Massenspektrometrische in-situ Messungen zur Bestimmung der chemischen Zusammensetzung von natürlichem und anthropogenem Aerosol

Für die vorliegende Arbeit wurde die chemische Zusammensetzung von natürlichen und anthropogenen Aerosolpartikeln untersucht. Zu diesem Zweck wurde das Aerosolmassenspektrometer (AMS) der Firma Aerodyne, Inc. eingesetzt, womit neben den chemischen Substanzen auch die Massengrößenverteilungen der einzelnen Komponenten der Aerosolpartikel in einem Größenbereich zwischen 20 und 1500 nm quantitativ gemessen werden können. Im Rahmen der HAZE2002-Messkampagne am Meteorologischen Observatorium Hohenpeißenberg wurden die Aerosolpartikel aus natürlichen Quellen untersucht. Diese Partikel bestanden aus Sulfat, Nitrat, Ammonium und organischen Komponenten (Organics). Sulfat, Nitrat und Ammonium wiesen den gleichen Durchmesser auf, was auf eine interne Mischung dieser drei chemischen Substanzen in den Partikeln hinwies. Die Organics hatten einen kleineren Durchmesser, was auf jüngere Partikel hindeutete. Die Analyse der organischen Substanzen in den Aerosolpartikeln zeigte, dass diese zu einem großen Teil aus oxidierten Kohlenwasserstoffen bestanden, die während den Nachmittagsstunden gebildet wurden. Die thermische Abhängigkeit der Bildung von Ammoniumnitrat konnte sowohl gemessen als auch mit Hilfe Konzentrationsberechnungen nach [Seinfeld und Pandis, 1998] nachvollzogen werden. Die gemessene Partikelneubildung konnte auf die ternäre Nukleation aus H2SO4/H2O/NH3 zurückgeführt werden. Aerosolpartikel aus anthropogenen Quellen, wie z.B. der motorischen Verbrennung, wurden während der Messungen in Zusammenarbeit mit dem Ford Forschungszentrum in Aachen (FFA) untersucht. Nukleationspartikel (D 45 nm) konnten bei Experimenten auf dem Rollenprüfstand nur bei einer ausreichend hohen Verdünnung, einem hohen Schwefelgehalt im Kraftstoff und einem hohen Lastzustand nachgewiesen werden. Die Messungen an der Autobahn A4 ergaben eine bimodale Massengrößenverteilung der organischen Partikel, wobei die erste Mode Partikeln aus der motorischen Verbrennungen zugeschrieben werden konnte. Aufgrund der guten Charakterisierung stellt das AMS ein vielseitig einsetzbares Aerosolmessgerät dar, welches in einer hohen Zeitauflösung eine quantitative, größenaufgelöste chemische Analyse der zu messenden Aerosolpartikel bereitstellt.

Complementary mass spectrometric techniques for the characterization of the organic fraction in atmospheric aerosols

Aerosol particles are strongly related to climate, air quality, visibility and human health issues. They contribute the largest uncertainty in the assessment of the Earth´s radiative budget, directly by scattering or absorbing solar radiation or indirectly by nucleating cloud droplets. The influence of aerosol particles on cloud related climatic effects essentially depends upon their number concentration, size and chemical composition. A major part of submicron aerosol consists of secondary organic aerosol (SOA) that is formed in the atmosphere by the oxidation of volatile organic compounds. SOA can comprise a highly diverse spectrum of compounds that undergo continuous chemical transformations in the atmosphere.rnThe aim of this work was to obtain insights into the complexity of ambient SOA by the application of advanced mass spectrometric techniques. Therefore, an atmospheric pressure chemical ionization ion trap mass spectrometer (APCI-IT-MS) was applied in the field, facilitating the measurement of ions of the intact molecular organic species. Furthermore, the high measurement frequency provided insights into SOA composition and chemical transformation processes on a high temporal resolution. Within different comprehensive field campaigns, online measurements of particular biogenic organic acids were achieved by combining an online aerosol concentrator with the APCI-IT-MS. A holistic picture of the ambient organic aerosol was obtained through the co-located application of other complementary MS techniques, such as aerosol mass spectrometry (AMS) or filter sampling for the analysis by liquid chromatography / ultrahigh resolution mass spectrometry (LC/UHRMS).rnIn particular, during a summertime field study at the pristine boreal forest station in Hyytiälä, Finland, the partitioning of organic acids between gas and particle phase was quantified, based on the online APCI-IT-MS and AMS measurements. It was found that low volatile compounds reside to a large extent in the gas phase. This observation can be interpreted as a consequence of large aerosol equilibration timescales, which build up due to the continuous production of low volatile compounds in the gas phase and/or a semi-solid phase state of the ambient aerosol. Furthermore, in-situ structural informations of particular compounds were achieved by using the MS/MS mode of the ion trap. The comparison to MS/MS spectra from laboratory generated SOA of specific monoterpene precursors indicated that laboratory SOA barely depicts the complexity of ambient SOA. Moreover, it was shown that the mass spectra of the laboratory SOA more closely resemble the ambient gas phase composition, indicating that the oxidation state of the ambient organic compounds in the particle phase is underestimated by the comparison to laboratory ozonolysis. These observations suggest that the micro-scale processes, such as the chemistry of aerosol aging or the gas-to-particle partitioning, need to be better understood in order to predict SOA concentrations more reliably.rnDuring a field study at the Mt. Kleiner Feldberg, Germany, a slightly different aerosol concentrator / APCI-IT-MS setup made the online analysis of new particle formation possible. During a particular nucleation event, the online mass spectra indicated that organic compounds of approximately 300 Da are main constituents of the bulk aerosol during ambient new particle formation. Co-located filter analysis by LC/UHRMS analysis supported these findings and furthermore allowed to determine the molecular formulas of the involved organic compounds. The unambiguous identification of several oxidized C 15 compounds indicated that oxidation products of sesquiterpenes can be important compounds for the initial formation and subsequent growth of atmospheric nanoparticles.rnThe LC/UHRMS analysis furthermore revealed that considerable amounts of organosulfates and nitrooxy organosulfates were detected on the filter samples. Indeed, it was found that several nitrooxy organosulfate related APCI-IT-MS mass traces were simultaneously enhanced. Concurrent particle phase ion chromatography and AMS measurements indicated a strong bias between inorganic sulfate and total sulfate concentrations, supporting the assumption that substantial amounts of sulfate was bonded to organic molecules.rnFinally, the comprehensive chemical analysis of the aerosol composition was compared to the hygroscopicity parameter kappa, which was derived from cloud condensation nuclei (CCN) measurements. Simultaneously, organic aerosol aging was observed by the evolution of a ratio between a second and a first generation biogenic oxidation product. It was found that this aging proxy positively correlates with increasing hygroscopicity. Moreover, it was observed that the bonding of sulfate to organic molecules leads to a significant reduction of kappa, compared to an internal mixture of the same mass fractions of purely inorganic sulfate and organic molecules. Concluding, it has been shown within this thesis that the application of modern mass spectrometric techniques allows for detailed insights into chemical and physico-chemical processes of atmospheric aerosols.rn

Visualization and validation of (Q)SAR models

Analyzing and modeling relationships between the structure of chemical compounds, their physico-chemical properties, and biological or toxic effects in chemical datasets is a challenging task for scientific researchers in the field of cheminformatics. Therefore, (Q)SAR model validation is essential to ensure future model predictivity on unseen compounds. Proper validation is also one of the requirements of regulatory authorities in order to approve its use in real-world scenarios as an alternative testing method. However, at the same time, the question of how to validate a (Q)SAR model is still under discussion. In this work, we empirically compare a k-fold cross-validation with external test set validation. The introduced workflow allows to apply the built and validated models to large amounts of unseen data, and to compare the performance of the different validation approaches. Our experimental results indicate that cross-validation produces (Q)SAR models with higher predictivity than external test set validation and reduces the variance of the results. Statistical validation is important to evaluate the performance of (Q)SAR models, but does not support the user in better understanding the properties of the model or the underlying correlations. We present the 3D molecular viewer CheS-Mapper (Chemical Space Mapper) that arranges compounds in 3D space, such that their spatial proximity reflects their similarity. The user can indirectly determine similarity, by selecting which features to employ in the process. The tool can use and calculate different kinds of features, like structural fragments as well as quantitative chemical descriptors. Comprehensive functionalities including clustering, alignment of compounds according to their 3D structure, and feature highlighting aid the chemist to better understand patterns and regularities and relate the observations to established scientific knowledge. Even though visualization tools for analyzing (Q)SAR information in small molecule datasets exist, integrated visualization methods that allows for the investigation of model validation results are still lacking. We propose visual validation, as an approach for the graphical inspection of (Q)SAR model validation results. New functionalities in CheS-Mapper 2.0 facilitate the analysis of (Q)SAR information and allow the visual validation of (Q)SAR models. The tool enables the comparison of model predictions to the actual activity in feature space. Our approach reveals if the endpoint is modeled too specific or too generic and highlights common properties of misclassified compounds. Moreover, the researcher can use CheS-Mapper to inspect how the (Q)SAR model predicts activity cliffs. The CheS-Mapper software is freely available at http://ches-mapper.org.