Ternary SiO₂@CuO/g-C₃N₄ Nanoparticles for Solar-Driven Photoelectrocatalytic CO₂-to-Fuel Conversion

Abstract

Electrocatalytic CO₂ reduction driven by renewable electricity offers a sustainable approach to producing valuable chemicals, though it is often hindered by low activity and selectivity. CuO, an important transition metal oxide, exhibits unique advantages in photoelectrocatalysis due to its high intrinsic catalytic activity and ability to serve as an active site for CO₂ reduction. SiO₂, a widely used substrate, facilitates Cu loading and increases the specific surface area of the catalyst. Meanwhile, g-C₃N₄ provides excellent visible-light responsiveness and efficient charge carrier mobility. Together, CuO, SiO₂, and g-C₃N₄ are earth-abundant, low-cost, and chemically stable, making them ideal for solar-to-fuel applications. Here, a novel ternary heterojunction photocatalyst was constructed using SiO₂, CuO, and g-C₃N₄. The heterostructure significantly improves light-harvesting efficiency, promotes efficient charge separation and transport, and simultaneously mitigates photogenerated carrier recombination and catalyst corrosion. The resulting SiO₂@CuO/g-C₃N₄ catalyst demonstrates outstanding CO₂ conversion performance, achieving a CO yield of 17 mmolg⁻¹h⁻¹ at 1.2 V_RHE with nearly 100% selectivity. Moreover, this work systematically investigates the electrocatalytic CO₂ reduction reaction (CO₂RR) mechanism on Cu-based catalysts, offering insights into the formation of high-value multicarbon products and highlighting the potential of rational heterojunction design in enhancing solar-driven fuel production efficiency.

1. Introduction

The research and development of green carbon dioxide conversion technologies were considered critically important for addressing global climate change, ensuring energy security, and promoting sustainable development under the dual carbon goals. Since the Industrial Revolution, atmospheric CO₂ concentrations have steadily increased, intensifying the greenhouse effect and contributing to global warming. As a result, CO₂ conversion was viewed as a key strategy for advancing “negative carbon technologies.” Breakthroughs in areas such as catalyst design, electrochemical reduction, and photocatalysis had significantly improved the efficiency and cost-effectiveness of CO₂ conversion [1,2,3]. “The catalytic reduction of CO₂ converts gaseous into valuable fuels and chemical feedstocks, simultaneously mitigating greenhouse gas emissions and enabling sustainable carbon resource cycling.” [1,4]. However, the high stability of CO₂ molecules (with a bond energy of 750 kJ/mol) necessitates significant activation energy. Although electrocatalysis requires significant electrical energy, photocatalysis faces limitations such as low efficiency (generally below 10%), insufficient selectivity, expensive catalysts, and difficulties in selectively producing high-value chemicals, making large-scale implementation challenging [5]. Therefore, by leveraging multi-disciplinary collaborative innovation, developing highly efficient catalysts, coupling light-electricity multi-energy drives, reducing energy consumption and breaking through thermodynamic limitations, the unity of environmental and economic benefits can be achieved [6].

Photocatalytic CO₂ reduction technology utilizes solar energy-driven catalysis to convert carbon dioxide into green fuels such as organic matter and methane, a method that mimics the natural process of photosynthesis. However, the efficiency of photocatalytic reduction is relatively low and challenging on catalysts [6,7]. Therefore, the development of efficient catalysts and novel catalytic reduction technologies is essential to improve conversion efficiency and product selectivity. Electrochemical reduction can be carried out at ambient temperature and pressure, and higher energy utilization efficiency can be achieved compared to other chemical conversion pathways [8,9]. Electrocatalytic (EC) reduction of CO₂ is a process in which an external electric field is used as the main energy source to drive a redox reaction at the electrode. It should be noted, however, that applying excessively high voltages may induce competing hydrogen evolution, thereby reducing the Faradaic efficiency of CO₂ electroreduction into fuels [10,11]. Given the practical limitations of standalone photocatalytic and electrocatalytic CO₂ reduction technologies, researchers have developed an innovative approach by integrating these methods into photoelectrocatalytic (PEC) systems, thereby enhancing both efficiency and overall performance. PEC is a method that utilizes sunlight to drive chemical reactions and is particularly suitable for CO₂ reduction. In PEC, light energy is used to excite carriers (electrons and holes) in semiconductor materials, which then participate in redox reactions. PEC CO₂ reduction can realize the rapid transfer of electrons at a low overpotential to reduce CO₂ on the photocathode, which greatly improves the reduction efficiency and ultimately realizes low-energy CO₂ reduction [12,13]. It has been shown that excellent PEC CO₂ reduction systems can not only utilize light energy to excite and generate carriers under light conditions, effectively reducing the energy input and energy consumption of external electrons but also utilize electrical energy to accelerate the separation and transfer of electrons-holes, thus facilitating the CO₂ reduction reaction [14,15].

In PEC CO₂ reduction systems, the photoelectrode material must exhibit an optimal band gap to generate photoexcited charge carriers for CO₂ reduction, while also demonstrating stability and high selectivity. During the PEC process, when the photoelectrode is subjected to light and an applied voltage, its electrons transition from the valence band (VB) to the conduction band (CB), generating excited state electrons [16,17]. The electrons and holes produced by light can participate in a series of chemical reactions on the surface of a catalyst or at the electrode/electrolyte interface. In a PEC CO₂ reduction system, photogenerated electrons migrate to the adsorbed CO₂ molecules on the catalyst surface, driving their conversion into organic products. Constructing heterojunctions in photocathodes is an effective way to improve catalytic efficiency and selectivity [15,18,19]. This typically involves integrating multiple semiconductors or catalytic active sites to promote the separation and efficient utilization of photogenerated charge carriers [8,19].

Halmann conducted pioneering research on PEC CO₂ conversion [20]. Since then, researchers have made significant and exciting advancements in this field. Xiong’s group developed a facile method for cladding MOF materials on the surface of Cu₂O cathodes, i.e., a photocathode with a columnar Cu₃(BTC)₂ MOF structure on the surface of a Cu₂O film. Christiane’s group investigated the growth of Ti-O-Cu nanotubes grown on a bimetallic alloy (Ti-x-Cu; x = 0.5, 5.5, and 10 at. %) for the PEC selectivity of the CO₂ reduction reaction [17,21]. Cu-based photoelectrodes exhibit great advantages in PEC CO₂ reduction. First, they have excellent light absorption capacity and can effectively utilize visible light to catalyze the reduction of carbon dioxide [18,19]. Secondly, some copper-based catalysts exhibit excellent electron transfer characteristics, which contribute to effective light absorption and electron transfer in PEC, thereby enhancing the efficiency of CO₂ reduction [1,20]. Finally, the active sites on the surface of Cu-based catalysts can be controlled through various methods to enhance catalytic performance and product selectivity [22]. In recent years, copper-based catalysts have developed rapidly, evolving from the initial pure copper oxide (CuO, Cu₂O) to the current multi-component copper oxide and copper sulfide, as well as copper-based composite materials. As a P-type semiconductor, Cu₂O can form various types of heterojunctions with semiconductors with appropriate band positions, including p-n junctions, P-N-P junctions, and Z-scheme heterojunctions, effectively suppressing the recombination of electrons and holes on the surface of photoelectrodes [23,24,25]. They play a crucial role in promoting the development of the PEC CO₂ reduction field.

SiO₂ is highly advantageous due to its earth abundance, non-toxicity, facile processing, and low production costs [5,6,7]. Numerous studies have explored SiO₂ as a functional additive to Cu-based catalysts, demonstrating its potential to enhance performance in photoelectrochemical CO₂ reduction reactions [8]. Zhang et al. found that the addition of SiO₂, TiO₂ or SiO₂-TiO₂ not only enhanced the surface area of metallic copper but also changed the adsorption performance of the catalyst [9]. Zhang et al. constructed the Cu⁰-Cu⁺-NH₂ composite reaction interface by taking advantage of the excellent properties of SiO₂: Firstly, it can uniformly coat inorganic nanoparticles; Second, it can be combined with silane coupling agents to introduce organic functional groups for effective electro-reduction of CO₂ to C₂₊ products [10]. Jia et al. found that the addition of SiO₂ significantly increased the specific surface area of BET and copper [11]. Among various semiconductor photocatalysts, graphitic carbon nitride (g-C₃N₄) exhibits distinctive advantages including remarkable chemical stability, an ideal band gap for visible-light absorption, non-toxicity, and low-cost fabrication. These superior properties make it a highly promising candidate for photocatalytic CO₂ reduction applications. Wang et al. explored the photocatalytic hydrogen evolution reaction of g-C₃N₄ under visible light for the first time [12]. Dong and Zhang were the first to use g-C₃N₄ to catalyze the reduction of CO₂ to CO under visible-light excitation in the presence of water vapor, demonstrating the potential of g-C₃N₄ for CO₂ reduction [13].

Here, SiO₂, CuO, and g-C₃N₄ were selected to construct a ternary heterojunction for photoelectrocatalytic CO₂ reduction. The efficient conversion of CO₂ to CO was successfully achieved. The SiO₂ support enhances both Cu loading and the catalyst’s specific surface area. Moreover, CuO exhibits excellent PEC performance, serving as an active site for CO₂ reduction. Meanwhile, g-C₃N₄, with its strong visible-light responsiveness and carrier transport capacity, acted as a photoresponsive material to effectively absorb light energy and facilitate charge transport. The integration of these three compounds into a ternary heterojunction enhanced light absorption, promoted efficient charge separation and transport, suppressed photogenerated carrier recombination, and improved catalyst stability. A provisional finding explored the photoelectrocatalytic reaction path on the SiO₂@CuO/g-C₃N₄ catalyst, and this work provides prospects for the interface design of ternary heterojunctions for efficient CO₂ conversion.

2. Discussion

2.1. Elemental Composition and Surface Chemistry

The elemental composition and crystal structure of the SiO₂@CuO/g-C₃N₄ electrode were analyzed by XRD, Raman and XPS characterization. The structure, phase and purity of SiO₂@CuO/g-C₃N₄ were investigated by X-ray diffraction analysis. Figure 1a–c indicates that the silica (SiO₂) nanoparticles are amorphous with a wide peak around 2θ ≈ 24.5 which agreement with other works. The XRD pattern showed the formation of the pure and polycrystalline phase of CuO nanoparticles with a monoclinic structure with (002), (111), (202), (113), and (311) orientations [5]. The characteristic peaks of SiO₂@CuO/g-C₃N₄ samples prepared with different CuO contents were basically the same, which proved that the phase structure of the metal did not change during the synthesis of the bimetallic catalyst. These results show the successful preparation of SiO₂@CuO/g-C₃N₄. The Raman spectrum of g-C₃N₄ shows a G band at around 1614 cm⁻¹ and a D band at around 1356 cm⁻¹ (Figure 1d). D peak was the vibration mode of phonons, which reflects the defect and disorder degree of carbon nanotubes. G and G′ peaks reflect the degree of order in the arrangement of carbon atoms. Besides the D and G bands, the additional Raman characteristic spectra of carbon at ~2700 cm⁻¹, known as 2D (or G′) band, was also weakly found in both samples. The intensity of the 2D band in both samples is very low and a highly broadened 2D band was also observed. The 2D band is Raman active for crystalline graphitic materials. Raman spectra of CuO exhibits a peak at 282 cm⁻¹ which is characteristic of Ag mode.

The XPS Cu 2p spectrum in Figure 2 shows the peaks at 934.1 and 954.2 eV which are assigned to Cu 2p_3/2 and Cu 2p_1/2. The appearance of shakeup satellites with peaks at ∼942 eV and ∼962 eV is also characteristic for the presence of Cu²⁺ and is unique for the Cu^IIO structure. Furthermore, the structure of CuO is also confirmed by the Auger peak analysis with the characteristic Cu at 917.1 eV. The XPS O 1s spectrum shows the presence of different peaks with binding energy (BE) of 529.7, 532.5, 533.2, and 534.3 eV. The peak at 529.7 eV is assigned to the bulk lattice oxygen of CuO, as established in the literature. Hence, with bulk lattice oxygen as the reference, the 3 shifts observed in the O 1s spectra are +1.7, +3.5, and +4.9 eV. These XPS shifts (and corresponding binging energy) are widely observed in the literature for CuO O 1s spectra. However, the assignment of these BE to surface moieties (and their adsorption configurations) is either absent, inconsistent or disputed. For instance, +1.7 eV can correspond to the oxygen adsorbed on the surface of CuO in molecular or dissociated form, +3.5 eV can correspond to the hydroxyl groups or H₂O adsorbed on the surface of CuO, and +4.9 eV can correspond to the hydroxyl groups or H₂O adsorbed on the surface of CuO [26].

The morphology and microstructure of the catalysts were observed by SEM and TEM. Figure 3 shows the SEM images of SCCN-10(SiO₂@CuO (10)/g-C₃N₄) and SCCN-15 (SiO₂@CuO (15)/g-C₃N₄). These images show that the prepared silica nanoparticles are spherical, regular, and dispersed with distinct boundaries. The average diameter of silica (SiO₂) nanoparticles was estimated to be about 15 nm. The resulting nanoparticles also have a uniform distribution. The FE-SEM image (Figure 3a,b) of SiO₂/CuO nanoparticles show that the SiO₂ spheres are completely covered with copper oxide nanoparticles and the distribution of nanoparticles is almost uniform, and the average size of these nanoparticles is estimated to be circa. 20 nm. At the same time, these uniformly distributed SiO₂/CuO nanoparticles are covered with layers of nitrogen carbide. The TEM images reveal the size and shape of the nanoparticles with an accuracy of about a tenth of a nanometer, depending on the type of material and device used. Figure 3e–g shows the TEM images of the SiO₂@CuO (10 wt%)/g-C₃N₄. The figure shows two sets of clear lattice fringes, with the corresponding crystal plane spacings of 0.27 nm and 0.275 nm, respectively, both marked as (110) crystal planes. This image shows high-resolution lattice fringes, indicating that the sample has good crystallinity [18].

