Decoration of Defective Sites in Metal–Organic Frameworks to Construct Tight Heterojunction Photocatalyst for Hydrogen Production

Zr–metal–organic frameworks (MOFs) have received much interest for their ultrahigh stability and are considered an up‐and‐coming class of catalysts for photocatalytic water splitting. However, their activity still needs to be improved. In this work, a series of defective UiO‐66‐NH2‐x@CdS nanoparticles (NPs) (x = 0, 50, 100, 150, 200, denotes the molar equivalent of the defect modulator) heterostructure photocatalysts is constructed for water splitting using defect engineering followed by a postmodification strategy. Defective structures are introduced to improve the photocatalytic activity of heterojunctions in the following ways: 1) modulating the energy band structure of UiO‐66‐NH2‐x, 2) providing chelate binding sites for modifying the bridging molecules and thus building strong interactions between UiO‐66‐NH2‐x and CdS NPs, and 3) serving as trapping sites to separate the photogenerated electron–hole pairs. Hence, this constructed series of UiO‐66‐NH2‐xCdS NPs exhibit ultrahigh photocatalytic water splitting for H2 production, especially UiO‐66‐NH2‐150@CdS NPs with moderately defective levels showing catalytic activity up to 2303 μmol g−1 h−1, which is 2.36 times higher than pure CdS NPs. Furthermore, the heterojunction catalysts with different defect levels exhibit a volcano‐type trend, demonstrating the feasibility of defect engineering. This work provides novel insights for developing advanced defect‐based MOF‐constructed composite photocatalysts.


Introduction
With the overexploitation and use of nonrenewable fossil energy sources, the increasingly severe energy crisis and environmental safety issues have received widespread attention.Exploring green and friendly new energy technologies has become an attractive way for humans to achieve sustainable development. [1,2][5][6] Therein, solar energy catalysis is a crucial technology to enable photovoltaic, photothermal, and photochemical conversion for stable energy, [7][8][9] among which photocatalysis should be praised as a power technology for energy conversion.Since titanium dioxide was reported to be a kind of ideal photocatalyst for water splitting, [10] different photocatalysts such as metal oxides, sulfides, and carbon nitrides have been successfully developed.In recent years, metalorganic frameworks (MOFs) materials have become the "star" materials in photocatalysis.13] However, MOFs still suffer from many drawbacks as ideal photocatalysts, such as limited light absorption range, severe charge carriers recombination, and poor stability. [14,15]Hence, many modification strategies based on MOFs, such as defect engineering (DE), crystal engineering, ligand engineering, guest engineering, and complex/derivative approaches, have received much attention in recent years. [16]E is one of the most effective strategies to modulate photoresponsiveness, [17] electronic structure, [18,19] active site, [20][21][22][23] Zr-metal-organic frameworks (MOFs) have received much interest for their ultrahigh stability and are considered an up-and-coming class of catalysts for photocatalytic water splitting.However, their activity still needs to be improved.In this work, a series of defective UiO-66-NH 2 -x@CdS nanoparticles (NPs) (x = 0, 50, 100, 150, 200, denotes the molar equivalent of the defect modulator) heterostructure photocatalysts is constructed for water splitting using defect engineering followed by a postmodification strategy.Defective structures are introduced to improve the photocatalytic activity of heterojunctions in the following ways: 1) modulating the energy band structure of UiO-66-NH 2 -x, 2) providing chelate binding sites for modifying the bridging molecules and thus building strong interactions between UiO-66-NH 2 -x and CdS NPs, and 3) serving as trapping sites to separate the photogenerated electron-hole pairs.Hence, this constructed series of UiO-66-NH 2 -xCdS NPs exhibit ultrahigh photocatalytic water splitting for H 2 production, especially UiO-66-NH 2 -150@CdS NPs with moderately defective levels showing catalytic activity up to 2303 μmol g À1 h À1 , which is 2.36 times higher than pure CdS NPs.Furthermore, the heterojunction catalysts with different defect levels exhibit a volcano-type trend, demonstrating the feasibility of defect engineering.This work provides novel insights for developing advanced defect-based MOF-constructed composite photocatalysts.specific surface area, [24] and charge carrier transport [25] of MOFs materials.[28][29] Using acetic acid (AcOH, AA) as the defect modulator, Jiang's group reported a series of Zr-UiO-66-NH 2 -x (x = 0, 50, 100, 150, 200, denoting the molar equivalents of AA) with different defect levels.The band and electronic structures of Zr-UiO-66-NH 2 can be tuned by modulating the defect levels.With moderate defects, the Pt NPs@Zr-UiO-66-NH 2 -100 displayed the best performance for photocatalytic H 2 production. [30]Then, with formic acid (HCOOH, FA) as the defect modulator, Sun's group reported Zr-UiO-66-NH 2 -ML/ MC-x (x = 0, 50, 100, 150, 200, denoting molar equivalents of FA) with miss ligands (ML) and clusters (MC) for photocatalytic CO 2 reduction. [31]The MC defects amplify the light absorption energy (E abs ), and the ML defects reduce the charge transfer energy (E LMCT ), as demonstrated by density functional theory (DFT) calculations.With the increase of ML, the decrease of E LMCT contributes to the rapid separation of photogenerated carriers, while MC defects tend to have a negative impact.With a medium degree of defects, UiO-66-NH 2 -ML-100 exhibited the best visible light CO 2 reduction performance.The volcanic-type catalytic performance trends exhibited by all the DE MOFs involved in the above cases indicate that defect chemistry has an optimal value in MOFs.
Besides, the defect chemistry not only enriches the accessible active sites, but also serve as a modification site to achieve further remodeling of MOFs.The design and application of open metal sites (OMS) or ligand-unsaturated sites (CUS), or open coordination sites (OCS), is an important part of MOF functionalization. [23]Demel and his colleagues obtained enhanced photosensitivity using a postsynthetic modification (PSM) strategy by incorporating a diphenylphosphinic acid ligand onto the Zr-oxo cluster of PCN-222. [32]In the work of Duan et al., the stereoselective L/D-pyrrolidin-2-ylimidazole (PYI) was integrated into the Zn-oxo cluster of Zn-MOF, and the terminal N atom of PYI was tightly linked to the metal center by coordination interactions, thus obtaining a novel tandem photocatalyst with a chiral structure. [33]Consequently, the rational design and modification of metal-centered clusters concerning DE MOFs is an impressive way to develop and enhance their photocatalytic performance.So far, we are not aware of any related works that directly utilize defective MOFs as good templates for preparing heterojunction catalysts.
To sum up, we successfully obtained a series of defective UiO-66-NH 2 -x (x = 0, 50, 100, 150, 200, x denotes the molar equivalent of AA) using acetic acid as the defect modulators.Then, with unsaturated metal centers as modification sites for -NH 2 , the -SH of DL-cysteine was grafted on the OMS of MOFs, thus providing anchoring sites to integrate with CdS.As we know, defects in MOFs have shown the advantages of modulating electronic structure, providing modification sites and inhibiting electron-hole pairs recombination.DL-cysteine acts as a 'bridge' for promoting photogenerated charge transfer from CdS to DE MOFs, breaking the interfacial binding barriers and thus establishing a heterojunction photocatalyst with a tightly connected surface and optimal energy band matching.The photocatalytic tests show that UiO-66-NH 2 -150 can be a better template, allowing UiO-66-NH 2 -150@CdS nanoparticles (NPs) to exhibit the highest photocatalytic performance for H 2 production (%2303 μmol g À1 h À1 ), which is 2.36 times higher than pure CdS NPs, surpassing many reported MOF-based photocatalysts.A series of characterizations demonstrate that the tight contact between CdS NPs and UiO-66-NH 2 -150 through strong coordination interactions forms a fast charge transfer channel.Thus, the feasibility of DE MOFs as excellent substrates for loading active species with stronger coordination interactions is investigated for the first time, promising to break through the interfacial binding barriers between different materials and construct heterojunction photocatalysts with optimal bandgap matching.
Next, the powder X-Ray diffraction (PXRD) test was performed to identify the purity and stability of the samples.As depicted in Figure 1a, the power patterns of the UiO-66-NH 2 -x are in good overlap with the simulated patterns, which proves their successful preparation.As shown in Figure 1b, all UiO-66-NH 2 -x@CdS NPs exhibit similar PXRD features, and the CdS NPs-related characteristic diffraction peaks appear at plausible positions, which proves that the anchoring of CdS NPs does not affect the framework structure of UiO-66-NH 2 -x and maintains good crystallinity after being integrated.Scanning electron microscopy (SEM) shows that the UiO-66-NH 2 -x series maintains good crystallinity and displays classical octahedral morphology, and the obtained crystal size grows with the increasing amount of AA (Figure 1c and S1, Supporting Information).The transmission electron microscopy (TEM) and SEM image of UiO-66-NH 2 -150@CdS NPs show the distribution of CdS NPs on the surface of UiO-66-NH 2 -150 (Figure 1d,h and S2, Supporting Information).Nonuniform distribution of CdS NPs was attributed to the defective distribution of MOFs, uneven grafting of DL-cysteine, and high initial concentration of Cd II , which reacted rapidly upon encountering the S source to form CdS and agglomerate.The enlarged high-resolution transmission electron microscopy (HRTEM) image shows the presence of distinctive CdS NPs' characteristic lattice stripes (d (200) = 0.281 nm and d (222) = 0.163 nm) on the surface of UiO-66-NH 2 -x, which proves the successful loading of CdS NPs (Figure 1f,g).To further confirm the successful preparation of the UiO-66-NH 2 -150@CdS NPs, the distribution of elements is investigated by SEM-energydispersive spectrometer (EDS) mapping (Figure S3, Supporting Information).Five elements (Zr, Cd, S, N, O, and C) are simultaneously present in the mapping diagram, which indicates that highly dispersed CdS NPs are assembled on the surface of UiO-66-NH 2 -150 as expected.
Thermogravimetric analysis (TGA) test is performed to identify the thermal stability of the samples.The thermal  decomposition curve of UiO-66-NH 2 -150@CdS NPs can be divided into three stages: first, when the temperature rose to the range of 55-120 °C, the DMF and H 2 O could be removed; second, when the temperature went up to 340 °C, AA and TAA were completely decomposed; finally, when the temperature reached 600 °C, the NH 2 -BDC and DL-cysteine decomposition led to the collapse of the skeletal structure, and the residues should be ZrO 2 and Cd 2 O 3 (Figure 2a). [24,31,34]Based on their thermal decomposition results, the loading amounts of CdS NPs in UiO-66-NH 2 -x@CdS NPs are estimated to be about 32.86 wt%, which is close to the theoretical content of 37.48 wt%. [35]The Brunauer-Emmett-Teller (BET) surface area of UiO-66-NH 2 -x gradually increased with increasing defects in their framework, in which the BET surface area of the perfect UiO-66-NH 2 -0 was estimated to be 572.6 m 2 g À1 , and UiO-66-NH 2 -150 with the maximum defects was increased to 756.5 m 2 g À1 (Figure 2b and S8, Supporting Information).After CdS NPs loading, the BET surface area of UiO-66-NH 2 -150 decreased from 756.5 to 107.6 m 2 g À1 , and the corresponding pore volume decreased from 0.45 to 0.12 cm 2 g À1 (Table S1, Supporting Information).Such a significant BET drop demonstrates the successful modification of DL-cysteine on the UiO-66-NH 2 -x backbone, where CdS NPs can be anchored inside the framework.
The Fourier-transform infrared spectroscopy (FTIR) is an effective tool for analyzing the presence of functional groups in MOFs.UiO-66-NH 2 -x with different defect levels are analyzed using FTIR (Figure S4, Supporting Information).The peak at 3500-3000 cm À1 is attributed to the stretching vibration of N─H bond, and its gradually decreasing peak area verifies the gradually increasing level of ligand defects.The peak at 1670 cm À1 is attributed to the C═O bond stretching vibration.The weak peak at 1621 cm À1 is attributed to the bending vibration of -NH 2 , and the sharp peak at 1580 cm À1 corresponds to the stretching vibration of the benzene ring and carbonyl group in NH 2 -BDC. [36,37]Additionally, the sharp peaks at 1260 and 1370 cm À1 are attributed to the C─N bond. [38]In addition, as the defective level in MOFs gradually increased, there is a significant intensity quenching of the individual feature peaks in FTIR, likes 3000-3500 cm À1 (Figure 2c), demonstrating that different levels of defects are successfully introduced into the framework.Notably, the corresponding C─N bond's vibrational characteristic peak at 1370 cm À1 is redshifted with increasing defect level, which can be attributed to the occupation of the defect sites by AA.The decoration of AA will lead to the inability of the ─COOH of NH 2 -BDC to coordinate with the Zr-oxo cluster, thus dispersing the charge on the C─N bond. [39]The characteristic peaks at 1480 and 485 cm À1 are attributed to C-S and Zr•••N, respectively, which proved the successful coordination of DL-cysteine.Also, Raman results are available at 485 cm À1 .As shown in Figure 2d, the shift of the D and G peaks of UiO-66-NH 2 -150@CdS NPs toward the lower wave number reflects the strong interaction between the Zr-oxo cluster and DL-cysteine, which will cause a change in the collaborative environment around the metal center. [40]Besides, the intense peak in the low-wavelength number band (%485 cm À1 ) can be attributed to the presence of CdS NPs, confirming the successful compounding of CdS NPs and UiO-66-NH 2 -x. [41]This indicates that DL-cysteine realizes coordination interactions with unsaturated Zr-oxo clusters via -NH 2 and not -COOH.
