Recently, the cover of Newsweek magazine featured a protostellar
disk in formation. This cover speaks to one of the most obvious questions
facing astronomers: "How do stars form and evolve?" While the evolution of
most stars can be understood with reasonable accuracy (see my edit "Why the Sun
Shines" from Sep. 25, 1997), the problem of forming stars remains a
head-scratcher. There are two main issues to contend with. First, space is
very empty, so how does one get such an intense concentration of matter as a star
in a relative vacuum? Second, once that matter is together, how to you get rid
of its angular momentum (tendancy to remain rotating, like a top), so that it can fall together.
The first problem is one of a suite of problems in astrophysics that basically
have to do with gravitational collapse within an intially uniform density
field. The basic solution is a simple one, and is due in large part to Jeans.
Jeans suggested that in a given volume of constant density, a small, local
increase in density would grow, because the overdensity would have more
gravitational attraction than the surrounding (relatively) underdense regions.
To stimulate growth, the increase in gravitational force has to be comparable
to the pressure force on the gas, otherwise the overdensity would propagate
outward and diffuse (as a sound wave) before any gravitational instability
could be useful. This basic construction allows one to define a characteristic
mass and length scale at which an overdensity will grow due to gravity, given
the temperature and density of the original volume. This "order-of-magnitude"
argument is born out in simulations, although the argument is by no means
exact. As the overdensity grows due to gravity, obviously some matter begins
to "fall in" toward the center of the overdensity, and make things denser
still, hence the runaway collapse into a very dense object.
The second problem is a much more difficult, and as yet unsolved problem.
Because the particles are moving about in random directions, they migrate into
the overdensity on random orbits and collide evermore vigorously with each
other as they fall in. However, inevitably, there is some residual angular
momentum in the inward flow, because the initial density field did not have
exactly random velocities relative to the overdensity's center. For
stars, the Jeans length scale is about a few to tens of light-years
(1017 meters), but the size of a planetary systems is only a few 100
astronomical units (1013 meters). Since angular momentum increases as the
square of radius, a tiny residual velocity at the Jeans length scale can result
in a huge angular momentum at the planetary disk scale.
