X-ray observations of WR 25

Phase-Resolved XMM-Newton and Swift Observations of Wr 25

J. C. Pandey, S. B. Pandey, and Subhajeet Karmakar Aryabhatta Research Institute of Observational Sciences (ARIES), Nainital-263 002, India; jeewan@aries.res.in

We present an analysis of long-term X-ray and optical observations of the Wolf-Rayet binary WR 25. Using archival data from observations with the XMM-Newton and the Swift observatories spanning over yr, we show that WR 25 is a periodic variable in X-rays with a period of days. X-ray light curves in the 0.5-10.0 keV energy band show phase-locked variability, where the flux increased by a factor of from minimum to maximum, being maximum near periastron passage. The light curve in the soft energy band (0.5-2.0 keV) shows two minima indicating the presence of two eclipses. However, the light curve in the hard energy band (2.0-10.0 keV) shows only one minimum during the apastron passage. The X-ray spectra of WR 25 were explained by a two-temperature plasma model. Both the cool and the hot plasmas were constant at and keV throughout an orbital cycle, where the cooler plasma could be due to the small scale shocks in a radiation-driven outflow and the high temperature plasma could be due to the collision of winds. The column density varied with the orbital phase and was found to be maximum after the periastron passage, when the WN star is in front of the O star. The abundances of WR 25 were found to be non-solar. Optical V-band data of WR 25 also show the phase-locked variability, being at maximum near periastron passage. The results based on the present analysis indicate that WR 25 is a colliding wind binary where the presence of soft X-rays is attributed to individual components; however, hard X-rays are due to the collision of winds.

Subject headings:
Star:individual (WR 25) – star:binary – star:X-ray – star:Wolf-Rayet – star:wind
slugcomment: The Astrophysical Journal, 787:1, 2014

1. Introduction

Massive O-type stars evolve into Wolf-Rayet (WR) phases when their hydrogen fuel has been consumed and products of nuclear fusion appear in their atmosphere, before ending their lives as core-collapse supernovae (Doom, 1987; Smartt et al., 2009; Smartt, 2009). Typically, the progenitors of WR stars have initial masses greater than 25 (Crowther, 2007), and they spend 10% of their 5 Myr lifetime in the WR phase (Meynet & Maeder, 2005). Spectra of WR stars are predominantly characterized by emission lines of He and N (WN stars), He and C (WC stars) and He and O (WO stars). In the evolutionary sequence of massive stars, WC and WO stars are expected to correspond to later stages than WN stars. WR stars are known to produce strong stellar winds driven by their strong radiation field. The stellar winds can reach velocities up to 1000-3000 km s with mass-loss rates of yr, depending upon mass and age (Hamann et al., 2006). These winds not only affect the evolution of WR stars but also have a tremendous impact on their ambient media. The detailed physical properties of WR stars are summarized in many past reviews (Abbott & Conti, 1987; van der Hucht, 1992; Crowther, 2007).

X-ray emissions from WR binaries may consist of the combined emission from intrinsic stellar wind shocks and colliding wind shocks between the two binary components (Prilutskii & Usov, 1976; Cherepashchuk, 1976; Luo et al., 1990; Usov, 1992; Stevens et al., 1992). Colliding wind binary systems often exhibit periodic X-ray modulation either because of the change in binary separation or due to the changing circumstellar opacity along the line of sight to the collision zone, resulting from the orientation of the system with respect to the observer. Serious investigations of these variations were carried out after the advent of high-quality facilities like XMM-Newton  and Chandra. Many O+O and WR+O binaries as observed with XMM-Newton, show different kinds of phase-locked modulations e.g., V444 Cyg and CD Cru (Bhatt et al., 2010a), HD 159176 (De Becker et al., 2004), HD 152248 (Sana et al., 2004), HD 93403 (Rauw et al., 2002), etc. It is believed that single WR and OB stars emit soft X-rays (kT 1 keV) via shocks that are set up by instabilities in their supersonic line-driven winds (Lucy & White, 1980; Lucy, 1982; Owocki et al., 1988). Gräfener & Hamann (2005) have shown that winds of WR stars can be driven by radiation pressure alone if multi-scattering effects are taken into account.

WR 25 (=HD 93162) is a bright (V 8.03 mag) WR binary located in the Carina Nebula region. WR 25 is classified as WN6h + O4f (van der Hucht, 2001). Gamen et al. (2006) detected periodic radial velocity variations in WR 25 and suggested that WR 25 is an eccentric binary system with a probable period of about 208 days. The basic parameters of WR 25 are summarized in Table 1. The origin of the large X-ray flux of WR 25 is suggestive of colliding wind emission in a binary system (Pollock, 1987; Raassen et al., 2003; Pollock & Corcoran, 2006). The X-ray to bolometric luminosity () ratio of (Seward & Chlebowski, 1982) for WR 25 is an order of magnitude higher than those observed for single massive stars. Raassen et al. (2003) noticed that the X-ray luminosity of WR 25 had remained relatively constant over a time span of 10 yr. Later, Pollock & Corcoran (2006) found an increase in X-ray luminosity of more than a factor of two and suggested that the observed X-ray variability is a result of colliding wind emission in a moderate eccentric binary. In this context, we use the full set of archival XMM-Newton and Swift X-ray observations of WR25 to further investigate the properties of the colliding winds in this system. Our aim also extends to search for variability in the V-band using the All Sky Automated Survey [ASAS; Pojmanski (2002)] archival data.

The paper is organized along the following lines: Section 2, describes the observations and data reduction. The light curve analysis is given in Section 3, X-ray spectral properties of WR 25 are described in Section 4. Section 5 describes the V-band optical observations. In Section 6, we present the discussion and conclusions.

Parameters Value Ref Parameters Value Ref
Period (d) 1 (km s) 2480 2
() 1 (M/yr) 10 2
V (km s) 1 dist.(kpc) 3.24 3
K (km s) 1 () 10 2
() 1 E(B-V) (mag) 0.63 3
T (HJD) 1 4
() 1

Here: - eccentricity, V - radial velocity, K - radial velocity amplitude, - orientation of periastron, T - Julian date of periastron passage, - semi major axis, - terminal velocity, - mass loss rate
References: (1) Gamen et al. (2006); (2) Crowther & Dessart (1998); (3) van der Hucht (2001); (4) Seward & Chlebowski (1982)

Table 1Basic parameters of WR+O binary, WR 25.

