2
$\begingroup$

The associate Legendre equation is given as:

$(1-x^2)\frac{d^2}{dx^2}y-2x\frac{d}{dx}y+\left[n(n+1)-\frac{m^2}{1-x^2}\right] y=0$

This becomes the standard unassociated Legendre equation for $m=0$. For positive m, the solutions to this equation can be expressed in terms of unassociated Legendre polynomials ($P_n(x)$) as:

$P_n^m(x)=(1-x^2)^{m/2}\frac{d^m}{dx^m}P_n(x)$

I would therefore expect that if I use this relationship as a solution to the associated Legendre equation that I should recover the unassociated Legendre equation.

I find the first and second derivative to be:

$\frac{d}{dx}P_n^m=(1-x^2)^{m/2}\left[\frac{d^{m+1}}{dx^{m+1}}P_n -\frac{mx}{1-x^2}\frac{d^m}{dx^m}P_n \right]$

$\frac{d^2}{dx^2}P_n^m=(1-x^2)^{m/2}\left[\frac{d^{m+2}}{dx^{m+2}}P_n-\frac{2mx}{1-x^2}\frac{d^{m+1}}{dx^{m+1}}P_n+\frac{m(m-1)x^2-m}{(1-x^2)^2}\frac{d^m}{dx^m}P_n \right]$

Using these two results with the associated Legendre equation and simplifying I find:

$(1-x^2)\frac{d^2}{dx^2}P_n-2x\frac{d}{dx}P_n+n(n+1)P_n-m(2x\frac{d}{dx}P_n+m(m+1)P_n)=0$

Clearly this is the unassociated Legendre equation with a little bit of extra stuff that didn't cancel. I am not seeing any way to make these terms vanish though--so what am I doing wrong here?

$\endgroup$
1
$\begingroup$

First of all, the result you achieve is wrong - somewhere along the line you have replaced $P_n^{(m)}$ (I use the shortcut notation $f^{(m)}$ for the $m^{th}$ derivative of $f$) by $P_n$.

The equation you should get is (in my notation): $$(1-x^2)P_n^{(m+2)}-2(m+1)xP_n^{(m+1)}+\{ n(n+1)-m(m+1)\} P_n^{(m)} = 0$$

Now this is not the unassociated Legendre equation, obviously, but it can be derived from that equation by differentiating it $m$ times.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.