Visible-light photocatalytic SiO₂/TiO_2−xC_x/C nanoporous composites using TiCl₄ as the precursor for TiO₂ and polyhydroxyl tannin as the carbon source

Abstract

We report an architecturally controlled synthesis of SiO₂/TiO_2−xC_x/C nanoporous composites that exhibit high absorption capability and efficient visible-light photocatalytic activity. The nanoporous composites are composed of silica particles as the cores and TiCl₄ as the precursor for the TiO₂ shell. Tannin is used as the binding agent between the core and the precursor shell, the carbon source, and the porosity promoter. The structure, crystallinity, morphology, and other physical–chemical properties of the samples are characterized by X-ray diffraction (XRD), transmission electron microscopy (TEM), high-resolution transmission electron microcopy (HRTEM), X-ray photoelectron spectroscopy (XPS), N₂ adsorption–desorption isotherm measurements, UV–vis diffuse reflectance spectroscopy (DRS), and photoluminescence (PL). The chemical contents of the SiO₂/TiO_2−xC_x/C nanoporous composites were also analyzed by energy dispersive X-ray spectra (EDX). The formation mechanism of the nanoporous composites was extensively discussed. Methylene blue (MB) solutions were used as model wastewater to evaluate the adsorption and photocatalytic activity of the samples under natural sunlight and visible light. Fourier transform-infrared spectroscopy (FT-IR) and mass spectrometry (MS) were used to investigate the photodegradated species on the photocatalysts and in solution, respectively. The SiO₂/TiO_2−xC_x/C nanoporous composite samples exhibit remarkably enhanced visible-light photoactivity than Degussa P25 and pure TiO₂, and can be readily collected for reuse by gravitational sedimentation.

---

1. Introduction

Domestic wastewater is first pretreated through coagulation, flocculation, sedimentation, and filtration to remove most of the solid debris and sludge. After that, the pretreated water still contains considerable amounts of small organic pollutants, such as pharmaceuticals (hormones, antibiotics, antiseptics etc.), food additives (food coloring, flavors etc.), cosmetic dyes, pigments, fragrancesetc. These small organic pollutants must be further disinfected, typically by chlorine-based bleaching agents. However, chlorine bleach is a consumable agent and leads to concerns regarding its production, storage, transportation, safety, costs, and health hazards. By contrast, semiconductor-catalyzed photo-oxidative degradation of organic pollutants powered by sunlight is a promising and cleaner scheme for energy and environment sustainable disinfection of wastewater.^1,2 The excellent chemical stability, nontoxicity, and photoactivity of anatase titania (TiO₂) have attracted much attention because of its potential use as a solid photocatalyst for environmental purification.^3,4 However, one of the major drawbacks that hinders the practical applications of pristine TiO₂-based photocatalytic oxidation is the large band gap of TiO₂ (∼3.2 eV) that only partially overlaps with the solar spectrum in the UV region. As a consequence, only less than 5% of the UV solar energy can be utilized to generate electron–hole pairs and to initiate the photocatalytic oxidation process. Hence, it is crucial to enhance the light harvesting of TiO₂ in the visible region which accounts for more than 43% of the total solar energy.⁵

In general, two methods are utilized to enhance the light harvesting of TiO₂ in the visible region. One method is to sensitize TiO₂ with color centers that have proper energy levels for photoelectron transfer between the color centers and TiO₂.^6,7 A cost-effective sensitization of TiO₂ is typically achieved by addition of non-metals such as I^8,9 and S.^10,11

The other way is to form a doping level through intercalation of proper atoms into the anatase form. For example, by doping carbon^12,13 and nitrogen^14–16 atoms into the TiO₂ lattice, a new, weak absorption is generated at lower energies, which is a promising approach for inducing effective vis-absorption. This is because the C_2p or N_2p orbital has a higher energy than that of O_2p. The C or N doping involves substitution of O atoms by C or N producing new energy states deep in the TiO₂ band gap, which are responsible for the visible light absorption. Direct doping of impurity into the lattice of anatase TiO₂ by using a chemical method is not feasible because it requires very high temperature, causing transition of photoactive anatase to non-active rutile, and thermal rupture of the desired TiO₂ structure. Thus, the dopants must be premixed with the precursor TiO₂ at a molecular level, followed by the conversion of the precursor TiO₂ to anatase. Thus, organotitanium compounds are often used as the precursor source for TiO₂ and carbon dopants. However, organotitanium is too expensive for mass production. Although doping by a gas-phase method can avoid this defect, the requirement of equipment is relatively high.¹⁶

In addition to the narrow band gap, the enhancement of surface area of the TiO₂ is another important aspect for the efficiency of the materials, because the desired photocatalytic reactions occur at the interfaces of the catalyst-pollute solution.¹⁷ Intuitively, a plausible approach is to maximize the photoactivity of TiO₂ through synthesis of small titania nanoparticles (NP) so as to possess a large surface area and a large number of reactive sites. However, the small particle size makes it a practical challenge for the reuse of the TiO₂-nanoparticle-based photocatalysts because the separation of the nanoparticles from the pollute solutions requires a costly centrifugation process. Moreover, reduction in particle size also introduces more internal crystal defects that act as charge trap states to diminish the effective transport of photoinduced charges to the external surface sites,^18,19 where the desired photocatalytic reaction occurs.

