Next Article in Journal
A New Regional Background Atmospheric Station in the Yangtze River Delta Region for Carbon Monoxide: Assessment of Spatiotemporal Characteristics and Regional Significance
Previous Article in Journal
Field Measurements of Building Air-Conditioning Heat Rejection and the Thermal Environment in Urban Areas
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Atmosphere
Volume 16
Issue 1
10.3390/atmos16010098
atmosphere-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Seasonally Dependent Daytime and Nighttime Formation of Oxalic Acid Vapor and Particulate Oxalate in Tropical Coastal and Marine Atmospheres

by
Le Yan
1,†,
Yating Gao
2,†,
Dihui Chen
3,
Lei Sun
2,
Yang Gao
1,2,4,
Huiwang Gao
2,4 and
Xiaohong Yao
1,2,4,*
1
Sanya Oceanographic Institution, Ocean University of China, Sanya 572024, China
2
Frontiers Science Center for Deep Ocean Multispheres and Earth System, Ministry of Education, Ocean University of China, Qingdao 266100, China
3
School of Environmental and Chemical Engineering, Shanghai University, Shanghai 200444, China
4
Laboratory for Marine Ecology and Environmental Science, Qingdao Marine Science and Technology Center, Qingdao 266237, China
*
Author to whom correspondence should be addressed.
†
These authors contributed equally to this work.
Atmosphere 2025, 16(1), 98; https://doi.org/10.3390/atmos16010098
Submission received: 5 December 2024 / Revised: 11 January 2025 / Accepted: 14 January 2025 / Published: 17 January 2025
(This article belongs to the Section Aerosols)
Browse Figures
Versions Notes

Abstract

:
Oxalic acid is the most abundant low-molecular-weight dicarboxylic acid in the atmosphere, and it plays a crucial role in the formation of new particles and cloud condensation nuclei. However, most observational studies have focused on particulate oxalate, leaving a significant knowledge gap on oxalic acid vapor. This study investigated the concentrations and formation of oxalic acid vapor and oxalate in PM2.5 at a rural tropical coastal island site in south China across different seasons, based on semi-continuous measurements using an Ambient Ion Monitor-Ion Chromatograph (AIM-IC) system. We replaced the default 25 μL sampling loop on the AIM-IC with a 250 μL loop, improving the ability to distinguish the signal of oxalic acid vapor from noise. The data revealed clear seasonal patterns in the dependent daytime and nighttime formation of oxalic acid vapor, benefiting from high signal-to-noise ratios. Specifically, concentrations were 0.059 ± 0.15 μg m−3 in February and April 2023, exhibiting consistent diurnal variations similar to those of O3, likely driven by photochemical reactions. These values decreased to 0.021 ± 0.07 μg m−3 in November and December 2023, with higher nighttime concentrations likely related to dark chemistry processes, amplified by accumulation due to low mixing layer height. The concentrations of oxalate in PM2.5 were comparable to those of oxalic acid vapor, but exhibited (3–7)-day variations, superimposed on diurnal fluctuations to varying degrees. Additionally, thermodynamic equilibrium calculations were performed on the coastal data, and independent size distributions of particulate oxalate in the upwind marine atmosphere were analyzed to support the findings.

1. Introduction

The tropics receive the highest solar irradiance on Earth, making tropical oceans the most significant sources of atmospheric moisture [1,2,3]. Aerosol-cloud interactions in this region play a critical role in modulating the radiation budget, thereby influencing the global climate [3,4,5]. Compared to mid-latitude atmospheres, tropical marine and coastal environments are generally less polluted by anthropogenic emissions. Therefore, the oxidation of ocean-derived biogenic gasses, such as dimethyl sulfide and isoprene, serves as an important pathway for generating secondary aerosols and cloud condensation nuclei (CCN) [3,6,7,8,9,10]. Oxalic acid vapor, one of the major end-products of isoprene oxidation [11,12,13], has been reported to substantially enhance the formation of ocean-derived secondary particles, as demonstrated through field observations, laboratory experiments, and theoretical studies [6,11,14,15,16]. Additionally, the high hygroscopicity of oxalic acid enables it to influence CCN by undergoing gas-particle condensation and/or secondary formation on atmospheric particles, thus impacting radiative forcing [17,18,19]. Despite its importance, most studies in the literature focus primarily on particulate oxalate, leaving a substantial knowledge gap regarding ambient oxalic acid vapor [20,21,22,23,24,25,26].
The vapor pressure of oxalic acid remains a subject of debate in the literature [27,28]. For instance, a value of (2.2 ± 1.2) × 10−2 Pa at 298 °C was measured using a Knudsen Effusion Mass Spectrometer [27], while Paciga et al. [28] reported a substantially lower value of (1.7 ± 0.6) × 10−6 Pa at 298 °C in the presence of atmospheric NH3, measured by a tandem differential mobility analysis system coupled with a thermodenuder. This result is indirectly supported by an experimental study on the volatility and yield of glyoxal secondary organic aerosol [29]. Theoretically, the gas–aerosol partitioning of oxalic acid vapor is strongly influenced by the pH and liquid water content of ambient aerosols, with partitioning enhanced by increasing relative humidity (RH) [24,25,26]. The low volatility due to the reactions with ammonia appears to be a unique property of oxalic acid and may not apply to most other atmospheric diacids [28]. The reported lower vapor pressure of oxalic acid is comparable to that of sulfuric acid [28,30]. If accurate, this would suggest that molecular concentrations of oxalic acid vapor in ambient air are similar to those of sulfuric acid vapor, which is considered a quasi-steady-state species due to its rapid and strong condensation sinks, and would not exceed (1−2) × 107 molecules cm−3 [31]. However, this hypothesis is inconsistent with reported field measurements [20,21,22,24,25,26]. Furthermore, the field measurements are also contentious due to the widespread use of wet denuder techniques involving the addition of H2O2 solutions to sample acidic species. Precursors such as glyoxylic acid and others can be rapidly oxidized to oxalic acid in the aqueous phase via in-cloud processing, especially given their highly effective Henry’s constants [11,32,33,34]. Consequently, oxalic acid vapor measured by H2O2-solution-added wet denuder techniques physically includes both oxalic acid vapor and that oxidized from its gaseous precursors in the solution, which is enriched in H2O2 and other oxidants. This mixture is referred to as oxalic-acid-vapor* in this study. The controversies complicate the analysis of the presence and abundance of oxalic acid vapor in ambient air, as well as the associated gas–aerosol partitioning and the formation mechanisms of both oxalic acid vapor and particulate oxalate.
In atmospheric chemistry, the formation of particulate oxalate is of significant interest because oxalic acid is the most abundant dicarboxylic acid found in atmospheric particles [11,35,36,37,38,39]. However, for most organic compounds in atmospheric particles, only functional groups can be identified and not individual species [13,40,41,42]. Over the past decades, significant efforts have focused on investigating the primary and secondary sources, as well as the formation pathways, of particulate oxalate in various atmospheres [32,34,35,36,37]. For instance, Carlton et al. [32] and Rinaldi et al. [34] reported that glyoxal, through aqueous-phase oxidation during in-cloud processing, is a potential precursor of oxalate and is considered an important source of submicron oxalate in the remote marine atmosphere. In fact, a good correlation between oxalate and sulfate in atmospheric particles has been widely observed and used to support the argument that both species undergo in-cloud processing. This is because (1) their respective precursors generally originate from different major sources, and (2) particulate sulfate in droplet mode is primarily derived from in-cloud processing [11,35,43,44]. The in-cloud processing of both species is further corroborated by their consistent size distributions observed in various atmospheres [11,20,43,44]. Moreover, Hoque et al. [38] proposed that the high concentrations of oxalate in atmospheric particles over the western North Pacific were associated with continental outflows, attributing their abundance to the aqueous-phase photo-oxidation of volatile organic carbons (VOCs) from the continent. Mochizuki et al. [12] found the oxidation of biogenic isoprene and α-pinene by O3 and other oxidants leads to the formation of particulate oxalate in a remote forest atmosphere. However, the studies mentioned above are limited by the lack of simultaneous measurements of both oxalic acid vapor and particulate oxalate.
To address these controversies and fill knowledge gaps related to oxalic acid vapor and particulate oxalate, this study focuses on less-polluted tropical coastal and marine atmospheres over several weeks during both wet and dry seasons of 2023. The strong ultraviolet (UV) radiation and high ambient temperatures in these regions favor significant photochemical formation of both oxalic acid vapor and associated particulate oxalate. In these relatively clean environments, secondary sources of oxalic acid and particulate oxalate typically dominate over primary sources [11], reducing the complexity of mixing various origins with varying formation mechanisms. For instrumentation, we employed an Ambient Ion Monitor-Ion Chromatograph (AIM-IC) system (Thermo Fisher, Waltham, MA, USA)with a 250 μL injection loop to semi-continuously measure gaseous oxalic acid vapor and particulate oxalate in PM2.5, achieving high signal-to-noise ratios. Notably, we switched from H2O2 solution to pure deionized water for sampling oxalic acid vapor on certain occasions. For modeling, we utilized the Extended Aerosol Inorganics Model (E-AIM) (https://www.aim.env.uea.ac.uk/aim/phpmain/select_compounds.php, accessed on 10 January 2025) ) to calculate the equilibrium concentrations of compounds in both the gas and particle phases [45]. Through this approach, we fill the knowledge gap regarding the gas–aerosol partitioning of oxalic acid vapor and particulate oxalate, while quantifying the measured gaseous oxalic acid vapor and its precursors. Considering the substantially lower ratios of oxalic acid vapor to particulate oxalate observed in April than in other months, we also included size-segregated atmospheric particle samples collected by a 14-stage Micro-Orifice Uniform Deposition Impactor (MOUDI) over the adjacent tropic sea (the South China Sea, SCS) in April 2017 to facilitate the analysis of particulate oxalate during that month. The SCS is located upwind of the coastal sampling site, regardless of the northeast and southeast monsoon, and the particulate oxalate in this region may represent antecedent particulate oxalate before it reaches the coastal site. The objectives of this study are (1) to characterize the concentration of oxalic acid vapor in the mixture with its precursors in tropical coastal atmospheres; (2) and to investigate the gas–aerosol partitioning of oxalic acid vapor and particulate oxalate, and to reevaluate the vapor pressure values used in thermodynamic models.

2. Methodology

The sampling site is located in a tropical high-tech park in the southeast corner of Hainan Island (18.328 °N, 109.169 °E, Figure 1a–e). This park supports research and development in deep-sea equipment and postgraduate education. Approximately 2 km to the northwest lies a national seed breeding base, which is expected to emit biogenic VOCs from agricultural activities, though no public data on emissions are available. The park is also about 1.9 km from the South China Sea, and both southeast and northeast monsoon winds, along with sea–land breezes, transport sea-salt aerosols and sea-derived VOCs to the site. To the south, the park is located ~1.2 km from a small hill with an elevation of 487 m. During the wet season (March to October), low clouds typically encounter this hill each morning, as documented in our regular photo records (Figure 1f). However, the frequency decreases somewhat in the dry season (November to February) because of the increasing atmospheric mixing layer height (MLH) (Figure S1).
The high-tech park is still under development, with large areas of construction land left vacant to the south and north. A major traffic road is located ~50 m from the site. The peak traffic flow at the site during weekdays is 810 vehicles per hour, with ~1/4 of the vehicles being zero-emission. Traffic flow begins to increase around 06:30–07:00, peaks at 09:00–09:30, and then slightly decreases. Traffic-related VOC emissions are expected to have a minimal impact. The park is located in Sanya City, Hainan Province, which boasts some of the best air quality in China. In 2021, the city’s annual averages for PM2.5, SO2, NO2 and CO were 12 µg m−3, 4 µg m−3, 8 µg m−3 and 0.51 mg m−3, respectively [46]. Due to the relatively low industrial activities and air pollutant emissions in the park, even better air quality can be anticipated compared to the city averages. Local air quality data from monitoring stations were also retrieved for correlation analysis in this study (https://aqicn.org/city/sanya/, accessed on 10 January 2025).
The AIM-IC system was housed in an air-conditioned research lab on the third floor of a research building in the park. A stainless-steel sampling probe, 2.5 m long and 3.5 cm in diameter, extended from the window to connect with ambient air. The sampling probe was positioned ~10 m above ground level. Measurements began in November 2022, with regular maintenance and calibration, but were intermittently interrupted due to the COVID-19 situation. Here, data from weekly observations in February and April 2023 (Period 1), as well as continuous observations from November to December 2023 (Period 2), were used for analysis. Details on the operation of the AIM-IC system can be found in our previous studies [7,47]. Briefly, the AIM-IC, equipped with a PM2.5 cyclone, measured semi-continuous concentrations of atmospheric gasses, such as oxalic acid, NH3, amines, SO2, HNO3, etc., and water-soluble ions in PM2.5 at a time resolution of one hour. A wet denuder with 0.02% H2O2 solution added was installed to absorb water-soluble gasses during Period 1 and from 17 November to 13 December in Period 2, although it also absorbed oxidants. To address this, the setup was switched to a pure-H2O-added wet denuder from 13 to 24 December during Period 2. To improve the signal-to-noise ratio, the system’s default 25 μL sampling loop was replaced with a 250 μL loop [48]. The ion chromatography (ICS-1100, Thermo Fisher, Waltham, MA, USA) of the AIM-IC was equipped with two analytical columns (Ion Pac CS20 (2 × 250 mm) for cations and Ion Pac AS11-HC (2 × 250 mm) for anions) and two guard columns (CG20 (2 × 50 mm) for cations and AG11-HC (2 × 50 mm) for anions). IC calibration with known concentrations of standard solution was conducted before and after each observation, and the R2 values of the calibration curves were 0.985 ± 0.013 for NH4+, 0.998 ± 0.003 for Cl−, 1.000 ± 0.0004 for SO42− 0.996 ± 0.007 for NO3− and 0.952 ± 0.036 for oxalate, respectively (Figure S2). Detection limits for oxalic acid and other ions ranged from 2 to 4 ng m−3 in ambient air. The uncertainties for ions detected by the IC system, arising from incomplete injections, varying separation efficiency caused by occasional clogs, and the inherent white noise of the electric conductivity detector, were assessed by repeatedly injecting identical concentration standards. The uncertainties for NH4+, NO3− and SO42− were within 5% when multiple injection solutions at different concentrations were tested. The uncertainties for oxalate and low-concentration ions were within 10% for their concentrations greater than 0.1 µg m−3 in ambient air. The uncertainties largely increased to 40% with concentrations in ambient air below 50 ng m−3. Both molecular and mass concentrations of oxalic-acid-vapor* were used in this study for comparison with related variables. As no standard method for generating oxalic acid vapor at ambient conditions is available in the research community, the collection efficiency of the vapor was not directly evaluated. However, a test for the collection efficiency of another sticky gas, atmospheric NH3, was conducted using the sampling probe, and the sampling loss was found to be negligible (Figure S3).
Note that the AIM-IC consists of two independent systems, i.e., a sampling system and an analytic system. The sampling system can be disconnected or shut down for calibration. In this case, the analytical system functions as a standard ion chromatograph (IC), which can be used for standard tests and offline sample analysis. The sample loop on the low-pressure injection valve is connected to the analytical system, and its volume is adjustable, e.g., ranging from 25 µL to 1000 µL. The credibility and accuracy of the IC measurements can be confirmed using the standard curve during regular system calibration, with no additional validation required. However, the frequency of incomplete injections increases when the volume exceeds 250 µL. In addition, the use of pure H2O instead of a 0.02% H2O2 solution does not affect the gas collection efficiency of the wet denuder. The efficiency of the wet denuder is primarily determined by the sampling air flow rate, the surface area of the denuder, and the velocity of the liquid film on the denuder surface. No additional validation is needed for using pure H2O instead of a 0.02% H2O2 solution, at least to our understanding of the AIM-IC. However, the use of pure H2O or a 0.02% H2O2 solution may influence the conversion of chemical species, e.g., from (HSO3− + SO32−) to SO42−, in the collected water pool, as a one-hour collection duration is used for each collection.
As in our previous studies [7], the Extended Aerosol Inorganics Model (E-AIM) (https://www.aim.env.uea.ac.uk/aim/aim.php, accessed on 10 January 2025) was employed to calculate the equilibrium concentration of compounds in both gas and particle phases [45]. The E-AIM model is currently the only model in the research community capable of simulating gas–aerosol equilibria for both inorganic and organic species. Like all gas–aerosol equilibrium thermodynamic models, E-AIM operates under two key assumptions: (1) that gas–aerosol equilibrium has been reached; (2) that the chemical components are internally mixed. Gas-particle equilibrium for NH3, HNO3, H2SO4 and oxalic acid vapor involved in the E-AIM model is listed in Text S1. Moreover, meteorological data were obtained from the China Meteorological Data Service Centre (https://data.cma.cn/, accessed on 10 January 2025). The ambient temperatures during the study were 24 ± 3 °C, 28 ± 2 °C, 26 ± 3 °C, 24 ± 3 °C (average ± standard deviation) in February, April, November and December 2023, respectively. The corresponding monthly averages for relative humidity and cloud coverage were 70 ± 12% and 4 ± 2, 78 ± 8% and 4 ± 2, 73 ± 13% and 3 ± 2, 71 ± 11% and 4 ± 2, respectively. Air mass back trajectories were modeled using the Hysplit model from the NOAA Air Resources Laboratory (https://www.ready.noaa.gov/HYSPLIT.php, accessed on 10 January 2025). Simulated heights were set to 100, 500, and 1000 m on the basis of the observed low mixing layer height (Figure S1), with a 24 h simulation time. Analysis was conducted at 0:00, 6:00, 12:00 and 18:00 each day.
Size-segregated atmospheric particles were collected during a cruise campaign over the SCS from 30 March to 1 May 2017 [49]. The average ambient temperature and relative humidity during the campaign were 28 ± 2 °C and 79 ± 6%, respectively. A total of nine sets of size-segregated samples, along with three additional sets of blank samples, were collected during the campaign. The duration of each sampling ranged from approximately 3 to 38 h, depending on weather conditions. All sample filters were wrapped in pre-combusted aluminum foils. For the measurements, size-segregated atmospheric particles were collected using a Nano MOUDI-II, operated at an airflow rate of 29.4 L min−1. Each set of samples consisted of 11 Teflon filters with cut sizes of 18, 10, 5.6, 3.2, 1.8, 1.0, 0.56, 0.32, 0.18, 0.10, and 0.056 μm and 3 Zefluor filters with cut sizes of 0.032, 0.018, and 0.010 μm. The Zefluor filter samples were excluded from the analysis in this study. The extraction and chemical analysis of ions in the samples by ion chromatography are described in detail in Guo et al. [43].
The vessel management system (VMS) recorded meteorological data, including wind speed, ambient temperature, and relative humidity. The wind speed was corrected by subtracting the effect of the vessel’s speed. UV radiation data over the SCS during the measurement period were obtained from (http://www.temis.nl/uvradiation, accessed on 10 January 2025). Daily chlorophyll-a (Chl-a) observations were obtained as L3 data from the Moderate Resolution Imaging Spectroradiometer (MODIS) onboard NASA’s EOS-Aqua satellite. The long-term seasonal averages of Chl-a in the SCS from 2002 to 2017 were analyzed by Yu et al. [50].
Additionally, ChatGPT-4 has been utilized for English language polishing in this manuscript. We trained ChatGPT-4 using review papers to learn appropriate expression styles, and subsequently refined the manuscript to meet the desired standards of scientific accuracy and presentation.

3. Results and Discussion

3.1. Overview of Oxalic-Acid-Vapor* and Particulate Oxalate in the Coastal Atmosphere

Figure 2a,b showed the hourly averaged molecular concentrations of oxalic-acid-vapor* and mass concentrations of particulate oxalate in PM2.5 during Period 1 and Period 2. The concentrations of oxalic-acid-vapor* were (4.4 ± 1.0) × 108 molecules cm−3 in February and (3.5 ± 0.7) × 108 molecules cm−3 in April during Period 1. Although the difference between the two average values was small, a significant difference in oxalic-acid-vapor* concentrations between February and April was observed through a t-test (p < 0.05). The concentrations decreased to (1.5 ± 0.5) × 108 molecules cm−3 during Period 2, with a significant difference compared to those in February and April during Period 1 (p < 0.01). In contrast, the coefficient of variation (CV) for oxalic-acid-vapor* concentrations slightly increased during Period 2 relative to Period 1. All these molecule concentrations of oxalic-acid-vapor* were 1–2 orders of magnitude higher than those of sulfuric acid vapor reported in ambient air [30,31], suggesting that either oxalic acid vapor is substantially more volatile than sulfuric acid regardless of atmospheric NH3, or the true concentration of oxalic acid vapor is lower than the detected value.
For particulate oxalate in PM2.5 (Figure 2a,b), the mass concentrations were 0.05 ± 0.03 μg m−3 in February and 0.07 ± 0.03 μg m3 in April during Period 1, with a significant difference between the two (p < 0.05). During Period 2, the mass concentrations of oxalate in PM2.5 were 0.03 ± 0.02 μg m−3, showing a significant difference from Period 1 (p < 0.01). When the values measured simultaneously in both phases were compared, the average mass concentration of oxalic-acid-vapor* was almost equal to that of oxalate in PM2.5 during Period 1 (0.059 μg m−3 vs. 0.057 μg m−3). During Period 2, the two averages were also comparable, with 0.021 μg m−3 for oxalic-acid-vapor* and 0.029 μg m−3 for oxalate in PM2.5. Section 3.3 will examine whether the ratios of oxalic-acid-vapor* to oxalate in PM2.5 are theoretically supported by thermodynamic equilibrium considerations.
The average concentrations of oxalic-acid-vapor* and oxalate in PM2.5 for both periods are listed in Table 1 and compared with those from other studies. The observed average concentration of oxalic-acid-vapor* during Period 1 was slightly lower than the 0.074 μg m−3 measured at a semi-urban site at Sydney, FL, from 27 April to 31 May 2002 [21], but nearly double the value observed in Yorkville, Georgia, from mid-August to mid-October 2016 [24,25]. In contrast, the observed average concentration of oxalate in PM2.5 during Period 1 was only approximately 20% of the average 0.29 μg m−3 measured in Sydney. The average concentration of oxalic-acid-vapor* during Period 2 was close to the value of 22 ± 8 ng m−3 measured at a rural site in Shanghai, China during the cold season from 8 November to 2 December 2020, using an identical AIM-IC system [22]. However, Yao et al. [22] reported oxalate concentrations in PM2.5 as high as 475 ± 266 ng m−3. Yao et al. measured oxalic-acid-vapor* at an urban site in Hong Kong using Na2CO3-coated dry denuders in December 2000 [20]. They reported only four values in the range of 0.02–0.05 μg m−3, with the rest of the measurements being below twice the field blank. The Na2CO3-coated dry denuder likely absorbed fewer gaseous precursors of oxalic acid and oxidants in comparison with the wet denuder used in the studies above, although no direct comparison test has been conducted. Compared to observations in other studies, the mass ratio of oxalic-acid-vapor* to oxalate in PM2.5 during both periods was substantially higher, leading to the speculation that either a high vapor pressure of oxalic acid in the tropical coastal atmosphere or a mixture oxalic acid vapor and its precursors was detected.
Periodic analysis of the two observed variables in Figure 2a,b revealed that the mass concentrations of oxalate in PM2.5 exhibited short-period variations (3–7 days), superimposed by higher-frequency diurnal variations in different extents during both Period 1 and Period 2. The (3–7)-day variations likely corresponded to planetary waves in the northern hemisphere, reflecting long-range transport of oxalate in PM2.5. For example, the strongest diurnal variations occurred during 18–22 April 2023, coinciding with a 5-day event (Figure 2a,b and Figure S4a). In contrast, much weaker diurnal variations were observed during 11–13 February and 18–23 February 2023, with other periods falling between these extremes. The concentrations of oxalic-acid-vapor*, however, exhibited distinct diurnal fluctuations (Figure 2a,b). During Period 1, the concentrations of oxalic-acid-vapor* peaked in the afternoon, in line with the O3 concentrations shown in Figure S4c. In Period 2, however, the highest concentrations of oxalic-acid-vapor* occurred later in the evening, except for an increase during the morning rush hours, corresponding to a trough in O3 levels and a slight rise in PM2.5 (Figure S4b,d). The lower mixing boundary layer in the morning favored the accumulation of air pollutants, enhancing the rush-hour effect. However, O3 concentrations still reached their peak later in the afternoon during Period 2 (Figure S4d). Further discussion of the diurnal fluctuations of oxalic-acid-vapor* and oxalate in PM2.5 will be provided in Section 3.2.
When analyzing the correlation between oxalic-acid-vapor* and particulate oxalate in PM2.5 mass concentration, a significant negative correlation (p < 0.01) was obtained during Period 1 (Figure 2c, black markers, and the black regression curve). This implied a source–sink relationship between them, which is further supported by the time delay in the peak concentration of oxalate in PM2.5 relative to that of oxalic-acid-vapor* in the diurnal variation curve (Figure S4a). In a closed system, the depletion of reactants theoretically leads to the enrichment of products due to the Law of Mass Conservation. According to this law, a perfect negative correlation should exist between the reactant and product in a unity stoichiometric ratio. However, external inputs of either reactants or products can eliminate this relationship, potentially switching it to a positive correlation. For oxalic acid vapor and its particulate counterpart in the tropical atmosphere, the gas–aerosol equilibrium time scale is expected to be on the order of minutes. Given that the resolution of the AIM-IC is one hour, the hourly averaged values should not be affected by the time scale of the gas–aerosol equilibrium. However, the correlation observed was relatively weak (R2 < 0.1), which could be attributed to several factors, e.g., (1) the presence of oxidizable precursors in oxalic-acid-vapor*; (2) fluctuating external inputs of particulate oxalate over the 3–7 day periods; and (3) secondary formation of oxalate in PM2.5 from its precursors, etc. Notably, a significant positive correlation between them was observed during certain short periods, such as 20–23 April 2023, when the secondary formation of oxalate in PM2.5 may have occurred simultaneously with that of oxalic-acid-vapor* (Figure 2c, red markers).
During Period 2, a significant positive correlation was obtained between oxalic-acid-vapor* and particulate oxalate in PM2.5, with a low R2 value (Figure 2d). These low R2 values occurred throughout Period 2, even during the period with the strongest diurnal variations in oxalate in PM2.5 (Figure 2d, red markers). Considering the negative correlation between oxalic-acid-vapor* and oxalate during Period 1, particulate oxalate may have acted as a condensation sink of oxalic acid vapor. The lack of negative correlation during Period 2 could be attributed to more freshly oxalic-acid-vapor* being formed. In other words, the formation effect overwhelmed the condensation effect during Period 2. Again, this implied that oxalic acid has a high vapor pressure and does not exist in a quasi-steady state. Theoretically, an increase in the source term relative to the sink term would strengthen the positive correlation. This positive correlation could also be explained by: (1) the formation rates of oxalic-acid-vapor* being the determining step for both species; (2) the inherent limitations of Eulerian observation at the fixed site, where oxalic-acid-vapor* and oxalate in PM2.5 might be affected by confounding factors, regardless of their origins or formation pathways. Unfortunately, tracking moving atmospheric air parcels to obtain the concentrations of either variable is practically impossible due to irregular air movement, limitation of measurement platforms, constraints of sampling area and time, etc.

