High catalytic activity and stability of visible-light-driven CO₂ reduction via CsPbBr₃ QDs/Cu-BTC core–shell photocatalysts

Abstract

Metal halide perovskites show great potential in photocatalysis, while intrinsic instability seriously hinders their application in photocatalytic CO₂ reduction. Coincidentally, metal–organic frameworks (MOFs) have garnered immense interest due to their unique characteristics of selective CO₂ absorption/activation, a large specific surface area, and highly active metal centers. Herein, in situ growth of a Cu-BTC coating on the surface of CsPbBr₃ quantum dots (CPB QDs) provides an effective photocatalyst for CO₂ reduction. The CPB QDs/Cu-BTC composites exhibit significant enhancements in moisture stability, CO₂ capture and activation capacity, and charge separation efficiency. Therefore, the CPB QDs/Cu-BTC heterojunction exhibits an enhanced CO production rate of 47.82 μmol g⁻¹ h⁻¹, which is 2.2- and 6.8-fold that of pristine CPB QDs and Cu-BTC, respectively. Moreover, a high CO selectivity of up to ∼100% is achieved. Based on in situ diffuse reflectance infrared Fourier transform (DRIFTS) spectra, CPB QDs/Cu-BTC composites facilitate the formation of HCO₃⁻ and ˙CO₂⁻ intermediates for converting CO₂ to CO through an adsorbed *COOH intermediate. This study sets up a new strategy to design excellent perovskite/MOF-based catalysts for promising catalysis.

---

1. Introduction

Excess carbon dioxide (CO₂) emissions have caused a serious greenhouse effect, which in turn makes the climate more extremely unstable.^1–3 Photocatalytic CO₂ reduction reactions have been devised to transform CO₂ into valuable chemicals and fuels, which is considered a potential strategy for mitigating global warming and energy shortages.^4,5 As a new candidate photocatalyst, perovskites have attracted wide attention due to their cost-effectiveness, high extinction coefficient, low exciton binding energy, defect tolerance, and adjustable band structure.⁶ In general, perovskite nanocrystals (NCs) exhibit low stability when exposed to light, oxygen, and heat, primarily due to their ionic crystal structure.^7,8 Consequently, enclosing these NCs within protective barrier hosts has emerged as a proven strategy to address this stability issue.⁹ Embedding within diverse porous substrates, such as polymeric matrices like polystyrene or polyethylene, as well as inorganic materials like TiO₂, SiO₂, Al₂O₃, and CaF₂, has been utilized for encapsulation and stabilization purposes.

Metal–organic frameworks (MOFs) represent innovative porous crystalline materials that are constructed through the integration of inorganic metal nodes and multifunctional organic linkers. Distinguishing themselves from conventional porous materials like carbon-based porous materials, mesoporous silica, and zeolites, MOFs boast tunable structures and pore diameters, large specific surface areas, customizable functionalities, and selective CO₂ absorption/activation.¹⁰ Consequently, MOFs have garnered significant interest for their potential applications in catalysis,¹¹ sensing technologies,¹² information security,¹³ gas adsorption,¹⁴ separation,¹⁵ and storage.¹⁶ Various perovskite@MOF hybrids have been synthesized through diverse methodologies. The first approach for the preparation of perovskite@MOF composites was through a simple mixing strategy, i.e., simply mixing pre-synthesized perovskite nanocrystals and MOFs. It is well known that the pore size of most MOFs is 2–3 nm. Wang and co-workers encapsulated CsPbX₃ QDs into mesostructured zinc-based materials (MOF-5), which is very difficult due to the relatively large size of CsPbX₃ NCs, approximately 10 nm.¹⁷ Moreover, the stepwise/sequential deposition method was used to grow perovskite NCs in the pores of MOFs. Adopting this approach, CH₃NH₃PbBr₃@MOF-5 hybrids were synthesized, which exhibited enhanced chemical and temperature stability.¹⁸ Owing to the microporous structure of MOF materials, the substantial diffusion barrier typically impedes the penetration of perovskite precursors into the MOF matrix. The direct conversion method was developed to grow perovskite NCs into an MOF matrix. Pb-based perovskite NCs could grow inside Pb-MOFs by mixing the halide salt and Pb-MOF.¹⁹ However, the available MOF materials are limited that can be directly converted into perovskite nanocrystals. Therefore, it cannot be widely applied. There is a pressing need to devise a strategy that enables precise control of the size and shape of perovskite NCs and enables their widespread application. Cu-BTC, as a representative MOF, exhibits unique properties, such as tunable structure, high porosity, excellent thermal and chemical stability, and simple synthesis process, which has led to its wide range of applications in the fields of energy storage and catalysis. Motivated by these research findings, it is envisioned that a well-conceived design of a perovskite/Cu-BTC composite structure may lead to the development of a highly efficient photocatalyst for CO₂ reduction.

Herein, a Cu-BTC-encapsulated CPB QD (CPB QDs/Cu-BTC) core/shell Z-scheme heterojunction is designed and fabricated through an in situ synthetic approach and exhibits excellent aqueous stability and remarkable photocatalytic CO₂ reduction performance. CPB QDs are uniformly coated with Cu-BTC under mild synthesis conditions, thereby resulting in greatly improved stability of CPB QDs. Meanwhile, the optical response, photogenerated carrier separation efficiency, and CO₂ capturing ability of the achieved CPB QDs/Cu-BTC hybrids can be efficiently enhanced. As a result, the optimized CPB QDs/Cu-BTC photocatalysts display remarkable activity and exceptional CO selectivity with a CO yield and selectivity of 47.82 μmol g⁻¹ h⁻¹ and 100%, respectively. Furthermore, a plausible mechanism of photocatalytic CO₂ reduction is systematically investigated using in situ DRIFTS spectra, X-ray photoelectron spectra (XPS), and photoelectrochemical characterization. This work proposes a new strategy to design perovskite/MOF-based catalysts for highly efficient CO₂ conversion.

2. Experimental section

2.1 Preparation of CPB QDs

CPB QDs were synthesized via a hot injection method according to a previous report.²⁰ The preparation of lead precursor solution: Cs-oleate precursor solution was first prepared by dissolving Cs₂CO₃ (0.814 g) in a mixture of octadecene (40 mL) and oleic acid (2.5 mL) in a three-neck bottle and dried under vacuum for 1 h at 120 °C. Then, the mixture was heated to 150 °C under N₂ until Cs₂CO₃ was completely dissolved.

2.2 CPB QD synthesis

PbBr₂ (207 mg, 0.56 mmol) and ODE (15 mL) are loaded into a 50 mL three-neck round-bottom flask and heated under vacuum to 120 °C for 60 min. Subsequently, 1.5 mL of OAm and 1.5 mL of OA are loaded at 120 °C under nitrogen. Once PbBr₂ was fully dissolved, the temperature was elevated to 150 °C, subsequently accompanied by a swift injection of 0.4 mL of Cs-oleate precursor. Then, the reaction was then immediately (within 5 s) quenched using an ice-water bath. The solution was centrifuged and the precipitate was collected and redispersed in ethyl acetate for further characterization.

2.3 Preparation of Cu-BTC

Cu(NO₃)₂·3H₂O (4.1 mmol, 100 mg) was loaded into 30 mL of ethyl acetate and 1,3,5-benzenetricarboxylic acid (2.5 mmol, 53 mg) was loaded into 30 mL of ethyl acetate. The copper nitrate solution was added dropwise into the 1,3,5-benzenetricarboxylic acid solution and stirred vigorously for 15 min at room temperature. A sky blue precipitate was observed. Then, the precipitate was filtered and washed 3 times with ethyl acetate and dried at 60 °C under vacuum.

2.4 Preparation of CPB QDs/Cu-BTC hybrids

The prepared CPB QDs (48.04 mg, 60.00 mg, 72.01 mg, 84.00 mg, and 96.08 mg, respectively) were dispersed in 20 mL ethyl acetate solution. Then, Cu(NO₃)₂·3H₂O (0.2 mmol, 50.0 mg) and 1,3,5-benzenetricarboxylic acid (0.125 mmol, 26.5 mg) were added to the mixture under continuous stirring for 1 h to grow Cu-BTC. The composites were collected by centrifugation and washed 3 times with n-hexane. The composite catalysts were denoted as CPB QDs/Cu-BTC-x (x = 0.4, 0.5, 0.6, 0.7, and 0.8).

3. Results and discussion

3.1 Structural characterization

The heterojunction of CPB QDs/Cu-BTC was prepared by the in situ growth of Cu-BTC on the surface of CPB QDs, as illustrated in Fig. 1a and S1.† The specific synthesis process can be briefly described as follows: presynthesized CPB QDs were dispersed directly in the precursor solution of the copper source (Cu(NO₃)₂·3H₂O) and organic component (C₆H₃(CO₂H)₃), and then Cu-BTC was directly coated on the surface of the CPB QDs to form a core@shell structure. First, CPB QDs were prepared by a hot-injection method according to a previous report.^20,21 The transmission electron microscopy (TEM) image of pristine CPB exhibited a cubic shape with a mean particle size of 8–9 nm (Fig. 1b). The preparation process of Cu-BTC is relatively simple. The metal salts and organic ligands were dispersed in ethyl acetate solution and further stirred at room temperature to generate Cu-BTC. Based on this feature, CPB QDs can be successfully encapsulated by the framework of Cu-BTC. The above-mentioned growth steps are described in the Experimental section. The morphology was further revealed by the scanning electron microscope (SEM) analyses. Top-view SEM images demonstrated that the morphology of Cu-BTC changed from a nanoparticle (Fig. S2a and b†) to a nanorod structure formed by accumulating nanoparticles (Fig. 1c and S2c and d†) after the recombination with CPB QDs. Furthermore, the color of CPB QDs/Cu-BTC hybrids changed significantly, changing from blue to pea green (Fig. S3a and b†). The TEM images were further analyzed to study the loading state of CPB QDs. For the CPB QDs/Cu-BTC composites, the spherical-like CPB QDs were completely coated with Cu-BTC, and CPB QDs displayed dispersion (Fig. 1d). As shown in Fig. 1d, S4 and S5,† the size of CPB QDs in the hybrid catalysts was not uniform (approximately 3–10 nm), which was attributed to the uneven size of the initial CPB QDs (Fig. 1b). The inset of Fig. 1d, S5a and S5b† display high-resolution TEM (HRTEM) images of CPB QDs, which were highly crystalline with a clear lattice fringe of 0.21 nm, assigned to the (220) plane of cubic-phase CPB.²² To further confirm the ubiquitous presence of different elements in the CPB QDs/Cu-BTC heterostructures, an energy-dispersive X-ray (EDX) analysis was carried out. As depicted in Fig. 1e, EDX elemental mappings of Cu, O, Cs, Pb, and Br revealed their existence and uniform distribution. The above-mentioned results strongly suggested that CPB QDs were evenly distributed in Cu-BTC.

Fig. 1 (a) Schematic illustration of CPB QDs/Cu-BTC hybrids. (b) TEM image and size distribution curve of CPB QDs. (c) SEM image of the CPB QDs/Cu-BTC composites. (d) TEM image of the CPB QDs/Cu-BTC composites. Inset shows the HRTEM image of pristine CPB QDs. (e) Elemental mapping of CPB QDs/Cu-BTC hybrids. (f) The XRD patterns of pristine CPB QDs, Cu-BTC, and CPB QDs/Cu-BTC hybrids.

The crystalline structure of CPB QDs, Cu-BTC, and CPB QDs/Cu-BTC was analyzed via X-ray diffraction (XRD) measurements. As displayed in Fig. 1f, the achieved CPB QDs displayed their characteristic diffraction peaks at 21.16°, 30.54°, 34.25°, 37.63° and 43.73°, which corresponded to the (110), (200), (210), (211), and (220) crystal planes, respectively. The XRD pattern of CPB QDs matched well with the standard card (JCPDS no. 01-087-4956) of cubic-phase CsPbBr₃.²³ Cu-BTC shows excellent crystallinity, in which the peaks at 11.62°, 19.10°, 24.09°, and 29.40° are ascribed to (222), (440), (551) and (751) crystal planes. The diffraction peaks were well-matched with previous results.²⁴ After the formation of CPB QDs/Cu-BTC, their diffraction peaks were observed and still sharp, indicating that the crystallinity of the CPB QDs and Cu-BTC was not affected. The abovementioned results demonstrated the successful synthesis of CPB QDs/Cu-BTC nanocomposites.

The optical properties of the as-prepared samples are preliminarily analyzed by UV-vis diffuse reflectance spectroscopy (DRS). As shown in Fig. 2a, CPB QDs and Cu-BTC both showed good visible-light response. Their corresponding Tauc plots are displayed in Fig. S6a and b,† showing estimated band gaps of 2.28 and 2.26 eV for CPB QDs and Cu-BTC, respectively. It should be noted that the absorption peaks centered at approximately 700 nm were associated with the d–d transition of Cu²⁺.²⁵ It is well known that the photoluminescence (PL) can reflect the interfacial charge separation in the semiconductor. Therefore, the separation efficiency of charge carriers in the CPB QDs/Cu-BTC heterojunctions was studied by steady-state PL spectroscopy. Under the excitation of 325 nm, pristine CPB QDs displayed a strong emission peak at 527 nm (Fig. 2b). For CPB QDs/Cu-BTC, the PL intensity was significantly quenched, confirming the effective separation of electron–hole pairs. Meanwhile, the peak signal of CPB QDs/Cu-BTC composites was detected at 520 nm (Fig. 2b), displaying a slight shift to higher energy, which was caused by the reduced size of CPB QDs. In addition, time-resolved photoluminescence decay (TRPL) was considered a powerful method to probe the charge transfer dynamics in a semiconductor. To reveal the photogenerated charge transfer behavior in the CPB QDs/Cu-BTC heterojunction composites, the TRPL spectra were monitored. Fig. 2c shows the TRPL spectra of CPB QDs and CPB QDs/Cu-BTC, which were fitted with a biexponential function. The average PL lifetimes decreased from 13.09 μs for CPB QDs to 5.31 μs for CPB QDs/Cu-BTC (Table S1†), aligning well with the PL spectra results. The aforementioned results suggested a close contact and efficient charge transfer between CPB QDs and Cu-BTC. The charge separation and transfer abilities of the samples were further investigated using photocurrent and electrochemical impedance spectroscopy (EIS) spectra. As revealed by the I–t curves (Fig. 2d), the CPB QDs/Cu-BTC heterojunction composites exhibited higher photocurrent density compared with those of individual CPB QDs and Cu-BTC. The notable photocurrent response, combined with the quenched PL intensity, indicated inhibited charge recombination and promoted charge transfer efficiency in CPB QDs/Cu-BTC. Besides, the EIS spectra showed that the CPB QDs/Cu-BTC heterojunctions displayed a smaller semicircular radius than those of pure CPB QDs and Cu-BTC (Fig. 2e), demonstrating a lower charge transfer resistance and promoting charge separation. CO₂ adsorption is an important condition for photocatalytic CO₂ reduction. Therefore, the CO₂ capture capability of CPB QDs, Cu-BTC, and CPB QDs/Cu-BTC-0.6 was evaluated based on the CO₂ adsorption isotherms at 273 K. As shown in Fig. 2f, the CO₂ uptake was 2.62 and 33.52 cm³ g⁻¹ for pristine CPB QD and Cu-BTC samples, respectively. For the CPB QDs/Cu-BTC-0.6 sample, the CO₂ uptake was 57.68 cm³ g⁻¹, which was 22.01-fold and 1.72-fold that of pristine CPB QDs and Cu-BTC. The results indicated that the Cu-BTC coating can enhance the enrichment of CO₂ on the surface of the photocatalysts, thus facilitating CO₂ reduction.

Fig. 2 (a) UV-vis absorption spectra of pristine CPB QDs, Cu-BTC, and CPB QDs/Cu-BTC heterojunctions with different ratios. (b) Steady-state PL spectra of pure CPB QD and CPB QDs/Cu-BTC-0.6 photocatalysts. (c) Time-resolved PL decays of pure CPB QD and CPB QDs/Cu-BTC-0.6 photocatalysts. (d) Photocurrent density–time curves of pure CPB QDs, Cu-BTC, and CPB QDs/Cu-BTC-0.6 heterojunction. The light intensity is 200 mW cm⁻². (e) EIS Nyquist plots of pure CPB QDs, Cu-BTC, and CPB QDs/Cu-BTC-0.6 heterojunction. (f) CO₂ adsorption–desorption isotherms of CPB QDs, Cu-BTC, and CPB QDs/Cu-BTC-0.6 at 273 K.

