Introduction

As catalytic processes continue to drive the vast majority of the chemical industry, heterogenous catalysis offers several advantages, including the ease of separating the catalyst from the bulk of reactants, the ability to regenerate the spent or deactivated catalyst, as well as enabling continuous process operation1. As the catalytic process in heterogeneous catalysis commences at the solid-liquid or solid-gas interface, maximizing the surface area of the catalyst is an important key towards boosting its activity and efficiency2. Size reduction of the catalyst, reaching down to micro- and nano-particles or even small clusters of catalytic sites, results in maximizing the surface-to-volume ratio of the nanocatalyst. While this approach is of great potential for impacting atom efficiency of catalysts, effective immobilizing such particles/clusters on suitable larger support is essential to avoid leaching into the reaction system, facilitate catalyst handling and processing, as well as avoid adverse health and environmental impacts associated with the miniscule size of the catalyst3,4. Though immobilizing catalysts on porous supports like silica and alumina is commonly explored, the self-supported catalyst is distinguished by size uniformity, homogeneous distribution of active sites, as well as better economics of operation3.

Metallic nanocatalysts, including gold, platinum, palladium, nickel, iron, and silver, demonstrate high catalytic performance in a large number of chemical processes, including, but not limited to, alkene hydrogenation, ammonia synthesis, hydrocarbon reformation, Fisher-Tropsch synthesis, methanol oxidation, among others5. A multitude of synthetic methods have been explored towards constructing tailorable supported nanocatalysts, including wet incipient impregnation, chemical vapor deposition (CVD) and atomic layer deposition (ALD), resulting in nanocatalysts deposited on solid support like zeolites6, silica7, alumina8, graphene9,10, and carbon nano tubes, allowing for controlled fabrication of nanoparticles with notable uniformity and control over the resulting size and shape of the supported nanoparticles11. Alternatively, a stable form of the nanocatalyst can be constructed without the need for solid support, known as a self-supported catalyst12.

For the past two decades, functional inorganic-organic hybrid frameworks have been extensively developed, utilizing ligands as Linkers and metal ions as nodes to form 1D, 2D, and 3D structures via coordination bonding. Due to their extended structures, these coordination polymer (CP) solids are typically insoluble in common reaction media, making them ideal candidates for heterogeneous catalysis when they incorporate appropriate catalytic functionalities. As a result, such materials can function as heterogeneous catalysts and serve as immobilized homogeneous catalysts without requiring external supports, effectively acting as self-supported catalysts12.

Recently, some interesting research has been published considering CP materials for different heterogeneous catalysis applications. A Bimetallic Mo–Zn infinite coordination polymer (ICP) nanoparticles were synthesized as a self-supported catalyst by post-modifying Zn-ICP with MoO2(acac)2. The resulting Mo–Zn-ICP showed high catalytic activity in olefin epoxidation, driven by cis-dioxo-Mo centers13. A single-site heterogeneous Pd(II)–NHC catalyst was developed on a Ru(II)–terpyridine-based coordination polymer. It efficiently promotes selective arene C–H mono-halogenation with high reusability. Control studies confirm the scaffold’s robustness and superior performance over the homogeneous analogue14. The sheet-like heterogeneous self-supported catalysts, dirhodium(ii) coordination polymers, were synthesized via ligand exchange using Rh2(TFA)4 or Rh2(OAc)4 with bdc linkers. The resulting Rh2-bdc frameworks retain the Rh–Rh bond and oxidation state. These self-supported catalysts exhibit high stability, reusability, and catalytic activity in styrene cyclopropanation15.

Self-supported coordination polymers capable of decomposing H2O2 are rarely reported. However, a notable study demonstrated the fabrication of self-supported Au–Pd@UiO-66-on-ZIF-L nanohybrids on flexible carbon cloth using a MOF-on-MOF strategy followed by Au electrodeposition. The resulting bimetallic system showed outstanding electrocatalytic activity for H2O2 reduction, achieving high sensitivity (390 µA mM‒1 cm‒2) and a low detection limit (21.2 nM)16.

Imidazole and its derivatives are prominent ligand candidates for CPs, metal-organic frameworks (MOFs), and zeolitic imidazolate frameworks (ZIFs) due to their two nitrogen donor coordination sites and form some of the most stable complexes among heterocyclic nitrogen ligands17. Silver(I)-imidazolate CP was first synthesized in 1995 and 1997, demonstrating a well-defined structure where single silver ions act as nodes17,18. Previous reports describe silver(I)-imidazole CP as a neutral Ag(I)-N bonding polymeric compound with silver(I) to imidazole in a 1:1 ratio19. IR analysis confirmed that imidazole coordinates Ag(I) as an anion rather than as a neutral ligand, while PXRD analysis revealed a polymeric chain structure with Linearly coordinated silver ions bridged by imidazolate Ligands, ensuring the formation of only cyclic, oligomeric species or 1D chains19.

