Photocatalytic CO₂ reduction on Cu single atoms incorporated in ordered macroporous TiO₂ toward tunable products

Abstract

The photocatalytic conversion of CO₂ to hydrocarbons is a fascinating strategy to defuse the growing energy and environmental dilemmas, but there are great challenges in improving photocatalytic efficiency and tuning product selectivity. Both are of equal importance to increase the yield of the desired product and clarify the photocatalytic CO₂ reduction mechanism. Herein, a Cu single-atom-incorporated three-dimensional-ordered macroporous TiO₂ (Cu_0.01/3DOM-TiO₂) photocatalyst was synthesized using a template-assisted in situ pyrolysis method. The Cu single atoms are uniformly anchored in a 3DOM TiO₂ matrix, which not only broadens the light absorption range but also provides specific active sites for the adsorption and transformation of CO₂ molecules. The photocatalytic CO₂ reduction reaction was conducted in gas–solid and liquid–solid systems, respectively, to explore the effects of reaction modes on the CO₂ conversion efficiency and product selectivity. The results indicate that the photocatalytic CO₂ reduction reaction in gas–solid system mainly produces methane (CH₄), with a high selectivity of 83.3% and a formation rate of 43.15 μmol g⁻¹ h⁻¹. In contrast, the main product in the liquid–solid system is ethylene (C₂H₄), with a selectivity of 58.4% and a formation rate of 6.99 μmol g⁻¹ h⁻¹. The Cu_0.01/3DOM-TiO₂ photocatalyst shows superior activity and selectivity in the gas–solid system while favorably producing C₂H₄ in the liquid–solid system. The possible photocatalytic mechanisms of CO₂ reduction in the two different reaction systems are discussed according to in situ infrared spectroscopy. This work provides some new information on promoting photocatalytic CO₂ reduction to desirable products by the rational design of photocatalysts as well as tuning the reaction conditions.

---

1. Introduction

Photosynthetic organisms harness solar energy to convert CO₂ and H₂O into carbohydrates and oxygen, which achieves energy transformation in nature and maintains the carbon–oxygen balance. Enlightened by natural photosynthesis, artificial photosynthesis for CO₂ reduction with H₂O powered by solar irradiation has attracted extensive interest, which is regarded as a double-benefit approach that converts greenhouse gas to chemical fuels and valuable chemicals by harvesting solar energy.¹ However, the reactivity and product selectivity of photocatalytic CO₂ reduction are always restricted by three main obstacles: (1) the difficult adsorption of reactants on photocatalysts in heterogeneous catalytic systems, (2) the high energy requirement for C–O bond activation up to 750 kJ mol⁻¹, and (3) the various intermediates and complex pathways of the CO₂ transformation process.^2,3 Hence, the rational construction of heterogeneous photocatalysts is imperative for improving the reactivity and product selectivity of CO₂ reduction, which requires the meticulous design and masterly integration of a light-harvesting component and catalytically active sites.⁴

In recent years, single-atom (SA) catalysts have emerged as new options for the design and preparation of cost-effective and high-efficiency photocatalysts, which provide 100% active metal dispersion and thus maximize metal utilization.⁵ The SA catalysts can be designed on the atomic scale by rationally embellishing the isolated specific metal-atom sites on the support materials.⁶ Based on the inherent functions of light absorption and photo-generated carriers supplying the support materials, SA active sites further offer catalytic centers for promoting charge separation/transfer and reactant adsorption/activation.⁷ Moreover, the definite active centers of SA photocatalysts enable improved understanding of the fundamental mechanism of photocatalysis, and allow us to draw more precise structure–performance correlations. Therefore, various metal cocatalysts including noble metals, transition metals and rare Earth metals, have been prepared in SA sites to increase the photocatalytic efficiency.^8–12 Among them, copper (Cu) has been identified as one of the most promising cocatalysts to convert CO₂ into various hydrocarbons owing to its optimum binding ability with CO₂ and reaction intermediates.^13–16 The individual Cu atoms immobilized on light-harvesting semiconductors can act as excellent active sites for activating CO₂ molecules and stabilizing reaction intermediates in the photocatalytic CO₂ reduction reaction.^17,18 Moreover, the introduction of highly dispersed Cu heteroatoms in light-harvesting semiconductors can broaden the light-adsorption range and enhance the photoexcited charge separation efficiency.¹⁹ Although numerous studies have highlighted the excellent catalytic activity of Cu SAs in electrocatalytic/photocatalytic CO₂ reduction,^16,20,21 the controllable and large-scale production of heterogeneous photocatalysts with Cu SAs is still challenging. Most strategies to anchor Cu SAs on light-harvesting semiconductors are explored as post-treatment methods (e.g. impregnation and photodeposition), where the Cu SAs tend to aggregate into nanoparticles or leach from supports during photocatalytic reactions because of the weak combination.^22,23 Hence, the in situ incorporation of highly stable Cu SAs in semiconductors is desirable. Meanwhile, the morphological features of the semiconductor support are a crucial factor to create stable Cu single atoms as well as heterogeneous photocatalysts with high performance, where the support not only determines the availability and stability of Cu SA sites but also supplies photogenerated electrons for CO₂ reduction. Three-dimensional ordered macroporous (3DOM) materials with an interpenetrating porous structure and high porosity can greatly facilitate CO₂ diffusion and enhance the light utilization simultaneously.^24,25 Moreover, the large specific surface area of the 3DOM support can afford more accessible Cu SA sites for CO₂ photoreduction.²⁶ Although some metal-modified 3DOM TiO₂ photocatalysts have been reported, the formation of metal nanoparticles with large sizes generally results in limited photocatalysis.^27,28 Our recent study indicated that Pd single atoms could be in situ anchored in 3DOM CeO₂via a template-assisted pyrolysis method owing to the strong metal–support interaction.²⁹ Inspired by this, the masterly integration of catalytically active sites (Cu SAs) and a light-harvesting semiconductor (3DOM TiO₂) is expected to achieve.

In addition, tuning the selectivity of CO₂ reduction reaction to the desired product is of great significance but remains a huge challenge.^30,31 Tan and coworkers found that the ethylene selectivity of electrocatalytic CO₂ reduction with the same catalyst could be tuned by modulating the local CO₂ concentration, where the kinetics of CO₂ reduction was substantially influenced by the modulation of CO₂ mass transport.³² Wang et al. reported that the introduction of N₂ into CO₂ could control the *CO coverage on Cu–Au bimetallic catalysts and thus facilitate *CO protonation for methane formation.³³ However, the effects of reaction conditions on product selectivity of CO₂ photoreduction are rarely studied. The photocatalytic CO₂ reduction reaction is commonly conducted in either a liquid–solid or a gas–solid system. Actually, the reaction system is a non-negligible factor influencing the product selectivity of CO₂ photoreduction. Therefore, exploring the effect of reaction systems on the product selectivity of photocatalytic CO₂ reduction is as important as the development of efficient photocatalysts. In this work, a heterogeneous photocatalyst of 3DOM TiO₂ doped with Cu single atoms (Cu_0.01/3DOM-TiO₂) is controllably synthesized via a facile template-assisted pyrolysis method. Taking this catalyst as a research model, the photocatalytic CO₂ reduction performance is studied in gas–solid and liquid–solid reaction systems, respectively, to explore the effects of reaction modes on the catalytic activity and product selectivity. Interestingly, the Cu_0.01/3DOM-TiO₂ photocatalyst exhibits higher activity and selectivity for methane (CH₄) production in the gas–solid system, while possessing a favorable capability to convert CO₂ into ethylene (C₂H₄) in the liquid–solid system. The possible photocatalytic mechanisms of CO₂ reduction in both catalytic systems are discussed according to in situ diffused reflection infrared Fourier transform spectroscopy (DRIFTS) measurements.

2. Experimental

2.1. Chemicals

Titanium butoxide (C₁₆H₃₆O₄Ti, ≥99.0%), anhydrous cupric chloride (CuCl₂, 98%), styrene (>99.5%, GC), potassium persulfate (K₂S₂O₈, AR, 99.5%), sodium hydroxide (NaOH, 97%), acetic acid glacial (CH₃COOH, ≥99.8%), ethanol (CH₃CH₂OH, ≥99.7%), methanol (CH₃OH, 99.5%), sodium sulfate anhydrous (Na₂SO₄, 99%), potassium iodide (KI, 99%), potassium hydrogen phthalate (C₈H₅KO₄, 99.8%), N₂ (≥99.99%), CO₂ (≥99.99%) and argon gas (≥99.99%) were obtained commercially and used without further purification.

2.2. Synthesis of Cu/3DOM-TiO₂ photocatalysts

Firstly, the monodisperse polystyrene (PS) sub-microspheres with diameters of about 260 nm were prepared using an emulsifier-free emulsion polymerization process and assembled into the PS colloidal crystal templates (CCTs) by centrifugation.²⁹ The as-prepared CCTs were immersed in methanol for 30 min and then dried for a moment to remove the superfluous methanol. Subsequently, the Cu/3DOM-TiO₂ photocatalysts were prepared using a template-assisted pyrolysis method. Specifically, methanol (0.5 mL), acetic acid glacial (0.5 mL), titanium butoxide (2.72 g) and a certain amount of CuCl₂ were mixed by vigorous stirring for 2 h to form a homogeneous precursor solution. The pretreated CCTs were immersed in the precursor solution under vacuum treatment. After being subjected to a vacuum for 1 h, the CCTs impregnated with precursor solution were collected, dried at room temperature, and calcined in air at 550 °C for 3 h (heating rate 1 °C min⁻¹) to obtain the Cu/3DOM-TiO₂ photocatalysts. The Cu content in the Cu/3DOM-TiO₂ photocatalysts was adjusted by changing the dosage of CuCl₂, and the as-synthesized photocatalyst was denoted as Cu_x/3DOM-TiO₂, where x represents the molar ratio of Cu to Ti based on the dosages of CuCl₂ and titanium butoxide. For example, the Cu_0.01/3DOM-TiO₂ photocatalyst was prepared by adding 10.8 mg CuCl₂ to the precursor solution. In addition, the other Cu_x/3DOM-TiO₂ samples (x = 0.0075, 0.0125, 0.02, 0.03, 0.04 and 0.05) were also synthesized by adding the corresponding dosage of CuCl₂. The actual Cu contents of Cu_0.0075/3DOM-TiO₂, Cu_0.01/3DOM-TiO₂, Cu_0.0125/3DOM-TiO₂, Cu_0.02/3DOM-TiO₂, Cu_0.03/3DOM-TiO₂, Cu_0.04/3DOM-TiO₂ and Cu_0.05/3DOM-TiO₂ were measured by atomic absorption spectroscopy (AAS) as 0.51 wt%, 0.69 wt%, 1.03 wt%, 1.33 wt%, 1.98 wt%, 2.75 wt% and 3.38 wt%, respectively.

2.3. Synthesis of control samples

Several control samples including 3DOM-TiO₂, Cu_0.01/TiO₂ and TiO₂ were prepared using a similar procedure to that for preparing the Cu_0.01/3DOM-TiO₂ photocatalyst, without adding CuCl₂, CCTs, CuCl₂ and CCTs, respectively.

2.4. Characterization

The crystalline components of photocatalysts were analyzed using a powder X-ray diffractometer (XRD, DX-2700) with Cu Kα radiation. The actual Cu contents of the Cu_x/3DOM-TiO₂ photocatalysts were measured on an atomic absorption spectrophotometer (Shimadzu AA-7000). The microstructures were observed using field emission scanning electron microscopy (FESEM, ZEISS Ultra-55) operated at 3.0 kV and transmission electron microscopy (TEM, JEOL-2010) working at 200 kV. Nitrogen adsorption/desorption isotherms were recorded on a Micromeritics ASAP 2460 apparatus at 77 K, to determine the specific surface areas and the pore size distributions of the photocatalysts. Thermogravimetric (TG) analysis was performed on a thermal gravimetric analyzer (TG209F3 Tarsus, Netzsch). Raman spectra were acquired on a Horiba Scientific LabRAM HR Evolution apparatus with a 514 nm laser as the excitation source. X-ray photoelectron spectra (XPS) were measured using a Thermo Scientific K-Alpha X-ray photoelectron spectrometer, and the binding energies were calibrated with a C 1s peak at 284.8 eV as the reference. UV–vis diffuse reflectance spectra (DRS) were taken on a Shimadzu UV-2600 spectrophotometer with BaSO₄ as the reference. Photoluminescence (PL) spectra were measured on a HORIBA Fluoromax-4 spectrophotometer at an excitation wavelength of 350 nm. The radical intermediates generated during photocatalytic CO₂ reduction were detected using in situ electron spin resonance (ESR) on a Bruker model JEOL JES-FA200 spectrometer, with 5,5-dimethyl-1-pyrroline-N-oxide (DMPO) as a radical-trapping reagent. The X-ray absorption of fine structures (XAFS) measurements of the Cu K-edge were conducted at the BL14W1 station at the Shanghai Synchrotron Radiation Facility (SSRF), China. The spectra were recorded in fluorescent mode using a Si (311) double-crystal monochromator and a 32-element Ge solid-state detector. The electron storage ring was operated at 3.5 GeV. Cu foil and CuO were employed as the reference samples. The XAFS data were processed using Athena and Artemis software, according to the standard procedures.

2.5. Photocatalytic CO₂ reduction tests

The photocatalytic CO₂ reduction reactions were carried out in a top-irradiated reactor with a volume of 170 mL. The simulated sunlight with a main irradiation range of 320–780 nm was provided by a Xe lamp (PLS-SXE300, Beijing Perfectlight) equipped with a specific reflector, and the optical power density irradiated on the reaction solution was 200 mW cm⁻². The photocatalytic CO₂ reduction reaction was conducted in gas–solid and liquid–solid systems, respectively. For the gas–solid reaction mode, the slurry prepared by dispersing 5 mg photocatalyst in 1 mL ethanol was coated on a glass plate (2 × 2 cm) and then dried at 60 °C. The glass-supported photocatalyst was placed at the bottom of the reactor, and 1 mL of deionized water was dripped beside the glass plate. After CO₂ gas was blown into the reactor and absorption equilibrium was achieved in dark, the photocatalytic CO₂ reduction reaction was started under simulated sunlight. For the liquid–solid reaction mode, the reaction suspension containing 5 mg of photocatalyst and 30 mL of water was transformed in the reactor, which was bubbled with CO₂ gas for 20 min. After the sealed reactor had been kept in dark for 30 min, the photocatalytic CO₂ reduction reaction was started under simulated sunlight from a Xe lamp. The photocatalytic reactions were conducted for 4 h to obtain the average generation rates of products. The gas products (CO, CH₄, C₂H₄) were quantitatively analyzed using a gas chromatograph (GC 9790 II, FuLi) equipped with both a thermal conductivity detector (TCD) and a flame ionization detector (FID). The liquid products were identified using ¹H nuclear magnetic resonance (NMR) spectrometry in D₂O, which was performed on a Bruker AVANCE AV 400 MHz spectrometer. For the recycling experiments, the reaction system was re-blown or re-bubbled with CO₂ gas after each reaction cycle and then irradiated for the next reaction cycle.

2.6. In situ DRIFTS measurement

The in situ diffused reflection infrared Fourier transform spectroscopy (DRIFTS) analysis was performed on a Bruker TENSOR II FTIR spectrometer equipped with a mercury–cadmium–tellurium (MCT) detector cooled with liquid nitrogen. For the gas–solid reaction mode, 30 mg of the photocatalyst was placed in the test chamber, which was then subjected to a vacuum. After the background spectrum was collected from 400 to 4000 cm⁻¹, humid CO₂ gas was introduced into the test chamber to achieve adsorption–desorption equilibrium in the dark for 30 min. Subsequently, the time-dependent DRIFTS spectra were recorded upon irradiation from the Xe lamp. For the liquid–solid reaction mode, the operations were the same as those described above, except that a drop of deionized water was dripped beside the photocatalyst on the sample holder.

3. Results and discussion

3.1. Structures and properties of photocatalysts

The Cu_x/3DOM-TiO₂ photocatalysts were synthesized via a template-assisted pyrolysis method as illustrated in Fig. 1a. The monodisperse PS sub-microspheres were assembled into colloidal crystal templates which were impregnated with the precursor solution containing Cu²⁺ ions and titanium butoxide. The Cu_x/3DOM-TiO₂ photocatalysts with different Cu contents were obtained via calcination of the colloidal crystal templates loaded with precursors in air at 550 °C for 3 h. The thermogravimetric (TG) analysis (Fig. S1, ESI†) indicated that the PS-sphere template decomposes at 400 °C in air and the precursors are completely converted into the final material below 500 °C. Therefore, it is reasonable to conduct the calcination at 550 °C for 3 h, during which the individual Cu single atoms are in situ anchored in the 3DOM-TiO₂ matrix. The XRD patterns of the typical Cu_0.01/3DOM-TiO₂ photocatalyst and the control samples (3DOM-TiO₂, Cu_0.01/TiO₂ and TiO₂) are shown in Fig. 1b. The photocatalysts obtained from the template-assisted synthesis (Cu_0.01/3DOM-TiO₂ and 3DOM-TiO₂) consist of anatase TiO₂ (JCPDS no. 01-562) accompanied by a small quantity of rutile TiO₂ (JCPDS no. 01-1292), while the Cu_0.01/TiO₂ and TiO₂ samples synthesized by the direct pyrolysis of precursors without the PS template present only the diffraction peaks of anatase TiO₂. The mass fractions of rutile TiO₂ (W_Rutile) in Cu_0.01/3DOM-TiO₂ and 3DOM-TiO₂ are estimated to be 13.86 wt% and 13.80 wt%, respectively, according to the previously reported formula:^34,35W_Rutile = A_Rutile/(0.884 A_Anatase + A_Rutile), where A_Anatase and A_Rutile are the integrated intensities of the anatase (101) and rutile (110) XRD peaks. The previous study demonstrated that the calcination process assisted by a PS colloidal crystal template could produce mixed phases of anatase and rutile.²⁷ As a light-harvesting material, TiO₂ containing both anatase and rutile phases is desirable because the anatase/rutile junction facilitates spatial charge separation, allowing more photogenerated charge carriers to migrate to the surface and participate in redox reactions.³⁶ The diffraction peaks of Cu species are not observed in the XRD patterns of both the Cu_0.01/3DOM-TiO₂ and the Cu_0.01/TiO₂ samples due to the high dispersity and/or low content of Cu. Compared with TiO₂ and 3DOM-TiO₂, the typical diffraction peaks of Cu_0.01/3DOM-TiO₂ and Cu_0.01/TiO₂ present obvious shifts to lower 2θ positions (Fig. S2, ESI†), which are caused by the doping of Cu cations with larger ionic radii (Cu²⁺, 0.73 Å; Cu⁺, 0.77 Å; and Ti⁴⁺, 0.61 Å) into TiO₂.^37,38 In addition, the anatase (101) diffraction peak of Cu_0.01/3DOM-TiO₂ and 3DOM-TiO₂ presents further shifts towards lower 2θ positions compared with those of Cu_0.01/TiO₂ and TiO₂, possibly because the special 3DOM structures result in little lattice distortion.^39,40 The SEM images (Fig. 1c; Fig. S3a, ESI†) affirm that the Cu_0.01/3DOM-TiO₂ photocatalyst exhibits a honeycomb morphology with 3D ordered macroporous structures, where the spherical voids have an average diameter of 160 nm and the walls have an average thickness of about 10 nm. These spherical voids are well arranged and interconnected. The 3DOM-TiO₂ sample (Fig. S3b, ESI†) presents a similar honeycombed morphology to that of the Cu_0.01/3DOM-TiO₂ photocatalyst. In contrast, Cu_0.01/TiO₂ (Fig. S3c, ESI†) and TiO₂ (Fig. S3d, ESI†) synthesized by the direct pyrolysis of precursors without a PS template are irregular bulks without porosity. The TEM image shown in Fig. 1d indicates that the 3DOM frameworks are constructed from numerous TiO₂ small particles. The HRTEM image (Fig. 1e) reveals the crystalline phase boundary of anatase and rutile, where the characteristic spacings of 0.352 and 0.229 nm are assignable to the (101) lattice of anatase TiO₂ and the (200) lattice of rutile TiO₂, respectively. The crystalline lattices of CuO or Cu₂O are not found in the HRTEM image of Cu_0.01/3DOM-TiO₂, suggesting that Cu element exists as highly dispersed species in this material.

Fig. 1 (a) Schematic diagram of the synthesis of Cu_0.01/3DOM-TiO₂ photocatalyst. (b) XRD patterns of different samples; (c) SEM, (d) TEM and (e) HRTEM images of Cu_0.01/3DOM-TiO₂.

The specific surface areas and pore size distributions of Cu_0.01/3DOM-TiO₂, 3DOM-TiO₂ and Cu_0.01/TiO₂ were determined using N₂ adsorption–desorption measurements (Fig. 2; Fig. S4, ESI†). Both Cu_0.01/3DOM-TiO₂ and 3DOM-TiO₂ samples display the typical type IV isotherm with a hysteresis loop, and the dominant adsorption of both samples can be identified as macropores corresponding to the step at P/P₀ = 0.8–0.995 (Fig. 2a).⁴¹ In contrast, Cu_0.01/TiO₂ presents an extremely weak N₂ adsorption capacity (Fig. S4, ESI†) due to its large bulk morphology without porosity. Cu_0.01/3DOM-TiO₂ presents a specific surface area of 52.80 m² g⁻¹ and a pore volume of 0.28 cm³ g⁻¹, which are higher than those of 3DOM-TiO₂ (Table S2, ESI†). The Cu_0.01/TiO₂ sample shows a very low specific surface area (3.74 m² g⁻¹) and a negligible pore volume (0.0014 cm³ g⁻¹). The large surface area and pore volume of Cu_0.01/3DOM-TiO₂ are conducive to the sufficient exposure of active sites and the easy diffusion of CO₂ and H₂O molecules during the photocatalytic CO₂ reduction reaction.⁴²

Fig. 2 (a) N₂ adsorption–desorption isotherms and (b) pore size distribution plots of Cu_0.01/3DOM-TiO₂ and 3DOM-TiO₂.

The Raman spectroscopy analysis shed light on the variations in local structures of TiO₂ in different photocatalysts. As shown in Fig. 3a, the Raman spectra of the three samples (Cu_0.01/3DOM-TiO₂, 3DOM-TiO₂ and Cu_0.01/TiO₂) exhibit five representative peaks of anatase TiO₂, which can be ascribed to the E_g(1), E_g(2) and E_g(3) modes (144, 197 and 639 cm⁻¹), the B_1g mode (397 cm⁻¹) and the doublet of the A_1g/B_1g mode (517 cm⁻¹), respectively.^41,43 Compared with Cu_0.01/TiO₂, 3DOM-TiO₂ shows a slight shift and broadening in the characteristic A_1g and B_1g vibration peaks due to the increased structural disorder resulting from the formation of the 3DOM structure. Cu_0.01/3DOM-TiO₂ presents further broadening of the characteristic peaks because of Cu doping in 3DOM TiO₂, but no peaks of copper oxide or copper hydroxide are observed. Moreover, both 3DOM-TiO₂ and Cu_0.01/3DOM-TiO₂ show a weak peak at 448 cm⁻¹ (inset in Fig. 3a), which corresponds to a characteristic vibration mode of rutile TiO₂, further confirming the coexistence of the anatase and rutile phases in the two samples.⁴⁴ The XPS technique was employed to acquire information on the surface elemental compositions and chemical states of the photocatalysts. For the three samples (Cu_0.01/3DOM-TiO₂, 3DOM-TiO₂ and Cu_0.01/TiO₂), each of the Ti 2p spectra (Fig. 3b) shows two characteristic peaks at binding energies of 458.6 and 464.3 eV, which are assignable to Ti 2p_3/2 and 2p_1/2 of Ti⁴⁺, respectively.^45,46 Compared with Cu_0.01/TiO₂, the Ti 2p_3/2 and 2p_1/2 peaks of both 3DOM-TiO₂ and Cu_0.01/3DOM-TiO₂ present a 0.1 eV positive shift, which is probably caused by the lattice distortions of 3DOM-TiO₂.⁴⁷ Two peaks are deconvoluted from each O 1s spectrum (Fig. 3c), where the main peak around 529.8 eV is assigned to the lattice oxygen of TiO₂ and the minor peak around 531.0 eV is attributed to the surface chemisorbed oxygen on oxygen-vacancy (V_O) sites.^48–50 Cu_0.01/3DOM-TiO₂ and 3DOM-TiO₂ have higher oxygen-vacancy contents (34.63% and 30.74%) than Cu_0.01/TiO₂ (24.87%), because the high specific surface area of 3DOM-TiO₂ exposes more surface Ti sites with unsaturated coordination.⁵¹ In Fig. 3d, two characteristic peaks centered at binding energies of 932.3 and 952.0 eV appear in the Cu 2p spectra of both Cu_0.01/3DOM-TiO₂ and Cu_0.01/TiO₂, which are approximate to the Cu 2p_3/2 and 2p_1/2 of Cu⁺ ions,^52,53 indicating that the Cu single atoms incorporated in TiO₂ have an electronic state close to that of the Cu⁺ cation. In addition, another pair of minor peaks centered at 934.0 and 954.3 eV deconvoluted from the Cu 2p spectrum of Cu_0.01/TiO₂ can be assigned to a Cu²⁺ cation with a higher oxidation state, probably due to the formation of CuO clusters.^54,55 Compared with 3DOM-TiO₂, the lack of peak shifts in the XPS spectra of Cu_0.01/3DOM-TiO₂ may be ascribable to the formation of the atomically dispersed Cu^δ+ in the lattice of 3DOM-TiO₂ during the high-temperature calcination process as well as the low Cu content. Furthermore, the electronic structure and local coordination environment of Cu atoms in the Cu_0.01/3DOM-TiO₂ photocatalyst were studied using X-ray absorption fine structure (XAFS) analysis, in comparison with Cu foil and CuO. Fig. 3e shows the normalized Cu K-edge X-ray absorption near-edge structure (XANES) spectra, where the adsorption edge of Cu_0.01/3DOM-TiO₂ is located between those of Cu foil and CuO, indicating that the electronic state of Cu in Cu_0.01/3DOM-TiO₂ lies between those of Cu⁰ and Cu²⁺.⁵² The adsorption peak of Cu_0.01/3DOM-TiO₂ is centered at 8997.1 eV and shifts a little to a lower energy compared with CuO, meaning a higher electron density of Cu single atoms in this photocatalyst than for Cu²⁺ in CuO.⁵⁶ The coordination configuration of Cu atoms in Cu_0.01/3DOM-TiO₂ is further revealed by the extended X-ray adsorption fine structure (EXAFS) spectrum (Fig. 3f). The Cu K-edge EXAFS spectrum of Cu_0.01/3DOM-TiO₂ presents two characteristic distances. The major peak centered at 1.4 Å corresponds to the direct binding of Cu atoms with lattice oxygen (Cu–O), while the minor peak located at 2.4 Å is assignable to Cu–Ti coordination in the TiO₂ environment.⁵⁷ No peak of Cu–Cu metallic binding is observed, indicating that the Cu atoms are individually doped into the lattice of TiO₂ and atomically distributed in the 3DOM-TiO₂ support.

Fig. 3 (a) Raman spectra of different photocatalysts. XPS spectra of (b) Ti 2p, (c) O 1s, and (d) Cu 2p of different photocatalysts. (e) Cu K-edge XANES and (f) EXAFS spectra of Cu_0.01/3DOM-TiO₂ with the reference samples of Cu foil and CuO.

High light-harvesting and charge separation efficiencies are fundamental requirements for efficient photocatalysis. The photoabsorption properties of bulk TiO₂, 3DOM-TiO₂, Cu_0.01/TiO₂ and Cu_0.01/3DOM-TiO₂ were measured using UV–vis diffuse reflectance spectroscopy (UV–vis DRS), as shown in Fig. 4a. The bulk TiO₂ shows absorption only in the ultraviolet region (λ < 400 nm) corresponding to the intrinsic transition from valence band (VB) to conduction band (CB). 3DOM-TiO₂ exhibits a great enhancement in its ultraviolet-harvesting ability owing to the unique 3DOM structure as well as additional absorption from 400 nm to longer wavelengths due to the formation of more oxygen vacancies.⁵⁸ With the introduction of Cu atoms, the photoabsorption above 400 nm becomes more conspicuous, suggesting the presence of more V_O defect states in TiO₂.¹⁹ Moreover, the Cu-incorporated TiO₂ samples (Cu_0.01/TiO₂ and Cu_0.01/3DOM-TiO₂) exhibit a near-infrared absorption band near 800 nm, corresponding to d–d transitions of the dopant Cu cations.⁵⁹ For Cu_0.01/TiO₂, the photoabsorption decreases in the ultraviolet region while increasing greatly in the visible region, which is attributable to the formation of CuO on TiO₂. Photographs of the different photocatalysts are provided in Fig. 4a and Fig. S12 (ESI†), respectively. Typically, the Cu_0.01/3DOM-TiO₂ photocatalyst is light yellow while the Cu_0.01/TiO₂ sample is dark green. The color of the Cu_x/3DOM-TiO₂ photocatalysts changes from light yellow to brown as the Cu content increases gradually (Fig. S12, ESI†). Photoluminescence (PL) spectroscopy is employed to evaluate the separation efficiency of the photogenerated electron–hole pairs under excitation of 300 nm. The Cu_0.01/3DOM-TiO₂ photocatalyst exhibits the weakest PL emission (Fig. 4b), suggesting the most efficient separation of photogenerated charge carriers.⁶⁰ The photocurrent responses and electrochemical impedance spectra (EIS) of photocatalysts were measured to further verify their photoexcited charge separation and transfer efficiencies. As shown in Fig. 4c, Cu_0.01/3DOM-TiO₂ shows the highest photocurrent response with a light on/off mode, reflecting the most efficient separation and transfer of photoexcited charge carriers. The EIS Nyquist plot of Cu_0.01/3DOM-TiO₂ presents the smallest arc radius (Fig. 4d), corresponding to the fastest charge transfer. These results demonstrate that the incorporation of Cu single atoms into 3DOM-TiO₂ affords outstanding superiority in enhancing light absorption and photogenerated charge separation/transfer efficiency. The highly dispersed Cu single-atom incorporation in TiO₂ enables efficient electron transfer via the Cu⁺–Cu²⁺ shift.²² The unique structure and properties of Cu_0.01/3DOM-TiO₂ make it a fascinating photocatalyst for CO₂ conversion.

Fig. 4 (a) Kubelka–Munk transformed UV–vis absorption spectra, (b) photoluminescence spectra, (c) transient photocurrent responses, (d) electrochemical impedance spectra of different photocatalysts.

3.2. Photocatalytic CO₂ reduction performance

Two typical heterogeneous photocatalytic reaction systems of gas–solid and liquid–solid modes were employed to evaluate the performance of the Cu_0.01/3DOM-TiO₂ photocatalyst toward CO₂ reduction with water under simulated sunlight. The effect of the Cu content in Cu_0.01/3DOM-TiO₂ on the photocatalysis was probed, as shown in Fig. S5,† and a volcano relationship was obtained for the photocatalytic CO₂ reduction performance relative to the Cu content. The Cu_0.01/3DOM-TiO₂ photocatalyst with a Cu/Ti molar ratio of 1% exhibited optimal performance in both reaction systems. Interestingly, the main product of the photocatalytic CO₂ reduction reaction in the gas–solid system was CH₄, while that in the liquid–solid system was C₂H₄ (Fig. S6, ESI†). Fig. 5a shows the time-dependent evolution of CH₄ and CO over the Cu_0.01/3DOM-TiO₂ photocatalyst in the gas–solid catalytic system, without H₂ evolution. As a main product, CH₄ was generated predominately from CO₂ reduction in the gas–solid catalytic reaction, with a remarkable production rate of 43.15 μmol g_cat⁻¹ h⁻¹ and a high selectivity of 83.3% (Fig. 5b). Notably, in the liquid–solid catalytic reaction, the Cu_0.01/3DOM-TiO₂ photocatalyst exhibits an excellent performance for CO₂ reduction to C₂H₄ (Fig. 5d), accompanied by the generation of CO and CH₄ by-products. The production rate of C₂H₄ reaches 6.99 μmol g_cat⁻¹ h⁻¹, with a selectivity of 58.4%, which is of great significance due to the crucial role of C₂H₄ in the chemical industry. No liquid product is detected in the reaction solution after 4 h of irradiation (Fig. S7, ESI†). In comparison, the Cu_0.01/3DOM-TiO₂ photocatalyst presents a higher activity and product selectivity for CO₂ reduction in the gas–solid catalytic reaction while producing more valuable C₂H₄ in the liquid–solid catalytic reaction. These results indicate that the photocatalytic CO₂ reduction reactions over Cu_0.01/3DOM-TiO₂ in the two catalytic systems actually follow different reaction pathways. The promotion effects of Cu doping and the 3DOM structure on CO₂ conversion are testified by the control experiments of photocatalytic reactions with 3DOM-TiO₂, Cu_0.01/TiO₂ and TiO₂, respectively. As shown in Fig. 5b and e, all the control catalysts without Cu doping or 3DOM structure present very poor photocatalysis toward CO₂ reduction in both the gas–solid and the liquid–solid systems, and none of them has the ability to convert CO₂ into CH₄ or C₂H₄, affirming that both Cu single-atom incorporation and the 3DOM structure are crucial for enhancing the photocatalytic CO₂ reduction performance. Subsequently, the durability of the Cu_0.01/3DOM-TiO₂ photocatalyst was examined by four cycles of the photocatalytic CO₂ reduction reaction with 4 h of irradiation for each cycle in the gas–solid and liquid–solid reaction systems (Fig. 5c and f), respectively. After each cycle, the photocatalyst was exposed to air for recovery. Both the excellent activity and the high selectivity were well maintained after four cycles of the photocatalytic reaction, indicating the good durability of the Cu_0.01/3DOM-TiO₂ photocatalyst. The XPS analyses (Fig. S8, ESI†) indicate that the surface composition of the Cu_0.01/3DOM-TiO₂ catalyst remains unchanged after the long-term photocatalytic CO₂ reduction reaction in both the gas–solid and liquid–solid systems, affirming the good stability of the photocatalyst. In addition, several control experiments of the photocatalytic CO₂ reduction reaction were carried out to provide evidence of the CO₂ conversion in both catalytic reaction systems. As shown in Fig. S9 (ESI†), the products are barely generated in the reaction system without CO₂, light or catalyst, indicating that all the products originate from the photocatalytic CO₂ reduction reaction. When CO₂ is replaced by N₂ in the reaction systems, small amounts of CO and CH₄ are produced at negligible rates over the Cu_0.01/3DOM-TiO₂ photocatalyst (Fig. S9b and S9d, ESI†), and the production stops after 2 h of irradiation. The slight production of CO and CH₄ can be ascribed to the reduction of surface-adsorbed CO₂ on the photocatalyst by air.

Fig. 5 Photocatalytic CO₂ reduction performance. (a–c) Gas–solid catalytic reaction: (a) product evolutions over Cu_0.01/3DOM-TiO₂, (b) generation rates of products over different photocatalysts, (c) cyclic test of Cu_0.01/3DOM-TiO₂. (d–f) Liquid–solid catalytic reaction: (d) product evolutions over Cu_0.01/3DOM-TiO₂, (e) generation rates of products over different photocatalysts, (f) cycling test of Cu_0.01/3DOM-TiO₂.

