Vector Analysis: Review

1
 < Vector Algebra >

First, Geometric representation of a vector

R or R Magnitude: length
O Direction: pointed by an arrow

Same vectors?
2
 Unit Vector

◼ Null (zero) vector : vector with magnitude 0

◼ Unit vector : vector with unit magnitude

A
aA = A = Aa A
A
· Unit vectors along x,y,z axes
𝑎Ԧ𝑥 , 𝑎Ԧ𝑦 , 𝑎Ԧ𝑧
𝑜𝑟 𝑥,
ො 𝑦,
ො 𝑧Ƹ

· Vector A can be written as

A = Ax a x + Ay a y + Az a z
3
 Vector Operations
1. Vector Addition

C = A+ B
C = A+ B

B
A

1. Addition of Vectors → is a Vector

2. Commutative 3. Associative

A+ B = B + A A + ( B + C ) = ( A + B) + C
4
 2. Vector Subtraction

· First,  
−B: B 
B 
Same magnitude −B
Opposite direction

· Then, Subtraction of vectors is,

D = A − B = A + (− B )

5
 Component Form of vector

A = Ax a x + Ay a y + Az a z
B = Bx a x + B y a y + Bz a z

Vector Addtion C = A+ B

՜
𝐶 = (𝐴𝑥 + 𝐵𝑥 ) ՜
𝑎 𝑥 + (𝐴𝑦 + 𝐵𝑦 ) ՜
𝑎 𝑦 + (𝐴𝑧 + 𝐵𝑧 ) ՜
𝑎𝑧

= 𝐶𝑥 ՜
𝑎 𝑥 + 𝐶𝑦 ՜
𝑎 𝑦 + 𝐶𝑧 ՜
𝑎𝑧
6
 • Two Vector Multiplication operations:
3. Dot product

Definition:
A  B = AB cos θ

1) Result of Dot Product is Scalar.

2) Geometrically defined as the product of magnitudes of A and B
and the cosine of the angle between them
3) dot product satisfies

Commutative: A B = B  A
Distributive: A  (B + C) = A  B + A  C
7
Scaling: k ( A  B) = (k A)  B = A  (k B)
Ex) Dot product of unit vectors

a x  a x = 1, a y  a y = 1, az  az =1
ax  ay = ay  az = az  ax = 0
• In Component Form,

( )(
A B = Ax a x + Ay a y + Az a z  Bx a x + By a y + Bz a z )
= (A a + A a + A a ) B a + (A a
x x y y z z x x x x )
+ Ay a y + Az a z  B y a y
+ (A a + A a + A a ) B a
x x y y z z z z

= Ax Bx + Ay B y + Az Bz 8
 • Scalar projection, Vector projection

· Scalar projection of · Vector projection of

   
A onto B onto
A B

A B
A cos θ =
B
= A aB AB = A cos θ a B =
A B
B
( )
aB = A aB aB
9
Component of A along x direction , Ax, is then...

Ax = Scalar projection of A
A ax
Ax = A cos θ = = A ax
ax

A = ( A  a x )a x + ( A  a y )a y + ( A  a z )a z

= AX a x + AY a y + AZ a z

A = AX + AY + AZ
2 2 2
Magnitude:

10
 4. Cross product

· Definition

C = A  B = AB sin θ a n
1) is a Vector whose

Magnitude: the area of parallelogram formed by A and B

Direction: in the direction of advance of right-handed screw as A is turned
into B

2) satisfies
՜ ՜ ՜ ՜
𝐴𝑛𝑡𝑖 − 𝐶𝑜𝑚𝑚𝑢𝑡𝑎𝑡𝑖𝑣𝑒: 𝐴 ×𝐵 =−𝐵 ⋅𝐴
՜ ՜ ՜ ՜ ՜ ՜ ՜
𝐷𝑖𝑠𝑡𝑟𝑖𝑏𝑢𝑡𝑖𝑣𝑒: 𝐴 × ( 𝐵 + 𝐶 ) = 𝐴 × 𝐵 + 𝐴 × 𝐶
՜ ՜ ՜ ՜ ՜ ՜
𝑛𝑜𝑛 − 𝑎𝑠𝑠𝑜𝑐𝑖𝑎𝑡𝑖𝑣𝑒 𝐴 × (𝐵 × 𝐶 ) ≠ (𝐴 × 𝐵 ) × 𝐶
11
Ex) Cross Product of Unit Vectors

ax  ax = ay  ay = az  az = 0
ax  ay = az, a y  az = ax, az  ax = ay

• In Component Form,

( )(
C = A  B = Ax a x + Ay a y + Az a z  Bx a x + By a y + Bz a z )
= (Ay Bz − Az B y )a x + ( Az Bx − Ax Bz )a y + (Ax By − Ay Bx )a z

ax ay az
C = A  B = Ax Ay Az
Bx By Bz Prob.1.2
 • Position vector, Displacement vector
and Separation Vector
1. Position vector:
Vector directed from origin O to the position P1

r = xa x + y a y + z a z
2. Displacement vector:
directed distance from the initial to the final position
Ex) Magnitude: distance from P1 to P2
Direction: parallel to P1P2
13

