Professional Documents
Culture Documents
NE
Vectors
Sk Jahiruddin
ide
Assistant Professor
Sister Nibedita Govt. College, Kolkata
gu
Author was the topper of IIT Bombay M.Sc Physics 2009-2011 batch
He ranked 007 in IIT JAM 2009 and 008 (JRF) in CSIR NET June
2011
ics
1
©Sk Jahiruddin, 2020 Vectors
Contents
T
NE
1 Vector Algebra 7
1.1 Basics . . . . . . . . . . . . . . . . . . . . . 7
1.3
1.4
ide
Vector product or cross product . . . . . . .
17
gu
1.5 Vector triple product . . . . . . . . . . . . . 20
1.6.1 Line . . . . . . . . . . . . . . . . . . 21
1.6.2 Plane . . . . . . . . . . . . . . . . . . 23
ys
1.6.3 Sphere . . . . . . . . . . . . . . . . . 25
T
1.7 Reciprocal vectors . . . . . . . . . . . . . . . 32
NE
1.8 Exercises . . . . . . . . . . . . . . . . . . . . 34
1.8.2 Solutions . . . . . . . . . . . . . . . . 37
2 Vector Calculus
ide 43
gu
2.1 Derivative of a vector . . . . . . . . . . . . . 43
2.5.1 Gradient . . . . . . . . . . . . . . . . 58
T
2.5.2 Divergence of a vector field . . . . . . 68
NE
2.5.3 Curl of a vector field . . . . . . . . . 70
2.7
ide
Spherical polar coordinates . . . . . . . . . . 82
2.8.2 Divergence . . . . . . . . . . . . . . . 90
ics
2.8.3 Laplacian . . . . . . . . . . . . . . . 91
ys
2.8.4 Curl . . . . . . . . . . . . . . . . . . 91
2.9 Exersizes . . . . . . . . . . . . . . . . . . . . 93
Ph
2.9.2 Solutions . . . . . . . . . . . . . . . . 98
T
3.1 Line Integral . . . . . . . . . . . . . . . . . . 116
NE
3.2 Green’s theorem in a plane . . . . . . . . . . 125
3.5
ide
Volume integrals . . . . . . . . . . . . . . . 145
T
3.8 Exercises . . . . . . . . . . . . . . . . . . . . 166
NE
3.8.1 Ans Keys . . . . . . . . . . . . . . . 172
4.1
ide
Ans key of More Exercises . . . . . . . . . . 202
197
gu
ics
ys
Ph
1 Vector Algebra
T
NE
1.1 Basics
ide
gu
ics
ys
T
NE
ide
Figure 1.2: Subtraction of vectors
(λ + µ)a = λa + µa
a = a1 e1 + a2 e2 + a3 e3
The three vectors e1 , e2 and e3 are said to form a basis
(for the three-dimensional space); the scalars a1 , a2 and a3 ,
T
ponents of the vector a with respect to this basis. We say
that the vector has been resolved into components.
NE
Most often we shall use basis vectors that are mutually
perpendicular, for ease of manipulation, though this is not
necessary. In general, a basis set must (i) have as many
basis vectors as the number of dimensions (in more formal
ide
language, the basis vectors must span the space) and (ii)
be such that no basis vector may be described as a sum
of the others, or, more formally, the basis vectors must be
linearly independent. Putting this mathematically, in N
gu
dimensions, we require
ics
c1 e1 + c2 e2 + · · · + cN eN 6= 0
for any set of coefficients c1 , c2 , . . . , cN except c1 = c2 =
· · · = cN = 0
ys
T
a = ax i + ay j + az k
NE
A vector in three-dimensional space thus requires three com-
ponents to describe fully both its direction and its magni-
tude. A displacement in space may be thought of as the
sum of displacements along the x − y− and z− directions .
For brevity, the components of a vector a with respect to a
ide
particular coordinate system are sometimes written in the
form (ax , ay , az ) . Note that the basis vectors i, j and k may
themselves be represented by (1, 0, 0), (0, 1, 0) and (0, 0, 1)
gu
respectively.
ics
ys
Ph
T
b is found by simply adding their components, i.e.
NE
a + b = ax i + ay j + az k + bx i + by j + bz k
= (ax + bx ) i + (ay + by ) j + (az + bz ) k
a − b = ax i + ay j + az k − (bx i + by j + bz k)
ide
= (ax − bx ) i + (ay − by ) j + (az − bz ) k
q
a ≡ |a| = a2x + a2y + a2z (1.2)
ys
a
â = (1.3)
|a|
T
NE
The scalar product (or dot product) of two vectors a and b
is denoted by a · b and is given by
i·i=j·j=k·k=1
Ph
i·j=j·k=k·i=0
T
scalar product is commutative and distributive over addi-
tion:
NE
a·b=b·a
a · (b + c) = a · b + a · c
·b = 1 × 2 + 2 × 3 + 3 × 4 = 20
√ √
|b| = 22 + 32 + 42 = 29
Thus,
Ph
20
cos θ = √ √ ≈ 0.9926 ⇒ θ = 0.12rad.
14 29
T
then clearly θ = 0 and we have
NE
a · a = |a|2
1.3
ide
Vector product or cross product
|a × b| = |a||b| sin θ
T
NE
ide
Figure 1.4: The vector product. The vectors a, b and a × b
form a right-handed set
gu
The vector product is distributive over addition, but an-
ticommutative and non-associative
ics
(a + b) × c = (a × c) + (b × c)
b × a = −(a × b) (1.5)
(a × b) × c 6= a × (b × c)
ys
If a 6= 0 and b 6= 0
Ph
a×a=0 (1.7)
T
finding the area, A, of a parallelogram with sides a and b,
using the formula
NE
A = |a × b|
and
ide
i×i=j×j=k×k=0
i × j = −j × i = k
j × k = −k × j = i
gu
k × i = −i × k = j
T
sides a = i + 2j + 3k and b = 4i + 5j + 6k
NE
Solution: The vector product a × b is given in compo-
nent form by
√
gu
p
A = |a × b| = (−3)2 + 62 + (−3)2 = 54
ics
[a, b, c] ≡ a · (b × c) (1.8)
Ph
T
area of the base. Further, v · c = vc cos φ.
NE
ide
Figure 1.5: The scalar triple product represents the volume
of a parallelepiped
gu
product is
T
a ay az
x
NE
a · (b × c) = bx by bz
cx cy cz
T
V = |a · (b × c)| = |(a × b) · c|
= |(−3i + 6j − 3k) · (7i + 8j + 10k)|
NE
= |(−3)(7) + (6)(8) + (−3)(10)| = 3
ide
(a × b) · (c × d) ≡ (a · c)(b · d) − (a · d)(b · c) (1.11)
T
form . It can also be shown that for any three vectors a, b, c
NE
a × (b × c) + b × (c × a) + c × (a × b) = 0
ide
Now that we have described the basic algebra of vectors, we
can apply the results to a variety of problems, the first of
which is to find the equation of a line in vector form.
gu
1.6.1 Line
ics
r = a + λb (1.12)
Ph
T
NE
ide
Figure 1.6: The equation of a line. The vector b is in the
direction AR and λb is the vector from A to R
gu
Taking the vector product of above equation with b and
remembering that b × b = 0 gives an alternative equation
for the line
ics
(r − a) × b = 0 (1.13)
r = a + λ(c − a)
T
x − ax y − ay z − az
= = = constant
NE
bx by bz
1.6.2 Plane
(r − a) · n̂ = 0 (1.14)
gu
ics
ys
Ph
T
R with position vector r is r − a; r will lie in the plane if
this vector is perpendicular to the normal to the plane.
NE
Rewriting equation (1.14) as r · n̂ = a · n̂, we see that
the equation of the plane may also be expressed in the form
r · n̂ = d, or in component form as
ide
lx + my + nz = d
T
r = αa + βb + γc
NE
where α + β + γ = 1
p = n1 × n2
ics
1.6.3 Sphere
Ph
T
|r − c|2 = (r − c) · (r − c) = a2 (1.16)
NE
where c is the position vector of the centre of the sphere
and a is its radius.
1.6.4 ide
Distance from a point to a line
T
The position vector of P is p = i + 2j + k and we find
NE
1
(p − a) × b̂ = √ [j × (2i − 3j + 3k)]
14
1
= √ (3i − 2k)
ide14
d = (a − p) · n̂ (1.18)
T
coordinates (1, 2, 3) to the plane that contains the points
A, B and C having coordinates (0, 1, 0), (2, 3, 1) and (5, 7, 2).
NE
Solution: Let us denote the position vectors of the
points A, B, C by a, b, c. Two vectors in the plane are
b − a = 2i + 2j + k and c − a = 5i + 6j + 2k
ide
and hence a vector normal to the plane is
gu
n = (2i + 2j + k) × (5i + 6j + 2k) = −2i + j + 2k
n 1
n̂ = = (−2i + j + 2k)
|n| 3
ys
d = (a − p) · n̂
1
= (−i − j − 3k) · (−2i + j + 2k)
3
2 1 5
= − −2=−
3 3 3
jahir@physicsguide.in 28 physicsguide CSIR NET, GATE
©Sk Jahiruddin, 2020 Vectors
T
a positive quantity. It follows from this that the original
point P with coordinates (1, 2, 3), for which d was negative,
NE
is on the opposite side of the plane from the origin.
ide
Consider two lines in the directions a and b, , since a × b is
by definition perpendicular to both a and b, the unit vector
normal to both these lines is
gu
a×b
n̂ =
|a × b|
ics
T
passes through the points (1, 2, 4) and (0, 0, 1). Find the
minimum distance between the two lines.
NE
Solution: The first line is given by
r1 = λ(i + j + k)
A vector between the two lines is, for example, the one con-
necting the points (0, 0, 0) and (0, 0, 1), which is simply k.
Thus it follows that the minimum distance between the two
Ph
lines is
1 1
d = √ |k · (i − 2j + k)| = √
6 6
T
NE
Let us consider the line r = a + λb. This line will intersect
any plane to which it is not parallel. Thus, if a plane has a
normal n̂ then the minimum distance from the line to the
plane is zero unless
b · n̂ = 0
ide
in which case the distance, d, will be
d = |(a − r) · n̂|
gu
where r is any point in the plane.
x + 2y + 3z = 6
ys
n = i + 2j + 3k
T
point on the line into the equation of the plane, obtaining
NE
1 + 4λ + 2(2 + 5λ) + 3(3 + 6λ) = 6 ⇒ 14 + 32λ = 6
a · a0 = b · b0 = c · c0 = 1 (1.20)
and
ys
a0 · b = a0 · c = b0 · a = b0 · c = c0 · a = c0 · b = 0 (1.21)
Ph
T
are given by
b×c
a0 =
NE
a · (b × c)
c×a
b0 = (1.22)
a · (b × c)
a×b
c0 =
a · (b × c)
ide
where a · (b × c) 6= 0. In other words, reciprocal vectors only
exist if a, b and c are not coplanar. Moreover, if a, b and
c are mutually orthogonal unit vectors then a0 = a, b0 = b
and c0 = c, so that the two systems of vectors are identical.
gu
Example: Construct the reciprocal vectors of a = 2i, b =
j + k, c = i + k
ics
= 2i · (i + j − k) = 2
Now find the reciprocal vectors by using the definition.
1 1
Ph
a0 = (j + k) × (i + k) = (i + j − k)
2 2
1
b0 = (i + k) × 2i = j
2
1
c0 = (2i) × (j + k) = −j + k
2
jahir@physicsguide.in 33 physicsguide CSIR NET, GATE
©Sk Jahiruddin, 2020 Vectors
1.8 Exercises
T
1.1. Which of the following statements about general vec-
NE
tors a, b and c are true?
(a) c · (a × b) = (b × a) · c
(b) a × (b × c) = (a × b) × c
(c) a × (b × c) = (a · c)b − (a · b)c
(d) d = λa + µb implies (a × b) · d = 0
ide
(e) a × c = b × c implies c · a − c · b = c|a − b|
(f) (a × b) × (c × b) = b[b · (c × a)]
1.2. Find the angle between the position vectors to the
gu
points (3, −4, 0) and (−2, 1, 0) and find the direction cosines
of a vector perpendicular to both.
1.3. Prove
ics
(a × b) × c = (a · c)b − (b · c)a
1.4. (a) Find the equation of the plane through the three
ys
1.6. Find the distance from P (1, 2, −1) to the line joining
T
P1 (0, 0, 0) and P2 (−1, 0, 2)
NE
1.7. Find the distance between the lines r = i − 2j + (i − k)t
and r = 2j − k + (j − i)t
1.8. (a) Find the direction of the line of intersection of the
planes x − 2y + 3z = 4 and 2x + y − z = 5.
(b) Find the cosine of the angle between the planes
ide
gu
ics
ys
Ph
T
NE
1.1. (c), (d) and (e)
−2
1.2. cos θ = √ 5
1.4. (a) 3x − 4y + z + 6 = 0, (b) r = (1, 0, −2) + (3, −4, 1)t
√
1.5. 11/ 14
p
1.6. 21/5
√
1.7.
1.8.
2/ 3 ide p
(a) −i + 7j + 5k (b) arccos 3/28
gu
ics
ys
Ph
1.8.2 Solutions
T
NE
Solution: 1.1.
(a) c · (a × b) = −c · (b × a) = −(b × a) · c 6= (b × a) · c
(e) a × c = b × c ⇒ (a − b) × c = 0 ⇒ a − bkc
⇒ (a − b) · c = c|a − b| ⇒ c · a − c · b = c|a − b|
ys
T
a = (3, −4, 0) and b = (−2, 1, 0) then its cosine is given by
NE
a · b −6 − 4 + 0 −2
cos θ = = √ =√
ab 5 5 5
giving θ = 153.4◦ A vector perpendicular to both a and b
is their cross product
ide
a × b = (0 − 0, 0 − 0, 3 − 8) = (0, 0, −5)
a × b = (ay bz − az by , az bx − ax bz , ax by − ay bx )
T
= bx (az cz + ay cy ) − ax (bz cz + by cy )
NE
= bx (az cz + ay cy + ax cx ) − ax (bx cx + bz cz + by cy )
= [(a · cz ) b − (b · cx )]x
we both added and subtracted ax bx cx on the RHS. This
establishes the result for the x -component and hence for all
three components.
ide
Solution: 1.4. (a) A vector joining any pair of the
−→
given points lies in the plane. Two such vectors are AB =
−→
(2, 3, 0) − (−1, 1, 1) = (3, 2, −1) and AC = (1, 0, −3). The
gu
cross product of these two vectors is perpendicular to the
plane. This is
i j k
ics
−→ −→
N = (AB) × (AC) = 3 2 −1 = −6i + 8j − 2k
1 0 −3
ys
−6(x − 2) + 8(y − 3) − 2z = 0 or 3x − 4y + z + 6 = 0
T
(x − 1) y (z + 2)
= =
NE
3 −4 1
The parametric equations of the line are r = i − 2k + (3i −
4j + k)t or, if you like, r = (1, 0, −2) + (3, −4, 1)t
−→
ide
this point Q. Then the vector from P to Q is
N = 3i − 2j + k
√
ics
vector along the line. Then a unit vector along the line is
√
u = (1/ 5)(−i + 2k). Let us take Q to be P1 (0, 0, 0). Then
−→
P Q = −i − 2j + k, so we get for the distance |P R| :
T
1
|P R| = √ |(−i − 2j + k) × (−i + 2k)|
NE
5
1 p
= √ | − 4i + j − 2k| = 21/5
5
√ √
= | − 1 + 4 − 1|/ 3 = 2/ 3
Solution: 1.8.
T
pendicular to the two normal vectors to the planes, namely
i − 2j + 3k and 2i + j − k. Then the direction of the line
NE
is that of the cross product of these normal vectors; this is
−i + 7j + 5k
2 Vector Calculus
T
NE
2.1 Derivative of a vector
ide
gu
ics
ys
T
a(u) = ax i + ay j + az k
NE
is given by
da dax day daz
= i+ j+ k
du du du du
The first application is in velocity and acceleration of a vec-
tor
dr dx dy dz
and
v(t) =
dt
=
dv d2 x
ide
dt
i+ j+ k
dt
d2 y
dt
d2 z
(2.2)
are given by
v(t) = dr
dt = 4ti + 3j + 6tk
Ph
a(t) = dv
dt = 4i + 6k
T
of t), and its component in the direction s is given by
√
NE
(4i + 6k) · (i + 2j + k) 5 6
a · ŝ = √ =
12 + 2 2 + 1 2 3
T
NE
ide
Figure 2.2: Unit basis vectors for two-dimensional Cartesian
and plane polar coordinates.
gu
ics
of the basis vectors êρ and êφ with respect to t are given by
dêρ dφ dφ
Ph
T
d d
v(t) = r = (ρê) = ρ̇êρ + ρê˙ ρ = ρ̇êρ + ρφ̇êφ (2.6)
NE
dt dt
The acceleration
d
a(t) = ρ̇êρ + ρφ̇êφ
dt
˙ide
= ρ̈êρ + ρ̇ρ + ρφ̂φ + ρφ̈êφ + ρ̇φ̇êφ
(2.7)
= ρ̈êρ + ρ̇ φ̇êφ + ρφ̇ −φ̇êρ + ρφ̈êφ + ρ̇φ̇êφ
2
= ρ̈ − ρφ̇ êρ + (ρφ̈ + 2ρ̇φ̇)êφ
gu
d da dφ
(φa) = φ + a
du du du
d db da
Ph
(a · b) = a · + ·b
du du du
d db da
(a × b) = a × + ×b
du du du
T
NE
One further useful result can be derived by considering the
derivative
d da
(a · a) = 2a ·
du du
since a · a = a2 , where a = |a|, we see that
a·
da
du
ide
= 0 if a is constant (2.8)
T
NE
ide
gu
T
F (x, y, z) = 0 G(x, y, z) = 0, where each equation repre-
sents a surface and the curve is the intersection of the two
NE
surfaces.
2
ds dr dr
= ·
du du du
T
Z u2 r
dr dr
s= · du (2.9)
NE
u1 du du
Rbp ide
along the curve between x = a and x = b is given by
s = a 1 + y 02 dx, where v 0 = dv/dx
r(u) = ui + y(u)j
ics
du du
from which we obtain
Ph
2
dr dr dy
· =1+
du du du
T
s 2
Z br Z b
dr dr dy
NE
s= · du = 1+ dx (2.10)
a du du a dx
ide
dr/du is a vector tangent to C at that point, in the direction
of increasing u. In the special case where the parameter u is
the arc length s along the curve then dr/ds is a unit tangent
vector to C and is denoted by t̂
gu
t̂ = dr/ds
ics
2
dt̂ d r̂
κ = = 2
ds ds
Ph
T
therefore have
dt̂
NE
= κn̂ (2.11)
ds
The unit vector b̂ = t̂ × n̂, which is perpendicular to the
plane containing t̂ and n̂, is called the binormal to C. The
vectors t̂, and b̂ form a right-handed rectangular cooordi-
nate system at any given point on C . As s changes so that
ide
the point of interest moves along C, the triad of vectors also
changes.
d db̂ dt̂
0= (b̂ · t̂) = · t̂ + b̂ ·
ds ds ds
db̂
= · t̂ + b̂ · κn̂
Ph
ds
db̂
= · t̂
ds
where we have used the fact that b̂ · n̂ = 0. Hence, since
T
db̂/ds ∝ n̂. The constant of proportionality is −τ
NE
so we finally obtain
db̂
= −τ n̂ (2.12)
ds
Taking the dot product of each side with n̂, we see that
ide
the torsion of a curve is given by
db̂
τ = −n̂ ·
ds
gu
We may also define the quantity σ = 1/τ, which is called
the radius of torsion.
ics
(2.13)
= −τ n̂ × t̂ + b̂ × κn̂
= τ b̂ − κt̂
T
proved. These are called the Frenet-Serret formulae,
NE
dt̂ dn̂ db̂
= κn̂, = τ b̂ − κt̂, = −τ n̂ (2.14)
ds ds ds
ide
Example: Show that the acceleration of a particle
traveling along a trajectory r(t) is given by
dv v2
a(t) = t̂ + n̂
gu
dt ρ
radius of curvature.
dr dr ds ds
v(t) = = = t̂
dt ds dt dt
Ph
T
dv dv dt̂
a(t) = = t̂ + v
NE
dt dt dt
but we note that
dt̂ ds dt̂ v
= = vκn̂ = n̂
dt dt ds ρ
Therefore, we have ide dv v2
a(t) = t̂ + n̂
dt ρ
gu
This shows that in addition to an acceleration dv/dt along
the tangent to the particle’s trajectory, there is also an ac-
celeration v 2 /ρ in the direction of the principal normal. The
ics
T
ation as
NE
2
d2 r du dr d2 u
dv d dr du
a= = = 2 +
dt dt du dt du dt du dt2
Z t p
s= |dR/dt|dt = (a2 + b2 ) t
0
ys
ds p
= (a2 + b2 )
dt
Ph
T
dR dR ds −a sin tI + u cos tJ + bK
T= = / =
NE
p
ds dt dt (a2 + b2 )
dT dT ds −a(cos tI + sin tJ)
= / =
ds dt
dt a2 + b 2
dT a
k = = 2 and N = −(cos tI + sin tJ)
ds u + b2
p
dB dB ds
ds
=
dt dt
ide
B = T × N = (b sin tI − b cos tJ + aK)/ (a2 + b2 )
b
τ= 2
gu
a + b2
2.5.1 Gradient
ys
∂x ∂y ∂z
T
field ∇φ should not be confused with the vector operator
φ∇, which has components (φ∂/∂x, φ∂/∂y, φ∂/∂z)
NE
Example: Find the gradient of the scalar field φ =
xy 2 z 3
∂φ ∂φ ∂φ
= dx + dy + dz
∂x ∂y ∂z
Ph
= dφ
T
dφ dr
= ∇φ ·
NE
du du
In the particular case where the parameter u is the arc
length s along the curve, the total derivative of φ with re-
spect to s along the curve is given by
dφ
ds
ide
= ∇φ · t̂
= ∇φ · â
ds
and is called the directional derivative. since â is a unit
vector we have
ys
dφ
= |∇φ| cos θ
ds
where θ is the angle between â and ∇φ as shown in figure
Ph
T
NE
ide
gu
(1, 2, −1), find its rate of change with distance in the di-
rection a = i + 2j + 3k. At this same point is the greatest
possible rate of change with distance and in which direction
Ph
does it occur?
T
∇φ = 2xyi + x2 + z j + yk
NE
The unit vector in the direction of a is
1
â = √ (i + 2j + 3k)
14
dφ
ide
so the rate of change of φ with distance s in this direction
is
1 10
= ∇φ · â = √ (4 + 6) = √
ds 14 14
gu
From the above discussion, at the point (1, 2, −1)dφ/ds will
be greatest in the direction of ∇φ = 4i + 2k and has the
√
value |∇φ| = 20 in this direction.
ics
T
field in moving from r to r + dr given in (dE = ∂E
∂x dx +
∂E ∂E
∂y dy + ∂z dz) as dE = (dr · ∇)E
NE
A second interesting geometrical property of ∇φ may be
found by considering the surface defined by φ(x, y, z) = c,
where c is some constant. If t̂ is a unit tangent to this surface
at some point then clearly dφ/ds = 0 in this direction and
from ( dφ ide
ds = ∇φ· t̂) we have ∇φ· t̂ = 0. In other words, ∇φ is
a vector normal to the surface φ(x, y, z) = c at every point,
as shown in figure 2.4. If n̂ is a unit normal to the surface
in the direction of increasing φ(x, y, z), then the gradient is
gu
sometimes written
∂φ
∇φ ≡ n̂
∂n
ics
origin,
T
(r − r0 ) · n0 = 0
NE
Similarly, if r is the position vector of any point on the
straight line passing through P (with position vector r0 ) in
the direction of the normal n0 then the vector equation of
this line is, from vector equation of a line
ide
(r − r0 ) × n0 = 0
at this point is
(r − r0 ) · 2ak = 0
This gives 2a(z−a) = 0 or z = a, as expected. The equation
ys
(r − r0 ) × 2ak = 0
Ph
T
NE
ide
gu
Figure 2.5: The tangent plane and the normal to the sur-
face of the sphere φ = x2 + y 2 + z 2 = a2 at the point r0 with
ics
coordinates (0, 0, a)
So in summary
ys
∂φ ∂φ ∂φ
∇φ = grad φ = i +j +k (2.16)
∂x ∂y ∂z
Ph
dφ
= ∇φ·u (directional derivative in the direction of u)
ds
(2.17)
The vector ∇φ is perpendicular to the surface φ = const
(2.18)
T
at (1, 2, −1) in the direction
NE
A = 2i − 2j + k
ide
1
u = (2i − 2j + k)
3
You get
∂φ ∂φ ∂φ
gu
∇φ =i +j +k = (2xy + z)i + x2 j + xk
∂x ∂y ∂z
∇φ at the point (1, 2, −1) = 3i + j + k
ics
T = x2 − y 2 + xyz + 273
Solution:
T
Here
NE
∇T = (2x + yz)i + (−2y + xz)j + xyk = 4i − 7j − 2k
at (−1, 2, 3)
to cold
Solution:
Ph
T
the equation of the tangent plane is
NE
9(x − 1) − 3(y + 2) + (z − 3) = 0
2.5.2
ide
Divergence of a vector field
gu
The divergence of a vector field a(x, y, z) is defined by
∂ax ∂ay ∂az
div a = ∇ · a = + + (2.19)
∂x ∂y ∂z
ics
x y i + y 2 z 2 j + x2 z 2 k
2 2
Solution:
T
of how much a vector field diverges (spreads out) or con-
verges at any given point. For example, if we consider the
NE
vector field v(x, y, z) describing the local velocity at any
point in a fluid then ∇ · v is equal to the net rate of outflow
of fluid per unit volume, evaluated at a point.
Solution:
∂ 2φ ∂ 2φ ∂ 2φ
∇2 φ = 2
+ 2 + 2 = 2xz 3 + 6xy 2 z .
∂x ∂y ∂z
jahir@physicsguide.in 69 physicsguide CSIR NET, GATE
©Sk Jahiruddin, 2020 Vectors
T
NE
The curl of a vector field a(x, y, z) is defined by
∂az ∂ay ∂ax ∂az
curl a = ∇ × a = − i+ − j
∂y ∂z ∂z ∂x
(2.21)
∂ay ∂ax
+ − k
∂x ∂y
ide
where ax , ay and az are the x− , y− and z - components of
a. The RHS can be written in a more memorable form as a
determinant:
gu
i j k
∂ ∂ ∂
∇ × a = ∂x ∂y ∂z (2.22)
ics
ax ay az
irrotational.
x2 y 2 z 2 i + y 2 z 2 j + x2 z 2 k
T
i j k
NE
∂ ∂ ∂
∇ × a = ∂x
∂y
2 2 2 2 2 2 2 ∂z
xy z y z xz
= −2 y 2 zi + xz 2 − x2 y 2 z j + x2 yz 2 k
i j k
∂ ∂ ∂
∇ × v = ∂x ∂y ∂z = 2ωk = 2ω
Ph
−ωy ωx 0
T
NE
2.5.4 Vector operator formula
Example:Prove
gu
∇ × (φa) = ∇φ × a + φ∇ × a
ics
∂az ∂ay ∂φ ∂φ
=φ − + az − ay
∂y ∂z ∂y ∂z
= φ(∇ × a)x + (∇φ × ax
Ph
T
Example: Prove that curl of grad of any scalar func-
tion is ZERO. That is
NE
∇ × (∇φ) = 0 (2.23)
Solution:
i j k
∇ × (∇φ) = ∇ ×
ide
∂φ
∂x
i+
∂φ
∂y
j+
∂φ
∂z
k
∂
= ∂x
∂φ
∂x
∂
∂y
∂φ
∂y
∂
∂z
∂φ
∂z
∂ ∂φ ∂ ∂φ
= − i
gu
∂y ∂z ∂z ∂y
∂ ∂φ ∂ ∂φ
+ − j
∂z ∂x ∂x ∂z
∂ ∂φ ∂ ∂φ
ics
+ − k
∂x ∂y ∂y ∂x
2
∂ 2φ
2
∂ 2φ
∂ φ ∂ φ
= − i+ − j
ys
T
Example: Prove that divergence of curl of any vector
function is also ZERO.
NE
∇ · (∇ × A) = 0 (2.24)
Solution:
i j k
∂
∇ · (∇ × A) = ∇ · ∂x
∂
ide ∂
∂y
Ax A2 ∂z
1
∂
∂z
∂
A1 A2 A3
gu
∂A3 ∂A2 ∂A1 ∂A3
=∇· − i+ − j
∂y ∂z ∂z ∂x
∂A2 ∂A1
+ − k
ics
∂x ∂y
∂ ∂A3 ∂A2 ∂ ∂A1 ∂A3
= − + −
∂x ∂y ∂z ∂y ∂z ∂x
ys
∂ ∂A2 ∂A1
+ −
∂z ∂x ∂y
∂ 2 A3 ∂ 2 A2 ∂ 2 A1 ∂ 2 A3
Ph
= − + −
∂x∂y ∂x∂z ∂y∂z ∂y∂x
∂ 2 A2 ∂ 2 A1
+ − =0
∂z∂x ∂z∂y
T
Example: Prove that
NE
Solution: We will prove for x component only as the
prove is little lengthy.
i j k
ide
∂ ∂ ∂
∇ × (∇ × A) = ∇ × ∂x ∂y ∂z
A1 A2 A3
∂A3 ∂A2 ∂A1 ∂A3 ∂A2 ∂A1
=∇× − i+ − j+ − k
∂y ∂z ∂z ∂x ∂x ∂y
gu
The x component is
∂ ∂A2 ∂A1 ∂ ∂A1 ∂A3
− − −
ics
∂y ∂x ∂y ∂z ∂z ∂x
∂ 2 A1 ∂ 2 A1 ∂ 2 A2 ∂ 2 A3
=− 2 − + +
∂y ∂z 2 ∂y∂x ∂z∂x
ys
∂2 ∂2 ∂2
∂ ∂A1 ∂A2 ∂A3
=− + + A1 + + +
∂x2 ∂y 2 ∂z 2 ∂x ∂x ∂y ∂z
jahir@physicsguide.in 75 physicsguide CSIR NET, GATE
©Sk Jahiruddin, 2020 Vectors
T
−∇2 A + ∇(∇ · A)
NE
You could have proved this using the triple product
A × (B × C) = B(A · C) − (A · B)C
∇ · (φa) = φ∇ · a + a · ∇φ (2.26)
ics
∇ × (φa) = ∇φ × a + φ∇ × a (2.27)
∇(a·b) = a×(∇×b)+b×(∇×a)+(a·∇)b+(b·∇)a (2.28)
ys
∇ · (a × b) = b · (∇ × a) − a · (∇ × b) (2.29)
∇×(a×b) = a(∇·b)−b(∇·a)+(b·∇)a−(a·∇)b (2.30)
Ph
∇ · (∇φ × ∇ψ) = 0
T
Solution: We already know
NE
∇ · (a × b) = b · (∇ × a) − a · (∇ × b)
T
and z respectively then we obtain the three vectors
NE
∂r
eρ = = cos φi + sin φj (2.33)
∂ρ
∂r
eφ = = −ρ sin φi + ρ cos φj (2.34)
∂φ
∂r
ez = =k (2.35)
ide
∂z
These vectors lie in the directions of increasing ρ, φ and z
respectively but are not all of unit length. We will make
them unit vectors by dividing each vector by its modulus
gu
êρ = eρ = cos φi + sin φj (2.36)
1
ics
length .
T
i = cos φêρ − sin φêφ
NE
j = sin φêρ + cos φêφ (2.39)
k = êz
ide
gu
ics
ys
T
∂r ∂r ∂r
dr = dρ + dφ + dz
NE
∂ρ ∂φ ∂z
= dρeρ + dφeφ + dzez (2.40)
T
ds2 = dx2 + dy 2 + dz 2 (2.42)
NE
We can also find the volume element in a cylindrical polar
system by calculating the volume of the infinitesimal par-
allelepiped defined by the vectors dρêρ , ρdφêφ and dzêz as
shown in the figure below
ide
dV = |dρêρ · (ρdφêφ × dzêz )| = ρdρdφdz (2.43)
gu
ics
ys
Ph
T
are as follows. These can be proved using the generalized
orthogonal curvilinear coordinate which will be discussed
NE
later.
∂Φ 1 ∂Φ ∂Φ
∇Φ = êρ + êφ + êz (2.44)
∂ρ ρ ∂φ ∂z
1 ∂
∇·a=
ide
(ρaρ ) +
1 ∂aφ ∂az
+ (2.45)
ρ ∂ρ ρ ∂φ ∂z
êρ ρêφ êz
gu
1 ∂ ∂ ∂
∇ × a = ∂ρ ∂φ ∂z (2.46)
ρ
aρ ρaφ az
1 ∂ 2Φ ∂ 2Φ
ics
2 1 ∂ ∂Φ
∇ Φ= ρ + 2 2 + 2 (2.47)
ρ ∂ρ ∂ρ ρ ∂φ ∂z
ys
T
of P can be written as
NE
r = r sin θ cos φi + r sin θ sin φj + r cos θk (2.49)
∂r ∂r
êθ = / = cos θ cos φi + cos θ sin φj − sin θk (2.51)
∂θ ∂θ
ys
∂r ∂r
êφ = / = − sin φi + cos φj (2.52)
Ph
∂φ ∂φ
T
NE
ide
gu
T
dV = |drêr · (rdθêθ × r sin θdφêφ )| = r2 sin θdrdθdφ
NE
(2.55)
ide
gu
ics
ys
Gradient
∂Φ 1 ∂Φ 1 ∂Φ
∇Φ = êr + êθ + êφ (2.56)
∂r r ∂θ r sin θ ∂φ
jahir@physicsguide.in 85 physicsguide CSIR NET, GATE
©Sk Jahiruddin, 2020 Vectors
Divergence
T
1 ∂ 2 1 ∂ 1 ∂aφ
∇·a = r ar + (sin θaθ )+ (2.57)
NE
2
r ∂r r sin θ ∂θ r sin θ ∂φ
Curl
êr rêθ r sin θêφ
1 ∂ ∂ ∂
∇×a= 2 (2.58)
∂r ∂θ ∂φ
r sin θ
Laplacian
ide
ar raθ r sin θaφ
∂ 2Φ
2 1 ∂ 2 ∂Φ 1 ∂Φ∂ 1
∇ Φ= 2 r + sin θ + 2 2
gu
r ∂r ∂r r2 sin θ ∂θ
∂θ r sin θ ∂φ2
(2.59)
we can rewrite the first term on the RHS of Laplacian as
ics
follows:
1 ∂2
1 ∂ 2 ∂Φ
r = (rΦ)
r2 ∂r ∂r r ∂r2
which can often be useful in shortening calculations.
ys
T
dinates x, y, z may be expressed in terms of the three curvi-
linear coordinates u1 , u2 , u3 , where
NE
x = x (u1 , u2 , u3 ) , y = y (u1 , u2 , u3 ) , z = z (u1 , u2 , u3 )
(2.60)
and inversely
T
for cylindrical and spherical polar coordinates were
NE
for cylindrical polars hρ = 1, hφ = ρ, hz = 1
for spherical polars hr = 1, hθ = r, hφ = r sin θ
(2.64)
dr =
∂r
ide
du1 +
∂r
du2 +
∂r
du3
∂u1 ∂u2 ∂u3
= du1 e1 + du2 e2 + du3 e3 (2.65)
gu
= h1 du1 ê1 + h2 du2 ê2 + h3 du3 ê3
T
dV = |du1 e1 · (du2 e2 × du3 e3 )|
= |h1 ê1 · (h2 ê2 × h3 ê3 )| du1 du2 du3 (2.67)
NE
= h1 h2 h3 du1 du2 du3
2.8.1 Gradient
ide
The change dΦ in a scalar field Φ resulting from changes
du1 , du2 , du3 in the coordinates u1 , u2 , u3 is given by,
∂Φ ∂Φ ∂Φ
dΦ = du1 + du2 + du3 (2.68)
gu
∂u1 ∂u2 ∂u3
For orthogonal curvilinear coordinates u1 , u2 , u3 we can write
this as
ics
dΦ = ∇Φ · dr (2.69)
where ∇Φ is given by
ys
1 ∂Φ 1 ∂Φ 1 ∂Φ
∇Φ = ê1 + ê2 + ê3 (2.70)
h1 ∂u1 h2 ∂u2 h3 ∂u3
This implies that the del operator can be written
Ph
2.8.2 Divergence
T
NE
A vector in terms of the basis vectors
The divergence is
1 ∂ ∂ ∂
∇·a =
h1 h2 h3 ∂u1 ide
(h2 h3 a1 ) +
∂u2
(h3 h1 a2 ) +
∂u3
(h1 h2 a3 )
(2.72)
We can prove this by following logic Let us consider the sub-
expression ∇ · (a1 ê1 ) . Now ê1 = ê2 × ê3 = h2 ∇u2 × h3 ∇u3 .
gu
Therefore
∇ · (a1 ê1 ) = ∇ · (a1 h2 h3 ∇u2 × ∇u3 )
ics
ê2 ê3 ê1
∇ · (a1 ê1 ) = ∇ (a1 h2 h3 ) · × = ∇ (a1 h2 h3 ) ·
h2 h3 h2 h3
Ph
letting Φ = a1 h2 h3 in ( ∇Φ = h11 ∂u
∂Φ
1
ê1 + h12 ∂u
∂Φ
2
ê2 + 1 ∂Φ
h3 ∂u3 ê3 )
and substituting into the above equation, we find
1 ∂
∇ · (a1 ê1 ) = (a1 h2 h3 )
h1 h2 h3 ∂u1
jahir@physicsguide.in 90 physicsguide CSIR NET, GATE
©Sk Jahiruddin, 2020 Vectors
Repeating the analysis for ∇ · (a2 ê2 ) and ∇ · (a3 ê3 ) , and
T
adding the results we get the final result.
NE
2.8.3 Laplacian
1 ∂ h2 h 3 ∂Φ ∂ h3 h1 ∂Φ
∇2 Φ =
gu
+
h1 h2 h3 ∂u1 h1 ∂u1 ∂u2 h2 ∂u2
∂ h1 h2 ∂Φ
+
∂u3 h3 ∂u3
ics
(2.73)
2.8.4 Curl
ys
The curl of a vector field a = a1 ê1 +a2 ê2 +a3 ê3 in orthogonal
Ph
T
expression ∇ × (a1 ê1 ) . since ê1 = h1 ∇u1 we have
NE
∇ × (a1 ê1 ) = ∇ × (a1 h1 ∇u1 )
= ∇ (a1 h1 ) × ∇u1 + a1 h1 ∇ × ∇u1
Letting Φ = a1 h1 in ( ∇Φ = h11 ∂u
∂Φ
1
ê1 + h12 ∂u
∂Φ
2
ê2 + h13 ∂u
∂Φ
3
ê3 )
gu
and substituting into the above equation, we find
ê2 ∂ ê3 ∂
∇ × (a1 ê1 ) = (a1 h1 ) − (a1 h1 )
h3 h1 ∂u3 h1 h2 ∂u2
ics
Similarly find ∇ × (a2 ê2 ) . you get terms in ê3 and ê1 ,. Find
∇ × (a3 ê3 ) and you get terms in ê1 and ê2 .
ys
2.9 Exersizes
T
2.1. For the twisted space curve y 3 + 27axz − 81a2 y = 0,
NE
given parametrically by
x = au 3 − u2 , y = 3au2 , z = au 3 + u2
equal
(e) any of the Frenet-Serret formulae that you have not al-
ready used directly are satisfied.
ys
2.2. The shape of the curving slip road joining two motor-
ways, that cross at right angles and are at vertical heights
Ph
T
(2h/π) csc(zπ/h) at height z and that the torsion τ = −1/ρ.
To shorten the algebra, set z = 2hθ/π and use θ as the pa-
NE
rameter.
2.3. Parameterising the hyperboloid
x2 y 2 z 2
+ − 2 =1
a2 b2 c
ide
by x = a cos θ sec φ, y = b sin θ sec φ, z = c tan φ, show that
an area element on its surface is
1/2
dS = sec2 φ c2 sec2 φ b2 cos2 θ + a2 sin2 θ + a2 b2 tan2 φ
dθdφ
gu
Use this formula to show that the area of the curved surface
x2 + y 2 − z 2 = a2 between the planes z = 0 and z = 2a is
√
ics
1
πa2 6 + √ sinh−1 2 2
2
2.4. For the function
ys
2 2
z(x, y) = x2 − y 2 e−x −y
p √
(a) Find the location , (s = x2 + y 2 = ρ) , at which the
Ph
T
(a) Simplify
NE
∇ × a(∇ · a) + a × [∇ × (∇ × a)] + a × ∇2 a
(b) By explicitly writing out the terms in Cartesian coordi-
nates, prove that
[c · (b · ∇) − b · (c · ∇)]a = (∇ × a) · (b × c)
ide
(c) Prove that a × (∇ × a) = ∇ 12 a2 − (a · ∇)a
1 2
u − v2
x = uv cos φ, y = uv sin φ, z =
2
(a) Show that the system of coordinates is an orthogonal
Ph
T
Cartesian coordinates as by
NE
x = cosh u cos v cos φ, y = cosh u cos v sin φ, z = sinh u sin v
T
NE
2.1.
(a) Evaluate (dr/du) · (dr/du)
(b) Integrate the previous result between u = 0 and u = 1
√ −1
(c) t̂ = 2 1 + u2 1 − u2 i + 2uj + 1 + u2 k . Use
dt̂/ds = (dt̂/du)/(ds/du) ρ−1 = |dt̂/ds|
−1
(d) n̂ = 1 + u2 −2ui + 1 − u2 j .
ide
√ −1 2
b̂ = 2 1 + u2 u − 1 i − 2uj + 1 + u2 k
2 3
1 − u2 i + 2u
−2 3 2a 1 + u
√ √
5± 17
2.4. (a) s = (3 ± 5)/2 , s = 1 or s = 4 ,
s = 0 or
ys
√
s = 1, s = (3 ± 5)/2
√ √
(b) 0.541, along the lines x ± y = ± 2 and x ± y = ∓ 2
Ph
2.9.2 Solutions
T
NE
Solution: 2.1. (a) We must first calculate
dr
= 3a − 3au2 , 6au, 3a + 3au2
du
from which it follows that
1/2
ds
du
=
dr dr
·
du du
ide 1/2
= 3a 1 − 2u2 + u4 + 4u2 + 1 + 2u2 + u4
√
= 3 2a 1 + u2
gu
0 3 0
Ph
(c) Using
dr dr du 3a
1 − u2 , 2u, 1 + u2
t̂ = = = √
ds du ds 3 2a (1 + u2 )
we find that
T
dt̂ dt̂ du
=
NE
ds du ds
d 1 − u2
1 1 d 2u
= √ √ , ,
3 2a (1 + u2 ) 2 du 1 + u2 du 1 + u2
d 1 + u2
du 1 + u2
1
=
6a (1 + u2 )3
−4u, ide
2 − 2u2
, 0
1 1
= κ = = =
ρ ds 6a (1 + u2 )3 3a (1 + u2 )2
ys
2 (1 + u2 )
1 2
n̂ = −2u, 1 − u , 0
1 + u2
T
b̂ = t̂ × n̂
NE
1 h
4 2
i
2 2
=√ u − 1, −2u 1 + u , 1 + u
2 (1 + u2 )2
1 u2 − 1 −2u
=√ , ,1
2 u2 + 1 u2 + 1
From this it follows that
db̂ db̂ du
=
ide
ds du ds !
1 1 4u 2 u2 − 1
= √ √ , ,0
gu
3 2a (1 + u2 ) 2 (1 + u2 )2 (1 + u2 )2
that
2
τ=
6a (1 + u2 )2
But
ys
1 1
κ= =
ρ 3a (1 + u2 )2
thus establishing the result that τ equals κ for this curve.
Ph
T
dn̂ dn̂ du
LHS = =
NE
ds du ds
d 1 − u2
1 d −2u
= √ , ,0
3 2a (1 + u2 ) du 1 + u2 du 1 + u2
2
1 2u − 2 −4u
= √ , ,0
3 2a (1 + u2 ) (1 + u2 )2 (1 + u2 )2
1
= √
3 2a (1 + u2 )3
2uide
2
− 2, −4u, 0
1 2
= √ − −4u,
ics
2u 2, 0
3 2a (1 + u2 )3
Thus, the two sides are equal and the unused formula is
verified.
ys
and
T
ds 1/2
= A tan2 θ + cot2 θ + 2
dθ
NE
1/2
= A sec2 θ + csc2 θ
A A
= 1/2
=
sin2 θ cos2 θ sin θ cos θ
Next,
dr dr dθ
t̂ = =
ide
ds dθ ds
√ sin θ cos θ
= A(− tan θ, cot θ, 2)
A
2 2
√
= − sin θ, cos θ, 2 sin θ cos θ
gu
from which it follows that
1 dt̂ dt̂ dθ sin θ cos θ √
n̂ = = = (− sin 2θ, − sin 2θ, 2 cos 2θ)
ics
ρ ds dθ ds A
Thus,
1 sin θ cos θ 1/2
sin2 2θ + sin2 2θ + 2 cos2 2θ
ys
=
ρ A
√
2 sin θ cos θ
=
A
Ph
T
1 √
n̂ = √ (− sin 2θ, − sin 2θ, 2 cos 2θ)
NE
2 √
2 2
t̂ = − sin θ, cos θ, 2 sin θ cos θ
2 2
√
⇒ b̂ = t̂ × n̂ = cos θ, − sin θ, 2 sin θ cos θ
db̂ dθ sin θ cos θ √
⇒ b̂ = = (− sin 2θ, − sin 2θ, 2 cos 2θ)
dθ ds √ A
=
ide
2 sin θ cos θ
A
From this it follows that
gu
√
db̂ − 2 sin θ cos θ π zπ 1
τ = −n̂ · = √ = − sin =−
ds 2h/π 2h h ρ
ics
dr
= (−a sin θ sec φ, b cos θ sec φ, 0)
dθ
Ph
dr
= a cos θ sec φ tan φ, b sin θ sec φ tan φ, c sec2 φ
dφ
T
dr dr
dS = × dθdφ
NE
dθ dφ
= bc cos θ sec3 φ, ac sin θ sec3 φ, −ab sec2 φ tan φ dθdφ
1/2
= sec2 φ c2 sec2 φ b2 cos2 θ + a2 sin2 θ + a2 b2 tan2 φ
dθdφ
1 1
sec2 φdφ = √ cosh ψdψ and sinh2 ψ
sec2 φ = 1 +
2 2
√
ys
T
Z Ψ 1/2
1 1 1
S = 2π √ cosh ψdψa2 1 + sinh2 ψ + sinh2 ψ
NE
0 2 2 2
√ Z Ψ
= 2πa 2
cosh2 ψdψ
0
√ 2 Z Ψ
2πa
cosh2 ψ + 1 dψ
=
2 0
√ 2
Ψ
=
πa
2
2
2πa sinh 2ψ
2
+ψide0
2 2
f (ρ, φ) = ρ2 cos2 φ − sin2 φ e−ρ = ρ2 cos 2φe−ρ
2 2
∇f = 2 ρ − ρ3 e−ρ cos 2φêρ − 2ρ2 e−ρ sin 2φêφ
h i
2 2 −2ρ2 2 2 2 2 2
|∇f | = 4ρ e 1 − ρ cos 2φ + ρ sin 2φ
= 4 se−2s (1 − s)2 cos2 2φ + s sin2 2φ
T
For the line of steepest gradient
NE
(i) ∂|∇f |2 /∂φ = 0 giving
(2.75)
+ s −2(1 − s) cos2 2φ + sin2 2φ = 0
T
stitution, are 0, 0.156 and 0.345 For cos 2φ = 0, sin2 2φ = 1
and the equation (2.75) reduces to
NE
(1 − 2s)s + s = 0 ⇒ s = 0 or s = 1
x±y =∓ 2
T
∇ × a(∇ · a) = ∇(∇ · a) × a + (∇ · a)(∇ × a)
NE
a × [∇ × (∇ × a)] = [a × ∇(∇ · a)] − a × ∇2 a
∇(∇·a)×a+(∇·a)(∇×a)+a×∇(∇·a)−a×∇2 a+a×∇2 a
ide
Thus the original expression is equal to (∇ · a)(∇ × a)
T
∂ax ∂ax ∂ay
(cx by − bx cy ) + (cx bz − bx cz ) + (cy bx − by cx )
∂y ∂z ∂x
NE
∂ay ∂az ∂az
+ (cy bz − by cz ) + (cz bx − bz cx ) + (cz by − bz cy )
∂z ∂x ∂y
∂ay ∂ax ∂az ∂ay
= (bx cy − by cx ) − + (by cz − bz cy ) −
∂x ∂y ∂y ∂z
∂ax ∂az
+ (bz cx − bx cz ) −
ide
∂z ∂x
=(b × c)z (∇ × a)z + (b × c)x (∇ × a)x + (b × c)y (∇ × a)y
=(b × c) · (∇ × a)
gu
(c) Consider the z-component of the LHS.
[a × (∇ × a)]z = ax (∇ × a)y − ay (∇ × a)x
∂ax ∂az ∂az ∂ay
ics
= ax − − ay −
∂z ∂x ∂y ∂z
∂ax ∂ay ∂az ∂az ∂az ∂az
= ax + ay + az − az − ax − ay
∂z ∂z ∂z ∂z ∂x ∂y
ys
1 ∂
a2x + a2y + a2z − (a · ∇)az
=
2 ∂z
1 2
= ∇ a − (a · ∇)a
Ph
2 z
= z -component of RHS.
In the third line az (∂az /∂z) was both added and subtracted.
The corresponding results for the x - and y -components
T
result.
NE
Solution: 2.6. (a) In Cartesian coordinates we need to
evaluate
2 ∂ 2ψ ∂ 2ψ ∂ 2ψ
∇ψ= + 2 + 2
∂x2 ∂y ∂z
The required derivatives are
∂ψ
∂x
=
2xz y 2 + z 2
(x2 + y 2 + z 2 )2
,
ide ∂ 2ψ
∂x2
=
y 2 + z 2 2zy 2 + 2z 3 − 6x2 z
(x2 + y 2 + z 2 )3
∂ψ −2x2 yz ∂ 2ψ 2zx2 x2 + z 2 − 3y 2
= , =−
∂y (x2 + y 2 + z 2 )2 ∂y 2 (x2 + y 2 + z 2 )3
gu
∂ψ x2 x2 + y 2 − z 2 ∂ 2ψ 2zx2 3x2 + 3y 2 − z 2
= , =−
∂z (x2 + y 2 + z 2 )2 ∂z 2 (x2 + y 2 + z 2 )3
Thus, writing r2 = x2 + y 2 + z 2
ics
2 2
2 2 2
4
2z y + z y + z − 3x − 4x
∇2 ψ =
(x2 + y 2 + z 2 )3
2z r2 − x2 r2 − 4x2 − 4x4
ys
=
r6
2z r2 − 5x2
=
Ph
r4
T
2
1 ∂ 2 ∂ψ
∇ψ r= 2 r
r ∂r ∂r
NE
2
= cos θ sin2 θ cos2 φ
r
1 ∂ ∂ψ
∇2 ψ θ = 2
sin θ
r sin θ ∂θ ∂θ
=
ide
1 cos2 φ ∂
r sin θ ∂θ
sin θ
∂
∂θ
2
cos θ sin θ
cos2 φ
4 cos3 θ − 8 sin2 θ cos θ
=
r
gu
1 ∂ 2ψ
∇2 ψ φ = 2 2
r sin θ ∂φ2
cos θ
−2 cos2 φ + 2 sin2 φ
ics
=
r
Thus, the full Laplacian in spherical polar coordinates is
cos θ
∇2 ψ = 2 sin2 θ cos2 φ + 4 cos2 θ cos2 φ
ys
r
−8 sin2 θ cos2 φ − 2 cos2 φ + 2 sin2 φ
cos θ
4 cos2 φ − 10 sin2 θ cos2 φ − 2 cos2 φ + 2 sin2 φ
Ph
=
r
cos θ
2 − 10 sin2 θ cos2 φ
=
r
2r cos θ r2 − 5r2 sin2 θ cos2 φ
=
r4
jahir@physicsguide.in 111 physicsguide CSIR NET, GATE
©Sk Jahiruddin, 2020 Vectors
T
nates, one finally obtains
NE
2 2
2z r − 5x
∇2 ψ =
r4
which establishes the equivalence of the two approaches.
e1 =
∂r
∂u
∂r
ide
= (v cos φ, v sin φ, u)
Thus
T
p
hu = hv = u2 + v 2 , hφ = uv
NE
(b) The u -component of ∇ × a is given by
hu ∂ ∂
[∇ × a]u = (hφ aφ ) − (hv av )
hu hv hφ ∂v ∂φ
1 ∂ ∂ p 2
= √ (uvaφ ) − u + v 2 av
2
uv u + v 2 ∂v ∂φ
=√
1
u2 + v 2 v
ide
aφ ∂aφ
+
∂v
−
1 ∂av
uv ∂φ
T
e1 · e2 = sinh u cos v cosh u sin v − cos2 φ − sin2 φ + 1 = 0
NE
e1 · e3 = sinh u cos2 v cosh u(− sin φ cos φ + sin φ cos φ) = 0
e2 · e3 = cosh2 u sin v cos v(sin φ cos φ − sin φ cos φ) = 0
As each is zero, the system is an orthogonal one.
ide
The scale factors are given by |ei | and are thus found
from:
= cosh2 u − cos2 v
|e2 |2 = cosh2 u sin2 v cos2 φ + sin2 φ + sinh2 u cos2 v
ics
= cosh2 u − cos2 v
ys
1/2
hu = hv = cosh2 u − cos2 v , hφ = cosh u cos v
1/2
An alternative form for hu and hv is sinh2 u + sin2 v
T
ψ(u), of u only, then all differentiation with respect to v and
φ can be ignored. The expression for ∇2 ψ reduces to
NE
1 ∂ h h
v φ ∂ψ
∇2 ψ =
hu hv hφ ∂u hu ∂u
1 ∂ ∂ψ
= cosh u cos v
cosh u cos v cosh2 u − cos2 v ∂u
∂u
ide
Laplace’s equation itself is even simpler and reduces to
∂ ∂ψ
cosh u =0
∂u ∂u
gu
This can be rewritten as
∂ψ k 2k 2keu
ics
= = = 2u
∂u cosh u eu + e−u e +1
u
Ae du −1 u
dψ =
u 2 ⇒ ψ = B tan e +c
1 + (e )
ys
T
grals
NE
3.1 Line Integral
C C
ide
a · dr,
Z
a × dr
C
(3.1)
C C C C
T
Z Z
φidx = i φdx
NE
which is allowable since the Cartesian unit vectors are of
constant magnitude and direction and hence may be taken
out of the integral. If we had been using a different coordi-
nate system, such as spherical polars, then, as we saw in the
ide
previous chapter, the unit basis vectors would not be con-
stant. In that case the basis vectors could not be factorised
out of the integral.
gu
The second and third line integrals can also be reduced
to a set of scalar integrals by writing the vector field a in
terms of its Cartesian components as a = ax i + ay j + az k,
ics
= (ax dx + ay dy + az dz)
Ph
Z C Z Z
= ax dx + ay dy + az dz
C C C
T
grals have properties that are analogous to those of ordinary
integrals. In particular, the following are useful properties
NE
(which we illustrate using the second form of line integral
but which are valid for all three types)
Z B
ide
evaluated has A and B as its end-points then
Z A
a · dr = − a · dr
A B
gu
This implies that if the path C is a loop then integrating
around the loop in the opposite direction changes the sign
of the integral.
ics
Z B Z P Z B
a · dr = a · dr + a · dr
A A P
T
point (0,0) to (2,1).
NE
ide
gu
ics
dW = F · dr
Ph
In the xy plane
We need to evaluate
T
Z
xydx − y 2 dy
W =
NE
(a) Along path 1, y = 12 x, dy = 12 dx.
W =
Z 2"
0
1
x · xdx −
2
ide1
2
2
1
# Z
x · dx =
2
2
0 8
3 2
2
x3
x dx = = 1
8 0
gu
Along path 2, y = 14 x2 , dy = 12 xdx
ics
Z 2 Z 2
1 1 1 1 3 1 5
W = x · x2 dx − x4 · xdx = x − x dx
0 4 16 2 0 4 32
2
x4 x6
2
= − =
ys
16 192 0 3
then from (0, 1) to (2, 1) and add the results. Along (0, 0)
to (0, 1), x = 0 and dx = 0 so we must use y as the variable.
Then we have
T
1
1
y 3
Z
2
1
0 · y · 0 − y dy = − = −
y=0 3 0 3
NE
Along (0, 1) to (2, 1), y = 1, dy = 0
2
2
x2
Z
(x · 1 · dx − 1 · 0) = = 2
2 0
x=0
Z 1 Z 1
3 2 2 4 12 2
12t7 − 2t5 dt = −
W = 2t · t · 6t dt − t · 2tdt =
ys
0 0 8 6
xdy − ydx
Z
I=
x2 + y 2
along each of the two paths indicated in Figure below from
(−1, 0) to (1, 0).
T
NE
ide
Solution: The problem may be given as: Evaluate
Z
I = F · dr with F = (−iy + xj)/ x2 + y 2
gu
So you need to identify the line integral sometimes.
x2 + y 2 1
At (−1, 0), θ = π; at (1, 0), θ = 0. Then we get
Z 0
I1 = dθ = −π
π
T
from (0, 1) to (1, 0) and add the results. The first straight
line has the equation y = x + 1; then dy = dx and the
NE
integral is
Z 0 Z 0 Z 0
xdx − (x + 1)dx −dx −2dx
2 2
= 2
= 2
−1 x + (x + 1) −1 2x + 2x + 1 −1 (2x + 1) + 1
= − arctan(2x + 1)|0−1
ide
= − arctan 1 + arctan(−1)
π π
=− + − =−
π
4 4 2
Along the second straight line y = 1 − x, dy = −dx, and the
gu
integral is
Z 1
xdx + (1 − x)dx
−
ics
0 x2 + (1 − x)2
Z 1
−2dx
= 2
= − arctan(2x − 1)|10
0 (2x − 1) + 1
π
ys
=−
2
Adding the results for the integrals along the two parts
Ph
of path 2, we get I2 = −π
T
cases the path may be subdivided into shorter line segments
along which one coordinate is a single-valued function of the
NE
other two.
T
I Z ap Z −a p
I= xdy = 2 2
a − y dy + (− a2 − y 2 )dy
NE
C −a a
Z ap
=4 a2 − y 2 dy = πa2
0
Suppose the functions P (x, y), Q(x, y) and their partial deriva-
tives are singlevalued, finite and continuous inside and on
the boundary C of some simply connected region R in the
Ph
T
over the enclosed region R. This theorem may be proved in
the following way.
NE
Consider the simply connected region R in figure 11.3,
and let y = y1 (x) and y = y2 (x) be the equations of the
curves ST U and SV U respectively. We then write
ZZ Z b Z y2 (x) Z b
∂P ∂P
R ∂y
dxdy =
a
Z b
dx
y1 (x)
ide
dy
∂y
=
a
y=y (x)
dx[P (x, y)]y=y21 (x)
= − P dx
ics
show that
ZZ Z d Z x2 (y) Z d
∂Q ∂Q x=x (y)
dxdy = dy dx = dy[Q(x, y)]x=x21 (y)
R ∂x ∂x
Ph
c x1 (y) c
Z d
= [Q (x2 (y), y) − Q (x1 (y), y)] dy
Zc c Z d I
= Q (x1 , y) dy + Q (x2 , y) dy = Qdy
d c C
T
plane.
NE
Example: show that the area of a region R enclosed by
a simple closed curve C is given by
I I I
1
A= (xdy − ydx) = xdy = − ydx
2 C C C
C R R
1
H
Therefore the area of the region is A = 2 C (xdy − ydx).
Alternatively, we could put P = 0 and Q = x and obtain
ys
H
A = C xdy, or put P = −y and Q = 0, which gives A =
H
− C ydx The area of the ellipse x = a cos φ, y = b sin φ is
given by
Ph
1 2π
I Z
1
ab cos2 φ + sin2 φ dφ
A= (xdy − ydx) =
2 C 2 0
ab 2π
Z
= dφ = πab
2 0
jahir@physicsguide.in 127 physicsguide CSIR NET, GATE
©Sk Jahiruddin, 2020 Vectors
T
I
I= [(ex y + cos x sin y) dx + (ex + sin x cos y) dy]
NE
C
R
For line integrals of the form C a · dr, there exists a class of
vector fields for which the line integral between two points is
Ph
RB
(i) The integral A a · dr, where A and B lie in the re-
T
gion R, is independent of the path from A to B. Hence the
H
integral C a · dr around any closed loop in R is zero.
NE
(ii) There exists a single-valued function φ of position
such that a = ∇φ
(iii) ∇ × a = 0
ide
(iv) a · dr is an exact differential.
a · dr = φ(B) − φ(A)
A
a · dr = dφ
T
ential: condition (iv). From ( ∇φ · dr = dφ) we can write
dφ = ∇φ · dr, and so we have
NE
(a − ∇φ) · dr = 0
F = 2xy − z 3 i + x2 j − 3xz 2 + 1 k
T
−∇φ
NE
Solution: Calculate the curl
i j k
∂ ∂ ∂
∇×F= =0
∂x ∂y ∂z
2xy − z 3 x2 −3xz 2 − 1
so F is conservative. Then
Z B Z B
ide
2xy − z 3 dx + x2 dy − 3xz 2 + 1 dz
W = F · dr =
A A
gu
is independent of the path.
