Introduction

Recently, clean water shortage is considered as a severe problem due to the rapid growth of industries and environmental pollution [1,2,3]. The pollutants in water include organic, inorganic, radioactive, and biological pollutants, which are harmful to human health. In recent years, the significant increase in the quantity and the variety of organic pollutants is a worrisome issue. Dyes, which are widely used in industry, are one of the most noticeable and harmful organic pollutants found in wastewater [4, 5]. Besides that, the existence of polycyclic aromatic hydrocarbons and phenolic compounds in the aquatic environment is a noticeable issue due to their mutagenic, carcinogenic potential [6]. Several water treatment methods have been studied and commercialized, including filtration, adsorption [7], coagulation/precipitation [8, 9], Fenton oxidation [10, 11], and biodegradation [12, 13]. Though these methods are economical and safe, their effectiveness on removal of organic pollutants is limited [14]. Advanced oxidation processes (AOPs) have appeared as groundbreaking technologies for water treatment, which realize the elimination of organic compounds resistant to the conventional treatment methods [15,16,17]. Heterogeneous photocatalysis is one of the most effective AOPs because of its low cost with the utilization of solar energy and its efficacy to convert a wide variety of recalcitrant organic pollutants into non-toxic compounds or mineralize them into carbon dioxide and water [18, 19].

TiO2 is a well-known photocatalyst due to its superior properties with a wide range of applications [17, 20,21,22,23]; however, its broad band gap, short lifetime, and low ability to adsorb hydrophobic pollutants limit the application of TiO2 in photocatalysis [23, 24]. These shortfalls can be indicated by paring TiO2 with other materials such as metals, e.g., Cu, Fe, Ag; non-metal, e.g., C, N, S; or other semiconductors, e.g., In2S3, CdS [25,26,27,28,29,30,31]. In particular, pairing TiO2 with graphene has arisen as one of the most effective methods to solve the pitfalls of TiO2, since it: (i) enhances the surface area of the catalyst, resulting in the introduction of more active site for the redox process; (ii) contributes to the extension of absorption edge to the visible-light region leading to the improvement in solar light utilization; (iii) hinders the recombination of photoinduced excitons as graphene works as an electron sink; and (iv) improves the ability of the catalyst to absorb organic pollutants due to the hydrophobic nature of graphene [32,33,34,35,36,37,38,39,40].

In order to improve the photocatalytic activity of TiO2/graphene photocatalysts, many synthetic and modification strategies have been studied to improve the uniformity or tailor the morphology of the photocatalysts, crystal size, and shape and to control the crystal facets of TiO2, resulting in adjustments in the charge carrier dynamics between TiO2 and graphene [41,42,43,44,45,46,47]. Consequently, many mechanisms for the photocatalytic activity improvement have been proposed. Addition of functional groups to graphene nanoplatelets to engineer the morphology of TiO2, i.e., shape and crystal facet, has been considered as a promising strategy to improve photocatalytic activity of the photocatalyst [41,42,43]. Sordello et al. reported that the migration of photogenerated electrons to low energy (101) facets and holes to high energy (001), (100), or (010) facets, when graphene bonded with (101) facets leading to the reduction of charge carrier recombination was the main mechanism for photocatalytic activity improvement of the photocatalyst [41, 42]. Some studies attempt to control the nanomaterial morphology by direct deposition of TiO2 with a specified morphology into graphene layers to form hierarchical 1D/2D, 2D/2D, or 3D/2D TiO2/graphene composites [44,45,46]. Among these, 2D/2D structure showed the best performance since the high-contact interface between TiO2 and graphene increased the injection speed of electrons from TiO2 to graphene and thus improved the formation of superoxide and hydroxyl radicals under UV irradiation [44]. Zhang et al. synthesized graphene/TiO2 continuous fiber with the aim of increasing broadband photocatalytic activity of the material, and they found that high adsorption ability, charge carrier separation, and transportation rate, which enhanced the formation of hole and hydroxyl radicals, were mainly contributed to the improvement in photocatalytic activity of the material [47]. However, most of the synthetic processes have utilized graphene oxide (GO) as a starting material, while the pitfalls of the usage of GO, e.g., defect-rich graphene, uncontrolled, and non-uniform growth of TiO2, have been figured out [48,49,50,51,52].

Recently, we have developed a novel GO-free route to hybridize TiO2 and graphene at a nanoscale by utilizing a graphene dispersion in titanium alkoxide as a starting material [48]. The advantage of this method is the ability of titanium alkoxide to exfoliate graphite into few-layer and defect-less graphene sheets. With nanolayer morphology and advanced optical properties, the TiO2/graphene nanocomposite exhibited an excellent visible-light photocatalytic activity, i.e., 15 times and 5 times higher activity compared to TiO2-P25 and the conventional TiO2/reduced graphene oxide (noted as TiO2/rGO) nanocomposite, respectively, in the photodegradation of methylene blue. Hence, in this paper, we aim to clarify the mechanism of the excellent photocatalytic activity of the TiO2/graphene nanocomposite via radical scavenger tests, an electron paramagnetic resonance measurement, and a photoluminescence measurement in the comparative way of TiO2-P25 and TiO2/rGO nanocomposite.

Methods

Materials

TiO2-P25, a commercial photocatalyst (Sigma Aldrich), was used after calcination at 400 °C for 2 h. A conventional TiO2/reduced graphene oxide nanocomposite denoted as TiO2/rGO was prepared by a hydrothermal method from GO dispersed in methanol and titanium tetra-n-butoxide (Ti(OnBu)4) [53]. These two samples were used as reference samples. Methylene blue (MB) (Kanto Chemical, ≥ 98.5%), phenol (Wako Pure Chemical, ≥ 99%), and naphthalene (Sigma Aldrich, ≥ 99%) were used as model pollutants in photocatalytic test. Tert-butyl alcohol (> 99%), ammonium oxalate (Wako Pure Chemical), and benzoquinone (Tokyo Chemical Industry) were utilized as radical scavengers for the investigation of active species.

Synthesis of TiO2/graphene nanocomposite

The synthesis of the TiO2/graphene nanocomposite was reported in our previous research [48]. In brief, a graphene dispersion in Ti(OnBu)4 was prepared by chemical exfoliation of graphite. 44.0 mg of graphite was sonicated in 5.5 mL of Ti(OnBu)4 under N2 atmosphere for 4 h at 60 °C and 43 kHz. A graphene dispersion in Ti(OnBu)4 at 0.18 mg mL−1 was obtained as a supernatant. Aqueous NH3 mixed with benzylamine at a volume ratio of 1:3 (total 5.0 mL), utilized as a catalyst for the sol–gel reaction, was dropwise added into 5.0 mL of the graphene dispersion in Ti(OnBu)4 under vigorous magnetic stirring and N2 atmosphere at room temperature. The grayish gel obtained after 2 h of the sol–gel reaction was washed with anhydrous hexane and acetone for three times alternatively, before being fully dried in vacuum for 12 h at 70 °C.

Photocatalytic test

The photocatalytic activity was evaluated by the photodegradation of aromatic compounds, i.e., naphthalene and phenol, in aqueous media at room temperature. A photocatalytic test was performed as follows. A specified amount of a catalyst was dispersed individually in a naphthalene (25 mg L−1) or phenol (100 mg L−1) solution prepared in the mixture of DI water and acetonitrile (v/v = 1:1). The dispersion was kept stirred in dark for 6 h to ensure the adsorption/desorption equilibrium. Then, the dispersion was irradiated for 4 h using a light irradiator (MAX-303, Asahi spectra) equipped with a 300 W xenon short arc lamp and a 422-nm longpass filter. The reaction was maintained at room temperature under the pressure of 0.1 MPa with O2 bubbling. The concentration of aromatic compounds was determined by gas chromatography (GC) using an internal standard. The activity of the catalyst was evaluated by the amount of a pollutant degraded per gram of the catalyst per hour.

Radical scavenger tests

Active species produced in a catalyst under the visible-light irradiation were investigated by adding different types of scavengers via the photodegradation of MB. In brief, 25 mg of a catalyst was dispersed in 25 mL of an aqueous MB solution at 100 mg L−1. After stirring in dark for 12 h to achieve the adsorption/desorption equilibrium, a specified amount of a scavenger was added into the dispersion before irradiation (24 μL of tert-butyl alcohol, 7.8 mg of ammonium oxalate, or 0.3 mg of benzoquinone). During the reaction, 2.0 mL of the MB solution was sampled to track the consumption of MB every 40 min.

