1

This question is a follow up to a previous question: Spherical Harmonic Identity.

Instead of using the above question's method, I tried something like this, but don't get the same result and I'm unsure why that is.

$$\partial_{\theta_0} \sum_{m=-\infty}^{\infty}Y_{\ell,m}^*(\theta,\phi)Y_{\ell,m}(\theta_0,\phi_0)=\partial_{\theta_0} Y_{\ell,0}^*(\theta,\phi)Y_{\ell,0}(\theta_0,\phi_0)+\partial_{\theta_0} \sum_{m=-\infty}^{-1}Y_{\ell,m}^*(\theta,\phi)Y_{\ell,m}(\theta_0,\phi_0)+\partial_{\theta_0} \sum_{m=1}^{\infty}Y_{\ell,m}^*(\theta,\phi)Y_{\ell,m}(\theta_0,\phi_0)$$ Then switching the final term to a sum over negative $m$, I get: $$\partial_{\theta_0} \sum_{m=-\infty}^{\infty}Y_{\ell,m}^*(\theta,\phi)Y_{\ell,m}(\theta_0,\phi_0)=\partial_{\theta_0} Y_{\ell,0}^*(\theta,\phi)Y_{\ell,0}(\theta_0,\phi_0)+\partial_{\theta_0} \sum_{m=-\infty}^{-1}Y_{\ell,m}^*(\theta,\phi)Y_{\ell,m}(\theta_0,\phi_0)+\partial_{\theta_0} \sum_{m=-\infty}^{-1}Y_{\ell,-m}^*(\theta,\phi)Y_{\ell,-m}(\theta_0,\phi_0).$$ The two sums can be joined. I also use the identity $Y_{\ell,-m}=(-1)^mY_{\ell,m}^*$. $$\partial_{\theta_0} \sum_{m=-\infty}^{\infty}Y_{\ell,m}^*(\theta,\phi)Y_{\ell,m}(\theta_0,\phi_0)=\partial_{\theta_0} Y_{\ell,0}^*(\theta,\phi)Y_{\ell,0}(\theta_0,\phi_0)+\partial_{\theta_0} \sum_{m=-\infty}^{-1}(Y_{\ell,m}^*(\theta,\phi)Y_{\ell,m}(\theta_0,\phi_0)+Y_{\ell,m}(\theta,\phi)Y_{\ell,m}^*(\theta_0,\phi_0)).$$ Expanding the final term to associated Legendre Polynomials, the result becomes $$\partial_{\theta_0} \sum_{m=-\infty}^{\infty}Y_{\ell,m}^*(\theta,\phi)Y_{\ell,m}(\theta_0,\phi_0)=\partial_{\theta_0} Y_{\ell,0}^*(\theta,\phi)Y_{\ell,0}(\theta_0,\phi_0)+\partial_{\theta_0}\sum_{m=-\infty}^{-1}\left(\frac{2\ell+1}{4\pi}\right)\frac{(\ell-m)!}{(\ell+m)!}P_{\ell,m}(\cos\theta)P_{\ell,m}(\cos\theta_0)(e^{i(\phi-\phi_0)}+e^{-i(\phi-\phi_0)}).$$ Then, using the convention $$P_{\ell,m}=(1-x^2)^{m/2}\frac{d^m}{dx^m}P_\ell(x),$$ and the identity $$\frac{d}{dx}P_{\ell,m}(x)=\frac{mx}{(1-x^2)^{1/2}}P_{\ell,m}(x)-P_{\ell,m+1}(x),$$ the formula above becomes: $$\partial_{\theta_0} \sum_{m=-\infty}^{\infty}Y_{\ell,m}^*(\theta,\phi)Y_{\ell,m}(\theta_0,\phi_0)=\partial_{\theta_0}Y_{\ell,0}^*(\theta,\phi)Y_{\ell,0}(\theta_0,\phi_0)+\sum_{m=-\infty}^{-1}\left(\frac{2\ell+1}{4\pi}\right)\frac{(\ell-m)!}{(\ell+m)!}P_{\ell,m}(\cos\theta)(m\cot\theta_0 P_{\ell,m}(\cos\theta_0)-P_{\ell,m+1}(\cos\theta_0))(2\cos(m\phi-m\phi_0)).$$ Finally, I set $\theta_0=0$, which reduces the above equation to: \begin{align} \partial_{\theta_0} \sum_{m=-\infty}^{\infty}Y_{\ell,m}^*(\theta,\phi)Y_{\ell,m}(\theta_0,\phi_0)&=\sum_{m=-\infty}^{-1}\left(\frac{2\ell+1}{4\pi}\right)\frac{(\ell-m)!}{(\ell+m)!}P_{\ell,m}(\cos\theta)\nonumber\\&(m\cot\theta_0 \delta_{m,0}-\delta_{m,-1})(2\cos(m\phi-m\phi_0))\nonumber\\ &=-2\left(\frac{2\ell+1}{4\pi}\right)\frac{(\ell+1)!}{(\ell-1)!}P_{\ell,-1}(\cos\theta)\cos(\phi-\phi_0)\nonumber\\ &=2\left(\frac{2\ell+1}{4\pi}\right)P_{\ell,1}(\cos\theta)\cos(\phi-\phi_0)\nonumber \end{align}

If there is an error here, I can't find it, maybe someone else can?

Karl
  • 217

1 Answers1

0

I think it might be a problem of the form of the associated Legendre function derivative... from the convention, it can't be mx/(1-x^2)^1/2, is it actually -mx/(1-x^2)? However if we replace x by costheta, and derive by theta, the formula seems the same.

  • Your answer could be improved with additional supporting information. Please [edit] to add further details, such as citations or documentation, so that others can confirm that your answer is correct. You can find more information on how to write good answers in the help center. – Community May 12 '22 at 03:09
  • Can you elaborate more on your answer, and if possible, can you use latex commands to typeset the response? – MathMinded May 12 '22 at 03:55