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

Behavior of Systems

  1. Jun 25, 2008 #1
    Why is it that systems organize? For example, the planets orbit around the sun and the stars orbit around in their respective galaxies. Another example on a smaller scale is the atom. The electrons move about in probable clouds around the nucleus (if I got this right). Why do systems do this?
  2. jcsd
  3. Jun 25, 2008 #2


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Well, most "systems" you talk about must have some form of stability, otherwise they would already be gone, right ? Turns out that as a function of certain interactions, stable systems, or at least systems with a long lifetime, take on certain forms of dynamics. In gravity for instance, orbital motion is the most common form. If planets weren't orbiting, they'd drop right into the sun and we wouldn't know they have existed. Also, there are certain conservation laws, like conservation of angular momentum, which mean that once a planet is on an orbital motion, it is not going to jump around, but rather remain in that orbital motion.
  4. Jun 25, 2008 #3
    Makes sense...what about on a subatomic scale? Why are atoms organized the way they are.
  5. Jun 25, 2008 #4


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    It's pretty similar: the stable or almost stable structures are given by the quantum mechanical description that results from the interaction between the components. BTW, quantum-mechanically, electrons don't turn around the nucleus as in a planetary system. That's a pretty rough and basically erroneous model of the atom (which nevertheless seems to be rooted in popular "knowledge"). The closest description would be that the electrons essentially "sit there".
  6. Jun 25, 2008 #5
    The answer is because 'systems undergo dynamic in such a way that they minimize action (or to be more precise such that they take the path of stationary action)'. This may not seem the most illumintating answer but for example, I do molecular simulations and all a simulation is doing is following the dynamics derived from calculus of variations and yet the result is a tendency to order. For your example of an electron, if say you have a nucleus which occasionally comes in contact with an electron, it is highly probably that the nucleus will 'capture' it and once an electron is 'attached' to a given atom is requires a large amount of energy to remove it (and thus this event is less likely at terrestrial temperature). Thus it's simply a matter of probability that we would expect most atoms to have full electron shells. (Of course things are more complicated then that but that's a sort of simplistic way to look at it)
  7. Jun 27, 2008 #6
    I don't agree. "Sit there" doesn't do a good job of capturing the kinetic energy or the concept of energy levels. Ordinary people without giant hulking brains need an approximation to understand things, or they would have to settle for "it works cause God says so". The orbit approximation helped me understand a lot of concepts throughout school, despite the fact that, as far as I know to date, I'm too dense to grasp QM.

    In fact, even the concept of quantization itself was explained in a way that fit neatly into my brain in this framework: "you know how particles act like waves with some frequency and interfere and stuff? well so if you have a wave orbiting around a nucleus then it interferes with itself so it's a standing wave right? so then it must have discrete energy levels" and that let me get a handle on things enough that I got through some pretty tough classes. It's an approximation and I know it, but why throw out my one and only dented, rusty shovel just because it ain't a backhoe?

    It may mislead me occasionally, but "everything is random and there are no rules la la la la la" is a lot worse.
  8. Jun 27, 2008 #7


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Well, in the scholarly hydrogen atom, in the ground state, there's one single electron in the s-state. The s-state has its highest probability of presence ON TOP of the nucleus, and no angular momentum (no rotational motion). Ok, it does have kinetic energy (equal to minus the total energy, as the potential energy is twice the total energy):
    KE = 13.6 eV
    PE = -27.2 eV
    Etot = -13.6 eV
  9. Jun 27, 2008 #8

    Andy Resnick

    User Avatar
    Science Advisor
    Education Advisor

    My desk is a counterexample- that system tends to complete disorganization.

    I wonder if you are confusing an analytical, abstracted, property of systems (periodicity, stability to perturbations, etc) with some "innate" property of objects.
  10. Jun 27, 2008 #9
    Well as far my knowledge of physics (which is not super) and the question of your is understood i would like to highlight following points:
    1.) System selection depends on problem concerned.For simplicity we select such system whis is organised,more properly symmetric.
    2.) Every system of universe tends toward lower potential energy,when it is being lowered system tends to be organised or symmetric.
    Please correct the answer if i made soethunf wrong.
  11. Jun 27, 2008 #10
    So are these systems just probable outcomes or are they improbable outcomes that exist only because they allow for existence?
  12. Jun 27, 2008 #11


    User Avatar

    What sort of systems? Where do you put the boundary in what is and is not allowed to be part of a "system"? Like someone said, a desk is a system but rarely is it organised :biggrin:

    I think it's the other way round, everything is inherently disorderly but we just highlight the more ordered situations.
  13. Jun 27, 2008 #12
    Heh...that makes sense.
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?

Similar Discussions: Behavior of Systems
  1. AntiSocial Behavior (Replies: 13)

  2. Ferrofluid behavior? (Replies: 8)