the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 19 Aug 2026
Furfural Exhibits Distinct Photooxidation and Secondary Organic Aerosol Formation Among Biomass Burning Related Furanoids
Abstract. Furanoids are a major compound class emitted during biomass burning and are highly reactive toward hydroxyl radicals (OH), suggesting their oxidation may contribute to secondary organic aerosol (SOA) formation. Previous furanoid studies focused on gas-phase kinetics, leaving SOA formation poorly understood. Here, we investigate the gas-phase oxidation and SOA formation from furfural, 2-methylfuran, and 3-methylfuran reactions with OH radicals in the presence of nitrogen oxides at RH<5 %. Furfural exhibits distinct oxidation chemistry compared with methylfurans: furfural alkyl radical preferentially reacts with NO2, whereas methylfuran oxidation proceeds via peroxy radical (RO2)+NO channel to form alkoxy radicals. Methylfuran isomers also differ in oxidation product speciation, with 2-methylfuran generating C4 and C5 compounds, whereas 3-methylfuran predominantly forms C5 compounds. C4H2O3, potentially maleic anhydride, increases continuously during all furanoid oxidations, supporting its potential as an aged biomass burning marker. Furfural produces the highest SOA mass and yield (66.3 µg⋅m⁻³, 9.8 %), followed by 2-methylfuran (10.3 µg⋅m⁻³, 1.4 %) and 3-methylfuran (7.2 µg⋅m⁻³, 1.1 %). Compared with methylfuran SOA, furfural SOA shows stronger signatures of lower-volatility oligomers, with higher contributions from large m/z fragments, a greater degree of unsaturation, broader thermograms, and more reduced CHON species (O/N<3). Accordingly, estimated particulate organic nitrate fraction is highest for 2-methylfuran SOA (~45 %), followed by 3-methylfuran (~36 %) and furfural (~11 %). These results show that both substituent type and position should be considered when representing furanoid oxidation in atmospheric models.
Competing interests: At least one of the (co-)authors is a member of the editorial board of Atmospheric Chemistry and Physics.
Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.- Preprint
(2866 KB) - Metadata XML
-
Supplement
(2975 KB) - BibTeX
- EndNote
Status: final response (author comments only)
- RC1: 'Comment on egusphere-2026-4881', Anonymous Referee #1, 20 Sep 2026
-
RC2: 'Comment on egusphere-2026-4881', Anonymous Referee #2, 02 Oct 2026
This work by Joo et al. investigates the OH-initiated photooxidation and SOA formation of three furanoids (furfural, 2-methylfuran, and 3-methylfuran) under high-NOx and dry conditions. The authors employ a comprehensive suite of gas- and particle-phase analytical techniques, including AMS, CIMS and LC-ESI-MS, to investigate oxidation pathways, SOA yields, volatility, oligomer formation and nitrogen-containing species. The results highlight substantial differences in the oxidation chemistry and SOA formation among the three furanoids, emphasizing the important roles of substituent type and position.
While the study provides a comprehensive dataset, several of the comparisons and conclusions rely on quantitative or semi-quantitative comparisons of mass spectrometric signals across different compound classes. However, the potential influence of compound-dependent ionization sensitivity and MS selectivity on these comparisons is not sufficiently discussed. In addition, although both gas- and particle-phase compositions were extensively characterized, their connection and implications for SOA formation could be better established.
Overall, the manuscript is well organized and provides valuable insights, particularly regarding the higher SOA yield from furfural and the distinct SOA composition among the three furanoids. The manuscript would be suitable for publication after several important concerns are adequately addressed.
1. P6, lines 151–153: I wonder how these specific adducts ([M + Na]+ and [M + NH4]+ in ESI positive mode, and [M + HCOO] − and [M + CH3COO]− in negative mode) were determined for formula assignment and what the sources of these adduct-forming species are.
2. P8, lines 194–197: The authors note that differences in precursor decay may partly reflect differences in effective OH exposure. Since the comparison of furfural and methylfuran OH oxidation is central to the study, can the authors estimate the OH exposure for each experiment?
3. P8, lines 203–208: The relative contributions of different compound classes are compared based on their intensity fractions of the total CIMS signal. However, CIMS sensitivity can vary substantially among compounds, and compound-specific calibrations are likely unavailable in this complex system. If possible, a calibration or sensitivity assessment based on major functional groups or representative compound classes would help better constrain these comparisons. Otherwise, the authors should discuss more explicitly how compound-dependent sensitivity may influence the reported signal fractions and their interpretation as relative abundances.
4. P8, lines 210–217: Please provide clearer criteria for identifying first-generation products. Was this classification based solely on the observed temporal profiles?
5. Figure 1a: The increase in NO at the onset of furfural oxidation is interesting. What causes this initial increase, and is it accompanied by a corresponding change in NO2?
6. Please check the numerical values reported throughout the manuscript. For example, line 221 states that the fraction of signals at m/z > 50 in furfural SOA (14%) is ~3 times higher than that in methylfuran SOA (11% for both). In addition, the Conclusions report SOA yields of 1.2% at 8.9 μg m-3 and 1.0% at 6.9 μg m-3 for 2- and 3-methylfuran, respectively, which differ from those reported in the Abstract and Table 1.
7. Figure 5: The proposed mechanisms provide a detailed picture of the gas-phase oxidation chemistry. However, it is not clear how these gas-phase pathways are connected to the observed particle-phase composition and SOA formation. Please further discuss this connection. In addition, labeling the reaction positions (e.g., position 2) for the three furanoids would help readers follow the proposed mechanisms.
8. Figure S5: Some molecular formulas appear in both the first- and second/later-generation product groups. Could these signals represent different isomers, or could the authors clarify why the same formulas are assigned to multiple generations? It would also be helpful to discuss the multi-stage temporal profiles observed for some carbonyl species, such as C5H6O2.
9. P14, line 362: Does “CxHyOz>1 species” refer to oxygenated organic compounds containing no nitrogen, or to AMS ion fragments?
10. How was the FIGAERO-CIMS desorption temperature program selected? Given the substantially different SOA compositions among the three furanoids, could the desorption conditions affect the detected oligomer contributions and introduce uncertainties in their comparison?
Citation: https://doi.org/10.5194/egusphere-2026-4881-RC2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 146 | 102 | 65 | 313 | 63 | 61 | 55 |
- HTML: 146
- PDF: 102
- XML: 65
- Total: 313
- Supplement: 63
- BibTeX: 61
- EndNote: 55
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Reviewer comments on “Furfural Exhibits Distinct Photooxidation and Secondary Organic Aerosol Formation Among Biomass Burning Related Furanoids” by Joo et al.
Joo et al. explores the gas phase oxidation and SOA formation potential of three compounds emitted during biomass burning: furfural, 2-methylfuran, and 3-methylfuran. Using chamber experiments, they show that gas and particle phase composition from each precursor varies substantially, with distinct radical branching, fragmentation, and oligomerization behaviors. SOA yields are highest for furfural, while formation of particulate organic nitrates is highest for 2-methylfuran. Taken together, these results indicate that biomass burning furanoids have different, substituent-dependent fates and impacts which require unique treatment in atmospheric models.
The study only covers 3 experiments, and it is not clear if replicates were performed. Additional replicates and additional variation of parameters would improve the manuscript. It would be helpful if the authors commented on reproducibility. However, the analysis presented is thorough and comprehensive.
The topic and work is a valuable addition to the literature and of interest to readers of ACP. Following minor revisions, I would recommend this manuscript for publication.
Specific Comments
Lines 138-141: For clarity, what compounds with known vapor pressures were used? I do not believe they are specified in Joo et al. (2019) or only their elemental formula is available in the SI. Is the same volatility calibration curve from Joo et al. (2019) applied to experiment here?
Line 217: Would CH2O2 be formic acid?
Lines 307-310: It is unclear why alky radicals from 2- and 3-methylfuran + OH reactions (MF2/3-R1/2) do not form alkoxy radicals via NO2 > NO pathway as shown for furfural + OH reactions which generate F-R1/2. Do computational chemistry or other experimental results show that MF2/3-R1/2 react much faster with O2 than with NO2/HO2/O3, as is the case with F-R3? Are these rates well known and are there reasonable conditions where the balance of alkyl radical fate would shift along these branches?
Line 343-344: How are thermal decomposition products of oligomers handled in the FIGAERO analysis performed here? Are they included in the discussed estimates of oligomer formation? Is there evidence that any precursor forms more thermally labile oligomer species (higher extent of fragmented products)?
Lines 377-379: Can the effect of NH3/NH4+ heterogenous chemistry be distinguished from gas phase RNO2 formation when attributing formation of reduced nitrogen in furfural SOA in these ammonium sulfate seeded experiments? Is such heterogenous chemistry occurring in 2-MF and 3-MF SOA?
Lines 441-443: Here you note that maleic acid increases continuously in the furfural + OH experiments, and later discuss that it increases in all experiments, showing that it is an appropriate aging tracer for biomass burning plumes. In line 217, you mention that only CH2O2 increases continually throughout the experiments, but maleic acid also does so. Maleic acid is discussed in both the abstract and here and I would suggest discussing it thoroughly in a dedicated paragraph if possible. Consider including maleic acid in the main figure time series traces in Figure 1.
Figures 1, 2, and 3: While I appreciate the consistent top to bottom ordering used throughout the paper, additional labels for which pairs of plots are related to each precursor would be helpful.
Figures 1b,d,f and S5: It is not clear why certain traces appear in the main text figure as opposed to in the supplement. I would recommend trying to include any species whose temporal behavior is discussed in the main text in Figure 1.
Figures 5 and S4: Are the oxidation pathways shown and discussed here entirely from literature? Are any mechanisms unique additions associated with this study or do existing mechanisms explain the measured products and apparent branching? From a flow/structure perspective, consider shifting some of the basic discussion of the varying reaction pathways into the introduction or earlier in the paper. The substantial fate differences dependent on furanoid substituents is certainly a new area to this reader and would be valuable to highlight early.
Figure 6: Move the marker size legend off the axes for improved legibility. What is the limit of detection of the FIGAERO CIMS?