3.3. Photocatalytic reaction mechanism

The photoinduced charge transfer from the catalyst to the reactant molecules is critical for photocatalytic reactions. To probe the interfacial charge kinetics, the photocurrent responses of photocatalysts were measured in different atmospheres. The transient photocurrent recorded under an Ar atmosphere is assignable to the electron transfer from catalyst to electrode. Notably, the photocurrent measured in the CO₂-saturated electrolyte is lower than that measured in the Ar-saturated electrolyte for each photocatalyst (Fig. 6a and b), which is caused by the competitive interfacial electron transfer from the photocatalyst to the surface-adsorbed CO₂ molecules.⁶¹ In comparison, the photocurrents of the Cu-incorporated photocatalysts (Cu_0.01/3DOM-TiO₂ and Cu_0.01/TiO₂) decline more than those of the photocatalysts without Cu (3DOM-TiO₂ and TiO₂), indicating that the incorporation of Cu is greatly beneficial to the electron transfer from the photocatalyst to the CO₂ molecules. In particular, the Cu_0.01/3DOM-TiO₂ photocatalyst shows the greatest decrease in photocurrent measured in the CO₂-saturated electrolyte, corresponding to the most efficient electron delivery from the catalyst to the CO₂ molecules. It is understandable that the high charge separation and transfer efficiencies as well as the favorable CO₂ adsorption ability of the Cu_0.01/3DOM-TiO₂ photocatalyst jointly contributes to enhancing the electron provision to CO₂ molecules. The significantly enhanced electron transfer efficiency can accelerate the CO₂ activation and the subsequent proton-coupled electron transfer process for hydrocarbon generation.⁶² Furthermore, in situ ESR detection reveals the generation of radical intermediates during photocatalytic CO₂ reduction with water over the Cu_0.01/3DOM-TiO₂ catalyst (Fig. 6c). After 10 min of irradiation, the characteristic signal of the carbon-centered radical is obviously detected, affirming the activation of chemisorbed CO₂ on the photocatalyst.⁶³ Meanwhile, the characteristic signal of the hydroxyl radical (˙OH) is also observed in the photocatalytic reaction system, which should stem from the oxidation of surface-adsorbed H₂O molecules by consuming photogenerated holes (h⁺).⁶⁴ Accordingly, hydrogen peroxide (H₂O₂) as the product of the oxidation half reaction in both the photocatalytic CO₂ reduction systems is determined by iodometry (Fig. 6d).⁶⁵ The H₂O₂ production rates in gas–solid and liquid–solid reaction systems of the Cu_0.01/3DOM-TiO₂ photocatalyst are 16.85 and 15.12 μmol g_cat⁻¹ h⁻¹, respectively, which are much higher than those of other photocatalysts (Fig. S10, ESI†). The results demonstrate that H₂O₂ has been produced from the oxidation of surface-adsorbed H₂O by the photogenerated holes and also confirm the superior activity of the Cu_0.01/3DOM-TiO₂ photocatalyst toward CO₂ reduction with water.

Fig. 6 (a, b) Transient photocurrents of different catalysts measured in the Ar or CO₂ saturated electrolyte, (c) in situ ESR detection of the photocatalytic CO₂ reduction system with Cu_0.01/3DOM-TiO₂, (d) UV–vis absorption spectra of the iodometry solution of photocatalytic CO₂ reduction over different photocatalysts in gas–solid and liquid–solid systems.

In situ DRIFTS spectra collected from the photocatalytic reaction systems are helpful to identify the surface-bound intermediates of CO₂ transformation and provide crucial clues for understanding the reaction pathway. In situ DRIFTS measurements were carried out for both the gas–solid and the liquid–solid catalytic reaction systems under Xe lamp irradiation. In both cases, four broad absorption bands appear in the 3750–3550 cm⁻¹ region (Fig. S11, ESI†). The bands centered at 3727 and 3707 cm⁻¹ are attributable to the stretching vibrations of surface-adsorbed –OH groups on Ti⁴⁺ and Ti³⁺, respectively.⁶⁶ The bands peaking at 3626 and 3600 cm⁻¹ correspond to the stretching vibrations of O–H in HCO₃⁻ and H₂O, respectively.⁶⁷ In comparison, the intensities of the four absorption bands in the liquid–solid catalytic system are much higher than those of the gas–solid catalytic system due to the water-rich environment of the former. The infrared adsorption signals of the CO₂-reduction intermediates are displayed in Fig. 7. The peak at 1542 cm⁻¹ in the gas–solid catalytic system is attributable to the surface-bound monodentate carbonate (m-CO₃²⁻), which is located at 1536 cm⁻¹ for the liquid–solid catalytic system.^68,69 Some infrared adsorption peaks associated with the CO₂ transformation intermediates are obviously detected in both photocatalytic systems. The broad peak at 1448 cm⁻¹ (Fig. 7a) or 1444 cm⁻¹ (Fig. 7b) is ascribed to *COOH, a typical one-electron-reduction intermediate.⁷⁰ The emergence of the peaks at 1346 and 1559 cm⁻¹ (1350 and 1548 cm⁻¹) is indicative of the formation of *HCOOH via the protonation process of *COOH.¹⁸ The adsorption bands in the region of 1614–1774 cm⁻¹ are assignable to *CHO, which is an intermediate generated from *HCOOH.^18,71,72 Notably, the infrared signals of the gas–solid catalytic system (Fig. 7a) are much stronger than those of the liquid–solid system (Fig. 7b), due to the higher conversion rate of the former. The low solubility of CO₂ in water restricts the reaction rate of the photocatalytic CO₂ reduction in the liquid–solid system. In addition, several distinctive signals are observed in the in situ DRIFTS of both catalytic systems, which provide crucial evidence to reveal the differences in the photocatalytic CO₂ reduction pathways. For the gas–solid catalytic system (Fig. 7a), the sharp peak at 1455 cm⁻¹ appearing after prolonged irradiation is attributed to the *CH₃ group originating from *CHO protonation,^38,73 which is generally considered as a key intermediate for CH₄ production. For the liquid–solid catalytic system (Fig. 7b), the characteristic peak of *CH₃ is absent, but a new peak at 2076 cm⁻¹ is increasingly apparent with the irradiation, which can be assigned to the chemisorbed *CO.^70,74 The previous report proved that the intermolecular interaction between surface-adsorbed CO and interfacial water is critical for the formation of ethylene during the electrochemical reduction of CO.⁷⁵ The interfacial water is helpful to stabilize the CO dimer, a key intermediate in ethylene formation, by hydrogen bonding to the terminal oxygens of the adsorbed CO molecules. The porous structure of the Cu_0.01/3DOM-TiO₂ photocatalyst provides a water-rich local environment during CO₂ reduction in the liquid–solid system, which is beneficial to the stabilization and dimerization of the *CO intermediate.⁷⁶

Fig. 7 In situ DRIFTS of the photocatalytic CO₂ reduction with Cu_0.01/3DOM-TiO₂ catalyst in two different reaction systems: (a) gas–solid catalytic system, (b) liquid–solid catalytic system.

According to the above analyses, the photocatalytic mechanism and possible CO₂ reduction pathways over the Cu_0.01/3DOM-TiO₂ photocatalyst in two different systems are illustrated in Fig. 8. 3DOM-TiO₂ is excited to generate electron–hole pairs under simulated sunlight, and the photogenerated electrons are efficiently captured by the positively charged Cu^δ+ (0 < δ ≤ 2) single-atom sites. The enhanced electron density at Cu sites facilitates the conversion of CO₂ into hydrocarbons (CH₄ or C₂H₄) via the multi-proton coupled multi-electron reduction process. Specifically, the adsorption and activation of CO₂ molecules are achieved at the Cu^δ+ active sites of photocatalyst, where the adsorbed CO₂ undergoes a continuous hydrogenation process to form *COOH, followed by the generation of *HCOOH and *CHO intermediates. In the gas–solid photocatalytic system (Fig. 8a), CH₄ is produced as the main product via the hydrogenation process of *CHO. On the other hand, it is widely accepted that the generation of the C₂ product undergoes three different reaction paths: (1) a *COCO intermediate from *CO dimerization, (2) a *COCHO intermediate from the dimerization between *CO and *CHO, and (3) a *CHOCHO intermediate from *CHO dimerization.^71,77,78 In the liquid–solid system of photocatalytic CO₂ reduction (Fig. 8b), the coexistence of *CO and *CHO intermediates is a crucial clue to speculate that the formation of C₂H₄ follows the pathway of the *COCHO intermediate from dimerization between *CO and *CHO. Therefore, the possible reaction pathways of photocatalytic CO₂ reduction in the present gas–solid and liquid–solid systems are described in eqn (1)–(5) and (6)–(11), respectively, where * denotes the catalytically active sites on the photocatalyst and ↑ represents the release of the gas product.

* + CO₂ + e⁻ + H⁺ → *COOH (1)

*COOH + e⁻ + H⁺ → *HCOOH (2)

*HCOOH + e⁻ + H⁺ → *CHO + H₂O (3)

*CHO + 4e⁻ + 4H⁺ → *CH₃ + H₂O (4)

*CH₃ + e⁻ + H⁺ → CH₄↑ (5)

* + CO₂ + e⁻ + H⁺ → *COOH (6)

*COOH + e⁻ + H⁺ → *CO + H₂O (7)

*COOH + e⁻ + H⁺ → *HCOOH (8)

*HCOOH + e⁻ + H⁺ → *CHO + H₂O (9)

*CHO + *CO → *CO–CHO (10)

*CO–CHO + 7e⁻ + 7H⁺ → C₂H₄↑ + 2H₂O (11)

Fig. 8 The possible mechanisms of photocatalytic CO₂ reduction over Cu_0.01/3DOM-TiO₂ catalyst in two different reaction systems: (a) gas–solid catalytic system, (b) liquid–solid catalytic system.

The oxidation half-reactions in both the catalytic systems are the formation of H₂O₂via the consumption of photogenerated holes by water (eqn (12) and (13)).

*H₂O → H⁺ + *OH⁻ (12)

2*OH⁻ + 2h⁺ → H₂O₂ (13)

The photocatalytic CO₂ reduction reaction in liquid–solid system is restricted by the low solubility of CO₂ in water (33 mM at 1 atm and 25 °C) and the poor mass transfer of CO₂ in the liquid phase.⁷⁹ However, the diffusion coefficient of CO₂ in the gas phase (∼0.1 cm² s⁻¹) is about 10 000 times that in the liquid phase (∼1 × 10⁻⁵ cm² s⁻¹), which allows the rapid delivery of CO₂ molecules to the surface of photocatalysts with uncovered active sites. The adsorption and activation of CO₂ and H₂O can be implemented more efficiently in the gas–solid system, and thus the photocatalyst shows higher activity.

4. Conclusion

In summary, the Cu single atom-incorporated 3DOM-TiO₂ photocatalyst was synthesized using a low-cost and mass-production template-assisted pyrolysis strategy and further demonstrated for solar-powered CO₂ conversion into hydrocarbons. The 3DOM structure with large surface area maximizes the exposed sites for anchoring Cu single atoms in a TiO₂ matrix, which enhances the light absorption as well as affording more accessible active sites for CO₂ reduction. The photocatalytic CO₂ reduction with H₂O was conducted in the gas–solid and liquid–solid systems, respectively, to explore the effects of reaction modes on CO₂ conversion efficiency and product selectivity. The Cu_0.01/3DOM-TiO₂ photocatalyst exhibits higher activity and selectivity for CH₄ production from CO₂ reduction in the gas–solid system while showing a favorable performance for converting CO₂ to C₂H₄ in the liquid–solid system. Both the excellent activity and the selectivity are well maintained after a long irradiation time, indicating the good durability of the Cu_0.01/3DOM-TiO₂ photocatalyst. Compared with the 3DOM-TiO₂-based photocatalysts reported previously (Table S8, ESI†), the Cu_0.01/3DOM-TiO₂ catalyst exhibits superior photocatalytic CO₂ reduction performance without using any noble metal cocatalysts. The photocatalytic mechanisms in both reaction systems were investigated by in situ DRIFTS analyses. It is proposed that CH₄ is generated as the main product via the *CHO intermediate in the gas–solid system while C₂H₄ is produced via the dimerization of *CO and *CHO in the liquid–solid system. This work develops a scalable strategy for preparing Cu-SA-incorporated photocatalysts but also offers a new idea for regulating the product selectivity of CO₂ photoreduction by altering the reaction system.

Author contributions

Cong Chen: methodology, investigation, data curation, visualization, writing – original draft. Ting Wang: methodology, investigation, validation. Ke Yan: investigation, validation. Shoujie Liu: software, data curation. Yu Zhao: software, funding acquisition. Benxia Li: supervision, conceptualization, writing – review & editing, funding acquisition.

Conflicts of interest

The authors declare no conflict of interest.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (21471004), the Open Research Fund Program of Key Laboratory of Surface & Interface Science of Polymer Materials of Zhejiang Province (SISPM-2021-02) and the Science Foundation of Zhejiang Sci-Tech University (17062002-Y).

