CaO-SiO2-B2O3 Glass as a Sealant for Solid Oxide Fuel Cells
Round 1
Reviewer 1 Report
The manuscript deals with study of silica based glass-ceramics for SOFC application.
Due to critical role of sealants in SOFCs, it is worth to study this topic.
However, I have following concerns about the paper
1. There are plenty of glass compositions already available in literature, why authors decided to study these particular glass compositions
2. Glass-ceramic shows low CTE (8.4) as compared with metallic interconnect (12-12.5) and electrolyte used in SOFC conditions. Did authors perform any thermal cycles to confirm the adhesion between glass and interconnect below Tg.
3. The results shown in paper are not sufficient to conclude about the quality of a glass to be used as SOFC sealant. Authors should perform long term resistivity analysis , preferably in SOFC conditions (reducing and oxidizing atm, high temperature and voltage).
4. Authors should discuss all the results in details and make a comparison with state of art literature
5. Figure 7: there should be a details discussion about different types of phases formed, their CTEs, and amount of residual glassy phase.
6. Figure 7: Point 2 seams to be SiO2 or cristobalite phase. Authors should address the relative issues related to cristobalite.
7. To make a clear comparison with Schott glass, it is important to show the composition of Schott glass. It is impossible to compare two glass systems without knowing their compositions.
Author Response
Thank you very much for your comments and recommendations, we have attempted to meet them the best we could. We have also proofread the manuscript to improve its language.
Below you may find detailed responses to each of your remarks, also all changes in the manuscript are marked with Microsoft Word editing tool.
Q1: There are plenty of glass compositions already available in literature, why authors decided to study these particular glass compositions
A1: Thank you for this remark. Main reason for our search of new composition was formulated in the Introduction (see fragment below):
“The dwell at elevated temperature may cause undesirable changes in structure of electrodes due to sintering processes and may lower power output of the cell.
It is thus desirable to use glass or glass-ceramic sealant that quickly reacts with material of a bipolar plate forming strong interface such that it remains stable under operating conditions.”
Therefore, we looked for a glass composition that would form strong interface with Crofer 22 APU after only a brief heat-treatment, so that no electrodes overheating and degradation takes place. We have chosen CaO-SiO2-B2O3 system because of the reported literature data (see Introduction section) reporting CTEs of 8-10∙10-6 1/K, and suitable softening point and because of the general idea that high boron oxide content will lead to higher chemical activity and consequently to faster formation of strong interface. As stated in the Introduction we are aware of the problems that high boron content can cause, and we hoped to bind most of the boron oxide in the crystalline phase and thus to lower its mobility.
If you find this explanation insufficient, please let us know how we can modify to make it more clear for the readers.
Q2: Glass-ceramic shows low CTE (8.4) as compared with metallic interconnect (12-12.5) and electrolyte used in SOFC conditions. Did authors perform any thermal cycles to confirm the adhesion between glass and interconnect below Tg
A2: Thank you for this remarks. While the paper was under review, we have finally received our data on the thermal cycling. We did not wait to include this data because we were not sure of obtaining the data in reasonable time. Now, however, we have obtained the necessary data. We have added a small fragment in the end of Methods and Materials section and a paragraph describing the results immediately after Figure 6.
Q3: The results shown in paper are not sufficient to conclude about the quality of a glass to be used as SOFC sealant. Authors should perform long term resistivity analysis , preferably in SOFC conditions (reducing and oxidizing atm, high temperature and voltage).
A3: Thank you for your remark. We agree that long term analysis of the properties of the studied sealant is necessary to make a conclusion on suitability of the sealant for application in SOFCs. We have stated so explicitly in Conclusion section, see citation below:
“We would like to note, however, that 100 h experiment gives only a preliminary information about stability of the sealant-steel interface and that of sealant itself. Further experiments with longer exposures should be carried out. Study of the sealants stability in humid environment at operation temperature will be of particular interest. Furhtermore, experiments studying long-term stability of SOFCs sealed with the proposed sealants should be conducted. These experiments will allow assessing the effect of high boron oxide content on electrode poisoning and studying stability of dielectric and thermomechanical properties of the sealant.”
Unfortunately, such experiments require access to certain experimental facilities in our institution that are currently heavily loaded, and we will not be able to perform such tests in the next half-year. We planned to prepare a number of other compositions of the same system and to test their adhesion and thermomechanical properties, and then start long-term tests on all samples. Presently we can only limit the manuscript to what data we have.
Q4: Authors should discuss all the results in details and make a comparison with state of art literature
A4: We have added more discussion and some comparison to popular barium and calcium aluminosilicate sealants in the end of the Results and Discussions section.
Q5: Figure 7: there should be a details discussion about different types of phases formed, their CTEs, and amount of residual glassy phase.
A5: We have added a large text fragment describing phase composition in more detail. The fragment contains also information on CTEs of the crystalline phases present and mentions detrimental role of crystallized SiO2 in the sealant. The fragment is located between Figures 4 and 5. We believe it is the best place to discuss the matter.
Q6: Figure 7: Point 2 seams to be SiO2 or cristobalite phase. Authors should address the relative issues related to cristobalite.
A6: Please see “A5” above. The discussion on cristobalite role is included in the mentioned text fragment.
Q7: To make a clear comparison with Schott glass, it is important to show the composition of Schott glass. It is impossible to compare two glass systems without knowing their compositions.
A7: Thank you for your comment. We believe that there is no actual need for the composition of Schott glass to be presented in the paper for two following reasons:
- We think that disclosure of the Schott 394 composition may violate certain regulations on the use of confidential information on composition of commercial materials.
- Furthermore, Schott 394 glass is supposed to work as a sort of “background” in the paper to show how our sealant works when compared to widely used commercial sealant. We had no plans on comparing composition of our sealant to that of Schott 394 glass. We rather explored idea of high-boron sealants by studying their adhesion and thermomechanical compatibility with steel interconnects.
Reviewer 2 Report
1. Better to show the real results of sealant on plate after heat treatment instead of using schematic depiction.
2. More experiment should be presented, like the stability under H2 and air, adhesion properties, and mechanical properties. In this way, glass sealing materials can be well evaluated.
Author Response
Dear Reviewer,
Below you may find detailed responses to each of your remarks, also all changes in the manuscript are marked with Microsoft Word editing tool.
Q1: Better to show the real results of sealant on plate after heat treatment instead of using schematic depiction.
A1: Thank you for your remark. We actually thought of adding photographs of the sealant-interconnect assemblies after heating to different temperatures. Unfortunately, we found that the photographs themselves provided little information on behavior of the sealants because of the limited quality of a photograph made on small samples. Some fine details, like texture of the sealant surface or thickness of the colored layer could not be adequately presented in the photographs we managed to prepare. Therefore, we decided that schemes with realistic proportions would be more illustrative to the evolution of the assembly.
Q2: More experiment should be presented, like the stability under H2 and air, adhesion properties, and mechanical properties. In this way, glass sealing materials can be well evaluated.
A2: You are right, for the material to be a suitable sealing agent for SOFCs, not only it needs to have acceptable thermomechanical and electrical properties along with good adhesion to interconnects and electrolytes, but also its properties and composition should be stable under operation condition. We understand it and have stated it clearly the Conclusion section (see fragment below).
“We would like to note, however, that 100 h experiment gives only a preliminary information about stability of the sealant-steel interface and that of sealant itself. Further experiments with longer exposures should be carried out. Study of the sealants stability in humid environment at operation temperature will be of particular interest. Furthermore, experiments studying long-term stability of SOFCs sealed with the proposed sealants should be conducted. These experiments will allow assessing the effect of high boron oxide content on electrode poisoning and studying stability of dielectric and thermomechanical properties of the sealant.”
If we understood you correctly, you also proposed adding some information on adhesion and mechanical properties of the sealant. We agree that such results are interesting indeed, but during preparation of the manuscripts we found that the reported results, although incomplete, seem wide enough to give idea of the sealant we studied, its strong sides, limitations, and its possible problems. We plan to conduct additional experiments, including long-term stability tests, on the studied sealant and similar compositions in about half a year. We also believe that it would be better to group these new results in a new work, rather than trying to crumple it in the present one.
Reviewer 3 Report
The authors claimed that “CaO-SiO2-B2O3 glasses are not presently used as sealants for SOFCs, ..” but there are several studies with similar subject:
- SiO2- CaO-B2O3 -Al2O3 glass ceramics as sealant for planar intermediate temperature solid oxide fuel cells, Journal of Inorganic Materials 19(1):37-42 (2004), Ruoyun Zheng et
- Evaluation and Application of a Novel BaO–CaO–SiO2–CoO–B2O3 Based Glass-Ceramic Sealing Material for Solid Oxide Fuel Cells, Journal of Electrochemical Energy Conversion and Storage 14(4) (2017) Zhaonan Li et al.
- Evaluation on Properties of CaO–BaO–B2O3–Al2O3–SiO2 Glass–Ceramic Sealants for Intermediate Temperature Solid Oxide Fuel Cells Fatemeh Heydari et al. Journal of Materials Science & Technology Volume 29, Issue 1, January 2013, Pages 49-54
-Evaluation and Application of a Novel BaO–CaO–SiO2–CoO–B2O3 Based Glass-Ceramic Sealing Material for Solid Oxide Fuel Cells Zhaonan LiJ. Electrochem. En. Conv. Stor. Nov 2017, 14(4): 041006
The authors have to clarify the differences, novelty and the advances compared to these one.
Moreover:
- The measurements tehniques are not clear: -Particle size distribution from Figure 1, the XRD patterns of (Figure 4) and dilatometric curve (Figure 5)
- The meaning of the white and blue areas from the discussion has to be explained
- For a clear assignment the XRD patterns of (Figure 4) must include the CaB2O4, SiO2 (Cristobalite) and CaSiO3 PDF-files or at least indicate the main peaks assignment.
Author Response
Dear reviewer,
Thank you very much for your comments and recommendations, we have attempted to meet them the best we could. We have also proofread the manuscript to improve its language.
Below you may find detailed responses to each of your remarks, also all changes in the manuscript are marked with Microsoft Word editing tool.
Q1: The authors claimed that “CaO-SiO2-B2O3 glasses are not presently used as sealants for SOFCs, ..” but there are several studies with similar subject:
- SiO2- CaO-B2O3 -Al2O3 glass ceramics as sealant for planar intermediate temperature solid oxide fuel cells, Journal of Inorganic Materials 19(1):37-42 (2004), Ruoyun Zheng et
- Evaluation and Application of a Novel BaO–CaO–SiO2–CoO–B2O3 Based Glass-Ceramic Sealing Material for Solid Oxide Fuel Cells, Journal of Electrochemical Energy Conversion and Storage 14(4) (2017) Zhaonan Li et al.
- Evaluation on Properties of CaO–BaO–B2O3–Al2O3–SiO2 Glass–Ceramic Sealants for Intermediate Temperature Solid Oxide Fuel Cells Fatemeh Heydari et al. Journal of Materials Science & Technology Volume 29, Issue 1, January 2013, Pages 49-54
-Evaluation and Application of a Novel BaO–CaO–SiO2–CoO–B2O3 Based Glass-Ceramic Sealing Material for Solid Oxide Fuel Cells Zhaonan Li, et al. J. Electrochem. En. Conv. Stor. Nov 2017, 14(4): 041006
A1: We have studied the articles you have listed in your comment. The first item has a most promising title, but we could not find page of the article in English (we used Google Scholar search engine), we only found Chinese version of the article, which none of the authors could read, unfortunately. The second item in the list describes glass where boron oxide content is about 7 wt.% (the paper gives H3BO3 content, we recalculated it to B2O3). The same goes for the item #3 in the list, there theoretical boron oxide content is stated explicitly to be 7.3 wt.%. The last item in the list is a copy of the second one. Our material contains 26 mol.% B2O3, which corresponds to roughly 32 wt.% in our composition, so the material studied significantly differs from the ones studied in the listed articles.
We have added a clarification stating: “CaO-SiO2-B2O3 glasses with boron oxide content above 7-8 wt. % are rarely used” near the end of Introduction section to underline that our material differs from the frequently reported compositions by high boron oxide content.
Q2: The measurements tehniques are not clear: -Particle size distribution from Figure 1, the XRD patterns of (Figure 4) and dilatometric curve (Figure 5)
A2: Thank you for noting. Our description of these experimental procedures was truly poor. We added some information in the Materials and Methods section, providing additional details on the measurement procedures. We hope that you find these changes adequate. Below the introduced changes are cited for the ease of tracking
“For the purpose of the measurement small quantity of ground powder was dispersedin bidistilled water, ultrasonicated to break any agglomerates and subjected to static light scattering measurement” – on particle size analysis.
“XRD analysis was performed on finely ground powder with copper X-ray tube in Bragg-Brentano reflective geometry. Peaks were identified with the use of PDF database” – on XRD.
“We also cut rectangular samples with approximate dimensions 3x1x1 mm from the heat-treated sealant and measured its CTE in air with dilatometer L75Vertical (LINSEIS, Germany).The samples were measured with heating and cooling rates of 3 °C/min and were held at 850 °C for 2 h. Ends of the samples were covered with thin zirconia pellets to minimize reaction of the sealant with the measurement chamber. Expansion of zirconia pellets was accounted for and subtracted from the final data.”
Q3: The meaning of the white and blue areas from the discussion has to be explained
A3: Thank you. We have added a few lines, explaining what white and blue areas are. You may find the added lines shortly before Figure 3.
Q4: For a clear assignment the XRD patterns of (Figure 4) must include the CaB2O4, SiO2 (Cristobalite) and CaSiO3 PDF-files or at least indicate the main peaks assignment.
A4: We have included numbers of PDF database entries used for peaks assignment in the text after Figure 4. We also indicated the most intensive peaks corresponding to each phase in Figure 4
Reviewer 4 Report
Dear Sir,
The paper is interesting and well prepared however it can be improved in some areas as follows:
The title
The full name of SOFCs should be added instead of its abbreviation..
In the abstract
- It is recommended to begin this part with general sentence about the main topic (SOFCs) to grasp the reader’s attention.
- Please state exactly the performed analysis regarding the characterization such as XRD, Dialtometric analysis etc.
- keywords should be arranged alphabetically and number of keywords can be increased to 5 words.
In other sections:
- At the end of the section of introduction the main aim should be summarized and focused to avoid any reader’s confusing thus compress it from line 82 to 96into bullets or combine it in any previous paragraphs.
- Schematic diagram for the experimental part could be useful for the reader
- All chemicals being used in the study should be addressed in materials’ section including where they are purchased or obtained.
- State the model, the company and country of origin of any instrument used in the research.
- Use right abbreviation for min instead of min line 104 and for hour too.
- Could the author give more information about the particle size distribution of the prepared samples?
- Provide more information about the mechanical test which preformed to ensure the good adhesion of the samples.
- Most of figures are not in a good quality and should be changed.
n Kindly, FORMAT the references correctly according to the author’s guide such as line 380 correct the subscript number of the chemical formula
Regard
Author Response
Dear reviewer,
Thank you very much for your comments and recommendations, we have attempted to meet them the best we could. We have also proofread the manuscript to improve its language.
Below you may find detailed responses to each of your remarks, also all changes in the manuscript are marked with Microsoft Word editing tool.
Q1: The full name of SOFCs should be added instead of its abbreviation.
A1: We have changed the title in accordance with your recommendation.
Q2. It is recommended to begin this part with general sentence about the main topic (SOFCs) to grasp the reader’s attention.
A2: We have modified the abstract accordingly
Q3. Please state exactly the performed analysis regarding the characterization such as XRD, Dialtometric analysis etc.
A3. We have added the necessary sentence to the Abstract.
Q4. keywords should be arranged alphabetically and number of keywords can be increased to 5 words.
A4. We have changed the keywords list accrodingly
Q5. At the end of the section of introduction the main aim should be summarized and focused to avoid any reader’s confusing thus compress it from line 82 to 96into bullets or combine it in any previous paragraphs.
A5. We have changed the end of the Introduction section and put main aims of the study into a list for the convenience of the readers. Thank you for the advice.
Q6. Schematic diagram for the experimental part could be useful for the reader
A6. Such a diagram would be useful, but it seems to us that it may be excessive in the paper already containing quite a number of figures, especially because we did not employ revolutionary methods or approaches requiring additional clarification via a figure.
Q7. All chemicals being used in the study should be addressed in materials’ section including where they are purchased or obtained.
A7. We have added information on the suppliers of the chemicals used in the study
Q8. State the model, the company and country of origin of any instrument used in the research.
A8. We have added information on all the equipment used in the research. Thank you for your remark we have really missed it out.
Q9. Use right abbreviation for min instead of min line 104 and for hour too.
A9. We have corrected the abbreviations, thank you.
Q10. Could the author give more information about the particle size distribution of the prepared samples?
A10. We have added some information on particle size distributions on the Figure 1 and in the text right before it. We also added some measurement details in the Materials and Methods section.
Q11. Provide more information about the mechanical test which preformed to ensure the good adhesion of the samples.
A11. Unfortunately, we cannot provide any additional details on the mechanical test, since it was done “by hand” with only visual observation used. We understand that measurements with registration of the applied load would provide more information, but we had no equipment necessary for such tests.
Q12. Most of figures are not in a good quality and should be changed.
A12. We have improved the quality of the figures.
Q13. Format the references correctly according to the author’s guide such as line 380 correct the subscript number of the chemical formula
A13. Thank you. We have corrected the mistakes in formatting of the references.
Round 2
Reviewer 1 Report
The authors have improved the paper significantly
there can be accepted in president form
Author Response
Thank you!
Reviewer 2 Report
The authors addressed the questions.
Author Response
Thank you!
Reviewer 3 Report
The authors provided proper answers.
Author Response
Thank you!
Reviewer 4 Report
Dear Author,
Thank you for the revised version however, some comments were not addressed for example the schematic diagram can be added as a supplementary materials and providing more information about the mechanical test.
Author Response
Thank you for your comments!
We have tried to address adequately the last of them. Please find our responses below.
- We totally forgot about an option of adding images as supplementary materials. We prepared a scheme of the sequence of the experiments conducted in our study and attached it as a supplementary figure S1. We also added a short paragraph at the end of Materials and Methods section, mentioning where the figure with detailed scheme of the experiment can be found. We believe that you can find the supplementary figure file attached to the latest version of the manuscript
- We measured the loads applied during mechanical tests of the adhesion strength and calculated flexural stresses generated. We have moved short description of the experimental procedure to the Materials and Methods sections. At the same time, we added more information on the experimental results, including some numerical values, obtained in this test. The resulting paragraph is located right after Figure 3.
