Analysis of hygroscopic cloud seeding materials using the Korea Cloud Physics Experiment Chamber (K-CPEC): A case study for powder-type sodium chloride and calcium chloride

Kim, Bu-Yo; Belorid, Miloslav; Cha, Joo Wan; Kim, Youngmi; Kim, Seungbum

doi:https://doi.org/10.5194/egusphere-2025-515

Preprints

https://doi.org/10.5194/egusphere-2025-515

Preprints

25 Apr 2025

| 25 Apr 2025

Status: this preprint is open for discussion and under review for Atmospheric Measurement Techniques (AMT).

Analysis of hygroscopic cloud seeding materials using the Korea Cloud Physics Experiment Chamber (K-CPEC): A case study for powder-type sodium chloride and calcium chloride

Bu-Yo Kim, Miloslav Belorid, Joo Wan Cha, Youngmi Kim, and Seungbum Kim

Abstract. In this study, we analyzed the particle characteristics and cloud droplet growth properties of NaCl and CaCl₂, which are powder-type hygroscopic materials applied in cloud seeding experiments, using the Korea Cloud Physics Experiment Chamber (K-CPEC) facility at the Korea Meteorological Administration/National Institute of Meteorological Sciences (KMA/NIMS) in South Korea. The aerosol chamber (volume 28.3 m³) enabled the observation of particle characteristics in an extremely dry environment (relative humidity (RH) < 1 %) that was clean enough to ignore the influence of background aerosols. The cloud chamber featured a double-structure design, with an outer (130 m³) and inner (22.4 m³) chamber. The inner chamber allowed the precise control of air pressure (1013.25–30 hPa) and wall temperature (–70–60 °C), facilitating cloud droplet growth through quasi-adiabatic expansion. In this study, a cloud chamber experiment was conducted to simulate both wet adiabatic and stable environmental lapse rate conditions. The experiments were initiated at low RH (< 60 %), and the variations in the cloud droplet concentration and diameter were observed as RH increased, leading to supersaturation (RH > 100 %) and subsequent cloud droplet formation. NaCl and CaCl₂ powders showed distinct particle growth behaviors owing to the differences in their deliquescence and hygroscopicity. The rate of cloud droplet formation in the NaCl powder experiments was slower than that for CaCl₂; however, the mean and maximum droplet diameters were approximately 2–3 μm and 10–20 μm larger, respectively. The droplet diameter varied from 1 to 90 μm, and large cloud droplets (30–50 μm) that served as the basis for drizzle embryo formation were also observed. Our study provides valuable insights for the development of new seeding materials and advanced cloud seeding experiments.

Received: 04 Feb 2025 – Discussion started: 25 Apr 2025

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 preprint. The responsibility to include appropriate place names lies with the authors.

Download & links

Bu-Yo Kim, Miloslav Belorid, Joo Wan Cha, Youngmi Kim, and Seungbum Kim

Status: open (until 01 Jun 2025)

Post a comment Subscribe to comment alert

CC1: 'Comment on egusphere-2025-515', Chao Peng, 27 Apr 2025 reply

I would like to bring into the authors' attention one paper which I co-author. It provided comprehensive analysis of hygroscopic properties and CCN activities of CaCl2.

Guo, L. Y., Gu, W. J., Peng, C., Wang, W. G., Li, Y. J., Zong, T. M., Tang, Y. J., Wu, Z. J., Lin, Q. H., Ge, M. F., Zhang, G. H., Hu, M., Bi, X. H., Wang, X. M., and Tang, M. J.: A comprehensive study of hygroscopic properties of calcium- and magnesium-containing salts: implication for hygroscopicity of mineral dust and sea salt aerosols, Atmos. Chem. Phys., 19, 2115-2133, 2019.

Reply

Citation: https://doi.org/10.5194/egusphere-2025-515-CC1
RC1:
'Comment on egusphere-2025-515', Anonymous Referee #1, 06 May 2025 reply

In the manuscript submitted by Kim et al., the authors built a Korea Cloud Physics Experiment Chamber and studied the hygroscopicity and cloud activation ability of two common salts used in warm cloud seeding experiments. I suggest this paper fits well within the scope of AMT. Nevertheless, I have a few comments for the authors before its acceptance for publication.

Major comments:

1. In the introduction section, the author highlighted the importance of cloud seeding in precipitation production. However, the outcomes of cloud seeding are not always positive and there is more literature reporting that cloud seeding fails to produce precipitation efficiently. The authors should also point this out and summarize the related studies.

2. The authors tested two types of powder, NaCl and CaCl2. However, they didn’t claim the motivation to use the two materials in warm cloud seeding. What are the pros and cons of using these materials compared to other materials?

3. One of the main conclusions from the authors' findings is that “The CaCl2 powder, with strong deliquescence, exhibited a high cloud condensation nuclei (CCN) activation fraction in low-RH conditions; the NaCl powder, with high hygroscopicity, produced larger cloud droplets under supersaturation conditions.” It is important to note that the two types of particles tested in this study differ in size, which significantly influences CCN activation and hygroscopicity. As a result, the comparison may not be an 'apple-to-apple' case. The authors may consider using alternative metrics instead of Fact to quantify the hygroscopicity of the produced aerosol, such as the kappa value, which accounts for particle size.

4. It is unclear how the authors define CCN in their cloud chamber experiments. For example, in Line 304, they stated that “Immediately after the beginning of the experiment, approximately 56 % of the number concentration was measured as CCNs (D > 0.3 μm) compared to the injected total number concentration under conditions of RH 55 %. ”. Did the authors define particles that can grow beyond 0.3 µm as CCNs? If so, what is the rationale behind this definition?

5. One of the major conclusions, “The CaCl2 powder, with strong deliquescence, exhibited a high cloud condensation nuclei (CCN) activation fraction in low-RH conditions” is not well supported. Based on aerosol chamber experiments, the Fact of CaCl2 was higher than NaCl at S of 0.1%. This means CaCl2 also has a stronger CCN activation ability at supersaturation conditions, not only at low RH conditions.

Also, the authors seem to equate the hygroscopic behavior of CaCl₂ at RH < 100% with CCN activation. However, CCN activation typically refers to particle growth under supersaturation (RH > 100%), whereas CaCl₂ undergoes deliquescence at lower RH levels. Could the authors clarify this distinction? For example, in Line 305: “This result contrasts with the higher Fact of approximately 77 % measured under S = 0.1 % condition, as fewer particles were activated as CCNs in low RH conditions. ”

6. The authors also demonstrated that “as the air in the inner chamber was evacuated using a vacuum pump, the CCNs or droplets in the chamber may have been lost”. Have the authors conducted experiments to quantify wall losses in aerosol and cloud chambers?

Specific comments:

1. What do authors mean by “until supersaturation (RH>100%) and droplet formation” (L15).

2. Quantitative results should be given in the abstract. For example, “large cloud droplets that served as…”; “low RH-conditions”; “supersaturation conditions”.

3. Section 2.2. Can authors comment on the homogeneity of temperature within their aerosol and cloud chambers? As the temperature of the chamber is only reported from few thermal couples located at four different places, would these temperature uncertainties cause the bias of the supersaturation calculation? How much would this be?

4. L169: Can authors briefly introduce this air jet milling process?

5. L191: Can authors schematically show the location of the four thermocouples in the inner chamber?

6. L211: Does the “remaining number concentration of aerosol” refer to aerosol remaining from the last cleaning procedure or last chamber experiments? The “experiment was performed once more” means the cleaning procedure will be performed once more, or the chamber experiment?

7. L225: step 5. Was the aerosol in the cloud chamber injected from the aerosol chamber or from the aerosol generator?

8. Section 4.1: SMPS measures the mobility diameter, while OPC gives the optical diameter of aerosol particles. Do authors assume they are the same, or how do they merge them? Please clarify this.

9. L255: The particle size in real-world seeding events can be smaller also because the greater coagulation process within the chamber experiments due to lower wind speed.

You can also cite: Critical Size of Silver Iodide Containing Glaciogenic Cloud Seeding Particles

Jie Chen, Carolin Rösch, Michael Rösch, Aleksei Shilin, and Zamin A. Kanji

10. L270: What's the critical size for NaCl to activate at S of 0.1%? Can you find this size in the literature to validate your conclusion?

11. Was the aerosol chamber experiment for each particle type conducted only once? If so, do the authors have repeated measurements to confirm the reproducibility of their results?

Reply

Citation: https://doi.org/10.5194/egusphere-2025-515-RC1
- EC1: 'Editor's comment', Mingjin Tang, 06 May 2025 reply
  
  Ref #1 contacted me (the handling editor) and told that the authors have addressed the comments he/she raised during the access review; as a result, these comments can be ignored.
  I also invited ref #1 to review the online preprint, and he/she will provide further review. I would like to thank the authors and ref #1 for their support!
  Mingjin Tang
  Handling editor
  
  Reply
  
  Citation: https://doi.org/10.5194/egusphere-2025-515-EC1
RC2:
'Comment on egusphere-2025-515', Anonymous Referee #3, 12 May 2025 reply
Summary:
This study analyzes the droplet growth and cloud formation properties of two hygroscopic compounds, NaCl and CaCl₂. Specifically, both compounds were analyzed for its application to cloud seeding in warm clouds. The authors conducted both aerosol experiments (CPC, CCNC) and cloud chamber measurements using the Korea Cloud Physics Experiment Chamber (K-CPEC). The authors observed smaller NaCl particles compared to CaCl₂ and greater growth for NaCl due to greater hygroscopic behavior. The authors concluded that cloud seeding for the analyzed compounds should be done in under more supersaturated environments to increase the fraction of CCN activation. This work provides greater insight into compounds' ability to help cloud formation using experimental techniques simulating atmospheric conditions. The results of this paper have greater implications for further cloud seeding performance and highlights the use of the K-CPEC for such studies. As a result, this work is well suited for AMT and should be published. A few clarifying questions are brought up before final publication:
Specific Questions:
Line 166: I understand that temperature was hard to control, but is there a temperature range that can be provided?

Line 170-174: Was a calibration performed for the CCNC ? Past studies have calibrated the CCNC using ammonium sulfate (Rose et al., 2008) - how were the instrument SS verified as being close to the SS input into the program? If a calibration was performed, please clarify in the text and put calibration results in an SI.

Line 195-198: It seems as though the N_CCN/N_CN results were obtained from scanning mobility CCN analysis (SMCA) method (scanning through range of diameters using SMPS then getting ratio of the distribution) as opposed to a stepping mode method (keeping diameter constant and varying SS% in CCN) - is this correct? If so, please clarify in the text, the authors can also cite Moore et al., 2010.

General questions/comments:
Line 124-125: Does having openings cause for any additional contamination/compounds other than the tested ones to enter the chamber and influence results?

Line 358: Authors state the degree of supersaturation can not be calculated effectively due to condensation - does this have major implications on the results of the cloud chamber (e.g., SS being 0.1% instead of 0.2%)? Are there ways to improve upon this with future work? If so, can the authors address this as a future work/implication in the conclusions

References:
Rose, D., Gunthe, S. S., Mikhailov, E., Frank, G. P., Dusek, U., Andreae, M. O., & Pöschl, U. (2008). Calibration and measurement uncertainties of a continuous-flow cloud condensation nuclei counter (DMT-CCNC): CCN activation of ammonium sulfate and sodium chloride aerosol particles in theory and experiment. Atmos. Chem. Phys., 8(5), 1153-1179. https://doi.org/10.5194/acp-8-1153-2008
Moore, R. H., Nenes, A., & Medina, J. (2010). Scanning Mobility CCN Analysis—A Method for Fast Measurements of Size-Resolved CCN Distributions and Activation Kinetics. Aerosol Science and Technology, 44(10), 861-871. https://doi.org/10.1080/02786826.2010.498715

