A Facile Strategy for the Preparation of N-Doped TiO2 with Oxygen Vacancy via the Annealing Treatment with Urea

Although titanium dioxide (TiO2) has a wide range of potential applications, the photocatalytic performance of TiO2 is limited by both its limited photoresponse range and fast recombination of the photogenerated charge carriers. In this work, the preparation of nitrogen (N)-doped TiO2 accompanied by the introduction of oxygen vacancy (Vo) has been achieved via a facile annealing treatment with urea as the N source. During the annealing treatment, the presence of urea not only realizes the N-doping of TiO2 but also creates Vo in N-doped TiO2 (N-TiO2), which is also suitable for commercial TiO2 (P25). Unexpectedly, the annealing treatment-induced decrease in the specific surface area of N-TiO2 is inhibited by the N-doping and, thus, more active sites are maintained. Therefore, both the N-doping and formation of Vo as well as the increased active sites contribute to the excellent photocatalytic performance of N-TiO2 under visible light irradiation. Our work offers a facile strategy for the preparation of N-TiO2 with Vo via the annealing treatment with urea.


Introduction
Titanium dioxide (TiO 2 ) has been widely investigated because of its excellent chemical stability, nontoxicity, and low cost [1,2].Although TiO 2 shows great potential in many fields [3][4][5], it can only absorb ultraviolet (UV) light [6], which limits its sufficient absorption of solar light because visible light (43%) takes up the majority of the solar spectrum [7].In addition, the fast recombination of photogenerated charge carriers in TiO 2 impedes its efficient use of solar energy [8,9].Thus, in order to improve the photocatalytic performance of TiO 2 , not only does the photoresponse range need to be extended but also the recombination of photogenerated charge carriers needs to be inhibited [10,11].
To extend the photoresponse range of TiO 2 from UV light to visible light, the doping strategy is commonly applied to adjust the intrinsic wide bandgap of TiO 2 [12,13].Among various doping strategies, it is reported that nitrogen (N) doping is an effective approach to reduce the bandgap of TiO 2 and enables its absorption of visible light [14][15][16].Generally, the preparation of N-doped TiO 2 can be realized by the annealing treatment with the presence of additional N sources such as urea and ammonia [17][18][19].To reduce the recombination of photogenerated charge carriers, the formation of defective structure is reported to be a useful strategy [6,20].The introduction of point defects into TiO 2 , such as oxygen vacancy (Vo), could trap the photogenerated electron inhibiting the recombination of the photogenerated charge carriers [10,21].Although many methods have been reported for the creation of Vo in TiO 2 [22][23][24], the annealing treatment of TiO 2 with an organic additive is reported to be an easy approach for the creation of Vo [25,26].Thus, it is of great interest to realize both the N-doping and creation of Vo to enhance the photocatalytic performance of TiO 2 .
To realize both the N-doping and introduction of Vo in one step, in this work, a facile strategy has been developed to prepare N-doped TiO 2 accompanied by the formation of Vo via the annealing treatment with urea.The effects of the annealing treatment on the photoresponse property and crystal structure of TiO 2 have been investigated as a function of the amount of urea.The photocatalytic performance of as-synthesized TiO 2 photocatalysts has been evaluated by the photocatalytic degradation of organic pollutants under visible light irradiation.

Materials
Hydrothermally prepared TiO 2 was obtained according to the reported synthetic procedures [27].Commercially available TiO 2 (P25, Degussa, Evonik, Resource Effiency GmbH, Essen, Germany) was purchased and used without further treatment.Custom-built high-borosilicate glass tubes were used for the annealing treatment.Methyl orange (MO, AR grade, Beijing Chemical Plant Co., Beijing City, China) was chosen as the model organic pollutant for evaluating the photocatalytic performance of all photocatalysts.

Photocatalytic Performance
Photocatalytic degradation of MO was carried out under visible light irradiation (Xenon lamp, 300 W, PerfectLight, Beijing, China) with a long-wavelength pass filter (>420 nm).TiO 2 photocatalysts (25 mg, 1 mg/mL) were added to aqueous solutions of MO (25 ppm, 25 mL) and stirred in dark for 60 min.The concentration variation of MO as a function of irradiation time was monitored by measuring its characteristic peak centered at 464 nm.The photocatalytic degradation ratio of MO was estimated by the expression: (C 0 − C)/C 0 × 100%, where C 0 is the initial concentration of MO and C corresponds to the concentration of MO at different time intervals.

