An IRAS two-color diagram can be used to roughly identify stars in different
stages of their evolution (cf. Olnon et al. 1984; Bedijn
1987; van der Veen & Habing 1988). Due to the
different properties of their dust shells, OH/IR stars, PPNe, and PNe
occupy slightly different regions in the diagram (although there is also an
overlap), and the diagram has been found to be especially useful in the
selection of PPNe (e.g., Hu et al. 1993a). Figure 3 (click here)
shows a plot of [F60/F25] versus [F25/F12] for the sources
in this sample and also from previous observations by Gomez et al.
(1990), Nyman et al. (1993a), and the OH/IR objects
detected by Jewell et al. (1991). The colors are defined from
the IRAS fluxes in the following way: .
Figure 3: IRAS two-color diagram of detected and
non-detected sources in our sample and from previous observations by
Gomez et al. (1990), Nyman et al.
(1993a), and Jewell et al. (1991)
As was also found by Gomez et al. (1990) very few sources are detected beyond [F25/F12]=1.2. Only two sources with redder colors have been found to have SiO masers: OH 231.8+4.2 and OH 15.7+0.8 which are discussed in the previous section. Bedijn (1987) suggests that sources with [F25/F12]>1.2 have ceased their mass loss and are surrounded by a detached envelope.
To study in more detail how the detection rate of SiO masers changes with IRAS colors, we have taken the data from our observations as well as from the papers listed above and calculated the detection rates as function of [F25/F12]. The data were separated into bins of 0.4 units in IRAS color, except for the last bin which is three times larger to contain more objects. The results are shown in Table 3 (click here). The statistical error in the detection rate is given within parenthesis in Table 3 (click here), and was calculated assuming that the detected sources have a Poisson distribution. As found by Gomez et al. (1990) there is a gradual decrease in detection rate with increasing IRAS color. The gradual decrease is partly due to a small decrease in the detection rate of the OH/IR stars, but mainly due to an increase in the number of PPNe and unclassified sources with increasing [F25/F12].
Our sample and data from | Our sample, | Our sample, | |||||||||
literature, all objects | all objects | OH/IR objects only | |||||||||
[F25/F12] | No. of | No. of | Det rate | No. of | No. of | Det rate | No. of | No. of | Det rate | ||
obj. | det. | (%) | obj. | det. | (%) | obj. | det. | (%) | |||
-0.4 - 0.0 | 22 | 15 | 68 (18) | 6 | 4 | 67 (33) | 3 | 3 | 100 (56) | ||
![]() | 46 | 21 | 45 (10) | 21 | 12 | 55 (16) | 18 | 11 | 61 (18) | ||
![]() | 43 | 18 | 42 (10) | 21 | 9 | 43 (14) | 10 | 8 | 80 (28) | ||
![]() | 46 | 11 | 24 (17) | 27 | 6 | 22 (9) | 12 | 5 | 42 (19) | ||
![]() | 13 | 1 | 8 (8) | 9 | 0 | 0 | 0 | ||||
![]() | 19 | 0 | 0 | 16 | 0 | 0 | 0 | ||||
![]() | 7 | 0 | 0 | 3 | 0 | 0 | 0 | ||||
![]() | 8 | 1 | 12 (12) | 5 | 1 | 20 (20) | 0 | ||||
![]() | 6 | 0 | 0 | 4 | 0 | 0 | 0 |