EDP Sciences logo
Subscriber Authentication Point
Search
Menu
Home All issues Volume 632 (December 2019) A&A, 632 (2019) L10 Full HTML
Free Access
Issue
A&A
Volume 632, December 2019
Article Number L10
Number of page(s) 5
Section Letters to the Editor
DOI https://doi.org/10.1051/0004-6361/201936638
Published online 10 December 2019

© ESO 2019

1. Introduction

The elemental abundances of carbon and oxygen in proto-planetary disks are a vital input to planet formation models and combined with characterization of exoplanets can tell us about the formation history of those planets. The simple picture of elemental abundances changing statically at the ice lines of the main chemical species (e.g. Öberg et al. 2011) is being increasingly enriched by a number of modeling studies and observations. Chemical evolution can be efficient in changing the composition of the disk gas, changing the major carriers of carbon and oxygen (e.g., Eistrup et al. 2016; Schwarz et al. 2018; Bosman et al. 2018a). Furthermore, disk dynamics and dust evolution can efficiently transport the volatile component of the disk, changing the elemental composition of the gas and ice (Kama et al. 2016; Booth et al. 2017; Bosman et al. 2018a; Krijt et al. 2016, 2018).

As the interplay of the physical and chemical processes is complex, observations are needed to benchmark disk models and provide much needed input for planet formation models. Observations of CO and H2O focusing on the outer regions of proto-planetary disks show that their abundances are up to two orders of magnitude lower than expected, implying that chemical and physical processes are indeed modifying the abundances of these species (Hogerheijde et al. 2011; Favre et al. 2013; Bergin et al. 2013; Miotello et al. 2017; Du et al. 2017). Nevertheless, planet formation models primarily need the composition around and within the H2O ice line, as this is where we expect giant planets to form and accrete their atmospheres (e.g. Kennedy & Kenyon 2008; Cridland et al. 2017; Dawson & Johnson 2018).

Therefore, we aim to constrain the elemental abundances in the inner disk of TW Hya. The outer disk of TW Hya has been well studied and the elemental composition of the gas has been constrained from a variety of observations, including HD to trace the total disk mass, allowing for the measurement of absolute abundances (Hogerheijde et al. 2011; Bergin et al. 2013; Kama et al. 2016; Schwarz et al. 2016; Trapman et al. 2017). These efforts have shown that both the volatile carbon and oxygen abundance in the outer disk are lowered by a factor of about 50 compared to the interstellar medium (ISM). Observations of 13C18O in TW Hya show that while some CO comes from the grains within the CO ice line, the CO abundance stays a factor of about 20 lower than the ISM abundance (Zhang et al. 2017), implying that a lot of the carbon is trapped on the grains or is converted into other species.

The physical and chemical structure of the inner disk of TW Hya has also been studied in detail. It is the disk that has been resolved at the highest physical resolution with ALMA (Andrews et al. 2016) and together with abundant photometry and infrared interferometry, there is a clear picture of the inner disk structure of TW Hya (Andrews et al. 2012; Menu et al. 2014; Kama et al. 2016). In the infrared, high signal-to-noise ratio Spitzer-IRS and VLT-CRIRES spectra have been taken. These observations include 3.2 km s−1 resolution observations of the CO v = 1 − 0 rovibrational band at 4.7 μm, and detections of the CO2 15 μm Q-branch, pure rotational lines lines of H2, H2O, and OH, as well as a number of atomic hydrogen lines (Najita et al. 2010; Pontoppidan et al. 2008). Modeling efforts have further constrained the inner disk gas mass using H2 (Gorti et al. 2011) as well as the H2O content of the inner disk (Zhang et al. 2013). These studies hint at abundances for CO and H2O in the inner disk that are lower than expected from inner disk chemistry for gas of ISM composition, implying that elemental carbon and oxygen are depleted relative to the ISM.

Here we build on these studies, using the better-constrained outer disk structure and composition from Trapman et al. (2017), with an updated inner disk gap from ALMA observations (Andrews et al. 2016). Using this disk structure and the thermo-chemical code DALI we constrain the elemental carbon and oxygen abundance in the inner disk of TW Hya.

2. Methods

The lines and features, and their fluxes, that we consider for our modeling comparison are tabulated in Table 1. The CO flux for TW Hya is taken from Banzatti et al. (2017) based on VLT-CRIRES (Kaeufl et al. 2004) spectra presented in Pontoppidan et al. (2008, program ID 179.C-0151). The fluxes for H2 and H2O are extracted from the Spitzer-IRS spectrum obtained from program GO 30300 using the method described in Banzatti et al. (2012). The spectrum has been published in Najita et al. (2010) and Zhang et al. (2013).

Table 1.

H2, CO, and H2O fluxes from observations and modeling.

We start with the Dust and LInes (DALI, Bruderer et al. 2012; Bruderer 2013) model of Trapman et al. (2017) for TW Hya (see also Kama et al. 2016, model parameters are given in Table A.1). This model fits the SED, many far-infrared and sub-millimeter lines, and the ALMA 12CO J = 3 − 2 image. The most important lines are the HD 112 and 56 μm lines, constraining the outer disk mass, and many CO (isotopologue) lines constraining the carbon abundance. On top of these, some atomic carbon and oxygen fine-structure lines have also been fit, which constrain the elemental carbon and oxygen abundances in the outer disk at C/H = 2.7 × 10−6 and O/H = 5.8 × 10−6 which are a factor 50 lower than expected for the ISM. In this work we update the inner disk structure moving the outer edge of the gap from 4 to 2.4 AU in accordance with the bright submillimeter ring seen by Andrews et al. (2016) at this radius (assuming a distance of 54 parsec).

Figure 1 shows the surface density structure in the inner disk. The inner disk mass is varied by varying δgas in the inner disk using a model that has constant, low elemental abundances. The predicted H2 lines for DALI are compared to the observed fluxes. The model that fits best is then used to constrain the elemental C and O abundances. We make the simplifying assumptions that the CO abundance in the inner disk scales linearly with the total elemental carbon abundance, that the H2O abundance scales with the elemental oxygen that is not locked in CO, and that changing the CO and H2O abundances does not significantly alter the gas temperatures. The elemental C and O abundances are varied by changing the CO and H2O abundances by a factor δC and δO, respectively, above gas temperatures of Tstep of 70, 150, and 500 K. Here, 70 and 150 K are the sublimation temperatures of CO2 and H2O respectively, whereas 500 K is a rough transition temperature for the release of carbon from a more refractory reservoir. The excitation is recalculated for these new abundance structures and the CO v1 P(10) line flux and the full H2O spectrum between 12 and 34 μm is extracted.

thumbnail Fig. 1.

Surface density structure in inner regions of the TW Hya model. δgas is the inner disk drop in gas surface density. The value δgas = 0.1 used here best fits the H2 observations. The red curves show the column averaged oxygen and carbon depletion factors in TW Hya. The red solid line shows a model with constant depletion. The dotted line shows the depletion profile assuming that carbon and oxygen return to the ISM values above Tgas = 150 K.

The excitation for H2 is done in local thermodynamic equilibrium (LTE) as there are no collisional rate coefficients available for H2; however, the low Einstein A coefficients of the H2 lines mean that deviations from LTE are expected to be negligible (Wolniewicz et al. 1998). For H2O and CO the local excitation–de-excitation balance is calculated explicitly. For CO, collision rate coefficients from Yang et al. (2010) for H2 and Song et al. (2015) and Walker et al. (2015) for H are used (see also Bosman et al. 2019). For H2O, the data file from the LAMDA database1 (Schöier et al. 2005) is used.

3. Results

Figure 2 compares the emitting regions of the different species in the constant abundance model with δgas = 0.1. There is a large overlap in the emitting areas, especially in the inner disk (< 2.4 AU). Water has the most confined emitting area and only probes the inner disk while CO and H2 both also probe the outer disk, where the composition of the gas is already strongly constrained (Bergin et al. 2013; Kama et al. 2016; Trapman et al. 2017; Zhang et al. 2017).

thumbnail Fig. 2.

Map of the gas temperature in the inner region of the model together with the emitting areas of the H2 S(1) line (black) the CO v1 vibrational lines (orange) and the strongest line of the H2O 33 μm feature (purple). The green line shows Tdust = 150 K, the approximate location of the H2O ice line. The emitting areas radially overlap in the inner disk.

Figure 3 compares the results of the DALI modeling with the observed fluxes. The H2 fluxes constrain the inner disk gas mass to be around 1.7 × 10−4 M⊙ (δgas ≈ 0.1), comparable to the value found in Gorti et al. (2011). The observations show a higher H2 S(2)/S(1) line ratio compared to the models, indicating that the average gas temperature in the inner disk of the model is too low. The upper level energies of the CO (∼3000 K) and H2O (∼1500 K) are similar to or higher than the upper level energies of the H2 lines (∼1000 and ∼1700 K), therefore a higher temperature in the inner disk would lead to even lower inferred values for C/H and O/H.

thumbnail Fig. 3.

Comparison of the H2 line fluxes (left), CO rovibrational line flux (middle), and the H2O 33 μm feature flux (right) between models and data. Left panel: the amount of gas within 2.4 AU is varied. Middle and right panels: δgas = 0.1 is used and C/H and O/H are varied respectively. Horizontal bands show the 1σ variation of the observed fluxes. The different coloured points denote different values for Tstep. The temperature threshold above which the abundances are varied. The vertical lines in the middle and right panel show the volatile carbon and oxygen abundances assumed for the ISM (Meyer et al. 1998; Lacy et al. 2017).

Both the CO lines and H2O lines fit well to an inner disk elemental abundance similar to the abundance found in the outer disk, which is a factor 50 lower compared to the ISM. Even a jump of a factor of two in the elemental carbon or oxygen abundance above 500 K can be ruled out based on the observed fluxes. The H2O fluxes in Table 1 show that this model underpredicts the H2O detections between 20 and 31 μm by a factor of about two, which is in line with the gas temperature in the model being too low.

4. Discussion

4.1. Constraining the inner disk chemical structure

Here, using a more complete physical and chemical structure of both the inner and the outer disk, we can quantify the total H2 mass, as well as the volatile carbon and oxygen abundance in the inner disk, confirming that the inner disk of TW Hya is both oxygen- and carbon-poor by a factor about 50 compared to the ISM. Furthermore, our modeling shows that there is no significant (factor 2 or more) increase in volatile carbon or oxygen in the inner disk. Thus, there is no sign of volatile release at the CO2 or H2O ice lines nor is there evidence of carbonaceous or silicate grain destruction at T <  500 K.

Both Gorti et al. (2011) and Zhang et al. (2013) studied the inner region of TW Hya using detailed modeling. They note that they overproduce the CO rovibrational lines by a factor of about two. Furthermore, Gorti et al. (2011) assume LTE excitation for CO, which generally underpredicts fluxes compared to models that include infrared pumping of the vibrational levels by the inner disk continuum emission (Bruderer et al. 2015; Bosman et al. 2017). These two factors together explain why our models need a CO or elemental carbon abundance that is almost two orders of magnitude lower than that of the ISM to fit the CO rovibrational line.

Zhang et al. (2013) use a low abundance in the inner disk xH2O <  10−6 with a ring of high abundance H2O at 4 AU (their ice line) to produce the Spitzer lines. Including such a ring and fitting the 33 μm (Eup = 1504 K) feature would lead to inferring lower oxygen abundances in the inner disk; however, it would also move the H2O emission to colder gas, increasing the discrepancy between model and observations for the 22 and 28 μm features. A constant-abundance model therefore fits the data better than a ring model.

4.2. Hiding C and O carriers?

The DALI models predict that CO and H2O are the dominant gas-phase oxygen and carbon carriers in the inner disk. However, it is possible that carbon and oxygen are locked in other gaseous species. There are a few obvious candidates for hiding more carbon and oxygen in molecular gas: CO2, C2H2, HCN, and CH4. There is a CO2 feature detected in the Spitzer spectrum of TW Hya. Applying the model results from Bosman et al. (2017) to the detected feature retrieves a CO2 abundance lower than 10−9 with respect to H2. The models in Bosman et al. (2017) do not have a gap, which if included would only increase the strength of the CO2 lines. As such the CO2 abundance < 10−9 is a stringent upper limit and this molecule is not an abundant carrier of either carbon or oxygen.

C2H2 and HCN are detected in many proto-planetary disks (Salyk et al. 2011), however neither are detected in the spectrum of TW Hya. As the Q-branches of HCN and C2H2 around 14 μm have similar upper level energies and Einstein A coefficients to the 15 μm Q-branch of CO2, both HCN and C2H2 should have been detected if they are as abundant as CO2. The lack of observed features implies that both HCN and C2H2 contain less than 1% of the volatile carbon. The final possible gaseous carbon carrier is CH4. Observations of this molecule have so far only been successful in one disk, in which CH4 lines have been detected in absorption (Gibb & Horne 2013). The CH4 abundance is therefore not strongly constrained by observations. However, from chemical models CH4 is not expected to be the most abundant carbon carrier in inner disk atmospheres (Walsh et al. 2015; Agúndez et al. 2018). As such, with CO and H2O we trace the bulk of gaseous carbon and oxygen in the inner disk.

4.3. Implications of uniform depletion

The low elemental abundances in the inner disk imply one of two things: either the entire disk is depleted in volatiles, implying dust grains with almost no ices, or volatiles are efficiently locked in ices on solids that are not transported through the disk.

To deplete the entire disk in volatiles, a period of very strong radial drift coupled with strong radial mixing is necessary (e.g., Booth et al. 2017). In such a scenario, the icy grains drift from the mass reservoir in the outer disk into the regions of the disk where the ices desorb. This leads to enhancements in C/H and O/H in the inner regions of the disk. After a few million years all of this high-C/H and high-O/H gas is accreted onto the star and the disk is left depleted in carbon and oxygen. However, a disk like this will also be strongly depleted in solid material as the grains that transport the ices inward will continue to drift into the star leading to dust depletions at least as high as the volatile element depletion. This scenario is therefore very unlikely for TW Hya, which has a massive dust component and a gas-to-dust ratio of around 100 (Bergin et al. 2013).

As efficient transport is excluded, this leaves 98% of the carbon and oxygen locked in solids that are not efficiently transported. Assuming that CO and H2O are the dominant oxygen and carbon carriers in the outer disk, it is necessary to stop CO from being transported over the CO ice line by a dust trap at a location larger than where the CO ice line is located. For TW Hya this would imply that one of the shallow millimeter continuum rings of TW Hya outside of 20 AU corresponds to an efficient dust trap. An efficient dust trap at > 20 AU should result in a disk with a large cavity. As dust is abundant down to 2.4 AU, a second dust trap would be necessary at that location, which would trap H2O-rich grains.

In this scenario H2O would be far more efficiently locked in large grains than CO as the former is frozen out in a larger fraction of the disk. However, if gaseous CO is efficiently converted into less volatile species, especially CO2 and CH3OH, then the CO- and H2O-depletion fractions are likely to be more similar and an efficient dust trap around 20 AU is no longer strictly necessary. The age of TW Hya of 10 Mr, is long enough to convert large amounts of CO into other species (Donaldson et al. 2016; Bosman et al. 2018b). In this case, the dust should not be allowed to pass the CO2 or CH3OH ice lines. These ice lines are at nearly the same location as the H2O ice line at the inner edge of the outer disk. As such, a dust trap at the innermost submillimeter ring would trap all of the icy CO2, CH3OH, and H2O in the outer disk. A schematic representation of this is given in Fig. 4.

thumbnail Fig. 4.

Schematic of the TW Hya disk showing the locations that should trap the oxygen- and carbon-bearing molecules. Oxygen needs to be trapped outside of 2.4 AU and is most likely trapped in the dust ring at that location. Carbon can either be trapped at the same location, with CO2 being the most probable carrier, or at larger radii outside the CO ice line, with CO being the most probable carrier.

In summary, we find that the elemental carbon and oxygen abundances in the inner disk are lower by a factor of approximately 50 compared to the ISM. Even at temperatures of 500 K, the gaseous carbon or oxygen elemental abundances cannot have increased by a factor of two, strongly constraining the release of volatiles and grain destruction up to 500 K. This is interpreted as the dust trap responsible for the dust-free cavity also trapping the major carbon- and oxygen-bearing ices outside of 2.4 AU. A planet currently accreting gas in the gap will accrete very low amounts of carbon and oxygen, while possibly accreting ISM concentrations of nitrogen and noble gasses. Depending on the accretion history this planet could have a substellar C/H and O/H abundance.

Acknowledgments

Astrochemistry in Leiden is supported by the Netherlands Research School for Astronomy (NOVA). This work is partly based on observations made with CRIRES on ESO telescopes at the Paranal Observatory under program ID 179.C-0151. This work is based in part 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. This project has made use of the SciPy stack (Virtanen et al. 2019), including NumPy (Oliphant 2006) and Matplotlib (Hunter 2007).

References

  1. Agúndez, M., Roueff, E., Le Petit, F., & Le Bourlot, J. 2018, A&A, 616, A19 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  2. Andrews, S. M., Wilner, D. J., Hughes, A. M., et al. 2012, ApJ, 744, 162 [NASA ADS] [CrossRef] [Google Scholar]
  3. Andrews, S. M., Wilner, D. J., Zhu, Z., et al. 2016, ApJ, 820, L40 [NASA ADS] [CrossRef] [Google Scholar]
  4. Banzatti, A., Meyer, M. R., Bruderer, S., et al. 2012, ApJ, 745, 90 [NASA ADS] [CrossRef] [Google Scholar]
  5. Banzatti, A., Pontoppidan, K. M., Salyk, C., et al. 2017, ApJ, 834, 152 [NASA ADS] [CrossRef] [Google Scholar]
  6. Bergin, E. A., Cleeves, L. I., Gorti, U., et al. 2013, Nature, 493, 644 [NASA ADS] [CrossRef] [PubMed] [Google Scholar]
  7. Booth, R. A., Clarke, C. J., Madhusudhan, N., & Ilee, J. D. 2017, MNRAS, 469, 3994 [NASA ADS] [CrossRef] [Google Scholar]
  8. Bosman, A. D., Bruderer, S., & van Dishoeck, E. F. 2017, A&A, 601, A36 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  9. Bosman, A. D., Tielens, A. G. G. M., & van Dishoeck, E. F. 2018a, A&A, 611, A80 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  10. Bosman, A. D., Walsh, C., & van Dishoeck, E. F. 2018b, A&A, 618, A182 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  11. Bosman, A. D., Banzatti, A., Bruderer, S., et al. 2019, A&A, 631, A133 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  12. Bruderer, S. 2013, A&A, 559, A46 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  13. Bruderer, S., van Dishoeck, E. F., Doty, S. D., & Herczeg, G. J. 2012, A&A, 541, A91 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  14. Bruderer, S., Harsono, D., & van Dishoeck, E. F. 2015, A&A, 575, A94 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  15. Cridland, A. J., Pudritz, R. E., Birnstiel, T., Cleeves, L. I., & Bergin, E. A. 2017, MNRAS, 469, 3910 [NASA ADS] [CrossRef] [Google Scholar]
  16. Dawson, R. I., & Johnson, J. A. 2018, ARA&A, 56, 175 [NASA ADS] [CrossRef] [Google Scholar]
  17. Donaldson, J. K., Weinberger, A. J., Gagné, J., et al. 2016, ApJ, 833, 95 [NASA ADS] [CrossRef] [Google Scholar]
  18. Du, F., Bergin, E. A., Hogerheijde, M., et al. 2017, ApJ, 842, 98 [NASA ADS] [CrossRef] [Google Scholar]
  19. Eistrup, C., Walsh, C., & van Dishoeck, E. F. 2016, A&A, 595, A83 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  20. Favre, C., Cleeves, L. I., Bergin, E. A., Qi, C., & Blake, G. A. 2013, ApJ, 776, L38 [NASA ADS] [CrossRef] [Google Scholar]
  21. Gibb, E. L., & Horne, D. 2013, ApJ, 776, L28 [NASA ADS] [CrossRef] [Google Scholar]
  22. Gorti, U., Hollenbach, D., Najita, J., & Pascucci, I. 2011, ApJ, 735, 90 [NASA ADS] [CrossRef] [Google Scholar]
  23. Hogerheijde, M. R., Bergin, E. A., Brinch, C., et al. 2011, Science, 334, 338 [NASA ADS] [CrossRef] [PubMed] [Google Scholar]
  24. Hunter, J. D. 2007, Comput. Sci. Eng., 9, 90 [NASA ADS] [CrossRef] [Google Scholar]
  25. Kaeufl, H. U., Ballester, P., Biereichel, P., et al. 2004, in Ground-based Instrumentation for Astronomy, eds. A. F. M. Moorwood, & M. Iye, Proc. SPIE, 5492, 1218 [NASA ADS] [CrossRef] [Google Scholar]
  26. Kama, M., Bruderer, S., van Dishoeck, E. F., et al. 2016, A&A, 592, A83 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  27. Kennedy, G. M., & Kenyon, S. J. 2008, ApJ, 673, 502 [NASA ADS] [CrossRef] [Google Scholar]
  28. Krijt, S., Ciesla, F. J., & Bergin, E. A. 2016, ApJ, 833, 285 [NASA ADS] [CrossRef] [Google Scholar]
  29. Krijt, S., Schwarz, K. R., Bergin, E. A., & Ciesla, F. J. 2018, ApJ, 864, 78 [NASA ADS] [CrossRef] [Google Scholar]
  30. Lacy, J. H., Sneden, C., Kim, H., & Jaffe, D. T. 2017, ApJ, 838, 66 [NASA ADS] [CrossRef] [Google Scholar]
  31. Menu, J., van Boekel, R., Henning, T., et al. 2014, A&A, 564, A93 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  32. Meyer, D. M., Jura, M., & Cardelli, J. A. 1998, ApJ, 493, 222 [NASA ADS] [CrossRef] [Google Scholar]
  33. Miotello, A., van Dishoeck, E. F., Williams, J. P., et al. 2017, A&A, 599, A113 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  34. Najita, J. R., Carr, J. S., Strom, S. E., et al. 2010, ApJ, 712, 274 [NASA ADS] [CrossRef] [Google Scholar]
  35. Öberg, K. I., Boogert, A. C. A., Pontoppidan, K. M., et al. 2011, ApJ, 740, 109 [NASA ADS] [CrossRef] [Google Scholar]
  36. Oliphant, T. 2006, NumPy: A Guide to NumPy (USA: Trelgol Publishing) [Google Scholar]
  37. Pontoppidan, K. M., Blake, G. A., van Dishoeck, E. F., et al. 2008, ApJ, 684, 1323 [NASA ADS] [CrossRef] [Google Scholar]
  38. Salyk, C., Pontoppidan, K. M., Blake, G. A., Najita, J. R., & Carr, J. S. 2011, ApJ, 731, 130 [NASA ADS] [CrossRef] [Google Scholar]
  39. Schöier, F. L., van der Tak, F. F. S., van Dishoeck, E. F., & Black, J. H. 2005, A&A, 432, 369 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  40. Schwarz, K. R., Bergin, E. A., Cleeves, L. I., et al. 2016, ApJ, 823, 91 [NASA ADS] [CrossRef] [Google Scholar]
  41. Schwarz, K. R., Bergin, E. A., Cleeves, L. I., et al. 2018, ApJ, 856, 85 [NASA ADS] [CrossRef] [Google Scholar]
  42. Song, L., Balakrishnan, N., Walker, K. M., et al. 2015, ApJ, 813, 96 [NASA ADS] [CrossRef] [Google Scholar]
  43. Trapman, L., Miotello, A., Kama, M., van Dishoeck, E. F., & Bruderer, S. 2017, A&A, 605, A69 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  44. Virtanen, P., Gommers, R., Oliphant, T. E., et al. 2019, ArXiv e-prints [arXiv:1907.10121] [Google Scholar]
  45. Walker, K. M., Song, L., Yang, B. H., et al. 2015, ApJ, 811, 27 [NASA ADS] [CrossRef] [Google Scholar]
  46. Walsh, C., Nomura, H., & van Dishoeck, E. 2015, A&A, 582, A88 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  47. Wolniewicz, L., Simbotin, I., & Dalgarno, A. 1998, ApJS, 115, 293 [NASA ADS] [CrossRef] [Google Scholar]
  48. Yang, B., Stancil, P. C., Balakrishnan, N., & Forrey, R. C. 2010, ApJ, 718, 1062 [NASA ADS] [CrossRef] [Google Scholar]
  49. Zhang, K., Pontoppidan, K. M., Salyk, C., & Blake, G. A. 2013, ApJ, 766, 82 [NASA ADS] [CrossRef] [Google Scholar]
  50. Zhang, K., Bergin, E. A., Blake, G. A., Cleeves, L. I., & Schwarz, K. R. 2017, Nat. Astron., 1, 0130 [NASA ADS] [CrossRef] [Google Scholar]

Appendix A: DALI model

Table A.1.

Parameters of the TW Hya model.

All Tables

Table 1.

H2, CO, and H2O fluxes from observations and modeling.

In the text
Table A.1.

Parameters of the TW Hya model.

In the text

All Figures

thumbnail Fig. 1.

Surface density structure in inner regions of the TW Hya model. δgas is the inner disk drop in gas surface density. The value δgas = 0.1 used here best fits the H2 observations. The red curves show the column averaged oxygen and carbon depletion factors in TW Hya. The red solid line shows a model with constant depletion. The dotted line shows the depletion profile assuming that carbon and oxygen return to the ISM values above Tgas = 150 K.
In the text
thumbnail Fig. 2.

Map of the gas temperature in the inner region of the model together with the emitting areas of the H2 S(1) line (black) the CO v1 vibrational lines (orange) and the strongest line of the H2O 33 μm feature (purple). The green line shows Tdust = 150 K, the approximate location of the H2O ice line. The emitting areas radially overlap in the inner disk.
In the text
thumbnail Fig. 3.

Comparison of the H2 line fluxes (left), CO rovibrational line flux (middle), and the H2O 33 μm feature flux (right) between models and data. Left panel: the amount of gas within 2.4 AU is varied. Middle and right panels: δgas = 0.1 is used and C/H and O/H are varied respectively. Horizontal bands show the 1σ variation of the observed fluxes. The different coloured points denote different values for Tstep. The temperature threshold above which the abundances are varied. The vertical lines in the middle and right panel show the volatile carbon and oxygen abundances assumed for the ISM (Meyer et al. 1998; Lacy et al. 2017).
In the text
thumbnail Fig. 4.

Schematic of the TW Hya disk showing the locations that should trap the oxygen- and carbon-bearing molecules. Oxygen needs to be trapped outside of 2.4 AU and is most likely trapped in the dust ring at that location. Carbon can either be trapped at the same location, with CO2 being the most probable carrier, or at larger radii outside the CO ice line, with CO being the most probable carrier.
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.