In the present study, due to its ease of synthesis, the aforementioned simple motif, and well-characterized structure, Ag(I)-imidazole CP was selected for synthesis for the conversion into a self-supported zero-valent silver catalyst. Herein, we describe the synthesis of Ag(I)-imidazole CP, compound 1, and its transformation into a self-supported microsphere heterogeneous catalyst, compound 2, via post-synthetic mild reduction by ascorbic acid. Upon reduction, it is suggested that Ag(0) nano-clusters were homogenously formed and distributed within the compound 2 motif, offering promising catalytic potential. To investigate such catalysis, the decomposition of H2O2 was performed, and the rate is given and discussed. Also, a proposal for the decomposition mechanism is properly given.

Experimental work

Materials

Imidazole (C₃H₄N₂, 99%, Sigma-Aldrich), ammonia solution (NH₃, 35%, Fisher Scientific), and silver nitrate (AgNO3, 99%, Sigma-Aldrich) were used as received without further purification. L-Ascorbic acid (C6H8O6, ≥ 99%, Sigma-Aldrich) was used for the reduction process. (Hydrogen peroxide (H2O2, 50% commercial). DI water was used throughout the synthesis. All chemicals were of analytical grade and purchased from commercial suppliers.

Synthesis of Ag-Imidazolate coordination polymer

To synthesize Ag-Imidazolate, compound 1, a beaker (100 ml) was loaded with imidazole (0.68 g/20 ml DI-water) and ammonia solution (35%, 1.3 ml). The solution was stirred on a hot plate adjusted at 90 °C for about five minutes to guarantee stabilized deprotonation of N-pyrrole of imidazole. Silver nitrate solution (AgNO3, 1.7 g/10 ml DI-water) was then added to the stirred ammoniated imidazolate solution, where immediately a dense white suspended cloud was formed occupying the whole solution volume. Stirring was allowed to continue for about an extra thirty minutes at the same temperature, where the white cloud volume was continually converting into particle aggregates. Stirring and heating were stopped, and the solution was left for complete settling down of the white product, compound 1. After cooling down, 1 powder was then separated by simple filtration from the mother solution and then washed 3 times using adequate amounts of DI-water. The wet solid compound 1 was then allowed to dry at ambient temperature. The measured yield of 1 was about 34%.

The novel compound 2 was thereafter prepared from 1 by reduction using L-ascorbic acid. A certain dried amount of 1 was treated by a low concentration of L-ascorbic acid aqueous solution with shaking for about 5 min at ambient temperature, producing compound 2. The powder of 2 was quickly separated from reducing solutions by simple filtration and then washed 3 times using adequate amounts of DI water. The wet solid reduced samples were then allowed to dry at ambient temperature. SEM/EDX (Zeiss EVO-10 microscopy), FTIR (KBr pellet method using Jasco FT/IR 4100), Powder X-ray diffraction (PXRD, Shimadzu XD-l), and XPS Thermo Fisher Scientific-USA, monochromatic Al-Kα, 1486.7 eV) were applied to characterize compounds 1 and 2.

To examine the catalytic decomposition of H2O2 by 2, a H2O2 solution (5000 ppm) was used as the starting material. The reagents included 0.1 N potassium permanganate (KMnO4) solution, 1 N sulfuric acid (H2SO4), and DI water. A known volume (50 ml) of the 5000 ppm H2O2 solution was transferred into a reaction vessel. A pre-determined quantity of the catalyst (10, 20, 30 mg) was introduced, and the reaction was allowed to proceed with decomposition. Every 10 min, 10 ml of the reaction mixture was transferred to a 100 ml conical flask. To this, 10 ml of 1 N H2SO4 was added to acidify the solution, ensuring complete reaction of H2O2 with KMnO4. A burette (50 ml) was filled with standardized 0.1 N KMnO4 solution. The acidified H2O2 solution was titrated by the gradual addition of KMnO4 with constant swirling until a persistent pink colour appeared, indicating the endpoint. The volume of KMnO4 solution used in the titration was recorded. The residual H2O2 concentration was calculated using the equation:

$$\begin{gathered} {C_{{\mathbf{H2O2}}}}={V_{{\mathbf{KMnO4}}}} \times {N_{{\mathbf{KMnO4}}}} \times {\mathbf{17}}/{V_{{\mathbf{sample}}}} \hfill \\ \hfill \\ \end{gathered}$$

Where CH2O2 is the concentration of residual H2O2 (mg/l), VKMnO4 is the volume of KMnO4 consumed (ml), NKMnO4 is the normality of KMnO4 solution (N), and Vsample is the volume of sample taken (50 ml). Titration was performed in triplicate to ensure accuracy and reproducibility. The obtained values were compared with the initial 5000 ppm concentration to determine the extent of catalysis.

