Covalently bonded interfaces with delocalized π electrons in a MOF-in-MOF heterojunction for efficient gas–solid phase CO₂ photoreduction

Abstract

The interfaces in heterostructure photocatalysts play an important role in charge migration, but the rational controllable interfacial construction remains a challenge. Herein, a covalently bonded interface with delocalized π bonds was controllably constructed via the encapsulation of UiO-66-NH₂ nanoparticles in MUV-10 skeletons. The UiO-66-NH₂/MUV-10 heterojunction enabled both efficient capture and conversion of CO₂ and H₂O. The delocalized π bond in the covalent interface efficiently accelerated interfacial charge migration, suppressed carrier recombination, and reduced the work function for electronic effusion. As a consequence, UiO-66-NH₂/MUV-10 exhibited superior CO₂ photoreduction activity with H₂O under visible light irradiation in the gas–solid phase. The CO evolution rate reached as high as 38.7 μmol g⁻¹ h⁻¹, which surpassed that of most commonly reported gas–solid phase photocatalysts. This work not only provides a feasible strategy for the controllable construction of interfaces but also demonstrates the great potential of interfacial engineering for efficient photocatalysis and other related areas.

---

Wenyu Yuan

Dr Wenyu Yuan is an associate professor at the School of Chemistry & Chemical Engineering in Shaanxi Normal University. He obtained his PhD degree from Northwestern Polytechnical University in 2019. After graduation, he worked as a postdoctoral researcher in the National Institute for Materials Science (NIMS) (Japan, 2019–2020) and Tokyo Metropolitan University (Japan, 2020–2021). His research interests are in two-dimensional heterostructures and metal–organic frameworks for energy conversion applications.

Introduction

Excessive CO₂ emission into the atmosphere leads to a series of serious environmental problems, such as the greenhouse effect and ocean acidification.^1–5 Therefore, the development of efficient CO₂ capture and conversion technologies (e.g., thermal catalysis, electrocatalysis, and photocatalysis) to address the energy and environmental crises has attracted wide attention.^6–15 Among them, converting CO₂ to renewable fuel via photocatalysis is considered an environmentally friendly and efficient technique.^16–18 Common CO₂ photoreduction reactions can be achieved in liquid–solid or gas–solid modes.¹⁹ Due to the low CO₂ adsorption capacity of the photocatalysts and poor CO₂ dissolution, the efficiency and practicability of the liquid–solid phase CO₂ photoreduction system is limited although the liquid solvent can partly restrict the self-decomposition of photocatalysts.²⁰ Gas–solid phase photoreduction with high reusability, efficient CO₂ capture ability, superior selectivity, and suppressed H₂ generation stands out. However, the efficiency for gas–solid CO₂ photoreduction is still unsatisfactory due to the following issues: (1) high energy for photo-induced carrier excitation, (2) challenge in carrier separation and transfer, and (3) poor gas capture and catalytic conversion ability.^21–26

To achieve efficient CO₂ photoreduction in gas–solid phase, tremendous efforts have been made to address the above challenges.^27–29 Among the many methods, heterojunctions are considered to be an extremely effective means to enhance the activity.³⁰ For instance, by gradually inserting Li ions into the TiO₂ lattice from the outside, Li et al. constructed a plaque-like Li₂TiO₃/TiO₂ heterostructure to effectively inhibit carrier recombination and improve the CO₂ adsorption capacity.³¹ Bi et al. proposed a novel COF/MOF heterojunction photocatalyst using an olefin (C=C)-linked covalent organic framework (TTCOF) and NH₂-UiO-66 as building units, which effectively inhibited carrier recombination and enhanced the reduction of CO₂ to CO.³²

Nevertheless, electron migration at the interface of heterojunction plays an important role in the development of high-performance heterojunction photocatalysts. The design of strong interface interactions and the overall improvement of their dynamic processes play an important role in promoting CO₂ photoreduction, while weak interface interactions will greatly limit the efficiency of electron migration and thus limit its activity. For instance, Wang et al. successfully constructed covalent bonds between CN and CQD in CN-CQD which facilitated efficient carrier migration and CO₂ activation. However, the electronegativity of different atoms in covalent bonds leads to electron localization, which affects electron transport.³³ Delocalized π bonds provide efficient electron transport channels and have been widely shown to be an effective way to facilitate charge transport (e.g., graphene).^34,35 Therefore, the construction of strong covalent interfaces containing delocalized π bonds is of great significance but challenging for photocatalysis.

Constructing amide bonds (–CO–NH–) at the interfaces is promising for overcoming the above challenge.^33,36 The π electron on the –CO–NH–carbonyl group and the lone pair electron on the nitrogen atom in the adjacent carbon–nitrogen bond together to form a three-centre four-electron delocalized π bond, which can effectively improve the interfacial charge transfer efficiency (Scheme 1). Herein, MUV-10 and UiO-66-NH₂ containing –COOH and –NH₂ functional groups were chosen to construct the desired heterojunction with strong covalent bonding. A MOF-in-MOF heterojunction, in which UiO-66-NH₂ nanoparticles were encapsulated in MUV-10, was successfully synthesized via a PVP-assisted in situ growth method, and an amide bonded heterogeneous interface was successfully constructed. The results showed that UiO-66-NH₂/MUV-10 exhibited superior gas–solid phase CO₂ photoreduction activity with a CO evolution rate of 38.7 μmol g⁻¹ h⁻¹ under visible light irradiation, which surpassed those of the most commonly reported gas–solid phase photocatalysts. Further analysis demonstrated that UiO-66-NH₂/MUV-10 with a Z-scheme junction exhibits superior gas capture ability toward both CO₂ and H₂O vapor, reduced work function, suppressed carrier recombination, and accelerated charge transfer efficiency. This study underscores the importance of interfacial bonding design in photocatalysis.

Scheme 1 Schematic for the interfacial covalent bonding design for efficient CO₂ photoreduction in the gas–solid phase.

Experimental section

Materials

Zirconium(IV) chloride (ZrCl₄), titanium isopropoxide (Ti(O_iPr)₄) and 1,3,5-benzoic acid (H₃BTC) were purchased from Aladdin Bio-Chem Technology Co. Ltd (Shanghai, China). MnCl₂·4H₂O, N,N-dimethylformamide (DMF), acetic acid (CH₃COOH) and methanol (CH₃OH) were purchased from Sinopharm Chemical Reagent Co. Ltd (Shanghai, China). 2-Amino-1,4-benzenedicarboxylic acid (NH₂-BDC) was purchased from Bide Pharmatech Co. Ltd (Shanghai, China). Polyvinylpyrrolidone (PVP) was purchased from Sahn Chemical Technology Co. Ltd (Shanghai, China). All chemicals were used directly without further purification.

Synthesis of UiO-66-NH₂

UiO-66-NH₂ was synthesized using the method reported in the literature:³⁷ ZrCl₄ (1.5 g), NH₂-BDC (1.07 g), CH₃COOH (44 mL), and deionized water (7.5 mL) were dissolved in DMF (100 mL) and then reacted at 120 °C for 15 min under stirring in an oil bath. The as-synthesized UiO-66-NH₂ was washed with DMF and methanol. Finally, the resultant yellow solid was dried overnight at 70 °C under vacuum.

Synthesis of PVP-modified UiO-66-NH₂

PVP-modified UiO-66-NH₂ was synthesized using the method reported in the literature.³⁸ The as-synthesized UiO-66-NH₂ nanoparticles were dispersed in water (10 mL) by stirring. An aqueous solution of PVP (2.5%, M_w = 55 000, 10 mL) was added into the suspension, and the mixture was stirred at room temperature for 24 hours. The PVP-modified UiO-66-NH₂ nanoparticles were collected by centrifugation at 9000 rpm for 3 minutes, washed three times with DMF, and finally redispersed in DMF.

Synthesis of 20-UiO-66-NH₂/MUV-10

MUV-10 was synthesized using the method reported in the literature:³⁹ H₃BTC (125.0 mg, 595 μmol) and MnCl₂·4H₂O (23.8 mg, 120 μmol) were dissolved in 12 mL mixture of DMF dispersed with 20 mg PVP-modified UiO-66-NH₂ and 3.5 mL of CH₃COOH in a 25 mL glass bottle. Subsequently, 36 μL of Ti(O_iPr)₄ (120 μmol) was added to the above solution. The bottle was sealed and ultrasonicated for 30 min for uniform mixing and then heated in an oven at 120 °C for 48 h. After naturally cooling down to room temperature, microcrystalline powder was obtained by centrifugation and washed with fresh DMF and CH₃OH.

Synthesis of x-UiO-66-NH₂/MUV-10

The synthesis of UiO-66-NH₂/MUV-10 with different contents of UiO-66-NH₂ (30-UiO-66-NH₂/MUV-10 and 40-UiO-66-NH₂/MUV-10) was done using the above procedure, but with 30 or 40 mg PVP-modified UiO-66-NH₂ instead of 20 mg PVP-modified UiO-66-NH₂.

Sample activation

The synthesized UiO-66-NH₂, MUV-10, and UiO-66-NH₂/MUV-10 samples were soaked in CH₃OH for four days with solvent exchange twice a day and then vacuumed at 150 °C for 10 h.

Characterization

Powder X-ray diffraction (PXRD) data of the samples were collected over the 2θ range of 5–50° on a MiniFlex 600 X-ray diffractometer using Cu Kα radiation at room temperature, and the diffractometer was operated at 40 kV and 15 mA. The size and morphology were investigated by field emission scanning electron microscopy (SU8220), transmission electron microscopy (TEM, Tecnai G2 F20, FEI, America), and high-angle annular dark-field scanning TEM (HAADF-STEM, S1806722, Thermo Fisher Scientific, The Netherlands) coupled with energy-dispersive X-ray (EDX) spectroscopy (Titan Cubed Themis G2 300, Thermo Fisher Scientific, The Netherlands). FT-IR measurements were performed on a TENSOR II spectrometer (Bruker Co. Ltd, Germany) with KBr pellets. The chemical composition and state of samples were investigated using X-ray photoelectron spectroscopy (XPS, AXIS ULTRA, Japan). The binding energy was calibrated using the C 1s peak (284.6 eV). UV-vis diffuse reflectance spectroscopy was performed on a UV-vis-near-infrared (UV-vis-NIR) spectrophotometer (Lambda 1050, PerkinElmer). The gas sorption isotherms of CO₂ at 298 K were measured by using an ASAP 2020 analyser. The gas sorption isotherms of H₂O were measured by using a JW-ZQ200C micropore and vapor adsorption instrument. All gases used in the adsorption experiment were of 99.99% purity. Prior to gas adsorption/desorption measurements and catalytic experiments, all the samples were activated. Ultraviolet photoelectron spectroscopy (UPS) spectra were collected on an ESCALAB Xi + instrument. The fluorescence spectra were recorded on a HORIBA JY-FluoroMax-4 fluorescence spectrometer. TRPL spectra were measured by using an FLS1000 fluorescence lifetime test system. Electron spin resonance (ESR) spectra were collected on a Bruker EMXplus-6/1 instrument, Germany. The ¹³C isotope labeling experiment using ¹³CO₂ was conducted and analysed by GC-mass spectrometry (Agilent 7890 GC system and Agilent 5975B GC/MSD) to unequivocally confirm the origin of the photocatalytic products.

Photocatalytic CO₂ reduction measurements

Before the photocatalytic reaction, all samples were activated to retain the original pore structure and avoid the side effect of small solvent molecules on the results. The CO₂ photoreduction properties were evaluated using a quartz photocatalytic reaction system. First, 10 mg of catalyst was placed on a quartz sand sheet and supported in a 180 mL quartz photocatalytic reactor containing a mixture of 30 mL of H₂O and 5 mL of TEOA. The reactor was vacuumed for 1 min after sealing, and then CO₂ gas was passed through the reactor for 15 min until the mixture was CO₂ saturated. A 300 W xenon lamp with a λ > 400 nm filter was used as the irradiation light source. The gas products were analysed by gas chromatography (GC), and the liquid products were analysed by ¹H nuclear magnetic resonance (¹H-NMR) spectroscopy. The incident light intensity was determined using a PL-MW2000 optical power meter. The AQE was assessed under identical reaction conditions.

Photoelectrochemical measurements

Photoelectrochemical experiments were performed in a three-electrode system with 0.2 M Na₂SO₄ as the electrolyte. For the preparation of the working electrode, 20 mg as-synthesized photocatalyst was dissolved in 10 μL Nafion solution and 1 mL methanol. Then, 200 μL of the aforementioned suspension was deposited on the FTO and dried at room temperature. An Ag/AgCl electrode and a platinum electrode were used as the reference electrode and the counter electrode, respectively. All electrochemical characterization tests were performed on a CHI 760E electrochemical workstation.

In situ FT-IR spectroscopy

In situ diffuse reflection infrared testing was conducted in an infrared spectrometer (Bruker INVENIO S) at room temperature. The sample powder was added to the high temperature diffuse reflection reaction tank, with KBr as the background, and then CO₂ gas was introduced under irradiation of a deuterium lamp, and the data were collected at different time points.

Results and discussion

Design, construction, and characterization

MUV-10 and UiO-66-NH₂, containing –COOH and –NH₂ functional groups, were selected to construct the desired heterojunction with strong covalent bonding. Both MUV-10 and UiO-66-NH₂ are semiconductors with moderate band gaps, enabling visible light harvesting and an enhanced built-in electric field for charge separation. Their uniform 3D microporous structures and high specific surface areas make them ideal gas adsorbents (Fig. S1†). More importantly, a –CO–NH– bonded heterogeneous interface can be formed via the condensation reaction between –NH₂ and –COOH groups. As discussed above, the delocalized π bond in the –CO–NH–carbonyl group effectively improves the interfacial charge transfer efficiency. Therefore, the heterostructure UiO-66-NH₂/MUV-10 not only possesses high gas uptake and wide-wavelength light absorption but also achieves suppressed charge recombination and fast charge migration, thereby leading to efficient CO₂ photoreduction activity.

Fig. 1a illustrates the strategy for synthesizing the UiO-66-NH₂/MUV-10 heterostructure. UiO-66-NH₂ nanoparticles were synthesized via a solvothermal reaction of ZrCl₄ and NH₂-BDC in a DMF/HAc/H₂O mixed solvent. The resulting UiO-66-NH₂ nanoparticles were then stirred in a PVP solution to obtain PVP-modified UiO-66-NH₂ nanoparticles. These PVP-modified nanoparticles were subsequently added to a mixture containing Ti(O_iPr)₄, H₃BTC, MnCl₂·4H₂O, DMF, and CH₃COOH. After the solvothermal reaction, the target product UiO-66-NH₂/MUV-10 was obtained. The PVP assisted in situ growth method can ensure the effective encapsulation of UiO-66-NH₂ in MUV-10, successfully constructing a heterogeneous interface with –CO–NH– bonding. A series of UiO-66-NH₂/MUV-10 heterojunction catalysts were obtained by adjusting the amounts of UiO-66-NH₂ nanoparticles (denoted as x-UiO-66-NH₂/MUV-10, where x represents the amount of UiO-66-NH₂/MUV-10).

Fig. 1 Materials synthesis and characterization. (a) The preparation process of the UiO-66-NH₂/MUV-10 heterojunction via the PVP-assisted in situ growth method. (b) XRD patterns of UiO-66-NH₂, MUV-10 and UiO-66-NH₂/MUV-10. SEM images of (c) UiO-66-NH₂, (d) MUV-10 and (e) UiO-66-NH₂/MUV-10. (f–h) TEM images of UiO-66-NH₂/MUV-10. (i) HAADF-STEM image and EDX mapping of elemental N, Mn, Ti, Zr, C, and O of UiO-66-NH₂/MUV-10 and their overlap image.

The crystalline structures of the synthesized samples were characterized using X-ray diffraction (XRD). As depicted in Fig. 1b, the diffraction peaks of the UiO-66-NH₂/MUV-10 heterojunction are in accordance with those of UiO-66-NH₂ and MUV-10, indicating that the structures of UiO-66-NH₂ and MUV-10 remain largely unchanged during heterojunction formation. The XRD patterns also suggest the successful encapsulation of UiO-66-NH₂ nanoparticles in MUV-10 skeletons (Fig. 1b and S2†). The morphology and microstructure of the samples were characterized using high-resolution field emission scanning electron microscopy (SEM), transmission electron microscopy (TEM), and high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM). SEM images (Fig. 1c, S3a and b†) show that the UiO-66-NH₂ nanoparticles are octahedral with a diameter of approximately 50 nm. MUV-10 is a regular octahedral crystal with a particle size of ∼1 μm (Fig. 1d, S3c and d†). After encapsulation, the morphology has been well preserved. A series of (20, 30, and 40)-UiO-66-NH₂/MUV-10 heterojunctions were successfully synthesized via the PVP-assisted in situ growth method, as shown in Fig. 1e and S4.† With increasing amounts of UiO-66-NH₂ nanoparticles, obvious agglomeration of UiO-66-NH₂ nanoparticles can be observed. TEM images (Fig. 1f–h) reveal that UiO-66-NH₂/MUV-10 forms an octahedron with a uniform size of ∼500 nm at low magnification. Under high magnification, UiO-66-NH₂ nanoparticles are encapsulated in MUV-10, indicating that this strategy successfully forms MOF-in-MOF heterojunctions. The HAADF-STEM image (Fig. 1i) shows uniform encapsulation of UiO-66-NH₂ in MUV-10 in the heterojunction catalyst. Clearly, the approximately octahedral UiO-66-NH₂ nanoparticles are well confined to both the surface and interior of MUV-10 in the UiO-66-NH₂/MUV-10 heterojunction (Fig. S5 and S6†). The corresponding energy dispersive X-ray (EDX) maps further confirm the nearly uniform distribution of N, Mn, Ti, Zr, C, and O elements and their overlap in a typical region (Fig. 1i).

Fourier transform infrared (FT-IR) spectroscopy was applied to analyze chemical bonds and functional groups in the sample (Fig. S7†). For MUV-10, a strong C=O vibration peak was observed at 1680 cm⁻¹, indicating that MUV-10 is rich in carboxyl groups. For UiO-66-NH₂, a strong absorption peak at 3000–3300 cm⁻¹ in UiO-66-NH₂ can be assigned to N–H vibration, indicating that UiO-66-NH₂ is rich in amino groups. 20-UiO-66-NH₂/MUV-10, 30-UiO-66-NH₂/MUV-10 and 40-UiO-66-NH₂/MUV-10 exhibit similar FT-IR spectra. The peaks at 1200–1600 and 1680 cm⁻¹ are assigned to the vibration of C–N and C=O, respectively, which may be caused by the amide bond formed at the hetero-interfaces.

UV-vis absorption spectra were recorded to detect the optical properties of these samples. As shown in Fig. 2a and S8,† all samples exhibit distinct light absorption in the region of 200–600 nm. Compared with the pure MUV-10, the light absorption window of UiO-66-NH₂/MUV-10 is significantly broadened due to the narrow band gap of UiO-66-NH₂. The light absorption wavelength is slightly red shifted. The corresponding band gaps were further calculated using the following equation: αhγ² = A(hγ − E_g), where α, h, γ, A, and E_g denote the absorption coefficient, Planck's constant, light frequency, fitting parameter, and bandgap, respectively. The plots of (αhγ)² against the photonic energy are shown in Fig. 2a and S9.† Pure MUV-10 exhibited the largest bandgap (2.98 eV), while UiO-66-NH₂ exhibited the smallest bandgap (2.85 eV). Three UiO-66-NH₂/MUV-10 samples showed similar band gap values (approximately 2.90 eV).

Fig. 2 Light harvest and gas adsorption performance. (a) UV-vis absorption spectra and plots of (αhγ)² against the photonic energy of MUV-10, UiO-66-NH₂ and UiO-66-NH₂/MUV-10. (b) N₂ adsorption–desorption isotherms at 77 K. (c) CO₂ and (d) H₂O gas adsorption–desorption isotherms at 298 K for MUV-10, UiO-66-NH₂ and UiO-66-NH₂/MUV-10.

The pore structures of the synthesized MUV-10 and UiO-66-NH₂/MUV-10 samples were further confirmed by N₂ adsorption–desorption isotherms at 77 K. MUV-10 exhibited a typical feature of microporous materials (Fig. 2b), and the calculated pore size distribution (PSD) further verified its microporous structure. Moreover, Brunauer–Emmett–Teller (BET) SSA was calculated to be 802.8 and 1225.8 m² g⁻¹ for MUV-10 and UiO-66-NH₂, while the SSA value for UiO-66-NH₂/MUV-10 was as high as 1092.6 m² g⁻¹, indicating that UiO-66-NH₂/MUV-10 showed a higher SSA than pure MUV-10. We further compared the total pore volume (P/P₀ = 0.99). The total pore volumes for MUV-10, UiO-66-NH_2, and UiO-66-NH₂/MUV-10 were 306.4, 667.9 and 390.9 cm³ g⁻¹, indicating that the embedding of UiO-66-NH₂ in MUV-10 does not greatly affect the pore structure. Fig. S10† shows the porous structures of MUV-10, UiO-66-NH₂ and UiO-66-NH₂/MUV-10, wherein the pores with diameters of 4.8, 4.5 and 4.4/5.1 Å could be clearly observed. Compared with pure MUV-10 and UiO-66-NH₂, the aperture of UiO-66-NH₂/MUV-10 increases.

Gas uptake ability, one of the key factors affecting the photocatalytic CO₂ reduction process, was further tested. The CO₂ adsorption capacity of UiO-66-NH₂, MUV-10, and UiO-66-NH₂/MUV-10 at 298 K was 75.8, 45.0, and 66.4 cm³ g⁻¹, respectively (Fig. 2c). Obviously, UiO-66-NH₂ showed the highest CO₂ absorption capacity among these samples, while UiO-66-NH₂/MUV-10 exhibited comparable CO₂ uptake ability. The adsorption properties of H₂O molecules directly affect the CO₂ photoreduction activity in a gas–solid system of CO₂ and water vapor. Through the water vapor adsorption performance test of UiO-66-NH₂, MUV-10 and UiO-66-NH₂/MUV-10 at 298 K, it was found that these samples have excellent water vapor adsorption performance, and the adsorption capacity reached up to 465.9, 403.1, and 368.0 cm³ g⁻¹, respectively (Fig. 2d). The PXRD spectra of the adsorbed samples (Fig. S11†) showed that the whole skeletons were well preserved during the gas adsorption process. The results showed that the pore volume increase of the heterojunction material is beneficial to enhancing the activity of the CO₂ reduction process.

X-ray photoelectron spectroscopy (XPS) was performed to study the chemical state of the elements in UiO-66-NH₂/MUV-10. All XPS spectra were calibrated to C 1s with a binding energy of 284.6 eV (Fig. S12†). Two peaks in the C 1s XPS spectrum at 285.9 and 288.6 eV are attributed to C–O and O–C=O bonds, respectively, indicating the presence of carboxyl groups (Fig. S12†).⁴⁰ Three peaks at 530.0, 531.5, and 532.8 eV in the high resolution O 1s XPS spectrum are assigned to C–O, O–H and M–O bonds, respectively (Fig. 3a). Notably, as the UiO-66-NH₂ content increases from 20-UiO-66-NH₂/MUV-10 to 40-UiO-66-NH₂/MUV-10, the intensity of O–H bonds decreases, which may be caused by the condensation reaction between the –NH₂ in UiO-66-NH₂ and the –COOH in MUV-10. In the N 1s XPS spectra (Fig. 3b), the peaks at 399.9 eV and 401.0 eV are assigned to (C)₃–N and N–H bonds, respectively. As the amount of UiO-66-NH₂ increases, no obvious intensity increase for the N–H bond can be observed, implying that the –NH₂ is involved in the condensation reaction, which suggests the generation of the –CO–NH bond. A plot of O–H/N–H against various amounts of UiO-66-NH₂ is shown in Fig. 3c. O–H/N–H decreases with the increasing amount of UiO-66-NH₂ in the UiO-66-NH₂/MUV-10 formulation, which is consistent with the breakage of –OH and the formation of –NH when the condensation reaction proceeds. All these XPS spectra suggest the successful construction of the interfacial amide bonds in UiO-66-NH₂/MUV-10 heterostructures. Two peaks of Zr 3d5/2 and Zr 3d3/2 at 182.9 eV and 185.2 eV can be observed in the Zr 3d XPS spectra (Fig. 3d).^36,41 As the amount of UiO-66-NH₂ increases, the peak intensity of Zr 3d5/2 and Zr 3d3/2 also increases, reflecting the growing presence of UiO-66-NH₂. In the XPS Mn 2p spectrum (Fig. 3d), one peak was observed at 653.1 eV, assigned to Mn 2p1/2. The Mn 2p3/2 spectrum was separated into two peaks at 641.9 and 643.1 eV, which were assigned to Mn³⁺ 2p3/2 and Mn⁴⁺ 2p3/2, respectively.⁴² The Ti 2p XPS spectrum showed typical peaks for Ti 2p3/2 and Ti 2p1/2, suggesting the oxidation state of Ti(IV) (Fig. 3d),⁴³ which was in accordance with the crystal structure of MUV-10. Both UiO-66-NH₂ and NUV-10 demonstrate weak ESR intensity. The signal at g = 2.01 for the UiO-66-NH₂/MUV-10 heterojunction can be clearly observed, which can be attributed to the presence of the π-conjugated electron system in amide bonds (Fig. S13†). The highly enhanced ESR spectra suggest that the heterostructure demonstrates a tight interaction due to the formation of interfacial amide bonds.

