Density Distribution and Grain Growth of Class 0...

1
4. Radial dependence of β : L1448 IRS 3B 2. Visibility comparison Observations and data Targets : 7 fields toward Serpens molecular cloud 13 fields toward Perseus molecular cloud Looney et al. 2000, Jorgensen et al. 2006, Enoch et al. 2009 2009 July : 1-mm data in E-configuration 2009 August-September : partial 3-mm data in D-configuration Synthesized beam ~ 5˝ 5˝ at both wavelengths Flux calibrator : Uranus for all data Abstract Introduction Survey Density Distribution and Grain Growth of Class 0 YSOs Woojin Kwon 1 ([email protected] ), Leslie W. Looney 1 , Lee G. Mundy 2 1 U of Illinois at Urbana-Champaign, 2 U of Maryland at College Park Acknowledgement: This study is based on CARMA observations. Support for CARMA construction was derived from the states of Illinois, California, and Maryland, the Gordon and Betty Moore Foundation, the Eileen and Kenneth Norris Foundation, the Caltech Associates, and the National Science Foundation. Ongoing CARMA development and operations are supported by the National Science Foundation under cooperative agreement AST-0540459, and by the CARMA partner universities. Class 0 YSOs are the youngest protostellar systems in the low mass star formation, which have massive envelopes and well- developed bipolar outflows (Andre et al. 1993). The envelopes have the conditions of the preceding molecular cores, the natal places of star formation. Motivation : results of Kwon et al. 2009 L1157 CARMA in E configuration From http://www.mmarray.org Targets 1.3 m mm 2.7 m mm Other names RA (J2000) Dec (J2000) RMS Flux RMS Flux α [mJy/bm] [mJy] [mJy/bm] [mJy] Serp01 Serp02 Serp03 Serp04 Serp05a b Serp06 Serp07a b Pers01 Pers02 Pers03 Pers04 Pers05 Pers06a b Pers07 Pers08 Pers09a b Pers10a b Pers11 Pers12 Pers13a b 18:29:09.101 0:31:31.056 6.4 279 3.0 47.5 2.48 FIRS1 18:29:49.785 1:15:20.523 32 1805 3.7 296 2.53 S68N 18:29:48.087 1:16:43.656 14 288 3.0 50.1 2.45 18:29:56.716 1:13:14.928 13 905 3.4 133 2.68 18:30:00.321 1:12:59.388 4.0 145 3.2 ----- ----- 18:30:00.567 1:12:50.504 4.0 67.5 3.2 ----- ----- 18:28:54.036 0:29:28.952 4.3 131 2.9 18.3 2.75 18:29:06.646 0:30:33.961 5.0 133 2.8 27.6 2.20 18:29:06.221 0:30:43.136 5.0 97.3 2.8 9.8 3.21 HH 211 3:43:56.789 32:00:49.939 8.4 280 1.9 43.9 2.59 B1-d 3:33:16.405 31:06:52.634 5.0 54.6 1.7 8.0 2.69 HH 7-11 MMS6 3:29:04.050 31:14:46.600 4.3 ----- 1.9 ----- ----- L 1448-mm 3:25:38.904 30:44:05.191 5.7 298 1.9 54.5 2.38 B1-c 3:33:17.879 31:09:31.971 6.9 424 1.7 55.9 2.83 3:33:21.361 31:07:26.509 12 356 3:33:21.164 31:07:43.426 15 312 3:29:07.776 31:21:57.030 5.3 110 IRAS 03292+3039 3:32:17.939 30:49:47.686 18 803 NGC 1333-IRAS 4B 3:29:12.011 31:13:07.979 32 1246 NGC 1333-IRAS 4C 3:29:12.865 31:13:06.905 32 323 3:29:11.284 31:18:31.196 6.7 170 3:29:10.690 31:18:20.168 8.7 172 NGC 1333-IRAS 2A 3:28:55.607 31:14:36.785 10 591 B1-a 3:33:16.660 31:07:55.200 4.8 ----- SVS 13A 3:29:03.733 31:16:03.262 14 589 SVS 13B 3:29:03.044 31:15:51.431 16 663 Dust continuum (λ = 1.3 and 2.7 mm) and dust opacity spectral index (β) maps of three targets. The synthesized beams are the same in the three maps of each target and shown on the bottom right of the β maps. Note : mostly β < 1 and β variation along radius. Visibility amplitude averaged in annuli (upper panels) and dust opacity spectral index β plots (lower panels) of the three targets as a function of uv distance. Note that larger uv distances correspond to smaller scales. Error bars indicate standard errors in each bin. Solid points on the β plots delimit the range of values expected based on absolute flux calibration uncertainties (15% at λ = 1.3 mm and 10% at λ = 2.7 mm). The curves in upper panels present the best fits. Note : β 1.0 and β decreasing on smaller scales (particularly, L1448 IRS 3B). No good fit with a constant β for L1448 IRS 3B (below left). Optically thick point source (“3. Modeling”, but no point source feature detected Looney et al. 2003) or radial dependent β (this new model) is required. A new model based on grain growth by gas accretion (Spitzer 1978) β(r) (ρv) -1 ρ -1 T -1/2 where r < R β β(r) = β out = 1.7 where r > R β The rest of 3-mm data toward Perseus will be taken in the next D configuration (from 2010 April 9). Modeling in uv space * No optically thin assumption nor Rayleigh-Jeans approximation * Power-law density and temperature distributions are assumed. 1. Image construction through radiative transfer 2. Primary beam correction (6m-6m, 6m-10m, 10m-10m) 3. Fourier transform 4. Sampling over the uv coverage of the observation data 5. Comparison with data (χ ν 2 calculation) (6). Likelihood calculation in (β, p), using exp(-χ ν 2 /2) Note that the modeling procedures are the same to Kwon et al. (2009) and parameter searching in a large space has not been done yet. Density distribution shedding light on star formation E.g., Ambipolar diffusion model : p ~ 1.7 (ρ r -p ) (e.g., Tassis & Mouschovias 2005 ) Inside-out model : p = 1.5 in the inside free-fall region (Shu 1977) p = 2.0 in the outer isothermal region Dust grain growth Dust opacity spectral index at sub/mm indicating grain sizes: β log(κ ν1 /κ ν0 )/log(ν 1 /ν 0 ) (e.g., Draine 2006) ISM : β ~ 1.7 (sub-micron size grains) Protoplanetary disks : β < 1 (mm size grains) 1. When have grains significantly grown? 2. What causes β(r)? E.g., dust grain segregation is predicted by the ambipolar diffusion model of star formation (Ciolek & Mouschovias 1996), inversely proportional to the initial mass-to-magnetic field flux ratios of cores. On the other hand, it could be due to faster grain growth at the dense central region. Preliminary results 1. Based on the total fluxes of targets at λ = 1.3 and 2.7 mm, β is likely around unity. Note that β < 1, when assuming optically thin and Rayleigh-Jeans approximations (Table): β = α - 2. Preliminary results 2. Current “eye” fitting results suggest that there are still two types of density distribution in Class 0 YSOs: p ~ 1.8 and p > 2.0. Caveat : a large parameter space has not been searched. 1. Image comparison β = α - 2 = log(F ν1 /F ν0 )/log(ν 1 /ν 0 ) - 2 (i.e., F ν ν β+2 ) Optically thin and Rayleigh-Jeans approximations 3. Modeling in uv space Targets p β M T (M ) R in (AU) R out (AU) F pt (Jy) L1448 IRS 2 1.79 0.88 1.36 20 5300 ----- L1448 IRS 3B 2.14 0.96 3.68 14 6300 0.099 L1157 1.72 0.91 0.59 20 2300 0.020 Targets p R β (AU) M T (M ) R in (AU) R out (AU) F pt (Jy) L1448 IRS 3B 2.59 420 2.51 10 5900 0.0 Color composite images taken by Spitzer IRAC: L1448 region (Tobin et al. 2007) and L1157 (Looney et al. 2007). L1448 IRS 3 L1448 IRS 2 L1448 mm Targets p β M T (M ) R in (AU) R out (AU) F pt (Jy) χ ν 2 Serp02 Serp03 Pers01 Pers04 2.3 0.8 3.9 10.0 3000 0.000 7.2 1.8 1.0 1.9 10.0 6000 0.003 3.7 1.8 0.9 0.4 10.0 1700 0.010 1.3 1.8 0.9 1.9 10.0 7000 0.020 1.5 References: Andre, Ward-Thompson, & Barsony, 1993, ApJ, 406, 122; Ciolek & Mouschovias, 1996, ApJ, 468, 749; Draine, 2006, ApJ, 636, 1114; Enoch, Evans, Sargent, & Glenn, 2009, ApJ, 692, 973; Jorgensen, Johnstone, Kirk, & Myers, 2006, ApJ, 656, 293; Kwon, Looney, Mundy, Chiang, & Kemball, 2009, ApJ, 696, 841; Looney, Mundy, & Welch, 2000, ApJ, 529, 477; Looney, Mundy, & Welch, 2003, ApJ, 592, 255; Looney, Tobin, & Kwon, 2007, ApJ, 670, L131; Shu, 1977, ApJ, 214, 488; Spitzer, 1978, PP in ISM; Tassis & Mouschovias, 2005, ApJ, 618, 783; Tobin, Looney, Mundy, Kwon, & Hamidouche, 2007, ApJ, 659, 1404 Circumstellar envelopes of young stellar objects (YSOs) at the earliest stage, the so-called Class 0 YSOs, harbor conditions of star formation. Recently, we studied the density distribution and the dust opacity spectral index (β) of three Class 0 YSOs (Kwon et al. 2009): L1448 IRS 2, L1448 IRS 3B, and L1157, through visibility modeling of the continuum data at λ = 1.3 mm and 2.7 mm taken by the Combined Array for Research in Millimeter-wave Astronomy (CARMA). We found that β is around 1 indicating significant grain growth at the earliest stage already and that there are two types of density distribution: density power-law index p ~ 1.8 in L1448 IRS 2 and L1157 and p ~ 2.6 in L1448 IRS 3B. Since L1448 IRS 3B is a binary system and younger in terms of the bipolar outflow kinematic time scale, it could indicate a discrepancy of binary star formation and/or an evolutionary change of density distribution. In addition, L1448 IRS 3B has a distinct radial dependence of β presenting dust grain segregation and/or faster grain growth at the central region, which is not clearly detected in the others. To investigate these interesting results further, we have been carrying out a survey of 20 YSO fields using CARMA. So far we have obtained a complete data sets of about half targets. Preliminary results include that the dust opacity spectral index seems to be around 1 in the larger sample, and Class 0 YSOs could still be divided into two types of density distribution. Physical properties of the targets will be obtained by modeling in a large parameter space. The density distribution and the grain growth should be explained by successful star formation theories.

