1 Introduction

Toxic organic pollutants, like hydrocarbons, esters, aldehydes, and phenols, emitted from chemical industries and modern life can lead to serious environmental pollution and health issues [1, 2]. Great efforts have been made to reduce the emission of organic pollutants, as well as their effective collection and degradation. Photocatalytic degradation of organic pollutants via catalytic oxidation process possesses two advantages of (i) renewable and clean solar energy as energy sources and (ii) the reaction products of CO2 and H2O showing no secondary pollution [3,4,5]. Therefore, the photocatalytic degradation strategy is considered as a promising and economic approach to resolve the toxic organic pollution. Efficient photocatalytic degradation of organic pollutants requires excellent photocatalyst to break the strong chemical bonds (e.g., C–C, C–H, and C = O bonds) of organic molecule [6]. The commonly used catalysts are indispensable active components of noble metals (likely Pd, Pt, Au, and Ag) coupled with semiconductors [7,8,9]. Exploration of alternative cost-efficient materials is also necessary to replace the noble catalysts. Titanium dioxide (TiO2) as the well-known photo(electro)catalyst is widely used in the fields of water splitting, methyl orange degradation, CO2 reduction, and so on [10,11,12,13,14,15]. However, the efficiencies of full-spectrum solar utilization are still very low because of the large band gap of 3.0–3.2 eV, poor visible (Vis) and even infrared (IR) light response, low carrier concentrations, weak separation and transportation of photoinduced electron–hole (e−−h+) pairs [16]. Therefore, the microstructures and physicochemical properties of the TiO2-based photocatalysts need to be improved markedly.

To narrow the intrinsic band gap and extend the response region of TiO2 to the Vis or IR light, large number of elements are doped to obtain colourful materials [10,11,12,13,14,15, 17, 18]. However, these strategies show little improvement in aspect of the catalytic efficiency owing to the enhanced recombination rate of photogenerated electrons and holes, and low carrier concentrations. Although Ti3+-self doped TiO2 can enhance the solar energy absorption, numerous bulk defects are unfavourable for charge transfer [19]. Surface defects like oxygen vacancies (Vo) over semiconductors are proved to efficiently facilitate the catalytic performance [20]. Moreover, an appropriate amount of the surface defects can induce the generation of intermediate states of Ti3+ species which serve as light-generated carrier sinks for retarding charge recombination [21]. During the search for a readily available and applicable strategy for the preparation of full-spectrum solar responsive TiO2-based materials for energy transformation and chemical conversion, we have recently demonstrated that several methods including H2-plasma, metal reduction, and element doping were widely used for the synthesis of various black, red, gray, and blue titania [22,23,24,25,26]. Compared to pristine TiO2, these colored samples with unique crystalline-core/amorphous-shell structure and large amount of surface Vo species can enhance the separation and transportation of photoinduced e−−h+ pairs and thus the carrier concentrations. Thus, the extraordinary structure and physicochemical properties of colored TiO2 endow its superior microwave and terahertz absorption [27,28,29,30,31,32], special solar light absorption and excellent photocatalysis performance.

Herein, we report that the conductive black titania (BT) fabricated via Al-reduction method shows exceptional photocatalytic efficiency for the degradation of organic pollutants (e.g., toxic toluene and ethyl acetate) under solar irradiation and mild reaction conditions. The key to the successful use of the designed photocatalysts is the unique crystalline-core/amorphous-shell structure and numerous surface Vo showing narrow band gap, high carrier concentration, prominent electron mobility, and excellent separation and transportation of photoinduced e−−h+ pairs. These special physicochemical properties of series BT samples favor efficient adsorption, chemical activation, and photocatalytic oxidation transformation of organic pollutant molecule. The engineered BT-500 catalyst achieves superior photocatalytic degradation rates for toluene and ethyl acetate under full-spectrum solar and even visible light irradiation, about three-fold higher than those of the pristine TiO2. In addition, the catalyst gives excellent photostability with high degradation efficiency of 93% for the 6th reuse. The present findings are promising not only for offering an effective photocatalyst system for degradation of toxic organic pollutants to environmentally friendly products, but also for obtaining novel insights into colored titania catalyzed reactions.

2 Experimental

2.1 Chemicals and Materials

TiO2 (P25 with 80% anatase phase and 20% rutile phase, 99.8 wt%) was obtained from Evonik. Al powders (99.98 wt%) and Pt wire (99.95 wt%) were supplied by Alfa Aesar. Toluene (99.5 wt%), ethyl acetate (99.5 wt%), and NaOH (98 wt%) were supplied by Sinopharm Chemical Reagent Co., Ltd (SCRC). All the chemicals were used without further purification.

2.2 Catalyst Preparation

The preparation of BT samples were conducted in a two-zone furnace (Fig. S1) [25], in which TiO2 and Al powders were separately placed at a pressure of < 0.5 Pa. Later, Al was heated to 800 °C, and the TiO2 was heated to 300–500 °C for 6 h. After the annealing and pressure release, the BT samples were carefully collected.

2.3 Catalyst Characterization

Brunauer–Emmett–Teller (BET) specific surface areas of BT catalysts were determined by N2 adsorption–desorption at 77 K, using a Micromeritics TriStar 3000 equipment. Sample degassing was carried out at 300 °C prior to acquiring the adsorption isotherm. Al content was measured by inductively coupled plasma-atomic emission spectroscopy (ICP-AES) using iCAP 6300 spectrometer. X-ray diffraction (XRD) information of the samples was carried out on a German Bruker D8 Advance X-ray diffractometer using nickel filtered Cu Kα radiation at 40 kV and 40 mA. X-ray photoelectron spectroscopy (XPS) spectra were recorded with a Perkin Elmer PHI 5000C system equipped with a hemispherical electron energy analyzer. The spectrometer was operated at 15 kV and 20 mA, and a magnesium anode (Mg Kα, hν = 1253.6 eV) was used. A JEOL 2011 microscope operating at 200 kV equipped with an EDX unit (Si(Li) detector) was used for the transmission electron microscopy (TEM) investigations. A JEM 2100F electron microscope operating at 200 kV equipped with an EDX unit (Si(Li) detector) was used for the high-resolution (HR)TEM investigations. The samples for electron microscopy were prepared by dispersing the powder in ethanol and applying a drop of very dilute suspension on carbon-coated grids. UV–VIS-NIR diffuse reflectance (DR) measurement of the solids were recorded at room temperature on a Hitachi U4100 Spectrometer equipped with an integrating sphere and using BaSO4 as reference. The solar absorption of TiO2 samples was attained by the following Eq. (1) [23]:

$$A = \frac{{\int {\left( {1 - T} \right) \cdot S \cdot d\lambda } }}{{\int {S \cdot d\lambda } }}$$
(1)

where A is the solar absorption, T is the reflectance of the sample, S is the solar spectral irradiance (W m−2 nm−1), λ is the wavelength (nm), and the (1−T) S represents the sample absorption of solar spectral irradiance. Photothermic effect of the cool-pressed TiO2 and BT-500 powders were investigated under the irradiation of an AM 1.5G Xe lamp solar light simulator for different times. Raman spectra were collected in a thermal dispersive spectrometer using a 10 mW laser with an excitation wavelength of 532 nm. Electron paramagnetic resonance (EPR) spectra were investigated using a Bruker EMX-8 spectrometer at 9.44 GHz and 300 K. Resistance measurement was conducted using standard four electrode method performed on physical property measurement system (PPMS) (DynaCool-9, Quantum Design Inc.) at room temperature. The carrier concentration was determined using Hall-effect measurement method at the magnetic field ranged from − 2 T to 2 T at room temperature. To characterize the transient photocurrent responses of the samples, a conventional three electrode system was used to conduct electrochemical measurements with an electrochemical workstation (CHI660B, CH Instruments). Pristine TiO2 films were prepared on FTO glass substrate by a spin-coating method. These films were first annealed at 500 °C for 1 h in air and then treated by Al reduction. The sample-coated FTO glasses were used directly as the working electrode, with a Pt wire and an Ag/AgCl (KCl saturated) electrode as counter and reference electrodes respectively in 1 M NaOH aqueous solution (pH 13.6). A set of transient photocurrent responses were recorded in dark and under irradiation. A 150 W Xe lamp was used as the light source to simulate the sunlight irradiation. The light intensity was measured by a calibrated Si photodiode. Photoluminescence (PL) was excited at 320 nm using a fluorescence spectrophotometer of MODEL Fluoro Max-3 (Horiba) at room temperature.

2.4 Adsorption Measurement

Adsorption measurement before catalysis was carried out on the conditions of dark and 25 °C. The initial concentration of aqueous toluene or ethyl acetate solution was 20 mg L−1. 50 mg sample was added to 100 mL solution under stirring. At given time intervals, the selected solution (ca. 1 mL) was analyzed using a gas chromatograph Agilent 7820A equipped with a HP-5 capillary column connected to a flame ionization detector.

2.5 Photocatalytic Degradation Measurement

Photocatalytic degradation was conducted by the decomposition of toluene or ethyl acetate in an aqueous solution with an initial concentration of 50 mg L−1. 50 mg catalyst was added to 100 mL solution. A 300 W Xe lamp (Aulight CEL-HX, Beijing) was used as a light source for photocatalytic reaction. At given time intervals, the selected solution (1 mL) was analyzed using a gas chromatograph Agilent 7820A equipped with a HP-5 capillary column connected to a flame ionization detector.

Photostability test of BT-500 was carried out on solar light irradiation and 3 h in each run. After each reaction, the centrifuged catalysts from parallel tests were collected and washed with distilled water several times, followed by air-drying at 100 °C overnight.

3 Results and Discussion

3.1 Structural Characterization and Physicochemical Properties

To examine the influence of Vo and the amorphous shell on the photocatalytic degradation activity of BT, the pristine TiO2 treatment temperature is mainly focused on 300–500 °C in this study (Fig. S1). Thermodynamic analysis demonstrates that Al-reduction temperature of 800 °C is sufficient for the generation of melting aluminum and aluminium oxides and thus the O-defective TiO2−x thin shell over crystalline TiO2 core. Note that the higher temperature of 600 °C can lead to the again crystallization of amorphous shell and the significant decrease of surface Vo amount [23]. The obtained catalysts constituting of TiO2@TiO2−x are denoted as BT-300, BT-400, and 500-BT at the treatment temperature of 300, 400, and 500 °C, respectively.

BT samples show no remarkable change in BET surface area and pore size properties relative to the pristine TiO2 (Fig. S3 and Table S1), which are consistent with the previous results [23]. It should be emphasized that there is no Al element in the final BT samples, as proved by XPS and ICP-AES analysis in Fig. S4. TiO2 powder was converted from white to gray of BT-300 and dark-blue of BT-400 and black of BT-500 by increasing the reduction temperature (left column in Fig. 1), suggesting that the enhanced Vis-light absorption for dark color samples [33]. TEM images in Figs. S5 and 1b show that the pristine TiO2 is completely crystalline, displaying well-defined lattice fringes and well-ordered crystal domains of around 25 nm. Compared with crystallized TiO2 (Fig. 1b), BT samples present a typical core–shell structure, as the HRTEM images depicted in Figs. 1d, f, and h, in which the core is remained the crystallized TiO2 with anatase (101) plane whereas the outer shell is amorphous TiO2−x. Note that the distance between the adjacent lattice planes for these samples is remained ca. 3.5 Ǻ. These results suggest that the Al-reduction process only affects the surface structure of pristine material. The thickness of the amorphous layer increases with the reduction temperature; that is, ca. 1 nm for BT-300 to 1–2 nm for BT-400 to 2–3 nm for BT-500 (right column in Fig. 1). It was demonstrated that the disordered structure in metal oxides is favorable for the generation of surface defects and the presence of the crystallized–amorphous heterojunction is beneficial to the separation and transportation of photoinduced e−−h+ pairs [22,23,24,25,26].

Fig. 1
figure 1

Photographs and HRTEM images of a, b TiO2, c, d BT-300, e, f BT-400, and g, h BT-500 samples

Full size image

XRD patterns in Fig. 2 reveal a negligible phase transition, in which BT samples remain high crystallinity with the co-existence of dominant anatase phase and rutile TiO2 after Al-reduction treatment. The crystallite size of BT samples was estimated using Scherrer equation to be about 25 nm, which is similar to the pristine TiO2. As the unique crystalline-core/amorphous-shell structure was formed, the colored titania was generated. For BT-500 catalyst, the typical core–shell structure also results in a narrow band gap of ca. 2.8 eV (Fig. S6), which is much lower than that of pristine TiO2. BT samples significantly enhance the absorption of Vis and even near IR light, as illustrated in Fig. 3. The pristine TiO2 only shows response on ultraviolet (UV) light. The absorption spectra demonstrate the increased enhancement with the Al-reduction temperature increasing. For example, BT-500 possesses far larger solar spectrum absorption (ca. 71% of solar spectral irradiance) than those of BT-400 (~ 60%) and BT-300 (~ 53%).

