Clouds, Clumps, Cores&Comets - a Cosmic Chemical Connection?

We discuss the connection between the chemistry of dense interstellar clouds and those characteristics of cometary matter that could be remnants of it. The chemical evolution observed to occur in molecular clouds is summarized and a model for dense core collapse that can plausibly account for the isotopic fractionation of hydrogen, nitrogen, oxygen and carbon measured in primitive solar system materials is presented.


Introduction
Comets have probably retained some material that originated in the molecular cloud from which the Sun formed. 1 Determining how much pristine interstellar material is in comets will help answer many important questions relating to the origin of our Solar System and, as comets are strong candidates for seeding planets with complex organic molecules, understanding the details of the interstellar-comet connection will have important implications for astrobiology. 2 Recent data from the Stardust mission, 3 and groundbased observations, 4,5 indicate that some materials present in cometary dust experienced very high temperatures (∼ 800 K), relative to those typically found in molecular clouds (∼ 10 K). Nevertheless, the organic inventory and isotopic signatures measured for cometary molecules do provide a tantalizing connection with interstellar chemistry.
In this paper we discuss the chemical structure and evolution of interstellar matter prior to its incorporation into protoplanetary disks. We outline a model whereby many of the key cosmogonic markers in cometary matter can be explained as being of interstellar origin. ever, these data only probe the outer (> 50AU) disk region. ALMA will have the sensitivity to detect and image all the dust in the disk down to the 1 AU scale. Through multifrequency observations, ALMA will also be able to measure the change of dust properties within disks, perhaps showing direct evidence for radius-dependent grain growth in the midplane, up to sizes of several mm. ALMA will have the sensitivity to map optically thick lines at a few AU resolution, providing information about the gas content and its chemistry and kinematics down to the planet-forming zones. The stellar masses inferred from kinematics form important direct tests of pre-main-sequence stellar evolution models. By imaging lines with different excitation conditions, maps of the H 2 density distribution will become possible. Together, such ALMA data can provide unbiased surveys of disks in different star-forming regions, down to an equivalent sensitivity of a few Earth masses of dust and gas, and probe the distribution of disk parameters with stellar mass, luminosity, age and environment. 5.3 Disk evolution and gap formation disks in Spitzer surveys show that there may be multiple evolutionary paths from the massive gas-rich disks to the tenuous gas-poor debris disks, involving both grain growth and gap opening. ALMA will be critical to study these transitional objects by imaging the holes or gaps in their dust disks down to a few AU and measure the remaining gas mass through tracers like CO and [C I], some of which may be left inside the holes. Giant planet formation, grain growth and photoevaporation are the three major contending theories for explaining holes in the dust disks but they have different predictions for the gas vs. large dust distribution. Overall, surveys for gaps in disks can provide statistics on the frequency and timescale for planet formation.

Conclusions
ALMA will be vital and unique to answer key questions in star-and planet formation, by resolving the physical processes taking place during the collapse of molecular clouds, imaging the structure of protostars and of protoplanetary disks, and determining the chemical composition of the material from which future solar systems are made. Many of the ALMA source lists will come from unbiased surveys being carried out now, in particular Spitzer, Herschel, near-infrared and single-dish submillimeter surveys. To extract information from ALMA data, however, sophisticated analysis and modeling tools are needed. The community needs to invest now in those tools to ensure that they are ready by the time that ALMA is fully commissioned.
Other major facilities in the timeframe of ALMA operations include the James Webb Space Telescope and ground-based extremely large optical telescopes (ELTs). These facilities will be highly complementary to ALMA, each addressing a different part of the star-and planet formation puzzle. There is no doubt, however, that ALMA will be the key instrument for much of the physics and chemistry associated with star-and planet formation.  Figure 1 illustrates the time-scales associated with protostar-disk mass accumulation (see Ref. 10). Detailed observational studies and chemical modelling have led to a deeper understanding of the chemistry of the evolutionary sequence from pre-collapse (and prestellar) cores, through to Class 0 and I sources, Class II and III objects and their related 'protoplanetary' and 'debris' disks 7,[12][13][14][15][16][17][18][19][20][21][22] Such studies can also shed light on the state of the interstellar material available to disks during the comet-formation epoch (e.g. Ref. 11). As most of the mass accreted up to early in the Class I epoch was consumed by growth of the protosun (Figure 1), the volatile disk material now retained in comets probably accreted during the Class I-Class II evolutionary phases, where the lower mass accretion rates also favour much weaker accretion shock strengths over the disk surface. 23 Thus, the chemistry in molecular clouds containing protostars at the Class 0/I boundary, and later, is that which may be best associated with the most pristine cometary matter.

Observations of cloud chemical evolution
The Barnard 5 molecular cloud in Perseus (B5) is a region in which lowmass protostars are forming 25, 26 . Of the four identified protostars, 27 B5 IRS 1 resides in the dense central core and is classified as a Class I source that may just be at the Class 0/I transition 28 . Interferometric maps of IRS1 confirm the presence of a circumstellar disk of ∼ 0.16 − 0.27 solar masses 29 . Mapping of clouds like B5 in various molecular lines 8,14,30 allows us to understand the physical and chemical evolution that occurs in cloud material prior to incorporation into a circumstellar disk.
High resolution C 18 O mapping of the B5 core ( Figure 2) shows a clumpy morphology for both the extended ambient gas and the dense central core. 24 Maps of the B5 core ( Figure 3) demonstrate that other molecular distributions are also very clumpy and chemically differentiated. 31 This chemical anticorrelation between the emission peaks of many molecules is evident in other high-resolution maps of the B5 core (e.g., CH 3 OH, H 13 CO + ), 24 and other dark clouds 30 . In particular, Figure 3 shows that emission from various carbon-chain species is anti-correlated with that of ammonia, as observed in TMC-1 and several other dense cores 8,32, 33 Chemical models indicate that carbon-chain species such as C 2 S, C 4 H, and the cyanopolyynes (HC 2n+1 N, n =1-5) are 'early time' species -they reach their peak abundances in ∼ 10 5 years -whereas 'late time' species, including N 2 H + and NH 3 , peak much later 34 . This has led to the idea that observed spatial compositional gradients (e.g. involving the carbon-chains and ammonia) are due to differing chemical ages within individual clouds. 8,33,[35][36][37] Thus, the two C-chain peaks (CP1&2), near δ ∼ +150 , are probably at an earlier stage of evolution relative to the material surrounding IRS1, as one would expect.
In B5 we can observe the chemistry at three distinct phases in the star formation sequence: the clumpy ambient medium, dense prestellar clumps and a core containing a Class I protostar. However, in B5 we do not appear to see the chemistry at two intervening phases.
Maps of starless globules in lines of carbon chain molecules show that they can have a reasonably smooth emission. 39,40 An apparently common subsequent step from such dense prestellar clumps is the formation of a depletion core where observations indicate that almost all the heavy molecules, particularly CO, become depleted into the solid state, [41][42][43] and strong emission from deuterated molecules becomes evident. 15 Hence, all the elemental C, O, N and S can be available for grain-surface chemistry. For example, Figure 4 shows the dense core of the starless dark globule TMC-1C mapped in its HC 3 N emission. 30 This morphology can be explained as resulting from the increasingly efficient sticking of HC 3 N molecules on dust as the density increases, the end-point being a core with all the HC 3 N molecules frozen out in the center. Comparison with an N 2 H + map of this source 38 indicates that the N 2 H + emission actually peaks at the central 'hole' in the HC 3 N emission (see Figure 4). This is an example of selective depletion where the enhancement of N-bearing species (such as N 2 H + and N 2 D + ) in CO-poor regions is thought to be related to the relative depletion of CO and N 2 [44][45][46] Although depletion cores are very difficult to study both in molecular lines (except perhaps for N 2 H + , N 2 D + , H 2 D + and HD + 2 ) and also in solid state absorption, the chemistry occurring in their ices can be revealed later in the star formation sequence. Many cores at the prestellar stage show evidence for gravitational infall 14 and the next step should be formation of a protostar in the deeply embedded Class 0 phase a . During the Class 0 phase, the protostar and disk accrete mass at a high rate; the consequent increase in luminosity and the present of outflows and shock waves means that it heats its immediate environment and a hot corino is formed. 15 Observations of hot corinos show that, like their massive counterparts, the hot molecular cores 2 they are very rich in complex organic molecules and exhibit extremely pronounced deuterium fractionation characteristics. 15 This is believed to be due to the evaporation and/or sputtering of molecular ices into the gas. 47 Well-studied 'hot corinos' are IRAS 16293-2422, NGC1333-IRAS4A, NGC1333-IRAS4B and NGC1333-IRAS2A. [48][49][50][51][52][53][54] Hence, the protostellar disk of a source such as B5 IRS 1 may, in principle, accrete material with chemical characteristics representative of a prestellar core, a depletion core, the Class 0 hot corino, and the remnant envelope at the Class I phase. The actual relative proportion of each chemistry that could end up unaltered in comets is unknown. Comparison with the measured volatile composition of comets can help us better understand this putative connection and to develop and constrain chemical models of the passage of interstellar material into disks.

Measured characteristics of cometary material
The inventory of molecules observed in comets is, in general, a subset of the ∼ 150 molecules detected in the ISM. Table 1 lists species detected in interstellar and circumstellar environments, with those also detected in comets highlighted. The extensive overlap between cometary and interstellar molecules has long suggested a connection between the two 55 . A thorough description of all cometary characteristics and their possible links with the ISM is beyond the scope of this paper, and has been discussed in detail elsewhere 1,2,56,57 . In the following discussion, we focus on isotopic ratios in specific molecules, and their potential role as tracers of the cometary-ISM connection.

Hydrogen and deuterium
Cometary D/H ratios are not significantly altered in the coma, 58 and are not thought to be altered during sublimation, or during their long deepa A candidate Class 0 source has recently been identified in B5 (J.V. Buckle, priv. comm.) that would be coincident with a putative depletion core identified in CH 3 OH maps 24  59 and the presence of highly-fractionated material in comets would be strong evidence for the preservation of interstellar material. D/H ratios have been observed in a handful of molecules in a total of five different comets, and are summarized in Table 2. In contrast, large numbers of deuterated species have been observed in the ISM, including several multiply-deuterated species 15 . Deuterium enrichments have also been seen in cometary dust grains returned by the Stardust mission, 60 and in interplanetary dust particles thought to originate from comets 61 . This D-enrichment is also thought to originate in the ISM, but as it resides in complex organic refractory material it is not possible to trace a direct link to specific interstellar molecules. With the exception of the large HDCO/H 2 CO measured by Kuan et al., 63 cometary D/H ratios are somewhat smaller than those measured in dark interstellar clouds and low-mass star-forming regions (hot corinos), 15,70 but are in line with those found in massive hot cores. 71 The HDO and DCN fractionation in the ISM is understood to result from ionmolecule isotope-exchange reactions at 10 K, whereas the cometary values are more in line with a somewhat warmer temperature of ≈ 30 K 72 .

Nitrogen
15 N/ 14 N ratios have only been observed in two cometary molecules: hydrogen cyanide and the related cyanogen radical. HC 15 N/HC 14 N ratios of 0.003 were seen in Hale-Bopp, 73,74 roughly equal to the terrestrial nitrogen isotope ratio, and slightly larger than the elemental protosolar ratio. 75 C 15 N/C 14 N ratios have been measured in eight comets, including Hale-Bopp, with an apparently constant value of 0.007 in every comet. 76 The discrepancy between the isotope ratios in HCN and CN shows that HCN cannot be the sole parent of CN, and demonstrates the importance of isotope ratios in testing putative 'parent-daughter' relationships between coma species. 77 The CN is thought to originate from the break-up of 15 N-enriched organic refractory (CHON) material in the coma. 76 Interestingly, 15 N-rich organics have been discovered in Stardust samples 60,78 , and also in meteorites and IDPs 79,80 .
Cometary (and meteoritic) 15 N fractionation is often assumed to origi-nate in the ISM via low-temperature chemistry, in analogy with enhanced D/H ratios, as there are no known nebular processes that can produce such effects. However, there have been very few observational studies of nitrogen fractionation in the ISM. Theoretical calculations predicted only small 15 N-enhancements of ∼ 25 per cent in a 'typical' dark interstellar cloud. 81 More recent studies indicate that, if CO is depleted from the gas-phase, much larger enhancements are possible 82,83 . In this scenario, 15 N is preferentially incorporated into ammonia ice, and bulk ice 15 N-enhancements of 80 per cent are possible. Following the isotopic ratios in individual monolayers (ML) as they accrete sequentially, these models also show that most of the 15 N-enhancement is due to highly-fractionated uppermost ML which accrete at late times. 84 These layers are also the most likely to be altered by subsequent processing of the ices by UV and cosmic ray irradiation into more refractory material, although the details of this processing are unclear. If this mechanism is in fact the origin of the 15 N anomalies in primitive solar system material, this indicates that at least some cometary organics originally formed in cold (10 K) interstellar gas.

Oxygen and carbon
Gas-phase 13 C/ 12 C ratios have been measured in cometary C 2 , CN, and HCN. To date, the only 18 O-bearing isotopologue detected in a comet is H 18 2 O. For both elements, the isotopic ratios are all 'normal' (i.e., Solar), 85 although the error bars in each case are sufficiently large that small deviations from the Solar ratio would not have been identified. More accurate laboratory analysis of the Stardust samples did in fact reveal the presence of some cometary grains with small 18 O and 17 O depletions, of the order forty parts per mil. 78 This material is isotopically identical to many particles found in meteorites and IDPs, 86 and presumably shares a common origin. Similarly small 13 C-anomalies are present in many Stardust particles, also in line with those observed in other primitive solar system material. 87 Some of these phases are apparently 'pre-solar' grains preserved intact since their formation in the outflows of red giant stars or supernovae (e.g., Ref. 88), but in other cases the fractionation may be due to chemistry in the ISM.
Low temperature interstellar fractionation of oxygen and carbon via ion-molecule reactions was modeled by Langer et al., 89 who showed that relatively small effects are to be expected. For carbon, 13 C becomes preferentially incorporated into CO, whereas other C-bearing species become 13 C-deficient. Subsequent freeze-out and surface chemistry will preserve these differences in specific molecules. 90 For example, formaldehyde and methanol are thought to be formed via hydrogenation of CO, so one would expect these species to have a slight 13 C excess. Ion-molecule reactions are apparently unable to account for the observed oxygen isotope anomalies, which instead have been interpreted as arising from self-shielding of the more abundant 12 C 16 O molecule compared to its isotopologues. 86,91 This shielding may have occurred at the surface of the interstellar cloud from which the solar system was formed, 92 or at the surface of the protosolar nebula. 86 In either case, the fractionation occurred in UV-irradiated gas, which models of photon-dominated regions (PDRs) predict to have temperatures of ∼ 30-70 K 93 . Thus, the presence of oxygen isotope anomalies in cometary material is evidence for the formation of some cometary matter in warm gas.

From the pre-Solar core to comets: tracing the chemical heritage of cometary material
In the previous section we described the isotopic ratios seen in cometary material, and the possible relations between cometary and interstellar material. In this section, we briefly describe a model of core collapse and isotope fractionation that can plausibly account for the observational data. Figure 5 illustrates the chemical structure of a pre-stellar core, immediately prior to the onset of collapse and beginning of the Class 0 phase. The core contains strong chemical gradients, and we expect these gradients will result in related chemical differentiation (both spatial and temporal) during the subsequent gravitational collapse. In the coldest, most dense, regions toward the center all heavy elements are frozen on to grains. Here, D + 3 dominates the gas-phase, 94 and accretion of gas with high atomic D/H ratios leads to ices containing highlydeuterated molecules 95 . However, once the core begins to collapse, most of the material in this centermost region will end up in the newborn protostar. Hence, we may expect the most heavily-deuterated molecules detected in the ISM to generally be only a minor constituent of comets. Surrounding this region, a shell of CO-depleted gas is the site of efficient 15 N-fractionation, which produces 15 N-enriched ammonia ice after about a few times 10 5 yr 82,83 -this time-scale is of sufficient duration that, for the collapse of a 1 M core, this shell will collapse during early Class I phase.
Farther out, CO remains in the gas phase and the chemical composition can be understood from 'standard' models of dark cloud chemistry. This region will however contain a temperature gradient, with those shells immediately behind the PDR being warmed by infrared reradiation of the  UV absorbed at the surface, and so having gas and dust temperatures of ≈ 25-35 K. As these regions are almost the last to accrete, during the Class I/ Class II epochs, they are a major source of material in the disk at the time when comets were formed and will be characterised by reduced deuterium fractionation and molecular ortho:para spin ratios characteristic of these temperatures 85 The interstellar UV field to which the surface of the core is exposed may actually be much stronger than the mean interstellar flux if the core is in close proximity to newborn massive O and B stars, 96 and so warm gas and dust may exist to greater depths. In any case, self-shielding of CO in these surface layers leads to mass-independent fractionation of atomic 13 C, 17 O, and 18 O. 86,92 The UV flux also results in increased gas and dust temperatures interior to the surface PDR, with T gas ∼ 70-150 K and T dust ∼ 30-35 K 93 . Previous models of CO self-shielding in clouds and disks have relied on rapid transport (e.g. turbulent diffusion) from the PDR to cold regions, where atom sticking and hydrogenation on dust fixes the oxygen isotopic anomalies in water 86,92,97 . At the elevated dust temperatures in our model ( Figure 5), atoms do not stick to the surfaces of dust grains, 45 so surface chemistry is unable to convert the isotope fractionation in the atoms into fractionation of more refractory species. However, as the gravitational col- lapse proceeds and the central luminosity increases, neutral chemistry in the hot infalling gas provides an alternative gas phase route to O → H 2 O conversion.
Modelling the chemistry in the collapsing protostellar cocoon requires a physical model in order to derive density, temperature, and infall velocity profiles in the envelope as a function of time. We have adapted the dynamical-chemical model of Rodgers and Charnley 21 , based on the 'insideout' collapse of Shu,98,99 to include a static PDR at the outer edge of the envelope. This work will be reported in detail elsewhere 100 and here we present some preliminary results to demonstrate how interstellar fractionation patterns in 13 C, 17 O, and 18 O may be transported from the distant outer envelope to the central protostar and disk. Figure 6 shows the density and temperature profiles used in this model, with conditions appropriate for a 1 M star accreting material at a rate of 10 −5 M yr −1 . Clearly, at late times when most of the mass is in the central star, the luminosity is large enough to significantly heat material in the innermost regions of the collapsing envelope. However, by this point the infall velocities are sufficiently large to ensure that the dynamical time-scale for material to pass through this hot zone is much less than the typical time-scales for chemical reactions to alter the composition of the gas 21 . Important exceptions are the reactions of O and OH with H 2 which rapidly convert atomic oxygen into water. This is illustrated in Fig. 7, which shows the C and O chemistry in material initially located at a radius of  Fig. 7. Chemistry in a collapsing protostellar envelope H 2 O near the protostar. Thus, the material that rains down onto the disk, during the epoch that cometesimals and planetesimals are forming, contains isotopically-enhanced water derived from the isotopic anomalies in neutral oxygen atoms generated by CO self-shielding in the surface PDR.

Conclusions
We have reviewed the putative contribution of interstellar chemistry to the volatile composition of comets. In doing so, we have purposely neglected discussion of the possible contribution of nebular chemistry as a source of cometary volatiles -either from the warm innner nebula or from an essential continuation of interstellar chemistry in the cold outer disk; these have been reviewed elsewhere. 11,101 Comets could contain material from several different stages of molecular cloud evolution and we have illustrated the related chemistry with some recent observations. Many of the molecules detected in comets, including several organics, are either directly observed in interstellar ices (e.g. methanol, formic acid) or are believed to form on interstellar grains (e.g. formamide). 2 A direct comparison of the organic inventories is however complicated by the fact that, in comets, additional sources of simple molecules appear to contribute to their coma abundances (e.g. CO, CS, CN, formaldehyde). These so-called extended sources are believed to be due to the thermal or photolytic break-up of large organic macromolecules 102 and have no readily identifiable parallel in interstellar chemistry. Interstellar isotopic fractionation in dense gas with temperatures in the range ∼ 10-35 K could account for the currently known, albeit meagre, fractionation ratios measured in comets.
We have presented the outline of a model that may account for all these characteristics as arising in the prestellar core from which the Sun formed. As observed in many galactic sources, such cores can develop very strong chemical gradients. Gravitational collapse of such a core may deliver chemically distinct regions (i.e. mass shells) onto the central disk during the Class 0/I epoch, and later, to the extent that all the 'interstellar' characteristics of comets are delivered during this infall.
Finally, future observations of more bright comets, especially shortperiod ones, as well as of protoplanetary disks, are necessary (e.g. 103). The advent of the Atacama Large Millimetre Array promises a great advances in further investigating the ISM-comet connection. 9,104