Gauss Law for Magnetic fields?

You probably know Gauss Law very well. It states that if you take any closed surface, the electric flux through this surface is proportional to the total charge enclosed. Mathematically: Φ E : = E d A = Q e n c ϵ 0 . \Phi_{E}:=\oint \vec{E} \cdot d\vec{A}=\frac{Q_{enc}}{\epsilon_{0}}. What about the magnetic flux? It turns out that Φ M : = B d A = 0 ( always! ) . \Phi_{M}:=\oint \vec{B} \cdot d\vec{A}=0 \quad (\textrm{always!}). This is a Law of Nature, equivalent to one of Maxwell's equations and it reflects the experimental fact that there are no magnetic charges . In particular, Φ M = 0 \Phi_{M}=0 implies that not every magnetic field configuration can be realized in nature. For example, one can show that it is impossible to have a magnetic field that increases along the z-axis having only a z-component.

Consider an axially symmetric field with z-component (the field is symmetric about the z-axis) given by B z = B 0 + b z B_{z}=B_{0}+ b z where B 0 = 2 μ T B_{0}= 2~ \mu \mbox{T} and b = 1 μ T/m b=1 ~\mu \mbox{T/m} .

Show that in addition to the z-component, this field must have a radial component B r B_{r} . Find B r |B_{r}| in Teslas at a point located 50 cm 50~\textrm{cm} away from the z-axis.


The answer is 2.5E-7.

This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try refreshing the page, (b) enabling javascript if it is disabled on your browser and, finally, (c) loading the non-javascript version of this page . We're sorry about the hassle.

5 solutions

Discussions for this problem are now closed

Josh Silverman Staff
Oct 7, 2013

When facing a physics problem, the first rule is do no harm.

Therefore, since we're told this field has a radial component, let us not pick a tetrahedron or a cube as a test surface to evaluate the flux integral. Instead, let us go with the cylindrically symmetric cylinder.

On the first accusation, we can easily see that a z z -component only field would violate the no-flux condition.

Say for sake of argument that the test cylinder is 2 units high (from z = 1 z=-1 to z = 1 z=1 ), and let's denote the normal vectors to the surfaces A ^ \hat{A} . Then the flux at the top is equal to

A ^ B = π R 2 ( B 0 + b ) \hat{A}\cdot\vec{B} = \pi R^2\left(B_0 + b\right)

and at the bottom is given by

A ^ B = π R 2 ( B 0 + b ) \hat{A}\cdot\vec{B} = \pi R^2\left(-B_0 + b\right)

for an overall flux of 2 π R b 2\pi R b piercing the wall, which is not zero.

The only surface on the cylinder that can save us with a negative compensating flux is the wall. Hence, the radial field is required.

Let us say there is a radial field and moreover that it points inwards (an outwards facing field will make more positive flux which is just further from our goal of zero flux).

Moreover let us assume the cylinder to have arbitrary height, z z . Our results cannot depend on the geometry of our arbitrarily chosen test surface and therefore had best not depend on the arbitrary height of the arbitrary shape of our arbitrary test surface.

The flux through the top and bottom is given by 2 π R b 2\pi R b as we showed above. Through the wall we have a field B r B_r penetrating the wall which has a total surface area 2 π R 2 z 2\pi R\cdot 2z making a flux contribution of 4 π R z B r -4\pi R z B_r

We want these two fluxes to sum to zero, therefore:

4 π R z B r = 2 π R b \displaystyle 4\pi R z B_r = 2\pi R b

and we have

B r = R b 2 = 2.5 E 7 T \boxed{\displaystyle B_r = \frac{Rb}{2} = 2.5\mbox{E}-7\mbox{ T}}

Nhat Le
May 20, 2014

Consider a cylindrical surface parallel to the z-axis, with radius 50cm. The two bases of the cylinder are at z=0 and z=t, respectively.

Total magnetic flux through the bases = Area x B = (pi r^2)(B0 + bt - B0) = (pi r^2)bt

Thus there must be an inward magnetic flux through the side of the cylinder, so that the total magnetic flux becomes 0

Let Br be the magnitude of the radial component. Then (Br)(2pi r) = (pi r^2)bt Br = rb/2 = 2.5E-7 T

Mark Hennings
Oct 9, 2013

For magnetostatic systems, the divergence . B \nabla . \mathbf{B} of B \mathbf{B} must be 0 0 . In cylindrical coordinates, so that B = ( B r , B θ , B z ) \mathbf{B} \,=\, (B_r,B_\theta,B_z) , this means that 0 = 1 r r ( r B r ) + 1 r B θ θ + B z z 0 \; = \; \frac{1}{r}\frac{\partial}{\partial r}(rB_r) + \frac{1}{r}\frac{\partial B_\theta}{\partial \theta} + \frac{\partial B_z}{\partial z} In our case, this means that 1 r r ( r B r ) + b = 0 \frac{1}{r}\frac{\partial}{\partial r}(rB_r) + b \; = \; 0 and hence r B r = α 1 2 b r 2 rB_r \,=\, \alpha - \tfrac12br^2 , and hence B r = α r 1 1 2 b r B_r \,=\, \alpha r^{-1} - \tfrac12br . Since we want B r B_r to be continuous, we must have α = 0 \alpha=0 , and hence B r = 1 2 b r B_r = -\tfrac12br .

When r = 0.5 r=0.5 , we obtain B r = 2.5 × 1 0 7 B_r = -2.5 \times 10^{-7} T.

Correction: . B = 0 \nabla . \mathbf{B} = 0 for all systems, not just magnetostatic ones.

Mark Hennings - 7 years, 8 months ago

I love formularies.

Alexander Bourzutschky - 7 years, 7 months ago
Yuchen Liu
Oct 7, 2013

Imagine a cylinder with radius R R and length 2 z 2z (along the z z axis), let the centre of mass of the cylinder be the origin, therefore

The magnetic flux through the top of the cylinder is π R 2 ( B 0 + b z ) \pi R^2 (B_0+bz)

The magnetic flux through the bottom of the cylinder is π R 2 ( B 0 + b z ) \pi R^2 (-B_0+bz) , because the direction of the field B 0 B_0 is towards the opposite direction at the bottom

Adding them up, magnetic flux along z z axis is π R 2 2 b z \pi R^2 *2bz

Since Φ M = 0 \Phi _ M = 0 ,

The radial component must be able to compensate for the flux along z z axis

The flux in radial direction (i.e through the walls of the cylinder) is the area of the rectangle times the strength of the magnetic field, 2 π R 2 z B r 2\pi R * 2z * B_r where B r B_r denotes the magnetic field along the radial direction

Equating the two fluxes,

π R 2 2 b z = 2 π R 2 z B r \pi R^2 *2bz = 2\pi R * 2z * B_r

B r = R b 2 = 0.5 m × 1 0 6 T / m 2 = 2.5 1 0 7 T B_r = \frac{Rb}{2} = \frac{0.5m \times 10^{-6}T/m}{2} = \boxed{2.5*10^{-7} T}

I think the more appropriate dimension used should be the area of the surface of the cylinder instead of the area of the rectangle.

Denver Wong - 7 years, 8 months ago

Haha sure but eventually it's the same idea.

Yuchen Liu - 7 years, 8 months ago
Jiahai Feng
Oct 7, 2013

Consider a cylinder with radius r r , height d z dz and centered at the z-axis. Let the base of the cylinder be at height z z From Gauss' Law, total magnetic flux entering and exiting this cylinder is 0. Hence, Φ s i d e + Φ t o p + Φ b o t t o m = 0 \Phi_{side} + \Phi_{top} + \Phi_{bottom}= 0

Φ t o p = π r 2 ( B 0 + b ( z + d z ) ) \Phi_{top} = -\pi r^2 \cdot (B_0 + b(z+dz)) and Φ b o t t o m = π r 2 ( B 0 + b z ) \Phi_{bottom} = \pi r^2 \cdot (B_0 + bz)

By symmetry, Φ s i d e = B r 2 π r \Phi_{side} = B_r \cdot 2 \pi r

Therefore, π r 2 ( B 0 + b ( z + d z ) ) + π r 2 ( B 0 + b z ) + B r 2 π r = 0 -\pi r^2 \cdot (B_0 + b(z+dz)) + \pi r^2 \cdot (B_0 + bz) + B_r \cdot 2 \pi r = 0

B r 2 π r = d z π r 2 B r = b r 2 = 1 × 1 0 6 0.5 2 = 2.5 × 1 0 7 B_r \cdot 2 \pi r = dz \cdot \pi r^2 \implies B_r = \frac{b \cdot r}{2} = \frac{1 \times 10^{-6} \cdot 0.5}{2} = 2.5 \times 10^{-7}

Correction: Φ s i d e = B r 2 π r d z \Phi_{side} = B_r \cdot 2 \pi r \cdot dz , but my final answer should still be the same.

Jiahai Feng - 7 years, 8 months ago

This is Faraday's Law (magnetic fields), not Gauss's (electric fields).

Mark Hennings - 7 years, 8 months ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...