ED_HW1
ED_HW1
We can solve this partial differential equation, the solution will be:
and the initial condition in real space is delta-function, , this lead the
, and:
The Fourier transfer in both space and time of can be written as:
because the . Then, take the inverse transfer of above expression, we can get:
This is a standard Fourier transform of the function , where , so the whole integral will be:
where is the Heaviside step function. Comparing both expressions of above and the result of part (a),
we observe that they match for , and for . Therefore, the diffusion Green's function
satisfies:
Because of the spherical symmetry of Green's function, which is only depends on the distance between two
points, we switch the system to spherical coordinates in - space, and represent as .
With the Euler's formula, , substitute this into the integral and simplify it:
So, we have:
For ,
so,
but the second term in integral of will be double, and finally, we have:
(a) Show that the momentum of these electromagnetic fields, (6.117), can be
transformed to
because we have the current , so we want represent above formula with in it, and we have the
vector identity:
Then,
The first term on the right-hand side will be zero because the field will be vanished at infinity(with the
Stokes theorem), and we have:
The first term will be zero as the current is localized and no net flow of charge into or out the distribution(
), so:
Again, because of the localization of current, we can infer that the is uniform over the localized current
distribution, and then , this indicates:
(c) Suppose the current distribution is placed instead in a uniform electric field
(filling all space). Show that, no matter how complicated is the localized ,
the result in part a is augmented by a surface integral contribution from infinity
equal to minus one-third of the result of part (b),yielding
(don’t use any expressions from Jackson other than green’s functions
expansions).
In part (a), we have assumed that filed will vanish at infinity, but we can take this assumption no longer
when the electric field is uniform everywhere and fill all space. That is to say, we should modified our result
in part (a):
for the second term, we have already known that is equal to the result in part (b), and now we deal with the
first term. It goes to be:
Substituting it into :
1. Find the static electric field and steady magnetic field when .
For the static electric field, we can consider the Gaussian theorem and divide the space into inside shell and
outside shell.
For the magnetic field, we can determine it by Ampère's Law(the slow spin-up rotated shell and its charge
can be regarded as steady current). The surface current is:
It equal to a solenoid with surface current , so if we take a circuit across the shell, we will find that:
Summarily,
and
2. Find everywhere during the spin-up and use it to find the
timedependent part of the electric field.
Here we have the line velocity of the charge , gives the current during the spin-up:
As a summary,
Solving the differential equation and applying the condition assumption of , we can get:
As a summary,
Those results are consistent with the Poynting's theorem, and the Poynting vector that represent surface
term of power is zero, means there is no surface term(which is normal at all the space integral).
The interward energy flow can be calculate with Poynting's theorem using a cylindrical volume with a radius
slightly smaller than the shell: