Sunlight-Driven Combustion Synthesis of Defective Metal Oxide Nanostructures with Enhanced Photocatalytic Activity.

Synthesis of metal oxide nanostructures through combustion routes is a promising technique owing to its simplicity, rapidity, scalability, and cost-effectiveness. Herein, a sunlight-driven combustion approach is developed to synthesize pristine metal oxides and their heterostructures. Sunlight, a sustainable energy source, is used not only to initiate the combustion reaction but also to create oxygen vacancies on the metal oxide surface. ZnO nanostructures are successfully synthesized using this novel approach, and the products exhibit higher photocatalytic activity in the decomposition of methyl orange (MO) than ZnO nanostructures synthesized by the conventional methods. The higher photocatalytic activity is due to the narrower band gap, higher porosity, smaller and more uniform particle size, surface oxygen vacancies, as well as the enhanced exciton dissociation efficiency induced by the sunlight. Porous Fe3O4 nanostructures are also prepared using this environmentally benign method. Surprisingly, few-layer Bi2O3 nanosheets are successfully obtained using the sunlight-driven combustion approach. Moreover, the approach developed here is used to synthesize Bi2O3/ZnO heterostructure exhibiting a structure of few-layer Bi2O3 nanosheets decorated with ZnO nanoparticles. Bi2O3 nanosheets and Bi2O3/ZnO heterostructures synthesized by sunlight-driven combustion route exhibit higher photocatalytic activity than their counterparts synthesized by the conventional solution combustion method. This work illuminates a potential cost-effective method to synthesize defective metal oxide nanostructures at scale.

Introduction

Nanostructured metal oxide semiconductors have impressive momentum in photocatalytic water treatment applications. However, pristine metal oxide photocatalysts suffer from some serious drawbacks, which negatively affect their photocatalytic performance, including short lifetime of photogenerated electrons and holes which are essential for the photocatalytic process and/or wide band gap, which negatively impacts the visible light response.[1,2] Different approaches have been introduced to overcome these drawbacks, including doping, co-doping, and construction of metal-oxide-based heterostructures.[3−7]

Surface defect engineering approach attracts recent research attention because it provides a platform to overcome the above-mentioned issues without adding any external dopants.[8−10] In this regard, surface oxygen vacancies (VOs) increase the valence band maximum (VBM) or decrease the conduction band minimum (CBM) of the defective metal oxides, leading to narrowed band gap and enhanced light absorption of longer wavelength. VOs also work as temporary traps for the photogenerated electrons, suppressing the electron–hole recombination. Moreover, VOs increase the adsorption ability of the defective metal oxides for the O2 and CO2 molecules.[11−15] Recent reports on this subject also reveal that the VOs are able to dissociate excitons into free electrons and holes leading to higher photocatalytic performance.[16] All of these remarkable characteristics of VOs substantiate its effect in regulating the photocatalytic activity of metal oxides.

Superior photocatalytic activity for a metal oxide is also attained by reducing its size to the nano range or by creating pores on its surface.[17−20] Both the size reduction and pore formation are often achieved by chemical or physical methods.[19,21−23] However, most of these methods are expensive and time-/energy-consuming. Additionally, generation of VOs is commonly achieved by post-treatment of the synthesized metal oxide, which is further time- and energy-consuming.[24]

Two-dimensional (2D) nanomaterials beyond graphene attracted widespread research attention due to their exceptional chemical and physical properties.[25] Two-dimensional metal oxide nanosheets are excellent candidates for photocatalytic applications due to their unique geometry with high specific surface area, which provides a large number of photocatalytic reaction sites and enables rapid charge photogeneration and high charge separation and transport efficiency.[26] In this aspect, 2D Bi2O3 nanosheets (pristine, doped, and heterostructures) have been synthesized using different routes and used for photocatalytic applications.[27−32] However, unscalable and/or unsustainable synthesis methods have been used to synthesize such 2D materials.

Solution combustion synthesis is a rapid and facile method to synthesize metal oxide nanostructures. In this process, an exothermic reaction takes place between a fuel (e.g., urea and glycine) and an oxidizer (normally metal nitrate), releasing a large amount of gases. Along with its simplicity and rapidity, higher product yield makes the method of solution combustion an ideal synthesis route, especially for industrial applications. In addition, the large volume of gases associated with this method develop a porous structure for the synthesized product.[33−38]

However, solution combustion is not completely reliable as it requires heat energy to initiate the exothermic combustion reaction. Also, applying electrical heating does not produce homogeneous preheating, which negatively affects both the particle size and pore size distribution.[33,39,40] Moreover, no such solution combustion synthesis routes are reported in the literature to directly produce VOs in metal oxides.

Herein, a solar-driven combustion synthesis to prepare metal oxides is introduced for the first time. A commercial convex lens is used to concentrate the normal sunlight irradiation (see Supporting Information Video 1). Sunlight irradiation initiates the exothermic reaction and creates oxygen vacancies on the metal oxide surface. Direct exposure to sunlight irradiation achieves uniform heating. It is worthy to note that the synthesis takes place at any time on a sunny day irrespective of the sunlight intensity. Moreover, the formation of nanoparticles immediately starts upon exposing the precursor to the sunlight and higher yields are obtained in short time (ca. 5 s). The process developed could be of high interest for industrial applications due to its simplicity, rapidity, sustainability, cost-effectiveness, and large-scale production. ZnO, Fe3O4, and Bi2O3 are selected as model metal oxides for the synthesis using the introduced method. The selection was based on their wider environmental photocatalytic applications.[41−45]

Experimental Section

Materials

Zinc nitrate hexahydrate (Alfa Aesar, 99%), bismuth nitrate pentahydrate (Sigma-Aldrich, 98%), ferric nitrate nonahydrate (Sigma-Aldrich, ≥98%), glycine (Sigma-Aldrich, 99%), sucrose (Sigma-Aldrich, 99.5%), 5,5-dimethyl-1-pyrroline N-oxide (Sigma-Aldrich, ≥97%, DMPO), 2,2,6,6-tetramethylpiperidine (Sigma-Aldrich, 98%, TEMP), and methyl orange (Loba Chemie) were used as received. The water used for synthesis and other experiments was deionized (DI).

Synthesis

Synthesis of ZnO

To synthesize ZnO nanoparticles, 0.297 g of zinc nitrate hexahydrate (oxidizer) and 0.065 g of sucrose (fuel) were mixed and ground for 10 min to obtain a homogeneous mixture. The mixture was exposed to concentrated sunlight. The sunlight intensity at the surface of the mixture was measured to be ∼1256 W m–2. The reaction started immediately producing very light nanopowder. The synthesis successfully took place in the morning, noon, and afternoon, indicating that a normal sunny day is enough to carry out the synthesis, irrespective of the light intensity. For comparison, ZnO nanoparticles were synthesized from the same precursor using the conventional solution combustion (CSC) synthesis on a hot plate. The corresponding sample was marked as ZnO-CSC.

ZnO nanoparticles were also synthesized using the hydrothermal method as follows. About 2.97 g of Zn(NO3)2·6H2O was dissolved in 100 mL of 0.5 M NaOH. The solution was stirred for 40 min at room temperature and then transferred to a Teflon-lined steel autoclave. The hydrothermal reaction was carried out at 120 °C for 8 h. Afterward, the autoclave was allowed to cool to room temperature. The product obtained was then washed three times with DI water and dried at 80 °C for 12 h.

Synthesis of Fe3O4

To validate the repeatability of the introduced synthesis method, Fe3O4 nanoparticles were also prepared following the same method. About 4.04 g of ferric nitrate nonahydrate and 0.95 g of glycine were ground, and the mixture was exposed to concentrated sunlight irradiation (Video 1). The sunlight intensity at the surface of the mixture was measured to be ∼1252 W m–2.

Synthesis of Few-Layer Bi2O3 Nanosheets

To synthesize few-layer Bi2O3, about 4.85 g of bismuth nitrate pentahydrate was mixed with 1.14 g of sucrose. The solid mixture was ground for 10 min to obtain a homogeneous mixture. The mixture was exposed to concentrated sunlight irradiation (Scheme ). The sunlight intensity at the surface of the mixture was measured to be ∼1264 W m–2.

Scheme 1

Schematic Illustration of the Setup of Sunlight-Driven Combustion (SDC) Synthesis of Metal Oxide Nanostructure

Synthesis of Bi2O3/ZnO Heterostructure

To assure its applicability to synthesize metal oxide heterostructures, SDC was used to synthesize Bi2O3/ZnO heterostructure. The precursors, bismuth nitrate pentahydrate and zinc nitrate hexahydrate, act as oxidizers, and sucrose as fuel. The Bi-to-Zn ratio was 1:0.7.

Characterizations

The as-prepared samples were analyzed using a Rigaku Smart Lab II (Cu Kα radiation, λ = 1.5414 Å) X-ray diffractometer to study the crystal structure, a Hitachi (S-3400N) scanning electron microscope, and a Jeol/JEM 2100 transmission electron microscope to investigate the morphology and structure; an ESCALAB 250Xi Thermo Scientific X-ray photoelectron spectroscope was used to analyze the surface electronic states and surface defects; and a Thermo Scientific Evolution 201 UV–vis diffuse reflectance spectrometer was used to record the absorption spectra.

Photocatalysis Test

For the photodegradation test, the mixture solution of methyl orange (MO, 10 mg L–1) or rhodamine B (RhB, 10 mg L–1) and the photocatalyst (0.5 g L–1) was stirred for 30 min in the dark to ensure the adsorption/desorption equilibrium before sunlight irradiation. The sunlight intensity at the surface of the suspension was measured to be ∼335 W m–2. The photodegradation of MO or RhB was evaluated by monitoring the change in absorbance of mixture solution with a Beckman Coulter DU 730 UV–vis spectrophotometer.

Mott–Schottky Measurements

To prepare the working electrodes, 5 mg of a photocatalyst was ultrasonically dispersed in 2 mL of isopropanol alcohol and 40 μm Nafion solution. The suspension was spin-coated on FTO glass substrates (1 × 2 cm2). The coated substrates were dried at ambient temperature in air followed by calcination at 400 °C for 30 min. A 608E CH instrument was used here to carry out the measurements. A platinum wire and an Ag/AgCl electrode were used as counter and reference electrodes, respectively. Na2SO4 aqueous solution (0.5 M) was used as electrolyte. The recorded potential values were converted to normal hydrogen electrode (NHE) scale. Measurements were recorded at a frequency of 1000 Hz, amplitude of 10 mV, and pH 7.

Electron Spin Resonance (ESR)-Trapping Tests

About 4 g L–1 of the sample was suspended in 50 μL of DI water followed by addition of 500 μL of the spin-trapping agent. The mixture was irradiated for 2 min and analyzed at the X-band frequency (9.4 GHz) at ambient temperature using a Bruker spectrometer (EMX plus model). The spin-trapping agents were an aqueous solution of TEMP (10 mM) and a methanol solution of DMPO (20 mM) for 1O2, and •O2, respectively.

Results and Discussion

Characterizations of ZnO

The sunlight-driven combustion synthesis route is schematically represented in Scheme . The convergence of solar radiation to the metal nitrate–glycine mixture and the formation of a porous nanostructure are clear from the scheme. Figures and S1 show the characteristic features of ZnO prepared by the sunlight-driven combustion (ZnO SDC), conventional solution combustion (ZnO CSC), and hydrothermal (ZnO Hy) routes. The X-ray diffraction (XRD) patterns in Figures a and S1a confirm the formation of pure hexagonal wurtzite phase of ZnO in all three samples (JCPDS 36-1451).[46] The ZnO SDC shows lower crystallinity with smaller crystallite size than ZnO CSC while ZnO Hy exhibits the highest crystallinity and the largest crystallite size. The scanning electron microscopy (SEM) images show the 3D spongelike structure of both ZnO SDC and ZnO CSC samples (see Figure b) and the irregular-shaped nanoparticles of ZnO Hy (see Figure S1b). Obviously, the ZnO SDC exhibits a much higher and uniform distribution of pores on its surface compared to ZnO CSC.

Figure 1

Structural and morphological comparison between ZnO nanostructure synthesized by sunlight-driven combustion (ZnO SDC) and its counterpart synthesized by a conventional solution combustion method (ZnO CSC). (a) X-ray diffraction (XRD) patterns, (b) scanning electron microscopy (SEM) images, (c) transmission electron microscopy (TEM) images, (d) X-ray photoelectron spectroscopy (XPS) survey spectra, (e) O 1s XPS images, and (f) Zn 2p XPS images of ZnO SDC and ZnO CSC.

The Brunauer–Emmett–Teller (BET) surface area and pore size distribution of the ZnO SDC and ZnO CSC samples were investigated. Both samples exhibit a typical type IV isotherm featuring a pronounced H3-type hysteresis loop demonstrating the existence of mesoporous structure (Figure S2a,b).[47] The BET surface area was found to be 31.46 and 32.14 m2 g–1 for ZnO CSC and ZnO SDC, respectively. The Barrett–Joyner–Halender (BJH) pore size distribution graph (inset of Figure S2a) reveals that the ZnO CSC sample has a broad pore size distribution ranging from 5 to 48 nm. The average pore diameter is measured to be 18.3 nm. For the sample ZnO SDC, the BJH pore size distribution was quite narrow (inset of Figure S2b) and the average pore diameter was measured to be 14.8 nm. These results are in good agreement with the SEM observations. It should be mentioned that the sharp peaks at about 3.6 nm in the pore size distribution curves for both samples are artifacts of the BJH method due to the tensile strength effect.[48,49]

ZnO SDC particles are more uniform in size and smaller than the ZnO CSC, as evident from the TEM image (see Figure c). These results can be attributed to the uniform heating induced by the direct exposure of sunlight radiation.

Both ZnO SDC and ZnO CSC samples were subjected to XPS analysis to confirm their purity as well as to investigate the presence of surface defects on their structure. Figure d displays the survey spectra of ZnO SDC and ZnO CSC. Only Zn, O, and a negligible amount of C are seen in the XPS survey spectra. The C peak is usually associated with the XPS sample holder. Figure e shows a comparison between the high-resolution O 1s XPS spectra for ZnO SDC and ZnO CSC and their fits. Each O 1s peak is fitted into two peaks: OI (530.18 and 530.13) is attributed to lattice oxygen and OII (531.56 and 531.53 eV) is attributed to VOs. Apparently, the OII/OI intensity ratio for ZnO SDC is higher than that of ZnO CSC, indicating the presence of a higher concentration of VOs in ZnO SDC.[50] Zn 2p XPS spectra of both samples show doublet peaks related to Zn 2p3/2 and Zn 2p1/2 (see Figure f) configurations. Zn 2p of ZnO SDC shows a downward shift compared to that of ZnO CSC, further confirming the presence of a considerable concentration of VOs in ZnO SDC. More clearly, the creation of VOs leads to an increase in the electron cloud density, resulting in the downward binding energy shift of Zn 2p peak.[51,52]

How does sunlight create VOs? It is widely accepted that at atmospheric temperature and pressure, UV–vis photons are able to produce VOs on the surface of metal oxide in the first step of the photothermochemical cycle.[53−55] The same mechanism is applicable here and thus photons of solar radiation produce VOs on the surface of metal oxide. The following reaction happens on the metal oxide (MO) surface when it is exposed to sunlight photons.Thus, under sunlight irradiation, MO is reduced to MO.

Photocatalytic Performance and Mechanistic Study

To further demonstrate the advantages of the method developed, the photocatalytic efficacy of ZnO SDC is compared to that of ZnO CSC, ZnO Hy and the standard photocatalyst TiO2-P25. Figure a depicts the adsorption (in the dark) and photocatalytic degradation of methyl orange (MO) over ZnO SDC, ZnO CSC, ZnO Hy and TiO2-P25 under natural sunlight irradiation as a function of irradiation time, and Figure b shows the photodegradation kinetics. A negligible decrease in the MO concentration is observed during the first 20 min of stirring in the dark because of the dye adsorption on the photocatalysts surfaces. On the other hand, in the absence of any catalyst, only 8% degradation of MO occurred under sunlight, which can be attributed to photolysis of MO molecules. It can also be seen that 89, 81, 71, and 45% of MO degraded over ZnO SDC, TiO2-P25, ZnO-CSC, and ZnO Hy, respectively, after 60 min of sunlight irradiation (Figure a). In fact, from Figure b, one can observe that the photodegradation rate constant for ZnO SDC (0.038 min–1) is ∼4.2 times that for ZnO Hy (0.009 min–1), ∼1.9 times that for ZnO CSC (0.02 min–1), and ∼1.3 times that for TiO2-P25 (0.029 min–1). Hence, ZnO SDC exhibits the highest photocatalytic performance.

Figure 2

Photocatalytic performance, optical properties, band positions, ESR test, and free radicals production mechanism. (a) Photocatalytic degradation curves of methyl orange (MO, C: initial concentration of MO, C: concentration of MO after t min of sunlight irradiation) and (b) degradation kinetic curves of MO over ZnO SDC, ZnO CSC, and TiO2 Degussa (P25) under sunlight irradiation. (c) Diffuse reflectance spectroscopy (DRS) spectra, (d) Tauc plot, (e) valence band XPS spectra, (f) Mott–Schottky plots at a fixed frequency of 1 kHz in the dark and a potential amplitude of 10 mV, (g) ESR spectra in the presence of TEMP in water and in the presence of DMPO in methanol, and (h) schematic diagram of energy band structure and free radicals production during the photocatalysis process for ZnO CSC and ZnO SDC.

The durable photostability of the photocatalyst is the key factor for its commercialization. The photostability was examined by recycling ZnO SDC for four sequential cycles, and the results are displayed in Figure S3a. A negligible decrease in MO degradation was observed after four consecutive cycles, confirming the photostability of ZnO SDC. To ensure the structural stability of ZnO SDC, XRD was reexamined before and after the recycling experiments. As shown in Figure S3b, the XRD pattern of ZnO SDC remains almost the same. Thus, ZnO SDC is photostable under long sunlight irradiation. This antiphotocorrosion property of ZnO SDC might be attributed to the presence of a considerable concentration of oxygen vacancies.[13,15]

To understand the mechanism behind the enhanced photocatalytic activity of ZnO SDC, the optical absorption, band gap, CBM potential, and VBM potential measurements were carried out using diffuse reflectance spectroscopy (UV–vis DRS) analysis. Compared to ZnO CSC, the light absorption of ZnO SDC is broadened to the visible light region (Figure c). The band gaps (Eg) calculated from Tauc plot are 3.07 and 3.19 eV for ZnO SDC and ZnO CSC, respectively (Figure d).

CBM and VBM are calculated according to the following equations[56]where χ is the ZnO electronegativity (5.79 eV) and EH denotes the energy of free electrons on the hydrogen scale (4.5 eV).[57] CBM and VBM are found to be, respectively, ca. −0.31 and 2.89 eV for ZnO CSC and ca. −0.25 and 2.83 eV for ZnO SDC.

VBM and CBM are further experimentally estimated using the Mott–Schottky and valence band XPS (VB-XPS) analyses. VP-XPS measurements show that VBM potential is 2.67 and 2.83 V for ZnO SDC and ZnO CSC, respectively (see Figure e). Also, the Mott–Schottky plots displayed in Figure f indicate that both ZnO SDC and ZnO CSC are n-type semiconductors as they have positive slopes. For n-type semiconductors, the Mott–Schottky flat-band potential is approximately equal to that of CBM. Therefore, the CBM potential for both ZnO SDC and ZnO CSC is equal to −0.36 V.[58]

To identify the different photocatalytic reactive oxygen species (ROS) products, electron spin resonance (ESR) measurement was employed. Here, 2,2,6,6-tetramethylpiperidine (TEMP) was used as the trapping reagent to verify the singlet oxygen (1O2) generation. As displayed in Figure g, the typical 1:1:1 triplet signal with a g-value of 2.0056 for both samples ZnO CSC and ZnO SDC agrees well with those of 2,2,6,6-tetramethylpiperidine-N-oxyl (TEMPO) verifying the 1O2 generation by both samples. The stronger ESR signal for ZnO CSC confirms its higher 1O2 generation.[16,59] On the other hand, 5,5-dimethyl-1-pyrroline N-oxide (DMPO) was employed as the trapping agent for •O2 detection in methanol. The sextet ESR signal for both samples ZnO SDC and ZnO CSC shown in Figure g verifies the DMPO–OOH formation, which verifies the •O2 generation by both samples. The stronger ESR signal for the sample ZnO SDC indicates its higher •O2 production.[16] Finally, the higher •O2 production and lower 1O2 generation exhibited by ZnO SDC indicate its higher exciton dissociation efficiency, which leads to higher concentration of free electrons and holes and thereby higher photocatalytic performance.

To determine the role of hydroxyl radicals (•OH) and holes (h+) in the MO photodegradation, potassium iodide (KI) and tert-butyl alcohol (TBA) were added into photocatalyst/dye suspension as scavengers to trap h+ and •OH, respectively, followed by the photocatalytic assessments.

It can be seen from Figure S4 that the addition of TBA (•OH scavenger) greatly decreased the photocatalytic performance of ZnO SDC and ZnO CSC, indicating that •OH is the main active species in the degradation reaction for both samples. In contrast, the MO degradation rate increased in the presence of KI (h+ scavenger); about 97 and 71% of MO were degraded by ZnO SDC and ZnO CSC, respectively, just in 40 min. This finding indicates that the h+ consumption in the KI solution suppressed the electron–hole recombination rate. Thus, the MO degradation was enhanced due to the increase in the photogenerated electron concentration.[60]

Based on the above results, the band structure diagrams of ZnO SDC and ZnO CSC are presented in Figure h. The standard redox potential of •OH/OH– is +1.99 V vs NHE, and the standard redox potential of O2/•O2 is −0.046 V vs NHE.[61,62] One can see that the CBM and VBM potentials for both ZnO SDC and ZnO CSC are enough to produce superoxide (•O2) and hydroxyl (•OH) free radicals. In addition, the VBM for ZnO SDC is higher than that for ZnO CSC. This is attributed to the overlapping of VOs state with VB leading to its upward shift in ZnO SDC. Generally, the narrower band gap is associated with higher visible light absorption and thereby higher solar photocatalytic activity. Moreover, the presence of a considerable concentration of VOs enhances the photocatalytic activity by lowering the electron–hole recombination rate, enhancing the exciton dissociation (as confirmed by ESR), and increasing the O2 adsorption at the defect sites.[63,64] Besides, the more uniform particle and pore size distribution increase the contact between the nanoparticles and the organic pollutant (MO), leading to the higher photocatalytic activity.

Repeatability of the Developed Method

Solar-driven combustion synthesis is also carried out to test the repeatability and reliability of the method by preparing Fe3O4. A recorded video demonstrating the simplicity of the method and its potential in synthesizing Fe3O4 is included as Supporting Information Video 1.

The XRD pattern for sunlight-driven combustion-synthesized Fe3O4 is shown in Figure S5. The peaks at 30.49, 35.78, 36.28, 43.25, 53.95, 57.27, 63.07, and 74.41° are related to the (220), (311), (222), (400), (422), (511), (440), and (533) planes of cubic spinel structure of Fe3O4 (ICDD no. 72-8151).[65] The XRD analysis confirms the production of Fe3O4 by sunlight-driven combustion method. The weak peaks at 37.22, and 50.19° are, respectively, related to the (110) and (024) planes of α-Fe3O4 (ICDD no. 85-0987),[65] indicating the presence of a small of amount of α-Fe2O3. Figure S6a,b displays the morphology of Fe3O4 SDC. The porous structure is the signature of released gases during the combustion. The photograph in Figure S6c shows a spongy dark red powder, indicating the presence of a small amount of α-Fe2O3.[66] This supports the XRD analysis. The consistency in the nanostructure formation for both ZnO and Fe3O4 confirms the reproducibility of the sunlight-assisted combustion method.

Synthesis of Few-Layer Bi2O3 Nanosheets

Surprisingly, we succeeded to synthesize few-layered Bi2O3 nanosheets using sunlight-driven combustion approach from bismuth nitrate pentahydrate (oxidizer) and sucrose (fuel). The XRD pattern of synthesized Bi2O3 is presented in Figure a, in which the peaks at 21.66, 25.90, 27.00, 27.49, 33.43, 35.2, 37.02, 46.44, 52.59, and 54.93° are, respectively, related to the (020), (002), (111), (120), (200), (−212), (121), (041), (−321), (−241) lattice planes of monoclinic α-Bi2O3 (JCPDS no. 71-2274).[41]Figure b shows the XPS survey spectrum of Bi2O3. All of the peaks are related to Bi and O elements except the one at 284.6 eV related to C 1s from the XPS sample holder confirming the purity of the prepared sample. Figure c shows the high-resolution XPS O 1s spectrum, in which the O 1s peak is fitted into two peaks at 528.9 and 531.0 eV related to lattice oxygen and VOs, respectively. The high-intensity VOs peak indicates that the Bi2O3 nanosheets are rich in VOs.[67] These results further support the potential of sunlight-driven combustion method to create oxygen vacancies on the surface of the synthesized oxides. High-resolution XPS 4f also shows peaks at 158.5 and 163.7 eV, respectively, related to Bi 4f7/2 and Bi 4f5/2, in agreement with the reported results (Figure d).[68]

Figure 3

Crystal structure, chemical structure, and morphology of few-layer Bi2O3. (a) XRD, (b) XPS survey spectrum, (c) O 1s XPS spectrum, (d) Bi 4f XPS spectrum, (e, f) SEM images, and (g, h) TEM images of Bi2O3 synthesized by sunlight-driven combustion method.

The SEM images in Figures e and S7 show the porous structure of Bi2O3, which is induced by the gas evolution during the combustion. In fact, the porous structure, gained from the gases releasing during the combustion process, is the signature of the combustion synthesis. The higher-resolution SEM image (Figure f) shows that the morphology of the as-prepared SDC Bi2O3 is nanosheets. It is also clear that the nanosheets are well separated. The nonagglomeration of the obtained Bi2O3 nanosheets may be induced by the gases evolved during the combustion. The TEM images show that the morphology of SDC Bi2O3 is a few-layered nanosheets (Figure g,h). This is yet another advantage achieved during the sunlight-assisted combustion synthesis.

Sunlight-Driven Synthesis of Bi2O3/ZnO Heterostructure

The proposed sunlight-driven combustion process was also used to prepare ZnO/Bi2O3 heterostructure. Later, the heterostructure was characterized by XPS to confirm the interaction between both metal oxides (Bi2O3 and ZnO) and by SEM to determine the morphology. Figure a shows the XPS survey spectrum for the Bi2O3/ZnO heterostructure, in which the Bi, Zn, and O peaks are present. The absence of any other elements except a small amount of C from the sample holder confirms again the phase purity. Figure b–d shows the high-resolution XPS spectra of Zn 2p, Bi 4f, and O 1s. The peaks at 1022.2 and 1045.2 eV are in agreement with those of Zn 2p3/2 and Zn 2p1/2, respectively (Figure b). The difference between Zn 2p3/2 and Zn 2p1/2 as 23 eV confirms the normal state of Zn2+ in ZnO.

Figure 4

Chemical structure and morphology. (a) XPS survey spectrum, (b) Zn 2p XPS image, (c) Bi 4f XPS spectrum, (d) O 1s XPS spectrum, and (e, f) SEM images of Bi2O3/ZnO heterostructure synthesized by sunlight-driven combustion method.

The two signals located at 158.64 and 164 eV correspond to the binding energies of Bi 4f7/2 and Bi 4f5/2. Compared to the pristine Bi2O3 (Section ), Bi 4f7/2 and Bi 4f5/2 of Bi2O3/ZnO heterostructure exhibit a shift to higher binding energy. Such a shift indicates that the photogenerated electrons transfer from Bi2O3 to ZnO which is in disagreement with the photogenerated electron transfer in the reported Bi2O3/ZnO heterojunctions.[68,69] Thus, the sunlight-driven combustion-synthesized Bi2O3/ZnO heterostructure is not a heterojunction system, but it is a direct Z-scheme system. In fact, Z-scheme charge transfer can improve the charge separation and redox ability for the developed material and thereby enhance the photocatalytic activity.[70] O 1s was fitted to two peaks at 529.04 and 530.46 eV related to lattice oxygen and oxygen atoms in the vicinity of VOs, respectively[68] (Figure d). Thus, XPS analysis confirms the successful formation of Bi2O3/ZnO heterostructure using sunlight-driven solution combustion process. The ZnO nanoparticles-decorated 2D nanosheet morphology for the Bi2O3/ZnO heterostructure is represented in Figure e,f.

Photocatalytic Activity of Bi2O3 Nanosheets and Bi2O3/ZnO Heterostructure

The photocatalytic activity of Bi2O3 nanosheets and Bi2O3/ZnO heterostructures prepared by sunlight-driven combustion (Bi2O3 SDC and Bi2O3/ZnO SDC) was compared to that of the counterparts synthesized by conventional solution combustion (Bi2O3 CSC and Bi2O3/ZnO CSC), and the results are shown in Figure a. It can be seen that about 64, 40, 96, and 82% of RhB were degraded by Bi2O3 SDC, Bi2O3 CSC, Bi2O3/ZnO SDC, and Bi2O3/ZnO CSC, respectively, after 80 min of sunlight irradiation. To compare the RhB photodegradation catalyzed by different photocatalysts, the degradation data were fitted according to the pseudo-first-order kinetic model as expressed by eq (Figure b)[71]where k, C0, and C are the kinetic rate constant, the RhB initial concentration, and the RhB concentration at time t, respectively. The k and correlation coefficient (R2) values are listed in Table . The R2 values for all of the samples are more than 98%, indicating that the data fitting using the pseudo-first-order kinetics model is reasonable. It can also be observed from Table that the samples prepared by sunlight-driven combustion method (Bi2O3 SDC and Bi2O3/ZnO SDC) exhibit a higher kinetic rate constant than those synthesized by the conventional solution combustion method (Bi2O3 CSC and Bi2O3/ZnO CSC). These findings further elucidate the importance of the developed synthesis method.

Figure 5

(a) Photocatalytic degradation curves and (b) degradation kinetic curves of RhB over Bi2O3/ZnO SDC, Bi2O3/ZnO CSC, Bi2O3 SDC, and Bi2O3 CSC under sunlight irradiation, and reusability of (c) Bi2O3 SDC and (d) Bi2O3/ZnO SDC (RhB concentration = 10 mg L–1, catalyst dosage = 0.5 g L–1, pH 7).

Table 1

Kinetics Reaction Constant (k) and Correlation Coefficient (R2) for RhB Degradation Catalyzed by Bi2O3 SDC, Bi2O3 CSC, Bi2O3/ZnO SDC, and Bi2O3/ZnO CSC

	R2	k
Bi2O3 SDC	0.99284	0.0123
Bi2O3 CSC	0.99202	0.00611
Bi2O3/ZnO SDC	0.98461	0.03754
Bi2O3/ZnO CSC	0.99486	0.02016

It is clear also that both Bi2O3 SDC and Bi2O3/ZnO SDC are photostable even after four cycles of photocatalytic degradation of RhB (Figure c,d); no remarkable decrease in the photocatalytic performance was observed after four cycles of RhB degradation under sunlight irradiation.

Z-Scheme Charge Transport Evidence

XPS analysis indicated that the charge migration follows the Z-scheme approach (Section ). Scavengers experiments were used here to further prove the Z-scheme charge transport in Bi2O3/ZnO heterostructure. p-Benzoquinone (BQ) and tert-butyl alcohol (TBA) were added into the photocatalyst/dye suspension to trap superoxide radicals (•O2–) and hydroxyl radicals (•OH), respectively, followed by the photocatalytic assessments. Figure a shows that both BQ and TBA significantly suppressed the RhB degradation by Bi2O3/ZnO under sunlight irradiation, indicating that both •O2– and •OH are produced during the photodegrading process.

Figure 6

(a) Effect of TBA and BQ on RhB degradation over ZnO/Bi2O3 SDC under sunlight irradiation and comparison of charge transport in Bi2O3/ZnO SDC heterostructure based on (b) heterojunction and (c) Z-scheme approaches.

The positions of CBM and VBM are, respectively, −0.36 and 2.67 V for ZnO, and 0.18 and 2.92 V for Bi2O3.[52] If charge transport follows heterojunction approach, the electrons will migrate to CB of Bi2O3 and holes to VB of ZnO (Figure b). The potential at CBM of Bi2O3 (0.18 V) is not enough to produce •O2– because O2/•O2 potential is −0.046 V. Thus, the heterojunction approach cannot explain the production of •O2– proved by scavengers experiments. Whereas, if charge transport follows the Z-scheme approach, the electrons will migrate to CB of ZnO and the holes will remain in the VB of Bi2O3 (Figure c). The CBM of ZnO has enough potential (−0.36 V) to produce •O2–, which is proved by scavengers experiments. Thus, the Bi2O3/ZnO heterostructure is a Z-scheme system. These results well agree with the XPS analysis.

Conclusions

In conclusion, we have developed a novel, facile, and sustainable sunlight-driven combustion method to synthesize porous metal oxide nanostructures. Surprisingly, in this method, the direct exposure to sunlight creates VOs on the surface of oxides and induces homogeneous heating, which facilitates the uniformity in particle size and pore size distribution. UV–vis DRS, Tauc plot, VB-XPS, and Mott–Schottky analyses indicated that ZnO SDC has a narrower band gap (3.07 eV) than ZnO CSC (3.19 eV). The ESR analysis confirmed that ZnO SDC exhibits higher exciton dissociation and thereby higher concentration of free electrons and holes than ZnO CSC. The band gap narrowing and the higher exciton dissociation for ZnO SDC were due to the higher concentration of VOs induced by sunlight irradiation. It was also demonstrated that few-layer Bi2O3 nanosheets and ZnO/Bi2O3 heterostructures can be synthesized using the introduced method. ZnO, Bi2O3, and ZnO/Bi2O3 synthesized by the sunlight-driven combustion method exhibit higher photocatalytic activity than their conventional counterparts. The cost-effectiveness of the sunlight-driven combustion method offers an alternative route for large-scale synthesis of porous metal oxide nanomaterials.