∆𝑟Ԧ = 𝑟Ԧ2 − 𝑟Ԧ1

3. Separation vector :
directed distance between two separate points,
(for example) source and field points 𝑅 = 𝑟Ԧ2 − 𝑟Ԧ1
 < Coordinate system>
Cartesian coordinate system

Constant coordinate surfaces

x = cons tan t
y = cons tan t
ax , a y , az
z = cons tan t
14
 Cylindrical Coordinate system

(x,y,z) → (r,f,z)
r = x2 + y2 x = ρ cos f
y
f = tan −1 ( ) y = ρ sin f
x
z=z z=z 15
 Unit Vectors in Cylindrical Coordinate system

• Unit vectors are perpendicular to …

r = x 2 + y 2 = constant ar
−1 y
f = tan ( ) = constant af
x
z = constant az
• Vector A (at a given point) in cylindrical
coordinate system can be written as

A = Ar a r + Af af + Az a z
16
• operations ? aρ aρ =1 af  af = 1 az  az = 1
a ρ  af = 0 af  a z = 0 az aρ = 0

aρ aρ = 0 af  af = 0 az  az = 0
a ρ  af = a z af  a z = a ρ a z  a ρ = af

• Addition, sub A  B = ( Aρ  Bρ )a ρ + ( Af  Bf )af + ( Az  Bz )a z

• Dot Product A  B = Aρ Bρ + Af Bf + Az Bz

• Cross Product aρ af az
A  B = Aρ Af Az
17
Bρ Bf Bz
Spherical coordinate system
( x, y, z ) → ( r, q, F )
x = r sin θ cosf
y = r sin θ sin f
z = r cos θ

r = x2 + y2 + z 2
z
q = cos−1[ ]
r
y
f = tan −1[ ]
x
18
Unit Vectors

• Unit vectors are perpendicular to

ar r = x2 + y2 + z 2
af z
q = cos−1[ ] = constant
aq r
y
f = tan −1[ ]
x

19
• Operation between unit vectors

ar  ar = 1 aθ  aθ = 1 af  af = 1
a r  aθ = 0 a θ  af = 0 af  a r = 0
a r  a θ = af a θ  af = a r af  a r = a θ

20
 < Differential Calculus >
• Differentiation of scalar function

• For 1D scalar function, • For 3D scalar function,

f = f (s )

f(s)
f = f ( x, y , z )
f

s s

df ( s) f ( s + s) − f ( s) f f ( x + x, y, z ) − f ( x, y, z )
= lim = lim
ds s →0 s x x→0 x
 • Differentiation of a vector function (field)

· Derivative of a vector function with respect to x, F ( x, y , z )

F F ( x + x, y, z ) − F ( x, y, z )
= lim
x x→0 x
 Fx ( x + x, y, z ), Fy ( x + x, y, z ), Fz ( x + x, y, z )  −  Fx ( x, y, z ), Fy ( x, y, z ), Fz ( x, y, z ) 
= lim 
x → 0 x
 F ( x + x, y, z ) − Fx ( x, y, z ) Fy ( x + x, y, z ) − Fx ( x, y, z ) Fz ( x + x, y, z ) − Fx ( x, y, z ) 
= lim  x , , 
x → 0
  x  x x 
 F ( x + x, y, z ) − Fx ( x, y, z ) Fy ( x + x, y, z ) − Fx ( x, y, z ) F ( x + x, y, z ) − Fx ( x, y, z ) 
=  lim x , lim , lim z 
 x →0 x x → 0 x x → 0 x 
 F Fy Fz 
= x, , 
  x x x 

22
 Del operator
<Question>
When x,y,z, changes by dx, dy, dz, how much does function f(x,y,z) change?

f f f
df = dx + dy + dz
x y z
 f f f 
=  a x + a y + a z   [dx a x + dy a y + dz a z ]
 x y z 

 f f f 
=  ax + a y + az   d
 x y z  Infinitesimal
displacement vector

23
• If we divide both sides of df by dl,

𝑑𝑓 𝜕𝑓 ՜ 𝜕𝑓 ՜ 𝜕𝑓 ՜ 𝑑ℓ
= 𝑎 + 𝑎 + 𝑎 ⋅
𝑑ℓ 𝜕𝑥 𝑥 𝜕𝑦 𝑦 𝜕𝑧 𝑧 𝑑ℓ

՜
= 𝑁⋅՜
𝑎ℓ Unit vector parallel to dl

where f f f
N = f = a x + a y + a z
x y z

  
Del operator:  = ax + ay + az
x y z
Vector Differential Operator 24
 Gradient of scalar function
f f f
N = f = ax + a y + az
x y z
df = f  d = |∇f|dl cos𝜃
1. Direction:
points in the direction of maximum increase of f !!!

Direction: normal to f(x,y,z)=constant

ex) f ( x, y , z ) = x 2 + y 2 + z 2
f f f
f = ax + a y + az
x y z f=constant
25

= 2 xa x + 2 ya y + 2 za z = 2r
2. Magnitude:
Slope along the MAX increase direction!!

• If 𝑑ℓ infinitesimal displacement vector parallel to ∇𝑓

df = f  d = |∇f|dl cos𝜃
df
= f  a  = f
d

f = (df / dl )max
Direction and magnitude of Maximum increase!!
26
 Gradient in cylindrical and spherical coordinate system

• For Cartesian coordinates

  
 = ax + ay + az
x y z
• For cylindrical coordinates
f 1 f f
f = ar + af + a z
r r f z

• For spherical coordinates

f 1 f 1 f
f = a r + aq + af
r r q r sin q f
27
 Divergence of vector function
• Def of Divergence: Scalar, only with magnitude
Fx Fy Fz
F = + +
x y z
• For cylindrical coordinates
1  1  
F = [ rFr ] + [ Ff ] + [ Fz ]
r r r f z
• For spherical coordinates
1  2 1  1 
 F = [ r F ] + [sin q Fq ] + [ Ff ]
r r r sin q q r sin q f
2 r

- But, what does it mean to take the divergence of a vector field?

(later) 28
 Curl of vector function

• Def of Curl: vector with magnitude and direction

 Fz Fy   Fx Fz   Fy Fx 

curl F =  −  ax +  −  ay +  −  az
 y z   z x   x y 

   
= a x + ay + a z   [ Fx a x + Fy a y + Fz a z ]
 x y z 
=  F
Or in determinant form,
ax ay az
  
 F =
x y z
29
Fx Fy Fz
• For cylindrical coordinates • For spherical coordinates

ar r af az ar r aq r sin q af
1    1   
 F =  F = 2
r r f z r sin q r q f
Fr rFf Fz Fr r Fq r sin q Ff

- But, what does it mean again taking the curl of vector field?
(later...)
30
 Second Derivatives
Laplacian operator

1. Laplacian of Scalar field

 2
 2

where  = 2 + 2 + 2
2

 2 f =   (f )
2

x y z
2 f 2 f 2 f
 f =   (f ) = 2 + 2 + 2
2

x y z
1  f 1 2 f 2 f
 f =
2
(r ) + +
r r r r 2 f 2 z 2
1  2 f 1  f 1 2 f
 f = 2
2
(r )+ (sin q )+ 2 2
r r r r 2 sin q q q r sin q f 2

2. Laplacian of Vector field

 2 F = a x  2 Fx + a y  2 Fy + a z  2 Fz
31
Show!!  F = (  F ) −   (  F )
2
 Some Properties of Del operator

• Products including a scalar

( fg ) = f g + g f
  ( f A) = f   A + A  f
  ( f A) = f   A + f  A

• Vector products

A  ( B  C ) = B  (C  A) = C  ( A  B) A  ( B  C ) = B( A  C ) − C ( A  B)

  ( A  B) = B  (  A) − A  (  B)

  ( A  B ) = A   B − B   A + ( B  ) A − ( A  ) B
32
 예제

• Show for continuously differentiable scalar and vector functions

f and A,

  (f ) = 0

  (  A) = 0

33
 < Integral Calculus >
Line, Surface, and Volume integral
• Line integral of 1-D scalar function
n
f ( x) dx = lim  f i xi
b
a n →
xi →0 i =1

1. Subdivide into infinitesimally small segments

2. Calculate each surface area of rectangles
3. Finally, add them up → integration

• In general, line integral in 3-D scalar function

n

 f d = lim  f
c n →
 i →0 i =1
i  i
n

 f d = lim  f 
c n →
 i →0 i =1
i i 34
 Line integral of a vector field
• Scalar line integral
F (r i )
Integral of tangential component along C
n

 F  d = lim  F
c n →
 i →0 i =1
i   i

 F  d =
L L
F cosqdl

• Vector line integral

 F  d = lim  F
c n →
 i →0 i =1
i   i

35
 Differential displacement

• rectangular coordinates • cylindrical coordinates • spherical coordinates

dl
dl dl

d = dr a r + r df af + dz a z
d = dx a x + dy a y + dz a z d = dr a r + r dq aq + r sin q df af
36
 Example
• An object rotates around z axis with radius
z
a under external force as given below.
Calculate the work done by the force per
rotation. a

F = (5 N )a r + (3N )af + (2 N )a z

՜
𝑊 = න 𝐹 ⋅ 𝑑ℓ = 6𝜋𝑎 (𝐽)
𝑐

37
 Surface integral
• For Vector field A defined on a smooth surface S

Surface integral
n

 A  ds = lim  A s
s n →
s i →0 i =1
i i

n
= lim  Ai cosqSi
n →
s i →0 i =1

=  A cos qdS
s

1. Divide into infinitely many, infinitesimal surfaces

2. Calculate scalar product,
3. Then add them up… 38
• In addition, we can also do

Scalar surface integral Vector surface integral

 F  ds = lim  F
n
i  s i  F  ds = lim  F i  s i
s n → s n →
s i →0 i =1 s i →0 i =1

39
 Here, Differential normal surface area

• rectangular coordinates • cylindrical coordinates • spherical coordinates

ds x = dy dz a x ds r = r df dz a r ds r = r 2 sin q dq df a r
ds y = dx dz a y dsf = dr dz af dsq = r dr sin q df aq
ds z = dx dy a z ds z = r dr df a z dsf = r dr dq af
40
 Example:

• Calculate the scalar surface integral of the vector fields below for
the surface of a sphere of radius R centered at origin.

A = r 2 ar

A = cos q ar

41
 Volume integral

• Volume integral of Scalar field f or Vector field F

 f dv = lim  f
v n →
vi →0 i =1
i vi

F dv = lim  F i  vi
v n→ 
 vi → 0 i =1

42
 Here, Differential volumes are

• rectangular coordinates • cylindrical coordinates • spherical coordinates

dv = dx dy dz dv = r dr df dz dv = r 2 dr sin q dq df

43
Example:

• What’s the volume of a sphere of radius R?

𝑉 = න 1 𝑑𝑣 𝑑𝑣 = 𝑟 2 𝑑𝑟 sin 𝜃 𝑑𝜃 𝑑𝜑
𝑉

𝜋 2𝜋 R
= න sin 𝜃 𝑑𝜃 න 𝑑𝜑 න 𝑑𝑟 2 𝑑𝑟
0 0 0

4𝜋R3
=
3
44
 < Divergence Theorem >
• Taking Divergence: Spreading out of vectors from a point !!

1 F
div F = lim
v →0 v
S
F  ds =   F
v
Divergence of vector field F at a given point P
= Outward flux per unit volume as the volume shrinks about P.

45
  F  ds = ?
S

• For surfaces perpendicular to x axis.…

1) x + 1/2 x surface
 Fx x 
F
 x x+  y z + higher
 2
+ 2) x - 1/2 x surface
 F x 
−  Fx − x  y z + higher
 x 2
Fx Fx
= xy z = v
x x
• If we do the same for y and z axis directions,
 Fx Fy Fz 
s F  ds =  x + y + z  v
1
= lim
v →0 v
S
F  ds =   F 46
 Divergence theorem

◼ For an arbitrary closed surface where vector fields are

defined

Total outward flux of a vector field through the closed surface S

= volume integral of the divergence of field

   F dv =  F  ds
v s

S
47
 Proof
1) Subdivide volume into large No. of small cells vi
1
  F i = lim
v →0 v
S
Fi  ds

2) Flux through ith cell

 F  ds =   F
si
i vi +  i vi
3) Summing the flux through all the cells
n n n 0
lim   F  ds = lim   F i vi + lim   i vi
n → si n → n →
i =1 i =1 i =1

 F  ds =    F dv
v
S

   F dv =  F  ds
v s
48
 Example

• For a vector field r = (x,y,z), show that the divergence theorem

holds for a closed spherical surface centered at origin and of radius a.

