Gradient of a scalar field

Let 𝜙(𝑥, 𝑦, 𝑧) is a scalar func.on defined everywhere in 3-d real space. We want to know the changes of
the func.on 𝜙 near a point 𝑃. We consider a neighboring point such that

𝑟⃗! = 𝑟⃗" + 𝛿𝑥𝚤̂ + 𝛿𝑦𝚥̂ + 𝛿𝑧𝑘2 = 𝑟⃗" + 𝛿 𝑟⃗

From Taylor’s series expansion,
𝜕𝜙 𝜕𝜙 𝜕𝜙
𝜙 𝑟⃗! − 𝜙 𝑟⃗" ≈ 𝛿𝑥 + 𝛿y + 𝛿𝑧 ≡ grad(𝜙 𝑟⃗" ⋅ 𝛿 𝑟⃗ = ∇𝜙 𝑟⃗" ⋅ 𝛿 𝑟⃗
𝜕𝑥 𝜕𝑦 𝜕𝑧
where
𝜕 𝜕 𝜕 𝑦
∇≡ 𝚤̂ + 𝚥̂ 2
+𝑘
𝜕𝑥 𝜕𝑦 𝜕𝑧
Is called that gradient operator and
grad 𝜙 ≡ ∇𝜙
𝑄
×

𝑃
×

𝑥
 along the direction of a tangent to the equipotential surface at P
and it attains its maximum value of |∇ψ | when û and ∇ψ are in
the same direction. The relation between these quantities is shown
in Figure 12.3.

Example 12.1
gure 12.3 Gradient and
ectional derivative, where If u = xy and w = yz, find expressions for grad (uw), u grad w and
e vector u indicates the w grad u.
ection of û defined in the
xt.
Solution
∂ ∂ ∂
grad (uw) = ∇(uw) = (xy2 z)i + (xy 2 z)j + (xy2 z)k
∂x ∂y ∂z
= y 2 z i + 2xyz j + xy 2 k,
! "
∂ ∂ ∂
u grad w = u∇w = u (yz)i + (yz)j + (yz)k
∂x ∂y ∂z
= uz j + uy k = xyz j + xy2 k,
and ! "
∂ ∂ ∂
w grad u = w∇u = w (xy)i + (xy)j + (xy)k
∂x ∂y ∂z
= wy i + wx j = y 2 z i + xyz j.

Example 12.2
A scalar field is given by ψ = x2 − y 2 z. Find ∇ψ at the point
(1, 1, 1); the directional derivative there in the direction i − 2j + k;
the equation of the line passing through and normal to the surface
ψ = 0 at the point (1, 1, 1).
 ∂x ∂y ∂z
= uz j + uy k = xyz j + xy2 k,
and ! "
∂ ∂ ∂
w grad u = w∇u = w (xy)i + (xy)j + (xy)k
∂x ∂y ∂z
= wy i + wx j = y 2 z i + xyz j.

Example 12.2
A scalar field is given by ψ = x2 − y 2 z. Find ∇ψ at the point
(1, 1, 1); the directional derivative there in the direction i − 2j + k;
the equation of the line passing through and normal to the surface
ψ = 0 at the point (1, 1, 1).

Solution
The gradient of ψ is
∇ψ = 2x i − 2yz j − y 2 k,
and at (1, 1, 1) this is ∇ψ = 2i − 2j − k. The directional derivative
is, from (12.8),
dψ/ds = ∇ψ · û,
√
where û = √16 (i − 2j + k), so that dψ/ds = 5/ 6. Since the direc-
tion of the normal is
∇ψ = 2i − 2j − k,
the equation of the line normal to the surface at (1,1,1) is
[cf. (8.37a)]
r = (xi + yj + zk) + t(2i − 2j − k),

where t is a variable parameter.

www.pdfgrip.com
2.2.1 Line integrals
Section
Line 8.3.1, we saw that the running vector
integral

A vector function 𝑟⃗ 𝑠 = r𝑟# =

𝑠a𝚤̂ ++𝑟$s 𝑠b 𝚥̂ + 𝑟% 𝑠 𝑘2 for a running parameter
(12.21)𝑠
define a curve in the space.
here a and b are fixed vectors, described a straight line passing
An infinitesimal vector directed along the tangent to the curve is given by
hrough the point r = a in the 𝑑 𝑟⃗direction
𝑑𝑟# b̂ as𝑑𝑟$the 𝑑𝑟scalar
%
parameter s
aried in the range −∞ <𝑑𝑟⃗s =<𝑑𝑠∞𝑑𝑠 =
. More 𝚤̂ +
𝑑𝑠generally,
𝑑𝑠
𝚤̂ + any𝚤̂ running
𝑑𝑠
𝑑𝑠 vector
s), where r is a differentiable function of s, will describe a curve in
pace and forsuppose
Now any given
we haves the differential
a vector ⃗ 𝑟).
function 𝐹( ⃗ We can define the line integral

= C 𝐹⃗ 𝑟⃗ ⋅ 𝑑 𝑟⃗
𝑊dr
dr = ds& (12.22)
ds ⃗
For example, 𝑊 can be interpreted as the work done by the force 𝐹 in moving
from initial position 𝑟⃗' to the final position 𝑟⃗( . We can reduce it to a one-
an infinitesimal Figure 12.4 Definition of a
dimensional vector directed
integral over 𝑠, along the tangent to the curve, as
space curve for the line
hown in Figure 12.4 for )" the point P corresponding to s = s . For
𝑑𝑟# 𝑑𝑟$ 𝑑𝑟% 0
𝑊=D 𝐹# 𝑟⃗ 𝑠 + 𝐹$ 𝑟⃗ 𝑠 + 𝐹% 𝑟⃗ 𝑠 𝑑𝑠 integral of a scalar product.
xample, in Cartesian co-ordinates
)! 𝑑𝑠 𝑑𝑠 𝑑𝑠 The path is defined in terms
of a parameter s, so that
r(s) = s i + (4as)1/2 j r = r(s).

www.pdfgrip.com
 directly together with the relations y = y(x), z = z(x) along the con-
tour C.
At this point we note that (12.26) is identical with the line inte-
gral (11.13) discussed in Section 11.1.3, if the functions Q, R, P are
replaced by functions Vx , Vy , Vz . Hence the methods and results dis-
cussed in Section 11.1.3 can be carried over, with a trivial relabeling,
to the line integrals (12.26) as we shall illustrate in the next section.

Example 12.6
The vector V is given by
V = (3x + 6y 2 )i − 4x2 yz j + 2xz 3 k.
Evaluate the line integral ˆ
V · dr,

from the point (0,0,0) to the point (1,1,1) to four significant fig-
ures, where the curve is given (a) by the parametric form
x = t2 , y = t, z = t3
and (b) by a straight line between the two end points.

Solution
The integral is
ˆ ˆ
V · dr = [(3x + 6y 2 )i − 4x2 yzj + 2xz 3 k] · (dx i + dy j + dz k)
ˆ ˆ ˆ
= (3x + 6y 2 )dx − (4x2 yz)dy + (2xz 3 )dz

(a) Using x = t2 , y = t, z = t3 gives

ˆ ˆ1
V · dr = (18t3 − 4t8 + 16t13 )dt = 4.484,
0
358 Mathematics for physicists
since (0,0,0) and (1,1,1) correspond to t = 0 and t = 1, respec-
tively.
(b) In parametric form, x = y = z = t and so
ˆ ˆ1 www.pdfgrip.com
V · dr = (3t + 6t2 − 2t4 )dt = 3.100
0

Example 12.7
Evaluate the integral
ˆ ˆ
V · dr = [ydx − y(x − 1)dy + y 2 z dz],
C C
(b) In parametric form, x = y = z = t and so
ˆ ˆ1
V · dr = (3t + 6t2 − 2t4 )dt = 3.100
0
Figure 12.5
Example 12.7
Evaluate the integral the path C of the integral, which is shown in blue. The point P is
ˆ ˆ
one such intersection point. We need to express C in a parametric
V · dr = [ydx − y(x − 1)dy + y 2 z dz], form, by using the fact that points on C satisfy
C C

where C is the curve given by the intersection of the sphere x2 + y 2 + z 2 = 4 and (x − 1)2 + y 2 = 1.
x2 + y 2 + z 2 = 4 Since the point P lies on the cylinder, we can write Vector
and the cylinder
(x − 1)2 + y 2 = 1, x − 1 = cos θ, y = sin θ,
in the positive octant x, y, z > 0, between the points A(2, 0, 0) and where θ is the angle shown in Figure 12.5 and 0 ≤ θ ≤ π in the
B(0, 0, 2). positive octant. Substituting into the equation for the sphere gives
www.pdfgrip.com
1 + 2 cos θ + cos2 θ + sin2 θ + z 2 = 4 ⇒ z = 2 sin(θ/2) (z > 0).
Solution
The solid black lines in Figure 12.5 show an octant of the sphere; Then the integral becomes
the dashed lines show the cylinder. Where they intersect defines
ˆπ
[sin θ d(1 + cos θ) − sin θ cos θ d(sin θ)
0
+2 sin2 θ sin (θ/2) d(2 sin θ/2)]
ˆπ
= (− sin2 θ − sin θ cos2 θ + sin3 θ)dθ = 2/3 − π/2,
0

where the final integral is evaluated using the methods of

Section 4.3.2.

12.2.2 Conservative fields and potentials

the path C of the integral, which is shown in blue. The point P is
one such intersection point. We need to express C in a parametric The result of a line integral of a vector between any two points will in
form, by using the fact that points on C satisfy general depend on the path taken between them. If, however, the line
integral is independent of the path for any choice of end points within
x2 + y 2 + z 2 = 4 and (x − 1)2 + y 2 = 1. the field, the vector field V is said to be conservative. Conservative
Since the point P lies on the cylinder, we can write fields play an important role in physics, as we shall now see.
Normal vector of a surface 𝑆
In 3d, a surface 𝑆 is defined by an equation of the form,
𝑓 𝑥, 𝑦, 𝑧 = 0.
and our goal is to find the normal vector 𝑛- at a point 𝑃 on this surface.