Characterizations of TiO 2 Photocatalysts
As shown in Figure 1, hydrothermally prepared TiO 2 appears as a white powder before the annealing treatment (Figure 1a and Figure S1a).In the absence of urea, TiO 2 Nanomaterials 2024, 14, 818 3 of 8 turns to a dark grey powder after the annealing treatment (Figure 1b and Figure S1e) due to the pyrolysis of the organic solvent [27,28].With the presence of urea (Figure S1b-d), TiO 2 is transformed to a brown powder after the annealing treatment (Figure S1f-h).This obvious color change indicates that N-doping of TiO 2 may happen during the annealing treatment with urea as the N source [29,30].Although the amount of urea increases, the colors of all N-TiO 2 samples are similar, implying that a low amount of urea is enough for N-doping (Figure 1c-e).This facile strategy is also suitable for the preparation of N-doped P25 (Figure S2) and white P25 changes to a brown powder after the annealing treatment with urea (Figure S3).Since the annealing treatment results in no clear color change of urea (Figures S4 and S5), it is reasonable to propose that the pyrolysis of urea leads to the N-doping of TiO 2 during the annealing treatment.

Characterizations of TiO2 Photocatalysts
As shown in Figure 1, hydrothermally prepared TiO2 appears as a white powder before the annealing treatment (Figures 1a and S1a).In the absence of urea, TiO2 turns to a dark grey powder after the annealing treatment (Figures 1b and S1e) due to the pyrolysis of the organic solvent [27,28].With the presence of urea (Figure S1b-d), TiO2 is transformed to a brown powder after the annealing treatment (Figure S1f-h).This obvious color change indicates that N-doping of TiO2 may happen during the annealing treatment with urea as the N source [29,30].Although the amount of urea increases, the colors of all N-TiO2 samples are similar, implying that a low amount of urea is enough for N-doping (Figure 1c-e).This facile strategy is also suitable for the preparation of N-doped P25 (Figure S2) and white P25 changes to a brown powder after the annealing treatment with urea (Figure S3).Since the annealing treatment results in no clear color change of urea (Figures S4 and S5), it is reasonable to propose that the pyrolysis of urea leads to the N-doping of TiO2 during the annealing treatment.To investigate the effects of the annealing treatment on the photoresponse property of TiO2, the DRS spectra of all samples were recorded.According to Figure 2a, both TiO2 and A-TiO2 show clear UV absorption and a negligible visible light response.This result is in agreement with the color of TiO2 and the reported phenomenon [27].Conversely, all N-TiO2 samples present obvious visible light absorption with almost identical absorbance, which further proves that a low amount of urea is enough for N-doping.The annealing treatment of TiO2 with urea definitely expands the photoresponse range of TiO2 to the visible light range suggesting that N-doping occurs [13,31].To further confirm the N-doping of TiO2, N1s XPS fine spectra of all samples were collected.Compared with TiO2 and A-TiO2, all N-TiO2 samples show a clear N1s peak which undoubtedly proves that the Ndoping of TiO2 occurs [32,33].In addition, the peak intensity increases as the amount of urea goes up, implying an increase in the N-doping level [18,34].Thus, the annealing treatment of urea offers a facile approach for the preparation of N-doped TiO2 to extend the photoresponse range of TiO2.To investigate the effects of the annealing treatment on the photoresponse property of TiO 2 , the DRS spectra of all samples were recorded.According to Figure 2a, both TiO 2 and A-TiO 2 show clear UV absorption and a negligible visible light response.This result is in agreement with the color of TiO 2 and the reported phenomenon [27].Conversely, all N-TiO 2 samples present obvious visible light absorption with almost identical absorbance, which further proves that a low amount of urea is enough for N-doping.The annealing treatment of TiO 2 with urea definitely expands the photoresponse range of TiO 2 to the visible light range suggesting that N-doping occurs [13,31].To further confirm the N-doping of TiO 2 , N1s XPS fine spectra of all samples were collected.Compared with TiO 2 and A-TiO 2 , all N-TiO 2 samples show a clear N1s peak which undoubtedly proves that the N-doping of TiO 2 occurs [32,33].In addition, the peak intensity increases as the amount of urea goes up, implying an increase in the N-doping level [18,34].Thus, the annealing treatment of urea offers a facile approach for the preparation of N-doped TiO 2 to extend the photoresponse range of TiO 2 .To further study the influences of the annealing treatment on the crystal structures of TiO2, XRD measurements were conducted and the results are shown in Figure 3.All the characteristic diffraction patterns of TiO2 photocatalysts are in accordance with the diffraction peaks of anatase, proving that the crystal phase of all TiO2 samples is anatase To further study the influences of the annealing treatment on the crystal structures of TiO 2 , XRD measurements were conducted and the results are shown in Figure 3.All the characteristic diffraction patterns of TiO 2 photocatalysts are in accordance with the diffraction peaks of anatase, proving that the crystal phase of all TiO 2 samples is anatase (JCPDS-21-1272) [25,35].In the absence of urea, the crystallinity of TiO 2 is improved