Enhanced photocatalytic removal of nitric oxide over Ag-decorated ZnSn(OH)6 microcubes

Presently, most of the population has been facing a string of severe air pollution problems that include the intensive emission of nitric oxide (NO), which requires a practical approach to sustain our living conditions. Herein, Ag nanoparticles (Ag NPs)-decorated ZnSn(OH)6 microcubes (Ag:cZHS) photocatalysts are synthesized and used for photocatalytic NO removal under solar light activation. The properties of the newly obtained photocatalysts are comprehensively characterized by a series of typical methods. The NO removal performance over the c:ZHS photocatalysts was increased markedly upon being combined with Ag NPs because of the surface plasmon resonance effect. The contribution of electron (e−), hole (h+), hydroxyl radical (•OH), and oxygen radicals (•O2) was investigated through trapping tests and electron spin resonance analysis. Also, the by-products and apparent quantum efficiency of the photocatalysts were thoroughly studied.


Introduction
In recent decades, with the rapid development of advanced industries, environmental pollution has become increasingly severe [1,2]. The mass emission of several toxic gases such as carbon dioxide, methane, and nitrogen oxide (NO x ) from manufacturing, transportation, factory, and power plant activities represents an pressing concern to the global environment and human health [3]. Arguably, the NO x is one of the leading greenhouse gases directly causing global warming [4,5].
Nitric oxide (NO) is a primary component of NO x released during combustion processes [6], the NO removal from the atmosphere is an important task for solving air pollution problems. Extensive research on photocatalytic NO removal from ambient air has been performed using various methods, such as selective catalytic reduction, selective non-catalytic reduction, biofiltration, adsorption, absorption, and photocatalytic technology [7][8][9][10]. Among them, photocatalytic technology has attracted much attention due to its great potential for pollutant degradation efficiency, low-cost, non-toxic, and ecofriendly [11,12]. Recently, a wide range of photocatalysts has been studied for NO gas removals, such as bismuth (III) oxide [13], zinc hydroxystannate microcubics (cZHS) [14], graphitic carbon nitride [15], tin oxide [16] and titanium dioxide [17]. Among these materials, cZHS presents a highly promising photocatalyst, especially in addressing air pollution. However, its wide bandgap (3 .7 eV) remains the key challenge for the widespread use of cZHS under visible light irradiation [14].
Therefore, several methods have been implemented to reduce the bandgap of cZHS materials, including the combination with noble metals (gold, copper, silver), metal oxides (zin oxide, tin oxide) [11,18,19], and nonmetals (carbon, sulfur) [20,21]. The metal decoration is one of the best ways to enhance the photocatalytic performance of the materials thank to the surface plasmon resonance (SPR) effect [22,23]. Herein, silver nanoparticles (Ag NPs) are well-known as a good combination with other semiconductors due to their excellent chemical and conductivity properties [24], which can reduce their bandgap and thereby improve photocatalytic efficiency. The SPR effect from Ag NPS enhances the lifetime of electron and hole pairs by accelerating the transfer of electrons on the surface of the materials. In addition, Ag is the most suitable metal due to its ease of preparation, well-studied, and reasonable cost.
In this work, Ag NPs are decorated on the surface of cZHS materials to enhance the photocatalytic performance through the SPR effect. Specially, cZHS materials are prepared by the hydrothermal method, and the silver nitrate (AgNO 3 ) is reduced by pluronic®127 (F127) to form Ag NPs at the high temperature (≥ 100°C) [25,26]. The photocatalytic NO removal is monitored by an analyzer that met the USEPA requirements for the RFNA-1289-074 method, and the international standard-ISO 22197-1. The photocatalytic mechanism of Ag:cZHS materials is determined experimentally by the photodegradation of NO under solar light. In addition, the characterizations of materials are studied by various techniques. Significantly, the presence of SPR of Ag NPs is identified by diffuse reflectance spectroscopy (DRS). The results of this work promise to provide a new and effective way to address NO pollution.

Preparation of ZnSn(OH) 6 microcubes
The F127 solution was prepared by adding 1.4 g of F127 (CAS: 9003-11-6, Sigma-Aldrich) into 300 mL of deionized water and 200 mL of 98% ethanol. The synthesis process of cZHS is illustrated in Fig. 1. In this work, 7.5 mL of 2 mM zinc acetate dihydrate (98%, C 4 H 10 O 6 Zn, CAS: 5970-45-6, ACROS Organics) and 7.5 mL of 2 mM tin (IV) chloride pentahydrate (SnCl 4 ·5H 2 O, CAS: 10026-06-9, Sigma-Aldrich) were mixed and stirred for 30 min until the solution becomes transparent. Next, 15 mL of 0.28 wt% F127 and 6 mL of 3 M sodium hydroxide solution were added into the mixture and stirred for 30 min to achieve a homogeneous solution, and a white precipitate was formed during this step. Finally, this solution was transferred into a 50 mL Teflonlined stainless steel autoclave for the hydrothermal step at 100°C for 10 h. The samples were rinsed with deionized water and ethanol until the pH reaches the value of about 6.0, followed by a drying step at 80°C to obtain cZHS.

Preparation of Ag nanoparticles-decorated ZnSn(OH) 6 microcubes
The preparation of Ag:cZHS is shown in Fig. 1. A Teflon-lined autoclave for the hydrothermal process was fastly cooled down by tap water. Then, the different amounts of 0.1 M AgNO 3 (CAS: 7761-88-8, Sigma-Aldrich) were added into the hot cZHS and stirred in 30 min. The solution turned to light-gray color after the addition of AgNO 3 . Then, the solutions were rinsed with ethanol and deionized water serval times and dried at 80°C to produce Ag:cZHS with 5, 10, 15, 20 and 30 wt% of the expected Ag denoted as 5% Ag:cZHS, 10% Ag: cZHS, 20% Ag:cZHS and 30% Ag:cZHS, respectively.

The photocatalytic experiments
For the photocatalytic test, 0.2 g of the catalysts was dispersed into a petri dish (d = 14 cm) by 10 mL of deionized water and treated by ultrasonicator for 3 min, followed by a drying step at 80°C for 30 min to eliminate water. Then, the catalyst was placed in a photocatalytic reactor tank (volume = 3 L). The reactor tank is made of stainless steel with a quartz-sealed top to allow light to penetrate the surface of the photocatalyst. The NO analyzer met the USEPA requirements for RFNA-1289-074. First, the NO 100 ppm (N 2 balanced, Ming Yang Co.) was diluted with zero air to reach a concentration of 500 ppb for the experimental process. The gas input was controlled by mass flow controllers; the pressure value of NO source and zero air source was 240 kPa and 310 kPa, respectively. The relative humidity and flow rate adjusted at 50% and 1.5 L min − 1 , respectively. The concentrations of NO, and nitrogen dioxide (NO 2 ) was recorded with a NO x analyzer (model 42c, Thermo-Science). To eliminate the adsorption capacity of the material, the fabric and NO gas interact with each other in the dark environment for about 10 to 15 min until the concentration of the gases reached equilibrium. Then, a 300 W xenon lamp (950 nm > λ > 200 nm) was turned on for 15 min., as the solar power source of this experiment. The full spectra of the xenon lamp are shown in Fig. S1 of Supplemental Materials.
The recycling test attested to the durability of the catalyst. Here, the catalyst was washed with water and dried after completion of each photocatalytic experiment, and repeated five times. The trapping test determined the critical factors of NO photocatalytic degradation. The potassium iodide (KI), isopropyl alcohol (IPA), and dichromate solution (K 2 Cr 2 O 7 ) scavengers were added to trap h + , •OH, and e − [13,27]. The NO degradation, NO 2 conversion, apparent quantum efficiency (AQE), and the kinetic reaction rate were evaluated by Eqs. (S1), (S2), (S3), and (S4), respectively [28,29].

Characterizations
Fourier transform infrared spectrophotometer (FTIR) analysis was established in the range of 400-4000 cm − 1 by using Jasco FT/IR-6500 to investigate molecular vibrations. X-ray diffraction (XRD) patterns were used to determine the materials' phase composition and crystal structure. This analysis was carried out using Bunaciu, Udristioiu, and Aboul-Enein 2015 X-ray diffractometer with the Cu K radiation (λ = 0.154 nm) and the scanning rate of 6 o min − 1 in the 2θ range of 10-80 o . The surface and crystal morphology of the materials was observed by scanning electron microscopy (SEM) and transmission electron microscopy (TEM). The SEM images were captured using Hitachi FE-SEM S-4800 N, and the TEM image was developed by the JEOL JEM 2000FXI model; samples were dispersed in ethanol and coated on a copper mesh for analysis. The high-resolution TEM (HR-TEM) was invested to determine the d-spacing of Ag NPs. X-ray photoelectron spectroscopy (XPS) analysis was also conducted for more investigation about the surface elemental composition and elemental states. Furthermore, the materials' optical property and surface area were examined using diffuse reflectance spectroscopy (DRS) analysis and Brunauer-Emmett-Teller (BET) analyzer, respectively. The electron spin resonance (ESR) was invested to determine the generation of oxygen radicals (•O 2 ) and •OH.

Photocatalytic activity
The photocatalytic NO removal efficiencies of the cZHS and Ag:cZHS materials under solar light are shown in Fig. 2a. The photocatalytic NO removal efficiencies of the blank, cZHS, 1% Ag:cZHS, 5% Ag:cZHS, 10% Ag: cZHS, 20% Ag:cZHS, and 30% Ag:cZHS materials are 13, 54, 52, 62, 68, 87, and 81%, respectively. The photocatalytic NO removal efficiency significantly increased upon being combined with Ag NPs, likely indicating the effect of the SPR of Ag NPs. In detail, the photocatalytic NO removal efficiency of the cZHS increases 8, 14, 33, and 27% with the 5, 10, 20, and 30% of Ag NPs decorated on the cZHS, respectively. However, the photocatalytic NO removal efficiency of the 1% Ag:cZHS decreases by about 2% compared to the cZHS. These results indicate that the photocatalytic activity of the materials can be decreased by adding too little or too much Ag NPs. Besides, the kinetic reaction rate constants are calculated by the Langmuir-Hinshelwood function (Fig.  S2). The reaction rate's constants of the cZHS, 1% Ag: cZHS, 5% Ag:cZHS, 10% Ag:cZHS, 20% Ag:cZHS, 30% Ag:cZHS are 0.14, 0.091, 0.13, 0.19, 0.32, and 0.25 min − 1 , respectively. In addition, the photocatalytic NO removal efficiency of the 20% Ag:cZHS reaches 88% after only 7 min and remains unchanged. In contrast, the photocatalytic NO removal efficiency of the cZHS increases slowly during the reaction and reaches the highest NO removal efficiency of 60% after 15 min. Thus, the 20% Ag:cZHS is the most efficient sample. Practically speaking, the reusability of materials is important, which is evaluated by a series of recycling tests. As shown in Fig. 2b, after the light is turned on, the NO concentration in the reaction decreases sharply and remained plateau for 15 min of the reaction until the light is turned off. The photocatalytic NO removal efficiencies of the 20% Ag:cZHS are 87, 85, 80, 79, and 78% at the first, second, third, fourth, and fifth cycle, respectively. The result shows that the photocatalytic NO removal efficiency of the 20% Ag: cZHS decreases from 87 to 78% after 4 times recycling at the same condition, an insignificant loss of NO removal efficiency after several times of reaction could be owing to the accumulation of degradation product generated from photocatalytic reactions [30]. The results indicates that the durability of the sample is promising, and its photostability is sufficient for long-term usage in reality. The NO 2 selectivity is shown in Fig. S3, the concentrations of NO 2 generated by the cZHS, 1% Ag:cZHS, 5% Ag:cZHS, 10% Ag:cZHS, 20% Ag:cZHS, 30% Ag: cZHS materials is 23, 109, 141, 189, 22, and 182 ppb, respectively. Based on the NO 2 selectivity (Fig. S3), the conversion of NO to NO 2 and other nitrogen species is also calculated as shown in Fig. 2c. In Fig. 2c, the 20% Ag:cZHS showed the lowest NO 2 generation (4%), and the highest other nitrogen species generation (83%). The NO 2 generation efficiencies of the cZHS, 1% Ag:cZHS, 5% Ag:cZHS, 10% Ag:cZHS, 30% Ag:cZHS is 5, 21, 38, 37, and 35%, respectively. Since the main components of NO x are NO and NO 2 , the low generation of NO 2 demonstrates that the material has a good photocatalytic ability to remove NO and NO 2 . The other nitrogen species (N 2 O, NO 2 − , NO 3 − , etc.) generation efficiencies of the cZHS, 1% Ag:cZHS, 5% Ag:cZHS, 10% Ag:cZHS, 30% Ag:cZHS are 54, 31, 24, 30, and 46%, respectively. The AQE is calculated by Eq. (S3) to understand the effect of photons on the photocatalytic ability of the Ag: cZHS and cZHS (Fig. 2d). The AQE (10 − 4 %) values of the cZHS, 1% Ag:cZHS, 5% Ag:cZHS, 10% Ag:cZHS, 20% Ag:cZHS, 30% Ag:cZHS are 2, 2, 2, 3, 4, and 3, respectively. Figure 3a shows the FTIR spectra of the 1% Ag:cZHS, 5% Ag:cZHS, 10% Ag:cZHS, 20% Ag:cZHS, 30% Ag:cZHS and cZHS, in the range from 400 to 4000 cm − 1 . The wide peak observed in 3800 to 2750 cm − 1 shows OH bending and stretching vibrations [11]. Besides, the sharp peak appears at 1170 cm − 1 , attributes to Sn-OH deformation vibration [31]. The stretching vibration of Sn-O-Sn is observed at 779 and 536 cm − 1 [32]. The results obtained confirm the success of the cZHS synthesis, but the signal of Zn and Ag are difficult to detect by the FTIR analysis. Therefore, the XRD diffraction is carried out to confirm the structure and crystallization of cZHS.

Material characterizations
The XRD patterns of all samples are shown in Fig. 3b [33]. However, the peaks that appeared on the XRD pattern could not prove the presence of Ag NPs in the Ag:cZHS samples due to peak overlap. Thus, the HR-TEM is also performed to clarify the issue.

Morphology of materials
The SEM images in Fig. 4 shows the cube shape of the sample before and after Ag NPs loading. As show in Fig.   4a, the cZHS is synthesized with precise cube shapes with smooth surfaces. The cZHS is dispersed well in the solvent, without any conglomeration. It could be seen in Fig. 4b that the original morphology of the cZHS is changed when Ag NPs are loaded onto the surface. The shape and structure of the Ag:cZHS become distinctive. However, the change is not much, which proves that this synthesis method do not change the morphology of the substrate material. The SEM demonstrates that the Ag NPs successfully loaded onto the cZHS surface. The EDS mapping and the wt% of the materials are shown in Fig. S4 and Table S1. Figure 5 shows the morphology and structure of the cZHS and 20% Ag:cZHS. As shown in Fig. 5a and b, the TEM images of the pristine cZHS show precise edges, and the length ranging from 300 to 600 nm matchs well with SEM images. It is observed in Fig. 5c and d that the presence of Ag NPs is  covered on the surface of the cZHS cubes with a diameter less than 10 nm. Furthermore, the dots that appear in Fig. 5d are predicted to be Ag NPs. In contrast, in Fig. 5b, the dots of Ag NPs do not appear; instead, the nanopores are created by the F127. These nanopores are filled with Ag NPs in the 20% Ag cZHS. As show in Fig. 5e and f, the d-spacing of Ag NPs is 0.236 nm, corresponding to the (111) plane of Ag NPs. The HR-TEM results confirm that F127 reduced the AgNO 3 at 100°C to form Ag NPs [34].

The specific surface area of the materials
The N 2 adsorption isotherms of the cZHS and 20% Ag: cZHS are shown in Fig. 6. The adsorbed quantity of the cZHS is higher than that of the 20% Ag:cZHS. In addition, the BET surface areas of the cZHS and 20% Ag:cZHS are 20 and 35 m 2 g − 1 , respectively. The total pore volume and average pore width of the cZHS and the 20% Ag:cZHS are 0.034 cm 3 g − 1 , 6.70 nm, and 0.027 cm 3 g − 1 , 3.97 nm, respectively. These results confirm that the higher total pore volume and average pore width of the cZHS are due to the presence of F127, which increase the porosity of the sample [35]. The hypothesis that the Ag NPs filled the nanopores of the cZHS has been determined by the lower total pore volume and average pore width of the 20% Ag:cZHS. The higher specific surface area of the 20% Ag:cZHS may be explained by the small pore size diameter (3.61 nm) and the nano size of the Ag NPs. The smaller particles have a higher specific surface area [36]. Herein, the Ag NPs with the nanosize affect the particular surface area of the material. These points can be understood as decreasing the particle size employed an increase in the specific surface area. The high specific surface area also confirms that Ag NPs are well dispersed on the surface of the materials [37].

Optical properties of materials
The optical absorption properties of the cZHS and 20% Ag:cZHS are shown in Fig. 7. In Fig. 7a, the 20% Ag: cZHS could reflect photons with a wavelength in the range of UV light (400 nm > λ > 200 nm). The wavelength of about 260 and 360 nm are the peaks of SPR, which are generated by Ag NPs. Besides, the formation of a peak of SPR in the range of UV (240 nm < λ < 260 nm) and the DRS tail of the 20% Ag:cZHS do not coincide with the Ox axis in the range of IR. In contrast, in the cZHS, there is a peak present at around 360 nm. The morphology of this peak is bizarre and attributed to the manipulation of the DRS operator. However, after repeated measurements of DRS, we conclude that the appearance of this peak of the cZHS sample is due to the influence of F127 on the surface of the material. At the wavelength of about 350 to 380 nm, there is a superposition between the two peaks, so the peak of F127 may have been masked by the SPR peak of Ag NPs. The previous studies also determined the superposition of the peak at 360 nm of F127. The peak at 360 nm could be covered by any peaks that have more vigorous intensity than this at the same wavelength [38,39]. As show in Fig. 7b, Kubelka-Munk plots of DRS reveal that the band The surface chemistry of materials is shown in Fig. 8. Figure 8a shows that the peaks of Sn 3d, O 1 s, and Zn 2p of cZHS. Furthermore, the peak around 370 eV of the 20% Ag:cZHS corresponds to the Ag 3d. The HR-XPS of Sn 3d5, O 1 s, and Zn 2p3 are shown in Fig. 8b, c, and d, respectively. The peaks at 485.3 eV, 493.8 eV correspond to Sn 3d 5/2 , Sn 3d 3/2 for Sn 4+ state in the cZHS, respectively [40]. The peak at 531 eV, 1022 eV correspond to O 1 s, and Zn 2p 3/2 , respectively. As show in Fig. 8c, the peak at 531.6 eV is assigned to the metal-oxygen bonds (Sn-O or Zn-O), the peak at 530.7, and 532 eV are assigned to the metal-hydroxyl bonds (Zn-OH or Sn-OH) and hydroxyl absorbed on the surface of the materials [41]. In Fig. 8d, the peak at 1021.8 eV is assigned to the Zn-hydroxyl (Zn-OH) bonding of the cZHS [42]. The HR-XPS of Ag 3d is shown in Fig. 8e; two peaks of Ag 3d at 367.6 and 373.6 eV, corresponding to Ag 3d 5/2 and Ag 3d 3/2 for Ag 2+ in the 20% Ag:cZHS, respectively. The peaks at 367.9 eV and 373.6 eV are assigned to the Ag + in silver-oxygen interaction (Ag-O) and metallic Ag [43]. In addition, XPS is a directly used tool to measure the valence-band maximum (VBM) of the materials. As show in Fig. 8f, the VBM of the cZHS is 2.55 eV, and 20% Ag:cZHS is 4.08, the VBM of 20% Ag:cZHS is 1.53 eV higher than the cZHS, it is revealed that the electrons provided by Ag lead to the enhancement of electron transport in the composites.

