The physics which governs the IR emission of a large isolated molecule (e. g. an interstellar PAH) after the absorption of an UV photon is described. We show that the simple thermal approximation is valid and we give the method to calculate emission spectra from absorption spectroscopy data.

This discussion followed the two talks, one by John Barker on “Grains or Molecules, Thermal or Non-Thermal,” the second, by Louis d'Hendecourt on “The Particle/Thermal Approach to the Physics of Emission from Interstellar PAHs.” The presentations discussed the factors involved in the process of absorption of an ultraviolet photon by a PAH molecule or ion, followed by infrared and/or visible emission. Model calculations were presented for various stages of those sequential events, as a function of PAH size. The material presented applied to large molecules in general and was not restricted to the PAH case. The models sought to investigate the radiative, and nonradiative intramolecular processes, including dissociation, that are competitive for conversion of the absorbed ultraviolet photon energy. The essential differences between the Barker and d'Hendecourt approaches were brought out by estimations of the range of physical parameters which delimited the approximate validity of a thermal treatment using Boltzmann statistics or the Planck function as compared with a more formally correct quantum statistical (i. e. non-thermal) approach to energy conversion and dissipation. It was concluded that the thermal approximation was reasonably satisfactory at high (absorbed) photon energies in explaining conversion to an infrared emission situation, but was unsatisfactory at low energies, when the (vibrational) excitation of the excited molecule was of the same order of magnitude as the emitted infrared photons. It was also not valid for discussions of the competitive unimolecular dissociation of excited PAHs, such as denydrogenation and fragmentation. In those cases the quantum statistical approach was required.

Interstellar grains serve as an intermediary in transferring stellar radiant energy or gas chemical energy into the heating of interstellar gas. Grain photoelectric heating, a process in which ultraviolet photons eject energetic electrons from grains into the gas, dominates the gas heating of the intercloud medium, diffuse clouds, and most photodissociation regions (PDRs); it is also significant in HII regions. Grain photoelectric heating in PDRs can explain the observed correlations of CII(158 μm) with CO J=1-0 intensities and CII(158 μm) + OI(63 μm) with infrared continuum (grain) luminosities. Gas-grain collisions and the infrared emission from grains can dominate the gas heating in molecular clouds with embedded stars. The ejection of newly-formed, vibrationally-excited H2 molecules from grains behind dense, dissociative shocks leads to postshock gas heating which produces strong far-infrared and millimeter line emission as well as H2O maser emission.

Recent observational and theoretical studies of dust in dense clouds are reviewed with an emphasis on the growth of dust grains through accretion and coagulation. IR reflection nebulae around protostellar objects are useful probes of grain sizes in dense clouds. For example, detailed studies of the IR reflection nebula surrounding OMC 2-IRS 1 show that the (scattering) grains are much larger (Ã 5000 å) than in the diffuse interstellar medium. Likewise, the presence of a weak shoulder at 2.95 μm on the 3.08 μm feature in BN indicates the importance of scattering by icy grains and implies a very similar increase in the grain size.

Theoretical studies of grain surface chemistry predict the possible presence of three distinctly different grain mantle components in dense clouds depending on the physical conditions in the gas phase. These are: 1) A hydrogenated mantle dominated by H2O and CH3OH; 2) An inert grain mantle dominated by CO and O2; and 3) An oxidized grain mantle dominated by CO2. Although the importance of H2O dominated grain mantles was known for 10 yrs, the presence of CH3OH was only recently confirmed. Furthermore, recent studies of the solid CO band have revealed the presence of at least two distinctly different interstellar grain mantle components along the line of sights towards most stars: One dominated by polar and one by non-polar molecules. Although specific identification of the molecules mixed in with the CO in these components is difficult, it is quite possible that the former component is dominated by H2O and the latter by CO itself, as suggested by theoretical models. Finally, the photochemical evolution of icy grain mantles is briefly reviewed and it is suggested that the resulting complex molecular mantles may evolve into amorphous carbon mantles in the diffuse ISM.

Grain-grain collisions can lead to large modifications of the interstellar grain size distribution. At high velocities (v ≳ 1 kms−1) shattering into many small fragments will be important, while at low velocities (v ≲ 10 ms−1) coagulation dominates. Both processes can play a role in dense molecular clouds. The sticking of grains at low velocities is discussed in some detail and it is concluded that coagulation in molecular clouds is only important if the colliding grains are covered by icy grain mantles.

Thus, a model for interstellar dust is proposed in which small (≲ 500 å) silicate and carbonaceous grains are “glued” together in large (Ã 3000å), open conglomerates by a polymerized, all enveloping grain mantle. This structure resembles that of certain interplanetary dust particles collected in the upper stratosphere.

A review is given of our current knowledge of the gas phase abundance of interstellar molecules. A comparison is made between cold dust clouds, where presumably gas phase processes dominate the chemistry, and regions of high mass star formation such as Orion where both shocks and evaporation of dust grain mantles may play a role. In dust clouds, carbon rich molecules often show enhanced abundances, whereas in hot turbulent star formation regions, saturated species such as methanol and ammonia are often favored. In both types of source, there is evidence for abundance gradients and inhomogeneities. A probable cause is that the chemistry is far from a steady state situation. A useful diagnostic in such cases is the relative abundance of deuterated species and our current knowledge concerning deuterium fractionation is discussed. Recent results show that the observed degree of deuterium enhancement in species such as water, methanol, and ammonia in the Orion-KL region is much too great to be explained on the basis of normal gas phase chemistry. It is argued that a plausible explanation of the observations is that one is observing material which has recently been evaporated from dust grain mantles. One important consequence of this hypothesis is that molecular line observations may yield important information on the composition of the solid phase.

At the beginning of this decade what we knew about polarization of far-infrared emission from dense clouds was that some very good observers had looked for it and had not found it. Gull et al. (1978) had shown that the degree of polarization in Orion was not more than 2%. That information provided an important guide but very little encouragement for later efforts. There was reason to doubt whether the mechanisms invoked to explain the alignment of dust grains in the diffuse intercloud medium would operate in dense clouds; whether the strengths of the magnetic fields in dense clouds would be sufficiently greater than those in the intercloud medium to overcome the higher rate at which gas collisions would destroy alignment; and whether the field, if sufficient locally, would have enough large-scale order to give measurable polarization in far-infrared observations with large beams and large column depths.

This review examines the question of how similar far infrared and submillimeter dust properties are among galaxies including the Milky Way, and within the Milky Way. The IRAS data are examined, and no evidence is found for variations in the broad-band dust properties between 10 and 100 μm. Submillimeter data are examined in a search for a power-law index describing the emissivity of dust at λ > 100 μm. The data are consistent with a universal value of the index between 1 and 2. The evidence for large amounts of cold dust (10 to 15 K) in galaxies is reviewed and found quite soft in light of conflicting data. In most spiral galaxies, the dust-to-gas ratios are comparable to their values in the Milky Way, with a few well-documented exceptions such as the Magellanic Clouds and Blue Compact Dwarf galaxies, which have a lower ratio. No compelling evidence is found for gradients in the dust-to-gas ratio within disks of galaxies. Very small grains with fluctuating temperatures appear to be present in all galaxies examined, and to behave just like those observed in the Milky Way.

Aspects of the infrared emission from galaxies obtained from ground-based observations are described. There are clear differences between the dust in galaxies with active nuclei and those dominated by nuclear HII regions. In particular the family of unidentified emission bands between 3 and 13 μm dominate the spectra in the latter objects, but are rarely evident in spectra of active nuclei. However results from spatial, spectral and polarization measurements show that many of the dust properties in galaxies are broadly similar to those in the Milky Way.

The interpretation of the interstellar λ2175å feature is reviewed. Various proposed explanations for the feature are examined, and difficulties with each of the proposed interpretations are identified.

Comparisons between measured extinction on small particle systems and the same quantity derived from scattering calculations using bulk optical constants often show substantial disagreements. Based on such comparisons, certain pitfalls which have been recognized in such a process, as it applies to the modelling of interstellar dust properties, are pointed out. Illustrations are taken from laboratory work on SiC, graphitic carbon, and olivine-type silicates.

Dust particles in the interstellar medium are almost certainly not spherical or any other shape which allows an analytical calculation of the extinction curve, even in the Rayleign limit. Particles in soot and interplanetary dust particles are aggregates formed by subclusters which stick together. This paper uses the discrete dipole approximation (DDA) to compute the absorption and extinction curves for fractal shapes generated by this clustering process. For fractals made from graphite the UV extinction curve shows a bump near the observed 220 nm feature, and a far infrared emission efficiency many times greater than that for spheres.

Historically there have been two different types of grain modelling: One of these basically uses particle populations which evolve essentially by coagulation (e. g., the MRN model: Mathis, Rumpl and Nordsieck, 1977); the other considers the physical and chemical evolution of the particles with a particular emphasis on changes not only in sizes but also in chemical and morphological structure (e. g. Greenberg, 1978; Williams, 1989). The model of Oort and van de Hulst (1946) was the first to consider that grains must evolve in interstellar space by treating both growth and destruction in clouds. The chemical properties had already been derived by van de Hulst (1946) and then later described as the dirty ice model which consisted of the saturated molecules H2O, CH4 and NH3 with trace constituents of other atoms and molecules resulting from surface reactions of atoms on the grains. How such grains could nucleate was left as an unsolved problem but the fact that, once formed, there did not seem to be any reason why they should not grow until they exhausted the condensable atoms in the gas led to the suggestion that a limiting destructive mechanism must be provided. This was assumed to be by grain-grain collisions within clouds moving at relative speeds of 10 km s−1. We thus had the first dynamical theory leading to a steady state distribution of grain sizes. This model provided for me the starting point of the core-mantle model of grains. The observations of the 60's and henceforth clearly showed the existence of other types of small particles, which have been invoked to explain the 2200 å hump (Stecher and Donn, 1965), the far ultraviolet (FUV) extinction (Greenberg and Chlewicki, 1983), and now certain infrared emission features. These other components notwithstanding, grains still account for the major fraction of the solid particle mass in space.

Several theories involving bare silicate grains, combined with carbon in either amorphous or graphitic form, are discussed and contrasted to the other grain theories outlined at this Symposium. The assumption of bare grains is justified, and power-law size distributions are strongly indicated from both theoretical and observational standpoints. Various successes and problems of these and other theories are discussed. All of the theories discussed in this Symposium can probably explain the observed extinction and polarization in general terms, but certain crucial observations, such as the wavelength behavior of the polarization and extinction in the wavelength range near the strong silicate absorption bands, will probably be able to discriminate among them.

A new model of interstellar grains is proposed in which grains in diffuse clouds consist of small (radius ≲ 10nm) and large asymmetric (0.05μm – 0.25μm) silicate cores thinly coated with mantles of amorphous carbon (thickness ≲ 5nm). This model can account successfully for many of the observed properties of interstellar dust and gives a simple interpretation of the interstellar extinction curve. The extinction bump at 220nm is attributed to absorption by finely divided silicates, as indicated by laboratory data. The large silicates provide a “grey” background extinction through the visual and UV, but the bulk of the extinction in this region is attributed, on the basis of laboratory data, to the amorphous carbon coatings. The average interstellar carbon depletion required in this model is about 50%.

Wide variations in the observed interstellar extinction along different lines of sight are well known. These variations have a natural explanation in this model in terms of two parameters: the thickness of the carbon coatings, and the proportion of graphitic to diamond-like structure within the amorphous carbon. The underlying silicate cores are generally unchanged in these variations. The average interstellar extinction curve requires roughly equal proportions of graphitic and diamond-like forms of amorphous carbon. A higher graphitic fraction produces more visual extinction and a higher diamond-like fraction produces more far UV extinction. Varying both the proportions, and the total carbon content encompasses a wide range of extinction curve shapes.

Amorphous carbon deposited at low temperature is generally diamond-like. Temperature excursions in the material reduce the hydrogen content and enhance the graphitic nature of the material. In the interstellar medium, this process has a time scale ≳ 106 yr. Carbon coatings are therefore expected to be deposited in the interstellar medium in diamond-like form and to be slowly converted to graphitelike. The growth of carbon mantles will be reversed by intermittent shocks Thus, interstellar space should contain grains with a range of thicknesses of carbon coatings, and compositions between graphitic and diamond-like.

The chemical and physical properties of amorphous carbon have been the subject of intensive laboratory study. These properties enable an understanding of a variety of observations of dust (especially in the IR) and lead to a number of predictions which are described in this paper.

A summary of the open discussion is provided. Topics ranged over composite particles, the 10 μm silicate feature, HAC, predictions, the 2200 å feature, carbon depletion, and far ultraviolet extinction.

Substantial enrichment of deuterium, D, in certain components of chondritic meteorites is interpreted as a record of isotopic fractionation during ion-molecule reactions at the very low temperatures characteristic of dense interstellar clouds. Whether those meteorites still contain the actual molecules that were synthesised in the presolar interstellar medium, or whether the interstellar material was recycled into a later generation of molecules within the early solar system is not known.

Well-preserved interstellar grains occur in carbonaceous chondrites. Diamond (10-100å) contains r- and p-process Xe (suggesting a supernovae connection), and appears to have formed by stellar condensation, not by interstellar shocks. SiC (mainly 0.1-1 μm) is labeled either with s-process Xe, Kr (red giant?) or with Ne22, N15 (nova?), and shows large isotopic variations in N (100x), C (16x), and Si (1.3x). As Si is unaffected by H, He-burning, the Si variations reflect the isotopic heterogeneity of the ISM on the scale of individual stars; the Si compositions found thus far require at least 6 separate stars.

During the past decade interplanetary dust particles (IDPs) have been collected in the earth's stratosphere. Isotopic studies of these particles have demonstrated that many of them are greatly enriched in deuterium and at least some of them carry this enrichment in smaller subcomponents. Deuterium enrichments of a similar magnitude are seen in simple molecules in interstellar clouds. Deuterium enrichment in IDPs can be taken as evidence for the presence of interstellar material. It is not clear at this time whether the carriers of the isotopic anomalies represent true, unaltered interstellar dust grains, or whether they represent an altered component with a molecular ‘memory’ of original interstellar grains. The spectra of different components in the collected dust provide suggestive matches to similar components evident in the astronomical spectra of dust in comets, dense molecular clouds, and emission nebulae. The known extraterrestrial nature of the particles, the possible presence of interstellar material in them, and their spectral similarity to many astronomical objects all argue that the collected IDPs provide useful analogs for the modelling of interstellar dust.

The Comet Halley campaigns resulted in an enormously better understanding of the composition of comets. Silicates, organic compounds, and volatile ices comprised of H2O, CO, CO2, S2, CN, and possibly CH4 and OCS occur in comet grains. These are all known or suspected constituents of interstellar dust. We review the chemical, elemental, and isotopic compositions of comet dust and compare this with interstellar matter. The many intriguing parallels suggest, but do not yet establish, a direct connection between comet dust and interstellar dust.

Grains are injected into the interstellar medium (ISM) from evolved stars and supernovae; in addition, supernova ejecta may condense onto pre-existing grains before becoming well-mixed with the interstellar gas. Once in the ISM, grains can grow by accretion, but are also subject to destruction by interstellar shocks. The current status of the theory of shock destruction of interstellar grains is reviewed briefly. Small grains are destroyed by thermal sputtering in fast, nonradiative shocks; large grains are destroyed by grain-grain collisions and eroded by nonthermal sputtering in radiative shocks. The dominant shocks in the ISM are from supernova remnants (SNRs), and the mass of grains destroyed is proportional to the energy of the SNR. In a multiphase ISM, these shocks destroy the grains at a rate proportional to the volume filling factor of the phase; since the density of the hot phase is too low for efficient grain destruction, most of the destruction occurs in the warm phase. Not all SNRs are effective at destroying grains, however: some are above the gas disk, and some —Type IPs in associations—are highly correlated in space and time. The galactic SN rate is observed to about 2.2 per century (van den Bergh, 1983), but the effective supernova rate for grain destruction is estimated to be only about 0.8 per century. As a result, the timescale for the destruction of a typical refractory grain in the ISM is inferred to be about 4 × 108 yr for either a two-phase or a three-phase ISM. Most of the refractory material in the ISM (other than carbon) is injected by supernovae, not evolved stars; the net injection timescale is estimated as about 1.5 × 109 yr. Comparison of the destruction and injection timescales indicates that the fraction of grains injected by stars which survive in the ISM is only about 20%. Most of the refractory material in interstellar grains must, therefore, have accreted onto the grains in the ISM. Nonetheless, a significant fraction of dust formed in stars survives in the ISM and may be detectable in meteorites and interplanetary dust particles.