after the annealing treatment as indicated by the increased intensity of the diffraction patterns which correspond to the (101) and (200) crystal planes.However, the crystallinity of all N-TiO 2 samples is similar to that of TiO 2 after the annealing treatment with urea.This unexpected phenomenon suggests that N-doping hampers the further crystallization of N-TiO 2 which may enlarge its specific surface area [3,36].In addition, new diffraction patterns appear close to the (101) crystal plane of N-TiO 2 (1:2), suggesting that a low amount of urea is enough for the N-doping.To further study the influences of the annealing treatment on the crystal structures of TiO2, XRD measurements were conducted and the results are shown in Figure 3.All the characteristic diffraction patterns of TiO2 photocatalysts are in accordance with the diffraction peaks of anatase, proving that the crystal phase of all TiO2 samples is anatase (JCPDS-21-1272) [25,35].In the absence of urea, the crystallinity of TiO2 is improved after the annealing treatment as indicated by the increased intensity of the diffraction patterns which correspond to the (101) and (200) crystal planes.However, the crystallinity of all N-TiO2 samples is similar to that of TiO2 after the annealing treatment with urea.This unexpected phenomenon suggests that N-doping hampers the further crystallization of N-TiO2 which may enlarge its specific surface area [3,36].In addition, new diffraction patterns appear close to the (101) crystal plane of N-TiO2 (1:2), suggesting that a low amount of urea is enough for the N-doping.The BET surface areas of all samples were measured to further analyze the effect of the annealing treatment on TiO2.From Table 1, it is clear that the BET surface area of hydrothermally prepared TiO2 (134.67 m 2 /g) is the largest whereas that of A-TiO2 is the smallest (40.21 m 2 /g).Compared with TiO2, the specific surface areas of N-TiO2 samples decrease and a decreasing tendency is observed as the amount of urea increases.Clearly, the annealing treatment of TiO2 with urea inhibits its crystallization which is in agreement with the results of XRD (Figure 3).For the moment, the reason for this phenomenon has The BET surface areas of all samples were measured to further analyze the effect of the annealing treatment on TiO 2 .From Table 1, it is clear that the BET surface area of hydrothermally prepared TiO 2 (134.67 m 2 /g) is the largest whereas that of A-TiO 2 is the smallest (40.21 m 2 /g).Compared with TiO 2 , the specific surface areas of N-TiO 2 samples decrease and a decreasing tendency is observed as the amount of urea increases.Clearly, the annealing treatment of TiO 2 with urea inhibits its crystallization which is in agreement with the results of XRD (Figure 3).For the moment, the reason for this phenomenon has been reported and it may result from the N-doping induced by the annealing treatment [3].This phenomenon is also observed in the P25 samples obtained after the annealing treatment with urea (Table S1).As a result, the annealing treatment with urea offers a facile approach for the preparation of N-doped TiO 2 with more active sites reserved.To further investigate the influences of the annealing treatment on the crystal structure of TiO 2 , EPR spectra of all samples were recorded and are shown in Figure 4.As shown in Figure 4, the characteristic peak with a g value of 2.002 corresponds to Vo [20,37].Compared with TiO 2 , A-TiO 2 contains a certain amount of Vo which may result from the both the crystallization of TiO 2 and the pyrolysis of the organic solvents [25,26].With the presence of urea, more Vo is introduced into the N-doped TiO 2 and its amount increases first and then decreases as the amount of urea increases.This result may be due to the N-doping process which not only induces the formation of Vo but can also occupy the Vo.The formation of Vo in N-doped TiO 2 may contribute to the separation of photogenerated charge carriers [38,39].Thus, the annealing treatment with urea not only induces the N-doping to extend the photoresponse range of TiO 2 but also creates Vo in N-doped TiO 2 which contributes to the separation of photogenerated charge carriers, which could both favor the enhancement of the photocatalytic performance of TiO 2 .
N-TiO2 (1:1) 73.36 N-TiO2 (1:1) 51.87 To further investigate the influences of the annealing treatment on the crystal structure of TiO2, EPR spectra of all samples were recorded and are shown in Figure 4.As shown in Figure 4, the characteristic peak with a g value of 2.002 corresponds to Vo [20,37].Compared with TiO2, A-TiO2 contains a certain amount of Vo which may result from the both the crystallization of TiO2 and the pyrolysis of the organic solvents [25,26].With the presence of urea, more Vo is introduced into the N-doped TiO2 and its amount increases first and then decreases as the amount of urea increases.This result may be due to the Ndoping process which not only induces the formation of Vo but can also occupy the Vo.The formation of Vo in N-doped TiO2 may contribute to the separation of photogenerated charge carriers [38,39].Thus, the annealing treatment with urea not only induces the Ndoping to extend the photoresponse range of TiO2 but also creates Vo in N-doped TiO2 which contributes to the separation of photogenerated charge carriers, which could both favor the enhancement of the photocatalytic performance of TiO2.

Photocatalytic Performance
The photocatalytic performance of all TiO2 photocatalysts was evaluated by the photocatalytic degradation of MO under visible light irradiation.From Figure 5, all TiO2 photocatalysts are capable of adsorbing a certain amount of MO and N-TiO2 (1:2) shows (1:1) the highest adsorption capacity for MO (12.9%).Compared with TiO2, the specific surface area of N-TiO2 (1:2) is smaller and its improved adsorption capacity of MO may be due to the formation of functional groups, which is in agreement with the XRD analysis (Figure 3).During visible light irradiation, both TiO2 and A-TiO2 are not able to degrade MO because they do not absorb the visible light (Figure 2a).With N-TiO2 as the photocatalyst,

Photocatalytic Performance
The photocatalytic performance of all TiO 2 photocatalysts was evaluated by the photocatalytic degradation of MO under visible light irradiation.From Figure 5, all TiO 2 photocatalysts are capable of adsorbing a certain amount of MO and N-TiO 2 (1:2) shows (1:1) the highest adsorption capacity for MO (12.9%).Compared with TiO 2 , the specific surface area of N-TiO 2 (1:2) is smaller and its improved adsorption capacity of MO may be due to the formation of functional groups, which is in agreement with the XRD analysis (Figure 3).During visible light irradiation, both TiO 2 and A-TiO 2 are not able to degrade MO because they do not absorb the visible light (Figure 2a).With N-TiO 2 as the photocatalyst, the photocatalytic degradation of MO is feasible and among the samples N-TiO 2 (2:1) presents the best photocatalytic performance.The photocatalytic performance of N-TiO 2 proves that the annealing treatment with urea extends the photoresponse range of TiO 2 to the visible light range [18,40].The photocatalytic performance of N-TiO 2 is comparable to the reported results (Table S2) which may be ascribed to the formation of Vo and the increased specific surface area [33,39,41,42].Based on the photocatalytic performance, the optimal weight ratio of TiO 2 to urea for the preparation of N-TiO 2 via the annealing treatment is found to be 2:1.
proves that the annealing treatment with urea extends the photoresponse range of TiO2 to the visible light range [18,40].The photocatalytic performance of N-TiO2 is comparable to the reported results (Table S2) which may be ascribed to the formation of Vo and the increased specific surface area [33,39,41,42].Based on the photocatalytic performance, the optimal weight ratio of TiO2 to urea for the preparation of N-TiO2 via the annealing treatment is found to be 2:1.

Conclusions
In summary, a facile strategy was developed for the preparation of N-TiO2 via the annealing treatment with urea.On the one hand, the photoresponse of TiO2 is extended by N-doping via the annealing treatment with urea as the N source.On the other hand, Vo is introduced into N-TiO2 which may contribute to the separation of photogenerated charge carriers.In addition, the specific surface area of N-TiO2 is enlarged with the presence of urea during the annealing treatment by inhibiting the crystallization of TiO2.Thus, more active sites could be reserved for photocatalytic reactions.All the above favorable aspects induced by the annealing treatment with urea contribute to the excellent photocatalytic performance of N-TiO2.This facile strategy is also suitable for other TiO2 photocatalysts such as P25 and, thus, our work offers a universal approach for the preparation of N-doped TiO2 via the annealing treatment with urea.The annealing treatment with other additives for different elements doping, not merely N doping, may be also possible and further work is underway.