Notes and references

1. X. B. Li, Z. K. Xin, S. G. Xia, X. Y. Gao, C. H. Tung and L. Z. Wu, Semiconductor nanocrystals for small molecule activation via artificial photosynthesis, Chem. Soc. Rev., 2020, 49, 9028–9056.
2. C. Gao and Y. Xiong, Solar-driven artificial carbon cycle, Chin. J. Chem., 2021, 40, 153–159.
3. E. Gong, S. Ali, C. B. Hiragond, H. S. Kim, N. S. Powar, D. Kim, H. Kim and S.-I. In, Solar fuels: Research and development strategies to accelerate photocatalytic CO₂ conversion into hydrocarbon fuels, Energy Environ. Sci., 2022, 15, 880–937.
4. T. Kong, J. Low and Y. Xiong, Catalyst: how material chemistry enables solar-driven CO₂ conversion, Chem, 2020, 6, 1035–1038.
5. Z.-H. Xue, D. Luan, H. Zhang and X. W. Lou, Single-atom catalysts for photocatalytic energy conversion, Joule, 2022, 6, 92–133.
6. S. Shin, R. Haaring, J. So, Y. Choi and H. Lee, Highly durable heterogeneous atomic catalysts, Acc. Chem. Res., 2022, 55, 1372–1382.
7. L. Zeng and C. Xue, Single metal atom decorated photocatalysts: Progress and challenges, Nano Res., 2020, 14, 934–944.
8. J. Wei, F. L. Meng, T. Li, T. Zhang, S. Xi, W. L. Ong, X. Q. Wang, X. Zhang, M. Bosman and G. W. Ho, Spontaneous atomic sites formation in wurtzite CoO nanorods for robust CO₂ photoreduction, Adv. Funct. Mater., 2021, 32, 2109693.
9. X. Xiong, C. Mao, Z. Yang, Q. Zhang, G. I. N. Waterhouse, L. Gu and T. Zhang, Photocatalytic CO₂ reduction to CO over Ni single atoms supported on defect-rich zirconia, Adv. Energy Mater., 2020, 10, 2002928.
10. Z. Chen, S. Wu, J. Ma, S. Mine, T. Toyao, M. Matsuoka, L. Wang and J. Zhang, Non-oxidative coupling of methane: N-type doping of niobium single atoms in TiO₂–SiO₂ induces electron localization, Angew. Chem., Int. Ed., 2021, 60, 11901–11909.
11. L. Luo, L. Fu, H. Liu, Y. Xu, J. Xing, C.-R. Chang, D.-Y. Yang and J. Tang, Synergy of Pd atoms and oxygen vacancies on In₂O₃ for methane conversion under visible light, Nat. Commun., 2022, 13, 2930.
12. S. Ji, Y. Qu, T. Wang, Y. Chen, G. Wang, X. Li, J. Dong, Q. Chen, W. Zhang, Z. Zhang, S. Liang, R. Yu, Y. Wang, D. Wang and Y. Li, Rare-Earth single erbium atoms for enhanced photocatalytic CO₂ reduction, Angew. Chem., Int. Ed., 2020, 59, 10651–10657.
13. S. Ali, A. Razzaq, H. Kim and S.-I. In, Activity, selectivity, and stability of Earth-abundant CuO/Cu₂O/Cu⁰-based photocatalysts toward CO₂ reduction, Chem. Eng. J., 2022, 429, 131579.
14. Z. Sun, Y. Hu, D. Zhou, M. Sun, S. Wang and W. Chen, Factors influencing the performance of copper-bearing catalysts in the CO₂ reduction system, ACS Energy Lett., 2021, 6, 3992–4022.
15. T. Yang, X. Mao, Y. Zhang, X. Wu, L. Wang, M. Chu, C.-W. Pao, S. Yang, Y. Xu and X. Huang, Coordination tailoring of Cu single sites on C₃N₄ realizes selective CO₂ hydrogenation at low temperature, Nat. Commun., 2021, 12, 6022.
16. J. Wang, T. Heil, B. Zhu, C.-W. Tung, J. Yu, H. M. Chen, M. Antonietti and S. Cao, A single Cu-center containing enzyme-mimic enabling full photosynthesis under CO₂ reduction, ACS Nano, 2020, 14, 8584–8593.
17. 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, Isolated square-planar copper center in boron imidazolate nanocages for photocatalytic reduction of CO₂ to CO, Angew. Chem., Int. Ed., 2019, 58, 11752–11756.
18. Y. Yu, X. Dong, P. Chen, Q. Geng, H. Wang, J. Li, Y. Zhou and F. Dong, Synergistic effect of Cu single atoms and Au–Cu alloy nanoparticles on TiO₂ for efficient CO₂ photoreduction, ACS Nano, 2021, 15, 14453–14464.
19. Y. Zhao, Y. Zhao, R. Shi, B. Wang, G. I. N. Waterhouse, L. Z. Wu, C. H. Tung and T. Zhang, Tuning oxygen vacancies in ultrathin TiO₂ nanosheets to boost photocatalytic nitrogen fixation up to 700 nm, Adv. Mater., 2019, 31, 1806482.
20. Q. Zhao, C. Zhang, R. Hu, Z. Du, J. Gu, Y. Cui, X. Chen, W. Xu, Z. Cheng, S. Li, B. Li, Y. Liu, W. Chen, C. Liu, J. Shang, L. Song and S. Yang, Selective etching quaternary MAX phase toward single atom copper immobilized MXene (Ti₃C₂Cl_x) for efficient CO₂ electroreduction to methanol, ACS Nano, 2021, 15, 4927–4936.
21. H. Bao, Y. Qiu, X. Peng, J.-A. Wang, Y. Mi, S. Zhao, X. Liu, Y. Liu, R. Cao, L. Zhuo, J. Ren, J. Sun, J. Luo and X. Sun, Isolated copper single sites for high-performance electroreduction of carbon monoxide to multicarbon products, Nat. Commun., 2021, 12, 238.
22. Y. Zhang, J. Zhao, H. Wang, B. Xiao, W. Zhang, X. Zhao, T. Lv, M. Thangamuthu, J. Zhang, Y. Guo, J. Ma, L. Lin, J. Tang, R. Huang and Q. Liu, Single-atom Cu anchored catalysts for photocatalytic renewable H₂ production with a quantum efficiency of 56%, Nat. Commun., 2022, 13, 58.
23. Y. Zhang, B. Xia, J. Ran, K. Davey and S. Z. Qiao, Atomic-level reactive sites for semiconductor-based photocatalytic CO₂ reduction, Adv. Energy Mater., 2020, 10, 1903879.
24. F. Wen and W. Liu, Three-dimensional ordered macroporous materials for photocatalysis: Design and applications, J. Mater. Chem. A, 2021, 9, 18129–18147.
25. A.-Y. Lo and F. Taghipour, Ordered mesoporous photocatalysts for CO₂ photoreduction, J. Mater. Chem. A, 2021, 9, 26430–26453.
26. X. Z. Zheng, J. Han, Y. Fu, Y. Deng, Y. Y. Liu, Y. Yang, T. Wang and L. W. Zhang, Highly efficient CO₂ reduction on ordered porous Cu electrode derived from Cu₂O inverse opals, Nano Energy, 2018, 48, 93–100.
27. J. Jiao, Y. Wei, Y. Zhao, Z. Zhao, A. Duan, J. Liu, Y. Pang, J. Li, G. Jiang and Y. Wang, AuPd/3DOM-TiO₂ catalysts for photocatalytic reduction of CO₂: High efficient separation of photogenerated charge carriers, Appl. Catal., B, 2017, 209, 228–239.
28. C. Wang, X. Liu, W. He, Y. Zhao, Y. Wei, J. Xiong, J. Liu, J. Li, W. Song, X. Zhang and Z. Zhao, All-solid-state Z-scheme photocatalysts of g-C₃N₄/Pt/macroporous-(TiO₂@carbon) for selective boosting visible-light-driven conversion of CO₂ to CH₄, J. Catal., 2020, 389, 440–449.
29. X. Tao, R. Long, D. Wu, Y. Hu, G. Qiu, Z. Qi, B. Li, R. Jiang and Y. Xiong, Anchoring positively charged Pd single atoms in ordered porous ceria to boost catalytic activity and stability in suzuki coupling reactions, Small, 2020, 16, 2001782.
30. H. S. Jeon, J. Timoshenko, C. Rettenmaier, A. Herzog, A. Yoon, S. W. Chee, S. Oener, U. Hejral, F. T. Haase and B. R. Cuenya, Selectivity control of Cu nanocrystals in a gas-fed flow cell through CO₂ pulsed electroreduction, J. Am. Chem. Soc., 2021, 143, 7578–7587.
31. J. Fu, K. Jiang, X. Qiu, J. Yu and M. Liu, Product selectivity of photocatalytic CO₂ reduction reactions, Mater. Today, 2020, 32, 222–243.
32. Y. C. Tan, K. B. Lee, H. Song and J. Oh, Modulating local CO₂ concentration as a general strategy for enhancing C–C coupling in CO₂ electroreduction, Joule, 2020, 4, 1104–1120.
33. X. Wang, P. Ou, J. Wicks, Y. Xie, Y. Wang, J. Li, J. Tam, D. Ren, J. Y. Howe, Z. Wang, A. Ozden, Y. Z. Finfrock, Y. Xu, Y. Li, A. S. Rasouli, K. Bertens, A. H. Ip, M. Graetzel, D. Sinton and E. H. Sargent, Gold-in-copper at low *CO coverage enables efficient electromethanation of CO₂, Nat. Commun., 2021, 12, 3387.
34. X. Wang, S. Shen, Z. Feng and C. Li, Time-resolved photoluminescence of anatase/rutile TiO₂ phase junction revealing charge separation dynamics, Chin. J. Catal., 2016, 37, 2059–2068.
35. Q. Xu, Y. Ma, J. Zhang, X. Wang, Z. Feng and C. Li, Enhancing hydrogen production activity and suppressing CO formation from photocatalytic biomass reforming on Pt/TiO₂ by optimizing anatase–rutile phase structure, J. Catal., 2011, 278, 329–335.
36. C. Peng, T. Zhou, P. Wei, H. Ai, B. Zhou, H. Pan, W. Xu, J. Jia, K. Zhang, H. Wang and H. Yu, Regulation of the rutile anatase TiO₂ phase junction in situ grown on -OH terminated Ti₃C₂T_x (MXene) towards remarkably enhanced photocatalytic hydrogen evolution, Chem. Eng. J., 2022, 439, 135685.
37. B.-R. Chen, V.-H. Nguyen, J. C. S. Wu, R. Martin and K. Kočí, Production of renewable fuels by the photohydrogenation of CO₂: effect of the Cu species loaded onto TiO₂ photocatalysts, Phys. Chem. Chem. Phys., 2016, 18, 4942–4951.
38. N. Li, B. Wang, Y. Si, F. Xue, J. Zhou, Y. Lu and M. Liu, Toward high-value hydrocarbon generation by photocatalytic reduction of CO₂ in water vapor, ACS Catal., 2019, 9, 5590–5602.
39. D. Qi, L. Lu, Z. Xi, L. Wang and J. Zhang, Enhanced photocatalytic performance of TiO₂ based on synergistic effect of Ti³⁺ self-doping and slow light effect, Appl. Catal., B, 2014, 160–161, 621–628.
40. M. Zhang, W. Liao, Y. Wei, C. Wang, Y. Fu, Y. Gao, L. Zhu, W. Zhu and H. Li, Aerobic oxidative desulfurization by nanoporous tungsten oxide with oxygen defects, ACS Appl. Nano Mater., 2021, 4, 1085–1093.
41. S. Wu, Z. Chen, W. Yue, S. Mine, T. Toyao, M. Matsuoka, X. Xi, L. Wang and J. Zhang, Single-atom high-valent Fe(IV) for promoted photocatalytic nitrogen hydrogenation on porous TiO₂-SiO₂, ACS Catal., 2021, 11, 4362–4371.
42. Y. Zhang, C. Cao, X.-T. Wu and Q.-L. Zhu, Three-dimensional porous copper-decorated bismuth-based nanofoam for boosting the electrochemical reduction of CO₂ to formate, Inorg. Chem. Front., 2021, 8, 2461–2467.
43. J. A. Benavides, C. P. Trudeau, L. F. Gerlein and S. G. Cloutier, Laser selective photoactivation of amorphous TiO₂ films to anatase and/or rutile crystalline phases, ACS Appl. Energy Mater., 2018, 1, 3607–3613.
44. Z. Luo, A. S. Poyraz, C.-H. Kuo, R. Miao, Y. Meng, S.-Y. Chen, T. Jiang, C. Wenos and S. L. Suib, Crystalline mixed phase (anatase/rutile) mesoporous titanium dioxides for visible light photocatalytic activity, Chem. Mater., 2014, 27, 6–17.
45. Y. Ma, X. Yi, S. Wang, T. Li, B. Tan, C. Chen, T. Majima, E. R. Waclawik, H. Zhu and J. Wang, Selective photocatalytic CO₂ reduction in aerobic environment by microporous Pd-porphyrin-based polymers coated hollow TiO₂, Nat. Commun., 2022, 13, 1400.
46. S. Jiang, K. Zhao, M. Al-Mamun, Y. L. Zhong, P. Liu, H. Yin, L. Jiang, S. Lowe, J. Qi, R. Yu, D. Wang and H. Zhao, Design of three-dimensional hierarchical TiO₂/SrTiO₃ heterostructures towards selective CO₂ photoreduction, Inorg. Chem. Front., 2019, 6, 1667–1674.
47. T. Wu, X. Zhu, Z. Xing, S. Mou, C. Li, Y. Qiao, Q. Liu, Y. Luo, X. Shi, Y. Zhang and X. Sun, Greatly improving electrochemical N₂ reduction over TiO₂ nanoparticles by Iron doping, Angew. Chem., Int. Ed., 2019, 58, 18449–18453.
48. J. Y. Do, R. K. Chava, K. K. Mandari, N.-K. Park, H.-J. Ryu, M. W. Seo, D. Lee, T. S. Senthil and M. Kang, Selective methane production from visible-light-driven photocatalytic carbon dioxide reduction using the surface plasmon resonance effect of superfine silver nanoparticles anchored on lithium titanium dioxide nanocubes (Ag@Li_xTiO₂), Appl. Catal., B, 2018, 237, 895–910.
49. M. Xiao, L. Zhang, B. Luo, M. Lyu, Z. Wang, H. Huang, S. Wang, A. Du and L. Wang, Molten-salt-mediated synthesis of an atomic nickel Co-catalyst on TiO₂ for improved photocatalytic H₂ evolution, Angew. Chem., Int. Ed., 2020, 59, 7230–7234.
50. Z. Wang, X. Mao, P. Chen, M. Xiao, S. A. Monny, S. Wang, M. Konarova, A. Du and L. Wang, Understanding the roles of oxygen vacancies in hematite-based photoelectrochemical processes, Angew. Chem., Int. Ed., 2019, 58, 1030–1034.
51. C. Liu, Y. Qin, W. Guo, Y. Shi, Z. Wang, Y. Yu and L. Wu, Visible-light-driven photocatalysis over nano-TiO₂ with different morphologies: From morphology through active site to photocatalytic performance, Appl. Surf. Sci., 2022, 580, 152262.
52. G. Wang, C.-T. He, R. Huang, J. Mao, D. Wang and Y. Li, Photoinduction of Cu single atoms decorated on UiO-66-NH₂ for enhanced photocatalytic reduction of CO₂ to liquid fuels, J. Am. Chem. Soc., 2020, 142, 19339–19345.
53. J. Yang, X. Zhu, Z. Mo, J. Yi, J. Yan, J. Deng, Y. Xu, Y. She, J. Qian, H. Xu and H. Li, A multidimensional In₂S₃–CuInS₂ heterostructure for photocatalytic carbon dioxide reduction, Inorg. Chem. Front., 2018, 5, 3163–3169.
54. J. Y. Do, N. Son, R. K. Chava, K. K. Mandari, S. Pandey, V. Kumaravel, T. S. Senthil, S. W. Joo and M. Kang, Plasmon-induced hot electron amplification and effective charge separation by Au nanoparticles sandwiched between copper titanium phosphate nanosheets and improved carbon dioxide conversion to methane, ACS Sustainable Chem. Eng., 2020, 8, 18646–18660.
55. A. A. Liu, L. C. Liu, Y. Cao, J. M. Wang, R. Si, F. Gao and L. Dong, Controlling dynamic structural transformation of atomically dispersed CuO_x species and influence on their catalytic performances, ACS Catal., 2019, 9, 9840–9851.
56. Y. Wang, Z. Chen, P. Han, Y. Du, Z. Gu, X. Xu and G. Zheng, Single-atomic Cu with multiple oxygen vacancies on ceria for electrocatalytic CO₂ reduction to CH₄, ACS Catal., 2018, 8, 7113–7119.
57. B.-H. Lee, S. Park, M. Kim, A. K. Sinha, S. C. Lee, E. Jung, W. J. Chang, K.-S. Lee, J. H. Kim, S.-P. Cho, H. Kim, K. T. Nam and T. Hyeon, Reversible and cooperative photoactivation of single-atom Cu/TiO₂ photocatalysts, Nat. Mater., 2019, 18, 620–626.
58. Z. Xu, C. Guo, X. Liu, L. Li, L. Wang, H. Xu, D. Zhang, C. Li, Q. Li and W. Wang, Ag nanoparticles anchored organic/inorganic Z-Scheme 3DOMM-TiO₂-based heterojunction for efficient photocatalytic and photoelectrochemical water splitting, Chin. J. Catal., 2022, 43, 1360–1370.
59. X. Lv, D. Y. S. Yan, F. L.-Y. Lam, Y. H. Ng, S. Yin and A. K. An, Solvothermal synthesis of copper-doped BiOBr microflowers with enhanced adsorption and visible-light driven photocatalytic degradation of norfloxacin, Chem. Eng. J., 2020, 401, 126012.
60. R. K. Chava, J. Y. Do and M. Kang, Enhanced photoexcited carrier separation in CdS–SnS₂ heteronanostructures: A new 1D–0D visible-light photocatalytic system for the hydrogen evolution reaction, J. Mater. Chem. A, 2019, 7, 13614–13628.
61. N. Zhang, A. Jalil, D. Wu, S. Chen, Y. Liu, C. Gao, W. Ye, Z. Qi, H. Ju, C. Wang, X. Wu, L. Song, J. Zhu and Y. Xiong, Refining defect states in W₁₈O₄₉ by Mo doping: A strategy for tuning N₂ activation towards solar-driven nitrogen fixation, J. Am. Chem. Soc., 2018, 140, 9434–9443.
62. X. Chang, T. Wang and J. Gong, CO₂ photo-reduction: Insights into CO₂ activation and reaction on surfaces of photocatalysts, Energy Environ. Sci., 2016, 9, 2177–2196.
63. S. Kreft, R. Schoch, J. Schneidewind, J. Rabeah, E. V. Kondratenko, V. A. Kondratenko, H. Junge, M. Bauer, S. Wohlrab and M. Beller, Improving selectivity and activity of CO₂ reduction photocatalysts with oxygen, Chem, 2019, 5, 1818–1833.
64. Y. Nosaka and A. Nosaka, Understanding hydroxyl radical (˙OH) generation processes in photocatalysis, ACS Energy Lett., 2016, 1, 356–359.
65. Z. Tian, C. Han, Y. Zhao, W. Dai, X. Lian, Y. Wang, Y. Zheng, Y. Shi, X. Pan, Z. Huang, H. Li and W. Chen, Efficient photocatalytic hydrogen peroxide generation coupled with selective benzylamine oxidation over defective ZrS₃ nanobelts, Nat. Commun., 2021, 12, 2039.
66. R. Long, Y. Li, Y. Liu, S. Chen, X. Zheng, C. Gao, C. He, N. Chen, Z. Qi, L. Song, J. Jiang, J. Zhu and Y. Xiong, Isolation of Cu atoms in Pd lattice: Forming highly selective sites for photocatalytic conversion of CO₂ to CH₄, J. Am. Chem. Soc., 2017, 139, 4486–4492.
67. Ş. Neaţu, J. A. Maciá-Agulló, P. Concepción and H. Garcia, Gold–copper nanoalloys supported on TiO₂ as photocatalysts for CO₂ reduction by water, J. Am. Chem. Soc., 2014, 136, 15969–15976.
68. Y. Huang, K. Li, J. Zhou, J. Guan, F. Zhu, K. Wang, M. Liu, W. Chen and N. Li, Nitrogen-stabilized oxygen vacancies in TiO₂ for site-selective loading of Pt and CoO_x cocatalysts toward enhanced photoreduction of CO₂ to CH₄, Chem. Eng. J., 2022, 439, 135744.
69. W. He, Y. Wei, J. Xiong, Z. Tang, W. Song, J. Liu and Z. Zhao, Insight into reaction pathways of CO₂ photoreduction into CH₄ over hollow microsphere Bi₂MoO₆-supported Au catalysts, Chem. Eng. J., 2022, 433, 133540.
70. W. Wang, C. Deng, S. Xie, Y. Li, W. Zhang, H. Sheng, C. Chen and J. Zhao, Photocatalytic C–C coupling from carbon dioxide reduction on copper oxide with mixed-valence copper(I)/copper(II), J. Am. Chem. Soc., 2021, 143, 2984–2993.
71. W. Ma, S. Xie, T. Liu, Q. Fan, J. Ye, F. Sun, Z. Jiang, Q. Zhang, J. Cheng and Y. Wang, Electrocatalytic reduction of CO₂ to ethylene and ethanol through hydrogen-assisted C–C coupling over fluorine-modified copper, Nat. Catal., 2020, 3, 478–487.
72. B. N. Choi, J. Y. Seo, Z. An, P. J. Yoo and C.-H. Chung, An in situ spectroscopic study on the photochemical CO₂ reduction on CsPbBr₃ perovskite catalysts embedded in a porous copper scaffold, Chem. Eng. J., 2022, 430, 132807.
73. R. Zhang, H. Wang, S. Tang, C. Liu, F. Dong, H. Yue and B. Liang, Photocatalytic oxidative dehydrogenation of ethane using CO₂ as a soft oxidant over Pd/TiO₂ catalysts to C₂H₄ and syngas, ACS Catal., 2018, 8, 9280–9286.
74. M. Wang, M. Shen, X. Jin, J. Tian, M. Li, Y. Zhou, L. Zhang, Y. Li and J. Shi, Oxygen vacancy generation and stabilization in CeO_2−x by Cu introduction with improved CO₂ photocatalytic reduction activity, ACS Catal., 2019, 9, 4573–4581.
75. J. Li, X. Li, C. M. Gunathunge and M. M. Waegele, Hydrogen bonding steers the product selectivity of electrocatalytic CO reduction, Proc. Natl. Acad. Sci. U. S. A., 2019, 116, 9220–9229.
76. P. B. Joshi, N. Karki and A. J. Wilson, Electrocatalytic CO₂ reduction in acetonitrile enhanced by the local environment and mass transport of H₂O, ACS Energy Lett., 2022, 7, 602–609.
77. J. X. Hao, D. Yang, J. J. Wu, B. X. Ni, L. J. Wei, Q. J. Xu, Y. L. Min and H. X. Li, Utilizing new metal phase nanocomposites deep photocatalytic conversion of CO₂ to C₂H₄, Chem. Eng. J., 2021, 423, 130190.
78. L. Hou, J. Han, C. Wang, Y. Zhang, Y. Wang, Z. Bai, Y. Gu, Y. Gao and X. Yan, Ag nanoparticle embedded Cu nanoporous hybrid arrays for the selective electrocatalytic reduction of CO₂ towards ethylene, Inorg. Chem. Front., 2020, 7, 2097–2106.
79. H. Huang, R. Shi, Z. Li, J. Zhao, C. Su and T. Zhang, Triphase photocatalytic CO₂ reduction over silver-decorated titanium oxide at a gas–water boundary, Angew. Chem., 2022, 61, e202200802.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d2qi01155g

---