the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 26 Apr 2023
Measurement Report: Carbonyl Sulfide production during Dimethyl Sulfide oxidation in the atmospheric simulation chamber SAPHIR
Domenico Taraborrelli
Sascha Alber
Birger Bohn
Hans-Peter Dorn
Hendrik Fuchs
Chenxi Qiu
Franz Rohrer
Roberto Sommariva
Fred Stroh
Zhaofeng Tan
Sergej Wedel
Anna Novelli
Abstract. Carbonyl sulfide (OCS), the most abundant sulfur gas in the Earth’s atmosphere, is a greenhouse gas, a precursor to stratospheric sulfate aerosol, and a proxy for terrestrial CO2 uptake. Estimates of important OCS sources and sinks still bear significant uncertainties and the global budget is not considered closed. One particularly uncertain source term, the OCS production during the atmospheric oxidation of dimethyl sulfide (DMS) emitted by the oceans, is addressed by a series of experiments in the atmospheric simulation chamber SAPHIR at conditions comparable to the remote marine atmosphere. DMS oxidation was initiated with OH and/or Cl radicals and DMS, OCS and several oxidation products and intermediates were measured, including hydroperoxymethyl thioformate (HPMTF) that was recently found to play a key role in DMS oxidation in the marine atmosphere. One important finding is that the onset of HPMTF and OCS formation occurred faster than expected from the current chemical mechanisms. In agreement with other recent studies, OCS yields between 9 and 12 % were observed in our experiments. Such yields are substantially higher than the 0.7 % yield measured in laboratory experiments in the 1990s that is generally used to estimate the indirect OCS source from DMS in global budget estimates. However, we do not expect the higher yields found in our experiments to directly translate to a substantially higher OCS source from DMS oxidation in the real atmosphere, where conditions are highly variable and, as pointed out in recent work, heterogeneous HPMTF loss is expected to effectively limit OCS production via this pathway. Together with other experimental studies, our results will be helpful to further elucidate the DMS oxidation chemical mechanism and in particular the paths leading to OCS formation.
-
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
(1091 KB)
-
Supplement
(428 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1091 KB) - Metadata XML
-
Supplement
(428 KB) - BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Marc von Hobe et al.
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2023-639', Anonymous Referee #1, 19 May 2023
General Comments and Overview
The work by Marc von Hobe et al. 2023 presents a selection of chamber experiments meant to better understand the connection between DMS oxidation and OCS production. This study utilizes an impressive suite of instrumentation and oxidative environments to better resolve the importance of DMS oxidation. The work compares its findings to previous work on DMS oxidation and the importance of the recently discovered molecule hydroperoxymethyl thioformate (HPMTF). This study presented not only validates the findings of previous work, but also expands the discussion by adding photolysis and chloride oxidation to the conversation. This work reports the first attempts to determine a potential experimental correlation between HPMTF and OCS through the lens of photolysis and chloride oxidation. This experimentation is important for progressing the understanding of DMS oxidation and as such the conclusion and results should expand and reiterate the photolysis and chloride findings. I recommend the publication of this work. I would recommend the elaboration and restructuring of the work before final publication. Generally, I would stress the importance of oxidative and environmental drivers of OCS production from DMS oxidation throughout the work as well as the role of chloride and photolysis within DMS oxidation. I would also recommend remaking the figures with updated legends to help the reader better understand which time traces are which.
Specific Comments
Line 112: A clarification of why photolysis is not a driving reaction would be greatly appreciated. 254 nm light is a higher energy light. Do you assume insignificant due to OH concentration or low light flux?
Table 2: TPA or Thioperfomric acid is not referenced before this point. Would recommend adding the full name of the species before adding the abbreviation.
Line 140: Ye et al and Jernigan et al found other DMS related products within their PTR-MS. Were any additional sulfur species found using the PTR implemented in this chamber experiment. Even uncalibrated signal could help understand when potential intermediates react to the perturbations within the chamber.
Line 168: If the signal from HPMTF is significant than the isotopic pattern of the sulfur from HPMTF should dominate over the potential pattern of N2O5. This could provide an experimental constraint on the identity of the 235 m/z species.
Line 175: I*SO2 is highly water dependent with the cluster only able to be used if the conditions are dry and the ratio of I*H2O to I are very low. The high and changing RH in the system would make this measurement hard.
Line 177: CO3- can arise from O2- chemistry and ozone. The variability on the SO2 measurement using this technique could be also due to the varying O3 concentration. See Novak et al (https://doi.org/10.5194/amt-13-1887-2020) for an in dept description of the CO3- chemistry and its correlation to O3. This interplay could change the sensitivity of SO2 in the CIMS.
Line 181: I understand the difficulty in calibrating the sulfur intermediates and understand the use of ncts for the observation of HPMTF and TPA. Were the ncts for HPMTF, TPA, and SO2 background subtracted? Showing the signal as a delta would help illustrate the change in signal over the course of the chamber experiment.
Line 183: How much did the humidity drop? Could this also affect the potential for humidity wall driven reactions throughout the chamber experiment?
I would reference supp. Figure 1 as it presents the change in RH over the experiment
Line 204: Looking at Figure S1 it seems that the temperature within the chamber changes by almost 10C over the time of the experiment and that temperature never seems to reach 25C. I would add a comment on this and how this much change in temp could alter the observed DMS + OH and the fraction of DMS shutting down either the OH addition and abstraction channels.
Line 205: How was dilution determined within the chamber? Were experiments performed elsewhere or was there a dilution tracer added and monitored within the experiment?
Line 213: Were any experiments done without sulfur (DMS) present while the system was oxidized utilizing the entire array of instrumentation. I am wondering if there is potential for the contribution of contaminants that could lead to prompt formation of OCS. You state that other VOC (line 208) may have off gassed leading to higher CO, could this lead to a prompt/background OCS?
Line 215: I think it is important to say that the OCS yield is an experimental yield that takes into the environmental parameters. This removes the ease to quickly see this value as a way to simplify the global yield of OCS from DMS oxidation.
Could you also describe how the OCS yield was calculated throughout the chamber experiment? Was the OCS yield calculated from comparing the DMS loss rate to OCS production rate as is shown in the figures or was the yield calculated from the change in signal of DMS and OCS (d[OCS]/d[DMS])? Adding a equation would help illustrate this cleanly.
Line 236: Dyke et al and others have proposed the reaction of DMS with Cl2 can lead to the formation of CH3SCH2Cl (DMS-Cl). Was there any evidence in your VOC measurements (PTR) that this intermediate arose upon the addition of Cl2? And could this intermediate and its subsequent photolysis or OH oxidation change the OCS yield? Lastly, Urbanski and Wine and Arsene et al found that the DMS + Cl could form an adduct with eventual addition of a O2. Is there any evidence in your PTR measurements of additional OH-addition like chemistry? (i.e. increased DSMO/MSIA/DMSO2)
Shallcross et al 2006 AtmEnv; Copeland et al 2013 EST; https://doi.org/10.1021/jp992682m (Wine and Urbanski)
Line 243: The temperature in the chamber is greater than 298K (S3 reads about ~305K). This elevated temperature would effect the fraction DMS oxidizing by the OH additional and abstraction as well as the isomerization rate. A comment about the varability in temperature across all experiments and if the variability in temperature would affect the product distribution would be greatly appreciated. OH-addition has a strong temperature dependence and as such would greatly change the amount of DMS shuttling down HPMTF and OCS.
Line 252: Barnes et al 1991 (https://doi.org/10.1002/kin.550230704) found that ClO can react with DMS at a rate constant of (9.5 ± 2.0) × 10−15 cm3 molecule−1 s−1. I think it is important to site this finding as support for the idea that this reaction could compete. They also suggest that the reaction would lead to the formation of DMSO, shutting of the potential production of HPMTF and OCS. If possible a small discussion/observations of chloride chemistry would be appreciated. Iodide CIMS has the potential to observe multiple halogens (i.e. Cl2, ClO, ClNO2). Do you have any observations of other halogen species (ClO, ClNO2) that could help understand the lower observed OCS yield and the mechanisms driving the DMS oxidation? With Cl2, JCl and O3 observations could you model/approximate the [ClO]?
Line 275: Could the drop in HPMTF also be associated with increased NO and thus less HPMTF (e.g RO2 + NO dominates/increases over isomerization). S3 shows a significant step change in NOx with a subsequent sustained concentration of O3 leading one to question the start of NOx cycling. Could NO chemistry start upon the opening of the roof?
Line 276: Could an approximate photolysis term for HPMTF be calculated using the decay in HPMTF? Using the assumption that HPMTF + OH is 1-2E-11 from Jernigan et al and Ye et al, could an approximate photolysis rate be determined from the missing HPMTF loss? This may be an over simplification of the observations, but an approximate vale could be compared to the values used in Khan et al to determine the weight of HPMTF photolysis and its potential roll in OCS formation.
Line 289: The clarifications done here are greatly appreciated in showing the complexity of OCS from DMS oxidation and the use of mechanism over a yield. I would still recommend the use of “experimental” or “chamber” as a clarifier for the observed OCS yields.
All Figures: If the OCS yield is calculated by relating the rates of DMS loss and OCS production, then I would recommend adding the yield as a function of time in the experiment. This could help show if the value is constant or variable depending on the various perturbations. I can also see that DMS and OCS may not be tightly correlated given the need to transition though intermediates, so this time dependent yield could be misleading. Any comments on this would be appreciated.
Figure 1: Do any species respond to the second addition of ozone? Upon first glance nothing in the SI or figure responds to the increased O3 and subsequent OH formation?
Figure 3: There seems to be an abrupt and unexplained drop in Cl2 at hour ~4.6 after DMS injection. Could you please explain what occurred at this hour and if that abrupt drop and loss of chloride could be used to understand a transition between Cl and OH dominated oxidation?
Technical comments
Line 165: Define TPA
Figures: I have trouble reading the axis titles, especially the mulipliter (i.e. 10-5). Could you please increase the font or resolution of the figures.
Figure 1: Which measurement for HCOH was used here. The table states two instruments were used. Please add for this and all future figures if the concentrations are dilution corrected.
Figure 4: I recommend adding labels for the three modeled lines in 4c. I see that the description is in the description, but a label will help guide the eye and discern colors.
Citation: https://doi.org/10.5194/egusphere-2023-639-RC1 -
AC1: 'Reply on RC1', Marc von Hobe, 18 Jul 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-639/egusphere-2023-639-AC1-supplement.pdf
-
AC1: 'Reply on RC1', Marc von Hobe, 18 Jul 2023
-
RC2: 'Comment on egusphere-2023-639', Anonymous Referee #2, 20 May 2023
Marc von Hobe et al. present experimental results of dimethyl sulfide oxidation in the chamber SAPHIR and quantified OCS yields. They observed that OCS formation start soon after DMS oxidation started, and the derived OCS yields are between 9% - 12%. They used both OH and Cl as the oxidant, and their results suggest that OCS can also be formed from DMS+Cl/Cl2. The paper is generally well written, and provide another important experimental dataset of OCS production from DMS oxidation that has a higher yield than previously thought. The paper is suitable for publication after the authors address the following comments.
Major comments:
1. The figures all have low resolutions. It is difficult to read the superscripts in the axis legend. Figures of a better quality need to be provided.
2 Are there any losses of precursor or products to the chamber wall?
3. In terms of the sulfur budget, were there any aerosol formations in these experiments?
4. In Figure 2, it seems that DMS started to decay before the UV light were turned on. Is the decay from DMS+O3? The figure also suggests that HPMTF formation started before light was on. Does this indicate HPMTF can also be generated from DMS+O3?
Technical comments:
5. Line 546, should be “OCS production rates given in panel d)”.
6. The OCS time series are quite fluctuated. Did the authors calculate OCS production rates using a longer period of time and what is the difference?
Citation: https://doi.org/10.5194/egusphere-2023-639-RC2 -
AC2: 'Reply on RC2', Marc von Hobe, 18 Jul 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-639/egusphere-2023-639-AC2-supplement.pdf
-
AC2: 'Reply on RC2', Marc von Hobe, 18 Jul 2023
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2023-639', Anonymous Referee #1, 19 May 2023
General Comments and Overview
The work by Marc von Hobe et al. 2023 presents a selection of chamber experiments meant to better understand the connection between DMS oxidation and OCS production. This study utilizes an impressive suite of instrumentation and oxidative environments to better resolve the importance of DMS oxidation. The work compares its findings to previous work on DMS oxidation and the importance of the recently discovered molecule hydroperoxymethyl thioformate (HPMTF). This study presented not only validates the findings of previous work, but also expands the discussion by adding photolysis and chloride oxidation to the conversation. This work reports the first attempts to determine a potential experimental correlation between HPMTF and OCS through the lens of photolysis and chloride oxidation. This experimentation is important for progressing the understanding of DMS oxidation and as such the conclusion and results should expand and reiterate the photolysis and chloride findings. I recommend the publication of this work. I would recommend the elaboration and restructuring of the work before final publication. Generally, I would stress the importance of oxidative and environmental drivers of OCS production from DMS oxidation throughout the work as well as the role of chloride and photolysis within DMS oxidation. I would also recommend remaking the figures with updated legends to help the reader better understand which time traces are which.
Specific Comments
Line 112: A clarification of why photolysis is not a driving reaction would be greatly appreciated. 254 nm light is a higher energy light. Do you assume insignificant due to OH concentration or low light flux?
Table 2: TPA or Thioperfomric acid is not referenced before this point. Would recommend adding the full name of the species before adding the abbreviation.
Line 140: Ye et al and Jernigan et al found other DMS related products within their PTR-MS. Were any additional sulfur species found using the PTR implemented in this chamber experiment. Even uncalibrated signal could help understand when potential intermediates react to the perturbations within the chamber.
Line 168: If the signal from HPMTF is significant than the isotopic pattern of the sulfur from HPMTF should dominate over the potential pattern of N2O5. This could provide an experimental constraint on the identity of the 235 m/z species.
Line 175: I*SO2 is highly water dependent with the cluster only able to be used if the conditions are dry and the ratio of I*H2O to I are very low. The high and changing RH in the system would make this measurement hard.
Line 177: CO3- can arise from O2- chemistry and ozone. The variability on the SO2 measurement using this technique could be also due to the varying O3 concentration. See Novak et al (https://doi.org/10.5194/amt-13-1887-2020) for an in dept description of the CO3- chemistry and its correlation to O3. This interplay could change the sensitivity of SO2 in the CIMS.
Line 181: I understand the difficulty in calibrating the sulfur intermediates and understand the use of ncts for the observation of HPMTF and TPA. Were the ncts for HPMTF, TPA, and SO2 background subtracted? Showing the signal as a delta would help illustrate the change in signal over the course of the chamber experiment.
Line 183: How much did the humidity drop? Could this also affect the potential for humidity wall driven reactions throughout the chamber experiment?
I would reference supp. Figure 1 as it presents the change in RH over the experiment
Line 204: Looking at Figure S1 it seems that the temperature within the chamber changes by almost 10C over the time of the experiment and that temperature never seems to reach 25C. I would add a comment on this and how this much change in temp could alter the observed DMS + OH and the fraction of DMS shutting down either the OH addition and abstraction channels.
Line 205: How was dilution determined within the chamber? Were experiments performed elsewhere or was there a dilution tracer added and monitored within the experiment?
Line 213: Were any experiments done without sulfur (DMS) present while the system was oxidized utilizing the entire array of instrumentation. I am wondering if there is potential for the contribution of contaminants that could lead to prompt formation of OCS. You state that other VOC (line 208) may have off gassed leading to higher CO, could this lead to a prompt/background OCS?
Line 215: I think it is important to say that the OCS yield is an experimental yield that takes into the environmental parameters. This removes the ease to quickly see this value as a way to simplify the global yield of OCS from DMS oxidation.
