PlanetMath (more info)
 Math for the people, by the people. Sponsor PlanetMath
Encyclopedia | Requests | Forums | Docs | Wiki | Random | RSS  
Login
create new user
name:
pass:
forget your password?
Main Menu
Owner confidence rating: High Entry average rating: No information on entry rating
[parent] proof to Cauchy-Riemann equations (polar coordinates) (Proof)

If $f(z)$ is differentialble at $z_0$ then the following limit \begin{eqnarray*} f'(z_0) & = & \lim_{\xi \to 0} \frac{f(z_0+\xi)-f(z_0)}{\xi} \end{eqnarray*}will remain the same approaching from any direction. First we fix $\theta$ as $\theta_0$ then we take the limit along the ray where the argument is equal to $\theta_0$ Then \begin{eqnarray*} f'(z_0) & = & \lim_{h \to 0} \frac{f(r_0e^{i\theta_0} + he^{i\theta_0})-f(r_0e^{i\theta_0})}{he^{i\theta_0}} \\ & = & \lim_{h\to 0} \frac{f((r_0+h)e^{i\theta_0})-f(r_0e^{i\theta_0})}{he^{i\theta_0}} \\ & = & \lim_{h\to 0} \frac{u(r_0+h,\theta_0) + iv(r_0+h,\theta_0) - u(r_0,\theta_0) - iv(r_0,\theta_0)}{he^{i\theta_0}}\\ & = & \frac{1}{e^{i\theta_0}} \Bigg[ \lim_{h\to 0} \frac{u(r_0+h,\theta_0) - u(r_0,\theta_0)}{h} + i \lim_{h\to 0} \frac{v(r_0+h,\theta_0) - v(r_0,\theta_0)}{h} \Bigg]\\ & = & \frac{1}{e^{i\theta_0}}\Bigg[ \frac{\partial u}{\partial r}(r_0,\theta_0) + i\frac{\partial v}{\partial r}(r_0,\theta_0) \Bigg] \end{eqnarray*} Similarly, if we take the limit along the circle with fixed $r$ equals $r_0$ Then

\begin{eqnarray*} f'(z_0) & = & \lim_{h\to 0} \frac{f(r_0e^{i\theta_0} + r_0e^{i(\theta_0+h)})-f(r_0e^{i\theta_0})}{r_0e^{i\theta_0}(e^{ih}-1)}\\ & = & \lim_{h\to 0} \frac{f(r_0e^{i(\theta_0+h)})-f(r_0e^{i\theta_0})}{he^{i\theta_0}}\\ & = & \lim_{h\to 0} \frac{u(r_0,\theta_0+h) + iv(r_0,\theta_0+h) -u(r_0,\theta_0) - iv(r_0,\theta_0)}{he^{i\theta_0}}\\ & = & \frac{1}{r_0e^{i\theta_0}} \Bigg[ \lim_{h\to 0} \frac{u(r_0+h,\theta_0) - u(r_0,\theta_0)}{h}\cdot \frac{h}{e^{ih}-1} + i \lim_{h\to 0} \frac{v(r_0+h,\theta_0) - v(r_0,\theta_0)}{h}\cdot \frac{h}{e^{ih}-1} \Bigg]\\ & = & \frac{1}{r_0e^{i\theta_0}} \Bigg[ \lim_{h\to 0} \frac{u(r_0+h,\theta_0) - u(r_0,\theta_0)}{h}\cdot \lim_{h\to 0} \frac{h}{e^{ih}-1} + i \lim_{h\to 0} \frac{v(r_0+h,\theta_0) - v(r_0,\theta_0)}{h}\cdot \lim_{h\to 0} \frac{h}{e^{ih}-1} \Bigg]\\ & = & \frac{1}{r_0e^{i\theta_0}} \Bigg[ \frac{\partial u}{\partial \theta}(r_0,\theta_0)\frac{1}{i} + \frac{\partial v}{\partial \theta}(r_0,\theta_0) \Bigg]\\ & = & \frac{1}{r_0e^{i\theta_0}} \Bigg[ \frac{\partial v}{\partial \theta}(r_0,\theta_0) - i\frac{\partial u}{\partial \theta}(r_0,\theta_0) \Bigg] \end{eqnarray*} Note: We use l'Hôpital's rule to obtain the following result used above $\lim_{h\to 0} \frac{h}{e^{ih}-1} = \frac{1}{i}$

Now, since the limit is the same along the circle and the ray then they are equal: \begin{eqnarray*} \frac{1}{e^{i\theta_0}}\Bigg[ \frac{\partial u}{\partial r}(r_0,\theta_0) + i\frac{\partial v}{\partial r}(r_0,\theta_0) \Bigg] & = & \frac{1}{r_0e^{i\theta_0}} \Bigg[ \frac{\partial v}{\partial \theta}(r_0,\theta_0) - i\frac{\partial u}{\partial \theta}(r_0,\theta_0) \Bigg]\\ \Bigg[ \frac{\partial u}{\partial r}(r_0,\theta_0) + i\frac{\partial v}{\partial r}(r_0,\theta_0) \Bigg] & = & \frac{1}{r_0} \Bigg[ \frac{\partial v}{\partial \theta}(r_0,\theta_0) - i\frac{\partial u}{\partial \theta}(r_0,\theta_0) \Bigg]\\ \end{eqnarray*}which implies that \begin{eqnarray*} \frac{\partial u}{\partial r} & = & \frac{1}{r}\frac{\partial v}{\partial \theta}\\ \frac{\partial v}{\partial r} & = & -\frac{1}{r}\frac{\partial u}{\partial \theta} \end{eqnarray*}QED




"proof to Cauchy-Riemann equations (polar coordinates)" is owned by Daume.
(view preamble | get metadata)

View style:


This object's parent.
Log in to rate this entry.
(view current ratings)

Cross-references: QED, implies, circle, argument, ray, fix, limit

This is version 2 of proof to Cauchy-Riemann equations (polar coordinates), born on 2004-01-06, modified 2004-03-02.
Object id is 5504, canonical name is ProofToCauchyRiemannEquationsPolarCoordinates.
Accessed 10883 times total.

Classification:
AMS MSC30E99 (Functions of a complex variable :: Miscellaneous topics of analysis in the complex domain :: Miscellaneous)

Pending Errata and Addenda
None.
Discussion
Style: Expand: Order:
forum policy
correctness by zbeekman on 2008-02-15 14:05:26
The second part, for varying theta is not correct.
[ reply | up ]

Interact
post | correct | update request | add example | add (any)