Massive obscured clusters in the milky way

Résumé

This thesis comprises of a full search in our galaxy for the hidden population of massive stars using the near-infrared photometry with a novel method and a quantitative spectroscopy study of an O star sample in the optical and near-infrared. Massive clusters are those with mass greater than 10^4 Solar Masses . The number of known massive clusters is significantly lower than the real prediction, as in general, due to the gas and dust that surround those clusters make diffcult to find them in the optical range and due to it is necessary to use near infrared to find and observe them. Under the MASGOMAS (MAssive Stars in Galactic Obscured MAssive clusterS) project, the main objective of this thesis is to find massive stars using 2MASS photometry on the north disc of the galaxy, using LIRIS@WHT (William Hershel Telescope) as the main telescope to make the follow-up. The method that we used to find the candidates for the massive stars was to develop an automatic detection algorithm, which mainly using photometric cuts (K , (J-K) and QIR), which eliminate a large part of the late not-massive population, and an overdensity algorithm over the stars which fulfill the photometric cuts. From this study is presented the association MASGOMAS- 6, that result in two populations at two di erent distances. To find high probability massive cluster candidates, we have developed a study that critically improves the way to select the overdensities. As result of this improvement we present the association MASGOMAS-10. In both associations the massive star population is confirmed, supporting the automatic method. The second part focused on the quantitative spectroscopy study of 12 O-type stars with high resolution optical spectra (R ~ 46000), which are also present in a survey of high resolution in the near-infrared (R ~ 12000). To estimate stellar parameters we use the IACOB Grid-Based Automatic Tool (IACOB-GBAT), an accurate and fast method that compares, by means of a chi^2 algorithm, observed and synthetic H and He line profiles in an automated way. Deriving the optical stellar parameters we test the viability of deriving parameters using only near-infrared. This study also lead us to study the not well know clumping condition for our sample, creating two full grids of clumping models Finally the clumping study results in a drastic improvement for several wind lines, especially for supergiant stars.