Plasma Polymerized Organosilicon Thin Films for Volatile Organic Compound (VOC) Detection

Round 1

Reviewer 1 Report

Sound and important work, well written.

I only indicate some minor revisions:

Check the latest literature (2021-2023), there has been important work on the plasma polymerization of HMDSO at soft and harsh conditions (over a broad W/F range) – well agreeing with your work.

As you are discussing, the refractive index contains information about the hydrocarbon content and the porosity of the films (free volume portions), increasing with CH content and decreasing with porosity (please add literature). Since the overall increase of refractive index with W/F well follows mass density, it might mainly indicate densification, e.g. instead of –CH3 terminal groups, crosslinking –CH2– groups are introduced.

Some work has been done using neutron reflectometry to study swelling of PP-HMDSO – you might check the literature.

Line 92 (and 96): "high amount of SiOC3 termination" – Please explainwhat you mean by SiOC3 termination (Si with three C and one O bond)?

Author Response

Thank you for your for your review.

Here are our answers to your questions/comments:

1. Several recent references have been added:

de Freitas, A.; Maciel, C.; Rodrigues, J.; Ribeiro, R.; Delgado-Silva, A.; Rangel, E., Vacuum, 2021, 194, 110556 - DOI 10.1016/j.vacuum.2021.110556

Avramov, I.; Radeva, E.; Lazarov, Y.; Grakov, T.; Vergov, L., Coatings 2021, 11, 1193

2. Concerning neutron reflectometry used to follow the swelling of PP-HMDSO films, the following references have been added:

Zhou, Y.; Josey, B. ; Anim-Danso, E.; Maranville, B.; Karapetrova, J.; Jiang, Z.; Zhou, Q.; Dhinojwala, A.; Foster, M. D., In Situ Nanoscale Characterization of Water Penetration through Plasma Polymerized Coatings, Langmuir, 2018, 34, 9634–9644. (https://doi.org/10.1021/acs.langmuir.8b01646)

Blanchard, N. E.; Naik, V. V.; Geue, T.; Kahle, O.; Hegemann, D.; Heuberger, M., Response of Plasma-Polymerized Hexamethyldisiloxane Films to Aqueous Environments, Langmuir, 2015, 31, 12944-12953

Reviewer 2 Report

MDPI Plasma states on its web page: "covering all aspects of plasma science." The manuscript does not cover any aspect of plasma science. I don't understand why the authors submitted the manuscript to Plasma. Although the film was deposited by plasma-enhanced CVD (PECVD), the experiment is not described in the text. The manuscript is focused on the performance of the sensors prepared by PECVD and hardly on how the plasma conditions and process reflect in the film structure.

In the Introduction, the authors describe previously obtained results on the film structure (ref. [27]). This information is essential to understand the current results because the film analyses described in the manuscript are insufficient and complete to understand the film structure that influences the sensor performance. Therefore, the authors should move the information to Discussion and clearly state what means soft/hard plasma conditions, i.e., what W/F is the border.

EDX is unsuitable for the elemental analysis of thin films (films with a thickness of about 300 nm) composed of light elements because it has considerable information depth and low sensitivity. The information about the substrate should be added, and the authors have to specify how they evaluated the EDX signals and subtracted the signal from the substrate that I suppose was silicon, i.e., one of the elements contained in the films.

The results on the sensor performances provide a detailed inside into the sensor functions and represent a valuable asset to the research on VOC sensors.

I have additional minor comments:

The SI units should be used in the text, e.g., in Table 2.

The quantities d and d0 should also be explained in the caption of Figure 4.

I suggest checking the manuscript with Grammarly, namely for using commas.

Author Response

Thank you for your review.

Here are our answers to your questions/comments:

1. As proposed, a short description of the PECVD process has been added to the manuscript: pareagraph 2.1.

2. Concerning EDX analysis, the following sentence has been added: All the thicknesses of the deposited PP-HMDSO films were compatible with the detection characterization volume of the EDX, i.e., much over ~500 nm in depth from the surface.

3. The figure caption 5 (previously 4) has been completed: d0 is the initial thickness of the film under vacuum and d is its thickness under P/P0 of toluene of ethanol.

4. English language has been revised, 

Reviewer 3 Report

Plasma polymerized organosilicon thin films for Volatile Organic Compounds (VOCs) detection 

1.       Bruker D8 diffractometer is an X-ray diffractometer (XRD). XRD is a common abbreviation for analyzing crystal structures among all research communities. If the author used XRD to measure the reflection, then some explanations such as Bragg Brentano measurements in reflection mode or Basic GID (grazing incidence) mode, and operational parameters may need to be provided in the context to avoid confusion.

2.       Please provide the raw data of XRD reflection, i.e. the intensity vs 2θ.

3.       Please add references for all uncited equations. Also please add equation numbers in each equation.

4.       What is “COV” for?

5.       Please explicitly specified “W” and “F” in the “W/F” notation in line 118. I understand this is a continuous study but authors have the responsibility to provide clear information to readers. Another question is why the selection of the W/F ratio is only restricted to 5, 10, and 20 in this study. I notice that the previous study has a ratio of 2.5-140 for various levels of power.

6.       I’m so curious about the accuracy of electron density ρ_e in both Eq. (1) and (2), can the authors discuss its possible errors because this term is so critical in this study?

7.       Clausius-Mosotti equation (5) need more discussions. I guess the dielectric constant ε is measured from ellipsometry and the polarizability α can be calculated. But then how many k terms are in equation (5)? (C, O, and Si?). If this is wrong, then α is estimated to calculate ε, then how is the fitting to the experimental measured refractive index? From this point, I strongly suggest that authors may need to provide more information about the measurement and calculations from the ellipsometry. Otherwise, readers have no clue about how and why authors interpret their results for the change in the refractive index of films.

8.       In equation (2), is Ci the atomic percentage in EDX? Please clarify in the context.

9.       For clarification, “Plasma power” should be the “power by power supply” (ECR antennas), right? There is no way to directly measure the power of bulk plasma.

10.   Low flow of HMDSO and high power supply would lead to more ionization, dissociation, or decomposition of molecules, which would be presented in the plasma field. In this respect, it seems that the absorption of heptane and ethanol is less influenced than toluene by the preset W/F ratio. Could the plasma play any role in the different absorption rates?

11.   Authors suggest that the bulk absorption of toluene (C6H5CH3) into the films is greater than the absorption of heptane (H3C(CH2)5CH3), and heptane is mostly absorbed on the surface due to their different molecular sizes. But ethanol (CH3−CH2−OH) is linear and much smaller, its absorption should be higher than that of toluene. What is the reason for the swelling of ethanol is smaller in Fig. 4? Since toluene has a ring structure, would that chemical structure affect absorption by chemical reactions?

12.   Authors suggested that the introduction of carbon into Si-O-Si chains can create new bonds such as Si-O-C and Si-CH2-Si. The density of film and polarizability would therefore enhance and so does the refractive index. However, the density of films is almost identical for 10 and 20 sccm but the refractive index still increases as shown in Fig. 1. It is very difficult to understand why polarizability is also enhanced by incorporating carbon atoms. I think authors should at least demonstrate using equation (5) to explain the variation of polarizability based on the experimental measure.

13.   The comparison of the adsorption of water vapor and toluene on the WF5 deposits in Fig. 5 seems not directly relevant to the effect of humidity because neither toluene dissolves in water nor water vapor could direct impact plasma polymerized SiOCH films. So it may need an explanation for this comparison and how this measure relates to the humidity in the air.

This extended study from an earlier report is supposed to have many in-depth findings. A quite special experiment is the in situ ellipsometry coupled to sorption measurements. But I think the authors skip too many details to confuse readers. The conclusion sounds rational but without quantitative support, which by and large should be in the authors’ hands for further calculations. I strongly recommend some decent quantitative analyses being added with illustrative figures, particularly using equations (1) – (5) and ellipsometry. It should be a major revision to the current manuscript, not a quick patch.

Please double-check if there are any typos or grammatical errors.

Author Response

Thank you for your for your review.

Here are our answers to your questions/comments:

1. The paragraph starting at line 134, i.e., "“X-ray reflectometry analysis (XRR) were carried out using a Bruker D8 diffractometer with Cu Kα1 (λ = 0.154 nm) radiation. Standard θ−2θ scans for the data collections were taken from 2θ = 0 to 4° with an angular resolution of 0.002°.” as been replaced by:

X-ray reflectometry (XRR) analyses were carried out using a Bruker D8 diffractometer with Cu-Kα1 (λ = 0.154 nm) radiation. Standard symmetric Bragg-Brentano θ−2θ scans were used for the data collections  from 2θ = 0 to 4° (q_max = 0.2844 Å^-1) with an angular stepsize of 0.002°, employing small opening and receiving slits and suitable attenuation foils for the measurements in the vicinity of the direct beam. X-ray reflectometry probes the electron density profile of a thin film perpendicular to the surface and therefore symmetric scans are adapted for these measurements as the momentum transfer q is perpendicular to the surface.”

2. A new figure has been added in the manuscript. Its caption is: X-ray reflectivity curves of the deposited films. The curves have been slightly displaced vertically for better visibility.

3. Equation numbers have been added.

4. "VOC" instead of "COV"

5. W et F have been explained in the new paragraph 2.1

6. The reasons for the possible errors are precisely given in lines 147-149.

7. Actually the same evolutions of the refractive index and the mass density can be explained by the fact that mass density is calculated from the electronic density by optical XRR measurements further corrected by the elemental composition.

8. Yes. This has been clarified in the manuscript.

9. Yes. This has been clarified in the manuscript.

10. In such low pressure MW-PECVD plasma, a higher W/F ratio lead to more inorganic (less organic) deposited SiOxCyHz material more crosslinked, less flexible with smaller interchains spaces. Thus, plasma operating conditions influence the ability of sorption that depends on the chemistry but also on the network rigidity and free volume. 

11. Yes. The bigger size of the sorbed molecule of toluene seems to be the main reason that the swelling of the film is more important compared to that of heptane, which is a smaller and linear molecule. But chemistry is also important: heptane and ethanol ad/absorption are different because of the hydrophobicity of the surface which depends on the plasma conditions: surface of WF5 is different from that of WF10 and that of WF20. 

12. This difference of evolutions in figure 2 (former figure 1) between refractive index and mass density is probably due to the scales of the vertical axes. If the red axis of mass density is rescaled from 1.33 to 1.52 instead of 1.30 to 1.55, the slight increase will be more visible.

13. The mass uptake depends on the concentration of the absorbate in the receiver PP-HMDSO network. But the absorbate comes from the surface, i.e., there is absorption if there is adsorption first. The surface of the PP-HMDSO is hydrophobic. Thus water concentration inside PP-HMDSO is weak.  

Round 2

Reviewer 3 Report

Please incorporate your replies 10-13 into appropriate sections in your manuscript. These comments touch on some core issues for your study. Although these inquiries in 10-13 are not fully answered, I won’t insist on digging out the root causes of everything.

However, from the revised manuscript, only limited revisions were made. I strongly recommend that authors constructively treat these comments. 

None.

Author Response

Dear reviewer,

Thank you for your second comments on the revised manuscript.

Reply #10 has been incoporated to the manuscript in the Results section as an introduction to explain why authors carried out QCM coupled with toluene, heptane and ethanol sorptions.

Replies #11 and #13 have been incorporated to the manuscript body in the Discussion section.

Finally, authors agree with the reviewer on the fact that the ellipsometry coupled with vapours sorption can give more information especially quantitative results but for a first approach, authors think that the content of this 13-pages long article may be sufficient. The next article will deeply present such results.

Author Response File: Author Response.docx