W = x2 y − xz 3 − z
φ = −W = −x2 y + xz 3 + z
T
following.
NE
We need to make the line integral independent of the
path. Let us choose the origin as our reference point and
RB
integrate the force ( A F · dr) from the origin to the point
(x, y, z).
ide
As the path of integration, we choose the broken line
from (0, 0, 0) to (x, 0, 0) to (x, y, 0) to (x, y, z). From (0, 0, 0)
to (x, 0, 0), we have y = z = 0, dy = dz = 0, so the integral
is zero along this part of the path.
gu
From (x, 0, 0) to (x, y, 0), we have x = const., z = 0, dx =
dz = 0, so the integral is
ics
Z y Z y
x2 dy = x2 dy = x2 y
0 0
From (x, y, 0) to (x, y, z) we have x = const., y = const.,
ys
dx = dy = 0, so the integral is
Z z
3xz 2 + 1 dz = −xz 3 − z
−
Ph
0
Adding the three results, we get
W = x2 y − xz 3 − z
or the potential
T
φ = −W = −x2 y + xz 3 + z
NE
3.4 Surface integrals
T
sent the surface in a simple way in any particular coordinate
system. In such cases, it is usual to work in Cartesian coor-
NE
dinates and consider the projections of the surface onto the
coordinate planes.
T
NE
ide
gu
dA dA
ys
dS = = (3.5)
| cos α| |n̂ · k|
So you have also seen that
Ph
1 1
sec α = = (3.6)
cos α |n · k|
T
n̂ = ∇f /|∇f | evaluated at that point. The scalar element
of surface area then becomes
NE
dA |∇f |dA |∇f |dA
dS = = = (3.7)
|n̂ · k| ∇f · k ∂f /∂z
don’t miss that
∂f ∂f ∂f ∂f
∇f · k = i +j
ide +k ·k=
∂x ∂y ∂z ∂z
ZZ ZZ
dA = sec αdxdy
ys
R
Example: Evaluate the surface integral I = S a · dS,
where a = xî and S is the surface of the hemisphere
x2 + y 2 + z 2 = a2 with z ≥ 0
T
figure below. In this case dS may be easily expressed in
spherical polar coordinates as dS = a2 sin θdθdφ, and the
NE
unit normal to the surface at any point is simply r̂. On the
surface of the hemisphere we have x = a sin θ cos φ and you
also know
so you get
ide
a · dS = x(î · r̂)dS = (a sin θ cos φ)(sin θ cos φ) a2 sin θdθdφ
gu
Therefore, inserting the correct limits on θ and φ, we have
Z π/2 Z 2π
2πa3
Z
3 3 2
ics
I= a · dS = a dθ sin θ dφ cos φ =
S 0 0 3
ys
Ph
T
NE
ide
gu
Figure 3.2: The surface of the hemisphere x2 + y 2 + z 2 =
a2 , z ≥ 0
ics
Now
T
∇f = 2xî + 2y ĵ + 2z k̂ = 2r
NE
so on the surface S we have |∇f | = 2a.
On S we also have
p
∂f /∂z = 2z = 2 a2 − x2 − y 2 and î · r̂ = x/a
ZZ
ide
Therefore, the integral becomes
x2
I= p dxdy
R a2 − x 2 − y 2
gu
Although this integral may be evaluated directly, it is
quicker to transform to plane polar coordinates:
ics
ρ2 cos2 φ
ZZ
I= p ρdρdφ
R0 a2 − ρ 2
Z 2π Z a
ρ3 dρ
ys
= cos2 φdφ p
0 0 a2 − ρ 2
Z 2π Z π/2
2 3 3 2πa3
I= cos φdφ a sin udu =
0 0 3
T
Example: Find the area cut from the upper half of the
sphere x2 + y 2 + z 2 = 1 by the cylinder x2 + y 2 − y = 0
NE
Solution : You need to first convince yourself that the
equation of the cylinder transform into
1 1
x2 + (y − )2 =
2 4
ide
So the base of the cylinder is in the xy plane. The area
cut from the upper half of the sphere will be the same as the
area on the sphere which projects onto the disk x2 +y 2 −y ≤
gu
0 in the (x, y) plane. Now the surface area
ics
ZZ ZZ
dA = sec θdxdy
where θ is the angle between the area vector and the z axis.
1 1
ys
T
NE
ide
gu
Figure 3.3:
| grad φ| 1p
sec θ = = (2x)2 + (2y)2 + (2z)2
ys
|∂φ/∂z| 2z
1 1
= =p
z 1 − x2 − y 2
Ph
The area is
Z 1 Z √y−y2
dxdy
A=2 p
y=0 x=0 1 − x2 − y 2
jahir@physicsguide.in 141 physicsguide CSIR NET, GATE
©Sk Jahiruddin, 2020 Vectors
T
NE
Z π/2 Z sin θ
rdrdθ
A=2 √
θ=0 r=0 1 − r2
√
Then change variable z = 1 − r2
Finally
A = −2
Z
ideπ/2 Z cos θ
dzdθ = π − 2
θ=0 z=1
gu
Example: Find the vector area of the surface of the
hemisphere
ics
x2 + y 2 + z 2 = a2
with z ≥ 0
ys
Solution:
Z
S= dS
S
T
coordinates. Therefore the vector area is given by
NE
ZZ
S= a2 sin θr̂dθdφ
S
!
Z 2π Z π/2
+ j a2 sin φdφ sin2 θdθ
0 0
!
Z 2π Z π/2
+ k a2 dφ sin θ cos θdθ
ys
0 0
=0 + 0 + πa2 k = πa2 k
Ph
T
Example: Find the vector area of the surface of the
hemisphere
NE
x2 + y 2 + z 2 = a2 , z ≥ 0,
r × dr = a cos φ a sin φ 0
−a sin φdφ a cos φdφ 0
Ph
T
NE
Volume integrals are defined in an obvious way and are gen-
erally simpler than line or surface integrals since the element
of volume dV is a scalar quantity. We may encounter vol-
ume integrals of the forms
Z Z
φdV, adV (3.8)
V ide V
T
rem
NE
Let A = velocity v at any point of a moving fluid. From a
part of the figure below we have: Volume of fluid crossing
dS in ∆t seconds = volume contained in cylinder of base dS
and slant height v∆t.
ide
gu
ics
Figure 3.4:
ys
T
RR
surface S = S v · ndS
NE
We know that ∇ · vdV is the volume per second of fluid
emerging from a volume element dV. Then Total volume
per second of fluid emerging from all volume elements in
RRR
S= V ∇ · vdV Thus
ZZ
v · ndS =
S
ide
ZZZ
∇ · vdV
V
gu
ZZZ ZZ
∇ · Vdτ = V · ndσ (3.9)
surface
volume τ inclosing τ
ics
density ρ(r, t) and velocity field v(r, t), in which fluid is nei-
ther being created nor destroyed, we will show that
∂ρ
+ ∇ · (ρv) = 0
∂t
T
mass tells us that the rate of increase or decrease of the
mass M of fluid in the volume must equal the net rate at
NE
which fluid is entering or leaving the volume, i.e.
I
dM
= − ρv · dS
dt S
d
R
Z
ide
V is simply M = V ρdV, so we have
I
ρdV + ρv · dS = 0
dt V S
gu
Taking the derivative inside the first integral on the RHS
and using the divergence theorem to rewrite the second in-
tegral, we obtain
ics
Z Z Z
∂ρ ∂ρ
dV + ∇ · (ρv)dV = + ∇ · (ρv) dV = 0
V ∂t V V ∂t
ys
∂ρ
+ ∇ · (ρv) = 0
∂t
This is known as the continuity equation. It can also be
applied to other systems, for example those in which ρ is
T
the flow of an incompressible fluid, ρ = constant and the
continuity equation becomes simply
NE
∇·v =0
Figure 3.5:
T
ZZZ ZZZ
∂A3 ∂A3
dV = dzdydx
∂z ∂z
NE
V V
Z Z "Z f2 (x,y) #
∂A3
= dz dydx
R z=f1 (x,y) ∂z
ZZ f2 ZZ
= A3 (x, y, z) dydx = [A3 (x, y, f2 ) −
R z=f1 R
ide
For the upper portion S2 , dydx = cos γ2 dS2 = k·n2 dS2 since
the normal n2 to S2 makes an acute angle γ2 with k
And
ZZ ZZ
A3 (x, y, f2 ) dydx − A3 (x, y, f1 ) dydx
Ph
ZR Z ZZ R
= A3 k · n2 dS2 + A3 k · n1 dS1
S2 S1
ZZ
= A3 k · ndS
S
So that
T
ZZZ ZZ
∂A3
dV = A3 k · ndS
V ∂z S
NE
Similarly, by projecting S on the other coordinate planes,
ZZZ ZZ
∂A1
dV = A1 i · ndS
V ∂x S
and
ZZZ
ide
∂A2
dV =
ZZ
A2 j · ndS
V ∂y S
Adding we get
gu
ZZZ ZZ
∂A1 ∂A2 ∂A3
+ + dV = (A1 i + A2 j + A3 k)·ndS
V ∂x ∂y ∂z S
ics
Hence ZZZ ZZ
∇ · AdV = A · ndS
V S
ys
Z
I= a · dS,
S
where
a = (y − x)i + x2 zj + z + x2 k
T
x2 + y 2 + z 2 = a2 , z ≥ 0
NE
Solution: We could evaluate this surface integral di-
rectly, but the algebra is somewhat lengthy. We will there-
fore evaluate it by use of the divergence theorem. since the
latter only holds for closed surfaces enclosing a non-zero vol-
ide
ume V, let us first consider the closed surface S 0 = S + S1 ,
where S1 is the circular area in the xy -plane given by
x2 + y 2 ≤ a2 , z = 0; S 0 then encloses a hemispherical vol-
ume V. By the divergence theorem we have
gu
Z I Z Z
∇ · adV = a · dS = a · dS + a · dS
V S0 S S1
ics
T
x2 + y 2 ≤ a2 . Transforming to plane polar coordinates we
have
NE
Z 2π Z a
πa4
ZZ
2 2 2 3
I= ρ cos φρdρdφ = cos φdφ ρ dρ =
R0 0 0 4
S ZV
2 (3.10)
= φ∇ ψ + (∇φ) · (∇ψ) dV
V
ys
I Z
ψ∇φ · dS = ∇ · (ψ∇φ)dV
S V
Z (3.11)
2
= ψ∇ φ + (∇ψ) · (∇φ) dV
V
T
I Z
φ∇2 ψ − ψ∇2 φ dV
(φ∇ψ − ψ∇φ) · dS = (3.12)
NE
S V
3.6.5
ide
Other Related Integral theorems
V S
Z I
∇ × bdV = dS × b (3.14)
V S
Ph
We will prove the first and leave the second to the stu-
dent
T
stant vector. We then have
NE
Z I
∇ · (φc)dV = φc · dS
V S
∇ · (φc) = φ∇ · c + c · ∇φ = c · ∇φ
Z
ide
since c is constant. Also, φc · dS = c · φdS, so we obtain
I
c · (∇φ)dV = c · φdS
V S
gu
since c is constant we may take it out of both integrals to
give Z I
ics
c· ∇φdV = c · φdS
V S
and since c is arbitrary we obtain the stated result in the
equation (3.13)
ys
a constant vector.
T
NE
Stokes’ theorem relates the integral of the curl of a vector
field over an open surface S to the line integral of the vector
field around the perimeter C bounding the surface.
T
ZZ ZZ
(∇ × A) · ndS = [∇ × (A1 i + A2 j + A3 k)] · ndS
NE
S I S
= A · dr
C
ide
gu
ics
ys
Ph
Figure 3.6:
RR
Consider first [∇ × (A1 i)] · ndS. since
T
S
i j k
NE
∂ ∂ ∂ ∂A1 ∂A1
∇ × (A1 i) = ∂x ∂y ∂z = j− k
∂z ∂y
A1 0 0
then
∂A1 ∂A1
[∇ × (A1 i)] · ndS = n·j− n · k dS (3.16)
ide ∂z ∂y
∂r ∂z ∂z
n· =n·j+ n·k=0 or n·j=− n·k
∂y ∂y ∂y
ys
∂z ∂y ∂z ∂y ∂y
or
∂A1 ∂A1 ∂z
[∇ × (A1 i)] · ndS = − + n · kdS (3.17)
∂y ∂z ∂y
T
∂A1 ∂A1 ∂z ∂F
∂y + ∂z ∂y = ∂y and (3.17) becomes
NE
∂F ∂F
[∇ × (A1 i)] · ndS = − n · kdS = − dxdy
∂y ∂y
then ZZ ZZ
∂F
S
ide
[∇ × (A1 i)] · ndS =
R ∂y
dxdy −
must have I I
F dx = A1 dx
Γ C
or ZZ I
ys
ZZ I
[∇ × (A2 j)] · ndS = A2 dy and
S C
ZZ I
[∇ × (A3 k)] · ndS = A3 dz
S C
Thus, by addition,
T
ZZ I
(∇ × A) · ndS = A · dr
NE
S C
T
Z Z 2π Z π/2
dθ −2a2 sin θ r̂ · k
(∇ × a) · dS = dφ
NE
S 0 0
Z 2π Z π/2 z
2
= −2a dφ sin θ dθ
0 0 a
Z 2π Z π/2
2
= −2a dφ sin θ cos θdθ = −2πa2
0 0
ide
We now evaluate the line integral around the perimeter
curve C of the surface, which is the circle x2 + y 2 = a2
in the xy -plane. This is given by
gu
I I
a · dr = (yi − xj + zk) · (dxi + dyj + dzk)
C IC
= (ydx − xdy)
ics
integral becomes
I Z 2π Z 2π
2 2 2 2
(ydx−xdy) = −a sin φ + cos φ dφ = −a dφ = −
Ph
C 0 0
since the surface and line integrals have the same value, we
have verified Stokes’ theorem.
T
NE
As for the divergence theorem, there exist two other integral
theorems that are closely related to Stokes’ theorem. If φ
is a scalar field and b is a vector field, and both φ and b
satisfy our usual differentiability conditions on some two-
sided open surface S bounded by a closed perimeter curve
C, then Z
S
ide I
dS × ∇φ =
C
φdr (3.18)
and Z I
(dS × ∇) × b = dr × b (3.19)
gu
S C
The proof of the first theorem is as follows
ics
S C
Expanding out the integrand on the LHS we have
∇ × (φc) = ∇φ × c + φ∇ × c = ∇φ × c
Ph
T
grals because it is constant, we find
NE
Z I
dS × ∇φ = c · φdr
S C
ide
Equation (3.19) may be proved in a similar way, by let-
ting a = b×c in Stokes theorem, where c is again a constant
vector.
gu
R
Example: Given V = 4yi + xj + 2zk, find (∇ × V) ·
ndσ over the hemisphere
ics
x2 + y 2 + z 2 = a2 , z ≥ 0
T
area inside the circle is also bounded by the same circle
which binds the hemisphere ! Since this plane area is in the
NE
(x, y) plane, we have
n = k, (∇ × V) · n = −3k · k = −3
so the integral is
Z
−3 ide
dσ = −3 · πa2 = −3πa2
and so
r z
(∇ × V) · n = −3k · = −3
a a
Ph
R
We want to evaluate −3(z/a)dσ over the hemisphere. In
spherical coordinates we have
z = r cos θ
dσ = r2 sin θdθdφ
T
NE
Z 2π Z π/2
a cos θ 2
−3 a sin θdθdφ
φ=0 θ=0 a
Z 2π Z π/2
2
= −3a dφ sin θ cos θdθ
0 0
1
= −3a2 · 2π · = −3πa2
ide 2
gu
ics
ys
Ph
3.8 Exercises
T
R
3.1. Evaluate the line integral I = C a · dr, where a =
NE
(x + y)i + (y − x)j, along each of the paths in the xy-plane
shown in figure below ,
ide
gu
ics
T
(i) C1 , given byx = cu, y = c/u, z = h
(ii) C2 , given by 2y = 3c − x, z = h
NE
(iii) Show that the vector field a is in fact conservative, and
find φ such that a = ∇φ
3.3. The vector field F is defined by
F = 2xzi + 2yz 2 j + x2 + 2y 2 z − 1 k
ide
Calculate ∇×F and deduce that F can be written F = ∇φ.
Determine the form of φ
gu
3.4. The vector field Q is defined by
Q = 3x2 (y + z) + y 3 + z 3 i + 3y 2 (z + x) + z 3 + x3 j
+ 3z 2 (x + y) + x3 + y 3 k
ics
that the integral of x − y over the upper half of the unit cir-
T
cle centred on the origin has the value − 23 . Show the same
result by direct integration in Cartesian coordinates.
NE
3.7. Evaluate the line integral
I
y 4x2 + y 2 dx + x 2x2 + 3y 2 dy
I=
C
Describe the shape of the path γ and show that the line
R
integral γ f · dr vanishes. Does this result imply that f is a
ics
conservative field?
3.9. A vector field a = f (r)r is spherically symmetric and
everywhere directed away from the origin. Show that a is
ys
R
3.10. Evaluate the surface integral r·dS, where r is the po-
sition vector, over that part of the surface z = a2 −x2 −y 2 for
which z ≥ 0, by each of the following methods. (a) Param-
eterise the surface as x = a sin θ cos φ, y = a sin θ sin φ, z =
T
r · dS = a4 2 sin3 θ cos θ + cos3 θ sin θ dθdφ
NE
(b) Apply the divergence theorem to the volume bounded
by the surface and the plane z = 0
3.11. (a) Prove
Z I
V
ide
∇ × bdV =
S
dS × b
αr
F=
(r2 + a2 )3/2
Ph
√
through the spherical surface |r| = 3a (b) by showing that
3αa2
∇·F=
(r2 + a2 )5/2
T
√
of the sphere |r| = 3a. The substitution r = a tan θ will
prove useful in carrying out the integration.
NE
3.13. A vector field F is defined in cylindrical polar coordi-
nates ρ, θ, z by
x cos λz y cos λz
F = F0 i+ j + (sin λz)k
a a
≡
F0 ρ
a
ide
(cos λz)eρ + F0 (sin λz)k
where i, j and k are the unit vectors along the Cartesian
axes and eρ is the unit vector (x/ρ)i + (y/ρ)j (a) Calculate,
gu
as a surface integral, the flux of F through the closed sur-
face bounded by the cylinders ρ = a and ρ = 2a and the
planes z = ±aπ/2 (b) Evaluate the same integral using the
ics
divergence theorem.
3.14. The vector field F is given by
F = 3x2 yz + y 3 z + xe−x i + 3xy 2 z + x3 z + yex j
ys
+ x3 y + y 3 x + xy 2 z 2 k
R
value of the line integral L F · dr, where L is the (three-
dimensional) closed contour OABCDEO defined by the
successive vertices (0, 0, 0), (1, 0, 0), (1, 0, 1), (1, 1, 1), (1, 1, 0), (0, 1
(0, 0, 0)
T
nates by
NE
3 2
y y xy/a2 xy x + y xy/a2
F = F0 + e +1 i+ + e j
3a3 a a3 a
z xy/a2 i
+ e k
a
Use Stokes’ theorem to calculate
I
ide
F · dr
L
T
3.1. (i) 11 13 , (ii) 10 23 , (iii) 8
NE
3.2. (i) c(h − 1), (ii) c(h − 1)
3.3. φF (r) = x2 z + y 2 z 2 − z
3.5. (a) c3 ln 2i+2j+(3c/2)k; (b) −3c4 /8 i−cj− c2 ln 2 k;
(c) c4 ln 2 − c
3.5. I = πa3 b/2 ide
3.13. 6πF0 a2 + 2a λπa
λ sin 2
e 5
3.14. 2 − 6
3.15. F0 a 2e3 − 4
gu
ics
ys
Ph
3.8.2 Solution
T
NE
Solution: 3.1. since each of the paths lies entirely in the
xy -plane, we have dr = dxi + dyj. We can therefore write
the line integral as
Z Z
I= a · dr = [(x + y)dx + (y − x)dy] (3.21)
C C
ide
We must now evaluate this line integral along each of the
prescribed paths. Case (i). Along the parabola y 2 = x we
have 2ydy = dx. Substituting for x in ( 11.3) and using just
gu
the limits on y, we obtain
Z (4,2)
I= [(x + y)dx + (y − x)dy]
(1,1)
ics
Z 2
1
y 2 + y 2y + y − y 2 dy = 11
=
1 3
ys
T
line integral in terms of the parameter u Along the curve
x = 2u2 + u + 1, y = 1 + u2 we have dx = (4u + 1)du and
NE
dy = 2udu
2
3u2 + u + 2 (4u + 1) − u2 + u 2u du = 10
=
0 3
gu
Case (iii). For the third path the line integral must be
evaluated along the two line segments separately and the
results added together. First, along the line y = 1 we have
ics
this into equation (3.21) and using just the limits on y for
this segment, we obtain
Z (4,2) Z 2
1
[(x + y)dx + (y − x)dy] = (y − 4)dy = −2
(4,1) 1 2
jahir@physicsguide.in 174 physicsguide CSIR NET, GATE
©Sk Jahiruddin, 2020 Vectors
The value of the line integral along the whole path is just the
T
sum of the values of the line integrals along each segment,
and is given by I = 10 12 − 2 12 = 8
NE
Solution: 3.2. Expanding out the integrand, we have
Z (2c,c/2,h)
xy 2 + z dx + x2 y + 2 dy + xdz
I=
(c,c,h)
ide
which we must evaluate along each of the paths C1 and C2
(i) Along C1 we have dx = cdu, dy = − c/u2 du, dz = 0,
and on substituting in the integral and finding the limits on
u, we obtain
gu
Z 2
2
I= c h − 2 du = c(h − 1)
1 u
ics
Z 2c
1 3 9 2 9 2
I= x − cx + c x + h − 1 dx = c(h − 1)
c 2 4 4
Ph
Hence the line integral has the same value along paths C1
and C2 . Taking the curl of a, we have
T
tween two points must be independent of the path taken.
since a is conservative, we can write a = ∇φ.
NE
Now you integrate each of the function and take the
repeating term only once as shown in the example before in
the chapter and show that
Z
1
Z
ide
xy 2 + z dx = x2 y 2 + zx
2
1
x2 y + 2 dy = φ = x2 y 2 + 2y
2
gu
Z
xdz = xz
hence
ics
1
φ = x2 y 2 + zx + 2y + k
2
where k is a constant.
ys
φ must satisfy
Ph
∂φ
= xy 2 + z
∂x
which implies that φ = 21 x2 y 2 +zx+f (y, z) for some function
f. Secondly, we require
T
∂φ ∂f
= x2 y + = x2 y + 2
NE
∂y ∂y
which implies f = 2y + g(z). Finally, since
∂φ ∂g
=x+ =x
∂z ∂z
we have g = constant = k.
So the function
ide
1
φ = x2 y 2 + zx + 2y + k
gu
2
Solution: 3.3.
ics
T
∂φ
⇒ φ(x, y, z) = x2 z + g(y, z)
2xz = Fx =
∂x
NE
∂
2yz 2 = Fy = x2 z + g(y, z) ⇒ g(y, z) = y 2 z 2 + h(z)
∂y
∂ 2
x2 + 2y 2 z − 1 = Fz = x z + y 2 z 2 + h(z)
∂z
⇒ h(z) = −z + k
ide
Hence, to within an unimportant constant, the form of φ is
φ(x, y, z) = x2 z + y 2 z 2 − z
gu
Solution: 3.4.
∂Q
(∇ × Q)z = ∂xy − ∂Q 2 2
2 2
∂y
x
= 3y + 3x − 3x + 3y =0
Hence, ∇ × Q = 0 which implies that Q is indeed a con-
Ph
f (x, y, z) = x3 (y + z) + x y 3 + z 3
⇒
T
From its y -component,
NE
∂g
x3 +3y 2 x+ = 3y 2 (z+x)+z 3 +x3 ⇒ g(y, z) = y 3 z+z 3 y
∂y
And finally, from its z-component,
∂h
x3 + 3z 2 x + y 3 + 3z 2 y + = 3z 2 (x + y) + x3 + y 3
∂z
Thus,
ide
⇒ h(z) = c
φ(x, y, z) = x3 (y + z) + y 3 + z 3 x + y 3 z + z 3 y + c
gu
= yz y 2 + z 2 + zx z 2 + x2 + xy x2 + y 2 + c
(−2 + c) = 54
the same path L, they are not necessarily of the same type.
The vector or scalar nature of the integral is determined
by that of the integrand when it is expressed in a form
Ph
T
L, before integrating:
NE
Z Z 2 3
c
Fdt = i + 2j + ctk dt
L 1 t
2
1
= c3 ln ti + 2tj + ct2 k
2 1
3
= c3 ln 2i + 2j + ck
ide 2
L 1
4 2
c 2c
= 2
i + j − c2 ln tk
2t t 1
4
3c
ys
= − i − cj − c2 ln 2k
8
Ph
T
Z Z 2 3
c c
F · dr = i + 2j + ctk · ci − 2 j + 0k dt
NE
L 1 t t
Z 2 4
c 2c
= − 2 dt
1 t t
2
4 2c
= c ln t +
t 1
ide
= c4 ln 2 − c
Solution: 3.6.
gu
To obtain the integral of x−y over the bounded region we
must choose Q(x, y) such that ∂Q/∂x is x, and P (x, y) such
that ∂P/∂y is y. Clearly Q(x, y) = 21 x2 and P (x, y) = 21 y 2
ics
T
Z π
I= sin2 θ(− sin θdθ) + cos2 θ(cos θdθ)
NE
0Z
π Z π
1 − cos2 θ sin θdθ + 1 − sin2 θ cos θdθ
=−
0 π 0
π
1 3 1 3
= cos θ − cos θ + sin θ − sin θ
3 0 3 0
2 4
= −2 + + 0 − 0 = −
3
ide3
The integral of x − y is therefore one-half of this, i.e. − 23
gu
As a double integral in Cartesian coordinates, we have
√
Z 1 Z 1−x2 p
Z 1
1
1 − x2 dx
dx (x − y)dy = x 1 − x2 −
2
ics
−1 0 −1
1 1
x3
1 2 3/2
1
= − 1−x − x−
3 −1 2 3 −1
1 2
=0−0−1+ =−
ys
3 3
Solution: 3.7.
Ph
T
might usefully be applied. The theorem states that
NE
I ZZ
∂Q ∂P
(P dx + Qdy) = − dxdy
C R ∂x ∂y
P (x, y) = y 4x2 + y 2
It follows that
ide
and Q(x, y) = x 2x2 + 3y 2
∂P ∂Q
gu
= 4x2 + 3y 2 and = 6x2 + 3y 2
∂y ∂x
leading to
ics
∂Q ∂P
− = 2x2
∂x ∂y
This can now be substituted into Green’s theorem and the y
ys
T
ZZ
I= 2x2 dxdy
NE
R
a 2 1/2
Z
x
= 2x2 2b 1 − 2 dx
−a a
Z 0
= 4b a2 cos2 φ sin φ(−a sin φdφ), on setting x = a cos φ
π
= −ba3
Z 0
π
ide1
sin2 (2φ)dφ = πba3
2
In the final line we have used the standard result for the
integral of the square of a sinusoidal function.
gu
Solution: 3.8.
ics
T
−c sin θdθ, c cos θdθ, c2 dθ
NE
and so the line integral of f along γ is given by
Z Z 2π
y(−c sin θ) − x(c cos θ) + c2 dθ
f · dr =
=
γ
Z 2π
0 ide
−c(b + c sin θ) sin θ − c(a − c + c cos θ) cos θ + c2 dθ
Z0 2π
−bc sin θ − c2 sin2 θ − c(a − c) cos θ − c2 cos2 θ + c2 dθ
gu
=
0
= 0 − πc2 − 0 − πc2 + 2πc2 = 0
ics
∇ × f = (0 − 0, 0 − 0, −1 − 1) = (0, 0, −2) 6= 0
Ph
T
êr rêθ r sin θêφ
1 ∂
NE
∂ ∂
∇×a= 2
∂r ∂θ ∂φ
r sin θ rf (r)
∂r 0 0
1 ∂(rf ) 1 ∂(rf )
= 0êr + êθ − êφ
r sin θ ∂φ r ∂θ
=0
ide
Hence a is irrotational. For it also to be solenoidal requires
that
1 ∂ 2 A
r rf (r) +0+0 ⇒ r3 f (r) = A ⇒ f (r) = 3
0 = ∇·a = 2
gu
r ∂r r
face appropriately:
= a2 1 − sin2 θ = a2 cos2 θ = z
T
for dS and hence r · dS in terms of θ and φ as follows:
NE
r = a sin θ cos φi + a sin θ sin φj + a2 cos2 θk
giving r · dS as
0 ≤ θ ≤ π/2 as follows:
T
Z Z 2π Z π/2
r · dS = a4 2 sin3 θ cos θ + cos3 θ sin θ dθ
dφ
NE
0 0
4 π/2 4
π/2 !
sin θ − cos θ
= 2πa4 2 +
4 0 4 0
3πa4
2 1
= 2πa4 + =
4 4 2
ide
(b) The divergence of the vector field r is 3, a constant,
R
and so the surface integral r · dS taken over the complete
surface Σ (including the part that lies in the plane z = 0) is,
gu
by the divergence theorem, equal to three times the volume
V of the region bounded by Σ. Now,
Z a2 Z a2
ics
1 1
πρ2 dz = π a2 − z dz = π a4 − a4 = πa4
V =
0 0 2 2
R
and so Σ r·dS = 3πa4 /2 However, on the part of the surface
ys
the normal to the face. The surface integral over this face
therefore contributes nothing to the total integral and the
value obtained is that due to the curved surface alone, in
agreement with the result in (a)
T
trary but fixed vector. Then, from the divergence theorem
NE
Z I
∇·a= a · dS
Z V IS
∇ · (b × c) = (b × c) · dS
Z V IS
[c · (∇ × b) − b · (∇ × c)]dV = (dS × b) · c
V
We have used
ide S
∇ · (a × b) = b · (∇ × a) − a · (∇ × b)
gu
And the cyclic property of a triple scalar product
But ∇ × c = 0 and so
ics
Z I
c · (∇ × b)dV = c · dS × b
V S
Z I
(∇ × b)dV = dS × b
V S
Ph
T
giving
NE
I Z
dS × u = ∇ × (ω × x)dV
S ZV
Thus
ics
I
1
ω=− u × dS
2V S
αr αr
F= = êr
(r2 + a2 )3/2 (r2 + a2 )3/2
Ph
T
the surface of the sphere is equal to the volume integral of
its divergence within the sphere. The divergence is given by
NE
!
2
1 ∂ 2 1 ∂ r αr
∇·F= 2 r Fr = 2
r ∂r r ∂r (r2 + a2 )3/2
" #
1 3αr2 3αr4
= 2 −
r (r2 + a2 )3/2 (r2 + a2 )5/2
=
ide
3αa2
(r2 + a2 )5/2
and on integrating over the sphere, we have
gu
Z √3a
3αa2
Z
∇ · FdV = 5/2
4πr2 dr,
V 0 (r2 + a2 )
π
ics
set r = a tan θ, 0 ≤ θ ≤
3
Z π/3 2
2 a tan θa sec2 θ
2
= 12παa dθ
0 a5 sec5 θ
ys
Z π/3
= 12πα sin2 θ cos θdθ
0
3 π/3 √ √
Ph
sin θ 3 3 3πα
= 12πα = 12πα = ,
3 0 8 2
as in (a).
T
Solution: 3.13. (a) The flux through the cylindrical
NE
surfaces
Z aπ/2 Z 2π ρ=2a
F0
= cos λzdz ρρdφ
−aπ/2 a 0 ρ=a
Z aπ/2
2πF0
(2a)2 − (a)2 cos λzdz
=
−aπ/2
= 6πF0 a
a
Z aπ/2
ide
cos λzdz
−aπ/2
12πF0 a λπa
gu
= sin
λ 2
λπa λπa
= π 4a2 − a2 F0 sin
− F0 sin −
2 2
λπa
= 6πa2 F0 sin
ys
2 2a λπa
Total flux = 6πF0 a + sin
λ 2
T
F0
∇·F= (cos λz + cos λz + aλ cos λz)
a
NE
which is independent of ρ. Thus the ρ and φ integrations are
trivial and the volume integral of the divergence reduces to
πF0 4a2 − a2 (2 + aλ)
Z Z aπ/2
∇ · FdV = cos λzdz
V −aπ/2 a
aπ/2
sin λz
2a
ide
= 3πF0 a(2 + aλ)
λ
−aπ/2
λπa
= 6πF0 + a2 sin , as in part (a)
λ 2
gu
Solution: 3.14. (a) This calculation is a piece-wise
evaluation of the line integral, made up of a series of scalar
products of the length of a straight piece of the contour
ics
Z 1 Z 1
−x −x 1
e−x dx = 1 − 2e−1
I1 = xe dx = −xe 0 +
0 0
Ph
T
Z 0
1
1 + 1 + z 2 dz = −1 − 1 −
I4 =
NE
1 3
T
NE
(∇ × F)x = x3 + 3y 2 x + 2yxz 2 − 3xy 2 − x3 = 2yxz 2
(∇ × F)y = 3x2 y + y 3 − 3x2 y − y 3 − y 2 z 2 = −y 2 z 2
(∇ × F)z = 3y 2 z + 3x2 z + yex − 3x2 z − 3y 2 z = yex
OABCDEO
Z Z
= (∇ × F)z dxdy − (∇ × F)x dydz
S S
Z 11 Z 1 Z 1 Z2 1
ys
T
2
y 1 xy/a2 (x + y)y xy/a2
(∇ × F)z =F0 + e + e
NE
a3 a a3
2
y 1 xy/a2 xy xy/a2
− F0 + e + 3e
a3 a a
2
F0 y 2
= 3 exy/a
a
ide
So, by Stokes’ theorem, the line integral has the same value
as this component of curl F integrated over the area of the
rectangle, i.e.
gu
Z a Z 3a
F0 y 2 xy/a2
I
F · dr = dx dy 3 e
L 0 a a
I Z
F · dr = 3 dy
L a a y
After simplification, this can be integrated by parts to yield
Ph
T
F0 3a y/a
I Z
F · dr = ye − y dy
NE
L a a
( Z 3a 2 3a )
F0 h y/a i3a y
= aye − aey/a dy −
a a a a a
F0 9 1
= 3a2 e3 − a2 e − a2 e3 + a2 e − a2 + a2
a 2 2
= F0 a 2e3 − 4 ide
zx2
φ(x, y, z) = 2
x + y2 + z2
(a) Directly in Cartesian coordinates
T
Exercise 4.3. Show that the force field
NE
F = ysin2xî + sin2 xĵ
x2/3 + y 2/3 = 4
Exercise 4.6. For the vector F~ = x2 î + y 2 ĵ + z 2 k̂ Find
→
−
the surface integral F~ .ds over the whole surface of the cube
ys
→
−
Exercise 4.8. Evaluate F~ .ds for F~ = xî + y ĵ over the part
T
of the surface z = 4 − x2 − y 2 that is above the (x,y) plane.
(Apply divergence theorem to the volume bounded by the
NE
surface and the piece that it cuts out of the (x,y) plane) .
Exercise 4.9. Find (∇ × V).ds for vector V~ = 4y î+xĵ +
R
dinates
T
1
x = (u2 − v 2 )
NE
2
y = uv
z=z
T
F = 2xy − z 3 î + x2 ĵ − 3xz 2 + 1 k̂
NE
→
−
is conservative, and find a scalar potential φ such that F =
→
−
− ∇φ.
RRR →− → −
Exercise 4.16. Evaluate ∇ · V dτ over the volume x2 +
→
−
x2 + y 2 (~ix + ~jy).
p
y 2 ≤ 4, 0 ≤ z ≤ 5, V =
ide
Exercise 4.17. Given the vector A = x2 − y 2 i + 2xyj
RR
[(a)] Find ∇ × A. [(b)] Evaluate (∇ × A) · dσ over
gu
a rectangle in the (x, y) plane bounded by the lines x =
H
0, x = a, y = 0, y = b . [(c)] Evaluate A · dr around the
boundary of the rectangle and thus verify Stoke’s theorem
ics
H
Exercise 4.20. Evaluate (2ydx−3xdy) around the square
T
bounded by x = 3, x = 5, y = 1 and y = 3.
NE
4.1 Ans keys of More Exercises
4.1. (a)i + j, |∇φ| = e 4.12.
(b) − 1/2
îu + ĵv
(c)i, |E| = 1
2 3
4.10. î(xz − y 3 /3)+ ĵ(−yz + 4.15. φ = −x y + xz + z
x3 /3)+ k̂(x + y)z + ∇u~ 4.16. 80π
4.11. −2π 4.17. 2ab2
T
+kezy + ∇u 4.20. 20
NE
ide
gu
ics
ys
Ph