ISM & AstrochemistryLecture 6

Stellar Evolution

Star Death

IRC+10216 (CW Leo)

• Nearby (~130 pc) high mass-loss carbon star (AGB)

• Brightest object in the sky at 2 microns – optically invisible

• Carbon dust envelope detected out to 200’’ = 25,000 AU ( ~ 1 lt yr)

• Molecular shells at ~ 1000 - 4000 AU

• >60 molecules detected: CO, C2H2, HC9N ...

• Newly discovered anions C8H-, C6H-,

C4H- , C3N- , C5N-

• Recent detections of H2O, OH and H2CO

Figures from Leao et al. (2006) Lucas and Guelin et al. (1999)

IRC+10216

HC3N J = 5-4 emission in plane of sky shown as contours

Molecular shells/arcs shown as thick green lines

Contour is optical V-band image in false colour

(Trung & Lim 2008, ApJ, 678, 303)

Molecular emission is correlated with dust shells/density enhancements

IRC+10216

Comparison of HC3N J = 5-4 (greyscale) with HC5N J = 9-8 (contours) using the VLA (Trung & Lim 2008)

Very close spatial correlation between HC3N and HC5N emission (and presumably abundances)

Chemical Structure of AGB CSE

LTE Chemistry in AGB Stars

CS

Maser emission is an excellent probe of conditions near the photosphere, particularly if driven by IR pumping.

In IRC+10216, HCN detected up to v=4, SiS detected up to v=0, J=15-14 (Exposito et al. 2006)

SO

Effect of C/O Ratio on Fractional Abundances

Dust Formation in AGB StarsDust formation in AGB stars requires:

(i) a cooling flow

(ii) high collisional rates

(iii) time

Shocks caused by pulsations are critical – they give large density enhancements, strong post-shock cooling, and lift material slowly away from stellar surface

Spatial scale is less than 5 stellar radii

Probe with vibrationally excited/maser species (line-widths less than terminal velocity/IR pumped) – eg thermal v=1SiS emission (Exposito et al. 2006)

SMA J = 5-4 SiO observations (Schoier et al. 2006) show SiO much more abundant than LTE value in IRC+10216 at 3-8 stellar radii

but lower at larger radii – condensation on to dust grains?

Pulsations and Shock WavesDensity and temperature profiles as a function of phase following a 20

km/s shock at a radial distance of 1.2 stellar radii

Shock propagates outward with shock velocity decreasing with increasing radial distance

Density 1-6 1015 cm-3 T = 4500-1500K

Shock ChemistryInput – LTE abundances for IRC+10216 parameters

Each cycle breaks down into three zones:

(i) Immediate post-shock – H2 and other LTE molecules destroyed

(ii) Post-shock – H2 and other molecules form

(ii) Dynamical time less than chemical time – freeze-out of abundances

(iv) No significant water formed in shock

Abundances at the end of the pulsation zone can be very different from LTE

SiO enhanced, HCN, CS destroyed in C-stars

(Willacy & Cherchneff 1998)

Shock ChemistryFormation of Molecules in O-rich and S-type Stars

Form carbon-bearing molecules in O-rich stars – C/O < 1:

(i) HCN increased by 6 orders of magnitude above LTE value

(ii) CO2 increased by 3 orders of magnitude

(iii) CS increased by 3 orders of magnitude

(Duari, Cherchneff & Willacy 1999)

For S-type stars (Chi Cygni) – C/O ~1:

HCN in inner envelope increased by 4 orders of magnitude (vibrationally excited HCN detected here – IR pumping within 33 stellar radii)

(Duari & Hatchell 2000)

Photochemistry in CSEsDestruction of Parents by IS UV Radiation Field

Self-Shielding

H2 – very reactive, daughter (H atoms) unreactive

CO – very unreactive, daughters (C, C+) very reactive

Cosmic ray ionisation

f(H3+) varies as r2

N(H3+) ~ 1012 cm-2 for CRI rate of 10-17 s-1, an order of magnitude less than

that detected in the interstellar medium

Photodissociation and photoionisation

Acetylene is the species which determines the complexity of the hydrocarbon chemistry