Photocatalytic mechanism over the materials
The trapping test considers the critical factors of photocatalytic degradation of NO, as show in Fig. 9. In the trapping test (Fig. 9a), the h + , e − , and •OH are trapped by adding the KI, K 2 Cr 2 O 7 , and IPA scavengers, respectively. The photocatalytic NO removal efficiencies of the 20% Ag:cZHS are 62, 57, and 37% after 15 min with the trapping of •OH, h + , e − , respectively. The trapping results indicate that e − and •OH almost have the same influence on the photocatalytic activity of 20% Ag:cZHS. The photocatalytic efficiency of the 20% Ag:cZHS is decreased dramatically (37%) by adding KI. The ESR is invested in determining the generation of radicals of 20% Ag:cZHS. Figure 9b and Fig. S5 shows that the 20% Ag:cZHS generate more •OH than •O 2 . However, the intensity of •O 2 and •OH are also strong. These results explain that in the 20% Ag:cZHS, the •OH is formed by h + , and the e − react with O 2 to form •O 2 at the same time.
Based on the experimental evidence and previous studies [11,14,44], the proposed mechanism on the photocatalytic NO removal of the Ag:cZHS under solar light is presented in Fig. 10. Under light activation, the e −h + pairs are generated in the cZHS. The photogenerated e − spontaneously move to the conduction band of the cZHS and is isolated by the interface between the cZHS and the Ag NPs [45]. These e − reduce the adsorbed O 2 on the surface to produce •O 2 , as evidenced from the ESR results. The photo-generated e − in the Ag NPs and cZHS is transported between the Ag NPs and cZHS to increase the lifetime of the e −h + pairs. The photo-generated h + in the valance band of the cZHS oxidize the adsorbed water on the surface of the cZHS to form •OH radicals. The •O 2 and •OH radicals assist in the removal of NO [46]. Thus, the introduction of the Ag NPs drastically enhance photocatalytic NO removal by the cZHS through the surface plasmonic effect, resulting in a better photoresponse and excellent electron-hole pairs separation. Besides, the NO degradation pathway is also shown in Fig. 10. Firstly, the •O 2   from NO and NO 2, is also predicted by other studies [47][48][49]. However, the selectivity of those gas is not recorded because of the limitation of the equipment.

Conclusions
In this work, the Ag:cZHS nanocomposite photocatalysts were prepared by a rapid and straightforward physical mixing process. The as-prepared Ag:cZHS nanocomposite photocatalysts showed a remarkable improvement of NO removal under solar light through the surface plasmonic effect from Ag NPs. With 20% of Ag loading, a photocatalytic performance of 87% was achieved, much higher than that of the bare cZHS (52%). This sample also had a good photostability and recycling ability with a photocatalytic performance of approximately 77% even after five repetitive runs. Besides, the effect of SPR on photocatalytic activity was confirmed by DRS. The peak of SPR appeared at the wavelength of about 360 nm and 260 nm. The trapping results and ESR determined that the h + and •OH were the main factors in the photocatalytic activity of the 20% Ag:cZHS. This study presents the promise of Ag:cZHS nanocomposite photocatalysts for addressing air pollution.