3.2 Photocatalytic CO₂ reduction performance

The photocatalytic performance of as-obtained CPB QDs/Cu-BTC was evaluated by photocatalytic CO₂ reduction in acetonitrile (see details in the ESI†). The time profiles of CO and CH₄ evolution showed that CPB QDs, Cu-BTC, and CPB QDs/Cu-BTC exhibited linear production of CO and CH₄ over 5 h (Fig. S7a and b†). As shown in Fig. 3a, pristine CPB QDs and Cu-BTC showed lower photocatalytic yields of CO (22 and 7 μmol g⁻¹ h⁻¹, respectively) due to rapid charge recombination. It was noteworthy that the CO production was significantly enhanced by the increased Cu-BTC loading, and the maximum CO yield and electron consumption rate of CPB QDs/Cu-BTC-0.6 were 47.82 and 95.64 μmol g⁻¹ h⁻¹, which were about 2.2 and 6.8 times higher than those of CPB QDs and Cu-BTC, respectively, due to its efficient charge separation, enhanced CO₂ uptake and visible-light-capturing ability. Compared to previously reported photocatalysts (Table S2†), the product yield and cycling stability of CPB QDs/Cu-BTC are among the best, demonstrating the considerable photocatalytic activity of CPB QDs/Cu-BTC for CO₂ reduction. Meanwhile, the obtained primary product was CO and the CO selectivity was almost 100% (Fig. 3b). However, a further increase in Cu-BTC loading would decrease CO production, resulting from the increased difficulty of photogenerated carriers migrating to the surface. To study the influencing factors in the photocatalytic test, a series of comparative experiments were carried out. Fig. 3c presents the performance of CPB QDs/Cu-BTC-0.6 under different conditions. Only a small amount of CO production was obtained under dark conditions, indicating the importance of light irradiation. Furthermore, CO was not produced in the absence of catalysts, demonstrating that photocatalysts were indispensable for photocatalytic reactions. To further confirm the origin of CO production, a comparative experiment was conducted using N₂ to replace CO₂. Almost no product was detected, confirming that the C in the product was derived from photocatalytic CO₂ reduction. The above experimental results indicated that the photocatalysts, light illumination, and CO₂ gas, acting as the driving force for photocatalytic reactions, were essential to achieve CO production. More importantly, the photocatalytic performance of CPB QDs + Cu-BTC samples achieved via the electrostatic self-assembly method was determined, and the conversion rate of CO was only 31.23 μmol g⁻¹ h⁻¹, which was significantly lower than the activity of the CPB QDs/Cu-BTC-0.6 heterojunction, indicating that the constructed hybrids have a tight contact interface rather than a simple physical coacervation. Furthermore, the impact of Cu-BTC on photocatalytic stability was evaluated by cycling experiments and long-term photocatalytic reactions. As displayed in Fig. 3d, the production yield of CO remained almost unchanged after ten consecutive reaction cycles. However, the photocatalytic activity of pure CPB QDs exhibited a significant drop over ten cycles under the same photocatalytic conditions, owing to its poor chemical and light stability (Fig. 3e). Intriguingly, the CPB QDs/Cu-BTC-0.6 heterojunction displayed linearly enhanced CO production for over 34 h of continuous irradiation (Fig. 3f), further indicating its excellent long-term stability. In addition, the CPB QDs/Cu-BTC catalysts were characterized by XRD and Raman analyses before and after photocatalytic CO₂ reduction. No significant changes were found, which further proved that CPB QDs/Cu-BTC has good structural stability (Fig. S8†). These results indicate that porous material encapsulation presents a promising approach for proposing novel high-performance photocatalysts.

Fig. 3 (a) CO₂ photoreduction activities of CPB QD, Cu-BTC, and CPB QDs/Cu-BTC-x catalysts with different ratios. (b) CO selectivity for CPB QD, Cu-BTC, and CPB QDs/Cu-BTC-x catalysts. (c) Gas evolution over CPB QDs/Cu-BTC-0.6 heterojunctions under different reaction conditions. (d) Cycling stability test of the CPB QDs/Cu-BTC-0.6 heterostructure. (e) Cyclic experiment of pure CPB QDs. (f) Long-term photocatalytic tests of the CPB QDs/Cu-BTC-0.6 heterostructure.

The stability of perovskite plays a crucial role in photocatalysis, as it determines the durability and efficiency of the catalyst under light irradiation. The more stable the perovskite material, the longer it can maintain its catalytic activity, thereby enhancing the overall effectiveness of photocatalytic processes. Therefore, the stability of CPB QDs and CPB QDs/Cu-BTC-0.6 composite catalysts under different conditions was investigated by fluorescence spectroscopy characterization. A 325 nm laser with the same laser power density served as the irradiation source for assessing the photochemical properties of the samples, as depicted in Fig. 4a and S9–S11.† Pure CPB QDs and CPB QDs/Cu-BTC-0.6 composites maintained 65% and 90% PL intensities after 6 h of light irradiation (Fig. 4a and S9a and b†), respectively, indicating that the composite catalysts enhanced photostability. Upon immersion in water for 2 h, CPB QDs without Cu-BTC shells displayed a PL intensity that was 38% of their initial value, whereas those encapsulated with Cu-BTC shells retained 78% of their original PL intensity (Fig. 4b and S9c, d†). The CPB QDs/Cu-BTC composites exhibited significantly enhanced resistance to water erosion compared to bare CPB QDs. This marked improvement in stability was attributed to the effective suppression of CPB QD decomposition in aqueous environments upon Cu-BTC encapsulation. The Cu-BTC shells serve as a barrier, minimizing direct exposure of the perovskite nanocrystals to water molecules. While some solvents may still penetrate the porous materials, causing minor degradation, the Cu-BTC coating ensures superior long-term stability even under extreme water exposure. It is hypothesized that the formation of a heterojunction between CPB QDs and Cu-BTC is crucial for stabilizing the perovskite materials, mitigating water adsorption on the perovskite surface and thereby safeguarding the CPB QD core from water-driven corrosion. Analogous to water molecules, polar solvents comprising short-chain alcohols, such as isopropanol (IPA) and ethyl alcohol (Et₂O), methylbenzene (MB), and chlorobenzene (C₆H₅Cl) can also induce perovskite structural damage. Consequently, we probed the temporal evolution of PL spectra for CPB QDs and CPB QDs/Cu-BTC composites dispersed in various typical polar solvents (Fig. 4c and S9e, f, S10a–c, and S11a–f†). Following 30–35 days of immersion in IPA, MB, Et₂O, or C₆H₅Cl, the PL intensities of bare CPB QDs diminished to 31%, 58%, 71%, and 50% of their initial levels, respectively. In contrast, when encapsulated within Cu-BTC shells, the PL intensities remained robust at 50%, 64%, 74%, and 54% under identical soaking conditions. It is evident that when the composite materials were immersed in IPA solution, the PL spectrum quenching of the composites was significantly suppressed compared to pristine CPB QDs. When the composite materials were immersed in solutions of MB, Et₂O, or C₆H₅Cl, the PL intensity decay of the single CPB QDs and the composite catalyst was comparable. Meanwhile, the XRD patterns of the single Cu-BTC material before and after immersion in solutions of water, IPA, MB, Et₂O, and C₆H₅Cl showed no significant changes (Fig. 4d). This indicates that the above five solvents have no destructive effect on Cu-BTC. It is speculated that the possible reason is that the hydrophilic and hydrophobic abilities of Cu-BTC towards the above solvents are different. As shown in Fig. S12,† for water, IPA, MB, Et₂O, and C₆H₅Cl, the contact angles of θ = 55.3°, 67.1°, 11.4°, 6.0° and 13.2° were observed, respectively. Therefore, the CPB QDs/Cu-BTC composites exhibited significantly enhanced resistance to water and IPA erosion compared to bare CPB QDs (Fig. 4e). In contrast, MB, Et₂O, and C₆H₅Cl are more conducive to adsorbing on the surface of the composites and entering the shell, resulting in the decomposition of perovskite (Fig. 4f).

Fig. 4 (a) PL intensity of CPB QDs and CPB QDs/Cu-BTC composites as a function of time under 325 nm laser irradiation. Time-dependent PL intensity in different solvents. (b) Water. (c) Isopropanol (IPA). (d) The XRD patterns of Cu-BTC soaked in different solvents. The diagram of CPB QDs/Cu-BTC composites in water and IPA (e), and MB, Et₂O, and C₆H₅Cl (f).

3.3 Enhanced photocatalytic activity mechanism

To further analyze the chemical composition, valence states, and interfacial interaction between CPB QDs and Cu-BTC, XPS spectra were recorded over CPB QDs, Cu-BTC, and CPB QDs/Cu-BTC heterojunctions. As shown in Fig. S13a,† the analysis of CPB QDs/Cu-BTC heterojunctions displayed the characteristic signals of Cs, Pb, Br, Cu, C, and O elements in the XPS spectrum, which further verified the formation of CPB QDs/Cu-BTC heterojunctions. The high-resolution XPS spectra of these elements were studied to explore the interfacial interaction between CPB QDs and Cu-BTC. As shown in Fig. 5a, two narrow peaks with binding energies of 723.94 eV and 737.97 eV were related to Cs 3d_5/2 and Cs 3d_3/2, respectively.²⁶ After coating with Cu-BTC, these peaks were positively shifted. The same positive shift can be observed in Pb 4f and Br 3d spectra (Fig. 5b and c). The binding energies of Pb 4f_7/2 and Pb 4f_5/2 appeared at 138.37 and 143.28 eV,²⁷ respectively, while those peaks in CPB QDs/Cu-BTC hybrids were located at higher values (138.86 and 143.74 eV) (Fig. 5b). The Br 3d spectrum of CPB QDs/Cu-BTC hybrids was deconvoluted into two peaks at 68.62 eV and 69.57 eV (Fig. 5c), corresponding to Br 3d_5/2 and Br 3d_3/2, respectively.²⁸ Compared with bare CPB QDs, these peaks positively shifted by ∼0.5 eV. The positive shift indicated electron deficiency. In contrast, the C 1s, O 1s, and Cu 2p spectra exhibited a contrasting trend. The binding energies of 284.95 and 288.86 eV corresponded to phenyl and carboxyl signals,²⁹ respectively, which were higher than those of CPB QDs/Cu-BTC hybrids by ∼0.12 eV (Fig. S13b†). The oxygen O 1s peak at 532.14 eV was assigned to Cu–O–C species (Fig. 5d),³⁰ which was 0.08 eV higher than that of CPB QDs/Cu-BTC hybrids. As shown in Fig. 5e, there were two strong peaks of Cu²⁺ at 934.76 and 954.78 eV, corresponding to Cu 2p_3/2 and Cu 2p_1/2, respectively.³¹ In contrast, for CPB QDs/Cu-BTC hybrids, the Cu 2p_3/2 and Cu 2p_1/2 peaks underwent a shift toward lower binding energies, specifically by approximately 934.66 eV and 954.61 eV, respectively. In addition, the in situ XPS spectra of the CPB QDs/Cu-BTC heterojunctions showed similar experimental results. The binding energies of Cs 3d, Br 3d, and Pb 4f shifted in the positive direction (Fig. S14a–c†) under illumination, while that of C 1s shifted in the negative direction (Fig. S14d†) under light irradiation. Additionally, the presence of Cu²⁺ species was evidenced by the shakeup satellite peaks observed near 962.17 eV and 942.09 eV,³² respectively. In the Cu LMM Auger spectrum, 917.62 eV was attributed to Cu²⁺ (Fig. 5f).³³ With the introduction of the CPB QDs, the valence state of Cu does not change significantly, while the core energy moves towards low binding energy. Consequently, Cu-BTC became an electron-rich state in the CPB QDs/Cu-BTC heterojunction. To balance the electron redistribution, CPB QDs adopted an electron-deficient state. These observations signify the establishment of a close interface and robust chemical coupling between CPB QDs and Cu-BTC within the heterostructure.

Fig. 5 XPS characterization of pristine CPB QDs, Cu-BTC, and CPB QDs/Cu-BTC composites. High-resolution XPS spectra of Cs 3d (a), Pb 4f (b), Br 3d (c), O 1s (d), Cu 2p (e) and Cu LMM (f).

Subsequently, the band positions of CPB QDs and Cu-BTC were examined employing the Mott–Schottky (M–S) methodology. As depicted in Fig. 6a and b, the M–S plots for both CPB QDs and Cu-BTC display a positive slope, signifying that both samples possess n-type semiconducting properties. The flat-band potentials (E_fb) of the CPB QDs and Cu-BTC were −0.79 V and −0.93 V (vs. SCE), respectively, thus corresponding to −0.55 V and −0.69 V (vs. NHE). It is well understood that the conduction band position (E_CB) was about 0.2 V higher than the E_fb for an n-type semiconductor,³⁴ and the E_CB of CPB QDs and Cu-BTC was roughly calculated to be −0.75 V and −0.89 V (vs. NHE), respectively. Based on the formula E_VB = E_CB + E_g, the valence band potential (E_VB) was 1.53 V and 1.37 V (vs. NHE) for CPB QDs and Cu-BTC, respectively. The XPS valence band spectra evaluated the energy difference between the valence band maximum and E_f.³⁵ As shown in Fig. S15a and b,† the valence band values with respect to the Fermi level have been ascertained as 1.63 eV for CPB QDs and 1.85 eV for Cu-BTC, respectively. Based on the above results, CPB QDs and Cu-BTC have staggered energy band alignments, and the band structure is displayed in Fig. 6c. Based on the above results, we deduce that the charge transfer dynamics at the heterojunction interface adhere to a Z-scheme mechanism. As depicted in Fig. 6c, upon light irradiation, electron–hole pairs are independently generated within both CPB QDs and Cu-BTC. It is hypothesized that the photoinduced holes in the valence band of Cu-BTC recombined with the electrons in the conduction band of CPB QDs, resulting in spatially segregated electrons and holes, respectively, within CPB QDs and Cu-BTC. Subsequently, these high-energy electrons and holes can be harnessed to facilitate CO₂ reduction and water oxidation processes, respectively. Under visible light irradiation, CPB, Cu-BTC and CPB QDs/Cu-BTC heterojunction samples displayed the characteristic ESR signal of DMPO−˙O₂⁻ adducts. The signal strength of the DMPO−˙O₂⁻adduct in CPB QDs/Cu-BTC was significantly stronger than those of pure CPB and Cu-BTC (Fig. S16†), indicating that there was effective carrier separation in CPB QDs/Cu-BTC hybrids. The Z-scheme reduced CO₂ to CO using water as an electron donor under visible light irradiation. This Z-scheme charge transfer paradigm imparts the CPB QDs/Cu-BTC heterojunctions with efficient charge separation capabilities, significantly enhancing their overall photocatalytic performance.

Fig. 6 Mott–Schottky plots of CPB QDs (a) and Cu-BTC (b). Schematic diagram of the interfacial charge transfer and accumulation on CPB QDs and Cu-BTC (c).

To gain a profound comprehension of the photocatalytic CO₂ reduction reaction (CRR) pathway and underlying mechanisms occurring on photocatalyst surfaces, we conducted in situ diffuse reflectance infrared Fourier transform spectra (DRIFTS) analysis, enabling precise monitoring of reaction intermediates throughout the adsorption, activation, and transformation stages of CO₂ on the surfaces of CPB QD, Cu-BTC, and CPB QDs/Cu-BTC catalysts. CPB QD, Cu-BTC, and CPB QDs/Cu-BTC catalysts were exposed to a CO₂/H₂O gas mixture to reach adsorption equilibrium, and the DRIFTS spectra were acquired separately under dark and light conditions (Fig. 7 and S17 and S18†). As shown in dynamic IR spectra in Fig. 7a–c, several carbonates and active ˙CO₂⁻ intermediates are formed on the surface of photocatalysts during the wet CO₂ adsorption process in the dark. As shown in Fig. S17a, c, and e,† the peaks from 2297 to 2383 cm⁻¹ corresponded to the asymmetric stretching of surface CO₂ molecules.³⁶ The strong adsorption peaks of CPB QDs/Cu-BTC hybrids exhibited their good CO₂ adsorption capacity. The overtone bands of gas-phase CO₂ molecules were also observed at 3593, 3624, 3705, and 3729 cm⁻¹ (ref. 37) (Fig. S17b, d, and f†). Compared with pure CPB QDs and Cu-BTC, the new characteristic vibration modes of ˙CO₂⁻ (1658 and 1685 cm⁻¹)³⁸ were detected (Fig. 7c). ˙CO₂⁻ is considered a pivotal intermediate for hydrogenation of CO₂, with the rate of its formation being a crucial step in the photocatalytic reduction of CO₂. In contrast, the pristine CPB QD and Cu-BTC surfaces lacked these distinct absorption peaks. The prominent peak observed at 1621 cm⁻¹ signified efficient adsorption of H₂O.³⁹ The other two characteristic peaks located at 1388 and 1455 cm⁻¹ belonged to bicarbonate (HCO₃⁻),⁴⁰ originating from the interfacial reactions between adsorbed CO₂ and H₂O on the surface. Theoretically, the transformation of HCO₃⁻ into ˙CO₂⁻ necessitates an energy barrier of approximately 0.27 eV (vs. NHE), whereas the conversion of linear CO₂ to ˙CO₂⁻ entails a significantly higher barrier of roughly 1.9 eV (vs. NHE).⁴¹ Consequently, the modification of the CPB surface with Cu-BTC facilitates the generation of HCO₃⁻, leading to a reduction in the energy barrier and an enhancement in the activity of CO₂ reduction. Besides, the peaks at 1324, 1558, and 1539 cm⁻¹ were assigned to bidentate carbonate species (b-CO₃²⁻), and the peaks at around 1750 and 1721 cm⁻¹ were attributed to the chelating-bridged carbonate (c-CO₃²⁻).^38,42 Of these, the dual-site bonding of CO₂ (as b-CO₃²⁻) was more favorable for enhancing the adsorption and subsequent activation of inert CO₂ molecules.³⁸ The presence of diverse carbonate species can be ascribed to the generation of carbonic acid and/or the adsorption of CO₂ onto the catalyst surfaces. Both carbonate and bicarbonate species serve as crucial intermediates in the CO₂ transformation process. HCO₃⁻ played a pivotal role in the transformation process of CO₂ into CO. Compared with CPB QDs and Cu-BTC, CPB QDs/Cu-BTC photocatalysts produced abundant intermediate species and a higher intensity of the formed ˙CO₂⁻ radical, confirming that the composite catalyst was more favorable for CO₂ catalytic conversion. After light was applied to the reaction cell, the above-mentioned absorption peaks still existed and their intensity increased gradually with time (Fig. 7d–f and S18a–f†), signifying the activation of the adsorbed CO₂ molecules. Most of the peaks on CPB QDs/Cu-BTC catalysts exhibited greater strength and can be identified, demonstrating that CPB QDs/Cu-BTC shows more efficient photoreduction ability than pristine CPB QDs and Cu-BTC. In addition, the characteristic peaks of *CO at 1715 cm⁻¹ (ref. 38) and *COOH at 1432 cm⁻¹ (ref. 43) can be detected and increased in intensity as the photocatalytic reaction continues. The formation of *COOH and *CO further proved that illumination of the photocatalysts led to the generation of CO by multistep hydrogenation of ˙CO₂⁻ radicals. Furthermore, the notably enhanced peak intensity observed in CPB QDs/Cu-BTC could potentially serve as a pivotal factor contributing to the increased selectivity towards CO products.

Fig. 7 In situ DRIFTS spectra for photocatalytic CO₂ reduction over CPB QD, Cu-BTC, and CPB QDs/Cu-BTC photocatalysts (a–c) in the dark and (d–f) under simulated solar illumination.

4. Conclusion

In summary, a unique Z-scheme CPB QDs/Cu-BTC heterostructure was constructed by directly coating the Cu-BTC shell on the surface of CPB QDs. Notably, the CPB QDs/Cu-BTC heterostructure demonstrated a substantial enhancement in both activity and moisture stability for CO₂ photoreduction compared to pristine CPB QDs. The electron consumption rate for CPB QDs/Cu-BTC composites was 95.64 μmol g⁻¹ h⁻¹, which was 2.2-fold that of naked CPB QDs. The activity of CPB QDs/Cu-BTC almost remained stable during 10 cycle experiments and 34 h of continuous irradiation. This enhancement can be attributed to the improved charge transfer efficiency, maintenance of high redox potential among charge carriers, improved visible light absorption, and enhanced CO₂ capture capabilities. Our research offers valuable insights into the strategic design and synthesis of highly efficient perovskite/MOF-based heterostructures tailored for photocatalytic applications.

Data availability

The data supporting this article have been included as part of the ESI.†

Author contributions

Yuanming Hou: investigation, data curation, visualization, and writing – original draft. Yanqing Zhang: formal analysis, resources, and data curation. Shilong Jiao: validation, methodology, supervision, and funding acquisition. Jingyi Qin: resources, data curation, and investigation. Luoyu Liu: formal analysis, investigation, and resources. Zhengzheng Xie: validation, formal analysis, and supervision. Zhongjie Guan: validation, formal analysis, and supervision. Jianjun Yang: validation, supervision, and methodology. Qiuye Li: validation, supervision, methodology, and funding acquisition. Xianwei Fu: conceptualization, methodology, supervision, funding acquisition, and writing – review & editing.

Conflicts of interest

The authors declare no competing financial interest.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (52102166 and 22405073), the China Postdoctoral Science Foundation (2021M701065, 2021M701071, and 2022T150187), the Science and Technology Development Plan Joint Fund of Henan Province (225200810051), the Key Research & Development and Promotion Project of Henan Province (Science and Technology Tackling Key Problems) under Grant No. 242300420534, the Key Science and Technology Research Project of Education Department of Henan Province (No. 24A430005), the Natural Science Foundation of Henan Province (222300420406 and 242300420531), and the Program for Innovative Research Team in University of Henan Province (21IRTSTHN009).

