Free Access
Issue
A&A
Volume 584, December 2015
Article Number L9
Number of page(s) 5
Section Letters
DOI https://doi.org/10.1051/0004-6361/201527237
Published online 24 November 2015

© ESO, 2015

1. Introduction

Classical T Tauri stars (CTTSs) are young stars accreting from their surrounding protoplanetary disk. The disk’s gas-to-dust ratio is challenging to measure directly. Therefore, the interstellar medium (ISM) ratio of 100:1 is regularly assumed (see review by Williams & Cieza 2011). In the standard magnetospheric accretion model, matter is channeled along magnetic field lines from the inner edge of the disk onto the stellar surface (Bouvier et al. 2007).

The young binary RW Aurigae (distance 140 pc) is among the well-studied CTTS with dense photometric and spectroscopic monitoring. It is composed of a jet-driving primary and a secondary separated by 1.̋4; the masses of both stars are thought to be approximately solar. The system (A+B combined) shows complex periodic optical variability. Its magnitude and color change by a few tenths of mag on time scales of 2–5 days (e.g., Petrov et al. 2001a,b). Additional irregular variability of up to two magnitudes along the reddening vector is superimposed onto this periodic variability (e.g., Grankin et al. 2007; Petrov & Kozack 2007). Observations of CO emission show that the primary’s disk is truncated at R< 60 AU and that diffuse emission extends well beyond the disk; both features are best explained by a close (~70 AU) approach of RW Aur B around A (Cabrit et al. 2006; Dai et al. 2015). A half-year dimming event in 2011 (ΔV ≈ 2 mag) was attributed to a “bridge” of material related to the extended CO structure connecting both components (Rodriguez et al. 2013; Dai et al. 2015).

Towards the end of 2014, the system dimmed again by 2 mag. The similarity of both dimming events could suggest that the “extra absorber” in 2014–2015 is again associated with the tidal stream. However, Petrov et al. (2015) found changes in certain wind features and that the absorber covered only the star and its immediate surroundings. They proposed that the occultation was related to an interaction between the stellar wind and the inner disk, which is supported by enhanced M- and L-band fluxes (Shenavrin et al. 2015). In this case, there might be similarities with UXor events observed in more massive stars (e.g., Grinin et al. 1998). No substantial increase in column density of cold gas towards the dimmed star was found (Petrov et al. 2015).

RW Aur has also been observed by Chandra with ACIS-S for 54 ks during the bright state (January 2013; Skinner & Güdel 2014). The A-B pair was resolved and the data revealed that the secondary is more X-ray luminous than the primary and that it is moderately variable (factor of 1.5 within 60 ks).

X-ray absorption is essentially independent of ionization, because only the inner atomic shells contribute to X-ray absorption. Typically, the ratio between X-ray absorption expressed as equivalent neutral hydrogen column density (NH) and optical extinction (AV) is used as a measure of the absorber’s gas-to-dust ratio. As RW Aur is already well characterized during the bright state, we use the increase in X-ray absorption to derive the gas and small grain content of the extra absorber, where “small” means that the grains are not opaque in X-rays (a few μm) .

We present new X-ray and optical/near-IR (NIR) data obtained during the dim state in Sect. 2, derive the corresponding column densities in Sect. 3, and compare both measurements in Sect. 4 to investigate the properties of the extra absorber and the nature of the dimming event.

2. Observations and data analysis

The new X-ray, optical, and NIR observations obtained to characterize the dim state are summarized in Table 1.

2.1. Chandra observation and data processing

The new X-ray observation (exposure time: 35 ks) was performed with the same setup as the 2013 observation of RW Aur during the bright state, i.e., 1/8 subarray readout was used to minimize the pileup of the X-ray bright B component (see Skinner & Güdel 2014). Data reduction was performed using ciao version 4.6 (Fruscione et al. 2006). Unless otherwise noted, the 0.3–10.0 keV energy range was used.

Table 1

Overview of the analyzed observations.

Figure 1 shows the X-ray images as well as source and background extraction regions. The location of the 2013 jet emission (green arrow in Fig. 1) is outside our source extraction region and so does not affect our analysis. The source region (radius: 0.̋54, PSF fraction: 75%) contains 51 photons and we estimate that PSF spillover by RW Aur B contributes 21.4 photons. We checked that the exact parameters of the regions do not significantly influence the spectral data.

We modeled the stellar X-ray emission in xspec using photoelectric absorption (phabs) and vapec models with plasma abundances from Skinner & Güdel (2014) for consistency. The equivalent hydrogen column density depends on the assumed absorber abundances; we used Anders & Grevesse (1989). Lower metallicities as suggested by Asplund et al. (2009) increase NH by about a factor of 1.5. To provide a robust estimate of the absorbing column density given the limited number of photons in 2015, we modeled the 2015 emission as a scaled version of the 2013 two component model plus extra absorption, i.e., we fixed the emission measure ratio (EMhot/EMcool) to values that accurately describe the 2013 Chandra data. We also investigated several other models (e.g., free EMhot/EMcool ratio, one component thermal plasma with free temperature) and find that the results are generally consistent with this approach.

thumbnail Fig. 1

X-ray images of RW Aur taken during the bright state in 2013 (top) and during the dim state in 2015 (bottom); the color bars indicate counts per pixel. The scaling reflects the difference in exposure time, i.e., similar count rates have similar colors. The inset shows the same sky area in the K-band obtained two days after the X-ray observation.

Open with DEXTER

Here, we are exclusively interested in the evolution of RW Aur A and note that the average X-ray flux of the B component increased by a factor 1.4 at similar spectral properties (unabsorbed LX = 1.9 × 1030 erg s-1) with respect to 2013 and that its count rate smoothly decreases by a factor of 1.3 during the exposure. X-ray emission from the RW Aur A jet as described by Skinner & Güdel (2014) is not significantly detected.

2.2. Ground-based data

Optical data in Johnson-Cousins filters were obtained with the Oskar-Lühning-Teleskop (OLT) at the Hamburger Sternwarte. Data reduction followed standard procedures. We also report AAVSO1V magnitudes that densely sample the optical brightness evolution of the system. When near-simultaneous OLT data are available, both datasets are compatible. Near-IR data were provided by IRTF and UKIRT. They were reduced using IRAF and Starlink2, respectively. The inset in Fig. 1 shows a UKIRT K-band image as an example of our NIR photometry. Absolute photometry was obtained by calibration against the 2MASS magnitudes of a check field and the standard star FS 12 (Hawarden et al. 2001) for the data from 20 March 2015 and 18 April 2015, respectively.

To estimate the brightness drop of RW Aur A, we compare the dim state values to its average bright state values, which we estimate from the ROTOR data (Grankin et al. 2007) transforming their Johnson R to Cousin R using the Bessell (1979) relations. For the NIR, we use the 2MASS magnitudes (Skrutskie et al. 2006) as the bright state reference. The contribution of the B component to the optical/NIR flux is below 10% during the bright state, but important during the dim state. For spatially unresolved observations, we subtract the average brightness of B calculated from resolved data to obtain the magnitude of A (see Table A.1).

3. Results

In the following, we show that the gas and dust absorption strongly increased between the bright and dim states.

thumbnail Fig. 2

X-ray spectra of RW Aur A with models. The APEC model uses the parameters provided in Table 2. The green shaded area shows the 2013 data scaled to match at Ephot> 3 keV including absorption; lower and upper bounds pertain to  cm-2and 4 × 1022 cm-2, respectively. Data was binned for display purposes.

Open with DEXTER

3.1. X-ray data

Figure 2 compares the X-ray spectra during the bright and dim states. The observed (absorbed) flux of the A component decreased by almost a factor of 20 (see Table 2) mainly owing to the decrease at soft photon energies. A significant source signal is only recorded above 1.9 keV. The number of photons at lower photon energies is compatible with a background fluctuation at the 90% confidence range. In addition, the observed flux above 2 keV is reduced by about a factor of three.

Table 2

X-ray properties of RW Aur A.

The absorbing column density is rather well constrained because a hot component is needed to describe the high-energy tail of the observed spectrum (Ephot> 2 keV) and emission from this component extends well into the soft part of the spectrum and drives the required absorbing column density. The fit results, summarized in Table 2, show that the absorbing column density increased by almost a factor of 30 to NH ≈ 2 × 1022 cm-2 in 2015. The temperature of the hot component was fixed as its precise value is unconstrained (already noted by Skinner & Güdel 2014). However, this does not affect the resulting NH significantly. The best fit column density corresponds to 5 × 10-2 g cm-2 for a mean molecular weight of μ = 1.4 and to AV = 13 mag for ISM-like absorption (NH = 1.8 × 1021 cm-2, Predehl & Schmitt 1995; Vuong et al. 2003).

thumbnail Fig. 3

Optical light curve of RW Aur around the X-ray observation (spatially unresolved).

Open with DEXTER

3.2. Optical and NIR data

Figure 3 shows the V-bandlight curve as well as the color evolution around the Chandra observation. Only minor variability is seen around the X-ray observation (standard deviation in V is 0.14 mag on a ten-day time scale). Attributing this variability to the primary suggests intrinsic changes by only 0.2 mag (color changes are <0.1 mag), i.e., the optical properties are very similar during the Chandra and UKIRT observations. This strongly suggests that the UKIRT data closely approximates the conditions during the X-ray observation. Compared to variability during the bright state, these differences are small so that we regard our X-ray, optical, and NIR datasets as effectively simultaneous.

Figure 4 shows the drop in brightness as a function of wavelength with respect to the mean bright state. The drop at optical () and NIR () wavelengths is almost identical, especially when considering that the bright state’s reference magnitudes are somewhat uncertain owing to intrinsic variability. Therefore, ISM-like absorption, or minor variations thereof, are incompatible with the data (see blue and red curves in Fig. 4 top). Scattering of stellar photons in the circumstellar environment might contribute to the observed optical flux, which would erroneously indicate only little optical extinction possibly explaining why optical and NIR fluxes show a similar drop. However, the NIR is essentially unaffected by scattering because of its strong wavelength dependence (σscat ~ λ-4). Thus, substantial dust extinction should offset RW Aur from the CTTS locus (Meyer et al. 1997) along the reddening vector while the RW Aur colors are below the CTTS locus (Fig. 4 bottom), i.e., the NIR colors indicate no significant reddening. Without scattering, the evolution of the RW Aur A optical/NIR magnitudes are almost wavelength independent, i.e., gray. This requires an absorber consisting mainly of large grains and only a small amount of small grains, which is compatible with the findings by Antipin et al. (2015) from their optical data.

thumbnail Fig. 4

Top: extinction by the extra absorber with different extinction curves. ISM-like models have Agray = 0 and RV = 3.1. The scattering model assumes ISM-like extinction with AV = 10 mag, a scattering efficiency of 13% in V, and no additional extinction along the scattering path. Photometric bands are shown at their effective wavelengths. Bottom: NIR color–color diagram.

Open with DEXTER

Gray absorption up to NIR wavelengths requires dust composed of grains with sizes 1 μm. The ratio between extinction and mass decreases with increasing grain size. Therefore, assuming grains of 1 μm in size provides a lower limit on the dust mass. These grains have absorption efficiencies around 2.5 (see, e.g., Draine & Lee 1984) so that the mass attenuation coefficient is 104cm2 g-1 for dust densities of ρ ≈ 2 g cm-3 (e.g., carbonaceous grains, which dominate ISM dust at these sizes). The required dust mass column density is about 2 × 10-4 g cm-2 for gray extinction of 2 mag.

4. Interpretation and conclusions

First, we assume the simplest scenario that reasonably explains our data: one single absorber, which contains the gas that causes the X-ray absorption and the large dust grains that cause the gray optical/NIR extinction. The requirement of predominately large grains is compatible with the 10/20 μm features seen in some IR spectra of CTTSs, which also suggest large grains in the μm-size range (e.g., Oliveira et al. 2010). Combining X-ray and dust absorption, we derive an upper limit on the gas-to-dust ratio for this absorber of 250:1 with about a factor of two uncertainty (a factor of 1.7 in X-ray column density and we assume a factor of 1.5 in dust mass density ρ). Thus, one absorber with an ISM-like gas-to-dust ratio can explain the X-ray and optical data for processed dust, i.e., for dust with sizes between about 1 and 10 μm. This suggests that grains grow from their ISM distribution with a peak in the few 0.1 μm range to sizes that are about ten times larger without strongly altering the gas-to-dust ratio. Without a substantial increase in low-ionization Na and K i absorption as suggested by Petrov et al. (2015), the strong increase in X-ray absorption indicates that the absorbing material is hot. This suggests that the absorbing material is located close to the star and might also be responsible for the enhanced L- and M-band fluxes measured by Shenavrin et al. (2015), e.g., the absorber might be a disk warp or an inner, dust-loaded wind as speculated by Petrov et al. (2015).

Second, we note that X-ray and dust extinction are not necessarily cospatial. The gas absorption might be caused by dust-depleted accretion streams or winds launched from the disk rim as in AA Tau where the innermost region (~0.1 au) is strongly gas-enhanced (ΔNH ≈ 1022 cm-2; see Schmitt & Robrade 2007; Grosso et al. 2007). In addition to this gas absorbing component, an opaque structure that partly occults RW Aur A causes the gray extinction extending from the NIR to the highest

energies in our X-ray spectrum. This opaque structure might be also located in the inner disk (0.1 au) as the gas is, but could also be located slightly farther out at a few au – like the absorber that causes the long-lasting dimming of AA Tau (Bouvier et al. 2013; Schneider et al. 2015) – or could even be part of the tidal stream. Depending on the actual location of the absorber, this likely requires a tilt between the inner and the outer disk resolved by the CO observations given that the upper limit on the inclination of the RW Aur A disk is i< 60°, while that of AA Tau is i ≈ 75°. Such a scenario appears attractive, because it releases the constraint that the grain size distribution is missing grains smaller than 1 μm and provides a uniform explanation for AA Tau-like CTTS and RW Aur A.


1

Observations from the AAVSO International Database, http://www.aavso.org

Acknowledgments

We thank B. Wilkes for granting the Chandra DDT observation and the referee, Prof. G. Gahm, for the careful and constructive report. P.C.S. and C.F.M. gratefully acknowledge an ESA Research Fellowship, H.M.G. is supported by NASA-HST-GO-12315.01, S.J.W. NASA contract NAS8-03060 (Chandra), and S.F. by an STFC/Isaac Newton Trust studentship. The results reported are based on observations made by the Chandra X-ray Observatory and by the United Kingdom Infrared Telescope (UKIRT) supported by NASA and operated under an agreement among the University of Hawaii, the University of Arizona, and Lockheed Martin Advanced Technology Center; operations are enabled through the cooperation of the Joint Astronomy Centre of the Science and Technology Facilities Council of the UK and by UKIRT. Some photometry was obtained at the Infrared Telescope Facility (IRTF), which is operated by the University of Hawaii under contract NNH14CK55B with the National Aeronautics and Space Administration (NASA).

References

  1. Anders, E., & Grevesse, N. 1989, Geochim. Cosmochim. Acta, 53, 197 [NASA ADS] [CrossRef] [Google Scholar]
  2. Antipin, S., Belinski, A., Cherepashchuk, A., et al. 2015, Information Bulletin on Variable Stars, 6126, 1 [NASA ADS] [Google Scholar]
  3. Asplund, M., Grevesse, N., Sauval, A. J., & Scott, P. 2009, ARA&A, 47, 481 [NASA ADS] [CrossRef] [Google Scholar]
  4. Bessell, M. S. 1979, PASP, 91, 589 [NASA ADS] [CrossRef] [Google Scholar]
  5. Bouvier, J., Alencar, S. H. P., Harries, T. J., Johns-Krull, C. M., & Romanova, M. M. 2007, Protostars and Planets V, 479 [Google Scholar]
  6. Bouvier, J., Grankin, K., Ellerbroek, L. E., Bouy, H., & Barrado, D. 2013, A&A, 557, A77 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  7. Cabrit, S., Pety, J., Pesenti, N., & Dougados, C. 2006, A&A, 452, 897 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  8. Cutri, R. M., Skrutskie, M. F., van Dyk, S., et al. 2003, VizieR Online Data Catalog: II/246 [Google Scholar]
  9. Dai, F., Facchini, S., Clarke, C. J., & Haworth, T. J. 2015, MNRAS, 449, 1996 [NASA ADS] [CrossRef] [Google Scholar]
  10. Draine, B. T., & Lee, H. M. 1984, ApJ, 285, 89 [NASA ADS] [CrossRef] [Google Scholar]
  11. Fruscione, A., McDowell, J. C., Allen, G. E., et al. 2006, in SPIE Conf. Ser., 6270, 1 [Google Scholar]
  12. Grankin, K. N., Melnikov, S. Y., Bouvier, J., Herbst, W., & Shevchenko, V. S. 2007, A&A, 461, 183 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  13. Grinin, V. P., Rostopchina, A. N., & Shakhovskoi, D. N. 1998, Astron. Lett., 24, 802 [NASA ADS] [Google Scholar]
  14. Grosso, N., Bouvier, J., Montmerle, T., et al. 2007, A&A, 475, 607 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  15. Hawarden, T. G., Leggett, S. K., Letawsky, M. B., Ballantyne, D. R., & Casali, M. M. 2001, MNRAS, 325, 563 [NASA ADS] [CrossRef] [Google Scholar]
  16. Meyer, M. R., Calvet, N., & Hillenbrand, L. A. 1997, AJ, 114, 288 [NASA ADS] [CrossRef] [Google Scholar]
  17. Oliveira, I., Pontoppidan, K. M., Merín, B., et al. 2010, ApJ, 714, 778 [NASA ADS] [CrossRef] [Google Scholar]
  18. Petrov, P. P., & Kozack, B. S. 2007, Astron. Rep., 51, 500 [NASA ADS] [CrossRef] [Google Scholar]
  19. Petrov, P. P., Gahm, G. F., Gameiro, J. F., et al. 2001a, A&A, 369, 993 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  20. Petrov, P. P., Pelt, J., & Tuominen, I. 2001b, A&A, 375, 977 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  21. Petrov, P. P., Gahm, G. F., Djupvik, A. A., et al. 2015, A&A, 577, A73 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  22. Predehl, P., & Schmitt, J. H. M. M. 1995, A&A, 293, 889 [NASA ADS] [Google Scholar]
  23. Rodriguez, J. E., Pepper, J., Stassun, K. G., et al. 2013, AJ, 146, 112 [NASA ADS] [CrossRef] [Google Scholar]
  24. Schmitt, J. H. M. M., & Robrade, J. 2007, A&A, 462, L41 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  25. Schneider, P. C., France, K., Günther, H. M., et al. 2015, A&A, 584, A51 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  26. Shenavrin, V. I., Petrov, P. P., & Grankin, K. N. 2015, Information Bulletin on Variable Stars, 6143, 1 [NASA ADS] [Google Scholar]
  27. Skinner, S. L., & Güdel, M. 2014, ApJ, 788, 101 [NASA ADS] [CrossRef] [Google Scholar]
  28. Skrutskie, M. F., Cutri, R. M., Stiening, R., et al. 2006, AJ, 131, 1163 [NASA ADS] [CrossRef] [Google Scholar]
  29. Vuong, M. H., Montmerle, T., Grosso, N., et al. 2003, A&A, 408, 581 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  30. White, R. J., & Ghez, A. M. 2001, ApJ, 556, 265 [NASA ADS] [CrossRef] [Google Scholar]
  31. Williams, J. P., & Cieza, L. A. 2011, ARA&A, 49, 67 [NASA ADS] [CrossRef] [Google Scholar]

Online material

Appendix A: RW Aur photometry

Table A.1 lists the optical/NIR magnitudes used in this study.

Table A.1

Optical and NIR magnitudes ordered by observing date.

All Tables

Table 1

Overview of the analyzed observations.

Table 2

X-ray properties of RW Aur A.

Table A.1

Optical and NIR magnitudes ordered by observing date.

All Figures

thumbnail Fig. 1

X-ray images of RW Aur taken during the bright state in 2013 (top) and during the dim state in 2015 (bottom); the color bars indicate counts per pixel. The scaling reflects the difference in exposure time, i.e., similar count rates have similar colors. The inset shows the same sky area in the K-band obtained two days after the X-ray observation.

Open with DEXTER
In the text
thumbnail Fig. 2

X-ray spectra of RW Aur A with models. The APEC model uses the parameters provided in Table 2. The green shaded area shows the 2013 data scaled to match at Ephot> 3 keV including absorption; lower and upper bounds pertain to  cm-2and 4 × 1022 cm-2, respectively. Data was binned for display purposes.

Open with DEXTER
In the text
thumbnail Fig. 3

Optical light curve of RW Aur around the X-ray observation (spatially unresolved).

Open with DEXTER
In the text
thumbnail Fig. 4

Top: extinction by the extra absorber with different extinction curves. ISM-like models have Agray = 0 and RV = 3.1. The scattering model assumes ISM-like extinction with AV = 10 mag, a scattering efficiency of 13% in V, and no additional extinction along the scattering path. Photometric bands are shown at their effective wavelengths. Bottom: NIR color–color diagram.

Open with DEXTER
In the text

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

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

Initial download of the metrics may take a while.