X-ray emission from O-type stars : DH Cep and HD 97434

X-ray emission from O-type stars : DH Cep and HD 97434


We present X-ray emission characteristics of the massive O-type stars DH Cep and HD 97434 using archival XMM-Newton observations. There is no convincing evidence for short term variability in the X-ray intensity during the observations. However, the analysis of their spectra reveals X-ray structure being consistent with two-temperature plasma model. The hydrogen column densities derived from X-ray spectra of DH Cep and HD 97434 are in agreement with the reddening measurements for their corresponding host clusters NGC 7380 and Trumpler 18, indicating that the absorption by stellar wind is negligible. The X-ray emission from these hot stars is interpreted in terms of the standard instability-driven wind shock model.

star:X-ray – star:binary – star:wind – star:individual (DH Cep and HD 97434)

1 Introduction

Massive, hot and luminous O-type stars have powerful stellar winds, with mass loss rates of the order of 10 to 10 M yr and velocities of several thousand km s. These winds play a crucial role both in the stellar evolution as well as the galactic evolution. The existence of strong winds is supposed to be a primary reason for X-ray emission in the early type stars. Study of the X-ray emission from O-type stars is important for understanding the high energy processes that generate hot gas in the wind or in the atmosphere of hot stars. X-ray emission from O-type stars was first discovered serendipitously with the EINSTEIN observatory (Harnden et al. 1979). Most O-type stars are reasonably bright ( ) and have a soft thermal spectrum with kT 1 keV. The X-ray luminosity of these sources is found to scale with their bolometric luminosity as to 10 (Long & White 1980, Pallavicini et al. 1981, Singh & Naranan 1982, Berghfer et al. 1997; Sana et al. 2006). It is generally thought to be produced as a result of shocks, with velocity jumps up to a few hundred km s , generated throughout the stellar wind because of dynamic instabilities (Lucy & White 1980; Owocki & Cohen 1999; Kudritzki & Puls 2000). This scenario is now called a standard model for X-ray emission from early-type stars (Rauw 2008). Although, most of the X-ray emission is dominated by soft X-rays as expected from wind-shock model (e.g. Feldmeier, Puls & Pauldrach 1997), a significant and an unexpected harder component is also seen in the X-ray spectra (e.g. Leutenegger & Khan 2003; Bross et al. 2007; Stelzer et al. 2005). High quality observational data from XMM-Newton and Chandra observatories are indeed crucial to provide a strong test of the latest models. In order to understand the X-ray emission processes from O-type stars, we have analyzed XMM-Newton data for two O-type stars DH Cep and HD 97434. The results of our analysis are presented here.

DH Cep is an O6 V + O7 V spectroscopic binary system and a member of the cluster NGC 7380 (distance = 3.7 kpc, age = 10 Myr; Underhill 1969; Massey, Johnson & De Gioio-Eastwood 1995; Hilditch, Harries, & Bell 1996). It has an orbital period of 2.11 d (Penny, Gies & Bagnuolo 1997). A low signal, Einstein X-ray observation is reported by Chlebowski, Harden & Sciortino (1989) with in the energy band 0.2–3.5 keV.

HD 97434 is an O9 V star (Skiff 2007), with a membership probability of 78% (Dias et al. 2006) in the cluster Trumpler 18 (distance = 1.5 kpc; age = 15 Myr; Vzquez & Feinstein 1990). The Einstein X-ray data for the source are cataloged by Chlebowski, Harden & Sciortino (1989) with (corrected for the distance of 1.5 kpc) in the energy band 0.2–3.5 keV.

2 XMM-Newton Observations and Data Reduction

We present analysis of the archival data obtained with the XMM-Newton observatory. Table 1 summarizes the XMM-Newton observations. Our analysis is based on CCD images, lightcurves and spectra from the European Photon Imaging Cameras (EPIC). Data were acquired simultaneously with EPIC-PN camera (Strder et al. 2001) and two nearly identical EPIC-MOS (MOS1 and MOS2;Turner et al. 2001) cameras. Data reduction followed standard procedures using XMM-Newton Science Analysis System software (SAS version 8.0.0) with updated calibration files (Ehle et al. 2004). A detailed description of the data reduction and analysis procedure is given in Bhatt et al. (2010). Owing to poor counts below 0.3 keV and above 7.5 keV, our analysis is limited to the energy band 0.3–7.5 keV. For the star HD 97434, the observations are not made with the MOS1 detector, therefore we used only the data from PN and MOS2 detectors for further analysis. For the star DH Cep, the observations have been taken during the binary phase of 0.08–0.25. The binary phases of DH Cep were determined using the ephemeris HJD = 2441905.805+2.110932E (Hilditch, Harries, & Bell 1996; Penny, Gies & Bagnuolo 1997).

Source photons were extracted from circular region of 20 around the source position to generate the light curves and spectra of the source DH Cep and HD 97434. Background was estimated from a number of empty regions close to the X-ray source in the detector. X-ray spectra of the sources were generated using SAS task especget, which also computes the photon redistribution as well as the ancillary matrix. Finally, the spectra were re-binned to have at least 20 counts per spectral bin for both the sources.

3 Analysis

Figure 1 shows the background and the background subtracted X-ray lightcurves of the stars DH Cep and HD 97434. We have binned the event lists into 1000 and 1500s time intervals for the stars DH Cep and HD 97434, respectively. We have performed test to measure the significance of the deviations from the mean count rates. The probability of the variability () to quantify the constancy of the data over the time scale of observations was calculated. The results are displayed in the respective panels of Figure 1. The value of was calculated to be % for the star DH Cep. However, it was calculated to be % for the star HD 97434. This indicates that the light curve of the star DH Cep shows significant variability. We further analysed the light curves of DH Cep and found that the observed variability may arise due to a few outlier points. If we neglect the four outlier points of the light curve of DH Cep, the probability of variability decreases to 80%. Therefore, we have plotted the limit in Figure 1 represented by a shaded region around the constant count rate, i.e., mean value of count rate for the data sample. Here, represents the standard deviation in count rates. This analysis shows that none of the data points in any of the light curves were lying outside of limit. Further, the variability trend seen in the PN data was not correlated with the MOS data and all the data points in the light curves were within limits. Therefore, it appears that the significance of the variability in the light curves is very poor and cannot be detected by the present data set.

X-ray spectra of DH Cep and HD 97434 are shown in Figure 2. Emission lines of Fe XVII (0.8 keV), Ne X (1.02 keV), Mg XII (1.47) and Si XIII (1.853 keV) are seen in their respective spectra. X-ray emission from massive stars is thought to be produced from thermal emission, generated from shock-heated plasma. These hydrodynamic shocks may occur in the stellar winds from massive stars. However, in massive binaries they may occur in wind-wind collision zone (Bhatt et al. 2010 and references therein). Therefore, we have modeled the spectra using (a) plane-parallel shock model (pshock; Borkowski, Lyerly & Reynolds 2001), and (b) models of Astrophysical Plasma Emission Code (apec; Smith et al. 2001), as implemented in the XSPEC version 12.3.0. A – minimization gave the best fitted model to the data. We corrected for the local absorption in the line-of-sight to the source () using the photoelectric absorption cross sections according to Baluciska-Church & McCammon (1992) and modeled it as phabs (photoelectric absorption screens). The abundances in the absorber were allowed to vary during the fitting procedure The abundances were varied with respect to solar abundances adopted from Lodders (2003).

First, we fitted pshock plasma model to derive their spectral features. The constant temperature pshock plasma model was considered without incorporating mass-loss and orbital parameters of massive stars. However, the model does account for non-equilibrium ionization effects and assumes an equal electron and ion temperature. The best fit pshock models to the data are shown in left and right top panels of Figure 2 and the best fit parameters are given in Table 2. The best-fit X-ray temperatures deduced from X-ray spectra are found to be keV and keV for DH Cep and HD 97434, respectively.