It has been confirmed that TiO₂/carbon composites or carbon-modified TiO₂ could increase adsorption capacity and improve photodegradation performance in the visible region.^20–25 Composites of TiO₂/C have also been shown to improve the energy and power density of electrochemical cells and enhance the energy-storage capacity.²⁶ Therefore, TiO₂ composited with different forms of carbon, such as activated carbon,^27,28graphite,²⁹ and carbon nanotubes,^30,31 have been fabricated to obtain novel structures and properties. Various organic compounds are suitable as carbon sources.^32,33 The prepared carbon-coated anatase nanocrystals exhibit fast adsorption of MB with high capacity.²⁹

Furthermore, as most photocatalytic reactions occur at the surface region of the TiO₂ particles, it is not an economical configuration to have a large amount of photocatalytic materials residing in the internal bulk of the particles which are not in direct contact with the pollute solutions.

In all these regards, it is exceptionally warranted to develop a TiO₂ matrix with a large surface area for adsorbing more pollute molecules and proper doping for visible light harvesting. However, simultaneously the TiO₂ matrix must maintain a large particle size so that the photocatalysts can be conveniently collected through gravitational precipitation. Logically, a larger particle size usually leads to less surface area, and these two factors are diametrically opposing. However, an effective way to decouple this inverse interlink is to construct highly porous large particles.

Hence, in this work, we aim to integrate (1) the synthesis of large TiO₂ particles with high porosity for enhanced adsorption of pollute molecules, (2) carbon doping in TiO₂ for photoactivity in the visible region, and (3) minimizing the usage of TiO₂ by replacing some internal TiO₂ with materials that are more cost-effective, e.g.silica. We chose silica sol as the internal stuffing material because of its rich surface hydroxyl groups that enable us to covalently bind precursor carbon dopants, and thus to spatially control the pathways of the doping process. Moreover, we also aim to coat a non-continuous ultrathin layer of amorphous carbon on the surface of TiO₂, which can further boost the adsorption of pollute molecules.^34,35 (4) The starting materials for TiO₂ should be largely available based on the current industrial infrastructure. In industry, TiCl₄ is an intermediate product during the mass production of TiO₂, known as a chloride process (Dupont method). This process is widely used in North America due to its relatively low energy consumption and better product consistency.^36,37 Therefore, for the commercialization of our photocatalysts, the use of TiCl₄ as the precursor for TiO₂ is more practical and attractive from an industrial point of view since it does not require additional production infrastructure. In contrast, the use of organotitanium as the precursor for TiO₂ will inevitably involve additional costs and complications for the mass synthesis of organotitanium.

In this work, we utilize tannin as the binding agent, carbon source, and porosity promoter to achieve SiO₂ (core)–C-doped-TiO₂ (shell) nanoporous composites. In addition, an ultrathin amorphous carbon layer forms on the surface of TiO₂ to further enhance the surface area. This structure is termed as SiO₂/TiO_2−xC_x/C. Tannin is an organic molecule with multiple hydroxyl groups that can be utilized as a binding agent to glue the precursor SiO₂ core with the precursor TiO₂ shell, since both SiO₂ and TiO₂ are abundant with surface hydroxyl groups. The gaseous species, resulting from the decomposition of tannin under heat treatment, can pulverize the TiO₂ shell to promote the formation of a nanoporous shell. Moreover, excessive tannin molecules can further anchor on the TiO₂ shell to form a precursor carbon layer that can be converted to an ultrathin layer of amorphous carbon upon carbonization. The extra SiO₂ particles in the composition serve as agglomeration centers in order to facilitate the interconnection between the particles. Thus, large TiO₂ particles with high porosity and SiO₂ cores are expected to form. As a proof-of-concept, we demonstrate that SiO₂/TiO_2−xC_x/C nanoporous composites can be synthesized, exhibit an excellent adsorption and photodegradation of model pollute molecules in aqueous solution under visible light radiation, and be readily collected via gravitational sedimentation.

2. Experimental

2.1 Materials and sample preparation

Commercial acidic silica sol (30 wt% silica with an average particle size of ∼10 nm) was purchased from Yancheng Junying Chemicals. Titanium tetrachloride (TiCl₄) and analytical reagent grade tannin (C₇₆H₅₂O₄₆) were purchased from Aldrich, without further purification. Commercial Degussa P25 with 80% anatase, 20% rutile, and a BET area of ca. 50 m² g⁻¹ was purchased from Degussa Co.

A typical procedure for preparing the SiO₂/TiO_2−xC_x/C nanoporous composites is described as follows. First, 20 g TiCl₄ was added drop-wise slowly into 80 mL of distilled water in an ice bath. The obtained product was labeled as the TiCl₄ stock solution. Then, 1.0 g tannin powder was dissolved in 30 mL of distilled water and the resulting solution was added drop-wise into 10 g 30 wt% silica sol. Under stirring, this mixture solution was warmed in a water-bath at 80 °C for 3 h. The above procedures lead to the formation of the SiO₂–tannin composites.

Then, the freshly prepared TiCl₄ stock solution was added slowly to the SiO₂–tannin composite solution and the resulting mixture was kept at 60 °C for 4 hours. After that, the mixture became a suspension, which was continuously stirred for an additional 12 h at room temperature. Finally, the mixed slurry was desiccated at 80 °C for 6 h and at 120 °C for 6 h to yield a solid composition. In order to obtain carbon-doped and carbon coated TiO₂, these solid composites were further heated at various temperatures under a nitrogen atmosphere. The heating rate was 5 °C min⁻¹ and the flow rate of nitrogen was 50 mL min⁻¹. In the final stage of the heat treatment, 5 v% of air was introduced and maintained for 20 min. After heat treatment, the resulting powders were cooled down to room temperature in a nitrogen atmosphere. The samples are now denoted as STC-T, where the second T refers to the applied temperature during the heat treatment. For comparison, pristine TiO₂ powder calcined at 500 °C (denoted as TiO₂-500) was also prepared without the addition of Si sol and tannin.

2.2 Sample characterization

The phases of the final products were identified using an X-ray diffractometer (XRD) (Rigaku D/max-2500VPC) with Ni-filtered Cu Kα radiation at a scanning rate of 0.02° s⁻¹ from 10° to 70°. The X-ray photoelectron spectra (XPS) measurements were carried out on an X-ray photoelectron spectrometer (ESCALAB MK II) using Mg Kα (1253.6 eV) X-rays as the excitation source, C1s (284.6 eV) was chosen as the reference. The morphology of the samples was observed using a Hitachi model H800 transmission electron microscope (TEM) and a JEOL Ltd model JEOL-2010 high-resolution transmission electron microscope (HRTEM). The porous structure and BET surface area were characterized by a N₂ adsorption–desorption isotherm (ASAP-2020 Micromeritics Co., USA). The samples were degassed at 180 °C prior to the BET measurements. The pore volume and diameter distribution were derived from the desorption branches of the isotherms by the Barrett–Joyner–Halenda (BJH) model. The BET surface area was calculated from the linear part of the Brunauer–Emmett–Teller (BET) plot. Fourier transform-infrared (FT-IR) spectra were recorded on a MAGNA 550 FT Infrared Spectrometer on samples embedded into KBr pellets. The photoluminescence (PL) spectra were measured at room temperature with a Perkin Elmer LS 55 Fluorescence Spectrometer, using pulsed Xenon discharge lamps as excitation sources. Ultraviolet–visible (UV-vis) diffuse reflection spectra (DRS) were recorded on a Shimadzu UV-2550 UV-vis spectrophotometer in the range of 200–800 nm at room temperature. The mass spectra (MS) were taken on an electrospray ionization mass spectrometer (Bruker ESQUIRE 3000).

2.3 Adsorption and photocatalytic activity measurements

The adsorption capability and photocatalytic activity of the SiO₂/TiO_2−xC_x/C nanoporous composites were evaluated by measuring the adsorption and decomposition rates of MB solution at room temperature. For comparison, the same measurements were also performed on the pure TiO₂-500 and commercial P25. Visible light was obtained from a 300 W tungsten arc lamp (Zhejiang Electric Co., Ltd.) with a glass filter (transparent for λ > 400 nm). For a typical absorption and photocatalytic experiment, 150 mg of the sample was added to 200 mL MB aqueous solution (5.0 × 10⁻⁴ mol L⁻¹) in a custom-made quartz reactor. The MB concentration was monitored by the UV–vis spectroscopy during the entire experiment. After 5 min intervals, during adsorption and visible light illumination, about 3 mL aliquots were taken out and centrifuged to remove the trace particles. The absorbance of the centrifuged solution was measured in the range of 500–800 nm using a UV–vis spectrophotometer (Shimadzu UV-2550).

3. Results and discussion

3.1 Characterization of samples

Powder X-ray diffraction (PXRD) is used to investigate the changes of structure and crystallite sizes of the STC-T samples obtained at different heat treatment temperatures. Fig. 1 shows the XRD patterns of the samples heated at various temperatures from 400 to 700 °C. The patterns show that all samples are anatase TiO₂. It is worth noting that no rutile phase is detected even for the sample calcined at 600 °C, indicating an excellent thermal stability of our samples against phase transformation, which can be ascribed to the SiO₂ composite and carbon coating.⁷ For sample STC-700, the predominant phase is still anatase TiO₂, which is accompanied by a very small portion of the rutile phase (corresponding to the peak at 2θ = 27.5°). As heat treatment is a necessary step for the formation of a porous structure, the high thermal stability of our samples is an important feature for maintaining anatase (the photoactive phase) under high temperature treatment. In addition, as the heat treatment temperature increases, the XRD peaks become sharper and more intense due to the formation of larger grains as summarized in Table 1.

Fig. 1 XRD patterns of STC-T porous composites heat treatment at different temperatures. (A = anatase; R = rutile).

Table 1 The average crystal size and physical properties of STC-T samples based on N₂ sorption analysis and XRD results^a

a Crystal size was determined based on the XRD peak (101) and the Debye–Scherrer equation. BET surface areas were calculated from the linear part of the BET plots. Total pore volumes were obtained from the volume of N2 adsorbed at P/P0 = 0.995. Average pore diameters were estimated using the desorption branch of the isotherm and the Barrett–Joyner–Halenda (BJH) formula.
Heat treatment temperature/°C	400	500	600	700
Crystalline size/nm	9.3	11.4	14.3	21.6
BET specific surface area/m^2 g^−1	392.4	366.8	305.7	263.5
BJH pore volume/cm^−3 g^−1	0.547	0.492	0.387	0.316
Average pore diameter/nm	8.423	9.745	11.612	13.327

---

In general, the crystallinity of anatase TiO₂ can be improved by high temperature treatment, but at the risk of phase transition of the photoactive anatase to a poor photoactive rutile. Carbon doping, coating, and SiO₂–TiO₂ composite can mitigate this dilemma by effectively stabilizing the anatase structure of TiO₂ under high temperature.^38,39 This is evidenced by the fact that our SiO₂/TiO_2−xC_x/C nanoporous composites calcined at 700 °C still retain the anatase phase.

