摘要:

Faraday Discuss. 1998 109 1»13 Introductory Lecture Frontiers of astrochemistry David A. Williams Department of Physics and Astronomy University College L ondon Gower Street L ondon UK W C1E 6BT The areas of astronomy in which astrochemistry plays a fundamental role are brie—y reviewed. It is argued that the astrochemical topic of which we have poorest understanding is the gas»dust interaction. It is shown that this interaction is important in many areas of astronomy. Some recent work in the general area of the gas»dust interaction is brie—y reviewed. 1 1 Introduction The title of this introductory lecture is perhaps rather pretentious ; but what I intend to do is»from a personal viewpoint»identify where the action is taking place in our subject at the present time.I shall try to give a sense of perspective highlight some areas of activity and pose some questions. The quality and variety of the papers which form the starting point of our Discussions ensure that many of these questions are addressed during the meeting. Our subject is interdisciplinary ; it involves astronomical observations astrophysical modelling and microscopic processes. Advances are occurring in all three areas. In astronomy advances in observational capability have opened up new wavebands (as for example with the Infrared Space Observatory) or provided (as the Hubble Space Telescope has done) greater sensitivity and angular resolution. In modelling much more realistic descriptions can be made thanks to advances in computing power.In the provision of fundamental data new approaches»both in the laboratory and on the computer»render previously intractable problems accessible to study. Therefore a huge variety of opportunities opens up in front of us. There are literally many thousands of processes that could be investigated to bene–t astrochemistry. I suggest to you that our response to this situation should be to remember that our subject sprang from the demands of astronomy and that we should be guided in what we do by what seems likely to yield the greatest scienti–c»i.e. astronomical»return. So my remarks will be framed around astronomical topics and I aim to show that very many current topics in modern astronomy can be and must be addressed by astrochemistry. Today astrochemistry has a marvellous opportunity.In choosing to be guided by the astronomical demands we can ensure that astrochemistry has an important and continuing role for astronomy in general. My plan is therefore as follows. In Section 2 I make some brief remarks about what seem to me to be the most active areas of astronomy in which astrochemistry plays a role. We shall see that this covers many of the important areas of modern astronomy. I shall note in particular the microscopic processes involved in the astrochemistry of each area ; and will argue in Section 3 that our greatest weakness and poorest understanding is in the general area of gas»dust interaction. I then discuss some recent eÜorts to remedy this weakness (Section 4) and make some general conclusions in Section 5.2 Frontiers of astrochemistry Table 1 Processes in astrochemistry timescale/yr mechanism process cr ionization molecular radiation gravity ion»neutral drift gas»grain collision surface chemistry ion»mol chemistry cooling collapse ambipolar diÜusion freeze-out on dust desorption B3]105 (10~17/f) B3]105 (at ca. 10 K) B108/n1@2 B4]105 X(i)/10~8 B3]109/n B3]105/n(H) 2) f is the cosmic ray (cr) ionization rate of hydrogen s~1; n\n(H) ]2n(H cm~3 is the total density of hydrogen nuclei ; n(X) is the number density of species X; and X(i)\n(ions)/n. As a –nal note to this introduction it is perhaps worthwhile to summarize the main processes and mechanisms that we have to deal with in the application of astrochemistry (see Table 1).In a real sense what we have to do is to choose among these and other processes and to identify which of them is signi–cant in the regimes we are describing. It is often the case that these regimes are time-dependent and dynamically evolving. As we shall see in Section 2 while the chemistry in many astronomical scenarios is driven by cosmic ray ionization in one situation (the early Universe) the ionization driving the chemistry is simply present because recombination in the expanding Universe is incomplete. In some other situations the initiating ionization is caused by stellar UV radiation or X-radiation from shocks. The chemistry is usually greatly enhanced by neutral» neutral exchanges. The molecular feedstock for the chemistry is almost always molecular hydrogen.The most efficient way of forming H in the temperatures and pressures occurring in many astronomical regimes appears to be in surface reactions on dust grains ; other routes 2 apply if grains are absent. Other species too may form in surface processes. In some circumstances the surfaces of cold dust grains become coated with ice as various species freeze-out. Solid state chemistry initiated by UV X-radiation or fast particles may form new and larger molecular species. The ices may be either thermally or non-thermally desorbed and contribute to the gas phase abundances. The chemistry has several important roles to play. Most obviously it provides molecules that are tracers of the densest coolest and most massive associations of nonstellar baryonic matter in the Universe.The largest entities that are known in the Galaxy for example are the giant molecular clouds GMCs which contain up to about a million solar masses of hydrogen. These are traced in radio emission lines of CO. Similar masses of gas exist in other galaxies. These and other emissions are also important cooling mechanisms for gas at low temperature and allow gravitational collapse in the formation of protogalaxies or protostars to continue by the conversion of gravitational potential energy through thermal energy to escaping radiation. Equally important in some circumstances can be the magnetic support which retards collapse. The magnetic –eld lines act on the largely neutral gas through the ionized component which is often determined by the chemical network.Dissociative recombination of molecular ions with electrons is usually rapid while radiative recombination of atomic ions is relatively slow. Hence the balance of atomic to molecular ions is the factor that controls the ionization fraction and the extent of magnetic support. 2 Areas of astrochemistry The areas of astrochemistry encompass almost all the main activities in astronomy. In most locations in the Universe that are cooler than sunspots it appears that molecules 3 D. A. W illiams are formed and usually have a signi–cant role to play. It is the task of astrochemistry to identify the origin of these molecules and to describe those roles. I will try to present these areas in some kind of logical sequence from large scale to small scale.2.1 The early Universe This is the era that gave rise to the main observable phenomena of the Universe on the larger scale. The formation of galaxies occurred around the time represented by redshift zB10 when the Universe was less than one tenth of its present age. However the era which we are discussing began much earlier around zB1300 (in the standard Big Bang Nucleosynthesis model) when matter and radiation became decoupled and from which the cosmic microwave background radiation originates (the age of the Universe then being ca. 0.3 My; see e.g. Lepp and Stancil1). This was the beginning of chemistry the Universe underwent a phase transition from being entirely ionized to being almost completely neutral.The chemistry in this mainly neutral gas was rudimentary since the elements available were H and 4He with mere traces of D 3He and 7Li. (The relative abundances of the elements H 4He D and Li were approximately 1 0.1 10~5 and 10~10 respectively.) Radiative associations and attachments initiated a short network of reactions that produced H with a fractional abundance of ca. 10~6. Gravitational collapse reheating of the gas by UV radiation from Population III stars (a postulated –rst 2 generation of stars of high mass) and by shocks perturbed this initial state and structure developed. Cooling by H through emission in vibration»rotation transitions was the only signi–cant means of losing gravitational potential energy and this cooling con- 2 trolled the nature of the objects formed at this phase»thought to be objects of the size of globular clusters (ca.105 solar masses M_). Other trace molecules such as HD or LiH might have played a cooling role once the H cooling lines became optically thick. Shocks in the intergalactic medium caused compression which drove the chemistry 2 faster and the H fraction was raised from 10~6 to ca. 10~3 making the cooling very rapid down to about 100 K. Additional ionization by quasars or in shocks enhanced H 2 formation the cooling and collapse. 2 Therefore the evolution of structure in the early Universe depended on H formation which is initiated by radiative processes. Chemistry has an ancient pedigree ! These 2 processes can be enhanced in radiatively stimulated situations.The signi–cance of stimulated processes has been assessed by Stancil and Dalgarno2 and shown to be small. Given the restrictions on the chemistry placed by the limited elemental abundances from the standard Big Bang Nucleosynthesis model the main uncertainties in studies of the early Universe appear to be associated with the physics rather than the chemistry. The pre-eminent role of H in determining the structure in the Universe however seems already established. 2 2.2 Starburst galaxies and active galactic nuclei Many starburst galaxies and active galactic nuclei (AGNs) have been studied in molecular line emissions (cf. Booth and Aalto3 and Lepp and Tineç 4). The current record for the most distant molecular object is the quasar BR1202-0725 with redshift z\4.69 (corresponding to an age of 109 yr post-Big Bang or 6% of the age of the Universe) in which CO and dust have been detected.5,6 The Hubble Space Telescope deep –eld studies have revealed that massive star formation in galaxies peaked at z\1 and is as large as at present back to z\4 (ref.7). Studies of c-ray bursts indicate that they are associated with massive star formation and originate in objects around z\1 and that at least one burst GRB 970828 is embedded in a dense molecular cloud.8 Evidently chemistry was proceeding efficiently even at this early epoch with some similarity to the chemistry in the Galaxy. There are however important diÜerences : 4 Frontiers of astrochemistry there were lower metallicities lower dust fractions diÜerent relative abundances intense UV and X-ray –elds»all of which had signi–cant consequences for chemistry and for dust formation.Black9 discusses the excitation of molecules in radiation –elds appropriate to AGNs. The diÜerences from the chemistry in our Galaxy make the CO/H ratio (needed to determine gaseous masses in AGNs) extremely uncertain. 2 The intense X-rays present in AGNs may promote a chemistry diÜerent to that driven by cosmic rays and by stellar UV such as generally occurs in the Galaxy. Hard X-rays may in—uence structures galaxy-wide and hard and soft X-rays may generate multiply charged ions opening up pathways that are not generally available in cosmic ray driven chemistry (cf. Lepp and Tineç 4). In X-ray dominated regions (XDRs) a UV radiation –eld is generated by the X-ray excitation of atoms and molecules and this UV –eld is ultimately transmitted into IR emission via molecules and dust.One particularly interesting diÜerence in chemistry between XDRs and UV photon dominated regions PDRs near hot stars concerns H pumping and destruction. Destruction of H is much 2 more likely in XDRs so that any vibrational emission in such locations is more likely to 2 be from the H formation process than pumping.4 The ionization of H by X-rays tends 2 3 ` which is now con–rmed to be widespread in the Galaxy 2 10 and pre- to promote H sumably may have a higher abundance in AGNs. 2.3 Star forming regions in the Galaxy Much of the star formation in the Galaxy occurs in giant molecular clouds (GMCs) the most massive non-stellar baryonic objects in the Galaxy (4106 M_).GMCs are collecnH of hydrogen nuclei is n tions of clumps (in which the number density H4103 cm~3) of gas embedded in a more tenuous background of neutral gas. Within a clump clusters (n of cores HB105 cm~3) may exist ; and from the cores protostars may evolve. Low mass stars (with masses 44 M_) are thought to form as a consequence of a gradual weakening of magnetic support in clumps that were originally magnetically supported against gravitational collapse. The formation of high mass stars however may be triggered by an abrupt increase of external pressure on a clump supported by magnetic –elds so that the clump becomes suddenly unstable. In both situations chemistry plays an important role in providing molecules that trace the gas and cool it during the collapse.In low mass star formation chemistry also controls the level of ionization and through it the rate of ambipolar diÜusion and the weakening of magnetic support. In clumps the main tracer and coolant is CO produced in a chemistry to which UV and cosmic rays both contribute (cf. van Dishoeck11). The ionization fraction is mainly atomic C` S`. In cores the UV is excluded and the chemistry is cosmic ray driven. The CO is optically thick and the main tracers are diatomic and polyatomic molecules such as CS C2S NH3 H2CO and HC At higher densities the gas»dust interaction is 3N. chemistry through loss of species especially coolants tracers and low ionization potenmore important for times longer than a freeze-out timescale tfo and may aÜect the tial species and may cause non-linear eÜects in the chemistry.For example in cores the important reactions aÜecting CO are summarized here HCO` »»»’ e CO He` H2O CO »»»’ C` »»»’ »»»’ H2 hydrocarbons »»»’ HC3N The normal route is the upper channel which essentially restores the CO that was lost by reaction with He`. However if the H2O is reduced in abundance by freeze-out onto dust then the slower lower channel operates and hydrocarbons may form. Hence molecules such as HC may appear as late-time species.12 A steep transition between high fractional ionization and low fractional ionization seems to occur at visual extinctions of 3N AvB2.5 magnitudes. This suggests a scenario for low mass star formation involving 5 D.A. W illiams collapse along –eld lines as magnetic turbulence is damped followed by collapse perpendicular to –eld lines as ambipolar diÜusion weakens the –eld strength.13 The observational determination of the fractional ionization is discussed by Plume et al.,14 and is shown to be consistent with models of chemistry driven by cosmic ray ionization. The formation of stars of all types seems to be accompanied by winds that are sometimes collimated into jets impacting with high velocity on circumstellar and interstellar gas; the jet trail can be as much as 10 light years long. However the formation of massive stars is accompanied by extremely powerful winds that rapidly disrupt the stellar environment. Transient debris left over from the massive star formation process is observed in the form of hot cores described in detail by Millar and Hatchell.15 It is evident that at the high number densities inferred for these objects (ca.107 cm~3) the gas»dust interaction dominates; the freeze-out timescale t is only about 102 yr. Therefore every species that can be frozen out on dust has been frozen out to be released fo (through thermal evaporation) when star formation occurs in the vicinity and warms the dust. The astrochemical problem is to determine the extent to which the original chemistry in the gas has been modi–ed by solid-state chemistry and by evaporation. The dynamical evaporation and thermal timescales need to be explored relative to the chemical timescale. As van Dishoeck has shown11 the ISO observations of solid state material may help to interpret the eÜects of freeze-out processing and evaporation.2.4 Circumstellar matter There are many types of situations under the general title of circumstellar matter. 2.4.1 Circumstellar envelopes. Cool molecular envelopes around stars have been very intensively studied. The interplay of observations modelling and microscopic processes has produced some of the best de–ned descriptions most reliable models and useful test-beds of astrochemical ideas (cf. Millar16). The basic picture is of parent molecules being injected from the stellar atmosphere into an out—owing envelope. Beyond a dust-forming zone the out—ow is penetrated by starlight and the photoionization of parent molecules (e.g.C2H2) gives rise to reactions forming transient daughter products. Both parents and daughter species can be identi–ed from observations (cf. Olofsson17). Parents are concentrated around the stellar source of the out—ow; daughters appear as concentric spherical shells. Such a geometrical association is clearly seen in the observation by Lindqvist et al.18 of parent HCN and daughter CN around the star CIT6. The richest chemistries are generally found in C-rich envelopes and the reaction networks that lead to the largest carbon molecules are still a matter for debate. Ion» molecule schemes apparently cannot supply the larger species because the reaction sequences are terminated by dissociative recombination rather than by chain growth. Current eÜorts on carbon chain growth are directed more toward neutral schemes (and relevant reactions are discussed by Chastaing et al.,19 Thaddeus et al.,20 and Kaiser et al.,21).Some discrepancies between observations and models remain and may require a contribution from carbon grain erosion.22 * ). O-rich circumstellar environments have been shown to have a fairly rich carbon chemistry in addition to the expected relatively simple oxygen chemistry. The source of the free carbon is unclear. It is possible that this source is CH4 formed as a result of surface reactions on dust.23 It seems clear that the CH (and some other species) in the 4 well-studied source IRC]10216 is not a parent,24 but appears further out in the out—ow (at radiusB125 R The outstanding problem for both C-rich and O-rich envelopes is the formation of dust.Gail and Sedlmayr15 discuss the formation of dust in O-rich envelopes and make the interesting proposal that»while SiO cannot nucleate in the conditions observed to give silicates»TiO clusters formed in homogeneous nucleation may be the seed nuclei 2 6 Frontiers of astrochemistry for the formation of silicates. The formation of C-rich dust in C-rich envelopes is equally important and has been much discussed (cf. CherchneÜ26). CherchneÜ et al.27,28 have proposed that chemistry occurring in cyclic gas dynamics deep in the envelope can lead to polycyclic aromatic hydrocarbon (PAH) formation. PAH clusters may then act as nucleation centres for carbon cluster growth. 2.4.2 Planetary nebulae.A planetary nebulae is a transient phenomenon occurring when the circumstellar envelope detaches from the star which itself contracts and brightens. Planetary nebulae oÜer a remarkable range of environments in which molecules may be found and for which they act as probes especially of high radiation intensity chemistry [see ref. 2 and 8(a)]. Howe and Williams29 have reviewed the chemistry occurring in these environments. In general terms the range of environments can be described in the interacting stellar winds model. The molecular out—ow (described in Section 2.4.1) becomes detached from the star which contracts brightens and develops a fast wind. The fast wind is arrested by a shock and the shocked wind expands supersonically into the slower molecular wind.The starœs radiation ionizes the zones created in this wind interaction. In the ionized shocked slow wind molecules such as H2 CO and CO` may occur. The unshocked slow wind may contain H HeH` OH and CH` may arise. Beyond this zone in a neutral shocked slow wind H2 2 CO and the photo-products of any molecules present in the slow wind now subject to the intense radiation of the central star. Dense small globules of molecular gas can also survive for a time within the ionized zone. 2.4.3 Novae and supernovae. Ejecta from novae and supernovae provide transient phenomena in which chemistry is important in providing tracers of conditions. Some attention has been given to the formation of dust in novae ejecta expected to be Crich. 30 Conditions in these ejecta are diÜerent from those in C-rich CSEs the gas is much denser with much higher radiation intensity and the impact of ionization fronts in the material limits the time available for dust growth.The chemistry in supernova ejecta31 is quite diÜerent from that in any other astrochemical environment. The material is chemically strati–ed and the extent of mixing between layers is unclear. It is very dense and hot and may be very H-poor; it is ionized through the decay of radioactive cobalt. Molecules CO and SiO have been detected in the ejecta of SN1987A in the Large Magellanic Cloud and the formation routes to these molecules is simply the direct radiative association of neutral atoms. The extent to which stimulated radiative association modi–es this picture is discussed by Stancil and Dalgarno,2 but appears to be small.2.4.4 Circumstellar disks. The chemistry in circumstellar disks is complex.32 In addition to cosmic ray driven chemistry freeze-out onto dust forming ice mantles and subsequent thermal desorption processes the chemical composition in the disk may be greatly aÜected by bulk motions and mass accretion towards the central star. This is a relatively new area for astrochemistry but provides an important interstellar/ interplanetary link. It also provides an opportunity to identify the sites of origin of comets in the Solar System. It seems inevitable that this will become an area of growing interest in astrochemistry. 2.5 Interstellar clouds 2.5.1 DiÜuse and dark clouds. The chemistry of diÜuse interstellar clouds was the starting point of astrochemistry.Both diÜuse and quiescent dark clouds have been well studied through observations and modelling and are reasonably well understood.33 It is established that the chemistry is activated by starlight and by cosmic rays and includes ion»molecule reactions neutral reactions surface chemistry adsorption onto and desorption from dust grains. The chemistry occurs mostly at low temperature (10»100 K), 7 D. A. W illiams though shocks may raise the temperature transiently to 4103 K with longer-lasting chemical than thermal signatures. The initial reactant in the ion»molecule chemistry has been postulated for many years to be H3 ` H generated by cosmic ray ionization of 2 . The detection of this ion in interstellar clouds10 con–rms»to the universal relief of astrochemists»the extensive edi–ce that has been erected on this postulate.The remaining problems in understanding the chemistry of interstellar clouds may lie in a failure to postulate the correct physical conditions rather than a serious fault in the chemistry. For example the pathological case CH` probably has its origin in a dynamical situation not yet correctly described.33 2 The origin of moderately sized molecular species (say 55 atoms) is also less well understood; the route of ion»neutral radiative association followed by dissociative recombination seems not to be successful in accounting for relative abundances and surface processes are now routinely invoked (cf. Hasegawa and Herbst34).Surface reactions are necessarily invoked for H formation and now also apparently for hydride 2 formation.35,36 In dark clouds there are problems to do with several key elements. The oxygen budget remains uncertain because of the difficulty of observing H2O O and the likely major species.32 For carbon the detection of relatively high fractions of neutral and ionized atomic carbon may imply signi–cant penetration of stellar UV into clouds requiring them to be highly fragmented. On the other hand this free carbon may be the consequence of star formation and the impact of stellar jets. Reactions of carbon atoms are therefore highly important in the context of dark cloud chemistry (see ref. 19»21). 2.5.2 Large interstellar molecules. The extent to which large molecules (410 atoms) may be present in interstellar clouds remains unclear.Such molecules may be the carriers of the diÜuse interstellar bands (DIBs) and the unidenti–ed infrared bands (UIBs). The existence of the DIBs has been known for much of this century and this unassigned spectrum now comprises several hundred features.37 The most signi–cant observational advances recently have been the high resolution studies by Hibbins et al.38 and the DIB mapping (in emission and absorption) of the Red Rectangle (Miles et al.39). Both these observational results support the view that large molecules are the carriers of at least some of the DIBs. The UIBs have been commonly attributed to polycyclic aromatic hydrocarbon molecules (PAHs). However in spite of valiant eÜorts no speci–c assignments have yet been made.Some astronomers are losing enthusiasm for the PAH hypothesis and other candidates have also been proposed (cf. Guillois et al.,40 see also Thaddeus et al.20). Sorokin et al.41 and Kirkwood et al.42 address speci–c assignments for the DIBs. A large eÜort is currently being expended on this longstanding and intractable astronomical problem. The astronomical return»or potential return»on this eÜort is at present unclear. If a large amount of carbon is presently locked in large molecules then that could have an important consequence for the carbon budget in the interstellar medium. Abundant large molecules may also aÜect the ion balance in molecular clouds. 2.6 Cometary chemistry and the interplanetary medium Studies of recent comets have indicated that comets have a rich chemistry with a close similarity to interstellar ices.43 Models of cometary chemistry are based on the assumption that parent molecules are injected into space by sublimation from cometary ices and that solar radiation ionizes the parents so that daughter species are created in the ensuing gas-phase ion»molecule chemistry (cf.Irvine et al.,44). The requirement for distributed sources seems to imply the ejection of solids from the cometary nucleus and subsequently sublimation pyrolysis or sputtering. Current questions in cometary chemistry are concerned with the extent to which comets have been processed and the information that comets can provide about the formation of the Solar System.Owen and environment early Universe AGNs star-forming regions circumstellar matter interstellar clouds comets a cr denotes cosmic ray. Table 2 Astrochemistry»a summary processes radiative association and detachment as galactic interstellar and circumstellar chemistry and intense radiation –elds cra]UV-driven chemistry freeze-out desorption surface chemistry cr]UV-driven chemistry dust formation and destruction surface chemistry ice processing]evaporation cr]UV-driven chemistry surface chemistry freeze-out and desorption sublimation photolysis ion»molecule chemistry dust gas»dust interaction role of intense radiation –elds and time dependence gas»grain interaction large molecules origin and properties problems (physical conditions) elements ? metallicity ? dust ? cr ? X-ray? magnetic and turbulent support; gas»dust interactions trapping of gas in ices sputtering dust]gas 8 Frontiers of astrochemistry 9 D.A. W illiams Bar-Nun45 explore the possibility that comets may be a source of volatiles on Earth. Apparently comets cannot supply all the oceans. Therefore the study of meteorites (and the volatiles trapped in them) remains of vital importance (literally) for that reason alone.46,47 3 Summary of processes and problems Astrochemistry contributes to an enormous range of topics in astronomy and there are many common themes in these studies. Table 2 summarizes the main processes occurring in each of the areas described in Section 2 and notes the important problems or particular uncertainties requiring further study.As noted earlier the chemical processes involved in the early Universe are probably de–ned and the uncertainty remains in the physical description. The chemistry of AGNs is probably similar to galactic chemistry but the galactic parameters for abundances intensities and —uxes almost certainly do not apply. However we can hope to use the techniques of astrochemistry to determine some of these parameters. Within our Galaxy many problems remain and are identi–ed in Table 2. One can be optimistic that they will be solved. The most poorly understood and recurring theme seems to be the nature of the gas»dust interaction.For many astrochemical applications our lack of knowledge about the gas»dust interaction limits our understanding of the physical situation and our ability to interpret the observational data. This is partly due to uncertainty in the composition of the dust but mainly due to the processes themselves. In the 1970s the identi- –cation of gas phase schemes to produce the variety of detected interstellar molecules stimulated a growth in the experimental and theoretical study of ion»molecule reactions and a huge amount of data was provided for interstellar and circumstellar work. With the recognition of the importance of neutral»neutral reactions recent attention has been directed to those types of reaction. We are now in an era in which the importance of surface reactions and solid state chemistry is becoming recognised.The remainder of this paper is therefore on some recent work in which some aspects of the gas»dust interaction are explored. 4 The gasñdust interaction 4.1 Formation of H2 H is involved in almost all of the astrochemical works and is generally assumed to form on the surfaces of dust grains where they are available and where the conditions 2 are appropriate. I shall therefore begin this section by describing some recent work concerning H formation. This work is experimental and theoretical. The surfaces of 2 relevance are amorphous ice carbons and silicates and graphite (e.g. Taylor et al.35). Pirronello et al.48,49 have measured the abundance of HD formed on an olivine slab at a temperature in the range 5»15 K from H and D atoms at a pressure of B10~10 Torr.Their main results were that the H formation rates deduced from the experiment are up to one order of magnitude smaller than predicted by the calculations of Hollen- 2 bach and Salpeter50,51 and Hollenbach et al. ;52 these theoretical values however were in harmony with the implications of the astronomical observations. Pirronello et al.48,49 also deduced that H atoms do not tunnel readily at temperatures less than ca. 10 K. Thermal activation is needed at the lowest temperatures. They have inferred a formation rate of H that applies in a low H coverage low H mobility case and this diÜers from previous descriptions which assumed high H coverage and mobility. 2 Takahashi et al.53 made a molecular dynamics (classical) study of H formation on 2 sions were that the sticking probabilities of H atoms on the ice surface are high at 10 K amorphous H2O-ice (itself formed through a molecular dynamics study).Their conclu-10 Frontiers of astrochemistry 2 and that the H molecules formed in diÜusion (Langmuir»Hinshelwood) and direct (Eley»Rideal) processes with high probability. These molecules are formed in highly 2 excited vibrational states with lAB5»8 and it is found that only 3»5% of the H bond energy is deposited in the surface of the ice. In a related but purely quantum mechani- 2 cal calculation Farebrother et al.54 have explored the formation of H at the surface of graphite. They use a density functional method and apply periodic boundary conditions.2 The Eley»Rideal (direct) process has been studied and results so far obtained are for a collinear CwHwH approach perpendicular to the graphite lattice. They conclude that there is a barrier-free potential-energy surface for H formation and they –nd that the 2 reaction probability is very close to unity. In this simulation Farebrother et al.54 –nd that there is considerable vibrational excitation in the newly formed H (lA\3 4). First results should be available later this year from an experiment to be performed on H formation on a variety of low-temperature surfaces with in situ determination of 2 the energy budget of H2 . The experiment is being performed under the auspices of the UCL Centre for Cosmic Chemistry and Physics. 4.2 Sticking probabilities The sticking of species at surfaces is an important process in astrochemistry.It is generally assumed that the probability of sticking of almost all species is close to unity on the likely materials of interstellar dust. While this may be probable for saturated molecules it is less clear what should happen when reactions occur; e.g. 2O[grain O]grain »»»’ H H »»»’ H OHC]grain and nothing is known (at least by astrophysicists) about the interaction of grains and ions X`]grain]X]grain` ]X`[grain ]X`]grain In addition grains may be charged; in dark clouds this charge is usually negative. It is not known how the grain charge in—uences the sticking process. For molecular ions the situation is even more complicated than for atomic ions. In interacting with a negatively charged grain does charge transfer take place at long range and if so do fragments of the recombination process arrive at and stick to the surface ? Or does recombination take place on the surface with the energy release desorbing the fragments? Such processes may have signi–cant eÜects especially in star forming regions ; however little is currently known about such interactions.The implications of freeze-out can be profound. The observations indicate that selective freeze-out occurs. For example consider the interstellar chemistry of sulfur. Sulfur is known to have its cosmic (solar) abundance in diÜuse clouds but is required by models to be very heavily depleted in dark clouds. If S is simply frozen-out at the same rate as C- and O-bearing species then huge over-abundances of sulfur molecules arise during the transition from diÜuse to dark clouds.Evidently sulfur must be removed from the system faster than C and O. In a model computation by Ruffle et al.,55 results indicate that if S is depleted faster than C- and O-species by a factor ca. 102 then realistic abundances are likely to result for S-bearing species. This seems to imply that S is incorporated into the dust ; if so how? It also implies that much of the O and C returns to the gas phase through desorption ; if so how? 11 D. A. W illiams Whatever are the conclusions to these considerations they must be compatible with the observations of S-bearing species in hot cores (cf. Millar and Hatchell15). 4.3 Desorption In addition to the models of sulfur chemistry referred to in Section 4.2 other models also suggest that the eÜective sticking probability may be less than the canonical value and that desorption may therefore be occurring (Howe et al.;56 Taylor et al.57 Willacy and Williams58). Desorption may be thermal or non-thermal. The latter process may use energy input from cosmic rays photons or chemistry and the mechanisms have been recently reviewed.35 The most recent work is by Takahashi.59 She computes the energy transfer and temperature into the amorphous H2O-ice surface in the vicinity of the H2 formation site. The results of Takahashi57 show that as the H»H interaction occurs the eÜective temperature of the immediate vicinity of the site (radius of several molecule spacings) rises to several hundred K albeit for a short time ca.100 fs. This indicates that the process originally described by Duley and Williams60 for the desorption of weakly bound species such as CO may operate. 4.4 Generalized models of surface chemistry The most basic mathematical models of surface chemistry simply assume that»after arriving at the surface»atoms and radicals are saturated with hydrogen. More sophisticated models have been developed (most recently Hasegawa and Herbst34) in which earlier treatments (Pickles and Williams61) are developed. These methods generate rate equations for surface chemistry that are conveniently similar to those describing the gas phase chemistry. However they involve many uncertain data on surface processes.A concern has been expressed by Tielens (see Caselli et al.62) that this approach does not take into account the discrete nature of the surface in interstellar clouds ; i.e. that species X and Y adsorbed on diÜerent dust grains cannot in fact interact. However a revised version of this treatment has now been tested against Monte Carlo calculations and shown to be in reasonable agreement.62 5 Conclusion Our subject astrochemistry began with the identi–cation of interstellar CH CH` and CN over half a century ago. The types of microscopic processes involved»including radiative association and recombination photoionization and photodissociation ion» molecule and neutral»neutral exchanges and surface processes»were almost entirely identi–ed by Bates and Spitzer63 in 1951.The subject has expanded enormously especially from around 1970 to the present day; no longer being a curious backwater of astronomy but occupying the mainstream of astronomy. The expansion has been largely driven by the astronomical observations and has created a huge demand for fundamental data that has to a very signi–cant extent been met. Much attention is now given to astronomical situations that are time-varying in which dynamics plays a role. Though the conditions in regions of current interest are usually very diÜerent from those in diÜuse clouds the ideas developed nearly half a century ago are providing a tool to help us understand the nature of the non-stellar baryonic matter in the Universe. It is a wonderful opportunity and we are lucky to be able to bring together a diverse range of skills to exploit it.References 1 S. Lepp and P. C. Stancil in T he Molecular Astrophysics of Stars and Galaxies ed. T. W. Hartquist and D. A. Williams Oxford University Press Oxford 1998 p. 37. 12 Frontiers of astrochemistry 2 P. C. Stancil and A. Dalgarno Faraday Discuss. 1998 109 61. 3 R. S. Booth and S. Aalto in T he Molecular Astrophysics of Stars and Galaxies ed. T. W. Hartquist and D. A. Williams Oxford University Press Oxford 1998 p. 437. 4 S. Lepp and S. Tineç in T he Molecular Astrophysics of Stars and Galaxies ed. T. W. Hartquist and D. A. Williams Oxford University Press Oxford 1998 p. 487. 5 K. Ohta T. Yamada K. Nakanishi K. Kohno M. Akiyama and R. Kawabe Nature (L ondon) 1996 382 426.6 A. Omont P. Petitjean S. Guilloteau R. G. McMahon P. M. Solomon and E. Pontal Nature (L ondon) 1996 382 428. 7 P. Madau H. C. Ferguson M. E. Dickinson M. Giavalisco C. C. Steidel and A. Fruchter MNRAS 1996 283 1388. 8 B. Paczynski and C. Kouveliotou Nature (L ondon) 1997 389 548. 9 (a) J. H. Black Faraday Discuss. 109 257; (b) J. H. Black in T he Molecular Astrophysics of Stars and Galaxies ed. T. W. Hartquist and D. A. Williams Oxford University Press Oxford 1998 p. 467. 10 B. J. McCall T. Oka K. H. Hinkle and T. R. Geballe Faraday Discuss. 1998 109 267. 11 E. F. van Dishoeck Faraday Discuss. 1998 109 31. 12 D. P. Ruffle D. A. Williams S. D. Taylor and T. W Hartquist MNRAS 1997 291 235. 13 D. P. Ruffle T. W. Hartquist J. M. C. Rawlings and D.A. Williams Astron. Astrophys. 1998 334 678. 14 R. Plume E. A. Bergin J. P. Williams and P. C. Myers Faraday Discuss. 1998 109 47. 15 T. J. Miller and J. Hatchell Faraday Discuss. 1998 109 15. 16 T. J. Millar in T he Molecular Astrophysics of Stars and Galaxies ed. T. W. Hartquist and D. A. Williams Oxford University Press Oxford 1998 p. 331. 17 H. Olofsson Astrophys. Space Sci. 1996 245 169. 18 M. Lindqvist R. Lucas H. Olofsson A. Omont K. Eriksson and B. Gustafsson Astrophys. Space Sci. 1995 224 501. 19 D. Chastaing P. L. James I. R. Sims and I. W. M. Smith Faraday Discuss. 1998 109 165 20 P. Thaddeus M. C. McCarthy M. J. Travers C. A. Gottlieb and W. Chen Faraday Discuss. 1998 109 121. 21 R. I. Kaiser C. Ochsenfeld M. Head-Gordon D.Stranges and Y. T. Lee Faraday Discuss. 1998 109 183. 22 M. Jura and H. W. Kroto Astrophys. J. 1990 351 222. 23 L. A. M. Nejad and T. J. Millar MNRAS 1988 230 79. 24 J. J. Keady and S. T. Ridgway Astrophys. J. 1993 406 199. 25 H.-P. Gail and E. Sedlmayr Faraday Discuss. 1998 109 303. 26 I. CherchneÜ in T he Molecular Astrophysics of Stars and Galaxies ed. T. W. Hartquist and D. A. Williams Oxford University Press Oxford 1998 p. 265. 27 I. CherchneÜ J. R. Barker and A. G. G. M. Tielens Astrophys. J. 1991 377 541. 28 I. CherchneÜ J. R. Barker and A. G. G. M. Tielens Astrophys. J. 1992 401 269. 29 D. A. Howe and D. A. Williams in T he Molecular Astrophysics of Stars and Galaxies ed. T. W. 30 J. M. C. Rawlings in T he Molecular Astrophysics of Stars and Galaxies ed.T. W. Hartquist and D. A. Hartquist and D. A. Williams Oxford University Press Oxford 1998 p. 347. Williams Oxford University Press Oxford 1998 p.393. 31 W. Lui in T he Molecular Astrophysics of Stars and Galaxies ed. T. W. Hartquist and D. A. Williams Oxford University Press Oxford 1998 p. 415. 32 Y. Aikawa T. Umebayashi T. Nakano and S. Miyama Faraday Discuss. 1998 109 281. 33 E. F. van Dishoeck in T he Molecular Astrophysics of Stars and Galaxies ed. T. W. Hartquist and D. A. Williams Oxford University Press Oxford 1998 p. 53. 34 (a) T. I. Hasegawa and E. Herbst MNRAS 1993 261 83; (b) T. I. Hasegawa and E. Herbst MNRAS 1993 263 589. 35 D. A. Williams and S. D. Taylor QJRAS 1996 37 565. 36 I. A. Crawford and D. A. Williams MNRAS 1997 291 L53.37 D. A. Williams Astrophys. Space Sci. 1996 237 243. 38 R. E. Hibbins J. R. Miles P. J. Sarre. S. J. Fossey and W. B. Somerville in T he DiÜuse Interstellar Bands ed. A. G. G. M. Tielens and T. P. Snow Kluwer Dordrecht 1995 p. 25. 39 J. R. Miles P. J. Sarre and S. M. Scarrott in T he DiÜuse Interstellar Bands ed. A. G. G. M. Tielens and T. P. Snow Kluwer Dordrecht 1995 p. 143. 40 O. Guillois G. Ledoux I. Nenner R. Papoular and C. Reynaud Faraday Discuss. 1998 109 335. 41 P. P. Sorokin J. H. Glownia and W. Ubachs Faraday Discuss. 1998 109 137. 42 D. A. Kirkwood H. Linnartz M. Grutter O. Dopfer T. D. Motylewski M. Pachkov M. Tulej M. Wyss and J. P. Maier Faraday Discuss. 1998 109 109. 43 J. Crovisier Faraday Discuss. 1998 109 437. 44 W. M. Irvine J.E. Dickens A. J. Lovell F. P. Schloerb and M. Senay Faraday Discuss. 1998 109 475. 45 T. Owen and A. Bar-Nun Faraday Discuss. 1998 109 453. 46 S. J. Clemett M. T. Dulay J. S. Gillette X. D. F. Chillier T. B. Mahajan and R. N. Zare Faraday Discuss. 1988 109 417. 13 D. A. W illiams 47 A. B. Verchovsky I. P. Wright C. T. Pillinger and A. V. Fisenko Faraday Discuss. 1998 109 403. 48 V. Pirronello C. Liu L. Shen and G. Vidali Astrophys. J. 1997a 475 L69. 49 V. Pirronello O. Biham C. Liu L. Shen and G. Vidali Astrophys. J. 1997b 483 L131. 50 D. J. Hollenbach and E. E. Salpeter J. Chem. Phys. 1970 53 79. 51 D. J. Hollenbach and E. E. Salpeter Astrophys. J. 1971 163 155. 52 D. J. Hollenbach M. W. Werner and E. E. Salpeter Astrophys. J. 1971 163 165. 53 J. Takahashi K. Masuda and M. Nagaoka Astrophys. J. 1998 in press. 54 A. Farebrother A. Fisher and D. C. Clary 1998 in preparation. 55 D. P. Ruffle T. W. Hartquist and D. A. Williams 1998 in preparation. 56 D. A. Howe S. D. Taylor and D. A. Williams MNRAS 1996 279 143. 57 S. D. Taylor O. Morata and D. A. Williams Astron. Astrophys. 1997 313 269. 58 K. Willacy and D. A. Williams MNRAS 1993 260 635. 59 J. Takahashi 1998 in preparation. 60 W. W. Duley and D. A. Williams MNRAS 1993 257 13P. 61 J. P. Pickles and D. A. Williams Astrophys. Space Sci. 1977 52 443. 62 P. Caselli T. I. Hasegawa and E. Herbst Astrophys J. 1998 in press. 63 D. R. Bates and L. Spitzer Jr. Astrophys J. 1951 113 441. Paper 8/04023K; Received 28th May 1998

ISSN:1359-6640    

DOI:10.1039/a804023k

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

Faraday Discuss. 1998 109 15»30 Chemical models of hot molecular cores Thomas J. Millar* and Jennifer Hatchell Department of Physics UMIST PO Box 88 Manchester UK M60 1QD We present the results of observational studies of a number of molecular clouds associated with ultracompact HII regions in regions of massive star formation. We derive molecular abundances and show that gas-phase-only chemical models of these regions are incapable of reproducing the observations. It appears that a rich chemistry occurs in the ice matrices frozen onto dust grains at high densities in the collapse of molecular clouds to form stars. The switch-on of a protostar can evaporate the ice matrices returning processed material to the gas phase. We present evidence that the surface chemistry can fractionate molecules in deuterium as well as form complex organic molecules.n\1»5. Speci–cally HMCs contain very large abundances of hydrogenated and together with larger species such as and Interstellar clouds in regions of massive star formation are known to contain very small (O0.1 pc; 3]1015 m) hot (T ca. 100»300 K) and dense [n(H2) ca. 1012»1014 m~3] clumps of gas which typically contain several tens of solar masses of gas and have large extinctions towards external radiation. In recent years molecular line observations have shown that these clumps known as hot molecular cores (HMCs) have a diÜerent chemical composition compared to that in cold (10 K) dense [n(H ca. 1010 m~3] dust clouds in which highly unsaturated species are found for example the cyanopolyyne chains 2) HC molecules including CH 2n`1N 4 NH3 H2O H2S CH OH C2H5OH (CH3)2O with abundances enhanced by factors of ca.103»105 over those in cold clouds. Very few unsaturated hydrocarbon chains are detected in 3 HMCs an exception being HC3N which is also observed to be vibrationally excited. Such a diÜerent composition may re—ect the temperature diÜerences between HMCs and cold clouds. However it appears impossible that gas-phase chemistry at 100»300 K is responsible for the bulk of the species observed in HMCs. This is because deuterium fractionation is observed in many molecules including HDO NH2D DCN HDCO and D CO with enhancements of 100»1000 over the cosmic D H ratio of 10~5 similar to those observed in cold clouds.Since the process of fractionation depends on small zero- 2 point energy diÜerences the observed fractionation in HMCs must have taken place at low temperatures (\20 K) and not at the gas kinetic temperature of the HMCs. Taken together with the enhancements in the abundances of hydrogenated molecules the implication is that signi–cant molecular processing has taken place on grain surfaces. Following this reasoning HMCs can be placed into an evolutionary picture of star formation. Cold and dense clouds are in an early or pre-collapse phase. Typical lifetimes before such clouds develop IR sources thought to be indicative of young stellar objects (YSOs) are around a few 105 years. At 10 K the time scale for accretion of gas onto the surfaces of dust grains is about 3]1015/n years where n is the hydrogen number density per m3.Thus on the time scale of cloud collapse and star formation material is expected to collide with and freeze onto the surfaces of cold dust grains. Note that at high number densities the accretion time scale is shorter than the collapse time scale so that all gas 15 16 Chemical models of hot molecular cores phase species with the exception of H2 He and their associated ions eventually –nd themselves incorporated into a frozen ice-like matrix surrounding a refractory grain core. These molecular ices can be detected through IR absorption spectroscopy once the YSO forms but there will be a period before this when both the gas and solid state material become ì invisible œ since gas phase molecules are frozen out and there is no YSO to provide an IR continuum against which to detect absorption.Once massive star formation occurs a hot OB star or more likely a cluster of OB stars will cause thermal heating or drive shock fronts into surrounding dense clumps thereby heating them and releasing the fragile ices into what is now a hot dense and essentially neutral gas. Since grain surface processes tend to produce stable neutrals (radicals are difficult to preserve in the ice phase which may last for 105»106 yr) the resulting clump of gas now an HMC takes over 104 yr to be altered chemically. Eventually parent species evaporated from grain ices are destroyed producing reactive daughter products which can drive a complex chemistry. In this article we shall brie—y review some recent observations of HMCs before discussing detailed chemical kinetic models.We shall show that sulfur species may be used as a chemical clock that the DCN HCN ratio may be used to derive information on the reaction between H and DCN and that models of HMCs need to consider a detailed solid-state chemistry which must be capable of producing molecules as complex as ethanol and dimethyl ether. and Observational results In this section we report on two observational programmes aimed at elucidating the chemical composition and physical condition in HMCs associated with ultracompact HII regions (UCHIIs) that is molecular gas associated with young bright hot stars able to ionise part of their natal molecular cloud.Physical conditions and hot core ages We used the James Clerk Maxwell Telescope (JCMT) to perform a molecular line survey toward 14 HMCs in up to ten 1 GHz bands between 200 and 350 GHz.1 In total some 150 molecular lines have been detected with great variation between sources. Some sources such as G34.3 G10.47 and G31.41 show emission from high excitation transitions of CH OH CH3CN CH3CCH and as well as transitions of complex species such 3 as C2H5OH (CH3)2O C and 2H5CN. Other sources such as G10.30 G13.87 G43.89 and G45.47 contain only a few lines from simple species such as C17O C18O C34S SO and low energy lines of CH Molecular abundances and excitation conditions have been derived using the rota- 3OH CH3CCH. 3OH and CH3CN are thermalized implies number n(H2) in excess of 1013 m~3 which can be reached in the centres of these and CH OH our observations of the 3 tion diagram approach2 which assumes that all emission lines are optically thin that the emission –lls the telescope beam and that the source is in thermal equilibrium at a single temperature.Such diagrams can be made when multiple transitions are observed for a particular molecule. In addition model line intensities can be calculated by adopting local thermodynamic equilibrium (LTE) and varying column density N kinetic temperature T and source size hS . In this case we can treat optically thick transitions. The requirement that the lines of CH densities HMCs. For several molecules such as CH isotopomers indicate that the main lines have signi–cant optical depth so that our LTE 3CN approach is to be preferred.Symmetric top molecules such as CH CN are particularly good tracers of temperature since rotational transitions are split by the K quantum number such that 3 several K components fall close together in frequency thereby minimising observational T . J. Millar and J. Hatchell temperature and source size (in arcsec). errors. We can use our observations of the J\19»18 transition of CH3CN at 349 GHz and the J\13»12 transition at 239 GHz to derive temperatures source sizes virial masses and lower limits to the beam-averaged fractional abundance of CH3CN (Table 1). Since the excitation energy of the observed lines range from 60 to 430 K emission is dominated by hot gas. The actual abundance of CH3CN must be much larger than given in this table because the lines have signi–cant optical depth and the source size is much smaller than the telescope beam.In the strongest sources of CH3CN emission the fractional abundance could be ca. 10~7 approximately 100 times larger than observed in cold dust clouds. Note that although the analysis assumes a constant temperature it appears that a temperature and/or density gradient exists within at least two sources since the J\19»18 transition requires a smaller source size and larger temperature than the J\13»12 transition. Similar results are obtained from analysis of CH OH although the derived temperatures are somewhat lower and the source sizes somewhat 3 larger than those derived from CH CN consistent with the excitation energies of the observed CH 3 3OH transitions and temperature gradients in these sources.Fig. 1 gives CH representative spectra of CH3CN toward G31.41 and Fig. 2 shows model –ts to the 3CN Several HMCs show no CH3CN emission and we derive upper limits to the column density of ca. 1017 m~2 for a temperature of 50 K. SO C34S SO and H2S lines have been detected in a number of HMCs in the 2 2 survey. The 220»211 line of H2S at 216.7 GHz the only accessible transition was detected in all sources observed at this frequency. SO was detected towards six sources with excitation ranging from 100 to 200 K while SO and C34S were detected toward nearly all sources observed. Table 1 CH3CN excitation temperatures source sizes virial masses and beamaveraged fractional abundances (x) in HMCs T /K h /arcsec S source 19»18 13»12 19»18 13»12 Mvir/solar masses 0.98 0.80 0.60 0.63 1.41 1.19 2.29 1.12 1.53 3.46 65 134 141 142 141 119 87 114 149 79 G9.62 G10.47 G29.96 G31.41 G34.3 J\19»18 spectrum with Gaussian –ts to the K\0 1 2 5 and 6 components.Fig. 1 CH3CN spectra towards G31.41. The observed spectra (thin lines) are overlaid with Gaussian –ts to the lines. (a) J\13»12 spectrum with Gaussian –ts to K\0 to 6 components. (b) 17 x]1011 [2 [ » 4 190 810 330 450 460 [7 [3 18 Fig. 2 CH3CN Although our sources were chosen because they are known to emit3 in high excitation lines of NH3 our observations indicate that they can be divided into two broad classes a line-rich class incorporating eight sources which show varying amounts of emission from high excitation lines with G34.3 G10.47 and G31.41 the strongest and G12.21 and G75.78 the least strong of these and a line-poor class which contains six sources.There appears to be a real physical diÜerence between these classes in that the line-rich sources show a hot dense core surrounded by cooler gas while the line-poor sources have evidence only for the cooler halo gas. The hot cores have angular sizes of the order of a few arcseconds corresponding to O0.1 pc temperatures [100 K densities [1014 m~3 and virial masses of the order of a few hundred solar masses. The halo gas is typically cool 10»20 K and dense ca.1010»1011 m~3. The hot cores appear to be centrally condensed in these objects as observations made at oÜsets of 20 arcsec show no emission from hot gas. In addition density and temperature gradients probably exist as can be inferred from the CH3CN observations as discussed above. Toward the cores we derive temperatures of ca. 50 K from CH OH gives temperatures ca. 100 K with increasing tem- while on smaller size scales CH 3 perature and smaller sizes for high excitation transitions. Vibrationally excited CH (and HCN) has also been detected. Emission from these species is pumped by IR absorp- 3CN Chemical models of hot molecular cores temperature and source size for several hot core sources (error bars are 1p) 3CN 19 T .J. Millar and J. Hatchell Fig. 3 The evolution of column densities of various species in the ultracompact component of G34.3 tion of photons emitted by hot dust grains at several hundred Kelvin. Such pumping is efficient over very small angular extents and future observations of vibrationally excited lines particularly with interferometers will be able to trace the hottest regions and hence presumably the source of excitation is these cores. The halo gas is very extended on the order of several parsecs in many cases and has a high column density ca. 1027» 1028 m~2. In the line-poor sources the halo gas appears to have a column density at least one order of magnitude smaller than in the line-rich sources. The lack of detectable hot cores in these sources may be due to beam dilution that is the hot gas has a small angular extent relative to the telescope beam size in which case the hot core sizes would either need to be \0.8 arcsec or that the original hot gas has cooled thus making these sources more evolved than the line-rich sources.Of course another possibility is that they are signi–cantly displaced by at least 10 arcsec from the UCHII regions in which case they would have been missed by our beam. We can also get an estimate of the evolutionary status of the sources through chemical modelling. Charnley4 has suggested that the relative abundances of the sulfurbearing molecules can be used to determine the time from which ice mantles are evaporated. In this picture H2S is formed in the mantles (it is known to have a very low fractional abundance in cold clouds ca.10~11) and evaporates. In addition to proton transfer reactions with H2O` it reacts with atomic H rapidly in hot gas H2S]H]HS]H2 HS]H]S]H2 where the –rst of these reactions has a relatively low activation energy barrier,5 ca. 350 20 Chemical models of hot molecular cores K. The atomic sulfur released by this then reacts with OH and O to form SO which is 2 itself converted to SO and CS. Thus one expects the abundances of H2S SO and SO 2 to trace increasing chemical evolution of evaporated materials from ice mantles. Fig. 3 2 shows the evolution of these abundances in the ultracompact core of G34.3 adopting a temperature of 300 K as a function of time since evaporation of the mantles. At such a high temperature H2S is destroyed completely within a few hundred years a very short time in astronomical terms so that one would never expect to detect any H2S emission from this region.Further out from the centre of the cloud the temperature decreases is this gas at around 100»150 K which gives us the most useful handle on chemical age. and H2S survives longer and contributes appreciably to the observed column density. It factor of 10 between the line-rich sources strongly suggesting that they are at similar In fact the relative abundances of H2S H2CS OCS and CS vary by less than a evolutionary stages. SO and SO abundances are somewhat larger in G5.89 G10.47 and G75.78 than in other sources. These sources may be slightly older or alternatively the 2 excess SO and SO emission may originate from molecular out—ows in or near hot 2 cores.The simultaneous detection of H2S SO and SO places bounds on the source 2 ages of a few thousand to 105 yr depending on the cosmic ray ionization rate and core temperature. Deuterated molecules in hot cores Table 2 presents a list of the deuterated species detected towards three sources including the hot core source in the Orion molecular cloud (OMC) and shows that fractionation is large despite the high kinetic temperature. We have recently begun a programme to study D H ratios in our HMC sample and in May 1997 detected the 3»2 transitions of HC15N and DCN in a number of sources with the JCMT. We also search for emission 20 arcsec oÜ the peak in order to (i) see if the emission was extended and (ii) to derive DCN HCN ratios in the halo gas.Choosing a line of HC15N ensures that emission is optically thin and allows us to directly derive the abundance ratio6 N N (HCN) (DCN)\ [ [ 14 15 N] N] I I D e3.93@Tx x H where [15N]/[14N] is the isotopic ratio of 15N to 14N and varies from source to source,7 but is always within 20% of 400 ID IH are the integrated intensities of the DCN and HC15N transitions respectively and T is the excitation temperature although one can see that the DCN HCN ratio is only weakly dependent on T x x . Table 3 gives the ratio for Tx\30 K. Ratios for Tx[100 K are about 10% larger. The absolute column densities of HC15N and DCN are of course much more sensitive to Tx a parameter we know poorly since we only have one transition observed.Gibb8 estimates T ca. 55 K on core for G34.3 decreasing to ca. 10»20 K at an oÜset of 15 arcsec. Our DCN HCN abundance ratios (1»4)]10~3 are consistent from source to source and an order of magnitude smaller than that observed in the cold dust cloud TMC-1 (0.023^0.001)9 and the Orion ridge cloud (0.02^0.01),10 and similar to the Orion hot core (0.003),10 while our oÜ-core abundance ratios are somewhat larger since we have detected DCN but not HC15N in some positions. It is known that the deuteration ratios in TMC-1 at 10 K and in the Orion ridge at 70 K are dominated by gas phase processes,11 whereas the Orion Hot Core and the HMCs in our survey contain material processed through a grain surface.The observation of DCN HCN ratios as low as 10~3 in these could arise for a number of reasons. Firstly it might re—ect the fact that the ratio was low when the gas accreted onto the ice. A ratio ca. 10~3 requires gas temperatures [50 K for which accretion is inefficient due to thermal desorption.11 T . J. Millar and J. Hatchell Table 2 Fractionation ratios for molecules in TMC-1 (10 K) the ridge clouds of OMC-1 (50»70 K) and the Orion hot core (150»200 K) molecule HDO DCO` N2D` DCN DNC NH C2D HDCO 2D DC C c-C HDCS 3HD 4D DC CH CH 3 5 N N 3 2 OD DOH 2 CH DCCH D2CO Secondly one could argue that a large ca. 10~2 DCN HCN ratio in cold gas could be diluted when frozen onto an ice surface.However detailed models of hydrogenation and deuteration on grain surfaces show that the formation of D-bearing molecules is enhanced over H-bearing molecules partly re—ecting the stronger bonds which D atoms form but also re—ecting the fact that the D H ratio on the surface is greatly enhanced over the cosmic value because it is enhanced in the gas phase and transfers this via collisions on the grain surface. Thus surface chemistry should enhance the DCN HCN abundance ratio.12h15 Finally it is possible that the DCN HCN ratio is large ca. 10~2 or greater on the ice mantles but gets altered on a rapid timescale in hot gas upon evaporation of the ice. Schilke et al.10 suggested that the reaction H]DCN]HCN]D can cycle deuterium between atomic D and DCN and estimated a rate coefficient of 10~16 m3 s~1 and an activation energy barrier E tion barrier the reaction with atomic H alters the initial DCN HCN ratio on a time- A/kB200 K.With such a small activascale of a few hundred years16 for a gas temperature of 100 K. If EA/kT [10 then the timescale for altering the ratio becomes longer than 104 yr incompatible with the ages Table 3 Fractionation ratios (]103) for DCN HCN in various HMCs for an excitation temperature of 30 K source G5.89 G9.62 G10.47 G29.96 G31.41 G34.3 hot core OMC-1 TMC-1 [0.002 0.002 0.003 0.02 0.01 0.045 0.015 \0.045 0.023 0.015 0.01 0.003 0.14 0.02 \0.02 0.015 0.02 0.08 0.004 0.06 0.013 0.03 0.04 0.054 \0.05 0.02 DCN HCN 2.1 2.2 3.2 2.7 2.3 1.4 21 22 Chemical models of hot molecular cores inferred for HMCs.This reaction is endothermic by about 800 K not large enough to prevent H atoms from destroying DCN in hot cores. However recent calculations of the potential surface for this reaction by Herbst and Talbi (this meeting) indicate that energy barriers of several thousand K are present. If this conclusion is correct reactions with H atoms cannot alter the DCN:HCN ratio. NH and Chemical kinetic models of HMCs The earliest models were developed by Brown et al.17 who considered a three stage process. In the –rst stage they followed the chemistry and mantle accretion of a spherical molecular cloud with initial atomic conditions as it underwent a gravitational collapse to high density .In this stage the accretion time was shorter than the collapse time (note taccPn~1 and tcollPn~1@2 where n is the total number density) so that the accreted grain mantle was rich in atoms of C N and O which were hydrogenated to form CH NH 4 and H2O in the second stage which followed a limited surface chemistry consisting essentially of hydrogenation of atomic species. Because of the high mobility of atomic 3 hydrogen (an H-atom scans the entire grain surface in about 10~6 s) hydrogenation is always rapid providing there is a sufficient —ux of H atoms to the surface which can be met at low number densities early in the collapse. At higher densities and later times the conversion of H atoms to H reduces their number and allows heavy atoms to combine.Finally the third stage investigated the chemistry of the evaporated ices in hot gas and 2 found that they survived for at least 104 yr. Brown and Millar13,14 extended this work to include deuteration. 2CO and in ices there is a region in which the hot gas once mantles have been evapo- 3 2CO CH3OH which can be found 3 breakdown leads to a rich N 3 Subsequently much work has concentrated on the third stage4,18h23 and shown that the evaporation of simple molecules from ices can lead to the formation of more complex species in the HMC. Most of these models have been investigated for conditions appropriate to the Orion Hot Core a clump of gas which shows large abundances of N-bearing molecules in contrast to a nearby HMC the Orion Compact Ridge which is rich in O-bearing molecules.The compositional diÜerences between these HMCs are puzzling. Caselli et al.20 suggested an answer through modelling the collapse of a molecular cloud into a central protostar with the inclusion of a radial temperature pro–le. Because of thermal desorption and the diÜerent binding energies involved the composition of the ice mantle varies with radial distance from the young star. In particular the low binding energy of CO means that it is easily evaporated close to the star whereas the much stronger bound NH can be retained. Since CO is hydrogenated to H CH rated contains NH and CO but little or no H 3OH further from the protostar.Because CO is unreactive chemistry close to the protostar in warmer gas while further from the star in relatively cooler gas and H2CO CH3OH can drive the formation of more complex molecules such as (CH3)2O and HCOOCH3 . cores. This is exactly the situation observed in the Orion hot The model developed by Caselli et al.20 used a reasonable description of the grain surface chemistry although the number of reactions included is an order of magnitude less than the number of gas phase reactions. In part this re—ects uncertainty about surface mobilities evaporation rates [which are exponentially dependent on the (uncertain) grain temperature] radical population H-atom —ux reactivities activation barriers and so on. Nonetheless for well observed sources one might hope to build a large enough model to show where gaps in knowledge are crucial.One such source is the HMC associated with the ultracompact HII region G34.3]0.15. Molecular line observations show that the molecular cloud has three main components (i) an ultracompact core, Table 4 Column densities calculated at 3.16]103 yr in the ultracompact core (UCC) and compact core (CC) components at 105 years in the halo UCC species H O CO OH2 NO 2.632]1018 2.700]1019 2.961]1016 1.070]1014 1.724]1016 6.059]1018 4 H CO2O CH2 HNC 2.427]1015 1.129]1017 2.658]1016 6.426]1012 1.196]1016 2.266]1017 HCO HC 2CO 2H2 C C3H2 C4H HC4 C 3N 3H4 1.156]1014 6.853]1014 5.399]1012 2.437]1015 5.573]1013 3.952]1011 He` N CN2 1.131]1019 5.089]1011 6.340]1015 8.181]1014 6.134]1013 5.138]1011 C C3N 3O C CH `2CO NH HCN C2H 2.707]1015 2.880]1016 8.348]1013 4.164]1018 2.254]1014 2.334]1017 NH 3CN CH3 CH3OH H3 ` HCO` N2H` 3.684]1012 8.138]1012 1.184]1012 5.216]1014 2.019]1014 1.453]1011 C2S S C ` 3 S HS H HCS` 2S 1.780]1013 8.942]1012 8.348]1010 3.268]1015 2.582]1017 4.555]1016 2 CS SO SO OCS H HCS 2CS C4S 2.716]1015 2.325]1015 1.542]1014 5.789]1013 2.715]1015 4.845]1013 C NS ONS Si CH CHO 3.077]1016 1.115]1017 2.242]1017 1.942]1016 1.682]1014 3 T .J. Millar and J.Hatchell halo CC 2.659]1020 1.646]1019 2.333]1018 1.336]1016 6.973]1016 2.210]1017 8.567]1018 1.346]1019 2.783]1017 1.877]1016 5.831]1016 2.468]1018 9.237]1015 6.257]1016 8.802]1014 6.583]1012 3.483]1015 7.567]1015 2.626]1016 5.804]1016 9.851]1015 4.996]1012 1.235]1016 1.170]1017 1.413]1015 5.261]1016 6.742]1014 1.657]1014 1.002]1013 4.208]1014 1.280]1015 3.184]1014 3.835]1012 1.928]1015 1.657]1013 3.325]1012 5.921]1018 1.855]1015 1.095]1014 6.096]1014 1.091]1014 9.185]1014 5.554]1018 4.846]1013 3.034]1015 8.763]1014 2.109]1013 3.127]1012 1.165]1014 6.373]1014 1.229]1015 1.324]1016 3.753]1013 4.222]1013 4.352]1015 7.299]1015 6.087]1014 2.085]1018 5.045]1013 1.137]1017 2.012]1015 1.151]1015 1.767]1014 1.861]1015 9.452]1014 3.566]1015 3.124]1013 4.925]1013 9.744]1012 1.695]1014 7.768]1013 1.832]1012 4.718]1012 1.785]1013 7.153]1014 8.707]1016 8.700]1016 2.799]1016 6.229]1014 1.716]1017 7.223]1011 1.619]1015 1.361]1016 1.414]1016 1.048]1015 2.089]1015 2.732]1013 3.715]1013 3.491]1012 2.683]1017 2.830]1013 3.899]1014 4.864]1012 5.795]1012 3.504]1015 9.056]1014 1.572]1019 4.651]1018 1.190]1017 1.465]1015 4.679]1012 1.029]1017 2.106]1017 6.394]1016 9.463]1015 9.400]1013 23 observedb totala [2.9]1019 2.771]1020 5.692]1019 2.641]1018 3.224]1016 1.453]1017 8.748]1018 [5.8]1015 1.3]1019 \3.3]1014 [5.3]1013 [2.3]1016 3.792]1016 2.335]1017 3.732]1016 1.801]1013 2.780]1016 3.512]1017 [2.5]1015 1.8]1016 2.808]1015 5.362]1016 6.835]1014 4.530]1015 8.232]1013 4.245]1014 [5.9]1014 6.7]1014 2.279]1019 1.904]1015 9.483]1015 2.304]1015 1.916]1014 9.221]1014 7.176]1015 3.674]1016 1.921]1015 6.262]1018 3.134]1014 3.471]1017 1.2]1015 [8.1]1015 [2.7]1018 2.4]1014 3.7]1016 [2.0]1015 2.047]1015 1.208]1015 1.876]1014 2.552]1015 1.225]1015 3.568]1015 6.454]1014 1.716]1017 7.161]1014 9.196]1016 3.588]1017 8.769]1016 [1.1]1016 [3.0]1013 1.0]1016 [2.4]1015 1.6]1016 [7.7]1015 [5.6]1016 [4.0]1013 3.792]1015 4.804]1015 1.864]1014 1.008]1014 6.223]1015 2.693]1017 1.585]1019 4.973]1018 4.071]1017 3.035]1016 2.668]1014 24 Chemical models of hot molecular cores Table 4»(Continued) halo CC UCC observedb species totala 3.5]1015 C2H5OH CH OCH 3 HCOOCH3 1.6]1016 3 1.175]1010 5.266]1014 2.647]1014 4.623]109 3.569]1015 8.875]1012 3.684]108 1.748]1011 1.006]1011 1.971]109 2.282]1012 1.595]106 1.026]1010 2.956]1014 7.101]1014 1.043]109 1.367]1015 4.236]1012 1.125]109 2.309]1014 1.936]1014 1.609]109 2.200]1015 4.639]1012 C2H6CO C C2H4 2H5 5.6]1023 C2H6 H electron 2 3.510]1015 5.706]1023 7.293]1016 0.000]10 1.657]1023 2.834]1016 1.206]1015 1.350]1023 3.403]1016 2.304]1015 2.700]1023 1.056]1016 a Total column density to the cloud centre.b Column density derived from observation. 2. H2O On the other hand with a size of 0.01 pc n(H2) ca. 2]1013 m~3 T ca. 300 K (ii) a compact core with size n(H 0.1 pc ca. 2]1012 m~3 T ca. 100»300 K and (iii) a molecular halo with size ca. 2) n(H 3.5 pc 2)Pr~2 and T (r)Pr~0.4. Macdonald et al.24 performed a 330»360 GHz spectral line survey using the JCMT and detected some 338 lines from 35 species and estimated molecular abundances. Further observations were done on this source as part of the HMC survey.1 These observations show evidence of all three components in G34.3 and a detailed chemical kinetic model has been constructed.25 The model follows the chemistry at 22 radial points.In the halo evolution from atomic constituents is followed whereas in the compact and ultracompact cores where mantle evaporation occurs the chemistry follows the breakdown of parent species. Except for H formation no grain surface chemistry is followed but the gas phase chemistry is extensive with ca. 2 2200 reactions covering 225 species. Because the chemistry is depth-dependent column densities can be calculated as a function of time. Fig. 3 shows how the column densities of selected species change in the ultracompact core. One sees that H2S is destroyed (by H atoms) within a thousand years to produce SO and SO which possesses a large activation barrier for reaction with H atoms is essentially unchanged.The ice mantles are assumed to be fairly simple and one sees that reasonable column densities of some complex molecules can be found in particular CH CHO HCOOCH 3 and (CH3)2O 3 but no C2H5OH is formed whereas the observational abundance is26 ca. 10~8»10~7 typical of many HMCs. This failure to reproduce the C2H5OH abundance may re—ect a lack of knowledge about its gas phase production in hot gas or more interestingly be an indication that C2H5OH is actually a parent molecule that is a product of grain surface chemistry itself. and The detailed model of G34.3 shows that good agreement can be found for most molecules if the time scale from evaporation is ca. 3000»104 yr and that for evolution of the halo gas is ca.105 yr (see Table 4 for a comparison for calculated column densities with those observed) but clearly points to species such as C2H5OH and HCOOCH3 which are ca. 105 and 102 short of the column densities observed. For this reason we have been developing a comprehensive time dependent chemical kinetic model which follows both the gas phase and grain surface chemistry in a collapsing cloud with a grain surface chemistry detailed enough to account for the synthesis of C2H5OH (CH3)2O.27,28 The need for surface chemistry to produce complex species can be in a number of HMCs. 2H5OH (CH3)2O inferred from Table 5 which gives observed fractional abundances29 for CH OH and C 3 In our model we have included 80 surface species and 200 gas phase species limited and (CH by 2000 gas phase reactions and about 80 surface reactions based on those described by Hasegawa and co-workers.30,31.However our surface reactions are optimised to allow the production of CH OH C2H5OH 3)2O on grains. In addition to simple 3 25 T . J. Millar and J. Hatchell hydrides we include the surface formation of CHO and H2CO CO2 by reactions such as *CO]*H]*HCO *HCO]*H]*CO]H2 *CH2]*O]*H2CO *H2CO]*H]*HCO]H2 2]*H *CO]*OH]*CO *HCO]*O]*CO2]*H where *M represents a chemical species M on the grain. However important reactions which mitigate against the formation of molecular backbones containing the CwC bond have been neglected in particular *C]*C]*C2 which could eventually lead to the production of ethane *C2H6 which is detected in cometary comae.32 Because of this neglect our models calculate the maximum abundances of complex molecules possible.The majority of reactions which build alcohols and ethers involve the reaction of *CH (n\0»3) with *OH (m\0 1) m n 3 *CHn]*OHm ]*CHnOHm OH. Routes to ethanol and followed by hydrogenation as necessary to produce *CH dimethyl ether involve forming a backbone of either CwCwO or CwOwC followed by hydrogenation. The backbones form via *CO]*C]*CCO *CO]*C]*COC A large number of model calculations have been performed. The models consider freefall gravitational collapse from initial densities n of 108 109 and 1010 m~3 as well as i retarded collapse models in which the collapse is slowed by a factor B. Free-fall times range from 4.4]106 yr for ni\108 m~3 to 4.4]105 yr for ni\1010 m~3.Changes in visual extinction and temperature are also followed during the collapse. One should note that in the late stages of collapse molecular line emission becomes optically thick and the cloud is not able to radiate away its gravitational energy. As a result it heats up T such that once dust[50 K mantles evaporate rapidly. In addition to this thermal desorption process the model also contains a number of other desorption mechanisms including mantle explosions direct UV photodesorption cosmic ray induced photodesorption and direct heating by cosmic rays. In addition some models not presented here consider desorption via periodic shock waves and thermal desorption caused by material in a rotating cloud passing close to a nearby star.Table 5 Observed fractional abundances of complex organic molecules in HMCs source C2H5OH CH3OH (CH3)2O SgrB2(N) OrionCR W51e1/e2 G34.3]0.15 NGC6334F 1.3]10~9 3.0]10~9 9.5]10~9 1.0]10~9 2.4]10~9 4.3]10~8 9.3]10~7 9.5]10~7 9.4]10~8 4.8]10~7 2.0]10~8 1.3]10~6 1.0]10~7 3.9]10~7 8.8]10~7 26 Chemical models of hot molecular cores Fig. 4 The evolution of fractional abundances of various gas phase species in the –rst 2]106 yr of a retarded collapse Fig. 4»6 show sample gas phase and grain surface fractional abundances for a retarded collapse from 109 m~3. In Fig. 4 the gas phase abundances are shown for times up to 2]106 yr. At times bigger than this accretion of the gas begins to dominate and abundances decrease (Fig.5) essentially to zero. Correspondingly in this period ice 27 T . J. Millar and J. Hatchell Fig. 5 The evolution of fractional abundances of various gas phase species between 2 and 4]106 yr in a retarded collapse 2H5OH\10~8 and of (CH3)2O ca. 10~6 (Fig. 5). mantles build up and surface chemistry builds complex species (Fig. 6). For t[4]106 yr thermal desorption liberates the mantles and returns material to the gas phase giving fractional abundances of C The calculations explicitly follow the gas and surface abundances and the exchange of material as a function of time so one can also obtain detailed information on the depth dependence of the composition of the ice mantle which in this case amounts to 28 Chemical models of hot molecular cores some 130 layers of material.Fig. 7 shows the percentage contribution to each layer for a variety of species as well as the variation of number of layers as a function of time. The dramatic increase of the number of layers at late times is a direct consequence of the retarded free-fall collapse for which the timescale to collapse to an in–nite density is proportional to n~1@2. Thus at low number densities the cloud evolves slowly. Once it reaches high density it both accretes material rapidly and collapses quickly. At early times the mantle is mainly composed of H2O since O atoms are the most abundant abundant CO is formed in the gas and accreted. The fractional abundance of *CH OH species colliding with the grains.At late times the contribution of H2O decreases as *C 3 2H5OH *(CH3)2O and together with their percentage contribution to the mantle are given in Table 6 for the model considered here. We note that on evaporation this would provide fractional abundances of CH OH C2H5OH and (CH3)2O 3 compatible with those observed (Table 5). in the gas phase Conclusions Our molecular line survey shows that hot molecular cores associated with UCHII regions can be rich sources of molecular emission with densities and temperatures much larger than is typical for molecular clouds. The chemical composition of the hot gas re—ects both the direct and indirect products of evaporated mantle ices and we have been able to use observations of sulfur-bearing molecules to constrain the ages of Fig.6 The evolution of fractional abundances of various grain mantle species between 2 and 4]106 yr of a retarded collapse. For times greater than 4]106 yr the mantles have evaporated. 29 T . J. Millar and J. Hatchell Fig. 7 The evolution of the number of mantle layers and the percentage contribution of various species to the mantle composition as a function of mantle layer number HMCs. In addition our observations of the DCN HCN abundance ratio shows evidence for the importance of the destruction of DCN by atomic hydrogen and allows inferences to be drawn about the likely activation energy barrier in this reaction. We have also reported on the results of detailed chemical kinetic models of hot cores and of the gas»grain interaction and surface chemistry in collapsing cores.Although there are several uncertainties in this latter model including the details of the surface chemistry in particular the possible presence of activation energy barriers stoichiometric eÜects binding energies and mobilities in realistic ices the neglect of a more comprehensive reaction network and the sensitivity to the details of the collapse since this determines the abundances of gas phase species which collide with the grains it is 30 *(CH3)2O % time/yr abundance 0.006 1.05]10~8 4.16]106 Chemical models of hot molecular cores Table 6 Calculated fractional abundances and percentage contribution of complex organic molecules in the ice mantle at the time of maximum mantle depth *C2H5OH *CH3OH % abundance % abundance 0.006 4.02 1.00]10~8 6.74]10~6 encouraging that a set of models does emerge for which reasonable agreement between both surface abundances when compared to those observed through IR absorption of protostellar sources and cometary ices which appear to be very similar to interstellar ices can be found.It appears that surface chemistry in cold interstellar ices is capable of producing a range of complex organic molecules. Astrophysics at UMIST is supported by a grant from PPARC. We are grateful to P. F. Hall G. H. Macdonald S. D. Rodgers and M. A. Thompson for collaborating in some of the work reported in this paper. Paper 8/00127H; Received 5th January 1998 References 1 J. Hatchell M. A. Thompson T. J. Millar and G. H.Macdonald Astron. Astrophys. Suppl. Ser. 1998 in press. 2 B. E. Turner Astrophys. J. Suppl. Ser. 1991 76 617. 3 R. Cesaroni C. M. Walmsley and E. Churchwell Astron. Astrophys. 1992 256 618. 4 S. B. Charnley Astrophys. J. 1997 481 396. 5 W. G. Mallard F. Westley J. T. Herron R. F. Hampson and D. H. Frizell NIST Chemical Kinetics Database V ersion 6.0 National Institute of Standards and Technology Gaithersburg MD 1994. 6 J. Hatchell T. J. Millar and S. D. Rodgers Astron. Astrophys. 1998 332 695. 7 G. Dahmen T. L. Wilson and F. Matteucci Astron. Astrophys. 1995 295 194. 8 A. G. Gibb personal communication. 9 A. Wootten in Astrochemistry ed. M. S. Vardya and S. P. Tarafdar Reidel Dordrecht 1987 p. 311. 10 P. Schilke G. P. Des Fo� rets E. RoueÜ D. R. Flower and S. Guilloteau Astron. Astrophys. 1998 256 595. 11 T. J. Millar A. Bennett and E. Herbst Astrophys. J. 1989 340 960. 12 A. G. G. M. Tielens Astron. Astrophys. 1983 119 177. 13 P. D. Brown and T. J. Millar Mon. Not. R. Astron. Soc. 1989 237 661. 14 P. D. Brown and T. J. Millar Mon. Not. R. Astron. Soc. 1989 240 25P. 15 B. E. Turner Astrophys. J. 1989 347 L39. 16 S. D. Rodgers and T. J. Millar Mon. Not. R. Astron. Soc. 1996 280 1046. 17 P. D. Brown S. B. Charnley and T. J. Millar Mon. Not. R. Astron. Soc. 1988 231 409. 18 T. J. Millar E. Herbst and S. B. Charnley Astrophys. J. 1991 369 147. 19 S. B. Charnley A. G. G. M. Tielens and T. J. Millar Astrophys. J. 1992 399 L71. 20 P. Caselli T. I. Hasagawa and E. Herbst Astrophys. J. 1993 408 548. 21 D. D. S. MacKay Mon. Not. R. Astron. Soc. 1995 274 694. 22 S. B. Charnley M. E. Kress A. G. G. M. Tielens and T. J. Millar Astrophys. J. 1995 448 232. 23 S. B. Charnley and T. J. Millar Mon. Not. R. Astron. Soc. 1994 270 570. 24 G. H. Macdonald A. G. Gibb R. J. Habing and T. J. Millar Astron. Astrophys. Suppl. Ser. 1996 119 333. 25 T. J. Millar G. H. Macdonald and A. G. Gibb Astron. Astrophys. 1997 325 1163. 26 T. J. Millar G. H. Macdonald and R. J. Habing Mon. Not. R. Astron. Soc. 1995 273 25. 27 P. F. Hall PhD Thesis UMIST 1997. 28 P. F. Hall and T. J. Millar 1998 in preparation. 29 M. Ohishi 1997 personal communication. 30 T. I. Hasagawa E. Herbst and C. M. Leung Astrophys. J. Suppl. Ser. 1992 82 167. 31 T. I. Hasagawa and E. Herbst Mon. Not. R. Astron. Soc. 1993 261 83. 32 M. Mumma et al. Science 1996 272 134

ISSN:1359-6640    

DOI:10.1039/a800127h

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

Faraday Discuss. 1998 109 31»46 What can ISO tell us about gasñgrain chemistry? Ewine F. van Dishoeck L eiden Observatory PO Box 9513 NL »2300 RA L eiden T he Netherlands Recent results of searches for IR absorption lines of gas-phase molecules in the ISO spectra of embedded massive young stars are discussed. Abundant highly excited gas-phase H2O is detected in the ìhot coresœ surrounding the young stellar objects but not in the colder regions. The abundance of gasphase CO is low in all sources in spite of the large observed abundance of solid CO 2 Gas-phase 2. C4 C2H2 H and HCN have been observed as well. The latter two species have high excitation temperatures of the order of 1000 K in some sources. The large abundance of CH suggests that the molecule is formed mainly through grain-surface reactions.CO may be 4 produced by grain-surface chemistry as well athough photochemical pro- 2 duction in ices can also play a role. Comparison of gas-phase with solidstate column densities for the same lines of sight allows accurate determination of gas solid state ratios which can be analyzed as functions of the temperature of the region. The data indicate that signi–cant evaporation of ices likely occurs in the hotter regions but that part of the observed gas-phase H2O may also be produced by high-temperature gasphase reactions. The results are discussed in the context of the physical evolution of the regions. 2 .9 1 Introduction Much of interstellar chemistry in the last 25 years has been concerned with gas-phase molecules which are readily detected at millimeter wavelengths.Detailed models have been developed to reproduce the observed abundances starting with the early work of Bates and Spitzer1 and Herbst and Klemperer.2 The most recent networks contain over 4000 gas-phase reactions between a few hundred species.3,4 Although it was recognized at an early stage that reactions on grain surfaces can also be signi–cant,5h8 they have been largely neglected in models except for the formation of H In the last decade several models have been developed which take account of both the gas-phase and the grain-surface processes.10h12 These have been stimulated by two sets of observational data. First owing to the rapid improvements in the sensitivity of IR detectors ground»based and airborne observations have revealed absorption of solid H CO CH3OH and other species along the lines of sight to bright IR sources.13,14 Second (sub-)millimeter emission data on complex saturated organic molecules such as 2O CH OCH3 HCOOCH3 and CH3CN in so-called ìhot coresœ around massive young 3 stellar objects15,16 have stimulated models in which grain mantles are evaporated into the gas at high temperatures and drive a rapid gas-phase chemistry.17,18 The short wavelength spectrometer (SWS)19 on board the IR space observatory (ISO)20 is particularly well-suited to address several aspects of gas»grain interactions in interstellar chemistry.First the ISO»SWS allows a complete inventory of the solid-state species in a variety of regions through full spectral scans from 2.5 to 45 lm at a resolving power R\j/*jB500 unhindered by the atmosphere.21h23 Secondly ISO can observe the vibration»rotation absorption lines of gas-phase species such as H2O CO2 31 32 ISO and gas»grain chemistry 2 Fig.1 Normalized ISO-SWS spectra in the wavelength region of the CO2 l3 (0,0,1)»(0,0,0) and CO 1»0 vibrational bands toward three embedded young stellar objects. The CO is primarily in solid form as revealed by its broad line shape. Toward NGC 7538 IRS1 CO is also mostly in solid form but toward W 3 IRS5 the ro-vibrational structure of gaseous CO is seen as well. Toward GL 4176 only highly excited gaseous CO is observed. The spectra have been shifted by 0.0 [0.8 and [1.8 for clarity. and CH which are thought to be among the dominant products of gas»grain chem- 4 istry but which cannot be observed from Earth or only with great difficulty.24h27 In addition accurate gas solid ratios can be determined for these species providing important constraints on the evaporation mechanisms and various aspects of ìhot-coreœ chemistry.Finally ISO allows more extensive searches for minor solid-state species especially nitrogen-containing molecules which may be produced by photoprocessing of ices. Together with observations of CO2 they can be used as diagnostics of the radiation –eld and give information on the relative importance of grain-surface reactions compared with photochemical reactions in ices. In this paper an overview of recent results with the ISO»SWS on gas-phase and solid-state species in massive star-forming regions is presented together with new observational data for several lines of sight.2 IR absorption vs. submillimeter emission observations The majority of the nearly 120 interstellar molecules identi–ed to date28,29 have been detected through (sub-)millimeter observations of pure rotational lines in chemically rich sources like TMC»1 Orion-IRc2/KL SgrB2(N) and IRC]10216. This technique has the advantage of very high spectral resolution (RP106) so that the line pro–les are 33 E. F. van Dishoeck resolved. In addition very small abundances down to 10~11 with respect to H2 can be detected and mapped. On the other hand IR observations have many advantages compared with millimeter data.30h32 First they probe the absorption of material along a pencil beam line of sight to the IR source (\1A) whereas the millimeter emission line data often refer to beams of at least 20A.Secondly the population distribution over the rotational energy levels can be directly constrained from a single IR vibration»rotation spectrum whereas often diÜerent receivers or telescopes are needed to determine the rotational excitation from submillimeter data. Also information on much higher energy levels is obtained up to J[20 for species like CO and HCN compared with J\7»9 from submillimeter data. The excitation temperature gives important information on the physical parameters of the gas where the molecule is located. Thirdly molecules without a dipole moment such as CO have strong IR vibration»rotation transitions 2 CH4 and C but negligible millimeter rotational emission.Finally not only gas-phase molecules but 2H2 also solid-state species can be detected at IR wavelengths. A vibrational band of a molecule in the solid-phase can be readily distinguished from that in the gas-phase because the former consists of a single broad spectral feature which lacks the characteristic rovibrational structure of the gas-phase spectrum and is slightly shifted in wavelength (see Fig. 1).33 The main drawback of the ISO data is their limited spectral resolution RB2000 for the grating mode on the SWS. This implies that the lines are unresolved and often optically thick when detected. In practice only abundances down to ca. 10~7 with respect to H can be probed and only lines with widths *V [2 km s~1 can be seen.2 3 Gasñgrain chemistry open issues 3.1 Basic processes The chemistry on the surfaces of interstellar grains has received ample discussion in the literature.7,8,11,12,34h36 Four diÜerent steps can be distinguished in the formation of a molecule (i) accretion ; (ii) diÜusion; (iii) reaction ; and (iv) ejection or evaporation. The timescale for a molecule to collide with a grain and accrete is given by tacc\2 ] y 109/nH yS yr where the sticking probability is thought to lie between 0.1 and 1.0 for S most species.37 Thus at typical densities in dense clouds of 104 cm~3 or more most species (except H and He) are expected to be condensed onto the grains on timescales 2 of less than 105 yr unless efficient desorption occurs.The residence time on the surface depends on the desorption mechanisms. For t is short for H (500 s) but very long ([1015 yr) t thermal evaporation the timescale TD\10 K. Thus all neutral species heavier than He can be con- for C and CH at evap 4 T sidered permanent residents at D\10 K. Once stuck on the grain a species can hop to the next site on a timescale thop which is signi–cantly shorter than the evaporation heavier species are immobile. H and H are so light that they can also tunnel quantum timescale. Light species such as H H2 C N and O can hop from site to site whereas mechanically to the next site. A species can scan the complete surface for a reaction 2 partner on a timescale scanB105thop determined by a random walk process.The efficiencies of the various reactions depend on relative timescales for accretion diÜusion and desorption. Two diÜerent regimes can be distinguished. In the ìaccretion limitedœ regime thop>tacc so that a species can diÜuse on the surface until it –nds a co-reactant. The chemistry is limited by the accretion rate of new species. In the ìreaction limitedœ regime the opposite holds thop?tacc so that a species trapped in a site can only react with migrating species that visit that site. At a typical density of 104 cm~3 a grain accretes only a few atoms or molecules per day. These species thus have a long time namely a day to scan the surface and react before a new species arrives, 34 ISO and gas»grain chemistry indicating that the chemistry is in the ìaccretion limitedœ regime.7 Many of the published gas»grain models have been formulated in the ìreaction limitedœ regime using rate equations for computational convenience.34 The rate equations have recently been reformulated to remedy their shortcomings,38 but the detailed eÜects on published models still remain to be assessed.Molecules can be returned to the gas phase by a variety of mechanisms.35,37,39,40 Most relevant for this paper is the thermal evaporation which is efficient only at higher T temperatures D[20 K and depends on the species involved. The sublimation temperatures of pure CO CO2 CH4 and H2O»ice under interstellar conditions are 20 45 20 and 90 K respectively.41 For traces (\5%) of CH and CO embedded in an H2O- 4 these temperatures increase to ca.90 K. Other desorption mechanisms in cold matrix 2 clouds include cosmic-ray spot heating42 and explosive desorption. The latter can be either cosmic-ray induced or triggered by grain»grain collisions at velocities greater than ca. 0.1 km s~1. In star-forming regions sputtering of icy mantles by shocks in the turbulent boundary layers of out—ows can be important as well. Most mechanisms are eÜective only for ìapolarœ ices containing CO O and N2 H but less for polar ices 2 which contain strong hydrogen bonds. Thus the desorption processes can shape the icy 2O-rich mantles into polar and apolar layers. Only sputtering in strong shocks is likely to remove the mantles completely. 3.2 Model predictions 3 TDB10 K. However in these experiments the atomic O was produced by 2 Because hydrogen is so abundant and mobile all models predict that grain surface chemistry leads primarily to hydrogenated species under hydrogen-rich conditions.Speci –cally hydrogen has enough time to tunnel through activation barriers as large as ca. 2000 K so that H2O NH3 and CH can be formed. Indeed the observed large abun- 4 dances of H2O ice in dark clouds have been argued to be strong evidence of its production by grain surface reactions.43 Hydrogenation of CO can lead to H2CO and CH OH although the efficiencies of these reactions are still subject of discussion.12 At higher densities the amount of atomic hydrogen in the gas phase decreases and reactions with atomic oxygen become important. The long accretion timescale allows reactions with barriers up to ca.400 K to proceed through thermal hopping. A particularly important case is the reaction of CO with O to form CO2 . Early laboratory experiments44 seemed to indicate that this reaction does not proceed at low temperatures photolysis of O2. O Since the was still abundant during the experiments most of the O 2 reacted with O to form O3 . Additional experiments of CO with O in the absence of abundant O are urgently needed. 2 Another process for producing new molecules is through photochemical reactions within the icy mantles. The UV photons can be provided by the external interstellar radiation –eld by the radiation from the young embedded star45 and by the interaction of cosmic rays with H resulting in UV photons.46 Photodissociation of molecules in ices produces radicals (e.g.H2O]OH]H O or ]H2) which can subsequently react 2 to form other molecules. This provides an important alternative route to the formation of solid CO through the CO]OH reaction. Indeed this process is well known from laboratory experiments on H2O»CO ice mixtures just a small amount of radiation 2 results in signi–cant CO production.47 Observations of solid CO in a variety of 2 regions may be able to distinguish between these two possible routes. 2 3.3 Hot core chemistry Once the molecules have been desorbed from the grains at high temperatures they drive a rapid chemistry in the gas phase. This so-called ìhot coreœ chemistry was –rst discussed 35 E. F. van Dishoeck can be reproduced if a mixture of simple ices containing3 H for the Orion hot core from which it derives its name.Recent models have shown that the observed abundances of complex organic molecules such as CH OCH and HCOOCH 3 2O CO 3 CH OH NH and/or HCN is evaporated into the hot gas.17,18,48h50 Reactions with 3 3 CH produce complex organic molecules on timescales of ca. 104 yr. In 3OH H2CO and addition the high temperature in the gas drives atomic oxygen into H2O through the 2 ]OH and OH]H2 ]H2O reactions at temperatures above 230 K.48,51 4 O]H Other molecules such as CO and CH do not participate in an active chemistry and are 2 destroyed on timescales of ca. 105 yr by normal ion»molecule and neutral»neutral reactions. Within 104 yr after evaporation the original composition of the ices is still re—ected in the gas-phase composition of the hot cores.3.4 Scenario for gas-grain interactions during star-formation IR absorption line observations are limited by the availability of a bright source located behind a large column of gas and dust. In practice most of the bright IR sources are deeply embedded massive young stars. The IR continuum radiation at 4»25 lm is due to emission from hot IR observations concern warm star-forming regions rather than cold quiescent clouds. (TDB200 K) dust located close to the young star. Thus most of the The processes discussed above lead to the following scenario for gas»grain inter- (TDB10 K) that most actions at various stages of the star-formation process.12,14,52h56 During the collapse phase the density increases and the temperature stays so low molecules freeze out onto the cold grains.Here the chemistry can be actively modi–ed by surface reactions resulting most likely in hydrogenation and oxidation of O C N and CO leading to and H2O CH4 NH3 CO2 CH3OH H2CO. After the new star has formed its radiation heats the surroundings (TDB20»100 K) and the molecules start to evaporate back into the gas phase probably in a sequence according to their sublimation temperatures. In addition the out—ows create shocks when they interact with the surrounding envelope which can drive high-temperature chemical reactions and return icy mantles and more refractory material containing silicon to the gas phase. These freshly evaporated molecules can then drive a ìhot-coreœ chemistry for a period of ca.105 yr. Finally the chemistry returns to the quiescent phase dominated by ion»molecule reactions. What can ISO contribute to this picture ? First it can con–rm that the major products of grain-surface chemistry under hydrogen-rich conditions are indeed H and NH through direct observation of the ices. Second constraints can be obtained on 2O CH4 the relative importance of grain-surface reactions compared with photochemical pro- 3 cesses in ices. Third information on the temperature structure and history of the region can be derived through determination of the various ice components (polar vs. apolar). Fourth the mechanisms for releasing molecules from the grains back into the gas can be tested by observing gas solid ratios for various species.Finally direct observation of gas-phase molecules in hot cores provides information on species which are evaporating from the ices and those produced in the gas by subsequent high-temperature reactions. This paper will address most of these issues except the diÜerent ice components which are discussed extensively in the paper by Ehrenfreund et al.23 4 ISO observations ISO observations were performed primarily with the SWS,19 which operates from 2.5 to 45 lm. The SWS06 grating mode was used which has a resolving power R ranging from 1350 to 2500. The aperture varies from 14A]20A at 2.4»12 lm up to 20A]33A at the longest wavelengths. Details of the data reduction and procedures for the removal of instrumental fringes are discussed elsewhere.26,57 36 ISO and gas»grain chemistry Searches for H2O lines were also performed with the long wavelength spectrometer, 58 which covers the wavelength range from 43 to 197 lm.Both the low-resolution grating mode LWS01 (RB500) and the high»resolution Fabry»Perot mode LWS04 (RB8000) were used. The LWS aperture is about 80»100A. Details of the reduction can be found in ref. 59. 5 Gas-phase molecules 5.1 Gas-phase CO High-resolution spectra of gas-phase 12CO and 13CO toward a number of massive young stellar objects have been obtained previously from the ground by Mitchell et al.30 Lines originating from levels up to J\24 have been detected indicating the presence of both warm (T \200»1000 K) and cold (T \60 K) gas along the lines of sight.Accurate CO column densities have been derived from the optically thin 13CO data which are important to constrain the total amount of H2 . The fraction of warm gas N the other sources. Thus most sources have already heated a substantial fraction of their warm(H2)/Ntot(H2) ranges from less than 5% for NGC 7538 IRS 9 to typically 50% for envelope. Although hampered by low spectral resolution gas-phase CO can also be detected with the ISO»SWS. Fig. 1 shows spectra in the region of the CO v\1»0 vibrational band at 4.68 lm. ISO is mostly sensitive to the warm gas with broad lines (*V [2 km s~1) which is clearly seen up to J[20 in the P- and R-branches in objects like GL 4176. These data provide information on the fraction of warm gas along the line of sight which is especially useful for sources which have not been observed previously from the ground (e.g.sources in the southern sky). 5.2 Hot abundant gas-phase water Helmich et al.24 and van Dishoeck and Helmich60 presented the –rst detection of IR absorption lines within the bending vibration of water at 6.2 lm toward four massive young stars. The observations are shown in Fig. 2 together with new data on three additional objects using the SWS06 grating mode. The strongest lines absorb 5»15% of the continuum. very diÜerent from that of a linear molecule such as CO and a detailed model is Because of its complex energy level structure the ro-vibrational spectrum of H2O is required to interpret the data. In its simplest form it is assumed that the water level populations can be characterized by a single excitation temperature which should be close to the kinetic temperature if collisions dominate the excitation.Alternatively the rotational level populations of the H2O molecule are readily coupled to the radiation –eld through near- and far-IR pumping in which case the excitation temperature is representative of the color temperature of the radiation –eld. In most regions a mixture of the two processes likely occurs. Under the assumption of LTE simulated spectra can be constructed in which the total H2O column density N the excitation temperature Tex and the Doppler parameter b\2)(ln2)*V are the only input parameters. Details of the method and references to the adopted molecular data can be found in ref.61.61 N(H2O)\2]1018 cm~2 Tex\300 K and Fig. 2 includes a model spectrum for b\5 km s~1. It is seen that this spectrum provides a good –t to the observations of the warm sources GL 2591 GL 2136 and GL 4176. However toward W 3 IRS5 S140 IRS1 and NGC 7538 IRS 1 at most a few H2O lines arising from the lowest energy levels are seen whereas no H2O is detected in the spectrum toward NGC 7538 IRS9. For these sources no constraints on the excitation temperature can be obtained. Gas-phase H2O has also been detected toward GL 7009S but the excitation temperature in this case appears to be low TexB25 K.27 Fig. 2 ISO-SWS normalized spectra of seven massive protostellar sources in the wavelength region of the H2O l2 bending mode showing absorption by hot abundant H2O in at least three sources.Model H N(H2O)\2]1018 cm~2 Tex\300 K and 2O spectra for a column density of Doppler parameter b\5 km s~1 and N(H are shown for comparison. 2O)\1]1018 The derived H2O column densities are summarized in Table 1. The results are accurate to a factor of two and are not sensitive to the adopted values of b and Tex as long as bZ2 km s~1. Since ISO cannot detect very narrow lines with b\1.5 km s~1 large amounts of cold H2O could potentially still be present. However there is no evidence for such narrow lines in submillimeter emission data of various species,62 nor in the 13CO IR absorption lines.30 Because of the low velocity resolution it is also not possible to distinguish whether the absorption occurs at or near the velocities of the cloud core or in the blue-shifted gas.Blue-shifted absorption due to high-velocity gas in the out—ow has been detected in several sources,30 but the fraction of the total column density Table 1 Gas-phase column densities toward massive young stellar objects object NGC 7538 IRS9 W 33A NGC 7538 IRS1 S140 IRS1 W3 IRS5 GL 2136 GL 4176 GL 2591 a From 13CO observations (warm]cold),30 assuming 12CO/13CO\60 and 12CO/H2\2]10~4.63 K. c Derived using b\3»7 km s~1 and Tex\100»300 K. d Results for Tex\100»250 2 e x NGC 7538 IRS9 and W 33A from ref. 26. warm(H2)\Nwarm(H2)/Ntot(H2) is the fraction of hot gas along the line of sight. f Derived from C17O 2»1 SEST spectrum and silicate optical depth.60 E. F. van Dishoeck H COa CO2 b 2Oc \3(17) 3(17) B1(17) 8(15) » 1(16) 2(15) 1(16) 2(16) 9.8(18) 2.6(19) 1.7(19) 7.4(18) 2.6(19) 2.2(19) B1(17) 3(17) 2(18) 2(18) 2(18) 1(16) 2(16) 1.6(19)f 1.9(19) b Derived from the l Q-branch using b\3»7 km s~1 and 37 cm~2 Tex\25 K and b\2 km s~1 CH4 d xwarm e H2 a 0.02 0.53 0.48 0.6 0.48 0.68 4.9(22) 1.3(23) 8.6(22) 3.7(22) 1.3(23) 1.1(23) 5(16) 1(17) 5(16) » » [5(16) [0.5 0.63 8.0(22)f 9.6(22) » [5(16) 38 ISO and gas»grain chemistry 2 l2 bending mode showing absorption by gas-phase CO2 in all sources.The Fig. 3 ISO»SWS normalized spectra of eight massive protostellar sources in the wavelength region of the CO dotted line indicates the position of the Q-branch.The observations have not been corrected for the (small) shifts due to the velocity of the source. contained in this component is only a few percent. If all of the observed H2O would be located in this component it would account for more than the interstellar oxygen abundance in the hottest sources. The H2O H abundances can be determined by comparison with the column den- 2 sities derived from the 13CO data along the same lines of sight assuming 12CO/ 13CO\60 and 12CO/H2\2]10~4.63 Table 1 lists both the total H column density 2 and the fraction of warm gas. For GL 2591 GL 2136 and GL 4176 the resulting H2O abundances range from (2»3)]10~5 if the H2O is homogeneously distributed throughout the source to (3»5)]10~5 if H2O is present only in the warm gas.In the latter case a signi–cant fraction ([5%) of the available oxygen is tied up in H2O. In the other sources the gas-phase H2O abundance integrated over the total line of sight is typically 1]10~6 and at most 5]10~6 in the warm gas. In order to further constrain the H2O abundance and excitation searches for the pure-rotational lines of H2O in emission or absorption have been made with the LWS toward several sources which show strong H2O 6 lm absorption.59 The lack of detected the beam but is present only in the small inner ìhot coreœ region which is typically only emission lines in GL 2591 GL 2136 and W 3 IRS5 suggests that the H2O does not –ll ca. 1016 cm in size (\1000 AU 1A at 1 kpc).CO2 5.3 Gas-phase CO and CH 2 4 l Searches for gas-phase CO l have been made in the asymmetric stretch and 2 3 2 bending modes at 4.3 and 15 lm.25 As Fig. 1 shows the 4.26 lm band is dominated by the strong solid CO feature and no hint of the ro-vibrational P- and R-branch structure of gas-phase N(CO2) \5]1016 is evident. This provides upper limits of 2 39 E. F. van Dishoeck CO2 l2 bending mode is less saturated and a weak sharp superposed 2 2 cm~2 corresponding to surprisingly low CO abundances of less than 5]10~7. The 15.2 lm solid absorption due to the Q-branch of gas-phase CO at 14.98 lm has been detected.25 In Fig. 3 new ISO data toward four additional sources are presented. The inferred column 2 densities are presented in Table 1 and are up to a factor of two higher than derived previously25 due to a better determination of the continuum.In all sources gas-phase CO is detected but at low abundances of (1»2)]10~7. Even toward Orion IRc2 a 2 similarly low value is found.64 No stringent limits on the excitation temperature can be T obtained although the best –ts to the pro–les are obtained for exB200 K or lower. The presence of gas-phase CO had been suggested indirectly from millimeter obser- 2 vations of the HOCO` ion,65 but except for the SgrB2 region these data also suggested low CO abundances of less than 10~6. Gas-phase CH has been identi–ed by Lacy et al.66 but the quality of the groundbased data is severely hindered by the Earthœs atmosphere. Although the ISO-SWS has 4 lower spectral resolution careful reduction of the data in the 7»8 lm region reveals the presence of gas-phase CH toward several sources (Fig.4).26,27,57 The resulting column densities are summarized in Table 1. The abundance of gas-phase CH with respect to 4 4 total H is ca. (0.5»1)]10~6. Comparison with simulated spectra indicates that the 2 CH gas is warm with an excitation temperature of 100^30 K. 4 and HCN and HCN. Lines of these species have been observed from and HCN may even be present,64 consistent with the 5.4 Gas-phase C2H2 The ISO spectra have also been inspected for the presence of a number of less abundant molecules including C2H2 the ground,31,32,67 but only in a few sources. Clear detections of the C2H2 l5 Q-branch at 13.7 l lm and the HCN Q-branch at 14.0 lm have been made in the ISO spectra of 2 half our sources (see Fig.5).57 For GL 2136 and GL 2591 the pro–les are very broad and high excitation temperatures are inferred of the order of 1000»1500 K. Absorption from vibrationally excited C2H2 K. The thick vertical lines above the spectra indicate the optical depths of the R(0) and R(2) Fig. 4 ISO-SWS spectrum toward the massive protostellar objects NGC 7538 IRS9 (top) and W33A (middle) showing the detection of solid CH (broad feature) with gas-phase CH lines CH 4 4 superposed.26 The lower –gure is a model gas-phase 4 100 spectrum for N\1017 cm~2 and Tex\ lines measured in ref. 66. 40 ISO and gas»grain chemistry 5 and the l Q-branch of HCN. The thin dashed line indicates the best –tting single com- N(C2H2)\2]1016 cm~2 Tex(C2H2)\1000 K and ex Fig.5 ISO-SWS normalized spectrum of GL 2136 showing the detection of the l Q-branch of C2H2 ponent model spectra for 2 N(HCN)\6.5]1016 cm~2 K T (HCN)\1500.57 Some of the remaining structure may be due to absorption from vibrationally excited C2H2 and HCN.64 detection of absorption from vibrationally excited CO in the same sources.30 Typical abundances with respect to total H are x(C2H2)B2]10~7 and x(HCN)B6]10~7 for the warm sources. For HCN the abundances are more than an order of magnitude 2 larger than obtained from submillimeter data suggesting again that the hot gas occupies only a small volume. 6 Interstellar ices The availability of complete spectral coverage from 2»45 lm with the ISO»SWS provides a landmark for the study of interstellar ices and allows for the –rst time an unbiased inventory.Overviews of recent results are given in several papers.14,21,22,68,69 Here we summarize only the conclusions for those species for which the gas-phase counterparts have been detected. A striking ISO result is the presence of strong solid CO l stretching mode at 4.27 lm (see Fig. 1) and the l bending absorption in the 2 3 2 mode at 15.2 lm along all lines of sight to embedded young stellar objects.22,70h72 The overall CO abundance in ices is ca. 15% relative to H2O making it one of the most abundant solid-state molecules. This corresponds to an abundance with respect to gas- 2 phase H of ca. 3]10~6»2]10~5 nearly two orders of magnitude larger than that of gas-phase CO 2 2.H2O Information on the amount of ice is available from ground-based observations of the 3 lm band.73 Additional measurements of the less saturated 6.0 lm bending mode allow a more accurate determination in some sources.74 Solid CO has been detected toward the colder sources (see Fig. 1).75 As discussed by Ehrenfreund et al.,23 the solid-state observations provide not only information on abundances but also on the ice environment since the shape and position of the bands are sensitive to the interactions of the molecules with their neighbors. Analysis of the pro–le shapes using laboratory data indicates the presence of grain mantles with distinct polar (H2O-rich) and non-polar (CO O2 and/or N -rich) 2 layers.75,76 CO appears to be present in both phases and may be ì segregatedœ or ìannealedœ in the warmer sources.2 41 E. F. van Dishoeck 2O C2 ice is about 15% similar as that toward young stellar objects andOis pri- 2 . Solid CO has also recently been detected toward Elias 16 a –eld star behind a 2 quiescent part of the Taurus molecular cloud.77 The derived abundance with respect to H marily in the H2O-rich phase. This important result indicates that no embedded energy source is needed for the production of solid CO An absorption feature near 7.67 lm has been detected in the ISO spectra toward 4 4 . is ca. 1% [ca. (1»2)]10~6 with respect to total H is also embedded in the polar H2O-rich ices. three deeply embedded objects (see Fig. 4).26,27,78 Comparison with laboratory spectra shows that it can be identi–ed with the l deformation mode of solid CH Its abundance with respect to H2O-ice 2] and CH Limits on the amount of solid C2H2 of 1»10% with respect to solid H2O have been 4 derived toward NGC 7538 IRS9.79 Because solid C2H2 does not have any strong bands when embedded in H2O-ice the limits are not very stringent.Information on the amount of solid HCN can be derived using recent laboratory data,80 giving typical upper limits of 3% with respect to H2O ice.81 7 ISOœs view on gasñgrain interactions 7.1 Hydrogenated molecules 3. H2O The high abundances of solid in dense clouds certainly support this As discussed in Section 3 one of the main predictions of grain-surface chemistry is the production of large amounts of hydrogenated species in particular H2O CH4 and NH theory.43 The observed abundances of gas]solid CH of ca.2]10~6 with ISO also 4 argue in favor of grain-surface chemistry.26 Time-dependent pure gas-phase chemistry models starting from a diÜuse atomic composition can approach such abundances but only for a very narrow time interval at tB105 yr.3,4 On the other hand conversion of C to CH on the grains is very efficient. The measured abundances suggest that the initial 4 atomic C abundance in the gas must have been low C/COB0.01 indicating that the solid CH was formed during a cold dense cloud phase. The fact that it is embedded in 4 an H2O-rich matrix suggests that the two were formed simultaneously. In contrast with CH4 C2H2 is thought to be produced primarily in the gas phase and then passively accreted onto grains.32 abundances measured in hot cores.14,83,84 The N/N ratio must have been NH 2 No stringent limits on the amount of solid NH are yet available from ISO because its bands are either blended with other strong features or occur in the deep silicate 3 feature where the signal-to-noise ratio of the ISO data is low.Ground»based limits are NH of the order of 5% with respect to solid H2O,82 which is comparable to the gas-phase 3/H2O low \0.1 consistent with the dense collapse phase. Gas-phase has been observed 3 toward GL 2591,85 but the deduced abundance in the 40A beam is low \10~8. Note that the observed abundances of NH in diÜuse clouds have also been cited as evidence of grain surface chemistry.86 Hydrogenation of solid CO may lead to solid and H2CO CH3OH.The latter species is known to be present in substantial abundances of a few percent with respect to H2Oice in these objects.87 Solid formaldehyde has possibly been seen in a few sources at a similar abundance.22,88 These abundances are larger than can be explained by pure gas-phase chemistry and subsequent freeze-out onto the grains suggesting grain-surface formation. 2 7.2 Solid CO2 The observed solid CO abundances of a few ]10~6 to 10~5 are more than an order of magnitude larger than those predicted from pure gas-phase chemistry.3,4 As discussed in 42 ISO and gas»grain chemistry 2 .89 Section 3.2 solid CO may be formed either by surface reactions of CO]O or through 2O»CO ice mixtures where the CO reacts with the O or OH liberated by photolysis of H 2 the photodissociation of other molecules.Can the ISO data distinguish between these two possibilities ? The presence of CO in grain mantles toward the –eld star Elias 16 in Taurus indi- 2 cates that no embedded source of luminosity is required to produce the molecule.77 However other sources of ultraviolet photons are available as well including the external interstellar radiation –eld and the cosmic ray induced photons. Quantitative estimates using laboratory data indicate that the latter route is insufficient to account for the observed abundance of solid CO2 .77 The total visual extinction AV\21 magnitudes of the cloud also appears to preclude the external interstellar radiation –eld.However if the cloud has an inhomogeneous structure and the typical A at any location along the path is about 5 mag this radiation may be sufficient. Alternatively ion bombardment of V ices by cosmic-ray particles may trigger the formation of solid-CO The current data cannot yet fully determine the relative importance of grain surface formation of CO compared with production through photolysis. Additional laboratory 2 experiments on the CO]O surface reaction as well as observations of other molecules which are thought to be produced through energetic processing such as ìXCNœ,90 NO2 and N2O are needed in a variety of sources. Note that XCN has not been detected toward Elias 16.90 7.3 Gas/solid-state abundance ratios In Table 2 the gas/solid state ratios derived from the ISO data for the major species CO CO2 H2O CH4 and are summarized.The temperatures listed in the last column refer to those derived from the gas-phase CO excitation.30 In all eight sources CO is principally in the gas phase although the gas/solid ratio still increases for the warmer sources (see Fig. 1). In contrast CO is primarily in the solid phase even in the warmest 2 regions. The gas/solid H2O ratio varies from less than 5% for NGC 7538 IRS9 and W 33A to more than unity for GL 2591 and GL 4176. The gas/solid CH ratio of 0.5 4 derived for the cold sources is considerably higher than that of O H2O C and but lower than that of CO. The gas/solid ratio of HCN is greater than 0.1 in sources in 2 which gas-phase HCN has been detected while no stringent limits are obtained for gas/solid C2H2 .What do these ratios imply for the source structure and evolution ? It should be recalled that the observed gas/solid ratios are averaged along the line of sight to the star. In reality strong gradients in the temperature and in the abundances of these species Table 2 Gas/solid state abundance ratiosa CO object CO2 H2O CH4 Twarm b/K 0.01 » 0.02 12 43 100 0.4 0.5 [1 180 120 176 390 577 580 \0.04 0.01 0.03 0.04 0.06 0.4 NGC 7538 IRS9 W 33A NGC 7538 IRS1 S140 IRS1 W 3 IRS5 GL 2136 » [ »1 0.01 0.01 0.03 0.08 0.07 121 163 122 [300 380 2.2 1.2 GL 4176 GL 2591 [500 B1000c [ »1 a Using the gas-phase column densities from Table 1 solid CO and H column densities from ref.70 and 75 solid CO from ref. 70 and 72 and 2O 2 solid CH from ref. 26. b From gas-phase CO excitation.30 c Two warm 4 components with T \200 and T \1000 K can be distinguished.30 43 E. F. van Dishoeck could result from outgassing of the icy mantles. exist both in the gas and on the grains.14,51,91 As the young star evolves its radiation heats up an increasing fraction of the surrounding gas and dust leading to enhanced evaporation of the ices and an increasing gas/solid ratio. Are the coldest sources such as NGC 7538 IRS9 therefore just younger versions of the hotter sources like GL 2591? Within a factor of two they have the same luminosity. Comparison of the column densities and abundances of the ices in NGC 7538 IRS9 with the values of the corresponding gas phase species in GL 2591 indicates that the latter are always lower suggesting that this scenario is quantitatively feasible all of the observed gas-phase H2O CO2 and CH4In more detail the trends for the gas/solid H H2O C and with increasing temperature indicate that outgassing of the icy mantles is signi–cant in the hotter regions.4 For NGC 7538 IRS9 and W 33A the CH excitation temperature of ca. 100 K and the 4 large CH gas/solid ratio compared with that of H2O suggest a grain temperature of the order of 80»90 K just below the sublimation temperature of solid H 4 2O. This would imply that the gas and dust temperatures are not fully coupled in spite of the high densities.Sputtering of grain mantles in the powerful out—ows is unlikely to be the dominant grain mantle removal mechanism since this process would have resulted in similar gas/solid ratios for all species. 2 produced by high-temperature gas-phase chemistry in the inner ìhot coreœ region initi- Part of the observed abundant gas-phase H2O in the warmer sources could also be ated by the O]H reaction. This path starts to become signi–cant at temperatures 2 greater than ca. 230 K,48,51 which could be reached by radiative heating close to the star. This possibility can be tested by observations of sources with a range of temperatures of the warm gas. In Fig. 6(a) the observed H2O abundances in the warm gas are shown as functions of the CO excitation temperatures30 assuming that all H2O is in the warm gas N(H O)/N 2 temperature.For three sources W 33A NGC 7438 IRS9 and IRS1 the temperatures are warm(H2). This excitation temperature should be close to the gas below 200 K whereas for the other sources they are signi–cantly higher. It is seen that there is a clear increase in the H2O abundance with increasing temperature. For the lower temperature sources the H2O abundance is around (3^2)]10~6 whereas for the warmer sources it is increased by an order of magnitude to (3^2)]10~5. The errors in the data are too large to determine whether there is a real dichotomy at around ca. 550 K or whether this is just a continuous trend. Does the lower value represent the amount of H2O that has evaporated from the grains and is the higher value representative of the gas-phase reactions ? Or is just the fraction of H2O-ice that has evaporated larger in the warmer sources ? In that case one may have expected that the gas-phase H2O abundances would have been even higher.Fig. 6(b) shows the observed H2O-ice abundances with respect to total H (open symbols) and with respect to cold H2 Ncold(H2)\Ntot(H2)[Nwarm(H2) (–lled symbols). As expected the ice abundance with respect to total H decreases from ca. 2]10~4 in the coldest sources to ca. 1]10~5 in 2 the warmer sources. However although the H2O-ice abundance in the cold gas also shows some decrease it is signi–cantly less and nearly constant at ca. 10~4 within the errors of a factor of two. These results imply that if the warmer sources evolved from a cold phase with H2O-ice abundances of typically 10~4 part of the oxygen must have been converted into another form after evaporation in the gas most likely O or O2 to account for the lower H2O abundances (gas]ice) in the warmer sources.Only at higher temperatures is part of the oxygen driven back into H2O again. Hot core models indeed predict a breakdown of H2O on a timescale of ca. 105 yr provided the temperature is not too high.48 2O However this is unlikely to be the explanation for the Gas-phase production of at least some of the H2O in the warmer sources would be consistent with the lack of abundant gas-phase CO2. H If the fraction of that results from outgassing of grain mantles is only ca. 5% this would be comparable with the amount of evaporated CO2 .44 ISO and gas»grain chemistry abundances in the warm gas Fig. 6 (a) Observed H N(H O)/N 2O warm(H2) H as functions of the gas 2 determined from the CO excitation.30 (b) Observed solid temperature T respect to total warm symbols) as functions of the gas temperature 2O abundances with H2 N(H2O)/Ntot(H2) H (open symbols) and the cold 2 N(H2O)/Ncold(H2) (–lled Twarm determined from the CO excitation.30 warmer sources where the data in Fig. 6 suggest signi–cant evaporation of H2O ice. Another possibility discussed in Section 6 is that most of the solid CO is in a segregated phase located in the outer part of the envelope which has not yet been heated. 2 Once in the gas-phase CO may be destroyed by gas-phase reactions on short time- 2 scales although efficient reactions with abundant species still remain to be identi–ed.25 Altogether the absence of abundant gas-phase CO is still not fully understood.2 8 Concluding remarks In summary the results illustrate that ISO can provide important new insights into the role of gas-grain interactions and the chemical and physical evolution of star-forming regions. Information on the major products of grain surface chemistry can be obtained the mechanisms for evaporation of icy mantles can be tested and the chemistry of hot core regions can be probed. IR absorption spectroscopy remains a unique tool for the study of both the gas-and the solid-state components along the same line of sight. Future work should include comparison with millimeter observations of these sources to obtain information on less abundant species.62,64 Progress in this –eld depends strongly however on the availability of basic molecular data which is often lacking for grain surface processes.Laboratory experiments on ice mantle spectroscopy as well as surface reactions remain essential for unraveling the complete story. The research described in this paper was carried out in close collaboration with J. H. Black G. A. Blake A. C. A. Boogert A. M. S. Boonman Th. de Graauw P. Ehrenfreund P. A. Gerakines F. P. Helmich F. Lahuis T. J. Millar W. A. Schutte W. F. Thi A. G. G. M. Tielens F. van der Tak D. C. B. Whittet and C. M. Wright. It was made possible thanks to the dedicated eÜorts of the SRON»MPE SWS instrument teams and the ISO»SIDT team.This work is supported by a grant from NFRA/NWO. 45 E. F. van Dishoeck References 1 D. R. Bates and L. Spitzer Astrophys. J. 1951 113 441. 2 E. Herbst and W. Klemperer Astrophys. J. 1973 185 505. 3 H. H. Lee R. P. A. Bettens and E. Herbst Astron. Astrophys. Suppl. 1996 119 111. 4 T. J. Millar P. R. A. Farquhar and K. Willacy Astron. Astrophys. Suppl. 1997 121 139. 5 W. D. Watson and E. E. Salpeter Astrophys. J. 1972 174 321; 175 659. 6 M. Allen and G. W. Robinson Astrophys. J. 1977 212 396. 7 A. G. G. M. Tielens and W. Hagen Astron. Astrophys. 1982 114 245. 8 L. B. dœHendecourt L. J. Allamandola and J. M. Greenberg Astron. Astrophys. 1985 152 130. 9 D. Hollenbach and E. E. Salpeter J. Chem. Phys. 1970 53 79.10 Dust and Chemistry in Astronomy ed. T. J. Millar and D. A. Williams IOP Bristol 1993. 11 E. Herbst in Dust and Chemistry in Astronomy eds. T. J. Millar and D. A. Williams IOP Bristol 1993 p. 183. 12 A. G. G. M. Tielens and S. B. Charnley Origin L ife 1997 27 23. 13 D. C. B. Whittet in Dust and Chemistry in Astronomy eds. T. J. Millar and D. A. Williams IOP 14 A. G. G. M. Tielens and D. C. B. Whittet in Molecules in Astrophysics Probes and Processes IAU Bristol 1993 p. 9. Symposium 178 ed. E. F. van Dishoeck Kluwer Dordrecht 1997 p. 48. 15 C. M. Walmsley and P. Schilke in Dust and Chemistry in Astronomy eds. T. J. Millar and D. A. Williams IOP Bristol 1993 p. 37. 16 L. E. Snyder Origin L ife 1997 27 115. 17 S. B. Charnley A. G. G. M. Tielens and T.J. Millar Astrophys. J. 1992 399 L71. 18 P. Caselli T. I. Hasegawa and E. Herbst Astrophys. J. 1993 408 548. 19 Th. de Graauw L. N. Haser D. A. Beintema et al. Astron. Astrophys. 1996 315 L49. 20 M. F. Kessler J. A. Steinz M. A. Anderegg et al. Astron. Astrophys. 1996 315 L27. 21 D. C. B. Whittet W. A. Schutte A. G. G. M. Tielens A. C. A. Boogert Th. de Graauw P. Ehrenfreund P. A. Gerakines F. P. Helmich T. Prusti and E. F. van Dishoeck Astron. Astrophys. 1996 315 L357. 22 L. B. dœHendecourt M. Jourdain de Muizon E. Dartois et al. 1996 Astron. Astrophys. 315 L365. 23 P. Ehrenfreund A. C. A. Boogert P. A. Gerakines and A. G. G. M. Tielens Faraday Discuss. 1998 109 463. 24 F. P. Helmich E. F. van Dishoeck J. H. Black Th. de Graauw D. A. Beintema A. M.Heras F. Lahuis P. W. Morris and E. A. Valentijn Astron. Astrophys. 1996 315 L173. 25 E. F. van Dishoeck F. P. Helmich Th. de Graauw et al. Astron. Astrophys. 1996 315 L349. 26 A. C. A. Boogert F. P. Helmich E. F. van Dishoeck W. A. Schutte A. G. G. M. Tielens and D. C. B. Whittet Astron. Astrophys. 1998 in press. 27 E. Dartois L. dœHendecourt F. Boulanger M. Jourdain de Muizon M. Breitfellner J.-L. Puget and H. J. Habing Astron. Astrophys. 1998 331 651. 28 M. Ohishi in Molecules in Astrophysics Probes and Processes IAU Symposium 178 ed. E. F. van Dishoeck Kluwer Dordrecht 1997 p. 48. 29 E. F. van Dishoeck in Molecules in Astrophysics Probes and Processes IAU Symposium 178 ed. E. F. van Dishoeck Kluwer Dordrecht 1997 p. 537 539; see http ://www.strw.leidenuniv.nl/Diau34 for links to databases.30 G. F. Mitchell J-P. Maillard M. Allen R. Beer and K. Belcourt Astrophys. J. 363 554; G. F. Mitchell J. P. Maillard and T. I. Hasegawa Astrophys. J. 1991 371 342. 31 N. J. Evans J. H. Lacy and J. S. Carr Astrophys. J. 1991 383 674. 32 J. Carr N. J. Evans J. H. Lacy and S. Zhou Astrophys. J. 1995 450 667. 33 A. G. G. M. Tielens and L. J. Allamandola in Interstellar Processes eds. D. J. Hollenbach and H. A. Thronson Reidel Dordrecht 1987 p. 397. 34 T. I. Hasegawa and E. Herbst MNRAS 1993 261 83; 263 589. 35 W. A. Schutte in T he Cosmic Dust Connection ed. J. M. Greenberg Kluwer Dordrecht 1996 p. 1. 36 D. A. Williams and S. D. Taylor QJRAS 1996 37 565. 37 D. A. Williams in Dust and Chemistry in Astronomy eds.T. J. Millar and D. A. Williams IOP Bristol 1993 p. 143. 38 P. Caselli T. I. Hasegawa and E. Herbst Astrophys. J. 1998 in press. 39 W. A. Schutte and J. M. Greenberg Astron. Astrophys. 1991 244 190. 40 B. Schmitt in Molecules and Grains in Space ed. I. Nenner AIP New York 1994 p. 735. 41 S. A. Sandford and L. J. Allamandola Icarus 1988 76 201. 42 A. Leç ger M. Jura and A. Omont Astron. Astrophys. 1985 144 147. 43 A. P. Jones and D. A. Williams MNRAS 1994 209 955. 44 R. J. A. Grim and L. B. dœHendecourt Astron. Astrophys. 1986 167 161. 45 M. Spaans M. R. Hogerheijde L. G. Mundy and E. F. van Dishoeck Astrophys. J. 1995 455 L167. 46 R. Gredel S. Lepp A. Dalgarno and E. Herbst Astrophys. J. 1989 347 289. 47 L. B. dœHendecourt L. J. Allamandola R. J.A. Grim and J. M. Greenberg Astron. Astrophys. 1986 158 119. 46 ISO and gas»grain chemistry 48 S. B. Charnley Astrophys. J. 1997 481 396. 49 T. J. Millar G. H. Macdonald and R. J. Habing Astron. Astrophys. 1997 325 1163. 50 F. P. Helmich T. J. Millar and E. F. van Dishoeck Astron. Astrophys. 1998 to be submitted. 51 C. Ceccarelli D. J. Hollenbach and A. G. G. M. Tielens Astrophys. J. 1996 471 400. 52 E. F. van Dishoeck G. A. Blake B. T. Draine and J. I. Lunine in Protostars and Planets III eds. E. H. Levy and J. I. Lunine University of Arizona Tucson 1993 p. 169. 53 E. F. van Dishoeck and G. A. Blake Astrophys. Spa. Sci. 1995 224 237. 54 E. F. van Dishoeck and G. A. Blake Annu. Rev. Astron. Astrophys. 1998 36 in press. 55 T. W. Hartquist P. Caselli J.M. C. Rawlings D. P. Ruffle and D. A. Williams to appear in Molecular Astrophysics II eds. T. W. Hartquist and D. A. Williams Oxford University Oxford 1998. 56 E. F. van Dishoeck F. P. Helmich W. A. Schutte P. Ehrenfreund F. Lahuis A. C. A. Boogert A. G. G. M. Tielens Th. de Graauw P. A. Gerakines and D. C. B. Whittet in Star Formation with the ISO Satellite eds. J. Yun and R. Liseau ASP Berkeley 1998 vol. 132 p. 54. 57 F. Lahuis and E. F. van Dishoeck in First ISO workshop on Analytical Spectroscopy ESA»SP 1998 419 Estec Noordwijk p. 275; and in preparation. 58 P. E. Clegg P. A. R. Ade C. Armand et al. Astron. Astrophys. 1996 315 L38. 59 C. M. Wright E. F. van Dishoeck F. P. Helmich F. Lahuis A. C. A. Boogert and Th. de Graauw in First ISO workshop on Analytical Spectroscopy ESA»SP 1998 419; Estec Noordwijk p.37. 60 E. F. van Dishoeck and F. P. Helmich Astron. Astrophys. 1996 315 L177. 61 F. P. Helmich 1996 PhD thesis University of Leiden; http ://www.strw.leidenunv.nl/Dfph/papers. 62 F. van der Tak E. F. van Dishoeck N. J. Evans and G. A. Blake Astrophys. J. 1998 to be submitted. 63 J. H. Lacy R. Knacke T. R. Geballe and A. T. Tokunaga Astrophys. J. 1994 428 L43. 64 A. M. S. Boonman F. van der Tak F. P. Helmich and E. F. van Dishoeck 1998 in preparation ; A. M. S. Boonman C. M. Wright E. F. van Dishoeck et al. 1998 in preparation. 65 Y. C. Minh M. K. Brewer W. M. Irvine P. Friberg and L. E. B. Johansson Astron. Astrophys. 1991 244 470. 66 J. H. Lacy J. S. Carr N. J. Evans F. Baas and J. Achtermann Astrophys.J. 1991 376 556. 67 J. H. Lacy N. J. Evans J. M. Achtermann et al. Astrophys. J. 1989 342 L43. 68 P. Ehrenfreund L. dœHendecourt E. Dartois M. Jourdain de Muizon M. Breitfellner J. L. Puget and H. J. Habing Icarus 1997 130 1. 69 L. J. Allamandola M. P. Bernstein and S. A. Sandford in Astronomical and biochemical origins and the search for life in the universe eds. C. B. Cosmovici S. Bowyer and D. Werthimer Ed. Comp. Bologna 1997 p. 23. 70 Th. de Graauw D. C. B. Whittet P. A. Gerakines et al. Astron. Astrophys. 1996 315 L345. 71 J. Gué rtler Th. Henning C. Koé mpe W. Pfau W. Kraé tschmer and D. Lemke Astron. Astrophys. 1996 315 L189. 72 P. A. Gerakines A. C. A. Boogert P. Ehrenfreund et al. 1998 in preparation. 73 S. P. Willner F. C. Gillett T.L. Herter et al. Astrophys. J. 1982 253 174. 74 W. A. Schutte A. G. G. M. Tielens D. C. B. Whittet A. C. A. Boogert P. Ehrenfreund Th. de Graauw T. Prusti E. F. van Dishoeck and P. R. Wesselius Astron. Astrophys. 1996 315 L333. 75 A. G. G. M. Tielens A. T. Tokunaga T. R. Geballe and F. Baas Astrophys. J. 1991 381 181. 76 S. A. Sandford L. J. Allamandola A. G. G. M. Tielens and G. J. Valero Astrophys. J. 1988 329 498. 77 D. C. B. Whittet P. A. Gerakines A. G. G. M. Tielens A. J. Adamson A. C. A. Boogert J. E. Chiar Th. de Graauw P. Ehrenfreund T. Prusti W. A. Schutte B. Vandenbussche and E. F. van Dishoeck Astrophys. J. 1998 489 L159. 78 A. C. A. Boogert W. A. Schutte A. G. G. M. Tielens D. C. B. Whittet F. P. Helmich P. Ehrenfreund P. R. Wesselius Th. de Graauw and T. Prusti Astron. Astrophys. 1996 315 L377. 79 N. Boudin W. A. Schutte and J. M. Greenberg Astron. Astrophys. 1998 in press. 80 M. P. Bernstein S. A. Sandford and L. J. Allamandola Astrophys. J. 1997 476 932. 81 W. A. Schutte 1997 personal communication. 82 R. G. Smith K. Sellgren and A. T. Tokunaga Astrophys. J. 1989 344 413. 83 F. Wyrowski and C. M. Walmsley Astron. Astrophys. 1996 314 265. 84 P. D. Gensheimer R. Mauersberger and T. L. Wilson Astron. Astrophys. 1996 314 281. 85 J. Harju C. M. Walmsley and J. G. A. Wouterloot 1988 Astron. Astrophys. Suppl. 98 51. 86 I. A. Crawford and D. A. Williams MNRAS 1997 291 L53. 87 L. J. Allamandola S. A. Sandford A. G. G. M. Tielens and T. Herbst Astrophys. J. 1992 399 134. 88 W. A. Schutte P. A. Gerakines T. R. Geballe E. F. van Dishoeck and J. M. Greenberg Astron. Astrophys. 1996 309 633. 89 M. E. Palumbo G. A. Baratta J. R. Brucato A. C. Castorina M. A. Satorre and G. Strazulla Astron. Astrophys. 1998 in press. 90 S. C. Tegler D. A. Weintraub T. W. Rettig et al. Astrophys. J. 1995 439 279. 91 S. D. Doty and D. A. Neufeld Astrophys. J. 1997 489 122. Paper 8/00815I; Received 29th January 1998

ISSN:1359-6640    

DOI:10.1039/a800815i

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

Faraday Discuss. 1998 109 47»60 Electron abundance in dense cloud cores Implications for star formation Reneç Plume Edwin A. Bergin Jonathan P. Williams and Philip C. Myers Harvard-Smithsonian Center for Astrophysics 60 Garden St. Cambridge MA 02138 USA Slog(X By combining observations of the J\1]0 transitions of C18O and H13CO` and the J\1]0 and 2]1 transitions of DCO` and a model of molecular cloud chemistry we have obtained the electron abundance (X cores. We –nd that the electron abundances are con–ned to a relatively e4ne/nH2 ) in a sample of 20 low-mass and 7 high-mass molecular cloud [7.5\log(X narrow range of e)\[6.5 with very little scatter about the mean values of e)T\[7.04 ; p\0.22 (low-mass cores) and Slog(Xe)T\[7.11 ; p\0.15 (high-mass cores).These values are consistent with the standard view that the ionization in dense cloud cores is dominated f by cosmic rays provided that H2 B5]10~17 s~1. These electron abundances also imply that the neutrals are only marginally coupled to the magnetic –eld (W B5»8) with only ca. 10% the core radius being cut oÜ from magneto-hydrodynamic (MHD) wave propagation. The coupling parameter (W ) also suggests that ambipolar diÜusion timescales are about an order of magnitude larger than the freefall timescale. 47 1 Introduction 1.1 Background AVB4 but since stars form in regions of high extinction the ionization In the standard model of star formation magnetic –elds can play an important role in the support of dense molecular cloud cores against gravitational contraction.1 In this model the diÜusion of magnetic —ux inside the core is controlled by the ionization fraction through ion»neutral collisions (called ambipolar diÜusion) and the initially magnetically subcritical core evolves to a supercritical state and collapses.2 However the interaction between magnetic –elds and the bulk of the gas occurs only indirectly through collisions between the neutral particles (which constitute the vast majority of the gas) and the ions which are tied to the magnetic –eld lines.Therefore the ionization fraction (or the electron abundance Xe4ne/nH2 where n is the number density of nH2 is the number density of hydrogen molecules) in the gas is a fundamen- electrons and e tal parameter of cloud dynamics.The ions in the cores may be created by cosmic rays or by energetic photons. McKee2 shows that in general photo-ionization by the ambient interstellar radiation –eld dominates the ionization in molecular clouds for low visual extinctions and cosmic rays for high extinctions. For a uniform homogeneous layer both sources contribute equally at balance should be dictated by cosmic rays alone. In this case X is proportional to H2 .3 However there are several lines of reasoning that suggest that the electron 1/Jn e fraction may be higher than the cosmic ray value. First newly formed low-mass stars are 48 Electron abundance in dense cloud cores proli–c X-ray sources which if sufficiently energetic (P1 keV) can penetrate and ionize regions of high visual extinction.4 Secondly molecular clouds are very inhomogeneous and UV radiation from hot stars can penetrate far deeper into clouds than a uniform medium.5,6 Thirdly turbulence in clouds might efficiently mix the UV-ionized surfaces of cores with their centres and enhance the electron fraction by more than an order of magnitude over that expected in the quiescent case.7 XeB10~6 was set by An upper limit to the ionization fraction in molecular clouds Dalgarno and Lepp8 who applied a model of deuterium fractionation (including for the –rst time reactions with atomic deuterium) to the DCO` and HCO` data of Wootten et al.9 At such low abundances magnetic –eld support is only temporary because the neutrals eventually slip past the ions10 through the process of ambipolar diÜusion.For typical conditions in molecular cloud cores the ambipolar diÜusion timescale (tADB 10(X /10~7)]106 years5) has been proposed as the valve that regulates the star forma- e tion rate.1 Measurements of the ionization fraction therefore may reveal star formation timescales. On the other hand if such measurements are incompatible with other estimates of star formation rates,11 then alternative mechanisms for core creation and support may be indicated. To examine the stability of dense cores and the coupling of the magnetic –eld to the gas we have undertaken an observational and theoretical study to survey the ionization fraction in dense cores with a variety of star-forming properties. The cores some of which contain embedded protostars and some of which do not have a range of physical properties.We separate our sample into low-mass cores and high-mass cores. The lowmass cores are located in regions that are known to form single isolated low-mass stars. The high-mass cores are located in regions associated with the formation of clusters of high-mass stars. The high-mass cores are also hotter denser and have larger linewidths than the low-mass cores. With the addition of ionization fractions determined for even more massive star forming sites,9,12,13 we will examine the abundance of electrons throughout the entire range of star formation activity currently found in the galaxy. The contents of this paper will be discussed in greater detail in a series of future articles.14,15 1.2 Observational and theoretical technique It is not a simple matter to measure the fractional ionization in a molecular cloud or core since most of the electrons are believed to be released from species with low ionization potentials such as Fe and other metals.16 These atomic species produce lines in the optical and therefore cannot be directly detected in regions of high visual extinction.Moreover their abundance is highly uncertain since the heavy elements readily attach to dust grains and can be severely depleted in dense cores. Therefore we use indirect determinations of the electron abundance through observations of molecular ions DCO` and isotopes of HCO` and the application of chemical models. This technique was originally proposed by Gueç lin et al.17 who showed that the ratio of the DCO` to HCO` abundances ([DCO`]/[HCO`]) is a sensitive probe of the electron fraction.Fig. 1 illustrates the chemical reaction network primarily responsible for the creation and destruction of DCO` and HCO` The formation of deuteriated molecules begins with the isotope exchange reaction (1) H3 `]HD7H2D`]H2]*E The forward reaction is exothermic by an amount *E/kB230 K and therefore at low temperatures the primary destruction mechanism for H2D` is via recombination with electrons. Formation of DCO` follows through the reaction (2) H2D`]CO]DCO`]H2 49 R. Plume et al. Fig. 1 Schematic diagram representing the chemical network responsible for the creation and destruction of DCO` and HCO` reproduced from ref. 19 with kind permission Similarly the main HCO` formation reaction is (3) (X H3 `]CO]HCO`]H2 The [DCO`]/[HCO`] ratio depends on the electron abundance because an increase in the electron density will lower the abundance of H2D` (through recombinations) and therefore the abundance of DCO` [through eqn.(2)]. Hence regions with high electron abundances will have smaller [DCO`]/[HCO`] abundance ratios than will regions with low electron abundances. This result was con–rmed by Gueç lin et al.18 who showed that the D/H ratios in –ve dense (2]103\n(H )/cm~3\105) clouds are more than 2 103 times greater than the typical interstellar ratios implying very low electron abundances eO10~7). Wootten et al.9 found a similar value of Xe in eight low-mass cores. Although this technique was eventually abandoned owing to several debilitating uncertainties,19 in the last few years two major sources of uncertainty have been addressed.Hubble Space Telescope (HST) observations have provided accurate measurements of the D/H ratio in the local interstellar medium,20 and laboratory measurements have agreed upon the previously controversial H3 ` and H2D` dissociative recombination rates.12,21 As a result observations of DCO` and HCO` can now be used with greater con–dence to estimate electron abundances in dense molecular cloud cores. 2 Observations To determine the ionization fraction in dense cores with a variety of star-forming properties we have selected a number of low-mass and high-mass cores. Some of these cores are known to have associated protostars whereas the remainder are considered ì starless œ (as determined by comparison with the IRAS point source catalog).For our sample of low-mass cores we selected 20 cores from the ammonia survey of Benson and Myers.22 The particular sources were selected randomly based upon which cores were observable during our observing times. For our sample of high-mass cores we have chosen seven sources in L1630 and L1641 from the IRAS-selected NH survey of Harju 3 et al.23 These high-mass cores are larger warmer denser and have larger velocity dispersions than the low-mass NH cores used in this study. None of the cores in our survey are associated with young stellar clusters ; our sample contains cores associated with 3 single IRAS sources or are starless.However the high-mass cores all have near neighbours that are currently forming clusters or associations. We observed our sample of cores in the J\1]0 transitions of C18O and H13CO` and the J\1]0 and 2]1 transitions of DCO` at the National Radio Astronomy Observatory 12 m telescope in 1996 and 1997. The beam size and velocity resolution 50 source low-mass cores L134A L183S L1696A L43 L260 L158 L234A L63 L483 L778 B335S L1152 L1155C L1082C L1082A L1228B B361 L1221 L1251 L1262 high-mass cores IRAS05302-0537 IRAS05369-0728A IRAS05369-0728B IRAS05389-0756A IRAS05389-0756D IRAS05403-0818 IRAS05399-0121 was 57A and 0.067 km s~1 for C18O 73A and 0.084 km s~1 for H13CO` and 87A and 0.10 km s~1 for DCO`.Further details of these observations as well as coordinates for the observed cores will be presented in future papers.14,15 3 Results 3.1 Column densities The observational constraints on our chemical models (described in the next section) are the column densities of C18O DCO` and H13CO`. These column densities were calculated using the assumptions that the gas is optically thin and in local thermodynamic equilibrium (LTE). The small optical depth of the observed transitions is supported by additional observations of C17O J\1]0 HC18O` J\1]0 and D13CO` J\1]0 in a few cores. The low intensity of these transitions implies that the C18O H13CO` and DCO` lines are indeed optically thin with opacities (q) less than unity.of 10 K24 for the low-mass cores and T (TK) K\15 We used a kinetic temperature K23 for the high-mass cores. However for these temperature and densities [n(H2)B (1»3)]104 cm~3 for low-mass cores and n(H2)B105 Electron abundance in dense cloud cores Table 1 Column densities DCO` C18O error log(N) log(N) 11.67 12.27 12.24 12.35 11.56 12.07 13.5 13.3 13.6 13.5 13.5 13.5 14.96 15.14 15.30 15.06 15.04 15.16 12.08 12.26 12.19 11.73 11.92 11.87 13.4 13.5 13.5 13.4 13.3 13.5 14.78 14.95 15.39 14.93 15.06 14.83 11.73 11.79 11.74 12.20 11.81 11.99 13.4 13.4 13.4 13.6 13.6 13.6 14.99 14.97 15.14 15.06 15.00 15.15 11.99 12.12 13.4 13.3 15.14 15.03 12.21 11.95 12.22 12.06 12.15 12.11 14.2 14.3 14.3 14.1 14.5 14.2 15.48 15.35 15.43 15.51 15.58 15.38 12.15 14.3 15.76 H13CO` error error log(N) 10.3 10.2 10.5 10.6 10.5 10.4 11.12 11.71 11.98 11.91 11.16 11.81 10.5 10.5 10.8 10.8 10.6 10.8 10.3 10.4 10.4 10.4 10.4 10.4 11.53 11.69 11.81 11.49 11.64 11.42 10.7 10.6 10.7 10.5 10.5 10.6 10.5 10.4 10.4 10.6 10.4 10.5 11.37 11.46 11.31 11.77 11.47 11.68 10.6 10.6 10.5 10.7 10.5 10.7 10.6 10.5 11.73 11.65 10.6 10.6 11.1 11.1 11.1 11.2 11.2 11.1 12.15 12.01 11.92 11.87 11.98 12.02 11.1 11.1 11.2 11.0 11.1 11.1 11.3 11.98 11.2 cm~3 for high-mass cores] the 51 R.Plume et al. level populations of HCO` and H13CO` will not be populated in a Boltzmann distribution but rather will be skewed toward the lower levels. Hence the assumption of LTE will tend to overestimate the column density necessary to account for the observed integrated intensity of the lines. Therefore we use statistical equilibrium calculations to derive a correction factor for the LTE column densities. The column densities and their observational errors are listed in Table 1. 3.2 Description of the chemical model We used the chemical model discussed in detail in Bergin et al.16 and later updated.25 This network utilizes the pure gas-phase UMIST RATE95 reaction set,26 although where mentioned we have used the gas-grain adaptation described in Bergin and Langer.25 Because the Millar et al.26 network does not include the chemical interactions that produce deuteriated molecules we have used an indirect method to estimate the deuterium fractionation of H3 ` and therefore HCO`.Essentially we use the equilibrium abundances of all non-deuteriated species as determined by our model and then compute the H2D`/H3 ` ratio by balancing of the formation and destruction rates of H2D`:8 (4) n(DCO`) n(HCO`) \1 3 n(H n(H2D`) 3 `) \ n n (HD) (H2) f where f is given by (5) kf f\ k X r] a(H n 2 (H D`)ne]; kX n n (H (X) 2) 2) k and k are the forward and backward coefficients of eqn.(4) a is the recombi- f Here r H nation rate for 2D` ne is the number density of electrons n(X) is the number density of neutral species that will react with H2D` (CO N2 O O 2 H2O C Fe S Mg Si) and k is the rate coefficient for those reactions. We use the rate coefficients for eqn. (4) x as given in de Boisanger et al.12 and the recombination rate for H2D` from Larsson et al.21 We wish to use our observations in combination with the chemical models to constrain the electron abundance as tightly as possible. Given the large set of potential variables we must appeal to a number of previously established observational results to constrain the parameter space over which we run the chemical models.3.3 Model parameters gas). For a given model calculation the adjustable variables are the gas kinetic temperature (T (A (nH2 ) the visual extinction and the volume density of molecular hydrogen gas) V) the factor (s) by which the interstellar radiation –eld is enhanced over the standard value of 1.6]10~3 erg s~1 cm~2 sr~1 the cosmic ray ionization rate for molecular hydrogen (fH2 ) and the initial abundances of atomic species (He C N O Si S Na Mg Fe P). (T For gas-grain models the dust temperature (Tdust) is set equal to the gas temperature For low-mass cores the ammonia survey of dense cores in dark clouds22 shows that the temperatures and densities do not vary greatly from core to core. Based on this we –x the gas temperature at 10 K and allow only two values for the density n(H2)\104 cm~3 and n(H NH observations) but we allow for the temperature being as high as 30 K in the 2)\3]104 cm~3.For high-mass cores we adopt Tk\15 K (based on chemical calculations. This range is permitted because the various tracers could poten- 3 52 Electron abundance in dense cloud cores Table 2 Initial abundances of species relative to H2 abundance species 0.28 1.33]10~10 1.46]10~4 He He(]) C(]) NO Si(]) S(]) Na(]) Mg(]) Fe(]) P(]) HD 4.28]10~5 3.52]10~4 4.0]10~8 4.0]10~8 4.0]10~9 6.0]10~9 6.0]10~9 6.0]10~9 2.8]10~5 Note that carbon and the other species with ionization potentials less than 13.6 eV are initially ionized as in diÜuse clouds.However with the large A in our models the system quickly reaches an equilibrium state v in which the majority of the carbon is neutral and the ionization fraction is substantially reduced. tially trace slightly diÜerent gas compared to the NH observations. The densities in the 2)\105 cm~3 on the basis of a multitransitional study of n(H high-mass cores are set at 3 CS.15 The visual extinction can be determined using the C18O column densities provided A X(C18O)\1.7]10~7N(H in Table 1. If we assume that 2)27 and N(H2)\1021Av then A extinctions range from v\3.5»35 magnitudes (mag) at the chosen positions. We have A adopted a value of v\7.5 mag for our analysis which is close to the low end of this range.As the radiation –eld is mostly extinguished at this depth higher values will not alter our results. As a test we have run additional models with Av\3 and have found that diÜerences between the v\3 and Av\7.5 models are negligible. Since the high-mass cores are located in the Orion cloud (a region of massive star formation) where the radiation –eld could be higher than average we examine two values of the UV enhancement factor s\1 and 500. Little diÜerence was found for these two models and therefore we only present results with the normal interstellar radiation –eld. Since all the low-mass cores are located in cold clouds with no nearby OB stars the standard UV radiation –eld (s\1) is used for these cores as well. Chemical interactions in the interior of a dense core are expected to be powered through cosmic ray ionization of molecular hydrogen.Thus the electron abundance is related to the cosmic ray ionization rate. In a starless core cosmic rays also provide the dominant heating source and an estimate of f (the cosmic ray ionization rate for molecular hydrogen) can be made by requiring thermal balance at a temperature of 8»20 H2 K. Using two diÜerent heating and cooling models,28,29 we –nd a permitted range models for fH2B(1»15)]10~17 s~1 for densities n(H2)\104»105 cm~3. Thus we have computed fH2B(1 5 10 15)]10~17 s~1 and we have run these models for all the 53 R. Plume et al. k\10 K fH2 B5]10~17 s~1 s\1 and Av\7.5. The logarithm of the electron fraction Fig. 2 Three-panel –gure displaying our low-mass core data (solid points) and the models (contour lines) which best –t the data (centre and left panels).The ordinate of the plot is log[N(HCO`)/N(CO)] and the abscissa is log[N(DCO`)/N(HCO`)]. The left panel plots the data and their associated error bars. The central and right panels overlay models for a density of T n H2\104 cm~3 and nH2\3]104 cm~3 respectively. The displayed models were calculated with [log(Xe)] ratio of carbon and oxygen abundances relative to the nominal value in Table 2) are labelled is labelled in each plot along the right-hand side of the contours. Variations in y (the along the top of each plot. H2B5 chosen densities. For the remainder of the paper we only present models with f ]10~17 s~1 because these results were in best agreement with our observations.Initial atomic abundances are typically determined from observations taken along lines of sight towards diÜuse clouds. However recent observations of diÜuse clouds suggest that carbon and oxygen are somewhat depleted.30 For this reason the initial C N and O abundances (listed in Table 2) have mild depletions from their solar abundances. We have also run models for values of initial carbon and oxygen abundances that are 0.5 0.75 1.0 and 1.2 times the values in Table 2. The missing carbon and oxygen is presumed to be incorporated into dust-grain cores or into CO and H2O which are in turn depleted onto grains.31,32 In addition previous chemical modelling eÜorts33 have shown that observed chemical abundances are reproduced best when the abundances of metal ions are depleted below the diÜuse cloud values.Therefore we have also run models that more severely deplete the metal abundances (varying the initial abundances of the Si S Na Mg Fe and P from 0.1 to 150 times the values in Table 2). These values were chosen for a number of reasons. First metal abundances are observed to vary substantially from cloud to cloud.34,35 Secondly since metals have an ionization potential less than that of hydrogen and their recombination rate is much slower than those of the lighter atomic species they will be ionized in the gas phase. The ionization fraction therefore depends critically on the abundance of metals in the gas phase.36 To summarize to account for the wide range of parameters more than 1000 models were run.For low-mass cores we chose a gas temperature of 10 K and densities of n(H2)\104 cm~3 and 2)\3]104 cm~3. For high-mass cores we adopt Tk\ n(H 2)\105 cm~3. In each model the radiation –eld and visual extinction 15»30 K and n(H through the core are –xed at s\1 and Av\7.5 respectively. For each value of Av n(H T and s the metal abundances vary by factors of 0.1 to 150 and C and O abundances by factors of 0.5 to 1.2 times their nominal values. We have also investigated 2), 54 Electron abundance in dense cloud cores Table 3 Electron abundances source log(Xe) min. log(Xe) max. log(Xe) [6.67 [7.06 [6.92 [7.46 [6.72 [6.88 [7.24 [7.23 [7.08 [6.82 [6.83 [7.18 [7.04 [7.17 [6.75 [7.33 [7.15 [7.22 low-mass cores L134A L183S L1696A L43 L260 L158 L234A L63 L483 L778 B335S L1152 L1155C L1082C L1082A L1228B B361 L1221 L1251 L1262 [6.84 [7.19 [7.22 [7.18 [7.28 [7.00 [7.04 [7.20 high-mass cores IRAS05302-0537 IRAS05369-0728A IRAS05369-0728B IRAS05389-0756A IRAS05389-0756D IRAS05403-0818 IRAS05399-0121 [6.87 variations in the cosmic ray ionization —ux between f fH2\5]10~17 s~1 after comparing the model results with our observations.3.4 Electron abundances 3.4.1 Low-mass cores. Fig. 2 plots the DCO` to HCO` ratio vs. the abundance of the principal molecular ion HCO` relative to CO for the low-mass cores. We have used isotope ratios [16O]/[18O]\500 [12C]/[13C]\60 to convert the Table 1 column densities of C18O to CO and H13CO` to HCO` respectively.The left panel plots the observations and their associated error bars. In the central and right panels the points denote the observations whereas the model outputs are represented as contours of fractional ionization. Models for a density of n(H2)\104 cm~3 are plotted in the central panel and models for n(H2)\3]104 cm~3 in the right panel. Xe (the electron fraction) decreases toward the top of each plot which corresponds to a higher HCO` to CO ratio (most of the electrons come from metals and the destruction rate of HCO` decreases as the abundance of free electrons decreases). The left and right (vertical) contour boundaries correspond to high and low C and O abundances respectively.The nominal C and O abundance is indicated by the heavy solid lines. Contours are nearly horizontal demonstrating that to –rst order the ionization fraction can be estimated from the HCO` abundance and the deuterium abundance is of secondary importance. There is good agreement between our observations and the models and most data points lie in a tight cluster that can be bounded by a fairly small range of Xe . [6.60 [7.04 [6.87 [7.38 [6.50 [6.83 [6.87 [7.11 [6.96 [7.53 [6.99 [6.95 [7.14 [7.15 [7.04 [6.72 [6.74 [7.02 [7.37 [7.32 [7.13 [6.90 [6.91 [7.32 [6.82 [7.05 [6.57 [7.23 [7.03 [7.13 [7.19 [7.27 [6.97 [7.42 [7.27 [7.29 [6.72 [7.08 [6.92 [7.29 [7.14 [7.05 [7.15 [6.83 [6.86 [7.08 [7.31 [7.33 [7.40 [7.17 [7.20 [7.31 [6.74 [7.03 H2B(1»15)]10~17 s~1 but adopt 55 R.Plume et al. Fig. 3 log[N(HCO`)/N(CO)] vs. log[N(DCO`)/N(HCO`)] for our sample of high-mass cores. The solid points with the error bars represent the data and the contours plot the models which n best –t the data. The displayed models were calculated with H2\105 cm~3 K Tk\15»30,fH2 B 5 ]10~17 s~1 s\1 and Av\7.5. The logarithm of the electron fraction [log(Xe)] is labelled in each plot along the right-hand side of the contours. Variations in the gas kinetic temperature are labelled along the top of each plot. The two starless cores are IRAS05369-0728B and IRAS05389- 0756D.Based on multi-transitional studies of these cores,22,37 we preferentially chose the higher density model to estimate the fractional ionization in each core. We refer to the n(H2)\104 cm~3 models only when the n(H2)\3]104 cm~3 models fail to –t the data. The contour range spanned by the error bars for each data point de–ne which can be –t by the models all have ionization fractions that lie within an order of the minimum and maximum values of Xe . Results are tabulated in Table 3. The 20 cores magnitude of each other with a mean value Slog(Xe)T\[7.04 and standard deviation 0.22. There is no signi–cant diÜerence in the ionization fraction between the cores with stars and those without. This somewhat counter-intuitive result was also found by Caselli et al.38 in an analysis of the observations of Butner et al.37 3.4.2 High-mass cores.Fig. 3 plots the DCO` to HCO` ratio vs. the abundance of the principal molecular ion HCO` relative to CO for our sample of high-mass cores in Orion. Comparing the data (solid points with error bars) to the low-mass core data (Fig. 2) shows that both samples have similar HCO` abundances. However the level of fractionation in deuterium is slightly diÜerent. Three high-mass cores all associated with stars have very low deuterium fractions [log(DCO`/HCO`)\[1.6]. The contours in Fig. 3 show the model results (in a fashion similar to Fig. 2). These models were calculated for fH2B5]10~17 s~1 c n(H2)\105m~3 and constant 56 Electron abundance in dense cloud cores 2D` [eqn. (1)] and eventually DCO` is temperature sensitive.carbon and oxygen abundances (set equal to the values in Table 2). This is a diÜerent approach from the low-mass core analysis. For low-mass cores we accounted for the horizontal spread in the data by varying the average density and the depletion of carbon and oxygen. For the high-mass cores similar models are unable to reproduce the lowest observed [DCO`]/[HCO`] ratios unless we raise the C and O abundances to much higher values (ones which are inconsistent with observed depletions towards diÜuse clouds). Fortunately we are able to use the higher gas temperatures of the high-mass cores to account for the horizontal spread of the data. This works because the main reaction that forms H The forward reaction is favoured at low temperatures but since H is ca.105 times more abundant than HD once the temperature becomes greater than ca. 20 K the reverse 2 reaction will begin to contribute and the [DCO`]/[HCO`] ratio will be lowered. Thus in Fig. 3 the left and right (vertical) contour boundaries correspond to T \30 K and T \15 K respectively. The carbon and oxygen abundances are still variables but we –nd that the data are best –t by models with ìnormalœ (see Table 2) carbon and oxygen abundances (although abundance variations of ^25% can still –t the data). The electron abundances derived for the high-mass cores are determined on the basis of the best –t to the models presented in Fig. 3. The values are given in Table 3. As in the case for low-mass cores there is no signi–cant diÜerence in the ionization fraction between the cores with stars and those without (see Fig.3). e 4 Discussion 4.1 Electron abundances and ionization We have calculated electron abundances in the largest sample of cores to date (20 lowmass cores and 7 high-mass cores). These estimates are the –rst to constrain the ion fraction to a relatively narrow range in a large sample of cores. Inspection of Table 3 Slog(X Xe . For the low-mass cores shows remarkably little scatter in the values of e)T\ [7.04 with a standard deviation (p) of only 0.22 and for the high-mass cores Slog(Xe)T\[7.11 with p\0.15. In contrast previous studies of the electron fraction reported much larger scatter in their estimates of X (ranging from ca. 10~7 to less than 10~8).9,17,18 Perhaps the most meaningful comparison between electron abundances among molecular cloud cores would be to compare the level of ionization to the density of each core.To extend our analysis to cores of higher density (and mass) than those in our sample we draw upon recent work from the literature. De Boisanger et al.12 determined X in two massive cores W3 and NGC2024 and we simply report their –ndings. To obtain the electron abundance in three more massive cores we use our models to inter- e pret the DCO` and HCO` observations of Wootten et al.9 in NGC1333 and Ser MC1 and the observations of Bergin et al.13 in OMC-1. The densities of the most massive cores have been determined via excitation analyses of various species with values typically ca.106 cm~3.39 The objects studied in this work have lower densities n(H2)\(3 ]104)»105 cm~3 but have not been systematically studied using one particular technique. Therefore there is no comprehensive examination of the density using similar models and molecular transitions for the entire range of objects. Since the density is not easily and consistently derived for each source we have used the virial mass of the cores as determined from ammonia observations. Each of these sources has been observed in NH and we compute a virial mass using the expressions 3 given in Goldsmith et al.40 assuming a density power law exponent a\2. We used NH observations from the literature to obtain the NH linewidth and core sizes for the cores3 3 in our sample,22,23 as well as for the de Boisanger Wootten and Ungerechts samples.57 R. Plume et al. vations of the ammonia core size and linewidth. Fig. 4 Electron abundance (X vs. the log of the virial mass for our sample of cores as well as for e) the de Boisanger12 Wootten9 and Ungerechts13 cores. Virial masses are calculated from obser- Fig. 4 plots the electron abundances as a function of the ammonia virial mass over a wide range of star forming regions (from low-mass isolated cores to extremely high-mass cores associated with OB star-forming regions). The low-mass cores which represents the largest sample are much more scattered and have an average electron abundance of SXeTB10~7. Fig. 4 illustrates the similarity in Xe between cores with stars and those without. Most striking however is the order of magnitude diÜerence in ionization fraction between the most massive and densest star forming cores and the lower mass/density cores.To be more quantitative cores with masses between 5 M_\ MVir(NH3)\20 M_ (where M is the mass of the sun) have an average electron abundance of 6]10~8 while the massive O B star forming sites (MVir(NH3)[100 M_) _ SX have eTB10~8. These results clearly suggest that the more massive sources which presumably have higher densities also have lower electron abundances. However these results must be viewed with caution as there are only three data points for very massive sources and the electron abundances were derived from alternate models. The fact that our models all of which were calculated with a normal radiation –eld (s\1) nicely –t the data (Fig.2 and 3) and that there is no diÜerence in X between cores with stars and those without (Fig. 4) suggests that the electron abundances can be e explained by the conventional picture of pure cosmic-ray ionization. Therefore there is no need to invoke strong internal or external ultraviolet or X-ray –elds to account for the observed electron abundances. However note that since young stars may be strong X-ray emitters,4 it will be necessary to include the eÜects of X-rays into the chemical models before such strong conclusions can be made. 4.2 Gasñgrain interactions Besides the uncertainties in the physical models for these sources (discussed earlier) there is an additional source of uncertainty in our use of a pure gas-phase chemical network without any interaction of gas-phase species with grain surfaces.In principle for lowmass cores we have accounted for this by varying the atomic carbon oxygen and metal abundances with the implicit assumption that these species have been removed from the gas phase either by incorporation into grain cores or accretion onto the surface. In this 58 Electron abundance in dense cloud cores fashion for Fig. 2 we account for the horizontal spread in the observations by variations in the carbon and oxygen abundances and the vertical spread by changes in metals and therefore electron abundances. Alternatively the horizontal spread could be the result of intrinsic variations in the cosmic ray —ux.38 If we use the gas»grain models in Bergin et al.16 for n(H2)\3]104 cm~3 Tdust\TK\10 K and the initial abundances from Table 2 we –nd that the depletion of molecular species such as CO and H2O will result in increasing the [DCO`]/[HCO`] ratio in the same fashion as found for lowering the atomic carbon and oxygen abundances.Thus the horizontal spread in the data can be accounted for by varying amounts of molecular depletion. However for gas»grain models the metal ions which account for the vast majority of the free electrons deplete so rapidly that the model cannot account for any vertical variations in the data (the model essentially collapses to the limit of totally depleted metals). There are at least two potential solutions to this dilemma. First there could be a more efficient desorption mechanism than used in the Bergin et al.16 models (cosmicray induced desorption) for cold (T \10 K) clouds such as grain mantle explosions IR radiation –elds and the heat of chemical reactions.Although alternative mechanisms could be unrecognized only high-energy cosmic ray impacts or grain»grain collisions appear to have sufficient strength to remove some of the tightly bound metal atoms.41 Secondly as discussed in greater detail in de Boisanger et al.12 a variable population of small charged grains in these cores could account for the changes in overall electron abundance. An additional complication is added when we consider the surface chemistry of deuteriated molecules. Including these eÜects can lower the [HD]/[H2] ratio in the gas phase.42 The exact eÜect of surface reactions on our results is difficult to quantify also change the [DCO`]/[HCO`] ratio as discussed above.If we simply use eqn. (4) because the abundances of other species such as O or H2O could be changed which will and (5) a lower HD to H ratio will directly lower the equilibrium [DCO`]/[HCO`] ratio. This requires greater carbon and oxygen depletions lower electron fractions 2 and/or higher densities and temperatures. If surface chemistry is active in cold cores such that the HD to H ratio is lowered then the ionization values derived here would be upper limits. Greater quanti–cation of these eÜects await combined models of gas and 2 surface chemistry. Despite these uncertainties we believe that the variations in ionization fraction found here are indeed real.This is supported by the clear changes in the abundance of HCO` from core to core (Fig. 2 and 3). One question is whether altering the ionization from core to core will result in observable chemical diÜerences. Indeed large changes in ion abundances have been associated with dramatic chemical variations towards the OMC-1 core.43 However it is probable that chemistry in OMC-1 is being altered by the intense radiation produced by the Orion A H II region. For the quiescent cores in our survey such dramatic variations are not likely although small variations in some species may be correlated with changes in the ion properties. 4.3 Stability of cores The electron abundance is a useful quantity because it allows an estimate of the ion» neutral collision rate and thus the strength of the coupling between the neutral gas and the magnetic –eld.This coupling is considered crucial to the support of molecular clouds against their self-gravity on the large scale where the coupling is strong and crucial to the gravitational collapse of dense cores on the small scale where the coupling is weak.1,44 The coupling of the neutral gas and the magnetic –eld can be described by several parameters including the ratio of ambipolar to free-fall time and the magnetic Reynolds number. All of these parameters are linearly proportional to the ion»neutral collision 59 R. Plume et al. frequency l or to the ion fraction Xe .5,45,46 Here we use the ratio of core size (R) to the in MHD cut-oÜ wavelength j0 .We denote this coupling parameter R/j as W . For strong 0 coupling W ?1 the –eld and the gas move together over a broad range of size scales a broad spectrum of MHD wavelengths can propagate above cut-oÜ and the gas can develop complex turbulent motions. For weak coupling W O1 the –eld and the neutrals can move independently wave motion is suppressed and the gas cannot be turbulent. Williams et al.20 derive W \5»8 for the low-mass cores in our sample. Following their formalism we –nd similar values of W for our high-mass cores. Thus the coupling in the typical low-mass core is marginal with a small but –nite range of MHD wavelengths allowed to propagate above cutoÜ in the core. W Bqff/qin where qff is the free-fall We can also write the coupling parameter as is the ion»neutral collision time and this ratio is nearly equal to q q time and AD/qff in q where is the timescale for ambipolar diÜusion.46 Thus the ìmarginal couplingœ deduced here is equivalent to the well known ratio of ca.10 which is used to justify the AD quasistatic treatment of ambipolar diÜusion in the formation of low-mass cores.1,47 These results imply that the cut-oÜ wavelength in the typical low-mass core is 0.01» 0.02 pc substantially less than the typical core radius. This implies that at most a small portion of a core can be completely cut oÜ from MHD wave propagation and that most of the core can propagate MHD waves. It is possible that this may also help account for the discrimination between high- and low-mass star-forming regions.The typical radius of our sample of low-mass cores is ca. 0.1 pc whereas the high-mass cores are ca. 50% larger. Therefore there is more gas in the high-mass cores that is decoupled from the magnetic –eld creating a larger reservoir of material with which to form massive stars or stellar clusters. Slog(X 5 Conclusions We have observed the J\1]0 transitions of C18O and H13CO` and the J\1]0 and 2]1 transitions of DCO` in 20 low-mass cores and 7 high-mass cores. We have converted these observations into CO HCO` and DCO` column densities assuming optically thin LTE conditions in the gas and the standard isotopic abundance ratios. The ratios of these column densities ([DCO`]/[HCO`] and [HCO`]/[CO]) were compared with the output of theoretical chemical models in order to determine the electron abundances (Xe) in dense cloud cores.Our results can be summarized as follows (1) With the use of observations to constrain the variables in our chemical calculations we have con–ned the electron abundances in our sample of dense molecular cloud cores to a relatively small range [7.5\log(Xe)\[6.5 with very little scatter about the mean values of e)T\[7.04 ; p\0.22 (for the low-mass cores) and Slog(Xe)T\[7.11 ; p\0.15 (for the high-mass cores). These results are similar to those determined in previous studies9,17,18 but we have tightened the range of X to an order of magnitude. e (3) The best-–t models use a normal radiation –eld (s\1) and there is no diÜerence in X between cores with stars and those without.Therefore we suggest that the electron abundances can be explained by the conventional picture of pure cosmic-ray ion- e ization (provided that fH2\5]10~17 s~1 which produces the best match between our models and observations) and there is no need to invoke strong internal or external ultraviolet or X-ray –elds to account for the observations. R/j or q (4) The coupling parameter (W ) measured as AD/qff is estimated to be 5»8 for the cores in our sample. This implies that the neutrals are only marginally coupled to 0 the –eld and that at most ca. 10% of a core radius can be cut oÜ from MHD wave propagation. These results justify the usual quasistatic treatment of ambipolar diÜusion in the formation of cores. 60 Electron abundance in dense cloud cores Paper 8/00076J; Received 2nd January 1998 References 1 F.H. Shu F. C. Adams and S. Lizano 1987 Annu. Rev. Astron. Astrophys. 1987 25 23. 2 C. F. McKee Astrophys. J. 1989 345 782. 3 B. G. Elmegreen Astrophys. J. 1979 232 729. 4 S. Casanova T. Montmerle E. D. Feigelson and P. Andreç Astrophys. J. 1995 439 752. 5 P. C. Myers and V. K. Khersonsky Astrophys. J. 1995 442 186. 6 J. Stutzki G. J. Stacey R. Genzel A. I. Harris D. T. JaÜe and J. B. Lugten Astrophys. J. 1988 332 379. 7 T. Xie M. Allen and W. D. Langer Astrophys. J. 1995 440 674. 8 A. Dalgarno and S. Lepp Astrophys. J. 1984 287 L47. 9 A. Wootten R. B. Loren and R. L. Snell Astrophys. J. 1982 255 160. 10 L. Mestel and L. Spitzer Mon. Not. R. Astron. Soc. 1956 116 503. 11 C. A.Beichman P. C. Myers J. P. Emerson S. Harris R. Mathieu P. J. Benson and R. E. Jennings Astrophys. J. 1986 307 337. 12 C. de Boisanger F. P. Helmich and E. F. van Dishoeck Astron. Astrophys. 1996 310 315. 13 E. A. Bergin P. F. Goldsmith R. L. Snell and W. D. Langer Astrophys. J. 1997 482 285. 14 J. Williams E. A. Bergin P. Myers R. Plume and P. Caselli Astrophys. J. 1998 in press. 15 E. A. Bergin R. Plume J. Williams P. Myers and P. Caselli Astrophys. J. 1998 in press. 16 E. A. Bergin W. D. Langer and P. F. Goldsmith Astrophys. J. 1995 441 222. 17 M. Gueç lin W. D. Langer R. L. Snell and H. A. Wootten Astrophys. J. 1997 217 L165. 18 M. Gueç lin W. D. Langer and R. W. Wilson Astron. Astrophys. 1982 107 107. 19 W. D. Langer in Protostars and Planets II ed.D. C. Black and M. S. Matthews University of Arizona Press Tucson 1985 p. 650. 20 J. L. Linsky A. Diplas B. E. Wood A. Brown T. R. Ayers and B. D. Savage Astrophys. J. 1995 451 L335. 21 M. Larsson S. Lepp A. Dalgarno C. Stroemhol G. Sundstriem V. Zengin H. Danared A. Kaellberg M. Af Ugglas and S. Datz Astron. Astrophys. 1996 309 1. 22 P. J. Benson and P. C. Myers Astrophys. J. Suppl. 1989 71 89. 23 J. Harju C. M. Walmsley and J. G. A. Wouterloot Astron. Astrophys. Suppl. 1993 98 51. 24 R. L. Dickman Astrophys. J. Suppl. 1978 37 407. 25 E. A. Bergin and W. D. Langer Astrophys. J. 1997 486 316. 26 T. J. Millar P. R. A. Farquhar and K. Willacy Astron. Astrophys. Suppl. 1997 121 139. 27 M. A. Frerking W. D. Langer and R. W. Wilson Astrophys. J. 1982 262 590.28 P. F. Goldsmith and W. D. Langer Astrophys. J. 1978 222 881. 29 D. A. Neufeld S. Lepp and G. J. Melnick Astrophys. J. Suppl. 1995 100 132. 30 U. J. Sophia J. A. Cardelli and B. D. Savage Astrophys. J. 1994 430 650. 31 D. C. B. Whittet M. F. Bode A. J. Longmore A. J. Adamson A. D. McFadzean D. K. Aitken and P. F. Roche Mon. Not. R. Astron. Soc. 1988 233 321. 32 D. C. B. Whittet A. J. Adamson W. W. Duley T. R. Geballe and A. D. McFadzean Mon. Not. R. Astron. Soc. 1989 241 707. 33 T. E. Graedel W. D. Langer and M. A. Frerking Astrophys. J. Suppl. 1982 48 321. 34 C. L. Joseph T. P. Snow II C. G. Seab and R. M. Crutcher Astrophys. J. 1986 309 771. 35 C. L. Joseph Astrophys. J. 1988 335 157. 36 E. B. Jenkins in Interstellar Processes ed. D. J. Hollenbach and H. A. Thronson Jr. Reidel Dordrecht 1987 p. 533. 37 H. M. Butner E. A. Lada and R. B. Loren Astrophys. J. 1995 448 207. 38 P. Caselli C. M. Walmsley R. Terzieva and E. Herbst Astrophys. J. submitted. 39 R. Plume D. T. JaÜe N. J. Evans II J. Martïç n-Pintado and J. Goç mez-Gonzaç lez Astrophys. J. 1997 476 730. 40 P. F. Goldsmith E. A. Bergin and D. C. Lis Astrophys. J. 1997 491 615. 41 O. M. Shalabeia and J. M. Greenberg Astron. Astrophys. 1995 303 233. 42 A. G. G. M. Tielens Astron. Astrophys. 1983 119 177. 43 H. Ungerechts E. A. Bergin P. F. Goldsmith W. M. Irvine F. P. Schloerb and R. L. Snell Astrophys. J. 1997 482 245. 44 C. F. McKee E. G. Zweibel A. A. Goodman and C. Heiles ref. 19 p. 327. 45 F. H. Shu Astrophys. J. 1983 273 202. 46 T. C. Mouschovias Astrophys. J. 1991 371 296. 47 T. C. Mouschovias Physical Processes in Interstellar Clouds Proceedings of the NATO Advanced Study Institute 1987 Reidel Dordrecht p. 453.

ISSN:1359-6640    

DOI:10.1039/a800076j

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

Faraday Discuss. 1998 109 61»69 Chemical processes in astrophysical radiation –elds P. C. Stancila and A. Dalgarnob* a Physics Division Oak Ridge National L aboratory P.O. Box 2008 Oak Ridge TN 37831-6372 USA b Harvard-Smithsonian Center for Astrophysics 60 Garden St. Cambridge MA 02138 USA The eÜects of stimulated photon emission on chemical processes in a radiation –eld are considered and their in—uence on the chemistry of the early universe and other astrophysical environments is investigated. Spontaneous and stimulated radiative attachment rate coefficients for H~ Li~ and C~ are presented. (1) (2) (3) X]Y]XY]hl X]e~]X~]hl X~]Y]XY]e~ (4) X]Y]hlb ]XY]hl]hlb (5) 1 Introduction In astrophysical environments lacking dust molecules can only be formed through gasphase chemistry.Two of the primary mechanisms involve radiative processes (i) the radiative association of two neutral species (or a neutral and an ion) and (ii) the negative-ion sequence of radiative attachment followed by associative detachment Until recently only spontaneous photon emission has been considered in investigations of reactions (1) and (2). If these processes occur in an environment with an intense IR radiation –eld there is a possibility that the reaction rate coefficients can be enhanced by stimulated eÜects and b X]e~]hlb ]X~]hl]hlb In the above processes l is the frequency of the stabilizing spontaneous emission photon and l the frequency of the radiation –eld. Dubrovich1 has suggested that the stimulated mechanism may operate in the recombination era of the early universe enhancing the abundance of LiH owing to the cosmic background radiation (CBR) –eld.We2 carried out explicit calculations for the formation of LiH via reaction (4) and found that indeed the rate coefficients are considerably enhanced; however other factors of the environment and chemistry conspired to keep 61 62 Chemical processes in astrophysical radiation –elds the stimulated eÜects from playing a signi–cant role in the formation of LiH in the early universe. Here we extend our work2 to the radiative attachment process (5) and investigate the in—uence of stimulated eÜects in the early universe supernova ejecta and other astrophysical environments. 1 (6) b (7) X~]hl]X]e~ El\hl\E]E0 E0 is the electron affinity of X and E is the (8) p7\Ak h cv l B2 g g0 p2 (9) p2(E)\(E 2 ]k c2 E E 0)2 g g ~ p7(E) (10) o(lb Tb)\8n c h 3 lb3 exp(hl 1 (11) p5(E)\ 8nhl3 p2(E) c3 2 Theory In ref.2 we outlined the extension of the theory of radiative association to include stimulated eÜects. Essentially the partial cross-section for a transition into a bound vibrational»rotational level of the resulting molecule is enhanced by the factor 1[exp([hl/kTb) for a blackbody radiation –eld characterized by the temperature Tb . The enhancement increases with increasing T or decreasing l. The theory has been applied to the formation of HD,3 HeH`,4 and LiH.2 The theory of stimulated radiative attachment is obtained in an analogous way.The negative ion X~ is created by reactions (2) and (5) and destroyed by the photodetachment process where the photon energy relative translational energy of the products X and e~. Cross-sections for reactions (2) and (7) are related through detailed balance5,6 0 ~ where g and g~ are the statistical weights of the atom and negative ion respectively k is the reduced mass and v the relative collision velocity. Eqn. (8) can be rewritten as 0 For a blackbody radiation –eld the photon density per unit frequency interval is b/kTb)[1 and the stimulated cross-section is related to the spontaneous cross-section by Combining eqn. (9)»(11) gives the total spontaneous plus stimulated radiative attachment cross-section p2`5(E Tb)\p2(E)]p5(E)o(l Tb) (12) \(E]E0)2 g~ 2kc2E p7(E) g 1[exp([hl/kTb) 0 63 P.C. Stancil and A. Dalgarno c T T and collision temperature can be b The total radiative attachment rate coefficient for written a 0/kTc) 2`5(Tc Tb)\ g g ~ c4nB1@2AkkT 1 B3@2 exp(E (13) ] 0P= El2p7(El)exp( c [El/kTc) dE 1[exp([El/kTb) l A 2 E0 Therefore given the photodetachment cross-section the electron affinity and the statistical weights rate coefficients can be readily obtained with the application of eqn. (13). 3 Results and Discussion Hó Fig. 1 presents the total radiative attachment rate coefficients for the formation of H~ obtained with the reliable photodetachment cross-sections of Sadeghpour et al.7 The cross-section consists of a combination of new R-matrix calculations and previous experimental and theoretical results and its accuracy is controlled by ensuring that a number of oscillator strength sum-rules are satis–ed.The spontaneous radiative attachment (Tb\0 K) results are in good accord with the previous calculations of Dalgarno and Kingston,8 however the rate coefficients used by Rawlings et al.9 are in serious disagreement. As the rate coefficients of Rawlings et al.9 have been adopted in the astrochemistry database of Millar et al.,10 the abundance of H~ may have been underestimated in some astrophysical models. Consequently H2 may have also been Fig. 1 Rate coefficients for the formation of H~ by spontaneous and stimulated radiative attach- T b\0 ment in a blackbody radiation –eld of temperature T b .Current results (thick lines) 3000 5000 10 000 12 500 15 000 20 000 30 000 40 000 50 000 K (bottom to top). Previous results for Tb\0 Dalgarno and Kingston8 (thin solid line) and Rawlings et al.9 (thin dot-dash line). 64 Chemical processes in astrophysical radiation –elds underestimated as it is primarily formed by (14) H~]H]H2]e~ in dust-free and relatively cool environments such as the early universe for redshifts z\100. Tb\3000 K but very pronounced The eÜect of stimulated emission is negligible for when Tb[10 000 K for all considered collision temperatures. Lió 7PE3@2. Total radiative attachment rate coefficients for the formation of Li~ are displayed in Fig.2. The rate coefficients were calculated using the R-matrix photodetachment results of Ramsbottom et al.11 We are unaware of any previous calculations except for the rate coefficients adopted by Stancil et al.12 for a model of the early universe lithium chemistry. Their adopted –t of 1.85]10~15 (T /300)0.62 exp([T /9300) cm3 s~1 should be c b\0 K) radiative attachment rate (T replaced by the –t to the present spontaneous c coefficients 1.53]10~15 (T /300)0.92 exp([T /4560) cm3 s~1. The temperature depen- c c dence is similar to that of H~. Since both species involve the photodetachment of an s electron resulting in a p-wave –nal state their photodetachment cross-section near threshold obeys the Wigner law p As for H~ the enhancement due to stimulated attachment is small for Tb\3000 K but increases signi–cantly for Tb[10 000 K.Có Total radiative attachment rate coefficients for the formation of ground state C~ (4S°) are presented in Fig. 3. The calculations used the measured photodetachment crossline is the adopted spontaneous Fig. 2 Rate coefficients for the formation of Li~ by spontaneous and stimulated radiative attach- Tb . Current results (thick lines) T ment in a blackbody radiation –eld of temperature b\0 3000 5000 7000 10 000 12 500 15 000 20 000 30 000 40 000 50 000 K (bottom to top). The thin solid Tb\0 K rate coefficient of Stancil et al.12 65 P. C. Stancil and A. Dalgarno b . Current results (thick Fig. 3 Rate coefficients for the formation of the ground state C~(4S°) by spontaneous and stimulated radiative attachment in a blackbody radiation –eld of temperature T lines) Tb\0 5000 7000 10 000 12 500 15 000 20 000 30 000 40 000 50 000 K (bottom to top).The thin solid line is the adopted spontaneous Tb\0 K rate coefficient of Millar et al.10 T Fig. 4 Rate coefficients for the formation of the metastable state C~(2D°) by spontaneous and stimulated radiative attachment in a blackbody radiation –eld of temperature Tb . Current results (thick lines) Tb\0 500 1000 2000 3000 5000 7000 10 000 12 500 15 000 20 000 30 000 40 000 50 000 K (bottom to top). The thin solid line is the adopted spontaneous b\0 K rate coefficient of Millar et al.10 for the ground state. 66 T Fig. 5 Radiative attachment rate coefficients for a model negative ion with arbitrary electron affinity for Tc\2000 K spontaneous attachment Tb\0 K (»»); spontaneous and stimulated attachment Tb\5000 K (… … …).The electron affinities of the –rst thirty elements are given along the top of the –gure (]). Explicit results from Fig. 1»4 are given for comparison Tb\0 ( K K) ; b K \5000 (+). The statistical weights are set to unity in all cases. (Tb\0 K) radiative attachment rate coefficients should be used for Tc\1000 K. section of Seman and Branscomb13 for E between 1.5 and 2.04 eV the R-matrix calculations of Ramsbottom et al.14 for El[2.04 eV and a –t to a Wigner threshold law for l El\1.5 eV. We are unaware of any previous calculations of the spontaneous radiative attachment rate coefficients but a temperature-independent value of 3.0]10~15 was adopted by Millar et al.10 The Millar et al.10 value is ca.30% larger than the current results suggesting that some astrochemistry models may have slightly overestimated the C~ abundance. The –t 2.25]10~15 exp([T /9000) cm3 s~1 for the present spontaneous c Results for the formation of the metastable C~(2D°) are presented in Fig. 4. The rate Similar to H~ and Li~ stimulated eÜects only become important for T coefficients were obtained using the R-matrix photodetachment cross-sections of Ramsbottom et al.14 for El[0.067 eV and a –t to a Wigner threshold law for smaller energies. b[10 000 K for the ground state but for the metastable state the enhancements can be signi–cant to T as low as ca. 500 K. As the electron affinity decreases (4S° 1.263 eV; 2D° 0.035 eV) the stimulated enhancement increases which is related to the appearance of El\E b ]E in the enhancement factor in eqn.(13). To further visualize this behaviour Fig. 5 displays the spontaneous and total radiative attachment rate coefficients for a model 0 negative ion with arbitrary electron affinity. 4 Astrophysical applications The early universe The rationale for investigating stimulated eÜects on molecular formation was motivated by studies of chemistry in the early universe. The early expanding universe is embedded Chemical processes in astrophysical radiation –elds 67 P. C. Stancil and A. Dalgarno in the CBR –eld left over from the big bang with a temperature that decreases with decreasing redshift z ; Tb\2.726(1]z) K.The temperature of the gas Tc is coupled to that of the CBR –eld until the beginning of the recombination era (zB1300) after T which b[TcP(1]z)2. The CBR –eld is sufficiently intense to prevent signi–cant formation of negative ions until after zB200 owing to the high efficiency of photodetachment. Since by then T had fallen to less than 600 K it is clear from Fig. 1 and 2 b that stimulated radiative attachment played an insigni–cant role in the formation of H~ and Li~ in the early universe. Additionally there is no increase in the abundances of H2 and LiH due to associative detachment process (3) following stimulated radiative attachment. Molecules do not begin to form in signi–cant abundances until z\500. At larger redshifts they are primarily destroyed by photodissociation.The eÜects of stimulated emission are generally larger for radiative association than radiative attachment for a comparable T and even appreciable for T as low as 500 K.2h4 This is because the b b (E energy of the stabilizing photons can be much smaller l\0.2 eV) for transitions into high-lying bound rotational»vibrational levels than the minimum photon energy for a negative ion the electron affinity. Since TbB1300 K at z\500 it is possible that stimulated radiative association may contribute to molecule formation. Stancil and Dalgarno2 and Zygelman et al.4 –nd that the abundances of LiH and HeH` are enhanced by ca. 25%. However this is only marginally signi–cant considering that the fractional abundances (with respect to hydrogen nuclei) are ca.10~17 and ca. 10~15 for LiH and HeH` respectively.12,15 Supernova ejecta While the environmental conditions of a supernova (SN) appear to be hostile to the formation of molecules CO and SiO were observed in the ejecta of the Type II object SN 1987A.16 The visible and IR spectrum of SN 1987A consisted of photospheric-like emission from the central region with an eÜective temperature ca. 5000 K and an IR excess whose intensity increased relative to the photosphere. The IR excess is believed to be due to dust which probably formed in the same region as the molecules the molecules acting as the dust nucleation sites. The chemistry of CO prior to the production of dust was described by Lepp et al.,17 the primary reactions being radiative association (15) C]O]CO]hl and the negative ion sequence (16) C]e~]C~]hl (17) C~]O]CO]e~ Stimulated eÜects are probably small for reaction (15) since CO has a large dissociation energy (11 eV compared to only 2.0 and 2.4 eV for HeH` and LiH respectively).Stimulated radiative attachment may play a role as suggested by Fig. 3 and 4. Wooden et al.18 have modelled the photosphere and dust emission by –tting the spectra of SN 1987A to scaled Planck functions. Using their –ts an estimate of a blackbody dilution factor consisting of geometric dilution and departure from a blackbody intensity can be made. At 260 days the dilution factor from the photosphere to the CwO region of the ejecta is ca. 2]10~3. Given the dilution factor the photospheric temperature of 5000 K and the CwO gas temperature of 2000 K,19 the total C~ (4S° and 2D°) radiative attachment rate is enhanced by \0.05%.As the dust is formed in the CwO region we take its dilution factor to be of the order of unity. Given the dust temperature of 700 K the formation of C~ is enhanced by 0.5%. While this suggests 68 Chemical processes in astrophysical radiation –elds that CO formation is not aÜected by stimulated eÜects due to the continuum radiation from the supernova line emission from the fundamental CO band (ca. 4.5 lm) may provide some enhancement as the line-to-continuum —ux ratio is ca. 10. The situation is similar for H~ which considering the photospheric and dust radiation is only enhanced by 0.03%. There is a possibility that the abundance of Fe~ may be aÜected by stimulated processes since the iron core of the ejecta is in closer proximity to the photosphere and its electron affinity (0.16 eV) is smaller than H~ or C~.Fig. 5 shows that enhancements due to stimulated radiative attachment increase with decreasing electron affinity. If the total dilution factor between the photosphere and iron zone is greater than 4]10~3 then the enhancement in the Fe~ formation is [1%. Two emission lines in SN 1987A have been tentatively identi–ed by Miller et al.20 to be due to HeH`. It is primarily formed in the hydrogen envelope of the ejecta by the radiative association process (18) He`]H]HeH`]hl but (19) He]H`]HeH`]hl may contribute. Zygelman et al.4 have calculated the rate coefficients for reaction (19) and for the stimulated radiative association reaction (20) He]H`]hlb ]HeH`]hl]hlb The analogous stimulated process for reaction (18) is probably negligible except for T large as the minimum stabilizing photon energy is 10.99 eV.Similar to the CwO b zone the dilution factor for the hydrogen envelope is ca. 8]10~4 from the photosphere and 0.6 from the dust at 260 days. Stimulated association enhances the formation of HeH` by 0.2% and 9% due to the photosphere and dust respectively compared to reaction (19). However reaction (18) produces ca. 85% of the HeH` (assuming a hydrogen ionization fraction of 0.01) so that the total HeH` production is only enhanced by 1.5%. Other environments Dubrovich and Lipovka21 have studied the formation of HeH` in high red-shift Lyman a clouds irradiated by a background quasar and suggest that HeH` could be detected with millimetre observations.However the HeH` abundance produced by spontaneous radiative association reactions (18) and (19) is probably not much larger than in the intergalactic medium (comparable to the early universe value of 10~15) and Dubrovich1 has suggested that HeH` could be enhanced by reaction (20) due to the quasar radiation. Their models suggest that the HeH` abundance is signi–cant if the Lyman a cloud is [1023 cm from the quasar. Assuming the quasar to radiate as a blackbody with a radius of 1019 cm one obtains a geometric dilution factor of 10~8 suggesting stimulated eÜects to be negligible. The situation is similar in a planetary nebulae but more severe.For the white dwarf eÜective temperature of 175 000 K for NGC 7027 molecules begin to form for distances greater than 1016 cm from the central star,22 corresponding to a dilution factor of 10~14. One possibility where stimulated eÜects may play a role is in quasar broad-line clouds. Kallman et al.23 have found that the fractional abundance of H can reach 0.5 at 2 a depth of 1014 cm into the cloud. Given that this distance is much smaller than the radius of the quasar there is no geometric dilution. For a quasar eÜective temperature of 50 000 K this implies that the abundances of H~ and HeH` could be enhanced by 69 P. C. Stancil and A. Dalgarno factors of up to 5 and 20 respectively. However a detailed model of a broad-line cloud will be necessary to test this prediction.5 Summary Recent work on stimulated emission eÜects in the formation of diatomic molecules has been extended to stimulated radiative attachment to form negative atomic ions. Rate coefficients for the production of H~ Li~ and C~ have been given for a range of collision and radiation temperatures. Corrections to currently adopted spontaneous radiative attachment rate coefficients have been indicated. Radiative association and attachment are signi–cantly enhanced by stimulated emission due to a blackbody radiation –eld with the enhancement to association occurring at somewhat smaller radiation temperatures than for attachment. However in the astrophysical environments considered thus far this enhancement is generally negated by other mitigating circumstances geometric dilution being the primary culprit.Geometric dilution appears to be particularly severe in planetary nebulae and quasar irradiated Lyman a clouds. Minor enhancements in molecule formation may be found in Type II supernova ejecta as the dilution is more moderate. While dilution is not a concern in the early universe only modest enhancements are obtained for molecules since large radiation temperatures are very eÜective at removing molecules and negative ions by photodestruction processes. As quasar broad-line clouds are thought to be nearly fully molecular at their core and are in close proximity to the intense quasar radiation they may be the prototypical environment for stimulated formation eÜects.We thank H. R. Sadeghpour for providing the H~ photodetachment cross-section. A.D. acknowledges support from NSF grant AST 95-317900. The work of P.C.S. was performed as a Eugene P. Wigner Fellow and staÜ member at the Oak Ridge National Laboratory managed by Lockheed Martin Energy Research Corp for the U.S. Department of Energy under Contract DE-AC05-96OR22464. References 1 V. K. Dubrovich 1996 personal communication. 2 P. C. Stancil and A. Dalgarno Astrophys. J. 1997 479 543. 3 P. C. Stancil and A. Dalgarno Astrophys. J. 1997 490 76. 4 B. Zygelman P. C. Stancil and A. Dalgarno Astrophys. J. in press. 5 H. S. W. Massey Negative Ions Cambridge University Press Cambridge 1950. 6 L. M. Branscomb in Atomic and Molecular Processes ed.D. R. Bates Academic Press New York 1962. 7 H. R. Sadeghpour A. Dalgarno and P. C. Stancil 1998 in preparation. 8 A. Dalgarno and A. E. Kingston Observatory 1963 83 39. 9 J. M. C. Rawlings D. A. Williams and J. Canto Mon. Not. R. Astron. Soc. 1988 230 695. 10 T. J. Millar P. R. A. Farquhar and K. Willacy Astron. Astrophys. Suppl. 1997 121 139. 11 C. A. Ramsbottom K. L. Bell and K. A. Berrington J. Phys. 1994 27 2905. 12 P. C. Stancil S. Lepp and A. Dalgarno Astrophys. J. 1996 458 401. 13 M. L. Seman and L. M. Branscomb Phys. Rev. 1962 125 1602. 14 C. A. Ramsbottom K. L. Bell and K. A. Berrington J. Phys. B 1993 26 4399. 15 P. C. Stancil S. Lepp and A. Dalgarno Astrophys. J. in press. 16 W. P. S. Meikle D. M. Rank J. D. Bregmau F. C. W. Heboru A. G. G. M. Tielens M. Cohen P. A. Pinto and T. S. Axelrod Mon. Not. R. Astron. Soc. 1989 238 193. 17 S. Lepp A. Dalgarno and R. McCray Astrophys. J 1990 358 262. 18 D. H. Wooden D. A. Allen J. Spyromilio and G. F. Varani Astrophys. J. Suppl. 1993 88 477. 19 W. Liu and A. Dalgarno Astrophys. J. 1995 454 472. 20 S. Miller J. Tennyson S. Lepp and A. Dalgarno Nature (L ondon) 1992 355 420. 21 V. K. Dubrovich and A. A. Lipovka Astron. Astrophys. 1995 296 309. 22 C. Cecchi-Pestellini and A. Dalgarno Astrophys. J. 1993 413 611. 23 T. Kallman S. Lepp and P. Giovannoni Astrophys. J. 1987 321 907. Paper 8/00074C; Received 2nd January 1998

