Astron. Astrophys. 336, 823-828 (1998)

5. Dust temperature

Based on the flux measurements, we can deduce a blackbody temperature from the L to L' ratio. These two bands are not the best choice for this, being close in wavelength, however, we recall that we did not detect NGC 7469 in the M band. We have estimated the measurement error on the flux to be 10%, which leads to an error on the temperature of 200 K.

Thanks to the high angular resolution achieved in the L and L' band images, we were also able to derive the dust temperature at increasing distances from the central source of the AGN. These results are presented in Fig. 3. The temperature decreases (from 900 K to 300 K) when the dust emission originates further away from the central source of the AGN (from r130 pc to r400 pc), suggesting that the central source is indeed responsible for the dust heating: heating by local sources (like starbursts) would lead to a more clumpy distribution in the dust temperature map.

Fig. 3. Dust temperature as a function of distance from the central source

Because the resolution is limited to 110 pc, we could not derive the dust temperature close to the central source (r110 pc), and the highest value of 900 K is therefore a mean value over a 130 pc radius region. This means that nearer to the central source, the temperature is probably higher. How high could the temperature be ? This answer depends on the dust composition. The sublimation temperature of dust grains is above 1500 K for graphite, 1200 K for silicates and 1000 K for PAH. Because they can survive in quite strong UV radiation field, graphite and/or silicates are more probable. By extrapolating the temperature curve, we find that the dust temperature at 15 pc from the central source could be as high than 1250 K, excluding the presence of PAH very close to the central engine. The PAH emission measured by Mazzarella et al. (1994) within a 2" diaphragm must be therefore in a ring like configuration.

A temperature above 1250 K for the inner dust is expected in torus models to explain the observed spectral energy distribution (Granato & Danese, 1994). In particular, the nuclear optical and IR fluxes show clear evidence for a 1µm minimum (Sanders et al., 1989), which can be explained by dust heated to its sublimation temperature 1500 K (Granato & Danese, 1994). With regard to dust temperature, our observations favor this interpretation.

The K band flux from Genzel et al. (1995) is given for a 1.4" diameter aperture. We have computed the corresponding flux in the L and L' bands for a 1.4" diameter aperture. Fitting the L and L' fluxes with a blackbody emission, we find a mean temperature of 900 K for this region. We then derive the K band flux corresponding to this blackbody emission. We find that the K band flux contributed by the dust component in this region must be mJy. Genzel et al. (1995) give a value of mJy for the K band flux, showing therefore that the hot dust component accounts for 363% of the near-infrared flux in a r225 pc region around the central source of the AGN. Based on the visible/infrared spectral energy distribution, Genzel et al. (1995) calculate that about one third to half of the K-band flux might be stellar. This suggests that the rest of the near-infrared flux (from 14% to 34%) is related to a non-thermal emission from the central engine.

© European Southern Observatory (ESO) 1998

Online publication: July 27, 1998
helpdesk.link@springer.de