Highly-active direct Z-scheme Si/TiO₂ photocatalyst for boosted CO₂ reduction into value-added methanol

Abstract

In the present study, direct Z-scheme Si/TiO₂ photocatalyst was synthesized via a facile hydrothermal reaction using tetrabutyl titanate and Si powder prepared from magnesiothermic reduction of SiO₂ nanospheres. The Si/TiO₂ nanospheres were composed of porous Si nanospheres with a diameter of ∼300 nm and TiO₂ nanosheets with a diameter of 50 nm and thickness of 10 nm, and demonstrated superior visible light harvesting ability to either Si nanospheres or TiO₂ nanosheets. CO₂ photocatalytic reduction proved that Si/TiO₂ nanocomposites exhibit high activity in conversion of CO₂ to methanol with the maximum photonic efficiency of 18.1%, while pure Si and TiO₂ catalyst are almost inactive, which can be ascribed to the integrated suitable band composition in the Si/TiO₂ Z-scheme system for CO₂ reduction. The enhanced photocatalytic property of Z-scheme Si/TiO₂ nanospheres was ascribed to the formation of Si/TiO₂ Z-scheme system, which improved the separation efficiency of the photogenerated carriers, prolonged their longevity, and therefore boosted their photocatalytic activity.

---

1. Introduction

High consumption of fossil fuels is creating not only an energy crisis but also environmental pollution and climate change due to excessive green house gas emissions to the atmosphere. As carbon dioxide (CO₂) is a major contributor to green house gases, attention is focused on its mitigation all over the world. Solar-driven photocatalytic conversion of CO₂ into added value fuels is the most promising proposition as it does not only remove CO₂ from effluent gases but also produces hydrocarbon fuels, which could be used to meet future energy needs. Thus, a lot of research efforts have focused on developing efficient catalysts for CO₂ photocatalytic reduction. Semiconductor photocatalysts (such as ZnO,^1,2 CdS,³ ZnGa₂O₄,⁴ Zn₂GeO₄,^5,6 WO₃,⁷ TiO₂ (ref. 8) etc.) have been explored. Among them, titanium dioxide (TiO₂) has been extensively studied as an important photocatalyst because of its low toxicity, low cost, superior photocatalytic activity and long-term chemical stability.^9–11 However, the wide band gap of TiO₂ (∼3.2 eV) limits its efficient utilization for solar energy conversation as TiO₂ only absorbs light with wavelengths shorter than ∼387 nm in the ultraviolet region. Moreover, after photoabsorption, the electrons are excited from the valence band of TiO₂ to the conduction band and the effective electron–hole pairs are generated.¹² Unfortunately, most of the effective electron–hole pairs are recombined and dissipated as heat before they arrive at the photocatalyst surface, which makes TiO₂ an inefficient photogenerated carrier hampering its charge separation ability.

Research scientists have devoted extensive efforts to address these problems. Introducing doping elements (such as S,¹³ N,¹⁴ and C (ref. 15)) into TiO₂ has been proven to be an effective approach to narrow the band gap, improve the visible light absorption and enhance the photocatalytic activity in CO₂ reduction. TiO₂ modification with metal particles (e.g., Ag,¹⁶ Au,¹⁷ Pt,¹⁸ and Cu (ref. 19)) has been reported to inhibit charge recombination probability because these metals serve as electron traps to suppress the recombination of the photogenerated electron–hole pairs and hence improve the photocatalytic activity. In addition, coupling TiO₂ with a narrower band gap semiconductor to construct heterojunctions is another effective approach to accommodate the visible-light photon energy and improve the photogenerated charge separation and CO₂ conversion efficiency. This coupling takes advantage of both the heterojunction to improve charge separation rate, and the narrow band gap of the coupled semiconductor to expand the light absorption region. TiO₂ based heterojunctions such as PbS/TiO₂,²⁰ CuO/TiO₂,²¹ FeTiO₃/TiO₂ (ref. 22) have been reported in recent years.

Similar to the heterojunction photocatalytic system, the Z-scheme photocatalytic system also features the spatial isolation of photogenerated electrons and holes, which reduces the bulk electron–hole recombination.²³ However, a Z-scheme photocatalytic system is generally constructed by employing a conductor as the electron mediator to form the known Ohmic contact with low contact resistance. Until now, many Z-scheme systems have been reported, such as TiO₂–Au–CdS system,²⁴ AgBr–Ag–AgI,²⁵ and ZnO–Au–CdS,²⁶ etc. Recently, studies on direct Z-scheme photocatalysts (such as g–C₃N₄–TiO₂ and WO₃–NaNbO₃ system) have been investigated for photocatalytic water splitting, CO₂ conversion and photocatalytic decomposition.^27–29

A narrow band gap semiconductor, silicon (1.12 eV), has drawn considerable interest because of its potential application in optoelectronic devices and integrated microelectronics. Recently, studies of silicon materials have reported its promising photocatalytic activity. Shao et al. prepared hydrogen-terminated Si nanowires (Si NWs) and noble metal-modified (Pt, Pd, Au, Rh, Ag) Si NWs substrates by oxide-assisted growth method, and investigated their performance for the degradation of Rhodamine B and oxidation of benzyl alcohol to benzoic acid under visible light irradiation.³⁰ Independently, Megouda et al. investigated the performance of hydrogen-terminated Si NWs and two kinds of metals (Ag, Cu) decorating Si NWs for the degradation of dye molecules.³¹ Furthermore, literature about Si/TiO₂ heterojunctions that achieve enhanced photochemical and photocatalytic properties have been reported.³² Wang et al. successfully deposited TiO₂ onto Si nanowire arrays to construct Si/TiO₂ heterojunctions using a surface reaction-limited pulsed chemical vapor deposition method, and tailored the electrical properties of TiO₂ for wider spectrum solar energy harvesting and conversion.³³ Li et al. attained a novel composite material of TiO₂ and porous silicon using a sol–gel method and found that it exhibits much higher photocatalytic activity for the degradation of RhB.³⁴ Recently, direct Z-scheme Si/TiO₂ tree-like heterostructures were constructed by Yang,³⁵ which is demonstrated to greatly improve the photocatalytic activity of H₂ evolution. Research on the direct Z-scheme system is just a recent work, and still needs further study.

In this work, we report a novel direct Z-scheme Si/TiO₂ photocatalyst synthesized via a facile hydrothermal method with tetrabutyl titanate and Si powder prepared from the magnesiothermic reduction of SiO₂ Stöber nanospheres.³⁶ The enhanced photocatalytic conversion of CO₂ reduction into value-added methanol is investigated.

2. Experimental section

2.1 Synthesis of SiO₂ nanospheres

The monodisperse silica spheres were prepared by hydrolysis and condensation of tetraethoxysilane (TEOS) in a mixture of water, ammonia, and ethanol. In a typical synthesis process, 9 mL 28 wt% ammonia water was mixed with 16.25 mL ethanol and 24.75 mL deionized water under a stirring condition (solution A). 4.5 mL TEOS was added to 45.5 mL ethanol under stirring (solution B). Here, we added B to A drop by drop and strewed for another 2 h at room temperature. SiO₂ nanospheres were centrifuged from the mixture, alternately washed with deionized water and ethanol 3 times and then dried at 100 °C for 12 h.

2.2 Synthesis of Si nanospheres

0.5 g SiO₂ nanospheres and 0.42 g Mg powder were ground for 5 minutes and then transferred into a crucible to calcine at 750 °C for 5 h under N₂ atmosphere. After cooling to room temperature, the powder was then added to 50 mL 32 wt% HCl solution with stirring for 24 h. Si nanospheres were then cleaned several times by centrifugation and water dispersion and finally dried into powders at 60 °C in a vacuum oven for 12 h.

2.3 Synthesis of direct Z-scheme Si/TiO₂ nanospheres

Direct Z-scheme Si/TiO₂ nanospheres were fabricated via a facile hydrothermal reaction with tetrabutyl titanate and Si nanospheres. In a typical process, 25 mL of Ti(OBu)₄ and 1.5 g Si nanospheres were added in a 100 mL Teflon pot and 3 mL of hydrofluoric acid was added dropwise with stirring. After stirring for 15 min at room temperature, the Teflon pot was sealed and kept at 200 °C for 24 h. Finally, the as-prepared Si/TiO₂ nanospheres were obtained after the resulting precipitate was centrifuged three times, washed with ethanol to remove the hydrofluoric acid and organics, and then dried in a vacuum oven for 12 h. Pure TiO₂ sample was prepared using the same hydrothermal reaction without Si nanospheres.

2.4 Characterizations

The crystal structure of all the samples was examined by means of X-ray diffraction analysis (XRD, Bruker D8 ADVANCE with Cu-Kα radiation, λ = 1.5418 Å). The morphology and particle size were determined by field emission scanning electron microscopy (FE-SEM, Hitachi S4800) and transmission electron microscopy (TEM, JEOL JSM-2010) with an accelerating voltage of 200 kV. UV-Vis absorption spectra were obtained using a UV-Vis spectrometer (Shimadzu UV-3600). Photoluminescence (PL) spectra were obtained with an Edinburgh Instrument FLS 920 spectrometer. The excitation wavelength, λ_ex, was 360 nm, and both the bandwidths of excitation and emission were 5 nm. The Brunauer–Emmet–Teller (BET) specific surface area of the samples were determined by a high speed automated area and pore size analyzer (ASAP 2010).

2.5 Photocatalytic reduction of CO₂ under 355 laser irradiation

The photocatalytic reaction cell and its setup have been described in our earlier publication.³⁷ The photocatalytic reactor is a cylindrical stainless steel cell with quartz windows on the top to enable the transmission of 355 nm pulse laser radiations. At the bottom of the cell there is a gas inlet with a needle valve which lets the CO₂ gas pass through the distilled water in the cell and there is an outlet fixed with the rubber septum in order to dispense the sample through the syringe almost at the same level at the opposite side. The whole cell is kept on a magnetic stirrer that constantly replenishes the photocatalyst in the path of laser radiations. Care has to be taken not to let the water level go much higher than the level of the catalyst platform in order to have better interaction of radiation with the photocatalyst. Since the quantity of sample taken for gas chromatographic analysis at each time was around 4.0 μL, the water level did not decrease due to sample withdrawing from the reaction cell. The reaction cell was cleaned, dried, then was tightly closed and checked for leaks up to 50 psi pressure after 1.0 g catalyst was loaded along with 100 mL distilled water. High purity CO₂ gas (99.99%) was introduced through reactor inlets and the reactor pressure was maintained at 50 psi. Prior to turning on the pulsed laser, CO₂ gas was purged into 100 mL water containing 1.0 g of catalyst for 30 min in order to saturate the contents of the reactor with CO₂. After a predetermined irradiation time, water samples were withdrawn from the reactor using a syringe without opening the reactor and subjected to GC analysis.

The laser (wavelength = 355 nm) used for this study was the third harmonic of the pulsed Nd:YAG laser (Model Spectra Physics GCR 250-10) operated at 10 Hz and the pulse width of ∼8 ns. Throughout this study, a laser pulse energy of 40 mJ was used. The laser beam was routed with high power UV reflecting mirrors/dichroic mirrors so that the beam enters from the top of the cell and an appropriate lens was also used to slightly expand the beam to the same diameter of the catalyst platform. Although the laser pulse energy was quite stable it was monitored throughout the experiment with the 50–50 beam splitter and the laser energy meter supplied by Coherent USA.

The water samples were analyzed for methanol and other hydrocarbons using a gas chromatograph equipped with flame ionization detector (FID). The separation was carried out on Rtx-Wax column (dimensions: 30 m × 0.32 mm × 0.32 mm) obtained from Restek, using temperature programmed conditions. For the analysis of end products, 4.0 μL of the laser irradiated sample was injected into the gas chromatograph and the operating conditions were as follows: oven temperature was set at 40 °C and was then increased to 90 °C at 5 °C min⁻¹ heating rate and increased to 180 °C at the rate of 50 °C min⁻¹ to elute all the components from the column before injecting another sample. The injector and detectors were both set at 200 °C and helium was used as the carrier gas. The total analysis run time was 11.8 min. A calibration plot was established for methanol standard solution in distilled water for calculating the amount of methanol produced as a function of irradiation time.

2.6 Photocatalytic reduction of CO₂ under Xe arc lamp irradiation

In a typical process, 0.1 g of the sample was uniformly dispersed on the glass reactor (4.2 cm²). A 300 W Xenon arc lamp was used as the light source. The reaction system (230 mL in volume) was vacuum-treated several times, and then the high purity of CO₂ gas was followed into the reaction setup to reach ambient pressure. 0.4 mL of deionized water was injected into the reaction system as reducer. The as-prepared photocatalysts were allowed to equilibrate in the CO₂/H₂O atmosphere for several hours to ensure that the adsorption of gas molecules was complete. During the irradiation, about 1 mL of gas was continually taken from the reaction cell at given time intervals for subsequent CH₄ concentration analysis by using a gas chromatograph (GC-2014, Shimadzu Corp., Japan). All samples were treated at 300 °C in nitrogen atmosphere for 2 h for removal of organic adsorbates before the photocatalysis reaction.

2.7 Photocatalytic degradation of aqueous RhB solution

The photocatalytic activity was measured as follows: 0.100 g of as-prepared TiO₂ and Si/TiO₂ samples were added to a 250 mL Pyrex glass vessel which contained 200 mL RhB solution (7.5 mg L⁻¹). The light source was a 300 W Xe arc lamp (CHF-XM500W, Beijing TrustTech Co. Ltd.) with an illumination intensity of 400 mW cm⁻². Prior to irradiation, RhB solution suspended with photocatalysts was stirred in the dark for 30 min to ensure that the surface of photocatalysts reaches the adsorption–desorption equilibrium. 3 mL of the suspension was withdrawn throughout the experiment after every 10 min. The samples were analyzed by a UV-Vis spectrophotometer after removing the catalyst powders by centrifugation.

3. Results and discussion

3.1 Phase and morphology analysis

X-ray diffraction patterns of as-prepared Si nanospheres, TiO₂ nanosheets and direct Z-scheme Si/TiO₂ nanospheres are depicted in Fig. 1. The diffraction peaks in the XRD curve marked in red at 28.4°, 47.3°, 56.1°, 69.1°, 76.3° and 88.0° can be assigned to (111), (220), (311), (400), (331) and (422) planes of Si (JCPDS Card 27-1402), respectively. The narrow broadness of diffraction peaks for Si nanospheres indicates that the Si nanospheres prepared via a magnesiothermic reduction method have a high purity and crystallinity. The diffraction peaks in the XRD curve marked in black at 25.3°, 37.8°, 48.0°, 55.0°, 62.6°, 70.3°, 75.0° and 82.1° are indexed to the (101), (004), (200), (211), (204), (220), (215) and (303) planes of TiO₂ (JCPDS Card 21-1272), respectively. From the diffraction peaks in the XRD curve of direct Z-scheme Si/TiO₂ nanospheres marked in blue, which contain both the Si and TiO₂ diffraction peaks, it can be seen obviously that the Si/TiO₂ product obtained via a hydrothermal method is composed of Si and TiO₂. The diffraction peaks of Si in the XRD curve of direct Z-scheme Si/TiO₂ nanospheres are very low, which may result from its low content.

Fig. 1 XRD patterns of the as-prepared Si nanospheres, TiO₂ nanosheets and direct Z-scheme Si/TiO₂ nanospheres.

As depicted in Fig. 2(a), the SiO₂ nanospheres prepared via the Stöber method are monodisperse and uniform with a smooth surface and an average diameter of about 300 nm. After the magnesiothermic reduction process, Si nanospheres were obtained with a porous structure and rough surface, resulting from the loss of O atoms from SiO₂ nanospheres captured by Mg under high temperature. The diameter of Si nanospheres remained mainly unchanged (as shown in Fig. 2(b)). The SEM images of the contrast TiO₂ samples prepared by a hydrothermal method without adding Si nanospheres are depicted in Fig. 2(c). It can be seen that the TiO₂ nanoparticles displayed a uniform sheet shape with an average edge length of about 100 nm and a thickness of about 10 nm. Fig. 2(d) shows the representative SEM image of direct Z-scheme Si/TiO₂ nanospheres, clearly indicating the Si nanospheres were coated with TiO₂ nanosheets.

Fig. 2 SEM images of as-prepared SiO₂ nanospheres (a), Si nanospheres (b), TiO₂ nanosheets (c), and direct Z-scheme Si/TiO₂ nanospheres (d).

In order to obtain further information on the structure of the samples, TEM observation of the TiO₂ nanosheets and direct Z-scheme Si/TiO₂ nanospheres was carried out. It can be clearly noticed from Fig. 3(a) that the as-prepared TiO₂ samples are composed of a large quantity of square nanosheets with an average edge length of about 100 nm and a thickness of about 10 nm, which is in good agreement with the result obtained from the SEM images. The TEM image of Si/TiO₂ direct Z-scheme nanocomposites in Fig. 3(b) clearly shows that the Si nanospheres were coated by TiO₂ nanosheets. Moreover, it can be observed that the Si nanospheres appear with a porous morphology resulting from the O element captured by Mg in the magnesiothermic reduction process. The magnified TEM image of the nanostructure of the Si/TiO₂ nanocomposites in Fig. 3(c) presents the TiO₂ nanosheets aggregation morphology on the Si porous nanospheres, revealing the formation of the Si/TiO₂ Z-scheme system. The high-magnification TEM in Fig. 3(d) depicts many different lattice fringes of the Si/TiO₂ nanocomposites. The fringes with lattice spacing of ca. 0.235 nm and 0.31 nm observed in the HRTEM image match those of the (001) and (111) crystallographic planes of anatase TiO₂ and Si nanoparticles, indicating the formation of Si/TiO₂ interfaces via a hydrothermal method, which may improve the photocatalytic properties of TiO₂.

Fig. 3 TEM images: (a) TiO₂ nanosheets, (b, c) direct Z-scheme Si/TiO₂ nanospheres, (d) high-magnification TEM image of direct Z-scheme Si/TiO₂ nanospheres.

3.2 Optical absorption properties

Fig. 4(a) shows the UV-Vis absorption spectra of the Si nanospheres, TiO₂ nanosheets and direct Z-scheme Si/TiO₂ nanospheres. It can be noticed from the spectra that the as-prepared TiO₂ and Si/TiO₂ nanocomposites exhibit similar absorption behaviour in the ultraviolet region. However, Si/TiO₂ Z-scheme nanospheres show an enhanced absorbance throughout the visible light region due to the existence of Si which is a visible light responsive material with a band gap of 1.12 eV. The improved visible light absorption explains the enhanced photocatalytic properties of the direct Z-scheme Si/TiO₂ nanospheres, as described later.

Fig. 4 UV-Vis absorption spectra of Si nanospheres, TiO₂ nanosheets and direct Z-scheme Si/TiO₂ nanospheres.

3.3 Photocatalytic reduction of CO₂

To evaluate the photocatalytic activity of Si nanospheres, TiO₂ nanosheets and direct Z-scheme Si/TiO₂ nanospheres, the conversion of CO₂ into hydrocarbon fuels in distilled water was investigated using a high power pulsed laser as light source at 355 nm wavelength. As we know, a number of reaction products (such as HCHO, CH₃OH, HCOOH, CO, CH₄, etc.) can be obtained during the CO₂ photoreduction process. The following reactions may be the pathways of CO₂ photoreduction into value added hydrocarbons.

Catalyst + hν → e⁻_cb + h⁺_vb

2H₂O + 4h⁺ → 4H⁺ + O₂ +0.82 V

CO₂ + 2H⁺ + 2e⁻ → HCOOH −0.61 V

CO₂ + 2H⁺ + 2e⁻ → CO + H₂O −0.53 V

CO₂ + 4H⁺ + 4e⁻ → HCHO + H₂O −0.48 V

CO₂ + 6H⁺ + 6e⁻ → CH₃OH + H₂O −0.38 V

CO₂ + 8H⁺ + 8e⁻ → CH₄ + H₂O −0.24 V

In this study, we were selective to obtain CH₃OH as the main product of CO₂ photocatalytic reduction over Si/TiO₂ nanocomposites. Our comparative tests demonstrated that nearly no product was found by using Si nanospheres or TiO₂ nanosheets as photocatalysts. This may be due to the low conductive band potential of TiO₂ and low valance band potential of Si, respectively. The conduction band (CB) and valence band (VB) edge of Si and TiO₂ semiconductors were calculated by the equation as follows³⁸ and are presented in Table 1.

E_CB = X − E_C − 1/2E_g

E_VB = E_g + E_CB

where X is the absolute electronegativity of the semiconductor; E_C is the energy of free electrons on the hydrogen scale (4.5 eV); and E_g is the band gap of the semiconductor.

Table 1 Relevant parameters of Ti, O, Si atoms (ionization energy, atomic electron affinity and absolute electronegativity) and TiO₂, Si semiconductors (absolute electronegativity, band gap and electrochemical potentials of CB/VB band edges)^a

Element	Si	Ti	O
a The relevant data were selected from handbook.^39
Atomic ionization energy (eV)	8.15168	6.82812	13.6182
Atomic electron affinity (eV)	1.38952	0.079	1.46111
Absolute electronegativity (eV)	4.7704	3.45356	7.53958

---

Catalyst	Si	TiO2
Band gap (eV)	1.1	3.2
Absolute electronegativity (eV)	4.7706	5.81193
CB band edge electrochemical potential (V vs. NHE)	−0.38	−0.29
VB band edge electrochemical potential (V vs. NHE)	0.721	2.912

---

It can be inferred from the table that, although the calculated VB band edge electrochemical potential of TiO₂ (2.91 V vs. NHE) is high enough to initiate the reaction of H₂O and h⁺ to form O₂ and H⁺ (0.82 V vs. NHE), the calculated CB band edge electrochemical potential (−0.29 V vs. NHE) is lower than the reaction potential needed for CO₂ transformation to CH₃OH with H⁺ and photogenerated e⁻ (−0.38 V vs. NHE), resulting in the thermodynamic impossibility of CO₂ photoreduction into CH₃OH over TiO₂ catalyst under 355 nm laser irradiation. In the same way, pure Si catalyst has an appropriate CB band edge electrochemical potential to reduce CO₂ into CH₃OH with H⁺ and e⁻, but its low VB band edge electrochemical potential (0.72 V vs. NHE) cannot transform any H⁺ from the reaction of H₂O oxidation, which enables the photoreduction of CO₂ into CH₃OH.

Gas chromatography (GC) was employed to verify and quantify the methanol products from the CO₂ reduction over direct Z-scheme nanospheres. It is depicted in Fig. 5(a) that the retention time for methanol standard is 2.46 min for the selected GC parameters and the used column. The relationship between GC peak area and methanol concentration was confirmed by using known methanol concentrations in a standard sample for calibration as depicted in Fig. 5(b), which exhibits a linear trend.

Fig. 5 (a) GC peak position of methanol standard, (b) calibration curve for methanol concentration vs. GC peak area.

Fig. 6(a) depicts the GC peaks of products for samples from CO₂ photoreduction, which are taken in 30 min intervals of irradiation with 355 nm laser by using Si/TiO₂ as the photocatalyst. It can be seen clearly that all the GC peaks in Fig. 6(a) appear at exactly 2.46 min of the retention time and no other GC peaks were detected, which suggests that the methanol is the only product obtained through the laser induced photocatalytic reduction of CO₂, possibly because of the sharp line width of the laser beam centred around 355 nm (highly monochromatic) of the pulsed laser radiation. Furthermore, it was shown that the GC peak areas of methanol from CO₂ photoreduction increase with the irradiation time (30 min, 60 min, 90 min, 120 min, 150 min) and reach the maximum in 150 min of irradiation after which they start to fall. Fig. 6(b) shows the concentration variation of the photocatalytic process of converting CO₂ into methanol with laser irradiation. It can be observed that the produced methanol concentration increases with laser irradiation time and reaches its maximum (197 μM/100 mL) at 150 min, but afterwards it declines. The decrease of the methanol concentration may be caused by the existence of photocatalytic oxidation effect of Si/TiO₂ composed semiconductor with positive VB position. When the methanol was produced in a substantial amount, it will be adsorbed on the surface of the photocatalysts and oxidized to inorganic matter, which is in agreement with the results obtained by, and explanation of, other groups.^37,40,41

Fig. 6 (a) GC peaks of methanol for sample taken every 30 min of irradiation with a laser pulse energy of 40 mJ per pulse at 355 nm radiation with 600 mg catalyst in 100 mL distilled water, 50 psi CO₂ pressure. (b) Concentration of produced CH₃OH and conversion efficiency with time.

In order to estimate the efficiency of CO₂ conversion into methanol using direct Z-scheme Si/TiO₂ nanospheres as the photocatalyst with 355 nm laser irradiation, the process of the CO₂ conversion efficiency was estimated. The amount of CO₂ dissolved in distilled water under our experimental conditions can be calculated from Henry’s law and the amount of CO₂ dissolved in 1 L water at atmospheric pressure is 34 mmol (Henry constant). As the pressure in the photocatalytic measurement is 50 psi (3.4 atm), the total CO₂ dissolved in 100 mL of water is 11.56 mmol. Once we know the methanol concentration at different irradiation times of the photocatalytic process, the CO₂ conversion efficiency can be calculated from the quotient of actual concentrations of methanol and CO₂ concentration (as shown in Fig. 6(b)). For that the maximum concentration of methanol is 197 μM/100 mL after 150 minutes of laser irradiation, the maximum CO₂ conversion efficiency is calculated (197/11 560) to be about 1.71%. However, according to Schüler’s experimental result,⁴² the CO₂ solubility in binary mixtures of water and methanol increases with increasing methanol content, resulting in the actual CO₂ conversion efficiency being slightly lower than the calculated efficiency (1.71%).

Moreover, we can calculate the photonic efficiency of the photocatalytic reduction of CO₂ from the number of methanol molecules produced for certain irradiation time and the number of consumed photons in the reaction. The number of methanol molecules can be estimated from the molar concentrations and Avogadro’s number. In the case of laser, the number of photons at 355 nm wavelength with the laser pulse energy of 40 mJ per pulse and repetition rate of 10 Hz can be calculated to be 4.286 × 10¹⁹ photons per min. The maximum rate of methanol is 1.294 × 10¹⁸ molecules per min at the irradiation time interval from 60 min to 90 min. As a single methanol molecule needs 6 photogenerated electrons, the maximum photonic efficiency (P.E.) of photoreduction of CO₂ can be calculated (6 × 1.294 × 10¹⁸/4.286 × 10¹⁹) to be about 18.1%. The achievement of high photonic efficiency of direct Z-scheme Si/TiO₂ nanospheres may be due to the construction nature of Si/TiO₂ direct Z-scheme system.

3.4 Mechanism analysis of the enhanced photocatalytic activity of Si/TiO₂

As shown in Fig. 7, photoluminescence spectroscopy was employed for further investigation of the photocatalytic activities of TiO₂ nanosheets and direct Z-scheme Si/TiO₂ nanospheres. Two major components of the spectrum of direct Z-scheme Si/TiO₂ nanospheres consisted of a strong peak at 540 nm and a weak, broad peak from 400 to 520 nm, which are attributed to Si and TiO₂, respectively. Moreover, it can be seen clearly that the peak intensities in photoluminescence intensity of Si/TiO₂ Z-scheme nanospheres are much lower in contrast to that of Si nanospheres and TiO₂ nanosheets. As the PL emission resulted from the recombination of photo-induced charge carriers and information regarding the efficiency of charge carrier trapping, and their recombination kinetics can be drawn from the PL spectra,⁴³ it can be inferred that direct Z-scheme Si/TiO₂ nanospheres have a higher efficient separation rate of photogenerated charge carriers than that of TiO₂ nanosheets, which can be attributed to the formation of the Si/TiO₂ direct Z-scheme system.

Fig. 7 Photoluminescence spectra of TiO₂ nanosheets and direct Z-scheme Si/TiO₂ nanospheres.

In order to confirm the effect of the Si/TiO₂ direct Z-scheme system on improving the charge separation efficiency, RhB was selected to be the target pollutant for degradation by Si nanospheres, TiO₂ nanosheets and direct Z-scheme Si/TiO₂ nanospheres under 300 W Xe arc lamp irradiation with an illumination intensity of 400 mW cm⁻². As shown in Fig. 8(a), it can be clearly seen that RhB molecules were completely decomposed by Si/TiO₂ direct Z-scheme nanospheres after 1 h Xe arc lamp irradiation, while only 88.5% were decomposed by TiO₂ nanosheets and 6.6% by Si nanospheres, indicating the enhanced photocatalytic activity of the Si/TiO₂ Z-scheme compared with the TiO₂ nanosheets and Si nanospheres. Fig. 8(b) depicts the kinetic study of photocatalytic degradation of RhB solution over the three photocatalytic materials. The linear relationship of ln(C₀/C) vs. irradiation time suggests that degradation of RhB is a first order reaction. The calculated rate constants for Si nanospheres, TiO₂ nanosheets and Si/TiO₂ Z-scheme nanospheres are 0.000772, 0.0354 and 0.074 min⁻¹, respectively, from which it can be seen that the Si/TiO₂ Z-scheme has the best photocatalytic activity, which is 2.09 times that of the TiO₂ nanosheets and 95.8 times the Si nanospheres. Therefore, construction of the Si/TiO₂ Z-scheme is beneficial to improve the photocatalytic activity of Si and TiO₂ extensively.

Fig. 8 Photocatalytic degradation rates (a) and the ln(C₀/C) vs. irradiation time curves of RhB (b).

Furthermore, the CO₂ photocatalytic reduction under Xe arc lamp irradiation was carried out to investigate the influence of light source and experimental conditions on the product. Fig. 9(a) shows that CO₂ can be photoreduced to CH₄ by using all the prepared samples as photocatalysts. It is obvious that the direct Z-scheme Si/TiO₂ nanocomposites exhibit much higher activity than pure TiO₂ nanosheets and Si nanospheres. The higher conversion is attributed to the improved photogenerated carriers separation efficiency of the Si/TiO₂ Z-scheme system. To further investigate whether CH₄ is a product of CO₂ photocatalytic reduction, we carried out a series comparative experiments to investigate the source of C and H in the produced CH₄ under Xe arc lamp irradiation. As shown in Fig. 9(b), the CH₄ detected in normal conditions is much higher than that in other experiments. It should be noted that the extremely low CH₄ concentrations in non-normal conditions are from the naturally occurring CH₄ in air (1–2 ppm). In other words, CH₄ can be produced only in the case of possessing all the conditions including catalyst, CO₂, H₂O and light. It can be inferred from the contrast experiment that C in the photoreduction product CH₄ is from CO₂, while H is from H₂O. Therefore, it can be confirmed that the detected CH₄ is from the photoreduction of CO₂, but not a product of organic oxidation at the Si/TiO₂ Z-scheme or release of surface bound organics.

Fig. 9 (a) CH₄ evolution with time under Xe arc lamp irradiation, (b) comparative experiments of the CO₂ photoreduction under different conditions.

Because the photocatalytic process is surface orientated, the specific surface areas of the prepared TiO₂ nanosheets and Si/TiO₂ nanocomposites were also measured by BET to study the actual exposed surface area in the photocatalytic reaction. As shown in Fig. 10, the Si/TiO₂ sample possesses a slightly higher specific surface area (165.4 m² g⁻¹) than TiO₂ nanosheets (154.3 m² g⁻¹) due to the surface structuring effect by the formation of spherical heterostructures with TiO₂ nanosheets well dispersed. In the photocatalytic reduction of CO₂ experiments, our results show that only the Si/TiO₂ sample is capable of producing methanol under laser irradiation due to thermodynamics, indicating that the exposed surface area of the catalyst has no effect on its photocatalytic reduction activity. In the photocatalytic degradation experiment, 0.1 g of the as-prepared Si, TiO₂ and Si/TiO₂ catalysts was adopted, respectively. The slight difference of the catalysts in specific surface area is as small as 6.6%, which is suggested to not be the main reason for the great improvement of photodegradation efficiency (more than twice). Hence, it can be concluded that the excellent performance of the Si/TiO₂ catalyst does not result from a surface structuring effect, but from the improved photogenerated carriers’ concentration and separation efficiency from the Si/TiO₂ direct Z-scheme system construction.

Fig. 10 N₂ adsorption–desorption isotherms of TiO₂ nanosheets and direct Z-scheme Si/TiO₂ nanospheres.

Fig. 11 illustrates the schematic charge flow in the Si/TiO₂ direct Z-scheme system under illumination. 355 nm lasers can be harvested by both Si and TiO₂ to generate e⁻–h⁺ pairs. As reported in Yang’s previous research,³² the photogenerated hole in TiO₂ (TiO₂h⁺) moves toward the TiO₂/electrolyte interface and oxidizes OH⁻ to oxygen, while photogenerated electrons in the TiO₂ (TiO₂e⁻) move away from the interface between TiO₂ and the electrolyte due to the schottky barrier. The potential barrier at the Si/TiO₂ interface reflects holes back into the TiO₂ layers. To complete the circuit, the photogenerated electrons in Si (_Sie⁻) move to the surface where the CO₂ reduction reaction takes place. The photogenerated hole in Si (_Sih⁺) moves towards the Si/TiO₂ interface and recombines with the TiO₂e⁻. Therefore, the direct Z-scheme Si/TiO₂ nanospheres show high activity towards CO₂ reduction into methanol since its band alignment at the junction helps reduce recombination under illumination. However, for individual Si photocatalysts, its VB potential (+0.721 V vs. NHE) is not high enough to achieve the oxidation reaction potential (O₂/H₂O 0.82 V vs. NHE). Similarly, the CB potential of TiO₂ (−0.29 V vs. NHE) is too low to initiate the CO₂ reduction reaction into methanol (CH₃OH/CO₂ −0.38 V vs. NHE). In other words, the individual Si nanospheres or TiO₂ nanosheets lack suitable VB or CB potential to photoreduce CO₂ into methanol.

Fig. 11 Schematic diagram of the enhanced photocatalytic property of the Si/TiO₂ direct Z-scheme system for CO₂ reduction.

4. Conclusion

In summary, we have presented a facile and low cost method to prepare a direct Z-scheme Si/TiO₂ nanostructure via hydrothermal reaction with tetrabutyl titanate and Si powder which was prepared from the magnesiothermic reduction of SiO₂ Stöber nanospheres. All the results indicate that Si/TiO₂ nanocomposites possess much higher photocatalytic activity than individual Si and TiO₂ samples for the CO₂ conversion and degradation of RhB. This excellent performance could be attributed to the integrated suitable conductive band of Si and valence band of TiO₂ for CO₂ reduction and improved light absorption ability, enhanced concentration of photogenerated carriers, and higher separation efficiency due to the elaborate construction of the Si/TiO₂ direct Z-scheme system.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (no. 51172109), the Fundamental Research Funds for the Central Universities (no. NS2014057), Funding of Jiangsu Innovation Program for Graduate Education (no. CXLX12_0148), and the Fundamental Research Funds for the Central Universities. This work is also supported by KFUPM through project # RG1011-1/2.

