the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 30 Apr 2026
Ice nucleating properties of α-pinene- and limonene-derived secondary organic aerosol under cirrus conditions
Abstract. The contribution of biogenic secondary organic aerosol (BSOA) to cirrus cloud formation remains unresolved, contributing to uncertainty in aerosol-cloud interactions in global climate models. Laboratory studies report highly variable ice nucleating efficiencies for BSOA, suggesting that these particles may act as either homogeneous or moderately effective heterogeneous ice nuclei. Here, we investigate the deposition ice nucleating properties of α-pinene- and limonene-derived BSOA, including both self-nucleated particles and BSOA coatings on ammonium sulfate and ammonium bisulfate seed particles. Deposition ice nucleation relevant to cirrus clouds (−45 °C, −40 °C, −35 °C; 1.0 ≤ Sice ≤ 1.6) was measured using the SPectrometer for Ice Nucleation (SPIN). Bulk physicochemical properties relevant to ice nucleation were characterized using aerosol mass spectrometry (AMS) and volatility distributions. Pre-cooling was applied to modulate phase state as inferred from glass transition temperature (Tg).
BSOA ice nucleating properties were strongly precursor dependent (p < 0.001). Tg was an unreliable predictor of freezing behavior, correctly anticipating freezing mode for only two of eleven particle combinations. Limonene-derived BSOA nucleated ice almost exclusively via heterogeneous freezing, with Sice onsets as low as 1.27±0.07 at -39.8±0.3 °C. α-pinene-derived BSOA predominantly nucleated ice homogeneously. BSOA coatings on ammonium bisulfate shifted freezing from homogeneous to heterogeneous, while the role of acid-catalyzed multiphase chemistry in ice nucleation remained inconclusive due to experimental limitations. These results demonstrate that cirrus-relevant BSOA parameterizations must explicitly account for precursor specific chemistry and broad classifications of BSOA ice nucleating abilities are inappropriate.
Competing interests: At least one of the (co-)authors is a member of the editorial board of Atmospheric Chemistry and Physics.
Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.- Preprint
(2185 KB) - Metadata XML
-
Supplement
(1432 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Reviewer comment', Anonymous Referee #1, 26 May 2026
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2026/egusphere-2026-2458/egusphere-2026-2458-RC1-supplement.pdfCitation: https://doi.org/
10.5194/egusphere-2026-2458-RC1 -
RC2: 'Comment on egusphere-2026-2458', Anonymous Referee #2, 17 Jun 2026
Review of the manuscript “Ice nucleating properties of α-pinene- and limonene-derived secondary organic aerosol under cirrus conditions” by C.N. Rapp et al.
Rapp et al. present laboratory measurements of the ice nucleation (IN) ability at cirrus conditions of aerosol particles composed of self-nucleated α-pinene (AP) and limonene (LIM) secondary organic aerosol (SOA), and of ammonium sulphate (AS) or partially neutralized sulphate (PNS) particles coated with AP or LIM. The experiments probe the role of the precursors' physicochemical properties, their estimated glass transition temperature (Tg), multiphase chemistry, and precooling.
The manuscript is well organized and clear to follow. The research hypotheses are well described and addressed. IN of SOA remains a genuinely difficult topic where instrument accuracy and reproducibility are compounded by complex, hard-to-constrain aerosol systems, and this study provides novel measurements that improve our understanding of the IN ability of SOA.
I recommend publication in ACP after the authors address the comments below.
General comments and questions
- In Section 2.6.3 you classify a freezing onset as heterogeneous when it falls below the lower bound of the homogeneous freezing region. However, at several points you note that even when the onset lies in the heterogenous regime, its uncertainty extends well into the homogeneous conditions (e.g. AP-PNS at -40°C). This makes it ambiguous whether such particles should be regarded as heterogeneous nuclei, poor heterogeneous nuclei, or effectively homogeneous. Is there additional evidence that could help resolve the freezing mode for points near the homogeneous threshold? In particular, the shape of the activation fraction spectrum may be informative: heterogeneous freezing tends to produce a gradual increase in AF with Sice, whereas homogeneous freezing tends to produce a sharp, threshold-like rise. The AF spectra in Figures S3 and S4 could be examined on this basis.
- The methods used to estimate the glass transition temperature (Tg) need a fuller description even with references to prior work. Specifically, I think it is important to show the relations between Tg and C* and Tg and molecular composition. How do errors in C* and bulk composition propagate into the Tg estimates? The uncertainties associated with the thermogram-derived Tg (Table 2) likely don’t reflect the variability of the parameterization and therefore underestimate the true error.
Are there multiple Tg parameterizations in the literature? Do they agree? What motivations brought you to use the two representations presented in the paper and not for example one based on the functional groups?
The estimated Tg for the pure and SOA-PNS systems are very different (Table 2), but the difference is considerably more pronounced in the bulk composition-derived parameterized Tg than for the thermogram-derived one. The bulk composition is very constant across the different systems, what then explains the difference?
For AP-PNS experiments, you mention that ice nucleation onset was preceded by an onset of water uptake. How can you unambiguously assign the morphology to a core-shell type and not to a homogeneously mixed particle? This leads to two additional questions: how can you assess that the PNS seeds had crystallized after generation? If not, shouldn’t you consider a glass transition temperature for the mixture?
Across different parts in the manuscript I’d soften the assertion that Tg (generally) fails to predict ice nucleation. I suggest instead saying that the chosen parameterizations applied to your data do not show it (whether this generalizes to all systems and all Tg parameterizations it is not shown in the manuscript). -
Referring to the seeds as ammonium bisulphate (ABS) in some parts of the manuscript is confusing, since by your own account only the first experiment with fresh ABS can be considered "true" ABS. For example, the coatings described in the abstract (L15) were not on ABS. Because you have AMS data, I would suggest labelling all the non-AS seeds as PNS consistently throughout and stating the neutralization degree explicitly for each experiment, including the AS, fresh ABS, and PNS cases.
-
Changes in aerosol concentration and temperature can shift the gas-particle partitioning of the organic components. Can you rule out changes in chemical composition between the inlets of the different instruments, or during the precooling phase of the IN measurements? Did you try measuring the aerosol composition before and after the PCU?
-
Can PALMS be used to say something about the distribution of the coating across the seed population?
-
Because you compare single-component and mixed particles with different size distributions, why not adopt an INAS-density framework for the heterogeneously freezing systems, which would normalize for surface area?
Please also clarify the D > 100 nm cut used in the AF denominator: how sensitive is the AF of the LIM experiments to this threshold? Given the low 0.5% AF threshold, could the small but efficient self-nucleated limonene fraction alone exceed it and account for the onset attributed to the coated seeds in the mixed experiments? -
Given the high variability across IN counters, more detail on the measurement strategy is needed. Please report (i) how often filtered, particle-free air was sampled during each measurement; (ii) the background counts for each measurement; (iii) the IN activity measured on the descending Sice branch (if the descending ramp returns to ice saturation, L303 should read “ice saturation” rather than “water saturation”); (iv) how often the chamber walls were re-iced or the ice layer refreshed; and (v) example ramps and ice onsets for the ammonium nitrate calibration experiments.
-
Section 4 (Experimental uncertainties) reads as a qualitative description of methodological limitations, much of which would be better integrated as concise uncertainty statements within the relevant sections.
Specific comments
L113 : Please add Bertozzi et al. (2021) on organic coatings of seed particles (relevant also to Figure 5).
Bertozzi B, Wagner R, Song J, Höhler K, Pfeifer J, Saathoff H, et al. Ice nucleation ability of ammonium sulfate aerosol particles internally mixed with secondary organics. Atmospheric Chem Phys. 2021 Jul 16;21(13):10779–98. doi:10.5194/acp-21-10779-2021L136 : This appears to be the only sentence describing how coating thickness was evaluated; it needs more description in the methods/aerosol-generation section. Did you compare the coating thickness for the different modes? Did you compare to the coating estimated from the AMS data assuming a core-shell morphology?
L220 : From the size-distribution plots it looks like there is an additional neutralizer in front of the SEMS. If this is correct, please mention it in the description.
L243 : Please define “asynchronously” specifying the time elapsed between the PALMS and IN measurements.
L271 : Citation format error.
L357 : 13% of particles are doublets and your ice onset is at <1%. On what basis can you exclude that these activated particles originate from the 13% doublets, i.e., that the freezing is driven by the larger seeds?
L374 : How can you establish core–shell morphology rather than a homogeneously mixed particle?
L380 : You state that no clear delineation between pre-cooling conditions was observed for the self-nucleated and LIM-PNS experiments. However, the LIM-PNS points at −45°C (Fig. 3) appear to sit very close to the homogeneous freezing region (comparable to the AP-AS experiments you classify as homogeneous) raising the same classification ambiguity discussed above. They also appear to show pre-cooling shifting the onset to higher Sice, opposite to the trend in all other systems and to the summary at L734. Can you provide a suggested explanation?
L390 : Why did you restrict the PALMS-NG analysis to the singlet particles and exclude the larger ones that could also contribute to the IN signal?
L448 / L689 : You report p = 0.28 at L448 and p = 0.59 at L689 for the seed-and-precursor combination, please reconcile.
L465 : Do you have AMS data to confirm the degree of neutralization?
L605 : As written, this reads as though you measured Tg directly, please clarify.
L647 : If LIM shows no water uptake but its glass transition is ~−70°C, what is the phase state of the particles?
L729 : Please reframe specifying which Tg parameterizations fail to predict IN behaviour.
L742 : Please clarify why the simultaneous presence of self-nucleated particles affects the multiphase-chemistry result specifically.
Figures
Figure 2 : An example of size distributions from a LIM self-nucleated and LIM-coating experiment would be helpful.
Figures 4 and 5 : Figures 4–6 are cited out of numerical order and are minimally described in the main text. Please cite them in sequence and describe their content/significance in the context.
Figure S1 : What does "heterogeneous AF" mean? Does it cover AF for Sice from 1.0 to 1.6? Are the AF values already background-subtracted? If so, how do you explain AF > 0 at Sice = 1.0? Does AF > 1 indicate a miscounted ice background?
Citation: https://doi.org/10.5194/egusphere-2026-2458-RC2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 296 | 80 | 14 | 390 | 31 | 17 | 19 |
- HTML: 296
- PDF: 80
- XML: 14
- Total: 390
- Supplement: 31
- BibTeX: 17
- EndNote: 19
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1