Imaging the Ionized Disk of the High-Mass Protostar Orion-I
aa r X i v : . [ a s t r o - ph ] A p r Imaging the Ionized Disk of the High-Mass Protostar Orion–I
M. J. Reid
Harvard–Smithsonian Center for Astrophysics, 60 Garden Street, Cambridge, MA 02138 [email protected]
K. M. Menten
Max-Planck-Institut f¨ur Radioastronomie, Auf dem H¨ugel 69, D-53121 Bonn, Germany [email protected]
L. J. Greenhill
Harvard–Smithsonian Center for Astrophysics, 60 Garden Street, Cambridge, MA 02138 [email protected]
C. J. Chandler
National Radio Astronomy Observatory, P.O. Box 0, Socorro, NM 87801 [email protected]
ABSTRACT
We have imaged the enigmatic radio source-I (Orion–I) in the Orion-KL neb-ula with the VLA at 43 GHz with 34 mas angular resolution. The continuumemission is highly elongated and is consistent with that expected from a nearlyedge-on disk. The high brightness and lack of strong molecular lines from Orion–I can be used to argue against emission from dust. Collisional ionization and H − free-free opacity, as in Mira variables, require a central star with > ∼ L ⊙ , whichis greater than infrared observations allow. However, if significant local heatingassociated with accretion occurs, lower total luminosities are possible. Alterna-tively, photo-ionization from an early B-type star and p + / e − bremsstrahlung canexplain our observations, and Orion–I may be an example of ionized accretiondisk surrounding a forming massive star. Such accretion disks may not be ableto form planets efficiently. Subject headings: infrared: stars — ISM: individual (Orion Kleinmann-Low) —stars: individual (I,IRc 2) — stars: formation — planetary systems: formation 2 –
1. Introduction
The nearest massive star forming region, the Kleinmann-Low (KL) nebula in the Orionmolecular cloud, is at a distance 480 pc (Genzel et al. 1981). While the brightest near-infrared source in Orion–KL (Kleinmann & Low 1967) is the Becklin-Neugebauer (BN)object, it contributes only a small fraction of the total nebular luminosity of ∼ L ⊙ (seeThronson et al. 1986 and references therein). Other young stellar objects (YSOs) are deeplyembedded and hidden from view at infrared (IR) wavelengths. An object giving rise to strongmid-IR emission, IRc 2, has been long suspected to be the dominant energy source in Orion–KL (Downes et al. 1981; Genzel et al. 1981; Shuping, Morris & Bally 2004), but hidden by >
60 mag of visual extinction (Gezari, Backman & Werner 1998; Greenhill et al. 2004b).However, IRc 2 breaks-up into several compact regions (Dougados et al. 1993; Greenhill et al.2004b). Moreover, Gezari (1992) and Menten & Reid (1995) showed that IRc 2 is offset fromthe compact radio source-I (hereafter Orion–I), and some components of IRc 2 could arisefrom reflected light. Orion–I is very deeply embedded and Greenhill et al. (2004b) estimateoptical depths >
300 at 8 and 22 µ m wavelengths.The radio source Orion–I is an enigmatic object. Proper motion measurements suggestthat Orion–I might have been part of a multiple system that disintegrated ≈
500 years ago(G´omez et al. 2005). Strong H O and OH (Cohen et al. 2006) masers are concentrated nearOrion–I, and the H O masers are distributed in an elongated pattern along position angle(PA) ≈ ◦ (East of North). The H O masers seem to be expanding about a central posi-tion in the general vicinity of Orion–I (Genzel et al. 1981). Orion–I also displays strong SiOmasers, which are usually associated with evolved asymptotic giant branch (AGB) or super-giant stars and are very rare in star forming regions (Hasegawa et al. 1985). Interferometricmaps of the SiO masers with the BIMA array at an angular resolution of ≈ ′′ indicatedthe possibility that they came from a rotating and expanding disk (Plambeck et al. 1995).Higher resolution observations with the VLA (Menten & Reid 1995) located Orion–I at thecenter of the SiO masers. Menten & Reid argued that the presence of vibrationally-excitedSiO masers, which require temperatures exceeding 1000 K (e.g., Lockett & Elitzur 1992) ata radius of ≈
50 AU from Orion–I, was strong evidence that it must be a very luminousobject ( ∼ L ⊙ ).Observations by Greenhill et al. (1998) with the VLBA at a resolution of ∼ ′′ “X”-like pattern.Since an X-like pattern has two symmetry axes, two models for the SiO masers have beenforwarded. The SiO masers could form in the limbs of a high velocity bi-conical outflowprojected along a NW–SE axis (Greenhill et al. 1998; Doeleman, Lonsdale & Pelkey 1999).In this model, the western (eastern) arms are moving away from (toward) the observer and 3 –hence are red-shifted (blue-shifted) with respect to the systemic velocity. Alternatively, theSiO masers could form in material being expelled from a rotating disk, whose spin axis isprojected NE–SW. In this model, the western (eastern) arms are moving away from (toward)the observer owing to rotation (Greenhill et al. 2004a). As discussed in §
3, evidence favorsthe latter model of a rotating, nearly edge-on, disk centered on Orion–I.We have used the VLA at 43 GHz and imaged the radio continuum emission associatedwith Orion–I and located it precisely toward the center of the SiO maser X-like pattern.These data together give the clearest picture yet presented of the nature of a high-mass, YSO,disk on scales of ≈
16 AU. In this paper we concentrate on understanding the properties andnature of the continuum emission from the disk-like structure of Orion–I.
2. Observations & Results
Our observations were made on 2000 November 10 with the NRAO Very Large Array(VLA) in a manner similar to those of Menten & Reid (1995). All 27 antennas had 40–50GHz receivers, compared to our previous observations with only 9 antennas. The recentlyinstalled receivers had better noise performance than the ones available in 1994 and wereplaced on antennas providing longer interferometer baselines. Thus, the present data yieldedlower noise levels and higher angular resolution compared to our previous data.In order to image the continuum emission from Orion–I and to locate this emissionwith respect to the SiO masers, we employed a dual-band continuum setup with a narrowband (1.56 MHz), covering red-shifted SiO v= 1, J=1–0 masers between LSR velocities of10.8 – 21.3 km s − (assuming a rest frequency of 43122.08 MHz), and a broad band (50MHz), centered at 43164.9 MHz on a line-free portion of the spectrum. Both frequencybands were observed in dual-circular polarizations. We observed from 0 h m to 10 h m localsidereal time. Absolute flux density calibration was obtained from an observation of 3C286,assuming the flux density spectrum of Baars et al. (1977). Observations of the quasar 0501–019, measured to be 0.74 Jy, were interspersed with Orion–I to monitor gain variations andto determine electronic phase-offsets between the bands.The narrow-band data were then “self-calibrated” with the very strong maser signal asa phase reference. The phase and amplitude corrections were then applied to the broad-band data, and a high quality map of the continuum emission was produced. A detailed The National Radio Astronomy Observatory (NRAO) is operated by Associated Universities, Inc., undera cooperative agreement with the National Science Foundation.
AIPS ) task
IMAGR . We producedmaps with two different weightings of the (u,v)-data, shown in Fig. 1. Using
IMAGR weight-ing parameter “
ROBUST =5” the dirty beam was 58 ×
45 mas at a PA of − ◦ , and werestored the image with a round beam of 50 mas (approximately the geometric-mean size).Using “ ROBUST =0” the dirty beam was 41 ×
28 mas at PA of − ◦ , and we restored theimage with a round beam of 34 mas. At a distance of 480 pc, 34 mas corresponds to 16 AU.These maps had rms noise levels of 0.13 mJy beam − and 0.14 mJy beam − , respectively.In order to image the SiO maser emission at high spectral resolution, several scans inspectral-line mode were interspersed with the dual-band continuum observations. We coveredall of the SiO maser emission with a bandwidth of 6.25 MHz and 128 spectral channels,which provided a velocity resolution of 0.34 km s − . The line data were self-calibrated bychoosing a channel with strong emission as a reference, and the resulting phase and amplitudecorrections were applied to the other channels. Scans of the strong extragalactic continuumsources 3C 84 and 3C 273 provided bandpass calibration. We produced a spectral-line datacube, which we restored with a 30 mas beam, taking advantage of the high signal-to-noiseratio and slightly “over-resolving” the dirty beam of 43 ×
27 mas.Alignment of the continuum and maser emission to about 5 mas accuracy was achievedby producing a pseudo-continuum map from the line data, using spectral channels coveringthe velocities that were within the 1.56 MHz passband of the dual-band continuum setup,and comparing it with the maps obtained from the dual-band data. Based on this cross-registration, we show SiO maser maps, smoothed to 5.43 km s − resolution, superposed onthe continuum emission of Orion–I in Fig. 2. This figure shows that the continuum emissionis precisely centered among the four SiO maser arms, whose innermost ends nestle tightlyagainst the disk-like continuum structure.