3.2. Daytime and Nighttime Formation of Oxalic-Acid-Vapor* and Particulate Oxalate

Based on the distinguishable diurnal variations in oxalic-acid-vapor* concentrations during Period 1 and its positive correlation with O3 (Figures S4a,c and S5), it can be reasonably inferred that oxalic-acid-vapor* is primarily produced through daytime photochemical reactions [11,51,52,53]. The lack of a positive response of oxalic-acid-vapor* to most (3–7)-day variations in oxalate in PM2.5 suggested that long-range transport of oxalic-acid-vapor* and its precursors made only a negligible contribution to the observed oxalic-acid-vapor*. For example, the lower concentrations of oxalate in PM2.5 were observed, with higher concentrations of oxalic-acid-vapor* during 10–13 February compared to other moments during Period 1. In addition, the increase in oxalic-acid-vapor* during morning rush hours implied a significant primary source for it [11] (Figure S4a). In fact, the primary source of oxalic-acid-vapor* in morning rush hours was also identifiable during Period 2 (p < 0.05, Figure S4b). However, the higher concentrations of oxalic-acid-vapor* occurred at nighttime during Period 2 (Figure 2d and Figure S4b), which contrasted with the even larger maximum concentrations of O3 occurring in the afternoon than Period 1 (Figure S4c,d). Figures S6–S8 show that the strong northeast monsoon winds blew from the sea during the daytime in Period 2, sweeping precursors of oxalic acid vapor emitted from local continental sources. While at night, the mixing layer height decreased by ~50%, allowing for the accumulation of local precursors and facilitating their chemical conversion into oxalic-acid-vapor* through dark chemistry [11]. The role of dark chemistry in this process warrants further investigation.
During Period 1, the (3–7)-day variations in oxalate concentrations in PM2.5 were likely driven by the long-range transport, as the air masses could have traveled over thousands of kilometers (Figures S6–S8). The superimposed diurnal variations in oxalate in PM2.5 were more likely attributed to the condensation of oxalic-acid-vapor* from local precursors through photochemical reactions, which is supported by the negative correlation and time delay between oxalic-acid-vapor* and oxalate in PM2.5, as mentioned earlier. When the diurnal variations in oxalate in PM2.5 were negligible, the concentrations of oxalate in PM2.5 were found to be highly correlated with SO42− concentrations in PM2.5, e.g., during 18–22 February 2023 (Figure 3a, blue markers). Following the logic from previous studies [11,35,43,44], this strong correlation highly indicated in-cloud processing of particulate oxalate, although it is unclear whether this process predominantly occurred during the daytime or nighttime. However, during periods of strong photochemical formation of oxalic acid vapor, followed by condensation on PM2.5 (e.g., 20–23 April), this significant but weak correlation between oxalate and SO42− in PM2.5 existed for the whole of Period 1 was even disrupted (Figure 3a).
During Period 2, a moderate correlation between oxalate and SO42− in PM2.5 also suggested that in-cloud processing of particulate oxalate was the dominant formation pathway. The percentage increase in oxalate concentrations at nighttime was notably larger than that of the total mass in PM2.5 (Figure S4b,d), which can be attributed to dark chemistry conversion, compounded by the low mixing layer height that facilitated the accumulation (Figure S4b). In addition, the strong northeast monsoon normally dominate the weakened sea–land breezes during this season (Figures S6–S8a). As with Period 1, the (3–7)-day variations in oxalate concentrations in PM2.5 during Period 2 were likely influenced by either daytime or nighttime in-cloud processing of particulate oxalate during long-range transport. Moreover, when we correlated the concentrations of oxalate in PM2.5 with ambient RH (Figure S9), no positive correlation was observed as Yang et al. [37] suggested, e.g., lower concentrations of particulate oxalate corresponding to higher RH (11–14 February), in contrast with higher concentrations corresponding to lower RH (21 February). This lack of correlation implied that aerosol droplet chemistry was not a major factor in the formation of particulate oxalate, at least in this study. Further analysis correlating the concentrations of oxalate in PM2.5 + oxalic acid vapor* with those of SO42− in PM2.5 (Figure S10) showed that the correlations were expectedly weaker during Period 1, as oxalic-acid-vapor* was likely generated via photochemical reactions in the gas phase. However, during Period 2, the determination coefficient remained nearly the same, with a doubled slope, indicating that oxalic-acid-vapor* might have been evaporated from the particulate phase and generated through a similar process as particulate oxalate during Period 2.

3.3. Thermodynamic Examination for True Oxalic Acid Vapor and Its Volatility

The E-AIM model was further employed to examine the gas–aerosol partitioning of oxalic acid species using the full ambient data during the study period (Table S1). For approximately 60% of the observational data (883 h), the model predicted that over 95% of (oxalic-acid-vapor + particulate oxalate) would be partitioned into the gas phase, leading to unpractically high partitioning ratio of particle phase. The reason will be analyzed later. For the remaining ~40% of the observational data (495 h), the model predicted that (oxalic-acid-vapor + particulate oxalate) were more evenly distributed between the two phases. The parts of the equilibrated concentrations of oxalate in PM2.5 predicted by the model in different periods were plotted against the observed concentrations, as shown in Figure 4a–d, and the differences between predicted oxalic-acid-vapor concentrations and the observed values were also plotted against the observed concentrations, as shown in Figure 4e–h. When considering a 10% analytical error for all chemical inputs, only a small fraction (~9%) of the equilibrated concentrations of oxalate in PM2.5 were consistent with the observed values within a 20% margin along the 1:1 line during different periods (Figure 4a–d). The consistency indicated that (1) gas–aerosol equilibrium has been achieved; (2) for certain cases in the coastal atmosphere, the vapor pressure of oxalic acid reported by Booth et al. [27] is applicable.
However, in ~85% of cases, a significant discrepancy was observed, where the equilibrated concentrations of oxalate in PM2.5 were over 20% larger than the observed values. Considering that the estimated short timescale for partitioning inorganic and organic volatility compounds into liquid particles in hot ambient air ranges from seconds to minutes [54,55], the gas–aerosol equilibrium should be reached. Notably, the equilibrated concentrations of oxalic acid vapor were lower than the observations in these cases. The deviation indicated that the observed oxalic-acid-vapor* might include a fraction of precursors to oxalic acid, rather than only the oxalic acid itself. The maximum deviation was observed in February 2023, when episodically high concentrations of oxalic-acid-vapor* were detected. By assuming that the overestimation of equilibrated oxalate concentrations in PM2.5 (more than 20%) was entirely due to the presence of oxalic-acid-vapor precursors, the contributions of the precursors to oxalic acid vapor were estimated to be approximately 69%, 57%, 73% and 70% for February, April, and before and after 15:00 on 13 December during Period 2, respectively. These results were partially supported by additional observations after 15:00 on 13 December 2023, where the observed oxalic-acid-vapor* concentrations, measured with pure H2O in the wet denuder, were (1.0 ± 0.3) × 108 molecules cm−3 and significantly lower than (1.6 ± 0.4) × 108 molecules cm−3 measured during other times in Period 2 (p < 0.01). Assuming that the difference in the average values was completely due to the presence of non-oxalic-acid-vapor (its precursors), it was estimated that these precursors accounted for approximately 1/3 of the observed oxalic-acid-vapor* before 15:00 on 13 December during Period 2. The estimated concentrations of oxalic acid vapor precursors may be lower than those in the actual ambient air, as the pure-H2O-added wet denuder can still absorb oxidants and precursors from the sampled ambient air, leading to the aqueous conversion of these precursors into oxalic acid. In the future, a reductant will be tested during specific periods to further investigate potential artifacts.
The equilibrated concentrations of oxalate in PM2.5 were more than 20% lower than observed in 6% of cases, where the modeled liquid water content (LWC) was generally less than 1 μg m−3. This underestimation strongly indicated the vapor pressure of ammonium oxalate, or a combination of ammonium oxalate and oxalic acid vapor pressures, should be used in place of oxalic acid alone for predicting the solid phase. A similar conclusion may apply to approximately 60% of cases, where the model predicted that over 95% of the (oxalic-acid-vapor + particulate oxalate) were partitioned in the gas phase. We analyzed the inputs and outputs for ~60% versus ~40% observational cases in Table S1, with the major difference being lower RH in the ~60% cases. However, the potential limitations of the E-AIM model in simulating dry aerosols cannot be ruled out.
When the modeled ratios of oxalic acid to particulate oxalate were plotted against aerosol pH in Figure 5a–d, the data with modeled LWC < 1 μg m−3 showed considerable scattered. Excluding these data, the ratios increased as pH decreased, ranging from < 0.01 to 1.0, with a median value of 0.3 in February 2023. The corresponding values for April were <0.01, 4.1 and 0.4; for November and December, they were <0.01, 2.1 and 0.2; and for December, with pure H2O used in wet denuder, the values were 0.01, 1.0 and 0.2. When comparing the modeled ratios to the corresponding observational values, the modeled ratios were consistently larger with the modeled LWC < 1 μg m−3. Again, the vapor pressure of ammonium oxalate, rather than oxalic acid, should be used in such cases [27,28]. For the cases with modeled LWC ≧ 1 μg m−3, smaller ratios were generally obtained, possibly due to artifacts related to the collected oxalic acid vapor precursors. Interestingly, the observed ratios in April 2023 were more narrowly distributed in the low-value range of 0.4–1.4, with a median value of 0.8, compared to other months, regardless of black markers or color markers in Figure 5f. It is noteworthy that the photochemical formation of oxalic acid vapor was significantly stronger in April than in November and December. This suggested a stronger source of particulate oxalate during April, which will be further explored in the next section.

3.4. Size Distributions of Oxalate in Atmospheric Particles over the SCS

To investigate the notably lower ratios of oxalic acid vapor to particulate oxalate observed in April than in other months, we further analyzed the formation of particulate oxalate in the upwind marine atmosphere using historical data from April of a different year. Figure 6 shows the mass size contributions of water-extracted oxalate in atmospheric particles over the SCS in April 2017. Although the mass concentrations of oxalate generally exhibited a bimodal distribution across particle sizes, the contribution of supermicron oxalate to the total particulate mass concentration varied significantly between samples. Specifically, supermicron oxalate exceeded submicron oxalate on the dates 2–6, 7–10, 15–17, 18 and 23–25 April 2017, with this distribution primarily occurring south of 15 °N latitude (Figure 6a,h). Conversely, the reverse was observed on other dates (Figure 6b,h). To investigate the origins and formation pathways of both submicron and supermicron oxalates, whether anthropogenic or natural, we performed a comparative analysis involving particulate non-sea-salt-SO42− (nss-SO42−), dimethylaminium (DMA+), Na+, NO3− and nss-K+ (Figure 6c–g).
The primary anthropogenic sources of particulate oxalate reportedly include emissions from vessel combustion [11]. To examine the impact of the vessel combustion emissions, samples were collected while the vessel was moored with its engines running on 15–17 and 18 April. Extremely high concentrations of nss-SO42−, apparently resulting from vessel combustion, were observed, predominantly in the size range below 0.32 µm (marked with blue inverted triangles and diamonds in Figure 6c). The contribution of nss-SO42− in the size ranges above 0.32 µm in the two moored samples were negligible, as these concentrations were lower than those observed in some size bins from other samples unaffected by vessel combustion emissions (Figure 6c). In contrast, an opposite pattern was observed for oxalate, with negligible amounts in the size range below 0.32 µm in the two moored samples, indicating that the contribution of the vessel combustion emissions to oxalate levels was minimal relative to the oxalate measured in this study. Similarly to the high concentrations nss-SO42−, DMA+ was also predominantly distributed in the size range below 0.32 µm (Figure 6d). DMA+ likely neutralized particulate inorganic and organic acids through gas-particle condensation [7,8,9]. In addition, heterogeneous conversions of glycolaldehyde and glyoxal, etc., on sea-salt or crustal aerosols, potentially enhanced by the photochemical degradation of NO3− to generate OH free radical in deliquesced sea-salt aerosol droplets under extremely strong UV radiation [56] (Figure S11b), could theoretically produce particulate oxalate in a supermicron mode, similar to NO3−, with a mode ranging between 2 and 6 µm [11]. However, this was not observed in this study. The observed particulate oxalate over the SCS was dominantly distributed in the 0.5–2.0 µm range, suggesting it was primarily formed through in-cloud processing of oxalate, i.e., 0.5–1.0 µm via mid-high cloud and 1.0–2.0 µm via low cloud (or fog) [32,34,35,43,57].
Rinaldi et al. [34] reported a similar finding of a supermicron mode of oxalate at 1–3 µm in coastal atmospheres, attributing it to the oxidation of natural marine precursors. A similar bi-modal pattern of particulate oxalate was also reported in marine atmospheres over the northwest Pacific Ocean [43,58]. As shown in Figure S11a, most oceanic zones over the SCS are part of the ocean desert zone, characterized by low chl-a concentrations and limited marine biogenic emissions. This restriction in oceanic biogenic VOC emissions to specific hotspots during that month (Figures S6–S8) is making locally emitted VOCs the primary precursors of oxalic-acid-vapor* and causing its strong diurnal variation in relation to O3 levels. Moreover, the absence of a supermicron mode of nss-SO42− with oxalate at 1–3 µm might explain the lack of a significant correlation between oxalate and SO42− in PM2.5 observed at the coastal site in April 2023. Whether the supermicron oxalate mode is related to fog processing of sea-spray organic aerosols requires further investigation. However, the strong supermicron mode of oxalate likely contributed to the low ratios of oxalic acid vapor to particulate oxalate observed in that month. In terms of natural sources, biomass burning has been considered a significant contributor to oxalate in marine atmospheric particles [58,59]. However, in this study, biomass burning appeared to be a negligible source of particulate oxalate over the SCS, as indicated by the low concentrations of nss-K+. The analysis over the SCS strongly supported the hypothesis that the observed (3–7)-day variations in particulate oxalate at the coastal island site were related to the in-cloud formation of oxalate in the upwind marine atmosphere, despite the fact that the former and latter observations were conducted at different times.

4. Summary and Atmospheric Application

This study accurately measured the concentrations and formation of oxalic-acid-vapor* and oxalate in PM2.5 at a rural tropical coastal island site in south China covering different seasons. The measurements were conducted using a large-volume injection loop integrated with the AIM-IC. The measurement results showed seasonally dependent daytime and nighttime patterns in the formation of oxalic-acid-vapor*. In February and April 2023, the measured concentrations of oxalic-acid-vapor* were 0.059 ± 0.15 μg m−3, exhibiting a consistent diurnal variation with that of O3, likely driven by photochemical reactions. In contrast, significantly lower concentrations of 0.021 ± 0.07 μg m−3 were observed for the vapor in November and December 2023. The vapor concentrations increased at night and were likely related to dark chemistry processes, amplified by the accumulation effect from a low mixing layer height. The concentrations of oxalate in PM2.5 were comparable to those of oxalic acid vapor in the two seasons, but exhibited (3–7)-day variations, superimposed on diurnal variations to different extents. The (3–7)-day variations in oxalate concentrations were likely due to long-range transport and in-cloud processing, as indicated by a good correlation with SO42− in PM2.5. The photochemical formation of oxalic-acid-vapor* in diurnal cycles, followed by condensation, largely reduced the correlation. However, the nighttime formation of oxalic-acid-vapor* had no detectable influence on the correlation, suggesting the evaporation of oxalic acid from in-cloud processing of oxalate aerosols as a potential nighttime source.
Additionally, the thermodynamic equilibrium calculations suggested that the gas–aerosol equilibrium was reasonably achieved in the tropical coastal atmosphere. However, appreciable positive sampling artifacts due to the oxidation of precursors of oxalic acid might exist in the observed concentrations of oxalic acid, leading to the overestimation of particulate oxalate in some cases. The modeling results strongly suggest that the gas–aerosol equilibrium of oxalic acid in the presence of atmospheric NH3 should be treated differently for dry aerosols. In these cases, the vapor pressure of ammonium oxalate should be alternatively adopted in modeling. Moreover, the independent size distributions of particulate oxalate in the upwind marine atmosphere in April suggested that fog-processing of VOCs in the absence of SO2 can be an important formation pathway of particulate oxalate. The formation pathway likely led to a narrow low-value range of ratios of oxalic-acid-vapors* to particulate oxalate in that month.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/atmos16010098/s1, Text S1: Equilibrium equations for NH3, HNO3, H2SO4 and oxalic acid vapor (H2C2O4) in the gas-aqueous-solid phases involved in the E-AIM model. Figure S1: Diurnal variations in monthly average atmospheric mixing layer height (MLH) over Sanya and CO concentrations in 2023 (a) February; (b) April; (c) November; (d) December; the MLH data and CO concentrations were downloaded from (https://cds.climate.copernicus.eu/cdsapp/dataset/reanalysis-era5-single-levelstab=form and https://aqicn.org/city/sanya/, respectively, accessed on 10 January 2025); Figure S2: Offline calibration curves detected by IC ((a) NO3−; (b) SO42−; (c) Cl−; (d) oxalate); Figure S3: An on-site comparison of NH3gas and NH4+ in PM2.5 measured by two dry denuder samplers with the AIM-IC in 2019 (the data from two dry denuders were averaged for the comparison); Figure S4: Diurnal variations in averaged concentrations of chemical species and PM2.5 (a) oxalic-acid-vapor* and oxalate in PM2.5 on 14–16 February, 14–16 and 18–20 April 2023; (b) same as a: but on 16–24 November and 19–24 December 2023; (c) O3 and PM2.5 on 14–16 February, 14–16 and 18–20 April 2023; (d) same as (c) but on 16–24 November and 19–24 December 2023; the shadows reflect the standard deviation); Figure S5: Correlation between the concentration of oxalic acid vapor* and that of O3 during Period 1; Figure S6: The calculated 24 hr air mass back-trajectories at 100 m in February (a) and April (b) during Period 1 and November (c) and December (d) during Period 2 (the trajectories in 14–16 February (a), 14–16 and 18–20 April (b) during Period 1, 16–24 November (c) and 19–24 December (d) during Period 2 were highlighted in red); Figure S7: The calculated 24 hr air mass back-trajectories at 500 m in February (a) and April (b) during Period 1 and November (c) and December (d) during Period 2 (the trajectories in 14–16 February (a), 14–16 and 18–20 April (b) during Period 1, 16–24 November (c) and 19–24 December (d) during Period 2 were highlighted in red); Figure S8: The calculated 24 h air mass back-trajectories at 1000 m in February (a) and April (b) during Period 1 and November (c) and December (d) during Period 2 (the trajectories in 14–16 February (a), 14–16 and 18–20 April (b) during Period 1, 16–24 November (c) and 19–24 December (d) during Period 2 were highlighted in red); Figure S9: Time series of particulate oxalate in PM2.5 and corresponding ambient RH and temperature during Period 1 (a), during Period 2 (b); Figure S10: Correlations between (oxalic acid vapor* + particulate oxalate) and SO42− in PM2.5 during Period 1 (a), during Period 2 (b); Figure S11: (a) Satellite-based monthly average of chlorophyll-a in April of 2017; (b) Geographical distributions of mean UV radiation over the sampling zones during the campaign from March 30 to May 1 in 2017. The dashed black line indicates unity of the mass ratio of oxalate in 1–3 μm particles and that in PM1.0. The black dots represent the sampling locations in Figure 6h; Table S1: The comparison of inputs and outputs of E-AIM modeling between two groups. (For 883 h, the E-AIM predicts over 95% of (oxalic-acid-vapor + particulate oxalate) to be partitioned in the gas phase and the results appeared to be unpractical; For 495 h, the model predicated (oxalic-acid-vapor + particulate oxalate) to be partitioned in the two phases comparably).

Author Contributions

Methodology, X.Y., Y.G. (Yang Gao) and H.G.; formal analysis, X.Y., Y.G. (Yang Gao) and H.G.; data curation, D.C. and L.S.; writing—original draft, L.Y. and Y.G. (Yating Gao); writing—review and editing, L.Y., Y.G. (Yating Gao). and X.Y. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by Hainan Provincial Joint Project (2021JJLH0050), Hainan Provincial Natural Science Foundation of China (No. 424CXTD429) and the Fundamental Research Funds for the Central Universities (No. 202461112).

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The original contributions presented in this study are included in the article/Supplementary Material. Further inquiries can be directed to the corresponding authors.

Acknowledgments

This work was supported by Hainan Provincial Joint Project of Sanya Yazhou Bay Science and Technology City, Hainan Provincial Natural Science Foundation of China and the Fundamental Research Funds for the Central Universities. During the preparation of this manuscript, the author(s) used ChatGPT-4 for the purposes of polishing. The authors have reviewed and edited the output and take full responsibility for the content of this publication.

Conflicts of Interest

The authors declare no competing financial interest.

Abbreviations

SCS: South China Sea; VOCs, volatile organic carbons; MOUDI, Micro-Orifice Uniform Deposition Impactor; MLH, mixing layer height; RH, Relative Humidity; MSC, the marginal seas of China; AIM-IC, Ambient Ion Monitor-Ion Chromatograph; nss-SO42−, non-sea-salt SO42−; nss-K+, non-sea-salt K+; VMS, vessel management system; Chl-a, chlorophyll-a; SSA, sea-salt aerosols; DMA+, dimethylaminium.

References

  1. Yu, L. Global air-sea fluxes of heat, fresh water, and momentum: Energy budget closure and unanswered questions. Annu. Rev. Mar. Sci 2019, 11, 227–248. [Google Scholar] [CrossRef]
  2. McPhaden, M.J.; Connell, K.J.; Foltz, G.R.; Perez, R.C.; Grissom, K. Tropical Ocean observations for weather and climate: A decadal overview of the global tropical moored buoy array. Oceanography 2023, 36, 32–43. [Google Scholar] [CrossRef]
  3. Williamson, C.J.; Kupc, A.; Axisa, D.; Bilsback, K.R.; Bui, T.; Campuzano-Jost, P.; Dollner, M.; Froyd, K.D.; Hodshire, A.L.; Jimenez, J.L.; et al. A large source of cloud condensation nuclei from new particle formation in the tropics. Nature 2019, 574, 399. [Google Scholar] [CrossRef]
  4. Good, P.; Chadwick, R.; Holloway, C.E.; Kennedy, J.; Lowe, J.A.; Roehrig, R.; Rushley, S.S. High sensitivity of tropical precipitation to local sea surface temperature. Nature 2021, 589, 408–414. [Google Scholar] [CrossRef] [PubMed]
  5. Leung, G.R.; Saleeby, S.M.; Sokolowsky, G.A.; Freeman, S.W.; van den Heever, S.C. Aerosol-cloud impacts on aerosol detrainment and rainout in shallow maritime tropical clouds. Atmos. Chem. Phys. 2023, 23, 5263–5278. [Google Scholar] [CrossRef]
  6. Guo, T.; Guo, Z.; Wang, J.; Feng, J.; Gao, H.; Yao, X. Tracer-based investigation of organic aerosols in marine atmospheres from marginal seas of China to the northwest Pacific Ocean. Atmo.s Chem. Phys. 2020, 20, 5055–5070. [Google Scholar] [CrossRef]
  7. Chen, D.; Yao, X.; Chan, C.K.; Tian, X.; Chu, Y.; Clegg, S.L.; Shen, Y.; Gao, Y.; Gao, H. Competitive uptake of dimethylamine and trimethylamine against ammonia on acidic particles in marine atmospheres. Environ. Sci. Technol. 2022, 56, 5430–5439. [Google Scholar] [CrossRef] [PubMed]
  8. Miller, R.M.; Rauber, R.M.; Di Girolamo, L.; Rilloraza, M.; Fu, D.; McFarquhar, G.M.; Nesbitt, S.W.; Ziemba, L.D.; Woods, S.; Thornhill, K.L. Influence of natural and anthropogenic aerosols on cloud base droplet sizedistributions in clouds over the South China Sea and West Pacific. Atmos. Chem. Phys. 2023, 23, 8959–8977. [Google Scholar] [CrossRef]
  9. Shen, X.; Chen, J.; Li, G.; An, T. A new advance in the pollution profile, transformation process, and contribution to aerosol formation and aging of atmospheric amines. Environ. Sci. Atmos. 2023, 3, 444–473. [Google Scholar] [CrossRef]
  10. Xiao, Q.; Zhang, J.; Wang, Y.; Ziemba, L.D.; Crosbie, E.; Winstead, E.L.; Robinson, C.E.; DiGangi, J.P.; Diskin, G.S.; Reid, J.S.; et al. New particle formation in the tropical free troposphere during camp2ex: Statistics and impact of emission sources, convective activity, and synoptic conditions. Atmos. Chem. Phys. 2023, 23, 9853–9871. [Google Scholar] [CrossRef]
  11. Kawamura, K.; Bikkina, S. A review of dicarboxylic acids and related compounds in atmospheric aerosols: Molecular distributions, sources and transformation. Atmos. Res. 2016, 170, 140–160. [Google Scholar] [CrossRef]
  12. Mochizuki, T.; Kawamura, K.; Miyazaki, Y.; Wada, R.; Takahashi, Y.; Saigusa, N.; Tani, A. Secondary formation of oxalic acid and related organic species from biogenic sources in a larch forest at the northern slope of mt. fuji. Atmos. Environ. 2017, 166, 255–262. [Google Scholar] [CrossRef]
  13. Mahilang, M.; Deb, M.K.; Pervez, S. Biogenic secondary organic aerosols: A review on formation mechanism, analytical challenges and environmental impacts. Chemosphere 2021, 262, 127771. [Google Scholar] [CrossRef] [PubMed]
  14. Arquero, K.D.; Gerber, R.B.; Finlayson-Pitts, B.J. The role of oxalic acid in new particle formation from methanesulfonic acid, methylamine and water. Environ. Sci. Technol. 2017, 51, 2124–2130. [Google Scholar] [CrossRef]
  15. Lee, S.; Gordon, H.; Yu, H.; Lehtipalo, K.; Haley, R.; Li, Y.; Zhang, R. New particle formation in the atmosphere: From molecular clusters to global climate. J. Geophys. Res.-Atmos. 2019, 124, 7098–7146. [Google Scholar] [CrossRef]
  16. Zhang, R.; Shen, J.; Xie, H.; Chen, J.; Elm, J. The role of organic acids in new particle formation from methanesulfonic acid and methylamine. Atmos. Chem. Phys. 2022, 22, 2639–2650. [Google Scholar] [CrossRef]
  17. Boreddy, S.K.R.; Kawamura, K. Investigation on the hygroscopicity of oxalic acid and atmospherically relevant oxalate salts under sub—And supersaturated conditions. Environ. Sci.-Proc. Imp. 2018, 20, 1069–1080. [Google Scholar] [CrossRef]
  18. Sorooshian, A.; MacDonald, A.B.; Dadashazar, H.; Bates, K.H.; Coggon, M.M.; Craven, J.S.; Crosbie, E.; Hersey, S.P.; Hodas, N.; Lin, J.J.; et al. A multi-year data set on aerosol-cloud-precipitation-meteorology interactions for marine stratocumulus clouds. Sci. Data 2018, 5, 1–13. [Google Scholar] [CrossRef] [PubMed]
  19. Tsigaridis, K.; Kanakidou, M. The Present and future of Secondary organic erosol direct forcing on climate. Curr. Clim. Change Rep. 2018, 4, 84–98. [Google Scholar] [CrossRef]
  20. Yao, X.; Fang, M.; Chan, C.K. Size distributions and formation of dicarboxylic acids in atmospheric particles. Atmos. Environ. 2002, 36, 2099–2107. [Google Scholar] [CrossRef]
  21. Martinelango, P.K.; Dasgupta, P.K.; Al-Horr, R.S. Atmospheric production of oxalic acid/oxalate and nitric acid/nitrate in the Tampa Bay airshed: Parallel pathways. Atmos. Environ. 2007, 41, 4258–4269. [Google Scholar] [CrossRef]
  22. Yao, Y.; Ye, X.; Chen, Y.; Zhou, Y.; Lv, Z.; Wang, R.; Zheng, H.; Chen, J. Gas-particle partitioning of low-molecular-weight organic acids in suburban Shanghai: Insight into measured Henry’s law constants dependent on relative humidity. Sci. Total Environ. 2024, 939, 173636. [Google Scholar] [CrossRef] [PubMed]
  23. Hodas, N.; Sullivan, A.P.; Skog, K.; Keutsch, F.N.; Collett, J.L.; Decesari, S.; Facchini, M.C.; Carlton, A.G.; Laaksonen, A.; Turpin, B.J. Aerosol liquid water driven by anthropogenic nitrate: Implications for fifetimes of water-soluble organic gases and potential for secondary organic aerosol formation. Environ. Sci. Technol. 2014, 48, 11127–11136. [Google Scholar] [CrossRef]
  24. Nah, T.; Guo, H.; Sullivan, A.P.; Chen, Y.; Tanner, D.J.; Nenes, A.; Russell, A.; Ng, N.L.; Huey, L.G.; Weber, R.J. Characterization of aerosol composition, aerosol acidity, and organic acid partitioning at an agriculturally intensive rural southeastern US site. Atmos. Chem. Phys. 2018, 18, 11471–11491. [Google Scholar] [CrossRef]
  25. Chen, Y.; Guo, H.; Nah, T.; Tanner, D.J.; Sullivan, A.P.; Takeuchi, M.; Gao, Z.; Vasilakos, P.; Russell, A.G.; Baumann, K.; et al. Low-molecular-weight carboxylic acids in the southeastern US: Formation, partitioning, and implications for organic aerosol aging. Environ. Sci. Technol. 2021, 55, 6688–6699. [Google Scholar] [CrossRef]
  26. Lv, S.; Wu, C.; Wang, F.; Liu, X.; Zhang, S.; Chen, Y.; Zhang, F.; Yang, Y.; Wang, H.; Huang, C.; et al. Nitrate-enhanced gas-to-particle-phase partitioning of water- soluble organic compounds in Chinese urban atmosphere: Implications for secondary organic aerosol formation. Environ. Sci. Technol. Let. 2023, 10, 14–20. [Google Scholar] [CrossRef]
  27. Booth, A.M.; Markus, T.; McFiggans, G.; Percival, C.J.; Mcgillen, M.R.; Topping, D.O. Design and construction of a simple Knudsen Effusion Mass Spectrometer (KEMS) system for vapour pressure measurements of low volatility organics. Atmos. Meas. Technol. 2009, 2, 355–361. [Google Scholar] [CrossRef]
  28. Paciga, A.L.; Riipinen, I.; Pandis, S.N. Effect of ammonia on the volatility of organic diacids. Environ. Sci. Technol. 2014, 48, 13769–13775. [Google Scholar] [CrossRef]
  29. Ortiz-Montalvo, D.L.; Haekkinen, S.A.K.; Schwier, A.N.; Lim, Y.B.; Mcneill, V.F.; Turpin, B.J. Ammonium addition (and aerosol pH) has a dramatic impact on the volatility and yield of glyoxal secondary organic aerosol. Environ. Sci. Technol. 2014, 48, 255–262. [Google Scholar] [CrossRef]
  30. Marti, J.J.; Jefferson, A.; Cai, X.P.; Richert, C.; McMurry, P.H.; Eisele, F. H2SO4 vapor pressure of sulfuric acid and ammonium sulfate solutions. J. Geophys. Res.-Atmos. 1997, 102, 3725–3735. [Google Scholar] [CrossRef]
  31. Yao, L.; Garmash, O.; Bianchi, F.; Zheng, J.; Yan, C.; Kontkanen, J.; Junninen, H.; Mazon, S.B.; Ehn, M.; Paasonen, P.; et al. Atmospheric new particle formation from sulfuric acid and amines in a Chinese megacity. Science 2018, 361, 278. [Google Scholar] [CrossRef] [PubMed]
  32. Carlton, A.G.; Turpin, B.J.; Altieri, K.E.; Seitzinger, S.; Reff, A.; Lim, H.; Ervens, B. Atmospheric oxalic acid and SOA production from glyoxal: Results of aqueous photooxidation experiments. Atmos. Environ. 2007, 41, 7588–7602. [Google Scholar] [CrossRef]
  33. Ip, H.S.S.; Huang, X.H.H.; Yu, J.Z. Effective Henry’s law constants of glyoxal, glyoxylic acid, and glycolic acid. Geophys. Res. Lett. 2009, 36, L01802. [Google Scholar] [CrossRef]
  34. Rinaldi, M.; Decesari, S.; Carbone, C.; Finessi, E.; Fuzzi, S.; Ceburnis, D.; O’Dowd, C.D.; Sciare, J.; Burrows, J.P.; Vrekoussis, M.; et al. Evidence of a natural marine source of oxalic acid and a possible link to glyoxal. J. Geophys. Res.-Atmos. 2011, 116, D16204. [Google Scholar] [CrossRef]
  35. Hilario, M.R.A.; Crosbie, E.; Banaga, P.A.; Betito, G.; Braun, R.A.; Cambaliza, M.O.; Corral, A.F.; Cruz, M.T.; Dibb, J.E.; Lorenzo, G.R.; et al. Particulate oxalate-to-sulfate ratio as an aqueous processing marker: Similarity across field campaigns and limitations. Geophys. Res. Lett. 2021, 48, e2021GL096520. [Google Scholar] [CrossRef] [PubMed]
  36. Hegde, P.; Boreddy, S.K.R.; Aswini, A.R.; Aryasree, S. Influence of south Asian outflow on secondary organic aerosol formation over the Indian Ocean: Inferences from water-soluble low molecular weight dicarboxylic acids and related organic compounds during ICARB 2018 experiment. Mar. Chem. 2022, 239, 104071. [Google Scholar] [CrossRef]
  37. Yang, C.; Zhou, S.; Zhang, C.; Yu, M.; Cao, F.; Zhang, Y. Atmospheric chemistry of oxalate: Insight into the role of relative humidity and aerosol acidity from high-resolution observation. J. Geophys. Res.-Atmos. 2022, 127, e2021JD035364. [Google Scholar] [CrossRef]
  38. Hoque, M.M.M.; Kawamura, K.; Uematsu, M. Spatio-temporal distributions of dicarboxylic acids, ω-oxocarboxylic acids, pyruvic acid, α- dicarbonyls and fatty acids in the marine aerosols from the North and South Pacific. Atmos. Res. 2017, 185, 158–168. [Google Scholar] [CrossRef]
  39. Boreddy, S.K.R.; Kawamura, K.; Gowda, D.; Deshmukh, D.K.; Narasimhulu, K.; Ramagopal, K. Sulfate-associated liquid water amplifies the formation of oxalic acid at a semi-arid tropical location over peninsular India during winter. Sci. Total Environ. 2023, 874, 162365. [Google Scholar] [CrossRef] [PubMed]
  40. Shah, V.; Jaeglé, L.; Jimenez, J.L.; Schroder, J.C.; Campuzano-Jost, P.; Campos, T.L.; Reeves, J.M.; Stell, M.; Brown, S.S.; Lee, B.H.; et al. Widespread pollution from secondary sources of organic aerosols during winter in the Northeastern United States. Geophys. Res. Lett. 2019, 46, 2974–2983. [Google Scholar] [CrossRef]
  41. Kuang, Y.; He, Y.; Xu, W.; Yuan, B.; Zhang, G.; Ma, Z.; Wu, C.; Wang, C.; Wang, S.; Zhang, S.; et al. Photochemical aqueous-phase reactions induce rapid daytime formation of oxygenated organic aerosol on the North China Plain. Environ. Sci. Technol. 2020, 54, 3849–3860. [Google Scholar] [CrossRef]
  42. Srivastava, D.; Vu, T.V.; Tong, S.; Shi, Z.; Harrison, R.M. Formation of secondary organic aerosols from anthropogenic precursors in laboratory studies. NPJ Clim. Atmos. Sci. 2022, 5, 22. [Google Scholar] [CrossRef]
  43. Guo, T.; Li, K.; Zhu, Y.; Gao, H.; Yao, X. Concentration and size distribution of particulate oxalate in marine and coastal atmospheres -implication for the increased importance of oxalate in nanometer atmospheric particles. Atmos. Environ. 2016, 142, 19–31. [Google Scholar] [CrossRef]
  44. Bian, Q.; Huang, X.H.H.; Yu, J.Z. One-year observations of size distribution characteristics of major aerosol constituents at a coastal receptor site in Hong Kong—Part 1: Inorganic ions and oxalate. Atmos. Chem. Phys. 2014, 14, 9013–9027. [Google Scholar] [CrossRef]
  45. Clegg, S.L.; Qiu, C.; Zhang, R. The deliquescence behaviour, solubilities, and densities of aqueous solutions of five methyl—And ethyl-aminium sulphate salts. Atmos. Environ. 2013, 73, 145–158. [Google Scholar] [CrossRef]
  46. Lin, Y.; Zhang, L.; Fan, Q.; Meng, H.; Gao, Y.; Gao, H.; Yao, X. Decoupling impacts of weather conditions on interannual variations in concentrations of criteria air pollutants in South China—Constraining analysis uncertainties by using multiple analysis tools. Atmos. Chem. Phys. 2022, 22, 16073–16090. [Google Scholar] [CrossRef]
  47. Chen, D.; Shen, Y.; Wang, J.; Gao, Y.; Gao, H.; Yao, X. Mapping gaseous dimethylamine, trimethylamine, ammonia, and their particulate counterparts in marine atmospheres of China’ s marginal seas—Part 1: Differentiating marine emission from continental transport. Atmos. Chem. Phys. 2021, 21, 16413–16425. [Google Scholar] [CrossRef]
  48. Yao, X.; Shairsingh, K.; Lam, P.H.; Evans, G.J. Underestimation of sulfate concentration in PM2.5 using a semi-continuous particle instrument based on ion chromatography. J. Environ. Monit. 2009, 11, 1292–1297. [Google Scholar] [CrossRef]
  49. Shen, Y.; Wang, J.; Gao, Y.; Chan, C.K.; Yao, X. Sources and formation of nucleation mode particles in remote tropical marine atmospheres over the South China Sea and the Northwest Pacific Ocean. Sci. Total Environ. 2020, 735, 139302. [Google Scholar] [CrossRef] [PubMed]
  50. Yu, Y.; Xing, X.; Liu, H.; Yuan, Y.; Wang, Y.; Chai, F. The variability of chlorophyll-a and its relationship with dynamic factors in the basin of the South China Sea. J. Marine Syst. 2019, 200, 103230. [Google Scholar] [CrossRef]
  51. Usukura, K.K. Distributions of low molecular weight dicarboxylic acids in the North Pacific aerosol samples. J. Oceanogr. 1993, 49, 271–283. [Google Scholar] [CrossRef]
  52. Cheng, C.; Li, M.; Chan, C.K.; Tong, H.; Chen, C.; Chen, D.; Wu, D.; Li, L.; Cheng, P.; Gao, W. Mixing state of oxalic acid containing particles in the rural area of pearl river delta, China: Implication for seasonal formation mechanism of Secondary Organic Aerosol (SOA). Atmos. Chem. Phys. Discuss 2016, 17, 9519–9533. [Google Scholar] [CrossRef]
  53. Hoffer, A.; Andreae, M.O.; Andreae, T.W.; Kawamura, K.; Kundu, S. Molecular distributions of dicarboxylic acids, ketocarboxylic acids and α-dicarbonyls in biomass burning aerosols: Implications for photochemical production and degradation in smoke layers. Atmos. Chem. Phys. 2010, 10, 2209–2225. [Google Scholar] [CrossRef]
  54. Wexler, A.S.; Seinfeld, J.H. The distribution of ammonium-salts among a size and composition dispersed aerosol. Atmos. Environ. Part A 1990, 24, 1231–1246. [Google Scholar] [CrossRef]
  55. Shiraiwa, M.; Seinfeld, J.H. Equilibration timescale of atmospheric secondary organic aerosol partitioning. Geophys. Res. Lett. 2012, 39, L24801. [Google Scholar] [CrossRef]
  56. Gen, M.; Liang, Z.; Zhang, R.; Go Mabato, B.R.; Chan, C.K. Particulate nitrate photolysis in the atmosphere. Environ. Sci. Atmos. 2022, 2, 111–127. [Google Scholar] [CrossRef]
  57. Mochida, M.; Umemoto, N.; Kawamura, K.; Uematsu, M. Bimodal size distribution of C 2 -C 4 dicarboxylic acids in the marine aerosols. Geophys. Res. Lett. 2003, 30, 2275. [Google Scholar] [CrossRef]
  58. Feng, J.L.; Guo, Z.G.; Zhang, T.R.; Yao, X.H.; Chan, C.K.; Fang, M. Source and formation of secondary particulate matter in PM 2.5 in Asian continental outflow. J. Geophys. Res.-Atmos. 2012, 117, D03302. [Google Scholar] [CrossRef]
  59. Van Pinxteren, M.; Fiedler, B.; van Pinxteren, D.; Iinuma, Y.; Koertzinger, A.; Herrmann, H. Chemical characterization of sub-micrometer aerosol particles in the tropical Atlantic Ocean: Marine and biomass burning influences. J. Atmos. Chem. 2015, 72, 105–125. [Google Scholar] [CrossRef]
Atmosphere 16 00098 g001
Figure 1. Map of the sampling site: (a) high-resolution terrains nearby from Google Earth (b,c); the photos were taken within ~1 km distance from the sampling site (d–g). Red stars in (a–c) represent the location of the sampling site.
Figure 1. Map of the sampling site: (a) high-resolution terrains nearby from Google Earth (b,c); the photos were taken within ~1 km distance from the sampling site (d–g). Red stars in (a–c) represent the location of the sampling site.
Atmosphere 16 00098 g001
Atmosphere 16 00098 g002
Figure 2. Time series of concentrations of oxalic-acid-vapor* and oxalate in PM2.5 during Period 1 (a) and Period 2 (b). The correlations between oxalic-acid-vapor* and oxalate in PM2.5 during Period 1 (c) and Period 2 (d). The diurnal variations in averaged concentrations of oxalic-acid-vapor* during Period 1 (e) and Period 2 (f) (the blue shadow in (e,f) represents the standard deviation).
Figure 2. Time series of concentrations of oxalic-acid-vapor* and oxalate in PM2.5 during Period 1 (a) and Period 2 (b). The correlations between oxalic-acid-vapor* and oxalate in PM2.5 during Period 1 (c) and Period 2 (d). The diurnal variations in averaged concentrations of oxalic-acid-vapor* during Period 1 (e) and Period 2 (f) (the blue shadow in (e,f) represents the standard deviation).
Atmosphere 16 00098 g002
Atmosphere 16 00098 g003
Figure 3. Correlations between oxalate and SO42− in PM2.5 during Period 1 (a) and during Period 2 (b). The blue and red dots in (a) represent the data obtained from 18 to 22 February and 20–23 April, respectively.
Figure 3. Correlations between oxalate and SO42− in PM2.5 during Period 1 (a) and during Period 2 (b). The blue and red dots in (a) represent the data obtained from 18 to 22 February and 20–23 April, respectively.
Atmosphere 16 00098 g003
Atmosphere 16 00098 g004
Figure 4. Comparisons of predicted concentrations of oxalate in PM2.5 with observed values (a) inFebruary; (b) in April, (c) November–December with H2O2 and (d) December with H2O; the difference in predicted oxalic-acid-vapor concentrations minus the observed values with the observed values, (e) in February; (f) in April, (g) November–December with H2O2; (h) December with H2O; black markers in (e–h) represent the cases with higher values in predicted vapor concentrations than the observations, and dark red markers represents the cases with lower predicted vapor values than the observations, respectively).
Figure 4. Comparisons of predicted concentrations of oxalate in PM2.5 with observed values (a) inFebruary; (b) in April, (c) November–December with H2O2 and (d) December with H2O; the difference in predicted oxalic-acid-vapor concentrations minus the observed values with the observed values, (e) in February; (f) in April, (g) November–December with H2O2; (h) December with H2O; black markers in (e–h) represent the cases with higher values in predicted vapor concentrations than the observations, and dark red markers represents the cases with lower predicted vapor values than the observations, respectively).
Atmosphere 16 00098 g004
Atmosphere 16 00098 g005
Figure 5. The modeled and observed ratios of oxalic acid vapor* to oxalate in PM2.5 varied with the modeled aerosol pH (a–d) modeled ratios in February, April, November–December and December with pure-H2O used in wet denuder; (e–h) same as (a–d) except for observed ratios; color bar represents LWC; black markers in (e–h) represent ~60% cases when the modeled unpractically high partitioning ratio of oxalic acid vapor).
Figure 5. The modeled and observed ratios of oxalic acid vapor* to oxalate in PM2.5 varied with the modeled aerosol pH (a–d) modeled ratios in February, April, November–December and December with pure-H2O used in wet denuder; (e–h) same as (a–d) except for observed ratios; color bar represents LWC; black markers in (e–h) represent ~60% cases when the modeled unpractically high partitioning ratio of oxalic acid vapor).
Atmosphere 16 00098 g005
Atmosphere 16 00098 g006
Figure 6. Mass size distributions of oxalate, nss-SO42− and DMA+ in atmospheric particles and Spatial-temporal variation in particulate oxalate observed over the SCS in 2017. (a) Oxalate with a dominant supermicron mode; (b) oxalate with a minor or comparable supermicron mode; (c) nss-SO42−; (d) DMA+; (e) Na+; (f) NO3−; (g) nss-K+; (f) spatial-temporal variation); (h) geographical distributions of oxalate mass concentrations in PM10 and the mass ratio of oxalate in 1–3 μm particles to that in PM1.0. Red star in (h) represents the coastal sampling site in Sanya during 2023–2024 observations.
Figure 6. Mass size distributions of oxalate, nss-SO42− and DMA+ in atmospheric particles and Spatial-temporal variation in particulate oxalate observed over the SCS in 2017. (a) Oxalate with a dominant supermicron mode; (b) oxalate with a minor or comparable supermicron mode; (c) nss-SO42−; (d) DMA+; (e) Na+; (f) NO3−; (g) nss-K+; (f) spatial-temporal variation); (h) geographical distributions of oxalate mass concentrations in PM10 and the mass ratio of oxalate in 1–3 μm particles to that in PM1.0. Red star in (h) represents the coastal sampling site in Sanya during 2023–2024 observations.
Atmosphere 16 00098 g006
Table 1. Concentrations of gaseous oxalic-acid-vapor* and particulate oxalate in PM2.5 in different atmospheres.
Table 1. Concentrations of gaseous oxalic-acid-vapor* and particulate oxalate in PM2.5 in different atmospheres.
RegionDateConcentration (ng m−3)Mass Ratio of
Oxalic-Acid-Vapor* to Oxalate in PM2.5		References
Oxalic-Acid-Vapor*Oxalate in PM2.5
Hainan Island, ChinaFebruary–April 2023
(Period 1)		59 ± 1557 ± 341.04 ± 0.44This study
November–December 2023
(Period 2)		21 ± 729 ± 230.72 ± 0.30
Hong KongDecember 200020–50
(4 data)		3500.06–0.14[20]
Sydney, FL, USApril–May 2002742900.26[21]
Shanghai, ChinaNovember–December 202022 ± 8475 ± 2660.05 ± 0.03[22]
Yorkville, Georgia, USAugust–October 200628 ± 1970 ± 50
in PM1		0.4 ± 0.4
in PM1		[24,25]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Yan, L.; Gao, Y.; Chen, D.; Sun, L.; Gao, Y.; Gao, H.; Yao, X. Seasonally Dependent Daytime and Nighttime Formation of Oxalic Acid Vapor and Particulate Oxalate in Tropical Coastal and Marine Atmospheres. Atmosphere 2025, 16, 98. https://doi.org/10.3390/atmos16010098

AMA Style

Yan L, Gao Y, Chen D, Sun L, Gao Y, Gao H, Yao X. Seasonally Dependent Daytime and Nighttime Formation of Oxalic Acid Vapor and Particulate Oxalate in Tropical Coastal and Marine Atmospheres. Atmosphere. 2025; 16(1):98. https://doi.org/10.3390/atmos16010098

Chicago/Turabian Style

Yan, Le, Yating Gao, Dihui Chen, Lei Sun, Yang Gao, Huiwang Gao, and Xiaohong Yao. 2025. "Seasonally Dependent Daytime and Nighttime Formation of Oxalic Acid Vapor and Particulate Oxalate in Tropical Coastal and Marine Atmospheres" Atmosphere 16, no. 1: 98. https://doi.org/10.3390/atmos16010098

APA Style

Yan, L., Gao, Y., Chen, D., Sun, L., Gao, Y., Gao, H., & Yao, X. (2025). Seasonally Dependent Daytime and Nighttime Formation of Oxalic Acid Vapor and Particulate Oxalate in Tropical Coastal and Marine Atmospheres. Atmosphere, 16(1), 98. https://doi.org/10.3390/atmos16010098

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop