Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

I Why hasn't the Oort cloud converged to a disk shape?

  1. Dec 4, 2018 #1
    The Oort cloud is often drawn as a sphere. Everything in it orbits the sun in circles or ellipses. Since the shape is not a disk, collisions are expected. Why hasn't the Oort cloud converged to the more stable, collision-free, disk shape?
  2. jcsd
  3. Dec 4, 2018 #2

    Buzz Bloom

    User Avatar
    Gold Member

    Hi @Jonathan212:

    Since quite a few people have already looked at your thread and did not respond, I decided to make a try at answering your question. I confess that I am not an expert, and the answer I will make is based mostly on reading other threads at this site.

    In general, a "cloud" configuration of stuff moving about a star is stable. What causes a spherically shaped collection of stuff to change its shape to an ellipsoidal shape, and an ellipsoidal shaped collection to flatten into a disk, is that the total dynamic energy of the stuff reduces. A disk of the same radius as a sphere has a lot less energy. The mechanism that causes energy to be lost is electromagnetic (EM). That is, energy in the form of photons carry energy away from the stuff. That will generally happen when the stuff is in the gaseous state. When the gas cools to a solid state the frequency of EM interactions reduces substantially. Stuff that is sufficiently close to the star (like the solar system) is heated by the star sufficiently to remain gaseous longer than stuff far away from the star (like to Oort cloud). When the stuff cools to a solid state, molecules hit molecules at a relatively low speed which results in attaching and making bigger molecules. Eventually, these attaching events produce dust. When the dust particles hit (slow enough) they attach into large dust, and so on until you get rocks and comets and asteroids, etc. These attaching interactions produce only a little dissipating heat photons which cause only a very little flattening of the overall shape.

    To the extent that the above answer has errors, I hope someone will correct them.

  4. Dec 4, 2018 #3


    User Avatar

    Staff: Mentor

    What is called dynamical friction plays a role. Dynamical friction occurs when bodies interact gravitationally when passing through some medium (like a dust cloud). In the early solar system the density of dust and stuff in the inner system was much higher than in the far reaches, so it had a chance to extract a lot of KE from things moving there. Things out in the far reaches have far fewer gravitational encounters to redistribute orbital energy and dump it into a medium.
  5. Dec 4, 2018 #4
    I'm trying to relate this to simulations of extremely large numbers of point masses. A dust cloud could be modeled as lots of point masses, if enough computing power and time is available. So it's all point masses colliding with each other, and they are all moving in elliptical segments between collisions, as per Newton's laws. What should happen when they collide? Are colliding dust particles or even molecules supposed to emit e/m waves and slow down as a result?
  6. Dec 4, 2018 #5
    Buzz, a disk could have an average radius equal to that of the sphere so it could have the same gravitational energy. The same kinetic energy too. Bouncing could even eject objects with escape velocity so there is no limit to how far they can get.
  7. Dec 4, 2018 #6


    User Avatar
    Gold Member

    Will point masses collide? Is it possible the sim needs to have non-zero masses in order to correctly predict the observed behavior?
  8. Dec 5, 2018 #7
    An obvious answer is that Kuiper disc already has, but Oort cloud is in the process of joining Kuiper belt, and has not yet.
    High inclination orbits are being cleared out by Kozai resonance. For long period comets, it is slow, and has not yet exhausted Oort cloud.
  9. Dec 5, 2018 #8
    You mean non-zero radius and finite small mass. Dealing with collisions just means adding repulsive electromagnetic forces that only apply at small distance or something like that. This is not a practical simulation because too much computing power and time would be needed, but for the purposes of understanding what is going on with the Oort cloud it may be helpful.
  10. Dec 5, 2018 #9
    I have a related question. If you launch 1000000 perfectly elastic balls in random directions and from random locations around the sun, would they still form a disk eventually? In other words, are plastic, not elastic collisions a necessary condition for disk formation?
  11. Dec 5, 2018 #10


    User Avatar
    Science Advisor
    Gold Member

    I could suggest that those objects were late arrivals to the Solar System from a random set of directions and with random speeds and their orbits haven't had time to be influenced by the large planets? The formation and maintenance of Saturns rings is 'explained' in terms of a number of small moons that 'shepherd' the particles in the rings and keep them within those well known ring patterns. Perhaps it's the same sort of thing on a bigger scale?
  12. Dec 5, 2018 #11
    Estimates have been made in the past of the total mass of Oort Cloud bodies. Here are a couple of estimates.

    "If Halley's Comet's mass is typical for comets, then the Oort Cloud could have a total mass between 4 and 80 Earth masses" [1].
    "Past estimates of the total mass of this Oort Cloud have ranged from about 40 times that of Earth to greater than that of Jupiter" (about 317 Earth masses) [2].

    These estimates are largely based on the masses of comets and dwarf planets which have been detected because, during parts of their orbits, they have reached the Inner Solar System, or at least the Kuiper Belt. All these bodies have had highly elliptical orbits, if they hadn't they would never have come within range.

    So such estimates do not take account of whatever bodies exist within the Cloud which do not have elliptical orbits, and so do not come into range. Taking these bodies into account might increase the estimated total mass of the Cloud.

    [1] Nick Stroebel. Comet Orbits---Oort Cloud and Kuiper Belt. http://www.astronomynotes.com/solfluf/s8.htm
    [2] Oort Cloud & Sol b? http://www.solstation.com/stars/oort.htm
  13. Dec 5, 2018 #12


    User Avatar
    Gold Member

    Oops. Yes. I meant non-zero size of course.

    Well, no. That was my point. Particle don't just bounce off one another - they collide in non-elastic collisions and stick together. See posts 2 and 3.

    I'm saying that that may well be important to the simulation in how it models the Oort cloud versus the Kuiper Belt.
  14. Dec 6, 2018 #13
    If the Oort cloud is getting added to the Kuiper belt why isn't there a continuity from one to the other?
  15. Dec 28, 2018 #14
    Unlike the galactic disk and the Kuiper belt, the Oort cloud has zero total angular momentum and thus no inclination to settle into a disk. The large radius of the Cloud guarantees little to no collisions, similar to the stability of the asteroid belt.
  16. Dec 28, 2018 #15


    User Avatar
    Science Advisor
    Gold Member

    I find that an interesting statement. How is that determined? By observing a statistically significant number of Oort objects (too small and too distant, I suspect) or just by observing comets? Presumably the "zero angular momentum" is referenced to the Sun as centre and assumes that the objects arrive symmetrically 'on either side' as they approach the Sun. As part of the galaxy, they presumably have the same mean angular momentum around the Galactic Centre as the Solar System. I say that because there can be nothing particularly special about the individual Oort objects; they didn't arrive from outside the Galaxy (presumably??) and just turn up 'here'.
  17. Dec 28, 2018 #16
    If there were more objects in the Oort cloud, would angular momentum arise from too many collisions?
  18. Dec 28, 2018 #17
    Another factor may be that the Oort Cloud keeps getting 'churned' by passing stars...

    IIRC, latest was Scholz's Star, about 70 kyear ago..

    Estimates indicate that the WISE 0720−0846 system passed about 52,000 astronomical units (0.25 parsecs; 0.82 light-years) from the Sun about 70,000 years ago.[2][5] [11] 98% of mathematical simulations of the star system's trajectory indicated it passed through the Solar System's Oort cloud, or within 120,000 AU (0.58 pc; 1.9 ly) of the Sun.[2] Comets perturbed from the Oort cloud would require roughly 2 million years to get to the inner Solar System.[2] At closest approach the system would have had an apparent magnitude of about 11.4, and would have been best viewed from high latitudes in the northern hemisphere, in the autumn mostly.[4] A star is expected to pass through the Oort Cloud every 100,000 years or so.[4] An approach as close or closer than 52,000 AU is expected to occur about every 9 million years.[2] In about 1.4 million years, Gliese 710 will pass somewhere between 8,800 and 13,700 AU from the Sun.

    FWIW, reading of this pass by Scholz's Star inspired me to write a short story, 'Hunter's Night'...
  19. Dec 28, 2018 #18
    Without non-elastic collisions the total energy will remain constant and you will get a spheroid with a ratio between kinetic and potential energy according to the virial theorem. That means, yes, you need non-elastic collisions. But I'm not sure if the collisions between your test particles are sufficient, because the number and the size of the particles cannot be realistic. How about a combination of your particle method with a continuum method for the collisions?
  20. Dec 28, 2018 #19
    Yes basically from the observation that comets have no preferential direction that they're coming from, they appear from all over they sky in a spherical distribution. Oort cloud objects are much too faint and distant to observe directly, and yes, I mean zero angular momentum with respect to the Sun.
  21. Dec 28, 2018 #20
    The direction is not sufficient to conclude that the total angular momentum is zero. The angular momentum needs to be equally distributed too. Is there a value available for the average orbital angular momentum of all known long-period comets?
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?