You are on page 1of 20

UIUC Physics 436 EM Fields & Sources II Fall Semester, 2015 Lect. Notes 13 Prof.

Steven Errede

LECTURE NOTES 13
ELECTROMAGNETIC RADIATION
In P436 Lect. Notes 4-10.5 (Griffiths ch. 9-10}, we discussed the propagation of macroscopic
EM waves, but we have not yet discussed how macroscopic EM waves are created. Using what
we learned in P436 Lect. Notes 12, we can now discuss how macroscopic EM waves are created.
Encrypted into Maxwells equations:

tot r , t B r , t
1) E r , t 3) E r , t
o t

E r , t
2) B r , t 0 4) B r , t o J tot r , t o o
t
is the physics associated with radiation of electromagnetic waves/electromagnetic energy,
arising from the acceleration {and/or deceleration} of electric charges (and/or electric currents).
In the P436 Lecture Notes #12, we derived the retarded electromagnetic fields associated with a
moving point charge q from the retarded Linard-Wiechert potentials:
q 1
Vr r , t where: r ctr c t tr , r r t r tr
4 o r

o q v tr
Ar r , t and: 1 rv tr c 1 r tr = retardation factor
4 r

With: Ar r , t tr Vr r , t c tr v tr c and: c 2 1 o o

We also derived the corresponding retarded electric and magnetic fields associated with a
moving point electric charge q:

Ar r , t
Er r , t Vr r , t
t

Coulomb
term for generalized term for radiation/


2 2
field/velocity field


acceleration field

q r
Er r , t c v t r u t r r u t r a t r
4 o r u t 3
r

1
Br r , t Ar r , t where: u tr cr v tr and: Br r , t r Er r , t
c

Coulomb
term for generalized term for radiation/


2 2
field/velocity field


acceleration field

1 q r
Br r , t r c v tr u tr r u tr a tr
c 4 o r u t 3
r

Professor Steven Errede, Department of Physics, University of Illinois at Urbana-Champaign, Illinois 1


2005-2015. All Rights Reserved
UIUC Physics 436 EM Fields & Sources II Fall Semester, 2015 Lect. Notes 13 Prof. Steven Errede

Microscopically:
The acceleration {and/or deceleration} of electric charges q and/or time-varying electric


current densities e.g. J nqv ; J t nq v t ~ nqa converts (a portion of the) virtual
photons (associated with the static Coulomb field, which individually have zero total
energy/zero-frequency) to real photons (which individually have finite total energy/finite
frequency f ), which then freely propagate outward/away from the source of time-varying electric
charge and/or electric current at the speed of light, c {in vacuum / free space}.
Since real photons individually carry energy/linear momentum/angular momentum,
macroscopic EM waves carry energy/linear momentum angular momentum away from the
source, in an irreversible manner these EM waves propagate away from the source time.
Energy/momentum must be input to the charged particle for this to happen energy/momentum
are {both} conserved in the radiation process.
{Note also that we can reverse the arrow of time t t in this process and thus learn about the
absorption of energy/linear momentum/angular momentum by electric charges/currents from
incoming/incident EM waves. . . .}

The total instantaneous power Pr r , t associated with radiation of EM waves from a source

(assumed to be localized) is obtained by integrating the retarded Poyntings vector S r r , t over
a large spherical shell of radius r a = characteristic dimension of a localized source this is
known as the far-field limit, when r :


E r , t B r , t da
1
Pr r , t S r r , t da
S o S
r r


The instantaneous power radiated is the limit of Pr r , t as r : Prrad t lim Pr r , t
r

The physical reason for this definition is simple. In the so-called near-zone, when r a ,

the (generalized) Coulomb field(s) (microscopically consisting of virtual photons) are dominant

in this region thus, time-varying but non-radiating E and B fields are present in proximity to
the source. The near-zone EM fields fall off/decrease/diminish as ~ 1 r 2 from the source.

In reality, for finite r , there is always a mixture of radiating and non-radiating EM fields
present that is associated with any source. Expressed in a graphical manner in terms of r a :

r a
. . . . 10 -4 -3
10 10 -2
10 -1 0
10 10 100 10 10 3 4 5. . . . .
10
=1
near-zone regime: far-zone regime
r r
1 1
a a
Generalized Coulomb field(s) Radiation/acceleration field(s)
dominant (virtual photons) dominant (freely propagating real photons)

2 Professor Steven Errede, Department of Physics, University of Illinois at Urbana-Champaign, Illinois


2005-2015. All Rights Reserved
UIUC Physics 436 EM Fields & Sources II Fall Semester, 2015 Lect. Notes 13 Prof. Steven Errede

The instantaneous EM power associated with the Generalized Coulomb field is:
GCF GCF
E
1
PrGCF r , t SrGCF r , t da r , t B r , t da
S o S
r r


But: ErGCF r , t ~ 1 r 2 (even faster than this, if the net charge = 0, e.g. for higher order EM
moments associated with electric dipoles, quadrupoles, octupoles, etc. )

And: BrGCF r , t ~ 1 r 2 (even faster than this, if the net charge = 0, e.g. for higher order EM
moments associated with magnetic dipoles, quadrupoles, octupoles, etc. )

S rGCF r , t ~ 1 r 4 (ever faster, for high-order EM moments than a point charge distribution)

But: Asphere 4 r 2 = area of sphere of radius r .

1 1 EM power associated with Generalized Coulomb


PrGCF r , t ~ 4 r 2 ~ 2
r r fields is only appreciable near the source.

Note that lim PrGCF r , t 0 i.e. no EM power is associated with G.C.F. at r
r

static sources do not radiate EM energy.


On the other hand, the instantaneous EM power associated with the radiation/acceleration fields is:
rad
E
1
Prrad r , t Srrad r , t da r , t Brrad r , t da
S o S
r


But: Errad ~ 1 r and Brrad ~ 1 r S rrad r , t ~ 1 r 2 , Asphere ~ r 2

Prrad r , t ~ 1 (i.e. Prrad r , t is independent of the radius of the enclosing surface S )

Thus, we can simply pick r to eliminate the PrGCF r , t contribution!!!

{n.b. for {unphysical} non-localized sources of time-varying EM radiation e.g. infinite planes,
infinitely long wires, infinite solenoids, etc. this requires a different approach altogether }
In general, arbitrary configurations of localized, time-dependent electric charge and/or

electric current density distributions, tr tr tr and J tr tr J tr can/do produce
EM radiation/freely-propagating EM waves.
As we learned in P435 (last semester), from the principal of linear superposition, we can
always decompose an arbitrary electric charge and/or current distribution into a linear
combination of EM moments of the electric charge/current distribution, i.e. electric monopole
(electric charge), electric and magnetic dipole, electric and magnetic quadrupole, etc.
moments. This is true {separately} for both static and time-varying EM moments of the electric
charge and/or current distribution(s).

Professor Steven Errede, Department of Physics, University of Illinois at Urbana-Champaign, Illinois 3


2005-2015. All Rights Reserved
UIUC Physics 436 EM Fields & Sources II Fall Semester, 2015 Lect. Notes 13 Prof. Steven Errede


For a point electric monopole field {E(0)}, i.e. r , tr q tr t r c 3 r

1 r , tr q tr t r c
Vr (E0)
r,t d
4 o v' r 4 o r

Where q tr total electric charge of the source at the retarded time tr . But electric charge is
(always) conserved, and furthermore, (by definition) a localized source is one that does not have
electric charge q flowing into or away from it. Therefore, the electric monopole moment
contribution/portion associated with the (retarded) potential(s) and EM fields is of necessity
static i.e. the electric monopole moment q has no EM radiation associated with it. In other
words, there can be no net transversely polarized EM radiation emitted from a spherically-
symmetric charge distribution! {See e.g. J. D. Jackson Classical Electrodynamics 3rd ed. p. 410
for additional/further details.}

The lowest-order electric multipole moment capable of producing EM radiation is that



associated with a time-varying electric dipole moment p r , tr qd r , tr or: q r , tr d .

Electric dipole (E1) radiation originates from r , tr

The lowest-order magnetic multipole moment capable of producing EM radiation is that



associated with a time-varying magnetic dipole moment m r , tr Ia r , tr or: I r , tr a .

Magnetic dipole (M1) radiation originates from J r , tr

Each time-varying, localized, higher-order EM moment contributes in alternating succession



between r , tr and J r , tr (i.e. electric vs. magnetic) multipole moment terms:

Time-varying localized Time-varying localized


electric moments: magnetic moments:

r , tr J r , tr
E 0 electric monopole q NO!
M 0 magnetic monopole g NO! no magnetic
charges anyways

E 1 electric dipole p qd

M 1 magnetic dipole m Ia


E 2 electric quadrupole QE 2qdd

M 2 magnetic quadrupole QM 2 Iaa

E 3 electric octupole
M (3) magnetic octupole
E 4 electric sextupole
etc M 4 magnetic sextupole

4 Professor Steven Errede, Department of Physics, University of Illinois at Urbana-Champaign, Illinois


2005-2015. All Rights Reserved
UIUC Physics 436 EM Fields & Sources II Fall Semester, 2015 Lect. Notes 13 Prof. Steven Errede

We will consider/discuss the case of EM radiation from an oscillating E(1) electric dipole and
then discuss case of radiation from an arbitrary localized source consisting of an arbitrary linear

combination of time-varying EM moments, a E n b M n , where E n and M n are
n 1
n n

nth-order time-varying electric and magnetic multipole moments, respectively.

E(1) Electric Dipole Radiation:



Consider an oscillating (i.e. harmonic/sinusoidally time-varying) electric dipole: p t qd t

where the charge separation distance varies in time: d t d t z d cos t z , 2 f

Then: p t qd cos t z p cos t z , with: p qd .

Equivalently, we can alternatively think of this as: p t q t d , with: d dz = constant,
and with time-varying/oscillating electric charge: q t q cos t .

Then: p t qd cos t z p cos t z , with: p qd . {n.b. same result!}

Either way one views/thinks about this, the physics associated with a harmonically time-

varying/oscillating electric dipole moment p t p cos t z qd cos t z is unchanged.
dq t
n.b. an electric current associated with the oscillating electric dipole: I t z , I t 0 0
dt
A picture of this, for a given moment/instant/snapshot in time t 0 is shown below:
z Observation
/ Field Point

r r r P r
q
z d 2
I t
r
r

r r r
p qd y

n.b. The choice of origin is deliberately chosen at
r the center of the localized charge distribution at
x the center of the oscillating electric dipole.
z d 2 q

n.b. exist (as always) some subtleties associated with the calculation of the retarded potentials
associated with moving point charges we will address these subsequently, but not right here /
right now so, well stick with the oscillating charge q t q cos t version for now

Professor Steven Errede, Department of Physics, University of Illinois at Urbana-Champaign, Illinois 5


2005-2015. All Rights Reserved
UIUC Physics 436 EM Fields & Sources II Fall Semester, 2015 Lect. Notes 13 Prof. Steven Errede


Now p t p cos t z refers to the time-dependence associated with itself. An observer at

field point P r at r sees the effects of the time-varying p t manifest themselves at a finite
time later, t tr r c or: tr t r c due to the retarded nature of this problem.

Thus, p t used in the formulae for the retarded scalar and vector potentials must be evaluated

at the retarded time tr , i.e. p t p tr q tr d qd cos tr z .

q tr q tr q cos tr q cos tr q cos tr cos tr




Vr E(1)
r,t
4 o r 4 o r 4 o r 4 o r 4 o r r




charge at charge at charge at charge at
+ d 2 z d 2 z + d 2 z d 2 z

I t dq tr
ArE(1) r , t o r d where: I tr q sin tr and: d dzz
4 r dt

Explicitly inserting the retarded time(s): tr t r c :

q cos t r c cos t r c
VrE(1) r , t
4 o r r
q z d 2 sin t r c
ArE(1) r , t o dz z
4 z d 2 r
Let us first focus our attention on calculating VrE(1) r , t . From the law of cosines {see P435
Lecture Notes 8 r.e. the derivation of the static multipole moment expansion}:

r r 2 rd cos d 2
2

However, we want to investigate EM radiation in the far zone where r d .


