Heterogeneous reactions of tropospheric trace-gases on solid model aerosol surface: a laboratory study

The first part of this laboratory study deals with heterogeneous reactions of bromine- and chlorine-containing species relevant to the atmosphere with solid model alkali salt substrates. The second part of this work treats the heterogeneous interaction of acidic compounds with various mineral dust substrates. The aim of this study is to investigate the role of adsorbed water, H2O(a) and the influence of defect surface sites on the reactions. Furthermore, reaction mechanisms are proposed for the different interactions. Steady state and pulsed valve uptake experiments have been performed in a low-pressure flow reactor. In order to characterise the solid samples images have been taken using a scanning electron microscope (SEM). Furthermore, a few scans using a photoacoustic cell coupled to a FTIR spectrometer have been performed. First, the interaction of gaseous Cl2, HCl, BrCl, Br2, Cl2O and HOCl with solid KBr and furthermore, the reaction of Cl2O and HOCl on solid NaCl have been investigated. The reaction products of the reactions of Cl2, Cl2O and HOCl with KBr have been Br2 and BrCl. The reaction of Cl2O with KBr leads to slow additional formation of HOCl, BrOCl and Cl2, and the reaction of HOCl with KBr leads to slow formation of HOBr, Cl2O, Br2O and BrOCl. The reaction of Cl2O on solid NaCl leads to formation of Cl2 but also a slow formation of HOCl has been observed which is attributed to the reaction with adsorbed H2O(a). The interaction of HOCl with NaCl was immeasurably slow under our conditions. In order to measure the amount of adsorbed H2O(a) desorption experiments have been carried out and a calibration curve has been obtained. While performing desorption experiments on ground KBr grains and thin KBr films sprayed on a gold-coated substrate desorption of HOBr has been observed which we attributed to the presence of surface adsorbed molecular bromine. Using the SEM we found that pumping and heating the substrate lead to a better-ordered surface. The exposure of the salt sample to an excess of chlorine or bromine leads also to a recrystallisation of the surface. Furthermore, all reactions on alkali salt samples have shown that adsorbed halogen species are retained on the surface. These adsorbed species play a crucial role for the reactions of halogens with alkali salt. In the case of HOCl on KBr we explain the autocatalytic behaviour of the reaction by the presence of adsorbed halogen species, particularly Br2. In the Cl2/KBr system a regeneration of the surface has been observed which we explain with desorption of adsorbed intermediates, but also by crystallisation of KCl that may lead to a recycling of reactive sites. In contrast to Cl2/KBr no regeneration of the substrate towards formation of gaseous bromine-containing species has been observed in the HOCl/KBr system. We explain this with the fact that KOH thermodynamically does not tend to crystallise in the presence of H2O as does the alkali halide. In the second part the reactions of CO2, SO2, HCl and HNO3 on various CaCO3 substrates such as roughened and polished marble and low- and high-ordered precipitated CaCO3 have been carried out. Additionally, a few HNO3 uptake experiments have been performed on dust substrates such as Saharan Dust, Kaolinite and Arizona Road Dust. On these substrates a fast uptake of HNO3 but no reaction products have been observed. In the case of SO2 and HCl interacting with precipitated CaCO3 the observed reaction product has been CO2. Whereas HNO3 uptake experiments on precipitated CaCO3 additionally lead to a slow formation of H2O. HCl uptake experiments on polished marble plates have not led to formation of any reaction product. However, HNO3 uptake experiments on polished and roughened marble led to formation of H2O but not of CO2. We explain these observations with the dependence of the rate of surface chemistry on the morphology of the sample. The HCl and HNO3 uptake experiments on precipitated CaCO3 have shown a delay in the formation of CO2 whereas it is immediately released after the exposure of the CaCO3 sample to SO2. Furthermore, an uptake of CO2 has been observed on precipitated CaCO3. All uptake experiments on CaCO3 have shown saturation due to a limited number of reactive sites. In order to explain these results we have speculated the existence of an intermediate species Ca(OH)(HCO3). This intermediate may be formed by exposure of the CaCO3 sample to ambient H2O and CO2. Acidic compounds such as HCl and HNO3 react rapidly with the basic OH part of the intermediate and subsequently more slowly with the bicarbonate part of the intermediate. This explains the delay of formation of CO2 in the reaction of HCl and HNO3 with CaCO3. In contrast to acidic species, SO2 attacks with preference the bicarbonate part of the intermediate. This leads to an immediate release of CO2 after exposure of the CaCO3 sample to SO2. The reaction of SO2 with the OH part of the intermediate leads to loss of SO2. The mass balance between loss of SO2 and yield of CO2 has shown a deficiency of CO2, which may be explained with a slow reaction with the OH part of the intermediate.

    Thèse École polytechnique fédérale de Lausanne EPFL, n° 2746 (2003)
    Section des sciences et ingénierie de l'environnement
    Faculté de l'environnement naturel, architectural et construit
    Institut des sciences et technologies de l'environnement
    Laboratoire de pollution atmosphérique et du sol
    Jury: Isabelle Bey, Christos Comninellis, John Crowley, Johannes Staehelin, Hubert van den Bergh

    Public defense: 2003-5-9


    Record created on 2005-03-16, modified on 2016-08-08

Related material