Review Reports - Trisferrocenyltrithiophosphite-Copper(I) Bromide Composites for Electrochemical CO<sub>2</sub> Reduction

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript describes the catalytic performance of trisferrocenyltrithiophosphite-copper(I) bromide composites in CO2 electroreduction. CO2 can be considered as a cheap source of carbon for organic synthesis; therefore, this area is of great interest; a number of publications on this topic is huge. Conversion of CO2 to C2 products is relatively rare but is the most interesting. Typically, C2 products are formed as the small admixture to more common C1 products. What’s why high selectivity for methanol and ethanol production declared in the manuscript is of great interest.

However, attentive examination of the manuscript raised many questions. Some comments and questions are given below.

The introduction is focused on the mentioning of the properties of various P-containing ligands whereas the main idea of the manuscript is CO2 conversion to C1 and C2 compounds. Therefore, analysis of the previous reports on the catalytic systems, which were previously successfully used for CO2 conversion to C1 and, especially, to Cn (n>1) alcohols, is strongly required, with the emphasis on the structural preconditions.
The voltammetric curves given in Fig.1 look strange. What about their reproducibility? The current density observed for 1 and 2 in Fig.2 contradicts the catalytic current values observed in Fig.1
The amounts of all ingredients of the composites should be clearly stated
The authors wrote about synergistic interaction between the metal center and the phosphorus-sulfur ligand framework. More evidence should be provided. The voltammogram for CuBr-doped composite should be added in Fig.1
How relative intensity of the signals corresponding to MeOH and EtOH were measured? The detailed description of the chromatographic analysis should be provided. There is not a word about this in the Experimental part. Current efficiency for the alcohols formation should be determined. It is not possible to declare selectivity based on a short fragment of the chromatogram.
The results should be compared to the literature data related to CO2 conversion to EtOH based on Cu-containing catalysts, to outline the benefits of the composite.
4: the size of the particles is not shown. What is illustrated in Fig.4? What does it mean: “structural integrity”?
The structural study of the composite is perfectly done. But what conclusions can be made from comparison of the morphology prior and after electrolysis? The changes are quite expected, but what do they prove? The text is full of generalities on this topic (“profound structural transformation of the precatalyst”, “dynamic changes in the surface composition”, “highly dispersed nanoclusters that are too small or disordered to coherently diffract X-rays”, “particularly rich in the crucial Cu+ species” etc. And most important, how does this relates to the mechanism? The interrelation between the results of the surface study and mechanism will be outlined. The detailed analysis based on literature data should be provided.

To summarize, the results will be interesting and publishable when the important points mentioned above will be carefully addressed:

Author Response

However, attentive examination of the manuscript raised many questions. Some comments and questions are given below.

Q1. The introduction is focused on the mentioning of the properties of various P-containing ligands whereas the main idea of the manuscript is CO₂ conversion to C₁ and C₂ compounds. Therefore, analysis of the previous reports on the catalytic systems, which were previously successfully used for CO₂ conversion to C₁ and, especially, to C_n (n>1) alcohols, is strongly required, with the emphasis on the structural preconditions.

Q6. The results should be compared to the literature data related to CO₂ conversion to EtOH based on Cu-containing catalysts, to outline the benefits of the composite.

The authors express their gratitude to the Reviewer for such important comments. We have combined the answers to two questions to ensure the most complete and high-quality editing of the manuscript.

The structural features of copper catalysts significantly influence product distribution. For example, monitoring the crystallographic facets of copper revealed that Cu(100) faces and high-index surfaces facilitate C-C interactions of adsorbed *CO fragments, promoting the formation of C₂ products [Hua, Y., Zhu, C., Zhang, L. and Dong, F. (2024). Designing Surface and Interface Structures of Copper-Based Catalysts for Enhanced Electrochemical Reduction of CO2 to Alcohols, Materials, 17, pp. 600. DOI: 10.3390/ma17030600.].

An alternative approach to the selective reduction of CO₂ is the use of molecular complexes and single-atom catalytic sites, where the coordination environment is strictly determined by the ligand. For example, complex catalysts based on transition metals with macrocyclic ligands are capable of multielectron reduction. Cobalt phthalocyanine (CoPc) is a prime example of a homogeneous-heterogeneous electrocatalyst for the six-electron pathway to methanol production [Wu, Y., Jiang, Z., Lu, X., Liang, Y. and Wang, H. (2019). Domino electroreduction of CO2 to methanol on a molecular catalyst, Nature, 575, pp. 639–642. DOI: 10.1038/s41586-019-1760-8.]. CoPc immobilized on carbon nanotubes provides a selectivity for CH₃OH of approximately 44% at -0.94 V (relative hydrogen potential).

In the case of copper, isolated Cu atoms on the support or in the coordination shell of a ligand are generally not able to directly generate C2 products - at least two adjacent Cu centers are usually required for C-C bond formation [Karapinar, D., Huan, N. T., Ranjbar Sahraie, N., Li, J., Wakerley, D., Touati, N., et al. (2019). Electroreduction of CO2 on Single‐Site Copper‐Nitrogen‐Doped Carbon Material: Selective Formation of Ethanol and Reversible Restructuration of the Metal Sites, Angewandte Chemie International Edition, 58, pp. 15098–15103. DOI: 10.1002/anie.201907994.] However, recent studies show that individual Cu atoms can be converted into active nanoclusters (Cu₃ and Cu₄) in situ.[ Xu, H., Rebollar, D., He, H., Chong, L., Liu, Y., Liu, C., et al. (2020). Highly selective electrocatalytic CO2 reduction to ethanol by metallic clusters dynamically formed from atomically dispersed copper, Nature Energy, 5, pp. 623–632. DOI: 10.1038/s41560-020-0666-x.] Similarly, a recent study successfully stabilized a pair of Cu₁ centers on a carbon support: at elevated Cu content (~13.3 wt% in the matrix), adjacent monoatomic centers formed active dimers.[ Xia, W., Xie, Y., Jia, S., Han, S., Qi, R., Chen, T., et al. (2023). Adjacent Copper Single Atoms Promote C–C Coupling in Electrochemical CO2 Reduction for the Efficient Conversion of Ethanol, Journal of the American Chemical Society, 145, pp. 17253–17264. DOI: 10.1021/jacs.3c04612.] This resulted in a faradaic efficiency of ~81.9% on ethanol (in 0.5 M CsHCO₃, -1.1 V vs. RHE) with a partial current of ~35.6 mA/cm². Operando XANES/XAFS confirmed the formation of Cu-Cu bonds at operating potentials, indicating the formation of Cu_n clusters as active species. Remarkably, after the potential is removed, the clusters dissociate back into individual Cu atoms, signifying the restoration of the catalyst structure. These studies demonstrate a new approach: homogeneous yet structurally dynamic systems in which distributed centers are brought together under voltage combine the advantages of molecular and metallic catalysis.

Compared to known copper-based electrocatalytic systems, the presented system differs in that selectivity is controlled by varying the precatalyst composition (the Cu:I ratio), rather than external parameters alone. A trisferrocenyltrithiophosphite-copper(I) bromide composites in a 1:1 ratio produces predominantly methanol, whereas with excess Cu (2:1), the product profile shifts sharply toward ethanol. This shift is unprecedented for copper catalysts: typically, altering selectivity requires a different material or surface modifier.

A catalyst based on readily accessible components (CuBr and an organic ligand) provides selectivity for alcohols (especially C₂) at a moderate overpotential, which is rare for pure copper, and maintains structural stability for at least several hours of continuous electrolysis. It is worth noting that a similar effect of the Pd:ligand ratio was observed in the homogeneous catalysis of Suzuki-Miyaura cross-coupling reactions [Scott, N. W. J., Ford, M. J., Jeddi, N., Eyles, A., Simon, L., Whitwood, A. C., Tanner, T., Willans, C. E. and Fairlamb, I. J. S. (2021). A Dichotomy in Cross-Coupling Site Selectivity in a Dihalogenated Heteroarene: Influence of Mononuclear Pd, Pd Clusters, and Pd Nanoparticles—the Case for Exploiting Pd Catalyst Speciation, Journal of the American Chemical Society, 143, pp. 9682–9693. DOI: 10.1021/jacs.1c05294.] on palladium, confirming our understanding of the possible influence of the metal:ligand ratio on reaction progression.

This approach, combining organic and inorganic components, expands the arsenal of tools for fine-tuning CO₂RR electrocatalysts and demonstrates a new concept: controlling the selectivity of ligand coordination in heterogeneous CO₂ catalysis. This distinguishes our system from traditional Cu nanocatalysts and can serve as a starting point for the creation of a whole class of tunable catalysts based on copper-ligand composites.

We have updated the article with this information on page 1 and 2. We have added references [11, 12, 13, 14, 15, 16, 17, 18, 25]

Q2. The voltammetric curves given in Fig.1 look strange. What about their reproducibility? The current density observed for 1 and 2 in Fig.2 contradicts the catalytic current values observed in Fig.1

We thank the reviewer for their careful examination of the data. We would like to clarify the points raised regarding the appearance of the curves in Fig. 1 and the relationship between the data in Fig. 1 and Fig. 2.

Regarding the appearance of the voltammetric curves in Fig. 1, the observed shapes are characteristic of measurements performed using carbon paste electrodes, where the electroactive material is immobilized within the paste matrix. We confirm that the results presented are reproducible across independent experiments.

The apparent discrepancy between the catalytic current values observed in Fig. 1 and the current density in Fig. 2 does not represent a contradiction; rather, it stems from the distinct experimental methodologies and electrode scales employed for these measurements.

Fig. 1 presents Linear Sweep Voltammetry (LSV) data, which are analytical measurements intended to study the electrochemical behavior of the modified electrode. As detailed in Section 3.1, these experiments were conducted using a fixed carbon-paste working electrode with a small surface area (1 mm²). Consequently, the results are reported as absolute current in microamperes (µA).

In contrast, Fig. 2 displays the results of potentiostatic testing conducted during preparative electrolysis. This setup utilized a significantly larger U-shaped glassy carbon paste electrode (48.00 cm², as detailed in Section 3.2). For these experiments, the results are reported as current density, meaning the total current was normalized relative to the large geometric surface area of the electrode.

Therefore, the values in Fig. 1 (absolute current) and Fig. 2 (normalized current density) are not directly comparable due to the vastly different electrode sizes and normalization methods.

Q3. The amounts of all ingredients of the composites should be clearly stated

We thank the reviewer for this important comment. We have now clearly specified the amounts of all ingredients used to prepare the composite electrodes. The base carbon paste consisted of graphite powder and a phosphonium-based ionic liquid in a 90:10 (w/w) ratio, consistent with our earlier RTIL-CPE methodology [Journal of Solid State Electrochemistry 2014, 19 (9), 2883-2890]. In the present work, 30 wt% of this base paste was replaced by the investigated component(s), giving final paste compositions of 63 wt% graphite, 7 wt% ionic liquid, and 30 wt% additive. We also explicitly indicated that the 30 wt% additive contained CuBr and P(SFc)3 in molar ratios of 1:1 (composite 1) and 2:1 (composite 2), and that the control electrodes were prepared in the same way.

We have added this information to the manuscript on page 11.

Q4. The authors wrote about synergistic interaction between the metal center and the phosphorus-sulfur ligand framework. More evidence should be provided. The voltammogram for CuBr-doped composite should be added in Fig.1

Under the conditions used in this study, the voltammetric response of the CuBr-doped composite does not show a distinctive characteristic change compared with the parent carbon paste electrode.

Q5. How relative intensity of the signals corresponding to MeOH and EtOH were measured? The detailed description of the chromatographic analysis should be provided. There is not a word about this in the Experimental part. Current efficiency for the alcohols formation should be determined. It is not possible to declare selectivity based on a short fragment of the chromatogram.

In the revised manuscript, we have added a dedicated subsection describing the chromatographic analysis of liquid products. The GC measurements were performed using an Agilent Technologies 6890N instrument with an FID detector and a 50 m PEG capillary column, and the relative intensities of MeOH and EtOH were evaluated from the integrated peak areas using external calibration with authentic standards

Q7. 4: the size of the particles is not shown. What is illustrated in Fig.4? What does it mean: “structural integrity”?

We thank the reviewer for this comment. We agree that the purpose of Fig. 4 was not sufficiently explained. Figure 4 presents SEM images and corresponding elemental (EDS) mapping of the electrode surfaces for the 2:1 and 1:1 CuBr–P(SFc)₃ composites after CO₂ electrolysis, aimed at demonstrating the surface morphology and the distribution of Cu, P, and S within the carbon paste matrix. Thus, this figure was intended as a qualitative confirmation of the formation of catalytically active domains and the homogeneous presence of key elements, rather than as a particle-size analysis.

Q8. The structural study of the composite is perfectly done. But what conclusions can be made from comparison of the morphology prior and after electrolysis? The changes are quite expected, but what do they prove? The text is full of generalities on this topic (“profound structural transformation of the precatalyst”, “dynamic changes in the surface composition”, “highly dispersed nanoclusters that are too small or disordered to coherently diffract X-rays”, “particularly rich in the crucial Cu+ species” etc. And most important, how does this relates to the mechanism? The interrelation between the results of the surface study and mechanism will be outlined. The detailed analysis based on literature data should be provided.

We agree that the comparison of surface morphology before and after electrolysis should not be overinterpreted as direct mechanistic proof. The purpose of the SEM/EDS section in our manuscript is primarily descriptive and supportive: to show that the composite electrode surface undergoes expected electrochemical conditioning under CO2RR conditions and the key elements associated with the composite (Cu, P, S) remain present on the electrode after operation, which supports the retention of the composite components during electrolysis. We also agree that several phrases in the original version were too general and could imply stronger mechanistic claims than our data allow. Therefore we have revised this part of the text to make it more precise and conservative. To address the Reviewer’s key point regarding the mechanism, we now explicitly state that the mechanistic interpretation in this work is based mainly on the comparative product profiles of the 1:1 and 2:1 composites, while the morphology study serves as supplementary evidence of the stability and persistence of the composite-derived surface features during catalysis. This clarification has been added to the revised Results and Discussion section to avoid overstatement.

Reviewer 2 Report

Comments and Suggestions for Authors

This manuscript reported the synthesis, characterization, and electrochemical evaluation of novel trisferrocenyltrithiophosphite–copper(I) bromide composites for CO₂ reduction. The composites show good catalytic activity, stability depending on the Cu-to-ligand ratio. However, several aspects require clarification and further evidence to strengthen the conclusions:

What are the Faradaic efficiencies for methanol and ethanol production? This is critical for evaluating catalytic selectivity.
How does the phosphonium-based ionic liquid influence the composite structure and catalytic behavior? A control without ionic liquid may be informative.
Was the ECSA of the composite electrodes measured to normalize current densities?
The current density drops by 9% over 5 hours. What is the performance beyond 5 hours? Is the deactivation reversible?
Some recently reported catalysts for CO2RR are suggested, such as DOI: 10.1016/j.cej.2025.164420; DOI: 10.1002/adfm.202420177
Have Tafel plots been constructed to gain insight into the rate-determining steps?

Author Response

This manuscript reported the synthesis, characterization, and electrochemical evaluation of novel trisferrocenyltrithiophosphite-copper(I) bromide composites for CO₂ reduction. The composites show good catalytic activity, stability depending on the Cu-to-ligand ratio. However, several aspects require clarification and further evidence to strengthen the conclusions:

Q1. What are the Faradaic efficiencies for methanol and ethanol production? This is critical for evaluating catalytic selectivity.

We thank the Reviewer for this important comment. The Faradaic efficiencies (FEs) for liquid products were calculated based on the GC-quantified amounts of alcohols and the total charge passed during potentiostatic electrolysis using.

With C, the Faradaic efficiency for methanol was 5.79% (5 µmol MeOH, ), while the Faradaic efficiency for ethanol was 9.26% (4 µmol EtOH, ). These values have been added to the revised manuscript to support the selectivity discussion.

We have added this information to the manuscript on page 7.

Q2. How does the phosphonium-based ionic liquid influence the composite structure and catalytic behavior? A control without ionic liquid may be informative.

The phosphonium-based ionic liquid is not an inert additive but an integral part of the electrode architecture. In our carbon paste configuration it serves as a conductive and chemically stable binder, ensuring homogeneous dispersion of the copper-ligand composite within the graphite matrix and providing reliable electron transport to the embedded catalytic moieties. This approach is supported by prior studies of phosphonium RTIL-based CPEs showing high conductivity, excellent time stability and reproducibility, and a very wide electrochemical window; importantly, direct comparison with paraffin-based CPEs demonstrates the clear advantages of the RTIL binder. The ionic liquid in our system primarily ensures a robust, conductive, and reproducible environment for probing the intrinsic behavior of the copper-ligand composite. A “control without ionic liquid” would necessarily require substitution with a different binder (e.g., paraffin oil), thereby altering conductivity and interfacial charge transfer. This would introduce a confounding variable and would not constitute an informative one-parameter control relative to our targeted question of stoichiometry-controlled selectivity. Therefore, we instead kept the ionic liquid content constant across all samples and varied only the CuBr-to-ligand ratio. Under these strictly identical electrode conditions, the observed shift in product distribution can be attributed to the composite composition rather than to the binder.

We have added this information to the manuscript on page 12.

Q3. Was the ECSA of the composite electrodes measured to normalize current densities?

Thank you for this important comment. In this work, the ECSA of the composite electrodes was not measured. Therefore, all current densities are reported with respect to the geometric electrode area. The working electrodes for LSV were prepared using identical procedures and the same geometric area, and the preparative electrolysis was performed on a U-shaped GC-CPE with a fixed area of 48.00 cm².

We note that quantitative ECSA determination for ionic-liquid-based carbon paste electrodes can be nontrivial, because the binder itself contributes to interfacial capacitance and can complicate a straightforward capacitance-to-ECSA conversion. The use of a phosphonium-based ionic liquid was chosen to ensure high conductivity, stability, and a wide potential window, as previously validated for RTIL-CPE systems. Importantly, the key comparisons in our study are made between composites prepared from the same carbon paste matrix and under identical experimental conditions, with only the CuBr:P(SFc)₃ ratio varied. Thus, even without ECSA normalization, the observed differences in activity and selectivity remain meaningful within the scope of this work.

Q4. The current density drops by 9% over 5 hours. What is the performance beyond 5 hours? Is the deactivation reversible?

In the present work, we did not extend the electrolysis beyond 5 h, since the main goal was to demonstrate stoichiometry-controlled selectivity toward C1 vs. C2 alcohols under identical conditions and to correlate this behavior with post-electrolysis structural/surface characterization obtained after a standardized 5 h test. Therefore, longer-duration stability and a dedicated reversibility study are beyond the scope of this manuscript.

Q5. Some recently reported catalysts for CO2RR are suggested, such as DOI: 10.1016/j.cej.2025.164420; DOI: 10.1002/adfm.202420177

We thank the reviewer for pointing out these recent and highly relevant contributions. In the revised manuscript, we have now cited both suggested works in the Introduction when discussing state-of-the-art CO₂ reduction electrocatalysts.

Specifically, Zhu et al. (Chem. Eng. J., 2025, DOI: 10.1016/j.cej.2025.164420) reported a Bi₂/BiVO₄₋ₓ heterostructure obtained by confining bismuth in oxygen-defective BiVO₄, achieving a high Faradaic efficiency for formate generation (up to ~97% at −0.8 V vs RHE) and large partial current density.

In addition, Zhu et al. (Adv. Funct. Mater., 2025, DOI: 10.1002/adfm.202420177) developed a VOₓ-mediated Bi–Sn alloy catalyst that delivers Faradaic efficiencies for formate above 90% over a wide potential window and excellent long-term stability.

These studies provide up-to-date benchmarks for highly efficient CO₂-to-formate electrocatalysts and help to contextualize the performance and design concept of our copper-based systems. We have briefly discussed their key results and highlighted the differences in catalyst composition, structure, and product distribution in the revised Introduction/Discussion.

We have added references 15 and 16

Q6. Have Tafel plots been constructed to gain insight into the rate-determining steps?

We thank the reviewer for this insightful suggestion. In the present work, we did not perform a systematic Tafel analysis, and therefore no Tafel plots are included in the manuscript. Our electrochemical data were acquired mainly as linear sweep voltammograms (Fig. 1) and potentiostatic electrolysis at fixed potentials (Fig. 2), which are not, by themselves, sufficient to construct reliable Tafel plots over an extended potential range. Moreover, the measurements were carried out using carbon paste electrodes under continuous CO₂ bubbling and in a cell geometry optimized for preparative electrolysis rather than detailed kinetic studies. Under these conditions, mass-transport limitations and uncompensated resistance effects can significantly affect the current-potential response, making it difficult to extract meaningful kinetic parameters or unambiguously assign a rate-determining step from a Tafel slope. For this reason, we chose not to include a formal Tafel analysis in order to avoid over-interpreting the available data.

Reviewer 3 Report

Comments and Suggestions for Authors

Review Report:

Manuscript Number: ijms-3982246

Trisferrocenyltrithiophosphite-Copper(I) Bromide Composites for Electrochemical CO₂ Reduction

Recommendation: Accepted after minor revision

Comments:

This manuscript investigates an interesting study on novel Cu(I)–trisferrocenyltrithiophosphite composites as tunable electrocatalysts for CO₂ reduction, highlighting methanol and ethanol selectivity depending on the Cu:ligand ratio. The concept of using redox-active ferrocenyl-thiophosphite ligands for tuning CO₂RR selectivity is interesting and relevant. The work combines electrochemical measurements, PXRD, SEM, and XPS characterization.

Given the comprehensive and detailed analysis, I believe this will be of interest to the readers of the International Journal of Molecular Sciences. However, I recommend acceptance of the review after minor revisions:

Clarify whether gaseous products (e.g., CO, CH₄, C₂H₄, H₂) were analyzed during CO₂ If not, please discuss whether these products were below detection limits or simply not measured, as including or excluding gas-phase products is essential for establishing a complete product distribution.
Please clarify whether other possible liquid-phase CO₂ reduction products (e.g., formate, acetate, or higher alcohols) were analyzed or ruled out.
PXRD patterns should be indexed to confirm phase purity and illustrate crystalline. Additionally, SEM images should include scale bars and magnification details, and elemental mapping should be supported with quantitative EDS spectra to verify the Cu:P:S stoichiometry and elemental uniformity of the composites.
Please provide a summary table comparing key literature studies (catalyst type, conditions, major products, selectivity, stability) with the present work to clearly highlight the novelty and performance advantage of this study.

Minor Comments

Units in electrochemical figures (Figure 1, 2) should be standardized (e.g., mA cm^-2).
Typographical errors:

(I) Remove electrochemistry as the last word of abstract.

(II) Introduction, para 5, correct the line “For example, electron-rich phosphines……, where electron donation.…… of the metal” with “For example, electron-rich phosphines……, whereas electron donation.…… of the metal”.

(III) Introduction, paragraph 5, Correct the notation “μ:η1,η1” to the proper IUPAC representation “μ–η1:η1”.

Figures 5, Y-axis labeling: Please revise the Y-axis label “Lin (counts)” to a scientifically standard form such as “Intensity (a.u.)”

After addressing the points above, the manuscript will be in excellent shape for publication.

Comments on the Quality of English Language

The manuscript is clearly written and scientifically sound, but minor grammatical and stylistic corrections are needed. A brief professional English proofreading is recommended to improve sentence flow and consistency in technical terms and figure captions.

Author Response

Manuscript Number: ijms-3982246

Trisferrocenyltrithiophosphite-Copper(I) Bromide Composites for Electrochemical CO₂ Reduction

Recommendation: Accepted after minor revision

This manuscript investigates an interesting study on novel Cu(I)-trisferrocenyltrithiophosphite composites as tunable electrocatalysts for CO₂ reduction, highlighting methanol and ethanol selectivity depending on the Cu:ligand ratio. The concept of using redox-active ferrocenyl-thiophosphite ligands for tuning CO₂RR selectivity is interesting and relevant. The work combines electrochemical measurements, PXRD, SEM, and XPS characterization.

Given the comprehensive and detailed analysis, I believe this will be of interest to the readers of the International Journal of Molecular Sciences. However, I recommend acceptance of the review after minor revisions.

We sincerely thank the Reviewer for their positive assessment of our work. We are pleased that the Reviewer considers the manuscript of interest to the readers of the International Journal of Molecular Sciences, and we will carefully address all minor comments to further improve the quality and clarity of the paper.

Q1. Clarify whether gaseous products (e.g., CO, CH₄, C₂H₄, H₂) were analyzed during CO₂ If not, please discuss whether these products were below detection limits or simply not measured, as including or excluding gas-phase products is essential for establishing a complete product distribution.

We thank the reviewer for this important question regarding the analysis of gas-phase products. Gaseous products formed during CO₂ reduction were analyzed by online gas chromatography using a certified reference gas mixture. The retention times and peak areas of the products formed in the CO₂ stream were compared with those of this standard mixture, which contained the expected CO₂RR products and was used for calibration. This is a standard procedure for quantitative analysis of gaseous electrolysis products.

The products shown in the gas chromatograms in the manuscript represent all species that were detected under our experimental conditions. Apart from these peaks, no additional signals attributable to other gaseous CO₂ reduction products (such as CO, CH₄, C₂H₄ or related light gases) were observed above the noise level of the instrument. Thus, if any other gaseous products were formed, their concentrations were below the detection limit of our GC setup.

We will clarify this point in the revised manuscript by explicitly stating that gas-phase products were monitored by GC using a reference gas mixture and only the products reported in the chromatograms were detected within the detection limits of the method.

Q2. Please clarify whether other possible liquid-phase CO₂ reduction products (e.g., formate, acetate, or higher alcohols) were analyzed or ruled out.

We thank the Reviewer for this comment. Under the analytical protocol used in this study, we screened for a broad range of possible CO₂ reduction products by gas chromatography. The GC system was calibrated/verified using a standard gas mixture, and the chromatographic analysis was performed to check for all expected gaseous and volatile liquid products. No additional peaks attributable to other possible products (including higher alcohols or other commonly reported CO₂RR products within the detection capability of our GC method) were observed. Therefore, besides methanol and ethanol, no other products were detected under our experimental conditions.

Q3. PXRD patterns should be indexed to confirm phase purity and illustrate crystalline. Additionally, SEM images should include scale bars and magnification details, and elemental mapping should be supported with quantitative EDS spectra to verify the Cu:P:S stoichiometry and elemental uniformity of the composites.

We thank the reviewer for these constructive suggestions. Concerning the PXRD data, we would like to note that the investigated materials are composite systems based on Cu(I) bromide dispersed in a carbon paste matrix and are not single-phase crystalline compounds. In this case, the diffraction patterns are used in a qualitative way, namely to confirm the presence of crystalline copper-containing phases and to monitor their evolution upon electrolysis. Because of the composite nature, the broad reflections, and the strong background from the carbon paste, an unambiguous indexing and a meaningful refinement of the PXRD patterns are not feasible and could give a misleading impression of phase purity. In the revised manuscript, we will clarify that the PXRD patterns are intended to provide qualitative structural information rather than a full crystallographic characterization.

Regarding the SEM data, the images were recorded with appropriate scale and magnification, and we will make sure that the corresponding scale bars and magnification values are clearly indicated in all SEM panels and figure captions in the revised version. The elemental mapping already demonstrates a homogeneous spatial distribution of Cu, P, and S within the composites. Quantitative EDS analysis on such rough composite carbon-paste surfaces is subject to significant systematic uncertainties; therefore, we have used EDS in a semi-quantitative manner. The obtained elemental ratios are consistent with the nominal Cu:P:S composition of the composites, and we will briefly mention this point in the text without overloading the figures with additional spectra.

Q4. Please provide a summary table comparing key literature studies (catalyst type, conditions, major products, selectivity, stability) with the present work to clearly highlight the novelty and performance advantage of this study.

We thank the Reviewer for this constructive suggestion. We have added a new summary table (Table 1) comparing representative Cu-based CO2RR systems for alcohol formation with our CuBr-P(SFc)3 composites. While several state-of-the-art Cu nanostructures and high-density/adjacent Cu single-atom systems can achieve very high ethanol or methanol Faradaic efficiencies under optimized conditions (e.g., up to ~91% FE for ethanol for dynamically formed Cu clusters and ~81.9% FE for ethanol on adjacent Cu–N₃ motifs with >25 h stability), our work offers a distinct conceptual advantage: selectivity is tuned by changing the CuBr:P(SFc)₃ ratio within the same ligand-salt composite platform. Under identical electrode architecture and electrolyte conditions, the 1:1 composite preferentially forms methanol, whereas increasing the copper content to 2:1 shifts selectivity toward ethanol. This stoichiometry-controlled handle provides a simple and generalizable route to modulate the C1 vs C2 alcohol pathway in a hybrid Cu(I)-ligand system.

Table added on pages 9-10.

Q5 Minor Comments

Units in electrochemical figures (Figure 1, 2) should be standardized (e.g., mA cm^-2).

Typographical errors:

(I) Remove electrochemistry as the last word of abstract.

(III) Introduction, paragraph 5, Correct the notation “μ:η1,η1” to the proper IUPAC representation “μ-η1:η1”.

Figures 5, Y-axis labeling: Please revise the Y-axis label “Lin (counts)” to a scientifically standard form such as “Intensity (a.u.)”

After addressing the points above, the manuscript will be in excellent shape for publication.

We thank the reviewer for these careful and constructive minor comments.

We have corrected the Figure 5 in accordance with the comments made.

Reviewer 4 Report

Comments and Suggestions for Authors

The study entitled "Trisferrocenyltrithiophosphite-Copper(I) Bromide Composites for Electrochemical CO2 Reduction" is novel and meaningful. The authors have systematically investigated the structure and properties of the materials and discussed the experimental results. However, there are still several points that could be further improved. Therefore, I recommend minor revision before publication. My specific comments are as follows:

1. The resolution of the data figures in the manuscript is insufficient and should be improved. In addition, it would be more aesthetically pleasing and clearer if the y-axes (including ticks and labels) were fully shown, which would help readers quickly obtain the necessary information.

2. In Figure 1, it would be clearer to write the full sample information directly in the figure rather than using only numbers 1, 2 and "+CO2".

3. For composites 1 and 2 with CuBr to P(SFc)3 molar ratios of 1:1 and 2:1, the authors provide FID chromatograms of the product distributions, which is very good. However, it would be more informative for readers if the quantitative results of the products could be indicated directly in the figure.

4. For Figure 5, I suggest revising the x- and y-axes by referring to Figure 6, so that the axes, scales and font sizes are consistent. The PXRD patterns show many peaks; it is recommended to label the key diffraction planes and indicate the corresponding phase information.

Author Response

Q1. The resolution of the data figures in the manuscript is insufficient and should be improved. In addition, it would be more aesthetically pleasing and clearer if the y-axes (including ticks and labels) were fully shown, which would help readers quickly obtain the necessary information.

We thank the Reviewer for this helpful suggestion. The figures in our manuscript were generated in Origin and inserted as high-resolution raster images. Therefore, the final production-quality files will have a resolution of no less than 1200 dpi, ensuring sufficient clarity for publication. In addition, we have checked the layout to ensure that the y-axes (including ticks and labels) are fully shown and not cropped, so that the data can be read more easily and consistently across all figures.

Q2. In Figure 1, it would be clearer to write the full sample information directly in the figure rather than using only numbers 1, 2 and "+CO2".

We thank the Reviewer for this helpful suggestion. We agree that the labeling in Figure 1 can be made clearer.

Q3. For composites 1 and 2 with CuBr to P(SFc)3 molar ratios of 1:1 and 2:1, the authors provide FID chromatograms of the product distributions, which is very good. However, it would be more informative for readers if the quantitative results of the products could be indicated directly in the figure.

We thank the Reviewer for this valuable suggestion.

Q4. For Figure 5, I suggest revising the x- and y-axes by referring to Figure 6, so that the axes, scales and font sizes are consistent. The PXRD patterns show many peaks; it is recommended to label the key diffraction planes and indicate the corresponding phase information.

We thank the reviewer for these suggestions. We have corrected the Figure 5 in accordance with the comments made. Concerning indexing of the PXRD data, we would like to note that the investigated materials are composite systems based on Cu(I) bromide dispersed in a carbon paste matrix and are not single-phase crystalline compounds. In this case, the diffraction patterns are used in a qualitative way, namely to confirm the presence of crystalline copper-containing phases and to monitor their evolution upon electrolysis. Because of the composite nature, the broad reflections, and the strong background from the carbon paste, an unambiguous indexing and a meaningful refinement of the PXRD patterns are not feasible and could give a misleading impression of phase purity. In the revised manuscript, we will clarify that the PXRD patterns are intended to provide qualitative structural information rather than a full crystallographic characterization.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors have made efforts to improve the manuscript. Analysis of the relevant literature and the key experimental details were added. The current efficiencies for methanol and ethanol formation were measured; they are low, ca. 5% and 9%, respectively. That means that the alcohols were the minor products whereas the other products of CO2 conversion were not detected. Therefore, it can hardly be said that the proposed composites are efficient and selective catalysts for CO2 electroreduction to alcohols, as it is written in the Abstract and Conclusion. Consequently, Abstract and Conclusion should be corrected; the CE data should be clearly indicated both in Abstract and in Conclusion, not to mislead the readers.

Additional comments:

Fig.3 which was added in the revised version is unclear. What does it mean “relative intensity”? It looks as if the alcohols are formed in the first minute of the electrolysis; afterwards their amount stays constant during the next 5 h.
How CEs for the alcohols formation were calculated? What chemical process (which consumed 50C of electricity) was considered ? How many electrons? The details of the calculation should be provided in the experimental part
The Authors wrote in reply to my previous comment Q3 which was related to Fig.1 : “Under the conditions used in this study, the voltammetric response of the CuBr-doped composite does not show a distinctive characteristic change compared with the parent carbon paste electrode”. It seems strange since copper centers participate in the catalytic process. This issue should be studied in more details.

Author Response

Comment: "The authors have made efforts to improve the manuscript. Analysis of the relevant literature and the key experimental details were added. The current efficiencies for methanol and ethanol formation were measured; they are low, ca. 5% and 9%, respectively. That means that the alcohols were the minor products whereas the other products of CO2 conversion were not detected. Therefore, it can hardly be said that the proposed composites are efficient and selective catalysts for CO2 electroreduction to alcohols, as it is written in the Abstract and Conclusion. Consequently, Abstract and Conclusion should be corrected; the CE data should be clearly indicated both in Abstract and in Conclusion, not to mislead the readers."

Reply: "We sincerely thank the Reviewer for this insightful and fair assessment. We agree that the Faradaic efficiency (FE) values of 5.79% for methanol and 9.26% for ethanol are modest when compared to state-of-the-art nanostructured copper systems. However, the primary focus of this study was not to achieve record-breaking efficiency, but to demonstrate a conceptually distinct approach to controlling reaction pathways. We show that by simply adjusting the copper-to-ligand stoichiometry in a hybrid system, we can shift the product distribution from C1 to C2 alcohols under identical conditions. We acknowledge that using terms like "efficient" or "high selectivity" without context may be misleading. In accordance with your suggestion, we have revised the Abstract and Conclusion to:Explicitly state the FE values for both methanol and ethanol. Replace "highly selective" with "tunable," emphasizing the control over product distribution rather than absolute efficiency."

Q1. Fig.3 which was added in the revised version is unclear. What does it mean “relative intensity”? It looks as if the alcohols are formed in the first minute of the electrolysis; afterwards their amount stays constant during the next 5 h.

Reply: We thank the Reviewer for the opportunity to clarify the presentation of the chromatographic data in Figure 3. The y-axis label 'Relative Intensity' refers to the GC-FID detector response presented in arbitrary units. Each chromatogram was individually normalized and vertically offset to clearly illustrate the presence of methanol and ethanol while allowing for an easy comparison of peak positions across different time intervals. This approach was chosen to confirm that the identity of the major product remains consistent throughout the electrolysis period for each composite. Because each trace has been scaled, the peak heights in Figure 3 are not intended for quantitative comparison of product amounts between different electrolysis times. This explains why the peaks appear to have a constant magnitude from the first minute to the end of the experiment. This method of displaying time-dependent chromatograms using normalization and offsets for qualitative comparison is a standard practice in the field of electrochemical CO2 reduction (The example: 10.1016/j.esci.2023.100143). For a quantitative assessment of the catalytic performance, we rely on the Faradaic efficiency calculations based on the total accumulated products after the 300-min period as described in the Experimental section. We have updated the caption of Figure 3 and the accompanying text in Section 2 to explicitly state that these traces are normalized and offset for clarity.

Q2. How CEs for the alcohols formation were calculated? What chemical process (which consumed 50C of electricity) was considered ? How many electrons? The details of the calculation should be provided in the experimental part

Reply: We thank the Reviewer for this important request for clarification. In the context of electrochemical CO2 reduction, the current efficiency (CE) reported in this manuscript corresponds directly to the Faradaic efficiency (FE) for a specific product. This value represents the fraction of the total charge passed through the cathode that is consumed during the formation of that product. We have added a dedicated subsection titled 3.3. Faradaic efficiency (current efficiency) calculations to the Experimental part. This section details the mathematical derivation and the stoichiometric assumptions used, specifically accounting for the multi-electron processes required for alcohol formation.

Q3. The Authors wrote in reply to my previous comment Q3 which was related to Fig.1 : “Under the conditions used in this study, the voltammetric response of the CuBr-doped composite does not show a distinctive characteristic change compared with the parent carbon paste electrode”. It seems strange since copper centers participate in the catalytic process. This issue should be studied in more details.

Reply: The catalytic activity is only "triggered" when the ligand is present to coordinate with the copper ions and form a more dynamic, amorphous active phase. This interpretation is supported by our post-electrolysis PXRD data, which show that the crystalline peaks of CuBr disappear only in the composite systems where the ligand can facilitate structural redistribution into a catalytically active state. Therefore, the lack of a distinctive signal in the CuBr-only control reinforces our conclusion that the observed catalytic performance is a result of a synergistic effect between the metal and the phosphorus-sulfur ligand framework rather than the properties of the individual components.

Reviewer 2 Report

Comments and Suggestions for Authors

This work is suitable for publication.

Author Response

We sincerely thank the Reviewer for the supportive assessment and for recommending this work for publication.