Figure 4 - available via license: Creative Commons Attribution 3.0 Unported
Content may be subject to copyright.
(a) Relative OA enhancement (ER OA = reactor OA / ambient OA) vs. estimated reactor photochemical age for the sampling period. (b) Absolute OA mass concentration enhancement (OA mass = reactor OA–ambient OA) vs. photochemical age. The data have been averaged into 6 % quantiles for day and night measurements, with vertical error bars indicating standard errors.
Source publication
Field studies in polluted areas over the last decade have observed large
formation of secondary organic aerosol (SOA) that is often poorly captured by
models. The study of SOA formation using ambient data is often confounded by
the effects of advection, vertical mixing, emissions, and variable degrees of
photochemical aging. An oxidation flow react...
Contexts in source publication
Context 1
... exp in Fig. 4a and b, respectively, for the sample period. OA mass is enhanced up to four times from ambient OA, with the majority of maximum ER OA peaking around a factor of two increase. OA enhancement peaks and plateaus between 0.8 and 6 days of OH aging, then decreases at higher aging, eventually showing net OA loss beyond 2 weeks of aging. When ...
Context 2
... OA mass is enhanced up to four times from ambient OA, with the majority of maximum ER OA peaking around a factor of two increase. OA enhancement peaks and plateaus between 0.8 and 6 days of OH aging, then decreases at higher aging, eventually showing net OA loss beyond 2 weeks of aging. When separated into daytime and nighttime ER OA and OA mass (Fig. 4), the qualitative trends are the same in both cases, but OA was more enhanced from reactor aging dur- ing nighttime by 5 µg m −3 , or a factor of 1.7 × of ambient. A smaller enhancement is observed during the day ∼ 2 µg m −3 , or a factor of 1.2 × of ambient, while at > 2 weeks of aging, day and night observations closely overlap, with ...
Context 3
... Pasadena, is ∼ 0.5 days ( Washenfelder et al., 2011;Hayes et al., 2013), with ambi- ent photochemical ages reaching ∼ 0.3 days. Thus most of the SOA precursors that can become SOA have already done so by the time the air was sampled in Pasadena and only about 20 % more SOA could be produced from the precur- sors that remained. The trends in Fig. 4 indicate increased oxidation transitioning from a dominance of functionaliza- tion reactions and condensation at low-to-moderate expo- sures to fragmentation-dominated reactions and evaporation of reaction products at the highest photochemical ages. Frag- mentation can occur in the gas phase by reactions of SVOCs with OH, leading to ...
Context 4
... but can be used as a proxy ambient photochem- istry, these results further confirm that as the degree of ambi- ent photochemical processing of the sampled air increases (during daytime), SOA production in the reactor becomes more limited, likely due to the depletion of reactive SOA pre- cursors in ambient air, consistent with the conclusions from Fig. ...
Similar publications
The particle size distribution of urban aerosols is typically dominated by ultrafine particles (UFP) originating from local sources such as traffic and industrial emissions. Due to their negligible mass they are clearly underrepresented by legislative PM10 mass measurements. This is one reason why their contribution to urban air quality is better c...
Citations
... The design, premise, and chemistry are described in detail elsewhere (Kang et al., 2007;Lambe et al., 2011;R. Li et al., 2015;Ortega et al., 2016;Peng & Jimenez, 2020). Briefly, ambient air was sampled directly into the cylindrical, aluminum reactor (∼13 L) via a 14 cm diameter opening at a rate of 4.4 L min 1 without inlet tubing (referred to as the "ambient OFR" in the campaign), corresponding to a residence time of ∼3 min. ...
... Additionally, there is some overlap in OH exposure between the least oxidized OFR measurements at night, and the highest OH exposure ambient measurements during the day. The oxidation of nighttime ambient air by the OFR serves as a surrogate for daytime ambient oxidation, as described by Ortega et al. (2016). There appears to be an offset between the OFR-and RT-Vocus enhancements, particularly for C8 and C9 aromatics. ...
The ozone air quality standard is regularly surpassed in the Los Angeles air basin, and efforts to mitigate ozone production have targeted emissions of precursor volatile organic compounds (VOCs), especially from mobile sources. In order to assess how VOC concentrations, emissions, and chemistry have changed over the past decade, VOCs were measured in this study using a Vocus‐2R proton‐transfer reaction time‐of‐flight mass spectrometer in Pasadena, California, downwind of Los Angeles, in summer 2022. Relative to 2010, ambient concentrations of aromatic hydrocarbons have declined at a similar rate as carbon monoxide, suggesting reduced overall emissions from mobile sources. However, the ambient concentrations of oxygenated VOCs have remained similar or increased, suggesting a greater relative importance of oxidation products and other emission sources, such as volatile chemical products whose emissions are largely unregulated. Relative to 2010, the range of measured VOCs was expanded, including higher aromatics and additional volatile chemical products, allowing a better understanding of a wider range of emission sources. Emission ratios relative to carbon monoxide were estimated and compared with 2010 emission ratios. Average measured ozone concentrations were generally comparable between 2022 and 2010; however, at the same temperature, daytime ozone concentrations were lower in 2022 than 2010. Faster photochemistry was observed in 2022, with average hydroxyl radical exposure being ∼68% higher during midday (statistically significant at 95% confidence), although this difference reduces to ∼35% when comparing observations at ambient temperatures of 25–30°C only. Future trends in temperature are important in predicting ozone production.
... As depicted in Fig. 4, the entire system necessitated over 100 min to achieve a thermal steady state. However, a much shorter durations (15-24 min) were usually applied in the laboratory and field studies when voltages were systematically varied (Link et al., 2016;Murschell and Farmer, 2018;Hu et al., 2016Hu et al., , 2022Ortega et al., 2016;Palm et al., 2016Palm et al., , 2018Saha et al., 2018;Shah et al., 2020). These brief constant lamp power periods led to diminished temperature variation between different power settings, as well as inconsistencies between maximum temperature peak with maximum power setting. ...
... For the lower driving voltage of 0.95 V for BHK lamps during a field campaign, a temperature increase of 1-2°C with an equivalent aging time of around 1.5 d was observed ( Fig. S7a and b). During that campaign, the OFR was operated mostly continuously at ∼ 1.5 d equivalent aging, since most SOA formation is often observed at these moderate exposures and is low enough that heterogeneous oxidation is not yet significant (e.g., Palm et al., 2016;Ortega et al., 2016;Hu et al., 2016). ...
... This impact of temperature varied depending on the specific precursors and reaction conditions. Simulation results suggested that the decreased ratio of SOA output to preexisting OA at higher OH exposures (e.g., equivalent aging time ≥ 5 d under low-NO x conditions) observed in previous studies (Hu et al., 2022;Lambe et al., 2015a;Ortega et al., 2016;Palm et al., 2016;Saha et al., 2018) may not only be due to heterogeneous reactions and/or enhanced gas-phase reaction. Instead, it could also be attributed to the lower SOA yield resulting from the temperature increase. ...
Oxidation flow reactors (OFRs) have been extensively utilized to examine the formation of secondary organic aerosol (SOA). However, the UV lamps typically employed to initiate the photochemistry in OFRs can result in an elevated reactor temperature when their implications are not thoroughly evaluated. In this study, we conducted a comprehensive investigation into the temperature distribution within an Aerodyne potential aerosol mass OFR (PAM-OFR) and then examined the subsequent effects on flow and chemistry due to lamp heating. A lamp-induced temperature increase was observed, which was a function of lamp-driving voltage, number of lamps, lamp types, OFR residence time, and positions within the PAM-OFR. Under typical PAM-OFR operational conditions (e.g., < 5 d of equivalent atmospheric OH exposure under low-NOx conditions), the temperature increase typically ranged from 1–5 °C. Under extreme (but less frequently encountered) conditions, the heating could reach up to 15 °C. The influences of the increased temperature over ambient conditions on the flow distribution, gas, and condensed-phase chemistry within PAM-OFR were evaluated. Our findings indicate that the increase in temperature altered the flow field, resulting in a diminished tail on the residence time distribution and corresponding oxidant exposure due to faster recirculation. According to simulation results from a radical chemistry box model, the variation in absolute oxidant concentration within PAM-OFR due to temperature increase was minimal (< 5 %). The temperature influences on seed organic aerosol (OA) and newly formed secondary OA were also investigated, suggesting that an increase in temperature can impact the yield, size, and oxidation levels of representative biogenic and anthropogenic SOA types. Recommendations for temperature-dependent SOA yield corrections and PAM-OFR operating protocols that mitigate lamp-induced temperature enhancement and fluctuations are presented. We recommend blowing air around the reactor's exterior with fans during PAM-OFR experiments to minimize the temperature increase within PAM-OFR. Temperature increases are substantially lower for OFRs utilizing less powerful lamps compared to the Aerodyne version.
