<p>A diesel fuel injector has been modified to allow rotationaround its axis, driven by an electric motor. Injections at upto 6000 rpm from the rotating injector have been investigatedunder the influence of air swirl on one optical research engineand one optically accessible heavy-duty diesel engine.</p><p>The experiments show that changing from a normal, staticinjection to a sweeping injection has profound effects on sprayformation, dispersion and penetration. This influences thefuel/air-mixing, autoignition, combustion rate and emissionsformation. The spray propagation is stronger influenced byinjector rotation than by air swirl.</p><p>The air entrainment into the spray increases forcounter-swirl rotation of the injector and this speeds up thevaporization and decreases the formation of soot. In addition,the oxidation of soot is enhanced since the counter-swirlinjection forces the intense fuel-rich and soot containingspray core to penetrate into fresh air instead of replenishingthe rich regions in the head of the spray. Fuel accumulationalong the piston bowl wall decreases as an effect of thereduced penetration with counter-swirl injection. Altogether,this decreases the smoke emissions for low and intermediateengine loads.</p><p>For the combustion system studied, counter-swirl rotation ofthe injector cannot decrease the smoke emissions at high engineload since the reduced spray penetration impairs the airutilization. Fast and efficient combustion at high loadrequires spray induced flame spread out into the squish region.Spray induced flow of cool fresh air from the bottom of thepiston bowl in towards the injector is also important for lowsoot formation rates.</p><p>Co-swirl rotation of the injector reduces the airentrainment into the spray and increases the soot formation.The increased smoke and CO emissions with co-swirl injectionare also attributed to the excessively large fuel-rich regionsbuilt up against the piston bowl wall.</p><p>Increased air swirl generally reduces smoke and COemissions. This is mainly an effect of enhanced burnout due tomore intense mixing after the end of fuel injection.</p><p>Changes in smoke as an effect of injector rotation aregenerally accompanied with opposite, but relatively small,changes in NO. Fast and efficient burnout is important for lowsmoke emissions and this raises both the temperature andproduction of NO. NO production is strongly influenced by thein-cylinder conditions during the latter part of themixing-controlled combustion and in the beginning of theburnout.</p><p><b>Keywords:</b>diesel spray combustion, rotating injector,air swirl, air/fuel-mixing, soot, NO, CO, flame visualization,Chemkin modeling, soot deposition</p>

Le travail présenté dans ce manuscrit concerne les mesures de taille de particule et de fraction volumique de suies dans des conditions moteur Diesel. Les techniques utilisées sont la laser-Induced incandescence (LII), la méthode d’extinction laser (LEM), la pyrométrie, et l’analyse d’images de microscopie électronique par transmission (TEM) d’échantillons prélevés in-Situ. Des stratégies de mesure de tailles de particules sont développées en se basant sur l’utilisation d’un modèle LII et en analysant la poly-Dispersion des tailles de particules, aussi bien à partir de signaux de LII résolu en temps (mesures ponctuelles) à pression atmosphérique, que d’informations résolues spatialement provenant d’images acquises a deux instants différents. Des mesures sont effectuées avec ces stratégies sur une flamme à pression atmosphérique et dans des conditions représentatives des conditions moteur Diesel pour évaluer leur applicabilité. Des mesures supplémentaires de température et de fraction volumique de suies sont aussi réalisés.Une nouvelle méthode, appelée two-Exponential reverse fitting (TERF) est introduite. Elle vise à extraire des informations sur la distribution de tailles de particules. Cette méthode est basée sur l’utilisation de fits mono-Exponentiel du signal de décroissance de LII à différents intervalles de temps. La distribution de tailles de particules est approximée par la combinaison de deux distributions de tailles de particules mono-Disperses : une petite et une large. Aucune hypothèse sur la forme de la distribution n’est nécessaire. La méthode permet aussi de fournir le ratio de la proportion respective des deux classes de particules. L’erreur systématique induite par la description mono-Exponentielle de la décroissance du signal de LII a été calculée et est inférieure à 2% pour des décroissances de signal de LII d’aggregats mono-Disperses avec des températures de chauffe pour lesquels la sublimation des suies est négligeable. La méthode a été appliquée à des données de LII obtenus sur une flamme laminaire atmosphérique éthylene/air à différentes hauteurs. Les résultats obtenus montrent un bon accord entre les tailles des grosses particules évaluées avec la méthode TERF et celle obtenue par analyse des images TEM. En revanche l’accord n’est pas obtenu pour les petites particules, ce qui est attribué à un manque d’information sur cette classe de particule dans l’analyse TEM.Des champs de fraction volumique de suies sont ensuite obtenus dans une cellule haute pression haute température dans les conditions opératoires du réseau ECN (Engine Combustion Network) par technique combinée de LEM et LII simultanées. Les mesures sont réalisées dans les conditions du spray A et incluent des variations paramétriques (température et dilution). La distance de Lift Off de la flamme est déterminée en parallèle par visualisation directe de la chimiluminescence OH. Des niveaux de fraction volumique maximale de 2-3ppm sont obtenus dans les conditions nominales du spray A (i.e. 900K), et peuvent atteindre 12 pmm à haute température (1030K). L’effet des variations de température et de concentration d’oxygène sur la formation et l’oxydation des suies et cohérente avec les résultats issus de la littérature.Une méthode d’imagerie de taille de particules est développée. Elle est basée sur l’acquisition de deux images de LII obtenues à deux instants différents après le pulse laser et l’analyse de ces images à l’aide de la simulation du signal LII pour déduire les tailles des particules à partir du rapport des images. Une stratégie basée sur une analyse par modèle LII est développée pour évaluer les incertitudes de mesure. La dépendance aux conditions limites de l’imagerie de taille de particule par LII est ainsi évaluée. [...] / This work focuses on measurements of soot particle size and volume fraction at Diesel engine conditions. A combination of laser-Induced incandescence (LII) imaging, line-Of-Sight laser extinction, soot pyrometry, and transmission electron microscopy (TEM) measurements of thermophoretically-Sampled soot was used. Particle sizing strategies were developed with LII model for the analysis of particle-Size poly-Dispersity with time-Resolved LII signal that is suitable for point-Wise measurements at atmospheric pressure, and for spatially-Resolved characterization with two-Time-Step LII imaging. Measurements were performed with these strategies in a flame at atmospheric pressure and in Diesel engine combustion to investigate their applicability. Additional measurements were performed for temperature and soot volume fraction.A novel method, called two-Exponential reverse fitting (TERF), is introduced to extract information about the size distribution. The method is based on mono-Exponential fits to the LII signal decay at a delayed time. It approximates the particle-Size distribution as a combination of one large and one small mono-Disperse equivalent mean particle size and does not require a distribution assumption. It also provides a ratio of the contribution of both size classes. The systematic error caused by de-Scribing LII signals by mono-Exponential decays was calculated as less than 2% for LII signals simulated for mono-Disperse aggregated soot with heat-Up temperatures for which evaporation is negligible. The method was applied to LII data acquired in a laminar non-Premixed ethylene/air flame at various heights above the burner. The particle size of the large particle-Size class evaluated with the method showed good consistency with TEM results, however the size of the small particle-Size class and its relative contribution could not be compared due to insufficient information in the TEM results for small particles. Simultaneous line-Of-Sight laser extinction measurements and LII imaging were performed to de-Rive the soot volume fraction in a high-Temperature high-Pressure constant-Volume pre-Combustion vessel under the Engine Combustion Network’s (ECN) "Spray A" conditions with parametric variations of gas temperature and composition. Extinction measurements were used to calibrate LII images for quantitative soot distribution measurements. OH-Chemiluminescence imaging was used to determine the lift-Off length, and used to interpret the soot measurements. Maximum soot volume fractions around 2–3 ppm were obtained at the nominal ambient temperature defined for Spray A (i.e. 900 K) that rise to 12 ppm at elevated temperature (1030 K). Variations of ambient temperature and oxygen concentration were carried out showing effects on soot formation and oxidation that are consistent with the literature.The method for particle-Size imaging is based on evaluating gated LII signals acquired with two cameras consecutively after the laser pulse and using LII modeling to deduce particle size from the ratio of local signals. A strategy was developed with a model-Based analysis: the dependence of LII particle-Size imaging on the assumed boundary conditions was identified such as bathgas temperature, pressure, particle heat-Up temperature, thermal accommodation coefficients, and soot morphology. Various laser-Fluence regimes and gas pressures were considered. Effects of laser attenuation were evaluated. A combination of one detection gate starting with the particle-Heating and the other starting with 11 ns delay with twice as long gate width was found to provide the highest sensitivity for particle sizing at 60 bar. The optimum gate delays for different pressures were calculated. The effects of timing jitter for laser pulse and poly-Dispersity were investigated. Systematic errors in pyrometry imaging at 60 bar was evaluated. [...]

Med skenade bunkerkostnader och överdimensionerade motorinstallationer har allt fler handelsfartyg börjat operera under benämningen slow steaming, vilket innebär att sotbränderna har ökat i avgaspannor/ekonomiser och i en del extrema fall har branden övergått till en vätgasbrand. Arbetet utreder främst dieselmotorer som använder restoljor som bränsle och opererar med en låg motorbelastning. Syftet med undersökningen var att se vad som leder till att oförbränt bränsle och cylinderolja ackumuleras i avgassystem samt beskriva vilka åtgärder som kan göras för att eliminera risken för att en brand uppstår i en avgaspanna alternativt en ekonomiser. Metoden som har används för undersökningen är en litteraturstudie genom datainsamling och facklitteratur inom området dieselmotorer och främst marina ångpannor. Resultaten som framkommit av undersökningen visar på svårigheten att köra en dieselmotor på ett lägre effektuttag då hela fartyget ofta är konstruerat för en drift med tjockolja där motorn belastas på ca 80% av sin effekt. Några av de slutsatser ur studien visar på att även när en felfri motor ej uppnått drifttemperatur sker en ofullständig förbränning av bränsle och smörjolja på grund av den låga temperaturen i förbränningsrummet. Detta kan medföra operativa risker genom att oförbränt bränsle och cylinderolja ackumuleras i avgassystem som blöta och sotiga beläggningar. Slutsatserna som kan dras utifrån de olika sotningsmetoder vilka har jämförts i resultatet har vattentvättning, ångsotning och rengöringsmedel med katalytisk verkan fungerat bäst på de flesta beläggningar. / With escalating bunker costs and oversized engine installations an increasing number of vessels have started to operate in a mode known as slow steaming. This means that the soot fires in exhaust gas boilers/economizers have increased and in some extreme cases led to hydrogen fires. This thesis focus on how diesel engines use residual fuels and operate with a low engine load. The purpose of this thesis is to investigate whether incomplete combustion caused by a low combustion temperature leads to the accumulation of unburned fuel and cylinder oil in the exhaust gas system and to describe which precautions eliminate the risk of a fire in the exhaust gas boiler or exhaust gas economizer. A literature review has been conducted and data has been collected in the field of diesel engine with priority given to marine steam boilers. The result of the study shows how difficult (problematic) it is to run a diesel engine at a lower power output, since the vessel often is designed for a heavy fuel oil operation when the engine load approximately 80 % output.                      Conclusions from the study show that even when a faultless engine has not reached operating temperature, incomplete combustion of fuel and lubricating oil occurs due to the low temperature in the combustion chamber. This can entail operational risks by accumulating unburned fuel and cylinder oil in exhaust systems such as wet and sticky carbon deposit. The result shows that water washing, steam cleaning and catalytic cleaning are the most efficient cleaning methods for most deposits in exhaust gas boilers and economizers.

Soot in coal combustion simulations is often ignored due to its computational complexity, despite significant effects on flame temperature and radiation. In this research, a 40 kW oxy-coal combustion system is modeled using Large Eddy Simulations (LES) and a semi-empirical monodisperse coal soot model. Simulation results are compared to experimental measurements of temperature, species concentrations, and soot concentration. Cases where soot is modeled are compared with cases where soot is neglected to determine the accuracy benefits of modeling soot. The simulations were able to replicate experimental results within an acceptable level of error. Including soot in the simulations did not consistently increase accuracy for the simulation setup and modeling assumptions used in this research.

Le carbone-suie, ou “black carbon” (BC), contribue au réchauffement climatique avec un forçage positif de l’ordre de +1.1 W m-2 dont l’incertitude reste haute (de l’ordre de 90%). Ce forçage s’effectue à travers l’interaction aérosol-radiation et l’interaction aérosols-nuage. Ces deux mécanismes sont affectés par le degré de mélange des particules du BC avec divers matériaux non-réfractaires et non-absorbants. Cependant, les estimations du forçage radiatif considèrent rarement les effets du mélange interne. Par ailleurs le rôle du BC comme noyau glaçogène qui influence l’interaction aérosol-nuage est largement inconnu. L’objectif de cette thèse est de mieux comprendre les mécanismes par lesquels le degré de mélange interne du BC influence la variabilité des propriétés optiques du BC et les propriétés d’activation des noyaux glaçogènes contenant du BC.Dans le premier chapitre de cette thèse, nous avons exploré la variabilité spatiale et saisonnière du coefficient d’absorption massique -mass absorption cross-section (MAC)- dans l’atmosphère en Europe. Les valeurs de MAC sont déterminées à partir de concentrations de carbone élémentaire et de coefficients d’absorption observée à différentes stations d’observation européenne du réseau ACTRIS (Aerosol, Cloud and Trace gases Research InfraStructure). Les résultats montrent une faible variabilité spatiale du MAC avec une moyenne de 10 ± 2.5 m2 g-1 à 637 nm de longueur d’onde qui peut être considérée comme représentative du BC en Europe. Le cycle saisonnier du MAC est probablement lié à la composition chimique de l’aérosol et son état de mélange, qui provoque une augmentation du MAC.Dans le second chapitre on s’est intéressé au lien entre l’absorption du BC et son état de mélange après transport sur longue-distance. Ce travail se base sur des mesures effectuées dans le cadre du projet CLIMSLIP (CLimate IMpact of Short-Lived Pollutants and methane in the Arctic). Une campagne de mesure a été conduite sur la station de recherche Zeppelin au Svalbard, Norvège en Avril 2012. Les données acquises avec un Single Particle Soot Photometer (SP2) révélaient que le BC est généralement présent en mélange interne dont l’épaisseur moyenne de la couche superficielle de matériel non-absorbant est de 47 nm pour des particules de BC de diamètre compris entre 170 et 280 nm. Ce mélange interne conduit à une augmentation d’absorption de 46%. Elle entraîne cependant une diminution relativement faible de l’albédo de simple diffusion, de l’ordre de 1%.Enfin, la capacité du BC à agir comme noyaux glaçogène pour la formation de cristaux de glace a été étudiée sur le site de haute altitude du Jungfraujoch (Suisse) dans le cadre du “cloud and aerosol characterization experiment” (CLACE) en 2013. Les différents éléments du nuage étaient séparé à partir d’une prise d’entrée type ice-CVI connectée au SP2. Ce dispositif permet de sélectionner uniquement les cristaux de glace et quantifier la fraction de BC activée. Une réduction de la présence de BC dans les résidus de glace a été observée. Des mesures de l’épaisseur de la couche de mélange interne des particules contentant du BC ont montré que les résidus de cristaux de glace présentaient des enrobages bien plus épais comparée à l’aérosol total.Les résultats obtenus au cours de ce travail ont permis de mieux comprendre l’impact du degré de mélange interne sur les propriétés optiques du BC et sur son rôle dans la formation de cristaux de glace. Les propriétés optiques du BC évoluent en fonction de la saison, tandis que la formation d’une couche superficielle amplifie sa capacité d’absorption du rayonnement solaire. De plus, cette étude souligne l’importance du vieillissement atmosphérique du BC sur sa capacité à servir de noyau de nucléation de la glace. Enfin, il fournit une avancée au sujet des propriétés sensibles mesurées dans l’atmosphère avec des techniques innovantes qui permettront la simulation plus précise du forçage radiatif. / Black carbon (BC) induces a warming effect (RFBC = +1.1 W m-2 ± 90%) through two main pathways: aerosol-radiation interaction (RFari) and aerosol-cloud interaction (RFaci). Both BC-radiation and BC-cloud interaction are affected by the mixing of black carbon with other non-refractory and non-absorbing matter present in the atmosphere. An estimation of the global radiative forcing of BC rarely accounts for internal mixing of BC while the net global cloud radiative forcing is sensitive to assumptions in the initiation of cloud glaciation, which is mostly unknown for black carbon particles. Within this thesis we investigated the variability of the light absorbing properties of black carbon, the mixing of black carbon, and the impact on light absorption and ice activation.In the first part of this thesis we investigated the spatial and seasonal variability of the mass absorption cross section (MAC) over Europe. MAC values were determined from ambient observations of elemental carbon mass concentrations (mEC) and absorption coefficients (σap). The data had been acquired during several years at different background ACTRIS supersites spread over Europe. Site specific MAC values were found to be spatially homogeneous, suggesting that the overall MAC average 9.5 ± 1.9 m2 g-1 at a wavelength of 637 nm might be representative of BC at European background locations. The MAC values showed a distinct seasonal cycle at every station. This seasonality might be related to chemical composition and aging. We observed that the MAC value has a linear and positive proportionality with the non-absorbing matter mass fraction.The second part of the work focuses on the coating acquisition of BC and the induced absorption enhancement after long-range transport. Within the CLIMSLIP (climate impact of short-lived pollutants and methane in the Arctic) project field experiments were conducted at the Zeppelin research site in Svalbard, Norway, during the Arctic spring. SP2 data were used to characterize the BC size distribution and mixing. BC containing particles having a core diameter between 170 and 280 nm were found to have a median coating thickness of 47 nm. The relationship between coating thickness and BC absorption was simulated. The observed coating thickness enhanced the mass absorption cross section by 46%, which led to a decrease of less than 1% in the single scattering albedo.In the final part of this work, the role of black carbon as ice nuclei in mixed phase clouds was investigated at the high elevation measuring site Jungfraujoch (Switzerland) during the cloud and aerosol characterization experiment (CLACE) held in 2013. The ice-CVI inlet and a single particle soot photometer were used to select and quantify the ice activated BC particles. According to the observations, BC containing particles were depleted in the ice residuals. The activation efficiency showed a size dependency, with larger BC containing particles being activated more efficiently compared to smaller ones. Activated BC cores having a diameter between 170 and 240 nm showed a larger coating thickness (median = 53 nm) compared to the total aerosol (median = 16 nm).The results obtained in this thesis shed new light on the effect of the mixing state on the optical properties and cloud activation of black carbon particles. Absorbing properties of BC showed a distinct seasonal pattern, while aging was found to consistently increase its absorption behavior. However, black carbon was found not to act as ice nuclei in low tropospheric mixed-phase clouds, where the coating thickness might play a role in the activation efficiency. This work provides freshly determined physical properties derived from ambient observations that will improve the accuracy of future aerosol and cloud radiative forcing estimations.

Combustion-generated carbon black nano particles, or soot, have both positive and negative effects depending on the application. From a positive point of view, it is used as a reinforcing agent in tires, black pigment in inks, and surface coatings. From a negative point of view, it affects performance and durability of many combustion systems, it is a major contributor of global warming, and it is linked to respiratory illness and cancer. Laser-Induced Incandescence (LII) was used in this study to measure soot volume fractions in four steady and twenty-eight pulsed ethylene diffusion flames burning at atmospheric pressure. A laminar coflow diffusion burner combined with a very-high-speed solenoid valve and control circuit provided unsteady flows by forcing the fuel flow with frequencies between 10 Hz and 200 Hz. Periodic flame oscillations were captured by two-dimensional phase-locked LII images and broadband luminosity images for eight phases (0°- 360°) covering each period. A comparison between the steady and pulsed flames and the effect of the pulsation frequency on soot volume fraction in the flame region and the post flame region are presented. The most significant effect of pulsing frequency was observed at 10 Hz. At this frequency, the flame with the lowest mean flow rate had 1.77 times enhancement in peak soot volume fraction and 1.2 times enhancement in total soot volume fraction; whereas the flame with the highest mean flow rate had no significant change in the peak soot volume fraction and 1.4 times reduction in the total soot volume fraction. A correlation (ƒv Reˉ1 = a+b· Str) for the total soot volume fraction in the flame region for the unsteady laminar ethylene flames was obtained for the pulsation frequency between 10 Hz and 200 Hz, and the Reynolds number between 37 and 55. The soot primary particle size in steady and unsteady flames was measured using the Time-Resolved Laser-Induced Incandescence (TIRE-LII) and the double-exponential fit method. At maximum frequency (200 Hz), the soot particles were smaller in size by 15% compared to the steady case in the flame with the highest mean flow rate.

Thermometry measurements were carried out using planar laser induced phosphorescence in conjunction with thermographic phosphors in heated turbulent jets and laminar flames in order to further develop the technique for usage in flames. Two dimensional thermometry measurements are essential to improve the understanding of combustion processes, as temperature governs soot pyrolysis, leading to soot formation. Two particular thermographic phosphors, BAM and YAG:Dy were tested and compared and it was found that they were unsuitable for gas phase flame thermometry measurements. Soot volume fraction measurements were carried out using planar two colour laser induced incandescence in gaseous and liquid fuel flames. The gas fuel flames were diluted with nitrogen, carbon dioxide and hydrogen individually and then with nitrogen and hydrogen together, as well as carbon dioxide and hydrogen together, separately. Results revealed the dilution effects of the gases on the soot formation process, where increasing nitrogen percentage in the flow decreased SVF, carbon dioxide reduced it further and hydrogen showed no marked difference. Biodiesels were compared with each other and with diesel in a wick burner in order to analyse their compositional effects on soot. Biodiesel composition was measured using gas chromatography. The sooting tendencies of the biodiesels were as expected, fuels with a longer average carbon chain length and a higher degree of unsaturation were found to produce more soot than shorter, more saturated fuels. Diesel was sootier than all of the biofuels tested, due to containing aromatics and a lower oxygen content. A pilot study was also done, where the performance and emissions of biofuels and biofuel-diesel blends were tested in a gas turbine engine, in order to relate the investigation to real world situations.

Fully understanding soot formation in flames is critical to the development of practical combustion devices, which typically operate at high pressures, and fire suppression systems in space. Flames display significant changes under microgravity and high-pressure conditions as compared to normal-gravity flames at atmospheric pressure, but the exact causes of these changes are not well-characterized. As such, the effects of gravity and pressure on the stability characteristics and sooting behavior of laminar coflow diffusion flames were investigated. To study these effects, a new highly-scalable combustion modelling tool was developed specifically for use on large multi-processor computer architectures. The tool is capable of capturing complex processes such as detailed chemistry, molecular transport, radiation, and soot formation/destruction in laminar diffusion flames. The proposed algorithm represents the current state of the art in combustion modelling, making use of a second-order accurate finite-volume scheme and a parallel adaptive mesh refinement algorithm on body-fitted, multi-block meshes. An acetylene-based, semi-empirical model was used to predict the nucleation, growth, and oxidation of soot particles. Reasonable agreement with experimental measurements for different fuels and pressures was obtained for predictions of flame height, temperature and soot volume fraction. Overall, the algorithm displayed excellent strong scaling performance by achieving a parallel efficiency of 70% on 384 processors. The effects of pressure and gravity were studied for flames of two different fuels: ethylene-air flames between pressures of 0.5–5 atm and methane-air flames between 1–60 atm. Based on the numerical predictions, zero-gravity flames had lower temperatures, broader soot-containing zones, and higher soot concentrations than normal-gravity flames at the same pressure. Buoyant forces caused the normal-gravity flames to narrow with increasing pressure while the increased soot concentrations and radiation at high pressures lengthened the zero-gravity flames. Low-pressure flames at both gravity levels exhibited a similar power-law dependence of the maximum carbon conversion on pressure which weakened as pressure was increased. This dependence decayed at a faster rate in zero gravity when pressure was increased beyond 1–10 atm.

Au cours d’un incendie dans une installation nucléaire de base (INB), les filtres à très haute efficacité (THE) sont colmatés par un dépôt (ou « gâteau ») de suies (des agrégats de nanoparticules carbonées). L’effet de l’humidité, observée au niveau du gâteau de suies par la présence d’un condensat, n’est pas encore pris en compte dans les modèles de colmatage développés dans la communauté scientifique. Dans ce contexte, la présente étude vise à mieux comprendre le phénomène de sorption de l’eau sur les suies. Pour ce faire, des suies dites « analytiques » ont été produites avec différents combustibles isolés, ceci à différentes teneurs en dioxygène, et des suies « d’incendie » ont été produites à partir d’essais de feux à grande échelle, ceci à différentes ventilations et avec des combustibles réalistes (boîte à gants, câbles électriques, huile hydraulique). Les propriétés physico-chimiques de ces suies (morphologie, porosité, surface spécifique, composition chimique et élémentaire) ont été déterminées parallèlement à l’obtention des isothermes de sorption d’eau, pour des suies sous forme de pastilles et de poudres non tassées. Les paramètres obtenus avec le modèle Dubinin-Serpinski pour une première catégorie de suies hydrophobes ont permis de proposer une modélisation pertinente des isothermes caractéristiques des suies analytiques. Par ailleurs, les isothermes de sorption de l’eau sur des gâteaux de suies dites hydrophiles et essentiellement issues de combustibles et situations réelles d’incendie, ont été modélisées à l’aide de l’équation de D’Arcy et Watt (DW). Pour cette seconde catégorie de suies, les paramètres du modèle DW apparaissent relativement dispersés, comparés à ceux obtenus pour les suies analytiques. Cette relative dispersion s’explique par des propriétés très variables des suies d’incendie et notamment par la présence importante d’oxygène et d’halogènes (chlore, phosphore). Cette étude a donc permis de mettre en évidence une adsorption plus importante pour les suies d’incendie, conduisant à la condensation capillaire, cette dernière étant favorisée pour les suies sous forme de pastilles. In-fine, la composition chimique et élémentaire des suies apparaît ainsi comme le paramètre prépondérant du phénomène de sorption de l’eau sur les suies / During a fire in a nuclear plan, the high efficiency particle air (HEPA) filters are clogged by a deposit (or “cake”) of soot, the latter corresponding to carbonaceous nanoparticles aggregates. The effect of humidity, observed on the filters by the presence of condensed water, is still not considered in the clogging models developed in scientific community. In this context, the aim of this study consists on a better understanding of the water sorption on the soot. The experimental approach was first the production of “analytical” soot with different isolated fuels and at different dioxygen concentrations, and of “fire” soot at large scale with different ventilations and complex elements (glove boxes, electrical cables, hydraulic oil). The physicochemical properties (morphology, porosity, specific surface area, elemental and chemical composition) and the water sorption isotherms, for samples at compacted pellet and powder state, have been determined. The parameters from the model of Dubinin-Serpinski, obtained for a first class of hydrophobic soot, enabled to propose a relevant model, characteristic of the analytical soot. Furthermore, water sorption isotherms on soot cake, coming from realistic fires and fuels, have been modeled with the D’Arcy and Watt (DW) equation. For this second class of hydrophilic soot, the DW parameters appear relatively more dispersed. This relative dispersion is due to the different properties of the fire soot and especially to the presence of high amounts of oxygen and halogens (chlorine, phosphor). This study enabled to highlight a more significant water adsorption on fire soot, leading to the capillary condensation, which is favored for soot compacted into pellet. Soot chemical a and elemental composition appeared to be the most influencing parameter on water sorption phenomenon

A emissão de poluentes em megacidades e áreas industriais pode ter fortes impactos no clima e na saúde. Nos últimos anos tem sido crescente a preocupação com emissões atmosféricas contendo partículas nanométricas, cuja presença, juntamente com compostos orgânicos voláteis, óxidos de nitrogênio e outros, pode resultar na formação de uma série de substâncias gasosas poluentes e na formação de aerossóis. Medições mais precisas da concentração e distribuição de tamanho de fuligem são importantes, não só do ponto de vista ambiental, mas também para a saúde humana. O objetivo deste trabalho foi determinar a distribuição de tamanho de partículas na chaminé de uma refinaria em Cubatão. Para isso foi utilizada uma abordagem baseada nos métodos de inversão, tradicionalmente usados para calcular parâmetros de aerossóis atmosféricos, ao contexto dos aerossóis de tochas industriais. Os resultados se mostraram consistentes com a literatura científica, sendo possível determinar alguns parâmetros da distribuição do tamanho de partículas provenientes da chama de uma tocha industrial utilizando um sistema lidar de três comprimentos de onda com um nível de discrepância aceitável. Um estudo do expoente de Angström foi realizado com o objetivo de validar o algoritmo de inversão desenvolvido neste trabalho. Os resultados deste estudo mostraram que os dados experimentais corroboram com as curvas teóricas e, portanto, o algoritmo pode ser utilizado como ferramenta para a medição de emissões atmosféricas provenientes de tochas industriais. O desenvolvimento deste projeto representará um passo importante, não somente do ponto de vista tecnológico, mas principalmente como recurso para tratar de problemas de emissões que futuramente poderão surgir, dentre as medidas voltadas ao controle de mudanças climáticas. / The emission of pollutants in megacities and industrial areas can have strong impacts on climate and health. In recent years there has been a growing concern about air emissions containing nanometric particles whose presence, together with volatile organic compounds, nitrogen oxides, and others, can result in the formation of a series of gaseous pollutants and aerosol. More accurate measurements of the concentration and size distribution of soot are important not only from an environmental point of view, but also to human health. The objective of this study was to determine the particle size distribution in the chimney of a refinery in Cubatao. For this it used an approach based on inversion methods traditionally used to calculate parameters of atmospheric aerosols, the context of aerosols of industrial torches. The results were consistent with the scientific literature, it is possible to determine some parameters of size distribution of particles from an industrial torch flame using a system handling three wavelengths with an acceptable level of mismatch. A study by the Angstrom exponent was carried out in order to validate the inversion algorithm developed in this work. The results of this study showed that corroborate experimental data with the theoretical curves and thus the algorithm can be used as a tool for measuring atmospheric emissions from industrial torches. The development of this project will be an important step, not only from a technological point of view, but rather as a resource to address emission problems that may arise in the future, among the measures aimed at controlling climate change.