First-Principles Investigation of the Diffusion of TM and the Nucleation and Growth of L12 Al3TM Particles in Al Alloys
Round 1
Reviewer 1 Report
1). I managed to read through this manuscript. However, this manuscript contains many typos, such as ‘slap models’ (line 20 in the Abstract), and grammatical mistakes. There are many terms which are not well defined. For example, line 37 the term ‘trace elements (TMs)’ is unclear and may mislead people. There are also many long sentences of confusions. The Abstract is lengthy and unclear. The Introduction can be improvements with addition of more recent literature about the L12-Al3TM and related phases from both experimental and theoretical studies. Overall, this manuscript looks more like a draft rather than one being ready for publication.
2). According to Refs. 15 and 16, the statement about the role of L12-Al3TM particles in the microstructure of the ally (pinning of the dislocations and grain boundaries) is true only for systems containing Sc. Meanwhile, there are no L12-Al3TM phases formed in many/most of the related Al-TM systems. Although, the obtained information here helps people to have a comprehensive knowledge about the role of the L12-Al3TM phase in a systematic way, this study may mislead readers. Moreover, adding the available experimental results into Table 1 is needed according to the text (line 150).
3). The authors used a cut-off energy of 500 eV, please specify it: is it for the wave functions or for the augmentation functions?
4). It seems that the authors used a 10×10×10 k-mesh for the conventional cell of L12-Al3TM which contains four atoms. This k-mesh might not be enough to provide accurate results. I’d like to suggest the authors give results of k-mesh tests to confirm the reliability of the results for readers.
5). I did not find the calculated results, including lattice parameters, formation energies, vacancy formation energies, etc. for the proposed cubic L12-Al3TM compounds. Moreover, comparison between the calculated results and experimental values in the literature if available should be also given. This information is important for readers to understand the reliability of the calculations and to understand the lattice misfits between the matrix Al and the L12-Al3TM particles (interface energies and strain effects).
6). The authors treated the nonstoichiometric {100} and {110} surfaces in an averaged way of two different surfaces: Al-terminated and AlTM-terminated. This is a wise way. Meanwhile, the values for the two types of surfaces might also be interesting to many readers, I’d like to suggest that the authors list them in Table 3.
7). There are negative values for the surface energies of several L12-Al3TM compounds (Table 3). I’d like to ask the authors to provide a clear picture about the physics behind the negative surface energies? This question is also true for the activation in Table 1.
this manuscript contains a lot of improper English expression, such as conclusion point 3: "And with the increase of atomic number, the diffusion rate Ds firstly 341 decreases linearly from Sc, Y and Hf to Mn, Ru and Ir, and then increases to Zn, Ag 342 and Au, for 3-5d TM elements, respectively."
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
This paper contains much important data. Some serious production issues must be fixed (especially the referencing). The paper’s impact may be enhanced if the data is interpreted so patterns in the results can be rationalized.
See accompanying file for details.
Comments for author File: Comments.pdf
English is not idiomatic but is serviceably clear and correct. Some small improvements are included in the detailed review file.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
The present work tries to show an investigation of the diffusion mechanism of TM as well as the nucleation and growth of L12 Al3TM in a Al matrix. For that, some calculation principles were used, and VESTA software was also used to show some details. The paper is interesting and has relevance to the field and is suitable to be published in Crystal. But, before a possible acceptance, I recommend a revision according to the following raised points:
In general, I feel a lack of discussion regarding nucleation and growth in a direct way related to the presented simulation. Authors should clearly discuss these mechanisms in the sections: 3.2. Nucleation hindrance and driven force; 3.4. Surface property.
Is there section 3.3? I did not see it.
Fist and second sentences of the abstract are too long. Each of them should be divided into two sentences.
Page 2, line 48. "1As we all known," Avoid commencing sentences in this way.
Page 2, line 64. "RE" Describe as it appears the first time in the paper. Rare Earth?
Page 2, line 89. "have obtained." or have been obtained?
Figure 1. (a) The diffusion model. Missed input letter a) in the first image
Page 6, line 230. "1. Li et al. adopted." Even though it is said before, you should include after al. the citation number.
"2. Although Mao et al." Same remark as the previous one
Some minor errors have been detected, and I suggest a deep reading in the whole text to correct them.
Author Response
Please see the attachment.
Author Response File: 
Author Response.docx
Round 2
Reviewer 1 Report
I have gone through the revision. The authors answered very well and made corresponding modification. The questions from my side is clear. I would recommend to publish it.
Author Response
Thank you very much for your review and recognition of our work!
Reviewer 2 Report
Comments on revised MS 2461949
The authors have responded conscientiously to the review of their MS. In this follow-up I will follow the numbering scheme used in their reply.
Question 1 (on referencing):
Corrections are not complete. For example, in the text (Line 61), Mao et al. is assigned reference #34. In the reference list, #34 is a citation of Wang and Wang.
Proper citation of VASP software [37] may be found here:
https://hpc.hku.hk/hpc/software/vasp/
A more general discussion of DFT should take the place of ref 15, such as N. Mardirossian and M. Head-Gordon, “Thirty Years of Density Functional Theory,” Mol. Phys. 2017, v115, pp 2315-2372
Question 2 (on references in Table 1 and axis labels in Fig 2c):
The corrections to references seem accurate. The y-axis label for Fig 1c is now ln(Q), but the numerical entries on the y-axis carry physical units so are not exponents (that is, pure numbers).
Question 3 (on the interpretation of data for Q in Fig 2b and 2d):
The authors acknowledge the inverse relation between the profiles for Q and D (or its power of ten, on the log scale) – higher barrier Q is associated with smaller (log) D. Two questions were posed: why do the profiles reach an extreme for half-full d shells, and why do the 3d profiles differ from the 4d and 5d profiles, which are similar? I suspect the explanation must lie in the degree of participation of nd electrons in bonding, which in turn depends on the relative energy of s, p, and d orbitals. (I see some mention of this possibility in the conclusion section, lines 376-382.)
In response to these points the authors have computed the electron localization function (ELF) for Sc and Cd species. But in what way does the ELF plots presented address the questions posed? Surely Ru (with its half-filled 4d shell) should be considered in any analysis.
Fig 2c for Sc and 2d for Cd containing ELF contours for the 010 face seem to show a zone of highly delocalized electron density connecting the vacancy and Sc, but no such feature for Cd. The authors note this difference. I do not understand the sentence beginning “The dissolute of Cd…” which is perhaps intended to explain the distinction.
Question 4 (on the interpretation of data for energetics in Table 2 and Fig 3):
The authors provide a discussion of nucleation theory which expresses the Gibbs energy for nucleation as composed of three components: interface energy, strain, and chemical formation. The complex behavior expressed in Fig 3(b-j) can presumably be traced to the variations in these components. A further complication arises from the choice to report the great bulk of data coming from description of energetics of three faces (001, 110, and 111).
The authors’ treatment of surface energy for Al3TM is analogous in presentation (Tables 3 and 4 and Fig 4 ) and style of discussion to that for the components of nucleation energy. Once again the ELF is shown (Fig 4 e and f) with a brief but useful comment (Lines 356-358), as well as the surface energy, which is shown for three surfaces for each of the TM rows (Fig 4 b-d). The caption needs minor correction.
Question 5 (on the conclusions):
The remarks focus on the data values. Some commentary on the significance of the values would be welcome. For example, mention of the desirable properties of the systems compared with undoped Al (see lines 49-53) would help make clear the accomplishment of this paper.
Comments for author File: Comments.pdf
Language is clear but contains occasional technical errors. Processing with grammar checking software is recommended.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
