Science Case   


MIRI  spectrometer for NGST

Overviews, Dutch participation and Science Case 

  Printing this frame may need "open in new window"; 
alternatively,  download Science Case only ( .pdf,  0.5MB)

See frame above for an always-available top-level Table of Contents (ToC).

All pictures are minimal for rapid loading; click to obtain a large version for best print.

For the project overall organization and time schedule, see ..............  

Most of the optical design and mechanical engineering in the Dutch contribution will take place at ASTRON, Dwingeloo, Netherlands.  ASTRON will use and extend the experience gained during the construction of VISIR, the mid-IR spectrometer/imager for ESO's VLT.

The present WWW document was extracted and marginally adapted from the proposal (funded in June 2002) to NWO (Netherlands Organization for Scientific Research),  for a Dutch contribution to the ESA-led European participation in the (NASA) NGST project. We have retained the original section numbering; some sections that are no longer relevant have been omitted. 


The mid-infrared wavelength range from 530 m is a key region for studying the origin and evolution of galaxies, stars and planetary systems.  A mid-infrared instrument on the Next Generation Space Telescope (NGST) can address and resolve outstanding scientific questions of interest to Dutch astronomers because of its orders of magnitude higher sensitivity and spatial resolution compared with all previous, existing and planned facilities. The Netherlands can make a crucial contribution to this project with a relatively small investment, which will ensure that the spectrometer part of the instrument is built.   This proposal requests NWO funding (granted, June 2002) to participate in the design and prototype studies of the instrument in the period 2002mid 2004 and in the construction phase 20042007. The project is a collaboration between astronomers at the four Dutch universities participating in the NOVA top-research school and the technical groups at the ASTRON and TNO-TPD institutes in association with SRON, and builds on the pioneering Dutch expertise at mid-infrared wavelengths developed in the 1980's and 1990's.



Executive summary

The Hubble Space Telescope (HST) is one of the most successful astronomical observatories ever deployed,  contributing profoundly to almost all branches of astronomy throughout the nineties.  In particular, deep surveys with HST have yielded the first view of galaxy evolution from small, high brightness building blocks at the edge of the universe to the majestic spirals and ellipticals seen today.  Images of the Orion nebula have revealed tiny disks of dust and gas encircling young stars, providing the raw material from which new solar systems are made. The results fascinate not only astronomers but also the general public.  

The successor to HST, the Next Generation Space Telescope (NGST), can be expected to be equally successful.  NGST is envisioned to be a 6.5m, passively cooled telescope optimized for diffraction limited performance in the infrared (Table 1).  NGST will have more than 100 times the sensitivity and 5 times the image sharpness of HST in the infrared. NGST is designed to detect first star light at the dawn of the universe, to trace the genesis and evolution of galaxies from their formation to the present, and to determine the nature of luminous galactic nuclei.  NGST will also revolutionize our view of star and planet-formation in the galaxy today, by unveiling the earliest protostellar phases, imaging the planet-forming disks around pre-main sequence stars, and even studying exo-solar giant planets  themselves.  Finally, with NGST a comprehensive understanding of the lifecycle of the elements can be developed from their injection into the interstellar medium by dying stars to their incorporation into planetesimals in the protoplanetary disks. The proposed science covers a large fraction of  the NOVA top research school program. 

 NGST is scheduled to be launched in 2009 and will have three instruments on board:  

  1. A Near-Infrared Wide Field Camera covering the wavelength range 0.65 m;

  2. A Near-Infrared Multiobject Spectrometer at 15 m; and

  3. A Mid-Infrared Camera/Spectrometer at 528 m. 


The mid-infrared instrument MIRI will be built on a 50%-50% basis by NASA and ESA with a small contribution from Canada. Europe will be reponsible for the optical design and fabrication, the relevant portions of calibration, integration and testing plus the associated electronics and software.  At the ESA council meeting in October 2000, the ESA participation in NGST was confirmed but it was decided that the European part of the mid-infrared instrument will have to be funded by the ESA member states directly, with ESA providing the cryostat and interface with the spacecraft.  Here we propose that the Netherlands makes a critical contribution to this European effort in the design, building and calibration of the mid-infrared spectrometer. 

MIRI will provide an excellent opportunity to maintain and extend a rich and pioneering tradition in building infrared instrumentation in the Netherlands, in particular in spectroscopy.  Over the last three decades, SRON has led the development of the IRAS-Low Resolution Spectrometer (LRS) and the ISO-Short Wavelength Spectrometer.  Presently, SRON is the PI-institute for HIFI the submillimeter heterodyne instrument for the Herschel Space Observatory   which hardware-wise will be delivered in mid-2004. Within ASTRON, mid-infrared instrumentation (VISIR, MIDI) has been developed for the ground-based Very Large Telescope (VLT) and its interferometer (VLTI) of the European Southern Observatory (ESO).  TPD-Delft also has much experience in this area, through participation in hardware development for the ISO-SWS and atmospheric missions. The mid-infrared instrument on NGST is related to these projects instrument-wise, science-wise, and project-wise. 

MIRI allows Dutch astronomy to participate in a highly visible space mission which is destined to revolutionize our understanding of the origin and evolution of the universe.  Driven by the Dutch instrumentation participation in IRAS, ISO and ESO, an active infrared astronomical community has developed in the Netherlands, which is internationally recognized as a leader in infrared spectroscopy.  For these reasons, the strategic plan of Dutch astronomy has put participation in this instrument as its top priority for new space projects (Appendix B).  


The main benefits for Dutch astronomers are: 

  1.    ensuring that the NGST mid-infrared instrument will have a proper spectrometer with the desired capabilities (see Table 2);

  2.    obtaining a fraction of the guaranteed NGST time;

  3.    gaining intimate knowledge of the instrument and its data reduction, essential as a headstart for subsequent open-time observing proposals; and

  4.    retaining unique mid-infrared technical expertise in the Netherlands. 


Participation in NGST is also on the scientific and technology path necessary to achieve ESA's next infrared mission, the infrared interferometer space mission IRSI/Darwin, slated for launch by the mid-2010's




 Cover of NWO proposal


Top: Mid-infrared 7 m image obtained with the camera on the Infrared Space Observatory ISO (orange image in center) and 20 cm radio continuum contours, superposed on an optical image of the nearest active galaxy Centaurus A.  Only the mid-infrared data reveal a warped disk of hot dust near the nucleus. Spectra taken at different positions can distinguish the energy sources responsible for the emission (Mirabel et al 1999).    

Bottom: Images of protoplanetary disks taken with the Hubble Space Telescope in near-infrared light, together with the mid-infrared spectrum of one such disk obtained with the ISO-Short Wavelength Spectrometer (Malfait et al. 1998). Note the wealth of mid-infrared features and the similarity with the spectrum of comet Hale-Bopp.


Top: Left: An optical (R-band) image of the field centered on the bright infrared source IRS2, in the Flame Nebula in Orion (North to the top, East to the left, image obtained from the Digital Sky Survey).  Apart from the bright emission from ionized hydrogen, virtually no stars are detected. In particular, the center of the field, at the location of IRS2, is totally obscured by surrounding gas and dust.  Right: A near-infrared image of the same field, which reveals the central cluster of young massive stars including IRS2, showing that at infrared wavelengths the extinction is much less severe (Lenorzer et al. 2001).  

Bottom: The cycle of organic molecules in the universe.  The molecules formed in interstellar clouds are transported through circumstellar disks to small solid bodies in new planetary systems such as comets and asteroids. Impacts of these objects may deliver these species on new planets. In the final stages of stars, dust and elements are returned to the interstellar medium (figure by R. Ruiterkamp, based on Ehrenfreund & Charnley 2000).


Applicants in the spectrometer funding proposal to NWO


Principal Investigators:


Prof. dr. L.B.F.M. Waters                           

Sterrenkundig Instituut Anton Pannekoek, Amsterdam




Prof. dr. E.F. van Dishoeck 

Sterrewacht Leiden




Dr. Th. de Graauw, SRON Groningen  

Dr. J.W. Pel, Kapteyn Laboratory Groningen   

Prof. dr. A. Quirrenbach, Sterrewacht Leiden    



Prof. dr. A.G.G.M. Tielens, Kapteyn Laboratory Groningen

Dr. P.P. van der Werf, Sterrewacht Leiden

Dr. D. de Winter,  TNO-TPD  Delft



NL Science Team:


E. Bakker, P. Barthel, P. Ehrenfreund, M. Franx, K. van der Hucht, F. Israel,W. Jaffe, L. Kaper,  A. de Koter,

H. Lamers, H.J. van Langevelde, H.Rttgering, M. Spaans, W. Schutte, P. Wesselius, P.T. de Zeeuw




I.  Introduction


Over the ages, mainly driven by new observational tools becoming available, astronomers have surveyed the sky in search of knowledge on the origin and evolution of the universe and its constituent galaxies, stars, and planets.  This quest started with the first telescopic observations by Galileo in the 17th century which firmly established the heliocentric Copernican view of the Solar system.   In the early parts of the last century, Edwin Hubble used the then largest telescope, the 100 inch at Mt.Wilson,  to discover  the expanding universe in which no place is special.  Over the last decade, the Hubble Space Telescope (HST) has expanded this view to include much of space and time as it enabled astronomers to look back to ever larger redshift and get the first tantalazing glimpses of galaxies in formation. Astronomy is poised to take another leap forward when new instruments, presently on the drawing boards, will become available by the end of this decade.   Foremost among these is the Next Generation Space Telesope, NGST, which will provide orders of magnitude increased sensitivity and spatial resolution in the infrared spectral window (see Table 1).

Table 1. Overview of the NGST mission  

Mission type: General purpose astronomical observatory
Telescope: Type      Three mirror anastigmat  
Primary         6.5 meter segmented mirror  
Image quality  Diffraction limited at 2 m (0.08'')  
Spacecraft: Orbit 

Sun-Earth Lagrange 2 halo orbit

Mission lifetime 5 years consumables for 10 years
Payload: Wavelengths 



Passively cooled to 50 K

Stray light   Sky-dominated below 10 m  
Sky coverage Instantaneous: >15%; Yearly: whole sky
Instrument modules: Type 

Near-infrared camera   0.65 m

Near-infrared multi-object spectrometer   15 m, R~1000  
Mid-infrared camera/spectrometer   528 m, R~3000 goal  
Total mass <1000 kg  
Expected launch date: 2009  


The key aim will be to chart the history of star formation in the universe all the way from the present, where the star- and planet formation process can be studied in detail, to the dawn of time when the first stars formed and galaxies were assembled. A large space telescope  at infrared wavelengths is the logical choice for these programs, since for the most distant galaxies, lines such as H Lyman-alpha at 1216 are shifted to the red part of the spectrum, where the ground-based sky brightness is high and the objects themselves are becoming very faint. This history of star formation is of course intimately interwoven with the origin of the elements in the universe because as stars evolve they enrich their environment with freshly synthesized elements which then become available for the next generation of stars.  Star formation is also closely linked to the formation of planetary systems which are now known to be widespread and the origin of life thereon.   

Much progress in these areas will have to come from the mid-infrared part of the spectrum as the formation of stars and planets occurs in the deeply enshrouded environment of molecular clouds and disks, hidden from visible view by copious amounts of small dust grains. The observations at long wavelengths allow astronomers not only to lift this obscuring veil (see back-cover), but also to study the dust itself. Indeed, the mid-infrared wavelength range is particularly rich in diagnostic spectroscopic features and it is the only region where the dominant molecule H2 the principal ingredient of gaseous giant planets and the solid state species can be studied directly.

The mid-infrared instrument on NGST is expected to revolutionize our understanding of these problems since it will have unprecedented sensitivity and spatial resolution.  Figure 1 illustrates the expected imaging and spectral sensitivity compared with other facilities. NGST  will be three orders of magnitude more sensitive than any ground-based telescope in the 530 m range, a large part of which (>50%) is completely inaccessible from the ground due to absorption by the Earth's atmosphere.  Moreover, even within the atmospheric windows, ground-based observations are limited by atmospheric conditions to less than 50% of the night time,  even at good mid-infrared sites.  Compared with SIRTF (the 85-cm NASA Space Infrared Telescope Facility, to be launched in 2002), NGST will have more than an order of magnitude increase in sensitivity and spatial resolution. Moreover, SIRTF will have only very low resolution spectroscopy R = lambda/delta(lambda) = 50100 in the important 510 m range, and only R = 600 in the 1038 m range.   Such low resolving powers are insufficient for many key scientific programs (section II.1).  Thus, NGST has the potential to perform unique science in a poorly explored wavelength range, increasing substantially the discovery space.



Figure 1. Imaging (left scale) and spectral (right scale) sensitivity  of NGST (orange line) compared with previous space missions (IRAS, ISO), future space missions (SIRTF) and future ground-based instruments (VISIR on the 8m ESO-VLT). The ISO-SWS stars (light-blue) refer to the  flux scale, all other facilities refer to the broad-band flux scale.


The mid-infrared instrument will be built jointly by NASA and ESA in a 50%50% partnership. The European consortium will be led by the UK and includes at least France, Germany and Italy.  At the NGST Mid-Infrared Steering Committee (MISC) meeting in July 2001 in the UK, the basic instrument characteristics (Table 2) and division of labor were agreed between the US and Europe. Here we request funding for the Netherlands to make a critical contribution to the European consortium which will ensure that the spectrometer part of the instrument is built with the desired spectral resolution and integral field capability.  The initial 7-month instrument design study (phase A)  will be partly financed by ESA and the applicants, and will start in Fall 2001. The subsequent Phase B/CD design and construction will start in the spring of 2003, with instrument delivery planned for 2007/2008.


Table 2. Top-level specifications of the NGST mid-infrared instrument





Wavelength range 527 m

Diffraction limited imaging, Nyquist sampled at no longer than 8 m 

Minimum field of view of 1.5 arcmin

Filter wheel with at least 8 spectral bands: 45 for SED definition,

   remainder for PAH isolation, brown dwarf atmosphere color identification

Simple coronograph

Larger imaging field of view a goal



Single object, 527 m, important goal to extend to 28.3 m

Resolution R ~100 from 510 m

Resolution R ~10003000 from 528.3m

Resolution close to 3000 (or slightly higher) to be a goal



II.  Science case  


The scientific case for the mid-infrared instrument on NGST has been described previously by Serabyn et al. (1999), Wright et al. (1999) and van Dishoeck (2000). The general NGST science program is contained in the Design Reference Mission, a set of strawman observing proposals written by the NGST Ad Hoc Science Working Group (ASWG) (see Appendix G).  It was most recently summarized in the NGST proposal to ESA in July 2000 (ESA-SCI(2000)9). This proposal led to the confirmation of the ESA participation in NGST in October 2000.  Dutch astronomers have actively participated in defining these science cases.

In the following, a brief overview of the rich variety of features at mid-infrared wavelengths will be given, followed by a selection of scientific topics which highlight the spectroscopic applications of interest to the NOVA research program. One of the main benefits of a Dutch investment in the mid-infrared instrument is that it will allow Dutch astronomers to be part of the mid-infrared science team that defines the observing programs for the guaranteed time. As always with new instruments, the most exciting science will likely come from unexpected discoveries in areas which have not even been considered.  The enormous increase in sensitivity and spatial resolution with NGST will virtually guarantee such serendipitous results.


Science Case subsections

II.1.   Mid-infrared features  

II.2.   Obscured star formation in galaxies    

II.3.   The high-redshift universe  

II.4.   Star formation 

II.5.   Protoplanetary disks  

II.6.   Giant exo-planets and brown dwarfs 

II.7.   Tracing the astrochemical evolution 

II.8.   The death of low mass stars

II.9.   Outer solar system: comets and KBOs

II.1.  Mid-infrared features

In the local universe, the continuum at mid-infrared wavelengths is dominated by thermal emission from warm dust with Tdust ~ 100300 K, such as found in star-forming regions and around late-type stars (red giants, so-called Asymptotic Giant Branch (AGB) stars).  The ultraviolet and optical radiation from these stars is absorbed by the dust and re-radiated at longer wavelengths. In addition, thermal emission from the atmospheres of cool stars, from brown dwarfs with effective temperatures Teff  ~ 1000 K, and from solar-system objects such as comets and Kuiper-Belt objects is significant at mid-infrared wavelengths. Any non-thermal (synchrotron) contribution is generally small.



Figure 2.   Complete ISOSWS grating spectrum centered at Orion IRc2 at a resolving power R=13002500. The principal absorption and emission features are indicated.  The ISO-SWS beam ranges from 14'' x 20'' to 20'' x 27'', and encompasses both the IRc2 and BN sources (van Dishoeck et al. 1998).


The mid-infrared is rich in spectral features, many of which are unique for this wavelength range. They are illustrated in the ISO Short Wavelength Spectrometer (SWS) spectrum of one of the brightest mid-infrared sources in the sky, the Orion-KL region (Figure 2) and are summarized in Table 3. Each of the categories of features is discussed below, together with their diagnostic value.

Table 3. Some important spectral features in the mid-infrared


Wavelength (m)   Species Diagnostica



7.0   [Ar II]   Radiation field 


9.0   [Ar III] Shocks


10.5 [S IV] Spectral index 


12.8   [Ne II]   Density 


14.3  [Ne V]     Metallicity


15.6     [Ne III]

18.7  [S III]      
25.2 [S I] 


25.9 [O IV]   


6.2, 7.7, 8.6, 1.3 Carbonaceous material,      
12.7, 14.2, 16.2 UV, Redshift obscured galaxies 

Silicates (Amorphous) 

9.7 Bulk of dust



Silicates (Crystalline)

10.0, 11.3, 16.3, 19.5 Mg2SiO4   Mineralogy, Heating events  


23.5,27.5 Solar system connection

18.5, 21.5, 24.5   (Mg,Fe)SiO3   (meteorites, IDP's, ...) 


11.6  Al2O3 


23   FeO 


23 FeS    Gas-solid chemistry 


6.8, 11.3 XCO3   Aqueous alteration, formation large bodies 


6.0 H2O Volatile solid material


6.8, 9.7 CH3OH   Building blocks complex organics 


7.7  CH4  Solar system connection (comets, KBOs) 


9.7  NH3      Thermal history 

15.2       CO2   


6.9, 8.0, 9.7, Mass + temperature of bulk of warm gas 


12.2, 17.0, 28.2 Photon- vs shock-heating 


19.4, 23.0, 28.5  [D]/[H]


6.0   H2   Temperature + density structure 


7.7         CH4  Building blocks complex organics 


13.7      C2H2  


14.0            HCN  
15.0 CO2

 a Diagnostic properties of category of species; individual species or lines probe a subset of these properties.


Atomic and ionic fine-structure lines: Fine-structure transitions within the lowest electronic term of most astrophysically relevant atoms and ions occur in the 530 m range. Important examples are the [Ar II] 7.0, [Ne II] 12.8, [Ne V] 14.3, [Ne III] 15.6,  [S I] 25.2, and [O IV] 25.9 m lines, but even higher ionization stages  ("coronal" lines) can be observed. These features are used as probes of the ionizing source, especially the hardness of the radiation field (e.g., starbursts vs. active nuclei),  and as diagnostics of heating mechanisms, in particular shock- vs. photon-heating. The strongest lines can also be used to trace the kinematics of gas in obscured regions.  

PAHs:  The CC and CH stretching and bending modes of polycyclic aromatic hydrocarbons (PAHs) at 6.2, 7.7, 8.6, 11.3, ...  m often dominate the mid-infrared emission. More recently, new features at longer wavelengths have been found in ISO-SWS spectra, which contain important diagnostic information about the mix of species involved (Tielens et al. 2000). Significant variations of factors of five or more can occur in the ratios of various bands (e.g. 8.6/7.7 and 11.3/7.7 m), which appear related to differences in the ionization and hydrogenation state of the carriers. PAHs contain a significant fraction of the carbon budget and play a role in the energy balance in the interstellar medium. They are also the best redshift indicators for distant, obscured galaxies.  

H2 and HD pure rotational lines: The lowest transitions of the most abundant molecule in the universe, H2, occur at mid-infrared wavelengths (see Figure 3). The fundamental J = 20 S(0) quadrupole line of para-H2 occurs at 28.22 m, whereas the next transition, the J = 31 S(1) line of ortho-H2, lies at 17.03 m.  Since the populations of these levels are thermalized under many conditions, the S(0) and S(1) lines provide a direct measure of the mass and temperature of the bulk of warm molecular gas at T = 50200 K. In contrast, the higher pure rotational lines, as well as the vibration-rotation lines at 2 m, probe only the small fraction (<1%) of photon- or shock-heated gas in the beam (Draine & Bertoldi 1999, Wright 2000).  The heavier isotope, HD, has a small dipole moment and its lowest J = 10 line at 112 m has been detected with the ISO-LWS (Wright et al. 1999).  The higher J = 43, 54 and 65 lines occur at 28.5, 23.0 and 19.4 m. The latter line was observed by Bertoldi et al. (1999) in the Orion shock and can, in combination with H2 data, be used to determine the [D]/[H] ratio in star-forming regions, providing constraints on the deuterium destruction in stars since its production in the Big Bang.  