The specific surface area and pore size of the samples were characterized by nitrogen adsorption–desorption isotherm measurements. Table 1 summarizes the BET specific surface area and the apparent average pore size of our samples. The pore volume and specific surface area decrease with increasing calcination temperature. However, the samples still maintain relatively high specific surface area and pore size even for sample STC-700. This high thermal stability of our sample is an important feature for maintaining high surface area since heat treatment is a necessary step for the formation of a porous structure.

XPS measurements were carried out to investigate the presence, contents, and chemical states of Ti, Si, C, and O in the samples. The XPS of the STC-500 composite with a Ti/Si ratio of 2.1 : 1 are presented in Fig. 2. In the survey spectra (Fig. 2(a)), the elements Ti, Si, O and C are clearly identified. Fig. 2(b) shows the high-resolution XPS (HR-XPS) of O1s, showing a broad peak that can be split into four peaks at 529.6, 531.2, 532.0, and 533.1 eV. These four peaks are attributed to Ti–O–Ti, Ti–O–Si, –OH, and Si–O–Si bonds, respectively.⁴⁰ The binding energy of the O1s electron of Ti–O and Si–O in the composite is 1.37 eV higher than that in pure TiO₂,⁴¹ but is 0.26 eV lower than that of Si–O in pure SiO₂. The shift of the binding energy of the O1s electron in our samples can be ascribed to the bond of Ti–O–Si.²² The HR-XPS of Si2p (Fig. 2(c)) show two peaks at 103.7 eV and 101.6 eV, which correspond to Si–O–Si and Si–O–Ti bonds, respectively.⁴²

Fig. 2 XPS spectra of STC-500 with a Ti/Si molar ratio of 2.1 : 1: (a) survey spectrum; (b)–(e) O1s, Si2p, Ti2p and C1s high-resolution spectrum, respectively. (f) EDX spectrum of the STC-500.

Fig. 2(d) and (e) show the HR-XPS of Ti2p and C1s core levels. The Ti2p spectrum can be deconvoluted into three peaks at 458.3, 459.0, and 459.5 eV, which can be attributed to Ti–C–Ti, Ti–O–Ti, and Ti–O–Si bonds, respectively.^43,44 The four peaks at 282.3, 284.9, 286.2, and 288.7 eV were observed for C1s. The C–Ti bond is located at ca. 282.5–281.5 eV and the C–Si bond is located at about 282.2–282.8 eV.⁴⁵ Accordingly, the C1s peak at 282.4 eV can be ascribed to the substitution of an oxygen atom by a carbon atom in the lattice of TiO₂, leading to an O–Ti–C bond.⁴⁶ The peak at 284.9 eV is due to the elemental carbon at the surface of TiO₂.⁴⁷ This is further evidenced by a TEM study, which will be described below. The peaks at 286.2 and 288.7 eV are likely due to the C–H and C–O bonds in residual tannin.

XPS analysis of the STC-500 sample demonstrated that the atomic concentrations of O, Ti, C, and Si are 67.38%, 15.12%, 12.04%, and 5.01%, respectively. The atomic ratio of titanium to silicon was about 3 : 1. In the sample preparation process, the ratio of titanium to silicon is ca. 2.1 : 1, the XPS results show that the ratio of titanium to silicon is significantly higher than the synthetic stoichiometry. Since XPS is a surface sensitive analysis technique, the higher content of Ti found in the XPS study suggests that Ti is preferred on the surface region while Si is more preferential in the bulk. As indirect and supportive evidence, EDX analysis in Fig. 2(f) shows that the atomic ratio of Ti to Si is 2.4 : 1, close to the synthetic stoichiometry. This is because EDX is not a surface-sensitive technique and it detects the elements in deeper layers. Thus, it is not surprising that the EDX result is closer to the stoichiometry based on the sample preparation. The XPS and EDX studies on other samples are summarized in Table 2. With an increase in the heat treatment temperature, the carbon content decreases gradually, indicating the loss of tannin at higher temperatures. Meanwhile, with the increase in the heat treatment temperature, the ratio of Ti to Si becomes closer to the synthetic stoichiometry. This can be understood as the result of sintering of the TiO₂ with SiO₂ at elevated temperature.

Table 2 The concentrations of O, Ti, C and Si of STC-T samples

	STC-400		STC-500		STC-600		STC-700
	XPS	EDX	XPS	EDX	XPS	EDX	XPS	EDX
O (%)	68.11	67.56	67.38	66.83	67.04	66.42	66.87	66.12
Ti (%)	14.53	15.32	15.12	16.92	15.98	17.46	16.75	17.98
C (%)	12.99	11.19	12.49	9.37	11.66	8.69	10.49	8.02
Si (%)	4.37	5.93	5.01	6.88	5.32	7.43	5.89	7.88

---

The typical microstructures of transparent silica sols and STC-T samples were characterized by TEM and HRTEM analysis, and the results are shown in Fig. 3. Also, Fig. 3a shows that the silica sol consists of silica nanospheres of ∼10 nm in diameter with a narrow size distribution. Fig. 3b shows the TEM image of sample STC-400, however, STC-400 does not exhibit an obvious porous structure, but the SiO₂ nanoparticles embedded in the TiO₂ can be observed. As the calcination temperature increases to 500 °C, the sample becomes more porous as shown in the TEM image (Fig. 3c). Moreover, there is a very thin layer of amorphous carbon covering the surface as shown in the HRTEM image (Fig. 3d). A further elevation of heat treatment to 600 °C yields a sample with a bimodal meso/nanoporous structure (Fig. 3e). More interestingly, we also observed the cavity core/shell structure (Fig. 3f) in sample STC-600. We assume that the complete decomposition of the tannin layer that is sandwiched between the SiO₂ nanoparticles and the outer TiO₂ layer led to the pulverization of the TiO₂ layer. When the heat treatment temperature reaches 700 °C, the nanoporous structure diminished due to sintering, while the mesopores remain as shown in Fig. 3g. In the HRTEM image of STC-700 (Fig. 3h), the lattice structure of TiO₂ can be clearly indexed. Carbon is clearly visible at the surface region of the composite. Overall, TEM images clearly demonstrate the temperature effect on the morphology of SiO₂/TiO_2−xC_x/C nanoporous composites.

Fig. 3 (a) and (b) TEM images of transparent colloidal silica sol and STC-400. (c) and (d) TEM and HRTEM images of STC-500. (e) and (f) TEM images of STC-600. (g) and (h) TEM and HRTEM images of STC-700.

3.2 SiO₂/TiO_2−xC_x/C nanoporous composite formation mechanism

Combining the results from XPS, EDX, TEM, FT-IR spectra, and BET analysis one could conclude that some C atoms are doped into the structures of TiO₂, while a significant amount of amorphous carbon is doped on the surface of TiO₂. Meanwhile, XPS and IR also suggest the presence of Ti–O–Si and Si–O–Si. Based on these results, we hereby propose a possible multistep formation mechanism that is schematically elucidated in Scheme 1. In this mechanism, tannin plays an important role in the formation of SiO₂/TiO_2−xC_x/C nanoporous composites.

Scheme 1 Schematic presentation of the fabrication of SiO₂/TiO_2−xC_x/C nanoporous composites.

First, there are many surface hydroxyl groups on the SiO₂ colloid particles, which provide sufficient sites to bind tannin molecules containing multi-hydroxyl groupsviahydrogen bonds (Schemes 1(a) and (b)). Upon heating, the surface active sites on silica particles can react with the attached tannin through dehydration. Note that there are still excessive tannin molecules in the solution phase as indicated in Scheme 1c.

As TiCl₄ solution was added into the above mixture, part of the TiCl₄ hydrolyzes at the surface of the SiO₂ nanoparticles due to the enriched surface hydroxyl groups provided by the surface tannin molecules. This leads to the formation of a shell of precursor TiO₂ that attaches to the SiO₂ nanoparticleviatannin molecules. Because of the abundant surface hydroxyl groups on the shell of the TiO₂ layer, the excessive tanning molecules in the solution phase can further attach to the surface of the TiO₂ shell viahydrogen bonds (Scheme 1d).^48,49 This structure is termed as SiO₂/tannin/TiO₂. However, we do not exclude the possibility that a certain amount of TiCl₄ hydrolyzed directly in the solution to form independent amorphous TiO₂ particles, and some SiO₂ particles are not wrapped by TiO₂.

Upon gradual evaporation of the solvent, the SiO₂/tannin/TiO₂ particles, the TiO₂ particles, and the SiO₂ particles aggregate to form the precursor composites (Scheme 1e).

The initial calcinations were carried out in a nitrogen atmosphere, which leads to the crystallization of TiO₂, carbonization of tannin (i.e. the formation of the amorphous carbon layer at the surface of the TiO₂ shell), carbon intercalation (i.e.carbon doping into the TiO₂ lattice), and the pulverization of the TiO₂ layer due to the release of gaseous species from the decomposition of the tannin layer in between the SiO₂ cores and the TiO₂ shell. In order to minimize the oxygen vacancies in the TiO₂ lattice upon heating, air was introduced during the final calcination stage. However, this can also lead to the loss of carbon, which is evident in the TEM and XPS studies.

3.3 Optical properties and photocatalytic activity

Fig. 4(a) shows the UV-Vis DRS spectra of the prepared STC-T composite samples, the pure TiO₂-500, and the commercial P25 TiO₂ powders. In comparison to the spectra of pure TiO₂-500 and P25, the spectra of the STC-T samples exhibit a strong broad absorption band between 400–700 nm, covering nearly the entire visible range. The intensity of the visible light absorption decreases with the increase in the heat treatment temperature. This can be ascribed to the loss of carbon at higher temperatures in the final heating stage in the presence of air, which is also evident from the color change of the samples as shown in Fig. 4(b). The sample obtained at 400 °C showed the maximum absorption in the visible region in comparison to other samples. The absorption in the visible-light region implies that the prepared samples can be activated by visible light and that more electrons and holes can be photogenerated and participated in the desired photocatalytic reactions.

Fig. 4 (a) UV-vis DRS of STC-T, pure TiO₂-500 and P25. (b) Photos of STC-T samples.

Photoluminescence (PL) emission spectra have been widely used to explore the efficiency of charge carrier trapping, migration, and recombination in order to understand the fate of electron–hole pairs in semiconductor particles.⁵⁰Fig. 5(a) shows the room-temperature PL spectra for pure TiO₂-500 and STC-T samples excited by UV light at 320 nm. It is clear that the PL intensity of the pure TiO₂-500 is higher than that in the spectra of STC-T composite samples. This indicates that carbon doping, SiO₂ composites, the surface amorphous carbon, and surface hydroxyl groups serve as charge trap states to suppress the radioactive electron and hole recombination.^51,52

Fig. 5 PL spectra of TiO₂-500 and STC-T samples under UV-320 nm (a) and visible light 520 nm (b).

However, for excitation with visible light at 520 nm, the STC-T samples exhibit a much better PL effect than pure TiO₂-500 (Fig. 5b). The disappearance of PL in TiO₂-500 is due to the insufficient energy of the visible photons that is lower than the band gap of pure TiO₂. In contrast, for the carbon doped STC-T samples, the reduced band gap allows absorption in the visible region to produce photoelectrons and holes.⁵³

Methylene blue (MB) is a brightly colored blue cationic thiazine dye and is often used as a test model pollutant in semiconductor photocatalysis. Fig. 6(a) shows the adsorption and photodegradation ability of methylene blue (MB) solutions by using our STC-T samples in comparison to P25 and pure TiO₂-500. A blank experiment was also carried out. The pure MB solution cannot decompose without a catalyst. Prior to the turn-on of the solar light, the concentrations of MB in the presence of the STC-T samples deplete much faster than those in the P25 and TiO₂-500 samples. This is especially seen for the STC-400 sample, nearly 60% of the dye molecules were adsorbed within 30 min. We ascribe this effect to the much larger surface area of the STC-T samples than that of the P25 or TiO₂-500 samples as characterized and discussed above. For STC-T samples, their adsorption capacity increases with the decrease in the heat treatment temperatures. This can be credited to the reduced surface area and the loss of the surface amorphous carbon upon sintering and oxidation of carbon at higher temperatures.

Fig. 6 (a) The blank MB solution, adsorption and photodegradation of MB solutions by using pure TiO₂-500, P25 and STC-T samples as photocatalysts under visible light irradiation. (b)–(f) The photographs of 20 mg STC-T and P25 in 100 mL MB solution under sunlight in the static state. The photograph of a pure MB solution is also shown for comparison.

Furthermore, Fig. 6(a) shows that under visible light illumination, the MB solutions containing the STC-T samples underwent significant degradation and became nearly transparent within 40 min. It should be noted that the STC-500 sample exhibits the best photocatalytic activity and within 23 minutes the absorption band due to MB nearly diminishes. In contrast, the MB solutions containing pure TiO₂-500 and P25 show very limited degradation.

Fig. 6(b)–(f) shows the real-time photographs of the samples under radiation by natural sunlight. It is striking that after initial stirring, the MB solutions containing STC-400 and STC-500 samples become nearly transparent and the powders are all precipitated. In fact, over 90% of the methylene blue was degraded after 25 min of sunlight irradiation at 18 °C.

In contrast, the MB solution containing P25 is still opaque and bluish. This simple separation by gravitational sedimentation exhibited by the STC-400 and STC-500 samples increases the potential use of our samples as practical photocatalysts for degradation of organic pollutants. In comparison, the pure TiO₂-500 sample and P25 showed very little MB adsorption and the suspensions have to be separated by centrifugation.

The cyclic stability test under the same conditions of the STC-500 sample was also examined and shows excellent photocatalytic activity in degrading MB solution, as shown in Fig. 7. No significant decrease in its photocatalytic oxidation activity is observed after being used 6 times, which makes it very promising for practical application.

Fig. 7 Recycling test results by using the STC-500 sample.

To eliminate the possibility that the enhanced photocatalytic activity of our STC-T samples mainly results from their enhanced physisorption alone, the FTIR and MS spectra of the STC-500 samples have been collected before and after the photodegradation of MB. As illustrated in Fig. 8, the FTIR spectra of sample STC-500 before (curve a in Fig. 8) and after degradation (curve b in Fig. 8) show no significant discrepancy. In contrast, the FTIR spectrum of MB (curve c in Fig. 8) shows more complicated vibrational modes due to the organic molecule. Therefore, it is unambiguous that the visible-light photodegradation on TiO₂ is the cause of the depletion of MB molecules.

Fig. 8 FT-IR spectra of STC-500 and pure methylene blue: (a) STC-500 before degradation, (b) STC-500 after degradation and (c) pure methylene blue.

For further confirmation, a mass spectrometer is used to study the species in the solution after photocatalysis. Fig. 9 suggests that the organic species in water undergo a structural degradation during the course of photocatalysis. After 30 minutes, the remaining species in solution are mainly formic acid and acetic acid.

Fig. 9 Mass spectra of (a) the methylene blue solution before photocatalysis, (b) after 10 minutes and (c) after 30 minutes photocatalysis with sample STC-500 under one Sun 1.5 AM G provided by a solar simulator (Photo Emission Inc. CA, model SS50B). The 47 and 61 peaks are attributed to the protonated formic acid and acetic acid respectively.

The high adsorption capacity of our nanoporous composite (the STC-T) samples is a product of several facts. First, the nanoporous structure resulting from the decomposition of tannin and the meso/nanopores composed of the voids between particles, together, provide a vast amount of surface active sites to adsorb reactive species. Moreover, the ultrathin amorphous carbon layer covering the surface of TiO₂ provides additional adsorption sites. The enhanced specific surface area consequently promotes the photocatalytic activity by increasing the concentration of contaminants and reaction intermediates near the surface of TiO₂. Carbon doping in the lattice of TiO₂ narrows the band gap, thus leading to visible light-induced charges that are essential for the desired enhanced photocatalytic oxidation. In addition, the near-surface doping of carbon in TiO₂ shortens the transport distance of photocharges to the surface region, where the desired photocatalytic reaction occurs.

4. Conclusions

SiO₂/TiO_2−xC_x/C nanoporous composites prepared by a simple carbonization have been confirmed to be visible-light-driven catalysts for photo-induced organic degradation. The addition of silica into titania and the carbonization of tannin can effectively suppress the transformation of titania from the anatase phase to the rutile phase and prevent the growth of TiO₂ grains during the heat-treatment process. The addition of silica and carbon into titania and carbon coating can increase the surface area of titania particles. The successful carbon doping of TiO₂ in this simple carbonization process is attributable to the retention of few-nanometre-sized TiO₂ clusters being highly reactive to carbon generated from the thermal decomposition of tannin. In the novel SiO₂/TiO_2−xC_x/C nanoporous composite, the high surface-area with excellent adsorption ability can concentrate organic molecules at the vicinity of the SiO₂/TiO_2−xC_x/C nanoporous composite. The SiO₂/TiO_2−xC_x/C nanoporous composite had good sedimentation ability. The technique presented here seems to be an economical and faster way for the preparation of a highly active photocatalyst. This may boost applications of the composite, particularly at the industrial scale.

Acknowledgements

TX and Liu thank NIU Summer Research and Artistry Grant and the U.S. Department of Energy under contract No. DE-AC02-06CH11357 for financial support. SG is grateful to National Basic Research Program of China (973 Program, Grant 2007CB613302). NSH thanks the support from the National Science Foundation (CHE-0906179).

