Forum Springer Astron. Astrophys.
Forum Whats New Search Orders

Astron. Astrophys. 331, 317-327 (1998)

Previous Section Next Section Title Page Table of Contents

6. Discussion

IRC+10420 is a peculiar star showing large mass loss rate (5 10 [FORMULA] yr-1) and high expansion velocity (35 kms-1). In the OH envelope, many complicated physical processes may come into play, for example the turbulence, the non-uniform distribution of OH molecules, the departure from spherical geometry, thus making the modeling of OH maser emission particularly difficult. The striking difference between the observed maser line profiles suggests that they originate from different regions of the envelope, contrary to our models which indicate that the 1612 MHz line and the main lines come from the same region (Fig. 10). The appearance of a middle peak close to the stellar velocity in the observed main line spectra (Fig.  8-9) probably indicates that the OH envelope has multiple shells. The fact that only model 1 produces main line maser emission proves that the standard model of OH main line masers (smooth outflow of material combined with FIR line overlap) is not realistic enough to explain the maser emission of IRC+10420. Recent observations with high angular and spectral resolution by Richards (1997) of water and OH masers from several supergiant stars show that OH masers come from clumps and the distribution of these clumps is not spherical. OH main line masers appear at smaller radii in comparison with the 1612 MHz satellite maser.

In addition, maser profiles calculated from either model 1 or model 2 are too narrow in comparison with observations. The wide maser peak ([FORMULA] 10 kms-1) cannot be explained by increasing the value of the velocity dispersion ([FORMULA]) in the envelope. Thus a deviation from smooth outflow and a possible NIR pumping mechanism must be considered. Alcock & Ross (1985) propose that mass loss occurs in the form of clumps of material ejected randomly from the central star. As a consequence, the emergent maser profile can be the superposition of the maser emission from individual clumps. Random distribution of clumps results in maser profiles with wide peaks. High sensitivity observations of OH masers and modeling by Zell & Fix (1990) seem to give some support to the clumpiness nature of the circumstellar envelope. On the other hand, clumpiness can also provide a natural explanation to the limit of the Doppler velocity between overlapping lines. This condition is necessary to reproduce main line masers when FIR line overlap is considered. If we assume that radiative interaction between clumps is negligible, FIR line overlap occurs essentially between OH molecules inside the clump. Consequently, the Doppler shift between overlapping lines is limited by the clump size and the expansion velocity. In the envelope of optically thin Mira variables where expansion velocity is small, the FIR overlap is a mechanism capable of producing main line masers and also contributes to the inversion of the satellite maser.

NIR overlap between a NIR [FORMULA] line and OH line at 2.8 µm was first suggested by Cimerman & Scoville (1980) to account for the OH main line emission from circumstellar envelopes. A velocity shift of 11 kms-1 between the [FORMULA] -NIR line and the OH-NIR line determined by Cimerman & Scoville (1980) seems unlikely to be realized in many OH maser sources. But recently, Collison & Nedoluha (1994) found another possible overlap between [FORMULA] and OH lines with a velocity shift of only about 6 kms-1. This mechanism may be at work in IRC+10420 due to its high expansion velocity.

Although we have adopted a non-local radiative transfer to treat the OH ground state maser lines, some uncertainties still remain in our model. The pumping conditions in the OH shell are determined using the LVG approximation. The thermal overlap between two close FIR hyperfine lines is not considered in our model. The fact that the Doppler shift between some OH-FIR lines is of only about 0.6 kms-1 together with a local linewidth of 1 kms-1 would result in a thermal overlap. A correct treatment of this effect will require an intricate direct integration of the radiative transfer equations of all OH-FIR hyperfine and microwave lines. For the moment, non-local FIR line overlap is the simplest way which allows us to determine the pumping conditions of ground state masers in the envelope.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 1998

Online publication: February 4, 1998