3. Results & Discussion
Our images of the continuum emission from Orion–I at 43 GHz are shown in Fig. 1.The top and middle panels of the figure are maps made with resolutions of 50 and 34 mas,respectively. The total flux density of Orion–I is 13 mJy and the peak brightness at 34 masresolution is 3.0 mJy beam − . The emission appears to be composed of a compact component,near the center of the source, and a component elongated NW–SE. Assuming the elongatedcomponent is approximately uniformly bright, it would contribute about 0.8 mJy beam − atthe center of the source, leaving 2.2 mJy for the compact component. Subtracting a 0.8 mJy 5 –point-source centered at the position of peak brightness, we obtain the image shown in thebottom panel of Fig. 1. This reveals a disk-like feature with a radius of ≈
35 AU and abrightness of about 1 mJy beam − . Away from the center of the source, the true brightnessis a lower limit, since the feature is not well resolved perpendicular to its elongation. Wealso note that the peak brightness along the disk-like feature does not follow a straight lineon the sky. Instead, it appears to bend as might be expected from a warped disk.In the following discussion, we assume that the compact emission comes from the im-mediate environment of a YSO, and that the elongated component traces a nearly edge-ondisk, whose spin axis is aligned northeast–southwest. Briefly, the evidence supporting thismodel is as follows. Greenhill et al. (2004a) report detection of a curved arc of SiO maseremission bridging the gap between the base of the south and west arms. Evidence of thisemission can also be seen in the − .
47 and 12 . − channel maps (Fig. 2). The bridgeemission displays a radial velocity gradient, and some features have tangential proper mo-tions, consistent with material rotating close to the nearside of a disk. Such emission isnot anticipated for a bipolar outflow. Additionally, Greenhill et al. (2004a) note that H Omaser emission comes from “caps” displaced predominantly 0. ′′ ′′ ◦ ). It is possible that a weak jet emanates from the YSO,perpendicular to the disk, resulting in the extended appearance at the center. Clearly, highersensitivity observations are needed to understand this structure.What are the physical conditions in the Orion–I source? To answer this question,one must know the emission mechanism (opacity source) for the cm-wave photons. Thecm-to-mm wavelength spectrum of the entire source can be characterized as a power lawwith flux density, S ν , rising with observing frequency, ν , as S ν ∝ ν . (Menten & Reid1995; Beuther et al. 2006), approaching that of a black body. Since the source is not wellresolved spatially at lower frequencies with the VLA, the spectral index does not allow usto discriminate between an inhomogeneous, single-component model (where the spectralindex is shallower than 2.0, because unity optical-depth occurs at a smaller radius at higherfrequencies) and a two-component model (with an optically thick central component and apartially optically thin disk-like structure).We think it unlikely that dust emission could be a dominant contributor to the cm- tomm-wavelength emission of Orion–I. A dense, warm, dusty disk would be expected to show 6 –a plethora of molecular lines at mm/sub-mm wavelengths. While Beuther et al. (2006) findnumerous, strong, molecular lines toward the nearby “hot core,” they find no strong linestoward the position of Orion–I (only weak SO lines and, of course, the strong SiO masersslightly offset from Orion–I). Thus, we look to other emission mechanisms to explain boththe YSO peak and the elongated disk components.The observations could be modeled with gas at ≈ < − free-free Opacity In several ways Orion–I appears similar to a Mira-like variable star. Such stars displayOH, H O, and SiO masers, as seen in Orion–I. In addition, Mira variables have continuumemission detectable with the VLA at cm wavelengths, with brightness temperatures of ≈ ≈ .
9) “radio photosphere” with characteristic temperature of ≈ ∼ cm − . Under these conditions the dominant opacity source is H-minusfree-free interactions, coming from free electrons interacting with neutral hydrogen (eitheratomic or molecular). This is analogous to normal proton-electron bremsstrahlung, exceptthat the interaction is about 10 weaker (Dalgarno & Lane 1966), requiring correspondinglyhigher densities. At these temperatures and densities, sufficient free electrons can be createdby collisional ionization of Na and K (Reid & Menten 1997).The SiO v= 1, J=1–0 maser emission at 43 GHz originates from the first vibrationally-excited state at ≈ ≈ ∼ − cm − for strong maser action(Alcolea, Bujarrabal & Gallego 1989; Lockett & Elitzur 1992; Bujarrabal 1994). Since thecontinuum emission region is more compact and has a higher (brightness) temperature thanrequired for SiO maser excitation, finding the loci of SiO maser emission extending outwardfrom the continuum, as shown in Fig. 2, is reasonable and as observed for Mira-like variables(Reid & Menten 2003). As both Miras and Orion–I display similar cm-to-mm wavelength 7 –spectral indexes and have a similar configuration of continuum and maser emission, there iscircumstantial evidence for similar physical conditions and mechanisms.We have explored models for the disk-like emission of Source-I owing to H − free-freeopacity. Assuming conditions similar to a Mira-like radio photosphere, material at density ∼ cm − and temperature ≈ ≈ ≈ . > ∼
450 K. This path length is about 10% of the disk radiusand, thus, is easily achievable. Such a model has the benefit that a single power-law canexplain the observed spectral energy distribution for the entire Orion–I source between 8 and350 GHz. However, recently Beuther et al. (2006) have measured the flux density of Orion–Iat 690 GHz to be between 3.5 and 9.9 Jy. Since the extrapolation of the cm-wave continuumspectrum to 690 GHz predicts under 2 Jy, an additional component (e.g., dust on a scale of0.2 ′′ to 2 ′′ ) seems required to explain the sub-mm wavelength spectrum of Orion–I, makinga single emission mechanism unlikely.For Orion–I, we observe a disk-like component that extends to about 0 . ′′ ( ≈
40 AU)from the star. We have attempted to model the brightness profile of such a disk in amanner similar to that done for the radio photospheres of Miras (Reid & Menten 1997),but with a disk geometry. Specifically, we assume an edge-on disk that is centrally heatedand is optically thick to most of the radiation from the YSO. In Fig. 3 we plot the observedbrightness temperature in the map with 34 mas resolution (middle panel of Fig. 1) as afunction of position along the disk elongation. The physical parameters of the central starand disk are listed in Table 1 for model A. In Fig. 3 we overplot the model brightness(blue dotted line) convolved with the observed restoring beam. While model A providesa reasonable fit to the observations, the model requires the central star to have a totalluminosity of ∼ × L ⊙ and a disk mass of ∼ ⊙ . These are general characteristicsof this class of models and are not sensitive to details of the parameters. Note that asimilarly large luminosity may also be required to explain the SiO masers (Menten & Reid1995). Reducing the stellar luminosity requires substantially increasing the disk mass (inorder to increase opacity and maintain a high disk brightness temperature). Thus, such amodel requires a fairly massive disk and a luminosity exceeding that from the IRc 2 region(Gezari, Backman & Werner 1998; Greenhill et al. 2004b) and, perhaps, even that of theentire Orion–KL nebula (Thronson et al. 1986). Since other energetic sources exist nearby(e.g., Source-n), we conclude that the disk component of Orion–I probably is not thermally(collisionally) excited by a central source.While central heating of the disk component (and also the SiO masers) may be ruled outon energetic grounds, the material in the disk and the SiO masers may be partially locally 8 –heated by accretion processes. Dissipation of energy in the disk could raise the temperatureabove that allowed by radiative equilibrium with the central star. Since the volume of thedisk can be considerably less than that of a sphere of the same radius, increasing the disktemperature in this manner can require less total energy than for central heating alone.Thus, we evaluated models that allowed the disk temperature to fall with radius, r , moreslowly than r − / . One such model, described in Table 1 as model B and shown in Fig. 3,fits the data well and requires a central star luminosity of 5 × L ⊙ , comfortably below theobservational limits.As pointed out by Menten & Reid (1995), the SiO maser excitation also requires a veryhigh luminosity source, were it to be centrally heated (i.e., assuming radiative equilibrium: L = σT πr ). However, infalling material might interact with outflowing material (andpossibly magnetic fields) in the conical walls of a bi-polar outflow. This may add heat,augmenting the central source and providing the necessary high temperatures ( ≈ p + / e − Bremsstrahlung
Given the high luminosities and disk masses characteristic of models involving thermalionization and H − free-free opacity, we now consider a hotter central star. The observedbrightness can be modeled with an early B-type star, which photo-ionizes a moderate densityplasma. Indeed, since proton-electron bremsstrahlung is ∼ times stronger than H − free-free per interaction, the disk plasma need only have a density ∼ − times lower.Assume that Orion–I contains a hot central star and a photo-ionized disk (or a photo-ionized surface layer). The disk may contain a neutral central layer, which provides a reser-voir for material that can be photo-ionized by the YSO (Hollenbach et al. 1994). The sub-mm wave spectrum of Orion–I, measured by Beuther et al. (2006), suggests a dust compo-nent dominates above 300 GHz, leaving a bremsstrahlung spectrum that becomes opticallythin above ≈
100 GHz. Such a turnover frequency can come from an electron density of ∼ cm − over a path length of 35 AU, comparable to the observed radius. These param-eters yield an excitation parameter, U , of ∼
10 pc cm − , which could come from a ZAMSB0–B1 star (Panagia 1973) of ≈
10 M ⊙ and a luminosity approaching 10 L ⊙ . A star of > ∼ ⊙ is consistent with the rotation and expansion seen in VLBA maps of the SiO masers(Greenhill et al. 2004a; Cunninghan, Frank & Hartmann 2005).In Fig. 3 (dashed green line), we present a simple model of a brightness temperatureprofile along an edge-on, photo-ionized disk with a constant temperature. The physical pa- 9 –rameters of the star and disk are given in the Table 1 as model C. This model provides areasonable fit to the data, demonstrating that a photo-ionized disk can explain our observa-tions.Recently, Keto (2002, 2003) and Keto & Wood (2006) have shown that the inner por-tion of a disk can be fully ionized and still allow for continued accretion onto a massiveprotostar. They point out that inside a critical radius r c = GM/ c s , where G is the gravi-tational constant, M is the mass of the central protostar, and c s is the sound speed in the(neutral or ionized) material, the protostar’s gravity exceeds the thermal pressure. For thestellar parameters given above for a ZAMS B0–B1 star, ionized accretion can proceed insideof the critical radius of about 25 AU. This critical radius is similar to the 35 AU radius ofthe disk observed at 43 GHz in Orion–I.Keto (2007) explored models of ionized accretion in the presence of significant angularmomentum. The example shown in the right-hand panel of his Fig. 1 corresponds to a starof 20 M ⊙ , an ionizing flux of 3 × photons s − (approximately a B0 – 09.5 star), andaccreting material with specific angular momentum of 0.16 km s − pc. The resulting criticalradius for accretion of ionized gas is 54 AU. Scaling this to a 10 M ⊙ star gives a critical radiusof 27 AU, reasonably consistent with our observations.An ionized accretion disk offers a natural explanation for the dearth of sub-mm wave-length spectral-lines, observed by Beuther et al. (2006) toward Orion–I, that would otherwisebe expected for a dense and warm neutral disk. Thus, Orion–I may be a good example of amassive YSO accreting material after ionizing its inner accretion disk. If the disk is main-tained at a temperature greater than ≈
4. Other High-mass Protostars
While Orion–I is probably the nearest high-mass protostellar object, more distant can-didates include CRL 2136, W 33, AFGL 2591, and NGC 7538/IRS 9. These objects displaycm-to-mm wavelength spectra closely resembling that of Orion–I, and whose continuum emis-sions are unresolved (or only marginally resolved) at 40 mas resolution (Menten & van der Tak 10 –2004; van der Tak & Menten 2005). While none of these candidates have been observed tohave SiO maser emission, Menten & van der Tak (2004) find that CRL 2136 has H O masersvery close (in projection) to the continuum emission; the masers might arise in dense, hotgas following an accretion shock.
5. Future Observations
Our current image of Orion–I, at a resolution of 34 mas ( ≈
16 AU), appears to show anionized disk around a massive YSO. While this may be the best image to date of such a disk,our data are limited both in sensitivity and angular resolution. We have only about five res-olution elements along the disk, and we have not clearly resolved the emission perpendicularto the disk. In order to improve significantly the sensitivity, of this image, we probably mustawait the completion of the EVLA phase-I project. Improved angular resolution, with therequired higher sensitivity, could be achieved with the planned increase in baseline length ofthe EVLA phase-II or, in the long-term, with the Square Kilometer Array.We thank L. Matthews for suggestions to improve the manuscript.
REFERENCES
Alcolea, J., Bujarrabal, V. & Gallego, J. D. 1989, A&AS, 211, 187Baars, J. W. M., Genzel, R., Pauliny-Toth, I. I. K & Witzel, A. 1977, A&AS, 61, 99Beuther, H. et al.2006, ApJ, 636, 323Bujarrabal, V. 1994, A&AS, 285, 953Cohen, R. J., Gasiprong, N., Meaburn, J. & Graham, M. F. 2006, MNRAS, 367, 541Cunningham, A., Frank, A. & Hartmann, L. 2005, ApJ, 631, 1010Dalgarno, A. & Lane, N. F. 1966, ApJ, 145, 623Dougados, C, L´ena, P., Ridgway, S. T., Christou, J. C. & Probst, R. G. 1993, ApJ, 406, 112Doeleman, S. S., Lonsdale, C. J. & Pelkey, S. 1999, ApJ, 510, L55Downes, D., Genzel, R., Becklin, E. E. & Wynn-Williams, G. C. 1981, ApJ, 244, 869 11 –Genzel, R., Reid, M. J., Moran, J. M. & Downes, D. 1981, ApJ, 244, 884Gezari, D. Y. 1992, ApJ, 396, L43Gezari, D. Y., Backman, D. E. & Werner, M. W. 1998, ApJ, 509, 283G´omez, L., Rodr´ıguez, L. F., Loinard, L., Lizano, S., Poveda, A. & Allen, C. 2005, ApJ,635, 1166Greenhill, L. J., Gwinn, C. R., Schwartz, C., Moran, J. M. & Diamond, P. J. 1998, Nature,416, 59Greenhill, L. J., Chandler, C. J., Reid, M. J., Diamond, P. J. & Moran, J. M. 2004a, in“Star Formation at High Angular Resolution,” Proceedings of IAU Symposium 221,eds. M. G. Burton, R. Jayawardhana & T. L. Bourke, (San Francisco: ASP), p. 155Greenhill, L. J., Gezari, D. Y., Danchi, W. C., Najita, J., Monnier, J. D. & Tuthill, P. G.2004b, ApJ, 605, L57Greenhill, L. J., Chandler, C. J., Reid, M. J., Diamond, P. J. & Moran, J. M. 2007, inpreparationHasegawa, T. et al.1985, in “Masers, Molecules and Mass Outflows in Star Forming Regions,”ed. A. Haschick (Haystack Obs., Westford, MA), p. 275Hollenbach, D., Johnstone, D., Lizano, S. & Shu, F. 1994, ApJ, 428, 654Keto, E. 2002, ApJ, 580, 980Keto, E. 2003, ApJ, 599, 1196Keto, E. 2007, submitted to ApJKeto, E. & Wood, K. 2006, ApJ, 637, 850Kleinmann, D. E. & Low, F. J. 1967, ApJ, 149, L1Lockett, P. & Elitzur, M. 1992, ApJ, 399, 704Menten, K. M. & Reid, M. J. 1995, ApJ, 445, L157Menten, K. M. & van der Tak, F. F. S. 2004, A&A, 414, 289Morino, J.-I., Yamashita, T., Hasegawa, T. & Nakano, T. 1998, Nature, 393, 340 12 –Panagia, N. 1973, AJ, 78, 929Plambeck, R. L., Wright, M. C. H., Mundy, L. G. & Looney, L. W. 1995, ApJ, 455, L189Reid, M. J. & Menten, K. M. 1997, ApJ, 476, 327Reid, M. J. & Menten, K. M. 2003, in Mass-losing pulsating stars and their circumstellarmatter, Astrophysics and Space Science Library, Vol. 283, eds. Y. Nakada, M. Honmaand M. Seki. (Kluwer Academic Publishers, Dordrecht), p. 283Shuping, R. Y., Morris, M. & Bally, J. 2004, ApJ, 128, 363Thronson, H. A. et al.1986, AJ, 91, 1350van der Tak, F. F. S. & Menten, K. M. 2005, A&A, 437, 947
This preprint was prepared with the AAS L A TEX macros v5.2.
13 –Table 1. Star & Disk Model Parameters
Parameter Units Model A Model B Model COpacity source .... H − free-free H − free-free p + / e − bremsstrahlungStellar Radius ( R ∗ ) .... AU 4 4 0.25Disk Temperature at R ∗ .. K 4500 3000 8000Disk Density at R ∗ .... cm − × × × Disk Thickness .... AU 25 25 6Temperature power law index .. − . − .
33 +0 . − . − . − . ⊙ × × × Disk Mass... M ⊙ + / e − Bremsstrahlung model assumes a constant electron temperature and is insensitive to the stellar radius; alsowhile the ionized disk mass is << ⊙ , it is possible to have a substantial neutral mass that does not contribute to theemission. Power law indexes are defined as ∝ r index .
14 –Fig. 1.— Continuum images of Orion–I at 43 GHz made with the VLA in the A-configuration.The image in the upper panel is at 50 mas (FWHM) resolution and the image in the middlepanel is at 34 mas resolution. The emission is elongated NW–SE and may be from an disksurrounding a massive YSO. The brightest component near the center of the disk may beunresolved. When we subtract a point-like 2.2 mJy component from the (13 mJy total)emission, we obtain the image in the lower panel.
In all images the contour levels are atinteger multiples of 0.5 mJy beam − . The FWHM of the restoring beams are shown in thebottom right corner of each panel. At a distance of 480 pc, 0.1 ′′ corresponds to 48 AU. 15 –Fig. 2.— Orion–I SiO v= 1, J=1–0 maser channel maps ( colored contours ) superposed onthe 34 mas resolution continuum emission ( heavy black contours ). The colors are chosento approximate those used in Fig. 1a of Greenhill et al. (1998); contouring levels start at3 Jy beam − and increase by factors of 2. Center LSR velocities of the 5.43 km s − widechannels are given in the lower left corner of each panel. Continuum contours are integermultiples of 0.5 mJy beam − . The bottom right panel shows a map of the integrated SiOemission, with light contours at integer multiples of 0.18 Jy beam − km s − . All position off-sets are relative to Orion–I, whose position is ( α, δ ) J2000 = (05 h m . s , − ◦ ′ . ′′ )at 2000 Nov. 13 (G´omez et al. 2005). 16 –Fig. 3.— Brightness temperature profiles of the 43 GHz continuum emission of Orion–I. Theobserved profile ( solid red line ) is along the elongation of the edge-on disk-like emission inthe 34 mas resolution image. Model profiles are shown for an edge-on disk with collisionalionization for central heating only (Model A, blue dotted line ) and with additional localheating (Model B, cyan dash-dotted line ; see § green dashed line ; see §§