To further identify the successful modification of DL-cysteine onto the unsaturated Zr-oxo cluster, the chemical states of the elements were investigated by high-resolution X-Ray photoelectron spectroscopy (XPS).The positions of the individual characteristic peaks in the survey spectrum reasonably correspond to the elements of Zr, C, O, N, Cd, and S in CdS NPs, UiO-66-NH 2 -150, and UiO-66-NH 2 -150@CdS NPs (Figure S7a, Supporting Information).In the high-resolution spectrum of S 2p, the characteristic peaks at 161.47 eV (161.5 eV) and 162.57eV (161.9 eV) are assigned to Cd-S (Figure 3a).The appearance of C-S (161.90 and 163.20 eV) peaks in UiO-66-NH 2 -150@CdS NPs proves the presence of DL-cysteine on UiO-66-NH 2 -150.Compared to pure CdS NPs, the shift of the Cd-S peak from 161.47 eV (162.67 eV) to 161.50 eV (162.60 eV) proves that the successful modification of DL-cysteine will change the charge density of S 2À . [42]ompared to UiO-66-NH 2 -150, the electron cloud density on Zr IV will be increased after being decorated with DL-cysteine.Therefore, Zr 3d 3/2 and Zr 3d 5/2 in UiO-66-NH 2 -150@CdS NPs display a distinct downward shift from 183.11 and 185.49 to 183.06 and 185.44 eV (Figure 3b), respectively.Depending on the electronegativity of the metal atoms (Zr 1.4 < Cd 1.66 (Pauling's Scale [43] )), although individual Cd II is more negative than Zr IV , the overall electronegativity of Zr 6 -O cluster is much higher than Cd II .Hence, the charge migration pathways from Cd II to the Zr-oxo cluster caused a decrease in charge density (metal-to-cluster charge transfer, MCCT) and an upshift of the Cd 3d 3/2 and Cd 3d 5/2 characteristic peaks from 405.17 and 411.94 to 405.51 and 412.23 eV occurs (Figure 3c). [35,44]ompared to CdS NPs, the XPS results again show that the -SH of DL-cysteine is involved in the nucleation process of CdS NPs, which means that CdS NPs and UiO-66-NH 2 -150 are not simply in physical contact, but in a much tighter strong coordination interaction.In addition, a characteristic peak at 399.30 eV in the high-resolution of N1s was ascribed to the Zr•••N bond, demonstrating the successful coordination of the DL-cysteine's -NH 2 (Figure S7b, Supporting Information).
All materials' UV/vis diffuse reflectance spectra (UV/vis DRS) are given to obtain the optical response characteristics and their bandgap (E g ).UiO-66-NH 2 -x series shows similar curves in the range of 200-800 nm, while the UiO-66-NH 2 -x@CdS NPs series obtains an extension of the light absorption region >430 nm due to the decoration of CdS NPs (Figure 4a and S5a, Supporting Information).According to the equation of (αhv) 1/n = A(hv-E g ), where α is the absorbance index, h is Planck's constant, v is the frequency, A is a constant, E g is the forbidden bandwidth of the semiconductor, and n is directly related to the semiconductor type (i.e., direct-type semiconductor (p, n = 1/2) or indirect-type semiconductor (n, n = 2)), the E g of all materials can be obtained based on the UV/vis DRS results.As shown in Figure 4b and S5b (Supporting Information), the E g of the UiO-66-NH 2 -x series is %3.0 eV and will not be changed by the introduction of defects. [30]he energy band structure is essential to explain the carrier's migration pathway and catalytic mechanism.The positive slopes of Mott-Schottky plots at different frequencies (500 and 1000 Hz) reveal the n-type semiconductor properties of all samples (Figure 4c,d, and S6, Supporting Information).The conduction band minimum (CBM) potential is close to the flat-band potential.Thus, the conduction band (CB) and the valence band (VB) of CdS NPs are estimated to be À0.9 and 1.64 V, respectively.Additionally, the lowest unoccupied molecular orbital (LUMO) of UiO-66-NH 2 -x (x = 0, 50, 100, 150, 200) series were calculated to be À0.37,À0.38, À0.39, À0.48, and À0.47 V versus the normalized hydrogen electrode (NHE), and the highest occupied molecular orbitals (HOMOs) of them were 2.53, 2.57, 2.55, 2.45, and 2.41 V versus NHE, respectively (Figure 4e).With the increasing defects (from 0%150), the HOMO of UiO-66-NH 2 will be upshifted.Thus, the bandgap between the HOMO of UiO-66-NH 2 and the CdS NPs of CB will be reduced, which is beneficial for charge migration between two materials.The energy band structure of all samples shows that CB potentials are more negative than the redox potential of H 2 O (0.0 V vs. NHE), which implies the possibility of the photocatalysts for water splitting. [41]ext, we explore the performance of all materials for photocatalytic H 2 production under simulated sunlight irradiation (>420 nm) using lactic acid and 10.0 mM H 2 PtCl 6 •6H 2 O as sacrificial agent and cocatalyst, respectively.UiO-66-NH 2 -x series is observed to produce no H 2 under visible light irradiation (Figure 5a).The pure CdS NPs exhibit H 2 production rate of 978 μmol g À1 h À1 , but UiO-66-NH 2 -0@CdS NPs with a perfect structure exhibit a lower yield (669 μmol g À1 h À1 ) than pure CdS NPs.This can be illustrated that the framework integrity limited the formation of strong coordinated bond between Zroxo cluster and CdS NPs.However, the photocatalytic H 2 production rates of UiO-66-NH 2 -50@CdS NPs, UiO-66-NH 2 -100@CdS NPs, UiO-66-NH 2 -150@CdS NPs, and UiO-66-NH 2 -200@CdS NPs can reach up to 1270, 2099, 2303, and 1839 μmol g À1 h À1 , respectively, showing a volcanic trend (Figure 5b).45][46][47][48][49][50][51][52] We speculate that the reduced photocatalytic efficiency of UiO-66-NH 2 -200@CdS NPs is attributed to the following reasons: 1) the excessive defective sites can be served as new recombination centers to reduce the electron-hole pair separation efficiency and 2) the excessive defects can also lower the stability of the MOFs materials.The enhanced photocatalytic performance can be ascribed to the decoration of DL-cysteine, which can be used as a precision nucleation site for Cd II , thus constructing tight heterojunctions between UiO-66-NH 2 -150 and CdS NPs.
Photoelectrochemical measurements are performed to probe the separation and transfer behavior of the photogenerated carriers between CdS NPs and UiO-66-NH 2 -x.Linear sweep voltammetry (LSV) records the variation of current with electrode potentials.When the potential is swept from 0.0 to À1.0 V versus NHE, the cathodic current increased around the onset potential (E on, HER ), where CdS NPs < UiO-66-NH 2 -0@CdS NPs < UiO-66-NH 2 -50@CdS NPs < UiO-66-NH 2 -100@CdS NPs < UiO-66-NH 2 -200@CdS NPs < UiO-66-NH 2 -150@CdS NPs (Figure 6a), which indicates that UiO-66-NH 2 -150@CdS NPs has the lowest electrocatalytic H 2 production potential on the ITO electrode. [46]econd, the transient photocurrent tests reflect the separation of the photogenerated electron-hole pairs.When the light is on, the anode current rapidly reaches a maximum value and then tends to a stable level (I s ), which indicates that the carrier separated rapidly and tended to equilibrate (Figure 6b).When the light is off, the photogenerated electron-hole pairs are inhibited causing the anode current to drop rapidly to almost zero.Notably, the current density of the UiO-66-NH 2 -150@CdS NPs is about 4.1 and 2.7 times higher than that of pure CdS NPs and UiO-66-NH 2 -0@CdS NPs, which can be attributed to the effective electron-hole pair migration and separation.The electrochemical impedance spectroscopy (EIS) results can also prove that DL-cysteine can serve as molecule linker to facilitate the photogenerated electron-hole pair separation.As depicted in Figure 6c, the interfacial charge transfer resistance of UiO-66-NH 2 -150@CdS NPs is lower than others, which indicates that DL-cysteine can establish strong interactions between UiO-66-NH 2 -150 and CdS NPs, thus making the electron-hole pair separation more efficient between two materials.The carrier's recombination information is also provided by the photoluminescence (PL) spectra.In comparison with UiO-66-NH 2 -150@CdS NPs, CdS NPs and UiO-66-NH 2 -150 show a strong PL strength, indicating a severe recombination of carriers (Figure S9, Supporting Information).Furthermore, the fitting of timeresolved PL spectra determines the PL lifetimes be 2650 AE 5 ps (CdS NPs), 663 AE 5 ps (UiO-66-NH 2 -150), and 450 AE 5 ps (UiO-66-NH 2 -150@CdS NPs), respectively (Figure 6d), which means that UiO-66-NH 2 -150@CdS NPs have more effective electronhole separation.Therefore, combining the LSV, photocurrent, EIS, and PL results, the important role of DL-cysteine for facilitating the photogenerated carriers' separation can be proved.
An open-circuit potential (OCP) experiment was conducted to mimic the photocatalytic reaction.Figure 6e shows the time profiles of OCP for CdS NPs, UiO-66-NH 2 -150, and UiO-66-NH 2 -150@CdS NPs, measured in 0.2 M Na 2 SO 4 .In the dark, each electrode had an OCP (V 0 ).When light was on, OCP curves immediately dropped down to a steady value (V hv ).The former is due to photogeneration electrons on the electrode.The latter is due to equilibrium being reached between the electron generation and consumption.Then the difference between V 0 and V hv corresponds to the neat number of electrons (e hv ) remaining on the electrode.Among the electrodes, the ehv value is CdS NPs≫ UiO-66-NH 2 -150@CdS NPs ≥ UiO-66-NH 2 -150.This means that CdS NPs transfer its electrons to UiO-66-NH 2 -150 for the hydrogen evolution reaction (HER), while UiO-66-NH 2 -150 transfers its holes to CdS NPs for consuming electron donors.As a result, UiO-66-NH 2 -150@CdS NPs has an improved efficiency of carriers' separation, an increased rate of surface reaction, and hence a smaller e hv , as compared to CdS NPs and UiO-66-NH 2 -150.In addition, the average photogenerated carrier lifetime (τ n ) proves that CdS NPs (%1.72 s) and UiO-66-NH 2 -150 (%46.67 s) show the shortest and longest photogenerated electron lifetimes, respectively (Figure 6f ).UiO-66-NH 2 -150@CdS (%9.51 s) suffered from strong coordination interactions, leading to a much lower photogenerated electron lifetime, indicating a lower carriers' recombination rate.Overall, we can speculate on the mechanism of photocatalytic water splitting for H 2 production on UiO-66-NH 2 -150@CdS NPs (Scheme 2).Under the visible light irradiation, CdS NPs possesses excellent visible-light harvesting ability to generate a mount of photogeneration electrons and transfer to its CB.S3 (Supporting Information).
After that, the electrons transfer directions can be divided into the LUMO of UiO-66-NH 2 -150 and further transfer to Pt cocatalyst to promote H2 production.On the other hand, due to the suitable band structure and close contact between two materials, the holes migrate from the HOMO of UiO-66-NH 2 -150 to the VB of CdS NPs to consume electron donors (lactic acid) and inhibit photogenerated electron-hole pairs' recombination.Hence, the photocatalytic 2 production was cycled by continuously consuming electrons and holes.

Conclusion
In summary, we successfully constructed the UiO-66-NH 2 -150@CdS NP heterojunction photocatalyst by DE and PSM.The unsaturated metal sites can modulate the electronic structure and form carrier recombination centers of UiO-66-NH 2 -x (x = 0, 50, 100, 150, 200) and act as effective modification sites.DL-cysteine is essential in tight heterojunctions, establishing new charge transfer channels and working as a molecular linker to develop strong interactions between the two materials.Compared with CdS NPs and UiO-66-NH 2 -0@CdS NPs, its charge separation and transfer ability can be significantly enhanced, thus showing higher photocatalytic activity for H 2 evolution.This unique approach may provide a new way to construct MOFs-based composite photocatalysts with solid interactions and improved photocatalytic activity.
Preparation of UiO-66-NH 2 -x@CdS NPs: Before CdS NPs anchoring, the obtained UiO-66-NH 2 -x (x = 0, 50, 100, 150, 200) was placed in a vacuum oven at 120 °C for 12 h to completely release the pore channels and the polar groups (e.g., ÀOH) or water molecules coordinated to the defect