Les travaux de cette thèse reposent sur l’étude des processus hétérogènes à la surface desparticules minérales en présence d’irradiation UV-A. Nous savons que les poussièresminérales contiennent des oxydes métalliques pouvant absorber la radiation solaire et ainsiactiver une chimie très différente de celle observée à l’obscurité. Un réacteur à écoulementd’aérosols a été utilisé pour étudier les interactions des gaz (SO2, NO2 et O3) avec devéritables poussières minérales, évitant ainsi les artéfacts de mesure liés à la naturemacroscopique des films comme dans les études précédentes.La mise en suspension des poussières minérales a permis d’observer une formation inattenduede nouvelles particules ultrafines en présence de SO2. Le mécanisme proposé pour expliquerce phénomène de nucléation suggère une désorption de radicaux OH photoproduits à lasurface des minéraux vers la phase gazeuse. Ce mécanisme a pu être corroboré par descampagnes de mesure en atmosphère réelle. Nous avons étudié la chimie des échantillons de réelles cendres volcaniques issus de la dernière éruption du volcan Eyjafjallajökull en Islande (2010). Ceci nous a permis d’élaborerdes cinétiques de capture du SO2 sur des films macroscopiques de cendres aboutissant à descoefficients de capture de l’ordre de 10-7. Ces cinétiques couplées à des analyses chimiquesont permis de proposer un mécanisme réactionnel expliquant la formation de sulfate de fer àla surface des cendres. Finalement, nous avons étudié les interactions photochimiques de O3 et NO2 sur les poussièresminérales dans le réacteur à écoulement mettant en évidence un bon accord avec des étudesantérieures sur des surfaces macroscopiques / The objective of this work is to study the heterogeneous processes of mineral dust surfacesunder UV-A radiation. It is know that mineral dust containing metal oxides which can absorbsolar radiation and therefore activate a different chemistry compared to that observed in thedark. In order to avoid measurement artifacts related to the nature of macroscopic films, anaerosol flow tube was developed during this work and applied to study the interactions ofSO2, NO2 and O3 with real mineral dust.An unexpected formation of new particles in the presence of SO2 was observed. In order toexplain this phenomenon, we suggest the desorption of OH radicals from the mineral dustsurface to the gas phase. This mechanism has also been supported by field campaigns.Using real samples of volcanic ash from the last eruption of Eyjafjallajökull in Iceland (2010)allowed us study capture of SO2 on macroscopic ashes films with uptake coefficient around10-7. Associated kinetic experiments combined with chemical analysis allowed us to propose areaction mechanism explaining the formation of iron sulfate on the surface of ashes.Finally, we investigated the photochemical interactions of O3 and NO2 with minerals dustaerosols in the flow tube reactor showing a good agreement with previous data obtained onmacroscopic surfaces.

<p>  </p> <p>Concrete is the second most consumed material in the world after water and is an essential element of constructed infrastructure. Over 14 billion m3 of concrete are being produced annually, resulting in a serious impact on the environment. The production of cement, which is the main component of concrete, is responsible for 5 – 8 % of global CO2 emissions. As a result, several global initiatives have been undertaken to achieve carbon neutrality by 2050. This carbon neutrality target coincides with the Paris Agreement's goal to limit global warming to 1.5 °C. A well-known, and successful strategy to reduce CO2 emissions in the concrete industry is to use supplementary cementitious materials (SCMs) as a partial replacement for cement. However, it is projected that in 2030 the demand for two of the most commonly used SCMs, fly ash and slag cement, will exceed their supply. Using nontraditional and natural pozzolans (NNPs) can help to close this supply gap, but there is a lack of knowledge regarding the reactivity and long-term performance of these materials.</p> <p>The purpose of this research was to perform experiments on several NNPs, some of which can be supplied in commercially viable quantities with the objective of evaluating their performance in cementitious systems (mortars and concretes) with the goal of accurately assessing their potential for use as alternative SCMs. The mortar study was performed using a total of 11 different NNPs, belonging to 4 distinctive groups and distributed as follows: 3 from the group of calcined clays (CCs) - CC1, CC2, and CC3, 3 from the group of natural pozzolans (NPs) - NP1, NP2 and NP3, 2 from the group of fluidized bed combustion (FBCs) ashes - FBC1 and FBC2, and 3 from the group of bottom ashes (GBAs) - GBA1, GBA2, and GBA3.</p> <p>The concrete study was performed on 4 different materials, one from each of the previously mentioned groups. The materials selected for concrete study were the worst-performing members of each group, as determined by the analysis of the test results obtained from mortars. These included CC2, NP3, FBC1, and GBA3 materials. This approach was adopted under the assumption that achieving adequate concrete characteristics with lowest-quality materials will all but assure satisfactory performance of concretes with higher-quality materials. </p> <p>The findings generated from this research indicate that several of the NNPs used in this study present a viable alternative to traditional SCMs. As an example, out of the 11 NNPS, 9 were found to conform to the requirements of the ASTM C618-19, the standard specification currently used to assess the suitability of coal fly ash and raw or calcined natural pozzolans for use in concrete. Results obtained from tests performed on mortars demonstrated that, when used at the replacement level of 25%, all 11 NNPs produced mixtures with characteristics similar to those obtained from the plain cement (OPC) mortar. For that reason, this level of replacement was selected to prepare concrete specimens. The results collected from concrete specimens showed that, when compared to plain concrete, mixtures with all 4 NNPs attained comparable (or improved) mechanical (compressive and flexural strength), durability (freeze-thaw resistance), and transport (formation factor and rate of water absorption) properties. As in the case of traditional SCMs, the mixtures with NNPs were found to require extended curing times to fully realize their property-enhancing potential associated with pozzolanic reactions. Overall, the best performing materials were those from the CCs group, followed by those belonging to, respectively, NPs, GBAs, and FBCs groups. </p>