Forum Springer Astron. Astrophys.
Forum Whats New Search Orders

Astron. Astrophys. 361, 895-900 (2000)

Previous Section Next Section Title Page Table of Contents

4. Discussion

Relation to extinction curve- The SMC is one of the most unusual environments in which to observe aromatic hydrocarbons in abundance because of two well-established facts. First, the metallicity of the SMC is lower than in our Galaxy by a factor of 10, and the C/O ratio is also lower than in our Galaxy, which could affect the abundance of aromatic hydrocarbon molecules, which composed almost entirely of C by mass (Dufour et al. 1982). Second, the measured extinction curve of SMC dust is significantly different from that of Galactic dust. In particular, most SMC extinction curves lack the prominent 2175 Å bump so obvious in the Galactic extinction curve (Savage & Mathis 1979). This bump is widely attributed to graphitic particles. Clusters of aromatic hydrocarbons may have similar spectral properties to graphite. Because the 2175 Å feature is missing from the SMC, while the aromatic hydrocarbon abundance is high in the SMC, our results suggest that aromatic hydrocarbons are not responsible for the 2175 Å bump in the Milky Way. However, at least one SMC star, Sk 143, has an extinction curve with the same shape as that of the Milky Way, which suggests strong regional variations in the SMC extinction curve (Lequeux et al. 1982, Gordon & Clayton 1998). Also, we have only detected the aromatic features from one quiescent cloud, which may or may not be representative of typical SMC material. Therefore, it is possible that we could be comparing the UV-visible extinction and mid-infrared emission for regions with different dust properties. This problem could be eliminated in the future, by observing the extinction curve through the same type of region from which mid-infrared emission in observed. The aromatic hydrocarbons must absorb a substantial fraction (see below) of the interstellar radiation field, but we cannot be sure what photon energy range is exciting them. Observations of reflection nebulae excited by stars with a range of spectral types indicate that aromatic hydrocarbons may not require far-ultraviolet photons for excitation (Uchida, Sellgren, & Werner 1998).

Comparison to Milky Way spectra- The mid-infrared spectrum of the SMC contains features at the same wavelengths as most galactic sources; however, the feature-to-feature ratios are quite distinct for the SMC. The spectrum of the diffuse cloud emission is well fit by a sum of 4 Lorentzians, representing the 6.2, 7.7, 11.3, and 12.6 µm features. A Lorentzian shape is a better fit and more physically justified than a Gaussian fit (Boulanger et al. 1998). Other than the wings of the Lorentzians and a broad pedestal under the 11.3 and 12.6 µm features, no continuum was detected toward the diffuse cloud (within the uncertainties). The relative strengths of the emission features are significantly different from those of Milky Way objects. Compared to the observations of H II regions, reflection nebulae, and diffuse clouds, summarized by Lu (1998), the ratio of (11.3)/(7.7) features is higher in SMCB1#1. Compared to the B-star excited emission from the [FORMULA] Oph molecular clouds (Boulanger et al. 1996, 1998), the ratio of (11.3)/(7.7) features is 3 times higher in SMCB1#1. Ionization of polycyclic aromatic hydrocarbon molecules (PAH) enhances the 6-9 µm features relative to the 11-14 µm features (Joblin et al. 1996 , Allamandola, Hudgins, & Sanford 1999), a signature that is opposite to our observations for the SMC. If the aromatic hydrocarbons in the [FORMULA] Oph region are ionized, then the comparison would suggest that the aromatic hydrocarbons in SMC B1#1 are more neutral . We suggest that another explanation may be more likely. The (11.3)/(7.7) feature ratio is proportional to the fraction of emission arising from C-H bonds as opposed to C-C bonds. The high value of this ratio in the SMC suggests there are relatively more C-H bonds. This seems consistent with the aromatic hydrocarbons forming in a more reducing environment (higher H/C abundance ratio) in the SMC, where the abundance of C in the gas from which the molecules form is a factor of 10 lower than in the Milky Way. If the hydrocarbons in the SMC are indeed more hydrogenated than those in the Milky Way, we predict a relatively bright 3.3 µm line from diffuse SMC gas. Our current understanding the nature of interstellar aromatic hydrocarbons is still in its early stages because of the complex inter-relationship between the excitation, chemistry, collisional destruction, and ionization. By providing a new aromatic hydrocarbon spectrum from a unique environment, the observations reported here should help toward disentangling the various processes that shape the aromatic hydrocarbon features and their carriers.

Comparison to Milky Way nebulae- We can put SMC B1#1 into context by comparing it to what known galactic nebulae would look like at the distance of the SMC. To this end, we used the IRAS ISSA (Wheelock et al. 1994) to make simulated maps of three galactic complexes: (1) the O-star-excited H II region around the Orion Trapezium, (2) the B-star-excited reflection nebula around [FORMULA] Oph, and (3) the star-forming complex in Taurus. The physical sizes of all three nebulae are in the same order of magnitude as SMC B1#1 ([FORMULA] pc). The 12 µm surface brightnesses at 1.7 pc physical resolution ([FORMULA] at SMC distance) of the objects are 350, 12, and [FORMULA] MJy sr-1, respectively; for comparison, the brightness of SMC B1#1 is about 2 MJy sr-1. It is clear that SMC B1#1 is significantly different from the Trapezium, both because it is much less bright in the mid-infrared, and because it has no associated radio or optical H II region. It is also evident that SMC B1#1 is different from the Taurus clouds, because it is brighter in the mid-infrared. The best Galactic analog for SMC B1#1 is [FORMULA] Oph. The smoothed image of [FORMULA] Oph, shown in Fig. 3, is comparable to that of SMC B1#1, with one bright point source and diffuse emission on a 10 pc size scale. The point-source in our ISOCAM image of SMC B1#1 could be a reflection nebula, analogous to the bright mid-infrared region in [FORMULA] Oph imaged by ISOCAM (Abergel et al. 1996), but with a rather different spectrum; or it could be an ultracompact H II region from a highly embedded late-type O star. Estimated abundance of aromatic hydrocarbons- The abundance of aromatic hydrocarbons in the SMC B1#1 cloud can be roughly estimated from the brightness of the infrared features in the diffuse emission from SMC B1#1, where the dust heating is due to the diffuse SMC interstellar radiation field. The total energy absorbed by a completely dark cloud is the integrated energy from the incident interstellar radiation field. The fraction of the radiation field that is absorbed by aromatic hydrocarbons can be calculated from the observed infrared brightness as


where [FORMULA] is the intensity of the observed aromatic hydrocarbon features, g is the fraction of all aromatic hydrocarbon emission that comes out in the observed features, and [FORMULA] is the intensity of the ultraviolet Solar Neighborhood radiation field (Mathis, Mezger, & Panagia 1983). In practice, the value of [FORMULA] obtained from this equation is a lower limit, because the cloud surface may not be completely dark and may not fill the beam uniformly. The radiation field of the observed cloud is [FORMULA], in units of [FORMULA]. Adding together the 1620 Å fluxes of the stars in the H II regions seen with the UIT (Cornett et al. 1997), and including the estimated fluxes of the H II regions DEM 21, 28 and 16 based on their H[FORMULA] brightness, we estimate that the radiation field at the location of SMC B1#1 has [FORMULA], with an uncertainty of about a factor of 2. If the exciting star of the point-like reflection nebula in SMC B1#1 is a B3 or later star, its flux would exceed that of the radiation field only within [FORMULA] pc of the star (or less, if extinction were included), consistent with the unresolved source in our image but of little or no importance for the diffuse emission. The ISOCAM wavelength range includes essentially all of the aromatic hydrocarbon emission, so [FORMULA], although it is possible that we miss some emission in a quasi-continuum under the features or outside of our observed wavelength range. Using the sum of the feature brightness from Table 1, we find [FORMULA].

[FIGURE] Fig. 3. IRAS Image of the [FORMULA] Oph region, as it would appear if it were observed at the distance of the SMC with [FORMULA] resolution. The bright and unrelated star Antares was deleted from the original IRAS map before projecting the map to the SMC distance. Contours are drawn at 2 MJy sr-1 intervals of IRAS 12 µm surface brightness. The brightest pixel in the map has an IRAS 12 µm band surface brightness of 22 MJy sr-1, while the periphery of the main cloud and the adjacent clouds in the image have brightnesses of a few MJy sr-1, comparable to the SMC B1#1 cloud that we observed with ISOCAM.


Table 1. Lorentzian fit to SMC infrared emission features

To estimate directly the energy emitted by large dust grains in the SMC, we use archival ISOPHOT (Lemke et al. 1996) observations. An image of the SMC-B1 region, covering [FORMULA] including the region of our ISOCAM observation, was made at 135 µm wavelength with the C2 detectors. The far-infrared emission is faint at the location of our ISOCAM image, with no evidence of a far-infrared peak corresponding to the mid-infrared peak. There is some far-infrared emission extending north of our target region, suggesting that SMC B1#1 is a clump at the southern edge of a large, cold cloud. The far-infrared surface brightness at the location of our mid-infrared peak SMC B1#1 is [FORMULA] MJy sr-1, measured with respect to nearby dark pixels in the ISOPHOT image. If the far-infrared spectrum is typical of large dust grains in the Milky Way (Boulanger et al. 1988, Dwek et al. 1997), then the integrated far-infrared emission is 2 times brighter than [FORMULA] at 135 µm, so the far-infrared surface brightness of our cloud is [FORMULA] nW m-2 sr-1. Comparing to the sum of the aromatic hydrocarbon feature brightnesses in Table 1, the relative amount of energy absorbed by aromatic hydrocarbon as compared to big grains is [FORMULA]. This is comparable to the fraction of energy absorbed by aromatic hydrocarbons in our Galaxy (Dwek et al. 1997). It is also consistent with our abundance estimate in the previous paragraph.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 2000

Online publication: October 10, 2000