ISSN:1359-6640    

DOI:10.1039/a800074c

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

Faraday Discuss. 1998 109 71»82 Low temperature electron attachment to polycyclic aromatic hydrocarbons Tou–k Moustefaoui Christiane Rebrion-Rowe Jean-Luc Le Garrec Bertrand R. Rowe and J. Brian A. Mitchell Physique des Atomes L asers Moleç cules et Surfaces U.M.R. du C.N.R.S. no. 6627 Universiteç de Rennes I Campus de Beaulieu 35042 Rennes France Electron attachment to anthracene vapour has been measured between 48 and 300 K using the CRESU technique. Preliminary measurements have found values between 1 and 3]10~9 cm3 s~1 that do not vary greatly with temperature. Polycyclic aromatic hydrocarbons (PAHs) are multi-ring molecular compounds that contain two or more benzene rings. They can take many isomeric forms and are made up of carbon and hydrogen atoms although variants on the basic form may include other species such as nitrogen oxygen sulfur halogen atoms etc.In recent years a strong interest in PAHs has arisen among the astrophysical community. They have been proposed as a candidate for the source of the so-called diÜuse interstellar bands seen in absorption spectra between 4000 and 13 000 ” from stars located behind interstellar clouds.1h4 The possibility that PAH type molecules could be the origin of the unidenti- –ed IR emission bands between 3 and 11 lm in the spectra of nebulae was –rst proposed by Duley and Williams5 and later ampli–ed by Leger and Puget,6 Allamandola et al.7 and by Duley and Jones.8 The interstellar environment is harsh being bathed in UV photons that can dissociate weakly bound molecular species.The source of these features must have a very stable structure and PAHs –t this bill very well. Their stability is very well recognized by their ability to survive and —ourish in the harsh environments found in combustion. Chromatographic studies of the emissions from hydrocarbon –res have identi–ed a number of PAHs (generally referred to as products of incomplete combustion or PICs) to be important components of smoke plumes,9 although it is interesting to note that despite the bewildering complexity of the chemistry of fuel-rich —ames only a few individual species tend to dominate suggesting that either their formation is energetically favored or that they exhibit superior thermal stability. Much experimental and theoretical eÜort has been mounted in order to provide an understanding of the physical and chemical processes governing the formation and lifecycle of PAHs in interstellar and stellar environments.Frenklach and Feigelson10 have presented a model (based upon previous combustion studies) which has shown that indeed PAHs can be formed in circumstellar envelopes in signi–cant quantities under favorable conditions. It was recognized early on in this story that in fact the source of the DIBs would actually be ionized PAHs due to the high —ux of ionizing photons typically encountered in interstellar media. In a synchrotron radiation experiment Jochims et al.11 have examined the photostability of a number of PAHs and have found that in an environment bathed in photons with energies up to 13.6 eV (i.e.typical of the interstellar environment) neutral PAHs with less than 30»40 carbon atoms are likely to be photodissociated. Their results on the other hand indicate that PAH cations are more likely to respond to UV photon absorption by radiative relaxation regardless of their size and therefore will be much more resistant to photodestruction. 71 72 L ow temperature electron attachment to polycyclic aromatic hydrocarbons One of the –rst examples of the modelling of the physical state of interstellar PAHs was that of Omont12 who computed the charge state distribution of these molecules assuming large values for the photoionization electron attachment and electron»ion recombination rates. A number of similar studies have followed.13h16 It is interesting to point out that in a recent model,16 the use of PAH electron»ion recombination rates measured in our laboratory17 to be considerably smaller than previously estimated,13 has led to the prediction of the positively charged fraction of small compact PAHs such as pyrene and coronene being increased from 2% to 60»70% at moderate hydrogen densities.This is a very clear indication of the need for accurate laboratory data as the basis for such models. It is also recognized that these molecules may attach electrons to form anions and that this may represent a signi–cant sink for electrons which would therefore not be available to participate in the gas phase ion»molecule chemistry.18h20 This would have a signi–cant impact on the various mechanisms that have been proposed and modelled for the formation of small interstellar molecules.The absence of free electrons would also have important implications for the magnetic structure and stability of interstellar clouds. Tielens21 and co-workers have modelled the non-dissociative attachment of electrons to PAHs represented by the reaction (1) PAH]e A8B (PAH~)* »»» kr ’ PAH~ kc c ka where k k k is the electron capture rate is the autoionisation rate and is the radiative a stabilisation rate by calculating a ì sticking coefficient œ given by r (2) S(e)\k kr r]ka The basis of their analysis is that the sticking coefficient depends mainly upon the electron affinity of the PAH since as this increases the density of states of the excited anion increases and the less likely it will be that all the energy will be located in the autoionisation channel within its radiative lifetime.Using this assumption Allamandola et al.22 found that the sticking coefficient becomes close to unity for electron affinities (Eeas) of the order of 1 eV. For PAHs with Eeas less than about 0.5 eV the sticking coefficient is estimated to be very small. Table 1 lists a number of PAHs with their measured or calculated electron affinities (in eV). This table is taken from ref. 23 and the values shown are the lowest and highest values quoted in Table II of that reference. The values display a great deal of scatter and following the lead of Tobita et al.,24 we have also listed the average of the values presented in ref.23. Tobita et al.,24 using an electron impact apparatus have performed a study of electron attachment to selected PAH compounds. These authors bombarded a number of PAH vapours with an energy selected electron beam and searched for the resulting production of negative ions using a quadrupole mass spectrometer. In this way they were able to detect the formation of negative ion resonances as a function of electron energy and to measure the peak formation cross section. They found that azulene (Eea\0.72 eV) anthracene (Eea\0.59 eV p\7.5]10~17 cm2) pyrene (Eea\0.56 eV p\1.5]10~17 cm2) displayed resonances at zero energy but with the exception of the latter compound the attachment cross sections were rather small. Thus it is to be expected that these compounds would have rather small attachment rates at low temperatures.Larger cross sections were found for the compounds —uoranthene (E eV p\1.1]10~15 cm2) perylene (E ea\0.77 ea\0.91 eV p\1.1]10~14 cm2) and tetracene (Eea\1.05 eV p\3.6]10~15 cm2) but these were for resonances occurring at energies above zero and so one would perhaps not expect that the low temperature attachment T . Moustefaoui et al. Table 1 Electron affinities (measured or calculated) for selected PAHsa formula compound naphthalene fulvalene azulene acenaphthalene biphenyl —uorene anthracene phenanthrene —uoranthene pyrene naphthacene 1,2-benzanthracene 3,4-benzophenanthrene chrysene triphenylene perylene 1,2-benzopyrene tetracene 4,5-benzopyrene anthanthrene indeno[1,2,3-cd]—uoranthene 1,2-benzotetracene 1,2-benzotetraphene pentacene 1,2,3,4-dibenzanthracene 1,2 :5,6-dibenzanthracene 1,2 :7,8-dibenzanthracene picene 2,3 :6,7-dibenzophenanthrene 2,3 :6,7-dibenzophenanthrene coronene C10H8 C10H8 C10H8 C12H8 C12H10 C13H10 C14H10 C14H10 C16H10 C16H10 C18H12 C18H12 C18H12 C18H12 C18H12 C20H12 C20H12 C20H12 C20H12 C22H12 C22H12 C22H14 C22H14 C22H14 C22H14 C22H14 C22H14 C22H14 C22H14 C22H14 C24H12 a Taken from ref.23; values in parentheses are the average of the values presented in that reference. (Eea\0.23 eV) and triphenylene (Eea\0.36 eV). rates would be very large. No long-lived low-lying negative ion resonances were found (E for compounds with low electron affinities such as naphthalene ea\0 eV) phenanthrene Experimental information concerning thermal rate coefficients for electron attachment to PAHs is rather limited and sometimes contradictory.Early studies of the attachment to a number of PAHs were performed by Christophorou and coworkers25,26 using a drift tube apparatus; they found the very small value of 3.1]10~13 cm3 s~1 for naphthalene at 400 K. This was in contrast to the 400 K value of 1.5]10~9 cm3 s~1 deduced by Johnsen and Sauer27 from pulse-radiolysis data. Given that naphthalene has a negative electron affinity23 it would seem that the drift tube data is more likely to be correct. Christophorou et al.26 found a 380 K value of about 2]10~9 cm3 s~1 for anthracene. The drift tube data displays a steep fall-oÜ with increasing electron energy but in fact little change in the ambient temperature rate coefficient was found25 when the temperatures of the vapour and of the electrons were simultaneously varied from 380»480 electron af–nity/eV [0.39»0.22 ([0.023) 1.73»1.75 (1.74) 0.52»1.03 (0.68) 0.42»0.98 (0.71) [0.23»1.02 (0.2) [0.4»0.28 (0.03) 0.4»1.0 (0.65) [0.02»0.69 (0.23) 0.63»0.92 (0.78) 0.37»0.81 (0.56) 0.79»1.3 (1.05) 0.42»0.66 (0.57) 0.18»0.55 (0.36) 0.26»0.52 (0.41) [0.28»0.29 (0.01) 0.71»1.06 (0.91) 0.35»0.61 (0.81) 0.79»1.3 (1.05) 0.35»0.61 (0.49) 1.02»1.41 (1.2) 1.33»1.37 (1.35) 1»1.11 (1.06) 0.76»0.8 (0.78) 1»1.52 (1.23) 0.34»0.57 (0.47) 0.37»0.98 (0.62) 0.33»1.1 (0.63) 0.23»0.98 (0.52) 0.44»0.7 (0.59) 0.26»0.37 (0.32) 0.18»1.88 (0.7) 73 74 L ow temperature electron attachment to polycyclic aromatic hydrocarbons K.Their result is in reasonable agreement with a more recent —owing afterglow Langmuir probe (FALP) measurement of Canosa et al.,28 who found a 300 K value of 1]10~9 cm3 s~1. Christophorou and Blaunstein25 found a very small rate for phenanthrene (as expected) but the measured rate for triphenylene was only about a factor of –ve less than that for anthracene in contradiction to what one would expect from the results of Tobita et al.24 and from its rather weak electron affinity. The rate for chrysene was similar to that of triphenylene and pyrene had a rate about a factor of two smaller again.Perylene on the other hand was found to display a rate about a factor of two higher than anthracene which one might perhaps not have expected given that Tobita et al. found that the negative ion resonance for this ion occurred at an energy above zero. If one examines the negative ion resonance that they found however which is centred around 0.5 eV one sees that it is very broad and extends down to zero energy. The peak cross section for this resonance is two orders of magnitude greater than that found for anthracene so this can account for the large measured rate for this species. Recent studies in this laboratory29 have shown that one can obtain very diÜerent results for electron attachment by lowering the temperature of the attaching molecules in step with the electron energy as opposed to varying the energy of electrons colliding with constant temperature molecules.For this reason therefore we decided to investigate the electron attachment to anthracene at low temperatures using the CRESU technique which is discussed in the next section. We have performed this experiment at 48 75 170 and 300 K and preliminary results are presented in this paper. Apparatus The use of the CRESU technique for the measurement of electron attachment rate coef- –cients has been described in detail previously29 and only its principal features are discussed here. An essential element of the apparatus shown in Fig. 1 is the Laval nozzle through which a buÜer gas (nitrogen in this case) —ows continuously leading to the Fig.1 Diagram of the CRESU apparatus showing the retractable nozzle the ionization region and the movable mass spectrometer/Langmuir probe assembly 75 T . Moustefaoui et al. Table 2 Aerodynamic characteristics of the nozzles used in this experiment nozzle temperature/K jet pressure/mbar density/1016 cm~3 jet velocity/104 cm s~1 7.11 6.73 5.08. 2.74 1.67 0.572 0.190 0.174 0.132 47.7 74.5 169.7 NN 2 45 K N 170 K 2 75 K 2 formation of a supersonic jet whose core is uniform and isentropic. The aerodynamic conditions in the jet are determined by the shape of the nozzle and a diÜerent nozzle is used for each temperature measurement. The characteristics of the nozzles used in this study are listed in Table 2. The nozzles are attached individually to a movable gas reservoir mounted inside the large vacuum chamber.To obtain good —ow uniformity the pressure in the chamber must be held equal to that of the jet (typically working pressures are of the order of several tenths of millibars) and in order that the —ow maintains an isentropic character at its centre the Reynolds number must stay sufficiently large which means that the width of the nozzle and the mass —ow must be quite high. From this it follows that a very large pumping capacity of 24 000 m3 h~1 must be employed in this apparatus. At ambient temperatures anthracene is a solid white powder. In order to inject it into the —ow in vapour form it must be heated to 190 °C in an oven. This vapour is entrained into a —ow of nitrogen gas which is then fed through a delivery tube and injected into the carrier gas —ow through an eight-needle entry port located at the upstream entrance to the nozzle.In order to avoid condensation of the anthracene vapour the delivery tube and injection needles are themselves heated to 200 °C. If qN2 is v the nitrogen —ow rate p the vapour pressure of anthracene at the oven temperature and ptot the total pressure in the oven the —ow rate of anthracene qant emerging from the oven can be estimated to be (3) qant\qN2 ptot[pv pv At the exit of the nozzle the jet is crossed at 90 degrees by a 10 lA beam of electrons at an energy of 10 keV and this results in the formation of a weakly ionised plasma with an initial electron density of 108»109 cm~3.Further downstream a mobile quadrupole mass spectrometer allows the ions present in the core of the —ow to be sampled as a function of distance downstream from the electron beam. Ions selected with a given m/z ratio are then detected using a channel electron multiplier coupled to a counting chain and data acquisition computer. A Langmuir probe attached to the outer casing of the mass spectrometer can be used to measure the electron density in the —ow also as a function of distance. (4) Measurement of the attachment rate coefficient (i) Langmuir probe method The equation that governs the development of the electron density ne in the presence of an attaching gas R can be written v dne dx \[b[R] where [R] is the density of the gas R and b is the electron attachment coefficient v is the —ow speed and x is the distance along the —ow axis.EÜects such as diÜusion and electron»ion recombination which could also diminish the electron density along the 76 L ow temperature electron attachment to polycyclic aromatic hydrocarbons —ow have been shown to be negligible in the CRESU apparatus29 under the conditions used in this experiment and so can be neglected. Since [R]A[ne] the value of [R] does not change along the —ow and so one works in pseudo –rst-order conditions. The value of the velocity is also constant due to the (5) uniformity of the —ow and therefore n ([R])\ integration n ([R] )exp isAstraightforward [ b[R]x yielding e e 0 v B For a given [R] therefore the logarithm of the decrease of the electron density with x is a straight line whose slope p is given by [b[R]/v.This is illustrated in Fig. 2. As discussed in the following section we were limited to using rather small —ow rates of anthracene in this experiment and since the attachment rate is not very large for this molecule the slope of this line was rather shallow. There was also quite a bit of scatter in the data points. In order to improve on the accuracy of our rate coefficient determination the experiment was repeated for diÜerent anthracene densities and then b[R]/v was plotted against [R] (see Fig. 3) in order to improve the statistics of the measurement. The velocities of the various —ows used have been well characterised previously (see Table 2) and the velocities are accurately known so that the attachment rate coefficients could be determined directly from this measurement.Results. This measurement proved to be problematic for a number of reasons. Firstly it was found that the Langmuir probe yielded non-reproducible values for the electron density during preliminary experiments. This was traced to the condensation of anthracene onto the surface of the probe thus producing an insulating layer and altering the collection efficiency for electrons. This problem was solved by using a continuously heated probe so that anthracene condensation did not occur on its surface. A more serious problem arose however which has been traced to the loss of anthracene from the —ow due to condensation onto the walls of the nozzle reservoir.(Indeed during one measurement using high anthracene —ow rates a nozzle throat became completely blocked.) In order to limit this problem the measurements were made over a rather restricted range of anthracene densities so that condensation onto the walls and successive build-up of a layer on the interior surface of the nozzle reservoir was minimised. This measurement was performed at a single temperature (75 K) and Fig. 2 shows the measured electron density as a function of anthracene —ow rate taken as a function of distance along the —ow. The solid dots correspond to measurements taken without anthracene injection and it can be seen that indeed loss of electrons due to diÜusion is minimal in this apparatus. The inverted triangles show data taken with an anthracene density of 1.46]1012 cm~3.It can be seen that the decrease is rather linear with distance. Other similar data (not shown) were taken for densities of 1.05]1012 and 2.47]1012 molecules of anthracene cm~3. Fig. 3 shows the ratio of the rate coefficient times the anthracene density over the —ow speed plotted as a function of anthracene density. From this plot a value of the attachment rate at 75 K equal to 1.0]10~9 cm3 s~1 was obtained. Because of the difficulties encountered with this method a second approach to the determination of attachment rates was applied. (ii) Comparative method In this method which we have used previously in the FALP measurement of the attachment rate of anthracene,28 a mass spectrometer is used to sample the negative ions formed as a result of electron attachment processes in the —ow due to the presence of two attaching gases.The increase of the two negative ion signals as a function of distance along the —ow is determined and from this a relative rate coefficient for the attachment reaction can be determined. The measurement of the rate of increase of a 77 T . Moustefaoui et al. Fig. 2 Langmuir probe measurement of electron density as a function of distance along the —ow. Solid circles Ö without anthracene; solid triangles @ with anthracene. Flow temperature 75 K. single negative ion in itself is sufficient for the determination of the attachment rate coefficient (in that it is complementary to the measurement of the decay of the electron density described above) but it suÜers from the same inaccuracy due to the weak attachment rate and the small quantities of anthracene used.The strength of a relative measurement in which the rate of attachment to one species is compared to that to a second species whose rate is accurately known is that one can be more con–dent that one has eliminated experimental artefacts from the measurement and again one can improve upon the statistical accuracy by making several comparisons. The main difficulty with Fig. 3 Plot of b[anthracene]/v vs. anthracene density for a —ow temperature of 75 K. The attachment rate coefficient is determined from the slope of this graph. 78 L ow temperature electron attachment to polycyclic aromatic hydrocarbons this method is due to the fact that one does not know a priori the collection and detection efficiency of the mass spectrometer for each individual ion and thus absolute concentrations of the negative ions cannot be determined.This can pose problems if there is a mass discrimination eÜect associated with the instrument where the collection efficiency for two ions of diÜering masses is not the same. One of the best characterised and most accurately measured electron attachment rate coefficients is that for the electrically insulating gas SF6 . There are basically two reasons for this the –rst being that it is an important practical quantity for high voltage electrical insulation applications ; the second being that it has a large value that does not vary greatly with temperature.The most accurate measurement is believed to be that of Petrovic and Crompton30 who found a 300 K value of 2.27]10~7 cm s~1. We have measured the rate coefficient for electron attachment to SF using the CRESU 6 apparatus29 and our results are tabulated in Table 3. Given that this is a well known quantity and that it is very easy to measure using our method it is possible to use it as a standard against which relative rate coefficients for other species can be compared and calibrated. The basic idea behind this method is to measure the negative ions formed via electron attachment in the —ow using the moveable mass spectrometer. If [M~] is the density of the corresponding negative ion created by electron attachment to molecule M where M is SF or anthracene then we can write 6 (6) v d[M~]\b[M][e] dx where v is the —ow velocity b the attachment rate coefficient and [e] is the electron density.Integrating eqn. (6) yields (7) [M~]\b[M] P[e] dx v with the condition that the neutral density is independent of x i.e. [M]A[e]. Hence at a –xed position of x by taking the individual equations for the two species one can write the ratio for the SF and anthracene negative ion densities thus 6 [SF6~] b 14H10~] \bant[C14H10] SF6 [SF6] Anthracene was introduced into the —ow as described previously. Since b (8) [C Since the densities of SF and anthracene are established by the entry —ow rates and 6 b is well known the attachment rate for anthracene can be determined from eqn. (8) by measuring the ratio of the negative ion densities.It should be noted that this mea- SF6 surement is independent of the —ow velocity. is so SF6 b much larger than only a small quantity (B0.01 cm3 min~1) ofFwas needed; ant S6 Table 3 Electron attachment rate coefficientsa FALP 300 K CRESU 300 K CRESU 169.7 K CRESU 74.5 K CRESU 47.7 K 2.75]10~7 1.0]10~9 » k(SF6)a k(ant)b k(ant) norm.c » 3.7]10~10 1.0]10~9 1.7]10~7 9.75]10~10 2.6]10~9 1.3]10~7 3.1]10~10 8.4]10~10 1.4]10~7 6.11]10~10 1.7]10~9 a For SF6 measured using the CRESU and FALP apparatuses.28,29 b For anthracene measured in this study using the CRESU and in a previous experiment28 using the FALP. c For anthracene normalised to the FALP study as discussed in the text.79 T . Moustefaoui et al. undergo electron attachment. thus a sample of SF diluted in nitrogen was used. The relative intensities of the two negative ions thus produced due to electron attachment were measured using the mass 6 spectrometer. A typical mass spectrum thus obtained is shown in Fig. 4. Fig. 5 shows the development of the intensities of the SF6~ and C14H10~ ion peaks as a function of distance along the —ow and it can be seen that although somewhat noisy both curves increase steadily indicating that both SF and C 6 The ratio given by eqn. (8) was then measured for several diÜerent anthracene den- 14H10 sities the SF density and the measuring position being kept constant. Fig. 6 shows the result of this measurement taken at a —ow temperature of 47.7 K.Similar plots were 6 Fig. 4 Mass spectrum taken with a mixture of SF and anthracene vapour injected into the —ow of nitrogen buÜer gas 6 Fig. 5 Measured ion density of SF6~ and C14H10~ as a function of distance along the —ow 80 L ow temperature electron attachment to polycyclic aromatic hydrocarbons 6 Fig. 6 Ratio of densities of C14H10~ to SF6~ plotted as a function of increasing anthracene density (the density of SF was kept constant) for a —ow temperature of 48 K and at a position 16 cm downstream from the electron beam obtained for —ow temperatures of 74.5 and 169.7 K. A fourth measurement was made in which the pressure in the expansion chamber of the CRESU was allowed to rise so that in fact a very weak expansion was produced across the Laval nozzle.Under this condition a sub-sonic —ow is produced whose temperature is ambient i.e. 300 K. The ratio of the SF and anthracene negative ion densities was again measured as a function of anthracene density. 6 Table 3 lists the values of the SF rate coefficient measured by us using the CRESU and FALP apparatuses in our laboratory and it can be seen that the 300 K result is 6 close to that obtained by Petrovic and Crompton and that the attachment rate for this molecule varies little with decreasing temperature. On the second line of Table 3 the values for the attachment rate to anthracene determined from these measurements and using eqn. (8) are listed and it can be seen that the value obtained at 300 K is a factor of ca.2.7 less than that previously measured by us using our FALP apparatus. This latter measurement is however expected to be more accurate. Firstly this experiment did not suÜer from any condensation eÜects since the anthracene was injected directly into the wide —ow tube without passing through a constrictive nozzle throat. In addition to this however the mass spectrometer used in that apparatus had undergone a rigorous calibration procedure involving the study of several ion»molecule reactions and so any mass discrimination eÜects had been well characterised. For this reason this value has been taken by us to be our standard and on line 3 of Table 3 normalised values for the anthracene attachment coefficient are listed for the four temperatures studied.It can be seen that there is a fair amount of scatter among the results but that overall no systematic temperature variation has been found for this process and that the values lie between 1 and 3]10~9 cm3 s~1. This is in agreement with the results of Christophorou et al.25,26 who found no variation in the rate between 380 and 480 K and a value for the rate lying in this range. Discussion Electron attachment occurs via the capture of an electron into an excited negative ion state of the parent molecule that may subsequently decay via autodetachment relaxation (naturally or in collision with another atom or molecule) or if the excited state is 81 T . Moustefaoui et al. [E repulsive via dissociation. For many polyatomic molecules (such as SF for example) where such negative ion states are energetically accessible the capture process at low 6 energies is dominated at low energies by an s-wave threshold law.31h33 This law predicts an E~1@2 energy dependence for the cross section for capture of an electron by a neutral molecule corresponding to a temperature independence for the rate coefficient.[The e pPE same law is responsible for the e~1 (aPT e~1@2) dependence often seen for the related process of the dissociative recombination of electrons and positive molecular ions.34] Since we see no apparent temperature dependence over the range studied it would seem reasonable to assume that (a) anthracene has low energy negative ion states available into which the capture can occur and (b) that its electron attachment reaction is described by just such an s-wave law.For those molecules that do not display zero energy resonances in their electron capture cross sections such as —uoranthene perylene and tetracene,24 one would not expect to –nd such a temperature independence and their capture will be described by p-wave or higher order scattering. This is very important as it means that even though these species have large attachment cross sections for energies greater than 0 eV their low temperature rate coefficients will be rather small. This point is very important for as discussed earlier many of the models of the physical state of PAHs12h16 and of the implications of PAHs on interstellar chemistry18h20 have assumed that as the size of the molecules gets larger and as the electron affinity increases the attachment rate will increase to very large levels such as several times 10~7 cm3 s~1.This hypothesis has been questioned recently by Smith and Spanel.35 They have pointed out that even though a molecule (such as for example WF6) may have a very large electron affinity (3.5 eV in that case) its electron attachment rate may be small. Conversely even though SF has a very large rate its electron affinity is rather (SF6)\1.06 eV]. In their study concerning electron attachment to C moderate 6 ea (Eea\2.65 eV) and C70 (Eea\2.68 eV) fullerenes they have found that while these 60 molecules are very efficient electron scavengers (via p-wave capture) at temperatures in excess of 2000 K their attachment rates are small at low temperatures since they do not have available negative ion s-states that can participate in the s-wave capture.Given that anthracene has a low temperature attachment rate of between 1 and 3]10~9 cm3 s~1 the question arises what eÜect this will have on the physical state of PAHs in interstellar clouds ? Certainly to obtain a clear answer to this will require detailed modelling such as that discussed in the introduction to this paper. If we assume that PAHs having such a rate account for say 10% of all PAHs so that they would have a relative abundance of perhaps 10~8 compared to hydrogen and if we take the case of a diÜuse interstellar cloud with a hydrogen density of 100 cm~3 and an electron density of perhaps 10~6 [nH] we can calculate the timescale for attachment to occur to these species using the expression (9) q~1\battach]ne This yields a value for q of 1013 s or 3]105 yr which is short compared to the typical lifetime of an interstellar cloud ([107 years) and so one would expect that a considerable fraction of PAHs would exist in anion form.At the same time we must recognise that this presupposes that the negative ions formed by electron attachment will stabilise radiatively so as not to undergo autodetachment and that subsequent photodetachment and collisional detachment processes do not destroy them. These are also topics that require experimental veri–cation. Given the preceding comments one must ask if we can suggest a candidate PAH that might display a considerably larger rate coefficient.In fact there is a good candidate and that is 1,2-benzanthracene C22H14 which though it has an electron affinity similar to anthracene has been found by Christophorou and co-workers25,26 to have an attachment rate about –ve times higher. It will also be interesting to examine substituted PAHs. Christophorou et al.26 have found that naphthoquinone C10H6O2 displays an 82 L ow temperature electron attachment to polycyclic aromatic hydrocarbons attachment rate 30 times that of anthracene. Compounds such as these will be the subjects of a subsequent study using the CRESU technique. In order to overcome the condensation problems that have hampered this particular measurement electrically heated nozzles will be fabricated and characterised and these will be used in the subsequent experiments.The authors would like to thank the Programmes Nationaux Franais de Planetologie et de Chimie du Milieu Interstellaire for their –nancial support and Dr Andre Canosa for helpful discussions. Paper 8/00490K; Received 19th January 1998 References 1 G. H. Herbig Annu. Rev. Astron. Astrophys. 1995 33 19. 2 A. Leger and L. dœHendecourt Astron. Astrophys. 1985 146 81. 3 M. K. Crawford A. G. G. M. Tielens and L. A. Allamandola Astrophys. J. 1985 293 L45. 4 G. P. Van der Zwet and L. A. Allamandola Astron. Astrophys. 1985 146 76. 5 W. W. Duley and D. A. Williams Mon. Not. R. Astron. Soc. 1981 196 269. 6 A. Leger and J. L. Puget Astron. Astrophys. 1984 137 L5. 7 L. J. Allamandola A. G. G. M. Tielens and J.R. Barker Astrophys. J. 1985 290 L25. 8 W. W. Duley and A. P. Jones Astrophys. J. 1990 351 L49. 9 M. Fingas F. Ackerman P. Lambert K. Li Z. Wang R. Nelson M. Goldthorp J. Mullin L. Hannon D. Wang A. Steenkammer S. Schuetz R. Turpin P. Campagna L. Graham and R. Hiltabrand Proceedings 19th Arctic and Marine Oilspill Prog. T ech. Sem. Environment Canada Ottawa Canada 1996 p. 907. 10 M. Frenklach and E. D. Feigelson Astrophys. J. 1989 341 372. 11 H. W. Jochims E. Ruhl H. Baumgartel S. Tobita and S. Leach Astrophys. J. 1994 420 307. 12 A. Omont Astron. Astrophys. 1986 164 159. 13 B. T. Draine and B. Sutin Astrophys. J. 1987 320 803. 14 E. L. O. Bakes and A. G. G. M. Tielens Astrophys. J. 1994 427 822. 15 C. Joblin L. dœHendecourt A. Leger and D. Defourneau Astron.Astrophys. 1994 281 923. 16 F. Salama E. L. O. Bakes L. J. Allamandola and A. G. G. M. Tielens Astrophys. J. 1996 458 621. 17 H. Abouelaziz J. C. Gomet D. Pasquerault B. R. Rowe and J. B. A. Mitchell J. Chem. Phys. 1993 99 237. 18 S. Lepp and A. Dalgarno Astrophys. J. 1988 324 553. 19 S. Lepp A. Dalgarno E. F. van Dishoeck and J. H. Black Astrophys. J. 1988 329 418. 20 G. Pineau de Forets D. R. Flower and A. Dalgarno Mon. Not. R. Astron. Soc. 1988 235 621. 21 A. G. G. M. Tielens in Dust and Chemistry in Astronomy ed. T.J. Millar and D.A. Williams IOP Publishing Bristol 1993 p. 99. 22 L. J. Allamandola A. G. G. M. Tielens and J. R. Barker Astrophys. J. Suppl. Ser. 1985 71 733. 23 A. A. Christodoulides D. L. McCorkle and L. G. Christophorou in Electron»Molecule Interactions and their Applications V ol II ed. L.G. Christophorou Academic Press Orlando 1984 p. 424. 24 S. Tobita M. Meinke E. Illenberger L. Christophorou H. Baumgartel and S. Leach Chem. Phys. 1992 161 510. 25 L. G. Christophorou and R. P. Blaunstein Radiat. Res. 1969 37 229. 26 L. G. Christophorou D .L. McCorkle and J .G. Carter J. Chem. Phys. 1971 54 253. 27 G. R. A. Johnsen and M. C. Sauer Jr. J. Chem. Phys. 1969 51 496. 28 A. Canosa D. C. Parent D. Pasquerault J. C. Gomet S. Laubeç and B. R. Rowe Chem. Phys. L ett. 1994 228 26. 29 J. L. le Garrec O. Sidko J. L. QueÜelec C. Rebrion-Rowe J. B. A. Mitchell and B. R. Rowe J. Chem. Phys. 1997 107 54. 30 Z. Li. Petrovic and R. N. Crompton J. Phys. B At. Mol. Opt. Phys. 1985 17 2777. 31 E. Wigner Phys. Rev. 1948 73 1002. 32 A. Chutjian A. Garscadden and J. M. Wadehra Phys. Rep. 1996 264 393. 33 J. P. Gauyacq and A. Herzenberg J. Phys. B At. Mol. Opt. Phys. 1984 17 1155. 34 J. B. A. Mitchell Phys. Rep. 1990 186 215. 35 D. Smith and P. Spanel J. Phys. B At. Mol. Opt. Phys. 1996 29 5199.

ISSN:1359-6640    

DOI:10.1039/a800490k

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

Faraday Discuss. 1998 109 83»108 General Discussion towards the BN/KL star forming core in Orion1 have recently 2O Dr Bergin opened the discussion of Prof. Millarœs and Prof. van Dishoeckœs papers ISO observations of H con–rmed the astrochemical prediction that water will be produced in abundance ([H O]/[H2][10~4) in shocked molecular gas. Since the cooling timescale after the shock passage is rather short (on the order of 100 years) an important question is what 2 happens to this large water abundance once the gas cools to typical values of TkB of cold dust grains. Chemical models have shown that this depletion will occur prior to 10»40 K. One likely possibility is that the H2O molecules will deplete onto the surfaces the re-assertion of the normal quiescent time-dependent chemistry.2 Thus this oÜers an alternate method to create the abundant water ice mantles that are observed in molecular clouds.This method is quite robust and is able to reproduce the [HDO]/[H2O] ratios observed both in hot cores and in cometary ices. Furthermore due to an imbalance created in the chemistry by the high water abundance there is a signi–cant increase in the abundance of OH in the post-shock gas which allows for the production of CO in the gas phase. The carbon dioxide molecules will deplete onto the grain mantles with2 an abundance that is 1»30% of H2O ice again an amount that is quite similar to observed values. These results do not suggest that surface chemistry does not occur but rather argue that care must be applied in the interpretation that ISM ices are the sole result of grain surface chemistry ; instead they may be created in shocks.1 Harwitt et al. Astrophys. J. L ett. 1998. 2 Bergin Melnick and Neufeld Astrophys. J. L ett. 1998 499 777. Prof. Millar responded Shocks may not play a dominant role in determining the composition of molecular ices because the degree of deuterium in hot cores appears to be relatively large e.g. for HDO and perhaps more importantly for D CO. However we need a much more detailed study of D-bearing molecules in hot cores to be certain of 2 this. Dr Hatchell said The shock tracer SiO is detected in single dish observations of many hot core regions. High resolution VLA§ observations of one such hot core G34.26 show that the SiO traces a possible out—ow and a widespread extended component but that no SiO is seen in the direction of the ammonia core (see Fig.1). Prof. Irvine said Last year my student Jimmy Dickens and I and colleagues in Sweden and Japan reported the identi–cation of a new ìhot coreœ molecule ethylene oxide in the Galactic center cloud Sgr B2 (N).1 This 3-membered ring c-C2H4O is only the third cyclic molecule discovered in interstellar clouds. However the richness of the Sgr B2 spectra and the width of the lines left some lingering doubt about the assignment. Quite recently additional observations at SEST have detected ethylene oxide in several other hot core sources thus absolutely con–rming the presence of this molecule in such regions. Fig. 2 shows seven transitions of ethylene oxide labeled with the relevant energy levels observed in the source NGC 6334F by Albert Nummelin from the Onsala Space Observatory and Jimmy Dickens from the University of Massachusetts.2 In addition several other large oxygen-containing organic molecules were observed in several transitions including acetaldehyde (CH CHO) a lower energy isomer of ethylene oxide (Fig.3 § The VLA is part of the National Radio Astronomy Observatory a facility of the National Science Foundation operated under co-operative agreement by Associated Universities Inc. 83 84 General Discussion 3 Fig. 1 SiO absorption (» » ») and emission (greyscale) towards the G34.26 hot core. The hot core itself is detected in NH (»»). 3). It is interesting that the ratio of to c-C2H4O CHCHO is quite well determined and 3 is rather constant in the sources observed to date (in the range CH CHO/c-C2H4OB 2»8).Hopefully these data will stimulate the modelers to include ethylene oxide in their 3 calculations and to see what constraints these observed values put on the models. 1 J. E. Dickens W. M. Irvine M. Ohishi M. Ikeda S. Ishikawa A. Nummelin and Aé . Hjalmarson Astrophys. J. 1997 489 753. 2 A. Nummelin J. E. Dickens P. Bergman Aé . Hjalmarson W. M. Irvine M. Ikeda and M. Ohishi Astron. Astrophys. 1998 in press. Dr Schutte communicated Some of the more complex molecules seen in hot cores such as CH3CH2OH (CH3)2O and could –rst be produced by grain surface chemistry and later released into the gas phase. In addition to surface chemistry complex species could be produced by UV irradiation of the ice mantles.As an example it was found that methyl formate (HCOOCH3) one of the products seen in hot cores which is not accounted for by the surface chemistry model is produced by irradiation of methanol ice.1 The production of ethanol was reported for UV irradiated H2O»CH3OH mixtures. 2 In this context it must be noted that methanol-rich ice is often observed in the direction of protostars.3,4 Thus photochemistry should be considered as a mechanism for the production of some of the more complex species seen in hot cores. 1 P. A. Gerakines W. A. Schutte P. Ehrenfreund and E. F. van Dishoeck 1996 Astron. Astrophys. 1996 312 289. 2 M. P. Bernstein S. A. Sandford L. J. Allamandola S. Chang and M.A. Scharberg Astrophys. J. 1995 454 327. 3 C. J. Skinner A. G. G. M. Tielens M. J. Barlow and K. Justanont Astrophys. J. 1992 399 L79. 4 M. E. Palumbo A. G. G. M. Tielens and A. T. Tokunaga Astrophys. J. 1995 449 674; M. E. Palumbo T. R. Geballe and A. G. G. M. Tielens Astrophys. J. 1997 479 839. Dr Palumbo asked In your paper you discuss that S-bearing molecules can be used to derive the age of HMCs. In this picture H2S is formed in the mantles and then evaporates. As far as I know H2S has never been detected in the solid phase. Why do you start with H2S? Why do you not consider OCS which has been identi–ed in the solid phase? General Discussion Fig. 2 Spectra towards NGC6334F with all seven detected c-C2H4O transitions indicated 85 86 General Discussion Fig.3 Three rotational transitions of ethylene oxide (c-C2H4O) and four of acetaldehyde (CH CHO) observed toward the hot core NGC 6334 (F) with the SEST telescope by A. Nummel in (Onsala Space Observatory) and J. Dickens (University of Massachusetts). Both these isomers 3 have very recently been found to be commonly present in such regions. sight but it is clear that it is not a major component of interstellar ices. In our models of Prof. Millar replied I believe that there is a claim for solid H2S along one line of hot molecular cores we –nd that our observations of many sulfur-bearing molecules are best –t by assuming an initial H abundance of ca. 10~7 that is about 100 times less than the cosmic abundances of sulfur. 2S 2 Prof.Black said The Introductory Lecture by Prof. Williams has called attention to a ì sulfur problemœ ; viz. our current inability to explain simultaneously the high abundance of atomic sulfur observed in diÜuse molecular clouds the chemistry of simple sulfur-bearing molecules (CS SO SO2 H2S and SO`) in dense clouds and the absence of large quantities of sulfur-containing ices (e.g. OCS and H S). Prof. Millarœs paper cites a possible role of H S-ice in the chemical evolution of interstellar gas. Do the recent ISO data on ices (Prof. van Dishoeckœs paper) provide an 2 improved limit on the abundance of solid H2S? A reason for raising this question is that Prof. Millarœs paper discusses the evaporation of solid H2S in regions where H-atom reactions H2S]H]SH]H2 SH]H]S]H2 87 General Discussion subsequently introduce S which in turn leads to formation of the widely observed CS SO and SO2 .If this scheme is important shouldnœt the SH/H2S abundance ratio approach unity ? Although SH has K-doubling and rotational transitions at unfavourable frequencies for ground-based astronomy searches for it might aÜord useful tests of this scheme. It is worth noting that the gaseous sulfur abundance may play a key role in controlling the electron abundance in dense clouds since atomic S is rather easily ionized and sulfur may not appear in molecular form to the extent that carbon does. Dr Schutte responded Several questions have been asked here but I will only address the solid phase abundance of H2S. The strongest feature of H2S is the combined l most sensitive search for this feature can be obtained with ground-based telescopes.No 3/l1 mode around 3.85 lm. Since this spectral region is accessible from the ground the feature that can be ascribed to solid H2S has yet been found in this spectral region. An earlier assignment of the weak absorption feature at 3.91 lm seen towards the high-mass convincingly showed that the band can be accounted for by solid methanol. So as of yet young stellar object W33A to solid H2S (ref. 1) was revised by Allamandola et al.,2 who scopic data of H no evidence for H2S in dense cloud ices is available. Unfortunately quantitative spectro- 2S in astrophysically relevant ice matrices is not yet available so no upper limit can be quoted here.1 T. R. Geballe F. Baas J. M. Greenberg and W. Schutte Astron. Astrophys. 1985 L6 146. 2 L. J. Allamandola S. A. Sandford A. G. G. M. Tielens and J. M. Herbst Astrophys. J. 1992 134 399. Prof. van Dishoeck added The ISO spectra do indeed provide the –rst opportunity to put quantitative limits on any sulfur-bearing ices because of the complete wavelength coverage (e.g. Whittet et al.,1 and dœHendecourt et al.2). Apart from the 6.8 lm feature discussed by Schutte et al.,3 the ISO spectra do not show any strong unidenti–ed bands. Thus if the sulfur is primarily in the solid phase in cold dense clouds it must be in a compound which does not have any strong IR bands. This conclusion was also reached by Palumbo et al.4 in their study of solid OCS. Detection and analysis of any weaker bands in the ISO spectra awaits improved data reduction software.1 D. C. B. Whittet W. A. Schutte A. G. G. M. Tielens et al. Astron. Astrophys. 1996 315 L357. 2 L. dœHendecourt M. Jourdain de Muizon E. Dartois et al. Astron. Astrophys. 1996 315 L365. 3 W. A. Schutte A. G. G. M. Tielens D. C. B. Whittet et al. Astron. Astrophys. 1996 315 L365. 4 M. E. Palumbo T. R. Geballe and A. G. G. M. Tielens Astrophys. J. 1997 479 839. Dr Hatchell commented I would like to show two –gures to illustrate the discussion on sulfur chemistry Fig. 4 shows a plot of the abundances in a number of hot core sources of several sulfur-bearing molecules; Fig. 5 is an illustration of the chemical evolution of sulfur species under typical hot core conditions.Prof. Thaddeus said The abundance ratio of most sulfur molecules in molecular clouds to their oxygen analogues is about that of the cosmic S O ratio 1 42. The most notable exception is CS which is very de–cient relative to CO. Taking into account limits which have been obtained for solid H2S how de–cient do you think H2S is relative to H2O? Prof. Millar responded In cold clouds the ratio of H2S H2O to 2O is hard to determine because the abundance of H2O H is difficult to estimate. However is likely to be reasonably abundant at least 1% of the total oxygen not locked up in CO and grains. H Upper limits to the H2S fractional abundance are very low and imply a ratio of on the order of 10 000»100 000. 2O :H2S 88 General Discussion Fig.4 Abundances of sulfur-bearing molecules in a number of hot cores associated with ultracompact HII regions is detectable but still appears to be a factor of at least 100 less In warm clouds H2S abundant than H2O. Dr Palumbo asked In your paper you discuss the formation of ice mantles during the collapse phase of a cloud. We know that ice mantles exist in quiescent molecular clouds. Reading other papers I understood that ice mantles already exist at the time the collapse starts. Why do you not consider pre-existing ice mantles? Alternatively is the scienti–c community changing its view on this subject ? Prof. Millar responded My belief is that the ice mantles in both cold dense clouds and in (pre-) hot molecular cores are very similar. Unfortunately IR observations are not sensitive to material with fractional abundances O10~6»10~7 relative to hydrogen.In cold clouds material is not returned to the gas whereas in hot cores we can see via gas-phase spectroscopy the ice mantles when they are returned to the gas phase. Fig. 5 Model of time evolution of sulfur species abundances for typical hot core conditions Tkin\100 K; nH2\2]105 cm~3 and cosmic ray rate Z\1.2]10~16 s~1 89 General Discussion Dr Ehrenfreund said ISO allowed a new de–nition of the composition of interstellar ices thermal processing in the protostellar vicinity and gas»grain chemistry. An important result of ISO was the ubiquitous presence of abundant CO ice in space. From the triple-peaked bending mode of CO ice we could derive for the –rst time that extensive 2 2 ice segregation is occurring in the environment of protostars.H2O CO2 and CH3OH ice seem to exist towards many protostars in temperature zones of B60 K in fragile ice layers where these molecules are loosely bound to each other. This has important consequences for their desorption in the gas phase as well as for outgassing of cometary ices. Prof. Chakarov said Photochemical reactions on the grain surfaces play an important role in the material evolution in space. Unfortunately the potential of photoinduced processes on surfaces is not yet fully understood and applied in the models attempting to reproduce the observed abundances. In the case of graphite (carbon) grains surface photoprocesses are especially efficient (much more than in the gas phase) due to the strong light absorption by the substrate and its relatively weak quenching of the excitations.Thus the surface photochemistry is driven by an efficient substrate mediated mechanism based on the creation transport and resonant scattering of hot charge carriers created in the bulk. Here (Fig. 6) we show experimental evidence for the photoproduction of a number of new products as well as photodesorption of the primary species during photon irradiation of water (ice) on potassium precovered graphite surface. These observations adequately support the model predictions from Prof. van Dishoeckœs paper. We found that UV irradiation (j\400 nm) of H2O and K coadsorbed on graphite (0001) at 85 K leads to the photodesorption of H H2 CH4 CO K and CO and the formation oxygen-rich KwOwC complexes on the surface.2 The primary step of these photoreactions involves selective photodissociation of H2O generating hot H atoms (signi–cantly more energetic than in thermal equilibrium) and OH in ground and excited vibrational states. Subsequent collisions with coadsorbed H2O K and with substrate (C) give rise to the observed photoproducts. It is worth noting that these reactions are possible in the dark only at temperatures above 1100 K. Dr Ehrenfreund communicated The results of photon-induced crystallization of thin ice –lms are very interesting and challenging and could have important consequences for gas»grain chemistry in the interstellar medium. It would be important to quantify how Fig.6 Direct observations of the products of photoinduced surface reactions in H2O»K hH2O\0.75 monolayers; hK\0.13 monolayers. coadsorbed layer on graphite surface at 85 K. Photon energy 4.3 eV (289 nm); —ux 1.6]1016 photons cm~2 s~1. 90 General Discussion much H2O ice can be converted into the crystalline form and to combine those experiments in the future with spectroscopic studies. In astronomical spectra our only possibility of proving the crystallization of water ice is to monitor the change in the IR spectrum. Prof. Chakarov added Astronomers believe that water ice forming the mantles covering interstellar dust particles is mainly in amorphous form. The belief is based not so much on spectroscopic evidence as on the well established fact that amorphous» crystalline [Ia ]Ic Ic stands for ice (cubic)] phase transition takes place at temperatures around 140 K in vacuo.This may turn out to be wrong taking into account the recent discovery of photon-induced crystallization of thin ice –lms. A new photoinduced surface process is observed under UV irradiation of thin ice –lms on graphite surface transition from amorphous to crystalline phase. The eÜect is con–rmed on the graphite/ice interface and –rst 3»4 monolayers. It has clear coverage and wavelength dependence with a threshold at ca. 4 eV. The mechanism we propose for the excitation of lattice vibrational modes includes both energy dissipation by the transient anion state and polarization eÜects. The observed eÜect has (at least) two important consequences (i) the sublimation rate of the crystalline ice is almost twice as low as those for amorphous ice (see Fig.7) thus directly aÜecting the H2O abundance; (ii) the surface structure (the termination) of the ice layer is diÜerent for the two phases thus dramatically changing the ability of water molecules to participate in reactions/transformations. The crystallization process can be called ìaccumulativeœ since ice grown (condensed) on already crystalline ice surfaces continues to grow as crystalline. Prof. van Dishoeck responded I do not agree with your statement that there is not much spectroscopic evidence for the fact that the water ice is in amorphous form. This has been demonstrated extensively through laboratory experiments starting with the work of Hagen et al.1 Crystalline and amorphous ice can clearly be distinguished at the spectral resolving power and signal to noise ratio of the observed IR spectra.1 W. Hagen J. M. Greenberg and A. G. G. M. Tielens Astron. Astrophys. Suppl. 1983 51 389. Fig. 7 Mass spectrometer (m/z\18) trace of the isothermal sublimation of 2.2 monolayers of water ice deposited on graphite (0001) surface at 89 K. The sublimation temperature of 134 K is approached by a linear T -ramp with b\0.1 K s~1. L trace recorded directly after the deposition ; Ö UV light irradiated layer (dose 5.5]1018 photons cm~2). The inset shows changes of the sublimation rate that are speci–c for the transition from amorphous to the crystalline state. General Discussion 91 Ms Trakhtenberg said We have investigated the adsorption of water ice on substrates of various roughness and chemical composition at 100 K.Amorphous quartz amorphous quartz covered with octadecyltrichlorosilan monolayer (OTS) amorphous quartz covered with mixed octadecyltrichlorosilan-docosyltrichlorosilan monolayer (OTS»DTS) single crystal silicon (100) wafer silicon wafer covered with OTS and silicon wafer covered with OTS»DTS were studied. It was found that on smooth substrates partially crystalline ice is formed however on rough substrates ice is amorphous. Hence the ice structure can provide information about the morphology of substrate on which ice is adsorbed. After the adsorption the samples were heated and the desorption of ice was observed. It was found that desorption of ice from smooth surfaces occurs at lower temperatures than from rough surfaces.In Fig. 8 the correlation between the ice desorption temperature and the substrate roughness is presented. These results can be rationalized in the following way. An ice cluster adsorbed on a rough substrate interacts with a larger number of sites than the ice cluster adsorbed on a smooth substrate. Therefore the interaction between the ice and the rough substrate is stronger than that between the ice and the smooth substrate. The ice clusters adsorbed on a smooth substrate are more mobile and they can rearrange themselves to a more stable crystalline state. Ice clusters adsorbed on rough substrates are trapped in the potential wells and therefore desorbed from the substrate at higher temperatures.Prof. van Dishoeck responded Thank you for presenting your interesting experiments on the dependence of the adsorbed ice phase on the roughness of the substrate. In the interstellar medium most of the silicate grains are amorphous and therefore have rough surfaces. The observed ices are indeed predominantly amorphous (e.g. Hagen et al.1). 1 W. Hagen J. M. Greenberg and A. G. G. M. Tielens Astron. Astrophys. Suppl. 1983 51 389. Fig. 8 Correlation between the substrate roughness ice phase and ice desorption temperature. Quartz slides and silicon wafers either bare or covered by self-assembled monolayers of octadecyltrichlorosilane pure (OTS) or mixed with decosyltrichlorosilane (OTS-DTS) were used as substrates. K = Deposited ice was partially crystalline ;,deposited ice was amorphous.92 General Discussion Dr McCoustra communicated The water ice temperature programmed desorption (TPD) data presented by Ms Trakhtenberg and co-workers contrasts markedly with measurements comparing the behaviour of amorphous and crystalline water ices that we have previously reported.1 Therein we presented evidence that clearly showed that a crystalline water ice –lm desorbs at a temperature typically 10 to 15 K above that of an amorphous –lm. Perhaps Ms Trakhtenberg would care to comment on this diÜerence. 1 P. Jenniskens S. F. Banham D. F. Blake and M. R. S. McCoustra J. Chem. Phys. 1997 107 1232. Ms Trakhtenberg responded The fact that the desorption rate of amorphous ice is higher than that of crystalline ice is well known.1 The reason is the excess free energy of the metastable amorphous ice phase vs.the crystalline ice phase. The hydrogen bonding within the ice crystal is known to be stronger than within the amorphous ice. However in this consideration one assumes that the interaction between the ice and the substrate is the same for both phases. Hence it is reasonable that in the study undertaken by Jenniskens et al.2 it was found that the desorption temperature of freshly deposited ice is lower than that of annealed ice due to the larger fraction of crystalline component in the latter. In both cases the same substrate has been used. On the contrary in our work3 we compared the desorption of ice which were deposited on substrates of various roughness.The substrate eÜect was enhanced by using thin ice –lms (5»10 monolayers thick vs. 70 monolayers thick –lms used in the study by Jenniskens et al.2). We found that the ice deposited on rough substrates was amorphous while ice deposited on smoother substrates was partially crystalline. From smooth surfaces ice was found to desorb at lower temperature than from rough surfaces. However it is important to realise that the desorption occurs after the onset of ice crystallisation. Therefore the ice which was deposited as amorphous became partially crystalline prior to desorption. It means that in both cases the desorption of partially crystalline ice was observed. We suggest that the desorption which we observed is the desorption of ice clusters.Therefore the main factor in this process is the interaction between the ice cluster and the substrate and not the interactions between the water molecules. The changes in desorption temperature of ice result from its interaction with the substrate and not from diÜerences in the ice phase during desorption. 1 R. J. Speedy P. G. Debenedetti R. S. Smith C. Huang and B. D. Kay J. Chem. Phys. 1996 105 240. 2 P. Jenniskens S. F. Banham D. F. Blake and M. R. S. McCoustra J. Chem. Phys. 1997 107 1232. 3 S. Trakhtenberg R. Naaman S. R. Cohen and I. Benjamin J. Phys. Chem. B 1997 101 5172. 3 2 H2CO 2H6 carbon monoxide CO formaldehyde ethane C is very sur- Dr Kaiser said From the chemical viewpoint the postulation of grain»surface diÜusion processes in the formation of methanol CH OH ethanol C2H5OH carbon dioxide CO prising and until now without scienti–c proof.Based on laboratory studies on the interaction of MeV cosmic ray particles such as protons (H`) and helium nuclei (He2`) and computer calculations of these processes there are strong alternative pathways to these grain surface processes. Inside dark clouds the ice mantle condensed on interstellar grains at about 10 K is processed by particles of the galactic cosmic radiation –eld leading to new molecules synthesized in the solid state. Since however typical carbon» hydrogen and carbon»carbon bond strengths in organic molecules range between 3 and 10 eV the cosmic ray particles are too energetic to form a stable chemical bond as implanted into the icy mantle.But upon interaction with the solid target each cosmic ray particle releases its excess energy to the target atoms in successive collisions via elastic and inelastic interactions. Here the elastic process transfers energy to the nuclei of the target atoms igniting primary knock-on particles (PKOs; 1st generation of knockon particles) if this amount is larger than the binding energy of the atom. MeV aparticles for example generate carbon PKOs with kinetic energies up to 10 keV. These knock-on particles can transfer their energy in consecutive encounters to the target 93 General Discussion Fig. 9 Derived pathways to form methanol (1) dimethylether (2) ethanol (3) carbon monoxide (4) formaldehyde (5) and carbon dioxide (6) and atoms resulting in a collision cascade of secondary tertiary etc.knock-on atoms. Moderated to about 1»10 eV (the so-called chemical energy range) these atoms are not in thermal equilibrium with the 10 K target (hence non-equilibrium or suprathermal particles) and can react –nally with the target molecules via elementary steps of bond insertion addition to double or triple bonds or hydrogen abstraction. The power of these suprathermal reactions to form new molecules at temperatures even as low as 10 K is based on their ability to overcome reaction barriers in the entrance channel easily since suprathermal species can impact their excess kinetic energy into the transition state of the reaction. Even reactions endothermic at 10 K are feasible and extend the synthetic power of this reaction class further beyond thermal processes of diÜusion controlled chemistry on interstellar grains.These unique aspects of suprathermal reactions result in reaction rate constants k orders of magnitude larger than their thermal counterparts. Detailed calculations on the reactions of e.g. 1 eV suprathermal hydrogen atoms with H2O C 4 H to form H and OH as well as CH 2 H2O)\1.7]10~11 3 4)\2.5]10~19 respectively depict k (1 eV H cm3 s~1 and k (1 eV H CH4)\5.0]10~11 cm3 s~1 vs. thermal H]H2O)\9.6]10~27 rate constants of k (293 K cm3 s~1 and k (293 K H ]CH cm3 s~1 a diÜerence of up to 16 orders of magnitude. An order of magnitude calculation shows that although the internal UV —ux in dense molecular clouds (/\103 cm~2 s~1) ranges two orders of magnitude higher than the cosmic ray particle —ux distribution maximum of /\101 cm~2 s~1 for 1»10 MeV protons and a-particles each MeV particle can generate about 100 suprathermal species in a 0.2 lm thick icy layer and 0.3 lm thick grain core.The —ux advantage of the internal UV –eld deep inside molecular clouds is clearly eliminated by the ability of a single MeV particle to induce collision cascades with up to 100 suprathermal species. Therefore an opening from ìphoton-only universeœ to process interstellar grains to include MeV ion synthesis is strongly encouraged. Fig. 9 summarizes derived pathways to form methanol (1) dimethylether (2) ethanol (3) carbon monoxide (4) formaldehyde (5) and carbon dioxide (6) in ices condensed on interstellar grains.The spin states of the carbon (C) and oxygen (O) atoms are not indicated. Square brackets indicate an internally/electronically excited intermediate which can transfer its excess energy to the surrounding matrix. Dr Schutte responded I agree that the matter of the mechanism for the formation of relatively complex species like formaldehyde methanol ethanol carbon dioxide etc. on 94 General Discussion interstellar grains is far from settled. Indeed though widely postulated in chemical models of gas»grain chemistry experimental proof (or disproof) for the required surface chemical reactions is still lacking. Therefore energetic processing i.e. by energetic ions or UV is often mentioned as an alternative. Such processing would involve very simple starting mixtures of H2O CO and possibly NH3 CH4 O2 and N2 .These species can either accrete directly from the gas phase (CO O2 N2) or be produced by a very basic surface chemistry network (H2O CH4 NH3). While experiments have shown that irradiation of such ices either by UV or energetic ions efficiently produces CO formaldehyde no indication of the formation of signi–cant quantities of 2CH and some 3OH or ethanol has yet been found.1h3 It must be noted however that the energy of the ions in these experiments (B1 MeV) was considerably lower than expected for cosmic rays. Nevertheless the experimental evidence at this point does not support the theoretical prediction that the formation of methanol and ethanol through energetic processing is feasible although carbon dioxide may be produced in this way.The matter of the relative importance of UV and ion processing is still open. Opposite opinions in this respect have been expressed with unfortunately so far little substantial evidence in the form of detailed published calculations from either side. What is necessary to address this matter is a self-consistent calculation which compares the energy deposited in dense cloud ices by an appropriate energy distribution of charged particles with the energy deposited by the UV radiation produced by the excitation of H by these charged particles as was described by Prasad and Tarafdar.4 2 1 W. Hagen 1982 PhD Thesis University of Leiden The Netherlands. 2 W. A. Schutte P. A. Gerakines T. R. Geballe E. F. van Dishoeck and J.M. Greenberg Astron. Astrophys. 1996 309 633. 3 M. H. Moore R. Uhanna and B. Donn J. Geophys. Res. 1991 966 17541. 4 S. S. Prasad S. P. Tarafdar Astrophys. J. 1983 267 603. Dr Kaiser commented I strongly recommend a report by Nebeling.1 Nebeling found that atomic carbon knock particles react with solid H2O and the following products are formed CO CH4 CO2 CH2O HCOOH and up to 60% CH OH. So there is experimental evidence that cosmic ray particle induced knock on carbon atoms could produce 3 methanol in solid H2O C in the icy mantles on interstellar grains. Regarding the synthesis this is only a theoretical but logical prediction based on collision cascade 2H5OH simulations (TRIM and MARLOWE code) and experiments in Juelich.1 1 B. Nebeling Forschungszentrum Juelich 1998 2245.Prof. Field commented One suggestion has been made that fast neutrals striking grains may in—uence the chemistry associated with grains. I should like just to mention that there is indeed an important astrophysical context in which fast neutrals may strike grains with many eV of energy. In magnetohydrodynamic shocks (MHD) there exists an extensive region in which there is diÜerential motion of the charged grains and the neutrals. This motion typically is a large factor of the shock velocity. In this regime therefore fast neutrals do indeed strike grains. One may however doubt that the grains maintain their molecular mantles in this environment. Prof. Smith said I wonder if Prof. Millar could clarify the kinetic information given in his paper for H]DCN]HCN]D.An activation energy of EA/kB200 K is given but for the reaction as written the endothermicity based on zero-point energy diÜerences corresponds to ca. 800 K. In the light of this is this reaction still likely to be important and would measurements of its rate coefficient at low temperature be valuable ? Prof. Herbst commented Talbi and Herbst1 have shown that the reaction H]DCN]HCN]D has large activation energy barriers in both the entrance and 95 General Discussion exit channels. This reaction which is endothermic by 800»850 K cannot occur efficiently in hot molecular cores unless the barriers do not exist. 1 D. Talbi and E. Herbst Astron. Astrophys. submitted. EA/kT . As long as this factor is smaller than about 5 Prof.Millar responded The factor of importance for determining whether atomic H can alter the DCN/HCH ratio is then the reaction can proceed at a fast enough rate to alter the DCN/HCN ratio on a time-scale less than those estimated for the ages of hot molecular cores. Thus for an endothermicity of ca. 800 K kinetic temperatures of greater than ca. 160 K are needed not a drastic constraint in terms of the temperatures of hot cores. However if the reaction posseses an energy barrier of the size discussed by Talbi and Herbst then this route will be unimportant in destroying DCN in hot cores. Prof. Zare said What kind of dust should we consider in gas»dust chemistry ? Can we de–ne a ìstandard dustœ or a series of standard dusts ? Prof. Papoular responded It is clear that a deep mental gap separates the large ìmolecularœ community from the small ìdustœ community.The former builds upon a solid foundation of well established facts and theories and can perform impressive computations none of which however applies to the solid state. The ìdustœ community on the other hand has not yet managed to bring to the forefront of the astrophysical literature the wealth of new experimental and theoretical results which are found in the general physical chemical and solid state literatures and may now pro–tably complement the seminal calculations of Salpeter.1 This conference is an opportunity to help bridge the gap. The most solidly established dust types are silicon carbide (SiC) and silicates crystalline and amorphous SiO (Mg Fe) and SiO (Mg Fe)2 .It does not seem that such 3 dust when bare enters into gas»solid reactions. Deep in molecular clouds however it 4 may be covered with condensed gases forming ices of H2O CO CO … … … . It is possible in turn that these ices interact chemically with the help of cosmic rays (see 2 CH4 remarks by Dr Kaiser at this meeting). The most reactive dust may turn out to be carbonaceous and hydrogenated like coal kerogen and hydrogenated amorphous carbon (HAC). A comparison of laboratory solid state carbonaceous models of cosmic dust can be found in Papoular et al.2 In view of the consensus that seems to have now been reached as to the functional groups associated with the UIBs (see ref. 3 and comments by Bernstein and Papoular at this meeting) a carbonaceous dust could be viewed macroscopically as —uÜy and porous.Microscopically it should include (in various proportions according to age and environment) conjugated aliphatic carbon chains and clusters of a few benzenic rings ; the rings would be more or less hydrogenated and substituted with aliphatic groups and heteroatoms (O N S . . .) and the clusters (graphenes) would be linked by oxygen bridges and more or less stacked together in turbostratic disorder. A schematic view of a model dust is presented by Papoular in a comment at this meeting. Such materials have been studied at length by coal and oil experts and their behaviour under heat and chemical treatments is now well documented. Banks of typical materials (ìstandardsœ) were established in the US and Europe.Clearly impinging atoms can react with attached H atoms and recombine into a molecule which is then easily expelled leaving behind a very reactive dangling bond. 1 E. E. Salpeter Astrophys. J. 1974 193 579; 585. 2 R. Papoular J. Conard O. Guillors I. Nenner C. Reynaud and J-N. Rouzard Astron. Astrophys. 1996 315 222. 3 C. Joblin Faraday Discuss. 1998 109 349. 96 General Discussion Dr Ali commented Professor Zare asked what the chemical composition and structure of interstellar dust is ? I think the question is very timely and most appropriate. We yet do not have any single de–nitive answer to this. Mass return from cool high luminous red giants and supergiants and more speci–cally from evolved thermally pulsing AGB stars is probably the single most important source of interstellar primary dust.Physical and chemical evolution of these stellar ejecta in the interstellar medium is far from being understood. The chemical composition and structure of interstellar primary grains will have a strong in—uence on the physics of gas»grain interaction in the interstellar medium. The problem of heterogeneous nucleation or particle growth in return addresses a question What does the surface of pertinent condensation nuclei look like ? At best we could expect that condensation at low temperature would result in a disordered molecular surface in dense molecular clouds. There is an uncertainty to a large extent in composition and structure of grains formed in stellar out—ows.M-type stars C/O\1 oxygen-rich element composition which may form some kind of iron»magnesium»silicate dust in expanding circumstellar shells around the stars. C-type stars C/O[1 carbon-rich systems which may form amorphous carbon grains perhaps the carbonaceous particles. We understand how gas phase small molecules are formed in interstellar clouds (see the papers by Lee and coworkers and Smith et al. presented at this meeting). These are certainly going to add another dimension to our knowledge of formation of small molecules in space. Yet there are no proper descriptions that take account of how dust grains are formed out of the small molecules in stellar out—ows. A theory guided by laboratory experiments on nucleation Cluster dynamics coupled with future observations at an increased resolution in spectroscopy and interferometric imaging with –ne spatial analysis of both scattered and emitted radiation from nucleation zone in circumstellar shells of evolved stars would be able to decide the size distributions chemical compositions and structures of grains formed.Inference on chemical and structural state of primary grains is crucial if we want to make further progress on chemical and physical evolution of interstellar dust. A. K. Speck said Many people ask what sort of grains should we be investigating. What are the grain surfaces where chemistry takes place ? No one seems to have caught onto the fact that we can get the answers to these questions from meteorites. There are samples of interstellar dust in the form of SiC (see ref.1) diamond and graphite2 and even aluminium oxide3 found in meteorites. From the isotopic studies it is clear that they are presolar. We can therefore study these grains in order to understand in detail which grains survive or are formed in the interstellar medium and what sort of grain surfaces are available. 1 P. Hoppe and U. Ott in Astrophysical Implications of the L aboratory Study of Presolar Materials ed. T. Bernatowicz and E. Zinner AIP Conference Proceedings American Institute of Physics New York 1997. 2 E. Anders and E. Zinner Meteoritics 1993 28 490. 3 L. Nittler C. M. OœD. Alexander X. Gao R. M. Walker and E. Zinner Astrophys J. 1997 483 475. Dr Ali responded The isotopic composition of a few elements in some meterotic minerals which is distinctly diÜerent from that in terrestrial samples of course bears the unambiguous evidence of heterogeneity of the solar nebula.However we do not have any clues as to what fraction of overall interstellar grain population constitutes such chemically distinct isotopically anomalous grains. The large degree of variations in cosmic abundances and physico-chemical conditions among various astrophysical objects lead to a question on the mechanism of formation of such grains (see the paper by Gail and Sedlmayr presented at this meeting) before passing a judgment based on isotopic anomalies in meteorites on the source medium where grains are formed. The presolar SiC has been astronomically observed and also graphite has been predicted in 97 General Discussion the interpretation of the hump at about 200 nm in the average interstellar extinction curve long before any meteoritic evidence of these grains.1 There are some major problems that concern most of us here in our understanding of the chemical and physical evolution of interstellar ìprimitiveœ dust and most notably what does the grain surface where chemistry occurs represent ? Immediate answers exist while it is hard to justify that graphites and diamonds are dominant as interstellar grains from such arguments based on nucleosynthetic isotopic anomalies alone.The 16O enrichments in refractory calcium»aluminum inclusions (CAI) in meteorites have been interpreted in terms of presolar grains with a separate history of nucleosynthesis for more than twenty years of research.2 An appeal was made that because chemical reactions could not produce such mass independent isotope eÜects the observed oxygen isotopic anomalies are purely correlated with nuclear astrophysical processes in the source medium.Now that it is known that the dynamics of nonequilibrium chemical reactions could indeed produce mass-independent isotopic anomalies strikingly similar to that observed in 16O-rich minerals in carbonaceous meteorites it is no longer possible to accept the notion that the 16O anomalies have always had a nuclear origin.3 In the light of this current understanding many of the 16O-rich thermocould be formed in the solar nebula by chemical processing and need not be considered dynamically stable refractory phases such as corundum (Al2O3) spinel (MgAl2O4) etc.as condensates in extrasolar system environments though it was originally proposed.4 Consider a speci–c reaction in solar nebular environment such as 18O]Si16O]O16xSixO18. The molecular symmetry induced stabilization of a nonsymmetric refractory trimer through short-lived transition state during a collision event at a certain energy would extract 17O and 18O from the gas-phase reservoir containing all three oxygen isotopic systems via preferential reaction of the SiO dimer to form the O16xSixO18 (or O17) trimer. The end result would be a gas phase enriched in 16O. Such an eÜect produces both positive and negative isotopic reservoirs and may account for the meteoritic O observations.Speci–c laboratory experiments would be required to demonstrate how symmetry could play a crucial role on the lifetimes of transition states during a chemical reaction event. The longer the transition states lifetimes the greater the probability of stabilization [see ref. 3(b)]. If a chemically produced non-mass dependent oxygen isotopic fractionation does indeed occur in the presolar nebula it establishes a direct link between observed refractory grains in meteorites with early solar system history. Oxide grains primarily corundum Al2O3 in primitive meteorites are recently reported to have a large excess of 17O (rather than 16O-rich) and a de–cit of 18O relative to solar system oxygen. From these unusual isotopic ratios and the presence of 26Al in grains it is conjectured that these grains are of circumstellar condensates from red giant and AGB stars.Many distinct nucleosynthetic sources with diÜerent masses and initial oxygen compositions were required to explain the range of isotopic diÜerences.5 However it is far from clear how the thermodynamically stable mineralogical solids such as corundum (Al2O3) spinel (MgAl2O4) etc. emerge from gas phase species as stellar condensates. There is not yet any de–nitive laboratory experiment on the formation of grains with variations in cosmic abundances and appropriate physicochemical conditions. No one has ever proposed an ab initio predictive theory of how dust formation occurs based on basic chemical processes in circumstellar shells of evolved stars. The study of isotopic anomalies and grain systematics in primitive meteorites is connected elsewhere both in testing models of nucleosynthesis and in the construction of models of origin and evolution of the solar system.However it is important to recognize that cosmic dust complex is primarily a many component system chemical details of which remain largely unknown. First dust in stellar out—ows grains are not primitive. The molecule/solid phase transition i.e. the formation of grains out of the gas phase 98 General Discussion Fig. 10 Schematic view of the UHV Paul-type trap used for the non-destructive mass determination of individual nm-sized grains. APD avalanche photo-diode PS particle source VP view U0\V cos(XHF t) applied potential (typically V \1»2 kV XHF/2p\0.2»2 port LD diode laser kHz) U1 ( gravitational compensation voltage U1\0.40 V).atomic and molecular species. Second dust in dense cool molecular clouds and in the diÜuse interstellar medium grain transformation and metamorphism i.e. the modi–- cation of existing primary dust by heterogeneous nucleation sputtering and recondensation coagulation etc. Third circumstellar dust around young stellar objects largely —uÜy and inhomogeneous aggregates. Structure and chemical composition of dust are expected to be diÜerent from those of dust in the diÜuse medium and the circumstellar shells around evolved stars. Dust grains play an important role for both the spectral appearance and the thermal dynamical and chemical evolution of astrophysical objects.Chemistry of interstellar dust is one of the exciting –elds of science and much remains to be learned from infrared observations speci–c laboratory experiments and theory on grain formation as well as evolution by interstellar surface processes. We still need answers to a difficult question relating to physics of interstellar grain surfaces.6 1 T. Stecher and B. Donn. Astrophys. J. 1965 142 1682. 2 R. N. Clayton L. Grossman and T. K. Mayeda Science 1973 182 485. 3 (a) M. H. Thiemens and J. E. Heidenreich III Science 1983 219 1073; (b) M. H. Thiemens Meteorites and the Early Solar System ed. J. F. Kerridge and M. S. Matthews University of Arizona Press Tuscson 4 D. D. Clayton in Cosmogonical Processes ed. W. D. Arnett C. Hansen J. Truran and S. Tsuruta VNU 5 I.D. Hutcheon G. R. Huss A. J. Fahey and G. J. Wasserburg Astrophys. J. 1994 425 L97; G. R. Huss AZ 1988 p. 899. Science Utrecht 1986 p. 101. A. J. Fahey R. Gallino and G. J. Wasserburg Astrophys. J. 1994 430 L81. 6 V. Buch and J. P. Devlin in Molecules in Astrophysics Probes and Processes ed. E. F. van Dishoeck IAU Kluwer Dordrecht 1997 p. 321. Dr Schlemmer commented In the –rst two papers in this session and in the Introductory Lecture by Prof. Williams the importance of gas»grain interaction was pointed out as one of the key frontiers in astrophysical chemistry. At the Technical University of Chemnitz we have built a new apparatus to carry out experiments answering some of the questions raised in the papers mentioned. The basic idea of this experimental set-up is to investigate individual single charged nm-sized particles which can be stored very well localized in an electrodynamical trap over days.Fig. 10 shows a schematic view of the apparatus. It is a modi–ed Paul-type trap1 mounted in an ultra-high-vacuum 99 General Discussion Fig. 11 Charge state of a single 500 nm diameter SiO particle as a function of time 2 chamber. Test particles are 500 nm in diameter SiO spheres. These particles are injected 2 from the source (PS) into the trap. For injection the pressure is increased up to 10~3 mbar. Owing to collisions with residual gas particles are decelerated into the eÜective potential of the electrodynamical trapping –eld. The apparatus will be described in detail in a forthcoming publication.2 For continuous monitoring a diode laser (LD) is focused into the center of the trap.Scattered light from the particle is collected with a lens through a –bre and detected by an avalanche photo-diode (APD). The scattered light is modulated by the motion of the particle in the center of the trap. Due to the harmonic eÜective potential of the trap the modulation is characterized by two secular frequencies u for the particleœs motion in the z-direction and u for the particleœs z r motion in the r-direction. These frequencies are proportional to the charge to mass ratio of the stored particle. They are extracted from the raw data by fast Fourier transform of the signal. By this method the particles charge to mass ratio can be followed over a long period of time with a current resolution of 3]10~4.In a –rst experiment the particleœs charge (mostly positive) has been changed with the aid of an electron gun. Due to the high resolution of the frequency measurement changing in steps of elementary charges can be observed. With this additional information the particleœs charge and thus also its mass can be determined. Fig. 11 shows the charge state of a single particle with a mass of 2.4]10~16 kg as a function of time. This experiment demonstrates that a mass determination of single particles in a mass range of \10~16 kg has become possible. In the future this technique of non-destructive mass determination will be used to detect the adsorption and desorption from single nano-grains with the aim of determining sticking coefficients under conditions relevant to astrophysical chemistry.1 W. Paul Rev. Mod. Phys. 1990 62 531 and references therein. 2 S. Schlemmer J. Illemann S. Wellert and D. Gerlich in preparation. Prof. Herbst opened the Discussion of Dr Plumeœs paper Your results indicate that the electron abundance in dense cloud cores is rather constant. Could you comment on the results of Caselli et al.,1 which indicate a much wider range of electron abundances? These authors varied both f (the cosmic ray ionization rate) and elemental depletions. 1 P. Caselli C. M. Walmsley R. Terzieva and E. Herbst Astrophys. J. 1998 499 234. 100 General Discussion Dr Plume responded Our approach and that of Caselli et al. diÜered in the way that cosmic rays were handled.Caselli et al. varied the cosmic ray —ux by two orders of magnitude using the cosmic ray —ux as a true variable in their model calculations (i.e. each core in their sample was –t with a diÜerent value). On the other hand we only varied the cosmic ray —ux to –t the entire sample. By constraining the gas temperature to a range of 8»20 K we used the thermal balance model of Neufeld et al.1 to limit the cosmic ray —ux to a range of 1»15]10~17 s~1. A cosmic ray —ux of 5]10~17 s~1 –t our sample of cores best. Therefore in our subsequent modeling we varied the density temperature and chemical abundances for each core individually but held the cosmic ray —ux constant at 5]10~17 s~1. We feel that this is a more appropriate approach for our cores since they lie in similar regions of the Galaxy (the low-mass cores all lie within a few hundred parsecs of each other and the high-mass cores are all in Orion).Therefore we believe there is no reason to assume that the cosmic ray —ux changes signi–- cantly from core to core. Thus it is only natural that our derived electron abundances show much less scatter than those in the paper as Caselli et al. since we have removed a free parameter whose value can change by two orders of magnitude. 1 D. A. Neufeld S. Lepp and G. J. Melnick Astrophys. J. Suppl. 1995 100 132. Prof. van Dishoeck asked In order to resolve the ambiguity in the modelling between varying the cosmic ray ionization rate or electron density it helps to have 2H` HCS`. Do you have data observations of other ions in the same clouds such as N on any of these species and are they consistent with the DCO`/HCO` analysis ? 2H` emission (opacities are esti- Dr Bergin responded In the low mass cores we have a sub-sample of eight cores in which we have reliable detections of optically thin N mated from the ratios of the hyper–ne components).A comparison of the H13CO` and N2H` abundances demonstrates that the regions with low H13CO` abundance (high electron fraction) also have low concentrations of N2H`. The opposite is true for sources with high H13CO` abundance. Thus for low mass cores the N2H` data support our derived electron fractions although the sample is not large. For high mass sources we have corresponding N2H` observations for –ve cores. For this small sample the N2H` abundance is also correlated with the H13CO` abundance.Dr RoueÜ commented Fractional ionisation may not be uniquely determined from the DCO`/HCO` ratio. The non-linear character of the chemical equation gives rise to possible multiple solutions re—ecting the sensitivity of the system to the initial conditions. 1 Fig. 12 shows the evolution of the ionisation fraction for eight diÜerent initial conditions whereas Fig. 13 displays the ensemble of the steady state solutions for m/nH ranging between 2]10~22 and 4]10~20 s~1 cm3. The vertical line corresponds to Fig. 12. In which way can this complex behaviour be taken into account in the analysis discussed by Plume et al. ? 1 J. Le Bourlot G. Pineau des Fore` ts and E. RoueÜ Astron. Astrophys. 1995 297 251; H.H. Lee E. RoueÜ et al. Astron. Astrophys. 1998 334 1047. Dr Bergin responded We agree that complex bistable solutions discussed in ref. 1 have dramatic eÜects on the ionization fraction and therefore on the coupling between ion and neutrals in cloud cores. In our models we varied the abundances of metals the cosmic ray ionization rate the temperature and the density. In all over 1200 models were run to –nd the best match to the data. In no case was a bistable solution found instead our results are consistent with the low ionization phase (LIP) solution. 1 J. Le Bourlot G. Pineau des Fore� ts and E. RoueÜ Astron. Astrophys. 1995 297 251. 101 General Discussion f/n Fig. 12 Evolution with time of the ionisation fraction for H\4]10~21 s~1 cm3. Eight evolution curves are displayed each one corresponding to diÜerent initial conditions.Prof. Herbst commented In addition to the work of Plume et al. on the ionisation fraction in dense cores a paper by Caselli et al.1 will shortly appear on the subject. In this paper the subject of bistability is mentioned and the following statement made ìWe point out that the parameter range (and in particular the abundances for refractory species) studied by us has the consequence that our solutions are not in general aÜected by conditions of bistability œ. The range of parameters in which bistability appears has been studied in great detail by Lee et al.2 and normally involves higher gas-phase elemental abundances of sulfur than used in most dense cloud models. 1 P.Caselli C. M. Walmsley R. Terzieva and E. Herbst Astrophys. J. 1998 499 234. 2 H-H. Lee E. RoueÜ G. Pineau des Fore� ts O. M. Shalabiea R. Terzieva and E. Herbst Astron. Astrophys. 1998 334 1047. Dr RoueÜ communicated The sulfur elemental abundance is an open problem as mentioned by Prof. Williams in his Introductory Lecture. No de–nitive statement can thus be derived on the absence of bistability. Dr Yates commented One of the most difficult problems we face in molecular line astrophysics is converting the photons we collect to abundances. Most of the molecular Universe is not in local thermodynamic equilibrium (LTE) and molecular clouds or ensemble of solutions displays a ì hysteresis curveœ. ionisation ratio to proton density. Each point is a solution of the steady state equations.The Fig. 13 Steady state values of the ionisation fraction as functions of f/nH the ratio of the cosmic 102 General Discussion envelopes around stars are radiatively coupled over most or all of their spatial extent. This should preclude the use of approximations such as LTE or the large velocity approximation. The historical need for these approximations is noted but there is no excuse for not using non-LTE radiation trant codes which are coupled to the kinetic master equations to compute simulated spectra/maps for comparison with your observed data. My question is therefore What radiation transport methods were used in interpreting the observed data and why are they valid ? Dr Yates added Using a spherical ALI code on a modern computer platform such as a Pentium 200-MHz machine one can compute a solution in minutes even if your observed data contains optically thick transitions as well as optically thin transitions.As an example see Fig. 14 which shows the computed line-shapes for HCO` emission from the protostar B335. The line-shapes show optically thick emission from the low lying J-state transitions and optically thinner emission from the high J-state transitions. Yet these spectra took 426 s to compute. A –nal plea Would modellers please remember that dust continuum emission is a very important source of photons and opacity and should never be excluded from molecular line radiation transport calculations. Prof. van Dishoeck commented I fully agree with Dr Yates that a non-local thermal equilibrium (LTE) non-local radiative transfer code is essential for the analysis of (sub)millimeter data of molecules and the derivation of abundances.For this reason we have developed a 1D and 2D Monte Carlo code,1 which has been tested extensively against the code by Choi et al.2 and is substantially faster. Comparison with your spherical ALI code would be very interesting as well. This new code will be used in all our subsequent analyses on molecular abundances and excitation by far-infrared radiation due to warm dust will be included. In general we derive abundances from optically thin lines wherever available. The situation for the near-infrared absorption lines presented in my paper is somewhat diÜerent from the case for millimeter lines discussed by Dr Yates.Here the analysis follows that commonly used for optical and ultraviolet absorption lines. Because of the low spectral resolving power of the ISO spectrometers the lines are unresolved and a curve-of-growth method or pro–le synthesis analysis must be used. The latter method requires speci–cation of the Doppler parameter (constrained by higher resolution ground-based CO data) the spectral resolution the total column density of the molecule and the populations in each of the rotational energy levels of the lowest vibrational state. In –rst approximation these are indeed assumed to be in LTE because of the high temperatures and densities in the regions but they could be further re–ned using the Monte Carlo program discussed above to obtain non-LTE values.This will indeed be the next step in our analysis. At the current signal-to-noise level of the data however no major departures from LTE are observed so that the column densities for the warm gas component should be reliable within the quoted errors of typically a factor of two. Note also that most infrared lines of and e.g. H2O COfor the warm component are not very optically thick except those arising from the lowest rotational energy levels. 2 1 M. R. Hogerheijde PhD Thesis University of Leiden 1998. 2 M. Choi N. J. Evans E. M. Gregersen and Y. Wang Astrophys. J. 1995 448 742. Dr Hatchell commented In calculating the excitation conditions we have assumed a homogeneous slab in local thermal equilibrium (LTE).For most molecules the assumption of LTE is reasonable at the high densities of these cores (105»109 cm~3). It would be nice to use a radiative transfer approach to unravel the temperature/density gradients in the cores. This may be possible for some of the simpler species but unfortunately for 103 General Discussion Fig. 14 Simulated multi-transition spectrum of the HCO` emission from the protostar B335. The –rst eight rotational transitions are shown here. The intensities are given as JCMT antenna temperatures assuming a dish efficiency of 0.6. The velocities are in km s~1. The physical model was kindly provided by Neal Evans (Austin Texas). more complex molecules such as CH OH which should give the most information on the excitation the collision cross-sections are not known.3 Dr Plume responded For C18O DCO` and H13CO` we used a simple optically thin LTE approach. This is likely true for C18O (due to its low isotopic abundance and the small critical density of this transition) but to test this assumption for DCO` and H13CO` we also observed the rarer isotopes of D13CO` and HC18O` in a small sub-sample of our cores. Comparing the isotopic lines ratios suggested that the opacities were less than unity for DCO` and H13CO`. We therefore assumed it to be true for all of our cores. To account for deviations from LTE due to the low densities used in our model calculations we used a statistical equilibrium calculation to correct the LTE 104 General Discussion derived column densities. For CS however we used a large velocity gradient (LVG) model to calculate the coupled equations of radiative transfer and statistical equilibrium.I agree that the more sophisticated Monte Carlo and accelerated lambda iteration techniques are better able to model the radiative transfer through clouds with realistic physical conditions. We are working on our own versions of these codes at the Harvard»Smithsonian Center for Astrophysics. I would like to point out however that the radiative transfer is only crucial if the opacities are large and our supplementary data indicates that they are likely to be quite small. Additionally since the observed spectral lines were weak (making a meaningful line-shape comparison between the observations and the models difficult) and the physical model for the cores (e.g.density and temperature pro–le) is unknown we feel that our approach is justi–ed. Prof. Irvine asked Dr Rowe In your paper you mention plans to carry out experiments with substituted PAHs. I wonder whether any such experiments have yet been made? Dr Rowe responded A heated nozzle is currently being developed and once this is in operation and characterised experiments on other condensable species such as substituted PAHs will commence. This should be in the next few months. Prof. Sarre said The experimental study of electron attachment to substituted PAHs would be interesting especially for those neutral molecules with large electric dipole moments which can support dipole-bound states of the molecular anions. These dipolebound excited states could also be important in the calculation of radiative association rates.Dr Rowe responded This is a very interesting suggestion and we shall bear it in mind in planning our future experimental programme. Prof. Clary commented A molecule with a dipole moment larger than 2.5 D can form an ìelectron-dipole bound state œ in which the geometry and rotational constants of the molecular anion are almost the same as the parent molecule.1 Over 30 such dipole bound anions have now been observed for systems ranging from acetone to uracil.2 These anions have very small electron ionisation potentials (typically less than 50 meV). The rotational spectra of these anions will be almost identical to the parent neutral molecules and vibrational excitation will normally dissociate the weakly bound electron.Therefore detection of these anions in the interstellar medium through their emission spectra is likely to be difficult even if there are conditions that enable them to be formed. 1 D. C. Clary J. Phys. Chem. 1998 92 3123. 2 C. Defrancois H. Abdoul-Carime and J-P. Schermann Int. J. Mod. Phys. B 1996 10 1339. Prof. Herbst commented The existence of excited electronic states of negative molecular ions can indeed increase calculated rates of radiative attachment if the excited states are connected by dipole transitions to the ground state. A theory concerning the amount of enhancement to be expected for the analogous process of radiative association has been published by Herbst and Bates.1 In the standard mechanism for radiative attachment and radiative association a complex is formed which can only be stabilized by the emission of vibrational (infrared) photons.The existence of bound electronic excited states permits stabilization by the emission of electronic (visible ultraviolet) photons which is generally far more rapid although Franck»Condon eÜects have to be considered. We propose to undertake calculations on the radiative attachment of electrons to bare carbon chains 105 General Discussion e~]C (e.g. 7 ]C7~) since the negative ions are known to possess bound excited electronic states. 1 E. Herbst and D. R. Bates Astrophys. J. 1988 329 410. 3H]e~]1-C3H~]c-C3H~]c-C3H]e~ Prof. Herbst added We have used a statistical theory of radiative attachment reactions to calculate rates of such reactions involving small radicals with large electron affinities.In general results for reactions such as C3N]e~]C3N~]hl show that abundances of small negative molecular ions in interstellar clouds are small. The electron-controlled catalytic reaction 1-C may be important in regulating the isomeric balance between and 1-C3H c-C in dense 3H clouds.1 1 S. Petrie and E. Herbst Astrophys. J. 1997 491 210. Prof. Zare asked What might be the role of dissociative electron attachment in generating negative ions ? How important might this process be as a function of the state of internal excitation of the target ? Prof. Herbst responded In cool molecular clouds the kinetic energy distribution of the electrons is nearly thermal.Under these conditions dissociative attachment must be exothermic to be rapid. This condition implies that the electron affinity of the negative ion product must be greater than the bond energy of the bond being broken. Although some such reactions1 have been suggested to occur in the low temperature interstellar medium and/or circumstellar envelopes viz. MgCN]e~]CN~]Mg they do not appear to produce copious quantities of negative ions. In regions where the electrons possess a signi–cant amount of kinetic energy then dissociative attachment will undoubtedly be more important especially if the targets are internally excited. 1 S. Petrie Mon. Not. R. Astron. Soc. 1996 282 807. Prof. Field commented Dissociative attachment may in principle be of importance in the ISM.Many small species such as H or CO undergo electron attachment via a 2 shape resonance at a few eV electron energy resulting in decay either to negative ion 2 products (H~ or O~ for example) or in vibrational excitation of the target. Crosssections tend however to be rather small especially for H2 . Prof. Field added May I take the opportunity to complement the excellent CRESU data from Rennes with data of our own which corroborates the electron attaching behaviour of PAHs. We have an experiment working on the ASTRID synchroton at ISA Aarhus and the Daresbury Synchrotron Radiation source which allows us to form electron beams down to a few meV energy with a few meV energy resolution. Data which are shown in Fig. 15 demonstrate that electrons are very strongly scattered out of the incident beam as they pass in this case through naphthalene vapour.Cross-sections are several hundred ”2 at 10»20 meV electron energy. Scattering events are very likely due to electron attachment. Further details may be found in a paper shortly to appear1 showing similar behaviour in benzene. We have similar data for a number of PAHs. 1 R. J. Gulley S. L. Lunt J-P. Ziesel and D. Field J. Phys. B. 1998 31 2735. Prof. Smith said The cross-sections for electron capture by naphthalene shown by Prof. Field exhibit a strong decrease with collision energy. Could he comment on the expected temperature dependence of the corresponding rate coefficients in the light of the result of Rowe and his co-workers of a temperature independent rate coefficient in the case of anthracene? 106 General Discussion Fig.15 Variation of the backward scattering cross-section for scattering of electrons by naphthalene as a function of electron impact energy. Prof. Field replied We do indeed –nd that the cross-section drops quite sharply in higher energy collisions between electrons and naphthalene. Of course we are looking at a temporary attachment process. The lifetime of the negative ion is however certainly increasing as the electron impact energy drops. The data of Rowe and co-workers involves the measurement of a third order rate coefficient as a function of temperature. In the low pressure limit of true third order the behaviour of this rate coefficient should indeed re—ect the lifetime of the anthracene negative ion (in this case).Therefore one might expect the third order rate coefficient to rise at very low temperatures. There are two points to consider however. Is it certain that the data of Rowe are in the third order limiting region ? (This is a question not a statement !) Also bulk data do need to be convoluted with the thermal spread of course. I would therefore not say at this stage that there is a clear discrepancy here but we should certainly look into it. The Chairman then invited discussion on any of the papers in the afternoon session and on the Introductory Lecture. Prof. Zare said to Prof. Williams I would like to encourage you to sharpen your de–nitions of astrochemistry. In your talk you de–ned astrochemistry as the intersection of observations processes and modeling.What diÜerence is there in your mind between astrophysics astrochemistry and astrobiology ? Is that diÜerence an important one? Prof. Williams responded Yes in my own mind I do make a distinction between astrophysics and astrochemistry. In astrophysics I regard the emphasis as being on using models to determine astronomical parameters such as density temperature mass dynamics etc. In astrochemistry I regard the emphasis as being on anything to do with molecules. The two do not entirely overlap. Astrobiology is surely an extension of both but there must be a great deal of commonality between the three. Prof. Vidali said I would like to make a few comments about the experimental results mentioned in Prof.Williamœs Introductory Lecture. We have built an apparatus to study hydrogen recombination reactions on surfaces in astronomically relevant conditions. Experiments were conducted using two low energy hydrogen and deuterium 107 General Discussion atomic beams impinging on a sample kept at 6»20 K in an ultra-high vacuum chamber. Our main results are (1) Recombination kinetics depend quadratically on reactant concentration at low coverage. (2) Mobility is dominated by thermally activated processes. Based on these results we propose a model consisting of a set of rate equations for the recombination of hydrogen in diÜerent astrophysical environments.1 The results quoted by Prof. Williams2 were for an olivine natural polycrystalline sample.Qualitatively similar results are found for a sample of amorphous carbon.3 The evidence is strong that thermal activation is the primary agent for diÜusion of H and D atoms. A somewhat longer recombination efficiency is seen for amorphous carbon with respect to olivine. 1 O. Biham I. Furman N. Katz V. Pirronello and G. Vidali Mon. Not. R. Astron. Soc. 1998 296 869. 2 V. Pirronello O. Biham C. Liu L. Shen and G. Vidali Astrophys. J. 1997 483 L131; V. Pirronello C. Liu L. Shen and G. Vidali Astrophys. J. 1997 475 L69. 3 V. Pirronello J. Roser and G. Vidali in preparation. Prof. Breç chignac commented Prof. Williams in his lecture has addressed the problem of the formation of molecular hydrogen at the surface of grains pointing out the experimental eÜorts being made in London as well as the results on the recombination efficiency obtained in Syracuse (USA) by Professor Vidali and his co-workers.He has shown some results of theoretical simulations for the case of an H2O-ice surface by Takahashi et al. and of preliminary quantum calculations for the case of a graphite surface in Prof. Claryœs group. In both cases the vibrational distribution of the newly formed molecule seems to peak at some vibrational quantum number of the order of 3 to 4. He also mentioned the poster by Dr Parneix and myself presenting the results of classical trajectory simulations for the recombination of a graphite surface. Since our results for the vibrational distribution seem to diÜer from those I wish to tell you more about what we did.1 We used a model potential energy surface for both the sticking of a single H atom to the graphite surface considered as a rigid hexagonal pavement of C atoms and the reactive part when a second H atom approaches.The model was adjusted on some available ab initio data for particular geometries. The Eley»Rideal process i.e. the reactive collision between an atom from the gas phase and a previously chemisorbed atom H]SwH]H2]S was simulated by running classical trajectories on this potential energy surface. Statistical averages were done over a large number of trajectories probing all the possible initial conditions. The results have been analysed as a function of the collision energy Fig. 16 Ensemble averages of the translational vibrational and rotational energies of the H mol- 2 ecule as a function of the initial collision energy of the impinging H atom 108 General Discussion (gas phase temperature) and of the initial coverage of the surface for the efficiency (reaction probability) the distribution of the excess energy among the various degrees of freedom and some observables which could be measured experimentally such as the angular distribution.The efficiency of the reaction has been found to be large (probability ranging from 0.3 to 0.6 in case of a 5% coverage) with a threshold in collision energy associated to the presence of a small barrier in the potential surface. It grows up to saturation as the coverage is increased. Fig. 16 shows the partitioning of the excess energy in the diÜerent degrees of freedom of the nascent H molecule.Clearly the distribution is almost independent of the collision energy which is consistent with the fact that most of the avail- 2 able energy comes from the binding energy of the H molecule. About 50% of the energy is transferred into the translation of the ejected molecule which in astrophysical condi- 2 tions must have an eÜect in the heating of the interstellar gas. About 20% of the energy is found in excited rotational states (peaking near J\10 or more) which may have a spectroscopic importance for the comparison with either experiment or astrophysical observations. The remaining 30% is found in vibrational excitation. The relative vibrational population distribution is shown in Fig. 17. Excitation up to the level v\6 is observed but with a smoothly decreasing distribution and no sign of any inversion. However I should mention that we have investigated the eÜect of a little change in the shape of the potential energy surface. Although no dramatic change was found an eÜect was seen on the vibrational distribution. Then these results on the vibrational distribution of H formed by recombination of H atoms at a graphite surface seem very diÜerent from the ones obtained by Prof. Clary. Many points seem to diÜer in 2 the two approaches (the potential energy surface the restricted geometry the quantum/ classical treatment of the dynamics) which will have in view of the strong astrophysical interest of the process to be considered in order to clear up possible discrepancies. It may be the beginning of a debate around this question on the theoretical expectations –rst with the hope that the experiment will help to approach the physical reality for the best of astrochemistry. 1 P. Parneix and Ph. Breç chignac Astron. Astrophys. 1998 334 363. population in v\0 Fig. 17 Vibrational population distribution of the newly formed H molecule relative to the 2

