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

Supermassive blackholes and galaxy formation.

  1. Aug 12, 2007 #1
    The question I'm about to asked has probably asked before many times b , but I keyed in a "supermassive blackholes and galaxy formation" in the PF search engine , and this question has yet to be asked:

    How can supermassive black holes be aid the formation of galaxies when they are suppose to absorbed everything in space where it cannot escape including anything particle (or photon) that travels at the speed of light? Shouldn't galaxies be absorbed by blackholes as well? Or perhaps they are being absorbed by blackholes, just at a slower rate than most matter coming into a black hole and beyond its event horizon?
  2. jcsd
  3. Aug 12, 2007 #2

    George Jones

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    What would happen to the Earth if, right noiw, the Sun turned into a one solar-mass black hole?
  4. Aug 12, 2007 #3


    User Avatar
    Science Advisor
    Gold Member
    Dearly Missed

    Attention everybody! the lights are going out in 8 minutes!
  5. Aug 12, 2007 #4


    User Avatar
    Science Advisor
    Homework Helper

    Utilisation of large optical telescopes would increase by 150%
  6. Aug 13, 2007 #5
    Black holes are ambush predators, not hunters. They only eat things unlucky enough to get into unstable orbits ridiculously close to the black hole.

    Otherwise, they act as an ordinary source of gravity, like the Sun or Jupiter but much bigger. As such, they can attract matter looking for a place to set up a stable orbit toward the center of the galaxy, or expel matter in an unstable orbit away from the center (or just leave it to escape to its freedom). The supermassive black hole can serve as the nucleus around which the central regions of a galaxy organizes itself, like a solar system orbiting a star.

    And as for that little bit of unlucky matter that they *do* eat, that matter first must dissipate all its angular momentum through intense radiative heat loss upon friction with other unlucky matter. This produces powerful radiation that can blow away stray gases and other small particles...and appears to us as a quasar or active galactic nucleus (a more general term referring to quasars and similar but weaker phenomena of supermassive black holes snacking). So the matter about to be "eaten" by the black hole can also be like a star, and its radiation pressure can clean out junk the middle of a galaxy, much as stars blow away gas in their birth places once they "turn on."

    So black holes are not simply destructive forces. At the right distance, they can act as organizing principles. And the vast majority of matter will be at the right distance, whether flung there by radiation pressure from matter on its way to being eaten or unstable orbital dynamics with other matter near the center, or never having ventured dangerously close in the first place.
  7. Aug 20, 2007 #6
    ok I understand that a BH doesnot reach out and suck stuff in

    a million solar mass's or even more
    and so early in the univerce's time span
    that they form the centers of galixicys

    does anyone think a few solar mass BH just grows
    into a million solar mass [without eating other BH's]
    by infalling dust and gas ?????????

    btw is there a size limit for a super nova, or the BH formed by one??
  8. Aug 20, 2007 #7
    How they get supermassive: I don't think anyone knows for sure. But it's not thought that these guys formed from any single star. I've heard that they may be the direct result of gas collapsing into the center of the galaxy, not necessarily having ever been a star or a group of stars first - just that you get so much gas so closely packed that it collapses in on itself before it could realistically blow itself apart.

    Because their mass is generally a roughly fixed proportion of the mass of the bulge of a galaxy (a roundish relatively dense region consisting mostly of old/cold stars with moderate to high heavy element abundances, found in many elliptical and spiral galaxies) - and galaxies with no bulges (e.g. M33/Triangulum, which has only a nuclear star cluster or pseudo-bulge) tend to have no supermassive black holes either - it's thought that the supermassive black holes may be a by-product of whatever forms bulge structures in galaxies. So, they don't form like regular stellar black holes do, and they probably form largely intact pretty early on.

    They are then fed by the earliest phases of star formation in the centers of galaxies, where it's all most intense and concentrated. They eat as much as they can of the gas left over from star formation, and blow the rest out. That helps them to become even more massive.

    The mass limit for black holes formed by a supernova will depend on the mass limit of stars, which as best we can tell is maybe a couple hundred solar masses at the VERY most, 40 solar masses more realistically speaking with today's stars. The very earliest stars were generally about a couple hundred solar masses, so they think.

    Some articles from a quick Google search:
    http://www.users.muohio.edu/whiteak1/Black%20hole%20formation.html [Broken]

    In the second link, I favor some combination of the second and third theories...a giant gas cloud or ultra-dense star cluster got the black hole started, and more gas at the center added to it.
    Last edited by a moderator: May 3, 2017
  9. Aug 21, 2007 #8
    You might find the episode 'Supermassive black holes' from the UK documentary series Horizon of interest regarding the search for a connection between SM black holes and galaxy formation. There are a couple of clips posted on YouTube, just type the above into the search.

  10. Aug 21, 2007 #9

    George Jones

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    As mass increases, the density of matter necessary to form a black hole decreases. If fact, if enough stars are used, they don't even have to touch for a black hole to form. In other words the stars have to be close together, but there still can be space between them. Below, I calculate a quantity that I'll call density, but, in reality, the quantity is only suggestive of density.

    Setting this "density" to the average density of the sun, about 1400 kg/m^3, gives a black hole mass of about 100 million solar masses. So, if more than 100 million or so (within an order of magnitude) sunlike stars congregate in the centre of a galaxy, they don't have to touch (initially) to form a black hole.

    The following calculation is only suggestive, and it is in no way rigorous. Because of the curvature and nature of spacetime, it probably doesn't make sense to calculate the spatial volume inside the event horizon of a black hole.

    Density is mass over volume, i.e.,

    [tex]\rho = \frac{M}{V},[/tex]

    and the volume of a spherical object of radius [itex]R[/itex] is given by [itex]4\pi R^3/3[/itex], so the density of a uniform sphere is

    [tex]\rho = \frac{3M}{4\pi R^3}.[/tex]

    A spherical black hole has event horizon (boundary) located at

    [tex]R = \frac{2GM}{c^2},[/tex]

    where [itex]G[/itex] is Newton's gravitational constant and [itex]c[/itex] is the speed of light.

    Subsituting this equation into the density equation for a spherical black hole gives

    [tex]\rho = \frac{3c^6}{32\pi G^3} \frac{1}{M^2}.[/tex]

    The first bit is just a constant number, while the second bit shows that the "density" of a spherical black rapidly decreases as mass increases.

    Inverting this equation gives

    [tex]M = \frac{c^3}{4}\sqrt{\frac{3}{2\pi G^3}}\sqrt{\frac{1}{\rho}},[/tex]

    and using the Sun's density for [itex]\rho[/itex] gives the result I mentioned at the top.
    Last edited: Aug 21, 2007
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook