the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 20 Jun 2023
Gas-particle partitioning of toluene oxidation products: an experimental and modeling study
Abstract. Aromatic hydrocarbons represent a large fraction of anthropogenic volatile organic compounds and significantly contribute to tropospheric ozone and secondary organic aerosol (SOA) formation. Toluene photooxidation experiments were carried out in an oxidation flow reactor (OFR). We identified and quantified the gaseous and particulate reaction products at 280, 285 and 295 K using a proton-transfer reaction time-of-flight mass spectrometer (PTR-ToF-MS) coupled to a CHemical Analysis of aeRosol ONline (CHARON) inlet. The reaction products accounted for both ring-retaining compounds such as cresols, benzaldehyde, nitrophenols, nitrotoluene, bicyclic intermediate compounds, as well as ring-scission products such as dicarbonyls, cyclic anhydrides, small aldehydes and acids. The chemical system exhibited a volatility distribution mostly in the semi-volatile (SVOCs – semi-volatile organic compounds) regime. The saturation concentration (Ci*) values of the identified compounds were mapped onto the two-dimensional volatility basis set (2D-VBS). Temperature decrease caused a shift of Ci* towards lower values while there was no clear relationship between Ci* and oxidation state. The CHARON PTR-ToF-MS instrument identified and quantified approximately 70–80 % of the total organic mass measured by an aerosol mass spectrometer (AMS). The experiments were reproduced by simulating SOA formation with the SSH-aerosol box model. A semi-detailed mechanism for toluene gaseous oxidation was developed. It is based on the MCM and GECKO-A deterministic mechanisms modified following the literature in particular to update cresols and ring-scission chemistry. The new mechanism improved secondary species representation with an increment of the major identified species (+35 % in number). Light compounds formation (i.e. m/z < 100) is enhanced and accumulation of heavy compounds (i.e. m/z ≥ 100) is reduced, especially in the gas phase. Additional tests on (i) partitioning processes such as condensation into aqueous phase, (ii) interactions of organic compounds between themselves and with inorganics and (iii) wall losses were also performed. When all these processes were taken into account the simulated SOA mass concentration showed a much better agreement with the experimental results. Finally, an irreversible partitioning pathway for methylglyoxal was introduced and considerably improved the model results, opening a way to further developments of partitioning in models. Our results underline that the volatility itself is not sufficient to explain the partitioning between gas and particle phase: the organic and the aqueous phases need to be taken into account as well as interactions between compounds in the particle phase.
-
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
(1800 KB)
-
Supplement
(3784 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1800 KB) - Metadata XML
-
Supplement
(3784 KB) - BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2023-1290', Anonymous Referee #1, 03 Aug 2023
This manuscript reports measurement and modeling of gas-phase and particle- phase composition and mass concentration of SOA formed in toluene+OH reaction system. Gas- and particle-phase components are measured by CHARON- PTR-MS. Gas- and particle-phase composition was modeled using SSH-aerosol model and updated chemistry of cresol and ring-scission chemistry and varying levels of complexity on gas-particle partitioning. It was found that SOA yields at different temperatures were consistent previous studies. According to measured volatility, most compounds belonged to SVOC. Volatility did not show clear relationship with oxidation state. PTR-MS quantified 70-80% of total organic mass measured by AMS. C7 compounds contributed the largest fraction in total products, both in the gas-phase and particle-phase. C1-6 compounds also contributed significantly. Models with varying levels of complexity on gas-particle partitioning did not well reproduce the measured gas-phase and particle-phase composition. Light compounds (m/z <100) were underestimated and the heavy compounds (m/z >100) were overestimated. However, the model including partitioning into aqueous phase, interactions between organics and between organics and inorganics achieved good agreement with measured SOA mass concentration. Irreversible uptake of methylglyoxal in the model further improved the simulation of SOA mass concentration.
Determining the reaction products of SOA and related gas-phase products on molecular level and elucidating key processes affecting its concentrations are key to predicting its ambient concentration and understanding its impact. This study addresses both aspects. Particularly, besides modeling SOA mass concentration, this study attempt to model the chemical composition of SOA via comparing with measurement, which is novel and not much reported in the literature. The manuscript is generally well-written. I have a few comments for the authors to consider before publication.
Specific comments
- Before digging into the reason for the difference between modeled and measured composition, one need be sure that the difference is minimally influenced by the measurement artefacts. Is this possible that small compounds, particularly those less oxygenated C1-C4 compounds in particle-phase, are fragment of parent products or decomposition during heating in CHARON? It is surprising to see these compounds in the particle-phase due to their high volatility. It would helpful to elaborate the method used to exclude the artefacts and/or discuss their influence on the measured chemical composition.
- In Fig. 2, no oligomers (C>7) compounds were shown. Are there any oligomers (C>7) detected? Can CHARON PTR MS see them?
- I suggest adding a figure of reaction schemes of toluene+OH reaction highlight the new reaction mechanisms added to the model in SI so that readers can better follow.
- L167, which data does this statement based on?
- L236, what does “long range interactions” mean? It would be helpful to briefly explain it.
Technical comments
- L35, it is not clear that what the “enhanced” is compared with.
- L287, “as” should be “and”?
- L407, “cercles” should be “circles”. Also it would helpful to clarify that it is the diameter or area of circles that denotes the mass in Fig. 3.
In the caption of Fig. 3, there are no “pie charts”. Please check.
- L451, the “of” after “24 %” should be deleted.
- L505, “47%”, is this correct or should it be 28%?
- L545, a comma is missing after “(a).
- L482, a space is missing after “distribution”.
Citation: https://doi.org/10.5194/egusphere-2023-1290-RC1 -
AC1: 'Reply on RC1', Victor Lannuque, 20 Sep 2023
We thank the reviewers for their comments and suggestions on the manuscript. We outline below responses to the points raised by each referee and summarize the changes made to the revised manuscript. We also provide a revised version of the manuscript with highlighted modifications.
Reponses to RC1
> 1. Before digging into the reason for the difference between modeled and measured composition, one need be sure that the difference is minimally influenced by the measurement artefacts. Is this possible that small compounds, particularly those less oxygenated C1-C4 compounds in particle-phase, are fragment of parent products or decomposition during heating in CHARON? It is surprising to see these compounds in the particle-phase due to their high volatility. It would helpful to elaborate the method used to exclude the artefacts and/or discuss their influence on the measured chemical composition.
Thermal decomposition is typically not observed in CHARON PTR-ToF-MS instruments (except for hydroperoxides and labile sugars). It has, however, been shown in previous work (e.g., Gkatzelis et al., 2018; Leglise et al. 2019) that at the conditions used in this study (E/N = 105 Td) scission of the C-C bond may indeed occur. Therefore a fraction of the low C compounds found in the particle phase may thus be measurement artifacts.
> 2. In Fig. 2, no oligomers (C>7) compounds were shown. Are there any oligomers (C>7) detected? Can CHARON PTR MS see them?
Oligomers can be observed with CHARON PTR-ToF-MS instruments. It is, however, likely that at the operating conditions used in this study (E/N = 105 Td) oligomers fragment upon protonation.
> 3. I suggest adding a figure of reaction schemes of toluene+OH reaction highlight the new reaction mechanisms added to the model in SI so that readers can better follow.
