星形成過程の観測的可視化 -- 対Entrainment Model大作戦 -- 富阪幸治（国立天文台）

偏光でさぐる星形成過程 富阪幸治（国立天文台）

NGC1333 IRAS 4A Polarization of dust thermal emission  Hourglass shape B-field Goncalves + 09

分子アウトフロー 2000 AU 原始星

L1551 IRS5 Molecular Outflow Optical Jets Snell, Loren, &Plambeck 1980 Saito, Kawabe, Kitamura&Sunada 1996 L1551 IRS5 Snell, Loren, &Plambeck 1980 Optical Jets

Uchida & Shibata (1985)

磁気加速によるアウトフローの形成 Initial Condition Numerical Method B W periodic Nested grid L=1 L=2 L=3 periodic boundary axisymmetric r perturbation nonaxisymmetric r perturbation

The coarsest grid Nested 4-times finer grid W r W r

Nested 28-times finer grid Just after the central density exceeds rA (first core formation), outflow begins to blow. (2) In this case, gas is accelerated by the magnetocentrifugal wind mechanism. (3) 10% of gas in mass is ejected with almost all the angular momentum. W r

偏光から磁場の方向を測定する (1) 吸収 (2) 熱放射 (3) 散乱

align align align ??? Alignment of Outflow and Magnetic Field local B global B IRAS 16293-2422 B Wolf et al (2003) align align 1800AU local B // Outflow but global B not // Outflow. B Outflow L1551 IRS5 Magnetic field B B How are the global and local magnetic field different? align ??? Polarization of thermal dust emission map SCUBA 850mm Tamura et al (1995) 1mm radio polarization

Alignment B-Field vs Jets and Disks around T Tauri Stars Menard & Duchene (2004) AAp, 425, 973 Misalignment angle measured from the direction of B-Field Jets 90 Disks w/o jets B-Field Jets Disks w/o jets Disk normal Jets are parallel and disks are perpendicular to B.  J // B Disks not perpendicular to B are not associated with jets.  J // B

Recent observations (1/2): B-fields, optical jets, and disks in Taurus Direction of magnetic fields inferred from polarized dust emission. Taurus-Auriga region Direction of optical jets Direction of disk normal CTTS are oriented randomly with respect to local magnetic fields! L1551-IRS5 Menard & Duchene (2004)

j j B j B Aligned Rotator Initial state Misaligned Rotator Spherical cloud B j B B-field disk This is result of 3D MHD simulation. In the case of aligned rotator, the angle between J and B theta equals 0, outflow Matsumoto, Tomisaka 2004 Matsumoto, Nakazato, Tomisaka 2006 　Disk perpendicular to not J but B-field (!) is formed.

MF45 L4 L9 L12 Disk perpto local B Cut A Global J L11 Local B-Field at the center outflow Cut A Cut B Axisymmetric!!

Disk, B Field and Rotation in Different Scales (Final state) Global J Initial B x16 x16 Disk oriantation, local B, and local J change their directions according to the scale.

Reconstruction of Polarization vectors at 5000 AU scale (Bave = 82 Reconstruction of Polarization vectors at 5000 AU scale (Bave = 82.8mG) MF45 Matsumoto et al. 2006 ApJ. 637, L105 B0=18.6mG yz xz xy 偏光度 高 偏光度 低 Green : mean direction of polarization vector Red : direction of the outflow (50AU scale B) Colors: column density ←　Three-dimensional structure Tree-dimensional angle between magnetic field and outflow is 12.4 deg. 4600AU The outflow is well aligned with the polarization vector.

Reconstruction of Polarization vectors at 5000 AU scale (Bave = 50 Reconstruction of Polarization vectors at 5000 AU scale (Bave = 50.1mG) WF45 B0=7.42mG yz xz xy Green : mean direction of polarization vector Red : direction of the outflow Colors: column density Three-dimensional angle between magnetic field and outflow is 53.5 deg. The alignment depends on the line of sight

Directions of B, W, and disk normal vectors: variation in scale. WF45 B0=7.42mG MF45 B0=18.6mG n F3D=12.4 deg. B W B n F3D=53.5 deg. W B：Ｍａｇｎｅｔｉｃ　Ｆｉｅｌｄ W: Rotation Axis n: Disk normal

Can we infer the central magnetic field near future? … by ALMA? Target: B335 @ 250 pc Resolution: 0.1” (25 AU) WF45 B0=7.42mG Yes, we can resolve the magnetic fields around the protostar. The outflow traces the direction of magnetic field at the cloud center.

双極分子流の起源? 磁場駆動 VS Entrainment 磁場駆動 磁気遠心力風 (Blandford&Peyne82) 磁気圧勾配 角運動量輸送 過剰な遠心力 rw2 アウトフロー 磁気圧勾配 -d/dz (Bf2/8p) [利点] 星形成時に余分の角運動量をアウトフローで捨てることが出来る。 not jet Kudoh & Shibata 97a,b Tomisaka 98

Segregation between angular momentum and mass by magnetically driven wind B-Field z ~99.99% Angular Momentum ~10% Mass ~0.01% Angular Momentum ~90% Mass positive torque negative torque Pseudo-disk r ~centrifugal radius Specific angular momentum is reduced a factor 10-4 Tomisaka 00

Two Types of Outflows strong B wide opening angle  U-type Three cloud has the same rotation rate Tomisaka 2002 but different magnetic field strength strong B wide opening angle  U-type  magnetocentrifugal wind Blandford & Peyne weak B narrow opening angle  I-type  magnetic pressure gradient consistent with jet’s simulation from Kepler disk (Kudoh & Shibata )

Observation of Magnetically Driven Molecular Outflow courtesy of Machida Low-velocity Flow from First core High-velocity flow from Protostar ~1000 times enlargement of central area 360 AU 0.35 AU voutflow~ 5 km/s vJet~50 km/s Two distinct flows appear in collapsing cloud First Core 　n~1011 cm-3, r~10-100 AU Protostar (Second Core) 　n~1021 cm-3, r~0.01 AU

観測とシミュレーションを比較することで、双極分子雲の駆動メカニズムを明らかにする。 Entrainment モデル 運動量が高速ジェットから周りの分子ガスに移される。例えばKH不安定性双極分子流 [問題]双極分子流の大きな開口角(Stahler 93) [欠点] 星形成過程で、分子雲コアの角運動量が星のそれに対して過剰である問題について解決とはならない。 観測とシミュレーションを比較することで、双極分子雲の駆動メカニズムを明らかにする。

(1) Turbulent entrainment Precession jet Induced turbulence Raga+93 A&A276,539 Masson & Chernin 93ApJ,414,230 (1) Turbulent entrainment (2) Entrainment through a bow shock Raga & Cabrit 93

We use linear polarization of dust thermal radiation Hour Glass 砂時計型磁場 We use linear polarization of dust thermal radiation SMA NGC1333IRS4 Polarization (E-vector) 400AU Expected Interstellar B-field Girart, Rao, Marrone 06

Hour glass 型 磁場ポロイダル Bp=(Br, Bz, 0) 磁場トロイダル Bt=(0, 0, Bf) 角運動量輸送 トルク　FfBp*Jp Bt 砂時計型磁力線 　　がトロイダル磁場を持つかどうか？

Magnetic Field Drives Outflow? --- Look for evidence of magnetic drive --- Search for rotation motion Intensity-weighted line-of-sight velocity 2000AU Launhardt +09 CB26 ANOTHER EVIDENCE IS MAGNETIC FIELD CONFIGURATION.

Shape of Magnetic Field? Magnetocentrifugal wind acceleration Strong field case Magnetic pressure gradient acceleration Weak field case

align align align ??? Relationship of Outflow and Magnetic Field local B global B IRAS 16293-2422 B Wolf et al (2003) align align 1800AU local B // Outflow but global B not // Outflow. B Outflow L1551 IRS5 Magnetic field B B How do the global and local magnetic field? align ??? Polarization of thermal dust emission map SCUBA 850mm Tamura et al (1995) 1mm radio polarization

7000AU Girart+ 09

Observational Visualization of Outflow Accretion onto a First core Runaway isothermal collapse Around 1st core molecular outflow is accelerated. 2D axisymmetric barotropic MHD simulation

Post process: Dust thermal emission Method Post process: Dust thermal emission Geometry of Observation W and B are taken in z-direction Line of sight: Polarization obs. Dust thermal rad. Line of sight unit vector of obs. grid:

Stokes parameter z y x z y x z V=0: linear polarization dx-dy=0,p y U=0  ex=0 or ey=0 Q=0  ex=ey x

Stokes parameter I, Q, and U Oblate/prolate dust is aligned in the B-field direction. (Draine & Lee 85, Fiege & Pudritz 2000) Talks by Lai, Mendez, … C: difference of cross sections perp and parallel to B R: reduction factor due to imperfect grain alignment F: reduction factor due to turbulent B-field c=r/nd g: angle b/w B and plane of the sky. y: angle b/w projection of B and h-axis Relative Stokes parameter (Wardle & Konigl 90)

