The manuscript “Characterization of the Vaporization Inlet for Aerosols (VIA) for Online Measurements of Particulate Highly Oxygenated Organic Molecules (HOMs)” report a systematic test of VIA used with NO3-CIMS to detect HOM, including the transmission efficiency, evaporation efficiency, quantification of particle-phase HOM as well as applicability for volatility measurement.

The authors found that transmission efficiency of particles (NaCl>50 nm) is >90%. Transmission efficiency of VOC was also high. Also the transmission loss for sulfuric acid vapors was negligible according to the evaporated AS particles measured by SMPS and sulfuric acid measured by NO3-CIMS. Adding a sheath flow after VIA reduced markedly the wall loss of HOM. The signal of HOM increased with T first and then decreased, indicating the loss of HOM in VIA. Tmax correlated with Tmax obtained from FIGAERO-I-CIMS, but much higher (~100-150 ºC) than Tmax from FIGAERO. The loss efficiency of HOM obtained by a one-dimensional model was high (3-9) and correction factor depended on molecular weight.

Determination of particle-phase organic components on-line and on molecular level is critical to understand the formation, fate and impacts of organic aerosol. In this regard, this study presents a valuable attempt to evaluate and to optimize VIA combined with NO3-CIMS to be used for HOM measurement, although there is a number of limitations and challenges to use VIA for the quantification of particle-phase HOM. This manuscript is generally well-written. I have a few comments for the authors to consider before its publication in AMT.

1. In this study, it was assumed that the loss of HOM was due to the collision with hot walls. What is the evidence for this assumption? It is possible that it was due to the decomposition in the air within VIA, which was not included in the model of this study as mentioned by the authors?
2. Another related question. How was the uncertainty in Fig. 9 derived? I suggest the authors to further discuss the uncertainty/limitations of correction factor, e.g. how the factors not considered in the model influence CF, as it is key to the quantification of particle-phase HOM.
3. Moreover, how applicable is the correction factor for one compound (molecular formula)? For example, it one does not ramp up temperature, can the correction factor be used (considering that ramping up temperature largely limits the time resolution of the method)? Or it has to be used with a thermogram? Does the correction factor depend on functional groups other than molecular weight as shown in Fig. 8b?
4. I would suggest the authors to briefly discuss the advantages and disadvantages in Sect. 3.4 compared with other techniques mentioned in the introduction part.
5. 2b, in the legend, is “140 ºC” the set temperature?
6. 7b, is the normalize frequency of ΔT obtained from each molecular formula? Can the difference in chemical composition at different aerosol loading influence the distribution of the frequency?
7. L433, Ren et al 2022 could be mentioned here.
8. L511, what does the “correlation coefficient” denote?

Reference

Ren, S., Yao, L., Wang, Y., Yang, G., Liu, Y., Li, Y., Lu, Y., Wang, L., and Wang, L.: Volatility parameterization of ambient organic aerosols at a rural site of the North China Plain, Atmos. Chem. Phys., 22, 9283–9297, https://doi.org/10.5194/acp-22-9283-2022, 2022.

Organic aerosols are a major contributor to total aerosol mass concentrations and have implications for both human health and climate change. However, the formation of organic aerosols involves a variety of chemical and physical processes in the atmosphere, resulting in complex particle compositions. Therefore, the measurement and quantification of particle composition, especially at the molecular level, has been a long-standing measurement technology challenge that is critical for a better understanding of the sources and formation mechanisms of organic aerosols.

This paper presents an improved thermal desorption technique, the Vaporization Inlet for Aerosols (VIA), coupled to the NO3-CIMS. The VIA inlet removes gas compounds with an activated carbon denuder, vaporizes particles in a heated tube, and transfers the thermally desorbed vapors to the NO3-CIMS with a newly designed sheath flow interface. The authors demonstrate that the VIA inlet can efficiently remove background gas compounds while maintaining high transmission of particles larger than 50 nm, and that the sheath flow interface achieves low detection limits of desorbed vapors due to reduced wall loss. In addition, the authors show that the VIA inlet can also be operated in a temperature ramping mode, where the volatility of particulate compounds can be probed through thermogram analysis. The scientific topic of this paper is important, the measurement technique is novel, and the technique characterization is comprehensive. Overall, this is a relevant study that fits within the scope of the AMT. However, some technical details need further clarification and discussion to make it more useful to the community. Here are my main questions/comments:

1. While the VIA-NO3-CIMS is an online technique when operating at a fixed T, it appears to have long duty cycles (hours) for T ramping. Are there limitations that prevent rapid ramping? If so, the authors should mention them in the main text, as volatility measurement is a key feature of this technique.
2. When operating in T-ramping mode, whether particles are fully evaporated can be judged from the shape of the thermograms. However, when operating at a fixed T for high time resolution, it’s less obvious to me how to tell if 0.1 s residence time is sufficient for complete evaporation, especially for aerosol loading in polluted environments. And this introduces quantification uncertainties into the online measurement. The authors should discuss this.
3. Thermogram analysis and the corresponding 1-D model are valid for a constant particle source. However, if the T ramp takes hours (or even 10s of mins), how would this technique account for variations in particle composition and size distribution for ambient measurements?
4. The authors attribute the decreasing HOM signals after reaching their maximums to the vapor wall loss in the vaporization tube. It’s true that molecular diffusion, and thus wall loss, increases with temperature, but I’m not entirely convinced that this can cause > 90% loss as shown in Fig S14. Could this decrease also be thermal decomposition? The lack of double modes in the thermograms may simply be that the decomposition products are less oxygenated, which escapes detection by NO3-CIMS. The authors would need to justify their conclusion.
5. If the vapor wall loss in the vaporization tube is indeed significant, this can introduce contaminations due to the wall memory effect when the VIA is cooled and heated again (e.g., Fig S10b). What level of quantification uncertainty would be introduced in the continuous operation of the T-ramping mode?
6. Fig 5b & c, why does the HOM trace in stepping mode seem less stable and smooth than in ramping mode?