The authors report on a new dissociation cavity enhanced absorption spectrometer for quantification of NO₂, RO₂NO₂ and RONO₂ in the atmosphere. The instrument relies on cavity-enhanced absorption spectroscopy to quantify NO₂ and NO₂ generated from organic nitrates by sampling through a heated inlet similar to what has been described by others (Thieser et al., 2016; Paul et al., 2009; Keehan et al., 2020; Chen et al., 2017; Wooldridge et al., 2010; Sadanaga et al., 2016; Di Carlo et al., 2013). Inlet characterization and sample field data are presented.

Overall, this is a well written manuscript suitable for publication for AMT after my comments below have been addressed by the authors.

Major comments

- Considering the large body of existing TD literature, a table comparing this new instrument to existing methods and a discussion of the differences, advantages and disadvantages should be added to the paper.

Specific comments

Title - replace 'detecting' with 'measuring' or 'quantification of' (the instrument does not merely detect the presence of PN and AN after all).

line 102. Since the instrument samples through a PTFE filter, the Mie scattering component should be zero.

line 110. Please comment on the precision of the output power of the stabilized source.

line 119. reported by whom? ATF?

line 155 please state the manufacturer and internal surface of the T-shaped solenoid valves.

line 160. Hg lamps tend to "run hot" which can affect the output of PAN, which is prone to thermal dissociation. Has the composition and purity of the PAN source been evaluated? Furthermore, what was the concentration or mixing ratio of acetone used?

line 166 which requirements?

line 182. Please describe how N₂ and He were delivered (sampled from the tip of the inlet, or statically).

line 226. Please justify omitting glyoxal from the fit and estimate the uncertainty introduced. Figure 4 suggests that the contribution of glyoxal was small but that may not always be the case. What (if anything) happens to the NO₂ retrieval when glyoxal is included in the fit?

line 236 - Figure 4. Please state the uncertainty of the NO₂ mixing ratios (16.2±? ppbv).

line 240 - Figure 5. It seems that the timing of the inlet temperature switch was off as there are blue data points at the same level as [NO₂]_ambient and red data points at the same level as 180 °C.

lien 248 It is 1 spectrum, but the plural should be 2 spectra (not "spectrums").

line 283. An alternative (and more likely) interpretation of the second plateau is the presence of alkyl nitrate impurity. What are the operating temperature and the output purity of the photochemical PAN source? Is [PAN]_out = [NO_x]_in ?

lines 283-284. The PAN dissociation temperature of 400 °C reported by Friedrich et al. is an outlier and inconsistent with every other paper on this subject. At a residence time of 142 ms (line 153) and using rate constant for unimolecular decomposition reported by (Kabir et al., 2014), PAN is predicted to be 99% dissociated at a temperature of "only" 127 °C.

line 286 - how much time is there for recombination to occur?

line 290. CH₃O₂NO₂ more readily dissociates than PAN; under the conditions of the authors' inlet, it is predicted to be >99% dissociated at a temperature of <50 °C.

line 291. The two plateaus can also be interpreted as a ~2:1 mixture of PAN and alkyl nitrates - can this be ruled out (see question for lines 160 and 283).

line 294. Please state how much time there is for recombination to occur.

line 318 Caption to Figure 7 - are gray/green and PAN/NO₂ backwards?

line 326 Filter and wall losses are small only if they the filter and wall material are made from Teflon. Please rephrase.

The statement is true for ANs such as methyl or ethyl nitrate; not sure the statement is true for isoprene nitrates that are prone to hydrolysis (Vasquez et al., 2020).

line 426 Replace MeN with Methyl nitrate (one is not supposed to start a sentence with an abbreviation or acronym).

line 448. It is worthwhile noting that these molecules are important only at night and early morning hours - for this reason, the Cohen group has generally not reported AN data at those times of day.

Please cite (Thaler et al., 2011) for ClNO₂ and (Womack et al., 2017) for N₂O₅.

line 462/465. It is 1 spectrum, and 2 spectra (not "spectrums").

line 465 "as shown in Fig. 9a. The 21077 spectrums" Figure 9a does not show this information (11a perhaps?). What is meant by 21077?

line 467 Fig. 9b should be 11b.

line 500 "up to 0.99" - please state the actual value of r

line 514 Figure 13. I am not sure what is plotted here. The GC-ECD data are labeled "PAN" on the left-hand side, but PNs on the right-hand side. Tthe TD-CEAS data are labeled PNs on the left but PAN on the right.

In principle, the GC-ECD can observe PAN, PPN etc. and those can be summed to ΣPN. Was this done?

Literature cited

Chen, J., Wu, H., Liu, A. W., Hu, S. M., and Zhang, J.: Field Measurement of NO₂ and RNO₂ by Two-Channel Thermal Dissociation Cavity Ring Down Spectrometer, Chin. J. Chem. Phys., 30, 493-498, 10.1063/1674-0068/30/cjcp1705084, 2017.

Di Carlo, P., Aruffo, E., Busilacchio, M., Giammaria, F., Dari-Salisburgo, C., Biancofiore, F., Visconti, G., Lee, J., Moller, S., Reeves, C. E., Bauguitte, S., Forster, G., Jones, R. L., and Ouyang, B.: Aircraft based four-channel thermal dissociation laser induced fluorescence instrument for simultaneous measurements of NO₂, total peroxy nitrate, total alkyl nitrate, and HNO₃, Atmospheric Measurement Techniques, 6, 971-980, 10.5194/amt-6-971-2013, 2013.

Kabir, M., Jagiella, S., and Zabel, F.: Thermal Stability of n-Acyl Peroxynitrates, Internat. J. Chem. Kin., 46, 462-469, 10.1002/kin.20862, 2014.

Keehan, N. I., Brownwood, B., Marsavin, A., Day, D. A., and Fry, J. L.: A thermal-dissociation–cavity ring-down spectrometer (TD-CRDS) for the detection of organic nitrates in gas and particle phases, Atmos. Meas. Tech., 13, 6255-6269, 10.5194/amt-13-6255-2020, 2020.

Paul, D., Furgeson, A., and Osthoff, H. D.: Measurements of total peroxy and alkyl nitrate abundances in laboratory-generated gas samples by thermal dissociation cavity ring-down spectroscopy, Rev. Sci. Instrum., 80, 114101, 10.1063/1.3258204 2009.

Sadanaga, Y., Takaji, R., Ishiyama, A., Nakajima, K., Matsuki, A., and Bandow, H.: Thermal dissociation cavity attenuated phase shift spectroscopy for continuous measurement of total peroxy and organic nitrates in the clean atmosphere, Rev. Sci. Instrum., 87, 074102, 10.1063/1.4958167, 2016.

Thaler, R. D., Mielke, L. H., and Osthoff, H. D.: Quantification of Nitryl Chloride at Part Per Trillion Mixing Ratios by Thermal Dissociation Cavity Ring-Down Spectroscopy, Anal. Chem., 83, 2761-2766, 10.1021/ac200055z, 2011.

Thieser, J., Schuster, G., Schuladen, J., Phillips, G. J., Reiffs, A., Parchatka, U., Pöhler, D., Lelieveld, J., and Crowley, J. N.: A two-channel thermal dissociation cavity ring-down spectrometer for the detection of ambient NO₂, RO₂NO₂ and RONO₂, Atmos. Meas. Tech., 9, 553-576, 10.5194/amt-9-553-2016, 2016.

Vasquez, K. T., Crounse, J. D., Schulze, B. C., Bates, K. H., Teng, A. P., Xu, L., Allen, H. M., and Wennberg, P. O.: Rapid hydrolysis of tertiary isoprene nitrate efficiently removes NOx from the atmosphere, Proc. Natl. Acad. Sci. U.S.A., 117, 33011-33016, 10.1073/pnas.2017442117, 2020.

Womack, C. C., Neuman, J. A., Veres, P. R., Eilerman, S. J., Brock, C. A., Decker, Z. C. J., Zarzana, K. J., Dube, W. P., Wild, R. J., Wooldridge, P. J., Cohen, R. C., and Brown, S. S.: Evaluation of the accuracy of thermal dissociation CRDS and LIF techniques for atmospheric measurement of reactive nitrogen species, Atmospheric Measurement Techniques, 10, 1911-1926, 10.5194/amt-10-1911-2017, 2017.

Wooldridge, P. J., Perring, A. E., Bertram, T. H., Flocke, F. M., Roberts, J. M., Singh, H. B., Huey, L. G., Thornton, J. A., Wolfe, G. M., Murphy, J. G., Fry, J. L., Rollins, A. W., LaFranchi, B. W., and Cohen, R. C.: Total Peroxy Nitrates (SPNs) in the atmosphere: the Thermal Dissociation-Laser Induced Fluorescence (TD-LIF) technique and comparisons to speciated PAN measurements, Atmos. Meas. Tech., 3, 593-607, 10.5194/amt-3-593-2010, 2010.

Summary: The authors present a new thermal dissociation cavity enhanced absorption spectrometer (TD-CEAS) for measurement of NO₂, peroxy nitrates (PNs), and alkyl nitrates (ANs). They demonstrate, through lab tests and box model simulations, that interferences can be corrected for, and that the instrument outputs accurate measurements when compared to a chemiluminescence detector. Finally, results from a measurement campaign in Chengdu, China are presented, demonstrating its effectiveness in ambient conditions.

In general, this manuscript successfully demonstrates the performance of this new instrument, carefully considering the difficulties in converting PNs and ANs to NO₂ in a thermal dissociation oven. There are some significant grammatical/English errors throughout the manuscript, which in some cases make the details difficult to understand but these can be fixed. I would recommend publication, after the authors address some comments below, as well as editing the English.

General comments:

The authors sometimes refer to the ANs channel, and sometimes to ONs channel, which is confusing. This should be made clear throughout the manuscript that these are different, but related to each other.

Section 3.4 demonstrates that the 180 degree oven is not sufficiently hot enough to prevent recombination of PA and NO2, with efficiencies ranging from 0.5 to 0.9 at 180 degrees. Are the authors correcting for this incomplete thermal dissociation in the rest of the paper? If so, this should be stated clearly. If not, this seems like a major inaccuracy of the measurement and should be addressed. Figure 9 makes it look like it isn’t being made, since the intercept at x = 0 doesn’t match what the legend says the input PAN concentration is.

The authors should more clearly demonstrate how they convert the measured α(λ) to [NO2]. Perhaps another equation would be helpful here in section 3.3, demonstrating that it is a linear fit of all the possible gas-phase absorbers in that wavelength region.

As the authors state on line 533, sudden changes in the ambient NO2 while this instrument is measuring from the PNs and ANs would pose a significant problem. Probably this instrument is only useful when a simultaneous measurement of NO2 is available. Most field campaigns do have NO2 measurements, so this likely isn’t a major issue, but the authors should address it anyway.

Many of the references have titles listed in all capital letters, which should be changed.

Specific comments:

Line 32: “One is peroxy acyl nitrates (PANs)…”. The other one is never defined. Is it the peroxy nitrates without an acyl group, as mentioned in line 34?

Line 33: Define PAN here, to differentiate from the more general PANs.

Line 46: “with a small branch ratio (1% - 30%)”: Which reaction (R3a or R3b) is defined as the one with the 1 – 30% branching ratio, and which is the 70 – 99% reaction?

Line 75: “the importance of PNs and ANs in regulating ozone formation has not been well studied [in China]”: The absence of citations here implies it has not been studied at all, which is not true. Examples include: Liu 2010, Zhang 2014, and Liu 2018. Some citations should be included here.

Line 92: Fig S1 only shows the wavelength range 430 – 460 nm, so this line should be changed to match.

Line 151: “… for the ANs and PNs channels are controlled at 180 degrees and 380 degrees, respectively”. These numbers appear to be backwards, as the ANs channel was at 380, not 180 degrees.

Line 154: Presumably the solenoid valves are made of stainless steel? Do the authors expect any NO2 losses on this steel?

Line 175: Define the MCM, and include a citation.

Line 229: “The corresponding fitting residual is in the range of 10 x 10^-9, suggesting the system can guarantee the accuracy…”. What is the meaning of this number and why does it imply the system’s accuracy? Wouldn’t it be better to compare the residuals between the two different fits to demonstrate they are similar in their magnitude?

Line 247: Move the “CONC” label to after “One is the differential concentration method” on line 245.

Line 245 – 255: In general, this paragraph is more confusing than it needs to be. You can simply state that there are two methods, one which calculates [NO2] in each channel from equation (1) using N2 as I0, then subtracts [NO2]_ambient to yield [ANs] and [PNs]. The other method uses I_ambient as I0 to first derive a corrected α(λ), and then uses this to calculate [PNs] and [ANs]. I do not think that equations 3 – 8 are necessary.

Line 254 and elsewhere: The “SPEC” method is often misspelled as “SEPC”.

Line 271 – 274: This is helpful information about why two different oven setpoints will yield PNs and ANs separately. It should be moved to earlier in the manuscript, perhaps in the introduction.

Lines 282: “platform” should be replaced with “plateau”.

Line 345: These interferences of a few percent, while not large, are still non-negligible. Are the measurements being corrected for these interferences? If so, that should be stated clearly.

Line 366: “as described above” should be “as described below”

Line 367: To stay consistent with previous sentence, replace “RO2” with “PA”

Lines 426 – 441 and equations 9 – 12: This is another example of a paragraph that is much more confusing than it needs to be. It seems that you could just say that to accurately measure ANs, you must first measure PNs in the 180 degree channel, apply a corrective factor based on the first look-up table, then subtract this from the raw ANs channel, then apply a second corrective factor based on the second look-up table. The way the authors have written it, with many new parameters such as [PNs_C] is just more confusing.

Line 472: “… the interference in the heated channels, which should be larger than 8%”. Where does this number come from?

Line 484 – 487: This is repeating how the corrections are made, and was already stated in the previous section, so it doesn’t need to be repeated here. Doing so implies that the technique is different here.

Line 535 – 538: These lines are introducing new information to the analysis, and should be included in the results and discussion section instead of the conclusions section.

Figure 2: Zoom in on the left-hand axis which shows reflectivity. It is difficult to see the full range of R.

Figure 3: Why do the authors expect the d_eff / L vs flow rate plot to be linear? A linear fit implies that at the intercept, where flow rate = 0, then d_eff / L will be 0.79, when in fact, d_eff / L should approach 0 as the flow rate decreases to 0. On the other end, as the flow rate gets larger, the d_eff / L will get larger, but will never get to 1 or higher, as a linear fit would imply. It seems that an exponential fit (d_eff / L = A – Be^(C*flow_rate)) would be more appropriate.

Figure 7: The caption states that the orange columns correspond to HNO3, but the legend indicates CH3O2NO2. Which is correct?

Figure S5: How was this simulated? Was it checked experimentally? How do the authors reconcile this non-uniform temperature profile with their statement on line 152 that “it is assumed that the temperature of the heating part is uniform”?

Figure 9: The y-axis label is confusing. Doesn’t using the SPEC method mean that the resulting [NO2] is simply [PN], without needing to subtract [NO2]_ref?

References

1. Liu et al, Evidence of reactive aromatics as a major source of peroxy acetyl nitrate over China, ES&T 44, 7017 (2010)
2. Zhang et al, Wintertime peroxyactyl nitrate (PAN) in the megacity of Beijing: Role of photochemical and meteorological processes, J. Environ. Sci, 26, 83 (2014)
3. Liu et al, Understanding unusually high levels of peroxyacetyl nitrate (PAN) in winter in Urban Jinan, China, J. Environ. Sci. 71, 249 (2018)

This paper describes a newly developed measurement system of NO₂, PNs and ANs in the atmosphere based on a thermal dissociation cavity enhanced absorption spectroscopy method (TD-CEAS). The authors evaluate characterization of this instrument and confirm the performance in field observations.

In an NO₂, PNs and ANs measurement system based on TD followed by NO₂ analyzer, NOx in the atmosphere interfere measured values of PNs and ANs. In this paper, in-depth evaluations for the interference were performed. As a result, TD-CEAS can measure ambient PNs and ANs concentrations using precise “correction factors”. I recommend the manuscript to be published in AMT. However, I found several concerns to be published in the present form, so the authors should perform appropriate revisions sufficiently.

NO₂ detection: The use of a CEAS method would be novel. But advantages of the use of a CEAS are unclear, so the authors should state the advantages of a CEAS. What are the advantages of CEAS over LIF, CRDS and CAPS?

Interference: The authors performed in-depth evaluations for the interference of NOx with PAN. But I could not find the evaluations for the interference with ANs. The authors should state the evaluations for the interference with ANs as well as PNs.

Other minor and technical comments:

lines 26-27, “alkyl nitrates (ANs, RONO₂)”: There are many kinds of RONO₂ other than “alkyl” nitrates.

Line 64, “and cavity enhanced spectroscopy”: Did the authors forget to delete?

Line 175: The authors should define MCM. (Master Chemical Mechanism?)

First paragraph on page 10: The authors explain that the reason for the insufficient decomposition efficiency of PAN at 180 ℃ is due to the recombination of PAN. I think the effect of the PAN recombination can be reduced by increasing the pyrolysis time. What is the reason for not doing that (and making corrections)?

Figure 7: Which is correct, CH3O2NO2 in the legend or HNO3 in the caption?

Line 465: Fig. 9a → Fig. 11a

Line 467: Fig. 9b → Fig. 11b