High Selectivity of Visible-Light-Driven La-doped TiO 2 Photocatalysts for NO Removal

Semiconductors mediated by rare earth metals (REMs) have attracted attention with regard to the degradation of pollutants. In order to enhance the visible response of TiO2, La-doped TiO2 (La-TO) photocatalysts with visible-light-driven capacity for NO removal were successfully synthesized in this study via a facile sol-gel method followed by calcination. A series of La-TiO2 samples with differing weight ratios were evaluated for their photocatalytic performances. It was found that 3% La integrated with TiO2 (in mass ratio) could enhance the removal efficiency of NO (up to 32%) under solar light, which is more than twice that seen with pure TiO2. The resulting products were characterized by a series of techniques, such as XRD, FTIR, UV-vis DRS, BET and (photo)electrochemical analysis. The results indicated that La-doped TiO2 can harvest visible light due to the relatively narrow band gap (from 2.98 to 2.75 eV). More importantly, La dopant improved electronhole separation and suppressed charge carrier recombination, due to the synergistic effect. Furthermore, La-doped TiO2 increased the photo-oxidation efficiency of the transformation from NO to NO3, owing to inhibition of the production of intermediate NO2 (0.02%). To the best of our knowledge, this study is the first time that La-doped TiO2 has been used to eliminate NO (at the ppb level) in the atmosphere. This study provides a facile and controllable route to fabricate La-TO photocatalyst for NO abatement with high selectivity of NO2 under visible light.


INTRODUCTION
NO x , which typically refers to NO and NO 2 , has been associated with Chinese haze and secondary organic aerosols (SOAs) in recent years (Huang et al., 2014;Shi et al., 2015;Fujitani et al., 2017).The previous literature notes the formation of fine particulate matter (PM, especially PM 2.5 ) generated from such chemical precursors, which form via photochemical reactions between NO x and volatile organic compounds (VOCs) (Li et al., 2015a;Wang et al., 2016a;Zhao et al., 2016).It is reported that anthropogenic sources of NO x , which come from stationary power (mainly coal) plants and automobile engines, contribute to the emission and accumulation of NO x in rapidly urbanizing cities (Ball et al., 1999;Lasek et al., 2013, Liu et al., 2017a).Over the past few decades, DeNO x catalysis techniques such as selective catalytic reduction (SCR), three-way catalysis (TWC) and wet scrubbing, have been developed to reduce the exhaust gas in after-treatment systems (Granger and Parvulescu, 2011;Lasek et al., 2013, Liu et al., 2017b).However, these technologies inevitably require the use of noble metals, high temperatures and high costs for the high-concentration NO x treatment that is part of the process.Photocatalysis, which mainly refers to heterogeneous catalysis, offers an innovative and promising approach to reduce NO x at the part-per-billion (ppb) level (Wang et al., 2016c;Zhang et al., 2016).Titania (TiO 2 ) is a well-known semiconductor material that has been employed commercially for a few years (Tong et al., 2012;Kaminski et al., 2015).Titania has been explored for applications in many areas, such as photocatalysis, photovoltaics, water splitting (Chen and Mao, 2007;Thunyasirinon et al., 2015).Owing to such features as abundance, long-term stability against photochemical corrosion and being non-poisonous, huge efforts have been dedicated to utilizing TiO 2 in photocatalysis (Devi and Kavitha, 2014;Yu et al., 2016).Nonetheless, it is not feasible to use TiO 2 to harvest visible light, because of the limitation of its bandgap (~3.0-3.2 eV).This means that TiO 2 can only absorb UV light with a wavelength < 380 nm, and UV light accounts for a very small portion (5%) of solar light.Two key aspects of photocatalysis are the effective harvesting of solar energy and suppression of charge carrier recombination.Therefore, one challenge in materials chemistry and physics is to design semiconductor-mediated photocatalysts with a visible-light-harvesting capability and high quantum yield (Wang et al., 2014;Li et al., 2015b).
With these considerations in mind, researchers have examined a number of strategies to broaden the spectral absorption range of TiO 2 .Numerous approaches have been proposed, including doping, the use of co-catalysts with foreign ions, the application of surface plasmon resonance (SPR) effects, Schottky junctions and the construction of heterojunctions (containing type-I and type-II heterojunctions, Z-scheme, and p-n heterojunctions), and these have been reported to effectively increase visible-light absorption, promote the separation and transportation of the charge carriers, and enhance photocatalytic stability (Rajh et al., 1999;Dinh et al., 2014;Ilkhechi et al., 2017).Among the various approaches, non-metal/metal doping into titania has drawn significant attention due to its ability to modify the surface-electronic properties that benefit the photocatalytic quantum yield.For non-metal ion dopants, substitution of p block elements (B, C, N, F, S, P and I) either at Ti 4+ and O 2-sites is an appealing option to tailor the bandgap absorption to the visible-light region, with a robust interfacial charge carrier transfer process (Devi and Kavitha, 2013).For instance, N-doping TiO 2 (N-TiO 2 ) generates the visible light response because of the occupied N 2p states above the VB edge, while the transition from the occupied π* character N-O localized state results in a visible light response for N-TiO 2 (Di Valentin et al., 2005).On the other hand, many transition metal ions have been introduced into titania to extend the visible absorption, which is attributed to the charge transfer transition between the d electrons of the dopant and CB (or VB) of TiO 2 .Moreover, metal ions modulate the charge carrier equilibrium concentration by serving as electron-hole traps (Bouras et al., 2007;Murakami et al., 2008;Yu et al., 2009).
In recent years, the doping of rare earth metals (REMs) has offered another simplified approach to extend the light absorption range and prevent the recombination of electronhole pairs (Yurtsever and Ciftcioglu, 2017).Previous studies showed that REMs having incompletely occupied 4f and empty 5d orbitals often act as promote catalysis.Doping TiO 2 with REMs can enhance the light absorption by tuning the phase structure, surface area and morphology (Yao et al., 2012).It has been shown that REM doping inhibits the phase transformation of TiO 2 from anatase to rutile, while lanthanum (La) doping can improve photocatalytic performance by increasing the surface area, pore volume, adsorption capacity and electron-hole separation effect (Raza et al., 2015;Shin et al., 2017).Overall, earlier studies with respect to La dopant have mostly focused on its application in the photocatalytic degradation of dyes in wastewater (Huixian et al., 2011;Wang et al., 2016b).
Herein, La-doped TiO 2 (La-TO) photocatalysis with visible-light-driven capacity was successfully synthesized via a facile sol-gel method followed by calcination.To the best of our knowledge, this is the first time that La-doped TiO 2 has been applied in eliminating NO (at the ppb level) in the atmosphere.The experimental results show that this approach has three advantages compared with pure TiO 2 : (1) La-doped TiO 2 can harvest visible light due to the relatively narrow band gap, which contributes to enhancing the removal efficiency of NO (up to 32%); (2) La dopant improves electron-hole separation and suppresses charge carrier recombination; (3) La-doped TiO 2 increases the photo-oxidation efficiency of the transformation from NO to NO 3 -, due to inhibiting the production of intermediate NO 2 (0.02%).Furthermore, a series of La-TO samples with differing weight ratios were evaluated with regard to their photocatalytic performances.

Preparation of Pristine TiO 2 and La-TiO 2 Samples
All the chemicals used in our experiment were of analytical grade and applied without further purification.The La-TiO 2 photocatalyst was prepared by a sol-gel method followed by calcination.In brief, 1.5 mL acetic acid and 18 mL absolute ethanol were mixed together (denoted as solution A).Two mL titanium(IV) isopropoxide (Ti(OBu) 4 ) was dissolved into 18 mL absolute ethanol with stirring for several minutes (denoted as solution B).Solution A was then added dropwise into solution B with strong stirring for 30 min.A transparent sol was then obtained and the gelation was finished by aging the sol for 2 h at room temperature.Solution C contained 10 mL H 2 O and different mass ratios of La(NO 3 ) 3 in the required stoichiometry, and this was slowly added dropwise into the above sol.The gel was dried at 100°C for 12 h after adding 1 mL nitric acid (68%).Finally, the powder samples were calcined at 500°C for 4 h and La-doped TiO 2 samples were obtained.For convenience and clarity, the synthesized La-TiO 2 with various mass ratios of La 3+ dopant were denoted as La x -TO, where x represents the mass ratio of La 3+ to La-TiO 2 .The resulting samples with x values of 0.5, 1, 2, 3 and 5% were thus labeled as La 0.5 -TO, La 1 -TO,La 2 -TO, La 3 -TO and La 5 -TO, respectively.For comparison, pure TiO 2 was prepared under the identical conditions without the addition of solution C.

Characterization
The crystalline phase of the as-obtained samples was determined by using an X-ray powder diffractometer (XRD; PANanalytical, X'pert, the Netherlands) with a Cu Kα radiation source (λ = 1.5406Å) at a scan step size of 0.017° in the 2θ range of 10°-80°.The Brunauer-Emmett-Teller (BET) surface area and pore structure of the samples were obtained from the adsorption branch of the isotherms and the Barrett-Joyner-Halenda formula at 77 K by using a Gemini VII 2390 analyzer (Micromeritics Instrument Corp., Norcross, GA, USA).The optical properties of the catalysts at 200-800 nm were obtained using a Varian Cary 100 Scan UV-Visible system (UV-Vis; Agilent Corp., USA) equipped with Labsphere diffuse reflectance device.Ba 2 SO 4 was used as a reflectance standard.Fourier transform infrared (FT-IR) spectra data were recorded on an FTIR absorption spectrometer (Vertex 70, Bruker), with KBr as the diluent.Photoluminescence (PL; F-7000, Hitachi, Japan) was also obtained to investigate the optical properties of the resulting samples.

Photocatalytic Activity Assessment
The photocatalytic activities of the as-prepared samples were investigated by degrading NO at ppb levels in a continuous flow reactor at ambient temperature under visible light and with solar light irradiation.The reaction chamber was a rectangular stainless-steel vessel (30 cm L × 15 cm W × 10 cm H) covered with a quartz window.Visible light was provided by a 300 W xenon lamp (Perfect Light MICROSOLAR 300, Beijing, China) with a UV cutoff filter (λ > 420 nm), which passed vertically through the quartz window.The commercial 300 W Xe arc lamp (MICROSOLAR 300 UV, Beijing) for the simulated solar light photocatalytic activity test was placed vertically 20 cm above the reactor.For each photocatalytic activity experiment, one sample dish (diameter = 12 cm) containing the photocatalyst powders was placed at the center of the reactor.The photocatalyst samples were prepared by coating an aqueous suspension of the products onto a glass dish.The weight of the photocatalysts used for each experiment was maintained at 100 mg.The dishes containing the photocatalyst were pretreated at 70°C for several hours until water in the suspension was completely removed.The pretreated dishes were then cooled to room temperature prior to the photocatalysis test.
NO gas at an original concentration of 50 ppm was obtained from a compressed gas cylinder.The original concentration of NO used in the photocatalysis test was diluted to 400 ppb with an air stream supplied by a zero air generator (Model 1001, Sabio Instruments LLC, Georgetown, TX, USA).The desired humidity level of the NO flow was controlled at 70% by passing the zero air streams through a humidification chamber.The gas streams were completely premixed by a gas blender, and the flow rate was controlled at 3 L min -1 by using a mass flow controller.After the catalyst achieved adsorption/desorption equilibrium, the xenon lamp was switched on.The NO and NO 2 concentrations were continuously measured with a chemiluminescence NO x analyzer (Model 42c, Thermo Environmental Instruments Inc., Franklin, MA, USA) during photocatalytic degradation at a sampling rate of 0.6 L min -1 .The reaction between NO and air was negligible in the control experiment performed with or without light in the absence of the photocatalyst.The removal ratio of NO at any onetime was noted as (1 -C/C 0 ), where C is the NO concentration of the outlet at any one time, and C 0 is the initial concentration of NO, ppb.The selectivity of NO 2 over different photocatalysts was calculated by the following equation (Zhang et al., 2016): where C NO 2 is the production of NO 2 , ppb.

Photoelectrochemical Measurements
The photoelectrochemical properties of the as-prepared samples were measured using an electrochemical workstation (PARSTAT-4000, Advanced Measurements Technology, Inc., USA) with a conventional three-electrode cell, which includes a platinum plate and an Ag/AgCl electrode that act as a counter electrode and a reference electrode, respectively.Afterward, La-TiO 2 (25 mg) was dispersed into 5 mL of 1 wt% Nafion ethanol solution to obtain a homogeneous suspension via bath sonication and thus fabricate a working electrode.The films were subsequently modified by a 10 mm × 10 mm fluorine-doped tin oxide conducting glass by using a dip coater (ZR-4200, Qingdao, China).The modified films were then arranged to dry under ambient conditions.A 300 W Xe arc lamp (Perfect Light MICROSOLAR 300, Beijing, China) with a UV cut-off filter (λ > 420 nm) was employed as a light source.Current-time curves were tested at 0.2 V versus Ag/AgCl in Na 2 SO 3 (0.1 mol L -1 ) at ambient temperature.

Characterization of La-TiO 2 XDR analysis
In order to determine the crystal phase of the prepared La-TO, the powder X-ray diffraction patterns were used to investigate the La 3+ doping on the anatase phase of TiO 2 at different calcination temperatures.Before doping with La 3+ , the effects of calcination temperature on the structure phase of pure TiO 2 were studied by XRD.As shown in Fig. 1(A), the peaks located at 25.3°, 37.8° and 48.1° show the characteristic diffractions of the ( 101), ( 004) and ( 200) anatase-type TiO 2 , which can be well-indexed as the phase of anatase (JCPD no.21-1272).This means that pure anatase was synthesized at 500°C, and the degree of crystallinity degree of TiO 2 was relatively poor at 400°C.Moreover, the phase transformation of TiO 2 from anatase to rutile occurred at 600°C.The Scherrer equation was applied to evaluate the crystallite sizes of the samples, using the following formula: where D is the average crystallize size (nm), λ is the wavelength of Cu Kα X-ray radiation (λ = 0.15406 nm), k is a coefficient usually taken as 0.94, β is the full width at half maximum (FWHM) intensity of the peak observed at 2θ.From the equation, the wider FWHM is, the smaller the crystallite size.Fig. 1(B) shows that La 3+ doping not only inhibited the phase transformation of TiO 2 from anatase to rutile at 600°C, but also decreased the crystallite size of La 3 -TO.According to previous research, this is because the surrounding La 3+ combined with Ti 4 + and formed a Ti-O-La bond.Furthermore, the La 2 O 3 lattice locks the Ti-O species at the interface, with titania domains preventing the nucleation that is necessary for the transformation of anatase to rutile, resulting in a decrease of the crystallite size of TiO 2 (Lin and Jimmy, 1998).Overall, La 3 -TO obtained at 500°C was chosen as the optimum sample due to the excellent crystallization of anatase and smaller size.

Surface Area and Pore Structure of Samples (BET)
As shown in Fig. 2, the N 2 adsorption-desorption isotherms were measured to analyze the surface area and porosity.The results showed that all isotherms are of a type Langmuir IV isotherm with an H2 hysteresis loop according to the IUPAC classification, which indicates the presence of mesopores (Thommes et al., 2015).Pure TiO 2 and La 3 -TO exhibited mesoporosity with average pore sizes approximating 3.45 nm and 5.47 nm, respectively.The BET specific surface areas, pore volume and pore diameters of the samples are given in Table 1.The specific surface area of the La 3 -TO calcined at 500°C is 101.90 cm 3 g -1 , which is four times larger than that of pure TiO 2 (23.47 cm 3 g -1 ).The relatively high surface area of La 3+ doped samples confirms that the frameworks of TiO 2 have better adsorption ability.This may be due to the linkage between the rare earth ions and titanium by the oxygen bridge, which effectively enhances the specific surface area of TiO 2 (Huixian et al., 2011).Overall, the results indicate that La 3 -TO has the better photocatalytic activity, which may be due to the existence of a mass of surface active sites.

FTIR and UV-vis Analysis
The FTIR spectra of pure TiO 2 and La 3 -TO are shown in Fig. 3(A), in which a series of typical absorption bands of pure TiO 2 and La 3 -TO can be observed.The absorption bands at 3414 cm -1 and 1635 cm -1 indicate the characteristic vibration mode of O-H, demonstrating that much more H 2 O adsorbed on the surface of La 3 -TO after doping with La 3+ , which was then capable of producing much more •OH radicals after light irradiation.The absorption band at 530 cm -1 is assigned to the characteristic vibration mode of Ti-O-Ti.To investigate the optical absorption properties of photocatalysts, the UV-vis DRS of pureTiO 2 and La 3 -TO in the range of 200-800 nm were measured and the results are shown in Fig. 3(B).The spectra of La 3 -TO show red shift compared with pure TiO 2 .This phenomenon can be attributed to the charge-transfer transition between 4f orbital electrons of rare earth ions and the TiO 2 conduction or valence band (Borgarello et al., 1982).Therefore, titania doped with La 3+ increased the absorption range in the visible-light region.Moreover, the band energies (Eg), which are estimated from the intercept of tangents to the plots of (αhν) 1/2 versus the photo energy, are 2.98 and 2.75 eV for pure TiO 2 and La 3 -TO, respectively, as illustrated in Fig. 3(B).It was thus found that La 3+ dopant narrowed the band gap of TiO 2 , which extended the optical response in the visible-light region.

Photocatalytic Activity Assessment
The photocatalytic activities of blank (direct photolysis), pure TiO 2 and La x -TO with different La-doped mass ratios (x = 0.5, 1, 2, 3 and 5%) were evaluated for the degradation of NO under visible light (λ > 420 nm) and under solar light irradiation.The NO removal rates against irradiation time over different photocatalysts are plotted in Figs.4(A) and 4(B).It can be seen that the pristine TiO 2 had poorer photocatalytic performance for NO removal under visible light (only 2.8%), as relatively broad bandgap of TiO 2 can only be stimulated by UV.In contrast, the use of La x -TO with varying amounts of dopant dramatically improved the photocatalytic efficiency.Moreover, the degradation efficiency increased along with the amount of La 3+ dopant.In particular, La 3 -TO exhibited the highest photocatalytic activity (up to 29%) and continued with the degradation even when irradiated for 30 min, with this activity being 10 times higher than that seen with pure TiO 2 .With respect to photocatalytic efficiency under solar light irradiation, La 3 -TO also showed 38.5% photocatalytic activity.As for pure TiO 2 , it became inactivate after being irradiated for 5 min due to the lack of sufficient •OH radicals.This is in accordance with the FTIR results and the higher specific surface areas of the La 3 -TO sample.Moreover, toxic intermediate NO 2 was also identified during the photocatalytic processes.As   ), the amount of NO 2 generated showed no significant difference under under visible light (λ > 420 nm).Nonetheless, La x -TO produced no toxic NO 2 under solar light irradiation (in Fig. 4(D)), compared with pure TiO 2 (approximately 26%).Therefore, La-doped TiO 2 increased the photo-oxidation efficiency of the transformation from NO to NO 3 -, as it inhibited the production of intermediate NO 2 .The doped metal species could trap and subsequently transfer the photoexcited electrons on the TiO 2 surface, accelerating the formation of hydroxyl radicals (•OH).•OH can then further convert intermediate NO 2 into NO 3 -.In Fig. 5(A), the NO 2 yields of as-prepared samples were calculated based on Eq. ( 1).The NO 2 yield of pure TiO 2 was up to 6.54% under solar light irradiation, which is much higher than that of La x -TO.Among the various results, La 3 -TO has the lowest minimum production rate of intermediate product (0.02%).This suggests that La x -TO is better for the removal of NO x under realistic conditions, due to the impact of La 3+ doping.It is thus found that La x -TO photocatalyst has excellent performance for NO abatement, with high selectivity of NO 2 under visible light.Various TiO 2 -based nanomaterials have been used for NO removal, and Table 2 summarized the details of some of these earlier efforts, including the methodology, initial concentration of NO, amount of catalyst, irradiation light and by-product (NO 2 ) yield.As a whole, La 3 -TO exhibited favorable capabilities with regard to eliminating NO (at the ppb level) with high selectivity of NO 2 under visible light.
Moreover, reproducibility and stability are important if photocatalysts are to be applied in realistic environments.
In order to determine the stability of the photocatalyst, multiple runs for the photocatalytic removal of NO with the La 3 -TO samples were tested under solar light irradiation.
As shown in Fig. 5(B), the NO removal ratio remained high after undergoing five repeated runs.This indicate that La 3 -TO has excellent stability with regard to photocatalytic activity, and can be used repeatedly in a realistic environment.

The Effects of Calcination Temperature on NO Removal
To investigate the effects of catalyst structure on photodegradation efficiency, the La 3 -TO samples calcined at different temperatures (400, 500, 600°C) were evaluated for the degradation of NO under visible light (λ > 420 nm) and under solar light irradiation (in Figs.6(A) and 6(B)).The results showed that the highest degradation efficiency was achieved with the sample calcined at 500°C, which benefited from the excellent crystallization of anatase and smaller size of La 3 -TO (as shown in Fig. 1).The photocatalytic activity of La 3 -TO fell when the calcination temperature was raised to 600°C and the rutile structure formed.This shows that the anatase crystal structure has better photocatalytic activity than that the rutile structure.Figs.6(C) and 6(D) show the corresponding NO 2 generation over La 3 -TO samples calcined at different temperatures and monitored simultaneously.The results indicate that La 3 -TO calcined at 500°C also released the minimum amount of NO 2 under visible light (λ > 420 nm).Moreover, the effects of calcination temperature on the generation NO 2 under solar light irradiation showed little significant difference.The temperature at 500°C was thus more appropriate for the calcination of La 3 -TO.

Charge Separation by Photoelectrochemical Analysis
Generally speaking, the charge separation efficiency is a critical factor to improve the activity of photocatalysts.To validate the effects of the La dopant, the transient photocurrent responses of pure TiO 2 and La 3 -TO electrodes in on-off cycles under visible light irradiation were examined.0.02% This work a f-TiO 2 : TiO 2 prepared by flame spray pyrolysis (FSP); b w-Pd/f-TiO 2 : 1 wt% of Pd was photodeposited on f-TiO 2 ; c K-MnOx/TiO 2 : potassium deactivation of MnOx/TiO 2 ; d SCR: selective catalytic reduction; e 15%MWCNTs/Cu-TiO 2 : 15 mol% Cu doped TiO 2 supported by MWCNTs; -: no relevant report.
As shown in Fig. 7, when the light source was turned on the photocurrent increased rapidly and was very stable, whereas the value decreased immediately after the lamp was turned off, suggesting that the as-prepared electrodes possess fast photoelectric response ability.The highest photocurrent density from La 3 -TO at the potential of 0.2 V (about 0.4 uA cm -2 ) is much higher than that from pure TiO 2 (0.13 uA cm -2 ).This distinctly enhanced photocurrent density indicates that the doping of La 3+ can efficiently accelerate the separation of photoexcited electron-hole pairs and inhibit the recombination efficiency of photogenerated carriers.

Possible Photocatalytic Mechanism
Based on the analysis presented above, we then considered the possible photocatalytic mechanism for NO removal over La-doped TiO 2 .The reaction processes of the degradation of NO over La-doped TiO 2 are thus explained in the following equations (Reactions 1-7).Once La-doped TiO 2 was irradiated by visible light, photoexcited electrons and holes were generated (R1).The resulting electrons could transfer onto the surface of La 3+ , which promotes the separation of electrons and holes.The electrons captured by La 3+ then reacted with the surrounding oxygen to form •O 2 -radicals (R2).The generated •O 2 -radicals and holes (h + ) could directly oxidize NO into the final product NO 3 -(R3 and R4).Moreover, H 2 O molecules on the surface of TiO 2 could be oxidized into •OH radicals (R5).Because there are much more •OH radicals, NO could be transferred into NO 3 -by these via two-step oxidation (R6 and R7), which prevents the production of intermediate NO 2 .The whole reaction mechanism for NO removal is shown in Scheme 1. Overall, La 3+ dopant could increase the amount of oxygen vacancies and/or surface defects in the TiO 2 photocatalysts that might capture photoelectrons and inhibit the recombination of photo-generated electrons-holes pairs, which results in greater quantum efficiency (Wang et al., 2011).In summary, La 3+ dopant showed a synergistic effect for NO removal under visible light.

CONCLUSIONS
In conclusion, La-doped TiO 2 (La-TO) photocatalysts were synthesized successfully in this work via a facile solgel method followed by calcination.Compared with pure TiO 2 , La-doped TiO 2 dramatically improved the visible light photocatalytic activity with regard to eliminating NO (at ppb level) in the atmosphere, and this is attributed to the fact that La-doped TiO 2 can harvest visible light because of the relatively narrow band gap.In addition, La-doped TiO 2 can promote electron-hole separation, as La 3+ can have synergistic effects and capture photoinduced electrons.Notably, La-doped TiO 2 increased the photo-oxidation efficiency of the transformation from NO to NO 3 -, due to inhibiting the production of intermediate NO 2 .In short, this study provides a facile and effective approach to prepare La-TiO 2 photocatalyst for NO abatement with high selectivity of NO 2 under visible light.

Fig. 4 .
Fig. 4. Photocatalytic activity for NO removal efficiency (C/C 0 ) against irradiation time over Blank (direct photolysis), pure TiO 2 and La x -TO with different La-doped mass ratios (A) under visible light (λ > 420 nm) and (B) under solar light irradiation; the corresponding NO 2 generation against irradiation time (C) under visible light (λ > 420 nm) and (D) under solar light irradiation.

Fig. 5 .
Fig. 5. (A) The corresponding NO 2 yields over pure TiO 2 and La x -TO under solar light irradiation.(B) Cycling runs for the photocatalytic degradation of NO with La 3 -TO under solar light irradiation.

Table 1 .
Summary of surface area, and pore diameter of pure TiO 2 and La 3 -TO.

Table 2 .
Comparison on NO removal efficiency of various TiO 2 -based nanomaterials.