The Synergistic Effect of Nitrogen Dopant and Calcination Temperature on the Visible-Light-Induced Photoactivity of N-Doped TiO 2

were characterized using X-ray diffraction spectrometer, high-resolution transmission electron microscope, X-ray photoelectron spectroscopy, UV-vis diffusion reflectance spectroscopy, and Brunauer-Emmett-Teller specific surface area. Finding showed that the photocatalytic activity ofN-dopedTiO 2 was greatly enhanced compared to pure TiO 2 under visible irradiation. N dopants could retard the transformation from anatase to rutile phase. Namely, N-doping effect is attributed to the anatase phase stabilization.The results showed nitrogen atoms were incorporated into the interstitial positions of the TiO 2 lattice. Ethylene was used to evaluate the photocatalytic activity of samples under visible-light illumination. The results suggested good anatase crystallization, smaller particle size, and larger surface are beneficial for photocatalytic activity of N-doped TiO 2 .TheN-doped TiO 2 catalyst prepared with ammonia to titanium isopropoxide molar ratio of 2.0 and calcinated at 400C showed the best photocatalytic ability.


Introduction
Titanium dioxide (TiO 2 ) is an effective photocatalyst due to its inexpensiveness, chemical stability, nontoxicity, biological and chemical inertness, and long-term stability against photo-corrosion and chemical corrosion [1,2].In the last two decades, many works have been done on expanding and augmenting the utility of TiO 2 in heterogeneous photocatalysis (e.g., water purification, air cleaning, and water splitting for the production of hydrogen) [2][3][4][5], self-cleaning surfaces [6], inactivation of bacteria and fungi [7,8], solar cells [9], anticorrosive coatings [10], and Li-ion batteries [11].TiO 2 has three phases in nature, brookite (orthorhombic), anatase (tetragonal), and rutile (tetragonal).It is amorphous at temperature up to 300 ∘ C, becoming anatase and rutile typically at 350 and 800 ∘ C, respectively [3].Anatase powder with good crystallinity, small grain size, and high specific surface area is desirable as long as the photocatalytic activity is concerned.However, its wide band gap of 3.2 and 3.0 eV for anatase and rutile polymorphs, respectively, requires UV light for the excitation of electron-hole pairs.It only uses 4-5% of the UV light in the solar irradiation.To utilize solar energy effectively, many attempts have been made to modify the properties of TiO 2 , such as doping with transition metal ions [12][13][14] or nonmetal anions [3,[15][16][17][18][19][20][21][22][23], and sensitization with organic dyes [24,25].Among the nonmetal-doping TiO 2 photocatalysts, the simplest and most feasible TiO 2 modification approaches for achieving visible-light-driven photocatalysis seem to be N-doping, that is, doping nitrogen atoms into interstitial (or substitutional) sites in the crystal structure of TiO 2 .Sato [4] was the first to report on nitrogendoped TiO 2 .It was treated with various nitrogen sources such as urea, ammonia, ammonium chloride, and nitric acid.Asahi et al. have prepared TiO 2− N  films whose International Journal of Photoenergy absorption activity not only was shifted toward the longer visible wavelength but also had higher thermal stability and less recombination centers of charge carries than the metaldoped TiO 2 [5].However, to date, little research has been conducted on gaining in-depth information on the properties of the interstitial (or substitutional) N-doped TiO 2 , as well as its photocatalytic activity.Different dopants result in TiO 2 of different properties and consequently alter the photocatalytic activity of the materials.Currently, few works have focused on the synergistic effects of nitrogen dopant and calcination temperature on the characteristics and visible-light-induced photoactivity of N-doped TiO 2 [6].
In this study, we focused on sol-gel synthesis process for the preparation of nitrogen-doped TiO 2 catalysts (N-TiO 2 ) because the wet processes have the advantages of low cost, ease of scale-up, and stablility for practical applications.The effects of N atom and calcination temperature on the structural, optical, and photocatalytic properties of N-TiO 2 were assessed.The gel samples were calcined at 400, 500, 600, 700, and 800 ∘ C. The photocatalysts were characterized using Xray diffraction (XRD), high-resolution transmission electron microscope (HRTEM), X-ray photoelectron spectroscopy (XPS), UV-vis diffusion reflectance spectroscopy (UV-vis DRS), Brunauer-Emmett-Teller (BET) specific surface area ( BET ), and thermogravimetric-differential thermal analysis (TG-DTA).The transition from amorphous to anatase and rutile phases at different calcination temperatures was also explored.Ethylene was chosen as a probe reactant because it is structurally simple, has relatively high reactivity, and is the parent compound of many widespread volatile organic compounds (VOCs) of environmental concern (e.g., TCE and tetrachloroethylene).The object of this study is to investigate the synergistic effects of nitrogen content and calcination temperatures in the synthesis process of N-doped TiO 2 prepared by sol-gel process and to investigate the influence of ethylene on photocatalytic decomposition.

Preparation of N-Doped TiO 2
Catalyst.Titanium isopropoxide (TTIP, C 12 H 28 O 4 Ti, 97%, Aldrich), a highly reactive alkoxide, was used as precursor.Aqua ammonia (NH 4 OH) was used as nitrogen source.A given amount of aqua ammonia was added to the distilled water while being stirred until it completely dissolved at room temperature.Then, the solution was added dropwise to a constant amount of titanium isopropoxide as to obtain an aqua ammonia to TTIP molar ratio (A/T) of 1 : 1; the TiO 2 sample was designated as N-TO 2 .While the mixture was being stirred vigorously, the urea solution was introduced to the TTIP.Hydrolysis and ploycondensation reactions occurred.The reactant was stirred continuously for one hour after all aqua ammonia solution was dissipated.The sol was aged in air for 24 hr to allow further hydrolysis.The final sol was left on the bench top to allow thickening.After aging, the sol was dried at 70 ∘ C for 24 hr to evaporate the solvent.The residual xerogel was crushed to fine powder before calcined in a furnace at 400, 500, 600, 700, and 800 ∘ C, respectively, for 3 hr.Pure TiO 2 powder was also prepared without the presence of aqua ammonia following the same procedures as described above.The nitrogen-doped samples were labeled AxTy, where  and  refer to the aqua ammonia to TTIP molar ratio (A/T) and calcining temperature, respectively.For example, A20T4 means the N-TiO 2 prepared with A/T molar ratio 2.0 and calcinated at 400 ∘ C.

Characterization.
The  BET of the samples was determined using nitrogen adsorption method measured with a Micromeritics ASAP 2020 adsorption apparatus.Pore size and pore volume were calculated by the BJH isotherm.The crystallization behavior and stability of N atoms in N-TiO 2 were monitored using a TG-DTA (STA 6000, PerkinElmer), while the sample was being heated from room temperature to 1000 ∘ C at 10 ∘ C/min.The morphology, structure, and grain size of the samples were characterized by a high-resolution transparent electromicroscopy (HRTEM, JEM-2010, JEOL).To study the optical response of N-TiO 2 samples ( = 200-800 nm), UV-vis DRS (U3900H, Hitachi) were recorded using a spectrophotometer with an integrating sphere attachment and Al 2 O 3 was used as the reference.The chemical environment information and the N concentration of all of the samples were measured by X-ray photoelectron spectra (XPS) (PHI 5000, VersaProbe/Scanning ESCA Microprobe with a C 60 ion gun).All samples were collected in vacuum and transferred to the main ultrahigh-vacuum (UHV) chamber for measurements.The shift of binding energy due to relative surface charging was corrected to the C 1s level at 285 eV as an internal standard.The XPS peaks were assumed to have Gaussian line shapes and were resolved into components by a nonlinear least-squares procedure after proper subtraction of the baseline.The crystallinity and phase of the samples were analyzed by XRD (PANalytical X'Pert-Pro MPD PW3040/60) with Cu-Ka radiation at a scan rate of 0.05 ∘ 1/s in the range from 20 to 80 ∘ .

Photooxidation Performance.
The experiment setup and operating conditions for photocatalytic activity tests were the same as those reported previously [7].The catalyst was prepared by depositing suspension of the as-prepared catalyst at the bottom surface of the reactor that was subsequently dried at ambient condition.The obtained catalyst film density on the bottom surface of reactor was kept at 4 mg/cm 2 .The experiment setup and operating conditions for photocatalytic activity tests were the same as those in our previous reports [7].The photocatalytic activity experiments of the as-prepared catalysts for the oxidation of ethylene in gas phase were performed at room temperature using a 250 mL photocatalytic reactor.Air with 60% relative humidity was passed through the reactor with the sample inside for 1 hr to reach equilibrium, and then a desired amount of ethylene was injected into the reactor with a gastight syringe.The ethylene inside the reactor was taken at regular intervals with gas-tight syringe and analyzed using gas chromatographs (PerkinElmer, Clarus 500, USA) equipped with a flame ionization detector (FID) and a thermal conductivity detector (TCD) detection.The GC instrument was equipped with an Elite-Plot-Q capillary column (Agilent Technologies).The ethylene was allowed to reach adsorption equilibrium with the catalyst in the reactor prior to irradiation.Two 500-W horizontal halogen lamps (9.5 cm in length, PHILIPS) were used to simulate the solar light.The lamp equipped with a cut-off filter ( > 400 nm) was vertically placed outside the photocatalytic reactor.Light intensity at the catalyst surface was 16 mW/cm 2 .Blank experiments using an uncoated (no catalyst) glass plate had no effect on the ethylene concentration and did not generate any carbon monoxide or carbon dioxide (see supplementary material online at http://dx.doi.org/10.1155/2013/268723).The photocatalytic activity of the catalyst can be quantitatively evaluated by comparing the apparent reaction rate constants.

Results and Discussion
3.1.Thermal Analysis.To investigate the process of phase transformation of titanium(IV)-oxide-hydroxide to titanium oxide and the effect of nitrogen doping content on the transformation from anatase to rutile, thermoanalytic techniques including TGA and DTA were conducted.Typically, the hydrothermal weight loss is attributed to a multistep of polycondensation of titanium(IV)-oxide-hydroxide and phase transformation of anatase into rutile or brookite from room temperature to 700-800 ∘ C as shown in Figure 1 and Table 1.DTA measurements established the transformation of TiO  (OH)  into TiO 2 including two endothermic and one exothermic peaks from room temperature to 398 ∘ C (Figure 1).A sharp endothermic peaks at 105 ∘ C due to the release of absorbed water, while the other minor peak at 253 ∘ C is referred to as the desorption of organic.There was a corresponding weight loss from TGA of about 19.8 and 9.4 wt%, respectively.In the region mentioned above, the finely exothermic peak around 276 ∘ C is ascribed to the thermal decomposition of unhydrolyzed TTIP with about 0.8 wt% weight loss.The weak thermal effect at 276-730 ∘ C is accompanied by obvious exothermic peak at 451 ∘ C which can be assigned to the crystallization of the amorphous phase to anatase.At around 730 ∘ C, an obvious exothermic peak was observed owing to the oxidation of carbon residue and evaporation of chemisorbed water, which can be assigned to the phase transformation from anatase titania to rutile TiO 2 .It can be concluded from the DTA result that the asprepared TiO 2 sample was amorphous.According to TGA, the weight loss roughly consists of three distinct stages.The first stage was up to 276 ∘ C, over which the mass loss is the greatest.A mass loss of up to 30% was observed, which is attributed to the evaporation of the physical adsorbed water and the release of organic.The second stage is from 276 to 451 ∘ C, where the mass loss is about 1.7%.This corresponds to the thermal decomposition of unhydrolyzed TTIP.The third stage is from 451 to 730 ∘ C and the mass loss is about 0.7%, which is attributed to the removal of chemisorbed water.At about 730 ∘ C, no mass loss was observed.The exothermic peaks of N-TiO 2 prepared at 0.5, 1.0, and 2.0 A/T ratios were at 710, 720, and 730 ∘ C, respectively (Table 1).The results indicated that nitrogen atoms doped in TiO 2 could prevent phase transition of anatase to rutile, because as far as pure TiO 2 was concerned, most TiO 2 particles were transformed into rutile at 700 ∘ C.

Phase and Structure.
The characteristics of N-TiO 2 at various calcination temperatures and molar ratios of A/T are summarized in Table 1. Figure 2 shows the XRD pattern of the N-TiO 2 samples at various A/T ratios and annealed at different temperatures.The presence of peaks (2 = 25.44 ∘ , 38.06 ∘ , and 48.24 ∘ ) was regarded as an attributive indicator of anatase titania, and the presence of peaks (2 = 27.54∘ , 36.14 ∘ , and 41.20 ∘ ) was regarded as an attributive indicator of rutile titania.The diffraction peaks for the anatase (JCPDS no.21-1272) and rutile (JCPDS no.21-1276) phases are marked with "A" and "R, " respectively, and the corresponding diffraction planes are given in parenthesis.No N-derived peaks were detected in all the patterns.This may be caused by the lower concentration of the doped species, and the limited dopants may have moved into either the interstitial positions or the substitutional sites of the TiO 2 crystal structure, which is consistent with [8,9].The average grain size was estimated at 28-131 nm, comparable to that from the HRTEM image.A comparison of XRD patterns clearly shows that nitrogen doping significantly changes the crystalline size and the crystal structure of TiO 2 .
The transformation from the amorphous to the crystalline form of TiO 2 usually requires temperatures close to 300 ∘ C [8].For the samples prepared with 2.0 A/T and calcinated at 400-500 ∘ C (Figure 2(a)), the intensities of the anatase peak are increased and the width of the (101) plane diffraction peak becomes narrower as the calcination temperature increases, implying an improvement and growth in crystallinity.Increasing the temperature to 800 ∘ C, the intensity of anatase peak decreases, but the rutile peaks appear, suggesting the transformation of anatase to rutile phase.The average crystalline size was 32, 85, and 103 nm and the mass percentage of anatase was 90, 60, and 6% for samples prepared at 2.0 A/T and calcination temperature of 400, 600, and 700 ∘ C, respectively.The maximum mean size of anatase and rutile crystallites was observed at 800 ∘ C. Results agreed with those reported that the crystalline size grows and the content of anatase diminishes as calcination temperature increases [8,[10][11][12][13].For the samples prepared at 0.5 A/T, the anatase peaks appear at temperatures 400 and 500 ∘ C. Increasing the temperature to 500 ∘ C, the intensity of anatase peak decreases, but the rutile peaks appear.When temperature is increased further to 600 ∘ C, rutile becomes a main phase.In this case, the anatase-to-rutile phase transformation temperature is between 400 and 500 ∘ C. XRD patterns of samples prepared at 3.0 A/T and annealed at 700 ∘ C show both anatase and a rutile phase, indicating that anatase-to-rutile phase transformation is shifted to higher temperature.It can be seen that the presence of nitrogen can restrain the formation and growth of TiO 2 crystal phase, thereby retarding the anatase-to-rutile phase transformation.This result is further supported by the appearance of rutile in XRD patterns of sample prepared at 0.5 and 1.5 G/T at 600 ∘ C and disappearance for samples prepared at G/T greater than 2.0 at 600 ∘ C (data not shown).Compared with P25, the peak position of the (101) plane of N-doped TiO 2 shifted slightly to higher 2 value, and the peak was broader as N/T ratio increases, suggesting distortion of the crystal lattice of TiO 2 by the incorporation of nitrogen.The crystalline sizes of samples calcinated at 400 ∘ C and prepared at 0.5, 1.5, 2.0, and 3.0 A/T were 28, 29, 32, and 37 nm, respectively.The average crystalline size of samples increases with increase in A/T, revealing that nitrogen can enhance the growth of N-TiO 2 crystals.Results agreed with those reported that the crystalline size grows as nitrogen to Ti proportion increases [14].The results reveal that the phase transformation temperature of anatase to rutile was progressively increased when the amount of N dopant was increased.Clearly, calcination temperature and A/T ratio play a significant role in the formed crystal structure and particle size (Figure 3).Higher temperature favors the growth of rutile structure and produces N-TiO 2 nanoparticles with larger particle size.Thus, a fine control of calcination temperature and A/T ratio is crucial for obtaining a pure phase of N-TiO 2 .
Further insights into the effect of calcination and doping on the morphology and structure of the N-TiO 2 samples can be obtained from HRTEM image (Figure 4).The selected area electron diffraction pattern and TEM confirmed its highly crystalline anatase or rutile structure.The observed d-spacing from HRTEM image is 3.522 Å and is comparable to 3.52 Å previously reported for (101) crystallographic plane of undoped anatase [15][16][17][18].The HRTEM image showed the nanocrystallines with primary grain size around 11-200 nm.From the TEM images, the particles in N-TiO 2 samples are monodispersed.The crystalline form aggregates with a mesoporous structure.The formation of mesoporous structure is similar to that reported in the literature [19,20].First, monodispersed amorphous titanium oxide sol particles are formed by the hydrolysis processes.Then the monodispersed sol particles are crystallized and aggregated under calcination treatment to form mesoporous crystalline.Based on the HRTEM results, the calcination temperature significantly changes the size of particles (Table 1).The grain size of crystalline is in consistent with the calculated value obtained from XRD patterns (Table 1).The grain size of samples increases with the calcination temperature.Morphological observation by TEM and XRD calculations reveals that the large particle size of A20T8 was mainly caused by agglomeration and the growth of crystallite size.are shown in Figure 5.The  BET of N-TiO 2 prepared with 2.0 A/T ratio at calcined 400, 500, 600, 700, and 800 ∘ C was 98, 62, 17, 4, and 2 m 2 /g, respectively.Obviously, the values of  BET and pore volume of N-TiO 2 prepared with the same A/T ratio decreased with increasing calcination temperatures.Such phenomenon could be attributed to the collapse of the mesoporous structure, the growth of TiO 2 crystallites, and the removal process of nitrogen core by high-temperature calcination.Consequently, N-TiO 2 calcined at 800 ∘ C has the least  BET .Based on the results obtained, the hysteresis loop of N-TiO 2 is significantly shifted in direction of higher relative pressures as calcination temperature increases, indicating much larger mesorpore.The average pore size of N-TiO 2 increased with increasing calcination temperature due to the formation of slit-like pores [21].

Surface Area and
The N-TiO 2 samples obtained at various calcination temperatures gave different nitrogen adsorption isotherms (Figure 5), implying differences in their porous structure.The N-TiO 2 samples prepared at low calcination temperature (i.e., 400 and 500 ∘ C) exhibited a type IV isotherm and a type H2 hysteresis loop at lower relative pressure region, which are typical characteristics of mesoporous structure with ink bottle pores.Note that the adsorption branches of these isotherms resembled type II, indicating the presence of some macropores.Otherwise the N-TiO 2 samples prepared at higher calcination temperature (i.e., 600-800 ∘ C) exhibited a type V isotherm and a type H3 hysteresis loop at higher relative pressure.The hysteresis loop at lower relative pressure region (0.4 < / 0 < 0.8) was attributed to a smaller mesopore, while that at the higher relative pressure (0.8 < / 0 < 1.0) was of larger mesopores.Thus samples prepared at high calcination temperature (>600 ∘ C) had a wide pore size distribution in the mesopores scale.The isotherm of N-TiO 2 prepared at higher calcination temperature was below that of lower calcination temperature (Figure 5), indicating lower surface area and pore volume of the former.Further observation from Table 1 indicated that  BET for the N-TiO 2 samples at the same A/T ratio increased with increasing calcination temperature (Figure 5(c)).Results reveal that all N-TiO 2 samples calcined at various temperatures show monomodal pore size distributions and mesopores may be assigned to the pores among interaggregated particles (data not shown).With calcination temperature increasing, there is a significant tendency of pore size distribution toward to the bigger pore size.The increase of pore size is due to the corruption of smaller pores during calcination as the smaller pores endured much greater stress than the bigger pores.Formation of bigger crystalline aggregation upon increase in temperature could form bigger pores also.Consequently, the pore size increases while the pore volume decreases with International Journal of Photoenergy calcinations temperature increasing.Based on the TEM and BET results, the mesopores of TiO 2 were decreased due to the heat shrinkage of calcination.The results indicated that calcination at high temperature and in the presence of nitrogen atoms will help to improve the thermal stability of the photocatalyst.While the calcination temperature was increased, both the specific surface area and the pore volume of N-TiO 2 samples decreased.This indicated that the average pore size increased while the pore volume and the specific surface area decreased with calcination temperature increasing.

Optical Properties.
The typical UV-vis diffuse reflectance spectra of N-TiO 2 are shown in Figures 6(a to that of the TiO 2 (i.e., P25).This phenomenon should be ascribed to either the nitrogen doping and/or sensitization by a surface-anchored group.The calcination temperature and A/T ratio are the crucial factor that can further affect the absorption as seen in Figure 6.Similar phenomenon had been reported by previous publications [17,[22][23][24][25].The absorption edges of the N-TiO 2 samples show continuous red shift as the calcination temperature changes from 400 to 700 ∘ C, which can be assigned to the intrinsic band gap absorption of TiO 2 .The shift is too minor to be observed when the calcination temperature changes from 500 to 600 ∘ C, which is possibly caused by the synergetic effect of the loss of the nitrogen and the increase of the TiO 2 crystallinity during the calcination process.The absorption threshold (  ) can be determined using the onset of the absorption edges by the section of the fitting lines on the upward and outward section of the spectrum (shown in the graph inserted in Figure 6) [23].Table 1 lists the determined   (nm) and the derived band gap energy values.The band gap of N-TiO 2 (  ) was calculated using the   = 1240/  equation.It also can be seen that N-TiO 2 samples have two characteristic light absorption edges.One of them corresponds to the band gap of TiO 2 ( 1 = 390∼430 nm) while the other originates from the N-induced midgap level ( 2 = 505∼517 nm).The color of the samples was different with calcination temperature and A/T ratio.The color of N-TiO 2 samples prepared with 2.0 A/T ratio and calcinated at 400, 600, and 700 ∘ C was vivid yellow, yellow, and white, respectively.There is no  2 when N-TiO 2 calcinated above 600 ∘ C. It can be ascribed to the amount of N dopant decreased.The color of N-TiO 2 prepared with 0.0, 1.0, 2.0, and 3.0 A/T ratio and calcinated at 400 ∘ C was white, pale yellow, yellow, and vivid yellow, respectively.The phenomenon that the A/T ratio increases the shifting of the spectra towards visible region was clearly observed (Figure 6(a)).The absorption threshold (  ) was increased with increasing A/T ratio.It reveals that more nitrogen doping in the matrix of TiO 2 structure results in a shift of absorbance region toward visible light wavelength.This is possible attribute to the nitrogen incorporated into TiO 2 lattice and formation of new electronic state above valence band caused by nitrogen doping to generate a red shift.

Analysis of Chemical State.
The chemical states of doping impurity are critical to the optical property, band gap, and photocatalytic activity of nitrogen-doped TiO 2 .To investigate the chemical states of the doping nitrogen, XPS spectra were applied to examine three regions: the Ti 2p core level near 460 eV, the O 1s core level near 530 eV, and the N 1s core level near 400 eV. Figure 7 shows the experimental observation of surface chemical composition and the electronic structures of N-TiO 2 samples using XPS.XPS peaks showed that the N-TiO 2 contained only Ti, O, N elements and a small quantity of carbon.The presence of carbon was ascribed to the residual carbon from the precursor solution and the adventitious hydrocarbon from the XPS instrument itself.In Figures 7(a) and 7(b), the binding energy (BE) peaks corresponding to N 1s core levels for N-TiO 2 are observed one major peak at 400 eV.The N 1s peak at 400 eV is ascribed to the presence of the oxidized nitrogen species, and the nitrogen may be incorporating into the interstitial positions of the TiO 2 lattice and to form a Ti-O-N structure [26,27].No other N 1s peaks were detected in N-TiO 2 samples.As shown in Figure 7, N 1s peak at 400 eV slightly decreased with increasing calcination temperature.This is different from the N-TiO 2 prepared by flame oxidation, in which nitrogen partially substitutes the oxygen sites in the TiO 2 [28].The preparation method plays an important role in determining the nitrogen state in TiO 2 band structure [26,29,30].Depending on the nitrogen source and experimental conditions N species are incorporated into TiO 2 in interstitial form or substitutional form [31,32].The interstitial N-doping is favored when the doping is carried out under oxygen-rich conditions and the substation-type Ndoping occurs under reducing conditions [31][32][33].Therefore, from the atomic N 1s XPS spectra the interstitial sites (400 eV) were observed under air atmosphere.
The XPS spectra of Ti 2p regions are shown in Figures 7(c) and 7(d).The peaks at around 458.5 and 464.2 eV for P25 are ascribed to Ti 2p 3/2 and Ti 2p 1/2 of TiO 2 (data not shown).The line separation between Ti 2p 1/2 and Ti 2p 3/2 was 5.7 eV, which is consistent with the standard binding energy.According to Figure 7, all spectra of Ti 2p 1/2 are symmetry.The spectra of Ti 2p 1/2 are consistent of single state and the banding energy of peaks is 458.0-458.3eV which shows a redshift of 0.5 eV compared to the binding energy of Ti 4+ in P25.This suggests that Ti 3+ was present in N-TiO 2 and the distance of peaks between Ti 2p 1/2 and Ti 2p 3/2 is 5.7-5.9eV.The 396 eV peak is assigned to the atomic b-N state and generally proves the presence of TiAN bonds formed when N atoms replace the oxygen in the TiO 2 crystal lattice [30].The peak characteristic for Ti-N (396 eV) is not present, indicating the absence of the TiN phase in the N-TiO 2 samples [34].The results show that the binding energies of Ti 2p peaks shift to lower energies with a negative shift of ∼0.5 eV for the N-TiO 2 due to the N doping [27].The lower binding energy than a typical Ti 2p signal of Ti 4+ oxidation state suggests that titanium cation has a considerable interaction with the doped nitrogen and thus TiO 2 lattice is modified [27,35].Nitrogen doping accompanies the formation of oxygen vacancies and/or Ti 3+ defects, resulting in slight shifts of the Ti 2p peak toward the lower binding energy [36][37][38].Hence, the red shifts of the Ti 2p peak in comparison with P25 are used to highlight the successful nitrogen doping into the TiO 2 lattice.The XRD results did not indicate the formation of TiN bonds.Therefore, results indicate that the titanium atom bonds with oxygen but not nitrogen.Thus, a shift toward lower binding energy of Ti 2p and the binding energy of N 1s upon nitrogen addition indicate that the nitrogen can be indeed incorporated into the interstitial positions of TiO 2 lattice and the formation of Ti-O-N bonding by partially interstitial with a carbon atom.
The incorporation of nitrogen into the oxide lattice should influence the BE of O 1s as well.The XPS spectra of the O 1s core level consists of the one peak (Figures 3(e) and 3(f)), at around 529.3-529.5 eV, which is ascribed to Ti-O bond in the TiO 2 lattice [30].The O 1s core level moved toward lower energy state from 529.6 eV in P25 to 529.4 eV in N-TiO 2 prepared with A/T ratio 3.0 and calcined at 400 ∘ C. The peak shifts by ∼0.2 eV which might be related to the creation of oxygen vacancies as a result of the nitrogen doping.The shifting of bind energy in both N 1s and O 1s region indicates that nitrogen was incorporated into the lattice.The shifts of Ti 2p 3/2 and O 1s peaks are due to the introduction of oxygen vacancies into the TiO 2 lattice [32].The presence of nitrogen dopant facilitates the formation of oxygen vacancies.The aforementioned results clearly indicate that nitrogen doping is accompanied by oxygen vacancy formation.The optical transitions between the dopant and dopant-induced level and Ti3d orbital explain the observed visible light absorptions of N-TiO 2 nanomaterials.The observed two absorbance edge features in the UV-vis spectra are directly related to the above modification of the electronic states from the dopants.

Visible Light Photocatalytic Activity.
To explore the photocatalytic activity of different N-doped TiO 2 at different calcination temperatures and A/T ratios, the degradation of ethylene under visible light with a cut-off wavelength of 400 nm was investigated.Carbon dioxide started being formed immediately after the visible light was turned on and ethylene was almost totally converted to CO 2 in all measurements.It is obviously that the N-TiO 2 shows significant progress in the photodegradation of ethylene compared to P25 in visible light system.P25 has no obvious photocatalytic activity (less than 1%) under visible light irradiation.Results suggest modification of TiO 2 provides visible-light-sensitive photocatalyst, which can be applied to the photooxidation of ethylene gas, as the band gap of TiO 2 was modified by nitrogen doping, evidenced by XPS and UV-visible DRS results.Figure 8 shows the activity of N-TiO 2 catalysts in the presence of visible light for photodegradation of ethylene.The reaction rate increases with the increase in the A/T ratio from 0.5 to 2.0, and then decreases in A/T ratio from 2.0 to 3.0.The enhancement of photocatalytic activity can be ascribed to an obvious improvement in anatase crystallinity and larger surface area.For the sample prepared with A/T ratio 2.0 and 3.0, however, the removal efficiency decreased rapidly, which can be due to its less surface area.The photocatalytic ability of N-TiO 2 samples calcinated at 400, 500, and 800 ∘ C is also examined.It was found that the photocatalytic activity decreased with increasing calcination temperature.The enhancement of photocatalytic activity can be ascribed to an obvious improvement in anatase crystallinity at 400 ∘ C calcination temperature.Good anatase crystallization is beneficial for reducing the recombination rate of the photogenerated electrons and holes due to the decrease in the number of the defects [39].Zhao and Yang showed that for photocatalytic oxidation application, anatase is superior to rutile: (a) the conduction band location for anatase is more favorable for driving conjugate reactions  involving electrons, and (b) very stable surface peroxide groups can be formed at the anatase during photo-oxidation reaction but not on the rutile surface [40].For the sample calcined at 800 ∘ C, the removal efficiency decreased rapidly, which can be due to its fewer amounts of anatase and small  BET .The result indicates that higher specific surface area, smaller crystallite size, and good anatase crystallization will promote the photocatalytic activity in visible light.The results indicate the N-TiO 2 catalyst prepared with A/T ratio 2.0 and calcinated at 400 ∘ C shows the best photocatalytic ability.

3.7.
The Formation Mechanism of N-Doped TiO 2 .For the pure TiO 2 , the energy of the visible light is not sufficient to excite electrons from the valence band (VB), as the intrinsic band gap of pure TiO 2 .Based on the results of UVvis DRS, XRD, and XPS, the nitrogen atoms were weaved into the crystal structure of TiO 2 with structure as Ti-O-N.N-TiO 2 was active in the visible light irradiation for the degradation of ethylene.The following explanation for the photocatalytic activity of N-TiO 2 in the visible light spectra can be considered.The nitrogen existing at interstitial positions of TiO 2 lattice induces localized occupied states and forms midband gap, leading to the red shift of absorption edge and photocatalytic activity under visible light irradiation [41].The electron (e − ) can be excited from the N-impurity level to the conduction band (CB).The exited electrons are trapped by O 2 adsorbed on the catalyst surface, producing O 2 −• superoxide anion radicals.After a series of reactions (1)-( 12), ethylene molecules are finally mineralized into CO 2 .DFT calcinations and experimental results indicate that Ndoping favored the formation of O vacancy (subband level O  ), which was below the bottom of the conduction band [42][43][44][45][46][47].Therefore, the electrons (e − ) are excited from the N-impurity level bond to the subband level (O  ) and then recombine with the hole (h + ) as proved in Figure 9.The holes and electrons react with OH − and O 2 molecules on the catalyst surface to form S-HO

Conclusions
The observed changes in the UV-vis DRS, XRD, and XPS spectra are providing consistent structural information for Ti-O-N formation; that is, the nitrogen can be incorporated into the interstitial positions of TiO 2 lattice, which leads to the enhanced photocatalytic activity in N-TiO 2 samples.It is concluded that the photocatalytic oxidation reactivity of N-TiO 2 under visible light illumination is mainly due to the presence of localized occupied states caused by interstitial nitrogen and the electron/hole pairs generated under visible light irradiation.It has been found that N dopant retarded the anatase to rutile phase transformation by forming Ti-O-N incorporated into the interstitial position of TiO 2 lattice.Moreover, the transformation temperatures of anatase-torutile progressively slightly increase when N dopant content is increased.The photocatalyst, N-TiO 2 prepared with A/T ratio 2.0 and calcinated at 400 ∘ C, shows the highest photocatalytic activity in the oxidation of ethylene.This N-TiO 2 provides an effective visible-light-responsive photocatalyst for future industrial applications in pollution control.

Figure 2 :
Figure 2: XRD patterns of N-TiO 2 samples prepared (a) at various calcination temperatures and (b) with various ammonia/TTIP ratios.

Figure 3 :
Figure 3: The anatase content of N-TiO 2 samples as a function of the various ammonia/TTIP molar ratios.

Figure 4 :
Figure 4: TEM images of N-TiO 2 prepared (a)-(e) at various calcination temperatures with 2.0 ammonia/TTIP ratio and (f) the selected area electron diffraction of A20T4.

Figure 5 :
Figure 5: Nitrogen adsorption and desorption isotherms of N-TiO 2 samples (a) calcined at various temperatures, (b) prepared with various ammonia/TTIP molar ratios; the surface area as a function of calcination temperature.

Figure 6 :
Figure 6: The typical UV-vis DR spectra of N-TiO 2 samples: (a) prepared with 2.0 ammonia/TTIP molar ratio and calcined at various temperatures; (b) prepared with various ammonia/TTIP molar ratios and calcined at 400 ∘ C.

Figure 9 :
Figure 9: Schematic diagram for the process in the photooxidation of ethylene and electron transfer process over N-TiO 2 under visible light illumination.

Table 1 :
Characteristics for N-TiO 2 at various ammonia/TTIP molar ratios under different calcination temperatures.