Ag co-catalyst prepared by ultrasonic reduction method for efficient photocatalytic conversion of CO₂ with H₂O using ZnTa₂O₆ photocatalyst

Abstract

Towards the realisation of carbon neutrality by utilising renewable energy sources, the photocatalytic conversion of CO₂ with H₂O—known as artificial photosynthesis—is important because H₂O is non-toxic or non-hazardous, and an abundant source of protons for CO₂ reduction. Many studies on the photocatalytic conversion of CO₂ have revealed that Ag nanoparticles are an effective co-catalyst for the selective conversion of CO₂ to CO in water. To improve the activity of the photocatalytic conversion of CO₂ in water, modifying the surface of the photocatalyst is essential to load small Ag nanoparticles with high dispersity, which is difficult to achieve using conventional methods. In this study, ultrasonic reduction (USR) was used as an advanced modification method for photocatalysts with an Ag co-catalyst. Ag/ZnTa₂O₆ prepared using the USR method exhibited good selectivity towards CO (>90%) evolution and a higher CO formation rate compared to those prepared using the conventional modification methods. High-resolution transmission electron microscopy images of the Ag co-catalyst revealed that Ag nanoparticles with a size of a single nanometre were loaded onto the surface of ZnTa₂O₆ by the USR method, whereas much larger Ag particles loaded onto it were observed in the case of other methods. Accordingly, a small Ag co-catalyst with a single nanometre size exhibits superior activity towards the selective conversion of CO₂ to CO. Thus, we successfully achieved a high CO formation rate with high selectivity using a Ag/ZnTa₂O₆ photocatalyst prepared via USR.

---

Introduction

Global warming is causing an increase in sea levels and abnormal weather, which are beginning to negatively impact people's lives and ecosystems. Carbon dioxide (CO₂) is a key greenhouse gas that drives the global climate changes by absorbing and radiating heat. However, the content of atmospheric CO₂ has increased by 50% since the Industrial revolution due to human activities such as burning of fossil fuels and large-scale deforestations, making CO₂ the largest contributor to global warming. While decreasing the use of fossil fuels can aid in reducing CO₂ emissions, developing advanced technologies for the capture and use of CO₂—known as carbon dioxide capture, utilisation, and storage (CCUS)—is more beneficial.¹ Among the CCUS technologies, artificial photosynthesis based on photocatalysis has attracted significant attention.^2–5 In particular, the photocatalytic conversion of CO₂ with H₂O is expected to have practical applications because it enables the direct reduction of CO₂ and the production of useful intermediate chemicals for industrial applications, such as CO, CHOOH, HCHO, CH₃OH and CH₄.^6–9 Furthermore, using H₂O as an electron donor is important for artificial photosynthesis, such as in plants, because water is harmless and an abundant reductant.

The photocatalytic reduction of CO₂ using H₂O can be described using the following equation: When the semiconductor photocatalyst is irradiated with ultraviolet (UV) light—which has greater energy than the band gap energy of the photocatalyst—excited electrons (e⁻) and holes (h⁺) are generated at the conduction and valence bands of the photocatalyst, respectively. Subsequently, CO₂ reacts with e⁻ to produce CO (eqn (1)), whereas H₂O is oxidised to O₂ (eqn (2)).

CO₂ + 2H⁺ + 2e⁻ → CO + H₂O (1)

2H₂O + 4h⁺ → O₂ + 4H⁺ (2)

The standard redox potential of CO₂ (E°(CO₂/CO) = −0.11 V versus standard hydrogen electrode (SHE)) is more negative than that of H⁺ (E°(H⁺/H₂) = 0.0 V versus SHE). This results in H₂ evolution caused by the reduction of protons (H⁺, eqn (3)), which is thermodynamically favourable compared with CO formation via CO₂ reduction.

2H⁺ + 2e⁻ → H₂ (3)

Appropriate modification of the photocatalyst surface is necessary to enhance the selectivity of the photogenerated electrons towards the less thermodynamically favoured CO₂ reduction. One method for the selective conversion of CO₂ is the modification of photocatalysts by loading them with metal co-catalysts.^10–12 In 2011, Kudo et al. reported that the BaLa₄Ti₄O₁₅ photocatalyst could be effectively used for the photocatalytic conversion of CO₂ in the presence of an Ag co-catalyst.⁶ Since then, many research groups, including our group, have concurred that the Ag co-catalyst is crucial for the selective conversion of CO₂ in water.^8,13–24 As a modification method of the photocatalyst with Ag co-catalyst, techniques such as impregnation (IMP), photodeposition (PD), chemical reduction (CR) and solution plasma methods have been widely used.^19–24 Building upon these methods, in this study, we investigated ultrasonic reduction (USR) as an advanced method for modification with the Ag co-catalyst. We believe that the preparation of an Ag co-catalyst loaded on a semiconductor photocatalyst with a small size and high dispersity is essential to improve the photocatalytic activity for the conversion of CO₂ with H₂O.

In 1991, amorphous iron nanoparticles were successfully synthesised for the first time by the ultrasonic irradiation of solutions containing volatile organometallic compounds.²⁵ These chemical effects of ultrasound were derived from local high temperatures (>5000 K) and high pressures (>1000 atmospheres), known as ‘hot spots’, formed by the collapse of cavitation bubbles generated in the system during acoustic cavitation.^25–29 Hot spots can result in both physical effects—which include crushing and mixing caused by microjets—and chemical effects—which include the induction of radical reactions and thermal decomposition.³⁰ Since the report by Suslick et al., several research groups have reported the synthesis of nanostructured metal particles via USR.^31–37

Hayashi et al. found that nanosized noble metals, such as Ag, Au, Pt and Pd, could be directly synthesized from the reaction of metal oxides (Ag₂O, Au₂O₃, PtO₂ and PdO) in alcohol using the USR method.^37,38 In addition, they synthesised nanocomposites containing Ag nanoparticles, such as Ag/carbon nanotube,³⁹ Ag/rubber⁴⁰ and Ag/BaTiO₃.⁴¹ Two mechanisms for Ag nanoparticle formation from Ag₂O via USR were proposed.^38,42,43 The first is the direct decomposition and reduction of Ag₂O in the hot spots via ultrasonic cavitation. In this process, hot spots act directly on the surface of the silver oxide particles, and the Ag₂O on the surface is thermally decomposed and reduced to Ag nanoparticles. When the Ag nanoparticles grew to approximately 50 nm on the surface, the interfacial stress between the Ag₂O and Ag nanoparticles reached a limit, and the Ag nanoparticles were desorbed. The second mechanism includes a reaction involving intermediate products such as silver acetate. Ethanol reacts with Ag₂O via ultrasonic waves to form silver acetate, which then decomposes at the hot spot to form Ag nanoparticles. Silver acetate is considered to form and decompose sequentially in small quantities, resulting in the formation of smaller nanoparticles compared to the first reaction mechanism.^42,43

In the application of the USR method to modify photocatalysts with co-catalysts, Pt/TiO₂ and AuPd/TiO₂ (core–shell bimetallic co-catalysts) were successfully prepared, demonstrating photocatalytic activity for H₂ evolution from an ethanol aqueous solution.^44,45 However, to the best of our knowledge, no studies have been reported on evaluating the activities for the photocatalytic conversion of CO₂ with H₂O over Ag-loaded photocatalysts modified by the USR method, except for our previous study.⁴⁶ The USR method is a simplified fabrication method for Ag nanoparticles synthesised by irradiating an alcohol solution containing Ag₂O as the silver precursor with an appropriate ultrasonic frequency. The USR process has many advantages for industrial applications. For example, because Ag₂O is used as a silver precursor and a reducing agent is not used, no contamination by impurities such as nitrate ions occurs. Therefore, the washing process of the obtained particles is easier, and the alcohol solvent can be reused. Unlike the PD method, the USR process does not require to control atmosphere. Furthermore, ultrasonic cleaners used in the USR method are widely used in industry, and technology for large-scale ultrasonic irradiation has been established. Therefore, USR is suitable for industrial applications.

In our previous study, we found that Ag nanoparticles with a particle size of approximately 20 nm were homogeneously loaded onto the surface of gallium oxide (Ga₂O₃) particles by irradiating an alcohol solution containing Ag₂O as a silver precursor and a Ga₂O₃ semiconductor photocatalyst with an appropriate ultrasonic frequency.⁴⁶ However, the characterisation of the Ag co-catalyst prepared using the USR method was insufficient, and the good photocatalytic activity demonstrated by the USR-modified photocatalyst has not yet been clarified.

In this study, the loading of an Ag co-catalyst using the USR method was applied to a zinc tantalate (ZnTa₂O₆) photocatalyst, and the photocatalytic conversion of CO₂ with H₂O over the Ag/ZnTa₂O₆ photocatalyst was investigated. As we reported previously, a ZnTa₂O₆ photocatalyst modified with an Ag co-catalyst showed good activity for the photocatalytic conversion of CO₂,¹⁴ and the photocatalytic activity of the ZnTa₂O₆ photocatalyst was successfully improved by appropriate surface modification with additional Zn species.⁴⁷ However, the loading method of Ag co-catalyst has not been optimised for the ZnTa₂O₆ photocatalyst. To improve the activity for the photocatalytic conversion of CO₂ with water, small Ag nanoparticles must be loaded onto the surface of ZnTa₂O₆ with high dispersity. Therefore, we investigated if such Ag co-catalyst-loaded ZnTa₂O₆ could be prepared using the USR method.

Experimental

Synthesis of ZnTa₂O₆

Zinc tantalate (ZnTa₂O₆) used in this study was synthesised via a typical solid-state reaction, as described in our previous study.⁴⁸ Stoichiometric amounts of ZnO (99.0%, FUJIFILM Wako Pure Chemical Corporation, Japan) and Ta₂O₅ (99.9%, Kojundo Chemical Laboratory Co., Ltd., Japan) were impregnated with 5 mL of pure water and ground in an aluminium mortar for 10 min. The wet mixtures were then dried overnight at 383 K, transferred to an aluminium crucible, and calcined for 50 h at 1273 K under an air atmosphere. X-ray diffraction measurements and UV-visible diffuse reflectance spectroscopy (UV-vis DRS) of the prepared ZnTa₂O₆ were performed, confirming its successful synthesis (Fig. S1 and S2,† respectively).

Modification of ZnTa₂O₆ with the Ag co-catalyst

Four modification methods with Ag co-catalysts were applied: USR, CR, IMP and PD. As a precursor of Ag, silver oxide (Ag₂O; 99.0%, FUJIFILM Wako Pure Chemical Corporation, Japan) and silver nitrate (AgNO₃; 0.1 M aqueous solution, FUJIFILM Wako Pure Chemical Corporation, Japan) were used for USR and the other conventional methods, respectively.

A mixture of ZnTa₂O₆, Ag₂O and ethanol (50 mL; 99.5%, Kanto Chemical Co., Inc., Japan) was sonicated at the frequencies of 28 and 45 kHz (frequency was switched every 0.2 s) for 3 h using an ultrasonic cleaner (WT-100-M, Honda Electronics Co., Ltd., Japan), and the electric power consumption was 100 W. The temperature of the solution was maintained in the range 313–318 K. After sonication, the solution was filtered and dried in air at 333 K for 1 h to obtain Ag/ZnTa₂O₆_USR.

To prepare Ag/ZnTa₂O₆ using the CR method (Ag/ZnTa₂O₆_CR), the AgNO₃ precursor was reduced using an aqueous NaH₂PO₂ solution (0.4 M prepared in house using a special-grade reagent; FUJIFILM Wako Pure Chemical Corporation, Japan) at 353 K for 1.5 h in an aqueous solution. The filtrate was then washed with Milli-Q water and dried at room temperature.

Ag/ZnTa₂O₆_IMP was prepared by impregnating ZnTa₂O₆ with an aqueous solution of AgNO₃ at 353 K for 10 min, followed by evaporation at 353 K for 1 h and drying at 353 K. The obtained sample was calcined at 723 K for 2 h under an air atmosphere.

ZnTa₂O₆ powder was dispersed in 1.0 L of ultrapure water, and the resulting suspension was thoroughly degassed under flowing Ar gas. A 0.1 M AgNO₃ aqueous solution was added to the suspension as a precursor of the Ag co-catalyst, and the resulting system was irradiated for 3 h through a cooling jacket fabricated from quartz glass using a high-pressure Hg lamp (400 W, Sen Lights Corporation, HL400BH-9). The obtained sample was named ‘Ag/ZnTa₂O₆_PD’. The amount of Ag species loaded onto the ZnTa₂O₆ photocatalyst was determined by inductively coupled plasma optical emission spectrometry (ICP-OES; iCAP 7400 ICP-OES DUO, Thermo Fischer Scientific Inc.; see Tables S1 and S2†).

Photocatalytic activity evaluation

The photocatalytic conversion of CO₂ in water was performed inside a quartz inner-irradiation-type reaction vessel using a quasi-flow batch system.⁴⁷ The synthesised photocatalyst (0.5 g) was dispersed in 1.0 L of 0.1 M NaHCO₃ aqueous solution (special grade reagent, Wako, Japan).⁸ High-purity CO₂ gas (99.999%) was continuously bubbled into the solution at a flow rate of 30 mL min⁻¹. After removing the dissolved air, the suspension was irradiated with a high-pressure Hg lamp (400 W, Sen Lights Corporation, HL400BH-9) through a quartz filter equipped with a cooling water system.⁴⁷ The filter was a cylindrical vessel with a double-layered structure.⁴⁷ The Hg lamp was placed on the innermost layer, and the outer layer was in contact with the reaction solution.⁴⁷ The cooling water flowed into the medium layer.⁴⁷ The wavelength of the light emitted from the light source was regulated by the material of the jacket; for example, light with wavelengths <300 nm was cut off in the case of Pyrex® glass, whereas light with wavelengths <300 nm was emitted in the case of quartz glass. The amounts of evolved H₂ and O₂ were detected using online gas chromatograph equipped with a thermal conductivity detector (GC-8A, Shimadzu Corporation, Japan; Ar carrier gas; molecular sieve 5 Å column), and CO was detected using online gas chromatograph equipped with a flame-ionisation detector with a methaniser (GC-8A, Shimadzu Corporation, Japan; N₂ carrier gas; Shincarbon ST column).

The selectivity towards CO formation in a mixture of CO and H₂ and the balance between the consumed electrons (e⁻) and holes (h⁺) were calculated using the following equations:

Selectivity towards CO formation (%) = 100 × R_CO/(R_CO + R_H2) (4)

Consumed e⁻/h⁺ = 2 × (R_CO + R_H2)/(4 × R_O2) (5)

where R_CO, R_H2 and R_O2 are the formation rates of CO, H₂ and O₂, respectively.

Characterization

To clarify the valence and chemical state of Ag co-catalyst, X-ray absorption spectroscopy was conducted in both transmission and fluorescence modes at the BL14B2 beamline of the SPring-8 synchrotron radiation facility (Hyogo, Japan). The photon energy was calibrated at the inflection point of the Ag K-edge X-ray absorption near edge structure (XANES) spectrum using a reference Ag metal foil (25 524 eV). The samples were analysed by UV-vis DRS using a JASCO V-670 spectrometer equipped with an integrating sphere. Transmission electron microscopy (TEM) images were captured using field-emission transmission electron microscope (JEM-2100F, JEOL Co., Ltd., Japan). The Ag cocatalysts were also observed at atomic resolution using high-resolution TEM (JEM-2200FS, JEOL Co., Ltd.).

Results and discussion

Fig. 1 shows the formation rates of H₂, O₂ and CO, as well as the selectivity towards CO evolution at 1 h for the photocatalytic conversion of CO₂ by H₂O over Ag/ZnTa₂O₆ prepared via various modification methods using the Ag co-catalyst. The e⁻/h⁺ balance calculated in eqn (5) was close to 1.0, indicating that H₂O served as an electron donor and proton source for the photocatalytic conversion of CO₂. The selectivity towards CO of Ag/ZnTa₂O₆_USR was >90%, and no obvious difference was observed between the four modification methods (i.e. USR, CR, IMP and PD) using Ag co-catalyst. However, the formation rate of CO varied significantly among the four modification methods using the Ag co-catalyst. The formation rate of CO over the Ag/ZnTa₂O₆_USR was the highest (69.6 μmol h⁻¹) among the four methods and was particularly 2.3 times higher than the formation rate of CO obtained using the conventional method (PD). This difference was not attributed to the amount of loaded Ag co-catalyst because the amount calculated by ICP-OES measurement was almost the same (see Table S1†).

Fig. 1 Formation rates of H₂ (blue), O₂ (green), and CO (red), and selectivity toward CO evolution (circle) at 1 h in the photocatalytic conversion of CO₂ by H₂O over Ag/ZnTa₂O₆ with 0.5 wt% Ag cocatalyst loaded by PD, IMP, CR, and USR.

Fig. 2 shows the Ag K-edge XANES spectra of Ag/ZnTa₂O₆ fabricated using the four modification methods. Each Ag K-edge spectrum closely resembles that of the Ag foil, rather than Ag₂O or AgNO₃, which were used as precursors. These results indicated that the Ag precursors were completely reduced to Ag⁰ and zero-valent Ag particles were loaded onto the surfaces of ZnTa₂O₆ regardless of the method used. In particular, the Ag co-catalyst prepared by the IMP method was in the metallic state, although calcination at 723 K for 2 h was included in the IMP process, which might have oxidised the Ag nanoparticles. However, the Ag nanoparticles on the other photocatalysts fabricated by the IMP and USR methods were also zero-valent (Fig. S3†). As reported by Yamamoto et al.,⁴⁹ the surface of the Ag co-catalyst prepared by the IMP method might be slightly oxidised, but the Ag species in Ag/ZnTa₂O₆_IMP were almost metallic, similar to bulk Ag. Consequently, the Ag K-edge XANES spectra, which provide the macroscopic average information of the bulk, suggested that the valency of the Ag particles was zero.

Fig. 2 XANES spectra of ZnTa₂O₆ with references. Ag species were modified by (a) ultrasonic reduction (USR), (b) chemical reduction (CR), (c) impregnation (IMP), and (d) photodeposition (PD) methods.

The morphology of the Ag co-catalyst on the surface of the photocatalysts was observed using TEM (Fig. 3(a), (c), (e) and (h)). As shown in the TEM images, differences were noted in the size and morphology of the Ag nanoparticles on the surface of ZnTa₂O₆ among the four types of Ag/ZnTa₂O₆. The sizes of the Ag nanoparticles on the surface of ZnTa₂O₆ prepared using the USR, CR and IMP methods were almost the same (Fig. 3(a), (c) and (e), respectively), whereas larger Ag particles were observed on Ag/ZnTa₂O₆_PD (Fig. 3(g)). However, only two Ag nanoparticles were observed on the Ag/ZnTa₂O₆_USR at the magnification and field displayed in Fig. 3(a), whereas more Ag nanoparticles were observed on the surface of ZnTa₂O₆ fabricated using the CR and IMP methods (Fig. 3(c) and (e), respectively). This suggested that much smaller Ag nanoparticles were loaded onto the surface of USR-modified ZnTa₂O₆.

Fig. 3 (a, c, e, and g) TEM and (b, d, f, and h) high-resolution TEM (HRTEM) images of Ag/ZnTa₂O₆ modified by (a and b) ultrasonic reduction (USR), (c and d) chemical reduction (CR), (e and f) impregnation (IMP), and (g and h) photodeposition (PD) methods.

To observe these small Ag nanoparticles, TEM observations were conducted at the atomic resolution level. As shown in Fig. 3(b), Ag nanoparticles with sizes ranging from 1 to 3 nm were present on the surface of ZnTa₂O₆ prepared using the USR method. In contrast, only Ag nanoparticles larger than 5 nm were observed on the surface of ZnTa₂O₆ loaded using other conventional methods (Fig. 3(d), (f), and (h)). Because many CO₂ reduction sites are present on Ag/ZnTa₂O₆_USR compared to other photocatalysts owing to the Ag nanoparticles with a size of a single nanometre, Ag/ZnTa₂O₆_USR exhibited a high formation rate of CO while maintaining high selectivity towards CO. Therefore, to obtain a high conversion of CO₂ using H₂O as an electron donor, the preparation of small Ag nanoparticles on the surface of ZnTa₂O₆ is crucial, which can be achieved using the USR method as a loading method for the Ag co-catalyst.

The UV-vis DRS profiles of Ag/ZnTa₂O₆ fabricated using the four different methods are shown in Fig. 4. The absorption peak and edge of ZnTa₂O₆ at wavelengths of 200–300 nm did not change, even after Ag was loaded using the four modification methods. The absorption peak at wavelengths longer than 300 nm was attributed to the surface plasmon resonance of the Ag nanoparticles.^20,23,50,51 The peak intensity assigned to plasmon absorption for Ag/ZnTa₂O₆_USR was lower than that of other conventional methods. Plasmonic absorption occurs when Ag particles are in the metallic state, and the particle size ranges from a few nanometres to hundreds.^20,51 Additionally, the plasmon absorption intensity depends on the amount of Ag particles and its size.⁵² The Ag K-edge XANES spectrum showed that the Ag co-catalyst on Ag/ZnTa₂O₆_USR, Ag/ZnTa₂O₆_CR, Ag/ZnTa₂O₆_IMP and Ag/ZnTa₂O₆_PD was in the metallic state. Moreover, the loading amount of the Ag co-catalyst determined by ICP-OES was the same for the photocatalysts modified by the four methods (Table S1†). Furthermore, Ag nanoparticles larger than 100 nm were not loaded onto ZnTa₂O₆, as observed by TEM. Thus, the small plasmon absorption suggests that many Ag nanoparticles with sizes in the single-nanometre range or even smaller were loaded onto ZnTa₂O₆ when using the USR method as the fabrication method.

Fig. 4 UV-vis diffuse reflectance spectra of Ag/ZnTa₂O₆ modified by (a) ultrasonic reduction (USR), (b) chemical reduction (CR), (c) impregnation (IMP), and (d) photodeposition (PD) methods.

As depicted in Fig. S4,† the formation rate of CO over Ag/ZnTa₂O₆_USR decreased more rapidly with increasing photoirradiation time compared to those prepared using other modification methods with Ag co-catalysts. Moreover, a reaction test for 10 h (Fig. S5†) showed that the formation rate of CO decreased rapidly in the photoirradiation time range of 0–4 h, but the decrease was slow after 4 h. This phenomenon can be attributed to a change in the size of the Ag nanoparticles on the surface during the reaction. Using TEM, larger Ag nanoparticles were observed after the photocatalytic reaction (Fig. S6†). Furthermore, the intensity of the broad peak in the UV-vis DR spectrum, assigned to plasmon absorption by the Ag nanoparticles, increased after the photocatalytic reaction (Fig. S7†). This increase suggested that the smaller Ag particles present before the reaction tended to aggregate into larger particles.¹⁹ Consequently, the rapid decrease in the formation rate of CO over Ag/ZnTa₂O₆_USR was attributed to the presence of smaller Ag nanoparticles before the photocatalytic reaction, thus corroborating the results depicted in Fig. 4.

Fig. 5 illustrates the formation rates of H₂, O₂ and CO, as well as the selectivity towards CO evolution at 1 h in the photocatalytic conversion of CO₂ by H₂O over Ag/ZnTa₂O₆ with various loadings of the Ag co-catalyst fabricated using the USR method. The formation rate of CO and selectivity towards CO increased from 13 μmol h⁻¹ and 46% to approximately 50 μmol h⁻¹ and 80%, respectively, upon modification with 0.1 wt% Ag co-catalyst. Furthermore, when the Ag loading was increased to 0.5 wt%, the formation rate of CO evolved increased, with a decrease in H₂ evolution. This improvement is attributed to an increase in the number of CO₂ reduction sites. The amount of Ag co-catalyst loaded onto ZnTa₂O₆, where the formation rate of CO was the maximum, was 0.5 wt% when using the USR method as a modification method with the Ag co-catalyst. In contrast, the optimal amount was 3.0 wt% when using the IMP and PD methods.^14,47,48 Using the USR method, higher photocatalytic activity can be achieved and the optimal amount of the Ag co-catalyst can be reduced. This is because single-nanometre Ag co-catalyst can be loaded onto ZnTa₂O₆ using the USR method.

Fig. 5 Formation rates of H₂ (blue), O₂ (green), and CO (red), and selectivity toward CO evolution at 1 h in the photocatalytic conversion of CO₂ by H₂O over Ag/ZnTa₂O₆ with various loading amounts of Ag fabricated by the USR method.

Modifications with higher amounts of Ag (>1.0 wt%) gradually decreased the activity for photocatalytic CO₂ reduction. This decrease could be attributed to the fact that the Ag nanoparticles act as recombination centres for electrons and holes and have a shielding effect on UV light.^53,54 As shown in Fig. S8,† absorption at approximately 300 nm appeared when the loading amount of the Ag co-catalyst was 2.0 wt% or higher. This photoabsorption band overlaps with the original photoabsorption band of ZnTa₂O₆, which may partly hinder the absorption of light by the photocatalyst, causing a decrease in photocatalytic activity. However, the formation rate of CO decreases more rapidly in the range of 0.5–2.0 wt% of Ag co-catalyst rather than in the range of 2.0–5.0 wt%, although Ag/ZnTa₂O₆ with 2.0–5.0 wt% of Ag co-catalyst exhibits more factors that contribute to the lowering of photocatalytic activity. Because Ag nanoparticles only contributed partly to the decrease in photocatalytic activity, another factor should be considered.

At 1.0 wt% or a higher amount of Ag co-catalyst, more Ag particles can be observed using TEM at the magnification displayed in Fig. S9.† Furthermore, the UV-vis DRS profiles shown in Fig. S8† indicate that the peak intensity assigned to the plasmon absorption of Ag nanoparticles increased as the amount of Ag co-catalyst increased, suggesting that the population of large Ag nanoparticles increased. Based on these characterisations, we consider that the decrease in photocatalytic activity is also attributable to the presence of more large Ag particles, rather than a single nanometre, as the loading amount of the Ag co-catalyst increases.

Fig. 6 shows the relationship between the average Ag particle size and the formation rate of CO over Ag/ZnTa₂O₆_USR with various loadings of the Ag co-catalyst. Although the formation rate of CO increased with the loading amount of 0.1 to 0.5 wt%, the average Ag particle size was almost the same. As mentioned above, this increase in photocatalytic activity was due to the presence of more CO reduction sites. However, considering the loading of 0.5 to 5.0 wt%, a clear correlation was observed between the formation rate of CO and the average size of Ag particles, suggesting that smaller Ag nanoparticles are more effective for the photocatalytic conversion of CO₂. Although several previous studies have demonstrated a volcano plot of the relationship between the formation rate of CO and the average size of Ag particles,^55–57 the linear relationship between them suggests that the average perimeter per Ag particle may be a key factor for the photocatalytic conversion of CO₂.⁵⁸ In particular, the peak top of the volcano plot for the USR method should be located in a much smaller region.

Fig. 6 Dependence of the average particle size of Ag on the formation rate of CO over Ag/ZnTa₂O₆ prepared by the USR method with various loading amount of Ag co-catalyst: (a) 0.1, (b) 0.25, (c) 0.5, (d) 1.0, (e) 1.5, (f) 2.0, (g) 3.0, and (h) 5.0 wt%.

As mentioned in the Introduction, two mechanisms of Ag nanoparticle formation from Ag₂O via USR are proposed: the first is the direct thermal decomposition and reduction of Ag₂O, and the second is reaction through intermediate products such as silver acetate.^37,42,43 The Ag particle size formed via the first mechanism was approximately 50 nm, which is much smaller than that formed via the second mechanism.^42,43 In the case of our photocatalyst, we considered the second mechanism to be dominant because few Ag nanoparticles with a size of approximately 50 nm or larger were observed regardless of the amount of Ag co-catalyst loaded (refer to the Ag particle size distribution on Ag/ZnTa₂O₆ with various loading amounts of the Ag co-catalyst shown in Fig. S10†). Hot spots produced by ultrasonic irradiation were randomly generated in an ethanolic solution. Therefore, heterogeneous nucleation of Ag in the solution occurred, followed by the growth and immobilisation of Ag particles on the surface of ZnTa₂O₆ while being uniformly dispersed. The increase in size with increasing loading of the Ag co-catalyst was due to the aggregation of the small Ag nanoparticles. Consequently, small Ag nanoparticles were loaded onto the surface of ZnTa₂O₆ when the Ag co-catalyst had a low loading amount. Therefore, we can conclude that to the preparation of small Ag nanoparticles on the surface of ZnTa₂O₆ is essential to obtain a high conversion of CO₂ using H₂O as an electron donor, which can be achieved using USR as a loading method for the Ag co-catalyst.

Conclusions

In this study, we found that the photocatalytic activity of the ZnTa₂O₆ photocatalyst for the conversion of CO₂ was significantly improved by modification with a Ag co-catalyst using the USR method. Ag nanoparticles with sizes ranging from 1 to 3 nm were loaded onto the surface of the ZnTa₂O₆ prepared using the USR method, although Ag nanoparticles with sizes of approximately 20 nm or larger were also loaded. Conversely, the average diameters of the Ag particles fabricated using the other methods were significantly larger than those produced using the USR method. Therefore, small Ag cocatalysts with single-nanometre dimensions exhibit superior activity towards the selective conversion of CO₂. By utilising the Ag/ZnTa₂O₆ photocatalyst prepared via the USR method, a high CO formation rate with high selectivity towards CO can be achieved.

Data availability

The datasets supporting this article have been uploaded as part of the ESI.†

Author contributions

K. Kawata: investigation, writing – original draft preparation; S. Iguchi and S. Naniwa: visualization, writing – editing; T. Tanaka and K. Teramura: visualization, supervision, funding acquisition; M. Nishimoto: supervision.

Conflicts of interest

The authors declare no competing financial interest.

Acknowledgements

The XAS measurements were performed at the BL14B2 of SPring-8 with the approval of the Japan Synchrotron Radiation Research Institute (JASRI) (Proposal No. 2022B1885). TEM observations at the atomic resolution level were conducted at the Advanced Research Infrastructure for Materials and Nanotechnology in Japan (ARIM Japan). This work was partially supported by “The Mitsubishi Foundation”.

Notes and references

1. J. R. Fernández, S. Garcia and E. S. Sanz-Pérez, Ind. Eng. Chem. Res., 2020, 59, 6767–6772.
2. K. Maeda, J. Photochem. Photobiol., C, 2011, 12, 237–268.
3. S. Xie, Q. Zhang, G. Liu and Y. Wang, Chem. Commun., 2016, 52, 35–59.
4. S. Chen, Y. Qi, C. Li, K. Domen and F. Zhang, Joule, 2018, 2, 2260–2288.
5. Y. Wang, D. He, H. Chen and D. Wang, J. Photochem. Photobiol., C, 2019, 40, 117–149.
6. K. Iizuka, T. Wato, Y. Miseki, K. Saito and A. Kudo, J. Am. Chem. Soc., 2011, 133, 20863–20868.
7. M. Yamamoto, T. Yoshida, N. Yamamoto, H. Yoshida and S. Yagi, e-J. Surf. Sci. Nanotechnol., 2014, 12, 299–303.
8. K. Teramura and T. Tanaka, Phys. Chem. Chem. Phys., 2018, 20, 8423–8431.
9. J. Artz, T. E. Müller, K. Thenert, J. Kleinekorte, R. Meys, A. Sternberg, A. Bardow and W. Leitner, Chem. Rev., 2018, 118, 434–504.
10. A. Kudo and Y. Miseki, Chem. Soc. Rev., 2009, 38, 253–278.
11. R. Daghrir, P. Drogui and D. Robert, Ind. Eng. Chem. Res., 2013, 52, 3581–3599.
12. H. Abdullah, M. M. R. Khan, H. R. Ong and Z. Yaakob, J. CO2 Util., 2017, 22, 15–32.
13. Z. Wang, K. Teramura, S. Hosokawa and T. Tanaka, J. Mater. Chem. A, 2015, 3, 11313–11319.
14. S. Iguchi, K. Teramura, S. Hosokawa and T. Tanaka, Catal. Sci. Technol., 2016, 6, 4978–4985.
15. Z. Huang, S. Yoshizawa, K. Teramura, H. Asakura, S. Hosokawa and T. Tanaka, ACS Sustainable Chem. Eng., 2018, 6, 8247–8255.
16. M. Takemoto, Y. Tokudome, S. Kikkawa, K. Teramura, T. Tanaka, K. Okada, H. Murata, A. Nakahira and M. Takahashi, RSC Adv., 2020, 10, 8066–8073.
17. S. Wang, K. Teramura, T. Hisatomi, K. Domen, H. Asakura, S. Hosokawa and T. Tanaka, Chem. Sci., 2021, 12, 4940–4948.
18. T. Nakamoto, S. Iguchi, S. Naniwa, T. Tanaka and K. Teramura, Catal. Sci. Technol., 2023, 13, 4534–4541.
19. K. Teramura, H. Tatsumi, Z. Wang, S. Hosokawa and T. Tanaka, Bull. Chem. Soc. Jpn., 2015, 88, 431–437.
20. M. Yamamoto, T. Yoshida, N. Yamamoto, T. Nomoto, Y. Yamamoto, S. Yagi and H. Yoshida, J. Mater. Chem. A, 2015, 3, 16810–16816.
21. M. Yamamoto, T. Yoshida, N. Yamamoto, T. Nomoto and S. Yagi, Nucl. Instrum. Methods Phys. Res., Sect. B, 2015, 359, 64–68.
22. R. Pang, K. Teramura, H. Asakura, S. Hosokawa and T. Tanaka, Appl. Catal., B, 2017, 218, 770–778.
23. T. Yoshida, N. Yamamoto, T. Mizutani, M. Yamamoto, S. Ogawa, S. Yagi, H. Nameki and H. Yoshida, Catal. Today, 2018, 303, 320–326.
24. Y. Kawaguchi, M. Akatsuka, M. Yamamoto, K. Yoshioka, A. Ozawa, Y. Kato and T. Yoshida, J. Photochem. Photobiol., A, 2018, 358, 459–464.
25. K. S. Suslik, S. Choe, A. A. Cichowlas and M. W. Grinstaff, Nature, 1991, 353, 414–416.
26. J. P. Loriner and T. J. Mason, Chem. Soc. Rev., 1987, 16, 239–274.
27. K. S. Suslick, Science, 1990, 247, 1439–1445.
28. K. S. Suslick and G. J. Price, Annu. Rev. Mater. Sci., 1999, 29, 295–326.
29. C. Kan, W. Cai, C. Li, L. Zhang and H. Hofmeister, J. Phys. D: Appl. Phys., 2003, 36, 1609–1614.
30. X. Hangxun, B. W. Zeiger and K. S. Suslick, Chem. Soc. Rev., 2013, 42, 2555–2567.
31. J. Zhang, J. Du, B. Han, Z. Liu, T. Jiang and Z. Zhang, Angew. Chem., Int. Ed., 2006, 45, 1116–1119.
32. N. A. Dhas, C. P. Raj and A. Gedanken, Chem. Mater., 1998, 10, 1446–1452.
33. Y. Mizukoshi, R. Oshima, Y. Maeda and Y. Nagata, Langmuir, 1999, 15, 2733–2737.
34. R. A. Caruso, M. Ashokkumar and F. Grieser, Langmuir, 2002, 18, 7831–7836.
35. C. H. Su, P. L. Wu and C. S. Yeh, J. Phys. Chem. B, 2003, 107, 14240–14243.
36. A. Nemamcha, J. L. Rehspringer and D. Khatmi, J. Phys. Chem. B, 2006, 110, 383–387.
37. Y. Hayashi, T. Sekino and K. Niihara, Trans. Mater. Res. Soc. Jpn., 2002, 27, 121–124.
38. Y. Hayashi, H. Takizawa, M. Inoue, K. Niihara and K. Suganuma, IEEE Trans. Electron. Packag. Manuf., 2005, 28, 338–343.
39. T. Yamada, Y. Hayashi and H. Takizawa, Mater. Trans., 2010, 51, 1769–1772.
40. Y. Hayashi, D. Ishikawa, H. Takizawa, M. Inoue, K. Suganuma and K. Niihara, Funtai oyobi Funmatsu Yakin, 2007, 54, 186–193.
41. Y. Hayashi, T. Shishido, K. Seki and H. Takizawa, J. Asian Ceram. Soc., 2023, 11, 464–471.
42. K. Toisawa, Y. Hayashi and H. Takizawa, Mater. Trans., 2010, 51, 1764–1768.
43. Y. Hayashi and H. Takizawa, Funtai oyobi Funmatsu Yakin, 2016, 63, 929–936.
44. Y. Mizukoshi, Y. Makise, T. Shuto, J. Hu, A. Tominaga, S. Shironita and S. Tanabe, Ultrason. Sonochem., 2007, 14, 387–392.
45. Y. Mizukoshi, K. Sato, T. J. Konno and N. Masahashi, Appl. Catal., B, 2010, 94, 248–253.
46. M. Nishimoto, Y. Abe, K. Teramura and T. Tanaka, Funtai oyobi Funmatsu Yakin, 2021, 68, 93–98.
47. S. Wang, K. Teramura, H. Asakura, S. Hosokawa and T. Tanaka, J. Phys. Chem. C, 2021, 125, 1304–1312.
48. X. Xu, K. Teramura, H. Asakura, S. Hosokawa and T. Tanaka, Appl. Catal., B, 2021, 298, 120508.
49. M. Yamamoto, T. Yoshida, N. Yamamoto, T. Nomoto, A. Yamamoto, H. Yoshida and S. Yagi, J. Phys.: Conf. Ser., 2016, 712, 012074.
50. X. Zhu, A. Anzai, A. Yamamoto and H. Yoshida, Appl. Catal., B, 2019, 243, 47–56.
51. J. Zhang, S. Xu and E. Kumacheva, Adv. Mater., 2005, 17, 2336–2340.
52. K. C. Song, S. M. Lee, T. S. Park and B. S. Lee, Korean J. Chem. Eng., 2009, 26, 153–155.
53. K. C. Christoforidis and P. Fornasiero, ChemCatChem, 2017, 9, 1523–1544.
54. A. Primo, T. Marino, A. Corma, R. Molinari and H. García, J. Am. Chem. Soc., 2011, 133, 6930–6933.
55. R. Kaur and B. Pal, J. Mol. Catal. A: Chem., 2012, 355, 39–43.
56. F. F. Schweinberger, M. J. Berr, M. Döblinger, C. Wolff, K. E. Sanwald, A. S. Crampton, C. J. Ridge, F. Jac̈kel, J. Feldmann, M. Tschurl and U. Heiz, J. Am. Chem. Soc., 2013, 135, 13262–13265.
57. M. Eder, C. Courtois, P. Petzoldt, S. Mackewicz, M. Tschurl and U. Heiz, ACS Catal., 2022, 12, 9579–9588.
58. M. Yamamoto, A. Kuwabara and T. Yoshida, ACS Omega, 2021, 6, 33701–33707.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4cy00564c

---