The formation of the major observational features of our Solar System can all be explained through a scenario of gradual growth through collisions. In this scenario, the Solar System started as a cloud of gas and dust that slowly collapsed into a flattening disk. (This is discussed more in the Solar Nebula article.) Initially, dust grains grew into larger and larger clumps through simple collisions, but over time objects grew large enough that they also began to grow by gravitationally attracting nearby material. This type of an accretion process tended to form bodies rotating in the same pro grade direction that the disk rotates. Dynamical studies also suggest typical rotation periods in our Solar System should be 5 to 20 hours, which is consistent with our observations of planets and asteroids.
Artist's concept of a protoplanetary disk. Click here for original source URL.
This model also explains the spacing of the planets. Studies suggest that if two large bodies started growing in orbits that were too close together, they would eventually grow large enough, as they gravitationally cleared out their orbit, to attract each other gravitationally, collide, and merge. In this way, the solar nebula divided into donut-shaped zones around the Sun, each about 1.5 to 2 times as wide as the one next closer to the Sun. The result is just one planet dominating each zone. (This is actually part of the definition of a planet, and is part of why Pluto in its shared orbit isn't a planet.) Johann Bode noticed this spacing several hundred years ago, although he could not explain it, and so we call the pattern Bode's Rule.
Johann Bode. Click here for original source URL.
The Russian scientist Victor Safronov was one of the first to work out the process of collisional accretion. As grains in the solar nebula collided and aggregated, they formed medium-sized planetesimals, ranging in size from millimeters to hundreds of kilometers. We know that large planetesimals were abundant throughout the young Solar System, based on the following three pieces of evidence: First, we observe craters on planets and moons that were caused by impacts with objects at least 100 kilometers in diameter. Second, many of the meteorites we’ve studied were once part of larger bodies up to a few hundred kilometers across. Lastly, we see asteroids and cometnuclei in the Solar System today that reach several hundred kilometers in diameter. These are surviving remnants of planetesimals. The newly formed planetesimals likely resembled many asteroids. Perhaps they had irregular shapes due to the asymmetric mergers of smaller bodies and fracturing and cratering that was caused by multiple collisions.
We can speculate how the accretion of planets took place in general. The details, however, are very complex. In the beginning, planetesimals orbiting the Sun were like race cars, moving on adjacent, parallel lanes of a circular racetrack. If they collided, they hit in low-speed (relative to each other) "sideswiping" collisions. Nothing initially disturbed their circular motions or forced them to "change lanes," so all the particles moved together on nearly circular paths. As the largest planetesimals grew, their gravity got stronger. That drastically perturbed the motion of any smaller planetesimal that passed nearby, sending it on a new course across the orbits of other bodies, like a car changing lanes. You can think of this in terms of conservation of energy: If a massive body transfers kinetic energy to a much smaller body, the massive body slows down slightly but the small body speeds up greatly. With forming planetesimals this type of energy transfer doesn’t occur through a collisions, but via gravitational interactions. In this way, the growing objects "pumped up" the collision speeds of smaller planetesimals. Later, their speeds got so high that smaller planetesimals sometimes smashed into each other at high enough speeds to self-destruct, ending the growth process for all except the largest objects.
How long did planet-building take? You might think it would take a very long time to build a planet by the accretion of tiny pieces. Let's imagine it was a linear process. In other words, growth occurred by adding one piece at a time. Say we start with 1 meter-sized chunks of rock. The biggest planetesimals are about 100 kilometers across. Volume increases with the cube of the diameter, so it would take (100 / 0.001)3 = 1015 small chunks to make a large planetesimal. A medium-size planet is about 10,000 kilometers across. This is another factor of (10,000 / 100)3 = 106 in volume. So a planet is made of 1015 × 106 = 1021, or a thousand billion billion small chunks. And if you build a planet one piece at a time, it will take a thousand billion billion times as long as it does for two small chunks to come together. It seems hopeless, yet accretion is actually a very efficient process.
What actually happened was that the largest planetesimals grew fastest, sweeping up the others. This process is in fact non-linear; the larger they grew, the more their gravitypulled in neighboring planetesimals, causing growth to accelerate. The time needed for this process can be measured using radioactive isotopes. Certain radioactive materials with half-lives of a few million years were trapped inside meteorite parent bodies before the isotopes decayed. Studies in the 1990s showed that the parent bodies of some meteorites had reached diameters of hundreds of kilometers, and had been partially melted, all within in a few million years of the Sun's formation. This is a strikingly short time compared to the history of the Solar System.
If you represent the 4.6 billion-year history of the Solar System as one Earth year, asteroid-like solid bodies had formed from nebular dust by noon on January 1st. Isotopic studies indicate that the largest of these bodies reached planet size in 50 to 100 million years, within the first one or two percent of solar system history. In the analogy just given, Earth and the other planets would be fully grown by around January 4th.
Using accretion, we can also explain the origin of the major groups of bodies in the solar system — terrestrial planets, giant planets, asteroids, and comets. What we can't explain is why in many solar systems other than our own, terrestrial planets seem to be absent. The simplistic model presented in this section is only a starting point, and new models are being worked on that can explain planet formation in all its diverse forms.
• Terrestrial Planets. In the inner part of the Solar System, the process of accretion continued until Mercury-sized to Earth-sized planets had formed. Most of the planetesimals in this region were made of silicate (rocky) material — this close to the Sun, it was too warm for ices to remain solid. The terrestrial planets grew in this ice-less environment until most of the silicate material in the area was swept up. An out-rushing of gas and radiation from the young Sun blew away the remaining gas and dust left behind.
• Giant Planets. The giant planets formed in the same way as the terrestrial planets, from accreting planetesimals. Farther from the Sun, the giant planet zone contained icy as well as rocky material, which augmented local planetesimal masses. Thus the embryo planets — called proto-planets — that would become Jupiter, Saturn, Uranus, and Neptune grew larger than Earth and the other terrestrial planets. When they reached about 10 to 15 times the mass of present-day Earth, their gravity was strong enough to pull in gas from the surrounding solar nebula. This is why they accreted not only solid planetesimals, but also massive atmospheres of gas with a composition approximately that of the nebular gas. The giant planets can be thought of as two-phase planets, with initial cores of icy and silicate materials, and massive hydrogen-rich atmospheres added on later. Notice that we can predict that the giant planets have solid cores, before we see evidence for them.
• Asteroid belt. Asteroids are planetesimals that never made it all the way to "planethood." Why were most asteroids stranded in the zone between Mars and Jupiter? Probably because it was the planet-forming zone closest to the largest planet in the Solar System. According to the rough geometric spacing expressed by Bode's rule, a planet should have grown here. Ceres, the largest asteroid, did grow to about 1000 kilometers in diameter. It stopped growing because nearby Jupiter had become so huge that its gravity disturbed the motions of the other asteroids and pumped up their collision speeds. This caused them to smash into innumerable fragments when they collided, instead of coalescing into an even-larger body. Its of interest to note that the Solar System’s "snow line" runs through the asteroid belt, allowing the asteroids furthest from the sun to have ices.
• Near-Earth Objects. The largest objects in the asteroid belt reached sizes of a few hundred kilometers across. Their internal heat caused them to melt and differentiate into metal and rock layer, just like the Earth. The transfer of energy from nearby Jupiter increased their speeds, causing some of them to shatter in collisions with other large and/or fast neighbors. Fragments of iron cores, mantles, and rocky surfaces scattered. Much of this debris left the asteroid belt, and a few pieces ended up in Earth-crossing orbits as Near-Earth Objects. Occasionally, these objects get too close to Earth and actually hit it, becoming meteorites. We can therefore explain the origins of stony, iron, and stony-iron meteorites. The violent origin of these rocks is particularly clear in brecciated meteorites, with their fused jumble of different rock types.
• Kuiper Belt. In the outer most parts of our Solar System, starting just inside the orbit of Neptune and continuing out out to more than 80 A.U., are icy bodies that range from few meters to a few 1000 km across. These objects include Pluto, Makemake, and other frozen bodies, including dwarf planets. These bodies, much like the asteroid belt, are fragments that couldn't gravitationally coalesce into a giant planet due to gravitational disruptions from the giant planets.
• Comets. Comets are icy planetesimals from the outer solar system. They have so far avoided direct collisions with the giant planets, but close encounters with these planets have flung them into the Kuiper Belt and the Oort cloud. It is unclear exactly what mechanisms are responsible for creating particular comets, but different causes include everything from gravitational interactions between our Solar System and other stars, collisions within the Kuiper belt, and resonances between the outer planets and Kuiper belt objects all can send chunks of ice into the inner Solar System.