Ultrafast synthesis of near-zero-cost S-doped Ni(OH)2 on C3N5 under ambient conditions with enhanced photocatalytic activity

Planting highly efficient and low-cost Ni-based noble-metal-free active sites on semiconductors is of great significance in the field of photocatalysis. Herein, taking wide visible-light-responsive 2D C3N5 as a model semiconductor, an impressive near-zero-cost 2D S-doped nickel hydroxide (S–Ni(OH)2) is grown on C3N5 ultrafast within 30 min under ambient conditions by facile reaction between extremely low-cost Ni(NO3)2 and Na2S in aqueous solution. The fabricated 2D S–Ni(OH)2–C3N5 hybrid exhibits enhanced photocatalytic performance for both H2 production from water and NO removal for air purification. The H2 production rate on S–Ni(OH)2–C3N5 is ∼7 times higher than that of Ni(OH)2–C3N5 and even slightly higher than that of Pt–C3N5, demonstrating its potential as a candidate for noble metal catalysts like Pt. In particular, an apparent quantum yield (AQY) value of 30.9% at 420 nm for H2 production is reached on 1.0 wt% S–Ni(OH)2–C3N5 due to quick internal charge transfer efficiency. In addition, ∼42% of NO can be purified in a continuous flow reaction system. This work affords a cost-efficient strategy to steer the photocatalytic property of Ni-based catalysts.


Introduction
In the past decades, the ever-increasing energy crisis and environmental pollution have drawn wide concern considering the sustainable development of human society and public health. [1][2][3] Among the various clean technologies, the environmentally friendly semiconductor-based photocatalytic oxidation and reduction is deemed as the most potential one. [4][5][6][7] In this eld, photocatalytic solar H 2 production and NO removal lie in the core attributed to the huge energy consumption and the high risk respiratory diseases. Up to now, different semiconductors including metal oxides/suldes (TiO 2 , CdS, ZnIn 2 S 4 , etc.), [8][9][10][11][12] metal-free materials (carbon nitride, COFs, conjugated polymers, etc.) [13][14][15][16][17] and heterojunctions are all widely investigated to solve these issues. 12,[18][19][20] Of these strategies, constructing a heterojunction between different materials to combine the superiority of each one can signicantly improve the whole photocatalytic efficiency. For example, K. Domen et al. selectively photodeposited Rh/Cr 2 O 3 and CoOOH on different facets of the SrTiO 3 :Al photocatalyst as cocatalysts for hydrogen and oxygen evolution respectively and an external quantum efficiency up to 96% at 350-360 nm was obtained. 21 Recently, carbon nitride has drawn enormous attention due to its low-cost, high stability and suitable band position for water splitting; [22][23][24][25] however, the band-gap of conventional carbon nitride ($2.7 eV for C 3 N 4 ) is still too wide to obtain ideal photocatalytic activity. To alter the photocatalytic efficiency of carbon nitride, structural and band position regulation by doping (S, O, P, N-or Crich modication) and fabricating heterojunctions (TiO 2 /C 3 N 4 , CdS/ C 3 N 4 , CNN/BDCNN, etc.) with type II or Z/S-scheme mechanisms are widely adopted. 15,[26][27][28][29][30][31] Recently, N-rich carbon nitride (i.e. C 3 N 5 ) was reported; [31][32][33][34][35][36] It possessed a narrower band-gap and higher catalytic activity than common C 3 N 4 , which represents a new direction for exploring carbon nitride based materials. For example, our previous work found that photocatalytic H 2 production activity over Pt-C 3 N 5 was $2.2 times higher than that of Pt-C 3 N 4 . 36 Nickel based materials exhibit excellent catalytic reduction and oxidation activities in the eld of both photocatalysis and electrocatalysis, [37][38][39][40][41][42][43] and thus can be used as co-catalysts on C 3 N 5 . Moreover, it has been reported that S doping can signicantly improve the electrocatalytic performance of Ni/Fe (oxy)hydroxide materials; 44 thus doping of S in nickel-based cocatalysts may further improve their performance. However, these Nibased species were commonly synthesized by hydrothermal or heattreatment at high temperature, which was time and cost exhaustive. Therefore, construction of highly effective Ni-based catalysts in short time with low cost and energy consumption is of great importance for potential industrialization of photocatalytic technology.