Polarization direction Polarization degree

Consider a ray n passing an observation grid point x0 Find a point of intersection toward the direction –n  x1 Find a point of intersection toward n  x2 Similar procedure to x1 Integrate q, u, S, and S2 from x1 to x2

Integration on the nested grid hierarchy L=0 For L=0,target-1 do begin integrate from outer to inner boundaries for grid L End for Integrate grid (target) For L=target-1,0,-1 do begin integrate from inner to outer L=1 L=2 Observation Grid (target)

3D simulation box Observation grid

偏光度は軸上から見たとき小さく、円盤上から見たときに大きくなる。 (2) 全強度分布はディスク形状を示す。 -- 円盤  低偏光度 ( ) 特徴: 偏光度は軸上から見たとき小さく、円盤上から見たときに大きくなる。 (2) 全強度分布はディスク形状を示す。 -- 円盤  低偏光度　　　　 ( ) (3) 偏光度分布は縦軸に対して対象でない -- q=30~60o 近辺で、低偏光度の領域は右上から左下に伸びる。 (q=0 と 90o は点対称と線対称) (4) 磁場は砂時計型形状を示す。 -- 砂時計の軸は、全強度分布の主軸と外れる場合もある。 L=3 星なし期 色：偏光度 等高線（黒）：面密度 ベクトル：偏光のBベクトル ~ 星間の磁場ベクトル 軸上 10000AU 円盤上

L=5 prestellar stage (close-up) 10000AU L=5 prestellar stage (close-up) Pole-on 2000AU edge-on

L=3 protostellar stage Pole-on c edge-on

(a) そのなかにより高い偏光度の領域（加速領域？）を 含むq=30~60deg. (b) 理由: 特徴： (1) アウトフローは低い偏光度の領域。 (a) そのなかにより高い偏光度の領域（加速領域？）を　 　　　　含むq=30~60deg. (b) 理由: * アウトフローの外は、磁場が円盤に垂直。 * 内部は、トロイダルの磁場が卓越、天球に垂直 (2) 円盤は低偏光度領域 (3) ただし、ｙ軸に対して対称ではない。 L=5 原始星期 色：　偏光度 等高線（黒）：　面密度 ベクトル： 偏光のBベクトル ~ 星間の磁場 L=5 protostellar stage 軸上 円盤面上 2000AU

What brings the asymmetry? Artificially only Troidal Field Bf was left Symmetric Artificially only Poloidal Field (Bz, Br) was left Symmetric

Reason why asymmetry against h-axis Rotating Pseudo-disk Looking from q~45deg

Weak B-Field Case Characteristics: (1) Intensity distribution  round (2) Polarization pattern shows asymmetry against h-axis. (3) B-field indicates hour glass shape -- the axis is not perpendicular to intensity’s major axis. Weak B-Field Case L=3 protostellar stage Pole-on 10000AU edge-on

L=5 protostellar stage Characteristics: (1) Outflow traces a low polarization region, (a) it contains a polarized region around the h-axis. (b) which comes from the pinched pol. field (2) Viewing from pole direction, azimuthal polarization pattern is emphasized. L=5 protostellar stage Pole-on 2000AU edge-on

L=6 protostellar stage Pole-on edge-on

結論 トロイダル磁場Bfの存在は、偏光パターンに現れる。 トロイダル磁場Bfは、アウトフローの磁気力加速の証拠

ダストの整列を仮定して、偏光マップを作成した。 (星無し期) 強度分布 ディスク構造. ( a=1モデル) ディスクは低偏光度 丸い分布 (弱い磁場 a=0.01 モデル) 偏光パタンはh-axis軸について非対称. q=30~60deg の場合：低偏光度領域が上右から左下へ向かって分布。 q=0 と 90o は点、軸対称 磁場は砂時計型。その軸は必ずしも偏光率分布のそれと一致しない。 (原始星期) 双極分子流は、低偏光度領域として観測。 この中に高偏光度の加速領域を含む。q=30~60o. η軸付近に横方向に偏光した領域を含む。 回転軸jから見ると φ方向の偏光パタンが観測される。 双極分子流内にBfが観測されれば  磁場が駆動するアウトフロー PASJ Feb.2011 in press

偏光観測の困難 ALMA Early Science Phase では努力目標。 我が国にはミリ波の偏光観測を専門とする研究者がいない。 バックアップ計画 Vφ測定　特に加速領域 Vφ entrainment<Vφmagnetic driven outflow Entrainment modelをどう作るか。 大半径sink cell 磁場無しモデル