ELECTRO MAGENTIC THEORY

ELECTROMAGNETICTHEORY
UNITI
ELECTROSTATIC
Boundary value problems and laplace equation – boundary conditions and uniqueness
theorem – Laplace equation in three dimension – Solution in Cartesian and spherical polar
coordinate – Examples of solutions for boundaryvalue problems.Polarization anddisplacement
vectors – Boundary conditions – Dielectric sphere in a uniform field – Molecular polarizability
and electrical susceptibility – Electro static energy in the presence of dielectric – Multi pole
expansion.

UNITII

MAGNETOSTATICS

Biot – Savart laws of induction – Magnetic vector potential and magnetic field of a
localized current distribution – Magnetic moment, force and torque on current distribution in an
external field – Magneto static energy – Magnetic induction and magnetic field in macroscopic
media – Boundary conditions – Uniformly magnetizes sphere.

UNIT III

MAXWELLEQUATIONS

Faraday’s laws of induction – Maxwell’s displacement current – Maxwell’s equations –

Vector and scalar potentials-Gauge invariance – Wave equation and plane wave solution –
Colom band Lorents gauges- Energy and momentum of the field- Pointing’s theorem- Lorents
force – Conservation laws for a system of charges and electro magnetic fields.

UNIT IV

WAVEPROPAGATION

Plane waves in non-conduction media – Linear and circular polarization, reflection and
refractionataplaneinterface-Wavesinconductionmedium–Propagationofwavesina
rectangularwaveguide.Inhomogeneouswaveequationandretardedpotentials–Radiation from a
localized source – Oscillating electric dipole.

UNIT V

ELEMENTARYPLASMAPHYSICS

The Boltzmann equation – Simplified magneto-hydrodynamic equations .Electron plasma

oscillations-The Debye shielding problem – Plasma confinement in a magnetic field – Magneto-
hydro dynamic waves – Alfiven waves and magneto sonic waves.
UNIT-1
 UNIT I
ELECTROSTATIC
Boundary value problems and laplace equation – boundaryconditions and uniqueness theorem –
Laplace equation in three dimension – Solution in Cartesian and spherical polar coordinate –
Examples of solutions for boundary value problems.Polarization and displacement vectors –
Boundary conditions – Dielectric sphere in a uniform field – Molecular polarizability and
electrical susceptibility – Electro static energy in the presence of dielectric – Multi pole
expansion.

Boundaryvalueproblemsandlaplaceequation:

TheLaplace'sequationisgiven

A2V=0
Thisequationisencounteredinelectrostatics,where Vistheelectricpotential,relatedtothe electric field by

E=−AVE=−AV;

ItisadirectconsequenceofGauss's law,

A⋅E=ρ/ϵ

in the absence of a chargedensity. WhereVis the gravitational potential, related to

thegravitational field by

g=−AVg=−AV

Inthermalphysics,withVplayingtheroleoftemperature,andinfluidmechanics, with V a
potential for the velocity field of an incompressible fluid.

Thelistcouldgoon.Laplace'sequationisavery importantequationinmany areasof physics.

 Boundaryconditionsanduniquenesstheorem:

The formulation of Laplace's equation in a typical application involves a number of

boundaries, on which the potential V is specified.

Examples of such formulations, known as boundary-value problems, are abundant in

electrostatics.

Here, a typical boundary-value problemforVbetween conductors, on whichVis

necessarily constant. In such applications, the surface of each conductor is a boundary, and by
specifying the constant value ofVon each boundary, we can find a unique solution to Laplace's
equation in the space between the conductors.

In other situations the boundary may not be a conducting surface, and Vmay not be
constant on the boundary. But the property remains, that once Vis specified on each boundary,
the solution to Laplace's equation between boundaries is unique.

We shall see this uniqueness property confirmed again and again in the boundary-value
problems examined. A general proof of the uniqueness theoremis not difficult to construct, but
we shall not pursue this here.

The superposition principle followsdirectlyfrom the factthat Laplace'sequation is linear in

the potentialV. Suppose thatV1,V2,V3, and so on, are all solutions to Laplace's equation, so that

A2Vj=0

Anysuperpositionofthe form

V=a1V1+a2V2+a3V3+⋯,

whereajareconstants,isalsoa solution,because

A2V=A2(a1V1+a2V2+a3V3+⋯)
 =a1A2V1+a2A2V2+a3A2V3+⋯=0.

This is the statement of the superposition principle, and it shall form an integral part of
our strategy to find the unique solution to Laplace's equation with suitable boundary conditions.

It is important to understand that the superposition principle applies to any numberof

solutions Vj; this number could be (and will be) infinite.

Laplaceequationinthreedimension:

ThegeneralformofLaplace’sequationis:

A2 V=0;

Itcontainsthelaplacian.Thissectionwillexaminetheformofthesolutionsof Laplacesequation in
cartesian coordinates and in cylindrical and spherical polar coordinates. Of course it is nice to
know how to solve Laplace’sequation inthese coordinate systems, particularlyrecallingthat the
choice of coordinate system is generally determined by the symmetry of the boundaryconditions.

SolutioninCartesiancoordinate:

Laplace's equation can be formulated in any coordinate system, and the choice of
coordinates is usually motivated by the geometry of the boundaries. When these are nice planar
surfaces, it is a good idea to adopt Cartesian coordinates, and to write

𝜕2𝑉 𝜕2𝑉 𝜕2𝑉

0=A 𝑉=2
+ +
𝜕𝑥2 𝜕𝑦2 𝜕𝑧2

forLaplace'sequation.

afactorizedsolutionofthe formis
 V=X(x)Y(y)Z(z)

involvingthreeindependentfunctionsofx,y,andz.

We cannot expectall solutionsto Laplace'sequationtobe ofthissimple,factorizedform; the

vast majority are not. The hope is that a superposition of factorized solutions will form the
unique solution to a given boundary-value problem.

𝑑2𝑋
𝑑2𝑌 𝑑2𝑍
0= 𝑌𝑍+𝑋
𝑑𝑥 2 1 𝑑𝑧2
+
𝑑𝑦2 𝑍

1𝑑2𝑋 1 𝑑2𝑌
0= + 𝑑2𝑍
𝑋𝑑𝑥 2 𝑌
1 𝑑𝑧2
+
𝑑𝑦2 𝑍

1𝑑2𝑋 1 𝑑2𝑌
− = 𝑑2𝑍
𝑋𝑑𝑥2
1 𝑑𝑧2
𝑌 +
𝑑𝑦2 𝑍

afterwe dividethroughbyXYZ.
AboveEquationstatesthatafunctionof xxonlyisequaltothesumofafunction of yy only and a
function of zz only. Introducing an obvious notation, the equation states that
f(x) =g(y)+h(z)f(x) =g(y)+h(z).
If yis changed, for example, the function g (y)is certainly expected to change, but this can
have no incidence on f(x), which depends on x, a completely independent variable.
Yet thisiswhat Laplace'sequation seemsto imply:a change in ymust produce a change inf,
because f = g+hf. The third variable zcannot come to the rescue here, because it is also
independent from x and y.
We have an intolerable contradiction, and the only way out is to declare that f(x),g(y),
and h(z) are all constant functions.
With this property we have that ggdoes not, in fact, change when yis changed, and the
tension with the equation
f=g+h
disappearsbecauseffalsowillnotchange.

Forconveniencewewrite
f(x) =α2=constant,
or

1𝑑2𝑋
𝑋𝑑𝑥 =−𝛼2
2

Thesolutionstothisordinarydifferentialequationare

𝑋(𝑥)=𝑒±𝑖𝛼𝑥

𝑋(𝑥)={ sin( 𝛼 𝑥)
cos(𝛼𝑥)

cos(𝘢x)= (ei𝝰x+e−i𝝰x),
1

sin(𝘢x)=− (ei𝝰x−e−i𝝰x),
1

e±i𝝰x=cos(𝘢x)±isin(𝘢x)

Proceedinginasimilarmannerforthefunctionofy,we write

G(y)=−β2=constant,

or

1d2Y
Ydy2 =−β
2

Y(y)=e±iβx

Y(y)={
cos(βx)
sin( β x)
Againwecangobackandforthbetweenthecomplexexponentialsandthetrigonometricfunctions, and the
sign in front of β2β2 can be altered by letting β→iβ.

1 𝑑2𝑍
𝑍 𝑑𝑧2 =𝛼 +𝛽
2 2

afterweinsertourpreviousresultsfor XandY. Inthiscasetheconstantontheright-handsideof

theequationisfullydeterminedintermsof ααandββ,andwiththepreviouschoicesofsign for α 2 and β2,
the constant is now positive. The solutions

𝑍
(𝑧)= 𝑒 ±√𝛼
2+𝛽2

cosh(√𝛼2+𝛽2)𝑧
𝑍(𝑧)= {
sin(√𝛼2+ 𝛽2)𝑧

Herewecanfreelygobackandforthbetweentheexponentialandhyperbolicformsofthe solutions.

Collectingresults,wefindthatthefactorizedsolutionstoLaplace'sequationinCartesiancoordinates are of
the form

Solutioninsphericalpolarcoordinate:

Sphericalboundariescallforsphericalcoordinates.

The Laplacianoperatorexpressedintermsof (r,θ,ϕ)wasobtainedbackinEquation,and we

have

0=A2= 𝛛𝑉
𝛛
(𝑠𝑖𝑛𝜃 )+ 1
𝛛2𝑉
(𝑟 )+
𝛛 2𝛛𝑉
1
1
𝑟2 𝛛 𝛛𝑟 𝑟2𝑠𝑖𝑛 𝛛 𝛛𝜃 𝑟2𝑠𝑖𝑛2 𝛛𝑡2
𝑟 𝜃 𝜃 𝜃
forLaplace'sequation.

Asusualwebeginwithafactorizedsolutionofthe form

V(r,θ,ϕ)=R(r)Y(θ, ϕ),

whichweshallinsertwithinLaplace'sequation.
We could factorize Y(θ, ϕ)further by writing it as Θ(θ) Φ(ϕ), but this shall not be
necessary. The notation might trigger some expectation that Ywill have something to do with
spherical harmonics.
SubstitutionintoLaplace'sequationyields

1𝑑 𝑑𝑅 1 1 𝛛 𝛛F
0= (𝑟2 )+ [ (𝑠𝑖𝑛𝜃 )+ ]
2
1 𝛛F

𝑅𝑑𝑟 𝛛𝑟 F 𝛛 𝛛𝜃 𝑠𝑖𝑛2 𝛛Ø2

𝜃 𝜃
𝑠𝑖𝑛𝜃

andthisequation informsusthata function ofronlymustbeequaltoa functionof θ and ϕ.

Asusualweconcludethateachfunctionmustbeaconstant,whichwedenote μ.We therefore obtain

the ordinary differential equation

𝑑 𝑑𝑅
(𝑟 2
)=𝜇𝑅
𝑑𝑟 𝜕𝑟
fortheradialfunction,andthepartialdifferentialequation
𝛛F
(𝑠𝑖𝑛𝜃 )+ =−𝜇𝑌
2
1 𝛛 1 𝛛F

𝑠𝑖𝑛𝜃 𝛛 𝛛𝜃 𝑠𝑖𝑛2 𝛛Ø2

𝜃 𝜃

for the angular function. This equation maybe compared withEquation, thedifferentialequation
satisfied by the spherical harmonics.
 They are identical in form, except that here, μis not necessarilyequal to ℓ(ℓ+1), with ℓ a
non negative integer.
 Itcanbeshownthatforagenericvalue ofμ,thesolutionstoEquationbecomeinfinite at θ = π
when they are required to be finite at θ=0.
AsimilarobservationisinthecontextofLegendrefunctions.Therethegeneralized Legendre

(1−µ2)f"−2uf′+λ(λ+1)f=0
equation,

hassolutionsthatmustblowupat 1u=−1evenwhentheyarefiniteat u=1,unlessλisa nonnegative

integer.
AgeneralizationtotheassociatedLegendre equation,

(1−µ2)f"−2uf′+[λ(λ+1)−
𝑚2 ]f=0
1−𝑢2

makesthe situationeven worse. Withu=cosθ , wearespeakingof functionsthatbecomeinfinite at

θ=π, just like the Y of Equation.
The singular solutions to Equationare simply not acceptable; our potential should be
nicely behaved everywhere in space, and it should certainly not go to infinity on the negative z-

=−l(l+1)Y
𝛛Y
(sinθ )+
axis. This compels us to set μ equal to ℓ(ℓ+1), because in this case Equation becomes <\p>
2
1 𝛛 1 𝛛Y

sinθ 𝛛 𝛛θ sin2 𝛛Ø2

θ θ

whichispreciselythedifferentialequationforthesphericalharmonics.Thesolutions areY ℓ (θ,ϕ),

withℓ=0,1,2,3,⋯
m andm=−ℓ,−ℓ+1,⋯,ℓ−1,ℓ, and these are allnicely behaved functions that don't
go infinite anywhere. With this restriction on μ, Equation becomes

d2R 𝜕r
r 2
+2𝑟 𝜕
dr2 R
 = 𝑙( 𝑙+1)𝑅

anditiseasytoshowthatthisdifferentialequationpossessesthe independentsolutions
 R= r ℓ
And
−(ℓ+1).
R=r

Polarizationanddisplacementvectors:

Dielectrics are materials which have no free charges; all electrons are bound and
associated with the nearest atom .

They behave as electrically neutral when they are not in an electric field.  An external
applied electric field causes microscopic separations of the centers of positive and negative
charges as shown in Figure1.1. These separations behave like electric dipoles, and this
phenomenon is known as dielectric polarization.

Figure1.1

Dielectrics may be subdivided into two groups :

• Non-Polar: dielectrics that do not possess permanent electric dipole moment. Electric
dipole moments can be induced by placing the materials in an externally applied electric
field.

• Polar: dielectrics that possess permanent dipole moments which are ordinarily randomly
oriented, but which become more or less oriented bythe application of an external electric
field. An example of this type of dielectric is water.
 Figure 1.2

Figure 1.3

However,thealignmentisnotcompleteduetorandomthermal motion.

The aligned molecules (induced dipoles) then generate an electric field that is
opposite to the applied field but smaller in magnitude.
 Figure 1.4

Suppose we have a piece of material in the form of a cylinder with area Aand height h,
and that it consists of N electric dipoles, each with electric dipole moment 𝑝 spread uniformly
throughout the volume of the cylinder.

Inthecaseof ourcylinder, whereall thedipolesareperfectlyaligned, themagnitudeof𝑃

isequalto

Figure 1.5

𝑃= 𝑁𝑝
𝐴ℎ
From the equivalence between Fig.1.5 (a) and (b), we have two net charges ±QP
whichproduce net dipole moment of QPh; hence
 𝑄𝑝= 𝑁
𝑝

ℎ
We note that the equivalent charge distribution resembles that of a parallel-plate
capacitor, with an equivalent surface charge density σ P that is equal to the magnitude of the
polarization:

𝜎𝑝= 𝑄 𝑁𝑝
= 𝑃
𝑝 𝐴ℎ
𝐴

Thus,ourequivalentchargesystemwillproduceanaverageelectricfieldofmagnitude
𝐸𝑃=𝑃/𝜖0.Sincethedirectionofthiselectricfieldisoppositetothedirectionof𝑃,invector notation, we

𝑃̅→
have

→
̅̅̅
𝐸̅̅̅ =−
𝑝 𝜀0

Thetotalelectricfield𝐸isthesumofthesetwofields:

𝐸̅→=𝐸̅̅̅0̅̅̅̅→+ 𝐸̅̅̅𝑝̅̅→
̅

=𝐸̅̅̅→
̅̅̅̅ − 𝑃̅→---------------------------------
(2)
0 𝗌0

In most cases, the polarization 𝑃 is not only in the same direction as 𝐸 but also linearly
proportional to 𝐸0 (and hence 𝐸.) This is reasonable because without the external field there
would be no alignment of dipoles and no polarization. We write the linear relation between andas
 ̅ P→ =ε0xeE̅→-----------------(3)

whereχe is calledtheelectricsusceptibility.Materials thatobey thisrelationare linear dielectrics.

CombingEqs. (2) and (3) gives
 𝐸̅̅̅̅0→ =(1+𝑥 𝑒 ) E̅→

=𝑥 𝑒 E̅→

𝜀𝑟=𝑥𝑒=1+𝑥𝑒

isthedielectricconstant.Thedielectricconstant𝜅𝑒isalways greaterthanonesince𝜒𝑒>0.

Boundaryconditions:

Let us first consider the interface between two dielectrics having permittivities 𝜀1 and 𝜀2
and occupying regions 1 and 2, as shown in the figure below. • We first examine the tangential
components by using
 Figure1.6

ф𝐸.𝑑𝐿=0

aroundthesmallclosedpath, obtaining

𝐸tan1∆𝑤+ 𝐸tan2∆𝑤 = 0

ThesmallcontributionbythenormalcomponentofEalongthesectionsoflength∆𝑤
becomesnegligibleas∆h decreasesandtheclosedpathcrowdsthesurface.Immediately,then,

𝐸tan1 =𝐸tan2

The boundary conditions on the normal components are found by applying Gauss’s
lawto the small “pillbox” shown in the figure below
 Figure1.7

Thesidesareagainveryshort,andthe fluxleavingthe topandbottomsurfacesisthe difference

𝐷𝑁1∆𝑆−𝐷𝑁2∆𝑆=∆𝑄=𝜌𝑠∆𝑆

𝐷𝑁1−𝐷𝑁2=𝜌𝑠
This charge maybe placed there deliberately, thus unbalancingthe total charge in and on
this dielectric body. Except for this special case, ρS is zero on the interface; Hence

𝐷𝑁1=𝐷𝑁2

𝜖1𝐸𝑁1=𝜖2𝐸𝑁2

LetD1(andE1)makeanangleθ1withanormaltothesurfaceasshowninthefigure.

Becausethe normalcomponentsofDarecontinuous,
 Figure1.8

But

𝐷tan 𝐷tan2
=𝐸tan1 =𝐸tan2
1
𝜖2
𝜖1 =

𝐷tan1 𝜖1
=
𝐷tan2 𝜖2

Thus

𝜖2𝐷1sin𝜃1=𝜖1𝐷2sin𝜃2

Andthedivisionoftheequation

tan𝜃
=
𝜖1
1tan
𝜖2
𝜃2
 Dielectricsphereinauniformfield:

Asphere of radius R, made of linear dielectric material is placed in an otherwise uniform

electric field E0.

Figure1.9

Find the resulting field inside and outside the sphere. The susceptibility is χ e. First,choose
the z-axis in the direction of the uniform electric field. Then the boundary condition at infinity is

V∞ =−E0r cosθ

SolvingtheLaplaceequationforthepotentialoutsidethesphere,wehave

𝑉 ∞
(𝐵𝑟,𝜃)=∑(𝐴𝑟𝑙+ )𝑃 (cos𝜃)
𝑙

𝑜𝑢𝑡
𝑟𝑙+1
� 𝑙
�
𝑙=0

Asr→∞,theBltermsbecomenegligiblysmall.Therltermsgrowlarge,butwemustnevertheless have a term

linear in r,

𝑉∞(𝑟,𝜃)=−𝐸0𝑟cos𝜃
 ∞

=∑𝐴𝑙𝑟𝑙𝑃𝑙(cos𝜃 )
𝑙=0

soA1=−E0,withallotherAl=0.Therefore,puttinginthesinglenonzeroAlterm explicitly, we have

∞
𝑉𝑜𝑢𝑡 𝐵𝑙
(𝑟,𝜃)= − �(cos𝜃)
𝑟cos𝜃+∑
𝐸0 𝑟
𝑙+
1
�
𝑙=0 𝑙

for theexteriorsolution.

TofindtheconstantsBl,wemustmatchthistotheinterior solution.

𝑥𝑒
Intheinteriorofthesphere,wehavethefreeandboundchargedensitiesrelatedby

𝜌𝑏= 𝜌𝑓
1−𝑥𝑒
butthereisnofreechargesobothvanish,andweagainsolvetheLaplaceequation,

𝑉 ∞
(𝑟,𝜃)=∑(𝐶𝑟
𝐷𝑙
𝑙
+ )𝑃 (cos𝜃)
𝑖𝑛 �
𝑟𝑙+1
𝑙
�
𝑙=0

Thistime,regularityat the origintellsusthatDl=0 foralll, so

𝑉𝑖𝑛(𝑟,𝜃)=∑𝐶𝑙𝑟𝑙𝑃𝑙(cos𝜃)
𝑙=0

Theboundaryconditionsatr =Rare:

𝐷
⊥
𝑜𝑢𝑡
−𝐷⊥𝑖𝑛=𝜎𝑓
𝐸
‖
𝑜𝑢𝑡
−𝐸‖ 𝑖𝑛=0
Now,computetheelectricfield:

𝐸𝑜𝑢𝑡=−A𝑉𝑜𝑢𝑡(𝑟,𝜃)

𝜕 ∞
𝐵𝑙
=−(𝑟^ ^ 𝜕
𝜕 1+𝜃 )(−𝐸0 𝑟cos𝜃+∑ )
𝑟 𝜕 𝑟𝑙+1
𝑟 𝜃
𝑙=0

(𝑙+1)𝐵
∞
𝐵𝑙
=𝑟^(𝐸 cos 𝜃+ ∑ 𝑙+2 𝑙 sin𝜃+ 𝜕𝑃𝑙
�)
𝑟
𝑟𝑙+2 𝜕𝜃 )
0 𝑙=0 �−𝜃^(𝐸0
𝑙

𝐸𝑖𝑛=−A𝑉𝑖𝑛(𝑟,𝜃)

𝜕
=−(𝑟^
∞
^ 𝜕
𝜕 +𝜃
1 )∑ 𝑟𝑙𝑃𝑙
𝑟 𝜕 𝑐𝑙
𝑟 𝜃
𝑙=0

=−𝑟^∑ 𝑙𝑐 𝑟𝑙 𝛛𝑃𝑙
∞
𝑙= 𝑟𝑙−1
𝜃^𝑐
𝛛𝜃
𝑃𝑙
-

𝑃𝑙
𝑙

where we use ∂P / ∂θ = 0 in the second sum. The parallel components are those in the θˆ
direction. For the parallel components at r = R

𝐸𝑜𝑢𝑡= 𝐸𝑖𝑛
‖ ‖
−θ^(
)=−θ^𝑐 𝑅𝑙−1𝛛𝑃𝑙
𝛛𝑃𝑙
𝐸 𝑠𝑖𝑛 𝜃+∑∞ 𝐵𝑙
0 𝑙
𝛛 𝛛𝜃
𝜃
𝑙=1𝑅𝑙+2
 (−𝐸 + )+∑ =𝑐 1 +𝑐𝑅𝑙−1
𝛛𝑃𝑙 𝐵𝑙𝛛𝑃𝑙 ∞ 𝐵𝑙 𝛛𝑃𝑙 𝛛𝑃𝑙 𝛛𝑃𝑙
0 𝑙=1𝑅𝑙+
𝛛𝜃 𝑅 𝛛 𝛛 𝛛𝜃 𝛛𝜃
3 𝜃 2 𝜃 𝑙

where we separate outthel=1terms.

The derivative terms are all independent (see optional section below), so each coefficient
must vanish.

Foreach l≥ 2, this implies

 𝑐𝑙 = 𝐵𝑙
𝑅2𝑙+1
whilefor thel = 1 term,

𝐵𝑙
𝑐𝑙=−𝐸0+
𝑅3
Thisfixesall of theClin termsof theBl.

Now we look at the normal components of D. Again, there is no free surface charge,
sowe have continuity of the normal components at R

𝐷⊥𝑜𝑢𝑡=𝐷⊥
𝑖𝑛

∞
𝜖 𝐸 𝑃+𝜖 (𝑙+1)𝐵𝑙
∑
𝑃 =−𝜖𝑙𝑐𝑅𝑙−1𝑃
0 0 0 𝑙 𝑙
𝑙
𝑙= 𝑅
𝑙
𝑙+2

wherewehavewrittenP1forcosθ.Forl=0,

𝐵0
𝜖0=
=0
𝑅2
𝐵0=0

Forl=1,

2𝐵𝑙
𝜖0𝐸0𝑃𝑙+𝜖 𝑃𝑙=−𝜖𝐶1𝑃𝑙
0 𝑅 3
2𝐵𝑙
𝜖0𝐸0+𝜖0 =−𝜖𝐶1
𝑅3
( 2𝜖0 )𝐵 =( 𝜖−𝜖 )𝐸
𝑅3
1 0 0
 𝜖−𝜖0
𝐵𝑙 = ( )𝐸0 𝑅3
2𝜖0+𝜖
Finally,fortheremainingl,

(𝑙+1)𝐵𝑙
𝜖 =−𝜖𝑙𝑐𝑙 𝑅𝑙−1
0 𝑅𝑙+2
Substitutingthevaluesfor Cl,

𝜖0 (𝑙+1)𝐵𝑙 𝐵𝑙
=−𝜖𝑙 𝑅𝑙−1
𝑅𝑙+2 𝑅2𝑙+1

𝐵 (𝜖 ( 𝑙+1)
𝑅𝑙−1)=0
1
+𝜖𝑙
𝑙 0
𝑅𝑙+ 𝑅2𝑙+1
2

1
𝐵𝑙(𝜖0(𝑙+1)+𝜖𝑙)=0
𝑅 𝑙+2

𝐵𝑙=0

𝐵𝑙
andthereforeBl= 0 andCl= 0 for alll≥ 2, whileforthel= 1 term,

𝑐1=−𝐸0+ 3
𝑅
=−𝐸0 1 𝜖−𝜖0 3
( 𝜖 )𝐸0
+ 3 2𝜖0+𝜖
𝑅
0
𝑅3
=− 𝐸0
2𝜖 +𝜖
0

Thepotentialis
 𝑉 (𝑟,𝜗)=−𝐸
𝜖−𝜖0 𝑅3
𝑟𝑐𝑜𝑠𝜃+( ) 𝐸 𝑐𝑜𝑠𝜃
𝑜𝑢𝑡 0 2𝜖0+ 0
𝑟2
𝜖
3𝜖0
𝑉 (𝑟,𝜗)=− 𝑟𝑐𝑜𝑠𝜃
𝐸
2𝜖0+𝜖
𝑜𝑢𝑡 0

𝑉 (𝑅,𝜗)=−𝐸 𝑅𝑐𝑜𝑠𝜃+(
𝜖−𝜖0
)𝐸 𝑅𝑐𝑜𝑠𝜃
Attheboundary,wecheckcontinuity:

𝑜𝑢𝑡 0 0
2𝜖0+𝜖

=−
3𝜖0 𝐸 𝑅𝑐𝑜𝑠𝜃
2𝜖0+𝜖
0

=𝑉𝑖𝑛(𝑅,𝜗)

Thefinalelectricfieldsare

𝐸𝑜𝑢𝑡=(1+ 2𝑅
( 2 𝜖−𝜖0 ) 𝐸0(cos𝜃𝑟^−𝑠𝑖𝑛𝜃𝜃^ ))
3

𝑟
2𝜖0+𝜖

𝑅3 𝜖−𝜖0
=(1 + 2 2 ( ^
)𝐸0𝑘
𝑟 2𝜖0 +𝜖
3𝜖0𝐸0
𝐸𝑖𝑛 = (cos𝜃𝑟^−𝑠𝑖𝑛𝜃𝜃^ )
2𝜖0+𝜖
3𝜖0𝐸0
𝐸𝑖𝑛 = 𝐸 𝑘^
2𝜖0+𝜖
0

Noticethatthepolarizationdensityisinthesamedirectionastheelectricfield.
 Electricalsusceptibility:

The electric susceptibility is generally defined as the constant of proportionality (whatcan

possibly be a matrix); these are related to an Electric Field E to the induced dielectric
polarisation density P. Electric susceptibility, which is also known as dielectric susceptibility, is
considered to be a dimensionless proportionalityconstant which is responsible for indicating the
degree of polarisation of a dielectric material, this phenomenon happens in response to anapplied
electric field.

Electricsusceptibilityisdirectlyproportionaltothe polarisationofa material.

Theformulaofelectricsusceptibilityisderivedasfollow:

P= ε₀ XₑE
Where,
P=Itisconsideredasthe polarisation density.
ε₀=Itisconsideredastheelectricpermittivityoffreespace. Xₑ = It
is considered as the electric susceptibility.
E=Itrepresentstheelectric field.
Thesusceptibilityisprovedtoberelatedtoitsrelativepermittivity,alsoknownas dielectric
constant εᵣ by:
Xₑ= εᵣ-1
Thereforeinthe case ofvacuum,

Xₑ= 0
Duringthistime,theelectricdisplacementDalsobecomesequaltothepolarisation density P by:

D
=
ε
Where, ₀
E+P
=ε
₀(1
 +
X
ₑ)
E
=
ε
ᵣ
ε=
εᵣε
₀
ε
ᵣ
=

(
1

+
X
ₑ
)
UNIT-II
 UNIT II

MAGNETOSTATICS

Biot – Savart laws– Magnetic vector potential and magnetic field of a localized current
distribution – Magnetic moment, force and torque on current distribution in an external field –
Magneto static energy – Magnetic induction and magnetic field in macroscopic media –
Boundary conditions – Uniformly magnetizes sphere.

Biot–Savart laws:

The Biot Savart Law is an equation describingthe magnetic field generated bya constant
electric current. It relates the magnetic field to the magnitude, direction, length, andproximityof
the electric current.

Biot–Savart law is consistent with bothAmpere’scircuital law andGauss’s theorem.The

Biot Savart lawisfundamentaltomagnetostatics,playinga role similartothat ofCoulomb’slaw in
electrostatics.

Figure2.1

BiotSavartlawwascreatedbytwo Frenchphysicists,JeanBaptisteBiotand FelixSavart

derived the mathematical expression for magnetic flux density at a point due a nearby current
carrying conductor, in 1820.
 Viewingthe deflection ofa magantic compass needle, these two scientists concluded that
anycurrent element projects a magnetic field into the space around it.Through observations and
calculationstheyhad derived a mathematicalexpression, which shows, themagnetic flux density of
which dB, is directlyproportional to the length of the element dl, the current I, the sine of the
angle and θ between direction of the current and the vector joining a given point of the field and
the current element and is inversely proportional to the square of the distance of the given point
from the current element, r.

BiotSavartLawStatement

ThisisBiotSavartlawstatement:

Hence,

𝐼𝑑𝑙𝑠𝑖𝑛
𝑑𝐵∝ 𝜃
𝑟2
𝑑𝐵=k
𝐼𝑑𝑙𝑠𝑖𝑛𝜃
𝑟2
Where,kisaconstant,dependinguponthemagneticpropertiesofthemediumand system of the
units employed.

𝜇0𝜇𝑟
InSIsystemofunit,

𝑘=
4𝜋
Therefore,finalBiotSavart lawderivationis,

𝑑𝐵= 𝐼𝑑𝑙𝑠𝑖𝑛𝜃
𝜇0𝜇𝑟 ×
𝑟2
4𝜋
Let us consider a long wire carrying a current I and also consider a point p in the space.
The wire is presented in the picture below, by red color.
 Let us also consider an infinitelysmall length of the wire dl at a distance r from the point
P as shown. Here, r is a distance vector which makes an angle θ with the direction of current in
the infinitesimal portion of the wire.

If you tryto visualize the condition, you can easilyunderstand the magnetic field density
at the point P due to that infinitesimal length dl of the wire is directly proportional to current
carried by this portion of the wire.

As the current through that infinitesimal length of wire is same as the current carried by
the whole wire itself, we can write,

𝑑𝐵∝I
It is also very natural to think that the magnetic field density at that point P due to that
infinitesimal length dl ofwireisinverselyproportionalto the squareofthe straightdistancefrom point

1
P to center of dl. Mathematically we can write this as,

𝑑𝐵∝
𝑟2

Figure1.2

Lastly, magnetic field density at that point P due to that infinitesimal portion of wire
isalso directly proportional to the actual length of the infinitesimal length dl of wire.
 As θ be the angle between distance vector r and direction of current through this
infinitesimal portion of the wire, the component of dl directlyfacing perpendicular to the point P
is dl sinθ,

Hence,

𝑑𝐵∝dlsinθ

𝐼𝑑𝑙𝑠𝑖𝑛𝜃
Now,combiningthesethreestatements,wecanwrite,

𝑑𝐵∝
𝑟2

𝐼𝑑𝑙𝑠𝑖𝑛𝜃
ThisisthebasicformofBiotSavart’sLawNow,puttingthevalueofconstantk

𝑑𝐵=k
𝑟2
𝑑𝐵= 𝐼𝑑𝑙𝑠𝑖𝑛𝜃
𝜇0𝜇𝑟 ×
𝑟2
4𝜋
Here, μ0 used in the expression of constant k is absolute permeability ofair or vacuum and
it’s value is 4π10-7 Wb/A-m in SI system of units.

μr of the expression of constant k is relative permeability of the medium. Now, flux

density(B) at the point P due to total length of the current carrying conductor or wire can be

𝜇0 𝜇𝑟
represented as,

𝐵=∫𝑑𝐵⇒𝑑𝐵=∫ 𝐼𝑑𝑙𝑠𝑖𝑛𝜃
4𝜋 ×
𝑟2
𝐼𝜇
𝑠𝑖𝑛𝜃𝜇
=
0 𝑟
∫ 𝑑𝑙
4𝜋 𝑟2
 Figure1.3

IfDisthe perpendiculardistanceofthepointPformthewire,then

𝑟𝑠𝑖𝑛𝜃=𝐷

𝑟= 𝐷
𝑠𝑖𝑛𝜃

𝐼𝜇0𝜇𝑟 𝑠𝑖𝑛𝜃
Now,theexpressionoffluxdensityBatpointPcanberewrittenas

𝐵= ∫ 𝑑𝑙
4𝜋 𝑟2
𝑠𝑖𝑛3𝜃
= 𝑑𝑙
𝐼𝜇0𝜇𝑟 𝐷 2
∫
4𝜋
𝑙
=cot𝜃
𝐷
 𝑙=𝐷cot𝜃

Asperthe figure1.3
Therefore,

𝑑𝑙=−𝐷𝑐𝑠𝑐2𝜃𝑑𝜃

Finallytheexpression of Bcomesas,

𝑠𝑖𝑛3
𝐵= 𝜃 (−𝐷𝑐𝑠𝑐 2
𝜃𝑑𝜃)
𝐼𝜇0𝜇𝑟
∫ 𝐷2
4𝜋
𝐼𝜇0𝜇𝑟
=− ∫𝑠𝑖𝑛3𝜃𝑐𝑠𝑐2𝜃𝑑𝜃)
4𝜋𝐷
𝐼𝜇0𝜇𝑟
=− ∫𝑠𝑖𝑛𝜃 𝑑𝜃
4𝜋𝐷
Thisangle θdependsuponthelengthofthe wire andthe positionofthe point P.

Say for certain limited length of the wire, angle θ as indicated in the figure above varies
from θ1 to θ2. Hence, magnetic flux density at point Pdue to total length of the conductor is,

𝐼𝜇0𝜇 𝜃2
𝐵=− ∫ 𝑠𝑖𝑛𝜃 𝑑𝜃
𝑟 4 𝜋 𝜃1

=− [−cos𝜃]𝜃2
𝐼𝜇0𝜇𝑟
4𝜋𝐷
 =− [cos𝜃 −cos𝜃2]
𝐼𝜇0𝜇𝑟
1

4𝜋𝐷
Let’simaginethewireisinfinitelylong,thenθwillvaryfrom0toπthatisθ1=0toθ2=
π.

PuttingthesetwovaluesintheabovefinalexpressionofBiotSavartlaw,weget,
 𝐼𝜇0𝜇𝑟
𝐵=− 4𝜋𝐷 [cos0−cos𝜋]

𝐼𝜇0𝜇𝑟
=− 4𝜋𝐷 [1− (−1)]

=
𝐼𝜇0𝜇
𝑟 2 𝜋
𝐷
Thisisnothingbutthe expressionofAmpere’sLaw.

Magneticvectorpotential:

TheMaxwellequationsformagnetostaticsare:

A̅→.𝐵̅→( 𝑟→ ) =0

A̅→×𝐵̅→( 𝑟→ ) =𝜇 0 𝚥→ ( 𝑟→ )

PuttingthesetwointoHelmholtsequationfor𝐵̅→( 𝑟→ ) gives

→
̅→ ̅
𝜇 𝚥→(𝑟̅̅→′)
𝐵( 𝑟→ ) =A× ∫𝑑 3 𝑟′
0

4𝜋 |𝑟→−𝑟̅̅→′|

Expressingthemagneticfieldasthecurlofvectorpotential𝐴→( 𝑟→ )

𝐵̅→( 𝑟→ ) =A̅→×𝐴→( 𝑟→ )

′
𝐴→(𝑟→)= 𝜇 𝚥→(𝑟̅̅→′)
∫𝑑 𝑟
3

04 |𝑟→−𝑟̅̅→′|
𝜋
 Notice that the dependence of this vector potential on the current distribution is
exactlyparallel to the dependence of the electrostatic (scalar) potential on the charge distribution

𝜑(𝑟→ )= 1 𝜌( 𝑟̅̅→′ )
4𝜋𝜀 ∫𝑑 𝑟
3 ′
|𝑟→−𝑟̅̅→′|
0

Magneticfieldof alocalizedcurrentdistribution:

consider the magnetic field generated by a steady current distribution confined to some
region of space, and how it looks from some distance away. In other words, we’ll look at a
multipole analysis of the magnetic field at a point r→r→ outside the region of current.

weusetheexpansion

1 1 𝑟→.𝑟̅̅→′
= +
| 𝑟→−𝑟̅̅→′| | 𝑟→ | +⋯..
| 𝑟→ | 3
togetforthei-componentofthe vectorpotential
 Magnetic moment, force and torque on current distribution in an external
field:

The force law of Lorentz, 𝑓 = 𝑞(𝐸 + 𝑉 × 𝐵), is the expression of the force 𝑓 exerted
on a point-charge𝑞movingwithvelocity𝑉intheexternalelectricandmagneticfields𝐸(𝑟,𝑡)and
𝐵(𝑟,𝑡).Astraightforwardgeneralizationyields

𝐹𝐿(𝑟,𝑡)=𝜌(𝑟,𝑡)𝐸(𝑟,𝑡)+𝐽(𝑟,𝑡)×𝐵(𝑟,𝑡)

𝑇𝐿(𝑟,𝑡)=𝑟×𝐹𝐿(𝑟,𝑡)

where 𝐹𝐿is the Lorentz force-density, 𝑇𝐿is the corresponding torque-density, 𝜌 is

electriccharge-density, and 𝐽 is electric-current-density.

There is no a priori reason to distinguish between free and bound charges, nor between
free and boundcurrents. In other words, 𝜌(𝑟,𝑡) and 𝐽(𝑟,𝑡) inthe above Lorentz
formulationcould arise from various combinations of free and bound electric charges.

Thus,thetotalelectriccharge-density

𝜌total=𝜌free−∙ 𝑃,

andthetotalelectric current-density

𝐽total =𝐽free+𝜕𝑃/𝜕𝑡+𝜇0−1×𝑀,

giverisetoanetforce-density

𝐹𝐿=𝜌total𝐸+𝐽total×𝐵.

InconjunctionwithMaxwell’smacroscopicequations,theLorentzforce-densityofEquationleads

1
to the Maxwell-Lorentz stress tensor and the Livens EM momentum-density,

𝐽̅̅ (𝑟,𝑡)= (𝜀𝐸.𝐸+𝜇−1𝐵.𝐵)𝐼→−𝜀𝐸.𝐸−𝜇−1𝐵.𝐵

→
̅̅
2
𝐿 0 0 0 0
 𝑝̅̅̅𝐿̅̅→(𝑟,𝑡)=𝜀0𝐸×𝐵

𝜕𝑝𝐿
whichsatisfythe followingcontinuityequation:

A̅→.𝐽̅̅→
̅̅ + +𝐹 =0
𝜕𝑡
𝐿 𝐿

Equationisthestatementofconservationoflinearmomentum,justasEquationisthe corresponding
statement of conservation of energy.

According toEquation,theforce-density 𝐹𝐿(𝑟,𝑡)istherateofexchangeofmomentum between

EM fields and material media.

NotethattheLivensmomentum-densityofEquationisrelatedtothePoyntingvector appearing in
Equation via 𝓅𝐿(𝑟,𝑡) = 𝑆2(𝑟,𝑡) / 𝑐2.

theEMfieldshouldhavetheAbrahammomentum-density

𝓅𝐴=𝑆1/𝑐2

=𝐸×𝐻/𝑐2.

This means that 𝓅𝐿 in Equation should be replaced by 𝓅𝐴, and the Lorentz force-
density should be augmented by the difference between the Livens and Abraham momenta, that
is, the actual force-density should be given by 𝐹𝐿− 𝜕(𝜀0𝑀 × 𝐸)/𝜕𝑡.

Thehiddenmomentum-densitytrappedinsidemagneticmaterialsisthusdefinedas
𝓅hidden(𝑟,𝑡)=𝜀0𝑀×𝐸.

In other words, whenever magnetic dipoles are subjected to an 𝐸-field, the time-rate-of-
change of hidden momentum must be subtracted from the Lorentz force-density 𝐹𝐿in order to
arrive at the actual force exerted on the material medium.
 Magnetostaticenergy–Magneticinductionandmagneticfieldin macroscopic
media:

𝜇0
Theexpressionforthevectorpotential.

A̅̅̅̅→= →
̅̅
4𝜋 ∫ 𝑑𝜏
|𝑟→−𝑟̅̅→′|

The current densityisdivided into 2 components,

1) aconductioncurrentdensity,J~cand

2) anatomiccurrentdensity,J~a.

Inanalogyto the electricfield case,identifyJ~aasabound current density.

̅A̅→= 𝜇0 𝚥̅̅𝐶̅̅̅→(𝑟′) 𝜇0 𝚥̅̅𝑎̅̅→

̅ (𝑟′)
∫ + ∫
4 𝑑𝜏 |𝑟→−𝑟̅̅ | 4 𝑑𝜏 |𝑟→−𝑟̅̅→′|
→′

𝜋 𝜋

Expand the second term by a multipole expansion and complete the integration to obtain
the magnetic moment. The first non-zero term is the dipole which for one atom is;

→
̅̅ ̅̅ ×(𝑟→−𝑟̅̅→′)
𝑚̅̅̅̅𝑦̅̅→
→
|𝑟→−𝑟̅ → |𝑟→−𝑟̅̅→′|
3
′
|
th
Here ~rj is the location of the j atom and ~mj is its magnetic moment. Now define a
magnetization, M~ , as the magnetic moment per unit volume, and change the sum over the
atoms (sum over j) to an integral.
M̅̅̅̅→
̅ =𝑁(𝑚̅̅→⟩
ThenaverageA~overtheatomsinasufficientlylargesample,whereNistheaveragenumberof dipoles per
unit volume. Substitute this into the expression for the vector potential;

̅A̅→= 𝜇0 𝚥̅̅𝐶̅̅̅→(𝑟′)+ 𝜇0 M̅̅̅̅→

̅ ×(𝑟→−𝑟̅̅→′ )
∫ ̅̅→′ ∫𝑑
4 𝜋𝑑𝜏 4 𝜋𝜏
3

|𝑟→−𝑟| |𝑟→−𝑟̅̅→′|

M̅̅̅̅→
̅ × (𝑟→−𝑟̅̅→′)
The contributiontoA~from the magnetization isre-writtenas follows.

1
∫𝑑𝜏 =∫𝑑𝜏′M̅̅̅̅̅→(𝑟
)×A̅→ |𝑟→−𝑟̅̅→′|
3
′ ′

|𝑟→−𝑟̅̅→ |
′

Useavector identity

A̅→×(𝛼M̅̅̅̅→
̅ )=[A̅→𝛼]×M̅̅̅̅→
̅ +𝛼A̅→×M̅̅̅̅→
̅

Toobtain;

A̅→×M̅̅̅̅→
̅ M̅̅̅̅→
̅
𝜇̅̅̅̅→
𝐴̅̅̅ = ∫𝑑𝜏 ′
− ∫𝑑𝜏′
𝜇0
0

| 𝑟→−𝑟̅̅→′| 4 (𝑟→−𝑟̅̅→′)
𝛼

𝜋
4𝜋

The second term on the right can be evaluated on a boundary surface at a large distance

𝜇0
and set equal to zero (M~ = 0). Combining the remaining two contributions one obtains;

A̅̅̅̅→= 𝚥̅̅ ̅→+(A̅→×M̅̅̅̅̅→)

4 𝜋 ∫𝑑𝜏 ′ 𝐶̅̅
 |𝑟→−𝑟̅̅→′|

Then identifyanewcurrentdensity;

Sothat,

𝚥̅̅*̅̅̅→=𝚥̅̅𝐶̅̅̅→+j𝛼*
Ontheotherhand,considerthesurface term;

M̅̅̅̅→
̅
∫𝑑𝜏 A̅→×( M̅̅̅̅→
̅
×𝑑̅̅̅𝑎̅→
′

|𝑟→−𝑟̅̅→ )=∫
|𝑟→−𝑟̅̅ →
′
| ′
|

A̅→×B̅̅̅→
̅ =𝜇0𝚥̅̅*̅̅→
̅

̅ +A̅→×M̅̅̅̅→
=𝜇0[𝚥̅̅𝐶̅̅→ ̅ ]

Whichcanbe written as;

A̅→×(B̅̅̅̅→−𝜇0M̅̅̅̅̅→)=𝜇0𝚥̅̅𝐶̅̅→
̅

Defineanewquantity,H:

𝜇0H̅̅̅̅̅→=B̅̅̅̅→−𝜇0M̅̅̅̅→
̅

The variable, H, is usually called the magnetic field and then B is indentified as the
magnetic induction. For comparison, the relationship between the electric displacement, electric
field, and the electric polarization is;

withJ~cthe conduction,orfree,currentdensity.

Boundaryconditions:
 The magnetic field, H~ , and magnetic induction, B~ , are not necessarily parallel in
contrast to a classAdielectric where the electric displacement and electric field are parallel
and lineraly related. The boundary conditions are derived from A ·B = 0 and A ×H = 0 (no
free currents). Using Gaussian surfaces with Gauss’ law and Amperian loops with Ampere’s
law, boundaryconditions at the interface between magnetic materials are obtained inthe same
wayas for the electric field. These are;

A̅→.𝐵̅→=0

Whichmeans𝐵⊥iscontinuious

A̅→×𝐻̅→=0

Whichmeans𝐻‖iscontinuious

Uniformlymagnetizessphere:

Now consider a sphere of magnetic material, radius R, placed in a uniform

magneticfield, B0zˆ. The sphere is assumed to have a permeability, µ. There are no free
currents, so the field equations for r > R are;

The magnetic field Hisgeneratedfromamagneticpotential, φM.

Solvethisproblemusingseparationofvariables.Usesphericalcoordinateswithazimuthal symmetry. The

solution for r > R is;
Thisformischosentohavethecorrectasymptoticbehavioras

Insidethespherethefieldequationsare;

this is Poisson’s equation. In this case, B~ = µH~ so there is only a surface current density. Use
the definition of the magnetic susceptibility , χM, so that M~ = χMH~ .

The sphere is then uniformlypolarized in the ˆz direction. This means that A × ~ M~ = 0

everywhere but on the suface, There is an equivalent magnetic charge density in analogy to that
produced by a uniformly electrically polarized dielectric sphere.

Then find asolutionbyseparation ofvariables. Note thatthefield isnot onlyuniform but

directed along, ˆz.

Thusthefieldinsidethe spheremusthavethe form;

Thisisadipolefield.Thedipolemoment,m,ismagnetization,M,timesthe volume.
UNIT-III
 UNIT III

MAXWELLEQUATIONS

Faraday’slawsof induction:

Faraday’s law of electromagnetic induction (referred to as Faraday’s law) is a basic lawof

electromagnetism predicting how a magnetic field will interact with an electric circuit to
produceanelectromotiveforce(EMF).Thisphenomenonisknownaselectromagneticinduction.

Faraday’s law states that a current will be induced in a conductor which is exposed to a
changing magnetic field.

Lenz’s law of electromagnetic induction states that the direction of this induced current
will be such that the magnetic field created by the induced current opposes the initial changing
magnetic field which produced it.

The direction of this current now can be determined using Fleming’s right-hand rule.
Faraday’s law of induction explains the working principle of transformers, motors, generators,
and inductors.

The law is named after Michael Faraday, who performed an experiment with a magnet
and a coil. During Faraday’s experiment, he discovered how EMF is induced in a coil when the
flux passing through the coil changes.

Figure3.1
 When the magnet is held stationaryat thatposition, the needle of galvanometer returns to
zero position.

Now when the magnet moves away from the coil, there is some deflection in the needle
butoppositedirection,andagainwhenthemagnetbecomesstationary, atthatpointrespecttothe coil, the
needle of the galvanometer returns to the zero position.

Similarly, if the magnet is held stationary and the coil moves away, and towards the
magnet, the galvanometer similarly shows deflection.

It is also seen that the faster the change in the magnetic field, the greater will be the
induced EMF or voltage in the coil.

Faraday’sFirstLaw

Any change in the magnetic field of a coil of wire will cause an emf to be induced in the
coil. This emf induced is called induced emf and if the conductor circuit is closed, the current
will also circulate through the circuit and this current is called induced current.

Methodtochangethemagneticfield:

1. Bymovingamagnet towardsor awayfromthecoil

2. Bymovingthe coil intooroutofthe magneticfield

3. Bychangingthe area ofacoilplacedinthe magneticfield

4. Byrotatingthe coilrelative tothemagnet

Faraday’sSecondLaw

It states that the magnitude of emf induced in the coil is equal to the rate of change of
u x that linkages with the coil.The ux linkage of the coil is the product of the number of turns in the
coil and ux associated with the coil.
 FaradayLawFormula:

Consider, a magnet is approaching towards a coil. Here we consider two instants at time
T1 and time T2. Flux linkage with the coil at time,

Fluxlinkagewiththe coilattime,

Changeinuxlinkage,

Letthischangeinuxlinkagebe,

So,the Change inuxlinkage

Nowtherateofchangeof uxlinkage

Takederivativeonright-handsidewewillget
Therate ofchange ofuxlinkage

But according to Faraday’s law of electromagnetic induction, the rate of change of ux linkage is
equal to induced emf.

ConsideringLenz’sLaw.Where:

FluxΦinWb=B.

AB=magneticfieldstrength A

= area of the coil

HowToIncreaseEMFInducedinaCoil

By increasing the number of turns in the coil i.e N, from the formulae derived above it is
easilyseen that if the number of turns ina coil is increased, the induced emf also gets increased.
By increasing magnetic field strength i.e B surrounding the coil Mathematically, if magneticfield
increases, ux increases and if ux increases emf induced will also get increased.

Theoretically, if the coil is passed through a stronger magnetic field, there will be more
lines of force for the coil to cut and hence there will be more emf induced.

By increasing the speed of the relative motion between the coil and the magnet – If the
relative speed between the coil and magnet is increased from its previous value, the coil will cut
the lines of ux at a faster rate, so more induced emf would be produced.
 Maxwell’sdisplacementcurrent:

Weknowthatanelectriccurrentproducesamagneticfieldaroundit.J.C.Maxwellshowed
thatforlogicalconsistency,achangingelectricfieldmustalsoproduceamagneticfield.

Further, since magnetic fields have always been associated with currents, Maxwell
postulatedthatthiscurrentwasproportionaltotherateofchangeoftheelectricfieldandcalledit
displacementcurrent.

Todetermine this, let’s look at theprocessof charging a capacitor. Further, we will apply
Ampere’scircuitallawtofindamagneticpointoutsidethecapacitor.

Thefigure3.3showsaparallelplatecapacitorconnectedinacircuitthroughwhichatime-
dependentcurrenti(t)flows.WewilltrytofindthemagneticfieldatapointP,intheregionoutside the
capacitor.

Figure3.2

Consider a plane circular loop of radius r centred symmetrically with the wire. Also, the
planeoftheloopisperpendiculartothedirectionofthecurrentcarryingwire.
 Figure3.3

Duetothesymmetry,themagneticfieldisdirectedalongthecircumferenceoftheloopand
hassimilarmagnitudeatall pointson theloop.

However, as shown in the Figure 3.3when the surface is replaced by a pot-like surface
whereitdoesn’ttouchthecurrentbuthasitsbottombetweenthecapacitorplatesoratiffin-shaped
surface(withoutthelid)andAmpere’scircuitallawisapplied,certaincontradictionsarise.

These contradictions arise since no current passes through the surface and Ampere’s law
doesnot take that scenario into consideration. This leadsusto understand that there issomething
missing in the Ampere’s circuital law.

Also,the missingterm is such which enablesusto get the same magnetic fieldat point P
regardless of the surface used

Ifwelookatthefigure3.3,wecanobservethatthecommonthingthatpassesthroughthe surface
and between the capacitor plates is an electric field. This field is perpendicular to the
surface,hasthesamemagnitudeovertheareaofthecapacitorplatsandvanishesoutsideit.
 Hence,the electric fluxthroughthe surface isQ/ε 0(using Gauss’s law). Further, since the
charge on the capacitor plates changes with time, for consistency we can calculate the current as
follows:

i=ε0(dQ/dt)

This is the missing term in Ampere’s circuital law. In simple words, when we add a term
which is ε0times the rate of change of electric flux to the total current carried by the conductors,
throughthesamesurface,thenthetotalhasthesamevalueofcurrent‘i’forallsurfaces.Therefore,
nocontradictionisobservedifweusetheGeneralizedAmpere’sLaw.

Hence,themagnitude of BatapointPoutsidetheplatesisthesameatapointjustinside.
Now,thecurrentcarriedbyconductorsduetotheflowofchargeiscalled‘Conductioncurrent’.The new
term added is the current that flows due to the changing electric field and is called
‘Displacementcurrent’orMaxwell’sDisplacementcurrent’.

Maxwell’sequations–Vectorandscalar potentials:

Maxwell’sequations

Equation1isGauss’slawfrom electrostatics,andrelatestheelectric fieldEtothecharge density

ρ.
 2 is derived from the Biot-Savart law, and expresses the experimental fact that there
are no magnetic monopoles.

3 isFaraday’slaw,andshowsthatachangingmagneticfieldinducesanelectric field.

4 is a combination of Ampère’s law for steady currents with Maxwell’s correction

to Ampère’s law, in which a changing electric field induces a magnetic field. These four
equations (and their variants for use in dielectrics) are the basis of all classical
electromagnetism

A·B= 0we canwriteBasthe curlofavectorfield

Thisisstilltrue inthe dynamic case.Substitutingthisinto3we get

The LHS is the curl of a vector field that combines the electric field and the magnetic
vector potential, and since it is zero even in the dynamic case, we can write this vector field as
the gradient of a scalar field V .

Equations5and9effectivelyreplace2and3.Using5and9wecaneliminateEandBfrom1
and4:
Wecanrearrangethelastequationusingtheidentity

A×(A×A)=A(A·A)−A2A

soweget

Tosummarize:

These two equations comprise 4 equations (one from 18 and one for each vector
component in 19) for four functions (V and A), and their solution allows us to calculate both E
and B by means of 9 and 5, so they form a complete replacement for the original set of 4
Maxwell equations that we started with. This reduces the number of equations to be solved from
6 (for the 3 components of each of E and B) to 4. We can write these equations in a more
compact form using the d’Alembertian operator
anddefiningthefunction

Weget

∎2𝑉=A2𝑉-𝜇0
𝜖
𝛛2
𝑉 -------(22)
0 𝛛𝑡

=A2𝑉− + (A.𝐴)-------------------------------------(23)
𝛛𝐿 𝛛

𝛛𝑡 𝛛𝑡

∎2𝑉+ =A2𝑉+ (A.𝐴)-------------------------------------------(24)

𝛛𝐿 𝛛

𝛛𝑡 𝛛
𝑡

=-𝜌
𝜖0
----(25)

using18inthe lastline.Also,

∎2𝐴=A2𝐴-𝜇0 𝜖
𝛛2
𝐴 ----(26)
0 𝛛𝑡

∎2𝐴-A𝐿=−𝜇0𝐽 ---(27)

𝜌
using19.Insummary

𝜕𝐿
∎2𝑉+ =−
 𝜕𝑡 𝜖0

∎2𝐴-A𝐿=−𝜇0𝐽
 Gaugeinvariance–Waveequationandplanewavesolution–
Coulomb and Lorentz gauges:
Two common gauges in electrodynamics are the Coulomb gauge and the Lorentz gauge.
Each gauge amounts to specifying a value for A·A. The Coulomb gauge sets A·A= 0, and is the
gauge we used when introducing the magnetic vector potential.

To see that it’s always possible to transform from an arbitrary gauge to the Coulomb
gauge, we need to find a function λ such that the transformation

givesA·A=0.Todothis,wemust have

ThelastlineisjustPoisson’sequationandforany“reasonable”originalpotentialA’itispossible to solve
it.

IntheCoulombgauge,thepotentialformsofMaxwell’sequations:

reduceto
 The scalar potential V is therefore also a solution of Poisson’s equation, and once we
have found it, we can, in principle, solve the second equation (which is a wave equation with a
complicated driving term on the RHS) for the vector potentialA. The Lorentz gauge sets

which transforms5 and 6 to

Intermsofthe d’Alembertianoperator

we canwritethesetwoequationsas

so that both potentials now become solutions of the wave equation with a driving term,
but now V and A are decoupled.
 Pointing’stheorem:

When electromagnetic wave travels in space, it carries energy and energy density is
always associated with electric fields and magnetic fields.

The rate of energy travelled through per unit area i.e. the amount of energy flowing
through per unit area in the perpendicular direction to the incident energy per unit time is called
poynting vector. Mathematically poynting vector is represented as

the direction of poynting vector is perpendicular to the plane containing E and H .

Poynting vector isalso calledasinstantaneousenergyflux density.Hererateof energytransferP
isperpendiculartoboth Eand H.

Since it represents the rate of energy transfer per unit area, its unit is W/m 2 . Poynting
theorem states that the net power flowing out of a given volume V is equal to the time rate of
decrease of stored electromagnetic energyin that volume decreased bythe conduction losses. i.e.

Total power leaving the volume = rate of decrease of stored electromagnetic energy –
ohmicpower dissipated due to motion of charge

Lorentzforce:

Consider a particle having charge q. The force Feexperienced by the particle in the presence of
electric field intensity E is

Fe=qE

TheforceFmexperiencedbythe particleinthe presenceofmagneticfluxdensity Bis

Fm=qv×B
where visthe velocityofthe particle.The Lorentzforce experiencedbythe particleissimplythe sum
of these forces; i.e.,

F=Fe+Fm

=q ( E+v× B)

The term “Lorentz force” is simply a concise way to refer to the combinedcontributions of
theelectric and magnetic fields.

Conservationlawsforasystemofchargesandelectromagnetic fields:

Law ofconservationofcharge saysthatthenet chargeofanisolatedsystem willalways

remain constant.[1]This means that any system that is not exchanging massorenergywith its
surroundings will never have a different total charge at any two times.

For example, if two objects in an isolated system have a net charge of zero, and one
object exchanges one million electrons to the other, the object with the excess electrons will be
negatively charged and the object with the reduced number of electrons will have a positive
charge of the same magnitude.

