the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 30 Sep 2024
A Comprehensive Characterization of Empirical Parameterizations for OH Exposure in the Aerodyne Potential Aerosol Mass Oxidation Flow Reactor (PAM-OFR)
Abstract. The oxidation flow reactor (OFR) has been widely used to simulate secondary organic aerosol (SOA) formation in laboratory and field studies. The extent of hydroxyl radical (OH) oxidation (or OH exposure, OHexp), normally expressed as the product of OH concentration and residence time in the OFR, is important in assessing the oxidation chemistry in SOA formation. Several models have been developed to quantify the OHexp in OFRs, and empirical equations have been proposed to parameterize OHexp. Practically, the empirical equations and the associated parameters are derived under atmospheric relevant conditions (i.e., external OH reactivity) with limited variations of calibration conditions, such as residence time, water vapor mixing ratio, O3 concentration, etc. Whether the equations or parameters derived under limited sets of calibration conditions can accurately predict the OHexp under dynamically changing experimental conditions with large variations (i.e., extremely high external OH reactivity) in real applications remains uncertain. In this study, we conducted 62 sets of experiments (416 data points) under a wide range of experimental conditions to evaluate the scope of the application of the empirical equations to estimate OHexp. Sensitivity tests were also conducted to obtain a minimum number of data points that is necessary for generating the fitting parameters. We showed that, for the OFR185 mode (185-nm lamps with internal O3 generation), except for external OH reactivity, the parameters obtained within a narrow range of calibration conditions can be extended to estimate the OHexp when the experiments are in wider ranges of conditions. For example, for water vapor mixing ratios, the parameters obtained within a narrow range (0.49–0.99 %) can be extended to estimate the OHexp under the entire range of water vapor mixing ratios (0.49–2.76 %) studied. However, the parameters obtained when the external OH reactivity is below 23 s-1 could not be used to reproduce the OHexp under the entire range of external OH reactivity (4–204 s-1). For the OFR254 mode (254-nm lamps with external O3 generation), all parameters obtained within a narrow range of conditions can be used to estimate OHexp accurately when experimental conditions are extended, but too-low lamp voltages should be avoided. Regardless of OFR185 or OFR254 mode, at least 20–30 data points from SO2 or CO decay with varying conditions are required to fit a set of empirical parameters that can accurately estimate OHexp. Caution should be exercised to use fitted parameters from low external OH reactivity to high ones, for instance, those from direct emissions such as vehicular exhaust and biomass burning.
- Preprint
(1774 KB) - Metadata XML
-
Supplement
(831 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2024-2721', Anonymous Referee #1, 04 Nov 2024
Liu et al. explicitly examined the OH exposure quantification in PAM-OFR by comparing the results of calibration experiments with a limited set of calibration conditions and a wide range of calibration conditions. Recommendations and cautions have been given for both modes of PAM-OFR when calibrating OH exposures in the laboratory. This paper has important implications for air quality and atmospheric chemistry studies considering the increasing number of PAM-OFR users in the community. I recommend its publication after the following minor comments are addressed.
Lines 103-104: Is this statement also true for other OFRs with different UV lamps and different designs of reactors (e.g. wall material, shape, or volume)?
Line 125: Please check if 0.1 ppm is a typo because 0.1 ppm of hydrocarbon seems high.
Line 200: What does a slight change of residence time mean here? 5% variation?
Line 250: Please define FPeOHR, 185.
Also, regarding the discrepancy for OH estimation between low OHR and extended high OHR, would the oxidation of SO2 by H2O2 in nucleated sulfuric acid aerosols contribute to such discrepancy as H2O2 would be formed in the OFR and could further oxidize SO2 in the aqueous sulfuric acid aerosols (Liu et al., 2020)?
The data points for OFR254 mode are more scattering. Are there any recommendations to improve the OH exposure estimation for the OFR 254 mode?
References:
Liu, T., Clegg, S.L. and Abbatt, J.P., 2020. Fast oxidation of sulfur dioxide by hydrogen peroxide in deliquesced aerosol particles. Proceedings of the National Academy of Sciences, 117(3), 1354-1359.
Citation: https://doi.org/10.5194/egusphere-2024-2721-RC1 -
RC2: 'Comment on egusphere-2024-2721', Anonymous Referee #2, 30 Dec 2024
This study assesses the accuracy of empirical equations for estimating hydroxyl radical exposure (OHexp ) in an oxidation flow reactor (OFR) under varied experimental conditions. Through 62 experiments, it was found that parameters derived from narrow calibration ranges, such as water vapor mixing ratios or low external OH reactivity, can be applied in broader scenarios for OFR254 mode but show limitations in OFR185 mode at high external OH reactivity. The authors found that at least 20–30 data points are necessary to derive reliable parameters. The findings highlight the need for caution when extrapolating parameters to conditions beyond their calibration range, particularly for scenarios involving high OH reactivity.
This study is planned well and the scope of having better characterized experiments when using OFRs is important. Furthermore, the manuscript is, for most parts, written well and clearly.
Although I have some comments that should be addressed before this study can be published.Specific comments:
Line 20: I would not suggest to introduce the OHexp, which is a main concept of that paper, in brackets. Change sentence to clarify.
Line 33: Give also RH values (at standard conditions), because a lot other studies use that, just for comparisons.
Line 37: “but too-low lamp voltages should be avoided.” For clarification add why.
Sentence starting at line 63 and general comment for introduction: It is definitely an argument that the PAM-OFR is widely used, but I find you should mentioned some other OFRs and their advantages and disadvantages and why you chose this one (you have some citations at line 53, but I think it is not enough to just cite them there). Almost all other OFRs have also published method paper where at least some of the aspects you discuss in this paper have also already been measured and you should cite them and describe the existing literature better. In line 104 you say other OFRs but without having a paragraph introducing them, it is unclear for the reader why this should be the case.
Line 86: “but this requires specific instruments» specify which instruments.
Paragraph 2.1 (and2.2) A schematic would help to better understand the setup and which instruments were used.
Line 96: Again maybe also mentioned other method (like deuterated butanol etc.) that other people are using.
Line 105 Methods: Have you considered temperature in the OFR? If so please state in the manuscript and if not I would highly suggest to do so. We saw that the temperature inside the OFR (which is not constant especially for 185 and 254 mode due to different power consumption) makes a massive difference in the OH concentration and that this is an important parameter that needs to be taken into account.
Methods: Add a small paragraph with chemicals (purity) and gases that you used.
Line 134 Clarify why “allowed to stabilize”
Line 156 and onward: It is confusing that a-f and a-c have same characters for different equations. I would suggest to use different letters for clarification.
Line 182: SO2 not defined, explain how SO2 used asOHRexp. This also needs to be done in the method part.
Figure 1:(and other figures) add error bars or at least explain why the data points don’t have error bars.
Explain 1:2 and 2:1 line in the caption and in the main text why you used this.
Figure 1: Figure 1 is very packed. It is hard to read the photochemical age in the last row and vice versa for OHexp, dec. Maybe you can think of making this figure a little bit less crowded… Three single figures? Also it would help if you remove the black lines around the data point and make them a little transparent.
Line 244: FP1OHR not defined. In general, I would recommend to overthink your abbreviations. I guess, for you they all make completely sense, but for an outside reader it is very confusing, also because they are very similar. You could make the paper more “reader-friendly” if you improve that. Having more than one information in the subscript is also not recommended. As example write instead of OHexp, est the estimated OHexp. It is a technical journal not one where you have a strict word limit, so you should invested a little bit more in clarity of the technical details.
Figure 2: Explain red “outliers” in (b)
Line 334: Clarify why not suitable
Line 357: Specify a threshold, “too low” is not very scientific.
Line 382: This paragraph contains methods and new results, so please move it up. Also figure 6 should be results. A conclusion should conclude and not show new results if possible.
Figure 6: use scatter plot with no lines, maybe then it is possible to see the data better. Also second Y-axis is different, use the same numbers for better comparison. (same for third Y-axis). Also some of the error bars are outside of the graph. Show all the data or at least justify why you don’t show all the error bars. Explain why some of them are very big, especially in the beginning.
Conclusion Try to give an outlook on how your findings can be used with other systems. This would improve the quality of the manuscript so that also “non PAM-OFR” users could profit from you work.Technical comments:
Sentence starting at line 30 need clarification.
Line 70-76 cite equations.
Line 226: OFR254 is not defined. It is clear from the context, but I would suggest to define it the first time you use it.
Line 250: OHext is not defined.Citation: https://doi.org/10.5194/egusphere-2024-2721-RC2 -
RC3: 'Comment on egusphere-2024-2721', Anonymous Referee #3, 02 Jan 2025
This article evaluates the impact of various factors on estimated OH exposure (OHexp) under the OFR185 and OFR254 operating modes of the PAM-OFR through a series of experiments. It provides valuable guidance for the broader application of OFR and the comparability of their results. However, several critical issues need to be addressed before publication.
OHexp can be divided into offline calibration and online calibration. The method introduced in this paper, which derives estimation equations through a series of univariate controlled experiments, belongs to offline calibration. On the other hand, the estimation of OHexp during field experiments, which involves observing the decay of highly reactive precursors inside the OFR, constitutes online calibration. The starting point of this study is to determine the minimum number of experiments required to obtain accurate offline calibration equations for OHexp. However, several issues need to be addressed:
1. Priority of Online Calibration: If VOC concentrations before and after the OFR can be measured online during field observations, this should be the preferred approach. Offline calibration cannot simulate all real-world conditions, so the applicability of the offline method should be clarified.
2. OHexp Range: Due to its portability, the OFR has significant advantages in field observations and is often used to simulate VOC oxidation and subsequent SOA formation under high OH exposure conditions. Therefore, the calibration experiments should cover the typical OHexp range used in field studies, potentially extending to several days. However, some single-variable experiments in the paper only achieve a maximum OHexp of 1 day. Additionally, readers are likely to be interested in the specific error values under different OHexp conditions, so quantitative results should be provided.
3. Exclusion of Irrelevant Experiments: In field observations, the OFR residence time is generally fixed, while other conditions, such as RH, may vary with the environment. When calculating the minimum number of experiments, were experiments that altered residence time excluded? Under OFR254 conditions, if UV lamp pressure and RH are kept constant while only the O3 concentration entering the reactor is varied, would it still be possible to achieve different OHexp? If so, can the number of experiments be further minimized? This paper should provide guidance on which variables should be set within a wider range and which variables only need to be adjusted within a narrower range, rather than prescribing a specific number of experiments.
4. Variation Ranges of Other Factors: The description of the variation ranges of other factors in the experiments investigating their effects on OHexp estimation is unclear. This information needs to be added.
5. Temperature Effects: Temperature might be an important influencing factor. Field experiments may occur under different locations and seasonal conditions, resulting in significant variation in sample temperature. At the same time, UV lamp pressure directly affects the internal OFR temperature. The reaction rates used in the paper are treated as constants—could this cause significant deviations? Please evaluate this potential impact.
6. Applicability Across Different OFR Designs: Is it feasible to apply the described minimum number of experiments to different types of OFRs? As far as I know, the designs of OFRs can vary significantly. For instance, cone-shaped inlets and premixing inlets can affect residence time distributions within the OFR. Can the OHexp estimation equations be applied universally to these designs? Detailed descriptions are required to clarify this point.
By addressing these questions, the study can provide a more robust and universally applicable framework for offline OHexp calibration in different experimental and field scenarios.
Specific comments:
L110-L112: In the PAM-OFR system, there are four ultraviolet lamps. Two of these lamps generate ozone, while the other two do not? Regardless of whether the system is operating in OFR185 or OFR254 mode, only two of these lamps are turned on at a time?L114: "quartz tubes" here is the sleeves in L112？
L125: "0.1ppm" here should be 1ppb, ref: http://www.bjkwnt.com/productinfo/803339.html
L147：In this paper, are the reaction rate constants of SO2 and CO with OH consistently using a constant value? As can be seen from Figure S2, the temperature inside the reactor varies significantly under different lamp pressures: 23-26°C for the OFR185 mode and 33-36°C for the OFR254 mode. The impact of temperature needs to be assessed, especially for the OFR254 mode.
It is crucial to note that while equation 3 captures the combined effect of light intensity and RH through rO3, experiments still need to be conducted at different humidity levels. Could you please inform the readers of the range of RH used in this study?
L161: equation 3: the sign after parameter a should be minus accoding to (Peng et al., 2015).
L186-200: This paragraph describes the effect of different residence times on the estimation of OHexp, finding that the equation fitting parameters at a residence time of 33s can be applied to longer residence times. However, as seen in Figure 1-a2, estimated OHexp at longer residence times (red points) seems to be higher than the shorter residence times (green points), meaning it deviates further from the 1:1 line. Although this is not obvious in logarithmic coordinates, a quantitative assessment of the error range, as described in L200, should be provided here. Subsequent descriptions on other factors should also include relevant quantitative statements, such as the error magnitude at specific OHexp values.
Regarding the study on residence time, the range of OHexp is within one day, but the highest OHexp in actual field observations should be at least greater than 3 days. What is the difference between the experiment in Figure S3 and that in Figure 1, and why aren't the data combined?
Furthermore, this paragraph only describes the range of residence times and does not detail the values of other experimental variables. Similar descriptions should be included in the subsequent paragraph about other factors.
L267-268: Here, the equations obtained by refitting the data for different OHRext ranges show good adaptability. If your experimental conditions are OHRext > 23 s^-1 but less than 198 s^-1, it is uncertain whether the fit will still be good, as the sensitivity of those parameters to OHRext does not appear to be linear.
L369，It is mentioned here that the parameters for OFR254 is not as good, but the abstract (L35-37) states that it is not bad. Please maintain consistent descriptions throughout the document.
Citation: https://doi.org/10.5194/egusphere-2024-2721-RC3 -
RC4: 'Comment on egusphere-2024-2721', Anonymous Referee #4, 14 Jan 2025
Comments on “A Comprehensive Characterization of Empirical Parameterizations for OH Exposure in the Aerodyne Potential Aerosol Mass Oxidation Flow Reactor (PAM-OFR)” by Liu et al.
Liu et al. re-calibrated the OH exposure (OHexp) Empirical Parameterizations with the experiment conditions (OH reactivity (OHR), RH, residence time, O3) range in large scales for OFR185 and OFR254 modes. They found the OHexp from the fitted parameters were less affected by the residence time, water vapor mixing ratio, and O3 concentration. When the OHR was much higher (e.g., more than 200 s-1), it would overestimate the OHexp if the parameters were obtained with the data from narrow experiment conditions (e.g., OHR less than 30 s-1) for OFR185 mode. They also pointed out that 20-30 data points were needed to obtain reliable parameters. Their results provide a reference for the estimation of OHexp and parameter fitting of OFR in different experiments, such as field observations or source emission experiments.
After reading through the manuscript, I was confused about the aim and novelty of this manuscript:
1） Since Rowe et al. 2020 have already published the parameterization results for the lamps produced by light source inc. Why did the authors demonstrate another set of parameters? Are they better compared with Rowe et al. 2020? The comparison with previous study shall be shown here.
2） The KinSim model can reproduce all the chemistry in the OFR. Do the experiment results done here agree well with the model results? Especially for the high OHR. The comparison of OH exp between the KinSim model and experiments is necessary. especially for the extreme high OHR cases. The parameterization in Rowe et al.2020 was fitted based on the KinSIM model output. The model output was constrained by experiment results.
3） Line 144, for the species with low kOH e.g., SO2 and CO, the OHexp calculated from plug low might agree with these from residence time. However, the residence time shows substantial influences on the species with high kOH. E.g., MT Isoprene (Palm et al., 2018). For these species, I do not think the conclusion shown here will be valid. The same conclusion is also applicable to the section which discussed the residence time influences to the OH exposure.
4) Line 110-111The lamps applied in this study are covered or not?
5) Line 202-204, RH shall also be shown.
6) Line 90-92: “Yet, it is unclear whether the fitted parameters obtained under certain conditions can still accurately estimate OHexp when experimental conditions, such as UV light intensity, water vapor mixing ratio, residence time, and external OH reactivity (OHRext), undergo significant changes.” I think at least the UV light intensity and water vapor mixing ratio should have been considered in most published papers.
7) line 110: was Š is
8) line 254-266: The author fitted the parameters based on the equation proposed by Li et al., (2015) and explained the deviation of OHexp under high OHR using the equation. They found the sum of parameters related to OHR first increased and then decreased with the increase of OHR (figure S4a3). The sum of the parameters related to OHR shows a monotonical decreasing trend when using the FPeOHR, 185. As the OHR from SO2 only reaches 204 s-1, will the curve in Figure S4b3 increase at higher OHR, as the one shown in Figure S4a3 (or what is the range of OHR applicable to FPeOHR, 185)?
9) The fitting parameters used in Figures 3 and 5 are not clearly stated.
References
Palm, B. B., de Sá, S. S., Day, D. A., Campuzano-Jost, P., Hu, W., Seco, R., Sjostedt, S. J., Park, J. H., Guenther, A. B., Kim, S., Brito, J., Wurm, F., Artaxo, P., Thalman, R., Wang, J., Yee, L. D., Wernis, R., Isaacman-VanWertz, G., Goldstein, A. H., Liu, Y., Springston, S. R., Souza, R., Newburn, M. K., Alexander, M. L., Martin, S. T., and Jimenez, J. L.: Secondary organic aerosol formation from ambient air in an oxidation flow reactor in central Amazonia, Atmos. Chem. Phys., 18, 467-493, 10.5194/acp-18-467-2018, 2018.
Citation: https://doi.org/10.5194/egusphere-2024-2721-RC4
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 307 | 52 | 22 | 381 | 22 | 3 | 6 |
- HTML: 307
- PDF: 52
- XML: 22
- Total: 381
- Supplement: 22
- BibTeX: 3
- EndNote: 6
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1