Crystal-Plane-Dependent Guaiacol Hydrodeoxygenation Performance of Au on Anatase TiO₂

Round 1

Reviewer 1 Report

Authors presented a systematic study on the crystalline plane effect on Au/TiO2 for guaiacol hydrodeoxygenation. The procedure of making the catalysts is not novel. However, the mechanism revealed is interesting and could be beneficial to the biomass upgrading. There are Au, Auδ, Ti4+, Ti(4+δ)+, O2-, O(2-δ)-, oxygen vacancies mentioned in the manuscript, but the effect of these sites to the mechanism is not described clearly enough. For example, which sites are bystanders. I suggest reconsider after major revision.

1)     Line 42 the paragraph should be more specific to the hydrodeoxygenation reaction, rather than the very general SMSI that could be dated back 1960s. The introduction part missing proper references for the effect of SMSI on TiO2 for hydrodeoxygenation of biomass or TiO2 of other groups. Please refer to:

·        R. Huang et.al, The effects of SMSI on m-Cresol hydrodeoxygenation over Pt/Nb2O5 and Pt/TiO2, J. Catal., Volume 398, June 2021, Pages 102-108, https://doi.org/10.1016/j.jcat.2021.04.012

·        X. Zhao et.al , Effect of Strong Metal-Support Interaction of Pt/TiO2 on Hydrodeoxygenation of m-Cresol, ChemistrySelect. 3 (37) (2018) 10364–10370, https://doi.org/10.1002/slct.201801147

2)     Line 71, it is improper to use SMSI here for Au/TiO2 system. If the metal particle is not partially covered or fully covered by the support material, then it would not be considered a case of strong metal-support interaction (SMSI). Suggest changing to EMSI.

3)     When preparing the Au/TiO2 catalysts, the catalysts were not treated in high temperatures. However, in the materials characterization or the reactivity studies, the catalysts were treated on/above 300 °C. What is the surface loss, could you provide evidence showing that the crystalline phase/morphology is stable after the reaction or characterization?

4)     Line 105, suggest the author points out the expected characteristic peak in the caption of the Figure S1. Although it is a flat line, it would be informative if the reader knows where Cl and F peaks are.

5)     Figure 4, is there any peak attributed to the H2 desorption from the chemisorbed H on Au?

6)     The experiment procedure for Figure 5 needs to be explained better in the methods section. It is not clear to me what the procedure is to obtain the 300C curve. Moreover, what is the temperature to obtain the Figure 5C?

7)     Figure 6, for the XPS, it would be more reader friendly if the author could use the Ti(4+δ)+, Ti4+ as the label rather than the energy level (458.9, 458.5). Same for Au and O.

8)     Figure 6, Rather than comparing Au/TiO2-101 with the Au/TiO2-001, could you compare Au/TiO2-101 wither the TiO2-101? And Au/TiO2-001 wither the TiO2-001 in the supplementary, to show the effect of the Au?

9)     Figure 6, S1, S5 Reverse x axis of Fig 4. XPS figures should always have lower binding energy on the right. Because binding energy comes from subtracting the photoelectron’s energy from the X-ray photon’s energy.

10)  Supplementary. “Based on the statistical analysis of TEM and HRTEM images of TiO2-101 and TiO2-001”, please include the standard deviation from the statistical analysis for the geometric dimensions.

11)  Line 336, it would be useful if the authors could point out and explain the implications of TiO2 is partially oxidized, which is contradictory to the Ni/TiO2 or Pt/TiO2 system.

12)  What is the catalytic performance if only Au is present? for example, you can put Au on carbon support effect, then the carbon support will be inert.

13)  Line 298, the Table 2 footnote says “ 700 RPM”, however, in the methodology section, it says “500 RPM”. Which one is correct?

14)  While H2 reduced TiO2 forming Oxygen Vacancies, why is Ti3+ not showed in the XPS?

15)  What is tricresol in Figure 7? This word is rarely used, and the chemical structure is poorly defined. Could you change to some common names?

Author Response

Reviewer: 1

 

Comments:

 

Authors presented a systematic study on the crystalline plane effect on Au/TiO2 for guaiacol hydrodeoxygenation. The procedure of making the catalysts is not novel. However, the mechanism revealed is interesting and could be beneficial to the biomass upgrading. There are Au, Auδ, Ti4+, Ti(4+δ)+, O2-, O(2-δ)-, oxygen vacancies mentioned in the manuscript, but the effect of these sites to the mechanism is not described clearly enough. For example, which sites are bystanders. I suggest reconsider after major revision.

 

1)     Line 42 the paragraph should be more specific to the hydrodeoxygenation reaction, rather than the very general SMSI that could be dated back 1960s. The introduction part missing proper references for the effect of SMSI on TiO2 for hydrodeoxygenation of biomass or TiO2 of other groups. Please refer to:

• R. Huang et.al, The effects of SMSI on m-Cresol hydrodeoxygenation over Pt/Nb2O5 and Pt/TiO2, J. Catal., Volume 398, June 2021, Pages 102-108, https://doi.org/10.1016/j.jcat.2021.04.012
• X. Zhao et.al , Effect of Strong Metal-Support Interaction of Pt/TiO2 on Hydrodeoxygenation of m-Cresol, ChemistrySelect. 3 (37) (2018) 10364–10370, https://doi.org/10.1002/slct.201801147

 

Response:

Thanks for your comment. The sentence "SMSI between metal and TiO₂ in the hydrodeoxygenation reaction is widely studied (Journal of Catalysis 398 (2021) 102-108. ChemistrySelect2018,3, 10364-10370), but EMSI between them in the hydrodeoxygenation is less reported." are added in line 51, paragraph 2. The two references are included in the Reference section.

 

2)     Line 71, it is improper to use SMSI here for Au/TiO2 system. If the metal particle is not partially covered or fully covered by the support material, then it would not be considered a case of strong metal-support interaction (SMSI). Suggest changing to EMSI.

 

Response:

 

Thanks for pointing out this error. I had changed the SMSI to EMSI in line 71. Intriguingly, little EMSI effect was observed on TiO₂-R (rutile) supported metal catalysts, which consequently exhibited poor catalytic activity in hydrodeoxygenation of guaiacol.

 

3)     When preparing the Au/TiO2 catalysts, the catalysts were not treated in high temperatures. However, in the materials characterization or the reactivity studies, the catalysts were treated on/above 300 °C. What is the surface loss, could you provide evidence showing that the crystalline phase/morphology is stable after the reaction or characterization?

 

Response:

 

Thanks for your comment. In order to clearly illustrate the morphology of the Au/TiO₂ catalyst before and after pre-treatment, we added the TEM images of Au/TiO₂ catalyst before and after pre-treatment (Figure S6) in section 2.3. The catalyst was pre-reduced at 300 °C before the catalytic reaction took place, however, the morphology of the catalyst did not change before and after the pre-treatment.

Figure S6 TEM images of Au/TiO₂ catalyst before and after pre-treatment. Au/TiO₂-101-before (A), Au/TiO₂-101-after (B), Au/TiO₂-001-before (C) and Au/TiO₂-001-after (D)

 

4)     Line 105, suggest the author points out the expected characteristic peak in the caption of the Figure S1. Although it is a flat line, it would be informative if the reader knows where Cl and F peaks are.

 

Response:

 

Thanks for your comment. To clearly show the location of the anion (Cl or F) peaks, we had marked the peak positions of Cl or F in Figure S1.

Figure S1 XPS spectrum of Cl 2p and F 1s in the Au/TiO₂-101(A) and Au/TiO₂-001(B)

 

5)     Figure 4, is there any peak attributed to the H2 desorption from the chemisorbed H on Au?

Response:

 

Thanks for your comment. We have added an explanation for the attribution peak without hydrogen adsorption on Au in the line 158. Lv et.al. (Surface Science 718 (2022) 122015) reported H₂ molecules desorption temperatures on Au surfaces are 276 K in the H₂-TPD. However, there is no any peak at this temperature in the Figure 4. It indicated that there was no H desorption on Au nanoparticles in this case.

 

6)     The experiment procedure for Figure 5 needs to be explained better in the methods section. It is not clear to me what the procedure is to obtain the 300C curve. Moreover, what is the temperature to obtain the Figure 5C?

 

Response:

 

Thanks for your comment. We elaborated on the operation process for Figure 5. In-situ DRIFT spectra of CO adsorption on the samples were collected using a Thermo Scientific Nicolet iS50 equipped with an MCT detector and the spectra were obtained as the average of 64 scans at a resolution of 4 cm^-1. Prior to the testing, the sample was pressed into the in-situ reaction cell, For the fresh sample, the high purity He was introduced for 3 min to collect background spectrum. Then, the mixture gas of 5 % CO/He passed into reaction cell at 30 °C for 3 min, and then the DRIFT spectrum was recorded after the chemisorption of CO. For the pre-treatment, the sample was heated in 10 % H₂/Ar at 10 °C/min to 300 °C for 1 h. After the treatment, the sample was purged in He and cooled to 30 °C to collect background spectrum. Subsequently, the sample was then exposed to 5 % CO/He at 30 °C for 3 min and the DRIFT spectrum was recorded after the chemisorption of CO.

 

7)     Figure 6, for the XPS, it would be more reader friendly if the author could use the Ti(4+δ)+, Ti4+ as the label rather than the energy level (458.9, 458.5). Same for Au and O.

 

9)     Figure 6, S1, S5 Reverse x axis of Fig 4. XPS figures should always have lower binding energy on the right. Because binding energy comes from subtracting the photoelectron’s energy from the X-ray photon’s energy.

 

Response:

 

Thanks for the reviewer’s suggestion. Here we address comment 7) and 9) together. We had revised the Figure 6, Figure S1 and Figure S5.

Figure 6 In-situ XPS of Au 4f (A), Ti 2p (B) and O 1s (C)

Figure S1 XPS spectrum of Cl 2p and F 1s in the Au/TiO₂-101(A) and Au/TiO₂-001(B)

Figure S5 XPS spectrum of Au 4f (A), Ti 2p (B) and O 1s (C) in Au/TiO₂ catalysts before in-situ reduction

 

8)     Figure 6, Rather than comparing Au/TiO2-101 with the Au/TiO2-001, could you compare Au/TiO2-101 wither the TiO2-101? And Au/TiO2-001 wither the TiO2-001 in the supplementary, to show the effect of the Au?

 

Response:

 

Thanks for your comment. Thanks for your comment. Effect of Au on TiO2 has been reported (ACS Catal. 2017, 7, 695-705. Journal of Catalysis 344 (2016) 136–140).   Pure TiO₂ exhibits very low activity for guaiacol hydrodeoxygenation. However, It presents excellent performance by supporting Au. This indicates that Au gives catalytic performance to Au/TiO2 catalysts. In this regard, effect of different crystal surfaces of TiO2, rather than Au, are well-studied in this work to further enhance performance.

 

10)  Supplementary. “Based on the statistical analysis of TEM and HRTEM images of TiO2-101 and TiO2-001”, please include the standard deviation from the statistical analysis for the geometric dimensions. hydrodeoxygenation

 

Response:

 

Thanks for your comment. We added the standard deviation in the Figure S3 and Figure S4

(1). TiO₂-101 nanocrystals

a=13.1 ± 0.18 nm

 

b=4.1 ± 0.16 nm

 

h=16.4 ± 0.35 nm

 

θ=68.3°

 

 

 

Figure S3 Geometric model of TiO₂-101 nanocrystals.

(2). TiO₂-001 nanocrystals

 

 

a=50.6 ± 0.34nm

 

b=47.5 ± 0.29nm

 

h=2.0 ± 0.22nm

 

θ=68.3°  

                                                                              

Figure S4 Geometric model of TiO₂-001 nanocrystals.

11)  Line 336, it would be useful if the authors could point out and explain the implications of TiO2 is partially oxidized, which is contradictory to the Ni/TiO2 or Pt/TiO2 system.

 

Response:

 

Thanks for your comment. The TiO₂ catalysts are not partially oxidized as the whole process during the reaction is conducted in hydrogen, and this is proved by XPS results. Comparing with the unreduced XPS spectrum, the peak position of Ti⁴⁺/O^2- in the in situ XPS spectrum is not shifted, and the new Ti^(4+δ)+/O^(2-δ)- peak is formed due to the interaction between Au and TiO₂, resulting in the transfer of electrons from Ti⁴⁺/O^2- to Au.

 

12)  What is the catalytic performance if only Au is present? for example, you can put Au on carbon support effect, then the carbon support will be inert.

 

Response:

 

Thanks for your comment. We added the explanation that Au/AC catalyst shown the very weak catalytic activity in the line 276. In our previous work (ACS Catal. 2017, 7, 695–705.), we reported that hydrodeoxygenation of guaiacol over Au/AC exhibited a marginal activity.

 

13)  Line 298, the Table 2 footnote says “ 700 RPM”, however, in the methodology section, it says “500 RPM”. Which one is correct?

 

Response:

 

Thanks for your comment. The suspension solution was stirred at a rate of 700 rpm. We had changed the 500 rpm to 700rpm in the line 298.

 

14)  While H2 reduced TiO2 forming Oxygen Vacancies, why is Ti3+ not showed in the XPS?

 

Response:

 

Thanks for your comment. We added the indication that the Ti³⁺ not showed in the XPS in the line 247. The absence of a Ti³⁺ signal in the XPS may be due to the oxygen vacancies is filled Au^δ- species or hydride. Taking all the XPS results together, the Au^δ- species and the Ti^(4+δ)+/O^(2-δ)- species were found in the pre-reduced Au/TiO₂-101 catalyst, however, only Au⁰ species and Ti⁴⁺/O^2- species discovered in the pre-reduced Au/TiO₂-001 catalyst. It may be indicated that electrons bound to oxygen and titanium ions shift toward Au atoms occupying the oxygen vacancies. Our group's previous studies (Catal. Sci. Technol. 2021, 11, 297–311.) have demonstrated that adsorbed H on the surface of TiO₂-A formed TiO-H species and Ti-H species for the pre-reduced Ni/TiO₂-A.

 

15)  What is tricresol in Figure 7? This word is rarely used, and the chemical structure is poorly defined. Could you change to some common names?

 

Response:

 

Thanks for your comment. Tricresol is a collective term, which includes 2,4,6-trimethylphenol, 2,3,6-trimethylphenol, 2,4,5-trimethylphenol etc.

 

Author Response File: Author Response.pdf

Reviewer 2 Report

Zhao et al describes the synthesis of ~1 wt.% Au on anatase TiO2 catalyst, with the TiO2 exposing mainly 101 or 001 surfaces respectively. The catalysts were extensively characterized and evaluated for the hydrodeoxygenation of guaiacol, showing that the 101 TiO2 is most active, due to stronger metal-support interactions and formation of oxygen vacancies in the TiO2 101 surface.

 

The paper is well written and easy to follow. It is an interesting study, which is well performed, and the experimental methods are mostly well described making it possible to reproduce the results. I recommend publication after minor revisions.

 

Specific comments:

 

L. 59: TiO2-A abbreviation is not explained.

 

L. 99-100: The sentence is a little confusing as it starts with stating “XRD patterns of TiO2 101”, but then the comparison is based on TiO2 001. Maybe the first half of line 99 is not necessary.

 

L. 115-116 the order of antisymmetric and symmetric bending vibration is reversed in the list.

 

Table 2 the first column has bold text and a horizontal line not needed between TiO2-101 and TiO2-001.

 

L. 370: Was the precipitate washed after or before centrifugation? Before does not really make sense?

 

L 419: Was the degassing in vacuum or flowing inert gas?

Author Response

Reviewer: 2

 

Comments:

 

Zhao et al describes the synthesis of ~1 wt.% Au on anatase TiO2 catalyst, with the TiO2 exposing mainly 101 or 001 surfaces respectively. The catalysts were extensively characterized and evaluated for the hydrodeoxygenation of guaiacol, showing that the 101 TiO2 is most active, due to stronger metal-support interactions and formation of oxygen vacancies in the TiO2 101 surface.

 

The paper is well written and easy to follow. It is an interesting study, which is well performed, and the experimental methods are mostly well described making it possible to reproduce the results. I recommend publication after minor revisions.

 

Specific comments:

 

1. 59: TiO2-A abbreviation is not explained.

 

 Response:

 

Thanks for your comment. The A in TiO₂-A represents anatase. We added the comment in the line 59.

 

1. 99-100: The sentence is a little confusing as it starts with stating “XRD patterns of TiO2 101”, but then the comparison is based on TiO2 001. Maybe the first half of line 99 is not necessary.

 Response:

 

Thanks for your comment. The sentence of line 99 is error. We had modified it in the line 99. “Obviously, the XRD patterns of TiO₂-101 is same as the purity anatase TiO₂,”.

 

1. 115-116 the order of antisymmetric and symmetric bending vibration is reversed in the list.

 

Response:

 

Thanks for your comment. Tian et.al. (J. Phys. Chem. C 2012, 116, 7515–7519.) reported that the B_1g peak is caused by symmetric bending vibration of O-Ti-O, the A_1g peaks at 515 cm^-1 is caused by antisymmetric bending vibration of O-Ti-O. To express more clearly, we had reworked this sentence in the line 115. “The vibrational mode of E_g peak at 144 cm^-1 and 638 cm^-1 are mainly caused by symmetric stretching vibration of O-Ti-O in TiO₂, the B_1g peak at 398 cm^-1 is caused by symmetric bending vibration of O-Ti-O, the A_1g peaks at 515 cm^-1 is caused by antisymmetric bending vibration of O-Ti-O.

 

 Table 2 the first column has bold text and a horizontal line not needed between TiO2-101 and TiO2-001.

 

Response:

 

 Thanks for pointing out this error. This error has been corrected in Table 2.

 

1. 370: Was the precipitate washed after or before centrifugation? Before does not really make sense?

 

Response:

 

 Thanks for your comment. The suspension first was centrifugated and then the precipitate obtained was washed using water and ethanol, finally the precipitate was centrifugated to obtain Ti(OH)₄ precursor. To make clearly, we revised this sentence in the line 370.

 

L 419: Was the degassing in vacuum or flowing inert gas?

 

Response:

 

Thanks for your comment. The degassing was in vacuum.

 

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The authors addressed all my comments. I suggest publish.