The catalytic degradation of methylene blue (MB) was conducted to evaluate the oxidative performance of the self-supported Ag(0) catalyst, compound 2. In a typical experiment, 50 mL of an aqueous MB solution (10 mg/l) was prepared and transferred into a 100 mL beaker. To this, 30 mg of the Ag(0) catalyst was added under constant magnetic stirring at room temperature. Subsequently, 1 mL of freshly prepared H2O2 (5 wt%) was introduced to initiate the reaction. The reaction mixture was kept under continuous stirring, and aliquots (3 mL) were withdrawn at regular time intervals (0, 5, 10, 15, 20, and 30 min). Each aliquot was immediately centrifuged at 5000 rpm for 5 min to remove catalyst particles. The clear supernatant was analyzed using UV-Vis spectroscopy by monitoring the absorbance at 664 nm, the characteristic absorption peak of MB. Control experiments were also performed in the absence of either the catalyst or H2O2 to confirm the role of each component in the degradation process. Additionally, radical scavenger experiments were conducted to investigate the active species involved in the catalytic mechanism.
The hydrothermal synthesis of Ag(I)-imidazolate coordination polymer, 1, utilized the affinity of the soft Lewis acid Ag(I) and soft Lewis base (imidazole), where Ag-N coordination interactions are expected to result in an extended network solid insoluble in the reaction medium. The scanning electron microscopy (SEM) images of 1 and its reduced form, 2, are shown in Fig. 1. Due to favourable solid-solid interactions, the resulting platelets of 1 self-assembled into microspheres of regular and uniform shape and size, Fig. 1(a). The Ag(I) ion within the coordinated structure in the microspheres of 1 can easily be reduced upon treatment with an aqueous solution of ascorbic acid, resulting in the reduced form maintaining the same size and morphology, 2 Fig. 1(b). Elemental mapping of C, N, and Ag was also recorded using energy dispersive X-ray analysis (EDX), where the elemental maps of the three selected elements trace the SEM images of 1 and 2, indicating uniformity of the Ag element in both samples, inserts in Fig. 1.
From Fig. 1(b), no Ag(0) micro-aggregates or micro-clusters are observed on the particle surface of 2. This suggests that no significant leaching of Ag(I) from the structure of 1 occurred during the reduction process. Consequently, the reduced silver did not materialize as surface-deposited Ag(0), supporting the feasibility of an alternative reduction mechanism. Specifically, the reduction likely proceeded via in-situ conversion of Ag(I) nodes within the pristine structural matrix, preserving their original coordination environment, and/or through the formation of internal Ag(0) nano-clusters within the framework of 2. In either case, the well-distributed metallic silver within 2 enhances its potential as a catalytic material for H₂O₂ reduction.
Figure 2 shows HRTEM images of compounds 1 and 2. Image 1 (pristine CP) shows a uniform distribution of dark spherical spots of similar contrast, which indicates these spots are of a specific moiety type. These spots are likely Ag(I)-rich domains or possibly lightly crystalline 1 regions. Also, the absence of lattice fringes suggests limited crystallinity or high dispersion of Ag(I) in the polymeric matrix. The soft contrast is consistent with apparently Ag(I) ions uniformly distributed/coordinated within the framework. On the other hand, the image of 2 (reduced coordination polymer) shows strong contrast with larger and darker spots. Clearly, some of these spots show obvious crystalline boundaries or lattice planes, which strongly indicate the formation of metallic Ag nanoparticles, Ag(0). The increased darkness strongly suggests partial reduction of Ag(I) to Ag(0). As a conclusion, the changes between the two images suggest that the reducing agent induced partial reduction of Ag(I) within the coordination polymer matrix to metallic Ag nanoparticles. This supports the proposal that the coordination polymer acts as an in-matrix for metallic nanocluster formation.
FTIR analysis was employed to further characterize compounds 1 and 2 (Fig. 3). Both compounds exhibited characteristic absorption bands at 1461 and 1475 cm, corresponding to C = N stretching, along with a sharp absorption at 3113 cm, characteristic of sp C-H stretching within the imidazolate ring. Notably, the FTIR spectra of the pristine compound 1 and the mildly reduced compound 2 are nearly identical, suggesting that the overall network of the pristine solid remains intact upon mild reduction. This preservation is likely attributed to minimal alterations in the coordination environment of the Ag centers, with the reduction of Ag(I) ions occurring primarily at the crystal surface rather than penetrating deeply into the bulk material. The compound 2 also shows good stability after 6 cycles of HO decomposition.
Raman is a powerful tool to investigate CPs and MOFs, and hence was applied to compounds 1 and 2. Figure 4 shows Raman spectra of imidazole and compounds 1 and 2. Regarding the crystallinity range, (0-300 cm), the imidazole spectrum has one strong peak at 121.5 cm and one weak peak at 139.5 cm. The compound 1 spectrum shows two strong peaks at 124.5 and 165 cm which indicates successful synthesis of compound 1 from imidazole. The fingerprint range (300-1800 cm) also indicates the successful synthesis of compound 1 from imidazole while conserving the imidazole structure. This is clear from the observed shifts of all peaks of imidazole in this range upon forming compound 1. The extended range (1800-3500 cm), more obviously, confirms the successful synthesis of compound 1 from imidazole. The characteristic strong peaks of imidazole at 3129 and 3146 cm completely vanished upon formation of compound 1. Finally, the spectrum of compound 2 is very similar and almost identical to that of compound 1, indicating that compound 1 withstands stable ascorbic acid reducing treatment, preserving structure and functionalities upon conversion into compound 2.
Figure 5 presents the XRD patterns of compounds 1 and 2, both exhibiting sharp peaks indicative of well-defined crystalline structures. The remarkable similarity between the two patterns suggests that the reduction treatment did not disrupt the ordered arrangement of silver atoms and imidazole moieties within the material's matrix. Furthermore, the absence of diffraction peaks corresponding to metallic Ag particles in the pattern of compound 2 implies that the reduction process occurred in highly localized regions, likely confined to the silver coordination nodes and/or the formation of Ag(0) nano-clusters within the structure.
All the above analyses suggest that compound 2 kept the morphological and topological natures of compound 1 upon reduction. The only possible change is the reduction of Ag(I) ions into Ag(0) atoms, preserving the coordination position and environment. The X-ray photoelectron spectroscopy may help support this proposal, and XPS analysis of 1 and 2 is shown in Fig. 6. The Ag 3 d line spectra indicated reduction of Ag(I) to Ag(0), the Ag 3d Line shift from 369.27 eV to 371.71 eV, a shift of + 2.44 eV, which is significantly greater than upshifts commonly recorded for the change of oxidation state. Additionally, the peak width increased upon reduction of Ag(I) to Ag(0). These two observations pointed out to reduction of Ag(I) to Ag(0) as well as the concomitant formation of ultra-small cluster size as shown in Fig. 6A.
For more explanation, the observed + 2.44 eV upshift in the Ag 3d/ XPS signal, from 369.27 eV in the pristine CP to 371.71 eV after mild reduction with ascorbic acid, deviates significantly from the conventional downshift typically associated with the reduction of Ag(I) to metallic Ag(0). This unusual behaviour can be rationalized by considering that both the initial and reduced CPs consist of nanoscale silver clusters embedded within an imidazole-based coordination polymer matrix. For compound 1, Ag(I) ions are strongly coordinated to nitrogen donors of the imidazole ligands, which can induce abnormally high binding energies due to electron withdrawal and poor core-hole screening. Upon reduction, metallic Ag(0) clusters are formed, but remain confined within the coordination polymer matrix as shown in Fig. 6A. These Ag(0) nanoclusters do not behave like bulk metallic silver; instead, they are subject to quantum confinement and limited electronic screening. These effects can lead to increased binding energies, occasionally even exceeding those of their oxidized counterparts. Therefore, the unexpected upshift is consistent with the formation of ultrasmall, matrix-confined Ag(0) clusters materials matrix.
As previously outlined, the screening of core holes in ultra-small metal nano-clusters becomes limited, which results in an upshift of core-level binding energy. In agreement with this proposed assignment, the N1s lines recorded for 1 and 2, Fig. 6b and c, respectively, indicated two distinctive chemical environments ascribed for pyridine-like N (399 eV) and pyrrole-like N (401.9 eV), with the later more pronounced in 2, indicating cleavage of some Ag-N bonds upon Ag(I) reduction with ascorbic acid.
To probe the catalytic activity of the self-supported Ag(0) catalyst, different masses of 2 were introduced into HO solution, and the remaining concentration of HO was monitored through iodometric titration, Fig. 7(D) demonstrating pseudo-first-order kinetics. A plot of initial reaction rate versus catalyst loading demonstrated a Linear behaviour, characteristic of surface-catalyzed reactions. Three different experiments utilizing catalyst loading of 10, 20, and 30 mg into 5,000 ppm HO solution were utilized, where the three reactions converged to full decomposition of HO at 60, 50, and 40 min, respectively, demonstrating the effective nature of the self-supported Ag(0) catalyst as shown in Fig. 7 (A, B, C). As a control experiment, compound 1, containing only the Ag(I) species, did not produce notable activity towards decomposition of HO even after 70 min of contact time with HO, confirming the catalytic activity of supported Ag(0) species in 2.
ICP analysis of the leachate revealed an Ag(I) concentration of 48.3 ppm, corresponding to a total leached amount of 2.42 mg in 50 mL. Considering the chemical yield of the sample preparation (34%), the estimated total silver content in the sample was 110 mg. Therefore, the percentage of silver leached over 6 h was calculated to be 2.2%. This low percentage suggests recognized stability of compound 2 against HO.
The catalytic performance of the material was evaluated in terms of both reactive species involvement and reusability, which are critical parameters for practical applications. To elucidate the underlying degradation mechanism of HO, scavenger experiments were conducted to identify the key reactive oxygen species (ROS) involved. As shown in Fig. 8A, the introduction of 2-propanol, a specific scavenger for hydroxyl radicals (OH), significantly reduced the MB removal efficiency from 98% (in the absence of any scavenger) to 57%. This result indicates that OH plays a major role in the decomposition process catalyzed by compound 2. Furthermore, when benzoquinone (BQ), a known superoxide radical (O⁻ and HO) scavenger, was added, the removal efficiency decreased even further to 22%. This substantial inhibition confirms the participation of both O⁻ and HO species as another dominant ROS in the system. These findings suggest a synergistic mechanism involving both hydroxyl and superoxide radicals in the catalytic breakdown of MB.
The prominent decline in performance upon scavenger addition provides strong evidence for a radical-mediated degradation pathway, facilitated by electron transfer processes on the surface of Ag(0) nanoclusters. The metallic silver likely acts as an electron relay center, activating HO and promoting the formation of ROS species.
In addition to its high catalytic activity, the recyclability of the catalyst was examined over six successive cycles, as shown in Fig. 8B. The catalyst maintained excellent stability, with HO decomposition efficiency decreasing only slightly from 97% in the first cycle to 92% by the sixth. This marginal drop indicates good structural integrity and durability of the catalyst under repeated use. The retained performance over multiple cycles designates the robust nature of the self-supported Ag(0) system, compound 2. No significant agglomeration or deactivation was observed, suggesting that the active silver sites remain accessible and functional throughout the reaction runs.
To assess the competitiveness of compound 2 relative to other similar catalysts, Table 1 presents a comparison based on their rate constants. A key advantage of compound 2 lies in its ability to achieve efficient catalysis using a nominal amount of active sites, Ag(0), in contrast to the higher metal loadings typically required by other materials.
The decomposition of H₂O₂ by compound 2 is driven by the interaction of Ag(I) and Ag(0) sites with HO. Ag(I) binds HO through oxygen atoms, weakening the O-O bond. Ag(0) adsorbs H₂O₂ via electronic interactions and transfers electrons, generating OH and HO radicals that desorb from the surface. Additionally, superoxide anions (O⁻) are also generated. The highly reactive OH radicals further interact with non-adsorbed HO molecules, leading to further decomposition as shown in Fig. 9 :
Ag(0)/Ag(I) redox cycle should also occur:
Further, Ag(I) can stabilize reaction intermediates, including OH and HO, acting as a Lewis acid to coordinate with HO and its decomposition products. This coordination influences the breakdown of HO into water and oxygen. The overall reaction for HO decomposition in the presence of compound 2 is summarized as follows:
A catalytic cycle is proposed, wherein Ag(I) is reduced back to Ag(0) by the decomposed HO or radical species, allowing continuous recycling of silver sites.