Recyclable and efficient silica-supported copper hydrogenation catalyst Authors: Anjali Katkar, Krishnamay Pal, Ajit Garade, Anant R Kapdi Published: 2025 ABSTRACT AbstractMany hydrogenation reactions still depend on noble-metal catalysts, which limits their broader use because of high cost and supply concerns. Developing effective catalysts from abundant materials that operate under mild reaction conditions remains a significant challenge. Here, we show the preparation of a copper silicate catalyst using an ammonia evaporation method that delivers exceptional performance in converting nitroarenes to anilines. The catalyst maintains high activity and stability, providing yields above 99 percent under mild conditions. Structural and surface analyses reveal well-dispersed copper species whose combined contributions promote both hydrogen activation and substrate conversion. This behaviour enables a cost-effective and sustainable alternative to noble-metal catalysts. The material also demonstrates robust recyclability, retaining activity over repeated use, and scale-up tests confirm its practical viability. These results highlight the promise of copper-based systems for efficient and industrially relevant hydrogenation processes.Subject terms: Heterogeneous catalysis, Catalyst synthesis (Hetero)Aromatic amines are ubiquitous structural motifs that have been found extensively in a variety of natural products, pharmaceutical intermediates, agrochemicals, and dyestuff intermediates such as Amprenavir, Sparfloxacin, Acetaminophen, Iclaprim, Flutoanil, Formetanate, Aniline Yellow, and Bismarck Brown (Fig. 1)1–6. Several synthetic methodologies are known for accessing the amine functionality from easily available starting materials7. Buchwald–Hartwig amination8 under palladium-catalyzed conditions has proven to be an important C―N bond-forming technology to access (hetero)aromatic amines from chloro(hetero)arenes9. However, despite its success in academia for accessing aromatic amines commercially, it does exhibit limitations. Accessing (hetero)aryl amines via the reduction of the -NO2 functionality in nitro(hetero)arenes is, even today, the most preferred pathway10. The conventional methodologies for -NO2 group reductions involve precious metal catalysts such as Pd, Pt, Ir, or Ru-based metal-supported catalysts11–16. Although these catalytic systems are selective, they are costly, toxic, and prone to sintering, which limits their economic application2,3. One of the most commercially successful -NO2 reduction catalyst is the Raney nickel catalyst17 which is proven to be highly effective at lower concentrations. However, several disadvantages related to its high-temperature synthesis process, the pyrophoric nature of the catalyst, ease of deactivation via poisoning (due to the presence of sulfur in the substrate), and difficulty in storage and disposal have become the contributing factors for researchers to look for alternatives that could overcome the above-mentioned limitations. To minimize these problems, Beller and co-workers recently reported intermetallic nickel-silica-supported nickel silicide catalysts, which are effective and selective towards nitro reductions18. (Hetero)aryl amine functionality in pharmaceutical intermediates, agrochemicals, and dyestuff intermediates. The catalyst exhibited good substrate scope and was shown to overcome most problems associated with the Raney nickel catalyst. However, the synthetic preparation process of nickel silicide, which involves the thermal reduction of Si-O bonds at a temperature of 1000 °C (promoting the formation of a strong Ni-Si bond)18, makes it less attractive commercially. In recent years, catalysts with transition metals19 like copper metal catalysts have attracted increasing attention as earth-abundant, inexpensive, and relatively benign alternatives for nitroarene hydrogenation20,21. Among them, copper silicate systems stand out because the silicate matrix provides high surface area, tunable acidity, and enhanced dispersion of active Cu species, which together promote both thermal stability and redox versatility22,23. The cooperative interplay of Cu⁺/Cu⁰ species with the silicate support has been shown to facilitate efficient H₂ dissociation and selective substrate activation, thereby minimizing undesired by-products24. Furthermore, the immobilization of copper within the silicate framework prevents particle agglomeration, granting exceptional recyclability compared to unsupported or conventionally impregnated Cu catalysts2,25. Despite these advances, many reported Cu-based systems still suffer from limitations such as systems frequently demanding harsher hydrogenation reaction conditions like elevated temperature and hydrogen pressure, (≥150 °C, ≥1.3 MPa H₂), high catalyst loading, poor recyclability, and loss of selectivity upon extended use20,21,26. To maximize industrial benefits, a catalyst should exhibit high activity under mild conditions and demonstrate the capability for multiple reuses without a significant loss in activity. The present study directly addresses these shortcomings by introducing a highly active and recyclable copper silicate catalyst that enables nitroarene hydrogenation under mild conditions (70 °C, bar-level H₂ pressure, methanol solvent) with 100% selectivity to aniline. Remarkably, the catalyst sustains >99% yield over eight consecutive recycles. Importantly, the catalyst combines high stability, green operation, cost-effectiveness, and non-pyrophoric, distinguishing it from existing precious metal and conventional Raney Ni, Ni–silicide, and noble-metal benchmarks. This positions Cu–silicate as a practical, scalable, underscoring outstanding durability. Through detailed structural characterization (SEM, EDS, HRTEM & HAADF-STEM, XRD, XPS, TPR, ICP-AES, AES), we elucidate the role of Cu oxidation states and the silicate matrix in driving this unprecedented combination of activity, stability, and selectivity. Results and discussions The structural and morphological characteristics of the synthesized catalysts were comprehensively analyzed by subjecting the synthesized catalyst materials to different characterization techniques such as Thermogravimetric analysis (TGA), X-ray diffraction (XRD), Scanning electron microscopy (SEM) coupled with energy-dispersive X-ray spectroscopy (EDS), Transmission Electron microscopy (TEM), X-photoelectron spectroscopy, Auger electron spectroscopy, FTIR spectroscopy and ICP-AES study. Also, to study the reduction behavior of the catalyst, hydrogen desorption studies (TPR) were conducted. Fourier transform infrared spectral studies The synthesized catalysts were identified using ATR-FTIR spectroscopy to identify their chemical interactions and functional groups ranging from 400 to 4000 cm−1. The Fourier-transform infrared (FTIR) spectrum (Supplementary Fig. 2) of silica support exhibits several characteristic absorption bands. Amorphous silica support showed peaks at 1044 cm−1, 800 cm−1, and 470 cm−1 corresponding to Si-O-Si asymmetric stretching, Si-O-Si symmetric stretching, and bending vibration of Si-O-Si bonds, respectively27–31. The synthesized catalysts (Cu-dried, CuSi600Cal, CuSi600Red, CuSi800Cal, and CuSi800Red) also showed a similar type of IR peaks (Supplementary Fig. 2); however adsorption band shifted from 1044 cm−1 to 1110 cm−1 in the case of CuSi600Cal, CuSi600Red, CuSi800Cal, and CuSi800Red samples, indicating a strong interaction of copper metal and silica matrix after calcination and reduction29,32. Overall, the FTIR analysis supports the structural integrity of the silicate phase even after high-temperature processing and reduction. Powder X-ray diffraction study XRD patterns of the silica support and prepared Cu/SiO2 samples are shown in Fig. 2. Silica support (Fig. 2A) shows an XRD pattern corresponding to the amorphous phase of silica. The feature at 2θ of around 21.8° and 20.5° in the case of prepared copper silicate samples came from silica support after calcination. Powder XRD analysis. A Sodium silicate. B Calcined and reduced CuSiO2 catalysts, such as CuSi600Cal, CuSi600Red and CuSi800Cal, CuSi800Red. The diffraction peaks of CuO (Fig. 2 Image- B) appeared at 35.5° and 38.6° in CuSi600Cal and CuSi800 Cal, which reduced significantly after reduction in CuSi600Red and CuSi800Red samples. The CuO crystallite sizes of the CuSi600Cal (CuO, tenorite) are 11.9 nm33, and the CuSi600Red sample leads to the formation of Cu₂O and Si-Cu phases, accompanied by a significant decrease in crystallite size to ~6 nm based on the Halder-Wagner method (Supplementary Table 1)34,35. The broad diffraction peaks at 32.2°, 58.2°, and 61.6° in the calcined samples suggest the presence of copper phyllosilicate32. In the CuSi600Red sample, three XRD diffraction peaks appearing at 2θ = 43.3°, 50.4°, and 74.1° were indexed as the (1 1 1), (2 0 0), and (2 2 0) planes of the face-centered-cubic (FCC) copper, respectively36. No diffraction peaks of copper hydroxides at 11.1 were detected, indicating that phase-pure metallic Cu particles. The XRD peak at 36.5° in CuSi600Red reduced samples could be attributed to the Cu2O phase37,38. At 800 °C, calcination (CuSi800Cal) results in further phase evolution, leading to the coexistence of CuO (Tenorite) and Tridymite phase of silica exhibiting a Tenorite-like crystal structure39. Copper silicate catalyst calcined at 800 °C after reduction (CuSi800Red) leads to the emergence of metallic Cu (Copper, syn), while CuO and SiO₂ persist, suggesting incomplete reduction and increased sintering effects40. Crystallite size evolution follows distinct trends under varying thermal conditions, likely due to structural rearrangements associated with CuO reduction41. At 800 °C, calcination induces substantial crystal growth (24.5 nm), and subsequent reduction results in a slight increase in size (25.8 nm), implying sintering effects40. dominate at higher temperatures, outweighing any reduction-induced size decrease. Higher temperatures (800 °C) promote crystallite growth through Ostwald ripening42, whereas calcination up to 600 °C followed by reduction leads to size reduction due to CuO transformation to Cu₂O. The progressive reduction pathway (CuO → Cu₂O → Cu) is evident under increasing temperature and reduction conditions43. Additionally, silica undergoes phase transitions, transforming from Cristobalite to Tridymite at 800 °C, indicating structural rearrangements within the support44. Transmission electron microscopy studies Transmission Electron microscopy analysis was conducted using Tecnai G2 Spirit Biotwin model (make FEI) at 100 kV. For sample preparation, 2 mg of sample was taken and diluted with 4.0 ml of ethanol solvent. The diluted sample (0.5 µl) was pipetted out and drop cast onto a copper-carbon grid (200 mesh), where it was dried for 10 minutes under an infrared lamp. Figure 3 (Image 1) shows aggregated clusters made of uniformly dispersed nanoparticles in a CuSi600Cal catalyst. The micrograph (C) shows that these particles are primarily spherical to somewhat irregular in shape, and their wide diameters range from 9 to 20 nm. The darker contrast areas show well-dispersed Cu-rich phases like CuO, while the lighter matrix in which the particles are embedded is probably amorphous silica. Images (A), (B), and (D) show bigger agglomerates with denser internal contrast, which suggests cluster formation during calcination. In the reduced copper silicate catalyst (CuSi600Red), the copper particles appear intimately anchored to the silica matrix without signs of any void formation, highlighting strong metal–support interactions (Fig. 3, Image 2). Also, slight aggregation is noticeable in field views (A, B); it remains limited and does not compromise the overall dispersion. This microstructural arrangement suggests that the reduced catalyst maintains structural stability and an effective distribution of active sites. TEM & HRTEM-SEAD pattern of the synthesized samples. (1) A–D CuSi600Cal, (2) A–D CuSi600Red, The images illustrate the morphological differences between the calcined and reduced samples, with particle distribution and agglomeration varying as a function of calcination temperature and treatment conditions. In CuSi800Cal (Supplementary Fig. 3, Image 3), Sheet-like nanostructures with dispersed darker regions were observed in low-magnification TEM images (A–C), which are indicative of copper-rich domains featured in the silica matrix, also supported by the intensity and spacing of diffraction rings. The crystal size is 24 nm. In this, the high-resolution images show distinct lattice fringes. As calculated, the lattice spacings in particles range from 0.084 to 0.292 nm (0.84–2.92 Å). These values correlate with the (111), (200), and denote planes of copper oxide and silicate phases, and are consistent with size-dependent lattice behavior. A polycrystalline structure is indicated by the clear concentric rings shown in the selected area electron diffraction (SAED) pattern (D). In a reduced catalyst, which was reduced at 350 °C after being calcined at 800 °C, CuSi800Red (Supplementary Fig. 3, Image 4), shows dispersion of Cu-rich phases over the amorphous silica substrate is reflected in the material’s appearance as aggregated nanostructures. All particles are 25.8 nm in size. The average measured interplanar spacings fell between 0.0711 and 0.0140 nm (0.711 and 0.140 Å) (Supplementary Fig. 3, Image 4 D). These values correspond to insufficient reduction after calcination due to the sintering effect at high temperature. So, in conclusion, the superior activity of CuSi600Red arises from its smaller crystallite size and favorable copper chemical state. XRD shows that calcination at 600 °C followed by reduction yields finely dispersed Cu+/0 and Cu₂O phases (~6 nm), while calcination at 800 °C promotes crystallite growth (~24–25.8 nm), sintering, and incomplete reduction, leaving residual CuO and SiO₂. TEM corroborates this, revealing uniformly dispersed, well-anchored Cu nanoparticles in CuSi600Red, in contrast to the agglomerated, poorly dispersed Cu-rich domains in CuSi800Red. These structural and morphological differences highlight those smaller, well-dispersed copper particles with strong metal–support interactions in CuSi600Red provide more accessible active sites, thereby accounting for its higher catalytic activity compared to CuSi800Red. Thus, the combined XRD and TEM results clearly establish the structure-activity relationship between copper crystallite size, oxidation state, and catalytic performance. Scanning electron microscopy studies The CuSi600Cal exhibits a porous and aggregated surface morphology of irregularly shaped particles2, according to the SEM-EDS study (Supplementary Fig. 4). With no discernible contaminants, the EDS spectrum verifies that the primary elements like Cu, Si, and O were present in the prepared sample (supplementary Tables 2–5). In this, Copper species are evenly distributed on or within the silica support, most likely forming small oxide clusters or copper silicate phases. The CuSi600Red sample in (Supplementary Fig. 5) exhibits a highly porous and agglomerated structure consisting of tiny, sponge-like particles within an intact silicate matrix. It further suggests that even after reduction, copper is perfectly bound, preserving the silicate’s structural integrity. The SEM micrographs of CuSi800Cal (Supplementary Fig. 6) show a dense, coarse structure with well-defined, compact agglomerates that have a wrinkled and rough surface texture. This prominent surface topography is likely the result of high-temperature sintering, which causes densification of the catalyst particles and is expected to result in lower catalyst activity. The SEM micrographs show that the CuSi800Red catalyst has a densely sintered morphology with thermally fused particles (Supplementary Fig. 7). This may decrease surface accessibility. The SEM micrograph displayed in Image (A) has a rough, porous surface texture with big, clumped particles and irregular spaces. These structural characteristics suggest sintering and solidifying of Silica support. X-ray