Fig. 2
figure 2

XRD patterns of TiO2 and BT samples

Full size image
Fig. 3
figure 3

UV-VIS–NIR Diffuse reflectance spectra of TiO2 and BT samples

Full size image

The enhanced solar absorption of BT materials was further investigated by the photothermic effect under the irradiation of an AM 1.5G Xe lamp solar light simulator. The investigated disks were cool-pressed from pristine TiO2 and BT-500 powders. As depicted in Fig. 4, after being irradiated for 60 s, the temperature of the BT-500 disk increased to 38 °C, which was much higher than the 30 °C for the pristine TiO2 disk. The accelerated heating rate of the BT-500 disk is ascribed to larger solar absorption illustrated by the absorption spectrum in Fig. 3, resulting in more electron excitation and transition and the enhanced heat emission. Thus, the conductive black titania materials can be widely applied as a solar absorber medium for solar thermal collection.

Fig. 4
figure 4

Thermal image map of cool-pressed disks of TiO2 and BT-500 samples under solar simulator irradiation for different times

Full size image

The surface amorphous layer and Vo (or Ti3+) can be proved by Raman spectroscopy. The strongest signal of Eg mode at 145 cm−1 depicted in Fig. 5a is assigned to the external vibration of Ti–O bond [25]. This demonstrates a visible blue-shift (from 145 of TiO2 to 148 cm−1 of BT-300 to 154 cm−1 of BT-500) accompanied the slight peak broadening along with the Al-reduction treatment at elevated temperatures. The latter is probably due to the existence of numerous surface defects (Ti3+ or Vo) in amorphous TiO2−x layer [34]. XPS results illustrated in Fig. 5b show a slight shift to lower energy for Ti2p peak of BT samples, which are consistent well with the Raman data (Fig. 5a). There is no appearance of Ti3+ species in pristine TiO2, while obvious signal of Ti3+ can be observed in BT samples (Fig. 5b) and the proportion of Ti3+ in total Ti species increases with Al-reduction temperature (Table S2). EPR as a highly sensitive technique is widely applied to probe paramagnetic species containing the unpaired electrons in Vo and/or Ti3+ [22,23,24,25,26]. There is a strong EPR signal with the g-value of 2.003 for BT catalysts, as depicted in Fig. 5c. This is ascribed to the unpaired electrons [35], indicative of the presence of Vo or Ti3+ species in surface amorphous TiO2−x layer. Moreover, the EPR intensity increases gradually with the Al-reduction temperature. In order to reveal the behavior of photogenerated holes and electrons, PL emission tests were carried out. Fig. S7 shows the PL spectra of pristine TiO2 and BT samples in the wavelength region of 350–650 nm. The intensities of BT catalysts are much weaker than that of pristine TiO2, indicative of a much lower recombination rate of the holes and electrons irradiated by UV light source for BT samples [22,23,24,25,26]. BT-500 shows the lowest PL peak intensity. The unique properties of the crystalline-core/amorphous-shell structured BT with abundant Vo or Ti3+ species can significantly reduce the recombination rate of e−−h+ pairs. Thus, efficiently promoting the separation and transportation of e−−h+ and the migration/relaxation of charge carriers within the BT matrix.

Fig. 5
figure 5

a Raman, b Ti2p XPS, and c EPR data of TiO2 and BT samples

Full size image

It was demonstrated that the Ti3+ species can generate 0.7–1.0 eV below the conduction band minimum (CBM) and just below the CB tail in the electronic structure of BT sample [23]. Mild thermal treatment conditions or even photons with relatively low energy can also induce the Ti3+ 3d1 electron to band tail or upper energy levels (Fig. S8). This shows a positive relationship to the highly enhanced electron concentration. In our previous study [23], the electron concentrations of BT-400 and BT-500 were recorded to be 8.87 × 1017 cm−3 and 7.41 × 1018 cm−3, respectively, according to the Hall measurements. Note that the electron concentrations of BT samples are about two orders of magnitude higher than that of TiO2. Meanwhile, the transport and accumulation of photogenerated charges as well as the heat produced in semiconductors (e.g., titania and colored titania) are essential in photocatalytic reactions [36,37,38]. Moreover, the current values are almost zero in dark while the photocurrent rapidly rises to a steady-state value with illumination. The observed steady-state photocurrent of BT-500 is 1.3 mA cm−2 (Fig. S9), which is approximately eight times larger than that of pristine TiO2 (0.17 mA cm−2), indicative of a much higher concentration of photo-induced carriers in BT samples due to the enhanced light-absorption as well as the effective separation and rapid transportation of light-generated carriers. These prominent properties can fulfill the potential requirement for photocatalytic degradation of organic pollutants over BT samples.

3.2 Photocatalytic Degradation of Organic Pollutants

The excellent photocatalytic performance of the BT samples was demonstrated by the degradation of toluene, as illustrated in Fig. 6. Note that the transformations did not occur at all without the use of BT catalysts. BT catalysts exhibit considerable full solar light and even Vis light activity, whereas the pristine TiO2 gives low activity under the same reaction conditions (Fig. 6). The catalytic efficiency of BT samples increases with the Al-reduction temperature, and BT-500 is the best. This is due to the more enhanced light absorption and lager number of surface Vo for BT-500 than other samples. It should be emphasized that BT-500 needs only 3 and 5 h for the degradation efficiency of more than 95% for toluene under full solar light and Vis light irradiation, respectively (Fig. 6). The degradation rates are about three-fold higher than those of the pristine TiO2. Moreover, BT samples exhibit significantly promoted activities for ethyl acetate degradation under full-spectrum solar irradiation with the similar tendency of toluene degradation, as the results depicted in Fig. 7. The degradation rate of ethyl acetate (BT-500 needs 6 h for the degradation efficiency of ca. 90% under full solar light irradiation) is much lower than that of toluene due to the complicated molecular structure of the former. These results demonstrate that BT catalysts, especially BT-500 with numerous surface Vo, which possesses a narrowed band gap and prominent properties on solar and even Vis light absorption, achieves a superior efficiency for photoinduced degradation of organic pollutants.

Fig. 6
figure 6

Photocatalytic activities of TiO2 and BT samples for toluene degradation under a full solar and b visible light irradiation

Full size image
Fig. 7
figure 7

Photocatalytic activities of TiO2 and BT samples for ethyl acetate degradation under full solar light irradiation

Full size image

Compared to the photocatalytic degradation of toluene and ethyl acetate under light irradiation, BT samples give low efficiency on adsorption capacity (Fig. S10). These results show that the physical adsorption process is easily saturated in limited time and the adsorbate is hard to be desorbed or converted downstream without external driving force [39,40,41,42]. As an important step of overall catalytic process, the adsorption with high efficiency is usually favorable for the catalytic transformation in practical conditions [43,44,45]. BT-500 sample with the largest amount of surface Vo shows the best adsorption efficiency for toluene and ethyl acetate (Fig. S10), thus displays the highest photocatalytic degradation for both organic pollutants under solar or Vis light irradiation (Figs. 6 and 7).

Having established that BT-500 catalyzed photoinduced degradation of toluene and ethyl acetate as a reliable and efficient catalysis system for oxidation removal of organic pollutants, we further studied the reusability of the photocatalyst under solar light irradiation. As illustrated in Fig. 8, BT-500 shows excellent stability for the degradation of toluene under full-spectrum solar light irradiation. This sample can be reused up to six cycles without remarkable loss in activity. The toluene degradation efficiency remains at a high level of 93% for the 6th reuse. Compared with fresh BT-500, there is no significant change on the aspects of the crystalline-TiO2-core/amorphous- TiO2−x-shell structure, black feature, abundant Vo, and prominent separation and transportation of photoinduced e−-h+ pairs of the used catalyst (Figs. 9, S11, and S12). It should be noted that the used BT-500 still shows distinct EPR signal corresponding to Vo or Ti3+ after photocatalytic degradation reaction for overall 18 h (Fig. 9b). These results further demonstrate the effectiveness of BT catalysts, especially BT-500 for full solar light excited photocatalytic degradation of organic pollutants under viable conditions.

Fig. 8
figure 8

Cycling tests of BT-500 for toluene degradation under full solar light irradiation

Full size image
Fig. 9
figure 9

a HRTEM image and b EPR spectrum of the used BT-500 sample

Full size image

4 Conclusions

Efficient conductive black titania nanomaterisls with typical crystalline-TiO2-core/amorphous- TiO2−x-shell structure were successfully fabricated by using two-zone Al-reduction strategy. Series BT catalysts were tested for the photocatalytic degradation of toxic organic pollutants (toluene and ethyl acetate) to environmentally friendly products under full solar or Vis light irradiation and mild reaction conditions. The unique core–shell structure and numerous surface Vo or Ti3+ species in the amorphous layer accompanying prominent physicochemical properties of narrow band gap, high carrier concentration, high electron mobility, and excellent separation and transportation of photoinduced e−−h+ pairs resulted in exceptional photocatalytic efficiency. The engineered BT-500 catalyst achieved superior photocatalytic degradation rates for toluene and ethyl acetate under full-spectrum solar and even Vis light irradiation, and an excellent photostability with high degradation efficiency of 93% for the 6th reuse. These findings are promising not only for offering an effective photocatalyst system for toxic organic pollutants degradation, but also for obtaining novel insights into colored titania catalyzed reactions.