Electromagnetic Fields and Propagation

Lecture 2

Fabrizio Di Pasquale
Scuola Superiore Sant’Anna
Pisa

Lecture content
Maxwell’s Equations

 Cartesian coordinate, vector differential operators, divergence & Stokes theorems

 Maxwell’s Equations in integral and differential forms
 Gauss’ Law
 Gauss’ Law for the magnetic field
 Faraday’s Law
 Ampere’s Law
 Maxwell’s Equations and light propagation
 Electrostatic, Magnetostatic, Electroquasistatic and Magnetoquasistatic

1
 The concept of Electric Field
Force between electric charges
q1q2
Coulomb’s experiment : f K force magnitude between two electric charges
r 2
q1 and q2 are the charge strengths, r is the distance,  is a constant representing the
effect of the medium and K a constant depending on the units.

In the SI unit system K=1/4 and  for vacuum is:  0  8.854 1012  1 109 F
36 m
For dielectric materials =r0 where r is the relative permittivity or dielectric constant
q1q2 
Direction information is included as: f  r r is a unit vector pointing from
4r 2 one charge to the other
The electric field can be introduced as a force per unit charge for each point of a
region influenced by charges. We introduce a charge test q small enough not to
disturb the charge distribution.
f (coulomb)(meter) Volt
The electric field is defined as: E E 
q
2
meter
( farad )(meter )
The electric field arising from a Electric Field lines
point charge in a homogeneous dielectric is: E 
q  around two
2 r 3
4r opposite charges

The concept of Magnetic Field

Magnetic forces may arise either from permanent magnets or from current flow.

The magnetic flux density B (vector field) is defined in terms of the force produced on
a small current element of length dl, such that:

df  Idl  B or df  IdlB sin 

where  is the angle between dl and B
f  q(E  v  B) Lorentz force law
f  q E  I dl  B
The magnetic field vector (or magnetic field intensity) is related to the magnetic flux
density: B: Tesla (Wb/m2)
B   H H: A/m

In free space:   0  4  10 7 H / m Henry is the unit of inductance

4

2
 Maxwell’s Equations
GENERAL FORM OF TIME VARYING MAXWELL’S EQUATIONS
(“Point” or Differential Form)

  Ε   B
t
  H  D  J
t
D  ρ
B  0
James Clerk Maxwell (1831-1879)

Physical Quantities, SI Units

QUANTITIES SI or MKS

E is the electric field intensity Volt/m

H is the magnetic field intensity Amp/m

D is the electric flux density Coul/m2

(also called displacement vector or electric induction)
B is the magnetic flux density Wb/m2 (Tesla)

J is the electric current density A/m2

 is the electric charge density Coul/m3

3
 Electromagnetic Field Sources

The sources of the electromagnetic fields are the electric current density J and the
electric charge density = (x,y,z,t) (macroscopic description: fields as continuos
functions of space and time)

charge in volume V

A magnetic current density could be introduced as a fictitious source for

mathematical convenience

However the real source of a magnetic current is always a loop of electric current
or a similar type of magnetic dipole (there is not flux of magnetic charge as magnetic
monopole charges do not exist)

We will not need to consider fictitious magnetic current sources for studying
propagation in optical fibers and waveguides.

Since electric current is really the flow of charge, the electric charge  is the
ultimate source of the electromagnetic field
7

Conduction Current Density (Ohm’s Law)

J  E A/m2 for a conductor where  is the conductivity in Siemens/meter

Convection Current Density

A/m2 for a charge density  moving with velocity v
J  v  unit is C/m3, v is in m/s, J is C/(m2s) i.e. A/m2

Displacement Current Density

D A/m2 it is important for example within a dielectric capacitor with V
t changing with time; it is negligible for many low frequency problems
(in good conductor it is negligible up to high frequencies !)

4
 Cartesian coordinate system

P In a Cartesian coordinate system for a

three-dimensional space the coordinates of
a point P can be taken as the distances
from P to the three planes defined by the
three axes.

VECTORS IN CARTESIAN COORDINATE SYSTEM A ( x, y , z , t )

A This notation represents a time varying vector field function of the
spatial coordinates x,y,z and the time variable t

Same mean with

different notations
for the unit vectors !
9

Cylindrical coordinate system

This coordinate system is important for studying
the propagation in optical fibers (optical fibers are z
cylindrical structures, invariant along the z axis)

.
z0

r0 y
x
0

VECTORS IN CYLINDRICAL COORDINATE SYSTEM

Same mean with

different notations
for the unit vectors !
10

5
 Spherical coordinate system

VECTORS IN SPHERICAL COORDINATE SYSTEM

Same mean with

different notations
for the unit vectors !

11

Vector differential operators

Rectangular coordinates
  x  y  z
  
Gradient (vector operator)
x y z
 f  f  f
Gradient of a scalar function f : f  x  y  z
x y z
The gradient of the scalar function f is a vector showing the direction and
magnitude of the maximum spatial variation of the scalar function.
The change in f, df, is given by the scalar product of the f and the vector dl,
so that for a given element of length dl, the maximum value of df is obtained
when that element is oriented to coincide with the direction of the gradient vector:
df  f  dl   A  (  x   y   z )  ( Ax x  Ay y  Az z ) 
     
x y z
Divergence of a vector field A
A A A
 x y z
x y z
The divergence of a vector field provides a description of the manner in which
the field varies in a point. It is the amount of flux per unit volume emerging from
an infinitesimal volume at a point:
Divergence Theorem
 Ad S
Intrinsic definition:   A  lim V 0 S
V
   Adv   A  d s
V S
12

6
 Divergence of an electrostatic field
D
Gauss Law in integral form:
 (x, y, z)

 D  d s   dV  Q
S: closed surface
surrounding a charge
distribution
S V

“the total electric flux coming out from a closed surface is equal to the total charge
enclosed by that closed surface”

 DdS  dV
lim
V  0
S
V
 lim
V  0 V
V
 D  
Gauss Law in differential form
Divergence of the electric flux density

Consider an infinitesimal volume V

as a rectangular parallelepided of
dimensions x, y, z
13

Gauss’s Law: physical understanding

D  

14

7
Gauss’s Law: physical understanding

15

Gauss’s Law: physical understanding

D  

16

8
 Gauss’s Law: physical understanding

Divergence Theorem

17

Gauss’ Law – Integral Form

 D  d s   dV  Q
S V
D

 ( x, y, z ) Carl Friedrich Gauss

(1777,1855 )
S: closed surface
surrounding a charge
distribution

Gauss’ Law: the total electric flux coming out of a closed surface is equal to the
total charge enclosed by that closed surface

Charges are “sources” or “sinks” of the electric field (charges produce or terminate
electric field lines)
If the total flux is positive the total charge enclosed by the surface is positive
If the total flux is negative the total charge enclosed by the surface is negative 18

9
 Divergence Theorem

For any vector field A( x, y , z , t ) z n

ds

   Adv   A  d s
V S
Surface S
y

x
Volume V

The flux of a vector though a closed surface S is

equal to the integral of the divergence of the vector
taken over the volume V enclosed by that close surface

19

Vector differential operators

Rectangular coordinates
Curl of a vector field

  A  (  x   y   z )  ( Ax x  Ay y  Az z )
     
x y z

  
x y z
A      x ( Az  Ay )  y ( Ax  Az )  z ( Ay  Ax )
x y z y z z x x y
Ax Ay Az

 A A  A A  A A
  A  x( z  y )  y( x  z )  z ( y  x )
y z z x x y

20

10
 Vector differential operators
Rectangular coordinates
Curl of a vector field

 A A  A A  A A
  A  x( z  y )  y( x  z )  z ( y  x )
y z z x x y

The curl of a vector A is defined as a vector function whose component in a point,

in a particular direction i, is found by orienting a small area Si normal to the desired
direction i at that point and finding the limit of the line integral divided by the area:

Intrinsic definition   Ai  lim S i  0

c A d l
Si

where i denotes a particular direction, Si is normal to that direction and the line
integral is taken in the right-hand sense with respect to the positive i direction
21

