Review of Hakkinen et al., “Online measurement of highly oxygenated compounds from organic aerosol”

Summary:

The authors characterize a new Vocus Inlet for Aerosols (VIA) inlet, which thermally evaporates particles for online sampling. Coupled with an NO3-CIMS, they demonstrate the use of the VIA for measuring the chemical composition of several different aerosol types. These measurements show that the VIA could be useful for making online speciated particle composition measurements. The authors also discuss several artifacts and possible unknowns that need to be further investigated before the VIA can really be trusted to provide accurate speciation of highly oxygenated organic compounds. It would be best if the authors could provide direct comparisons of speciated VIA-NO3CIMS measurements with one or more of the other particle phase measurements discussed in the introduction (see my last comment below), which could give the reader a better idea of how well this method works in comparison to other methods, but that may not have been feasible. After addressing my comments below, I recommend for publication.

Comments:

Line 25 and throughout: Why have you chosen to use the term “highly oxygenated compounds” throughout the manuscript instead of HOMs? Is there a technical difference between highly oxygenated compounds and HOMs that you can give in the text to clarify this, or can you instead refer to them as HOMs everywhere?

Line 32: I’m not sure what you mean by “form larger components of SOA”. I think you mean form larger molecules through oligomerization, but this could also mean they just accumulate until they comprise a large component of the SOA by mass. Please clarify.

Line 136: This paragraph is a bit confusing. First you say the internal temperature is not known, then you say you measured it and it internally reaches the setpoint by the outlet, then you go back to using the external temps and saying you use the setpoints and they aren’t necessarily the same as actual internal temps. I think it would be best if you show/state the internal measurements, but it would also be fine to just replace this paragraph with a simple sentence saying the temps are controlled externally so the internal temps may vary slightly.

Line 160: By “decreasing trend”, do you mean the fact that the linear regression slopes are less than 1? This language could probably be clarified. But my main question here is what role could wall losses of H2SO4 or highly oxidized organics to the tubing between the VIA and the NO3-CIMS be playing? Could that be causing this effect at higher concentrations?

Line 163: How does this calibration value of 2x10^10 cm^-3 compare to calibrations previously determined for NO3-CIMS instruments? Is it perhaps lower due to sampling line losses between the VIA and the NO3-CIMS in this setup? Please comment on this.

Line 173: Can you apply your calibration factor to the organic compounds here to give a rough estimate of mass concentration measured by the NO3-CIMS?

Line 174: The detection limit will depend on the dilution amount relative to sample flow, right? Since you’re diluting here by a factor of 4 (2 lpm sample flow and 8 lpm dilution flow), does that mean the limit of detection should be stated as 0.25 ug m-3 mass concentration at the NO3-CIMS inlet?

Line 222: Does this sentence refer to the data in Fig. 9 and the discussion in Sect. 4.1.1? If so, I suggest removing this statement from here. You are already discussing this concept of comparing the oxygen number between gas and particle measurements in 4.1.1. If you are actually referring to something else in this sentence, please provide a citation or show data to back up this assertion.

Line 243: Is there any literature on how fast these particle phase reactions could be at room temperature? That could help inform whether these particle phase reactions are occurring before the particles enter the VIA, or if the heat in the VIA is the real cause of particle phase changes. If the latter, that’s not so great for trying to identify specific compounds in the particle phase.

General comment: I’m wondering if you have tried sampling the gas phase (without seed aerosol) through the VIA with heating, to compare with normal gas phase sampling without heating. Does the heat lead to gas phase reactions?

Line 257: Can you compare this 10-20% number with previous literature or expectations?

Line 271: Could the thermal decomposition be not just breaking C=C bonds, but loss of water, CO, CO2, etc leaving you with less oxygenated species that are no longer clustering with NO3-? I’d suggest adding a sentence discussing this here.

Another general comment: In the introduction, you summarized well the existing particle phase speciation techniques. I am wondering if there are opportunities to directly compare measurements taken with the VIA and NO3-CIMS to some of these other techniques. For example, are there any published spectra of e.g. alpha pinene ozonolysis SOA that measure similar compounds, that you could show a direct comparison with? The ionization techniques might be different, but it would be interesting to see such a comparison. Sampling the same aerosol with multiple techniques would of course be the best way to directly compare, but may not be feasible depending on the authors’ access to the other techniques.

Review of Hakkinen et al., “Online measurement of highly oxygenated compounds from organic aerosol”

In this manuscript rather new technique - Vocus Inlet for Aerosols (VIA) is briefly described and coupled with NO3-CIMS applied for measurement and characterization of organic aerosol. VIA is traditionally coupled with Vocus PTR, therefore this study represents an important extension of VIA’s usability in atmospheric science by showing successful real time measurement of complex composition of secondary aerosol with impressive detection limits. Authors have put significant thought into the experimental set up, background corrections and themselves highlight the pros and cons of such technique for reliable aerosol measurement. It would be interesting to see such measurement in comparison to techniques targeting similar goal such as EESI or Figaero. As first attempt to make aerosol measurements with this coupling I find the study well done and recommend for publication after addressing minor comments.

135: The temperature profile inside the unit was not uniform’’  - Could authors provide the numbers of these measurements?  I assume 300C is the maximum temperature VIA can reach and at recommended flow not sufficient for total evaporation as authors state. Would not decreased flow through VIA lead to more efficient evaporation? Did authors try that? Or would this be compensated through losses and decreased particle transmission. The particle transmission was measured at 1 l/min however 1.5 l/min or 2 l/min were used throughout the experiment without clarification why. Again, could authors provide results of the size dependance measurement.

160: The way the calibration is presented is a bit confusing. Shall the reader understand that three time-separate calibrations were made corresponding to presented calibration factors of 1- 3 x10^10 cm^-3? It is not clear that these calibration factors originate from separate measurements. How do these calibration factors compare to literature? How do authors explain the non linear step between 0.8 – 1 ng/m3 for all the curves? Could losses explain the discrepancy in the calibrations? Did authors calibrate higher than 1 ng/m3? Please clarify.

165: For readers not having experience with Nitrate Cims, please explain the unit for the calibration factor.

170: What role does the gas-particle equilibrium play? This equilibrium is distorted 1. Via the denuder and 2. via the dilution with clean air on potentially still present particles. Would this somehow effect your results? Or is this effect negligible?

175: In this case is the author talking about background (filter) NO3-CIMS corrected signal?

190: With detection limits as low as 1 ug/m3 why did authors choose to investigate SOA at such high loadings?

215: There has been studies showing that the prevalence of oligomers is linked to total SOA mass. I encourage authors to consider this option and provide references.

250: What does prevent the author to use calibration factor for sucrose from direct sucrose calibration? This would be technically possible the same way as for ammonium sulfate and could diminish the discrepancy in the figure at high temperature