References

1. Y. Wang, J. Wang, M. Zhang, S. Zheng, J. Wu, T. Zheng, G. Jiang and Z. Li, Small, 2023, 19, 2300841.
2. W. Zhan, H. Gao, Y. Yang, X. Li and Q. Zhu, Adv. Energy Sustain Res., 2022, 3, 2200004.
3. W. Zhang, A. R. Mohamed and W.-J. Ong, Angew. Chem., Int. Ed., 2020, 59, 22894–22915.
4. Y. Hou, M. Dong, J. He, J. Sun, C. Sun, X. Li, X. Wang and Z. Su, Inorg. Chem. Front., 2023, 10, 5098–5110.
5. S. Chen, H. Yin, P. Liu, Y. Wang and H. Zhao, Adv. Mater., 2023, 35, 2203836.
6. H. Zhang, X. He, H. Wang, L. Chen, G. Xu, N. Zhang, K. Qu, Q. He, Y. Peng and J. Pan, Nanotechnology, 2023, 34, 245601.
7. S. Huang, Z. Li, B. Wang, N. Zhu, C. Zhang, L. Kong, Q. Zhang, A. Shan and L. Li, ACS Appl. Mater. Interfaces, 2017, 9, 7249–7258.
8. Y. Wei, Z. Cheng and J. Lin, Chem. Soc. Rev., 2019, 48, 310.
9. W. Lv, L. Li, M. Xu, J. Hong, X. Tang, L. Xu, Y. Wu, R. Zhu, R. Chen and W. Huang, Adv. Mater., 2019, 31, 1900682.
10. L. Jiao, J. Y. R. Seow, W. S. Skinner, Z. U. Wang and H.-L. Jiang, Mater. Today, 2019, 27, 43–68.
11. Y. Shan, G. Zhang, Y. Shi and H. Pang, Cell Rep. Phys. Sci., 2023, 4, 101301.
12. S. Zhang, M. Wang, X. Wang, J. Song and X. Yang, Biosensors, 2024, 14, 42.
13. J. W. Oh, S. L and H. Han, Light Sci. Appl., 2023, 12, 226.
14. M. Fakhraei Ghazvini, M. Vahedi, S. Najafi Nobar and F. Sabouri, J. Environ. Chem. Eng., 2021, 9, 104790.
15. Q. Qian, P. A. Asinger, M. J. Lee, G. Han, K. M. Rodriguez, S. Lin, F. M. Benedetti, A. X. Wu, W. S. Chi and Z. P. Smith, Chem. Rev., 2020, 120, 8161–8266.
16. T. Jia, J. Environ. Chem. Eng., 2022, 10, 108300.
17. J. Ren, T. Li, X. Zhou, X. Dong, A. V. Shorokhov, M. B. Semenov, V. D. Krevchik and Y. Wang, Chem. Eng. J., 2019, 358, 30–39.
18. D. Zhang, Y. Xu, Q. Liu and Z. Xia, Inorg. Chem., 2018, 57, 4613–4619.
19. C. Zhang, B. Wang, W. Li, S. Huang, L. Kong, Z. Li and L. Li, Nat. Commun., 2017, 8, 1138.
20. K. Vighnesh, S. Wang, H. Liu and A. L. Rogach, ACS Nano, 2022, 16, 19618–19625.
21. D. Zhang, S. W. Eaton, Y. Yu, L. Dou and P. Yang, J. Am. Chem. Soc., 2015, 137, 9230–9233.
22. Y. Su, Q. Zeng, X. Chen, W. Ye, L. She, X. Gao, Z. Ren and X. Li, J. Mater. Chem. C, 2019, 7, 7548–7553.
23. P. Cottingham and R. L. Brutchey, Chem. Mater., 2018, 30, 6711–6716.
24. Y. Liu, P. Ghimire and M. Jaroniec, J. Colloid Interface Sci., 2019, 535, 122–132.
25. D. Tuncel and A. N. Ökte, J. Chem. Technol. Biotechnol., 2023, 98, 2925–2942.
26. Y. Jiang, R. Zhou, Z. Zhang, Z. Dong and J. Xu, J. Mater. Chem. C, 2023, 11, 2540–2551.
27. Z.-C. Kong, J.-F. Liao, Y.-J. Dong, Y.-F. Xu, H.-Y. Chen, D.-B. Kuang and C.-Y. Su, ACS Energy Lett., 2018, 3, 2656–2662.
28. 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.
29. X. Li, H. Liu, X. Jia, G. Li, T. An and Y. Gao, Sci. Total Environ., 2018, 621, 1533–1541.
30. S. Wan, M. Ou, Q. Zhong and X. Wang, Chem. Eng. J., 2019, 358, 1287–1295.
31. L. Zhang, C. Guo, T. Chen, Y. Guo, A. Hassan, Y. Kou, C. Guo and J. Wang, Appl. Catal. B Environ., 2022, 303, 120888.
32. Y. Tang, N. Rong, F. Liu, M. Chu, H. Dong, Y. Zhang and P. Xiao, Appl. Surf. Sci., 2016, 361, 133–140.
33. X. Dong, F. Li, N. Zhao, Y. Tan, J. Wang and F. Xiao, Catal. Lett., 2017, 147, 1235–1242.
34. W. Xiong, Y. Dong and A. Pan, Nanoscale, 2023, 15, 15619–15625.
35. Z. Dong, J. Zhou, Z. Zhang, Y. Jiang, R. Zhou and C. Yao, ACS Appl. Energy Mater., 2022, 5, 10076–10085.
36. H.-X. Zhang, Q.-L. Hong, J. Li, F. Wang, X. Huang, S. Chen, W. Tu, D. Yu, R. Xu, T. Zhou and J. Zhang, Angew. Chem., Int. Ed., 2019, 58, 11752–11756.
37. X. Yang, S. Wang, N. Yang, W. Zhou, P. Wang, K. Jiang, S. Li, H. Song, X. Ding, H. Chen and J. Ye, Appl. Catal. B Environ., 2019, 259, 118088.
38. 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.
39. H. Ran, D. Wu, W. Chen, Y. Liu, L. Gao, J. Zhou, B. Gao, J. Lai, H. Luo, F. Kuang, M. Mo, Z. Luo, F. Dong, H. Ma, Q. Zhang, F. Ling, B. Sun and X. Tang, Appl. Catal. B Environ., 2024, 352, 124048.
40. Q.-L. Mo, J.-L. Li, S.-R. Xu, K. Wang, X.-Z. Ge, Y. Xiao, G. Wu and F.-X. Xiao, Adv. Funct. Mater., 2023, 33, 2210332.
41. Y. Li, Q. Wu, Y. Chen, R. Zhang, C. Li, K. Zhang, M. Li, Y. Lin, D. Wang, X. Zou and T. Xie, Appl. Catal. B Environ., 2021, 290, 120058.
42. W. Tao, Q. Tang, J. Hu, Z. Wang, B. Jiang, Y. Xiao, R. Song and S. Guo, J. Mater. Chem. A, 2023, 11, 24999–25007.
43. J. Wei, S. Zhang, J. Sun, T. Liang, Z. Li, Z. Li, X. Yi, R. Xiong, J. Deng, Z. Yu, S. Wang and Y. Hou, J. Colloid. Interface Sci., 2023, 629, 92–102.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4ta07190e

---