To evaluate the silver retention of compound 2 within its matrix, 0.15 g of the material was immersed in 50 mL of deionized water containing 10% H2O2. The sample was left undisturbed for 6 h to allow potential leaching of Ag(I) ions, after which inductively coupled plasma (ICP) analysis was conducted.

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.

Results and discussion

Characterization of compounds

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 process20. 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.

Fig. 1
figure 1

SEM images of (a) compound 1 and (b) after mild treatment with ascorbic acid as reducing agent to form compound 2, with inserts showing EDX mapping of C, N, and Ag elements

Full size image

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.

Fig. 2
figure 2

HRTEM image of (A) compound 1, and (B) compound 2

Full size image

FTIR analysis was employed to further characterize compounds 1 and 2 (Fig. 3). Both compounds exhibited characteristic absorption bands at 1461 and 1475 cm–1, corresponding to C = N stretching, along with a sharp absorption at 3113 cm–1, characteristic of sp2 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 H2O2 decomposition.

Fig. 3
figure 3

FTIR spectra of 1 (black line), 2 (red line) indicate the maintained structure of 1 upon surface reduction with ascorbic acid, and (blue) after 6 cycles of H2O2 decomposition

Full size image

Raman is a powerful tool to investigate CPs and MOFs, and hence was applied to compounds 1 and 221,22,23. Figure 4 shows Raman spectra of imidazole and compounds 1 and 2. Regarding the crystallinity range, (0–300 cm–1), the imidazole spectrum has one strong peak at 121.5 cm–1 and one weak peak at 139.5 cm–1. The compound 1 spectrum shows two strong peaks at 124.5 and 165 cm–1 which indicates successful synthesis of compound 1 from imidazole. The fingerprint range (300–1800 cm–1) 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 124,25. The extended range (1800–3500 cm–1), more obviously, confirms the successful synthesis of compound 1 from imidazole. The characteristic strong peaks of imidazole at 3129 and 3146 cm–1 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.

Fig. 4
figure 4

Raman spectra of imidazole, compound 1, and compound 2

Full size image

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.

Fig. 5
figure 5

XRD patterns of compounds 1 and 2 indicate maintained crystallinity and absence of diffraction peaks characteristic of Ag particles

Full size image

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 3d5/2 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. 6A26.

For more explanation, the observed + 2.44 eV upshift in the Ag 3d5/2 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 matrix27. 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 screening28. 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 energy26,29,30,31. 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.

Fig. 6
figure 6

XPS analysis of 1 and 2, (a) Ag 3 d line scan indicating reduction of Ag(I) in 1 to Ag(0) in 2, (b) N1s line in 1 demonstrating two nearly equivalent nitrogen environments, while (c) the N1s line scan in 2 demonstrating pronounced pyrrolic N environment upon surface treatment with reducing agent

Full size image

Decomposition of H2O2 by compound 2

To probe the catalytic activity of the self-supported Ag(0) catalyst, different masses of 2 were introduced into H2O2 solution, and the remaining concentration of H2O2 was monitored through iodometric titration32, 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 H2O2 solution were utilized, where the three reactions converged to full decomposition of H2O2 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 H2O2 even after 70 min of contact time with H2O2, confirming the catalytic activity of supported Ag(0) species in 2.

Fig. 7
figure 7

Catalytic decomposition of H2O2 utilizing compound 2 with different catalyst loading (A, B, C), and (D) linear fit with different mass loading

Full size image

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 H2O2.

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 H2O2, 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 (O2⁻• and HO2•) scavenger, was added, the removal efficiency decreased even further to 22%. This substantial inhibition confirms the participation of both O2⁻• and HO2• 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 MB32.

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 H2O2 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 H2O2 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.

Fig. 8
figure 8

(A) Effect of radical scavengers on H₂O₂ removal efficiency: 2-propanol (•OH scavenger) and benzoquinone (O₂⁻• and HO2• scavenger), confirming the involvement of hydroxyl and superoxide radicals in the catalytic mechanism, and (B) Recyclability of the self-supported Ag(0) catalyst for H2O2 decomposition over six consecutive cycles, demonstrating high catalytic stability

Full size image

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.

