Continuous Reactive-Roll-to-Roll Growth of Carbon Nanotubes for Fog Water Harvesting Applications

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors The manuscript by Meunier et al. presented a study on the preparation of CNTs using reactive-roll-to-roll (RR2R) method catalyzed by the stainless-steel substrate and the application of CNTs for fog water harvesting. It is nice to see the large-scale production of CNTs with potential practical applicability. I see this work useful and interesting to certain readerships and therefore would like to recommend the publication of this work. Some minor comments for the authors to consider: 1.  What is the size of the production setup? Length and width. Maby it is better to be illustrated in Figure 3. 2.  Can the CNTs be easily detached from the stainless-steel substrate? 3. Be careful with the screenshot in Figure 7. 4. The numbering of the sub-figures in each Figure should be unified.

Author Response

Response to Reviewer 1:

We thank the reviewer for the comments and suggestion presented. Below are our responses to the three comments given by the reviewer:

Comment 1 on size of production setup: 

Answer: The size of the tubular furnace tube is in fact indicated in the manuscript in section 3.1, 2^nd paragraph, line 4:  "… The reactor used was a quartz tube (OD=58 mm; ID=52 mm; L=1.2 m) mounted in the tubular furnace, and the SS wire is suspended and fed along the central axis of the quartz tube. …".  This size can also be inferred from Figure 6 showing a Temperature-Position plot throughout the whole tube reactor with indications of the positions of the hot section of the tube and of the tube ends.

Comment 2 asking if the CNTs can easily be detached from the SS substrate:

Answer: No, the principal objective of the direct-growth technique developed is exactly to prevent an easy detachment of the CNTs from their SS substrate, as discussed in points (1) and (5) of the Introduction; primary bonds with the SS substrate provide this strong anchorage. CNT detachment is problematic for many potential applications.   However, we also indicate in the paper that such strong attachment enables easy manipulation of the CNT-SS structures with easy post-processing possibilities to detach the CNTs if desired. The typical example is for nanofluid generation in which the CNTs are removed from the surface in a second step using ultrasounds in water. The ultrasound simply cuts the CNTs above the SS surface, leaving only the strongly anchored CNT ‘trunks’ on the SS surface.

For more info, see:   N. Hordy, S. Coulombe and J. L. Meunier, "Plasma functionalizationof carbon nanotubes for the synthesis of stable aqueous nanofluids and poly(vinyl alcohol) nanocomposites," Plasma Processes and Polymers, vol. 10, no. 2, pp. 110-118, 2013.     Bottom of Form

Comments 3 & 4 relative to Figure 7 and Figures sub-numbering:

Thank you for the recommendation, we will make sure in the reviewed file that Figure 7 is clear, and that sub-numbering is consistent.

Reviewer 2 Report

Comments and Suggestions for Authors

The CVD growth of carbon nanotubes on stainless steel without introducing catalyst was realized for fog water capturing. In general, the work is inspiring, the procedures are logic, and the conclusions are robust. Before becoming fully acceptable, the manuscript is suggested to be modified according to following comments:

1. As stated repeatedly, the initial fabrication process of SS wire results in different grain sizes and morphologies. Is there any clearer presentation of relevant data revealing the varying trend?

2. The CVD process follows SS wire heat-treatment. Are the two processes combined or separately carried out? In other words, after heat-treatment, is the SS wire quenched or naturally cooled? How does the heat-treatment process proceed in an R2R manner? In fact, from the results, the injection of C2H2 and purging by Ar is also hard to understand for an R2R process;

3. From Figure 6, it can be found that there is a lack of isothermal region in the tubular furnace, how can the CVD process be realized in a precisely controlled manner?

4. As for the CNTs grown on SS, are there any data revealing their size, length, size distribution, water contact angle, or coverage on SS? How is their anchorage measured? 

Comments on the Quality of English Language

A few English expression defects should be corrected, such as "Climate change... have...", "... must be manage", or "the... technique...provide...".

Author Response

We thank the reviewer for the comments and suggestion presented. Below are our responses to the three comments and suggestions given by the reviewer:

Comment 1. As stated repeatedly, the initial fabrication process of SS wire results in different grain sizes and morphologies. Is there any clearer presentation of relevant data revealing the varying trend?

Answer: This question intrinsically relates to metallurgical deformations, solid flow, thermal effects in solids, dislocation movement leading to recrystallization, among others… It is extremely vast and definitely out of scope of the present paper and review process. The main aspects we can say is that we exploit cases where nano-scale grain sizes may be easily generated, examples being metallurgical extrusion (i.e. with solid flow) such as fine wires, dendritic structures such as spherical powder generation, or flat surfaces having a strong defect concentration from a high density of dislocations for example. Acid attack then generates etching around these defects, resulting in the nanoscale structures. For CNT generation, the nanoscale structure needs to be at or below the 50 nm scale size. The "varying trend" discussed by the reviewer is not linked to some wire size for example, but on questioning oneself on what metallurgical transformation(s) will enable nanoscale surface defects on some specific surface.     

Comment 2. The CVD process follows SS wire heat-treatment. Are the two processes combined or separately carried out? In other words, after heat-treatment, is the SS wire quenched or naturally cooled? How does the heat-treatment process proceed in an R2R manner? In fact, from the results, the injection of C2H2 and purging by Ar is also hard to understand for an R2R process;

Answer:  Both are possible and demonstrated in the manuscript. In the so-called "static" cases, a long length of the wire may be rolled on itself and left on a boat inside the furnace, or held along the center-line of the reactor. The wire is then simply brought to recrystallization temperature, then naturally cooled down to room temperature and removed from the furnace.  Such wire is then ready for future use for CNT forest generation.  The other approach in a RR2R pattern is also possible and made/demonstrated in the continuous "dynamic" experiments. This is a case where the temperature profile in the reactor is of primary importance as the recrystallization temperature needs to be attained in the wire along it’s moving path so that the nanoscale defects are generated for CNT growth.

In the RR2R process, the manuscript discusses the questions of the Ar purging step and shows (see Fig. 7b where no argon is being used) that this step used in static mode can be eliminated in R2R mode. The Ar purging is made ONLY upon the start of a R2R sequence in order to eliminate oxygen in the reactor. In other words in RR2R, THERE IS NO gas or temperature change made throughout the synthesis process. Once startred, it is a continuous process with the CNT-covered wire drawn for the length desired.    

Comment 3. From Figure 6, it can be found that there is a lack of isothermal region in the tubular furnace, how can the CVD process be realized in a precisely controlled manner?

Answer:  One beauty of this "direct CNT growth process" now used in static mode on a variety of surfaces for over 15 years, is that it does not require a very precise and isothermal zone in the furnace. Modeling of the wire temperature in the furnace indicated a temperature of 725°C was needed for CNT growth in order to compensate from the heat losses by conduction along the wire. Figure 6 indicates that such temperature is attained and very stable over at least some 20 cm of axial length. This is more than enough to attain a good recrystallization step and CNT growth during passage of the wire. 

Furthermore, Figure 8 shows a series of different CNT growth patterns along the wire depending on growth conditions, or more likely on the wire extrusion process during its fabrication. These patterns are of strong interest in the fog water harvesting application, particularly for the fog water removal step. The control of these patterns may become a key and very interesting issue. We believe the initial wire surface topography is playing THE major role in these patterns.  

Comment 4. As for the CNTs grown on SS, are there any data revealing their size, length, size distribution, water contact angle, or coverage on SS? How is their anchorage measured? 

Answer:  We can refer the reviewer here to a series of papers published on these aspects since the Baddour paper (ref [35] in the Manuscript) published in 2009, and follow-up papers in the last 15 years (see also for example references [34, 40, 41, 42, 50, 53, 55]). Typical MWCNT diameter ranges from 50nm and below, while their lengths are most often in the range from 1-10 μm. As seen in Figs 5, 7, 8, the coverage varies strongly on the SS substrate surface morphology. This is a benefit in the present application as a tuning process of hydrophilic/hydrophobic zones can be made by taking advantage of the various coverage topology.

CNT anchorage measurements are not made here, however manipulations of CNT-covered wires and grids generated by this Direct-Growth method are easily and routinely made in subsequent processing steps, the main example being for nanofluid generation. In fact, the strength of the anchorage is well demonstrated by the nanofluid generation process. In this case, the CNT-covered grids are immersed in water and treated by ultrasonic processing for removal of the CNTs from the grids for dispersion in the water bath. In this process, the CNTs are literally cut above the surface, leaving only CNT “trunks” on the SS surface like a forest that has been harvested. In particular, we do not see sings of detachment, i.e. uprooting, of the CNTs from the surface. This is contrary to a normal nanoparticle-based catalyst growth process which takes out both the nanoparticle and the CNT away from the surface. The nanoscale bumby structure in our process provides the primary bonds to the metallic surface from the diffusion processes in the eutectic transformation generating the CNT growth.   

Comments 5: A few English expression defects should be corrected, such as "Climate change... have...", "... must be manage", or "the... technique...provide...".

Answer:  We thank the reviewer for highlighting the language errors. The updated version of the manuscript has been thoroughly reviewed and a large number of errors have been corrected, as well as some other language modifications made to increase clarity.

 

Reviewer 3 Report

Comments and Suggestions for Authors

The authors show a simple method to fog water harvesting experimentally, but don't go through the underlying mechanisms in this manuscript. The experimental design is reasonable and the results are well presented. However, the introduction part is too lengthy. It is also necessary to provide explanation why the hydrophobic CNT is strongly anchored on the hydrophilic stainless steel surfaces. There are some grammar errors as well.  More, the graphic abstract don't attractively exhibit the novelty of the method. 

Comments on the Quality of English Language

There are some grammar errors and typos, for example, "The process enables very larges surfaces".

Author Response

Response to questions and comments from Reviewer 3:

We thank the reviewer for the comments and suggestion presented. Below are our responses to the comments given by the reviewer:

Comment 1: On the Reviewer’s first sentence indicating the underlying mechanisms are not covered:

Answer:  It is difficult to see which underlying mechanisms are referred to by the reviewer’s comment.  The manuscript covers and brings together three very different topics, namely (i) the growth of CNT with strong primary bonds attachment to the surface and the method of enabling this strong attachment, (ii) the process of enabling a continuous growth for generating very large surfaces for a variety of applications, and (iii) the ability and demonstration for enabling fog water harvesting with the structures generated. Mechanisms are discussed in each three of these topics, as well as a vast literature being referenced on these topics.

It is important to note that linking three very different themes for generating an entirely new process and device requires very extended background which obviously cannot fit in a regular publication. We thus provided what is felt necessary for an understanding of the topics, and extended the literature references for making sure the reader could find additional background on these mechanisms. 

  

Comment 2: on the Introduction being too lengthy:

Answer:  We agree with the reviewer. As indicated in our answer to Comment 1, a paper linking themes that are so different as the "continuous RR2R of CNT" and "fog harvesting" requires to introduce the specificity and some background on both themes, which leads to a long Introduction.

Based on the Reviewer’s comment however, we judge beneficial to reduce the length of the Introduction by removing the 19 lines of text starting with "Benefits of a RR2R generation particularly for a fog harvesting application can be summarized as: …", and ending with "… on particles, flat surfaces, and grids."  This corresponds to the five (5) numbered points stated in the Introduction, and essentially reduces the length of the Introduction by essentially one full page. These five summary points were transferred to the Conclusion of the manuscript, which seems a much better fit than in the Introduction.

Comment 3: Explanation on why the hydrophobic CNT is strongly anchored on the hydrophilic stainless steel surfaces.

Answer:  The explanations for the strong anchoring of the CNT to the SS surface in fact constitute a full section of the manuscript (Section 3 having close to three pages, including Figure 2). References [34, 35] are also providing specific details on this strong CNT-SS interaction based on the generation of primary bonds at the root of the CNTs.

The reviewer seems to confuse two very distinct topics in the paper, (a) the SS-CNT root interaction generating strong CNT-roots attachment bonds (i.e. Section 3), which has nothing to do with (b) the CNT-forest hydrophobic regions for fog droplet capture, and the bare (uncovered) SS surface (hydrophilic) for efficient water recovery (i.e. Figure 8 and Section 5).

Comment 4:  Grammar errors and graphical abstract

Answer: We thank the reviewer for highlighting some language errors. We made a thorough review of the manuscript for eliminating these errors, and in fact also modified a large number of sentences in the text to facilitate the reading.

We totally agree with the comment on the graphical abstract, unfortunately the short timescale for the manuscript review submission will not allow any changes to be made on this figure. In fact the novelty is not on the R2R geometry, but on the nanoscale effects making the process posible, and on the fog harvesting application. This is why we simply show the R2R geometry, the CNT forests on wires, and the harp-based fog harvester.

Round 2

Reviewer 3 Report

Comments and Suggestions for Authors

The authors have replied to all concerns.