Preparation of Al2O3 Multichannel Cylindrical-Tube-Type Microfiltration Membrane with Surface Modification
Round 1
Reviewer 1 Report
Comments to the Authors,
This manuscript has reported a multichannel cylindrical tube-type support with a microfiltration (MF) coating using alumina powder of various particle sizes with minimum linear shrinkage, and the effectiveness of SiO2 modification on pristine alumina membranes was demonstrated. I gone through this manuscript and found it to be some interesting. However it requires to be improved with major revision and main remarks are following:
1. What is the basis for selecting the size of S/L-Al2O3 for 4.8/15.66 μm to prepare the specimens? Have studies shown that this particle size and the collocation of them is conducive to the preparation and performance improvement of specimens?
2. In the Materials and Methods, you mention that “specimens were calcined at 400 °C for 4 h to burn off the binder”. However, due to the porous structure of the specimens, does the volatilization of PVA after 400 °C leave residues in the specimens? Will this affect the filtration of water later?
3. The filtration effective of the specimens prepared in this paper is good or not compared to the filters currently used in commercial or in other studies? Can the authors list relevant data in the article?
Comments for author File: 
Comments.pdf
Author Response
Answer to Comments from Reviewer#1:
Respectfully, I would like to thank the reviewer to spare sometime to review this article.
This manuscript has reported a multichannel cylindrical tube-type support with a microfiltration (MF) coating using alumina powder of various particle sizes with minimum linear shrinkage, and the effectiveness of SiO2 modification on pristine alumina membranes was demonstrated. I gone through this manuscript and found it to be some interesting. However, it requires to be improved with major revision and main remarks are following:
- What is the basis for selecting the size of S/L-Al2O3 for 4.8/15.66 μm to prepare the specimens? Have studies shown that this particle size and the collocation of them is conducive to the preparation and performance improvement of specimens?
Answer: I would like to thank to the reviewer for raising this concern. A literature survey was conducted prior to the experimental study. It was found out that alumina particles with particle size up to 1 μm leads to very high linear shrinkage rate (>10%). Based on this investigation AM210 powder was selected for our study as it shows shrinkage rate of <2%. Moreover, as mentioned in the manuscript, AM210 powder particles are supposed to fill the pores among large sized particles in order to inhibit the shrinkage. It means the gap created by the large sized particles must be at least 5 μm. To achieve this requirement, alumina powder of average particle size of 15.66 μm was selected as it is expected to provide sufficient inter-particle pore size for the deposition of AM210 particles.
The comparison with previous studies as given in Table 2, justifies the use of AM210 powder for the fabrication of alumina support layer.
- In the Materials and Methods, you mention that “specimens were calcined at 400 °C for 4 h to burn off the binder”. However, due to the porous structure of the specimens, does the volatilization of PVA after 400 °C leave residues in the specimens? Will this affect the filtration of water later?
Answer: I would like to thank the reviewer for this comment. The specimens are calcined at 400 °C for the time period of 4 h to allow the binder to decompose completely. Moreover, the binder burns off in the form of gas so it could not possibly block the porous structure. Moreover, after calcination, the specimens are sintered at 1500 °C, high enough temperature to burn off any residual binder, if remaining.
- The filtration effective of the specimens prepared in this paper is good or not compared to the filters currently used in commercial or in other studies? Can the authors list relevant data in the article?
Answer: I would like to thank the reviewer for highlighting this concern. One of the main targets of this study was to prepare the water-filtration membrane with multi-channel tubular configuration for commercial applications. In order to successfully fabricate desired filtration membrane, an approach of mixing alumina powders of different particle sizes was adapted to reduce the linear shrinkage rate. It is expected that higher the shrinkage rate, higher would be the probability of crack generation in commercial-sized specimen. Table 2 in the manuscript, shows the comparison of support layers fabricated with different size of alumina particles.
It is apparent that fabrication of support with single-sized alumina particles sintered at relatively low sintering temperature compared to this work, leads to very high shrinkage rate.
Author Response File: 
Author Response.docx
Reviewer 2 Report
In this work, the authors prepared a multi-channel cylindrical ceramic support using different-sized Al2O3 particles and then coated a microfiltration (MF) layer using multiple dip-coating method. Also the membrane surface was modified using SiO2 sol-gel technique to enhance the antifouling property. Some statements require further characterizations to support. Therefore, a major revision is required before it is recommended for publication in the journal.
1. Is there any relationship between pores and shrinkage for the ceramic tube? The variation of the open porosity and pore size with increasing L-Al2O3 amount should be given.
2. SEM image of an alumina support layer without L-Al2O3 should be provided in Figure 1b.
3. It is difficult to identify α-alumina phase from SEM image, XRD pattern should be added.
4. It was confused why there are four similar SEM pictures in Figure 3 without one magnified one.
5. There is an obvious boundary in the cross-sectional SEM image of Figure 4, how about the adhesion between the coated support layers and the Al2O3 support?
6. The membrane fouling test method (line 235-247) should be moved to experimental section. And what about the fouling is the modified membrane for other foulants?
Author Response
Answer to Comments from Reviewer#2:
Respectfully, I would like to thank the reviewer to spare sometime to review this article.
In this work, the authors prepared a multi-channel cylindrical ceramic support using different-sized Al2O3 particles and then coated a microfiltration (MF) layer using multiple dip-coating method. Also the membrane surface was modified using SiO2 sol-gel technique to enhance the antifouling property. Some statements require further characterizations to support. Therefore, a major revision is required before it is recommended for publication in the journal.
- Is there any relationship between pores and shrinkage for the ceramic tube? The variation of the open porosity and pore size with increasing L-Al2O3 amount should be given.
Answer: I would like to thank the reviewer for this suggestion. In this study, it was expected that addition of large-sized alumina particles would generate an inter-particle gap to accommodate the small-sized particles. As a result, the shrinkage rate would be controlled while the pore size would be unaffected. This can be supported by the mercury porosimetry results for alumina support of different L/S Al2O3 particle concentration. It can be observed that average pore size of alumina support layer, regardless of L/S Al2O3 particle concentration is 1 μm. However, the pore size distribution gets broader as the concentration of L-Al2O3 particles increases. It shows that S-Al2O3 particles were successfully accommodated among the inter-particle space created by L-Al2O3 particles thus maintaining the average pore size.
It can be said that there is no such relationship between pore size and shrinkage rate for alumina support layer. Due to this reason these results were not included in the manuscript.
- SEM image of an alumina support layoer without L-Al2O3 should be provided in Figure 1b.
Answer: I would like to thank the reviewer for this suggestion. Required changes has been made in the manuscript.
- It is difficult to identify α-alumina phase from SEM image, XRD pattern should be added.
Answer: I would like to thank the reviewer for this comment. Basically, S/L-Al2O3 powder (4.8/15.66 μm) used for the fabrication of support layer has intrinsically α-alumina phase. In order to clarify the α-alumina phase formation, XRD has been given below.
- It was confused why there are four similar SEM pictures in Figure 3 without one magnified one.
Answer: I would like to thank the reviewer for highlighting this point. One of the main objectives of this work was to successfully fabricate multichannel cylindrical tube-type microfiltration membrane. So in order to determine the uniformity of membrane across the cross-section of tube-type support, SEM images were taken in this manner for one of the channels as highlighted in the given figure.
The respective figure has been changed in the manuscript as well.
5. There is an obvious boundary in the cross-sectional SEM image of Figure 4, how about the adhesion between the coated support layers and the Al2O3 support?
Answer: I would like to thank the reviewer for this comment. The adhesion between the coated support layers and the Al2O3 support was determined by conducting the pure water permeability test. The stable permeate curves over the period of time for single and double coted layer refers to the strong adhesion between the membrane layer and support. If there would have been any defects, the curves would not be stable.
- The membrane fouling test method (line 235-247) should be moved to experimental section. And what about the fouling is the modified membrane for other foulants?
Answer: I would like to thank the reviewer for this suggestion. The required changes have been made in the manuscript.
The main reason for only using humic acid as a foulant was just to determine the effect of surface modification on the microfiltration membrane layer. It opens the door for surface modification to be applied at commercial-scale for better performance. Moreover, in our previous studies, in addition to humic acid, BSA (Bovine Serum Albumin) solution was also used as a model foulant for coated disk-type supports. (Reference [21] in the manuscript)
(J. Lee, J. H. Ha, I. H. Song, and J. W. Park, “Effect of SiO2 coating on alumina microfiltration membranes on flux performance in membrane fouling process,” J. Ceram. Soc. Japan, pp. 35–43, 2019, doi: 10.2109/jcersj2.18124.)
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
Can be published now though it can still be improved.
Reviewer 2 Report
The manuscipt can now be accepted for publicaiton
