This reviewed paper is an accompanying paper of “Soot aerosol from commercial aviation engines are poor ice nucleating particles at cirrus cloud temperatures” (doi: 10.5194/egusphere-2023-2441), which is still under review at the time of submission of these comments. Testa et al., in this follow-up study of aircraft-engine emitted soot aerosol, describe the reactivation ability of contrail-processed soot. The experiments were conducted on the ground with 2 HINC ice nucleation chambers, which simulated two consecutive steps of cloud formation with a sublimation step between them, at a range of cirrus-relevant temperatures and RHi. This study claims to advance the current knowledge in the indirect climate effects of soot emitted from jet engines, which are responsible for anthropogenic direct injection of fresh soot into higher altitude. Using real engines, the authors have demonstrated an improvement of the jet combustion soot simulation, in comparison to previous studies conducted with soot surrogates. The authors conclude that jet-engine cloud-processed combustion-soot in its compact form is poor INP, similar to activation of the unprocessed soot, despite the expected increase in activation via the PCF mechanism, so heavily reported in previous studies. The authors underline the importance of sulfuric acid and organic volatiles presence in the activation process and their inhibiting impact. The paper is well written and structured in a logical way.

I don’t have much comments on the writing or presentation quality but I do have major concerns about the scientific significance and the implied generalized representation of real contrail processes in this and the accompanying paper, both fail to recognize the limitations of such ground experiments even though some of the major uncertainties are mentioned. While such steady flow studies might be relevant for natural cirrus formation, I find there are still major gaps for proper simulation of contrail formation process on the ground.

This accompanying paper focuses on the reactivation however, the evaluation of this sequential study can't be done in isolation and it is impossible to avoid the review of the initial soot generation and sampling methodology. It must be done in order to properly evaluate the conclusions and the usefulness of the presented results for future contrail research.

Essentially this study dismisses the relevance of PCF ice to contrails (many of the earlier studies cited in the manuscript), implying that contrail formation processes shouldn’t be discussed anymore in the context of PCF. This is a strong claim, which requires a deeper substantiation and discussion of the limitations of this study and further clarification on the applicability of this study to actual contrails. To demonstrate the relevance of this study to high altitude contrails, the authors should have a more detailed description of timescales and rates of processes in their experiment with a comparison to rates known from real contrails.

Below I listed some major comments, unanswered questions, and some minor comments. Thus, after a major revision is complete, I’d recommend this manuscript for publication. Please see my comments below:

Major comments:

1. The current design involves a stirred tank for offline characterization with isokinetic flow instruments and filter sample collection. I think the experiments could have been designed better to have a real-time sampling, similar to what was done by Korhonen et al. 2022 however, this would not solve all the problems either. Additionally, I think expansion chambers  (e.g. Crawford et al. 2011). might be more representative of the actual dynamic process in contrails. For the current ground setup, there should at least pictures, simulation of the exhaust plume,description of the ambient conditions- was the experiment conducted outdoors, wind/temperature/humidity information, was the heat-resistant alloy probe, located ∼1 m downstream of the engine exhaust nozzle, sampling from the middle of the plume, at an angle, etc.
2. Line 42-43: The gap between the exhaust and the inlet, introducing dilution and turbulence: would it impact the coagulation in the tank? would a higher concentration sampled directly from the engine have resulted in a faster coagulation, a different size mode? if so are you properly simulating the contrail formation process?
3. The aforementioned meter gap between the engine exhaust and the sampling inlet: how injection of ambient air downstream of jet fuel combustion would impact soot emissions, concentrations, oxidation etc. I would like to see a discussion referring to findings reported by Kelesidis et al. 2023.
4. What was the temperature of air intake by the engine? Would room temperature air intake affect the combustion products? The characteristics of soot particles formed? How different would it be in comparison to intake of -60C cold air?
5. Given the fact that the engines tested had differences between the produced soot, can the results of this study be generalized to the extent expressed in the title?
6. Sulfuric acid was mentioned as one of the main factors limiting the PCF. This is more obvious in a long and slow cooling process that allows for more efficient condensation. What about an instant temperature and significant instant pressure drop between the combustor and the high-altitude environment, would that impact the uniformity of sulfuric acid vapors condensation? The boiling point of sulfuric acid is 166 C, would thermodynamics and cooling rate in your experiment, dictate the condensation of sulfuric acid and impact its degree of soot surface coating, which in turn impacts the reactivation capacity due to the chemical content present in the voids?
7. The dynamics, rates of processes are extremely important, dropping the particle and air temperature from thousands of degrees, to hundreds and then to subzero temperature during tens of minutes to several hours is not the same as an almost instantaneous transition from thousands of degrees to -60 C, would soot particle temperature be different than its carrying air temperature, would heat dissipation be slower at low pressure, would that impact INP activity and how would that compare to this slow ground test? Moreover, with the isokinetic flow into the cloud chamber slowly transitioning from room temperature around the inlet of CFDC into -60 C, would you expect glass transition of some of the coatings at higher cooling rates in the atmosphere, would that impact the INP activation? good to discuss the differences.
8. The importance of surface properties is discussed e.g., hydrophilic surface oxygenated functionalities, polar groups etc. What’s the impact on surface properties of soot aerosol under increased ozone and UV-b, UV-a conditions once emitted at high altitude in comparison to this ground experiment? would proper simulation with exposure to ozone and UV-b would change the results of this experiment? After contrail sublimation multiple things can happen to the soot particles, depending on the timespan, but even in the case of almost immediate reactivation, would ozone and UV change surface properties and the results of the next activation cycle?
9. On the same topic of surfaces charges, which seem to play a role in the ice nucleation process. Charging is often observed in expansion chambers experiments (e.g., AIDA). I haven’t seen any discussion about charging and charge transfer of soot in the exhaust ejection process, would it play a role in measured INP activity in comparison to long mixing in the tank on the ground and isokinetic low flow aspiration with HINC?
10. Is there any impact of exhaust inhomogeneity and selected sampling inlet location/orientation?
11. Does this setup represent well the impact of shear forces trapping the exhaust in vortices behind the aircraft, would this impact the results? could the vortices create multiple sharp supersaturation and sublimation cycles? How does that compare to the timespan of this experiment?
12. A key achievement in this study (e.g., lines 31-33,224-225 and elsewhere) is the production of jet engine soot aerosol that unlike previously studied BC proxies is more representative of the contrail ice nucleation followed by “contrail” sublimation and reactivation. While the combusted jet fuel better represents the real content of the exhaust, anything downstream the exhaust is still significantly different on many dimensions in comparison to the physics and thermodynamics of high altitude. The author’s claim of a greater relevance needs to be further substantiated, alternatively the claims should be toned down (as should be the accompanying paper) on its relevance to actual contrails or discuss in greater breadth the limitations of this study. These questions have to be addressed before the second step of contrail processing can be properly evaluated.
13. Was the CATZ time scale representative of contrail sublimation rate? Do you think liquid/solid phase chemistry could play a role here, in the context of sulfuric acid dilution?
14. The paragraphs and sentences dedicated to hypotheses about alternative jet fuel activation e.g., in the atmospheric implications section. In my opinion, these should be avoided. There is no need to hypothesize about experiments that have not been done, especially given the questions remaining about the applicability of the presented ground test results to real contrails. I’d recommend focusing on the applicability and substantiation of the results of this study with the fuels and engines actually tested.

Minor comments:

