Scribd
Upload
20 views

Classical Electrodynamics: Lecture Note 9

This document discusses electromagnetic potentials, including scalar and vector potentials. It introduces the electric scalar potential Φ and shows that the electric field E can be written as the negative gradient of Φ. It also introduces the magnetic vector potential A and shows that the magnetic field B can be written as the curl of A. Finally, it notes that while the vector potential A is useful, a magnetic scalar potential is of limited interest due to inconsistencies with Ampere's law.

Uploaded by

140557
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PDF, TXT or read online on Scribd
20 views

Classical Electrodynamics: Lecture Note 9

This document discusses electromagnetic potentials, including scalar and vector potentials. It introduces the electric scalar potential Φ and shows that the electric field E can be written as the negative gradient of Φ. It also introduces the magnetic vector potential A and shows that the magnetic field B can be written as the curl of A. Finally, it notes that while the vector potential A is useful, a magnetic scalar potential is of limited interest due to inconsistencies with Ampere's law.

Uploaded by

140557
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 7

Classical Electrodynamics

Lecture Note 9

Jingbo ZHANG [email protected]

Chapter 3 Electromagnetic Potentials


3.1 Scalar and Vector Potentials

The Maxwell equations read E =


0

(3.1) B t E t (3.2) (3.3)


0

E = B = 0

B = 0 j + 0

(3.4)

3.1.1

Electric Potential

In electrostatics, the electric field E is not just any old vector function. It is a very special kind of vector function. E = 0. Its curl is always zero, i.e. (3.5)

In this section, we are going to reduce a vector problem down to a much simpliar scalar problem. 1

Theorem: scalar,

Any vector whose Curl is Zero is equal to the Gradient of some

f = 0.

(3.6)

So, we can define a function (x), named electric scalar potential by E(x) (x). We can also introduce this potential function in another way. field in static case, the integral around a closed path is Zero, E dl = 0.
P

(3.7) For electric

(3.8)

Because the line integral of E around any closed loop is zero, the integral from A to B is independent of Path. of path. Thats mean
B A

Edl is independent

So, we can define a scalar function


P

(P )
O

E dl,

(3.9)

here, O is some standard reference point, then depends only on the point P. Evidently, the potential difference between two points A and B is
B A

(B ) (A) =
O B

E dl +
O O

E dl E dl
A

=
O B

E dl E dl
A

Now, the fundemental theorem for gradient states that


B

(B ) (A) =
A

dl

So
B B

dl =
A A

E dl

Since, fanally, this is true for any points A and B, the integrands must be equal: E = This is the differential version. gradient of a scalar potential. equation (3.10) distract you. just a matter of tradition. (3.10)

It says that the electric field is the By the way, dont let the minus sign in

It carries over from (3.9) and is really

3.1.2

Poissons Equation and Laplaces Equation

We found that the electric field can be written as the gradient of a scalar potential (3.10). The question arisses: How do the fundamential equations

for E read in terms of ?

From the equation (3.10) in steady case, apart

from that persistent minus sign, the divergence of E is just the Laplacian of . Gausss law then says 2 = .
0

(3.11)

This is known as Poissons equation. (x) = 1 4 0

Its solution is called Comlombs potential: (3.12)

(x ) 3 d x. |x x |

In regions where is no charge, so that = 0, Poissons equation reduces to Laplaces equation. 2 = 0. 3 (3.13)

So much for Gausss law. What about the Curl law? Zero. This says that E = must equal

But thats no condition on , ---Curl of Gradient is always Zero.

Ofcause, we used the Curl law to show that E could be expressed as the gradient of a scalar, so its not really very surprising that this works out.

3.1.3

Magnetic Vector Potential

Just as E = 0 permitted us to introduce a scalar potential in electrostatics, E = . So, B = 0 invites the introduction of a vector potential

A in magnetostatics, B(x) = A(x). The former is authorized by Theorem 1, the latter by Theorem 2: of a curl is always Zero. care of B = 0. There remains Amperes law, B = ( A) = ( A) 2 A = 0 j . Now, the electric potential had a built-in ambiguity: (3.15) You can add to (3.14) the divergence

So, the potential formulation automatically takes

any function whose gradient is Zero (which is to say, any constant) without altering the physical quantity E. Likewise, you can add to the magnetic

potential any function whose Curl vanishes (which is to say, the gradient of any scalar), with no effect on B. the divergence of A: A = 0. (3.16) We can exploit this freedom to eliminte

For suppose our original potential, A, is not divergence-less. to it the gradient of , so that A = A + . The new divergence is A = A + 2 .

If we add

(3.17)

We can accommodate, provided a function can be found that statifies 2 = A. But this is mathematically identical to Poissons equation (3.11), with A in place of / equation.
0

as the "source".

And we know how to solve Poissons

Thats what electrostatics is all about (given the charge distribution, The essential point remains: It is always possible

find the potential).

to make the vector potential divergence-less. To put it the other way arround: the diffinition B = A specifies We are

the curl of A, but it doesnt say anything about the divergence.

at liberty to pick that as we see fit, and zero is evidently the simplest choise. With this condition on A, Amperes law becomes 2 A = 0 j . (3.18)

This again is nothing but Poissons equation, or rather, it is three Poissons equations (one for each Cartesian component). This time, 0 j is the source.

Assuming j goes to zero at infinity, we can read off the solution: A(x) = 0 4 j (x ) 3 d x. |x x | 5 (3.19)

3.1.4

Magnetic Scalar Potential


For one thing, its still

It must be said that A is not as useful as . a vector.

Although (3.19) is somewhat easier to work with than the Biot-Savat

law, you still have to fuss with components. It would be nice if we could get away with a scalar potential, B = m . (3.20)

But this is in consistent with Amperes law, since the curl of a gradient is always zero. A magnetostatics scalar potential m can be used, if you stick scrupulously to simply-connected, current-free regions, but as a theoretical tool it is of limited interest. Moreover, since magnetic forces do not work, A does not admit a simple physical interpretation in terms of the work done per unit charge. Nevertheless, the vector potential has substanitial importance, as you will see in next section.

You might also like