Efficient Liquid Multicarbon Production from Overall CO2 Photoreduction by Creating a Superhydrophobic Gas Storage Space

doi:10.21203/rs.3.rs-2455626/v1

Download PDF

Article

Efficient Liquid Multicarbon Production from Overall CO₂ Photoreduction by Creating a Superhydrophobic Gas Storage Space

https://doi.org/10.21203/rs.3.rs-2455626/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

It is exceedingly desired yet challenging to steer the photocatalytic overall CO₂ reduction reaction (CRR) toward high-value liquid products, especially multicarbon products like ethanol. Numerous attempts have been reported such as developing new materials, cocatalyst engineering, etc. Here, we find that by just curling the superhydrophobic Cu₂O to create an enclosed Cu₂O hollow structure, the selectivity of ethanol (59.59%) is impressively elevated by ~ 9 times higher than that without the enclosed hollow structure. Further investigations indicate that the superhydrophobic hollow structure can efficiently repel water out and thus create an enclosed gas storage space, enabling the high concentration of CO₂ molecules. With such a strategy, an impressive ethanol generation rate of 996.18 µmol g^− 1 h^− 1 is achieved. The mechanism is also investigated based on theoretical calculations and corresponding characterizations. This strategy provides a geometry-modulated method with high compatibility, which is promising to develop more efficient photocatalysts for multicarbon production combined with other traditional methods.

Physical sciences/Energy science and technology/Renewable energy/Solar energy/Artificial photosynthesis

Physical sciences/Chemistry/Green chemistry/Renewable energy/Solar energy/Artificial photosynthesis

Physical sciences/Engineering/Chemical engineering

Physical sciences/Materials science/Materials for energy and catalysis/Photocatalysis

overall CO2 photoreduction

multicarbon

superhydrophobic

gas storage

Cu2O

The ever-increasing worldwide consumption of fossil fuels depletes these finite natural resources, but more urgently leads to overproduction of the greenhouse gas carbon dioxide (CO₂).¹ Hence, solutions to reduce CO₂ emissions or, at best, utilization of CO₂ in a sustainable energy cycle are highly relevant.² Overall photocatalytic CO₂ reduction reaction (CRR) without sacrificial agents is regarded as one of the most promising solutions, because it can convert CO₂ and H₂O into high value-added products (e.g. carbon monoxide (CO), methane (CH₄), ethylene (C₂H₄), methanol (CH₃OH), ethanol (CH₃CH₂OH), etc.) while only consuming renewable solar energy.³ Among the products, liquid carbon derivatives (such as methanol, ethanol, etc.) are desirable because of the high energy density and transportation convenience.⁴ Specially, the multicarbon products obtained from the carbon-carbon (C-C) coupling process attract more attention in recent years, since they have higher economic potential and can directly serve as building blocks for the synthesis of long-chain hydrocarbon fuels and oxygenates.^{4, 5} Belonging to both liquid and multicarbon product, ethanol has been widely used as industrial chemical and liquid fuel, which has been paid special attention in this report.

The C-C coupling process often suffers from a high energy barrier, which makes the generation of multicarbon products difficult.^{6, 7} Meanwhile, the generation of multicarbon products also involves complex multiple electron coupled to proton transfer reactions (e.g., twelve-electron transfer for ethanol).^{8, 9} This inevitably creates larger overpotentials and a greater variety of products, leading to poor activity and selectivity of multicarbon products.⁴ Cu₂O is one of the most promising materials for liquid multicarbon production from photocatalytic CRR, whose low-coordinated Cu⁺ surface atoms could facilitate the adsorption of *CO, which is a key intermediate for C-C coupling during CRR.^{10, 11} Based on Cu₂O material, numerous methods aiming to elevate the generation of multicarbon products such as heterojunction construction,¹² cocatalyst engineering,¹³ etc. have been adopted, which have been proven to be efficient.

Besides these traditional methods, researchers realize that higher concentration of CO₂ near active sites is another significant factor for elevating the activity and selectivity of CRR products, especially for multicarbon products.¹⁴ However, due to the low solubility and poor diffusion coefficient in aqueous solution, the CO₂ molecule is very difficult to transfer to the surface active sites.^{15, 16} Numerically, the ratio of CO₂ to H₂O molecules is 1:1300 at 1 atm pressure, resulting in the serious competitive hydrogen evolution reaction (HER) and low selectivity of carbon derivatives from CRR.¹⁷ Recently, researchers have developed a series of hydrophobically modified photocatalysts for CO₂ mass transfer enhancement, elevating the CO₂ concentration near active sites to some extent, and successfully promoting the generation of carbonaceous products. However, C₁ products such as CO, CH₄, CH₃OH, etc. account for the major proportion. One possible reason is that the concentration of CO₂ is still insufficient to support the production of liquid multicarbon products.^{14, 18} Thus, the local CO₂ concentration should be further elevated to improve the performance of multicarbon production by mass transfer enhancement.

Here, we find that by just curling the superhydrophobic Cu₂O to create an enclosed gas storage space (superhydrophobic Cu₂O hollow structure, O-CHS), the selectivity of ethanol production from overall CO₂ photoreduction is impressively elevated by ~ 9 times compared with the superhydrophobic Cu₂O losing the gas storage space. This structure can efficiently repel water out and store CO₂ inside. Meanwhile, the stored CO₂ can pass through the mesoporous shell to the catalyst surface again to supplement its consumption, greatly shortening the mass transfer distance of CO₂ and increasing the local CO₂ concentration, thereby ensuring the rich enough triphase microenvironments during CRR (Fig. 1). With such a strategy, the ethanol generation rate up to 996.18 µmol g^-1 h^-1 has been achieved. Considering the methanol production, the selectivity and apparent quantum efficiency (AQE) of liquid carbon derivatives can be up to 99.97% and 1.63%, respectively. The further mechanism investigations based on in situ Fourier transform infrared spectroscopy (FTIR) and density functional theory (DFT) calculation clarify the reaction paths and the advantages of the gas storage space in O-CHS. The geometric operation to facilitate the liquid products from photocatalytic overall CRR, especially multicarbon products, is very compatible, which is promising to develop more efficient photocatalysts combined with other traditional methods.

Morphological And Structural Characterization

Cu₂O hollow structure with enclosed space (denoted as CHS, Supplementary Fig. 1a) was firstly synthesized by an in situ one-pot chemical transformation procedure (see Section 1 of Methods).¹⁹ The as-synthesized CHS exhibits a hydrophilic property with a water contact angle (WCA) of 43.96° (Fig. 2g). Then the target catalyst O-CHS was obtained by hydrophobic modification of CHS using tiny amounts of 1-dodecanethiol (see Section 4 of Methods). After modification, the O-CHS exhibits a superhydrophobic property with a WCA of 152.24° (Fig. 2h).²⁰ To investigate the effect of the enclosed space surrounded by the superhydrophobic Cu₂O and the advantages of the hollow structure, we also synthesized another three kinds of reference catalysts without the enclosed space. The first one was the hydrophilic Cu₂O solid sphere (denoted as CSS, Supplementary Fig. 1b, see Section 2 of Methods) with a WCA of 40.71° (Fig. 2m), which was synthesized by a modified method according to previous studies.²¹ The second one was the superhydrophobic Cu₂O solid sphere (denoted as O-CSS, Supplementary Fig. 1b) with a WCA of 152.17° (Fig. 2n), which was also obtained through hydrophobic modification of CSS by 1-dodecanethiol. Last but not least, the third one was the superhydrophobic Cu₂O nanosheet without the enclosed space (denoted as O-CNS, Supplementary Fig. 1c, see Section 3 and 4 of Methods), which came from the broken CHS and was also hydrophobically modified by 1-dodecanethiol. It should be emphasized that the only difference between O-CHS and O-CNS lies in that the former structure possesses an enclosed gas storage space, while the latter loses this space. The WCA of O-CNS is detected to be 152.39° (Fig. 2o), demonstrating the superhydrophobic property similar to that of O-CHS and O-CSS.

The transmission electron microscopy (TEM) and high-resolution TEM (HRTEM) images of the as-synthesized CHS are shown in Fig. 2a-c. The representative TEM image in Fig. 2a combined with the lattice spacing shown in Fig. 2b reveals the successful synthesis of the Cu₂O hollow structure with an enclosed space. The universal existence of CHS can be demonstrated by the TEM image of a larger vision (Fig. 2c). The TEM and HRTEM images of O-CHS shown in Fig. 2d-f are similar to corresponding images of CHS, indicating that the hollow structure of O-CHS remains intact during the hydrophobic modification. The scanning electron microscopy (SEM) image of O-CHS (Supplementary Fig. 2) also shows the uniformly synthesized hollow structures. Figure 2i-l show the Energy-dispersive X-ray spectroscopy (EDS) mapping images of O-CHS, where the mapping of S present in 1-dodecanethiol overlaps with that of Cu, indicating that 1-dodecanethiol is uniformly distributed on the O-CHS surface. Additionally, the TEM and HRTEM images of CSS (Fig. 2m, Supplementary Fig. 3a), O-CSS (Fig. 2n, Supplementary Fig. 3b) and O-CNS (Fig. 2o, Supplementary Fig. 3c) also demonstrate the successful syntheses of corresponding structures.

The chemical composition of every sample (CHS, O-CHS, CSS, O-CSS, and O-CNS) is detected by X-ray diffraction (XRD). As shown in Fig. 3a, all five samples exhibit quite similar XRD patterns, which are well matched with the cubic phase Cu₂O (JCPDS Card No. 78-2076). The chemical composition can be further detected by X-ray photoelectron spectroscopy (XPS). As shown in the Cu 2p XPS spectra (Fig. 3b), for all the samples, two strong peaks centered at ~ 932.4 and ~ 952.4 eV can be observed, which should be attributed to Cu^I combined with the results of Cu LMM Auger spectra (Fig. 3c)^{22, 23}. Although the signals at ~ 933.7 and ~ 953.6 eV belonging to Cu^II can also be observed, they are much weaker than those of Cu^I, indicating the main component of the samples is indeed Cu₂O. In particular, the Cu 2p peak positions show no obvious change before and after hydrophobic modification, indicating that the electronic structures are maintained during the hydrophobic process.²⁴

In addition, the signals of S present in the hydrophobic modifier (1-dodecanethiol) are detected in XPS survey spectra (Supplementary Fig. 4a) for O-CHS, O-CSS and O-CNS and the existing state of the hydrophobic modifier on the Cu₂O surface can be inferred based on the S 2p XPS spectra. As shown in Fig. 3d, for all the O-CHS, O-CSS and O-CNS, a strong peak at ~ 162.3 eV can be observed, suggesting the formation of S-Cu^I bond.^{25, 26} Thus, it can be inferred that the 1-dodecanethiol is grafted on the Cu₂O surface through the chemical bond between S and Cu atoms. Whereas S 2p_1/2 peaks at 163.6 eV are attributed to the residual unbound thiols.^{27, 28} This primarily proves that chemical adsorption takes place upon the modification of the Cu₂O structure by 1-dodecanethiol, which is further confirmed by FTIR. As shown in Fig. 3e, in the FTIR spectra of 1-dodecanethiol modified O-CHS, O-CSS and O-CNS, four new peaks at 719, 2852, 2922 and 2958 cm^− 1 appear, which can be assigned to the rocking vibration of long-chain alkyl, the symmetric stretching of -CH₂-, the asymmetric stretching of -CH₂- and the asymmetric stretching of -CH₃, respectively, and all the four peaks belong to 1-dodecanethiol.²⁹ However, the peak at 2570 cm^− 1 belonging to S-H stretching vibration cannot be observed in O-CHS, O-CSS and O-CNS, indicating that S-H is destroyed.²⁸ Thus, it is safe to infer that during the hydrophobic modification process, the S-H bond in 1-dodecanethiol is broken firstly, and then S-Cu^I bond formed to graft 1-dodecanethiol on the surface of Cu₂O (Supplementary Fig. 5)²⁷. Generally, compared with physical adsorption, species adsorbed to surfaces via chemical bonds are more difficult to desorb, ensuring good hydrophobic stability of catalysts, which is favorable for CRR. As a result, even after the Xe lamp irradiation for 5 hours, or a long storage time of 20 days, the hydrophobicity can still be well maintained (Supplementary Fig. 6). It should be noted that the amount of 1-dodecanethiol grafted on the surface of the target catalyst is small enough to avoid the hindrance of the surface charge transfer during the reaction (Supplementary Fig. 4b), which can be well proved by two Ag⁺ probe experiments (Supplementary Figs. 7 and 8, see Section 5 of Methods).³⁰

Subsequently, the ultraviolet-visible diffuse reflectance spectra (UV-vis) were adopted to investigate the situation of light absorption for the five samples (CHS, O-CHS, CSS, O-CSS, and O-CNS). Results show that all samples can absorb not only ultraviolet but also visible light, indicating a sufficient utilization of solar energy (Supplementary Fig. 9a).³¹ Furthermore, their bandgaps estimated from the Tauc plots (Supplementary Fig. 9b) are similar with a value of ~ 2.19 eV, manifesting their silimar light absorption abilities. Additionally, the efficiency of charge separation can be evaluated by photoluminescence (PL) spectra (Supplementary Fig. 9c), with stronger PL intensity indicating poorer stability of the separated charges.³² The slightly lower PL intensities of CHS, O-CHS and O-CNS than those of O-CSS and CSS indicate the marginally higher charge-separation efficiency, which could be attributed to the shortened migration distance resulting from the provided thin-walled structures of hollow structures or nanosheets. Additionally, the PL intensities of 1-dodecanethiol treated O-CHS and O-CSS are similar to those of unmodified CHS and CSS, implying that the modification has almost no effect on the charge separation of catalysts. Nitrogen adsorption-desorption isotherms (Supplementary Fig. 10) show the type IV curves with H3 hysteresis for all catalysts, confirming the mesoporous structures of all samples, which allow efficient mass transport during the reaction.³³

In order to verify the existence of the gas storage space for O-CHS, we applied total internal reflection fluorescence microscope (TIRFM) to directly observe the local distributions of gas and liquid surrounding the nanostructure of the catalyst for the first time (see Section 6 of Methods). The water phase was labeled with water-soluble Cy5 NHS ester dye (fluorescent molecule), of which a strong red emission under 635 nm excitation can outline the boundary morphology of the water phase. Thus, a region with the fluorescent signal means the presence of water, while no fluorescent signal means no water distribution (Supplementary Fig. 11). As shown in Fig. 3f, the TIRFM images of O-CHS show a nearly dark field inside the hollow shell, while outside the hollow shell, the strong fluorescent signal can be observed. This indicates that water can be repealed outside the inner space, which can be used for gas storage. However, for CHS, the obvious fluorescent signal can be seen both inside and outside the hollow space, suggesting that the hydrophilic shell of CHS cannot repel water out from the inner space (Fig. 3g). It should be pointed out that the fluorescent signal near the Cu₂O shell is always stronger than that in aqueous solution, which may be attributed to the adsorption effect of the mesoporous Cu₂O shell on dyes.^{34, 35} The results suggest that O-CHS with superhydrophobic surfaces indeed has the capability of gas storage, providing the basis for high local CO₂ concentration on the catalyst surface.

Photocatalytic Crr Performance

The photocatalytic overall CRR performances for the target catalyst (O-CHS) and all the reference catalysts (CSS, O-CSS, CHS and O-CNS) are evaluated in aqueous solution under visible-light irradiation without any sacrificial agent (see Section 7 of Methods). The products are detected by gas chromatography (GC) and calculated by a general algorithm based on original data (Supplementary Figs. 12 and 13). The results are shown in Fig. 4a.

The main product of CSS is CH₃OH with a generation rate of 205.45 µmol g^− 1 h^− 1. Only a few H₂ with 13.50 µmol g^− 1 h^− 1 can be detected, resulting in a good selectivity of carbon derivatives of 99.36% (Supplementary Fig. 14a), which can be attributed to the specific inhibiting effect of the Cu₂O material on HER.³⁶ After the hydrophobic modification to form O-CSS, the generation rate of CH₃OH is elevated to 473.27 µmol g^− 1 h^− 1, and the selectivity of carbon derivatives is also increased to 99.88%. This result can be attributed to the mass transfer enhancement of CO₂. The hydrophobic Cu₂O surface enables higher CO₂ and lower H⁺ concentration, and thus the HER can be further suppressed, while the performance of the CRR is improved.¹⁶

For CHS, the main product is also CH₃OH with a relatively higher generation rate of 797.71 µmol g^− 1 h^− 1, which can be attributed to the shortened charge transfer distance, alleviated carrier recombination and weakened self-corrosion brought by the thin shell of the hollow structure.³⁷ In addition, as evidenced by TIRFM measurements (Fig. 3f and g), even if a hollow space appears in CHS, the CHS has no gas storage capability. However, for O-CHS, the superhydrophobic shell endows this structure gas storage capability, enabling the high concentration and short mass transfer distance for CO₂ molecules surrounding active sites. Finally, different from CSS, O-CSS and CHS, the O-CHS (the target catalyst in this work) exhibits an impressive ethanol generation rate of 996.18 µmol g^− 1 h^− 1, which is 156, 131 and 132 times higher than that of CSS, O-CSS and CHS. Such a high activity makes the O-CHS one of the most effective heterogeneous photocatalysts for ethanol production at ambient temperature pressure (generally 0.37 ~ 545 µmol g^− 1 h^− 1, as shown in Supplementary Fig. 15 and Supplementary Table 8). In addition, with O-CHS, the selectivity of ethanol is up to 59.59%, which is 21, 10 and 31 times higher than that of CSS, O-CSS and CHS (Fig. 4a). The excellent CO₂-to-ethanol performance of O-CHS can be further evidenced by the results of time-dependent activity tests. As shown in Fig. 4b, the CH₃CH₂OH generation over O-CHS increases rapidly with reaction time, which is remarkably faster than all the reference catalysts. Additionally, with O-CHS, the CH₃OH generation is also elevated to 1073.04 µmol g^− 1 h^− 1, enabling high selectivity and AQE of liquid carbon derivatives with 99.97% and 1.63%, respectively, because of the high local CO₂ concentration.

The powerful effect of the superhydrophobic gas storage space can be further evidenced by the CRR performance of O-CNS. As mentioned above, the only difference between O-CNS and O-CHS is that the former no longer has the enclosed gas storage space (Fig. 2o). Figure 4a shows that after the loss of this space, O-CNS cannot maintain the encouraging performance like O-CHS, especially for the ethanol generation.

Concomitantly, oxygen (O₂) in synchrony with carbon derivatives is detected, suggesting that the designed photocatalyst enables the photocatalytic overall CRR coupling the H₂O oxidation with CO₂ reduction, albeit in a substoichiometric ratio (Fig. 4a). Substoichiometric O₂ production is typically less than that of the main catalytic products³⁸. Some of the photogenerated holes may be consumed in the photo-induced disproportionation of Cu₂O nanoparticles, the formation of H₂O₂ (see Section 10 of Methods and Supplementary Fig. 14b) and/or the re-oxidation of carbon derivatives.³⁹ To prove that catalysts have a suitable electronic band structure for the CO₂ reduction and H₂O oxidation reactions, the valence-band XPS spectra of O-CHS and CHS are examined.⁴⁰ As shown in Fig. 4c, the valence-band (VB) positions are determined to be 1.18 eV for O-CHS and CHS, consistent with the published literature.⁴¹ By incorporating the bandgaps calculated from Supplementary Fig. 9, the band structure alignments are schematically presented in Fig. 4d. The conduction-band (CB) positions are more negative than all of the reduction potentials of CRR products, which could provide a strong driving force for photocatalytic CRR thermodynamically.⁴² Strikingly, the VB edges are also located below the oxidation potential of H₂O (E (H₂O/O₂) = 0.82 V at pH = 7).^{41, 43} Therefore, the oxygen evolution reaction (OER) driven by photogenerated holes is also a thermodynamically feasible process, in accord with the O₂ generation shown in Fig. 4a.

The control experiment with the target catalyst (O-CHS) under the atmosphere of Ar instead of CO₂ is implemented to investigate the carbon source of generated carbon derivatives. Results show that with the absence of CO₂, no carbon derivates can be detected (Supplementary Fig. 14c), proving that the carbon derivates originate from CO₂ reduction. The carbon source can be further investigated by an isotope experiment using ¹³CO₂ as the reactant. Results obtained from gas chromatography-mass spectrometry (GC-MS) show the peaks with mass-to-charge ratio (m/z) of 32, 47, and 48 for ethanol (Supplementary Fig. 16a), corresponding to the fragments of ¹³CH₂OH, ¹³CH₃¹³CHOH, and ¹³CH₃¹³CH₂OH of ¹³CH₃CH₂OH.^{6, 44} Meanwhile, the peaks with m/z = 32 and 33 for methanol are assigned to the fragments of ¹³CH₃O and ¹³CH₃OH (Supplementary Fig. 16b).⁴⁵ These results confirm that carbon derivates indeed come from CO₂ rather than other carbon sources in the reaction. Other control experiments of no irradiation or catalyst are also implemented, and results indicate that the CRR is indeed driven by light irradiation and the photocatalyst.

The stability of O-CHS and CHS are also investigated under the reaction condition within 6 hours. Results (Supplementary Fig. 14d) show that the generation rate of carbon derivatives declines after three hours for both O-CHS and CHS, which is a general phenomenon for Cu₂O-based photocatalysts.⁴¹ However, the O-CHS exhibits better stability than CHS, since the generation rate of carbon derivatives can still maintain 74.49% of the top rate after 6 h with O-CHS, while it can only maintain 7.18% with CHS. The reason is that the morphology and chemical composition of O-CHS can be better maintained after the reaction, which can be supported by results of SEM, XRD and XPS for spent O-CHS and CHS after the reaction (Supplementary Fig. 17). This fact can be attributed to its strong ability to repel H₂O, which often participates in the self-corrosion reaction of Cu₂O^{46, 47}. Meanwhile, the faster reaction rate over O-CHS can avoid the accumulation of charge carriers on Cu₂O, which is also beneficial for the inhibition of self-corrosion^{48, 49}.

Mechanism Study

The mechanism is investigated to illustrate the enhanced C-C coupling during photocatalytic overall CRR by the creation of the superhydrophobic gas storage space. In situ FTIR was firstly employed to detect the reaction intermediates for O-CHS and CHS (see Section 12 of Methods). As depicted in Fig. 5a-b, there is no obvious peak before light irradiation (at 0 min), but several peaks gradually appear and intensify between 1220 to 2120 cm^− 1 with an extension of the irradiation time. The main peak at 1647 cm^− 1 can be attributed to the signal for the carbonyl group absorption of CO₂.^{50, 51} Another main peak at 1545 cm^− 1 ascribed to the *COOH group (the asterisk denotes the surface adsorption site) demonstrates that *COOH is a key intermediate of the CO₂ conversion process.^{40, 52} The interaction between CO₂ and H₂O co-adsorbed on the Cu₂O surface also leads to the monodentate carbonate (m-CO₃²⁻) peaks at 1507 and 1558 cm^− 1.^{40, 53, 54} The absorption peak at 2077 cm^− 1 corresponds to the *CO specie, which is generated through a further *COOH intermediate protonation process.⁵⁵ *CHO is also observed at 1733 and 1760 cm^− 1, proving that the hydrogenation of *CO occurs in the photocatalysis.^{29, 56} More importantly, two obvious peaks at 1313 and 1520 cm^− 1 also be observed after 70 min of illumination, which can be attributed to the vibrational signature of *OCCHO, implying the occurrence of C-C coupling.^{57, 58} What’s more, the peak at 1338 cm^− 1 is indexed to the absorbed *OC₂H₅, consistent with the experimental results of ethanol generation.^{59, 60} It should be noted that peaks existing in Fig. 5a also appear in Fig. 5b, indicating that the same kind of intermediates is generated over O-CHS and CHS during the reaction. However, the intensities of peaks in Fig. 5a and Fig. 5b are different, indicating that concentrations of intermediates are different under the same irradiation time. As shown in Fig. 5c, concentrations of both *COOH and *CO increase rapidly from 0 to 20 min for both O-CHS and CHS. After that, the peak intensities are nearly stable for CHS. While for O-CHS, the peak intensities still increase rapidly with the extension of the reaction time. This indicates that the effective CO₂ adsorption in the form of *COOH is indeed enhanced over O-CHS, leading to the enhanced formation of *CO. This is a result of the high-efficient mass transfer of CO₂ to the surface and further improves the CO₂ reduction performance.^{31, 40} Additionally, the boosted mass transfer of CO₂ for O-CHS can also be evidenced by the results of CO₂ dissolution tests (see Section 13 of Methods and Supplementary Fig. 18). It can be seen that the amount of CO₂ in the gas phase decreases more quickly in the system containing O-CHS than that of CHS, corresponding to the rapid increase of dissolved CO₂ in the liquid phase. This suggests that O-CHS in the suspension accelerates the dissolution and diffusion of CO₂, consistent with the analyses of in situ FTIR spectra.

Based on the above analyses of FTIR spectra, the promotion of the generation of carbon derivates by highly concentrated local CO₂ over Cu₂O surface can be further proved by DFT calculations (see Section 14 of Methods, Supplementary Fig. 19–24 and Supplementary Table 1–5). For the reaction path of ethanol generation or methanol generation, it is found to be the same on different catalysts and it is also summarized in Fig. 5d (Supplementary Tables 6 and 7) according to the DFT calculations.

Since O-CHS which is surrounded by higher CO₂ concentration could enrich more *CO on the surface compared with CHS, as proved by Fig. 5c, different *CO coverages are adopted to model the reaction processes under different local CO₂ concentrations.^{61, 62}. As revealed in Fig. 5d and e, no matter for the high *CO coverage (HCC) or the low *CO coverage (LCC), ethanol is formed following the same process, and so does methanol. For ethanol generation, the C-C coupling step exhibits the highest energy barrier (denoted as ΔG_RDS1), and thus it is regarded as the rate-determining step (RDS) for ethanol formation. Similarly, for the methanol generation, the *CH₃OH desorption step exhibits the highest energy barrier (denoted as ΔG_RDS2), and thus it is regarded as the RDS for the methanol formation. As shown in Fig. 5f, compared with the LCC state, the HCC state exhibits lower ΔG_RDS1 and ΔG_RDS2, benefiting to the generation of ethanol and methanol, respectively. In addition, the difference of ΔG_RDS1 and ΔG_RDS2 (ΔG_RDS1-ΔG_RDS2) can be used as the descriptor for the tendency to ethanol vs. methanol generation, which can reasonably describe the CRR selectivity, where a lower ΔG_RDS1-ΔG_RDS2 value means a higher selectivity toward the generation of ethanol.^{10, 62} The result is that the HCC state exhibits a lower ΔG_RDS1-ΔG_RDS2 value, indicating its higher selectivity of ethanol compared with the LCC state, which is in accord with the experimental results (Fig. 4a). Overall, based on the DFT calculation results we can confirm that higher local CO₂ concentration is beneficial for the generation of carbon derivates, especially for the ethanol generation, reinforcing our conclusion that the gas storage space induced sufficient CO₂ makes the O-CHS more efficient for the formation of carbon derivates, especially for the ethanol formation.

In summary, we report a strategy to improve the activity and selectivity for the production of liquid multicarbon products represented by ethanol, by just curling the superhydrophobic Cu₂O to create an enclosed gas storage space. The designed O-CHS can efficiently repel water out, which can be well proved for the first time by TIRFM, and thus store CO₂ into its interior space. While the stored CO₂ can realize continuous CO₂ supply from the interior to the catalyst surface, shortening the mass transfer distance of CO₂ and balancing the consumption and supply of CO₂, thereby ensuring the high local CO₂ concentration on the catalyst surface. Further investigations based on in situ FTIR and DFT indicate that the enriched CO₂ concentration on O-CHS lowers the energy barrier of the rate-determining steps during the generation of both ethanol and methanol, enhancing the photocatalytic activity for both ethanol and methanol generation, and also the enriched CO₂ concentration dramatically elevates the selectivity towards ethanol. Finally, the target catalyst O-CHS exhibits high AQE of liquid products with 1.63%, high activity and selectivity of ethanol generation with 996.18 µmol g^− 1 h^− 1 and 59.59%, respectively, during overall photocatalytic CRR. Through this investigation, we provide a compatible geometric operation to facilitate the liquid products, especially multicarbon products, which is promising to develop more efficient photocatalysts for multicarbon product generation combined with traditional methods.

Section 1: Synthesis of Cu ₂ O hollow sphere (CHS). An aqueous solution was first prepared by mixing 20 mL of DI water, 0.4 mL of PVP solution (M.W = 30000, 1.0 wt %), 2.0 mL of (C₄H₉)₄NBr solution (0.1 M) and 0.2 mL of CuCl₂ solution (0.1 M). Then 1.0 mL of ascorbic acid solution (0.1 M) was added to the solution. The mixture was stirred for 3 min and became turbid shortly. Afterwards, 1.0 mL of NaOH solution (0.2 M) was added to the above turbid precursor solution under stirring, and the resulting solution was then kept for 15 min at room temperature without stirring. The resultant Cu₂O product was collected by centrifugation, washed with DI water and ethanol several times, and dried in a vacuum oven.

Section 2: Synthesis of Cu ₂ O solid sphere (CSS). 28.6 mg of Cu(NO₃)₂·3H₂O and 57.8 mg of (NH₄)₂SO₄ were added into 50 mL of DI water and stirred for 5 min. Then 2.0 mL of 0.2 M NaOH solution was dropwise injected into the above solution and stirred magnetically for 2 min, followed by the dropwise addition of 3.5 mL of ascorbic acid (0.1 M). Finally, the solution was stirred for another 13 min. The resultant Cu₂O product was collected by centrifugation, washed with DI water and ethanol several times, and dried in a vacuum oven.

Section 3: Synthesis of Cu ₂ O nanosheets derived from damaged CHS (CNS). The synthetic process of Cu₂O NP was similar to the synthesis of CHS. The difference was that the standing time was extended from 15 min to 45 min.

Section 4: Modified the surfaces to synthesize superhydrophobic Cu ₂ O structures. The as-prepared Cu₂O samples of CHS, CSS, and CNS were all treated with 1-dodecanethiol to obtain superhydrophobic surfaces. These superhydrophobic Cu₂O structures were denoted as O-CHS, O-CSS, and O-CNS, respectively. In a typical procedure, 10 mg of Cu₂O was ultrasonically dispersed in 10 mL of ethanol, followed by the injection of 7 µL of 1-dodecanethiol. The mixture was then kept for 15 min at room temperature under gentle stirring. The resultant superhydrophobic Cu₂O was collected by centrifugation, washed with ethanol several times, and dried in a vacuum oven.

Section 5: Ag ⁺ Probe Experiment. Probe experiments were conducted to verify the hydrophobic treatment could not hinder the migration of charge carriers. Ag⁺ (from AgNO₃) was used as a probe ion and methanol was used as a positive charge scavenger. Typically, 15 mg of O-CHS was dispersed into 30 mL of H₂O. Then 15 mL of methanol and 250 µL of AgNO₃ solution (20 mg mL^− 1) were added. After 0.5 h stirring and evacuation, the mixture was irradiated by a 300 W Xe lamp (λ > 420 nm) for 3 h. After centrifugation, washing, and drying procedures, the Ag-loaded O-CHS sample was obtained. The protocol for the preparation of the Ag-loaded CHS sample was the same for comparison.

Section 6: Total internal reflection fluorescent microscope measurement. Total internal reflection fluorescence microscope (TIRFM) measurements were carried out on Nikon A1. The specific process was as follows. Firstly, Cy5 NHS ester was dissolved in DMSO with a concentration of 85 mM. Then 1 µL of Cy5 NHS ester solution (85 mM in DMSO) was added into 2 mL of H₂O to form aqueous solution A. Meanwhile, 5 mg of catalyst was dispersed into 1 mL of H₂O and then 10 µL of solution A was added to form solution B. After that, solution B was wrapped in tinfoil to avoid light and placed in the refrigerator overnight. Then take 100 µL of obtained solution B and drop it on the cleaned coverslip. After 2 h, the coverslip was placed on TIRFM for imaging using direct stochastic optical reconstruction microscopy with an excitation of 635 nm.

Section 7: CO ₂ photoreduction tests. Photocatalytic performances of catalysts were completed using Labsolar 6A (PerfectLight, China) and gas chromatography (GC, Agilent 8890, America).

Before illumination, 20 mg of photocatalyst was suspended in 100 mL of KHCO₃ (0.1 M) solution without using hole scavenger under magnetic stirring in the double-layer Pyrex glass reactor (PerfectLight, China). Connect the reactor to Labsolar 6A, constituting the system for photoreduction tests (Supplementary Fig. 18a). Then the system was evacuated to remove air. Subsequently, compressed CO₂ (99.999%), which was controlled by a mass flow controller (LZB-3WB, China), was bubbled into the reactor to raise the system pressure to 80 kPa and then the system was evacuated again. Repeat such operation three times to thoroughly eliminate air and dissolved oxygen. Finally, the pure CO₂ gas was continuously bubbled through the solution again at a 40 mL min^− 1 flow rate for about 30 min to raise the CO₂ pressure to 80 kPa under a constant stirring speed of 250 r min^− 1. Then it was maintained for 40 min to establish an adsorption-desorption balance and saturate the solution with CO₂. There was no more CO₂ injected into the closed system during the reaction and also powders were continuously agitated by a magnetic stirrer. A 300 W Xe lamp (PLS-SXE300, PerfectLight, China) equipped with a cutoff filter (λ > 420 nm) was then used as the visible light source to illuminate the reactant through the quartz window. Using cooling water circulation controlled by a chiller (CNSHP DC-0506, China), the photoreaction temperature was maintained at 5°C. The photocatalytic reaction was typically performed for 6 h and 0.5 mL of liquid product was taken out using a syringe every one hour for later analysis.

The amount of gas and liquid products evolved was all determined using GC. For gas products, they were detected directly by GC, which was equipped with a thermal conductivity detector (TCD), a flame ionization detector (FID) and a methanation reactor. For liquid products, the samples obtained each hour were injected into GC through a headspace sampler (APL HS-20, China) for analysis. Specifically, 200 µL of the extracted liquid was mixed with 5 mL of H₂O in each headspace bottle and then place the headspace bottle into the headspace sampler. The liquid products were automatically injected into the GC for detection. Three replicates were collected for each catalyst and an external standard was used to calculate the amount of every component from the peak area of GC.

Under the same experimental conditions, the comparative experiments were carried out without Xe lamp irradiation or without catalysts. Meanwhile, using Ar instead of CO₂ gas flow as feed gas was also conducted as a comparative experiment to verify that the carbonaceous products were from CO₂ photocatalytic conversion.

Section 8: Calculation of the selectivity. The selectivity of CO₂ photoreduction on an electron basis (6e⁻ for CH₃OH, 12e⁻ for CH₃CH₂OH, 2e⁻ for H₂) was calculated using the following equation:

$$\text{Selectivity of carbon derivatives}\text{ }\left(\text{\%}\right)\text{=}\frac{\text{6}\text{n}\text{(}{\text{CH}}_{\text{3}}\text{OH)+12}\text{n}\text{(}{\text{CH}}_{\text{3}}{\text{CH}}_{\text{2}}\text{OH)}}{\text{6}\text{n}\text{(}{\text{CH}}_{\text{3}}\text{OH)+12}\text{n}\text{(}{\text{CH}}_{\text{3}}{\text{CH}}_{\text{2}}\text{OH)+2}\text{n}\text{(}{\text{H}}_{\text{2}}\text{)}}\text{×100\%}$$

$$\text{Selectivity of ethanol }\left(\text{\%}\right)\text{=}\frac{\text{12}\text{n}\text{(}{\text{CH}}_{\text{3}}{\text{CH}}_{\text{2}}\text{OH)}}{\text{6}\text{n}\text{(}{\text{CH}}_{\text{3}}\text{OH)+12}\text{n}\text{(}{\text{CH}}_{\text{3}}{\text{CH}}_{\text{2}}\text{OH)+2}\text{n}\text{(}{\text{H}}_{\text{2}}\text{)}}\text{×100\%}$$

where n(CH₃OH), n(CH₃CH₂OH) and n(H₂) were the amounts (moles) of CH₃OH, CH₃CH₂OH and H₂ formed within a certain period of time, respectively.

Section 9: Apparent quantum efficiency (AQE) test for photocatalytic CO ₂ reduction. The photocatalytic activity of the target catalyst at a specific excitation wavelength was obtained by a specific band pass filter (λ = 510 nm) with a 300W Xe lamp in the same apparatus as the photocatalytic CO₂ reduction reaction. Generally, the AQE is calculated as follows:

$$\text{AQE }\left(\text{\%}\right)\text{=}\frac{\text{number of the electrons taking part in reaction}}{\text{number of incident photons}}\text{×}\text{100\%}$$

Thus, for the photocatalytic reduction of CO₂ to ethanol and methanol, the AQE is calculated by the following equation:

$$\text{AQE }\left(\text{\%}\right)\text{=}\frac{\left[\text{6}\text{n}\left({\text{CH}}_{\text{3}}\text{OH}\right)\text{+12}\text{n}\left({\text{CH}}_{\text{3}}{\text{CH}}_{\text{2}}\text{OH}\right)\right]\text{×}{\text{N}}_{\text{A}}}{{\text{N}}_{\text{P}}}\text{×}\text{100\%}$$

where n(CH₃OH) and n(CH₃CH₂OH) are the amounts (moles) of CH₃OH and CH₃CH₂OH formed within 1 h, respectively. N_A is the Avogadro constant (6.02×10²³ mol^− 1), and N_P is the number of incident photons. In the actual measurement, it is assumed that all incident photons are absorbed by the suspension; the calculation formula of N_P is as follows:

$${\text{N}}_{\text{P}}\text{=}\text{PSt}\frac{\text{λ}}{\text{hc}}$$

where P is the average intensity of the radiation (P = 0.36×10^− 3 W cm^− 2), S is the incident irradiation area (S = 38.465 cm²), t is the irradiation time (t = 1×3600 s), λ is the incident wavelength, h is the Planck constant (6.626×10^− 34 J s), and c is the speed of light (3×10⁸ m s^− 1).

Section 10: Detection of hydrogen peroxide (H ₂ O ₂ ) in the reaction solution. The detection of H₂O₂ in the reaction solution was conducted through a chemiluminescence approach by scopoletin assay.^{63, 64} Firstly, the detecting solution was prepared by mixing 20 mL of H₂O, 5 mL of HCl (0.1 M), 500 µL of scopoletin solution (0.7 mg L^− 1), and 1 mL of horseradish peroxidase solution (0.16 mg mL^− 1). Then, 10 mL of the reaction solution after photocatalytic CRR with different catalysts was added into the detecting solution. After incubation for 15 min, the final solution was measured by fluorescence spectrophotometer with excitation of 350 nm light. The reference experiment was conducted in the same manner except that adding 10 mL of H₂O instead of the reaction solution.

Section 11: Isotope-labelling Testing. The isotope labeling experiment was executed as follows. Firstly, 30 mg of O-CHS was dispersed in 50 mL of NaH¹³CO₃ (0.1 M) solution under ultrasound for 20 min. Then the reactor with O-CHS suspension was connected to Labsolar 6A, sealed, and purged by flowing Ar for 60 min to remove air. After that, ¹³CO₂ (99.999%) was pumped into the reactor and the final pressure for the reaction was raised to 80 kPa. The reactor was allowed to stabilize for 40 min and then the photocatalytic CRR was performed for 4 h using a 300 W Xe lamp with a cutoff filter (λ > 420 nm). The products containing C-isotope were analyzed by GC-MS (Shimadzu QP-2010SE).

Section 12: In situ FTIR test. Diffuse reflection in situ FTIR was used to investigate the effective CO₂ adsorption on the catalyst and detect possible intermediates during the photocatalytic CRR. 50 mg of catalyst was loaded on the sample cell of diffuse reflectance attachment and make sure the flat of the top surface. A cover was fixed on the sample holder to form a reaction space, which was then cleaned by argon. Then the FTIR data was collected as background. Subsequently, visible light was introduced to the reaction space by the observation window, and CO₂ was introduced through a container. It should be noted that the container was filled with water, making sure the formation of water film on the sample powders. Then FTIR data were collected every 10 min, and the change could be recorded.

Section 13: Determination of CO ₂ dissolution rate in the suspension of catalysts. To verify that the presence of hydrophobic materials in the water environment (the suspension of catalysts) was conducive to the dissolution and diffusion of CO₂, the temperature and pressure changes of the photoreduction measurement system during the dissolution of CO₂ in the catalysts suspension were recorded in detail. Then the amounts of CO₂ in the gas and liquid phases of the system at different dissolution times were calculated. As we described in Section 7: CO₂ photoreduction tests, the system experienced several processes of evacuation, bubbling CO₂ and finally the solution was saturated with CO₂. Thus, the pressure value displayed on the actuator of Labsolar 6A (Supplementary Fig. 18b) was the real-time CO₂ pressure in the gas phase of the system. So, if the change of pressure and temperature with time during CO₂ dissolution could be recorded, the change in the amount of CO₂ in the gas phase could be calculated using the van der Waals equation. Since the system was composed of Labsolar 6A and a reactor, the temperatures of the two parts were not the same. Therefore, the system should be divided into two parts to calculate. For Labsolar 6A, the corresponding temperature was displayed on the actuator (Supplementary Fig. 18b). While for the reactor (Supplementary Fig. 18c), it was kept constant at 5°C due to the cooling water circulation. It was noted that for the comparability of different catalysts, the CO₂ flow rate (40 mL min^− 1), total system pressure (80 kPa), magnetic stirring speed (250 r min^− 1) and cooling water temperature (5°C) were consistent. The van der Waals equation was as follows:

$$\left(\text{p}\text{+}\frac{{\text{n}}^{\text{2}}\text{a}}{{\text{V}}^{\text{2}}}\right)\left(\text{V}\text{-}\text{nb}\right)\text{=}\text{nRT}$$

where p was the pressure of the system displayed on the actuator of Labsolar 6A. n was the amount of CO₂. V was the volume of gas. Rwas molar gas constant. T was the temperature. a and b were van der Waals constants. For CO₂, a = 0.3658 Pa·m⁶ mol^− 2, b = 0.429×10^− 4 m³·mol^− 1. n was calculated using the software of Mathematica.

For the part of Labsolar 6A, the V₁ was 100 mL and p₁, T₁ were all displayed on the actuator. For the part of the reactor, the total volume of the reactor was 370 mL, but the suspension of catalysts occupies a volume of 100 mL. Thus, the volume occupied by the CO₂ in the gas phase of the reactor (V₂) was 270 mL. Meanwhile, T₂ of this part was 5°C. Since the pressure of the system was in equilibrium, p₂ was equal to p₁. The sum of corresponding n₁ and n₂ was the total amount of CO₂ in the gas phase. Therefore, the total amount of CO₂ in the gas phase at different dissolution times could be calculated. The decrease of the total amount of CO₂ in the gas phase was the amount of CO₂ dissolved into the liquid phase. Thus, the relationship between the amount of CO₂ dissolved into the liquid phase and its dissolution time could be obtained, and that was the dissolution rate of CO₂.

Section 14: Computational methods. All the optimized geometries and electronic energies based on the spin-polarized DFT were performed using the Vienna ab initio simulation package. Exchange correlation functional was treated within the generalized gradient approximation (GGA) and parameterized by Perdew-Burke-Ernzerhof (PBE) formula to describe electronic exchange and correlation effect. The van der Waals interactions were described using the empirical correction in Grimme’s scheme (DFT-D3). The cutoff energy for the plane wave basis sets was 500 eV, and methfessel-paxton (width = 0.1 eV) was used to calculate the partial occupancies. And a vacuum layer with a thickness of 20.00 Å was placed along the vertical direction in order to avoid spurious interactions. The Brillouin zone integration was performed using Monkhorst-Pack special k-points mesh of 3×3×1 for all the calculation models. The Cu₂O (111) surface was constructed and a 3×3 supercell with three layers was selected as the computational model. During the optimization, the atoms in the bottom two layers were fixed to maintain the bulk structure and the other atoms in the top layer were allowed to fully relax.

Acknowledgements

We acknowledge the National Natural Science Foundation of China (22008121, 11774173), the National Outstanding Youth Science Fund Project of National Natural Science Foundation of China (T2125004), the Fundamental Research Funds for the Central Universities (30920032204, 30920041115), the Foundation of State Key Laboratory of High-efficiency Utilization of Coal and Green Chemical Engineering (Grant No. 2022-K12), Funding of NJUST (No. TSXK2022D002) for financial support.

Author contributions

Hailing Huo performed, analyzed experiments and prepared the manuscript. Zhengwei Du conducted the DFT calculation. Hua He and Xin Guan performed the TIRFM measurements. Hongbin Xing helped with the synthesis of catalysts. Chengxi Huang and Yongping Du provided important suggestions for the DFT calculation. Fang Wu and Jingjing Ma provided suggestions for experimental characterizations. Ang Li conceived the project and Erjun Kan revised the manuscript.

