EDP Sciences logo
Subscriber Authentication Point
Search
Menu
Home All issues Volume 596 (December 2016) A&A, 596 (2016) A44 Full HTML
Free Access
Issue
A&A
Volume 596, December 2016
Article Number A44
Number of page(s) 7
Section Stellar atmospheres
DOI https://doi.org/10.1051/0004-6361/201629320
Published online 28 November 2016

© ESO, 2016

1. Introduction

Strong magnetic fields have been detected in a dozen O stars during the last decade (Petit et al. 2013; Fossati et al. 2015, and references therein). In such objects, the stellar wind flows are channelled towards the equator (Babel & Montmerle 1997a), creating a dense region of confined winds. Part of this material is shock-heated to higher temperatures, leading to the emission of X-rays, while the cooler plasma is easily detected in the visible (e.g. through Balmer emission lines). When the rotation axis is not perfectly aligned with the magnetic axis, a periodic modulation of the electromagnetic emission associated with these confined winds is observed. Variations of the longitudinal field, of the broad-band photometry, of the visible emission lines, and of the X-ray flux thus occur simultaneously, as was found for example for θ1 Ori C (Donati et al. 2002; Gagné et al. 2005) or HD 191612 (Donati et al. 2006; Nazé et al. 2007, 2010; Howarth et al. 2007).

Table 1

List of the new X-ray observations.

Spectropolarimetric observations found Tr16-22 (O8.5V) to be strongly magnetic (Nazé et al. 2012, 2014a). In the X-ray range, Tr16-22 displays a bright, variable, and hard emission atypical of single, “normal” massive stars (Evans et al. 2004; Antokhin et al. 2008; Nazé et al. 2011, 2014a), but in line with theoretical expectations for confined winds (Nazé et al. 2014b). The Fourier analysis of the X-ray data clearly indicated the presence of a periodicity; the favoured value was ~54 d but many aliases were present, leaving some ambiguity on the actual period (Nazé et al. 2014a). Therefore, to better constrain its physical properties, we needed additional data of Tr16-22.

In this paper, we analyse these new data. Section 2 presents the observations used in the study, Sect. 3 provides the results, and Sect. 4 reports our conclusions.

2. Observations

2.1. X-ray observations

Being close to η Carinae, Tr16-22 is frequently observed at high energies. Since the observations presented in Nazé et al. (2014a), three new XMM-Newton exposures are available in the archives and an additional one was kindly provided by Dr Kenji Hamaguchi (Table 1). In these observations, the target only appears on the MOS2 camera: it falls outside the field of view of the other two cameras (because of dead CCDs in MOS1 or the use of the small window mode in MOS1 or pn). These new data were reduced with SAS v14.0.0 using calibration files available in December 2015 and following the recommendations of the XMM-Newton team1. Only the best quality data (PATTERN of 0–12) were kept and background flares were discarded. A source detection was performed on each EPIC dataset using the task edetect_chain on the 0.4–10.0 keV energy band and for a likelihood of 10 to find the best-fit position of the target in each dataset. We note that the source is not bright enough to present pile-up. EPIC spectra were extracted using the task especget for circular regions of 35′′ radius centred on these best-fit positions and, for the backgrounds, at positions as close as possible to the target considering crowding and CCD edges. The spectra were finally grouped, using specgroup, to obtain an oversampling factor of five and to ensure that a minimum signal-to-noise ratio of three (i.e. a minimum of 10 counts) was reached in each spectral bin of the background-corrected spectra.

The X-ray spectra were then fitted within Xspec v12.7.0 using the same models as in Nazé et al. (2014a), i.e. absorbed optically thin thermal plasma models, i.e. wabs × phabs × ∑ apec, with solar abundances (Anders & Grevesse 1989). The first absorption component is the interstellar column, fixed to 4.4 × 1021 cm-2 (a value calculated using the colour excess of the star and the conversion formula 5.8 × 1021 × E(B−V) cm-2 from Bohlin et al. 1978), while the second absorption represents additional (local) absorption. For the emission components, we considered either two or four temperatures, as in Nazé et al. (2014a). Table 2 yields the results of our X-ray fittings. Since the 2012 December exposure is long, we also extracted the X-ray lightcurve of Tr16-22 using the same regions as for spectrum extraction. Dedicated χ2 tests were then performed and did not reveal any significant variability of Tr16-22 during this observation.

Table 2

Results from the X-ray spectral fitting.

2.2. Spectropolarimetry

Spectropolarimetric data of Tr16-22 had been obtained with the Very Large Telescope equipped with FORS2 in 2011 (Nazé et al. 2012) and 2013 (Nazé et al. 2014a). In view of the variations detected in X-rays, we requested a monitoring of the star (Prog. 095.D-0082, PI Nazé): seven additional observations were taken in service mode between April and June 2015 (Table 3). These new data were taken with the red CCD (a mosaic composed of two 2k×4k MIT chips) without binning, with a slit of 1′′ and the 1200B grating (R ~ 1400). The observing sequence consisted of 8 subexposures of 240 s duration (except for one night, 2015 June 06, where it was pushed up to 500 s) with retarder waveplate positions of + 45°, − 45°, − 45°, + 45°, + 45°, − 45°, − 45°, + 45°. We reduced these spectropolarimetric data with IRAF2 as explained by Nazé et al. (2012): aperture extraction radius fixed to 20 px, subtraction of nearby sky background, and wavelength calibration from 3675 to 5128 Å (with pixels of 0.25 Å) considering arc lamp data taken at only one retarder waveplate position (in our case − 45°). This allowed us to construct the normalized Stokes V/I profile, as well as a diagnostic “null” profile (Donati et al. 1997; Bagnulo et al. 2009). We note that the signal-to-noise ratio of the derived I spectra near 5000 Å was about 1100–1200. Finally, the associated longitudinal magnetic field was estimated by minimizing with yi either V/I or the null profile at the wavelength λi and (Bagnulo et al. 2002). This was done for xi in the interval between − 10-6 to + 10-6 (an interval where the vast majority of good points are available, avoiding potential problems in slope determination due to a few isolated data points at extreme xi, although we note that enlarging the interval, e.g. ± 1.6 × 10-6, does not significantly change the reported values) after discarding edges and deviant points, after rectifying the Stokes profiles, and after selecting spectral windows centred on lines (see Nazé et al. 2012, for further discussion). Table 3 yields the resulting field values, along with the values for the old datasets (recalculated for the same spectral windows).

Table 3

Results from the spectropolarimetry (using rectification and within the same spectral windows).

2.3. High-resolution spectroscopy

Data used in this work correspond to spectra collected in three different observatories over a time span of eighteen years. Ten spectra covering 3600–6100 Å with a resolving power of 15 000 were obtained with the REOSC échelle spectrograph3 attached to the 2.15 m Jorge Sahade telescope at Complejo Astronómico El Leoncito (CASLEO, Argentina), during 1997, 1998, 2011, and 2015. Thirteen spectra covering 3450–9850 Å with a resolving power of 40000 were gathered with the échelle spectrograph attached to the 2.5 m du Pont telescope at Las Campanas Observatory (LCO, Chile), in 2010, 2013, 2015, and 2016. For these CASLEO and LCO spectra, calibration lamp exposures were secured immediately before or after each star integration at the same sky position. Data were reduced and processed in a standard way using the usual IRAF routines. In addition, two spectra of Tr16-22 were obtained with the FEROS spectrograph attached to the 2.2 m telescope at ESO La Silla Observatory (Chile) in 2011 (086.D-0997B) and 2015 (089.D-0975A). These spectra, covering 3570–9210 Å with a resolving power of 46 000, were gathered following the standard ESO procedures, and they were reduced using the FEROS pipeline provided for MIDAS4. We note that spectra obtained after 2010 were taken in the framework of the program “OWN Survey”, a spectroscopic monitoring of southern Galactic O- and WN-type stars (Barbá et al. 2010), while the CASLEO spectra obtained in the nineties are part of the XMEGA project (Corcoran et al. 1999; Albacete Colombo et al. 2001).

3. Results and discussion

In the X-ray range, the new data are in line with previous observations. First, the new data appear amongst the old ones in the flux-hardness relation (see Fig. 1). Second, the periodogram does not change much if the new data are included: ~54 d is still the favoured period and folding with this timescale yields the same, coherent behaviour as shown before (see right panel of Fig. 1, and Nazé et al. 2014a). However, ambiguities on the period remain because aliases are still numerous (typically spaced by 4 × 10-4 d-1). As mentioned in Nazé et al. (2014a), the datasets recorded in 2003 provide a hint that the variation timescale may not be very long as the observed flux increased that year by about 30% over two weeks at the end of July and beginning of August, then went back to its level in mid-August. Nevertheless, no full cycle was observed in a single observing run, and only observations with a higher cadence might be able to remove the ambiguity on the periodicity.

thumbnail Fig. 1

Left and middle: evolution of the hardness ratios (HR = H/S = FISMcor(2.−10.0 keV) /FISMcor(0.5−2. keV)) as a function of observed fluxes. Red triangles correspond to the 2003 XMM-Newton data (the year with the largest number of observations), green crosses to the new data, and black open circles to the other XMM-Newton observations. The values shown in the left (resp. middle) panel correspond to the results from 4T fits (resp. 2T fits). Right: evolution of the observed fluxes from 2T fits as a function of phase (with 1 /P = 0.01838 d-1 and T0 = 2 451 751.707), along with the best-fit sinusoid (dotted line).

Concerning magnetic field measurements, the 2015 FORS2 data formally provide only three secure detections (i.e. detections at 5−6σ; see Bagnulo et al. 2012). Nevertheless, these new data sample very different phases of the favoured X-ray timescale, allowing it to be tested (see phases in the second column of Table 2). Unfortunately, the longitudinal field values do not show coherent variations when folded with this timescale (see Fig. 2): secure detections and non-detections are both notably found around phase 0.3. The problem remains even when using only the 2015 data, where the precision of the ephemeris has less impact on the folding; for example, only the first of two consecutive days (May 19 and 20) provide a secure detection. Formally, a χ2 test finds a very small chance (0.6%) for the Bz values to be constant, but in fact the errors are underestimated – only 5σ detections are secure with FORS2 spectropolarimetry when 3σ is usually sufficient (Bagnulo et al. 2012). Indeed, the largest difference between the 2015 values actually only amounts to ~300 G, i.e. around 3σ. Therefore, we can conclude that, within the errors, the field has remained constant over the three months of observations, i.e. variations may be possible, but only with a very small amplitude (<300 G). Furthermore, the 2015 longitudinal field strengths are similar to those based on 2011 or 2013 data, hinting at long-term constancy, though no clear conclusion can be drawn on possibly longer timescales as their sampling is far from perfect. For completeness, we have performed a period search5 on Bz and Nz values, but the derived periodograms are similar hence no significant periodicity can be identified for the stellar field measurements. These results appear somewhat at odds with the X-ray variations. In the context of an oblique rotator model, the flux doubling recorded at high energies suggested a high i + β value for the dipolar field, with large changes in Bz being expected. On the contrary, as large magnetic field variations over tens of days appear excluded by observations, this rather suggests either a pole-on configuration (i ~ 0), an alignment of rotational and magnetic axes (β ~ 0), or a very long rotation period.

thumbnail Fig. 2

Longitudinal field values (filled symbols) and the associated values for the null profiles (dotted empty symbols) as a function of phase (with 1 /P = 0.01838 d-1 and T0 = 2 451 751.707). The black circles correspond to 2011 data, the red triangles to 2013 data, and the green squares to 2015 data.

thumbnail Fig. 3

Left and middle: RVs (left panel) and EWs (middle panel) of selected lines (He ii 4686 shown as open red circles, O iii 5592 as blue crosses, He i 5876 as cyan asterisks, Na i 5890 as green stars, and Hα as black dots) as a function of time or phase (with 1 /P = 0.01838 d-1 and T0 = 2 451 751.707), respectively. Typical errors are <5 km s-1 for RVs and <10% for EWs (see Table 4). Right: Hα profile observed in 2010 (black, HJD = 2 453 40.538), 2011 (red, HJD = 2 455 642.689), 2013 (green, HJD = 2 456 498.481), and 2015 (blue, HJD = 2 457 116.539).

For the optical spectra, the radial velocities and equivalent widths (EWs, estimated by integrating the line over a fixed interval) were measured for lines of different origins (interstellar, photospheric, circumstellar); the results are provided in Table 4. For Hα, this was done after the narrow nebular component was subtracted, which could only be done on the high-resolution spectra, but this procedure is imperfect as demonstrated by e.g. residual [O iii] emissions. The radial velocities of all stellar lines yield coherent results (left panel of Fig. 3). In particular, the systematic, coherent trend towards more negative velocities since 2010 (at a rate of ~3.5 km s-1 per year) should be noted. Formally, velocities are significantly variable: they display a larger dispersion than that of the interstellar Na i line (~7 km s-1 vs. 0.2 km s-1 in high-resolution data) and the maximum RV difference amount to about 20 km s-1 corresponding to ~14σ since typical RV errors on high-resolution spectra are only 1 km s-1 (i.e. RV variability criterion of Sana et al. 2013 is fulfilled). As no obvious line profile changes are detected, these RV variations indicate that Tr16-22 is a probable SB1 with a long period. To find the binary timescale, we applied the same set of period search algorithms as used before. Unfortunately, no unambiguous periodicity could be pinpointed for the RVs measured on Tr16-22, most probably because of the sparse sampling of the optical data. In view of the measured values, however, long periods (≫tens of days, clearly incompatible with the 54 d putative X-ray period) are favoured, and may be more precisely constrained by further monitoring the system.

While detecting RV variations is an interesting result, our main objective was to study magnetically confined winds, and this is performed through analysing line strengths of circumstellar emissions. It is now well known that Balmer lines, He ii 4686, and He i lines are particularly sensitive indicators of confined winds (see e.g. the case of HD 191612 in Nazé et al. 2007). We found that, for any given line (interstellar, stellar, or circumstellar), the EWs always remain similar, with a small dispersion (<0.01 Å on high-resolution data). Only the Hα line displays some significant variations, and Fourier analyses tentatively yield a best period of about 8 d for them. However, aliases are numerous and the current sampling of high-resolution spectra does not enable us to really test such a short period. Furthermore, folding magnetic field values or X-ray fluxes with this period results in a very dispersed graph, hence no clear timescale can be definitely identified. Finally, it should be noted that the Hα EW variations are of small amplitude (0.5 Å difference between the extreme EW values, Fig. 3), an amplitude similar in magnitude to the 0.15 Å difference between extreme EW values recorded for the residual [O iii] λ5007; both EWs actually appear correlated, indicating some impact of the nebular contamination (which varies because of the changing seeing) on the Hα variations: the real Hα variations are thus most probably of even smaller amplitude. These low-amplitude changes (or quasi constancy) seen in optical emission lines agree with the magnetic field results, but are at odds with the large variations detected in X-rays.

Table 4

Radial velocities (in km s-1) and equivalent widths (in Å) measured for some important lines.

4. Conclusion

Spectropolarimetric, X-ray, and high-resolution spectroscopic data of Tr16-22 have been obtained. They show a good agreement with previous results: the X-ray properties are in line with those already reported and the magnetic field values are similar (within errors) to previous measurements. However, there are also surprises.

The optical spectroscopy indicates the presence of RV changes in Tr16-22, suggesting that it is a long-term SB1, but the sparse sampling of the optical data prohibits us from deriving a full orbital solution. The probable binary nature of Tr16-22 cannot explain its exceptional X-ray emission, however. First, colliding-wind emission in the system could be seen as one possible source of X-ray variability, but (1) the long timescale of the RV variations may be difficult to reconcile with the shorter timescale detected in X-ray data and (2) such late-type O stars do not exhibit bright colliding-wind X-ray emission. Second, a low-mass (PMS) companion could be envisaged as the source of the X-ray variations because such objects display X-ray flares, but (1) the observed variations are large (about a factor of 2 corresponding to an increase in flux of >1032 erg s-1), an extreme value for typical PMS flares, (2) PMS flares occur on relatively short timescales while the 2003 data indicate that a month was needed to return to “normal” flux levels, and (3) the combination of a long period and a large RV amplitude (at least 20 km s-1) of the O-type star rule out a companion with very low mass. Finally, it should also be noted that the average luminosity of Tr16-22 (log [LX] ~ 32.3) is similar to expectations from confined wind models (in particular, see the right panel of Fig. 6 in Nazé et al. 2014b, where Tr16-22 is #8).

