摘要：利用取之不盡的太陽能資源進行光催化水裂解制氫是緩解全球能源危機、實現(xiàn)碳中和戰(zhàn)略的一項有前景的技術。石墨相氮化碳（g-C3N4）因成本低且穩(wěn)定性高在光催化產氫領域備受關注。然而，純g-C3N4存在表面積小、電子轉移慢、光生載流子復合快等缺陷，產氫性能通常不佳。本研究通過直接熱解硫酸銨和三聚氰胺混合物，成功實現(xiàn)硫物種對g-C3N4氮位點的原位取代，開發(fā)出一種高效的硫摻雜g-C3N4 （S-g-CN）光催化劑。系列結構和光譜表征證實硫的成功摻雜。密度泛函理論的第一性原理計算表明S活性位對氫的吸附吉布斯自由能近乎為零（～0.26 eV），揭示S摻雜在優(yōu)化H活性中間體吸附和解吸過程中起著重要作用。透射電子顯微鏡和原子力顯微鏡測試結果表明，S-g-CN具有超薄的納米片狀結構，其片層厚度約為2.5 nm。隨后的氮氣吸脫附等溫線和光電化學性質測試結果表明，S摻雜不僅可顯著增大g-C3N4比表面積，而且還能有效提高其光生電子-空穴對的轉移、分離和氧化還原能力。得益于材料良好的結構特性，S-g-CN的光催化產氫速率高達4923 μmol?g?1?h?1，是原始g-C3N4的28倍，超越諸多最近報道的其它S摻雜g-C3N4光催化劑。而且，S-g-CN的表觀量子效率高達3.64%。本研究除了開發(fā)一種高效的光催化劑，還將為高性能g-C3N4基催化劑的設計提供有益借鑒。

關鍵詞：理論預測；硫摻雜；g-C3N4；產氫；光催化

中圖分類號：O649；O644

S-Doping of the N-Sites of g-C3N4 to Enhance Photocatalytic H2 Evolution Activity

Abstract： The use of solar energy as an inexhaustible resource to conductphotocatalytic water splitting in hydrogen （H2） production can alleviate theworldwide energy crisis and achieve carbon neutrality. However， research inphotocatalytic H2 evolution reaction （HER） is extremely challenging in terms ofexploring the current development of an active and durable graphitic carbon nitride（g-C3N4）-based photocatalyst. Several parameters of pristine g-C3N4 requirestructural， physical， and chemical improvements， such as optimization of thesurface area， electron transfer， and photo-generated carrier recombination， torender the g-C3N4 suitable for photocatalysis. In this study， the development of anefficient and robust S-doped g-C3N4 （S-g-CN） catalyst was pursued that involvesdoping nitrogen （N） active sites of g-C3N4 with sulfur （S） dopants via one-stepcalcination of the sulphate and melamine precursors. A combination of structural and spectroscopic fingerprints wasestablished to distinctly determine the realization of S-doping onto the g-C3N4 structure. We obtained the optimum Gibbsfree energy of adsorbed hydrogen （ΔGH*） for S-g-CN at the S active sites， which is nearly zero （～0.26 eV）， suggesting thatthe filled S dopants play an essential role in optimizing the adsorption and desorption processes of H-active intermediates.The results of atomic force and transmission electron microscopies （AFM and TEM） demonstrated that the produced S-g-CN catalyst has an ultrathin nanosheet structure with a lamellar thickness of approximately 2.5 nm. A subsequent N2sorption isotherms test revealed a substantial increase in the specific surface area after the integration of S dopants intothe g-C3N4 nanoskeleton. Moreover， the incorporation of S atoms into the g-C3N4 significantly increased the carrierconcentrations， improving the transfer， separation， as well as the oxidation and reduction abilities of the photo-generatedelectron-hole pairs. Leveraging the favorable material characteristics of the S-doped two-dimensional nanostructures， theresulting S-g-CN achieved a high H2 evolution rate of 4923 μmol·g?1·h?1， which is 28 times higher than that of the pristineg-C3N4. Additionally， the developed S-g-CN possessed a high apparent quantum efficiency （3.64%） at visible-lightirradiation. When compared to the recently reported S-doped g-C3N4-based photocatalysts， our optimal S-g-CN catalyst（S-CN1.0） showed one of the best HER catalytic activities. Our rational design is based on an effective strategy that notonly explored a promising HER photocatalyst but also aimed to pave the way for the development of other highperformanceg-C3N4 based catalysts.

Key Words： Theoretical prediction; S-doping; g-C3N4; Hydrogen evolution; Photocatalysis

1 Introduction

The excessive fossil fuels consumption as the primarynonrenewable energy resources has triggered imminentenvironmental crisis that become alarming issues. Theseinevitable challenges require serious attentions towardsustainable and efficient energy sources 1，2. Hydrogen （H2） withthe remarkable characteristics of high calorific value， zeroemission and renewable， has been considered to be a promisingenergy source in the 21st century， capable of poweringequipment from portable electronic devices to vehicles 3. Todate， the conventional H2 production is mainly produced via thechemical conversion of natural gas/coal method， which is limitedby the fossil energy consumption and moreover this route has ledto severe environmental contaminations 4. Based on theseintriguing backgrounds， we believe a reliable method is urgentlyrequired to produce environmentally-friendly， inexpensive， andsustainable H2 gas.

Photocatalysis technology， which takes advantage ofinexhaustible solar energy resources， is a promising option for astrategy to mitigate the global energy crisis and eventuallyachieve zero carbon emissions 5，6. Therefore， photocatalyticwater splitting for hydrogen production has attracted extensiveattention from researchers in the field of hydrogen energy 7–9.Prior to the photocatalytic process of H2 evolution reaction（HER）， a suitable catalyst must be designed to provide an efficient H2 gas production 10，11. Although many usefulsemiconductors， including metal （oxy） sulfides， metal oxides，and metal （oxy） nitrides， are constructed as photocatalysts foroverall water splitting， the high-cost， complicated synthesisprocess， and mediocre photocatalytic performance all restricttheir applications 12–14.

Graphitic carbon nitride （g-C3N4） has drawn considerablecenter of interest as a candidate of metal-free photocatalysttowards hydrogen production， due to its wide band gap， robustchemical stability and tunable composition 15，16. However， thephoto-induced carriers of pristine g-C3N4 are strictly confined inthe triazine unit based on the theoretical calculations. This ismainly governed by the excited electrons that are not capable ofbridging N atoms， nor being transferred from one heptazine（C6N7） unit to an adjoining unit 17–19. Furthermore， the relativelow surface area， rapid electron-hole pair recombination， andinadequate light absorption of pristine g-C3N4， results inunsatisfactory photocatalytic HER performance 20，21.

To overcome the aforementioned issues， several approachesare introduced in this research area such as an attempt to performthe shape and size manipulation， element doping， heterojunctionstructure， and composites， etc. Amongst， non-metal elementdoping in g-C3N4， especially sulfur （S） element， has beengenerally considered to be an effective candidate to regulate itsband gap that plays crucial role for the light harvesting and photocatalytic process under visible light region 22–26. Forexample， Wang et al. have fabricated S-doped g-C3N4nanosheets by self-assembling melamine and tri-thiocyanuricacid to study its photocatalytic activity for hydrogen evolution 27.The H2 evolution rate of g-C3N4 after S doping is 11 times higherthan that of g-C3N4. Li et al. verified the outcome of S-doped g-C3N4 can cause the modification of intrinsic electron structureand specific surface area， thus enhancing visible lightabsorption， reactive sites and catalytic properties 28. However，the above previous reports focused almost exclusively on thesynthesis， characterization and catalytic performance of S-dopedg-C3N4. In addition， the chemical nature of S dopants into the g-C3N4 remains elusive. Moreover， the deterministic spatiallocation of S-dopants in the g-C3N4 molecular structure isrequired to identify its specific contribution to the photocatalyticHER performance.

In this work， density functional theory （DFT） calculationsof g-C3N4 are simulated to introduce S dopants into the Nsites，and resulting S-doped g-C3N4 （S-g-CN） can serve as anefficient and robust HER photocatalyst. To confirm theproposed structural model， several experimental results areunambiguously demonstrated that the filling of S-dopants intothe N-sites of g-C3N4 significantly enhance specific surfacearea， regulate carrier concentrations， and improve transfer，separation as well as oxidation and reduction ability of photogeneratedelectron-hole pairs. Based on DFT calculations， itturns out that the filled S-dopants contribute significantly inthe photocatalytic enhancement of HER activity byoptimizing the Gibbs free energy of adsorbed hydrogen（ΔGH*）. Therefore， we present an optimum S-g-CN catalystthat demonstrates an excellent photocatalytic HER activity of4923 μmol?g?1?h?1 compared to its pristine counterparts.Moreover， we achieve the apparent quantum efficiency （AQE）is even up to 3.64% （λ = 420 nm）.

2 Experimental section

2.1 Chemicals

Ammonium sulfate （NH4）2SO4 （AR， ≥ 99.0%）， ammoniumcarbonate （NH4）2CO3 （AR， ≥ 99.0%） and melamine （AR， ≥99.0%） were obtained from Sinopharm Chemical Reagent Co.Ltd.

2.2 Synthesis of S-doped g-C3N4 （S-g-CN）

Typically， different amount of （NH4）2SO4 （0.5， 0.75， 1.0， 1.25，and 1.5 g） and 1 g melamine （MA） were grinded to non-granularpowder， respectively. The resulting mixtures were then annealeddirectly at 550 °C for 4 h under air atmosphere to prepare a seriesof S-g-CNx materials. According to the dosage of （NH4）2SO4， thesamples were respectively named S-g-CN0.5， S-g-CN0.75， S-g-CN1.0， S-g-CN1.25， and S-g-CN1.5. The g-C3N4 was prepared in asimilar way without the addition of （NH4）2SO4 that intended fora comparative material. For comparison， the g-C3N4-（NH4）2CO3control sample was also fabricated through a similar way with（NH4）2CO3 instead of （NH4）2SO4.

3 Results and discussions

The optimized atomic configurations of g-C3N4 and S-g-CNare simulated initially by first-principle DFT calculations asexhibited in Fig. 1a，c. Subsequently， the values of ΔGH* for g-CN， and S-g-CN samples are also calculated to evaluate theirintrinsic HER catalytic activity. We carefully optimized H*active intermediate that adsorbed onto different active sites （Nor S sites） of g-CN and S-g-CN. For g-C3N4， the H* adsorptionmodel is established at N active sites （Fig. 1b）. Meanwhile， weconsider three H* adsorption models are constructed at N and Ssites （Fig. 1d–f）， corresponding to S-g-CNN1， S-g-CNN2 and Sg-CNS， respectively. The targeted ΔGH* value should be close to0 eV， which represents the optimum H* adsorption/desorptionprocesses over an efficient HER catalyst 29–31. According to thecalculation， the value of ΔGH* for g-C3N4 and S-g-CN at N activesites are found to be ?1.92 eV （g-C3N4）， ?1.59 eV （S-g-CNN1），and ?1.55 eV （S-g-CNN2）. Notably， the optimum value of ΔGH*for S-g-CN at the S active sites （S-g-CNS） is calculated to be?0.26 eV， which is approximately close to zero in comparison tothe ΔGH* of g-C3N4 and S-g-CN at N sites （Fig. 1g）. The aboveDFT results confirm the significant role of filling S-dopants intothe N-sites in promoting the HER catalytic activity of g-C3N4.

Guided by theoretical predictions， a versatile and eco-friendly fabrication strategy is employed to prepare S-g-CN by using（NH4）2SO4 as a non-toxic S doping source， while g-C3N4 sampleis also prepared via a typical thermal polycondensation processof MA molecules （Fig. 2a，b）. The sample morphologies of g-C3N4 and S-g-CN are determined by transmission electronmicroscopy （TEM） technique. Fig. 2c displays the TEM imageof g-C3N4 displays a well-defined layered structure. On the otherhands， The TEM observation in Fig. 2d displays the retainedlamellar framework of S-g-CN upon S-doping. We note that thestructural difference in the S-g-CN possess an ultrathinnanosheet structure compared to its counterparts. To confirm thenanosheet thickness of the S-g-CN， the cross-sectional analysisusing atomic force microscope （AFM） is demonstrated as theevidence for the formation of free-standing nanosheets，indicating the lamellar thickness （～2.5 nm） associated to the Sg-CN shown in Fig. 3a. This result corresponds to the thicknessas thin as four layers of g-C3N4 is formed after filling S-dopants 32.

The electron spin-resonance spectroscopy （ESR） is employedto study the evolution of substitution N sites by S dopants in g-C3N4 （Fig. 3b）. Obviously， the intensity of ESR for S-g-CN ismuch higher than that of g-C3N4， inferring the C3N4 matrix hasmore defects with poorer crystallinity upon S-doping. Fig. 3cdisplays the X-ray diffraction （XRD） results of g-C3N4， and S-g-CN obtained at different dosage of （NH4）2SO4， in which thepeaks centered at ～13° and 27° are assigned to the （100） and（002） reflection planes of g-C3N4， which correspond to in-planeand inter-planar stacking of aromatic units， respectively 33，34.Both reflection planes remain at the similar positions after g-C3N4 phase is converted by the S dopants. The gradual reductionof （100） diffraction peak intensity of S-g-CN demonstrates that it s framework rigidity is reduced， which is consistent to thereduced layer thickness after introducing S dopants as discussedabove 35.

To obtain the porosity textural features of g-C3N4 and serial Sg-CN samples， we further analyzed the N2 sorption isothermsprofile as displayed in Fig. 3d. Noteworthy， the specific surface areas of all S-g-CN samples are generally larger than that oforiginal g-C3N4， which could be assigned to the stripping effectof the gas generated by the decomposition of （NH4）2SO4 duringsample preparation process. Particularly， the S-g-CN1.0 samplepossesses the highest specific surface area of 216.23 m2?g?1 thatprovide many catalytic active sites for the interface reaction（Table S1）， thus expecting to become an excellent photocatalysttowards HER 36，37.

X-ray photoelectron spectroscopy （XPS） technique isemployed to investigate the influence of filling S-dopants intothe N-sites of g-C3N4 （Table S2）. The prominent peaks observedin Fig. 4a are assigned to C 1s and N 1s. A distinct feature was collected compared to the XPS spectra of g-C3N4， here wemeasured that the chemical states corresponding to thecharacteristic of S 2p state at 165.4 eV peak has emerged in theXPS spectra of S-g-CN1.0， suggesting the S-doping formation.Furthermore， the atomic ratios of C ： N in S-g-CN （0.806） ishigher than that of g-C3N4 （0.727）， revealing that some N activesites are replaced by S dopants. The spectral characteristic of C1s and N 1s in these samples are similar in their lineshapes andbinding energies. To outline their chemical states， the C 1s， N 1sand S 2p spectra for the g-C3N4 and S-g-CN1.0 are studiedsystematically. The C 1s core level spectra are deconvoluted intothree major peaks located at 284.4， 287.2， and 287.8 eV，corresponding to the C―C， C―NH2， and N―C＝N in triazineskeleton rings of g-C3N4 38，39， respectively （Fig. 4b）. For the Sg-CN1.0， the peak intensity of N―C＝N is decreased relative tothat of g-C3N4， implying the successful S doping into g-C3N4.Both the spectral features recorded in the g-C3N4 and S-g-CN1.0showed the analogous N 1s XPS profiles （Fig. 4c） in which thecorresponding peaks at 398.0， 398.6， and 400.3 eV can beascribed to C―N＝C， N―（C）3 and C―NH2， respectively 40，41.We note that the peak intensity of C―N＝C for S-g-CN1.0 islesser than that of g-C3N4， suggesting several of sp2 hybridizedN atoms are substituted by the S atoms 42. Fig. 4d depicts theS 2p spectra of S-g-CN1.0， where the peaks at 165.4 eV suggestthe presence of C―S―C bond 43. The chemical states of C―S―C have shown that S atom is successfully incorporated intothe N-sites of g-C3N4 structural unit.

The photocatalytic HER performance of as-fabricatedcatalysts is assessed in a triethanolamine solutions the visiblelight irradiation （≥ 420 nm， Experimental detail， SupportingInformation）. As expected， pristine g-C3N4 can only release atrace amount of H2 （Fig. 5a，b）， while the significantly enhancedH2 generation efficiency is reached over all g-CNx catalyst，which indicates the significant role of S-doping in improving thephotocatalytic activity. Particularly， the optimized catalyst of SCN1.0achieved the champion H2 production rate of 4923μmol?h?1?g?1， which is about 28 times larger than that of g-C3N4（173 μmol?h?1?g?1）. Moreover， the S-CN1.0 catalyst holds one ofthe best catalytic activity compared to other S-doped C3N4-basedphotocatalysts reported recently （Table 1） 44–54. Fig. S1 displayedthe specific surface area and photocatalytic HER activity of theg-C3N4-（NH4）2CO3. Notably， the g-C3N4-（NH4）2CO3 alsopossesses a large specific surface area （186.37 m2?g?1）， but itscatalytic activity is much lower than that of S-g-CN （Fig. 5a）.We believe that the incremental changes of surface area upon Sdopingcould be the turnover factor to improve the photocatalyticHER activity of g-C3N4. Therefore， we used g-C3N4-（NH4）2CO3as the comparative sample via direct pyrolysis of melamine with（NH4）2CO3. According to these results， we suggest that S-dopingis the promising route to achieve the highly-efficient reaction of photocatalytic HER activity of g-C3N4 rather than the increaseof specific surface area.

Additionally， the apparent quantum efficiency （AQE） of S-g-CN1.0 is measured as a function of different incident illuminationwavelengths by using band-pass filters. One should considerAQE is an indispensable indicator to measure the photocatalyticefficiency of a catalyst. The larger AQE indicates a prominentaccess to identify the separation efficiency that influences thephoto-induced charge pairs， thus achieving superior HERphotocatalytic activity. We present Fig. 5c to outline theimplication of incident light wavelength on HER catalyticactivity， which suggests that the HER photocatalysis is mainlydriven by the incident photons. Impressively， the AQE of S-g-CN1.0 reaches to 3.64% at 420 nm， and even at 550 nm， the AQEis still 0.37. Apart from the ideal photocatalytic activity， the longtermstability is a deterministic feature of the catalyst that mightbe another major concern to consider 55. Therefore， the durabilityof S-g-CN1.0 is also studied by comparing the activity attenuationafter six catalytic cycles. As exhibited in Fig. 5d， the H2generation rates of S-g-CN1.0 remains unperturbed after sixcycles， demonstrating their robust photocatalytic durability.

To explore further insight of the photocatalytic activityenhancement upon the filling S-dopants into the N-sites of g-C3N4， we systematically investigate the carrier concentration， thetransfer， and the recombination ability of photo-induced carriersby recording the Mott-Schottky （M-S） plots， electrochemicalimpedance spectra （EIS）， photoluminescence （PL） andphotocurrent （PC） responses. Fig. 6a exhibits the M-S plots of g-C3N4 and serial S-g-CNx materials， where the present positiveslopes indicate the typical n-type semiconductors feature 56.Among them， S-g-CN1.0 delivers the smallest slope，demonstrating the abundant carrier concentration and low photogeneratedelectron/hole pair recombination， which is anindispensable reason for its superior catalytic activity 57.Meanwhile， the typical Nyquist plots of g-C3N4 and serial S-g-CNx materials are exhibited in Fig. 6b. Obviously， the interfacialcharge transfer resistance （the diameter of the semicircle in theEIS spectra） of the S-g-CN1.0 is significantly smaller than thoseof g-C3N4， verifying the fastest charge transport after filling Sdopantsinto the N-sites of g-C3N4 58. Such an effective chargetransfer capacity of developed S-g-CN1.0 can also be furthersupported by the smaller overpotential （Fig. 6c） and Tafel slopetowards electrocatalytic HER in comparison to those of g-C3N4（Fig. S2）. In addition， PL and spectra analyses are carried out toinvestigate the effect of filling S-dopants on the recombinationefficiency of photo-excited carriers. Generally， strong PL peakintensity reveals the rapid recombination of photo-excitedcarriers 59. The PL spectra of g-C3N4 and serial S-g-CNx catalystsall show the emission peak in the range from 415–530 nm at theexcitation wavelength of 275 nm （Fig. 6d）. As expected， the Sg-CN1.0 possesses the lowest PL emission peak， which indicatesthat suitable filling S-dopants in g-C3N4 can effectively preventthe recombination of photo-generated electron-hole pairs， thusendowing superior catalytic HER activity. The separation abilityof electron-hole pairs of g-C3N4 and S-g-CN1.0 are evaluated byrecording their PC curves. As shown in Fig. 6e， the PC responseof S-g-CN1.0 is much stronger than that of g-C3N4， indicating theenhanced separation efficiency of photon-generated carriersafter S doping. In addition， the energy gap changes of g-C3N4after S-doping is disclose on the basis of M-S plots. Fig. 6fdepicts the estimated band gaps energy （Eg） of g-C3N4 and S-g-CNx， which can be calculated from the corresponding UV-Visdiffuse reflectance spectra （UV-Vis DRS， Fig. S3）. Obviously，the S-g-CN1.0 possesses the largest Eg value of is determined tobe 2.74 eV， revealing the enhanced oxidation and reductioncapacity of photo-generated holes and charges after filling Sdopantsinto g-C3N4 structural unit 60，61， which is essential for thepromotion of HER photocatalytic activity. According to theabove discussions， the integration of S dopants not onlyenhances the carrier concentration profile， but also improves thecharge carrier transfer and their splitting event as well asoxidation and reduction ability of photo-generated electron-hole pairs， thereby strengthening photocatalytic H2 evolutionperformance.

4 Conclusion

In summary， guided by the theoretical predictions， we havepresented experimentally a direct demonstration of S-dopantsfilling into the N-sites of g-C3N4 under multitude spectroscopictechniques. Here the rational chemical design of S-g-CN impliesthat the one-step calcining the mixture of sulphate and melaminewithout the usage of toxic agent is a promising route toward itsphotocatalytic response. Owing to the typical lamellarnanostructures with large specific surface area， moderateintegrating S dopants， abundant carrier concentration， as well aseffective carrier separation and transfer efficiency， thedeveloped S-g-CN presents a conspicuously improved HERphotocatalytic activity and durability relative to g-C3N4. Asdisclosed by the DFT calculations， the ΔGH* of S-g-CN at thefilled S active sites is approximately close to zero （?0.26 eV），strongly confirms the significant role of filling S-dopants intothe N-sites in enhancing the HER catalytic activity. Importantly，filling N-sites of g-C3N4 with S-dopants may open a new avenueto design high-performance photocatalyst for hydrogenproduction and other photo（electro）chemical conversionprogress.

Author Contributions： H.W. and J.J. designed the proposaland wrote the manuscript. H.W. and L.Y. performed thesynthesis， characterization and the catalytic measurements.H.W.， J.J.， Arramel， and J.Z.， contributed to revising themanuscript. All authors discussed the results and reviewed themanuscript.

Supporting Information： Supplementary data associatedwith this article is available free of charge via the internet athttp：//www.whxb.pku.edu.cn.， including： Physical characterizations，photoelectrochemical measurements， photocatalytic hydrogenevolution tests， details of theoretical calculations， N2 sorptionisotherms， XPS results， Tafel plots， UV-Vis DRS spectra.

References

（1） Zhao， Z.; Wang， Z.; Zhang， J.; Shao， C.; Dai， K.; Fan， K.; Liang， C.Adv. Funct. Mater. 2023， 33， 2214470. doi： 10.1002/adfm.202214470

（2） Li， Y.; Zhang， M.; Zhou， L.; Yang， S.; Wu， Z.; Ma， Y. Acta Phys. -Chim.Sin. 2021， 37， 2009030. [李云鋒， 張敏， 周亮， 楊思佳， 武占省，馬玉花. 物理化學學報， 2021， 37， 2009030.]doi： 10.3866/PKU.WHXB202009030

（3） Wang， J.; Jiang， J.; Li， F.; Zou， J.; Xiang， K.; Wang， H.; Li， Y.; Li， X.Green Chem. 2023， 25， 32. doi： 10.1039/D2GC03160D

（4） Wang， X.; Wang， X.; Huang， J.; Li， S.; Meng， A.; Li， Z. Nat.Commun. 2021， 12， 4112.doi： 10.21203/rs.3.rs-208751/v1

（5） Wang， Z.; Liu， R.; Zhang， J.; Dai， K. Chin. J. Struct. Chem. 2022， 41，2206015. doi： 10.14102/j.cnki.0254-5861.2022-0108

（6） Liu， T.; Li， Y. F.; Sun， H. J.; Zhang， M.; Xia， Z. L.; Yang， Q. Chin. J.Struct. Chem. 2022， 41， 2206055.doi： 10.14102/j.cnki.0254-5861.2022-0152

（7） Li， X.; Liu， J.; Huang， J.; He， C.; Feng， Z.; Chen， Z.; Wan， L.; Deng，F(xiàn). Acta Phys. -Chim. Sin. 2021， 37， 2010030. [李喜寶， 劉積有， 黃軍同， 何朝政， 馮志軍， 陳智， 萬里鷹， 鄧芳. 物理化學學報，2021， 37， 2010030.] doi： 10.3866/PKU.WHXB202010030

（8） Li， X.; Luo， Q.; Han， L.; Deng， F.; Yang， Y.; Dong， F. J. Mater. Sci.Technol. 2022， 114， 222. doi： 10.1016/j.jmst.2021.10.030

（9） Shen， R.; Ren， D.; Ding， Y.; Guan， Y.; Ng， Y.; Zhang， P.; Li， X. Sci.China Mater. 2022， 63， 2153. doi： 10.1007/s40843-020-1456-x

（10） Zhang， S.; Dong， H.; An， C.; Li， Z.; Xu， D.; Xu， K.; Wu， Z.; Shen， J.;Chen， X.; Zhang， S. J. Mater. Sci. Technol. 2021， 75， 59.doi： 10.1016/j.jmst.2020.10.030

（11） Li， F.; Jiang， J.; Wang， J.; Zou， J.; Sun， W.; Wang， H.; Xiang， K.; Wu，P.; Hsu， J. P. Nano Res. 2023， 16， 127.doi： 10.1007/s12274-022-4799-z

（12） Liu， S; Wang， K; Yang， M; Jin， Z. Acta Phys. -Chim. Sin. 2022， 38，2109023. [劉珊池， 王凱， 楊夢雪， 靳治良. 物理化學學報， 2022，38， 2109023.] doi： 10.3866/PKU.WHXB202109023

（13） Jiang， J.; Zou， Y.; Arramel; Li， F.; Wang， J.; Zou， J.; Li， N. J. Mater.Chem. A 2021， 9， 24195. doi： 10.1039/d1ta07332j

（14） Jiang， J.; Xiong， Z.; Wang， H.; Xiang， K.; Wu， P.; Zou， J. Sci. ChinaTechnol. Sc. 2022， 65， 3020. doi： 10.1007/s11431-022-2192-6

（15） Tao， S. R.; Wan， S. J.; Huang， Q. Y.; Li， C. M.; Yu， J. G.; Cao， S. W.Chin. J. Struct. Chem. 2022， 41， 2206048.doi： 10.14102/j.cnki.0254-5861.2022-0068

（16） Yang， H.; Dai， K.; Zhang， J.; Dawson， G. Chin. J. Catal. 2022， 43，2111. doi： 10.1016/s1872-2067（22）64096-8

（17） Zhao， Z.; Dai， K.; Zhang， J.; Dawson， G. Adv. Sustain. Syst. 2023， 7，2100498. doi： 10.1002/adsu.202100498

（18） Li， Y.; He， Z.; Liu， L.; Jiang， Y.; Ong， W.; Duan， Y.; Ho， W.; Dong， F.Nano Energy 2023， 105， 108032. doi： 10.1016/j.nanoen.2022.108032

（19） Zou， J.; Liao， G.; Wang， H.; Ding， Y.; Wu， P.; Hsu， J. P.; Jiang， J.J. Alloy. Compd. 2022， 911， 165020.doi： 10.1016/j.jallcom.2022.165020

（20） Zou， J.; Liao， G.; Jiang， J.; Xiong， Z.; Bai， S.; Wang， H.; Wu， P.;Zhang， P.; Li， X. Chin. J. Struct. Chem. 2022， 41， 2201025.doi： 10.14102/j.cnki.0254-5861.2021-0039

（21） Fu， J.; Xu， Q.; Low， J.; Jiang， C.; Yu， J. Appl. Catal. B Environ. 2019，243， 556. doi： 10.1016/j.apcatb.2018.11.011

（22） Shen， R.; Hao， L.; Chen， Q.; Zheng， Q.; Zhang， P.; Li， X. Acta Phys. -Chim. Sin. 2022， 38， 2110014. [沈榮晨， 郝磊， 陳晴， 鄭巧清， 張鵬，李鑫. 物理化學學報， 2022， 38， 2110014.]doi： 10.3866/PKU.WHXB202110014

（23） Liu， Y.; Zheng， Y.; Zhang， W.; Peng， Z.; Xie， H.; Wang， Y.; Guo， X.;Zhang， M.; Li， R.; Huang， Y. J. Mater. Sci. Technol. 2021， 95， 127.doi： 10.1016/j.jmst.2021.03.025

（24） Wang， J.; Wang， G.; Cheng， B.; Yu， J.; Fan， J. Chin. J. Catal. 2021，42， 56. doi： 10.1016/s1872-2067（20）63634-8

（25） Chen， Y.; Su， F.; Xie， H.; Wang， R.; Ding， C.; Huang， J.; Xu， Y.; Ye，L. Chem. Eng. J. 2021， 404， 126498. doi： 10.1016/j.cej.2020.126498

（26） Jiang， J.; Xiong， Z.; Wang， H.; Liao， G.; Bai， S.; Zou， J.; Wu， P.;Zhang， P.; Li， X. J. Mater. Sci. Technol. 2022， 118， 15.doi： 10.1016/j.jmst.2021.12.018

（27） Wang， H.; Bian， Y. R.; Hu， J. T.; Dai， L. M. Appl. Catal. B Environ.2018， 238， 592. doi： 10.1016/j.apcatb.2018.07.023

（28） Zhou， Y.; Lv， W.; Zhu， B.; Tong， F.; Pan， J.; Bai， J.; Zhou， Q.; Qin， H.ACS Sustain. Chem. Eng. 2019， 7， 5801.doi： 10.1021/acssuschemeng.8b05374

（29） Bai， S.; Yang， M.; Jiang， J.; He， X.; Zou， J.; Xiong， Z.; Liao， G.; Liu，S. npj 2D Mater. Appl. 2021， 5， 78. doi： 10.1038/s41699-021-00259-4

（30） Liu， D.; Xu， G.; Yang， H.; Wang， H.; Xia， B. Y. Adv. Funct. Mater.2023， 33， 2208358. doi： 10.1002/adfm.202208358

（31） Jiang， J.; Bai， S.; Yang， M.; Zou， J.; Li， N.; Peng， J.; Wang， H.; Xiang，K.; Liu， S.; Zhai， T. Nano Res. 2022， 15， 5977.doi： 10.1007/s12274-022-4276-8

（32） Qin， Z.; Wu， J.; Li， B.; Su， T.; Ji， H. Acta Phys. -Chim. Sin. 2021， 37，2005027. [秦祖贈， 吳靖， 李斌， 蘇通明， 紀紅兵. 物理化學學報，2021， 37， 2005027.] doi： 10.3866/PKU.WHXB202005027

（33） Zou， J.; Wu， S.; Liu， Y.; Sun， Y.; Cao， Y.; Hsu， J. P.; Wee， A. T. S.;Jiang， J. Carbon 2018， 130， 652. doi： 10.1016/j.carbon.2018.01.008

（34） Feng， C.; Tang， L.; Deng， Y.; Wang， J.; Liu， Y.; Ouyang， X.; Yang， H.;Yu， J.; Wang， J. Appl. Catal. B Environ. 2021， 281， 119539.doi： 10.1016/j.apcatb.2020.119539

（35） Deng， Y.; Zhou， Z.; Zeng， H.; Tang， R.; Li， L.; Wang， J.; Feng， C.;Gong， D.; Tang， L.; Huang， Y. Appl. Catal. B Environ. 2022， 320，121942. doi： 10.1016/j.apcatb.2022.121942

（36） Wang， H.; Qiu， X.; Peng， Z.; Wang， W.; Wang， J.; Zhang， T.; Jiang，L.; Liu， H. J. Colloid Interface Sci. 2020， 561， 829.doi： 10.1016/j.jcis.2019.11.065

（37） Hua， J.; Wang， Z.; Zhang， J.; Dai， K.; Shao， C.; Fan， K. J. Mater. Sci.Technol. 2023， 156， 64. doi： 10.1016/j.jmst.2023.03.003

（38） Jiang， J.; Lei， O. Y.; Zhu， L.; Zheng， A.; Zou， J.; Yi， X.; Tang， H.Carbon 2014， 80， 213. doi： 10.1016/j.carbon.2014.08.059

（39） Su， T.; Hood， Z. D.; Naguib， M.; Bai， L.; Luo， S.; Rouleau， C. M.;Ivanov， I. N.; Ji， H.; Qin， Z.; Wu， Z. Nanoscale 2019， 11， 8138.doi： 10.1039/c9nr00168a

（40） Hu， Y.; Li， X.; Wang， W.; Deng， F.; Han， L.; Gao， X.; Feng， Z.; Chen，Z.; Huang， J.; Zengi， F.; et al. Chin. J. Struct. Chem. 2022， 41，2206069. doi： 10.14102/j.cnki.0254-5861.2022-0103

（41） Wu， M.; Yan， J.; Tang， X.; Zhao， M.; Jiang， Q. ChemSusChem 2014，7， 2654. doi： 10.1002/cssc.201402180

（42） Lin， Y. R.; Dizon， G. V. C.; Yamada， K.; Liu， C. Y.; Venault， A.; Lin，H. Y.; Yoshida， M.; Hu， C. J. Colloid Interface Sci. 2020， 567， 202.doi： 10.1016/j.jcis.2020.02.017

（43） Wang， H.; Qiu， X.; Wang， W.; Jiang， L.; Liu， H. Front. Chem. 2019，7， 855. doi： 10.3389/fchem.2019.00855

（44） Bai， J.; Zhou， P.; Xu， P.; Deng， Y.; Zhou， Q. Ceram. Int. 2021， 47，4043. doi： 10.1016/j.ceramint.2020.09.275

（45） Jiao， Y.; Liu， M.; Qin， J.; Li， Y.; Wang， J.; He， Z.; Li， Z. J. ColloidInterface Sci. 2022， 608， 1432. doi： 10.1016/j.jcis.2021.10.084

（46） Fei， T.; Qin， C.; Zhang， Y.; Dong， G.; Wang， Y.; Zhou， Y. Int. J.Hydrog. Energy 2021， 46， 20481. doi： 10.1016/j.ijhydene.2021.03.148

（47） Li， J.; Liu， X.; Liu， C.; Che， H.; Li， C. J. Taiwan. Inst. Chem. E 2020，117， 93. doi： 10.1016/j.jtice.2020.12.001

（48） Zhang， T.; Cai， X.; Lin， X.; Jiang， Z.; Jin， H.; Huang， Z.; Gan， T.; Hu，H.; Zhang， Y. Sep. Purif. Technol. 2023， 314， 123618.doi： 10.1016/j.seppur.2023.123618

（49） Niu， L.; Du， J.; Tian， X.; Jiang， D.; Gu， L.; Yuan， Y. Mater. Lett. 2021，300， 130120. doi： 10.1016/j.matlet.2021.130120

（50） Fang， K.; Chen， Z.; Wei， Y.; Fang， S.; Dong， Z.; Zhang， Y.; Li， W.;Wang， L. J. Alloy. Compd. 2022， 925， 166257.doi： 10.1016/j.jallcom.2022.166257

（51） Long， D.; Wang， L.; Cai， H.; Rao， X.; Zhang， Y. Catal. Lett. 2020，150， 2487. doi： 10.1007/s10562-020-03156-5

（52） Ahmad， K.; Khan， M. Q.; Alsalme， A.; Kim， H. Synth. Met. 2022，288， 117100. doi： 10.1016/j.synthmet.2022.117100

（53） Zhou， P.; Meng， X.; Li， L.; Sun， T. J. Alloy. Compd. 2020， 827，154259. doi： 10.1016/j.jallcom.2020.154259

（54） Feng， C.; Tang， L.; Deng， Y， Wang， J.; Liu， Y.; Ouyang， X.; Yang， H.;Yu， J.; Wang， J. Appl. Catal. B Environ. 2021， 281， 119539.doi： 10.1016/j.apcatb.2020.119539

（55） Zou， J.; Zou， Y.; Wang， H.; Wang， W.; Wu， P.; Arramel; Jiang， J.; Li，X. Chin. Chem. Lett. 2023， 34， 107378.doi： 10.1016/j.cclet.2022.03.101

（56） Zhao， Z.; Li， X.; Dai， K.; Zhang， J.; Dawson， G. J. Mater. Sci.Technol. 2022， 117， 109. doi： 10.1016/j.jmst.2021.11.046

（57） Che， W.; Cheng， W.; Yao， T.; Tang， F.; Liu， W.; Su， H.; Huang， Y.;Liu， Q.; Liu， J.; Hu， F.; et al. J. Am. Chem. Soc. 2017， 139， 3021.doi： 10.1021/jacs.6b11878

（58） Gao， C.; Wei， T.; Zhang， Y.; Song， X.; Huan， Y.; Liu， H.; Zhao， M.;Yu， J.; Chen， X. Adv. Mater. 2019， 31， 1806596.doi： 10.1002/adma.201806596

（59） Ruan， X.; Huang， C.; Cheng， H.; Zhang， Z.; Cui， Y.; Li， Z.; Xie， T.;Ba， K.; Zhang， H.; Zhang， L.; et al. Adv. Mater. 2023， 35， 2209141.doi： 10.1002/adma.202209141

（60） Xia， P.; Cao， S.; Zhu， B.; Liu， M.; Shi， M.; Yu， J.; Zhang， Y. Angew.Chem. Int. Ed. 2020， 59， 5218. doi： 10.1002/anie.201916012

（61） Wang， H.; Jiang， J.; Yu， L.; Peng， J.; Song， Z.; Xiong， Z.; Li， N.;Xiang， K.; Zou， J.; Hsu， J.-P.; et al. Small 2023，doi： 10.1002/smll.202301116

國家自然科學基金（62004143）， 湖北省重點研發(fā)計劃（2022BAA084）， 湖北省自然科學基金（2021CFB133）， 國家重點研發(fā)計劃 （2022YFC3902703）， 磷資源開發(fā)利用教育部工程研究中心創(chuàng)新項目（LCX2021003），能量轉換與存儲材料化學教育部重點實驗室開放基金（2021JYBKF05）資助