Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.sao.ru/Doc-en/Science/Public/Conf/magstars-2006/p218.pdf Дата изменения: Fri Dec 26 12:11:16 2008 Дата индексирования: Mon Feb 4 15:55:37 2013 Кодировка:

Physics of Magnetic Stars, 2007, pp. 218­222

X-ray emission and the incidence of magnetic fields in the massive stars of the Orion Nebula Cluster
V. Petit
1

1,2

, G.A. Wade2 , T. Montmerle3 , L. Drissen1 , N. Grosso3 , F. Menard

3

Dґ epartement de Physique, de Gґ enie Physique et d'Optique and Observatoire du mont Mґ egantic, Universitґ e Laval, Quґ ec, QC G1K 7P4, Canada eb 2 Department of Physics, Royal Military College of Canada, PO Box 17000, Station `Forces', Kingston, Ontario, K7K 4B4, Canada 3 Laboratoire d'Astrophysique de Grenoble, Universitґ Joseph-Fourier, F-38041 Grenoble, France e

Abstract. Magnetic fields have b een frequently invoked as a likely source of variability and confinement of the winds of massive stars. To date, the only magnetic field detected in O-typ e stars are those of 1 Ori C (HD 37022; Donati et al. 2002), the brightest and most massive memb er of the Orion Nebula Cluster (ONC), and HD 191612 (Donati et al. 2006). Notably, 1 Ori C is an intense X-ray emitter, and the source of these X-rays is thought to b e strong sho cks o ccurring in its magnetically-confined wind (Bab el & Montmerle 1997a, Donati et al. 2002). Recently, Stelzer et al. (2005) have found significant X-ray emission from all massive stars in the ONC. Perio dic rotational mo dulation in X-rays and other indicators suggested that 1 Ori C may b e but one of many magnetic B- and Otyp e stars in this star-forming region. In 2005B we carried out sensitive ESPaDOnS observations to search for direct evidence of such fields, detecting unambiguous Zeeman signatures in two ob jects. Key words: stars: early-typ e - stars: magnetic fields - stars: mass-loss

1

Intro duction

The existence of magnetic fields in early-typ e stars remains a mystery. The intermediate-mass magnetic Ap-Bp stars, with observed kG dip olar fields, are well known, but b ecause they have radiative envelop es, they cannot host the dynamo-generated surface magnetic fields found in latetyp e convective stars like the Sun. These fields are b elieved to b e fossil remnants of either interstellar magnetic fields swept up during the star formation pro cess or fields pro duced by a pre-main sequence envelop e dynamo that has since turned off. In more massive stars, magnetic fields have only b een discovered recently, mostly via clues provided by unusual X-ray prop erties. Traditionally, the X-ray emission from O and B stars, with a typical level LX /Lbol 10-7 , has b een explained by radiative instabilities, via a multitude of sho cks in the wind (Lucy & White 1980, Owo cki & Cohen 1999). However, the very strong and rotationally mo dulated X-ray emission of the brightest Trap ezium star, 1 Ori C (O7, P=15.4 d), was explained by Bab el & Montmerle (1997a, 1997b) in terms of the "magnetically confined wind sho ck" mo del (MWCS). In this mo del, the stellar magnetic field is sufficiently strong, and the radiative wind sufficiently weak, to allow a dip olar magnetic field to confine the outflowing wind 218


X-RAY EMISSION AND THE INCIDENCE OF MAGNETIC FIELDS IN...

219

in the immediate circumstellar environment, resulting in a closed magnetosphere with a large-scale equatorial sho ck which heats the wind plasma. In this way, the X-ray emission is enhanced and may b e mo dulated by stellar rotation. Bab el and Montmerle thereby predicted the existence of a dip olar magnetic field in 1 Ori C; such a field was subsequently discovered by Donati et al. (2002). One can therefore conclude that an unusually high, and/or rotationally mo dulated X-ray emission, is a strong indication of the presence of closed magnetospheres in massive stars. An exciting application of these results has b een provided by the "Chandra Orion Ultradeep Pro ject" (COUP). This program consisted of a 9.7 day almost continuous observation with the ACIS detector (FOV 17 в 17 ), centered on the Orion Nebula Cluster (ONC). In all, 1616 sources were detected, characterized and identified (Getman et al. 2005). The COUP study of the OBA p opulation (20 stars) by Stelzer et al. (2005) was aimed at disentangling the resp ective roles of wind and magnetic fields in X-ray emission of these stars. X-ray emission by standard radiative sho cks is found to b e the dominant mechanism for the subsample of 9 high-mass O to early-B stars which have strong winds. However, 3 of the high-mass stars showed X-ray prop erties suggestive of the presence of magnetic fields: 1 Ori C (O7; discussed ab ove), 1 Ori A (B0), and JW660 (B3). On the other hand, a fourth star, Par 1772 (B2), a tentative helium-strong star, showed only a very weak X-ray emission. We have undertaken a study with ESPaDOnS to explore the role of magnetic fields in pro ducing this diversity of X-ray b ehaviours.

2

Observations

In January 2006 we conducted circular p olarisation observations with ESPaDOnS at CFHT to search for direct evidence of magnetic fields in the ONC massive stars. We obtained single high S/N Stokes V sp ectra of 8 of the massive OB stars. The mean Stokes I and V profiles were extracted with the Least Square Deconvolution technique (LSD) of Donati et al. (1997), which allows the use of many lines to increase the level of detection of a magnetic field Stokes V signature. The LSD line masks were carefully chosen, in order to maximize the resultant LSD S/N. Several exp eriments were attempted, including the removal of contaminated lines and helium lines, along with comparison with synthetic sp ectra. We detected clear magnetic (Stokes V ) signatures in sp ectra of 2 stars: 1 Ori C, which was already known to b e magnetic, and Par 1772, a new detection (Fig. 1). For the remaining 6 stars, we observed no Stokes V detection. Table 1 gives a summary of the target prop erties, along with the longitudinal field error bar that we achieved. Table 1: Target information and longitudinal field error bars achieved V 5.1 5.1 6.7 6.7 6.8 6.0 8.4 9.7 Typ e O7 O9.5 B0 B0.5 B1 B1 B2 B3 vsini (km/s) 50 110 115 50 160 25 120 210 Obs. Time 0h53 1h20 2h40 2h40 2h40 0h40 2h40 4h25 (G) 37 346 225 30 116 12 72 460

B

1 Ori C 2 Ori A 1 Ori A 1 Ori D NU Ori 2 Ori B Par 1772 JW 660


220

PETIT ET AL.

Figure 1: Least Square Deconvolved Stokes I and V profiles. The thin lines show the mean Stokes I profile (b ottom), the mean Stokes V profile (top) and the N diagnostic null profile (middle). The thick line is the b est dip ole mo del fit from our magnetic analysis.

3

Magnetic analysis

To constrain admissible surface magnetic field configurations of these stars, we have compared the observed Stokes V sp ectra with synthetic profiles computed assuming a large grid of dip olar magnetic field mo dels. Those mo dels are characterized by different values of the magnetic tilt angle and dip olar intensity Bd . By computing the reduced 2 of the mo del with resp ect to the observations, we have constructed confidence maps of admissible (and rejectable) field configurations for each star. One such map is illustrated in Fig. 2 for Par 1772. Our single observation of this star constrains the surface magnetic field to b e ab ove 400 G with 99% confidence. However, we have weak constraint on the dip ole tilt and on the maximum field strength. This b ehaviour results from the limited information afforded by a single observation. Analogous maps have b een obtained for all observed stars. For those in which no Stokes V detection is obtained, our maps allow us in principle to determine upp er limits on admissible field configurations. This is crucial: a magnetic field of sp ecific minimum strength is required to confine a wind of known mass flux. Hence if we can determine an upp er limit on admissible magnetic fields of the undetected stars, we can p otentially rule out magnetically-confined wind sho cks as the source of their X-ray emission and variability. In addition, these upp er limits provide unique data with which to confront mo dels of magnetic field origin in neutron stars and magnetars, such as that prop osed by Ferrario & Wickramasinghe (2006). Unfortunately, our confidence maps of the undetected stars suffer from the same characteristics as the map of Par 1772: the maximum admissible field strength (upp er limit) is p o orly constrained. Because when = 90-i, geometries exist for which there is no Stokes V signature, notwithstanding the field strength. Fortunately, our numerical exp eriments show that by obtaining at least one additional Stokes V observation of each star, we can constrain significantly the orientation of the dip ole, leading to a dramatic reduction in the range of probable field strengths by a full order of magnitude. Nevertheless, we can still give a rough estimate of the


X-RAY EMISSION AND THE INCIDENCE OF MAGNETIC FIELDS IN...

221

Table 2: First estimates of the field strengths and upp er limits for the ONC massive stars. Calculated from 1 contours at = 90 Bd (G) 1100±100 (3 ) <600 <300 <150 TBD <150 800-2500 <4000

1 Ori C 2 Ori A 1 Ori A 1 Ori D NU Ori 2 Ori B Par 1772 JW 660

magnetic field, by taking reasonable values for and i (ie, by ignoring the degenerate configuration). Those first estimates are given in Table 2. A more robust analysis is currently underway, along with the planning of multi-phase sp ectrop olarimetric observations.

4

Conclusion

Using Stokes V mesurements obtained with ESPaDOnS sp ectrop olarimeter at CFHT, we were able to detect the presence of a new magnetic star in the Orion Nebula Cluster. Par 1772 has a dip ole field comp onent greater than 400 G (at 99% confidence). However, the precise magnetic configuration cannot b e fully constrained with a single observation; neither can the maximum field strengh in the non-detection cases. However, a more detail analysis, combined with multi-phase observations will enable more precise estimates. Once completed, this study of the Orion stellar cluster will represents a complete magnetic survey of a co-evolved and co-environmental p opulation of massive stars. A imp ortant issue of this study regards magnetic fields in even more massive stars. It is related to neutron star p opulation studies. Ferrario & Wickramasinghe (2006) explored the hyp othesis that neutron star magnetic fields are of fossil origin, and predicted a field distribution for the OB star progenitors (ab ove 8 solar masses). According to their mo del, 7% of the main sequence OB stars should have a magnetic field in excess of 1kG. Interestingly, we find 2 stars (25%) with fields ab ove 1 kG in the ONC. As the fossil hyp othesis dep ends on the environment (in the case of a interstellar magnetic fields swept up), it could b e p ossible that the Orion Nebular Cluster is overly magnetized. Similar studies of other young OB clusters would therefore b e really interesting. On the other hand, if the magnetic field origin is intrinsic to the star itself (envelop e dynamo remnant coming from a pre-main sequence phase for exemple) the fraction of magnetized stars should b e similar in all OB clusters.

References
Lucy L.B. & White R.L., 1980, ApJ, 241, 300 Owocki S.P. & Cohen D.H., 1999, ApJ, 520, 833 Babel J., & Montmerle T., 1997a, A&A, 323, 121 Babel J., & Montmerle T., 1997b, ApJ, 485, L29 Donati J.-F., Babel J., Harries T.J., et al., 2002, MNRAS, 333, 55 Getman K.V., Feigelson E.D., Grosso N., et al., 2005, ApJS, 160, 353 Stelzer B., Flaccomio E., Montmerle T., et al., 2005, ApJS, 160, 557


222

PETIT ET AL.

Figure 2: Confidence map for admissible dip ole field strength B p and obliquity . There is 99% probability that the real surface field strength and obliquity are within the 99% contour. From this map, constructed from a single observation, we can say that the field is stronger than 400 G (at 99% confidence), but more observations are required to significantly constrain the dip ole inclination and the field strength upp er limit.
Ferrario L. & Wickramasinghe D.T., 2006, MNRAS, 367, 1323 Donati J.-F., Semel M., Carter B.D., et al., 1997, MNRAS, 291, 658 Donati J.-F., Howarth I. D., Bouret J.-C., et al., 2006, MNRAS, 365, L6