Efficient Electronic Modulation of g-C 3 N 4 Photocatalyst by Implanting Atomically Dispersed Ag 1 -N 3 for Extremely High Hydrogen Evolution Rates

Open AccessCCS ChemistryRESEARCH ARTICLE5 Aug 2022Efficient Electronic Modulation of g-C3N4 Photocatalyst by Implanting Atomically Dispersed Ag1-N3 for Extremely High Hydrogen Evolution Rates Guanchao Wang†, Ting Zhang†, Weiwei Yu†, Zhe Sun†, Xiaowa Nie, Rui Si, Yuefeng Liu and Zhongkui Zhao Guanchao Wang† State Key Laboratory of Fine Chemicals, Department of Catalysis Chemistry and Engineering, School of Chemical Engineering, Dalian University of Technology, Dalian 116024 †G. Wang, T. Zhang, W. Yu, and Z. Sun contributed equally to this workGoogle Scholar More articles by this author , Ting Zhang† State Key Laboratory of Fine Chemicals, Department of Catalysis Chemistry and Engineering, School of Chemical Engineering, Dalian University of Technology, Dalian 116024 †G. Wang, T. Zhang, W. Yu, and Z. Sun contributed equally to this workGoogle Scholar More articles by this author , Weiwei Yu† State Key Laboratory of Fine Chemicals, Department of Catalysis Chemistry and Engineering, School of Chemical Engineering, Dalian University of Technology, Dalian 116024 †G. Wang, T. Zhang, W. Yu, and Z. Sun contributed equally to this workGoogle Scholar More articles by this author , Zhe Sun† State Key Laboratory of Fine Chemicals, Department of Catalysis Chemistry and Engineering, School of Chemical Engineering, Dalian University of Technology, Dalian 116024 †G. Wang, T. Zhang, W. Yu, and Z. Sun contributed equally to this workGoogle Scholar More articles by this author , Xiaowa Nie State Key Laboratory of Fine Chemicals, Department of Catalysis Chemistry and Engineering, School of Chemical Engineering, Dalian University of Technology, Dalian 116024 Google Scholar More articles by this author , Rui Si *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] E-mail Address: [email protected] Shanghai Synchrotron Radiation Facility, Shanghai Institute of Applied Physics, Chinese Academy of Sciences, Shanghai 201204 Google Scholar More articles by this author , Yuefeng Liu *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] E-mail Address: [email protected] Dalian National Laboratory for Clean Energy (DNL), Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Dalian 116023 Google Scholar More articles by this author and Zhongkui Zhao *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] E-mail Address: [email protected] State Key Laboratory of Fine Chemicals, Department of Catalysis Chemistry and Engineering, School of Chemical Engineering, Dalian University of Technology, Dalian 116024 Google Scholar More articles by this author https://doi.org/10.31635/ccschem.021.202101191 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesDownload CitationsTrack Citations ShareFacebookTwitterLinked InEmail Developing an efficient method to improve the photocatalytic efficiency of graphitic carbon nitride (g-C3N4) is of great significance for solar H2 production. Electronic structure modulation has been considered one of the most crucial strategies to improving the photocatalytic efficiency of g-C3N4, but how to efficiently modulate its electronic structure remains a huge challenge. Herein, we, for the first time, report a facile and highly-efficient approach to modulating the electronic structure of g-C3N4 through single Ag atom implantation with a Ag1-N3 coordination configuration into the g-C3N4 framework. Due to the remarkably promoted light absorption and notably improved charge separation resulting from efficient electronic structure modulation, the Ag1-N3 sites embedded hollow g-C3N4 sphere (Ag1N3-HCNS) shows an unprecedentedly high visible-light photocatalytic H2 evolution rate (HER) of 17.95 mmol g−1 h−1 under an atmospheric pressure with a remarkable apparent quantum yield (AQY) of 23.6% at 420 nm. Owing to different test apparatuses and conditions in different literature, neither the absolute HER value nor AQY could be used as a comparative indicator. Generally, times (tsHER) regarding the