1

Why equal signs weren't used for limits(r < a, a < r < b, r > b instead of r <= a, a <= r <= b, r >= b)?

Two concentric spherical surfaces of radius a and b = 2a are in vacuum and they are evenly charged with charge quantities Qa = Q and Qb = 2Q. The reference point of zero potential is at infinity. Gauss's law was used to determine field in each area

Electric field:

  1. (r < a) E = 0
  2. (a < r < b) E = Q / 4πε0 * r * r
  3. (r > b) E = 3Q / 4πε0 * r * r

Potential:

  1. (r > b) Ф = 3Q / 4πε0r
  2. (a < r < b) Ф = Q * (a / r + 1) / 4πε0a
  3. (r < a) Ф = Q / 2πε0a enter image description here

enter image description here

Another example: The conducting sphere of radius a = 10 cm is surrounded by a concentric layer of dielectric of relative dielectric constant εr = 2 of radius b = 20 cm and is in the air. The sphere is loaded with charge Q = 10 nC

Electric field:

  1. (r < a) E = 0
  2. (a <= r < b) E = Q / 4πε0εr * r * r
  3. (r > b) E = Q / 4πε0 * r * r

Potential:

  1. (r <= a) Ф = (Q * ((b - a) / εr * a + 1))) / 4πε0b
  2. (a <= r <= b) Ф = (Q * ((b - r) / εr * r + 1))) / 4πε0b
  3. (r >= b) Ф = Q / 4πε0r

enter image description here

enter image description here

Third example: The conductive sphere of radius a, loaded with the amount of electricity Q, is located inside a concentric metal spherical shell of radius b and thickness c - b. The system is in a vacuum.

Electric field

  1. (r < a) E = 0
  2. (a <= r <= b) E = Q / 4πε0 * r * r
  3. (b < r < c) E = 0
  4. (r >= c) E = Q / 4πε0 * r * r

Potential:

  1. (r >= c) Ф = Q / 4πε0r
  2. (a <= r <= c) Ф = (Q * ((b - a) / rb + 1 / c)) / 4πε0r

enter image description here

enter image description here

Pavle
  • 23
  • 4
  • You are partially right. Currently, E at a and b are not clearly defined. Point#2 could have been modified as a <= r <= b. However, if point#1 was also modified as r <= a, then E at a would have been defined twice, and that too, inconsistently (0 versus Q/(4pi e a^2)). – AJN Aug 04 '20 at 07:24
  • However, it so happens that the finite valued electric field at the infinitesimally thin thickness defined as r=a (as well as r=b) doesn't change the final answer for this particular question. – AJN Aug 04 '20 at 07:29
  • Your figures do not show the variables b and c. There is a mismatch between the figures and the numbered lists above it. – AJN Aug 04 '20 at 07:32
  • I edited them now – Pavle Aug 04 '20 at 07:48
  • Below functions it says that the function of electric field is discontinuous at points where charges are located and that it's wrong to use r <= a, a <= r <= b and r >= b and that function of potential is continuous but it's not differentiable at points where charges are located – Pavle Aug 04 '20 at 07:52
  • How can be determined if function is differentiable at points where charges are located? – Pavle Aug 04 '20 at 08:05
  • I think that your question has some missing info, particularly the charge distribution. "electric field is discontinuous at points where charges are located" implies that there are locations having infinite charge density (C/m^3). More info is required to give a clear answer. – AJN Aug 04 '20 at 08:12
  • I updated info now. Value of Q is not given – Pavle Aug 04 '20 at 08:25
  • I also added another example which has defined value for Q – Pavle Aug 04 '20 at 08:56

1 Answers1

0

Electric field

Lets consider a simpler example. A spherical surface of radius \$a\$ with uniformly distributed charge \$Q\$ on it. Let the sphere be centred at the origin for convenience. Let the space be vacuum. The charge density is

\$\frac{Q}{4\pi\ a^2} \mathrm{\frac{coulomb}{m^2}}\$.

Electric field E is given by

\$ E = \begin{cases} 0, & r < a & \text{inside}\\ \frac{Q}{4\pi\varepsilon_0\ r^2} & r > a & \text{outside}\\ ? & r=a & \text{on the sphere} \end{cases} \$

The question as I have understood is: "Where to place the equals symbol ?".

Clearly, the case with equals symbol on both cases

1

\$ E = \begin{cases} 0, & r \leq a \\ \frac{Q}{4\pi\varepsilon_0\ r^2} & r \geq a \end{cases} \$

is wrong since E is defined as 0 as well as \$\frac{Q}{4\pi\varepsilon_0\ a^2} \neq 0\$ for the case \$r = a\$. Electric field can't have two different values at the same location at the same time (in classical mechanics?). So this option is ruled out. In other words, for discontinuous functions like the one shown here for electric fields, you cannot put the equals sign in both places at the same time since the values calculated from both equations are different and a field cannot have two values at same time at same location.

2

Another option is not to put the equals sign in either case. This is not wrong, but E is left undefined at \$r=a\$. In certain scenarios, this is okay. For example, to find the potential V, we need to integrate E from \$r = \infty\$ to \$r=\mathrm{some value}\$. Even if the value of E is undefined at \$r=a\$, we can still perform the integration if the value of E at \$r=a\$ is finite as \$\int_{a^+}^{a^-}E(r) dr = 0\$ if \$E(a)<\infty\$.

Two other options are either

3

\$ E = \begin{cases} 0, & r \leq a \\ \frac{Q}{4\pi\varepsilon_0\ r^2} & r > a \end{cases} \$

or

4

\$ E = \begin{cases} 0, & r < a \\ \frac{Q}{4\pi\varepsilon_0\ r^2} & r \geq a \end{cases} \$

To see if #3 or #4 is more appropriate, consider the case \$r=a\$. To see if the electric field is zero consider placing a test charge at \$r=a\$ and find the force acting on it. From the below figure, we can see that the forces from all parts of the figure are pointing right-to-left and there is no left-to-right force to cancel them. So the Electric field is not zero. That rules out option #3 also. We have not established if the field is \$\frac{Q}{4\pi\varepsilon_0\ r^2}\$. If we can do that, then option #4 is the correct way to place the equals sign. You can also note that this is followed in the second example you have supplied (for \$a\$, but for \$b\$ there is no equals sign at all).

sphere of charges

Potential

For V there is not such confusion since the plot of V is a continuous line and hence even if we put equals sign on all inequalities, the result from the different equations are the same for this particular example. So, for potential V, #1 (equals sign in both places) is also correct.

AJN
  • 3,831
  • 1
  • 10
  • 20
  • Why is a <= r < b instead of a <= r <= b and r > b instead of r >= b in the second example that I posted? – Pavle Aug 06 '20 at 08:19
  • Because Electric field can't have two different values at the same time right? – Pavle Aug 06 '20 at 08:38
  • I am not very sure about the r=b missing case in the second example. – AJN Aug 06 '20 at 08:43
  • @Pavle I would also suggest posting or migrating this question to physics.SE if the answer is not satisfactory. Remember to include all the details in the question if you choose to do so. IMO, the value at exactly r=a or r=b wont affect further calculations in practical cases as an infinitesimaly thin sphere of charge doesn't exist in practice. – AJN Aug 06 '20 at 08:54
  • Can you please also check third example that I posted? In this example metal shell is used – Pavle Aug 06 '20 at 08:57
  • Why a <= r <= b instead of a <= r < b and r >= c instead of r > c? – Pavle Aug 06 '20 at 09:03
  • In the third case, using the same argument as in the figure, I would imagine that the correct way to write would be r<a, a<= r <b, b <= r <c, c <= r. Again I'm not sure about it though. What justification does the text book give for this ? My guess is that the answer would depend on how they evaluate the integral used to find E $\int_\theta \int_\phi q(a,\theta,\phi)/(\dots) d\theta d\phi$. – AJN Aug 06 '20 at 09:12
  • They calculate electric field like this: ∮EdS=ΣQ -> E∮dS = E * S -> E * S = Q -> E * 4πε0 * r * r = Q -> E = Q / 4πε0 * r * r – Pavle Aug 06 '20 at 09:26
  • This is Gauss's law. The imaginary closed surface (S) used for the integral divides the charges into two parts: inside and outside the surface. For the case r=a, the surface and the charge are perfectly overlapping. Is Gauss's law applicable for that case ? I do not think so. You have to use the regular Coulomb's law for such cases. – AJN Aug 06 '20 at 09:37
  • Is r <= b in a <=r <= b because there is value in inner side of b surface(b-) E(r=b-)? – Pavle Aug 06 '20 at 09:40