The Carriers of the Interstellar Unidentified Infrared Emission Features: Aromatic or Aliphatic?
aa r X i v : . [ a s t r o - ph . GA ] O c t The Astrophysical Journal Letters , in press
The Carriers of the Interstellar Unidentified Infrared Emission Features:Aromatic or Aliphatic?
Aigen Li and B.T. Draine ABSTRACT
The unidentified infrared emission (UIE) features at 3.3, 6.2, 7.7, 8.6, and 11.3 µ m,commonly attributed to polycyclic aromatic hydrocarbon (PAH) molecules, have beenrecently ascribed to coal- or kerogen-like organic nanoparticles with a mixed aromatic-aliphatic structure. However, we show in this Letter that this hypothesis is inconsistentwith observations. We estimate the aliphatic fraction of the UIE carriers based on theobserved intensities of the 3.4 µ m and 6.85 µ m emission features by attributing themexclusively to aliphatic C–H stretch and aliphatic C–H deformation vibrational modes,respectively. We derive the fraction of carbon atoms in aliphatic form to be < Subject headings: dust, extinction — ISM: lines and bands — ISM: molecules
1. Introduction
The “unidentified infrared emission” (UIE) bands, a distinct set of spectral features at wave-lengths of 3.3, 6.2, 7.7, 8.6, 11.3 and 12.7 µ m, dominate the mid-infrared spectra of many brightastronomical objects. They are ubiquitously seen in the interstellar medium (ISM) of our owngalaxy and star-forming galaxies, both near and far, and account for over 10% of their total in-frared (IR) luminosity (see Joblin & Tielens 2011). Although the exact nature of the carriersremains unknown, the UIE bands are commonly attributed to polycyclic aromatic hydrocarbon(PAH) molecules (L´eger & Puget 1984, Allamandola et al. 1985). The identification of the UIEbands is important as they are a useful probe of the cosmic star-formation history, and their carriersare an essential player in galactic evolution.Very recently, Kwok & Zhang (2011; hereafter KZ11) argue that the UIE bands arise fromcoal- or kerogen-like organic nanoparticles, consisting of chain-like aliphatic hydrocarbon material Department of Physics and Astronomy, University of Missouri, Columbia, MO 65211, USA; [email protected] Princeton University Observatory, Princeton University, Princeton, NJ 08544, USA; [email protected] § §
3, astronomical observations show that if aliphatic hydrocarbon units are presentin the UIE carriers, they must be a minor constituent. Further, their arguments against the PAHmodel do not seem to pose a problem (see §
2. Constraints on the Aliphatic Fraction from the 3.4 µ m Feature
KZ11 argue that the material responsible for the UIE features has a substantial aliphaticcomponent, based on the mid-IR spectra of NGC 7027 (a planetary nebula), IRAS 22272+5435 (aprotoplanetary nebula), and the Orion bar (a photodissociation region in the Orion nebula). Theydecompose the 3–20 µ m spectra of these objects into three components: the UIE bands, broadplateaus (several µ m in width) peaking at 8 and 12 µ m, and a thermal continuum. They attributethe broad plateau features (which account for ∼ µ m power of these objects) toaliphatic branches of the UIE carriers, similar to the coal model for the UIE bands (Guillois et al.1996). Recognizing the challenge of the coal model in being heated to emit the UIE bands (Pugetet al. 1995), KZ11 hypothesize that the coal-like UIE carriers are nanometer in size or they areheated by the chemical energy released from the H + H → H reaction (Duley & Williams 2011).Aliphatic hydrocarbon has a band at 3.4 µ m due to the C–H stretching mode (Pendleton& Allamandola 2002). In some HII regions, reflection nebulae and planetary nebulae (as well asextragalactic regions, e.g., see Yamagishi et al. 2012, Kondo et al. 2012), the UIE near 3 µ m exhibitsa rich spectrum: the dominant 3.3 µ m feature is usually accompanied by a weaker feature at 3.4 µ malong with an underlying plateau extending out to ∼ µ m. In some objects, a series of weakerfeatures at 3.46, 3.51, and 3.56 µ m are also seen superimposed on the plateau, showing a tendencyto decrease in strength with increasing wavelength (see Figure 1 and Geballe et al. 1985, Jourdainde Muizon et al. 1986, Joblin et al. 1996). While assignment of the 3.3 µ m emission feature to thearomatic C–H stretch is widely accepted, the precise identification of the 3.4 µ m feature (and theaccompanying weak features at 3.46, 3.51, and 3.56 µ m and the broad plateau) remains somewhatcontroversial. By assigning the 3.4 µ m emission exclusively to aliphatic C–H, one can place anupper limit on the aliphatic fraction of the emitters of the UIE features.Let I . and I . respectively be the observed intensities of the 3.4 µ m and 3.3 µ m emission 3 –features. In interstellar and circumstellar environments, I . /I . typically ranges from ∼ ∼ A . and A . respectivelybe the band strengths of the aliphatic and aromatic C–H bonds. We take A . = 2 . × − cmper aliphatic C–H bond, averaged over ethane, hexane, ethyl-benzene, and methyl-cyclo-hexane(d’Hendecourt & Allamandola 1986, Mu˜noz-Caro et al. 2001). We take A . = 4 . × − cm peraromatic C–H bond for small neutral PAHs (Draine & Li 2007).Let N H , aliph and N H , arom respectively be the numbers of aliphatic and aromatic C–H bonds inthe emitters of the 3.3 µ m UIE feature. We obtain N H , aliph /N H , arom ≈ ( I . /I . ) × ( A . /A . ) ≈ .
30, taking I . /I . = 0.2 [KZ11 estimate I . /I . ≈ .
22 for NGC 7027, and I . /I . ≈ .
19 forthe Orion bar]. We assume that one aliphatic C atom corresponds to 2.5 aliphatic C–H bonds(intermediate between methylene -CH and methyl -CH ) and one aromatic C atom corresponds to0.75 aromatic C–H bond (intermediate between benzene C H and coronene C H ). Therefore,in the UIE carriers the ratio of the number of C atoms in aliphatic units to that in aromaticrings is N C , aliph /N C , arom ≈ . × (0 . / .
5) = 0 .
09, showing that the aliphatic component isonly a minor part of the UIE emitters. KZ11 take I . /I . ≈ .
88 for the protoplanetary nebulaIRAS 22272+5435 (but much smaller I . /I . ratios have also been reported for this source; seeGoto et al. 2003). So far only a few sources (exclusively protoplanetary nebulae) are reported tohave I . /I . & ∼ µ m and ∼ µ m, while typicalUIE spectra have distinctive peaks at 7.7, 8.6, and 11.3 µ m (see Tokunaga 1997). We note that N C , aliph /N C , arom = 0.09 is an upper bound as the 3.4 µ m emission feature couldalso be due to anharmonicity of the aromatic C–H stretching mode (Barker et al. 1987). Let ν be the vibrational quantum number. In a harmonic oscillator, the level spacing is constant;the ∆ ν = 1 transition between high ν levels results in the same spectral line as for the ν = 1 → ν levels, and the ∆ ν = 1transitions between higher ν levels occur at longer wavelengths. The anharmonicity model explainsthe weaker features (at 3.40 µ m, 3.51 µ m, ...) as “hot bands” ( ν = 2 → ν = 3 →
2, ...) of the 3.3 µ mfundamental ν = 1 → µ m emission feature could also bedue in part to “superhydrogenated” PAHs in which some peripheral C atoms have two H atoms Typical type II Kerogens have A . ≈ . × − cm per C atom while the 3.3 µ m aromatic feature is barelyvisible (see Figure 2 in Papoular 2001). This clearly shows that kerogen – at least this type – is not a good explanationfor the 3.3 µ m and 3.4 µ m emission features. In coals, the 3.3 µ m aromatic feature is usually weaker than the 3.4 µ maliphatic feature except for those with high ranks (i.e., more evolved, more ordered, with lower H/C and O/C ratios).As coal evolves, the progressive release of heteroatoms (decreasing H/C and O/C) leads to formation of planar clustersof benzene-type rings followed by stacking of these aromatic sheets to form disordered stacks of graphitic planes (seePapoular 2001). As a result of the progressive aromatization, A . /A . decreases as coal evolves. The aliphatic C–Hdeformation band at 6.85 µ m band disappears in highly evolved coal (e.g., anthracite, see Papoular 2001). Overall, the IR spectra of coals or kerogens resemble that of atypical sources (e.g., some protoplanetary nebulae;see Guillois et al. 1996). They do not resemble the UIE features seen in the interstellar sources except for highlyevolved (i.e., highly aromatized) coals (see Papoular 2001). µ m and 3.5 µ m, with the formermore intense than the latter, consistent with astronomical observations (Bernstein et al. 1996). The3.4 µ m feature may also result from aliphatic sidegroups attached as functional groups to PAHs(see Figure 2; Duley & Williams 1981, Pauzat et al. 1999, Wagner et al. 2000). The C–H stretchingmodes of methyl (-CH ), methylene (-CH -), and ethyl (-CH CH ) sidegroups on PAHs fall near theweaker satellite features associated with the 3.3 µ m band. All these possibilities (i.e., anharmonicity,superhydrogenation, and aliphatic sidegroups) probably contribute to the 3.4 µ m emission, theextent of each depending on conditions in the local environment. Sandford (1991) argued that thesatellite features at 3.40, 3.46, 3.51, and 3.56 µ m in NGC 7027 cannot be predominantly due toaliphatic sidegroups on PAHs.KZ11 note that the 3.4 µ m aliphatic C–H stretching mode is commonly observed in absorption in the diffuse ISM. If the UIE carriers have the same mixed aromatic-aliphatic structure as thebulk of the hydrocarbon material, then in heavily obscured regions, both the 3.3 µ m band and the3.4 µ m band would show up in absorption , with the 3.4 µ m absorption band much weaker than the3.3 µ m absorption band. However, astronomical observations have actually shown the opposite (seeFigure 1): the 3.4 µ m absorption band is much stronger than the 3.3 µ m absorption band (e.g. inthe Galactic center source GCS 3, the 3.4 µ m absorption band is stronger than the 3.3 µ m band bya factor of 35; Chiar et al. 2000). Therefore, the bulk of the 3.4 µ m absorber in the ISM must behydrocarbon material in the larger grains, evidently more strongly aliphatic than the UIE carriers(Dartois et al. 2007).
3. Constraints from the 6.85 µ m Feature
In addition to the 3.4 µ m C–H stretching mode, aliphatic hydrocarbon materials also have twoC–H deformation bands at 6.85 µ m and 7.25 µ m. These two bands have been observed in weakabsorption in the diffuse ISM (Chiar et al. 2000). They are also seen in emission in interstellar andcircumstellar UIE sources. Their strengths (relative to the nearby 7.7 µ m C–C stretching band)also allow an estimate of the aliphatic fraction of the UIE carrier. One may argue that in the KZ11-type coal- or kerogen-like material, the aliphatic C–H bands may not occurat the same wavelengths as for pure aliphatics or PAHs with simple aliphatic sidegroups: the aliphatic H atomsoccupy a broad range of local chemical environments, subject to hydrogen bonding perturbations by nearby O andS atoms. Such interactions could conceivably shift the C–H frequencies from their “normal” aliphatic positions.However, laboratory measurements have shown that the aliphatic C–H bands in coal or kerogen do occur at 3.4 µ mand 6.85 µ m, displaying little wavelength shift compared to that of pure aliphatics (see Papoular 2001). Coals or kerogens do not exhibit a distinct band at 7.7 µ m and thus one cannot infer their aliphatic fractionsfrom I . /I . . Comparison of the 3.15–3.65 µ m emission spectrum of the Orion Bar (position 4; black; Sloan et al. 1997)with the optical depth ( absorption ) spectra of GCS 3 (a Galactic center source; blue; Chiar et al. 2000). The weaknessof the 3.4 µ m feature in the emission spectrum indicates that the aliphatic component must be minor, even assumingthat the 3.4 µ m emission is exclusively due to aliphatic C–H (i.e., neglecting anharmonicity and superhydrogenation).In contrast, the absorption spectrum of the diffuse ISM toward GCS 3 is dominated by aliphatic hydrocarbon. Fig. 2.—
Examples of “superhydrogenated” PAHs with methyl (-CH ) aliphatic sidegroups. In addition to an-harmonicity, superhydrogenation and methyl-like aliphatic sidegroups attached to PAHs may contribute to the weak3.4 µ m emission feature accompanying the 3.3 µ m feature. I . and I . be the observed intensities of the 6.85 µ m and 7.7 µ m emission features. Let A . and A . be the strengths of the 6.85 µ m aliphatic C–H band and the 7.7 µ m aromatic C–Cband. We take A . = 2 . × − cm per CH or CH group, an average of that measured formethylcylcohexane ( A . = 3 . × − cm per CH group; d’Hendecourt & Allamandola 1986)and for hydrogenated amorphous carbon ( A . = 1 . × − cm per CH or CH functional group;Dartois & Mu˜noz-Caro 2007). We take A . = 5 . × − cm per C atom for charged aromatic molecules (Draine & Li2007). Let N ′ C , aliph and N ′ C , arom respectively be the numbers of aliphatic and aromatic C atomsin the emitters of the 6–8 µ m UIE bands. Let B λ ( T ) ∝ λ − / [exp ( hc/λkT ) −
1] be the Planckfunction at wavelength λ and temperature T (where h is Planck’s constant, c is the speed oflight, and k is Boltzmann’s constant), with B . /B . ≈ . ± . < T < N ′ C , aliph /N ′ C , arom ≈ ( I . /I . ) × ( A . /A . ) × ( B . /B . ) ≈ .
10 for NGC 7027 ( I . /I . ≈ . N ′ C , aliph /N ′ C , arom ≈ .
14 for the Orion bar ( I . /I . ≈ . I . /I . ≈ . µ m UIE bands are predominantly aromatic, with <
15% of the C atoms in aliphatic form. The aliphatic fraction, while still small, appears to behigher than estimated for the 3.3–3.4 µ m band carriers, consistent with increased aromatization ofthe smallest particles, which are heated to the highest temperatures.
4. Discussion
KZ11 attribute the broad plateau emission around 8 and 12 µ m to the aliphatic component ofthe UIE carreirs. They hypothesize that the clustering of aromatic rings may break up the simplemethyl- or methylene-like sidegroups and hence the aliphatic components may take many otherforms (e.g., -CH=CH , -CH=CH-, C=CH , C=C-H). They speculate that the in-plane and out-of-plane bending modes of these sidegroups may combine to form the plateau. We note that thePAH model naturally accounts for the so-called “plateau” emission through the combined wings ofthe C–C and C–H bands. We also note that the clustering of aromatic rings and aliphatic chainswould be accompanied by forming new C–C bonds and losing H atoms. Laboratory measurementsof coals have shown that lowering the H content leads to aromatization (see Papoular 2001).KZ11 claim that the PAH hypothesis postulates that the UIE emission is excited exclusively byfar-UV photons, and that this is inconsistent with observation of UIE emission in reflection nebulaeexcited by cool stars (Sellgren et al. 1990). However, Li & Draine (2002) explicitly considered the Typical type II Kerogens have A . ≈ . × − cm per C atom (Papoular 2001). If 15% of the C atoms inkerogens are in aliphatic form, for kerogens we would have A . ≈ . × − cm per aliphatic C atom. This is closeto that adopted in this work: A . ≈ . × − cm per aliphatic C atom. ∼ µ m has been further experimentally verified(Mattioda et al. 2005). KZ11 note the constancy of UIE band ratios in regions (e.g. the Carinanebula) where the radiation intensity changes by orders of magnitude. This is precisely what oneexpects if the emission comes from single-photon heating of nanoparticles [see Figure 13 of Li &Draine (2001), Figure 4b of Li & Draine (2002), Figure 1f of Draine & Li (2007)].KZ11 note that of the more than 160 molecules identified in circumstellar and interstellarenvironments, none is a PAH. This is true, but also not surprising because the mid-IR UIE bands –the major observational information – are representative of functional groups and do not fingerprintindividual PAH molecules. KZ11 argue that the carrier of the UIR features cannot be a “purearomatic compound”. Proponents of the identification of the astronomical UIE features as comingfrom PAHs do not claim that the emitting material is “pure aromatic compound”, as strictlydefined by a chemist. The astronomical material may well include a minor aliphatic component,as well as defects, substituents (e.g., N in place of C), partial dehydrogenation, and sometimessuperhydrogenation (Tielens 2008). Some of the nanoparticles may be multilayer aggregates ofPAHs.KZ11 state that PAH molecules have strong and narrow absorption features in the UV whereasthe search for characteristic absorption features of PAHs superposed on the interstellar extinctioncurves was not successful (e.g., see Clayton et al. 2003). For individual small
PAHs, this is true.However, in the PAH hypothesis it is natural to expect that there will be a large number of distinctspecies present in the ISM, and no single UV band may be strong enough to be identified inthe UV. This also explains why laboratory-measured spectra of individual
PAHs do not preciselymatch the observed UIE features in band widths and peak wavelengths, while combined laboratoryspectra of neutral PAHs and their ions can successfully reproduce the UIE bands associated withmany different interstellar objects (Allamandola et al. 1999). There are, in fact, over 400 diffuseinterstellar bands (DIBs) in the optical that remain to be identified (Sarre 2006, Salama et al.2011). Many of these may eventually be found to be produced by specific PAHs, but at this timewe lack the laboratory spectroscopy to make the identifications. The lack of identification of anyspecific PAH is not a fatal problem for the PAH hypothesis, at least at this time. As we developa better knowledge of the gas-phase spectroscopy of the larger PAHs, this story may change. Ifthe DIBs are electronic transitions of PAHs, they hold great promise for identifying specific PAHmolecules, as the electronic transitions are more characteristic of a specific PAH molecule than the The far-IR bands are more sensitive to the skeletal characteristics of a molecule, and hence are more diagnosticof the molecular identity and more powerful for chemical identification of unknown species. In principle, far-IRspectroscopy could be used to test the KZ11 hypothesis: the KZ11-type material has an extremely “floppy” structurecompared to the more rigid PAHs, and therefore there would be many low frequency skeletal bends and very low-frequency pseudo-rotations about bond axes. However, there is little information on the far-IR spectroscopy of coalor kerogen. Even for PAHs, this information is very limited (e.g., see Joblin et al. 2009, Zhang et al. 2010). blend of π – π ∗ absorptionbands from the entire population of PAHs, with the fine structures from individual PAH moleculessmoothed out. Furthermore, internal conversion may lead to extreme broadening of the UV ab-sorption bands in larger PAHs, which may account for absence of recognizable absorption featuresshortward of 200 nm. This has been demonstrated both experimentally and theoretically. L´egeret al. (1989) measured the absorption spectra of mixtures of over 300 neutral PAH species with ∼ ∼ ∼ µ m UIE bands havemost of the PAH mass in PAHs with >
100 C atoms, see Li & Draine 2001, Draine & Li 2007).Indeed, Steglich et al. (2010) showed that larger PAHs indeed provide better fits to the observed217.5 nm feature. Cecchi-Pestellini et al. (2008) also showed that a weighted sum of 50 neutral andionized PAHs in the size range of ∼
5. Conclusion
We examine the hypothesis of mixed aromatic-aliphatic organic matter as the UIE carriers.We place an upper limit on the aliphatic fraction of the UIE carriers based on the observed weakintensities of the 3.4 µ m and 6.85 µ m emission features. By attributing them exclusively to aliphaticC–H stretch and aliphatic C–H deformation, we derive the fraction of carbon atoms in aliphatic formto be < REFERENCES
Allamandola, L.J., Tielens, A.G.G.M., & Barker, J.R. 1985, ApJ, 290, L25Allamandola, L.J., Hudgins, D.M., & Sandford, S.A. 1999, ApJ, 511, 115Barker, J.R., Allamandola, L.J., & Tielens, A.G.G.M. 1987, ApJ, 315, L61 9 –Bernstein, M.P., Sandford, S.A., & Allamandola, L.J. 1996, ApJ, 472, L127Cecchi-Pestellini, C., Malloci, G., Mulas, G., Joblin, C., & Williams, D.A. 2008, A&A, 486, L25Chiar, J.E., Tielens, A.G.G.M., Whittet, D.C.B., et al. 2000, ApJ, 537, 749Clayton, G.C., et al. 2003, ApJ, 592, 947Cody, G.D., et al. 2011, PNAS, 108, 19171Dartois, E., & Mu˜noz-Caro, G.M. 2007, A&A, 476, 1235Dartois, E., Geballe, T.R., Pino, T., et al. 2007, A&A, 463, 635d’Hendecourt, L.B., & Allamandola, L.J. 1986, A&AS, 64, 453Derenne, S., & Robert, F. 2010, Meteorit. Planet. Sci., 45, 1461Draine, B. T., & Li, A. 2007, ApJ, 657, 810Duley, W.W., & Williams, D.A. 1981, MNRAS, 196, 269Duley, W.W., & Williams, D.A. 2011, ApJ, 737, L44Geballe, T.R., Lacy, J.H., Persson, S.E., McGregor, P.J., & Soifer, B.T. 1985, ApJ, 292, 500Goto, M., Usuda, T., Takato, N., et al. 2003, ApJ, 589, 419Guillois, O., Nenner, I., Papoular, R., & Reynaud, C. 1996, ApJ, 464, 810Hrivnak, B.J., Geballe, T.R., & Kwok, S. 2007, ApJ, 662, 1059Joblin, C., Bern´e, O., Simon, A., & Mulas, G. 2009, Cosmic Dust – Near and Far (ASP Conf. Ser.414), ed. Th. Henning, E. Gr¨un, & J. Steinacker (San Francisco, CA: ASP), 383Joblin, C., & Tielens, A.G.G.M. 2011, PAHs and the Universe: A symposium to Celebrate the 25thAnniversary of the PAH Hypothesis, EAS Publ. Ser., 46Joblin, C., L´eger, A., & Martin, P. 1992, ApJ, 393, L79Joblin, C., Tielens, A.G.G.M., Allamandola, L.J., & Geballe, T.R. 1996, ApJ, 458, 610Jourdain de Muizon, M., Geballe, T.R., d’Hendecourt, L.B., & Baas, F. 1986, ApJ, 306, L105Kondo, T., Kaneda, H., Oyabu, S., et al. 2012, ApJ, 751, L18Kwok, S., & Zhang, Y. 2011, Nature, 479, 80 (KZ11)L´eger, A., & Puget, J. 1984, A&A, 137, L5 10 –L´eger, A., Verstraete, L., d’Hendecourt, L., D´efourneau, D., Dutuit, O., Schmidt, W., & Lauer,J. 1989, in Interstellar Dust (IAU Symp. 135), ed. L.J. Allamandola & A.G.G.M. Tielens(Dordrecht: Kluwer), 173Li, A., & Draine, B.T. 2001, ApJ, 554, 778Li, A., & Draine, B.T. 2002, ApJ, 572, 232Mattioda, A.L., Allamandola, L.J., & Hudgins, D.M. 2005, ApJ, 629, 1183Mu˜noz-Caro, G.M., Ruiterkamp, R., Schutte, W.A., Greenberg, J.M., & Mennella, V. 2001, A&A,367, 347Papoular, R. 2001, A&A, 378, 597Pauzat, F., Talbi, D., & Ellinger, Y. 1999, MNRAS, 304, 241Pendleton, Y.J., & Allamandola, L.J. 2002, ApJS, 138, 75Puget, J.L., L´eger, A., & d’Hendecourt, L. 1995, A&A, 293, 559Salama, F., Galazutdinov, G. A., Kre lowski, J., Biennier, L., Beletsky, Y., & Song, I.-O. 2011, ApJ,728, 154Sandford, S.A. 1991, ApJ, 376, 599Sarre, P.J. 2006, J. Mol. Spec., 238, 1Schutte, W.A., Tielens, A.G.G.M., & Allamandola, L.J. 1993, ApJ, 415, 397Sellgren, K., Luan, L., & Werner, M.W. 1990, ApJ, 359, 384Sloan, G.C., Bregman, J.D., Geballe, T.R., Allamandola, L.J., & Woodward, C.E. 1997, ApJ, 474,735Steglich, M., J¨ager, C., Rouill´e, G., Huisken, F., Mutschke, H., & Henning, T. 2010, ApJ, 712, L16Tielens, A. G. G. M. 2008, ARA&A, 46, 289Tokunaga, A.T. 1997, in ASP Conf. Ser. 124, Diffuse Infrared Radiation and the IRTS, ed. H.Okuda, T. Matsumoto, & T. Roellig (San Francisco, CA: ASP), 149Wagner, D.R., Kim, H., & Saykally, R.J. 2000, ApJ, 545, 854Zhang, J., Han, F., Pei, L., Kong, W., & Li, A. 2010, ApJ, 715, 485Yamagishi, M., Kaneda, H., Ishihara, D., et al. 2012, A&A, 541, A10