1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Solution to EM wave

  1. Jan 15, 2009 #1


    User Avatar

    I know that in free space, the general solution of the wave equation about electric field is of sine and cosine form. One can also write it in complex form as

    [tex]E = E_0 \exp(i\vec{k}\cdot\vec{r} - i\omega t)[/tex]

    I have two queations about this solution

    1) If consider the polarization, how should I rewrite the solution?

    2) In the presence of polarization, will the angular frequency [tex]\omega[/tex] depend on polarization?
  2. jcsd
  3. Jan 15, 2009 #2


    User Avatar

    I just find the solution involves polarization in another textbook, it reads

    [tex]\vec{E} = \tilde{E} \exp(i\vec{k}\cdot\vec{r} - i\omega t)[/tex]

    where [tex]\tilde{E}[/tex] is complex amplitude represent the amplitude and the polarization. The book read: for finding the general multi-mode solution, we have to add up all possible solutions of above form

    [tex]\sum_{k}\sum_p\tilde{E}_{k,p} \exp(i\vec{k}\cdot\vec{r} - i\omega_{k, p} t)[/tex]

    Now the complex amplitude depends on wavenumber k and p (polarization I guess). Well, I don't understand why a 'p' there? If we need to sum over p, so what values of p is allowed? In the book, it said p can be 1 or 2, but why is that? The last question is about the frequency, why it also depends on p?
  4. Jan 16, 2009 #3
    That is because the EM wave is a transverse wave. The direction of the wave propagation is "k". So the E vector is in the plane perpendicular to the "k" vector. So E can be given in a basis of two vectors in the plane, representing polarization.
    Let [tex]\textbf{n}[/tex] be a unit vector parallel to the direction of wave propagation, that is parallel to [tex]\textbf{k}[/tex]. This means that we can write [tex]\textbf{k}=k\textbf{n}[/tex] where k is the absolute value of the k vector.

    Now lets take a unit vector in the plane perpendicular to the k vector, that is let:

    [tex]\textbf{k}\cdot \textbf{e}^1(\textbf{n})=0[/tex]

    We want an orthonormal basis in this plane. But we know that we can generate a unit vector orthogonal to e^1 via the cross product, so let:

    [tex] \textbf{e}^2(\textbf{n})=\textbf{n}\times\textbf{e}^1(\textbf{n})[/tex]

    Now since the E vector is in this plane, it can be given as a linear combination of the above two unit vectors. So we have for E:


    Using summation notation:


    So the polarization state can be represented by the vector (A_1,A_2). This is called the Jones vector. hence the p index.
  5. Jan 16, 2009 #4


    User Avatar

    Thanks a lot. Very clear :)

    But in your explanation, it seems that the angular frequency only depends on k but not polarization (so no p index in [tex]\omega[/tex])?

    By the way, if it is not plane wave, do we still have something like polarization?
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook