the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 28 Aug 2023
Modeling the influence of carbon branching structure on SOA formation via multiphase reactions of alkanes
Abstract. Branched alkanes represent a significant proportion of hydrocarbons emitted in urban environments. To accurately predict the SOA budgets in urban environments, these branched alkanes should be considered as SOA precursors. However, the potential to form SOA from diverse branched alkanes under varying environmental conditions is currently not well understood. In this study, the Unified Partitioning Aerosol Phase Reaction (UNIPAR) model is extended to predict SOA formation via the multiphase reactions of various branched alkanes. Simulations with the UNIPAR model, which processes multiphase partitioning and aerosol phase reactions to form SOA, require a product distribution predicted from an explicit gas kinetic mechanism, whose oxygenated products are applied to create volatility-reactivity based lumping species array. Due to a lack of practically applicable explicit gas mechanisms, the prediction of the product distributions of various branched alkanes was approached with an innovative method that considers carbon lengths and branching structures. The lumping array of each branched alkane was primarily constructed using an existing lumping array of the linear alkane with the nearest vapor pressure. Generally, the vapor pressures of branched alkanes and their oxidation products are lower than those of linear alkanes with the same carbon length. In addition, increasing the branching of an alkane can also decrease the ability of alkanes to undergo autoxidation reactions that tend to form low-volatility products and significantly contribute to alkane SOA formation. To account for this, an autoxidation reduction factor, as a function of the degree and position of branching, was applied to the lumped groups which contain autoxidation products. The resulting product distributions were then applied to the UNIPAR model for predicting branched alkane SOA formation. The simulated SOA mass was compared to SOA data generated under varying experimental conditions (i.e., NOx levels, seed conditions, and humidity) in an outdoor photochemical smog chamber. Branched alkane SOA yields were significantly impacted by NOx levels but insignificantly impacted by seed conditions or humidity. The SOA formation from branched and linear alkanes in diesel fuel was simulated to understand the relative importance of branched and linear alkanes in a wide range of carbon numbers. Overall, branched alkanes account for more SOA mass than linear alkanes due to their higher contribution to diesel fuel. As anthropogenic emissions of hydrocarbons decrease, biogenic precursors tend to become increasingly important with regard to atmospheric organic aerosol. Unlike aromatics, which are almost exclusively sourced from anthropogenic emissions, alkanes can also be emitted from biogenic sources (i.e. plant wax) and they will remain a significant source of SOA.
- Preprint
(996 KB) - Metadata XML
-
Supplement
(1013 KB) - BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on egusphere-2023-1500', Anonymous Referee #1, 19 Sep 2023
Madhu et al. investigated the effects of carbon branching structure on SOA formation through both outdoor chamber experiments and modeling approaches. This comprehensive study employs multiple measurement techniques and examines various aspects of SOA formation within the UNIPAR model. The model performs well in reproducing chamber experiments. I believe this paper fits within the scope of ACP. However, there are a few concerns that need addressing before its publication in ACP.
Major comments.1) L219: It's unclear why branched alkanes have lower volatility compared to linear alkanes. I think methods for estimating vapor pressure like SIMPOL (Pankow and Asher, 2008) calculate the same value for both linear and branched alkanes. This is a crucial point throughout the paper and requires further justification.
2) L329-330, Figure S2: When comparing C19B and C19C, the seed conditions are the only differing factors, yet large discrepancies exist in ozone simulation capability. What could be the underlying reason? Are there measurement errors or issues with the CB06 mechanism? A similar issue is observed between C16B and C16C, although in this case, initial NOx/HC concentrations also vary.
3) L383-385: Alkanes with 1 and 2 branches show no significant difference. However, changes are distinct for others (0 vs 1, 2 vs 3, 3 vs 4). I initially thought ARF values for 1 and 2 would be similar, but Table S4 suggests otherwise. Could the authors explain this discrepancy?
4) L505: The results presented in this study are based on a timescale of less than half a day. In the real world, these compounds can undergo further aging. The authors should discuss this point when it comes to the real-world implications.
5) Data & Code Availability: Restricting access to data and code by simply stating they are "available upon request" seems inequitable, particularly in the year 2023. While it's understood that chamber data can be provided upon request, the rationale for not making the code publicly available is unclear. This lack of transparency is inconsistent with Copernicus's data policy (https://www.atmospheric-chemistry-and-physics.net/policies/data_policy.html#data_availability). I recommend utilizing free public repositories, such as Zenodo, for FAIR principles.
Minor comments.1) L46-47: Caplain et al. (2006) do not seem to discuss that branched alkanes represent larger proportions compared to linear alkanes. And I think it depends on the species. For example, n-butane is higher than isobutane but isopentane is higher than n-pentane.
2) L51: The proper name for the model is GECKO-A, not GECKOa.
3) L106: Which protocol was used for OC/EC measurements? Thermal optical transmittance or thermal optical reflectance? OC/EC concentrations depend on the protocol, so some discussion is needed for their quantifications.
4) L146-148: What about decomposition? Alkoxy radicals from branched alkanes could undergo decomposition, resulting in higher vapor pressures.
5) L171: A reference for DSMACC is missing.
6) L175: A detailed description of the CB6 mechanism is needed. There are several versions of CB6, such as CB6r1, CB6r2, CB6r3, and CB6r4. The authors may want to include a section describing the specific characteristics of the CB6 mechanism they used.
7) Table S1: Any reason for using Kwon and Atkinson, 1995, over a more recent framework like Jenkin et al. (2018)?
8) L370: Why does C12 show higher nitrate functional groups compared to C15 and C19? I think alkyl nitrate yield increases with increasing carbon numbers?
9) L400-401: While I agree with this point, how does the fate of RO2 change under high NOx conditions, particularly reactions like RO2 + HO2 vs RO2 + NO vs RO2 + RO2? I guess more RO2 + NO path under high-NOx conditions can enhance autooxidation?
10) L421-424: The logic is clear, but I can't see significant differences from Figure 8. To me, SOA changes due to temperature appear similar, regardless of NOx conditions.
11) L455: Genter et al. -> Gentner et al.
12) L462: Fig. 1 -> Fig. 11
13) L501-502: The temperature dependency could vary significantly based on the enthalpy of vaporization parameter. What value was used in this study?
Pankow, J. F. and Asher, W. E.: SIMPOL. 1: a simple group contribution method for predicting vapor pressures and enthalpies of vaporization of multifunctional organic compounds, Atmos. Chem. Phys., 8(10), 2773–2796, 2008.
Jenkin, M. E., Valorso, R., Aumont, B., Rickard, A. R. and Wallington, T. J.: Estimation of rate coefficients and branching ratios for gas-phase reactions of OH with aliphatic organic compounds for use in automated mechanism construction, Atmos. Chem. Phys., 18(13), 9297–9328, 2018.Citation: https://doi.org/10.5194/egusphere-2023-1500-RC1 -
RC2: 'Comment on egusphere-2023-1500', Anonymous Referee #2, 05 Nov 2023
Madhu et al. have used a process-level model to study SOA formation from branched alkanes and evaluated their model against outdoor chamber experiments performed on several long-chain branched (and linear) alkanes. They found that the model performed modestly in reproducing the observed SOA. Sensitivity simulations revealed a stronger dependence on NOx, relative to seed conditions or relative humidity. They performed a case study with diesel fuel, a potential source for branched alkane emissions, to highlight its importance for ambient air quality.
Branched alkanes, as precursors for SOA (and ozone), are ubiquitous and likely come from a variety of fossil fuel and biomass combustion sources. The study, which aims to study SOA mechanisms for this class of SOA precursors, is well motivated. The manuscript is easily understood, but some of the methods, model application, and data analysis need more work; see comments below. I do not recommend publication in ACP until the editor (and/or I) have had a chance to review the updated manuscript.
Major comments
- Introduction: My impression of UNIPAR is that it is a ‘medium complexity’ model that can be leveraged to study aerosol systems as a box model but too computationally expensive to be added to a chemical transport or chemistry climate model. I am inferring this from the model description. I would strongly encourage the authors to consider adding text that places UNIPAR as an appropriate model for the science to be addressed and how findings from UNIPAR can be used to build reduced-form models for atmospheric models.
- Table 1: If one is to peruse the molecules found in gasoline and diesel exhaust (one could also do this for oil and gas and volatile chemical products – other important sources for atmospheric alkanes), one would be hard pressed to find the types of molecules shown in Table 1. See, for instance, the work from Allen Robinson, Albert Presto, Drew Gentner, and Allen Goldstein between 2010 and 2020; I apologize for not listing other important contributors to this since I am going off of memory. Most branched alkanes seem to be much more ‘branched’ with a much smaller linear carbon backbone. A rationale is needed for why the species in Table 1 are relevant for the real atmosphere.
- Line 188: The comment that there are no explicit mechanisms for branched alkanes is not true. I see several branched alkane species in MCM (albeit not as heavily branched as those found in vehicle exhaust); see: https://mcm.york.ac.uk/MCM/browse. Similarly, I see pre-determined mechanisms for several branched alkanes for GECKO-A, suggesting that mechanism generation is possible for the branched alkanes studied in this work; see: https://www.acom.ucar.edu/gecko/output-library.shtml.
- ‘Lumping array’: My understanding is that MCM mechanisms were used to simplify the product distribution from oxidation of linear alkanes and adjusted for branched alkanes to create one of the key inputs (i.e., alpha?) to the UNIPAR model. This approach is very poorly described, and I encourage the authors to improve this. Also, how the mechanism is used to inform a dynamically varying product distribution using the aging factor is unclear.
- Line 230 onwards: It’s unclear what the basis for the ARF rules is. Do these come from more fundamental studies that explore the probability of isomerization modulated by branching?
- Equation 6: While I appreciate the treatment of separate phases based on the organic aerosol composition and varying RH, I was somewhat surprised to see Raoult’s law being used directly to simulate the partitioning of organic species into an inorganic phase. Is this approach informed using more chemically-resolved models (e.g., AIOMFAC)?
- Figures 2-3: The model-measurement comparison is less than stellar and there are experiments where the model deviates significantly from the observations (12A, 16B). Since the results in Section 4.3 and beyond are based entirely on model simulations, how does one interpret the sensitivity results given the imperfect model comparisons in Figures 2 and 3. Are any of the model inputs/processes tweaked for the branched alkane results presented in this manuscript, relative to the linear alkane system? Can these tweaks, if they improve model performance, tell us something about mechanistic differences between linear and branched alkanes?
- Sections 4.3-4.4: Do the results shown in Figure 5, qualitatively, align with those from chamber studies of branched alkanes (Lim and Ziemann, ES&T, 2009; Tkacik et al., ES&T, 2012; Loza et al., ACP, 2014)? Loza et al. (ACP, 2014) found that SOA mass yields were higher for alkanes under high NOx conditions, which contrasts with that shown in Figure 6. How does one reconcile these differences? How was the enthalpy of vaporization modeled for the SOA species, which should strongly influence the temperature sensitivity results shown in Figure 8? How important are oligomers in the end-of-experiment SOA? Is the low sensitivity to oligomerization rates shown in Figure 9 simply because oligomers don’t account much for the total SOA? Overall, there are details missing in these sections that help place the results in context.
- Section 4.6: I don’t think the case study with diesel fuel is atmospherically relevant since diesel fuel rarely evaporates to emit alkanes. What could be useful is to use a speciation of diesel (or gasoline) exhaust to look at the relative importance of linear and branched alkanes. This analysis also misses the point that a lot of diesel fuel and diesel exhaust is comprised of branched-cyclic species, yields for which are relatively less certain than those for linear and branched alkanes.
Minor comments:
- Line 29: The statement about biogenic VOCs being more important in the future needs backing. Also, are plant wax emissions of alkanes – assuming most of these are branched alkanes – are a significant source relative to other biogenic VOCs (isoprene, terpenes)?
- Introduction: The importance of branched alkanes as SOA precursors to the atmosphere could be better stated by presenting findings from the literature. The current discussion is too qualitative.
- Line 104: Could density estimation methods (e.g., from H:C and O:C data) or previous literature be used instead to determine SOA density?
- Line 140: Please define ‘functionality distributions’.
- Line 145: It’s a little unclear if the lumping is performed for the precursor or oxidation products or both. Clarify what ‘lumping array’ means. I also think ‘lumped array’ is a better use of the term.
- Line 183: Are there experimental or theoretical (e.g., molecular dynamics) studies that provide evidence for autooxidation pathways for linear alkanes? Is autooxidation also desirable for branched alkanes?
- Equations 1-3: A rationale needs to be provided for this simplification in treating aging reactions and an example of how the aging kernel works. Citing previous work is necessary but not sufficient.
- Sections 3.6-3.7: As most of these formulations to simulate particle-phase reactions and chamber artifacts come from previous work from this group, text needs to be added to summarize if these formulations have been evaluated against direct observations and what the remaining uncertainties are.
- Line 367: Is there literature evidence for this statement? If not, can you describe the mechanism?
Citation: https://doi.org/10.5194/egusphere-2023-1500-RC2 -
AC1: 'Comment on egusphere-2023-1500', Myoseon Jang, 18 Dec 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-1500/egusphere-2023-1500-AC1-supplement.pdf
Status: closed
-
RC1: 'Comment on egusphere-2023-1500', Anonymous Referee #1, 19 Sep 2023
Madhu et al. investigated the effects of carbon branching structure on SOA formation through both outdoor chamber experiments and modeling approaches. This comprehensive study employs multiple measurement techniques and examines various aspects of SOA formation within the UNIPAR model. The model performs well in reproducing chamber experiments. I believe this paper fits within the scope of ACP. However, there are a few concerns that need addressing before its publication in ACP.
Major comments.1) L219: It's unclear why branched alkanes have lower volatility compared to linear alkanes. I think methods for estimating vapor pressure like SIMPOL (Pankow and Asher, 2008) calculate the same value for both linear and branched alkanes. This is a crucial point throughout the paper and requires further justification.
2) L329-330, Figure S2: When comparing C19B and C19C, the seed conditions are the only differing factors, yet large discrepancies exist in ozone simulation capability. What could be the underlying reason? Are there measurement errors or issues with the CB06 mechanism? A similar issue is observed between C16B and C16C, although in this case, initial NOx/HC concentrations also vary.
3) L383-385: Alkanes with 1 and 2 branches show no significant difference. However, changes are distinct for others (0 vs 1, 2 vs 3, 3 vs 4). I initially thought ARF values for 1 and 2 would be similar, but Table S4 suggests otherwise. Could the authors explain this discrepancy?
4) L505: The results presented in this study are based on a timescale of less than half a day. In the real world, these compounds can undergo further aging. The authors should discuss this point when it comes to the real-world implications.
5) Data & Code Availability: Restricting access to data and code by simply stating they are "available upon request" seems inequitable, particularly in the year 2023. While it's understood that chamber data can be provided upon request, the rationale for not making the code publicly available is unclear. This lack of transparency is inconsistent with Copernicus's data policy (https://www.atmospheric-chemistry-and-physics.net/policies/data_policy.html#data_availability). I recommend utilizing free public repositories, such as Zenodo, for FAIR principles.
Minor comments.1) L46-47: Caplain et al. (2006) do not seem to discuss that branched alkanes represent larger proportions compared to linear alkanes. And I think it depends on the species. For example, n-butane is higher than isobutane but isopentane is higher than n-pentane.
2) L51: The proper name for the model is GECKO-A, not GECKOa.
3) L106: Which protocol was used for OC/EC measurements? Thermal optical transmittance or thermal optical reflectance? OC/EC concentrations depend on the protocol, so some discussion is needed for their quantifications.
4) L146-148: What about decomposition? Alkoxy radicals from branched alkanes could undergo decomposition, resulting in higher vapor pressures.
5) L171: A reference for DSMACC is missing.
6) L175: A detailed description of the CB6 mechanism is needed. There are several versions of CB6, such as CB6r1, CB6r2, CB6r3, and CB6r4. The authors may want to include a section describing the specific characteristics of the CB6 mechanism they used.
7) Table S1: Any reason for using Kwon and Atkinson, 1995, over a more recent framework like Jenkin et al. (2018)?
8) L370: Why does C12 show higher nitrate functional groups compared to C15 and C19? I think alkyl nitrate yield increases with increasing carbon numbers?
9) L400-401: While I agree with this point, how does the fate of RO2 change under high NOx conditions, particularly reactions like RO2 + HO2 vs RO2 + NO vs RO2 + RO2? I guess more RO2 + NO path under high-NOx conditions can enhance autooxidation?
10) L421-424: The logic is clear, but I can't see significant differences from Figure 8. To me, SOA changes due to temperature appear similar, regardless of NOx conditions.
11) L455: Genter et al. -> Gentner et al.
12) L462: Fig. 1 -> Fig. 11
13) L501-502: The temperature dependency could vary significantly based on the enthalpy of vaporization parameter. What value was used in this study?
Pankow, J. F. and Asher, W. E.: SIMPOL. 1: a simple group contribution method for predicting vapor pressures and enthalpies of vaporization of multifunctional organic compounds, Atmos. Chem. Phys., 8(10), 2773–2796, 2008.
Jenkin, M. E., Valorso, R., Aumont, B., Rickard, A. R. and Wallington, T. J.: Estimation of rate coefficients and branching ratios for gas-phase reactions of OH with aliphatic organic compounds for use in automated mechanism construction, Atmos. Chem. Phys., 18(13), 9297–9328, 2018.Citation: https://doi.org/10.5194/egusphere-2023-1500-RC1 -
RC2: 'Comment on egusphere-2023-1500', Anonymous Referee #2, 05 Nov 2023
Madhu et al. have used a process-level model to study SOA formation from branched alkanes and evaluated their model against outdoor chamber experiments performed on several long-chain branched (and linear) alkanes. They found that the model performed modestly in reproducing the observed SOA. Sensitivity simulations revealed a stronger dependence on NOx, relative to seed conditions or relative humidity. They performed a case study with diesel fuel, a potential source for branched alkane emissions, to highlight its importance for ambient air quality.
Branched alkanes, as precursors for SOA (and ozone), are ubiquitous and likely come from a variety of fossil fuel and biomass combustion sources. The study, which aims to study SOA mechanisms for this class of SOA precursors, is well motivated. The manuscript is easily understood, but some of the methods, model application, and data analysis need more work; see comments below. I do not recommend publication in ACP until the editor (and/or I) have had a chance to review the updated manuscript.
Major comments
- Introduction: My impression of UNIPAR is that it is a ‘medium complexity’ model that can be leveraged to study aerosol systems as a box model but too computationally expensive to be added to a chemical transport or chemistry climate model. I am inferring this from the model description. I would strongly encourage the authors to consider adding text that places UNIPAR as an appropriate model for the science to be addressed and how findings from UNIPAR can be used to build reduced-form models for atmospheric models.
- Table 1: If one is to peruse the molecules found in gasoline and diesel exhaust (one could also do this for oil and gas and volatile chemical products – other important sources for atmospheric alkanes), one would be hard pressed to find the types of molecules shown in Table 1. See, for instance, the work from Allen Robinson, Albert Presto, Drew Gentner, and Allen Goldstein between 2010 and 2020; I apologize for not listing other important contributors to this since I am going off of memory. Most branched alkanes seem to be much more ‘branched’ with a much smaller linear carbon backbone. A rationale is needed for why the species in Table 1 are relevant for the real atmosphere.
- Line 188: The comment that there are no explicit mechanisms for branched alkanes is not true. I see several branched alkane species in MCM (albeit not as heavily branched as those found in vehicle exhaust); see: https://mcm.york.ac.uk/MCM/browse. Similarly, I see pre-determined mechanisms for several branched alkanes for GECKO-A, suggesting that mechanism generation is possible for the branched alkanes studied in this work; see: https://www.acom.ucar.edu/gecko/output-library.shtml.
- ‘Lumping array’: My understanding is that MCM mechanisms were used to simplify the product distribution from oxidation of linear alkanes and adjusted for branched alkanes to create one of the key inputs (i.e., alpha?) to the UNIPAR model. This approach is very poorly described, and I encourage the authors to improve this. Also, how the mechanism is used to inform a dynamically varying product distribution using the aging factor is unclear.