The suggested figure was added to SI (Fig. S6) and is now mentioned in the text.
> 4. L167, which data does this statement based on?
According to Jenkin et al. (2019) SAR, the RO2 + RO2 reactions mainly lead to the formation of alkoxy radicals as RO2 + NO reactions. Under atmospheric conditions, the RO2 + CH3O2 (the most abundant RO2 in atmosphere) reaction represents about 10 % of the RO2 reactivity (Lannuque et al., 2018). If we have here decided to not consider all RO2 + RO2 reactions it is to limit the already relatively large size of our mechanisms. We are aware that this means ignoring the formation of certain minor secondary compounds, notably C>7. This choice was made considering (1) the high NOx levels limiting the importance of RO2 + RO2 reactions and (2) the absence of detection of such compounds in our experiments.
> 5. L236, what does “long range interactions” mean? It would be helpful to briefly explain it.
Middle- and long-range interactions correspond to interactions between the organic compounds and charged molecules (especially inorganic ions, the only considered here) in water phase. This term is used by opposition of the interactions between the different uncharged organic molecules which occur at a shorter range.
> 1. L35, it is not clear that what the “enhanced” is compared with.
It is compared to a reference mechanism using MCM and GECKO-A. It is now specified in the text.
> 2. L287, “as” should be “and”?
The text was modified.
> 3. L407, “cercles” should be “circles”. Also it would helpful to clarify that it is the diameter or area of circles that denotes the mass in Fig. 3.
The area of the circles has been used. The text was modified and the legend of Fig. 3 was detailed.
> In the caption of Fig. 3, there are no “pie charts”. Please check.
The mention to pie charts was removed.
> 1. L451, the “of” after “24 %” should be deleted.
The text was modified.
> 2. L505, “47%”, is this correct or should it be 28%?
You are right, the mistake was corrected and 47% was replaced by 30% (28 + 2 %).
> 3. L545, a comma is missing after “(a).
> 4. L482, a space is missing after “distribution”.The text was corrected.
References :
Gkatzelis, G. I., Tillmann, R., Hohaus, T., Müller, M., Eichler, P., Xu, K.-M., Schlag, P., Schmitt, S. H., Wegener, R., Kaminski, M., Holzinger, R., Wisthaler, A., and Kiendler-Scharr, A.: Comparison of three aerosol chemical characterization techniques utilizing PTR-ToF-MS: a study on freshly formed and aged biogenic SOA, Atmos. Meas. Tech, 11, 1481–1500, https://doi.org/10.5194/amt-11-1481-2018, 2018.
Jenkin, M. E., Valorso, R., Aumont, B., and Rickard, A. R.: Estimation of rate coefficients and branching ratios for reactions of organic peroxy radicals for use in automated mechanism construction, Atmos. Chem. Phys., 19, 7691–7717, https://doi.org/10.5194/ACP-19-7691-2019, 2019.
Lannuque, V., Camredon, M., Couvidat, F., Hodzic, A., Valorso, R., Madronich, S., Bessagnet, B., and Aumont, B.: Exploration of the influence of environmental conditions on secondary organic aerosol formation and organic species properties using explicit simulations: development of the VBS-GECKO parameterization, Atmos. Chem. Phys., 18, 13411–13428, https://doi.org/10.5194/acp-18-13411-2018, 2018.
Leglise, J., Müller, M., Piel, F., Otto, T., and Wisthaler, A.: Bulk Organic Aerosol Analysis by Proton-Transfer-Reaction Mass Spectrometry: An Improved Methodology for the Determination of Total Organic Mass, O:C and H:C Elemental Ratios, and the Average Molecular Formula, Anal. Chem., 91, 12619–12624, https://doi.org/10.1021/acs.analchem.9b02949, 2019.
Citation: https://doi.org/10.5194/egusphere-2023-1290-AC1
-
RC2: 'Comment on egusphere-2023-1290', Anonymous Referee #2, 09 Aug 2023
This manuscript describes an experimental and modeling study of the formation and gas-particle partitioning of organic compounds from the photooxidation of toluene. The toluene photooxidation experiments were conducted in a Pyrex OFR, with the gaseous and particulate organic products measured simultaneously by two separate PTR-ToF-MS without and with a thermal desorption inlet (CHARON), respectively. It is shown that the particulate organic species quantified by CHARON PTR-MS can explain approximately 70% of the secondary organic aerosol (SOA) mass measured by an AMS. A semi-explicit gas-phase oxidation mechanism of toluene was developed to reproduce the formation and distribution of organic products in both gas and particle phases. It is found that with the modifications of cresol chemistry, aromatic ring opening and furan formation chemistry, as well as relevant reaction kinetics based on a structure-activity relationship, the model can better represent the speciation of the oxidation products in the gas phase, but does not improve the prediction of the product distribution in the particle phase. Further simulations with an aerosol box model (SSH), incorporated with the newly developed gas-phase oxidation mechanism, show that the formation of SOA is highly sensitive to the parameterization of the phase partitioning processes, vapor wall losses in the OFR, and the irreversible uptake of small molecules such as methylglyoxal. Overall, this study is interesting and provides valuable insights into the physicochemical processes associated with toluene SOA formation as well as the further development of SOA models. The manuscript can be considered for publication in ACP after the following comments are addressed.
Line 98: Additional experimental details such as the rate and residence time of the gas flow in the OFR and the initial concentrations of toluene should be provided in the experimental section.
Line 132: Have any experiments been done to verify that all the organics in SOA are thermally desorbed at a temperature of 413 K in the CHARON inlet? Was the fragmentation of organic compounds during the thermal desorption of SOA significant?
Line 145: Are the C* values calculated from the measured partitioning coefficients using Equation 2? What are the influences of the potential non-equilibrium phase partitioning (for example under low RH conditions) and non-ideality of the condensed phase on such calculations?
Line 210-214: The formation pathways of highly oxygenated organic molecules (HOMs) from toluene photooxidation are also included in the newly developed mechanism. Do they make a significant difference to the speciation of the gaseous and particulate organic species, especially for those with high oxygen numbers?
Line 280: What are the size distributions of toluene SOA formed under typical experimental conditions? Are the formed SOA particles monodispersed as assumed here? Given the low pre-existing seed aerosol concentrations (~9 μg/m3), I would expect that there is significant nucleation and new particle formation during toluene photooxidation. As a result, polydispersed aerosols are likely generated, e.g., a larger particle mode comprised of inorganic-organic mixed particles and a smaller particle model primarily with pure SOA particles. If this is the case, how would it affect the phase partitioning behaviors of organic compounds as well as the irreversible uptake of small species such as methylglyoxal.
Line 287: These parameters are expected to significantly impact the gas-phase oxidation chemistry, phase partitioning, and/or particle-phase chemistry (e.g, irreversible uptake of methylglyoxal), but it is found in the present study that they have little impact on the overall speciation of the oxidation products. The authors only present the data obtained at different temperatures in the manuscript. I suggest also providing the data obtained for other parameters in the SI.
Line 356: In addition to losing a H2O molecule, how likely do the toluene oxidation products fragment via C-C scission during PTR-MS measurements? How would it affect the discrimination between parent and fragment ions?
