Laser-Assisted Surface Texturing of Ti/Zr Multilayers for Mesenchymal Stem Cell Response
, Janez Kovač 3, Miodrag Mitrić 1, Jelena Potočnik 1, Zlatko Rakočević 1 and Emmanuel Stratakis 2Round 1
Reviewer 1 Report
In the paper entitled: " Laser-assisted surface texturing of Ti/Zr multilayers for osteoblast cell response" by Suzana Petrović et. al. the authors report on (Ti/Zr)/Si thin films deposited by ion sputtering on Si substrate followed by surface texturing by femtosecond laser processing. The surface of Ti/Zr multilayer with different composition and topography has been tested as biocompatible material to understand the effect of topography on the survival, proliferation and differentiation of the murine MSCs cells.
The paper is clear and readable and English good.
-In Fig.3 the label (a) and (b) are present in the caption but not in beside the two graphs. Please add the labels to the relative figures.
-The same must be done for Fig.4.
-The scale bars of Fig. 5 b and c are not well readable, please increase their dimension.
-The same must be done for Fig.6 b and d.
-Why the authors do not also show SEM and fluorescence images relative to day 2? It would be interesting to see the cell orientation and proliferation in the intermediate situation.
-A curiosity, can the (Ti/Zr)/Si multilayer structure be a mechanical stimulus also for MCSs differentiation? Are there in literature evidences about this in similar materials?
Author Response
Dear reviewer
We are grateful to the reviewers for very useful comments and suggestions. We have accepted all of them and gave our best to respond to questions and clarify certain ambiguities. All corrections and changes in the text are marked in pink.
In the paper entitled: " Laser-assisted surface texturing of Ti/Zr multilayers for osteoblast cell response" by Suzana Petrović et. al. the authors report on (Ti/Zr)/Si thin films deposited by ion sputtering on Si substrate followed by surface texturing by femtosecond laser processing. The surface of Ti/Zr multilayer with different composition and topography has been tested as biocompatible material to understand the effect of topography on the survival, proliferation and differentiation of the murine MSCs cells.
The paper is clear and readable and English good.
-In Fig.3 the label (a) and (b) are present in the caption but not in beside the two graphs. Please add the labels to the relative figures.
We thank the reviewer for this remark. We have corrected the omission that was made in Figure.
-The same must be done for Fig.4.
We thank the reviewer for this remark. We have corrected the omission that was made in Figure.
-The scale bars of Fig. 5 b and c are not well readable, please increase their dimension.
We have made scale bar readable in the fluorescence images.
-The same must be done for Fig.6 b and d.
We have made scale bar readable in the fluorescence images.
-Why the authors do not also show SEM and fluorescence images relative to day 2? It would be interesting to see the cell orientation and proliferation in the intermediate situation.
We decided to monitor the cell behavior after one and three days based on our previous experience. Namely, after one day, cell positioning / retention sites should be noted, which can be estimated cell adhesion. To test cell behavior, three days are sufficient to determine the sure cell proliferation. For this reason, we have omitted the research of the cell behavior after two days, since no significant changes were observed early, and a large amount of samples were also required.
-A curiosity, can the (Ti/Zr)/Si multilayer structure be a mechanical stimulus also for MCSs differentiation? Are there in literature evidences about this in similar materials?
We did not find study on the mechanical stimulus for MCSs differentiation in literature.
Author Response File: 
Author Response.docx
Reviewer 2 Report
This paper describes a laser-processing technique to produce nanotextured line patterns of Ti/Zr alloys for controlling cell proliferation. The authors prepared the alternating layers of Ti and Zr films on silicon substrates by ion sputtering and introduced textured alloys by pulsed laser. The structures were characterized by scanning electron microscopy, energy-dispersive X-ray spectroscopy, and X-ray diffraction. The authors also compared proliferation of osteoblast cell on the patterned surface with flat multilayer structures and observed preferred cell orientation along the line patterns. However, I cannot support the manuscript for publication because the manuscript did not show clear scientific and technical advances. Detailed comments are provided point by point:
The introduction tried to explain the importance of titanium-based materials (and alloys) for biological applications and the demand for surface processing. However, it is unclear what the limitations of existing Ti materials and alloys are and how the manuscript overcomes these limitations. For example, the second paragraph mentioned the importance of matching Young’s modulus (line 49), but the manuscript did not show any measurements of mechanical properties. The authors need to provide specific examples. Also, a review of alternative surface engineering techniques needs to be compared with the laser-processing strategies in terms of throughput, cost, and resolution. The fundamental mechanism for controlling the cell growth by textured alloys is unclear because the authors demonstrated one design of alloy texture. How does the size of nanoscale and microscale features influence cell growth? How important is the alloy composition and why was a 1:1 Ti/Zr ratio selected? Comparison between the laser-processed surface with flat surface provides limited insight because the flat surface is different in both composition (purely Ti or Zr) and structure. Experiments with different Ti/Zr ratios and Ti/Zr film thicknesses (total and each layer) need to be added. The abstract mentioned the study of cell differentiation (line 24) on the Ti/Zr alloys, but I did not find any data and discussion on cellular differentiation from the results. There are well-established assays for cell viability. These tests are more statistically reliable than microscopy images presented in Figures 5 and 6 which can have spatially varying cell densities. I would suggest further support the claims about cell growth rate with these assays. Why are osteoblast cells selected for the tests? Is controlling the growth and orientation of these cells potentially interesting, or is it considered as an ideal model for fundamental biology studies? Comments on writings and figures: First, the introduction needs to be more focused and specific. For example, the first paragraph is very confusing and most information is not related to the results of the manuscript, which targets Ti-based thin films for biological systems. Avoid grammar issues. Without careful reading, I already found: “surface properties are a key factor ..” (line 41), “Young modul” (line 55), “there are requires more and more” (line 60), and “technologies usually based on chemical” (line 62). Keep font sizes and scales consistent in microscopy images in Figures 5 and 6.
I still think that the alloy structures have interesting properties and great potential applications, but the merits of the work must be justified by additional experiments and significantly improved writing. Therefore, I recommend the authors to resubmit after major revisions with additional experiments that address the listed points.
Author Response
This paper describes a laser-processing technique to produce nanotextured line patterns of Ti/Zr alloys for controlling cell proliferation. The authors prepared the alternating layers of Ti and Zr films on silicon substrates by ion sputtering and introduced textured alloys by pulsed laser. The structures were characterized by scanning electron microscopy, energy-dispersive X-ray spectroscopy, and X-ray diffraction. The authors also compared proliferation of osteoblast cell on the patterned surface with flat multilayer structures and observed preferred cell orientation along the line patterns. However, I cannot support the manuscript for publication because the manuscript did not show clear scientific and technical advances. Detailed comments are provided point by point:
The introduction tried to explain the importance of titanium-based materials (and alloys) for biological applications and the demand for surface processing. However, it is unclear what the limitations of existing Ti materials and alloys are and how the manuscript overcomes these limitations. For example, the second paragraph mentioned the importance of matching Young’s modulus (line 49), but the manuscript did not show any measurements of mechanical properties. The authors need to provide specific examples.
We completely agree with the reviewer's remark that we probably did not show well our goal. Specifically, we take into account the fact that the TiZr alloy has good mechanical properties from literature. However, this TiZr alloy was formed in the localized sub-surface region by combining the deposition of the multilayer Ti/Zr structure and laser processing. A much thicker sample is required for precise determination of the mechanical properties without the influence of the substrate, which is not the case with the sample obtained by the shown procedure. Our main goal is the formation of the alloy in the sub-surface region and the formation of Ti and Zr oxides on the surface, together with the creation of the specific surface morphology, suitable for cell growth, by the laser processing of the multilayer Ti / Zr system. We have added clarification to our research objectives.
Also, a review of alternative surface engineering techniques needs to be compared with the laser-processing strategies in terms of throughput, cost, and resolution.
For the reviewer's suggestion, we have compared traditional chemical and physical surface engineering techniques with the advantages of laser processing technique in more details.
The fundamental mechanism for controlling the cell growth by textured alloys is unclear because the authors demonstrated one design of alloy texture. How does the size of nanoscale and microscale features influence cell growth?
We have added how micro and nano-dimensional surface morphological characteristics affect to the cell response.
How important is the alloy composition and why was a 1:1 Ti/Zr ratio selected? Comparison between the laser-processed surface with flat surface provides limited insight because the flat surface is different in both composition (purely Ti or Zr) and structure. Experiments with different Ti/Zr ratios and Ti/Zr film thicknesses (total and each layer) need to be added.
The explanation, why we chose the multilayer Ti/Zr structure with thickness of the individual layers about 20 nm and a total thickness of 1 μm with a ratio of 1: 1 Ti:Zr before laser modification. During laser processing to achieve well-defined surface texturing, the multilayer Ti/Zr structure is ablated depending on the laser pulse fluence. Sometimes ablation can be pronounced and surface roughness is expressed, to ensure the surface modification on Ti / Zr system without reaching the Si substrate, the total thickness of Ti/Zr multilayer is 1μm. Due to the use of femtosecond laser pulses, the HAZ (heat affected zone) is about 40 nm, which would mean that the intermixing of the Ti and Zr components was happened in 2-3 layers (sub-surface region). Also, the removal of Zr was higher than the Ti component, which is confirmed after laser modification. In subsurface region with TiZr alloy is almost achieved Ti-based alloy with a lower content of Zr, like in bulk alloy where addition component has content up to 25 %. Because of the pronounced effect of ablation and the shallow region of laser interaction, the initial Ti: Zr concentration ratio does not have much effect on such a thin alloy layer.
The abstract mentioned the study of cell differentiation (line 24) on the Ti/Zr alloys, but I did not find any data and discussion on cellular differentiation from the results.
The reviewer is right, without thinking we used cell differentiation when enumerating processes involving the study of cellular response/behavior. However, our testing of cell behavior on laser-modified Ti/Zr multilayer structures involved cell adhesion and proliferation. Thanks for noticing the omission, we have now corrected.
There are well-established assays for cell viability. These tests are more statistically reliable than microscopy images presented in Figures 5 and 6 which can have spatially varying cell densities.
Our study of cell behavior was focused on cell adhesion and proliferation on laser-modified Ti/Zr multilayer samples, the quantification from confocal microscope images yielded a relatively reliable result of cell proliferation.
I would suggest further support the claims about cell growth rate with these assays. Why are osteoblast cells selected for the tests? Is controlling the growth and orientation of these cells potentially interesting, or is it considered as an ideal model for fundamental biology studies?
Mesenchymal stem cells are multipotent stromal cells that can differentiate into a variety of cell types, including osteoblasts were used for monitoring the proliferation along with possible cell orientation, which was the main task of this experimental study on cell behavior. Their differentiation into osteoblasts is only taken as a hypothesis. However, the results presented in the manuscript show good adhesion, intense proliferation and cell orientation along laser-induced periodic surface structures on Ti/Zr multilayer samples. We have incorporated this comment in the manuscript.
Comments on writings and figures: First, the introduction needs to be more focused and specific. For example, the first paragraph is very confusing and most information is not related to the results of the manuscript, which targets Ti-based thin films for biological systems.
At reviewer's suggestion that the first paragraph in the introduction is confused and it is not the scope of the study of the proposed manuscript, we have removed it.
Avoid grammar issues. Without careful reading, I already found: “surface properties are a key factor ..” (line 41), “Young modul” (line 55), “there are requires more and more” (line 60), and “technologies usually based on chemical” (line 62). Keep font sizes and scales consistent in microscopy images in Figures 5 and 6.
We have corrected all these omissions.
I still think that the alloy structures have interesting properties and great potential applications, but the merits of the work must be justified by additional experiments and significantly improved writing. Therefore, I recommend the authors to resubmit after major revisions with additional experiments that address the listed points.
We are truly grateful for the help and comments, and hope that our corrections have now created a better paper.
Author Response File: Author Response.docx
Reviewer 3 Report
The authors used a Si(100) wafer as a substrate that was further coated by the Ti/Zr multilayer thin film. The total thin film thickness was 1 um. The surface was further processed by using a polarized fs laser pulses. This resulted in formation of the laser-induced periodic surface structures that are clearly visible on Fig. 1. I highly recommend the authors to avoid an extensive usage of “creation of micro and nanometer morphological features”, since this is very vague term. Use the term “LIPSS” instead, since LIPSS is a well-known and a well-defined phenomenon. However, the term LIPSS is not used until the Results and Discussion.
When the authors describe LIPSS, they should refer to the following paper, DOI: 10.1109/Jstqe.2016.2614183. Moreover on p. 4., lines 144-145 they simply explain LIPSS as a consequence of interference effect, although several fundamentally different mechanisms are still under the debate (see DOI: 10.1016/j.apsusc.2005.08.120 & DOI: 10.1016/j.apsusc.2016.06.174). When discussing the formation of HSFL (lines 156-160), the authors should compare their results by the results in DOI: 10.1016/j.apsusc.2016.06.174 (Figs. 1&2).
On the laser-modified surfaces, the authors have grown the osteoblast MSCs cells. In Materials and Methods section (lines 118-120) the authors should explain if they used static conditions or if the medium was shook. Additionally, the authors have not count the cell numbers on the surfaces after 1 and 3 days, but their results rely only on visual observations, without any statistical analysis of the cell adhesion. Additionally, the biocompatibility was evaluated only through the adhesion experiments, without any cell viability assays (dead/alive cells, Resazurin, NRU and CBB assay, MTT test, etc.).
In lines 101-102 the authors should be more consistent and should explain that moving the surface out of the focal position, directly influence the pulse fluence on the surface and only indirectly influences the width of the ablated area.
In lines 171-179 the authors discuss the chemical composition made by EDS. The EDS is not an appropriate method for making ANY conclusions of the chemical composition of the thin films after laser processing (e.g., see DOI: 10.1016/j.apsusc.2019.06.068 (subsection 3.4), also the supporting information). Instead, the authors should use the AES or similar method, as described in DOI: 10.1063/1.4993128.
The conclusions are simplified and the discussion of the influence of the surface chemistry and wettability on the cells behavior is missing. The authors should at least provide the contact angle of the surfaces after the laser processing.
Author Response
The authors used a Si(100) wafer as a substrate that was further coated by the Ti/Zr multilayer thin film. The total thin film thickness was 1 um. The surface was further processed by using a polarized fs laser pulses. This resulted in formation of the laser-induced periodic surface structures that are clearly visible on Fig. 1. I highly recommend the authors to avoid an extensive usage of “creation of micro and nanometer morphological features”, since this is very vague term. Use the term “LIPSS” instead, since LIPSS is a well-known and a well-defined phenomenon. However, the term LIPSS is not used until the Results and Discussion.
We have accepted the reviewer's suggestion and the surface structures described in more detail through LIPSSs in the introduction.
When the authors describe LIPSS, they should refer to the following paper, DOI: 10.1109/Jstqe.2016.2614183. Moreover on p. 4., lines 144-145 they simply explain LIPSS as a consequence of interference effect, although several fundamentally different mechanisms are still under the debate (see DOI: 10.1016/j.apsusc.2005.08.120 & DOI: 10.1016/j.apsusc.2016.06.174). When discussing the formation of HSFL (lines 156-160), the authors should compare their results by the results in DOI: 10.1016/j.apsusc.2016.06.174 (Figs. 1&2).
We have taken in consideration of the proposed papers for a better explanation of the formation mechanisms for LSFL and HSFL ripples. These explanations, we have incorporated in introduction and results parts.
On the laser-modified surfaces, the authors have grown the osteoblast MSCs cells. In Materials and Methods section (lines 118-120) the authors should explain if they used static conditions or if the medium was shook.
We have given the explanation that experiments with Mesenchymal stem cells which can differentiate into the osteoblasts were carried out only at static conditions.
Additionally, the authors have not count the cell numbers on the surfaces after 1 and 3 days, but their results rely only on visual observations, without any statistical analysis of the cell adhesion. Additionally, the biocompatibility was evaluated only through the adhesion experiments, without any cell viability assays (dead/alive cells, Resazurin, NRU and CBB assay, MTT test, etc.).
We fully agree with the reviewer that we did not provide quantitative results for cellular behavior testing. Now we have shown the number of cells per observed surfaces as well as MTT analysis after 1 and 3 days. Based on these statistically results, we can conclude that the number of cells on the laser-processed surfaces after the first day was higher, which indicates better cell adhesion in the presence of cell orientation along the LIPSSs. The conclusion is that laser-modified Ti/Zr samples do not increase cell proliferation, but only contribute to the initial higher number of attached cells on the surface with apparent elongation in the direction of LIPSSs
In lines 101-102 the authors should be more consistent and should explain that moving the surface out of the focal position, directly influence the pulse fluence on the surface and only indirectly influences the width of the ablated area.
We have accepted the reviewer's remark, that the irradiation was done in defocus mode by positioning the samples out of focus at 4 cm, thereby it was achieved a pulse fluence of 0.4 J cm-2 with a spatial extension of the Gaussian beam profile. Indirectly, under the given laser irradiation conditions, the width of line with value of 75 μm is adjusted by defocusing of the laser beam, while distances between lines keep constant at value of 80 μm. The lines were direct laser writing at scan velocity of 2 mm s-1. Also, we have put the figure with experimental laser setup.
In lines 171-179 the authors discuss the chemical composition made by EDS. The EDS is not an appropriate method for making ANY conclusions of the chemical composition of the thin films after laser processing (e.g., see DOI: 10.1016/j.apsusc.2019.06.068 (subsection 3.4), also the supporting information). Instead, the authors should use the AES or similar method, as described in DOI: 10.1063/1.4993128.
We fully agree with the reviewer, that the EDS technique is not suitable for the precise study of the thin film composition. With this technique we wanted to indicate changes in the chemical composition, which is oxidation in zone of the laser modification and relatively higher ablation of Zr in comparison with Ti component. The main reason for using the EDS technique is to check the degree of ablation during laser irradiation in order to avoid, for example, complete ablation of the thin film. In the text of the manuscript, we have added that the EDS method was used only for first roughly estimate the composition changes. For a more accurate estimation of composition changes after laser processing, we have used SIMS technique and we have given depth profiles on Figure 5.
The conclusions are simplified and the discussion of the influence of the surface chemistry and wettability on the cells behavior is missing. The authors should at least provide the contact angle of the surfaces after the laser processing.
We thank the reviewer for this helpful remark. We have added the result of the measurement of contact angle for Ti / Zr multilayer structure before and after laser treatment. Also, we have added a comment in the manuscript regarding the effect of surface composition on the cells behavior.
Author Response File: Author Response.docx
Reviewer 4 Report
The paper contributes to the laser processing of titanium-zirconium multilayer thin films. The reviewer recommends some major edits prior to publication:
Applications of Ti/Zr should be presented in the introduction section. A schematic diagram of the laser texturing setup SEM Magnification of Figures 2b, c, d, e The EDS results (Figure 1) should be presented in a separate figure, showing the location of the EDS area of interest. The accuracy of the EDS should be mentioned in the paper. How the fracture surface was prepared? (Figure 2) Why the authors did not try several laser processing conditions? And why specifically they used 0.4 J cm-2. The main contribution needs to be discussed in the conclusions section.
Author Response
The paper contributes to the laser processing of titanium-zirconium multilayer thin films. The reviewer recommends some major edits prior to publication:
Applications of Ti/Zr should be presented in the introduction section.
We thank the reviewer for this suggestion. Now, we have added a concrete application of Ti-Zr alloy as implant with improved properties in the introduction part of the manuscript.
A schematic diagram of the laser texturing setup
We have incorporated the schematic representation of the experimental setup for laser irradiation as new figure in the manuscript.
SEM Magnification of Figures 2b, c, d, e
In a new arrangement of figures, the figures that referring to the SEM analysis are Fig.2 and Fig. 4, in their description, we have added the magnification at which they were recorded.
The EDS results (Figure 1) should be presented in a separate figure, showing the location of the EDS area of interest.
We have extracted the EDS analysis into a separate figure together with a SEM image showing the positions of the recorded EDS spectra.
The accuracy of the EDS should be mentioned in the paper.
We have included that despite the used instrument analytical accuracy of 2σ = 2%, the obtained results should be taken as semi-quantitative because of the small thickness of 30x(Ti/Zr) multilayer system.
How the fracture surface was prepared? (Figure 2)
The native cross-section of the sample was simply prepared by made fracture transversely to the laser-written lines.
Why the authors did not try several laser processing conditions?
A few laser texturing experiments were performed before we decided on the laser modification conditions shown in the manuscript. We were made a series of laser modifications of the samples, where the following parameters were varied: scanning speed (writing lines with a laser beam), distancing from focus point and pulse energy. Finally, we chose laser modification conditions to obtain as wide a line as possible with well-defined regular LIPSSs without significantly ablation of the thin film.
And why specifically they used 0.4 J cm-2.
We decided for the laser fluence of 4 J cm-2, which is slightly higher than the ablation threshold because in this case we have obtained the well-defined LIPSS without pronounced ablation of the multilayer structure. Otherwise, at higher laser fluence, the intensive ablation occurs and the depth between the LIPSS can reach the Si substrate. On the other hand, at lower laser fluence, no regular LIPSS shaped as LSFL are obtained which have a periodicity close to the laser wavelength.
The main contribution needs to be discussed in the conclusions section.
We have added a comment in the conclusion of the manuscript regarding the effect of surface composition and surface topography on the cells behavior.
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
The paper, after the modification suggested by the referees, can be now published.
Author Response
We thank the reviewer 3 for this remark. We have corrected the omission that was made in presenting the results of contact angle measurements. The experimental measurement of the contact angle were included a statistic of 5 measurements, with an accuracy of 5%. Based on these results and accuracy, we have decreased the number of significant digits for showing values of contact angles. Now we have not made new measurements of the contact angle, because these values were obtained just before seeding the cell, we just adjusted the display of the results.
Reviewer 2 Report
The revised version provides additional results and discussion that improve the quality of the work significantly. Therefore, I support acceptance now.
Author Response
We thank the reviewer 3 for this remark. We have corrected the omission that was made in presenting the results of contact angle measurements. The experimental measurement of the contact angle were included a statistic of 5 measurements, with an accuracy of 5%. Based on these results and accuracy, we have decreased the number of significant digits for showing values of contact angles. Now we have not made new measurements of the contact angle, because these values were obtained just before seeding the cell, we just adjusted the display of the results.
Reviewer 3 Report
I checked the author's response. The authors have answered all my questions. I also briefly scrolled the reviewed version of the manuscript. The only minor issue I found is connected with the contact angle. The authors have reported the so-called "static" contact angle by using 4 significant digits.
Although this is not a critical issue, I would like to warn the authors that the static contact angle is a misleading term referring to measuring the CA “as is” – namely as the drop happens to land on the solid surface. Unfortunately, this is a meaningless measurement, since there is a random element involved in the process of drop landing. Consequently, the “static” contact can be any value within the range of advancing and receding contact angels (e.g., see: DOI: 10.1680/jsuin.17.00002). Such problem can be solved by repeating the measurements several times and by revealing the average and standard deviation of such measurement. Additionally, where applicable, is in this context worth to provide the roll-off angle.
However, I believe that at least the authors should reduce the number of significant digits, since I do not believe that they are able to measure the CA by 0.01% accurracy (especially not in a context of the above explained problems of the "static" CA).
Author Response
We have corrected the omission that was made in presenting the results of contact angle measurements. The experimental measurement of the contact angle were included a statistic of 5 measurements, with an accuracy of 5%. Based on these results and accuracy, we have decreased the number of significant digits for showing values of contact angles. Now we have not made new measurements of the contact angle, because these values were obtained just before seeding the cell, we just adjusted the display of the results.
Reviewer 4 Report
Accept
Author Response
We thank the reviewer 3 for this remark. We have corrected the omission that was made in presenting the results of contact angle measurements. The experimental measurement of the contact angle were included a statistic of 5 measurements, with an accuracy of 5%. Based on these results and accuracy, we have decreased the number of significant digits for showing values of contact angles. Now we have not made new measurements of the contact angle, because these values were obtained just before seeding the cell, we just adjusted the display of the results.