... They come from a variety of places, including as natural sources like vegetation, automobile emissions, and industrial operations. VOCs have a role in the production of secondary organic aerosols, these aerosols can contribute to the particulate matter, including PM 10 , suspended in the air Ortega et al., 2016). In addition, VOCs can adsorb onto particulate matter surfaces. ...
... example, isopréne, pantane and formaldéhyde) can be explained by OH and O 3 oxidation of VOC, that can produce secondary organic aerosols (SOA), that represent an important fraction of PM10 (sbai et al., 2022;Ortega et al., 2016;Palm et al., 2016) ...
Please cite this article as: Salah Eddine, S., Drissi, L.B., Mejjad, N., Mabrouki, J., Romanov, A.A., Machine learning models application for spatiotemporal patterns of particulate matter prediction and forecasting over Morocco in north of Africa, Atmospheric Pollution Research, https://doi. 5 ORCID ID Salah eddine sbai: https://orcid.org/0000-0001-9607-7253 Nezha Mejjad : https://orcid.org/0000-0002-6750-6781 Aleksey A. Romanov https://orcid.org/0000-0001-9612-4240 Abstract 8 Atmospheric air pollution exposure raises morbidity and mortality rates and is a major cause of 9 the world's illness burden. In this context, we explored spatial and temporal trends in particulate 10 matter PM10 from 2003 to 2020 over Morocco to assess air pollution exposure. We use the 11 capabilities of ML models to study PM10 trends using 26 predictor variables, including 12 meteorological parameters, volatile organic compounds, atmospheric oxidants, and aerosol 13 optical depth data from the Copernicus Atmosphere Monitoring Service (CAMS). For this 14 purpose, three ML models were built: Multiple Linear Regression (MLR), Random Forest 15 Regression (RFR), and Generalized Additive Model (GAM). To match and optimize these 16 models, a set of ML algorithms has been coupled with each model. The results show all these 17 models are highly accurate in predicting and forecasting PM10 total column trends. Cross-18 J o u r n a l P r e-p r o o f 2 validation showed that GAM had better prediction ability for the PM10 total column with R 2 19 =0.994 and a very low root mean squared error (RMSE) not exceeding 0.046×10-16 kg/m 2. The 20 GAM model showed much higher predictive ability and lower bias than the other models. This 21 finding can be explained by the advantages of GAMs, including their ability to capture complex 22 and non-linear patterns in the data, making them particularly useful when relationships are not 23 easily represented by linear models. This study has presented a comprehensive methodology for 24 predicting the spatiotemporal variability of PM10. The proposed methodology holds potential 25 applicability across all regions, facilitating the generation of high-resolution PM10 monitoring 26 and the establishment of systems for the early detection of air pollution incidents in Morocco. 27 Furthermore, the developed models exhibit versatility, enabling their application for estimating 28 future trends of individual pollutants or making real-time predictions of air quality levels. This 29 research contributes to advancing the understanding and proactive management of air quality in 30 the context of Morocco, offering valuable insights for pollution control efforts. 31 Keywords: Machine Learning Model, Multiple Linear Regression Model, Random Forest 32 Model, PM 10 , Generalized Additive Model, Volatile organic compounds, atmospheric 33 oxidants 34
... The 185-nm photon flux is typically on the order of 1-10% of the 254nm photon flux when using this method [27]. The unique feature of OFR185 lies in its capability to generate OH without relying on external O 3 , thus enhancing its adaptability for field studies [28,29] and ability to more closely atmospheric organic proxy radical chemistry [30]. ...
... In addition, Tkacik et al. [107] extended the application of OFR to a tunnel in Pennsylvania to investigate SOA formation from on-road vehicles, highlighting the important contribution of vehicle emissions to SOA formation in urban areas. A more comprehensive investigation was conducted by Ortega et al. [28], who utilized an OFR along with an AMS and PTR-MS in Los Angeles, which significantly advanced our understanding of OFR as a tool for in situ evaluation of the SOA formation and aging from ambient air. Meanwhile, Palm et al. [108] found in a Colorado forest that the contribution of unmeasured S/IVOCs to SOA formation in OFR was 4.4 times more than that of measured VOCs, emphasizing the importance of S/IVOCs. ...
... Most campaigns observed a rapid increase followed by a net mass loss at high OH exposures during the SOA formation and aging process. This trend is qualitatively consistent with SOA yields obtained in numerous laboratory OFR studies in which the SOA yield increases and then decreases as a function of photochemical age due to a transition from functionalization to fragmentation reactions [28,124]. Additionally, SOA formation in ambient air is highly correlated with precursor concentrations. ...
Purpose of Review
This review aims to provide a comprehensive examination of oxidation flow reactor (OFR) studies and their applications in both laboratory and field investigations. OFRs play a crucial role in understanding secondary organic aerosol (SOA) formation and aging processes in the atmosphere. By evaluating the advancements and limitations of OFR technology, this review seeks to identify key research directions and challenges for future studies in atmospheric chemistry and air quality research.
Recent Findings
In recent years, OFR has emerged as an encouraging alternative to smog chambers for SOA study. The high oxidative capacity and short residence time of OFR enable its wide application in both laboratory and field studies. Research utilizing OFR has uncovered the critical role of semi-volatile and intermediate-volatility organic compounds (S/IVOCs) in the formation of SOA from various sources, including vehicle emissions, biomass burning, cooking activities, and non-traditional emissions such as volatile chemical products. Notably, field studies have observed considerable variability in the SOA formation potential across different environments globally, generally showing higher formation potential in urban areas compared to rural and forest regions.
Summary
OFR studies have significantly advanced our understanding of SOA formation and aging processes, identifying key precursors, evaluating influencing factors, and quantifying SOA formation potential. However, challenges remain in unraveling detailed mechanisms due to the complexity of SOA sources and properties. Future OFR research should focus on innovations in OFR design, study non-traditional emissions, conduct long-term field observations, develop standardized calibration procedures, and establish SOA yield parameterization schemes for S/IVOCs.
... Another application of this technique would involve sampling air into an oxidation flow reactor (OFR) through one or several conductive polymer tubes. An OFR can be used to quantify the amount of secondary aerosol mass formed from gas-phase precursors (Ortega et al., 2016;Kang et al., 2007). When put in front of an OFR, conductive polymer tubes act as volatility high-pass filters, only transmitting highervolatility VOCs while also transmitting particles. ...
Previous studies have demonstrated volatility-dependent absorption of gas-phase volatile organic compounds (VOCs) to Teflon and other polymers. Polymer–VOC interactions are relevant for atmospheric chemistry sampling, as gas–wall partitioning in polymer tubing can cause delays and biases during measurements. They are also relevant to the study of indoor chemistry, where polymer-based materials are abundant (e.g., carpets and paints). In this work, we quantify the absorptive capacities of multiple tubing materials, including four nonconductive polymers (important for gas sampling and indoor air quality), four electrically conductive polymers and two commercial steel coatings (for gas and particle sampling). We compare their performance to previously characterized materials. To quantify the absorptive capacities, we expose the tubing to a series of ketones in the volatility range 104–109 µgm-3 and monitor transmission. For slow-diffusion polymers (e.g., perfluoroalkoxy alkane (PFA) Teflon and nylon), absorption is limited to a thin surface layer, and a single-layer absorption model can fit the data well. For fast-diffusion polymers (e.g., polyethylene and conductive silicone), a larger depth of the polymer is available for diffusion, and a multilayer absorption model is needed. The multilayer model allows fitting solid-phase diffusion coefficients for different materials, which range from 4×10-9 to 4×10-7 cm2 s-1. These diffusion coefficients are ∼ 8 orders of magnitude larger than literature values for fluorinated ethylene propylene (FEP) Teflon film. This enormous difference explains the differences in VOC absorption measured here. We fit an equivalent absorptive mass (CW, µgm-3) for each absorptive material. We found PFA to be the least absorptive, with CW ∼ 105 µgm-3, and conductive silicone to be the most absorptive, with CW ∼ 1013 µgm-3. PFA transmits VOCs easily and intermediate-volatility species (IVOCs) with quantifiable delays. In contrast, conductive silicone tubing transmits only the most volatile VOCs, denuding all lower-volatility species. Semi-volatile species (SVOCs) are very difficult to sample quantitatively through any tubing material. We demonstrate a system combining several slow- and fast-diffusion tubing materials that can be used to separate a mixture of VOCs into volatility classes. New conductive silicone tubing contaminated the gas stream with siloxanes, but this effect was reduced 10 000-fold for aged tubing, while maintaining the same absorptive properties. SilcoNert (tested in this work) and Silonite (tested in previous work) steel coatings showed gas transmission that was almost as good as PFA, but since they undergo adsorption, their delay times may be humidity- and concentration-dependent.
... The potential aerosol mass (PAM) reactor is a highly versatile oxidation flow reactor (OFR) that offers both portability and the ability to simulate multigenerational oxidation, which is equivalent to several days of aging. As a result, the PAM reactor has become a popular choice for field deployments in studies investigating the formation and aging of SOA in ambient air (Ahlberg et al., 2019;Hu et al., 2022;Liao et al., 2021;Ortega et al., 2013Ortega et al., , 2016Palm et al., 2016Palm et al., , 2018Saha et al., 2018;Sbai et al., 2021;Tkacik et al., 2014;Xu et al., 2022) and individual or mixed VOCs (Coggon et al., 2019;Lambe et al., 2012;Srivastava et al., 2022). It's worth noting that the results of the studies mentioned above show significant variability in SOA formation and aging depending on location and source strength. ...
... It's worth noting that the results of the studies mentioned above show significant variability in SOA formation and aging depending on location and source strength. For example, Ortega et al. (2013Ortega et al. ( , 2016 observed substantial enhancement of organic aerosols (OA) in ambient forest air, while Ahlberg et al. (2019) did not detect any OA enhancement at a rural site in northern Europe. Furthermore, Xu et al. (2022) reported highest OA enhancement at ~3.9 equivalent photochemical days in Beijing, which is shorter than the duration in Lyon city (~5 days) (Sbai et al., 2021) but longer than in Hong Kong city (1-2 days) . ...
... days aging as ~83.5 μg/m 3 and 1.84, respectively. The absolute enhancement is very high compared to other studies reported in Beijing, Guangzhou, North Carolina, and Los Angeles (Li et al., 2019;Ortega et al., 2016;Saha et al., 2018;Xu et al., 2022). While the relative enhancement is similar to those observed in Los Angeles , and Guangzhou (Hu et al., 2022) but higher than in Beijing (Xu et al., 2022). ...
Secondary aerosols constitute a significant fraction of atmospheric aerosols, yet our understanding of their formation mechanism and fate is very limited. In this work, the secondary organic aerosol (SOA) formation and aging of ambient air of Delhi are studied using a potential aerosol mass (PAM) reactor, an oxidation flow reactor (OFR), coupled with aerosol chemical speciation monitor (ACSM), proton transfer reaction time of flight mass spectrometer (PTR-ToF-MS), and scanning mobility particle sizer with counter (SMPS + C). The setup mimics atmospheric aging of up to several days with the generation of OH radicals. Variations in primary volatile organic compounds (VOCs) and oxygenated volatile organic compounds (OVOCs) as a function of photochemical age were investigated. Primary VOCs such as benzene, toluene, xylene, trimethyl benzene, etc. decrease and OVOCs like formic acid, formaldehyde, acetone, ethanol, etc. increase substantially upon oxidation in OFR. The highest organic aerosol (OA) enhancement was observed for the 4.2 equivalent photochemical days of aging i.e., 1.84 times the ambient concentration, and net OA loss was observed at very high OH exposure, typically after 8.4 days of photochemical aging due to heterogeneous oxidation followed by fragmentation/evaporation. In ambient air, OA enhancement is highest during nighttime due to the high concentrations of precursor VOCs in the atmosphere. SMPS + C results demonstrated substantial new particle formation upon aging and decrement in preexisting aerosol mass. This is the first experimental study conducting an in-situ evaluation of potential SOA mass generated from the ambient aerosols in India.
... This means the impact over a particular city depends on the relative magnitude of these three components of UCMF and how they act on SOA. The UCMF also caused a relatively large impact over rural areas, which can also be explained by the fact that SOA is formed more readily in an aged urban plume (Ortega et al., 2016), so, probably, the SOA precursors removed by the increased vertical eddy diffusion were transported over rural areas, while they were oxidized and condensed into SOA. Regarding the impact of increased PM deposition velocities due to urbanization, their impact follows that of secondary inorganic aerosol or PEC; i.e. the urban concentrations decreased (by around −0.02 µg m −3 ). ...
Rural-to-urban transformation (RUT) is the process of turning a rural or natural land surface into an urban one, which brings about important modifications in the surface, causing well-known effects like the urban heat island (UHI), reduced wind speeds, and increased boundary layer heights. Moreover, with concentrated human activities, RUT introduces new emission sources which greatly perturb local and regional air pollution. Particulate matter (PM) is one of the key pollutants responsible for the deterioration of urban air quality and is still a major issue in European cities, with frequent exceedances of limit values. Here we introduce a regional chemistry–climate model (regional climate model RegCM coupled offline to chemistry transport model CAMx) study which quantifies how the process of RUT modified the PM concentrations over central Europe including the underlying controlling mechanisms that contribute to the final PM pollution. Apart from the two most studied ones, (i) urban emissions and (ii) urban canopy meteorological forcing (UCMF; i.e. the impact of modified meteorological conditions on air quality), we also analyse two less studied contributors to RUT's impact on air quality: (iii) the impact of modified dry-deposition velocities (DVs) due to urbanized land use and (iv) the impact of modified biogenic emissions due to urbanization-induced vegetation modifications and changes in meteorological conditions which affect these emissions. To calculate the magnitude of each of these RUT contributors, we perform a cascade of simulations, whereby each contributor is added one by one to the reference state, while focus is given on PM2.5 (particulate matter with diameter less then 2.5 µm). Its primary and secondary components, namely primary elemental carbon (PEC), sulfates (PSO4), nitrates (PNO3), ammonium (PNH4), and secondary organic aerosol (SOA), are analysed too.
The validation using surface measurements showed a systematic negative bias for the total PM2.5, which is probably caused by underestimated organic aerosol and partly by the negative bias in sulfates and elemental carbon. For ammonium and nitrates, the underestimation is limited to the warm season, while for winter, the model tends to overestimate their concentrations. However, in each case, the annual cycle is reasonably captured.
We evaluated the RUT impact on PM2.5 over a sample of 19 central European cities and found that the total impact of urbanization is about 2–3 and 1–1.5 µgm-3 in winter and summer, respectively. This is mainly driven by the impact of emissions alone causing a slightly higher impact (1.5–3.5 and 1.2–2 µgm-3 in winter and summer), while the effect of UCMF was a decrease at about 0.2–0.5 µgm-3 (in both seasons), which was mainly controlled by enhanced vertical eddy diffusion, while increases were modelled over rural areas. The transformation of rural land use into an urban one caused an increase in dry-deposition velocities by around 30 %–50 %, which alone resulted in a decrease in PM2.5 by 0.1–0.25 µgm-3 in both seasons. Finally, the impact of biogenic emission modifications due to modified land use and meteorological conditions caused a decrease in summer PM2.5 of about 0.1 µgm-3, while the winter effects were negligible. The total impact of urbanization on aerosol components is modelled to be (values indicate winter and summer averages) 0.4 and 0.3 µgm-3 for PEC, 0.05 and 0.02 µgm-3 for PSO4, 0.1 and 0.08 µgm-3 for PNO3, 0.04 and 0.03 µgm-3 for PNH4, and 0 and 0.05 µgm-3 for SOA. The main contributor of each of these components was the impact of emissions, which was usually larger than the total impact due to the fact that UCMF was counteracted with a decrease. For each aerosol component, the impact of modified DV was a clear decrease in concentration, and finally, the modifications of biogenic emissions impacted SOA predominantly, causing a summer decrease, while a very small secondary effect of secondary inorganic aerosol was modelled too (they increased).
In summary, we showed that when analysing the impact of urbanization on PM pollution, apart from the impact of emissions and the urban canopy meteorological forcing, one also has to consider the effect of modified land use and its impact on dry deposition. These were shown to be important in both seasons. For the effect of modified biogenic emissions, our calculations showed that they act on PM2.5 predominantly through SOA modifications, which only turned out to be important during summer.
... Mobile sources are also known to contribute to secondary organic aerosol (SOA) formation, 13,14 which unlike primary emissions, is not a regulated metric in any region of the world. SOA is a major constituent of ambient aerosol mass and forms in the atmosphere through oxidation reactions of VOC precursors with varying volatilities. ...
... The Hg UV fluorescent lamps used in this study are commercial products (GPH436T5VH/4 or GPH436T5L/4, Light Sources, Inc.) which are the default light bulbs chosen by ARI for the OFR (Fig. S2). In addition, the temperature within an OFR with the Penn State low-pressure Hg UV lamps BHK Inc.), which are also widely used in PAM-OFR (Khalaj et al., 2021;Xu and Collins, 2021;Lambe et al., 2011b;Siemens et al., 2022;Hu et al., 2016;Ortega et al., 2013;Link et al., 2016;Palm et al., 2016;Kang et al., 2018;Mitroo et al., 2018;Kang et al., 105 2007), are also measured. For the Light Sources lamps, the light intensity of each lamp was changed through the AC voltage input to the lamp ballast, controlled by a computer with settings ranging from 0 V (minimum) to 10 V (maximum, full AC output). ...
... The Hg UV fluorescent lamps used in this study are commercial products (GPH436T5VH/4 or GPH436T5L/4, Light Sources, Inc.) which are the default light bulbs chosen by ARI for the OFR (Fig. S2). In addition, the temperature within an OFR with the Penn State low-pressure Hg UV lamps BHK Inc.), which are also widely used in PAM-OFR (Khalaj et al., 2021;Xu and Collins, 2021;Lambe et al., 2011b;Siemens et al., 2022;Hu et al., 2016;Ortega et al., 2013;Link et al., 2016;Palm et al., 2016;Kang et al., 2018;Mitroo et al., 2018;Kang et al., 105 2007), are also measured. For the Light Sources lamps, the light intensity of each lamp was changed through the AC voltage input to the lamp ballast, controlled by a computer with settings ranging from 0 V (minimum) to 10 V (maximum, full AC output). ...
... Fig. 4 shows that the entire system needs more than 100 minutes to achieve a thermal steady state. However, a much shorter time (15-24 minutes) is usually applied in the laboratory and field studies when voltages are systematically varied (Link et 255 al., 2016;Murschell and Farmer, 2018;Hu et al., 2016;Ortega et al., 2016;Palm et al., 2016;Saha et al., 2018;Shah et al., 2020;Hu et al., 2022). These short constant lamp power times result in reduced temperature variation between different power settings, as well as inconsistency between maximum temperature peak with maximum power setting. ...
Oxidation flow reactors (OFRs) have been widely used to investigate the formation of secondary organic aerosol (SOA). However, the UV lamps that are commonly used to initiate photochemistry in OFRs can lead to increases in the reactor temperature with consequences that have not been assessed in detail. In this study, we systematically investigated the temperature distribution inside an Aerodyne Potential Aerosol Mass OFR and the associated impacts on flow and chemistry arising from lamp heating. A lamp-induced temperature enhancement was observed, which was a function of lamp driving voltage, number of lamps, lamp types, OFR residence time, and positions inside OFR. Under common OFR operational conditions (e.g., < 5 days of equivalent atmospheric OH exposure under low-NOx conditions), the temperature enhancement was usually within 1–5 °C. Under extreme (but less commonly used) settings, the heating could reach 15 °C. The influence of the increased temperature over ambient conditions on the flow distribution, gas, and condensed phase chemistry inside OFR was evaluated. We found that the increase in temperature changes the flow field, leading to a reduced tail on the residence time distribution and corresponding oxidant exposure due to faster recirculation. According to simulation results from a box model using radical chemistry, the variation of absolute oxidant concentration inside of OFR due to temperature increase was small (<5 %). The temperature influences on existing and newly formed OA were also investigated, suggesting that the increase in temperature can impact the yield, size, and oxidation levels of representative biogenic and anthropogenic SOA types. Recommendations for temperature-dependent SOA yield corrections and OFR operating protocols that mitigate lamp-induced temperature enhancement and fluctuations are presented. We recommend blowing air around the outside of the reactor with fans during OFR experiments to minimize the temperature increase inside OFR. Temperature increases are substantially lower for OFRs using less powerful lamps than the Aerodyne version.
... Another application of this technique would involve sampling air into an oxidation flow reactor (OFR) through one 375 or several conductive polymer tubes. An OFR can be used to quantify the amount of secondary aerosol mass formed from gas-phase precursors (Ortega et al., 2016;Kang et al., 2007). When put in front of an OFR, conductive polymer tubes act as volatility high-pass filters, only transmitting higher volatility VOCs, while also transmitting particles. ...
Previous studies have demonstrated volatility-dependent absorption of gas-phase volatile organic compounds (VOCs) to Teflon and other polymers. Polymer-VOC interactions are relevant for atmospheric chemistry sampling, as gas-wall partitioning in polymer tubing can cause delays and biases during measurements. They are also relevant to the study of indoor chemistry, where polymer-based materials are abundant (e.g. carpets, paints). In this work, we quantify the absorptive capacities of multiple tubing materials, including 4 non-conductive polymers (important for gas sampling and indoor air quality), as well as 4 electrically-conductive polymers and 2 commercial steel coatings (for gas and particle sampling). We compare their performance to previously-characterized materials. To quantify the absorptive capacities, we expose the tubing to a series of ketones in the volatility range 104–109 μg m−3 and monitor transmission. For slow-diffusion polymers (e.g. PFA Teflon and nylon), absorption is limited to a thin surface layer, and a single layer absorption model can fit the data well. For fast-diffusion polymers (e.g. polyethylene and conductive silicone) a larger depth of the polymer is available for diffusion, and a multilayer absorption model was needed. The multilayer model allows fitting solid phase diffusion coefficients for different materials, which range from 4 × 10−9 to 4 × 10−7 cm2 s−1. These diffusion coefficients are ~8 orders of magnitude larger than literature values for FEP Teflon. This enormous difference explains the differences in VOC absorption measured here. We fit an equivalent absorptive mass (Cw, μg m−3) for each absorptive material. We found PFA to be the least absorptive, with Cw ~105 μg m−3, and conductive silicone to be the most absorptive, with Cw ~1013 μg m−3. PFA transmits VOCs easily, and intermediate-volatility species (IVOCs) with quantifiable delays. In contrast, conductive silicone tubing transmits only the most volatile VOCs, denuding all lower volatility species. Semi-volatile species (SVOCs) are very difficult to sample quantitatively through any tubing material. We demonstrate how to use a combination of slow- and fast-diffusion tubing to separate a mixture of VOCs into volatility classes before analysis. New conductive silicone tubing contaminated the gas stream with siloxanes, but this effect was reduced by 10,000-fold for aged tubing, while maintaining the same absorptive properties. SilcoNert (tested in this work) and Silonite (tested in previous work) steel coatings showed gas transmission that was almost as good as PFA, but since they undergo adsorption, their delays times may be humidity and concentration dependent.