Line 401-402: Please clarify what type of heterogeneous reactions on the reactor wall can produce these small molecules.
Line 405: Although the authors are able to determine the gas-particle partitioning coefficients (Kp) of the toluene oxidation products, they did not present such data in the manuscript. As the Kp can offer additional insights into the phase partitioning behaviors of the oxidation products, I suggest the authors add an analysis of those data as well as their measurement-model comparisons in the manuscript.
Line 413: It seems that some content is missing before the sentence “Other 20 ions were detected…”. Please check.
Line 425: Figure 3 does not include any pie charts described in the figure caption. Also, there are no descriptions/discussions regarding the comparisons between organic compounds measured by CHARON PTR-MS and the SOA mass measured by AMS in the text. Please double check.
Line 430-431: The volatility is affected not only by the oxidation state, but also by the molecular size. Therefore, a trend for OSc versus C* may be observed if the data are further discriminated by the carbon number. In addition, Fig. S6 should be Fig. S5.
Line 608: Why does the inclusion of the wall loss parameterization lead to the reduction of the concentration of all m/z values in the particle phase but only the heavier compounds in the gas phase?
In addition, the following language errors should be corrected.
Line 91: produced -> produce
Line 249: wall losses -> wall loss effects
Line 273: Delete “controlled”.
Line 287: as -> as well as
Line 298: as -> such as
Line 346: by -> followed by
Line 407: cercles -> circles
Line 442: that -> than
Line 443: Delete either of “only”.
Citation: https://doi.org/10.5194/egusphere-2023-1290-RC2 -
AC2: 'Reply on RC2', Victor Lannuque, 20 Sep 2023
We thank the reviewers for their comments and suggestions on the manuscript. We outline below responses to the points raised by each referee and summarize the changes made to the revised manuscript. We also provide a revised version of the manuscript with highlighted modifications.
Reponses to RC2
> Line 98: Additional experimental details such as the rate and residence time of the gas flow in the OFR and the initial concentrations of toluene should be provided in the experimental section.
Experiments were carried out using a total flow varying from 1.3 to 1.8 L/min corresponding to a residence time of approximately 10-13 minutes. These details have been introduced in the experimental section.
> Line 132: Have any experiments been done to verify that all the organics in SOA are thermally desorbed at a temperature of 413 K in the CHARON inlet? Was the fragmentation of organic compounds during the thermal desorption of SOA significant?
It has been shown in the literature that at the temperatures used in this study SVOCs, LVOCs and even ELVOCS are vaporized (e.g. Piel et al., 2021). It is important to note that vaporization is affected on the walls of the thermal desorption unit at reduced at reduce pressure (few mbar). Piel et al. (2021) also mention that certain types of thermal decomposition such as the decarboxylation of organic acids does not occur at the temperatures used in this study. The same authors found evidence for the thermal decomposition of hydroperoxides in the CHARON inlet.
> Line 145: Are the C* values calculated from the measured partitioning coefficients using Equation 2? What are the influences of the potential non-equilibrium phase partitioning (for example under low RH conditions) and non-ideality of the condensed phase on such calculations?
The C* were calculated using equations 1 and 2. According to Shiraiwa and Seinfeld (2012): the equilibrium time is achieved from seconds to minutes for relatively high volatility organic compounds into liquid particles. However, equilibrium time can increase to hours or days for organic aerosol when the particles are large, semi-solid particles, of low volatility, and low mass loadings. Instantaneous equilibrium partitioning may lead to substantial overestimation of particle mass concentration and underestimation of gas-phase concentration.
In our study most of the identified SOA compounds identified are semi-volatile (Table 1). Moreover, monodispersed seeds+SOA had a size was about 150-216 nm so our particles are quite small to establish an equilibrium with the gas phase. In addition, the total mass concentration was on average 13.9±3.4 µg m-3 for the experiments at 295 K and 17.4±4.9 µg m-3 for the experiments at 280 K, the mass loading is high enough for the establishment of the equilibrium. Indeed according to Figure 4 of Shiraiwa and Seinfeld (2012), for semi-solid SVOCs with a diameter of 150 nm and organic mass concentration 14-17 µg m-3 the equilibrium timescale of SOA partitioning is about 10 min. Given the fact that the residence time in the OFR was about 11-13 minutes we conclude that particle and gas phases were practically in equilibrium. Furthermore, in preliminary modeling tests, gas-particle partitioning was represented using both methods: one considering thermodynamic equilibrium at each time step, and one following a dynamic phase transfer approach. The two methods showed no significant differences in partitioning under our simulation conditions.
The non-ideallity, for individual organic compounds in different mixtures is quantified by the activity coefficient γ. Liu at et. (2020) found that the γ of bulk SOA (produced by OH oxidation of a mixture of 1-alkanols) increased from near 1 to 5 as the seeds and the SOA have more differing polarities. A high γ value of 74 was found for a wet ammonium sulfate-SOA system indicating phase separation. Ammonium sulfate is a polar compound. We have to check the polarity of the SOA compounds, if the majority of their mass have polarities close to ammonium sulfate, then we say that γ is around 1 and we are close to ideality. If the polarity is quite different, then the seeds are less hospitable to the condensing vapors and the compounds tend to stay in the gas phase increasing the C* by up to 5 times (and so logC* will increase half an order of magnitude).
Tests on non-ideality carried out in the modeling section clearly show the effect that considering it can have on the partitioning of different compounds. However, these tests highlight the large uncertainties that remain in the calculation of non-ideality, with effects that can be contrary depending on the phases and compounds. The lack of interaction data between the different functional groups adds even more uncertainties. It is partly for this reason that we have chosen not to take non-ideality into account for the experimental C* in Fig. 3.
> Line 210-214: The formation pathways of highly oxygenated organic molecules (HOMs) from toluene photooxidation are also included in the newly developed mechanism. Do they make a significant difference to the speciation of the gaseous and particulate organic species, especially for those with high oxygen numbers?
In our study, the species referred as “HOMs” correspond to stable species that have undergone at least two successive autoxidation steps after the addition of a first OH radical. Other highly oxygenated compounds (which could have higher oxygen number than these “HOMs”) are formed in a more conventional way by successive oxidation with OH, O2 and then NO or HO2. “HOMs” are negligible in our simulations. This observation is not discussed too much in the article, as it could be due to the high concentrations of radicals (NO and HO2) in our experimental conditions, which are unfavorable to successive autoxidation pathways. In line with this, recent use of the mechanism in simulations under atmospheric conditions shows that “HOMs” are not negligible in SOA formation when there are few radicals (which is not, however, the most common case for toluene, generally emitted at the same time as many NOx).
> Line 280: What are the size distributions of toluene SOA formed under typical experimental conditions? Are the formed SOA particles monodispersed as assumed here? Given the low pre-existing seed aerosol concentrations (~9µg /m3), I would expect that there is significant nucleation and new particle formation during toluene photooxidation. As a result, polydispersed aerosols are likely generated, e.g., a larger particle mode comprised of inorganic-organic mixed particles and a smaller particle model primarily with pure SOA particles. If this is the case, how would it affect the phase partitioning behaviors of organic compounds as well as the irreversible uptake of small species such as methylglyoxal.
Typically monodisperse seeds were injected ranging from 120 to 170 nm (mobility diameter). As a function of the experimental conditions more or less condensed organics were formed in the particle phase. The bottom plots show SMPS size distribution before and during photooxidation of toluene. Yes, some nucleation is observed at the beginning of the oxidation, after few minutes as the surface condensation increases nucleation is suppressed at least under our experimental conditions. Below size distribution (SMPS) for seeds and during toluene photooxidation. Partitioning has been calculated during stable periods in absence of nucleation events. (see SMPS data in supplemnt of this comment)
> Line 287: These parameters are expected to significantly impact the gas-phase oxidation chemistry, phase partitioning, and/or particle-phase chemistry (e.g, irreversible uptake of methylglyoxal), but it is found in the present study that they have little impact on the overall speciation of the oxidation products. The authors only present the data obtained at different temperatures in the manuscript. I suggest also providing the data obtained for other parameters in the SI.
We agree that RH, initial seeds, toluene and IPN (i.e. OH radical) concentrations can impact phase partitioning. Seed impact has been shown in the SI section (Figure S2). The figure shows as increased sees surface favors SOA formation. But it also clearly suggests that experimental temperature has a stronger impact on SOA mass loading with respect to seed concentration. Therefore, temperature variation has been considered a key parameter. Other parameters have not been extensively investigated in this work. Under our experimental conditions, however, we did not observe a significant impact on gaseous chemistry and therefore overall speciation (i.e. in both phases). This is reflected in the identification of the same m/z (with different intensities), whatever the experimental conditions.
The core of the modeling work was to reproduce the global speciation of secondary compounds before reproducing and analyzing their partitioning. As particle-phase chemistry was not represented, this work consisted in developing new chemical mechanisms for gaseous oxidation, on which RH and the initial concentrations of toluene, seeds and IPN had no impact. Only experiments at two different temperatures were therefore reproduced.
The most debatable choice is probably not to reproduce experiments at two different RHs, but as the RH range covered during the experimental campaign was restricted (from 24 to 40%) and the artifacts generated by its variations in the tube little analyzed (e.g. RH-dependent wall condensation), this choice was made. Regarding the impact of RH on methylglyoxal partitioning, as mentioned in the article, the parameterization is very simplified. It is based mainly on field data where temperature and RH are dependent, and leads to potential discrepancies by not integrating either temperature or OH concentrations in its calculation. In this respect, the test is mainly exploratory.
Now that an oxidation mechanism has been proposed, an analysis based on a new campaign of experiments focusing on the partitioning of compounds as a function of conditions should be envisaged, in particular to reduce the uncertainties associated with the processes considered.
> Line 356: In addition to losing a H2O molecule, how likely do the toluene oxidation products fragment via C-C scission during PTR-MS measurements? How would it affect the discrimination between parent and fragment ions?
Yes, it has been shown in the literature (Gkatzelis et al., 2018; Leglise et al., 2019) have shown that at the conditions used in this study (E/N = 105 Td) scission of the C-C bond may occur. Part of the low C compounds found in the particle phase may thus be measurement artifacts.
> Line 401-402: Please clarify what type of heterogeneous reactions on the reactor wall can produce these small molecules.
Typical reaction occurring on the wall of OFR or chamber give rise to HONO, HCHO, HCOOH formation. In general adsorbed molecules on the chamber or OFR tube can undergo chemical reaction and produce such small molecules (Doussin et al., 2023).
> Line 405: Although the authors are able to determine the gas-particle partitioning coefficients (Kp) of the toluene oxidation products, they did not present such data in the manuscript. As the Kp can offer additional insights into the phase partitioning behaviors of the oxidation products, I suggest the authors add an analysis of those data as well as their measurement-model comparisons in the manuscript.
The experimentally derived Kp are now in table S3.
Concerning modeled Kp, we chose not to detail or represent their comparison with experimental data for several reasons: the uncertainty of partitioning according to the processes considered, and the impossibility of calculating a Kp for (1) many major compounds of the modeled SOA due to their total transfert in the condensed phase, or (2) compounds whose Kp is measured experimentally but present only in the gas phase in the simulations (i.e. light compounds). A bubbles plot comparison figure would have compared different species that do not necessarily represent the major part of the secondary mass produced.
> Line 413: It seems that some content is missing before the sentence “Other 20 ions were detected…”. Please check.
Nothing is missing but the sentence was not clear, text was modified.
> Line 425: Figure 3 does not include any pie charts described in the figure caption. Also, there are no descriptions/discussions regarding the comparisons between organic compounds measured by CHARON PTR-MS and the SOA mass measured by AMS in the text. Please double check.
The mention to pie charts was removed.
> Line 430-431: The volatility is affected not only by the oxidation state, but also by the molecular size. Therefore, a trend for OSc versus C* may be observed if the data are further discriminated by the carbon number. In addition, Fig. S6 should be Fig. S5.
The numbering error was corrected.
> Line 608: Why does the inclusion of the wall loss parameterization lead to the reduction of the concentration of all m/z values in the particle phase but only the heavier compounds in the gas phase?
Wall losses of compounds depend on their volatility, whatever their m/z. It's therefore logical to see compounds in the particulate phase (i.e. the least volatile) all impacted to a greater or lesser extent. Here, the impact on the gas phase is minimal in all cases. It is concentrated on two types of compounds: (1) those of low volatility, which are present in both phases (the decrease in particulate concentration goes hand in hand with the decrease in gas concentration) and (2) compounds formed after several oxidation steps, whose parent species have seen their concentrations decrease due to wall losses (these compounds are potentially the heaviest due to the successive addition of functions). This does not mean that there is no effect on other gaseous compounds, but it is negligible.
> In addition, the following language errors should be corrected.
Line 91: produced -> produce
Line 249: wall losses -> wall loss effects
Line 273: Delete “controlled”.
Line 287: as -> as well as
Line 298: as -> such as
Line 346: by -> followed by
Line 407: cercles -> circles
Line 442: that -> than
Line 443: Delete either of “only”.The text was modified.
References:
Doussin, J.-F., Fuchs, H., Kiendler-Scharr, A., Seakins, P., and Wenger, J. (Eds.): A Practical Guide to Atmospheric Simulation Chambers, Springer International Publishing, Cham, https://doi.org/10.1007/978-3-031-22277-1, 2023.
Gkatzelis, G. I., Tillmann, R., Hohaus, T., Müller, M., Eichler, P., Xu, K.-M., Schlag, P., Schmitt, S. H., Wegener, R., Kaminski, M., Holzinger, R., Wisthaler, A., and Kiendler-Scharr, A.: Comparison of three aerosol chemical characterization techniques utilizing PTR-ToF-MS: a study on freshly formed and aged biogenic SOA, Atmos. Meas. Tech, 11, 1481–1500, https://doi.org/10.5194/amt-11-1481-2018, 2018.
Leglise, J., Müller, M., Piel, F., Otto, T., and Wisthaler, A.: Bulk Organic Aerosol Analysis by Proton-Transfer-Reaction Mass Spectrometry: An Improved Methodology for the Determination of Total Organic Mass, O:C and H:C Elemental Ratios, and the Average Molecular Formula, Anal. Chem., 91, 12619–12624, https://doi.org/10.1021/acs.analchem.9b02949, 2019.
Liu, X., Day, D. A., Krechmer, J. E., Ziemann, P. J., and Jimenez, J. L.: Determining Activity Coefficients of SOA from Isothermal Evaporation in a Laboratory Chamber, Cite This Environ. Sci. Technol. Lett, 8, 212–217, https://doi.org/10.1021/acs.estlett.0c00888, 2021.
Piel, F., Müller, M., Winkler, K., Skytte Af Sätra, J., and Wisthaler, A.: Introducing the extended volatility range proton-transfer-reaction mass spectrometer (EVR PTR-MS), Atmos. Meas. Tech, 14, 1355–1363, https://doi.org/10.5194/amt-14-1355-2021, 2021.
Shiraiwa, M. and Seinfeld, J. H.: Equilibration timescale of atmospheric secondary organic aerosol partitioning, Geophys. Res. Lett., 39, 1–6, https://doi.org/10.1029/2012GL054008, 2012.
-
AC2: 'Reply on RC2', Victor Lannuque, 20 Sep 2023
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2023-1290', Anonymous Referee #1, 03 Aug 2023
This manuscript reports measurement and modeling of gas-phase and particle- phase composition and mass concentration of SOA formed in toluene+OH reaction system. Gas- and particle-phase components are measured by CHARON- PTR-MS. Gas- and particle-phase composition was modeled using SSH-aerosol model and updated chemistry of cresol and ring-scission chemistry and varying levels of complexity on gas-particle partitioning. It was found that SOA yields at different temperatures were consistent previous studies. According to measured volatility, most compounds belonged to SVOC. Volatility did not show clear relationship with oxidation state. PTR-MS quantified 70-80% of total organic mass measured by AMS. C7 compounds contributed the largest fraction in total products, both in the gas-phase and particle-phase. C1-6 compounds also contributed significantly. Models with varying levels of complexity on gas-particle partitioning did not well reproduce the measured gas-phase and particle-phase composition. Light compounds (m/z <100) were underestimated and the heavy compounds (m/z >100) were overestimated. However, the model including partitioning into aqueous phase, interactions between organics and between organics and inorganics achieved good agreement with measured SOA mass concentration. Irreversible uptake of methylglyoxal in the model further improved the simulation of SOA mass concentration.
Determining the reaction products of SOA and related gas-phase products on molecular level and elucidating key processes affecting its concentrations are key to predicting its ambient concentration and understanding its impact. This study addresses both aspects. Particularly, besides modeling SOA mass concentration, this study attempt to model the chemical composition of SOA via comparing with measurement, which is novel and not much reported in the literature. The manuscript is generally well-written. I have a few comments for the authors to consider before publication.
Specific comments
- Before digging into the reason for the difference between modeled and measured composition, one need be sure that the difference is minimally influenced by the measurement artefacts. Is this possible that small compounds, particularly those less oxygenated C1-C4 compounds in particle-phase, are fragment of parent products or decomposition during heating in CHARON? It is surprising to see these compounds in the particle-phase due to their high volatility. It would helpful to elaborate the method used to exclude the artefacts and/or discuss their influence on the measured chemical composition.
- In Fig. 2, no oligomers (C>7) compounds were shown. Are there any oligomers (C>7) detected? Can CHARON PTR MS see them?
- I suggest adding a figure of reaction schemes of toluene+OH reaction highlight the new reaction mechanisms added to the model in SI so that readers can better follow.
- L167, which data does this statement based on?
- L236, what does “long range interactions” mean? It would be helpful to briefly explain it.
Technical comments
- L35, it is not clear that what the “enhanced” is compared with.
- L287, “as” should be “and”?
- L407, “cercles” should be “circles”. Also it would helpful to clarify that it is the diameter or area of circles that denotes the mass in Fig. 3.
In the caption of Fig. 3, there are no “pie charts”. Please check.
- L451, the “of” after “24 %” should be deleted.
- L505, “47%”, is this correct or should it be 28%?
- L545, a comma is missing after “(a).
- L482, a space is missing after “distribution”.
Citation: https://doi.org/10.5194/egusphere-2023-1290-RC1 -
AC1: 'Reply on RC1', Victor Lannuque, 20 Sep 2023
We thank the reviewers for their comments and suggestions on the manuscript. We outline below responses to the points raised by each referee and summarize the changes made to the revised manuscript. We also provide a revised version of the manuscript with highlighted modifications.
Reponses to RC1
> 1. Before digging into the reason for the difference between modeled and measured composition, one need be sure that the difference is minimally influenced by the measurement artefacts. Is this possible that small compounds, particularly those less oxygenated C1-C4 compounds in particle-phase, are fragment of parent products or decomposition during heating in CHARON? It is surprising to see these compounds in the particle-phase due to their high volatility. It would helpful to elaborate the method used to exclude the artefacts and/or discuss their influence on the measured chemical composition.
Thermal decomposition is typically not observed in CHARON PTR-ToF-MS instruments (except for hydroperoxides and labile sugars). It has, however, been shown in previous work (e.g., Gkatzelis et al., 2018; Leglise et al. 2019) that at the conditions used in this study (E/N = 105 Td) scission of the C-C bond may indeed occur. Therefore a fraction of the low C compounds found in the particle phase may thus be measurement artifacts.
> 2. In Fig. 2, no oligomers (C>7) compounds were shown. Are there any oligomers (C>7) detected? Can CHARON PTR MS see them?
Oligomers can be observed with CHARON PTR-ToF-MS instruments. It is, however, likely that at the operating conditions used in this study (E/N = 105 Td) oligomers fragment upon protonation.
> 3. I suggest adding a figure of reaction schemes of toluene+OH reaction highlight the new reaction mechanisms added to the model in SI so that readers can better follow.
The suggested figure was added to SI (Fig. S6) and is now mentioned in the text.
> 4. L167, which data does this statement based on?
According to Jenkin et al. (2019) SAR, the RO2 + RO2 reactions mainly lead to the formation of alkoxy radicals as RO2 + NO reactions. Under atmospheric conditions, the RO2 + CH3O2 (the most abundant RO2 in atmosphere) reaction represents about 10 % of the RO2 reactivity (Lannuque et al., 2018). If we have here decided to not consider all RO2 + RO2 reactions it is to limit the already relatively large size of our mechanisms. We are aware that this means ignoring the formation of certain minor secondary compounds, notably C>7. This choice was made considering (1) the high NOx levels limiting the importance of RO2 + RO2 reactions and (2) the absence of detection of such compounds in our experiments.
> 5. L236, what does “long range interactions” mean? It would be helpful to briefly explain it.
Middle- and long-range interactions correspond to interactions between the organic compounds and charged molecules (especially inorganic ions, the only considered here) in water phase. This term is used by opposition of the interactions between the different uncharged organic molecules which occur at a shorter range.
> 1. L35, it is not clear that what the “enhanced” is compared with.
It is compared to a reference mechanism using MCM and GECKO-A. It is now specified in the text.
> 2. L287, “as” should be “and”?
The text was modified.
> 3. L407, “cercles” should be “circles”. Also it would helpful to clarify that it is the diameter or area of circles that denotes the mass in Fig. 3.
The area of the circles has been used. The text was modified and the legend of Fig. 3 was detailed.
> In the caption of Fig. 3, there are no “pie charts”. Please check.
The mention to pie charts was removed.
> 1. L451, the “of” after “24 %” should be deleted.
The text was modified.
> 2. L505, “47%”, is this correct or should it be 28%?
You are right, the mistake was corrected and 47% was replaced by 30% (28 + 2 %).
> 3. L545, a comma is missing after “(a).
> 4. L482, a space is missing after “distribution”.The text was corrected.
References :
Gkatzelis, G. I., Tillmann, R., Hohaus, T., Müller, M., Eichler, P., Xu, K.-M., Schlag, P., Schmitt, S. H., Wegener, R., Kaminski, M., Holzinger, R., Wisthaler, A., and Kiendler-Scharr, A.: Comparison of three aerosol chemical characterization techniques utilizing PTR-ToF-MS: a study on freshly formed and aged biogenic SOA, Atmos. Meas. Tech, 11, 1481–1500, https://doi.org/10.5194/amt-11-1481-2018, 2018.
Jenkin, M. E., Valorso, R., Aumont, B., and Rickard, A. R.: Estimation of rate coefficients and branching ratios for reactions of organic peroxy radicals for use in automated mechanism construction, Atmos. Chem. Phys., 19, 7691–7717, https://doi.org/10.5194/ACP-19-7691-2019, 2019.
Lannuque, V., Camredon, M., Couvidat, F., Hodzic, A., Valorso, R., Madronich, S., Bessagnet, B., and Aumont, B.: Exploration of the influence of environmental conditions on secondary organic aerosol formation and organic species properties using explicit simulations: development of the VBS-GECKO parameterization, Atmos. Chem. Phys., 18, 13411–13428, https://doi.org/10.5194/acp-18-13411-2018, 2018.
Leglise, J., Müller, M., Piel, F., Otto, T., and Wisthaler, A.: Bulk Organic Aerosol Analysis by Proton-Transfer-Reaction Mass Spectrometry: An Improved Methodology for the Determination of Total Organic Mass, O:C and H:C Elemental Ratios, and the Average Molecular Formula, Anal. Chem., 91, 12619–12624, https://doi.org/10.1021/acs.analchem.9b02949, 2019.
Citation: https://doi.org/10.5194/egusphere-2023-1290-AC1
-
RC2: 'Comment on egusphere-2023-1290', Anonymous Referee #2, 09 Aug 2023
This manuscript describes an experimental and modeling study of the formation and gas-particle partitioning of organic compounds from the photooxidation of toluene. The toluene photooxidation experiments were conducted in a Pyrex OFR, with the gaseous and particulate organic products measured simultaneously by two separate PTR-ToF-MS without and with a thermal desorption inlet (CHARON), respectively. It is shown that the particulate organic species quantified by CHARON PTR-MS can explain approximately 70% of the secondary organic aerosol (SOA) mass measured by an AMS. A semi-explicit gas-phase oxidation mechanism of toluene was developed to reproduce the formation and distribution of organic products in both gas and particle phases. It is found that with the modifications of cresol chemistry, aromatic ring opening and furan formation chemistry, as well as relevant reaction kinetics based on a structure-activity relationship, the model can better represent the speciation of the oxidation products in the gas phase, but does not improve the prediction of the product distribution in the particle phase. Further simulations with an aerosol box model (SSH), incorporated with the newly developed gas-phase oxidation mechanism, show that the formation of SOA is highly sensitive to the parameterization of the phase partitioning processes, vapor wall losses in the OFR, and the irreversible uptake of small molecules such as methylglyoxal. Overall, this study is interesting and provides valuable insights into the physicochemical processes associated with toluene SOA formation as well as the further development of SOA models. The manuscript can be considered for publication in ACP after the following comments are addressed.
Line 98: Additional experimental details such as the rate and residence time of the gas flow in the OFR and the initial concentrations of toluene should be provided in the experimental section.
Line 132: Have any experiments been done to verify that all the organics in SOA are thermally desorbed at a temperature of 413 K in the CHARON inlet? Was the fragmentation of organic compounds during the thermal desorption of SOA significant?
Line 145: Are the C* values calculated from the measured partitioning coefficients using Equation 2? What are the influences of the potential non-equilibrium phase partitioning (for example under low RH conditions) and non-ideality of the condensed phase on such calculations?
Line 210-214: The formation pathways of highly oxygenated organic molecules (HOMs) from toluene photooxidation are also included in the newly developed mechanism. Do they make a significant difference to the speciation of the gaseous and particulate organic species, especially for those with high oxygen numbers?
Line 280: What are the size distributions of toluene SOA formed under typical experimental conditions? Are the formed SOA particles monodispersed as assumed here? Given the low pre-existing seed aerosol concentrations (~9 μg/m3), I would expect that there is significant nucleation and new particle formation during toluene photooxidation. As a result, polydispersed aerosols are likely generated, e.g., a larger particle mode comprised of inorganic-organic mixed particles and a smaller particle model primarily with pure SOA particles. If this is the case, how would it affect the phase partitioning behaviors of organic compounds as well as the irreversible uptake of small species such as methylglyoxal.
Line 287: These parameters are expected to significantly impact the gas-phase oxidation chemistry, phase partitioning, and/or particle-phase chemistry (e.g, irreversible uptake of methylglyoxal), but it is found in the present study that they have little impact on the overall speciation of the oxidation products. The authors only present the data obtained at different temperatures in the manuscript. I suggest also providing the data obtained for other parameters in the SI.
Line 356: In addition to losing a H2O molecule, how likely do the toluene oxidation products fragment via C-C scission during PTR-MS measurements? How would it affect the discrimination between parent and fragment ions?
Line 401-402: Please clarify what type of heterogeneous reactions on the reactor wall can produce these small molecules.
Line 405: Although the authors are able to determine the gas-particle partitioning coefficients (Kp) of the toluene oxidation products, they did not present such data in the manuscript. As the Kp can offer additional insights into the phase partitioning behaviors of the oxidation products, I suggest the authors add an analysis of those data as well as their measurement-model comparisons in the manuscript.
Line 413: It seems that some content is missing before the sentence “Other 20 ions were detected…”. Please check.
Line 425: Figure 3 does not include any pie charts described in the figure caption. Also, there are no descriptions/discussions regarding the comparisons between organic compounds measured by CHARON PTR-MS and the SOA mass measured by AMS in the text. Please double check.
Line 430-431: The volatility is affected not only by the oxidation state, but also by the molecular size. Therefore, a trend for OSc versus C* may be observed if the data are further discriminated by the carbon number. In addition, Fig. S6 should be Fig. S5.
Line 608: Why does the inclusion of the wall loss parameterization lead to the reduction of the concentration of all m/z values in the particle phase but only the heavier compounds in the gas phase?
In addition, the following language errors should be corrected.
Line 91: produced -> produce
Line 249: wall losses -> wall loss effects
Line 273: Delete “controlled”.
Line 287: as -> as well as
Line 298: as -> such as
Line 346: by -> followed by
Line 407: cercles -> circles
Line 442: that -> than
Line 443: Delete either of “only”.
Citation: https://doi.org/10.5194/egusphere-2023-1290-RC2 -
AC2: 'Reply on RC2', Victor Lannuque, 20 Sep 2023
We thank the reviewers for their comments and suggestions on the manuscript. We outline below responses to the points raised by each referee and summarize the changes made to the revised manuscript. We also provide a revised version of the manuscript with highlighted modifications.
Reponses to RC2
> Line 98: Additional experimental details such as the rate and residence time of the gas flow in the OFR and the initial concentrations of toluene should be provided in the experimental section.
Experiments were carried out using a total flow varying from 1.3 to 1.8 L/min corresponding to a residence time of approximately 10-13 minutes. These details have been introduced in the experimental section.
> Line 132: Have any experiments been done to verify that all the organics in SOA are thermally desorbed at a temperature of 413 K in the CHARON inlet? Was the fragmentation of organic compounds during the thermal desorption of SOA significant?
It has been shown in the literature that at the temperatures used in this study SVOCs, LVOCs and even ELVOCS are vaporized (e.g. Piel et al., 2021). It is important to note that vaporization is affected on the walls of the thermal desorption unit at reduced at reduce pressure (few mbar). Piel et al. (2021) also mention that certain types of thermal decomposition such as the decarboxylation of organic acids does not occur at the temperatures used in this study. The same authors found evidence for the thermal decomposition of hydroperoxides in the CHARON inlet.
> Line 145: Are the C* values calculated from the measured partitioning coefficients using Equation 2? What are the influences of the potential non-equilibrium phase partitioning (for example under low RH conditions) and non-ideality of the condensed phase on such calculations?
The C* were calculated using equations 1 and 2. According to Shiraiwa and Seinfeld (2012): the equilibrium time is achieved from seconds to minutes for relatively high volatility organic compounds into liquid particles. However, equilibrium time can increase to hours or days for organic aerosol when the particles are large, semi-solid particles, of low volatility, and low mass loadings. Instantaneous equilibrium partitioning may lead to substantial overestimation of particle mass concentration and underestimation of gas-phase concentration.
In our study most of the identified SOA compounds identified are semi-volatile (Table 1). Moreover, monodispersed seeds+SOA had a size was about 150-216 nm so our particles are quite small to establish an equilibrium with the gas phase. In addition, the total mass concentration was on average 13.9±3.4 µg m-3 for the experiments at 295 K and 17.4±4.9 µg m-3 for the experiments at 280 K, the mass loading is high enough for the establishment of the equilibrium. Indeed according to Figure 4 of Shiraiwa and Seinfeld (2012), for semi-solid SVOCs with a diameter of 150 nm and organic mass concentration 14-17 µg m-3 the equilibrium timescale of SOA partitioning is about 10 min. Given the fact that the residence time in the OFR was about 11-13 minutes we conclude that particle and gas phases were practically in equilibrium. Furthermore, in preliminary modeling tests, gas-particle partitioning was represented using both methods: one considering thermodynamic equilibrium at each time step, and one following a dynamic phase transfer approach. The two methods showed no significant differences in partitioning under our simulation conditions.
The non-ideallity, for individual organic compounds in different mixtures is quantified by the activity coefficient γ. Liu at et. (2020) found that the γ of bulk SOA (produced by OH oxidation of a mixture of 1-alkanols) increased from near 1 to 5 as the seeds and the SOA have more differing polarities. A high γ value of 74 was found for a wet ammonium sulfate-SOA system indicating phase separation. Ammonium sulfate is a polar compound. We have to check the polarity of the SOA compounds, if the majority of their mass have polarities close to ammonium sulfate, then we say that γ is around 1 and we are close to ideality. If the polarity is quite different, then the seeds are less hospitable to the condensing vapors and the compounds tend to stay in the gas phase increasing the C* by up to 5 times (and so logC* will increase half an order of magnitude).
Tests on non-ideality carried out in the modeling section clearly show the effect that considering it can have on the partitioning of different compounds. However, these tests highlight the large uncertainties that remain in the calculation of non-ideality, with effects that can be contrary depending on the phases and compounds. The lack of interaction data between the different functional groups adds even more uncertainties. It is partly for this reason that we have chosen not to take non-ideality into account for the experimental C* in Fig. 3.
> Line 210-214: The formation pathways of highly oxygenated organic molecules (HOMs) from toluene photooxidation are also included in the newly developed mechanism. Do they make a significant difference to the speciation of the gaseous and particulate organic species, especially for those with high oxygen numbers?
In our study, the species referred as “HOMs” correspond to stable species that have undergone at least two successive autoxidation steps after the addition of a first OH radical. Other highly oxygenated compounds (which could have higher oxygen number than these “HOMs”) are formed in a more conventional way by successive oxidation with OH, O2 and then NO or HO2. “HOMs” are negligible in our simulations. This observation is not discussed too much in the article, as it could be due to the high concentrations of radicals (NO and HO2) in our experimental conditions, which are unfavorable to successive autoxidation pathways. In line with this, recent use of the mechanism in simulations under atmospheric conditions shows that “HOMs” are not negligible in SOA formation when there are few radicals (which is not, however, the most common case for toluene, generally emitted at the same time as many NOx).
> Line 280: What are the size distributions of toluene SOA formed under typical experimental conditions? Are the formed SOA particles monodispersed as assumed here? Given the low pre-existing seed aerosol concentrations (~9µg /m3), I would expect that there is significant nucleation and new particle formation during toluene photooxidation. As a result, polydispersed aerosols are likely generated, e.g., a larger particle mode comprised of inorganic-organic mixed particles and a smaller particle model primarily with pure SOA particles. If this is the case, how would it affect the phase partitioning behaviors of organic compounds as well as the irreversible uptake of small species such as methylglyoxal.
Typically monodisperse seeds were injected ranging from 120 to 170 nm (mobility diameter). As a function of the experimental conditions more or less condensed organics were formed in the particle phase. The bottom plots show SMPS size distribution before and during photooxidation of toluene. Yes, some nucleation is observed at the beginning of the oxidation, after few minutes as the surface condensation increases nucleation is suppressed at least under our experimental conditions. Below size distribution (SMPS) for seeds and during toluene photooxidation. Partitioning has been calculated during stable periods in absence of nucleation events. (see SMPS data in supplemnt of this comment)
> Line 287: These parameters are expected to significantly impact the gas-phase oxidation chemistry, phase partitioning, and/or particle-phase chemistry (e.g, irreversible uptake of methylglyoxal), but it is found in the present study that they have little impact on the overall speciation of the oxidation products. The authors only present the data obtained at different temperatures in the manuscript. I suggest also providing the data obtained for other parameters in the SI.
We agree that RH, initial seeds, toluene and IPN (i.e. OH radical) concentrations can impact phase partitioning. Seed impact has been shown in the SI section (Figure S2). The figure shows as increased sees surface favors SOA formation. But it also clearly suggests that experimental temperature has a stronger impact on SOA mass loading with respect to seed concentration. Therefore, temperature variation has been considered a key parameter. Other parameters have not been extensively investigated in this work. Under our experimental conditions, however, we did not observe a significant impact on gaseous chemistry and therefore overall speciation (i.e. in both phases). This is reflected in the identification of the same m/z (with different intensities), whatever the experimental conditions.
The core of the modeling work was to reproduce the global speciation of secondary compounds before reproducing and analyzing their partitioning. As particle-phase chemistry was not represented, this work consisted in developing new chemical mechanisms for gaseous oxidation, on which RH and the initial concentrations of toluene, seeds and IPN had no impact. Only experiments at two different temperatures were therefore reproduced.
The most debatable choice is probably not to reproduce experiments at two different RHs, but as the RH range covered during the experimental campaign was restricted (from 24 to 40%) and the artifacts generated by its variations in the tube little analyzed (e.g. RH-dependent wall condensation), this choice was made. Regarding the impact of RH on methylglyoxal partitioning, as mentioned in the article, the parameterization is very simplified. It is based mainly on field data where temperature and RH are dependent, and leads to potential discrepancies by not integrating either temperature or OH concentrations in its calculation. In this respect, the test is mainly exploratory.
Now that an oxidation mechanism has been proposed, an analysis based on a new campaign of experiments focusing on the partitioning of compounds as a function of conditions should be envisaged, in particular to reduce the uncertainties associated with the processes considered.
> Line 356: In addition to losing a H2O molecule, how likely do the toluene oxidation products fragment via C-C scission during PTR-MS measurements? How would it affect the discrimination between parent and fragment ions?
Yes, it has been shown in the literature (Gkatzelis et al., 2018; Leglise et al., 2019) have shown that at the conditions used in this study (E/N = 105 Td) scission of the C-C bond may occur. Part of the low C compounds found in the particle phase may thus be measurement artifacts.
> Line 401-402: Please clarify what type of heterogeneous reactions on the reactor wall can produce these small molecules.
Typical reaction occurring on the wall of OFR or chamber give rise to HONO, HCHO, HCOOH formation. In general adsorbed molecules on the chamber or OFR tube can undergo chemical reaction and produce such small molecules (Doussin et al., 2023).
> Line 405: Although the authors are able to determine the gas-particle partitioning coefficients (Kp) of the toluene oxidation products, they did not present such data in the manuscript. As the Kp can offer additional insights into the phase partitioning behaviors of the oxidation products, I suggest the authors add an analysis of those data as well as their measurement-model comparisons in the manuscript.
The experimentally derived Kp are now in table S3.
Concerning modeled Kp, we chose not to detail or represent their comparison with experimental data for several reasons: the uncertainty of partitioning according to the processes considered, and the impossibility of calculating a Kp for (1) many major compounds of the modeled SOA due to their total transfert in the condensed phase, or (2) compounds whose Kp is measured experimentally but present only in the gas phase in the simulations (i.e. light compounds). A bubbles plot comparison figure would have compared different species that do not necessarily represent the major part of the secondary mass produced.
> Line 413: It seems that some content is missing before the sentence “Other 20 ions were detected…”. Please check.
Nothing is missing but the sentence was not clear, text was modified.
> Line 425: Figure 3 does not include any pie charts described in the figure caption. Also, there are no descriptions/discussions regarding the comparisons between organic compounds measured by CHARON PTR-MS and the SOA mass measured by AMS in the text. Please double check.
The mention to pie charts was removed.
> Line 430-431: The volatility is affected not only by the oxidation state, but also by the molecular size. Therefore, a trend for OSc versus C* may be observed if the data are further discriminated by the carbon number. In addition, Fig. S6 should be Fig. S5.
The numbering error was corrected.
> Line 608: Why does the inclusion of the wall loss parameterization lead to the reduction of the concentration of all m/z values in the particle phase but only the heavier compounds in the gas phase?
Wall losses of compounds depend on their volatility, whatever their m/z. It's therefore logical to see compounds in the particulate phase (i.e. the least volatile) all impacted to a greater or lesser extent. Here, the impact on the gas phase is minimal in all cases. It is concentrated on two types of compounds: (1) those of low volatility, which are present in both phases (the decrease in particulate concentration goes hand in hand with the decrease in gas concentration) and (2) compounds formed after several oxidation steps, whose parent species have seen their concentrations decrease due to wall losses (these compounds are potentially the heaviest due to the successive addition of functions). This does not mean that there is no effect on other gaseous compounds, but it is negligible.
> In addition, the following language errors should be corrected.
Line 91: produced -> produce
Line 249: wall losses -> wall loss effects
Line 273: Delete “controlled”.
Line 287: as -> as well as
Line 298: as -> such as
Line 346: by -> followed by
Line 407: cercles -> circles
Line 442: that -> than
Line 443: Delete either of “only”.The text was modified.
References:
Doussin, J.-F., Fuchs, H., Kiendler-Scharr, A., Seakins, P., and Wenger, J. (Eds.): A Practical Guide to Atmospheric Simulation Chambers, Springer International Publishing, Cham, https://doi.org/10.1007/978-3-031-22277-1, 2023.
Gkatzelis, G. I., Tillmann, R., Hohaus, T., Müller, M., Eichler, P., Xu, K.-M., Schlag, P., Schmitt, S. H., Wegener, R., Kaminski, M., Holzinger, R., Wisthaler, A., and Kiendler-Scharr, A.: Comparison of three aerosol chemical characterization techniques utilizing PTR-ToF-MS: a study on freshly formed and aged biogenic SOA, Atmos. Meas. Tech, 11, 1481–1500, https://doi.org/10.5194/amt-11-1481-2018, 2018.
Leglise, J., Müller, M., Piel, F., Otto, T., and Wisthaler, A.: Bulk Organic Aerosol Analysis by Proton-Transfer-Reaction Mass Spectrometry: An Improved Methodology for the Determination of Total Organic Mass, O:C and H:C Elemental Ratios, and the Average Molecular Formula, Anal. Chem., 91, 12619–12624, https://doi.org/10.1021/acs.analchem.9b02949, 2019.
Liu, X., Day, D. A., Krechmer, J. E., Ziemann, P. J., and Jimenez, J. L.: Determining Activity Coefficients of SOA from Isothermal Evaporation in a Laboratory Chamber, Cite This Environ. Sci. Technol. Lett, 8, 212–217, https://doi.org/10.1021/acs.estlett.0c00888, 2021.
Piel, F., Müller, M., Winkler, K., Skytte Af Sätra, J., and Wisthaler, A.: Introducing the extended volatility range proton-transfer-reaction mass spectrometer (EVR PTR-MS), Atmos. Meas. Tech, 14, 1355–1363, https://doi.org/10.5194/amt-14-1355-2021, 2021.
Shiraiwa, M. and Seinfeld, J. H.: Equilibration timescale of atmospheric secondary organic aerosol partitioning, Geophys. Res. Lett., 39, 1–6, https://doi.org/10.1029/2012GL054008, 2012.
-
AC2: 'Reply on RC2', Victor Lannuque, 20 Sep 2023
Peer review completion
Journal article(s) based on this preprint
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 395 | 145 | 25 | 565 | 66 | 13 | 18 |
- HTML: 395
- PDF: 145
- XML: 25
- Total: 565
- Supplement: 66
- BibTeX: 13
- EndNote: 18
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Evangelia Kostenidou
Florian Couvidat
Alvaro Martinez-Valiente
Philipp Eichler
Armin Wisthaler
Markus Müller
Brice Temime-Roussel
Richard Valorso
Karine Sartelet
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1800 KB) - Metadata XML
-
Supplement
(3784 KB) - BibTeX
- EndNote
- Final revised paper