the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 18 Oct 2023
Production of oxygenated volatile organic compounds from the ozonolysis of coastal seawater
Abstract. Dry deposition of ozone (O3) to the ocean surface and the ozonolysis of organics in the sea surface microlayer (SSML) is a potential source of volatile organic compounds (VOC) to the marine atmosphere. We use a gas chromatography system coupled to a Vocus proton transfer reaction time-of-flight mass spectrometer to determine the chemical composition and product yield of select VOC formed from ozonolysis of coastal seawater collected from Scripps Pier in La Jolla, California. Laboratory-derived results are interpreted in the context of direct VOC vertical flux measurements made at Scripps Pier. The dominant products of laboratory ozonolysis experiments and the largest non-sulfur emission fluxes measured in the field correspond to Vocus CxHy+ and CxHyOz+ ions. GC analysis suggests that C5–C11 oxygenated VOC, primarily aldehydes, are the largest contributors to these ion signals. In the laboratory, using a flow reactor experiment, we determine a VOC yield of 0.43–0.62. In the field at Scripps Pier, we determine a maximum VOC yield of 0.04–0.06. Scaling the field and lab VOC yields for an average O3 deposition flux and an average VOC structure results in an emission source of 12.6 to 136 Tg C yr-1, competitive with the DMS source of 21.1 Tg C yr-1. This study reveals that O3 reactivity to dissolved organic carbon can be a significant carbon source to the marine atmosphere and warrants further investigation into the speciated VOC composition from different seawater samples, and the reactivities and secondary organic aerosol yields of these molecules in marine-relevant, low NOx conditions.
-
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
(1050 KB)
-
Supplement
(922 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1050 KB) - Metadata XML
-
Supplement
(922 KB) - BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2023-2210', Anonymous Referee #2, 31 Oct 2023
This manuscript presents a unique combination of field observations of reactive organic carbon (ROC) fluxes and laboratory experiments probing reactive organic carbon emissions from a quiescent seawater microlayer reacting with gas phase ozone. The authors provide VOC yield estimates from both in situ ROC fluxes and laboratory ozone oxidation, and demonstrate that real world VOC yields are likely much smaller than laboratory estimations suggests. This is a very important point to make given that laboratory VOC yields from similar experiments have already been scaled to global emissions in some modelling studies. However, even scaling the much lower field VOC yield by an average ozone deposition flux points to an important source of ROC over the world's oceans. This work also makes two additional very useful points: (1) that the authors see no evidence for direct photochemistry driving VOC fluxes from the sea surface, and (2) that C5H9+ signals from PTR-type instruments should be interpreted with extreme caution in marine environments (i.e., building on Coggon et al., 2023 for urban centers).
General Comments1. (L19, L444) 21.1 Tg-C/yr from marine DMS emissions. It is difficult to understand where this number comes from. It is not cited, and doesn't appear to line up with current climatologies. For example, the third revision of the DMS climatology (Hulswar et al., 2022 https://essd.copernicus.org/articles/14/2963/2022/) estimates a global DMS emission of ~27.1 Tg-S/yr. Scaling to carbon in DMS gives ~20.3 Tg-C/yr. While this is relatively close to the number stated in the abstract and conclusions, this highlights the ambiguity in the origin for this comparison. The authors should cite recent climatologies for this comparison and make clear that this number is both not exactly known and arises from averaging over significant regional variability. Otherwise, this may not be a useful benchmark against which to judge the importance of VOC emissions driven by O3 deposition to the sea surface.
2. Use of the term heterogeneous (L29, L369 and elsewhere). Given the reactions studied occur at liquid interfaces, "multiphase" may be more appropriate. e.g., see https://acp.copernicus.org/articles/23/9765/2023/ for current suggestions on revisions to terminology. It is more useful to consider a multiphase reaction that may have substantial interfacial and bulk components, then it is to consider a perfectly "heterogeneous" surface-only reaction versus and entirely bulk phase reaction.
3. The estimation of multiphase O3 reactivity and k_DOC (L299 - L324) requires further, and more nuanced, discussion.
(a) The authors apportion O3 reactivity into an I- reactivity of 266 s^-1 and a DOC reactivity of 177 s^-1. Can the authors propagate uncertainties in the measurements that go into this estimation of O3 reactivity? Further, this O3 reactivity assumes a bulk rate coefficient and a bulk iodide concentration, and would be more accurately referred to as a "bulk reactivity," because (as the authors state later in the paper) the O3 reactions likely have a significant interfacial component.
(b) In addition to assuming the surface I- is not depleted by reaction, do the authors also assume that the surface I- concentration is equal to the bulk I- concentration? I- ions are known to have affinity for the air-water interface, and a Langmuir adsorption isotherm, with literature constraints on the bulk-surface partition coefficient, could be used to better constrain the magnitude of potential surface concentrations in the present experiments.
(c) Similarly, while the estimated k_DOC uses a bulk [DOC], it also uses an observed gas phase O3 loss and so inherently incorporates both interfacial and bulk O3 reactions of unknown relative importance. It cannot be simply stated (L311) that this reaction "likely happens at the interface," rather for a bimolecular rate coefficient on the order of 5e6 M^-1 s^-1, it is possible that interfacial reactions will make a substantial contribution to O3 loss, though that will depend on the interplay between O3 and DOC partitioning to the interface (e.g., https://pubs.acs.org/doi/full/10.1021/acs.jpca.2c03059)
(d) The authors go on (L317) to compare their mixed bulk/interfacial k_DOC to a set of strictly bulk rate coefficients for species used to approximate DOC. These bulk, bimolecular rate coefficients span 3 orders of magnitude, and may not represent species that are relevant to marine DOC. The comparison to an "authentic marine sample" (L319) is presumably that from Shaw & Carpenter 2013 (2.6e7 M^-1 s^-1), and is therefore a more appropriate comparison as this literature value arises from similar O3 deposition experiments and is thus a combination of bulk and interfacial O3 reactivity.4. Figure 4: Pie charts an interesting choice here as you can't show a range or variability in these contributions. A bar chart with error bars that correspond to some measure of variability in contributions to your observed yields would be much more informative.
5. The authors use field and lab average yields to bracket the range of possible VOC yields in the conclusions and abstract. It may be more appropriate to include the range of yields for both lab and field data, show in Table 1, into these discussions.
Specific CommentsL36: "This collection of VOC" -- this wording is somewhat confusing; which VOCs specifically?
L195: "Ions without an expected molecule" -- does this mean ions without a known, or single, contributing molecule?
L302: k_iodide = 2.4e9 M^-1 s^-1, presumably this value is from Magi et al., 1997? More recent determinations exist e.g., https://pubs.acs.org/doi/10.1021/ic000919j (for pH 6.7, 1.2 (+/-0.1)e9 M^-1 s^-1). Further, given the pH dependence of O3 + I-, is the Magi 1997 rate coefficient applicable to seawater pH? This should be discussed further.
L321: Alkanes reacting with ozone?
L332: "..range in yield reflects standard deviations in O3." -- +/- 1 sigma, or more?
L342-344: What fraction of organic carbon goes to the gas phase in the experiment? The authors appear to have the data to estimate this from the DOC measurement in sea water together with integrated delta-VOC over time.
L355: "..based on standard deviation of O3 mixing ratios measured in 2019" -- is the data normally distributed? Is this an appropriate measure of your uncertainty?
L356: why not include these additional uncertainties, or an estimate, to provide a more true measure of the uncertainty in your field VOC yield?
L376-378: "any O3 source variability was not directly measured simultaneously during experiments, meaning that quicker O3 fluctuations than what occur in the ambient could have heightened laboratory yields." -- do the authors expect this is a significant source of higher VOC yield when delta-VOC mirrors delta-O3? If so, this should be discussed further.
L380: "[I-] a factor of three lower" -- clarify if this is measured or inferred.
L385: "caution our ability" -- should be "complicate our ability"?
L405: (Figure 5 caption) Regression should be relationship, or similar? You are not quantitatively assessing the relationship between two variables (i.e., you are not applying a regression model)
L440: "a couple of hundred" -- avoid vague language, and give the range from the literature you cite and their uncertainties.
Citation: https://doi.org/10.5194/egusphere-2023-2210-RC1 -
RC2: 'Comment on egusphere-2023-2210', Anonymous Referee #1, 06 Nov 2023
General Comments
This paper by Kilgour et al. explores the emission of volatile organic compounds (VOCs) from the ozonolysis of seawater with a combination of field and laboratory measurements. From this work, the authors determine the approximate yield of VOCs from ozone deposition, which is about a factor of 10 larger in the lab experiments than at Scripps Pier during field measurements. Using a proton transfer reaction mass spectrometer, the authors also tentatively identify the emitted VOCs as being primarily aldehydes, although there are limitations to their analytical method.
This work shows the yield of VOCs is competitive with DMS, which suggests that it could be an important source to consider when thinking about secondary aerosol formation or evaluating the oxidative capacity of the marine atmosphere.
General comments:
I am very interested to know more about the difference between the laboratory and field measurements. The authors provide 4 potential reasons for why the laboratory measurements had a higher ozone yield for VOCs relative to the field measurements, however some ideas could be discussed more thoroughly.
- The authors first reason for lower field yields is the relatively stable SSML in the lab relative to the more dynamic real SSML in the ocean. Did they authors specifically sample the SSML at Scripps? My impression is they collected the underlying seawater, which contained some unsaturated/insoluble species which then formed a new microlayer in the lab. Do the authors think that the composition between the laboratory SSML and the authentic SSML composition could impact the VOC yield? Two factors come to mind, including the composition of the SSML (i.e. the reactive component) and changes to the physical partitioning of the VOCs from the aqueous phase into the gas-phase.
- The second reason the authors hypothesize difference is the presence of photochemistry, which would be occurring in the environment and could reduce the lifetime of the VOCs in seawater to lower the flux. (L373) From Figure 1, the emission of VOCs has a diurnal profile which peaks in the afternoon, which is likely related to the ozone flux. Can the authors show the VOC yield from ozonolysis changes during the daytime and nighttime during their field study? This would perhaps disprove their second point.
- Is there any evidence towards biological factors reducing the lifetime of VOCs in the aqueous phase? For example, biological processes have been shown to be a sink of acetone (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7983863/ ), and I assume other VOCs as well based on recent reviews and studies, (https://www.sciencedirect.com/science/article/pii/S0012825223000491 , https://www.nature.com/articles/s41564-020-00859-8 )
- The role of downward mixing in the ocean vs the lab. In the iodide-ozone model built by Carpenter et al. (2013; https://www.nature.com/articles/ngeo1687 ), they found the rate of emission also depends on the downward mixing of the oxidized iodine, which sequesters the volatile products in the bulk ocean. This seems likely to be occurring with the organics, as well and should be proposed as a limitation of the laboratory measurement that could contribute to differences in yield.
The authors also make an important point that under marine conditions, the C5H9+ ion cannot be interpreted as being isoprene, since it is a fragment of many different, larger aldehydes. The authors used a GC-Vocus system to resolve the parent ions, thus separating the different soures of the C5H9+ ion fragment. Does the results from these experiments match the frgment library put together by Pagonis et al. (2019)? Why or why not? What parent ion contributes the most (if any) to the total C5H9+ ion signal?
Specific comments:
L75 – Authors state that the flux of carbon from oxidation is “competitive with the carbon mass flux from BVOC and a proposed photochemical source”. The authors provide an estimated range from a previous study for oxidation, but not for photochemical or BVOC flux; perhaps it would be clearer if all the ranges (and the limitations of their estimates, perhaps) were presented.
L235 – The authors state that degassing experiments observed BVOC like DMS, and reference Figure S3 which shows the emission of DMS and isoprene. Are these the only two BVOC ions observed? How were these attributed to BVOCs and not photochemical sources or other sources?
L263 – How was this threshold value (of 50 cps) chosen?
L439 – Why 8 carbons? Previously, Novak & Bertram (2021) used 5 carbons.
Errata
L270 – different color?
Citation: https://doi.org/10.5194/egusphere-2023-2210-RC2 -
AC1: 'Comment on egusphere-2023-2210', Delaney Kilgour, 16 Jan 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-2210/egusphere-2023-2210-AC1-supplement.pdf
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2023-2210', Anonymous Referee #2, 31 Oct 2023
This manuscript presents a unique combination of field observations of reactive organic carbon (ROC) fluxes and laboratory experiments probing reactive organic carbon emissions from a quiescent seawater microlayer reacting with gas phase ozone. The authors provide VOC yield estimates from both in situ ROC fluxes and laboratory ozone oxidation, and demonstrate that real world VOC yields are likely much smaller than laboratory estimations suggests. This is a very important point to make given that laboratory VOC yields from similar experiments have already been scaled to global emissions in some modelling studies. However, even scaling the much lower field VOC yield by an average ozone deposition flux points to an important source of ROC over the world's oceans. This work also makes two additional very useful points: (1) that the authors see no evidence for direct photochemistry driving VOC fluxes from the sea surface, and (2) that C5H9+ signals from PTR-type instruments should be interpreted with extreme caution in marine environments (i.e., building on Coggon et al., 2023 for urban centers).
General Comments1. (L19, L444) 21.1 Tg-C/yr from marine DMS emissions. It is difficult to understand where this number comes from. It is not cited, and doesn't appear to line up with current climatologies. For example, the third revision of the DMS climatology (Hulswar et al., 2022 https://essd.copernicus.org/articles/14/2963/2022/) estimates a global DMS emission of ~27.1 Tg-S/yr. Scaling to carbon in DMS gives ~20.3 Tg-C/yr. While this is relatively close to the number stated in the abstract and conclusions, this highlights the ambiguity in the origin for this comparison. The authors should cite recent climatologies for this comparison and make clear that this number is both not exactly known and arises from averaging over significant regional variability. Otherwise, this may not be a useful benchmark against which to judge the importance of VOC emissions driven by O3 deposition to the sea surface.
2. Use of the term heterogeneous (L29, L369 and elsewhere). Given the reactions studied occur at liquid interfaces, "multiphase" may be more appropriate. e.g., see https://acp.copernicus.org/articles/23/9765/2023/ for current suggestions on revisions to terminology. It is more useful to consider a multiphase reaction that may have substantial interfacial and bulk components, then it is to consider a perfectly "heterogeneous" surface-only reaction versus and entirely bulk phase reaction.
3. The estimation of multiphase O3 reactivity and k_DOC (L299 - L324) requires further, and more nuanced, discussion.
(a) The authors apportion O3 reactivity into an I- reactivity of 266 s^-1 and a DOC reactivity of 177 s^-1. Can the authors propagate uncertainties in the measurements that go into this estimation of O3 reactivity? Further, this O3 reactivity assumes a bulk rate coefficient and a bulk iodide concentration, and would be more accurately referred to as a "bulk reactivity," because (as the authors state later in the paper) the O3 reactions likely have a significant interfacial component.
(b) In addition to assuming the surface I- is not depleted by reaction, do the authors also assume that the surface I- concentration is equal to the bulk I- concentration? I- ions are known to have affinity for the air-water interface, and a Langmuir adsorption isotherm, with literature constraints on the bulk-surface partition coefficient, could be used to better constrain the magnitude of potential surface concentrations in the present experiments.
(c) Similarly, while the estimated k_DOC uses a bulk [DOC], it also uses an observed gas phase O3 loss and so inherently incorporates both interfacial and bulk O3 reactions of unknown relative importance. It cannot be simply stated (L311) that this reaction "likely happens at the interface," rather for a bimolecular rate coefficient on the order of 5e6 M^-1 s^-1, it is possible that interfacial reactions will make a substantial contribution to O3 loss, though that will depend on the interplay between O3 and DOC partitioning to the interface (e.g., https://pubs.acs.org/doi/full/10.1021/acs.jpca.2c03059)
(d) The authors go on (L317) to compare their mixed bulk/interfacial k_DOC to a set of strictly bulk rate coefficients for species used to approximate DOC. These bulk, bimolecular rate coefficients span 3 orders of magnitude, and may not represent species that are relevant to marine DOC. The comparison to an "authentic marine sample" (L319) is presumably that from Shaw & Carpenter 2013 (2.6e7 M^-1 s^-1), and is therefore a more appropriate comparison as this literature value arises from similar O3 deposition experiments and is thus a combination of bulk and interfacial O3 reactivity.4. Figure 4: Pie charts an interesting choice here as you can't show a range or variability in these contributions. A bar chart with error bars that correspond to some measure of variability in contributions to your observed yields would be much more informative.
5. The authors use field and lab average yields to bracket the range of possible VOC yields in the conclusions and abstract. It may be more appropriate to include the range of yields for both lab and field data, show in Table 1, into these discussions.
Specific CommentsL36: "This collection of VOC" -- this wording is somewhat confusing; which VOCs specifically?
L195: "Ions without an expected molecule" -- does this mean ions without a known, or single, contributing molecule?
L302: k_iodide = 2.4e9 M^-1 s^-1, presumably this value is from Magi et al., 1997? More recent determinations exist e.g., https://pubs.acs.org/doi/10.1021/ic000919j (for pH 6.7, 1.2 (+/-0.1)e9 M^-1 s^-1). Further, given the pH dependence of O3 + I-, is the Magi 1997 rate coefficient applicable to seawater pH? This should be discussed further.
L321: Alkanes reacting with ozone?
L332: "..range in yield reflects standard deviations in O3." -- +/- 1 sigma, or more?
L342-344: What fraction of organic carbon goes to the gas phase in the experiment? The authors appear to have the data to estimate this from the DOC measurement in sea water together with integrated delta-VOC over time.
L355: "..based on standard deviation of O3 mixing ratios measured in 2019" -- is the data normally distributed? Is this an appropriate measure of your uncertainty?
L356: why not include these additional uncertainties, or an estimate, to provide a more true measure of the uncertainty in your field VOC yield?
L376-378: "any O3 source variability was not directly measured simultaneously during experiments, meaning that quicker O3 fluctuations than what occur in the ambient could have heightened laboratory yields." -- do the authors expect this is a significant source of higher VOC yield when delta-VOC mirrors delta-O3? If so, this should be discussed further.
L380: "[I-] a factor of three lower" -- clarify if this is measured or inferred.
L385: "caution our ability" -- should be "complicate our ability"?
L405: (Figure 5 caption) Regression should be relationship, or similar? You are not quantitatively assessing the relationship between two variables (i.e., you are not applying a regression model)
L440: "a couple of hundred" -- avoid vague language, and give the range from the literature you cite and their uncertainties.
Citation: https://doi.org/10.5194/egusphere-2023-2210-RC1 -
RC2: 'Comment on egusphere-2023-2210', Anonymous Referee #1, 06 Nov 2023
General Comments
This paper by Kilgour et al. explores the emission of volatile organic compounds (VOCs) from the ozonolysis of seawater with a combination of field and laboratory measurements. From this work, the authors determine the approximate yield of VOCs from ozone deposition, which is about a factor of 10 larger in the lab experiments than at Scripps Pier during field measurements. Using a proton transfer reaction mass spectrometer, the authors also tentatively identify the emitted VOCs as being primarily aldehydes, although there are limitations to their analytical method.
This work shows the yield of VOCs is competitive with DMS, which suggests that it could be an important source to consider when thinking about secondary aerosol formation or evaluating the oxidative capacity of the marine atmosphere.
General comments:
I am very interested to know more about the difference between the laboratory and field measurements. The authors provide 4 potential reasons for why the laboratory measurements had a higher ozone yield for VOCs relative to the field measurements, however some ideas could be discussed more thoroughly.
- The authors first reason for lower field yields is the relatively stable SSML in the lab relative to the more dynamic real SSML in the ocean. Did they authors specifically sample the SSML at Scripps? My impression is they collected the underlying seawater, which contained some unsaturated/insoluble species which then formed a new microlayer in the lab. Do the authors think that the composition between the laboratory SSML and the authentic SSML composition could impact the VOC yield? Two factors come to mind, including the composition of the SSML (i.e. the reactive component) and changes to the physical partitioning of the VOCs from the aqueous phase into the gas-phase.
- The second reason the authors hypothesize difference is the presence of photochemistry, which would be occurring in the environment and could reduce the lifetime of the VOCs in seawater to lower the flux. (L373) From Figure 1, the emission of VOCs has a diurnal profile which peaks in the afternoon, which is likely related to the ozone flux. Can the authors show the VOC yield from ozonolysis changes during the daytime and nighttime during their field study? This would perhaps disprove their second point.
- Is there any evidence towards biological factors reducing the lifetime of VOCs in the aqueous phase? For example, biological processes have been shown to be a sink of acetone (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7983863/ ), and I assume other VOCs as well based on recent reviews and studies, (https://www.sciencedirect.com/science/article/pii/S0012825223000491 , https://www.nature.com/articles/s41564-020-00859-8 )
- The role of downward mixing in the ocean vs the lab. In the iodide-ozone model built by Carpenter et al. (2013; https://www.nature.com/articles/ngeo1687 ), they found the rate of emission also depends on the downward mixing of the oxidized iodine, which sequesters the volatile products in the bulk ocean. This seems likely to be occurring with the organics, as well and should be proposed as a limitation of the laboratory measurement that could contribute to differences in yield.
The authors also make an important point that under marine conditions, the C5H9+ ion cannot be interpreted as being isoprene, since it is a fragment of many different, larger aldehydes. The authors used a GC-Vocus system to resolve the parent ions, thus separating the different soures of the C5H9+ ion fragment. Does the results from these experiments match the frgment library put together by Pagonis et al. (2019)? Why or why not? What parent ion contributes the most (if any) to the total C5H9+ ion signal?
Specific comments:
L75 – Authors state that the flux of carbon from oxidation is “competitive with the carbon mass flux from BVOC and a proposed photochemical source”. The authors provide an estimated range from a previous study for oxidation, but not for photochemical or BVOC flux; perhaps it would be clearer if all the ranges (and the limitations of their estimates, perhaps) were presented.
L235 – The authors state that degassing experiments observed BVOC like DMS, and reference Figure S3 which shows the emission of DMS and isoprene. Are these the only two BVOC ions observed? How were these attributed to BVOCs and not photochemical sources or other sources?
L263 – How was this threshold value (of 50 cps) chosen?
L439 – Why 8 carbons? Previously, Novak & Bertram (2021) used 5 carbons.
Errata
L270 – different color?
Citation: https://doi.org/10.5194/egusphere-2023-2210-RC2 -
AC1: 'Comment on egusphere-2023-2210', Delaney Kilgour, 16 Jan 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-2210/egusphere-2023-2210-AC1-supplement.pdf
Peer review completion
Journal article(s) based on this preprint
Data sets
Production of oxygenated volatile organic compounds from the ozonolysis of coastal seawater Delaney B. Kilgour, Gordon A. Novak, Timothy H. Bertram http://digital.library.wisc.edu/1793/84597
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 387 | 184 | 18 | 589 | 46 | 9 | 18 |
- HTML: 387
- PDF: 184
- XML: 18
- Total: 589
- Supplement: 46
- BibTeX: 9
- EndNote: 18
Viewed (geographical distribution)
| Country | # | Views | % |
|---|---|---|---|
| United States of America | 1 | 218 | 37 |
| China | 2 | 79 | 13 |
| Germany | 3 | 45 | 7 |
| United Kingdom | 4 | 45 | 7 |
| France | 5 | 25 | 4 |
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
- 218
Cited
1 citations as recorded by crossref.
Delaney B. Kilgour
Gordon A. Novak
Megan S. Claflin
Brian M. Lerner
Timothy H. Bertram
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1050 KB) - Metadata XML
-
Supplement
(922 KB) - BibTeX
- EndNote
- Final revised paper