<?xml version="1.0" encoding="UTF-8"?>

<record version="13" id="100">
 <title>Laplacian in Spherical Coordinates</title>
 <name>LaplacianInSphericalCoordinates</name>
 <created>2005-12-30 20:39:07</created>
 <modified>2006-01-02 19:45:29</modified>
 <type>Definition</type>
 <creator id="1" name="bloftin"/>
 <modifier id="1" name="bloftin"/>
 <author id="1" name="bloftin"/>
 <classification>
	<category scheme="msc" code="02.40.Dr"/>
 </classification>
 <preamble>% this is the default PlanetMath preamble.  as your knowledge
% of TeX increases, you will probably want to edit this, but
% it should be fine as is for beginners.

% almost certainly you want these
\usepackage{amssymb}
\usepackage{amsmath}
\usepackage{amsfonts}

% used for TeXing text within eps files
%\usepackage{psfrag}
% need this for including graphics (\includegraphics)
\usepackage{graphicx}
% for neatly defining theorems and propositions
%\usepackage{amsthm}
% making logically defined graphics
%\usepackage{xypic}

% there are many more packages, add them here as you need them

% define commands here</preamble>
 <content>The Laplacian operator in spherical coordinates is 
\begin{equation}
\nabla _{sph}^{2} = \frac{1}{r^2} \frac{\partial}{\partial r}(r^2 \frac{\partial}{\partial r}) + \frac{1}{r^2 sin\theta} \frac{\partial}{\partial \theta} ( sin \theta \frac{\partial}{\partial \theta}) + \frac{1}{r^2 sin^2 \theta} \frac{\partial^2}{\partial \phi^2}
\end{equation}

The derivation is fairly straight forward and begins with locating a vector {\bf r} in spherical coordinates as shown in the figure.

\begin{figure}
\includegraphics[scale=.698]{SphericalCoordinates.eps}
\caption{Spherical Coordinates}
\end{figure}

The z component of the unit vector in direction of {\bf r} is given from the simple right triangle

$$ cos \theta = \frac{z}{| \hat{r} |} $$

Since a unit vector has a length of 1, the z component is

$$ z = cos \theta $$

To get the x component, we need to get the $\hat{r}$ projected onto the xy-plane

$$ cos( 90 - \theta) = \frac{ |\hat{r}_{xy}|}{ |\hat{r}| } $$

using the trig identity

$$ cos( 90 - \theta ) = sin \theta $$

the projected unit vector is

$$ |\hat{r}_{xy}| = sin \theta $$

Finally, the x component is reached through the right triangle 

$$ cos \phi = \frac{x}{|\hat{r}_{xy}| }$$

giving 

$$ x = cos \phi \, sin \theta $$

the y component follows the x component through

$$ sin \phi = \frac{y}{|\hat{r}_{xy}|} $$

which yields

$$ y = sin \phi \, sin \theta $$

The vector in spherical coordinates is then
\begin{equation}
{\bf r} = r \hat{r} = r[ sin \theta \, cos \phi \, \hat{i} + sin \theta \, sin \phi \, \hat{j} + cos \theta \, \hat{k}]
\end{equation}

Now describing the unit vectors of a moving particle $ \hat{r}, \hat{\theta}, \hat{\phi}$ shown in the spherical coordinates figure is a little more tricky.  If a particle moves in the $\hat{r}$ direction, it only moves in or out along r so

$$ \frac{\partial {\bf r}}{ \partial r} = k \hat{r} $$

For $\hat{\theta}$ and $\hat{\phi}$, think of uniform circular motion like a record, so they can be calculated from

$$ \hat{\theta} = \frac{\frac{ \partial {\bf r}}{\partial \theta}}{| \frac{ \partial {\bf r}}{\partial \theta}|} $$


$$ \hat{\theta} = \frac{\frac{ \partial {\bf r}}{\partial \phi}}{| \frac{ \partial {\bf r}}{\partial \phi}|} $$

Next, take the partial derivatives of (2) and then calculate their magnitudes to get the above unit vectors.

$$ \frac{ \partial {\bf r}}{\partial \theta} = r [cos \theta \, cos \phi \hat{i} + cos \theta \, sin \phi \hat{j} - sin \theta \hat{k}] $$

$$ \frac{ \partial {\bf r}}{\partial \phi} = r [-sin \theta \, sin \phi \hat{i} + sin \theta \, cos \phi \hat{j}] $$

$$ | \frac{ \partial {\bf r}}{\partial \theta}| = \sqrt{ {\bf x} \cdot {\bf x}} = \sqrt{r^2 [cos^2 \theta \, cos^2 \phi + cos^2 \theta \, sin^2 \phi + sin^2 \theta]} $$

$$ | \frac{ \partial {\bf r}}{\partial \theta}| = r \sqrt{cos^2 \theta (cos^2 \phi + sin^2 \phi) + sin^2 \theta]} $$

Using the basic trig identity

$$ | \frac{ \partial {\bf r}}{\partial \theta}| = r $$

$$ | \frac{ \partial {\bf r}}{\partial \phi}| = r \sqrt{sin^2 \theta (sin^2 \phi + cos^2 \phi)]} $$

$$ | \frac{ \partial {\bf r}}{\partial \phi}| = r sin \theta $$

Then plug in these values to get

$$ \hat{\phi} = -sin \phi \hat{i} + cos \phi \hat{j}$$
$$ \hat{\theta} = cos \theta \, cos \phi \hat{i} + cos \theta \, sin \phi \hat{j} - sin \theta \hat{k} $$

and from before we had

$$ \hat{r} = sin \theta \, cos \phi \hat{i} + sin \theta \, sin \phi \hat{j} + cos \theta \hat {k} $$

The generalized differential for curvilinear coordinates is

$$ d {\bf r} = \frac{ \partial {\bf r}}{\partial u_1} du_1 + \frac{ \partial {\bf r}}{\partial u_2} du_2 + \frac{ \partial {\bf r}}{\partial u_3} du_3 $$

For spherical coordinates we have

$$ u_1 = r $$
$$ u_2 = \theta $$
$$ u_3 = \phi $$

and from earlier we learned

$$ | \frac{ \partial {\bf r}}{\partial r}| \hat{r} = \frac{ \partial {\bf r}}{\partial r} $$

$$ | \frac{ \partial {\bf r}}{\partial \theta}| \hat{\theta} = \frac{ \partial {\bf r}}{\partial \theta} $$


$$ | \frac{ \partial {\bf r}}{\partial \phi}| \hat{\theta} = \frac{ \partial {\bf r}}{\partial \phi} $$

Plugging these values into the generalized differential yields

$$ d {\bf r} = dr \hat{r} + r d \theta \hat{\theta} + r sin \theta d \phi \hat{\phi} $$

The next major step is to see how the gradient fits into the definition of the differential for a function f

$$ df = \frac{ \partial f}{\partial r} dr + \frac{ \partial f}{\partial \theta} d \theta + \frac{ \partial f}{\partial \phi} d \phi $$
$$ df = \nabla f \cdot d {\bf r} $$

So we see that the following must be equal and we need to solve for the gradient's components.

$$ \frac{ \partial f}{\partial r} dr = \nabla_r f dr $$
$$ \frac{ \partial f}{\partial \theta} d \theta = \nabla_{\theta} f d \theta $$
$$ \frac{ \partial f}{\partial \phi} d \phi = \nabla_{\phi} f d \phi $$

Therfore our scale factors are

$$ \nabla_r = 1 $$
$$ \nabla_{\theta} = \frac{1}{r} $$
$$ \nabla_{\phi} = \frac{1}{r sin \theta} $$

which finally gives us the gradient in spherical coordinates

$$ \nabla_{sph} = \frac{\partial}{\partial r} \hat{r} + \frac{\partial}{\partial \theta} \frac{\hat{\theta}}{r} + \frac{\partial}{\partial \phi} \frac{\hat{\phi}}{rsin \theta} $$

The last tedious calculation is then the Laplacian, which is our goal

$$ \nabla^2 = \nabla \cdot \nabla = ( \frac{\partial}{\partial r} \hat{r} + \frac{\partial}{\partial \theta} \frac{\hat{\theta}}{r} + \frac{\partial}{\partial \phi} \frac{\hat{\phi}}{rsin \theta}) \cdot (\frac{\partial}{\partial r} \hat{r} + \frac{\partial}{\partial \theta} \frac{\hat{\theta}}{r} + \frac{\partial}{\partial \phi} \frac{\hat{\phi}}{rsin \theta}) $$

Let us break it up by components to make it easy to view, so carrying out only part of the dot product our 1st term is

$$ \hat{r} \cdot ( \frac{\partial \hat{r}}{\partial r} \frac{\partial}{\partial r} + \hat{r} \frac{\partial^2}{\partial r^2} + \frac{\partial \hat{\theta}}{\partial r} \frac{1}{r} \frac{\partial}{\partial \theta} +\hat{\theta} \frac{\partial}{\partial r} \frac{1}{r} \frac{\partial}{\partial \theta} + \frac{\hat{\theta}}{r} \frac{\partial^2}{\partial r \partial \theta} + \frac{\partial \hat{\phi}}{\partial r} \frac{1}{r sin \theta} \frac{\partial}{\partial \phi} + \hat{\phi}\frac{\partial}{\partial r}(\frac{1}{r sin \theta}) \frac{\partial}{\partial \phi} + \frac{\hat{\phi}}{r sin \theta} \frac{\partial^2}{\partial r \partial \phi} $$

While this looks a little scary, all but the 2nd term is zero.  The first term is zero because there is no r in $\hat{r}$


$$ \hat{r} = sin \theta \, cos \phi \hat{i} + sin \theta \, sin \phi \hat{j} + cos \theta \hat {k} $$

so taking the derivative with respect to r yeilds zero.  Similarly, the 3rd and 6th term are zero when taking the partial derivatives.  The 4th,5th,7th and 8th terms are zero because the dot product of two orthogonal vectors ($90^o$) is zero so

$$ \hat{r} \cdot \hat{\theta} = 0 $$
$$ \hat{r} \cdot \hat{\phi} = 0 $$</content>
</record>