We consider a neighboring point 𝑄 (𝑟⃗! = 𝑟⃗" + 𝛿 𝑟)

⃗ on the same surface such that
𝑓 𝑟⃗" = 𝑓 𝑟⃗! . From the definition of gradient, we have
0 = 𝛿𝑓 = 𝑓 𝑟⃗! − 𝑓 𝑟⃗" = ∇𝑓 ⋅ 𝛿 𝑟⃗
Therefore, we can conclude that the vector ∇𝑓 is perpendicular to the vector 𝛿 𝑟.
⃗
𝑦
In other words, ∇𝑓 is normal to the surface. And the normal unit vector (which
perpendicular to the surface 𝑓 𝑥, 𝑦, 𝑧 = 0) is defined,
𝑄
∇𝑓 𝛿 𝑟⃗ ×
𝑛- =
|∇𝑓| 𝑃× 𝑓=0

𝑥
Surface integral
We can now introduce the surface integral over a surface 𝑓 = 0,

7 𝐸 ⋅ 𝑑 𝑆⃗ ,
#
∇%
where 𝑑 𝑆⃗ = 𝑛𝑑𝑆
- = 𝑑𝑆.
|∇%|

For example, if one is integrating over the 𝑥 − 𝑦 plane, we have

𝑓 𝑥, 𝑦, 𝑧 = 𝑧 = 0. and. 𝛻𝑓 = 𝑘>
Therefore, the normal vector of the surface is
𝑛- = 𝑘>
and

7 𝐸 ⋅ 𝑑 𝑆⃗ = @𝑑𝑥 @𝑑𝑦 𝐸' (𝑥, 𝑦, 0)

#
This general formula can be used to convert an integral over a curved
surface S to an integral over A in the xy plane, as illustrated in
Example 12.11 below.

Example 12.10
Evaluate the integral ¨
I= V · ds,
s

where V = xi and S is the surface of the hemisphere

x2 + y 2 + z 2 = a2 , z≥0

Solution
In spherical polar co-ordinates, the surface is r = a with
0 ≤ θ ≤ π/2 and 0 ≤ φ ≤ 2π, while the surface element
ds = r2 sin θ dθ dφ er . Hence the integral is
ˆπ/2 ˆ2π
I = a2 dθ (V · er ) sin θ dφ,
0 0

where, for r = a,
V · er = a sin θ cos φ(i · er ) = a sin2 θ cos2 φ
using (11.41) for er . Therefore,

ˆπ/2 ˆ2π
3 2 2 2πa3
I = −a sin θ dcos θ cos φ dφ = .
3
0 0

www.pdfgrip.com
Divergence

Now we consider a vector func.on (or some.mes we call it vector field) 𝑣( ⃗ which is defined
⃗ 𝑟)
for the en.re 3-d real space.

The defini.on of divergence is defined by

𝜕 𝜕 𝜕 𝜕𝑣# 𝜕𝑣$ 𝜕𝑣%
𝛻 ⋅ 𝑣⃗ = 𝚤̂ + 𝚥̂ + 𝑘2 ⋅ 𝑣# 𝚤̂ + 𝑣$ 𝚥̂ + 𝑣% 𝑘2 = + +
𝜕𝑥 𝜕𝑦 𝜕𝑧 𝜕𝑥 𝜕𝑦 𝜕𝑧
Just like the dot (scalar) product between the gradient operator ∇ and the vector field 𝑣.
⃗

The divergence 𝛻 ⋅ 𝑣⃗ is a measure of how much the vector 𝑣⃗ spreads

out (diverges) from the point in ques.on. For example, the figure in (a)
has a large (posi.ve) divergence and (b) has zero divergence, and (c) has
a posi.ve divergence.
 A vanishing value of div F indicates that the field
Fy(x, y + ∆y, z) sinks.
Fy(x, y, z)
Having a good grasp of the concept, we now must
∆z
compute the divergence.
Fz(x, y, z)
Let us consider a small cuboid of dimensions ∆x, ∆
x ∆x
y
!
Divergence div A = lim
1
F · dA
∆y V →0 V

The surface integral is given by the sum of the flo

Fig.get
In 17.25
a better understanding of the divergence. We consider a surface integral overone
a the
infinitesimal
on each vector field will be constant, approxim
volume.
As the components of F are parallel to the surface element vectors, we get: z
Fz(x, y, z + ∆z)
!
1 1
F · dA ≈ {[Fx (x + ∆x, y, z) − Fx (x, y, z)]∆y∆z
V ∆x∆y∆z
Fy(x, y, z) Fy(x, y + ∆y, z)
+[Fy (x, y + ∆y, z) − Fy (x, y, z)]∆x∆z
∆z Fz(x, y, z)
+[Fz (x, y, z + ∆z, z) − Fz (x, y, z)]∆x∆y} x ∆x
y

Fx (x + ∆x, y, z) − Fx (x, y, z)
= ∆y

∆x Fig. 17.25
Fy (x, y + ∆y, z) − Fy (x, y, z)
+
∆y As the components of F are parallel to the surface
Fz (x, y, z + ∆z) − Fz (x, y, z) !
+ 1 1
{[Fx (x + ∆x, y, z)
∆z V
F · dA ≈
∆x∆y∆z
In the limit V → 0 i.e. ∆x → 0, ∆y → 0, ∆z → 0, we end up with three partial +[Fy (x, y + ∆y, z) − Fy (x,
differentials. +[Fz (x, y, z + ∆z, z) − Fz (x
1 Fx (x + ∆x, y, z) − Fx (x, y
𝛻 ⋅ 𝐹⃗ = C 𝐹⃗ ⋅ 𝑑 𝐴⃗ (over an infinitesimal volume) =
∆x
𝑉
Fy (x, y + ∆y, z) − Fy (x,
+
∆y
Fz (x, y, z + ∆z) − Fz (x,
+
∆z
In the limit V → 0 i.e. ∆x → 0, ∆y → 0, ∆z → 0,
Geometric meaning of the divergence

1
𝛻 ⋅ 𝐹⃗ = C 𝐹⃗ ⋅ 𝑑 𝐴⃗
𝑉

In other words, ∇ ⋅ 𝐹⃗ at a point 𝑟⃗ is the flux per unit volume out of an infinitesimal volume 𝑉 surrounding 𝑟.
⃗
 n
∑ div F (xi , yi , zi
i =1

It is quite apparent that all adjacent f

Divergence Theorem values which cancel out. So just the valu
to be summed up. In the limit n → ∞ and
We have
n
1
⃗ ( , 𝑦( , 𝑧( ) ≈
𝛻 ⋅ 𝐹(𝑥 H 𝐹⃗ ⋅ 𝑑 𝐴⃗ lim
n→∞
∑ div F · (xi , yi ,
Δ𝑉( i =1
Multiplication by Δ𝑉( and adding the 𝑛 values yields and furthermore
+ + n !
⃗ ( , 𝑦( , 𝑧( ) Δ𝑉( ≈ J H 𝐹⃗ ⋅ 𝑑 𝐴⃗
J 𝛻 ⋅ 𝐹(𝑥 lim
n→∞
∑ F·
()* ()*
i =1 Ai
It is quite apparent that all adjacent faces of the Δ𝑉( in the interior give rise to values
which cancel out. So just the values of 𝐹⃗ ⋅ 𝑑 𝐴⃗ on the surface 𝐴 are left over to be
summed up. In the limit 𝑛 → ∞ and Δ𝑉( → 0, we arrive at
+
⃗ ( , 𝑦( , 𝑧( ) Δ𝑉( = @ 𝛻 ⋅ 𝐹⃗ 𝑑𝑉
lim J 𝛻 ⋅ 𝐹(𝑥
+→- .
()* dAi+1= –dAi dAi
and
+

lim J H 𝐹⃗ ⋅ 𝑑 𝐴⃗ = H 𝐹⃗ ⋅ 𝑑 𝐴⃗
+→- ∆Vi
()* /
–∆Vi+1

⇒ @ 𝛻 ⋅ 𝐹⃗ 𝑑𝑉 = H 𝐹⃗ ⋅ 𝑑 𝐴⃗
. /
Fig. 17.27

All this adds up to the following theor

Gauss law in differential form
Consider the Gauss law,
𝑄*+,-
C 𝐸 ⋅ 𝑑 𝑆⃗ =
𝜖.
Notice that by the divergence theorem,
C 𝐸 ⋅ 𝑑 𝑆⃗ = D𝛻 ⋅ 𝐸𝑑𝑉

and
𝑄*+,- = D𝜌𝑑𝑉
We have
𝜌
D𝛻 ⋅ 𝐸𝑑𝑉 = D 𝑑𝑉
𝜖.
for any Gaussian volume.
𝜌
⇒𝛻⋅𝐸 =
𝜖.
Example:

Suppose the electric field in some region is found to be 𝐸 = 𝑥 Q 𝚤̂ + 𝑥 𝚥.̂

(a)Find the charge density 𝜌(𝑟)

⃗
(b)Find the total charge contained in a box of length 2, centered at the origin. (Do it
two different ways)
(a) By the Gauss law,
𝜕𝐸#
𝜌 𝑟⃗ = 𝜌(𝑥, 𝑦, 𝑧) = 𝜖. 𝛻 ⋅ 𝐸 = 𝜖. = 2𝑥𝜖.
𝜕𝑥

(b) Method 1:
The total charge is
0 0 0 0
0
𝑄 = D 𝑑𝑥 D 𝑑𝑦 D 𝑑𝑧 𝜌 𝑥, 𝑦, 𝑧 = 4 D 2𝑥𝜖. 𝑑𝑥 = 4𝜖. 𝑥 1 /0 =0
/0 /0 /0 /0

Method 2:
By the gauss law,
𝑄
= C 𝐸 ⋅ 𝑑 𝐴⃗
𝜖
0 0 0 0
= D 𝑑𝑦 D 𝑑𝑧 𝐸# 1, 𝑦, 𝑧 − 𝐸# −1, 𝑦, 𝑧 + D 𝑑𝑥 D 𝑑𝑧 𝐸$ 𝑥, 1, 𝑧 − 𝐸$ 𝑥, −1, 𝑧
/0 /0 /0 /0
0 0 0 0
= D 𝑑𝑦 D 𝑑𝑧 11 − −1 1
+ D 𝑑𝑥 D 𝑑𝑧 𝑥 − 𝑥 = 0
/0 /0 /0 /0
In the previous example, the gradient operator ∇ is acFng on a vector 𝑣⃗ to give a scaler number (divergence)
𝜕𝑣0 𝜕𝑣1 𝜕𝑣'
∇ ⋅ 𝑣⃗ ≡ + +
𝜕𝑥 𝜕𝑦 𝜕𝑧

In fact, the gradient operator can also act on a scaler funcFon 𝑓(𝑥, 𝑦, 𝑧) resulFng a vector.

𝜕𝑓 𝜕𝑓 𝜕𝑓
∇𝑓 𝑥, 𝑦, 𝑧 = 𝚤̂ + 𝚥̂ + 𝑘>
𝜕𝑥 𝜕𝑦 𝜕𝑧
Example: X
What is 𝛻 ?
Y

We can calculate the gradient using the Cartesian coordinate. (In fact, using the spherical coordinate is
faster and simpler)

𝜕 1 𝜕 1 𝑥 𝑥
= = =
𝜕𝑥 𝑟 𝜕𝑥 𝑥1 + 𝑦1 + 𝑧1 𝑥1 + 𝑦1 + 𝑧1 2/1 𝑟2

1 𝜕 1 𝜕 1 𝜕 1 𝑥𝚤̂ + 𝑦𝚥̂ + 𝑧𝑘2 𝑟̂

⇒𝛻 = 𝚤̂ + 𝚥̂ + 2
𝑘= =
𝑟 𝜕𝑥 𝑟 𝜕𝑦 𝑟 𝜕𝑧 𝑟 𝑟2 𝑟1

1 𝑟̂
𝛻 = Q
𝑟 𝑟
Electric dipole - potential

𝑞 1 1
𝑉 𝑟⃗ = −
4𝜋𝜖2 𝑑 𝑑
𝑟⃗ − 𝚥̂ 𝑟⃗ + 𝚥̂
2 2
𝑞 1 1
⇒ 𝑉 𝑟⃗ = −
4𝜋𝜖2 𝑑3 𝑑3
𝑟 3 − 𝑑 𝑟⃗ ⋅ 𝚥̂ + 𝑟 3 + 𝑑 𝑟⃗ ⋅ 𝚥̂ +
4 4
𝑞 1 1
⇒ 𝑉 𝑟⃗ = −
4𝜋𝜖2 𝑟 𝚥̂ 𝑑3 𝚥̂ 𝑑3
1 − 𝑑 𝑟̂ ⋅ + 3 𝑟 3 + 𝑑 𝑟̂ ⋅ + 3
𝑟 4𝑟 𝑟 4𝑟
𝑞 1 𝚥̂ 1 𝚥̂ 𝑟⃗
⇒ 𝑉 𝑟⃗ ≈ 1 + 𝑑 𝑟̂ ⋅ − 1 − 𝑑 𝑟̂ ⋅
4𝜋𝜖2 𝑟 2 𝑟 2 𝑟 +
1 𝑟̂ ⋅ 𝑞𝑑 𝚥̂
⇒ 𝑉 𝑟⃗ = 𝑝⃗ = 𝑞𝑑 𝚥̂
4𝜋𝜖2 𝑟3

1 𝑝⃗ ⋅ 𝑟̂ −
⇒ 𝑉 𝑟⃗ =
4𝜋𝜖_ 𝑟 Q
Electric dipole – electric field
The electric field is defined by
1 𝑝⃗ ⋅ 𝑟̂ 1 𝑝# 𝑥 + 𝑝$ 𝑦 + 𝑝% 𝑧
𝐸 𝑟⃗ = −𝛻𝑉 𝑟⃗ = − 𝛻 =− 𝛻 2
4𝜋𝜖. 𝑟1 4𝜋𝜖. 1 1 1
𝑥 +𝑦 +𝑧 1
1 𝑝# 3 2𝑥(𝑝# 𝑥 + 𝑝$ 𝑦 + 𝑝% 𝑧)
⇒ 𝐸# = − 2 − 4
4𝜋𝜖. 1 1 1 2 1 1 1
𝑥 +𝑦 +𝑧 1 𝑥 +𝑦 +𝑧 1
1 3 𝑝⃗ ⋅ 𝑟⃗ 𝑥 𝑝#
⇒ 𝐸# = 4 − 2
4𝜋𝜖. 1 1 1 1 1 1
𝑥 +𝑦 +𝑧 1 𝑥 + 𝑦 + 𝑧 1
Similarly, we obtain 𝐸$ and 𝐸% .

To conclude,
1 3 𝑝⃗ ⋅ 𝑟⃗ 𝑟⃗ 𝑝⃗ 1 3 𝑝⃗ ⋅ 𝑟̂ 𝑟̂ − 𝑝⃗
𝐸 𝑟⃗ = − =
4𝜋𝜖. 𝑟4 𝑟2 4𝜋𝜖. 𝑟4
Electric dipole
Now we consider a (large) volume of dielectric such that in each infinitesimal volume, there is a dipole
moment 𝑝⃗ = 𝑃 𝑟′⃗ 𝑑 2 𝑟′. The total potential is
1 𝑢c ⋅ 𝑃(𝑟⃗ 6 ) 2
𝑉 𝑟⃗ = D 𝑑 𝑟′
4𝜋𝜖. 5 𝑢1
where 𝑢 = 𝑟⃗ − 𝑟′.
⃗ Now notice that
1 𝑢c
𝛻′ = 1
𝑢 𝑢
We have
1 1
𝑉 𝑟⃗ = D 𝛻6 ⋅ 𝑃(𝑟⃗ 6 )𝑑 2 𝑟′ 𝑢
4𝜋𝜖. 5 𝑢
Using the product rule, we have 𝑟⃗
1 𝑃(𝑟′)
⃗ 1 6 𝑟′
⃗
⇒ 𝑉 𝑟⃗ = D 𝛻6 ⋅ 𝑑2 𝑟 6 − D 𝛻 ⋅ 𝑃 𝑑 2 𝑟′
4𝜋𝜖. 5 𝑢 5𝑢

Using the divergence theorem, we finally obtain

1 1 1 6
⇒ 𝑉 𝑟⃗ = ⃗ ⋅ 𝑑 𝑆⃗ − D
D 𝑃(𝑟′) 𝛻 ⋅ 𝑃(𝑟′
⃗ 𝑑𝑉
4𝜋𝜖. 7 𝑢 5 𝑢
Electric dipole

1 1 1 6
𝑉 𝑟⃗ = D 𝑃(𝑟′)
⃗ ⋅ 𝑛𝑑𝑆
c −D 𝛻 ⋅ 𝑃(𝑟′
⃗ 𝑑𝑉
4𝜋𝜖. 7 𝑢 5 𝑢

surface charge bulk (volume) charge

𝜎 𝑟⃗ 6 = 𝑃 𝑟⃗ 6 ⋅ 𝑛 ̂ 𝜌 𝑟⃗ 6 = −𝛻 6 ⋅ 𝑃(𝑟⃗ 6 )

𝑢
𝑝⃗X 𝑝⃗Q 𝑟⃗
𝑟′
⃗
− + − +
 nsider a dielectric sphere of radius Ä and permittivity p in a uniform electric
the polarization of the dielectric material, the electric field in the sphere and
is modified. The polarization of the dielectric sphere is due to mobile
ifted in the uniform electric field. Here we model the dielectric effects by
harged spheres with radius Ä and charge density ±1 being displaced along Example:
ents ±I/2 respectively (see Fig. 1).
rq Ä p
B

±1 rq ±I/2,

#⃗
Q4-2 English (Official)
𝜎`aYb = 𝑃 ⋅ 𝑛@ = 𝑃 cos 𝜃

!"
Fig.1: Metal ball under the uniform electrostatic field.

ball behavior under the plane wave field of frequency and amplitude . The radius of
ppears that in case when wavelength and field penetration depth are both much greater
the ball, one can consider the metal ball behavior as a dielectric ball under the uniform
d. But one have to take (analogous to one expressed in the previous part) in place
Pan Pearl 2018 – Paper 2 – Q2

As wave propagates the external electric intensity alternates cos . The dipole
ternates cos .

tch qualitatively the field lines (inside, near and far from the ball) in the sys- 0.5pt
assuming .
A uniformly polarized sphere of radius 𝑅 (𝑃 = 𝑃 𝑧)̂ induced surface bound charge

ermine , the value of (given that ) when the internal electric 0.1pt
nsity increases without limit.

ase in the electric field amplitude with frequency equaling is called plasmon reso-
g there is no energy dissipation we have the infinite answer for . Now take account
n. Assume that the major loss of energy is coming from dipole radiation.

oscillating dipole emits energy. Estimate the power of this energy loss 0.4pt
Curl
4 4 4
̂ 4' 𝑘> and a
Basically, the divergence is defined as the scaler product between the gradient operator ∇= 40 𝚤̂ + 41 𝚥+
>
vector function (or sometimes we call it vector field) 𝑣⃗ 𝑟⃗ = 𝑣0 𝚤̂ + 𝑣1 𝚥̂ + 𝑣' 𝑘.

In fact, we can also define the vector cross product between ∇ and a vector field 𝑣,
⃗

