the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 18 Aug 2022
Modeling the Influence of Chain Length on SOA Formation via Multiphase Reactions of Alkanes
Azad Madhu
David Deacon
Abstract. Secondary Organic Aerosol (SOA) from diesel fuel is known to be significantly sourced from the atmospheric oxidation of aliphatic hydrocarbons. In this study, the formation of linear-alkane SOA was predicted using the Unified Partitioning Aerosol Phase Reaction (UNIPAR) model that simulated multiphase reactions of hydrocarbons. In the model, the formation of oxygenated products from the photooxidation of linear alkanes was simulated using a near-explicit gas kinetic mechanism. Autoxidation paths integrated with alkyl peroxy radicals were added to the Master Chemical Mechanismv3.3.1 to improve the formation of low volatility products in the gas phase and better predict SOA mass. The resulting gas products were then classified into volatility-reactivity based lumping groups that are linked to mass-based stoichiometric coefficients. The SOA mass in the UNIPAR model is produced via three major pathways: partitioning of gaseous oxidized products onto both the organic and wet inorganic phases; oligomerization in organic phase; and reactions in the wet inorganic phase (acid-catalyzed oligomerization and organosulfate formation). The model performance was demonstrated for SOA data that were produced through the photooxidation of a homologous series of linear alkanes ranging from C9 to C15 under varying environments (NOx levels, temperature, and inorganic seed conditions) in a large outdoor photochemical smog chamber. The product distributions of linear alkanes were mathematically predicted as a function of carbon number using an incremental volatility coefficient (IVC) to cover a wide range of alkane lengths. The prediction of alkane SOA using the incremental volatility-based product distributions, which were obtained with C9–C12 alkanes, was evaluated for prediction of C13 and C15 chamber data and further extrapolated to predict the SOA from longer chain alkanes (C15) that can be found in diesel. The model simulation of linear alkanes in diesel fuel suggests that SOA mass is mainly produced by alkanes C15 and higher. Alkane SOA is insignificantly impacted by the reactions of organic species in the wet inorganic phase due to the hydrophobicity of products, but significantly influenced by gas-particle partitioning.
-
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
(1184 KB)
-
Supplement
(513 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1184 KB) -
Supplement
(513 KB) - BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Azad Madhu et al.
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2022-681', Anonymous Referee #1, 14 Sep 2022
This paper presents a comparison between a modeling study and chamber measurements. It reaches conclusions that should be of interest to the atmospheric chemistry modeling community with regard to the relative importance of SOA mass creation processes. The paper is well written, with clear figures and discussion. I have two requests for improvement:
1) This work is a development of a previous work by some of the same workers (Zhou et al 2019), and it is not always immediately apparent what is new work and what is a restatement of that previous work. The text should clarify the distinction. The restated sections should be condensed and appropriate reference should be made to the prior work.
2) This study should add at least one sensitivity test that attempts to model the morning spike seen in the chamber observations, to support the authors’ assertion about its origin.
The standard review questions follow:
1. Does the paper address relevant scientific questions within the scope of ACP?
Yes, this work is well within the shpere of interest of ACP. The paper presents a process-model simulation of the formation of SOA from a selection of precursors representative of diesel fuel emissions. The authors use parameterized product distributions, and extrapolate these distributions to larger precursors than those available in the model’s chemical mechanism.
2. Does the paper present novel concepts, ideas, tools, or data?
The simulations using the extrapolated parameterizations are compared to new chamber data. The comparisons are improved by the addition of the auto-oxidation process to the model training set, and by the consideration of wall losses in the chamber.
3. Are substantial conclusions reached?
The authors conclude that reactions in the wet inorganic phase make no significant contribution to SOA mass from the species considered: gas-particle partitioning is far more important. This result should be useful to future modeling studies.
4. Are the scientific methods and assumptions valid and clearly outlined?
The scientific method combines several previously-published concepts and equations, which are clearly outlined.
5. Are the results sufficient to support the interpretations and conclusions?
The interpretations and conclusions follow logically from the model/chamber comparisons. The Figures illustrate the analysis nicely.
- Line 300: discussion of Figure 3. I agree that C9D, C10D and C11C show good model-data comparisons. Comparisons C9C and C10B are not so good, and look more like comparisons C11A and C11B, which are said to be poor. The text should be clearer about this distinction. This comment des not detract from the paper’s overall conclusions.
- IMPORTANT. This study should add at least one sensitivity test that attempts to model the morning spike seen in the chamber observations, to support the authors’ assertion about its origin. (Line 325)
6. Is the description of experiments and calculations sufficiently complete and precise to allow their reproduction by fellow scientists (traceability of results)?
