Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.stsci.edu/jwst/overview/history/drm/drm15.pdf Дата изменения: Unknown Дата индексирования: Tue Feb 5 14:23:27 2013 Кодировка: Поисковые слова: m 17

Next Generation Space Telescope

Ad-Hoc Science Working Group Design Reference Mission Proposal

The Physics of Star Formation: Understanding the Youngest Protostars
Program contacts: Tom Greene, Michael Meyer, Ewine van Dishoeck, John Bally, Suzan Edwards Scienti c category: STAR FORMATION Instruments: MIR CAM, MIR SPEC Days of observation: 53

Abstract
We propose to bring the power of NGST to bear on one of the most important problems in star formation understanding protostellar formation from parent molecular cores and subsequentevolution towards becoming young stars. A few extremelyyoung 104 105 yr protostars have nally been discovered with sub-millimeter telescopes in recentyears, but the high extinctions Av 400 mag towards their central regions have prevented detailed study of their inner structures and mass accretion properties. Thus manykey questions of the star formation process remain unanswered, such as: What determines the nal mass of a forming star; do initial conditions of the ambient gas play the dominant role, or does feedback from the process itself x stellar mass? How do protostellar envelopes evolve from pre stellar cores to star+disk systems, and is the actual process consistent with our current theoretical models? When and under what conditions do binary systems form? What are the accretion histories of protostellar ob jects? When do protostellar disks form and when are jets rst launched from the system? We propose to answer these questions by undertaking a program of mid-IR imaging and spectroscopic observations of protostellar sources with NGST. We will probe the infalling envelopes, central ob jects, accretion disks, and accretion powered jets from protostars which are still accreting the bulk of their masses. These observations will be used to infer the morphologies and density structures of infalling circumstellar envelopes, measure the binary frequency of protostars, determine envelope infall rates from shocked line emissions, and evaluate mass out ow rates and accretion histories from observations of jets. This information will be crucial in determining the process by which stars and disks are assembled from molecular cloud cores and for understanding what mechanisms regulate stellar masses and determine the initial mass function.


ASWG DRM Proposal The Physics of Star Formation: Understanding the Youngest Protostars

Observing Summary:
Target 120 CLASS 0 I YSOS 120 CLASS 0 I YSOS 120 CLASS 0 I YSOS

RA Dec galactic plane QAB galactic plane NAB galactic plane QAB

mAB =19 , 22 17 =19 , 22

Con guration mode Days MIR CAM R 5 10 MIR SPEC R3000 18 MIR CAM R100-300 25 Grand total days 53

2


ASWG DRM Proposal The Physics of Star Formation: Understanding the Youngest Protostars

Scienti c Ob jectives
Over the past 20 years, a theoretical paradigm of low-mass star formation has emerged e.g., Shu 1977; Terebey et al. 1984; Shu et al. 1987. This standard model has successfully interpreted the IR spectral energy distributions SEDs of young stellar ob jects YSOs as an evolutionary sequence from highly embedded protostars still accreting mass Class I SED to young stars surrounded by substantial accretion disks Class II SED to revealed young stars whichhave dissipated their disks and mayhave completed their planet formation Class III SED. Although wehave made considerable progress in observing and understanding pre main sequence evolution Class II and III phases, the earliest phases of star formation the events which lead to Class I YSOs remain highly enigmatic. In recentyears, several protostars in the earliest stage of assembly have been discovered. e Known as Class 0" ob jects Andr et al. 1993, many were initially identi ed from their powerful and highly collimated jets whichwere found to be emerging from dense molecular cloud cores. Most have only been detected at far IR e.g., 25 , 60m IRAS data and sub mm wavelengths. They appear to be protostars that haveyet to accrete the bulk of their masses and are characterized by very cold T 30 K, extended r 1000 AU envelopes and powerful bipolar out ows. Their ages, inferred from their SEDs and number frequency, are less than 105 yr, and the dynamical ages of their out ows length velocity 104 yr are often muchyounger than those of the more evolved Class I YSOs. The Class 0 YSOs represent the most crucial phase in the star formation process, when the physics that determines the mass of the nal star is imprinted on the system. Key questions regarding Class 0 sources are: Is the mass infall rate onto the protostar+disk system determined solely by the isothermal sound speed as predicted in the inside-out" collapse models? What is the initial ratio of protostellar to disk mass and how doesitevolve with time? Does fragmentation of massive accretion disks lead to the formation of binary systems? What determines the onset of disk accretion and how are the powerful out ows triggered? Does out ow disrupt the envelope, terminating infall and thereby in uencing the nal mass of the forming star? Is the mass of the star built up through steady accretion of material from its disk, or is most of its mass accumulated in short bursts with very high rates of mass accretion? In order to study the energetic processes that occur deep in the potential well of these collapsing systems, mid IR observations that trace the emission of the inner regions directly are required. We propose here a suite of imaging and spectroscopic NGST investigations of Class 0 I YSOs and their out ows that weanticipate will provide the rst detailed understanding of these enigmatic ob jects and the star assembly process.

1 Protostellar Sources and Envelopes: Physical Structures
Sub-mm observations have been able to constrain the luminosities and overall sizes of Class 0 YSO systems but have yielded few details of the central protostars or their inner circumstellar regions. The luminosity of Class 0 YSOs emerges predominantly in the sub-mm wavelength 3


ASWG DRM Proposal The Physics of Star Formation: Understanding the Youngest Protostars region as a result of re-processing of shorter wavelength radiation by grains in their extended protostellar envelopes. The sub-mm spectra of these ob jects are feature-less and similar to blackbodies T ' 20 30 K see Figure 1 and thus convey little information on their warmer driving central protostars or inner circumstellar regions. Mass accretion rates and luminosities are highly dependentonlocal volume densities through the sound speed of the gas. As a protostar evolves, the density and temperature pro les of its inner envelope may change e.g., see Andr et al. 1996, thereby changing the accretion rate onto the central e source. We propose to characterize this evolution by conducting a multi color imaging survey of a large sample of 120 Class 0 I ob jects found in nearby star forming regions selected for intrinsic luminosity i.e. mass and accretion rate, SED shape i.e. evolutionary state, and star forming environment i.e. initial conditions. Class 0 YSOs are expected to have uxes 100 or more times higher than cool T = 20 30 K blackbodies in the mid-IR Figure 1. This is because radiation from warm central sources and scattering o dust grains in inner circumstellar envelopes are strongest at short" wavelengths, 10 50 m. The exact amount of scattering is a strong function of the density and geometry of inner YSO envelopes, so departures from blackbody emission are highly diagnostic of envelope structures density pro les etc. and warm central sources. Radiative transfer models e.g., Wol re & Cassinelli 1986 also predict strong grain and gas features in mid-IR spectra which further constrain protostellar envelope radiation environments and geometries. Knowledge of envelope structures i.e., r and T r leads to direct predictions of mass accretion rates onto the central protostars. We will compare these predicted rates to those measured from emission line observations described below and rates inferred from protostellar luminosities. This comparison will clearly reveal whether actual protostars accrete mass at rates predicted by the standard model or whether re nements of the model are needed. Spatially resolved 50 AU at = 10 m is possible for many nearby protostars images will also provide needed constraints to the source, envelope, and disk geometries for a given mid-IR energy distribution Yorke et al. 1995. Furthermore, these mid-IR images will be the best and most universal probes of protostellar multiplicity on scales of 100 AU and smaller. This is very signi cant for testing whether fragmentation of a massive accretion disk leads to the formation of binaries with separations 50 100 AU, typical of multiple systems observed in the eld.

