摘要：探索高效水分解光催化劑具有獲得氫能源的巨大潛力。調(diào)控異質(zhì)結(jié)界面可以有效地促進(jìn)電荷載流子的分離和太陽能的利用，從而提高光催化活性。本工作使用了一種機械混合輔助自組裝方法來構(gòu)建NiPS3 （NPS）納米片（NSs）/C3N5（CN） NSs （NPS/CN）異質(zhì)結(jié)，即在二維（2D） CN NSs表面緊密沉積2D NPS NSs以形成2D/2D異質(zhì)結(jié)構(gòu)。在可見光下，通過在去離子水和海水中分解水生成氫氣來評價樣品的光催化性能。與CN NSs和NPS NSs相比，NPS/CN復(fù)合材料顯示出較高的光催化產(chǎn)氫（PHE）活性，這是由于光捕獲能力增加和異質(zhì)結(jié)形成的協(xié)同作用所致。然而，過量的NPS NSs沉積在CN NSs表面會降低NPS/CN中CN NSs組分的光吸收，從而降低NPS/CN復(fù)合材料的PHE活性。這表明，NPS/CN復(fù)合材料要獲得良好的光催化活性，需要兩個組分之間適當(dāng)?shù)馁|(zhì)量比。優(yōu)化后的光催化劑（3-NPS/CN）具有良好的結(jié)構(gòu)穩(wěn)定性，在可見光下PHE效率最高，為47.71 μmol?h?1，是CN NSs的2385.50倍。此外，3-NPS/CN在海水中也表現(xiàn)出良好的PHE活性，反應(yīng)速率為8.99 μmol?h?1。采用光電化學(xué)、穩(wěn)態(tài)光致發(fā)光（PL）、時間分辨光致發(fā)光（TR-PL）、穩(wěn)態(tài)表面光電壓（SPV）和時間分辨表面光電壓（TPV）技術(shù)研究了不同光催化劑上的電荷分離和遷移。根據(jù)表征結(jié)果提出了一種可能的PHE機理。在NPS/CN光催化劑中，由于CN NSs和NPS NSs之間的電位差和強的界面電子耦合，光生電子從CN NSs的導(dǎo)帶迅速遷移到NPS NSs的導(dǎo)帶。然后，聚積在NPS NSs組份導(dǎo)帶上的光生電子可以有效地還原質(zhì)子生成氫氣分子。同時，在三乙醇胺（TEOA）分子存在下，CN NSs和NPS NSs的價帶上的光生空穴被消耗。本研究提供了一種簡單的2D/2D異質(zhì)結(jié)構(gòu)光催化劑制備方法，該方法對于構(gòu)建高效二維異質(zhì)結(jié)光催化劑在能源領(lǐng)域中的應(yīng)用具有重要價值。

關(guān)鍵詞：C3N5納米片；NiPS3納米片；光催化；產(chǎn)氫；異質(zhì)結(jié)

中圖分類號：O643

Interfacial Engineering of Ultrathin 2D/2D NiPS3/C3N5 Heterojunctions for Boosting Photocatalytic H2 Evolution

Abstract： This study focuses on exploring efficientphotocatalysts for water splitting， which holds great potentialfor harnessing hydrogen （H2） as a renewable energy source.Modulating the heterojunction interface is known to enhancecharge carrier separation and solar energy utilization，thereby boosting photocatalytic activity. In this work， amechanical mixing-assisted self-assembly approach wasdeveloped to construct a heterojunction between NiPS3（NPS） nanosheets （NSs） and C3N5 （CN） NSs. Specifically，two-dimensional （2D） NPS NSs were tightly deposited on 2DCN NSs surface to gain a 2D/2D heterostructure. The photocatalytic performance of the synthesized photocatalysts wasdetermined by their ability to generate H2 through water splitting， both in deionized （DI） water and seawater， under visiblelight. The resulting NPS NSs/CN NSs （NPS/CN） composites possessed boosted photocatalytic hydrogen evolution （PHE）activity related to CN NSs and NPS NSs. This improvement was assigned to the synergistic effect of increased lightharvestingcapacity and heterojunction formation. Nevertheless， an excessive amount of deposited NPS NSs on thesurface of CN NSs was found to reduce the light absorption of the CN NSs component in the NPS/CN composites， resultingin decreased PHE activity. Therefore， it was determined that an appropriate mass ratio between the two components isnecessary to achieve excellent photocatalytic activity for the NPS/CN composites. The optimized photocatalyst， referredto as 3-NPS/CN， demonstrated the highest visible-light-driven PHE efficiency of 47.71 μmol?h?1， which was 2385.50 timeshigher than that of CN NSs. Moreover， 3-NPS/CN also exhibited excellent PHE activity in seawater， with a rate of 8.99μmol?h?1. The photoelectrochemical， steady-state photoluminescence （PL）， time-resolved PL （TR-PL）， steady-state surfacephotovoltage （SPV） and time-resolved surface photovoltage （TPV） techniques were performed to investigate the chargeseparation and migration behaviors of various photocatalysts. Based on the characterization results， our group proposeda reasonable PHE mechanism. In the NPS/CN photocatalysts， photo-induced electrons rapidly migrated from theconduction band （CB） of CN NSs to the CB of NPS NSs due to the potential difference and strong interfacial electroniccoupling between the two materials. The photogenerated electrons accumulated on the CB of the NPS NSs componentefficiently reduced protons to generate H2 molecules. Concurrently， photogenerated holes on the valence band （VB） of CNNSs and NPS NSs were consumed with the assistance of triethanolamine （TEOA） molecules. This study presents a facilemethod for fabricating 2D/2D heterostructured photocatalysts， which hold promise for efficient and robust implementationin energy applications.

Key Words： C3N5 nanosheets; NiPS3 nanosheets; Photocatalysis; Hydrogen production; Heterojunction

1 Introduction

Recently， the escalating energy and environmental challengeshave become crucial global issues. Solar-driven hydrogen （H2）evolution using semiconductor photocatalysts has emerged asa promising approach to harness solar energy 1，2. Titaniumdioxide （TiO2） is a well-known photocatalyst， first reported in1972， and has been commonly used because of its non-toxicity，chemical resistance， abundance and low cost 3–5. However， itswide bandgap restricts its excitation to UV-light only， and itsrapid charge carrier recombination hinders its practicalapplications 4. Therefore， there is a pressing need to developnovel photocatalysts for the effective utilization of the solarenergy 6.

Graphitic carbon nitride photocatalysts， as typical polymersemiconductors， have garnered increasing attention inphotocatalytic fields. They possess nonmetal characteristics，are environmentally friendly， stable， and have tunablebandgaps 7–9. Among them， C3N5 （CN）， with the higher nitrogencontent， shows great potential for visible-light-drivenphotocatalytic water splitting due to its improved sunlightabsorption and more negative CB potential compared tographitic carbon nitride （g-C3N4）. However， pristine CN suffersfrom severe charge carrier recombination and limited lightharvesting ability， mainly attributed to its intrinsic band energystructure 10. To overcome these drawbacks， different modification approaches have been explored， includingmorphology control 11，12， cocatalyst deposition 13， elementaldoping 14，15 and heterojunction formation 16，17. Heterojunctionconstruction， in particular， has been widely employed to designnovel CN-based photocatalysts for efficient hydrogenproduction. Examples include NH2-UiO-66/CN 18，LaCoO3/CN 17， C3N4/rGO/CN 19， CdS/CN 20， and CN/poly（triazine imide） 21. Despite the impressive performanceachieved by CN-based heterostructured photocatalysts， thedevelopment of high-efficiency CN-based photocatalysts withefficient charge separation and transfer remains a challengingissue.

Two-dimensional （2D） transition metal phosphoruschalcogenides （MPCx）， such as NiPS3 （NPS）， have garneredsignificant interest in the fields of catalysis 22–28，optoelectronics 29，30 and sensing 31 due to their abundantavailability， unique structure， excellent activity， and tunableproperties 32. NPS can be exfoliated into nanosheets （NSs），offering a multifunctional platform for enhancing photocatalytichydrogen evolution （PHE） efficiency 33. The thin thickness ofNPS NSs facilitates charge carrier separation and transport，while the large surface area promotes interfacial electroniccoupling and provides abundant surface reactive sites forcatalysis. Considering the properties of NPS and CN， it isreasonable to expect that constructing heterostructured photocatalysts by combining CN NSs with NPS NSs is expectedto promote visible light-driven PHE performance. Notably， noprevious reports have focused on designing composites based onCN NSs and NPS NSs for PHE.

In this context， a self-assembly method was employed tocombine NPS NSs with CN NSs. The incorporation of NPS NSsoffers two significant advantages： increased light harvesting andenhanced charge separation. The photocatalytic performance ofthe synthesized NPS/CN composites was determined by visiblelight-driven PHE reaction. The optimal loading amount of NPSNSs in the composites proved crucial in prolonging lifetimes andfacilitating the charge carrier separation， ultimately resulting inenhanced PHE performance. A reasonable PHE mechanism waspresented based on the experimental results.

2 Experimental section

The characterization details of the synthesized samples are displayed in the Supporting Information.

2.1 Materials

Sodium hydroxide （NaOH， ≥ 96%）， sodium sulfite anhydrous（Na2SO3， ≥ 98%）， ethanol （C2H5OH， ≥ 99.7%）， triethanolamine（TEOA， ≥ 99.9%）， 3-amino-1，2，4-triazole （C2H4N4， ≥ 96%），potassium ferricyanide （K3FeN6C6， ≥ 99.5%）， potassiumferrocyanide trihydrate （K4FeN6C6?3H2O， ≥ 99%）， sodiumsulfate anhydrous （Na2SO4， ≥ 99%） and L-lactic acid （C3H6O3，90%） were purchased from Aladdin Reagent Co.， LTD.，Shanghai， China. Bulk NiPS3 （BNPS） with a high purity（≥ 99.999%） were directly bought from Shenzhen Six CarbonTechnology Co.， LTD. Methanol （CH3OH， ≥ 99.5%） waspurchased from Jiangsu Tongsheng Chemical Reagent Co.， LTD.Sodium sulfite nonahydrate （Na2S·9H2O， ≥ 98%） were providedby Macklin. Potassium bromide （KBr， ≥ 99%）， magnesiumchloride hexahydrate （MgCl2?6H2O， ≥ 98%）， potassium chloride（KCl， ≥ 99.5%）， calcium chloride anhydrous （CaCl2， ≥ 96%） andsodium chloride （NaCl， ≥ 99.5%） were bought from SinopharmChemical Reagent Co.， LTD.， Shanghai， China.

2.2 Preparation of C3N5 （CN） nanosheets （NSs）

To synthesize bulk C3N5 （BCN） 34， 10.0 g of 3-amino-1，2，4-triazole was placed in a covered crucible for the next heating at 550 °C with a rate of 5.0 °C?min?1 for 3 h.

To obtain C3N5 （CN） nanosheets （NSs）， 1.0 g of BCN and 0.36g of NaOH were separately added to 80 mL of deionized （DI）water. The mixture was stirred for 0.5 h and then transferred to aTeflon-lined autoclave. After heating at 120 °C for 12 h， theresulting sediment was collected by centrifugation and washedwith DI water several times until the pH of the suspensionreached approximately 7. The collected solid was dried undervacuum at 60 °C overnight and subsequently calcined in air at500 °C for 2 h with a rate of 5.0 °C?min?1 to obtain CN NSs.

2.3 Preparation of NiPS3 （NPS） NSs

A liquid exfoliation method was employed to prepare nickelphosphorus sulfide NiPS3 （NPS） NSs. Initially， 0.1 g of bulkNiPS3 （BNPS） was added into 40 mL of absolute ethanol. The mixture was subjected to magnetic stirring for 0.5 h， resulting ina homogeneous suspension. Subsequently， continuousultrasonication was carried out for 10 h using a sonicatoroperating at 120 W output power， while maintaining an ice-waterbath. Afterward， the suspension containing BNPS wascentrifuged at a speed of 3000 r?min?1 for 8 min. The resultingsupernatant was collected to obtain a suspension of NPSnanosheets （NPS NSs）. It is worth noting that the unexfoliatedBNPS was carefully collected and dried under vacuum at 80 °Covernight. The concentration of the NPS NSs suspension wasestimated to be 23.5 μg?mL?1.

2.4 Synthesis of NPS NSs/CN NSs （NPS/CN）composites

To fabricate composites of nickel phosphorus sulfidenanosheets/carbon nitride nanosheets （NPS NSs/CN NSs）， 0.6 gof CN NSs and a specific amount of NPS NSs suspension （23.5μg?mL?1） were added to an agate mortar. The mixed suspensionswere continuously ground while being exposed to infrared lightirradiation until the absolute ethanol was evaporated completely.The resulting products were designated as x-NPS/CN （where x =1， 3， 5）， representing the mass ratio of NPS NSs to CN NSs. Forcomparison， two additional samples， 3-BNPS/CN NSs （3-BNPS/CN） and 3-NPS NSs/BCN （3-NPS/BCN） were preparedusing the same method as for 3-NPS/CN， with the exception ofincorporating BNPS and BCN， respectively.

2.5 Synthesis of NPS NSs/g-C3N4 NSs composites

Two types of graphitic carbon nitride nanosheets （g-C3N4NSs） were prepared using melamine and urea as raw materials，respectively. The g-C3N4 NSs derived from melamine weresynthesized based on a previously reported method 9. In a typicalsynthesis， 10.0 g of melamine was placed in a covered crucibleand directly calcined at 550 °C for 2 h in air with a rate of2.3 °C?min?1. Subsequently， 3.0 g of the obtained powder wassubjected to further calcination at 500 °C for 2 h while exposedto air. The resulting powder was collected and labeled as CNMNSs for subsequent use.

Besides that， g-C3N4 NSs were fabricated using urea as thestarting material 8. Typically， 10.0 g of urea was added to acovered ceramic crucible and subjected to a heating treatment at550 °C （heating rate： 2.5 °C?min?1） for 4 h under ambientatmosphere. The resulting solid sample was ground into apowder. Then， 1.5 g of the obtained powder was uniformly addedto a boat crucible and heated at 500 °C for 4 h. The resultingsample was designated as CNU NSs.

Two composites， 3-NPS NSs/CNM NSs （3-NPS/CNM） and 3-NPS NSs/CNU NSs （3-NPS/CNU）， were synthesized using thesame procedure as described for 3-NPS/CN， but with theincorporation of CNM NSs and CNU NSs， respectively.

2.6 Photocatalytic H2 evolution measurement

Photocatalytic hydrogen evolution （PHE） tests of sampleswere conducted in a photocatalytic online reaction andevaluation system （OLPCRS-3， Shanghai Boyi ScientificInstrument Co.， China） equipped with a quartz side irradiation reactor （refer to Fig. S1）. A 300 W Xenon lamp （Ceaulight， CELHXF300，China） was used as the visible light source， along witha 420 nm cut-off filter. Typically， 40.0 mg photocatalyst， 20 mLTEOA （hole sacrificial regent） and 180 mL DI water wereindividually added to the quartz reactor， and then sonicated for2 min to obtain a uniform suspension. To remove any remainingair， the PHE system was evacuated and purged with argon gasfor several cycles. The amount of produced H2 was measuredusing online gas chromatography （Shimadzu， GC-2014， Japan），equipped with a thermal conductivity detector （TCD）. The PHEexperiments with 3-NPS/CN were performed over five cycles toevaluate the photocatalytic stability. Furthermore， PHEexperiments in seawater were conducted using the sameprocedure as described above， with the addition of naturalseawater sourced from Qingdao， Shandong， China.

The apparent quantum yield （AQY） values of 3-NPS/CN weremeasured under the same experimental conditions as the PHEexperiments， but with the use of monochromatic light sources（λ = 400， 420， 450， 500 nm）. The corresponding averageirradiation intensities were measured and are presented in TableS1. The AQY values were achieved based on the followingequation 35：

where nH2 was the molar mass of generated H2 molecules， h wasthe Planck constant， NA was Avogadro constant， c was the speedof light， P was the intensity of irradiation light， S meant theirradiation area， t was the photocatalysis time， and λ representedthe wavelength of monochromatic light.

2.7 Photoelectrochemical measurements

Photoelectrochemical measurements were carried out on aCHI660E electrochemical workstation （Shanghai Chenhua，China） in a standard three-electrode system. Pt foil served as thecounter electrode， while Ag/AgCl （saturated KCl） was used asthe reference electrode. The working electrode was preparedthrough the following steps： 1.0 mg photocatalyst was mixedwith 1 mL Nafion reagent （0.5 wt%， mass fraction） and treatedwith ultrasonication. The resulting slurry （40 μL） was coatedonto the conductive surface of an FTO （Fluorine Tin Oxide）glass plate， with an approximate area of 1 cm2 （1 cm × 1 cm）.Linear sweep voltammetry （LSV） measurements wereconducted to study overpotential， and transient photocurrentresponses were recorded in a 0.2 mol?L?1 Na2SO4 aqueoussolution. Mott-Schottky plots were determined in a 0.5 mol?L?1Na2SO4 solution （pH = 2， H2SO4） at a frequency of 1 kHz. Forelectrochemical impedance spectroscopy （EIS） tests， the electrolyteused was a mixed solution of K3[Fe（CN）6]/K4[Fe（CN）6]/KCl （0.01mol?L?1/0.01 mol?L?1/0.5 mol?L?1）. The potential measuredusing Ag/AgCl （saturated KCl） was converted to the normalhydrogen electrode （NHE） scale according to the Nernstequation as follows 36：

ENHE = EAg/AgCl - 0.197 + 0.059 × PH

3 Results and discussion

The synthesis process of NPS NSs/CN NSs （NPS/CN）composites is depicted in Fig. 1. Initially， 3-amino-1，2，4-triazolewas heated to release ammonia-based gas during pyrolysis，leading to the formation of bulk C3N5 （BCN）. The obtained BCNunderwent alkali-assisted hydrothermal treatment andsubsequent calcination to achieve the thermal oxidative etchingof BCN， resulting in the formation of CN NSs 18. Meanwhile，NPS NSs were obtained through the liquid exfoliation methodassisted by ultrasonic treatment， taking advantage of their lowinterlayer force 33. Finally， the mixed suspension containing NPSNSs and CN NSs was continuously ground to establish a strongconnection between these two components through Van derWaals forces.

The field emission scanning electron microscopy （FESEM）image of BCN displays a layered structure with curved sheetaccumulation （see Fig. S2）. Following the two-stephydrothermal and calcination treatments， the resulting CN NSsexhibits an irregular granular distribution and smaller particlesize compared to BCN. Agglomeration of CN NSs is observeddue to the thermal etching effect （Fig. 2a，b）.

Commercial BNPS displays a two-dimensional （2D） structurewith granular sizes in the micrometer range （Fig. 2c，d）. Uponcombining NPS NSs with CN NSs， the morphology of 3-NPS/CNremains almost unchanged compared to CN NSs （Fig. 2e，f）. Thiscan be attributed to the low loading amount and uniform dispersion of NPS NSs. Additionally， the textures of 1-NPS/CNand 5-NPS/CN are similar to those of CN NSs and 3-NPS/CN（Fig. S3）. It implies that the introduction of NPS NSs with asmall loading amount （lt; 5 wt%） has no significant effect on themicrostructure of CN NSs.

The morphologies of CN NSs， NPS NSs and 3-NPS/CN werealso investigated using transmission electron microscopy（TEM）. CN NSs exhibit a 2D microstructure with stacked sheets（Fig. 3a）. The absence of clear lattice fringes in the highresolutionTEM （HRTEM） image confirms their amorphousnature 37 （Fig. 3b）. The TEM image of NPS NSs shows distinctnanosheets with a smooth surface （Fig. 3c）. The HRTEM imageof NPS NSs reveals lattice fringes with a spacing of 0.17 nm，corresponding to the （060） crystal plane of NiPS3 （Fig. 3d） 32.

Upon coupling NPS NSs with CN NSs， the morphology of 3-NPS/CN remains similar to that of CN NSs， confirming theminimal impact of NPS NSs loading on the microstructure of CN NSs （Fig. 3e）. The HRTEM image of 3-NPS/CN displays a clearlattice boundary at the junction， indicating the formation of a2D/2D heterostructure between NPS NSs and CN NSs （Fig. 3f）.The observed lattice fringes of 0.17 nm in the enlarged HRTEMimage of 3-NPS/CN is attributed to the crystal plane of NiPS3（060）. Moreover， the absence of lattice fringes on one side of theimage indicates the presence of amorphous CN NSs. Thissuggests that 2D NPS NSs are closely decorated on the surfaceof 2D CN NSs to produce 2D/2D heterostructure between thesetwo components.

The STEM-HAADF and corresponding elemental mappingimages （Fig. 3g–l） clearly demonstrate the uniform spatialdistribution of carbon （C）， nitrogen （N）， nickel （Ni）， phosphorus（P） and sulfur （S）. This confirms the successful integration ofNPS NSs with CN NSs， forming a 2D/2D NPS/CNheterojunction. This structure enables efficient charge carrierseparation.

The crystal structures of BNPS， CN NSs and x-NPS/CN （x =1， 3， 5） composites were confirmed through X-ray diffraction（XRD） patterns. For CN NSs and BCN （Fig. S4）， two diffractionpeaks at 2θ = 13.1°， 27.7°， corresponding to the （100） and （002）planes， respectively. These peaks are associated with in-planestructural ordering and interlayer stacking of the aromaticsystem in graphite materials， respectively 38，39. Compared toBCN， CN NSs exhibit relatively low peak intensities at 2θ =13.1°， 27.7°， indicating a reduced in-plane periodicity andincreased interlayer spacing， confirming the successfulexfoliation from BCN to CN NSs 40. The XRD pattern of BNPS（Fig. 4a） matches well with monoclinic NiPS3 （PDF card No.33-0952）， with three prominent peaks at 14.0°， 28.2° and58.1° due to the （001）， （002） and （004） planes， respectively24，41. With the increase in NPS NSs loading content， the peakintensities at 14.0°， associated with the （001） plane of NPS，gradually increase in the x-NPS/CN （x = 1， 3， 5） composites.Importantly， the XRD patterns of x-NPS/CN （x = 1， 3， 5）composites contain the characteristic peaks of both NPS and CNNSs， demonstrating the successful preparation of the NPS/CNcomposites.

Fourier transform infrared （FT-IR） spectra were measured forCN NSs and x-NPS/CN （x = 1， 3， 5） composites to identify theirchemical bonds （Fig. 4b）. In the spectrum of CN NSs， a sharpsignal at ～811 cm?1 is related to bending vibrations of N―H 42.Several characteristic peaks observed at 1244， 1323， 1410， 1461，and 1639 cm?1 are assigned to triazine ring stretching， while a prominent signal at 3172 cm?1 originates from N―H stretchingvibration 38，43. The x-NPS/CN （x = 1， 3， 5） composites displaysimilar peak positions to CN NSs， with no significant signalsderived from NPS， indicating the small deposition amount andhigh dispersion of NPS NSs in the x-NPS/CN （x = 1， 3， 5）composites. These results manifest that NPS NSs loading inNPS/CN composites has a negligible effect on the chemicalstates of CN NSs.

The Brunauer-Emmett-Teller （BET） surface area and poresize distribution of BCN， CN NSs and 3-NPS/CN were recordedand shown in Fig. 5a. The BCN， CN NSs and 3-NPS/CN exhibitsimilar type IV isothermal curves， with hysteresis loop shapesresembling the H3 type at relatively high pressures （P/P0 = 0.5–1.0） 44. The pore size distribution curves （Fig. 5b） indicate astructure for 3-NPS/CN， attributed to the restacking ofnanosheets 6，45. The BET surface areas of BCN， CN NSs and 3-NPS/CN are measured to be 8.9， 37.1 and 31.8 m2?g?1，respectively. Compared to BCN， the significantly increased BETsurface area of CN NSs indicates the successful exfoliation intoCN NSs. The BET surface area of 3-NPS/CN is slightlydecreased compared to CN NSs due to the presence of NPS NSs，which can block existing pores in CN NSs. The high BETsurface area and mesoporous structure of 3-NPS/CN arebeneficial to PHE reaction as they provide abundant activesites 35.

X-ray photoelectron spectroscopy （XPS） analysis wasconducted to verify the elemental composition and chemical states of the samples. The XPS survey spectrum of 3-NPS/CN（Fig. 6a） reveals the presence of Ni， P， S， O， C and N elements，indicating the combination of NPS NSs with CN NSs in 3-NPS/CN. However， weak signals of Ni， P， and S are detected in3-NPS/CN due to the low loading amount of NPS NSs. The XPSsurvey spectrum of NPS NSs shows additional peaks of C andO， likely originating from adventitious carbon and adsorbedgaseous molecules， respectively. The C 1s XPS spectrum of CNNSs （Fig. 6b） exhibits two prominent peaks at 284.8 and 287.9eV， corresponding to adventitious carbon （C―C） and N＝C―Ngroups， respectively 46. In the XPS N 1s spectrum of CN NSs（Fig. 6c）， three characteristic peaks are observed at 398.4， 400.2and 404.0 eV， corresponding to C ＝ N―C， C―N ＝N―C/C―NH2， and π electron delocalization， respectively47. XPS C 1s and N 1s peaks of 3-NPS/CN show slight shiftstoward higher binding energies compared to those of bare CNNSs， confirming the strong interaction between NPS NSs andCN NSs 48.

Furthermore， the XPS Ni 2p spectrum of NPS NSs （Fig. 6d）exhibits two sets of peaks. One of the triplet peaks at lowerenergies corresponds to the Ni 2p3/2 spin-orbit， with an apparentsignal at 855.0 eV， while the other two weak peaks around 859.7and 864.9 eV are attributed to the satellite peaks of Ni 2p3/2 49.Similarly， the triplet peaks at 872.4， 876.7 and 882.2 eV areassociated with the Ni 2p1/2 spin-orbit 32. The binding energiesof Ni 2p in 3-NPS/CN shift to lower values compared to NPSNSs， suggesting the presence of Ni2+ state in NPS NSs and 3-NPS/CN. The multiple satellite peaks indicate the shake-uptransitions of ligand-to-metal charge transfer 49.

The high-resolution XPS P 2p spectrum of NPS NSs showstwo divided peaks at 132.0 and 132.9 eV， corresponding to the2p3/2 and 2p1/2 states， respectively （Fig. 6e） 50. Similarly， the XPSS 2p spectrum of NPS NSs exhibits two characteristic peakslocated at 162.5 eV for the S 2p1/2 state and 163.7 eV for the S 2p3/2 state （Fig. 6f）， which can be attributed to the ―PS3 group 50.Importantly， compared to NPS NSs， the binding energies ofNi 2p， P 2p and S 2p in 3-NPS/CN shift to lower values，indicating a strong interaction between CN NSs and NPS NSs.These negative shifts in the XPS Ni 2p， P 2p and S 2p bindingenergies of 3-NPS/CN also imply an increased electron densityfor the NPS NSs component， confirming the efficient transportof photo-induced electrons from CN NSs to NPS NSs across thestrong interfaces.

Visible-light-driven PHE experiments were performed toassess photocatalytic performance of samples， with each sampletested three times for error estimation. The time course of H2generation using triethanolamine （TEOA） as a hole scavenger isshown in Fig. 7a. CN NSs and NPS NSs exhibit negligible H2generation owing to their quick recombination rates of photoinducedcharge carriers （Figs. 7a and S5）. However， thecomposites of 1-NPS/CN， 3-NPS/CN and 5-NPS/CN show alinear increase in H2 production with light irradiation time，indicating stable H2 production. However， a slight decreasetendency in PHE activity can be observed for all samples in thefifth hour due to the consumption of the sacrificial agent.

Among all samples， 3-NPS/CN exhibits the highest PHEefficiency of 47.71 μmol·h?1， which is 2385.50， 13.04 and 1.72times higher than CN NSs， 1-NPS/CN and 5-NPS/CN，respectively （Fig. 7b）. The improved PHE activity of 3-NPS/CNis mainly due to the combined effects of widened lightabsorption region and efficient charge carrier separation through the deposition of NPS NSs. However， excessive loading amountof NPS NSs in 5-NPS/CN restrains the light harvesting of theCN NSs component， resulting in reduced charge carriergeneration and decreased PHE activity 8. Therefore， theappropriate loading amount of NPS NSs in NPS/CN compositesis crucial to strike a balance between light harvesting and chargecarrier separation， thereby enhancing the solar energy utilizationand improving PHE activity. Moreover， the PHE activity of 3-NPS/CN surpasses that of most reported materials， assummarized in Table S2.

The wavelength-dependent AQY values over 3-NPS/CN wererecorded under various monochromatic light sources （Fig. 7c）.The measured AQY values over 3-NPS/CN are 8.33% at 400 nm，3.22% at 420 nm， 0.50% at 450 nm and 0.08% at 500 nm， whichare consistent with its optical absorption spectrum. Thisdemonstrates the close relationship between PHE performanceand light-harvesting capacity.

The selection of a suitable hole scavenger is essential for aneffective evaluation of PHE activity. Taking the optimal sampleof 3-NPS/CN as an example， no significant H2 generation isobserved in the absence of hole scavengers because of the rapidrecombination of photogenerated electrons （Fig. S6）. Besides，PHE experiments were conducted using different holescavengers. The corresponding PHE efficiencies measured usingNa2S/Na2SO3， triethanolamine （TEOA）， ethanol （EtOH），methanol （MeOH） and lactic acid （LA） as hole scavengers are0.27， 47.71， 0.51， 0.63， and 0.33 μmol·h?1， respectively. These results indicate that TEOA is the suitable hole scavenger for thePHE reaction in the NPS/CN system.

For comparison， the PHE performance of 3-NPS/BCN and 3-BNPS/CN composites was also investigated. It is evident that thePHE efficiency of 3-NPS/CN is higher than that of 3-NPS/BCNand 3-BNPS/CN （Fig. S7）. Generally， the BET surface area ofthe nanosheets-based materials is larger than that ofcorresponding bulk materials， leading to the higher BET surfacearea value for NPS NSs and CN NSs. After coupling NPS NSswith CN NSs， the resulted 3-NPS/CN exhibits the increased BETsurface area， which can provide more reactive sites forenhancing PHE activity compared to 3-NPS/BCN and 3-BNPS/CN. Furthermore， the uniform distribution of NPS NSson the CN NSs surface， forming a 2D/2D heterojunction in 3-NPS/CN， contributes to improved PHE activity compared tocomposites based on their bulk materials.

To emphasize the superiority of CN NSs， g-C3N4 nanosheetswere used as starting materials to synthesize 3-NPS/CNU and 3-NPS/CNM under the same conditions as 3-NPS/CN. The PHEefficiency of 3-NPS/CN （47.71 μmol·h?1） is found to be 11.1 and16.8 times higher than that of 3-NPS/CNU （4.29 μmol·h?1） and3-NPS/CNM （2.85 μmol·h?1）， respectively （Fig. S8）. The lightharvesting capacity of 3-NPS/CN in the visible-light region isalso higher than that of 3-NPS/CNU and 3-NPS/CNM （Fig. S9）.These results， combined with previous works 46，51， suggest thatC3N5 materials， such as CN NSs， offer significant potential forimproving PHE performance.

The stability of 3-NPS/CN during the PHE cyclingexperiment was evaluated， as shown in Fig. 7d. The PHE rate of3-NPS/CN remains at 89% efficiency compared to the first cycleafter five cycles， with a rate of 42.76 μmol·h?1. The slightdecrease in PHE rate is mainly attributed to the consumption ofthe sacrificial agent 52. The structural stability of 3-NPS/CNduring the PHE reaction was confirmed by XRD patterns， FT-IRspectra and XPS measurements （Figs. S10 and S11）. Nosignificant changes in the crystal structure or surface functionalgroups are observed before and after the PHE reaction for fivecycles， indicating the good structural stability of 3-NPS/CN. Itdemonstrates that the integration of NPS NSs with CN NSs in 3-NPS/CN composites leads to improved PHE efficiency，excellent stability， and efficient charge separation， making it apromising photocatalyst for practical applications.

The practical value of PHE experiments in seawater was alsodemonstrated （Fig. 7e）. While CN NSs shows negligible H2generation due to rapid charge carrier recombination， theincorporation of NPS NSs with CN NSs in the composites of 1-NPS/CN， 3-NPS/CN and 5-NPS/CN results in stable H2production in seawater. This indicates that NPS NSs caneffectively enhance the hydrogen production activity of CN NSseven in seawater. The PHE efficiencies measured in both DIwater and seawater follows the same order： CN NSs lt; 1-NPS/CNlt; 5-NPS/CN lt; 3-NPS/CN. However， the PHE efficiencies in DIwater are higher than those in seawater （Fig. 7f）. To investigate the reasons for the reduced PHE activity in seawater， comparablePHE experiments were conducted with the addition of variouscompounds present in natural seawater， such as NaCl， KCl， KBr，CaCl2， MgCl2 and Na2SO4 53. The results show that the presenceof these salts significantly reduced the PHE rates compared tofreshwater conditions （Fig. S12）. This suggests that thecompetitive adsorption of salts on active sites hinders the PHEreaction over 3-NPS/CN 54. Thus， an appropriate loading massof NPS NSs in NPS/CN composites can effectively boost PHEactivity in both seawater and freshwater， making it promising forpractical applications.

To further investigate the charge separation and migrationbehaviors， electrochemical impedance spectra （EIS） andtransient photocurrent response （TPR） measurements wereperformed on various photocatalysts. Smaller arc radii in theNyquist plot indicate higher charge transfer efficiency and lowerinterfacial resistance 55，56. The arc radiuses of x-NPS/CN （x = 1，3， 5） composites are found to be smaller than those of CN NSsand NPS NSs （Fig. 8a）， indicating that the heterojunctionformation between NPS NSs and CN NSs facilitates reducedresistance for interfacial charge transfer 57，58. Among all thesamples， 3-NPS/CN exhibits the smallest arc radius， indicatingexcellent charge carrier transfer efficiency. The transientphotocurrent response （TPR） intensities of x-NPS/CN （x = 1， 3，5） composites are also higher than those of CN NSs and NPSNSs （Fig. 8b）， reflecting their good separation rates ofphotogenerated charge carriers 59，60. Notably， the optimal sampleof 3-NPS/CN possesses the largest photocurrent intensity（～0.035 μA?cm?2）， which is 3.5 times higher than CN NSs （～0.01μA?cm?2）， further confirming the most efficient charge carrierseparation in 3-NPS/CN.

To further confirm the charge transfer property， steady-statephotoluminescence （PL） and time-resolved PL （TR-PL） spectrawere recorded with an emission peak at 457 nm. In the PLspectra （Fig. 8c）， CN NSs exhibit the strongest emission signal，indicating rapid charge carrier recombination 61. On the otherhand， BNPS shows only weak PL signals when excited at 272nm （Fig. 8c） and 501 nm （Fig. S13）. After the introduction ofNPS NSs， the PL intensities of x-NPS/CN （x = 1， 3， 5）composites are significantly reduced with the lowest signalobserved for 3-NPS/CN. This suggests that the formation of aheterojunction between NPS NSs and CN NSs promotes theseparation of photo-induced electrons and holes， leading toreduced recombination rates 62.

The results from time-resolved PL （TR-PL） measurementsfurther support these findings （Fig. 8d）. The average emissionlifetime （τ） can be determined using the following equation，based on our previous works 63.

τ =A1τ12 + A2τ22／A1τ1 + A2τ2

where A1 and A2 mean the weight factors， and τ represents thefluorescence lifetime 64. Obviously， the average lifetime of 3-NPS/CN （3.74 ns） is longer than that of CN NSs （2.88 ns）.

The surface photovoltage （SPV） technique， which includessteady-state SPV and time-resolved surface photovoltage （TPV）responses， is a highly sensitive and non-destructive method forinvestigating the photo-physics of photogenerated charges insemiconductor photocatalysts 65，66. In the case of 3-NPS/CN， itexhibits a higher SPV response compared to CN NSs （Fig. 8e），indicating enhanced photogenerated charge separation. Thisconfirms that the heterojunction formation between NPS NSsand CN NSs greatly enhances the photogenerated chargeseparation.

Furthermore， TPV spectra were tested to validate thephotogenerated charge separation and transfer mechanism （Fig.8f）. The heterojunction formation between NPS NSs and CNNSs in 3-NPS/CN accelerates the charge transport， as evidencedby the quicker charge extraction time （t2） compared to CN NSs（t1） 67. This， combined with the results from EIS， TPR， PL andTR-PL， suggests that the formed heterojunction between NPSNSs and CN NSs prolongs charge carrier lifetimes， promotescharge separation， and enhances PHE activity 68.

The effect of NPS NSs deposition on hydrogen evolutionkinetics in x-NPS/CN （x = 1， 3， 5） composites was studied usinglinear sweep voltammetry （Fig. 9a） 69. The current density of thesamples follows the order of CN NSs lt; 1-NPS/CN lt; 5-NPS/CNlt; 3-NPS/CN， indicating that the loading of NPS NSs reduces theoverpotential and favors hydrogen production 70.

Ultraviolet-visible diffuse reflectance spectroscopy （UV-Vis DRS） spectra of samples are displayed in Fig. 9b. CN NSs andNPS/CN composites exhibit strong light absorption in both theultraviolet and visible regions. The absorption edges of CN NSsand BNPS are observed around 487 and 967 nm， respectively（Fig. S14）. With an increase in the proportion of NPS NSs in theNPS/CN composites， the color of the composites darkens，indicating enhanced light harvesting. The band gap values of CNNSs and NPS NSs， as indirect semiconductors， are determinedto be 2.20 and 1.38 eV， respectively， based on previous works 18.

Electron spin resonance （ESR） spectra were recorded todetermine the active species during PHE reactions （Fig. 9d）.Under visible light， characteristic peaks of DMPO-·O2? areobserved， indicating the positive contribution of ·O2? radicals tothe PHE reaction. Additionally， characteristic signals ofTEMPO-h+ are detected both in the dark and under visible light.After visible light irradiation for 3 min， the reduced intensity ofTEMPO-h+ signals means the formation of h+ （Fig. 9e）. TheseESR results confirm the presence of photo-generated h+ and ·O2?radicals during PHE reactions over 3-NPS/CN. It is proposedthat photogenerated electrons react with protons to generate H2，while the remaining h+ is eliminated using a hole scavenger suchas TEOA 45.

The electronic band structures of CN NSs and NPS NSs aredetermined using Mott-Schottky plots to gain deeper insight intothe PHE mechanism of NPS/CN composites （Fig. 9f，g）. Theslopes of both CN NSs and NPS NSs indicate that they are n-type semiconductors. The flat band potentials （Efb） of CN NSsand NPS NSs are estimated to be ?0.73 and ?0.57 V （vs.Ag/AgCl， pH ≈ 2）， respectively， based on the tangent interceptsof the plots. By applying the Nernst equation， the Efb values ofCN NSs and NPS NSs can be converted to ?0.81 and ?0.65 V（vs. NHE）， respectively. Generally， the conduction band （CB）potential （ECB） of an n-type semiconductor is approximately 0.1–0.3 V more negative than its flat band potential （Efb） 71. Thus， theECB positions are calculated to be ?1.01 V （vs. NHE） for CN NSsand ?0.85 V （vs. NHE） for NPS NSs by assuming a voltagedifference of 0.2 V between ECB and Efb. Accordingly， thevalence band potentials （EVB） of CN NSs and NPS NSs aredetermined to be +1.19 eV and +0.53 eV （vs. NHE）， respectively，using the formula （EVB = ECB + Eg）， where Eg is the band gapenergy.

According to the above discussion， we propose a possiblePHE mechanism over NPS/CN heterojunctions （Fig. 9h）. Undervisible light irradiation （λ ≥ 420 nm）， both CN NSs and NPS NSsin the NPS/CN composite are excited， generating electrons andholes that reside on the conduction band （CB） and valence band（VB） positions of the respective components. Due to thepotential difference and strong interfacial electronic couplingbetween CN NSs and NPS NSs， the photogenerated electronsquickly migrate from the CB of CN NSs to the CB of NPS NSs.Subsequently， the accumulated photogenerated electrons on theCB of NPS NSs can efficiently reduce protons to generate H2molecules. Meanwhile， the photogenerated holes on the VB ofCN NSs and NPS NSs are consumed in the presence of TEOAmolecules， producing oxidation products 33. Thus， the NPS/CNheterojunction facilitates the efficient charge carrier dissociationand migration， leading to enhanced PHE efficiency.

4 Conclusions

In conclusion， a mechanical mixing-assisted self-assembly approach was successfully employed to fabricate a 2D/2DNPS/CN heterostructured photocatalyst for visible light-drivenPHE. The deposition of NPS NSs onto CN NSs led to thebroadening of the light response region and tuning of the energyband structure in the NPS/CN heterojunctions. The experimentalresults， including photo-electrochemical measurements，photoluminescence， time-resolved photoluminescence andsurface photovoltage， demonstrated that the formation of the2D/2D heterostructure between NPS NSs and CN NSs improvedthe lifetimes of charge carriers and their separation rates. Theenhanced light-harvesting capacity and efficient charge carrierseparation synergistically contributed to the enhanced PHEactivity of CN NSs upon NPS NSs loading. Among the samplestested， the optimized 3-NPS/CN composite exhibited the highestPHE efficiency of 47.71 μmol·h?1 and an AQY value of 8.33%at 400 nm. However， excessive loading of NPS NSs hindered thelight absorption of the CN NSs component， leading to a decreasein PHE efficiency. Therefore， achieving an appropriate massratio of NPS NSs to CN NSs is crucial for balancing lightabsorption and charge separation. Moreover， the PHEefficiencies measured in DI water were higher than those inseawater for all samples， indicating the influence of competitiveadsorption between salts and sacrificial agents on the activesites. Based on the characterization results， a plausible PHEmechanism was proposed for the 3-NPS/CN composite.Considering the excellent PHE activity and good stability ofNPS/CN composites， it is anticipated that these composites holdsignificant potential for PHE applications in both DI water andseawater environments.

Author Contributions： Conceptualization， C.L. and Q.Z.;Methodology， J.H.， K.X.， A.Y. and Z.Z.; Software， J.H.;Validation， W.X.， C.L. and Q.Z.; Formal Analysis， C.L.;Investigation， C.L.; Resources， C.L.; Data Curation， J.H.;Writing-Original Draft Preparation， J.H.; Writing-Review amp;Editing， C.L.; Visualization， J.H.; Supervision， C.L.; ProjectAdministration， Q.Z.; Funding Acquisition， C.L.

Supporting Information： available free of charge via theinternet at http：//www.whxb.pku.edu.cn.

References

（1） Gao， Y.; Xu， B.; Cherif， M.; Yu， H.; Zhang， Q.; Vidal， F.; Wang， X.;Ding， F.; Sun， Y.; Ma， D.; et al. Appl. Catal. B： Environ. 2020， 279，119403. doi： 10.1016/j.apcatb.2020.119403

（2） Liu， C.; Zhang， Q.; Zou， Z. J. Mater. Sci. Technol. 2023， 139， 167.doi： 10.1016/j.jmst.2022.08.030

（3） Fujishima， A.; Honda， K. Nature 1972， 238， 37.doi： 10.1038/238037a0

（4） Tong， Z.; Yang， D.; Xiao， T.; Tian， Y.; Jiang， Z. Chem. Eng. J. 2015，260， 117. doi： 10.1016/j.cej.2014.08.072

（5） Fu， C.; Wu， T.; Sun， G.; Yin， G.; Wang， C.; Ran， G.; Song， Q. Appl.Catal. B： Environ. 2023， 323， 122196.doi： 10.1016/j.apcatb.2022.122196

（6） Liu， C.; Zhang， Y.; Wu， J.; Dai， H.; Ma， C.; Zhang， Q.; Zou， Z.J. Mater. Sci. Technol. 2022， 114， 81.doi： 10.1016/j.jmst.2021.12.003

（7） Gao， Z.; Chen， K.; Wang， L.; Bai， B.; Liu， H.; Wang， Q. Appl. Catal.B： Environ. 2020， 268， 118462. doi： 10.1016/j.apcatb.2019.118462

（8） Qin， Y.; Li， H.; Lu， J.; Feng， Y.; Meng， F.; Ma， C.; Yan， Y.; Meng，M. Appl. Catal. B： Environ. 2020， 277， 119254.doi： 10.1016/j.apcatb.2020.119254

（9） Lin， B.; Li， H.; An， H.; Hao， W.; Wei， J.; Dai， Y.; Ma， C.; Yang， G.Appl. Catal. B： Environ. 2018， 220， 542.doi： 10.1016/j.apcatb.2017.08.071

（10） Yin， H.; Yuan， C.; Lv， H.; Zhang， K.; Chen， X.; Zhang， Y.; Zhang， Y.Powder Technol. 2023， 413， 118083.doi： 10.1016/j.powtec.2022.118083

（11） Wang， H.; Li， M.; Lu， Q.; Cen， Y.; Zhang， Y.; Yao， S. ACS Sustain.Chem. Eng. 2019， 7， 625. doi： 10.1021/acssuschemeng.8b04182

（12） Huang， L.; Liu， Z.; Chen， W.; Cao， D.; Zheng， A. J. Mater. Chem. A2018， 6， 7168. doi： 10.1039/c8ta01458b

（13） Teng， M.; Shi， J.; Qi， H.; Shi， C.; Wang， W.; Kang， F.; Eqi， M.;Huang， Z. J. Colloid Interface Sci. 2022， 609， 592.doi： 10.1016/j.jcis.2021.11.060