Notes and references

1. C. J. Wang, O. Ranasingha, S. Natesakhawat, P. R. Ohodnicki, M. Andio, J. P. Lewis and C. Matranga, Nanoscale, 2013, 5, 6968–6974.
2. F. C. Meunier, Angew. Chem., Int. Ed., 2011, 50, 4053–4054.
3. Y. S. Chaudhary, T. W. Woolerton, C. S. Allen, J. H. Warner, E. Pierce, S. W. Ragsdale and F. A. Armstrong, Chem. Commun., 2012, 48, 58–60.
4. S. C. Yan, S. X. Ouyang, J. Gao, M. Yang, J. Y. Feng, X. X. Fan, L. J. Wan, Z. S. Li, J. H. Ye, Y. Zhou and Z. G. Zou, Angew. Chem., Int. Ed., 2010, 49, 6400–6404.
5. Q. Liu, Y. Zhou, J. H. Kou, X. Y. Chen, Z. P. Tian, J. Gao, S. C. Yan and Z. G. Zou, J. Am. Chem. Soc., 2010, 132, 14385–14387.
6. N. Zhang, S. X. Ouyang, T. Kako and J. H. Ye, Chem. Commun., 2012, 48, 1269–1271.
7. P. Q. Wang, Y. Bai, P. Y. Luo and J. Y. Liu, Catal. Commun., 2013, 38, 82–85.
8. T. Wang, X. G. Meng, P. Li, S. X. Ouyang, K. Chang, G. G. Liu, Z. W. Mei and J. H. Ye, Nano Energy, 2014, 9, 50–60.
9. S. S. Tan, L. Zou and E. Hu, Catal. Today, 2006, 115, 269–273.
10. O. Ozcan, F. Yukruk, E. U. Akkaya and D. Uner, Top. Catal., 2007, 44, 523–528.
11. K. R. Thampi, J. Kiwi and M. Graetzel, Nature, 1987, 327, 506–508.
12. A. L. Linsebigler, G. Q. Lu and J. T. Yates, Chem. Rev., 1995, 95, 735–758.
13. S. T. Hussain, K. Khan and R. Hussain, J. Nat. Gas Chem., 2009, 18, 383–391.
14. O. K. Varghese, M. Paulose, T. J. Latempa and C. A. Grimes, Nano Lett., 2009, 9, 731–737.
15. J. G. Yu, G. P. Dai, Q. J. Xiang and M. Jaroniec, J. Mater. Chem., 2011, 21, 1049–1057.
16. K. Koci, L. Matejova, L. Obalova, S. Krejckov, Z. Lacny, D. Placha, L. Capek, A. Hospodkovad and O. Solcova, Appl. Catal., B, 2010, 96, 239–244.
17. W. B. Hou, W. H. Hung, P. Pavaskar, A. Goeppert, M. Aykol and S. B. Cronin, ACS Catal., 2011, 1, 929–936.
18. K. P. Yu, W. Y. Yu, M. C. Kuo, Y. C. Liou and S. H. Chien, Appl. Catal., B, 2008, 84, 112–118.
19. Y. Li, W. N. Wang, Z. Zhan, M. H. Woo, C. Y. Wu and P. Biswas, Appl. Catal., B, 2010, 100, 386–392.
20. C. J. Wang, R. L. Thompson, P. Ohodnicki, J. Baltrus and C. Matranga, J. Mater. Chem., 2011, 21, 13452–13457.
21. S. Qin, F. Xin, Y. Liu, X. Yin and W. J. Ma, J. Colloid Interface Sci., 2011, 356, 257–261.
22. Q. D. Truong, J. Y. Liu, C. C. Chunq and Y. C. Ling, Catal. Commun., 2011, 19, 85–89.
23. P. Zhou, J. G. Yu and M. Jaronniec, Adv. Mater., 2014, 26, 4920–4935.
24. H. Tada, T. Mitsui, T. Kiyonaga, T. Akita and K. Tanaka, Nat. Mater., 2006, 5, 782–786.
25. H. Lin, J. Cao, B. Luo, B. Xu and S. Chen, Catal. Commun., 2012, 21, 91–95.
26. Z. B. Yu, Y. P. Xie, G. Liu, G. Q. Lu, X. L. Ma and H. M. Cheng, J. Mater. Chem. A, 2013, 1, 2773–2776.
27. J. G. Yu, S. H. Wang, J. X. Low and X. Wei, Phys. Chem. Chem. Phys., 2013, 15, 16883–16890.
28. M. Miyauchi, Y. Nukui, D. Atarashi and E. Skai, ACS Appl. Mater. Interfaces, 2013, 5, 9770–9776.
29. Y. C. Wang, J. Tang, Z. Peng, Y. H. Wang, D. S. Jia, B. Kong, A. A. Elzatahry, D. Y. Zhao and G. F. Zheng, Nano Lett., 2014, 14, 3668–3673.
30. M. W. Shao, L. Cheng, X. H. Zhang, D. D. D. Ma and S. T. Lee, J. Am. Chem. Soc., 2009, 131, 17738–17739.
31. N. Megouda, Y. Cofininier, S. Szunerits, T. Hadjersi, O. ElKechai and R. Boukherroub, Chem. Commun., 2011, 47, 991–993.
32. Y. J. Hwang, A. Boukai and P. D. Yang, Nano Lett., 2009, 9, 410–415.
33. J. Shi and X. D. Wang, Energy Environ. Sci., 2012, 5, 7918–7922.
34. Q. D. Wu, D. Z. Li, Z. X. Chen and X. Z. Fu, Photochem. Photobiol. Sci., 2006, 5, 653–655.
35. C. Liu, J. Y. Tang, M. Chen, B. Liu and P. D. Yang, Nano Lett., 2013, 13, 2989–2992.
36. W. Stöber, A. Fink and E. J. Bohn, J. Colloid Interface Sci., 1968, 26, 62–69.
37. M. A. Gondal, M. A. Ali, X. F. Chang, K. Shen, Q. Y. Xu and Z. H. Yamani, J. Environ. Sci. Health, Part A, 2012, 47, 1571–1576.
38. X. F. Chang, J. Huang, C. Cheng, Q. Sui, W. Sha, G. B. Ji, S. B. Deng and G. Yu, Catal. Commun., 2010, 11, 460–464.
39. D. R. Lide, Handbook of Chemistry and Physics, CRC Press, Florida, 87th edn, 2006–2007.
40. S. Suzuki, T. Tsuneda and K. Hirao, J. Chem. Phys., 2012, 136, 024706.
41. C. Y. Wang, J. Rabani, D. W. Bahnemanna and J. K. Dohrmann, J. Photochem. Photobiol., A, 2002, 148, 169–176.
42. N. Schüler, K. Hecht, M. Kraut and R. Dittmeyer, J. Chem. Eng. Data, 2012, 57, 2304–2308.
43. L. Q. Jing, Y. C. Qu, B. Q. Wang, S. D. Li, B. J. Jiang, L. B. Yang, W. Fu, H. G. Fu and J. Z. Sun, Sol. Energy Mater. Sol. Cells, 2006, 90, 1773–1787.

---