Table 1 Comparison of the catalytic decomposition of hydrogen peroxide by compound 1 with others in the literature.
Full size table

Proposed mechanism of H2O2 decomposition over compound 2

The decomposition of H₂O₂ by compound 2 is driven by the interaction of Ag(I) and Ag(0) sites with H2O2. Ag(I) binds H2O2 through oxygen atoms, weakening the O–O bond. Ag(0) adsorbs H₂O₂ via electronic interactions and transfers electrons, generating •OH and HO2• radicals that desorb from the surface36. Additionally, superoxide anions (O2•⁻) are also generated. The highly reactive •OH radicals further interact with non-adsorbed H2O2 molecules, leading to further decomposition37 as shown in Fig. 9 :

$$\begin{array}{*{20}{l}} {2{H_{2}}{O_2}{\text{ }} - {\text{ }}Ag\left( 0 \right){\text{ }}{ \to ^ \bullet }OH\,+\,H{O_2}^{ \bullet }+{\text{ }}{H_2}O} \\ {{O_2}+{\text{ }}Ag\left( 0 \right){\text{ }} \to {\text{ }}{O_2}^{{ \bullet - }}{\text{ }}+{\text{ }}Ag\left( I \right)} \\ {^{ \bullet }OH\,+\,{H_2}{O_2}{\text{ }} \to {\text{ }}{H_2}O\,+\,H{O_2}^{ \bullet }} \end{array}$$

Ag(0)/Ag(I) redox cycle should also occur38:

$$\begin{array}{*{20}{l}} {Ag\left( 0 \right)\,+\,{H_2}{O_2}{\text{ }} \to {\text{ }}Ag\left( I \right)\,+\,{H_2}O{\text{ }}+{\text{ }}\raise.5ex\hbox{$\scriptstyle 1$}\kern-.1em/ \kern-.15em\lower.25ex\hbox{$\scriptstyle 2$} {\text{ }}{O_2}} \\ {Ag\left( 0 \right)\,+\,{O_2}^{{ \bullet - }} \to {\text{ }}Ag\left( 0 \right){*^ - }+{\text{ }}{O_2}} \\ {Ag\left( I \right)\,+\,Ag\left( 0 \right){*^ - } \to {\text{ }}2Ag\left( 0 \right)} \\ {Ag\left( I \right){\text{ }}{+^ \bullet }OH{\text{ }} \to {\text{ }}Ag\left( 0 \right)\,+\,O{H^ - }} \\ {Ag\left( I \right)\,+\,2H{O_2}^{ \bullet } \to Ag\left( 0 \right)\,+\,2{O_2}\,+\,{H_2}} \\ {Ag\left( I \right)\,+\,O{H^ - } \to {\text{ }}Ag\left( 0 \right){\text{ }}{+^ \bullet }O{H^{}}} \end{array}$$

Further, Ag(I) can stabilize reaction intermediates, including •OH and HO2•, acting as a Lewis acid to coordinate with H2O2 and its decomposition products. This coordination influences the breakdown of H2O2 into water and oxygen. The overall reaction for H2O2 decomposition in the presence of compound 2 is summarized as follows:

$${H_2}{O_2}{\text{ }}+{\text{ }}Ag\left( I \right){\text{ }} - {\text{ }}Ag\left( 0 \right)/Ag\left( I \right) \to {\text{ }}{H_2}O{\text{ }}+{\text{ }}\raise.5ex\hbox{$\scriptstyle 1$}\kern-.1em/ \kern-.15em\lower.25ex\hbox{$\scriptstyle 2$} {O_2}$$

A catalytic cycle is proposed, wherein Ag(I) is reduced back to Ag(0) by the decomposed H2O2​ or radical species, allowing continuous recycling of silver sites38.

Fig. 9
figure 9

Schematic illustration of the proposed mechanism of H2O2 Decomposition over Compound 2

Full size image

Conclusion

This work demonstrated the generation of highly efficient self-supported Ag(0) nanocatalysts from a simple structured Ag(I)-based coordination polymer through mild reduction. Ag(0) moieties within the coordination polymer should be nano-clusters or even Ag(0) nodded atoms. For both cases, metallic Ag is well distributed within the coordination polymer motif, giving a high chance for effective catalysis performance, especially when the coordination polymer is of micro-size with a featured platelet surface. Hence, the large library of previously reported coordination polymers or Metal-Organic Frameworks is an excellent candidate for this approach, where pre-designed catalytic metallic sites can be readily generated from ligand-metal coordination complex at the solid-liquid interface in a mild reducing solution.