1 Introduction

Metallic nanoparticles (NPs) production is one of the most exciting scientific achievements of the last two decades so far, and it will be of much interest in the future due to their distinctive characteristics, such as small particles’ size, their shapes, and their chemistry, which is reflected in the wide range of their applications [1]. Most of the chemical and physical properties of metallic NPs depend on their size and morphology, which are affected by temperature, period of reaction, concentration, and the ratio of reagents [2]. For these reasons, much attention is currently devoted to the production of the smallest particle size with wonderful morphologies useful in many applications.

Silver NPs have been tremendously investigated due to their superior chemical, physical, and biological characteristics [2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22]; for this reason, they are the focus of our attention in this research. Silver NPs have revolutionized the field of nanotechnology due to their application in several fields that are dependent on unique size-dependent magnetic properties such as physical, chemical, optical, surface Plasmon resonance properties, and electrical resistance. Ag NPs are used in several fields; for instance, Pandey and Ramontja [3] reported antibacterial applications of Ag NPs in biotechnology and biomedical fields. Rajiri et al. [4] demonstrated that Ag NPs have anticancer and therapeutic potential against breast cancer and pancreatic cells. Fayaz et al. [5] addressed the Ag NPs’ antimicrobial activity towards gram-positive and gram-negative bacteria. Prasher et al. [6] reported the effectiveness of Ag NPs as drug delivery carriers. Kim et al. [7] synthesized nanosilver and carbon nanocomposite membrane for enhanced water treatment. The various electronic applications of Ag NPs were investigated by Wang et al. [8]. Not only are these mentioned uses for nanosilver, but there are also a lot of uses in our lives, such as bio-sensing [9] and the production of high-performance cotton fabrics [10].

The preparation methods of Ag NPs play a crucial role in their applications because the method of preparation is a key factor in controlling the particles’ size and shape, which is reflected in their applications. Numerous routes for Ag NPs synthesis can be found in the open literature. These include chemical etching, sputtering, ball/mechanical milling, thermal/laser ablation, lithography, thermal decomposition, vapor deposition, atomic/molecular condensation, chemical and electrochemical precipitation, the sol–gel process, spray pyrolysis, and biological-based methods [12,13,14,15,16,17].

The combustion method is one of the most important methods for preparing nanometals because it has quick and easy production, operational simplicity, and the utilization of low environmental impact reagents. In the combustion method, a metallic precursor acts as an oxidizer (normally metal nitrate) that reacts with an organic fuel as a reducing agent (such as sugar, amino acid, urea, ethylene urea, glycine, cetyltrimethylammonium bromide (CTAB), etc.) below a specified degree of combustion temperature to produce metal nanoparticles [23,24,25,26,27,28,29,30]. Several factors impact the properties of nanometals, such as calcination temperature, type of fuel, pH, combustion pressure, the ratio between oxidizer and fuel, and type of solvent. In this way, Sharma et al. [21] showed that, in comparison to glycine, citric acid has the advantage of forming more spherical (diameters of approximately 40 nm) and symmetrical Ag NPs. Spherical Ag NPs parties have also been obtained using gallic acid as a fuel with diameters in the range of 61–400 nm [31]. Employing poly N-vinylpyrrolidone (PVP) and lactic acid as fuels, spherical Ag NPs have been obtained [32]. Moreover, raising the calcination temperature from 350 to 500 °C has led to a particle size increase from 22 to 32 nm of these Ag NPs [32]. A flameless combustion route was used to prepare quasi-spherical Ag NPs with dimensions in the 20–30 nm range, starting with cellulose nitrate and silver carbonate at 350 °C [33].

Reviewing the available literature resources revealed a growing interest in the Ag NPs synthesis in general and utilizing the combustion route in particular. In the open literature, there is a lack of knowledge regarding the usage of urea as a combustion fuel during Ag NPs synthesis. Therefore, the present research involves a systematic investigation that addresses the role of the combustion fuel (urea) as well as the calcination temperature (400–700 °C) on the formation of Ag NPs. Various characterization tools will be devoted to the precise elucidation of the crystal structure, the morphologies, and the surface properties of the as-prepared Ag NPs. Finally, the activity of the formed nanomaterials will be evaluated for NaBH4 decomposition.

