the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 28 Sep 2026
Phase-resolved molecular signatures associated with black carbon across biomass fuels and ignition temperatures
Abstract. Biomass combustion is an important source of black carbon (BC), yet the phase-resolved molecular patterns accompanying black-carbon evolution remain poorly constrained. In particular, detectable molecular signal abundance may not track black-carbon concentration across combustion conditions. Here, we investigated how detectable molecular profiles covary with BC by combining real-time measurements from a proton transfer reaction time-of-flight mass spectrometer (PTR-TOF-MS), for both gas-phase and particle-associated organics, with simultaneous aethalometer (AE-33) BC readings. Experiments were conducted across two biomass fuels (wheat straw and poplar wood) and three ignition temperatures (300, 500, and 700 °C). More than 200 retained nominal-m/z features were classified, and a prespecified block-level screen selected four BC-associated marker families, including furan-ring fragments, oxygenated aromatic, single-ring aromatic (SRA), and polycyclic aromatic hydrocarbons (PAHs), represented by 12 molecular markers. Particle-associated family signals show stronger within-event associations with BC (median block-level Spearman ρ = 0.86–0.92) than corresponding gas-phase signals. Particle-phase molecular profiles improve out-of-experiment BC prediction, increasing R² from 0.29 to 0.56 beyond a baseline model containing fuel, ignition temperature, and modified combustion efficiency. However, increasing ignition temperature enhanced BC production without a proportional increase in detectable BC-linked marker abundance, indicating a temperature-driven shift from precursor accumulation toward a more conversion-dominated regime. The molecular changes accompanying BC enhancement were fuel-specific, reflecting distinct marker compositions. These findings suggest that biomass BC formation is influenced not simply by detectable marker signal, but by temperature-dependent transformation of fuel-specific molecular pools, providing new constraints for interpreting molecular precursors in biomass-derived BC formation.
- Preprint
(1554 KB) - Metadata XML
-
Supplement
(511 KB) - BibTeX
- EndNote
Status: open (until 12 Nov 2026)
- RC1: 'Comment on egusphere-2026-5314', Anonymous Referee #1, 22 Oct 2026 reply
-
RC2: 'Review of egusphere-2026-5314', Anonymous Referee #2, 03 Nov 2026
reply
This manuscript presents gas and particle mass spectra (instrument: PTR-MS with CHARON inlet) and multi-wavelength filter-based light absorption measurements (instrument: AE33) of two types of biomass during their heating in a furnace at three temperatures. Only entire burns are analyzed; there is no explicit identification of flaming combustion (which forms BC) versus smouldering (which does not).
The manuscript aims to make inferences about the chemical formation mechanisms of black carbon based on this dataset. However, light absorption is not directly proportional to black carbon mass, and the mass absorption cross-section of the emitted particles will change with their degree of graphitization. Therefore, the current data set can be used to make inferences about the nature of light-absorbing PM, but not the formation mechanisms of BC mass. For this reason, I cannot recommend publication of this manuscript in its current form, in spite of the very interesting topic.
Minor comments
Line 86. Define "clean" compressed air?
Line 81. The citation to Cui et al. (2023) doesn't seem to demonstrate the previous statement. Please add a clarification of what information is being cited.
Citation: https://doi.org/10.5194/egusphere-2026-5314-RC2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 294 | 140 | 56 | 490 | 65 | 44 | 51 |
- HTML: 294
- PDF: 140
- XML: 56
- Total: 490
- Supplement: 65
- BibTeX: 44
- EndNote: 51
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Peng et al. assess the connections of organic molecules in gas and particle phase with particulate black carbon during the combustion of two biomass species (straw and wood) at three ignition temperatures. They provide novel information on the complex combustion processes and assess the relations of the different emission components with a reasonably novel statistical framework. The results can improve constraining the emission formation processes in biomass burning, making the topic suitable for publication at ACP. The work is overall well written and visualizations have suitable scientific quality. However, there are several methodological issues, especially considering data handling and emission normalization. These have to be clarified in revision before interpretation of the results is fully possible.
Methodology:
- It is recommended to denote black carbon concentration determined based on optical attenuation (i.e., when measured by aethalometer) as “equivalent BC” (eBC) to avoid ambiguity in terminology (see, e.g., Petzold et al., 10.5194/acp-13-8365-2013, that you had cited for MAC).
- What multiple scattering coefficient (C) was used? Please clarify in methodology. I assume a constant as instrument standard, whereas in reality C might be different for the different particle materials (Yus-Díez et al.; https://doi.org/10.5194/amt-14-6335-2021). This should be noted as an uncertainty and the potential consequences discussed.
- A PTR-TOF-MS was used for the VOC measurements. However, only nominal masses are reported. Why not specify the exact mass? Several biomass related peaks have identical nominal masses but exact masses distinct enough to be typically separated by time-of-flight PTR. For example furan and isoprene might both be important constituents of mz 69 and likely to have distinct connections with soot emissions. A list of all the peaks assigned and used for further analyses should be included in e.g., the supplement, together with the applied reaction rate constants when applicable.
- l. 78. Clarify here what is meant by “four datasets”. I assume this is to tell that both gas and particle phase were measured by PTR for the two fuels? Were the ignition temperatures tested consecutively or in separate experiments?
- Duplicates are mentioned on l. 94 but not elaborated on in results. Biomass burning can be very variable. Please consider the potential variation between the burns based on e.g. these replicates.
- l. 98-99 reads as if separate instruments were used. Do these settings from l. 100 onwards apply for both PTR-TOF and CHARON-PTR measurements?
- Are the concentrations presented in the manuscript as in raw exhaust conditions, or as dilution corrected values? How well was the dilution ratio maintained within your setup? Additionally, there is bound to be variance in e.g., internal dilution, which will skew the values if not accounted for. This is especially important for (intermediately / semi-volatile) organic species that can be present in either gas or particle phase depending on existing particle populations. Presenting the results in e.g., per g of fuel would be best for interpretation and enhance the comparability and later use of the results. In Implications you mention that this is not the aim of these experiments; nevertheless, the differences in experimental conditions have to be sufficiently explained and accounted for.
- l. 135. “21 valid blocks”. Explain why only 21, although four fuels x three temperatures x two phases would result in 24 conditions.
Results:
- Fig.1 and the related discussion. The separation of BC to “particle” and “gas” in the discussion is confusing. BC is obviously always in particle phase. What is the difference for the setups regarding BC for the “particle” and “gas”? If the experiments were as reproducible as suggested, the aethalometer results should be roughly identical across the experiments with the same fuel/temperature.
- It is imperative to normalize the emissions for presentation of absolute concentrations (e.g., Fig. 1) For example, if higher T leads to higher combustion rate and lower air-to-fuel ratio, the normalized BC values (e.g., ng/g of fuel) would proportionally change compared to lower T with hypothetically lower air-to-fuel ratio.
- Fig 1c. “Particle phase organic species” might be the correct term here (and elsewhere in the manuscript). VOCs are by definition volatile species that are in gas phase at ambient conditions; if these species were measured from the particles they are semivolatile at best. A full list of ions would be helpful here for interpretation, preferably with estimations of the volatility of each species.
- Very volatile species are likely not much affected by changes in the particle concentrations. This is no longer true when species are semivolatile, which seems to be the range here. Please consider how this might affect the interpretation of the results. For example PAHs are additionally prone for adsorption onto soot. This might enhance their proportional share in the particle phase at higher BC (higher T) conditions and contribute to the low correlation of gas phase species with BC.
- Fig. 3. What is indicated by the shaded areas in (a) and (c)?
- Clarify the “700-300C contrast”. Does this imply the relative difference or absolute change, or…? Is it the same as “BC-marker ratio” on l. 265?
- On l. 282: “wood exhibited higher signals than straw”; or Fig. 4. Without proof of sufficient corrections in data analysis, I assume the differences in concentrations to be simply because of differences in air-to-fuel ratio (causing internal dilution within the combustion). Although the relative shares in different exhaust emissions would remain the same at all gas conditions, the raw concentration values are not meaningful for any further realistic interpretations.
- l.308; Fig. 5. Clarify what is meant by “low/high-mz PAH”, or ‘aromatic advancement’.