Keywords

8.1 Preparation of Visible Light-Responsive TiO2 Photocatalysts by Chemical Doping Modification Methods

In order to improve the visible light activity of TiO2, many modification methods have been developed in recent years, like doping impurities, coupling semiconductors, dye sensitization, and so on. After the modification, the visible light-driven TiO2 can use the solar energy in dealing with the environmental pollution and new energy development. For instance, the dye-sensitized TiO2 has been widely used in preparing solar cells, owing to its strong visible light absorption ability. In 1991, Michael Grätzel and Brian O’Regan described a photovoltaic cell, created from low- to medium-purity materials through low-cost processes, which exhibited a commercially realistic energy-conversion efficiency [1]. The device is based on a 10-μm-thick, optically transparent film made with TiO2 particles having a few nanometers in size, coated with a monolayer of dye to sensitize the film for light harvesting. The overall light-to-electric energy conversion is 7.1–7.9% for simulated solar light and 12% for diffuse daylight. The large current densities and high stability of the solar cells based on dye-sensitized colloidal TiO2 films are making practical applications feasible. Since then, the record of energy conversion for the dye-sensitized TiO2-based solar cells is constantly improved year by year [2,3,4,5,6,7,8]. In 1998, the solar cell based on dye-sensitized mesoporous TiO2 films converts photons to electric current with a high yield of 33%, which is also achieved by Prof. Michael Grätzel [9]. At present, the photoelectric conversion efficiency of TiO2-based solar cells has stabilized to be more than 10%, but its cost of manufacturing is still high.

Some other modified visible light-responsible TiO2 especially the doping-modified TiO2 also attracted much attention and have been applied to organic pollutant photodegradation and water splitting reaction. Hence, in this section, we would like to introduce the modification methods related to dope trace impurity into TiO2, including chemical synthesis like high-temperature sintering in an atmosphere, wet chemical methods such as sol–gel processes and hydrothermal treatment, spray pyrolysis, and supercritical methods.

These chemical synthesis methods used in doing TiO2 are summarized in Table 8.1. As can be seen in Table 8.1, all the doped TiO2 synthesized by different chemical preparation methods have a high photocatalytic activity for the degradation of organic dye pollutants and water splitting. In addition to be divided in different synthesis method, the chemical doping modification on TiO2 can also be divided into metal doping and nonmetal doping. In metal doping method, a certain amount of metal ions such as Fe3+ [10,11,12,13,14], Cr3+ [15, 16], Ru2+ [17, 18], Ce4+ [19, 20], La3+ [21,22,23], V5+ [24,25,26], are introduced into TiO2 (Table 8.1), as the active “small oxide islands”, affecting the lifetime of the photo-formed electrons and holes their-transfer processes, thus affecting the photocatalytic activity of TiO2. Furthermore, the impurity induced by metal doping into the TiO2 could efficiently narrow its bandgap and extend the absorption edge into the visible light range. Many studies have demonstrated that metal doping could effectively improve the photocatalytic activity of TiO2 under the visible light irradiation [27,28,29,30]. However, metal doping also showed several drawbacks: thermal instability of doped TiO2, electron trapping by the metal centers, introduction of the electrons, and hole recombination centers [31]. It is worth mentioning that, in addition to the above traditional impurity metal ion doping, more and more research has focused on the investigation of Ti3+ self-doped TiO2 in recent years [32,33,34,35,36]. Recent research work has found that excessive Ti3+ would not easily introduce the electron and hole recombination centers in TiO2 [33].