 r  ds =  aa  ds ar = a  ds = 4a 3
r
s s s

   =    =  =  3
r dv r dv 3 dv 4 a
v v v

49
 Stokes Theorem
• Taking curl: Measure TWIST of vector field(function)

• Curl of a vector field is a vector

1 an
(curl F )  a n = lim  F  d
s →0 s c
s
c

Magnitude: Maximum circulation of vector per unit area as the area tends to 0

Direction: Normal direction of the area which makes the circulation Maximum
50
 Derivation: Curl of a Vector field
• Line- integral along a small closed loop

 F  d =  F  d +  F  d +  F  d +  F  d
c c1 c2 c3 c4

• Using Taylor expansion:

Along c2
 Fy 
Fc2  d = c2 Fy ( x + x, y, z)a x  dyax = Fy ( x + x, y, z)y =  Fy + x x y
Along c4

 F  d = 
c2 c2
 
Fy ( x, y, z )a x  (−dy a x ) = − Fy ( x, y, z )y = − Fy y

Summing up 1  Fy  Fy

= lim  xy =
s →0 s
 x  x 51
• In the same way,
1  Fx  F
we obtain c1 and c3. ➔ = lim  xy = x
s →0 s
 y  y
• Adding up,
1 1  Fy Fx  Fy Fx
s →0 s c
lim F  d = lim  − xy = −
s →0 s  x y  x y

Fy Fx
(curl F )  a z = (curl F ) z = −
x y

• x, y components of curl are obtained in the same way..

 Fz Fy   Fx Fz   Fy Fx 

curl F =  −  ax +  −  ay +  −  az
 y z   z x   x y 
52
 Stokes Theorem
• For a vector field defined on an arbitrary closed path

 (  F ) ds =  F  d
s c

Surface integral of the curl of the

vector field over open surface S

= Circulation(Line integral around a closed path)

of the vector field around boundary? path C
53
 Proof

1) Subdivide the open surface into small cells. For the ith cell,

(  F )  a s si =  F  d + i si
ci
2) Summing over all cells
n
 n n

lim 
si →0 i =1
(  F )  s i = lim   F  d +   i si 
si →0  i =1
ci
i =1 
n → n →

=  (  F ) ds =  F  d
s ci

Therefore

 (  F ) ds =  F  d
s c
54
 Example

• For the following vector field and a circle of radius a, show that
Stoke’s theorem holds.

F = r cos f a r + sin f af

55
 Classification of Vector Fields

Solenoidal Irrotational
 
F = 0,  F =0
 
F  0,  F =0
 
F = 0,  F 0
 
F  0,  F 0
◼ Class I field: electrostatic fields in charge-free medium and
magnetic fields in current-free medium
◼ Class II fields
◼ Electrostatic fields in a charged region
◼ Class III fields
◼ Magnetic field within a current-carrying conductor
◼ Class IV fields
56
◼ Hydrodynamic fields in a compressible medium
 < Dirac Delta Function >

• Consider vector field 1

F = 2 rˆ
r
- Divergence theorem says,

   F dv =  F  ds
v s
0 1 1
1  2 1
s R 2 r  dsr = R 2 sds = 4
ˆ ˆ
F = 2 (r 2 ) = 0
r r r

?? What’s wrong? F  0 at origin 57

 ՜
∇⋅ 𝐹
1. =0 everywhere except at origin

2. When Integrated ➔ finite value, not 0

Dirac delta function

1-D Dirac Delta Function f(x)

a)
 ( x) = 0 if x  0 b) 

 if x = 0
&
−
 ( x) dx = 1
58
 Properties of delta function

• For ordinary function f(x)

f ( x) ( x) = f (0) ( x)
 
 f ( x) ( x) = f (0)   ( x) = f (0)
Then
− −

Also therefore

 ( x − a) = 0 if x  a 

 if x = a
&  −
 ( x − a) f ( x)dx = f (a)

Ex14, Ex15 59
 3D Dirac Delta Function

• Delta function in 3D

 3 (r ) =  ( x) ( y ) ( z )
• Therefore
  
  (r )d r =   ( x)dx   ( y )dy   ( z )dz = 1
3
allspace − − −


allspace
 3 (r − a) f (r )d r = f (a)

1 1 
• For F = 2 rˆ   2 r̂  = 4 3 (r )
r r 
1 1  1 2 1 
 (r ) =
3
   2 r̂  = −   
4 r  4  r 
60
will be used very, very often!!!