Ionized gas in massive star

forming regions

Guido Garay

Universidad de Chile

Great Barriers in High-Mass Star Formation

Townsville, September 15, 2010

Outline

HII regions Origin: molecular gas surrounding a high-mass star

Jets Origin: gas from putative star-disk system

Review the characteristics of sources of ionized gas within massive star forming regions. Depending on the origin of the gas, we distinguish two type of sources: HII regions and Jets

Aim

HII regions

Physical parameters

Source of ionization: UV photons emitted by embedded young luminous high-mass stars.

Based on their sizes, densities and emission measures, three classes of HII regions have been identified: Class Diameter Density EM Ref.

(pc) (cm-3) (pc cm-6)

Compact 0.1 < D < 0.5 >103 >106 Mezger et al. (1967)

Ultracompact 0.02 < D < 0.1 >104 >107 Wood & Churchwell (1989)

Hypercompact D < 0.02 >105 >108 Kurtz (2004) Hoare et al. (2007)Time discovery line increase in angular resolution, observing frequency and sensitivity.

Rather than discrete, there is a continuous distribution in the value of the parameters

Garay & Lizano (1999)(see also Churchwell 2002)

UCHII

CHII

HCHII

UCHII

CHII

HCHII

Rather than discrete, there is a continuous distribution in the value of the parameters

… + more data from recent surveys:

DePree et al. (2004) Sewilo et al. (2004) Garay et al. (2006) Murphy et al. (2010)

Garay & Lizano (1999)(see also Churchwell 2002)

HCHII

HCHII

HCHII are uncommon

There is a significant correlation between the parameters

How do we explain these trends?

Classical model: spherical bubble expanding in a uniform density medium (Spitzer 1978)

7

6

7

4

4

712)(

4

71)(

s

soe

s

sS

R

tcntn

R

tcRtR

no: ambient density, cs: sound speed

Rs: initial Stromgren radius pc

n

cm

s

NR

o

US

3

2363

1

148

10

100032.0

Lines indicate model relations for: no=106 cm-3 and Nu=3x1048 s-1 (upper)

no=106 cm-3 and Nu=3x1045 s-1 (lower)

Evolutionary sequence

O7.5

B1

O7.5

B1

This simple dynamical model

HII regions reach pressure equilibrium with ambient medium in a time scale of a few 104 yrs.

Age of compact HII regions could be much larger than this value.

Hypercompact are the youngest, smaller and denser versions of UCHII regions.

They should give us information about the process of high-mass star formation in the earliest evolutionary stages.

: single star, no dust (see Churchwell’s talk for dust considerations)

Massive stars are born in a high density ambient medium.

Densities are similar to those of hot molecular cores Hot cores are the precursors of UCHII regions.

1.2mm dust continuum

Hypercompact HII regions

Characteristics of their large scale (~1 pc) surroundingsDust continuum and molecular line observations in high densitytracers HCHII are found inside massive and dense cores.

1 pc

Massive and dense cores

Very dark even at IR (IRDCs)

8 μm MSX

Massive and dense core

e.g., IRAS 16272-4837

Physical parameters: R ~ 0.4 pc M ~ 4x103 M

n(H2) ~ 6x105 cm-3

Δv ~ 6 km s-1

Highly centrally condensed

n r -1.5

Optically thick lines

Optically thin lines

large scale infalling motions

About thirty massive dense cores known with infalling motions Snell & Loren 1977, Welch et al. 1988, Garay et al. 2002, 2003, Wu & Evans 2003

Massive and dense core undergoing intense accretion phase

Vinf ~1 km s-1 Minf ~ 1x10-3 M yr-1

Dynamical state:

Most in virial equilibrium

Few undergoing large scale inflow motions

e.g., IRAS 16547-4247

Whether massive stars are formed at the center or migrate there, is still an open question.

Images: 4.8 GHz emission (HCHII region) Contours: 1.2-mm emission (Massive core)

IRAS 13291-6249 IRAS 15520-5234 IRAS 17016-4124

HCHII regions typically found at the center of massive cores

1 pc 1 pc 1 pc

Where are HCHIIs located within massive and dense cores?

Continuum spectraDue to their high emission measure HCHIIs are expected to have turnover frequencies, νto , greater than 10 GHz.

Range of power-law too wide to correspond to the transition from optically thick to optically thin regimes in a constant density region.

=1.0±0.1

Franco et al. (2000)

Below νto, HCHII regions frequently show power-law spectra over a

wide frequency range, Sνν , with typically ~1.

GHzK

T

cmpc

EM eto

64.0

4

48.0

69 10100.16

Possible explanations for the power-law:

HCHIIs possess density gradients.