The experiments appear to be well documented.
7. Do the authors give proper credit to related work and clearly indicate their own new/original contribution?
In general, yes, credit is given. However this work is a development of a previous work by the same group (Zhou et al 2019), and it is not always immediately apparent what is new work and what is a restatement of that previous work. The model description, especially Sections 3.2 & 3.4, appears to be taken almost verbatim from Zhou et al 2019. That work should be acknowledged early in each relevant section and the authors should clearly say in the model description what parts of the model system are reused from that prior work, and what parts are new. Consideration should be given as to whether any of these sections can be simplified and shortened by referring the reader to the previous work. The same applies to Figure 1, which appears to be almost identical to Figure 1 of Zhou (2019)
8.Does the title clearly reflect the contents of the paper? 9.Does the abstract provide a concise and complete summary?
The title and abstract are clear and complete. I suggest that the model name should be added to the paper title.
10. Is the overall presentation well structured and clear?
The presentation and structure are generally clear. For specific minor points, see my list below.
11. Is the language fluent and precise?
The language is mostly fluent, precise and clear. I have a couple of minor requests for clarity, detailed below.
12. Are mathematical formulae, symbols, abbreviations, and units correctly defined and used?
- The symbols and abbreviations are much easier to follow after studying Figure 1.
- Section S4, Table S2: It is unfortunate that the accommodation coefficient alpha_w,i, and the polarizability alpha_i have such similar symbols, and that these could be confused with the mass-based stoichiometric coefficient alpha_i introduced in section 3.2. It might be appropriate to change one of these symbols, or at least to explicitly acknowledge the possibility for confusion.
13. Should any parts of the paper (text, formulae, figures, tables) be clarified, reduced, combined, or eliminated?
- The model description could be condensed by properly acknowledging and referring to the previous work, as already mentioned.
- Line 131-132, 197, 215 etc: This reviewer finds the abbreviations “or” and “in” confusing in the text because they can be mistaken for ordinary words. Please spell out the words “organic” & “inorganic” unless the abbreviations are used in combination with other symbols? Or, at a minimum, give them bold italic font to differentiate them from normal text.
14. Are the number and quality of references appropriate?
This is generally a good (but not exhaustive) guide to the recent literature on the relevant topics.
- The following works are cited in the text but are missing from the reference lists: Line 198: Pankow (1994); Line 272: Yap (2011); SI: Roldin (2019).
- Line 142: Pye 2019 and Xavier 2019 are not appropriate references for the identification of auto-oxidation reactions: they are modeling papers. It would be better to cite an early lab study (e.g. Sahetchian et al, Combust. Flame 1991; Crounse & Nielsen. J. Phys. Chem. Lett., 2013)
15. Is the amount and quality of supplementary material appropriate?
The supplementary material is helpful and appropriate.
- Tables S4, S5: please provide a key to or explanation of the chemical compound code names?
Other specific comments (major):
- Line 158: Please briefly discuss the conceptual movement of the reaction products between reactivity levels in your 51-box matrix. What does this represent, in a chemical-physical sense? Are there any constraints on the sum of all the dynamic alpha_sub_i values?
Specific comments (minor):
- Line 45: please correct the reference (Robinson et al., 2007)
- Lines 46-73: Please mention the names of the models used in Hodzic (2010), Cappa (2013), Zhang & Seinfeld (2013)
- Line 50: Note that, since Lee-Taylor et al used the larger linear alkanes as surrogates for all larger species, the fact that the majority of SOA precursors in their model were linear alkanes is a logical outcome of those SOA species having larger carbon numbers.
- Lines 61-65: Perhaps “neglect” would be kinder than “fail to consider”? The previous authors are likely well aware of the limitations of their approaches, and (at least in one case) mention those limitations explicitly.
- Line 86: “… the typical ozone mechanisms..” This seems vague. Did this study use several ozone mechanisms or just one?
- Line 133: “OMAR” needs subscripts.
- Lines 272-274: This sentence is confusing. Please rearrange it?
- Line 305: Do you mean “from” instead of “form”?
- Line 321, 323: Do you mean figure 5 instead of fig 6?
- Line 322: Do you mean figure 4 instead of fig 5?
- Line 341: why is it notable that previous experiments used H2O2 as a low-NOx OH source? What are the implications? Please discuss.
- Line 357: please specify whether you mean the relative or absolute contribution of OMAR?
- Line 370: do you mean Figure S6 instead of Fig 5?
- Line 625, 655: “OCEC is missing its “/”
Citation: https://doi.org/10.5194/egusphere-2022-681-RC1 -
RC2: 'Comment on egusphere-2022-681', Anonymous Referee #2, 20 Sep 2022
Madhu et al. present their work on modeling SOA yields from the oxidation of linear C10 through C20 alkanes. The model is described well and a lot of effort was invested in the research. However, the use of an outdoor chamber to conduct systematic tests of NOx and particle seed type on SOA yields was a flaw in the experiment design. The day-to-day variability in conditions, namely temperature and light, clouded the very impact of these variables the authors sought to test on SOA yields. Perhaps as a result, the demonstration of model peformance on SOA loadings (shown in Figures 2 through 5) is unconvincing. For instance, the explanation of the observed morning SOA burst in some of the experiments that the model fails to capture was seriously lacking.
Likewise, the discussion on the effect of NOx on autoxidation, the main driver of SOA formation they report, is thin. What is the implemented rate of autoxidation for each alkane precursor? A sensitivity test of the rate of autoxidation on SOA yield is reported on line 370 to be shown in Figure 5, which does not show that at all (only the sensitivity results of wall loss rates). What are the levels of Hox and Nox in the chamber for the duration of each experiment? Both would undoubtedly affect RO2 fates and lifetimes, neither of which is discussed in the section of Nox impact on autoxidation. Figure 6 shows that SOA yields increase with decreasing Nox. However, Figure 2 shows that the model underestimates SOA loading at high Nox and overestimates at low Nox. This, as a result, possibly undercuts the key model results of SOA yield vs Nox shown in Figure 6. Proper explanation of glaring caveats is not discussed. Authors go on to attribute the decrease in SOA yield with Nox to the partitioning of non-autox products, but that explanation is unclear and buried in a Table in the SI. A clearer demonstration with a graphic is needed.
I would have like to have seen optimization of model parameters (volatility, reactivity, and/or aging) using the suite of observations made. Or at the very least, show Figures 6 and 7 (which are the key figures) but with observations shown as well. This would ground the model and allow the authors to make conclusions that currently seem like extreme extrapolations. For instance, the levels of SOA in the chamber are beyond atmospherically relevant. The high loadings most likely affected the relative contribution of gas partitioning versus heterogeneous chemistry to SOA loading. Additionally, the authors conclude from this modeling exercise that straight chain alkanes are important for urban SOA - without showing much evidence. This is not backed up by the results shown.
It is my opinion that major revisions are required for this work to be published in ACP.
Minor
Line 153: Why 51 species? Where is this from?
Line 24: symbol in front of C15 did not render
Line 125: multiplication symbol did not render properly
Line 128: the “i” in alpha_i needs subscript
Figure S1. Is that starting from an alkoxy radical? Initial compound is blurry. How does the hydroperoxy group convert to a carbonyl? Please explain in the caption.
Instead of “or” for organic phase, “in” for inorganic, consider using “org” and “inorg” for clarity
Figure 2. Consider not using black and grey for MCM and MCM+autox. Something more different.
Line 321. Should cite Fig 5 not Fig 6?
Citation: https://doi.org/10.5194/egusphere-2022-681-RC2 - AC1: 'Comment on egusphere-2022-681', Myoseon Jang, 08 Nov 2022
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2022-681', Anonymous Referee #1, 14 Sep 2022
This paper presents a comparison between a modeling study and chamber measurements. It reaches conclusions that should be of interest to the atmospheric chemistry modeling community with regard to the relative importance of SOA mass creation processes. The paper is well written, with clear figures and discussion. I have two requests for improvement:
1) This work is a development of a previous work by some of the same workers (Zhou et al 2019), and it is not always immediately apparent what is new work and what is a restatement of that previous work. The text should clarify the distinction. The restated sections should be condensed and appropriate reference should be made to the prior work.
2) This study should add at least one sensitivity test that attempts to model the morning spike seen in the chamber observations, to support the authors’ assertion about its origin.
The standard review questions follow:
1. Does the paper address relevant scientific questions within the scope of ACP?
Yes, this work is well within the shpere of interest of ACP. The paper presents a process-model simulation of the formation of SOA from a selection of precursors representative of diesel fuel emissions. The authors use parameterized product distributions, and extrapolate these distributions to larger precursors than those available in the model’s chemical mechanism.
2. Does the paper present novel concepts, ideas, tools, or data?
The simulations using the extrapolated parameterizations are compared to new chamber data. The comparisons are improved by the addition of the auto-oxidation process to the model training set, and by the consideration of wall losses in the chamber.
3. Are substantial conclusions reached?
The authors conclude that reactions in the wet inorganic phase make no significant contribution to SOA mass from the species considered: gas-particle partitioning is far more important. This result should be useful to future modeling studies.
4. Are the scientific methods and assumptions valid and clearly outlined?
The scientific method combines several previously-published concepts and equations, which are clearly outlined.
5. Are the results sufficient to support the interpretations and conclusions?
The interpretations and conclusions follow logically from the model/chamber comparisons. The Figures illustrate the analysis nicely.
- Line 300: discussion of Figure 3. I agree that C9D, C10D and C11C show good model-data comparisons. Comparisons C9C and C10B are not so good, and look more like comparisons C11A and C11B, which are said to be poor. The text should be clearer about this distinction. This comment des not detract from the paper’s overall conclusions.
- IMPORTANT. This study should add at least one sensitivity test that attempts to model the morning spike seen in the chamber observations, to support the authors’ assertion about its origin. (Line 325)
6. Is the description of experiments and calculations sufficiently complete and precise to allow their reproduction by fellow scientists (traceability of results)?
The experiments appear to be well documented.
7. Do the authors give proper credit to related work and clearly indicate their own new/original contribution?
In general, yes, credit is given. However this work is a development of a previous work by the same group (Zhou et al 2019), and it is not always immediately apparent what is new work and what is a restatement of that previous work. The model description, especially Sections 3.2 & 3.4, appears to be taken almost verbatim from Zhou et al 2019. That work should be acknowledged early in each relevant section and the authors should clearly say in the model description what parts of the model system are reused from that prior work, and what parts are new. Consideration should be given as to whether any of these sections can be simplified and shortened by referring the reader to the previous work. The same applies to Figure 1, which appears to be almost identical to Figure 1 of Zhou (2019)
8.Does the title clearly reflect the contents of the paper? 9.Does the abstract provide a concise and complete summary?
The title and abstract are clear and complete. I suggest that the model name should be added to the paper title.
10. Is the overall presentation well structured and clear?
The presentation and structure are generally clear. For specific minor points, see my list below.
11. Is the language fluent and precise?
The language is mostly fluent, precise and clear. I have a couple of minor requests for clarity, detailed below.
12. Are mathematical formulae, symbols, abbreviations, and units correctly defined and used?
- The symbols and abbreviations are much easier to follow after studying Figure 1.
- Section S4, Table S2: It is unfortunate that the accommodation coefficient alpha_w,i, and the polarizability alpha_i have such similar symbols, and that these could be confused with the mass-based stoichiometric coefficient alpha_i introduced in section 3.2. It might be appropriate to change one of these symbols, or at least to explicitly acknowledge the possibility for confusion.
13. Should any parts of the paper (text, formulae, figures, tables) be clarified, reduced, combined, or eliminated?
- The model description could be condensed by properly acknowledging and referring to the previous work, as already mentioned.
- Line 131-132, 197, 215 etc: This reviewer finds the abbreviations “or” and “in” confusing in the text because they can be mistaken for ordinary words. Please spell out the words “organic” & “inorganic” unless the abbreviations are used in combination with other symbols? Or, at a minimum, give them bold italic font to differentiate them from normal text.
14. Are the number and quality of references appropriate?
This is generally a good (but not exhaustive) guide to the recent literature on the relevant topics.
- The following works are cited in the text but are missing from the reference lists: Line 198: Pankow (1994); Line 272: Yap (2011); SI: Roldin (2019).
- Line 142: Pye 2019 and Xavier 2019 are not appropriate references for the identification of auto-oxidation reactions: they are modeling papers. It would be better to cite an early lab study (e.g. Sahetchian et al, Combust. Flame 1991; Crounse & Nielsen. J. Phys. Chem. Lett., 2013)
15. Is the amount and quality of supplementary material appropriate?
The supplementary material is helpful and appropriate.
- Tables S4, S5: please provide a key to or explanation of the chemical compound code names?
Other specific comments (major):
- Line 158: Please briefly discuss the conceptual movement of the reaction products between reactivity levels in your 51-box matrix. What does this represent, in a chemical-physical sense? Are there any constraints on the sum of all the dynamic alpha_sub_i values?
Specific comments (minor):
- Line 45: please correct the reference (Robinson et al., 2007)
- Lines 46-73: Please mention the names of the models used in Hodzic (2010), Cappa (2013), Zhang & Seinfeld (2013)
- Line 50: Note that, since Lee-Taylor et al used the larger linear alkanes as surrogates for all larger species, the fact that the majority of SOA precursors in their model were linear alkanes is a logical outcome of those SOA species having larger carbon numbers.
- Lines 61-65: Perhaps “neglect” would be kinder than “fail to consider”? The previous authors are likely well aware of the limitations of their approaches, and (at least in one case) mention those limitations explicitly.
- Line 86: “… the typical ozone mechanisms..” This seems vague. Did this study use several ozone mechanisms or just one?
- Line 133: “OMAR” needs subscripts.
- Lines 272-274: This sentence is confusing. Please rearrange it?
- Line 305: Do you mean “from” instead of “form”?
- Line 321, 323: Do you mean figure 5 instead of fig 6?
- Line 322: Do you mean figure 4 instead of fig 5?
- Line 341: why is it notable that previous experiments used H2O2 as a low-NOx OH source? What are the implications? Please discuss.
- Line 357: please specify whether you mean the relative or absolute contribution of OMAR?
- Line 370: do you mean Figure S6 instead of Fig 5?
- Line 625, 655: “OCEC is missing its “/”
Citation: https://doi.org/10.5194/egusphere-2022-681-RC1 -
RC2: 'Comment on egusphere-2022-681', Anonymous Referee #2, 20 Sep 2022
Madhu et al. present their work on modeling SOA yields from the oxidation of linear C10 through C20 alkanes. The model is described well and a lot of effort was invested in the research. However, the use of an outdoor chamber to conduct systematic tests of NOx and particle seed type on SOA yields was a flaw in the experiment design. The day-to-day variability in conditions, namely temperature and light, clouded the very impact of these variables the authors sought to test on SOA yields. Perhaps as a result, the demonstration of model peformance on SOA loadings (shown in Figures 2 through 5) is unconvincing. For instance, the explanation of the observed morning SOA burst in some of the experiments that the model fails to capture was seriously lacking.
Likewise, the discussion on the effect of NOx on autoxidation, the main driver of SOA formation they report, is thin. What is the implemented rate of autoxidation for each alkane precursor? A sensitivity test of the rate of autoxidation on SOA yield is reported on line 370 to be shown in Figure 5, which does not show that at all (only the sensitivity results of wall loss rates). What are the levels of Hox and Nox in the chamber for the duration of each experiment? Both would undoubtedly affect RO2 fates and lifetimes, neither of which is discussed in the section of Nox impact on autoxidation. Figure 6 shows that SOA yields increase with decreasing Nox. However, Figure 2 shows that the model underestimates SOA loading at high Nox and overestimates at low Nox. This, as a result, possibly undercuts the key model results of SOA yield vs Nox shown in Figure 6. Proper explanation of glaring caveats is not discussed. Authors go on to attribute the decrease in SOA yield with Nox to the partitioning of non-autox products, but that explanation is unclear and buried in a Table in the SI. A clearer demonstration with a graphic is needed.
I would have like to have seen optimization of model parameters (volatility, reactivity, and/or aging) using the suite of observations made. Or at the very least, show Figures 6 and 7 (which are the key figures) but with observations shown as well. This would ground the model and allow the authors to make conclusions that currently seem like extreme extrapolations. For instance, the levels of SOA in the chamber are beyond atmospherically relevant. The high loadings most likely affected the relative contribution of gas partitioning versus heterogeneous chemistry to SOA loading. Additionally, the authors conclude from this modeling exercise that straight chain alkanes are important for urban SOA - without showing much evidence. This is not backed up by the results shown.
It is my opinion that major revisions are required for this work to be published in ACP.
Minor
Line 153: Why 51 species? Where is this from?
Line 24: symbol in front of C15 did not render
Line 125: multiplication symbol did not render properly
Line 128: the “i” in alpha_i needs subscript
Figure S1. Is that starting from an alkoxy radical? Initial compound is blurry. How does the hydroperoxy group convert to a carbonyl? Please explain in the caption.
Instead of “or” for organic phase, “in” for inorganic, consider using “org” and “inorg” for clarity
Figure 2. Consider not using black and grey for MCM and MCM+autox. Something more different.
Line 321. Should cite Fig 5 not Fig 6?
Citation: https://doi.org/10.5194/egusphere-2022-681-RC2 - AC1: 'Comment on egusphere-2022-681', Myoseon Jang, 08 Nov 2022
Peer review completion
Journal article(s) based on this preprint
Azad Madhu et al.
Azad Madhu et al.
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 269 | 101 | 11 | 381 | 32 | 3 | 10 |
- HTML: 269
- PDF: 101
- XML: 11
- Total: 381
- Supplement: 32
- BibTeX: 3
- EndNote: 10
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1184 KB) - Metadata XML
-
Supplement
(513 KB) - BibTeX
- EndNote
- Final revised paper