First Detection of SiO J=7-6 Maser Emission

During 1995 June, we devoted several shifts to observing the 321- and 325-GHz water masers in the well- known symbiotic binary, R Aquarii, with the VLA simultaneously monitoring the 22-GHz transition. At these frequencies, even in the best of weather conditions it can feel like you are using JCMT as the world's most expensive water-vapour radiometer, and so during one of the many periods of unsuitable weather, we turned to our umpteenth backup programme: a continuation of our search for J = 7-6, v = 1 and 2 SiO maser emission.

R Aqr is unusual for a symbiotic Mira, in that it exhibits maser emission. In this respect, it resembles isolated Miras: it has been detected as a source of SiO maser emission in the J = 1-0 and J = 2-1 lines of both the v = 1 and 2 states; indeed, the maser is often used as a pointing source by, for example, SEST. It is also one of only two symbiotic systems known to support water masers (Ivison, Seaquist & Hall 1994) hence our primary programme on this occasion. The resemblance of the SiO emission in R Aqr to that in isolated Miras is probably due to the large binary separation and the low luminosity of its hot companion star compared with the other symbiotics observed by Seaquist, Ivison & Hall (1995). In terms of our continuing search for J = 7-6 SiO maser emission, R Aqr seemed as good a target as any, especially since we were already pointing at it.

Figure 1: R Aqr SiO maser transitions: (top) v = 1, J = 7-6 (301.814 GHz) - the highest frequency SiO maser yet detected; four-channel averages are shown; (bottom) v = 2, J = 7-6 (299.704 GHz) - the highest temperature (E/k) SiO maser transition yet detected; six-channel averages are shown.

The observations were carried out during 1995 June 9.8 UT. The rest frequencies used were 301.814 GHz for v = 1 and 299.704 GHz for v = 2, with a velocity of -25.0 km/s relative to the LSR. The channel spacing was 0.156 MHz (or 0.156 km/s). For each transition, we integrated for 30 min, though the lines were apparent after only a few cycles. The resulting spectra were of excellent quality (Figure 1) and clearly show SiO maser emission from the highest lying rotational transitions detected to date.

The theoretical models used to predict the existence of these masers indicate that they arise from a predominantly collisional pumping mechanism in very dense regions of the circumstellar envelope. Using these models, in conjunction with a pulsation model of a 1-Mo Mira (Bowen 1988), we find a maximum amplification factor of 170.

Assuming that the maser amplifies a black-body background at a typical inner envelope temperature of 1500 K, the amplification factor is around 170, and that the spot size of a 300-GHz maser is about 0.5 mas, we find that the maser beam-angle must be 4.8x10-4 sr to match the observed flux - well within the range appropriate for geometrical beaming arising from alignment of small clumps of amplifying gas along the line of sight. Our detection therefore adds credence to the most recent (and optimistic) theoretical work.

It is interesting that the new maser transitions were detected in R Aqr, whilst those in R Leo and VY CMa went undetected. The most likely reason is a strong stellar phase effect on the intensities of the higher rotational state masers. Support for this idea comes from calculations (presently in progress) using a phase- dependent Mira model, and from our non-detection of the J = 6-5 transition in R Leo (the transition was seen at an earlier epoch by Jewell et al. 1987).

It could also be argued that the detection of J = 7-6 SiO masers towards R Aqr has more to do with the influence of the Mira's hot companion than with its pulsational phase, i.e. that the J = 7-6 maser may be pumped in some way by R Aqr's white-dwarf companion. Maybe J = 7-6 SiO masers can only exist in symbiotic Mira systems, where the companion causes temperature and density perturbations in the CS envelope? Comparative observations of a control sample of isolated Miras and a small sample of symbiotic Miras (say, H1-36 and R Aqr, those with known low-J SiO masers) should provide the answer.

It may be that the most interesting consequence of our detection is that it confers the ability to test the 'clump model' for maser emission. Clumps supposedly form through thermal instabilities resulting from infrared band cooling by CO and SiO: a bifurcated envelope structure is the result, with phases which differ in kinetic temperature by typically 200-300 K and in molecular abundance. One phase forms the clumps and the other a hotter background medium.

Models predict that the number of clumps contributing to a SiO maser falls with increasing rotational excitation, and so we would expect clumps emitting in J = 7-6 lines to be much rarer than those emitting in, say, the J = 2-1 lines. Our prediction for the J = 7-6 lines would therefore be for an almost catastrophic form of variability, since the loss of 2 or 3 dominant clumps could eradicate the entire spectrum in these transitions for some fraction of the stellar cycle.

Simultaneous monitoring of J = 7-6 and J = 2-1 SiO transitions would allow comparison of maser-feature lifetimes against those predicted by the model. New spectral features should appear on the cooling time (< 106 s) and should last until a clump is re-heated by the passage of a shock-wave (i.e. every 2x106 to 107 s); thus, a study of the changes in SiO maser features from R Aqr over a pulsational period (386 d) will act as a severe test of the clump model.

References

Allen D.A., Hall P.J., Norris R.P., Troup E.R., Wark R.M., Wright A.E., 1989, MNRAS, 236, 363.

Bowen G.H., 1988, ApJ, 329, 299.

Doel R.C., Gray M.D., Humphreys E.M.L., Braithwaite M.F., Field D., 1995, A&A, in press.

Gray M.D., Humphreys E.M.L., Field D., 1995, ApSS, 224, 63.

Ivison R.J., Seaquist E.R., Hall P.J., 1994, MNRAS, 269, 218.

Jewell P.R., Dickinson D.F., Snyder L.E., Clemens D.P., 1987, ApJ, 323, 749.

Seaquist E.R., Ivison R.J., Hall P.J., 1995, MNRAS, 276, 867.

R. J. Ivison, ROE,

J. A. Yates, M. D. Gray, E. M. L. Humphreys & D. Field, University of Bristol &

P. J. Hall, Australia National Telescope Facility