Visible-Light-Driven Zr-MOF/BiOBr Heterojunction for the Efficient Synchronous Removal of Hexavalent Chromium and Rhodamine B from Wastewater.

With the rapid industrial development, the coexistence of multiple pollutants in wastewater has become a common phenomenon. Thus, developing highly efficient decontamination methods is imperative. In this work, a string of UiO-66-NH2/BiOBr heterojunctions with varying ratios of BiOBr were prepared and applied to remove hexavalent chromium Cr(VI) and rhodamine B (RhB). The possible growth process of BiOBr nanosheets on UiO-66-NH2, removal activity of contaminants, and photocatalysis mechanism were investigated. When the mass ratio of UiO-66-NH2 to BiOBr reaches 1:0.75, the heterojunction (NB-75) shows optimal photocatalytic activity. After 30 min of adsorption, the total removal rates of Cr(VI) (50 mg/L) and RhB (10 mg/L) over NB-75 (0.25 g/L) reaches 96.7% within 120 min of illumination and 98.9% within 80 min of illumination, respectively. For the removal process, there are two factors. The first is the high adsorption capacity for RhB and Cr(VI) owing to the high porosity of UiO-66-NH2 and interlayer surface positive charge of BiOBr. The second is the improved visible-light photocatalytic performance of the UiO-66-NH2/BiOBr heterojunction via rapid separation of photoinduced carriers. In addition, the active species capture study reveals that the electrons (e-) and the superoxide radicals (•O2 -) play key roles in Cr(VI) reduction, while the holes (h+) are major reactive groups participating in the degradation of RhB. This work demonstrated a kind of promising MOF-based photocatalysis material for eliminating Cr(VI) and RhB simultaneously.

Introduction

With the continuous development of industrialization worldwide, a growing number of various harmful contaminants enter the water ecosystem, including chemicals such as heavy metals and organic compounds.[1,2] Typically, the contaminants hexavalent chromium (Cr(VI)) and Rhodamine B (RhB) produced in leather and dyeing industries have certain genotoxicity and carcinogenicity.[3,4] In many cases, Cr(VI) usually coexists with an organic contaminant in wastewater, and it is of great significance to remove Cr(VI) and organic pollutants simultaneously. Recently, numerous studies about wastewater purification methods were reported, including adsorption,[5,6] membrane separation,[7,8] catalytic degradation,[9,10] and chemical precipitation.[11,12] Among these methods, as an environmentally friendly technology, photocatalysis, especially visible-light catalysis, has gained considerable focus in reducing Cr(VI) to Cr(III) and degrading organic pollutants.[13,14]

Semiconductors including metal oxides,[15] metal phosphides,[16] metal sulfides,[17,18] and carbon-based composites[19,20] are considered as prospective photocatalytic materials. However, the practical applications of single-component photocatalysts still face challenges to achieve efficient Cr(VI) reduction and organic degradation simultaneously due to unmatched energy band, rapid recombination of photoinduced carriers, and small specific surface area problems.[21] Fabricating heterojunction photocatalytic systems is an effective method to improve the photocatalytic performance and gain strong redox capacity simultaneously.[22,23] Meanwhile, because photocatalysis is an interfacial reaction, good adsorption performance of photocatalysts will benefit pollutants’ rapid diffusion to the surface of the catalyst.[24] Hence, it is necessary to explore heterojunction photocatalysts with excellent adsorption-photocatalysis properties.

Metal–organic frameworks (MOFs) have aroused a lot of interest because of their exceptional porosity, high specific surface area, and abundant active functional groups.[25] With the merits of these outstanding properties, MOFs play significant roles in separation,[26] adsorption,[27] catalysis,[28] sensing,[29] etc. Specifically, photoresponsive MOFs have attracted more attention due to the abundant adsorption channels and carrier charge transfer pathways.[30−34] Moreover, owing to the organic ligands and metal clusters, MOFs act as reception aerials to capture light to generate the excited hole–electron pairs, which can effectively promote light utilization.[35,36] In particular, Zr-based MOFs have unique aqueous stability, and UiO-66-NH2 possesses high visible-light photocatalytic ability due to the introduction of amino groups.[37,38]

