Carbon Black-Doped Anatase TiO2 Nanorods for Solar Light-Induced Photocatalytic Degradation of Methylene Blue.

In this work, C-doped TiO2 nanorods were synthesized through doping carbon black into hydrothermally synthesized solid-state TiO2 nanowires (NWs) via calcination. The effects of carbon content on the morphology, phase structure, crystal structure, and photocatalytic property under both UV and solar light by the degradation of methylene blue (MB) were explored. Besides, the photoelectrochemical property of C-TiO2 was systematically studied to illustrate the solar light degradation mechanism. After doping with C, TiO2 NWs were reduced into nanorods and the surface became rough with dispersed particles. Results showed that C has successfully entered the TiO2 lattice, resulting in the lattice distortion, reduction of band gap, and the formation of C-Ti-O, which expands TiO2 to solar light activation. Comparing with P25 and anatase TiO2 NWs, doping with carbon black showed much higher UV light and solar light photocatalytic activity. The photocatalytic activity was characterized via the degradation of MB, showing that K ap was 0.0328 min-1 under solar light, while 0.1634 min-1 under UV irradiation. The main free radicals involved in methylene blue degradation are H+ and OH•-. Doping with carbon black led to the reduction of photocurrent in a long-term operation, while C-doping reduced the electron-hole recombination and enhanced the carrier migration.

Introduction

Titanium dioxide (TiO2) possesses outstanding photocatalytic activity,[1] which has been extensively used in wastewater purification,[2−5] water splitting,[6−8] volatile organic compound degradation,[9] CO2 reduction,[10,11] etc. Nevertheless, TiO2 can only be activated by ultraviolet light (UV, 5–9% in solar light) due to its large band gap (3.0–3.2 eV)[12] and fast recombination rate while most of the solar light is wasted.[13] To efficiently utilize the light energy, expanding the photoexcitation spectra of TiO2 to solar light is more desirable. A series of techniques have been developed to expand TiO2 to be active under solar light (or visible light) through grafting (polymers or nanoparticles),[14,15] doping,[16−18] coating,[19] etc. Among of these methods, doping is extensively used to extend TiO2 to the visible light spectrum by narrowing the band gap, separating photoelectrons, and reducing the recombination rate of valance hole.[17,18]

A variety of materials including transition metals and nonmetal elements (e.g., C,[20−23] N,[24] S,[25] B,[26] I,[27] etc.) have been doped into TiO2 to achieve solar light activation.[28,29] However, doping with metal elements may cause poor thermal stability[30] and an increment of recombination centers due to aggregation and larger particle size.[30,31] The critically thin doping (10–20 nm) layer at the interface fails to act as both electron and hole traps under thermal treatment.[31] In comparison, doping with nonmetal elements might generate the impurity band gap[32] and create an overlap between the intrinsic band gap,[33] thereby expanding TiO2 to visible light activation. Asahi et al.[34] first reported that visible light excitation of TiO2could be achieved by N-doping. N enters the TiO2 lattice and replaces O, resulting in the formation of crystal defects, and the N 2p doping level overlaps with O 2p, which narrows the band gap, thereby switching TiO2 into visible light activation.[35,36] Since then, a variety of nonmetal-doped TiO2 photocatalysts have been developed. Sakthivel et al.[37] compared the visible light degradation efficiency of 4-chlorophenol, showing that C-doped TiO2 was 5 times more active than N-doped TiO2. Carbon is widely used as a doping agent due to its large electron storage capacity, capability of absorbing visible light in a wide range (400–800 nm), high efficiency in separating photogenerated carriers, stabilization on the anatase phase, and conductivity improvement.[38−41] Besides, C enters the lattice and generates a C–Ti–O bond and forms a hybrid orbital above the valance band, and enhances visible light adsorption of TiO2.[42,43] Zainullina et al.[44] reported that doping with C resulted in visible excitation due to the formation of Ti 3d-C 2p-O 2p hybrid orbital states transition from the valance band to the impurity band. Dong and co-workers[45] mixed Ti(SO4)2 and C12H22O11 with water to synthesize C-doped mesoporous TiO2 with visible light photocatalytic activity via hydrothermal treatment. Colombo et al.[46] reported the fabrication of C-doped TiO2 nanoparticles with visible light activity by decreasing the recombination rate of electron/holes through sol–gel doping glucose into TiO2. Wang et al.[47] utilized benzoic acid as a carbon source for doping into platelike TiO2, resulting in a lattice expansion, thereby improving the visible light photocatalytic activity. Wu et al.[48] synthesized carbon-doped TiO2 nanoparticles through the hydrothermal reaction between titanium tetra-n-butoxide and ethanol obtaining visible light activity. Though carbon-doped TiO2 has been synthesized from sol–gel or hydrothermal reactions between organic carbon and tetrabutyl orthotitanate or titanium compounds, introducing the C atom into the TiO2 lattice to obtain uniformly doped TiO2 and stable catalysts still remains a challenge. Especially, doping carbon particles into the solid-state TiO2 powder is difficult.

