Forum Springer Astron. Astrophys.
Forum Whats New Search Orders

Astron. Astrophys. 362, 310-324 (2000)

Previous Section Next Section Title Page Table of Contents

3. The fine-structure line

The two fine-structure line [Ne III ] 15.6 µm and [S IV ] 10.5 µm are prominent in the CVF spectra (see Fig. 6). The [Ne II ] line at 12.8 µm is blended with the infared band (IR) at 12.7 µm, and this blend is quite faint everywhere with respect to the [Ne III ] line. However, the distribution of the 12.7 IB + [NeII] emission line (not shown here) and that of the [Ne III ] are very different, the former being very similar to the distribution of the 11.3 and 6.2 µm IBs. This suggests that the emission feature at 12.7-12.8 µm is dominated by the IB and that the [NeII] fine structure line is everywhere negligible. A strict lower limit for the [Ne III ]/[Ne II ] intensity ratio is 1.0. We can compare this lower limit to the [Ne III ] 15.5µm/[Ne II ] 12.8 µm intensity ratios found in other sources with ISO-SWS. Contrary to the ISOCAM instrument, SWS can separate the [Ne II ] 12.8 µm emission line from the 12.7µm IB. The values found are all less than in N66, ranging from [FORMULA]0.8 in the overlapping region of the Antennae galaxy to [FORMULA]0.2 in the Galactic Center (Moorwood et al. 1996, Lutz et al. 1996, Kunze et al. 1996). On the other hand, higher [Ne III ]/[Ne II ] values have been observed with ISOCAM in a sample of dwarf galaxies from Madden (2000). The exceptionally large strength of the [S IV ] line and the very high [Ne III ]/[Ne II ] ratio confirm that N 66 is a region of particularly high excitation, due to the large number of extremely hot O stars that it contains (Massey et al. 1989). This was previously known from the high electron temperatures (12000-14000 K) and the high ratio [O III ][FORMULA]5007/H[FORMULA] [FORMULA] 5 measured at different points of the nebula (Peimbert & Torres-Peimbert 1976, Dufour & Harlow 1977, Pagel et al. 1978, Dufour et al. 1982). Fig. 3 is a map in the [S IV ] 10.5 µm line superimposed on the [Ne III ] 15.6 µm map.

[FIGURE] Fig. 3. CVF map of N  66 in the [S IV ] 10.5 µm line (contours) superimposed on the [Ne III ] 15.6 µm line map. Coordinates are J2000. Notice the holes in the [Ne III ] 15.6 µm map. There are large differences between the distributions of the two line. The position of star W 3, classified O3 III(f*), is indicated by a white cross. The strong [S IV ] emission to the East is presumably due to the supernova remnant SNR 0057-7226, whose center is indicated by a black cross. The solid line represents the cut along which we evaluated the line intensity profiles of Fig. 4.

By comparison with the H[FORMULA] map of Fig. 1 and the radio continuum map of Taisheng Ye et al. (1991), it appears that the fine structure line emission is associated with N 66, extending over our field of view. We estimate from the radio data of Taisheng Ye et al. (1991) that the amount of energy in the H[FORMULA] line is [FORMULA] 5 10-13 W m-2 in this field. This of course is intrinsically corrected for interstellar extinction. Extinction is low (E(B-V)=0.14 according to Massey et al. (1989)), and we neglect it when considering the mid-IR observations.

An interesting feature of the [Ne III ] map is the presence of several holes in the distribution of the ionized gas (Fig. 3). These holes are presumably due to previous supernova explosions or to the effects of stellar winds. The latter explanation is probably true for the central and more pronounced hole, which is near the hottest star in the N 66 OB association (an OIII(f*) star marked on Fig. 3). The differences in the distribution of the [Ne III ] and [S IV ] line are noteworthy. The emission in the [Ne III ] line follows roughly that of H[FORMULA] as far as one can judge given the different angular resolutions (compare Fig. 1 and Fig. 3). This is not the case for the [S IV ] line. Most of the differences in the main emitting region are probably density effects: model calculations e.g. by Stasinska (1984) show that the [Ne III ]/[S IV ] line intensity ratio is decreased in regions of lower densities, the other parameters being the same. This might explain why the holes are more visible in the [Ne III ] than in the [S IV ] line. There is relatively less [Ne III ] emission in the eastern part of the field where the emission of [S IV ] is substantial; here the [S IV ] line intensity is roughly twice that of [Ne III ]. This region contains a faint H[FORMULA] filament (not visible on Fig. 1) which is a part of the supernova remnant SNR 0057-7226 (Taisheng Ye et al. 1991). The position of the center of this remnant is indicated on Fig. 3. The difference in the distributions of the [Ne III ] and of the [S IV ] line is best illustrated by Fig. 4 which shows the fine structure line-intensity profiles along the direction marked on Fig. 3. It appears that the emission of [S IV ] is enhanced compared to that of [Ne III ] by preferential shock ionization of S with respect to Ne: in high-excitation conditions Ne III  is the dominant Ne ion and its abundance can only be decreased by collisional ionization, while S III  and S IV  have roughly the same abundances and S III  will be ionized in the shock.

[FIGURE] Fig. 4. The logarithmic profile in the [Ne III ] 15.6 µm line (dashed line) and the [S IV ] 10.5 µm line (solid line) along the cut shown in Fig. 3. Coordinates are J2000. Note that the [S IV ] line intensity is always higher than that of [Ne III ], the [S IV ]/[Ne III ] line intensity ratio reaching a maximum at a position closest to SNR.

By integrating over the whole map, we obtain total fluxes of [FORMULA] 7.3 10-14 W m-2 and [FORMULA] 8.8 10-14 W m-2 in the [Ne III ] and [S IV ] line respectively. The [Ne III ]/H[FORMULA] and the [S IV ]/[Ne III ] ratios are [FORMULA] 0.1 and [FORMULA]1.2 respectively and together with the optical line ratios obtained by various authors cited above, they can be compared with the results of photoionization models. We used the models of Stasinska (1982, 1984, 1990), of Stasinska & Leitherer (1996) and of Schaerer & de Koter (1997). A fair agreement can be reached for the relative intensities of the [Ne II ], the [Ne III ] and the [S IV ] line as well as of the [O III ][FORMULA]5007, 4959 Å and 4363 Å and other visible line, using models with Teff of exciting stars 40 000 to 45 000K, abundances 1/10 solar, density 10 to 100 electrons cm-3. These parameters are reasonable from what we know otherwise of the H II  region with its very hot exciting stars. Direct abundance effects cannot be invoked to explain the particularly high [S IV ]/[Ne III ] ratios, the abundance ratios [Ne/O] (Pagel et al. 1978) and [S/O] (Dennefeld & Stasinska 1983) being approximately the same in the H II  regions of the SMC and of the Galaxy. However, most of these models yield ratios of the mid-IR line of [Ne III ] and [S IV ] to H[FORMULA] too high by a factor 2 with respect to the observed ratios, with the exception of the old models of Stasinska (1982), which are based on the model atmospheres of Mihalas. But one should note that the optical measurements refer to the central H II  region while the [Ne III ] and [S IV ] line intensity ratios to H[FORMULA] are global. Clearly more detailed optical studies are required in order to reach more definitive conclusions. Given the lack of data for the supernova remnant, it is premature to try to model the intensities of the [Ne III ] and [S IV ] line in its direction.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 2000

Online publication: October 30, 19100