Although it is thought that UiO-66-NH2 might be a good host matrix to achieve some desirable performance due to its photoresponse characteristic and distinguished structure feature for reactant diffusion and adsorption, the photocatalytic oxidation property of pure UiO-66-NH2 is still not satisfactory, and the photocatalytic performances of the single MOF photocatalyst are generally inhibited by its poor conductivity.[39] Currently, heterojunction composites were designed with the aid of narrow-gap semiconductors to increase the transportation and separation of photongenerated carriers.[40−42] As a kind of semiconductor, bismuth oxyhalides (BiOX, X = halogens) belong to the tetragonal crystal system with layered structures composed of X– and [Bi2O2]2+ layers.[43,44] Different electric densities between layers lead to orbital polarization, which in turn produces a self-built electric field and further promotes the separation of photoinduced carriers.[45] Among these compounds, BiOBr has attracted extensive interest for its visible-light catalytic activity.[46−49] It is worth noting that BiOBr has been used to modify the traditional catalysts such as CoS,[50] Sb2WO6,[51] g-C3/N4,[52] and SnO2,[53] which significantly enhances the visible-light reactivity. Thus, the scholars began to think whether BiOBr could enhance the photocatalysis of the new MOF-based photocatalytic systems as well.

In most of the previous studies, the Zr-MOF was mostly used as a kind of modifier in a small amount to improve the oxidative degradation performance of bismuth oxyhalide materials.[54−57] In our study, UiO-66-NH2 was used as the host matrix and BiOBr was introduced to construct a heterojunction composite system. This not only retains the original excellent specific area property of the MOF but also enhances visible-light photocatalytic oxidation and reduction performance. It is found that UiO-66-NH2/BiOBr heterojunctions in our work can effectively prevent BiOBr flakes from aggregation and demonstrate a more efficient removal ability toward Cr(VI) and RhB. Our study on the photoredox of UiO-66-NH2/BiOBr would promote the development of highly efficient MOF-based photocatalysts.

Materials and Methods

Reagents and Materials

Zirconium chloride (ZrCl4), 2-aminoterephthalic acid (BDC-NH2), acetic acid (HAc), potassium bromide (KBr), bismuth nitrate pentahydrate [Bi(NO3)3·5H2O], kalium dichromicum (K2Cr2O7), rhodamine B (RhB), NaIO3, p-benzoquinone (BQ), tertiary butanol (TBA) and ETDA-2Na were purchased from Macklin Chemical Co., Ltd. Hydrochloric acid (HCl), dimethylformamide (DMF), methanol, ethanol, ethylene glycol (EG), and tertiary butanol (TBA) were provided by Hangzhou Gaojing Fine Chemical Co. Ltd. All chemicals were used without any purification.

Preparation of UiO-66-NH2 Nanoparticles

UiO-66-NH2 was synthesized according to the previous study.[26] ZrCl4 (0.5 g), 0.536 g of BDC-NH2, and 4 mL of HAc were added to 60 mL of DMF. The mixed liquid was sonicated for 15 min and subsequently poured into a high-pressure PTFE-lined reactor and stored at 80 °C for 12 h. The powder was separated by centrifugation and was successively washed with DMF and methanol two times to completely dissolve the unreacted ligand. Finally, the powders were activated at 150 °C for 4 h.

Synthesis of UiO-66-NH2/BiOBr Series

UiO-66-NH2/BiOBr samples were prepared in the presence of UiO-66-NH2, as described in Figure . UiO-66-NH2 (200 mg) and 0.165, 0.330, 0.495, and 0.660 mmol Bi(NO3)3·5H2O samples were dispersed in 40 mL of EG, separately. After 20 min of mechanical stirring, KBr in the same molar amount as Bi(NO3)3·5H2O was added in the suspension and continued to agitate for another 20 min. Then, the suspension was held at 120 °C for 5 h in a Teflon-lined reactor (50 mL). The precipitates were washed three times with ethanol and dried at 80 °C for 6 h. The theoretical formation mass of BiOBr was taken into UiO-66-NH2/BiOBr heterojunctions, and mass ratios of mUiO-66-NH/mBiOBr were 1:0.25, 1:0.50, 1:0.75, and 1:1. The obtained samples were labeled as NB-25, NB-50, NB-75, and NB-100, respectively. Pristine BiOBr was also synthesized through a similar procedure in the absence of MOFs.

Figure 1

Schematic illustration for the preparation process of UiO-66-NH2/BiOBr heterojunctions.

Characterization

The morphologies of the catalysts were determined by scanning electron microscopy (SEM, Zeiss Ultra55) and transmission electron microscopy (TEM, JEOL). The crystal phase patterns of samples were collected by powder X-ray diffraction using Cu Kα (XRD, AXS). X-ray photoelectron spectroscopy was used to characterize the component and chemical states of catalysts using a K-Alpha (XPS, Thermo Fisher Scientific). UV–vis diffuse reflection spectroscopies (DRS, JASCO) were adopted to investigate the light-harvesting properties. Brunauer–Emmett–Teller (BET) surface areas were characterized with a Micromeritics ASAP 2020. Total organic carbon (TOC) was measured with a TOC analyzer (Elementar Vario TOC). The photoelectrochemical properties of the catalysts were tested on an electrochemical workstation (CHI 660B) with a standard three-electrode system, and the Na2SO4 (0.2 mol/L) was used for bath solution. The steady-state photoluminescence spectra were recorded with a fluorescence spectrophotometer (PL, FluoroMax-4) with an excitation wavelength of 350 nm. The electron paramagnetic resonance signals were collected at a spectrometer (EPR, Bruker EMX PLUS) with a radical trapping reagent DMPO in water and methanol solution.

Evaluation of Photocatalytic Measurement

The photocatalytic performances of UiO-66-NH2, BiOBr, and UiO-66-NH2/BiOBr catalysts were investigated under 300 W Xe lamp irradiation (λ > 400 nm). The catalysts were employed for the photocatalytic elimination of the model contaminants, including RhB aqueous solution, Cr(VI) aqueous solution, and their mixed solution. Typically, 25 mg of photocatalysts was added to 100 mL of 10 mg/L RhB aqueous solution (pH = 7) and 100 mL of 50 mg/L K2Cr2O7 solution (pH = 3) separately. In another experiment, 50 mg of photocatalysts was added to 100 mL of mixed solution (pH = 3) consisting of 10 mg/L RhB and 50 mg/L K2Cr2O7. Before illumination, the above experimental systems were agitated for 30 min without light to achieve the absorption equilibrium and then illuminated with a xenon lamp. At regular intervals, a 5 mL aliquot was centrifuged for UV–vis spectroscopy analysis, and the concentration was calculated according to the peak intensity. The involvement of active species hole (h+), electron (e–), superoxideradical (•O2–), and hydroxyl radical (•OH) was investigated by introducing EDTA-2Na (1 mM), NaIO3 (1 mM), BQ (1 mM), and TBA (1 mM) into the RhB and K2Cr2O7 solutions.

Results and Discussion

XRD Analysis

The XRD results of the as-synthesized samples are shown in Figure . The peaks of UiO-66-NH2 at 2θ = 7.4, 8.5, 12.1, 25.7, 30.7, 43.4, and 50.4° match well with the previous reports,[26] indicating the formation of UiO-66-NH2. The peaks of pristine BiOBr at 2θ = 10.8, 25.2, 31.7, 32.2, 46.3, and 57.1° suggest the successful fabrication of BiOBr.[43] By comparison, characteristic diffraction peaks of UiO-66-NH2 and BiOBr also appear in the NB-25, NB-50, NB-75, and NB-100 samples, suggesting that the integration of BiOBr causes little damage to the UiO-66-NH2 crystal structure. It is also found that as the amount of BiOBr increases, the intensities of the UiO-66-NH2 diffraction peaks at 2θ = 7.4 and 8.5° gradually decrease and BiOBr diffraction peaks at 2θ = 10.8 and 25.2° increase, confirming the efficient loading of BiOBr in heterojunctions.