𝚤̂ 𝚥̂ 𝑘>
𝜕 𝜕 𝜕 𝜕𝑣' 𝜕𝑣1 𝜕𝑣0 𝜕𝑣' 𝜕𝑣1 𝜕𝑣0
∇×𝑣⃗ = = − 𝚤̂ + − 𝚥̂ + − 𝑘>
𝜕𝑥 𝜕𝑦 𝜕𝑧 𝜕𝑦 𝜕𝑧 𝜕𝑧 𝜕𝑥 𝜕𝑥 𝜕𝑦
𝑣0 𝑣1 𝑣'
is called the curl of 𝑣.
⃗
 Vector calculus 351
Example 12.3
Example 12.3 Given that
Given that
A = xy i − y 2 z j + xz 2 k, B = xy 2 i + xz j − 3xy k, φ = 2xyz 2 ,
A = xy i − y 2 z j + xz 2 k, B = xy 2 i + xz j − 3xy k, φ = 2xyz 2 ,
find: (a) ∇φ, (b) divA, (c) curl B, (d) (∇ · B)A, (e) (B · ∇)A,
find: (a) ∇φ, (b) divA, (c) curl B, ∇(2∇
(f)(d) φ, ·(g) ∇2(e)
B)A, A. (B · ∇)A,
(f) ∇2 φ, (g) ∇2 A.
Solution
Solution
∂φ ∂φ ∂φ
∂φ ∂φ ∂φ 2 ∇φ =
(a) 2 i+ j+ k = 2yz 2 i + 2xz 2 j + 4xyz k,
(a) ∇φ = i+ j+ k = 2yz i + 2xz j +
∂x4xyz k,
∂y ∂z
∂x ∂y ∂z
∂ ∂ 2 ∂z ∇ 2· A = ∂ (xy) + ∂ (−y 2 ) + ∂z (xz 2 ) = y − 2yz + 2xz,
(b)
(b) ∇ · A = (xy) + (−y ) + (xz ) = y −∂x2yz + 2xz,∂y ∂z
∂x ∂y ∂z
! ! ! !
! i j k ! ! i j k !
! ! ! !
!
! ∂
!
!
! !
∂ ∂ ! ∂ ∂ ∂ !
(c) ∇ × B = !! ! =(c)
! ( − 3x
∇ −
× B
x)i =
− (
! −3y)j ! = (−3x − x)i − (−3y)j
∂x ∂y ∂z ! ∂z !!
!
!
!
! ! ∂x ∂y
! xy 2 xz −3xy ! ! !
! xy 2 xz −3xy !
+(z − 2xy)k = −4xi + 3y j + (z − 2xy)k.
+(z − 2xy)k = −4xi + 3y j + (z − 2xy)k.
(d) (∇ · B)A = (y 2 )A = xy3 i − y 4 z j + xy2 z 2 k
" (d) (∇ ·#B)A = (y 2 )A = xy3 i − y 4 z j + xy2 z 2 k
∂ ∂ ∂ " #
(e) (B · ∇)A = xy 2 + xz − 3xy A ∂ ∂ ∂
∂x ∂y (e) (B∂z· ∇)A = xy 2 + xz − 3xy A
∂x ∂y ∂z
= xy 2 (y i + z 2 k) + xz(xi − 2yz j)
2 2
 where th
screw ru
The l
circulatio
∇×𝑣⃗ =the
0 surfa
Stokes’ theorem the boun
curl. To
surface e
Given a closed loop 𝐶, spanned by a surface 𝑆, and a vector field 𝑣( ⃗ defined on 𝑆, then
⃗ 𝑟)
this limi
Stokes’ theorem states that the left-h
⃗ ⋅ 𝑑 𝐴⃗ = H 𝑣⃗ ⋅ 𝑑 𝑟⃗
7 (∇×𝑣)
# 5
Where the sense of the vector element 𝑑 𝐴⃗ is given by a right-handed rule with respect to
the direction of integration around 𝐶. In other
around t
point r.
The line integral on the right-hand side is called the circulation of 𝑣⃗ around the loop 𝐶. ∇×𝑣⃗ ≠where
0 ρ
The theorem states that the surface integral of curl 𝑣⃗ is equal to the circulation of 𝑣⃗ uniform
around the boundary curve 𝐶. and V is
vortex, l
We can apply the Stokes’s theorem to a loop 𝐶 that encloses a small surface element non-zero
𝑑 𝐴⃗ = 𝑛𝑑𝐴,
- which shrinks to a point when 𝑑𝐴 → 0. In this limit, we have since the
∇×𝑣⃗ curl
≠ 0 V "=
1
∇×𝑣⃗ ⋅ 𝑛- = lim H 𝑣⃗ ⋅ 𝑑 𝑟⃗ Figure 12.12 Flow of a fluid. to, trans
6/→2 𝑑𝐴 The coloured lines are the would ha
flow lines; the arrows show to rotate
In other words, ∇×𝑣⃗ at a point 𝑟⃗ is the circulation per unit area around the boundary of the direction of the vector
an infinitesimal surface 𝑑𝐴. field V. Their lengths show
curl V =
the relative magnitudes of V. ures 12.1
In the
then con

12.4.1
We start
S paralle
 mittivity and permeability of the medium.
(a) Show that Maxwell’s equations imply that ρ is the density of a
conserved charge.
(b) In a conductor, the current obeys Ohm’s law j = σE, where σ
is the conductivity. Find the charge density ρ as a function of time
if ρ(r, t = 0) = ρ0 (r).
12.25 Verify Stokes’ theorem for the vector

A = (2x − y)i − yz 2 j − y 2 z k,

where S is the surface of the hemisphere

x2 + y 2 + z 2 = 1, z > 0,

and C is the boundary of S.

12.26 Use Stokes’ theorem (12.68) to prove the relation
¨ ˛
ds × ∇φ = φ ds,
S
C

where φ is a scalar fields and S is an open surface bounded by

a closed curve C. [Hint: apply Stokes’ theorem to the vector field
V = φ c, where c is a constant vector.]
2 2
The boundary is defined as 𝐶: 𝑥 3 +𝑦 3 = 1. Hence the line integral is

H 𝐴⃗ ⋅ 𝑑 𝑟⃗
5
The line integral can be computed using the polar coordinates:
𝑥 = cos 𝜃 , 𝑦 = sin 𝜃
The vector field (when 𝑧 = 0) becomes,
𝐴⃗ = 2 cos 𝜃 − sin 𝜃 𝚤̂
And the displacement vector 𝑑 𝑟⃗ reads,
𝑟⃗ = cos 𝜃 𝚤̂ + sin 𝜃 𝚥̂ ⇒ 𝑑 𝑟⃗ = − sin 𝜃 𝚤̂ + cos 𝜃 𝚥̂ 𝑑𝜃
Finally,
37 37 37
H 𝐴⃗ ⋅ 𝑑 𝑟⃗ = @ 2 cos 𝜃 − sin 𝜃 𝚤̂ ⋅ − sin 𝜃 𝚤̂ + cos 𝜃 𝚥̂ 𝑑𝜃 = − @ 2 cos 𝜃 − sin 𝜃 sin 𝜃 d𝜃 = @ sin3 𝜃 𝑑𝜃 = 𝜋
5 2 2 2
On the other hand,
𝜕𝐴' 𝜕𝐴1 𝜕𝐴0 𝜕𝐴' 𝜕𝐴1 𝜕𝐴0
∇×𝐴⃗ = − 𝚤̂ + − 𝚥̂ + − 𝑘> = −2𝑦𝑧 + 2𝑦𝑧 𝚤̂ + 0 − 0 𝚥̂ + 1 𝑘> = 𝑘>
𝜕𝑦 𝜕𝑧 𝜕𝑧 𝜕𝑥 𝜕𝑥 𝜕𝑦

Hence
⃗ ⋅ 𝑑 𝑆⃗ = 7 𝑑𝑆' = projected area of a hemisphere = area of a spher = 𝜋
7 (∇×𝐴)
# #

The Stokes’ theorem is valid.

Integral Form of Maxwell’s equations

In phase 1, we learn the Maxwells’ equations and they are represented in integral form/
Differen>al Form of Maxwell’s equa>ons

In fact, the Maxwell’s equations can be rewritten using the gradient operator ∇.

Let’s look at the Gauss’s law,

𝑄8+9 1
H 𝐸 ⋅ 𝑑 𝐴⃗ = = @𝜌𝑑𝑉
𝜖2 𝜖2
We apply the divergence theorem on the LHS,
1
H 𝐸 ⋅ 𝑑 𝐴⃗ = @∇ ⋅ 𝐸 𝑑𝑉 = @𝜌𝑑𝑉
𝜖2
𝜌
⇒∇⋅𝐸 =
𝜖2
Differen>al Form of Maxwell’s equa>ons

We can also rewrite the Ampere’s law,

dΦ: 𝜕𝐸
H 𝐵 ⋅ 𝑑 𝑙⃗ = 𝜇2 𝑖9 + 𝜇2 𝜖2 = 𝜇2 7(⃗𝚥9 +𝜖2 ) ⋅ 𝑑 𝐴⃗
𝑑𝑡 𝜕𝑡
where 𝚥⃗9 is the current density of the charge current. Apply Stokes’ theorem on the LHS, we get
𝜕𝐸
H 𝐵 ⋅ 𝑑 𝑙⃗ = 7(∇×𝐵) ⋅ 𝑑 𝐴⃗ = 𝜇2 7(⃗𝚥9 +𝜖2 ) ⋅ 𝑑 𝐴⃗
𝜕𝑡
𝜕𝐸
⇒ ∇×𝐵 = 𝜇2 𝚥⃗9 + 𝜇2 𝜖2
𝜕𝑡
Differential Form of Maxwell’s equations
To summarize,

𝜌
∇⋅𝐸 =
𝜖2
𝜕𝐵
∇×𝐸 = −
𝜕𝑡
∇⋅𝐵 =0
𝜕𝐸
∇×𝐵 = 𝜇2 𝚥⃗9 + 𝜇2 𝜖2
𝜕𝑡

For examples, we can expand the Faraday’s law and obtain,

 C S

where S is any surface spanning the loop C. This is called Ampère’s

law and it states that the line integral of B around a closed loop is
equal to µ0 times the total current Iencl flowing through the loop. It
enables the magnetic field to be calculated quickly in symmetrical
situations, as we shall illustrate.

Example 12.17
An infinitely long thin wire is aligned along the z-axis and carries
a current I. Find the form of the generated magnetic field B,
assuming that B → 0 infinitely far from the wire.

Solution
We have cylindrical symmetry about the z-axis, so that if we use
cylindrical polar co-ordinates, B must be independent of z and φ,
so that
B = Bρ (ρ)eρ + Bφ (ρ)eφ + Bz (ρ)ez , (12.78)
where eρ , eφ , ez are the unit vectors shown in Figure 11.13.
We next impose (12.75a) and (12.76) on (12.78) at r "= 0, where
j = 0. Using the forms of div and curl in cylindrical polar co-
ordinates given in Table 12.1, this gives
∂Bρ
∇·B = =0
∂ρ
and
∂Bz 1 ∂
∇ × B = −eφ + ez (ρBφ ) = 0,
∂ρ ρ ∂ρ
so that
∂Bz ∂Bρ ∂
= 0, = 0, (ρBφ ) = 0.
∂ρ ∂ρ ∂ρ
This can only be satisfied, subject to the boundary condition
B → 0 as ρ → ∞, by
Bz = 0, Bρ = 0, Bφ = k/ρ,

where k is a constant. This can now be determined by applying

Ampère’s law (12.77) to a circle in the x–y plane shown in Fig-
ure 12.15, giving ˛
B · dr = 2πρBφ (ρ) = µ0 I, Figure 12.15 The circuit C
C used to derive (12.79), where
so that, finally, the current I at the centre is
µ0 I directed out of the page.
B(r) = eφ . (12.79)
2πρ

www.pdfgrip.com