Emissions from natural ecosystems, broadly classified as biogenic volatile organic compounds (BVOCs), contribute 90\% to the VOC budget. Individual BVOCs vary widely in their reaction rates with atmospheric oxidants, making their atmospheric impact highly dependent on VOC composition. Their emissions are also dependent on vegetative make up and a number of meteorological and ecological variables. However, the ecological and physical drivers of their emissions is becoming more variable in a changing climate, leading to greater uncertainties in models. Increasing the monitoring of individual compounds can improve our understanding of the drivers of these emissions and the impact of individual chemical species on atmospheric composition. Improved understanding of BVOC composition can better emission models and, SOA and ozone formation predictions. To study the atmospheric impacts and physical drivers of BVOCs, a GC-FID was adapted for automated hourly sampling and analysis. The details of the hardware and software used for the system are described in detail to enable future long-term BVOC measurements in additional locations. The instrument was deployed at a measurement tower in a forest in central Virginia for year-round collection of BVOC concentrations. Using two years of collected hourly data, this work assesses the chemical impacts of individual BVOCs on time scales ranging from hour to year. This work identifies the importance of both concentration and chemical structure in determining atmospheric impacts. Additionally, seasonality in the concentration of some biogenic species has large implications for atmospheric reactivity in the warmest months of the year, particularly ozone reactivity. Using ecological and meteorological data collected at the site in conjunction with the BVOC data, the drivers of BVOC concentrations and their seasonality are identified. Comparison between this data and current models, reveal important deviations which may lead to large modeled uncertainties. Furthermore, the collected data has been made publicly available to aid in future research regarding BVOCs. / Doctor of Philosophy / The earth hosts a number of sources of atmospheric emissions. These range from human-driven sources such as vehicles and factories, to natural sources such as trees and grass. The content of these emissions, amongst others, become a part of a large reactor (the atmosphere), that interact with each other. The interaction of these emissions with atmospheric oxidants forms a gas (ozone) with implications for human and ecosystem health, and secondary organic aerosol (the leading component to smog). However, the extent to which these emissions react with atmospheric oxidants is largely dependent on the structure of individual compounds. A major focus of this dissertation is to show that compounds with reactive structures can have a large impact on atmospheric composition, and that the quantity of emissions can be as important as compound structure. Understanding the impact of individual compounds in the atmosphere requires improved measurement techniques, capable of detecting the compounds of interest over long time periods. Therefore, another focus of this work was the adaptation and deployment of an instrument capable of detecting some of the most reactive species in the atmosphere, volatile organic compounds emitted from forests. The instrument deployed in this work was a gas chromatography flame ionization detector (GC-FID), which detects compounds largely composed of carbon and hydrogen. The instrument was adapted to run automatically through the development of an electronics box and software program interfaced with the GC-FID. Following development, the instrument was deployed to a remote forest research site for two years. The data collected from this work was used to determine the impact of individual compounds on atmospheric composition. Findings from this work could be used to improve a range of atmospheric models. Small changes in emissions (human or plant) contribute to the total VOC budget which can have large implications for the formation of ozone and SOA. Therefore, increased understanding of the BVOC concentrations and emission driver will aid in predicting these atmospheric components.

Cette thèse porte sur l'étude de dégradation atmosphérique de quelques composés organiques volatils oxygénés par les principaux oxydants atmosphériques OH, Cl, et NO3. Ce travail, comporte, d'une part la détermination des spectres d'absorption UV-Visible des composés hydroxycarbonylés (Hydroxyacétone (HAC), 4-hydroxy-2-butanone (4HB), 3-hydroxy-2-butanone (3HB) et 3-methyl-3-hydoxy-2-butanone (3H3M2B) et d'autre part les mesures cinétiques de la réaction du composé HAC avec le chlore atomique et le radical nitrate. Les réactions entre 4HB et 3H2B avec les radicaux OH et le chlore atomique ont aussi été étudié. Les mesures d'absorption ont été réalisées dans la gamme de température 250 et 363 K à l'aide d'une lampe D2 couplé à un monochromateur. Les valeurs de section obtenues sont utilisées pour calculer la constante de photolyse et d'estimer les durées de vie dans la troposphère des composés étudiés. Les résultats suggèrent que la photolyse pourrait être un processus important pour ces espèces dans la troposphère. Les résultats des études cinétiques, montrent une légère variation des constantes de vitesse avec la température. Nous avons aussi étudié les composés dicétones : 2,4-pentanedione et 2,3-pentanedione. Cette étude porte sur la détermination des spectres d'absorption UV-Visible des composés dicétones et l'étude cinétique de leur réaction avec les radicaux OH en fonction de température. Pour les deux composés une variation de température de plus de 60 % est observée. Les calculs théoriques ont été effectués avec deux programmes différents. Avec logiciel Gaussian 03 nous avons réalisé une optimisation de la géométrie des minima, complexes et états de transition au niveau B3LYP/6-311G++(2d, pd). La méthode composite de haut niveau CBS-QB3 a été utilisée. Et avec le logiciel ChemRate pour les calculs cinétique, et la détermination des constantes de vitesse en phase gazeuse de la réaction étudié en fonction de la température. Les calculs des niveaux d'énergies montrent l'existence d'une étape réactionnelle où il se produit d'un complexe intermédiaire suivi d'une abstraction d'un atome d'hydrogène. / This thesis focuses on the study of atmospheric degradation of some Oxygenated Volatile Organic Compounds by major atmospheric oxidants OH, Cl, and NO3. This work comprises firstly determining the spectra of UV-Visible absorption hydroxycarbonyl compounds (hydroxyacetone (HAC), 4-hydroxy-2-butanone (4HB), 3-hydroxy-2-butanone (3HB) and 3-methyl-3-hydoxy-2-butanone (3H3M2B) and secondly measures the kinetics of the reaction of the compound HAC with the atomic chlorine and nitrate radical.Reactions between 4HB and 3H2B with OH radicals and atomic chlorine were also studied.The UV absorption cross-sections of hydroxyacetone, 3-hydroxy 2-butanone, 4-hydroxy 2-butanone and 3-hydroxy 3-methyl 2-butanone have been measured. The experiments have been carried out between 250 and 363 K using a D2 lamp coupled to a monochromator. This work provides the first UV cross-section measurements for 3-hydroxy 2-butanone, 4-hydroxy 2-butanone and 3-hydroxy 3-methyl 2-butanone. The obtained cross-section values are used to calculate the photolysis rates and to estimate the tropospheric lifetimes of the studied compounds. The results suggest that photolysis could be an important removal process for these species in the troposphere. The results of kinetic studies show a slight variation of the rate constants with temperature.We also studied the diketones compounds: 2,4-pentanedione and 2,3-pentanedione. This study focuses on the determination of UV-Visible absorption spectra of diketones compounds and kinetic study of their reaction with OH radicals according to temperature. For both compounds a temperature variation of more than 60% is observed.The theoretical calculations were performed with two different programs: With Gaussian 03 software we performed a geometry optimization minima complex and transition states at B3LYP/6-311G + + level (2d, pd). The high-level composite method CBS-QB3 was used. And with the software ChemRate for kinetic calculations and the determination of rate constants for gas phase reaction according to the studied temperature. Calculations of energy levels show the existence of a reaction step where there is an intermediate complex followed by abstraction of a hydrogen atom.