2 Protostellar Disks and Mass Accretion
After a central protostellar core develops, theory predicts that higher angular momentum material continues to infall from the circumstellar envelope along stream lines whicheventually miss" the central ob ject. A circumstellar disk is formed with mass comparable to the central protostar. Such a massive disk quickly becomes unstable and can rapidly transport mass onto the central ob ject while redistributing angular momentum outward. This process _ repeats until infall onto the protostar+disk system terminates M=M 105 years, signaling the end of the Class I phase. It is only during these early stages, when AV 100 mag, 4


ASWG DRM Proposal The Physics of Star Formation: Understanding the Youngest Protostars

Figure 1: Spectral Energy Distribution of Class 0 protostar VLA 1623 from Andr et al. 1993. e Points are existing observations, solid line is a T=20 K blackbody, and dashed line is the prediction of the Wol re & Cassinelli 1986 radiative transfer code with r,0:5 . Observations in the 10 50 m region are required to reveal details of the inner infalling envelope. The sensitivity of NGST T=50 K primary at R=1000 for S N=10 and 104 s exposure time is shown over 5 m 30 m as a thick curve. Most known Class 0 YSOs are equally or somewhat less luminous and approximately a factor of 3 more distant. Thus they are about a factor of 10 fainter but should still be observable continua at R 5 and lines at R 1000.

5


ASWG DRM Proposal The Physics of Star Formation: Understanding the Youngest Protostars that the the physics of the collapse process can be investigated. We propose to quantify mass infall rates from collapsing envelopes by observing mid IR emission line diagnostics of envelope material crashing onto protostar+disk systems in accretion shocks. We will observe a large sample of ob jects as outlined above and described later in the Observing Strategy section in nearby star forming regions. Our goals are: i to test whether infall rates and disk radii evolve with time during the protostellar Class 0 I phase; ii to determine whether infall rates are consistent with those predicted by the observed envelope structures and cloud conditions; and iii to explore whether higher mass protostars are assembled in a way fundamentally di erent from low mass ones. As illustrated in Figure 2, shocks propagating into a dense molecular medium can produce a wide variety of emission line diagnostics. The relative strengths of various emission lines ro vibrational modes of H2, and vibrational bands of H2O and CO indicate the shock velocities of the impacting material Neufeld & Hollenbach, 1994. For a xed velocity, the line luminosities are directly proportional to the infall rate. Further, the shockvelocity provides an indication of at what radius the infalling material is impacting the circumstellar disk. Thus measuring the line ratios and uxes of these diagnostic features will enable us to infer the mass accretion rates and infall radii for our sample of protostars as a function of mass, evolutionary state, and initial conditions. In order to determine infall rates within 10, line uxes must be measured with S N 10 against the continuum of the underlying protostar. Spatially resolved spectra of nearby protostars will separate infall and out ow regions and also yield the physical extent of active accretion onto disks. These data would allow us to test in detail models of infall radius and calibrate our shock diagnostics for application to more distant regions. Analysis of solid state ice features in protostellar envelopes acquired simultaneously with the emission lines will provide information on present and past envelope temperatures, further constraining accretion properties.

3 Formation and Histories of Protostellar Jets
Stellar out ows and jets provide fossil records of the mass loss history of young stars, so they are also very useful in chronicling the very earliest phases of star formation. Attention was rst drawn to many Class 0 protostars by the detection of out ows from unidenti ed sources which were not visible in the IRAS data see Bachiller 1996 for a recent review. Once attention was drawn to the points of symmetry of these `sourceless' out ows, deep sub-mm or radio continuum observations were used to directly detect these protostars. Out ow observations can probe several fundamental questions of star formation: exactly when do protostellar out ows rst turn on does the formation of a massive accretion disk instantly ignite a powerful out ow? What determines the energetics of a protostellar out ow - stellar mass, accretion history, or cloud conditions? Where exactly does the out ow jet form? How much momentum do out ows deposit in clouds? Is the out ow responsible for terminating infall, thus determining the nal mass of the forming star? Mid-IR images and spectra of out owing protostellar jets can address these questions. It appears that nearly all protostars show out ow activity as soon as a central source 6


ASWG DRM Proposal The Physics of Star Formation: Understanding the Youngest Protostars

Figure 2: Shock excited emission lines due to ro vibrational modes of H2 17.0, 12.3 m, etc., and vibrational bands of H2 O 6 7 m and CO fundamental band at 4.6 m. This ISO SWS spectrum from van Dishoeck et al. 1998 was obtained toward a dense embedded cluster associated with the very luminous protostellar ob ject IRc 2 Lbol 104L . These line uxes observed with a 2000 input aperture diagnose the impact of a large scale out ow with ambient dense gas in the molecular cloud core n 106 cm,3 . Calculations by Neufeld & Hollenbach 1994 predict that infall onto a disk surrounding a solar mass protostar will produce emission line strengths orders of magnitude lower, impossible to observe without NGST. However, this gure does illustrate the richness of mid-IR spectral diagnostics of energetic processes in star forming environments.

7


ASWG DRM Proposal The Physics of Star Formation: Understanding the Youngest Protostars has formed; i.e. accretion and out ow appear to go hand-in-hand. If out ow and infall are indeed causally linked, can the knot structure in out ows reveal details of a protostar's accretion history? Indeed, many jets show bipolar symmetry in their ejecta structures, so jet morphologies do appear to be related to events which occurred at the central protostars. The size of out ows, their symmetries, and velocity elds provide constraints on their ages, momenta, and formation locations mechanisms. ISO has shown that the rotational transitions in the = 0 , 0 vibrational lines of H2 between 28 and 6m provide the most sensitive tracer of out ows powered by the youngest stars, those that are heavily obscured, or those that power only relatively weak v 5km s,1 shocks. In order to address the above questions of jet ages, formation, and energies, we propose NGST imaging and spectroscopic observations of H2 emission from the jets of our Class 0 and I YSO sample. Via these jet observations, NGST will probe protostars and their environmental interactions in conditions that range from dense cluster forming regions to relatively sparse isolated groups.

References

NGST Uniqueness Relationship to Other Facilities

Andr P., Ward-Thompson, D., & Barsony, M. 1993, ApJ, 406, 122 e, Andr P., Ward-Thompson, D., & Motte, F. 1996, A&A, 314, 625 e, Bachiller 1996, ARA&A, 34, 111 Ceccarelli, C., et al. 1998, A&A, 331, 372 Neufeld, D. A., & Hollenbach, D. J. 1994, ApJ, 428, 170 Shu, F. H. 1977, ApJ, 214, 288 Shu, F. H., Adams, F. C., & Lizano, S. 1987, ARA&A, 25, 23 Terebey, S., Shu, F. H., & Cassen, P. 1984, ApJ, 286, 529 van Dishoeck, E. F. et al. 1998, ApJ, 502, L173 Wol re, M. G., & Cassinelli, J. P. 1986, ApJ, 310, 207 Yorke, H., Bodenheimer, P., & Laughlin, G. 1995, ApJ, 443, 199

NGST is the only existing or planned facility which can make these observations. Its large aperture and low background produce high sensitivities in the mid-IR where it has superb spatial resolution 0:003or 50 AU for nearby clouds for the continuum and line emission imaging studies. NGST will be sensitive enough to image hundreds of low-mass Class 0 protostars within 1 kpc at S N 10 in t 103,4 sat wavelengths 5 m 40m. SIRTF and IRIS will identify new Class 0 YSOs but will lack the spatial spectral resolution and sensitivity to study them in the mid-IR regime. SOFIA will lack the sensitivity and spatial resolution required for these observations. Although ISO performed a pioneering observation of the 63 m OI feature in an accretion shock for the most luminous Class 0 source IRAS 16293 2422 Ceccarelli et al. 1998, it did not have the sensitivity to detect line emission from the hotter shock diagnostics described above and was unable to perform this observation on any of fainter Class 0 sources. We anticipate that a few pioneering observations of this type will be conducted with SOFIA, but NGST will be required to 8


ASWG DRM Proposal The Physics of Star Formation: Understanding the Youngest Protostars perform the systematic program capable of addressing the fundamental questions asked above. NGST will achieve S N = 10 detections of emission line diagnostics at R=3000 against the continuum of Class 0 and 1 YSOs in approximately 104 seconds in nearbyd 500 pc regions. Ground-based observations with the Keck or GEMINI telescopes are impossible as they would require over 1000 times as much time even when NGST is source-noise limited! Also, only NGST can observe the key H2O bands at 6 m and the =6 , 28m H2 lines which are crucial to the success of this program. Our approach is to conduct an aggressive observing campaign of a representative sample of 120 ob jects using the best practical techniques for the goals of the investigation. This sample will contain several 10 protostars in each of 12 categories as a function of intrinsic source luminosity greater less than 3 L to probe underlying mass and accretion rates, SED shape Class 0 I to trackevolutionary state, and star forming environment 3 regions within 1 kpc distance with di erent global cloud properties. Broad-band R = 3 5 photometric observations of Class 0 and I protostars are needed in at least 4 lters in the 10 28 m range. Extension to 40 mwould provide considerably more scienti c return. These spatially resolved multi color data will provide constraints on extinction and envelope properties i.e., r and T r and will also be combined with far-IR SIRTF, sub-mm, and mm energy distributions to best determine source luminosities and large scale envelope properties. This will require 10 days of observing time for our sample of 120 ob jects. These same YSO elds will also be imaged in narrowR 100 H2 to observe their jets over linear elds of 10 , 40 each. The best transitions for such observations are the 28 and 17 m S0 and S1 lines in the = 0 , 0 vibrational states since these lines are easily excited by shocks that heat the medium to only a few hundred degrees or less. Thus these observations will be sensitive to cold 40 T 100 K gas, which constitutes the bulk of the material in molecular out ows. 1 hour of integration in each of these lines will detect 10,3 , 10,5 M per square arcsecond, several orders of magnitude better than ISO and by far the most sensitive observations of this material. Observing the ob ject sample in both line lters will require an additional 25 days of observation. These observations will be performed in advance of the spectroscopic survey outlined below so that slit orientations can be set appropriately to include spatial information on the extended jet emission. We shall also acquire mid-IR spectroscopic observations of protostellar envelopes and disks to measure mass infall rates and constrain envelope temperatures. We will observe the same sample of 120 protostars divided into 12 categories as described above. In order to obtain S N 10 in a line, we require S N 80 on the continuum of a typical Class 0 I YSO at R = 3000. This results in an integration time of 1:5 104 s when we model the central ob ject as a very activeT Tauri star extinguished byan envelope with Av = 400 mag. We wish to observe several lines and bands over the 6 12 m range whichwe expect to cover in a single exposure. We estimate that spectroscopic observations of 120 ob jects 9

Observing Strategy


ASWG DRM Proposal The Physics of Star Formation: Understanding the Youngest Protostars will require 18 days with 2 wavelength settings per ob ject. Lower spectral resolution results in much longer integration times needed to detect these unresolved features i.e. over 10 increase at R = 1000 for the emission line program outlined here. Minimum Spatial Resolution: 300 mas at 10 m Minimum Spectral Resolution: 3000 at 10 m Minimum FOV: 4 arcmin2 at 17 m

Special Requirements

Precursor Supporting Observations
Millimeter Array and NGST observations of protostellar envelopes and jets are complimentary. For example, Andr et al. 1993 noted that the dust envelope structure and accretion e properties of protostars are best constrained by tting the spectral energy distributions and radial pro les simultaneously over mid-IR through mm wavelengths. The spatial resolution of the MMA at 350 m and NGST in the near IR will be comparable. However, the MMA will probe cooler envelope material at r 20 AU from the protostar compared to the observations of warm inner regions which are actually accreting onto the protostar described here. Accretion rates derived from infalling gas spectrally and spatially resolved MMA observations can be compared with shock diagnostics NGST as can mass loss rates from the jet components. Thus both mid-IR 10 50m NGST and existing future sub-mm observations are required to understand fully protostars and their circumstellar environments. Source lists of Class 0 I ob jects will be provided by surveys that will be conducted with SIRTF and IRIS. SOFIA and ground based 8 10 m class telescopes will conduct initial spectroscopic studies of several interesting ob jects which will serve as precursors to the kind of systematic program enabled by NGST as outlined here.

10