1. “Contrail processed….” – the title is misleading, the ground/laboratory experiment doesn’t cover the full complexity of the contrail formation process at high altitude nor does it simulate the transition rates in a “violent” contrail formation but rather presents a cloud formation in a steady state environment, feeding well stirred combusted soot at room temperature into an ice nucleation chamber. Please come up with a title that properly describes the ground simulation that you present.
2. Figure 1 caption: “incomplete combustion… emits” - combustion produces, or aircraft emits
3. Line 14: “and they exhibit” – they hint on high likelihood of poor ice nucleation ability at cirrus relevant….
4. Line 28: the accompanying study (not published) states 2-50 nm. Which one is correct?
5. Figure 1 caption: “forming a contrail cloud due to high concentration of water vapor and cold temperatures” – in the aerosol reservoir depicted, aren’t you removing water vapor down to less than 10%? How does that correspond to the process in the atmosphere?
6. Line 151-152, similarly compacted regardless of their size: would a higher sublimation rate cause a greater compaction or was the maximum compaction reached?
7. Same for line 159, was maximum compaction reached?
8. Figure 5: a colorbar should be included in the main plot and x axis titles on the adjacent box plots.
9. Line 223 unlikely to promote ice nucleation…
10. Line245-259, see major comment 14, this doesn’t fit into atmospheric implications of the current study. It looks more like a discussion paragraph about future work that can be done. I would personally prefer to see a future work discussion that will present design insights from this study that can be modified to improve the simulation or guidelines to build a future ground facility for more accurate simulations of contrail formation. Alternativly, in this section there could be more discussion about the impact of the results on radiative transfer (depicted in Figure 1), what is the difference in radiative transfer between what was previously assumed and with the new findings.
11. Line 268 strong compaction (65% convexity increase) – is that the max compaction that could be achieved? What would be the typical difference between a strong and a weak compaction?
12. Line 271: In real contrails, would this condensation be impacted by the rate of the process, would the phase of the organic matter be impacted by the rate of cooling e.g. Zhang et al. 2019. Would that in turn influence the INP activity?

References:

• Korhonen, K., Kristensen, T. B., Falk, J., Malmborg, V. B., Eriksson, A., Gren, L., Novakovic, M., Shamun, S., Karjalainen, P., Markkula, L., Pagels, J., Svenningsson, B., Tunér, M., Komppula, M., Laaksonen, A., and Virtanen, A.: Particle emissions from a modern heavy-duty diesel engine as ice nuclei in immersion freezing mode: a laboratory study on fossil and renewable fuels, Atmos. Chem. Phys., 22, 1615–1631, https://doi.org/10.5194/acp-22-1615-2022, 2022.
• Crawford, I., Möhler, O., Schnaiter, M., Saathoff, H., Liu, D., McMeeking, G., Linke, C., Flynn, M., Bower, K. N., Connolly, P. J., Gallagher, M. W., and Coe, H.: Studies of propane flame soot acting as heterogeneous ice nuclei in conjunction with single particle soot photometer measurements, Atmos. Chem. Phys., 11, 9549–9561, https://doi.org/10.5194/acp-11-9549-2011, 2011.
• Georgios A. Kelesidis, Amogh Nagarkar, Una Trivanovic, and Sotiris E. Pratsinis, Environmental Science & Technology 2023 57 (28), 10276-10283, DOI: 10.1021/acs.est.3c01048
• Yue Zhang, Leonid Nichman, Peyton Spencer, Jason I. Jung, Andrew Lee, Brian K. Heffernan, Avram Gold, Zhenfa Zhang, Yuzhi Chen, Manjula R. Canagaratna, John T. Jayne, Douglas R. Worsnop, Timothy B. Onasch, Jason D. Surratt, David Chandler, Paul Davidovits, and Charles E. Kolb, Environmental Science & Technology 2019 53 (21), 12366-12378, DOI: 10.1021/acs.est.9b03317

In this study, Testa et al. investigate ice nucleation abilities of aerosols emitted by six different commercial aircraft engines fueled with Jet A-1 and running while on the ground. The authors further subject the aircraft-engine emitted soot aerosols to catalytic stripping and/or cloud chamber processing to simulate contrail clouds conditions and evaluate their impact on the aerosols’ ice nucleation activity at temperatures relevant for cirrus formation. The authors investigate the effects of H₂SO₄ and volatile organic coating, the effect of morphology and compaction of soot aggregates on their ice nucleation ability. The authors conclude that H₂SO₄ and volatile organic coatings inhibit ice nucleation in preventing PCF to occur and that the compaction observed after “contrail-like processing” remained inefficient at promoting PCF, which makes aircraft-engine emitted unprocessed and processed soot unlikely INPs.

The manuscript is of good quality and the study relevant. However, the manuscript would benefit from a clear assessment of its potential limitations with respect to its representativity of real contrail processing conditions and thus cannot be generalized to the extent expressed in the title. I would therefore recommend this manuscript for publication after this major concern has been addressed, particularly in the experimental section.

Some of the comments and questions below can help address this concern.

I fell the title is misleading as the aerosols were flown through experimental chambers to simulate contrail processing but only to some extent since the experimental conditions were still far from what would happen in real conditions (e.g., high-altitude temperature, pressure, background gas or aerosols, versus ground). Maybe change to a humbler title?

Line 40: several engines’ models from Pratt&Whitney and CFM international fueled with Jet A-1 were used to produce soot particles. How are these engines representative of the fleet? What about the fuel? Since their ice nucleation response can be different (as shown in this manuscript), it may be relevant to get an idea from the get go. Please specify.

Line 42: In which medium were the aerosols collected 1m downstream the exhaust nozzle, ambient air and room temperature? Do you expect this difference with real conditions to affect aerosols’ properties, if yes how so?

Line 43: the exhaust is directed to an aerosol reservoir; do you expect any change in the aerosols’ mixing state in the tank? Is this directly comparable to what happens in real conditions? What is the temperature in the tank? Are the air composition/energy carriers (e.g. UV photons) simulated the same way as in real conditions? This should be discussed so the reader can understand the potential limitations.

Line 49-51: how long do the aerosols remain in the first cloud chamber? In the subsaturated flow tube? Do these residence times have an influence?

Line 64: what is the efficiency of the catalytic stripper, it is not specified?

Line 70: do you expect any sampling-induced change in morphology when collecting particles on TEM grids? This should be discussed as results interpretations are based on this morphology analysis.

Line 106: “trigger modest PCF at 5% RHi […]”: Why is this qualifier used? Modest compared to what? The nucleation onset is reached at RHi below RHhom, so it does trigger PCF, doesn’t? If it is implicitly compared to other potential INPs, please make it explicit.

Line 112: Were all the engines used for each sample type to derive the results shown in Figure 2? It is not specified.

Line 125: “for all engines, the soot aggregate mass increases for given sizes”. However, it seems that the error bars displayed in Figure 3b prevent any comparison between CP soot and unprocessed soot as the error bars fully overlap. Same goes for the distinction between 150 nm particles and smaller ones regarding their compaction upon “contrail processing”. The interpretations should be moderated accordingly.

Line 139-141: here again comparisons are done between processed and unprocessed soot for different engines but the stated results (e.g., ΔDm is larger for CP soot compared to CS CP soot) rely on differences that fall within measurement uncertainties (clearly said this time). More caution should be taken in such case and the interpretations should be moderated accordingly.

Line 148: “[…] for all investigated engine types, which is indicative of aggregate compaction”. What about the previous limitation for particles below 150 nm that showed no compaction in mass measurements; TEM results do not show such limitation?

Line 162-163: “CS-CP soot undergo […] than CP soot upon processing”. Which processing? Both samples have already been processed. Please clarify/reword.

Line 160-169: “CS-CP-soot show morphology changes similar to CP soot” […] CS-CP-soot undergo smaller size reduction than CP-soot upon processing” […] “CS-CP soot are on average more hydrophobic. This is due to the removal of H₂SO₄ upon catalytic stripping”. I do not understand what parameter is preponderant here to explain why CS-CP-soot need the lowest %RHi to activate ice nucleation compared to CP-soot (Fig 2), is it because of the pore size, hydrophobic character and/or H2SO4 removal? Could you please clarify?

Line 177: how is “the inability of CP-soot to promote ice nucleation […] contrail processing does not generate pores relevant to PCF” similar to observing moderate enhancement for contrail-processed propane soot. Please remove “similarly” or reword.

Line 182: “CFM56-7B“, should it be “CFM56-7B26/3” instead?

Line 203: possess

Line 223: again the very strong statement “aviation soot does not promote ice nucleation below […] RH_hom” tends to generalize this to all engines, while in the same paper (Testa et al. 2023) a subset of engines (2/10) were shown to activate ice via PCF when large soot aggregates were emitted. The statement needs to be moderated not to elude this result.