（14） Sun， D.; Zhang， X.; Shi， A.; Quan， C.; Xiao， S.; Ji， S.; Zhou， Z.; Li，X.; Chi， F.; Niu， X. Appl. Surf. Sci. 2022， 601， 154186.doi： 10.1016/j.apsusc.2022.154186

（15） Li， K.; Cai， W.; Zhang， Z.; Xie， H.; Zhong， Q.; Qu， H. Chem. Eng. J.2022， 435， 135017. doi： 10.1016/j.cej.2022.135017

（16） Meng， Q.; Yang， X.; Wu， L.; Chen， T.; Li， Y.; He， R.; Zhu， W.; Zhu，L.; Duan， T. J. Hazard. Mater. 2022， 422， 126912.doi： 10.1016/j.jhazmat.2021.126912

（17） Wang， R.; Zhang， K.; Zhong， X.; Jiang， F. RSC Adv. 2022， 12， 24026.doi： 10.1039/d2ra03874a

（18） Wu， B.; Sun， T.; Liu， N.; Lu， L.; Zhang， R.; Shi， W.; Cheng， P. ACSAppl. Mater. Interfaces 2022， 14， 26742.doi： 10.1021/acsami.2c04729

（19） Liu， D.; Yao， J.; Chen， S.; Zhang， J.; Li， R.; Peng， T. Appl. Catal. B：Environ. 2022， 318， 121822. doi： 10.1016/j.apcatb.2022.121822

（20） Shi， J.; Wang， W.; Teng， M.; Kang， F.; E’qi， M.; Huang， Z. J. ColloidInterface Sci. 2022， 608， 954. doi： 10.1016/j.jcis.2021.10.027

（21） Xiong， Z.; Liang， Y.; Yang， J.; Yang， G.; Jia， J.; Sa， K.; Zhang， X.;Zeng， Z. Sep. Purif. Technol. 2023， 306， 122522.doi： 10.1016/j.seppur.2022.122522

（22） Zhang， X.; Zhao， X.; Wu， D.; Jing， Y.; Zhou， Z. Adv. Sci. 2016， 3，1600062. doi： 10.1002/advs.201600062

（23） Wang， J.; Li， X.; Wei， B.; Sun， R.; Yu， W.; Hoh， H. Y.; Xu， H.; Li， J.;Ge， X.; Chen， Z.; et al. Adv. Funct. Mater. 2020， 30， 1908708.doi： 10.1002/adfm.201908708

（24） Wang， F.; Shifa， T. A.; He， P.; Cheng， Z.; Chu， J.; Liu， Y.; Wang， Z.;Wang， F.; Wen， Y.; Liang， L.; et al. Nano Energy 2017， 40， 673.doi： 10.1016/j.nanoen.2017.09.017

（25） Shifa， T. A.; Wang， F.; Cheng， Z.; He， P.; Liu， Y.; Jiang， C.; Wang，Z.; He， J. Adv. Funct. Mater. 2018， 28， 1800548.doi： 10.1002/adfm.201800548

（26） Gusmao， R.; Sofer， Z.; Sedmidubsky， D.; Huber， S.; Pumera， M. ACSCatal. 2017， 7， 8159. doi： 10.1021/acscatal.7b02134

（27） Cheng， Z.; Shifa， T. A.; Wang， F.; Gao， Y.; He， P.; Zhang， K.; Jiang，C.; Liu， Q.; He， J. Adv. Mater. 2018， 30， 1707433.doi： 10.1002/adma.201707433

（28） Barua， M.; Ayyub， M. M.; Vishnoi， P.; Pramoda， K.; Rao， C. N. R.J. Mater. Chem. A 2019， 7， 22500. doi： 10.1039/c9ta06044h

（29） Jenjeti， R. N.; Kumar， R.; Austeria， M. P.; Sampath， S. Sci. Rep.2018， 8， 8586. doi： 10.1038/s41598-018-26522-1

（30） Chittari， B. L.; Park， Y.; Lee， D.; Han， M.; MacDonald， A. H.;Hwang， E.; Jung， J. Phys. Rev. B 2016， 94， 184428.doi： 10.1103/PhysRevB.94.184428

（31） Chu， J.; Wang， F.; Yin， L.; Lei， L.; Yan， C.; Wang， F.; Wen， Y.;Wang， Z.; Jiang， C.; Feng， L.; et al. Adv. Funct. Mater. 2017， 27，1701342. doi： 10.1002/adfm.201701342

（32） Fang， L.; Xie， Y.; Guo， P.; Zhu， J.; Xiao， S.; Sun， S.; Zi， W.; Zhao， H.Sustain. Energy Fuels 2021， 5， 2537. doi： 10.1039/d1se00110h

（33） Ran， J.; Zhang， H.; Fu， S.; Jaroniec， M.; Shan， J.; Xia， B.; Qu， Y.; Qu，J.; Chen， S.; Song， L.; et al. Nat. Commun. 2022， 13， 4600.doi： 10.1038/s41467-022-32256-6

（34） Li， S.; Cai， M.; Liu， Y.; Zhang， J.; Wang， C.; Zang， S.; Li， Y.; Zhang，P.; Li， X. Org. Chem. Front. 2022， 9， 2479.doi： 10.1039/D2QI00317A

（35） Zhang， Q.; Gu， H.; Wang， X.; Li， L.; Zhang， J.; Zhang， H.; Li， Y.-F.;Dai， W.-L. Appl. Catal. B： Environ. 2021， 298， 120632.doi： 10.1016/j.apcatb.2021.120632

（36） Liu， C.; Xiao， W.; Yu， G.; Wang， Q.; Hu， J.; Xu， C.; Du， X.; Xu， J.;Zhang， Q.; Zou， Z. J. Colloid Interface Sci. 2023， 640， 851.doi： 10.1016/j.jcis.2023.02.137

（37） Han， L.; Peng， C.; Huang， J.; Wang， S.; Zhang， X.; Chen， H.; Yang，Y. Rsc Adv. 2021， 11， 36166. doi： 10.1039/d1ra07275g

（38） Zhang， J.; Jing， B.; Tang， Z.; Ao， Z.; Xia， D.; Zhu， M.; Wang， S.Appl. Catal. B： Environ. 2021， 289， 120023.doi： 10.1016/j.apcatb.2021.120023

（39） Kumar， P.; Vahidzadeh， E.; Thakur， U. K.; Kar， P.; Alam， K. M.;Goswami， A.; Mahdi， N.; Cui， K.; Bernard， G. M.; Michaelis， V. K.;et al. J. Am. Chem. Soc. 2019， 141， 5415.doi： 10.1021/jacs.9b00144

（40） Sun， Z.; Luo， E.; Meng， Q.; Wang， X.; Ge， J.; Liu， C.; Xing， W. ActaPhys. -Chim. Sin. 2022， 38， 2003035. [孫志聰， 羅二桂， 孟慶磊， 王顯， 葛君杰， 劉長鵬， 邢巍. 物理化學(xué)學(xué)報， 2022， 38， 2003035.]doi： 10.3866/PKU.WHXB202003035

（41） Wang， J.; Wang， T.; Shi， X.; Wu， J.; Xu， Y.; Ding， X.; Yu， Q.; Zhang，K.; Zhou， P.; Jiang， Z. J. Mater. Chem. C 2019， 7， 14625.doi： 10.1039/c9tc04722k

（42） Zhang， J.; Tao， H.; Wu， S.; Yang， J.; Zhu， M. Appl. Catal. B：Environ. 2021， 296， 120372. doi： 10.1016/j.apcatb.2021.120372

（43） Li， S.; Wang， C.; Cai， M.; Liu， Y.; Dong， K.; Zhang， J. J. ColloidInterface Sci. 2022， 624， 219. doi： 10.1016/j.jcis.2022.05.151

（44） Zhang， G.; Wang， Z.; He， T.; Wu， J.; Zhang， J.; Wu， J. Chem. Eng. J.2022， 442， 136309. doi： 10.1016/j.cej.2022.136309

（45） Liu， C.; Han， Z.; Feng， Y.; Dai， H.; Zhao， Y.; Han， N.; Zhang， Q.;Zou， Z. J. Colloid Interface Sci. 2021， 583， 58.doi： 10.1016/j.jcis.2020.09.018

（46） Sun， T.; Li， C.; Bao， Y.; Fan， J.; Liu， E. Acta Phys. -Chim. Sin. 2023，39， 2212009. [孫濤， 李晨曦， 鮑鈺鵬， 樊君， 劉恩周. 物理化學(xué)學(xué)報， 2023， 39， 2212009.] doi： 10.3866/PKU.WHXB202212009

（47） Che， H.; Wang， J.; Gao， X.; Chen， J.; Wang， P.; Liu， B.; Ao， Y. J.Colloid Interface Sci. 2022， 627， 739.doi： 10.1016/j.jcis.2022.07.080

（48） Zhan， X.; Zheng， Y.; Li， B.; Fang， Z.; Yang， H.; Zhang， H.; Xu， L.;Shao， G.; Hou， H.; Yang， W. Chem. Eng. J. 2022， 431， 134053.doi： 10.1016/j.cej.2021.134053

（49） Zhao， L.; Lei， S.; Tang， C.; Tu， Q.; Rao， L.; Liao， H.; Zeng， W.; Xiao，Y.; Cheng， B. J. Colloid Interface Sci. 2022， 616， 401.doi： 10.1016/j.jcis.2022.02.089

（50） Vedhanarayanan， B.; Chiu， C.-C.; Regner， J.; Sofer， Z.; SeethaLakshmi， K. C.; Lin， J.-Y.; Lin， T.-W. Chem. Eng. J. 2022， 430，132649. doi： 10.1016/j.cej.2021.132649

（51） Mane， G. P.; Talapaneni， S. N.; Lakhi， K. S.; Ilbeygi， H.; Ravon， U.;Al-Bahily， K.; Mori， T.; Park， D.-H.; Vinu， A. Angew. Chem. Int. Ed.2017， 56， 8481. doi： 10.1002/anie.201702386

（52） Bai， J.; Chen， W.; Hao， L.; Shen， R.; Zhang， P.; Li， N.; Li， X. Chem.Eng. J. 2022， 447， 137488. doi： 10.1016/j.cej.2022.137488

（53） Sun， H.; Shi， Y.; Shi， W.; Guo， F. Appl. Surf. Sci. 2022， 593， 153281.doi： 10.1016/j.apsusc.2022.153281

（54） Chen， K.; Shi， Y.; Shu， P.; Luo， Z.; Shi， W.; Guo， F. Chem. Eng. J.2023， 454， 140053. doi： 10.1016/j.cej.2022.140053

（55） Yu， G.; Zhang， Y.; Du， X.; Wu， J.; Liu， C.; Zou， Z. J. ColloidInterface Sci. 2022， 623， 205. doi： 10.1016/j.jcis.2022.05.040

（56） Xiong， Z; Hou， Y.; Yuan， R.; Ding， Z.; Ong， W. J.; Wang， S. ActaPhys. -Chim. Sin. 2022， 38， 2111021. [熊壯， 侯乙東， 員汝勝， 丁正新， 王偉俊， 汪思波. 物理化學(xué)學(xué)報， 2022， 38， 2111021.]doi： 10.3866/PKU.WHXB202111021

（57） Yu， G.; Hu， F.; Cheng， W.; Han， Z.; Liu， C.; Dai， Y. Acta Phys. -Chim. Sin. 2020， 36， 1911016. [郁桂云， 胡豐獻(xiàn)， 程偉偉， 韓字童，劉超， 戴勇. 物理化學(xué)學(xué)報， 2020， 36， 1911016.]doi： 10.3866/PKU.WHXB201911016

（58） Cai， X.; Du， J.; Zhong， G.; Zhang， Y.; Mao， L.; Lou， Z. Acta Phys. -Chim. Sin. 2023， 39， 2302017. [蔡曉燕， 杜家豪， 鐘光明， 張一鳴，毛梁， 婁在祝. 物理化學(xué)學(xué)報， 2023， 39， 2302017.]doi： 10.3866/PKU.WHXB202302017

（59） Cai， M.; Liu， Y.; Wang， C.; Lin， W.; Li， S. Sep. Purif. Technol. 2023，304， 122401. doi： 10.1016/j.seppur.2022.122401

（60） Liu， C.; Xu， Q.; Zhang， Q.; Zhu， Y.; Ji， M.; Tong， Z.; Hou， W.;Zhang， Y.; Xu， J. J. Mater. Sci. 2019， 54， 2458.doi： 10.1007/s10853-018-2990-0

（61） Xie， Y.; Zhang， Q.; Sun， H.; Teng， Z.; Su， C. Acta Phys. -Chim. Sin.2023， 39， 2301001. [謝垚， 張啟濤， 孫宏麗， 滕鎮(zhèn)遠(yuǎn)， 蘇陳良.物理化學(xué)學(xué)報， 2023， 39， 2301001.]doi： 10.3866/PKU.WHXB202301001

（62） Liu， C.; Xiao， W.; Liu， X.; Wang， Q.; Hu， J.; Zhang， S.; Xu， J.;Zhang， Q.; Zou， Z. J. Mater. Sci. Technol. 2023， 161， 123.doi： 10.1016/j.jmst.2023.04.007

（63） Liu， C.; Feng， Y.; Han， Z.; Sun， Y.; Wang， X.; Zhang， Q.; Zou， Z.Chin. J. Catal. 2021， 42， 164. doi： 10.1016/S1872-2067（20）63608-7

（64） Dang， X.; Xie， M.; Dai， F.; Guo， J.; Liu， J.; Lu， X. J. Mater. Chem. A2021， 9， 14888. doi： 10.1039/D1TA02052H

（65） Zhang， X.; Hu， K.; Zhang， X.; Ali， W.; Li， Z.; Qu， Y.; Wang， H.;Zhang， Q.; Jing， L. Appl. Surf. Sci. 2019， 492， 125.doi： 10.1016/j.apsusc.2019.06.189

（66） Wang， J.; Qin， C.; Wang， H.; Chu， M.; Zada， A.; Zhang， X.; Li， J.;Raziq， F.; Qu， Y.; Jing， L. Appl. Catal. B： Environ. 2018， 221， 459.doi： 10.1016/j.apcatb.2017.09.042

（67） Li， L.; Zhang， R.; Lin， Y.; Wang， D.; Xie， T. Chem. Eng. J. 2023，453， 139970. doi： 10.1016/j.cej.2022.139970

（68） 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(xué)學(xué)報， 2022，38， 2109023.] doi： 10.3866/PKU.WHXB202109023

（69） Guo， S.; Li， Y.; Xue， C.; Sun， Y.; Wu， C.; Shao， G.; Zhang， P. Chem.Eng. J. 2021， 419， 129213. doi： 10.1016/j.cej.2021.129213

（70） Cheng， C.; Zhang， J.; Zeng， R.; Xing， F.; Huang， C. Appl. Catal. B：Environ. 2022， 310， 121321. doi： 10.1016/j.apcatb.2022.121321

（71） Zheng， J.; Lei， Z. Appl. Catal. B： Environ.2018， 237， 1.doi： 10.1016/j.apcatb.2018.05.060

國家自然科學(xué)基金（51902282， 12274361）， 江蘇高校青藍(lán)工程，江蘇省自然科學(xué)基金（BK20211361）， 江蘇省高校自然科學(xué)研究項目（20KJA430004）和江蘇省生態(tài)環(huán)境材料重點實驗室開放課題資助項目