Carbon Nitride Quantum Dots Modiﬁed TiO 2 Inverse Opal Photonic Crystal for Solving Indoor VOCs Pollution

: Indoor toxic volatile organic compounds (VOCs) pollution is a serious threat to people’s health and toluene is a typical representative. In this study, we developed a composite photocatalyst of carbon nitride quantum dots (CNQDs) in situ-doped TiO 2 inverse opal TiO 2 IO for efﬁcient degradation of toluene. The catalyst was fabricated using a sol-gel method with colloidal photonic crystals as the template. The as-prepared catalyst exhibited excellent photocatalytic performance for degradation of toluene. After 6 h of simulated sunlight irradiation, 93% of toluene can be converted into non-toxic products CO 2 and H 2 O, while only 37% of toluene is degraded over commercial P25 in the same condition. This greatly enhanced photocatalytic activity results from two aspects: (i) the inverse opal structure enhances the light harvesting while providing adequate surface area for effective oxidation reactions; (ii) the incorporation of CNQDs in the framework of TiO 2 increases visible light absorption and promotes the separation of photo-generated charges. Collectively, highly efﬁcient photocatalytic degradation of toluene has been achieved. In addition, it can be expanded to efﬁcient degradation of organic pollutants in liquid phase such as phenol and Rhodamine B. This study provides a green, energy saving solution for indoor toxic VOCs removal as well as for the treatment of organic wastewater.


Introduction
Nowadays, people spend most of their time (up to 90%) in an indoor environment. Thus, indoor air quality (IAQ) has a significant influence on human health, comfort and productivity [1,2]. Indoor air pollution is now considered among the top five environmental risks to public health according to the Environmental Protection Agency (EPA), which declared the air to be two to five times more polluted indoors than outdoors [3]. Interior flooring, adhesives, etc., used in interior decoration are sources of volatile organic compounds (VOCs) [4,5]. Exposure to toxic VOCs is seriously harmful to human health. In the past, benzene was widely used as an organic solvent for adhesives and paints. Nowadays, benzene has been replaced by relatively less toxic solvents, such as toluene, owing to its carcinogenic property [6]. However, toluene emitted from building materials The specific surface area is one of the vital factors that affects the photocatalytic performance of materials. Generally, the photocatalysts with larger surface areas possess more adsorption and active sites for photocatalytic process, which enhances the photocatalytic activity. N2 sorption was carried out to determine the specific surface area as well as corresponding pore structure and to calculate the corresponding pore size distributions. As shown in Figure 2a, all samples exhibit type IV isotherms. TCN IO and TiO2 IO display higher adsorption capacities at high relative pressures (P/P0 > 0.8) than bulk-TiO2, which provides evidence for the existence of macropores. The BJH pore size distribution ( Figure 2b) reveals that the majority of the macropores have an average pore size of around 60 nm which provides larger surface areas and pore volumes than those of common bulk-TiO2. The corresponding structural parameters of the as-synthesized samples obtained from the adsorption isotherms are summarized in Table 1. As shown in Table 1, the BET surface areas are approximately 9.0, 50.8, 47.7 and 62.6 m 2 g −1 for bulk-TiO2, P25, TiO2 IO and TCN IO, respectively. These results show that the inverse opal structure can increase the specific surface area of the material, while the introduction of CNQDs can further increase it. This could be attributed to the fact that na- The specific surface area is one of the vital factors that affects the photocatalytic performance of materials. Generally, the photocatalysts with larger surface areas possess more adsorption and active sites for photocatalytic process, which enhances the photocatalytic activity. N 2 sorption was carried out to determine the specific surface area as well as corresponding pore structure and to calculate the corresponding pore size distributions. As shown in Figure 2a, all samples exhibit type IV isotherms. TCN IO and TiO 2 IO display higher adsorption capacities at high relative pressures (P/P 0 > 0.8) than bulk-TiO 2 , which provides evidence for the existence of macropores. The BJH pore size distribution ( Figure 2b) reveals that the majority of the macropores have an average pore size of around 60 nm which provides larger surface areas and pore volumes than those of common bulk-TiO 2 . The corresponding structural parameters of the as-synthesized samples obtained from the adsorption isotherms are summarized in Table 1.
