Research and Optimization of the Influence of Process Parameters on Ti Alloys Surface Roughness Using Femtosecond Laser Texturing Technology

Round 1

Reviewer 1 Report

Title: Research and Optimization of the Influence of Process Parameters on Ti-Alloys Surface Roughness Using Femtosecond Laser Texturing Technology

MS No.: coatings-2422115-peer-review-v1

This work investigated the influence and the optimization of processing parameters on the roughnesses of Ti alloys by using femtosecond laser texturing technology. The surface roughness of Ti alloys is introduced for the purposes of the preparation of coating, which would have a significant influence on the coating adhesion. The authors used the design of experiments and Monte Carlo simulations to provide an interval with a significance level of 95% for the surface roughness. The results are interesting and valuable. However, some revisions are still recommended.

1.The aim of this work should be extended and revised. Why is this article written? What is the difference between the present submitted article from the previous works? What is needed for this article? Explain all these questions at the end of the introduction part of the article.

2. In the Page 2, “In [24] authors examined …” should be “in Ref. [24], the authors examined…”

3. For different surface modification methods of Ti alloys and for different purposes, the surface roughnesses have various requirements. When the authors gave examples in the Introduction, can authors also mention the surface modification methods besides the specific coating? The surface modification methods of Ti alloys can be referred to “Advanced Engineering Materials 22, 2020, 1901258”.

4. Can authors provide the processing state of the sample in this work. As-cast? Forged? Or others?

5. The scale bar in Figure 1b is too small to be clearly seen, including other SEM images.

6. High quality and clear of Figure 2, Figure 4, Figure 7, Figure 10, and Figure 14 are required.

7. The Figure 3 can be provided in text. In the current version, it is a screenshot from other literature. It has a significantly low quality.

8. The authors should discuss the results and compare the results with the previous studies and mention coherent/incoherent points with the possible reasons.

9.The typesetting of mathematical quantities must be corrected. Variables must be typeset in italics font, sub-, superscripts other than tensorial (counting) indices, and mathematical operators must be typeset in the upright font.

10.The manuscript needs to be revised concerning the style of references because it contains many mistakes and inconsistencies.

11.Grammar errors and typos are found. Please check the article carefully.

Minor editing of English language required

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The manuscript is devoted to the optimization of process parameters, such as the number of passes and the average power of a femtosecond laser on the surface roughness of a titanium alloy. A promising goal was to increase wear resistance by improving the adhesion of the protective coating. The optimization used a Design of Experiments and Monte Carlo simulations.

The manuscript is convincing, very detailed and meticulously prepared. I would like to criticize only the style of presentation in the form of diagrams overflowing with parameters, which divert the reader's attention from the physics of phenomena. Redundant numerical data and calculations are best to be displayed in the separate appendices at the end of the paper. Conclusions are difficult to understand, are written as a squeeze from the previous 18 pages and are full of numerical data, but it would be better to see a brief and comprehensive analysis of the investigation. The manuscript deserves publication after correcting these and the list of shortcomings below:

1. For the Figure 3 between the lines 228-229: “Contribution - %”: please, translate the following text to English; “p-Value”: please, translate the following text to English; the same for parameter S;
2. Resolution of SEM images a2, b2 and c2 is too low on Figure 11 to be able to evaluate.
3. Lines 409-410: “The coupling of input factors (Pmc, Wmc) with the response variable (Ra) by using the equation (8).” - It seems that this sentence is incomplete, since it does not contain a verb;
4. Line 459: “Typically, the value of the Test data for R^2 was lower than R^2” – please, consider rephrasing for better clarity.

Minor editing of English language required

Author Response

Reviewer 2:

The manuscript is devoted to the optimization of process parameters, such as the number of passes and the average power of a femtosecond laser on the surface roughness of a titanium alloy. A promising goal was to increase wear resistance by improving the adhesion of the protective coating. The optimization used a Design of Experiments and Monte Carlo simulations.

The manuscript is convincing, very detailed and meticulously prepared. I would like to criticize only the style of presentation in the form of diagrams overflowing with parameters, which divert the reader's attention from the physics of phenomena.

• Redundant numerical data and calculations are best to be displayed in the separate appendices at the end of the paper.

It is not clear what redundant data the reviewer is referring to. The authors ask for additional clarification. However, the authors suggest that the numerical data and calculations remain in the format as is. It would be difficult for the reader to follow the text of the paper, if the tables or pictures were transferred to the appendix of the article.

• Conclusions are difficult to understand, are written as a squeeze from the previous 18 pages and are full of numerical data, but it would be better to see a brief and comprehensive analysis of the investigation.

The authors agree with the reviewer's remark that it is probably difficult for the reader to follow the text of the paper due to the large amount of data and that he may not have a complete overview of the paper before reading the conclusion. Because of this, the authors concluded that something like an overview image of the work should be added.

For additional clarification, a new Figure 1. Overview of research steps and methods in the introductory part of the article has been added.

Due to the added Figure 1., the authors suggest that the conclusion remain as it is.

The manuscript deserves publication after correcting these and the list of shortcomings below:

• For the Figure 3 between the lines 228-229: “Contribution - %”: please, translate the following text to English; “p-Value”: please, translate the following text to English; the same for parameter S;

Updated and redesigned.

• Resolution of SEM images a2, b2 and c2 is too low on Figure 11 to be able to evaluate.

Updated and redesigned.

• Lines 409-410: “The coupling of input factors (Pmc, Wmc) with the response variable (Ra) by using the equation (8).” - It seems that this sentence is incomplete, since it does not contain a verb;

Corrected.

• Line 459: “Typically, the value of the Test data for R^2 was lower than R^2” – please, consider rephrasing for better clarity.

Reformulated.

Reviewer 3 Report

In this work, authors provide the optimization of femtosecond laser machining parameters by design of experiment method, Mote Carlo simulation, and linear regression model.

1. In the introduction references in introduction seems to be irrelevant to the subject. This work is more based on the statistics not on scientific investigation on femtosecond laser machining.

2. In the experiment section, experimental setup should be clearly presented by showing some plots in order for other researchers to reproduce the experiment.

Author Response

Reviewer 3:

In this work, authors provide the optimization of femtosecond laser machining parameters by design of experiment method, Mote Carlo simulation, and linear regression model.

1. In the introduction references in introduction seems to be irrelevant to the subject. This work is more based on the statistics not on scientific investigation on femtosecond laser machining.

The authors believe that the mentioned references are in accordance with the topic of the research, which is, first of all, research on the influence and optimization of the process parameters of Ti-Alloys surface roughness using fs laser texturing technology. The proposed way of optimizing the parameters of the surface texturing process and introducing the performance capability index-C_pk, according to the mentioned references, has not been implemented so far. The authors are of the opinion that the work presented has all the elements of a scientific research work, from setting up a hypothesis to proving it. Statistical methods were used due to the stochastic nature of laser texturing, and we wanted to investigate the influence of the main parameters of the texturing process on the roughness R_a.

2. In the experiment section, experimental setup should be clearly presented by showing some plots in order for other researchers to reproduce the experiment.

For additional clarification, a new image (Figure 1.) has been added to the paper, which shows the steps and research methods.

Reviewer 4 Report

The paper is heavy on statistical analysis of a very specific narrow application of laser texturing.  It would be helpful if the authors could discuss how their finding apply more generally.

Lines 90-92: The different R values -Not clear what these are and how they differ for someone not working in this field. Could a diagram or equations be used to describe them?

Line 97: how was the spot size measured?

lines 98-99: the spot size, line velocity and spacing are also experimentally controlled input variables.  So there is a much larger parameter space to be investigated other than just number of passes and average power.  Explain why you limited yourselves to varying just 2 input parameters.

Fig. 1a x axis tick mark values need repair

Fig.3 Some non-English words are present

line 377: The topography scans and the SEM photos don't seem to agree.  The SEM photo looks like mud that has dried and contracted leaving cracks between islands of relatively flat material.  But the topography looks like sharp peaks rising ABOVE flat regions.  Should the sign of the topography scans be reversed?

line 378: The roughness scans and SEM morphologies for the higher powers are quire different from the low power one.  Ra is no longer a good description for them since you appear to have mostly flat regions with deep cracks that increase in number with increasing power.  There is probably a change in physical process going on - perhaps a change from local softening spreading and rehardening to melting and volatile droplet emission. Please comment on this in the manuscript.

Fig. 11 SEM scale is too small to read.

Minor improvements in grammar such as changing "as less rough" in line 39 to "as low in roughness" and "author" to "the authors" in line 44 could be made but the English is understandable.

Author Response

Reviewer 4:

• The paper is heavy on statistical analysis of a very specific narrow application of laser texturing. It would be helpful if the authors could discuss how their finding apply more generally.

Statistical analysis was necessary considering that the research is related to industrial application. The first goal set by the authors was to investigate whether the parameters number of passes and laser power affect the amount of surface roughness of Ti-Alloys samples using Femtosecond laser texturing technology. As another goal, it was to obtain a model that, based on the given value of R_a, would give the necessary values of the number of passes and the amount of power needed to properly set up the laser machine. Given the stochastic nature of the process itself, optimization of the process parameters at the 95% level of statistical significance was also carried out in the research. A general and detailed discussion of the research method is presented in the introductory part of the paper.

For additional clarification, a new image (Figure 1.) has been added to the paper, which shows the steps and research methods.

• Lines 90-92: The different R values -Not clear what these are and how they differ for someone not working in this field. Could a diagram or equations be used to describe them?

Diagram added (Figure 2.)

• Line 97: how was the spot size measured?

The laser pulses were focused on the sample surface to a 170 μm radius spot. The size of the spot was measured by strongly attenuating the beam and then sending the focused beam to a CCD camera.

A sample was fixed to an XYZ computer-controlled translational stage, with the Z axis being the laser beam direction and the surface of the sample being in the XY plane. The texturing was performed by scanning the sample in the XY plane, line by line, at a con-stant line velocity of v_x = 10 mm/s and with an inter-line (y) spacing of h = 250 μm.

For preliminary and DOE experiments laser fluences of 0.057, 0.139, 0.205 and 0.297 J/cm² were used, which correspond to average laser powers of 52, 126, 186 and 270 mW, respectively. For each fluence 3, 5, and 7, consecutive layers were made as aforementioned. In the rest of the paper, we refer to average laser powers for easier calculations.

• lines 98-99: the spot size, line velocity and spacing are also experimentally controlled input variables. So there is a much larger parameter space to be investigated other than just number of passes and average power.  Explain why you limited yourselves to varying just 2 input parameters.

According to [34,35] by changing the femtosecond laser parameters such as speed, power, and size of the laser beam, it is possible to produce surface roughness that can be hydrophobic or hydrophilic. Therefore we decided to change two laser parameters that were changed in the mentioned works as well.

• 1a x axis tick mark values need repair.

Updated.

• 3 Some non-English words are present.

Updated and redesigned.

• line 377: The topography scans and the SEM photos don't seem to agree. The SEM photo looks like mud that has dried and contracted leaving cracks between islands of relatively flat material.  But the topography looks like sharp peaks rising ABOVE flat regions.  Should the sign of the topography scans be reversed?

Sample labels were not replaced. Roughness profiles (surface topography) and SEM images were measured that way. The ridges seen in the SEM images are the deposited microdroplets (molten material) that was created during laser texturing. Such protrusions (hills) cannot be avoided because they are an inevitable event during the melting of the surface of the material. They are present on all the samples used in this work at the given laser texturing parameters.

• line 378: The roughness scans and SEM morphologies for the higher powers are quire different from the low power one. Ra is no longer a good description for them since you appear to have mostly flat regions with deep cracks that increase in number with increasing power.  There is probably a change in physical process going on - perhaps a change from local softening spreading and rehardening to melting and volatile droplet emission. Please comment on this in the manuscript.

The authors agree with the reviewer's remark in accordance with the ref. [23,27,34,35]. However, the basic idea of the work was not a study of the microstructure at the molecular level and the analysis of the physical properties (effects) that can influence the adhesiveness of the coating. The paper was written for the purpose of engineering application of femtosecond laser in surface modification of Ti alloy before the coating process. The aim of this paper was to define an acceptable interval of roughness R_a by changing the number of passes of the laser beam (P) and laser power (W), for the preparation of the surface substrates in order to achieve better adhesion of the thin hard tribological coating and the metal surface. The proposed peer reviews can be the subject of future research in collaboration with colleague physicists.

• 11 SEM scale is too small to read.

Updated.

Reviewer 5 Report

Dear Authors!

I have revised the manuscript, and made the following comments and suggestions:

Figure 3. It is blurring. It is recommended to  replace it with the typed text

Figure 4. Improve the quality

Line 373-375. Please, in the manuscript (not only here) replace Laser power by laser power density (W/cm² ; W/cm), like that.

Line 431. ...was 0.0139 [mW]...  0.0139 mW

Line 499  samples? Or iterations?

The results of experimental data (performed on titanium alloy samples, not simulations) should be described in a separate subsection.  The comparison between experimental and simulation results should be made (convergence of the results)

Is it possible to describe the experimental results as a graph or a surface plot?

Did You check the simulation results experimentally? 

Also, I think that this manuscript does not meet the topical requirements of the "Coatings".

Author Response

Reviewer 5:

I have revised the manuscript, and made the following comments and suggestions:

1. Figure 3. It is blurring. It is recommended to replace it with the typed text.

Updated and redesigned.

2. Figure 4. Improve the quality.

Updated and redesigned.

3. Line 373-375. Please, in the manuscript (not only here) replace Laser power by laser power density (W/cm² ; W/cm), like that.

The laser pulses were focused on the sample surface to a 170 μm radius spot. The size of the spot was measured by strongly attenuating the beam and then sending the focused beam to a CCD camera.

A sample was fixed to an XYZ computer-controlled translational stage, with Z axis being the laser beam direction and the surface of the sample being in the XY plane. The texturing was performed by scanning the sample in the XY plane, line by line, at a constant line velocity of v_x = 10 mm/s and with an inter-line (y) spacing of h = 250 mm.

For preliminary and DOE experiments laser fluences of 0.057, 0.139, 0.205 and 0.297 J/cm² were used, which correspond to average laser powers of 52, 126, 186 and 270 mW, respectively. For each fluence 3, 5, and 7, consecutive layers were made as aforementioned. In the rest of the paper, we refer to average laser powers for easier calculations.

4. Line 431. ...was 0.0139 [mW]... 0139 mW

Updated.

5. Line 499 samples? Or iterations?

Updated.

6. The results of experimental data (performed on titanium alloy samples, not simulations) should be described in a separate subsection. The comparison between experimental and simulation results should be made (convergence of the results).

The comparison of research results (preliminary, DOE, MC) is shown in Figure 17. Results of Multiple Sample Comparison of Roughness-R_a.

7. Is it possible to describe the experimental results as a graph or a surface plot?

The comparison of research results is shown in Figure 17. Results of Multiple Sample Comparison of Roughness-R_a in tabular and graphical form.

8. Did You check the simulation results experimentally?

The generated regression models (R_aoLM and R_amc) were tested on 15% of the data that did not enter to the regression model training phase. Results of Multiple Sample Comparison of Roughness-R_a show that the regression models fit well the values of the experimental data obtained by the DOE/MC experiment. The values for the preliminary data are somewhat weaker, which is understandable because the number of data is smaller. Detailed analysis is presented in 3.4. Multiple sample comparison of roughness.

Also, I think that this manuscript does not meet the topical requirements of the "Coatings".

The authors cannot comment on the above because it is not clear in which part the presented work does not meet the topical requirements of the "Coatings".

Round 2

Reviewer 3 Report

1.     Line 61: Clarify the meaning of “the treated zone is much smaller than in nanosecond and microsecond laser pulses”. Thermally or dimensionally?

2.     Line 101: 17 x 17 x 7 mm should be 17 x 17 x 7 mm³

3.     Line 132: In “for each fluence 3, 5, and 7,” a unit is missing.

4.     Experimental setup for the femtosecond laser surface machining setup should be clearly presented with parameters and specifications in order for other researchers to reproduce the experiment.

5.     Line 152: Units for Po and Wo are missing.

6.     From Line 199 to 206: They should be tablelized.

7.     Line 232 to Line 235: They are confused with a and b in Eq. (3).

8.     Above line 248: The formulas should be presented in the equations with reference.

9.     Figure 5 is not appropriate. They should be presented in equations.

10.  From Line 264 to 281: they should be tablelized.

11.  From 294 to 307: they are not appropriate to be included in text. Present in other forms. (flow chart, or table).

12.  The characters and numbers are too small to be notified In Figs. 7 and 8.

13.  Paper is not well organized.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Round 3

Reviewer 3 Report

All the questions are adequately addressed. However, I still think that all the equations in the figure should go to supplementary materials for the readability and conciseness of the paper.

Author Response

Please see the attachment

Author Response File: Author Response.docx