The detection and research of exoplanets is one of the most forward and the hottest topics in the world. The observational data of exoplanets increased enormously over the last two decades. However, the formation and evolution of exoplanets are still unclear. Many observational findings on exoplanets even are beyond the predictions by the current formation and evolution theory of exoplanets. In this thesis, by studying the mass transfer process of a Roche-lobe filling planet and its host star, we have established an angular momentum evolution model for the mass transfer process of a planet-star system. Additionally, we have developed a distortion model with investigating the deformation and disruption of a planet spiraling into its host star. Several possible evolutionary fates of planets are presented here. Our results are expected to play a key role in further study and predicting the observational properties of planetary evolution. Moreover, our studies provide important theoretical predictions for the effect of planet engulfment on the properties of its host star. The main results are summarized as follows: (1) We have investigated the mass transfer from a planet to its host star, with emphasis on the conditions under which Roche lobe overflow becomes dynamically unstable. We also have systematically analyzed the effect of the planet-star system structure and stellar structure on the angular momentum evolution during the mass transfer process. We find that the orbital angular momentum of a planet is too small to affect the rotation of its (main sequence) host star. This leads to that most of the angular momentum of the mass-transferring material from the Roche-lobe filling planet is accreted by its host star. We have built an improved model of the evolution of (planet-star) system angular momentum for the case of mass transfer from a Roche lobe filling planet to a main sequence (or larger) star, which is called as the 'minimal assumption' model. Our model is expected to be an important theoretical basis for studying the mass transfer process of a planet-star system. (2) Based on the 'minimal assumption' model, we explored the conditions for dynamically unstable Roche lobe overflow as a function of planetary mass and mass and radius of host star and equation of state of planet. It is found that gas giant planets in a range of mass and entropy can undergo dynamically unstable mass transfer (in a Sun-like host star case). Dynamical mass transfer of rocky planets depends somewhat sensitively on the equation of state used. Silicate planets in the range 1 ME < MP < 10 ME (ME is Earth mass) typically go through a phase of dynamically unstable mass transfer before settling to slow overflow when their mass drops to less than 1 ME. The higher likelihood of dynamically unstable Roche lobe overflow in our model significantly increases the possibility of discovering a planet during the process of rapid mass transfer. (3) We have investigated the processes leading to the deformation and destruction of the planet which spirals into the convective envelope of its host star. We also have systematically analyzed the possible physical processes during the spiral-in process of the planet: the compression, accretion and ablation. We have developed the distortion and disruption model, and then simulated the planetary trajectories and distorted shapes during the spiral-in processes of several kinds of planets. We find that the process of ablation (slow peeling of the surface) is ineffective and the actual disruption of the planet is likely to take place in the form of a global deformation (splitting) rather than by ablation during its spiral-in process. Before disruption of the planet, it is compressed by the ram pressure and deformed into a flattened shape. The compression increases the planetary density and its gravitational binding energy, delaying the disruption of the planet. The timescale of spiral-in process of the planet is of the order of the planetary initial orbital period (about 104 s). For some combinations of mass and composition, a planet can survive its path through the entire convective envelope and disrupt finally in the radiative interior. In this way, planets can increase the metallicity in the interior rather than the convective envelope. Our results show that planet engulfment may not exhibit the observational metal- rich properties of the surface of host stars. (4) We have estimated the depths of disruption of rocky (iron) planets by ram pressure, and have analyzed the subsequent fragmentation evolution of the planetary remnants. We find that instead of quickly mixing through the convection zone, the debris of rocky or iron planets sink below the base of the convection zone. The sinking timescale is of the order of a few orbital periods of the planet. Therefore, the planet is engulfed by its host star and is disrupted in the stellar convection zone, although the metallicity of stellar interior will be enhanced by planetary material and the metal enhancement will not be seen on the surface of its host star. Our results show that a sufficient mass of rocky and/or iron planet(s) polluting the interior of the Sun may provide an explanation for the current discrepancy between helioseismic evidence and the solar interior models.
修改评论