Prof.

Alessandro Galli

MICROWAVES
9-ECTS Course of the Master Degree in Electronics Engineering
Lecture 03 – EM Fields and Maxwell’s Equations

AG-01/18
L03 Outline & Intro
LECTURE 03 – EM FIELDS AND MAXWELL’S EQUATIONS
- Electromagnetic fields and classical electrodynamics (03-04) The electromagnetic field is, for the modern
- Fundamental laws of electromagnetism (05-11) physicist, as real as the chair in which he sits.
- Units of measurements for EM quantities (12) (A. Einstein and L. Infeld, ‘The Evolution
- Standard physical formulation and fundamental equations (13-14) of Physics’)
- Formulations based on duality principle & impressed terms (15-17)
- Symbols and constants for EM (18)

War es ein Gott der diese

< Zeichen schrieb!
(L. Boltzmann after W. Goethe, ‘Faust’)

3rd Lecture Art Masterpiece - 3rd Lecture Movie -

‘La terrazza dell’infinito’ at Villa Cimbrone, A. Hitchcock:
Ravello, I (ca. XX c.) ‘Vertigo’ (1958)

The Beatles: ‘Strawberry Fields Forever’ Sting -‘Fields of Gold’ AG-02/18

Electromagnetic Fields: Lorentz Force Equation
The usual applications of Engineering Electromagnetics can efficiently be analyzed by means of a ‘classical’ field modeling
James C. Maxwell
described collectively by the well-known Maxwell’s Equations (‘MEs’).
(1831-1879):
In this context we skip all the details to describe the development of these equations (assumed known from the part of Physics Scottish scientist,
courses on electrology): our main purpose here is just to recall the basic info of their significance in connection with the fundamental he made studies
experiments carried out by several investigators mainly during XIX century and the relevant mathematical formulation. on colour vision,
thermodynamics,
In the following, we refer to EM fields that, in addition to the space dependence (through the coordinates of a point P, or,
and, as is well-known,
concisely, through the position vector r in an arbitrary coordinate system), generally depend also on the time variable (instant t): electromagnetism. His
this is the typical and useful situation described by ‘electrodynamics’ (i.e., in conditions that generally are not static). scientific contributions
To emphasize the time dependence, the quantities here introduced will be represented by script symbols, instead of roman-type can be considered of the
(or block) letters, which will be used again for frequency-domain quantities – cf. 2nd Week). same importance as those
So here we introduce quantities of the type A ( r , t ) etc. of Newton or Einstein.
The possible independence of time of the EM fields is a mathematically idealized particular situation, which gives rise to the well-known
separated branches of electrostatics and magnetostatics, whose basic properties have already been considered in basic Physics courses too.
It is known that, from a physical viewpoint, the EM field is generated by electric charges in rest or in motion. Hendrik Lorentz
The EM field is the fundamental quantity which enables us to quantify the EM forces or interactions. (1853-1928):
To introduce operatively the concept of EM field, we consider the fundamental relationship that allows Dutch
physicist,
for identifying the presence of the ‘electromagnetic force’ through the well-known Lorentz force equation:
F ( r , t )  q E  v  B 
Nobel prize,
he gave
fundamental
The EM force F ( r , t ) acting on a point test charge q located in r at the instant t, generally moving at velocity v, is proportional contributions to
to an electric vector quantity, E ( r , t ) , usually called ‘electric field’ (or, sometimes, ‘electric field strength’, or ‘electric field intensity’) electrodynamics,
and to a magnetic vector quantity, B ( r , t ) , usually called ‘magnetic induction’ (or ‘magnetic flux density’ or also simply ‘magnetic field’). underpinning,
e.g., Einstein’s
The physical dimension of [E] is ‘Force/Charge’ [F/Q] and the unit of measurement (‘UoM’) in the International System (‘SI’) theory of relativity.
is newtons per coulomb (N/C), that is equivalent to volts per meter (V/m), which is the common SI-UoM of E.
It is seen that [B] is dimensionally as [E/v], so that for the SI-UoM it is V∙s/m2 = Wb/m2 = T, where Wb = V∙s = weber, and T = Wb/m2 = tesla.
- Additional info on the SI-UoM of EM quantities is listed inside this lecture.

AG-03/18
Classical Electrodynamics
Some considerations are noteworthy about the modeling of EM, generally involving the mathematical idealization of the physical reality.
Lorentz force equation, which operatively allows for the definition of the EM field, is a mathematical idealization of the physical phenomenon.
In fact, to evaluate the ‘exact’ fields in one point, we have in principle to assume that the test charge is dimensionless and also it should have a
negligible value so that it does not modify the EM field it is testing. F ( r, t )
So, we can operatively define, e.g., the electric field in terms of the force on a point infinitesimal electric charge in rest as: E ( r , t )  lim
q0 q
If the charge is in motion, assuming known its exact velocity, there is another force contribution that allows for the definition of the magnetic
field. Incidentally we note that the force due to E is able to give energy by possibly moving the test charge, whilst the force due to B, being always
perpendicular to the movement of the charge (presence of the cross product) can influence only the trajectory of the charge but not its energy.
In nature, we remind that the value of the electric charge is ‘quantized’ and in principle it cannot be arbitrarily small:
the minimum value of the charge can be associated to the quark, i.e., the 1/3 or 2/3 fraction of the electron charge -e = -1.6.10-19 C.
Even considering a ‘small’ charge such as an electron, also its ‘classical’ physical dimension is non-zero, although practically negliglible.
As a reference, the typical dimensions of the atoms are of the order of angstrom (1 Å = 10-10 m), and a rough evaluation of the dimension
of an electron is of the order of 10-14 m.
As will be commented further, when we consider EM interactions in the matter, we have to consider that, on an ‘extremely small’ (‘microscopic’)
scale (that is comparable or less than atomic dimension), the analysis of the EM field is in practice a nonsolvable task.
Spatio-temporal ‘regular’ behavior of the EM fields in a medium can be obtained only if a spatial average on a volume containing a ‘sufficiently
large’ number of atoms of the matter is performed: for most material cases, a typical linear minimum dimension which regularizes the EM field is
of the order of 10-8 m: thus, we can imagine that in a cube of this side we have about 106 atoms to be spatially averaged in their overall ‘smooth’
field effect. This reference minimal dimension can be viewed as the inferior limit for treating EM phenomena with a ‘classical’ field approach.
In Engineering Electromagnetics, all the common devices, being ‘much larger’ than individual charged particles, automatically ‘average’ the
irregular EM fields rapidly varying at microscopic scale, thus allowing for viewing a ‘continuous’ environment for the EM phenomena.
When we have EM interactions with matter below such reference minimal dimensions, classical electrodynamics in terms of fields is generally
no longer useful: in this case the study of EM phenomena usuallly requires more involved approaches based on quantum electrodynamics (QED).

AG-04/18
Electric Flux in Vacuum: Basic Gauss’ Law in Terms of Charges
Fundamental quantitative relationships in electrology go back to well-known experiments on the force among electric charges due to Coulomb.
Electrostatic studies emphasized that an electric charge of proper sign is a source (or sink) of electric fields Pierre-Augustin
(electric field lines diverge/converge from a positive/negative electric charge). de Coulomb
(1736-1806)
Experimentally it was then established a fundamental integral relationship, known as electric Gauss’ Law, French physicist
which relates the flux of the electric field from a closed surface S, border of a finite (macroscopic) volume t and engineer,
(n being as usual the outward normal unit vector to S) to the total quantity of net charge q inside the volume. he developed the
first fundamental law of
In vacuum, it is found: q
 S
E ( r, t )  n d S 
0
electricity regarding the
force due to static charges.

The flux of E is directly proportional to the total net charge inside the volume through the well-known constant parameter  0 ,
which is the so-called ‘dielectric constant’ or ‘permittivity’ of vacuum: in SI, it results: 109
 12
 0  8 .8 5 4  1 0  fa ra d / m
36
If the charge is not necessarily ‘discretized’ in specific points but can be seen instead as distributed continuously in the space,
a general ‘volume electric charge density’ r can formally be defined by:
dq The electric charge density, defined as the electric charge per unit volume, has UoM of ‘coulombs per cubic meter’ (C/m3).
 
d  We just note here that, ideally, the charge can be distributed not only in a volume (3D), but also on a surface (2D), or on a line (1D),
or just in a point (‘0D’). At due time, we will see the most convenient mathematical form (Dirac delta) to express such distributions.
1
Therefore, a more general form of Gauss’ Law in vacuum is obtained:
 S
E ( r, t )  n d S 
0  
 d

Such ‘large-scale’ integral relation can be reduced to a ‘small-scale’ local differential form by simply exploiting the Divergence Theorem:
1 
 S
E ( r, t )  n d S   
  E ( r, t ) d  
0  
 d   
(   E ( r, t ) 
0
)d   0
 ( r, t )
  E ( r, t ) 
From the integral equality, it derives the equality of integrands in any point. In vacuum, it is therefore: 0

AG-05/18
Electric Flux in the Matter: Free and Polarization Charges
As is well-known, in the presence of matter, the electric field is also due to the modification of the electric charge distribution at microscopic
level (referred to as electric ‘polarization’ charges, labeled with subscript ‘p’), generated by the ‘external’ electric charges (referred to as ‘free’
charges, labeled with subscript ‘f ’).
In principle, if we would be able to have exact knowledge of all the polarization charges in a medium,
we could predict the behavior of the electric field as could be done in vacuum, either in integral or in differential form:
qf  qp 1 1  f ( r, t )  p ( r, t )
S E ( r, t )  n d S     E ( r, t ) d  
0

0   f
 d  
0   p
 d     E ( r , t ) 
0

0
Unfortunately, as said, the distribution of the polarization charges is extremely irregular and can be defined only in terms of ‘macroscopic’
quantities, which take into account a process of ‘regularization’ with a spatial average on a sufficiently large number of ‘microscopic’ charges.
Then it is convenient to introduce a polarization vector, P ( r , t ) , able to account for the reaction of a medium in the presence of an external field.
For a fixed external electric field, the specific behavior of the polarization is a characteristic of each medium: this issue will be considered in
detail in the 3rd Week, that deals with the constitutive relations of media.
The polarization vector is defined so that its divergence is related to the ‘averaged’ polarization charge density by means of the relationship:
  P ( r, t )    p ( r, t )
 f ( r, t )   P ( r, t )
Therefore, referring directly to the differential expression, we obtain:   E ( r, t )      [  0 E ( r, t )  P ( r, t ) ]   f
0 0
It is convenient to define an overall auxiliary vector, D ( r , t )   0 E ( r , t )  P ( r , t ) , said ‘electric displacement’.
Dropping now the subscript ‘f ’ for the free charges, it is achieved a ‘div’ relation that has the advantage to depend on sources that are given
by free charges only, even though there is the need of introducing an auxiliary vector D ( r , t ) intrinsically related to the specific medium:
  D ( r, t )   ( r, t )
This is the common way to describe the flux features of electric quantities in a general medium. In vacuum we have obvioulsy: D ( r , t )   0 E ( r , t )
It is easily seen that both the fields P ( r , t ) and D ( r , t ) have a SI-UoM given by coulombs per square meter (C/m2).

AG-06/18
Magnetic Flux: Solenoidality and Magnetic Gauss’ Law
The study on magnetic effects, which dates back to very ancient times, emphasized
that the magnetic field, which can be generated both by magnets and by electric currents
(as verified experimentally by different scientists since the XIX century),
has the property of having force lines that are always closed.
The magnetic field is in fact solenoidal, and no specific physical elements have been found till now in nature that can behave
as ‘monopoles’, that is a source (or a sink) of magnetic field lines (as occurs instead for electric positive or negative charges).
- We will add however some considerations on this issue in the next lesson devoted to the formulations of Maxwell’s Equations.
As a consequence of the magnetic-field solenoidal nature, the flux of the induction field from the surface enclosing any arbitrary volume
is always zero. The Gauss’ Law for the magnetic field in integral form has therefore the simplified form:

 S
B ( r, t )  n d S  0
We emphasize that this relationship is valid both for vacuum and for any arbitrary physical medium filling an arbitrary volume t.
By applying again the Divergence Theorem, it is easily found:

 S
B ( r, t )  n d S   
  B ( r, t ) d   0
Then, from the ‘large-scale’ behavior, it is derived the ‘small-scale’ local differential relation
for the magnetic induction, which has the well-known form in any medium:

  B ( r, t )  0

AG-07/18
Faraday’s Law on Magnetic Induction and Electromotive Force
One of the key moments in the development of electrodynamics was the discovery by Faraday that the time rate of change for the magnetic
induction flux generates an electric field, able to give an electromotive force in a circuit. E.g., this principle is at the basis of electric engine action.
If applied to a ‘large-scale’ domain (any surface S limited by the contour s, with usual conventions on the right-hand orientation of the related
unit vectors), this feature is expressible in integral form showing that the circulation on the line s of the electric field is equal to the opposite of
the time-derivative of the magnetic induction flux through S (this expression is also known as Faraday-Neumann-Lenz law):
d
s dt 
E ( r, t )  s 0 d s   B ( r, t )  n d S

It should be noted that such expression is properly valid for non-moving surfaces S (it involves a ‘total’ temporal derivative of the flux).
For possibly moving surfaces, a more general expression can be found, where the time derivative is directly on B ( r , t )
inside the integral (and therefore it should now be considered as a ‘partial’ temporal derivative):

 B ( r, t )
 s
E ( r, t )  s 0 d s   
 t
 n dS

Taking advantage of Stokes’ Theorem, it is possible to pass from the ‘large-scale’ integral relation to the ‘small-scale’ differential expression:
 B ( r, t )
s
E ( r, t )  s 0 d s     E ( r, t )  n d S 
 


t
 n dS

From the last equality, we have one fundamental differential equation of electrodynamics, which shows as a time-varying magnetic field is an
origin, as a vortex, of an electric field (note also that this relation maintains the same form in any medium):
 B ( r, t )
  E ( r, t )  
t

AG-08/18
Ampere’s Law on Electric Currents and Magnetic Fields
Since XIX century, it was shown experimentally that electric currents are able to generate magnetic fields, whose lines of force are always
closed around the flowing of the current ic (these are usually the conduction currents in metals or in some cases could be convection
currents related to movement of charge densities). This is the well-known Ampere’s Law , which in vacuum assumes the form:
The circulation of B is directly proportional to the total linked electric current through the well-known constant
s B ( r , t )  s 0
d s   i
0 c
parameter m0, which is the so-called ‘permeability’ of vacuum: in SI, it results:  0  4   1 0  7 h e n r y / m
The current is measured in the fundamental unit of SI (MKSA), the ‘ampere’ (incidentally, its definition has recently been re-formulated).
It is important to note that the conduction/convection currents are not necessarily discretized elements given by charges 1 A = electric current
flowing along lines, but rather they can generally be associated to continuous net charge densities r in a volume moving related to the flow of
with velocity v along an oriented direction through a surface S. It can be defined a vector electric current density J ( r , t ), 1/(1.602176 634∙10−19)
directed along the local flow of (positive) charges and with an amplitude given by the current passing through a unit area: elementary charges
J ( r , t )  d i c / d S . Such currents can be generally expressed as J ( r , t )   v . in 1 second.
This volume electric current density in SI is measured as ampere per square meter (A/m2). There are cases for which the current flows on
surfaces (the section being a line), and it is said ‘surface electric current density’ (measured in A/m). If the current flows along a line (the section
being a point) we have a ‘linear electric current density’ (measured in A, as the current). We will consider again these quantities at due time.
Based on these considerations, the Ampere’s Law can be written in the more general form, which considers André-Marie Ampère
(1775-1836)
the electric current passing through a section S as the flux of the electric current density; still for vacuum, we have: French physicist, he

 B ( r, t )  s 0 d s   0ic   0  J ( r, t )  n d S made fundamental

s  discoveries on
electricity
From this relationship, it is possible to get also a local differential form, by using the circulation Stokes’ Theorem: and magnetism,

    B ( r, t )  n d S   0  J ( r, t )  n d S    B ( r, t )   0 J ( r, t )
thus developing
B ( r, t )  s 0 d s  electrodynamics.
s  

We can see that, even if derived in vacuum, such expression is incomplete in electrodynamic conditions: an additional term has to be considered.
In order to better understand such issue, it is opportune to complete first the analysis of the link between electric charges and currents.

AG-09/18
Charge Conservation and Continuity of the Current
Due to the fact that the electric current is related to the movement of charges, if we consider a volume t bounded by a closed surface S,
due to the principle of conservation of the electric charge, when a certain flux of current exists through S, this is balanced by a variation of
charge in the volume t. It results therefore an important equation which expresses the continuity of the electric current in any medium:
d   ( r, t )
S J ( r, t )  n d S  
dt 
 ( r , t )d    
t
d
where the last equality can hold, as usual, also for moving regions. The minus sign is related to the convention of the normal, for which
if a positive current flows outward through the surface, this has to be balanced by a decreasing of the positive charge inside the volume.
Note that in discretized form the previous relationship could be written in the well-known form that relates electric current and charges:
d dq dq
S J ( r , t )  n d S  i c
 
dt 
 ( r , t )d   
dt
 i c
 
dt
With standard developments, exploiting again the Divergence Theorem, we obtain the local differential form of the current continuity:

  ( r, t )   ( r, t )
 S
J ( r, t )  n d S   
  J ( r , t )d    

t
d    J ( r, t )  
t

Only in static conditions (  /  t  0 ), it results that the electric current density is solenoidal everywhere (   J ( r , t )  0 ).
We note now that, if the ‘curl’ eq. for B be valid (   B ( r , t )   0 J ( r , t ) ), by taking a further ‘div’, for the known differential
identity, we would obtain:   [   B ( r , t ) ]    [  J ( r , t ) ]  0    J ( r , t )  0
0

This is manifestly incorrect in electrodynamic conditions (as just seen, it should be:   J ( r , t )     ( r , t ) /  t ).
Therefore the circulation of the magnetic field given in terms of electric currents only as in Ampere’s Law is in general not correct:
something to make consistent this relation is missing, as discovered by Maxwell with the contribution of the displacement current…

AG-10/18
Displacement Current and Ampere-Maxwell’s Current Law
The fundamental Maxwell’s contribution to the EM laws was the observation, based on speculative considerations, that an additional current contribution
was needed in the Ampere’s Law, related to the rate change of displacement field and therefore called displacement current (identified by the subscript ‘d’):
D D
Jd ( r, t ) 
t
,Id   
Jd ( r , t )  n d S  
 t
 n dS
The circulation of the magnetic field that correctly takes into account both conduction and displacement currents is generally
known as Ampere-Maxwell’s Law. In vacuum, being D ( r , t )   0 E ( r , t ) , it assumes the form:
E
 s
B ( r, t )  s 0 d s   0 I c   0I d   0  J ( r, t )  n d S   0  0 
  t
 n dS
By using the circulation Stokes’ Theorem, the large-scale relation assumes the local differential form in vacuum:
E E
 s
B ( r, t )  s 0 d s   
  B ( r, t )  n d S   0  J ( r, t )  n d S   0  0 
  t
 n dS    B ( r, t )   0 J ( r, t )   0  0
t
In a generic medium different from vacuum, it is noted that the phenomenon of magnetic polarization, usually said ‘magnetization’, occurs, related to the possible
generation of elementary currents on an atomic scale, which modify the external magnetic field generated by the macroscopic free currents.
The magnetic polarization is taken into account with a vector field M (‘magnetization’), averaged on an adequate scale to make regular its spatio-temporal behavior.
M is related to the additional vortices generated by the atomic structure averaged on a ‘large’ scale (magnetic polarization current density J m p ),
satisfying the property:   M   0 J m p .
In a medium, Ampere-Maxwell’s Law would assume the form:
 D ( r, t )
 s
B ( r, t )  s 0 d s   
  B ( r, t )  n d S   0 I f   0I m p   0 I d   0  J f ( r, t )  n d S   0  Jm p ( r, t )  n d S   0 
   t
 n dS 
 D ( r, t )
 0  J f ( r, t )  n d S     M ( r, t )  n d S   0   n dS
   t
B ( r, t )  M ( r, t )  D ( r, t )  D ( r, t ) B  M
    H ( r, t )  J f ( r, t )   J ( r, t )   H  , B   0H  M
0 t t 0
Note that the definition of M is here different from the usual convention in physics for the presence of the term m0. The two forms are conceptually equivalent, being
the present choice, commonly used by engineers (with M is homogeneous to B instead of H), is formally dual w.r.t. the case of the electric case involving polarization.
The final differential form of the Ampere-Maxwell’s Law in arbitrary media is therefore the following:   H ( r , t )  J ( r , t )   D
t

