Theoretical Study on the Lewis Acidity of the Pristine AlF3 and Cl-Doped α-AlF3 Surfaces
Round 1
Reviewer 1 Report
The article deals with theoretical studies of various models of alpha-AlF3 surface and its active centers in terms of Lewis acidity. Chlorinated equivalents that describe experimental AlClxF3-x systems are also studied. The Lewis acidity of the active sites is a measure of the reactivity with electron-rich compounds. Lewis acidity is tested by adsorbing small molecules such as CO and NH3. Authors use the periodic DFT method with the PBE functional and dispersion corrections. Calculations show that the tetragonal coordination of aluminum exhibits the largest shielding against small molecule adsorption. The calculated frequency shifts for CO and NH3 agree with the experiment. Authors conclude that the high catalytic activity is related to the amorphous nature of the catalyst. Models of active centers obtained in the paper can be used to study the mechanisms of catalytic reactions.
The work deals with important issues related to catalytic systems activating C-H and C-F bonds. The article is written clearly and interestingly. In my opinion, the article is suitable for publication in its current form.
Author Response
Authors reply:
We thank the reviewer for the recommendation.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
In the manuscript " Theoretical study on the Lewis acidity of the pristine AlF3 and
5 Cl-doped α-AlF3 surfaces" authors investigated the bonding of CO and NH3 on AlF3 and ACF surfaces by computing adsorption energies and vibrational frequencies.
I don't understand the purpose of the work. Authors report a series of adsorption values and IR shifts, but do not provide a rationale other than the well-known dangling bond reasoning. Also, authors claim to provide structural insights on the adsorption of CO and NH3 on model surfaces, but the exploration of the adsorption modes is far from being complete and bonds are described only on the basis of adsorption energies and vibrational frequency shifts, while details on the electronic structure would be affordable and will provide a better understanding of the bonding modes of CO and NH3.
I have also the following remarks on the methodology:
- At page 3, line 112, authors state that molecular probes were placed at x=0.5 and y=0.5 fractional coordinates and then allowed to relax. However, using whatever geometry optimization algorithm, the probe will end up in the closest local minimum and not in the global minimum as it would make sense in this context. A better exploration of the PES is demanding to evaluate the overall properties of surfaces. Besides, I'm not sure that the statement is true, because for the (11-22) surface two binding sites were investigated.
- At page 5, line 161, authors report an adsorption energy of -1.29 eV for ammonia on the (01-10) surface and claim that this is due to three hydrogen bonds. By no means the formation of three hydrogen bonds can justify such a value.
- there is a huge difference in surface area among different models and in a few cases the lateral distance between molecular probes is short enough to allow their interaction. This prevents the direct comparison of adsorption energies, because differences can arise due to the lateral interactions.
- Adsorption modes at the (0001)-1x1 surface, as shown in fig.5, seem unreasonable from a chemical point of view. Why molecules do not adsorb on top of Al as one may expect? Moreover, the chosen surface model conflicts with the one used in reference 20, where Bailey et al used a 1x2 reconstruction, that is said to be more stable than the pristine surface model.
- Surface models were taken from a previous work of the same authors, where they found that the most stable surface is the (01-10). However, ref 19-21 report that the most stable surface of AlF3 is the (01-12). This latter surface model should be at least included in the work, also to extend the comparison of adsorption energies with previous works (e.g. ref. 20).
Author Response
Reviewer 2
Comments and Suggestions for Authors
In the manuscript " Theoretical study on the Lewis acidity of the pristine AlF3 and 5 Cl-doped α-AlF3 surfaces" authors investigated the bonding of CO and NH3 on AlF3 and ACF surfaces by computing adsorption energies and vibrational frequencies.
I don't understand the purpose of the work. Authors report a series of adsorption values and IR shifts, but do not provide a rationale other than the well-known dangling bond reasoning. Also, authors claim to provide structural insights on the adsorption of CO and NH3 on model surfaces, but the exploration of the adsorption modes is far from being complete and bonds are described only on the basis of adsorption energies and vibrational frequency shifts, while details on the electronic structure would be affordable and will provide a better understanding of the bonding modes of CO and NH3.
Authors reply:
We thank the reviewer for the recommendation and agree to the suggestion of the expandable electronical investigation. We clarified the purpose of the work by following sentence in the introduction:
“For the analysis of the electronic structure the Bader charges for all surface models and density of states for selected ones are added to the SI. In none of the cases interesting electronic contributions to binding were found, therefore a more detailed analysis of the electronic structure with other methods like ELF or bond analysis on surfaces is not necessary.[24] Both are not straight-forward methods with a plane-wave code used in this work.”
We carefully tried to get insides with the analysis of the bader charges and projected density of states as now implemented in the SI:
“ In Table S2 we present the Bader charges of the adsorbed CO on different surface models. Compared to the free molecule, only for the adsorption on the threefold coordinated Al-sites changes larger than 0.1 e are observed. Therefore, it is concluded, that there is only a negligible covalent contribution to binding for the four- and fivefold adsorption sites.
In Table S3 we present the Bader charges of the adsorbed CO on different surface models. Compared to the free molecule, only for the adsorption on the unchlorinated threefold coordinated Al-sites changes larger than 0.2 e are observed. Therefore, it is concluded, that there is only a negligible covalent contribution to binding for the four- and fivefold adsorption sites.
To elucidate the electronic structure further we present the projected density of states for the terminal chlorinated α-AlF3-(011̅0) surface (see fig. S3 with adsorbed CO and the plane surface in fig. S4). It was found that the unoccupied states of carbon monoxide are slightly below the conduction band of the surface. The occupied states of CO are well embedding in the valence band. We found no signatures for any hybridisation and therefore also the projected density of states gives no hint for a covalent contribution to binding.”
We also extended the introduction of the paper to make the purpose clear.
I have also the following remarks on the methodology:
• At page 3, line 112, authors state that molecular probes were placed at x=0.5 and y=0.5 fractional coordinates and then allowed to relax. However, using whatever geometry optimization algorithm, the probe will end up in the closest local minimum and not in the global minimum as it would make sense in this context. A better exploration of the PES is demanding to evaluate the overall properties of surfaces. Besides, I'm not sure that the statement is true, because for the (11-22) surface two binding sites were investigated.
• Authors reply: At page 3, line 112, we clarified the process of the positioning of the probe molecules at the start of the investigation and added the following sentence:
• “Additional starting structures different to the one with which the first minimum structure was reached were applied, especially if multiple possible active centers are present. With this approach a wider range of adsorption positions of certain surface models was included.”
• At page 5, line 161, authors report an adsorption energy of -1.29 eV for ammonia on the (01-10) surface and claim that this is due to three hydrogen bonds. By no means the formation of three hydrogen bonds can justify such a value.
• Authors reply:
At page 5, line 161, we added a reference to the work of Steiner et al. to classify the strength of the hydrogen bonds:
“Therefore, it can be assumed that each hydrogen bond add -0.43 eV respectively to the overall adsorption energy of the molecule. This is within the range of medium sized hydrogen bonds in small molecules.[46]”
• there is a huge difference in surface area among different models and in a few cases the lateral distance between molecular probes is short enough to allow their interaction. This prevents the direct comparison of adsorption energies, because differences can arise due to the lateral interactions.
• Authors reply: To check the possible interactions of the adsorbed molecules at the different sized surfaces we carefully checked the influence of the different distances between the molecules in different surface cells and added the following sentence:
“Although different surface unit cells have different surface areas, in all cases the adsorbed molecules are so far apart that their interaction energy is less than 5 meV and therefore below the accuracy which can be achieved with the current set-up.”
• Adsorption modes at the (0001)-1x1 surface, as shown in fig.5, seem unreasonable from a chemical point of view. Why molecules do not adsorb on top of Al as one may expect? Moreover, the chosen surface model conflicts with the one used in reference 20, where Bailey et al used a 1x2 reconstruction, that is said to be more stable than the pristine surface model.
• Authors reply: As pointed out in the remark on the 1x1-(0001) surface we were surprised by ourselves and double checked this structure. Therefore, the structure of CO directly on top of the aluminium site were found to be energetical less favoured by 0.2 eV. Also, the purpose of the investigation of this surface was to include threefold coordinated aluminium sites since four- and fivefold coordinated aluminium centers are already included in our work.
• Surface models were taken from a previous work of the same authors, where they found that the most stable surface is the (01-10). However, ref 19-21 report that the most stable surface of AlF3 is the (01-12). This latter surface model should be at least included in the work, also to extend the comparison of adsorption energies with previous works (e.g. ref. 20).
• Authors reply: To improve the comparison of our results with the previous ones, we improved the discussion of the work of Bailey et al. on the CO adsorption at their surfaces in the introduction:
“In the dissertation by Bailey [20] a few surface structures of α- and β-AlF3
were characterized via the adsorption of CO at the model surfaces.[20] In their work the adsorption of CO at the fourfold coordinated aluminium centers on the α-AlF3-(0001) surface within a 2x1 cell and on the α-AlF3- (011̅2) surface with a 1x1 and a √2 x √2 surface cell were investigated under the usage of the Crystal code with the B3LYP basis set. The α-AlF3-(0001) surface model shows adsorption energies between -0.16 to -0.30 eV with blue shifts of the CO stretch vibration between 78 and 88 cm-1. For the α-AlF3- (011̅2) adsorption energies between -0.52 eV for the 1x1 and -0.21 eV for the √2 x √2 surface were found together with blue shifts of the CO stretch vibration between 72 and 87 cm-1 . For both surface models the theoretical results are in the range of the experimental
data.[20] Our previous investigations[21] have shown, that the (011̅2) surface is quite high in energy and only another layer of fluorides and surface reconstruction stabilizes the α-AlF3- (011̅2) surface. Locally the structural motifs found by Bailey[20] and ours agree well, proven by similar frequency shift obtained for CO. Contrary to the model of Bailey et al. the (0001) surface model used in this work shows threefold coordinated aluminium centers due to the usage of a different theoretical set-up and no additional fluorides.[21]”
Again, we want to thank for the detailed recommendation, which helped us to improve our work and to clarify our focus.
Author Response File: Author Response.pdf
Reviewer 3 Report
Work written correctly. The presented results confirm well the conclusions presented at the end of the paper.
Elements that require improvement:
- the figuress, especially those that are an enlargement of a selected fragment of the structure, should contain the bond lengths described in the text (it would make it much easier to follow the author`s thinking);
- the discussion is limited to the geoemtic parameters and IR spectra, while the title of the paper suggests a full theoretical description - in my opinion a description of the electronic structure of systems should be added (influeance of CO and NH3 adsorption) or the title of the paper should be clarified.
Author Response
Reviewer 3
Work written correctly. The presented results confirm well the conclusions presented at the end of the paper.
Elements that require improvement:
• the figuress, especially those that are an enlargement of a selected fragment of the structure, should contain the bond lengths described in the text (it would make it much easier to follow the author`s thinking);
• the discussion is limited to the geoemtic parameters and IR spectra, while the title of the paper suggests a full theoretical description - in my opinion a description of the electronic structure of systems should be added (influeance of CO and NH3 adsorption) or the title of the paper should be clarified.
Authors reply:
We thank the reviewer for the recommendation. We appreciate the suggestion by the reviewer but including the distances in the figures would make them even more crowded. We have added some comments on the electronic structure in the manuscript and provided data on Bader charges and density of states in the SI.
Author Response File: Author Response.pdf