Solid-state vibrational bands: The characteristic vibrational bands of ices, silicates and oxides occur uniquely at mid-infrared wavelengths. Solid-state species can be distinguished from gas-phase molecules because their bands lack the characteristic ro-vibrational structure and are broadened (see Figure 4).  The presence of H2O and CO ice mantles surrounding grain cores was established from ground-based observations, but  only ISO revealed the complete interstellar ice content and the rich spectral detail. Similarly, the SiO stretching and bending bands of amorphous silicates at 9.7 and 18 m were well  known, but features of crystalline silicates and oxides at longer wavelengths were discovered only recently by ISO in circumstellar matter around young and old stars (Waters et al. 1996,  Waelkens et al. 1996). This wealth of mid-infrared solid-state features was one of the big surprises and legacies of the ISO-SWS. Complete wavelength coverage of the mid-infrared wavelength range allows the full inventory of ices and silicates to be made, whereas their bandshapes and abundances are sensitive diagnostics of  heating events and the temperature history of the region.  


Figure 3.    Energy level diagram of the H2 and HD molecules, illustrating the fundamental pure rotational transitions


Gas-phase vibrational bands:  Fundamental vibrational transitions of key molecules such as H2O, CH4, C2H2, HCN and CO2 occur at 6.0, 7.7, 13.7, 14.0 and 15.0 m.  Symmetric molecules like CH4 and C2H2 have no dipole moment and cannot  be observed through rotational transitions at millimeter wavelengths. CO2 and H2O are so abundant in the Earth's atmosphere that they can only be detected from space. In addition to its importance in organic chemistry, CH4 is also an excellent diagnostic of the atmospheres of brown dwarfs and giant exo-planets (Burrows et al. 1997).  

Red-shifted bands:  For objects at high redshifts, the features discussed above are shifted further into the mid-infrared wavelength range, and the potential of NGST to detect these lines in distant galaxies is mostly limited by the available wavelength range rather than the sensitivity. For example, the strong [Ne II] 12.8 m line can be observed for z < 1.4, whereas the 6.2 m PAH band is available for z < 3.8. Strong near-infrared bands such as  Bracket-gamma can be searched up to z ~ 12, whereas  Halpha enters the mid-infrared range at z > 6.6.

The features and science discussed above and in section II.29 lead to the desired instrumental characteristics summarized in Table 2. A brief summary of the arguments for medium spectral resolution and integral field spectroscopy is given below.


Need for medium spectral resolution:  A resolving power R = lambda/delta(lambda) ~ 20003000 is required because:  

(i)                  the ro-vibrational bands of gas-phase molecules become apparent only at R >= 2000 (see Figure 4);

(ii)                sub-structure in the solid-state bands is an important diagnostic tool and requires at least R > 1000. The PAH 7.7 m band, which separates into at least 5 independent components at R = 2000, is a case in point. Also, many solid-state bands overlap at longer wavelengths and cannot be distinguished at R < 1500;

(iii)               detection of intrinsically weak and unresolved lines such as the H2 S(0) and S(1) lines on top of strong mid-infrared dust continuum emission requires R ~ 3000 to obtain sufficient line/continuum ratio; 

(iv)              the kinematics of galaxies (e.g., stellar mass determinations from the red-shifted CO 2.3 m band) and outflows from protostars require velocity resolutions of ~100 km s-1, corresponding to R ~ 3000.  


Figure 4.   ISO-SWS spectra toward the young stellar object Elias 29 in the region of the CO vibrational band at R = 400 and 2000. At R = 400, solid CO at 4.67 m is detected, but the intrinsically much narrower lines of gas-phase CO between 4.5 and 5 m only become visible at R = 2000, illustrating the need for medium resolution spectroscopy (Boogert 1999).


Case for integral field spectroscopy: While a conventional long-slit spectrometer is a minimum requirement for the NGST mid-infrared spectrometer, an integral field unit  (in which a spectrum is obtained at each pixel in a two-dimensional array) would provide a much more powerful instrument for addressing the scientific problems. From limited ISO imaging, it is clear that many objects which appear spherical or axi-symmetric at optical wavelengths are highly asymmetric and 'blobby' at mid-infrared wavelengths.  For example, in the nuclei of galaxies, PAHs and ionic lines have very different off-center distributions which would be missed in a single long-slit spectrometer setting (e.g. Le Floc'h et al. 2001). Nearby protoplanetary disks often have a clumpy asymmetric distribution (e.g., Greaves et al. 1998). The sizes of these objects are typically a few arcsec in radius. Thus, a long-slit spectrometer covering ~1' would be a very inefficient use of the array, whereas an integral field unit covering ~ 4'' x 4'' would be well tailored to the science.  



II.2.  Obscured star formation in galaxies  


The major science goal of NGST is to trace the star-formation history of the universe. It is now becoming clear that this is not a gradual process, but that most galaxies undergo bursts of intense star formation during their evolution. Extreme starbursts in distant high redshift galaxies may be responsible for much of the stellar populations of present-day galaxies. Remarkably, such starbursts are totally dust-enshrouded, and emit most their energy in the mid-infrared. Less extreme starbursts occur during the evolution of galaxies, often as the result of interactions with other galaxies.  During these episodes, galaxies evolve rapidly in stellar content, in gas content, in spectrophotometric properties, in metallicity, in luminosity, and often also in morphology. Starbursts are therefore a fundamental driver of the evolution of the galaxy population, and mid-infrared observations are a key to their study (Genzel & Cesarsky 2000).



Figure 5.  Various tracers in the NGC4038-4039 system (the "Antennae"). In the lower right frame the large-scale structure of the system is shown with X-ray (red) and H ~ I (black) contours superposed. The other 3 frames show the central 10 kpc of the merger. The underlying HST optical color picture reveals numerous young blue globular clusters, located mainly in an extended arm around the northern galaxy. However, the 15 m image from ISOCAM (contours in lower left panel, Mirabel et al.1998) shows that the most intense star formation occurs in an  optically obscured region where the disks interact. One remarkable cluster in the south-east region generates 15% of the total luminosity but is optically invisible. The present 15 m image has 6'' resolution, as have the CO and radio continuum images shown in the other panels. NGST will produce imaging spectroscopy at a factor of 10 better spatial resolution.


At low redshift, the most extreme starbursts are found in mergers of gas-rich galaxies, where the dissipative gas components quickly sink to the center of the potential well, resulting in an intense burst of star formation. The ultraluminous infrared galaxies discovered by IRAS are a manifestation of this phenomenon and approach quasar-like luminosities up to 1014 LSun , which is almost entirely emerging in the infrared regime. Based on these high luminosities, it has been suggested that powerful active nuclei and black holes  are born in the obscured cores of gas-rich mergers.    

A dramatic illustration of this effect is found in the nearby merging system NGC 4038-4039 (the "Antennae"). Here HST has produced spectacular images of the optical emission in the remnant host galaxies (see Figure 5), revealing numerous compact young star clusters. These sources represent young globular clusters formed as a result of the merger event. However, mid-infrared images obtained with the camera on ISO, ISOCAM, show that most of the star formation takes place in an optically obscured region where the two disks interact directly.  One spectacular star cluster in this region accounts for more than 15% of the total luminosity, yet is totally invisible optically (Mirabel et al. 1998).  High spatial resolution mid-infrared (1030 m) data will thus be needed to study the true, obscured, star formation properties of these rapidly evolving systems. Starburst phenomena occur even in our nearest neighbours (the Magellanic Clouds), as well as in other nearby galaxies such as M 33. Well-developed nuclear starbursts are found in very nearby spirals such as NGC 253, NGC 1808 and M 82. The availability of such local "laboratories" that can be studied in detail as templates for more distant sources makes this field an important science goal for NGST. The capabilities of NGST to probe the hot dust emission ideally complement those at longer wavelengths provided by Herschel and ALMA: whereas the hot dust probes violent (star formation) activity, ALMA and Herschel trace the cool dust, representing the pre-stellar material in galaxies.  

The key role of the NGST mid-infrared instrument  will be in spectroscopy in the 530 m region. The numerous bright forbidden fine-structure lines of abundant ions (see Table 3) are excellent diagnostics of local kinetic temperatures and gas densities, and of the temperature of the exciting radiation field. The latter is related to the range of stellar masses present in the galaxy, the so-called Initial Mass Function (IMF).  PAH features are also unique tracers of starbursts. Multi-line spectra can be used to separate the starburst and active nucleus as the power source. The potential of this approach has been demonstrated with (spatially unresolved) ISO spectra of nearby (up to z ~ 0.1) ultraluminous infrared galaxies (e.g., Genzel et al. 1998, Lutz et al. 1998, see Figure 6).  While this work forms one of the principal heritages of the ISO mission, it also highlights the most outstanding questions in this field. For instance,  many ultraluminous infrared galaxies contain an extreme starburst as well as an active nucleus (AGN) powered by a  black hole.  Does the occurrence of an extreme starburst trigger the formation of an AGN? How does this depend on the parameters of the starburst? Do globular clusters form in these immense compact star clusters?  ISO did not have the spatial resolution required to separate these components, but NGST can provide spatially resolved information and probe much more distant galaxies.



Figure 6.   Diagnostic diagram showing the extinction-corrected  25.9 m [O IV]/12.8 m [Ne II] line ratios as functions of the  strengths of the 7.7 m PAH feature. Starburst galaxies are marked as open triangles, ultraluminous galaxies as filled circles and AGNs as crossed triangles. In the left figure, the individual galaxies are labeled. In the right figure, the areas of the diagram dominated by star formation and by AGNs are indicated (Genzel et al. 1998)


These projects require spectroscopy at R ~ 3000 in the 1030 m range, at the spatial resolution provided by NGST.  Since the diagnostic lines are the major cooling lines for active massive star-forming regions, they are very luminous and detectable with NGST out to the redshifts where they shift out of the spectrometer passband.  For instance, the very moderate starburst galaxy M 82 would be detectable in the [NeII] 12.8 m line out to z = 1.3 (where the line shifts beyond 30 m) at the 20-sigma level in only 1 hour for a 6.5m NGST with a 50 K focal plane. The Milky Way would be detectable out to the same redshift to the 5-sigma level in 1 hour under the same assumptions. Spatially, a 3 kpc starburst ring surrounding an AGN, such as that of the well-known nearby active galaxy NGC 1068 could be separated from the nucleus for galaxies out to 50 Mpc with NGST.



II.3.   The high-redshift universe


Without observing the universe at wavelengths above 5 m, NGST would miss much of the cosmic star formation history. Several independent observations show that most of the star formation taking place in early, distant galaxies is dust-enshrouded (Blain et al. 1999, Chary & Elbaz 2001).  A result of paramount importance is the spectrum of the integrated extragalactic background, where the power in the infrared part due to re-radiation of absorbed starlight by dust exceeds the direct optical starlight by about a factor of 2 (Dwek et al. 1998, Gispert et al. 2000).  Because at low redshifts the integrated infrared light is only  ~30% of the optical light, this implies that the universe was much more obscured at high redshift than it is now. Quantitatively, an increase in infrared energy production out to z ~ 2 by a factor 30 is implied, and the infrared energy production should remain constant from z = 2 out to at least z = 5. Thus, star formation rates at high z derived from optical data alone will be greatly underestimated.  

ISOCAM has lifted the tip of the veil in this spectral region.  Deep ISOCAM images of the Hubble Deep Field South   one of the deepest images  ever taken with HST reveal typically 60 times fewer galaxies per unit area than HST, but these few galaxies produce at least a third of the integrated optical light (see Figure 7).  Clear evidence for an increase in the infrared number counts is seen at the faint end, indicative of this new dusty population of galaxies at z > 0.2 (Elbaz et al. 1999).  Interestingly, most ISOCAM-detected galaxies have optical/near-infrared counterparts that have totally normal optical colors; thus, it would not be possible to select these galaxies from optical/near-infrared data alone.


  Figure 7.   ISOCAM 15 m image (color, 7' x 7') of the Hubble Deep Field South (HDF-S) WFPC2 field, with contours of the 6.7 m image superposed (Oliver et al. 2001). This is one of the deepest mid-infrared images to date. The objects seen at mid-infrared wavelengths, in particular that indicated with the white arraow,  are inconspicuous on the optical HDF-S data. The objects studied spectroscopically so far are classified as dusty starburst galaxies with redshifts up to z ~1.3. NGST will take the study of distant starburst galaxies out to much higher z.


A key NGST program beyond 5 m will be the identification of this new population of galaxies accounting for most of the cosmic star formation through their dust emission. The mid-infrared spectra of dusty galaxies are dominated by  deep silicate and ice absorption features and broad emission features attributed to PAHs, which are detectable out to high z. A modest 1 x 1011 LSun starburst galaxy would be detectable out to z ~ 3.5 in the 6.7 m band at 10 sigma  in 3 hours.  

For a 6.5m NGST, the confusion limit at 25 m is at 5 x 105 sources per deg2, which is very close to the optical source density in the Hubble Deep Field: 1 x 106 per deg2. Current infrared source population models indicate that this source density is reached at the 2 Jy flux density level at lambda = 25 m, which takes approximately 100 hours of NGST time. While SIRTF will make efficient large area surveys, its impact will be significantly limited by confusion in the 8 times larger beam.   

Mid-infrared observations also offer the unique opportunity of sampling the rest-frame 2 m region at z > 2, which is dominated by an evolved  stellar population with mean age >50100 Myr that cannot be isolated at shorter wavelengths, but which would trace a previous episode of star formation. For example, the masses of the so-called Lyman break-galaxies at z ~ 3 are very uncertain since it is not known whether or not an older underlying stellar population is present (Papovich et al. 2001).  NGST will resolve the ambiguity between age and reddening which plagues the analysis of near-infrared colors.  Moreover,  spectroscopy of the CO band heads at rest-frame 2.3 m will provide a measurement of the velocity dispersions and hence dynamical masses of early spheroids. This will take the study of the cosmic star formation history from the measurement of light to the measurement of mass, which will be a major breakthrough. Medium spectral resolution observations at R ~ 3000 are essential for this project.  

The lambda > 5 m region also gives access to the most important nebular lines at the highest redshifts, in particular Halpha at z > 6.6. This redshift regime corresponds to the epoch of recombination, where cosmic first light ionizes the gaseous universe. The Halpha line is crucially important since (unlike Lyalpha) it is not affected by resonant scattering and therefore forms a direct and quantitative probe of the ionizing flux. The star formation rate of the first galaxies can thus be directly measured using redshifted Halpha , and any dust extinction can be corrected using the Halpha /Hbeta  line ratio. Since the epoch of cosmic first light is one of the cornerstone science targets of the NGST, this application is of key importance to the mission (Haiman & Loeb 1998,  Ciardi & Ferrara 2001).



II.4.   Star formation


The processes by which stars, protoplanetary disks and planets are formed remain poorly understood. NGST will play a vital role  in this area, in concert with complementary facilities such as ALMA and Herschel.  The tremendous sensitivity of NGST at mid-infrared wavelengths is essential because protostars and disks are relatively cool with huge dust extinctions, so that they are impossible to penetrate at shorter wavelengths. The high spatial resolution of NGST is needed to zoom in on protostellar disk formation, to image gaps in disks around more mature pre-main sequence stars and to detect faint brown dwarfs and extrasolar giant planets close to their bright parent stars. Because star- and planet formation is accompanied by huge changes in the physical conditions, with densities ranging 104 to 1013 cm-3 and temperatures from 10 to 10,000 K, there are a wealth of atomic and molecular lines available with which to unravel the physical structure and evolution. Moreover, these lines can be used to trace the chemical composition of gas and dust through the various stages of the star formation process, thus providing an inventory of the building blocks available for new solar systems (Figure 8).  


Figure 8.    The ISO-SWS midinfrared spectra  of newly-formed stars in different stages of formation and circumstellar disks (Gibb et al. 2000, van den Ancker et al. 2000, Malfait et al. 1998). From  top to bottom in a rough evolutionary sequence   the spectra change from dominated by solid state absorption features and gas (shock) emission lines, to featureless, to PAH features and PDR lines, to amorphous and crystalline silicates (F = Forsterite) with H ~ I recombination lines. The ISO-SWS spectrum of comet Hale-Bopp is shown for comparison (Crovisier et al. 1997).


Deeply-embedded protostars:  Mid-infrared observations with NGST will be particularly powerful to provide insight into the physical processes occurring in the deeply embedded protostellar phase when the star is still being assembled through accretion of material from the circumstellar disk. The fragmentation of the collapsing cloud into binary or multiple star systems and the formation of the disk itself are key questions, which can likely only be addressed properly through combined NGST mid-infrared and ALMA data.  Mid-infrared continuum images at >20 m probe the warm part of the accretion disk and the inner envelope,  constraining the geometry and the elusive mass infall rate (see simulations in Serabyn et al. 1999).  Imaging in different spectral features such as the H2 pure rotational lines and the [S I] 25.2 m line probe the physics of the accretion shock at the disk surface, as well as the interaction of the outflow with the inner envelope as it starts to clear the surroundings (van den Ancker et al. 2000, Figure 8).  Periodic structures in the outflow jets may be used to trace the recent accretion history of the central protostar.   The PAH features track the importance of ultraviolet radiation in dispersing the envelope. Together, such data can trace the protostellar evolution from the earliest collapse to the phase where the young stars emerge from their natal cocoons.  

The high sensitivity and spatial resolution of NGST are essential for these projects, since the youngest protostars have mid-infrared continuum fluxes of less than a mJy (see Fig. 2  in Andr et al. 2000) and the disks have sizes of at most a few arcsec (a few hundred AU) for typical distances of 150300 pc.  The SIRTF legacy program of Evans et al. (2001) (including van Dishoeck as co-I) will detect all protostars with luminosities down to 1.5 x  10-3 LSun as well as young stars and substellar objects down to 5 MJ in five nearby molecular clouds, providing a prime database for NGST high-resolution imaging and spectroscopy follow-up.  

Massive young stars:  More distant, massive protostars are also prime targets for NGST mid-infrared observations. Little is known about the earliest stages of massive stars, for example whether their formation is accompanied by circumstellar disks as in the case of low-mass stars. Also, traditional models of massive star formation through the collapse of a single cloud core are being challenged by alternative models of merging of lower-mass stars in the central potential of very dense clusters (Stahler et al. 2000). Since this latter mechanism would occur in the deeply embedded phase, only sensitive high spatial resolution NGST observations, combined with ALMA data, will be able to resolve such clusters of  ~10100 low-mass stars crammed into a ~1000 AU region out to a few kpc.  Once formed, the massive young stars enter the ultra-compact H II region phase, in which they ionize the surrounding gas. A circumstellar disk consisting of hot dust and gas (~4000 K) may still exist in regions shielded from the intense ultraviolet radiation of the central star (Hanson et al. 1997, Lenorzer et al. 2001). High-spatial resolution infrared spectra are required to study the distribution and physical nature of this hot dust and gas and confirm the presence of disks.



II.5.   Protoplanetary disks


One of the most exciting developments in astrophysics in the last decade has been the definite detection of extrasolar planets around nearby stars via the radial velocity technique (Mayor & Queloz 1995, Marcy& Butler 1996). More than 60 of these exo-planets are now known, showing that planet formation is common and reviving age-old questions about their formation.  The planets are thought to have originated in the disks or 'pre-solar nebulae' around young stars, first postulated by Kant in 1755 for the origin of our solar system.  The existence of such disks was inferred from pioneering infrared and millimeter observations in the early 1990's and it is now commonly accepted that more than 50% of young stars have disks (see Beckwith & Sargent 1996 for a review).  Disks are also beautifully seen as silhouettes against a bright background in HST optical (e.g., McCaughrean & O'Dell 1996, Burrows et al. 1996), and near-infrared images (e.g., Padgett et al. 1999).  The sizes of the disks are typically a few hundred AU, comparable to that of our own solar system.  

Toward the end of the star formation process, the accretion of matter onto the star through the disk stops and the dust in the disk begins to coagulate to form larger and larger bodies, becoming the seeds for planetesimal formation. Once these are self-gravitating, they accrete even faster and may develop into proto-planets.  However, a detailed understanding of the processes by which disks turn into planets, the time scales involved, the frequency of planet formation and its dependence on external influences, and the evolution and dissipation of the primordial gas and dust in the disk, is still lacking. The fact that none of the known exo-planetary systems resembles our own solar system indicates that existing theories need to be revised (e.g.  Artymowicz 2001). The combination of NGST mid-infrared and ALMA submillimeter data will allow major steps forward in our understanding of protoplanetary disk evolution.  

Young massive disks:  The gas-rich disks seen around pre-main sequence T Tauri and Herbig Ae stars with ages of a few Myr have gas + dust masses of 0.01 MSun , similar to that of our primitive solar nebula.  Once accretion stops and the dust starts to coagulate, both the geometry and composition of the disk undergo substantial changes.  These changes can be studied best in the mid-infrared spectral region, since at these wavelengths the inner disk, where planet formation is expected to occur, dominates the spectrum. In particular, changes in dust scale height affect the strength of the mid-infrared flux emerging from the disk, while structural modifications in the dust grains result in mineral formation traced by mid-infrared spectra (e.g. Chiang et al. 2001, Meeus et al. 2001, Dullemond et al. 2002, see section II.7).  The possibility of obtaining spatially resolved images with NGST and ALMA down to ~10 AU will allow not only the dust settling, but also the relative settling of the dust versus the gas in gas-rich disks to be determined and the earliest stages of planetesimal formation to be followed up to roughly millimeter sizes. Indirect evidence for planet formation will be provided by observations of gaps in disks, where an unseen giant planet has cleared out a ring of material. The power of mid-infrared data for these studies was demonstrated by Koerner et al. (1998) for HR4796A. For the brightest systems in the nearest star-forming regions, MIDI on the VLTI can zoom in even closer and detect the warm material in the 110 AU range.  

Most disks have been studied through their dust emission, but 99% of the mass is in H2 gas. Not only does this gas affect the dynamics of the dust, but it is also the principal ingredient for giant Jovian planet formation. It has recently become clear that CO is not a good tracer of the gas in disks due to the combined effects of photodissociation in the surface layers and freeze out in the cold midplane (van Zadelhoff et al. 2001, Aikawa et al. 2001).  In contrast, H2 traces the bulk of the warm gas at T > 50 K directly. The ISO-SWS provided the first opportunity to search for the principal H2  J = 20 S(0) and 31 S(1) lines at 28 and 17 m,  and detected lines not only from disks around young pre-main sequence stars (Thi et al. 1999), but also tentatively from disks around 1020 Myr old stars such as Beta Pictoris (Thi et al. 2001a,b, see Figure 9).  This would indicate that these latter objects are better thought of as the final stages of gas-rich accretion disks. The time scale over which this gas clearing occurs and the mechanisms by which it operates are critical to both planetary formation and migration models.  Currently there is still no consensus whether giant planets originate through the formation of a rock-ice core followed by gravitational gas accretion or by instabilities in the outer disk and fragmentation.  Combined spatially resolved observations of both the dust and H2 as functions of age over the 1100 Myr range will be powerful tools to distinguish between these models.  


  Figure 9.   ISO-SWS observations of H2 toward the 'debris' disks  around Beta Pictoris, 49 Ceti and HD 135344 (Thi et al. 2001a,b).  The Beta Pictoris 1.25 m  scattered light image is based on Mouillet et al. (1997). The 20 m thermal emission image of 49 Ceti is from Koerner (priv. comm.) and that at 12 m of HD 135344 from Blake & Kessler (priv. comm.).  

NGST will be sensitive to gas with masses of 10-310-7  MSun   at Tgas = 50200 K (see Figure 12 of van Dishoeck 2000). The high spectral (R ~ 3000) and spatial resolution of the NGST mid-infrared instrument is essential for this project: even though SIRTF will attempt to observe H2 from disks, the low line/continuum ratio at R = 600 will prevent detection in most cases. 


Figure 10.   Left: Flux density distributions for model debris disks around stars of various spectral types in nearby star-forming regions. The models assume 0.1 MMoon of 30-m-sized dust grains in a zone extending from 30 to 60 AU radius, using the Draine & Lee (1984) dust properties.   The NGST 5-sigma sensitivity in 0.7 hr is indicated. Right: Flux density distributions for zodiacal dust (1 m-sized grains between 35 AU) around a G-type star at 25 pc. The masses are 10-5, 3 x10-5, 10-4 and 5 x 10-4 MMoon (Dominik, priv. comm.).


Debris disks:  Current near-infrared surveys for excess emission indicate that the massive disks disappear in a few Myr (Haisch et al. 2001). However, these data are only sensitive to very warm dust close to the star.  Longer wavelength observations are needed to trace the evolution of the bulk of the dust beyond the classical T Tauri stage. The SIRTF guaranteed and legacy programs of Meyer et al. and Evans et al. will survey several hundreds of young stars with ages ranging from a few to several hundred Myr, including weak-line T Tauri stars associated with the closest star-forming clouds (ages 120 Myr), X-ray bright young stars in the solar neighborhood for which accurate distances are known from Hipparcos (ages ~10100 Myr), and young main sequence stars in open clusters with age >30 Myr.  Together, they will produce a phenomenal database for subsequent NGST spectral imaging of disk evolution around solar-type stars in the planet-forming phase. NGST can detect disks with only a fraction of a lunar mass throughout the solar neighborhood (see Figure 10).   

Eventually, the original gas and dust in the disk will have dissipated and/or coagulated into larger bodies.  The disk will then continue to evolve collisionally over a period of typically 400 Myr, resulting in the destruction of larger bodies and the production of new so-called 'debris' dust, first detected by IRAS (Aumann et al. 1984) and surveyed by ISO (Habing et al. 1999).  A similar period of collisions occurred in the early solar system, as evidenced from, for example, the formation of the Moon and the impact record of asteroids on the Moon. The zodiacal dust in the inner regions of the solar system is the result of fairly recent collisions of asteroids in the asteroid belt, whereas a colder component resides in the Kuiper belt.  Such exo-Kuiper belts have been imaged in a few cases at submillimeter wavelengths by SCUBA on the JCMT (Holland et al. 1998, Greaves et al. 1998), but those data are not sensitive to the warm inner solar system dust.  Even though the zodiacal grains compose only a very small fraction of the mass in our solar system (<10-10, they intercept and re-radiate effectively a fraction ~10-7 of the Sun's energy and therefore account for the lion's share of its far-infrared radiation.  Thus, even 4.5 Gyr after its formation, zodiacal dust provides the more readily detectable indirect evidence of a solar system like our own.  

NGST will have the sensitivity to study the composition and geometry of exo-zodiacal dust near other stars in great detail (Figure 10).   IRAS and ISO showed that ~15% of stars have more than 100 x  the amount of cold dust in our own solar system, but these statistics are based on a small sample and limited entirely to  the more massive A-type stars. NGST would allow solar-system levels of zodiacal dust to be detected around Sun-like stars out to distances of at least 25 pc. More importantly, its high spatial resolution will allow the disk to be distinguished from the stellar photospheric emission and the surrounding interstellar 'cirrus' emission. Knowledge about exo-zodiacal mid-infrared dust emission will also be essential for planning future missions such as IRSI-DARWIN and TPF, since a strong exo-zodiacal light component will severely restrict the possibilities to detect Earth-like planets in habitable zones.  



II.6.   Giant exo-planets and brown dwarfs


The primary goal for future space missions such as ESA's DARWIN/IRSI and NASA's TPF in the middle of the next decade is the direct detection of Earth-like planets and the probing of their atmospheres for signs of life. NGST will form an important bridge toward this aim, not only by enhancing our understanding of the mechanisms by which planets form from disks (see section II.5), but also by allowing detection and characterization of extrasolar giant planets and their atmospheres.  

The spatial resolution and sensitivity of NGST will allow a direct spectroscopic detection of Jupiter-like giant planets in wide (>5 AU) orbits around the nearest stars (<10 pc). These observations are best done  near 5 m, where the molecular opacity in the outer atmosphere of a gas giant is low so that warm thermal emission can emerge.  Searches for these objects may be carried out with a coronograph in the NGST near-infrared camera, but a simple coronographic option in the mid-infrared camera is being considered as well. Once found, follow-up mid-infrared spectroscopy can characterize the planetary atmospheres and study their chemistry as a function of mass, age and temperature (Figure 11). With large enough statistics, such observations  may also give insight into the formation history and possible migration of giant gas-rich planets. For instance, planets formed late in the evolution of the protoplanetary disk are expected to have less gas and a different composition.  


Figure 11.    Model atmospheres of a 5 MJ exo-solar giant planet at a distance of 10 pc at different ages. Note the strong spectral evolution with time in the mid-infrared part of the spectrum. Only NGST has the sensitivity to follow this evolution for Jupiter-type objects (based on Burrows et al. 1997).


NGST will also be enormously powerful for studying other substellar objects such as brown dwarfs with masses in the 0.010.07 MSun range. Thanks to the 2MASS, DENIS and SLOAN surveys, more than a hundred of these objects have now been discovered (e.g., Kirkpatrick 2001), and ground-based surveys with large optical telescopes will pin down the initial mass function in this regime in young clusters.  However, several questions can only be addressed by  space-based mid-infrared imaging and spectroscopy. For example, are young brown dwarfs surrounded by their own disks during formation, i.e., do they indeed form as stars do, or do they form as companions to stars? Also, are their atmospheres dusty, i.e., can we see 10 m silicate emission? Theoretical models predict significant changes in the atmospheric composition with age and mass, especially around 8 m due to NH3 formation (Burrows & Sharp 1999). SIRTF will provide a first glimpse of these spectra, but lacks the sensitivity and spectral resolution for a proper survey.



II.7.   Tracing the astrochemical evolution


Generations of low- and high-mass stars have converted the primordial hydrogen and helium into successively heavier elements including carbon, oxygen and nitrogen (see section II.8). These elements are mixed into the diffuse and dense interstellar medium and may eventually end up in the envelopes and disks around young stars. Tracing the lifecycle of the gas and dust from the pre-stellar cores to their eventual incorporation into planetary systems where they may be the stepping stone toward prebiotic molecules, is a major goal of astrochemistry and the new discipline of astrobiology (see reviews by van Dishoeck & Blake 1998, Waters 2000, Ehrenfreund & Charnley 2000, van Dishoeck & Tielens 2001).  

In the cold quiescent clouds prior to star formation, gas-phase molecules collide with the grains and stick, forming icy mantles around the silicate cores. Indeed, this freeze-out is predicted to be so efficient that it has been a puzzle for decades why any gas-phase molecules are seen at all in cold regions.  Recent indirect evidence based on near-infrared star counts and millimeter data suggests that the amount of depletion inside dense cores can be substantial, with more than 90% of the heavy elements (including oxygen and carbon) frozen out onto the grains (Lada et al. 1999, Kramer et al. 1999). NGST will be the first instrument capable of probing this depletion process directly, through near- and mid-infrared observations of solid-state absorption bands toward hundreds of background stars located behind dark clouds, providing "maps" of the freeze-out process.    

The ices in the cold outer envelopes around protostars can be observed through medium-resolution absorption spectroscopy.  In the earliest collapse stages, more of the heavy elements are frozen out and the chemical composition is modified through grain-surface reactions. As the surroudings are heated by the protostar, the ices gradually evaporate back into the gas phase (Figure 4).  Figure 8 includes the ISO-SWS mid-infrared spectrum toward the embedded massive protostar W33A, which is particularly rich in various (organic) features (Gibb et al. 2000). ISO has been able to obtain such high-quality spectra for only a dozen massive protostars with luminosities of 103105 LSun , but information on low-mass protostars is lacking almost entirely. Since SIRTF lacks the required spectral resolution, only NGST will be able to perform a complete census of solid-state material as it enters the planet-forming disks around solar-type stars.   

The ice and gas-phase data also reveal the thermal and irradiation history of the protostellar envelope. The solid CO2 bending mode at 15 m is a particularly sensitive diagnostic of the ice environment, with a characteristic double-peaked structure appearing upon heating  by the young star (Ehrenfreund et al. 1998, Gerakines et al. 1999).  Other features, such as the OCN band at 4.62 m and the unidentified 6.85 feature, are sensitive to ultraviolet irradiation (Schutte & Greenberg 1997, Schutte et al. 2001). Since the changes in the band profiles are irreversible, they record the cumulative events experienced during the protostellar phase such as FU-Orionis-type outbursts related to disk accretion.  The analysis of the NGST data will greatly benefit from the involvement of the Raymond & Beverly Sackler/NOVA Laboratory for Astrophysics at Leiden.  

Part of the gas and dust from the protostellar envelope will be incorporated into the circumstellar disk, where it can be further processed by ultraviolet radiation, X-rays and thermal processes.  In the cold midplane, most molecules are expected to be frozen out on the grains and NGST will have the sensitivity to probe these ices in absorption against the star in edge-on geometries. Emission from gas-phase molecules such as H2, CO, CH4 and C2H2 from the warm inner regions can be searched for at the highest spectra resolution.  

The dominant spectral emission features from disks are expected to be due to PAHs, ices and silicates, both in amorphous and crystalline form. These dust particles are the building blocks of planets, asteroids and comets. Our solar system has a very rich mineralogy which holds the formation record of its solid bodies; examples are the olivines (one of the major constituents of the Earth' crust), silica, iron sulfide and carbonates. The latter material is linked to the presence of liquid water on large parent bodies.  The minerals are believed to have played a key role in the formation of life on Earth through their ordered lattice structure. In contrast,  interstellar clouds have so far revealed only a limited number of fairly simple materials such as amorphous silicates and carbon, demonstrating that substantial processing of the dust must occur as solar system planetesimals are formed. Grains can stick together, they can be heated by electric current discharges or by stellar radiation which alters their lattice structure, and gas-solid reactions can modify their chemical composition. Mid-infrared spectroscopy of protoplanetary disks with NGST can probe these changes and trace the formation of planetesimals and other solid bodies that would be undetectable directly due to their small radiating surfaces. In combination  with laboratory light scattering experiments of cosmic dust analogues, both the composition and size of the dust particles can be established. Such light scattering measurements are being performed using the University of Amsterdam-NOVA/FOM-AMOLF light scattering experiment.  

The ISO-SWS has allowed a first glimpse of the rich mid-infrared spectra of disks around pre-main sequence stars. Indeed, one of the major results is that solid-state evolution of grain minerals is observed (e.g., Malfait et al. 1998, Meeus et al. 2001). Some objects show only amorphous dust emission, whereas others have clear signatures of crystalline silicates and/or PAHs similar to those found in comets (see Figure 8 and front cover).  These variations appear related to the processes in grain coagulation, settling and thermal annealing in the disk (Bouwman et al. 2001). The ISO-SWS  did not have the sensitivity to probe disks around solar-type stars. SIRTF will provide the first data on lower-mass objects, but will be spatially unresolved and has the sensitivity to do only R = 100 spectroscopy on the more evolved debris disks. Only NGST will be able to follow the entire dust evolution in the 1600 Myr range for Sun-like stars.  


Figure 12.    The 12-17 m spectrum of CRL 618  obtained by the SWS on ISO illustrating the molecular complexity of AGB ejecta.   The model (red) includes the ro-vibrational transitions of acetylenic chains (C2H2 and C4H2) and their nitrile derivatives (HC3N, HC5N), hydrogen cyanide, and benzene (C6H6) (Cernicharo et al. 2001).}



II.8.   The death of low mass stars


Low mass stars such as the Sun (<8 MSun) end their lives by returning most of their mass to the interstellar medium in the form of a gentle wind (velocity ~10 km s-1) while evolving on the so-called Asymptotic Giant Branch (AGB).  At that point, all of their hydrogen in the core has burned to helium during the main sequence stage of evolution, and subsequently to carbon and oxygen during the horizontal branch stage of evolution.  The central core is then essentially a white dwarf surrounded by a large convective envelope. While these stars derive their energy from burning alternatingly hydrogen and helium in a shell around the core, stellar evolution during this stage is driven by mass loss from the star rather than nuclear burning.  Eventually, when the whole envelope is lost, the central white dwarf will evolve rapidly to the blue and ionize and heat the previously ejected material.  This will set it aglow, forming a planetary nebula.  The dust and gas in the ejecta absorb much of the visible and near-infrared radiation from the star and reradiate it in the mid-infrared, so NGST is particularly well suited to study the death struggle of stars.  

Studies of the mass loss during the AGB phase are important not only to study stellar evolution, but also because these stars dominate the mass budget of the interstellar medium.  The gas in AGB winds consists largely of molecules ranging from the mundane H2 and CO to the exotic HC9N and PAHs. The ejecta from these stars are a major source of carbon as well as the source of all the s-process elements in the interstellar medium.  Most of the heavy elements are returned to the interstellar medium in the form of small dust grains, which then form the building blocks of new stars and planets. Indeed, even in the heavily processed environment of the inner solar system, relic stardust grains have been recovered (Bernatowicz et al. 1996, Tielens 2001).    

During this mass loss phase, the mid-infrared spectra of AGB stars are dominated by the vibrational signatures of the surrounding dust and gas.  As an example, Figure 12  shows a small portion of the infrared spectrum of the heavily enshrouded source, CRL 618, which has just left the AGB, illustrating the molecular richness of the envelope.  The composition of the dust is set to a large extent by the elemental composition of the ejecta.  If the star is oxygen-rich, silicates and oxides are formed.  In carbon-rich ejecta, carbonaceous materials, such as carbides and hydrogenated amorphous carbon grains condense.  The dust formed also depends on the density and temperature in the condensation zone.  Hence, AGB stars and their descendants show a bewildering zoo of dust types.  Figure 13 illustrates the spectral richness of stardust.  While most spectra follow this dichotomy of O- versus C-rich dust, an increasing number of objects shows evidence for mixed chemistry.  It it generally thought that the O-rich material represents dust formed during an earlier evolutionary phase of the star, possibly stored in a long-lived circumstellar disk.   


Figure 13.   ISO-SWS spectra of a variety of objects illustrating the spectral diversity of stardust formed in the ejecta of AGB stars (Molster 2000).

After the AGB phase, as the central object evolves to the blue, the increased far-ultraviolet radiation and the strong outflows driven by its powerful winds process the envelope so that the spectral signatures of the dust and gas change.  At this point, the infrared features of PAH molecules start to dominate the near- and mid-infrared spectra of C-rich objects.  Many O-rich objects start to show evidence for crystalline silicates in their spectra (Figure 13).

Mid-infrared spectroscopy with NGST will revolutionize our understanding of the composition and evolution of the ejecta of AGB stars.  Ground-based studies are limited to two atmospheric windows which miss much of the spectral richness.  ISO studies of (post-)AGB objects have been limited to bright objects in the solar neighborhood, whereas SIRTF lacks spectral and spatial resolution.  The mid-infrared instrument on NGST will be able to survey a large sample of late-type objects in galaxies from the local group out to the Virgo cluster. In this way a census of stardust can be built up and its dependence on the physical conditions (effective temperature, mass loss rate) including metallicity can be determined.  These spectral characteristics can then also be compared to those of the interstellar medium in the host galaxy to determine their interrelationship and investigate the dust mass balance of the galaxy.  

High-mass stars evolve much more quickly and end their lives through a spectacular supernova explosion, expelling many heavy elements. NGST will have the sensitivity and spatial resolution to observe the in-situ formation of dust in knots with greatly different elemental abundances, such as the famous SN 1987A in the Large Magellanic Cloud.



II.9.   Outer solar system: comets and KBOs


Comets are the most primitive objects in the solar system, formed in the outer regions of the solar nebula (see reviews by Irvine et al. 2000, Ehrenfreund & Charnley 2000).  They therefore contain a precious record of the conditions that prevailed in the primitive solar nebula. These initial conditions are important constraints on models that describe the evolution of the solar nebula and the formation of the planets. So far, only limited information about the gas-phase and solid composition of comets has been obtained from mid-infrared observations, because of lack of instrumentation.  

Infrared observations of comet Hale-Bopp with ISO provided CO2 outgassing rates at different heliocentric distances, as well as data on the silicate emission (Crovisier et al. 1997).  The dust shows (sometimes strong) evidence for substantial processing, and the mineral formation may point to complex radial mixing or heating processes in the solar nebula.  New ground-based high spectral resolution near-infrared instruments are starting to provide information on minor organic species in comets, and it is becoming apparent that chemical differentiation exists amongst the comet population. For example, recent data from comet C/1999 S4 (LINEAR) show a greatly different composition compared with that of comets Halley, Hyakutake and Hale-Bopp (Mumma et al. 2001).  Understanding this differentiation can put constraints on the place of cometary origin and on the chemical history of the organic material which formed them.  

The NGST mid-infrared instrument will be able to observe comets when still far from perihelion when most volatiles are still in the cometary nucleus, and it will be able to monitor the development of the cometary tail as the temperature of the comet increases and dust and gas are released. It will have the sensitivity to obtain spectra of comets whose emission is a factor of 1000 times weaker than that of comet Hale-Bopp; several such comets will be visible to NGST each year.  

The outer regions of the solar system are densely populated by a large number of bodies orbiting the Sun beyond Neptune.  Up to 105 objects are estimated to orbit in a distance of 3050 AU from the Sun. Currently more than 400 Kuiper belt objects (KBO's) are detected and most of them have a diameter between 100 and 1200 km.  These size estimates of KBO's are based on an assumed albedo of 0.04 (as for cometary nuclei) and those estimates may be considerably in error. The combination of optical data with simultaneous measurements of the thermal emission in the mid-infrared can determine both their size and albedo (Jewitt et al. 2001).  



IV.  The NGST mid-infrared instrument

The information below is approximately correct, but the design is still evolving (August 2002).

The mid-infrared science goals described in section II call for both high spatial resolution imaging with as wide a field as practicable, and for moderate resolution spectroscopy. In addition, a coronographic capability is warranted for observations of exo-solar planets. The top-level requirements for the NGST mid-infrared instrument have been investigated by several groups independently, most recently by the international NGST Mid-Infrared Steering Committee (MISC) in July 2001, and agree excellently among each other. Thus, there is widespread consensus within the astronomical community on the desired capabilities of the instrument. The MISC recommendations have been approved by NASA and ESA and are summarized in Table 2 , and they form the baseline for the phase A  design study started in fall 2001.  

Since 1999, various concept studies for the instrument have been carried out sponsored by ESA and NASA (see reports by Wright et al. 1999, Serabyn et al. 1999, MISC report). The earlier studies had 46 modules in the instrument, with two modules each for the camera and spectroscopy channels to accomodate different plate scales and cope with the different background levels. The most recent studies focussed on simplification and lightweighting of the instrument; the minimum design that satisfies all the science goals has just one module each for the camera and spectrometer, plus a module containing the common fore-optics, with a minimum number of mechanisms. For the spectroscopy channel, both traditional long-slit designs and more innovative integral-field units (IFU) have been considered.  From the science point of view, an IFU is preferred and is endorsed by the MISC report.  

An optical pre-design study of this system has been made, primarily at Service d'Astrophysique (SAP, Saclay, France) for the camera subsystem and at the UK Astronomy Technology Centre (ATC, Royal Observatory, Edingburgh, UK) for the spectrograph.  All optical elements (except filters and grisms) are reflective. 


Figure 14.   Spectrometer concept with 4 wavelength channels, 2 detectors.  Most of the Dutch contribution is in the centre (slits to cameras).

The studies and the current design have drawn heavily on the European experience in building mid-infrared instruments for ISO, Herschel and for ground-based telescopes such as the VLT (VISIR), UKIRT (image-slicer) and Gemini (MICHELLE). In fact, most of the elements have already been incorporated in a similar form in other instruments. Thus, the risk in the design and development is minimized in this approach.

The heart of the instrument is formed by the detectors. Here the NGST project profits heavily from the detector development for SIRTF, which has produced 256 x 256 pixel formats with high quantum efficiency and encouragingly low dark currents. The NASA-NGST project has  funded development to extend the detector device format to 1024 x 1024 pixels, with the aim to also bring down the read-noise and dark currents to levels such that NGST is background limited at low spectral resolution as short as 6 m. Thanks to these rapid developments in mid-infrared detector arrays, the enormous jump forward in sensitivity and observing speed is possible; indeed, there are few other wavelength ranges where such large improvements can still be achieved.  The detectors for the camera and spectrometer will be procured by the US part of the consortium.  

For optimal performance and reduction of thermal background emission, the optics should be as cold as possible ( <24 K), with an enclosure temperature of ~18 K and the detectors themselves cooled to < 7.5 K.  This will be accomplished by a solid-hydrogen cryostat, to be procured by ESA. Fig 15 is an overview of the cryogenic requirements.              


Figure 15.    Cryogenic requirements for MIRI.  "FPA" = Focal Plane Assembly, i.e. the detectors.



VII.  Project organization and management


This chapter is being implemented.




Aikawa, Y., van Zadelhoff, G.J., van Dishoeck, E.F., Herbst, E.  2001, A &A, submitted

Andr, P., Ward-Thompson, D., Barsony, M.  2000, in Protostars  & Planets IV, eds. V.G. Mannings et al. (Univ.  Arizona), p. 59

Artymowicz, P.  2001, in  Planetary systems in the universe, IAU Symposium 202, eds.  A.  Penny et al.  (San Francisco: ASP), in press

Aumann, H.H., Gillett, F.C., Beichman, C.A., et al.  1984, ApJ, 278, L23

Bacon, R., Copin, Y., Monnet, G., et al.  2001, MNRAS, 326, 23

Beckwith, S.V.W., Sargent, A.I.  1996, Nature, 383, 139

Bly, P., et al.  1998, NGST monograph no. 3 (Washington: GSFC)

Bernatowicz, T.J., Cowsik, R, Gibbons, P.C.  et al.  1996, ApJ, 472, 760

Bertoldi, F., Timmermann, R., Rosenthal, D., Drapatz, S., Wright, C.M.  1999, A &A, 346, 267 

Blain, A.W., Kneib, J.-P., Ivison, R.J., Smail, I.  1999, ApJ, 512, L87

Boogert, A.C.A.  1999, PhD Thesis, Univ. of Groningen

Bouwman, J., Meeus, G., de Koter, A., Hony, S., Dominik, C., Waters, L.B.F.M.  2001, A &A, in press

Burrows, A., Sharp, C.M.  1999, ApJ, 512, 843

Burrows, A., Marley, M., Hubbard, W.B., et al.  1997,  ApJ, 491, 856

Burrows, C.J., Stapelfeldt, K.R., Watson, A.M. et al.  1996, ApJ, 473, 437

Cernicharo, J., Heras, A.M., Tielens, A.G.G.M., Pardo, J.R., Herpin, F., Gulin, M., Waters, L.B.F.M.  2001, ApJ, 546, L123

Chary, R., Elbaz, D.  2001, ApJ, 556, 562

Chiang, E.I., Joung, M.K., Creech-Eakman, M.J., Qi, C., Kesser, J.E., Blake, G.A., van Dishoeck, E.F.  2001, ApJ, 547, 1077

Ciardi, B., Ferrara, A. 2001, MNRAS in press (astro-ph/0005461)

Crovisier, J., Leech, K., Bockele-Morvan, D., et al.  1997, Science, 275, 1904 

Draine, B.T., Bertoldi, F.  1999, in 'The Universe as seen by ISO',  ESA SP-427, p.  553

Draine, B.T., Lee, H.M.  1984, ApJ, 285, 89

Dullemond, C., Dominik, C., Natta, A.  2001, ApJ, submitted

Dwek, E., Arrendt, R.G., Hauser, M.G.  et al.  1998, ApJ, 508, 106

Ehrenfreund, P., Charnley, S.  2000, ARA &A, 38, 427

Ehrenfreund, P., Dartois, E., Demyk, K., d'Hendecourt, L.  1998, A &A, 339, L17

Elbaz, D., Cesarsky, C.J., Fadda, D., et al.  1999, A &A, 351, L37

Evans, N.J., Allen, L.E., Blake, G.A., et al.  2001, AAS meeting 198, 25.05 (

Genzel, R., Cesarsky, C.  2000, ARA &A, 38, 761

Genzel, R., Lutz, D., Sturm, E., et al.  1998, ApJ, 498, 579

Gerakines, P.A., Whittet, D.C.B., Ehrenfreund, P., et al.  1999, ApJ, 522, 357

Gibb, E., Whittet, D.C.B., Gerakines, P., et al.  2000, ApJ, 536, 347

Gispert, R., Lagache, G., Puget, J.L.  2000, A &A, 360, 1

Greaves, J.S., Holland, W.S., Moriarty-Schieven, G., et al.  1998, ApJ, 506, L133

Habing, H.J., Dominik, C., Jourdain de Muizon, M., et al.  1999, Nature, 401, 456

Haiman, Z., Loeb, A.  1998, ApJ,  503, 505

Haisch, K.E., Lada, E.A., Lada, C.J.  2001, ApJ, 553, L153

Hanson, M.M., Howarth, I.D., Conti, P.S.  1997, ApJ, 489, 698

Holland, W.S., Greaves, J.S., Zuckerman, B., et al.  1998, Nature, 392, 788

Irvine, W.M., Schloerb, F.P., Crovisier, J., Fegley, B., Mumma, M.J.  2000, in Protostars  & Planets IV, eds.  V.G.  Mannings et al.  (Univ.  Arizona), p.  1159

Jewitt, D., Aussel, H., Evans, A.  2001, Nature, 411, 446

Kirkpatrick, J.D.  2001, in 'Galactic structure, stars and the interstellar medium', eds. C.E. Woodward et al.  ASP Vol. 231 (San Francisco: ASP), p.  17

Koerner, D.W., Ressler, M.E., Werner, M.W., Backman, D.E.  1998, ApJ, 505, 358

Kramer, C., Alves, J., Lada, C.J., et al.  1999, A &A, 342, 257

Lada, C.J., Alves, J., Lada, E.A.  1999, ApJ, 512, L250

Langer, W.D., van Dishoeck, E.F., Blake, G.A., Bergin, E., Tielens, A.G.G.M., Velusamy, T., Whittet, D.C.B.  2000, Protostars  & Planets IV,  eds.  V.  Mannings, A. Boss, S. Russell (Tucson: Univ.  of Arizona), p.  27

Le Floc'h, E., Mirabel, I.F., Laurent, O.  et al.  2001, A &A, in press (astroph/0101038)

Lenorzer, A., Bik, A., Kaper, L., et al. 2001, A &A, to be submitted

Lutz, D., Spoon, H.W.W., Rigopoulou, D., Moorwood, A.F.M., Genzel, R.  1998, ApJ, 505, L103

Malfait, K., Waelkens, C., Waters, L.B.F.M., Vandenbussche, B., Huygen, E., de Graauw, Th.,  1998, A &A,  332, L25

Marcy, G.W., Butler, R.P.  1996, ApJ, 464, L147

Mayor, M., Queloz, D.  1995, Nature, 378, 355

McCaughrean, M.J., O'Dell, C.R.  1996, AJ, 111, 1977

Meeus, G., Waters, L.B.F.M., Bouwman, J., van den Ancker, M.E., Waelkens, C., Malfait, K.  2001, A &A, 365, 476

Mirabel, I.F., Vigroux, L., Charmandaris, V., Sauvage, M., Gallais, P., Tran, D., Cesarsky, C., Madden, S.C., Duc, P.A. 1998, A &A, 333, L1

Mirabel, I.F., Laurent, O., Sanders, D.B.  et al.  1999, A &A, 341, 667

Molster, F.J., Yamamura, I., Waters, L.B.F.M. et al.  1999, Nature 401, 563

Molster, F.J.  2000, PhD Thesis, University of Amsterdam

Morris, P.W., Waters, L.B.F.M., Barlow, M.J., Lim, T., de Koter, A., Voors, R.H.M., Cox, P., de Graauw, Th., Henning, Th., Hony, S., Lamers, H.J.G.L.M., Mutschke, H., Trams, N.R.  1999,  Nature, 402, 502

Mouillet, D., Larwood, J.D.,  Papaloizou, J.C.B.,  Lagrange, A.M.  1997, MNRAS, 292, 896.

Mumma, M.J., Dello Russo, N., DiSanti, M.A., et al.  2001, Science, 292, 1334

Oliver, S., Mann, R.G., Carballo, R., Franceschini, A., Rowan-Robinson, M., Kontizas, M., Dapergolas, A., Kontizas, E., Verma, A., Elbaz, D., Granato, G.-L., Silva, L., Rigopoulou, D., Gonzalez-Serrano, J.I., Serjeant, S., Efstathiou, A., Van der Werf, P.P., MNRAS, submitted

Padgett, D.L., Brandner, W., Stapelfeldt, K.R., Strom, S.E., Terebey, S., Koerner, D.  1999, AJ, 117, 1490

Papovich, C.J., Dickinson, M., Ferguson, H.C.  2001, ApJ, in press (astro-ph/0105087)

Schutte, W.A., Greenberg, J.M.  1997, A &A, 317, L43

Schutte, W.A., et al.  2001, A &A, submitted

Serabyn, E.  et al.  1999, A mid-infrared camera for NGST, (

Stahler, S., Palla, F., Ho, P.T.P.  2000, in Protostars  & Planets IV,  eds.  V.  Mannings, A. Boss, S. Russell (Tucson: Univ.  of Arizona), p.  327

Stockman, P., ed., 1997, 'NGST: Visiting a time when galaxies were young' (Associated Univ.)

Thi, W.F., van Dishoeck, E.F., Blake, G.A., van Zadelhoff, G.J., Hogerheijde, M.R.  1999, ApJ, 521, L63

Thi, W.F., Blake, G.A., van Dishoeck, E.F., et al.  2001a, Nature, 409, 60

Thi, W.F., van Dishoeck, E.F., Blake, G.A. et al.  2001b, ApJ, in press (astro-ph/0107006)

Tielens, A.G.G.M.  2001, in 'Galactic structure, stars and the interstellar medium', ed.  C.E.  Woodward et al.  (San Francisco: ASP), p.  92

Tielens, A.G.G.M., van Kerckhoven, C., Peeters, E., Hony, S.  2000, in 'Astrochemistry: from molecular clouds to planetary systems', IAU Symposium 197, eds.  Y.C. Minh  & E.F. van Dishoeck (San Francisco: ASP), p.  349

van den Ancker, M.E., Wesselius, P.R., Tielens, A.G.G.M.  2000, A &A, 355, 194

van Dishoeck, E.F.  2000, in 'NGST science and technology exhibition', eds.  E.  Smith and K.  Long, ASP Vol.  207 (San Francisco: ASP), p.  85

van Dishoeck, E.F., Blake, G.A.  1998, ARA &A, 36, 317 

van Dishoeck, E.F., Wright, C.M., Cernicharo, J., Gonzlez-Alfonso, E., Helmich, F.P., de Graauw, Th., Vandenbussche, B.  1998, ApJ, 502, L173 

van Zadelhoff, G.J., Thi, W.F., van Dishoeck, E.F., Blake, G.A.  2001, A &A, in press 

von Helden, G., Tielens, A.G.G.M., van Heijnsbergen, D., Duncan, M.A., Hony, S., Waters, L.B.F.M., Meijer, G.  2000,  Science, 288,  313

Waelkens, C., Waters, L.B.F.M., de Graauw, Th., et al.  1996, A &A, 315,  L245 

Waters, L.B.F.M., Waelkens, C.  1998, ARA &A, 36, 233

Waters, L.B.F.M., Molster, F.J., Waelkens, C.  1999, in 'Solid interstellar matter: the ISO revolution', eds L. d'Hendecourt et al. (EDP, Springer), p. 219 

Waters, L.B.F.M., Molster, F.J., de Jong, T., et al.  1996, A &A, 315, L361

Wright, C.M.  2000, in 'Astrochemistry: from Molecular Clouds to Planetary Systems', eds.  Y.C.  Minh  & E.F.  van Dishoeck, IAU Symposium 197  (ASP), p.  177

Wright, C.M., van Dishoeck, E.F.,  Cox, P., Sidher, S.D., Kessler, M.F.  1999, ApJ, 515, L29

Wright, G., Posselt, W.  et al.  1999, MIRCAM and MIRIFS, (




APPENDIX F. Abbreviations



Asymptotic Giant Branch 


Active Galactic Nucleus


Assembly, Integration and Testing   


Atacama Large Millimeter Array


Institute for Atomic and Molecular Physics 


Stichting ASTRonomisch Onderzoek in Nederland


Ad-Hoc Science Working Group of NGST 


Astronomy Technology Centre in Edinburgh, UK 


Association of Universities for Research in Astronomy 


Astronomy Working Group of ESA 


Astronomical Unit 


Brown Dwarf 


Bidirectional Reflection Distribution Function 


Computer Aided Design / Computer Aided Manufacturing


Computer Numerically Controlled


Canadian Space Agency


ESA's infrared space interferometry mission (also known as IRSI)


Deep Near Infrared Southern Sky Survey 


Design Reference Mission of NGST


European Coordinating Facility


Extrasolar Giant Planet


Earth Observing System


European Space Agency


European Southern Observatory


European Space Research and Technology Centre in Noordwijk


Far-Infrared and Submillimeter Space Telescope (renamed to Herschel)   


Fundamenteel Onderzoek der Materie 


Goddard Space Flight Center   


Global Ozone Monitoring Experiment


Hubble Deep Field   


Heterodyne Instrument for the Far-Infrared on Herschel


Hubble Space Telescope 


Herschel Space Observatory (formerly known as FIRST)


Interplanetary Dust Particle   


Integral Field Unit   


Initial Mass Function   


InfraRed Array Camera on SIRTF 


InfraRed Astronomical Satellite   


InfraRed Spectrograph on SIRTF   


InfraRed Space Interferometry mission of ESA (also known as DARWIN)  


Infrared Space Observatory


CAMera on ISO


Short Wavelength Spectrometer on ISO


Interim Science Working Group of NGST


James Clerk Maxwell Telescope   


Kuiper Belt Object   


Klein Mechanische Werkplaats Eindhoven   


Large Magellanic Cloud   


Low Resolution Spectrometer (IRAS)   


Long Wavelength Spectrometer (ISO)   


2 Micron All Sky Survey   


MIdinfrared eCHELLE spectrometer on UKIRT/Gemini   


MID-infrared Interferometric instrument for the VLTI   


Multiband Imaging Photometer on SIRTF   


Mid-InfraRed Partnership Planning Group of NGST   


Mid-Infrared Steering Committee of NGST 


Meteorological Operational polar satellites of EUMETSAT


National Aeronautics and Space Administration of the USA


Next Generation Space Telescope   


Near Infrared Camera and Multi-Object Spectrometer on HST   


Nationaal Lucht en Ruimtevaartlaboratorium


Nederlandse Onderzoekschool voor Astronomie


Optical Ground Support Equipment   


Ozon Monitoring Instrument   


Product Assurance   


Polycyclic Aromatic Hydrocarbon   


Photo-Dissociation Region or Photon-Dominated Region


Principal Investigator   


Quality Assurance   


Rutherford Appleton Laboratory 


RijksUniversiteit Groningen   


Service d'Astrophysique in Saclay   


Spectrographic Areal Unit for Research on Optical Nebulae   


SCanning Imaging Absorption SpectroMeter for Atmospheric CHartographY


Submillimeter Common User Bolometer Array (JCMT)   


Spectral Energy Distribution   


Space InfraRed Telescope Facility   


Square Kilometer Array 


Sloan Digital Sky Survey   


Stratospheric Observatory For Infrared Astronomy   


Space Research Organization Netherlands 


Toegepast Natuurwetenschappelijk Onderzoek   


Technisch Physische Dienst   


Terrestrial Planet Finder 


United Kingdom InfraRed Telescope 


Universiteit Leiden   


UltraLuminous InfraRed Galaxy   


Universiteit van Amsterdam   


VLT Mid Infrared Spectrometer/Imager 


Very Large Telescope of ESO


Very Large Telescope Interferometer of ESO   


Work Breakdown Structure   


Wide Field Planetary Camera on HST


William Herschel Telescope on La Palma   


Westerbork Synthese Radio Telescoop   


APPENDIX G.   Relevant Web sites

Much of the material below dates from 2001. The list will evolve; please email useful additions to me
The main NGST Web pages with detailed information about the project can be found at:


The science programs contained in the NGST design reference mission (DRM) can be downloaded through:


The ESA and NASA instrument studies carried out in 1999 can be found at:


The Web pages of the applicants and their relation to the NOVA science program can be accessed through:


APPENDIX H. Some astronomical units     


 Solar Mass 

MSun  1.989  x  1033     gr

 Jupiter Mass   

MJupiter 1.899  x  1030   gr

 Earth Mass   

MEarth    5.974  x  1027    gr   

 Moon Mass  

MMoon       7.349  x  1025    gr   

 Astronomical Unit 

AU    1.496  x  1013    cm

 Light Year   

ly   9.461  x  1017  cm   


pc     3.086  x  1018 cm   


Jy      10-23    erg s-1 cm-2 Hz-1

 Solar Luminosity 

LSun  3.85    x  1033   erg s-1   

 Solar Radius 

RSun  6.963  x  1010  cm 



Email:         ASTRON project manager: Lars Venema             This page: Jaap Tinbergen

     Dwingeloo,  Netherlands.

  (ASTRON is the official name of the Netherlands Foundation for Research in Astronomy NFRA).

URL: file:///C:/jaap/MIRI-NGST/frontpage_miri-website/public/science/phase-a/text.htm

Last modified by: Tinbergen    Last Update: 25 sep 2002 15:22