0% found this document useful (0 votes)
47 views5 pages

Electro PDF

This document discusses three methods for solving problems in magnetostatics involving boundaries: 1. Using Maxwell's equations and the vector potential to solve problems involving currents and boundaries. 2. Expressing the vector potential as a sum of spherical harmonics and applying boundary conditions to solve problems involving charges distributions. 3. Using a scalar potential and applying boundary conditions involving the normal and parallel components of the magnetic field to solve problems involving charge distributions and boundaries.

Uploaded by

Jhon
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
0% found this document useful (0 votes)
47 views5 pages

Electro PDF

This document discusses three methods for solving problems in magnetostatics involving boundaries: 1. Using Maxwell's equations and the vector potential to solve problems involving currents and boundaries. 2. Expressing the vector potential as a sum of spherical harmonics and applying boundary conditions to solve problems involving charges distributions. 3. Using a scalar potential and applying boundary conditions involving the normal and parallel components of the magnetic field to solve problems involving charge distributions and boundaries.

Uploaded by

Jhon
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/ 5

Class 6 in Electrodynamics

Based on course by Yuri Lyubarsky and Edited By Avry Shirakov


Department of Physics, Ben-Gurion University, Beer-Sheva 84105, Israel

This exercise pool is intended for an undergraduate course in “Electrodynamics 1”.

Magnetostatics with a boundary values

Maxwell eq. for a ”static” case are:


ρ
∇·E= Gauss’s law ; ∇ × E = µ0 J
0

∇ · B = 0 ; ∇ × B = µ0 J Ampere’s law

Since ∇ · B = 0 a vector potential can be defined in magtetostatics, A, such that:

B=∇×A

here we can add to the magnetic potential any function whose ∇ × f = 0 ,thus we can eliminate the div of A, ∇ · A = 0
and consequently we get the 3D Poissons eq.

∇2 A = −µ0 J

We can write (assuming J goes to zero at infinity)

J(r0 ) 0
Z
µ0
A(r) = dV so generally A k J
4π V 0 r
1
Generally the Green Function G(r, r0 ) = can be written (as you already know),
||r − r0 ||
∞  n
1 1 X r0
= Pn (cos(θ0 ))
||r − r0 || r n=0 r

∞ X
l l
1 X 4π r<
= Y ∗ (θ0 , φ0 )Ylm (θ, φ)
l+1 lm
||r − r0 || 2l + 1 r>
l=0 m=−l

while
 
R r>R r r>R
r< = and for r> =
r r<R R r<R

and general relation of the spherical harmonic functions


Z

Ylm (θ0 , φ0 )Yij (θ0 , φ0 ) sin(θ0 )dθ0 dφ0 = δli δmj

Another way to approach problems in magneto-statics: if J = 0 except the boundary, ∇ × B = 0 there exist a scalar
potential Φm so that B = −∇Φm. This is similar solution to electrostatics, with the boundary conditions:
⊥ ⊥
B> = B<

k k
B> − B< = µK while B k ⊥ K

Problem 0311
2

A thick slab extending from z = −a to z = a carries a uniform volume current J~ = J x̂. Find the magnetic field as a
function of z using vector potential. We start at the −a < z < a range, so the J = J0 x̂ and at this range we have

∇2 A = −µ0 J

by choosing A k J the eq. becomes

∇2 Ax = −µ0 J0

Since B = ∇ × A and we choose A = Ax , the magnetic field can be written as

0 · x̂ + ∂zAx ŷ − ∂yAx ẑ = B

here we will set the magnetic field to be in the y direction,as a result Ax is only z dependant (that the derivative
won’t be zero).

d2 d2 d2
A x (z) + A x (z) + Ax (z) = Jx
dx2 dy 2 dz 2

d2 −µ0 Jx z 2
A x (z) = Jx ; general solution → A x (z) = + C1 z + C2
dz 2 2
while outside z > a range the solution is Ax (z) = C3 z + C4 . Now we should find all the constants.
1. Ax is continues at z = a:

−µ0 Jx a2
+ C1 a + C2 = C3 a + C4
2
2. Continuousness of A0x at z = a:

−µ0 J0 a + C1 = C2

3. From the geometry of the problem we can see that B(z = 0) = 0, this implies that C1 = 0.

C3 = −µ0 J0 a

µ0 J0 a2
C4 − C4 =
2
we are left with two equations and 3 constants, we gauge Ax (z = a) = 0, this implies C2 = 0 and we are done. The
last constant is
µ0 J0 a2
C4 =
2
We can write the vector potential Ax :

µ0 J0 a2
−µ0 J0 az + a<z



 2
µ0 J0 z 2

Ax (z) = − −a < z < a

 2 2
 µ0 J0 az + µ0 J0 a


 z < −a
2
(here we applied same procedure for the z < −a range).
The magnetic field out of this vector potential is (B = ∂z Ax (z)ŷ)

−µ0 J0 a · ŷ a<z
B = −µ0 J0 z · ŷ −a < z < a
 µ J a · ŷ z < −a
0 0
3

Problem 0312

A spherical shell of radius R, carrying a uniform surface charge σ is set spinning at angular velocity ω. Find the
vector potential it produces at point r. We will solve this question in 3 different ways.
Stating from the Poisson’s eq. for vector potential

∇2 A = −µ0 J

we let the r lie on the z axis, so that ω is tilted at angle ψ, moreover we may make the orientation of x so that ω lies
in the xz plane. According to
K(r0 )
Z
µ0
A(r) = da0
4π a0 ||r − r0 ||
p
where K = σv, ||r − r0 || = R2 + r2 − 2Rr cos(θ0 ) and da0 = R2 sin(θ0 )dθ0 dφ0 . The velocity of a point (r0 ) is ω × r0
in this case
 
x̂ ŷ ẑ
v = ω × r0 = det  ω sin(ψ) 0 ω cos(ψ) 
R sin(θ ) cos(φ ) R sin(θ ) sin(φ ) R cos(θ0 )
0 0 0 0

= Rω [− (cos(ψ) sin(θ0 ) sin(φ0 )) x̂ − (sin(ψ) cos(θ0 ) − cos(ψ) sin(θ0 ) cos(φ0 )) ŷ + (sin(ψ) sin(θ0 ) sin(φ0 )) ẑ]

Notice that each term that has sin(φ0 ) or cos(φ0 ) equals to zero since:
Z 2π Z 2π
0 0
sin(φ )dφ = cos(φ0 )dφ0 = 0
0 0

Just one term survives, and we can write the integral as


!
π
µ0 R3 σω sin(ψ) cos(θ0 ) sin(θ0 )
Z
A(r) = − p dθ0 ŷ
2 0 R2 + r2 − 2Rr cos(θ0 )

defining u = cos(θ0 ) the integral becomes


Z 1 
u R2 + r2 + Rru p 2
√ du = − 2 r2
R + r2 − 2Rri|1−1
−1 R 2 + r 2 − 2Rru 3R

1  2
R + r2 + Rr |R − r| − R2 + r2 − Rr (R + r)
  
=− 2 2
3R r
If r lies inside the sphere R > r and the above expression reduces to (2r/3R2 ), if r lies outside the sphere R < r it
reduces to (2R/3r2 ).

− µ0 Rσω sin(ψ)r R > r

A(r) = 4
3
− µ0 R σω sin(ψ) R < r

3r2
Noting that ω × r = −ωr sin(ψ)ŷ, we rewrite A(r),

µ Rσ
 0
 (ω × r) R > r
A(r) = µ R 34
 0 σ (ω × r) R < r

3r3
The second way to solve this problem is by using Legandre polynomials. Everywhere, except r = R, J = 0. Therefore
∇ × B = µ0 J = 0, in this case, there is a scalar potential Φm so that B = −∇Φm and ∇B = ∇2 Φm = 0. Now we are
left with applying the appropriate boundary values:
1. Physical behaviour on the boundaries.
4

⊥ ⊥ k k
2. B> = B< and B> − B< = µK
In our case K = σv = σωR sin(θ)φ̂.
The general solution is of the form:
X Bn

Φ(r) = An rn + n+1 Pn (cos(θ))
n
r

Applying the first condition,


X Bn
Φ> (r) = Pn (cos(θ)); for outside the sphere r > R
n
rn+1

X
Φ< (r) = (An rn Pn (cos(θ)); for inside the sphere r < R
n

∂Φ
and the second one B⊥ =
∂r
∂Φ> ∂Φ<
|r=R − |r=R = 0
∂r ∂r

 
Bn n−1
−(n + 1) n+2 − nAn R Pn (cos(θ)) = 0
R
1 ∂Φ
Bk = Bθ =
r ∂θ
1 ∂Φ> 1 ∂Φ<
| +− | − = µ0 σωR sin(θ)
r ∂θ R r ∂θ R
noting that K ≈ cos(θ) therefore n = 1 and
B1
sin(θ) − A1 sin(θ) = µ0 σωR sin(θ)
R3

 
B1 2B1
−2 3 − A1 cos(θ) = 0 so A1 = − 3
R R

µ0 σωR4 2µ0 σωR


and B1 = and as a consequence A1 = − The scalar vector potential is:
3 3
4

 µ0 σωR cos(θ) r > R

3r 2
Φ(r) =
− 2µ0 σωRr cos(θ) r < R

3
∂Φm 1 ∂Φm
Now if we want to find the magnetic field B we should use B = −∇Φm which at this case is r̂ + θ̂ so, ..
∂r r ∂θ
4 

 µ0 σωR 2 cos(θ)r̂ + sin(θ)θ̂ r > R


3r 3
B(r) =
 2µ0 σωR cos(θ)r̂ − sin(θ)θ̂
 
r<R

3
The third way to solve this problem is through Spherical Harmonic polynomials. As mentioned at the beginning, we
can write the Green function as
∞ X
l l
1 X 4π r<
= Y ∗ (θ0 , φ0 )Ylm (θ, φ)
0
||r − r || 2l + 1 r> lm
l+1
l=0 m=−l
5

and then plug it into the eq.


K(r0 )ds0
Z
µ0
A(r) =
4π ||r − r0 ||

as at the previous solutions K = σv = σωR sin(θ)φ̂ but now we expand the φ̂ into x̂ and ŷcoordinates, K =
σωR sin(θ) [− sin(φ)x̂ + cos(φ)ŷ]
1
(ix̂ + ŷ)eiφ + (−ix̂ + ŷ)e−iφ

[− sin(φ)x̂ + cos(φ)ŷ] =
2
r
1 3
Now we should write K in the Ylm expansion, we remember that Y11 = − sin(θ)eiφ and Y1−1 =
r 2 2π
1 3
sin(θ)e−iφ
2 2π
" r ! r !#
σωR sin(θ)  iφ −iφ
 σωR sin(θ) 2 2π 2 2π
K= (ix̂ + ŷ)e + (−ix̂ + ŷ)e = (ix̂ + ŷ) − Y11 + (−ix̂ + ŷ) Y1−1
2 2 sin(θ) 3 sin(θ) 3

r

= σωR [−(ix̂ + ŷ)Y11 + (−ix̂ + ŷ)Y1−1 ]
3
applying this into vector potential integral
r Z XX l
2π µ0 4π r<
A= σωR Y ∗ (θ0 , φ0 )Ylm (θ, φ) [−(ix̂ + ŷ)Y11 (θ0 , φ0 ) + (−ix̂ + ŷ)Y1−1 (θ0 , φ0 )] R2 sin(θ0 )dθ0 dφ0
l+1 lm
3 4π m
2l + 1 r >
l

for r < R, A is
r
rl
Z
2π µ0 X X 4π
= σωR l+1
Ylm (θ, φ) [−(ix̂ + ŷ)Y11 (θ0 , φ0 ) + (−ix̂ + ŷ)Y1−1 (θ0 , φ0 )] Ylm

(θ0 , φ0 )R2 sin(θ0 )dθ0 dφ0
3 4π m
2l + 1 R
l

the result of the integral is (here we used Spherical Harmonics polynomials properties)
R2 [−(ix̂ + ŷ)δl1 δm1 + (−ix̂ + ŷ)δl1 δm−1 ]

r
2π µ0 4π r 2
A(r < R) = σωR R [−(ix̂ + ŷ)Y11 (θ, φ) + (−ix̂ + ŷ)Y1−1 (θ, φ)]
3 4π 3 R2
writing it back in the θ, φ terms
σωRrµ0 σωRrµ0
A(r < R) = sin(θ) [− sin(φ)x̂ + cos(θ)ŷ] = sin(θ)φ̂
3 3
l
r< R
and for the case r > R the only difference is in the l+1
term, which in this case (l = 1) is 2 , so
r> r

σωR4 µ0
A(r > R) = sin(θ)φ̂
3r2
to find B we should imply ∇ × A.
1 ∂ 1 ∂
∇A = Aφ r̂ − (rAφ )θ̂
r sin(θ) ∂θ r ∂r

4 

 µ0 σωR 2 cos(θ)r̂ + sin(θ)θ̂ r > R


B(r) = 3r3
2µ 0 σωR
 
cos(θ)r̂ − sin(θ)θ̂ r<R


3

You might also like