References

1. M. Fujihira, Y. Satoh and T. Osa, Nature, 1981, 293(5829), 206–208.
2. M. R. Hoffmann, S. T. Martin, W. Y. Choi and D. W. Bahnemann, Chem. Rev., 1995, 95, 69–96.
3. J. Zhang, Q. Xu, Z. Feng, M. Li and C. Li, Angew. Chem., Int. Ed., 2008, 47, 1766–1769.
4. A. L. Linsebigler, G. Q. Lu and J. T. Yates, Chem. Rev., 1995, 95, 735–758.
5. S. Livraghi, M. C. Paganini, E. Giamello, A. Selloni, C. D. Valentin and G. Pacchioni, J. Am. Chem. Soc., 2006, 128, 15666–15671.
6. G. Liu, Y. Liu, G. Yang, S. Li, Y. Zu, W. Zhang and M. Jia, J. Phys. Chem. C, 2009, 113, 9345–9351.
7. S. Usseglio, P. Calza, A. Damin, C. Minero, S. Bordiga, C. Lamberti, E. Pelizzetti and A. Zecchina, Chem. Mater., 2006, 18, 3412–3424.
8. P. Xu, J. Lu, T. Xu, S. Gao, B. Huang and Y. Dai, J. Phys. Chem. C, 2010, 114, 9510–9517.
9. S. Usseglio, A. Damin, D. Scarano, S. Bordiga, A. Zecchina and C. Lamberti, J. Am. Chem. Soc., 2007, 129, 2822–2828.
10. P. Xu, T. Xu, J. Lu, S. Gao, N. S. Hosmane, B. Huang, Y. Dai and Y. Wang, Energy Environ. Sci., 2010, 3, 1128–1134.
11. H. X. Li, X. Y. Zhang, Y. N. Huo and J. Zhu, Environ. Sci. Technol., 2007, 41, 4410–4414.
12. S. Sakthivel and H. Kisch, Angew. Chem., Int. Ed., 2003, 42, 4908–4911.
13. G. Wu, T. Nishikawa, B. Ohtani and A. Chen, Chem. Mater., 2007, 19, 4530–4537.
14. J. Wang, D. N. Tafen, J. P. Lewis, Z. Hong, A. Manivannan, M. Zhi, M. Li and N. Wu, J. Am. Chem. Soc., 2009, 131, 12290–12297.
15. I. C. Kang, Q. W. Zhang, S. Yin, S. T. Sato and F. Saito, Environ. Sci. Technol., 2008, 42, 3622–3626.
16. R. Asahi, T. Morikawa, T. Ohwaki, K. Aoki and Y. Taga, Science, 2001, 293, 269–271.
17. Y. Ide, Y. Nakasato and M. Ogawa, J. Am. Chem. Soc., 2010, 132, 3601–3604.
18. Z. Yang, T. Xu, Y. Ito, U. Welp and W. K. Kwok, J. Phys. Chem. C, 2009, 113, 20521–20526.
19. Z. Yang, T. Xu, S. Gao, U. Welp and W. K. Kwok, J. Phys. Chem. C, 2010, 114, 19151–19156.
20. Y. C. Hsu, H. C. Lin, C. W. Lue, Y. T. Liao and C. M. Yang, Appl. Catal., B, 2009, 89, 309–314.
21. I. Moriguchi, R. Hidaka, H. Yamada, T. Kudo, H. Murakami and N. Nakashima, Adv. Mater., 2006, 18, 69–73.
22. G. S. Shao, L. Liu, T. Y. Ma, F. Y. Wang, T. Z. Ren and Z. Y. Yuan, Chem. Eng. J., 2010, 160, 370–377.
23. Y. P. Cheng, H. Q. Sun, W. Q. Jin and N. P. Xu, Chem. Eng. J., 2007, 128, 127–133.
24. M. Janus, M. Inagaki, B. Tryba, M. Toyoda and A. W. Morawski, Appl. Catal., B, 2006, 63, 272–276.
25. M. Janus, B. Tryba, M. Inagaki and A. W. Morawski, Appl. Catal., B, 2004, 52, 61–67.
26. H. S. Cheng, L. Chen, A. C. Cooper, X. W. Sha and G. P. Pez, Energy Environ. Sci., 2008, 1, 338–354.
27. W. D. Wang, C. G. Silva and J. L. Faria, Appl. Catal., B, 2007, 70, 470–478.
28. T. Tsumura, N. Kojitani, I. Izumi, N. Iwashita, M. Toyoda and M. Inagaki, J. Mater. Chem., 2002, 12, 1391–1396.
29. Y. C. Hsu, H. C. Lin, C. W. Lue, Y. T. Liao and C. M. Yang, Appl. Catal., B, 2009, 89, 309–314.
30. K. Woan, G. Pyrgiotakis and W. Sigmund, Adv. Mater., 2009, 21, 2233–2239.
31. J. Q. Geng, Z. Y. Jiang, Y. B. Wang and D. Yang, Scr. Mater., 2008, 59, 352–355.
32. D. H. Wang, D. W. Choi, J. Li, Z. G. Yang, Z. M. Nie, R. Kou, D. H. Hu, C. M. Wang, L. V. Saraf, J. G. Zhang, I. A. Aksay and J. Liu, ACS Nano, 2009, 3, 907–914.
33. W. J. Ren, Z. H. Ai, F. L. Jia, L. Z. Zhang, X. X. Fan and Z. G. Zou, Appl. Catal., B, 2007, 69, 138–144.
34. A. Mishra, S. Banerjee, S. K. Mohapatra, O. A. Graeve and M. Misra, Nanotechnology, 2008, 19, 445607.
35. S. Shanmugam, A. Gabashvili, D. S. Jacob, J. C. Yu and A. Gedanken, Chem. Mater., 2006, 18, 2275–2282.
36. N. S. Subramanian, R. P. Bernard, Y.-H. S. Hsu, C. D. Musick, K. Ogunde and J. N. Tilton, US Patent, 7 476 378 B2, 2009.
37. H. S. Holger, US Patent, 2488440, 1949.
38. M. Janus, M. Inagaki, B. Tryba, M. Toyoda and A. W. Morawski, Appl. Catal., B, 2006, 63, 272–276.
39. D. W. Lee, S. K. Ihm and K. H. Lee, Chem. Mater., 2005, 17, 4461–4467.
40. H. Zhang, X. Luo, J. Xu, B. Xiang and D. Yu, J. Phys. Chem. B, 2004, 108, 14866–14869.
41. R. Jin, Z. Wu, Y. Liu, B. Jiang and H. Wang, J. Hazard. Mater., 2009, 161, 42–48.
42. S. Ren, X. Zhao, L. Zhao, M. Yuan, Y. Yu, Y. Guo and Z. Wang, J. Solid State Chem., 2009, 182, 312–316.
43. J. Zhang, Z. Liu, B. Han, Z. Li, G. Yang, J. Li and J. Chen, J. Supercrit. Fluids, 2006, 36, 194–201.
44. W. Ren, Z. Ai, F. Jia, L. Zhang, X. Fan and Z. Zou, Appl. Catal., B, 2007, 69, 138–144.
45. C. Beck, T. Mallat, T. Bürgi and A. Baiker, J. Catal., 2001, 204, 428–439.
46. Y. H. Zhang, P. Xiao, X. Y. Zhou, D. W. Liu, B. B. Garcia and G. Z. Cao, J. Mater. Chem., 2009, 19, 948–953.
47. C. S. Kuo, Y. H. Tseng, C. H. Huang and Y. Y. Li, J. Mol. Catal. A: Chem., 2007, 270, 93–100.
48. F. Mei, C. Liu, L. Zhang, F. Ren, L. Zhou, W. K. Zhao and Y. L. Fang, J. Cryst. Growth, 2006, 292, 87–91.
49. L. Zhao, J. Yu and B. Cheng, J. Solid State Chem., 2005, 178, 1818–1824.
50. F. Z. Dong, W. R. Zhao and Z. B. Wu, Nanotechnology, 2008, 19, 365607.
51. Y. Huang, W. Ho, S. Lee, L. Zhang, G. Li and J. C. Yu, Langmuir, 2008, 24, 3510–3516.
52. C. Di Valentin, G. Pacchioni and A. Selloni, Chem. Mater., 2005, 17, 6656–6665.
53. F. Dong, H. Wang and Z. Wu, J. Phys. Chem. C, 2009, 113, 16717–16723.

---