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

Definition of the eigenfunction?

  1. Mar 11, 2009 #1

    I need another Definition of the eigenfunction

    and the book the include this Definition


    Those values of Lamda for which nontrivial solutions do
    exist are called eigenvalues; the corresponding
    nontrivial solutions are called eigenfunctions.
  2. jcsd
  3. Mar 11, 2009 #2


    User Avatar
    Science Advisor

    Well, that's pretty much the definition you will find in any book. I don't know exactly what your question is. Perhaps some examples would help.

    The problem [itex]d^2y/dx^2+ \lambda y= 0[/itex] with boundary conditions y(0)= 0, y(1)= 0 has the "trivial solution" y(x)= 0 for all x. Does it have any other, "non-trivial", solutions?

    The way we write the general solution depends on [itex]\lambda[/itex]. If, for example, [itex]\lambda= 0[/itex], the equation is [itex]d^2y/dx^2= 0[/itex] and integrating twice,[itex]y= C_1x+ C_2[/itex]. Then [itex]y(0)= C_2= 0[/itex] and [itex]y(1)= C_1+ C_2= C_1= 0[/itex] so both constants are 0 and y(x)= 0 for all x, the trivial solution. 0 is NOT an eigevalue.

    If [itex]\lambda[/itex] is negative then we can simplify the problem by writing [itex]\lambda= -\alpha^2[/itex] where [itex]\alpha[/itex] is any non-zero number. The equation becomes [itex]d^2y/dx^2- \alpha^2 y= 0[/itex]. That has characteristic equation [itex]r^2- \alpha= 0[/itex] which has roots [itex]r= \alpha[/itex] and [itex]r= -\alpha[/itex] so the general solution is [itex]y(x)= C_1e^{\alpha x}+ C_2e^{-\alpha x}[/itex] so [itex]y(0)= C_1+ C_2= 0[/itex] and [itex]y(1)= C_1e^{\alpha}+ C_2e^{-alpha}= 0[/itex]. From the first equation, [itex]C_2= -C_1[/itex]. Putting that into the second equation and factoring out [itex]C_1[/itex], we have [itex]C_1(e^{\alpha}- e^{-\alpha}= 0[/itex]. Because [itex]\alpha[/itex] is not 0, [itex]\alpha[/itex] and [itex]-\alpha[/itex] are different and, since [itex]e^x[/itex] is a "one-to-one" function, [itex]e^\alpha[/itex] and [itex]e^{-\alpha}[/itex] are different. [itex](e^{\alpha}- e^{-\alpha}[/itex] is not 0 so [itex]C_1[/itex] must be 0 which means [itex]C_2= -C_1[/itex] is also zero. That is [itex]y(0)= 0(e^{\alpha x})+ 0(e^{-\alpha x})= 0[/itex] for all x. Again, y(x)= 0 is the only solution so NO negative number is an eigenvalue.

    If [itex]\lambda[/itex] is positive, we can write [itex]\lambda= \alpha^2[/itex] so the equation becomes [itex]d^2y/dx^2+ \alpha^2y= 0[/itex] which has characteristic equation [itex]r^2- \alpha^2[/itex] with solutions [itex]r= i\alpha[/itex] and [itex]r= -i\alpha[/itex] so the general solution is of the form [itex]y(x)= C_1 cos(\alpha x)+ C_2 sin(\alpha x)[/itex]. Now, since cos(0)= 1 and sin(0)= 0, [itex]y(0)= C_1= 0[/itex]. [itex]y(1)= C_2 sin(\alpha)= 0[/itex]. If [itex]sin(\alpha)[/itex] is not 0, then [itex]C_2[/itex] is 0 and again we get y(x)= 0 for all x. But sine is 0 for any multiple of [itex]\pi[/itex], so if [itex]\alpha[/itex] is, say, [itex]n\pi[/itex], [itex]C_2[/itex] does NOT have to be 0: [itex]y(x)= C sin(n\pi x)[/itex] satisfies both the differential equation and the boundary conditions. [itex]n\pi[/itex] is an eigenvalue for for any integer n. The corresponding eigenfunctions are [itex]y(x)= C sin(n\pi x)[/itex] for C any non-zero number.

    Notice by the way, that all of this depends not just on the equation but also on the boundary conditions. Also, as I said, y(x)= 0 for all x always satifies this equation so being an eigenvalue requires that the solution NOT be unique.
  4. Mar 11, 2009 #3

    thank you very very much
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook