Spitzer Spectroscopy of the Circumprimary Disk in the Binary Brown Dwarf 2MASS J04414489+230151311affiliation: Based on observations made with the Spitzer Space Telescope, which is operated by the Jet Propulsion Laboratory, California Institute of Technology under a contract with NASA.
Using the Spitzer Infrared Spectrograph, we have performed mid-infrared spectroscopy on the young binary brown dwarf 2MASS J04414489+2301513 (15 AU) in the Taurus star-forming region. The spectrum exhibits excess continuum emission that likely arises from a circumstellar disk around the primary. Silicate emission is not detected in these data, indicating the presence of significant grain growth. This is one of the few brown dwarf disks at such a young age ( Myr) that has been found to lack silicate emission. To quantitatively constrain the properties of the disk, we have compared the spectral energy distribution of 2MASS J04414489+2301513 to the predictions of our vertical structure codes for irradiated accretion disks. Our models suggest that the remaining atmospheric grains of moderately depleted layers may have grown to a size of µm. In addition, our model fits indicate an outer radius of 0.2–0.3 AU for the disk. The small size of this circumprimary disk could be due to truncation by the secondary. The absence of an outer disk containing a reservoir of small, primordial grains, combined with a weak turbulent mechanism, may be responsible for the advanced grain growth in this disk.
Grain growth, crystallization, and settling occur in primordial circumstellar disks around both stellar and substellar objects. These processes produce several observational signatures. The mid-infrared (IR) silicate emission becomes weaker as the grains in the upper layers grow and settle towards the midplane while crystallization modifies the shape of the band. IR continuum emission is reduced at longer wavelengths with grain growth and settling because the intercepted stellar flux decreases as the disk flattens. In the substellar domain, these hallmarks of dust evolution have been reported previously for disks at ages Myr (Apai et al., 2004, 2005; Scholz et al., 2007; Morrow et al., 2008; Pascucci et al., 2009; Riaz, 2009; Riaz et al., 2009). For instance, disks around the youngest brown dwarfs exhibit evidence of small, crystalline silicate grains (Apai et al., 2005; Riaz, 2009) while older brown dwarf disks have weaker silicate emission that indicates the presence of large grains (Scholz et al., 2007; Morrow et al., 2008; Riaz et al., 2009). The evolution of grains in the inner annuli of these disks, e.g., the zone where the 10 µm silicate band emerges (within AU, Kessler-Silacci et al., 2007), appears to occur more rapidly than in disks around solar-mass stars. One possible explanation for this phenomena is that the low levels of turbulence for brown dwarf disks produce lower mass accretion rates and reduce the replenishment of the inner disk with primordial grains from outer annuli (Sicilia-Aguilar et al., 2007). Also, the collisional rate between large size grains would be small, preventing the fragmentation to small size grains.
During a mid-IR spectroscopic survey of young brown dwarfs with the Spitzer Space Telescope (Werner et al., 2004), we have identified a brown dwarf disk that shows evidence of advanced grain evolution at a relatively young age of Myr. This brown dwarf, 2MASS J04414489+2301513 (catalog ) (Luhman, 2006, henceforth 2M J04414489), lies in the Taurus star-forming region and was recently shown to harbor a companion with a projected separation of ( AU) and a mass of 5–10 (Todorov et al., 2010). In this Letter, we use our mid-IR spectrum of this system and our models of settled, irradiated accretion disks to constrain the properties of the circumstellar disk that resides around the primary.
We observed 2M J04414489 with the Spitzer Infrared Spectrograph (IRS; Houck et al., 2004) on 2009 April 20 as a part of the Guaranteed Time Observations of the IRS instrument team. We collected data with both of the low-resolution IRS modules, Short-Low and Long-Low, which cover 5.3-14 and 14-40 µm, respectively, with a resolution of . We used the basic calibrated data produced by version S18.7 of the Spitzer Science Center pipeline. Before extraction of the spectrum, bad and “rogue” pixels were fixed by interpolating adjacent, good pixels in the dispersion direction. 2M J04414489 was then extracted using the optimal extraction tool in SMART (Higdon et al., 2004; Lebouteiller et al., 2010). We chose the default calibration, which is based on a relative spectral response function derived from three calibrator stars. The young low-mass star 2MASS J04414565+2301580 is from 2M J04414489 and fell within the slit during the LL observations. We performed optimal extraction at the nominal nod position of 2M J04414489 to extract its spectrum, which should result in minimal contamination from 2MASS J04414565+2301580. Indeed, the resulting flux levels of the data at SL and LL match at 14 µm without applying any scaling factors. The IRS spectrum includes the flux of both 2M J04414565 A and B, but it should be dominated by the former, as explained in Section 3.1.
3.1 Disk Emission
We show the spectral energy distribution (SED) of 2M J04414489 in Figure 1, which was constructed with our IRS data, 0.8–2.5 µm spectroscopy from Luhman (2006), and 3–8 µm photometry from Luhman et al. (2010). We have included in Figure 1 the SED for a diskless young brown dwarf near the same spectral type as 2M J04414489 (2MASSW J1139511-315921, Morrow et al., 2008). Relative to the stellar photosphere of the latter object, the SED of 2M J04414489 exhibits significant excess emission that indicates the presence of a dusty, optically thick disk, which is consistent with earlier results based on the mid-IR photometry (Luhman et al., 2010). The disk is probably located around the primary because a disk around the secondary would be too faint to account for the observed excess emission based on the flux ratio at near-IR wavelengths (, Todorov et al., 2010) and the typical near- to mid-IR colors of brown dwarfs with disks (Luhman et al., 2010). A circumbinary disk is not plausible since its inner hole would be too large for the disk to produce significant mid-IR emission. Finally, the mid-IR excess emission cannot be due to the photosphere of the secondary since its IR colors should be similar to those of the primary’s photosphere. For instance, given the spectral type of M8.5 for the primary and an expected type of M9.5-L0 for the secondary (Luhman, 2006; Todorov et al., 2010), their photospheres should have colors near and 0.83, respectively (Luhman et al., 2010). Combining those colors with the flux ratio at suggests that the secondary’s photosphere produces an excess of only 0.02 mag in for the system relative to the photospheric colors of the primary.
3.2 Model Parameters
The optical spectrum of 2M J04414489 exhibits H emission (Luhman, 2006), indicating the presence of ongoing mass accretion. Therefore, we have modeled the mid-IR excess emission from 2M J04414489 in terms of emission from a settled, irradiated accretion disk following the procedures of D’Alessio et al. (1998, 2005, 2006) adapted for substellar objects (Adame, 2010). For the stellar photosphere, we adopt an effective temperature of 2555 K, which corresponds to its M8.5 spectral type (Luhman et al., 2003; Luhman, 2006). We also adopt a luminosity of 0.004 based on its magnitude and a mass of 0.025 M based on theoretical evolutionary models (Chabrier et al., 2000). In our models, we solve the vertical structure of the disk while assuming a constant turbulent-viscosity parameter and a uniform mass accretion rate, , through the disk. For most of the brown dwarfs in Taurus, assuming is suitable (Adame, 2010), although we allow to range from to 0.01 for our models of 2M J04414489. In each annulus, the mass surface density depend on the ratio . We compute models with , , and M yr, which are typical values measured for young substellar objects (Natta et al., 2004; Muzerolle et al., 2005; Mohanty et al., 2005; Herczeg & Hillenbrand, 2008; Herczeg et al., 2009).
The adopted dust model is composed of segregated spheres of astronomical silicates and graphite with abundances and optical constants from Draine & Lee (1984) and Weingartner & Draine (2001) and a size distribution of (Mathis et al., 1977), where . To account for grain growth, we have varied the maximum grain size in the upper layers and the inner wall, adopting values of , 1, 5, 10, 100 and 1000 µm. Meanwhile, we assume mm for the dust population near the midplane. As grains grow and settle towards the midplane, the atmosphere is depleted of small grains. The model of D’Alessio et al. (2006) quantifies this depletion through the parameter , which is the dust-to-gas mass ratio of the atmosphere relative to the standard ratio (with and , Draine & Lee, 1984). We have considered , 0.01, and 0.1 in our models.
We also calculate the structure and SED of the inner wall , which is heated by radiation from the stellar photosphere (D’Alessio et al., 2005). The wall is located at radius , which can be the dust destruction radius (assuming a sublimation temperature of K) or the magnetospheric radius, whichever is larger. We explore maximum dust temperatures of K to 1000 K (in steps of 100 K), corresponding to an inner dust gap with to 6 R. The outer disk radius may be set by interaction with the secondary (Papaloizou & Pringle, 1977; Artymowicz & Lubow, 1994). Assuming a mass ratio of the secondary to the primary of , a projected semimajor axis of AU (Todorov et al., 2010), and a mean Reynolds number of (see Artymowicz & Lubow, 1994), the truncation radius is predicted to be AU for eccentricities between and (a large circumprimary radius corresponds to a small eccentricity). Therefore, we fixed to an initial value of 5 AU; for promising models, we have selected values of that range from 0.1–5 AU. Finally, we adopt inclination angles of , 20, and . The disk SED library consists of some 700 models.
3.3 Model Fits
We proceed to identify the combinations of disk parameters that can reproduce the SED of 2M J04414489. We perform a reduced- goodness-of-fit test between each model and the IRS data (e.g. van Boekel et al., 2005), where indicates a good match. We find that models with , M yr, and a wall temperature of K match the SED at µm better than other values of these parameters. The radius of this inner wall is R AU. We allow the wall height to vary from H to H, where H is the gas scale height at . The models with these values of , , and that produce are 21. An important constrain for our model SED selection is the value of the mass accretion rate. From the relationship of Herczeg & Hillenbrand (2008), the mass accretion rate onto this low mass brown dwarf would be , thus models with M yr are in the higher end of the expected mass accretion rate. On the other hand, the equivalent width of the H emission line of 2M J04414489 (Luhman, 2006) is 100 Å, which corresponds to a 10% width of H of km s (Muzerolle et al., 2005), and to a mass accretion rate of M yr (Natta et al., 2004). Hence we discard the high mass accretion models, and select the remaining four models as our best-fit models, which are listed in Table 1.
The SED predicted by model A, which has the lowest value of , is compared to the observed SED in Figure 1. The variations in the disk parameters among the best-fit models in Table 1 are qualitatively explained by the fact that the mid-IR emission is weaker for smaller and and larger and , causing a reduced optical depth. For instance, the grain growth of model C and the depletion of model D must be balanced by increasing the dust mass content in order for the models to produce sufficient emission to match the data. Meanwhile, models with mm and most of the models with M yr produced emission that was too weak to reproduce the data ().
We compare the silicate emission predicted by our best-fit models to the IRS data for 2M J04414489 in Figure 2. We normalize the IRS spectrum and the models with a continuum interpolated from a third-order polynomial constructed by fluxes between and µm and between and 16 µm (e.g. Furlan et al., 2006). The strength of the silicate feature relative to the continuum is obtained by subtracting the continuum to the feature, integrating this residual flux and normalize it to the integrated continuum flux between and µm (Furlan et al., 2006). The strength of the observed feature is 1.03, whereas the strengths of our models features range from 1.024 to 1.032. Thus, the observed and modeled silicate features are similar, but also, they are extremely weak. The absence of detectable silicate emission in the data indicates that grains in the disk atmosphere have grown to sizes of µm, which are larger than the typical ISM grains.
The dust masses of our final set of disk models are listed in Table 1, using (Draine & Lee, 1984). The masses of the models range from to , which are more than an order of magnitude lower than the values estimated from millimeter measurements of other brown dwarf disks in Taurus (Scholz et al., 2006). This result is not surprising given that close binary stars exhibit low disk masses compared to single stars and wide binaries (Jensen et al., 1994).
3.4 Size of the Mid-IR Emission Region
The mid-IR emission from an isolated disk irradiated by a M8.5 brown dwarf typically forms within AU. The exact size of this region depends the size of the grains (Kessler-Silacci et al., 2007), but also on the dust mass surface density through the variation of , or . We find that decreasing or increasing or result in a wider emitting region since those variations increase the irradiation surface and the impinging irradiation flux at any given radius. On the other hand, decreasing will increase the optical depth to the incident radiation at the upper layers, which in turn extends vertically outward the irradiation surface, allowing more distant annuli to contribute to the emission.
If the outer radius of the disk is smaller than the nominal radius of the mid-IR emitting region, then the mid-IR SED is sensitive to the value of the outer radius. For instance, the radii for all of the models in Table 1 are constrained to be AU by the mid-IR SED of 2M J04414489, which is smaller than the expected radius given by the binary tidal interaction (–7 AU). However, if the eccentricity of the binary is high enough, the periastron would be close to 1 AU if a semi-major axis of 15 AU is still assumed. The dependence of the predicted SED on the outer radius is illustrated in Figure 3 for model A. Although the disk radii for our models of 2M J04414489 are roughly similar to that derived for the circumprimary disk of SR 20 (0.39 AU, for a G7 primary and a binary projected separation of AU, McClure et al., 2008), the former has a redder SED (i.e., less affected by truncation) because the mid-IR emitting region of a brown dwarf is much smaller than that of a star.
We find that the dust in the disk around 2M J04414489 is relatively evolved for the age of the system ( Myr). Based on the slope of its SED and the absence of silicate emission, most of the initial atmospheric grains have settled towards the midplane () and the remaining grains may have experienced significant growth ( µm), representing one of brown dwarf disks in Taurus that show evidence of accelerated dust evolution (see KPNO 6111For the other objects reported by Riaz (2009), silicate emission is detected at weak levels in the same spectra that were re-reduced (Furlan et al. 2011, in preparation)., Riaz, 2009). This degree of dust evolution has been observed primarily for brown dwarfs at ages of 5–10 Myr (Scholz et al., 2007; Morrow et al., 2008).
Some mechanisms to explain grain growth and settling in disks around either single or multiple late-type M-dwarfs have been put forward. Meeus et al. (2003) suggested that binary interaction may accelerate the dust evolution through an enhanced vertical stirring of grains from the midplane. However, Pascucci et al. (2008) and Furlan et al. (2011, in preparation) found no significant differences between the silicate feature for early type single and binary star systems in Taurus, revealing that binary interaction may not affect the inner disk region, although the outermost region is indeed influenced. The interaction truncates the disk to a given radius (Papaloizou & Pringle, 1977; Artymowicz & Lubow, 1994), decreasing the emission from a certain wavelength range. For instance, the submillimeter and millimeter emission from the circumprimary disk is diminished for projected binary separations of —100 AU (Jensen et al., 1994, 1996), while for smaller separations ( AU), the mid-IR emission of the outer disk decreases with respect to that of the innermost region (McClure et al., 2010). The mid-IR emission of J04414489 is the result of this truncation, whereas the fluxes near µm are unlikely influenced by the binary interaction. Instead, the inner region may be probably affected by the turbulence as most of the eddie dissipation occurs within those annuli ( AU), which represents —50% of the total radial extension of the disk.
Grain growth and settling should be efficient in regions close to the star, where temperatures and densities are larger, and the turbulent movements are higher (Beckwith et al., 2000; Dullemond & Dominik, 2004; Nomura & Nakagawa, 2006, references therein). For this weakly turbulent disk (), the coagulation rate may be slowed down, but also the relative velocities of the grains, thus preventing their fragmentation. The smaller velocity gradients can work to preserve the large grains once formed. A radial mixing mechanism may exist that moves evolved grains outward and replenishes the inner disk with primordial dust (e.g. turbulent diffusion, convection, gravitational torques, among others, see Gail, 2001; Bockelée-Morvan et al., 2002; Boss, 2004). The primordial dust reservoir is limited by the outer radius of a disk. If a disk is outwardly truncated to a small enough radius (e.g., ), the outermost region will contain evolved grains, which return to the innter region to be processed further. The characteristic timescale for this radial mixing is approximately the viscous timescale (Bockelée-Morvan et al., 2002), which for an -disk, . Here, is the sound speed at the midplane and the Keplerian angular speed, both evaluated at (the initial outer radius of the disk). We evaluate at the outer radii listed in Table 1 assuming a central temperature at the midplane of —80 K. The mixing timescale derived is —2 Myr, which is comparable to the age of 2M J04414489. The radial mixing may be at an earliest stage, but its influence over the dust processing cannot be neglected. Therefore, low levels of turbulence combined with a highly truncated disk could explain the deficit of primordial dust that we observe in the disk of 2M J04414489.
We have presented a mid-IR spectrum of the binary brown dwarf 2M J04414489 in the Taurus star-forming region. The spectrum exhibits excess emission that we attribute to a circumprimary disk, but silicate emission at 10 µm is not present, indicating that significant grain growth has occurred, leading to a disk on which small grains no longer exist ( µm). This is one of brown dwarf disks at such a young age ( Myr) that have been found to lack silicate emission. Our models of the SED of 2M J04414489 suggest that the outer radius of its disk is rather small (–0.3 AU), which may reflect truncation by the binary companion. The absence of an outer disk with a reservoir of primordial grains, combined with weak turbulence, could explain the advanced grain growth in the inner disk of 2M J04414489.
- Adame (2010) Adame, L. 2010, Ph. D. Thesis, Universidad Nacional Autónoma de México
- Apai et al. (2004) Apai, D., Pascucci, I., Sterzik, M. F., van der Bliek, N., Bouwman, J., Dullemond, C. P., & Henning, T. 2004, A&A, 426, L53
- Apai et al. (2005) Apai, D., Pascucci, I., Bouwman, J., Natta, A., Henning, T., & Dullemond, C. P. 2005, Science, 310, 834
- Artymowicz & Lubow (1994) Artymowicz, P., & Lubow, S. H. 1994, ApJ, 421, 651
- Beckwith et al. (2000) Beckwith, S. V. W., Henning, T., & Nakagawa, Y. 2000, Protostars and Planets IV, 533
- Bockelée-Morvan et al. (2002) Bockelée-Morvan, D., Gautier, D., Hersant, F., Huré, J.-M., & Robert, F. 2002, A&A, 384, 1107
- Boss (2004) Boss, A. P. 2004, ApJ, 616, 1265
- Chabrier et al. (2000) Chabrier, G., Baraffe, I., Allard, F., & Hauschildt, P. 2000, ApJ, 542, 464
- D’Alessio et al. (1998) D’Alessio, P., Cantó, J., Calvet, N., & Lizano, S., 1998, ApJ, 500, 411
- D’Alessio et al. (2005) D’Alessio, P., et al. 2005, ApJ, 621, 461
- D’Alessio et al. (2006) D’Alessio, P., Calvet, N., Hartmann, L., Franco-Hernández, R., & Servín, H. 2006, ApJ, 638, 314
- Draine & Lee (1984) Draine, B. T., & Lee, H. M. 1984, ApJ, 285, 89
- Dullemond & Dominik (2004) Dullemond, C. P., & Dominik, C. 2004, A&A, 421, 1075
- Furlan et al. (2006) Furlan, E., et al. 2006, ApJS, 165, 568
- Gail (2001) Gail, H.-P. 2001, A&A, 378, 192
- Herczeg et al. (2009) Herczeg, G. J., Cruz, K. L., & Hillenbrand, L. A. 2009, ApJ, 696, 1589
- Herczeg & Hillenbrand (2008) Herczeg, G. J., & Hillenbrand, L. A. 2008, ApJ, 681, 594
- Higdon et al. (2004) Higdon, S. J. U., et al. 2004, PASP, 116, 975
- Houck et al. (2004) Houck, J. R., et al. 2004, ApJS, 154, 18
- Jensen et al. (1994) Jensen, E. L. N., Mathieu, R. D., & Fuller, G. A. 1994, ApJ, 429, L29
- Jensen et al. (1996) Jensen, E. L. N., Mathieu, R. D., & Fuller, G. A. 1996, ApJ, 458, 312
- Kessler-Silacci et al. (2007) Kessler-Silacci, J. E., et al. 2007, ApJ, 659, 680
- Lebouteiller et al. (2010) Lebouteiller, V., Bernard-Salas, J., Sloan, G. C., & Barry, D. J. 2010, PASP, 122, 231
- Luhman (2006) Luhman, K. L. 2006, ApJ, 645, 676
- Luhman et al. (2010) Luhman, K. L., Allen, P. R., Espaillat, C., Hartmann, L., & Calvet, N. 2010, ApJS, 186, 111
- Luhman et al. (2003) Luhman, K. L., Stauffer, J. R., Muench, A. A., Rieke, G. H., Lada, E. A., Bouvier, J., & Lada, C. J. 2003, ApJ, 593, 1093
- Mathis et al. (1977) Mathis, J. S., Rumpl, W., & Nordsieck, K. H. 1977, ApJ, 217, 425
- McClure et al. (2008) McClure, M. K., et al. 2008, ApJ, 683, L187
- McClure et al. (2010) McClure, M. K., et al. 2010, ApJS, 188, 75
- Meeus et al. (2003) Meeus, G., Sterzik, M., Bouwman, J., & Natta, A. 2003, A&A, 409, L25
- Mohanty et al. (2005) Mohanty, S., Jayawardhana, R., & Basri, G., 2005, ApJ, 626, 498
- Morrow et al. (2008) Morrow, A. L., et al. 2008, ApJ, 676, L143
- Muzerolle et al. (2005) Muzerolle, J., Luhman, K. L., Briceño, C., Hartmann, L., & Calvet, N. 2005, ApJ, 625, 906
- Natta et al. (2004) Natta, A., Testi, L., Muzerolle, J., Randich, S., Comerón, F., & Persi, P. 2004, A&A, 424, 603
- Nomura & Nakagawa (2006) Nomura, H., & Nakagawa, Y. 2006, ApJ, 640, 1099
- Papaloizou & Pringle (1977) Papaloizou, J., & Pringle, J. E. 1977, MNRAS, 181, 441
- Pascucci et al. (2008) Pascucci, I., Apai, D., Hardegree-Ullman, E. E., Kim, J. S., Meyer, M. R., & Bouwman, J. 2008, ApJ, 673, 477
- Pascucci et al. (2009) Pascucci, I., Apai, D., Luhman, K., Henning, T., Bouwman, J., Meyer, M. R., Lahuis, F., & Natta, A. 2009, ApJ, 696, 143
- Riaz (2009) Riaz, B. 2009, ApJ, 701, 571
- Riaz et al. (2009) Riaz, B., Lodieu, N., & Gizis, J. E. 2009, ApJ, 705, 1173
- Scholz et al. (2006) Scholz, A., Jayawardhana, R., & Wood, K. 2006, ApJ, 645, 1498
- Scholz et al. (2007) Scholz, A., Jayawardhana, R., Wood, K., Meeus, G., Stelzer, B., Walker, C., & O’Sullivan, M. 2007, ApJ, 660, 1517
- Sicilia-Aguilar et al. (2007) Sicilia-Aguilar, A., Hartmann, L. W., Watson, D., Bohac, C., Henning, T., & Bouwman, J. 2007, ApJ, 659, 1637
- Todorov et al. (2010) Todorov, K., Luhman, K. L., & McLeod, K. K. 2010, ApJ, 714, L84
- van Boekel et al. (2005) van Boekel, R., Min, M., Waters, L. B. F. M., de Koter, A., Dominik, C., van den Ancker, M. E., & Bouwman, J. 2005, A&A, 437, 189
- Weingartner & Draine (2001) Weingartner, J. C., & Draine, B. T. 2001, ApJ, 548, 296
- Werner et al. (2004) Werner, M. W., et al. 2004, ApJS, 154, 1
|Modelbb, M yr, , R for all models.||z|