Preparation and Characterization of New Electrically Conductive Composites Based on Expanded Graphite with Potential Use as Remote Environmental Detectors

Round 1

Reviewer 1 Report

This paper presents the preparation and characterization of electrically conductive composites with potential applications as remote environmental detectors. As a general comment, I find that this paper is well written and interesting as it complements previous studies by the same authors, using extended graphite as a filler for elastomeric matrix material. However, in the current paper, the authors explore how certain factors, such as filler particle size, film thickness, detector length, temperature, and oil quantity, affect the response rate of the detector. In addition, the authors investigate the adhesion of prepared detector films to different materials that may mimic the materials found for example in a boat in the context of an environmental incident involving oil contamination occurring in the sea.

I would however suggest the text to be revised in the following aspects:

• I do not agree with the title. The authors claim that they have “conductive nanocomposites based on graphene”. Although the word “graphene” and “nanocomposites” is very appealing to be used in a research, one should use it when there is a real proof of its presence. The authors have used expanded graphite (EG) average particle size 5 μm (GFG 5), 50 μm (GFG 50), 200 μm (GFG 200), and 500 μm (GFG 500) used as fillers. EG is a loose structure and porous material consisting of numerous graphite sheets of thickness in nanometers and micrometer in diameter. In the experimental part of this work no surface treatment of EG (e.g. exfoliation) is mentioned to obtain separation of graphene sheets from the expanded graphite. Moreover, no complementary characterization techniques like electron microscopy, x-ray diffraction (XRD) or Raman spectroscopy were used on this work. So, without further proof I find speculative to refer the materials reported as ‘nanocomposites’ or to the filler as ‘graphene’. Therefore, the authors should alter the title and replace the words nanocomposites and graphene by expanded graphite.

• Throughout the manuscript the insertion of automatic references to the figures failed and a sentence “Error! Reference source not found. “appears – the authors should correct this.

• Lines 29 and 30 – a space is missing before the references

• Line 168 – I don’t understand what the authors mean with the sentence “The pale areas around the curves indicate the standard deviations of the measurements.”. There are no pale areas around the curves…are the authors referring to the error bars? Please rephrase.

• Figures 4,5,6,7 and 9 – the decimal places in the numbers should follow the English notation so instead of “,” authors should use “.”. Correct this in all axis of the figures mentioned. Moreover, in all the figures where error bars are shown (Figures 4 to 10) the figure captions should include the number of experiments at which these error bars are referred two. Something like: “data points correspond to mean ± STD of x experiments”

• Figure 7 – The legend inside the graph should be relocated to a different place as it gets in the way of data points’ visualization
  

• Line 239. Authors say there is a good correlation of the I0 data with a fitted exponential decay function, but the R2 of the fitting is 0.6639 which is not indicative of a good fitting. Please comment.

• Line 249 – instead of saying that “temperature of the surroundings can vary in a relatively broad range (from freezing during winter to hot summer climate) and quite often (during day and night).” I would suggest using more objective/restrained language and say instead: “temperature of the surroundings can vary in a relatively broad range (from lower temperatures during winter to higher temperatures during summer) and quite often (during day and night).”

• Line 255 – Authors say there is a “linear decrease of the parameter t1/2”. Although it is visible a decrease, it does not follow a linear trend, so correct.

• Line 265 – I do not understand the sentence: “…and if it could be used to repeating determine the scale of contamination…”. Do the authors mean “…and if it could be used to repeatedly determine the scale of contamination…”?

• Line 268, 275, 331 and x axis of Figure 9, according to the SI should be mL instead of ml.

• Line 291 – I think that is missing a sentence before the sentence in this line. Maybe a sentence saying that results in fig 9 show high values of peel force/width indicative that there is good adhesion between composite layer and the different materials.

Author Response

remote environmental detectors. As a general comment, I find that this paper is well written and interesting as it complements previous studies by the same authors, using extended graphite as a filler for elastomeric matrix material. However, in the current paper, the authors explore how certain factors, such as filler particle size, film thickness, detector length, temperature, and oil quantity, affect the response rate of the detector. In addition, the authors investigate the adhesion of prepared detector films to different materials that may mimic the materials found for example in a boat in the context of an environmental incident involving oil contamination occurring in the sea. 

 We thank the reviewer for the thorough review of our manuscript and we address the individual comments in more detail below. 

I would however suggest the text to be revised in the following aspects: 

• I do not agree with the title. The authors claim that they have “conductive nanocomposites based on graphene”. Although the word “graphene” and “nanocomposites” is very appealing to be used in a research, one should use it when there is a real proof of its presence. The authors have used expanded graphite (EG) average particle size 5 μm (GFG 5), 50 μm (GFG 50), 200 μm (GFG 200), and 500 μm (GFG 500) used as fillers. EG is a loose structure and porous material consisting of numerous graphite sheets of thickness in nanometers and micrometer in diameter. In the experimental part of this work no surface treatment of EG (e.g. exfoliation) is mentioned to obtain separation of graphene sheets from the expanded graphite. Moreover, no complementary characterization techniques like electron microscopy, x-ray diffraction (XRD) or Raman spectroscopy were used on this work. So, without further proof I find speculative to refer the materials reported as ‘nanocomposites’ or to the filler as ‘graphene’. Therefore, the authors should alter the title and replace the words nanocomposites and graphene by expanded graphite. 

After reviewing the manuscript title, we agree that since we did not include techniques to prove the presence of graphene, we propose changing the title to ‘Preparation and Characterisation of New Electrically Conductive Composites Based on Expanded Graphite with Potential Use as Remote Environmental Detectors’ in order to avoid confusion. 

• Throughout the manuscript the insertion of automatic references to the figures failed and a sentence “Error! Reference source not found. “appears – the authors should correct this. 

We thank the reviewer for pointing out this issue and we confirm that this has been resolved in the revised manuscript. 

• Lines 29 and 30 – a space is missing before the references 

This typographical error has been amended in the revised manuscript. 

• Line 168 – I don’t understand what the authors mean with the sentence “The pale areas around the curves indicate the standard deviations of the measurements.”. There are no pale areas around the curves…are the authors referring to the error bars? Please rephrase. 

Yes, the sentence refers to error bars used in the graphs and the figure captions were rephrased to reflect this in the revised manuscript. 

• Figures 4,5,6,7 and 9 – the decimal places in the numbers should follow the English notation so instead of “,” authors should use “.”. Correct this in all axis of the figures mentioned. Moreover, in all the figures where error bars are shown (Figures 4 to 10) the figure captions should include the number of experiments at which these error bars are referred two. Something like: “data points correspond to mean ± STD of x experiments” 

We thank the reviewer for pointing this out and we have amended the aforementioned figures along with their captions.  

• Figure 7 – The legend inside the graph should be relocated to a different place as it gets in the way of data points’ visualization 

The legend was repositioned in the revised manuscript to avoid obscuring data points. 

• Line 239. Authors say there is a good correlation of the I0 data with a fitted exponential decay function, but the R2 of the fitting is 0.6639 which is not indicative of a good fitting. Please comment. 

We appreciate that the presented value of the fitting coefficient is not sufficiently high to consider this a good fit and the text was revised correspondingly in order to avoid misleading the reader. We kept the power law fit line as a visual guide. 

• Line 249 – instead of saying that “temperature of the surroundings can vary in a relatively broad range (from freezing during winter to hot summer climate) and quite often (during day and night).” I would suggest using more objective/restrained language and say instead: “temperature of the surroundings can vary in a relatively broad range (from lower temperatures during winter to higher temperatures during summer) and quite often (during day and night).” 

We thank the reviewer for this suggestion and we have amended the manuscript accordingly. 

• Line 255 – Authors say there is a “linear decrease of the parameter t1/2”. Although it is visible a decrease, it does not follow a linear trend, so correct. 

Reviewing the presented data, we acknowledge that the wording was not reflecting the observed trend and we have changed the phrasing to reflect this.  

• Line 265 – I do not understand the sentence: “…and if it could be used to repeating determine the scale of contamination…”. Do the authors mean “…and if it could be used to repeatedly determine the scale of contamination…”? 

We thank the reviewer for indicating this suggestion and we agree that the suggested wording is more clear for the reader and we have, thus, amended the aforementioned sentence. 

• Line 268, 275, 331 and x axis of Figure 9, according to the SI should be mL instead of ml. 

We appreciate the reviewer’s suggestion to follow the SI notation and we have corrected the units to follow this format. 

• Line 291 – I think that is missing a sentence before the sentence in this line. Maybe a sentence saying that results in fig 9 show high values of peel force/width indicative that there is good adhesion between composite layer and the different materials. 

We thank the reviewer for highlighting this omission and we have added an introductory sentence to this paragraph in the revised manuscript. 

Reviewer 2 Report

This study investigated impacts of filler particle size, film thickness,
detector length, temperature and the amount of oil the film being exposed to on the performance of the electrically conducted films regarding the response rate of the film. The study presents a broad scope of variables but failed to illustrate the significance and novelty of the study comparing to existing studies. The data analysis can be improved starting from better addressing the error bars, especially in the cases where the significance of difference is weak. I also find it problematic to refer to unpublished work. It should be either replaced with other similar published references or included as a part of the current study. I think the request of publication of this manuscript should be reconsidered after addressing the following comments:

1. In the Introduction section, clarify what in the pollutant do the detecting systems respond to. Including more specific details of the previous studies, for example materials involved and limitations, would add some depth to the text and distinguish the study presented here. State the reason for selecting current materials for the film in this study.
2. Line 70, missing a "," after "multimeter".
3. Line 104, indicate the provider of the adhesive. 
4. Explain why t_1/2 and I₀ were chosen to be the criteria and how they relate to the performance of the films.
5. Please clarify if t_1/2 is correlated to response rate. The following comments are based on the assumption that it is. Also provide values from other studies for reference if available. 
6. According to Fig. 5, the average t_1/2 of GFG500 lies in between GFG 5 and 50, so the response rate does not decrease with increasing filler particle size as stated in the text.
7. According to the film thickness data provided in Table 1, it's not very convincing that the thickness of the films are actually different from each other judging from the standard deviation, especially for the thickness measured from 0.028 mm to 0.039 mm. The standard deviations are greater than the average measured film thickness between the films (0.028 mm and 0.030 mm; 0.037 mm and 0.039 mm).
8. In Fig. 6, I don't think film applicator gap is a good representation of the film thickness for the same reason mentioned above.
9. Line 239, a fitting with R² of 0.6639 doesn't seem to be a good correlation to me and the exponential fitting doesn't seem to serve any purpose in the study. Please provide stronger evidences and explain why this information is relevant.
10. The explanation from line 238 to 242 is a bit hard to follow. Is it suggesting that there is correlation between the value of I₀ and how fast it drops regardless of the length?
11. The experimenting temperature range of 5 ºC to 35 ºC does not seem to be wide enough for practice in a lot of water bodies and if the author thinks it is, references should be provided to support that.
12. Which film was selected for the temperature study?
13. Based on the error bar of the data point at 5 ºC, it's telling me that the film is not reliable at this temperature.

Author Response

This study investigated impacts of filler particle size, film thickness, detector length, temperature and the amount of oil the film being exposed to on the performance of the electrically conducted films regarding the response rate of the film. The study presents a broad scope of variables but failed to illustrate the significance and novelty of the study comparing to existing studies. The data analysis can be improved starting from better addressing the error bars, especially in the cases where the significance of difference is weak. I also find it problematic to refer to unpublished work. It should be either replaced with other similar published references or included as a part of the current study. I think the request of publication of this manuscript should be reconsidered after addressing the following comments: 

We thank the reviewer for taking their time for a thorough review of our manuscript and we address their individual comments below. 

1. In the Introduction section, clarify what in the pollutant do the detecting systems respond to. Including more specific details of the previous studies, for example materials involved and limitations, would add some depth to the text and distinguish the study presented here. State the reason for selecting current materials for the film in this study. 

In order to make the points mentioned above more clear to the reader, we have included additional information in the Introduction section of the revised manuscript.  

1. Line 70, missing a "," after "multimeter". 

We thank the reviewer for pointing out this omission and we have addressed this in the revised manuscript. 

1. Line 104, indicate the provider of the adhesive.  

This adhesive was produced in-house and we have added this information to the revised manuscript. 

1. Explain why t1/2 and I0 were chosen to be the criteria and how they relate to the performance of the films. 

The response of the proposed system has to be relatively fast in order to be advantageous compared to existing methods of pollution detection. In this work, we used the time factor t1/2 due to the shape of the obtained electric current data (initial fast decrease followed by a slower change), allowing a simple way of comparing different data sets, and thus, material performance, without the need to wait for extended periods of time for the current to reach zero value. The value of initial current I0 is an operational factor that needs to be optimised with regards to minimising electric current consumption, while retaining sufficiently high value, allowing it to be distinguished from noise. We have expanded this explanation in the revised manuscript to make it more clear for the reader.  

1. Please clarify if t1/2 is correlated to response rate. The following comments are based on the assumption that it is. Also provide values from other studies for reference if available.  

Yes, the aforementioned parameter corresponds to response rate. As the current decreases faster, the value of the parameter decreases, allowing a simpler comparison of different data sets. 

1. According to Fig. 5, the average t1/2 of GFG500 lies in between GFG 5 and 50, so the response rate does not decrease with increasing filler particle size as stated in the text. 

We thank the reviewer for pointing out this discrepancy and we have amended the wording to reflect this in the revised manuscript.  

1. According to the film thickness data provided in Table 1, it's not very convincing that the thickness of the films are actually different from each other judging from the standard deviation, especially for the thickness measured from 0.028 mm to 0.039 mm. The standard deviations are greater than the average measured film thickness between the films (0.028 mm and 0.030 mm; 0.037 mm and 0.039 mm). 

While we appreciate that for some thickness values the differences are comparable with the standard deviation values, there is still a general increase in measured dry film thickness with increasing film applicator gap even with taking the standard deviation values into account. In addition, the same increasing trend observed for parameters t1/2 and I0 supports this correlation.  

1. In Fig. 6, I don't think film applicator gap is a good representation of the film thickness for the same reason mentioned above. 

As mentioned above, the effect of increasing thickness is discernible despite the standard deviation values. The value of applicator gap is a manufacturing variable and we believe it shows noticeable correlation with film thickness. Due to the correlation between the two values, we do not believe that changing the film applicator gap for dry film thickness would provide a better representation of the data and would show similar trend to Figure 6.  

1. Line 239, a fitting with R2 of 0.6639 doesn't seem to be a good correlation to me and the exponential fitting doesn't seem to serve any purpose in the study. Please provide stronger evidences and explain why this information is relevant. 

We appreciate that the presented value of the fitting coefficient is not sufficiently high to consider this a good fit and the text was revised correspondingly in order to avoid misleading the reader. We kept the power law fit line as a visual guide. 

1. The explanation from line 238 to 242 is a bit hard to follow. Is it suggesting that there is correlation between the value of I0 and how fast it drops regardless of the length? 

We have rephrased the aforementioned sentence in the revised manuscript to make it more clear for the reader. The aforementioned paragraph aims to explain the low value of t1/2 for the longest detector, which is caused by the low value of the electric current in contrast to the shorter detector films. 

1. The experimenting temperature range of 5 ºC to 35 ºC does not seem to be wide enough for practice in a lot of water bodies and if the author thinks it is, references should be provided to support that. 

The lower temperature limit was chosen to be close to the freezing point of water, as going below that would mean the detector would not be submerged in liquid water anymore. As the detector was designed with liquid water in mind, we do not believe lowering the temperature further would provide further insight into the detector performance. Reaching the upper temperature limit in Figure 8, there seems to be a diminishing effect of temperature on the t1/2 value, again, limiting the additional benefit of experiments above this temperature. Additionally, the used temperature range is in agreement with a study by Fondriest Environmental Inc., where temperature of the most lakes and rivers is reported to fall between 4 °C and 35 °C (Fondriest Environmental, Inc. “Water Temperature.” Fundamentals of Environmental Measurements. 7 Feb. 2014. Web. https://www.fondriest.com/environmental-measurements/parameters/water-quality/water-temperature/). 

1. Which film was selected for the temperature study? 

The same film as for the particle size determination. The sentence “The samples were prepared using the film applicator with 0.24 mm wet film thickness.” was added in order to clarify this in the revised manuscript. 

1. Based on the error bar of the data point at 5 ºC, it's telling me that the film is not reliable at this temperature.
   

While we appreciate a larger value of standard deviation at the lowest temperature studied here, we do not believe this makes the detector unreliable since it still provides sufficient response in under 15 minutes.

Reviewer 3 Report

The authors presented a kind of electrically conductive composite as an environmental detector based on an elastomeric matrix and graphene from expanded graphite as the filler. The influence of filler particle size, film thickness, detector length, temperature and the amount of oil on detector response rate were discussed. This detector might be useful to the detection of water contamination caused by petroleum and its constituents and derivatives; however, it is difficult to understand the advantages of this detector comparing with other types of detectors from this manuscript. In addition, the results showed that the tested polymer composite was suitable just for detecting the presence of the contamination, but not its scale due to the unsensitive to the different oil amounts. This would be a problem to limit its practical application. Therefore, I do not think this manuscript is suitable to be published at this moment.

Author Response

We thank the reviewer for taking their time to review our manuscript. While we appreciate the reviewer’s concern regarding the practical application of the studied material without the ability to discern the extent of pollution, we believe there is still benefit in cheap and fast detector that could be used to indicate the presence of pollution without the need to utilise more sophisticated, and usually more expensive, techniques, allowing a potential for wider applicability of the studied detector material with possible faster indication of pollution even in remote areas.  

Reviewer 4 Report

The manuscript presenting study on electrically conductive composites based on an elastomeric matrix and expanded graphite as the filler is well presented and written. A potential application of the obtained composite as an environmental remote detector was well studied and presented. I recommend it for publication in present form.

Author Response

We thank the reviewer for taking time to assess our work and their positive review of our manuscript. 

Round 2

Reviewer 2 Report

Most of the comments from last review were properly addressed and the minor errors were corrected. I recommend this study to be published at present form.

Reviewer 3 Report

I have reviewed the revised version. I think this manuscript can be accepted for publication although I still concern the practicality.