photoelectron spectroscopy study XPS profiles of the CuSi600Cal, CuSi600Red, CuSi800Cal and CuSi800Red performed to understand the electronic nature of Cu, Si, O. The deconvoluted spectra of Si 2p (Supplementary Fig. 9), shows an XPS peak at a binding energy of 103 eV, suggesting that surface Si species (SiO2) are in the silicate structure45. The O 1s spectra (Supplementary Fig. 8) in the supported document) of all samples deconvolute into two peaks at a low BE component (530.6–530.8 eV) assigned to Cu–Si–O species46 and a higher BE peak (532.6–532.8 eV) attributed to surface oxygen in CuO/SiO₂. Calcined samples show stronger oxide oxygen, while reduced samples exhibit enhanced Cu–Si–O contributions, confirming reduction-induced changes in oxygen environment. In all spectra, intense signals were observed in the 930–933 eV region47, which can be assigned to overlapping contributions of metallic copper (Cu⁰) or cuprous oxide (Cu⁺)45 or both in addition to copper oxide (Cu²⁺). The presence of these low-binding-energy components indicates partial reduction of CuO during calcination or reduction treatment. For the calcined CuSi600Cal sample (Fig. 4A), deconvolution reveals major contributions at 932.8 eV, attributed to Cu⁰/Cu⁺ species, and the presence of satellite at 940.5 eV characteristic of Cu²⁺. After reduction of the sample (CuSi600 Red, Fig. 4B), the Cu⁰/Cu⁺ feature remains dominant (932.3 eV), while the shake-up satellite at ~941.3 eV becomes less intense, confirming a decrease in Cu²⁺ content upon hydrogen treatment. XPS deconvolution profiles of Cu 2p and Cu LMM for the CuSiO₂-based catalyst samples. Panel (a) shows the Cu 2p spectra for (A) CuSi600Cal, (B) CuSi600Red, (C) CuSi800Cal, and (D) CuSi800Red. Panel (d) presents the corresponding Cu LMM spectra for (A) CuSi600Cal, (B) CuSi600Red, (C) CuSi800Cal, and (D) CuSi800Red. In contrast, the CuSi800Cal sample (Fig. 4C) displays peaks centered at 930.9 eV and 932.9 eV, along with a broad satellite at ~940.8 eV. This suggests a presence of Cu⁰/Cu⁺ species with a significant Cu²⁺ component. The persistence of the Cu²⁺ signal is consistent with stronger Cu–SiO₂ interactions formed during high-temperature calcination, which hinder complete reduction. Upon reduction (CuSi800Red, Fig. 4D), the spectra retain both Cu⁰/Cu⁺ signals (932.0 eV) and clear Cu²⁺ satellites at ~941.0 eV, indicating that a fraction of Cu remains oxidized even after hydrogen treatment. Finally, the shake-up satellite features between 940–942 eV, arising from the 2p → 3 d transition of Cu²⁺, are evident in all samples but show marked differences in intensity & they are more pronounced in the calcined samples (especially CuSi800Cal) and significantly attenuated in the reduced CuSi600Red, consistent with partial reduction of Cu²⁺ species. The Auger Electron Spectroscopy (AES) analysis was also conducted to differentiate various Cu species (Cu⁰/Cu⁺/Cu²⁺) present in the catalyst. The Cu LMM AES spectra (Fig. 5) and (Supplementary Table 6) showed the presence of Cu⁰/Cu⁺/Cu²⁺ species and their relative ratio, respectively in the catalysts. Peaks centered at ~918.5 eV correspond to metallic Cu⁰, while those at ~916–917 eV and ~914–915 eV could be assigned to Cu⁺ (Cu₂O) and Cu²⁺ (CuO), respectively,48,49. Relative ratios of Cu oxidation states derived from AES is presented in Supplementary Table 6. In CuSi600Cal, the Cu⁰/Cu⁺/Cu²⁺ ratio was 0.0/0.51/0.49, showing an almost equal distribution between Cu⁺ and Cu²⁺. Reduction to CuSi600Red significantly increased Cu⁰ species, with Cu⁰/Cu⁺/Cu²⁺ = 0.31/0.36/0.33, consistent with the attenuation of the Cu²⁺ XPS satellite. For CuSi800Cal, the Cu²⁺ and Cu⁺ state overwhelmingly dominated (Cu⁰/Cu⁺/Cu²⁺ = 0.0/0.50/0.50), in line with the CuO/Cu2O reflections in XRD. Even after hydrogen treatment, the CuSi800Red sample retained substantial Cu²⁺ and Cu⁺ contributions (0.0/0.50/0.50), indicating difficulty in reduction due to strong metal support interaction45,50. H2-TPR and H2-adsorption/desorption analysis of the calcined sample. H₂-TPR and H₂-adsorption/desorption profiles of the calcined CuSiO₂-based catalysts. The reduction and adsorption/desorption behaviors of the CuSi600Cal (A) and CuSi800Cal (B) calcined catalysts are compared to evaluate differences in surface hydrogen uptake and reducibility. Inductively coupled plasma-mass spectrometry (ICP-AES) The elemental composition of the catalyst samples was determined and captured (Supplementary Table 7) using Inductively Coupled Plasma Atomic Emission Spectroscopy (ICP-AES). Approximately 0.1 g of each catalyst sample was accurately weighed and subjected to acid digestion. The digestion process involved treating the solid material with a mixture of concentrated nitric acid (HNO₃) and hydrochloric acid (HCl) in a 3:1 volume ratio (aqua regia). The resulting clear solution was analyzed using an ICP-AES instrument. Calibration was carried out using certified standard solutions of copper in the concentration range of 0.1–10 ppm. Each measurement was performed in triplicate to ensure reproducibility and accuracy. The detection wavelength for copper was set at 324.754 nm. The copper loading on the SiO₂ support was in the range of 21–23 wt%, which closely aligns with the theoretical loading of 20 wt%, indicating efficient deposition and minimal loss during synthesis. Hydrogen desorption studies TPR The TPR profile, which shows the reduction behavior of catalysts designated as CuSi600Cal and CuSi800Cal, is shown in Fig. 5, images A & B, respectively. CuSi600Cal catalyst shows a sharp reduction peak around 250–270 °C. The occurrence of this sharp peak at relatively low reduction temperature indicates the uniform distribution of easily reducible CuO/Cu2O species36,47,51–53. This also indicates that the copper species are well dispersed and interact relatively weakly with silica support, making it easier to reduce. In contrast, CuSi800Cal shows a broad and low-intensity peak compared to CuSi600Cal. This suggests that copper species in the calcined catalyst at 800 °C are more difficult to reduce, which could be due to stronger interactions between copper/cupric oxides and the silica support. Secondly, high calcination temperature can also lead to sintering of copper species, leading to larger CuO/Cu2O crystallites and thus making it slightly difficult to reduce compared to CuSi600Cal. These distinctions arise primarily from differences in calcination temperature, which influence copper particle growth, dispersion, and interaction with the silica matrix, and restrict the development of catalytically active Cu⁰ species. The improved performance observed with CuSi600Cal than CuSi800Cal after reduction could be attributed to the presence of easily reducible copper species in the temperature range of 250–270 °C. Catalytic activity evaluation The catalytic performance was evaluated using nitrobenzene as a probe molecule. The reactions were carried out using an SS CAT-7 reactor (supplementary Fig. 49) with small glass tubes covered with Teflon cork, each having a capacity of 10 ml. Typically, 61.5 mg of nitrobenzene and 45 mg of the catalyst were suspended in 2 mL of methanol. The reaction was maintained at 60 °C for 10 h under constant stirring at 300 RPM. Post-reaction, the mixture was cooled, and the catalyst was separated by filtration. The reaction mixture was analyzed using gas chromatography GC, YL-6500 with a flame ionization detector. Methanol was used as an internal standard. The column used for the study was a Restek MXT®-1 column (capillary column) of 30 m, and nitrogen was used as a carrier gas. Evaluation of different catalysts for nitrobenzene hydrogenation The performance of various catalysts (CuSi600Cal, CuSi600Red, CuSi800Cal, and CuSi800Red) were evaluated for the catalytic hydrogenation of nitro aromatic compounds in MeOH under a H2 gas pressure of 10 kg at 70 °C using a catalyst concentration of 0.023 g/cm3 (Supplementary Fig. 10). Among the various catalysts, CuSi600Red catalyst showed excellent activity for the hydrogenation of nitrobenzene, achieving 100% selectivity and over 99% isolated yield of aniline. Its smaller crystal size of around 6 nm and the uniform distribution of copper within the silica matrix contributed significantly to its performance. Furthermore, the presence of easily reducible copper species (CuO/Cu₂O) in conjunction with a favorable ratio of Cu⁰, Cu⁺, and Cu²⁺ species (0.31:0.36:0.33) made CuSi600Red catalyst more active compared to the remaining catalysts54,55. It is well reported in the literature that Cu⁰ species serve as active centers for hydrogenation reactions, while Cu⁺ and Cu²⁺ act as Lewis acid sites, facilitating substrate activation and polarization. In contrast, the CuSi600Cal catalyst displayed negligible activity due to the absence of Cu⁰, having only Cu⁺ and Cu²⁺ species in a ratio of 0.51:0.49. Similarly, CuSi800Cal also lacked Cu⁰ and exhibited a higher crystal size (~25 nm) attributed to higher calcination temperatures, leading to significant particle sintering. These factors inhibited the formation of active Cu⁰ even after hydrogen treatment, leading to negligible catalyst activity. The superior performance of CuSi600Red over CuSi800Red can also be explained by the nature of their respective copper species after treatment. The more easily reducible copper oxides found in CuSi600Red, compared to the more inert species in CuSi800Cal suggest that temperature and calcination conditions critically affected the catalysts’ ability to generate active forms of copper species. The performance of CuSi₆₀₀Red catalyst is also compared with the Raney Ni catalyst, which is widely used worldwide and is also the leading catalyst in the chemical industries. Under identical reaction conditions, Raney Ni which used as the reference catalyst achieves complete conversion within 6 h compared to CuSi₆₀₀Red catalyst (Supplementary Fig. 11), which took 10 h to achieve similar conversion (Supplementary Table 8). However, CuSi600Red catalyst offers a safer and greener alternative to the conventional Raney Ni catalyst, which poses a safety problem due to its difficulty in handling, instability and pyrophorocity (8–10%). In terms of turnover number (TON), CuSi₆₀₀Red exhibited a cumulative TON of 31 after eight consecutive reaction cycles without any measurable decline in catalytic performance. In comparison, Ni silicide catalysts reported in the literature¹³ achieved a TON of ~25, though their synthesis required temperatures as high as 1000 °C, thereby limiting their industrial scalability and energy efficiency. By contrast, commercial Raney Ni displayed a relatively modest TON of ~3.3 under identical reaction conditions, likely due to its partial site passivation in liquid-phase hydrogenation. Beyond performance, CuSi₆₀₀Red has excellent thermal and chemical stability, maintaining structure and oxidation states even after multiple recycles. This combination of high catalyst activity, operational safety, and recyclability underscores CuSi₆₀₀Red as a robust and sustainable hydrogenation catalyst for selective transformations. Since CuSi600Red showed excellent activity compared to other catalysts, further studies were conducted using this catalyst. Effect of catalyst concentration The effect of catalyst concentration on the activity-selectivity pattern was studied using CuSi600Red catalyst (Fig. 6). The conversion of nitrobenzene increases from 21.3% to 100%, and selectivity remains almost constant (100%) with an increase in catalyst concentration from 0.0025 to 0.023 g/cm3. The conversion was increased from 21.3% to 92.7% with an increase in catalyst concentration from 0.0025 to 0.15 g/cm3, thereafter, it increased marginally from 92.7% to 100% with a further increase in catalyst concentration from 0.15 to 0.023 g/cm3. relatively low. The catalyst concentration of 0.023 g/cm³ was found to be optimum for achieving complete conversion and product selectivity. Catalyst concentration optimization. Reaction condition as given below : 0.5 mmol Nitrobenzene, 0.023 g/cm³ of catalyst (CuSi600Red), 2.0 mL of MeOH, 70 °C, 10 h, 10 kg H₂, 300 RPM. Effect of solvent on catalyst activity The polarity of the solvent highly influenced catalytic performance. In methanol, a highly polar solvent, the catalyst (CuSi600Red) exhibited outstanding efficiency, achieving a complete conversion and aniline selectivity (Supplementary Fig. 12). Because the solvation of reactants and intermediates could be better, allowing efficient access to the catalyst’s active sites. Conversely, in aprotic polar solvents such as ethyl acetate, acetonitrile, and non-polar solvents like hexane, isopropyl alcohol, the catalyst showed no or nearly less activity, which could be due to the insolubility of reactants and as well as hydrogen gas. Also, in terms of ethanol solvent, even though it is polar due to its high boiling point than the reaction conditions are not suitable for the reaction. In a water-methanol system (50:50), the catalyst showed 65% conversion with complete aniline selectivity. The lower conversion with the water-methanol system could be attributed to the competition of water molecules with nitrobenzene for active catalytic sites, as nitrobenzene is insoluble in water. So, the combined effects of solvent polarity, catalyst-solvent interactions, adsorption dynamics, and hydrogen solubility in solvents might affect the catalyst’s activity-selectivity in terms of solvent choice. Substrate scope for the reduction of nitro to amino group using CuSi600Red After a detailed study of the characterization of the catalyst and the suitable protocol for converting nitrobenzene to aniline, we focused on the catalyst’s (CuSi600Red) efficiency in reducing substituted nitroarenes into aromatic amines (Fig. 7). We performed a substrate scope on the catalyst under the conditions mentioned below. Different nitroarenes were taken as starting materials to check the reduction efficiency of the catalyst (CuSi600Red). We measured the after-column yields, varying from 35% to 97%. For the product 2b & 2m, due to the OH group present in the ortho & para positions respectively, it increases the electron density by conjugation to the carbon attached with the NO2 group and reduces the binding efficiency of the molecule with the catalyst surface, and hence the reaction yield also decreases. The same kind of effect can be seen in the case of 2l as well, due to the presence of an amino group at the ortho position. But for ortho methyl product (2c & 2e), electron donation due to hyperconjugation is less effective, so it does not show a significant effect towards catalyst binding as well as the reaction yield. Another major effect that decreases the catalyst’s reactivity is the substrate’s bulkiness. In the case of nitropyrene to aminopyrene (2n), the bulky pyrene group shows resistance to binding with the catalyst surface, as a result, the reaction yield is the least. So, from various substrate scopes, we concluded that for better reactivity, the carbon atom attached to the NO2 group should be electron-deficient, and the substrate should not have bulkiness. Substrate scope for nitroarene reductions using CuSi600Red catalyst. Reaction condition: 0.5 mmol Nitroarene, CuSi600Red (0.023 g/cm³), H₂ (10 bar), 70 °C, 10 h, 2 ml of MeOH (GC yield (%); Isolated yield (%). a : Reaction temperature 95.5 °C. Scale-up reactions To evaluate the scalability and efficiency of the optimized copper silicate catalytic system (CuSi600Red), a hydrogenation reaction was performed using nitrobenzene as the substrate under standard reaction conditions involving the use of 67 mL of methanol for dissolving, 2200 mg (17.9 mmol) of nitrobenzene, which was then hydrogenated with 1600 mg of CuSi600Red catalyst in this scale-up investigation (Supplementary Scheme 1). The reaction was carried out for 10 h at 70 °C and an initial hydrogen pressure of 10 kg. The process delivered 99.0% isolated yield of aniline as product, corresponding to 1750 mg (18.2 mmol), while leaving only 0.2% (4.4 mg or 0.036 mmol) of unreacted nitrobenzene. These numbers show outstanding selectivity toward the intended amine product and almost complete conversion. Catalyst reusability assessment The durability and reusability of the prepared 20 wt% copper on silica catalyst (CuSi600Red) was evaluated by recovering the catalyst after each reaction cycle. The used catalyst was filtered, thoroughly washed with methanol after each reaction cycle, and freed from the solvent using a centrifuge. The catalyst was then utilized in further catalytic cycles under the identical reaction conditions. A slight loss of 2–3% might be possible during decantation after centrifugation; however, no additional fresh catalyst was introduced. The catalytic activity and selectivity were measured over eight recycles to determine the catalyst’s stability and lifespan. The catalyst was successfully reused up to the 8th recycling (Fig. 8). Recyclability studies. Reaction conditions: 0.5 mmol Nitroarenes, 45 mg of CuSi600Red, 2 ml of MeOH, 70 °C, 10 h, 10 kg H₂ P, 300 RPM. Characterization study of the spent catalyst Adsorption study of the spent catalyst of CuSi600Red was done using an ATR-FTIR instrument (Supplementary Fig. 13). After 8th recycles, the catalyst was washed with methanol solvent twice and centrifuged to separate excess solvent. FT-IR spectra of the recycled catalyst showed an FT-IR pattern similar to fresh catalyst, showing a sharp band at 470 cm−1 corresponding to the bending vibration of Si-O-Si bonds and the most intense band at around 1092 cm−1 ascribed to the Si-O asymmetric stretching vibration of silicate tetrahedrons. The remaining bands at 800 cm−1 correspond to the symmetric stretching vibrations of silicate chains, further confirming that the SiO₂ network remains structurally intact after repeated use45. Comprehensive post-reaction analyses establish that the catalyst framework and copper active sites remain largely intact after recycling. XRD patterns of the recycled spent CuSi600Red catalyst preserve the characteristic reflections of metallic Cu at 43.3° (111), 50.4° (200) and 74.1° (220), identical to the 20 wt% Reduced CuSi₆₀₀Red phase, with no detectable peak shifts or new phases (Supplementary Fig. 14). TEM micrographs (Supplementary Fig. 15) confirm retention of the nanostructured morphology with a modest increase in crystalline size from ~6 nm to ~16–17 nm, however, no drop in activity was observed up to 8th recycles. Surface-sensitive analyses provide further evidence of the redox stability of copper sites. XPS analysis (Fig. 9C) of the Cu 2p region of recycled catalyst demonstrates the presence of mixed Cu⁰/Cu⁺/ Cu2⁺ species similar to fresh reduced catalyst, while Cu LMM Auger spectra (Fig. 9D) confirm an unchanged Auger parameter relative to the fresh reduced sample, affirming stability of the reduced copper states. These structural and spectroscopic findings are consistent with the catalytic results45,48, where complete selectivity is retained over eight cycles and the yield decreases only marginally to 93% in the ninth run with full selectivity retained50,56. Taken together, these results demonstrate that despite slight particle growth, the crystallographic framework and active copper states are preserved under turnover, ensuring long-term functional stability of the catalyst. XPS and AES analyses of the fresh and spent CuSi600Red catalyst. C Deconvoluted XPS spectra of the spent CuSi600Red catalyst. D AES analysis comparision of fresh CuSi600Red catalyst (A) and the spent CuSi600Red catalyst (B). Conclusion The requirement for a carbon-free catalyst alternative to reduce nitroarenes to anilines has both academic and industrial interest. Replacing precious transition metals with earth-abundant and readily available ones could further enhance the appeal of the reduction catalyst. We herein report a carbon-free copper silicate hydrogenation catalyst for reducing substituted nitroarenes to anilines under relatively lower temperature and H2 gas pressure than those reported in the literature. The developed catalyst has been characterized using various techniques such as powder XRD, SEM, TEM, XPS, AES, FTIR, ICP-AES, and TPR. The derived catalyst was tested against a wide range of nitroarenes, providing excellent yields in most cases of the reduced amine products. Recyclability of the copper silicate has also been examined, with the catalytic activity being retained over 8 recycles. The scale-up possibility was also tested for nitrobenzene reduction, providing a quantitative aniline product. Methods Chemical requirement The following reagents were procured from different vendors and used as such: Copper nitrate trihydrate Cu(NO3)2·3H2O (purity >98%, Sigma-Aldrich), Sodium Silicate (Assay: 22.44%, commercial source), 25% Ammonia solution (Loba Chemie). Catalyst synthesis Here, we synthesized a copper silicate catalyst through a meticulously controlled Ammonia Evaporation-deposition method. In this method, first, 15.25 g of Cu(NO3)2·3H2O (0.0631 mol) was dissolved in 150 mL of deionized water in a round-bottom flask at ambient temperature to obtain an aqueous solution. To this solution, 38.2 mL (34.3 g, 0.5048 mol) of 25% aqueous ammonia was added and stirred for 30 min. The initial pH of the solution was between 11–13. Then, 62.5 g of sodium silicate was added to the copper ammonia complex solution, and the resultant suspension was subjected to continuous stirring for 4 more hours at ambient temperature using a mechanical overhead stirrer (Ika-Nanostar 7.5 digital) at 500 RPM. After stirring for 4.0 h, the suspension was transferred to an oil bath preheated at 75–80 °C to allow for the evaporation of ammonia and the reduction of slurry pH to ~8.15–8.30, for the deposition of copper species on silica. Once the pH of the suspension had been decreased to ~8.15–8.30, the evaporation process was terminated. The reaction mass was filtered, and the obtained cake was washed with 500 mL of deionized water (5 times) and dried at 120 °C overnight in a static oven. The dried sample was crushed and homogenized. It was further subjected to thermogravimetric analysis to ascertain the stability of the synthesized material before it could undergo calcination and reduction. Supplementary Fig. 1 depicts the temperature analysis of the dried sample involving the TGA studies on the synthesized 20 wt% Cu on silica catalyst. The study was conducted by subjecting 10 mg of the dried catalyst sample to TG analysis, with the degradation pattern of the material monitored over a temperature range of 35–700 °C at a ramping rate of 10 °C/min. The endothermic mass loss below 200 °C is about 2.0–2.7%, which could be due to the loss of residual water or moisture that was adsorbed on the surface of the catalyst. The TG curve of the synthesized dried sample shows weight loss of 4.1% in the temperature range between 200 °C and 635 °C, suggesting the dihydroxylation of structural OH groups and the possible presence of a phyllosilicate phase. As per literature reports, the natural phyllosilicate chrysocolla decomposes slowly by eliminating hydroxyls, resulting in the exothermal production of crystalline CuO between 600–680 °C57. It has also been discussed that the calcination temperatures, when applied below 330 °C, resulted in no dehydroxylation of the silica-supported chrysocolla-like structure. Accordingly, TGA performed on the synthesized sample also did not detect any distinct stages for the decomposition of copper–ammonia complexes between the temperature range of 300–450 °C. Hence, the above TGA result suggests that the synthesized copper silicate material exhibits thermal decomposition behavior like that of a chrysocolla-like phyllosilicate phase. The fascinating aspect is that such material was obtained by the ammonia evaporation method using a commercially available and cheap silica support with a relatively low specific surface area at 75 °C within 8–9 h. After analyzing the dried sample using TGA, it was further subjected to calcination at 600 °C for 4 h in a muffle furnace (CuSi600Cal), followed by reduction. The calcined sample was reduced by placing 1 g of the calcined catalyst in a ceramic boat and subsequently pressuring the tube reactor to 50 bar on the hydrogen and nitrogen lines, respectively. The temperature of 350 °C in the tube furnace was achieved in 35 min by ramping it at 15 °C/min. The calcined catalyst was reduced in a tube reactor under a 50:50 H₂:N₂ gas atmosphere for a period of 4 h and finally cooled to room temperature, providing the reduced catalyst (CuSi600Red). Figure 10 shows the formation of olive green, or light green colored dried catalyst samples, often attributed to the presence of copper phyllosilicates. After calcination at 600 °C, the color of the catalyst material changed from olive green to dark gray, indicating the possible presence of CuO with some degree of aggregation in the silica matrix. Further reduction of the calcined material provided dark-colored catalyst material. Similarly, we have calcined the dried material at 800 °C (CuSi800Cal) and reduced it at 350 °C for 4 h in a 50:50 H₂:N₂ gas atmosphere (CuSi800Red). Synthesis of calcined and reduced Cu/Silicate catalyst. Left to right are dried uncalcined catalyst, calcined catalyst, and reduced catalyst. Top row : calcination at 600 °C; Bottom row: calcination at 800 °C. Reduction performed at 350 °C under 50:50 H2:N2 for 4 h. Supplementary information Transparent Peer Review file Supplementary information Acknowledgements This work was supported by the Institute of Chemical Technology, Matunga, with the collaboration of Evonik Catalyst India Pvt. Ltd, Dombivli, India. We gratefully acknowledge the support of the IIT Bombay SAIF facility for instrumental analysis. The authors would also like to thank Prof. Kenji Hara for technical discussions. Author contributions Conceptualization: A.R.K., A.G., and A.K.; A.K. and K.P. conducted the experiments; A.K. and K.P. analyzed the results and characterized the compounds; A.K. and K.P. wrote the manuscript and was edited by A.R.K. and A.G.; All authors have reviewed the manuscript. Peer review Peer review information Communications Chemistry thanks the anonymous reviewers for their contribution to the peer review of this work. A peer review file is available. Data availability The data that support the findings of this study are available from the corresponding author upon reasonable request. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Supplementary information The online version contains supplementary material available at 10.1038/s42004-025-01867-7. References Associated Data Supplementary Materials Transparent Peer Review file Supplementary information Data Availability Statement The data that support the findings of this study are available from the corresponding author upon reasonable request.