2.2. Optical Properties

The effects of CuO content and g-C₃N₄ introduction on the photoresponse of the catalyst were investigated by photophysical characterization. The UV/Vis diffuse reflectance spectroscopy (DRS) shows SiO₂@CuO catalysts (Figure 4a) display greatly enhanced absorption of visible and near-infrared light with the increase in CuO content. Compared to the SiO₂@CuO catalysts, the UV-visible spectrum of SiO₂@CuO/g-C₃N₄ (Figure 4b) shows complete visible spectral absorption, possibly due to the darker color of the catalyst sample and the introduction of g-C₃N₄. The UV-VIS absorption spectrum of the sample was tested after grinding, and it was found that the absorption range of visible light increased with the increase in Cu doping.

2.3. Electrochemical Properties

Figure 5 depicts the electrochemical impedance spectroscopy (EIS) and transient photocurrent response (TPR) of the photocatalyst. Electrochemical impedance spectroscopy (EIS) was used to reveal the transfer behavior of photocarriers. The arc radius of SCCN-10 heterostructure is significantly smaller than that of SCCN-15, indicating that the charge transfer rate of SCCN-10 catalyst is accelerated. In addition, the TPR intensity of SiO₂@CuO/g-C₃N₄ ternary heterojunction catalyst is SCCN-20 > SCCN-10 > SCCN-15, indicating that SCCN-10 catalyst improves the electron transport efficiency.

During the photoelectrocatalytic reduction of carbon dioxide, the SiO₂@CuO/g-C₃N₄ composite material has significant advantages over the individual CuO/g-C₃N₄ or SiO₂/CuO. The binary system of CuO/g-C₃N₄, where the catalyst is directly anchored on the carrier, leads to agglomeration [27]. Therefore, the direct use of CuO₂ is not pursued, and the covering of a layer of SiO₂ on another layer of g-CN (CuO/g-C₃N₄) without the support of SiO₂ is not adopted. Amorphous SiO₂ typically possesses a rich mesoporous structure, which can significantly increase the specific surface area of the material, providing more anchoring sites for the dispersion of CuO and exposing more catalytic active centers [28]. Instead, a layer of graphitic carbon nitride is chosen to be added to form a ternary SiO₂@CuO/g-C₃N₄ composite material, because graphite carbon nitride is a good electron transfer material and is used to form a complex of photosensitizer and photocatalyst to achieve the electroreduction of carbon dioxide [29]. The SiO₂@CuO and SiO₂@CuO/g-C₃N₄ samples were characterized by ultraviolet-visible diffuse reflectance spectroscopy (UV-Vis DRS), and the influence of the introduction of g-C₃N₄ on the photocatalytic response characteristics of the catalyst was studied. The experimental results show that compared with SiO₂@CuO, the SiO₂@CuO/g-C₃N₄ sample (Figure 4b) has a wider visible light absorption range, which may be due to the expansion of the light response range caused by the introduction of g-C₃N₄. SiO₂@CuO/g-C₃N₄ combines the structural advantages of SiO₂ and the semiconductor properties of g-C₃N₄ (wide spectral response, efficient charge separation) [30]. Therefore, it has more advantages in photocatalytic carbon dioxide reduction compared to the individual CuO/g-C₃N₄ or SiO₂/CuO materials, highlighting its strong potential for efficiently converting solar energy into chemical fuel.

2.4. Photoelectrocatalytic Performanc for CO₂RR

2.4.1. PEC Performance and Faraday Efficiency of the Catalysts

To study the influence of the photoelectric effect on the catalytic performance of the samples, the photoelectric reduction of CO₂ was evaluated in the electrochemical analysis workstation (CHI 760) under the dual drive of simulated sunlight and electrical energy. Figure 6a shows the CO₂ photoreduction activities of various catalysts under photoelectric dual drive conditions. It can be clearly seen that all the samples have a high selectivity for CO₂-to-CO and only produce a small amount of C₂H₄. SCCN-10 exhibited a relatively high CO yield, while SCCN-20 showed a higher C₂H₄ yield. This might be because the Cu content comprehensively affects the activity and selectivity of CO₂RR by regulating the properties of active sites, electronic structure, competitive reactions and mass transfer processes [31,32,33]. Optimizing the dispersion degree, valence state and interaction with the carrier of Cu is the key. An increase in Cu content may provide more active sites, but it is not a linear relationship. When the Cu content is low, the active sites are insufficient, limiting the reaction rate. When it is too high, it may lead to the agglomeration of Cu particles and reduce the effective surface area. An appropriate amount of Cu (SCCN-10) can optimize the adsorption of *COOH intermediates and promote the generation of CO. Also, large-particle Cu may tend to compete for hydrogen evolution reaction (HER), reducing the selectivity of CO.

2.4.2. Comparison of PEC Performance at Different Potentials

To reveal the relationship between the activity of the catalyst and the overpotential, the overpotential (η) is the difference between the actual potential that drives the reaction and the thermodynamic equilibrium potential (for example, the equilibrium potential of CO₂/CO is −0.11 V vs. RHE). The minimum overpotential required for the catalyst to achieve a high current density (activity) and measure its energy efficiency. The electrocatalytic activity of the catalyst was evaluated, respectively, under different overpotentials. Figure 7 showed photoelectrocatalytic performance at three different potentials. It can be seen from Figure 7a that the CO yield of SCCN−10 is the highest at −1.2 V_RHE, reaching 17 mmol g⁻¹ h⁻¹. It shows better performance advantages compared with the reported literature (Table 1). The ethylene yields gradually increased with the increase in potential and reached 4.3 mmol g⁻¹ h⁻¹under the condition of −1.4 V_RHE. The Faraday efficiency graph in Figure 7b shows that CO exhibits the highest Faraday efficiency at −1.0 V, approaching 17%. At −1.2 V_RHE, the highest FE_C2H4 = 12% was exhibited.

2.4.3. Comparison of Photoelectrocatalytic Performance and Stability Test

In the photoelectrocatalytic CO₂ reduction reaction (PEC CO₂RR), the role of light is far more than just providing energy. It profoundly affects the reaction pathway, catalytic activity and product selectivity by influencing the generation and separation of photogenerated carriers and the adsorption of reaction intermediates [26,34]. Taking the photocatalyst SCCN-10 as the optimal catalyst, a comparative experiment on electrocatalytic and photocatalytic performance was conducted, as shown in Figure 7e,f. It can be clearly seen that under photoelectrocatalysis, the yields of CO and C₂H₄ have significantly increased, as the introduction of light further promotes the CO₂ reduction reaction. Taking the catalyst SCCN-10 as the optimal catalyst, the stability of the catalyst was tested, and the results are shown in Figure 7f. After the photoelectrocatalytic CO₂ reduction reaction for 16 h, the catalyst still maintained relatively stable catalytic activity. The production rates of both products over the SCCN-10 catalyst exhibit minor fluctuations. Overall, the CO production rate increases, while the C₂H₄ production rate shows a slight decrease. This is beneficial in terms of the selectivity of CO. The KHCO₃ electrolyte may affect the reaction environment pH due to the co26nsumption of HCO₃⁻ and the migration of K⁺. Such changes may influence the adsorption strength of intermediates (such as COOH and CO), thereby altering the selectivity of CO products. In the reported literature, the valence state of Cu and the physical or chemical changes in the catalyst carrier affect the product selectivity [35,36]. At the same time, during the long-term stability reduction reaction process, catalytic performance also typically shows stable fluctuations in both directions [37,38]. Overall, during the 16 h reduction reaction process, the CO production rate increases, while the C₂H₄ production rate shows a slight decrease, which is beneficial in terms of the selectivity of CO.

Table 1. CO₂RR properties of Cu-based electrocatalysts in different electrolytes.