Herein, taking 2D C 3 N 5 as a model semiconductor, 2D Sdoped Ni(OH) 2 (denoted as S-Ni(OH) 2 ) was facilely planted on its surface by a one-step precipitation method between Ni(NO 3 ) 2 and Na 2 S in water (co-existence of S 2À and OH À ) under stirring within 30 minutes at room temperature, as revealed in Scheme 1. The hybrid photocatalyst (S-Ni(OH) 2 -C 3 N 5 ) was evaluated by photocatalytic activity for H 2 production and NO oxidation removal. Experimental results depicted that S-Ni(OH) 2 can promote the photogenerated e À / h + separation of C 3 N 5 aer light excitation and favor the generation of active oxygen species (ROS) participating in the subsequent photocatalytic procedure. In particular, the visible-light-induced photocatalytic H 2 production rate over S-Ni(OH) 2 -C 3 N 5 is even higher than that of Pt-C 3 N 5 , while the cost of S-Ni(OH) 2 (taking the H 2 production test as an example: in 45 mg 1.0 wt% S-Ni(OH) 2 -C 3 N 5 ) is calculated to be just $0.0032 U or $0.00049 $ based on the price of Ni(NO 3 ) 2 and Na 2 S. The present work provides new insights into the design of low-cost and noble-metal-free catalysts for energy and environmental applications.

Material preparation
All chemicals used were purchased from Aladdin and used without purication. C 3 N 5 was prepared by heating 3-amino-1,2,4-triazole at 500 C in air according to our previous work. 36 Generally, S-Ni(OH) 2 -C 3 N 5 was synthesized by mixing a certain amount of C 3 N 5 , Ni(NO 3 ) 2 and Na 2 S in water and stirring for 30 min. And then, the precipitate was washed with distilled water and absolute ethanol several times and dried in a vacuum at 80 C for 12 h. The loading amount of S-Ni(OH) 2 on C 3 N 5 was calculated based on the initial input nickel mass fraction in the hybrid, and a series of S-Ni(OH) 2 -C 3 N 5 hybrids with S-Ni(OH) 2 loading amount varying between 0.5 and 3.0 wt% were prepared by changing the amount of Ni(NO 3 ) 2 and Na 2 S accordingly. Pristine S-Ni(OH) 2 was also prepared by the same method in the absence of C 3 N 5 . By the way, Ni(OH) 2 -C 3 N 5 was prepared by the same method using NaOH instead of Na 2 S.

Catalyst characterization
XRD patterns were recorded on a powder X-ray diffractometer with Cu Ka radiation (D8 Advance Bruker Inc., Germany). TEM and HRTEM images were acquired on an FEI TALOS F200. SEM and EDS mapping images were acquired on a Gemini Sigma 300. The Brunauer-Emmett-Teller (BET) specic surface area was evaluated using nitrogen adsorption-desorption apparatus (ASAP 2040, Micrometrics Inc., USA) with all samples degassed at 120 C for 12 h prior to measurements. Valence state of each element in the catalyst was analyzed by X-ray photoelectron spectroscopy (XPS) on a Thermo Fisher ESCALAB 250Xi X-ray photoelectron spectroscope. A Shimadzu UV-3100 recording spectrophotometer and uorescence spectrometer (F-4600, Hitachi Inc., Japan) were used to record UV-vis diffuse reectance (DRS) and PL spectra measurements respectively. Time-resolved uorescence spectra (TRFS) were obtained on an Edinburgh FLS 1000. Electron paramagnetic resonance (EPR) spectra were recorded on a JES-FA200 EPR spectrometer. 5,5-Dimethyl-1-pyrroline-N-oxide (DMPO) was used as a radical spintrapping reagent for cOH and cO 2 À . 2,2,6,6-Tetramethylpiperidine (TEMP) was used as the trapping agent for 1 O 2 . 3 The photocurrent and electrochemical impedance measurements were recorded on a CHI 660D electrochemical workstation (Chenhua Instrument, Shanghai, China) with a conventional three electrode conguration. Pt foil and Ag/AgCl (saturated KCl) were used as the counter electrode and reference electrode respectively. The working electrode was made by spreading a catalyst/Naon slurry on FTO glass and 0.1 M Na 2 SO 4 aqueous solution was used as the electrolyte.

Photocatalytic activity tests and photoelectrochemical measurements
The photocatalytic H 2 production rate and AQY measurements were carried out on a top-irradiated reaction vessel containing the catalyst and sacricial reagent (TEOA) connected to a closed gas system with a 300 W Xe-lamp (PLS SXE300, Beijing Perfectlight Inc., China) equipped with a cutoff lter (l > 420 nm) or band-pass lters (l ¼ 420, 500 nm, etc.); the photocatalytic H 2 production system is shown in Fig. S1. † The H 2 production rate was detected using a GC (SP7820, TCD detector, 5Å molecular sieve columns, and Ar carrier), and AQY values were calculated according to our previous work: 17 For photocatalytic NO removal, a continuous ow reactor under ambient conditions was adopted according to our previous work. 7 The volume of the rectangular reactor was 4.5 L (30 cm Â 15 cm Â 10 cm). 25 mg catalyst was dispersed into the mixture of ethanol and water and then transferred into a culture dish with a diameter of 12 cm, and then the dish was placed in the reactor aer the solvent was dried at 45 C. A 30 W visible LED (General Electric) was used as a light source. The gas (containing NO and air) ow rate through the reactor was controlled at 1000 mL min À1 using a mass ow controller with an initial NO concentration of $600 ppb. The NO and NO 2 concentrations were recorded on a NO x analyzer model T200 (Teledyne API). The generation of NO 3 À was detected by ion chromatography on a Thermo DIONEX ICS-900.

Results and discussion
Crystal phase and micro morphology analyses The pristine S-Ni(OH) 2 prepared from the ultrafast reaction between Ni 2+ and Na 2 S in water under ambient conditions within 30 min is rstly detected by XRD, as depicted in Fig. S2, † which demonstrates the formation of S-Ni(OH) 2 with low diffraction peaks corresponding to the PDF card of 73-1520 for Ni(OH) 2 , indicating the low crystallinity of the present S-Ni(OH) 2 material. Herein, the conrmation of S doping will be illustrated in the following section. It should be noted that pristine Ni(OH) 2 without S doping is prepared by reaction between Ni 2+ and NaOH for comparison, and the diffraction peaks of this Ni(OH) 2 are also in agreement with the same PDF card (Fig. S2 †), indicating that the comparison is reasonable. For the XRD patterns of C 3 N 5 and S-Ni(OH) 2 -C 3 N 5 hybrid (Fig. 1a), only two distinct peaks corresponding to (100) and (002) planes of C 3 N 5 , 32,35,36,45 are observed even with a high loading amount of S-Ni(OH) 2 up to 3.0 wt%, which should relate to the low crystallinity of present S-Ni(OH) 2 . In addition, no obvious shi happens in the main peaks of C 3 N 5 aer the loading of S-Ni(OH) 2 , revealing that the procedure for hybrid preparation does not destroy the basic structure of C 3 N 5 .
The element component of 1.0 wt% S-Ni(OH) 2 -C 3 N 5 is analyzed by elemental analysis and ICP-MS (Table S1 †). From the elemental analysis result, the C/N ratio of the catalysts was determined to be 3 : 4.7, which is much lower than that of C 3 N 4 and very close to that of C 3 N 5 , and the result is in good agreement with previous literature. 35 As revealed in Fig. 1b, S-Ni(OH) 2 -C 3 N 5 is composed of C, N, Ni, O and S elements, demonstrating the existence of S-doping in Ni(OH) 2 during the reaction between Ni 2+ and Na 2 S preliminarily. In addition, the result depicts the successful combination of S-Ni(OH) 2 and C 3 N 5 . The high resolution spectra of C 1s (Fig. S3a †) and N 1s (Fig. S3b †) in S-Ni(OH) 2 -C 3 N 5 are similar to that of pristine C 3 N 5 shown in our previous work, 36 and no obvious peak shis are found, revealing that the structure of C 3 N 5 is well kept during the loading of S-Ni(OH) 2 . The existence of Ni ions is conrmed by the high resolution spectrum of Ni 2p, in which two distinct spin-orbit peaks at $855.8 (Ni 2p 3/2 ) and $873.5 eV (Ni 2p 1/2 ) along with two satellite peaks (identied as ''Sat.'') are clearly observed (Fig. 1c). 40,42 The O 1s spectra mainly consist of a peak at 531.79 eV, (see Fig. S3c †) which can be assigned to the O 2À species in Ni(OH) 2 . 46 Most importantly, two peaks are observed in the high resolution spectrum of S 2p, in which the peak centered at $164 eV represents the metal-sulfur bond and S 2À , while the other peak in the high binding energy region can be considered as residual SO 4 2À adsorbed on the material (Fig. 1d). 40,42,47 That is, the existence of the Ni-S bond can support the formation of S-doping in Ni(OH) 2 .
To further conrm the successful fabrication of the S-Ni(OH) 2 -C 3 N 5 hybrid, TEM and SEM images are acquired and shown in Fig. 2. It has been reported in our previous work that the micro morphology of pristine C 3 N 5 is nanosheet. 36 Fig. 2a and b display the TEM and HRTEM images of S-Ni(OH) 2 . As shown in Fig. 2a, S-Ni(OH) 2 also exhibits a micro morphology of nanosheets and a clear lattice fringe of 0.23 nm corresponding to the (011) plane of S-Ni(OH) 2 is observed (Fig. 2b). In particular, no lattice fringe representing nickel sulde is found, demonstrating the existence state of S-doping in S-Ni(OH) 2 . Fig. 2c displays the TEM image of 1.0 wt% S-Ni(OH) 2 -C 3 N 5 , in which 2d nanosheets containing S-Ni(OH) 2 and C 3 N 5 are observed. In addition, S-Ni(OH) 2 and C 3 N 5 come into tight contact with each other and it is difficult to distinguish each material due to their similar micro morphology. The high resolution region (Fig. 2d) of S-Ni(OH) 2 -C 3 N 5 can be clearly divided into an amorphous region corresponding to C 3 N 5 and a crystalline region for S-Ni(OH) 2 with a clear lattice fringe of 0.23 nm corresponding to the (011) plane of S-Ni(OH) 2 . By the way, the two different regions in Fig. 2d come into tight contact with each other, which will be favorable for the internal transfer of photogenerated carriers. Fig. 2e depicts the SEM and element mapping images of 1.0 wt% S-Ni(OH) 2 -C 3 N 5 . As revealed, the hybrid is basically composed of nanosheets containing S-Ni(OH) 2 and C 3 N 5 , which is in good agreement with the above results of TEM. The element mapping images convey that the hybrid contains C, N, Ni, O and S, and all of the elements disperse uniformly in the observed region. Moreover, the existence of S-doping in S-Ni(OH) 2 is conrmed from the mapping of S in the material. Based on the abovementioned XRD, XPS, TEM and SEM analyses, the product of reaction between Ni 2+ and Na 2 S is named S-Ni(OH) 2 considering that no any information corresponding to nickel sulde is detected  while the Ni-S bond exists in the hybrid, and the hybrid of 2D S-Ni(OH) 2 -C 3 N 5 is successfully constructed.

DRS and e À /h + separation analyses
In the eld of photocatalysis, the light responsive region plays a signicant role in the performance of catalysts. The visible light absorption properties of S-Ni(OH) 2 , C 3 N 5 and S-Ni(OH) 2 -C 3 N 5 are compared in Fig. 3. In brief, C 3 N 5 possesses a narrow band-gap of $2.25 eV according to our previous work, 36 and can absorb the visible light region of 400-600 nm with high intensity, while pristine S-Ni(OH) 2 exhibits a strong absorption across the spectral range of 200-2400 nm (Fig. S6 †). For the hybrid material, 1.0 wt% S-Ni(OH) 2 -C 3 N 5 exhibits a similar visible light absorption region to C 3 N 5 , but the absorption intensity is slightly increased by the hybridization of black S-Ni(OH) 2 . In a word, this wide and strong visible light absorption property of the hybrid is favorable for achieving high photocatalytic performance.
Light absorption is the rst step for a photocatalytic procedure, and the subsequent photogenerated e À /h + separation is of great importance to the whole performance. Monitoring of the PL behavior of photocatalysts can nd out some information on e À /h + separation aer light excitation. 26,48,49 As revealed in the steady-state PL spectrum in Fig. 4a, C 3 N 5 exhibits strong PL intensity, demonstrating its high photogenerated e À /h + recombination rate aer light excitation. However, the emission intensity of C 3 N 5 can be efficiently quenched by S-Ni(OH) 2 . The lower PL intensity of 1.0 wt% S-Ni(OH) 2 -C 3 N 5 than C 3 N 5 demonstrates that the recombination of e À /h + pairs on C 3 N 5 aer excitation is restrained, which should be caused by the electron transfer from C 3 N 5 to S-Ni(OH) 2 since Ni-based species such as nickel sulde and Ni(OH) 2 are oen used as co-catalysts for photocatalytic H 2 production. 43,50-52 Mott-Schottky analysis was further conducted on S-Ni(OH) 2 to verify whether the CB position of S-Ni(OH) 2 is sufficient to capture the photogenerated electrons from C 3 N 5 (Fig. S7 †). The CB position of S-Ni(OH) 2 was determined to be À0.58 V vs. NHE, which is less negative than that of C 3 N 5 (À0.98 V vs. NHE). 36 So, under irradiation, the photogenerated electrons of C 3 N 5 could migrate to S-Ni(OH) 2 and be captured. The time-resolved uorescence spectra (TRFS) of C 3 N 5 and 1.0 wt% S-Ni(OH) 2 -C 3 N 5 are compared in Fig. 4b to further illustrate this viewpoint. As depicted, the PL intensity of C 3 N 5 decays quickly aer excitation, indicating its low uorescence lifetime which is not favorable for photogenerated e À /h + separation. However, the decay tendency of 1.0 wt% S-Ni(OH) 2 -C 3 N 5 exhibits a clear slowdown under the same conditions, demonstrating that the uorescence lifetime is prolonged, which should be related to the internal electron transfer from C 3 N 5 to S-Ni(OH) 2 . 18,53 For comparison, the uorescence lifetime of C 3 N 5 and 1.0 wt% S-Ni(OH) 2 -C 3 N 5 is calculated to be $8.50 and 14.02 ns respectively via a biexponential tting. The enhanced e À /h + separation efficiency of 1.0 wt% S-Ni(OH) 2 -C 3 N 5 compared to pristine C 3 N 5 is further conrmed by photocurrent-time and EIS behaviors. As revealed in Fig. 4c, sharp photocurrent is generated under light irradiation and then decreases suddenly when light is turned off, demonstrating that the current is lightswitched and produced by the photogenerated e À /h + separation aer the semiconductor is excited. 20,54,55 In comparison, 1.0 wt% S-Ni(OH) 2 -C 3 N 5 exhibits a much higher photocurrent value than pristine C 3 N 5 , indicating that the separation of photogenerated e À /h + on C 3 N 5 is facilitated by loading of the S-Ni(OH) 2 co-catalyst. In addition, a smaller arc radius is recorded on the S-Ni(OH) 2 -C 3 N 5 electrode, revealing the smaller resistance for charge transfer which is favorable for e À /h + separation. 53,56 All these results convey that the S-Ni(OH) 2 -C 3 N 5 hybrid possesses quick internal charge transfer efficiency, which will facilitate the subsequent photocatalytic performance.

Photocatalytic H 2 production performance
In consideration of the wide visible light absorption region and quick photogenerated e À /h + separation efficiency of the S-Ni(OH) 2 -C 3 N 5 hybrid, its photocatalytic performance for H 2 production from water with TEOA as a sacricial reagent under visible light irradiation (l > 420 nm) is evaluated rstly. A blank experiment shows that almost no H 2 production is observed on pristine C 3 N 5 due to the lack of a co-catalyst. And pristine S-Ni(OH) 2 also cannot produce H 2 under these conditions.  However, as displayed in Fig. 5a, the H 2 production rate is dramatically enhanced even with a low loading amount of 0.5 wt% S-Ni(OH) 2 on C 3 N 5 . Generally, a volcano-type relationship between the loading amount of a cocatalyst and the whole photocatalytic activity will be obtained, 57 suggesting that the fraction of S-Ni(OH) 2 in the hybrid must be screened to explore the highest H 2 production rate. As shown in Fig. 5b, the maximum H 2 production rate is 1450 mmol h À1 on the catalyst of 1.0 wt% S-Ni(OH) 2 -C 3 N 5 . Notably, excessive loading of S-Ni(OH) 2 will inuence the visible light absorption of C 3 N 5 , and then the whole photoactivity is decreased. In addition, the initial concentration of Na 2 S in the procedure of catalyst preparation will inuence the concentration of OH À and S 2À , which participate in the formation of S-Ni(OH) 2 . The inuence of initial mol ratio between Ni 2+ and Na 2 S (Ni/S) in the preparation procedure on the photoactivity is shown in Fig. S4, † and the selected condition for S-Ni(OH) 2 preparation is Ni/S ¼ 1/3.
To explore the inuence of S doping on the photocatalytic H 2 production performance, the comparison of 1.0 wt% S-Ni(OH) 2 -C 3 N 5 and 1.0 wt% Ni(OH) 2 -C 3 N 5 is shown in Fig. 5b. Unexpectedly, 1.0 wt% Ni(OH) 2 -C 3 N 5 exhibits much lower H 2 production activity (209 mmol h À1 ), which is only one seventh of the value over the 1.0 wt% S-Ni(OH) 2 -C 3 N 5 hybrid. That is to say, a novel method is explored to prepare a highly efficient Ni(OH) 2 -based co-catalyst by simply changing the initial reactant from NaOH to Na 2 S that reacts with Ni 2+ under ambient conditions ultra-fast. The photocatalytic performance is also evaluated using Pt nanoparticles as a reference since Pt is generally considered as the most efficient co-catalyst for photocatalytic H 2 production. 15,23 As depicted in Fig. 5b, the H 2 production rate on 1.0 wt% S-Ni(OH) 2 -C 3 N 5 is slightly higher than that of 1.0 wt% Pt-C 3 N 5 , demonstrating that the present S-Ni(OH) 2 is a highly efficient co-catalyst for H 2 production and is promising in consideration of the facile preparation procedure and low cost of Ni-based materials compared with noble metals like Pt.
The AQY values of 1.0 wt% S-Ni(OH) 2 -C 3 N 5 under a series of monochromatic light irradiation are shown in Fig. 6a, in which the AQY values decrease gradually with the increase of light wavelength and the highest AQY value (30.9%) is obtained at 420 nm. The change tendency of AQY is in agreement with the DRS spectrum of S-Ni(OH) 2 -C 3 N 5 , demonstrating that the photocatalytic performance of S-Ni(OH) 2 -C 3 N 5 is determined by its light absorption behavior. Long-term stability of a photocatalyst is an important norm in the eld of H 2 production. Fig. 6b reveals the cycling performance of 1.0 wt% S-Ni(OH) 2 -C 3 N 5 and no obvious attenuation of H 2 production rate happens aer 16 h and 4 runs of photocatalytic test, demonstrating the excellent photocatalytic stability of the present S-Ni(OH) 2 -C 3 N 5 hybrid. To further evaluate the stability of S-Ni(OH) 2 -C 3 N 5 in photocatalysis, the samples are collected aer long-term light irradiation and monitored by XRD and DRS. As depicted in Fig. 6c and d, no obvious change happens in the XRD peaks of S-Ni(OH) 2 -C 3 N 5 experiencing the photocatalytic test, indicating that the basic structure of the hybrid is maintained. Meanwhile, the collected sample exhibits a similar visible light absorption region compared with fresh S-Ni(OH) 2 -C 3 N 5 , demonstrating the stability of the hybrid again.

Photocatalytic NO oxidation performance
Except for H 2 production, the photocatalytic performance of S-Ni(OH) 2 -C 3 N 5 is also tested for NO oxidation to demonstrate the potential of the catalyst in purication of atmospheric pollution. A control experiment shows that pristine S-Ni(OH) 2 has no activity for NO removal. As revealed in Fig. 7a, the oxidation of NO can reach equilibrium in 25 min in the present continuous ow reaction system. Briey, NO removal efficiencies on C 3 N 5 and S-Ni(OH) 2    for NO removal, and only a slight decrease in NO removal ratio is observed during 120 nm light irradiation, demonstrating the good stability of the hybrid. Generally, the main products of photocatalytic NO oxidation are NO 2 and NO 3 À according to literature reports. [58][59][60] The detection of NO 2 is carried out on the present NO x analyzer, and the concentration of NO 2 along with NO oxidation is shown in Fig. S5. † It is found that most NO is converted to NO 3 À based on the amount of NO removal and NO 2 production. Therefore, the conrmation of produced NO 3 À in the reaction is shown in Fig. 7c with NaNO 3 as a reference by ion chromatography, in which a clear signal corresponding to NO 3 À is observed for the sample prepared with 1.0 wt% S-Ni(OH) 2 -C 3 N 5 experiencing photocatalytic NO oxidation reaction. It has been reported that activation of molecular O 2 and generation of a series of ROS (i.e. cOH, cO 2 À and 1 O 2 ) play the main role in the procedure of photocatalytic NO oxidation. [61][62][63] In addition, photogenerated h + residual in the VB of the semiconductor is also of great importance. 64,65 Thus, the photocatalytic NO removal performance of 1.0 wt% S-Ni(OH) 2 -C 3 N 5 in the presence of a series of trapping agents for different active species is compared in Fig. 7d. As displayed, when Na 2 C 2 O 4 , tert-butanol (TBA), p-benzoquinone (PBQ) and b-carotene are used as scavengers for h + , cOH, cO 2 À and 1 O 2 respectively, the photocatalytic performance of 1.0 wt% S-Ni(OH) 2 -C 3 N 5 is seriously restrained, demonstrating that NO is removed by these ROS and residual h + in the VB of C 3 N 5 . The NO 2 concentration during the photocatalytic NO removal on 1.0 wt% S-Ni(OH) 2 -C 3 N 5 was also monitored with the NO x analyzer (see Fig. S8 †). The corresponding residual ratio of NO 2 is provided in Table S2. † From these results, it can be seen that the cOH and 1 O 2 play important roles in the conversion of NO 2 to NO 3 À . Electron spin-resonance spectroscopy (ESR) tests are conducted to monitor the change of signals corresponding to cOH, cO 2 À and 1 O 2 . As revealed in Fig. 8, almost no signals are detected in the dark on both 1.0 wt% S-Ni(OH) 2 -C 3 N 5 and C 3 N 5 , but strong peaks corresponding to distinct DMPO-cOH, DMPO-cO 2 À and TEMPO (product of TEMP oxidized by 1 O 2 ) signals are detected, 3,7,61 demonstrating that the generation of these ROS is light controllable. Moreover, the light induced ESR signal intensity on 1.0 wt% S-Ni(OH) 2 -C 3 N 5 is higher than that of C 3 N 5 , revealing that more amount of ROS is generated on S-Ni(OH) 2 -C 3 N 5 due to the quick internal charge transfer. So, the process of photocatalytic NO removal on 1.0 wt% S-Ni(OH) 2 -C 3 N 5 should be as follows: S-Ni(OH) 2 -C 3 N 5 + hv / e À + h + (2) h + + cO 2 À / 1 O 2 (4) Based on all of the above results, the mechanism for photocatalytic H 2 production or NO oxidation over the S-Ni(OH) 2 -C 3 N 5 hybrid is proposed. Under visible light irradiation, C 3 N 5 is excited and generates e À /h + pairs. S-Ni(OH) 2 can promote the internal charge separation and harvest the e À on the CB of C 3 N 5 , and then the trapped electrons participate in subsequent surface reactions. For photocatalytic H 2 production (Fig. 9a), the collected electrons on S-Ni(OH) 2 react with adsorbed H 2 O molecules and H 2 is produced. The residual h + on the VB of C 3 N 5 is consumed by TEOA, and then the whole photocatalytic H 2 production procedure is accomplished. However, the situation for photocatalytic NO oxidation is different (Fig. 9b). O 2 is rstly activated and a series of ROS (cO 2 À , 1 O 2 and cOH) can be generated. Generally, cO 2 À is produced by reaction between e À  and adsorbed O 2 , while the reaction between cO 2 À and h + can generate 1 O 2 . 3,61 cOH should originate from the oxidization of H 2 O by cO 2 À since the VB position of C 3 N 5 cannot satisfy the requirement of cOH generation based on our previous work. 33,66 In addition, the generation of these ROS is enhanced by S-Ni(OH) 2 , as discussed in the above section. And then, these ROS and the residual h + on the VB of C 3 N 5 participate in the NO oxidation procedure along with the formation of NO 3 À and NO 2 .

Conclusions
In summary, near-zero-cost 2D S-Ni(OH) 2 active sites are planted on wide visible light responsive C 3 N 5 ultrafast within 30 min at room temperature by reaction between Ni(NO 3 ) 2 and Na 2 S in aqueous solution. The loading of S-Ni(OH) 2 can greatly enhance the photogenerated e À /h + separation efficiency of C 3 N 5 aer light excitation. Due to the quick internal charge transfer, the S-Ni(OH) 2 -C 3 N 5 hybrid is highly efficient as a multifunctional catalyst in various photocatalytic applications: H 2 production from water and NO removal. Most impressively, the H 2 production activity on S-Ni(OH) 2 -C 3 N 5 is even higher than that of Pt-C 3 N 5 and an AQY value of 30.9% at 420 nm is achieved. This work brings new insights into the design of low-cost noble-metal-free co-catalysts on semiconductors for photocatalytic applications.

Conflicts of interest
There are no conicts to declare.