Table 8.1 Doping-modified TiO2 photocatalysts prepared by different chemical methods
Full size table

On the other hand, nonmetal doping is another technology to modify TiO2, which could achieve the substitution of lattice oxygen by nonmetal elements [31, 37,38,39]. Since a work was investigated by Asahi et al. [40] in 2001 in which they reported that nitrogen doping could enhance the photocatalytic activity of TiO2 for the photodegradation of methylene blue and gaseous acetaldehyde in the visible light irradiation, though the photocatalytic activity in UV light regions decreases. Since then, many researchers have reported about various nonmetal-doped TiO2, such as N [41, 42], B [43,44,45,46], C [21, 47,48,49], F [50,51,52,53,54], S [55,56,57,58,59], and P [60]. Although nonmetal doping modification could change the band structure of TiO2 and affect the transfer of electrons and holes, the origin of its visible light photoactivity is still in debate [31], especially the photocatalytic mechanism of nitrogen doping. Recent experimental and theoretical studies suggest that the N doping does not cause the narrowing of the band gap of the TiO2 but the formation of localized midgap states above the valance band of TiO2 which is the reason for its enhancing visible light responsiveness and photoactivity [31, 61].

In this section, we mainly introduce and highlight the chemical-synthesized visible light-responsible TiO2 photocatalysis with doping modification, including metal doping, nonmetal doping, and co-doping modification. The influencing factors on doping modification, the research development of doping and photocatalytic mechanism, and novel investigation of synergistic effect between different elements are also discussed in this section.

8.1.1 Metal Doping Modification

As can be seen from the reaction mechanism of TiO2 photocatalysis, it can be concluded that an increase in the number of photo-formed electrons and holes participating in the photoredox reaction is the key factor to improve the visible light-responsive TiO2 photocatalyst’s performance. In order to achieve the substantive enhancement of the quantum yield of photocatalytic reaction for TiO2 in the visible light irradiation, two issues must be overcome. One issue is how to produce much more photo-formed electrons and holes; another issue is how to improve the separation efficiency of electrons and holes. It is well known that appropriate amount of transition metal ions doped into TiO2 can introduce electron capture centers and change the crystallinity of TiO2 and then produce some defects, which results in an decrease in photo-formed electron and hole recombination centers [11, 27,28,29,30, 33, 34, 102]. Thus, metallic ion doping is recognized as an effective modification method for improving the reactivity of the visible light-responsive TiO2.

The methods using metallic ions selected for appropriate doping into TiO2 can be divided into three types. The first type is the transition metal ions. Most early, Choi et al. [103] studied the photoactivity of TiO2 doped with 21 transition metal ions by using the model reaction of photocatalytic oxidation of chloroform and photocatalytic reduction of carbon tetrachloride. The researchers discovered that the Fe3+, Mo5+, Ru3+, Os3+, Re5+, V4+, and Rn3+ ion doping modification is beneficial to the enhancement of photodegradation of chloroform for TiO2. Recently, Yan et al. [104] synthesized the TiO2 nanoparticles doped with different content of cerium ion by a sol–gel method. The Ce-doped TiO2 act as the capture of photo-formed holes and decrease the recombination of photo-generated electrons and holes, leading to better visible light absorption and photocatalytic degradation of methylene blue than pure TiO2 (Fig. 8.1). Zhang and coworkers have also done many works on the transition of metal-doped TiO2 and its application to various photocatalytic reactions [14, 15, 19, 20, 24, 29, 105,106,107,108,109,110]. Iron ion-doped anatase TiO2 were prepared by hydrothermal hydrolysis and crystallized in octanol–water solution [105]. The results of photodegradation of active yellow XRG dye indicated that the amount of doped iron ion plays a significant role in affecting the photocatalytic activity and iron doped with optimum content enhances the photocatalytic activity, especially under visible light irradiation. When the Fe3+-doped TiO2 photocatalysts were prepared by combining sol–gel method with hydrothermal treatment [106], Fe was found to exist in trivalent ionic state and substitute Ti4+ in TiO2, and its concentration was found to decrease from the surface to the deep bulk of TiO2. As a result, the synthetic method of Fe doping in TiO2 has a significant influence on the doping structures and concentration. In addition to iron, chromium and vanadium are also widely used as dopant elements into TiO2. Due to the excitation of 3d electron of Cr3+ to the conduction band (CB) of TiO2, Cr–TiO2 always shows a good ability for absorbing the visible light to induce the photodegradation of XRG [15]. V4+ ions are also successfully incorporated into TiO2 by flame spray pyrolysis (FSP) technique [24], sol–gel [26, 111], and other chemical methods [25]. V-doping into TiO2 leads to change the bandgap of TiO2, leading to an extension of the absorption regions to visible light region, resulting in the improvement of the visible light-driven photocatalytic activity of TiO2 [26].

Fig. 8.1
figure 1

UV–vis absorption spectra of Ce–TiO2 samples with different Ce ion doping concentrations (a). Curves of methylene blue degradation constant by Ce–TiO2 samples with different Ce ion doping concentrations (b) (Reprinted with permission from ref. [104]. Copyright 2015, Elsevier)

Full size image

The second type is the rare-earth metal doping, and the lanthanide-doped TiO2 have been accounted for the majority [21, 22, 112,113,114]. Sun et al. [112] investigated the effects of substitutional La doped on the electronic structures and photocatalytic activity of TiO2 by the density function theory (DFT) calculation method. Their calculation results indicated that the enhanced absorption in the visible light region for La–TiO2 was attributed to the adsorptive of La doping rather than the substitutional La. Differently, Anandan et al. [115] believed that the rapid mineralization of monocrotophos over La-doped TiO2 under the light irradiation could be associated with the suppression of the electrons and holes recombination by La3+ doped into TiO2 and generation of more number of •OH radicals by oxidation of holes. Recently, Zhang and coworkers have studied some other lanthanide metals such as Eu-, Yb-, and Sm-doped TiO2 and their photocatalytic activities under the visible light irradiation [116,117,118]. Samarium-doped TiO2 (Sm–TiO2) was successfully prepared via a chemical coprecipitation method. The curve in the Sm3d XPS spectrum was found to fit into two peaks [118]. The peak at 1084.3 eV corresponds to the bond of Sm–O. And another peak at 1082.2 eV corresponds to the bond of Sm–O–Ti. Although the ionic radius of Sm3+ (1.08 Å) is bigger than the ionic radius of Ti4+ (0.68 Å) and the Sm3+ ions cannot enter into the lattice of TiO2, however, Ti4+ ions may enter into the lattice of the Sm2O3 leading a change in the electronic field of Sm3+ and an increase in the electron density and decrease in the binding energy of Sm3+. Well-ordered mesoporous TiO2 doped with ytterbium was also successfully synthesized by an evaporation-induced self-assembly process [116]. The Yb dopant was beneficial in stabilizing the mesoporous structure and reduces the recombination of photo-generated electrons and holes, being beneficial to its visible light-driven photoactivity. Europium-doped TiO2 was synthesized by the precipitation–peptization method and used as the photocatalyst to degrade the salicylic acid [117]. The results showed that the doping of Eu was beneficial to the adsorption of salicylic acid and the separation of photo-formed holes and electrons.

In addition to the transition metal ion doping and rare-earth metal ion doping, the studies on some other metal ion-doped TiO2 can be seen. Stannum doping TiO2 and Ti (III) self-doping TiO2 were also reported. The dopant Sn4+ substituted Ti4+ in the lattice of TiO2, which was reflected in the lattice expansion in both a- and c-direction and change in the binding energy [119]. Different from other metal ion doping, the doping of Sn4+ in anatase TiO2 would result in a blue shift of absorption edge and enhance the amount of surface hydroxyl and oxygen vacancies in the UV light region. On the other hand, Sn doping has also been demonstrated as an effective modification method to enhance the visible light response of TiO2 [120]. Tin would improve the photocatalytic activity of TiO2 by enhancing the separation rate of photo-generated electrons and holes on the surface of TiO2. Due to the Fermi level of SnO2 lower than that of TiO2, the photo-generated electrons easily transfer from TiO2 to SnO2, resulting in a reduction of the number of photo-generated on the surface of TiO2. In recent years, Ti3+-doped TiO2 has attracted much interest, since it has been demonstrated to exhibit visible light absorption [121, 122]. Sasikala et al. [122] have found that the surface Ti3+ and oxygen vacancies may be responsible for the enhanced visible light absorption of the TiO2–SnO2 composite. However, the surface Ti3+ and oxygen defects on the TiO2 are usually not stable enough in air, since the surface Ti3+ is easily oxidized into Ti4+ by the dissolved oxygen in water [32, 123]. Most research has focused on Ti3+ self-doped TiO2, which exhibits better chemical stability and is active for photocatalytic activity. Zuo et al. [32] have reported a one-step calcination method to synthesize Ti3+ self-doped TiO2 having high stability and found that it exhibits improved visible light absorption and efficient photocatalytic hydrogen production capacity from water under visible light irradiation (Fig. 8.2, top). It has been also reported a vacuum activation method for modifying P25 to obtain Ti3+ self-doped TiO2 with high stability, visible light absorption, and photocatalytic activity under visible light irradiation (Fig. 8.2, bottom) [36]. In addition to the vacuum activation method, a simple one-step solvothermal method with low-cost NaBH4 added as a reductant was also reported to successfully synthesize a series of TiO2 catalysts self-doped with Ti3+ which also exhibited strong visible light absorption and enhanced photocatalytic activity [33].

