Answer courtesy of Wayne Hayes, University of Toronto.
(begin quote)
A good article is:
I just finished sitting in on a short course on planetary formation by Scott Tremaine here in Toronto. In a nutshell:
The old idea of planets condensing in situ is almost certainly wrong. This is called the "giant gaseous protoplanet" hypothesis and has fallen out of favour for various quite convincing reasons discussed in Lissauer's paper above. It is essentially a "top-down" model, because it hypothesizes that big clouds of gas condense in situ into planets. The currently accepted model is "bottom-up" because it builds planets via the route dust particles growing via dust and gas accumulation and "sticking" -> pebbles growing by sweeping up dust and gas -> boulders -> moon-sized objects which grow by collision with each other -> planets. The amazing thing about this theory is that we start with gas/dust grains of mass 10^-12 g, and get to planets with mass 10^30 grams, so we need to understand the detailed physics over a mass scale of 10^42, which is quite amazing! So, a bit more detail:
A slowly spinning gas + dust cloud flattens into a disk. Eventually the dust starts to accumulate into small particles. (Not the gas --- you need strong gravity for that, and you can't hold onto gas until you're at least the size of the Moon.) The particles sweep up the dust, growing incredibly fast --- about 10cm/year in radius. These particles grow by surface accumulation of surrounding dust, and this process can be modelled using statistical physics and sticky particles. When the particles reach meter-sized or so, there is a problem: particles of order 1m in diameter spiral into the Sun within about 100 years, due to drag from the surrounding gas cloud. So, we're pretty sure the 1-meter sized particles must collide and stick together to form kilometer sized objects within about 100 years, or else we wouldn't be here. Km sized objects aren't affected much by the gas drag. The problem is we don't understand this process well enough to say that it's possible... but of course it must be possible or we wouldn't be here... that's an application of what's called "The Anthropic Principle." We don't understand the physics of building km sized objects from 1m sized ones in only 100 years, but we know that it must have happened somehow.
So anyway, now we have km-sized objects, and from this point on it's basically an N-body problem. These km-sized objects may accumulate some more dust from the nebula, but for the most part they grow by collision with each other. The final stage occurs when we get about 10^4 objects of approximately lunar-mass ("planetary embryos") that are all in roughly circular orbits. All the previous stages were pretty fast, but this final stage lasts tens or hundreds of millions of years, because these 10^4 lunar-mass objects are all in short-term stable, non-intersecting orbits, and they can only collide when their orbits get slowly perturbed towards each other.
That's a really short-and-dirty crash course on solar system formation.
|
|
|---|