Astronomy & Astrophysics
Heidelberg, Germany
F. Motte,
P. AndrÈ,
Astronomy & Astrophysics:
F. Rostas, press contact,
C. Bertout, Editor-in-Chief,
January 8, 2001
A&A press release 2001-1

Early stages of star formation are now better understood, following an extensive millimeter-wave study of protostars, which are young stellar objects still deeply embedded in their parent molecular cloud. Thanks to their unprecedented high-resolution maps of the circumstellar environment of many young stellar objects, astronomers FrÈdÈrique Motte and Philippe AndrÈ find that the radial density in the circumstellar envelopes of protostars depends on the nature of their environment. In isolated protostars, this is well described by a model of spontaneous gravitational collapse proposed by Shu, Adams, and Lizano in 1987, but it is quite different for protostars that form in groups, suggesting a more dynamic collapse driven by external perturbations such as those possibly caused by neighbouring stars in formation. This shows for the first time that the environment of forming stars may partly control the process of star formation itself. These results, obtained by F. Motte [1] and P. AndrÈ [2] using a matrix of bolometric detectors [3] at the IRAM [4] 30-meter radiotelescope, will appear in Astronomy and Astrophysics 365, 440 (January 2001).
[1] Max-Planck-Institut f¸r Radioastronomie, Postfach 2024, D-53010 Bonn,     Germany
[2] Service d’Astrophysique, Commissariat ý líEnergie Atomique,
    C.E.N. Saclay, F-91191 Gif-sur-Yvette, France
[3] Bolometers are temperature sensitive detectors that measure directly     the energy flux carried by the radiation. In the millimeter wavelength     range this technique is more effective than converting photons to     electrons through the photoelectric effect, a technique used in the     visible and near infrared wavelength ranges (for example in CCD     cameras). In the millimeter wavelength range, each photon carries     thousand times less energy than in the visible range and it is thus     impossible to find efficient photoelectric materials.
    The MAMBO bolometer array used in this work was developed at the     Max-Planck-Institut f¸r Radioastronomie
    ( Pictures of this     innovative instrument are available on that site.
[4] Institut de Radioastronomie MillimÈtrique, 300 rue de la piscine,     Domaine Universitaire, 38406 Saint Martin d’HËres, France or
Motte, F. and AndrÈ, P., 2001, A&A, 365, 440
Shu, F.H., Adams, F.C., Lizano, S., 1987, ARA&A 25, 23
Illustrations []
Figure 1. Dust continuum emission in the circumstellar environment of embedded young stellar objects located in the Taurus molecular cloud obtained at a wavelength of 1.3 mm with the MAMBO bolometer array mounted at the IRAM 30-meter radiotelescope.
The sky region displayed here contains two objects corresponding to different stages in the star forming process: L 1535-NE is a pre-stellar dense core that has not yet developed an inner stellar nucleus. IRAS 04325 is a protostar, corresponding to the next stage in the star formation process where a stellar nucleus has formed and attracts the surrounding envelope. The 1.3 mm emission recorded here is due to the dust that is intimately mixed with the interstellar gas. Assuming a uniform mixing, the intensity maps can be analyzed to derive gas density distributions such as those displayed on the next figure.
Figure 2. Examples of density profiles in the circumstellar environment of protostars, derived from 1.3 mm continuum maps such as shown in Figure 1.
Pre-stellar cores (e.g. L1535-NE) and protostellar envelopes (e.g. L1527 in Taurus and L1448-N in Perseus) are clearly resolved by the 30 m telescope, in contrast to the circumstellar envelope of the older pre-main sequence star HL Tau. The telescope resolution limit is shown as a dashed line marked IRAM 30m beam. The singular isothermal sphere (SIS) model (Shu et al. 1987) accounts fairly well for the radial profiles of isolated protostellar envelopes in Taurus (here L1527), but is too centrally condensed to represent pre-stellar cores (here L1535-NE). In contrast, the envelopes of Perseus protostars (here L1448-N) are one order of magnitude denser than the SIS model and their radial intensity profiles merge into the cluster cloud emission at about 10000 astronomical units from the protostarís center.
The study of the earliest stages of star formation has long been hampered by the fact that the cold and dense gas found in dense cloud cores is opaque to visible and near infrared radiation. The recent advent of sensitive matrices (or arrays) of (sub-)millimeter bolometers has opened the possibility of mapping the thermal continuum emission of the dust component around 1 mm where the medium is transparent. Assuming plausible values for the physical properties of the circumstellar dust, the intensity maps can be converted to density maps and thus the density structure of the dense cloud cores can be studied and compared to theoretical models describing the early phase of star formation. Using the MAMBO bolometer array on the IRAM 30m radiotelescope, a sample of 49 embedded young stellar objects (YSOs) has been mapped at a resolution of 11 arcseconds. This corresponds, at the object distances (from 140 to 460 parsecs), to spatial scales of 1500 to 5000 AU.
Star formation scenarios
Solar-type stars result from the collapse of a dense fragment of molecular gas (called dense core) under its own gravity. The cloud fragment evolves towards higher degrees of central condensation until a star forms in its center. The newborn protostar then builds up its mass from the surrounding infalling envelope until the protostellar envelope dissipates, revealing the central object as a pre-main sequence star. After additional contraction taking place over a few million years, the star begins turning hydrogen into helium by thermonuclear fusion reactions and enters its adult life, becoming a main-sequence star.
A model was proposed in the 80s to explain the formation of isolated low-mass stars (Shu et al. 1987). This so-called "standard model" describes the spontaneous gravitational collapse of an idealized, isolated cloud core. However, we now know that many stars form in a clustered environment where individual cloud fragments occupy a finite volume of space and various external disturbances due for example to other forming stars in the neighborhood can induce core collapse. Because of the complexity of the problem, there is little theoretical work on the formation of stars in groups.
In order to constrain the initial conditions for collapse and develop more realistic star formation models, the astronomers must observe the density and velocity structure of a variety of cloud cores and protostellar envelopes. Although theoretical models have been around for a while, they could not be tested by comparison with observations until the advent of bolometer arrays. Past studies only investigated a few protostellar envelopes and yielded contradictory, inconclusive results on their density structure.
The high sensitivity and angular resolution of the MAMBO bolometer array allowed F. Motte and P. AndrÈ to study a broad sa
mple of nearby protostars forming either in isolation or in groups (or clusters). A map example is shown in Figure 1. Motte and AndrÈ mapped all 27 protostars in the Taurus molecular cloud along with 9 isolated
globules detected by the IRAS infrared satellite and 9 protostars in the Perseus molecular cloud complex.
For the first two sets, made up of isolated objects, the authors conclude that the density structure of protostellar envelopes (such as L1527 in Figure 2) is well described by the standard model. In the last set, where the YSOs belong to clusters, the protostellar envelopes (such as L1448 in Figure 2) are found to be denser and finite sized. The authors suggest that this results from a more dynamical collapse initiated in compact cloud fragments. The collapse initial conditions prevailing in protoclusters thus appear to differ markedly from the idealized case assumed in the standard protostellar model.
Future developments
Because protostars are enshrouded in dusty molecular gas that obscure them from view at infrared and shorter wavelengths, submillimeter mapping appears as the only way to observe protostars in their buildup phase. The capability to make such studies will improve drastically in the next decade as two major international facilities will start operation. The Herschel/FIRST satellite (Far Infra-Red and Submillimeter Telescope) to be launched in 2007 and the ground-based interferometer ALMA (Atacama Large Millimeter Array) to be completed around 2010. By making deep, unbiased surveys of molecular clouds and protostars in the submillimeter band, these two instruments might revolutionize our understanding of the earliest stages of star formation.