Figure 2

XRD patterns of UiO-66-NH2, BiOBr, and UiO-66-NH2/BiOBr heterojunctions.

Morphological Characterization

Figure shows the SEM and TEM morphologies of the pure UiO-66-NH2, BiOBr, and their composites. UiO-66-NH2 shows a well-crystallized regular octahedron structure with a particle size of about 150 nm (Figure a). In Figure b, the pure BiOBr sample exhibits a flowerlike microsphere structure formed by the self-assembly nanosheets with a thickness of about 80–100 nm. Figure c–e shows the SEM photos of UiO-66-NH2/BiOBr with different ratios (0.25, 0.75, and 1) of BiOBr. Compared with the pure BiOBr sample, the size and thickness of BiOBr nanosheets of all ratios in composites greatly decrease. As for the NB-25 and NB-75 samples, it is found that BiOBr nanosheets uniformly disperse on the UiO-66-NH2 matrix. However, the uneven distribution of BiOBr in UiO-66-NH2/BiOBr appears in Figure e when the ratio of BiOBr increases to 1 (NB-100), which is mostly caused by the self-assembly of BiOBr nanosheets. EDS analysis of UiO-66-NH2/BiOBr reveals the homogeneous element signals of C, O, Bi, Br, and Zr (Figure f), which indicates the successful growth of BiOBr along the surface of the UiO-66-NH2 crystal. In addition, the hierarchical microstructure of NB-75 is further observed by TEM of low magnification and HRTEM of high resolution. The TEM image shows the formation of a tight heterojunction between UiO-66-NH2 and BiOBr. The observed lattice spacing of 0.270 nm in the HRTEM image is consistent with the (1 1 0) crystal plane of BiOBr.

Figure 3

SEM images of (a) UiO-66-NH2, (b) BiOBr, (c) NB-25, (d) NB-75, and (e) NB-100. (f) EDS spectrum of NB-75, and (g–i) TEM images of NB-75.

Chemical State Analysis

X-ray photoelectron spectroscopy (XPS) spectra of the prepared samples were collected to explore the presence of interfacial chemical connections in the composites. Survey spectra of UiO-66-NH2, BiOBr, and NB-75 are shown in Figure a. The presence of C, O, N, Zr, Bi, and Br in NB-75, which is consistent with the analysis of EDS analysis, confirms the successful preparation of the composite catalysts. The Bi 4f and Br 3d spectra of NB-75 exhibit 0.7 and 0.6 eV offset toward the lower binding energy compared with the pure BiOBr (Figure b,c). However, the N 1s spectrum of NB-75 (Figure d) exhibits a higher peak position at 401.3 eV compared to pure UiO-66-NH2, which confirms the presence of the extra electronic binding of N with other atoms increasing the electron density. This probably could be attributed to the fact that the ionized protons in glycol make the initial solution become acidic (pH = 3), and then amino groups in UiO-66-NH2 are protonated to form −NH3+, attracting Br– by ionic electrostatic interactions, which can make BiOBr nucleate at the surface of UiO-66-NH2. In summary, it is plausible to suppose that UiO-66-NH2/BiOBr is not just a physical bonding but also a composite with chemical interfacial interactions.

Figure 4

XPS spectra of samples: (a) survey, (b) Bi 4f, (c) Br 3d, and (d) N 1s.

Specific Surface Areas

N2 adsorption–desorption isotherms and the corresponding porosity parameters of as-prepared samples are shown in Figure and Table S1 to gain a deep insight into specific surface area and porous nature. The BET surface area of the pure BiOBr synthesized is only 17.11 m2/g, and its isotherm manifests a type IV isotherm shape with a hysteresis loop (Figure a), indicating that only a mesoporous structure formed by crystal stacking was observed. Compared with single UiO-66-NH2, although the specific surface area of NB-75 decreases, it maintains a high surface area of 284.34 m2/g and a pore volume of 0.3277 cm3/g and exhibits a composite of types I and IV isotherm shape (Figure b). It is known that the type I isotherm is representative of micropores related to the MOF moiety, providing more active sites for the adsorption and facilitating contact with the pollutants.

Figure 5

BET isotherm and pore-size distribution (insert) of (a) BiOBr and (b) NB-75.

Removal Activity and Stability

The removal abilities of UiO-66-NH2, BiOBr, and UiO-66-NH2/BiOBr samples for Cr(VI) and RhB were investigated, as shown in Figure . C/C0 represents the ratio of instantaneous concentrations (C) to the initial concentrations (C0) of contaminants in aqueous solution. Before illumination, the as-prepared materials adsorbed pollutants under darkness for 30 min to reach the adsorption–desorption equilibrium. As seen in Figure a,c, there is no significant decrease in the concentration of Cr(VI) and RhB under direct photolysis conditions. It is also found that UiO-66-NH2/BiOBr composites demonstrate good removal capabilities for Cr(VI) and RhB. In practice, the photocatalytic reaction was conducted parallelly with the adsorption reaction at the same time. The total removal efficiency is the result of the synergistic effect between adsorption and the photocatalytic process. To further study the adsorption and catalytic properties of as-synthesized samples, the adsorption and photocatalysis removal efficiencies were calculated according to C/C0 in Figure b,d.

Figure 6

Adsorption-photocatalysis performance of various samples (a) Cr(VI) and (b) RhB and the corresponding removal efficiency and k values of (c) Cr(VI) and (d) RhB.

As shown in Figure b, the adsorption efficiencies of the pure UiO-66-NH2 and BiOBr for Cr(VI) were 10.6 and 6.9%, respectively. Then, with the increase of BiOBr, the UiO-66-NH2/BiOBr samples exhibit superior adsorption performance for Cr(VI) in the order NB-25 (14.3%) < NB-50 (17.5%) < NB-75 (18.3%) < NB-100 (23.5%); that is to say, the combination of two materials improves the adsorption performance of Cr(VI). The increased adsorption may be attributed to the presence of interlayer surface positive charge on BiOBr nanosheets and the intercalated Cr2O72– adsorption that occurs between the lattice layers.[58] According to SEM analysis in Section , the smaller size and scattered distribution of BiOBr on UiO-66-NH2 can provide more adsorption sites for Cr2O72–. Then, during the photocatalytic reduction process, the removal efficiency of UiO-66-NH2 can reach 73.7%, and the NB-75 exhibits a higher photocatalytic removal efficiency of 77.9%. Taken together, after 150 min of adsorption–degradation process, the total removal efficiency of NB-75 reaches 92.6%, demonstrating the best removal performance.

Furthermore, as shown in Figure d, in the absence of light, the pure BiOBr can only adsorb 10.9% of RhB, while UiO-66-NH2 can adsorb 35.2% of RhB, exhibiting greater adsorption performance than the former. Meanwhile, with the increasing contents of BiOBr, UiO-66-NH2/BiOBr samples show decreased adsorption capacity for RhB in the order NB-25 (32.18%) > NB-50 (27.2%) > NB-75 (25.3%) > NB-100 (23.3%). Considering the porous structure of the UiO-66-NH2 sample, the adsorption capacity for RhB mainly depends on the high specific surface area and pore volumes, which contribute to providing more adsorption reactive sites and contaminant transport paths. After the subsequent photocatalytic degradation in visible light, the total RhB removal rates of pure UiO-66-NH2 and BiOBr were 56.3 and 75.8%, respectively. Moreover, it can be seen that with increasing weight ratio of BiOBr, the total removal efficiencies of UiO-66-NH2/BiOBr heterojunctions gradually increase to 98.9% for NB-75 but decrease to 84.7% for NB-100, consistent with the result of the Cr(VI) removal process. The trend shows the enhanced synergy between UiO-66-NH2 and BiOBr with an increase of the BiOBr ratio, but when the BiOBr ratio increased to 1, the synergistic effect decreased due to the agglomeration of BiOBr nanosheets.

For better comparison, the photocatalysis reaction kinetics were analyzed according to the pseudo-first-order kinetic model ln(C0/Ct) = kt. The corresponding linear fitting results and reaction rate constant k values are shown in Figures S1 and 6b,d. Apparently, in Figure b,d, NB-75 exhibited the highest k value in Cr(VI) reduction and RhB degradation, which further proves that the NB-75 could effectively improve visible-light photoreduction and photodegradation activity.

Figure shows the UV–visible absorption spectra of Cr2O72– and RhB solution after the adsorption and catalytic processes by NB-75. In Figure a, the peak intensity (354 nm) of Cr2O72– decreased significantly, indicating the elimination of the Cr(VI). Figure b shows that after the adsorption and catalytic processes, the peak intensity (553 nm) of RhB decreased significantly and the peak position blue-shifted, indicating the decomposition of RhB molecules. Total organic carbon (TOC) is adopted to assess the organic pollutant removal performance of NB-75 toward RhB degradation. The initial TOC was 28.19 mg/L, and then the TOC reduced to 5.62 mg/L after 80 min of illumination. The high TOC removal efficiency of 80.06% revealed the significant removal ability of NB-75. Meanwhile, being compared with the photocatalysts reported in the references, the NB-75 in our study exhibited competitive performance toward Cr(VI) and RhB removal (Table S2).

Figure 7

Time-dependent absorption-photocatalysis UV–visible spectra of (a) Cr(VI) and (b) RhB over NB-75.

Experiments to explore the effect of solution pH on the photocatalytic activity of the NB-75 sample were carried out. Figure S2 shows that a lower solution pH is conducive to the reduction of Cr(VI) and degradation of RhB. This could be attributed to the fact that a lower solution pH was able to accelerate the reduction of Cr(VI) as proceeded in the following equation: Cr2O72– + 14H+ + 6e– → 2Cr3+ + 7H2O. RhB is a kind of a weak basic dye, which is easy to ionize and thus interact with photocatalysts in a weak acid solution.[59]

Based on the above research results, we formulated a mixed solution (pH = 3) containing 50 mg/L Cr(VI) and 10 mg/L RhB as the target pollutants and studied the adsorption and photocatalytic performance of NB-75. Figure a shows the ultraviolet absorption spectra of the mixed solution. As we can see, after 30 min of adsorption in the absence of illumination, the characteristic ultraviolet absorption peaks of Cr(VI) and RhB at 354 and 553 nm are obviously weakened, and then the intensity of those characteristic peaks continued to decrease after the subsequent 90 min of photocatalytic process. This suggests that the as-prepared NB-75 sample is effective in the purification of the wastewater containing Cr(VI) and RhB by adsorption and photocatalysis. Furthermore, the reusability of heterojunctions for the mixed solution was tested. As illustrated in Figure b, their first-time total removal efficiencies were 85.48 and 80.32% for Cr(VI) and RhB, respectively. After five cycles, the corresponding removal efficiencies maintained at 74.38% and 69.56%, respectively. In addition, the XRD patterns and XPS spectra of NB-75 before and after use were characterized to further explore the photocatalytic stability. As shown in Figure S3, the crystallinity and surface compositions were almost unchanged throughout five cycles, which implies that the NB-75 sample possesses good reusability during the photocatalytic process.

Figure 8

(a) Ultraviolet absorption spectra of Cr(VI) and RhB in the mixed solution and (b) the cycling removal experiments for Cr(VI) and RhB in the mixed solution by NB-75.

Possible Mechanism on Enhanced Visible-Light Photocatalytic Activity

Figure a shows the ultraviolet–visible diffuse reflectance spectroscopies (DRS) of samples. The absorption threshold values are around 450 nm for UiO-66-NH2 and 425 nm for BiOBr, and the absorption edge of UiO-66-NH2/BiOBr occurs between pure UiO-66-NH2 and BiOBr. Furthermore, band gap energies were calculated according to the Kubelka–Munk equation αhv = A(hv – Eg), where α, hν, and Eg stand for the absorption coefficient, photon energy, and the band gap, respectively. The n value depends on the optical transition type of the semiconductor. Herein, UiO-66-NH2 is a direct semiconductor (n = 1), whereas the n value of BiOBr is 4 for indirect transition.[60] According to Figure b, the band gap energies of pure UiO-66-NH2 and BiOBr were 2.82 and 2.69 eV. Meanwhile, band gaps of NB-25 (2.80 eV), NB-50 (2.78 eV), NB-75 (2.75 eV), and NB-100 (2.74 eV) suggest that after the modification by BiOBr, UiO-66-NH2/BiOBr heterojunctions have a narrower band gap than pure UiO-66-NH2 and exhibit higher utilization efficiency for visible light.

Figure 9

(a) UV–vis diffuse reflection spectra; (b) (αhν)2 vs hν curves of different samples; (c) transient photocurrent response under irradiation of visible light with 20 s light on/off cycles; and (d) EIS Nyquist plots.

Then, photoelectric chemical characters of UiO-66-NH2, BiOBr, and NB-75 are characterized by transient photocurrent curves and electrochemical impedance spectroscopy (EIS). In Figure c, a stable photocurrent response was observed for each light-on and light-off on the samples, and the photocurrent intensity of NB-75 was higher than those of the other two pure samples. As a rule, the separation efficiency of photogenerated carriers is positively correlated with photocurrent density. EIS spectra, as shown in Figure d, exhibited a smaller radius of NB-75 compared with UiO-66-NH2 and BiOBr, implying faster interfacial charge migration and more effective separation of photogenerated carriers.[61] In addition, the photogenerated carrier separation performance was evaluated by photoluminescence (PL) spectra. As seen in Figure S4, the NB-75 exhibited a weaker PL intensity compared with pure UiO-66-NH2, indicating that the formation of heterojunctions accelerated photoinduced electron–hole pair separation.[62] Therefore, the above results account for the good photocatalytic capability of UiO-66-NH2 equipped with BiOBr.

Trapping experiments were conducted to explore the active species in the photocatalytic reaction of UiO-66-NH2/BiOBr heterojunctions. EDTA-2Na, tert-butanol (TBA), NaIO3, and p-benzoquinone (BQ) were utilized to trap holes (h+), hydroxyl radicals (•OH), electrons (e–), and superoxideradical (•O2–), respectively.[62,63] For Cr(VI) reduction (Figure a), NaIO3 and BQ could effectively inhibit the removal rate, which implies e– and •O2– play crucial roles in the Cr(VI) removal process. However, the photoreduction performance of Cr(VI) is enhanced with the addition of EDTA-2Na. This could be due to holes being captured by EDTA-2Na to facilitate the separation of photogenerated electron–hole pairs. Figure b shows that the degradation efficiency of RhB significantly decreased with the addition of EDTA-2Na, illustrating that h+ is the dominant active species. Meanwhile, the EPR technique was adopted to explore the main active species generated by NB-75. As shown in Figure c,d, strong DMPO–•O2– signals and negligible DMPO–•OH signals were observed over NB-75 under visible-light irradiation. This indicates that the dissolved oxygen in water easily receives e– on the heterojunction to generate •O2–, but the h+ is difficult to oxidize water molecules or hydroxides to produce •OH. This is consistent with the results of radical quenching experiments.[53] Moreover, compared with pure UiO-66-NH2 and BiOBr (Figure S5), a stronger DMPO–•O2– signal intensity for NB-75 was found, indicating the efficient separation of photogenerated carriers.

Figure 10

Trapping experiments of the different scavengers on (a) Cr(VI) reduction (pH = 3) and (b) RhB degradation (pH = 7) over NB-75; EPR spectra of NB-75 under visible-light irradiation: (c) DMPO–•O2– and (d) DMPO–•OH.

The Mott–Schottky curves are shown in Figure S6 to study the band structure of UiO-66-NH2 and BiOBr. The positive slopes of UiO-66-NH2 and BiOBr show that they are all n-type semiconductors, and the flat band potential is at −0.74 and −1.20 V vs Ag/AgCl (−0.53 and −0.99 V vs NHE), respectively.[64] In general, the conduction band (CB) potential is more negative by about 0.1 V than the flat band potential for n-type semiconductors.[65] Therefore, the CB potential of UiO-66-NH2 and BiOBr was calculated to be −0.63 and −1.09 eV. In addition, the valence band (VB) of UiO-66-NH2 (2.18 eV) and BiOBr (1.60 eV) was calculated via the equation: Eg = EVB – ECB.

Based on the above analysis, Figure shows a reasonable photocatalytic mechanism of the NB-75 heterojunctions. Since the EVB of BiOBr and UiO-66-NH2 is less positive than H2O/•OH (2.40 V vs NHE), it is difficult for NB-75 to oxidize H2O to •OH.[66] However, the RhB molecules can be directly oxidized by holes because the RhB/RhB+ (1.43 V vs NHE) is lower than UiO-66-NH2 and BiOBr.[67] Moreover, the electrons are more vulnerable to being caught by O2 to create •O2– due to the potential of UiO-66-NH2/BiOBr being more negative than that of O2/•O2– (−0.33 V vs NHE).[68] The above analyses are consistent with the results of active species capture and EPR experiments shown in Figure . When visible-light irradiation occurs, hole–electron pairs are excited on the UiO-66-NH2/BiOBr heterojunction. Then, the electrons of BiOBr flow into the CB of UiO-66-NH2 to reduce Cr(VI) to Cr(III). Meanwhile, the holes of UiO-66-NH2 can be transferred to the VB of BiOBr to oxidize RhB. This inhibits the recombination of photogenerated charge pairs and promotes their migration across the surface of the coupled nanocomposite photocatalysts, which can increase the possibility of electron–hole pairs participating in surface redox reactions. More importantly, the intrinsic porosity, large specific area surface area of UiO-66-NH2, and the close coupling with BiOBr nanoflakes can make more catalytic activity sites in the heterostructures, which are favorable for water purification.

Figure 11

Possible mechanism for Cr(VI) reduction and RhB degradation over the UiO-66-NH2/BiOBr heterojunctions under visible-light irradiation.

Degradation Pathway of RhB

In addition, the possible intermediates during RhB degradation were identified by MS/HPLC-MS spectroscopy (Figures S7 and S8). Based on the analysis of spectra, the possible RhB degradation route is proposed in Figure . RhB molecules (m/z = 443) are first degraded to produce N-demethylated intermediates (m/z = 331.27). Afterward, the intermediates are further oxidized and three benzoic intermediates including amino phthalic acid (m/z = 182.19), hydroxybenzoic acid (m/z =138.24), and dihydroxy-benzene (m/z = 110.00) are identified. Finally, ring-opening reactions occur and result in the formation of small-molecule acids with m/z values of 100.12, 90.93, 88.04, 74.00, and 61.88. These acids will be eventually decomposed into CO2 and H2O.[46]

Figure 12

Proposed photocatalytic degradation pathway of RhB.

Conclusions

In our study, UiO-66-NH2/BiOBr heterostructured photocatalysts have been prepared through a two-step solvent thermal synthesis method, in which BiOBr nanosheets combine with UiO-66-NH2 by chemical interactions. The heterojunction shows outstanding removal properties for simultaneous Cr(VI) reduction and RhB oxidation by the synergy of adsorption and visible-light photocatalysis. The formed structure facilitates the synergistic transfer of photogenerated electron-holes in UiO-66-NH2/BiOBr and ensures their lifetime to improve the photocatalytic activity. The as-prepared NB-75 photocatalyst has remarkable simultaneous Cr(VI) reduction and RhB degradation performance and is superior to the pure UiO-66-NH2 under visible-light irradiation. The study can provide a valuable strategy for the development of an efficient MOF-based visible-light photocatalyst for the removal of heavy metals and organics in wastewater.