Photochemistry in CSEsShell distributions – the photodestruction of acetylene

IRC+10216

Acetylene has a relatively large photoionisation cross-section

Ions and radicals form in outer CSE where both density and UV field are relatively large

(Millar, Herbst & Bettens 2000)

Hydrocarbon formationShell distributions – rapid formation of hydrocarbons

IRC+10216C2H and C2H2

+ are both very reactive with acetylene and derivative species

Peak abundances occur at slightly larger radii as size increases

Degradation of grains may give inverse behaviour

(Millar, Herbst & Bettens 2000)

Cyanopolyyne formationShell distributions – rapid formation of hydrocarbons

IRC+10216Neutral chemistry important in forming cyanopolyynes and other molecules

(Millar, Herbst & Bettens 2000)

(Guelin et al. 2000)

Anion formationAnion column densities ~ 0.1 - 0.2 neutral column densities

IRC+10216High electron fraction and high density allow formation of abundant anions

C4H-, C6H-, C8H-, C3N- recently detected in IRC+10216

(Millar, Herbst & Bettens 2000)

Modelling the circumstellar shellsModelling the circumstellar shells

Leao et al (2006):Leao et al (2006): Concentric ringsConcentric rings

H2 radial density profile time-evolution

Density-enhanced packets of gas and dust move outwards through the CSE, their chemistry evolves with time Cordiner & Millar, ApJ, 697, 68 (2009)

2 arcsec thick shells (90 yr enhanced mass-loss)

12 arcsec apart (540 yr between events)

Density enhancement = 5

IRC+10216 – Standard vs Rate06 + Rings

Rings provide additional extinction – so parents are destroyed further out

HC3N no longer most abundant cyanopolyyne – effect of Rate06

Anion results

IRC+10216 results

Hydrocarbon radicals

No shells – number density

Shells – 3mm emission

No shells

Shells

3mm emission

Hydrocarbon anionsHC3N and HC5N

Comparison with 3mm emission observations

The effect of anions in the CSE

Free electron abundance reduced increases number of cations

Anion abundances > electron abundance within part of envelope

Formation of Water

Willacy, ApJL, 600, L87 (2004)

In C-rich source (IRC+10216),

f(H2O) ~ 0.4-2.4 10-6 and

f(OH) ~ 4 10-8.

Possibility – evaporation of comets

(formed in O-rich material in the neighbourhood – presence of HDO?)

Shock chemistry ? No

Formation of water by Fischer-Tropsch catalysis

CO + 3H2 CH4 + H2O

on iron grains

Need few percent of Fe in iron or F-Ni alloy.

IRC+10216 summary

• Successfully model C6H-, C8H-, C3N- abundances• C6H- and C8H- formed by radiative electron

attachment, C3N- formed by CnH- + N• C4H- overproduced by ~ 100 times• CN- and C2H- predicted to be observable in the inner

CSE even though the electron attachment rate is very low, due to reaction of H- with HCN and C2H2

• Density-enhancements in the CSE may result in narrow emission rings / arcs of carbon-chain species and their anions

Photochemistry in O-rich CSEsDestruction of Parents by IS UV Radiation Field

Self-Shielding

H2 – very reactive, daughter (H atoms) unreactive

CO – very unreactive, daughters (C, C+) very reactive

H2O – can self-shield but only weakly, daughter (OH) very reactive

Photodissociation and photoionisation

Water drives chemistry through production of OH

Water photoionisation small so neutral-neutral chemistry dominates

Photochemistry in O-rich CSEsRadical reactions produce oxides

OH reacts rapidly with other daughters

O to produce O2

N to produce NO

S to produce SO

SO to produce SO2

Si to produce SiO

P to produce PO, etc

ProblemsC-rich AGB CSEs

Detection of water, formaldehyde and OH in outer envelope of IRC+10216

Pulsational shocks – no, too little formed – and in inner region

Photochemistry – no, not enough O freed from parents

Fisher-Tropsch-Type chemistry (grains) – possibly

Evaporation of comets - possibly

O-rich AGB CSEs

Formation of carbon-bearing molecules

HCN, HNC, CN, CS, H2CO, OCS

Not enough carbon released by photodissociation of CO

Pulsational shocks – can help, but in inner region

Grain chemistry – not yet studied