X-ray Line Emissions of \thetatwo

Detailed X-Ray Line Properties of \thetatwo in Quiescence

Arik W. Mitschang1 , Norbert S. Schulz2 , David P. Huenemoerder2 , Joy S. Nichols1 , Paola Testa1
1affiliation: Smithsonian Astrophysical Observatory (SAO), Cambridge, MA
2affiliation: MIT Kavli Institute for Astrophysics and Space Research, Cambridge, MA
Abstract

We investigate X-ray emission properties of the peculiar X-ray source \thetatwo in the Orion trapezium region using more than 500 ksec of HETGS spectral data in the quiescent state. The amount of exposure provides tight constraints on several important diagnostics involving O, Ne, Mg, and Si line flux ratios from He-like ion triplets, resonance line ratios of the H- and He-like lines and line widths. Accounting for the influence of the strong UV radiation field of the O9.7V star we can now place the He-like line origin well within two stellar radii of the O-star’s surface. The lines are resolved with average line widths of 38 km s confirming a line origin relatively close to the stellar surface. In the framework of standard wind models this implies a rather weak, low opacity wind restricting wind shocks to temperatures not much larger than 2 K. The emission measure distribution of the X-ray spectrum, as reported previously, includes very high temperature components which are not easily explained in this framework. The X-ray properties are also not consistent with coronal emissions from an unseen low-mass companion nor with typical signatures from colliding wind interactions. The properties are more consistent with X-ray signatures observed in the massive Trapezium star Ori C which has recently been successfully modeled with a magnetically confined wind model.

stars: magnetic fields – stars: winds, outflows – X-rays: stars – stars: individual (\thetatwo)

1 Introduction

Ori A is a triple star system at the heart of the Orion Nebula Cluster (ONC), Its massive primary has been identified as a 5th magnitude O9.5 V star (Abt et al., 1991) with a mass of 25 M (Preibisch et al., 1999), making it the second most massive star in the ONC next to the 45 M O5.5 V star of Ori C. A more recent photometric study provides an optical identification of O9 V and a total system mass of 39 M (Simón-Díaz et al., 2006). The studies of Abt et al. (1991) and Preibisch et al. (1999) show that this system includes two close intermediate mass companions at 174 AU and 0.47 AU separation with mass estimates between 7 and 9 M for each.

Ori A has been extensively monitored in X-rays with The Chandra X-ray Observatory and has shown its fair share of odd behavior. Observations in 2000 found that the X-ray source exhibited unusual and dramatic variability with a 50 flux drop in less than 12 hours accompanied by multiple small flares with only a few hours durations (Feigelson et al., 2002). Such behavior in an early type stellar system is surprising since this can not be explained by the standard wind shock models for X-rays in early type stars (Lucy, 1982; Owocki et al., 1988), nor by the magnetically confined wind models (MCWMs; Babel & Montmerle (1997)). While the MCWM can produce hard X-ray emission like observed in Ori C (Schulz et al., 2000, 2003; Gagné et al., 2005) and Sco (Cohen et al., 2003)., it does not explain the observed variability in \thetatwo. At the time, the suggestion was made that such emission could be the result of magnetic reconnection events. To add to this excitement, a specifically powerful X-ray flare from \thetatwo, seen with the Chandra High Energy Transmission Grating Spectrometer (HETGS), surprised observers in 2004 (Schulz et al., 2006) and produced a total power output exceeding 10 ergs s. Considering the orbital phase of the close spectroscopic companion, the low He-like forbidden/intercombination line ratios, and the fact that all lines remained unresolved led to the argument that these events are triggered by magnetic interactions with the close companion. A sub-pixel re-analysis of a similar flare event which appeared during observations in the Chandra Orion Ultradeep Project (COUP) (Stelzer et al., 2005; Schulz et al., 2006) in 2003, however, seem to indicate that these events may originate from the companion instead (M. Gagne, priv. comm.). An unseen T Tauri companion appears unlikely due to the observed peculiar line properties.

In contrast to that observed in the elevated states, the quiescent spectrum of \thetatwo exhibits temperatures above 25 MK and has line ratios which suggest that the X-ray emitting plasma is close enough to the stellar surface of the massive star to argue for some form of magnetic confinement (Schulz et al., 2006). The argument is strengthened by the fact that the line widths, quite in contrast to the narrow line widths observed during the outbursts, seem broadened to the order of 300 km s. These properties are very reminiscent of the MCWM results obtained in Ori C (Gagné et al., 2005), where, through detailed simulations, it was demonstrated that the bulk of the emitting plasma is close to the photosphere, or within 2 R, and line widths are 400 km s. However, in spite of these apparent differences, the properties of the quiescent state remained fairly unconstrained with respect to precise line ratios and widths. Ori A’s X-ray luminosity is about an order of magnitude lower than than that observed during outburst and the study remained statistically limited.

The Chandra Data Archive (CDA)111http://cxc.harvard.edu/cda/ now contains an additional 300 ks on Ori A between 2004 and 2008 and in this paper we present a full analysis of the quiescent spectrum allowing us to derive much better constrained line properties. The results are also used to test the hypothesis that the X-ray emission from Ori A is consistent with predictions from the MCWM. The paper is structured as follows; in Section 2 we discuss the observations and analysis methods, in Section 3 we discuss the results of our emission line measurements, and finally we summarize our findings in Section 4

2 Observations and Analysis

We have retrieved Chandra HETGS data in the vicinity of the ONC, which were originally observed as a part of the HETG Orion Legacy Project (Schulz et al., 2008), from the CDA. There are now seventeen separate Chandra observations which include \thetatwo within an off-axis angle suitable for extraction. See Table 1 for a list of the included observations and selected properties. Noting that this study is focused on the quiescent state spectrum and that ObsID 4474 was not included in any analysis in the current study due to the substantially elevated count rate during its entire exposure, we have accumulated 520 ks of exposure time on \thetatwo in the quiescent state. Figure 1 shows the total combined counts spectrum using the 520 ks on \thetatwo.

Figure 1: Counts spectrum from the total combined data on \thetatwo (MEG+HEG).
Sequence ObsID Start Date Start Time Exposure Offsetaa\thetatwo zeroth order position offset from nominal pointing Phase RangebbAssuming a 20.974 day period and periastron passage at HJD=2440581.27 (Abt et al., 1991)
Number (UT) (UT) () arcmin
200001 3 1999-10-31 05:58:56 49.6 2.42 0.76-0.79
200002 4 1999-11-24 05:39:24 30.9 2.28 0.92-0.99
200175 2567 2001-12-28 12:25:56 46.4 1.98 0.99-1.01
200176 2568 2002-02-19 20:29:42 46.3 2.10 0.53-0.55
200242 4473 2004-11-03 01:48:04 49.1 1.26 1.00-1.03
200243 4474cc4474 is included here only for reference, no analysis herein utilized it due to the extremely elevated count rate during its entirety 2004-11-23 07:48:38 50.8 1.39 0.96-0.99
200420 7407 2006-12-03 19:07:48 24.6 1.64 0.27-0.29
200423 7410 2006-12-06 12:11:37 13.1 3.02 0.40-0.41
200421 7408 2006-12-19 14:17:30 25.0 2.08 0.02-0.04
200422 7409 2006-12-23 00:47:40 27.1 2.30 0.19-0.21
200424 7411 2007-07-27 20:41:22 24.6 3.94 0.53-0.54
200425 7412 2007-07-28 06:16:09 25.2 4.39 0.55-0.56
200462 8568 2007-08-06 06:54:08 36.1 2.53 0.98-1.00
200462 8589 2007-08-08 21:30:35 50.7 2.53 0.10-0.13
200478 8897 2007-11-15 10:03:16 23.7 3.37 0.80-0.81
200477 8896 2007-11-30 21:58:33 22.7 2.34 0.54-0.55
200476 8895 2007-12-07 03:14:07 25.0 1.74 0.84-0.85
Table 1: Observation Log

As noted, none of these observations were targeted at \thetatwo; indeed no Chandra gratings observations have ever targeted \thetatwo. However using the suite of advanced extraction tools provided by the Chandra Transmissions Grating Catalog and Archive (TGCat; Huenemoerder et al. (2010); Mitschang et al. (2010))222http://tgcat.mit.edu, extraction of the dispersed counts of off-axis X-ray source positions proved to be trivial.

Grating spectra were extracted and responses computed using TGCat software to locate the optimal centroid position of \thetatwo and apply proper calibration for each observation. In a crowded field such as the Orion Trapezium, careful attention must be made during analysis to contamination from other zeroth order counts lying close to or on top of dispersion counts and dispersion arms crossing one another at critical points. To this end, we reviewed order sorting images (ACIS CCD event energy vs. gratings order wavelength, or specifically FITS-file columns vs. ) for each observation and identified potential contamination. In this view, the source traces two hyperbolas centered on (e.g. see Chandra POG333http://cxc.harvard.edu/proposer/POG/ Fig 8.13); traces from confusing sources show as offset hyperbolas (dispersed) or vertical lines (zeroth order). We found no significant source of contamination in the regions used for line fitting; See Table 2 for details on the locations of these regions. Similarly when fitting the continuum we used a set of wavelength ranges containing few lines, the “line free regions”, in which we found little contamination. See Section 2.1 for a more detailed discussion on the continuum modeling and Section 2.2 on line fitting. Line width analysis is treated separately in Section 2.3.

All fitting of data was done using the Interactive Spectral Interpretation System (ISIS; Houck & Denicola (2000))444http://space.mit.edu/CXC/isis, along with the Astrophysical Plasma Emission Database (APED; Smith et al. (2001))555http://cxc.harvard.edu/atomdb/sources_aped.html for line emissivities and continuum modeling.

2.1 Continuum

The continuum emission of \thetatwo was modeled by fitting a single temperature APED model to the combined MEG+HEG counts for all observations to improve statistics. In order to fit only the continuum emission, we selected a set of narrow bands, considered free of significant line emission, specifically 2.00-2.95Å, 4.4-4.6Å, 5.3-6.0Å, 7.5-7.8Å, 12.5-12.7Å and 19.1-20 Å (e.g. see Testa et al. (2007)). We assumed a hydrogen column density (N) of cm. Potential contamination resulting from cross-dispersion or zeroth order confusion was mitigated in these regions by simply ignoring the affected region of an individual order during the computation of the fit. The resulting continuum model was then used when fitting lines.

2.2 Line Fluxes & Ratios

Figure 2: Fit to the fir triplet - and -ratios for, clockwise from top left, Si xiii, Mg xi, O vii, Ne ix showing the predicted line profile in red, data and errors in black and gray respectively. The line centroid positions for each component are given here for clarity [,,]: Si xiii [6.65,6.68,6.74], Mg xi [9.17,9.24,933], O vii [21.60,21.80,22.11] and Ne ix [13.45,13.55,13.69].
Figure 3: Confidence contours for measured - and -ratios for, clockwise from top left, Si xiii, Mg xi, O vii, Ne ix. The red inner contours show 1, green middle contours show 2 and outer blue contours show 3 confidences.

The fir (forbidden, intercombination, and resonance) line ratios given by and have been shown to be probes of both density () and temperature () (Gabriel & Jordan, 1969) in X-ray emitting plasmas, and in the presence of a strong UV radiation field, such as is typical in O stars like \thetatwo, Waldron & Cassinelli (2001) demonstrated that the  value rather acts as a proxy for the radial distance of X-ray emission from the stellar surface. Specifically for the -ratio, it is also important to make the comparison between the observed ratio and that of the low density limit. Blumenthal et al. (1972) showed that

(1)

where is a measure of the photo-excitation, is a measure of the density and , , and depend only on atomic parameters and temperature. It is easily seen from Eq. 1 that, ignoring photo-excitation, = when and thus represents the low density limit. We have computed using emissivities in APED and temperatures derived from the  ratio given in Table 3 for each triplet and list them in Table 2. In order to test the MCWM predictions we derive the radial distance () using these ratios, a surface temperature of 30,000 K for the 09.7V star and photo-excitation and decay rates from Blumenthal et al. (1972). Figure 4 shows the dependence curves with measured values and 90% confidence intervals over-plotted.

In cases where there were significant contributions from other lines or line groups, as in the case of Ne ix  triplet where Fe xix and Fe xxi converge and blend, those lines were included in the model. A special case is Ne x which is unresolvably blended with FeXVII. In this case we assumed the Fe component contributed flux equaling 13% of the flux of a prominent FeXVII line at 15.01Å (e.g. see Walborn et al. (2009)). Additionally, the Mg Ly-series converges at the centroid position of the Si xiii f-line where we assumed, based on the theoretical relative line strengths, the observed flux was overestimated by 10% of the measured flux of the isolated H-Like Mg xii Ly line.

When fitting the He-like triplet lines, the relative separation of the lines was fixed and the positions of the resonance lines were constrained by their rest positions. Where available, we fit using both MEG and HEG counts, where MEG counts were rebinned onto the HEG grid whose intrinsic channel size is half that of the MEG. Fits were performed by applying Gaussian functions for each contributing line. Several contaminating lines known to be in the vicinity were included as well. The - and -ratios were computed directly during the fitting procedure and the fluxes were treated co-dependently. The instrumental profile was included as calibration data while the excess width was included as a gaussian turbulent broadening term (). In Figures 2 and 3 the triplet regions are shown with residuals, over-plotted models, and computed confidence contours.

2.3 Line Widths

Due to degradation of Chandra image quality at off-axis angles, the HETG resolving power likewise decreases. Though the PSF is well defined across the ACIS detector, this degradation becomes a problem for gratings because, owing to the complexity of modeling, responses are only calibrated for zeroth order positions at the instrument nominal pointing.

This effect can be critical in line width measurements which may include a significant instrumental broadening signature. Our flux measurements are unaffected by the broadening, and we have utilized as much available data as possible to improve statistics. Four of our observations are at off-axis angles greater than the others, in particular \thetatwo is greater than 3 off-axis in obsids 7410, 7411, 7412, and 8897. We have chosen to ignore counts in these obsids during computation of line width parameters. There are two exceptions. Si xiv and Mg xii where statistics are too poor in the absence of extra counts to obtain reasonable measurements. In these cases we provide upper limits on the line widths.

The average offset of our data is 2.1 which is around the location where degradation becomes noticeable. Based on analysis of ACIS zeroth order Line Response Functions (LRFs) at large axial offsets (e.g. see Chandra POG), we estimate that our reported line widths are on the order of up to 5% broader than that of identical on-axis profiles.

ION aaMeasured position of resonance line for He-like triplet line groups fluxbbflux is that of the resonance line only for He-like triplet line groups Line Ratiosccfor H-Like Ly lines this is the ratio of the H-Like Ly flux to He-Like resonance line flux of the corresponding ion
(Å) () ddComputed from APED emissivities according to Eq. 1 at -ratio derived temperatures (see Table 3) ()
He-Like Lines
Si xiii 6.650 1.30.3 1.10.3 1.70.4 2.4 491120
Mg xi 9.171 4.80.6 1.00.1 0.30.1 3.1 43253
Ne ix 13.448 28.93.3 1.20.1 0.10.04 2.8 22834
O vii 21.602 147.340.9 0.90.3 0.09 4.1 27483
           H-Like Ly Lines         
Si xiv 6.187 0.40.2 0.30.2 686
Mg xii 8.423 0.80.3 0.20.1 518
Ne x 12.133 20.02.9 0.70.1 31543
O viii 18.971 164.722.3 1.10.4 32753
Table 2: Line Measurements

