Combustion-Derived Organic Aerosols Enhance PM<sub>2.5</sub> Oxidative Potential in a Medium-Sized North China Plain City

Cai, Jing; Liu, Yongchun; Zheng, Feixue; Ding, Dian; Du, Wei; Yuan, Liang; Zou, Yingjie; Suolang, Lazong; Lu, Rongxiang; Li, Wanwei; Liu, Chao; Fan, Xiaolong; Zha, Qiaozhi; Luan, Mengxiao; Wang, Jiandong

doi:10.5194/egusphere-2026-4106

Preprints

https://doi.org/10.5194/egusphere-2026-4106

Preprints

14 Jul 2026

| 14 Jul 2026

Status: this preprint is open for discussion and under review for Atmospheric Chemistry and Physics (ACP).

Combustion-Derived Organic Aerosols Enhance PM_2.5 Oxidative Potential in a Medium-Sized North China Plain City

Jing Cai, Yongchun Liu, Feixue Zheng, Dian Ding, Wei Du, Liang Yuan, Yingjie Zou, Lazong Suolang, Rongxiang Lu, Wanwei Li, Chao Liu, Xiaolong Fan, Qiaozhi Zha, Mengxiao Luan, and Jiandong Wang

Abstract. Organic aerosol (OA) is a major component of PM_2.5 with significant health impacts, particularly in the highly populated NCP. However, characterizing OA composition, toxicity, and sources remains challenging. To address this, we conducted an intensive campaign throughout the 2023–2024 heating season in Weifang, a representative medium-sized NCP city. Despite overall air quality improvements, severe winter PM_2.5 pollution events still occurred, with a maximum concentration reaching 985 µg m^-3. Using a Chemical Ionization Time-of-Flight Mass Spectrometer equipped with a Filter Inlet for Gases and AEROsols (FIGAERO-CIMS), we obtained molecular composition and identified six clusters, linking them to sources resolved by positive matrix factorization (PMF) applied to the online dataset, including secondary inorganics, biomass burning, vehicle emissions, coal combustion, and dust. To evaluate toxicity, we determined the oxidative potential (OP) in Weifang. While volume-normalized OP (OP_v) increased with both the degree of unsaturation and O:C ratios of OA, the compounds most positively correlated with mass-normalized OP (OP_m) exhibited an average carbon oxidation state of -0.5 and O:C of 0.5, indicating that moderately oxygenated OA possesses high OP activity. Multiple linear regression revealed that two combustion-derived groups drive OP: highly unsaturated C_10-15 compounds with ~10 oxygen atoms from solid fuel burning, and C_>15 compounds with <5 oxygen atoms or nitrogen-containing compounds from traffic emissions, which exhibited OP_m up to 23 times higher than the ambient average. Our study highlights the continued need for emission controls, particularly those targeting solid fuel combustion in the NCP region beyond its megacities.

Received: 09 Jul 2026 – Discussion started: 14 Jul 2026

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.

Download & links

Preprint (PDF, 2544 KB)

Supplement (942 KB)

Download & links

Jing Cai, Yongchun Liu, Feixue Zheng, Dian Ding, Wei Du, Liang Yuan, Yingjie Zou, Lazong Suolang, Rongxiang Lu, Wanwei Li, Chao Liu, Xiaolong Fan, Qiaozhi Zha, Mengxiao Luan, and Jiandong Wang

Status: open (until 30 Sep 2026)

Post a comment Subscribe to comment alert

RC1:
'Comment on egusphere-2026-4106', Anonymous Referee #1, 26 Jul 2026 reply
This manuscript investigates the molecular characteristics of combustion-derived organic aerosols (OA) and their contribution to PM_2.5 oxidative potential (OP) in Weifang, a medium-sized city in the North China Plain. The authors combine FIGAERO-CIMS molecular characterization, PMF source apportionment, hierarchical cluster analysis, and dithiothreitol (DTT)-based oxidative potential measurements to identify OA components associated with coal combustion, biomass burning, vehicle emissions, and secondary formation. The manuscript argues that combustion-related OA, particularly high-DBE aromatic compounds, plays a dominant role in enhancing PM_2.5oxidative potential. The monitoring technique is novel as the authors integrated online CIMS, PMF, and OP measurements. Although the topic is relevant and the dataset contains potentially valuable molecular information, the major scientific claims rely heavily on correlations, indirect source assignments, and assumptions about molecular toxicity without sufficient experimental validation. The work would be improved if further convincing mechanistic evidence can be provided besides the currently represented analytical evidence.

Major comments

The manuscript repeatedly claims that combustion-derived OA enhances PM_2.5 OP and identifies high-DBE compounds as key contributors. However, most evidence is based on statistical correlations between FIGAERO-CIMS signals and DTT-based OP measurements. Correlation between molecular abundance and OP does not establish that these compounds are chemically responsible for oxidative activity. There may be covary issues due to common emission sources or atmospheric processing pathways. For example, the manuscript states that high-DBE compounds, likely aromatic derivatives, are major contributors to OPv based on enhanced correlations between DBE and OP. However, DBE is only a structural indicator and cannot distinguish among redox-active quinones, nitroaromatics, polycyclic aromatic hydrocarbons, sulfur-containing compounds, or thermally generated fragments. Without targeted quantification of representative compounds, authentic standard testing, or fractionation experiments, the proposed molecular drivers remain speculative and may be an overestimation.

The identification of coal combustion-, biomass burning-, and vehicle-related OA compounds is primarily based on correlations between molecular formulas and PMF-derived source factors. This approach has substantial uncertainty because PMF factors themselves are derived from bulk chemical tracers rather than direct OA molecular measurements. The assignment of sulfur-containing compounds to coal combustion is particularly questionable. The manuscript argues that increased sulfur oxidation states indicate coal combustion-related chemistry, but atmospheric oxidation and heterogeneous processing can generate similar sulfur-containing products from multiple sources. Likewise, compounds correlated with biomass burning are interpreted as combustion tracers, but biomass burning and coal combustion are known to co-occur in the North China Plain, making source separation difficult. The moderate correlation of levoglucosan with biomass burning is acknowledged by the authors, but the manuscript does not adequately quantify the uncertainty introduced by mixed combustion sources.

The manuscript combines several established methods, including FIGAERO-CIMS molecular identification, PMF source apportionment, hierarchical clustering, and DTT assays. However, the conceptual advance is unclear. Previous studies have already demonstrated that oxygenated organic aerosols, combustion-derived aromatic compounds, and nitroaromatic species influence aerosol oxidative potential. The manuscript does not introduce a new mechanistic framework, a new molecular toxicity indicator, or a predictive model that substantially improves understanding beyond previous work. The authors concluded that combustion-derived OA contributes to PM toxicity simply based on existing literature. To justify publication in a high-impact journal, the authors need to demonstrate what fundamentally new knowledge is obtained from this dataset.

The interpretation of oxidation state and the role of oxygen on OP is internally inconsistent. The manuscript reports that OPv increases with increasing O ratio, while OPm increases initially and then decreases at higher O ratios. The authors attribute this to highly oxygenated OA becoming less toxic during haze evolution. However, this explanation is not sufficiently supported. Oxidation state alone does not determine oxidative potential. Aging can both generate and remove redox-active structures depending on precursor composition and reaction pathways. The manuscript does not distinguish between increased oxygenation caused by functionalization and oxygenation caused by fragmentation. Additional evidence, such as Van Krevelen analysis combined with molecular-level identification of redox-active species, is required.

The manuscript lacks stronger validation linking ambient OA composition to health relevance. The DTT assay provides a chemical measure of oxidative potential, but the manuscript interprets OP changes as indicators of aerosol toxicity. This connection requires caution. DTT activity represents only one oxidative mechanism and does not directly represent cellular toxicity, inflammatory response, or health impacts. The conclusions should be limited to oxidative potential rather than implying broader toxicological effects. Ideally, the study should include complementary oxidative assays or biological responses to support stronger health-related claims.

Minor comments

Lines 146–150 (Page 6): The authors introduce the integration of FIGAERO-CIMS, OP assays, cluster analysis, and PMF source apportionment as the methodological framework. However, the rationale for combining these approaches is not sufficiently explained. The authors should clarify how each method addresses a distinct scientific question rather than simply providing parallel datasets.

Lines 168–177 (Page 6): The sampling description lacks important QA/QC information. Although the authors state that “all data have undergone strict quality control,” specific details regarding field blank correction, filter storage conditions, extraction recovery, and possible sampling artifacts should be provided.

Lines 204–223 (Page 7): The manuscript identifies 1630 compounds using FIGAERO-CIMS and applies semi-quantitative calibration based on levoglucosan and succinic acid. However, the uncertainty associated with applying only two calibration standards across a broad chemical space (m/z 120–500) should be discussed. The authors should provide an estimate of uncertainty introduced by this calibration approach.

Lines 212–220 (Page 7): The statement that CHOX species can be regarded as oxygenated compounds because I⁻ ionization favors moderately oxygenated OA requires more careful explanation. The authors should clarify the potential bias against highly oxygenated or less oxygenated compounds introduced by the ionization scheme.

Lines 224–250 (Page 8): The DTT assay methodology is described, but the manuscript should clarify whether extraction efficiency was evaluated. Since the authors acknowledge that water-insoluble components, including EC, insoluble organics, and insoluble metals, were excluded (Lines 232–234), the potential impact of this limitation on total PM oxidative potential should be discussed more explicitly.

Lines 252–264 (Page 8): The HCA and PMF methodology requires additional statistical justification. The authors should provide sensitivity tests showing whether the six HCA clusters and selected PMF factor number remain stable under different clustering parameters or factor numbers.

Lines 317–326 (Page 10–11): The comparison among Weifang, Beijing, and Wangdu is potentially valuable, but the authors should clarify whether differences in sampling periods, meteorological conditions, and sampling locations may influence the observed OA composition differences.

Lines 429–432 (Page 14): Figure 3 contains a large number of molecular formulas, but important marker compounds are not explicitly labeled. The authors should annotate representative compounds to improve interpretation and readability.

Lines 480–485 (Page 16): The criteria for selecting high-abundance compounds (>0.05% of total CHOX signal and twice background level) should be justified. The authors should explain whether different thresholds significantly alter the HCA results.

Lines 513–518 (Page 17): The authors interpret Cluster F compounds as thermal fragments based on thermogram behavior. Since thermal decomposition may affect a broader range of FIGAERO-CIMS compounds, the authors should discuss whether other clusters may also be influenced by thermal artifacts.

Lines 815–823 (Page 28): The authors acknowledge that PMF and HCA cannot fully separate primary emissions from aging processes. This limitation should be emphasized earlier in the discussion because it directly affects the interpretation of source-specific OP contributions.

Figures 5 and 7 (Lines 570–575 and 719–722): The figures contain extensive molecular-level information but lack sufficient guidance for readers. The authors should improve figure legends by defining abbreviations, explaining color coding, and identifying representative compounds.

Reply
Citation: https://doi.org/10.5194/egusphere-2026-4106-RC1
RC2: 'Comment on egusphere-2026-4106', Anonymous Referee #2, 12 Aug 2026 reply

The manuscript by Cai et al. investigates the chemical composition, sources, and toxicity of organic aerosol and PM2.5 in a representative medium-sized NCP city, based on FIGAERO-CIMS mass spectrometer and DTT measurements in addition to source apportionment and data analysis methods including HCA, PMF, and MLR. The topic of this manuscript is interesting. However, the manuscript needs major improvement in the texts to support its statements, as a lot of the statements didn’t show any data or figures. Besides, major revision on the discussions and interpretation are also needed before its possible publication on ACP. Please see my comments and questions below.
Major:
The manuscript is about dominant contributors from different OA types to OP, but then the OA source apportionment is done through linking/correlating HCA clusters to PM2.5 PMF source factors with input data of BC, TC, metals, and secondary inorganics. It would be make more sense to (try to) run PMF on the OA measured by FIGAERO-CIMS even though it’s probably only ~40 filters in the 4 months campaign. This would be more convincing than the correlations of OA compounds with the PM2.5 source factors since for example OA cannot originate from secondary inorganics or dust.
Besides, the authors have concluded several times (e.g. Line 302-307, Line 536-542) about dominant role based on the strong correlation coefficient, such as key OA role in PM2.5 and OC role in OPm. However, strong correlations only mean variations in A parameter are more closely associated with variations in B parameter, but does not necessarily imply about it being a key/dominant driver. Please avoid overinterpretation of the correlations for dominance. The manuscript may focus on what the manuscript and introduction says about the research question/gap on dominant contributors from different OA types to OP at this site.
Specific:
Line 37. Is the overall air quality improvement compared to other cities at the same measurement period or same location but in previous years? Please specify.
Line 41-43. The linking part of OA sources to the PM2.5 sources. Does it mean none of the OA measured come from any biogenic origins? And also, how can the OA sources originate from secondary inorganics and dust? It seems weird to me to do the source apportionment of OA based on correlations with PM2.5 sources with PMF input data of BC, TC, metals, and secondary inorganics.
Line 80-82. Do quinones and semiquinone generate ROS without ascorbic acid and iron? In Line 75-77, it seems they can generate ROS on their own.
Line 130-131. Are the results of SOA dominating OP in Beijing from Zhou et al (2019) contradictory with the results from Ma et al (2018) in Line 126-128 who identifies solid fuel combustion, vehicle emissions and secondary formation as the dominant drivers of OP?
Line 167. Is the heightened exposure to elevated concentrations of fine particulate matter compared to other cities in NCP region or same location but in previous years? Please specify.
Line 175-177. What do you mean “strict quality control”? Please elaborate a bit more.
Line 215-220. Please add the sensitivity you obtained from the calibrations of levoglucosan and succinic acid. Also, please discuss the uncertainties in CHOX mass concentrations calculated using this method described here.
Line 232-235. Could you comment on the toxicity of these water-insoluble components including HULIS and how do they compare to the water-soluble ones, since you talked about ROS from HULIS in Line 76 and 82?
Line 277-283. Please add the gaseous pollutant plots in the manuscript or SI to show what you are talking about.
Line 294-297. Why did Weifang exhibit this distinct daily cycle? Is this distinct daily cycle only shown after January 2024, and if so why? Please also add the daily cycle both from Weifang before January 2014 and from Beijing to show that Weifang daily cycle after January 2024 is distinct.
Besides, please add the gaseous pollutants diurnal cycle in Fig S2 for better comparison. Also, why is Fig. 3 mentioned here to show the emission influences? I didn’t get it. Please clarify.
Line 302-307. Please add a pie chart for PM2.5 contributions, and also scatter plot and the correlations for PM2.5 vs OA and PM2.5 vs nitrate to show what you are talking about.
Besides, a stronger correlation between OA and PM2.5 does not necessarily imply that OA emissions are the key driver. It just indicates that variations in OA are more closely associated with variations in PM2.5 than are variations in nitrate. One example is in Line 391-395, as you stated secondary inorganics dominated the pollution.
Line 335. Please add the residential burning fraction in Weifang.
Line 336-338. What are the compounds with the largest signal in Weifang in Fig 1c (i.e., the compound in blue and purple with DBE<2)? They don’t seem to be the levoglucosan nor compounds with DBE>4.
Besides, why are you using number fraction of CHOX? I think signal fraction would make more sense.
Line 350-353. The naming of the factors for Fig S3 was described in SI. But their high correlation with organic and inorganic tracers as stated in Line 357 and SI should be included to support.
Besides, in SI it says “A total of seven sources were resolved in this study, including secondary inorganic aerosols, vehicle emissions, coal combustion, biomass burning, fireworks and related emissions, industrial emissions, and dust.” But then in the two paragraphs afterwards, it says “The variation in Q/Qexp values suggested that five factors could explain the majority of the residuals”. It’s confusing how many factors were resolved.
Line 377-381. CO seems to reach max at 8-9 am instead of max vehicle emissions at 10 am. Is CO the best indicator still? How about the timing of the max NOx in the morning? Would it match better the timing of the max vehicle emissions?
Line 391-395. On Jan 12, solid fuel burning didn’t dominate the pollution based on Fig 2, as secondary inorganics still dominated with a contribution of 46%. Besides, solid fuel burning contributions seem not to be 36% if I summed up the coal combustion of 9% with biomass burning of 25%.
Line 395-399. For biomass burning and coal combustion comparisons, it seems heating season didn’t include the spring-heating season, while for dust comparisons the spring-heating season was included in the heating season. Is spring period of March 1-March 15 (Line 384) heating season or not? It’s quite confusing.
Line 400-402. Please show the PM10 data to support your statement. I only found PM2.5 in figs 1 and 2. PM10 measurement was never mentioned in the method as well.
Line 406. It would be nice to show the correlations between identified OA compounds with the dust source in the SI, at least for post-heating season when dust source is important.
Line 423-425. Why not showing the thermograms with Tmax to validate this for these ions?
Line 434-436. Here you are choosing the well-known biomass burning tracers to compare even though their correlations are below 0.75. However, for coal combustion case the OA compound correlations discussed are above 0.75 (Line 406-409). Same for vehicle emission comparisons for the C15H20NO5 and C17H24O5 which seem to have correlations below 0.6 (Line 449-452) from Fig 3d. It feels subjective what compounds to choose to compare. Maybe PMF can do a better job for OA source apportionment.
