Applications of Gauss Law

The Gauss Law of electrostatics relates the net electric field flux through a complete
surface S of some volume V to the net electric charge inside that volume,
ZZ
def
E [S] = E dA = 4k Qnet [inside V ]. (1)
S

For a highly symmetric configuration of electric charges, the Gauss Law can be used to
obtain the electric field E without taking any hard integrals. Instead, one uses a Gaussian
surface whose symmetry alone assures that E is normal to the surface and has constant
magnitude along the surface. Consequently, the flux through such a surface is simply

E [S] = E A(S), (2)

hence by Gauss Law (1),

4k
E = Q[surrounded by S]. (3)
A(S)

In these notes, I shall illustrate this method for configurations of electric charges which have
spherical, cylindrical, or planar symmetry.

1
 Spherical Symmetry
The spherically symmetric charge configurations include a point charge, a uniform spher-
ical shell of charges, a uniformly charged solid ball, several concentric spherical shells, as well
as more exotic configurations where the 3D charge density depends on the radius but not
on the direction of the radius-vector. In spherical coordinates (r, , ),

dQ = (r) dVolume = (r) dr r 2 d (4)

where d = d sin d is the infinitesimal solid angle (in steradians) and (r, \, \) depends
only on the radius but not on the angular coordinates or .

By symmetry, the electric field E of spherically symmetric charges always points in the
radial direction towards the origin, or away from the origin while its magnitude depends
only on the radius,

E(r, , ) = E(r only)~r . (5)

To find the radial dependence of E(r) we use the Gauss Law. Let S be a sphere of some
radius r centered at the origin. By symmetry, the radial electric field is always perpendicular
to S, while its magnitude E(r) stays constant along S. This makes S a Gaussian surface, so
the flux through S is simply

E [S] = E(r) A(S) = E(r) 4r 2 . (6)

Hence, by Gauss Law,

E(r) 4r 2 = 4k Qnet [inside S] (7)

and therefore
 
k def k inside sphere
E(r) = 2 Q(r) = 2 Qnet . (8)
r r of radius r

2
Example: Point Charge.

For an isolated point charge Q, any sphere surrounding the charge contains the same net
charge Q(r) = Q, hence eq. (8) reproduces the Coulomb Law,

kQ
E(r) = . (9)
r2

Example: Thin Spherical Shell.

Now consider a thin spherical shell of radius R and uniform surface charge density

dQ Qnet
= = . (10)
dA 4R2

For this shell, a Gaussian sphere of radius r < R contains no charge at all, while a Gaussian
sphere of radius r > R contains the whole charge Qnet of the shell, thus

0 for r < R,

Q(r) = (11)
Qnet for r > R,

and therefore

E(r)

0 for r < R,

E(r) = (12)
kQnet for r > R,
r2

In other words, inside the shell there is no electric field, but outside the shell the electric field
is the same as if the whole charge was at the center.

3
Example: Solid Ball.

For our next example we take a solid ball of radius R and uniform 3D charge density

dQ Qnet
= = 4 3 . (13)
dV 3 R

Again, a Gaussian sphere of radius r > R contains the entire charge of the ball, hence

kQnet
for r > R, E(r) = , (14)
r2

outside the ball, its electric field is the same as if the whole charge was at the center of the
ball. This is an example of a general rule: outside any spherically symmetric charged body,
its electric field is the same as if the whole charge was at the center.

But inside the body, only the charges inside radius r contribute to the E(r). For the
solid ball, the Gaussian sphere of radius r < R contains charge

r3
 
ball of 4 3
Q(r) = V = r = Qnet 3 . (15)
radius r 3 R

Consequently, the electric field at radius r < R is

k Qnet r 3 kQnet r
E(r) = 2
3
= 2
. (16)
r R R R

Thus, inside the ball, the electric field increases linearly with the radial coordinate, while
outside the ball it decreases as 1/r 2 . Altogether,

E(r)

(17)

4
Example: Non-Uniform density.

Finally, consider a solid ball of radius R whose charge density depends on the radial
coordinate as

(r) = C r (18)

for some constants C and . The net charge of such a ball is

ZR ZR
Qnet = (r) dVolume(r) = Cr 4r 2 dr
0 0
ZR (19)
R+3
= 4C r +2 dr = 4C
+3
0
4
= CR+3 .
+3

Again, outside the ball the electric field is simply the Coulomb field of the net charge at
the center of the ball,
kQnet
for r > R, E(r) = . (20)
r2

But inside the ball, we have a more complicated formula. Indeed, the charge inside a Gaussian
sphere of radius r < R is

Zr Zr

Q(r) = (r ) dVolume(r ) = Cr 4r 2 dr
0 0 (21)
4  r +3
= Cr +3 = Qnet ,
+3 R

hence
kQ(r)  r +3 1 kQnet  r +1
E(r) = = kQnet = . (22)
r2 R r2 R2 R

5
 Cylindrical Symmetry

Cylindrically symmetric configurations of electric charges include a long thin rod, a long
hollow cylinder, several coaxial cylindrical shells (like a coaxial cable), as well as more general
systems which are uniform and infinitely long in one dimension and axially symmetric in the
other two dimensions. In cylindrical coordinates (rc , , z),

dQ = (rc ) dV = (rc ) drc rc d dz (23)

where the charge density (rc , \, \z ) depends only on the cylindrical radius rc but not on the
angle or the lengthwise coordinate z.

By symmetry, the electric field of a cylindrically symmetric charge configuration always

points directly away from the cylinders axis or directly towards the axis, i.e. it has no z or
components but only the rc component. Also, the magnitude of the electric field depends
only on the cylindrical radius rc ,

E(rc, , z) = E(rc only)~rc . (24)

The Gaussian surfaces for such an electric field are cylinders C of generic radii r and lengths
L but always coaxial with the charge distribution. For completeness sake, such a surface
comprises both the outside cylinder at fixed rc = r and the circular endcaps at fixed z = z1
or z = z2 ,

However, on the endcaps the electric field (24) lies parallel to the endcaps instead of crossing
them, so the flux through the endcaps is zero. Instead, the entire flux comes through the

6
outer cylinder where E stays normal to the surface and its magnitude stays constant E(r).
Altogether, the net flux through the cylinder C is

E [C] = E(rc = r) A(C) = E(rc = r) 2rL. (25)

By Gauss Law, this implies

 
inside cylinder of
2rL E(rc = r) = 4k Qnet (26)
length = L and radius = r

and therefore
 
2k inside cylinder of
E(rc ) = Qnet (27)
rL length = L and radius = rc
By translational symmetry along the cylinder axis, the charge inside the cylinder is always
proportional to its length L, so let us define the linear charge density inside a given radius
rc ,
 
1
def inside cylinder of
(rc ) = Qnet (28)
L length = L and radius = rc
In terms of this linear density,
2k (rc )
E(rc ) = . (29)
rc

Example: Thin Rod.

Consider an infinitely long, infinitely thin rod of uniform linear charge density . Any
Gaussian cylinder containing this rod has net charge Q = L regardless of the cylinders
radius. In terms of eq. (28), this means (rc ) for any rc > 0, hence by the Gauss Law
equation (29)
2k 2k
E(rc ) = = E = ~r . (30)
rc rc c
Note the agreement of this formula with the result of direct integration of Coulomb fields
over the length of the rod. But thanks to the Gauss Law, this time we did not have to do
any integrals.

7
Example: Thin Cylindrical Shell.

For our next example, consider an infinitely long thin cylindrical shell of radius R with
a uniform surface charge density

dQ net
= . (31)
dA 2R

For this shell, a Gaussian cylinder of radius rc < R contains no electric charge at all, while
a cylinder of radius rc > R contains charge Q = L. In terms of (rc ) this means

0 for r < R,

(rc ) = (32)
net for r > R,

and therefore

E(rc )

0 for rc < R,

E(rc ) = (33)
2knet for rc > R,
rc

rc

In other words, inside the shell there is no electric field, but outside the shell the electric field
is the same as if the whole charge was on the axis.

In general, outside any cylindrically symmetric charged body, the electric field is the
same as outside a thin rod of the same net charge density net .

Example: Solid Cylinder.

Now consider an infinitely long solid cylinder of radius R and uniform 3D charge density

dQ net
= . (34)
dV R2

This time, a Gaussian cylinder of radius rc > R contains charge Q = L net , while a

8
Gaussian cylinder of radius rc < R contains non-zero but smaller charge

rc2
Q = rc2 L = Lnet (35)
R2

In terms of (rc ), this means

2
r

for rc < R,
(rc ) = net R2 (36)

1 for rc > R,

and therefore

rc2 1 2knet rc
for rc < R, E(rc ) = 2knet = ,
R 2 rc R R
(37)
2knet
for rc > R, E(rc ) = ,
rc

inside the cylinder the electric field grows linearly with the radius as rc while outside the
cylinder it decreases as 1/rc ,

E(rc )

(38)

rc

9
Example: Thick Cylindrical Shell.

This time, consider a thick cylindrical shell of inner radius R1 and outer radius R2 ,

R1

R2

and uniform 3D charge density , hence net linear charge density

net = R22 R12 .

(39)

This time, a Gaussian cylinder of radius smaller than the inner radius of the shell contains
no electric charge at all, and there is no electric field in the hollow inside the cylinder,

for rc < R1 , (rc ) = 0 = E(rc ) = 0, (40)

On the other hand, the Gaussian cylinder of radius larger than the outer radius of the shell
contains the entire linear charge density of the shell, thus

2knet
for rc > R2 , (rc ) = net = E(rc ) = , (41)
rc

the electric field outside the shell is is similar to the field of a thin rod of the same net .

Finally, to find the electric field within the thickness of the shell, we use a Gaussian
cylinder of radius R1 < rc < R2 . The linear charge density inside this Gaussian cylinder is

rc2 R12
(rc ) = rc2 R12


= net , (42)
R22 R12

hence
rc2 R12 1
E(rc ) = 2knet . (43)
R22 R12 rc

10
Altogether,
E(rc )

(44)

rc

Planar Symmetry

Charge configurations with planar symmetry include uniform 2D sheets, uniform slabs
of finite thickness, as well as sandwiches of such sheets and slabs. In Cartesian coordinates
(x, y, z), a general configuration with planar symmetry has 3D charge density (x, y, z) which
depends only on z but not x or y. By symmetry, the electric field of such a configuration is
parallel to the z axis and its magnitude depends only on the z coordinates but not on x or
y,

E(x, y, z) = Ez (z only)~z . (45)

For simplicity, let me assume an additional upside-down symmetry z z, thus

(z) = (+z) = Ez (z) = Ez (+z). (46)

For charge configurations with planar and upside-down symmetries, the simplest Gaus-
sian surfaces to use are brick-like parallelepipeds with top and bottom surfaces at +z and
z and horizontal area A = x y. Since the electric field is vertical everywhere, it has
no flux through the four vertical sides of such a Gaussian brick, so the entire flux through
the brick comes from the horizontal top and bottom surfaces. Thus,
ZZ ZZ
E [brick] = dx dy Ez (x, y, +z) dx dy Ez (x, y, z) (47)
top bot

where the sign between the two integrals comes from opposite directions of the inside-
to-outside normals to the top and bottom surfaces, ntop = (0, 0, +1), nbot = (0, 0, 1).

11
Fortunately, the integrals in eq. (47) are trivial since the electric field does not depend on x
and y, hence


E [brick] = A Ez (+z) Ez (z) . (48)

Finally, in light of the upside-down symmetry (46),

E [brick] = 2A Ez (+z). (49)

By Gauss Law, this flux is related to the net electric charge of the Gaussian brick,
 
inside brick
2AEz (+z) = 4k Qnet (50)
A (z to + z)

Since the net charge inside such a brick is always proportional to the bricks horizontal area
A, lets define the net surface density of charge between z and +z,
  Z+z
1
def inside brick
(z) = Qnet = (z ) dz . (51)
A A (z to + z)
z

In terms of this surface density,

Ez (+z) = +2k (z), Ez (z) = 2k (z). (52)

Example: Thin Sheet.

As a first example, consider an infinitely thin charged sheet of uniform surface charge
density net . For this sheet, any Gaussian brick has net charge Q = net A and therefore
(z) net . Consequently, the electric field of this sheet is

Ez (+z) = +2knet , Ez (z) = 2knet , for any z > 0, (53)

or in other words,

E = 2ktot sign(z)~z . (54)

Note: this is the same formula we had earlier obtained by integrating the Coulomb field of
infinitesimal point charges all over the infinite 2D sheet. But this time, we did not have to
integrate anything.

12
Example: Thick Slab.

For our second example, consider a slab of finite thickness D and uniform 3D charge
density
net
= . (55)
D

For this charge configuration, a Gaussian brick of area A and thickness 2z < D has net
charge Q = A 2z, hence

2z
(z) = 2z = net (56)
D

and therefore

D 2z +z z
for 0 < z < , Ez (+z) = +2knet = 4knet , Ez (z) = 4knet .
2 D D D
(57)
In other words, the electric field inside the slab is

D D z
for < z < + , E(z) = 4kz~z = 2knet ~z . (58)
2 2 D/2

On the other hand, the Gaussian brick or thickness 2z > D has net charge Q = net A,
hence (z) = net and therefore

D
for z > , Ez (+z) = +2knet , Ez (z) = 2knet . (59)
2

In other words, the electric field outside the slab is

D
for |z| > , E(z) = 2knet sign(z)~z (60)
2

which is the same as the field of a thin sheet of the same net surface charge density net .

13
Altogether,
Ez

Generalization: No Upside-Down Symmetry.

Now consider a configuration of electric charges which has planar symmetry but not
the upside-down z z symmetry (46), for example an asymmetric sandwich of several
charged slabs with different thicknesses and charge densities. The electric field of such a
configuration has direction along the z axis and depends only on the z coordinate as in
eq. (45), but there is no general relation between E(+z) and E(z).

The Gaussian surfaces for such electric fields are bricks with bottom surface at z1 and
top surface at z2 , but now z1 and z2 are no longer related by a symmetry. For any such
brick, the electric flux through the 4 vertical sides vanishes since the electric field is vertical,
so the net flux comes through the top and bottom surfaces,
ZZ ZZ
E [brick] = dx dy (0, 0, +1) E(x, y, z2 ) + dx dy (0, 0, 1) E(x, y, z1 )
top bot (61)
= +A Ez (z2 ) A Ez (z1 ).

The Gauss Law relates this flux to the net electric charge inside the brick, or after dividing
by the horizontal area A, to the

  Zz2
def 1 inside
(z1 , z2 ) = = Q = (z) dz. (62)
A brick
z1

Specifically,
 
inside
E [brick] = 4k Q = 4kA (z1 , z2 ) (63)
brick

14
and therefore
Zz2
Ez (z2 ) Ez (z1 ) = 4k(z1 , z2 ) = 4k (z) dz. (64)
z1

By itself, this formula does not give us the electric field. But if we know the field at any
particular value of z, then eq. (64) gives us the electric field everywhere else.

Mathematically, eq. (64) is the formal solution of the differential equation

dEz
= 4k (z), (65)
dz

which is the special case (for systems with planar symmetry) of the Maxwells electrostatic
equation
Ex Ey Ez
E = + + = 4k(x, y, z). (66)
x y z

15