From the journal Environmental Science: Atmospheres Peer review history

Development of volatility distributions for organic matter in biomass burning emissions

Round 1

Manuscript submitted on 07 Jul 2022
 

26-Jul-2022

Dear Dr Grieshop:

Manuscript ID: EA-ART-07-2022-000080
TITLE: Development of Volatility Distributions for Organic Matter in Biomass Burning Emissions

Thank you for your submission to Environmental Science: Atmospheres, published by the Royal Society of Chemistry. I sent your manuscript to reviewers and I have now received their reports which are copied below.

I have carefully evaluated your manuscript and the reviewers’ reports, and the reports indicate that major revisions are necessary.

Please submit a revised manuscript which addresses all of the reviewers’ comments. Further peer review of your revised manuscript may be needed. When you submit your revised manuscript please include a point by point response to the reviewers’ comments and highlight the changes you have made. Full details of the files you need to submit are listed at the end of this email.

Please submit your revised manuscript as soon as possible using this link:

*** PLEASE NOTE: This is a two-step process. After clicking on the link, you will be directed to a webpage to confirm. ***

https://mc.manuscriptcentral.com/esatmos?link_removed

(This link goes straight to your account, without the need to log on to the system. For your account security you should not share this link with others.)

Alternatively, you can login to your account (https://mc.manuscriptcentral.com/esatmos) where you will need your case-sensitive USER ID and password.

You should submit your revised manuscript as soon as possible; please note you will receive a series of automatic reminders. If your revisions will take a significant length of time, please contact me. If I do not hear from you, I may withdraw your manuscript from consideration and you will have to resubmit. Any resubmission will receive a new submission date.

The Royal Society of Chemistry requires all submitting authors to provide their ORCID iD when they submit a revised manuscript. This is quick and easy to do as part of the revised manuscript submission process. We will publish this information with the article, and you may choose to have your ORCID record updated automatically with details of the publication.

Please also encourage your co-authors to sign up for their own ORCID account and associate it with their account on our manuscript submission system. For further information see: https://www.rsc.org/journals-books-databases/journal-authors-reviewers/processes-policies/#attribution-id

Environmental Science: Atmospheres strongly encourages authors of research articles to include an ‘Author contributions’ section in their manuscript, for publication in the final article. This should appear immediately above the ‘Conflict of interest’ and ‘Acknowledgement’ sections. I strongly recommend you use CRediT (the Contributor Roles Taxonomy from CASRAI, https://casrai.org/credit/) for standardised contribution descriptions. All authors should have agreed to their individual contributions ahead of submission and these should accurately reflect contributions to the work. Please refer to our general author guidelines http://www.rsc.org/journals-books-databases/journal-authors-reviewers/author-responsibilities/ for more information.

I look forward to receiving your revised manuscript.

Yours sincerely,
Dr Tzung-May Fu
Associate Editor
Environmental Science: Atmospheres
Royal Society of Chemistry

************


 
Reviewer 1

The paper describes a method to derive the VBS of biomass burning aerosols using a thermal desorption method in which particulate matter and higher volatility gas-phase compounds (up to about log10(C*) = 6) are collected using a combination of a quartz filter punch and a sorbent (in this case Carbotrap). The filter punch and the sorbent are placed in the same thermal desorption tube and are desorbed together during the analysis using TD-GC-MS. Deuterated alkane compounds are used to estimate the volatility of other compounds based on their retention time and to estimate their amount in the sample. This data, corrected using thermos-optical OC measurements, is then used to populate the VBS bins. The method was applied to several laboratory and field measurements of biomass burning emissions and provided a reasonable agreement with VBS of biomass burning aerosols measured using other techniques. The presented method has a significant potential to expand our ability to measure VBS of biomass burning aerosols and understand their chemical transformations in the atmosphere. The presented VBS data from several laboratory and ambient experiments is also very useful for the atmospheric chemistry community. I recommend this paper to be published after the authors address relatively minor issues listed below.

Specific comments:

Some aspects of the presented method need to be better described. For example, I assume the filter punch is placed upstream of the sorbent, but this is not mentioned in the text. It is not clear when the tube is spiked with deuterated standards, before or after sampling? Spiking prior to sampling could provide a better measure of recovery that includes any potential losses during sampling. The sorbent used is Carbotrap F and C. What is the ratio of these, were the tubes purchased pre-filled? How was the column bleed determined and used to correct the TIC chromatogram?

The correction using thermo-optical measurements (Section S1) is rather confusing: were OC measurements artifact corrected or not? "Background" correction could be interpreted as artifact correction or it could be media blank correction. This needs to be clarified. If there was no artifact correction, then an estimate of the artifact should be provided.

The authors should discuss the reason for sorbent selection. For example, some compounds could be reactive during sampling when using Carbotrap (Rothweiler et al., 1991). The sorbent selection could also affect the recovery of some of the (heavier and more polar) compounds during sampling and desorption.

Were the tubes tested for any breakthrough during sampling? Lighter and more polar compounds could suffer from breakthrough losses. This could provide an explanation for some of the observed discrepancy with the thermo-optical measurements. Potential losses during sampling should be added to the limitations of the method.

Given the confidence range of the response factor fit (up to a factor of 2), the error bars in Figure 3 should be larger than the variability of the triplicate measurements.

Related to the above point, the magnitude of the correction using OC measurements (Section S1) should be mentioned in the text, not only in a figure caption. It is very significant (up to an order of magnitude for the field measurements). This could introduce an additional uncertainty to both the VBS shape and the EF determined using the method.

Since there is almost an order of magnitude more material in the log10C = 5 and 6 bins, the amount of material from those bins in the PM could be comparable to that of the log10C = 4 bin. It would be helpful to see an order of magnitude analysis whether the omission of the higher volatility bins could bias the comparison with May et al.


Reference

Rothweiler H., Wäger P.A, Schlatter C. (1991) Comparison of Tenax Ta and Carbotrap for sampling and analysis of volatile organic compounds in air, Atmospheric Environment. Part B.
V25, 231-235.

Reviewer 2

Wildfires have recently occurred more common and intense in the US, making it increasingly important to understand the chemical and physical properties of biomass burning (BB) emissions. This study investigated the volatility of organic emission from different types of BB using the filter-in-tube sorbent tube sampling method. The volatility distribution showed that the IVOCs accounted for a large fraction of total captured organic matter, followed by SVOCs and LVOCs. A phase equilibrium equation was applied to predict gas-to-particle partitioning and emission factors (EFs) from sampling data, and the results were basically consistent with previous experiments. The paper is relatively well organized and the results are of interest to the atmospheric community. However, some aspects need additional discussion, especially for the methods of sample analysis and predicted EFs from equilibrium equation. Major revision is recommended. Detailed comments are listed below.
1. Insufficient information related to sample collection and analysis.
a) Line 181-182: The author used a secondary dilution system to dilute the RWC emissions. However, no information was provided for the dilution ratio. Besides, the author also conducted a chamber experiment to simulate open burning. Why were the samples for this experiment not diluted? More importantly, the emission factors and gas-to-particle partitioning of IVOCs may be related to the dilution process during sample collection as indicated in previous studies (Qian et al., 2021; Shrivastava et al., 2006). The argument would become more compelling if there are some discussions about the influence of dilution processes on the PM emission factors and volatility distributions.
b) A TD/GC/MS was applied to measure the TIC of BB samples. It must be emphasized that the combustion of biomass will also emit abundant oxygenated IVOCs (OIVOCs) such as organic acids, oxygenated aromatic hydrocarbons, alcohols, and so on (Liang et al., 2021), while the TD/GC/MS can only partially measure these polar compounds. Besides, OIVOCs with higher polarity are important in further SOA formation through gas-to-particle partitioning. Will the missing measurement affect the results of BB volatility distribution profiles? I noticed in Section 3.1 that the author tried to use authentic standards to obtain the response factors for different classes of compounds. However, there are few polar substances among the authentic standards. Even if the TIC can somehow reflect the OIVOCs, it typically has lower response factors than hydrocarbons. Therefore, I wonder if it is appropriate to convert the entire IVOC TIC signal to mass concentrations using hydrocarbon response factors.
2. The author quantitatively analyzed the LVOCs, SVOCs, and IVOCs in different volatility bins. Is it possible to calculate the EFs for LVOCs, SVOCs, and IVOCs under different BB types in Section 3.2 (Line 329, it should be Section 3.2 but there was a typo?). Moreover, the author only discussed the relationship between PM2.5 EFs with OC/EC EFs. How about the relationship between IVOCs, SVOCs, and LVOCs EFs with OC/EC EFs? Further, the author also used authentic standards for calibration, while no information was provided for speciated organic compounds. It would be interesting if these data can be discussed and will be beneficial for reducing the uncertainty of emission inventory in modelling studies.
3. Section 3.4: The author tried to use the volatility distributions to predict the EFs. However, I wonder if this equation holds for any sampling conditions? This uncertainty of this method is also indicated by the large deviations (mostly more than 50%) of the POA EFs predictions at 10 μg/m3. More specifically, if we apply the predicted results to other emission-based models, will this uncertainty be magnified or reduced? More discussion should be presented related to the predicted EFs.

Minor Comments:
1. Line 194-195: The author used the deuterium substitutes to track the analyte recovery, what is the TD recovery of samples?
2. Figure 1B: Why the saturated acids were excluded from response factor analysis?
3. Line 303-305: It seems improper to use the ∗ bins to define SOA production in this study, as the author directly collected the emission samples.
4. Figure S3: It is difficult to get the information from Figure S3 that the OC EFs are negatively correlated to MCE, and the EC EFs are positively correlated to MCE. Could the author add a dashed line to the fitting curve?
5. Figure 3: What is the difference between Woodstove and RWC samples? If they all represent the same experiment, why distinguish them in legend?
6. Figure S5: The author lumped the mass fractions for different BB emissions, then conducted a correlation analysis. If you lumped the source emission, why not directly analyze the correlation for OC/EC among different sources? It would be more insightful if the author could compare mass fractions in different sources based on volatility distribution.

Reference:
Liang, Yutong, Coty N. Jen, Robert J. Weber, Pawel K. Misztal, and Allen H. Goldstein. "Chemical composition of PM 2.5 in October 2017 Northern California wildfire plumes." Atmospheric Chemistry and Physics 21, no. 7 (2021): 5719-5737.
Qian, Zhe, Yingjun Chen, Zeyu Liu, Yong Han, Yishun Zhang, Yanli Feng, Yu Shang et al. "Intermediate volatile organic compound emissions from residential solid fuel combustion based on field measurements in rural China." Environmental Science & Technology 55, no. 9 (2021): 5689-5700.
Shrivastava, Manish K., Eric M. Lipsky, Charles O. Stanier, and Allen L. Robinson. "Modeling semivolatile organic aerosol mass emissions from combustion systems." Environmental science & technology 40, no. 8 (2006): 2671-2677.


 

This text has been copied from the PDF response to reviewers and does not include any figures, images or special characters.

Response to reviewer comments:
Development of Volatility Distributions for Organic Matter in Biomass Burning Emissions Submitted to Environmental Science: Atmospheres

To Whom It May Concern:

We would like to thank the editor and reviewers for their thoughtful and thorough review, and constructive remarks. We have modified the manuscript based on these comments to improve and clarify the text. Please find below detailed responses in blue text (with direct quotes from the revised manuscript shown in blue, indented and italic text) to the comments and suggestions offered by the reviewers (shown in normal text). All line numbers in our responses correspond to the tracked-changes version of the revised manuscript.

Best regards,

The Authors


REVIEWER REPORT(S):
Referee: 1

Comments to the Author
The paper describes a method to derive the VBS of biomass burning aerosols using a thermal desorption method in which particulate matter and higher volatility gas-phase compounds (up to about log10(C*) = 6) are collected using a combination of a quartz filter punch and a sorbent (in this case Carbotrap). The filter punch and the sorbent are placed in the same thermal desorption tube and are desorbed together during the analysis using TD-GC-MS. Deuterated alkane compounds are used to estimate the volatility of other compounds based on their retention time and to estimate their amount in the sample. This data, corrected using thermos-optical OC measurements, is then used to populate the VBS bins. The method was applied to several laboratory and field measurements of biomass burning emissions and provided a reasonable agreement with VBS of biomass burning aerosols measured using other techniques. The presented method has a significant potential to expand our ability to measure VBS of biomass burning aerosols and understand their chemical transformations in the atmosphere. The presented VBS data from several laboratory and ambient experiments is also very useful for the atmospheric chemistry community. I recommend this paper to be published after the authors address relatively minor issues listed below.

Specific comments:

Some aspects of the presented method need to be better described. For example, I assume the filter punch is placed upstream of the sorbent, but this is not mentioned in the text. It is not clear when the tube is spiked with deuterated standards, before or after sampling? Spiking prior to sampling could provide a better measure of recovery that includes any potential losses during sampling. The sorbent used is Carbotrap F and C. What is the ratio of these, were the tubes purchased pre-filled? How was the column bleed determined and used to correct the TIC chromatogram?

We thank the reviewer for their comments and agree that these are important details that should be included in the manuscript. The quartz filter was placed upstream of the sorbent - this is highlighted on line 218 of Section 2.1 in the main paper:

“Sorbent tube samples with quartz fiber filter punches upstream of the sorbent material (“filter-in-tubes”) were analyzed using a thermal desorption gas chromatography mass spectrometer (TD/GC/MS) (Gerstel TDS2, MD, USA) measuring total ion chromatograms (TIC), which we use to indicate the mass of gas- and particle-phase (i.e., total) S/I-VOCs sampled.”


Tubes were spiked with deuterated standard prior to sampling - so our estimates of recovery should include potential losses during sampling. This is now mentioned in Line 214 of the main paper:

“Prior to sampling and subsequent TD/GC/MS analysis, each tube was spiked with a microliter of a deuterated internal standard (obtained from the Wisconsin State Lab of Hygiene) to track analyte recovery. The internal standard solution contained six deuterated n-alkanes ranging from n-Pentadecane-d32 (C15H32) to n-Hexatriacontane-d74 (C36H74).”

Some signals are known to be attributable to column bleed contamination like m/z 73 and m/z 207. The column bleed signal begins to make a significant contribution to the TIC signal for retention times > 26 minutes. This contamination signal needs to be removed so that it doesn’t contribute a positive bias to the VBS measurements as well as bias originating from sampling at longer retention times (lower C*). Here, a correction factor is calculated as the ratio of the contribution of m/z 207 to the total ion chromatogram (TIC) signal in the chromatogram retention time window between 40 and 45 minutes where column bleed dominates the TIC signal. The correction factor is used to determine the column bleed contribution to the TIC and subtract its contribution from the TIC signal. The TIC signal was recalculated as:

f(rt) = max{TICrt - (207rt/correction factor), 0} for rt>26,
TICrt , for rt <=26

where rt represents the retention time. The figures below show the raw chromatogram and the column bleed-corrected TIC using the above process.

This correction is now described in the main paper on line 240.

“First, the TIC of the biomass burning sample was corrected for column bleed by deriving a correction factor, calculated as the ratio of the contribution of m/z 207 to the TIC signal in the chromatogram retention time window between 40 and 45 minutes, where column bleed dominates the signal. The column bleed corrected TIC was then divided into 29 bins in GC retention time space.”


The details of the ratio of Carbotrap F and C are now included on line 207 of the manuscript:

“S/I-VOC sampling media consisted of 40 mm of Carbotrap-F (20/40 mesh) and 20 mm of Carbotrap-C (20/40 mesh) sorbents (purchased already filled in sorbent tubes by Sigma-Aldrich) with previously baked (550° C for 8 hours) quartz filter punches (0.385 cm2) placed in the Gerstel tube upstream of the sorbent beds to capture both particle and gas phase emissions.”

The correction using thermo-optical measurements (Section S1) is rather confusing: were OC measurements artifact corrected or not? "Background" correction could be interpreted as artifact correction or it could be media blank correction. This needs to be clarified. If there was no artifact correction, then an estimate of the artifact should be provided.

In most cases, particulate matter OC measurements were corrected for positive artifact due to gas phase absorption by subtracting results from a quartz-behind teflon (QBT) filter from those from a bare quartz (BQ) filter (Subramanian et al. 2004). Note that the QBT filters were not collected for field samples. Additionally, background filter samples were also collected at all the test sites. The emission factors presented are corrected to account for the background concentrations obtained in these measurements. Line 17 in Section S1 has been reworded to better represent this information.

“Combustion emissions were sampled onto bare quartz (BQ) filters and analyzed using thermo-optical methods to quantify organic carbon content. Primary organic carbon (OC) was estimated for positive artifact due to vapor absorption using a quartz-behind teflon (QBT) filter method.1 The artifact-corrected OC is calculated as the difference between the OC concentration on a front quartz filter and the OC concentration on a parallel quartz filter positioned behind a teflon filter. Note that the QBT filters were not collected for field samples and so estimates from BQ filters were used directly. The artifact percentage (QBT/BQ ✕ 100) is less than 1% for burnhut, 17% for RWC-Startup and 14% for RWC-high fire. The OC concentrations were also background corrected using measurements taken from background filter samples collected at the test sites.”

The authors should discuss the reason for sorbent selection. For example, some compounds could be reactive during sampling when using Carbotrap (Rothweiler et al., 1991). The sorbent selection could also affect the recovery of some of the (heavier and more polar) compounds during sampling and desorption.

The reviewer highlights an important consideration. A comprehensive polar compound analysis would require additional modifications to the sample preparation practices and GC-MS methodology in addition to tube media selection. We focused on the measurement of hydrocarbon SVOCs as they form the basis of our volatility distribution estimation method and thus chose a sorbent that has traditionally measured them well. As the reviewer notes, these sorbents may not be as efficient for collecting all types of polar compounds. However, our standards testing shows us that they were capable of measuring methoxy phenols, which are relatively polar SVOCs. Having said that, the overall efficacy is difficult to estimate since we can only verify using individual compound standards, which are never fully representative of the thousands of compound structures in combustion emissions. Line 210 was added to explain the sorbent selection:

“This sorbent material was chosen as the measurement of hydrocarbons was a focus in our experiments as they formed the basis of our volatility distribution estimation method (described in detail in Section 2.3).”

Were the tubes tested for any breakthrough during sampling? Lighter and more polar compounds could suffer from breakthrough losses. This could provide an explanation for some of the observed discrepancy with the thermo-optical measurements. Potential losses during sampling should be added to the limitations of the method.

We have tested the tubes for breakthrough across several emissions testing environments, including those examining wildfire, residential burning, and vehicle emissions. Breakthrough
testing is normally accomplished by positioning two tubes in series and analyzing both using identical GC-MS methodology. Because naphthalene is amongst the most volatile compounds measured in these VBS test cases, we typically use that compound as our basis for judgment, although C10-C38 hydrocarbon compound data were also collected). At the flow rates, dilution ratios, and concentrations in the present study, we have not observed any compound breakthrough in the second tube. In addition, the recovery of the internal standards spiked before sampling is monitored, where the d8-naphthalene internal standard recovery would indicate breakthrough issues during each sampling event. Such breakthrough has not been observed.

Given the confidence range of the response factor fit (up to a factor of 2), the error bars in Figure 3 should be larger than the variability of the triplicate measurements.

The n-alkane response factors fit by the polynomial in Figure 1B were derived from pooled data across several alkane calibration runs through the TD/GC/MS system. In practice, a blank and deuterated-alkane standard calibration is first performed before analyzing sorbent tubes containing samples from an emissions event. The recovery factors suggested from this ‘run-specific’ calibration were applied to the corresponding sample tubes. This approach thus accounts for the variability observed when looking at response factors averaged across many runs. Stated another way, the mean response factors in Figure 1B were not applied to all the sample tube data and are shown to be representative of overall consistency in response.

Related to the above point, the magnitude of the correction using OC measurements (Section S1) should be mentioned in the text, not only in a figure caption. It is very significant (up to an order of magnitude for the field measurements). This could introduce an additional uncertainty to both the VBS shape and the EF determined using the method.

Line 40 in Section S1 was amended to highlight the divergence in the field measurements:

“The divergence was largest for open field burns (burn-average correction factor of 10.36 – reflecting an order of magnitude divergence) and smallest for residential woodstove emissions (test average correction factor of 2.82)”

Since there is almost an order of magnitude more material in the log10C* = 5 and 6 bins, the amount of material from those bins in the PM could be comparable to that of the log10C* = 4 bin. It would be helpful to see an order of magnitude analysis whether the omission of the higher volatility bins could bias the comparison with May et al.

This figure is now included in the SI as Figure S8 and referenced in the text on Line 457.

“Figure S8: Partitioning plot showing the particle mass fraction (XP) calculated using Equation 1 and volatility distributions shown in Figure 3A (containing bins 5 and 6) indicated by the dashed and dotted lines and Figure S6 (re-normalized after removing bins 5 and 6) indicated by the solid lines vs organic aerosol concentration (COA). The biomass burning sources are delineated by color. Also plotted are observations from May et al.2 The curves including bins 5 and 6 show the same general trends as those using renormalized distributions but are shifted to lower XP across the COA¬¬ range, as these distributions encompass more material. This shift in XP is less pronounced at higher COA values as more of the IVOC material partitions into the particle phase under these conditions.”


Reference

Rothweiler H., Wäger P.A, Schlatter C. (1991) Comparison of Tenax Ta and Carbotrap for sampling and analysis of volatile organic compounds in air, Atmospheric Environment. Part B.
V25, 231-235.


Referee: 2

Comments to the Author
Wildfires have recently occurred more common and intense in the US, making it increasingly important to understand the chemical and physical properties of biomass burning (BB) emissions. This study investigated the volatility of organic emission from different types of BB using the filter-in-tube sorbent tube sampling method. The volatility distribution showed that the IVOCs accounted for a large fraction of total captured organic matter, followed by SVOCs and LVOCs. A phase equilibrium equation was applied to predict gas-to-particle partitioning and emission factors (EFs) from sampling data, and the results were basically consistent with previous experiments. The paper is relatively well organized and the results are of interest to the atmospheric community. However, some aspects need additional discussion, especially for the methods of sample analysis and predicted EFs from equilibrium equation. Major revision is recommended. Detailed comments are listed below.
1. Insufficient information related to sample collection and analysis.
a) Line 181-182: The author used a secondary dilution system to dilute the RWC emissions. However, no information was provided for the dilution ratio. Besides, the author also conducted a chamber experiment to simulate open burning. Why were the samples for this experiment not diluted? More importantly, the emission factors and gas-to-particle partitioning of IVOCs may be related to the dilution process during sample collection as indicated in previous studies (Qian et al., 2021; Shrivastava et al., 2006). The argument would become more compelling if there are some discussions about the influence of dilution processes on the PM emission factors and volatility distributions.

We thank the reviewer for highlighting this gap. All samples had some level of dilution due to excess air in the combustion and exhaust collection systems. Additional sample dilution was used in some cases to be able to sample over the entire test phase and obtain a comparable mass loading in the different scenarios. Additional detail on the sampling strategy was added starting at Line 139:

“Sampling times and dilution ratios for the S/I-VOC samples were designed to obtain approximately the same amount of sample mass and capture the duration of each process, while avoiding breakthrough conditions. Breakthrough testing across several emissions testing environments has been conducted, including those examining wildfire, residential burning, and vehicle emissions. Breakthrough testing is normally accomplished by positioning two tubes in series and analyzing both using identical GC-MS methodology. At the flow rates, dilution ratios, and concentrations in the present study, compound breakthrough in the second tube has not been observed. In addition, the recovery of the internal standards spiked before sampling was monitored, where the d8-naphthalene internal standard recovery would indicate breakthrough issues during each sampling event.”

Information on the dilution system was added to line 192:

“A secondary dilution system (DI-1000, Dekati) was used to further reduce pollutant concentrations so that SVOC samples could be collected over the entire test phase duration and have concentrations comparable to the field and burnhut samples. SVOC and black carbon (MA350, Aehtlabs) sampled from a manifold connected to the secondary dilution system. Emissions were sampled from the exhaust duct with a stainless-steel probe (3/8”) using an eductor (DI-1000, Dekati) supplied with 20 lpm of dilution air dried and scrubbed of CO2 (Van Air Compressed Gas Dryer) and supplied to a stainless-steel (Schedule 40 2” pipe) sampling manifold. The eductor provides a nominal 8:1 dilution ratio; but is dependent on the absolute pressure in the exhaust so the CO2 concentration in the diluted air was measured continuously (LiCor, Li-820) to determine the dilution ratio.’’

We agree with the reviewer that the emission factors and gas-to-particle partitioning of IVOCs may be related to the dilution process. However, measuring partitioning was not the focus of our experiments. We were primarily interested in capturing the total mass included in the gas- and particle-phases and measuring them accurately. Once captured in our filter-tube assembly, the mass was reassigned by volatility using chromatography which simulates distillation. Volatility distributions are then inferred from these chromatography measurements.


b) A TD/GC/MS was applied to measure the TIC of BB samples. It must be emphasized that the combustion of biomass will also emit abundant oxygenated IVOCs (OIVOCs) such as organic acids, oxygenated aromatic hydrocarbons, alcohols, and so on (Liang et al., 2021), while the TD/GC/MS can only partially measure these polar compounds. Besides, OIVOCs with higher polarity are important in further SOA formation through gas-to-particle partitioning. Will the missing measurement affect the results of BB volatility distribution profiles? I noticed in Section 3.1 that the author tried to use authentic standards to obtain the response factors for different classes of compounds. However, there are few polar substances among the authentic standards. Even if the TIC can somehow reflect the OIVOCs, it typically has lower response factors than hydrocarbons. Therefore, I wonder if it is appropriate to convert the entire IVOC TIC signal to mass concentrations using hydrocarbon response factors.

The reviewer raises important concerns about compound polarity and method suitability for polar constituents, some of which are addressed in our response to reviewer one. We agree that polar compounds are likely important to the SOA formation process. IVOCs from biomass burning emissions are challenging to measure in a comprehensive way as they include a wide range of compound classes. With the measurements collected in our study, it is difficult to ascertain if compounds are missing from the TIC signal altogether or if they are present in an unresolved state (in the UCM). The standards used in the present study were meant to be indicative of the potential differences in response when considering oxygenated molecules in the PM from biomass burning. Assuming polar compound recovery factors are lower than hydrocarbons, the amount of mass in the log10C* bins 4-6 would be underestimated by using hydrocarbon response factors. This is now acknowledged on line 558 in Section 4:

“One of the key assumptions of this method is that compounds in the UCM behave like n-alkanes and have similar response factors. Analysis of calibration samples showed that polynomial fits describing response factors for a set of standard calibration compound classes were roughly constrained within the variation between different compounds. The resulting fit was best constrained between C15 and C30 while predicted response factors on either end of the volatility range were more uncertain. However, combustion of biomass can also emit oxygenated IVOCs68 which may not be measured by the TD/GC/MS or remain in an unresolved state. The application of hydrocarbon response factors to these compounds may underestimate the mass in the IVOC bins as polar compounds tend to have lower response factors than hydrocarbons.”

2. The author quantitatively analyzed the LVOCs, SVOCs, and IVOCs in different volatility bins. Is it possible to calculate the EFs for LVOCs, SVOCs, and IVOCs under different BB types in Section 3.2 (Line 329, it should be Section 3.2 but there was a typo?). Moreover, the author only discussed the relationship between PM2.5 EFs with OC/EC EFs. How about the relationship between IVOCs, SVOCs, and LVOCs EFs with OC/EC EFs? Further, the author also used authentic standards for calibration, while no information was provided for speciated organic compounds. It would be interesting if these data can be discussed and will be beneficial for reducing the uncertainty of emission inventory in modelling studies.

The section number has also been fixed in the updated manuscript - we thank the reviewer for bringing this to our attention. To address the reviewer’s question on the relationship between IVOCs, SVOCs and LVOCs EFs with OC/EC EFs and MCE, we assembled the figures below. Our assessment is that showing these fractions do not add information of sufficient value to warrant their inclusion as a cohesive/consistent trend is hard to discern from these figures.






We recognize the value of speciated data in emission inventories, but we feel that their inclusion would require substantial additional length, and so they will be the focus of a future publication.

3. Section 3.4: The author tried to use the volatility distributions to predict the EFs. However, I wonder if this equation holds for any sampling conditions? This uncertainty of this method is also indicated by the large deviations (mostly more than 50%) of the POA EFs predictions at 10 μg/m3. More specifically, if we apply the predicted results to other emission-based models, will this uncertainty be magnified or reduced? More discussion should be presented related to the predicted EFs.

We assume that the deviations that the reviewer is referring to in this comment are those referenced on Lines 482-483 of the manuscript. These deviations in the POA EF are to be expected at different sampling conditions as per the partitioning equation (Eq. 1 in the manuscript). This section shows the value of knowing the volatility distribution of an emissions source. Knowledge of the volatility distributions alongside the sampling conditions are together sufficient in estimating the influence of partitioning on POA EFs. Part of the uncertainty in the predicted EFs would thus arise from the uncertainty in the volatility distributions. Improvement in measurement methods could reduce this uncertainty. In general, the predicted EFs should reduce uncertainty in emission-based models as they would account for partitioning based on the OA concentration using biomass burning specific volatility distributions. This effect has only fairly recently been recognized (Robinson et al. 2010) and this paper provides additional constraint on real-world volatility distributions and, importantly, demonstrates a fairly straightforward method to do so for other sources/source classes.

Minor Comments:
1. Line 194-195: The author used the deuterium substitutes to track the analyte recovery, what is the TD recovery of samples?

The internal standard (IS) solution containing a suite compounds is spiked onto the sample tubes prior to sampling to gauge analytical recovery during TD analysis. The instrument responses for the spiked IS compounds were compared against IS responses from the daily calibration check standard that was analyzed with the samples. The IS responses in the samples were typically within 50% of the responses for the calibration standard.


2. Figure 1B: Why the saturated acids were excluded from response factor analysis?

Figure 1A has saturated acid data from fortuitous data collection for another analysis. We did not perform the response factor analysis experiment with a saturated acid standard. This is highlighted in the Figure 1 caption on Line 290:

“Note: C* vs GC retention time data for saturated acids were collected fortuitously from another calibration exercise; a separate calibration for response factor was not completed and hence these compounds are missing from Figure 1B.”


3. Line 303-305: It seems improper to use the C∗ bins to define SOA production in this study, as the author directly collected the emission samples.

This line has now been reworded as the following on line 327 in the updated manuscript:

“Quantifying emissions in bins up to log10C*=6 can include capturing S/IVOCs, which likely represent a substantial fraction of SOA precursors.”


4. Figure S3: It is difficult to get the information from Figure S3 that the OC EFs are negatively correlated to MCE, and the EC EFs are positively correlated to MCE. Could the author add a dashed line to the fitting curve?

As per the suggestion of the reviewer, these fits have now been included in Figure S3.



5. Figure 3: What is the difference between Woodstove and RWC samples? If they all represent the same experiment, why distinguish them in legend?

We thank the reviewer for bringing this to our attention, as these are the same. The legend in Figure 3 has been updated in the revised manuscript.


6. Figure S5: The author lumped the mass fractions for different BB emissions, then conducted a correlation analysis. If you lumped the source emission, why not directly analyze the correlation for OC/EC among different sources? It would be more insightful if the author could compare mass fractions in different sources based on volatility distribution.