Thetotalchargeofthesystemhasnotandwillnever change.
UNIT-IV
 UNIT IV

WAVEPROPAGATION

Planewavesinnon-conduction media:

Let us consider a plane EM wave.Aplane wave can be described as a wave where phase
isconstant over asetof parallelplanesinthe direction ofpropagation.Auniformplane wave isa wave
where magnitude and phase both are constant.

We know that, transverse electromagnetic waves in free space are uniform plane waves.
The electric and magnetic field vectors are mutually perpendicular to each other and to the
direction of propagation of the wave. T

The electromagnetic energy associated with the wave is equally shared between the
electric and magnetic field. The phases and their magnitudes are always constant. The EMenergy
is transmitted in the direction of propagation.

A non uniform plane EM wave is a wave where amplitudes of may vary within plane
normal to the direction of propagation. As a consequence H and E, are no longer in phase with
each other

Figure. 4.1

Letusconsider,auniformplaneEMwave travellinginthepositiveZ-directionasshown.Since, there is

no variation of electric field and magnetic field intensity in a plane normal to the direction of
propagation of the wave. We have,
 𝜕E 𝜕E
= =0
𝜕x 𝜕y

𝜕H 𝜕H
= =0
𝜕x 𝜕y

Letusassume that,

𝐸̅=𝚤^𝐸𝑥

𝐻̅=𝚥^𝐻𝑦

That is electric field vector is chosen along x- axis while magnetic field vector is chosen
along y axis so that, the

𝑆̅=𝐸̅×𝐻̅

=(i×j)𝐸𝑥𝐻𝑦

=𝚥 ^. (𝐸𝑥𝐻𝑦)

pointsinthe z-direction.

Let us assume the sinusoidally time varying nature of Electromagnetic fields, expressed
in phasor form as

𝐸𝑥=𝐸0𝑒j(𝜔𝑡−𝛽𝑧)------------------A(a)

𝐸𝑥=𝐸0𝑒 j(𝑡−𝛽𝑧)
𝜔

Similarly,

𝐻𝑦 =𝐻0𝑒j(𝜔𝑡−𝛽𝑧)----------------A(a)

ConsiderMaxwell’sThirdEquation,

𝜕𝐵̅
A×𝐸̅=−
𝜕𝑡
 𝛛𝐻̅ 𝛛𝐻̅̅
RHS= -𝜇
𝛛𝑡
=−𝚥^𝜇 −− −− −−−(4𝑎)
𝑦̅̅

𝛛𝑡
 𝚤^ 𝚥^ 𝑘^
𝜕 𝜕 𝜕
+𝑘^ 𝛛y
𝛛𝐸𝑥
LHS= =-𝚥^
𝜕 𝜕y 𝜕z
x 𝛛𝐸𝑧
𝛛x
𝐸𝑥 0 0
A×𝐸=−𝛛𝐸𝑥
𝛛z
+0 --(4b)

Thenwe have Equation 4(a)=4(b), hence

-𝚥^ =− 𝛛𝐻
𝚥^𝜇
𝛛𝐸𝑥
𝛛z 𝑦

𝛛𝑧
𝛛𝐸
𝑥 ---(5)
𝛛z 𝛛𝐻𝑦
=𝜇 𝛛z

DifferentiatingequationA(a)w.r.tz,andsubstitutinginequation(5)weobtain,

𝜕[𝐸 𝜕𝐻𝑦
𝑥]
=𝜇
𝜕z
𝜕z

𝛛[𝐸0𝑒j(𝜔𝑡−𝛽𝑧)]
=-j𝜇0𝐻𝑦
𝛛z

-j𝛽𝐸0𝑒 =−-j𝜇0𝐻𝑦
j(𝜔𝑡−𝛽𝑧)

𝛽𝐸𝑥= 𝜔𝜇0𝐻𝑦

𝐻𝑦
=
𝛽
𝜔𝜇
-- - - -(6)
0
 𝛽=
2𝜋

𝜆
Now

2𝜋𝑓
=
𝜆𝑓
 𝜔
=
𝑐

=𝜔√𝜇0𝜀0

Usingthevalueof

𝛽
=√𝜇0𝜀0
𝜔
inequation(6)weget,

𝐻𝑦= 𝛽 √𝜇0𝜀
.𝐸𝑥= .𝐸𝑥
𝜔𝜇0 0

𝜇0

𝐻 =√𝜀0
𝜇0
𝑦 .𝐸𝑥

Notethat,theratioEx/Hyhasthedimensionofimpedance.Itiscalledastheintrinsic impedance of
the free space to the EM wave and is denoted by 0 then we have,

𝜇0
η =√ =√
4𝜋×10−7
8.85×10−12 𝛺
0 𝗌0

η0 ≈ 377𝛺

Let us further assume that the EM wave is propagating through a lossless dielectric, i.e.
let  =0, then the wave equation reduces to

𝛛 Ex
2
=
𝛛 2E
𝛛z2 x
----(1)
𝛛t2
µ0ε0
𝛛 Hy
2
 = 𝛛2Hy-----------------------------------------------------
𝛛z2 µ0ε0 𝛛t2
(2)
 1
FromtheseequationsweunderstandthatthephasevelocityofEMwaveunder

considerationis 1 .
√µ0ε0

Thus, is the case of a plane EM wave propagating in free space, we have the
followingquantities

(i) α-attenuation constant= 0

(ii) 𝛽= 𝜔√𝜇0𝜀0 =phaseconstant radian/meter

(iii) 𝑉𝑝=phasevelocity= 1/√𝜇0𝜀0= 3×108m/s

𝜇 =377𝛺
√
(iv) 0
=intrinsicimpedance=
0
𝗌
0

In general,forlosslessmedium(𝛼=0);
SForauniform planeEMwave,we havethe followingrelations

𝘢=0,

β=ω√µ0ε
𝜂= 𝜇0
√ 𝜂0
𝜀 = √𝜀𝑟

𝜀
𝑤ℎ𝑒𝑟𝑒 /𝜀0 =𝜀𝑟(dielectricconstant )
1
𝑉 𝑝=
√𝜇0𝜀
𝑐
=
𝜀0

The quantityisalsoknown asRefractive Index of themedium.

Thus,we have
 R.I=√εr=
c
Vp

Linearpolarization:

Linear polarization refers to the orientation of the electric field vector in an

electromagnetic wave. In a linearly polarized wave, the electric field oscillates in a single plane,
while the magnetic field oscillates perpendicular to it.

The direction of polarization is determined by the direction of the electric field vector.
Linear polarization can be achieved bypassing unpolarized light through a polarizingfilter or by
using certain types of antennas.

AdvantagesofLinearPolarization

1. Longer read range: Linear polarization typically provides a longer read range compared
to circular polarization. This is because the energy of the RF signal is concentrated in a
single direction, allowing for greater distance between the reader and the tag.

2. Better penetration: Linear polarization is betterat penetratingcertain materials, such as

liquids and metals. This makes it suitable for applications where tags may be placed on or
near these materials.

3. Lower cost: Linear polarization antennas are generally less expensive to manufacture
compared to circular polarization antennas.

Circularpolarization:

Circular polarization is a property of electromagnetic waves, such as light, where the

direction of the electric field vector rotates in a circular pattern as the wave propagates.

In circularly polarized light, the electric field vector traces out a helix or spiral shape in
space. This is in contrast to linear polarization, where the electric field oscillates in a straightline,
and elliptical polarization, where the electric field traces out an ellipse.
 Circular polarization can be either right-handed or left-handed, depending on the
direction of rotation of the electric field vector.

AdvantagesofCircularPolarization

1. Better tag orientation: Circular polarization is less affected by the orientation of the tag.
This means that tags can be read from any angle, making it easier to achieve consistent and
reliable reads.

2. Multiple tag reading: Circular polarization allows for the simultaneous reading of
multiple tags, even if theyare indifferent orientations. Thismakesit idealforapplicationswhere
multiple tags need to be read quickly and efficiently.

3. Improved performance in multi-path environments: Circular polarization is less affected

by multi-path interference, where signals bounce off objects and create interference. This
makes it suitable for applications in environments with many reflective surfaces.

Reflection andrefractionataplaneinterface:

The normal laws of reflection and refraction which we know can be derived when theEM
wave is polarized perpendicular to the plane of incidence.
Using boundary conditions on field vectors, the expression for reflection coefficient Rand
transmission coefficient T can be derived in this case. Let us see the results for normal incidence
and oblique incidence.
In the case of a plane EM wave with E vector polarized perpendicular to the plane of
incidence , the reflection coefficient R and transmission coefficient T can be derived using
boundary conditions . For normal incidence, the results are as shown.

𝑅⊥
𝑛1−𝑛2 2
=( 𝑛1+𝑛)2

𝑇⊥
𝑛2 2𝑛1 2
= 𝑛 𝑛
( 1+𝑛)2
1
 Wavesinconductionmedium:

A sugar stone moving vertically down through a water column slowly dissolves in it due
to its high solubility.
Similarly, whenanEMwave ispropagatingthrougha conductingmedium,the amplitude of
Vectors and hence the EM energy goes on decreasing and finally the wave is absorbed in the
conducting medium.

Conduction current term dominates the displacement current term in the wave equation

𝜕𝐸̅
for E. the wave equation reduces to the one as shown, H and E

A2𝐸̅ ≈ 𝜇𝜎
𝜕𝑡
`Skindepthisdefinedasthedistanceaftertravellingwhichtheamplitudeofthe electricfield vector reduces
to 1/e times the original amplitude.
The loss of amplitude of E vector is measured in terms of a physical quantity known
as“Skin depth” .

𝛿=√ 2
𝜔𝜇𝜎

From the expression it is clear that for higher value of frequency of wave and
conductivity of medium, the skin depth is smaller.

Sea waterhasconductivitydue toitsmineral contents.Hence EMwave cannot propagate

over a larger distance through it.
Due to this, submarine communication is not possible with EM waves. We have to use
ultrasonic waves which are sound waves of frequency about 30 kHz for submarine
communication.
A good conductor has very high conductivity. Hence the conduction current is much
higher than the displacement current. Thus, we have𝜎≫𝜔 𝜀.
Inthiscase,the propagationconstantisgivenby
 𝛤=𝛼+j𝛽

𝛼=𝛽=√𝜋𝑓𝜇𝜎

Theskindepthisgivenby

1
𝛿=
√𝜋𝑓𝜇𝜎

2
=√
𝜔𝜇𝜎

Thesurfaceresistance Rsofthe filmisgivenby

Rs ω
µ =√
2σ

Theintrinsicimpedanceofagoodconductorisgivenby

jωµ
𝜂=√ σ

=(1+j)𝑅𝑠

Generally, all the dielectric materials have some conductivity. But the conductivities are
very small. However, when the conductivities cannot be neglected , the intrinsic impedance of
the dielectric to EM wave is given by

jω µ
𝜂=√
σ+jωε
 𝜇 𝜎
=√ [1− j
𝜀 ω ]−1/2
ε

σ
The term
ω𝝴
iscalled asthelosstangentsand isdefined as,

𝜎
tan𝜗=
ωε
Ifthelosstangentissmall,thentheintrinsicimpedanceisapproximatelygivenby,

𝛾= 𝛼+𝑖𝛽

Where

𝛼≅ 𝜎 𝜇
√
2 𝜀

𝛼≅ ω√𝜇𝜀

Notethatthederivationsoftheapproximateexpressionsfor𝑎and𝛽makeuseofbinomial expansions.

Propagationofwavesinarectangularwaveguide:

In electromagnetics, a waveguide confines electromagnetic signals within the structure,

preventing spreading, losses, andsignal transmissionfrom one point to another. Usually, a basic
waveguide can be constructed from a hollow conducting tube. If the conducting tube has a
rectangular cross-section, then it forms the rectangular waveguide. In the section below, we will
discuss various aspects of rectangular waveguide theory.

TheStructure ofaRectangular Waveguide:

Rectangular waveguides are the most commonly used waveguides. They consist of a
hollow metallic structure with a rectangular cross-section. A rectangular waveguide is usually
constructed with a length of a > b, where b is the breadth of the rectangle. A common trend for
the dimension of a rectangular waveguide is
 a=2b.

AdvantagesofRectangularWaveguides:

Theadvantagesofrectangularwaveguidesinclude:

 Wide frequencybandwidthforsingle-modepropagation
 Lowattenuation
 Excellentmodestabilityfor fundamentalpropagation modes

Inhomogeneouswaveequation andretarded potentials:

More generallythe retarded potentialscanbewritten as

[𝜌]𝑑3𝑟′
Ø(𝑟,𝑡)=∫
|𝑟−𝑟′|

[𝐽]𝑑3𝑟′
1
𝐴(𝑟,𝑡)= ∫
𝑐 |𝑟−𝑟′|

Thenotation[Q]meansthatQistobe evaluatedatthe retardedtime

|
′
[Q] =Q(𝑟′,t-|𝑟−𝑟 )
𝑐

Here | r – r’ | is the distance from the source point r 0 to the field point r.Now
electromagnetic ”news” travel at the speed of light.

When the source charge and current distributions are time dependant , it is not the
instaneous status of the source distribution that matters, but rather its condition at some earlier
time(called the retarded time) when the ”message” left.

Since this message must travel a distance |r − r ‘ |, the delay is ( r−r’ ) / cBecause the
integrands inand are evaluated at the retarded time ,these are called retarded potentials.
 Note that the retarded time is not the same for all points of the source distribution, the
most distant parts of the source have earlier retarded times than the nearby ones. Also note that
the retarded potentials reduce properly to

∫ [𝞀]r′
Ø(r)=
|r−r′|
1 [J]r′
A(r)= ∫
c |r−r′|

Asimilarsetofsolutions withthepotentialsevaluatedattheadvancedtime

𝑡𝑎𝑑𝑣=𝑡+ |𝑏𝑓𝑟−𝑟′|
𝑐

arealsomathematicallyvalidsolutionsoftheinhomogeneouswaveequationsand.

These areaccordinglycalledtheadvancedpotentials.

Although the advanced potentials are entirely consistent with Maxwell’s equations, they
violate the most sacred tenet in all of physics: the Principle of Causality.

They suggest that the potentials now depend on what the charge and current distribution
will be at some time in the future-the effect, in other words, precedes the cause.

Although the advanced potentials are of some theoretical interest, they have no direct
2
physical signifance. (Because the D’Alembartianinvolvest instead oft, the theoryitselfistime
reversal invariant, and does not distinguish ”past” from ”future”.)

Time assymetry is introduced when we select the retarded potentials in preference to the
advanced ones, reflecting the reasonable belief that electromagnetic influences propagate
forward, not backward, in time).
 Radiationfromalocalized source:

Like all electromagnetic fields the source of electromagnetic waves is some arrangement
of electric charge.But a charge at rest does not generate electromagnetic waves: nor does a
steady current.

It takes accelerating charges and changing currents to produce em waves, i.e., to radiate.
Once established, the em waves in vacuum propagate out to infinity, carrying energy with them;
the signature of radiation is this irreversible flow of energy away from the source.

Imagine a spherical shell at radius r; the total power P(r) passing out through this surface
is the integral of the Poynting vector:

P(r)=∫𝑆.𝑑𝐴

𝑐
=
4𝜋 ∫(𝐸×𝐵).𝑑𝐴

Thepowerradiatedisthelimitofthisquantityasrgoestoinfinity:

Prad= Limitr → ∞P(𝑟)

Thisisthe energy(perunit time)thatistransportedto infinity,

Oscillatingelectricdipole:

To understand the meaning of oscillating electric dipole, first we take two opposite
charges located at a distance d along z-direction at points A and B. Basically, this pairs ofcharges
form an electric dipole with associated dipole moment (p) being directed along z- direction (as
shown in figure ).
 Now,if we somehowchange thechargesat pointAandBwith time.

Moreover,ifthechangeissuchthatchargesatpointAandBbecometheoscillatory function of time

i.e., for example:

𝑞(𝑡)=𝑞0sin(𝜔𝑡)

whereωrepresentstheangularfrequencyoftheoscillation.Insuchasituation,thedipole moment associated

with the dipole is:

𝑝(𝑡)=𝑞0𝑑sin( 𝜔𝑡) 𝑒^𝑧

𝑝(𝑡)=𝑝0sin(𝜔𝑡)𝑒^𝑧

wherep0=q0drepresentsthemaximumvalueofdipole moment.

From the above equation, it is clear that the dipole moment also becomes the oscillatory
function of time. Such an electric dipole with the associated dipole moment oscillates with timeis
known as an oscillatory electric dipole.

RadiationfromanOscillatingElectric Dipole:

To calculate the radiation emitted by an oscillating electric dipole, we take an oscillating

dipole system as shown in the below figure.
 Now, our aim is determine the expression for the radiation at point P located at r1
distance from +q charge and r2 distance from -q charge (see the figure).

Moreover, the position vector r represents the position of point P with respect to the
center O of the dipole (midpoint on the axis of the dipole). The vector r is oriented at angle θwith
the axis.

Now,fromthe previouslectures,theretardedpotentialdue to+qchargeatpointPis:

𝑉1= 1
𝑞 cos[𝜔(𝑡− 1)]
0
r
𝑐
4𝜋𝗌 𝑟1
----(1)
0

In the above expression, potential is calculated at retared time tr = t – r1/c. Similarly, the
retarded potential due to -q charge at point P is

𝑉2=− 1
𝑞 cos[𝜔(𝑡− 2)]
0
r
𝑐
4𝜋𝗌 𝑟2
----(2)
0

Thenetretardedpotentialduetoboththechargesis:

𝑉(𝑟,𝑡)=
1
4𝜋𝗌0 𝑞
r
cos[𝜔(𝑡− 1)]
0
( 𝑐
−
𝑟1 c
𝑞0
o )-----------------
s[
𝜔
(𝑡
−
r2

)]
�
�
�
�
2
Now note that for the triangle AOP, length of the sides are AO=d/2, OP=r, AP=r1. So from the
law of cosines (also called as the cosine rule) for the triangle AOP:

𝑑 𝑑
𝑟2=𝑟2+( )2−2𝑟( )cos 𝜃
1
2 2

𝑟1 =√𝑟2+
𝑑2

4
−𝑟𝑑cos𝜃----------------------------------(4)
Similarly,forthetriangle BOP,usingthe lawofcosines,we canhave:

𝑑 𝑑
𝑟2=𝑟2+( )2−2𝑟( )cos(𝜋−𝜃
2
2 2

=√𝑟2+𝑑2+𝑟𝑑cos𝜃 (5)
UNIT V
 ELEMENTARYPLASMAPHYSICS

Electronplasmaoscillations:

Consider a small spherical region inside plasma and suppose that a perturbation in the
form of an excess of negative charge is introduced in this small region.
Because of spherical symmetry, the corresponding electric field is radial and points
towards the center, forcing the electrons to move radially outward.
After a small time interval, since the electrons gain kinetic energy in the course of their
motion, more electrons leave the spherical plasma region (due to their inertia) than is necessaryto
resume the state of electrical neutrality.
An excess of positive charge results, therefore, inside this region and the reversed
(outward, now) electric field causes the electrons to move inward.
This sequence of outward and inward electron movement in the spherical plasma region
continues periodically, resulting in electron plasma oscillations.
Inthiswaythe plasma maintainsitsmacroscopic neutralityonthe average,since the total
charge inside the spherical region, averaged over one period of these oscillations, is zero.
The frequency of these oscillations is usually very high, and since the ions (in view of
their much higher mass) are unable to follow the rapidity of the electron oscillations, theirmotion
is often neglected.
Figure5.1
To study the characteristics of the electron plasma oscillations we can use the cold plasmamodel,
in which the particle thermal motion and the pressure gradient force are not taken into account.

ne(r,t)=n0+ne′(r,t)-------------------------(1)
We shall neglect ionmotionand assume a verysmall electron densityperturbation such that

wherenoisaconstantnumberdensityand
ln~l«no.
Similarly,weassumethattheelectricfieldproduced,E(r,t),andtheaverageelectron velocity,
Ue(r,t),arefirst-orderperturbations,sothatthelinearizedequationscanbeused.Thelinearized
continuity and momentum equations become, respectively,

𝛛ne′
(𝑟,𝑡) +n0 A.ue (r,t)=0------------------------(2)
𝛛𝑡

=−
𝛛ue(𝑟,𝑡 𝑒
) 𝐸(𝑟,𝑡)---------------------------(3)
𝑚𝑒
𝛛𝑡

In the momentum equation we have assumed that the rate of momentum loss from the
electron gasduetocollisionsisnegligible. Consideringsinglycharged ions, thecharge densityis

𝜌(𝑟,𝑡)=−𝑒[𝑛0+ne (𝑟,𝑡)]+𝑒𝑛0
given by
′

=−𝑒ne′(𝑟,𝑡)-------------------(4)
where the ion densitywas considered to be constant and uniform, and equal to no(neglectingion
motion). Therefore,

A.E(r,t)=
𝜌(𝑟,𝑡)
=− ne′(𝑟,𝑡)---------------------------(5)
𝑒
𝜖0 𝜖0
Equationsareconstitute acomplete setofequationstobesolvedforthe variablesn~(r,t), ue(r,t),
andE(r,t). Takingthe divergenceof(1.3)andusing(1.2)to substitute forweobtain

𝛛2ne′

− 𝑚𝑒 A.𝐸(𝑟,𝑡)=0------------(6)
(𝑟,𝑡) 𝑒𝑛0
𝛛𝑡2

Combining(5)and(6)toeliminatedel·E, yields

𝛛2ne′
ne (𝑟,𝑡)=0---------------(7)
+𝜔𝑝 2
′
(𝑟,𝑡)
𝛛𝑡2 𝑒

𝜔𝑝𝑒
=( 𝑚𝑒𝜖
𝑛0𝑒2 )1/2-----------------(8)
0

is called theelectron plasma frequency.Equationshows thatn~ ( r,t)varies harmonically in time at

the electron plasma frequency,

ne′(𝑟,𝑡)=ne′(𝑟)exp(𝑖𝜔𝑝𝑒𝑡)-------------(9)
Infact,allfirst-orderperturbationshaveaharmonictimevariationattheplasma frequency Wpe·
To justify this statement it is convenient tostart with the assumption that all first-
orderquantities varyharmonicallyin time, as exp ( -I w t). Eqquation. (2) and (3) become,in this
case,

𝑖
ne′=− 𝑛 A.𝑢
𝜔
--(10)
0 𝑒

𝑢 𝐸---------------------------------(11)
=−
𝑖𝑒
𝜔𝑚
𝑒
𝑒

n =−
𝑛0𝑒
A.𝐸
′
whichcanbecombined into

e 2
𝜔 𝑚𝑒
--(12)
Substitutingthisexpressionforninto(5), yields

𝜔𝑝𝑒2

)A.𝐸
(1−
𝜔2
--(13)
 whichshowsthatanontrivialsolutionrequiresw=Wpe·
Therefore, all the perturbations vary harmonically in time at the electron plasma
frequency. Further, for all variables thereis no change in phase from point to point,implyingthe
absence of wave propagation. The oscillations are therefore stationary.
Also,(11) showsthattheelectronvelocityisinthesamedirectionastheelectricfield,so that
these oscillations are longitudinal.
Theelectronplasmaoscillationsarealsoelectrostaticincharacter.
In order to show this aspect of the oscillations, consider Maxwell curl equations with a

A×𝐸=𝑖𝜔𝐵
harmonic time variation,

---(14)
A×𝐵=𝜇0(𝐽− 𝑖𝜔𝜖0𝐸) ---(15)

𝐽=−𝑒𝑛0𝑢𝑒
Theelectriccurrentdensityis givenby

= 𝜔 𝑚𝑒
𝑖𝑛0𝑒2
(16)

where we have used(1.11)forUe.Therefore,

A×𝐵=−𝑖𝜇0𝜔𝜖0𝜖𝑟𝐸) ---(17)

wherewehavedefinedarelativepermittivityby

𝜖𝑟=1− 𝜔𝑝𝑒2

𝜔2
----(18)

Forthe electronplasmaoscillationswe havew=Wpe,sothat Er=0,and(17)reducesto

A×𝐵=0------------------------(19)
Sincethecurl ofthe gradientofanyscalarfunctionvanishes identically,
we maywrite

B=A 𝚿---------------------(20)
where psiis a magnetic scalar potential. Substituting (20) into (14) and taking the divergence of
both sides, we obtain the Laplace equation

A.(A𝚿)=A2𝚿=0----------------(21)
sincethedivergenceofthecurlofanyvectorfunctionvanishesidentically.
Theonlysolutionofthisequation,whichisnotsingularandfiniteatinfinity, is
psi=constant, so that B=0.
Hence,thereisnomagneticfieldassociatedwiththesespacechargeoscillations.

TheDebyeshieldingproblem:

To examine the mechanism by which the plasma strives to shield its interior from a
disturbing electric field, consider a plasma whose equilibrium state is perturbed by an electric
field due to an external charged particle.
For that matter, this electric field mayalso be considered to be due to one of the charged
particles inside the plasma, isolated for observation.
For definiteness, we assume this test particle to have a positive charge +Q, and choose a
spherical coordinate system whose origin coincides with the position of the test particle.
Weareinterestedindeterminingtheelectrostatic potential¢(r)thatisestablishednearthe test
charge Q, due to the combined effects of the test charge and the distribution of charged particles
surrounding it.
Since the positive test charge Q attracts the negatively charged particles and repels the
positivelycharged ones, the numberdensities oftheelectrons ne( r)and oftheionsni( r) will be
slightly different near the origin (test particle), whereas at large distances from the origin the
electrostatic potential vanishes, so that ne ( ∞) = ni (∞) = n 0.
Sincethisisasteady-stateproblemundertheactionofaconservativeelectricfield,we
have
 𝐸(𝑟)=−A𝜑(𝑟)------------------(1)

𝑛𝑒 ( ) exp[
𝑒𝜑(𝑟)
]
𝑟 =𝑛0 𝑘𝑇
---(2)

𝑛𝑖 ( ) exp[−
𝑒𝜑(𝑟)
]
𝑟 =𝑛0 𝑘𝑇
---(3)

𝜌(𝑟)=−𝑒[𝑛𝑒(𝑟) −𝑛𝑖(𝑟)] +𝑄𝛿(𝑟)---------------(4)

wherewehaveassumedthattheelectronsandions(ofchargee)havethesame temperature T.

𝑒𝜑(𝑟) 𝑒𝜑(𝑟)
𝜌(𝑟)=−𝑒𝑛 {exp[ ]−exp[− ] }+𝑄𝛿(𝑟)--------(5)
Thetotalelectriccharge densityp(r),includingthe testchargeQ,canbe expressedas

0
𝑘𝑇 𝑘𝑇

Substituting(1)and(5)intothe followingMaxwellequation,
𝜌 ( 𝑟)
A.𝐸(𝑟)=
𝜖0
----(6)

givesthedifferentialequation
𝑒𝜑(𝑟) 𝑒𝜑(𝑟)
A2𝜑(𝑟)−
𝑒𝑛0 Q
{exp[ ]−exp[− ] }=− 𝛿(𝑟)---(7)
𝜖0 𝑘𝑇 𝑘𝑇 𝜖0

whichallowstheevaluationoftheelectrostatic potential¢(r).
Inordertoproceedanalytically,we assume nowthat theperturbingelectrostatic potential is

𝑒𝜑(𝑟)≪𝑘𝑇---------------------------(8)
weak so that the electrostatic potential energyis much less than the mean thermal energy, that is,

Underthisconditionwe canusetheapproximation(makinga seriesexpansion)

𝑒𝜑(𝑟) 𝑒𝜑(𝑟)
exp[± 𝑘𝑇
]≈1± 𝑘𝑇
------(9)
Therefore,equation(7)simplifiesto

A2𝜑(𝑟)−
2
𝜑(𝑟)=−
Q
𝛿(𝑟)---------------(10)
𝜆𝐷
2 𝜖0

whereADdenotestheDebyelength

𝜖0𝑘𝑇 1
𝜆𝐷=(
𝑛 2)2
𝑒
0
= 𝑇𝑒𝑘 )2----------------(11)
1

1 (
𝜔𝑝𝑒 𝑚𝑒

Sincetheproblemhassphericalsymmetry,theelectrostaticpotentialdependsonlyonthe radial
distance r measured from the position of the test particle, being independent of the spatial orientation
of r.
Thus,usingspherical coordinates, (10)can bewritten (forr= 0) as

𝜑(𝑟)]−
1 𝑑 2
𝑟2 𝑑
𝑑 [𝑟2 𝜑(𝑟)=0--------------------(12)
𝑑𝑟 𝜆𝐷
2
𝑟

Inordertosolvethisequationwenoteinitiallythatforanisolatedparticleofcharge
+Q,infree space,the electricfieldisdirectedradiallyoutwardandisgivenby
𝐸 ( r) =
1
Q
^r
------(13)

4π𝛜 r2
0

sothattheelectrostaticcoulombpotentialpi c(r)duetothisisolatedchargedparticlein free

space is

𝜑 𝑐 ( r) =
1
Q ------(14)

4π𝛜0r
 Intheverycloseproximityofthetestparticletheelectrostaticpotentialshouldbethe same as that for
an isolated particle in free space.
Hence,itisappropriatetoseekthe solutionof(12)inthe form

𝜑(𝑟)=𝜑𝑐 (r)F(r)= Q
F(r
4 π𝛜0 ) -- - -(15)
r
 wherethe functionF(r)mustbesuchthatF(r) →1whenr→0.
Furthermore,theelectrostaticpotential¢(r)isrequiredtovanishatinfinity,thatis,¢→
0when r→∞.

𝐹(𝑟)
Substituting(15)into(12)yieldsthe followingdifferentialequationforF( r):

=
2F(r) 2
𝑑
d𝑟2
𝜆𝐷
-------(16)

ThissimpledifferentialequationforF(r)hasthe solution

Theconditionthat¢(r)vanishesforlargevaluesofrrequiresA=0.Also,theconditionthatF(r)
tendsto one whenrtendsto zerorequiresB=1.Therefore,thesolutionof (2.12)is

Thisresultiscommonlyknownasthe Debyepotential, sincethisnonrigorousderivation was

first presented by Debye and Ruckel in their theory of electrolytes.

Substituting¢(r) bythe Debye potential(18),we obtain

Toobtainthetotalchargeqt weintegrate(2.20)overall space,

Sincethe firstintegralgives-Q,whereasthe secondoneisequalto+Q,we findqt=0.

An important point to be noted in the result (18) is that, for r → 0, the Debye potential becomes

verylarge and the assumption e¢(r) « kTis unlikelyto be fulfilled. To verifythe validityof this
approximation, and consequently of (18), note that using (18) with Q =e we have

where N Dis the number of electrons inside a De bye sphere. Since N nis verylarge for virtually all
plasmas, itisevident thattheratio given in (2.22)ismuchless than one,exceptwhen ris less than
An/Nn. Therefore,
As a final point, we note that in the derivation of the Debye potential that appears extensivelyin
the literature, it is usual to ignore ion motion and to assume a constant ion number densityequal
to the unperturbed electron number density. In this case the factor of 2 disappears from (10), and
the expression for the Debye potential becomes
 Plasmaconfinementinamagneticfield:

Consider,forsimplicity,thespecialcaseinwhichthemagneticfieldisalongthe zaxis,thatis B = Bz,

so thatsimplifies to

from whichwe obtain

Also,from𝛻·B=0,wehave

since, in the local coordinate system, B is along thez axis. This last equation, together with (4),
implies that both p and B do not vary in

thezdirection.The solutionsof(2)and(3),combinedwiththisresult,give
Therefore, in the presence of an externally applied magnetic field, if the plasma is bounded, the
plasmakineticpressuredecreasesfromtheaxisradiallyoutwards,whereasthemagneticpressure
increases in the same direction in such a manner that their sum remains constant at each point,

The plasma kinetic pressure can be forced to vanish on an outer surface if the applied
magnetic field is sufficiently strong, with the result that the plasma is confined within this outer
surface by the magnetic field.
Let Bobe the value of the magnetic induction at the plasma boundary. Since the kinetic
pressure at the plasma boundary is zero (ideally), we can evaluate the constant in (6) from the
pressure equilibrium condition at t he plasma boundary. Therefore,

Themaximumfluidpressure thatcanbeconfinedforagivenappliedfieldBois, consequently,

A parameter beta, defined as the ratio of the kinetic pressure at a point inside the plasma,
to the confining magnetic pressure at the plasma boundary, is usuallyintroduced as a measure of
the relative magnitudes of the kinetic and magnetic pressures. It is given by

Notethatbetarangesbetween0and1,sincethe fieldinsidethe plasmaislessthanB0.

wecanalsoexpresstheparameterbetaas
An important property of a plasma is its diamagnetic character. Equation (7.7) implies that the
magnetic field inside the plasma is less than its value at the plasma boundary. As the kinetic
pressure increases inside the plasma, the magnetic field decreases. Under the action of the
externally applied B field, the particle motions give rise to internal electric

currents that induce a magnetic field opposite to the externally applied field. Consequently, the
resultant magnetic field inside the plasma is reduced to a value less than that at the plasma
boundary. The electric current , induced in the plasma, depends on the number density of the
charged particles and on their velocity. Therefore, as the plasma kinetic pressure increases, the
induced electric current and the induced magnetic field also increase, thus enhancing the
diamagnetic effect.

Magneto-hydrodynamicwaves:

Whenthefluidisnotperfectlyconducting,buthasafiniteconductivity,orifviscous effects are

present, the MHD waves will be damped
ManonmaniamSundaranarUniversity,DirectorateofDistance&ContinuingEducation, Tirunelveli

𝛛
𝛛𝑡
Forplanewavesolutions,thedifferentialoperators
andAarereplacedby-iwandik,
respectively,sothatthesetofdifferentialequations(7.1)to(7.3)arereplacedbyacorresponding set of
algebraic equations. Thus, we obtain

Substituting(4)and(6)into(5),andrearranging,we get

Comparingthisequationforu1with(19),weseethatweobtainthesameresultsasbefore, except that w2must

be multiplied by the factor

Alfivenwaves:

ForthecaseoftransverseAlfvenwavespropagatingalongB0,therelationbetweenwandk
becomes

ElectromagneticTheory
ManonmaniamSundaranarUniversity,DirectorateofDistance&ContinuingEducation, Tirunelveli

In order to simplify this result we shall assume that the correction terms corresponding to the
kinematicand magnetic viscosities are small, so that the term in the right-hand side of (8) can be
neglected. Thus,

where we have replaced w / k, in the right-hand side, by the first-order result (VA)· Using the
binomial expansion approximation
/
(1+x)1 2~1+x/2, valid

for x « 1, (9) can be further simplified to

The positive imaginary part in this expression for k(w) implies in wave damping. This is easily
seen bywriting k = kr+ i ki, with krand kireal numbers, and noting that

which represents a wave propagating along the z axis with wave number kr, but having an
exponentiallydecreasing amplitude, which falls to 1/e of its original intensityin a distance of1/ ki.

ElectromagneticTheory
ManonmaniamSundaranarUniversity,DirectorateofDistance&ContinuingEducation, Tirunelveli

Magnetosonicwaves:

ForlongitudinalmagnetosonicwavespropagatingacrossB0,thedispersionrelationbecomes

Tosimplifythisexpressionweconsiderthatthekinematicandmagneticviscositiesaresmalland neglect
the term involving the product Hence, afterrearrangement

In thetermsin theright-handside of(7.15) we canreplace bythe approximate result

so that (7.15) can be further simplified to give the following dispersion relation:

𝑘=
[𝜂𝑘
𝜔
]
2 𝜂𝑚
𝜔 2 1/2
) 2 2 1/
)
22(𝑉 +
+
7 𝑠2
2 )
+
-- - -(16)
(𝑉+𝑉
𝑉 𝑠
𝑠
𝐴 𝐴 (1+
7𝐴2
Thus, the attenuation of magnetosonic waves also increases with frequency and with
kinematic and magnetic viscosities, but decreases with increasing magnetic field strength.

ElectromagneticTheory