Chemical composition of giant planet hosts

Sammanfattning: Does the formation of planets give imprints on the chemical composition of the host star? In order to test this theory a differential abundance analysis (i.e. comparing two stars' abundances to each other) was carried out on a few stars with and without known planets, to see if any differences were found. The analysis was based on high resolution, high signal to noise ratio (SNR) spectra of four stars: HD 102117, HD 90722, HIP 79672 (18 Sco) and the Sun, where two of the stars host known planets (Sun and HD 102117). Between two solar spectra, observed with two different instruments, no systematic offset was identified, and thus my result does not rely on the choice of the different spectrographs. A comparison between HIP 79672 and the Sun was performed and our results agree well with that from previous literature. When HD 102117 was compared to its stellar twin in the sample, HD 90722, I found that HD 102117 is more metal-poor by 0.05 dex. This difference could possibly be due to the formation of its gas giant. As a rough estimate, adding 7-9 M$_\oplus$ of an equal mixture of earth-like material and meteorite material to the convective zone of HD 102117 would even out the difference in average refractory abundance, in comparison to HD 90722. A trend with condensation temperature (T$_c$, the temperature at which an element condensate) was found as well. Since the mass of the planet's core is small, the imprinted T$_c$ trend is not very clear, but could be improved with higher precision.