For a region in which the electron density goes as n r -β Flux density depends with ν as Sν ν, with =(4β-6.2) /(2β-1) Angular size depends with ν as θν νγ, with = -2.1/(2β-1)

e.g. HCHII G28.20-0.04:

Sν ν1.1 n r -2.8

( ν-0.5)

Is the expected size dependence with ν actually observed?

e.g., Avalos et al. (2005)

Shell model: β = 2.8 Ri = 0.0063 pc

Ro = 0.055 pc

ni = 6x105 cm-3

HCHIIs are hierarchically clumped structures.

Ignace & Churchwell (2004)

Ensemble of clumps with a distribution of optical depths produce:

Power-law spectral index covering a wide frequency range

No dependence of angular size with frequency

Caveat: Contribution from dust and free-free emission at frequencies of ~50 GHz can be of the same order, affecting the spectral index interpretation.

e.g., G75.78+0.34-H2O

n r - 4

Franco et al. (2000)

Cte. density HCHII region + hot dust diskKurtz (2010)

Radio recombination lines

HCHII regions often have broader line widths than UCHII regions.

Origin of line broadening?

Possible mechanisms:

Large-scale organized motions: rotation expansion infall

Pressure broadening v n7.4 ne

vHCHII > 40 km s-1

vCHII ≈ 30 km s-1

Velocity gradient of 103 km s-1 pc-1. Rotating torus with a velocity of 5 km s-1 at 0.005pc

High angular resolution observations indicate that ordered motions are present.

H92

H92

e.g. G28.20-0.04 N H53

Sewilo et al. (2004)

v74 km s-1 Sewilo et

al. (2008)

VLA observations in two epochs with Δt = 5 years,

Expansion motions with a velocity of about 35 km s-1.

G5.89-0.39

Acord et al. (1998)

In most cases the bulk motions are not able to explain the observed linewidths.

Line v ne

(km s-1) (cm-3)H66 50.9 2.2e6H53 32.5 4.7e6H30 26.8 ---

Pressure broadening seems the most important contributor.

In addition to the high densities, the RRL observations indicate the presence of density gradients ( at the higher frequency seeing deeper into the region).

Line v ne

(km s-1) (cm-3)H92 74 .4 3e5H76 57.6 9e5H53 39.7 7e6

e.g. W51e2

For G28.20-0.04 N:

H66H53 H30

Models

Accretion flow (Keto 2002) Ingredients: Gravity + Ionization

Ri : radius of ionization equilibrium

Rg : gravitational radius

(escape radius for ionized gas)

Two characteristicradii

If Ri < Rg ionized gas can not expand

gravitationally trapped HII region

pcM

M

c

GMR

sg

200007.0

2pcn

cm

s

NR

o

Ui

3

2363

1

148

10

100032.0

However, Rg < RHCHII

Accretion disk (Keto 2007) Ingredients: Gravity + Ionization + Rotation of primordial cloud

radius at which centrifugal and gravitational forces balance

Third characteristicradius

.6

53

2

M

L

GRd

Actual situation depends on the relative values of the three R’s.

e.g., if Ri > Rd, Rg

radial wind driven by the thermal pressure of the ionized gas.

Density gradient of photoevaporatedwind produces spectral index of ~ 0.8.

Photoevaporating disks Ingredients: Keplerian circumstellar disk + luminous YSO

Hollenbach et al. (1994)Lizano et al (1996)

Rg = 0.0007 pc , Rd = 0.0015 pc

Lugo et al. (2004)

MWC 349A

In the case of low-mass protostars, observations clearly showa Disk-Jet symbiosis.

Disk: Protostar grows by accreting from disk.

Jet: Carries away angular momentum and mechanical energy from disk into the surroundings, allowing accretion to proceed.

Accretion disk + jet

Ingredients: Gravity + Ionization + Rotation + Magnetic field

If this symbiosis also holds for high-mass protostars, then

Jets

Jets ≡ Highly collimated ionized flows that emanate from YSOs.

Emission mechanism: thermal free-free emission. Detectable as weak radio sources.

Source of ionization: UV photons from shocks produced by the impact of neutral collimated wind on the surrounding high density material.

Jets are almost always found in the case of low-mass protostars.

Thought to be the “base” of the large scale outflow phenomena like the bipolar outflows and HH systems.

Are young massive stars associated with ionized jets?

Garay & Lizano (1999) reported a handful of ionized thermal radio jets associated with massive YSOs, all of which have luminosities < 2x104 L.

Source Lumin. S References (L) (GHz) (mJy)

Cepheus A HW2 1.0x104 8 10 0.6 Rodríguez et al. 94

IRAS 20126+4104 1.3x104 8 0.2 -- Hofner et al. 99

W75N(B) VLA1 1.5x104 8 4 0.7 Torrelles et al. 97

IRAS 18162-2048 1.7x104 5 5 0.2 Martí et al. 95

Observed flux density and size dependence with frequency

biconical thermal jet

S0.7

-0.6

Cepheus A HW2 jet

Rodriguez et al. 1994

L = 700 AU M v = 6x10-4 M yr-1 km s-1

The number of detections has increased during the last decade and detections have been made towards progressively more luminous YSOs.

Source Lumin. S References (L) (GHz) (mJy)

G35.2-0.7 N 1.6x104 9 0.4 >1.3 Gibb et al. 03

IRAS18089-1732 3.2x104 9 1.1 0.58 Zapata et al. 06

CRL2136-RS4 5.0x104 9 0.56 1.2 Menten & Tak 04

IRAS 16547-4247 6.2x104 9 6 0.5 Garay et al. 03

G345.494+1.468 7.0x104 9 9 0.85 Guzman et al. 10

G331.512-0.103 2.0x105 9 166 1.1 Bronfman et al. 08

Triple radio continuum source toward IRAS 16547-4247 (L = 6104 L)

Garay et al (2003)

0.3 pc

Triple radio continuum source toward

IRAS 16562-3959 (L = 7104 L)

Guzman et al. (2010)

0.07 pc

ATCA survey of radio continuum emission toward luminous massive proto-stellar objects

Jets are found associated with luminous YSOs.

Thermal jet

Lobes

4.8 GHz

Possible explanations:

Different formation mechanism?

Obscured by bright CHII region?

Short timescale for jet phase?

Bipolar outflows in high mass protostar have dynamic ages of 105 yrs >longer than the K-H time of the jet/disk stage of 104 yrs. jet may turn off and the large scale outflow will still persist as a fossil for a relatively long time.

Molecular outflows (large scale) are relatively frequent towards high-mass YSOs, however jets are still rare…

Characteristics of jets associated with high-mass YSOs

Velocity : 1000-3000 km s-1

Size : 0.01 pc

Momentum rate: 10-2 - 10-1 M km s-1 yr-1

Jet luminosity

103 times more luminous and energetic than low-mass jets !

Rodriguez et al. 2007

High-mass jets

Low-mass jets

Jets associated with luminous YSOs are powerful

Mom

entu

m r

ate

Summary HII regions HCHIIs are probes of the earliest phase of evolution of regions of ionized gas excited by young high-mass stars.

However, still far from understanding them:

density gradients : n r -2.5 ? Sν ν1.0 ensemble of clumps ?

Disk rotation ? Broad linewidths Outflows ? Pressure broadening ? Morphologies ?

Are the high-mass stars exciting HCHII regions formed at the center of massive and dense cores or they migrate there?

Ionized jets

Jets are found associated with high-mass YSOs (up to luminosities of 2x105 L).

They are 103 times more energetic and luminous than low-mass jets.

They are rare.

Open questions:

Disk wind?

Which is the nature of the driving mechanism?

X-wind?

Which are their lifetimes?

Do jets rotate?

To address these questions we need to probe HCHIIs and ionized jets with high spatial resolutions (< 2 AU) and high sensitivities.

ALMA

Variability of radio emission

Lobes show a clear decrease in radio flux density of 20-30% over 12 years.

Franco-Hernandez & Rodriguez (2004)

e.g., NGC 7538 IRS1

Possible explanations:

Ionizing photon flux from star is decreasing.

Inyection of fresh gas in core steals ionizing photons for the lobes

Spectral energy distribution At ν < 30 GHz ( > 1cm): free-free emission from ionized gas

Ultracompact HII regions

SEDs of UCHII regions

Kurtz et al. 1994Modified Planck function

ν -0.1

ν2

1

1

1

At ν > 300 GHz ( < 1 mm): thermal dust emission from warm cocoon

Shell 28%

Cometary 14%

Core-halo 16% Bipolar 8%

Morphologies

+ spherical, irregular, and unresolved morphologies (Churchwell 2002)

UCHII regions exhibit a variety of morphologies

Morphologies depend on the characteristics of the exciting star and of theenvironment, as well ason their interaction.

Observational consideration

yrM

M

SUNHK

3

4

20107

Kelvin-Helmholtz time

Massive stars spend short time in the pre-main sequence:

3

20003.0)(

SUNMM

MN

Rate of massive star formation in the Galaxy:

6

20200)()(

SUNHKPMS M

MMNMN

Massive protostars are very rare

How many massive protostars we expect to see in our Galaxy?