Дисертації з теми "Stellar feedbacks"

Les conditions de formation des étoiles est un sujet central en astrophysique. Le taux de formation stellaire (SFR) est relié à la masse de gaz moléculaire par la relation de Schmidt-Kennicutt. Une étoile modifie son nuage parent grâce aux vents, jets et à son rayonnement, balayant son environnement, détruisant des sites de formation d’étoiles, mais pouvant aussi en compresser et déstabiliser, déclenchant la formation de nouvelles étoiles. Ma thèse s’est concentrée sur la rétroaction radiative, largement dominée par celle des étoiles massives. Cela crée une région ionisée en expansion au plus près de l’étoile, suivie d’une région où l’hydrogène moléculaire est dissocié (photodissociation region en anglais, ou PDR), trop chaude pour former des étoiles. De nombreux modèles physico-chimiques des PDRs cherchent un état stationnaire, négligeant la dynamique du gaz. Des observations Herschel en CO excité et ALMA (Atacama Large Millimeter Array) en CH+ et SH+ ont changé la vision stationnaire de la structure des PDRs en soulignant le rôle de la dynamique du gaz. Le bord des nuages se trouve être à haute pression, fortement corrélée à l’intensité du champ UV incident. Le mécanisme de photo-évaporation peut reproduire ces caractéristiques: avec l’évaporation à haute vitesse du gaz chaud ionisé, l’effet fusée fait se propager une onde de pression dans le nuage, expliquant les hautes pressions observées. Par l’érosion du nuage, la frontière avec le milieu ionisé, le front d’ionisation (IF), avance dans le milieu neutre. Les modèles PDRs tant numériques que théoriques doivent être mis à jour pour prendre en compte cette propagation de l’IF. Nous avons d’abord construit un modèle semi-analytique de la transition entre le gaz atomique et moléculaire (H/H2) tenant compte de l’avancement de l’IF. Nous avons montré que la largeur de la région atomique est réduite comparé à des modèles statiques. Elle peut même disparaître si la vitesse de l’IF dépasse une valeur seuil, menant à la fusion de l’IF et de la transition H/H2. Nous avons trouvé des formules pour estimer ce seuil ainsi que la colonne densité totale de H atomique. En comparant notre théorie avec des observations de PDRs, nous avons montré que les effets de la dynamique sont forts, en particulier pour les PDRs faiblement illuminées comme la nébuleuse de la Tête de Cheval. En préparation des observations JWST de H2, nous avons implémenté le calcul des populations des niveaux de H2 dans le code Hydra, un code hydro-dynamique dépendant du temps modélisant les PDRs en photo-évaporation. L’étude précédente nous a permis de conclure que les effets dynamiques amène du H2 dans une région plus chaude et plus illuminée. Le rapprochement de la transition H/H2 réduit l'intensité absorbée par les poussière, qui est alors convertie en pompage UV de H2 (amplification d'un facteur 6 trouvé pour la Barre d'Orion mais peu efficace dans la Tête de Cheval). En addition, nous avons étudié des observations ALMA de la Tête de Cheval à haute résolution spatiale montrant une grande proximité entre l’IF et la molécule CO, présente habituellement profondément dans le nuage. Nous trouvons une borne supérieure à la largeur de la région atomique à quelques centaines d’unités astronomiques. Nous trouvons que le code PDR statique et stationnaire de Meudon reproduit la largeur de la région atomique sous la contrainte, tout comme les modèles dynamiques. Ces observations ne permettent donc pas de contraindre les effets dynamiques.Nous avons effectué une étude d’observations à haute résolution spectrale de raies d’émission de H2 faites par le spectrographe IGRINS. Nous montrons que les rapports de raies contraignent peu les conditions physiques, mais que le peuplement des états de H2 est fortement influé par des relaxations induites par collision, contrairement à l'image classique d'une cascade majoritairement radiative après pompage UV<br>The conditions of formation of stars is a fundamental question of astrophysics. The star formation rate (SFR) is linked to the mass of molecular gas by the Schmidt-Kennicutt relation. However, a star applies some feedbacks on its parent cloud in the form of winds, jets and radiation. They sweep their environment, destroying other star formation sites, but can also compress and destabilize them, triggering the formation of new stars. My thesis focused on the radiative feedback, which is vastly dominated by the one of massive stars. It creates an expanding region where the gas is ionized close to the star, followed by a region where the chemistry is dominated by photons capable of dissociating molecular hydrogen (photodissociation region, or PDR) which includes a layer of atomic hydrogen, which is too hot to form stars. Its width informs us about the fraction of gaz unable to form stars. Numerous models describe the physics and chemistry of PDRs by looking for a stationary state, and neglecting the gas dynamics. However, new observations made by Hershel in excited CO, and by the Atacama Large Millimeter Array (ALMA) in CH+ and SH+ have changed the stationary vision of PDR structure by highlighting the role of the gas dynamics. The edge of clouds is found to be a high-pressure environment, which is strongly correlated to the impinging UV field intensity. The photo-evaporation mechanism is capable of reproducing those features: with the high-speed evaporation of hot ionized gas, the rocket effect makes a pressure wave propagate inside the cloud, explaining the high pressures observed. By the erosion of the cloud, the border withe the ionized medium, the ionization front (IF) advances into the neutral medium. PDR models have to be updated to take into account the propagation of the IF.We built a semi-analytical model of the transition between atomic and molecular gas (H/H2) including the advancing IF. We obtained that the width of the atomic region is reduced compared to static models. It can also disappear if the IF velocity exceeds a threshold value, leading to the merging of the IF and the H/H2 transition. We found analytical formulas to estimate this threshold as well as the total column density of atomic H. By comparing our theory to PDRs observations, we showed that the dynamical effects are strong, especially in the case of weakly illuminated PDRs such as the Horsehead.To prepare for the JWST observations of H2, we have implemented the computation of H2 levels in the Hydra code, which is a hydro-dynamic, time dependent code that models the physics and chemistry of photo-evaporating PDRs. The precedent study allowed to conclude that dynamical effects bring some H2 in a hotter and more illuminated region. The reduction of the IF-H/H2 distance reduces the intensity absorbed by dust, which is then converted to UV-pumping of H2 (amplification by a factor 6 for the Orion Bar, but not efficient in the Horsehead).In addition, we studied ALMA observations of the Horsehead with high spatial resolution. They show a great proximity between the IF and the CO line emission, usually present deep in the cloud. We find an upper limit of a few hundred astronomical units for the width of the atomic region. We find that isobaric, static and stationary Meudon PDR models reproduce the width of the atomic region within the limit found, and so does the dynamical models. These observations therefore do not allow us ton constrain dynamical effects.We performed a study on high spectral resolution observations of rotation-vibration lines of H2 made by the IGRINS spectrograph. We show that the line ratios do not constrain well the physical conditions, but that the population of the states of H2 are much influenced by relaxation rates induced by collisions, unlike the classical picture of a cascade mainly dominated by radiation after the UV pumping

This thesis contains a study of the mechanical feedback from winds and supernovae on inhomogeneous molecular material left over from the formation of a massive stellar cluster. Firstly, the mechanical input from a cluster with three massive O-stars into a giant molecular cloud (GMC) clump containing 3240M⊙ of molecular material within a 4 pc radius is investigated using a 3D hydrodynamcial model. The cluster wind blows out of the molecular clump along low-density channels, into which denser clump material is entrained. The densest regions are surprisingly resistant to ablation by the cluster wind, in part due to shielding by other dense regions close to the cluster. Nonetheless, molecular material is gradually removed by the cluster wind during which mass-loading factors in excess of several hundred are obtained. Because the clump is very porous, 60-75% of the injected wind energy escapes the simulation domain. After 4.4Myrs the massive stars in the simulation start to explode as supernovae. The highly structured environment into which the SN energy is released allows even weaker coupling to the remaining dense material and practically all of the SN energy reaches the wider environment. Secondly, the X-ray emission from the simulated stellar cluster is presented. The GMC clump causes short–lived attenuation effects on the X-ray emission of the cluster. However, once most of the material has been ablated away by the winds the remaining dense clumps do not have a noticable effect on the attenuation compared with the assumed interstellar medium (ISM) column. The evolution of the X-ray luminosity and spectra are presented, and synthetic images of the emission are generated. The X-ray luminosity is initially high whilst the winds are “bottled up”, but reduce to a near constant value once the GMC clump has been mostly destroyed. The luminosity decreases slighly during the red supergiant phase of the stars due to the depressurization of the hot gas. However, the luminosity dramatically increases during the Wolf-Rayet stage of each star. The X-ray luminosity is enhanced by 2-3 orders of magnitude for at least 466 yrs after each supernova explosion, at which time the blast wave leaves the grid. Comparisons between the simulated cluster and both theoretical models and observations of young stellar clusters are presented. Thirdly, the radio emission from the simulated cluster is presented. Similar to the X-ray emission, the thermal radio emission is intially high when the winds are confined in the GMC clump and reduce as the material is ablated away. The evolution of the radio flux density and spectra are presented, and synthetic images of the emission are generated. The radio emission is compared with the X-ray results throughout the evolution of the cluster. The flux density increases during the RSG phase, and remains high during the WR phgase of the stars. The radio flux density is enhanced by three orders of magnitude during the first supernova explosion. Comparisons between the simulated cluster and observations of young stellar clusters are made. Finally, a preliminary investigation of the interaction of stellar winds within a massive cluster are presented. The hydrodynamcial simulations examine the energy and mass input of a stellar cluster into the ISM.