Reply
Citation: https://doi.org/10.5194/egusphere-2025-515-RC2
RC3: 'Comment on egusphere-2025-515', Anonymous Referee #2, 16 May 2025 reply

This manuscript describes the Korea Cloud Physics Experiment Chamber (K-CPEC) as a new cloud chamber facility for developing cloud seeding technology and investigating on aerosol-cloud Interactions. The chamber is equipped with a promising set of instruments to characterize the CCN activation ability of aerosol particles and the microphysical processes with regard to cloud droplet activation. The authors present how the aerosol and the cloud chambers can be used to study warm clouds by testing powder-type hygroscopic materials applied in cloud seeding experiments. Since the deviations from a ideal expansion process for both air pressure and temperature may affect CCN activation and cloud droplet growth inside the inner chamber, further consideration will be needed to evaluate the performance of these concept experiments and discuss on validating the results.
In the aerosol chamber experiments on the characteristics of NaCl and CaCl2 powders, the effects of PSD should be considered. It is speculated that the difference in the activation fraction of CCNs at low supersaturation (0.1 %) seems to be due to the difference in mode diameter. The gaps in the activation diameter between two sample powders was quite small over the supersaturation range (0.1–1 %), so it is unclear whether it is possible to distinguish the suitable environments for each powder in cloud seeding field experiments as proposed in this paper.
Regarding the cloud chamber experiments on the observations of cloud droplet formation, it is essential to discuss phenomena occurring under water sub-saturation and supersaturation separately so that the onset of cloud droplet formation and the measured RH at those time should be clarified. The existence of cloud droplets at the under-saturated stage (RH ≤ 85 %) and pre-saturated stage (85 % < RH ≤ 100 %) are questionable without noteworthy explanation. When discussing measurements under sub-saturated conditions, it is necessary to explain how to identify pre- and post-deliquescent salt particles. The main discussion should be focused on how such salt particles lead to their CCN activation and readily grow into larger cloud droplet up to drizzle embryo size at the super-saturated stage. From this perspective, it is necessary to consistently clarify the relationship between the deliquescence and hygroscopic properties of each salt particle type and the number concentration and size distribution of the induced cloud droplets.
Major comments:
L20-21: Regarding the notation “The droplet diameter varied from 1 to 90 μm”, it may include a reflection of size changes due to deliquescence, especially for particles of a few microns in size. It is not clear how to distinguish between aerosols and aqueous solutions, so for “The droplet diameter”, how about expressing it as “the particle diameter including aerosols and droplets”?
L258-260: Regarding “In the cloud chamber experiment, the particles measured by the OPC were assumed to be water droplets. Therefore, the size parameter of the OPC was set based on the RI value of 1.33 (water).”, the timing of onset of CCN activation is a critical issue in the cloud chamber experiment, and so it is incomprehensible to treat all OPC measurement data in the sub-saturated region as water droplets and to compare and discuss the cloud droplet formation process of different materials without applying the specific method of distinguishing aerosols from cloud droplets. How should we interpret the identification of water droplet formation in the sub-saturated region?
L306-309: For S ≥ 0.2%, some D_act values for NaCl powder are relatively smaller than those for CaCl2 powder, so it is not possible to say clearly about the differences in target clouds in field cloud seeding experiments. It should be noted that the results depend on the PSD and mode diameter of the sample particles used in this study. Because the PSD in the aerosol chamber changes with time, and micron-sized particles in particular fall out quickly, using the initial PSD value will affect the calculation of D_act values for higher S. Was the updated PSD data applied to the calculation of the D_actvalues?
L347-351: In OPC measurements, the RI value of 1.33 (water) is applied even in the unsaturated region. Unless it is any evidence that the particles observed during the deliquescence transition are cloud droplets, the particle size measurements may not be accurate. If the cloud droplets (several tens of μm in size) measured with the CPI under sub-saturated environment are correct, the spatial representativeness of the RH measurement or the accuracy of the RH values is required to be precise (is RH>100% unevenly distributed?). Regarding Figure 4i, in the other three cases, the timing at which the CPI first measured was at 85% RH (near the vertical blue lines), but is it relatively earlier?
L351-354 Isn't "the most significant decrease in mean absolute humidity" a contradiction to the increase in RH? Also, when RH exceeds 100%, is the transition to condensational growth unclear?
L469-482: Since the definition of cloud droplets and the timing of their generation (onset) in this study are not clear, it is difficult to understand the results that cloud droplets were induced even in the S2 stage (unsaturated region). If the particles captured by CPI in the S2 stage are cloud droplets, how can their formation process be explained? Considering the relationship between the temperature lapse rate and the updraft velocity in the atmosphere, is what the authors described here about the applicable atmospheric conditions as the scope of these experimental setup appropriate?
Specific comments:
L67-70: Considering the timing of conducting cloud seeding experiments for suitable warm clouds, the occurrence characteristics of each cloud type in each season should also be described.
L122-124: How long does it take for the heat transfer fluid to pass from inlet to outlet inside the cloud chamber? Also, what is the approximate elapsed time for the heat transfer fluid to circulate through the entire path of the cooling system?
L127-129: To what extent does the evacuation rate corresponding to the controlled flow rate with the vacuum pump and the opening rate of the SV cover the range of the updraft velocity(m/s)?
L129-134: What models of a triple filter in the dry air system and the pure water system products are used? What are the humidity control ranges (min/max dew points) and how long does it take for conditions inside the chamber to reach these values?
L154-155: What is the typical rate (cm³/min) at which the aerosol generator injects the aerosol into the inner chamber?
L156-158: The measurement range of OPC covers not only cloud droplets but also aerosol sizes. If so, how can these particles be identified? How to calculate total cloud droplets number concentration? Similarly, for L217-218.
L163-165: Is the CPC measurement not affected by the pressure difference between inlet and outlet? Why not sample the air like other measurement instruments?
L170-172: Describe more precisely the roles of the OPC and the SMPS in measuring aerosol size.
L208-209: I don't understand why it is necessary to mention "the growth of cloud droplets was observed at 900 and 840 s, respectively" in this paragraph.
L212-214: Does the T_wall measurement positions take into account not only the geometric representative points of the inner chamber, but also the transition time of the heat transfer fluid through the wall panel of the inner chamber?
L234-236: Is ΔP kept at 30 hPa during aerosol injection, or does it decrease gradually according to the injection rate without exhausting?
L238-239: Are there any significant changes in CN number concentration and PSD during the 1-hour observation period?
L245-246: How was the pressure in the aerosol chamber controlled during SMPS and OPC measurements, which required a certain total sample flow rate?
L253-254: After the humidification, was ΔP adjusted to be 30hPa, as same as the procedure in the aerosol chamber experiment?
L258: Was there little effect on drying due to aerosol injection?
L262-263: The initial values of Tair in NaCl Exp. #1 and #2 are approximately 0.4°C lower than those of T_wall. The differences between them were relatively large compared to the two cases in CaCl2 experiment. Could this be due to some difference in the procedure?
L273: At what timing were these PSDs measured?
L323-324: Is it inevitable that the operation of the SV and vacuum pump will have an effect after the experiment has started? If necessary, describe it as a part of the experimental procedure.
L326-328: Excluding the transitional period up to 150 seconds after the start of each experiment, the cooling rate was almost uniform at about -2K/min. Even if there is a significant difference in updraft velocity between SV 20% and 50%, the greater the updraft velocity, the greater the deviation from adiabatic expansion, so it is unclear whether the experiment conforms to the conditions proposed here.
L329-330: Does the time it takes for the fluid to circulate in the chamber affect the homogeneity of the wall temperature? Also, are there any adjustments being made to the flow of the heat transfer fluid into the chamber in terms of the wall temperature control?
L330-331: In the case where Tair-Tdew is small (initial value of RH is high), is there a possibility that it will affect the variation in air temperature? How can it be reduced?
L333-337: Did the authors confirm that the PSDs of each sample introduced into the cloud chamber were the same as that measured in the aerosol chamber? Also, the number concentration and PSD change with time in the chamber experiment, but have these been taken into consideration? The same applies to lines 390-392.
L365-366: Since the evacuation by a vacuum pump simulates the expansion process, does it take into account the reduction (or loss) of aerosols containing CCNs due to the dilution? Is an additional consideration the loss due to falling out micron-sized CCNs and cloud particles?
L369-372: Shouldn't the size distributions at least in stages S1 and S2 be aerosol and cloud droplet size distribution rather than cloud DSD since it contains aerosol particles that are not activated as CCNs? The same shall apply hereinafter.
L376-377: In Fig. 6a, an increase in the number concentration of particle in sizes measured with CPI can be confirmed in S3 compared to S2 and, but the difference in the “right tail of the bimodal distribution” of S2 and S3 does not seem to be clear.
L384-385: As for NaCl, is it possible that there are a relatively large number of particles that are slightly below the OPC detection limit (D > 0.3 μm)? If so, should it be considered that those relatively small particles act as CCNs?
Technical corrections:
L145: Table 2. -> Table 1.
L214: Tdew -> dew point temperature (Tdew)
L269: Table 6. Experimental conditions -> Experimental initial conditions
L313: DDact -> Dact
L319: “The air pressure, air, dew point,” -> “The air pressure, air temperature, dew point,”
L324-326: To clarify that this is the case for NaCl Exp. #2.
L359-360: Add “not only in case of NaCl Exp. #1 but also in the other three cases”.
L373: shown in Fig. 4g -> shown in Fig. 4i
L384-385: cloud droplets were observed -> cloud particles at below freezing point were observed
L386: (Figs. 4g and 4h) -> (Figs. 4i and 4j)
L425-449: The legends in Figures 4 and 5 overlap with the plots and there are some unclear parts, so they should be corrected.
L464-466: Add “in NaCl Exp. #2”

Reply

Citation: https://doi.org/10.5194/egusphere-2025-515-RC3

Bu-Yo Kim, Miloslav Belorid, Joo Wan Cha, Youngmi Kim, and Seungbum Kim

Viewed

Total article views: 186 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	BibTeX	EndNote
145	32	9	186	2	2

HTML: 145
PDF: 32
XML: 9
Total: 186
BibTeX: 2
EndNote: 2

Views and downloads (calculated since 25 Apr 2025)

Month	HTML	PDF	XML	Total
Apr 2025	75	12	4	91
May 2025	70	20	5	95

Cumulative views and downloads (calculated since 25 Apr 2025)

Month	HTML	PDF	XML	Total
Apr 2025	75	12	4	91
May 2025	70	20	5	95

Viewed (geographical distribution)

Total article views: 187 (including HTML, PDF, and XML) Thereof 187 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 16 May 2025

Short summary

This study analyzed NaCl and CaCl₂, powder-type hygroscopic materials for cloud seeding, using the Korea Cloud Physics Experiment Chamber (K-CPEC) facility in South Korea. The aerosol chamber enabled particle analysis in an extremely dry environment, while the cloud chamber, with precise pressure and temperature control, facilitated droplet growth through adiabatic expansion. Our study provides valuable insights for the development of new seeding materials and advanced cloud seeding experiments.


Total:	0
HTML:	0
PDF:	0
XML:	0