Extrasolar Planets:The (more or less) Standard Theory for Planet Formation[Textbook is weak on planet formation. But see pp. 14, 20-21 (especially study thebeautiful illustration on p.21), and 264-265. Read this texbook material before studyingthese more detailed notes.Also, these notes are meant to be read in conjunction with the lecture presentation.A pdf of the powerpoint presentation is available online, and a hardcopy version will beavailable at PMA.]Theories for planet formation involve many thorny problems and hotly-contested ideas,but they all star with a rotating disk, and now we know observationally that this is a goodassumption.Minimum mass of the disk: probably ~ 0.01MO , similar to masses of observed disks(although these are very uncertain). Note that the sum of our solar system’s planets’ masses isabout 0.001Mo (mostly Jupiter), but this is partly because most of the hydrogen and helium havebeen lost (escaped from planetary atmospheres because so light). Some models do assume moremassive disks.The main features to remember are that:a. In all models, the temperature and density decrease with distance as you move out fromthe central star. (Observations suggest slower decrease of temperature than modelspredict, but both are uncertain.) This will make it fairly easy to understand most of thecomposition differences in terms of a “condensation sequence”: different kinds of solidsvaporize at different temperatures, so only “rocky” material can survive in the inner disk,while both rock and icy material can survive in the outer disk. (Illustrations in class inpresentation on disks.)b. These disks are mostly gaseous (~99% by mass), with a “cosmic composition” like thesun’s and most other stars, but about 1% of the mass is in the form of microscopic dustgrains, which play a fundamental role in the evolution of the model planetary systems.Presumably these grains are interstellar grains that happened to be dragged along with thegas into the protostellar disk.The more-or-less standard “core-accretion” theoryMain stages in evolution of a disk toward a planetary system: (things in braces are references totechnical papers that you can ignore)1. Settling of the dust through the gas into a thin layer at the disk mid-plane. (Because gas issupported in vertical direction by pressure, dust isn’t. Explained in class.).Problem: How can the dust settle if the gas in the disk is turbulent? It would constantly be“stirred up.”2. Collisional “accretion” or “accumulation” of dust particles by other dust particles, leading tokilometer-size planetesimals in ~ 104 yr.Problem: particle-particle collision velocities are large enough (~0.1 to 10 m/sec for sizes~1mm to 1cm) that collisions should lead to shattering, not coalescence. So how can you getto 1 km planetesimals?{Wurm et al. 2001 Icarus, 151, 318 offer a solution; many papers on this since then.}3. Gravitationally-aided collisional accumulation of planetesimals through a process of“runaway growth” (explained in class), resulting in planetary embryos (Mercury- to Mars-size)in ~ 105 yr.{“Dynamical friction” causes energy equilibration, so large bodies have enhanced rate ofaccumulation of other large bodies—this is a runaway process; new runaway due to size-dependent gravitational perturbations and gas drag: Kortenkamp et al. 2001, Science,293, 1127.}4. Giant impacts between embryos, resulting in full-size terrestrial planets in ~ 107 to 108 years(problem?—remember lifetimes of disks estimated above), but also causing large disturbancesand destruction (e.g. formation of our Moon).5. Farther out in disk, ices survived (because cooler), so more solids, and embryos may reachabout 10 Earth masses in about 106 yr. After reaching this mass, the bodies can gravitationallyaccrete ~100 Earth masses of disk gas to produce giant planets like Jupiter and Saturn in about107 yr. (Again, potential problem with timescales for disk disruption.)Pictures of simulations of steps 4 and 5 are shown in lecture: result is formation of planetsthat look roughly like our solar system (except not enough time to make Uranus andNeptune—some people think they formed closer in and were “scattered” or “migrated” out totheir present positions).An alternative to this “core-accretion” mechanism (5) for forming giant planets: theyform by the gravitational collapse of Jupiter-mass clumps of gas and dust in the disk. Thismight only require about 100 years. Pictures of simulations of this process are shown in thelecture presentation. The problem for this process is that it requires a fairly massive disk, andno one knows if the disk masses are sufficient (hard to measure from dust emission or COemission lines).If this model is correct, then giant planets formed before terrestrial-size planets. In thatcase the evolution of planetesimals was dominated not by their own (fairly feeble)interactions, but by much stronger gravitational perturbations from the massive planets. Thismight be a problem if you want to form terrestrial planets, because many giant planetsundergo migration (see below).Several illustrations of simulations of this process will be shown in class.So the planetesimals are probably acted on by two competing forces:1. Gravitational perturbations from the massive bodies “scatter” the planetesimals,increasing their orbital eccentricities and inclinations; in fact many of them must havebeen “kicked out” of our solar system (and probably others) by such a process, to accountfor the “Oort cloud” of comets (which are supposedly planetesimals scattered into veryeccentric orbits). Also undoubtedly formed the “Kuiper belt” of comets (beyondNeptune). Some bodies must have been kicked into interstellar space (consider: “rogueplanets” with no parent star). We’ll call this process “gravitational scattering.”2. The planetesimals move around the central star faster than the gas (because the gas haspressure as a partial support, as well as rotation, while the planetesimals do not have apressure). The resulting “gas drag” is strongest for smallest planetesimals. The dragforce reduces orbital eccentricity and inclination, and also removes angular momentum,causing planetesimal orbits to slowly decay toward the star. This might cause some starsto “cannibalize” their own planets, or at least planetesimals.Both processes cause orbital migration. (There are other
View Full Document
Unlocking...