Review Reports

Round 1

Reviewer 1 Report

The paper describes a new approach for heterogeneous nucleation of a solid from a liquid on generic inoculant particles. The authors use an MD technique with a large system to investigate, how the liquid rearranges into a solid and can derive from their results a new approach to understand solidification initiated by solid substrates in a melt. Their work is ingenious, performed with well established methods, which are carefully described. They cite all relevant literature and discuss their observations very carefully. They are able for the first time to solve the long standing debate and understanding of heterogeneous nucleation and can derive new ideas, a new picture which helps to understand the action of incolants in metallic melts. Their observation of three different processes, namely first a 3-layer structure on the substrate which helps to overcome the misfit between substrate and growing solid  leading to a 2D nucleus which grows into a hemisphere and shows a distinct different behaviour depend on the size of the substrate and the size of the 2D-nucleus. Their idea of  having the new concept of grain initiation which has to be distinguished from nucleation (and is often mixed in the literature with heterogeneous nucleation itself). This grain intitiation can need an undercooling larger than the free growth undercooling or and undercooling being smaller gives the essential hint that there is either constraint growth or barrier less growth. Especially figure 18 and the final discussion of their wonderful MD simulation shows, that this study and paper is a real step forward to unify the various approaches to so-called heterogeneous nucleation.

 

The paper should be published as it is. It is in all aspects a perfect paper and will stimulate many discussions in the area of solidification and crystal growth.

Author Response

We would like to thank Reviewer 1 for his/her effort to review our manuscript, and for his/her constructive and positive comments.

Reviewer 2 Report

The topic is, in general, very important and interesting. The prenucleation that is demonstrated numerically is the result of a long and persistent activity. There are some issues that need addressing.

1) Equation formatting does not apply to the required standard

2) Equation 1 is flawed and needs correction (Signs)

3) From a historical standpoint, there is a paper by Larson and Garside that is missing related to the interface thickness/roughness suggesting small clusters on the surface (DOI:10.1016/0022-0248(86)90013-8)

4) Similarly on the experimental evidence supporting both ordering and clustering similar to for instance DOI10.1016/j.matlet.2014.06.051 could also be used to substantiate the simulations.

Author Response

We would like to thank Reviewer 2 for his/her effort to review our manuscript. Here are our rely to the comments raised by Reviewer 2:

1. “Equation formatting does not apply to the required standard”.

We have reformatted the equations in the revised manuscript.

2. “Equation 1 is flawed and needs correction (Signs)”.

 It was a typo in Eq. 1 in Line 42, page 1, which has been corrected in the revised manuscript.

3. “From a historical standpoint, there is a paper by Larson and Garside that is missing related to the interface thickness/roughness suggesting small clusters on the surface (DOI:1016/0022-0248(86)90013-8).”

 In the paper, Larson et al suggested that the clusters with sub-critical size in the liquid can make significant contribution to the crystal growth or even heterogeneous nucleation. However, this is not in line with our understanding on atomistic mechanism of heterogeneous nucleation or crystal growth. In other words, the substrate induced atomic ordering at the liquid/substrate interface through structure templating is critical to heterogeneous nucleation, although atomic clusters in the liquid may affect structural templating.

4. “Similarly on the experimental evidence supporting both ordering and clustering similar to for instance DOI1016/j.matlet.2014.06.051 could also be used to substantiate the simulations.”

In this paper, Yang et al reported that an epitaxial layer for heterogeneous nucleation was obtained through a controlled nucleation process. From our best understanding, we don’t believe that this layer with a thickness of 71 nm is relevant to the three-layer nucleation mechanism.