- Line 230 onwards: It’s unclear what the basis for the ARF rules is. Do these come from more fundamental studies that explore the probability of isomerization modulated by branching?
- Equation 6: While I appreciate the treatment of separate phases based on the organic aerosol composition and varying RH, I was somewhat surprised to see Raoult’s law being used directly to simulate the partitioning of organic species into an inorganic phase. Is this approach informed using more chemically-resolved models (e.g., AIOMFAC)?
- Figures 2-3: The model-measurement comparison is less than stellar and there are experiments where the model deviates significantly from the observations (12A, 16B). Since the results in Section 4.3 and beyond are based entirely on model simulations, how does one interpret the sensitivity results given the imperfect model comparisons in Figures 2 and 3. Are any of the model inputs/processes tweaked for the branched alkane results presented in this manuscript, relative to the linear alkane system? Can these tweaks, if they improve model performance, tell us something about mechanistic differences between linear and branched alkanes?
- Sections 4.3-4.4: Do the results shown in Figure 5, qualitatively, align with those from chamber studies of branched alkanes (Lim and Ziemann, ES&T, 2009; Tkacik et al., ES&T, 2012; Loza et al., ACP, 2014)? Loza et al. (ACP, 2014) found that SOA mass yields were higher for alkanes under high NOx conditions, which contrasts with that shown in Figure 6. How does one reconcile these differences? How was the enthalpy of vaporization modeled for the SOA species, which should strongly influence the temperature sensitivity results shown in Figure 8? How important are oligomers in the end-of-experiment SOA? Is the low sensitivity to oligomerization rates shown in Figure 9 simply because oligomers don’t account much for the total SOA? Overall, there are details missing in these sections that help place the results in context.
- Section 4.6: I don’t think the case study with diesel fuel is atmospherically relevant since diesel fuel rarely evaporates to emit alkanes. What could be useful is to use a speciation of diesel (or gasoline) exhaust to look at the relative importance of linear and branched alkanes. This analysis also misses the point that a lot of diesel fuel and diesel exhaust is comprised of branched-cyclic species, yields for which are relatively less certain than those for linear and branched alkanes.
Minor comments:
- Line 29: The statement about biogenic VOCs being more important in the future needs backing. Also, are plant wax emissions of alkanes – assuming most of these are branched alkanes – are a significant source relative to other biogenic VOCs (isoprene, terpenes)?
- Introduction: The importance of branched alkanes as SOA precursors to the atmosphere could be better stated by presenting findings from the literature. The current discussion is too qualitative.
- Line 104: Could density estimation methods (e.g., from H:C and O:C data) or previous literature be used instead to determine SOA density?
- Line 140: Please define ‘functionality distributions’.
- Line 145: It’s a little unclear if the lumping is performed for the precursor or oxidation products or both. Clarify what ‘lumping array’ means. I also think ‘lumped array’ is a better use of the term.
- Line 183: Are there experimental or theoretical (e.g., molecular dynamics) studies that provide evidence for autooxidation pathways for linear alkanes? Is autooxidation also desirable for branched alkanes?
- Equations 1-3: A rationale needs to be provided for this simplification in treating aging reactions and an example of how the aging kernel works. Citing previous work is necessary but not sufficient.
- Sections 3.6-3.7: As most of these formulations to simulate particle-phase reactions and chamber artifacts come from previous work from this group, text needs to be added to summarize if these formulations have been evaluated against direct observations and what the remaining uncertainties are.
- Line 367: Is there literature evidence for this statement? If not, can you describe the mechanism?
Citation: https://doi.org/10.5194/egusphere-2023-1500-RC2 -
AC1: 'Comment on egusphere-2023-1500', Myoseon Jang, 18 Dec 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-1500/egusphere-2023-1500-AC1-supplement.pdf
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 296 | 90 | 26 | 412 | 49 | 14 | 19 |
- HTML: 296
- PDF: 90
- XML: 26
- Total: 412
- Supplement: 49
- BibTeX: 14
- EndNote: 19
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1