Influence of Growth Defects on the Oxidation Resistance of Sputter-Deposited TiAlN Hard Coatings

Round 1

Reviewer 1 Report

The manuscript submitted is intended to discuss how growth defects influence the oxidation resistance of TiAlN coatings. It has a long list of more or less big deficiencies. 

1. The abstract is too long. The authors write "Therefore we belive that pores occuring in the upper aluminum oxide layer at temperature 850 ºC have a greather impact on oxidation resistance than growth defect. " This is not core of the manuscript and is not relevant for the abstract.
2. The introduction is lacking some references. Also the motivation is unclear until the 10th paragraph ("There are only few papers in the literature dealing with the role of growth defects..."). I personally would have  liked less descriptions of deposition strategies to improve the oxidation resistance (which is not the focus of the manuscript) but rather a more detailed description of growth defects behaviour in oxidising environments for different material systems for example
3. References 27, 38 and 39 are not mentioned in the text. Also the spelling of certain references should be checked. For example ref 4: P. H. Mayrhofer and not Mayerhofer.
4. The Materials and methods section is incomplete. Please include the provider of the substrates, a detailed description of the cleaning method (including the detergents used and time spent in the ultrasonic bath), the glas flows of Ar, N2 and Kr and the deposition duration. 
5. In the second paragraph of the Materials and methods section, the authors wrote "Coated substrates were isothermally oxidized in ambient atmosphere at temperature 800 °C and 850 °C for times ranging from 5 min to 20 h using a conventional tube furnace." 
   1. First is there a specific motivation behind the decision of oxidizing the coatings at 800 and 850 degC? The 50degC difference does not seem that significant considering the potential temperature gradient in the tube furnace. 
   2. Second, I am assuming that the times of 5 min to 20 h are typos as the shortest time discussed is 15 min at 850degC and 120 min at 580degC.
6. The results need a lot of work to make this section more fluid for the reader. As a general comment, it is very uncomfortablöe and frankly annoying to have to come back and forth to figure 1 throughout the text. I would suggest reworking the figures and regrouping as deposited micrographs and EDS maps in one or two figures, then the ones corresponding to the oxidized samples at 800degC, etc. 
7. More specifically, the authors would need to discuss all figures presented. For example I am lacking information on Figure 1e1 where we can see that the oxide layer is now continuous but the authors do not mention whether it has delaminated or not. At the end of the manuscript they mention the film has fully oxidised, however, the oxide scale seem dense enough that it seems somewhat surprising.
8. The authors have included an etching step at 3/4 of the deposition time. It does not seem as an important step or even a necessary one. Indeed, the motivation behind it is to induce a microstructural change from a coarsed grain/columnar structure to a finer equiaxed one. That is making the film more prone to oxidation as the grain boundaries near the surface are more numerous and constitute paths for inward oxygen diffusion down to a certain depth. Furthemore, it is clear from Figure 3 that the etching step has no effect on the formation of growth defects as these appear to nucleate at the substrate/film interface. 
9. Was there a comparison made of the coatings roughness with and without the etching step during deposition? How does that affect the oxidation in terms of time (a fully columnar film oxidizes faster? slower? please discuss). 
10. The manuscript is also lacking XRD data before and after oxidation - Please include them as they will support even more the conclusions. 
11. What is the composition of the deposited film? The composition is often relevant to the oxidation mechanism - see the case of MAX phases (X. Li, Journal of the European Ceramic Society 36 (2016) 3311–3318). 
12. It would be easier to compare the EDS maps that are shown if the colours are consistent. Consider using the same colour for all the same element in all figures. While a detail, it makes the figures easier to follow. 
13. Page 10, when discussing the formation of TiO2 crystals the authors support that the "penetration of oxygen at these sites (i.e. sites where these oxide clusters are observed) is faster". What led to this conclusion? In H.J. Yang et al. Scripta Materialia 65 (2011) 135–138, they observe similar oxide scales for the Ti2AlC MAX phase. One of their conclusions is that pores and cavities lead to the formation of randomly oriented Al2O3 grains, rather than a continuous protective oxide scale, which promotes outward diffusion of Ti. Would you care to comment on that? One the same subject, page 12 last paragraph, the authors also write "During growth a large Ti-O crystals penetrated into the upper Al-O layer." I have trouble understanding how these crystals would penetrate the layer. Please rephrase.
14. Please remain consistent with the use of TiO2 or Ti-O, and Al-O or Al2O3 throughout the manuscript. Both are often used to describe the same crystals or layers.
15. Regarding the oxidation of 120 min, you mention that the film has been completely oxidized whereas for half the time an oxide scale of approx. 1 micron is formed. What do you have to say about the oxidation kinetics?
16. In general the manuscript is lacking a proper comparison of the oxidation behaviour of the TiAlN coating in a region with growth defects and one without. The authors might need to re-analyze their samples or deposit, oxidize and analyze new ones. 

Author Response

see attached file

Author Response File: Author Response.docx

Reviewer 2 Report

This paper report results of an investigation of oxidation of sputter deposited TiAlN hard coating in air at temperatures 800 and 850 °C for times ranging from 15 minutes to 2 hours. This content is well organized. Here list some minor errors needed to be modified.

1. Subscript error

            Line 87 TiO2

            Line 126 LaB6

            Line 167 N2

2. Reference error: missing item number
3. It is well known that magnetron sputtering made good surface roughness of coating. Why there is significant particle on top surface as shown in Fig. 1?
4. Line 17 : the surface area covered by growth defects is relatively low less than 3% …., Is this value critical to resist oxidation or just a special case?

Author Response

see attached file

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

The revised version of this manuscript is clearer and contains more information, discussions and conclusions which are supported by the authors' observations. The revisions have indeed strengthened the focus of this work. It has also become more enjoyable to read despite the presence of few typos and grammar errors (see below). Thank you for your efforts to address the issues in the previous version of this manuscript.

1. The authors discuss that the oxidation at growth defects is different and depends on the shape and type of defect. Was there a conclusion drawn as to which kind of defect (Protrusion, nodular defects or pinholes?) Indeed in section 3.2 dark patches (rich in Ti) are discussed and are said to not form around all defects. was there a particular defect that promoted the formation of these patches?

Below some formatting issues that the authors may want to take into consideration: 

1. There are a few typos and sentences that need rephrasing for better clarity: 
   1. line 139-140: Remove the "(grown on droplets)" which does not contribute to the text. 
   2. line 233 "fracture cross-section SEM images" (stay consistent throughout the text).
   3. lines 301-302: "Whether the diffusion..." is not clear - please rephrase. 
   4. In section 3.3, line 339, the authors write ".. can be indentified, but there is still no alumina diffraction pattern". please rephrase. There are no crystalline alumina contributions however, there seem to be amorphous alumina as evidenced from elemental mapping and mentioned elsewhere in the text
   5. lines 346-350: "Namely the position... ... or microcracks formed in the top alumina layer" both sentences need to be rephrased for clarity.
2. Please specify the alumina phase in the XRD and the text (is it γ or α - probably α)
3. Figure 12 contains very nice micrographs and lots of information, however the presentation is somewhat confusing. Indeed, there are two fig.12b - Maybe replace the letters b,c,d,e of each line by a number to avoid confusion? Also it would be helpful to add a square of the area corresponding to the high magnification b figure in the low magnification b figure. 
4. Still in Figure 12, what does the A3 stand for?
5. In Figure 18, please make sure there is a better contrast between the letters and the micrographs - The a and Pt in Fig.18a are barely noticeable. 

Author Response

see attached file

Author Response File: Author Response.docx