3

Problem:

Use Stokes’ theorem to evaluate the integral $I = \int\limits_C \textbf{F} \centerdot \textbf{ds}$ when $\textbf{F}$ is the vector field $\textbf{F} = 3zx\textbf{i} + 3xy\textbf{j} + yz\textbf{k}$ and C is the path consisting of the three edges of the triangle $∆ABC$ shown in

enter image description here

formed by the portion of the plane $x + y + z = 1$ in the first octant of 3-space, oriented as shown.

Attempt at Solution:

I know that by Stokes' theorem, $\int\limits_C \textbf{F} \centerdot \textbf{ds} = \iint\limits_S (\nabla \times \textbf{F}) \centerdot \textbf{dS}$, where $S$ is the surface closed by $C$. So, evaluating $\nabla \times \textbf{F}$, I found it to be equal to $z\textbf{i} + 3x\textbf{j} + 3y\textbf{k}$. Substituting $z = 1 - x - y$, $\nabla \times \textbf{F} = (1 - x - y)\textbf{i} + 3x\textbf{j} + 3y\textbf{k}$.

Next, to find $\textbf{dS}$, I took the gradient of $x + y + z = 1$, which turned out to be equal to $<1, 1, 1>$. Thus, $(\nabla \times \textbf{F}) \centerdot \textbf{dS} = 1 + 2x + 2y$.

For the limits of integration, I used the projection of the surface onto the $xy$-plane, which produced the image of a triangle bounded by $x = 0$, $y = 0$, and $y = -x$. Accordingly, I chose the limits of integration as follows: $\int_0^1 \int_0^{-x} (1 + 2x + 2y)dydx$. After performing integration, I found the result to be $-\frac{5}{6}$. However, this answer was incorrect. Where did I go wrong?

Swamp G
  • 337
  • was it off by a factor of $\sqrt{3}$? – user66081 Aug 13 '14 at 22:16
  • Unfortunately, I am not sure. The possible answers (multiple choice) are $I = $ 1, $\frac{1}{2}$, $\frac{5}{6}$, $\frac{2}{3}$, or $\frac{7}{6}$. I attempted $\frac{5}{6}$ thinking I had misplaced a negative sign in my calculations, but I was incorrect. – Swamp G Aug 13 '14 at 22:22
  • 1
    it is $y = 1-x$ – user66081 Aug 13 '14 at 22:29
  • Why is it not off by a factor of root $3$? You are supposed to dot the curl with a unit normal and $<1,1,1>$ is not unit – Sorfosh Aug 19 '18 at 03:02
  • @Sorfosh But then you will have to calculate the scalar $dS$, which, with the parametrization $(x, y, 1 - x - y)$, will give $dS = \sqrt 3 ,dx dy$. – Maxim Sep 24 '18 at 21:24

1 Answers1

1

As user66081 pointed out, I made the simple mistake of setting the upper $y$-boundary in the iterated integral to $-x$ instead of $1-x$. Correcting this, I found that $\int_0^1\int_0^{1-x} (1 + 2x + 2y)dydx = \frac{7}{6}$. Thank you!

Swamp G
  • 337