2. Observations and data reduction

2.1. XMM-Newton

WR 25 was observed with the XMM-Newton  satellite using various detector configurations on twenty occasions from the year 2000 to 2009, spanning 8.5 yr. The XMM-Newton  satellite is composed of three coaligned X-ray telescopes (Jansen et al., 2001), which simultaneously observe a source, accumulating photons in three CCD-based instruments, namely the nearly-identical MOS1 and MOS2 (Turner et al., 2001) detectors and the PN (Strüder et al., 2001) detectors, which comprise the European Photon Imaging Camera (EPIC). The EPIC instrument provides imaging and spectroscopy in the energy range from 0.15 to 15 keV with an angular resolution of 4.5-6.6 arcsec and a spectral resolution () of 20-50. Exposure times for these observations were in the range of 6-60 ks. A log of observations is provided in Table 2.

The data were reduced with standard XMM-Newton Science Analysis System (SAS) software, version 12.0 using version 3.1 calibration files. The pipeline processing of raw EPIC Observation Data Files was done using the epchain and emchain tasks which allow calibrations both in energy and astrometry of the events registered in each CCD chip. We have restricted our analysis to the energy band 0.5 - 10.0 keV as the background contribution is particularly relevant at high energies where stellar sources have very little flux and are often undetectable. Event list files were extracted using the SAS task evselect. The epatplot task was used for checking for pile-up effects and no observations were affected by pile-up. Data from the three cameras were individually screened for the time intervals with high background. The observations affected by high background flaring events were excluded (see Table 2). X-ray light curves and spectra of WR 25 were generated from on-source counts obtained from circular regions with a radius of 30 around the source. The background was chosen from several source-free regions on the detectors surrounding the source. We used the tool epiclccorr to correct for good time intervals, dead time, exposure, PSF, quantum efficiency and background subtraction. The SAS task especget was used to generate the spectra, which also computes the photon redistribution as well as the ancillary matrix. Finally, the spectra were rebinned to have at least 20 counts per spectral bin.

2.2. Swift

WR 25 has also been regularly monitored by the Swift X-ray telescope (XRT) since 2007 November. The XRT observes from 0.3 to 10 keV using CCD detectors, with energy resolution of eV at keV (Burrows et al., 2005). The exposure times for the XRT observations varied from 1 to 26 ks. All XRT data were collected in photon counting mode. A log of observations is given in Table 3. We have excluded observations with exposure times less than 2 ks due to poor or no signal. In order to produce the cleaned and calibrated event files, all the data were reduced using the Swift  xrtpipeline task (version 0.12.6) in which standard event grades of 0-12 were selected, and calibration files were used from the CALDB 2.8 release111http://heasarc.gsfc.nasa.gov/docs/heasarc/caldb/caldb_intro.html.

For every observation, images, light curves and spectra were obtained with the xselect (verson 2.4) package. For each observation, source spectra and light curves were extracted from a circular region with a radius of 30. For the background estimation, an annular region with inner and outer radii of 69 and 127 were used around the source region. The spectra were grouped to have a minimum of twenty counts per energy bin with grppha. The spectra were corrected for the fractional exposure loss due to bad columns on the CCD. For this, we created exposure maps with the xrtexpomap task, which is used as an input to generate the ARF with the xrtmkarf task. For the RMF, the latest version was used from the HEASARC calibration data base.

Data Rev. Observation Detector Obs. Date U.T. Exposure Bkg flare Effective Off axis
set ID (Filter) (YYYY-MM-DD) (hh:mm:ss) Time(s) from-to (ks) exposure(s) ()
x01 115 0112580601 M1(TH) 2000-07-26 05:07:47 33589 32.8-33.6 32761 6.864
M2(TH) 06:00:45 30488 29.6-30.5 29585
PN(TH) 05:48:51 31198 30.3-31.2 30298
x02 116 0112580701 M1(TH) 2000-07-27 23:57:54 10991 - 10991 6.877
M2(TH) 2000-07-28 00:50:51 7991 - 7991
PN(TH) 2000-07-28 00:38:57 8999 - 8999
x03 283 0112560101 M1(TH) 2001-06-25 07:34:37 34249 0-9.6 24681 0.066
M2(TH) 07:40:30 33894 0-9.6 24325
PN(TH) 06:51:26 33133 0-8.7 24433
x04 284 0112560201 M1(TH) 2001-06-28 07:29:20 35886 30.0-35.9 30000 0.094
M2(TH) 07:29:20 35890 30.0-35.9 30000
PN(TH) 07:22:56 37364 27.7-37.4 27733
x05 285 0112560301 M1(TH) 2001-06-30 04:45:30 37107 33.6-37.1 33636 0.066
M2(TH) 04:45:28 37108 33.6-37.1 33636
PN(TH) 05:23:05 34526 31.4-34.5 31381
x06 573 0145740101 M1(TH) 2003-01-25 12:58:20 6812 - 6812 7.912
M2(TH) 12:58:20 6813 - 6813
x07 574 0145740201 M1(TH) 2003-01-27 01:03:43 6961 - 6961 7.855
M2(TH) 01:03:43 6961 - 6961
x08 574 0145740301 M1(TH) 2003-01-27 20:37:04 6812 - 6812 7.856
M2(TH) 20:37:05 6812 - 6812
x09 575 0145740401 M1(TH) 2003-01-29 01:40:32 8311 - 8311 7.824
M2(TH) 01:40:33 8311 - 8311
x10 575 0145740501 M1(TH) 2003-01-29 23:55:06 6811 - 6811 7.815
M2(TH) 23:55:07 6811 - 6811
x11 640 0160160101 M1(TH) 2003-06-08 13:30:32 28415 0-1.4,8.9- 12735 5.965
M2(TH) 13:30:22 28414 as above 12735
x12 662 0145780101 M1(TH) 2003-07-22 01:51:36 8300 - 8300 6.335
M2(ME) 01:51:29 8300 - 8300
x13 668 0160560101 M1(TH) 2003-08-02 21:01:14 17649 4.7-13.7 8649 6.516
M2(ME) 21:01:03 17667 4.7-13.7 8649
x14 671 0160560201 M1(TH) 2003-08-09 01:44:20 12652 9.5-10.3 11863 6.557
M2(ME) 01:44:08 12652 9.5-10.3 11863
x15 676 0160560301 M1(TH) 2003-08-18 15:23:34 18850 - 18850 7.070
M2(ME) 15:23:22 18850 - 18850
x16 1126 0311990101 M1(TH) 2006-01-31 18:04:22 63034 - 63034 7.656
M2(TH) 18:04:20 66036 - 66036
PN(TH) 18:26:30 30000 0-3.0 26961
x17 1662 0560580101 M1(TH) 2009-01-05 10:23:02 14602 - 14602 8.025
M2(TH) 10:22:53 14602 - 14602
x18 1664 0560580201 M1(TH) 2009-01-09 14:28:07 11603 - 11603 7.999
M2(TH) 14:27:58 11603 - 11603
x19 1667 0560580301 M1(TH) 2009-01-15 11:22:56 26552 - 26552 7.942
M2(TH) 11:22:46 26552 - 26552
x20 1676 0560580401 M1(TH) 2009-02-02 05:31:18 23903 - 23903 7.757
M2(TH) 04:46:09 26613 - 26613
PN(TH) 04:45:24 24669 - 24669

M1 and M2 stand for MOS1 and MOS2, respectively. TH and ME stand for thick and medium filters, respectively. Exposure start time. Background proton flare duration during the observations.
PIs of observations were Dr. Albert Brinkman, Dr. Michael Corcoran, Dr. Fred Jansen, and Dr. Kenji Hamaguchi

Table 2Log of observations of WR 25 from XMM-Newton.
Set Observation Obs. Date U. T. Exp. offset
ID (YY-MM-DD) (hh:mm:ss) Time(s) ()
s01 00037049001 2007-12-11 14:09:00 4490.5 6.339
s02 00037049002 2008-01-19 06:08:01 3246.6 4.682
s03 00031097001 2008-01-25 02:00:01 4440.5 1.342
s04 00031097002 2008-01-26 02:10:01 3717.2 0.387
s05 00031097003 2008-01-27 00:44:00 3452.8 3.892
s06 00031097004 2008-01-28 00:57:00 3787.7 0.584
s07 00031097006 2008-01-30 17:18:01 3876.7 0.803
s08 00031097007 2008-01-31 06:17:01 2585.1 1.518
s09 00031097008 2008-02-01 00:04:00 26394.8 0.182
s10 00031097009 2008-02-02 23:59:01 22568.9 0.999
s11 00031097010 2008-02-05 16:40:01 3468.7 1.396
s12 00031097011 2008-02-06 00:27:26 3332.3 3.913
s13 00031097012 2008-02-07 00:34:46 3372.7 1.334
s14 00031097013 2008-02-09 21:33:01 2785.1 0.923
s15 00031097016 2008-02-11 17:02:01 2478.9 2.168
s16 00031097017 2008-02-12 00:45:01 2967.9 1.300
s17 00031097018 2008-02-13 07:15:01 2598.9 1.590
s18 00031097019 2008-02-14 15:22:01 3382.0 0.412
s19 00031097020 2008-02-15 12:29:01 2422.4 1.264
s20 00031097022 2008-02-17 12:52:01 3692.9 3.464
s21 00031097023 2008-02-18 17:30:00 3389.9 1.234
s22 00031097025 2008-02-20 06:15:01 3392.9 2.647
s23 00031097026 2008-02-21 07:57:01 3646.4 2.027
s24 00031097028 2008-02-23 14:33:01 2977.3 1.415
s25 00031097029 2008-02-24 16:14:00 2349.4 0.638
s26 00031097030 2008-02-25 11:35:01 2408.8 0.816
s27 00031097031 2008-02-26 05:10:01 3835.2 2.159
s28 00031097034 2008-03-16 08:35:01 9246.5 1.688
s29 00031308001 2008-12-14 03:17:01 6758.9 6.929
s30 00090033001 2009-03-18 08:44:01 14363.4 6.457
s31 00031097035 2009-04-01 12:10:22 9225.9 1.270
s32 00031097036 2009-04-02 08:59:01 12517.4 1.741
s33 00031097037 2009-04-03 09:10:01 10208.9 1.064
s34 00031097038 2009-04-07 09:24:01 8674.8 0.911
s35 00031097039 2009-04-08 07:53:01 8259.7 1.191
s36 00031097040 2009-04-09 13:04:00 9845.3 1.465
s37 00031097041 2009-06-03 01:46:01 3790.9 0.348
s38 00031097042 2009 06 04 06:51:00 3923.2 1.858
s39 00031097043 2009-06-05 06:58:00 3364.8 1.595
s40 00031097044 2009-06-06 06:50:01 4194.7 3.771

Proposers of observations were F. Senziani, A. Pollock, E. Pian and M. Corcoran.

Table 3Log of observations of WR 25 from Swift.

2.3. ASAS V-band observations

WR 25 was observed from 2000 December 3 to 2009 December 3 by ASAS222http://www.astrouw.edu.pl/asas. We have used only ‘A’ and ‘B’ grade data within 1 arcsec to the target of WR 25. ASAS photometry provides five sets of magnitudes corresponding to five aperture values varying in size from 2 to 6 pixels in diameter. For bright objects, Pojmanski (2002) suggested that the magnitudes corresponding to the largest aperture (i.e. the aperture diameter of 6 pixels) are useful. Therefore, we took magnitudes corresponding to the largest aperture for further analysis.

(a) XMM-Newton-MOS
(b) Swift-XRT
Figure 1.— X-ray light curves in total (0.3-10.0 keV), hard (2.0-10.0 keV) and soft (0.3-2.0 keV) energy bands as observed from (a) XMM-Newton-EPIC and (b) Swift-XRT.

3. X-ray light curves and period analysis

The background-subtracted X-ray light curves as observed from XMM-Newton-EPIC, and Swift-XRT in the total (0.5-10.0 keV), hard (2.0-10.0 keV) and soft (0.5-2.0 keV) energy bands are shown in Figures 1 (a) and (b), respectively running from top to bottom. The light curves were binned at 2000 s intervals. The variability in each band is clearly seen. For additional confirmation of the variability in all bands, the significance of deviations from the mean count rate were measured using the - test, defined as


where is the average count rate, C is the count rate of observations and is the error corresponding to C. The statistic was compared against a critical value () for 99.9% significance level, obtained from the -probability function. For the XMM-Newton MOS light curves, values were obtained to be 6440, 4403 and 4499 in the total, hard and soft bands, respectively. These values of are very large in comparison to the of 1430 for 1599 degrees of freedom. For Swift -XRT light curves in total, hard and soft bands, values of 5188, 5113 and 5124 were obtained, respectively, with 393 for 483 degrees of freedom. This indicates that WR 25 is essentially variable in all X-ray bands.

The present long-term X-ray data as observed from XMM-Newton permit us to derive the orbital period of WR 25. We have performed a period analysis of light curves in all bands with a Lomb-Scargle periodogram (Lomb, 1976; Scargle, 1982). The plots from top to bottom in Figure2 show the power spectra in the total, hard and soft bands, respectively. The frequency corresponding to the highest peak was found to be cycles day in all bands. Other frequencies were also noticed in the power spectra but are not consistent with each other among all the energy bands. Thus, the corresponding period of days appears to be real. Also, the derived period is very similar to the orbital period (= days) derived by Gamen et al. (2006) using radial velocity measurements.

Further, the X-ray light curves as observed from XMM-Newton-MOS and Swift-XRT in the total, hard and soft energy bands were folded using the ephemeris HJD = 2451958.0 + 207.85E (Gamen et al., 2006) and are shown in Figures 3(a) and (b), respectively. The zero phase in the folded light curves corresponds to the time of periastron (HJD 2451958.0). X-ray light curves in each band from both observations show similar behavior. The light curves in the individual bands show phase-locked variability. In the soft and total-band light curves, the count rates decrease when going from orbital phase 0.0 (i.e., near the periastron passage) to orbital phase then increase up to phase . Count rates further decrease to phase before reaching a maximum value near periastron passage. However, in the hard band, the count rates were systematically decreasing when going from periastron to apastron. The ratio of maximum to minimum count rates in total, soft and hard bands were found to be 1.9, 2.6, and 1.7 for XMM-Newton-MOS observations, and 2.5, 2.6, and 2.9 for Swift-XRT observations, respectively.

The hardness ratio (HR) defined as (Hard -Soft)/(Hard+Soft) can reveal information about the spectral variations. The HR curve displayed in plot (d) of Figure 3 exhibits similar behavior to that of the light curves in the hard band. The maximum value of the HR during periastron passage indicates a harder spectrum.

Figure 2.— Lomb-Scargle power spectra of X-ray data in different energy bands as observed from EPIC detectors of XMM-Newton.
Figure 3.— Folded light curves using the ephemeris derived by Gamen et al. (2006) in different energy bands as observed from (a) XMM-Newton (b) Swift.

4. X-ray Spectral Analysis

X-ray spectra of WR 25 as observed by the EPIC detector of XMM-Newton at different orbital phases are shown in Figure 4. Below 1 keV, the X-ray spectra were found to be affected by high extinction. Strong emission lines like Fe XVII (0.8 keV), Ne X (1.02 keV), Mg XII (1.47 keV), Si XIII (1.853 keV), S XV (2.45 keV), Ar XVII (3.12 keV),Ca XIX+XX (3.9 keV), and Fe XXV (6.67 keV) were identified in the X-ray spectra of WR 25. In order to derive the spectral parameters at different orbital phases, we performed spectral analysis of each data set corresponding to different orbital phases using simultaneous/joint fitting of EPIC data by models of the Astrophysical Plasma Emission Code [APEC; Smith et al. (2001)], as implemented in the X-ray spectral fitting package xspec (Arnaud, 1996) version 12.7.1. For spectral fitting, we adopted the similar approach of Raassen et al. (2003) and Pollock & Corcoran (2006). The form of the model used was wabs(ism)*wabs(local)*(vapec+vapec). A minimization gave the best fit model to the data. The presence of interstellar material along the line of sight and the local circumstellar material around the stars can modify the X-ray emission from massive stars. We have applied the local absorption in the line of sight to the source using photoelectric absorption cross-sections according to Morrison & McCammon (1983) and modeled as wabs with two absorption components i.e. interstellar medium () and local () hydrogen column densities. Assuming the E(B-V ) 0.63 mag (van der Hucht, 2001) and a normal interstellar reddening law towards WR 25, and using the relation given by Gorenstein (1975), the   was estimated to be cm. For the first stage of spectral fitting, we fixed the  and abundances of He(=2.27), C(=0.15), and N(=5.9), and varied other parameters for all phases. The values of He, C, and N abundances were adopted from the optical spectrum of WR 25 (Crowther et al., 1995). The temperatures for both components were found to be constant within the level at all phases. However, abundances of Ne, Mg, Al, Si, Ca, Ar, Fe, and Ni were found to be constant within the 1-2 level. Average values of elemental abundances and temperatures are given in Table 4. In the next stage of spectral fitting, we fixed temperatures and abundances of all elements at their average values for all phases along with , and varied , and normalizations of both components. Many spectra were taken at the same orbital phase; therefore, joint spectral fittings were performed for those spectra, which were observed within a difference of in the orbital phase. In this way, from XMM-Newton observations, we have nine data points over an orbital cycle of WR 25. The final set of the best-fitted parameters is given in Table 5.

(a) Phase - 0.060
(b) Phase - 0.141
(c) Phase - 0.360
(d) Phase - 0.438
(e) Phase - 0.589
(f) Phase - 0.617
(g) Phase - 0.735
(h) Phase - 0.779
(i) Phase - 0.988
Figure 4.— X-ray spectra of WR 25 as observed from MOS and PN detectors along with the best fit folded 2T vapec model at nine epochs. WR 25 was observed with the PN detector only at three phases, namely 0.360, 0.438 and 0.735. The PN spectra are shown on the top at these three epochs. The lower panels show the residual in terms of the ratio of the data and model. The emission lines of NeX, Mg XI+XII, Si III+IV, S XV+XVI, Ar XVII+XVIII, Ca XIX+XX and Fe (K-shell) are seen in the spectra.

The method used for the fitting of XMM-Newton-EPIC spectra was also applied for the fitting of spectra observed from the Swift-XRT. Since WR 25 was observed more than 40 times by the Swift-XRT and many observations were taken with short exposures, the spectra with a phase difference of 0.02 were fitted jointly. In this way, we have spectra at 12 different phases over an orbital cycle from Swift observations. For the spectral fitting of Swift-XRT data, we have fixed temperatures and abundances to values as obtained from spectral fitting of XMM-Newton-EPIC data. However,  and normalizations of both components were kept as free parameters in the spectral fitting. The best fit parameters from the spectral fitting of Swift -XRT data are also given Table 5. We have compared various parameters derived from both observatories. It appears that the X-ray fluxes and other spectral parameters derived from spectral fitting of both satellite data are consistent. Comparison of spectral parameters derived from both telescopes may not be possible due to the lack of simultaneous observations. However, Plucinsky et al. (2012) has shown that the spectral parameters from both observations are in agreement within .

The X-ray fluxes of WR 25 are estimated using the cflux model in xspec and are corrected for . The X-ray luminosities of WR 25 in soft (), hard () and total () energy bands were derived by using the corresponding unabsorbed flux values and a distance of 3.24 kpc. The EMs (EMs), EM and EM,  corresponding to both the cool and the hot plasma components are derived from the normalization parameters. We have plotted , , , , EM, and EM as a function of orbital phase in Figure 5 (a) and 5 (b) for observations from XMM-Newton and Swift, respectively. The maximum value of  was found near periastron passage and dropped suddenly to a phase and then increased to phase . After phase ,  decreased to phase . It appears that  peaked twice during an orbital cycle. The minimum value  was observed just after periastron passage, which was times lower than the observed maximum value. A similar tendency of X-ray luminosity was also observed in the total energy band, where the maximum to minimum flux ratio was found to be . Apart from the periastron passage, no other peak of  was noticed in an orbital cycle of WR 25. The minimum value of  was observed near phase and was times lower than that observed during the periastron passage. The value of  was observed to be maximum after the periastron passage at phase 0.03. However, the minimum value of  was observed at phase . The EMs, EM and EM corresponding to the cool and the hot temperature components were also found to be phase dependent, with maximum at periastron passage and minimum at apastron passage.

Parameters Value Parameters Value
kT (keV) 0.628 Si 1.07
kT (keV) 2.75 S 1.59
O Ar 1.30
Ne 1.60 Ca 1.67
Mg 1.38 Fe 0.68
Al 4.42 Ni 3.93
Table 4Average values of temperatures and abundances of WR 25 as obtained from X-ray spectral fitting. Abundances are in units of Solar photospheric values.
(a) XMM-Newton EPIC observations
(b) Swift XRT observations
Figure 5.— Spectral parameters as a function of orbital phase as observed from (a) XMM-Newton-EPIC and (b) Swift-XRT, where X-ray luminosities are in units of ,  is in units of cm, and EM and EM are in units of 10 cm.

5. Optical V-band light curve

The V-band light curve of WR 25 as obtained from the ASAS archive is plotted in Figure 6 (a). In ASAS observations, error bars are very large (i.e. 0.035 to 0.1 mag). Therefore, it is difficult to search for any small scale variability that might be present. In order to reduce the short term fluctuations, a moving average of 10 data points in the forward direction was performed. The light curve of the moving-averaged data points is shown in Figure 6 (b). The V-band light curve does not appear to be constant over the time span of the observations; however, variations are not statistically significant. The was found to be 270 for 956 degrees of freedom (see Equation 1). This value of is very low in comparison to the of 826 corresponding to the 99.9% significance level. Furthermore, we have folded the V-band data with phase bins of 0.1 and using the ephemeris given by Gamen et al. (2006). The folded V-band light curve of WR 25 is shown in Fig 6 (c). It appears that variability is present in the light curve. The variability amplitude was found to be mmag. The maximum brightness was seen near periastron passage and after that the brightness decreased toward apastron passage, being minimum near phase 0.75.

(a) V-band light curve
(b) V-band light curve (running average)
(c) Folded light curve
Figure 6.— (a) V-band light curve WR 25 as observed in ASAS, (b) light curve after smoothing (see the text for details) (c) folded light curve of WR 25 using the ephemeris HJD = 2451598.0 + 207.85 E as derived by Gamen et al. (2006).

6. Discussion and conclusions

We have carried out X-ray and optical studies of WR 25 using X-ray data from the XMM-Newton and Swiftsatellites and optical data from the ASAS archive. The X-ray spectra of WR 25 at all orbital phases are well-explained by a two-temperature plasma model. The temperatures of both components were found to be constant throughout an orbital cycle. The temperature values are similar to those derived by Pollock & Corcoran (2006) and Raassen et al. (2003) using XMM-Newton data. The temperature of the cool component (=0.628 keV) of WR 25 is comparable to that of other WN-subtype WR+O binaries: for example WR1 [0.56-0.67 keV; (Ignace et al., 2003)], WR 22 [0.6 keV; (Gosset et al., 2009)] WR 47 [ keV; (Bhatt et al., 2010a)], WR 139 [ (Bhatt et al., 2010a)], and WR 147 [0.7-0.8 keV; (Skinner et al., 2007)]. A plasma temperature of 0.6-0.8 keV is also dominant for massive OB stars (Sana et al., 2006; Nazé, 2009; Bhatt et al., 2010b). The soft X-ray component may originate in the winds of the individual components of WR 25 via radiation driven instability shocks. The ratio between the wind momentum () and the momentum of the radiation field () for WR 25 was derived to be (Hamann et al., 2006), indicating that wind of WR 25 is driven by radiation pressure. Furthermore, the derived temperature of the cool component appears to be realistic for radiation-driven wind shocks. The derived plasma temperature for WR 25 can be used to estimate the pre-shock velocity using the relation keV (Luo et al., 1990), where is mean mass per particle in units of the proton’s mass (1.16 for a WN star and 0.62 for an O-type star), and is shock velocity in units of 1000 km s. The pre-shock velocities corresponding to the cool temperature for the WN and O components of WR 25 were found to be 527 and 721 km s, respectively. These values are about a factor of more than those predicted by the radiative shock model of Lucy (1982). The advanced version of the wind-shock model by Owocki et al. (1988) predicts X-ray emission up to 1 keV. The hydrodynamic shocks, which are expected to occur within unstable stellar winds of massive O-type stars, also show similar pre-shock velocities (Feldmeier et al., 1997). Observationally some apparently single WN stars (e.g. WR 1, WR 6, and WR 110) show intrinsic X-ray emission (Skinner et al., 2002a, b; Ignace et al., 2003) while apparently single WC stars and the WN-subtype WR star WR 40 have not been detected in X-rays thus far (Oskinova, 2005; Gosset et al., 2005).

The temperature corresponding to hot the plasma of WR 25 is intermediate compared to those observed for other similar WR binaries (Zhekov, 2012). Using the mean particle weight for WN stars, the maximum shock temperature on the line of centers for an adiabatic shock corresponds to a pre-shock wind velocity of km s (see Luo et al. 1990), which is % of the observed terminal velocity (Crowther & Dessart, 1998; Niedzielski & Skorzynski, 2002) of WR 25. This could be due to an oblique wind collision over most parts of the shock, which occurs with a lower velocity normal to the shock and thus leads to a lower plasma temperature. The derived values of  and X-ray luminosities exhibit phase-related time variability. We found  to be maximum after periastron passage and minimum when  is maximum during periastron passage. It also appears that  reached a peak value during the eclipse, indicating that the extra absorption could be due to winds from the WN star. The column density along the line of sight through the WR wind to the region of stellar wind collision is (Usov, 1992), where is the distance between the binary components. Using the parameters of WR 25,  was estimated to be cm, which is quite similar to the observed maximum value (see Table 5).

Set Phase EM EM (dof)
x13,x14 0.060 0.62 6.25 5.45 4.82 4.91 9.74 1.18( 598)
x06,x07,x08, 0.141 0.45 6.72 4.03 6.33 3.87 10.20 1.07(1048)
x03,x04,x05 0.360 0.34 5.27 2.87 6.22 2.80 9.03 1.23(2956)
x16 0.438 0.32 5.00 2.97 6.22 2.85 9.08 1.32(965)
x17 0.583 0.29 4.20 2.80 5.84 2.72 8.56 1.37(227)
x18,x19 0.617 0.28 4.47 3.36 6.62 3.22 9.84 1.12( 554)
x01,x02,x20 0.735 0.32 5.89 3.46 7.42 3.35 10.77 1.30(2336)
x11 0.779 0.29 5.42 3.36 7.40 3.29 10.69 1.28(270)
x12 0.988 0.68 12.77 7.73 7.86 7.18 15.04 1.25(281)
s15,s16,s17, 0.009 0.98 13.04 5.34 4.45 5.03 9.49 1.40 (178)
s19,s20,s21, 0.028 1.09 13.37 5.01 3.85 4.72 8.57 1.21 (191)
s22,s23,s24, 0.057 0.99 12.72 4.21 4.11 4.12 8.23 1.15 (103)
s28 0.163 0.52 8.22 3.22 6.13 3.27 9.41 1.17 ( 49)
s37,s38,s39, 0.306 0.32 6.34 2.96 7.76 3.01 10.77 1.12 ( 85)
s29 0.476 0.23 4.90 2.32 7.47 2.38 9.85 1.16 ( 31)
s01 0.703 0.46 7.18 3.13 6.05 3.07 9.12 1.50 ( 20)
s02 0.889 0.40 11.86 4.59 11.48 4.76 16.23 1.11 ( 25)
s03,s04,s05, 0.924 0.40 13.19 5.89 13.14 5.95 19.10 1.05 ( 93)
ss07,s08,s09 0.952 0.54 14.57 6.69 10.82 6.60 17.42 1.38 (423)
s11,s12,s13 0.975 0.56 11.25 7.74 9.04 7.22 16.26 1.14 ( 92)
s14, s31 0.995 0.77 10.73 7.11 5.87 6.47 12.34 1.15 ( 88)

Notes: EM and EM are in units of cm,  is in units of cm, and ,  and   are unabsorbed X-ray luminosities in soft, hard and total bands in units of 10 . Abundances are in units of solar photospheric values. is per degree of freedom and dof is the degree of freedom.

Table 5Best Fit Parameters Obtained From X-ray Spectral Fitting of WR 25.

The EMs corresponding to the cool and hot temperatures change substantially, reflecting the variations in X-ray luminosities in the soft and hard bands. The X-ray luminosities observed for WR 25 were found to be more than that for other close WR+O binaries and single WN stars (Zhekov, 2012; Skinner et al., 2010). X-ray light curves of WR 25 as observed from XMM-Newton  and Swift  show similar behavior. However, in terms of phase coverage, the light curves from Swift are much better than those observed from XMM-Newton. The deficit in X-ray flux just after periastron passage could be due to the eclipse of the wind interaction zone by the wind of the WN star. In all bands, the excess emission is strongest near periastron passage. The X-ray enhancement after the eclipse in the soft band indicates that besides the colliding wind, individual components of the WR 25 system also contribute to the X-ray emission. The soft X-ray flux peaked near phase 0.5, which further supports that both of the components of WR 25 are sources of soft X-rays, enhancing the combined soft X-ray flux when both of the stars are completely visible to the observer. During phase , the soft X-rays further decrease to the minimum value indicating the possibility of a secondary eclipse, when the O-star is in front of the primary WN star. The deeper primary eclipse at phase could be due to the larger opacity of the WN wind. On the other hand,  was found to be at minimum during the phase . Both components of WR 25 are farthest apart at phase 0.5; therefore, the collision is weak, generating fewer X-rays. This indicates that the hard X-rays originate from the collision zone of the wind giving enhanced flux during periastron passage. Stevens et al. (1992) and many other authors showed that strong winds from massive stars collide and generate hard X-rays, in addition to softer X-ray components due to intrinsic or individual components (Berghoefer et al., 1997; Sana et al., 2006). The phase-locked X-ray variability could be a result of changing separation in an eccentric orbit of WR 25. Fig 7 shows that the X-ray luminosity varies as a function of the inverse binary separation [1/(D/a)]. It appears that soft X-ray luminosity does not depend on the binary separation. However, it is clearly seen that the hard X-ray luminosity is linearly dependent on the inverse of the binary separation. The Pearson correlation coefficients for  versus 1/(D/a) and  versus 1/(D/a) were derived to be -0.02 and 0.87 with a probability of no correlation found to be 0.94 and , respectively. This indicates that the hard X-rays in WR 25 are due to the collision of winds. The wide massive binary systems whose X-ray luminosities follow the 1/D relation are Cyg OB2 # 9 (Nazé et al., 2012) and HD 93205 (Antokhin et al., 2003). Cyg OB2 # 8A and WR140 are the massive binary systems which deviate from expectations at periastron, since the collisions then become radiative (De Becker et al., 2006; Pollock, 2012; Corcoran et al., 2011). However, there are a few wide binary systems that do not follow the 1/D variation e.g., WR 11 (Rauw et al., 2000), WR 22 (Gosset et al., 2009). The present X-ray light curves of WR 25 strongly support the idea that during the rise in X-ray emission around periastron passage the X-ray emission is primarily due to colliding winds (Willis et al., 1995; Stevens & Pittard, 1999).

Using V mag, a distance of kpc, a bolometric correction of -4.5 (Hamann et al., 2006) a,nd an anomalous reddening of 4.5 mag (Hamann et al., 2006), the bolometric luminosity of WR 25 is calculated to be . The maximum and minimum values of  for WR 25 are thus calculated to be and , respectively. The derived value of  during periastron passage is similar to that derived by Seward & Chlebowski (1982) using Einstein observations. During the phases and , the possible primary and secondary eclipse positions where the WN and O-type stars contributed more in soft X-rays, respectively, the  are calculated to be and , respectively. These values of  are similar to those obtained for other single WR and O-type stars (Skinner et al., 2010; Sana et al., 2006).

The hard X-rays from WR 25 appear to be a result of collision of winds from binary companions. The gas in the colliding wind regions could either be adiabatic or radiative depending on the cooling parameter () as (Stevens et al., 1992), where is in units of 10 km s, is the distance from the contact to star in units of km and is in units of yr. For , the wind can be assumed to be adiabatic while for , it is roughly isothermal. The cooling parameter is directly proportional to the distance of contact (); as a consequence of this, (Stevens et al., 1992). This means that the shocked region will be adiabatic for longer period binaries. WR 25 is one of the longer orbital period binaries, indicating the shocked region is adiabatic. Furthermore, if we assume that the contact of winds is close to the O-type star of WR 25, the value of is calculated to be , indicating an adiabatic wind in the shocked region. In our analysis, the abundances of O, Ne, Mg, Si, S, Ca, Ar, Fe and Ni were found to be non-solar, which is expected from WR stars because they are in advanced evolutionary stages. The non-solar abundances of these elements could be due to the presence of hot plasma near the surface of the O4 star, due to the collision of strong supersonic winds from the WN6 star with the less powerful supersonic wind from O4 star. For wide binaries, Pittard & Stevens (2002) showed that WR winds dominate the X-ray luminosity. Luo et al. (1990) and Myasnikov & Zhekov (1993) have also found that the shocked WR stellar wind dominates the X-ray emission for WR+O systems in the adiabatic limit. Furthermore, the shocked volume and the emission measure near the O-type star may be dominated by WN winds; therefore, the X-ray emitting plasma also shows non-solar abundances. When we compared the derived values of abundances of different elements for WR 25 with other WN-type stars, we found that the abundances of Ne and S in WR 25 are similar to that of WN-type stars (Smith & Houck, 2005; Ignace et al., 2007)

Figure 7.— X-ray luminosity as a function of inverse of binary separation in soft and hard X-ray bands.

The light curve of WR 25 in the V band shows minimum light near phase 0.75, which could be possible position of eclipse when the WN star is behind the O-star. It appears that the WN star is brighter than the O-type star in the WR 25 system; therefore, may observe an eclipse near phase 0.75. Hamann et al. (2006) also showed that O-type stars contribute a minimum of 15% to the total light of the system in the V band and Gosset et al. (1994) had also noticed the photometric variation in WR 25 with a variability amplitude of 0.02 mag in the Strömgren b-band on a time scale of 250 days. Many other WR binaries also show phase-locked variability in the optical band and the variability is attributed to their binary nature (Lamontagne et al., 1996). The presence of eclipses in X-ray and optical light curves indicates that the orbital plane of WR 25 has a high inclination angle. The minimum of the V-band light curve is consistent with the phase minimum at of the soft X-ray light curve. However, unlike to the soft X-ray light curve we could not see any other minimum in the V-band light curve near phase 0.03, where  was also found to be maximum. This could be due to the poor phase coverage of the V-band light curve.

The analysis of the present data shows that WR 25 is a colliding wind binary with an orbital period of days, where the hard X-rays could be due to the collision zone while soft X-rays could be attributed to individual components.

We thank the referee for his/her comments and suggestions that helped to considerably improve the manuscript. This work uses data obtained by XMM-Newton, an ESA science mission with instruments and contributions directly funded by ESA Member States and the USA (NASA). We acknowledge the Swift data archive and the UK Swift Science Data Center. Facilities: XMM-Newton, Swift.


  • Abbott & Conti (1987) Abbott, D. C., & Conti, P. S. 1987, ARA&A, 25, 113
  • Antokhin et al. (2003) Antokhin, I. I., Rauw, G., Vreux, J.-M., & van der Hucht, K. A. 2003, Magnetic Fields in O, B and A Stars: Origin and Connection to Pulsation, Rotation and Mass Loss, 305, 383
  • Arnaud (1996) Arnaud, K. A. 1996, Astronomical Data Analysis Software and Systems V, 101, 17
  • Berghoefer et al. (1997) Berghoefer, T. W., Schmitt, J. H. M. M., Danner, R., & Cassinelli, J. P. 1997, A&A, 322, 167
  • Bhatt et al. (2010a) Bhatt, H., Pandey, J. C., Kumar, B., Singh, K. P., & Sagar, R. 2010a, MNRAS, 402, 1767
  • Bhatt et al. (2010b) Bhatt, H., Pandey, J. C., Kumar, B., Sagar, R., & Singh, K. P. 2010b, New A, 15, 755
  • Burrows et al. (2005) Burrows, D. N. et al. 2005, Space Sci. Rev., 120, 165
  • Cherepashchuk (1976) Cherepashchuk, A. M. 1976, Pisma v Astronomicheskii Zhurnal, 2, 356
  • Corcoran et al. (2011) Corcoran, M. F., Pollock, A. M. T., Hamaguchi, K., & Russell, C. 2011, arXiv:1101.1422
  • Crowther et al. (1995) Crowther, P. A., Hillier, D. J., & Smith, L. J. 1995, A&A, 293, 403
  • Crowther & Dessart (1998) Crowther, P. A., & Dessart, L. 1998, MNRAS, 296, 622
  • Crowther (2007) Crowther, P. A. 2007, ARA&A, 45, 177
  • De Becker et al. (2004) De Becker, M., Rauw, G., Pittard, J. M., et al. 2004, A&A, 416, 221
  • De Becker et al. (2006) De Becker, M., Rauw, G., Sana, H., et al. 2006, MNRAS, 371, 1280
  • Doom (1987) Doom, C. 1987, Reports on Progress in Physics, 50, 1491
  • Feldmeier et al. (1997) Feldmeier, A., Puls, J., & Pauldrach, A. W. A. 1997, A&A, 322, 878
  • Gamen et al. (2006) Gamen, R., Gosset, E., Morrell, N., et al. 2006, A&A, 460, 777
  • Gorenstein (1975) Gorenstein, P. 1975, ApJ, 198, 95
  • Gosset et al. (1994) Gosset, E., Rauw, G., Manfroid, J., Vreux, J.-M., & Sterken, C. 1994, NATO ASIC Proc. 436: The Impact of Long-Term Monitoring on Variable Star Research: Astrophysics, 101
  • Gosset et al. (2005) Gosset, E., Nazé, Y., Claeskens, J.-F., et al. 2005, A&A, 429, 685
  • Gosset et al. (2009) Gosset, E., Nazé, Y., Sana, H., Rauw, G., & Vreux, J.-M. 2009, A&A, 508, 805
  • Gräfener & Hamann (2005) Gräfener, G., & Hamann, W.-R. 2005, A&A, 432, 633
  • Hamann et al. (2006) Hamann, W.-R., Gräfener, G., & Liermann, A. 2006, A&A, 457, 1015
  • Ignace et al. (2003) Ignace, R., Oskinova, L. M., & Brown, J. C. 2003, A&A, 408, 353
  • Ignace et al. (2007) Ignace, R., Cassinelli, J. P., Tracy, G., Churchwell, E., & Lamers, H. J. G. L. M. 2007, ApJ, 669, 600
  • Jansen et al. (2001) Jansen, F., Lumb, D., Altieri, B., et al. 2001, A&A, 365, L1
  • Lamontagne et al. (1996) Lamontagne, R., Moffat, A. F. J., Drissen, L., Robert, C., Matthews, J. M.  1996, AJ, 112, 2227
  • Lomb (1976) Lomb, N. R. 1976, Ap&SS, 39, 447
  • Lucy & White (1980) Lucy, L. B., & White, R. L. 1980, ApJ, 241, 300
  • Lucy (1982) Lucy, L. B. 1982, ApJ, 255, 286
  • Luo et al. (1990) Luo, D., McCray, R., & Mac Low, M.-M. 1990, ApJ, 362, 267
  • Meynet & Maeder (2005) Meynet, G., & Maeder, A. 2005, A&A, 429, 581
  • Morrison & McCammon (1983) Morrison, R., & McCammon, D. 1983, ApJ, 270, 119
  • Myasnikov & Zhekov (1993) Myasnikov, A. V., & Zhekov, S. A. 1993, MNRAS, 260, 221
  • Nazé (2009) Nazé, Y. 2009, A&A, 506, 1055
  • Nazé et al. (2012) Nazé, Y., Mahy, L., Damerdji, Y., et al. 2012, A&A, 546, A37
  • Niedzielski & Skorzynski (2002) Niedzielski, A., & Skorzynski, W. 2002, Acta Astron., 52, 81
  • Owocki et al. (1988) Owocki, S. P., Castor, J. I., & Rybicki, G. B. 1988, ApJ, 335, 914
  • Oskinova (2005) Oskinova, L. M. 2005, MNRAS, 361, 679
  • Pittard & Stevens (2002) Pittard, J. M., & Stevens, I. R. 2002, A&A, 388, L20
  • Plucinsky et al. (2012) Plucinsky, P. P., Beardmore, A. P., DePasquale, J. M., et al. 2012, Proc. SPIE, 8443
  • Pojmanski (2002) Pojmanski, G. 2002, Acta Astron., 52, 397
  • Pollock (1987) Pollock, A. M. T. 1987, ApJ, 320, 283
  • Pollock & Corcoran (2006) Pollock, A. M. T., & Corcoran, M. F. 2006, A&A, 445, 1093
  • Pollock (2012) Pollock, A. M. T. 2012, Proceedings of a Scientific Meeting in Honor of Anthony F. J. Moffat, 465, 308
  • Prilutskii & Usov (1976) Prilutskii, O. F., & Usov, V. V. 1976, Soviet Ast., 20, 2
  • Raassen et al. (2003) Raassen, A. J. J., van der Hucht, K. A., Mewe, R., et al. 2003, A&A, 402, 653
  • Rauw et al. (2000) Rauw, G., Stevens, I. R., Pittard, J. M., & Corcoran, M. F. 2000, MNRAS, 316, 129
  • Rauw et al. (2002) Rauw, G., Vreux, J.-M., Stevens, I. R., et al. 2002, A&A, 388, 552
  • Sana et al. (2004) Sana, H., Stevens, I. R., Gosset, E., Rauw, G., & Vreux, J.-M. 2004, MNRAS, 350, 809
  • Sana et al. (2006) Sana, H., Rauw, G., Nazé, Y., Gosset, E., & Vreux, J.-M. 2006, MNRAS, 372, 661
  • Scargle (1982) Scargle, J. D. 1982, ApJ, 263, 835
  • Seward & Chlebowski (1982) Seward, F. D., & Chlebowski, T. 1982, ApJ, 256, 530
  • Skinner et al. (2002a) Skinner, S. L., Zhekov, S. A., Güdel, M., & Schmutz, W. 2002a, ApJ, 579, 764
  • Skinner et al. (2002b) Skinner, S. L., Zhekov, S. A., Güdel, M., & Schmutz, W. 2002b, ApJ, 572, 477
  • Skinner et al. (2007) Skinner, S. L., Zhekov, S. A., Güdel, M., & Schmutz, W. 2007, MNRAS, 378, 1491
  • Skinner et al. (2010) Skinner, S. L., Zhekov, S. A., Güdel, M., Schmutz, W., & Sokal, K. R. 2010, AJ, 139, 825
  • Smartt et al. (2009) Smartt, S. J., Eldridge, J. J., Crockett, R. M., & Maund, J. R. 2009, MNRAS, 395, 1409
  • Smartt (2009) Smartt, S. J., 2009, ARA&A, 47, 63
  • Smith et al. (2001) Smith, R. K., Brickhouse, N. S., Liedahl, D. A., Raymond, J. C. 2001, ApJ, 556, 91
  • Smith & Houck (2005) Smith, J.-D. T., & Houck, J. R. 2005, ApJ, 622, 1044
  • Stevens et al. (1992) Stevens, I. R., Blondin, J. M., & Pollock, A. M. T. 1992, ApJ, 386, 265
  • Stevens & Pittard (1999) Stevens, I. R., & Pittard, J. M. 1999, Wolf-Rayet Phenomena in Massive Stars and Starburst Galaxies, 193, 289
  • Strüder et al. (2001) Strüder, L., Aschenbach, B., Bräuninger, H., et al. 2001, A&A, 375, L5
  • Turner et al. (2001) Turner, M. J. L., Abbey, A., Arnaud, M., et al. 2001, A&A, 365, L27
  • Usov (1992) Usov, V. V. 1992, ApJ, 389, 635
  • van der Hucht (1992) van der Hucht, K. A. 1992, A&A Rev., 4, 123
  • van der Hucht (2001) van der Hucht, K. A. 2001, New A Rev., 45, 135
  • Willis et al. (1995) Willis, A. J., Schild, H., & Stevens, I. R. 1995, A&A, 298, 549
  • Zhekov (2012) Zhekov, S. A. 2012, MNRAS, 422, 1332
Comments 0
Request Comment
You are adding the first comment!
How to quickly get a good reply:
  • Give credit where it’s due by listing out the positive aspects of a paper before getting into which changes should be made.
  • Be specific in your critique, and provide supporting evidence with appropriate references to substantiate general statements.
  • Your comment should inspire ideas to flow and help the author improves the paper.

The better we are at sharing our knowledge with each other, the faster we move forward.
The feedback must be of minimum 40 characters and the title a minimum of 5 characters
Add comment
Loading ...
This is a comment super asjknd jkasnjk adsnkj
The feedback must be of minumum 40 characters
The feedback must be of minumum 40 characters

You are asking your first question!
How to quickly get a good answer:
  • Keep your question short and to the point
  • Check for grammar or spelling errors.
  • Phrase it like a question
Test description