Reactive Oxygen Species (ROS) Measurements in Laboratory and Ambient Studies
Open access
Author
Date
2018Type
- Doctoral Thesis
ETH Bibliography
yes
Altmetrics
Abstract
Breathing pristine clean air, a fundamental human right ever since the world was created, is threatened now and then on the road of human society development. To create the environment that supports our life in a sustainable way, we must solve the air pollution problem and at the same time continue our technologically based civilization. Although air pollution has attracted attention for centuries, the air-cleaning task is still arduous and needs long-term efforts to be settled. Understanding the composition of the atmospheric air, what causes the air pollution and its adverse health effects to human beings, is of highest importance to make decisions in preventing air pollution.
The chemical composition of ambient air, including both the gas and the particle phase, is changing due to human activities. Aerosol particles are defined as the tiny particles suspended in a gas, where particulate matter (PM) with an aerodynamic diameter d < 10 µm (PM10) often represents the aerosol particles of interest in atmospheric studies. Exposure to PM2.5 (PM with d < 2.5 µm) was linked to 3.15 million premature deaths worldwide in 2010. Our study focused on the tropospheric aerosols, which are the major factors in climate change and health effects. One of the most important pathways of particulate PM leading to deleterious impacts on health is believed to be the induced oxidative stress by the generation of reactive oxygen species (ROS) from the inhaled PM in vivo (endogenous ROS), or by the transportation of particle-bound ROS (PB-ROS) into the lungs (exogenous ROS). Therefore the understanding of the ROS formation and decay in PM is of utmost importance for mitigating their impact on health. In our work, to enable a rapid screening of the ROS content of PM, we have developed and characterized a highly sensitive ROS analyzer using a 2’,7’-dichlorofluorescin (DCFH) based assay, which can be used either online or offline. The online ROS analyzer greatly improved the quality of the real-time ROS monitoring, and provided a reliable ROS quantification by reducing the losses of the short-lived ROS. This instrument was used in laboratory studies to measure PB-ROS in the exhaust of different emission sources, including the primary emissions and secondary formation of wood and coal combustion, as well as secondary formation from α-pinene oxidation. To study this secondary formation, two atmospheric aging simulators, a smog chamber (SC) and a potential aerosol mass reactor (PAM) chamber, were employed. Further, the PB-ROS content in ambient aerosols was quantified in-situ at two contrasting places: Beijing and Bern. The ROS data were complemented by data from an aerosol mass spectrometer (AMS) in most of our studies, and advanced source apportionment approaches (positive matrix factorization and multilinear engine) were used to identify the major sources of the organic fraction in the ambient studies. The combination of PB-ROS measurements on ambient aerosols and different emission sources revealed the main PB-ROS contributors in the ambient aerosol. The aims of this thesis were:
1) Build a robust and reliable fast online ROS analyzer and characterize its performance. Quantify the PB-ROS contributions from different emission sources including wood combustion, coal combustion as well as in the photooxidation of α-pinene.
2) Investigate the influence of aging conditions on PB-ROS formation of different emission sources by simulating the atmospheric aging process.
3) Confirm the main acellular sources of PB-ROS in ambient air by combining in-situ ambient measurements with laboratory studies.
During the course of this study, it was demonstrated that the online ROS analyzer is sufficiently sensitive and robust to be applied to routine analysis of ROS emissions in both laboratory and ambient studies. A characterization of the instrument with tested model organic compounds showed that only peracetic acid were quantitatively measured, while large organic peroxides or those with bulky functional groups (i.e., tert-butyl and phenyl) showed a strongly reduced fluorescence response of the DCFH assay. Potential interferences from gas-phase O3 and NO2 were not observed and matrix effects of particulate SO42- and NO3- were not statistically significant. No interference of Fe3+ was detected, while high concentrations of soluble Fe2+ reduced the ROS signal. The comparison of online and offline ROS measurement demonstrated the degradation of the highly reactive ROS fraction, and the offline method generally underestimated the ROS concentration, on average by 60 ± 20 %, suggesting that the fast online method presented in this study is advantageous. However, the ROS signal from ambient aerosols may be below or around the instrument detection limit at cleaner sites (such as in Bern). In such cases a versatile aerosol concentration enrichment system (VACES) was successfully used to enhance the sensitivity.
PB-ROS from wood combustion emissions varied for different combustion devices and technologies, different fuel types, operation methods, combustion regimes, combustion phases and aging conditions. For all tested eight combustion devices (within different technologies), primary PB-ROS emissions substantially increased upon aging. The primary and secondary PB-ROS emission factors (EFROS) were dominated by the combustion devices (within different combustion technologies) and by the combustion regimes (expressed as the air to fuel ratio lambda). EFROS from automatically operated combustion devices were on average one order of magnitude lower than those from manually operated appliances, indicating that automatic combustion devices operated at optimum conditions to achieve near-complete combustion, is most effective to minimize PB-ROS emissions. The variability of EFROS within one device was much higher than the variability from different manual devices. In general, an increase of aged EFROS was observed from optimal to high lambda values, with ~2-80 times higher aged EFROS values under bad combustion conditions than under optimum combustion conditions. The PB-ROS content in the secondary organic aerosol (SOA) (represented as fROS-SOA), increased with the SOA oxidation state, which increased with OH exposure and decreased with the additional partitioning of semi-volatile components with lower PB-ROS content at higher OA concentrations, while further aging seemed to result in a decay of PB-ROS.
Further, the PB-ROS concentrations of five types of coal used in residential heating in different regions in China were investigated, including three types of bituminous coal and two types of anthracite coal. The primary EFROS of the three bituminous coals were not statistically different. Primary EFROS of the two types of anthracite coal were not detectable. The EFROS of the bituminous coals further increased upon aging, and their secondary EFROS were ~ 7 times higher than of the anthracite coal, indicating the importance of the type of coal used for the combustion. The primary EFROS from the wood combustion were significantly higher than those of the bituminous coal, while the aged EFROS from wood combustion were on average comparable to those from bituminous coal. The fROS-SOA initially increased upon aging, and decreased with even higher OH exposure. For all three types of bituminous coal, variable fROS-SOA were observed under the same OH exposure, which was dominated by the OA loading: fROS-SOA was higher at lower SOA loading for each individual type of coal.
Atmospheric aerosol measurements at the two contrasting locations Beijing (China) and Bern (Switzerland) showed large predominance of SOA to PB-ROS activity in fine aerosol. During the campaign in Beijing (from January to February 2015), the main OA composition was a mixture of hydrocarbon-like OA and coal combustion OA (HOA+CCOA), cooking emissions OA (COA), biomass burning OA (BBOA), as well as of oxygenated OA (OOA). In Bern, the main OA sources were HOA, COA, BBOA and OOA in November, 2014. The combination of these source apportionment results with observed PB-ROS by a multiple linear regression model (MLRM) revealed that the main parameter affecting the ROS concentration was OOA at both locations. In Bern, OOA contributed on average to more than 52 % of the explained ROS by MLRM, followed by HOA (19 %), BBOA (24 %) and COA (5 %). In Beijing, OOA explained 77% of the observed PB-ROS, while the contribution of primary OA sources to PB-ROS activity could not be retrieved within our uncertainties, when these sources were considered individually or lumped together. The PB-ROS content in the OOA derived from the ambient measurements was comparable to those from the direct source emissions obtained from the laboratory studies, where the PB-ROS contents in SOA emissions are about 4 to 25 times higher than those in the corresponding primary samples, confirming the importance of the secondary organic aerosol for the PB-ROS level in the ambient atmosphere. The source-to-source variation of the PB-ROS content in SOA from laboratory studies could be explained by the different SOA precursors.
This study describes a much improved state-of-the-art estimate of the PB-ROS, a large fraction of which is labile. By coupling the PB-ROS with the revealed dominant sources and components of fine particles, the result clearly suggests the SOA, which is formed from precursors of different anthropogenic and biogenic emissions, plays a predominant role to PB-ROS activity in fine aerosol. The findings and special data analysis may provide a guideline for future ROS analyses and facilitate further toxicological and epidemiological studies in related fields. Show more
Permanent link
https://doi.org/10.3929/ethz-b-000281449Publication status
publishedExternal links
Search print copy at ETH Library
Contributors
Examiner: Baltensperger, Urs
Examiner: McNeill, Kristopher
Examiner: Dommen, Josef
Examiner: Kalberer, Markus
Publisher
ETH ZurichSubject
Reactive oxygen species (ROS); Air pollution; Emission sources; Atmospheric chemistry; Health effectsOrganisational unit
02350 - Dep. Umweltsystemwissenschaften / Dep. of Environmental Systems Science
More
Show all metadata
ETH Bibliography
yes
Altmetrics