Mesoporous TiO₂ Implanted ZnO QDs for the Photodegradation of Tetracycline: Material Design, Structural Characterization and Photodegradation Mechanism

Questions/suggestions

Response

The paper reports on the template-assisted synthesis of a composite material consisting of ZnO nanoparticles supported on mesoporous titania. The material was successfully applied as catalyst for promoting the photo-degradation of tetracycline. Hypothesis for reaction pathways was suggested based on the characterization of intermediate products. The experimental work and the interpretation of the results was obviously done with expertise. What is missing is a EDX spectroscopy investigation and the proof of the presence of TiO2 and ZnO. In particular the presence ZnO has to be confirmed, and the speculations in lines 137-141 need to be reconsidered based on sound experimental evidence. The manuscript can be accepted for publication if this and the following minor points were satisfactorily addressed:

Thank you for the comments and the suggestions.

The proof of ZnO presence in the matrix of TiO₂ can be seen from the characterization data of meso-TiO₂ and TZQ. The changes in the intensity of the XRD peak will likely not occur without the presence of ZnO.

The lattice fringes related to ZnO is also visible in the HRTEM of TZQ together with TiO₂. The deconvolution of PL spectrum TZQ also shows spectral evidence for ZnO.

Furthermore XPS analysis detected the presence of Zn and O which confirmed the presence of ZnO in the TiO₂ matrix. Compared to EDX, XPS is more sensitive in detecting chemical species. Hence, we have decided not to perform EDX analysis.

The reasons given for the absence of ZnO peaks in the XRD diffractogram of TZQ were not merely speculative. TZQ was prepared via one pot synthesis method. During the synthesis, in addition to TTIP, a template (starch) was used, the pH was adjusted, and high calcination temperature was used. Moreover, during the hydrolysis and condensation processes, Ti-O-Zn structures will be formed. Hence, this may account for the reason for the structure of ZnO QDs to change.  It has been reported that XRD will not detect metal oxide with lower concentration. Based on these arguments, we would like to request to keep the points. To avoid ambiguity, we have replaced the word “possibilities” in “The absence of the peaks can be due to several possibilities”…(line 138) to “several reasons”.

Line 93: “Fukui function approach and the Wheland localization” Mentioned in two subsequent sentences. Consider a more concise rephrasing.

Thank you for the suggestion. We have removed the sentence. Please refer to line 97-98.

105: “…small crystallite size and low crystallinity” small YES but low crystallinity is questionable. The peaks are there, and this clearly indicates the presence of crystals with a small size.

Thank you for the comment. The reduction in crystallite size and crystallinity is indicated by the reduction in the band intensity and appearance of line broadening. The XRD bands of bulk ZnO will be intense and narrow due to its high crystallinity. The arrangement of atoms in materials with poor crystalline materials are random, hence the scattering angle will vary resulting in broader bands.  This is evidenced in the diffractogram of meso- TiO₂ and TZQ. Compared to meso-TiO₂, the bands attributed to the TiO₂ appear to be broader and less intense. The change indicates that the TiO₂ in TZQ has lower crystallinity compared to meso-TiO₂.  

108: The Scherrer equation is well-known and must not explained here

Thank you for the comment. We agree with the reviewer, and we have removed the equation.

130: It is certainly not necessary to list all reflections here.

Thank you for the comments. We have included the reflection peaks to give better understanding in the revised manuscript.

134: “…relate to the presence of Ti-O-Zn bonds.” How is that possible? Broadness of reflections is caused by small size. ZnO and TiO2 are present is separated crystallites as evidenced by HRTEM.

Thank you for the question. As mentioned before, structural disorder in the materials will give different scattering angle hence leading to broader lines. Smaller crystallite size can also lead to broader features. The synthesis of TZQ is carried out in one pot synthesis method. Hence, during the condensation and hydrolysis of TiO₂, it can react with the Zn(OH)₂ (in alkaline pH, ZnO can be converted to Zn(OH)₂₎ to form Ti-O-Zn bonds. Due to different size and bond length, the crystallinity can be reduced. We have revised the language in the revised manuscript to provide better explanation. Please refer to line 135.

144: change to: ca. 11.4 nm to 9.3 nm appears.

Thank you for the suggestion. The correction has been made. Please refer to line 145-146.

162: “along an apical direction leading” Pure speculation

Thank you for the comment. The statement was included based on the data that we have collected. The XRD analysis indicate the presence of anatase TiO₂.  Based on our experimental method, this is the best account that we can propose based on our results.

169: It is unclear where the ZnO is. Thus, EDX spectroscopic characterization is indispensable.

The proof of ZnO presence in the matrix of TiO₂ can be seen from the characterization data of meso-TiO₂ and TZQ. The changes in the intensity of the XRD band will not occur in the presence of ZnO.

The lattice fringes related to ZnO is also visible in the HRTEM of TZQ together with TiO₂. The deconvolution of PL spectrum TZQ also shows peak related to ZnO.

Furthermore, XPS analysis detected the presence of Zn and O which confirmed the presence of ZnO in the TiO₂ matrix. Compared to EDX, XPS is more sensitive in detecting chemical species. Hence, we have decided not to perform EDX analysis.

The reasons given for the absence of ZnO peaks in the XRD diffractogram of TZQ were not merely speculative. TZQ was prepared via one pot synthesis method. During the synthesis, in addition to TTIP, a template (starch) was used, the pH was adjusted, and high calcination temperature was used. Moreover, during the hydrolysis and condensation processes,  Ti-O-Zn structures will be formed. Hence, they can be related to the change in the crystallinity of the ZnO QDs.  It has been reported that XRD will not detect metal oxide with lower concentration. Based on these arguments, we would like to request to keep the points. Updates to the revised manuscript are highlighted in the markup version of the revised manuscript (line 138).

Figure 3: 0.351 nm – more accurate than possible to measure; change to 0.35 nm. This is also the case for the particles size, e. g. 12.54±2.05 -> 12.5±2

Designation of the figure incomplete add (d),(e),(f)

Thank you for the comment. The particle size was changed. Please refer to line 215 and line 219.

The capacity to edit images is somewhat limited at the present time; however, the values of the lattice fringes. Lattice fringes have been frequently reported as 3 significant figures. The designation of the figure was changed as recommended by the reviewer.

306: What is the reason for the reduction of Ti4+?

Thank you for the question. Ti⁴⁺ in the manuscript does not refer to the reduction. It refers to the Ti⁴⁺ in TiO₂.

390: was used TO assess

Thank you for the comments. The sentence has been corrected. Please refer to line 392.

810: Microscopes used for TEM and SEM should be given.

Thank you for the comments. The brand of the microscopes have been included. Please refer to line 834-836.

Questions/suggestions

Author Response

The authors have synthesised a mesoporous TiO2 implanted ZnO quantum dot (TZQ) as a photocatalyst using a sol-gel method for the photodegradation of tetracycline (TC) under fluorescent light irradiation. The main novelty of the work is the way of synthesis. However, the introduction has not covered enough discussion about the templated synthesis, especially the hard template. Hard/soft templated synthesis of porous materials should be mentioned in the introduction. I suggest the authors to use some review papers that have focused on the hard template synthesis of nanoporous materials, such as Hard-templated synthesis of metal-organic frameworks.

Thank you for the recommendation. We have updated the introduction section. However, we have decided to focus on sol-gel method. To the best of our knowledge, we could not find relevant work that discusses the synthesis of mesoporous TiO₂/ZnO QDs nanocomposites.

The XRD discussion needs to be more rigorous. Especially the one relating to TZQ. Out of the scenarios offered for not detecting any ZnO peaks, which one do the authors think is more plausible and why? Maybe they can re-examine the TEM images for extra clues.

We appreciate the reviewer comments. The rationale was based on data obtained in this study. The interpretation provided is also supported with references from the current literature. The revised manuscript highlights changes made to address the reviewer query. Please refer to line 138.

Regarding the XPS, can it be Zn2+ substituting Ti4+ creates an oxygen vacancy without Ti3+? Can the authors explain, without any doubt, the origin of Ti3+? Clarification can be done by quantifying the level of O vacancy and demonstrating that the existing Zn substitutes can not neutralise all the vacancies, and therefore, Ti3+ ought to be created for maintaining charge neutrality.

Thank you for the questions. We are aware  that the variation in the area below the bands for Ti⁴⁺ and Ti³⁺ can be used to show the removal of oxygen to form oxygen vacancies. However, due to limited funding, we are only able to perform the TiO₂ survey and the element quantification. Hence, we are unable to quantify and provide clarification. However, referring to the survey spectra (Fig 4a), the peaks in TZQ were shifted compared to meso-TiO₂. The shifting indicates that Zn²⁺ may have substituted the Ti atom to form Ti-O-Zn structures. Since only a small amount of ZnO was added relative to TiO₂, it is unlikely for the Zn²⁺ to neutralize vacancies. The deconvolution of XPS and PL analysis indicate the presence of Ti⁴⁺ and Ti³⁺. We have modified the write-up for XPS in the revised manuscript. Please refer to Section 2.1.3.

It is well-known that electrons and holes in heterojunctions and nanostructures suffer lower mobility due to boundary scattering, high defect concentrations, and interface resistance. Can the authors elaborate on why their composite material has a better photocatalytic activity in TC degradation? Can it be because all Zn ions were homogeneously distributed in the titania matrix, reducing the resistivity? The authors can refer to these two relevant publications in developing their arguments:

Doustkhah et al. (2021) Applied Catalysis B: Environmental, https://doi.org/10.1016/j.apcatb.2021.120380

Assadi & Hanaor (2016) Applied Surface Science, https://doi.org/10.1016/j.apsusc.2016.06.178

Thank you for the comments. The reason for the high photocatalytic activity of TZQ over P25 and meso-TiO₂ are due to the presence of quantum confinement effect of ZnO QDs. The QCE will be able to suppress the recombination rate of the highly energetic photogenerated e-/hole pairs. In addition to the QCE, the higher photocatalytic activity of TZQ is due to the formation of z-scheme which also will suppress the recombination rate. The third factor that contributes to the increased photocatalytic activity is the presence of adequate amount of various types of defects. Please refer to paragraph 1, Section 2.2.1 and 2.2.7. 

The suggested papers by the reviewer are very interesting and intriguing. However, we feel that the discussions in the cited work extends beyond the key objectives of this study and will be further considered in future studies. The Fukui function approach and the Wheland localization approach was used to investigate the interaction between radicals and TC.

Furthermore, the quality of some figures is poor and it seems the authors have copy-pasted from somewhere. For instance, Figure 8, should be re-written.

Thank you for the comment. We have replaced figure 8 and improved the quality of other images.

Questions/suggestions

Response

Please, kindly highlight the innovative component of study, in the abstract, within the context of water/environment protection.

Thank you for the comment. We have revised the abstract according to the suggestions made by reviewer #3 to address the concerns.

Please, pay attention in terms of English Grammar.

For instance, in the title, use caps lock for all nouns.

Thank you for the suggestion. The manuscript was sent for proofreading to improve the quality of the English language.

The authors should clearly identify a clear and coherent objective of the research. Please, stress some key points, as well.

Please, write a more concise state of the art, highlighting:

1. What is the importance and scope of this work in the context of this journal? Need to discuss the relation/impact of the proposed work with the scope of Catalyst in introduction briefly.
2. Abstract would be kept simple and precise with the novelty of the paper.

Thank you for the comments and the suggestions given to improve the quality of the manuscript. The abstract and the introduction have been revised to addres the reviewer comments and suggestions.

Questions/suggestions

Response

The authors have not revised the paper based on the reviewer’s comments and furthermore, there are some serious flaws in authors claim that is totally opposite of the what is shown in the analysis, more importantly, authors claim that their material is mesoporous, however, based on the BJH the average pore size is more than 60 nm that is classified as macroporous material. Furthremore, the mesoporosity can not be seen in any other characterization. 

We appreciate the reviewer’s concern. The manuscript was revised according to the comments as best as possible based on the data we have collected.

In lines 399-400, the unit for the pore size distribution if given in angstrom (Å) not nm. According to IUPAC, mesopores is between 2-50 nm (20-500 Å)[please refer to https://doi.org/10.1351/pac199466081739]. Hence, the pore size that we have stated is in the mesopores range.

The widely used and accepted method to determine the pore architecture of a material is through N₂ adsorption-desorption analysis which we had used. The BJH method can be used to determine the pore size distribution in the mesopore range, whereas mercury porosimetry can be used to determine macropores accurately.

The interpretation that we have made are based on the data that we have collected. We have also included relevant references where necessary. Hence, we did not make any claims that are opposite to the presented analyses.

How the authors have detected the enegy potential levels?

Thank you for the question. In this study we have calculate the energy potential levels value, which are highest occupied molecular orbital (HOMO), the lowest unoccupied molecular orbital (LUMO), the singly occupied molecular orbital (SOMO) of the Tetracycline compound by using DFT Gaussian. Please refer to Section 3.6 for more details.

The curve-fitting in PL spectra are not reliable, and cannot be acceptable.

We appreciate the reviewer’s concern. The deconvolution of the PL spectra was carried out with care, where favourable fits were obtained that match the XPS results. The goodness-of-fit criteria exceeded 0.99 for the adjusted values of R² for the meso-TiO₂ and TZQ.

Where is the source of Ti3+?? it should be explained. This big fraction cannot be orginated from spantenous process in the clacination in air without adding a reducing agent or a Ti3+ source.

Thank you for the questions. It is well known that Ti³⁺ can be formed in the presence of a reducing agent or thermal treatment such as calcination. This was outlined in a recent review (Chem. Eng. J. 2020, 382, 123011;  see the following: https://doi.org/10.1016/j.cej.2019.123011) This point was inadvertently omitted in the previous revision but it is now included it in the revised manuscript. Please refer to the markup version for updates.

SEM shows two types of morphology for ZnO loaded sample. More SEM images should be shown to prove that the ZnO is synthesized in the TiO2 vacant space. 

The SEM analysis is used to observe the morphology or texture of a material. In some cases, the different phases can be identified; however, it is unlikely to detect the presence of vacant space using SEM analysis to the best of our knowledge. Even if EDX analysis was carried out, it could only give the distribution of the elements. We have substituted EDX using XPS which provide more information.

Due to the limited research budget, we are unable to carry out further imaging. The images that we have selected represent the overall morphology of TZQ and meso-TiO₂.

Adsorption of tetracycline should be examined and shown in the MS.

The photocatalytic analyses showed the adsorption of tetracycline is not very large. Hence it was not made as the main objective of this research. We have decided to focus on the photocatalytic activity.

More importantly, the larger peaks of ZnO in the final sample cannot be observed, i.e., 100 and 101 planes’ peaks. How the authors claim that ZnO is formed??

Thank you for pinpointing the mistake we have made. We have removed the “…the (100) planes of the Wurtzite…” (line 201) to avoid confusion.

Even though the XRD of TZQ was carried out multiple times (cheaper compared to SEM in our institution), the ZnO phase was not detected.  We have cited several reasons for this observation (refer to to the revised markup version of the manuscript).

If the authors belive that they have synthesized QD, they should show the QDs in the TEM image, otherwise it is not acceptable to claim without evidence. The current TEM images don’t show any QD.

Quantum dots are defined as semiconductor nanocrystalline within the size range of 1.5 - 10 nm. The particle size that we have determined is 4.12 nm which is less than 10 nm. Hence, it is qualified to be categorized as quantum dots.

Questions/suggestions

Response

The BJH of TiO₂ here shows a peak below 50 nm, but the peak is very broad and shows that the peak can be due to the network porosity, therefore, it should be proved.

Moreover, the peak shifts to upper than 50 nm in TiO2-ZnO composite that you can never claim it is mesoroporous material.

The most important is the final material that it cannot be mesopore, according to the BJH. 

Thank you for the concern. We agree with Reviewer #2 that the pore size distribution peak is broad. This is why we used the word “centered” in describing the pore size distribution. Please refer to lines 372-372. The unit used for the pore size distribution is Å not nm, which falls into the category of characterized by such mesoporous materials.

The pore size distribution of TZQ is less than 40 Å (4 nm) which resides in the mesopore range according to IUPAC.  The hysteresis loop can be observed clearly in the sorption isotherms of the meso-TiO₂ and TZQ in the characteristic relative pressure range characteristic of mesoporous materials. Hysteresis loops of this type fall into the Type IV category, according to the IUPAC classification, and can only be observed if the material is mesoporous. Hence, we are confident that the materials we have synthesized are mesoporous.

N₂ adsorption at lower relative pressure were not observed in the isotherms, hence the meso-TiO₂ and TZQ do not support evidence for any significant micropore features.

As mentioned in previous response, macropores can only be accurately detected using mercury porosimetry. BJH method could not detect the presence of macropores. Finally, with reference to Fig. S2 a-b, in the Supporting Material, the pore size distribution (PSD) is shown for meso-TiO₂ and TZQ. The reduction in the PSD of meso-TiO₂ (ca. 7.5 nm) upon implantation of the ZnO QDs into meso-TiO₂ (ca. 3.4 nm) is consistent with the reduction in the PSD noted for TZQ (cf. Fig. S2(b) and Table 3). The breadth of the PSD profile relates to the amorphous nature of the materials, according to the XRD results (Fig. 1) and the FESEM images (cf. Fig. 2).

I still cannot accept that the authors have loaded ZnO on TiO2 according to the XRD, at least teh 101 plane of ZnO should appear in the composite, but it cannot be observed. 

We appreciate the constructive comments given by the reviewer. The XRD results in Fig. 1 provide support that TZQ is more amorphous relative to its constitutent components (ZnO QDs and meso-TiO₂ single phase materials). The broadening of the XRD features of meso-TiO₂ provide support for this conclusion, along with evidence for the loss of crystallinity of the ZnO QDs, where the latter is of lower composition in the binary composite. The FESEM results (Fig. 2) show support for ZnO loading onto TiO₂ along with the trends in PSD values obtained from the nitrogen adsorption isotherm results, described above. Additional support for the foregoing concurs with the results obtained from XPS, HR-TEM and PL. In summary, the authors suggest that the experimental evidence that support the conclusions related to implantation of ZnO QDs onto meso-TiO₂. See also, the offset between the TOC removal for P25 and TZQ (cf. Fig. 7c).

SEM shows the presence of another out-of-shape particles that shows the morpholgy is not uniform. And the authors did not provide more SEM images to show if the particles are uniform. 

Thank you for this comment. In line with the revised manuscript and the above discussion, we agree with the reviewer that the particles are not uniform in size. This concurs with the results from the structural characterization presented in this study. The changes that occur upon incorporation of ZnO QDs onto TiO₂ is highlighted in Section 2.1.2. The word “uniform” was not mentioned in any parts of the manuscript. The manuscript was further edited to check for any such discrepancies.