We are not totally clear on what is being suggested by the reviewer in this comment. This figure shows scatter plots of the contributions in the volatility bins comparing different combustion types. One of our objectives was to understand differences between the VBS distributions for different biomass burning sources, and that was the motivation for constructing these plots. These figures do not ‘lump’ emissions from different source types (e.g. RWC and Field burns), but rather compare the relative contributions from the different VBS bins for different sources. For example, Figure S5F shows that the volatility distributions for RWC startup and high power are very similar, suggesting that a common distribution can probably be used to describe partitioning of emissions for both operation conditions. We’re not clear how this should be linked to OC/EC, and as shown above (response to comment 2) there was limited correspondence between emission factors of volatility ranges and OC/EC.

Reference:
Liang, Yutong, Coty N. Jen, Robert J. Weber, Pawel K. Misztal, and Allen H. Goldstein. "Chemical composition of PM 2.5 in October 2017 Northern California wildfire plumes." Atmospheric Chemistry and Physics 21, no. 7 (2021): 5719-5737.
Qian, Zhe, Yingjun Chen, Zeyu Liu, Yong Han, Yishun Zhang, Yanli Feng, Yu Shang et al. "Intermediate volatile organic compound emissions from residential solid fuel combustion based on field measurements in rural China." Environmental Science & Technology 55, no. 9 (2021): 5689-5700.
Shrivastava, Manish K., Eric M. Lipsky, Charles O. Stanier, and Allen L. Robinson. "Modeling semivolatile organic aerosol mass emissions from combustion systems." Environmental science & technology 40, no. 8 (2006): 2671-2677.


References in Response:
Subramanian, R., et al. "Positive and negative artifacts in particulate organic carbon measurements with denuded and undenuded sampler configurations special issue of aerosol science and technology on findings from the fine particulate matter supersites program." Aerosol Science and Technology 38.S1 (2004): 27-48.

A. L. Robinson, A. P. Grieshop, N. M. Donahue and S. W. Hunt, Updating the Conceptual Model for Fine Particle Mass Emissions from Combustion Systems Allen L. Robinson, J. Air Waste Manag. Assoc., 2010, 60, 1204–1222

References in quoted text:
1I.P.C.C., Ed., The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change, Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, 1535 pp, Geneva, Switzerland, 2014.
2A. A. May, E. J. T. Levin, C. J. Hennigan, I. Riipinen, T. Lee, J. L. Collett, J. L. Jimenez, S. M. Kreidenweis and A. L. Robinson, Gas-particle partitioning of primary organic aerosol emissions: 3. Biomass burning, J. Geophys. Res. Atmospheres, 2013, 118, 11327–11338.
68Y. Liang, C. N. Jen, R. J. Weber, P. K. Misztal and A. H. Goldstein, Chemical composition of PM&lt;sub&gt;2.5&lt;/sub&gt; in October 2017 Northern California wildfire plumes, Atmospheric Chem. Phys., 2021, 21, 5719–5737.




Round 2

Revised manuscript submitted on 07 Sep 2022
 

06-Oct-2022

Dear Dr Grieshop:

Manuscript ID: EA-ART-07-2022-000080.R1
TITLE: Development of Volatility Distributions for Organic Matter in Biomass Burning Emissions

Thank you for submitting your revised manuscript to Environmental Science: Atmospheres. I am pleased to accept your manuscript for publication in its current form. I have copied any final comments from the reviewer(s) below.

You will shortly receive a separate email from us requesting you to submit a licence to publish for your article, so that we can proceed with the preparation and publication of your manuscript.

You can highlight your article and the work of your group on the back cover of Environmental Science: Atmospheres. If you are interested in this opportunity please contact the editorial office for more information.

Promote your research, accelerate its impact – find out more about our article promotion services here: https://rsc.li/promoteyourresearch.

We will publicise your paper on our Twitter account @EnvSciRSC – to aid our publicity of your work please fill out this form: https://form.jotform.com/211263048265047

How was your experience with us? Let us know your feedback by completing our short 5 minute survey: https://www.smartsurvey.co.uk/s/RSC-author-satisfaction-energyenvironment/

By publishing your article in Environmental Science: Atmospheres, you are supporting the Royal Society of Chemistry to help the chemical science community make the world a better place.

With best wishes,

Dr Tzung-May Fu
Associate Editor
Environmental Science: Atmospheres
Royal Society of Chemistry


 
Reviewer 2

The authors have well addressed all my questions and comments. I have no further comments.

Reviewer 1

The authors have adequately addressed the reviewer comments.




Transparent peer review

To support increased transparency, we offer authors the option to publish the peer review history alongside their article. Reviewers are anonymous unless they choose to sign their report.

We are currently unable to show comments or responses that were provided as attachments. If the peer review history indicates that attachments are available, or if you find there is review content missing, you can request the full review record from our Publishing customer services team at RSC1@rsc.org.

Find out more about our transparent peer review policy.

Content on this page is licensed under a Creative Commons Attribution 4.0 International license.
Creative Commons BY license