On the other hand, neither the magnetic field values nor the optical spectroscopy display an obvious modulation with the favoured (54 d) timescale derived from X-rays. Moreover, the measured magnetic field is compatible with a constant value, or at most with a low-amplitude modulation, while the optical line strengths also remain remarkably constant in our data. This near-constancy is reminiscent of HD 148937 (Nazé et al. 2008, 2014b; Wade et al. 2012). It would imply a system with a long period, always seen close to pole-on, or with magnetic and rotational axes aligned, but in these latter cases the X-ray emission of the confined winds would also remain stable, as significant occultation of the X-ray emitting regions by the stellar body would not occur, and this clearly contradicts the X-ray observations of Tr16-22. However, many aliases were present in the X-ray flux periodogram, rendering the choice of the best period difficult and the new data did not fully resolve this ambiguity.

The new data have thus brought additional questions. The variation timescales of the X-ray emission, optical emission line strength, and longitudinal magnetic field remain poorly constrained and – worse – their correlated behaviour has not yet been established, while it is clearly seen in all other magnetic O stars. The near-constancy of magnetic field values and optical emission strengths may even be difficult, if not impossible, to reconcile with the much larger variations detected at high energies. Only additional optical and X-ray observations, carefully scheduled and with high signal-to-noise ratios, would be able to clarify the situation and firmly confirm whether there is a mismatch for Tr16-22 with the usual oblique rotator scenario for confined winds in massive stars.

2

http://iraf.noao.edu/ IRAF is distributed by the National Optical Astronomy Observatories, which are operated by the Association of Universities for Research in Astronomy, Inc., under cooperative agreement with the National Science Foundation.

3

On long-term loan from the University of Liège.

5

We applied several period search algorithms: (1) the Fourier algorithm adapted to sparse/uneven datasets (Heck et al. 1985; Gosset et al. 2001, a method rediscovered recently by Zechmeister & Kürster 2009− these papers also note that the method of Scargle 1982, while popular, is not fully correct, statistically); (2) two different string length methods (Lafler & Kinman 1965; Renson 1978); (3) three binned analyses of variances (Whittaker & Robinson 1944; Jurkevich 1971; which is identical but with no bin overlap, to the “pdm” method of Stellingwerf 1978; and Cuypers 1987; which is identical to the “AOV” method of Schwarzenberg-Czerny 1989); and (4) conditional entropy (Cincotta et al. 1999; Cincotta 1999, see also Graham et al. 2013). Each of these methods has its advantages and its drawbacks; the most reliable is the Fourier method, while the fastest are usually analyses of variances – but the multiple identification of the same signal secure a detection.

Acknowledgments

The authors acknowledge Dr Nolan Walborn for fruitful discussions, Dr Kenji Hamaguchi for sharing his private X-ray data, as well as Francesco Di Mille and Claudio Germana who were observing on the du Pont telescope and kindly obtained the Feb. 2016 observation under NM’s request. Y.N. acknowledges support from the Fonds National de la Recherche Scientifique (Belgium), the Communauté Française de Belgique, the PRODEX XMM contract (Belspo), and an ARC grant for concerted research actions financed by the French community of Belgium (Wallonia-Brussels federation). R.H.B. is grateful for financial support from FONDECYT Regular Project No. 1140076. R.G. was supported by grant PIP 112-201201-00298 (CONICET). R.G. and N.M. were visiting Astronomer, Complejo Astronómico El Leoncito operated under agreement between the Consejo Nacional de Investigaciones Científicas y Técnicas de la República Argentina and the National Universities of La Plata, Córdoba, and San Juan.

References

  1. Albacete Colombo, J. F., Morrell, N. I., Niemela, V. S., & Corcoran, M. F. 2001, MNRAS, 326, 78 [NASA ADS] [CrossRef] [Google Scholar]
  2. Anders, E., & Grevesse, N. 1989, Geo. Chim. Acta, 53, 197 [Google Scholar]
  3. Antokhin, I. I., Rauw, G., Vreux, J.-M., van der Hucht, K. A., & Brown, J. C. 2008, A&A, 477, 593 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  4. Babel, J., & Montmerle, T. 1997a, A&A, 323, 121 [NASA ADS] [Google Scholar]
  5. Bagnulo, S., Szeifert, T., Wade, G. A., Landstreet, J. D., & Mathys, G. 2002, A&A, 389, 191 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  6. Bagnulo, S., Landolfi, M., Landstreet, J. D., et al. 2009, PASP, 121, 993 [NASA ADS] [CrossRef] [Google Scholar]
  7. Bagnulo, S., Landstreet, J. D., Fossati, L., & Kochukhov, O. 2012,A&A, 538, A129 [Google Scholar]
  8. Barbá, R. H., Gamen, R., Arias, J. I., et al. 2010, Rev. Mex. Astron. Astrofis. Conf. Ser., 38, 30 [NASA ADS] [Google Scholar]
  9. Bohlin, R. C., Savage, B. D., & Drake, J. F. 1978, ApJ, 224, 132 [NASA ADS] [CrossRef] [Google Scholar]
  10. Cincotta, P. M. 1999, MNRAS, 307, 941 [NASA ADS] [CrossRef] [Google Scholar]
  11. Cincotta, P. M., Helmi, A., Mendez, M., Nunez, J. A., & Vucetich, H. 1999, MNRAS, 302, 582 [NASA ADS] [CrossRef] [Google Scholar]
  12. Corcoran, M. F., Pittard, J. M., Marchenko, S. V., & Xmega Group 1999, Wolf-Rayet Phenomena in Massive Stars and Starburst Galaxies, Proc. Symp. IAU, 193, 772 [Google Scholar]
  13. Cuypers, J. 1987, A&AS, 69, 445 [NASA ADS] [Google Scholar]
  14. Donati, J.-F., Semel, M., Carter, B. D., Rees, D. E., & Collier Cameron, A. 1997, MNRAS, 291, 658 [NASA ADS] [CrossRef] [MathSciNet] [Google Scholar]
  15. Donati, J.-F., Babel, J., Harries, T. J., et al. 2002, MNRAS, 333, 55 [NASA ADS] [CrossRef] [MathSciNet] [Google Scholar]
  16. Donati, J.-F., Howarth, I. D., Bouret, J.-C., et al. 2006, MNRAS, 365, L6 [NASA ADS] [CrossRef] [Google Scholar]
  17. Evans, N. R., Schlegel, E. M., Waldron, W. L., et al. 2004, ApJ, 612, 1065 [NASA ADS] [CrossRef] [Google Scholar]
  18. Fossati, L., Castro, N., Schöller, M., et al. 2015, A&A, 582, A45 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  19. Gagné, M., Oksala, M. E., Cohen, D. H., et al. 2005, ApJ, 628, 986; Err, 634, 712 [NASA ADS] [CrossRef] [Google Scholar]
  20. Gosset, E., Royer, P., Rauw, G., Manfroid, J., & Vreux, J.-M. 2001, MNRAS, 327, 435 [NASA ADS] [CrossRef] [Google Scholar]
  21. Graham, M. J., Drake, A. J., Djorgovski, S. G., Mahabal, A. A., & Donalek, C. 2013, MNRAS, 434, 2629 [NASA ADS] [CrossRef] [Google Scholar]
  22. Heck, A., Manfroid, J., & Mersch, G. 1985, A&AS, 59, 63 [NASA ADS] [Google Scholar]
  23. Howarth, I. D., Walborn, N. R., Lennon, D. J., et al. 2007, MNRAS, 381, 433 [NASA ADS] [CrossRef] [Google Scholar]
  24. Jurkevich, I. 1971, Ap&SS, 13, 154 [NASA ADS] [CrossRef] [Google Scholar]
  25. Lafler, J., & Kinman, T. D. 1965, ApJS, 11, 216 [NASA ADS] [CrossRef] [Google Scholar]
  26. Nazé, Y., Rauw, G., Pollock, A. M. T., Walborn, N. R., & Howarth, I. D. 2007, MNRAS, 375, 145 [NASA ADS] [CrossRef] [Google Scholar]
  27. Nazé, Y., Walborn, N. R., Rauw, G., et al. 2008, AJ, 135, 1946 [NASA ADS] [CrossRef] [Google Scholar]
  28. Nazé, Y., ud-Doula, A., Spano, M., et al. 2010, A&A, 520, A59 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  29. Nazé, Y., Broos, P. S., Oskinova, L., et al. 2011, ApJS, 194, 7 [NASA ADS] [CrossRef] [Google Scholar]
  30. Nazé, Y., Bagnulo, S., Petit, V., et al. 2012, MNRAS, 423, 3413 [NASA ADS] [CrossRef] [Google Scholar]
  31. Nazé, Y., Wade, G. A., & Petit, V. 2014a, A&A, 569, A70 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  32. Nazé, Y., Petit, V., Rinbrand, M., et al. 2014b, ApJS, 215, 10; Err, 224, 13 [NASA ADS] [CrossRef] [Google Scholar]
  33. Petit, V., Owocki, S. P., Wade, G. A., et al. 2013, MNRAS, 429, 398 [NASA ADS] [CrossRef] [Google Scholar]
  34. Renson, P. 1978, A&A, 63, 125 [NASA ADS] [Google Scholar]
  35. Sana, H., de Koter, A., de Mink, S. E., et al. 2013, A&A, 550, A107 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  36. Scargle, J. D. 1982, ApJ, 263, 835 [NASA ADS] [CrossRef] [Google Scholar]
  37. Schwarzenberg-Czerny, A. 1989, MNRAS, 241, 153 [NASA ADS] [CrossRef] [Google Scholar]
  38. Stellingwerf, R. F. 1978, ApJ, 224, 953 [NASA ADS] [CrossRef] [Google Scholar]
  39. ud-Doula, A., Owocki, S. P., & Townsend, R. H. D. 2009, MNRAS, 392, 1022 [NASA ADS] [CrossRef] [Google Scholar]
  40. ud-Doula, A., Owocki, S., Townsend, R., Petit, V., & Cohen, D. 2014, MNRAS, 441, 3600 [NASA ADS] [CrossRef] [Google Scholar]
  41. Wade, G. A., Grunhut, J., Gräfener, G., et al. 2012, MNRAS, 419, 2459 [NASA ADS] [CrossRef] [Google Scholar]
  42. Whittaker, E. T., & Robinson, G. 1944, The calculus of observations; a treatise on numerical mathematics, eds. E. T. Whittaker, & G. Robinson (London: Blackie) [Google Scholar]
  43. Zechmeister, M., & Kürster, M. 2009, A&A, 496, 577 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]

All Tables

Table 1

List of the new X-ray observations.

In the text
Table 2

Results from the X-ray spectral fitting.

In the text
Table 3

Results from the spectropolarimetry (using rectification and within the same spectral windows).

In the text
Table 4

Radial velocities (in km s-1) and equivalent widths (in Å) measured for some important lines.

In the text

All Figures

thumbnail Fig. 1

Left and middle: evolution of the hardness ratios (HR = H/S = FISMcor(2.−10.0 keV) /FISMcor(0.5−2. keV)) as a function of observed fluxes. Red triangles correspond to the 2003 XMM-Newton data (the year with the largest number of observations), green crosses to the new data, and black open circles to the other XMM-Newton observations. The values shown in the left (resp. middle) panel correspond to the results from 4T fits (resp. 2T fits). Right: evolution of the observed fluxes from 2T fits as a function of phase (with 1 /P = 0.01838 d-1 and T0 = 2 451 751.707), along with the best-fit sinusoid (dotted line).

In the text
thumbnail Fig. 2

Longitudinal field values (filled symbols) and the associated values for the null profiles (dotted empty symbols) as a function of phase (with 1 /P = 0.01838 d-1 and T0 = 2 451 751.707). The black circles correspond to 2011 data, the red triangles to 2013 data, and the green squares to 2015 data.

In the text
thumbnail Fig. 3

Left and middle: RVs (left panel) and EWs (middle panel) of selected lines (He ii 4686 shown as open red circles, O iii 5592 as blue crosses, He i 5876 as cyan asterisks, Na i 5890 as green stars, and Hα as black dots) as a function of time or phase (with 1 /P = 0.01838 d-1 and T0 = 2 451 751.707), respectively. Typical errors are <5 km s-1 for RVs and <10% for EWs (see Table 4). Right: Hα profile observed in 2010 (black, HJD = 2 453 40.538), 2011 (red, HJD = 2 455 642.689), 2013 (green, HJD = 2 456 498.481), and 2015 (blue, HJD = 2 457 116.539).
In the text

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.