Fig. 3 XPS spectra of (a) O 1s and (b) N 1s. (c) Schematic of the interfacial covalent bonding formation for the breakage of –OH and the formation of –NH. (d) XPS spectra of Mn 2p, Ti 2p and Zr 3d in 20-UiO-66-NH₂/MUV-10, 30-UiO-66-NH₂/MUV-10, and 40-UiO-66-NH₂/MUV-10.

Photocatalytic, photoelectrochemical, and optical properties

The photocatalytic CO₂RR activity of these samples was evaluated under visible light irradiation in a gas–solid system with water. The gas products of photocatalysis were quantitatively identified via gas chromatography. A representative gas chromatogram is shown in Fig. S14, and the ESI† describes the relevant calculation procedures for CO and CH₄ evolution rates. ¹H nuclear magnetic resonance (¹H-NMR) spectra are shown in Fig. S15,† and no liquid products have been detected. As shown in Fig. 4a, after 3 h of continuous visible light irradiation, UiO-66-NH₂/MUV-10 demonstrates superior photocatalytic activity, achieving high CO and CH₄ yields of 116.1 and 3.4 μmol g⁻¹, respectively (Fig. S16†). Its performance significantly surpasses those of UiO-66-NH₂ (CO: 67.2 μmol g⁻¹) and MUV-10 (CO: 30.6 μmol g⁻¹), indicating that heterojunctions with strong covalent bonding can enhance the photocatalytic activity. A further study of CO evolution of various amounts of UiO-66-NH₂ in UiO-66-NH₂/MUV-10 showed that 30-UiO-66-NH₂/MUV-10 exhibited the highest activity among all samples (Fig. 4b). 30-UiO-66-NH₂/MUV-10 achieved the maximal CO and CH₄ yields, which are almost 1.5 and 1.7 times higher than those of 20-UiO-66-NH₂/MUV-10 and 40-UiO-66-NH₂/MUV-10, respectively (Fig. S17†). Excess UiO-66-NH₂ may lead to inevitable agglomeration, leading to reduced activity. Fig. 4c clearly shows that UiO-66-NH₂/MUV-10 exhibited the highest CO production rate of 38.7 μmol g⁻¹ h⁻¹, which surpasses those of UiO-66-NH₂ (22.4 μmol g⁻¹ h⁻¹), MUV-10 (10.2 μmol g⁻¹ h⁻¹) and the mechanical mixture of UiO-66-NH₂ and MUV-10 (Fig. S18†). This result further demonstrates the indispensable role of interface charge transfer regulation. The AQEs of 30-UiO-66-NH₂/MUV-10 at different wavelengths were measured to be 0.006% (380 nm), 0.005% (420 nm), and 0.001% (500 nm) (Fig. S19 and Table S1†). ¹³C isotope labeling experiments were carried out to confirm the origination of CO products. The signal at m/z = 29 is attributed to ¹³CO, confirming that the CO product originates from CO₂ photoreduction, rather than from decomposition or other impurities. No signal at m/z = 44 for CO₂ can be observed, implying that the structure of MOFs has been well preserved.

Fig. 4 Photocatalytic CO₂ reduction performance. (a) Photocatalytic CO evolution over time for MUV-10, UiO-66-NH₂ and UiO-66-NH₂/MUV-10 with a λ > 400 nm filter. (b) Photocatalytic CO evolution over time for 20-UiO-66-NH₂/MUV-10, 30-UiO-66-NH₂/MUV-10 and 40-UiO-66-NH₂/MUV-10 with a λ > 400 nm filter. (c) Mass-normalized CO₂RR rates of MUV-10, UiO-66-NH₂, 20-UiO-66-NH₂/MUV-10, 30-UiO-66-NH₂/MUV-10 and 40-UiO-66-NH₂/MUV-10 with a λ > 400 nm filter. (d) Mass spectrometry analysis of ¹³CO₂ products by UiO-66-NH₂/MUV-10 photoreduction. (e) Cycling stability of UiO-66-NH₂/MUV-10. (f) Summary of the photocatalytic activity of our photocatalyst along with the recently reported photocatalysts in direct gas–solid-phase photocatalytic conversion of CO₂.

In addition, the photocatalytic stability of UiO-66-NH₂/MUV-10 was examined using a recycling test (Fig. 4e and S20†). During 6 cycles of photocatalytic reaction, the CO₂RR activity of UiO-66-NH₂/MUV-10 can be well preserved. The SEM image in Fig. S21a† suggests that the structure of the UiO-66-NH₂/MUV-10 heterojunction basically remains intact, confirming its high stability. XRD and XPS were utilized to characterize the UiO-66-NH₂/MUV-10 sample after the photocatalytic reaction (Figs. S21b and S22†). Compared with UiO-66-NH₂/MUV-10 before the reaction, no obvious changes were observed in the sample after CO₂ photoreduction, verifying its excellent photostability. The enhanced intensity of the M–O bond may be caused by the SiO₂ impurities from the quartz substrate. Its superior stability originated from the high thermal stability of MUV-10 and UiO-66-NH₂, while the gas–solid photocatalytic system enabled excellent recycling features.⁴⁴ Overall, the obtained UiO-66-NH₂/MUV-10 demonstrated superior CO₂ photoreduction activity due to the covalently bonded interfaces with delocalized π electrons. Table S2† shows that UiO-66-NH₂/MUV-10 outperforms most reported excellent photocatalysts in the gas–solid phase system, including TiO₂, C₃N₄, PCN-601, and 2D Mn-MOFs (Fig. 4f).

The photogenerated current as an important indicator of photoelectric response was assessed using a three-electrode system. The photocurrent of UiO-66-NH₂/MUV-10 reached as high as ∼0.13 mA cm⁻² (Fig. 5a), which is 10-fold and 5.2-fold higher than that of MUV-10 and UiO-66-NH₂, respectively, suggesting the positive effect of the delocalized π bond between the UiO-66-NH₂ and MUV-10 interface on photoelectron excitation. Among all heterojunction samples, 30-UiO-66-NH₂/MUV-10 demonstrated the highest photocurrent, which is consistent with the CO₂ photoreduction results (Fig. S23†). Electrochemical impedance spectroscopy (EIS) spectra and the Nyquist plots show that UiO-66-NH₂/MUV-10 has the lowest charge transfer resistance, indicating the enhanced charge transfer efficiency (Fig. S24, S25 and Table S3†). The suppressed charge-carrier recombination was then probed using the steady-state photoluminescence (PL) technique. As shown in Fig. 5b, the PL intensity is drastically quenched compared with those of UiO-66-NH₂ and MUV-10, indicating that the recombination of photoexcited charge carriers has been efficiently suppressed (Fig. S26†). Time resolved photoluminescence spectroscopy (TRPL) was further applied to study the charge carrier dynamics and the lifetime of photoinduced carriers. As shown in Fig. 5c, the average fluorescence lifetimes of UiO-66-NH₂, MUV-10, 20-UiO-66-NH₂/MUV-10, 30-UiO-66-NH₂/MUV-10, and 40-UiO-66-NH₂/MUV-10 were 4.71, 9.33, 4.16, 3.95, and 5.17 ns, respectively. 30-UiO-66-NH₂/MUV-10 had the shortest survival time, suggesting the efficient separation of photogenerated carriers (Fig. S27†). The above results indicate that the Z-scheme UiO-66-NH₂/MUV-10 heterojunction is capable of inhibiting charge-carrier recombination and facilitating charge transfer due to the delocalized π bond formed at the interfaces.

Fig. 5 Photoelectrochemical and optical measurements and the proposed photocatalytic mechanism. (a) Transient photocurrent density curves, (b) PL spectra, and (c) TRPL spectra of MUV-10, UiO-66-NH₂ and UiO-66-NH₂/MUV-10. (d) Time-dependent in situ FT-IR spectra of MUV-10, UiO-66-NH₂ and UiO-66-NH₂/MUV-10 under photoirradiation in a CO₂ atmosphere. (e) The enlarged in situ FT-IR spectra of MUV-10, UiO-66-NH₂ and UiO-66-NH₂/MUV-10 after photoreduction in a pure CO₂ atmosphere without water for 20 min. (f) UPS spectra of MUV-10, UiO-66-NH₂ and UiO-66-NH₂/MUV-10. (g) Band structures of MUV-10, UiO-66-NH₂ and the charge transfer pathway of the heterojunction under light irradiation.

Mechanism study

In situ FT-IR spectroscopy was employed to explore the pathways of CO₂ adsorption and activation on UiO-66-NH₂, MUV-10 and UiO-66-NH₂/MUV-10 (Fig. 5d). In situ FT-IR spectra certified that several active *CO₂ intermediates were formed on the surface of UiO-66-NH₂/MUV-10 during the CO₂ adsorption process under visible light irradiation.⁴⁰ The peaks at 1284–1365 cm⁻¹ belonged to the b-CO₃²⁻ intermediate, and the peaks at 1424–1390 cm⁻¹ belonged to the vibration of HCO₃⁻. In addition, the absorption peaks at 1724 and 1557 cm⁻¹ arose from the *COOH intermediate, while the peak at 1178 cm⁻¹ belonged to CO (C–O), which imply the formation of CO and the adsorption of CO₂. In particular, the peaks at 1870, 1796, and 1734 cm⁻¹ belonged to the transition state CO species (*C=O and C=O), implying the efficient activation and conversion of the adsorbed CO₂ into CO. Notably, surface-bonded OH⁻ from adsorbed H₂O (∼3412 cm⁻¹) can be clearly detected in UiO-66-NH₂/MUV-10. Interestingly, when UiO-66-NH₂/MUV-10 was again exposed to CO₂ gas without H₂O, the H₂O peak can still be observed, indicating that the adsorption of H₂O was strengthened in UiO-66-NH₂/MUV-10. In addition, the peak intensity of the *COOH and *CO intermediates in UiO-66-NH₂/MUV-10 was significantly higher than that in UiO-66-NH₂ and MUV-10, suggesting the high CO₂RR activity (Fig. 5e). These results imply that the adsorption of gas molecules was strengthened, while their kinetics accelerated.

The electronic structure and charge migration pathway were analysed via valence-band XPS and UPS. The valence-band potentials of MUV-10, UiO-66-NH₂, 20-UiO-66-NH₂/MUV-10, 30-UiO-66-NH₂/MUV-10 and 40-UiO-66-NH₂/MUV-10 were determined from valence-band XPS spectra to be 2.42, 2.20, 2.35, 2.37, and 2.38 eV, respectively. Based on the formula E_CB = E_g − E_VB, the conduction potentials of MUV-10, UiO-66-NH₂, 20-UiO-66-NH₂/MUV-10, 30-UiO-66-NH₂/MUV-10 and 40-UiO-66-NH₂/MUV-10 can be obtained (Fig. S28†). The UPS measurements were calibrated using a gold reference sample, and the working function values of the sample were correctly obtained in our setup. The work functions of 20-UiO-66-NH₂/MUV-10 (2.11 eV), 30-UiO-66-NH₂/MUV-10 (2.02 eV) and 40-UiO-66-NH₂/MUV-10 (2.34 eV) were lower than those of MUV-10 (2.36 eV) and UiO-66-NH₂ (2.50 eV) (Fig. 5f). Notably, the work function of UiO-66-NH₂/MUV-10 is significantly reduced to 2.02 eV, indicating that the delocalized π bonds in the covalent interfaces can significantly reduce the energy barrier for electron effusion (Figs. S29†).

To further understand the above CO₂ and H₂O adsorption performances, GCMC simulations were performed to analyse the adsorption sites and distribution density of CO₂ and H₂O molecules. The adsorption sites and strength were measured by the distance between the guest molecule and the skeleton via loading one molecule of CO₂ or H₂O into the skeleton. In MUV-10 (Fig. S30a†), CO₂ underwent C_CO2⋯O and O_CO2⋯C electrostatic interaction with uncoordinated carboxylic O and C (C_CO2⋯O = 3.57–3.58 Å and O_CO2⋯C = 3.50–3.88 Å); whereas, in UiO-66-NH₂, the distances between CO₂ and the framework are 3.56 (C_CO2⋯O) and 3.15–3.98 Å (O_CO2⋯C) (Fig. S30c†). In addition, the distances between the H₂O molecule and adsorption sites were measured to be 3.01–3.63 Å (O–H⋯O) for MUV-10, while that for UiO-66-NH₂ are 2.92–3.77 Å (O–H⋯N), and 2.87–3.50 Å (O–H⋯O) (Fig. S30b and d†). Nevertheless, the calculation results reveal that the CO₂ loading density in UiO-66-NH₂ was lower than that in MUV-10 (Fig. S31 and S32†), which was due to the presence of–NH₂ in UiO-66-NH₂ that hinders the entry of gas molecules. The obtained results suggest that both MUV-10 and UiO-66-NH₂ possess high adsorption ability toward both CO₂ and H₂O via the C⋯O interaction and N⋯H interaction.

Based on the above results, the proposed photocatalytic mechanism is shown in Fig. 5g. Upon photoexcitation, the electrons of UiO-66-NH₂ transition from the valence band to the conduction band, which then quickly transfer to MUV-10 in the Z-type heterojunction through the interface and stimulate the CO₂ reduction. The CO₂ molecule is first adsorbed and activated, which then receives electrons and protons to form *COOH intermediates,⁴⁵ further accepts a proton to remove a molecule of water to form *CO, and then desorption of CO occurs to realize the entire photocatalytic CO₂ reduction process.⁴⁶ The covalent π electron interface plays three important roles: (1) inhibition of electron–hole recombination, (2) reduce the electron effusion work function and achieve efficient electron interface transfer, and (3) enhance the capture, adsorption and conversion of gas molecules. In conclusion, the covalent π electron interface of heterojunctions can effectively enhance the photocatalytic CO₂ reduction activity.

Conclusions

In summary, a covalent bond interface with delocalized π bonds was successfully constructed via the encapsulation of UiO-66-NH₂ nanoparticles in MUV-10 skeletons. The delocalized π bond in the covalent interface on the one hand enables efficient electronic excitation and interfacial charge migration and suppresses the carrier recombination. On the other hand, the gas capture and conversion efficiency can be significantly enhanced. Benefiting from the amide bonded heterointerfaces, the UiO-66-NH₂/MUV-10 Z-scheme heterojunction can efficiently convert CO₂ and H₂O under visible light irradiation. The yield of CO reached up to 38.7 μmol g⁻¹ h⁻¹, which surpasses those of the most commonly reported gas–solid phase photocatalysts. This work not only provides a feasible strategy for the controlled construction of interfaces but also shows the great potential of interfacial engineering in efficient photocatalysis and other related fields.

Data availability

All data are available in the main text or the ESI.†

Conflicts of interest

The authors declare that they have no competing interests.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (No. 52302103), Natural Science Foundation of Shaanxi Province (No. 2023-JC-QN-0522), China Postdoctoral Science Foundation (No. 2022TQ0200), Shaanxi Association for Science and Technology Youth Talent Support Program (No. 20230616), and the Youth Innovation Team of Shaanxi Universities (2023).

Notes and references

1. Q. Ma, Y. Gao, B. Sun, J. Du, H. Zhang and D. Ma, Nat. Commun., 2024, 15, 8243.
2. S. Ian, G. Andrey, A. D. Ibadillah, L. Xueqian, A. A. Harry, A. V. David and C. Xiang, Nat. Catal., 2021, 4, 952–958.
3. M. Li, E. Irtem, H.-P. Iglesias van Montfort, M. Abdinejad and T. Burdyny, Nat. Commun., 2022, 13, 5398.
4. D. Segets, C. Andronescu and U.-P. Apfel, Nat. Commun., 2023, 14, 7950.
5. T. Zheng, K. Jiang and H. Wang, Adv. Mater., 2018, 30, 1802066.
6. J. Lee, W. K. Chong, S. H. W. Kok, B. J. Ng, X. Y. Kong, S. P. Chai and L. L. Tan, Adv. Funct. Mater., 2023, 33, 2303430.
7. T. Nakajima, Y. Tamaki, K. Ueno, E. Kato, T. Nishikawa, K. Ohkubo, Y. Yamazaki, T. Morimoto and O. Ishitani, J. Am. Chem. Soc., 2016, 33, 2303430.
8. Y.-X. Pan, Y. You, S. Xin, Y. Li, G. Fu, Z. Cui, Y.-L. Men, F.-F. Cao, S.-H. Yu and J. B. Goodenough, J. Am. Chem. Soc., 2017, 139, 4123–4129.
9. K. Sun, Y. Huang, Q. Wang, W. Zhao, X. Zheng, J. Jiang and H.-L. Jiang, J. Am. Chem. Soc., 2024, 146, 3241–3249.
10. H. Terholsen, H. D. Huerta-Zerón, C. Möller, H. Junge, M. Beller and U. Bornscheuer, Angew. Chem., Int. Ed., 2024, 136, e202319313.
11. S. S. Hannah and Y. Y. Jenny, Nat. Catal., 2021, 4, 928–933.
12. D. Li, Y. Huang, S. Li, C. Wang, Y. Li, X. Zhang and Y. Liu, Chin. J. Catal., 2020, 41, 154–160.
13. W. Ziqi, Y. Zhongqing, F. Ruiming, Y. Yunfei, R. Jingyu and Z. Li, Chem. Eng. J., 2021, 429, 132322.
14. A. Kumar, R. Bhardwaj and J. Choudhury, ACS Catal., 2023, 13, 927–933.
15. C. Xu, J. Cai, Z. Wang, M. Ni, K. Cen and Y. Zhang, Environ. Sci. Technol., 2019, 53, 12091–12100.
16. X. Ma, H. Liu, W. Yang, G. Mao, L. Zheng and H.-L. Jiang, J. Am. Chem. Soc., 2021, 143, 12220–12229.
17. S. Naghdi, A. Cherevan, A. Giesriegl, R. Guillet-Nicolas, S. Biswas, T. Gupta, J. Wang, T. Haunold, B. C. Bayer, G. Rupprechter, M. C. Toroker, F. Kleitz and D. Eder, Nat. Commun., 2022, 13, 282.
18. M. Xu, D. Li, K. Sun, L. Jiao, C. Xie, C. Ding and H.-L. Jiang, Angew. Chem., Int. Ed., 2021, 133, 16508–16512.
19. H. N. Wang, Y. H. Zou, H. X. Sun, Y. Chen, S. L. Li and Y. Q. Lan, Coord. Chem. Rev., 2021, 438, 213906.
20. Y. Ma, S. Wang and X. Duan, Chem. Eng. J., 2023, 455, 140654.
21. R. Chen, Z. Ren, Y. Liang, G. Zhang, T. Dittrich, R. Liu, Y. Liu, Y. Zhao, S. Pang, H. An, C. Ni, P. Zhou, K. Han, F. Fan and C. Li, Nature, 2022, 610, 296–301.
22. Y. Wang, G. Fan, S. Wang, Y. Li, Y. Guo, D. Luan, X. Gu and X. W. Lou, Adv. Mater., 2022, 34, 2204865.
23. Y. He, L. Yin, N. Yuan and G. Zhang, Chem. Eng. J., 2024, 481, 148754.
24. S. Gong, X. Teng, Y. Niu, X. Liu, M. Xu, C. Xu, L. Ji and Z. Chen, Appl. Catal. B Environ., 2021, 298, 120521.
25. G.-Z. Chen, K.-J. Chen, J.-W. Fu and M. Liu, Rare Met., 2020, 39, 607–609.
26. J. Low, J. Yu, M. Jaroniec, S. Wageh and A. A. Al-Ghamdi, Adv. Mater., 2017, 29, 1601694.
27. J. M. Chang, T. H. Lin, K. C. Hsiao, K. P. Chiang, Y. H. Chang and M. C. Wu, Adv. Sci., 2024, 11, 2309526.
28. P. Chen, X. A. Dong, M. Huang, K. Li, L. Xiao, J. Sheng, S. Chen, Y. Zhou and F. Dong, ACS Catal., 2022, 12, 4560–4570.
29. S. S. Bhosale, A. K. Kharade, E. Jokar, A. Fathi, S. M. Chang and E. W. G. Diau, J. Am. Chem. Soc., 2019, 141, 20434–20442.
30. L. Wang, B. Zhu, J. Zhang, J. B. Ghasemi, M. Mousavi and J. Yu, Matter, 2022, 5, 4187–4211.
31. W. Bi, Y. Hu, N. Jiang, L. Zhang, H. Jiang, X. Zhao, C. Wang and C. Li, Appl. Catal. B Environ., 2020, 269, 118810.
32. Q. Niu, S. Dong, J. Tian, G. Huang, J. Bi and L. Wu, ACS Appl. Mater. Interfaces, 2022, 14, 24299–24308.
33. W. Hou, H. Guo, M. Wu and L. Wang, ACS Nano, 2023, 17, 20560–20569.
34. F. D. Novaes, R. Rurali and P. Ordejón, ACS Nano, 2010, 4, 7596–7602.
35. E. Bekyarova, S. Sarkar, F. Wang, M. E. Itkis, I. Kalinina, X. Tian and R. C. Haddon, Acc. Chem. Res., 2013, 46, 65–76.
36. Q. Su, Y. Li, R. Hu, F. Song, S. Liu, C. Guo, S. Zhu, W. Liu and J. Pan, Adv. Sustainable Syst., 2020, 4, 2000130.
37. Z. Ai, L. Jiao, J. Wang and H.-L. Jiang, CCS Chem., 2022, 4, 3705–3714.
38. G. Lu, S. Li, Z. Guo, O. K. Farha, B. G. Hauser, X. Qi, Y. Wang, X. Wang, S. Han, X. Liu, J. S. DuChene, H. Zhang, Q. Zhang, X. Chen, J. Ma, S. C. J. Loo, W. D. Wei, Y. Yang, J. T. Hupp and F. Huo, Nat. Chem., 2012, 4, 310–316.
39. J. Castells-Gil, N. M. Padial, N. Almora-Barrios, J. Albero, A. R. Ruiz-Salvador, J. González-Platas, H. García and C. Martí-Gastaldo, Angew. Chem., Int. Ed., 2018, 130, 8589–8593.
40. W. Yuan, L. Cheng, Y. Zhang, H. Wu, S. Lv, L. Chai, X. Guo and L. Zheng, Adv. Mater. Interfaces, 2017, 4, 1700577.
41. X. Zhang, M. Liu and R. Han, J. Environ. Chem. Eng., 2021, 9, 106672.
42. R. Ramachandran, M. Saranya, A. N. Grace and F. Wang, RSC Adv., 2017, 7, 2249–2257.
43. N. Li, X.-P. Zhai, B. Ma, H.-J. Zhang, M.-J. Xiao, Q. Wang and H.-L. Zhang, J. Mater. Chem. A, 2023, 11, 4020–4029.
44. H. He, W. Zhai, P. Liu and J. Wang, Mater. Today, 2025 DOI:10.1016/j.mattod.2024.12.019.
45. L. Shi, P. Wang, Q. Wang, X. Ren, F. Ichihara, W. Zhou, H. Zhang, Y. Izumi, B. Cao, S. Wang, H. Chen and J. Ye, J. Mater. Chem. A, 2020, 8, 21833–21841.
46. L. Yuan and Y.-J. Xu, Appl. Surf. Sci., 2015, 342, 154–167.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5ta00012b

---