Interstellar and circumstellar grain formation and

Downloaded from rsta.royalsocietypublishing.org on July 23, 2011

Interstellar and circumstellar grain formation and survival Anthony P. Jones Phil. Trans. R. Soc. Lond. A 2001 359, 1961-1972 doi: 10.1098/rsta.2001.0890

Email alerting service

Receive free email alerts when new articles cite this article - sign up in the box at the top right-hand corner of the article or click here

To subscribe to Phil. Trans. R. Soc. Lond. A go to: http://rsta.royalsocietypublishing.org/subscriptions

This journal is © 2001 The Royal Society


10.1098/rsta.2001.0890

Interstellar and circumstellar grain formation and survival By A n t h o n y P. J o n e s Institut d’Astrophysique Spatiale, Universit¶e Paris Sud, Bâtiment 121, 91405 Orsay Cedex, France ([email protected]) Dust formation is primarily associated with stars in their dying throes, e.g. when low-mass stars reach the red-giant or asymptotic-giant branch (AGB) phase of their evolution, or when massive stars explode as supernovae (SNe). While the contribution of AGB stars to the galactic dust budget is signi cant, both in terms of variety and quantity, that due to SNe is not yet clear. AGB stardust formation includes grains of amorphous and crystalline silicates, hydrogenated carbons, silicon carbide and graphite. However, not all of these materials have yet been detected in circumstellar regions or in the interstellar medium (ISM). The derived lifetimes for these materials in the ISM appear to be short compared with the time-scale for the formation of new dust. Thus a grain lifetime and propagation problem is posed. Apparently, it is also necessary to reform and grow grains in the ISM, through accretion and coagulation processes, in order to explain interstellar dust observations. This paper discusses dust formation in circumstellar and interstellar environments, dust sources and their contributions to the galactic dust budget, and dust survival and propagation in the ISM. Keyword s: interstellar dust; circumstellar dust; dust composition; dust formation; dust processing in the ISM

1. Introduction Dust is primarily formed in the shells around stars in the red-giant and asymptoticgiant branch (AGB) phases of their evolution (e.g. M giants, carbon stars and radio luminous OH/IR stars), but some small fraction is also formed in the circumstellar shells around supergiants, novae, planetary nebulae (PNe), WC stars and in the ejecta of supernovae (supernovae (SNe) types Ia and II). The presence of circumstellar dust is generally revealed by infrared (IR) thermal continuum emission from the dust near the star and also by the opacity of this dust. On entering the interstellar medium (ISM), i.e. through the e¬ects of stellar winds, dust formed in circumstellar regions is subject to processing in the ISM. This processing may include erosion, fragmentation and destruction in SN-generated shock waves, grain growth via mantle accretion and coagulation in quiescent clouds, and fragmentation/coagulation in turbulent interstellar clouds. (a) The composition of interstellar dust In the ISM we have direct evidence for amorphous silicate grain materials from the Si{O bond stretching and O{Si{O bending modes at 10 and 20 m m, respecPhil. Trans. R. Soc. Lond. A (2001) 359, 1961{1972

1961

® c 2001 The Royal Society


A. P. Jones

1962

Table 1. Interstellar dust properties composition

nature

evidence

àromatic’ carbons (hydrogenated)

small heat capacity, stochastically heated

depletion of carbon, extinction, àromatic’ emission bands

carbonaceous (hydrogenated)

amorphous grains (not in mantles)

depletion of carbon, extinction, unpolarized aliphatic C{H band

silicate/metal oxide

amorphous, aspherical, magnetic inclusions

Si, Mg and Fe depletions, extinction, Si{O bands at 10 and 20 m m, optical and IR polarization

amorphous ices (H2 O, CO, CO2 , CH3 OH, H2 CO, etc.)

mantles on silicate grains

IR absorption bands, polarized 3 m m H2 O ice band

tively, observed in absorption toward stars without dusty circumstellar envelopes (see, for example, Mathis 1990). There is also evidence for hydrocarbon grains from the aliphatic 3.4 m m C{H stretching vibration, which is seen in absorption in the di¬use ISM, and also from the ubiquitous IR emission bands seen at 3.3, 6.2, 7.7, 8.6, 11.3 and 12.7 m m, which are attributed to C{H and C{C modes in an aromatic hydrocarbon dust component (polycyclic aromatic hydrocarbons (PAHs) or hydrogenated carbon particles (see, for example, Dorschner & Henning 1995)). In addition, the absorption bump in the extinction curve at 217.5 nm is also attributed to carbonbearing dust, generally of graphitic composition (see, for example, Draine 1995). We also have indirect evidence for carbonaceous, silicate and Fe/Mg oxide dust from the observed depletion of elements in the di¬use ISM, i.e. C, Si, Mg, Fe, Fe and O (see, for example, Savage & Sembach 1996). Although graphite is often used in interstellar dust models, there is currently no direct evidence for this material in the ISM. Table 1 summarizes the nature and composition of the dust in the di¬use and dense ISM, as inferred from observations of interstellar extinction, scattering, emission and polarization, and also from elemental depletion studies (see, for example, Whittet 1992; Evans 1994; Dorschner & Henning 1995). This dust has a continuous size distribution ranging from the smallest|ca. 1{10 nm grains (aromatic hydrocarbons/PAHs and amorphous carbons)|up to ca. 1 m m sized grains (amorphous silicate). (b) The composition of circumstellar dust The broad and featureless amorphous silicate bands at 10 and 20 m m have been observed in absorption and emission in dusty stellar envelopes (see, for example, Dorschner & Henning 1995), and also in regions close to bright stars (see, for example, Cesarsky et al. 2000). These bands were long ago attributed to silicates (Woolf & Ney 1969). However, Infrared Space Observatory (ISO) observations of AGB circumstellar shells show additional narrower emission bands in the 10{45 m m region. These bands are attributed to crystalline silicates, e.g. Mg-rich olivine (forsterite) and Mg-rich pyroxene (enstatite) silicates. The detection of these crystalline silicates is evident only in high mass-loss rate shells (Waters et al. 1996), but this may simply be due to sensitivity-limit and radiative transfer e¬ects (Kemper et al. 2001). Phil. Trans. R. Soc. Lond. A (2001)


Interstellar and circumstellar grain formation

1963

Table 2. Typical pre-solar grains found in primitive meteorites (All data are taken from Frisch et al. (1999).) abundance (ppm)

isotopic tracersa

stellar sources

Xe{HL

type II SNe

composition

typical radii

diamond

2 nm 0.05{10 m m

1400 14

13

0.4{6 m m

10

12

C, 1 8 O extinct 4 4 Ti

type II SNe, (Wolf{Rayet stars)

0.15{2.5 m m

0.3

17

red giant, AGB stars

SiC (X grains)

0.25{5 m m

0.1

12

C, 1 5 N, 2 8 Si extinct 2 6 Al, 4 4 Ti

type II SNe

Si3 N4

ca. 0.5 m m

0.002

12

type II SNe

SiC (mainstream) graphite Al2 O3

C,

O,

14

18

N,

22

Ne

O

C, 1 5 N, 2 8 Si extinct 2 6 Al

C-rich AGB stars

a

Bold face (normal) type indicates an isotopic enhancement (depletion) with respect to the Solar System isotopic composition, except for extinct species.

It is interesting to note that crystalline silicates have only been unambiguously observed in circumstellar shells and not in the ISM. Apparently, all interstellar silicates are amorphous (see, for example, Mathis 1990); this could be explained by the amorphitization of crystalline silicates in interstellar shock waves (Demyk et al . 2001). Another Si-rich material, silicon carbide (SiC), has long been detected in circumstellar shells (see, for example, Tre¬ers & Cohen 1974), but, as is the case for crystalline silicates, it is not seen in the ISM. The composition of the observed circumstellar SiC appears to be consistent with the meteoritic SiC grain composition, i.e. -SiC (Speck et al. 1999). Diamond is yet another material whose presence is only indicated in circumstellar regions. Its presence has been inferred from an emission band seen at ca. 21 m m in some protoplanetary nebulae (PPNe) (Hill et al . 1998) and from emission bands at 3.43 and 3.53 m m (Guillois et al. 1999). (c) Interstellar and circumstellar dust in the Solar System Pre-solar dust (see table 2) includes nanometre-sized grains of diamond and micrometre-sized grains of silicon carbide (SiC), graphite, corrundum (Al2 O3 ) and silicon nitride (Si 3 N4 ). TiC has also been found as inclusions in the pre-solar graphite grains (Bernatowicz 1997). For a review of the characteristics of the three most abundant pre-solar grain types (i.e. diamond, SiC and graphite), see the review by Anders & Zinner (1993). The pre-solar grains, ranging from nanometres to tens of micrometres in size, have been extracted from primitive meteorites, i.e. those that underwent little thermal alteration during the formation of the Solar System (see, for example, Cameron 1973). The extracted pre-solar grains are all of very refractory materials, a property that ensures their survival in the ISM and through the laboratory extraction processes. They have isotopic anomalies that clearly indicate their pre-solar origins. For example, the diamonds are associated with type II SNe (Lewis et al . 1987), most SiC grains were formed in the atmospheres of AGB stars (Lewis et al . 1994), Phil. Trans. R. Soc. Lond. A (2001)


A. P. Jones

1964

the graphite grains formed around massive stars (Hoppe et al . 1992), and the oxide grains were formed around red-giant and AGB stars (Nittler 1997). The fact that pre-solar circumstellar dust is found in the Solar System implies that these grains must have been transported through the ISM and incorporated into the solar nebula 4.5 billion years ago (see, for example, Cameron 1973). Note, however, that the extracted pre-solar grains are compositionally very di¬erent from interstellar grains (see table 1). It has been suggested (Bradley 1994) that the grains known as GEMS (glass with embedded metal and sulphides) found in interplanetary dust particles represent silicate stardust that has been processed in and transported through the ISM. This intriguing suggestion has yet to be con rmed. The Solar System is today actually passing through a cloud of interstellar gas and dust. The evidence for this comes from the detectors on the Ulysses spacecraft (Grun et al . 1993, 1994; Frisch et al . 1999) and radar studies of meteoroids in the Earth’s atmosphere (Taylor et al . 1996). In both cases, particles with velocities in excess of the escape velocity for the Solar System at the point of detection were observed. These grains must therefore also be of interstellar origin. The particles detected by the Ulysses experiments (mean size ca. 0.8 m m) are somewhat larger than the typical interstellar grains (ca. 0.005{0.5 m m), and the meteoroids detected in the radar studies are signi cantly larger (sizes ca. 15{40 m m).

2. Dust formation in circumstellar environments Observationally, stars with the highest mass-loss rates show the presence of dust, implying a link between high mass-loss rates and the formation of dust (see, for example, Jura 1987). In essence, high mass-loss rates correspond to high densities near the photosphere, which are conducive to dust formation. (a) The dust formation and growth The critical step in circumstellar grain formation is the production of the nuclei that seed grain growth in the dense photosphere. Early studies of particle nucleation in the astrophysical context showed that approaches based on the classical homogeneous nucleation theory are not valid (see, for example, Nuth & Donn 1983). Nucleation is probably the least understood step in the dust-formation process despite early progress made in this eld (see, for example, Frenklach & Feigelson 1989; Woitke et al . 1993) and will not be discussed here. Based on ISO data, Demyk et al. (2000) have shown that the dust composition around OH/IR stars is consistent with the oxygen-rich dust condensation sequence. Hence it seems that once oxygen-rich grains nucleate and grow, they do so in accordance with expectations. Once formed in circumstellar shells, grains are driven away from the central star by a radiation pressure force, which is counteracted by gas-drag. The net e¬ect of these two forces determines the terminal velocity of the grains, and therefore the rate of grain growth in a circumstellar shell (Tielens 1983; Dominik et al. 1989). In this process, the grains couple with the gas through gas{grain collisions and drive stellar mass loss (see, for example, Dominik et al. 1989). Following Spitzer (1978), the radiation pressure force on a grain in a circumstellar region is given by Fp r = º a2 hQp r ifL? =(4º r 2 c)g, where a is the grain radius, hQp r i Phil. Trans. R. Soc. Lond. A (2001)



1965

Table 3. The contributions of stellar sources to dust in the ISM (See, for example, Dorschner & Henning (1995), Jones et al . (1997) and Dwek (1998).)

stellar source

( 10¡

M giants RL OH/IR stars C stars supergiants novae PN WC stars SN type II SN type Ia

contribution 6 M kpc¡ 2 yr¡ 3 3 2 0.2 0.003 0.2 0.03 0.03 0.15 14 0.03 2.3

1

)

type(s) of dust formed silicates silicates SiC, carbonaceous silicates silicates, SiC, carbonaceous carbonaceous carbonaceous silicates, carbonaceous silicates, carbonaceous

is the radiation pressure e¯ ciency factor, L? is the stellar luminosity, r is the distance from the star and c is the velocity of light. The opposing gas-drag force is approximately given by Fd = º a2 » gas vd 2 , where » gas is the gas density, and vd is the dust drift velocity with respect to the gas. The stellar mass-loss rate is given by M_ = 4º » gas r 2 v, where v is the terminal velocity of the gas, and we can then write the dust drift velocity (cf. Habing et al . 1994) as µ ¶ µ ¶1=2 hQp r iL? v 1=2 a vd ’ ’ 1:9 km s¡ 1 ; (2.1) _ 50 nm Mc where typical cool stellar radiation eld parameters (i.e. v = 10 km s¡ 1 , L? = 5000L and M_ = 10¡ 5 M yr¡ 1 ) and a radiation pressure e¯ ciency appropriate for silicates (hQp r i ’ 0:02a=50 nm) have been adopted. Carbonaceous particles have slightly higher e¯ ciencies (hQp r i ’ 0:08a=50 nm) and hence slightly higher drift velocities. Equation (2.1) indicates that the drift velocity increases with decreasing massloss rate and, for a mass-loss rate of 10¡ 8 M yr¡ 1 , it reaches a limiting value of ca. 40 km s¡ 1 for 50 nm silicate grains. In terms of the gas and dust balance in the ISM, the relatively few sources with the highest mass-loss rates (greater than 10¡ 5 M yr¡ 1 ) can dominate dust input to the ISM (see, for example. Jura 1987). Typical drift velocities for these types of envelopes are of order 1 km s¡ 1 , which is less than the threshold velocity for the sputtering of dust materials, and destruction of the grains is therefore unimportant. However, the lower threshold velocities (ca. 2 km s¡ 1 ) for shattering in grain{grain collisions implies that grain fragmentation could be important (Jones et al . 1996; see also x 3). (b) Dust sources and the galactic dust budget In table 3 the contributions of the major sources of dust in the ISM are given (see, for example, Dorschner & Henning 1995; Jones et al. 1997; Dwek 1998). The dust from these circumstellar sources is injected into the ambient ISM by stellar winds (see x 2 a). The dust-formation rate is thus of order 8{30 10¡ 6 M kpc¡ 2 yr¡ 1 averaged over the Galaxy. The exact rate depends on the e¯ ciency of dust formation in SNe, which is not yet known (see, for example, Jones et al . 1997; Dwek 1998). Phil. Trans. R. Soc. Lond. A (2001)


A. P. Jones

1966

However, SNe certainly do contribute some dust because grains with type II SN isotopic composition are recovered from meteorites. Additionally, even if SNe do not directly contribute a signi cant fraction to the dust budget, they do provide the elemental building blocks for refractory dust (e.g. the Si and Fe atoms that can form silicate grains). However, to date, no Fe-containing pre-solar grains have been identi ed. Assuming that the interstellar dust mass is 1% of the galactic gas mass (i.e. 0:01 5 109 M ), the entire galactic dust mass can be replenished on a time-scale of the order of 3 109 yr (see, for example, Jones & Tielens 1994).

3. Dust survival, formation and propagation in the ISM Pre-solar dust studies clearly show that some circumstellar dust grains survived residence in, and transport through, the ISM before their eventual incorporation into Solar System meteorites. However, these grains are of refractory materials (e.g. diamond, SiC, graphite and Al2 O3 ) and are not typical interstellar grains (see x 1 c). If grains of less refractory materials, e.g. crystalline and amorphous silicates, and amorphous carbons, are present, they are lost during the laboratory extraction process. In any event, it is not clear that these less-refractory grains could ever retain an isotopic memory of their formation site. This is because, in the ISM, grains undergo erosional reprocessing in high-energy shocks, followed by reformation in quiescent regions. Thus, in time, they should become chemically and isotopically homogenized. True interstellar grains will therefore not be easy recognize in meteorites. (a) Dust survival in the ISM and its consequences Dust in the ISM is subject to processing in interstellar shock waves arising from cloud{cloud collisions and SNe. Observations show that strong shocks destroy dust (see, for example, Routly & Spitzer 1952; Cowie 1978; Crinklaw et al . 1994; Savage & Sembach 1996). Shocks drive gas{grain collisions, which result in the sputtering and erosion of grains, and also grain{grain collisions, which lead to the vaporization and fragmentation of grains (see, for example, Jones et al . 1994, 1996, 1997). The fragmentation process may indeed be at the origin of the approximate power-law dust-size distribution inferred for the di¬use ISM (see, for example, Bierman & Harwitt 1980). Thus the interstellar dust-size distribution must evolve both temporally and spatially, and its dynamical evolution is determined by the balance between the physical processing in shock waves in the low-density ISM (dominated by erosion and fragmentation) and more benign processing in dense clouds (dominated by accretion and coagulation). Let us rst consider the erosional processing of interstellar dust in shock waves. The subject of dust destruction in shocks has been the focus of many theoretical studies (see Jones et al . (1997) and the references therein for a review of this subject). In the most recent of these studies (Jones et al . 1994, 1996), grain lifetimes of the order of 4{6 108 yr are indicated. In contrast, the injection time-scale for dust formed by AGB stars is of order 3 109 yr (see, for example, Jones & Tielens 1994). The di¬erence between the inferred destruction and injection time-scales is at odds with the observed depletions of the refractory elements in the ISM (Mathis 1990). Thus, based on the results of these theoretical models, we are drawn to the conclusion that dust must also be formed in situ in the ISM. Phil. Trans. R. Soc. Lond. A (2001)



1967

The problem of grain lifetimes in the ISM is further compounded if one considers the e¬ects of grain shattering in grain{grain collisions. The shattering of grains into smaller fragments, upon collision with other grains, occurs at relatively low velocities (v > 2 km s¡ 1 ), compared with vaporization in grain{grain collisions and sputtering in atom/ion{grain collisions (v > 20 km s¡ 1 ). The time-scale for grain disruption in shocks, it turns out, is about an order of magnitude shorter than the destruction timescale (Jones et al. 1996). Thus, in addition to the problem of reforming grains in the ISM through accretion, large grains must also be reformed in some phase of the ISM. In order to explain the observed interstellar extinction, a large fraction (ca. 40%) of the grain mass must be in particles larger than 100 nm. If grain shattering is e¯ cient in shock waves, then, necessarily, the coagulation of the fragments into large grains in dense clouds must also be e¯ cient. This scenario leads to the conclusion that the large interstellar grains must be porous or fractal in nature because the packing e¯ ciency in coagulated grains is not 100% e¯ cient. Interestingly, the inferred ages for the meteoritic SiC grains are ca. 1:3 108 { 2 109 yr (Lewis et al. 1994), i.e. much larger than the typical interval between SN shock waves (ca. 107 yr). It therefore seems that pre-solar grains that traversed the ISM before their incorporation into the Solar System should retain some memory of their time in the ISM, i.e. evidence of exposure to sputtering and/or surface pitting due to the cratering impact of small grains in shock waves. However, in contrast to expectations, the surfaces of the extracted SiC and graphite grains are relatively clean and, in the case of SiC, may even show crystallographic faces (Bernatowicz 1997). The most likely solution, particularly for the large SiC and graphite grains, is that the grains that survive are those that never saw a strong shock in the ISM. The dust that has been processed by shocks in the ISM has probably lost the ìsotopic memory’ of its birth-site, due to ion implantation, erosion and re-accretion, and is therefore no longer isotopically discernible in meteorites. (b) Dust formation in the ISM Field (1974) showed that elements with the highest condensation temperatures generally have the highest depletions in the ISM and that this is consistent with grain condensation in cool stellar atmospheres. However, the observed e¬ect is also consistent with the lower-temperature process of grain growth via the re-accretion of eroded atoms/ions (see, for example, Savage & Sembach 1996). Thus the elemental depletion patterns observed in di¬use clouds could arise from the selective accretion of elements with high condensation temperatures. The accretion time-scale for a cloud of density nH is ca. 109 (1 cm¡ 3 =nH ) yr. For a di¬use cloud with a typical density of a few tens of H atoms/cm3 , the accretion time-scale is therefore of the order of 107 {108 yr, i.e. similar to the neutral ISM cycling time-scale of 3 107 yr, which is driven by massive star formation in molecular clouds (see, for example, McKee 1989). Thus accretion onto grains in the low-density ISM probably has only a minor role in determining the elemental depletions. However, in higher-density regions, accretion onto pre-existing grains is faster, and under such conditions mantles will be formed far from equilibrium. Grain mantles will therefore probably be amorphous and chemically heterogeneous. Also, because the Si/C abundance ratio is much less than unity, no silicates or SiC grains are expected to form in the ISM. Interstellar dust, the dust which resides in the ISM, is then a mixture of circumstellar dust, reprocessed circumstellar dust and dust that is formed in situ in the ISM. Phil. Trans. R. Soc. Lond. A (2001)


A. P. Jones

1968

In the densest interstellar regions, i.e. the shielded environments of molecular clouds, the grains may form icy mantles of simple molecules such as H2 O, CO, CO2 and CH3 OH (see table 1 or, for example, Schutte (1999)). In these regions, icy mantle formation will probably be accompanied by relatively low-velocity grain{grain collisions, leading to grain growth via coagulation. (c) Dust propagation in the ISM Grains formed in a circumstellar shell move away from the central star through the relatively gentle outward force of radiation pressure (see x 2 a). As dust propagates away from the star, further processing can occur in the shock front (v ’ 5{20 km s¡ 1 ), where the ejecta merge with the ambient ISM. Neglecting shattering, the processing of dust in the transition region between the stellar wind and the ISM has been considered (Woitke et al . 1993), and it has been concluded that there is little dust destruction in these environments. Thus stellar winds generally provide a gentle mechanism for dust propagation into the ISM. SN shock waves, on the other hand, are e¯ cient at propagating grains over large distances through the ISM because they sweep up interstellar gas and dust, and accelerate them to velocities of the order of 10{100 km s¡ 1 . These shocks provide the kinetic energy that maintains the turbulent motions in the ISM, and are thus responsible for the turbulent di¬usion of dust in the ISM. In cold atomic and molecular clouds, where most of the dust mass resides, shocks are typically of order 10 km s¡ 1 , i.e. fast enough to shatter dust but not destroy it. However, shocks in the warm intercloud medium are faster, and the large di¬erential gas{grain and grain{grain velocities that are generated behind the shock front can lead to grain destruction and reprocessing. Thus shock waves are essential in transporting dust through the ISM, but not all of the dust survives this process (see x 3 a).

4. Summary New dust is principally formed in the circumstellar shells around AGB stars, although SNe could also be a very signi cant source. SNe undoubtedly contribute to the elemental building blocks of dust (i.e. Si, Fe, etc.), but SN-generated shock waves are the major destroyers of pre-existing interstellar dust. Currently, our knowledge of the refractory interstellar and circumstellar dust components comes from observations of dust extinction, scattering, absorption, emission and polarization, and from elemental depletion studies. However, in the pre-solar grains extracted from primitive meteorites, we also have a direct sample of grains formed in circumstellar shells. Table 4 summarizes what we currently know of the interstellar and circumstellar dust composition. The entries in table 4 indicate that interstellar and circumstellar dust compositions generally seem to be rather similar, except for the Mg-rich crystalline silicate and SiC grains. On the other hand, apart from the SiC grains in common with circumstellar dust, the pre-solar grains appear to represent completely di¬erent grain species. However, this di¬erence is almost certainly due to a selection e¬ect, i.e. the fact that only grains of the most refractory materials have so far been extracted from primitive meteorites. The less refractory components (e.g. silicates and carbonaceous materials) are currently lost during the extraction process. Phil. Trans. R. Soc. Lond. A (2001)



1969

Table 4. Current know ledge of interstellar, circumstellar and pre-solar dust components and grain composition (Square brackets indicate dust components that have not been unequivocally identi¯ed in a given phase.) interstellar

circumstellar

pre-solar

aromatic hydrocarbons aliphatic hydrocarbons

aromatic hydrocarbons aliphatic hydrocarbons

[aromatic hydrocarbons]

amorphous silicates

[Mg and Fe oxides]b

a b

[diamond] amorphous silicates [aluminosilicate] a crystalline silicates (Mg-rich olivine, Mg-rich pyroxene) [Fe oxide]a -SiC

graphite diamond

Al2 O3 -SiC TiC Si3 N4

Demyk et al . (1999). Inferred from depletion studies (see, for example, Spitzer & Fitzpatrick 1993).

Clearly, pre-solar circumstellar dust grains have somehow been transported through the ISM and incorporated into the Solar System. Stellar winds are the means by which newly formed circumstellar dust nds its way into the ambient ISM around the source. The action of these winds on the dust is relatively benign and thus it is likely that large amounts of circumstellar dust are transferred into the ISM by this means. However, stellar winds cannot propagate the dust over large interstellar distances. For transport over large distances, SNe must play a major role, but they will heavily process, or even destroy, much of the dust in the process of transporting it. Thus interstellar dust will be processed as it is propagated throughout the Galaxy by the e¬ects of SN shock waves. Most of the dust in the ISM is therefore probably not in the pristine state that it was formed in. This may explain why crystalline grains are only found near to their sites of formation, i.e. in AGB stardust shells. In the ISM, the dust is probably rendered amorphous by the e¬ects of atom/ion implantation in SN shock waves. Additionally, the lifetime for dust in the ISM is shorter than the time-scale to reform it. Hence we are drawn to the inescapable conclusion that a signi cant fraction of dust must also be formed, or reformed, in the ISM through the e¬ects of mantle accretion and coagulation in grain{grain collisions. The derived ages of the pre-solar SiC grains extracted from primitive meteorites (ca. 1:3 108 to 2 109 yr) indicate that these are grains that survived for a long time in the ISM, and also survived in the solar nebula and eventual incorporation into meteorites. Thus some fraction of pre-solar circumstellar dust can apparently survive almost unscathed in the Galaxy for at least a hundred million years. This survival is remarkable given the hostility of the galactic ISM that they must have traversed. However, the survival of these grains is evidently aided by their very refractory nature. Hence the pre-solar SiC gains are those grains that would have been the most resistant to processing in interstellar shocks. Thus the analysis of Phil. Trans. R. Soc. Lond. A (2001)


1970

A. P. Jones

the less refractory meteoritic components, i.e. the carbon, silicate and oxide particles that could be the proto-typical interstellar grains, is eagerly awaited. Will these grain components be distinct (i.e. isotopically anomalous), or isotopically `normal’ through having been completely homogenized by the processes of destruction and re-accretion in the ISM? If interstellar grains in primitive meteorites are isotopically `normal’, then these grains can probably only be recognized as being of interstellar origin through their heavily processed morphologies (i.e. implanted, cratered, fragmented, etc.) or through spectroscopy (i.e. direct comparison of meteoritic grain spectroscopy with observations of dust in the ISM). Neither of these options will be easy and it may therefore be some time before we are able to recognize and isolate `real’ interstellar grains from primitive meteorites.

References Anders, E. & Zinner, E. 1993 Interstellar grains in primitive meteorites. Meteoritics 28, 490{514. Bernatowicz, T. J. 1997 Presolar grains from meteorites. In From stardust to planetesimals (ed. Y. J. Pendleton & A. G. G. M. Tielens). ASP Conference Series, vol. 122, pp. 227{251. San Francisco, CA: Astronomical Society of the Paci¯c. Bierman, P. & Harwitt, M. 1980 On the origin of the grain-size spectrum of interstellar dust. Astrophys. J. 241, L105{L107. Bradley, J. P. 1994 Chemically anomalous, preaccretionally irradiated grains in interplanetary dust from comets. Science 265, 925{929. Cameron, A. G. W. 1973 Interstellar grains in musuems? In Interstellar dust and related topics (ed. J. M. Grrenberg & H. C. van de Hulst), pp. 545{547. Dordrecht: Reidel. Cesarsky, D., Jones, A. P., Lequeux, J. & Verstraete, L. 2000 Silicate emission in Orion. Astron. Astrophys. 358, 708{716. Cowie, L. L. 1978 Refractory grain destruction in low-velocity shocks. Astrophys. J. 225, 887{ 892. Crinklaw, G., Federman, S. R. & Joseph, C. L. 1994 The depletion of calcium in the interstellar medium. Astrophys. J. 424, 748{753. Demyk, K., Jones, A. P., Dartois, E., Cox, P. & d’ Hendecourt, L. 1999 The chemical composition of the silicate dust around RAFGL7009S and IRAS 19110+1045. Astron. Astrophys. 349, 267{275. Demyk, K., Dartois, E., Wiesemeyer, H., Jones, A. P. & d’ Hendecourt, L. 2000 Structure and chemical composition of the silicate dust around OH/IR stars. Astron. Astrophys. 364, 170{ 178. Demyk, K., Carrez, Ph., Leroux, H., Cordier, P., Jones, A. P., Borg, J., Quirico, E., Raynal, P. I. & d’ Hendecourt, L. 2001 Structural and chemical alteration of crystalline olivine under low energy He + irradiation. Astron. Astrophys. 368, L38{L41. Dominik, C., Sedlmayr, E. & Gail, H.-P. 1989 The size distribution of dust particles in a dustdriven wind. Astron. Astrophys. 223, 227{236. Dorschner, J. & Henning, T. 1995 Dust metamorphosis in the Galaxy. Astron. Astrophys. Rev. 6, 271{333. Draine, B. T. 1995 Properties of dust in di® use clouds. In The physics of the interstellar medium and intergalactic medium (ed. A. Ferrara, C. F. McKee, C. Heiles & P. R. Shapiro). ASP Conference Series, vol. 80, pp. 133{147. San Francisco, CA: Astronomical Society of the Paci¯c. Dwek, E. 1998 The evolution of the elemental abundances in the gas and dust phases of the Galaxy. Astrophys. J. 501, 643{665. Evans, A. 1994 The dusty Universe. Wiley. Phil. Trans. R. Soc. Lond. A (2001)



1971

Field, G. B. 1974 Interstellar abundances: gas and dust. Astrophys. J. 187, 453{459. Frenklach, M. & Feigelson, E. D. 1989 Formation of polycyclic aromatic hydrocarbons in circumstellar envelopes. Astrophys. J. 341, 372{384. Frisch, P. (and 15 others) 1999 Dust in the local interstellar wind. Astrophys. J. 525, 492{516. Grun, E. (and 22 others) 1993 Discovery of Jovian dust streams and interstellar grains by the ULYSSES spacecraft. Nature 362, 428{430. Grun, E., Gustafson, B., Mann, I., Baguhl, M., Mor¯ll, G. E., Staubach, P., Taylor, A. & Zook, H. A. 1994 Interstellar dust in the heliosphere. Astron. Astrophys. 286, 915{924. Guillois, O., Ledoux, G. & Reynaud, C. 1999 Diamond infrared emission bands in circumstellar media. Astrophys. J. 521, L133{L136. Habing, H. J., Tignon, J. & Tielens, A. G. G. M. 1994 Calculations of the out° ow velocity of envelopes of cool giants. Astron. Astrophys. 286, 523{534. Hill, H. G. M., Jones, A. P. & d’ Hendecourt, L. B. 1998 Diamonds in carbon-rich proto-planetary nebulae. Astron. Astrophys. 336, L41{L44. Hoppe, P., Amari, S., Zinner, E. & Lewis, R. S. 1992 Large oxygen isotopic anomalies in graphite grains from the Murchison meteorite. Meteoritics 27, 235. Jones, A. P. & Tielens, A. G. G. M. 1994 Interstellar dust|physical processes. In The cold Universe (ed. T. Montmerle, C. J. Lada, I. F. Mirabel & J. Tran Thanh Van), pp. 35{44. Gif-sur-Yvette: Editions Frontiers. Jones, A. P., Tielens, A. G. G. M., Hollenbach, D. J. & McKee, C. F. 1994 Grain destruction in shocks in the interstellar medium. Astrophys. J. 433, 797{810. Jones, A. P., Tielens, A. G. G. M. & Hollenbach, D. J. 1996 Grain shattering in shocks: the interstellar grain size distribution. Astrophys. J. 469, 740{764. Jones, A. P., Tielens, A. G. G. M., Hollenbach, D. J. & McKee, C. F. 1997 The propagation and survival of interstellar grains. In The astrophysical implications of the laboratory study of presolar materials (ed. T. J. Bernatowicz & E. Zinner), pp. 595{613. New York: American Institute of Physics. Jura, M. 1987 The Milky Way as a galaxy. In Interstellar processes (ed. D. J. Hollenbach & H. A. Thronson), pp. 3{17. Dordrecht: Reidel. Kemper, F., Waters, L. B. F. M., de Koter, A. & Tielens, A. G. G. M. 2001 Crystallinity versus mass-loss rate in asymptotic giant branch stars. Astron. Astrophys. 369, 132{141. Lewis, R. S., Ming, T., Wacker, J. F., Anders, E. & Steel, E. 1987 Interstellar diamonds in meteorites. Nature 326, 160{162. Lewis, R. S., Amari, S. & Anders, E. 1994 Interstellar grains in meteorites. II. SiC and its noble gases. Geochim. Cosmochim. Acta 58, 471{494. McKee, C. F. 1989 Dust destruction in the interstellar medium. In Interstellar dust (ed. L. J. Allamandola & A. G. G. M. Tielens), pp. 431{443. Dordrecht: Kluwer Academic. Mathis, J. S. 1990 Interstellar dust and extinction. A. Rev. Astron. Astrophys. 28, 37{70. Nittler, L. R. 1997 Presolar oxide grains in meteorites. In The astrophysical implications of the laboratory study of presolar materials (ed. T. J. Bernatowicz & E. Zinner), pp. 59{82. New York: American Institute of Physics. Nuth, J. & Donn, B. 1983 Nucleation theory is not applicable to the condensation of refractory grains in the primitive solar nebula. Lunar Planet. Sci. XIV, 570{571. Routly, P. M. & Spitzer, L. 1952 A comparison of the components in interstellar Na and Ca. Astrophys. J. 115, 227{243. Savage, B. D. & Sembach, K. R. 1996 Interstellar abundances from absorption-line observations with the Hubble Space Telescope. A. Rev. Astron. Astrophys. 34, 279{329. Schutte, W. A. 1999 Ice evolution in the interstellar medium. In Solid interstellar matter: the ISO revolution (ed. L. d’ Hendecourt, C. Joblin & A. Jones), pp. 183{198. Les Ulis: EDP Sciences. Phil. Trans. R. Soc. Lond. A (2001)


1972

A. P. Jones

Speck, A. K., Hofmeister, A. M. & Barlow, M. J. 1999 The SIC problem: astronomical and meteoritic evidence. Astrophys. J. 513, L87{L90. Spitzer Jr, L. 1978 Physical processes in the interstellar medium. Wiley. Spitzer, L. & Fitzpatrick, E. L. 1993 Composition of interstellar clouds in the disk and halo. III. HD 149881. Astrophys. J. 445, 196{210. Taylor, A., Baggaley, W. J. & Steel, D. I. 1996 Discovery of interstellar dust entering the Earth’ s atmosphere. Nature 380, 323{325. Tielens, A. G. G. M. 1983 Stationary ° ows in the circumstellar envelopes of M giants. Astrophys. J. 271, 702{716. Tre® ers, R. & Cohen, M. 1974 High-resolution spectra of cool stars in the 10- and 20-micron regions. Astrophys. J. 188, 545{552. Waters, L. B. F. M. (and 36 others) 1996 Mineralogy of oxygen-rich dust shells. Astron. Astrophys. 315, L361{L364. Whittet, D. C. B. 1992 Dust in the galactic environment. Bristol: Institute of Physics. Woitke, P., Dominik, C. & Sedlmayr, E. 1993 Dust destruction in the transition region between stellar wind and interstellar medium. Astron. Astrophys. 274, 451{464. Woolf, N. J. & Ney, E. P. 1969 Circumstellar infrared emission from cool stars. Astrophys. J. 155, L181{L184.

Discussion A. Witt (The University of Toledo, Toledo, USA). I would like to suggest that the idea of grain shattering and random reassembly in the ISM could be tested with a main spectrometer such as the one available on the dust detector on the Cassini spacecraft. If this is indeed a dominant process, detected single grains should display a mixture of elements not representative of speci c chemical formation conditions in stellar out®ows. A. P. Jones. This is indeed a good idea to test the prediction of interstellar dust homogenization arising from the conclusions of the most recent dust lifetime determinations (Jones et al . 1996). If the comet Halley ®y-by results from the Vega and Giotto dust detectors (see, for example, Schulze et al . 1997) can tell us anything, i.e. because the analysed particles include fossil interstellar grains, then we may already have some indications. The Halley results show that the detected grains always contain an organic carbonaceous component, and that the overall dust composition is chondritic, as would be expected for e¯ cient dust homogenization in the ISM. However, the results also indicate that there are reasonably well-de ned mineral assemblages (similar to the interplanetary dust-particle compositions), and that almost half of the detected particles are dominated by single mineral grains. This would support the case for inef cient dust processing in the ISM. Thus the comet Halley dust results are somewhat inconclusive. However, the comet Halley dust particles may have no relationship to interstellar dust at all! Additional reference Schulze, H., Kissel, J. & Jessberger, E. K. 1997 Chemistry and mineralogy of comet Halley’ s dust. In From stardust to planetesmials (ed. Y. J. Pendleton & A. G. G. M. Tielens). ASP Conference Series, vol. 122, pp. 397{414. San Francisco, CA: Astronomical Society of the Paci¯c.

Phil. Trans. R. Soc. Lond. A (2001)