2 Experimental

2.1 Combustion Procedure

Analytical-grade precursors of the oxidized (silver nitrate, AgNO3) and the organic fuel (urea, CH4N2O2) were used as received. One materials series has been prepared that focuses on the effect of the temperature of calcination on the phase formed, crystallite size, and morphology of Ag NPs. In this series, the fuel/oxidizer ratio was adjusted to 0.50, and the calcination temperature was selected at 400, 500, 600, and 700 °C. The precursors have been dissolved in the least amount of distilled water, mixed, and dried gently over a hotplate. The furnace was adjusted to the desired temperature (500–700 °C). Then, the dried materials were, finally, subjected to calcination at that temperature in static air for 1 h. The produced solids have been referred to by the abbreviation Ag–x, where the x-value denotes the temperature of calcination.

2.2 Characterization Tools

Powder X-ray diffraction patterns have been performed on monochromatized Cu-Kα radiation (λ = 0.154060 nm). The operation condition used was 30 kV and 30 mA with a scanning rate of 0.02136° s–1. Morphology identification of some selected Ag NPs was performed using scanning and transmission electron microscopies. The models used were JEOL (JSM-5400 LV) and (JEMTH-100 II), respectively. In the scanning mode, the samples have been measured after performing gold sputtering. Before the transmission inspection, the samples have been sonicated (for 30 min), loaded on Cu-grid, and left to dry. Surface inspection using the XPS technique has been examined on the K-ALPHA (Thermo Fisher Scientific) instrument. The conditions utilized were monochromatic X-ray Al K-alpha radiation, spot size 400 μm, binding energy (BE) range (10 to 1350 eV), and pressure of 1 × 10–9 m bar. The C 1 s peak (at 284.8 eV) was used for calibrating the binding energies in the XPS spectra.

2.3 Activity Assessment

The hydrolysis of NaBH4 experiments over the various Ag NPs was conducted in a 100 ml flask reactor made of Pyrex glass, whereas the volume of the produced hydrogen was measured using the water displacement method [34]. The reactor was immersed in a 1 L Pyrex glass beaker. The temperature of the reactor was controlled and monitored using an IKA-RCT hotplate connected to a contact thermometer. 0.1 g Ag catalyst was introduced to the reactor, which contains a thermostated 20 ml NaBH4 solution (1.5 wt%). The reaction started, and the generated hydrogen volume was followed up by the water displacement method over a reaction temperature range of 35–50 °C.

3 Results and Discussion

3.1 Phase Formation

The recorded X-ray powder diffraction patterns over the 2 θ range of 20–90° are presented in Fig. 1. The obtained pattern for the 400 °C calcined samples reveal the crystalline nature of this sample, together with the presence of a large number of peaks. Analyzing these peaks revealed the presence of major peaks at 2 θ = 21.74°, 24.34°, 29.70°, 31.93°, 32.82°, 39.12°, 40.13°, 43.52°, 46.25°, 49.77°, and 53.87°. These peaks match those of orthorhombic AgNO3 as a major phase (JCPDS file No. 00-006-0363). However, these peaks showed a slight 2 θ shift of 0.05–0.12° towards higher values. In this context, it is to be mentioned that the various reported JCPDS peaks of AgNO3 vary in the range of 0.70° from each other. The detection of such a phase agrees well with the reported data for the complete decomposition of silver nitrate at approximately 500 °C [35]. An additional factor can be raised by analyzing the rest of the peaks in this sample. Another peak can be observed at 2 θ = 38.18°, 44.35°, 64.51°, 77.45°, and 81.61°, which could be anticipated to be the presence of a minor amount of Ag (JCPDS file No. 01-089-3722). A trace amount of carbon can also be determined by the presence of the reflections at 2 θ = 44.35° and 75.22° (JCPDS file No. 00-001-1249). Therefore, it is reasonable to relate the observed slight shift in the 2 θ values of AgNO3 to the presence of these phases. In other words, the co-existence of the Ag and C phases in the constituents of the Ag-400 sample could be responsible for a slight d-spacing shift. In this context, Wu et al. [36] reported a shift in the peak positions of their Ag NPs prepared by the hydrothermal route. Singh and Khanna [37] reported a little change in the Ag NPs lattice spacing from 2.365 to 2.382 Å because of increasing the Ag loading from 2 to 10% in their Ag/polymethylmethacrylate (PMMA) composite. Considering the thermal decomposition of urea, it proceeds with the formation of a wide spectrum of products, including biuret, cyanuric acid, cyanic acid, ammonia, ammonium cynate, cyanamide, ammelide, ammeline, and melamine [23, 38,39,40,41]. The reaction covers a wide range of temperatures (from > 130 °C to approximately 400 °C) with complex reaction pathways [23, 38,39,40,41]. Krum et al. [40] demonstrated that bare urea undergoes full decomposition after isothermal heating at 380 °C for 55 h. This agrees well with our XRD data for the Ag-400 sample, where only a trace amount of carbon was found in that sample.