Line 443-445. What do you mean of “one in fourth of the levoglucosan-to-OA ratio”? Please rephrase it.
Line 460-461. It’s not very straightforward in Fig 3c to see the differences of DBE and MW between different panels of fig 3. Maybe can show pearson’s r-weighted DBE vs pearson’s r-weighted molecular masses for OA with different PM2.5 sources as the Fig 3?
Line 480-487. Do the compounds discussed in Fig3 (in Line 409-410, 420-421, 435-436, 452, and 462) fall into this 274 high-abundance compound list?
Besides, I think Fig 4 says more than Fig 3 in terms of OA sources and correlations with PM2.5 source factors. Is it really necessary to have Fig 3 and relevant discussions? However, the best way to talk about OA sources would be to run PMF on OA and show factor correlations, instead of OA cluster correlations to PM2.5 PMF factors.
Line 519-530. Dust itself is most minerals, but dust transport can also bring over OA together as you stated in Line 654 when they are mix with anthropogenic emissions in NCP (Zhang et al., 2026). Have you tried to run the HCA for the post-heating season to check, since this period was dust dominated?
Line 536-542. Does Fig S6 show the correlations of all days or excluding Feb 11 since it’s not very clear from the text and Fig S6 caption?
Besides, a stronger correlation between OPm and OC does not necessarily imply that OC are the key driver. It just indicates that variations in OC are more closely associated with variations in OPm than are variations in transition metals. As you also stated in Line 538, metals are typically one of the OP drivers; if you focus on Feb 11 firework case, then your conclusion would be different.
Line 543-554. Please show the OP time series figure. It’s difficult to relate if there is no data supporting the statement. Also, is Feb 11 the only day having elevated SO2 and metal levels since you excluded only Feb 11 due to this reason?
Line 555-556. I don’t understand why Fig S7 shows some OA species are important in driving the OPm? Fig 5c-d probably show better? Fig S7 only shows the signal ratio of Feb 17 to Jan 12. Why are these two days chosen and why are OA compound signal ratios larger than 1 are shown and how does this relate to importance in OPm? Please explain more.
Figure 5c-d. I assume the size of the compound markers refer to the spearman correlation coefficient? Does smallest marker refer to strongly negative correlation and biggest one strongly positive correlation? Please specify.
Line 579-581. I can see the relationship from more aged OA to higher O:C ratio, but more discussions are needed to explain why is there a difference trend for OPm vs O:C compared to OPv vs O:C, because the difference of OPm and OPv is probably the density (Line 237-247). Maybe some literature on OPv vsO:C from the literature similar to your discussions of OPm vs O:C from Liu et al (2024a).
Line 593-597. How would it look like when correlating with OPv? Could add it in SI.
Line 597-603. Could you explain more the connection of group (1) C10-15HO10 compounds and group (2) C>15HO<5 compounds to solid fuel burning emissions and traffic emissions, respectively from section 3.3? Based on fig 5b, group (1) may be from BB only since cluster C and E correlated with BB also exhibited long carbon chains of C9-10. However, group (2) doesn’t seem to relate with vehicle emissions which only exhibited relatively higher correlations with cluster A and D with carbon chain of C8,11.
Line 610-613. The four days are also not the same four days in Fig 2a. It seems the day choice of data is also a bit subjective. Would it make more statistical sense if you just do the average for the autumn-winter heating season, spring-heating season, and post-heating season? Or give out the reasoning for different choices of days selected.
Line 615. Are 76% and 47% the average contributions for the two episodes?
Line 619-626. OPm from vehicle emissions also seem to be elevated on these two days (seen also in Fig 6a). Would it make sense to include it when talking about dominant contributors to OPm? In particular, compounds from this group also correlated with OPm as stated in Line 686-688.
Line 639-644. Continuing about OPm from vehicle emissions: do you mean the concurrent accumulation of secondary inorganic formations cancelled out some OPm from vehicle emissions at high vehicle emission source contributions?
Line 647-654. If the anthropogenic emissions are related to the OPm decrease at high dust source contributions, are they likely to be the primary solid fuel combustion as stated in Line 625?
Line 677-683. Please show the data. Besides, are these high O:C compounds significantly correlating with OPm contradictory to what was shown in Fig 5b and Line 581-590 about decline in OPm at higher O:C?
Line 704. Can you show the temperature and RH data to support?
Fig 8: Feb 11 (one day after Chinese New Year) exhibited high OPm caused by transition metals during firework events (in Line 543-554). Where would this date sit in this plot? And for this day, I assume the conclusion of dominant OPm contributors would be different since you also stated in Line 538 that transition metals are typically being considered as one of the drivers of OP?
Line 770-774. It’s difficult to see from Fig 8. The different color shades illustrating the standard deviation ranges of previously reported OPm values from various sources are overlapping on top of each other, in particular biomass burning, traffic emissions, and SOA_bio. Please try to make the figure more clear.
Line 774-783. If the biomass and coal fraction are positively scaled with OPm plus primary OA such as coal combustion exhibited so low OPm compared to other sources (Fig 8), is it still solid to say primary solid fuel combustion (stated in Line 625) is the dominant driver of OPm? Have you tried plotting biomass fraction alone since it’s OPm is much higher? Aren’t primary biomass burning and all anthropogenic SOA the dominant contributors to OPm based on the color shades in Fig 8?
Line 790-795. Aren’t the OPm differences and comparisons just due to measurements in different years which related to different stages of air quality mitigation implementation, instead of emission source differences in megacities vs median/small-sized cities?
Line 803-804. Is this statement based on higher correlations of OP with OA (fig S6)? But can high correlation necessarily lead to high importance? As you also stated in Line 543-554, Feb 11 (one day after Chinese New Year) exhibited similarly high OPm caused by transition metals during firework events.
Line 824-826. Considering the similar OPm in Fig 8 for Beijing and Weifang, is it still solid to say OP activity vary significantly across NPC region between Beijing and Weifang? Instead, would it mean OP can be very similar despite of different OA composition and origins?
Technical:
Line 68-70. It seems repetitive from the sentence in Line 62-64. You may consider keeping only one of them.
Line 151. What do you mean “near-molecular”?
Line 213-216. It seems repetitive from the sentence in Line 204-209. Please remove one of them.
Figure 1. why is the size area of the dots in panel c-e proportional to the “fourth” square root of corresponding integrated signals (Line 313)?
Also why are the data shown in panel c-e “in autumn” (Line 313) since in Line 321 the data from Wangdu are from January 2019?
Line 543. I think you meant fig 1a here instead of fig 1b.

Reply

Citation: https://doi.org/10.5194/egusphere-2026-4106-RC2

Jing Cai, Yongchun Liu, Feixue Zheng, Dian Ding, Wei Du, Liang Yuan, Yingjie Zou, Lazong Suolang, Rongxiang Lu, Wanwei Li, Chao Liu, Xiaolong Fan, Qiaozhi Zha, Mengxiao Luan, and Jiandong Wang

Supplement

https://doi.org/10.5194/egusphere-2026-4106-supplement

Jing Cai, Yongchun Liu, Feixue Zheng, Dian Ding, Wei Du, Liang Yuan, Yingjie Zou, Lazong Suolang, Rongxiang Lu, Wanwei Li, Chao Liu, Xiaolong Fan, Qiaozhi Zha, Mengxiao Luan, and Jiandong Wang

Viewed

Total article views: 248 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	Supplement	BibTeX	EndNote
168	61	19	248	34	18	20

HTML: 168
PDF: 61
XML: 19
Total: 248
Supplement: 34
BibTeX: 18
EndNote: 20

Views and downloads (calculated since 14 Jul 2026)

Month	HTML	PDF	XML	Total
Jul 2026	150	48	16	214
Aug 2026	18	13	3	34

Cumulative views and downloads (calculated since 14 Jul 2026)

Month	HTML	PDF	XML	Total
Jul 2026	150	48	16	214
Aug 2026	18	13	3	34

Viewed (geographical distribution)

Total article views: 178 (including HTML, PDF, and XML) Thereof 178 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 23 Aug 2026

Short summary

Characterizing organic aerosol (OA) molecular composition, sources and toxicity in the North China Plain remains difficult. Through comprehensive online and offline measurements in Weifang, a representative medium-sized city, we identify combustion-derived OA with a moderate oxidation state as the key drivers of oxidative potential (OP). These insights are critical for health risk assessments amidst the region’s ongoing reduction in particulate mass due to emission controls.


Total:	0
HTML:	0
PDF:	0
XML:	0

Combustion-Derived Organic Aerosols Enhance PM2.5 Oxidative Potential in a Medium-Sized North China Plain City

Supplement

Viewed

Viewed (geographical distribution)

Combustion-Derived Organic Aerosols Enhance PM_2.5 Oxidative Potential in a Medium-Sized North China Plain City