Li, A., et al. Tunable syngas production from photocatalytic CO₂ reduction with mitigated charge recombination driven by spatially separated cocatalysts. Chem. Sci. 9, 5334–5340 (2018).
Lin, H., Luo, S., Zhang, H., Ye, J. Toward solar-driven carbon recycling. Joule 6, 294–314 (2022).
Ran, J., Jaroniec, M., Qiao, S. Z. Cocatalysts in semiconductor-based photocatalytic CO₂ reduction: achievements, challenges, and opportunities. Adv. Mater. 30, 1704649 (2018).
Zheng, Y., et al. Understanding the roadmap for electrochemical reduction of CO₂ to multi-carbon oxygenates and hydrocarbons on copper-based catalysts. J. Am. Chem. Soc. 141, 7646–7659 (2019).
Das, R., Das, K., Ray, B., Vinod, C. P., Peter, S. C. Green transformation of CO₂ to ethanol using water and sunlight by the combined effect of naturally abundant red phosphorus and Bi₂MoO₆. Energ. Environ. Sci. 15, 1967–1976 (2022).
Shi, H., et al. Atomically dispersed indium-copper dual-metal active sites promoting C-C Coupling for CO₂ photoreduction to ethanol. Angew. Chem. Int. Ed., e202208904 (2022).
Yang, B., et al. Accelerating CO₂ electroreduction to multicarbon products via synergistic electric-thermal field on copper nanoneedles. J. Am. Chem. Soc. 144, 3039–3049 (2022).
Albero, J., Peng, Y., García, H. Photocatalytic CO₂ reduction to C₂₊ products. ACS Catal. 10, 5734–5749 (2020).
Xie, S., et al. Photocatalytic and electrocatalytic transformations of C₁ molecules involving C-C coupling. Energ. Environ. Sci. 14, 37–89 (2021).
Wang, S., et al. Selective visible-light driven highly efficient photocatalytic reduction of CO₂ to C₂H₅OH by two-dimensional Cu₂S monolayers. Nanoscale Horiz. 6, 661–668 (2021).
Ding, L., et al. Over 70% faradaic efficiency for CO₂ electroreduction to ethanol enabled by potassium dopant-tuned interaction between copper sites and intermediates. Angew. Chem. Int. Ed. 61, e202209268 (2022).
Li, P., Liu, L., An, W., Wang, H., Cui, W. Efficient photothermal catalytic CO₂ reduction to CH₃CH₂OH over Cu₂O/g-C₃N₄ assisted by ionic liquids. Appl. Surf. Sci. 565, 150448 (2021).
Hurtado, L., Natividad, R., García, H. Photocatalytic activity of Cu₂O supported on multi layers graphene for CO₂ reduction by water under batch and continuous flow. Catal. Commun. 84, 30–35 (2016).
Bai, S., et al. Porous fixed-bed photoreactor for boosting C-C coupling in photocatalytic CO₂ reduction. eScience 2, 428–437 (2022).
Pham, T. H. M., et al. Enhanced electrocatalytic CO₂ reduction to C₂₊ products by adjusting the local reaction environment with polymer binders. Adv. Energy Mater. 12, 2103663 (2022).
Huang, H., et al. Triphase photocatalytic CO₂ reduction over silver-decorated titanium oxide at a gas-water boundary. Angew. Chem. Int. Ed. 134, e202200802 (2022).
Li, J., et al. Efficient electrocatalytic CO₂ reduction on a three-phase interface. Nat. Catal. 1, 592–600 (2018).
Lum, Y., Yue, B., Lobaccaro, P., Bell, A. T., Ager, J. W. Optimizing C-C coupling on oxide-derived copper catalysts for electrochemical CO₂ reduction. J. Phys. Chem. C 121, 14191–14203 (2017).
Gao, J., et al. One-pot synthesis of uniform Cu₂O and CuS hollow spheres and their optical limiting properties. Chem. Mater. 20, 6263–6269 (2008).
Larmour, I. A., Bell, S. E., Saunders, G. C. Remarkably simple fabrication of superhydrophobic surfaces using electroless galvanic deposition. Angew. Chem. Int. Ed. 46, 1710–1712 (2007).
Xiong, L., et al. Geometric modulation of local CO flux in Ag@Cu₂O nanoreactors for steering the CO₂RR pathway toward high-efficacy methane production. Adv. Mater. 33, e2101741 (2021).
An, X., Li, K., Tang, J. Cu₂O/reduced graphene oxide composites for the photocatalytic conversion of CO₂. ChemSusChem 7, 1086–1093 (2014).
Kim, C., et al. Cu/Cu₂O interconnected porous aerogel catalyst for highly productive electrosynthesis of ethanol from CO₂. Adv. Funct. Mater. 31, 2102142 (2021).
Wang, H., et al. Electrochemical tuning of vertically aligned MoS₂ nanofilms and its application in improving hydrogen evolution reaction. P. Natl. Acad. Sci. USA 110, 19701–19706 (2013).
Volmer, M., Stratmann, M., Viefhaus, H. Electrochemical and electron spectroscopic investigations of iron surfaces modified with thiols. Surf. Interface Anal. 16, 278–282 (1990).
Wakerley, D., et al. Bio-inspired hydrophobicity promotes CO₂ reduction on a Cu surface. Nat. Mater. 18, 1222–1227 (2019).
Yamamoto, Y., Nishihara, H., Aramaki, K. Self-assembled layers of alkanethiols on copper for protection against corrosion. J. Electrochem. Soc. 140, 436 (1993).
Kong, L.-H., Chen, X.-H., Yu, L.-G., Wu, Z.-S., Zhang, P.-Y. Superhydrophobic cuprous oxide nanostructures on phosphor-copper meshes and their oil-water separation and oil spill cleanup. ACS Appl. Mater. Interfaces 7, 2616–2625 (2015).
Speight, J. G. Lange's Handbook of Chemistry. McGraw-Hill Education (2017).
Liu, X., et al. Enhancement of mass and charge transfer during carbon dioxide photoreduction by enhanced surface hydrophobicity without a barrier layer. Chem. Eur. J. 28, e202201034 (2022).
Li, A., et al. Three-phase photocatalysis for the enhanced selectivity and activity of CO₂ reduction on a hydrophobic surface. Angew. Chem. Int. Ed. 58, 14549–14555 (2019).
Li, A., et al. Thin heterojunctions and spatially separated cocatalysts to simultaneously reduce bulk and surface recombination in photocatalysts. Angew. Chem. Int. Ed. 128, 13938–13942 (2016).
Wei, Y., et al. Steering hollow multishelled structures in photocatalysis: optimizing surface and mass transport. Adv. Mater. 32, e2002556 (2020).
Kuwahara, Y., et al. TiO₂ photocatalyst for degradation of organic compounds in water and air supported on highly hydrophobic FAU zeolite: structural, sorptive, and photocatalytic studies. J. Catal. 285, 223–234 (2012).
Gao, C., et al. Integrated oil separation and water purification by a double-layer TiO₂-based mesh. Energ. Environ. Sci. 6, 1147–1151 (2013).
Yang, P., et al. The functionality of surface hydroxy groups on the selectivity and activity of carbon dioxide reduction over cuprous oxide in aqueous solutions. Angew. Chem. Int. Ed. Engl. 57, 7724–7728 (2018).
Xiao, M., et al. Hollow nanostructures for photocatalysis: advantages and challenges. Adv. Mater. 31, e1801369 (2019).
Chueh, W. C., et al. High-flux solar-driven thermochemical dissociation of CO₂ and H₂O using nonstoichiometric ceria. Science 330, 1797–1801 (2010).
Wu, Y. A., et al. Facet-dependent active sites of a single Cu₂O particle photocatalyst for CO₂ reduction to methanol. Nat. Energy 4, 957–968 (2019).
Shangguan, W., et al. Molecular-level insight into photocatalytic CO₂ reduction with H₂O over Au nanoparticles by interband transitions. Nat. Commun. 13, 1–11 (2022).
Yu, L., et al. Enhanced activity and stability of carbon-decorated cuprous oxide mesoporous nanorods for CO₂ reduction in artificial photosynthesis. ACS Catal. 6, 6444–6454 (2016).
Yu, L., et al. Visible-light driven CO₂ reduction coupled with water oxidation on Cl-doped Cu₂O nanorods. Nano Energy 60, 576–582 (2019).
Abdul Nasir, J., et al. Photocatalytic Z-scheme overall water splitting: recent advances in theory and experiments. Adv. Mater. 33, 2105195 (2021).
An, B., et al. Cooperative copper centres in a metal-organic framework for selective conversion of CO₂ to ethanol. Nat. Catal. 2, 709–717 (2019).
Wang, Y., et al. Unique hole-accepting carbon-dots promoting selective carbon dioxide reduction nearly 100% to methanol by pure water. Nat. Commun. 11, 2531 (2020).
Toe, C. Y., Scott, J., Amal, R., Ng, Y. H. Recent advances in suppressing the photocorrosion of cuprous oxide for photocatalytic and photoelectrochemical energy conversion. J. Photoch. Photobio. C 40, 191–211 (2019).
Zhonghai Zhang, R. D., Lianbin Zhang, Haibo Zhu, Hongnan Zhang, and Peng Wang. Carbon-layer-protected cuprous oxide nanowire arrays for efficient water reduction. ACS Nano 7, 1709–1717 (2013).
Toe, C. Y., et al. Photocorrosion of cuprous oxide in hydrogen production: rationalising self-oxidation or self-reduction. Angew. Chem. Int. Ed. 57, 13613–13617 (2018).
Xia, Y., et al. Improving artificial photosynthesis over carbon nitride by gas-liquid-solid interface management for full light-induced CO₂ reduction to C₁ and C₂ fuels and O₂. ChemSusChem 13, 1730–1734 (2020).
Yang, Z.-Z., et al. Coordination effect-regulated CO₂ capture with an alkali metal onium salts/crown ether system. Green Chem. 16, 253–258 (2014).
Gonglach, S., et al. Molecular cobalt corrole complex for the heterogeneous electrocatalytic reduction of carbon dioxide. Nat. Commun. 10, 3864 (2019).
Li, X., et al. Ultrathin conductor enabling efficient IR light CO₂ reduction. J. Am. Chem. Soc. 141, 423–430 (2019).
Jiao, X., et al. Partially oxidized SnS₂ atomic layers achieving efficient visible-light-driven CO₂ reduction. J. Am. Chem. Soc. 139, 18044–18051 (2017).
Yu, H., et al. Three-in-one oxygen vacancies: whole visible-spectrum absorption, efficient charge separation, and surface site activation for robust CO₂ photoreduction. Angew. Chem. Int. Ed. 58, 3880–3884 (2019).
Wu, J., et al. Efficient visible-light-driven CO₂ reduction mediated by defect-engineered BiOBr atomic layers. Angew. Chem. Int. Ed. 57, 8719–8723 (2018).
Kim, Y., et al. Time-resolved observation of C-C coupling intermediates on Cu electrodes for selective electrochemical CO₂ reduction. Energ. Environ. Sci. 13, 4301–4311 (2020).
Perez-Gallent, E., Figueiredo, M. C., Calle-Vallejo, F., Koper, M. T. Spectroscopic observation of a hydrogenated CO dimer intermediate during CO reduction on Cu(100) electrodes. Angew. Chem. Int. Ed. 56, 3621–3624 (2017).
Ou, H., et al. Atomically dispersed Au-assisted C-C coupling on red phosphorus for CO₂ photoreduction to C₂H₆. J. Am. Chem. Soc. 144, 22075–22082 (2022).
Zhu, S., Li, T., Cai, W.-B., Shao, M. CO₂ electrochemical reduction as probed through infrared spectroscopy. Acs Energy Lett. 4, 682–689 (2019).
Wang, P., et al. Boosting electrocatalytic CO₂-to-ethanol production via asymmetric C-C coupling. Nat. Commun. 13, 1–11 (2022).
Zhang, X. Y., et al. Selective methane electrosynthesis enabled by a hydrophobic carbon coated copper core-shell architecture. Energ. Environ. Sci. 15, 234–243 (2022).
Wang, X., et al. Efficient methane electrosynthesis enabled by tuning local CO₂ availability. J. Am. Chem. Soc. 142, 3525–3531 (2020).
Perschke, H., Broda, E. Determination of very small amounts of hydrogen peroxide. Nature 190, 257–258 (1961).
Zhang, P., Wang, T., Chang, X., Zhang, L., Gong, J. Synergistic cocatalytic effect of carbon nanodots and Co₃O₄ nanoclusters for the photoelectrochemical water oxidation on hematite. Angew. Chem. 128, 5945–5949 (2016).

There is NO Competing Interest.

SupplementaryInformation.docx
Supplementary Information

Download PDF

Version 1

posted

You are reading this latest preprint version

Efficient Liquid Multicarbon Production from Overall CO₂ Photoreduction by Creating a Superhydrophobic Gas Storage Space

Status:

Version 1

Abstract

Figures

Introduction

Results And Discussion

Morphological And Structural Characterization

Photocatalytic Crr Performance

Mechanism Study

Conclusions

Methods

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1