Second, the simplest spectral apec model was used for the spectral fitting. Here, we first considered the case of an isothermal hot ionized gas as the one temperature (1T) apec model. This model is expressed as phabs apec. The 1T apec models with either solar or subsolar photospheric abundances were found to be unacceptable for both the stars due to high value of . Therefore, for the next level of complexity, we used models with two temperatures (2T) expressed as phabs(apec+apec). In 2T models the first component represents the ”cooler” ionized gas and the second component represents the ”hotter” ionized gas. The 2T models were found to be acceptable but required subsolar abundances, i.e., . The subsolar abundances have also been reported for 15 other massive stars by Zhekov & Palla (2007). The best fit 2T models are displayed as histograms along with the data in the left and the right bottom panels of Figure 2 for the stars DH Cep and HD 97434, respectively. The best-fit parameters for these models were obtained by -minimization technique and are given in Table 2 with their corresponding . We noted that 2T apec model fitted these spectra better than pshock model with a smaller value of (see Figure 2 and Table 2). The cooler and hotter X-ray temperatures deduced from X-ray spectra are found to be respectively keV and keV for DH Cep and keV and keV for HD 97434, respectively. The best fit values of for DH Cep and HD 97434 are in agreement with the values of derived from optical data for NGC 7380 ( ; Massey, Johnson & De Gioio-Eastwood 1995), and Trumpler 18 ( ; Vzquez & Feinstein 1990), respectively. The from optical data was estimated using the relation, (Vuong et al. 2003), where , assuming a normal interstellar reddening law towards the direction of the cluster. The extra absorbing hydrogen column density in addition to the interstellar contribution was not needed in the spectral fits implying that the absorption by the stellar wind is negligible (see also Zhekov & Palla 2007; Sana et al. 2006).

4 Discussion

4.1 X-ray Variability

The wind-shock model predicts a significant short-term variability in the X-ray flux as the shocks fade and grow on short time scales. However, it has not been observed so far. We have also explored the possibility of short-term X-ray variability of DH Cep and HD 97434 over the observation time span (40 ks). The timing analysis reveals a lack of short term variability in the X-ray intensity for both the massive stars. The lack of short term variability could be due to the wind fragmentation, so that individual X-ray fluctuations are smoothed out over the whole emitting volume, leading to a rather constant output ( Feldmeier, Puls & Pauldrach 1997). Several other massive stars, e.g., HD 159176 (De Becker et al. 2004) and HD 47129 (Linder et al. 2006) have also shown lack of variability in their light curves.

The longer-term variability can not be probed using present data. The long-term variability (few days) from O-type stars was reported by Snow, Cash & Grady (1981) for the first time, but their results were based on the low sensitivity detectors available at that time. No significant variability was detected by the ROSAT All-Sky Survey in the X-ray emission of 57 OB stars over a time-scale of 2 d (Berghfer & Schmitt 1995).

4.2 X-ray temperatures of plasma

Our analysis of XMM-Newton data reveals reveal two-component X-ray temperature structure for massive stars. The temperatures corresponding to the cool and the hot components of HD 97434, and the cool component of DH Cep are found to be less than 1 keV. However, the temperature corresponding to the hot component of DH Cep was found to be more than 1.89 keV. The derived values of temperatures are similar to the O-type stars in the NGC 6231 cluster (Sana et al. 2006) and in the Carina OB1 association (Antokhin et al. 2008). Generally for O-type stars, the best fits to good quality data can usually be achieved by the sum of two thermal components at about 0.3 and 0.7–1 keV (Gdel & Naz 2009). The observed soft X-ray emission could originate from radiation-driven instabilities in stellar winds (Lucy 1982). The wind-shock model predicts the intrinsic instability of the line driving mechanism. Indeed, the velocity in an unstable wind is not the same everywhere and the fast-moving parcels of material will overcome the slow moving material, generating shocks between them. This model estimates the shock velocities (v) by the relation (Lucy 1982, Luo, McCray & Maclow 1990)


Adopting the mean particle weight 0.62 for O-type stars (Cassinelli et al. 2008) we derive the ”average” value of shock velocities correspond to the cool components (kT) and for the stars DH Cep and HD 97434, respectively. However, the shock velocities correspond to the hot components (kT) are found to be and for DH Cep and HD 97434, respectively. These values are about a factor of 2 larger than the predicted by radiative shock model of Lucy (1982). However, the evolved version of the standard model by Owocki, Castor & Rybicki (1988) predicts X-ray emission upto 1 keV from wind shocks model. The temperature corresponding to the hot component of DH Cep is found to be more than 1.89 keV. It may arise from wind collision zone of the binary system (see Bhatt et al. 2010 and references therein), but we are not able to draw any firm conclusion on the basis of it alone. Therefore, it appears that the cool as well as hot temperature components from DH Cep and HD 97434 could be generated by instabilities in radiation-driven wind shocks.

4.3 X-ray Luminosity

For massive stars DH Cep and HD 97434, the ratios of X-ray to bolometric luminosities, ), are found to be -6.7 and -7.3 (see Table 2 ) in the energy band 0.3-7.5 keV, respectively, which are broadly consistent with the relation derived for similar kind of O-type stars (Sana et al. 2006).

X-ray luminosities derived from XMM-Newton data are found to be an order of magnitude lower than the derived from Einstein observations for both the massive stars (see Table 2 and §1). These discrepancies can not be explained by difference in sensitivity ranges of the various instruments. A similar results have been found by De Becker et al. (2004) for the star HD 159176 after the comparison of XMM-Newton, Einstein and ROSAT data. We found the X-ray luminosity and in energy range 0.3–2.0 keV for the DH Cep and HD 97343, respectively, using XMM-Newton data. The choice of the energy range is nearly a similar to the energy range in Einstein observations, i.e., 0.2–3.5 keV. We further converted the Einstein IPC count rates into flux using the WebPIMMS2, where we assumed a thermal plasma model with temperature of 0.5 keV, and the hydrogen column density of for DH Cep and for HD 97434 (see Table 2). The Einstein IPC count rates are 0.0166 and counts s for the stars DH Cep and HD 97343, respectively (Chlebowski et al. 1989). The X-ray luminosity thus calulated was and for the stars DH Cep and HD 97434, respectively. These value are consistent with that obtained from XMM-Newton obsertions. Therefore, it appears that the model used by Chlebowski et al. (1989) to convert the Einstein count rates into luminosities may be responsible for these discrepancies.

5 Summary and Conclusions

Using the archival XMM-Newton observations, we present, for the first time, the temporal and spectral analysis of X-ray emission from O-type stars DH Cep and HD 97434. The main results are as follows:

  1. There is no firm evidence for short term variability in the X-ray intensity during the observations time span ( 40 ks) for both the stars.

  2. The values of derived from the best fit X-ray spectra are consistent with the optical estimates of reddening.

  3. X-ray spectra of these stars are fitted consistently with two-temperature plasma models. The X-ray emitting plasma is found to be generated at a temperature of lesser than 1 keV. It could originate from small shocks in the radiation-driven outflows. A hotter component is indicated in DH Cep.

  4. The best-fit values of abundances are found to be 0.2 times of the solar abundances.

Object Name DH Cep HD 97434
Observation ID 0205650101 0051550101
Exposure Time (s) 31413 40822
Start time UT (hh:mm:ss) 19 Dec 2003 02:02:12 06 Feb 2002 01:13:19
Usable time(ks) (MOS1) 25.23
Usable time(ks) (MOS2) 25.46 39.20
Usable time(ks) (PN) 19.52 35.85
EPIC filter Thick Medium
Offset from Target (arcmin) 0.000 3.68

: The observations have not done in Prime Full window mode for MOS1 detector.

Table 1: Journal of XMM-Newton Observations of the objects.
Parameters DH Cep HD 97434
pshock 2T apec pshock 2T apec
1.32(181) 1.18(180) 1.36(75) 1.31(74)
0.33 0.34 0.08 0.08
log 32.73 32.75 31.31 31.32
-6.73 -6.71 -7.29 -7.29
-6.77 -6.77 -7.31 -7.31
-7.76 -7.55 -8.56 -8.48
Table 2: The best-fit spectral parameters of the stars DH Cep and HD 97434.
Figure 1: X-ray light curves for the stars DH Cep (left panel) and HD 97434 (right panel). The background subtracted lightcurves are shown by top set of circles and the background in the same XMM-Newton/EPIC field of view represented by bottom set of circles in each panel.

Figure 2: Best-fit X-ray spectra of massive stars. left panel: for DH Cep. right panel : for HD 97434. The solid line histograms represent the best fit pshock (in upper panels) and 2T apec model (in lower panels) with absorption model PHABS for PN, MOS1 and MOS2 data.


  1. journal: New Astronomy
  2. http://heasarc.gsfc.nasa.gov/Tools/w3pimms.html


  1. Antokhin I. I., Rauw G., Vreux J-M, van der Hucht K. A., Brown J. C., 2008, A&A, 477, 593
  2. Baluciska-Church M., McCammon D., 1992, ApJ, 400, 699
  3. Berghfer T. W., Schmitt J. H. M. M., 1995, AdSpR, 16, 163
  4. Berghfer T. W., Schmitt J. H. M. M., Danner, R., Cassinelli J. P., 1997, A&A, 322, 167
  5. Bhatt H., Pandey J. C., Kumar B., Singh K. P., Sagar, R., 2010, MNRAS, 402, 1767
  6. Borkowski K. J., Lyerly W. J., Reynolds S. P., 2001, ApJ, 548, 820
  7. Broos P. S., Feigelson E. D., Townsley L. K., Getman K. V., Wang J., Garmire G. P., Jiang Z., Tsuboi Y., 2007, ApJ, 169, 353
  8. Cassinelli J. P., Ignace R., Waldron W. L., Cho J., Murphy N. A., Lazarian A., 2008, ApJ, 683, 1052
  9. Chlebowski T., Harnden F. R. Jr., Sciortino S., 1989, ApJ, 341, 427
  10. De Becker M., Rauw G., Pittard J. M., Antokhin I. I., Stevens I. R., Gosset E., Owocki S. P., 2004, A&A, 416, 221
  11. Dias W. S., Assafin M., Flrio V., Alessi B. S., Lbero V., 2006, A&A, 446, 949
  12. Ehle M. et al., 2004, User’s Guide to XMM-Newton Science Analysis System.
  13. Feldmeier A., Puls J., Pauldrach A. W. A., 1997, A&A, 322, 878
  14. Gdel M., Naz Y., 2009, A&ARv, 17, 309
  15. Harnden F. R. Jr. et al. 1979, ApJ, 234, L51
  16. Hilditch R. W., Harries T. J., Bell S. A., 1996, A&A, 314, 165
  17. Kudritzki R -P, Puls J., 2000, ARA&A, 38, 613
  18. Leutenegger M. A., Kahn S. M., 2003, ApJ, 585, 1015
  19. Linder N., Rauw G., Pollock A. M. T., Stevens I. R., 2006, 370, 1623
  20. Lodders K., 2003, ApJ, 591, 1220
  21. Long K. S., White R. L., 1980, ApJ, 239, 65L
  22. Lucy L. B., White R. L., 1980, ApJ, 241, 300
  23. Lucy L. B., 1982, ApJ, 255, 286
  24. Luo D., McCray R., Mac Low M., 1990, ApJ, 362, 267
  25. Massey P., Johnson K. E., Degioia-Eastwood, K., 1995, ApJ, 454, 151
  26. Owocki S. P., Castor J. I., Rybicki G. B., 1988, ApJ, 335, 914O
  27. Owocki S. P., Cohen D. H., 1999, ApJ, 520, 833
  28. Pallavicini R., Golub L., Rosner R., Vaiana G. S., Ayres T., Linsky J. L., 1981, ApJ, 248, 279
  29. Penny L. R., Gies D. R., Bagnuolo W. G. Jr., 1997, ApJ, 483, 439
  30. Rauw G., 2008, in Benaglia P., Bosch G. L., Cappa C. E., eds, Massive Stars: Fundamental Parameters and Circumstellar Interactions, 33, p.59
  31. Sana H., Rauw G., Naz Y., Gosset E., Vreux J.-M., 2006, MNRAS, 372, 661
  32. Singh K. P., Naranan S., 1982, A&A, 113, 167
  33. Skiff B. A., 2007, yCat, 102023S, Available at VizieR On-line Data Catalog http://webviz.u-strasbg.fr/viz-bin/VizieR
  34. Smith R. K., Brickhouse N. S., Liedahl D. A., Raymond J. C., 2001, ApJ, 595, 365
  35. Snow T. P. Jr., Cash W., Grady C. A., 1981, ApJ, 2446, 19
  36. Stelzer B., Flaccomio E., Montmerle T., Micela G., Sciortino S., Favata F., Preibisch T., Feigelson E. D., 2005, ApJ, 160, 557
  37. Strder L. et al. 2001, A&A, 365, 18
  38. Turner M. J. L. et al. 2001, A&A, 365, 27
  39. Underhill A. B., 1969, A&A, 1, 356
  40. Vzquez R. A. & Feinstein A., 1990, A&AS, 86, 209
  41. Vuong M. H., Montmerle T., Grosso N., Feigelson E. D., Verstraete L., Ozawa H., 2003, A&A, 408, 581
  42. Zhekov S.A., Palla F., 2007, MNRAS, 392, 1124
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