Catalysts	Conditions	Reactor	CO	C2H4	Ref.
Cu2O/Zn-Cr-LDHs	200 W Hg-Xe lamp	H-type Cell	1.3125 μmol/g/h		[1]
Cu2O/Cu/Cu3V2O7(OH)2·2H2O	300 W Xe lamp (λ > 400 nm)	H-type cell	6.97 μmol/g/h		[2]
3D porous Cu2O	300 W Xe lamp (λ > 420 nm)	H-type cell	26.8 μmol/g/h	0.66 μmol/g/h	[3]
BiVO4/C/Cu2O nanowires	300 W Xe lamp (λ > 420 nm)	H-type cell	3.01 μmol/g/h		[4]
g-C3N4/Cu2O		Flow cell	8.182 μmol/g/h		[5]
FeCuP	−1.6 V	Flow cell	1.44 μmol/h		[11]
Co-ZIF-9/g-C3N4	NA	Flow cell	495 μmol/g/h		[13]
NH2/MOF	−1.2 V	Flow cell	1.7 mmol/g/h		[14]
Re@NH2-MOF	−1.2 V	Flow cell	CO = 14.85 mmol/g/h		[14]
Cu2O/Sn/PTFE	−1.2 V	Flow cell	CO = 68.31 μmol/h/cm2		[15]
Au-Cu/SrTiO3/TiO2	NA	Flow cell	3.7mmol/gcat/h		[16]
Ni/g-C3N4	300 W Xe lamp	Flow cell	19.85 μmol/g/h		[17]
α-Fe2O3/g-C3N4	300 W Xe lamp	Flow cell	27.2 μmol/g/h		[18]
Cu/ZnO/g-C3N4	400 W Hg lamp	H-type cell	65.1 μmol/g/h		[39]
Ru/Zn-g-C3N4-1/20	300 W Xe lamp	H-type cell	288.42 μmol/g/h		[40]
Pt/TiO2/g-C3N4	300 W Xe lamp	H-type cell	3.8 μmol/g/h		[41]
SiO2@CuO (10 wt%)/g-C3N4	300 W, −1.2 VRHE, 2 h, 0.1 M KHCO3	H-type cell	17 mmol/g/h	2305.1 μmol/g/h	This work

Table 1. CO₂RR properties of Cu-based electrocatalysts in different electrolytes.

Catalysts	Conditions	Reactor	CO	C2H4	Ref.
Cu2O/Zn-Cr-LDHs	200 W Hg-Xe lamp	H-type Cell	1.3125 μmol/g/h		[1]
Cu2O/Cu/Cu3V2O7(OH)2·2H2O	300 W Xe lamp (λ > 400 nm)	H-type cell	6.97 μmol/g/h		[2]
3D porous Cu2O	300 W Xe lamp (λ > 420 nm)	H-type cell	26.8 μmol/g/h	0.66 μmol/g/h	[3]
BiVO4/C/Cu2O nanowires	300 W Xe lamp (λ > 420 nm)	H-type cell	3.01 μmol/g/h		[4]
g-C3N4/Cu2O		Flow cell	8.182 μmol/g/h		[5]
FeCuP	−1.6 V	Flow cell	1.44 μmol/h		[11]
Co-ZIF-9/g-C3N4	NA	Flow cell	495 μmol/g/h		[13]
NH2/MOF	−1.2 V	Flow cell	1.7 mmol/g/h		[14]
Re@NH2-MOF	−1.2 V	Flow cell	CO = 14.85 mmol/g/h		[14]
Cu2O/Sn/PTFE	−1.2 V	Flow cell	CO = 68.31 μmol/h/cm2		[15]
Au-Cu/SrTiO3/TiO2	NA	Flow cell	3.7mmol/gcat/h		[16]
Ni/g-C3N4	300 W Xe lamp	Flow cell	19.85 μmol/g/h		[17]
α-Fe2O3/g-C3N4	300 W Xe lamp	Flow cell	27.2 μmol/g/h		[18]
Cu/ZnO/g-C3N4	400 W Hg lamp	H-type cell	65.1 μmol/g/h		[39]
Ru/Zn-g-C3N4-1/20	300 W Xe lamp	H-type cell	288.42 μmol/g/h		[40]
Pt/TiO2/g-C3N4	300 W Xe lamp	H-type cell	3.8 μmol/g/h		[41]
SiO2@CuO (10 wt%)/g-C3N4	300 W, −1.2 VRHE, 2 h, 0.1 M KHCO3	H-type cell	17 mmol/g/h	2305.1 μmol/g/h	This work

2.5. DFT Calculations

To further explore the photoelectrocatalysis reduction reaction path of CO₂ on SCCN-10 catalyst, the DFT method was used to calculate the energy required to convert CO₂ to *CO. CO₂ forms the activation energy of *COOH (0.532 eV) on SCCN-10, and *COOH intermediate reacts with a proton to release a water molecule and generate CO. Based on the DFT, a probable CO₂ conversion pathway is proposed: * + CO₂ —→ *CO₂ + H⁺ → *COOH + H⁺ →*CO + H₂O → CO. In general, CO₂ can be reduced first to CO through the *COOH pathway, and the adsorbed CO is a common intermediate of C₂₊ products in CO₂RR [42]. The DFT results show that the hydrogenation of *CO to *CHCO plays an important role in the enhancement of C₂₊ products. In this case, the possible reaction pathways in the C-C coupling step are as follows: *CO + CO → *2CO → *OCCO + H⁺ → *HOCCO + H⁺ → *CCO + H⁺ → *CHCO + H⁺ → *CHCHO + H⁺ → *CH₂CHO + H⁺ → *CH₂CHOH + H⁺ → *CH₂CH₂OH + H⁺ → * + C₂H₄ + H₂O. Correspondingly, the Gibbs free energy change (ΔG) for each step is calculated, as shown in Figure 8. For SCCN-10, the conversion of *HOCCO to *CCO is identified as the rate-limiting step.

3. Materials and Methods

3.1. Materials

Tetraethoxysilane (TEOS), Ethanol (C₂H₆O), Copper (II) Chloride Dihydrate (CuCl₂·2H₂O) and Ammonium hydroxide (NH₃·H₂O) was purchased from Shanghai Titan Technology Co., Ltd. (Shanghai, China). Melamine was provided by Beijing Aladdin Future Technology Co., Ltd. (Beijing, China). Deionized water was used in the experiments. potassium bicarbonate (KHCO₃) was provided by Titan Co., Ltd. (Shanghai, China). All the chemical substances used in this study were of analytical grade and were not further purified. The Nafion was purchased from Titan Co., Ltd. (Shanghai, China).

3.2. Synthesis of the SiO₂@Cu and SiO₂@CuO/g-C₃N₄ Catalyst and Working Electrodes

Synthesis of SiO₂@Cu. 300 mL of ethanol and 56 mL of water were placed in a 500 mL three-necked flask, 7 mL of ammonia water was added and stirred at 30 °C for 1 h. Subsequently, 12 mL of tetraethyl orthosilicate (TEOS) was added and continued to react for 20 h. A white colloidal solution was formed. Then 5 wt.% Cu was added to the above solution mixture, the solution changes from white to dark blue. At the same time, an additional 10 mL of ammonium hydroxide (with a concentration of 25%) was added was heated to 85 °C and stirred for 6 h. Once the reaction ended, it was washed several times with water and ethanol, respectively, and then dried at 60 °C overnight. The Cu content was changed to 10 wt.%, 15 wt.%, and 20 wt%, respectively, to obtain catalysts SiO₂@Cu (x wt.%) with different Cu contents.

Synthesis of SiO₂@CuO/g-C₃N₄. 0.5 g of SiO₂@Cu and 2.5 g of melamine powder were thoroughly mixed and ground using a pestle and mortar. The powder mixture was subsequently heated in a muffle furnace at a heating rate of 5 °C/min up to 550 °C and maintained for 3 h to obtain SiO₂@CuO/g-C₃N₄.

Synthesis of working electrodes. The electrochemical experiments were performed in a three-electrode cell using a potentiostat (CHI 760). A reversible hydrogen electrode (RHE) immersed in the working electrolyte and a graphite rod were used as reference and counter electrodes, respectively. The working electrodes were prepared as follows: 4 mg of the electrocatalyst was suspended and ultrasonic in 1 mL of a 0.2 wt.% Nafion and 20 wt.% isopropanol aqueous solution. 300 μL of the 4 mg mL⁻¹ suspension was dropped-cast onto the carbon paper.

3.3. Catalysts Characterization

X-ray powder diffraction (XRD) spectra were performed on benchtop Aeris equipped with Cu Kα radiation with the scanned range of 10–80°. The morphology was characterized by Scanning electron microscope (SEM) (Regulus 8100, JEOL Ltd., Tokyo, Japan) and High-Resolution Transmission Electron Microscope (HRTEM) (JEM-F200, JEOL Ltd., Tokyo, Japan) and Energy Dispersive X-Ray Spectroscopy (EDX) (JEM-F200, Japan) were conducted on Fei G2 F30 to examine morphologies and elemental compositions. The visible absorption range and band gap were recorded by UV-vis diffuse reflectance spectroscopy (DRS) (Shimadzu UV3600, Shimadzu Corporation, Tokyo, Japan). Raman measurements were carried out using a Horiba LabRAM HR Evolution Raman microscope (LabRAM HR Evolution, Horiba Ltd., Paris, France). A 532 nm laser was used, and signals were recorded using a 20 s integration and by averaging two scans. X-ray photoelectron spectroscopy (XPS) was recorded on Thermo Fisher ESCALAB XI⁺ (Thermo Fisher, Waltham, MA, USA). UV-vis diffuse reflectance spectroscopy (UV-DRS) spectra were tested on Shimadzu UV-2600i. Transient photoluminescence (PL) spectra were obtained at room temperature using a FLS-1000 spectrometer (FLS1000, Edinburgh Instruments, Edinburgh, UK). Steady PL spectra were determined with an RF-5301PC Spectro fluorophotometer (Shimadzu RF-5301PC, Shimadzu Corporation, Tokyo, Japan). Inductively coupled plasma optical emission spectrometry (ICP-MS) measurements were carried out using Agilent 720 (Tokyo, Japan).

3.4. Electrochemical Test

Photoelectrochemical measurements of the samples were analyzed with an electrochemical workstation (CHI 760) using a conventional three-electrode system The Pt electrode and Ag/AgCl electrode (saturated with KCl) were used as the counter electrode and reference electrode, respectively. The working electrode was prepared on 10 × 15 mm conductive carbon paper (TGP-H-060). The method was as follows: 4 mg of the sample was dispersed in 1mL of a solution containing ethanol, water and Nafion, and then ultrasonically dropped onto the conductive carbon paper after 30 min. The electrochemical characterization test of instantaneous photoelectrocatalytic CO₂ reduction performance was carried out in 0.1 M potassium bicarbonate electrolyte.

3.5. Photoelectrocatalytic CO₂ Reduction

The CO₂RR testing was carried out in a gastight H-type electrolytic cell separated by a Nafion 117 membrane. In both chambers, 0.1 M KHCO₃ were used as cathode and anode electrolyte. A three-electrode system, in which SiO₂@CuO/g-C₃N₄ used as work electrode, Ag/AgCl (3 M KCl) and Pt (1 cm × 2 cm) plate as reference and counter electrode with an electrochemical station (CHI 760) was adopted to electrochemical measurements and a circulating water device to maintain a constant temperature of 25 °C. CO₂ (99.999%) was passed into the cathode at a rate of 20 mL/min for 30 min, followed by constant potential photoelectrochemical CO₂ reduction. A 300 W xenon lamp (λ ≥ 420 nm) with 570 mW/cm² was used for the PEC CO₂ reduction tests. The gaseous products were detected by gas chromatography (Fuli GC7970, Ar as a carrier gas) equipped with a flame ionization detector and a column (TDX-01). The column temperature was maintained at 180 °C and the detector temperature at 200 °C. In this study, the potentials were converted by the following equation:

E_RHE = E_Ag/AgCl + 0.0591 _× pH + 0.199 V

V % of gaseous products can be obtained from the GC peak areas and calibration curves for the TCD detector. Since the flow rate of the outlet was monitored to be constant, the moles of gaseous products can be calculated. The Faradaic efficiency of gaseous product is

FE = models of product/Q/nF × 100%

(Q: charge (C); F: Faradaic constant (96,485 C/mol); n: the number of electrons required to generate the product)

3.6. DFT Calculations Method

The DFT implemented in the Vienna Ab initio Simulation Package (VASP) is used in all calculations [14]. The Perdew–Burke–Ernzerhof generalized gradient approximation (GGA-PBE) is utilized to describe the exchange-correlation potential [15]. The interaction between the ionic core and valence electrons is treated using the projection-enhanced wave (PAW) method. The plane wave cutoff energy is fixed at 500 eV and relaxes the given structural model until the Hermann–Feynman force is less than −0.02 eV/Å and the energy change is less than 10⁻⁵ eV. Artificial interactions between the periodic images were avoided by adding a 20 Å vacuum layer perpendicular to the sheet in the heterostructure. And the dispersive interactions were corrected by the DFT-D3 method [16]. The Gibbs free energy (G) is computed by correcting the DFT energy (E) with the following formula:

$$G=E+ZPE+\mathsf{\Delta }{H}_t-T\mathsf{\Delta }S$$

where T is thermodynamic temperature (set to 298.15 K), ZPE is zero-point energy, $\mathsf{\Delta }{H}_t$ is enthalpy change and $T\mathsf{\Delta }S$ is the entropy correction.

4. Conclusions

In conclusion, A unique ternary heterojunction electrocatalyst for photoelectrocatalytic CO₂ reduction, composed of SiO₂, CuO, and g-C₃N₄, was developed. SiO₂ enhanced Cu dispersion and increased the catalyst’s specific surface area; CuO served as an active site with excellent photoelectrocatalytic performance; and g-C₃N₄, with strong visible-light responsiveness and carrier mobility, acted as a photoresponsive component to facilitate light absorption and charge transport. The synergistic integration of these three earth-abundant and stable materials enabled efficient CO₂ conversion, achieving a CO yield of 17 mmolg⁻¹h⁻¹ at 1.2 V_RHE with nearly 100% selectivity—surpassing many reported systems. In addition to demonstrating high performance, this study provided mechanistic insights into CO₂ electroreduction over Cu-based catalysts, including the formation pathways of C₂₊products. This work offers a promising strategy for the rational design of advanced ternary heterojunctions and broadens the prospects for efficient, scalable photoelectrocatalysis and solar-to-fuel technologies.