improvement in HER compared to bulk g-C3N4 under the same test apparatus and conditions are presented in the literature. Therefore, the tsHER can be used as an indicator for comparisons of the photocatalytic performance of the developed catalyst. Ag1N3-HCNS shows an unprecedented 193-fold higher HER than bulk g-C3N4 under the same measurement conditions, remarkably outperforming the previously reported g-C3N4 photocatalysts. This work presents a new horizon for designing excellent g-C3N4 photocatalysts through efficient electronic structure modulation of tri-s-triazine by implanting single-atom metals with strong metal-N bonding. Download figure Download PowerPoint Introduction Hydrogen is a promising second energy source and considered an important key building-block material in many modern chemical processes, such as Fischer–Tropsch synthesis and ammonia reactions.1 Solar photocatalytic water splitting into hydrogen fuels has been considered a promising strategy to alleviate the growing energy crisis and environmental challenges caused by fossil fuel burning.1–4 To realize the practical applications concerning green hydrogen production via photocatalytic water splitting, intense efforts have been devoted to designing more active photocatalysts and systems.5–7 Recently, a variety of semiconductors have been reported for water splitting.8–10 Among them, graphitic carbon nitride (g-C3N4) has attracted more attention due to its low cost precursor, suitable band gap, unique N/C coordinating network consisting of a tri-s-triazine structure, high physical and chemical stability, and good optical electronic structure.11–15 However, bulk C3N4 (BCN) still suffers from inferior photocatalytic efficiency owing to its low charge separation and transfer, insufficient light absorption, and low specific surface area, limiting its further development.16,17 Aimed at compensating for the intrinsic drawbacks of g-C3N4 to enhance its photocatalysis, many strategies like nanostructure design, electronic structure modulation, crystal-structure engineering, and heterostructure construction have been developed,16,18–21 among which electronic structure modulation can be considered as one of the most crucial and effective approaches to improving the photocatalytic efficiency of g-C3N4.20,22–24 Although many efforts to modulate the electronic structure of g-C3N4 to improve its photocatalysis have been made,20,22–29 the development of a facile and efficient electronic regulation method remains a sizable challenge. From references,20,30,31 compared to non-metallic element doping, metal element-doping has emerged as a promising approach for the electronic modulation of g-C3N4. Unfortunately, the incorporation of metal nanoparticles (NPs) has not worked well thus far. It is highly desirable to develop a facile and efficient method for electronic structure regulation via metal element-doping of g-C3N4. Recently, single-atom metals (SAMs), as catalytically active sites to lower the H2 evolution energy barrier, have attracted great interest in g-C3N4 photocatalysis.32–37 Although the H2 evolution rate (HER) per gram metal is efficiently improved owing to atomic dispersion, the HER per gram g-C3N4 catalyst remains quite low. To achieve ultrahigh loading of SAM remains a challenge. Now that the doping of metal NPs can effectively adjust the electronic structure of g-C3N4, the implantation of SAM could be proposed as a sophisticated strategy for electronic modulation to improve the HER per gram g-C3N4 catalyst. However, nowadays, rare reports on the SAM-doping of g-C3N4 regarding the modulation of electronic structures can be found. From references,38–41 the implantation of SAM via strong covalent metal-N bonding is more efficient than that by a weak electrostatic interaction for modulating the electronic structure of g-C3N4 photocatalysts. Fu et al.38 presented a single-atom Fe-doped [email protected]3N4 joint electronic system, exhibiting 18.3 times higher HER of BCN. We previously reported a atomically dispersed Cu1-N3 embedded g-C3N4, generating an outstanding photocatalyst for hydrogen evolution with 35.4 times higher HER than BCN.39 Subsequently, Xiao et al.40 reported Cu single atom-incorporated g-C3N4 tubes, showing a 30 times higher HER than BCN. However, regarding the future applications of g-C3N4 in solar H2 production, further improvements in photocatalytic efficiency still remain a challenge. Like Cu, Ag belongs to the IB group, which has a similar configuration of extra-nuclear electrons (4d105s1) to Cu (3d104s1). However, the Ag atom has a distinct atomic radius and proton number from Cu, where a new horizon may emerge by embedding single Ag atoms into g-C3N4 framework via strong Ag-N bonding for the modulation of its electronic structure. To the best of our knowledge, no reports regarding this issue can be found. Herein, we, for the first time, have prepared a single-atom Ag implanted hollow carbon nitride sphere (Ag1N3-HCNS) with Ag1-N3 coordination through a supramolecular assembly of melamine and melamine-Ag complex (Mel-Ag) with cyanuric acid followed by a pyrolysis process in N2 atmosphere. The atomically dispersed Ag1-N3 moieties embedded within nanosheets of g-C3N4 via Ag-N bonding were unambiguously confirmed by high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM), X-ray absorption near-edge structure (XANES), and X-ray photoelectron spectroscopy (XPS) with/without Ar plasma etching. The combination of experimental results and density functional theory (DFT) calculations clearly demonstrate that the implantation of single-atom Ag within the g-C3N4 framework remarkably promotes visible light absorption and notably improves the separation efficiency of charge carriers, resulting from the highly efficient modulation of electronic structure of g-C3N4 via strong covalent Ag-N bonding. Asa result, Ag1N3-HCNS shows an unprecedentedly high visible-light photocatalytic HER of 17.95 mmol g−1 h−1 under an atmospheric pressure, and a remarkable apparent quantum yield (AQY) of 23.6% at 420 nm. The achieved HER over Ag1N3-HCNS is 193 times higher than that on BCN, notably outperforming the previously reported g-C3N4 photocatalysts so far in the literature. This work opens a new window for improving the photocatalytic efficiency, a bottleneck issue, of g-C3N4 photocatalysts via efficient electronic structure modulation of tri-s-triazine frameworks. Experimental Methods Synthesis of pristine HCNS According to references,39,42 0.5 g melamine was dissolved in dimethyl sulfoxide (DMSO) to obtain solution A, and 0.51 g cyanuric acid was dissolved in DMSO to obtain solution B. Then, the resulting solution B was added into the above solution A under vigorous stirring for 10 min to form the precursor of HCNS. After filtration and washing, the resulting precipitate was dried at 60 °C for 12 h and then calcined at 550 °C for 4 h in a nitrogen atmosphere to obtain the final HCNS. Synthesis of HCNS embedded with single Ag atoms in nanosheets (Ag1N3-HCNS) In detail, the complex melamine-Ag (denoted as Mel-Ag) was presynthesized. After that, 4.3 mg Mel-Ag complex and 0.5 g melamine were dissolved together to obtain solution C, and cyanuric acid was dissolved in DMSO to obtain solution B. Then, solution B was added to solution C under stirring to form the precursor of Ag1N3-HCNS. After filtering and washing, the resulting precipitate was then dried at 60 °C for 12 h and calcined at 550 °C for 4 h under nitrogen atmosphere to obtain the final Ag1N3-HCNS catalyst. Synthesis of HCNS embedded with single Cu atoms in nanosheets (Cu1N3-HCNS) According to references,39,42 45.6 mg Cu(NO3)2·3H2O and 0.5 g melamine were dissolved in DMSO to obtain solution D, and 0.51 g cyanuric acid was dissolved in DMSO to obtain solution B. Then, solution B was added into the solution D under vigorous stirring for 10 min to form the precursor of Cu1N3-HCNS. After filtration and washing, the precipitate was dried at 60 °C for 12 h and then calcined at 550 °C for 4 h under nitrogen to obtain the final Cu1N3-HCNS sample. Synthesis of HCNS embedded with Ag nanoparticles in nanosheets (AgNP-HCNS) 0.5 g