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

More clues about structure formation (e.g. minimum galaxy mass)

  1. Sep 3, 2008 #1


    User Avatar
    Science Advisor
    Gold Member
    Dearly Missed

    Structure formation is arguably the central problem in cosmology. How did the dilute gas manage to coagulate. On what timetable did stars and galaxies form? What explains the distribution of galaxy sizes and composition--clustering--the wispy cobwebby structure with its various size voids? Cosmic structure formation models play out against a background of assumptions about dark energy and dark matter. The observed structure gives information about how these things must behave. Here's a new study that might provide another piece of the puzzle.

    http://today.uci.edu/news/release_detail.asp?key=1802 [Broken]

    this is a 27 August press release. the article was in 28 August Nature.

    If modified gravity still had a chance, what with news like the bullet cluster and weak lensing maps, even that slim chance now seems wiped out. Depending on where you look, the ratio of dark matter to visible matter can be as little as 10 to one and as large as 1000 to one---it varies especially widely in dwarf galaxies. These have been observed to have a minimum mass of 10 million solar, regardless of how little visible matter they contain. The fact that there are clumps of dark matter with this mass, containing very little visible, suggests that there may be clumps of dark which contain no stars at all.

    This new study (Strigari et al) is surprising enough that I guess a critical look is needed before crediting it---but publication in Nature is already some reassurance. Assuming they are right, why would there be a threshhold mass for clumping of dark matter?

    Here's the Nature article preprint
    Last edited by a moderator: May 3, 2017
  2. jcsd
  3. Sep 3, 2008 #2


    User Avatar
    Science Advisor
    Gold Member
    Dearly Missed

    The key idea this and similar research is that observing structure can actually tell us stuff about dark matter----what it must be and how it must behave in order for the largescale structure we see to have evolved (in the context of the standard cosmology model). Here is a quote from Strigari et al that illustrates this point:

    ...The gravity of dark matter overwhelms that of the normal atoms and molecules and hence governs the formation and evolution of galaxies and large-scale structure [8–10]. In the currently favored dark matter models, structure in the Universe forms hierarchically with smaller gravitationally bound clumps of dark matter – haloes– merging to form progressively larger objects. The mass of the smallest dark matter halo is determined by the particle properties of dark matter. Dark matter candidates characterized as cold dark matter can form haloes that are many orders of magnitude smaller than the least luminous haloes that we infer from observations. Cosmological simulations of cold dark matter predict that galaxies like the Milky Way should be teeming with thousands of dark matter haloes with masses ∼ 10^6 M⊙, with a steadily increasing number as we go to the smallest masses [11–14]. A large class of dark matter candidates characterized as “warm” would predict fewer of these small haloes [15]. However, even for cold dark matter it is uncertain what fraction of the small dark matter haloes should host visible galaxies, as the ability of gas to cool and form stars in small dark matter haloes depends on a variety of poorly-understood physical processes...

    to paraphrase very roughly, computer simulations of structure formation suggest that there should be little dabs of dark matter with masses like a million solar scattered round and about---lots of them. And those would presumably tend to attract some ordinary matter and fire up as dwarf galaxies. But these astronomers found that the minimum mass of dwarf galaxies (even very dim ones consisting mostly of dark matter with very little ordinary) is actually more like TEN million solar. Is there some reasonable explanation? Or could the computer simulations be wrong?

    Or maybe the computer simulations are OK, and there ARE a bunch of dark matter blobs with mass only a million solar. But the catch could be that star formation requires a more massive blob, like ten million solar. Maybe because in order to start star formation you need a first generation of massive stars to cook oxygen and carbon and explode, seeding the original gas with those elements which facilitate condensation. Could it take a more massive, ten million solar, dark matter blob in order to get that process going? I'm just paraphrasing and tossing out ideas to illustrate the considerations that come up in studying structure formation.

    It's a central question in cosmology and essentially it boils down to asking how must dark matter behave in order for the universe to look like it does?
    Last edited: Sep 3, 2008
  4. Sep 3, 2008 #3


    User Avatar
    Science Advisor
    Gold Member
    Dearly Missed

    BTW here is a thread in Beyond the Standard Model forum about a minimal dark matter extension of the standard particle model which provides a dark matter candidate
    In the context of this extension of the standard model, one
    predicts that a certain excess of positrons would be observed in space,
    resulting from dark matter annihilation. And there are suggestions that such an
    abundance of positrons may have been observed by the PAMELA satellite.
    This remains to be checked out and confirmed. I don't know even that it makes sense,
    but there is a Nature News article about it and something published in Physical Review B (2006) and a current preprint. Links are in that other thread, in case anyone is curious.
  5. Sep 5, 2008 #4
    A satisfying feature of cosmology that distinguishes it from the frontiers of much of modern physics is that it is fed by a continuous stream of new observations from ground- and space-based instruments, funded by taxpayers.

    First: thanks, all you taxpayers, and thanks Marcus, for making this modern version of Zwicky's observations so quickly known to folk like me in remote places.

    Second: I've long been intrigued by the importance of the Virial Theorem in structure formation. Because dark matter gravitates I assume that this theorem also applies to aggregates of dark matter, like the haloes of small galaxies orbiting the Milky Way, as observed by UCI scientists like James Bullock, together with Joshua Simon of the California Institute of Technology, Marla Geha of Yale University, Beth Willman of the Harvard-Smithsonian Center for Astrophysics, and Matthew Walker of the University of Cambridge.

    Is my assumption correct?

    Third: When applied to structure formation the Virial Theorem quantifies for specific force-laws a quite general phenomenon, namely that if a structure is to condense from a more diffuse state, there must exist a mechanism for disposing of surplus kinetic energy, so that the condensing structure can preserve virial equilibrium. In specific situations the mechanism can tell us something about the condensate, and vice-versa. Examples are:

    (a) When hydrogen atoms 'condense' into their ground states out of a plasma of free electrons and protons, surplus kinetic energy is carried away by emitted photons. This spectral light tell us a lot about the energy states of hydrogen.

    (b) When a star 'condenses' out of a a gas/dust cloud, as in say the Orion Nebula, surplus kinetic energy of the condensate must be removed by starlight, throughout the process. This tells us quite a lot about the nascent star and its environment.

    Fourth: In the case of dark matter condensing into galactic structures, how is its surplus kinetic energy removed? Dark matter is thought to be exotic stuff immune to electomagnetic interactions (as verified by the bullet cluster observations), so preumably it can't emit photons. Is virial equilibrium maintained by tidal heating of ordinary matter by the gravity of dark matter? Or is some more exotic mechanism thought to be at work here?
  6. Sep 5, 2008 #5


    User Avatar
    Science Advisor
    Gold Member
    Dearly Missed

    Oldman, I am very glad to see you---who have as I recall made a number of interesting comments and contributed excellent questions over the past couple of years. Also it was you who referred to the condensation of stars and other structure as curdling. A descriptive term.

    It is an interesting question. I wont try to speculate but I will add one cooling consideration.

    The expansion of space (the pattern of increasing distances) does itself deprive particles of kinetic energy. I don't have Weinberg's new cosmology book but I gather that he gives a proof of this at a textbook level. I think it is well-known.

    So that means that a PART of the cooling could happen like this: two galaxies each surrounded by their clouds of dark matter collide and merge and, as we see in computer simulations, some energy is ejected in a big sloppy DOLLOP of stars and dark matter, while the majorities on both sides coallesce. So the bills have been paid by having SOME dark matter take on EXTRA kinetic energy. So the books balance, but what happens to this erstwhile cold dark matter which is now in a sense hotter?

    What could happen is that the sacrificial scapegoat dark matter which has absorbed surplus kinetic is gradually cooled back down by the expansion of space. so it eventually gets another chance to join in a curdle. The same would apply to the ejected stars.

    I can also imagine processes whereby gravitational interaction between dark and ordinary matter dumps some energy into the ordinary matter----which can then get rid of it as radiation in the usual ways-----like collisions freaking out CO molecules which then radiate away the jitters of unwanted energy
  7. Sep 6, 2008 #6
    I do appreciate your kind remarks. Thanks, Marcus. The idea you express above is along the same lines as the suggestion I made about tidal interactions: equilibrium being maintained by gravitational inteactions between dark and normal matter. Seems reasonable.

    The more I think about this "interesting question", the more intrigued I get. Consider what would happen if dark matter were to collapse in the absence of normal matter. If indeed dark matter truly is"...hypothetical matter that does not interact with the electromagnetic force, but whose presence can be inferred from gravitational effects on visible matter " (Wikipedia), as many who write about it seem to speculate, there may be surprising consequences.

    Without electromagnetic interactions with normal matter, the attainment of virial equilibrium as collapse proceeds is inhibited. What then is to terminate collapse with the formation of a stable structure? (Contrast the way first gas heating and later nuclear ignition converts the kinetic energy of collapse of normal matter into escaping radiation and so enables the formation of a stable star). Pehaps something along the lines you suggested -- "The expansion of space (the pattern of increasing distances) does itself deprive particles of kinetic energy." becomes important, although this is perhaps a bit on the slow side?

    But there is also the possibility that for pure dark matter that feels only gravitation, collapse can't be stopped until it's too late and an event horizon forms.

    Which brings me to an interesting suggestion. Maybe it is the salting of dark matter with ordinary matter that produces galaxies as we know them; luminous ordinary matter embedded in dark matter haloes centred on black holes that are perhape generated by the collapse of dark matter that has too little salt. There's a crazy speculation for you!
  8. Sep 6, 2008 #7
    Here is an additional thought. If dark matter particles can only interact gravitationally with normal particles and with other dark matter particles then dark matter can not gravitationally collapse. The reasoning goes like this. Imagine an object that has been dropped down a hole drilled through a planet. It falls straight through until it pops out the other side and then turns around and falls back to its original starting point and then continues to oscillate indefinitely like this, in a manner that can be described as simple harmonic motion. I am of course assuming there is no atmosphere to create friction and damp the oscillation eventually bringing the object to rest at the centre of the planet. That is sort of the point of this post. Without some non-gravitational interaction causing friction there is no damping and no gravitational collapse, particles just pass straight through and continue to oscillate forever.

    [EDIT] This I think is basically what oldman was referring to in relation to the virial theorem and the requirement for kinetic energy to be dispersed in order for collapse to take place. [/EDIT]

    So none-interacting dark matter evenly distributed in the early universe shows no tendency to clump like normal matter. Weakly interacting dark matter might have some tendency to clump but not to the extent of normal matter. This is partly supported by the observation that the dark matter of galaxies tends to extend way beyond the radius of normal luminous matter in a galaxy.

    It also raises the question as to why dark matter seems to be centred on galaxies and does not remain evenly distributed throughout the universe. Since science does not seem to have a definite answer as to what exactly dark matter is composed of, I hope I am allowed to make this speculation. It has been observed that the amount of dark matter in galaxies is related to the size of the supper massive black holes at their centres, so is it possible that dark matter is generated by those black holes? One idea is inspired by discussions of Hawking radiation. Virtual particles that pass through a event horizon take on "real" qualities and real particles can become virtual. What if black holes somehow radiate virtual particles?

    The half life of virtual particles is determined by the Heisenberg uncertainty principle and this is typically extremely brief period. What if these ejected virtual particles are gravitationally accelerated to extreme velocities by the black holes so that there half life is greatly extended by time dilation and with their extreme speeds they can survive to great distances before expiring? This would create a sphere around the black hole (and galaxy) that has a definite radius that can be calculated from the expected energies of the virtual particles and the size of the black hole, the radius being determined by the expected half life of the relativistic virtual particles. This idea has an analogue in the virtual particles that are thought to be responsible for the strong force in the nucleus of atoms and this idea nicely explains the very limited radius of the strong force. The volume and surface area of the black hole presumably determines the amount of virtual particles that are generated by the black hole as random production of virtual particles is related to the volume of vacuum under consideration. So in this idea, rather than clumping of dark matter seeding the clumping of normal matter, normal matter clumps according to small random fluctuations in the mass distribution of the early universe (possibly assisted by primordial black holes created during the big bang that would survive much longer due to the great amount of back ground radiation preventing their immediate evaporation). Once normal matter clumps and initial black holes form, the generation of dark matter by those black holes accelerates the clumping and structure forming process, possibly exponentially. Just food for thought. I hope it is of interest and inspirational to someone ;)
    Last edited: Sep 6, 2008
  9. Sep 6, 2008 #8


    User Avatar
    Science Advisor
    Gold Member
    Dearly Missed

    kev, your conclusion is not logical. please read some of the preceding posts in this thread.

    several mechanisms have been suggested by which dark matter, while undergoing clumping---collecting into large clouds---could dispose of excess kinetic energy.
    Last edited: Sep 6, 2008
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook