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

Close to Sturm-Liouville form

  1. Feb 25, 2010 #1
    Close to Sturm-Liouville form....

    I got an O.D.E down to the form

    fll(x) + ([tex]\lambda[/tex] - 16x2)f(x) = 0

    I omitted some constants to make it look simple. What I'm trying to do is find a function f(x) to normalize. Solving by using roots ended up giving me an exponential function I am unable to solve. However I think if I could convert the above to proper Sturm-Liouville form I might find an alternative expression for y(x) so that I could normalize its square. Any thoughts? I'm not completely competent in the aspects D.E as of yet so I may have missed a simpler route. Input is greatly appreciated.
     
  2. jcsd
  3. Feb 25, 2010 #2
    Re: Close to Sturm-Liouville form....

    You look into modified Bessel, Parametric Bessel equation? I don't have the answer, just look very similiar to one of those.
     
  4. Feb 26, 2010 #3
    Re: Close to Sturm-Liouville form....

    I don't know what you mean by using roots, but your equation is inhomogeneous. This is actually a rescaled version of the quantum harmonic oscillator, the solutions of which are given in terms of Hermite functions. Look here for a concise reference:
    http://www.fisica.net/quantica/quantum_harmonic_oscillator_lecture.pdf
     
  5. Feb 26, 2010 #4
    Re: Close to Sturm-Liouville form....

    The general solution to your ODE is as follows

    [tex]f(x) = \frac{1}{\sqrt{x}}[C_1 WhittakerM(\frac{\lambda}{16},\frac{1}{4},4x^2)+C_2 WhittakerW(\frac{\lambda}{16},\frac{1}{4},4x^2)][/tex]

    where C1 and C2 are arbitrary constants.
     
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook