Where is the warm H2 ? A search for H2 emission from disks around Herbig Ae/Be stars
C. Martin-Zaidi, J-.C. Augereau, F. Menard, J. Olofsson, A. Carmona, C. Pinte, E. Habart
aa r X i v : . [ a s t r o - ph . S R ] A p r Astronomy&Astrophysicsmanuscript no. Martin-Zaidi˙fin c (cid:13)
ESO 2018October 30, 2018
Where is the warm H ?A search for H emission from disks around Herbig Ae/Be stars C. Martin-Za¨ıdi , J-.C. Augereau , F. M´enard , J. Olofsson , A. Carmona , C. Pinte , and E. Habart Universit´e Joseph Fourier - Grenoble 1 / CNRS, Laboratoire d’Astrophysique de Grenoble (LAOG) UMR 5571, BP 53, 38041Grenoble Cedex 09, Francee-mail: [email protected] ISDC Data Centre for Astrophysics & Geneva Observatory, University of Geneva, chemin d’Ecogia 16, 1290 Versoix, Switzerland School of Physics, University of Exeter, Stocker Road, Exeter EX4 4QL, United Kingdom Institut d’Astrophysique Spatiale, Universit´e Paris-Sud, 91405 Orsay Cedex, FranceReceived ... / Accepted ...
ABSTRACT
Context.
Mid-infrared (mid-IR) emission lines of molecular hydrogen (H ) are useful probes to determine the mass of warm gaspresent in the surface layers of circumstellar disks. In the past years, numerous observations of Herbig Ae / Be stars (HAeBes) havebeen performed, but only two detections of H mid-IR emission toward HD 97048 and AB Aur have been reported. Aims.
We aim at tracing the warm gas in the circumstellar environment of five additional HAeBes with gas-rich environments and / orphysical characteristics close to those of AB Aur and / or HD 97048, to discuss whether the detections toward these two objects aresuggestive of peculiar conditions for the observed gas. Methods.
We search for the H S(1) emission line at 17.035 µ m using high-resolution mid-IR spectra obtained with VLT / VISIR ,and complemented by CH molecule observations with VLT / UVES . We gather the H measurements from the literature to put the newresults in context and search for a correlation with some disk properties. Results.
None of the five
VISIR targets shows evidence for H emission at 17.035 µ m. From the 3 σ upper limits on the integratedline fluxes we constrain the amount of optically thin warm ( > K ) gas to be less than ∼ . M Jup in the disk surface layers. Thereare now 20 HAeBes observed with
VISIR and
TEXES instruments to search for warm H , but only two detections (HD 97048 andAB Aur) were made so far. We find that the two stars with detected warm H show at the same time high 30 / µ m flux ratios andlarge PAH line fluxes at 8.6 and 11.3 µ m compared to the bulk of observed HAeBes and have emission CO lines detected at 4 . µ m.We detect the CH 4300.3 Å absorption line toward both HD 97048 and AB Aur with UVES . The CH to H abundance ratios that thiswould imply if it were to arise from the same component as well as the radial velocity of the CH lines both suggest that CH arisesfrom a surrounding envelope, while the detected H would reside in the disk. Conclusions.
The two detections of the S(1) line in the disks of HD 97048 and AB Aur suggest either peculiar physical conditionsor a particular stage of evolution. New instruments such as
Herschel / PACS should bring significant new data for the constraints ofthermodynamics in young disks by observing the gas and the dust simultaneously.
Key words. stars: circumstellar matter – stars: formation – stars: pre-main sequence – ISM: molecules
1. Introduction
Planets are supposed to form in circumstellar disks composed ofgas and dust around stars in their pre-main sequence phase. Atthis evolutionary stage, the disk mass is essentially dominated bythe gas (99%), especially molecular hydrogen (H ). Although H is the principal gaseous constituent in disks, it is very challeng-ing to detect. Molecular hydrogen is an homonuclear moleculewhich means that its fundamental transitions are quadrupolar innature. Hence their Einstein spontaneous emission coe ffi cientsare small and produce only weak lines. For circumstellar disksanother challenge is that the weak H lines must be detected ontop of the strong dust continuum emission. In the mid-infrared,high spectral resolution instruments like VISIR at the VLT arerequired to disentangle these weak lines from the continuum.However, one should bear in mind a fundamental issue con-cerning the structure of a gas-rich optically thick disk. Molecularlines are produced in the hot upper surfaces of the disk wherethe molecular gas and accompanying dust are optically thin.
Send o ff print requests to : C. Martin-Za¨ıdi Therefore, molecular line emission is not sensitive to and doesnot probe the mid-plane interior layers of the disk because it isoptically thick. Because the amount of molecular gas in the opti-cally thin surface layers is small, the expected H line fluxes arevery weak. As an example of this, Carmona et al. (2008) calcu-lated the expected H line fluxes from typical Herbig Ae disksassuming a two-layer Chiang & Goldreich (1997) disk model, T gas = T dust , a gas-to-dust-ratio of 100, LTE emission and a dis-tance of 140 pc. They found that the expected line H fluxes aremuch fainter than the detection limits of current instrumentation.Indeed, numerous non-detections of H mid-IR pure rotationallines in the circumstellar environment of young stars have beenreported in the past few years (Bitner et al. 2008; Carmona et al.2008; Martin-Za¨ıdi et al. 2008b, 2009a). Nevertheless, lines canreach a detectable level if the gas-to-dust to ratio is allowed tobe higher than 100 and / or the T gas > T dust in the surface layersof the disk.Molecular hydrogen mid-IR lines have been detected in twoHerbig Ae / Be stars, namely HD 97048 and AB Aur, respectivelyobserved with VLT / VISIR (Martin-Za¨ıdi et al. 2007) and
TEXES
C. Martin-Za¨ıdi et al.: Where is the warm H ? Table 1.
Astrophysical parameters of the sample stars.
Star Sp. T ef f A v v rad( a ) d Disk Evidence forType (K) (mag) (km s − ) (pc) resolved gas-richenvironmentHD 142527 F6 IIIe 6300 (1) (1) -3.5 (2) (3) yes (4) HD 169142 A8 Ve 8130 (5) (6) -3.0 (7) (5) yes (8) yes (9)
HD 150193A A1 Ve 9300 (1) (1) -6.0 (10) (1) yes (11)
HD 163296 A1 Ve 9300 (1) (1) -4.0 (2) (1) yes (12) yes (13)
HD 100546 B9 Vne 10470 (1) (1) + (13) (1) yes (14) yes (13)( a ) radial velocity of the star in the heliocentric rest frame.References: (1) van den Ancker et al. (1998); (2) SIMBAD database; (3) de Zeeuw et al. (1999); (4)
Fukagawa et al. (2006); (5)
Acke et al. (2005); (6)
Malfait et al. (1998); (7)
Dunkin et al. (1997); (8)
Kuhn et al. (2001); (9)
Pani´c et al. (2008); (10)
Reipurth et al. (1996); (11)
Fukagawa et al.(2003); (12)
Grady et al. (2000); (13)
Lecavelier des Etangs et al. (2003); (14)
Augereau et al. (2001).
Table 2.
Summary of the observations.
Star t exp Airmass Optical Standard Airmass Optical Asteroid Airmass Optical(s) Seeing Star Seeing Seeing(”) (”) (”)HD 142527 3600 1.10-1.28 0.71-1.08 HD 211416 1.23-1.25 0.79-1.02 HEBE 1.19-1.20 0.97-1.06HD 169142 3600 1.17-1.50 0.71-1.01 HD 211416 1.23-1.25 0.79-1.02 HEBE 1.19-1.20 0.97-1.06HD 150193A 3600 1.31-1.88 0.85-1.40 HD 211416 1.24-1.25 0.81-0.99 PSYCHE 1.13-1.16 0.75-0.83HD 163296 3600 1.60-2.70 0.79-1.19 HD 211416 1.24-1.25 0.81-0.99 IRIS 1.00-1.01 0.99-1.28HD 100546 3600 1.44-1.52 0.73-1.09 HD 89388 1.31-1.32 0.61-0.73 HEBE 1.19-1.20 0.97-1.06The airmass and seeing intervals are given from the beginning to the end of the observations. (Bitner et al. 2007). These detections imply particular physicalconditions for the gas and dust, such as T gas > T dust , as men-tioned above. These conditions may be created by gas heatedby X-rays or UV photons in the surface layers of the disks. Wenote that these two detections indicate that the gas has not com-pletely dissipated in the inner part of these disks in a lifetime ofabout 3 Myrs (ages of the stars), while photoevaporation of thegas is expected to clear up this inner region within roughly thesame time ( < around HD 97048 was more likely distributed in an extendedregion within the inner disk, between 5 AU and 35 AU of thedisk, and Bitner et al. (2007) concluded from their observationof the disk of AB Aur that the H emission arised around 18 AUfrom the central star.We present new observations with the VISIR high-resolutionspectroscopic mode, to search for the mid-IR H emission lineat 17.035 µ m, the most intense pure rotational line observablefrom the ground, in a sample of well studied Herbig Ae / Be starsknown to harbor extended gas-rich circumstellar disks. The maingoal of these observations is to enlarge the global sample ofHAeBes observed at 17.035 µ m, and to better constrain the par-ticular physical conditions observed in the disks of HD 97048and AB Aur. In order to better understand the detections of H inthe disks of HD 97048 and AB Aur, we also analyze VLT / UVES spectra of these two stars to observe spectral lines of the CHand CH + molecules which are linked to the formation and ex-citation of H . In Sects. 3 and 4 we present the analysis of the VISIR spectra of the five HAeBes of our sample and relate it tothe context of the global search for molecular hydrogen at mid-IR wavelengths. We perfom a statistical analysis of the wholesample of HAeBes where H emission has been searched for in the mid-IR, and explore the possible link between the H and thedusty disk properties. Finally, we discuss the possible origin ofthe mid-IR H emission in the disks of HD 97048 and AB Aur.
2. Observation and data reduction
We selected a sample of five Herbig Ae / Be stars with spatiallyresolved disks (Table 1). They were selected using the follow-ing criteria: (i) proximity to the Sun ( < (ii) resolved disk and / or firm evidence of gas-richcircumstellar environment, and / or properties close to that ofHD 97048 and / or AB Aur, for which the 17.035 µ m line has al-ready been detected; (iii) well known Herbig Ae / Be stars with arich set of existing observations, particularly including gas-linesspectroscopy observations (Far-UV H lines, CO lines...; e.g.Thi et al. 2001; Martin-Za¨ıdi et al. 2008a; van der Plas et al.2009) and PAHs (Polycyclic Aromatic Hydrocarbons) features(e.g. Habart et al. 2004; Acke & van den Ancker 2006). Thefundamental parameters of each target are presented in Table 1.The target stars were observed on 2009 June 14 and 15,with the high spectral resolution long-slit mode of the VISIR ( ESO VLT Imager and Spectrometer for the mid-InfraRedLagage et al. 2004) instrument at the VLT. We focussed on theobservations of the S(1) pure rotational line of H at 17.035 µ mbecause it is the most intense line observable from the ground,and because it is the only line ever detected with VISIR in disksof Herbig Ae / Be stars (Martin-Za¨ıdi et al. 2007). The centralwavelength of the observations was thus set to be 17.035 µ m.We used the 0.75” slit, providing a spectral resolution of about14 000. The observation conditions are summarized in Table 2.In order to correct the spectra from the Earth’s atmospheric ab- . Martin-Za¨ıdi et al.: Where is the warm H ? 3 Fig. 1.
Spectra obtained for the H S(1) line at 17.035 µ m. Left panel: continuum spectra of the asteroid and of the target beforetelluric correction.
Middle panel: full corrected spectra: dotted lines show spectral regions strongly a ff ected by telluric features. Right panel: zoom of the region where the H lines should be observed (dashed vertical lines). A Gaussian of width FWHM = − and integrated line flux equal to the 3 σ line-flux upper limits is overplotted. The spectra were corrected neither for the radialvelocity of the targets nor the Earth’s rotation velocity. C. Martin-Za¨ıdi et al.: Where is the warm H ? sorption and to obtain the absolute flux calibration, asteroids andstandard stars were observed at nearly the same airmass andseeing conditions as the objects, just before and after observ-ing the science targets. For this purpose, we also used the mod-eled spectra of the standard stars (Cohen et al. 1999), and thoseof the asteroids (for HEBE and IRIS see Mueller & Lagerros1998, 2002, and for PSYCHE, Mueller 2009, priv. comm.). Fordetails on the observations and data reduction techniques seeMartin-Za¨ıdi et al. (2007) and Martin-Za¨ıdi et al. (2008b). To better understand the detections of H towards HD 97048and AB Aur, we observed spectral lines of the CH and CH + molecules in the optical range at high spectral resolution using UVES (ESO program numbers 075.C-0637 and 078.C-0774).The blue arm of
UVES (3700Å - 5000Å) gives acces to theelectronic transitions of the CH and CH + molecules that aregood tracers of the H formation and excitation (Federman 1982;Mattila 1986; Somerville & Smith 1989). The formation of CHis predicted to be controlled by gas-phase reactions with H .CH is thus a good tracer of H and their abundances are gen-erally strongly correlated. The formation of the CH + moleculethrough the chemical reaction C + + H needs a temperature ofabout 4500 K to occur. Thus, the CH + molecule is a probe ofhot and excited media, which could be interpreted as materialclose to the star, and this would allow us to better constrain theexcitation of H .The spectra were reduced using the UVES pipeline v3.2.0(Ballester et al. 2000) based on the ESO Common LibraryPipeline, and were corrected from the Earth’s rotation in order toshift them in the heliocentric rest frame. To analyze the absorp-tion lines in the observed spectra we used the O wens package(for details, see Lemoine et al. 2002; Martin-Za¨ıdi et al. 2008a),which allows us to derive the characteristics of the gaseous com-ponents, i.e. column density N, radial velocity v rad and broaden-ing parameter b .
3. Results
We observed five well studied Herbig Ae / Be stars with the highresolution spectroscopic mode of
VISIR to search for the H purerotational S(1) line. None of the observed sources show evidencefor H emission at 17.035 µ m (Fig. 1). In HD 169142, the ex-pected centroid for the H line corresponds to an emission fea-ture in the spectrum. However, this feature has a full-width-at-half maximum (FWHM) smaller than a spectral resolution ele-ment ( ∆ v ∼
21 km s − ), and an amplitude of about 1 σ that makesit insignificant in terms of detection (for details on the calcula-tion of the standard deviation σ , see Martin-Za¨ıdi et al. 2008b).We did not consider it further as a detection. As described pre-viously, we insist that mid-IR H lines only probe warm gaslocated in the surface layers of the disks because of the largeopacity in the interior layers due to the dust. The surface lay-ers of the disks, from which solid-state emission features alsoarise, surrounding our target stars therefore do not contain su ffi -cient warm gas to enable it to be detected in emission at mid-IRwavelengths.Then, we derived 3 σ upper limits on the integrated linefluxes and upper limits for the total column densities and massesof H as a function of the temperature for all the spectra by in-tegrating over a Gaussian of FWHM corresponding to the VISIR spectral resolution, and an amplitude of 3 σ , centered on the ex-pected wavelength for the S(1) line (see Table 3 and Fig. 2).For this purpose, we used the method detailed inMartin-Za¨ıdi et al. (2008b, 2009a), by assuming that the line isoptically thin at local thermodynamic equilibrium (LTE) and thatthe radiation is isotropic (for details on the formulae, see alsovan Dishoeck 1992; Takahashi 2001). Fig. 2.
Upper mass limits of optically thin H derived from H S(1) line flux limits as a function of the assumed LTE tempera-ture.From the 3 σ upper limits to the emission line flux, we cal-culated upper limits on the column density and mass of H foreach star. We found that the column densities should be lowerthan ∼ cm at 150 K, and lower than ∼ cm at 1000 K.The corresponding upper limits to the masses of warm gas inthe surface layers of the inner disk were estimated to be in therange from ∼ × − to ∼ M Jup (1 M Jup ∼ − M ⊙ ), assumingLTE excitation, and depending on the adopted temperature (seeTable 3 and Fig. 2). detectionstatistics In the past three years, numerous observations have beenperformed to observe the pure rotational mid-IR emissionlines of molecular hydrogen in the circumstellar environ-ments of HAeBes.
VISIR observations of 15 HAeBes havebeen conducted (Carmona et al. 2008; Martin-Za¨ıdi et al. 2007;Martin-Za¨ıdi et al. 2008b, and this work) and only one sourceout of the 15, namely HD 97048, presents a clear evidence forH emission at 17.035 µ m. In addition, in their sample of fiveHAeBes observed with TEXES at 17.035 µ m, Bitner et al. (2008)have reported only one detection of the S(1) H line in the diskof AB Aur. This leads to a detection statistics of the S(1) linefrom the ground of 10% only in disks about Herbig Ae / Be stars.However, this statistic su ff ers from the fact that the 3 σ limitsin our sample are inhomogeneous. Indeed, due to the di ff er-ent detection limits of each instrument (and exposure time foreach target), the 3 σ upper limits on the line flux from VISIR ob-servations cannot be directly compared to those obtained with
TEXES . However, our sample is quite representative of the vari-ety of HAeBes observable from the ground, including stars withspectral type from F6 to B2, with very young, massives or tran- . Martin-Za¨ıdi et al.: Where is the warm H ? 5 Table 3.
Results of the H S(1) line analysis.
Star λ (a) obs Integrated flux (b)
Intensity (b) N (H ) upper limits (b) H mass upper limits (b , c) HD ( µ m) (ergs s − cm − ) (ergs s − cm − sr − ) (cm − ) ( M Jup )150K 300K 1000K 150 K 300 K 1000 K142527 17.0351 < × − < × − × × × × − × − × − < × − < × − × × × × − × − × − < × − < × − × × × × − × − × − < × − < × − × × × × − × − × − < × − < × − × × × × − × − (a) expected position of the line in the observed spectra by correcting the radial velocity of the star for the epoch of observation.(b) 3 σ upper confidence limits.(c) masses of H are calculated assuming the distances quoted in Table 1. sitional disks. Therefore, the 10% statistics can be regarded as alower limit on the number of H detections with current ground-based mid-IR instrumentation. This is consistent with the mod-els by Carmona et al. (2008), who show that mid-IR H linescannot be detected with the existing instruments for disks underLTE conditions. Those authors estimated that sensitivities downto 10 − ergs s − cm − must be reached to detect the S(1) line,while VISIR typically reaches 10 − ergs s − cm − .The numerous non-detections of the S(1) line in circumstel-lar environments of HAeBes could imply that the physical con-ditions of the warm gas in most disks are consistent with thoseassumed in such a model, i.e. gas and dust well-mixed, a gas-to-dust ratio of about 100, and T gas = T dust .
4. Discussion and disk properties The challenge now is to explain the two detections of the S(1)line in the disks of HD 97048 and AB Aur. If the HD 97048and AB Aur disks trace a particular evolutionary stage, one mayexpect some connection between the mass of warm H and thedisk properties.The slope of the spectra corresponding to the ratio of the F and F indexes, i.e. continuum fluxes at 30 and 13 µ m respec-tively, has been used by Brown et al. (2007) to identify the popu-lation of cold disks. In addition, young flared disks tend to showrising spectra in the mid-IR with a much higher F / F ratiothan the more evolved, settled systems with self-shadowed diskswith much flatter mid-IR spectra (Dominik et al. 2003). The F / F ratio diagnostic is therefore not unique, but allows us toidentify a population with properties that depart from those ofmost young stars with disks. Indeed, in their sample of T Tauristars, Brown et al. (2007) have shown that the majority of theirsample had an emission that increased by a factor of 2.3 ± µ m, while the cold, possibly transitional, disksrose by factors of 5 − SpitzerSpace Telescope , by computing F which is the mean flux value(in Jy) in the range 13 µ m ± µ m, and F the mean fluxvalue (in Jy) between 29 µ m and 31 µ m (Table 4, for detailssee Olofsson et al. 2009). The data reduction was performedwith the “c2d legacy team pipeline” (Lahuis et al. 2006) withthe S18.7.0 pre-reduced (BCD) data.Figure 3 shows the upper limits of H mass measured fromthe S(1) line flux versus the slope of the spectra ( F / F ratio). Table 4. F / F indexes, and H mass upper limits at 600 K. Star F / F H mass and upperlimits at T =
600 K ( M Jup )(This work)HD 142527 4.86 < × − HD 169142 7.12 < × − HD 150193A 1.44 < × − HD 163296 1.54 < × − HD 100546 3.32 < × − (Martin-Za¨ıdi et al. 2008b)HD 98922 0.70 < × − HD 250550 2.67 < × − HD 259431 2.78 < × − HD 45677 (a) (Carmona et al. 2008)UX Ori (a)
HD 100453 5.41 < × − HD 101412 0.92 < × − HD 104237 1.28 < × − HD 142666 1.55 < × − (Bitner et al. 2008)49 Cet 2.31 < × − MWC 758 4.00 < × − V892 Tau 3.72 < × − VV Ser 0.76 < × − (Martin-Za¨ıdi et al. 2009a)HD 97048 (b) ± × − (Bitner et al. 2007)AB Aur (b) ± × − (a) Spitzer data not available.(b) masses of H (not upper limits) because the S(1) has been de-tected.We used distances and integrated fluxes values from the papers ref-erenced in the table. For comparison with our targets, we plotted all the stars observedat 17.035 µ m by VISIR and by
TEXES , including HD 97048 andAB Aur (Martin-Za¨ıdi et al. 2007; Martin-Za¨ıdi et al. 2008b,2009a; Carmona et al. 2008; Bitner et al. 2008). For this task,we considered only the upper limits on the mass of H at 600 Kfor each target, because the temperature of the observed H isaround 600 K for AB Aur (Bitner et al. 2007) and lower than600 K for HD 97048 (Martin-Za¨ıdi et al. 2009a).Most of our sample stars have an emission that increases bya median factor of 2.2 ± µ m, which is C. Martin-Za¨ıdi et al.: Where is the warm H ? Fig. 3.
Slope of the Spitzer spectra ( F / F ratio) versus uppermass limits of optically thin H at 600 K derived from H S(1)line flux limits. In this diagram are plotted all the Herbig Ae / Bestars observed at 17.035 µ m with TEXES and
VISIR . Trianglesrepresent the two stars for which the S(1) line has been detected.Squares represent the Herbig Be stars for which a circumstellardisk has never been clearly detected. Plain circles represent theAe stars of the sample (see text).fully consistent with the results of Brown et al. (2007). However,four stars in our sample, namely HD 169142, HD 100453,HD 142527, and HD 97048, have significantly higher F / F ratios. We stress that AB Aur is marginally higher than the bulk.In addition, the four spectra are also characterized by the ab-sence of 10 µ m amorphous silicate features. This is consistentwith the fact that the strong wings of the amorphous silicatefeature at 10 µ m increase the continuum flux around 13 µ m andthus decrease the F / F ratio. The F / F ratio of these fourstars correspond to the region of the plot where Brown et al.(2007) have identified the transitional cold disks in their sample,which also do not present the 10 µ m amorphous silicate featurein their spectra. However, HD 97048 is supposed to be a youngstar harboring a massive gas-rich flared disk (Lagage et al. 2006;Doucet et al. 2007), as compared to the other three stars thatare known to have more evolved and less massive disks thanHD 97048 (e.g. Grady et al. 2007). Its position in our Fig. 3raises numerous questions about its status. Below, we exploredi ff erent ways to understand the origin of the detected H andthe status of HD 97048 (and marginally AB Aur). relatedto PAHs? As we did for the F / F ratio, we searched for the possiblecorrelation between the presence and line flux of PAHs featuresand the upper limits on H mid-IR lines. For this task we com-pared the line flux of the PAHs feature at 8.6 µ m and 11.3 µ m(called 11 µ m complex due to the possible blend with features ofamorphous and crystalline silicate; e.g. Acke & van den Ancker2004) found in the literature (Table 5), with the H upper masslimits (Table 4). With regard to the F / F ratio, AB Aur andHD 97048 seem to depart from the bulk, and appear in the top ofthe plots where the PAHs line fluxes are high (Fig. 4 and Fig. 5).In the upper disk surface layers, photoelectric heating is verye ffi cient on small grains such as PAHs, and can play a signifi- Fig. 4.
Line flux of the PAHs feature at 8.6 µ m versus upper masslimits of optically thin H at 600 K derived from H S(1) line fluxlimits. Same legend as for Fig. 3.
Fig. 5.
Line flux of the PAHs complex at 11 µ m versus uppermass limits of optically thin H at 600 K derived from H S(1)line flux limits. Same legend as for Fig. 3.cant role in the gas heating process (Kamp & Dullemond 2004;Jonkheid et al. 2007). The PAHs emission depends on the geom-etry of the disk, i.e. the emission is proportional to the amountof PAHs that is directly illuminated by the UV light from thestar. Thus the PAHs emission depends both on PAHs abundanceand on geometry, and it is unclear which of these two e ff ectsis responsible for the higher PAH emission strength in AB Aurand HD 97048 compared to other Herbig stars in the sample.Irrespective of its origin, the strong emission of PAHs appearsrelated to H emission. Is there a link between their excitationand heating mecanisms ? This question must be studied further.As shown in Fig. 4 and Fig. 5, the spectra of the other starsof the sample present weaker PAH features. One could assumethat for these stars photoelectric heating is less e ffi cient and doesnot excite enough the H to be detected in the mid-IR range.In both plots HD 100546 also departs from the bulk. Thismay be because the disk of this star has a particular structure . Martin-Za¨ıdi et al.: Where is the warm H ? 7 Table 5.
Line fluxes of the PAHs features at 8.6 µ m and linefluxes of the PAHs complex at 11 µ m. star PAHs 8.6 (W / m ) Comp 11 (W / m ) refHD 142527 < × − × − × − × − < × − × − < × − × − × − × − × −
2, 3HD 250550 n n 4HD 259431 – – –HD 45677 – – –UX Ori < × − < × − < × − × − × −
2, 3HD 104237 < × − < × − < × − × −
149 Cet – – –MWC 758 – – –V892 Tau – – –VV Ser < × − × − × − × − × − × −
1– Data not available.“Y” (yes): presence of PAHs features but flux not measured,“N” (no): absence of PAHs features.References: 1: Acke & van den Ancker (2004); 2: Geers et al.(2006); 3: Geers et al. (2007); 4: Habart et al. (2004); with an inner ring, a gap between ∼ emissionarise fromthe envelope? Both HD 97048 and AB Aur are known to possess extendedenvelopes with a significant contribution in the various obser-vations of gas and dust (e.g. Hartmann et al. 1993; Grady et al.1999). Could the observed H mid-IR emission be due to thegaseous component of the envelopes of these stars?This scenario can certainly excluded because, firstly, the H S(1) lines observed in the spectra of HD 97048 and AB Aurare not spatially resolved, which constrains the emitting regionto be very close to the central stars. Given the weakness ofthe detected H emission, at least for AB Aur, it is also un-likely that had any extended emission is present. This wouldhave been detected. Secondly, electronic transitions of H havebeen detected in absorption with the FUSE ( Far UltravioletSpectroscopic Explorer ) satellite towards AB Aur, likely arisingfrom the extended envelope surrounding the star (Roberge et al.2001; Martin-Za¨ıdi et al. 2008a). The column density of the J = measured in the FUSE spectrum is 10 timeslower than that derived from the mid-IR observations, implyingthat we clearly do not observe the same regions around the star.The source HD 97048 has never been observed with FUSE ,which precludes any other analysis of the H lines than that of VISIR observations. We thus analyzed the spectra of HD 97048and AB Aur obtained with VLT / UVES . For the two stars wedetected absorption lines of the CH and CH + molecules (seeFig. 6), as well as absorption lines of Ca I, Ca II and K I. Due tothe high inclination angles of the disks with respect to the lines Table 6.
Column densities, radial velocities and b parameters ofCH and CH + for HD 97048 and AB Aur. Star log N(CH) log N(CH + ) v rad b (cm − ) (cm − ) (km s − ) (km s − )HD 97048 13.1 ± ± ± ± ± ± ± ± of sight, the observed gas cannot be in the disks, but likely arisesfrom the surrounding envelope. We measured the radial veloci-ties of these lines in the UVES spectra of the two stars (Table 6).All these lines in the
UVES spectra have the same radial velocity,implying that we probed a single gaseous component along eachline of sight. This is confirmed by the measured b -values of CHand CH + lines that are in the range expected for pure thermalbroadening, as generally seen in di ff use ISM clouds where typ-ical b -values are about 1.5 up to 3 km s − (e.g. Gry et al. 2002),and which would be much higher if several gaseous componentswere present along the line of sight. In addition, the radial ve-locities of the S(1) lines observed for HD 97048 and AB Aur,with VISIR and
TEXES respectively (Martin-Za¨ıdi et al. 2007;Bitner et al. 2007) di ff er by about 10 km s − from those mea-sured for the gas in the UVES spectra. This is an additional cluethat we probed di ff erent regions of the environment of the starsin the mid-IR and visible ranges, and that the H emission in themid-IR does not come from the envelope, but from the disk.Finally, it is important to note here that in their sampleof Herbig Ae / Be stars observed both with
FUSE and
UVES ,Martin-Za¨ıdi et al. (2009b,c) showed that when H and CHare both observed, their column densities are correlated inproportions consistent with those observed in the di ff use in-terstellar medium (N(CH) / N(H ) ∼ − , e.g. Welty et al. 2006;She ff er et al. 2008). It is important to note that for these stars,including AB Aur, the FUSE H data implied an interstellar ori-gin for the gas or a remnant of the parent molecular cloud and / orcircumstellar envelope (Martin-Za¨ıdi et al. 2008a). Here, the to-tal column densities of H derived from the mid-IR observationsgive a N(CH) / N(H ) ratio higher by a factor 10 to 10 than theinterstellar ratio, again suggesting a di ff erent location, the H de-tected in the mid-IR being in the disk, CH being in the envelope.Cold molecular hydrogen is very likely present in the envelopeas probed by FUSE , but does not possess the physical propertiesrequired to emit at mid-IR wavelengths.
An important tracer of gas in the inner disks of T Tauri andHerbig Ae / Be stars are the fundamental CO ro-vibrational emis-sion bands (e.g., ∆ v =
1) present at 4.7 µ m. CO emission at 4.7 µ m traces warm and hot gas at temperatures ranging from hun-dreds to a few thousands Kelvin (Najita et al. 2003). For HerbigAe stars, CO has been detected in almost all sources with op-tically thick disks (i.e. E(K-L) > C. Martin-Za¨ıdi et al.: Where is the warm H ? Fig. 6.
Fits of the CH line at 4300.3 Å in the
UVES spectra of HD 97048 (left) and AB Aur (right) with the O wens profile fittingprocedure. Stellar continuum: dashed line; intrinsic line profile: dotted line; resulting profile convolved with the line-spread function:thick line.cence for flaring disks (van der Plas et al. 2010, in prep). High-excitation CO emission lines (up to v = mid-IR emission detections, we find thatin all the Herbig Ae / Be stars where H mid-IR emission has beendetected their disks are flared, and CO emission at 4.7 µ m hasbeen detected too. Note however that the opposite is not true, asthis and other studies show, not all the sources with CO 4 . µ memission display H mid-IR emission. Bitner et al. (2008) showthat when H mid-IR is detected, the gas temperatures are sig-nificantly higher than the equilibrium temperatures expected forthe emitting regions, thus suggesting that the gas temperature ishigher than the dust and that processes such as UV-X-rays oraccretion heating may be important (a conclusion in agreementwith the conclusions of Carmona et al. 2008, when explainingnon-detections of H emission).If we extend the idea of an additional source of heating tocollisions (Bitner et al. 2008; Carmona et al. 2008) to analyzethe CO and H mid-IR detections together (note that mid-IRH traces warm gas and CO 4.7 µ m hotter gas), we may ex-pect that the sources with detected H mid-IR emission wouldhave CO emission at temperatures higher than the equilibriumtemperature for the respective emitting regions. In other words,we will expect for sources with H mid-IR emission to ob-serve CO lines at high excitation transitions. This argumen-tation works at least for HD 97048, where CO emission upto v = CO emission at 4.7 µ maround HD 97048, observed with the VLT / CRIRES, with ahomogeneous Keplerian disk. Their modeling of CO emissionyields an inner radius of 11 AU. Those authors concluded a de-pletion of CO in the inner region of the disk, probably due tophotodissociation. Our previous estimate of the location of theH mid-IR emission ( > = mid-IR emission with VISIR. But then, there is alarge gap (r ∼
14 AU) in the dust disk of HD 100546, with very little material inside (Bouwman et al. 2003; Benisty et al. 2010).This is likely di ff erent from HD 97048 and AB Aur.In addition, Brittain et al. (2003) conducted observationstoward AB Aur at 4.7 µ m of CO (v = ∼ > observed in absorption by FUSE , as well as forthe CH observed with
UVES . On the other hand, Bitner et al.(2007) showed that the H emission at 17 µ m in the disk ofAB Aur arised around 18 AU from the central star, with an exci-tation temperature of about 670 K. Their derived gas temperatureand distance from the central star fall between the hot and coldcomponents seen in the CO observations (Brittain et al. 2003),suggesting that the H and CO IR-emission does not come fromthe same region of the disk.It is hard to drive definitive conclusions based on very fewdetections. Further searches of high-excitation CO emission us-ing high-spectral resolution in sources with H mid-IR detec-tions (most notably AB Aur) and further searches of H mid-IRemission in sources with high-excitation CO will be required totest the idea whether H mid-IR emission and high-excitationCO emission are correlated. For instance, the main conclusionthat we can arrive at is that Herbig Ae stars appear to be in gen-eral a uniform group concerning CO µ m emission and H emis-sion in the sense that CO is detected in optically thick disks andthat H mid-IR emission is very weak (i.e. not detected), exceptfor those sources with flaring disks where an additional mech-anism heats the H to a detectable level. The use of full disksmodels (coupling gas and dust; Woitke et al. 2010) could helpus to understand the possible correlation between CO and H ,but this is beyond the scope of this paper. . Martin-Za¨ıdi et al.: Where is the warm H ? 9
5. Conclusion
We reported here on a search for the H S(1) emission lineat 17.035 µ m in the circumstellar environments of five wellknown HAeBes with the high resolution spectroscopic mode of VISIR . No source shows evidence for H emission at 17.035 µ m. From the 3 σ upper limits on the integrated line fluxes,we derived limits on column densities and masses of warmgas as a function of the temperature. The present work bringsto 18 the number of non-detections of the H S(1) line ina global sample of 20 Herbig Ae / Be stars observed with
VISIR and
TEXES (Martin-Za¨ıdi et al. 2007; Bitner et al. 2007;Carmona et al. 2008; Martin-Za¨ıdi et al. 2008b; Bitner et al.2008).The detections of H emission at 17.035 µ m by Bitner et al.(2007) and Martin-Za¨ıdi et al. (2007) show that at least a fewcircumstellar disks have su ffi ciently high H mid-infrared emis-sion to be observed from the ground. The most likely explana-tion for this is that the optically thin surface layers of the diskhas T gas > T dust and that the gas-to-dust ratio is higher than thecanonical ratio of 100 (Carmona et al. 2008; Martin-Za¨ıdi et al.2007). Indeed, in the surface layers of the disk, low densities ordust settling and coagulation may conduct to a spatial decou-pling between the gas and the dust. Photoelectric heating canthus play a significant role in the gas heating process and thephysical conditions may rapidly di ff er from the LTE ones, i.e., T gas > T dust (Kamp & Dullemond 2004; Jonkheid et al. 2007).On the other hand, UV and X-ray heating can be responsible forthe excitation of the observed gas and can heat the gas to temper-atures significantly hotter than the dust (Nomura & Millar 2005;Glassgold et al. 2007; Ercolano et al. 2008). In any case, onewould need to observe the lower H J -levels (i.e. J = J =
1) to definitively constrain the kinetic temperature of thegas, and better understand the excitation mechanisms responsi-ble of the mid-IR emission.As a second step, we also performed a statistical analysisof the whole sample of Herbig Ae / Be stars observed at 17.035 µ m with VISIR as well as with
TEXES . This analysis allowedus to identify a population of stars, including HD 97048 andmarginally AB Aur, with properties that depart from the bulkof our sample. This raises numerous questions about the originof the detected gas and the status of HD 97048 and AB Aur.From our
UVES observations we clearly demonstrated that theobserved mid-IR H emission does not come from the envelope,but from the disk.Are HD 97048 and AB Aur peculiar stars (and why)? Due tothe similarities of these two stars ( T e f f , age, mass, disk size...),one would expect that the physical conditions of their circum-stellar gas are typical of a particular (short) stage of evolutionof the disks. However, to confirm this assumption, we wouldneed to observe other similar HAeBes. Unfortunately, no othernearby Herbig Ae / Be star observable with the existing instru-ments presents the same observational properties. A global diag-nostic of the gaseous content of the disks is thus now required.To better constrain the physical and chemical properties of thegas, multi-wavelengths observations and a deep modeling wouldbe very useful. In this context, the
Herschel satellite will be veryhelpful because it will allow us to constrain the thermodynamicsin young disks by observing the gas and the dust simultaneously.
Acknowledgements.
This work is based on observations obtained at ESO / VLT(Paranal) with
VISIR , program number 083.C-0910, and with
UVES programnumbers 075.C-0637 and 078.C-0774. We warmly thank G. van der Plas forfruitful discussions about CO observations. C.M-Z. warmly thanks A. Smette(ESO) for discussions about
VISIR observations. We thank ANR for financialsupport through contract ANR-07-BLAN-0221 (Dusty Disks). A.C acknowledge support from a Swiss National Science Foundation grant (PP002–110504). C.M-Z. was supported by a CNES fellowship.
References
Acke, B. & van den Ancker, M. E. 2004, A&A, 426, 151Acke, B. & van den Ancker, M. E. 2006, A&A, 457, 171Acke, B., van den Ancker, M. E., & Dullemond, C. P. 2005, A&A, 436, 209Alexander, R. 2008, New Astronomy Review, 52, 60Augereau, J. C., Lagrange, A. M., Mouillet, D., & M´enard, F. 2001, A&A, 365,78Ballester, P., Modigliani, A., Boitquin, O., et al. 2000, The Messenger, 101, 31Benisty, M., Tatulli, E., M´enard, F., & Swain, M. R. 2010, A&A, 511, A75 + Bitner, M. A., Richter, M. J., Lacy, J. H., et al. 2007, ApJL, 661, L69Bitner, M. A., Richter, M. J., Lacy, J. H., et al. 2008, ApJ, 688, 1326Blake, G. A. & Boogert, A. C. A. 2004, ApJL, 606, L73Bouwman, J., de Koter, A., Dominik, C., & Waters, L. B. F. M. 2003, A&A, 401,577Brittain, S. D., Rettig, T. W., Simon, T., et al. 2003, ApJ, 588, 535Brittain, S. D., Simon, T., Najita, J. R., & Rettig, T. W. 2007, ApJ, 659, 685Brown, J. M., Blake, G. A., Dullemond, C. P., et al. 2007, ApJL, 664, L107Carmona, A., van den Ancker, M. E., Henning, T., et al. 2008, A&A, 477, 839Chiang, E. I. & Goldreich, P. 1997, ApJ, 490, 368Cohen, M., Walker, R. G., & Witteborn, F. C. 1999, LPI Contributions, 969, 5de Zeeuw, P. T., Hoogerwerf, R., de Bruijne, J. H. J., Brown, A. G. A., & Blaauw,A. 1999, AJ, 117, 354Dominik, C., Dullemond, C. P., Waters, L. B. F. M., & Walch, S. 2003, A&A,398, 607Doucet, C., Habart, E., Pantin, E., et al. 2007, A&A, 470, 625Dullemond, C. P. & Dominik, C. 2004, A&A, 417, 159Dunkin, S. K., Barlow, M. J., & Ryan, S. G. 1997, MNRAS, 286, 604Ercolano, B., Drake, J. J., Raymond, J. C., & Clarke, C. C. 2008, ApJ, 688, 398Federman, S. R. 1982, ApJ, 257, 125Fukagawa, M., Tamura, M., Itoh, Y., Hayashi, S. S., & Oasa, Y. 2003, ApJL, 590,L49Fukagawa, M., Tamura, M., Itoh, Y., et al. 2006, ApJL, 636, L153Geers, V. C., Augereau, J., Pontoppidan, K. M., et al. 2006, A&A, 459, 545Geers, V. C., van Dishoeck, E. F., Visser, R., et al. 2007, A&A, 476, 279Glassgold, A. E., Najita, J. R., & Igea, J. 2007, ApJ, 656, 515Grady, C. A., Devine, D., Woodgate, B., et al. 2000, ApJ, 544, 895Grady, C. A., Schneider, G., Hamaguchi, K., et al. 2007, ApJ, 665, 1391Grady, C. A., Woodgate, B., Bruhweiler, F. C., et al. 1999, ApJL, 523, L151Gry, C., Boulanger, F., Nehm´e, C., et al. 2002, A&A, 391, 675Habart, E., Natta, A., & Kr¨ugel, E. 2004, A&A, 427, 179Hartmann, L., Kenyon, S. J., & Calvet, N. 1993, ApJ, 407, 219Jonkheid, B., Dullemond, C. P., Hogerheijde, M. R., & van Dishoeck, E. F. 2007,A&A, 463, 203Kamp, I. & Dullemond, C. P. 2004, ApJ, 615, 991Kuhn, J. R., Potter, D., & Parise, B. 2001, ApJL, 553, L189Lagage, P.-O., Doucet, C., Pantin, E., et al. 2006, Science, 314, 621Lagage, P. O., Pel, J. W., Authier, M., et al. 2004, The Messenger, 117, 12Lahuis, F., Kessler-Silacci, J. E., Evans, N. J., I., et al. 2006, c2d SpectroscopyExplanatory Supplement, Tech. rep., Pasadena: Spitzer Science CenterLecavelier des Etangs, A., Deleuil, M., Vidal-Madjar, A., et al. 2003, A&A, 407,935Lemoine, M., Vidal-Madjar, A., H´ebrard, G., et al. 2002, ApJS, 140, 67Malfait, K., Bogaert, E., & Waelkens, C. 1998, A&A, 331, 211Martin-Za¨ıdi, C., Deleuil, M., Le Bourlot, J., et al. 2008a, A&A, 484, 225Martin-Za¨ıdi, C., Habart, E., Augereau, J.-C., et al. 2009a, ApJ, 695, 1302Martin-Za¨ıdi, C., Lagage, P.-O., Pantin, E., & Habart, E. 2007, ApJL, 666, L117Martin-Za¨ıdi, C., Le Bourlot, J., Roue ff , E., Hily-Blant, P., & Gry, C. 2009b,in SF2A-2009: Proceedings of the Annual meeting of the French Society ofAstronomy and Astrophysics, held 29 June - 4 July 2009 in Besanc¸on, France.Eds.: M. Heydari-Malayeri, C. Reyl´e and R. Samadi, p.295, ed. M. Heydari-Malayeri, C. Reyl’E, & R. Samadi, 295– + Martin-Za¨ıdi, C., Le Bourlot, J., Roue ff , E., Hily-Blant, P., & Gry, C. 2009c,A&AMartin-Za¨ıdi, C., van Dishoeck, E. F., Augereau, J., Lagage, P., & Pantin, E.2008b, A&A, 489, 601Mattila, K. 1986, A&A, 160, 157Meeus, G., Waters, L. B. F. M., Bouwman, J., et al. 2001, A&A, 365, 476Mueller, T. G. & Lagerros, J. S. V. 1998, A&A, 338, 340Mueller, T. G. & Lagerros, J. S. V. 2002, A&A, 381, 324Najita, J., Carr, J. S., & Mathieu, R. D. 2003, ApJ, 589, 931Nomura, H. & Millar, T. J. 2005, A&A, 438, 923Olofsson, J., Augereau, J., van Dishoeck, E. F., et al. 2009, A&A, 507, 327Pani´c, O., Hogerheijde, M. R., Wilner, D., & Qi, C. 2008, A&A, 491, 219 ? Reipurth, B., Pedrosa, A., & Lago, M. T. V. T. 1996, A&AS, 120, 229Roberge, A., Lecavelier des Etangs, A., Grady, C. A., et al. 2001, ApJL, 551,L97She ffff