Astron. Astrophys. 363, 917-925 (2000)

8. Dust masses

For each knot we determined the dust mass , adopting the model parameters of Chini et al. (Chini et al. 1986):

where D is the distance, the flux density, the Planck function, and the mass absorption coefficient of the dust at a reference wavelength. The fit of the modified Planck functions to the data yields equations

where and are the scaling factors of the cold and warm component for each knot. Assuming the same value of for the two components, this yields

We adopted = 0.03 m²/kg for a reference wavelength of 1.3 mm, which has already been successfully used for star forming regions and cold cloud fragments (Chini et al. 1987; Krügel & Chini 1994). This value is well in between those derived from other investigations and computed to 1.3 mm, i.e. from Sodroski et al. (1994) (for DIRBE/COBE FIR data) and from Fich & Hodge (1991) (for IRAS and mm data). For the distance of M 31 we took . The resulting mass values are given in Column (11) of Table 1.

Note, that these values are only lower limits of the dust masses, as dust with temperatures below about 12 K cannot be detected at 180 but only at longer wavelengths. The presence of such cold dust has already been found in Galactic molecular cloud complexes (Krügel & Chini 1994; Ristorcelli et al. 1998) as well as in star forming regions (Ristorcelli et al. 1999), and might therefore also be common in M 31's FIR sources. However, for the Chameleon region Toth et al. (2000) have shown that the number of such cold protostellar cores is very low, and their mass does not contribute significantly to the total mass of the molecular cloud complex. Hence, we assume that in case there is any dust below 12 K in M 31 the increase of the derived masses will not be very high nor critically influence our discussions.

The integrated mass of all knots (each seen within 5´ aperture) is , which is only about 6% of the total dust mass of M 31 (, when recalculated with the same model parameters; Haas et al. 1998computed the dust mass slightly differently according to Hildebrandt 1983). It will be interesting to compare the dust masses of M 31 with the masses of the involved gas. For the southwest part of the galaxy, Pagani et al. (1999) have found an excellent correlation between the column density of neutral gas and the mid IR intensity. Apparently, the gas of the 10 kpc ring is dominated by H I whereas little H I is found inside, in the inter-ring region. However, the conversion factor between CO and H₂ is not very well known which might change the total column density of neutral gas. Thus, the comparison of M 31's dust and gas masses derived from H I and CO surveys will be performed in a future detailed work.

© European Southern Observatory (ESO) 2000

Online publication: December 5, 2000
helpdesk@link.springer.de