logo

Proof of the Pappus-Guldin Theorem 📂Vector Analysis

Proof of the Pappus-Guldin Theorem

Theorem

20180907\_120248.png

Let the area of a shape FF on the plane be denoted as AA, and let the volume of the solid of revolution WW obtained by rotating FF around axis zz be denoted as VV. If the distance between the center of mass of FF and the axis zz is denoted as rr, then V=2πrA V = 2 \pi r A

Description

The Pappus-Guldin Theorem is often mentioned by teachers when learning about solids of revolution, even though it cannot be proved at the high school level. Surprisingly, even at the undergraduate level of mathematics where it can be proved, it is rarely needed.

Proof

Claim: Since the center of mass of FF is at r=FydAA\displaystyle r = {{ \iint_{F} y dA } \over {A}}, it suffices to show that V=2πFydA\displaystyle V = 2 \pi \iint_{F} y dA .


Part 1.

If function Φ1:FF\Phi_{1} : F ' \to F is defined as Φ(u,v)=(0,y(u,v),z(u,v))\Phi (u,v) = (0,y(u,v),z(u,v)), then Φ1\Phi_{1} becomes bijective. By using this to change coordinates, the Jacobian is det[yuyvzuzv]=yuzvyvzu \left| \det \begin{bmatrix} \displaystyle {{\partial y } \over { \partial u }} & \displaystyle {{\partial y } \over { \partial v }} \\ \displaystyle {{\partial z } \over { \partial u }} & \displaystyle {{\partial z } \over { \partial v }} \end{bmatrix} \right| = \left| {{\partial y } \over { \partial u }} {{\partial z } \over { \partial v }} - {{\partial y } \over { \partial v }} {{\partial z } \over { \partial u }} \right| Therefore, the area of FF is A=FdA=Fyuzvyvzududv A = \iint_{F} dA = \iint_{F ' } \left| {{\partial y } \over { \partial u }} {{\partial z } \over { \partial v }} - {{\partial y } \over { \partial v }} {{\partial z } \over { \partial u }} \right| du dv


Part 2.

Let’s denote W:=F×[0,2π] W’ := F ' \times [0, 2 \pi ]. If function Φ2:WW\Phi_{2} : W’ \to W is defined as Φ(u,v,θ)=(y(u,v)cosθ,y(u,v)sinθ,z(u,v))\Phi (u,v, \theta ) = (y(u,v) \cos \theta ,y(u,v) \sin \theta ,z(u,v)), then Φ2\Phi_{2} becomes bijective. By using this to change coordinates, the Jacobian is det[yucosθyvcosθysinθyusinθyvsinθycosθzuzv0]=yuzvycos2θ+yvzuycos2θ(yuzvysin2θyvzuysin2θ)=yuzvyyvzuy=yyuzvyvzu \begin{align*} &\left| \det \begin{bmatrix} \displaystyle {{\partial y } \over { \partial u }} \cos \theta & \displaystyle {{\partial y } \over { \partial v }} \cos \theta & \displaystyle -y \sin \theta \\ \displaystyle {{\partial y } \over { \partial u }} \sin \theta & \displaystyle {{\partial y } \over { \partial v }} \sin \theta & \displaystyle y \cos \theta \\ \displaystyle {{\partial z } \over { \partial u }} & \displaystyle {{\partial z } \over { \partial v }} & 0 \end{bmatrix} \right| \\ =& \left| - {{\partial y } \over { \partial u }} {{\partial z } \over { \partial v }} y \cos^2 \theta + {{\partial y } \over { \partial v }} {{\partial z } \over { \partial u }} y \cos^2 \theta - \left( {{\partial y } \over { \partial u }} {{\partial z } \over { \partial v }} y \sin^2 \theta - {{\partial y } \over { \partial v }} {{\partial z } \over { \partial u }} y \sin^2 \theta \right) \right| \\ =& \left| {{\partial y } \over { \partial u }} {{\partial z } \over { \partial v }} y - {{\partial y } \over { \partial v }} {{\partial z } \over { \partial u }} y \right| \\ =& y \left| {{\partial y } \over { \partial u }} {{\partial z } \over { \partial v }} - {{\partial y } \over { \partial v }} {{\partial z } \over { \partial u }} \right| \end{align*} The first equality is due to the Laplace expansion related to the third row. Therefore, the volume of WW is V=WdV=Wyyuzvyvzududvdθ V = \iiint_{W} dV = \iiint_{W’} y \left| {{\partial y } \over { \partial u }} {{\partial z } \over { \partial v }} - {{\partial y } \over { \partial v }} {{\partial z } \over { \partial u }} \right| du dv d \theta


Part 3.

Combining the results obtained above V=Wyyuzvyvzududvdθ=02πFyyuzvyvzududvdθ=2πFyyuzvyvzududv=2πFydA \begin{align*} V =& \iiint_{W’} y \left| {{\partial y } \over { \partial u }} {{\partial z } \over { \partial v }} - {{\partial y } \over { \partial v }} {{\partial z } \over { \partial u }} \right| du dv d \theta \\ =& \int_{0}^{2 \pi } \iint_{F’} y \left| {{\partial y } \over { \partial u }} {{\partial z } \over { \partial v }} - {{\partial y } \over { \partial v }} {{\partial z } \over { \partial u }} \right| du dv d \theta \\ & = 2 \pi \iint_{F’} y \left| {{\partial y } \over { \partial u }} {{\partial z } \over { \partial v }} - {{\partial y } \over { \partial v }} {{\partial z } \over { \partial u }} \right| du dv \\ =& 2 \pi \iint_{F} y dA \end{align*}