| Summary: | High-mass stars play a key role in the energetics and chemical evolution
of molecular clouds and galaxies. However, the mechanisms that allow
the formation of high-mass stars are far less clear than those of
their low-mass
counterparts. Most of the research on high-mass star formation has focused
on regions currently undergoing star formation. In contrast, objects
in the earlier prestellar stage have been more difficult to identify.
Recently, it has been
suggested that the cold, massive, and dense Infrared Dark Clouds (IRDCs) host
the earliest stages of high-mass star formation.
The chemistry of IRDCs remains poorly explored. In this dissertation, an
observational program to search for chemical
variations in IRDC clumps as a function of their age is described.
An increase in N2H+ and HCO+ abundances
is found from the quiescent,
cold phase to the protostellar, warmer phases, reflecting chemical
evolution. For HCO+ abundances, the observed trend is consistent with
theoretical predictions. However, chemical models fail to explain the observed
trend of increasing N2H+ abundances.
Pristine high-mass prestellar clumps are ideal for testing and constraining
theories of high-mass star formation because their predictions differ
the most at the early stages of evolution. From the initial IRDC sample,
a high-mass clump that is the best candidate to be in the prestellar phase
was selected (IRDC G028.23-00.19 MM1). With a new set of observations,
the prestellar nature of the clump is confirmed. High-angular resolution
observations of IRDC G028.23-00.19 suggest that in
order to form high-mass stars, the detected cores have to accrete a large
amount of material, passing through a low- to intermediate-mass phase
before having the necessary mass to form a
high-mass star. The turbulent core accretion model
is inconsistent with this observational result, but on the other hand, the
observations support the competitive accretion model. Embedded cores have
to grow in
mass during the star-formation process itself; the mass is not set at early
times as the turbulent core accretion model predicts.
The observed gas velocity dispersion in the cores is transonic and mildly
supersonic, resulting in low virial parameters (neglecting magnetic fields).
The turbulent core accretion model assumes highly supersonic linewidths and
virial parameters $sim$1, inconsistent with the observations, unless
magnetic fields in the cores have strengths of the order of 1 mG.
|
|---|