 (  A )  d s   A  d l
S C
Stokes Theorem

Stokes Theorem

For any vector field A( x, y, z , t )

 (  A)  d s   A  d l
S C
C

The line integral of a vector over a closed

contour C is equal to the surface integral of the
curl of the vector over any arbitrary surface
S bounded by the closed contour
22

11
Curl of a vector field

23

Curl of a vector field

24

12
 Curl of a vector field

Curl and  are equivalent formalisms indicating the curl of a vector field
25

Stoke’s Theorem

The line integral of a vector over a closed

contour is equal to the surface integral of the
curl of the vector over any arbitrary surface
26
S bounded by the closed contour

13
 Stoke’s Theorem

  H  D  J
t

 H  d l  t  D  d s   J  d s  t  D  d s  I
C S S S

Ampere’s Law: the line integral of the magnetic field over a closed contour C is
equal to the total current plus the time rate of change of the total electric flux that
goes through any arbitrary surface bounded by the closed contour

 Electric currents and time changing electric fields can generate magnetic fields
(this is the only known way to generate magnetic fields)
 The positive direction for the surface normal vector ds and of the contour are
related by the right hand rule
27

Vector differential operators

Rectangular coordinates
 f  f  f
2 2 2

Laplacian operator  f  2  2  2
2

x y z

Vector Laplacian   
operator  2 A  x 2 Ax  y 2 Ay  z  2 Az

 2 A    A      A

28

14
 Vector differential operators
Cylindrical coordinates

 f  f  f
Gradient of a scalar function f : f  r  z
r  z

A A
Divergence of a vector field A :   A  1  (rAr )  1   z
r r r  z

 Az A  A A   (rA ) Ar

Curl of a vector field A :   A  r (1  )  ( r  z )  z 1(  )
r  z z r r r 

29

Vector differential operators

Cylindrical coordinates

f  f  f
2 2

 f  1  (r )  12
2

r r r r  2 z 2

 2 A    A      A

30

15
 Vector differential operators
Spherical coordinates


 f  1 f  f
Gradient of a scalar function f : f  r  
r r  r sin  

1  2 1  1 A
Divergence of a vector field A :   A  2 r (r Ar )  r sin   (sin A )  r sin  
r

Curl of a vector field A :

  
A  r [  ( A sin  )  A ]   [ 1 Ar   (rA )]   [  (rA )  Ar ]
r sin     r sin   r  r r  
31

Vector differential operators

Spherical coordinates

f  (sin  f )   f
2

 f  12  (r 2 )  2 1
2 1
r r r r sin    r sin   2
2 2

 2 A    A      A

32

16
 Some Vector Identities
Divergence of the curl:

Curl of the gradient:

Divergence of the gradient:

Curl of the curl:

Distributive property:

Vector cross product:

Product of a scalar and a vector:

33

Solenoidal and irrotational Vectors

Solenoidal vector fields

vector potential

Irrotational vector fields

A conservative vector field is a vector field that is the gradient of a scalar function

As every conservative vector field is also an irrotational vector field

34

17
 Maxwell’s Equations
MAXWELL’S EQUATIONS ARE NOT INDEPENDENT OF EACH OTHER !

Taking the divergence of   Ε   B since  Ε  0

t

  B  cos t B  0 if fixing the constant to zero !

CONTINUITY EQUATION:
 H  0
Taking the divergence of   H  D  J since
t D  
 Continuity Equation
J  0
t Charge is conserved ! Current is continuos !

J represents the outflow of current at a point and /t represents the charge
build-up with time at the same point.
35

Continuity Equation
CONTINUITY EQUATION:
 H  0
Taking the divergence of   H  D  J since
t D  
 Continuity Equation
J  0
t Charge is conserved ! Current is continuos !


V t
dV     JdV  0
V


t V
dV   J  ndS J represents the outflow of current at a
point and /t represents the charge
S build-up with time at the same point.
Q
 I
t
The charge variation in a volume is equal to the current entering the volume
Electric charges can neither be created nor destroyed (a charge variation can only
be due to a flux of current through the surface)

18
 Conservation of Charge
The concept of Displacement Current

Although the charge density  varies with space or time, we expect that the total
charge is conserved.

If current flows out of any volume the amount of charge inside must decrease and if the
currunt flows in, the amount of charge inside increases.

Considering a smaller and smaller volume we get: J  
t  ( x, y, z)

If we consider the equation from magneto-static :   H J

we achieve :  J   H  0
This does not agree with the continuity argument and conservation of charge !

Maxwell recognized that eq. H  J from magneto-static was not complete for

time-varying fields and postulated an added term  D   H  D  J

t t

In this way continuity is satisfied   J    (  D )   37
t t

Conservation of Charge

The concept of Displacement Current

The term  D contributes to the curl of the magnetic field as the actual conduction
t
current density (motion of charge in conductors) or convection current density
(motion of charge in space) and it is called displacement current term.

Displacement current is negligeble with respect to conduction current in good

conductors up to optical frequencies

Displacement current becomes however important in many situations as the

frequency of the time varying phenomena is increased.

For example the displacement current is important within the dielectric of a capacitor
whenever the capacitive voltage changes in time

38

19
 Maxwell’s Equations in Free-Space

In free space the following relations are valid between the electric and magnetic
field intensities and flux densities :

B  0 H 0  4  107 Henry/m (permeability of free space)

D  0 E  0  8.854  10 12
Farad/m (permittivity of free space)

 0 H
Ε 
t
 E
H  0  J
t
 0 E  ρ
  0 H  0 39

Maxwell’s Equations
INTEGRAL AND DIFFERENTIAL FORMS

Using the divergence theorem:

   Adv   A  d s
V S

and the Stokes’ theorem:  (  A)  d s   A  d l

S C

Differential Form Integral Form

D  ρ Gauss’ Law  D  d s   dV  Q

S V
total charge contained
in the volume V
B  0 Gauss’ Law  Bd s  0
S

  Ε   B Faraday’s Law 
 E  d l   t  B  d s
t C S

  H  D  J Ampere’s Law  
C H  d l  t S D  d s  S J  d s  t S D  d s  I
t

I  Jds : Total electric current flowing through the surface S 40

S

20
 Gauss’ Law – Integral Form

 D  d s   dV  Q
S V
D

 ( x, y, z ) Carl Friedrich Gauss

(1777,1855 )
S: closed surface
surrounding a charge
distribution

Gauss’ Law: the total electric flux coming out of a closed surface is equal to the
total charge enclosed by that closed surface

Charges are “sources” or “sinks” of the electric field (charges produce or terminate
electric field lines)
If the total flux is positive the total charge enclosed by the surface is positive
If the total flux is negative the total charge enclosed by the surface is negative 41

Gauss’ Law for the Magnetic Field

Integral Form

Bds  0
S

S: closed surface of arbitrary shape

Gauss’ Law for the magnetic field: the total magnetic flux coming out of
a closed surface is always zero

Magnetic charges do not exist (they can not generate or terminate magnetic
field lines)
If the magnetic field is not zero, then the flux into any closed surface must be equal
to the flux out of it, so that the net flux coming out is zero 42

21
 Faraday’s Law – Integral Form

 E  d l   t  B  d s
C S

C: closed contour
Michael Faraday
(1791-1867)

Faraday’s Law: the line integral of the electric field over a closed contour C is
equal to the negative of the time rate of change of the total magnetic flux that goes
through any arbitrary surface bounded by the closed contour

 A time changing magnetic field can generate electric fields

 The positive direction for the surface normal vector ds and of the contour are
related by the right hand rule

43

Ampere’s Law – Integral Form

 H  d l  t  D  d s   J  d s  t  D  d s  I
C S S S

D
J electric
electric flux
current density
A.M. Ampere density C: closed contour
(1775-1836)

Ampere’s Law: the line integral of the magnetic field over a closed contour C is
equal to the total current plus the time rate of change of the total electric flux that
goes through any arbitrary surface bounded by the closed contour

 Electric currents and time changing electric fields can generate magnetic fields
(this is the only known way to generate magnetic fields)
 The positive direction for the surface normal vector ds and of the contour are
related by the right hand rule
44

22
 Displacement current
Physical Picture
Consider a circuit including an ac generator and a capacitor:

b
S2

 H  d l  t  D  d s   J  d s  t  D  d s  I
a c

S1
d C S S S

The line integral of the magnetic field around the loop a-b-c-d
should be the current through any surface of which the loop is a boundary !

For S1 which cuts the wire, a finite value is obtained for the line integral (conduction
current).
For S2 which does not cut the wire but passes through the plates of the capacitor,
the line integral would be zero if we only consider conduction current (static analysis).

The displacement current term is the one which preserves the continuity of the
current between the plate of the capacitor.
45

Displacement current
Physical Picture
Consider an ideal parallel plate capacitor of capacitance C with applied voltage:

A
V0 sin t
d A

The charging current is: I c  C dV  CV0 cos t (from circuit model theory)
dt
Field inside the capacitor: E=V/d so the displacement current density  D is:
t
V
J d   E   0 cos t
t d
The displacement current flowing between the two plates is the current density
multiplied by the area of the plates:
I d  AJ d   ( A )V0 cost  CV0 cos t
d
The total displacement current between the capacitor plates is exactly the same as
the value of the charging current calculated by the circuit model theory
46
(the displacement current acts to complete the circuit)

23
 Maxwell’s Equations and Light
Propagation
In free-space :
 0 H
Ε 
  0 E  ρ t

  0 H  0
 0 E
H  J
t
 Maxwell’s equations are linear but not independent of each other
 Time varying electric and magnetic fields are coupled
 This coupling explains the propagation of electromagnetic waves
Assuming   J  0 in free space:
0 H    H c 1  3  108 m / s
  0 0  E
2
  (  Ε)    ( ) 0 0 0
t t t 2

  (  Ε )   12  E
2
Equation for a wave traveling at the speed c
c t 2 47

Electrostatic and Magnetostatic

We consider time independent situations

(stationary condition  0 )
t
Equations for Electric and Magnetic Fields are independent !

Equations for Electrostatic Equations for Magnetostatic

 0 E  ρ H  J
Ε  0   0 H  0
 Electric fields are only produced  Magnetic fields are only produced
by electric charges by electric currents

 The Electric Field has to be  The Magnetic Field has to be

determined given some charge determined given some current
distribution distribution
48

24
 Electroquasistatic and
Magnetoquasistatic
We consider “slow” time variations

Equations for Electroquasistatic Equations for Magnetoquasistatic

 0 E  ρ   0 H  0
Ε  0 H  J
 0 E  0 H
H  J Ε 
t t
 Electric fields are only produced  Magnetic fields are only produced
by electric charges by electric currents
 Once the Electric Field is known  Once the Magnetic Field is known
the Magnetic field is determined the Electric field is determined
by the last equation by the last equation 49

Electroquasistatic and
Magnetoquasistatic
The question is: when can we consider a phenomenum to be “slow” ?

Time variations are considered slow if the time scales over which things
are changing are much longer than the time taken by the light to travel
over the length scale of the problem !
Electromagnetic waves at frequency f are related to the speed of the wave
c by the relation:

f  c
If L is the length scale of the problem and T its time scale (T~1/f) the condition
for quasistatic analysis is:
T  L
c
The wavelength of the wave at the
cT  L
operating frequency is much longer
c  L
f than the length scale of the problem !

  L 50

25
 Electroquasistatic and
Magnetoquasistatic
EXAMPLE

Amplifier chip operating from 100 MHz to 10 GHz

 100 MHz operation: Time scale of the problem (1/100 MHz) = 10 nsec
Length scale of the problem 3 cm
Time taken by the light to travel 3 cm: 0.1 nsec
Since 10 nsec >> 0.1 nsec quasistatic can be used (: 3m >> 3 cm)

 10 GHz operation: Time scale of the problem (1/10 GHz) = 0.1 nsec
Length scale of the problem 3 cm
Time taken by the light to travel 3 cm: 0.1 nsec
Quasistatic can not be used (: 3 cm ~ length scale of the problem !)
51

26