J. Maldonado1, E. Villaver2, C. Eiroa2, G. Micela1

1INAF - Osservatorio Astronomico di Palermo, Piazza Parlamento 1, 90134 Palermo, Italy. E-mail: [email protected]

2Universidad Autónoma de Madrid, Dpto. Física Teóri ca, Facul tad de Ciencias, Campus de Cantoblanco, 28049 Madrid, Spain

The role of the host star's metallicity in planet formation has been largely discussed in the framework of the so-called gas-giant/planet metallicity correl ati on. However, previous works are mainly focused on particular kinds of stars or planets. In this contribution we aim to put together all the pieces of the planet formation puzzle by analysing in the most homogeneous possible way a large sample of stars (without any restriction on spectral type or evolutionary status) showing all the possible outcomes of the planet formation process (from debris discs to massive brown dwarfs).

Our sa mp le is co m po s ed of 551 stars : 71 F-type, 261 G- type, 166 K-type, and 53 early-Ms. 373 stars are in the Main-Se q ue n ce , 63 are subgiants, and 115 ar e giants.

The di ffer en t ar chi te ctur es of the planetary systems harboured by our sample are listed in Ta bl e 1, while figure 1 shows the HR diagram. Fig. 1.. Luminosity versus Teff diagram for the observed stars. Green: stars with debris discs; filled circles: stars with planets; filled stars: stars with brown dwarf companions. Fig. 3. Stellar [Fe/H] of the host stars as a function of the minimum mass of the substellar companions. Different colours and symbols indicate the mass of the host star. Our sa m pl e is selected fr om our previous works (Maldonado et al.2012, 2013, 20 1 5a, 2015b,20 1 8 ; Maldonado & Villaver 2016, 2017). Metallicities ar e de te rm in e d by using To give a br oa de r context, Figure 4 shows the host star me tal li ci ty as a function of the the ir on ionisation and equi l ibri um conditions, except for the M dwarfs for which a substellar companion mass, including da ta from low-mass bi n ar ie s and debris disc’s methodology based on the use of pseudo equi vale nt widths of spectral features were masses when available. Th e data for low-ma ss K, M binaries comes from Ma n n et al. used. (2013).

The figure shows that the tendency of a wider range of metallicities (l owe r values) towards more massive objects continues inthe low-ma ss stellar range.

Neither low-mass ho st stars, nor stars wi th di scs show metal enri chme nt. In fact, a tendency of wider me tal l i ci tie s towards less ma ssi ve planet formation’s outcomes ma y Table 1. Architecture of the pl an e ta r y systems also be present inthe data. in our stellar sample. We conclude that there is no an un i ver sal pl an e t formation me ch an i sm . Di ffer e nt mechanisms ma y operate altogether and their relati ve efficiency change with the mass of the substellar object that is formed.

The core-accretion me cha n i sm for planet formation would ha ve it highestefficiency for forming planets wi th masses ar ound 1 Jupiter’s mass. Towar d s mo re ma ssi ve su b s tel la r Figure 2 compares the cumulative distribution function of the stellar me ta ll i ci ty of objects the efficiency of the co re -accretion di m in i sh e s and the pr ob ab i li ty of forming different subsamples: fr o m stars hosting low-mass planets to stars with massive objects in a similar way as stars incre ase s. brown dwarfs. The figure sh o w s a tr a n si t i on towards lower metallicities as the ma ss of the substellar object increases, be in g the stars har b ou ri n g low-ma ss planets the only exception to this trend. On the other side, small planets and planetesimals may still be formed by core-accretio n arou nd Figure 3 sh o w s the ho st star low-met allicity stars. The low- metallicity enviro nment im plies metallicity as a function of the long times for forming a core able (minimum mass) of the to accret gas before the disc’s substellar comapnions. Colours dissipati on, so only small planets and symbos indicate the mass of and planetesimals can be formed. the host star. Two ma i n conclusions can be dra wn fr o m this figure:

o There is a tendency of lower Fig. 4. [Fe/H] of the host stars host star’s me tal l i ci tie s as the as a function of the debris mass of the substellar disc’s ma ss (green), the companion increases. In (minimum) ma ss of the subtellar co mp an io n s (open agreement with Figure 2. blue circles), and the ma ss of the low-mass K and M stellar o That more massive planets companions (r ed filled circles). tend to or bi t around more massive stars. Acknowledgements: This research was supported by the Italian Ministry of Education, University, and Fig. 2. [Fe/H] cumulative frequencies for different subsamples. Research through the PREMIALE WOW 2013 project. J. M. acknowledges support from the Ariel ASI-INAF agreement. Background im ag e : Artist's impression of a baby Solar System, showing the formation of gas giants from the surrounding nebula. Image credit: NASA/JPL-Caltech.