Summary:

In the presented study the ice nucleation properties of two types of kaolin minerals are investigated, which are chemically identical but have different morphologies. While kaolinite forms flat platelets and has a more constrained ice nucleation behaviour (e.g., freezing onset temperature in the range of 243.3 K to 244 K using freshly prepared samples), halloysite has a variety of different morphologies, e.g. tubes, and shows a more diverse ice nucleation activity with ice onset temperatures ranging from 238.2 K to 244.9 K. To better understand the role of morphology, the samples are milled and reveal a clear decrease in the ice nucleation ability of halloysite, while kaolinite samples are rather unaffected. By determining the pore size distributions and pore volumes of the samples before and after milling, it is shown that the halloysite tubes are destroyed, and thus it is suggested that they are likely involved in ice nucleation processes. The authors provide a detailed discussion about the surface type of the mineral causing the ice formation and conclude that hydroxylated particle edges are the most likely location for ice nucleation.

The study is well conceived and I enjoyed reading the manuscript which is generally very well written. I only have minor comments.

General comments:

Do you have suggestions for further studies to test your hypothesis that the hydroxylated edges of the kaolin minerals are causing the ice nucleation, e.g., molecular dynamics studies, or other laboratory studies?

Abstract: You might want to consider mentioning that the milling leads to an increase in specific surface area.

Line 63: An early study by Vonnegut (1947) should be referenced here as well.

Section 2.: It might be helpful to include a figure showing the structure of kaolinite and halloysite.

Lines 175 and 189: Shouldn’t other studies next to Klumpp et al. (2022) be referenced here as well?

Lines 201 to 202: Is there a reason why those halloysite samples were chosen for milling (e.g., the content of impurities)?

Lines 208 to 210: Can you explain in more detail this equation, and also provide uncertainty estimates for your measurements for pore volume distributions?

Line 277: The description of the experiments using ammonia/ammonium is missing in the methods.

Line 357 and following: You might want to consider referencing Fig. 1 here.

Section 4.3: I recommend naming this section slightly differently, to indicate that this is a discussion and not a results section (e.g., “likely location of ice nucleation”).

Figs 6, 7, 10 and 11: Could you indicate the uncertainties in your measurements by error bars?

Technical comments:

Line 3: “1” is missing in the authors’ name for their affiliation.

Figs. 3 and 4: While in Fig. 3 the untreated samples are labeled “pure”, there are not specifically labeled in Fig. 4.

Lines 421 and 423: A bracket is missing at the end of the sentence.

Reference

Vonnegut, B.: The Nucleation of Ice Formation by Silver Iodide, J. Appl. Phys., 18, 593-595, 10.1063/1.1697813, 1947.

Review of the manuscript “Comparing the ice nucleation properties of the kaolin minerals kaolinite and halloysite” by Klumpp et al.  

General comments: The extensive work by Klumpp et al. describes the heterogeneous ice nucleation properties of two mineral dust types, kaolinite and halloysite, that are chemically identical and have the same crystal structure, but differ in their morphologies. Comparison of their freezing properties before and after physical modification by milling showed that the morphology plays an important role affecting the chemistry as well. This study has an important contribution to the ice nucleation field of study and it improves our understanding of why a certain surface produce better ice nucleating sites compare to another. The manuscript is within the scope of ACP. The experiments were well designed, the results are well presented and well interpreted. The introduction is very thorough and focused. I recommend to publish the manuscript in ACP after the authors will address the following comments:

Major comments :

1. The authors suggest the notion of surface hydroxylated edges as ice nucleating sites, however, the current work did not provide a direct evidence for that. So, there is no mechanism which was found and I think this should be highlighted in the conclusion and discussed. Is there scientific way to establish this for example? A way to quantify that?
2. What are the atmospheric implications? kaolin particles are common in the atmosphere and will atmospheric transport or cloud processing can affect its ice nucleating abilities in light of the results of this study?  
3. In many cases along the manuscript the data presented without the uncertainties. Please make sure you report the uncertainties.
4. There is no information about sample preparation and measurement process for surface area in the BET. Degassing was done? At which conditions? How many times each powder was measured?

Minor comments:

1. A figure describes kaolinite and halloysite structures can be included in the text or in the supplementary.
2. Line #71: What is the source of difference between results of H₂O and N₂ surface area and pore size?
3. Line #249: Please detail to which sample variability you refer to?
4. Line #250: Please explain what is the cause for the 1 K difference in the homogeneous freezing temperature between the different measurements.

Technical comments:

Line #83-84: Please provide a reference.

Line #184: For uniformity, notice the use of backslash in temperature change rate here while in the rest of the text there is a use of superscript. 

Fig 2: missing units to the y-axis for Thet.

Fig 4: missing units to the y-axis for Thet.

Fig 8. Missing bracket in the caption.

Fig 9. Missing bracket in the caption.