 101, 200, and 192 min; and TOFs were 0.051, 0.034, and 0.032 (mg MO/mg catalyst/min).

The catalysts were regenerated for the first time by repeatedly rinsing them with methanol and water after the experiment was finished, and they were then utilized in another experiment. Following the initial regeneration, the corresponding numbers of cycles for Co₃O₄, Co₃O₄/PC, and Co₃O₄/NS were 10, 12, and 12; total duration times were 95, 176, and 183 min; TONs were 4.8, 5.8, and 5.7 mg of 4-NP/mg catalyst; and TOFs were 0.051, 0.033, and 0.031 (mg MO/mg catalyst/min) (Figure 13b). Following the first regeneration, the catalysts were washed with methanol and water multiple times for reactivation and use in a second-generation experiment. Co₃O₄, Co₃O₄/PC, and Co₃O₄/NS successfully completed 9, 11, and 11 cycles, respectively. TONs were 4.10, 5.10, and 5.05 mg of 4-NP/mg catalyst with total times of 110, 128, and 129 min, and TOFs were 0.037, 0.040, and 0.039 (mg MO/mg catalyst/min) (Figure 13c).

2.9. Mechanism of Reduction of PNP by Cobalt Oxide Nanocomposites

Our reduction experiment in an aqueous solution required self-generated hydrogen, which was produced by the hydrolysis of sodium borohydride. However, we found in our work that the reduction process of PNP in an aqueous solution containing sodium borohydride and a catalyst resulted in excessive generation of hydrogen bubbles, compared to a process involving an aqueous solution of sodium borohydride without a catalyst, demonstrating that the catalyst cobalt oxide catalyzed the generation of hydrogen molecules in the presence of sodium borohydride, probably through a hydrogen mediator (cobalt boride species) [35]. Furthermore, p-nitrophenol absorbed onto the surface of cobalt oxides by physical bonds, by chemical bonds through π-π interactions between their aromatic rings and the aromatic rings of porous carbon support in Co₃O₄/PC, and by hydrogen bonds between the hydroxyl group of PNP and oxygen atoms of silica support in Co₃O₄/NS.

We now outline the stepwise hydrogenation mechanism, highlighting how Co₃O₄ nanocomposites activate the hydrogen donor (NaBH₄) and shuttle hydrogen atoms to p-nitrophenol (PNP), in agreement with established findings in the literature, as follows: (1) Activation of hydrogen by Co₃O₄: On exposure to NaBH₄ in aqueous solution, hydride ions (H⁻) react to generate molecular hydrogen (H₂) and borate species. Co₃O₄ nanostructures—especially those with abundant defect sites and high redox activity—cleave H₂ heterolytically to form surface Co-H and Co-OH species. These surface hydrides act as the immediate hydrogen source for subsequent PNP reduction steps [36,37]. (2) Stepwise hydrogen transfer to the nitro Group: The reduction of the nitro moiety proceeds via a classic four-hydrogen addition pathway, as corroborated by DFT and experimental studies on similar oxide catalyst [38], as follows: (a) Nitro to Nitroso: Two hydrogen atoms transfer from Co-H to the –NO₂ group of adsorbed PNP, yielding a nitroso intermediate (–NO) with concomitant elimination of one H₂O molecule. (b) Nitroso to hydroxylamine: A further two-hydrogen transfer reduces the –NO to hydroxylamine (–NHOH). (c) Hydroxylamine to amine: Finally, two more hydrogen atoms add to –NHOH, accompanied by loss of H₂O, to furnish p-aminophenol (PAP) as the end product. This pathway mirrors the sequence outlined for copper oxide catalysts, supporting its validity for Co₃O₄ systems as well [39].

2.10. Comparative Studies

The present advanced study of cobalt oxide nanocomposites underscores the synergistic interaction between electroactive cobalt oxides and supporting materials such as porous carbon and nano silica. This integrated architecture significantly enhances catalytic activity and imparts competitive performance characteristics, making these nanocomposites promising candidates for advanced electrochemical and catalytic applications.

The catalytic performance of cobalt oxide nanocomposites was benchmarked against various reported cobalt catalysts (

Table 2. Comparison of the rate constant values obtained in the current work with previous reports.

Catalyst *	Rate Constant	Reference
quasi-CoMOF (TMU-10)	1.68 min−1	[40]
Co–NP	0.210 min−1	[41]
CoC-NC	0.128 min−1	[42]
Co3O4NC	0.0925 min−1	[21]
Co3O4NP (meso-Co-150)	3.20 × 10−4 mol m−2 min−1	[22]
Co3O4NP	0.46 min−1	Present work
Co3O4/PC	0.41 min−1
Co3O4/NS	0.38 min−1

* MOF: metal–organic framework; NP: nanoparticles; NC: nanocomposites, PC: porous carbon, NS: nanosilica

). The pseudo-first-order rate constant (k₁) for p-nitrophenol reduction catalyzed by Co₃O₄, Co₃O₄/PC, and Co₃O₄/NS averaged 0.42 min⁻¹. This value surpasses the k₁ reported for quasi-cobalt organometallic framework (quasi-CoMOF) [40], cobalt nanoparticles (CoNP) [41], cobalt–carbon nanocomposites (CoC-NC) [42], Co₃O₄ nanocomposites (Co₃O₄-NC) [21] and Co₃O₄ nanoparticles (Co₃O₄-NP) referred to (meso-Co-150) [22].

Thus, variations in catalytic efficiency stem from three principal factors: (a) Active site density: a greater number and optimal nature of surface active sites directly boost intrinsic catalytic activity. (b) Particle size and morphology: smaller particles with high surface-to-volume ratios increase the availability and accessibility of active sites. (c) Support material: supports such as graphitic matrices or silica that offer high surface areas and tailored pore architectures improve dispersion of the active phase and enhance catalyst stability.

3. Experimental Section

3.1. Chemicals

Sodium borohydride (NaBH₄) (98%) (Sigma-Aldrich, St. Louis, MO, USA), cobalt(II) chloride hexahydrate (98%, Panreact), N,N-Dimethylformamide (DMF, 99.9%, Panreact), para-nitrophenol p-NP (>99%, Sigma-Aldrich), and BTC (95%, Sigma-Aldrich) were used in the present study.

3.2. Characterizations

UV–visible absorption spectra were recorded using a UV–Vis spectrophotometer, (Shimadzu, UV 1800, Kyoto, Japan). The infrared spectra of compounds were measured in the range of 300–4000 cm⁻¹ using a Shimadzu 8300 FTIR Spectrophotometer (Kyoto, Japan). TEM transmission electron microscopy images were taken using JEOL-JEM-1011 (Tokyo, Japan). An SEM scanning electron microscope with EDS facility, JEOL JSM-6380 LA (Tokyo, Japan) was used to analyze the morphology of the samples. Powder X-ray Diffraction (XRD) patterns were performed with ARL EQUINOX 1000 (Thermo Scientific, Basel, Switzerland).

3.3. Synthesis of Cobalt Complexes

-

Synthesis of Co(BTC) complex:

The Co(BTC) complex was prepared following the procedure described in [43] with slight modification. First, 0.714 g (3.0 mmol) of CoCl₂.6H₂O was dissolved in 7.5 mL of water under continuous stirring. In a separate flask, 0.210 g (1.0 mmol) of benzene tricarboxylic acid (BTC) was dissolved in 5 mL of DMF under stirring. These solutions were mixed and then transferred to a stainless-steel autoclave with an inner Polytetrafluoroethylene PTFE chamber with a volume of 100 mL. The mixture was heated at 150 °C for two days. After that, the reaction mixture was cooled slowly to room temperature. Purple crystals were obtained, filtered, washed with DMF and distilled water, and then dried at 70 °C in a forced-air oven for 2 days. The yield was 0.34 g.

-

Synthesis of Co(BTC)/PC complex:

Co(BTC)/PC complex was prepared following a similar protocol to that used for Co(BTC) complex, but with the initial addition of porous carbon as support material. To prepare Co(BTC)/PC, the process began by dissolving 0.012 g (1 mmol) of porous carbon (PC) in 5 mL of water. This solution underwent sonication for 15 min to ensure uniform dispersion of the PC. After that, the procedure continued as described for the preparation of Co(BTC)/PC. The product was then cooled to room temperature and centrifuged. It was subsequently washed with water and ethanol. The final product was left to dry at room temperature for several days to obtain 0.35 g of gray powder.

-

Synthesis of Co(BTC)/NS complex:

A similar procedure to that used for Co(BTC)/PC was followed but with the addition of 0.03 g (0.5 mmol) of nano silica (NS) instead of 1 mmol PC. The yield of light-purple powder was 0.036 g.

3.4. Synthesis of Co₃O₄ and Nanocomposites

-

Synthesis of Co₃O₄: 0.1 gm of Co(BTC) complex in a crucible covered with a lid was treated in an air furnace oven at 425 °C for 1 h with a heating ramp rate of 10 °C/min, then cooled slowly to room temperature to obtain black powder of Co₃O₄ (yield 0.042 gm).

-

Synthesis of Co₃O₄/PC: 0.05 gm of Co(BTC)/PC complex in a crucible covered with a lid was treated in an air furnace at 425 °C for 1 h. After cooling to room temperature, a fine black powder (0.0147 gm) was obtained.

-

Synthesis of Co₃O₄/NS: 0.153 gm of Co(BTC)/NS complex in a crucible covered with a lid was treated in an air furnace at 425 °C for 1 h, then cooled slowly to room temperature yielding a fine black powder (0.05 gm).

3.5. Catalytic Reduction Experiment

Para-nitrophenol (PNP) catalytic reductions into para-aminophenol (PAP) were investigated in aqueous solution containing sodium borohydride (NaBH₄) as a hydrogen source. In the first cycle, 2.8 mL of distilled water in a tube was mixed with 0.4 mg of the catalyst and 4 mg of NaBH₄, followed by the addition and mixing of 0.028 mg of PNP, forming 0.072 mM solution. Only an additional, comparable amount of PNP was added in the subsequent cycle. The reduction reaction’s progress was assessed by measuring the absorbance of the UV–visible spectra at λ_max (i.e., 400 nm) of the formed p-nitrophenolate at specific time intervals. When the absorbance at the corresponding λ_max approached zero, the cycle was deemed finished.

4. Conclusions

This study demonstrates a straightforward, solvent-free pyrolytic route to synthesize cobalt oxide nanocomposites with finely tuned morphologies on porous carbon (PC) and nano silica supports (NS). The Co₃O₄/PC and Co₃O₄/NS materials exhibited distinctive microsphere, microfiber, and hierarchical nanosphere architectures confirmed by FTIR, UV–Vis, SEM, and TEM analyses, directly correlating structural features with catalytic performance.

The supported catalysts achieved exceptional p-nitrophenol reduction kinetics (k averaged 0.395 min⁻¹) and high turnover numbers (average 6.5 mg 4-NP per mg catalyst), surpassing many reported cobalt-based systems. Importantly, both Co₃O₄/PC and Co₃O₄/NS retained over 95% activity after multiple regeneration cycles, underscoring their durability and recyclability.

By combining easy synthesis, precise morphological control, and robust catalytic activity, these cobalt oxide composites offer a promising platform for scalable, regenerable catalysts in environmental remediation. Future work will explore reactor integration and broaden pollutant scopes to advance practical wastewater treatment applications.