Fig. 1
figure 1

The recorded X-ray patterns for the silver-urea precursor calcined at 400 (a), 500 (b), 600 (c), and 700 °C

Full size image

Raising the temperature of calcination from 400 to 500 °C is associated with the disappearance of all the previously mentioned peaks due to AgNO3 and C phases and the development of the five peaks located at 2 θ = 38.03°, 44.19°, 64.31°, 77.29°, and 81.40°. These peaks could be allocated to the (111), (200), (220), (311), and (222) planes of the cubic Ag (space group Fm-3m, JCPDS file No. 01-089-3722), respectively. Increasing the temperature of calcination to 700 °C resulted in the persistence of these five peaks, indicating the detection of only one phase, viz., Ag NPs. It can be stated that within the examined temperature range, Ag NPs represent the only phase detected during the 500–700 °C range. The crystallite size of the formed Ag NPs, within the calcination temperature range of 500–700 °C, has been calculated using the Scherrer equation employing the (111) reflection, which is usually used for estimating the silver crystallite size [14, 36, 37]. The obtained values were 40, 37, and 36 nm for the Ag-500, Ag-600, and Ag-700 samples, respectively. Two points could be raised here: (i) these values show only a very slight decrease with increasing the calcination temperate, and (ii) the estimated values were higher than those resulting from silver acetate thermal degradation (13–22 nm) [14] and silver in Ag/PMMA composite (23–25 nm) [37] and lower than those reported for Ag NPs prepared using PVP (78–113) [36]. Using in situ X-ray diffraction, Yanase et al. [42] reported that on heating metallic silver in an oxygen atmosphere, the silver lattice becomes more stressed, which eventually leads to failure and a decrease in crystallite size. Nagy and Mestl [43] reported a striking reduction in the silver crystallite size during the time on stream experiments on treating silver catalyst in 10 Vol.% oxygen at 750 °C. They suggested that this decrease is due to the formation of interstitially located bulk-dissolved oxygen species. TPO experiments suggested the presence of the Ag → Ag2O → Ag redox cycle on heating Ag-based catalysts over the temperature range of 375–625 °C [44]. Based on nonisothermal TGA-DTA measurements, chemisorbed oxygen on Ag was released via an exothermic peak at 700–780 °C [45]. Hong et al. [46] reported an Ag structure deterioration for the CrNx/Ag nanocomposite (20.2 at.% Ag) calcined at 800 °C. Based on these literature data, it is plausible to attribute the observed slight crystallite size decrease of Ag NPs decrease with increasing calcination temperature (Fig. 1) to the formation of bulk-dissolved oxygen species, formed during calcination in air, which results in the deterioration of metallic silver structure with temperature.

3.2 Characterization of Ag NPs

In the preceding section, it was found that the reaction between the mixture constituents (AgNO3 and CH4N2O2) takes place at 500 °C. Therefore, and in combination with the activity data (vide infra), Ag-500 and Ag-700 catalysts have been subjected to characterization using SEM, TEM, and XPS techniques. The obtained micrograph for the Ag-500 catalyst (Fig. 2a) shows a compact arrangement of particles with sphere-like morphology and sizes ranging from 5 to 20 μm. Various pores can be observed, ascribable to the various gases released accompanying the precursor’s combustion. It is evident that increasing the calcination temperature to 700 °C is associated with particle welding (Fig. 2b), which leads to the loss of the sphere-like particles of the Ag-500 catalyst. The Ag-700 catalyst shows clear evidence of particle welding, as indicated by the arrows, where the various particles are linked to each other, forming bigger ones. The morphology of the Ag-700 catalyst is similar to that observed for other Ag NPs prepared via hydrogen reduction of Ag2O [47], the solvothermal method [48], and thermal decomposition of silver acetate [14].

Fig. 2
figure 2

SEM micrographs obtained for Ag-500 (a) and Ag-700 (b) catalysts

Full size image

TEM images of Ag-500 and Ag-700 catalysts are shown in Fig. 3, which are magnified by a factor of 19 kx. For both catalysts, abundant pores are distributed among the Ag NPs, giving a membrane-like network. The image of Ag-500 catalysts (Fig. 3a) shows a set of two-layered membranes with elongated pores. In this context, silver-based membranes have been previously reported in the literature and have important energy applications [49, 50]. With the aid of the XRD findings and TEM inspection, we may suggest the following reasons for Ag NPs formation. First, Wu et al. [36] reported that the viscosity increase of the water–ethanol solution retards the Ag+ ion diffusion, which is associated with an anisotropic growth of Ag nanocrystals. Knowing that the viscosity of the urea-water solutions is higher than that of pure water [51]. Thus, we may suggest that one role of urea is to reduce silver ions diffusion, which favors the anisotropic growth of Ag NPs with membrane-like morphology up on the precursor decomposition. Second, it is generally agreed that, among the Ag nanocrystal planes, the (1 1 1) plane processes the lowest surface energy [36, 52, 53]. This plane tends to adsorb molecules such as polymers and ammonia, which results in additional surface energy lowering on this plane [36, 52]. Therefore, it is reasonable to propose that during the precursor preparation, the urea molecules are adsorbed on the (1 1 1) plane, leading to a decrease in both its surface energy and its growth rate. This, in turn, would lead to lower adsorption of urea on other faces, leading to a higher growth rate, which favors the formation of the 2D morphology upon precursor decomposition. A similar argument was used to interpret the formation of Ag NPs with plate-like morphology [36, 52]. Increasing the calcination temperature results in noticeable morphological modifications. The photograph of the Ag-700 catalyst (Fig. 3b) indicates the presence of a porous membrane of Ag NPs with smaller diameters and wider distribution. Moreover, the pores show a decrease in size either by splitting into smaller pores (as shown by the α–arrows) or via pore narrowing (as shown by the β–arrows), which eventually leads to pore blocking as shown by the γ–arrows. From the electron microscopy examinations, it is obvious that, in addition to the role of the precursors (type of fuel and fuel/oxidizer ratio [21]), the calcination temperature plays a crucial role in adjusting the morphology and texture of the prepared Ag NPs.

Fig. 3
figure 3

TEM micrographs obtained for Ag-500 (a) and Ag-700 (b) catalysts

Full size image

Based on the activity assessment, the Ag-500 and Ag-700 samples were further investigated using XPS analysis. In the survey spectra of both catalysts (not shown), only silver, oxygen, and carbon elements have been detected; no other element, such as nitrogen, was detected on the surface of both catalysts. The detection of carbon at the surface of both silver samples is not surprising, which probably originated from the urea decomposition. In this context, the detection of both O and C has been reported for the Ag NPs obtained from the degradation of silver acetate [14]. The fitted Ag 3d XPS spectrum of the Ag-500 catalyst is shown in Fig. 4a. Two peaks located at 368.15 and 374.10 eV with a splitting BE of approximately 6 eV can be seen in the original spectrum. From the literature database, these peaks could be indexed to Ag 3d5/2 and Ag 3d3/2 electronic levels, respectively [14, 20, 22, 54,55,56]. Peak deconvolution of the Ag 3d5/2 level yields three features at 367.61, 368.11, and 369.29 eV. Similar features can be obtained by the peak fitting of the Ag 3d3/2 level, which appear at 373.13, 374.12, and 375.32 eV. For Ag 3d5/2 and Ag 3d3/2 electronic levels, the sharp peaks at 368.11 and 374.12 eV and the weak ones at 369.29 and 375.32 eV could be assigned to the metallic silver [14, 20, 54,55,56]. It was reported that oxidizing metallic silver is associated with a binding energy lowering of both levels [14, 54]. Therefore, the small-intensity features at 367.61 and 373.13 eV, respectively, could be related to the presence of Ag+ species, in the form of Ag2O, at the surface of the prepared nanomaterial [14, 20, 54]. The detection of Ag+ species agrees with the work of Sun et al. [55], who observed the development of a minor amount of oxidic silver in their Ag/graphene composites. Their formation was attributed to the possible partial Ag NPs oxidation upon exposure of the prepared composites to atmospheric air [55]. Consistent data were reported by Albiter et al. [56], who detected Ag2O on the surface of the Ag/TiO2 photocatalyst via Ag0 oxidation upon exposure to the air environment.

Fig. 4
figure 4

Fitted Ag 3d (a), O 1 s (b), and C 1 s (c) XPS spectra of Ag-500 catalyst

Full size image

Peak-deconvolution of the O 1s core level of the Ag-500 catalyst (Fig. 4b) gives two peaks at BEs of 530.85 and 531.78 eV. The first peak could be indexed to the lattice oxygen of Ag2O, whereas the second peak could be indexed to the surface hydroxyl oxygen and/or C=O groups [14, 20, 54, 57]. Three features can be seen in the deconvoluted asymmetric C 1s core level of the Ag-500 catalyst (Fig. 4(c)), which suggests the existence of various carbon-containing function groups at the prepared Ag NPs. The first one, which is located at 284.68 eV, may be due to C–H and/or C=C bonding [14, 57]. The second feature appearing at BE of 285.22 eV could be indexed to C–O and/or C–C bonding [14, 57]. The lowest intensity peak observed at BE 287.79 eV could be assigned to C=O bonding [14, 57].

The deconvoluted Ag 3d, O 1 s, and C 1 s core levels of the Ag-700 catalyst are presented in Fig. 5. The same species identified in the case of the Ag-500 catalyst can be identified here. However, some modifications can be identified. The intensity of the Ag 3d spectrum of the Ag-700 catalyst (Fig. 5a) shows a noticeable decrease. Meanwhile, the relevant XPS spectra of O 1s (Fig. 5b) and C 1s (Fig. 5c) reveal a noticeable intensity decrease. A quantitative analysis of the three elements at the surface of both catalysts can be seen in Table 1. It is reasonable to correlate the obtained decrease in the C and O concentrations of Ag-700 catalyst with the expected increased burning rate of the carbonaceous species located at its surface upon increasing the calcination temperature. It is to be mentioned that the first peak intensity in the fitted spectrum of C 1 s of the Ag-700 catalyst is higher than those of the other two peaks (Fig. 5c). This indicates that the surface of this sample has a higher content of C–H and/or C=C species.

Fig. 5
figure 5

Fitted Ag 3d (a), O 1 s (b), and C 1 s (c) XPS spectra of Ag-500 catalyst

Full size image
Table 1 XPS surface binding energies and atomic percentages of silver, oxygen, and carbon of Ag-500 and Ag-700 catalysts
Full size table

It is obvious from Figs. 4 and 5 that the intensity of the main peak assignable for metallic silver is higher than that of Ag+ for both Ag-500 and Ag-700 catalysts. It is plausible to suggest that, however, no diffraction peaks were observed for silver oxide phases in the XRD patterns of both catalysts. The XPS detected Ag+ species suggests that a portion of the formed metallic silver is oxidized by the atmospheric oxygen, yielding surface Ag2O at the surface of the prepared Ag NPs.

3.3 NaBH4 Hydrolysis Activity

Figure 6 shows the results of NaBH4 hydrolysis experiments (VH2 vs. time) for the Ag NPs calcined at 500, 600, and 700 °C. It can be seen that, all over the tested reaction temperatures (35–50 °C), there is no induction (or activation) period. In other words, there was no considerable hydrogen generation at the beginning of the reaction. Moreover, a linear dependence of the produced hydrogen volume with time can be observed. Furthermore, a continuous activity increase is associated with an increase in the reaction temperature. The obtained activity order is Ag-500 > Ag-600 > Ag-700. For instance, 35.2, 38.5, and 44.5 min are required for producing 75 ml of hydrogen at 35 °C over Ag-500, Ag-600, and Ag-700 catalysts, respectively. This indicates that the HGR significantly depends on the catalyst calcination temperature. All over the tested reaction temperatures, a continuous HGR decrease with the calcination temperature can be seen.

Fig. 6
figure 6

Activity measurements in terms of V-t plots obtained for the Ag-500, Ag-600, and Ag-700 catalysts at 35 °C (a), 40 °C (b), 45 °C (c), and 50 °C (d)

Full size image

The rate constants and the activation energies of the NaBH4 hydrolysis over the three Ag catalysts can be computed from the obtained data. The hydrogen generation rate (HGR) has been computed from the slopes of the VH2 vs. time curves, and the obtained values are shown in Fig. 7a. It is obvious that the HGR responds positively to the reaction temperature and negatively to the calcination temperature. The highest values (21.5, 32.2, 57.6, and 70.7 ml min–1 g–1 at 35, 40, 45, and 50 °C, respectively) are obtained over Ag-500 at 50 °C. These HGRs are much lower compared to those of Co3O4/C (1821 ml min–1 g–1 at 35 °C) [34], Co/CNTs (410 ml min–1 g–1 at 35 °C) [57], CoB (1735 ml min–1 g–1 at 20 °C) [58], Co thin films (7900 ml min–1 g–1 at 25 °C) [59], and Co3O4 NPs (1776 ml min–1 g–1 at 25 °C) [60]. These catalysts are known to show excellent activity for this reaction. On the other hand, the lie approximately within the same activity range for the reaction performed over Pd/MWCNs (20.7 ml min–1 g–1 at 30 °C) [61], Ag/MWCNs (17.4 ml min–1 g–1 at 30 °C) [62], Ru/Al2O3 (34.9 ml min–1 g–1 at 25 °C) [63], and chitin/Cu hydrogel nanocomposite (23.8 and 46.2 ml min–1 g–1 at 35 and 55 °C, respectively) [64]. Regarding the effect of the calcination temperature on the activity performance, Durano et al. [60], working on Co3O4 NPs calcined at 600–800 °C, reported the highest activity for the 700 °C calcined Co3O4 NPs. They ascribed the poor activity of the 800 °C calcined catalyst to the active sites’ destruction at such a high temperature. For Co3O4/C hierarchical nanocomposites, the highest HGR was obtained over the 500 °C calcined catalyst [34]. Higher calcination temperatures are associated with a decrease in the textural parameters and the active centers. In the present work, it seems that 500 °C is the optimal temperature for the formation of Ag NPs and their highest performance for NaBH4 hydrolysis.

Fig. 7
figure 7

a Dependence of the hydrogen generation rate (HGR) on the reaction temperature, b Arrhenius plots, and c Constable-plot obtained for Ag-500, Ag-600, and Ag-700 catalysts

Full size image

Applying the Arrhenius equation, ln k = ln (A)−(Ea/RT), to the data presented in Fig. 7a, enables the calculation of the activation energies for the hydrolysis reaction for the three catalysts. Figure 7b depicts the dependence of ln HGR vs. the reciprocal absolute temperatures. Good linearity can be seen for all catalysts. By the compensation in the Arrhenius equation, the computed activation energies were 68.8, 69.9, and 74.8 kJ mol–1 K–1 for Ag-500, Ag-600, and Ag-700 catalysts, respectively. By plotting ln A vs. Ea, for the three Ag catalysts, the Constable plot can be obtained as indicated in Fig. 7c. Good linearity can be seen with a positive slope, which belongs to case “A” of Bond’s et al. classification [65]. This, in turn, highlights the presence of a compensation effect among these catalysts. In other words, hydrogen is generated over these Ag NPs catalysts from the same active sites and via the same mechanism.

Four kinetic steps have been reported for the catalyzed hydrolysis of NaBH4, which implies the adsorption of both NaBH4 and H2O on active metal sites [34, 57, 61, 62, 66,67,68]. In the first step, BH4– ion is chemisorbed on the active site, which is followed by the H– ion transfer to another active site. In the third step, hydrogen is evolved via the reaction between the adsorbed H– and H2O, leaving behind OH– ion. The final step involves the reaction of OH– with the adsorbed BH3, forming the BH3(OH)– ion. The formed BH3(OH)– undergoes successive cycles of reactions, eventually leading to the formation of B(OH)4−. The desorption of such a product leads to the regeneration of the active center. Huff et al. [62] reported the mechanism for the NaBH4 hydrolysis over AgMWCNs, indicating the role of silver in the interaction with the BH4– ion. However, they did not indicate the role of silver species or the type of active centers in the BH4– and H2O adsorption and reaction. Based on XPS investigations for the NaBH4 hydrolysis over Co-B [66] catalysts, it was concluded that the electron-enriched metal is crucial, which is responsible for facilitating the reaction by providing the required electrons. In the proposed mechanism by Andrieux et al. [67], BH4– and H2O are adsorbed on atoms “A” and “B”, respectively. Atom “A” is an electron-rich center that can donate its electron density to the boron atom. On the other hand, atom “B” is an electron-deficient center that attracts the electron density of the oxygen atom in the water molecule. Over the Co NPs catalyst, atoms “A” and “B” have been proposed to be Co0 and Coδ+, respectively [67]. Over Co/MWCNTs and hierarchical Co3O4/C NPs, these atoms were suggested to be the Co2+ and Co3+ species, respectively [34, 57]. Over Ru-RuO2/C catalyst, Ru0 and RuO2 have been suggested as adsorption sites for BH4– and H2O, respectively [68]. Our XPS analysis revealed the presence of Ag0 and Ag+, as indicated by the peak deconvolution of the 3d electronic level (Figs. 4a and 5a), on the surface of Ag-500 and Ag-700 catalysts. Accordingly, it is plausible to assign these species as electron-rich and electron-deficient adsorption sites, respectively. Moreover, it was found that increasing the temperature of calcination from 500 to 700 °C is associated with an increase in the BE of Ag metal (Table 1). In other words, the lower binding energy of Ag metal in the Ag-500 catalyst indicates a higher electron density on Ag NPs, which could be the reason for the higher activity of the Ag-500 catalyst compared to the Ag-700 catalyst.

The reusability tests have been performed for seven cycles. The obtained V-t plots vs. time and the HGR vs. the number of cycles for the NaBH4 hydrolysis over the optimal catalyst, i.e., Ag-500, are shown in Fig. 8 (conditions: 1.5 wt% NaBH4, 0.2 g catalyst, 40 °C reaction temperature).

Fig. 8
figure 8

Recycling tests: a V-t plots versus time and b HGR versus number of cycles performed using Ag-500 catalyst

Full size image

It can be seen that the activity decreases gradually with cycles. However, the magnitude of this decrease is less after the 4th cycle. The catalyst is still showing 67 and 58% of its initial activity after the 4th and 7th cycles, respectively. This activity decrease agrees with that reported for other catalysts. For instance, around 70% retaining of the initial activity was reported for the NaBH4 hydrolysis after five cycles over hierarchical Co3O4/carbon NPs [34], 2.8CoB/Ag-TiO2 [58], and cobalt-based thin films (14%) [59]. Lower activities have been reported over other catalysts such as Co/MWCNTs (34%) [57] and intrazeolite Co0 nanoclusters (59%) [69]. This activity decrease could be ascribed to: (i) the development of strongly adsorbed borate complex; (ii) the induced structural and/or morphological modifications in the catalysts [57]; and (iii) the possible loss of the catalyst during recycling.

Figure 9 shows the XRD patterns for the Ag-500 and Ag-700 spent catalysts after seven catalytic cycles. In comparison with the fresh catalysts, it is evident that the spent ones still possess reflections due to the cubic Ag (space group Fm-3 m, JCPDS file No. 01-089-3722) indicating that the structure remained intact after contact with NaBH4 for several runs.

Fig. 9
figure 9

XRD patterns of the spent Ag-500 and Ag-700 catalysts after seven cycles of the catalytic activity tests

Full size image

4 Conclusions

Based on the previous findings, we may suggest the following points: The combustion route, using urea as a fuel, is a novel one-pot technique for preparing Ag NPs, which adopt the cubic structure and have a crystallite size in the range of 36–40 nm. The role of the used combustion fuel is not only to minimize the temperature required for phase formation but also to favor the reduction of silver ions and act as a structure-directing agent for morphology control. The prepared Ag NPs are characterized by a porous structure with a membrane-like morphology. No diffraction peaks were observed for silver oxide phases in the XRD patterns. However, Ag+ species were detected at the surface of the prepared Ag NPs using XPS investigation. The activity of the prepared nanomaterials was evaluated for NaBH4 hydrolysis. The 500 °C calcination temperature represents the balance temperature for the formation of Ag NPs and their highest activity performance. Finally, in addition to its catalytic application for the NaBH4 hydrolysis, the porous morphology nominates the prepared Ag NPs for the application(s) as a membrane.