AG-11/18
SI Units of Measurement in EM and Other Symbols

AG-12/18
Maxwell’s Equations: Standard ‘Physical’ Formulation
It has been seen that from the integral relations of Gauss law (for electric and magnetic fields) and of Faraday and Ampere-Maxwell laws,
exploiting the flux and circulation theorems, four fundamental equations for the local EM fields are achieved:
  D   In vacuum:    E   /  0

  
B  0    B  0
NB:
  1
 B   E    B  00  , c : s p e e d o f lig h t in v a c u u m
  E   
t c2
 t  1
  D    B   J     E c   2 .9 9 7 9 2 4 5 8  1 0 8 m / s  3  1 0 8 m / s
 H  J   0 0 0
t  00
t
This is the set of Maxwell’s Equations (‘MEs’) in the fundamental form commonly used nowadays in physics.
The set of equations developed by Maxwell in the 1860-70s is what Einstein called ‘the most important invention since Newton’s time’.
It consists of four linear differential equations with respect to space and time coordinates, involving vector and scalar EM fields:
E, B, D , H , J, 
MEs provide information on divergence and curl, that is on sources and vortices, which are the mathematical origins of the unknown fields.
Since, as seen, in order to keep MEs in a simple form related to only ‘free’ and generally ‘macroscopic’ electric charges and currents,
the introduction of ‘auxiliary’ EM vectors are needed, the formal problem of fixing all the origins of the EM fields for their determination is not
solved until the distinctive features of the medium are not known.
This issue will be addressed further in the 3rd Week, involving the ‘constitutive relationships’ of the media.
In vacuum, being fixed the links between E,D and B,H, MEs assume a form that allows for mathematically determining in principle
the EM fields in terms of physical electric charges and currents, since both sources (‘div’) and vortices (‘curl’) of E,B are given.

AG-13/18
Interdependence Among Differential Field Equations
In addition to MEs, it is possible to reach a local form of the integral continuity equation (‘CE’) of current, again through divergence theorem:

 J =  Hermann Schwarz
t (1843-1921):
It is possible to see that, in general, the 4 MEs are fundamental, since e.g. the CE can be derived from them. German mathematician,
he mainly worked
In fact, operating the div on 4th ME, using 1st ME, the CE is achieved: on subjects of complex analysis
D     D   
and differential geometry.
    H    J       0    J   0   J   0   J = 
  t  t t t
where the known Schwarz theorem on the exchange of the orders of derivation has been used (‘equality of mixed partials’):
A general space derivative (𝛻) of a time-differentiated ( 𝜕/𝜕𝑡 ) function is equal to a time derivative ( 𝜕/𝜕𝑡) of a space-differentiated (𝛻) function
(and vice-versa).
Nevertheless, in electrodynamic problems (i.e., when electric and magnetic phenomena are inter-related in the 3rd & 4th ME), it is possible to
consider as fundamental equations the 3rd & 4th MEs plus the CE, from which both the 1st and the 2nd MEs can be derived as a consequence:
In fact, it is seen with simple procedures similar to that presented above:
  D      D  
    H    J       
  t  t

t

t
     D   0    D    c o n s t (t )  0    D = 
  B  
    E       
  t  t
  B   0    B  c o n s t (t )  0
In the previous relations, the temporally-constant values const (t) for the div operations on D and B have to be fixed reasonably at zero,
since it should be assumed that physical EM fields cannot be present from infinitely remote instants.
   D      B  0
Note anyway that in electrostatic and magnetostatic problems (   t  0 ), the 4 MEs are always  ;  ; J = 0
the fundamental ones, being the electric and the magnetic phenomena completely separated:    E  0    H  J
 

AG-14/18
Symmetry of Maxwell’s Equations: Duality Principle
It can be seen that MEs show some formal symmetries between electric and magnetic relations, which anyway Paul Dirac (1902-1984):
are not complete, due to the absence of the corresponding (‘dual’) to the electric charge, i.e. the ‘magnetic charge’. English theoretical
physicist, Nobel prize,
In fact, single magnetic charges in rest (monopoles) or in motion (magnetic currents) have not yet been isolated regarded as one of the
in nature, even though a famous work by Dirac in 1931 showed theoretically that the possible existence most important scientists of XX c.,
of just a magnetic monopole in the universe could explain one of the great puzzles of science, he made very important
that is the quantization of all the electric charge. discoveries in quantum
mechanics and other areas.
Apart from these considerations, from a ‘formal’ viewpoint, it is possible to introduce fictitious
magnetic equivalents to the electric charges and currents and make completely symmetric MEs (and CE as well). As an intuitive example of
So, we can write MEs & CE in the following ‘dual’ form: use of duality, we can refer
   D   to the magnetic field
 distribution of an electric
   B   Even though such magnetic charges and currents are fictitious,
 m
being nonphysical quantities, their use in conveniently solving current loop (e.g., a circle of
 B radius a): at a distance
   E   J m  engineering EM issues (e.g., classes of radiation problems, etc.)
 t from the loop (r>>a), it is
is extremely powerful, as will be manifest after the analysis of
   H  J   D seen that the magnetic field
 the so-called ‘equivalence principles’ (4th Week): there, in fact is analogue to that
t
 it will be seen that magnetic current densities can profitably generated by a suitable
   J =    be associated to the knowledge of specific electric-field distributions magnetic dipole. This case
  t
 in suitable surface regions. is dual of the electric dipole.
  m
   J m = 
 t
It is seen that the system of MEs & CE commute in the same system if the following substitutions are made, which establish the ‘duality principle’:
D  B , E  H ,    m , J  Jm (All the electric quantities commute in the corresponding magnetic ones,
B   D , H   E ,  m    , Jm   J whilst magnetic quantities commute in the corresponding electric ones with changed sign).
- Duality has many useful applications to solve classes of symmetric EM problems, as will be emphasized in some cases in this Course as well.

AG-15/18
Maxwell’s Equations: Formulation with Impressed Terms
The ‘physical origins’ that produce the EM fields are electric charges in rest or in motion, the charges in motion being usually schematized in terms of currents.
In a local form, we deal in general with charge and current densities.
In many cases, such charges and currents can be considered as fixed ‘primary’ quantities, i.e., as known or assigned or ‘impressed’ terms from the mathematical
viewpoint. However, it is possible that, in media different from vacuum, there also exist physical electric charges and currents related to the field generated by the
impressed terms: such ‘secondary’ quantities, which are induced by the impressed terms, are unknown terms (as the EM fields) from the mathematical viewpoint.
The presence of these induced terms is particularly significant in conductive media (e.g., metals), where an electric current can be induced and is properly identified
as ‘conduction current’; in other materials, the induced movement of free electric charges (e.g. in a host medium as vacuum) generates what is also said ‘convection
current’. Conduction and convection currents are examples of unknown induced terms that are generated by other known impressed terms.
For modeling purposes, in MEs it is then possible to formally separate the impressed (known) terms from the induced (unknown) ones: e.g., we can add a subscript
‘i’ to the impressed charge and current densities (ri, Ji) to distinguish them from the unknown quantities, which we can leave with no subscript (r, J).
If the dual principle is applied , also magnetic charge and current densities can be introduced in MEs. In this case, however, due to the absence of physical net
magnetic charges, the magnetic charge and current densities have to be considered only as impressed (known) terms.
The set of MEs (and CE) can therefore be written in a modified reference form particularly useful for ‘engineering’ problems, commonly used in the following:


  D  i   As an intuitive example of the
 distinction between impressed and

   B  mi
 induced currents, let us consider an

  B

   E   Jm i  EM problem with a transmitting (Tx)
  t
 radio-base station (RBS) (the electric

  D
   H  Ji  J  currents of the relevant antennas

 t being Ji ) and a receiving (Rx) mobile

  (    )

   ( Ji  J ) =  i phone, capturing the radiated EM
  t field, which is converted in an induced


   current J by a metallic antenna inside
   Jm i =  mi


 t the phone.
Since in electrodynamics the continuity equation gives a relationship for the motion of charges in terms of currents, it is usually possible to consider only currents
as origins. Note that the impressed electric/magnetic current density can be seen as the equivalent of a current/voltage generator of circuit theory.

AG-16/18
Characteristics of Maxwell’s Equations
MEs originally proposed by Maxwell (in different scientific contributions, in 1861-62, 1864-65, and 1873) had Oliver Heaviside
a form quite different from the well-known present one, usually attributed to Heaviside , who gave a modern (1850-1925),
English electrical
form of vector representation, and to Minkowski , who avoided to explicitly refer them to the medium features. engineer, he made
useful mathematical
The 4 MEs represents the founding mathematical frame for modeling all the ‘classical’ static and electrodynamic
innovations to applied
problems (not involving ‘quantum electrodynamics’). ME can potentially describe all the engineering EM phenomena. sciences (EM, TLC, etc.).
Starting from the basic forms given by the set of partial differential equations (‘PDE’), various mathematical manipulations
are possible in order to find the most appropriate technique for solving (analytically and/or numerically) specific EM problems.
We will examine further (next Week), after having completed other fundamental issues, what kind of additional information
is needed to make the mathematical model based on MEs ‘well posed’, i.e., fully consistent and practically useful.
We just notice here a particular feature of MEs from the physico-mathematical viewpoint: Hermann Minkowski
it results that the differential form of MEs is ‘invariant’ with respect to any inertial reference frame, i.e., (1864-1909),
German mathematician,
MEs maintain the same expressions in any system moving with a fixed velocity v w.r.t. another one. he gave important contributions
The measurements of EM quantities by observers in different reference frames should be the same. to number theory and relativity.
This important characteristic, which had fundamental consequences e.g. for the general theory of relativity, is known as covariance principle.
The PDE model of MEs in principle allows for the possible exact determination of EM fields in any point in vacuum (cf. Helmholtz theorem, etc.).
In a general medium, if seen at ‘microscopic’ (atomic or sub-atomic) scale, the EM field shows extremely irregular behaviors in space and time,
since the matter is formed by an extremely high number of elementary charges and currents, which also contribute in complex way to the field.
The EM field vectors defined in MEs can be ‘regularized’ (through statistical spatial averages on a scale typically of the order of 10-8 m at least)
in the so-called ‘macroscopic’ regime, where the ‘granular’ features of the matter is no longer revealed (we then refer to ‘continuous media’):
this allows for treating problems through ‘regular’ fields also in the matter, i.e. with ‘well-behaved’ mathematical functions of space and time.
This issue will be discussed next by considering the EM features of the matter (next Week).

AG-17/18
Symbols, Prefixes, and Physical Constants for EM
GREEK ALPHABET
A a alpha
B b beta
G g gamma
D d delta
E e epsilon
Z z zeta
H h eta
Q q theta
I i iota
K k kappa
L l lambda
M m mu
N n nu
X x xi
O o omicron
P p pi
R r rho
S s sigma
T t tau
U u upsilon
F F phi
C c chi
Y y psi
W w omega

AG-18/18