Estimating the binary fraction of central stars of planetary nebulae

Planetary nebulae are the end-products of intermediate-mass stars evolution, following a phase of spherical expansion of their atmospheres at the end of their lives. Observationally, it has been estimated that 80% of them have non-spherical shapes. Such a high fraction is puzzling and has occupied the planetary nebula community for more than 30 years. One scenario that would allow to justify the observed shapes is that a comparable fraction of the progenitors of central stars of planetary nebula (CSPN) are not single, but possess a companion. The shape of the nebulae would then be the result of an interaction with this companion. The high fraction of non-spherical planetary nebulae would thus imply a high fraction of binary central stars of planetary nebula, making binarity a preferred channel for planetary nebula formation. After presenting the current state of knowledge regarding planetary nebula formation and shaping and reviewing the diverse efforts to find binaries in planetary nebulae, I present my work to detect a near-infrared excess that would be the signature of the presence of cool companions. The first part of the project consists in the analysis of data and photometry acquired and conducted by myself. The second part details an attempt to make use of archived datasets: the Sloan Digital Sky Survey Data Release 7 optical survey and the extended database assembled by Frew (2008). I also present results from a radial velocity analysis of VLT/UVES spectra for 14 objects aiming to the detection of spectroscopic companions. Finally I give details of the analysis of optical photometry data from our observations associated to the detection of companions around centrals star of planetary nebula using the photometric variability technique. The main result of this thesis is from the near-infrared excess studies which I combine with previouslypublished data. I conclude that if the detected red and NIR flux excess is indicative of a stellar companion then the binary fraction is larger than what we may expect based on the main-sequence progenitor population binary fraction and therefore conclude that binarity is a preferential channel for the formation of planetary nebula. I finish by underlining the need for a sample size of ~150 objects to decrease the uncertainty on the planetary nebula population binary fraction and increase the statistical significance of this result.