Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://www.atnf.csiro.au/research/masermeeting/web_papers/507_Bujarrabal.pdf Äàòà èçìåíåíèÿ: Mon Mar 26 03:46:10 2007 Äàòà èíäåêñèðîâàíèÿ: Sun Dec 23 12:41:55 2007 Êîäèðîâêà: Ïîèñêîâûå ñëîâà: annular solar eclipse

THE STUDY OF THE INNER SHELLS AROUND EVOLVED STARS FROM SiO MASER EMISSION

ValentÌn Bujarrabal
Observatorio AstronÑmico Nacional, Spain v.bujarrabal@oan.es

+

J. Alcolea, F. Colomer, J.F. Desmurs, J.R. Pardo, C. SÀnchez Contreras, R. Soria-Ruiz


SiO MASERS: BASIC OBSERVATIONAL RESULTS

Rotational transitions: J=1­0, 2­1, up to 8­7 (345 GHz) EXCITED vibrational states, v = 1, 2, 3, 4 ( 1800 K) in more than 500 O-rich stars, and a few S-type objects Bright AGBs (Miras, semiregulars, irregulars), OH/IR, SGs, a few PPNe v=1 J=1­0, 2­1 and v=2 J=1­0 dominate Spiky, narrow profiles, around stellar velocity (V 2 ­ 10 km/s) Good correlation with IR (8 µ v=1) also 29SiO and 30SiO (mainly v=0) (e.g. Bujarrabal et al., A&A, 175, 164; Alcolea et al. A&A, 211, 187; A&A, 231, 431)


SiO MASERS: BASIC PUMPING PROPERTIES

v 1 states populated from v = 0

+ fast radiative de-excitation, A 5 s-1 !!

> 1 => Prad A A/ 1/gu

=> selectively over populates levels with higher J => chain of masers (mutually reinforced)

High excitation required

=> close to the star or radiative excitation (observational suppor t) ?

Collisional (theoretical effor t)

Effects of line overlap ? (with lines of H2O or of SiO itself)


VARIABILITY OF SiO MASERS FROM AGB STARS

(AGB stars are strong pulsators, with periods 1 yr.) SiO masers in them are also strongly variable with the same periods. 11-years, shor t spaced monitoring: 15­20 sources; J=1-0, v=1,2; 4000 spectra => Systematic phase lag with respect to optical cycle: + 0.1-0.15 (= IR/optical lag). Systematically IN phase with IR cycle (within 0.05). (Pardo et al., A&A, 424, 145) Very difficult to explain if the pumping is collisional: A shock needs several periods to reach the masing shell from the photosphere, and such a time must vary from star to star. Naturally explained if radiatively pumped.




PHASE OF SiO MASER MAXIMA, v=1,2 J=1-0, in AGB stars from Fourier analysis v=1 v=1 v=2 v=2 average period name

peak area peak area SiO max. phase w.r.t. op. max.: 0.09 0.08 0.15 0.14 0.15 0.12 0.17 0.19 0.13 0.13 0.12 0.10 0.19 0.14 0.01 0.07 0.10 0.09 0.11 0.13 0.11 0.11 0.12 0.13 0.10 0.17 0.18 0.03 0.05 0.12 0.12 0.10 0.14 0.13 0.07 0.14 0.18 0.09 0.09 0.16 0.10 0.13 0.13 0.16 0.11 0.13 0.17 0.05 0.08 0.13 0.12 0.01 -0.02 -0.01 467.5 IK Tau (IR var.) 650.5 IRC+10011 (IR var.) 461.5 GY Aql (smr) 397.0 362.0 434.5 331.5 403.5 384.5 314.0 374.5 R Aqr R Cnc R Cas o Cet U Her W Hya (smr?) R Leo R LMi 542.5 TX Cam

average : 0.06 0.03 -0.01 -0.04 0.04 0.02 -0.06 -0.07 average :

SiO max. phase w.r.t. IR max.:


RELATIVE SPATIAL DISTRIBUTIONS OF THE DIFFERENT MASER LINES

In general: rings at a few stellar radii

(see e.g. Diamond et al. 1994, ApJ, 430, L61; Cotton et al. 2006, A&A, 456, 339)

v=1, J=1­0

vs

v=2, J=1­0 :

Nearby, but ver y rare coincidences v=2 J=1­0 slightly (but systematically) inner ring Each transition 'avoids' the other Expected from theor y(ies): require very different excitations (Desmurs et al. 2000, A&A 360, 189)


TX Cam v=1,2 J=1­0
(Desmurs et al. 2000, A&A 360, 189)


IRC +10011 v=1, J=1­0: greys v=2, J=1­0: contours
(Soria-Ruiz et al. 2004, A&A 424, 145)


RELATIVE POSITIONS OF THE DIFFERENT MASER LINES

v=1, J=1­0

vs

v=1, J=2­1 :

Again, concentric rings. But very different distributions ! v=1 J=2­1 outwards (in the few observed cases)

Totally unexpected from theoretical considerations (maser chains, similar excitation) Invoke line overlap ?


IRC +10011 v=1, J=1­0: greys v=1, J=2­1: contours
(Soria-Ruiz et al. 2004, A&A 424, 145)


RELATIVE POSITIONS OF THE DIFFERENT MASER LINES

29

SiO v=0, J=1­0

Again, concentric rings. TOO SIMILAR TO THE OTHERS ! In spite of the ver y low excitation required A result in fact expected from low-resolution observations (systematically similar spectra as in other lines) Completely unexpected from theory: Masers at long distances, radial amplification => double-peak profiles (Robinson & Van Blerkom ApJ, 249, 566; Deguchi & Nguyen-Q-Rieu, A&A, 117, 314) Overlap between 28SiO and 29SiO ro-vibrational lines ? (GonzÀlez-Alfonso & Cernicharo, 1997, A&A, 322, 938)


IRC +10011 (new epoch)
28

SiO lines

+ 29SiO v=0, J=1­0
(Soria-Ruiz et al. 2005, A&A 432, L39)


SOME REMARKS ON LINE OVERLAP AND SiO MASER PUMPING

LINE OVERLAP Two lines, perhaps of different molecules, coincide in frequency. The external photon field 'seen' by one of the lines can be strongly affected. In SiO: always ro-vibrational lines affected

OBSERVATIONAL FACTS THAT COULD BE EXPLAINED BY LINE OVERLAP: H2O v2=0,127,5 ­ v2=1,116,6 +
28

SiO v=2,J=1 ­ v=1,J=0

· Anomalously weak emission in v=2 J=2­1 (in O-rich stars) Olofsson et al. (1985, A&A, 169, 179) · Anomalous distribution of 28SiO v=1, J=2­1 Increase of photons due to emission in H2O v2=0,127,5 ­ v2=1,116,6 => overpopulation of SiO v=2,J=1 and depopulation of SiO v=1,J=0


------ · · · · · · ·· greys : ­­­­

v=2, J=1­0 v=2, J=2­1 v=1, J=1­0 v=1, J=2­1

Predictions for H2O-poor (no overlap, Cyg) or -rich (overlap, IRC +10011) CSEs. explains the v=1 J=1­0, J=2­1; v=1 J=2­1 distributions and the weak v=2 J=2­1 (Soria-Ruiz et al., A&A, 426, 131; Bujarrabal et al., A&A, 314, 883)


SOME REMARKS ON LINE OVERLAP AND SiO MASER PUMPING

EFFECTS ON 29SiO and 30SiO MASERS: · Could explain the 29SiO v=0, J=1­0 maser, in par ticular its distribution
28

SiO v=2,J=4 ­ v=1,J=3 +

29

SiO v=1,J=1 ­ v=0,J=0

· And the other 29SiO and 30SiO masers? At least :
29 29 29 30 30 30

SiO v=0 J=1­0, J=2­1, J=4­3, J=5­4 SiO v=1 J=1­0, J=3­2, J=4­3, J=6­5 SiO v=2 J=6­5 SiO v=0 J=1­0, J=2­1, J=4­3, J=5­4 SiO v=1 J=4­3 SiO v=2 J=4­3 P 1/100 (for a given line of 29SiO, with 28SiO).

Too many coincidences?

Why not in other molecules?


OH 231.8+4.2: SiO MASERS IN YOUNG PNe (PPNe)

Alcolea et al. (2001, A&A 373, 932), Bujarrabal et al. (2002, A&A 389, 217)


SiO v=1, J=2­1 MASER (86 GHz) IN OH 231.8+4.2

Remarkably stable three-peak structure. Not due to lucky clump combination in the inner envelope (Vt/R > 1) . Expected from masers in rotating disks (e.g. Deguchi et al. 1983, ApJ, 264, L65)

unpublished data from SÀnchez Contreras et al., in preparation


OH 231.8+4.2: VLBI MAPPING OF SiO v=2 J=1­0 (43 GHz)

Spatial distribution and velocities compatible with rotation for a reasonable stellar mass (plus infall?) Perpendicular to the nebular axis! But, unfor tunately the central peak was undetected Single-dish monitoring plus maps => rotating tiny disk

SÀnchez Contreras et al. (2002, A&A, 385, L1)


OH 231.8+4.2: ABSOLUTE POSITION MAPPING OF SiO AND H2O MASERS

H2O placed along the axis => suppor ts the rotating disk ­ axial structure interpretation Desmurs et al. (2007, A&A, in press; SEE ALL DETAILS IN POSTERs)


SEARCH FOR ROTATING DISKS IN YOUNG PNe

INNER ROTATING DISKS IN PPNe: IMPORTANT FEATURES !

· Explain the ubiquitous bipolar ejections in PPNe : ­ implying enormous amounts of momentum (radiation pressure not enough) ­ leading the shaping of PNe (systematically axial) · Explain other obser vational results : ­ NIR excess due to hot dust in stable reservoirs (i.e. keplerian disks) ­ Anomalous atmospheric abundances probably due to reaccretion ­ ...

But only well detected in one PPN: the Red Rectangle (from CO mapping, A&A, 441, 1031)


SEARCH FOR ROTATING DISKS IN YOUNG PNe

Inverse angular momentum problem: Gas ejected by an AGB star must win momentum from a companion! Detection in thermal emission require big disks (> 1 ), like in the Red Rectangle. But such big disks require a stellar companion with a relatively high mass ( 0.5 M ) Substellar companions or planets cannot accelerate so strongly the circumstellar gas => the disk sinks to a small size (R M2) Companion
<

0.1 M => very small disks

=> VLBI obser vations of masers Observations of SiO masers around post-AGB stars may be the key ... work in progress