Transcript of Density Distribution and Grain Growth of Class 0...

Page 1: Density Distribution and Grain Growth of Class 0 YSOsconference.astro.ufl.edu/STARSTOGALAXIES/science_final/talks/kwon_w.pdf · 3. Fourier transform 4. Sampling over the uv coverage

4. Radial dependence of β : L1448 IRS 3B

2. Visibility comparison

Observations and dataTargets : 7 fields toward Serpens molecular cloud 13 fields toward Perseus molecular cloud Looney et al. 2000, Jorgensen et al. 2006, Enoch et al. 20092009 July : 1-mm data in E-configuration2009 August-September : partial 3-mm data in D-configurationSynthesized beam ~ 5˝ ⨉ 5˝ at both wavelengthsFlux calibrator : Uranus for all data

Abstract Introduction Survey

Density Distribution and Grain Growth of Class 0 YSOsWoojin Kwon1([email protected]), Leslie W. Looney1 , Lee G. Mundy2 1 U of Illinois at Urbana-Champaign, 2 U of Maryland at College Park

Acknowledgement: This study is based on CARMA observations. Support for CARMA construction was derived from the states of Illinois, California, and Maryland, the Gordon and Betty Moore Foundation, the Eileen and Kenneth Norris Foundation, the Caltech Associates, and the National Science Foundation. Ongoing CARMA development and operations are supported by the National Science Foundation under cooperative agreement AST-0540459, and by the CARMA partner universities.

Class 0 YSOs are the youngest protostellar systems in the low mass star formation, which have massive envelopes and well-developed bipolar outflows (Andre et al. 1993). The envelopes have the conditions of the preceding molecular cores, the natal places of star formation.

Motivation : results of Kwon et al. 2009

L1157

CARMA in E configuration From http://www.mmarray.org

TargetsTargetsTargets Other names RA (J2000) Dec (J2000)1.3 mm1.3 mm 2.7 mm2.7 mm

αOther names RA (J2000) Dec (J2000) RMS Flux RMS Flux αOther names RA (J2000) Dec (J2000)[mJy/bm] [mJy] [mJy/bm] [mJy]

α

Serp01 Serp02 Serp03 Serp04 Serp05a b Serp06 Serp07a b Pers01 Pers02 Pers03 Pers04 Pers05 Pers06a b Pers07 Pers08 Pers09a b Pers10a b Pers11 Pers12 Pers13a b

18:29:09.101 0:31:31.056 6.4 279 3.0 47.5 2.48

FIRS1 18:29:49.785 1:15:20.523 32 1805 3.7 296 2.53

S68N 18:29:48.087 1:16:43.656 14 288 3.0 50.1 2.4518:29:56.716 1:13:14.928 13 905 3.4 133 2.6818:30:00.321 1:12:59.388 4.0 145 3.2 ----- -----18:30:00.567 1:12:50.504 4.0 67.5 3.2 ----- -----18:28:54.036 0:29:28.952 4.3 131 2.9 18.3 2.7518:29:06.646 0:30:33.961 5.0 133 2.8 27.6 2.2018:29:06.221 0:30:43.136 5.0 97.3 2.8 9.8 3.21

HH 211 3:43:56.789 32:00:49.939 8.4 280 1.9 43.9 2.59B1-d 3:33:16.405 31:06:52.634 5.0 54.6 1.7 8.0 2.69

HH 7-11 MMS6 3:29:04.050 31:14:46.600 4.3 ----- 1.9 ----- -----L 1448-mm 3:25:38.904 30:44:05.191 5.7 298 1.9 54.5 2.38

B1-c 3:33:17.879 31:09:31.971 6.9 424 1.7 55.9 2.833:33:21.361 31:07:26.509 12 3563:33:21.164 31:07:43.426 15 3123:29:07.776 31:21:57.030 5.3 110

IRAS 03292+3039 3:32:17.939 30:49:47.686 18 803NGC 1333-IRAS 4B 3:29:12.011 31:13:07.979 32 1246NGC 1333-IRAS 4C 3:29:12.865 31:13:06.905 32 323

3:29:11.284 31:18:31.196 6.7 1703:29:10.690 31:18:20.168 8.7 172

NGC 1333-IRAS 2A 3:28:55.607 31:14:36.785 10 591B1-a 3:33:16.660 31:07:55.200 4.8 -----

SVS 13A 3:29:03.733 31:16:03.262 14 589SVS 13B 3:29:03.044 31:15:51.431 16 663

Dust continuum (λ = 1.3 and 2.7 mm) and dust opacity spectral index (β) maps of three targets. The synthesized beams are the same in the three maps of each target and shown on the bottom right of the β maps. Note : mostly β < 1 and β variation along radius.

Visibility amplitude averaged in annuli (upper panels) and dust opacity spectral index β plots (lower panels) of the three targets as a function of uv distance. Note that larger uv distances correspond to smaller scales. Error bars indicate standard errors in each bin. Solid points on the β plots delimit the range of values expected based on absolute flux calibration uncertainties (15% at λ = 1.3 mm and 10% at λ = 2.7 mm). The curves in upper panels present the best fits. Note : β ≤ 1.0 and β decreasing on smaller scales (particularly, L1448 IRS 3B).

No good fit with a constant β for L1448 IRS 3B (below left).Optically thick point source (“3. Modeling”, but no point source feature detected Looney et al. 2003) or radial dependent β (this new model) is required.A new model based on grain growth by gas accretion (Spitzer 1978) β(r) ∝ (ρv)-1 ∝ ρ-1T-1/2 where r < Rβ

β(r) = βout = 1.7 where r > Rβ

The rest of 3-mm data toward Perseus will be taken in the next D configuration (from 2010 April 9).

Modeling in uv space* No optically thin assumption nor Rayleigh-Jeans approximation* Power-law density and temperature distributions are assumed.1. Image construction through radiative transfer2. Primary beam correction (6m-6m, 6m-10m, 10m-10m)3. Fourier transform4. Sampling over the uv coverage of the observation data5. Comparison with data (χν2 calculation)(6). Likelihood calculation in (β, p), using exp(-χν2/2)Note that the modeling procedures are the same to Kwon et al. (2009) and parameter searching in a large space has not been done yet.

Density distribution shedding light on star formationE.g., Ambipolar diffusion model : p ~ 1.7 (ρ ∝ r-p) (e.g., Tassis & Mouschovias 2005 ) Inside-out model : p = 1.5 in the inside free-fall region (Shu 1977) p = 2.0 in the outer isothermal region

Dust grain growthDust opacity spectral index at sub/mm indicating grain sizes: β ≡ log(κν1/κν0)/log(ν1/ν0) (e.g., Draine 2006) ISM : β ~ 1.7 (sub-micron size grains) Protoplanetary disks : β < 1 (mm size grains)1. When have grains significantly grown?2. What causes β(r)?E.g., dust grain segregation is predicted by the ambipolar diffusion model of star formation (Ciolek & Mouschovias 1996), inversely proportional to the initial mass-to-magnetic field flux ratios of cores. On the other hand, it could be due to faster grain growth at the dense central region.

Preliminary results 1.Based on the total fluxes of targets at λ = 1.3 and 2.7 mm, β is likely around unity. Note that β < 1, when assuming optically thin and Rayleigh-Jeans approximations (Table): β = α - 2.

Preliminary results 2.Current “eye” fitting results suggest that there are still two types of density distribution in Class 0 YSOs: p ~ 1.8 and p > 2.0. Caveat : a large parameter space has not been searched.

1. Image comparisonβ = α - 2 = log(Fν1/Fν0)/log(ν1/ν0) - 2 (i.e., Fν ∝ ν β+2)Optically thin and Rayleigh-Jeans approximations

3. Modeling in uv space

Targets p β MT (M) Rin (AU) Rout (AU) Fpt (Jy)

L1448 IRS 2 1.79 0.88 1.36 20 5300 -----

L1448 IRS 3B 2.14 0.96 3.68 14 6300 0.099

L1157 1.72 0.91 0.59 20 2300 0.020

Targets p Rβ (AU) MT (M) Rin (AU) Rout (AU) Fpt (Jy)

L1448 IRS 3B 2.59 420 2.51 10 5900 0.0

Color composite images taken by Spitzer IRAC: L1448 region (Tobin et al. 2007) and L1157 (Looney et al. 2007).

L1448 IRS 3

L1448 IRS 2

L1448 mm

Targets p β MT (M) Rin (AU) Rout (AU) Fpt (Jy) χν2

Serp02Serp03Pers01Pers04

2.3 0.8 3.9 10.0 3000 0.000 7.21.8 1.0 1.9 10.0 6000 0.003 3.71.8 0.9 0.4 10.0 1700 0.010 1.31.8 0.9 1.9 10.0 7000 0.020 1.5

References: Andre, Ward-Thompson, & Barsony, 1993, ApJ, 406, 122; Ciolek & Mouschovias, 1996, ApJ, 468, 749; Draine, 2006, ApJ, 636, 1114; Enoch, Evans, Sargent, & Glenn, 2009, ApJ, 692, 973; Jorgensen, Johnstone, Kirk, & Myers, 2006, ApJ, 656, 293; Kwon, Looney, Mundy, Chiang, & Kemball, 2009, ApJ, 696, 841; Looney, Mundy, & Welch, 2000, ApJ, 529, 477; Looney, Mundy, & Welch, 2003, ApJ, 592, 255; Looney, Tobin, & Kwon, 2007, ApJ, 670, L131; Shu, 1977, ApJ, 214, 488; Spitzer, 1978, PP in ISM; Tassis & Mouschovias, 2005, ApJ, 618, 783; Tobin, Looney, Mundy, Kwon, & Hamidouche, 2007, ApJ, 659, 1404

Circumstellar envelopes of young stellar objects (YSOs) at the earliest stage, the so-called Class 0 YSOs, harbor conditions of star formation. Recently, we studied the density distribution and the dust opacity spectral index (β) of three Class 0 YSOs (Kwon et al. 2009): L1448 IRS 2, L1448 IRS 3B, and L1157, through visibility modeling of the continuum data at λ = 1.3 mm and 2.7 mm taken by the Combined Array for Research in Millimeter-wave Astronomy (CARMA). We found that β is around 1 indicating significant grain growth at the earliest stage already and that there are two types of density distribution: density power-law index p ~ 1.8 in L1448 IRS 2 and L1157 and p ~ 2.6 in L1448 IRS 3B. Since L1448 IRS 3B is a binary system and younger in terms of the bipolar outflow kinematic time scale, it could indicate a discrepancy of binary star formation and/or an evolutionary change of density distribution. In addition, L1448 IRS 3B has a distinct radial dependence of β presenting dust grain segregation and/or faster grain growth at the central region, which is not clearly detected in the others. To investigate these interesting results further, we have been carrying out a survey of 20 YSO fields using CARMA. So far we have obtained a complete data sets of about half targets. Preliminary results include that the dust opacity spectral index seems to be around 1 in the larger sample, and Class 0 YSOs could still be divided into two types of density distribution. Physical properties of the targets will be obtained by modeling in a large parameter space. The density distribution and the grain growth should be explained by successful star formation theories.