Fig. 8.2
figure 2

The top figure is the time course of evolved H2 under visible light (>400 nm) irradiation and the UV−vis diffuse reflectance spectra for commercial anatase TiO2 (solid line) and Ti3+ self-doped TiO2 (dash line) (the top figures). (Reprinted with permission from ref. [32]. Copyright 2010, American Chemical Society) The bottom figure is (A) photooxidation of 5 mg L−1 MO before (a) and after (b) visible light (>420 nm) irradiation for 3 h by the sample after vacuum activation for 180 min. (B) Photooxidation of 20 mg·L−1 phenol under visible light (>420 nm) irradiation for 10 h. (C) UV–vis diffuse reflectance spectra for pure P25 and the vacuum-activated samples (Reprinted with permission from ref. [36]. Copyright 2011, Royal Society of Chemistry)

Full size image

8.1.2 Nonmetal Doping Modification

Although the metal doping modification could promote the absorption of TiO2 for the visible light, the metallic ions also would induce the poor thermostability of TiO2 and introduce some recombination centers with an excess doping concentration. Since in 2001 Asahi et al. [50] found that the nitrogen-doped TiO2 exhibited the visible light absorption and photocatalytic activity, the study on nonmetal doping modification in TiO2 has been a research hotspot.

Taking into consideration of the poor thermostability caused by the metal doping and the high cost in modification, the increasing number of nonmetal elements is used as the dopant to modify the bandgap of TiO2 in recent years. Nonmetal doping mainly consists of the N, C, F, B, and other elements having the similar atomic radius with O atom. Among them, nitrogen has attracted much attention and been widely studied. In some of recent work [51, 52], a new approach to synthesize the N-doped TiO2 nanocrystals was developed, and the relationship of doped nitrogen species to visible light photoactivity was investigated. It was found that the nitrogen introduced into TiO2 is beneficial to the visible light photodegradation of 2,4-dichlorophen and the nitrogen species chemically adsorbed on catalyst surface are harmful to the photoactivity, as shown in Fig. 8.3 [42]. Moreover, the nitrogen source also plays an important role in the N doping effect. Urea, triethylamine, thiourea, and hydrazine hydrate all could be used as the nitrogen source to prepare the nitrogen-doped TiO2 by a microemulsion–hydrothermal method [41]. Among the above nitrogen sources, triethylamine was a more effective nitrogen source, which was used to synthesize the N-doped TiO2 with the highest visible light photocatalytic degradation of rhodamine B (Fig. 8.3). Yates et al. [124] used NH3 gas as the nitrogen source to prepare the N-doped TiO2 with enhanced visible light photoactivity by a calcination method. However, the TiO2 treated by the N2 had decreased photoactivity due to the broadening of the bandgap of TiO2.

