Scribd
Upload
31 views

Stokes Theorem Ii Maths21a, O. Knill: y Z Z X X y X y Z

Stokes theorem relates the curl of a vector field integrated over a surface S to the line integral of the vector field itself over the boundary curve C of the surface. In 3 dimensions, if S is a surface with boundary curve C, then the surface integral of curl(F) over S equals the line integral of F around C. Stokes' theorem generalizes this relationship to higher dimensions, where the curl and integrals are defined in higher dimensional forms.

Uploaded by

martim_martins
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
31 views

Stokes Theorem Ii Maths21a, O. Knill: y Z Z X X y X y Z

Stokes theorem relates the curl of a vector field integrated over a surface S to the line integral of the vector field itself over the boundary curve C of the surface. In 3 dimensions, if S is a surface with boundary curve C, then the surface integral of curl(F) over S equals the line integral of F around C. Stokes' theorem generalizes this relationship to higher dimensions, where the curl and integrals are defined in higher dimensional forms.

Uploaded by

martim_martins
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 1

STOKES THEOREM II Maths21a, O.

Knill
curl: curl(M, N, P) = (P
y
N
z
, M
z
P
x
, N
x
M
y
).
gradient: grad(f) = (f
x
, f
y
, f
z
).
Fundamental theorem of line integrals
curve r(t) : [a, b] C with boundary {r(a), r(b)}.

b
a
grad(f)(r(t)) r

(t) dt = f(r(b)) f(r(a))


Stokes theorem:
surface r(u, v) : R S with boundary r(t) : [a, b] C

R
curl(F)(r(u, v)) (r
u
r
v
)dudv =

b
a
F(r(t)) r

(t) dt
REMINDERS:
CURL: curl(F) = F
GRAD: grad(f) = f
FTL:

C
grad(f) dr = f(B) f(A)
STOKES:

S
curl(F) dS =

C
F dr
CURL(GRAD)=0: f =

0
CONSERVATIVE FIELDS. F is conservative if is a gradient eld F = f. The fundamental theorem of line
integral implies that the line integral along closed curves is zero and that line integrals are path independent.
We have also seen that F = grad(f) implies that F has zero curl. If we know that F is conservative, how do we
compute f? If F = (M, N, P) = (f
x
, f
y
, f
z
), we got f by integration. There is an other method which you do
in the homework in two dimensions: to get the potential value f, nd a path C
P
from the origin to the point
P = (x, y, z) and compute

C
F dr. Because line integrals are path independent, the fundamental theorem of
line integrals gives

C
F dr = f(P).
CONNECTED. A region is called connected if one can connect any two points in the region with a path.
SIMPLY CONNECTED. A region is called simply connected if it is connected and every path in the region
can be deformed to a point within the region.
EXAMPLES 2D:
Simply connected. Not simply connected. Simply connected. Not simply connected.
EXAMPLES 3D:
A solid ball is simply con-
nected.
A solid torus is not simply
connected.
The complement of a solid
ball is simply connected.
The complement of a solid
torus is not simply con-
nected.
THEOREM.
In a simply connected region D, a vector eld F is conservative if
and only if curl(F) =

0 everywhere inside D.
Proof. We already know that F = f implies curl(F) =

0. To
show the converse, we verify that the line integral along any closed
curve C in D is zero. This is equivalent to the path independence
and allows the construction of the potential f with F = f.
By assumption, we can deform the curve to a point: if r
0
(t) is
the original curve and r
1
(t) is the curve r
1
(t) = P which stays at
one point, dene a parametrized surface S by r(t, s) = r
s
(t). By
assumption, curl(F) =

0 and therefore the ux of curl(F) through


S is zero. By Stokes theorem, the line integral along the boundary
C of the surface S is zero too.
THE NASH PROBLEM. Nash challenged his multivariable class in the movie A beautiful mind with a
problem, where the region is not simply connected.
Find a region X of R
3
with the property that if V is the set of
vector elds F on R
3
\ X which satisfy curl(F) = 0 and W is the
set of vector elds F which are conservative: F = f. Then, the
space V/W should be 8 dimensional.
You solve this problem as an inclass exercise (ICE). The problem
is to nd a region D in space, in which one can nd 8 dierent
closed paths C
i
so that for every choice of constants (c
1
, ..., c
8
),
one can nd a vector eld F which has zero curl in D and for
which one has

C1
F dr = c
1
, ...,

C8
F dr = c
8
.
One of the many solutions is cut out 8 tori from space. For each
torus, there is a vector eld F
i
(a vortex ring), which has its
vorticity located inside the ring and such that the line integral of
a path which winds once around the ring is 1. The vector eld
F = c
1
F
1
+... +c
8
F
8
has the required properties.
CLOSED SURFACES. Surfaces with no boundaries are called closed surfaces. For example, the surface of a
donought, or the surface of a sphere are closed surfaces. A half sphere is not closed, its boundary is a circle.
Half a doughnut is not closed. Its boundary consists of two circles.
THE ONE MILLION DOLLAR QUESTION. One of the Millenium problems is to determine whether any three
dimensional space which is simply connected is deformable to a sphere. This is called the Poincare conjecture.
LINEINTEGRAL IN HIGHER DIMENSIONS. Line integrals are dened in the same way in higher dimensions.

C
F dr, where is the dot product in d dimensions and dr = r

(t)dt.
CURL IN HIGHER DIMENSIONS. In d-dimensions, the curl is the eld curl(F)
ij
=
xj
F
i

xi
F
j
with

d
2

components. In 4 dimensions, it has 6 components. In 2 dimensions it has 1 component, in 3 dimensions, it


has 3 components.
SURFACE INTEGRAL IN HIGHER DIMENSIONS. In d dimensions, a surface element in the ij-plane is
written as dS
ij
. The ux integral of the curl of F through S is dened as

curl(F) dS, where the dot
product is

i<j
curl(F)
ij
dS
ij
. If S is given by a map X from a domain R in the 2-plane to R
d
, U =
u
X and
V =
v
X are tangent vectors to that plane and dS
ij
(u, v) = (U
i
V
j
U
j
V
i
) dudv.
STOKES THEOREM IN HIGHER DIMENSIONS. If S is a two dimensional surface in d-dimensional space
and C is its boundary, then

S
curl(F) dS =

C
F ds.

You might also like