Electron paramagnetic resonance measurement

Electron paramagnetic resonance (EPR) experiments were performed on a Bruker EMX, CW-EPR, at 77 K. Light-induced EPR experiments were performed by in situ irradiation of the sample using UV light (ORIEL Xe lamp 1000W). UV light was used in order to fully photoactivate TiO2/rGO and the TiO2/graphene nanocomposite since the absorption edge of TiO2/rGO was recorded at 396 nm and the ones of the TiO2/graphene nanocomposite were 360 and 578 nm [48].

Photoluminescence measurement

A photoluminescence (PL) emission measurement was taken using a spectrofluorometer (FP-6500, JASCO) equipped with a powder holder accessory. The excitation wavelength was set at 300 nm, and the emission spectra were collected from 360 to 520 nm at the scanning rate of 50 nm min−1, the excitation and emission bandwidths of 5 nm, and the response time of 2 s. The charge carrier dynamics of a photocatalyst was investigated by a time-resolved photoluminescence (TRPL) measurement. A powder sample was dispersed in ethanol and then casted onto a glass slide. The glass slide was dried at room temperature overnight. A picosecond laser pulse at 374 nm was utilized for excitation. The PL spectrum was recorded on an imaging spectrograph (Hamamatsu Photonics C5094). The transient PL signal was measured by a single-photon-counting mode using a streak camera (Hamamatsu Photonics C5680).

Results and discussion

Photocatalytic activity of the TiO2/graphene nanocomposite

In our previous report, we performed the GO-free route to synthesize the TiO2/graphene nanocomposite featured with many advantages for an excellent visible-light photocatalyst (Fig. 1) including (i) sheet-like morphology in which TiO2 anatase (101) nanolayer thinly and uniformly covered graphene sheets (Figure S1); (ii) defect-less graphene sheets; and (iii) a notable extension of the absorption edge into visible-light region [33]. With these advanced properties, the TiO2/graphene exhibited photocatalytic activity 15 and 5 times greater than that of TiO2-P25 and TiO2/rGO nanocomposite in visible-light photodegradation of MB [48].

Fig. 1
figure 1

Via chemical exfoliation of graphite in Ti(OnBu)4 and sol–gel reaction, a TiO2/graphene nanocomposite with advanced properties for visible-light photocatalysis had been synthesized. The TiO2/graphene nanocomposite exhibited an excellent visible-light photocatalytic activity in the photodegradation of MB

Full size image

In this time, the performance of the TiO2/graphene nanocomposites was challenged by the photodegradation of aromatic compounds in aqueous media. In general, the existence of graphene enhanced the ability of the catalyst to adsorb organic pollutants (Figs. 2a and S2a). This is because of the enhancement in surface area and/or π−π interaction between aromatic rings in naphthalene and phenol molecules and the aromatic domains of graphene potentiated their conjugation [54, 55]. Obviously, the TiO2/graphene nanocomposite performed far the highest adsorption ability. The large difference in the adsorption capacity was normalized by adjusting the amount of the catalysts to unify the initial concentration of the reactants before irradiation. The photocatalytic activity of the three catalysts was quantified by the amount of reactant degraded per gram of catalyst per hour. As shown in Fig. 2b, the TiO2/graphene nanocomposite performed far the highest photocatalytic activity, which was 3 and 16 times higher than those of TiO2/rGO and TiO2-P25, respectively. Similar results were observed in the photodegradation of phenol (Figure S2b). In order to determine the origin of the excellent photocatalytic activity of the TiO2/graphene nanocomposite, the activity of the catalysts was evaluated on the basic per adsorption amount rather than per gram of catalyst (Fig. 2c). It was found that in the photodecomposition of aromatic compounds under oxygen atmosphere, besides the great adsorption ability, the greatest photocatalytic activity of the TiO2/graphene nanocomposite was contributed by other factors. The contribution ratio between adsorption ability and other factors was about 1:1.

Fig. 2
figure 2

a Equilibrium adsorption of naphthalene on the photocatalysts. 20.0 mg of a catalyst was placed in 20 mL of naphthalene solution (100 mg mL−1) in dark for 6 h under constant stirring. After 6 h, the naphthalene concentration in the supernatant was measured in order to evaluate the amount of adsorption. b Photocatalytic activity per gram catalyst. The activities were evaluated from the loss of the naphthalene concentration after 4-h irradiation. c Photocatalytic activity per naphthalene adsorption amount. The activity per gram catalyst in b was divided by the naphthalene adsorption capacity per gram catalyst in a

Full size image

Insight into an excellent photocatalytic activity of the TiO2/graphene nanocomposite

In order to understand the mechanism of the excellent photocatalytic activity of the TiO2/graphene nanocomposite, the contribution of different active species in the visible-light photocatalysis was examined. The photodegradation of MB was chosen for this test due to the ease of reaction process and analysis. In general, different types of active species can be formed throughout the photocatalytic reaction, such as electrons, holes, hydroxyl radicals, and superoxide radicals [56,57,58]. The roles of these active species in the photodegradation reactions are different. Photoexcited electrons can reduce the oxygen molecules adsorbed on the catalyst surface to produce superoxide radicals (O2 (ads) + e− → O2·−) [59]. Photoinduced holes can contribute to the formation of hydroxyl radicals (HO− + h+  → HO·) and/or can be directly involved to oxidize organic pollutants (R· + h+  → R+  → intermediate(s)/final degradation products). Subsequently, the generated hydroxyl and superoxide radicals directly participate in the degradation of organic pollutants [60, 61]. Thus, in this photocatalyst systems for the MB photodegradation, holes, hydroxyl, and superoxide radicals were considered as potential active species. Tert-butyl alcohol, ammonium oxalate, and benzoquinone were used as scavengers for hydroxyl radicals, holes, and superoxide radicals, respectively. A scavenger which lets the activity of a catalyst reduced by its addition dictates the importance of the corresponding active species. Figure 3 presents the impact of the addition of different scavengers on the activity of photocatalysts. The impact was quantified by the loss of photocatalytic activity in the presence of a radical scavenger compared to the one in the absence of any radical scavengers. It can be seen that the addition of ammonium oxalate, a hole scavenger, exhibited the largest impact on the activity for all the three catalysts, indicating the importance of holes in the photocatalytic degradation of MB. However, the impact was most pronouncedly observed for the TiO2/graphene nanocomposite. The addition of tert-butyl alcohol, a scavenger for hydroxyl radicals, reduced the activity of TiO2-P25 and TiO2/rGO in comparison with that of the TiO2/graphene nanocomposite. Last, the addition of benzoquinone, a superoxide radical scavenger, slightly influenced the photocatalytic performance of TiO2-P25 and TiO2/rGO nanocomposite. Contrarily, the activity of the TiO2/graphene nanocomposite was obviously inhibited by the existence of benzoquinone. It can be considered that in the case of TiO2-P25 and TiO2/rGO, holes and hydroxyl radicals were mainly involved in the photocatalytic reaction. On the other hand, in the case of the TiO2/graphene nanocomposite, holes and especially superoxide radicals played more important roles for the photodegradation reaction, whereas hydroxyl radicals seemed not to be the major oxidation species in the TiO2/graphene nanocomposite.

Fig. 3
figure 3

Effect of radical scavengers on the photocatalytic efficiency of TiO2-P25, TiO2/rGO, and TiO2/graphene nanocomposites. The effect is represented by the loss of activity of a photocatalyst in the MB photodegradation in the presence of a radical scavenger compared to the one without any radical scavengers

Full size image

The active species study on the catalysts partly revealed the ability of the TiO2/graphene nanocomposite to activate oxygen molecules resulting in the formation of superoxide radicals, which was considered as an advantage in the photocatalytic activity of the nanocomposite. The oxygen activation ability of the TiO2/graphene nanocomposite was also confirmed by EPR experiments. Figure 4 presents the EPR spectra of TiO2/graphene and TiO2/rGO nanocomposites recorded under three conditions: (a) in dark, (b) under UV irradiation in vacuum, and (c) under UV irradiation in the presence of molecular oxygen. No relevant signal was observed in both catalysts prior illumination. Upon UV irradiation in vacuum of the TiO2/rGO sample, an intense signal resonating at ≈ 345 mT (g = 1.99) is observed, which is assigned to Ti3+ species [62, 63]. The EPR fingerprint of Ti3+ species is observed also in the case of UV irradiation of the TiO2/graphene nanocomposite (Fig. 4). In this case, the signal is broader and characteristic of surface Ti3+ species [64]. Irradiation under oxygen atmosphere leads to drastically different results on the two different photocatalysts. In the case of the TiO2/rGO sample, the presence of oxygen leads to an EPR silent system, indicating a diamagnetic reactive pathway; on the contrary, the same experiment performed in the TiO2/graphene nanocomposite leads to the appearance of the typical spectrum of a superoxide radical anion [62]. It is known that under irradiation of UV light on the TiO2 surface, photoexcited electron–hole pairs are generated. The photoexcited electrons and holes can react with adsorbed molecular species, leading to the well-established photoactivity of TiO2. Oxygen molecules typically act as electron scavengers. The first step in this reduction process is the formation of superoxide radicals. In the case of the TiO2/graphene nanocomposite, due to the special sheet-like morphology, photoexcited electrons from the conduction band of TiO2 are available as reducing agents and thanks to the excellent adsorption ability of the TiO2/graphene nanocomposite, the superoxide radicals are stabilized on the surface, as proven by the observation of the EPR signal. In the case of TiO2/rGO nanocomposite, photoexcited electrons appear to react with oxygen; however, the reaction does not stop with the formation of O2− radicals but evolves toward diamagnetic species such as peroxide ions. This may be due to the different surface morphology and unavailability of surface trapping sites capable of trapping the radical. Thus, it appears that the two catalysts both promote the activation of oxygen molecules but to different extents and through different reactive pathways, highlighting the role of surface structure and morphology in tuning the photocatalytic properties of heterogeneous photocatalysts.

Fig. 4
figure 4

EPR spectra of a TiO2/graphene and b TiO2/rGO nanocomposites recorded in dark (black trace) after UV irradiation in vacuum (red trace) and UV irradiation under oxygen atmosphere (blue trace). All spectra were recorded at 77 K. (Color figure online)

Full size image

According to the behavior of the TiO2/graphene nanocomposite in the photodegradation of aromatic compounds and especially under the oxygen atmosphere, the prolongation of charge carrier lifetime is essential for an excellent photocatalyst. The efficiency of the charge carrier trapping, transfer, and separation in the catalysts was studied by PL emission measurement (Fig. 5a). Since the photoemission arises from the recombination of the excited electrons and holes, a lower PL reveals lower recombination rate of the excited electrons and holes. The PL spectra of the three samples show six emission peaks. The peak at 398 and 420 nm corresponds to the band gap energy of TiO2 and the hydroxylated Ti3+ surface complexes, respectively [58, 59]. The four peaks in the range of 450–493 nm were from excitonic PL of non-stoichiometric TiO2 and surface oxygen vacancies of TiO2 [60]. Compared to TiO2-P25, the PL intensity of the TiO2/rGO and TiO2/graphene samples was dramatically decreased. This reveals the suppression of electron–hole recombination in TiO2 in the presence of graphene, and it is more significant in the TiO2/graphene nanocomposite.

Fig. 5
figure 5

a PL emission spectra and b TRPL spectra of TiO2-P25, TiO2/rGO, and TiO2/graphene nanocomposites

Full size image

The specific charge carrier dynamics of the catalysts were probed by the time-resolved PL spectroscopy (Fig. 5b). The emission decay curves of the two samples were fitted well by biexponential kinetic function (Eq. 1).

$$\mathrm{I}\left(t\right)={A}_{1}\,\mathrm{exp}\left(-t/{\tau_{1}}\right)+{A}_{2}\,\mathrm{exp}\left(-t/ {{\tau }_{2}}\right)$$
(1)

where \({\tau }_{1}\) and \({\tau }_{2}\) are emission lifetimes and A1, A2 are the corresponding amplitudes. The longer lifetime \({\tau }_{1}\) is originated from the free-exciton recombination in TiO2 and \({\tau }_{2}\) is attributed to the surface-related non-radiative recombination of the charge carriers [61, 65,66,67]. The average emission lifetime representing the overall emission decay behavior of samples was calculated by the following equation (Eq. 2),

$${\tau }_{\mathrm{ave}.}=\frac{{A}_{1}{\tau }_{1}^{2}+{A}_{2}{\tau }_{2}^{2}}{{A}_{1}{\tau }_{1}+{A}_{2}{\tau }_{2}}$$
(2)

As shown in Table 1, the incorporation of graphene decreased the photoluminescence lifetime of TiO2 suggesting the formation of electron transfer channel from TiO2 to graphene in a non-radiative quenching pathway [60, 61, 65,66,67,68]. The dynamics process of photoexcited charge carriers can be further investigated via the electron transfer rate (ket) calculated by Eq. 3.

Table 1 Parameter of photoluminescence decay analysis of the photocatalysts
Full size table
$${k}_{\mathrm{et}}= \frac{1}{{\tau }_{{\mathrm{TiO}}_{2}/\mathrm{graphene}}}-\frac{1}{{\tau }_{{\mathrm{TiO}}_{2}}}$$
(3)

The electron transfer rate of the TiO2/graphene nanocomposite was 0.157 ns−1, which was 9 times faster than the electron transfer rate of TiO2/rGO nanocomposite (0.016 ns−1). It can be considered that the special sheet-like structure of the TiO2/graphene nanocomposite facilitates the transportation of photoexcited electrons from the conduction band of TiO2 to graphene. This results in a high efficiency of charge carrier separation in the TiO2/graphene nanocomposite.

According to the above results, the photocatalytic mechanism of the TiO2/graphene nanocomposite in the degradation of organic compound in water is illustrated in Fig. 6. In general, under light irradiation, the photoexcited electrons and holes were formed (Reaction 1) and underwent a series of reactions (Reactions 2–4). With high efficiency in suppressing electron–hole recombination, the large amount of photoexcited electrons subsequently reduced the adsorbed oxygen molecules to form superoxide radicals (Reaction 5), trapped on the surface of the TiO2/graphene nanocomposite, and then directly degraded organic pollutant molecules to produce CO2 and H2O (Reaction 9). The trapped superoxide radicals are hypothesized as the distinct active species, explaining the excellent photocatalytic activity of the TiO2/graphene nanocomposite. In addition, with the excellent adsorption ability, organic pollutants strongly adsorb on the surface of TiO2/graphene, which is expected to facilitate the direct degradation of organic pollutants by photoinduced holes (Reaction 10) and instead decrease a chance of hydroxyl radical formation (Reactions 6 and 7). The mechanism of photocatalysis seems not very different between TiO2-P25 and TiO2/rGO. Similar to the TiO2/graphene nanocomposite, photoinduced electrons and holes underwent Reactions 2–6. However, more hydrophilic nature of TiO2-P25 and TiO2/rGO likely causes the adsorption of water molecules, and their reactions with photoinduced holes to form hydroxyl radicals and protons (Reaction 7) prior to the protonation of superoxide radicals to form hydroperoxide radicals (Reaction 11), which would generate hydroxyl radicals through a series of reaction (Reactions 12–14). Thus, in the photocatalysis of TiO2-P25 and TiO2/rGO nanocomposite, the contribution of photoinduced holes and hydroxyl radicals was predominant (Reactions 8 and 10), while the involvement of superoxide radicals was minor [47, 69, 70].

$${\text{Photoexcitation}}:{\text{ TiO}}_{{2}} + {\text{ h}}\nu \, \to {\text{ e}}^{-} + {\text{ h}}^{ + }$$
((Reaction 1))
$${\text{Charge carrier trapping of e}}^{ - } :{\text{ e}}^{ - }_{{{\text{CB}}}} \to {\text{ e}}^{ - }_{{{\text{TR}}}}$$
((Reaction 2))
$${\text{Charge carrier trapping of h}}^{ + } :{\text{ h}}^{ + }_{{{\text{VB}}}} \to {\text{ h}}^{ + }_{{{\text{TR}}}}$$
((Reaction 3))
$${\text{Electron}}{-}{\text{hole recombination}}:{\text{ e}}^{ - }_{{{\text{TR}}}} + {\text{ h}}^{ + }_{{{\text{VB}}}} \left( {{\text{h}}^{ + }_{{{\text{TR}}}} } \right) \, \to {\text{ e}}^{ - }_{{{\text{CB}}}} + {\text{ heat}}$$
((Reaction 4))
$${\text{Photoexcited electron scavenging}}: \, \left( {{\text{O}}_{{2}} } \right)_{{{\text{ads}}}} + {\text{ e}}^{ - } \to {\text{ O}}_{{2}}^{ \bullet - }$$
((Reaction 5))
$${\text{Oxidation of hydroxyls}}:{\text{ HO}}^{ - } + {\text{ h}}^{ + } \to {\text{ HO}}^{ \bullet }$$
((Reaction 6))
$${\text{Breakage of water molecule}}:{\text{ H}}_{{2}} {\text{O }} + {\text{h}}^{ + } \to {\text{ HO}}^{ \bullet } + {\text{ H}}^{ + }$$
((Reaction 7))
$${\text{Photodegradation by HO}}^{ \bullet } :{\text{ R}}{-}{\text{H }} + {\text{ HO}}^{ \bullet } \to {\text{ R}}^{ \bullet } + {\text{ H}}_{{2}} {\text{O}}$$
((Reaction 8))
$${\text{Photodegradation by O}}_{{2}}^{ \bullet - } :{\text{ R}}{-}{\text{H }} + {\text{ O}}_{{2}}^{ \bullet - } \to {\text{ CO}}_{{2}} + {\text{ H}}_{{2}} {\text{O}}$$
((Reaction 9))
$${\text{Direct oxidation by}} {\text{photoinducedholes}}:{\text{ R}}^{ \bullet } + {\text{ h}}^{ + } \to {\text{ R}}^{ + } \to {\text{ intermediate}}\left( {\text{s}} \right)/{\text{final degradation products}}$$
((Reaction 10))
$${\text{Protonation of superoxide}}:{\text{ O}}_{{2}}^{ \bullet - } + {\text{ H}}^{ + } \to {\text{ HOO}}^{ \bullet }$$
((Reaction 11))
$${\text{Co}} - {\text{scavenging of e}}^{ - } :{\text{ HOO}}^{ \bullet } + {\text{ e}}^{ - } \to {\text{ HOO}}^{ - }$$
((Reaction 12))
$${\text{Formation of H}}_{{2}} {\text{O}}_{{2}} :{\text{ HOO}}^{ - } + {\text{ H}}^{ + } \to {\text{ H}}_{{2}} {\text{O}}_{{2}}$$
((Reaction 13))
$${\text{Reduction ofH}}_{{2}} {\text{O}}_{{2}} :{\text{H}}_{{2}} {\text{O}}_{{2}} + {\text{e}}^{ - } \to {\text{ HO}}^{ \bullet } + {\text{ HO}}^{ - }$$
((Reaction 14))
Fig. 6
figure 6

Photocatalytic mechanism of the TiO2/graphene nanocomposite in the degradation of organic compounds in aqueous media. Photoinduced holes and trapped superoxide radicals were the main active species contributed to the photocatalytic reaction

Full size image

Conclusion

In conclusion, the TiO2/graphene nanocomposite synthesized using a graphene dispersion in Ti(OnBu)4 exhibited superior photocatalytic activity in degradation of organic pollutants in aqueous media. Through photocatalytic test, radical scavenger tests, electron paramagnetic resonance measurements, and photoluminescence measurements, the TiO2/graphene nanocomposite exhibited (i) an excellent capacity to adsorb organic pollutants; (ii) a great efficacy to separate photoexcited electrons and holes; (iii) a great ability to activate adsorbed oxygen molecules; and (iv) ability to trap superoxide radicals. Based on these finding, it can be concluded that the excellent visible-light photocatalytic activity of the TiO2/graphene nanocomposite is originated from the great ability to activate the adsorbed molecular oxygen and subsequently trap the superoxide radicals, which are considered as essential active species for the great photocatalytic activity of the TiO2/graphene nanocomposite.