Effect of B2O3 and Basic Oxides on Network Structure and Chemical Stability of Borosilicate Glass
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThe work presented by Ming et al. focuses on the Effect of B2O2 on the Network Structure and Chemical Stability of Borosilicate Glass .
While the scientific work, as well as the used characterization methods, are well performed, however, not recommend acceptance of the manuscript in its present form.
In many places, the manuscript presently has a structure more similar to an extensive lab journal than a scientific article, where findings, results and interpretations are frequently mixed and presented in chronological order, most likely as the experiments were conducted. This makes the article much longer than necessary and will not add value to the readers.
It is not possible to extract the main novelty of the work. I would suggest to re-structure the work, put the key information in the main manuscript and add other information in the supplementary information.
There are several parts of the manuscript and aspects of the scientific work that I believe the authors should improve before acceptance. Some parts relate to the overall manuscript structure, while others refer to the motivation of the work and the conclusions made. A few of the most important points are specified below.
1. Please specify which rare earths were included.
2. Please provide the advantage to use the Igor software for fitting the Raman spectra.
3. Please re-organize the Table numeration, you start to enumerate and explain Table 3.
4. There is no Table 3 in the manuscript. Authors have addressed to this table3 couple of times, but doesnt exits. It is not possible to correlate the the glass transition temperature .
5. If its not possible to extract that the CTE increases by checking the graph, it seems that all curves have same slope it means same CTE, so please add the calculated values of CTE for each sample A1-to A6.
6. Same as tables, please organize, you starte to explain Fugure 5 without explaining figure 3 and 4.
7. Authors talks about EDS experiments done however didnt give the results of the composition done by EDS. So please add the EDS results and where were done to judge the difference of composition in the dendritic and outside dendritic forms.
8. Please add references to all statements related to the proportion of non-bridging 232 oxygen atoms in the glass network structure in page 8.
9. Please add which type of fitting where used for Raman analysis, type of peak gaussian pseudo Voigt, lorentzian... etc, including all their parameter for doing deconvolution.
10. Please provide a table of Raman peaks with all peaks assigned and their location. If It not possible to judge the shift of peaks if you dont give in-depth info of the Raman analysis.
11. Add references why you have employed the proposed formula to determine the fraction of non-bridging oxygen.
12. The author presents values of CTE that ranges from 5.5 to 9 ppm, please add units. It is quite strange how you get this values showing similar slopes in CTE curves.
The findings presented by the authors are certainly interesting and should eventually be published. In my opinion, however, the structure (story) of the manuscript can (and should) be improved to increase the overall appearance before acceptance.
Comments on the Quality of English Language
English can be improved
Author Response
Comments 1: Please specify which rare earths were included.
Response 1: Detailed chemical formula (LaCePrZr) O2
Comments 2: Please provide the advantage to use the Igor software for fitting the Raman spectra.
Response 2: Igor software is easy to operate in the data processing of Raman spectroscopy, and can better process substrates and perform multi peak fitting simultaneously.
Comments 3: Please re-organize the Table numeration, you start to enumerate and explain Table 3.
Response 3: We fcel sorry for our carelessness. In our resubmitted manuscript The description of table 3 was deleted Thanksforyour corection.
Comments 4:There is no Table 3 in the manuscript. Authors have addressed to this table3 couple of times, but doesnt exits. It is not possible to correlate the the glass transition temperature .
Response 4: We fcel sorry for our carelessness. In our resubmitted manuscript The description of table 3 was deleted. The glass transition temperature is reflected in Table 2 in relation to the other parameters.
Comments 5: If its not possible to extract that the CTE increases by checking the graph, it seems that all curves have same slope it means same CTE, so please add the calculated values of CTE for each sample A1-to A6.
Response 5: Regarding the CTE curve, the vertical coordinate of the graph indicates the coefficient of thermal expansion of the sample rather than the slope. I am sorry for your misunderstanding, but the exact variation curve is shown in Fig. 8, and accordingly I have added a paragraph in the article to describe it.
Comments 6-7: Same as tables, please organize, you starte to explain Fugure 5 without explaining figure 3 and 4. Authors talks about EDS experiments done however didnt give the results of the composition done by EDS. So please add the EDS results and where were done to judge the difference of composition in the dendritic and outside dendritic forms.
Response 6-7:Thank for your feedback and we have modified. Figures 3-5 all show the micro-morphology of the samples after corrosion, and Figure 5 shows the EDS results, all with the aim of demonstrating the different degrees of phasing that occur in different samples, and I have added a description of this paragraph in the article accordingly.
Comments 8: Please add references to all statements related to the proportion of non-bridging 232 oxygen atoms in the glass network structure in page 8.
Response 8: Thank for your suggestion and we have added add references to all statements related to the proportion of non-bridging 232 oxygen atoms in the glass network structure in page 8.
Comments 9: Please add which type of fitting where used for Raman analysis, type of peak gaussian pseudo Voigt, lorentzian... etc, including all their parameter for doing deconvolution.
Response 9: In order to characterize the modification of the glass network, a multimodal Gaussian fitting approach was employed.The glass phases were analysed using a Renishaw inVia micro confocal laser Raman spectrometer at a wavelength of 532 nm and at a magnification of X50. The spectral range was 100-2000 cm-1 with 10% energy, the collection time for each spectrum was 60 S, and the laser spot size was about 2-3 μm. Igorpro software was used to perform a split-peak fitting process on the collected data and subtracted the baseline. Spectra of glass-phase Raman spectra after baseline subtraction and Gaussian fitting. Two vibrational peaks appeared at 500 cm-1 and 1000 cm-1 , which originated from the bending and stretching modes of [SiO4]n- . After the multi-peak fitting process, the stretching vibrational peaks can be separated into four components, the Raman shift of Qn of each sample adopts a uniform standard as follows: the band at 780 cm−1 can be attributed to Q1 (three non-bridging oxygens in the tetrahedron);The band at 900cm−1 can be assigned to Q2 (two non-bridging oxygens in each tetrahedron);The band at 990 cm−1 is due to Q3 (Si–O band stretching vibration in each tetrahedron with a non-bridging oxygen) and the band at 1060 cm−1 is Q4 (Si–O band stretching vibration in fully polymerized unit) results.
Comments 10: Please provide a table of Raman peaks with all peaks assigned and their location. If It not possible to judge the shift of peaks if you dont give in-depth info of the Raman analysis.
Response 10: We feel great thanks for your professional review work on our article and we have tried our best to add explanations in the revised manuscript.The main emphasis of the article is on peak broadening, and its purpose is to demonstrate that an increase in the area of the modified peak is also an increase in Qn, and the specific parameters of the Qn area are captured in Table 2.
Comments 11: Add references why you have employed the proposed formula to determine the fraction of non-bridging oxygen.
Response 11: Thanks for your good comments. The significance of this formula is to calculate the amount of bridging oxygen in the glass network and thus to analyse the relationship between the degree of corrosion resistance of the glass and the bridging oxygen, with reference to the relevant parts of reference 23.
Comments 12: The author presents values of CTE that ranges from 5.5 to 9 ppm, please add units. It is quite strange how you get this values showing similar slopes in CTE curves.
Response12: Added coefficient of thermal expansion in °C-1 to the article Regarding the CTE curve, the vertical coordinate of the graph indicates the coefficient of thermal expansion of the sample rather than the slope. I am sorry for your misunderstanding, but the exact variation curve is shown in Fig. 8, and accordingly I have added a paragraph in the article to describe it. And the full text of the English writing was polished
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsAn interesting article presenting original research on the properties of borosilicate glass like thermal stability, thermal expansion, degree of polymerization of the glass network and acid resistance. Improvement of these properties was expected after adding alkali metal oxides (Na2O, K2O), alkaline earth oxide(BaO), Al2O3, TiO2, ZrO, Li2O. Boron oxide B2O3 content was used as a main factor of glass formulation with an expansion coefficient closer to that of tantalum metal.
The article was written in correct English and contains extensive, significant and important results, justifying their publication. The experimental part is sufficiently detailed and the literature listed is extensive.
Before final acceptance we suggest the Authors to check the following issues:
- Please check if the equation for thermal expansion coefficient, given in paragraph 2.5, lacks a Greek symbol “delta”; Equation is not numbered. Correct if necessary.
- Titles of section 3. “Related performance analysis” and section 3.1.”Intrinsic performance analysis” sound ambiguous. Therefore, please use "plain English". The main purpose of publishing a scientific article is to spread knowledge in a way that is understandable to the majority of the scientific and engineering community.
- Figure 2. Glass transition temperature (left) – the vertical axis (ordinate axis) is not titled (DTA signal?); Moreover, in the Abstract the differential scanning calorimetry (DSC) is announced, but in the text body – DTA. Both methods are used for thermal analysis but they are different. Please make corrections where necessary.
- Deconvolution of the Raman spectra. The symbol “Qn” appears first in the Abstract, then in line 256 the following explanation is given:” Qn represents different bridging oxygens”.
- How the deconvolution was made? What software was used? What function was supposed to model Raman peaks in Figure 6? (A multimodal Gaussian fitting approach was used in Figure 7). Please explain why do you consider 4 functions (Q1, Q2, Q3, Q4) while the deconvolution yielded 6 peaks contributing to the Raman spectra given in Figure 6.
- Please explain the use of the aggregation index. Figure 6 shows that aggregation index can be calculated for Q3.
Please explain the statement in line 263: “As shown in Table 2, the number of Ip ranges from 2.26 to 2.58, i.e. belongs to a group.” Given in Table 2 Ip values range from 1.98 to 3.82.
Author Response
Comments 1: Please check if the equation for thermal expansion coefficient, given in paragraph 2.5, lacks a Greek symbol “delta”; Equation is not numbered. Correct if necessary.
Response 1:Thanks for your good comments. we have checked and added the appropriate public number to the article.
Comments 2: Titles of section 3. “Related performance analysis” and section 3.1.”Intrinsic performance analysis” sound ambiguous. Therefore, please use "plain English". The main purpose of publishing a scientific article is to spread knowledge in a way that is understandable to the majority of the scientific and engineering community.
Response 2:We think this is an excellent suggestion. As suggested by the reviewer, We modified the relevant headline content and highlighted it in the article.
Comments 3: Figure 2. Glass transition temperature (left) – the vertical axis (ordinate axis) is not titled (DTA signal?); Moreover, in the Abstract the differential scanning calorimetry (DSC) is announced, but in the text body – DTA. Both methods are used for thermal analysis but they are different. Please make corrections where necessary.
Response 3:Thanks for your good comments. Differential thermal analysis (DTA) and coefficient of thermal expansion (CTE) of borosilicate glasses of different compositions. (DTA) is to indicate the specific glass transition temperature Tx, while (CTE) is to test the trend of the coefficient of thermal expansion of the glass.
Comments 4: Deconvolution of the Raman spectra. The symbol “Qn” appears first in the Abstract, then in line 256 the following explanation is given:” Qn represents different bridging oxygens”.
Response 4: We think this is an excellent suggestion.We have explained Qn, which first appeared in the abstract.
Comments 5: How the deconvolution was made? What software was used? What function was supposed to model Raman peaks in Figure 6? (A multimodal Gaussian fitting approach was used in Figure 7). Please explain why do you consider 4 functions (Q1, Q2, Q3, Q4) while the deconvolution yielded 6 peaks contributing to the Raman spectra given in Figure 6.
Response 5:In order to characterize the modification of the glass network, a multimodal Gaussian fitting approach was employed.The glass phases were analysed using a Renishaw inVia micro confocal laser Raman spectrometer at a wavelength of 532 nm and at a magnification of X50. The spectral range was 100-2000 cm-1 with 10% energy, the collection time for each spectrum was 60 S, and the laser spot size was about 2-3 μm. Igorpro software was used to perform a split-peak fitting process on the collected data and subtracted the baseline. Spectra of glass-phase Raman spectra after baseline subtraction and Gaussian fitting. Two vibrational peaks appeared at 500 cm-1 and 1000 cm-1 , which originated from the bending and stretching modes of [SiO4]n- . After the multi-peak fitting process, the stretching vibrational peaks can be separated into four components, the Raman shift of Qn of each sample adopts a uniform standard as follows: the band at 780 cm−1 can be attributed to Q1 (three non-bridging oxygens in the tetrahedron);The band at 900cm−1 can be assigned to Q2 (two non-bridging oxygens in each tetrahedron);The band at 990 cm−1 is due to Q3 (Si–O band stretching vibration in each tetrahedron with a non-bridging oxygen) and the band at 1060 cm−1 is Q4 (Si–O band stretching vibration in fully polymerized unit) results. Whereas the red curve in Fig. 6 is the original spectrum, it is the spectrum that emerges from deconvolution that is below the red curve.
Comments 6: Please explain the use of the aggregation index. Figure 6 shows that aggregation index can be calculated for Q3
Response 6: Thank you for your valuable comments.Qn significance is to calculate the bridging oxygen
Comments 7: Please explain the statement in line 263: “As shown in Table 2, the number of Ip ranges from 2.26 to 2.58, i.e. belongs to a group.” Given in Table 2 Ip values range from 1.98 to 3.82.
Response 7: Thanks for your good comments.The purpose of stating that the IP values 2.26 to 2.58 belong to a group is to indicate that glass components with similar IP values and similar corrosion resistance can be grouped together, but we have changed the expression in the article to make it clearer.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Comments and Suggestions for AuthorsThere are several parts of the previous review that the authors did not reply correctly. The manuscript can not be published if they authors did not take the comments in account
The authors did not reply question 5 in the initial review.
In the comment of figure 2 you state:
“Thermal expansion coefficients are shown as functions of temperature in Figure 2 (right), which reveals that the expansion coefficients of A1–A6 gradually increase with temperature; hence, the surface tension of the glass increases gradually with increasing temperature and is generally negatively correlated with the silica content.”
You are not representing the CTE, the curve shows the length change as a function of the temperature . The physical (differential) coefficient of linear thermal expansion is the slope of the expansion curve at a given temperature. So you need to clarify basic concepts.
The statement above is wrong. And you can not see that the CTE “gradually increase with temperature;”
Regarding comment 10. You should add an explanation of all abbreviations that you use in table 2 (From AQ1 to Tx).
The authors did not assign the peaks as it was required in the revision. Please add a table with Raman peaks with all peaks assigned and their location.
Comment 12 from previous review. In the manuscript is not explained Figure 8, the left axis related to CTE. In the corrected version I don’t see that you have added the units of CTE.
And again the author presents values of CTE that ranges from 5.5 to 8 ppm, 8 It is quite strange how you get this values showing similar slopes in CTE curves. Please add the calculations of CTE in the supplementary information.
The manuscript need a major revision and comments should be answer.
Comments on the Quality of English Language
Extensive editing of English language required
Author Response
Comments :
You are not representing the CTE, the curve shows the length change as a function of the temperature . The physical (differential) coefficient of linear thermal expansion is the slope of the expansion curve at a given temperature. So you need to clarify basic concepts.
The statement above is wrong. And you can not see that the CTE “gradually increase with temperature;”
Regarding comment 10. You should add an explanation of all abbreviations that you use in table 2 (From AQ1 to Tx).
The authors did not assign the peaks as it was required in the revision. Please add a table with Raman peaks with all peaks assigned and their location.
Comment 12 from previous review. In the manuscript is not explained Figure 8, the left axis related to CTE. In the corrected version I don’t see that you have added the units of CTE.
And again the author presents values of CTE that ranges from 5.5 to 8 ppm, 8 It is quite strange how you get this values showing similar slopes in CTE curves. Please add the calculations of CTE in the supplementary information.
Response : Thank for your suggestion ,I'm very sorry I didn't realise last time that there was a huge problem with the description of the thermal expansion curve in my manuscript. I have redrawn the relevant images and added detailed values for the coefficient of thermal expansion at 300°C in Table I. The modifications are highlighted in blue.
I have listed a detailed explanation from AQ1 to Tx and added detailed positions of the Raman peaks of the samples in Table 2, and I apologise for the oversight last time, I have added the units of CTE in the manuscript.
Finally regarding the CTE values, I have chosen 300°C (metal tantalum capacitor operating temperature) for the calculation of the slope, and I am very sorry for not describing it in detail before. Finally, I sincerely thank you for your valuable comments.
Author Response File: Author Response.pdf
Round 3
Reviewer 1 Report
Comments and Suggestions for AuthorsThe authors somehow replied the comments. A better understanding is needed. But I dont think authors will do at this stage. So I accept the paper, please check grammatical mistakes and units along all the manuscript.
Comments on the Quality of English LanguagePlease revise English for example: "coeffificient", "specifified" and more mistakes.
Author Response
We have studied the comments carctully and have made revision in this paver. The responses to the comments from two reviewers are offered separately, and the corresponding modification position page in the revised manuscript.
Reviewer :
Material preparation section (2.1) – a reader looking in this section would like to get a precise information on what kind of samples were investigated in this study. Please add a general composition of the glass at the beginning of this section.
Response :
I don't quite understand what you mean by adding ordinary borosilicate glass properties at the beginning of 2.1, so I added the following paragraph at the beginning of the second paragraph.
Low thermal expansion borosilicate glass occupies an important position in glass research because of its unique chemical structure, which is significantly different from conventional sodium-calcium silica glass. In the borosilicate glass system, the addition of alkali metal oxide drives the borooxygen trihedron [BO3] to the borooxygen tetrahedron The transformation of [BO4], this process not only transforms the boron from the layered structure to the frame structure, but also promotes the formation of [BO4] and the silicon oxygen tetrahedral [SiO4], thus enhancing the overall structural compactness and performance of glass.
Reviewer :
Moreover, the following sentence is not clear “Among these, three sets (B1–B3) of borosilicate glass…” – what are B samples? Be precise in samples presentation and avoid misleading information.
Response :
Thank you very much for your feedback. I have rephrased this paragraph to avoid the occurrence of three groups of glass that cannot be formed due to high viscosity at high temperatures.
Reviewer :
Some other phrases given here are not clear, such as: “All data in this study include quality percentages.” or “Detailed chemical formula (LaCePrZr)O2.” (this is a formula of … what? surely it is not correct; moreover, if it contains RE ions, Zr should not belong here). What is the role of RE ions in the glass? Were there intentionally added? I do not find this information.
Response :
I will rephrase this paragraph in the article, which mentions rare earth compounds with the aim of adjusting the thermal expansion coefficient of the glass.
Reviewer :
Figure 4 – what present insets in images (those in green)? Explanation in the caption is needed.
What is the difference between Figs. 6 and 7? It should be clear from the figures captions. “A1000” is not clear without referring to the main body text. Why Fig. 10 is a separate figure? What does is represent?Caption of figure 11 is not clear.
Response :
The green color inserted in Figure 4 represents the EDS surface scan results of the boron element with phase separation, and is explained in the title.
To avoid ambiguity, I have re annotated Figures 6 and 7 and changed their titles
Reviewer :
Check carefully the whole manuscript and correct text and mistakes (e.g. grammar mistakes, Table II should be Table 2, dots are used instead of commas, etc.). “Acid-resistance data for glasses formulated with each set of raw-material components161 are shown in Figure 2.” – figure 2 is not correct here.
Response :
The corresponding paragraphs in the article have been modified.
Reviewer :
Two paragraphs in lines 245-270 contain similar information and should be rewritten. Additionally, discussion on BO an NBO, that is related to the description of the glass structure, is included in lines 346-363. Would it be useful to combine these paragraphs? Finally, why the text related rather to glass structure (l. 346-392) is included in the paragraph “3.2. Corrosion performance analysis”? With this structure
Response :
The discussion about BO and NBO is related to the description of glass structure. This paragraph explains the relationship between glass structure and acid resistance of glass. Therefore, I have made modifications to the title of this section. All changes to the article are highlighted in red.Finally, I sincerely thank you for your valuable comments.
Author Response File: Author Response.pdf
