Stellar Xray sources as diagnostic of the local star formation history

1
Stellar X-ray sources as diagnostic of the local
star formation history
KLUTSCH Alexis (Strasbourg, France)?

• PhD Thesis Directors
• P. Guillout, R. Freire Ferrero (Strasbourg)?
• Collaborators
• A. Frasca, E. Marilli, K. Biazzo (Post-Doc) and
  G. Mignemi (PhD Student) (Catane, Italy)?

2
Outline

• Introduction (Stellar Activity and Kinematic
  Groups)?
• RasTyc sample (Major Aims, selection and
  observations)?
• Astrophysical parameters determination
• Results
• Conclusions Perspectives

3
Stellar Activity
Guillout et al., 1999

• Einstein mission has discovered the universality
  of stellar X-ray emission throughout the HR
  diagram (Vaiana et al., 1981, Pallavicini et al.,
  1981),
• confirmed by ROSAT survey,
• Stellar X-ray luminosity can exceed mean solar
  luminosity by 3 orders of magnitude,

• Most of stars detected by ROSAT are younger than
  1 Gyr (Motch et al., 1997),

Possibility to extract in a novel way large
samples of young stars

• The distinct group of objects within the Giant
  branch known as the clump clearly appears.

4
Kinematic groups

• Some X-ray and EUV active and lithium-rich stars
  are members of Stellar kinematic groups (SKGs)
  see, e.g., Jeffries (1995)
• SKG a kinematically coherent group of stars
  that could share a common origin (e.g. the
  evaporation of an open cluster),
• Eggen (1994)
• supercluster (SC) group of stars
  gravitationally unbound that share the same
  kinematics and may occupy extended regions in the
  Galaxy,
• moving group (MG) part of the SC that enters
  the solar neighborhood and can be observed all
  over the sky,
• Well-known members of these SKGs are mainly
  early-type stars and few studies have been
  centered on late-type stars,
• Montes et al. (2001) Identification of a
  significant number of late-type members of these
  young SKGs would be extremely important to
• study their chromospheric and coronal activity
  and their age evolution.
• understand the local star formation history in
  the solar neighborhood.

5
Major Aims of RasTyc Sample
XMM Predictions (contamination of RasTyc sample
by evolved stars)?
Follow-up of stars showing interesting
characteristics
Local galactic structure evolution (SFR and scale
height of young stars)?
Correlation between chromospheric and coronal
activity
Dynamo Theory (constraints on some parameters)?
6
Identification of stellar X-ray sources
In small size regions Motch et al. (1997) et
Zickgraf et al. (1997).
Correlation between catalogues The RasTyc
sample is the result of the cross-correlation of
the ROSAT All-Sky Survey (RASS) with the Tycho
catalogue (Guillout et al. 1999) and represents
the largest (13875 active stars) and most
comprehensive sample of stellar X-ray sources
constructed so far.
Selection we have started a spectroscopic
observations campaign aimed at a deep
characterization of a representative sub-sample
of the RasTyc population in the northern
hemisphere.
7
T1.5m Observations T1.9 m and T3.6 m
Observations Stars with measurements from the
literature
Sub-sample of RasTyc stars in the northern
hemisphere 1097 sources (VT ? 10.5
mag). Observations of about 800 of these
sources. Characterization of the optically bright
sample (704 sources with VT ? 9.5 mag) Guillout
et al., to be submitted.
Guillout et al., to be submitted
T1.5m Observations T1.9 m and T3.6 m
Observations Stars with measurements from the
literature
Guillout et al., to be submitted
Guillout et al., to be submitted
8
Observations

• High resolution spectroscopy (main program)?
• OHP1 T1.5 m telescope (R 40000)?
• 154 nights in 2001 - 2005 period.
• 1250 spectra ( 500 RasTyc stars).
• H? region (6490 6630 Å)
• Li region (6650 6780 Å)?
• OHP T1.9 m telescope (R 42000)
• 110 stars observed in 2000 - 2001.
• Echelle spectra (3900 6800 Å)?
• TNG2 T3.6 m telescope (R 46000)?
• 182 stars observed in 2007 (in progress).
• Echelle spectra (4620 7920 Å)?
• Medium resolution spectroscopy (follow-up)?
• OAC3 T0.9 m telescope (R 22000)?
• Echelle spectra (4300 - 6800 Å)?
• Photometric data (follow-up)?
• OAC3 T0.9 m telescope

1 Observatoire de Haute Provence 2 Telescopio
Nazionale Galileo 3 Osservatorio Astrofisico di
Catania
800 sources already observed
35 binaries monitored
9
Determination of RV and Vsini
RV

• Automatic IDL procedure dedicated to Aurélie
  spectra
• Cross-correlation method (based on synthetic
  spectra)?
• Search significant local minima of the
  cross-correlation function (CCF)
• Fit these peaks with multiple gaussian functions
  or multiple rotation profile functions according
  to multiplicity
• Analyze residuals to find peaks with a amplitude
  greater than 3s
• Determination of radial velocity (RV) and
  projected rotational velocity (vsini) for single
  and components of multiple systems

10
RV measurements accuracy

• Using RV standard stars.
• Comparison of our RVs measurements with those
  quoted in the literature.
• Dotted lines delimit a 1.5 km s-1 region
  around the one-to-one relation (continuous red
  line).
• Standard RV error 1.5 km s-1.

Guillout et al., to be submitted
11
Detection of SB1 systems

• Each star was observed at least 2 times.
• Take advantage of RV inconsistencies from the H?
  and Li spectra to gather spectroscopically
  unresolved binaries
• ?RV RVH?-RVLi
• ?RV gt ?RVH??RVLi
• a real Doppler shift.

Stars classified as SB1 systems
Guillout et al., to be submitted
12
Astrophysical Parameters
Single stars or SB1 systems

• IDL code developed by us (Frasca et al., 2006),
  ROTFIT.
• Determination simultaneously of Teff, log(g),
  Fe/H and vsini by the best fit
  (?2-minimization) of the observed spectrum with a
  grid of stellar spectra retrieved from the Élodie
  Archive (Prugniel Soubiran, 2001).

Klutsch et al. (2007), Poster of "14th Cambridge
Workshop on Cool Satrs, Stellar Systems, and the
Sun "
13
SB2 systems

• IDL code developed by us (Frasca et al. 2006),
  COMPO2.
• Determination of the best combination of two
  standard-stars spectra able to reproduce the
  observed spectrum of the SB2 system.
• Input parameters radial velocity and vsini of
  the two components.
• Output parameters spectral types and fractional
  flux contributions that reproduce the observed
  spectrum better, i.e. which minimize the
  residuals in the collection of difference
  (observed - synthetic) spectra.

14
H? Luminosity
15
Lithium abundance
Determination of EWLi According to Pavlenko
Magazzù (1996) NLTE calculations, we derived the
Li abundance
16
Summary files
Physical parameters are presented on a summary
file created automatically for each star and
stored in a major database.
Guillout et al., to be submitted
17
Searching for Moving Group members (1)?
Method 2D
Membership criteria Pmember ? 10 and ?MG ?
210
Moving Groug (MG) i stars ( )? MG barycenter
( )? MG candidate ( )? Determination of
VS and ? for each MGs star where VS
((Ui-U)2 (Vi-V)2 (Wi-W)2)1/2 Vb
((UB-U)2 (VB-V)2 (WB-W)2)1/2 ? angle
between Vb and VS ?MG max(?) min(?)?
Possible membership Pmember ? 10 and 210 ?
?MG ? 170
XE
YE
VS sin(?)?
YE
?
VS cos(?)?
VS
XE
Klutsch et al., in preparation
18
Searching for Moving Group members (2)?
Membership's probability Pmember - 2?(k)
(2/?)1/2 k e -0.5 k2 where k2 U 2 V
2 W 2 and ?(k) is the area under the normal
curve from 0 to k.
Method 3D
Fit each histogram in U, V and W of all MG's
members with a gaussian. Determination of sU, sV
and sW.
Membership criteria Pmember ? 10
U
W
V
V (km s-1)?
W (km s-1)?
U (km s-1)?
Klutsch et al., in preparation
19
Results

• Multiple systems
• Statistics
• Binaries (SB1 SB2)?
• Spectroscopic triple systems (SB3)
• Evolved stars
• Very young stars
• Different group
• Kinematics and memberships
• Post TTauris Candidates Stars

20
Multiples Systems including very active binaries
(RS CVn et BY Dra)? 36
Klutsch et al., to be submitted
21
Binaries (SB1 SB2)
PhD thesis of G. MIGNEMI
Giants
(Frasca et al., AA, 2006)?
Some indications for a significant fraction of
evolved stars (RS CVn systems ? ) in our sample
XMM-SSC Meeting - Strasbourg 18/10/2007
22
Spectroscopic triple systems (SB3)?
Klutsch et al., to be submitted
23
(No Transcript)
24
Guillout et al., to be submitted
Guillout et al. 1999
Evolved Stars 18
If we restrict to K stars only, this fraction
increases to 52 .
MV
(B-V)j
25
Very young stars 48
If we consider stars younger than 1 Gyr, this
fraction increases to 70 .
26
PMS-Like Stars
2D membership
3D membership
27
PMS-Like Stars
2D membership
3D membership
28
Pleiades-Like Stars

• Young Disk (YD) 10(2) stars
• IC2391 1 gt 8-10
• Castor 2(1) gt 17-20
• Pleiades 6 gt 50-60
• UMa 0(1) gt 0-8
• Hyades 1 gt 8-10
• Old Disk (OD) 7 stars ( 37 )?

29
Post TTauris candidates stars

• For 2 PMS-like stars we have good Hipparcos
  parallaxes
• Pmember 0 ? Giant
• Pmember gt 90 ? PMS (1M035 Myrs)?

MV
MV
(B-V)j
(B-V)j
30
Some open questions
MV
MV
(B-V)j
(B-V)j
31
Conclusions Perspectives

• Importance fraction of binaries and multiple
  systems
• understand the formation of close binaries
• Study the recent local star formation history.
• Contamination by active giants (RS CVn).
• Very young stars (70 of stars younger than 1
  Gyr)
• New members of Moving Groups.
• Post TTauris candidates stars (exoplanets
  candidates).

• The investigation of chromospheric and coronal
  activity evolution with age and its dependency on
  stellar parameters such as stellar rotation and
  mass.
• Both X-ray and H? data for a large stellar sample
  can contribute to constrain conveniently
  theoretical dynamo parameters.
• Besançon stellar X-ray population model versus
  observations better understand the local
  galactic structure evolution (SFR and scale
  height of young stars).
• During my stay at Madrid, I am working with Dr.
  David MONTES to know correctly the MGs members,
  to apply the Eggens criteria on the RasTyc
  sources, in particular on the PMS-like and
  Pleiades-like stars, to determine their
  membership with a already known MG and to find
  possibly new young MGs.

32
Besançon stellar X-ray population model
Although the star formation history in the Galaxy
is rather well constrained with optical star
counts, the last gigayear local star formation
rate is still poorly known. However young stars
represent the overwhelming population at low
galactic latitudes in soft X-ray surveys which
are heavily biased towards the most actives
stars. This bias offers the opportunity to
overcome difficulties encountered at optical
wavelength as illustrated in the figure at right
and in the table below.
Relative number of Pleiades-like, Hyades-like and
Old stars at the RASS sensitivity for the star
formation rates presented in the right figure.
Log(N)-Log(S) curves computed in the direction of
the galactic plane (l 90, b 0) for
populations of disc stars. N is the number of
stars per square degree in the direction l,b,
detected above a given PSPC count rate S (in the
0.1-2.4 keV band) as a function of S. We have
tested stellar formation rates decreasing by a
factor 2 during the last gigayears (D07 left
panel) and slightly increasing by 30 (I07
right panel) with respect to the standard
constant rate presented in the middle panel
(C07). These curves clearly show the impact of
the recent local star formation rate on the
fraction of Pleiades-like (red line), Hyades-like
(green line) and old (blue line) stars (see also
expected fraction of populations in the left
table). In both figures, dots symbols show the
observed stellar density of the Rass in a low
latitude area covering 64.5 square degrees
studied by Motch et al. 1997.
33

• Thank you for your attention.

Gracias por su atención.