ISSN:1359-6640    

DOI:10.1039/FD109083

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

Faraday Discuss. 1998 109 109»119 Electronic spectroscopy of carbon chains and relevance to astrophysics David A. Kirkwood Harold Linnartz Michel Grutter Otto Dopfer Tomasz Motylewski Mikhail Pachkov Marek Tulej Muriel Wyss and John P. Maier* Institute for Physical Chemistry University of Basel Klingelbergstrasse 80 CH-4056 Basel Switzerland Laboratory measurements of electronic transitions in carbon chains are presented. Guided by the results found in neon matrices gas-phase spectra have been obtained with two diÜerent experimental approaches. Electronic transitions of neutral carbon chains are detected in a supersonic slit jet plasma by cavity ring-down (CRD) spectroscopy and those of mass-selected anions are probed by resonant two-colour electron photodetachment.The successful combination of matrix and gas-phase experiments is demonstrated for the examples of C6H C and These and other previous measure- 14H~. ments on electronic transitions of long carbon chains in the gas phase are compared with the tabulated wavelengths of the diÜuse interstellar bands. In A2% the case of the u»X2%g transition of C7~ the origin and three intense bands in the gas-phase spectrum match within ^0.2 nm error limits diÜuse interstellar bands which have comparable FWHMs and equivalent widths. Carbon chains are considered among the attractive candidates as carriers of some of the diÜuse interstellar bands (DIBs).1 Following the initial suggestion and arguments by Douglas,2 this concept has received further encouragement on one hand by the actual detection of molecules with a carbon-chain skeleton in dense interstellar clouds by microwave spectroscopy3 and on the other by understanding the spectral properties of the characteristic p»p electronic transitions.4 The latter has come about following the detection of such spectra in absorption for a number of homologous series using the technique which combines mass-selection with matrix isolation.5 Thus to answer questions as to which types and sizes of carbon chains their ions and simple derivatives with H N and O need to be spectroscopically characterized in the gas phase the following strategy has been pursued.The initial step is the observation and identi–cation of the vibronic spectra of the species under study in neon matrices.With this information in hand the planning of gas-phase experiments becomes a realistic proposition. In fact the gas-phase results reported in this article as well as those in recent publications on carbon anions,6 substantiate this philosophy. Independent of such measurements the patterns in the spectroscopic signatures of carbon chains can then be considered together with astrophysical constraints as well as chemical ones to decide which types and sizes should be measured in the gas phase. Clearly the systematic detection of the electronic spectra in the gas phase for numerous species as has proven possible in neon matrices is currently not trivial. The dependence of the electronic spectra of carbon chains in neon matrices on the size of the system has led to the conclusion that species such as C (n\8»18 or so) 2n`1 and their isoelectronic analogues would have their strong p»p electronic transitions in 109 110 Electronic spectroscopy of carbon chains and of the largest species to date the 400»900 nm wavelength range,4 where diÜuse interstellar absorptions are mainly observed.7 In this article the emphasis is placed on the observation of the gas-phase electronic transitions of a few species which however from size and stability considerations alone would not necessarily be a priori the ones selected for comparison with astrophysical data.Rather owing to the difficulties of gas-phase studies these should be regarded as pilot approaches to show that such measurements have now become feasible.In particular this has led to the –rst observation of electronic spectra of neutral carbon chains C C14H~ in the gas phase. nH Two diÜerent experimental strategies are considered (1) the detection of neutral carbon chains in absorption by using cavity ring-down (CRD) spectroscopy in a supersonic slit jet expansion plasma and (2) resonant two-colour electron photodetachment of mass-selected anions. In both cases the observation of electronic transitions in a neon matrix was the starting point. In the –nal section selected gas-phase data on electronic transitions of carbon chains and their ions are directly compared with the catalogue of DIBs.7 Gas-phase electronic spectrum of C6H The 2%^X2% electronic absorption spectrum of C6H was initially identi–ed in a 5 K neon matrix using the mass-selective technique.5 Based on this information this transition was detected in the gas phase by means of CRD spectroscopy in a hollow cathode discharge cell.8 The approach has now undergone further development increasing the sensitivity of the measurement by using a plasma generated in a supersonic slit jet expansion.Experimental CRD spectroscopy has proven in recent years to be a powerful technique for direct absorption spectroscopy. The intrinsic insensitivity to light-source —uctuations and the very long absorption lengths make it ideally suited for the study of weak molecular transitions or of molecules that are difficult to generate in large abundance. Since its introduction as a spectroscopic tool in 1988,9 it has been applied in diÜerent –elds trace-gas detection combustion discharges and molecular beams.10,11 Molecular ions could also be detected in a pulsed hollow-cathode discharge,12 leading to estimated detection limits of around 107 molecules cm~3.This set-up was the one used to detect the 2%»X2% electronic transition of C sented whereby the hollow cathode has been replaced by a supersonic expansion 6H.8 Here the extension of this approach is preplasma. In the last few years much eÜort has been put into the development of techniques that combine the advantages of a two-dimensional supersonic expansion through a small slit (high densities adiabatic cooling and complexation) and plasma generation (ion and radical production). Direct absorption spectra of several rotationally cold molecular ions radicals and an ionic complex have been reported using production techniques ranging from ablation,13 photolysis14 and discharge,15 to electron impact ionization.16 Impressive have been the observations of rotationally resolved IR spectra of carbon chains C nO13 following excimer laser ablation of a carbon rod posi- n tioned in the throat of the slit.17 The current experimental set-up consists of a standard CRD unit sampling a supersonic slit jet (Fig.1). The latter is located in a large stainless-steel cross-piece evacuated by a Roots blower. The mirror housings are connected with —exible bellows to the opposite sides of the cross-piece. Two anti-re—ection coated quartz windows are –xed outside the cavity and seal the chamber.The deposition of contaminants on the mirrors is restricted by a curtain of helium. The nozzle is mounted in the centre of the chamber 111 D. A. Kirkwood et al. Fig. 1 Schematic of the experimental set-up for the measurement of electronic absorption spectra of neutral carbon chains by CRD spectroscopy with its slit parallel to the laser beam. The distance of the ori–ce to the optical axis can be varied from 1 to 15 mm. The gas (0.2 to 1% C2H2 or C2D2 in helium 5 bar backing pressure) is expanded through a 3 cm]250 lm pulsed slit (Fig. 1). The pressure in the chamber during operation is around 150 mTorr. The ori–ce of the slit comprises an insulator metal plate second insulator and two sharp jaws to which a pulsed negative voltage of 800»1200 V (50 mA 100 ls) is applied.The inner metal plate is grounded while the body —oats. This con–guration leads to a discharge over the total length of the slit. Furthermore carbon dust is not produced inside the slit and thus the discharge is stable for many hours. Although the volume in the slit has been kept small the eÜective gas pulse has a small delay and smears out the valveœs nominal opening from 350 to typically 500»750 ls. The whole experiment runs at 30 Hz; a master signal is sent to a delay generator from which the excimer laser gas valve and discharge are independently triggered. With typical ring-down times of 40 ls eÜective absorption lengths of ca. 1 km are achieved. 2PñX2P Electronic transition of C6H The initial observation of the 2%»X2% transition in the gas phase was in a hollowcathode discharge.8 The resulting spectrum (T rotB350 K) reveals only the unresolved R-branches of the two spin»orbit components 2%1@2»X2%1@2 and 2%3@2»X2%3@2 [Fig.2(a)]. Consequently it was possible to determine neither accurate rotational constants nor the spin»orbit constant A@ for the excited state. In contrast the ground electronic state parameters are well characterized from the microwave spectrum of C6H.18 Fig. 2(b) shows a low-resolution (0.1 cm~1 laser bandwidth) spectrum in the region of the origin band of the 2%»X2% electronic transition of C6H recorded in the slit jet expansion. The rotational temperature is ca. 45 K. The P and R branches of both spin»orbit com- 2% ponents are visible as well as the Q-branch of the 3@2»X2%3@2 sub-band.This is because of the more favourable Hoé nl»London factors compared to the X\1/2 system. T Furthermore at rot\45 K the population density for the X\3/2 component is ca. 30% higher than for X\1/2 ; the ground state spin»orbit splitting is ca. [15 cm~1. The Q-branches are not observed in the warm spectrum [Fig. 2(a)] due to the population distribution being over many rotational levels. The Q-heads of the origin bands for the X\3/2 and 1/2 components are observed and estimated at 18 985.5(2) and 112 Electronic spectroscopy of carbon chains Fig. 2 (a) Spectrum of the origin band of the 2%»X2% transition of C6H measured in a hollowcathode discharge at 350 K. The two heads are in the R-branches of the X\3/2 and 1/2 com- T rot\45 K measured in a supersonic slit nozzle plasma.P and R ponents. (b) Spectrum at branches are visible as well as the Q-branch of the 2%3@2»X2%3@2 band. the origin bands are shifted by 6H. 18 994.7(4) cm~1 respectively. In the spectrum of C6D 51.3 cm~1 to higher energies compared with C An upper-state spin»orbit constant A@\[23.5(2) cm~1 is obtained assuming a ground-state value AA\[15.09 cm~1 determined from the microwave spectrum.18 In the analysis of the spectrum in the hollow-cathode discharge,8 only oA@»AA o could be deduced because the Q-branches were not discernible and it was assumed that A@\AA. However the spectrum in Fig. 2 reveals clearly that A@[AA. Consequently the blue shift of the origin band (X\3/2 component) in the gas phase relative to its position in the neon matrix is 131 cm~1 rather than the 142 cm~1 value quoted previously.8 This shift can be compared with the gas»neon matrix shift of 122 cm~1 observed for the 2%»X2% transition of the isoelectronic triacetylene cation.4 component.A standard 2%»X2% Hamiltonian was used to –t more than The spectral measurements in the slit nozzle are nearly Doppler free and thus the rotational structure of the transition could be resolved. With an intracavity etalon a laser bandwidth of 0.04 cm~1 can be achieved which is comparable with the rotational constant of C6H. Fig. 3 presents the rotationally resolved origin band of the 2%3@2»X2%3@2 113 D. A. Kirkwood et al. 2% Fig. 3 Rotational structure of the origin band in the 3@2»X2%3@2 electronic transition of C6H measured by CRD spectroscopy in a slit nozzle plasma 30 observed transitions in the X\3/2 system mainly occurring in the P-branch.In the –t the accurate set of spectroscopic constants inferred from the microwave spectrum was used for the ground electronic state. The upper-state rotational constant is determined as B@\0.045 563(23) cm~1 where D@ has been kept –xed to the ground-state value. In the case of the X\1/2 system the lower signal-to-noise ratio precluded the accurate measurement of the line positions. This technique has now also been applied successfully to detect the corresponding 2%»X2% electronic transitions of longer members of this homologous series. The centre frequencies of the unresolved origin bands of C8H C10H and have been identi–ed at 15 973.5(1.0) cm~1 and 14 000(3) cm~1 respectively.19 (n\2»7),21 C2nH (n\3»8),20 Electronic spectrum of C14Hó Absorption spectrum in a neon matrix The combined technique of mass-selection with matrix isolation has been used to identify the characteristic electronic transitions of a number of homologous series of carbon chains.These include neutrals C2n (n\2»5),20 C2n`1 (n\2»7),22 anions HC2n`1H Cn~ HCnH` (nO16).25 In the (nO20),23,24 and cations measurements described here C14H~ was chosen as an example of a longer species with the goal of detecting its electronic transitions in the gas phase. As a –rst step the electronic absorption spectrum in a neon matrix was sought. A hot-cathode anion source was fed with a diacetylene»argon mixture (1 5) and the resulting ion beam was massselected and codeposited with neon at 6 K on a rhodium-coated sapphire plate.C2nH~ are formed for nO7 whereas the larger anions contain Mainly the species predominantly 2 and 4 hydrogen atoms. Following the deposition of mass-selected C14Hx~ ions (x\0»4 with 1 2 being the strongest) the absorption spectrum was measured with the waveguide technique26 and is shown in Fig. 4 (top). The mass-resolution 114 Electronic spectroscopy of carbon chains was deliberately degraded to obtain higher anion currents. When the resolution was increased to select a speci–c C14Hx~ species the observed band system was strongest for x\1. The observation that the system is also weakly present when depositions with x\2 4 are performed can be understood in terms of fragmentation due to the high kinetic energy of the ions.The 420 nm band system can be eliminated through irradiation of the matrix with photons from a medium-pressure mercury lamp after which the known 2%^X2% electronic transition of neutral C14H becomes apparent. with origin band at 865 nm,20 14H in neon matrices27 and in solution up to n\12,28 but they lie this band system is ca. 1.5 eV 14H~. This diÜerence is quite large for isoelectronic species and conse- The observed electronic absorption spectrum (Fig. 4) shows features characteristic of p»p transitions of carbon chains a strong origin band and vibrational excitation of a number of stretching modes. As is often the case in matrices degenerate bending modes in even quanta are probably also excited.Corresponding band systems have been observed for other members of this series C2nH~ (n\3»9) and a plot of the wavelength of their origin bands as a function of the number of carbon atoms shows an approximately linear dependence,27 a characteristic feature of a homologous series. Because the observed band systems of the C2nH~ molecules are intense comparable with the other allowed p»p ones of carbon chains hitherto observed it is supposed that they are due to 1&`^X1&` electronic transitions. These have also been observed for the isoelectronic polyacetylenes HC to higher energies. For example in the case of HC 2nH above that of C quently the assignment of the band system (Fig.4) to a forbidden 1*^X1&` transition was considered. The latter band system is observed for the smaller members of the HC series n\2 3 in the gas phase29 and for nO5 in solution.30 Extrapolation of 2nH these transition frequencies to n\7 leads to energies close to the observed band of 14H~. Thus the 1&`^X1&` assignment C14H~. However the 1*^X1&` band system is not observed for the longer polyacetylenes and no second intense band system is apparent in the neon matrix spectrum of above the electron detachment threshold of C C14H~ at energies higher than that of the 420 nm one. Transitions above 5 eV would lie is preferred. Another explanation for the unusually large shift (ca. 1.5 eV) relative to 3 Fig. 4 Electronic transition of C14H~ recorded by direct absorption at 6 K in the neon matrix (top trace) and resonant two-colour electron photodetachment spectroscopy in the gas phase at T B100 K (bottom trace).The asterisk denotes the absorption peak of C formed by fragmentation. 115 D. A. Kirkwood et al. HC14H C may be that the electronic structure of 14H~ is more cumulenic than acetylenic in character or that the geometric structure may not be linear e.g. bent or ringchained as has been observed for similar neutral species.3 Consequently the nature of the observed transition in C14H~ cannot be de–nitively determined. in the gas phase (1) (2) Electronic spectrum of C14Hó Experimental 4~»C9~.6,31,32 Based on the information on the electronic transition of C14H~ observed in a neon matrix the search in the gas phase was undertaken.The C14H~ anions were generated in a pulsed DC gas discharge through a mixture of acetylene and argon followed by a supersonic expansion. This produces anionic carbon species of the general form CnHm~ which were subsequently mass-selected in a Wiley»MacLaren time-of-—ight spectrometer. The resonant two-colour electron photodetachment (R2CPD) technique was then employed to measure their electronic transitions. This approach has already been used C to measure the 2%»X2% electronic transitions of linear carbon anions C2~ The R2CPD approach is a pump and probe experiment in which two laser beams intersect a mass-selected anion beam. Mass selectivity is obtained by varying the delay of the laser pulses such that they interact with only the carbon anion of choice.The two photons (hl and hl 1 2) are generated by an excimer pumped dye and a Nd YAG laser respectively with l tuned over the expected wavelength range of the transition of the 1 mass-selected ion (1). The second photon hl from the excited species but not from the ground state (2) 2 has sufficient energy to detach an electron M~*]hl M~]hl1 ]M~* 2 ]M]e~ The resultant neutral product is then detected at a multichannel-plate array. By applying a retarding voltage to a grid above the detector which delays the arrival of the anion beam with respect to the neutrals it is possible to detect parent ions and product neutrals thus enabling normalisation of the spectra for —uctuations in the production of parent cluster ions in the source.Gas-phase spectrum The R2CPD spectrum of the expected band system of C14H~ is shown in the lower trace of Fig. 4. There is a good correspondence between the stronger gas-phase and the neon matrix bands. The gas-phase spectrum however yields a greater number of spectral features but those which correlate with matrix bands indicate a neon-environmentinduced blue shift of 50^20 cm~1. It should however be pointed out that this shift refers to the maxima of the bands observed in the neon matrix. Owing to the width of these peaks the shift with respect to the zero-phonon line (which is not discernible) will be less. Blue shifts on passing from the gas phase to the neon environment are unusual being only observed for electronic transitions of a few of the smaller carbon anions Cn~ (n\2 4 5).For neutral and cationic carbon chains red shifts are the rule.4 The positions of the band maxima are listed in Table 1 together with possible vibrational assignments. As in the neon matrix at 6 K the relatively low temperature (ca. 100 K) of the anions in the gas phase produced in the supersonic expansion means that the prominent bands in the absorption spectrum originate from the lowest vibrational level in the ground electronic state. The Franck»Condon accessible vibrational levels in 116 Electronic spectroscopy of carbon chains the upper electronic state are then mapped out in the spectrum. The choice of the speci –c stretching modes excited in the upper electronic state was made on the basis of HF/6-31G** calculations carried out for the ground electronic state.Thus the vibrational assignment has to be considered as tentative. Table 1 Band maxima of vibrational transitions in the electronic absorption spectrum of C neon matrix at 6 K and in the gas phase at ca. 50 K transition lgas/cm~1 00 0 23 13 10 0 2 0 1 0 1 100 1 130 1 80 1 or 160 2 30 1 100 2 80 1 210 2 or 160 2 210 2 10 1 23 775 23 782 24 258 24 351 25 047 25 104 25 338 25 373 25 445 25 519 25 606 25 622 25 761 26 051 26 106 26 152 26 210 26 324 26 430 26 578 26 697 26 902 26 954 27 021 27 065 100 3 Tentative vibrational assignments are given based on HF/6-31G** ground-state calculations.The error in the gas-phase values is \5 cm~1 and ^15 cm~1 in the matrix. Relation to astrophysical observations In order to compare electronic transitions of carbon-chain type molecules observed in the gas phase with the DIBs,7 it is necessary that the band maximum can be de–ned within certain error limits. A complication which exists in comparing the laboratory and astronomical data is the eÜect of temperature on the band pro–les. The band maximum estimated from the unresolved rotational contour will shift with temperature. This can be seen for example in measurements on the origin band of the A2%u»X2%g transition of diacetylene cation in a supersonic free jet.33 A shift of 5 cm~1 would be apparent on passing from 300 to 10 K.In the compilation of Table 2 only the species for which the band maximum can be 14H~ observed in a lmatrix/cm~1 23 848 24 307 24 387 25 086 25 483 25 676 25 833 26 086 26 267 26 751 26 984 27 482 determined within ^10 cm~1 are listed but a larger uncertainty is given to accommodate for possible band maxima shifts. The data include the 2%»X2% electronic transitions of the polyacetylene cations HC2nH` (n\2 3 4) and the related cyano-derivatives HC4CN` NCC2nCN` (n\2 3) which were detected as emission spectra.4 The most recent additions are the 2%»X2% band systems of the carbon anions In the case of C Cn~ (n\4»8),6,31,32 as well as the presently reported system of C14H~ all detected by resonant electron photodetachment spectroscopy.The 2%»X2% band systems of three neutral carbon chains C2nH (n\3 4 5) are the measurements via CRD absorption spectroscopy. In all cases coincidences were sought with DIBs that are claimed to have been identi–ed (designated with]in ref. 7). For most of the molecules listed in Table 2 no 6~ and coincidences were found except for C 7~. 0 The 0 0 bands in the electronic spectra C of these anions at 606.8(4) nm (C6~) ( and 627.1(2) nmC7 ~) show coincidences with DIBs situated at 606.538(4) and 627.01(3) nm respectively. In order to check whether this is accidental the other intense bands in the laboratory absorption spectrum were then compared. 7~ the match between the laboratory frequencies of the origin and vibrationally excited bands with DIB positions is striking.This is illustrated in Fig. 5 where the laboratory spectrum (top trace) can be compared with a simulation of the tabulated values of the DIBs (bottom trace).7 It has been proposed that in the case of several coincidences for one speci–c species two criteria should be considered.34 First the DIBs picked out should have comparable half-widths (FWHM) and secondly the ratio of the relative intensities of the laboratory bands should show a similar trend to that of the equivalent widths of the corresponding DIBs. As can be seen in Table 3 this is approximately the case for four bands of C7~. The FWHMs of the DIBs are comparable and their equivalent widths are not too dissimilar to the relative laboratory intensities.The only other intense band 10 130 1 observed in the laboratory spectrum lies in a region of an intense broad DIB at 544.96 nm. Owing to its FWHM and intensity this belongs to another carrier and may obscure a weaker narrow band. Table 2 Wavelength of the origin bandœs maximum in the electronic transitions of carbon chains in the gas phase known with accuracy O10 cm~1 molecule C4~ HC4H` NC4N` HC4CN` C C6~ 6H HC6H` NC6N` C 8~ 7 ~ C C 8H C C10H 14H~ a The uncertainty given in parentheses is estimated from the data shown in the cited publications. 117 D. A. Kirkwood et al. ref. transition kmax/nm 2%u ^X2%g u ^X2%g 6 4 4 2 2 % % g ^X2%u 2% ^X2% 4 32 2%g ^X2%u 2% ^X2% 2% ^X2% 4 4 32 31 2% ^X2% 2%u ^X2%g 2%u ^X2%g 19 19 457.1(1) 506.9(3) 595.8(4) 581.6(2) 606.8(4) 526.6(1) 599.7(4) 655.1(4) 627.1(2) 771.2(6) 625.9(1) 714.1(3) 427.7(1) 2% ^X2% 2% ^X2% 1&`^X1&` 118 Electronic spectroscopy of carbon chains length region simulated from the compilation in ref.7 A2% Fig. 5 Gas-phase spectrum of the u ^X2%g electronic transition of C7~ measured in the laboratory (top trace) reproduced from ref. 31 with kind permission and the DIBs in this wave- A higher-lying electronic transition of C7~ (B 2%»X2%) has been observed in the laboratory,31 but the bands are too broad for a meaningful comparison with DIBs. This is also the case for the bands in the electronic spectra of the related anions C5~ C9~ and C11 ~ .31,32 Thus on the experimental side a goal should be to record the transitions of the latter anions under conditions similar to those achieved for the C7~ spectrum leading to sufficiently narrow bands for a direct comparison with astronomical data.It has been suggested that the carriers of DIBs must have a certain size to be photostable in the interstellar radiation –eld. Estimates of the likely size of the molecule necessary for stability were proposed earlier as ca. 10»15 atoms and more recently as at least 50 atoms.35 In view of the remarks made in the introduction of this article concerning the conclusions drawn from the observations made on the electronic spectra of carbon chains characterized in neon matrices another goal should be the measurement of the spectra of species such as C2n`1 (n[7) or of their isoelectronic analogues in the gas phase.Table 3 Comparison of the laboratory wavelengths (kmax) and relative inten- A2%u ^X2%g electronic transition of C7~ sities (I) of vibrational bands in the with astronomical data7 astronomy laboratory EW FWHM/nm DIB/nm transition I kmax/nm 627.1(2) 606.4(2) 574.7(2) 0.08(2) 0.014(3) 0.044(1) 0.10(2) 0.09(1) 0.23(3) 627.01(3) 606.538(4) 574.781(8) 0.035(8) 0.17(4) 560.996(23) 1.0 0.3 0.5 0.6 0.5 561.2(2) 544.8(2) 00 0 30 1 20 1 10 1 10 1 30 1 Band positions FWHM and equivalent widths (EW) of the matching DIBs are tabulated.119 D. A. Kirkwood et al. This research has been supported by the Swiss National Science Foundation (project no. 20-29104.96). Paper 8/00072G; Received 2nd January 1998 References 1 T he DiÜuse Interstellar Bands ed. A. G. G. M. Tielens and T. P. Snow Kluwer Dordrecht 1995. 2 A. E. Douglas Nature (L ondon) 1977 269 130. 3 M. C. McCarthy M. J. Travers A. Kovacs C. A. Gottlieb and P. Thaddeus Astrophys. J. Suppl. 1997 113 105. 4 J. P. Maier Chem. Soc. Rev. 1997 26 21. 5 D. Forney J. Fulara P. Freivogel M. Jakobi D. Lessen and J. P. Maier J. Chem. Phys. 1995 103 48. 6 Y. Zhao E. de Beer and D. M. Neumark J. Chem. Phys. 1996 105 2575. 7 P. Jenniskens and F-X. Deç sert Astron. Astrophys. Suppl. 1994 106 39. 8 M. Kotterer and J. P. Maier Chem.Phys. L ett. 1997 266 342. 9 A. OœKeefe and D. A. G. Deacon Rev. Sci. Instrum. 1988 59 2544. 10 J. J. Scherer J. B. Paul A. OœKeefe and R. J. Saykally Chem. Rev. 1997 97 25. 11 R. Engeln G. Berden E. van den Berg and G. Meijer J. Chem. Phys. 1997 107 4458. 12 M. Kotterer J. Conceicao and J. P. Maier Chem. Phys. L ett. 1996 259 233. 13 J. R. Heath and R. J. Saykally in On Clusters and Clustering from Atoms to Fractals Elsevier Amsterdam 1993 pp. 7»21. 14 R. F. Curl K. K. Murray M. Petri M. L. Richnow and F. K. Tittel Chem. Phys. L ett. 1989 161 98. 15 G. Hilpert H. Linnartz M. Havenith J. J. ter Meulen and W. L. Meerts Chem. Phys. L ett. 1994 219 384. 16 T. Ruchti A. Rohrbacher T. Speck J. P. Connelly E. J. Bieske and J. P. Maier Chem. Phys. 1996 209 169; T.Speck H. Linnartz and J. P. Maier J. Chem. Phys. 1997 107 8706. 17 T. F. Giesen A. van Orden H. J. Hwang R. S. Fellers R. A. Provenc” al and R. J. Saykally Science 1994 265 756. 18 J. C. Pearson C. A. Gottlieb D. R. Woodward and P. Thaddeus Astron. Astrophys. 1988 189 L13. 19 H. Linnartz T. Motylewski and J. P. Maier unpublished work. 20 P. Freivogel J. Fulara M. Jakobi D. Forney and J. P. Maier J. Chem. Phys. 1995 103 54. 21 D. Forney P. Freivogel M. Grutter and J. P. Maier J. Chem. Phys. 1996 104 4954. 22 J. Fulara P. Freivogel D. Forney and J. P. Maier J. Chem. Phys. 1995 103 8805. 23 D. Forney M. Grutter P. Freivogel and J. P. Maier J. Phys. Chem. A 1997 101 5292. 24 P. Freivogel M. Grutter D. Forney and J. P. Maier J. Chem. Phys. 1997 107 4468. 25 P. Freivogel J. Fulara D. Lessen D. Forney and J. P. Maier Chem. Phys. 1994 189 335. 26 R. Rossetti and L. E. Brus Rev. Sci. Instrum. 1980 51 467. 27 M. Grutter M. Wyss J. Fulara and J. P. Maier unpublished work. 28 R. Eastmond T. R. Johnson and D. R. M. Walton T etrahedron 1972 28 4601. 29 R. E. Bandy C. Lakshminarayan and T. S. Zwier J. Phys. Chem. 1992 96 5339. 30 E. Kloster-Jensen H. J. Haink and H. Christen Helv. Chim. Acta 1974 57 1731. 31 M. Tulej D. A. Kirkwood G. Maccaferri O. Dopfer and J. P. Maier Chem. Phys. 1998 228 293. 32 Y. Zhao E. de Beer C. Xu T. Travis and D. M. Neumark J. Chem. Phys. 1996 105 4905. 33 R. Kuhn J. P. Maier and M. Ochsner Mol. Phys. 1986 59 441. 34 J. P. Maier Nature (L ondon) 1994 370 423. 35 T. Allain S. Leach and E. Sedlmayr Astron. Astrophys. 1996 305 602.

ISSN:1359-6640    

DOI:10.1039/a800072g

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

Faraday Discuss. 1998 109 121»135 New carbon chains in the laboratory and in interstellar space P. Thaddeus M. C. McCarthy M. J. Travers C. A. Gottlieb and W. Chen Center for Astrophysics and Division of Applied Sciences Harvard University Cambridge MA 02138 USA Twenty seven new carbon-chain molecules have been detected over the past two years with a Fourier-transform microwave (FTM) spectrometer 13 of which are reported here for the –rst time. Of the 27 11 are closed shell polyynes 9 are free radicals 2 are cumulene carbenes and 5 are carbenes formed by substituting carbon chains for one of the hydrogen atoms of the tional transitions [including hyper–ne structure (hfs) for the radicals] of the three member carbene ring C3H2 . All the astronomically interesting rotaentire set have either been measured to high precision or are readily calculated to comparable accuracy from the spectroscopic constants derived from the laboratory data.On the basis of this data 4 of the chains (C H 7H C8H 2C6 HC11N) and have already been detected in astronomical sources and with large radio telescopes under construction or the discovery of better astronomical sources it is possible that nearly all can be found. Astronomical detection is aided by the apparent high polarity of all unsymmetrical chains. The sensitivity of the present liquid-nitrogen-cooled spectrometer is far from fundamental limits ; an increase by one to two orders or magnitude is possible with liquid helium cooling and other re–nements and better precursor gases may be found.Many of the chains here probably have lowlying isomers and ionized and radical variations which may be detected by the present techniques. Carbon chains are hard to stretch but easy to bend centrifugal distortion is well described by a classical model according to which all chains distort under rotation like classical thin rods with the same Youngœs modulus E\1.7]1013 dyn cm~2 larger than that of diamond; the longest two chains we have detected HC15N H17N and C have low frequency bending vibrations which lie within the range of existing high altitude radio telescopes. Finally it is pointed out that the density of the chains at the limit of detection in our spectrometer ca. 108 cm~3 is high by the standards of modern laser spectroscopy so that optical detection of many should be possible.At least 114 molecules have now been identi–ed in the interstellar gas or in circumstellar shells. Many of the smaller ones are inorganic compounds such as H2 H OH and NH 2O 3 but in space as on Earth Nature employs the carbon bond when she chooses to assemble large structures all but 3 of the 64 interstellar molecules with more than three atoms are organic in the strict sense of the term. The largest is the linear carbon chain HC11N with a molecular weight of 147 twice that of the simplest amino acid glycine. Most of the smaller polyatomic molecules are also carbon chains the great majority acetylenic in structure with alternating single and triple CC bonds but a few cumulenic with successive CC double bonds.It is astonishing that highly unsaturated carbon of this kind should be so conspicuous in regions where hydrogen by a very large factor is the most abundant chemically active element. With a Fourier-transform microwave (FTM) spectrometer,1 we have recently been able to identify in the laboratory a large number of carbon chain molecules of astro- 121 New carbon chains in laboratory and interstellar space 122 physical interest. The total found over the last two years stands at 27 11 closed-shell polyynes (three larger than HC N) 9 radicals and 7 carbenes. Three of the carbenes were proposed as possible astronomical molecules at the 1992 Faraday Symposium on 11 Chemistry in the Interstellar Medium,2 and –ve are unusual structures containing threemember carbon rings.Fig. 1 shows the entire set and contains references for those already published ; nearly one half are new and are reported here for the –rst time. All or nearly all of the 27 probably exist in astronomical sources at some level of abundance and the four marked in Fig. 1 with asterisks have already been detected with Fig. 1 Twenty seven new carbon chains. References to published papers are given after or below the chemical formula with astronomical detections underlined. 123 P. T haddeus et al. Fig. 1 Continued radio telescopes either in the rich molecular cloud TMC-1 or in the circumstellar shell of the nearby carbon star IRC]10216 the two best carbon-chain sources so far discovered. Seven of the 13 molecules not previously reported contain nitrogen the two isoand and the two ring chains HC 4NC HC6NC 4N H6N.and 3C12N cyanopolyynes HC and H the three methylcyanopolyynes H3C8N H3C10N C The ease with which the isocyanopolyynes were made suggests that it might be possible to insert nitrogen more or less at random into the chains and ring-chains in Fig. 1 to produce a large number of 124 New carbon chains in laboratory and interstellar space isomers but we have deferred a study of this problem until ab initio calculations are available to guide an experimental search. Insertion of nitrogen is one way in which hypothetical non-polar monocyclic rings of carbon20 might be made sufficiently polar to detect by means of FTM spectroscopy. By analogy with SiC4 ,21 there may be chains and ring-chains containing silicon (or other elements) which might be detectable by the present techniques.Because of limited space it is not possible to give here a detailed spectroscopic description of the unpublished new chains or much in the way of experimental details. We will instead make some general observations about the whole set and will discuss the prospects for further laboratory and astronomical detections and identi–cations. T Spectroscopic characterization The radio spectroscopy of all the molecules in Fig. 1 is complete in the sense that all the astronomically interesting rotational transitions have either been measured to high precision or are readily calculated to comparable accuracy from the spectroscopic constants derived from the laboratory data.In addition to the rotational constants these include in all cases the leading one or two centrifugal distortion constants and for the carbon-chain radicals like C8H the several constants which determine the hfs and lambda doubling caused by the unpaired electron. 2% ladder when n is even and the 2% ladder when 3@2 C is odd. As a result A cannot be determined on the basis of data from our FTM 1@2 7H C8H and however we were able to detect both rotational is ca. 100 K and rot The short carbon chain radicals CCH and C4H have 2& electronic ground states but all longer ones including those here have 2% states with inverted –ne structure and a negative –ne structure constant A when the number of carbon atoms n is even but normal –ne structure and A positive when n is odd.At the low rotational temperature C of our molecular beam only the lowest lying of the two –ne structure ladders in the C ground state is populated the nC spectrometer alone. For C ladders in our free-space millimetre-wave spectrometer where since A is readily found when the rotational constants in both ladders are known [via the relation B1@2 3@2\B0(1^B0/A)] we were able to obtain accurate values of A for these two chain radicals. All the transitions of the new chains which are at all likely to be detected in space in the foreseeable future are now known to at least 1 km sv1 in equivalent radial velocity (i.e. 3 ppm) and often much better than that since lines have generally been measured and –t to 0.3 ppm or better.A detailed account of the data and the spectroscopic analysis for the molecules already reported can be found in the references cited in Fig. 1. For the new molecules here a full account of experimental and spectroscopic details will be given elsewhere. 2Cn a remarkably high polarity for exhibit a Dipole moments Unsymmetrical carbon chains like those here are extremely polar molecules which is one of the reasons why they are readily detected in space and in the laboratory. Dipole moments have not been measured for any of the chains in Fig. 1 but ab initio calculations which are probably accurate to a few percent have been done for all the cumulene carbenes,22,23 the shorter acetylenic radicals,24 and the shorter cyanopolyynes.23 In all three families the dipole moment is found to increase steadily with chain are the most polar with a moment of 6 D for 2C5 H2C10 rising to over 10 D for the still undetected length.10 The cumulene carbenes H H hydrocarbons.The acetylenic radicals HC and the cyanopolyynes HC n nN similar increase in polarity with length but are less polar by ca. 20% and 40% respec-125 P. T haddeus et al. tively. It will be interesting to see from further ab initio calculations if this monotonic increase in polarity with length extends to chains as long as HC17N. Sensitivity far from fundamental limits Our FTM spectrometer currently works in the microwave band from ca. 6»26 GHz. A pulsed supersonic molecular beam of an organic vapour such as acetylene diacetylene or cyanoacetylene heavily diluted in either Ne or A is produced by a commercial solenoid valve (Fig.2). As the beam expands through one of the mirrors of the large confocal Fabry»Perot cavity typically a 1000 V discharge is applied synchronously with the gas pulse which is ca. 400 ls long. At a pressure behind the pulsed valve of 2 atm the discharge is stable during the gas pulse and the rotational temperature of all molecules including any made in the discharge has dropped to only a few K by the time the molecules have moved only a few cm downstream from the nozzle. This low rotational temperature with the concomitant reduction of the rotational partition function by two orders of magnitude for a linear molecule is one of the main reasons the present technique is so sensitive for carbon chains.As the molecules traverse the high-Q Fabry»Perot cavity they are irradiated by a short (1 ms) microwave pulse resonant with one of their rotational transitions to which Fig. 2 Schematic diagram of the FTM spectrometer 126 New carbon chains in laboratory and interstellar space one of the Fabry»Perot modes has been mechanically tuned. The subsequent freeinduction decay is detected with a sensitive microwave receiver which like the mirrors of the resonator has been cooled to the temperature of liquid nitrogen to reduce thermal and ampli–er noise. The Fourier transform of the decaying free induction yields the power spectrum shown in Fig. 3 for one of the rotational transitions of C7H2 . The line shape is double peaked the result of the Doppler shift of the two travelling waves that compose the standing Fabry»Perot mode with respect to the typically Mach 2 molecular beam.This peculiar line shape however is only a minor inconvenience; in practice the resolution of line structure (e.g. hfs or K-doubling as shown in Fig. 4) is usually limited by the small width of the two Doppler components whose fractional width *l/l may be as small as 7]10~7 (0.2 km s~1). Fig. 3 Lower rotational levels of C7H2 showing the transitions detected. The double-peaked line pro–le is instrumental in origin the Doppler splitting that results when the Mach 2 axial molecular beam interacts with the standing wave in the confocal Fabry»Perot cavity (see text). The spectrum shown above is the result of a 2 min integration.P. T haddeus et al. (T Fig. 4 Typical rotational transitions of the two new methylpolyynes showing the resolution of K-structure. The K\1 ladder although lying ca. 8 K above the K\0 ladder is metastable owing to spin symmetry and therefore well populated in the rotationally cold molecular beam rot\1»3 K). The largest member of each family of chains in Fig. 1 is close to the limit of detectability of the present liquid-nitrogen-cooled FTM spectrometer but the sensitivity of this instrument is far from fundamental limits. Liquid-helium cooling of the optics and the –rst stage of receiver ampli–cation might improve the sensitivity by nearly an order of magnitude and a further improvement by a factor of ca. –ve might be gained for molecules with 10 or more heavy atoms by lowering the frequency of operation to the low-GHz range where lines are strongest at the rotational temperature of 1»3 K characteristic of our supersonic molecular beam.While technically challenging both improvements are quite feasible particularly if a closed resonant cavity is used instead of the large open Fabry»Perot resonator of our present device. Another factor favours the 127 128 New carbon chains in laboratory and interstellar space Fig. 5 Relative intensities of the strongest rotational lines of the cyanopolyynes (Ö) and relative abundances (L) from the FTM spectrometer as a function of the number of carbon atoms in the chain. Error bars are estimated 2p uncertainties. The indicated detection limit is that achieved in an integration of ca.3 h. detection of large molecules with our discharge source the line intensity and abundance decrement of the cyanopolyynes markedly —atten beyond about nine carbon atoms as Fig. 5 shows; HC17N for example turned out to be much easier to detect than was expected by extrapolation from the short members of the series.18 If a similar eÜect exists for the other types of chains in Fig. 1 and liquid-helium cooling can be exploited much larger chains than those here may be within reach. Production of carbon chains All the carbon chains in Fig. 1 were produced when simpler unsaturated molecules such as acetylene and diacetylene heavily diluted in an inert gas were subject to a small dc discharge in the throat of the supersonic nozzle of our molecular beam spectrometer.The choice of the precursor and inert gases the precise dilution factor and the parameters of the electrical discharge are all highly empirical guided largely by trial and error since no adequate theoretical model (or understanding) of the synthetic process exists. In the past acetylene and allene were shown in this laboratory and others25 to be a good source of carbon chains and the C3H2 ring in the large glow discharges used in millimetre-wave free-space spectrometers and these were employed in the present FTM instrument with some success. Better results were achieved however with diacetylene which to avoid explosions was synthesized in small quantities and immediately diluted in Ne or A.Cyanoacetylene was also synthesized and proved to be a good source of nitrogen-containing chains. Further details on the particular sources used for each chain can be found in the articles cited in Fig. 1 or for the previously unreported molecules will appear elsewhere. The point which deserves emphasis here is that our production techniques are still primitive largely hit or miss and with systematic investigation may be subject to considerable improvement. Only a few have been tried among the large number of possible production schemes and it is likely that ones we have not yet employed such as laser ablation or discharges through even more unstable precursor gases will yield signi–- cantly higher concentrations of carbon chains than any of the methods so far adopted.A systematic theoretical investigation of the problem is needed but we have no illusions that that would be easy or would quickly yield results to guide experiment. 129 P. T haddeus et al. Astronomical detections were identi–ed in astronomical sources before being detected on Earth but with the Many of the –rst carbon chains (e.g. the sequence of free radicals CCH CCCH… … …C6H) work here the laboratory spectroscopy is now well ahead of the radio astronomy allowing new astronomical molecules to be found without searches in frequency that with large radio telescopes can be prohibitive in time and cost. Any carbon chain which can be observed in space can probably be detected in the laboratory with the present techniques or extensions of these which are planned.All or most of the chains in Fig. 1 are candidates for astronomical detection because at least one shorter member of each type has already been detected in at least one source and as mentioned four of the present chains were quickly found once laboratory frequencies were in hand. With existing telescopes detecting the larger ones will be difficult or impossible in even the best presently known astronomical sources (integrations of roughly 30 h per line were required to detect HC with modest signal-to-noise14) but larger and better telescopes (e.g. the Green Bank 100 m telescope and the resurfaced 11N Arecibo 305 m telescope) will soon be available to work in the band from 1»10 GHz where the larger molecules in Fig. 1 are expected to have their strongest astronomical lines.There is moreover the possibility that signi–cantly better astronomical sources of large molecules will be found. Molecule-rich circumstellar shells of stars comparable in distance to IRC]10216 are rare and it is quite likely that this object is the best of its kind. TMC-1 is an entirely diÜerent matter; it is merely one rather nondescript molecular clump or ìcoreœ among many in the extensive collection of dark nebulae in Taurus covering some 150 square degrees which at 140 pc is one of the closest regions of molecular gas and star formation to the sun. Very few of the many other molecular cores in Taurus have been studied in any molecule heavier than HC3N and there are comparably extensive molecular clouds elsewhere in the sky which have been even less studied although not much further away than those in Taurus e.g.the well known molecular complexes associated with the dark nebulae in Auriga Perseus Aquila and Ophiuchus. More distant molecular clouds are almost entirely terra incognita with respect to large molecules. Carbon monoxide the most readily observed interstellar molecule has been detected in over one-half of the 3600 square degrees of the sky that lies within 5° of the Galactic plane,26 but almost none of the hundreds or thousands of dense generally distant molecular clouds which lie in this wide band have been studied in any molecule larger than HC3N. Once a few good sources were found the radio astronomers simply returned to these locations again and again in their search for still larger molecules unwilling to devote scarce telescope time to slow and frustrating attempts to –nd still better locations.A sensitive new 115 GHz CO survey of the Taurus Dark Clouds is currently being done by Megeath with the CfA 1.2 m telescope at full angular resolution (8.7@) to update the original low-resolution (30@) survey done with this instrument over a decade ago.27 In an attempt to –nd a better source of large molecules than TMC-1 the most promising CO cores will be studied at higher resolution with larger instruments –rst in HC3N then in longer carbon chains. Finally the most promising locations may be observed with the Arecibo telescope which because of its very large collecting area and high resolution at frequencies below 7 GHz may be the most powerful instrument with which to detect the longer carbon chains in Fig.1. Isomers With increasing number of atoms the number of low lying isomers of a molecule which one might hope to detect in the laboratory and in space tends to increase very rapidly. 130 New carbon chains in laboratory and interstellar space Fig. 6 Some possible low-lying isomers of C7H2 ; –lled circles are C open circles H. Energies and dipole moments are from an ab initio calculation.25 The C ring-chain has been calculated ab initio to be the most stable isomer with that 7H2 elemental formula but –ve others have been calculated to lie within 1 eV,28 and there are still others which may be comparably stable or only slightly less so (Fig. 6).Such calculations to our knowledge have not been done for C9H2 or the longer carbon chains in Fig. 1 but by analogy with C7H2 these must all possess a large number of low-lying isomers many of which are probably quite stable and candidates for laboratory and astronomical detection. The existence of fairly energetic molecular isomers in space in regions where the kinetic and rotational temperature is only ca. 10 K is striking evidence of how far the chemical processes there depart from thermal equilibrium. HNC for example is more energetic than HCN by 0.6 eV and in thermal equilibrium at the temperature of a typical cold molecular cloud say 15 K the abundance of HNC relative to HCN is entirely negligible exp(E/kT )B10~200. The observed ratio sometimes in fact approaches unity because the ion»molecule reactions which make both isomers couple the chemistry to the very high eÜective temperature reservoir of the cosmic rays which permeate the interstellar gas.Ionized and radical ring-chains main source of the widely distributed cyclic C The C3H3 ` ion is familiar in the laboratory as the mass 39 peak in mass spectrometers and it is almost certainly present in the interstellar gas where it is thought to be the 3H2 carbene via dissociative recombi-131 P. T haddeus et al. Fig. 7 Hypothetical radical and ionic ring-chains nation. Owing to its planarity and D3h symmetry however it possesses no electric dipole moment and no microwave spectrum (although C3H2D` has a weak rotational spectrum which we can probably detect in the laboratory).On substituting carbon chains for one of the H atoms however the symmetry is broken and one obtains as shown in Fig. 7 a family of ring-chain ions analogous to the carbene ring-chains of Fig. 1. Because these too are likely to be highly polar and because positive ions are readily formed in the interstellar gas this so-far-unobserved sequence is of considerable astronomical interest. Laboratory detection is again a prerequisite for an astronomical search but there does not appear to be any fundamental reason why this cannot be accomplished with an FTM instrument such as ours. known astronomical molecule which in the laboratory produces very strong lines in The same substitution as Fig. 7 also shows can be done with the radical c-C3H a our spectrometer.By analogy with c-C3H such radical ring-chains are expected to be planar with 2B electronic ground states and the unpaired electron in each case is expected to produce well resolved –ne and hyper–ne structure. The –rst member of this 2 hypothetical sequence C5H has been calculated by Stanton and Crawford (personal communication) to lie only ca. 5 kcal above the linear isomer a molecule which has been found in space and which also yields very strong lines in our spectrometer. Carbon chains as classical elastic rods Centrifugal distortion in long linear molecules like those here has not been systematically studied in the past. It turns out to be remarkably simple and readily understood in terms of a simple semi-classical theory.In Fig. 8 the ratio of the centrifugal distortion constant to the rotational constant is shown on a log»log plot as a function of chain length for a number of the molecules in Fig. 1. The –t in Fig. 8 is a one parameter linear –t only the vertical displacement of the line has been varied to best match the data; the slope has been set at exactly [4. To understand this simple L~4 dependence on length let us suppose that a carbon chain is a uniform thin elastic rod of cross-sectional area p and density per unit length o with a Youngœs modulus E that is independent of length and independent of the type of chain. Without centrifugal distortion the moment of inertia is I\oL 3/12 and it is 132 New carbon chains in laboratory and interstellar space Fig. 8 D/B for a number of carbon chains as a function of chain length L .(Ö) Radicals (HCn) ()) cumulenes and (H2Cn) (>) cyanopolyynes (HC2n`1N). readily shown from the theory of elasticity29 that the fractional increase in I produced by centrifugal distortion is *I/I\(144/5Epo)J2L~4. To –rst order this will lower the energy of rotation J2/2I by an amount (864/5Epo2)J4L~7. Again from the theory of elasticity,29 it is readily calculated that exactly one-half of this energy is stored in the elastic deformation of the rod. When both angular momentum and energy are conserved e.g. if the molecule is held rigid spun up and then allowed to centrifugally distort the rest is lost as vibration. Taking the quantum limit J2](h/2n)2J(J]1) it is then found that (1) E\(3h2/2n2oL 3)J(J]1)[(27h4/5n4Epo2L 7)J2(J]1)2 which if written in terms of the rotational and centrifugal distortion constants E\BJ(J]1)[DJ2(J]1)2 immediately yields (2) D/B\(18h2/5n2Epo)L ~4 exhibiting the observed dependence on the inverse fourth power of the length.It is remarkable that this expression applies over such a large range of length and with a single value of the Youngœs modulus E holds for all the chains we have studied. One might expect it to hold only in the limit of long chains or one might expect that the stiÜness would depend on the type of chain i.e. the various end groups and valence structures in Fig. 1 but that does not appear to be the case to any appreciable extent. Instead to an accuracy of a few per cent the model seems to hold to the very shortest chains studied and to –t all types of chain equally well.A full quantum mechanical treatment would presumably show that D/B depends on the full spectrum of vibrational frequencies which undoubtedly depends on chain type but evidently not strongly enough to make much of a diÜerence. The practical value of eqn. (1) is considerable when it comes to identifying new chains. It immediately allows a bona –de chain to be distinguished from weakly bonded complexes which are readily formed and observed in molecular beams such as ours (e.g. HCCCCH… … …H2O) and by making a speci–c prediction of the centrifugal distortion 133 P. T haddeus et al. once the rotational constant is determined it simpli–es signi–cantly the process of identi –cation.It is interesting to see just how stiÜ carbon chains are relative to familiar substances. To do that we must assign to them a cross sectional area p. We adopt the reasonable value of 5.6]10~16 cm2 obtained on the assumption of a hypothetical bulk density equal to that of graphite (unfortunately not a number likely to be measured because of the tendency of carbon chains to explosive polymerization !). Assuming an average bond length of 1.3 ” we then obtain from the empirical –t log(D/B)\[3.98[4 log L for the Youngœs modulus E\1.7]1013 dyn cm~2 which is ca. 8 times that of steel and 1.5 times that of diamond. Carbon chains are evidently stiÜ indeed. Low-frequency bends at radio frequencies 15N H3N should be 16 times less than that ofCwhich is at 220 and should lie at ca.11 cm~1 or 330 GHz. Long carbon chains may be difficult to stretch but they are easy to bend and the lowest frequency bends of the two longest chains in Fig. 1 HC15N HC17N and are expected to lie in the submillimetre band where they are within range of existing high-frequency radio telescopes. A fairly good estimate of the frequencies is provided by a simple classical calculation. The bending modes of a thin elastic rod scale in frequency as the inverse square of the rodœs length,29 so one expects the more ì classical œ of the transverse vibrations of a long chain the lowest-frequency bend especially to scale with length in approximately that way. The low-frequency bends measured for HCN30 and HC3N,31 and those calculated ab initio for HC5 and N HC7 N,32,33 –t this scaling law to ca.15%. Extended to the chains here such scaling predicts that the frequency of the lowestfrequency bend of HC cm~1. It should therefore lie at ca. 14 cm~1 or 420 GHz. Similarly that of HC should be 20 times less than that of HC 17N Both these frequencies lie comfortably within the range of existing radio telescopes 3N 2O O2 line at 325 GHz and an even sharper line at 425 GHz but for example the Caltech submillimetre telescope and the UK James Clark Maxwell telescope on Mauna Kea where the atmosphere is fairly transparent at both frequencies (except for a sharp H these contaminate only a fraction of the range of uncertainty of either bend and so may prove no hindrance to astronomical detection).Precise laboratory measurement of these vibrational transitions is a prerequisite for an astronomical search because the estimated frequencies are probably uncertain to ca. 15%. The transition dipole moments for these low frequency bends are even more uncertain than the frequencies. Estimates by Andrew Cooksy diÜer by an order of magnitude from ca. 1.3 to 0.1 D the uncertainty resulting from the largely unknown distribution of the dipole moment along the long conjugated chain. Ab initio calculations of the structures of HC15N HC17N and might yield considerably better transition moments and it would be desirable to have these in hand before attempting to observe the undoubtedly weak bends in astronomical sources. Weakly bonded molecules The FTM technique applied to supersonic molecular beams has been a very powerful tool for the study of van der Waals complexes which are readily formed at the characteristically low kinetic and rotational temperature of the beam.We have identi–ed one such complex without yet arriving at a certain identi–cation. It may be a stable chain such as diacetylene weakly attached to a second. There are probably many others which we could detect but which have not been pursued because of the limited astrophysical interest in such systems. No van der Waals complex has been detected in space and it is difficult in the almost total absence of three-body collisions there to imagine how they might be formed at a rate adequate to maintain a detectable population. 134 New carbon chains in laboratory and interstellar space Optical detection Carbon chains because of their conjugated electrons are similar to organic dye molecules and similarly they are expected to possess strong electronic optical transitions which move to the red as the chain length is increased.All or nearly all of the carbon chains in Fig. 1 therefore are likely to possess intense optical transitions but with the exception of C8H C10H and (J. Maier personal communication) none of these has been observed in the gas phase. At the densities which we have been able to produce almost all are probably detectable with modern laser techniques laser-induced —uorescence (LIF) spectroscopy for example for those transitions which —uoresce and resonanceenhanced multiphoton ionization (REMPI) spectroscopy for those which do not.Carbon chains such as HC17N at the limit of detection of the present FTM spectrometer (at an integration time of ca. 1 h) have a density of ca. 108 cm~3 in our molecular beam which is well above the threshold of detection by either laser technique. Aside from the interest to radio astronomy a major motivation of the present work has been to learn how to produce and to identify large carbon chains preparatory to optical experiments designed identify the carriers of the diÜuse interstellar bands. Note added in proof Five additional carbon chains have now been detected. Three of these are the free rad- CCCCH and CCCN respectively for the terminal CN group of H icals H2C5H H2C7H and H2C6N obtained by substituting the linear groups CCH 2C4N in Fig.1. The fourth is a bent carbene isomer of C5H2 the analog of isomer 7 in Fig. 6 with a CCH group substituted for the CCCCH acetylenic group. The –fth is the long methylpolyyne H3C13H. CH Two molecules here the cyanopropynyl radical H2C4N and the methylcyanopolyyne 3C7N were –rst detected by one of us (W.C.) working at Wesleyan University in the laboratory of Prof. Stewart Novick; his support and collaboration are gratefully acknowledged as are contributions by J-U. Grabow. We are indebted to W. Klemperer for a number of useful discussions and to A. Cooksy for his estimates of vibrational transition moments. References 1 T. J. Balle and W. H. Flygare Rev. Sci. Instrum. 1981 52 33. 2 P. Thaddeus C. A.Gottlieb R. Mollaaghababa and J. M. Vrtilek J. Chem. Soc. Faraday T rans. 1993 89 2125. 3 M. C. McCarthy M. J. Travers A. Kovaç cs C. A. Gottlieb and P. Thaddeus Astron. Astrophys. 1996 309 L31. 4 J. Cernicharo and M. Gueç lin Astron. Astrophys. 1996 309 L27. 5 M. J. Travers M. C. McCarthy C. A. Gottlieb and P. Thaddeus Astrophys. J. L ett. 1996 465 L77. 6 M. Gueç lin J. Cernicharo M. J. Travers M. C. McCarthy C. A. Gottlieb P. Thaddeus M. Ohishi S. 7 M. C. McCarthy M. J. Travers P. Kalmus C. A. Gottlieb and P. Thaddeus Astrophys. J. L ett. 1996 8 M. J. Travers M. C. McCarthy P. Kalmus C. A. Gottlieb and P. Thaddeus Astrophys. J. L ett. 1996 9 M. J. Travers M. C. McCarthy P. Kalmus C. A. Gottlieb and P. Thaddeus Astrophys. J. L ett. 1996 10 M. C. McCarthy M.J. Travers A. Kovaç cs W. Chen S. E. Novick C. A. Gottlieb and P. Thaddeus Saito and S. Yamamoto Astron. Astrophys. 1997 317 L1. 467 L125. 469 L65. 472 L61. Science 1997 275 518. 11 M. C. McCarthy M. J. Travers P. Kalmus C. A. Gottlieb and P. Thaddeus Chem. Phys. L ett. 1997 264 252. 135 P. T haddeus et al. 13 M. C. McCarthy M. J. Travers A. Kovaç cs C. A. Gottlieb and P. Thaddeus Astrophys. J. Suppl. 1997 12 W. D. Langer T. Velusamy T. B. H. Kuiper R. Peng M. C. McCarthy M. J. Travers A. Kovaç cs C. A. Gottlieb and P. Thaddeus Astrophys. J. L ett. 1997 480 L63. 113 105. 14 M. B. Bell P. A. Feldman M. J. Travers M. C. McCarthy C. A. Gottlieb and P. Thaddeus Astrophys. J. L ett. 1997 483 L61. 18 M. C. McCarthy J-U. Grabow M. J. Travers W.Chen C. A. Gottlieb and P. Thaddeus Astrophys. J. 15 M. J. Travers M. C. McCarthy C. A. Gottlieb and P. Thaddeus Astrophys. J. L ett. 1997 483 L135. 16 M. C. McCarthy M. J. Travers C. A. Gottlieb and P. Thaddeus Astrophys. J. L ett. 1997 483 L139. 17 W. Chen M. C. McCarthy M. J. Travers E. W. Gottlieb M. R. Munrow S. E. Novick C. A. Gottlieb and P. Thaddeus Astrophys. J. 1998 492 849. L ett. 1998 494 L231. 19 M. C. McCarthy M. J. Travers W. Chen C. A. Gottlieb and P. Thaddeus Astrophys. J. L ett. 1998 498 L89. 20 G. von Helden N. G. Gotts and M. T. Bowers Nature (L ondon) 1993 363 60. 21 M. Ohishi N. Kaifu K. Kawaguchi A. Murakami S. Saito S. Yamamoto S. Ishikawa Y. Fugita Y. Shiratori and W. M. Irvine Astrophys. J. L ett. 1989 345 L83. 22 M. Oswald and P. Botschwina J. Mol. Spectrosc. 1995 169 181. 23 P. Botschwina personal communication 1997. 24 D. E. Woon Chem. Phys. L ett. 1995 244 45. 25 M. Bogey C. Demuynck and J. L. Destombes Chem. Phys. L ett. 1986 125 383. 26 T. M. Dame H. Ungerechts R. S. Cohen E. de Geus I. A. Grenier J. May D. C. Murphy L-A. Nyman and P. Thaddeus Astrophys. J. 1987 322 706 supplemented by extensive new CO survey data. 27 H. Ungerechts and P. Thaddeus Astrophys. J. Suppl. 1987 63 645. 28 K. Aoki and S. Ikuta J. Mol. Struct. (T heochem.) 1994 310 229. 29 L. D. Landau and E. M. Lifshitz T heory of Elasticity Pergamon Press 1959 p. 109. 30 H. C. Allen E. D. Tidwell and E. K. Plyler J. Chem. Phys. 1956 25 302. 31 P. D. Mallinson and A. Fayt Mol. Phys. 1976 32 473. 32 P. Botschwina A. Heyl M. Oswald and T. Hirano Spectrochim. Acta A 1997 53 1079. 33 P. Botschwina M. Horn K. Markey and R. Oswald Mol. Phys. 1997 92 381. Paper 8/00286J; Received 9th January 1998

ISSN:1359-6640    

DOI:10.1039/a800286j

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

Faraday Discuss. 1998 109 137»163 Current status of the H diÜuse interstellar bands theory 2 1 Introduction In a recently published article,1 a new theory was presented attributing the diÜuse optical interstellar bands (DIBs) (Fig. 1) to resonantly enhanced simultaneous twophoton (VUV]VIS) absorption of light from a bright star by H contained in a thin tenuous cloud located relatively near the illuminating star. While it was shown in ref. 1 that the wavelengths of over 70 narrow DIBs could be matched to within the obserconsensus of most astronomers (e.g. ref. 2) remained that these spectral matches were vational limits of error to calculated wavelengths of inter-Rydberg transitions in H2 the simply happenstance coincidences that did not provide any real credibility for the model.Within the past year additional investigations were made with the aim of further testing the viability of the H DIBs model. On the one hand laboratory spectra were acquired [at Vrije Universiteit (VU)] of H inter-Rydberg transitions that a priori would have been expected to correspond to DIBs according to the non-linear H DIBs model of ref. 1. On the other hand important changes in the structure of this model were proposed (at IBM) in an attempt to overcome some of the de–ciencies that had become apparent in the original model. In making such changes we have availed ourselves of a Assignments of the kk5780 6284 4428 4882 large-equivalent-width diÜuse interstellar bands P. P. Sorokin,a§ J. H. Glowniaa§ and W. Ubachsb a IBM Research Division T .J. W atson Research Center Y orktown Heights NY 10598-0218 USA b L aser Centre V rije Universiteit De Boelelaan 1081-1083 1081 HV Amsterdam Netherlands 2 2 2 2 2 Major changes recently incorporated into the structure of the non-linear H DIBs model (P. P. Sorokin and J. H. Glownia Astrophys J. 1996 473 900) are outlined. To account for observed relative intensities of narrow DIBs assigned in the original model to H inter-Rydberg transitions it has become necessary to postulate a form of lightwave coherence existing in the H -containing cloud of the model with numerous coherent 2n wave-mixing processes each initiated by broadband stimulated Raman scattering (SRS) occurring on various H transitions. By considering such wave-mixing processes it becomes possible both to interpret archived VUV spectra of B-type supergiants and to easily assign large-equivalent-width DIBs (e.g.kk5780 6284 4428 4882) whose origins cannot be understood using the original ìpassiveœ H two-photon absorption model. The wave-mixing schemes discussed also enable one to understand why DIB intensities in given lines-ofsight correlate well with the amount of reddening observed in the same lines-of-sight. 2 2 2 2 § E-mail addresses sorokin=watson.ibm.com; glownia=us.ibm.com; wimu=nat.vu.nl 137 138 Current status of the H DIBS theory 2 Fig. 1 ” (DIB) spectrum from 3800»8680 for a line-of-sight of unit reddening. Spectrum synthesized from data contained in Table 3 of ref. 4. variety of published spectral data that were not considered in ref.1. These range from recent high-resolution spectral measurements of the lineshapes of the k5797 and k6379 DIBs to vacuum ultraviolet (VUV) spectra of mid-/late-B supergiants recorded in the 1970s. In this paper we focus primarily on the changes that have been made in the past year in the original H DIBs model and on the reasons why such changes were required. 2 2 The H DIBs model1 2 Postulated in the model of ref. 1 was the existence of a thin H -containing cloud located ca. 0.01»10 pc from a bright early-type star and intersected by our line-of-sight to the 2 star. It was argued that in a collision-free medium (a good approximation for the cold cloud with density ca. 104 cm~3 assumed in the model) elastic (i.e.resonance Rayleigh) scattering of near-resonant VUV starlight should completely dominate inelastic scattering (i.e. ìabsorptionœ leading to real excitation of B- and C-state quantum levels) of the same light. (The theory of resonance —uorescence and light scattering is thoroughly discussed in ch. 8 of ref. 3 for example.) Consequently near-resonant VUV light coming from the nearby hot star diÜuses throughout the entire cloud resulting in the steadystate —ux column densities of this light along the line-of-sight to the star becoming greater than the values that would prevail in the total absence of the cloud. Conservation of energy dictates that ìabsorptionœ of near-resonant VUV photons whose freone-photon transitions. (For simplicity Doppler broadening in the cloud of the model is quencies lie outside the ca.0.003 cm~1 H2 natural linewidth cannot occur via singlet-gerade-state quantum levels involving both near-resonant VUV photons trapped assumed to be zero.) However simultaneous two-photon absorption transitions to H2 by elastic scattering within the cloud and visible photons directly coming from the star can exactly conserve energy and were assumed to occur in the model producing the DIBs. On the basis of the simple scheme described above it became immediately possible to assign several dozen narrow DIBs to H inter-Rydberg transitions. An example is the 2 narrow DIB shown in Fig. 2. According to Table 3 of ref. 4 the frequency of this (]) (i.e. ìcertainœ) DIB is 12 455.7^0.2 cm~1 (8026.21^0.15 ”) its linewidth is 1.2^0.1 cm~1, 139 P.P. Sorokin et al. Fig. 2 Diagram on right shows ref. 1 assignment of narrow DIB observed at ca. 12 456 cm~1. An expanded portion of the spectrum in Fig. 1 containing this DIB is shown on the left. In all synthesized DIB spectra shown in this paper DIBs with wavenumbers indicated were assigned in ref. 1. and its equivalent width (in HD 183143) is 52 m”. (Although the equivalent width is here relatively modest the sharpness of this DIB allows it to be clearly seen in Fig. 1.) The exact frequency of the H inter-Rydberg transition to which this DIB is assigned in Fig. 2 can be calculated from the H quantum level energies tabulated in ref. 5»7. The 2 2 value is 12 455.79 cm~1 well within the narrow error limits given in ref.4. In ref. 8 the frequency of this DIB is listed as 12 455.82^0.06 cm~1. Both the VUV and visible transitions shown in Fig. 2 have very large calculated radiative transition probabilities. For the C1»0R1 transition which originates (as required in the model of ref. 1) from the very lowest quantum level of orthohydrogen (JA\1 of X0) the value calculated9 is 0.14]9 s~1. The strength of the visible transition shown is indicated by its large ìadiabatic transition moment (ATM)œ. (ATM values are calculated and tabulated for a variety of H transitions in ref. 10.) At –rst glance the assignment in Fig. 2 seems quite compelling. However a serious difficulty is present here 2 that also occurs in most of the DIB assignments given in ref. 1. Brie—y stated no DIBs have thus far been observed that can be assigned to many other strong transitions originating from the same resonant intermediate-state quantum level J@\2(]) of C1.For example no DIB is observed at 12 518.54 cm~1 the calculated frequency of the GK37C1 R(2) transition. To us this strongly implied that a mechanism based entirely upon ìpassiveœ two-photon absorption by H molecules cannot adequately explain the origin of the DIBs. As will be explained in detail below we currently assume that some 2 form of coherent lightwave structure is present in the H -containing cloud of our model. In the case of Fig. 2 we would now suggest that broadband IR stimulated emission 2 spectrally centred on a transition connecting the GK3 J\1 quantum level with some lower-lying B- or C-state quantum level must be occurring.We would now additionally postulate that this IR stimulated emission occurs as a step in an existing parametric oscillation cascade that ìcomes from aboveœ i.e. that starts at a higher-lying quantum level. Absorption of ca. 12 456-cm~1 photons would then occur via a process that should 140 Current status of the H DIBS theory 2 probably be termed ìfour-photon inverse Raman absorptionœ a four-photon analogue of the inverse Raman eÜect.11 This mechanism should produce much stronger absorption of light at the DIB frequencies than passive two-photon absorption. Thus in our current view DIB selectivity can be the result of the operation of speci–c enhancement mechanisms. Another DIB which was assigned in ref.1 on the basis of a close frequency match is the narrow one (*lB2 cm~1) appearing in Fig. 3 at ca. 16 135 cm~1. In ref. 4 the frequency of this much studied DIB is given as 16 134.4^0.9 cm~1. In ref. 8 the value listed is 16 135.12^0.05 cm~1. The value calculated for the H transition to which the 2 DIB is assigned in Fig. 3 is 16 135.06 cm~1. In terms of a frequency match with existing observations the assignment is again seen to be a good one. However in trying to account for the observed DIB intensity simply on the basis of a passive two-photon absorption model one again encounters difficulty. The nature of this difficulty can be seen by comparing the two assignments shown in Fig. 2 and 3. For the B3»0P1 transition which also originates from the lowest level of orthohydrogen the calculated radiative transition probability12 is 0.10]9 s~1 about the same as for C1»0R1.However the ATM for the visible transition in Fig. 3 is seen to be more than ten times smaller than that for the visible transition in Fig. 2. Nonetheless the equivalent width of the DIB at 16 135 cm~1 is ca. 80 m” roughly 1.6 times greater than that of the DIB at 12 456 cm~1. ” one of the most widely studied of the DIBs was not actually 2 ”) H of the closest transition i.e. the By discussing together the cases of the strong but narrow k5797 and k6379 DIBs one can again illustrate both the strengths and the weaknesses of the H DIBs model of ref. 1. The DIB at 5797 assigned in ref. 1 since the air wavelength (5796.96 2 one originating from J@\2 of B0 and terminating on J\3 of EF12 appeared to lie outside the k5797 DIB wavelength measurement error limits (5797.11^0.05 ”) listed in ref.4. However subsequent very-high-resolution measurements of the k5797 DIB presented in ref. 13 show this DIB to be ca. 2.4 wavenumbers wide and to be roughly peaked at 5797.0 ” a point at which the DIB pro–le displays a distinct sharp ìnotchœ. Fig. 3 Ref. 1 assignment and expanded spectrum for the DIB at ca. 16 135 cm~1 141 P. P. Sorokin et al. This ìnotchœ is clearly seen in both lines-of-sight (i.e. towards l Sgr and towards f Oph) for which the k5797 DIB pro–le was recorded in ref. 13. The wavelength of this DIB is stated in ref. 8 to be 5796.98^0.04 ”. Thus the more recent k5797 DIB wavelength measurements lend strong support to the H2-based assignment.The k6379 DIB was assigned in ref. 1 to the transition (EF10 J\1)^(B0 J@\2) that is it was assigned to a transition having the same intermediate-state quantum level as in the above mentioned k5797 DIB assignment. The calculated (air) wavelength of this transition is 6379.23 ” within the error limits of ref. 4 (6379.27^0.08 ”) and very nearly within those of Ref. 8 (6379.20^0.02 ”). Some observations were recently made that lend more interest to the fact that our assignments of the k5797 and k6379 DIBs feature the same intermediate-state quantum level. In ref. 14 the pro–le of the sharp k6379 DIB was recorded at very high resolution in the line-of-sight to f Oph. After telluric correction this DIB showed a substructure described by the authors of ref.14 as consisting of two equal peaks. However the k6379 DIB pro–le shown in this work can equally well be described as having a distinct sharp notch near its peak. In a still more recent high-resolution study of DIBs,15 the presence of a clearcut notch at ” ca. 6379.25 on the k6379 DIB was con–rmed. An earlier study16 was evidently the –rst to suggest that the k6379 DIB possessed a two-peak structure. The serious difficulty appearing to discredit the H -based assignments of both of these DIBs is again the lack of an explanation for the observed DIB intensities. Both 2 k5797 and k6379 are among the strongest of the narrow DIBs. (In ref. 8 the equivalent widths in HD 183143 for k5797 and k6379 are listed as 238 and 123 m” respectively.) Yet the strengths of the simultaneous two-photon absorption transitions involved in the DIB assignments are in both cases relatively weak.The B0»0R1 transition is the weakest of all Bn»0R1 transitions lying below the Lyman limit being ca. 40 times weaker than C1»0R1 for example. In the simultaneous two-photon absorption model of ref. 1 the strength of the VUV transition in—uences both the intrinsic strength of the simultaneous two-photon absorption process and the density of near-resonant VUV photons trapped in the cloud by elastic scattering. In addition to the VUV transition being weak the ATMs for the transitions (EF12 J\3)^(B0 J@\2) and (EF10 J\1)^(B0 J@\2) are also comparatively modest being 0.04 and 0.15 respectively. The processes illustrated in Fig.4 represent a possible solution to the intensity anomaly problem discussed above. However several unusual features requiring considerable explanation are involved. To begin with as will be noted below there exists fairly compelling spectral evidence that broad-band coherent VUV Stokes-wave emission occurs on the transition B0»0P3 in H -containing clouds near B-type giants and 2 supergiants. This radiation would raise orthohydrogen molecules in JA\3 of X0 to ì virtual states œ lying near J@\2 of B0. From these virtual states the molecules could be induced to follow pathways similar to the ones shown in Fig. 4. In Fig. 4 it is assumed that both quantum levels EF12 J\3 and EF10 J\1 are already involved in parametric oscillation cascades proceeding downward from higher levels.[For simplicity in Fig. 4 we have speci–cally assumed a common IR cascade for the two levels proceeding through J@\2 of B15. If the transition from EF12 J\3 to EF10 J\1 happened to occur as a degenerate two-photon emission process the broad-band IR generated would be spectrally centred at 787 cm~1 the frequency of one of the ìunidenti–ed IR emission bands (UIBs)œ discovered in ref. 17.] With strong broad-band light waves present that are spectrally centred on the topmost two IR transitions in Fig. 4 and with strong broad-band VUV light around B0»0P3 also present one expects that light around 5797 ” will be absorbed through the same ìfour-photon inverse Raman absorptionœ process already brie—y mentioned above.A similar process would produce the k6379 DIB. From the unspecifed singlet-ungerade-state quantum level shown in Fig. 4 the downward parametric oscillation cascade would continue in some manner eventually returning the molecules to JA\3 of X0. 142 Current status of the H DIBS theory 2 Fig. 4 Wavemixing processes incorporating inverse Raman absorption steps proposed in text to explain selective enhancement of k5797 and k6379 DIB intensities On the basis of the scheme shown in Fig. 4 one can thus account for the relatively strong k5797 and k6379 DIB intensities that are observed for the fact that the intensities of these DIBs are strongly correlated with reddening in the line-of-sight (explanation to be given below) and for the fact that there here again must be some mechanism that restricts the number of DIBs.No DIBs are observed that correspond to the frequencies of the transitions (EF12 J\1)^(B0 J@\2) or (EF10 J\3)^(B0 J@\2) for example. In the present instance it is again seen that selectivity results from the presence of enhancement mechanisms. In global terms the model outlined in ref. 1 is far more inadequate than has thus far been represented. We have described the severe intensity anomalies present in some of the assignments of narrow DIBs. However on the basis of the model of ref. 1 most of the prominent large-equivalent-width DIBs appearing in Fig. 1 cannot be assigned at all. Several of the DIBs have very broad linewidths of the order of 50 wavenumbers or more. In ref. 1 only speculative arguments were oÜered to explain the origin of DIBs with such large linewidths.It was suggested that the cause of this width might be found in the rapid rates of predissociation and/or autoionization of singlet-gerade-state quantum levels lying higher in energy than 118 375.6 cm~1 the threshold for dissociation into H(1s)]H(2s 2p). The various double-resonance measurements described in the next section have generally dispelled both this notion and several others that were contained in ref. 1. 3 H double resonance measurements 2 In ref. 18 double resonance (VUV]VIS) ionization spectra of H were reported which showed a number of coincidences (i.e. frequency and linewidth matches) with modest- 2 equivalent-width DIBs (Fig. 5). The matching DIBs have only been reported in ref.4 however. Both C5 and C6 quantum levels were excited from X0 with the VUV laser 143 P. P. Sorokin et al. Fig. 5 Double-resonance ionization spectra probing singlet-gerade-state quantum levels lying above the ionization threshold. The VUV laser was tuned to C5 intermediate states (left) and C6 intermediate states (right). The bottom spectra show expanded portions of the DIB spectrum of Fig. 1 corresponding to the same wavelength ranges. separately tuned to each of the R(0) R(1) and Q(1) transitions. With the VUV laser frequency set to coincide with a given absorption line a tunable visible laser was then scanned in frequency and a photoionization spectrum was recorded. Singlet-geradestate quantum levels lying in a region just above the H2 autoionization threshold (124 418.4 cm~1) were probed.Somewhat remarkably it was found that in a spectral range spanning ca. 1000 cm~1 in an energy region where the H density-of-states might a priori be expected to be high only the strong resonances shown in Fig. 5 were seen. Four of these transitions termin- 2 ate on a singlet-gerade-state quantum level of orthohydrogen (con–guration unknown) located at 124 752.5 cm~1. The other two both terminate on a quantum level of parahydrogen (presumably a diÜerent rotational level belonging to the same con–guration as above) at 124 701.9 cm~1. The linewidths of all six resonances and of the –ve DIBs which seemed to match were all between 2 and 3 wavenumbers. These widths undoubtedly re—ect the rates of autoionization of these two quantum levels.One can attempt to discuss the resonances appearing in Fig. 5 in terms of the basic model of ref. 1. To begin with all three C5»0 transitions [i.e. R(0) R(1) and Q(1)] lie below the Lyman limit (91.2 nm). These transitions should therefore be potentially excitable in the H2-containing cloud assumed in the model of ref. 1. The equivalent widths (in HD 183143) of the two DIBs which seem to correspond to the top and bottom C5 resonances in Fig. 5 are stated in ref. 4 to be 24 and 12 m” respectively. Since the ratio of these equivalent widths is about the same as the ratio (1.82) of the C5»0Q1 and C5»0R1 transition probabilities it would seem at –rst glance that this particular pair of DIB assignments can be more-or-less understood on the basis of the original ìpassiveœ two-photon absorption model of ref.1. The middle C5 resonance is much too far oÜ resonance (ca. 9 cm~1 diÜerence) from the closest DIB appearing in Fig. 5 to be even remotely considered a frequency match. This DIB which has a frequency of 15 331.5 cm~1 and an equivalent width of 30 m” was reasonably assigned in ref. 1 as the R(2) transition of EF137B2. However one is now required to explain the absence of an observed DIB corresponding to the middle C5 resonance. At this point, 144 Current status of the H DIBS theory 2 from ref. 21.) Fig. 6 Double-resonance ionization spectrum probing singlet-gerade-state quantum levels lying below the ionization threshold. The VUV laser was here tuned to the C2»0R1 line.(Spectral data we can only postulate that broad-band stimulated Raman scattering (SRS) involving C5 J@\1([) as resonant intermediate state must be occurring in the H -containing clouds in which the DIBs originate. According to ref. 19 when the intermediate quantum level 2 in a resonant two-photon absorption process is radiatively coupled by a light wave to a fourth quantum level the probability of simultaneous two-photon absorption occurring from the ground level to the two-photon level may be reduced by many orders of magnitude. This eÜect is a two-photon analogue of ìelectromagnetically induced transparency (EIT)œ a topic of current theoretical interest. (See e.g. ref. 20.) It is thus seen that with the assumption of a coherent lightwave structure present in the H2-containing cloud there is the potential for both DIB-enhancing and DIB-attenuating mechanisms to exist.It is more awkward to try to explain the three apparent coincidences of DIBs with transitions involving C6 seen in Fig. 5. [The narrow DIB at ca. 13 718 cm~1 was assigned in ref. 1 to the strong (ATM\0.95) transition HH7C2 Q(2).] All three C6»0 transitions lie at shorter wavelengths than the VUV cut-oÜ imposed by the Lyman limit. We speculate that virtual states near the C6 levels are excited by VUV light existing in the cloud at certain wavelengths longer than 91.2 nm (vide infra). This VUV light promotes transitions of molecules existing in vibrationally excited levels of the X state. The latter are populated in primary VUV SRS processes of the type to be discussed in detail below.From the strong resonances seen in Fig. 5 one would guess that the potential curve for the unknown singlet gerade state is similar in shape to the C-state potential curve only vertically shifted from the latter to higher energies. This in turn would imply that strong transitions between the same two states could in principle occur from lower C-state vibronic levels to correspondingly lower vibronic levels of the singlet gerade state. However the energies of these lower singlet-gerade-state vibronic levels would lie below the threshold for H autoionization. One would therefore not expect signals to be generated at resonances using the experimental arrangement employed in ref. 18. 2 However by modifying the technique used in ref.18 to include an extra pulse for ionizing H(2s,2p) atoms produced in H(1s)]H(2s,2p) dissociation channels the group at VU has now shown21 that transitions from C-state vibronic levels to singlet-gerade-state 145 P. P. Sorokin et al. vibronic levels lying between the energy thresholds for dissociation into H(1s)]H(2s,2p) and autoionization into H2 `]e~ can easily be observed. (In unpublished earlier work,22 several H resonances in this same range were observed with use of a similar technique.) 2 Fig. 6 shows an example of an ionization spectrum obtained in this manner.21 Although striking resonances are seen here it turns out that there are no correspondences with DIBs. One can again reasonably propose that SRS involving C2 J@\2(]) as resonant intermediate state might be occurring and that consequently (again by the main result derived in ref.19) the intensity of two-photon absorption from JA\1 of X0 to any of the singlet-gerade-state quantum levels located by the resonances in Fig. 6 will be greatly diminished. However considering all the double-resonance experiments which have here been described one senses the futility of attempting to explain the origins of large-equivalent-width DIBs (especially the very broad ones) on the basis of rapidly dissociating and/or autoionizing singlet-gerade-state quantum levels existing in heretofore unexplored energy ranges. In the remainder of this paper we outline a promising new approach we have recently been following in seeking to provide an H -based answer to this question.2 4 Evidence for nonlinear generation of VUV light in H -containing clouds 2 near B-type supergiants If mechanisms involving coherent light generation are required to explain the observed DIB relative intensities it becomes a reasonable investigative step to examine archived VUV spectra of bright stars for some evidence of VUV light generation actually occurring in these stars. In a manuscript recently submitted for publication,23 we indicate that it is indeed possible to interpret some of the spectral data obtained by various VUV space probes (e.g. Copernicus and ORFEUS-124) to show that in clouds surrounding some B-type stars both SRS and a related process known as four-wave parametric oscillation (FWPO) ubiquitously occur on H transitions in a manner closely resembling some of the scenarios that were brie—y considered in ref.1 and that these 2 processes can actually represent ìmain eventsœ determining the spectral character of these objects in the VUV. We began outlining this evidence in ref. 23 by showing that for some stars located high above the galactic plane (ìhalo stars œ) whose lines-of-sight would thus be less likely to be intersected by interstellar H clouds that are far removed from the stars themselves the observed intensity ratios of individual Bn^X0 R(0) R(1) and P(1) transitions 2 cannot be comprehended on the basis of linear spectroscopy. This can be seen for example in the ORFEUS-1 spectrum of the B1 Ib halo star (d\1.7 kpc) HD 214080 (Fig. 7 and 8).Comparing these two –gures one observes that the B0»0 absorption components are clearly stronger than those of B4»0. Yet the calculated radiative transition probabilities12 of the latter are ca. 16 times those of the former. From this clear anomaly which is also apparent in the VUV spectra of several other B-type stars that were probed by ORFEUS-1 we infer that there exist ubiquitously in our galaxy spectrally accessible clouds of H for which paradigms based upon the Copernicus high- 2 resolution VUV spectrum of f Oph shown in ref. 25 are not relevant. The tentative explanation we oÜer for the strong enhancement of individual R(0) R(1) and P(1) Bn»0 absorption components that is observed for select low n in many B-type stars involves SRS processes occurring on some of these transitions.This will be discussed further in what follows. In ref. 26 Copernicus low-resolution (0.2 ”) observations of 16 O-type stars and 10 B-type stars (mostly supergiants) covering the range 1000 ” »1200 are presented in the form of a classical spectral atlas. In ref. 23 we focussed on what appeared to be a few 146 Current status of the H DIBS theory 2 Fig. 7 Expanded view of a section of the ORFEUS-1 HD 214080 spectrum. Absorptions marked ì s œ are of known stellar origin. Two interstellar N I absorption lines are indicated. dozen emission lines rising well above the background continuum in the mid-/late-B supergiant spectra shown in ref. 26 (Fig. 9). There seemed to exist no speci–c identi–cations or assignments of these apparent emission features in the scienti–c literature.However we have recently learned that it has long been the general opinion of the astrophysical community (see for example ref. 27) that (1) no emission lines are present in spectra such as those shown in Fig. 9; (2) the background VUV continuum level should always be placed at the level of the very highest peaks that are observed; and (3) the resulting large fraction of continuum light which is thus inferred to be missing is marked. Fig. 8 Expanded view of a section of the ORFEUS-1 HD 214080 spectrum adjacent to that shown in Fig. 7. Absorptions marked ì s œ are of known stellar origin. An interstellar O I line is 147 P. P. Sorokin et al. Fig. 9 Copernicus low-resolution VUV spectra of four mid-/late-B supergiants as shown in Fig.6 of ref. 26 Ref. 27 features a high-resolution (0.05 absorbed by over 1000 blended stellar photospheric absorption lines mostly due to ions of transition metals. Despite this interpretation we here proceed to describe some of the interesting consequences that follow if one assumes that emission peaks are indeed present in the spectra of Fig. 9. Fig. 10 shows a wavelength-expanded portion of Fig. 9 containing one of the most intense ìemissionœ peaks appearing in the latter. This peak appears in each spectrum of Fig. 10 at a wavelength that is very close to 1135.9 ”. Making the assumption that this peak must be nearly centred on the wavelength of an allowed transition of H2 and that low rotational quantum numbers must be involved one concludes from inspection of Table 61 of ref.28 that the most likely transition is B8»3R0 at 1135.90 ”. In ref. 23 it was assumed that this emission peak was generated as a Stokes wave in a resonantly enhanced stimulated Raman scattering process. This SRS process would result from broadband optical pumping of ground-state parahydrogen molecules by VUV photons in the H -containing cloud that are nearly resonant with B8»0R0. As has already been explained it is generally assumed in all of our DIB models that enhanced densities of 2 nearly-resonant VUV photons exist throughout the cloud as a result of elastic (i.e. Rayleigh) scattering. ” i.e. 5 cm~1 at ca. 1000 ”) Copernicus scan ” ” . In this scan the width of the 1135.9 peak appears to of b Ori A from 999 to 1561 be 50»60 cm~1.Since Doppler broadening of VUV —uorescent lines emitted by H2 molecules residing in the thin cloud of our model should be at least 10 to 100 times smaller than this one can conclude that if the 1135.9 ” emission band is indeed due to H2 it is generated not through normal —uorescence but through a mechanism such as broad-band SRS or FWPO. Another prominent emission peak in Fig. 10 is seen to occur ubiquitously at ca. 1116 ”. This emission band also appears strongly in all the ORFEUS-1 spectra of B-type giants and supergiants we have examined. (See e.g. Fig. 11.) From Table 58 of ref. 28 we assign this peak to B0»0P3 at 1115.9 ”. We speculated in Section 2 that generation of coherent light on this transition should play an important role in determining the absorption strengths of the k5797 and k6379 DIBs.In Fig. 11 the ORFEUS-1 spectrum of the (B1 Ib) star HD 214080 is compared with the Copernicus spectrum of b Ori A in the same wavelength range shown in Fig. 10. 148 Current status of the H DIBS theory 2 Fig. 10 Expanded view of a section of the Copernicus spectra shown in Fig. 9 While the (emission peak height/ background continuum level) ratio is clearly less in the case of the much hotter ORFEUS-1 star the major peaks are seen to occur at approximately the same wavelengths as in Fig. 10. One becomes more easily convinced of the existence of emission peaks viewing the b Ori A spectrum however. All other assignments of peaks shown in Fig. 11 result from analysing the higher-resolution b Ori A spectrum shown in ref.27 with the same method used above in assigning the 1135.9 ” peak. In Fig. 12 wavelength-expanded portions of two of the spectra appearing in Fig. 9 are shown in another spectral range. One sees again that the major peaks occur at exactly the same wavelengths in these two diÜerent temperature stars. Two of the emission peaks in this –gure (at ” ” ca. 1043.9 and ca. 1065.2 ) are assigned as Stokes waves in SRS processes having the same initial and –nal states as the SRS process proposed above as the generating mechanism for the 1135.9 ” emission peak. However in order for SRS to occur on a continuous basis some eÜective mechanism must exist to depopulate the terminal state and repopulate the ground state.A possible mechanism in this case would be that shown in Fig. 13. The essential idea here is that additional SRS and/or inverse Raman and/or parametric oscillation steps are required in order to deplete continuously the population of molecules pumped to JA\0 of X3 and to provide a return path for this population back to JA\0 of X0. With the incorporation of such additional steps the overall process can no longer be described as being simple broad-band SRS. It must instead be viewed as a type of hybrid wave-mixing process in which SRS pumping absorption resulting from inverse Raman scattering or parametric oscillation can each occur at more than one step. In ref. 23 we employed the term ìSRS»FWPOœ to describe such hybrid processes. The hybrid process may involve four 149 P.P. Sorokin et al. latter spectrum. Fig. 11 Expanded Copernicus b Ori A spectrum of Fig. 10 compared with the ORFEUS-1 spectrum of HD 214080 over the same spectral range. Two interstellar N I lines are indicated in the Fig. 12 Expanded view of a section of the Copernicus spectra of two of the stars shown in Fig. 9. A few emission bands discussed in the text are shown. 150 Current status of the H DIBS theory 2 Fig. 13 An SRS-FWPO wavemixing scheme that can account for continuous SRS Stokes-wave emission on B8»3R0 C4»3R0 and C3»3R0. waves six waves eight waves or more. In Fig. 13 it is assumed that a downward parametric oscillation cascade continues from an unspeci–ed singlet-gerade-state quantum level (the ì s.g. level œ in Fig.13) eventually terminating on JA\0 of X0. All waves participating in SRS»FWPO processes of the type shown in Fig. 13 must be colinear a basic requirement that follows from momentum conservation. In the thin H -containing cloud of our model this implies that all waves participating in SRS» FWPO processes must propagate in the plane of the cloud. This restriction will form the 2 basis for the explanation proposed in the next section for the universally observed dependence of DIB intensity upon reddening in the line-of-sight. In Fig. 13 it is assumed that intense pumping radiation in the vicinity of Ly-a exists in the cloud. A relatively high density of H atoms is assumed to exist in the cloud and elastic scattering of VUV radiation near Ly-a Ly-b . . . etc.by these atoms will result in there being present in the cloud greatly concentrated densities of photons that are nearly resonant with these transitions. Note that owing to the isotropic nature of the elastic scattering process there will always be Ly-a Ly-b Ly-c . . . etc. photons propagating in the plane of the cloud. These photons can therefore participate in SRS»FWPO processes. Let us restate the general hypothesis that has evolved from the discussion of Fig. 13 just presented. If there happens to exist a near-coincidence in frequency between either Ly-a Ly-b or a higher-lying H-atom resonance line and a strong transition to some higher-lying singlet-ungerade-state quantum level from the particular X-state quantum level that is populated by a given Stokes-wave emission an eÜective SRS»-FWPO pathway capable of depleting the X-state quantum level may exist allowing the VUV SRS to occur.On the basis of this general hypothesis it appears to be possible to justify most of the H2-based assignments we have thus far made. Consider for example the emission peak appearing at ” ca. 1126 in Fig. 11 which we have assigned as a broadband SRS Stokes-wave emission occurring on B16»5P1. (The exact wavelength of the B16»5P1 transition is 1126.085 ”.) The primary SRS process would here be pumped by VUV photons nearly resonant with the strong transition B16»0P1. According to our model the densities of such photons everywhere in the cloud should be greatly enhanced 151 P. P. Sorokin et al. Fig. 14 Expanded view of another section of the Copernicus spectra of two of the stars shown in Fig.9. Several emission bands assigned in the text are shown. through elastic scattering. There are several B16»nP1 transitions considerably stronger than B16»5P1 which should have correspondingly higher Stokes-wave gains but to which none of the emission peaks in the spectra of late-/mid-B supergiants actually seem to correspond. A plausible explanation can be oÜered for this unusual selectivity in terms of the general SRS»FWPO scheme here being proposed. If one adds the energy of a Ly-a photon (82 259 cm~1) to the energy29 of the X5 JA\1 quantum level (18 582 cm~1) the result is 100 841 cm~1. The energy of the B9 J@\0 quantum level is 100 843 cm~1 just two wavenumbers away. The B9 J@\0 level could thus eÜectively serve as the resonant intermediate state in a secondary SRS process originating from X5 JA\1.The secondary Stokes wave could be generated either in the IR for example on the (ATM\0.13) transition (B9 J@\0)](EF1 J\1) at ca. 1467 cm~1 or in the VUV on a transition that drops the molecules back to some vibronic level of the X state. Since Stokes-wave gain generally varies as o2l where o is the transition dipole moment and l is the frequency of the Stokes-wave transition VUV generation would a priori be heavily favoured except that once more some mechanism must exist that is capable of depopulating the terminal state of the secondary SRS process. Relaxation of the secondary SRS terminal state is not a problem in the case of IR Stokes-wave generation since EF vibronic levels can decay on a nanosecond timescale to lower-lying B- and C-state quantum levels.We are unable to specify with any accuracy an exact pathway for the SRS»FWPO process in the present instance. However a reasonable speculation would be that the secondary Stokes-wave emission occurs on B9»4P1 at 1169.6 ”. There is an emission peak seen at this wavelength in the spectra of the mid-/late-B supergiants. Moreover when one adds the energy of a Ly-a photon to the energy of the X4 J\1 quantum level (15 346 cm~1) one gets 97 605 cm~1 just 14 wavenumbers away from B6 J@\0 at 97 591 cm~1. The Stokes wave in a tertiary SRS process employing B6 J@\0 as resonant intermediate state would necessarily have to occur in the VUV since there 152 Current status of the H DIBS theory 2 are no EF vibronic levels below B6 J@\0.We speculate that the Stokes-wave emission might here occur on B6»11P1 at 1546.7 ”. This has a large calculated radiative transition probability and an emission peak is seen at this wavelength in ref. 27. Moreover from X11 J\1 a Ly-a photon reaches an energy of 115 197 cm~1 just 25 wavenumbers away from C9 J@\2(]) at 115 222 cm~1. This quaternary SRS process involving orthohydrogen molecules thus perhaps proceeds in a manner generally similar to the tertiary SRS process involving parahydrogen molecules shown in Fig. 13. In Fig. 14 wavelength-expanded portions of two of the spectra shown in Fig. 9 are presented in yet another spectral range. On the basis of the high-resolution b Ori A spectrum contained in ref.27 we have again assigned many of the most prominent peaks shown here. Four of the transitions assigned as Stokes-wave emissions in this –gure (B9»2R1 B8»2R1 B7»2R1 and B7»2P1) have the same SRS terminal state JA\1 of X2. Molecules brought to this terminal state by the primary SRS processes can be redirected back to the SRS ground state through the SRS»FWPO process shown in Fig. 15. In this case pumping light for the secondary SRS process is provided by the enhanced densities of photons in the cloud that are nearly resonant with Ly-b. An emission band whose frequency corresponds to the last leg (B10»0P1) of the SRS»FWPO process illustrated in Fig. 15 can be seen in ORFEUS-1 spectra of B-type stars (Fig. 16). In Fig.16 a notch can be discerned on the emission band at the exact frequency of the B10»0P1 line centre. Parametric oscillation normally develops in such a manner that gain is optimized and loss is minimized. Therefore to avoid loss associated with real absorption occurring within the natural linewidth or elastic scattering occurring at larger frequency oÜsets a notch in the VUV emission band forms at the exact position of the B10»0P1 line centre. In ref. 23 another SRS»FWPO pathway capable of depleting the population in X2 JA\1 was discussed. In this alternative scheme the secondary SRS Stokes-wave emission occurs in the VUV. The secondary SRS intermediate state is the same J@\1(]) of C3. The SRS process which produces the prominent Stokes-wave emission peak assigned as B16»4P4 in Fig.14 is somewhat unusual. In this case the molecules that are excited to undergo SRS are parahydrogen molecules occupying the X0 JA\2 quantum level. Fig. 15 A possible SRS»FWPO process discussed in the text capable of generating emission on the transitions B9»2R1 B8»2R1 B7»2R1 and B10»0P1 153 P. P. Sorokin et al. 2 Fig. 16 Expanded view of a section of the ORFEUS-1 spectrum of HD 214080. The marked emission peaks and absorption regions are assigned in the text to the H transitions indicated. The pumping transition is B16»0R2 at 933.2 ”. Some evidence of non-linear absorption occurring around this wavelength can be seen in the ORFEUS-1 spectra of B-type stars (Fig. 17). However adding the photon energies of either Ly-a Ly-b or Ly-c to the energy of the X4 JA\4 quantum level (16 191 cm~1) results in an energy that is far removed from that of any singlet-ungerade-state quantum level optically connected to JA\4 of X4.To achieve resonance one has to add the energy of a Ly-d photon 105 292 cm~1. The frequency of the strong D4»4Q4 transition is 105 289 cm~1.30 It is also possible to explain some of the absorption and emission features seen in the Copernicus and ORFEUS-1 spectra using FWPO processes of the simpler type that were considered in ref. 1. For example Fig. 18 shows a simple FWPO process characterized by high radiative transition probabilities which should lead to emission around 2 Fig. 17 Expanded view of a section of the ORFEUS-1 spectrum of HD 214080 extending to the Lyman limit (91.2 nm).The H transitions marked are discussed in the text. 154 Current status of the H DIBS theory 2 Fig. 18 A possible simple FWPO process that could generate VUV emission on B13»0R0 954.4” ” and absorption around 929.5 . In Fig. 19 a prominent emission band is seen corresponding to the B13»0R0 wavelength but the observed absence of any hint of a notch is somewhat puzzling. In Fig. 17 there appears to be a fairly well de–ned absorption region corresponding to C4»0R0 at 929.5 ”. There may also exist some evidence of C4 ” »0Q1 absorption at 930.6 in Fig. 17 but this is somewhat hard to discern in view of the H(7p) absorption occurring at 930.75 ”. Signi–cantly there appears to be no hint of absorption corresponding to C4»0R1 at 929.7 ” although the calculated radiative transition probabilities of C4»0R0 and C4»0R1 are roughly similar.We will refer again to this observation in discussing our assignment of the k4428 and k4882 DIBs in the next section. 2 Fig. 19 Expanded view of a section of the ORFEUS-1 spectrum of HD 214080. An emission peak and several regions of absorption are assigned in the text to the H transitions indicated. 155 P. P. Sorokin et al. 5 Assignments of the kk5780 6284 4428 4882 large-equivalent-width DIBs Two prominent emission bands appearing in all the low-resolution Copernicus mid-/ late-B supergiant spectra and in the high-resolution Copernicus spectrum of b Ori A occur at 1188.3” ” and at 1244.0 . In ref. 23 these bands were assigned as Stokes-wave emissions on the transitions C3»6R0 and B10»6R0 respectively.Following the same line of argument employed in Section 4 one can here again account for removal of molecules from the SRS terminal state JA\0 of X6 by a plausible SRS»FWPO scheme (Fig. 20). It will now be shown that by combining this particular SRS»FWPO scheme with another one to be shortly described it is possible to make an assignment for what is probably the most famous DIB of all the one at 5780 ”. ative transition probabilities that could be occurring in H2 The energy level diagram of Fig. 21 shows a speci–c H process having high radi- 2-containing clouds around B-type giant and supergiant stars. From (EF4 J\0) a downward parametric oscillation cascade terminating on JA\0 of X0 could occur which would deplete the EF4 J\0 quantum level.However such a parametric cascade is not required for the SRS process illustrated in Fig. 21 to occur since the EF4 J\0 quantum level could otherwise —uoresce to lower-lying B-state quantum levels in strongly allowed transitions. The X0 JA\0 quantum level could thus be repopulated through purely —uorescent channels. For the moment we assume the existence of the SRS process illustrated in Fig. 21 without further comment about possible optical pathways leading back to JA\0 of X0. The existence of a region of absorption near 972.0 ” in Fig. 16 should however be noted. The BA-state quantum level shown at the top left of Fig. 21 is the very same one appearing at the top right of Fig. 20. In the scheme shown in Fig.20 this quantum level is shown as the resonant intermediate state in a secondary broad-band SRS process driven by Ly-b photons trapped in the H -containing cloud by H atoms also present in the cloud. It now becomes apparent that the secondary SRS process shown in Fig. 20 2 also creates a new SRS»FWPO pathway for the population excited to EF4 J\0 by the SRS process shown in Fig. 21. V ia the ìinverse Raman eÜectœ (ref. 11) it now becomes possible for visible photons coming from the illuminating star to be absorbed in a band Fig. 20 SRS»FWPO processes explaining some observed strong emission peaks appearing in Copernicus VUV spectra of mid-/late-B supergiants 156 Current status of the H DIBS theory 2 Fig. 21 SRS»FWPO process (when combined with the process shown in Fig.20) proposed in text as origin of k5780 DIB centred at the wavelength of the BA17EF4 R(0) transition with H molecules that 2 have been excited to EF4 J\0 now being driven to the (unspeci–ed) singlet-geradestate quantum level shown in Fig. 20. In Fig. 21 we have indicated the exact energies of both the BA1 J@\1 and EF4 J\0 quantum levels as listed in ref. 31 and in ref. (6 and 7) respectively. When one converts the diÜerence in energies of these levels 17 294.92 cm~1 to air wavelengths one obtains 5780.45 ”. In Table A1 of his recent review article on the DIBs,8 Herbig states the rest frame air wavelength of the k5780 DIB to be 5780.45^0.02 ”. Jenniskens and Deç sert4 list the wavelength of this DIB as 5780.59^0.05 ”. An important point to note concerning the SRS»FWPO mechanisms invoked in the above proposed k5780 DIB assignment is the following.If the H population that is excited to the unspeci–ed singlet-gerade-state quantum level shown at the top right in 2 Fig. 20 makes its way back to JA\0 of X0 entirely via a downward parametric oscillation cascade the summed photon momenta of all the light waves comprising either the SRS»FWPO process pictured in Fig. 20 or the one just discussed that combines transitions shown in Fig. 21 and 20 must be conserved i.e. ; k ki\0 where is the wave- i vector of the ith wave. Hence all the light waves participating in such SRS»FWPO processes must be coplanar i.e. they must all lie in the ìplaneœ of the thin H cloud of our model. On the basis of this consideration an explanation can be oÜered for 2-containing the known strong dependence (see e.g.ref. 8) of DIB intensity upon reddening in the line-of-sight. Reddening is generally believed by astronomers to be caused by dust particles of some kind. The presence of dust in the thin H -containing cloud of our model would cause visible photons coming from the illuminating star to be scattered into the 2 plane of the cloud thereby allowing the photons to participate in SRS»FWPO processes. Without the presence of such dust in the cloud all visible photons would simply traverse the thin cloud along the line-of-sight since there are no H or H absorption lines in the visible region of the electromagnetic spectrum to produce elastic scattering at 2 these wavelengths.In the discussion presented above it was implied that generation of a broad-band Stokes wave in an SRS process pumped by an intense —ux of Ly-b photons results in absorption of ca. 5780 ” photons through the inverse Raman eÜect. That is it was 157 P. P. Sorokin et al. tacitly assumed that the photon —ux at ” ca. 5780 propagating in the plane of the cloud is not intense enough to produce SRS by itself. On general grounds one might expect the linewidth of an absorption resulting from the inverse Raman eÜect to be much narrower than the spectral linewidth of an SRS Stokes wave generated by broad-band pumping since the former is basically only sensitive to the dispersion in the Raman susceptibility s(3) whereas the latter depends both on s(3) being large and the linear dispersion being small.32 With this in view it is interesting to note that the linewidths of both the k5780 and k6284 DIBs are ca.6 cm~1 whereas the linewidth of the k4428 DIB is roughly ten times greater. In the k6284 DIB assignment to be given below absorption could again occur via a plausible inverse Raman eÜect process. However in the case of the k4428 DIB assignment that will also be oÜered absorption via the inverse Raman eÜect is not possible and one must assume that broad-band light propagating in the plane of the H -containing cloud at wavelengths near 4428 ” is sufficiently intense to reach the SRS threshold. If this speculation should turn out to be correct the linewidth 2 of a large-equivalent-width DIB should provide an indication of the mechanism through which the DIB is generated i.e.SRS or inverse Raman eÜect. Of all the DIBs k4428 has the largest equivalent width being almost four times that of k5780 for example. The ATMs calculated in ref. 10 in principle apply to J\0 singlet-gerade-state quantum levels only. In Fig. 21 the large value indicated (0.5) for the IR transition shown should therefore be reasonably accurate. In ref. 10 ATMs are listed for transitions occurring between EF and B@ quantum levels but not for transitions between EF and BA quantum levels. However the potential curve for BA is similar in shape to that for B@ being only vertically displaced from the latter to somewhat higher energies. Therefore ATMs listed for EF»B@ transitions may be used as a guide in obtaining relative ATMs of EF»BA transitions.The ATM listed for transitions between B@1 and EF3 is very large (2.4) but the ATM listed for transitions between B@1 and EF4 is 60 times smaller (0.04) implying a very small value for the 17 295-cm~1 transition in Fig. 21. At –rst glance this appears to damage seriously the credibility of the k5780 DIB assignment here being proposed. The EF4 J\0 quantum level is also known as the F2 J\0 quantum level. The wavefunction for this quantum level is primarily localized in the outer (ìFœ) well of the double-welled EF-state potential curve. The wavefunction for the EF3 J\0 quantum level also known as the E1 J\0 quantum level is primarily localized in the inner (ìEœ) well of the EF-state potential curve. While the energy of F2 J\0 exceeds that of E1 J\0 by 204 cm~1 the energies of the E1 quantum levels increase with J faster than the energies of the F2 quantum levels.A cross-over occurs and at J\3 the energy of the E1 quantum level exceeds that of F2 by 80 cm~1. It is well known that this near degeneracy in energies leads to tunnelling through the barrier existing between the E and F wells. According to ab initio calculations presented in ref. 33 the vibrational wave functions of the outer well show a small but signi–cant tunnelling into the inner well and thus into a region that has good Franck»Condon overlap with the vibrational wavefunctions of the B@1 and BA1 states. Experimental veri–cation of tunnelling occurring between E- and F-state quantum levels was reported in ref.34. The authors of ref. 34 excited various E,F-state quantum levels by two-photon absorption from thermally populated levels of X0. The non-linear absorption was detected by monitoring the subsequent photoionization of the excited E,F quantum levels. The ion signal amplitude for the Q(0) transition to F2 was observed to be 87 times less than that for the Q(0) transition to E1. In each case the ion signal should vary as the amplitude squared of the rovibronic transition dipole moment between the two-photon-state quantum level and some intermediate-state quantum level. Hence for the F2 J\0 quantum level this transition dipole moment should be about nine times smaller than for the E1 J\0 quantum level. Thus one can estimate the ATM value for the 17 295-cm~1 transition in Fig.21 to be ca. 0.3. 158 Current status of the H DIBS theory 2 In Fig. 22 an assignment is proposed for the k6284 DIB that bears some resemblance to the k5780 assignment just discussed. In this assignment parahydrogen molecules are again the ìDIB carriers œ. The EF4 J\4 quantum level can be equally well described as the E1 J\4 quantum level that is its wavefunction is primarily localized in the E well. (The cross-over between E1 and F2 occurs between J\2 and J\3.) Therefore in obtaining ATMs for transitions between EF4 J\4 and B- B@- and C-state quantum levels one uses the values calculated in ref. 10 for transitions involving EF3. Table 1 of ref. 10 indicates very strong coupling between EF3 and C1 (ATM\3.3) C2 (ATM\0.2) B@1 (ATM\2.4) and B@2 (ATM\1.3) so that a large ATM for EF37D2 can reasonably be assumed.Using the term energy values given in ref. 31 for D2 J@\5([) (117 989.67 cm~1) and in ref. 6 and 7 for EF4 J\4(]) (102 081.05 cm~1) one obtains 15 908.62 cm~1 for the diÜerence frequency. This corresponds to an air wavelength of 6284.17 ” well within the limits (6284.31^0.33 ”) for the k6284 DIB given in ref. 4 but somewhat outside the limits (6283.86^0.02 ”) given in ref. 8. In ref. 30 results of new measurements of the line positions of some B@7X and D7X tran- D2(1% sitions are presented. In Table 7 of ref. 30 the energy for the u `) J@\5([) quantum level is given as 117 990.25 cm~1. This would predict the wavelength of the k ” 6284 DIB to be 6283.93 much closer to the value listed in ref.8. In the scheme shown in Fig. 22 no indication is given of how the SRS process that generates a Stokes wave on the C27EF4 P(4) transition is actually pumped. A simple two-photon SRS process would obviously have to start from JA\2 of X0. At –rst glance it is reasonable to assume that a sizeable steady-state population of parahydrogen molecules might exist in this level and there is some evidence in the ORFEUS-1 spectra of absorptions occurring from this level for example on the transitions B8»0P2 (Fig. 16) and C3»0Q2 (Fig. 19). In Section 4 we also saw that SRS Stokes-wave emission on B16»4P4 resulted from optical pumping of molecules in JA\2 of X0. However no absorption is seen in Fig 19 at the critical wavelength (965.8 ”) of C2»0R2.One therefore concludes that excitation of the EF4 J\4 quantum level does not occur via a simple SRS process starting from JA\2 of X0. We propose that excitation of the EF4 J\4 quantum level occurs through the cascaded SRS process diagrammed in Fig. 23. There is abundant evidence in the Coper- Fig. 22 SRS»FWPO process (when combined with the process shown in Fig. 23) proposed in text as origin of k6284 DIB 159 P. P. Sorokin et al. Fig. 23 SRS»FWPO processes proposed in text as mechanisms for excitation of EF4 J\4 quantum level nicus and ORFEUS-1 spectra that a speci–c set of parallel SRS processes is occurring in the H2-containing clouds near the stars studied with each process starting from JA\0 of X0 and terminating on JA\2 of X6. One –nds in the Copernicus spectra emission bands at B2»6P2 (1406.8 ”) B4»6P2 (1359.8 ”) B5»6P2 (1338.4 ”) B8»6P2 (1281.1 ”) B11»6P2 (1232.7 ”) B14»6P2 (1191.4 ”) B15»6P2 (1179.2 ”) B17»6P2 (1156.5 ”) B18 ) ” ” »6P2 (1145.9 emission band may also be C0»3Q1) C4»6P2 (1165.2 C2»6P2 ”) (1221.4 C1»6P2 (1254.7 ”) and C0»6P2 (1292.1 ”).The strongest of these transitions are shown in Fig. 23. Regions of absorption corresponding to the R(0) Bn7X0 and Cn7X0 pumping transitions are easily identi–ed in the ORFEUS-1 spectra. One should keep in mind that in broad-band SRS it is the total pump power at all wavelengths that determines the Stokes-wave gain at any particular wavelength. Thus for example the pump power around B5»0R0 contributes to the gain of the Stokes wave around B14»6P2.The secondary SRS process shown in Fig. 23 could deplete the terminal state (JA\2 of X6) of the primary SRS process. One sees in this –gure that Ly-a is nearly resonant with the strong (0.13]9~1) transition C2»6R2. The secondary SRS Stokes wave is assumed to occur on the C27EF4 P(4) transition as shown in both Fig. 23 and 22. The sequence of two-photon transitions that begins as in Fig. 23 and that then proceeds to some other singlet-gerade-state quantum level via inverse Raman absorption of ca. 6284 ” photons with J@\5([) of D2 being the resonant intermediate state is assumed to be part of an overall SRS»FWPO process that returns molecules to JA\0 of X0. In the scheme shown in Fig. 22 absorption of photons near 6284 ” probably occurs via inverse Raman absorption.However unlike the case of the scheme shown in Fig. 20 it is here difficult to isolate an independent SRS process that can generate broad-band coherent radiation centred about the transition between J@\5 of D2 and a lower-lying singlet-gerade-state quantum level. The energies obtained by subtracting Ly-a Ly-b Ly-c or Ly-d from the energy of D2 J@\5 are nowhere close to those of X-state quantum levels except JA\4 of X13. However D27X13 transitions are extremely weak relatively speaking. This implies that broad-band coherent radiation centred on the transition from D2 J@\5 to a lower-lying singlet-gerade-state quantum level must 160 Current status of the H DIBS theory 2 ìcome from aboveœ that is it must represent a component of a parametric oscillation cascade that originates at some higher-lying level.The existence of such a higher-lying level emerges from our (combined) assignments of the k4428 and k4882 DIBs to be discussed next. Our assignments for both the k4428 and k4882 large-equivalent-width DIBs are shown in Fig. 24. From inspection of this –gure several unusual features are seen to be involved. (1) The uppermost quantum level shown in the –gure could be either D8(1%u `) J@\2(]) or BA6(1&u `) J@\2(]). According to ref. 31 both quantum levels have exactly the same energy 127 320.11 cm~1 and both are seen to be optically connected with the two EF-state quantum levels shown although none of the relevant ATMs were calculated in ref. 10. It is at this point impossible to say which of the two equi-energetic upper levels plays the dominant role in the case of each of the two DIBs.(2) In Fig. 24 excitation of the quantum level EF11 J\3 is shown occurring via an SRS process that starts from X0 JA\3. Strong supporting evidence that this process actually occurs can be seen from inspection of Fig. 17. We remarked at the end of Section 4 that no evidence of C4»0R1 absorption exists in this –gure. However the reader should now take note of the intense region of absorption appearing at the C4» 0P3 wavelength. In general absorptions originating from JA\3 of X0 appear quite prominently in the ORFEUS-1 spectra suggesting the existence of a relatively large population in this level perhaps fed by Stokes-wave emissions. In Section 4 a prominent emission band was assigned to B0»0P3.From Fig. 19 one sees that strong absorption on either C3»0R1 C3»0R0 or on both of these transitions could be occurring but it is hard to distinguish among these possibilities since the frequencies of C3»0R0 and C3»0R1 diÜer by only 4 cm~1. In principle the EF8 J\1 state could also be populated by an SRS process involving C3»0Q1 as the pumping transition. While there is no strong evidence of absorption occurring at C3»0Q1 in the case of HD 214080 the ORFEUS-1 spectrum of HD 149881 shown in Fig. 25 gives some indication that SRS excited via C3»0Q1 may be occurring in the H -containing cloud near this star. (The wavelength scale for the HD 149881 spectrum is less accurate than that for HD 214080.) The non-linear nature of the 2 H absorptions which appear in the ORFEUS-1 spectra of these two B-type halo stars is again illustrated by the fact that the radiative transition probability of C3»0Q1 is 3.5 2 times that of C3»0R1.Fig. 24 SRS»FWPO processes proposed in text as origins of j4428 and j4882 DIBs 161 P. P. Sorokin et al. Fig. 25 Expanded views of ORFEUS-1 spectra of two stars (HD 214080 and HD 149881) that include the wavelength region of the C3»0 transitions discussed in the text. All the pumping and Stokes-wave transitions shown in Fig. 24 have very large transition dipole moments making them very favourable for SRS processes. All the quantum level energies have been rounded oÜ in Fig. 24 since the DIBs here are very broad and the frequencies at their maxima are therefore much less accurately known than in the case of narrower DIBs.The exact frequencies and air wavelengths of the two transitions in Fig. 24 to which the two DIBs are assigned are 22 572.78 cm~1 (4428.89 ”) and 20 480.45 cm~1 (4881.36 ”). Ref. 4 lists the wavelengths linewidths and equivalent widths (in HD 183143) of the two DIBs as 4428.88^1.35 ” 12.33^2.06 ” and 2488 m”; 4881.83^0.65 ” 19.67^2.38 ” m and 810 ”. In ref. 8 the values of the same quantities are given as 4428^1 ” 18 ” m and 3400 ”; 4882 ” 25 ” m and 890 ”. In wavenumber units the linewidths of these two DIBs given in ref. 4 are 63^10 cm~1 for the k4428 DIB and 82^10 cm~1 for the k4882 DIB. Within the limits of error the linewidths could thus possibly be the same.We hypothesize that enough blue and blue-green light is scattered by dust into the plane of the H -containing cloud to pump the secondary SRS processes shown in Fig. 24 beyond threshold thereby opening up new hybrid SRS»FWPO optical pathways. 2 One should here bear in mind that the —ux emitted by a T \15 000 K radiator when expressed in units of power per unit area per unit frequency interval peaks somewhere near the blue region of the spectrum. It is also a feature of broad-band SRS that the Stokes-wave gain is remarkably —at even when the pump radiation spectrally spans a resonance line.32 The secondary SRS process shown in Fig. 24 is hypothesized to continue in a downward parametric oscillation cascade. If the optical pathway of this downward cascade were to involve for example the D2 J@\5([) quantum level in Fig.22 this would lead to absorption of photons near 6284 ” by the inverse Raman eÜect. 6 Conclusions In trying to account for major de–ciencies apparent in the H DIBs model of ref. 1 we have called attention to what we believe is VUV spectral evidence indicating that there 2 ubiquitously exist thin cold tenuous H -containing clouds located close enough to B-type giant and supergiant stars that the 2H molecules in these clouds are driven into a 2 162 Current status of the H DIBS theory 2 pattern of coherent oscillations by the VUV blackbody continua emitted by the illuminating stars. At present we are unable to estimate either the actual distances between the H -containing clouds and the nearby stars or the shapes of the clouds.Our overall conclusion regarding the generation of coherent light in these clouds mostly derives 2 from the following sequence of observations. (1) Emission bands seem to be clearly present in archived (900»1450 ”) spectra of B-type supergiants. (2) These bands appear at exactly the same wavelengths in the spectra of all the mid-/late-B supergiants examined. (3) On the basis of wavelength positions alone these bands appear to be readily assignable to low-JA H transitions on which one would expect SRS Stokes-wave emission to occur if viable mechanisms existed to deplete the populations of the terminal 2 states. We then proposed that such mechanisms are provided by the presence of greatly enhanced densities of Ly-a Ly-b Ly-c .. . etc. photons trapped in the H -containing cloud via elastic scattering by H atoms also assumed to be present at high densities in 2 the cloud. It was shown how pumping by the Ly-a Ly-b Ly-c . . . etc. —ux can initiate secondary SRS processes which deplete the primary SRS terminal states. It is assumed in our model that the secondary SRS processes then typically initiate downward parametric oscillation cascades which ultimately repopulate the ground state. The overall optical pathways in such schemes can best be described as SRS»FWPO mixing processes in which SRS inverse Raman absorption or parametric oscillation can occur at one or more of the 2n (nP2) steps. In such SRS»FWPO schemes conservation of photon momentum holds i.e. i\0 ; 2 n i/1 References k This necessitates that all the light waves participating in SRS»FWPO processes be coplanar.Assignments for several large-equivalent-width DIBs are shown to follow from consideration of speci–c hybrid SRS»FWPO processes. In order that visible photons emitted by the illuminating star can be absorbed in individual SRS or inverse Raman steps of a hybrid SRS»FWPO process the photons must be propagating in the ìplaneœ of the thin H2-containing cloud. This can only happen if there is also present dust in the cloud to scatter the visible light coming from the star into the plane of the cloud. This provides a simple explanation for the general observation that DIB intensities in given lines-of-sight correlate well with reddening observed in same lines-of-sight.We are especially grateful to Dr Ralph C. Bohlin for sending us his ASCII-format data –les of Copernicus spectra of mid-/late-B supergiant stars. We have bene–ted from several conversations with Drs Donald S. Bethune and Rodney T. Hodgson and Prof. Stephen E. Harris. P.P.S. and J.H.G. acknowledge the U.S. Army Research Office for having partially supported their work. W.U. acknowledges the Vrije Universiteit and the Research Institute for Condensed Matter and Spectroscopy (COMPAS) for a USFproject grant. 1 P. P. Sorokin and J. H. Glownia Astrophys. J. 1996 473 900. 2 T. P. Snow Nature (L ondon) 1996 384 406. 3 R. Loudon T he Quantum T heory of L ight Oxford University Press New York 2nd edn. 1983. 4 P. Jenniskens and F-X. Deç sert Astron.Astrophys. Suppl. Ser. 1994 106 39. 5 H. Abgrall E. RoueÜ F. Launay J-Y. Roncin and J-L. Subtil J. Mol. Spectrosc. 1993 157 512. 6 S. Yu and K. Dressler J. Chem. Phys. 1994 101 7692. 7 S. C. Ross and Ch. Jungen Phys Rev. A 1994 50 4618. 8 G. H. Herbig Annu. Rev. Astrophys. 1995 33 19. 9 H. Abgrall E. RoueÜ F. Launay J-Y. Roncin and J-L. Subtil Astron. Astrophys. Suppl. Ser. 1993 101 323. 163 P. P. Sorokin et al. 10 M. Glass-Maujean P. Quadrelli and K. Dressler At. Data Nucl. Data T ables 1984 30 273. 11 R. A. McLaren and B. P. StoicheÜ Appl. Phys. L ett. 1970 16 140. 12 H. Abgrall E. RoueÜ F. Launay J-Y. Roncin and J-L. Subtil Astron. Astrophys. Suppl. Ser. 1993 101 273. 13 P. J. Sarre J. R. Miles T. H. Kerr R. E. Hibbins S. J. Fossey and W. B. Somerville Mon. Not. R. Astron. Soc. 1995 277 L41. 14 P. Ehrenfreund and B. H. Foing Astron. Astrophys. 1996 307 L25. 15 J. Krelowski and M. Schmidt Astrophys. J. 1997 477 209. 16 B. E. Westerlund and J. Krelowski Astron. Astrophys. 1988 203 134. 17 P. F. Roche D. K. Aitken and C. H. Smith Mon. Not. R. Astron. Soc. 1989 236 485. 18 W. Ubachs P. C. Hinnen and E. Reinhold Astrophys. J. 1997 476 L93. 19 G. S. Agarwal and W. Harshawardhan Phys. Rev. L ett. 1996 77 1039. 20 S. E. Harris Phys. T oday 1997 50(7) No. 7 36. 21 W. Ubachs and E. Reinhold 1997 unpublished data. 22 R. H. Page Ph.D. Thesis University of California Berkeley 1987. 23 P. P. Sorokin and J. H. Glownia submitted for publication. 24 M. Hurwitz S. Bowyer R. Kudritzki and D. J. Lennon Astrophys. J. 1995 450 149. 25 D. C. Morton Astrophys. J. 1975 197 85. 26 N. R. Walborn and R. C. Bohlin Pub. Astron. Soc. Paci–c 1996 108 477. 27 J. B. Rogerson and W. L. Upson Astrophys. J. Suppl. Ser. 1982 49 353. 28 J-Y. Roncin and F. Launay J. Phys. Chem. Ref. Data (Monograph No. 4) 1994 American Physical Society Woodbury NY and American Chemical Society Washington. 29 I. Dabrowski Can. J. Phys. 1984 62 1639. 30 H. Abgrall E. RoueÜ F. Launay and J-Y. Roncin Can. J. Phys. 1994 72 856. 31 A. Mon–ls J. Mol. Spectrosc. 1965 15 265. 32 D. S. Bethune J. R. Lankard M. M. T. Loy and P. P. Sorokin IBM J. Res. Develop. 1979 23 556. 33 W. Kolos and L. Wolniewicz J. Chem. Phys. 1969 50 3228. 34 E. E. Marinero R. Vasudev and R. N. Zare J. Chem. Phys. 1983 78 692. Paper 8/00224J; Received 7th January 1998

ISSN:1359-6640    

DOI:10.1039/a800224j

出版商:RSC    

年代:1998

数据来源: RSC