# Gauss' Law and charges

1. Oct 5, 2011

### Harrisonized

1. The problem statement, all variables and given/known data

Gauss' Law states:

∫∫ E.dS = ∫∫∫ div(E) dV = Qenc/ε₀

The proof is as follows (this is from Marsden's Vector Calculus 5e):

Let M be a elementary region in ℝ3. Then if (0,0,0) ∉ ∂M, we have:
∫∫∂M r.n/r3 dS
= 4π if (0,0,0) ∈ M
= 0 if (0,0,0) ∉ M

Construct a sphere of radius δ with boundary ∂N such that the sphere is contained completely within M. Let W be the region between ∂N and ∂M. W has a divergence of 0.

Note that ∂N has a normal vector pointing inward toward the deleted point (0,0,0) and ∂M has a normal vector pointing outward away from the deleted point (0,0,0).

Since the total flux out of W must be 0 based on the divergence theorem, the flux out of ∂N and the flux out of ∂M must be equal. The flux out of ∂N is 4π, since

-∫∫∂N r.n/r3 dS = ∫∫∂N δ24 dS = 1/δ2 ∫∫∂N dS

∫∫∂N dS = 4πδ2, the surface area of a sphere. Take the limit as the sphere approaches 0.

Therefore, the flux out of ∂M containing the deleted point (0,0,0) is 4π.

Let E = Q/(4πε0) r/r3. Then

∫∫ E.dS = Q/(4πε0) ∫∫∂M r.n/r3 dS = Q/ε0

My questions are as follows:

1. Let M be a convex region of any finite size with a charge distribution. From the exterior of M, does the force follow Coulomb's law as though M were a point charge acting from the centroid? (I'm inclined to believe yes, because of the proof of Gauss' Theorem, but I can't be certain.)

2. Suppose I have a sphere with a conical "bite" on its side. Like this: (< The centroid is thus to the left of the sharp tip of the "cone". Suppose I put a test charge somewhere inside the cone, such that it's not touching the surface of the object. What force does it experience? (I'm inclined to believe that F = q0Q/(4πε0r2) toward the centroid.)

3. Suppose I have an off-center cavity near the surface of the sphere, such that it doesn't pass through the centroid of the shape. What force does a test charge placed in this cavity experience?

4. In my physics class, I'm told that Gauss' Law may only be used when there is a high degree of symmetry. However, the proof I mentioned above allows me to use Gauss' law for ANY shape, given that its centroid is contained within its boundary. Why is Gauss' Law used only for symmetrical regions? What allows me to use Gauss' Law for shapes that don't have centroids, such as infinite planes or infinite lines?

5. Can anyone tell me what's so special about a solid angle? I really don't understand proofs of Gauss' law that just mention the solid angle, such as this one:

http://faculty.cua.edu/sober/536/Gauss_solid_angle.pdf [Broken]

I see in too many physics textbooks and it's driving me nuts because it just seems so... unsatisfactory.

Last edited by a moderator: May 5, 2017