3 Discussion

The exposure obtained from the Chandra archive of \thetatwo represents the deepest combined high resolution spectroscopic dataset on this young massive O-star to date. The long exposure provides high statistics in critical emission lines, allowing to diagnose its X-ray stellar wind properties beyond the level. In a previous study Schulz et al. (2006) provided some preliminary results for the quiescent state for less than half of the current exposure. This limited measurements of critical line fluxes and widths to uncertainties larger than 50. Our new analysis greatly reduces these uncertainties to the order of 20. For example, while the previous analysis could only speculate about possible line broadening of the order of 300 km s, we now clearly resolve the lines to values between 22834 km s for Ne ix and 491120 km s for Si xiii, with an average of all lines of 34138 km s. Likewise critical line ratios such as the -ratios are significantly improved, specifically for the cases of Mg xi with 0.30.09 and Si xiii with 1.70.4; for the case of O vii  since its -line was not detected, we now also have an upper limit.

The measured -ratios are significantly less than (Table 2). In early type stars this is due to the substantial UV radiation field provided by blackbody radiation of the hot surface temperature (Kahn et al., 2001; Gabriel & Jordan, 1969), which for the O9.7V star in \thetatwo is about 30,000 K. In this case the  ratio maps the distance of emission from the stellar surface and we utilize this relation to show that the X-ray line emission from \thetatwo is indeed located close to the O-star’s surface. Schulz et al. (2006) estimated that the emissions could be within several stellar radii, Table 2 and Figure 4 show emission origins within two stellar radii (dotted line in Figure 4) for Si xiii, Mg xi, and Ne ix with their 90 uncertainties.

ION Log T(H/He) Log T(G)
O vii 6.41 (6.36, 6.44) 6.3 (6.0, 6.5)
Ne ix 6.62 (6.60, 6.64) 6.1 (6.0, 6.3)
Mg xi 6.70 (6.65, 6.72) 6.5 (6.3, 6.6)
Si xiii 6.96 (6.87, 7.02) 6.4 (6.0, 6.8)
Table 3: Derived Temperatures
Figure 4: Dependence of  ratios on the distance to the stellar surface of the emission from Si xiii, Mg xi, Ne ix, O vii. The diamonds show the best fit and highlighted lines show 90% confidence limits projected on to dependence curves computed for a stellar surface temperature of 30,000 K. The vertical dashed line at 2R represents the approximate theoretical limit for generation of X-rays under the MCWM.

Another important result of our analysis is that the measured line centroid positions shown in Table 2 are, with quite high accuracy, at the expected ion rest wavelengths indicating that there are no line shifts within the Chandra sensitivity. This is an important result because any shift would indicate fast outward moving sources in a high density wind. The line profiles appear symmetric, supporting a low density wind assumption even though at such low broadening, profile deviations are almost impossible to trace even at our data quality.

In the case of Ori, Waldron & Cassinelli (2001) find that lines are resolved with comparatively low Doppler velocities of around 900 km s, -ratios that are characteristic of several stellar radii, an extremely small Si xiii -ratio, and symmetric and unshifted lines. Except for the extremely small Si xiii -ratio, our results seem very similar, if not more extreme with respect to line widths and -ratios. Our line widths indicate an even lower shock jump velocity than in the case of Ori making the formation of the observed ionization states even more difficult. At 350 km s shock temperatures are expected to not exceed 2 K. This discrepancy is supported by the emissivity distribution of the spectrum which includes X-ray temperatures greater than 25 MK (Schulz et al., 2006). These results are difficult to reconcile within the standard wind model. In this respect we conclude that a picture of a low density wind with shocks produced near its onset is not particularly convincing.

There are not many scenarios left which could explain our findings. We can rule out significant contributions of unseen low-mass pre-main sequence companions by the level of the line broadening. Standard coronal emission would show unresolved lines or moderate broadening due to orbital motion (Brickhouse et al., 2001; Huenemoerder et al., 2006); neither is the case here. Colliding winds are ruled out simply by the fact this would require an unseen massive companion with a much earlier type than the O9.7, which would be impossible to hide.

We find, however, quite strong similarities to the most massive star in the Orion Trapezium Ori C (Schulz et al., 2003; Gagné et al., 2005). In the magnetically confined wind scenario, field lines of the magnetic dipole act to channel emitted material from either pole toward the magnetic equator. Simulations by Gagné et al. (2005) demonstrate that these two components meet at the magneto-equator and wind plasma with high tangential velocities reaching up to 1000 km s collides generating strong shocks and elevate gas temperatures to tens of millions of degrees, thus producing the observed hard X-ray emission. Gagné et al. (2005) further demonstrate that the conditions for X-ray production are quite specific; the post shock in-falling material is rather cool, and the outflowing material’s density is too low to produce sufficient X-rays. This places a relatively tight constraint on the location of the hard X-ray emission around R2R.

Another result of the simulations by Gagné et al. (2005) states that the post shock-heated material is moving slowly, thus generating observed line profiles much narrower than expected for non-magnetic shock-heated X-ray production in O stars (Lucy, 1982; Waldron & Cassinelli, 2001). In order to quantify the expected broadening, Gagné et al. (2005) recreated emission measure and line profiles from the simulations and found that the turbulent broadening is expected to be on the order of 250 km s, with little to no blueshift in the line centroid position, which is very close to what we observe in \thetatwo.

4 Conclusion

We have analyzed high resolution X-ray spectra from Chandra on the young massive O star \thetatwo, totaling over 500 ks in the quiescent state, and computed line widths and line ratios for a series of prominent emission lines appearing in its spectrum. The resulting measurements show relatively narrow lines at an average width of 34138 km s and -ratio derived X-ray emitting origin within 2 stellar radii. Comparing these results to the simulation results of Gagné et al. (2005) for Ori C, we argue that the X-ray production mechanism in \thetatwo is most likely via magnetic confinement of its stellar wind outflows.

We have explored other possibilities, including standard O-star wind models and close companions, for the the generation of X-rays in \thetatwo but find that none of these are ideal for explaining the observed spectral properties. Observed line widths are too low, while shock temperatures too high to satisfy model predictions in most of these cases.

Finally, we note that this is a comparative analysis, and sets up a case for a more rigorous analysis specifically aimed at magneto-hydrodynamical modeling (MHD) using the MCWM similar to that under-taken for Ori C.

  This research has made use of data obtained from the Chandra Data Archive and software provided by the Chandra X-ray Center (CXC) in the application package CIAO. This research also made use of the Chandra Transmission Grating Catalog and archive http://tgcat.mit.edu. Chandra is operated by the Smithsonian Astrophysical Observatory under NASA contract NAS 8-03060. This work was supported by NASA through the Smithsonian Astrophysical Observatory (SAO) contracts NAS 8-03060 and SV3-73016 for the Chandra X-Ray Center and Science Instruments. Facilities: Chandra

References

  • Abt et al. (1991) Abt, H. A., Wang, R., & Cardona, O. 1991, ApJ, 367, 155
  • Babel & Montmerle (1997) Babel, J., & Montmerle, T. 1997, A&A, 323, 121
  • Blumenthal et al. (1972) Blumenthal, G. R., Drake, G. W. F., & Tucker, W. H. 1972, ApJ, 172, 205
  • Brickhouse et al. (2001) Brickhouse, N. S., Dupree, A. K., & Young, P. R. 2001, ApJ, 562, L75
  • Cohen et al. (2003) Cohen, D. H., de Messières, G. E., MacFarlane, J. J., Miller, N. A., Cassinelli, J. P., Owocki, S. P., & Liedahl, D. A. 2003, ApJ, 586, 495
  • Feigelson et al. (2002) Feigelson, E. D., Broos, P., Gaffney, III, J. A., Garmire, G., Hillenbrand, L. A., Pravdo, S. H., Townsley, L., & Tsuboi, Y. 2002, ApJ, 574, 258
  • Gabriel & Jordan (1969) Gabriel, A. H., & Jordan, C. 1969, MNRAS, 145, 241
  • Gagné et al. (2005) Gagné, M., Oksala, M. E., Cohen, D. H., Tonnesen, S. K., ud-Doula, A., Owocki, S. P., Townsend, R. H. D., & MacFarlane, J. J. 2005, ApJ, 628, 986
  • Houck & Denicola (2000) Houck, J. C., & Denicola, L. A. 2000, in Astronomical Society of the Pacific Conference Series, Vol. 216, Astronomical Data Analysis Software and Systems IX, ed. N. Manset, C. Veillet, & D. Crabtree, 591–+
  • Huenemoerder et al. (2006) Huenemoerder, D. P., Testa, P., & Buzasi, D. L. 2006, ApJ, 650, 1119
  • Huenemoerder et al. (2010) Huenemoerder, D. P., et al. 2010, AJ, in prep.
  • Kahn et al. (2001) Kahn, S. M., Leutenegger, M. A., Cottam, J., Rauw, G., Vreux, J., den Boggende, A. J. F., Mewe, R., & Güdel, M. 2001, A&A, 365, L312
  • Lucy (1982) Lucy, L. B. 1982, ApJ, 255, 286
  • Mitschang et al. (2010) Mitschang, A. W., Huenemoerder, D. P., & Nichols, J. S. 2010, ArXiv e-prints
  • Owocki et al. (1988) Owocki, S. P., Castor, J. I., & Rybicki, G. B. 1988, ApJ, 335, 914
  • Preibisch et al. (1999) Preibisch, T., Balega, Y., Hofmann, K., Weigelt, G., & Zinnecker, H. 1999, New Astronomy, 4, 531
  • Schulz et al. (2003) Schulz, N. S., Canizares, C., Huenemoerder, D., & Tibbets, K. 2003, ApJ, 595, 365
  • Schulz et al. (2000) Schulz, N. S., Canizares, C. R., Huenemoerder, D., & Lee, J. C. 2000, ApJ, 545, L135
  • Schulz et al. (2008) Schulz, N. S., Testa, P., Huenemoerder, D. P., & Canizares, C. R. 2008, in AAS/High Energy Astrophysics Division, Vol. 10, AAS/High Energy Astrophysics Division, 13.12–+
  • Schulz et al. (2006) Schulz, N. S., Testa, P., Huenemoerder, D. P., Ishibashi, K., & Canizares, C. R. 2006, ApJ, 653, 636
  • Simón-Díaz et al. (2006) Simón-Díaz, S., Herrero, A., Esteban, C., & Najarro, F. 2006, A&A, 448, 351
  • Smith et al. (2001) Smith, R. K., Brickhouse, N. S., Liedahl, D. A., & Raymond, J. C. 2001, ApJ, 556, L91
  • Stelzer et al. (2005) Stelzer, B., Flaccomio, E., Montmerle, T., Micela, G., Sciortino, S., Favata, F., Preibisch, T., & Feigelson, E. D. 2005, ApJS, 160, 557
  • Testa et al. (2007) Testa, P., Reale, F., Garcia-Alvarez, D., & Huenemoerder, D. P. 2007, ApJ, 663, 1232
  • Walborn et al. (2009) Walborn, N. R., Nichols, J. S., & Waldron, W. L. 2009, ApJ, 703, 633
  • Waldron & Cassinelli (2001) Waldron, W. L., & Cassinelli, J. P. 2001, ApJ, 548, L45
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
Cancel
Loading ...
133170
This is a comment super asjknd jkasnjk adsnkj
Upvote
Downvote
""
The feedback must be of minumum 40 characters
The feedback must be of minumum 40 characters
Submit
Cancel

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
Test description