Could you also describe how the OCS yield was calculated throughout the chamber experiment? Was the OCS yield calculated from comparing the DMS loss rate to OCS production rate as is shown in the figures or was the yield calculated from the change in signal of DMS and OCS (d[OCS]/d[DMS])? Adding a equation would help illustrate this cleanly.
Line 236: Dyke et al and others have proposed the reaction of DMS with Cl2 can lead to the formation of CH3SCH2Cl (DMS-Cl). Was there any evidence in your VOC measurements (PTR) that this intermediate arose upon the addition of Cl2? And could this intermediate and its subsequent photolysis or OH oxidation change the OCS yield? Lastly, Urbanski and Wine and Arsene et al found that the DMS + Cl could form an adduct with eventual addition of a O2. Is there any evidence in your PTR measurements of additional OH-addition like chemistry? (i.e. increased DSMO/MSIA/DMSO2)
Shallcross et al 2006 AtmEnv; Copeland et al 2013 EST; https://doi.org/10.1021/jp992682m (Wine and Urbanski)
Line 243: The temperature in the chamber is greater than 298K (S3 reads about ~305K). This elevated temperature would effect the fraction DMS oxidizing by the OH additional and abstraction as well as the isomerization rate. A comment about the varability in temperature across all experiments and if the variability in temperature would affect the product distribution would be greatly appreciated. OH-addition has a strong temperature dependence and as such would greatly change the amount of DMS shuttling down HPMTF and OCS.
Line 252: Barnes et al 1991 (https://doi.org/10.1002/kin.550230704) found that ClO can react with DMS at a rate constant of (9.5 ± 2.0) × 10−15 cm3 molecule−1 s−1. I think it is important to site this finding as support for the idea that this reaction could compete. They also suggest that the reaction would lead to the formation of DMSO, shutting of the potential production of HPMTF and OCS. If possible a small discussion/observations of chloride chemistry would be appreciated. Iodide CIMS has the potential to observe multiple halogens (i.e. Cl2, ClO, ClNO2). Do you have any observations of other halogen species (ClO, ClNO2) that could help understand the lower observed OCS yield and the mechanisms driving the DMS oxidation? With Cl2, JCl and O3 observations could you model/approximate the [ClO]?
Line 275: Could the drop in HPMTF also be associated with increased NO and thus less HPMTF (e.g RO2 + NO dominates/increases over isomerization). S3 shows a significant step change in NOx with a subsequent sustained concentration of O3 leading one to question the start of NOx cycling. Could NO chemistry start upon the opening of the roof?
Line 276: Could an approximate photolysis term for HPMTF be calculated using the decay in HPMTF? Using the assumption that HPMTF + OH is 1-2E-11 from Jernigan et al and Ye et al, could an approximate photolysis rate be determined from the missing HPMTF loss? This may be an over simplification of the observations, but an approximate vale could be compared to the values used in Khan et al to determine the weight of HPMTF photolysis and its potential roll in OCS formation.
Line 289: The clarifications done here are greatly appreciated in showing the complexity of OCS from DMS oxidation and the use of mechanism over a yield. I would still recommend the use of “experimental” or “chamber” as a clarifier for the observed OCS yields.
All Figures: If the OCS yield is calculated by relating the rates of DMS loss and OCS production, then I would recommend adding the yield as a function of time in the experiment. This could help show if the value is constant or variable depending on the various perturbations. I can also see that DMS and OCS may not be tightly correlated given the need to transition though intermediates, so this time dependent yield could be misleading. Any comments on this would be appreciated.
Figure 1: Do any species respond to the second addition of ozone? Upon first glance nothing in the SI or figure responds to the increased O3 and subsequent OH formation?
Figure 3: There seems to be an abrupt and unexplained drop in Cl2 at hour ~4.6 after DMS injection. Could you please explain what occurred at this hour and if that abrupt drop and loss of chloride could be used to understand a transition between Cl and OH dominated oxidation?
Technical comments
Line 165: Define TPA
Figures: I have trouble reading the axis titles, especially the mulipliter (i.e. 10-5). Could you please increase the font or resolution of the figures.
Figure 1: Which measurement for HCOH was used here. The table states two instruments were used. Please add for this and all future figures if the concentrations are dilution corrected.
Figure 4: I recommend adding labels for the three modeled lines in 4c. I see that the description is in the description, but a label will help guide the eye and discern colors.
Citation: https://doi.org/10.5194/egusphere-2023-639-RC1 -
AC1: 'Reply on RC1', Marc von Hobe, 18 Jul 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-639/egusphere-2023-639-AC1-supplement.pdf
-
AC1: 'Reply on RC1', Marc von Hobe, 18 Jul 2023
-
RC2: 'Comment on egusphere-2023-639', Anonymous Referee #2, 20 May 2023
Marc von Hobe et al. present experimental results of dimethyl sulfide oxidation in the chamber SAPHIR and quantified OCS yields. They observed that OCS formation start soon after DMS oxidation started, and the derived OCS yields are between 9% - 12%. They used both OH and Cl as the oxidant, and their results suggest that OCS can also be formed from DMS+Cl/Cl2. The paper is generally well written, and provide another important experimental dataset of OCS production from DMS oxidation that has a higher yield than previously thought. The paper is suitable for publication after the authors address the following comments.
Major comments:
1. The figures all have low resolutions. It is difficult to read the superscripts in the axis legend. Figures of a better quality need to be provided.
2 Are there any losses of precursor or products to the chamber wall?
3. In terms of the sulfur budget, were there any aerosol formations in these experiments?
4. In Figure 2, it seems that DMS started to decay before the UV light were turned on. Is the decay from DMS+O3? The figure also suggests that HPMTF formation started before light was on. Does this indicate HPMTF can also be generated from DMS+O3?
Technical comments:
5. Line 546, should be “OCS production rates given in panel d)”.
6. The OCS time series are quite fluctuated. Did the authors calculate OCS production rates using a longer period of time and what is the difference?
Citation: https://doi.org/10.5194/egusphere-2023-639-RC2 -
AC2: 'Reply on RC2', Marc von Hobe, 18 Jul 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-639/egusphere-2023-639-AC2-supplement.pdf
-
AC2: 'Reply on RC2', Marc von Hobe, 18 Jul 2023
Peer review completion
Journal article(s) based on this preprint
Marc von Hobe et al.
Data sets
Atmospheric simulation chamber study: Dimethyl sulfide + OH - Gas-phase oxidation - product study A. Novelli, M. von Hobe, H. Fuchs, F. Rohrer, and S. Wedel https://doi.org/10.25326/667N-KA95
Atmospheric simulation chamber study: Dimethyl sulfide + OH - Gas-phase oxidation - product study A. Novelli, M. von Hobe, S. Alber, H. Fuchs, Y. Li, C. Qiu, F. Rohrer, S. Wedel, and F. Stroh https://doi.org/10.25326/74PS-NM77
Atmospheric simulation chamber study: Dimethyl sulfide + OH - Gas-phase oxidation - product study A. Novelli, M. von Hobe, S. Alber, H. Fuchs, Y. Li, C. Qiu, F. Rohrer, S. Wedel, F. Stroh, B. Bohn, H.-P. Dorn, R. Sommariva, and Z. Tan https://doi.org/10.25326/57DX-WR36
Atmospheric simulation chamber study: Dimethyl sulfide + OH - Gas-phase oxidation - product study A. Novelli, M. von Hobe, S. Alber, H. Fuchs, Y. Li, C. Qiu, F. Rohrer, S. Wedel, F. Stroh, B. Bohn, and H.-P. Dorn https://doi.org/10.25326/BYAV-YK31
Marc von Hobe et al.
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 371 | 177 | 17 | 565 | 44 | 6 | 8 |
- HTML: 371
- PDF: 177
- XML: 17
- Total: 565
- Supplement: 44
- BibTeX: 6
- EndNote: 8
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1091 KB) - Metadata XML
-
Supplement
(428 KB) - BibTeX
- EndNote
- Final revised paper