Fig. 8.3
figure 3

The left figure is the nitrogen forms doped in TiO2 (Reprinted with permission from ref. [42]. Copyright 2009, Elsevier); the right figure is the photocatalytic activities of TiO2 prepared from different nitrogen sources with the optimal doping value (Reprinted with permission from ref. [41]. Copyright 2007, American Chemical Society)

Full size image

In addition to nitrogen, carbon is also demonstrated to be an effective doping element to modify the visible light absorption and photoactivity of TiO2. Nagaveni et al. [125] successfully prepared the C-doped TiO2 by using a sol–gel method, which exhibited a high photodegradation of methylene blue under the visible and UV light irradiation. Kamisaka et al. [126] investigated the affection of C doping on the structure and optical property of TiO2 by the density functional theory (DFT) calculation method. They assumed the carbon atom could substitute four sites of titanium and oxygen to obtain four corresponding C-doping structures. The DFT calculation results indicated that the substitution of C for Ti could not cause the visible light response of TiO2 because of the formation of titanate anion. On the contrary, the substitution of C for O was beneficial to the visible light absorption of TiO2 and did not change its crystal structure. Recently, Bai et al. [127] prepared monodisperse, carbon-doped rutile TiO2 single crystal with exposed (110) facets, which possessed hierarchical structure and highly efficient H2 generation activity. Yu et al. [128] also fabricated novel carbon self-doped TiO2 sheets with exposed (001) facets by a hydrothermal treatment. The C-doped TiO2 sheets had enhanced absorption in the whole visible light region and a significant redshift at the absorption edges. It was also showing a high photocatalytic degradation of methylene blue under the visible light irradiation. Lin et al. [129] prepared a visible light-driven C-doped mesoporous TiO2 films by a sol–gel method combined with a hydrothermal treatment. The C-doped TiO2 film had high surface area and excellent photodegradation of dye reactive brilliant Red X-3B in the UV and visible light irradiation.

Fluorine-doped TiO2 as one of the widespread modified methods has attracted much more attention in recent years, but conventional F-doped TiO2 is hard to effectively achieve the enhanced UV and visible light photoactivity, and its photocatalytic mechanism still remains controversial [130,131,132]. In previous work, NH4F was used as a hydrophobic modifier, and isopropanol was used as the solvent to prepare super-hydrophobic mesoporous MCF loaded with fluorinated TiO2 nanoparticles [50], through a simple one-step solvothermal method (Fig. 8.4, top). The prepared catalyst exhibited permanent and excellent super-hydrophobic property, high adsorption capacity, and photocatalytic activity for rhodamine B degradation (Fig. 8.4, bottom). However, through the solvothermal method, the F ion could only adsorb on the surface of the catalyst but not be introduced into TiO2 lattice. Many researchers also have reported the F-doped TiO2 having high photocatalytic activity [50, 132]. And the standard substitutional of TiO2 with F is similar to F-doped SnO2, with generation of impurity levels close to the conduction band [133,134,135]. However, its electronic structures and photocatalytic mechanism are still unclear. It has been demonstrated that F substitution for lattice O could not introduce impurity level inside TiO2 band gap, as well as shift its absorption edge into visible region by the first-principles calculation [136]. That is because of the absence consideration of F substitution for oxygen vacancy during the calculation process. Some studies have reported that F substitution for oxygen vacancy could introduce acceptor impurity level inside ZnO or SnO2 bandgap [137, 138]. Hence, it was concluded that the achievement of abundant fluorine substitution for lattice oxygen vacancy played a very important role in the diminishing of vacancy-induced recombination sites and the introduction of impurity level inside the TiO2 bandgap [53].

Fig. 8.4
figure 4

The top figure is the schematic diagram of the fluorination reaction that occurred in the pore channels of MCF; the bottom figure is adsorption capacities of RhB (20 mg/L, catalyst concentration: 0.25 g/L) on different samples (a); visible light photocatalytic activities of different samples (b) (Reprinted with permission from ref. [50]. Copyright 2012, Elsevier)

Full size image

Compared with the abovementioned nonmetal elements, the study on boron doping in TiO2 is relatively rare in recent years. In some studies, it was suggested that the boron doping in TiO2 could lead to a redshift of the absorption band of TiO2 to the visible light region, because of the overlapping of the impurity levels caused by boron with the 2p electronic states of oxygen [139, 140]. By contrast, some other studies reported that the boron incorporation into TiO2 lattice could induce a blue shift rather than redshift due to the decrease of the crystal size [141, 142]. The quantization effect of the crystal size would result into a blue shift of absorption band to the UV light region. Even so, the boron has still been used as an important co-dopant together with other nonmetals for modification of TiO2. Hence, more and more researchers began to study the co-doping modification of boron with other elements in TiO2 [70, 77, 143, 144].

8.1.3 Co-doping Modification

In order to further improve the photocatalytic activity of TiO2 and make effective use of sunlight in visible light region, multicomponent modified TiO2 such as co-doped with different elements has attracted much more attention in the photocatalysis fields. Some studies found that co-doped TiO2 with appropriate elements could exhibit a much higher photocatalytic activity than single-doped catalyst, because of the existence of synergistic effect of doped elements which could promote the visible light absorption and facilitate the separation efficiency of photo-formed electrons and holes [11, 25, 35, 76, 77, 112, 116, 145, 146]. Herein, we will focus on summary of the research status of the co-doping modification on TiO2 photocatalyst in recent years, the modification mechanism of co-doping, and the synergistic effect existing in the co-doped TiO2.

According to the band theory of semiconductors, the conduction band of TiO2 semiconductors is mainly determined by the Ti3d orbital energy level, and the valence band mainly depends on the energy level of O2p orbital. Compared with the O2p orbital, nonmetal elements such as N, C, S, and P have the 2p orbital with relatively high energy levels; hence, it is possible to enhance the visible light photocatalytic activity of TiO2 by the doping N, C, S, P, and other nonmetal elements, due to the increase of electric potential of valence band by the partial substitution of impurity dopants for lattice oxygen.

Compared with other nonmetallic element co-doping modification, nitrogen co-doped with other nonmetal modification is very important, owing to the significant synergistic effect for visible light response. Cong et al. [147] proposed that the energy level of N doping could connect with the states of C doping and facilitate the overlap of C1s and N1s with the VB states of TiO2, as shown in Fig. 8.5. The co-doping of N and C further narrowed the band gap of TiO2 and improved the visible light photocatalytic activity of TiO2 [147]. Meanwhile, boron has also been used as an important co-dopant together with nitrogen for co-doping modification of TiO2 [69, 76, 77, 99, 143]. In et al. [99] and Liu et al. [76] proposed that B and N co-doped TiO2 show a high UV and visible light photocatalytic activity, probably due to the existence of a synergistic effect between boron and nitrogen by the formation of Ti–B–N structure at the catalyst surface. However, there is still no detailed illustration of the B–N synergistic effect and its effect on the photocatalytic activity of TiO2. Xing et al. [77] have illustrated the exact role of synergistic effect in optical absorbance and photocatalytic activity of B and N co-doped TiO2. Various co-doped TiO2 was systematically prepared by using the double hydrothermal method. Different new bonds were determined to form on the surface of TiO2 when the order of boron and nitrogen addition was changed, and they could significantly affect the photoactivities of the materials. The abovementioned experimental results are further illustrated by the DFT calculation. Gombac et al. [148] found that the surface N doping did not appreciably modify the TiO2 structures and texture, and boron incorporation in TiO2 indeed inhibited the TiO2 crystallite growth and increased the surface area of TiO2. Only when B was present in excess with respect to N, a remarkable photoactivity improvement could be obtained. DFT calculation method was used to interpret the observed behavior. And the B in molar excess with respect to N led to the generation of Ti3+ sites, which might further induce the generation of reactive superoxide species. Different from the surface bond structures and the reactive Ti3+ sites induced by B and N co-doping, Czoska et al. [70] considered that the lattice center (labeled [NOB]·) based on the presence of interstitial N and B atoms both bound to the same lattice oxygen ion could introduce an energy level near the edge of VB of TiO2. [NOB]· can easily trap one electron to produce a diamagnetic center at about 0.4 eV above the top of the VB (Fig. 8.6), which can contribute to the visible light photoactivity.

Fig. 8.5
figure 5

Mechanism for photocatalytic degradation of organic pollutants over C–N–TiO2 photocatalyst under visible light irradiation (Reprinted with permission from ref. [31]. Copyright 2010, Royal Society of Chemistry; and reprinted with permission from ref. [147]. Copyright 2006, Chemical Society of Japan)

Full size image
Fig. 8.6
figure 6

Schematic representation of interplay between the [NOB]˙ and [NiO]˙ centers in N–B co-doped TiO2 (Reprinted with permission from ref. [70]. Copyright 2011, Royal Society of Chemistry)

Full size image

Recently, nitrogen and sulfur co-doped TiO2 was successfully immobilized on the surface of nitride Ti substrate, which exhibited high photodegradation of methylene blue in the visible light irradiation [149]. It was estimated that N and S co-doping in the anodic TiO2 narrowed the band gap of TiO2 and enhanced its visible light absorption and photocatalytic activity. In order to eliminate the recombination centers induced by the nonmetal doping in TiO2 nanoparticles, Yang et al. [150] synthesized the fluorine and sulfur co-doped mesoporous TiO2. The ability to control the morphology and chemical composition of the mesoporous TiO2 could be beneficial to improve the light-harvesting capacity and decrease the recombination centers. The F and S co-doping in TiO2 can redshift the threshold of the TiO2 absorption into the visible light region and improve the photocatalytic efficiency for the degradation of organic pollutants. As well as our previous work, some nonmetal co-doped TiO2 are successfully synthesized for the photodegradation of organic dyes under the visible light irradiation. N- and F co-doped TiO2 microspheres were prepared by ethanol solvothermal method [151]. It was found that the co-doped catalyst with mesoporous structure exhibited a significant synergistic effect of N and F doping, which led to the high photocatalytic activity of the degradation of AO7 in the visible light irradiation. Moreover, carbon and boron co-doped TiO2 were also synthesized firstly by the gel–hydrothermal method [144], that is, prepared through sol–gel process followed by hydrothermal in the glucose solution. The experimental results indicated that the coke carbon generated on the co-doped catalyst surface acted as a photosensitizer and had a photosensitization effect under the visible light. And the boron doping could effectively narrow the band gap of TiO2, which induced the easier transition of photo-formed electrons from the boron dopant level to the Ti3+ level. The synergistic effect of B and C is responsible for its excellent visible light photocatalytic activity.

Either metal doping or nonmetal doping, they both will change the electric structure of TiO2 and create a new doping level inside the band gap of TiO2. Because of the different position of doping level in TiO2, it is generally considered that the doping level induced by the substitution of metallic ions for Ti4+ is close to the CB of TiO2. And the impurity level caused by the doping of nonmetallic ions into the oxygen vacancy sites is nearby the VB of TiO2. These doping levels are located inside the band gap of TiO2, which can accept the photo-formed electrons from the VB or absorb the photos with longer wavelength and extend the range of absorption spectrum of TiO2. The synergistic effect between the metal and nonmetal could promote the separation of electrons and holes, resulting into the improvement of the visible light photocatalytic activity of TiO2.

Generally speaking, the synergistic effect between metal and nonmetal is mainly shown as follows: the nonmetallic ion doping can enhance the absorption of TiO2 in the visible light region, and the metallic ion doping can introduce traps for electrons and decrease the recombination of electrons and holes. Vanadium and nitrogen co-doped TiO2 was synthesized by the sol–gel method, and the catalyst showed a high visible light photocatalytic activity for the degradation of RhB [111]. The visible light absorption efficiency of V–N co-doped TiO2 was better than the V or N single-doped TiO2, because of the effective narrowing of the band gap induced by the simultaneous incorporation of V and N in TiO2 lattice, as shown in Fig. 8.7. The energy levels inside the TiO2 band gap can act as traps for photo-formed holes and electrons thus decrease the recombination between photo-formed charges. The narrowed band gap and enhanced charge separation exhibit synergistic effect to improve the visible light photoactivity of the co-doped TiO2. Wei et al. [152] considered that the synergistic effect between nitrogen and lanthanum in La and N co-doped TiO2 was responsible for the high photocatalytic activity. The N doping decreased the band gap of TiO2 and increased the absorption intensity of TiO2 in visible light region. And the La3+ doping could not only increase the surface area of TiO2 but also restrain the recombination of electrons and holes, due to the electron capturing capacity of La3+. In order to minimize the role of metallic ions as recombination centers, Kim and coworkers synthesized boron and iron co-doped TiO2 using a modified sol–gel method, and the presence of boron and iron caused a redshift in the absorption band of TiO2 [153]. Cong et al. [79] also have successfully prepared the nanosized TiO2 catalyst co-doped with nitrogen and iron, which exhibited a higher photocatalytic activity than the single doping catalyst under the visible light irradiation. Xing et al. [21] prepared the highly dispersed carbon and lanthanum co-doped TiO2 crystals with exposed (001) facets by using glucose as the carbon-doping source and (001)-facet-controlling agent through a simple one-step hydrothermal method. The C doping was responsible for the strong absorption in the visible light region, and the La doping acted as the electron capture center. The synergistic effect of C and La is the reason for the high visible light photocatalytic activity of the co-doped TiO2.

Fig. 8.7
figure 7

Schematic diagram for photocatalytic degradation of RhB dye over V2–N4-co-doped TiO2 photocatalyst under light irradiation. And the comparison of photocatalytic degradation of RhB under visible light source in the presence of undoped, V2-doped, N4-doped, and V2–N4-co-doped TiO2 powders (Reprinted with permission from ref. [111]. Copyright 2012, Elsevier)

Full size image

However, in addition to the abovementioned synergistic effect, there are also some other forms of synergistic effects between metal and nonmetal, such as the increase of the surface active species including the hydroxyl groups and hydrogen peroxide in TiO2. The role of surface hydroxyl groups and hydrogen peroxide is mainly including two aspects: one is involving in the reaction with the photo-formed holes to generate the hydroxyl radicals; another function is changing the adsorption forms of reactant and acting as the active center to influence the photocatalytic reaction of the reactant molecules. Wei and coworkers found that the content of hydroxyl groups on the surface of sulfur and iron co-doped TiO2 was increasing, as can be seen in the XPS characterization [154], which is beneficial to its visible light photocatalytic activity. The difference of surface –OH between pure TiO2 and single-doped TiO2 is very small, while the amount of surface –OH on Fe and S co-doped TiO2 is much higher than single-doped catalyst, resulting from the synergistic effect between Fe and S. The increase of surface –OH is in favor of the enhancing of the photoactivity. Gomathi Devi et al. [155] prepared the Ag and nitrogen co-doped TiO2 by grinding sol–gel titania with urea followed by a photoreduction process. The as-prepared Ag–TiO2–xNx exhibited much higher visible light photocatalytic activity than the single nitrogen doped TiO2, which can be accounted to the synergistic effect of Ag loading with N doping. Strongly interacting electron accepting species of hydrogen peroxide at the catalyst surface are acting as the surface states enabling inelastic transfer of electrons from the CB to the oxidizing species. Additionally, the synergistic effect also can be shown as the electrons transfer between the co-dopant states. Hoang et al. [35] reported a synergistic effect involving Ti3+ and nitrogen in TiO2 nanowire arrays, which exhibited an enhanced water photooxidation performance in the visible light irradiation. The authors proposed a reversible electron transfer between the paramagnetic bulk species of N (Nb) and Ti3+ centers forming the diamagnetic bulk species of Nb − and Ti4+. That means the existence of interaction between Ti3+ and N in TiO2 (Fig. 8.8). The lower oxidation states of the substitutional N in the co-doped TiO2 might be resulted from the electron transfer form Ti3+. Because of the Columbic repulsion, the lower oxidation states of N in co-doped TiO2 have higher energy than that of the N single-doped TiO2, thus enabling excitation with photons of longer wavelengths.

Fig. 8.8
figure 8

Proposed mechanism for the interaction between Ti3+ and substitutional N (Reprinted with permission from ref. [35]. Copyright 2012, American Chemical Society)

Full size image