The diffuse interstellar bands (DIBs) are absorption features in the interstellar medium (ISM) with rest-frame optical wavelengths. They are commonly observed toward background reddened stars in the Milky Way galaxy. Since their discovery by Heger (1922), there have been over 300 DIBs discovered and studied (Galazutdinov et al. 2000; Jenniskens & Desert 1994; Tuairisg et al. 2000; Weselak et al. 2000; Hobbs et al. 2008). Understanding the source carriers of the DIBs has immense importance in understanding the nature of a galaxy's ISM as well as possible significance for astrobiology. To date, no DIB has been positively identified with a carrier atom, molecule, or dust grain. However, there are candidate carriers that are actively being researched. See the Astrobiology Motivation section for a discussion on possible DIB carriers. The largest DIB bands are observed throughout the optical, as seen in Figure 1 below.
The observed Galactic spectral characteristics of the six DIBs labeled in Figure 1 are tabulated in Table 1. The data are compiled from the DIB catalogue of Jenniskens & Desert (1994), who used observations toward four reddened stars in the Galaxy. Listed, in columns from left to right, are the rest wavelength in air, the rest vacuum wavelength, the rest full width half-maximum (FWHM), and the rest average equivalent width. The equivalent width is an average of the equivalent widths (EWs) of the DIB toward each star, after being normalized to the λ5780 DIB equivalent width and the reddening, E(B-V), along the line of sight (Jenniskens & Desert 1994). The λ4428 DIB is very broad which lends itself to greater uncertainties in finding the profile centroid, FWHM, and equivalent width.
The λ5780, λ5797, and λ6284 DIBs have been widely observed in the Galaxy, and several correlations of those DIBs have been published. The DIB-Galactic correlations I employ in my work are those of the DIB EW strengths with neutral hydrogen column density, N(HI); reddening, E(B-V); and neutral sodium column density, N(NaI) (e.g. Herbig 1993, 1995; Cox et al. 2006b; Welty et al. 2006; Ellison et al. 2008).
Adapted from Welty et al. (2006) and plotted in Figure 2 is the log EW versus the log HI column density, N(HI), of the λ5780, λ5797, and λ6284 DIBs. The circles are from Large Magellanic Cloud (LMC) sightlines, the triangles are from Small Magellanic Cloud (SMC) sightlines, and the remaining points are Galactic sightlines (Welty et al. 2006). The DIB strengths from the LMC and SMC sightlines are lower than expected from the Galactic best-fit lines. These correlations with HI gas, and the fact that DIB strengths are not correlated with molecular hydrogen (Herbig 1993), are why DIBs are typically associated with the diffuse interstellar medium and not molecular clouds (see Herbig 1993, 1995).
The λ5780, λ5797, and λ6284 DIBs are also widely found to be correlated with reddening. Shown in Figure 3, adapted from Welty et al. (2006), is the log EW versus the log E(B-V) of the λ5780, λ5797, and λ6284 DIBs. The circles are the LMC sightlines, the triangles are the SMC sightlines, and the remaining points are Galactic sightlines. The extragalactic sightlines appear to more closely follow the Galactic relations relative to the DIB-N(HI) relations shown in Figure 2 Ellison et al. (2008) have shown that extragalactic sightlines do appear to follow the Galactic DIB-E(B-V) relation quite well for the λ5780 DIB.
A relatively weak Galactic DIB-N(NaI) correlation is also recognized for the λ5780, λ5797, and λ6284 DIBs. Adapted from Welty et al. (2006) and shown in Figure 4 are the log EW versus log N(NaI) relations for the λ5797 and λ6284 DIBs. The circles are LMC sightlines, triangles are SMC sightlines, and the remaining points are Galactic sightlines. The extragalactic DIBs are underabundant relative to the Galactic best-fit lines, for the same column densities of NaI, much like they are with N(HI) (see Figure 2). Although, the scatter about the Galactic DIB-N(NaI) relation is larger than for the Galactic DIB-N(HI) relation, rms~0.10 and rms~0.22, respectively (Welty et al. 2006).
Plotted in Figure 5, as adapted from Ellison et al. (2008), is the log N(NaI) versus the log EW of the λ5780 DIB. The diamonds and triangles are LMC and SMC sightlines (Welty et al. 2006; Vladilo et al. 1987; Vidal-Madjar et al. 1987); squares are Galactic sightlines (Herbig 1993); crosses are other extragalactic sightlines (Sollerman et al. 2005; D'Odorico et al. 1989; Heckman & Lehnert 2000), and the best-fit line is to the Galactic sightlines (Ellison et al. 2008). Similarly with Figure 4, the DIB strengths appear to be systematically weaker in LMC, SMC, and other extragalactic sightlines relative to Galactic sightlines. The extragalactic sightlines include starburst galaxies (Heckman & Lehnert 2000), galaxy Centaurus A (D'Odorico et al. 1989), and galaxy NGC 1448 (Sollerman et al. 2005).
Many researchers have found that at least some of the DIBs are possibly due to ionized molecules (see Snow 2001a; Ehrenfreund & Charnley 2000; Herbig 1995). It is not immediately clear that the DIBs should correlate with neutral atomics like NaI, which has a relatively low first ionization level (~5.1 eV). Herbig (1995) makes the argument that so long as there is a relative constancy in the degree of first ionization, the DIBs will show an equally good correlation with ionized species.
DIBs have been extensively studied in the Galactic environment for nearly a century (see reviews by Herbig 1995; Snow 2001a). To date there have been few extragalactic studies of DIBs. A recent spate of extragalactic DIB studies has been undertaken in the past decade and is growing as a subfield of DIB research. One such group, including T. P. Snow, C. W. Churchill, and myself, was created after the realization that extragalactic environments have varying gas abundances, dust contents, and metallicities that can provide unique insights to the molecules possibly responsible for the DIBs (see Snow 2001b). I explore these avenues by examining DIB strengths and limits in galaxies from low to moderate redshifts. Prior to my work (up through 2008), DIBs are known outside of the Galaxy in the following 16 systems:
Due to their relatively weak strengths, targeted DIB surveys outside of the Galaxy require high signal-to-noise. Thus, extragalactic DIB surveys are currently rare. Furthermore, researchers executing intensive observations or surveys outside of the Galaxy rarely are searching their data for DIBs. There is tremendous room for studies of extragalactic DIBs, and there are many existing spectra of extragalactic sources that can be explored. The two extragalactic samples in my work (DLAs and starbursts), were obtained through dedicated observations and via collaborations with researchers who have existing data. To maximize the opportunity for DIB detections and to study known Galactic DIB correlations, I have selected samples of HI abundant galaxies, known as damped Lyman α systems (DLAs), and NaI abundant starburst galaxies (see Figure 2 and Figure 4). I compare the DIB equivalent width strengths and limits with HI column densities, reddenings, and gas-to-dust ratios in the DLAs. In the starbursts, I compare the DIB equivalent width strengths, limits, and velocities with NaI column densities, H column densities, and NaI velocities.