melamine and NaBH4 were dissolved in 20 mL DMSO to obtain solution E. 15.8 mg AgNO3 and 0.51 g cyanuric acid were dissolved in 10 mL DMSO to obtain solution F. Then, the obtained solution F was added into solution E under magnetic stirring. The precipitate was kept with magnetic stirring for 10 min. After filtration and washing, the precipitate was dried at 60 °C for 12 h and then calcined at 550 °C for 4 h under nitrogen atmosphere to obtain AgNP-HCNS. Synthesis of BCN BCN was prepared on the basis of reference,39 where 1 g melamine was calcined at 550 °C for 4 h under nitrogen to obtain BCN. Characterizations of materials The morphologies of the samples were characterized using a JEOL JSM-5600LV (JEOL, Akishima Shi, Japan) emission scanning electron microscopy (SEM). HAADF-STEM and transmission electron microscopy (TEM) characterization were conducted on a JEOL-2100F (JEOL, Akishima Shi, Japan) field-emission transmission electron microscope (FETEM) with an acceleration voltage of 80 kV. XPS was acquired using an ESCALAB 250 XPS (Thermo Scientific, Waltham, MA, USA) system under the Al Kα X-ray source. Elemental analysis of N and C was collected using an elemental analyzer (Vario EL, Elementar, Heraeus, German). Powder X-ray diffraction (XRD) patterns were performed on an X-ray diffractometer (Rigaku Corporation SmartLab 9, Rigaku, Akishima-shi, Tokyo, Japan) using Cu Kα radiation, operating at 40 kV and 40 mA, ranging from 5° to 80° with a scanning speed of 8°/min. N2 adsorption–desorption measurements were performed on a Beishide instrument (Beishide Corp., Beijing, China). The surface area was calculated using the model of 3H-2000PSI system (Beishide Corp., Beijing, China), and sample pore sizes were obtained from the adsorption data using the t-Plot method. UV–vis diffuse reflectance spectra (DRS) were recorded on a JASCO V-550 UV–vis spectrometer (JASCO, Hachioji-shi, Tokyo, Japan). Photoluminescence (PL) spectra were recorded on a Hitachi F-7000 fluorescence spectrometer (Hitachi, chiyoda-ku, Tokyo, Japan; exciting samples by 376 nm photons). Time-resolved PL measurement was performed on FLS-920 transient steady-state fluorescence spectrometer (Edinburgh Instruments, Livingston, Edinburgh, UK). The amounts of Ag or Cu in the samples were analyzed by an inductively coupled plasma atomic emission spectrometer (ICP-AES) on an Optima 7300 DV (PerkinElmer, Waltham, MA, USA). X-ray absorption fine structure (XAFS) spectra at the Cu K (E0 = 8979 eV) edge and Ag K edge (E0 = 25514 eV) were measured at the BL14W1 beam line of Shanghai Synchrotron Radiation Facility (SSRF). The storage ring of SSRF was working at an energy of 3.5 GeV under “top-up” mode with a constant current of 260 mA. The XAFS data were monochromatized with a Si (111) monochromator and Lytle-type ion chamber. The energy was calibrated according to the absorption edge of pure copper foil. The extended XAFS (EXAFS) data were processed using the Athena and Artemis codes. For the part of XANES spectrum, the change in experimental absorption coefficient with energy μ (E) was processed through background subtraction and normalization procedures and reported as “normalized absorption.” For the normalized XANES profiles, the oxidation states of copper and silver were determined by a linear combination fit with a large number of references (Cu2O for Cu+, CuO for Cu2+, and AgNO3 for Ag+). For the EXAFS portion, the Fourier transformed (FT) data in R space were analyzed by applying a C3N4-like model by replacing the Cu/Ag center for Cu-N/Cu-C or Ag-N/Ag-C and first-shell approximate model for Cu-O/Ag-O contributions, respectively. The parameters describing the electronic properties (e.g., correction to the photoelectron energy origin, E0) and local structure environment including coordination number (CN), bond distance (R), and Debye–Waller factor around the absorbing atoms were allowed to vary during the fit process. The fitted ranges for k and R spaces were selected to be k = 3–11 Å−1 with R = 0.8–3.0 Å (k3 weighted). DFT calculations DFT calculations were performed using the Vienna Ab initio Simulation Package (VASP) with ion cores represented by the projector augmented wave (PAW) potentials implemented by VASP. The exchange-correlation interactions were described with the generalized-gradient approximation (GGA) approach and the spin polarized Perdew–Burke–Ernzerhof (PBE) functional. The plane wave basis was set with a cut off energy of 400 eV for all calculations. Geometries of the catalyst models were fully relaxed using a damped molecular dynamics method until the forces on all atoms were less than 0.03 eV Å−1. The lattice constants of the Ag1N3-HCNS and Cu1N3-HCNS models were calculated to be a = b = 14.32 Å and c = 20.00 Å. The k-space integration was sampled using a 3*3*1 Monkhorst–Pack grid. A 20 Å vacuum layer in the z direction was built to avoid interactions between repeating slabs. For the orbital distributions of highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO), the value of iso-surface was set to 0.015 e Å−3. Photoelectrochemical measurements Photoelectrochemical measurements were performed on a CHI660E electrochemical workstation using a conventional three-electrode cell. The working electrodes were each fabricated by coating an indium-doped tin oxide (ITO) glass with an as-prepared catalyst. Experimental conditions for the photocurrent test: 0.5 M Na2SO4 aqueous solution was used as the electrolyte (pH 7.1, 25 °C); Ag/AgCl electrode was used as the reference electrode, and Pt plate was used as the counter-electrode. Experimental conditions for the Mott–Schottky plots were the same as the aforementioned process. The Ag/AgCl electrode system can be converted to a reversible hydrogen electrode (RHE) system by the following equation: V RHE = V Ag/AgCl + V Ag/AgClVSNHE 0 + 0.059 pH (1)where V Ag / AgClVSNHE 0 is 0.1976 V at 25 °C. Photocatalytic performance tests All photocatalytic hydrogen evolution experiments were carried out in a top-illuminated vessel in N2 atmosphere (at atmosphere pressure condition) by visible light irradiation. In detail, 10 mg photocatalyst was dispersed in 100 mL triethanolamine (TEOA) aqueous solution (10 mL TEOA/90 mL water). Pt (3 wt %) was loaded on the photocatalysts as a cocatalyst by a general photodeposition method. The reaction solution was evacuated several times to completely remove air, and then was re-filled with N2 to maintain an atmospheric pressure condition. The reactor was irradiated by a 300 W Xe lamp equipped with a 420 nm cut-off filter (PLSSXE300/300UV, Perfectlight, Beijing, China). The temperature of the reactant solution was maintained at 15 °C of constant temperature by a flow of cooling water during the whole reaction test process. The resulting hydrogen product was analyzed by a GC-9790 gas chromatograph with a thermal conductive detector. Wavelength-dependent AQY of H2 evolution was obtained under different wavelengths of light (400, 420, 450, and 520 nm). The AQY were calculated by the following equation: AQY ( % ) = 2 × Amount of H 2 molecules evolved Number of incident photons × 100 (2) Results and Discussion Structural characterization of as-prepared materials In this work, a single-atom Ag implanted HCNS (Ag1N3-HCNS) was prepared by the supramolecular assembly process of melamine and Mel-Ag with cyanuric acid followed by a pyrolysis process in N2 (Figure 1a). According to our previous reports,39,42 the single-atom Cu1-N3 species embedded HCNS (Cu1N3-HCNS), pristine HCNS, and AgNP-HCNS were prepared for comparison. The intercalation-structured HCNSs composed of carbon nitride nanosheets were observed on SEM (Figures 1b–1d and Supporting Information Figure S1a) and TEM (Figures 1e–1g) images. The corresponding elemental mappings verify that the Ag and Cu atoms were homogeneously distributed in the entire carbon nitride nanosheets ( Supporting Information Figures S2 and S3), and the Ag and Cu contents of Ag1N3-HCNS and Cu1N3-HCNS were 3.78 and 0.85 wt %, respectively, as determined by ICP-AES. The high-resolution TEM (HRTEM) images ( Supporting Information Figures S1b and S1c) show that no Ag or Cu NPs were present on the Ag1N3-HCNS and Cu1N3-HCNS, while Ag NPs can be obviously observed on AgNP-HCNS ( Supporting Information Figure S1d). The atomically dispersed Ag (Cu) atoms in Ag1N3-HCNS (Cu1N3-HCNS) are identified by HAADF-STEM images as bright spots (Figures 1h and 1i, highlighted by yellow circles). Supporting Information Figure S4 and Table S1 implies no obvious influence on the textural properties and N/C ratio after the introduction of a low amount of Ag or Cu atoms within the framework. Similar to the pristine HCNS, Ag1N3-HCNS and Cu1N3-HCNS show two weak XRD peaks at 13.1° and 27.4° (Figure 1j ), indexed as g-C3N4,14 suggesting that the g-C3N4 matrix is not obviously affected by the addition of SAM. Figure 1 | (a) Schematic illustration of preparation of Ag1N3/Cu1N3-HCNS. (b–d) SEM images of Ag1N3-HCNS (b), Cu1N3-HCNS (c), and pristine HCNS (d). (e–g) TEM images of Ag1N3-HCNS (e), Cu1N3-HCNS (f), and HCNS (g). (h and i) HAADF-STEM images of Ag1N3-HCNS (h) and Cu1N3-HCNS (i). (j) XRD patterns of Ag1N3-HCNS, Cu1N3-HCNS, HCNS, and AgNP-HCNS. Download figure Download PowerPoint Chemical state and coordination characterization XPS analysis provides more structural details of the samples (Figures 2a–2c and Supporting Information Figure S5). Compared to pristine HCNS, both Ag1N3-HCNS and Cu1N3-HCNS feature a shift of higher binding energy of pyridinic N on N 1s XPS spectra ( Supporting Information Figure S5), implying the possible coordination of pyridinic N with the embedded metal via metal-Nx bonding.39,42–44 A new peak deconvoluted on N1s XPS spectra of Ag1N3-HCNS and Cu1N3-HCNS (Figures 2a–2c) can be assigned to Ag–N bonding and Cu–N bonding, respectively.39,40 The lowering content of pyridinic N ( Supporting Information Table S2) is a further indicator of coordination of Ag or Cu with pyridinic N.39 Furthermore, the remarkably increased intensity of Ag 3d and Cu 2p XPS peaks after Ar plasma etching ( Supporting Information Figure S6) indicates the implantation of Ag and Cu atoms within the carbon nitride nanosheets of Ag1N3-HCNS and Cu1N3-HCNS, respectively.39 Figure 2 | (a–c) N 1s XPS spectra of Ag1N3-HCNS (a), Cu1N3-HCNS (b), and HCNS (c). (d and e) Normalized Ag K-edge XANES spectra of Ag1N3-HCNS (d) and Cu K-edge XANES spectra of Cu1N3-HCNS (e). (f and g) k3-weighted Fourier transform spectra from Ag K-edge EXAFS of Ag1N3-HCNS (f) and Cu K-edge EXAFS of Cu1N3-HCNS (g). (h and i) EXAFS fitting curve of Ag1N3-HCNS (h) and Cu1N3-HCNS (i) in R space. Inset in Figure 2d: magnified pre-edge of XANES curves. Inset in Figure 2h: structure model of Ag1N3-HCNS. Inset in Figure 2h: structure model of Cu1N3-HCNS. Download figure Download PowerPoint The electronic structure and local coordination environment of Ag and Cu species are further investigated by X-ray absorption spectroscopy (XAS). The pre-edge of XANES curve for Ag1N3-HCNS is different from that of AgNO3 and Ag foil (Figure 2d), indicating that the valence state for the single Ag atom may be changed because of the electron transfer caused by interactions of single Ag atoms with g-C3N4 via Ag-N bonding. The inset in Figure 2d shows that the re-edge of XANES curve for Ag1N3-HCNS is quite close to Ag foil, implying the oxidized state of Ag is near 0. The valence state of the single Ag atom was further investigated by XPS analysis. As shown in Supporting Information Figure S7, the two peaks at approximately 367.8 and 373.8 eV can be attributed to Ag 3d5/2 and Ag 3d3/2, respectively. These two peaks can be deconvoluted into two peaks at about 367.6/368.4 and 373.6/374.2 eV, respectively. The peaks at 367.6 and 373.6 eV are attributed to the Ag0 (64%), and those at 368.4 and 374.2 can be assigned to the Ag+ (36%). The results indicate that the single-atom Ag in the Ag1N3-HCNS mainly exists in the state of zero valence, which is consistent with the results from XAS. From the Cu K-edge spectra, the absorption threshold of Cu1N3-HCNS between CuO and Cu2O (Figure 2e) indicates the oxidation state of Cu (Cuδ+, 1 < δ < 2) in Cu1N3-HCNS.39,42 k3-weighted Fourier transform spectra from Ag K-edge EXAFS of Ag1N3-HCNS (Figure 2f) and Cu K-edge EXAFS of Cu1N3-HCNS (Figure 2g) exhibit no obvious Ag–Ag and Cu–Cu coordination bond existing on the two samples, suggesting that Ag and Cu atoms are atomically dispersed in the HCNS matrix. Furthermore, the EXAFS fitting results and DFT calculations give the coordination model structures of Ag-N and Cu-N over HCNS matrix with three N atoms anchoring one M (Ag or Cu) atom (Figures 2h and 2i and Supporting Information Table S3). Photocatalytic H2 evolution The test of photocatalytic H2 evolution reaction performance was performed to check the modulating effect of implantation of atomically dispersed Ag1-N3 moieties within the g-C3N4 matrix by using TEOA as a hole sacrificial reagent and 3 wt % Pt as co-catalyst under visible light irradiation and atmospheric pressure conditions. From Figure 3a, the developed Ag1N3-HCNS shows an unprecedented high HER of 17.95 mmol g−1 h−1, which is 193 times higher than that of BCN (0.093 mmol g−1 h−1) under the same test conditions. However, AgNP-HCNS only displays trace H2 evolution (0.015 mmol g−1 h−1), which is even lower than that of BCN. To make clear the reason for the poor photocatalytic performance of AgNP-HCNS, electron paramagnetic resonance (EPR), electrochemical impedance spectra (EIS), and transient photocurrent response measurements were performed. From EPR spectrum (Figure 4e), different from the HCNS and the single-atom Ag or Cu-modified HCNS with a strong EPR signal centered at a g value of 2.0082, AgNP-HCNS shows no EPR signal on the EPR spectrum, suggesting that the π-conjugated electronic system of g-C3N4 is destroyed during the pyrolysis process of the Ag NPs-containing melamine-cyanuric assembly at 550 °C for 4 h, although no obvious change happens on the XRD pattern. As a consequence, AgNP-HCNS shows a quite higher charge transfer resistance (Figure 4d) and a much worse charge photocurrent (Figure 4e) than the other g-C3N4 photocatalysts. The poor separation and transfer of photogenerated charge carriers of AgNP-HCNS leads to its poor photocatalytic activity for hydrogen evolution. Moreover, Ag1N3-HCNS and Cu1N3-HCNS show 14.6 and 2.7 times higher HER than pristine HCNS. It indicates the notable promoting effect of implantation of Ag and Cu of IB group on g-C3N4 regarding photocatalytic H2 evolution, and Ag is far superior to Cu for the photocatalysis promotion of g-C3N4. When the silver content is lowered from 3.78 to 1.0 wt %, the HER still remains as high as 15.9 mmol g−1 h−1 ( Supporting Information Figure S8). To the best of our knowledge, such excellent photocatalytic hydrogen evolution performance is far superior to the most reported photocatalysts under atmospheric pressure conditions ( Supporting Information Table S4). To compare the photocatalytic performance of the developed catalyst with that reported in literature, the BCN (generally reaching ca. 0.1 mmol g−1 h−1 HER rate) can be used as a benchmark catalyst, and a promoting factor (PF) can be defined. The ratio of the HER rate on the designed g-C3N4 to that on the BCN under the same test conditions presents an indicator for evaluating the advancement of the developed new catalysts, which can be defined as PF. From Supporting Information Table S4, the developed Ag1N3-HCNS catalyst shows a PF of 193 (193 times higher HER rate than BCN under the same reaction conditions), remarkably outperforming the reported g-C3N4 based catalysts in the literature. To make clear whether the implanted single-atom acts as an active site or not for photocatalytic hydrogen production, besides serving as a metallic dopant to modulate electronic structure of g-C3N4, the photocatalytic performance test of the as-prepared g-C3N4-based photocatalysts was performed without Pt deposition. Supporting Information Figure S9 shows that the Ag1N3-HCNS and Cu1N3-HCNS only exhibit a low HER of 0.021 and 0.015 mmol g−1 h−1, respectively, although both of them are superior to AgNP-HCNS, HCNS, and BCN. It presents evidence that the embedded single-atom Ag or Cu do not act as acti