Catalysts 2021, 11, x FOR PEER REVIEW nosized CNQDs exhibit relatively high specific surface area. Additionally, the sponding pore volume increases from 0.022 cm 3 g −1 for bulk-TiO2 to 0.245 cm 3 g −1 f IO and to 0.254 cm 3 g −1 for TCN IO while the pore volume of P25 is 0.167 cm 3 g −1 . T be attributed to the introduction of inverse opal structure, whose three-dimens dered pore structure benefits the enlargement of pore volume [43].    As shown in Table 1, the BET surface areas are approximately 9.0, 50.8, 47.7 and 62.6 m 2 g −1 for bulk-TiO 2 , P25, TiO 2 IO and TCN IO, respectively. These results show that the inverse opal structure can increase the specific surface area of the material, while the introduction of CNQDs can further increase it. This could be attributed to the fact that nanosized CNQDs exhibit relatively high specific surface area. Additionally, the corresponding pore volume increases from 0.022 cm 3 g −1 for bulk-TiO 2 to 0.245 cm 3 g −1 for TiO 2 IO and to 0.254 cm 3 g −1 for TCN IO while the pore volume of P25 is 0.167 cm 3 g −1 . This can be attributed to the introduction of inverse opal structure, whose three-dimensionordered pore structure benefits the enlargement of pore volume [43]. Figure 3 shows the X-ray diffraction (XRD) patterns of bulk-TiO 2 , TiO 2 IO and TCN IO porous composites. For all samples, the XRD patterns exhibited strong diffraction peaks at 25.3 • and 48 • , indicating TiO 2 in the samples are all in anatase phase [25,29,44]. For CNQDs loaded TiO 2 IO composites (TCN IO), no diffraction peaks were observed for g-C 3 N 4 , most probably due to low weight loading and high dispersion of CNQDs in the catalyst. nosized CNQDs exhibit relatively high specific surface area. Additionally, the corresponding pore volume increases from 0.022 cm 3 g −1 for bulk-TiO2 to 0.245 cm 3 g −1 for TiO2 IO and to 0.254 cm 3 g −1 for TCN IO while the pore volume of P25 is 0.167 cm 3 g −1 . This can be attributed to the introduction of inverse opal structure, whose three-dimension-ordered pore structure benefits the enlargement of pore volume [43].  Figure 3 shows the X-ray diffraction (XRD) patterns of bulk-TiO2, TiO2 IO and TCN IO porous composites. For all samples, the XRD patterns exhibited strong diffraction peaks at 25.3° and 48°, indicating TiO2 in the samples are all in anatase phase [25,29,44]. For CNQDs loaded TiO2 IO composites (TCN IO), no diffraction peaks were observed for g-C3N4, most probably due to low weight loading and high dispersion of CNQDs in the catalyst. X-ray photoelectron spectroscopy (XPS) spectra were used to analyze the elemental composition and the chemical state of elements in TCN IO. As Figure 4a shows, the XPS survey spectrum of TCN IO confirmed the existence of Ti, O, C and N elements. C 1s (Figure 4b) peak at 284.5 eV is assigned to the C-C bond in the turbostratic CN structure [36]. The C 1s peak at 285.8 eV is attributed to the sp2 C atoms bonded to N inside the aromatic structure. The peak at 288.5 eV is linked to the sp3 C-N bond of the sp3 bonded composition [45]. The N 1s peaks (Figure 4c) contain three components concentrated at 399.1 eV, 400.0 eV and 401.3 eV, which are identified as the C-N-C, (N (C)3) and C-N-H groups, respectively [46]. The O 1s core level peak at 529.7 eV belongs to Ti-O-Ti linkages in TiO2 (Figure 4d) [47]. After the modification of CNQDs, the peak at 531.4 eV, coming from OH functional groups on the surface of CNQDs, can be clearly identified. This latter X-ray photoelectron spectroscopy (XPS) spectra were used to analyze the elemental composition and the chemical state of elements in TCN IO. As Figure 4a shows, the XPS survey spectrum of TCN IO confirmed the existence of Ti, O, C and N elements. C 1s (Figure 4b) peak at 284.5 eV is assigned to the C-C bond in the turbostratic CN structure [36]. The C 1s peak at 285.8 eV is attributed to the sp2 C atoms bonded to N inside the aromatic structure. The peak at 288.5 eV is linked to the sp3 C-N bond of the sp3 bonded composition [45]. The N 1s peaks ( Figure 4c) contain three components concentrated at 399.1 eV, 400.0 eV and 401.3 eV, which are identified as the C-N-C, (N (C) 3 ) and C-N-H groups, respectively [46]. The O 1s core level peak at 529.7 eV belongs to Ti-O-Ti linkages in TiO 2 ( Figure 4d) [47]. After the modification of CNQDs, the peak at 531.4 eV, coming from OH functional groups on the surface of CNQDs, can be clearly identified. This latter peak is much higher than that of TiO 2 IO, revealing the successful loading of CNQDs to TiO 2 IO [48]. The XPS results of C 1s and N 1s further indicate the successful combination of CNQDs and TiO 2 IO. Catalysts 2021, 11, x FOR PEER REVIEW 5 of 12 peak is much higher than that of TiO2 IO, revealing the successful loading of CNQDs to TiO2 IO [48]. The XPS results of C 1s and N 1s further indicate the successful combination of CNQDs and TiO2 IO. The separation and transfer of electron-hole pairs of the samples can be analyzed via Electrochemical Impedance Spectroscopy (EIS) [49]. The intersection of the main highfrequency semicircle contribution with the x-axis in the Nyquist plots in EIS tests, corresponds to the charge transfer resistance of the catalysts. Figure 5a shows the Nyquist plots of TiO2 IO and TCN IO electrodes, respectively. The EIS plot of TCN IO under simulated solar light illumination shows a smaller semicircular diameter compared with TiO2 IO. This result indicates that TCN IO possess a smaller charge transfer resistance than that of TiO2 IO without CNQDs modification. This suggests that the modification of CNQDs facilitates the charge separation efficiency of the catalyst. Photoluminescence (PL) spectra were analyzed ( Figure 5b) to investigate the migration, transfer and recombination processes of the photo-generated electron-hole pairs in photocatalysts. Notably, the PL spectrum characteristics of the TiO2 IO was markedly weaker than bulk-TiO2. After the modification of CNQDs, PL intensity of TCN IO sample further weakens, which indicates a low recombination rate of photo-generated electrons and holes, and a favorable contact between CNQDs and TiO2 IO. The construct of inverse opal structure and the introduction of CNQDs facilitate the faster separation of photo-generated charges, contributing to enhanced photocatalytic activity. The separation and transfer of electron-hole pairs of the samples can be analyzed via Electrochemical Impedance Spectroscopy (EIS) [49]. The intersection of the main high-frequency semicircle contribution with the x-axis in the Nyquist plots in EIS tests, corresponds to the charge transfer resistance of the catalysts. Figure 5a shows the Nyquist plots of TiO 2 IO and TCN IO electrodes, respectively. The EIS plot of TCN IO under simulated solar light illumination shows a smaller semicircular diameter compared with TiO 2 IO. This result indicates that TCN IO possess a smaller charge transfer resistance than that of TiO 2 IO without CNQDs modification. This suggests that the modification of CNQDs facilitates the charge separation efficiency of the catalyst. Photoluminescence (PL) spectra were analyzed (Figure 5b) to investigate the migration, transfer and recombination processes of the photo-generated electron-hole pairs in photocatalysts. Notably, the PL spectrum characteristics of the TiO 2 IO was markedly weaker than bulk-TiO 2 . After the modification of CNQDs, PL intensity of TCN IO sample further weakens, which indicates a low recombination rate of photo-generated electrons and holes, and a favorable contact between CNQDs and TiO 2 IO. The construct of inverse opal structure and the introduction of CNQDs facilitate the faster separation of photo-generated charges, contributing to enhanced photocatalytic activity.
Samples were used for toluene degradation under AM1.5 simulated solar light irradiation ( Figure 6). After 1 h dark adsorption, we assumed that the gas-solid adsorption equilibrium was reached, and the simulated sunlight irradiation began. The result of blank experiment (without any catalysts) confirms that the concentrations of CO 2 and toluene are stable (Figure 6a). In the dark, without the irradiation of simulated sunlight, the concentration of CO 2 did not increase, while that of toluene decreased. This is most probably due to the adsorption of the porous structure of TCN IO and the adsorption efficiency of TCN IO is 17% (Figure 6b). As shown in Figure 6c,d, the toluene concentration decreased as the illumination continued, and the concentration of CO 2 increased gradually, indicating that toluene was oxidized into CO 2 . After 5h irradiation, 95% of toluene had been removed over TCN IO (Figure 6e), whereas the toluene removal rates of TiO 2 IO, bulk-TiO 2 and P25 were 88%, 77% and 60%, respectively. After 6 h irradiation, the concen-tration of toluene is below the detection limit over TCN IO, whereas C t, C7H8 drops to 17%, 2% and 37% of original concentration over bulk-TiO 2 , TiO 2 IO and P25, respectively. Since the adsorption efficiency of TCN IO in the dark system is just 17%, the high removal rate of toluene obtained in the light irradiant system (100%) is mainly due to the photocatalytic degradation. Additionally, the calculated photocatalytic degradation efficiency of toluene into CO 2 ( t, toluene ) over bulk-TiO 2 is 60%, t, toluene is greatly improved to 82% after the introduction of inverse opal structure (Table S1). These results demonstrate that TiO 2 with inverse opal structure exhibits a significantly enhanced photocatalytic activity for the degradation of toluene under simulated sunlight compared to bulk-TiO 2 . TCN IO shows the highest value of t, toluene , up to 93%, while that of P25 is only 37%, revealing that the incorporation of inverse opal structure and CNQDs promotes the photocatalytic performance of the catalyst. The highly efficient activity of TCN IO can be explained by (i) the inverse opal structure which provides an adequate surface area for the adsorption and the oxidation of toluene; (ii) and the incorporation of CNQDs in the framework of TiO 2 which promotes the separation of photo-generated charges. Samples were used for toluene degradation under AM1.5 simulated solar light irradiation ( Figure 6). After 1 h dark adsorption, we assumed that the gas-solid adsorption equilibrium was reached, and the simulated sunlight irradiation began. The result of blank experiment (without any catalysts) confirms that the concentrations of CO2 and toluene are stable (Figure 6a). In the dark, without the irradiation of simulated sunlight, the concentration of CO2 did not increase, while that of toluene decreased. This is most probably due to the adsorption of the porous structure of TCN IO and the adsorption efficiency of TCN IO is 17% (Figure 6b). As shown in Figure 6c,d, the toluene concentration decreased as the illumination continued, and the concentration of CO2 increased gradually, indicating that toluene was oxidized into CO2. After 5h irradiation, 95% of toluene had been removed over TCN IO (Figure 6e), whereas the toluene removal rates of TiO2 IO, bulk-TiO2 and P25 were 88%, 77% and 60%, respectively. After 6 h irradiation, the concentration of toluene is below the detection limit over TCN IO, whereas Ct, C7H8 drops to 17%, 2% and 37% of original concentration over bulk-TiO2, TiO2 IO and P25, respectively. Since the adsorption efficiency of TCN IO in the dark system is just 17%, the high removal rate of toluene obtained in the light irradiant system (100%) is mainly due to the photocatalytic degradation. Additionally, the calculated photocatalytic degradation efficiency of toluene into CO2 (ŋt, toluene) over bulk-TiO2 is 60%, ŋt, toluene is greatly improved to 82% after the introduction of inverse opal structure (Table S1). These results demonstrate that TiO2 with inverse opal structure exhibits a significantly enhanced photocatalytic activity for the degradation of toluene under simulated sunlight compared to bulk-TiO2. TCN IO shows the highest value of ŋt, toluene, up to 93%, while that of P25 is only 37%, revealing that the incorporation of inverse opal structure and CNQDs promotes the photocatalytic performance of the catalyst. The highly efficient activity of TCN IO can be explained by (i) the inverse opal structure which provides an adequate surface area for the adsorption and the oxidation of toluene; (ii) and the incorporation of CNQDs in the framework of TiO2 which promotes the separation of photo-generated charges.   Under the irradiation of simulated solar light, samples were also used for the degradation of liquid phase pollutants represented by RhB and phenol. As shown in Figure 7, the degradation rate of dye and phenol reached above 97% after 75 min and 100 min of illumination, respectively. The efficient degradation of phenol and RhB provides evidence of the high performance of TCN IO and proves that as-prepared catalysts could also be utilized to solve the problem of water pollution. Under the irradiation of simulated solar light, samples were also used for the degradation of liquid phase pollutants represented by RhB and phenol. As shown in Figure 7, the degradation rate of dye and phenol reached above 97% after 75 min and 100 min of illumination, respectively. The efficient degradation of phenol and RhB provides evidence of the high performance of TCN IO and proves that as-prepared catalysts could also be utilized to solve the problem of water pollution.
Under the irradiation of simulated solar light, samples were also used for the degradation of liquid phase pollutants represented by RhB and phenol. As shown in Figure 7, the degradation rate of dye and phenol reached above 97% after 75 min and 100 min of illumination, respectively. The efficient degradation of phenol and RhB provides evidence of the high performance of TCN IO and proves that as-prepared catalysts could also be utilized to solve the problem of water pollution. The reusability and stability of the catalyst is an important parameter to evaluate its practical application potential. After illuminating the substrates by a solar simulator (AM 1.5) equipped with a 300 W Xenon lamp for 30 min and overnight ventilating without washing, the catalyst was recycled. The recycling experiment was repeated four times to test the stability of TCN IO for toluene degradation. As shown in Figure 8, a near 88% degradation ratio of 665 ppm toluene can be observed in the fourth degradation process, revealing the excellent reusability and stability of the as-prepared catalyst TCN IO. Figure  S2 shows XRD of TCN IO sample after the recycle experiments had been conducted ( Figure S2). The slight decrease in crystallinity and part of the toluene which occupies the active sites of the catalyst and partially inhibits the reaction, probably caused the monotonous but small drop in activity. The reusability and stability of the catalyst is an important parameter to evaluate its practical application potential. After illuminating the substrates by a solar simulator (AM 1.5) equipped with a 300 W Xenon lamp for 30 min and overnight ventilating without washing, the catalyst was recycled. The recycling experiment was repeated four times to test the stability of TCN IO for toluene degradation. As shown in Figure 8, a near 88% degradation ratio of 665 ppm toluene can be observed in the fourth degradation process, revealing the excellent reusability and stability of the as-prepared catalyst TCN IO. Figure S2 shows XRD of TCN IO sample after the recycle experiments had been conducted ( Figure S2). The slight decrease in crystallinity and part of the toluene which occupies the active sites of the catalyst and partially inhibits the reaction, probably caused the monotonous but small drop in activity.

Synthesis of PS Colloidal Crystal Template
Polystyrene spheres (PS) suspension with controllable size were synthesized according to the reported emulsion polymerization method [20,[23][24][25]. Styrene monomers were alternately washed with 0.5 M NaOH solution and deionized water in a separate funnel with a 1:1 volume ratio 3 times to remove the polymerization inhibitor in styrene. 0.45 g of sodium dodecyl sulfate (SDS) and 0.6 g of K2S2O8 were mixed in 288 mL of deionized water and 32 mL of ethanol (EtOH) with magnetic stirring to form a homogeneous solution; this was followed by injecting 36 mL of washed styrene at nitrogen atmosphere. A milky product was obtained after heating at 71 °C for 19 h. The as-prepared polystyrene emulsion was transferred into 250 mL beakers with a height of ~ 2 cm and then put into a 70 °C oven to evaporate the solvent.

Synthesis of Graphitic Carbon Nitride Quantum Dots (CNQDs)
CNQDs were synthesized through a simple low-temperature solid-phase method as

Synthesis of PS Colloidal Crystal Template
Polystyrene spheres (PS) suspension with controllable size were synthesized according to the reported emulsion polymerization method [20,[23][24][25]. Styrene monomers were alternately washed with 0.5 M NaOH solution and deionized water in a separate funnel with a 1:1 volume ratio 3 times to remove the polymerization inhibitor in styrene. 0.45 g of sodium dodecyl sulfate (SDS) and 0.6 g of K 2 S 2 O 8 were mixed in 288 mL of deionized water and 32 mL of ethanol (EtOH) with magnetic stirring to form a homogeneous solution; this was followed by injecting 36 mL of washed styrene at nitrogen atmosphere. A milky product was obtained after heating at 71 • C for 19 h. The as-prepared polystyrene emulsion was transferred into 250 mL beakers with a height of~2 cm and then put into a 70 • C oven to evaporate the solvent.

Synthesis of Graphitic Carbon Nitride Quantum Dots (CNQDs)
CNQDs were synthesized through a simple low-temperature solid-phase method as reported [42]. 0.101 g of urea and 0.081 g of sodium citrate were mixed and grounded to powders in an agate mortar, then the mixture was transferred into a Teflon-lined stainlesssteel autoclave (20 mL capacity) and followed by thermal treatment at 180 • C for 2 h in an oven. The resultant mixture was alternately washed with ethanol and centrifuged three times (12,000 rpm for 10 min). A yellowish CNQDs solution was obtained by dialyzing against 20 mL of deionized water through a dialysis membrane with 3500 molecule weight cut-off (MWCO) for 24 h.

Synthesis of TiO 2 IO and TCN IO
The synthesis process of CNQDs incorporated TiO 2 inverse opal (TCN IO) is shown in Figure 9. In detail, 5.6 mL of titanium isopropoxide (TTIP), 45 mL of EtOH and 1 mL of acetylacetone (AcAc) were mixed and stirred. Then, 0.85 mL of hydrochloric acid and 4.6 mL of CNQDs aqueous solution were added into the solution, which was continually stirred for 2 h. Then the precursor solution was dropped on the PS colloidal crystal solids. After hydrolysis for 8 h at room temperature, the samples were calcined at 500 • C for 2 h (heating rate was 2 • C/min). On the other hand, pure TiO 2 inverse opals (TiO 2 IO) were prepared via the same method by replacing CNQDs with 5 mL of deionized water in precursor solution. Additionally, bulk-TiO 2 was synthesized without templates by using the same precursor solution as TiO 2 IO.

Characterization of Materials
X-ray diffraction (XRD) was conducted with a Shimadzu XRD-7000 XRD diffractometer (Shimadzu, Kyoto, Japan) using Cu Kα (λ= 0.15406 nm) radiation. Other test conditions included a current of 100 mA, an operating voltage of 40 kV, a scanning range between 2θ = 5-75°and a scan rate of 0.02°/2 s. Scanning electron microscopy (SEM) was conducted by a JSM-6360 LV electron microscope (Jeol, Tokyo, Japan). A routine analysis consisted of sprinkling the sample on a conductive tape and spraying with gold under a 15 kV work voltage. The transmission electron microscopy (TEM), which was used to characterize the samples' morphologies, was performed on a Jeol JEM-2011 transmission electron microscope (Jeol, Tokyo, Japan) with a 120-200 kV work voltage. The morphologies of commercial and homemade TiO2-based samples were further characterized by HRTEM using a Jeol JEM-2100 (Jeol, Tokyo, Japan). X-ray photoelectron spectroscopy (XPS) was tested with a PerkinElmer PHI 5000C ESCA system(PerkinElmer, Waltham, MA, USA) with Al K radiation (250 W). The Brunauer−Emmett−Teller (BET) surface area of all photocatalysts was obtained via nitrogen adsorption at 77 K by a Micromeritics ASAP2010 (Micromeritics, Norcross, GA, USA). The photoluminescence (PL) spectra of samples were measured with a RF-5301 spectrofluorophotometer (Shimadzu, Kyoto, Japan). With an electrochemical analyzer CHI 660 D electrochemical station (CH Instruments Inc., Bee Cave, TX, USA), the electrochemical experiments were carried out in a cell with a standard three-electrode system. It consisted of a working electrode (as-prepared samples were coated on a square fluoride-tin oxide (FTO) with an area of ca. 0.5 cm −2 ), a Pt wire serving as the counter electrode and a saturated Ag/AgCl acting as the reference electrode. Electrochemical impedance spectroscopy (EIS) of different samples were obtained in the frequency range from 100 kHz to 0.1 Hz under an amplitude of 10 mV using N2-saturated potassium ferricyanide-mixed electrolyte. were coated on a square fluoride-tin oxide (FTO) with an area of ca. 0.5 cm −2 ), a Pt wire serving as the counter electrode and a saturated Ag/AgCl acting as the reference electrode. Electrochemical impedance spectroscopy (EIS) of different samples were obtained in the frequency range from 100 kHz to 0.1 Hz under an amplitude of 10 mV using N 2 -saturated potassium ferricyanide-mixed electrolyte.

Photocatalytic Degradation of VOCs and Dyes
The photocatalytic activities of the prepared photocatalysts were evaluated by the degradations of the gas phase VOCs toluene, the liquid phase VOCs phenol and a typical dye RhB under simulated sunlight irradiation. The photocatalytic reactions were conducted under simulated solar light by a 300 W Xe lamp CEL-HXF300 (Beijing Jin Yuan Technology Co., Beijing, China) with a cut-off filter (AM 1.5), of which the UV-Visible emission spectrum is 350-780 nm.
In the case of photocatalytic degradation of toluene, the distance between the light source and the substrate coated with photocatalysts was 20 cm where the light intensity was 612 mW/cm 2 . For preparation of the substrate, 100 mg of photocatalyst was dispersed in 3.5 mL of EtOH via sonication for 15 min and the obtained suspension was slowly and evenly dropped on a square quartz substrate with an area of 64 cm 2 . The substrate was deposited in a 70 • C oven over night for drying and stabilization. A self-developed Pyrex reactor (total volume of 1.735 L) with a flat quartz window on the top was used, wherein the prepared substrate was placed on a quartz holder inside the reactor ( Figure S3). The reactor was sealed and flushed with air for 30 min, and then 5 µL of liquid toluene were injected into the reactor in a vacuum state and vaporized into gas phase, which corresponded to the initial concentration of 665 ppm (2500 mg/m 3 ). Before the lamp was switched on, the gassolid adsorption equilibrium was reached after 1 h. The photocatalytic oxidation of toluene was evaluated by toluene and CO 2 detection at different time intervals on a Inesa GC126N gas chromatograph (Inesa Analytical Istrument Co., Shanghai, China) equipped with a flame ionization detector (FID), and a methane-reforming furnace [44]. The adsorption efficiency of toluene (ξ t, toluene ) in the dark condition and the photocatalytic degradation efficiencies of toluene into CO 2 ( t, toluene ) over different photocatalysts were calculated according to the following formulas: t, toluene = (C t, CO2 − C 0, CO2 )/7C 0, C7H8 , where C 0, C7H8 and C 0, CO2 are the initial concentrations of C 7 H 8 and CO 2 , respectively, C t, C7H8 and C t, C7H8 are the concentrations of C 7 H 8 and CO 2 at reaction time (t), respectively. For the degradation of phenol and RhB, 50 mg of photocatalyst was added into a quartz reactor containing 50 mL of 10 mg/L phenol solution and 20 mg/L RhB, respectively. Prior to the photocatalytic reaction, the suspension was stirred for 30 min in the dark to achieve the adsorption-desorption equilibrium of organic contaminants on the surface of the catalysts. At the given time interval, the analytical sample was taken from the mixture solution and immediately centrifuged. The concentrations of phenol were analyzed by a Shimadzu SPD-M20A (Shimadzu, Kyoto, Japan) high-performance liquid chromatograph (HPLC), while those of RhB were measured with a Shimadzu 2450 UV-vis spectrophotometer (Shimadzu, Kyoto, Japan).
All tests of photocatalytic activities over samples were conducted three times and the mean values were reported. No significant deviations between the three tests were found.

Conclusions
In this study, carbon nitride quantum dots for in situ-loading TiO 2 inverse opal structures were designed and prepared for photocatalytic degradation of gaseous toluene under simulated sunlight irradiation. Compared with the common catalyst bulk-TiO 2 , TiO 2 IO and commercial P25, the as-prepared TCN IO significantly promoted the degradation of toluene, and a 93% mineralization rate of 665 ppm toluene was achieved within 6 h. The small material can be used for the degradation of liquid phase pollutants such as RhB and phenol. Based on the specific surface area, photoelectrochemical measurements and PL spectra, it can be concluded that the modification of CNQDs and inverse opals structure not only increases the surface area, along with the concentration of active sites, but also decreases the recombination rate of photo-generated electrons and holes. This material, with good stability and green environmental protection, provides a practical solution to the problem of indoor toxic VOCs pollution and wastewater treatment.