2 Observations and spectroscopic analysis

2.1 Observations and data reduction

High resolution spectra for more than 50 planet hosts stars were obtained during several runs using 6 different spectrographs. The general characteristics of these instruments are presented in Table 1.

 

 

Table 1: Spectrographs used for the current study and their spectral coverage and resolution.

Spectrograph/Telescope	Resolution	Coverage
	($\lambda$/ $\Delta\lambda$)	(Å)
CORALIE/1.2-m Euler Swiss	50 000	3 800-6 800
FEROS/1.52-m ESO	48 000	3 600-9 200
UES/4-m William Hershel	55 000	4 600-7 800
SARG/3.5-m TNG	57 000	5 100-10 100
UVES/VLT 8-m Kueyen UT2	110 000	4 800-6 800
ELODIE/1.93-m OHP	48 000	3 800-6 800

The spectra have in general a S/N ratio between 150 and 400, but are as high as 1000 for the UVES spectra. Except for the UES spectra, all the others cover very well all the spectral domain without any significant gaps, permitting us to measure the Equivalent Widths for most of the spectral lines used (see Santos et al. (2000) - hereafter Paper I - and Paper II). But even in this case, the gaps did not imply any strong limitations, since the available lines still have a wide variety of equivalent widths and lower excitation potentials, essential to the precise determination of the stellar parameters (see next section).

 

 

Table 2: Stellar parameters derived in the current study. $\xi _t$ denotes the microturbulence parameters. For a list of the planet discovery papers see tables at http://obswww.unige.ch/exoplanets/ and http://cfa-www.harvard.edu/planets/. A more complete table with the number of Fe I and Fe II lines used and the dispersions will be available at the CDS via anonymous ftp to cdsarc.u-strasbg.fr (130.79.128.5) or via http://cdsweb.u-strasbg.fr/cgi-bin/qcat?J/A+A/398/363

Star	$T_{{\rm eff}}$	$\log{g}$	$\xi _t$	[Fe/H]	Inst.	$M_\star^{\dag\dag\dag }$	Star	$T_{{\rm eff}}$	$\log{g}$	$\xi _t$	[Fe/H]	Inst.	$M_\star^{\dag\dag\dag }$
	(K)	(cgs)	(km s-1)		(a)	($M_{\odot}$)		(K)	(cgs)	(km s-1)		(a)	($M_{\odot}$)
HD 142	6290	4.38	1.91	0.11	[2]	1.26	HD 117176	5530	4.05	1.08	-0.05	[4]	0.92
HD 2039	5990	4.56	1.24	0.34	[1]	1.20	HD 128311	4950	4.80	1.00	0.10	[3]	0.76
HD 4203	5650	4.38	1.15	0.40	[2]	0.93	HD 130322	5430	4.62	0.92	0.06	[4]	1.04
HD 4208	5625	4.54	0.95	-0.23	[2]	0.86	HD 134987	5780	4.45	1.06	0.32	[4]	1.05
HD 8574	6080	4.41	1.25	0.05	[4]	1.17	HD 136118	6175	4.18	1.61	-0.06	[4]	1.28
HD 9826	6120	4.07	1.50	0.10	[4]	1.29	HD 137759	4750	3.15	1.78	0.09	[4]	-
HD 10697	5665	4.18	1.19	0.14	[4]	1.22	HD 141937	5925	4.62	1.16	0.11	[3] $^{\dag\dag }$	1.10
HD 12661	5715	4.49	1.09	0.36	[3]	1.05	HD 143761	5835	4.40	1.29	-0.21	[4]	0.95
HD 19994	6165	4.13	1.49	0.23	[1]$^\dag$	1.34	HD 145675	5255	4.40	0.68	0.51	[4]	0.90
	6250	4.27	1.56	0.30	[2]$^\dag$	1.35	HD 147513	5880	4.58	1.17	0.07	[1]	1.11
	6105	4.02	1.51	0.18	[5]	1.34	HD 150706	6000	4.62	1.16	0.01	[3]	1.21
(average)	6175	4.14	1.52	0.21		1.34	HD 160691	5820	4.44	1.23	0.33	[1] $^{\dag\dag }$	1.10
HD 20367	6100	4.55	1.31	0.14	[6]	1.17	HD 168443	5600	4.30	1.18	0.06	[4]	0.96
HD 23079	5945	4.44	1.21	-0.11	[2]	1.00	HD 177830	4840	3.60	1.18	0.32	[4]	1.03
HD 23596	6125	4.29	1.32	0.32	[3]	1.30	HD 179949	6235	4.41	1.38	0.21	[1] $^{\dag\dag }$	1.25
HD 27442	4890	3.89	1.24	0.42	[2]	0.83	HD 186427	5765	4.46	1.03	0.09	[4]	0.99
HD 30177	5590	4.45	1.07	0.39	[1]	1.00	HD 187123	5855	4.48	1.10	0.14	[4]	1.05
HD 33636	5990	4.68	1.22	-0.05	[2]	1.12	HD 190228	5360	4.02	1.12	-0.24	[3]$^\dag$	0.84
HD 37124	5565	4.62	0.90	-0.37	[3]	0.76		5325	3.95	1.10	-0.23	[4]	0.82
HD 39091	5995	4.48	1.30	0.09	[1]$\dag$	1.10	(average)	5340	3.99	1.11	-0.24		0.83
HD 46375	5315	4.54	1.11	0.21	[3]	0.83	HD 190360	5590	4.48	1.06	0.25	[3]	0.96
HD 50554	6050	4.59	1.19	0.02	[3]	1.11	HD 192263$^{\star}$	4995	4.76	0.90	0.04	[2]	0.75
HD 74156	6105	4.40	1.36	0.15	[2]	1.27	HD 195019	5845	4.39	1.23	0.08	[4]	1.06
HD 75732A	5307	4.58	1.06	0.35	[3]	0.88		5832	4.34	1.24	0.09	[1] $^{\dag\dag }$	1.05
HD 80606	5570	4.56	1.11	0.34	[3]	1.03	(average)	5840	4.36	1.24	0.08		1.06
HD 82943	6025	4.54	1.10	0.33	[1]$^\dag$	1.15	HD 196050	5905	4.41	1.40	0.21	[1]	1.10
	6025	4.53	1.15	0.30	[5]	1.15	HD 209458	6120	4.56	1.37	0.02	[5]	1.15
(average)	6025	4.54	1.12	0.32		1.15	HD 210277	5575	4.44	1.12	0.23	[2]$^\dag$	0.94
HD 92788	5820	4.60	1.12	0.34	[1]	1.10		5560	4.46	1.03	0.21	[4]	0.93
HD 95128	5925	4.45	1.24	0.05	[4]	1.05	(average)	5570	4.45	1.08	0.22		0.94
HD 106252	5890	4.40	1.06	-0.01	[1] $^{\dag\dag }$	1.02	HD 213240	5975	4.32	1.30	0.16	[1]	1.22
HD 108874	5615	4.58	0.93	0.25	[3]	0.96	HD 216435	5905	4.16	1.26	0.22	[1]	1.26
HD 114386	4875	4.69	0.63	0.00	[1]	0.68	HD 216437	5875	4.38	1.30	0.25	[1]	1.06
HD 114729	5820	4.20	1.03	-0.26	[3]	0.94	HD 217014	5805	4.51	1.22	0.21	[2]	1.04
HD 114762	5870	4.25	1.28	-0.72	[5]	0.80	HD 222582	5850	4.58	1.06	0.06	[3]	1.02
HD 114783	5160	4.75	0.79	0.16	[4]	0.88

^a The instruments are [1] CORALIE, [2] FEROS, [3] UES, [4] SARG, [5] UVES, and [6] ELODIE.
$\dag$ Already published in Santos et al. (2001a) (Paper II).
$\dag\dag$ Already published in Santos et al. (2001b).
$\dag\dag\dag$ From the isochrones of Schaller et al. (1992), Schaerer et al. (1993) and Schaerer et al. (1992).
$\star$ The existence of a planet around HD192263 was recently put in cause by Henry et al. (2002); we think, however,
that these authors have not shown enough evidences against the presence of a planet, and thus we prefer to keep
this star in the planet-hosts sample (Santos et al., in preparation).

Data reduction was done using IRAF tools in the echelle package. Standard background correction, flat-field, and extraction procedures were used. In all the cases, the wavelength calibration was done using a ThAr lamp spectrum taken during the same night.

We have compared the Equivalent Widths (EW) for some stars for which we have obtained spectra using different instruments to check for possible systematics. In all cases, the average difference of the EWs is within 1-2 mÅ, and usually lower than 1 mÅ. As can be also verified from Table 2 of this article and Table 2 from Paper II, these possible small systematics do not seem to affect significantly the analysis of the atmospheric parameters and metallicity. The only star that has large variations in the derived atmospheric parameters is HD 19994. This variation might be connected with the fact that this late F dwarf has a high rotational velocity $v~\sin{i}=$ 8.1 km s^-1 (from the calibration of the CORALIE Cross-Correlation Function presented in Santos et al. 2002a), a sign of relative youth and (most probably) activity related phenomena.

2.2 Stellar parameters and chemical analysis

In this paper we use the same technique, line-lists, and model atmospheres as in Papers I and II. The abundance analysis was done in standard Local Thermodynamic Equilibrium (LTE) using a revised version of the code MOOG (Sneden 1973), and a grid of Kurucz (1993) ATLAS9 atmospheres.

The atmospheric parameters were obtained from the Fe I and Fe II lines by iterating until the correlation coefficients between $\log{\epsilon}$(Fe I) and $\chi_l$, and between $\log{\epsilon}$(Fe I) and $\log{({W}_\lambda/\lambda)}$ were zero, and the mean abundance given by Fe I and Fe II lines were the same. This procedure gives very good results since the set of Fe I lines has a very wide range of excitation potentials.

Figure 1: Left: metallicity distribution for stars with planets (hashed histogram) compared with the same distribution for the field dwarfs presented in Paper II (empty histogram). The vertical lines represent stars with brown dwarf candidate companions. Right: the cumulative functions of both samples. A Kolmogorov-Smirnov test shows the probability for the two populations being part of the same sample is around 10-7.

The results of our analysis are presented in Table 2. The number of measured Fe I and Fe II lines is always between 24 and 39, and 4 and 8, respectively. The rms around the mean individual abundances given by the lines has values between 0.03 and 0.07 dex in most cases. The errors in $T_{{\rm eff}}$, $\log{g}$, $\xi _t$ and [Fe/H] were computed as in Gonzalez & Vanture (1998). For a typical measure the uncertainties are usually lower than 50 K, 0.15 dex, 0.10 km s^-1, and 0.06 dex, respectively (see Paper II). The only important exceptions are the cases of HD 20367 (for which the lower quality ELODIE spectra with S/N $\sim$ 80-100 were responsible for errors of the order of 100 K, 0.20 dex, 0.15 km s^-1, and 0.10 dex in $T_{{\rm eff}}$, $\log{g}$, $\xi _t$ and [Fe/H], respectively) and for HD 137759, a giant star for which the dispersion in the [Fe/H] values for individual lines was quite high. The masses were then determined from the theoretical isochrones of Schaller et al. (1992), Schaerer et al. (1993) and Schaerer et al. (1992), using M_V computed from Hipparcos parallaxes (ESA 1997) and $T_{{\rm eff}}$ obtained from spectroscopy. We adopt a typical error of 0.1 $M_{\odot}$ for the masses.