Electric field of an infinite plane and a conductor

1. Jun 19, 2008

Niles

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

I have two questions.

1) The electric field of a uniformly charged, infinite plane is given by:

$$E = \frac{\sigma }{{2\varepsilon _0 }}$$

The electric field just outside a conductor is given by:

$$E = \frac{\sigma }{{\varepsilon _0 }}$$

Why are these expressions not the same? Isn't a plane with a surface charge a conductor?

2)
The electric field just outside a conductor is given by:

$$E = \frac{\sigma }{{\varepsilon _0 }}$$

Lets say I have two conducting plates situated a distance d from eachother. Is the electric field between the two conducting planes given by

$$E = \frac{\sigma }{{\varepsilon _0 }}$$

OR

$$E = \frac{2\sigma }{{\varepsilon _0 }}$$

I can't find an answer for these questions in my book nor on the WWW. Help is very much appreciated!

2. Jun 19, 2008

tmc

The answer to your first question may be near the middle / bottom of this page:
http://farside.ph.utexas.edu/teaching/em/lectures/node58.html; [Broken] specifically equations (623) onwards.

This might allow you to answer your second question.

Last edited by a moderator: May 3, 2017
3. Jun 19, 2008

rohanprabhu

no. a charged plane referred to earlier has a uniform charge on one of it's surface only. The other side of it is uncharged. However, it it is a conductor, the other side HAS TO have a negative charge of the equal magnitude. This is because the field inside a conductor is zero [that is how a conductor is defined].

The electric field between them is given by $$E = \frac{\sigma }{{\varepsilon _0 }}$$. This follows from a simple application of Gauss Law.

Also, do note here that if two conducting plates are placed closed to each other, the surfaces facing each other will develop opposite and equal charges, no matter what charge they initially had. Same applies to the surfaces facing away.

4. Jun 19, 2008

Niles

This I cannot see how - don't the electric fields just add?

5. Jun 19, 2008

Dick

The electric field between two equally charged conductors (or planes) will be zero.

6. Jun 19, 2008

Niles

Say that I have two conducting planes separated with a vertical distance d. Plate 1 at z=0 has a potential 0, and plate 2 at z=d has a potential V maintained by a battery. How come there is an electric field E = V/z there then and doesn't this equal E = sigma/epsilon_0?

Last edited: Jun 19, 2008
7. Jun 19, 2008

Dick

In that case the charge on the two plates isn't equal. The battery pumps charge from one plate to the other creating the charge difference which in turn creates the field.

8. Jun 19, 2008

Niles

And this electric field does not equal $$E = \frac{\sigma }{{\varepsilon _0 }}$$?

9. Jun 19, 2008

Dick

It's the difference of two expressions like that with two different charge densities. As you said, it's also V/d.

Last edited: Jun 19, 2008
10. Jun 19, 2008

Niles

If $$E = \frac{\sigma }{{\varepsilon _0 }}$$ is the difference between the electric field of the upper and lower plate, then must it not equal the total field?

11. Jun 19, 2008

Dick

In between the two plates the field from the upper plate and the field from the lower plate point in opposite directions. If the charges are equal, they cancel, if they aren't, then it's the difference that determines the net field.

12. Jun 19, 2008

Niles

Hmm, I think there is a contradiction in what you are telling me and what my book is saying. According to my book, the force that the two plates are attracting eachother with is given by

F = sigma^2 / (2*e_0) - which is OK

and they say that E = sigma / e_0 => sigma = V*e_0 / z. So they say that the electric field is $$E = \frac{\sigma }{{\varepsilon _0 }}$$

13. Jun 19, 2008

Staff: Mentor

OK.

I suspect you are thinking of two oppositely charged plates close together (like a parallel plate capacitor). The field between them is given by the first expression.

Note that you can consider the field between the plates as the sum of the fields created by each sheet of charge:

$$E = \frac{\sigma }{2\epsilon _0} + \frac{\sigma }{2\epsilon _0} = \frac{\sigma }{\epsilon _0}$$

14. Jun 20, 2008

Niles

Thanks!

15. Jun 20, 2008

Niles

Wait, I haven't understood this thoroughly.

We have two conducting plates. I am not told anything about the charges on the plate, only that the lower plate is grounded and the upper plate has a potential V. So the electric field goes from the upper plate to the lower, and this is given by

E = V/z, where z is the distance between the plates.

For this setup I have the following questions:

1) Can I say anything about the amount of charge on each plate from these informations?

2) If I want to find the force that each plate exerts on eachother, according to my book (Griffiths' page 103, eq. 2.51) I should use the expression

$$F = PA = A\frac{{\varepsilon _0 }}{2}E^2$$
where A is the area of the plate.

How can I justify using this expression, when we are dealing with planes? When we are told they are conducting planes, what expression for the electric field am I supposed to use?

16. Jun 20, 2008

Staff: Mentor

Sure. If you have the potential and the distance between the plates, then you have the field between the plates. If you have the field, then you have the charge density.

Realize that in this expression for force, E stands for the total field between the plates, not just the field from one plate. In deriving this expression, use the field from one plate to calculate the force on the other.

17. Jun 20, 2008

Niles

For 1 and 2) This is what I am confused about. In this setup, am I supposed to use the electric field for two conductors (so the total field is E = 2*sigma / e_0) or the field from two oppositely charged infinitely large planes E = sigma/e_0?

I am just told that it is two conducting planes with a potential difference. Nothing else.

18. Jun 20, 2008

Staff: Mentor

I think I see what's messing you up. Realize that a conducting plate with a total charge Q will have different surface charge densities in these two situations:

(a) The conducting plate is isolated from the world. The charge is evenly divided on both sides of the plate so the surface charge density on each side is Q/(2A). The field from that plate is Q/(2 A e_0).

Note that this is identical to the field from a sheet of charge Q. (Which makes sense, right?)

(b) The conducting plate is parallel to another oppositely charged conducting plate. In this case, the entire charge is on the inner surface, so the surface charge density is Q/A. The field between those plates is Q/(A e_0), which is twice what it would be if there were just one plate.

To answer your question, the field between the two conducting plates is sigma/e_0.

That's all the information you need.

19. Jun 20, 2008

Niles

I hope it is alright that I just comment on your post this way.

20. Jun 20, 2008

Staff: Mentor

It makes it a more difficult to respond to your comments.

This is the field due to both conducting plates. (The field from a single sheet of charge would only be half that.) Realize that when deriving the field from a conductor using Gauss's law, you are including the effects of both plates at once.

It is double! (Double the field that would exist if there were only one plate with the same charge.)

Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook