A New Atomistic Mechanism for Heterogeneous Nucleation in the Systems with Negative Lattice Misfit: Creating a 2D Template for Crystal Growth
Round 1
Reviewer 1 Report
The Paper is focused on the heterogeneous nucleation of Aluminum with negative lattice misfit substrate via MD simulations. Results are interesting to the reader of Metals journal.
Conclusions are clear. Authors simulated Al crystallization on the different templates with negative misfit and generalized their conclusion to other systems. However, only Aluminum was simulated. In the case of other materials, the interaction potential is different and it can be expected that also a mechanism of crystal formation could differ.
It would be mentioned in the summary.
Comments for author File: 
Comments.pdf
Author Response
We would like to thank all the reviewers for their helpful and constructive comments, which we mostly agree and have taken into consideration during our revision. Here is a summary of our responses to their comments.
Response to Reviewer 1
C1) In the summary, it will be appropriate to emphasize that the results are valid for Aluminum.
Response: The reviewer suggests ‘Aluminum has an extremely high growth rate, and the mechanism of crystal formation could differ when the interaction potential is different’. In this paper, we intend to establish the atomistic mechanism of heterogeneous nucleation. As we have stated clearly in Section 2.1 that this simulation system is generic for metallic elements. In general, the mechanism of nucleation as the first stage of the solidification is different from that of the crystal growth. The crystal growth rate is very high for all pure metals (not only for aluminum), being close to the speed of sound. The topic of the crystal growth is beyond the scope of this paper.
C2) Nucleation temperature increases rapidly with negative misfit (Fig. 8). At higher misfit, the homogeneous nucleation could occur. It would be interesting to the reader if homogeneous nucleation was observed in MD simulations.
Response: We did observe that the nucleation undercooling increases rapidly with the magnitude of negative misfit while the misfit is small (|f| < 12.5%, Fig. 8). We had carried out studies on the nucleation for the system with higher lattice misfit, and we will publish this result elsewhere. With increasing lattice misfit, the nucleation behaviour becomes more complicated than just simply extrapolate to homogeneous nucleation, which is beyond the scope of this paper.
Reviewer 2 Report
It is a nice paper to read, clearly described and presented.
Why is equation in line 145 not given as all others in the paper?
I believe there should be a space between numbers and units. Also for °, °C, %,...
What do the empty areas in fig. 4 without atoms, especially in 4a-L1/L0 and L2/L1 down to 4c represent. Are these voids or liquid atoms. It should be explained. If they are voids is it also connected with fig. 5a where L1 never reaches 1?
Markings of C and B are not visible in figure 4, same for fig. 6e. If C represents the edge dislocations, what does B stands for - I haven't noticed the description in the text!
In line 354 the referencing to fig. 6b is wrong. Should it be 3a?
Figure caption for fig. 6 should give the information about time - 50 ps
I am not sure to see the lattice twist in L2 at fig. 10a and also not for 10b. Can you enlarge the picture to better see the presented idea or somehow show this phenomena in a better way? This is the lack for the confirmation of suggested mechanism.
Check the degree sign in line 831
Author Response
Response to Reviewer 2
Q1) Why is equation in line 145 not given as all others in the paper?
Response: We had put this equation as Eq(1) in a separate line, and modified the sequence numbers of other equations in the manuscript.
Q2) I believe there should be a space between numbers and units. Also for °, °C, %.
Response: We checked the publications in metals, and there should be a space between numbers and unit, but this is not applicable to ° and %. These have been corrected in the revised manuscript…
Q3) What do the empty areas in fig. 4 without atoms, especially in 4a-L1/L0 and L2/L1 down to 4c represent. Are these voids or liquid atoms. It should be explained. If they are voids is it also connected with fig. 5a where L1 never reaches 1?
Response: As we stated in Fig. 4 that only solid atoms have been displayed in the simulation system. For clarity, we added ‘where the liquid atoms have been removed from the snapshot’ in the caption of Fig. 4 of the revised manuscript.
Q4) Markings of C and B are not visible in figure 4, same for fig. 6e. If C represents the edge dislocations, what does B stands for - I haven't noticed the description in the text!
Response: We had stated that ‘“C” marks the stacking fault that is surrounded by partial edge dislocations (the solid lines)’ in the caption of Fig. 3. To make it clear, we added ‘“C” and “B” mark the stacking fault that is surrounded by partial dislocations (the solid lines) in L1 and L2, respectively.’ in the caption of Fig. 4.
Q5) In line 354 the referencing to fig. 6b is wrong. Should it be 3a?
Response: we had changed Fig. 6b into Fig. 3a.
Q6) Figure caption for fig. 6 should give the information about time - 50 ps
Response: We added at ‘t = 50 ps’ in the caption of Fig. 6.
Q7) I am not sure to see the lattice twist in L2 at fig. 10a and also not for 10b. Can you enlarge the picture to better see the presented idea or somehow show this phenomena in a better way? This is the lack for the confirmation of suggested mechanism.
Response: Fig. 10 confirms the phenomenon of the twist between the new phase relative to the TiB2 substrate, as we observed in the simulation results. The twist angle is measured by double tilting experiments with HRTEM, and this has been clearly stated in Section3.4. The TiB2 is in the focus and the Al is off the focus due to the twist in Fig. 10a, vice versa in Fig. 10b.
Q8) Check the degree sign in line 831.
Response: The sign is correct for either misfit or temperature in the manuscript.
