the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 08 Jul 2026
First field deployment of VIA-NO3-CIMS for molecular characterization and source apportionment of particle-phase oxygenated organic molecules in eastern China
Abstract. Secondary organic aerosol (SOA) is a major component of fine particulate matter, yet its molecular-level characterization remains challenging. Here, we systematically evaluated the performance of a Volatilization Inlet for Aerosols coupled with nitrate chemical ionization mass spectrometry (VIA-NO3-CIMS), including particle transmission, volatilization behavior, quantitative response, and operational stability. Following instrument characterization, the first field deployment of VIA-NO3-CIMS was conducted at the SORPES station in Nanjing, eastern China. A high-confidence dataset containing 1,695 particle-phase oxygenated organic molecules (OOMs) was established. The summed OOM concentration correlated strongly with AMS-derived organic aerosol concentrations (r = 0.88), demonstrating the quantitative capability of VIA–NO3–CIMS. Compared with gas-phase species, particle-phase OOMs exhibited larger carbon numbers, higher unsaturation, and lower volatility, with low-volatility compounds dominating the particle phase. Positive matrix factorization resolved seven SOA factors, including three biogenic-dominated, two anthropogenic-dominated, one regional background, and one pollution-episode factor. These factors contributed approximately 36.5 %, 35.4 %, 15.6 %, and 12.1 % of the total particle-phase OOM signal, respectively. Their molecular characteristics and temporal behaviors indicate important influences from daytime photochemistry, nighttime NO3 oxidation, regional transport, and episodic pollution accumulation. Factor contributions evolved systematically with PM2.5, with freshly formed SOA dominating under clean conditions and the pollution-episode factor becoming increasingly important as pollution levels increased, eventually dominating under the most polluted conditions. These results demonstrate that VIA-NO3-CIMS enables quantitative molecular-level characterization and source apportionment of particle-phase SOA, providing new insights into the composition and evolution of urban organic aerosols.
- Preprint
(4154 KB) - Metadata XML
-
Supplement
(25382 KB) - BibTeX
- EndNote
Status: open (until 19 Aug 2026)
- RC1: 'Comment on egusphere-2026-3647', Anonymous Referee #1, 28 Jul 2026 reply
-
RC2: 'Comment on egusphere-2026-3647', Anonymous Referee #2, 04 Aug 2026
reply
This manuscript presents a field application of VIA-NO3-CIMS at SORPES to characterize particle-phase OOMs and explore their potential source/process contributions. The dataset is interesting, and the combination with AMS measurements could provide useful molecular-level insight into SOA composition. However, several central interpretations are currently not sufficiently supported. In particular, the source-oriented PMF factor names, the non-concurrent gas-particle comparison, and the quantitative interpretation of VIA-NO3-CIMS signals require substantial clarification. I recommend major revision before the conclusions on particle-phase OOM sources and gas-particle differences can be accepted.
Major comments
1 The PMF factor names “biogenic-dominated” and “anthropogenic-dominated” are too strong. The authors acknowledge that PMF factors are mixed and not pure source categories. After applying the DBE-Oeff screening, the contribution of the biogenic-related factors is also substantially reduced, from about 51% in the PMF factors to 37%. Have the authors checked whether the biogenic ions enhanced in the PMF factors are consistent with the molecular classification criteria used in Nie et al. 2022? If not, the factors should be renamed more conservatively.
In addition, two factors with pronounced daytime peaks are both classified as biogenic factors. I do not think AVOC oxidation during daytime can be ignored. The authors should justify why these daytime factors are assigned mainly to biogenic sources rather than mixed BVOC/AVOC photochemical SOA.
2 The particle-phase OOMs were measured in 2024, whereas the gas-phase OOM reference dataset appears to be from 2022. This makes several gas-particle comparisons rather weak. Similar meteorology alone is not enough to prove that gas-phase HOM/OOM composition was stable across years. If the authors can show multi-year evidence that gas-phase composition is seasonally reproducible, the comparison would be more convincing. The measurement years should also be clearly marked in Figs. 2 and 3 and in their captions to avoid misunderstanding. The Supplement also contains apparent 2022/2023/2024 inconsistencies that need correction.
Since VIA-NO3-CIMS uses the same nitrate CIMS principle as gas-phase HOM measurements, the authors should also explain why switching or bypass measurements were not available during the 2024 campaign.
3 Line 165: The use of SMPS volume loss to infer organic volatilization is not fully convincing, because ambient particles contain both organic and inorganic components. For example, NH4NO3 can evaporate readily, whereas sulfate is much less volatile. Since AMS was operated in parallel, why did the authors not use AMS to measure the change in organic mass during VIA heating directly? A component-resolved check would be more appropriate than SMPS volume loss alone.
4 The correlations between AVOC or BVOC proxies and selected PMF factors are not sufficient unless specificity is shown. The authors should show the same O3 x AVOC, ALWC x AVOC, and BVOC-related proxy correlations for all factors. Otherwise, these correlations may simply reflect common diurnal cycles, boundary-layer evolution, or pollution accumulation rather than source-specific formation.
In addition, the number of data points in Fig. S25 is much smaller than expected from the online dataset. The authors should explain how these points were averaged or selected. They should also specify which AVOCs and BVOCs were included in the proxy calculations.
5 The VIA was operated at 300 °C, but the manuscript does not sufficiently address whether thermal decomposition or fragmentation affects the observed formulas. This is especially important for organic nitrates, peroxides, oligomers, and thermally labile HOMs. The authors should clarify whether the detected ions mainly represent intact molecules or may include thermal decomposition products.
6 It is not clear how the whole VIA-NO3-CIMS system was calibrated for particle-phase organic measurements. The manuscript describes a sulfate/sulfuric-acid-based empirical calibration, but this does not fully establish the response of the complete system to organic aerosol, including particle transmission, evaporation efficiency, vapor losses, ionization efficiency, and compound-dependent sensitivity. The authors should clarify what exactly was calibrated, which parts of the system were included, and how the calibration factor should be applied to diverse particle-phase OOMs. The statement that VIA-NO3-CIMS detects about 22% of AMS-derived SOA should be interpreted with this limitation in mind.
Other minor comments：
7 Line 113: The manuscript should clarify what “first” means: first in China, first in eastern China, or first at Nanjing/SORPES. The title, abstract, and conclusions should use a consistent and accurate scope.
8 The VIA setup needs more complete reporting, including the manufacturer/brand or whether it was custom-built. The authors should also state how often the gas denuder was replaced or regenerated, and how denuder breakthrough was checked during the campaign.
9 Line 364: The authors report 1,695 high-confidence OOMs, but it is unclear how many peaks were initially detected, how many were fitted, how many meaningful peaks remained unfitted, and what fraction of the total signal they represented. This information is needed to judge how representative the fitted molecular dataset is.
Citation: https://doi.org/10.5194/egusphere-2026-3647-RC2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 172 | 78 | 23 | 273 | 54 | 15 | 14 |
- HTML: 172
- PDF: 78
- XML: 23
- Total: 273
- Supplement: 54
- BibTeX: 15
- EndNote: 14
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
This study couples a VIA with NO3-CIMS and presents its first deployment for continuous measurements of ambient particles at a suburban site in Nanjing. The authors investigate particle-phase OOMs from three perspectives: molecular composition, gas–particle compositional differences, and PMF-based source apportionment. The online measurement of particle-phase highly oxygenated organic molecules at high time resolution has clear methodological value. The integration of VIA-NO3-CIMS measurements with HR-ToF-AMS, VOC, meteorological, and trajectory data also gives this dataset potentially important scientific significance. The manuscript is generally well structured and contains a comprehensive set of figures. If the key methodological issues are adequately resolved, this work could become an important case study for the application of VIA-NO3-CIMS to ambient aerosol measurements.
However, the current conclusions require more robust support from instrument characterization and statistical analyses, or alternatively, more cautious wording. I would be willing to recommend this study for publication after the issues below have been addressed.
The authors derived an empirical calibration coefficient by regressing the normalized HSO4⁻ signal measured by VIA-NO3-CIMS against particulate sulfate mass concentrations measured by the AMS, and subsequently applied this single coefficient to all detected organic molecules. This approach raises several concerns, such as the conversion efficiency of particulate sulfur-containing salts into detectable H2SO4/HSO4⁻ at 300 °C, post-evaporation wall losses, and compound-dependent NO3⁻ ionization responses for different organic molecules. The response to sulfuric acid cannot automatically be assumed to represent the responses of multifunctional organic compounds, peroxides, and organic nitrates. Moreover, the contribution of OOMs to AMS-derived SOA additionally depends on both the AMS-PMF separation of SOA and the VIA calibration. I recognize that quantifying all uncertainties associated with the VIA-NO3-CIMS analytical and quantification process, as well as the AMS-PMF analysis, is challenging. Nevertheless, I suggest that the uncertainties that can be reasonably estimated should be explicitly reported in the results. Accordingly, the statements in the manuscript (including the Abstract and Conclusions) regarding "quantitative capability" and absolute concentration or contribution values should be explicitly described as “semi-quantitative” or “empirical estimates”.
The total particle-phase OA signal reached a maximum at 300 °C (Fig. S1a). Therefore, the authors selected 300 °C as the operating temperature for field measurements. However, maximizing the total signal does not necessarily mean that the original particle-phase molecules are measured most accurately. The continued increase in signal from above 200 °C to 300 °C may reflect more complete evaporation, but it may also arise from thermal decomposition of oligomers, peroxides, carboxylic acids, or organic nitrates, as well as dehydration, decarboxylation, and fragmentation or recombination processes. Have the authors compared the complete molecular spectra at different temperatures? The authors state that thermal decomposition cannot be fully excluded during the desorption process, but that this effect is generally considered limited under the current operating conditions. The authors should provide a more substantial justification or supporting evidence for this assertion.
Minor comments
Abstract, L30–31: an r value of 0.88 only indicates that the two signals co-vary over time; it does not demonstrate the accuracy of the absolute concentrations. Therefore, the statements that this calibration “demonstrates quantitative capability” (L30-31) and that the study provides “quantitative molecular-level characterization” (L44) are not sufficiently accurate.
L168–170: The main text reports an average evaporation efficiency of approximately 70%, whereas Sect. S1.5 reports a range of 70%–92%. Please reconcile these values. The description of particle-volume loss as direct volatilization of organic material should also be corrected.
L249–255: Please explain how the alternating 10 min sample and background periods were paired. Background interpolation during rapidly changing conditions may introduce either positive or negative biases. A representative measurement cycle and the stability of the zero-air/HEPA-filter background should be shown.
L334–339: An r2 value of 0.70 between PM1 and PM2.5 is insufficient on its own to demonstrate that “submicron particles dominated fine particulate mass.” The PM1/PM2.5 mass ratio and its distribution should be reported.
L357–360: The cross-reference to “Sect. 2.5.2” is incorrect and should be Sect. 2.4.2.
Supplement: Figure S26 is used twice. The second occurrence should be renumbered as Fig. S27.