To dope element C into TiO2, several methods have been reported, e.g., doping carbon into Ti in an oxygen atmosphere,[49] TiC oxidation.[50,51] Shen et al.[49] oxidized TiC powder in an air atmosphere at different temperatures expanding TiO2 to visible light irradiation. Varnagiris et al.[50] synthesized carbon black-doped TiO2 (with the mixed-phase composition of anatase and rutile) through magnetron sputtering, where the carbon powder was placed on the Ti cathode and sputtered under DC current in an oxygen atmosphere. The sputtering time was varied, and different amounts of C were doped into TiO2. The highest degradation rate constant under visible light irradiation of methylene blue was 1.14 × 10–3 min–1 after being sputtered for 180 min. Nevertheless, the doping content of the carbon powder in TiO2 is difficult to precisely control. Besides, the effects of doping content of carbon powder on the phase behavior, crystal structure, and morphology are rarely reported.

In this work, C-doped TiO2 nanorods were synthesized through doping carbon black into hydrothermally synthesized solid-state TiO2 nanowires (NWs) via calcination. The effects of carbon content on the morphology, phase structure, crystal structure, and the photocatalytic property under both UV and solar light by the degradation of methylene blue (MB) were explored. Besides, the photoelectrochemical property of C-TiO2 was systematically studied to illustrate the solar light degradation mechanism. The photocatalytic activity of varied amounts of carbon black-doped TiO2 nanorods was compared to those of P25 and anatase TiO2 NWs, indicating that doping with carbon black resulted in much higher UV light and solar light photocatalytic activity via the degradation of MB. The generated free radicals and the dominant free radicals on MB degradation were also characterized. Results showed that C-doping into TiO2 leads to the lattice distortion and the formation of O–Ti–C bond and reduces the band gap, thereby extending TiO2 to the solar light activation region.

Results and Discussion

The alkali hydrothermal method[52] was utilized to synthesize TiO2 NWs and carbon black-doped TiO2 (C-TiO2) nanorods from commercial TiO2 (P25). A certain amount of P25 was well dispersed into 10 M NaOH and hydrothermally treated at 260 °C for 24 h in a Teflon autoclave (Figure ). The obtained intermediate was washed with 0.1 M HCl and DI water to remove Na+ and adjust the pH to around 7. Then, it was dried completely at 80 °C and heat-treated in an airtight tube furnace to obtain TiO2 NWs under 700 °C for 2 h. The as-prepared TiO2 NWs were mixed with a certain amount of carbon black and then heat-treated at 700 °C for 2 h in a N2-filled airtight tube furnace to obtain C-doped TiO2 nanorods (Figure ).

Figure 1

Schematic Illustration of the Fabrication Process of C-Doped TiO2.

Scheman class="Chemical">tic Illustration of the Fabrication Process of C-Doped TiO2.

Since P25 is composed of 20% rutile and 80% anatase, the phase composition of carbon black-doped TiO2 nanorods was investigated via XRD. As shown in Figure a, only anatase was detected after doping with carbon black, and increasing the doping content of carbon black does not lead to peak shifting. The characteristic peaks of anatase were observed at 2θ values of 25.26, 36.88, 48.04, 53.80, 55.00, 62.62, 68.76, 70.32, and 75.04°, which correspond to (101), (004), (200), (105), (211), (204), (116), (200), and (215). The relatively strong intensity of these peaks indicates the good crystallinity of C-doped TiO2, and the increment of the doping content of C generated a negligible effect on the phase structure of TiO2. Further characterization was performed by calculating the crystallite size. The crystallite size of TiO2 nanowires—(0.2% C)-TiO2, (0.5% C)-TiO2 and (1.0% C)-TiO2—was calculated at 31.92 nm, 31.66 nm, 33.21 nm and 30.48 nm, respectively. This result illustrates the negligible effects of carbon content on the crystallite size of TiO2.

Figure 2

(a) XRD analysis of C-TiO2 nanorods at varied carbon doping contents. (b) Raman shift of TiO2 NWs and C-doped TiO2 nanorods at different doping contents.

(a) XRD analysis of C-TiO2 nanorods at varied n class="Chemical">carbon doping contents. (b) Raman shift of TiO2 NWs and C-doped TiO2 nanorods at different doping contents.

Figure b shows the Raman spectra of TiO2 NWs and C-doped TiO2 nanorods showing that doping with C caused obvious peak shifting and widening (Figure b). Comparing with TiO2 NWs, significant peak shifting was observed at 135, 488, and 605 cm–1 after doping with carbon black. Furthermore, the peak shifting became more obvious with the increment of C-doping content (Figure b). In detail, at the peak shift of 135 cm–1, the peaks after C-doping shifted to the right at 137, 139, and 142 cm–1 corresponding to (0.2% C)-TiO2, (0.5% C)-TiO2, and (1.0% C)-TiO2 (Figure b inset (1)). However, the peaks shift to the left at 488 and 605 cm–1 (Figure b inset (2)) after C-doping. The 488 peak shifts to 487, 486, and 482 cm–1, while the peak of 605 cm–1 shifts to 604, 602, and 600 cm–1 after doping with 0.2, 0.5, and 1.0% C correspondingly (Figure b inset (2)). Moreover, the peak shifts at 135, 488, and 605 cm–1 became wider with the increment of the doping content of carbon black (Figure b insets). These results illustrate that doping with carbon black induced the generation of defects or impure states, and this phenomenon becomes more apparent with the increasing doping content of carbon black.[53,54] Furthermore, the intensity of these peaks is much stronger than that of TiO2 NWs (Figure b insets), indicating the good crystallization of TiO2 nanorods after doping with carbon black.[54]

The morphology of as-prepared specimens was characterized by scanning electron microscopy (SEM) (Figure ). The TiO2 NWs synthesized by the hydrothermal reaction are relatively smooth with width and length of 200 nm and several millimeters, respectively (Figure a,b). However, notable changes have been observed from the SEM that the long nanowires were turned into short nanorods after doping with carbon black (Figure c–e). Furthermore, the surface of C-doped TiO2 nanorods became rough and dispersed with many nanoparticles (Figure c–e). This phenomenon became more significant with the increment of C-doping content. The composition of these dispersed nanoparticles on (1.0% C)-TiO2 nanorods was analyzed by energy-dispersive X-ray spectroscopy (EDX), which shows that the main composition included C, Ti, and O (Figure f). Moreover, the content of C in C-doped TiO2 increased with the increment of the doping load of carbon black (Figure S1). A continuous increase in the doping load of C (2.0 and 3.0 wt %) caused a significant reduction of the aspect ratio of TiO2 nanorods, and more particles were observed on the surface of TiO2 (Figure S2). The result shows the significant effects of carbon black on the morphology change after doping.

Figure 3

Morphology of the as-prepared TiO2 NWs and C-doped TiO2 through SEM characterization: (a, b) TiO2 NWs, (c) (0.2% C)-TiO2, (d) (0.5% C)-TiO2, (e) (1.0% C)-TiO2, and (f) EDS analysis of (1.0% C)-TiO2.

Morphology of the n class="Chemical">as-prepared TiO2 NWs and C-doped TiO2 through SEM characterization: (a, b) TiO2 NWs, (c) (0.2% C)-TiO2, (d) (0.5% C)-TiO2, (e) (1.0% C)-TiO2, and (f) EDS analysis of (1.0% C)-TiO2.

To further investigate the surface structure of carbon black-doped TiO2 nanorods, transmission electron microscopy (TEM) was used to analyze the surface morphology and the results are compared to the original TiO2 nanorods, which shows that the original TiO2 is smooth and well crystallized (Figure a,b) while there are many nanoparticles dispersed on the surface (Figure c,d). Some of the carbon black nanoparticles coated onto TiO2 nanorods (Figure c, yellow square) form a thin layer (around 7 nm) at the interface, which can act as electron acceptors to separate electron and holes and protect Ti3+ from oxidation.[23] The high-resolution image of the selected square in Figure c shows that there is a disconnected area (Figure d) that might be formed by the doping of carbon black. Similarly, at the edge of TiO2 nanorods (Figure e, selected area), the crystal faces were not well aligned, showing a slight distortion (Figure f) due to the doped carbon black.

Figure 4

(a, b) TEM images of original TiO2 nanorod. TEM image of (0.5% C)-TiO2 nanorods: (c) carbon black coated onto TiO2 nanorods, and (d) carbon black forming a thin layer on the surface of TiO2 nanorod, and (e, f) the edge of TiO2 nanorod is not well aligned after doping with C.

(a, b) TEM images of original TiO2 nanorod. TEM image of (0.5% C)-n class="Chemical">TiO2 nanorods: (c) carbon black coated onto TiO2 nanorods, and (d) carbon black forming a thin layer on the surface of TiO2 nanorod, and (e, f) the edge of TiO2 nanorod is not well aligned after doping with C.

The effects of C loading on the chemical composition of C-doped TiO2 nanorods were analyzed via X-ray photoelectron spectroscopy (XPS), which shows that a O–Ti–C bond was formed after doping with 0.2% carbon black and that the intensity of the O–Ti–C bond becomes stronger with the increment of the doping amount of C (Figure a–c). The content of C=C and C=O decreased with doping more C, while the content of C–O increased with doping more C (Figure a–c and Table ). The atomic percentages of O–Ti–C (283.0 eV), C=C (284.6 eV), C–O (286.1 eV), and C=O (288.4 eV) are 9.95, 44.45, 33.91, and 11.69% in 0.5% C-TiO2 (Table ). In comparison, the intensity of O–Ti–C in (1.0% C)-TiO2 (10.72%) is higher than that of the O–Ti–C bond in (0.2% C)-TiO2 (5.76%). Interestingly, the intensities of the C–O bonds in (0.5% C)-TiO2 (33.91%) are the strongest (Table ). It can be concluded that carbon black and N2 provided a reduction atmosphere, which enhances the doping of C into the TiO2 lattice. Figure d–f illustrates the effects of C dosage on the O 1s, showing that the content of the lattice O-ion state (O2–) at 529.5 eV and defects of O* at 531.6 eV increase with the increment of doping loading of C (from 0.2 to 1.0%). Results showed that the generation of the oxygen defects and oxygen vacancies (O*) lowers the energy band through doping C in TiO2 NWs. The intensity of the lattice O– in (1.0% C)-TiO2 is much higher than that of (0.2% C)-TiO2, indicating that doping more C enhances the generation of lattice O2–. Lattice oxygen vacancies can be used as an effective transfer medium and electron acceptor, which can efficiently suppress the electrons and holes in the compound, promoting the catalytic process. The XPS results confirmed that carbon black doped into TiO2 nanorods via the C atom entering the lattice and the formation of the C–Ti–O bond at the interface of carbon black and TiO2 nanorods, which showed negligible effects of Ti 2p (Figure S3).

Figure 5

Chemical states of C 1s and O 1s at varied carbon addition: (a)–(c) C 1s of (0.2% C)-TiO2 nanorods, (0.5% C)-TiO2 nanorods, and (1.0% C)-TiO2 nanorods, respectively. (d)–(f) O 1s of (0.2% C)-TiO2 nanorods, (0.5% C)-TiO2 nanorods, and (1.0% C)-TiO2 nanorods, respectively.

Table 1

Atomic Percent (%) of Different C 1s Chemical States of As-Prepared Samples

sample	(0.2% C)-TiO2	(0.5% C)-TiO2	(1.0% C)-TiO2
C 1s (C=C)	66.14	44.45	66.92
C 1s (C–O)	15.82	33.91	16.24
C 1s (C=O)	12.28	11.69	11.11
C 1s (O–Ti–C)	5.76	9.95	10.72

Chemical states of C 1s and O 1s at varied n class="Chemical">carbon addition: (a)–(c) C 1s of (0.2% C)-TiO2 nanorods, (0.5% C)-TiO2 nanorods, and (1.0% C)-TiO2 nanorods, respectively. (d)–(f) O 1s of (0.2% C)-TiO2 nanorods, (0.5% C)-TiO2 nanorods, and (1.0% C)-TiO2 nanorods, respectively.

The effects of C-doping content on the photocatalytic activity were characterized under UV and solar light via UV-DRS (Figure ). A relatively strong absorption peak in UV spectral range (380–400 nm) and certain visible light absorption in the visible light range (more than 400 nm) are observed in C-doped TiO2 nanorods (Figure a). The absorption intensity of C-doped TiO2 nanorods increases with the increasing of the doping content of C under solar light. According to the turning point in these curves, a blueshift is observed, indicating the generation of electron–hole pairs and the stronger ability of photocatalytic oxidation in C-doped TiO2 nanorods. In comparison, the TiO2 NWs show a weak UV light adsorption peak and a lower solar light adsorption intensity (Figure a). The band gap of TiO2 NWs and C-doped TiO2 nanorods decreased with increasing doping amount of C. The band energy (Figure b) of doped samples was narrowed to 2.40–3.20 eV (Table S1 and Figure S4). This might be attributed to the doping of the C atom in the TiO2 lattice and the replacement of the O atoms, which narrow the band gap of TiO2.

Figure 6

Diffuse reflectance spectra of TiO2 NWs and C-doped TiO2 nanorods between 200 and 600 nm (a) and the corresponding band energy (b).

The photocatalytic property of TiO2 NWs and C-doped TiO2 nanorods is evaluated via the degradation of MB in terms of ln(C0/C) and Kap under UV light and solar light irradiation, respectively (Figures and S5). The photocatalytic degradation efficiency of C-doped TiO2 nanorods increases notably with the increment of C-doping content under UV light, and the Kap value reaches to as high as 0.1634 min–1 (Figure a) and the MB solution turned transparent after degradation for 15 min (Figure a, top right). The effect of C on enhancing the solar light photocatalytic performance of C-TiO2 was also investigated. The Kap value of TiO2 NWs is 0.0112 cm–1 (Figure b), and the MB solution remained blue after degradation of 120 min (Figure b, bottom right). Compared to TiO2 NWs, the highest Kap (0.0328 cm–1) was achieved after doping 1.0% C (Figure b), showing the vital role of C in enhancing the solar light photocatalytic activity. Furthermore, MB was quickly decomposed within 30 min under solar light using (1.0% C-TiO2) (Figure b, top right).

Figure 7

Photocatalytic activities of TiO2 NWs and C-doped TiO2 nanorods under (a) UV light and (b) solar light.

Photocatalytic acn class="Chemical">tivities of TiO2 NWs and C-doped TiO2 nanorods under (a) UV light and (b) solar light.

To verify the active species participating in MB photodegradation, several charge-trapping agents were utilized to capture specific radical species. It is well known that photoinduced charges, i.e., photoinduced h+ and photoinduced e–, are generated through the photoexcitation of TiO2. The photoinduced e– can react with the surface-absorbed O2 and transform into intermediate species of superoxide radical anions O2–, or recombine with photoinduced h+. The photoinduced h+ can participate in the degradation or react with H2O/OH– producing active hydroxyl radicals OH•. In this work, AgNO3, n-butyalcohol, and (NH4)2C2O4 were used as e–, OH–, and h+ scavengers, respectively. (0.5% C)-TiO2 nanorods were used as a model to illustrate the generated active species and were irradiated under 365 nm UV irradiation for 60 min following the same procedure of C-TiO2 photodegradation of MB.

As shown in Figure a, the degradation rate (D%) of 0.03 mM MB was 94.5% (black), 95.9% with e– trapping agent (AgNO3), 21.2% with OH– trapping agent (n-butyalcohol), 15.8% with h+ trapping agent ((NH4)2C2O4), and 5.3% with OH– + h+ trapping agents after UV radiation for 40 min. The photoinduced e– was transformed to O2– and then reacted with H2O generating H2O2, thereby producing OH– with the addition of trapping agent n-butyalcohol OH•. However, the remaining h+ may still act as active species (Figures a-(2), 21.2%). After adding (NH4)2C2O4 to trap h+, the generation of OH– originated from photoinduced h+ was largely restricted, while the remaining OH– or O2– (produced by photoinduced e–) decomposed 15.8% MB, verifying that OH– or O2– may also act as active species. With both OH– + h+ trapping agents added in the solution, a slight degradation (5.3%) was observed, verifying that the intermediate species of O2– was not the main reaction species. The contribution rates of all species are listed in Table S2. Figure b shows the digital image of the degraded MB solution with varied scavengers at different degradation times (0, 10, 20, 30, 40, 50, 60 min). The results verified that the main free radicals in our study were OH– and the photoinduced h+, which also play an important role in the photodecomposition of MB.

Figure 8

(a) Photocatalytic activities of (0.5%C)-TiO2 under UV light for 40 min with AgNO3, n-butyalcohol, (NH4)2C2O4, and n-butylalcohol + (NH4)2C2O4 as charge-trapping agents and (b) the corresponding degradation process in (a).

(a) Photocatalytic acn class="Chemical">tivities of (0.5%C)-TiO2 under UV light for 40 min with AgNO3, n-butyalcohol, (NH4)2C2O4, and n-butylalcohol + (NH4)2C2O4 as charge-trapping agents and (b) the corresponding degradation process in (a).

The photocurrent of the samples under UV irradiation was characterized by performing six on–off cycles[55] with a duration of 120 s (Figure a). The average stable photocurrent of nanowire series samples was close to each other and much lower than that of P25. P25 exhibited the highest photocurrent (8.0 μA/cm2) in the first 20 s, which then dropped and remained at 3.4 μA/cm2 with increasing operation time, while the photocurrents of TiO2 NWs and varied C-doped TiO2 were 6.0 μA/cm2 (TiO2 NWs), 7.2 μA/cm2 (0.2% C), 7.2 μA/cm2 (0.5% C), and 6.4 μA/cm2 (1.0% C), which then decreased and remained stable at 0.67, 0.3, 0.8, and 0.9 μA/cm2, respectively. This result illustrates that the instantaneous photocurrents in the first 20 s of TiO2 NWs and C-TiO2 were close to that of P25, while the long-term photocurrent of C-TiO2 dropped sharply (Figure a). The long-term photocurrent of C-TiO2 was restricted by the doping of carbon black, implying the low transmission speed of photocurrent in the long-term operation.

Figure 9

Photoelectric properties of P25, TiO2 NWs, and C-doped TiO2 nanorods: (a) photocurrent spectra, (b) EIS Nyquist spectra, and (c) PL spectra.

Photoelectric properties of n class="Gene">P25, TiO2 NWs, and C-doped TiO2 nanorods: (a) photocurrent spectra, (b) EIS Nyquist spectra, and (c) PL spectra.

Electrochemical impedance spectroscopy (EIS) is one of the most powerful tools for studying the electrochemical processes occurring at the electrode/electrolyte interface (medium-frequency region, MFR, 104–102).[56] The radius of AC impedance represents the resistance of carrier migration (1/2Rct) and reveals the transportation of charge carriers. Generally, a larger AC impedance radius stands for the higher resistance of the charge carrier migration. The EIS Nyquist spectra in Figure b show that the catalysts of C-doped TiO2 had a lower Rct than those of P25 and TiO2 NWs, indicating the rapid electron transport of electron–hole pairs through the electrode reaction. According to the EDX and XPS results, the generated defects and the coated carbon black nanoparticle (oxycarbides, C–O bond) might contribute to the decrease of the charge transfer resistance. Although the as-prepared TiO2 NWs showed a lower generation of electron–hole pairs (Figure a), C-doping reduced the resistance of carrier migration and enhanced the transport of electron–hole pairs.

Photoluminescence (PL) spectra were measured to approximately characterize the recombination rate of photoinduced charges; a higher PL intensity normally indicates a higher carrier recombination. As shown in Figure c, the high intensity in P25 verified the high recombination rate of photoinduced carriers, which was in accordance with the results of photocurrent in Figure a. The intensity was largely reduced in TiO2 NWs and C-TiO2 (0.2, 0.5, 1.0%). The lower PL intensity of C-doped samples revealed an efficient separation of charge carriers. Though doping with carbon black led to the reduction of photocurrent in the long-term operation, C-doping reduced the electron–hole recombination and enhanced the carrier migration.

Conclusions

In summary, solar light-activated TiO2 nanorods were fabricated by doping carbon black into solid-state TiO2 NWs under thermal heating treatment. The effects of doping content on the surface morphology, crystal structure, photocatalytic activity under both UV and solar light, and photoelectrochemical property were systematically studied. Results showed that doping with C caused peak shifting and widening, and this phenomenon became more apparent with increasing doping loading of C. The doping of C also caused the transfer of TiO2 NWs into short nanorods, and the aspect ratio decreased with the increment of the doping content of C. The chemical composition of TiO2 after doping with C showed that C entered the crystal lattice and alternated O to generate O–Ti–C and oxygen vacancies. The photocatalytic activities of doped TiO2 were investigated under full-spectra solar light and UV light by degrading methylene blue. Comparing with P25 and TiO2 NWs, UV and solar light photocatalytic activity can be enhanced through C-doping, and the degradation efficiency of MB under solar light reached as high as 0.0328 min–1 in (1.0% C)-TiO2. Charge scavenger tests indicated that the major active species were OH– and photogenerated holes during the photodegradation of methylene blue. The EIS Nyquist spectra confirmed a faster surface charge charier transport, and the PL spectra demonstrated an efficient charge carrier separation in C-doped TiO2 catalysts, which contributed to the enhancement of UV and solar light photocatalysis.

Experimental Section

Chemicals

Commercial P25 (Evonik, composites of rutile and anatase), NaOH (Sigma-Aldrich), N2 (99.9%), carbon black (99.95%, Sigma-Aldrich), deionized water (DI water, supplied by the lab), HCl (20 wt %, Chemistore University of Waterloo), and methylene blue (MB, Sigma-Aldrich).

Synthesis of Materials

Synthesis of TiO2 Nanowires (TiO2 NWs)

Commercial P25 was employed as the initial raw material to prepare TiO2 nanowires, which were synthesized in an alkali hydrothermal solution. Typically, 2.0 g of P25 powder was completely dispersed in a 10 M NaOH solution, followed by magnetic stirring for 1 h. Then, it was put into a Teflon autoclave and reacted under 260 °C for 24 h. The precursor was collected by centrifugation after the complete reaction and washed with DI water and 0.1 M HCl solution for complete ion exchange. After that, the as-prepared product was placed in an oven to dry completely at 80 °C. The dried powder was then put in a ceramic crucible and thermally treated in a muffle furnace under 700 °C for 2 h to obtain the anatase TiO2 nanowires, which were denoted as TiO2 NWs.

Synthesis of Carbon Black-Doped TiO2 (C-TiO2) Nanorods

The as-prepared TiO2 NWs were used as a titanium source and carbon black (99.95%, Sigma-Aldrich) was used as a C source (the weight ratios between carbon black and TiO2 NWs are 0, 0.2, 0.5, and 1.0%). The obtained specimens were labeled as TiO2 NWs, (0.2% C)-TiO2, (0.5% C)-TiO2, and (1.0% C)-TiO2. In a typical experiment, carbon black nanoparticles and the as-prepared TiO2 NWs were mixed homogeneously in anhydrous ethanol with magnetic stirring. After drying at 80 °C for 12 h, the mixture was transferred into an airtight tube furnace and thermally treated at 700 °C for 2 h under a nitrogen (N2, 99.9%) atmosphere.

Characterization

The phase composition and crystal behavior of TiO2 NWs and C-TiO2 nanorods were characterized by X-ray diffraction (XRD, Bruker D8 FOCUS) and Raman spectroscopy (Raman, equipped with a He–Ne ion laser at 633 nm, Renishaw, U.K.). The microstructure of the synthesized samples was characterized by field emission scanning electron microscopy (SEM, LEO-Ultra, Gemini, Germany). The surface chemical composition of these specimens was characterized using X-ray photoelectron spectroscopy (XPS, Thermo VG Scientific ESCALAB 250, Al Kα radiation, 5 × 10–9 mbar of chamber vacuum, 0.2 mA of emission current) equipped with a hemispherical analyzer (150 nm in radius). The surface composition of these as-prepared specimens was calibrated according to the standard C 1s at a fixed binding energy of 284.6 eV before fitting into the point peak software (CasaXPS). The optical property and band gap energy of the prepared samples were measured by an ultraviolet–visible diffuse reflection spectroscope (UV-DRS, UV-2501PC, Shimadzu, Japan, 200–800 nm) with solid BaTiO3 as a reference. The photoelectric properties of the specimens were characterized by an Autolab Electrochemical workstation (Nova2.1.1 software, Metrohm, China). EIS was measured by a three-electrode system, glassy carbon electrode was selected as a counter electrode, saturated calomel electrode as a reference electrode, stainless steel mesh (10 μm) as the base of the working electrode, a 1.0 M KOH solution as an electrolyte, and the light source was a 300 W xenon lamp (PLS–SXE300/300UV, PerfectLight, Beijing).[57,58] During the test of AC impedance, the frequency range and amplitude were set at 100–10 MHz and 0.01 V, respectively. The PL spectra were measured by a steady/transient fluorescence spectrometer (FLS 980-STM, Edinburgh, U.K.), and the excitation wavelength was set at 250 nm.

Photocatalytic Performance Characterization

The photocatalytic property of TiO2 NWs and C-TiO2 nanorods was evaluated by the degradation of methylene blue (MB, Sigma-Aldrich) solution under UV light and solar light. Before adding TiO2 and C-doped TiO2 into the methylene blue solution, a photolysis experiment was carried out to verify if methylene blue can be degraded by solar light or UV light. The result showed that methylene blue cannot be degraded by UV and solar light (Figure S6). Then, TiO2 photocatalysts were added into the methylene blue solution. In a typical test, 20 mg of TiO2 NWs and C-TiO2 nanorods were dispersed into 50 mL of a 0.03 mM MB solution, respectively. Adsorption equilibrium was obtained before illumination under magnetic stirring for 1 h in a dark environment. After 1 h adsorption in the dark environment, the adsorption reached the equilibrium state because the adsorption lines overlapped at varied adsorption time (Figure S7). Photocatalytic degradation was carried out under the irradiation of a UV lamp (Philips, the Netherlands, 400 W, 365 nm) or a xenon lamp (Newport, 4.4 W/m2,150 W). In a typical degradation experiment, 5 mL aliquots were taken out (the left solution was vigorously stirred to maintain the homogeneous dispersion of the nanorods) using a pipette at specific time points of 0, 10, 20, 30, 40, 50, 60, 80, 100, and 120 min to characterize the degradation rate of MB by UV–Vis spectroscopy (UV-2501PC, Shimadzu, Japan) between 800 and 200 nm.

The photoactivity of C-TiO2 nanorods was quantitatively evaluated by calculating the apparent reaction rate constant (Kap) via the pseudo-first-order kinetic reactionwhere c0 and c are the original and final concentrations of the MB solution, respectively, and t stands for the relative irradiation time. The generated active species were characterized using different types of scavengers. In this work, AgNO3, n-butyalcohol, and (NH4)2C2O4 were used as e– scavenger, OH– scavenger, and h+ scavenger, respectively, during the degradation of 0.03 mM MB under UV light irradiation.