the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 18 Jan 2024
Secondary Organic Aerosols Derived from Intermediate Volatility n-Alkanes Adopt Low Viscous Phase State
Abstract. Secondary organic aerosol (SOA) derived from n-alkanes, as emitted from vehicles and volatile chemical products, is a dominant component of anthropogenic particulate matter, yet its chemical composition and phase state are poorly understood and hardly constrained in aerosol models. Here we provide a comprehensive analysis of n-alkane SOA by explicit chemistry modeling, machine learning, and laboratory experiments to show that, counterintuitively, n-alkane SOA adopt low viscous semisolid or liquid states. Our study underlines the complex interplay of molecular composition and SOA viscosity: n-alkane SOA with higher carbon number mostly consists of less functionalized first-generation products with lower viscosity, while the lower carbon number SOA contains more functionalized multigeneration products with higher viscosity. This study opens up a new avenue for analysis of SOA processes and the results indicate little kinetic limitations of mass accommodation in SOA formation, supporting the application of equilibrium partitioning for simulating n-alkane SOA formation in large-scale atmospheric models.
-
Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
-
Preprint
(2987 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(2987 KB) - Metadata XML
- BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2024-51', Anonymous Referee #1, 05 Feb 2024
In the manuscript " Secondary Organic Aerosols Derived from Intermediate Volatility n-Alkanes Adopt Low Viscous Phase State, " the authors reported that n-alkane SOA with higher carbon number mainly consists of less functionalized first-generation products with lower viscosity, while the lower carbon number SOA contains more functionalized multigeneration products with higher viscosity based on the GECKO-A box model simulation and chamber experiments. In general, I am very supportive of the hypothesis that the increase of the alkyl group may reduce the viscosity of SOA from n-alkanes, and the topic of this work is interesting to readers of this subject.
Only the functional group information was provided in this study, which can indeed help understand SOA's total properties. However, since there is no specific molecular information in the chamber data, it is very difficult to evaluate the mechanisms generated by the GECKO-A. Using the GECKO-A box model isn't helpful unless you know which products are really forming in the gas phase and subsequent particle-phase chemistry in the aerosol. Thus, for the publication of this manuscript, the following points should be addressed.
Line 19，in the abstract section, the authors stated that SOA derived from n-alkanes is the dominant component of anthropogenic particulate matter; however, it is generally believed that aromatics and alkenes are the main precursors of anthropogenic aerosols.
In the methods section, the authors showed that GECKO-A generated the chemical mechanisms of n-alkanes; however, the details about the mechanisms were not provided. Additionally, the mechanism's performance should be carefully characterized by chamber experiments before it is used to evaluate the molecular properties of SOA.
Line 123: How many species were involved during the SOA formation process? Detailed information about the box model should be provided.
In line 233, the authors only compared the yields of SOA between chamber data and model simulations; without further comparisons of chemical compositions, it is hard to conclude that particle-phase oligomerization contributes minor. Additional information or references are needed to support this statement.
Line 261, why does the Tg,org decrease for C8-12 in Figure 2a? Some explanation should be provided.
Figure 2b, some semi-volatile compounds may escape from particles into the gas phase in the aerodynamic lens due to the high vacuum; hence, Tg may be overestimated, and the effect of vacuum on SOA compositions should be discussed.
In Line 397, It shows that the model performs well on SOA yields by comparing the chamber and the GECKO-A box model; still, the time profiles of SOA mass concentrations should be provided to evaluate the performance of the box model.
Citation: https://doi.org/10.5194/egusphere-2024-51-RC1 -
AC1: 'Reply on RC1', Manabu Shiraiwa, 06 Mar 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2024/egusphere-2024-51/egusphere-2024-51-AC1-supplement.pdf
-
AC1: 'Reply on RC1', Manabu Shiraiwa, 06 Mar 2024
-
RC2: 'Comment on egusphere-2024-51', Anonymous Referee #2, 12 Feb 2024
GENERAL COMMENTS:
In this study, the authors predict the viscosity of n-alkane SOA under high NOx conditions using a box model (GECKO-A) combined two different Tg-predictors: the compositional parameterization (CP) and a machine-learning based approach (tgBoost). This paper is within the scope of ACP, and the results are relevant and useful for the atmospheric chemistry community in terms of better understanding the various trends of SOA viscosity. This study provides a clear benchmark for future work to test the viscosity predictions of tgBoost and CP to further validate the author’s conclusions. However, some parts of the paper need further explanation and I also suggest rewording some parts of the paper for clarity, as discussed below. Overall, this paper is acceptable for publication in ACP after revision of the below points.
SPECIFIC COMMENTS:
- The authors point to previous studies that describe in detail the limitations and assumptions of using tgBoost and CP. This is normal practice in general, but it would strengthen the paper to add a brief summary of the main limitations and assumptions of each model here. The authors do mention the types of reactions that tgBoost does not explicitly consider, but nothing about other assumptions made for tgBoost or for combining GECKO-A with CP. A short discussion will add some needed context for the readers.
- More details need to be provided for the calculation of viscosity and bulk diffusivity from Tg,org. What parameters were used in the VFT equation to calculate viscosity (e.g., what was Df?). What parameters were used in the fractional Stokes-Einstein equation to calculate bulk diffusivity (e.g., what hydrodynamic radii was used for the n-alkane SOA)? Please at least provide the parameters for this calculation. This could either go in the main text or supplement.
- In many cases, the authors describe their results as “counter-intuitive”, “surprising”, and “remarkable”. I don’t necessarily think these kinds of superlatives are appropriate in this case. From what I can tell, the authors describe three things as “remarkable”: 1) that tgBoost and CP don’t agree on the viscosity trend (e.g. Line 259, 272), 2) that the viscosity of n-alkanes decrease with increasing n (Line 275), and 3) that tgBoost predicts the viscosity very well (Line 449). For points (1) and (3), the authors themselves state on Lines 352-355 that “as tgBoost considers molecular structure…. It should make better predictions for multi-functionalized compounds,” which I agree with – tgBoost seems like a more sophisticated model, so I think it’s reasonable that it would both perform better than CP and perform quite well when comparing to measurements. For point (2), I believe it’s relatively well-hypothesized that SOA viscosity is not only dependent on molecular weight, but also functionality and structure. I suggest either revising or removing this type of language throughout.
- The authors mention studies where CP viscosity predictions have agreed with viscosity measurements (Line 378-380), but if I recall the literature correctly, there have been previous studies where CP (i.e., the DeRieux et al., 2018 paramterization) has not agreed with experimental viscosities. Mentioning some of these cases would strengthen the conclusion that GECKO-A + tgBoost is a better tool for predicting SOA viscosity.
- Lines 323-327. The authors note that the trend of N:C and O:C is consistent with previous studies, which seems correct. However, they also mention that the simulated values are 15-45% lower than these measured values. A short discussion of why these values are lower is warranted.
- The mass loadings of seed particles in the chamber were ~200-400 ug/m3, but I don’t see anywhere that states the mass loadings of SOA in the chamber. Is it assumed that the SOA mass loading is equal to the seed particles mass loading?
- Lines 243-245: the authors mention that particle-phase chemistry was shown to be substantial in n-alkane SOA formation for low NOx conditions. If I understood correctly, GECKO-A models gas-phase chemistry only, and not particle-phase chemistry. The authors do not further discuss the possibility of particle-phase chemistry under high NOx conditions and how this may affect their results.
TECHNICAL NOTES:
- Line 76: Long-chain not defined. What value of n differentiates long-chain from medium-chain?
- Line 92: move “to date” later in the sentence – “The GECKO-A model is one of the most comprehensive generators of gas-phase chemical schemes to date…”
- Line 107: add “the” before “oxidation”.
- Line 152: Should be Fig. A3a, not S3a.
- Line 164: Add a comma after “In this study”
- Line 165: Make “prediction” plural.
- Line 194: I could be wrong, but doesn’t Tenax need a registered (R) symbol following it? “Tenax®”.
- Line 231: “lower volatility” not “volatility lower.
- Line 247: Add "an" before “effective mass accommodation”.
- Line 247: Change “that” to “which”.
- Line 251: Glass transition temperature was already defined, so the symbol Tg can be used here.
- Line 252: Text suggests to look at the "green line," but no figure has been mentioned for a while. Please indicate which figure/panel.
- Figure 2 - The text of secondary y-axis on panel (a) and the y-axis of panel (b) are too close together.
- Figure 1 – The ordering of panels is not intuitive with (d) being the top right panel.
- Figure 1c - The orange line is essentially not visible behind the green line. Make some note of this in the caption or text, or visually show it some other way.
- Line 340-341 – Add "that the" before “five species”.
- Line 366: Add "the" before SOA.
- Line 431: Reword the start of the sentence. Either “IVOCs have gained..” or “IVOCS are gaining attention..”
Citation: https://doi.org/10.5194/egusphere-2024-51-RC2 -
AC2: 'Reply on RC2', Manabu Shiraiwa, 06 Mar 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2024/egusphere-2024-51/egusphere-2024-51-AC2-supplement.pdf
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2024-51', Anonymous Referee #1, 05 Feb 2024
In the manuscript " Secondary Organic Aerosols Derived from Intermediate Volatility n-Alkanes Adopt Low Viscous Phase State, " the authors reported that n-alkane SOA with higher carbon number mainly consists of less functionalized first-generation products with lower viscosity, while the lower carbon number SOA contains more functionalized multigeneration products with higher viscosity based on the GECKO-A box model simulation and chamber experiments. In general, I am very supportive of the hypothesis that the increase of the alkyl group may reduce the viscosity of SOA from n-alkanes, and the topic of this work is interesting to readers of this subject.
Only the functional group information was provided in this study, which can indeed help understand SOA's total properties. However, since there is no specific molecular information in the chamber data, it is very difficult to evaluate the mechanisms generated by the GECKO-A. Using the GECKO-A box model isn't helpful unless you know which products are really forming in the gas phase and subsequent particle-phase chemistry in the aerosol. Thus, for the publication of this manuscript, the following points should be addressed.
Line 19，in the abstract section, the authors stated that SOA derived from n-alkanes is the dominant component of anthropogenic particulate matter; however, it is generally believed that aromatics and alkenes are the main precursors of anthropogenic aerosols.
In the methods section, the authors showed that GECKO-A generated the chemical mechanisms of n-alkanes; however, the details about the mechanisms were not provided. Additionally, the mechanism's performance should be carefully characterized by chamber experiments before it is used to evaluate the molecular properties of SOA.
Line 123: How many species were involved during the SOA formation process? Detailed information about the box model should be provided.
In line 233, the authors only compared the yields of SOA between chamber data and model simulations; without further comparisons of chemical compositions, it is hard to conclude that particle-phase oligomerization contributes minor. Additional information or references are needed to support this statement.
Line 261, why does the Tg,org decrease for C8-12 in Figure 2a? Some explanation should be provided.
Figure 2b, some semi-volatile compounds may escape from particles into the gas phase in the aerodynamic lens due to the high vacuum; hence, Tg may be overestimated, and the effect of vacuum on SOA compositions should be discussed.
In Line 397, It shows that the model performs well on SOA yields by comparing the chamber and the GECKO-A box model; still, the time profiles of SOA mass concentrations should be provided to evaluate the performance of the box model.
Citation: https://doi.org/10.5194/egusphere-2024-51-RC1 -
AC1: 'Reply on RC1', Manabu Shiraiwa, 06 Mar 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2024/egusphere-2024-51/egusphere-2024-51-AC1-supplement.pdf
-
AC1: 'Reply on RC1', Manabu Shiraiwa, 06 Mar 2024
-
RC2: 'Comment on egusphere-2024-51', Anonymous Referee #2, 12 Feb 2024
GENERAL COMMENTS:
In this study, the authors predict the viscosity of n-alkane SOA under high NOx conditions using a box model (GECKO-A) combined two different Tg-predictors: the compositional parameterization (CP) and a machine-learning based approach (tgBoost). This paper is within the scope of ACP, and the results are relevant and useful for the atmospheric chemistry community in terms of better understanding the various trends of SOA viscosity. This study provides a clear benchmark for future work to test the viscosity predictions of tgBoost and CP to further validate the author’s conclusions. However, some parts of the paper need further explanation and I also suggest rewording some parts of the paper for clarity, as discussed below. Overall, this paper is acceptable for publication in ACP after revision of the below points.
SPECIFIC COMMENTS:
- The authors point to previous studies that describe in detail the limitations and assumptions of using tgBoost and CP. This is normal practice in general, but it would strengthen the paper to add a brief summary of the main limitations and assumptions of each model here. The authors do mention the types of reactions that tgBoost does not explicitly consider, but nothing about other assumptions made for tgBoost or for combining GECKO-A with CP. A short discussion will add some needed context for the readers.
- More details need to be provided for the calculation of viscosity and bulk diffusivity from Tg,org. What parameters were used in the VFT equation to calculate viscosity (e.g., what was Df?). What parameters were used in the fractional Stokes-Einstein equation to calculate bulk diffusivity (e.g., what hydrodynamic radii was used for the n-alkane SOA)? Please at least provide the parameters for this calculation. This could either go in the main text or supplement.
- In many cases, the authors describe their results as “counter-intuitive”, “surprising”, and “remarkable”. I don’t necessarily think these kinds of superlatives are appropriate in this case. From what I can tell, the authors describe three things as “remarkable”: 1) that tgBoost and CP don’t agree on the viscosity trend (e.g. Line 259, 272), 2) that the viscosity of n-alkanes decrease with increasing n (Line 275), and 3) that tgBoost predicts the viscosity very well (Line 449). For points (1) and (3), the authors themselves state on Lines 352-355 that “as tgBoost considers molecular structure…. It should make better predictions for multi-functionalized compounds,” which I agree with – tgBoost seems like a more sophisticated model, so I think it’s reasonable that it would both perform better than CP and perform quite well when comparing to measurements. For point (2), I believe it’s relatively well-hypothesized that SOA viscosity is not only dependent on molecular weight, but also functionality and structure. I suggest either revising or removing this type of language throughout.
- The authors mention studies where CP viscosity predictions have agreed with viscosity measurements (Line 378-380), but if I recall the literature correctly, there have been previous studies where CP (i.e., the DeRieux et al., 2018 paramterization) has not agreed with experimental viscosities. Mentioning some of these cases would strengthen the conclusion that GECKO-A + tgBoost is a better tool for predicting SOA viscosity.
- Lines 323-327. The authors note that the trend of N:C and O:C is consistent with previous studies, which seems correct. However, they also mention that the simulated values are 15-45% lower than these measured values. A short discussion of why these values are lower is warranted.
- The mass loadings of seed particles in the chamber were ~200-400 ug/m3, but I don’t see anywhere that states the mass loadings of SOA in the chamber. Is it assumed that the SOA mass loading is equal to the seed particles mass loading?
- Lines 243-245: the authors mention that particle-phase chemistry was shown to be substantial in n-alkane SOA formation for low NOx conditions. If I understood correctly, GECKO-A models gas-phase chemistry only, and not particle-phase chemistry. The authors do not further discuss the possibility of particle-phase chemistry under high NOx conditions and how this may affect their results.
TECHNICAL NOTES:
- Line 76: Long-chain not defined. What value of n differentiates long-chain from medium-chain?
- Line 92: move “to date” later in the sentence – “The GECKO-A model is one of the most comprehensive generators of gas-phase chemical schemes to date…”
- Line 107: add “the” before “oxidation”.
- Line 152: Should be Fig. A3a, not S3a.
- Line 164: Add a comma after “In this study”
- Line 165: Make “prediction” plural.
- Line 194: I could be wrong, but doesn’t Tenax need a registered (R) symbol following it? “Tenax®”.
- Line 231: “lower volatility” not “volatility lower.
- Line 247: Add "an" before “effective mass accommodation”.
- Line 247: Change “that” to “which”.
- Line 251: Glass transition temperature was already defined, so the symbol Tg can be used here.
- Line 252: Text suggests to look at the "green line," but no figure has been mentioned for a while. Please indicate which figure/panel.
- Figure 2 - The text of secondary y-axis on panel (a) and the y-axis of panel (b) are too close together.
- Figure 1 – The ordering of panels is not intuitive with (d) being the top right panel.
- Figure 1c - The orange line is essentially not visible behind the green line. Make some note of this in the caption or text, or visually show it some other way.
- Line 340-341 – Add "that the" before “five species”.
- Line 366: Add "the" before SOA.
- Line 431: Reword the start of the sentence. Either “IVOCs have gained..” or “IVOCS are gaining attention..”
Citation: https://doi.org/10.5194/egusphere-2024-51-RC2 -
AC2: 'Reply on RC2', Manabu Shiraiwa, 06 Mar 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2024/egusphere-2024-51/egusphere-2024-51-AC2-supplement.pdf
Peer review completion
Journal article(s) based on this preprint
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 368 | 121 | 19 | 508 | 11 | 10 |
- HTML: 368
- PDF: 121
- XML: 19
- Total: 508
- BibTeX: 11
- EndNote: 10
Viewed (geographical distribution)
| Country | # | Views | % |
|---|---|---|---|
| United States of America | 1 | 239 | 46 |
| China | 2 | 73 | 14 |
| France | 3 | 35 | 6 |
| Germany | 4 | 24 | 4 |
| South Korea | 5 | 21 | 4 |
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
- 239
Tommaso Galeazzo
Bernard Aumont
Marie Camredon
Richard Valorso
Yong B. Lim
Paul J. Ziemann
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(2987 KB) - Metadata XML