One-Pot Synthesis of Nitrogen-Doped TiO2 with Supported Copper Nanocrystalline for Photocatalytic Environment Purification under Household White LED Lamp

Developing efficient and cheap photocatalysts that are sensitive to indoor light is promising for the practical application of photocatalysis technology. Here, N-doped TiO2 photocatalyst with loaded Cu crystalline cocatalyst is synthesized by a simple one-pot method. The structure is confirmed by transmission electron microscopy and X-ray photoelectron spectroscopy analysis, which exhibit that Cu metal nanocrystalline is uniformly deposited on the surface of N-doped TiO2 material. UV-Vis absorption spectra illustrate that the modified samples possess favorable visible light absorption properties and suppressed-electron hole separation. The as-fabricated Cu-loaded N-TiO2 materials show high activity in photocatalytic decomposing isopropanol and inactivating E. coli under the irradiation of a household white LED lamp. The developed synthetic strategy and photocatalytic materials reported here are promising for indoor environment purification.


Introduction
Photocatalysis is regarded as a promising approach for removing pollutants and harmful microbes [1][2][3][4]. To achieve the widespread practical application of photocatalytic environment remediation, efficient photocatalysts that are sensitive to weak visible light, especially indoor light, are required [5,6]. Among the various semiconductor photocatalysts, titanium dioxide (TiO 2 ) is one of the most promising materials for air purification and antibacterial functions due to its unique electronic band structures, prominent chemical stability, nontoxicity, and low cost [7,8]. However, TiO 2 is a typical wide bandgap semiconductor (~3.2 eV), and it is only activated by ultraviolet (UV) irradiation [9,10]. Doping TiO 2 with other elements is realized as an effective method to change the electronic band structure and, subsequently, the light absorption properties [11,12]. Over the past decades, the N-doped TiO 2 , which possesses hybridization of N 2p and O 2p orbits with re-constructed valence band, shows decreased bandgap and visible light sensitivity [13,14]. Up to now, the N-doped TiO 2 is extensively studied as promising catalysts to decompose pollutants and microbes [15][16][17]; however, the activity is still unfavorable due to the easy recombination between photogenerated electrons and holes in bare materials [18,19]. To solve such a problem, loading cocatalysts such as Pt, Au, Ag, and other metals are effective in relieving the recombination of electron-hole pairs [20]. Especially, some non-noble metals such as Cu are more concerned as high activity and inexpensive cocatalyst materials to enhance the photocatalytic activity [21,22].
Up to now, the use of Cu metal cocatalysts to improve the photocatalytic activities over TiO 2 materials has been widely reported. For example, Wu et al. prepared Cu particles deposited TiO 2 by 400 • C calcination/reduction under H 2 (3 mol% in N 2 ) atmosphere after incipient-wetness impregnation, and the photocatalytic activity for hydrogen evolution was significantly enhanced [23]. Chiang et al. synthesized Cu nanoparticles deposited TiO 2 nanorod composites by microwave-assisted sol-gel method and chemically reduced them with sodium borohydride as the reducing agent. The modified TiO 2 enhanced the photocatalytic degradation of bisphenol A [24]; however, previous fabrication of cocatalyst of Cu metal on TiO 2 is fabricated by mixture TiO 2 and Cu 2+ based salt at high temperature in the presence of H 2 . The TiO 2 can also be reduced into TiO 2−x and thus possibly decrease the intrinsic photocatalytic activity. The solvent reduction in the presence of sodium borohydride to reduce Cu 2+ into Cu at mild conditions also seems effective; however, the process is sophisticated, and the reduced Cu metals usually have weak interaction with TiO 2 , which is not beneficial to the photocatalytic process. Developing a facile method to fabricate Cu catalysts on TiO 2 with favorable photocatalytic activity is still limited.
In this study, the Cu-loaded on N-doped TiO 2 were synthesized by the one-pot reaction method. In such a fabrication process, the TiO 2 and Cu 2+ salt simply mixed and then be nitridized in the presence of NH 3 at 775 K. The NH 3 at a high temperature can not only nitridize the TiO 2 into N-doped TiO 2 , but also reduce the Cu 2+ salt into Cu metal. The Cu metal loaded on N-TiO 2 could thus be achieved by the one-pot procedure. The interaction between catalyst and cocatalyst was difficult using such a synthetic process, which benefits the photocatalytic reactions. The UV-Vis absorption spectrum shows that fabricated materials exhibit favorable visible light absorption. The modified samples have a remarkable enhancement in remediating gaseous pollutants and microbes under indoor commercial LED irradiation. The developed one-pot synthetic strategy and Cu-loaded N-TiO 2 materials reported here are promising for the development of efficient photocatalysts for indoor environment purification.

Results and Discussion
The one-pot synthetic process is illustrated in Figure 1. Firstly, the Cu(NO 3 ) 2 and TiO 2 were mixed and ground; the Cu(NO 3 ) 2 thus adsorbed on the surface of TiO 2 nanoparticles. After that, the mixture was calcined under an ammonia atmosphere. The Cu(NO 3 ) 2 was reduced to Cu metal particles on the surface of TiO 2 in such a process: NH 3 → H 2 + N 2 and H 2 + Cu 2+ → Cu + 2H + . Meanwhile, a slight amount of N was doped into TiO 2 to generate N-doped TiO 2 (N-TiO 2 ). Finally, the Cu metal-supported N-doped TiO 2 was produced by the designed one-pot synthetic method. The products were expressed as xCu-N-TiO 2 (the x represented the weight ratio of loaded Cu to N-TiO 2 ). them with sodium borohydride as the reducing agent. The modified TiO2 e photocatalytic degradation of bisphenol A [24]; however, previous fabricati lyst of Cu metal on TiO2 is fabricated by mixture TiO2 and Cu 2+ based salt at ature in the presence of H2. The TiO2 can also be reduced into TiO2−x and t decrease the intrinsic photocatalytic activity. The solvent reduction in the pr dium borohydride to reduce Cu 2+ into Cu at mild conditions also seems ef ever, the process is sophisticated, and the reduced Cu metals usually have w tion with TiO2, which is not beneficial to the photocatalytic process. Develo method to fabricate Cu catalysts on TiO2 with favorable photocatalytic activi ited.
In this study, the Cu-loaded on N-doped TiO2 were synthesized by the tion method. In such a fabrication process, the TiO2 and Cu 2+ salt simply mi be nitridized in the presence of NH3 at 775 K. The NH3 at a high temperature nitridize the TiO2 into N-doped TiO2, but also reduce the Cu 2+ salt into Cu m metal loaded on N-TiO2 could thus be achieved by the one-pot procedure. Th between catalyst and cocatalyst was difficult using such a synthetic process, fits the photocatalytic reactions. The UV-Vis absorption spectrum shows th materials exhibit favorable visible light absorption. The modified samples ha able enhancement in remediating gaseous pollutants and microbes under ind cial LED irradiation. The developed one-pot synthetic strategy and Cu-loade terials reported here are promising for the development of efficient photoca door environment purification.

Results and Discussion
The one-pot synthetic process is illustrated in Figure 1. Firstly, the Cu(N were mixed and ground; the Cu(NO3)2 thus adsorbed on the surface of TiO2 n After that, the mixture was calcined under an ammonia atmosphere. The C reduced to Cu metal particles on the surface of TiO2 in such a process: NH3 → H2 + Cu 2+ → Cu + 2H + . Meanwhile, a slight amount of N was doped into TiO N-doped TiO2 (N-TiO2). Finally, the Cu metal-supported N-doped TiO2 was the designed one-pot synthetic method. The products were expressed as xCu x represented the weight ratio of loaded Cu to N-TiO2).  The X-ray diffraction (XRD) patterns of the pristine TiO 2 , N-TiO 2 , and xCu-N-TiO 2 samples are shown in Figure 2. All peaks of the as-prepared samples consist of anatase phases TiO 2 (JCPDS card. 21-1272) without other impurity peaks observed [25]. The half-peak width of all products is not changed significantly, indicating that the crystallinity and grain size are almost kept. The copper and copper oxide are not observed, which is probably due to the fact that the amount of loaded Cu-based materials is too low to be detected by the XRD technical.
The X-ray diffraction (XRD) patterns of the pristine TiO2, N-TiO2, and samples are shown in Figure 2. All peaks of the as-prepared samples cons phases TiO2 (JCPDS card. 21-1272) without other impurity peaks observed peak width of all products is not changed significantly, indicating that th and grain size are almost kept. The copper and copper oxide are not obser probably due to the fact that the amount of loaded Cu-based materials is detected by the XRD technical. Scanning electron microscopy (SEM) and transmission electron micro were employed to characterize the morphologies and compositions of th samples. Figures 3a and S1 show the typical SEM images, which demonstra N-TiO2, N-TiO2, and pristine TiO2 photocatalysts have nanoparticle-like where the size of these particles is about 20-50 nm. After nitridation and C size and morphology of the TiO2 are not changed obviously. The TEM imag exhibits that materials of 1%Cu-N-TiO2 are uniformly dispersed nanoparti identical to the SEM results. The selected area electronic diffraction (SAED plays well resolved (101), (004), and (205) diffraction rings (inset of Figure 3 responds to the structural characteristic of anatase TiO2 [26]. The enlarged T Figure 3c further shows that the shape of each particle is an irregular shape w distributed size from 20-50 nm. The high-resolution TEM (HRTEM) image o ( Figure 3d) shows that the lattice fringe distance of the nanoparticles was 0.35 and 0.36 nm, which corresponds to the (101) and (101 ) face of anatase tively. In addition, some nanoparticles with a size of about 4 nm are loaded of TiO2. As the HRTEM image illustrated, the lattice fringe is measured as 0 is consistent with the copper phase of Cu metal (111). Such a result indicat metal nanoparticles are loaded on TiO2 nanocrystals. To further con Cu(NO3)2·3H2O can be converted into Cu metal under the designed synthe proper Cu(NO3)2·3H2O powder was calcined in ammonia atmosphere unde out the addition of TiO2. The XRD pattern in Figure S2 shows that all peaks w to Cu metal. So, the Cu nanocrystalline loaded on N-doped TiO2 nanoparti Scanning electron microscopy (SEM) and transmission electron microscopy (TEM) were employed to characterize the morphologies and compositions of the as-prepared samples. Figure 3a and Figure S1 show the typical SEM images, which demonstrate that 1%Cu-N-TiO 2 , N-TiO 2 , and pristine TiO 2 photocatalysts have nanoparticle-like morphology where the size of these particles is about 20-50 nm. After nitridation and Cu loading, the size and morphology of the TiO 2 are not changed obviously. The TEM image in Figure 3b exhibits that materials of 1%Cu-N-TiO 2 are uniformly dispersed nanoparticles, which is identical to the SEM results. The selected area electronic diffraction (SAED) pattern displays well resolved (101), (004), and (205) diffraction rings (inset of Figure 3b), which corresponds to the structural characteristic of anatase TiO 2 [26]. The enlarged TEM image in Figure 3c further shows that the shape of each particle is an irregular shape with a narrow-distributed size from 20-50 nm. The high-resolution TEM (HRTEM) image of the product (Figure 3d) shows that the lattice fringe distance of the nanoparticles was measured as 0.35 and 0.36 nm, which corresponds to the (101) and (101) face of anatase TiO 2 , respectively. In addition, some nanoparticles with a size of about 4 nm are loaded on the surface of TiO 2 . As the HRTEM image illustrated, the lattice fringe is measured as 0.20 nm, which is consistent with the copper phase of Cu metal (111). Such a result indicates that the Cu metal nanoparticles are loaded on TiO 2 nanocrystals. To further confirm that the Cu(NO 3 ) 2 ·3H 2 O can be converted into Cu metal under the designed synthetic condition, proper Cu(NO 3 ) 2 ·3H 2 O powder was calcined in ammonia atmosphere under 500 • C without the addition of TiO 2 . The XRD pattern in Figure S2 shows that all peaks were assigned to Cu metal. So, the Cu nanocrystalline loaded on N-doped TiO 2 nanoparticles is reasonable in our materials synthesis route.  In the following study, the Cu states in 1%Cu-N-TiO2 were deter Cu 2p XPS spectra. As shown in Figure 4a, the product of 1%Cu-N-TiO peak located at 932.47 eV, which could be assigned to Cu 0 state [21]. Su that the synthesized product is loaded with Cu metal. XPS element con S1) exhibits that the amount of loaded Cu is about 2 wt%, which is h signed amount in our synthetic experiment. The variety in the amoun because the XPS is a surface analysis technology, which detects the surface, which leads to the higher value of loaded Cu. The formed Cu on TiO2 may partly be oxidized into the oxidation state of Cu2O or CuO to air conditions; however, these characterizations show that the oxid such as the CuO and Cu2O, are not obvious. Figure 4b presents the state in N-TiO2 and 1%Cu-N-TiO2 samples. The binding energy range o 404 eV, which is a typical N 1s XPS spectra for N-doped TiO2 [27]. Th of both samples are at 399.6 eV, which is considered as the form of Ti-So, the N element has been successfully doped into the TiO2. Figure 4 XPS spectra of Ti 2p and O1s state in N-TiO2 and 1%Cu-N-TiO2 strength and peak position of the Ti 2p and O1s XPS spectra of the almost unchanged, indicating that Cu nanoparticles were only depos of N-TiO2 photocatalyst, which did not influence the chemical states o In the following study, the Cu states in 1%Cu-N-TiO 2 were determined through the Cu 2p XPS spectra. As shown in Figure 4a, the product of 1%Cu-N-TiO 2 displays one main peak located at 932.47 eV, which could be assigned to Cu 0 state [21]. Such a result suggests that the synthesized product is loaded with Cu metal. XPS element content analysis (Table S1) exhibits that the amount of loaded Cu is about 2 wt%, which is higher than the designed amount in our synthetic experiment. The variety in the amount of loaded may be because the XPS is a surface analysis technology, which detects the atoms on and near surface, which leads to the higher value of loaded Cu. The formed Cu metal nanoparticle on TiO 2 may partly be oxidized into the oxidation state of Cu 2 O or CuO once it is exposed to air conditions; however, these characterizations show that the oxidation states of Cu, such as the CuO and Cu 2 O, are not obvious. Figure 4b presents the XPS spectra of N1s state in N-TiO 2 and 1%Cu-N-TiO 2 samples. The binding energy range of N1s peaks is 396-404 eV, which is a typical N 1s XPS spectra for N-doped TiO 2 [27]. The binding energies of both samples are at 399.6 eV, which is considered as the form of Ti-N-O bonds [27,28]. So, the N element has been successfully doped into the TiO 2 . Figure 4c and d present the XPS spectra of Ti 2p and O1s state in N-TiO 2 and 1%Cu-N-TiO 2 samples. The peak strength and peak position of the Ti 2p and O1s XPS spectra of the two samples were almost unchanged, indicating that Cu nanoparticles were only deposited on the surface of N-TiO 2 photocatalyst, which did not influence the chemical states of Ti and O. The UV-Vis absorption spectra of as-prepared materials were studied. Figure 5a shows that pristine TiO2 have an absorption edge at about 380 nm. The N-doped TiO2 products have much wider absorption edges. Figure 5b shows the enlarged areas from Figure 5a. It is observed that the N-TiO2 and the Cu loaded can N-TiO2 reach the absorption edge at about 540-560 nm. So, the light absorption ability is significantly enhanced. This can be attributed to the formation of a local intermediate band (N 2p) energy level at the top of the O 2p valence band in TiO2 by the N introduction. The band gap of the semiconductor is narrowed so that longer wavelengths of light can be absorbed to form photogenerated electrons and holes [29,30]. It is worth noting that the loaded Cu metal particles have no obvious enhancement on the UV-Vis absorption spectra, which is attributed to the fact that the Cu element was only loaded on the surface of TiO2, which did not change the electronic energy band structure of TiO2.  The UV-Vis absorption spectra of as-prepared materials were studied. Figure 5a shows that pristine TiO 2 have an absorption edge at about 380 nm. The N-doped TiO 2 products have much wider absorption edges. Figure 5b shows the enlarged areas from Figure 5a. It is observed that the N-TiO 2 and the Cu loaded can N-TiO 2 reach the absorption edge at about 540-560 nm. So, the light absorption ability is significantly enhanced. This can be attributed to the formation of a local intermediate band (N 2p) energy level at the top of the O 2p valence band in TiO 2 by the N introduction. The band gap of the semiconductor is narrowed so that longer wavelengths of light can be absorbed to form photogenerated electrons and holes [29,30]. It is worth noting that the loaded Cu metal particles have no obvious enhancement on the UV-Vis absorption spectra, which is attributed to the fact that the Cu element was only loaded on the surface of TiO 2 , which did not change the electronic energy band structure of TiO 2 .  The UV-Vis absorption spectra of as-prepared materials were studied. Figure 5a shows that pristine TiO2 have an absorption edge at about 380 nm. The N-doped TiO2 products have much wider absorption edges. Figure 5b shows the enlarged areas from Figure 5a. It is observed that the N-TiO2 and the Cu loaded can N-TiO2 reach the absorption edge at about 540-560 nm. So, the light absorption ability is significantly enhanced. This can be attributed to the formation of a local intermediate band (N 2p) energy level at the top of the O 2p valence band in TiO2 by the N introduction. The band gap of the semiconductor is narrowed so that longer wavelengths of light can be absorbed to form photogenerated electrons and holes [29,30]. It is worth noting that the loaded Cu metal particles have no obvious enhancement on the UV-Vis absorption spectra, which is attributed to the fact that the Cu element was only loaded on the surface of TiO2, which did not change the electronic energy band structure of TiO2.   The degradation of gaseous isopropanol (IPA) under white LED light irradiation is evaluated to reflect the photocatalytic performance of the prepared materials. IPA was chosen as the target molecule because it is a representative volatile organic compound for gaseous pollutants [31]. The wavelength of the indoor commercial white LED light used as a light source with a wavelength of 400-700 nm and light intensity of approximately 8 mW/cm 2 ( Figure S3). Figure 6a shows a comparative study of the photocatalytic activity of acetone increases at full speed in the first 80 min over N-TiO 2 and different Cu loaded N-TiO 2 products. It could be found that the evolution rates of acetone over 2%Cu-N-TiO 2 , 1%Cu-N-TiO 2 , 0.5%Cu-N-TiO 2 , and N-TiO 2 reach about 420.0, 662.3, 364.5, and 70.0 ppm h −1 , respectively. The 1%Cu-N-TiO 2 shows the best activity in the decomposing IPA, which is about 9.5 folds as N-TiO 2 . Obviously, Cu nanoparticles on the surface of N-TiO 2 can act as cocatalysts and suppresses the recombination of the photogenerated hole-electron pairs, which improve the photocatalytic activity [32]. Figure 6b shows a complete process of IPA degradation under an LED light. Firstly, the concentration of IPA decrease fast with and the concentration of acetone increase rapidly (CH 3 CHOHCH 3 + e − + O 2 + H + → CH 3 COCH 3 + HO· + H 2 O or CH 3 CHOHCH 3 + h + → CH 3 COCH 3 + 2H + + e − ) [33]. When the concentration of acetone reaches a maximum level, the concentration of acetone starts to be decomposed into CO 2, and the evolved CO 2 increases rapidly. Finally, the evolved concentration of CO 2 reaches about three times as initial concentration of IPA, indicating that 1%Cu-N-TiO 2 photocatalyst can mineralize IPA to CO 2 completely (CH 3 CHOHCH 3 + 5H 2 O + 18h + → 3CO 2 + 18H + ) [33,34]. The degradation of gaseous isopropanol (IPA) under white LED light irradiation is evaluated to reflect the photocatalytic performance of the prepared materials. IPA was chosen as the target molecule because it is a representative volatile organic compound for gaseous pollutants [31]. The wavelength of the indoor commercial white LED light used as a light source with a wavelength of 400-700 nm and light intensity of approximately 8 mW/cm 2 ( Figure S3). Figure 6a shows a comparative study of the photocatalytic activity of acetone increases at full speed in the first 80 min over N-TiO2 and different Cu loaded N-TiO2 products. It could be found that the evolution rates of acetone over 2%Cu-N-TiO2, 1%Cu-N-TiO2, 0.5%Cu-N-TiO2, and N-TiO2 reach about 420.0, 662.3, 364.5, and 70.0 ppm h −1 , respectively. The 1%Cu-N-TiO2 shows the best activity in the decomposing IPA, which is about 9.5 folds as N-TiO2. Obviously, Cu nanoparticles on the surface of N-TiO2 can act as cocatalysts and suppresses the recombination of the photogenerated hole-electron pairs, which improve the photocatalytic activity [32]. Figure 6b shows a complete process of IPA degradation under an LED light. Firstly, the concentration of IPA decrease fast with and the concentration of acetone increase rapidly (CH3CHOHCH3 + e − + O2 + H + → CH3COCH3 + HO· + H2O or CH3CHOHCH3 + h + → CH3COCH3+ 2H + + e -) [33]. When the concentration of acetone reaches a maximum level, the concentration of acetone starts to be decomposed into CO2, and the evolved CO2 increases rapidly. Finally, the evolved concentration of CO2 reaches about three times as initial concentration of IPA, indicating that 1%Cu-N-TiO2 photocatalyst can mineralize IPA to CO2 completely (CH3CHOHCH3 + 5H2O + 18h + → 3CO2 + 18H + ) [33,34]. The Brunner−Emmet−Teller (BET) specific surface areas of the 1%Cu-N-TiO2 and N-TiO2 were determined using N2 adsorption and desorption isotherms ( Figure S4). The BET surface area of N-doped TiO2 and 1%Cu-N-TiO2 are approximate 29.7 and 31.9 m 2 g −1 , which are quite closed. So, the degradation of activity is thus mainly influenced by the loading of Cu rather than the surface areas.
In the following study, the sterilization performance of 1%Cu-N-TiO2 was evaluated under irradiation of indoor LED light. E. coli is one of the most common bacteria as a model microbe to evaluate the inactivation performance for many sterilization materials [35]. Figure 7a shows the corresponding activities of 1%Cu-N-TiO2 and N-TiO2. The survival rate of E. coli over 1%Cu-N-TiO2 reaches 10 −7 after 100 min, shows that the E. coli is completely sterilized. For comparison, the survival rate of E. coli over N-TiO2 is only 0.33 in 100 min. So, the 1%Cu-N-TiO2 shows much enhanced activity compared to N-TiO2. In addition, the control experiments were carried out to prove that the sterilization effect is caused by the photocatalytic process over 1%Cu-N-TiO2. When the 1%Cu-N-TiO2 photocatalyst is placed in the bacterial culture medium under dark conditions, the bacteria The Brunner−Emmet−Teller (BET) specific surface areas of the 1%Cu-N-TiO 2 and N-TiO 2 were determined using N 2 adsorption and desorption isotherms ( Figure S4). The BET surface area of N-doped TiO 2 and 1%Cu-N-TiO 2 are approximate 29.7 and 31.9 m 2 g −1 , which are quite closed. So, the degradation of activity is thus mainly influenced by the loading of Cu rather than the surface areas.
In the following study, the sterilization performance of 1%Cu-N-TiO 2 was evaluated under irradiation of indoor LED light. E. coli is one of the most common bacteria as a model microbe to evaluate the inactivation performance for many sterilization materials [35]. Figure 7a shows the corresponding activities of 1%Cu-N-TiO 2 and N-TiO 2 . The survival rate of E. coli over 1%Cu-N-TiO 2 reaches 10 −7 after 100 min, shows that the E. coli is completely sterilized. For comparison, the survival rate of E. coli over N-TiO 2 is only 0.33 in 100 min. So, the 1%Cu-N-TiO 2 shows much enhanced activity compared to N-TiO 2 . In addition, the control experiments were carried out to prove that the sterilization effect is caused by the photocatalytic process over 1%Cu-N-TiO 2 . When the 1%Cu-N-TiO 2 photocatalyst is placed in the bacterial culture medium under dark conditions, the bacteria surviving is closer to 100% after 120 min. So, there is no toxic effect of Cu-N-TiO 2 on E. coli cells under dark conditions. When the bacterial medium is placed under LED light irradiation without catalyst, there is no obvious decrease in E. coli cells after 120 min. So, the activity is certain from the photocatalytic effects. Figure 7b,c show the images of E. coli colonies on agar plates before (diluted 10 5 times) and after (not diluted) LED light irradiation for 100 min in the presence of 1%Cu-N-TiO 2 . During the photocatalytic process, the photogenerated electrons and holes pairs react with water to produce reactive oxygen species. The reactive oxygen species include HO • , O 2 •− , HO 2 • , etc., and can destroy the E. coli cell and results in bacterial inactivation [3]. surviving is closer to 100% after 120 min. So, there is no toxic effect of Cu-N-TiO2 on E. coli cells under dark conditions. When the bacterial medium is placed under LED light irradiation without catalyst, there is no obvious decrease in E. coli cells after 120 min. So, the activity is certain from the photocatalytic effects. Figure 7b,c show the images of E. coli colonies on agar plates before (diluted 10 5 times) and after (not diluted) LED light irradiation for 100 min in the presence of 1%Cu-N-TiO2. During the photocatalytic process, the photogenerated electrons and holes pairs react with water to produce reactive oxygen species. The reactive oxygen species include HO • , O2 •− , HO2 • , etc., and can destroy the E. coli cell and results in bacterial inactivation [3].     surviving is closer to 100% after 120 min. So, there is no toxic effect of Cu-N-TiO2 on E. coli cells under dark conditions. When the bacterial medium is placed under LED light irradiation without catalyst, there is no obvious decrease in E. coli cells after 120 min. So, the activity is certain from the photocatalytic effects. Figure 7b,c show the images of E. coli colonies on agar plates before (diluted 10 5 times) and after (not diluted) LED light irradiation for 100 min in the presence of 1%Cu-N-TiO2. During the photocatalytic process, the photogenerated electrons and holes pairs react with water to produce reactive oxygen species. The reactive oxygen species include HO • , O2 •− , HO2 • , etc., and can destroy the E. coli cell and results in bacterial inactivation [3].

Preparation of Photocatalysts
In a typical process, 1.0 g TiO 2 (anatase phase, 10-25 nm grain size, Sigma-Aldrich Co. St. Louis, MO, USA) and Cu(NO 3 ) 2 ·3H 2 O (AR, Sinopharm Chemical Reagent Co., Ltd., Shanghai, China) were well mixed with a certain ratio and ground. The mixture was calcined at 773 K for about 3 h under an ammonia atmosphere to obtain Cu metal-loaded N-doped TiO 2 nanoparticles (xCu-N-TiO 2 ). For comparison, pure TiO 2 powder was also calcined at 773 K in an ammonia atmosphere for 3 h without Cu, which was expressed as N-TiO 2 .

Photocatalytic Degradation of Pollutants
The photocatalytic decomposition of gaseous isopropanol (IPA) was evaluated under white LED illumination. Figure S5 shows the reaction cell used in IPA degradation experiments. In a typical procedure, a 600 mL glass container was used as the photocatalytic vessel reactor. A commercial white LED (30 W, OPPLE, Shanghai, China) was located at 10 cm from the vessel reactor. About 100 mg photocatalysts were dispersed in a 9.5 cm 2 circular glass dish, which was located in the center of the vessel reactor. The reactor was filled with fresh synthetic air (20% oxygen and 80% nitrogen) for 1 h. The pressure inside the reactor was kept at 1 Kpa (Atmospheric pressure). Then, about 1000 ppm of gaseous IPA was injected into the reactor. The reactor was placed in darkness for 3 h to reach adsorption equilibrium. When the concentration of IPA remained constant, the adsorption equilibrium of IPA was considered to have been reached. During the process of IPA adsorption equilibrium, no products such as acetone or CO 2 were detected, indicating that the gaseous IPA could not be degraded by photocatalyst under dark conditions. The reactor was then irradiated with an LED lamp, and 0.5 mL products along the reactions were regularly extracted from the reactor. The concentrations of IPA, acetone, and CO 2 were measured by a gas chromatograph (GC-2014, Shimadzu, Kyoto, Japan).

Evaluation of Anti-Bacteria
The anti-bacteria activities of photocatalysts were evaluated by killing E. coli cells. The E. coli was cultured in Luria Bertani (LB) nutrient medium at 37 • C and shocked for 24 h. Then, the cultured E. coli cells were washed by centrifugation at 5000 rpm and were resuspended and diluted to~1 × 107 cfu (colony forming unit)/mL in sterilized 0.9% saline solution. All glass apparatuses and consumables used in the experiments were sterilized in an autoclave at 121 • C for 25 min. About 30 mg photocatalyst was added to a glass reactor containing 30 mL diluted E. coli suspension. The reaction mixture was covered with quartz glass and stirred during the experiment. The light source was also produced from a commercial white LED (30 W, OPPLE) which was located 15 cm from the E. coli solution. The experimental temperature of photocatalysis was kept at about 25 • C by circulating cooling water. At regular intervals, an aliquot of the reaction solution was collected and diluted with saline solution, and 0.1 mL of the diluted solution was spread on a nutrient agar plate. The diluted E. coli cells were incubated at 37 • C for 24 h and then the number of viable cells was determined by counting colonies.

Conclusions
In summary, the Cu metal nanocrystalline-loaded N-TiO 2 photocatalysts were synthesized by a simple one-pot method. TEM, XPS, and UV-Vis absorption spectra confirm that Cu metal nanocrystalline is successfully deposited on N-doped TiO 2 photocatalysts and exhibits favorable visible light absorption ability. Under the irradiation of indoor white LED light, the IPA photodegradation rate over optimal 1%Cu-N-TiO 2 is about 662.3 ppm h −1 , which is about 9.5 folds as N-doped TiO 2 (70.0 ppm h −1 ). The E. coli could be completely killed by 1%Cu-N-TiO 2 under LED light irradiation in 120 min, which is significantly improved in comparison with N-TiO 2 . The developed synthetic strategy and photocatalytic materials reported here are promising for the development of photocatalysts for indoor environment purification.