For this situation we can make the following approximation:
2

r r 1 cos r 1 cos
d 1 d d
But: 1 1 12 for 1 .
r 4 r r

1d d
Thus: r r 1 cos r 1 cos for r d .
2 r 2r

Similarly/correspondingly:

1 1 1 d 1
1 cos for r d , since: 1 for 1 .
r 1d r 2r 1
r 1 cos
2 r

6 Professor Steven Errede, Department of Physics, University of Illinois at Urbana-Champaign, Illinois


2005-2015. All Rights Reserved
UIUC Physics 436 EM Fields & Sources II Fall Semester, 2015 Lect. Notes 13 Prof. Steven Errede

Likewise, for the cos t r c term, for the far zone, when r d we have:

r d r d
cos t r c cos t 1 cos cos t cos
c 2r c 2c
r d r d
cos t cos cos sin t sin cos
c 2c c 2c

In order to proceed further, we need to make an additional simplifying assumption, namely


that the characteristic spatial dimension a of the source (here, a = d) is wavelength of
the emitted radiation, i.e. d { c f }. Thus we have: d c f { f 2 }, or:
d 2 c or: d c .

n.b. This assumption is tantamount/physically equivalent to saying that we will neglect any/all
time-retardation effects associated with finite EM propagation delay times over the dimensions
characteristic of/associated with the source i.e. changes in charge/current are essentially
coherent/instantaneous over the {small} spatial dimensions of the source, relative to the
wavelength of the emitted radiation.

Suppose we have a source (e.g. an atom) with a d 1 nm 10 emitting a f 1 Hz sine-
wave. Since EM radiation travels propagates at 1 ft 30 cm per nanosecond, a 1 nm dimension
source doesnt run into finite propagation decay time problems until:
109
ct a d (here) i.e. ct 1 nm t 0.3 1017 sec f 3 1017 Hz
3 108
d
Thus, provided that we additionally are in the regime of d , or d c , i.e. 1.
c
d d
Noting that if: 1 , then: cos 1 { 0 }.
c 2c

Then from the Taylor series expansions of cos x 1 and sin x x for very small x 1 , we see that:

d d d
cos cos cos 0 1 and: sin cos cos
2c 2c 2c

r d r
Thus: cos t r c cos t cos sin t
c 2c c
Thus:

q 1 d r d r
VrE(1) r , , t 1 cos cos t cos sin t
4 o r 2r c 2c c
1 d r d r
1 cos cos t cos si n t c
r 2r c 2c

Professor Steven Errede, Department of Physics, University of Illinois at Urbana-Champaign, Illinois 7


2005-2015. All Rights Reserved
UIUC Physics 436 EM Fields & Sources II Fall Semester, 2015 Lect. Notes 13 Prof. Steven Errede

Expanding this out:

q 1 r d r
VrE(1) r , , t cos t cos cos t
4 o r c 2r c
d r d 2 r
cos sin t cos 2 sin t
2c c 4rc c
r d r
cos t cos cos t
c 2r c
d r d 2 r
cos sin t cos 2 sin t
2c c 4rc c

q 1 d r d r
VrE(1) r , , t cos cos t cos sin t
4 o r r c c c
Thus:
qd cos 1 r r
cos t sin t
4 o r r c c c

But: p p qd . Hence in the far-zone d r and d :

p cos r 1 r
VrE(1) r , , t sin t cos t
4 o r c c r c

In the far-zone d r , with the additional restriction that weve also imposed on the source
EM radiation: d . We now additionally require/impose a third restriction that the far-zone
also be such that r , thus we have the hierarchical relation: d r for far-zone EM
c 1 1 1
radiation, namely that for r r , then i.e. for r .
c r r

Thus for the far-zone, when d r we can neglect the second term in the above expression
for VrE(1) r , , t .

p cos r
Then: Vr r , , t sin t in the far-zone, for d r .
E(1)

4 o c r c

Note that in the static limit, when 0 it is necessary to retain the second term in the above
p cos
expression; we obtain in this limit: VrE(1) r , {cf w/ P435 Lect. Notes same!}
4 o r 2

8 Professor Steven Errede, Department of Physics, University of Illinois at Urbana-Champaign, Illinois


2005-2015. All Rights Reserved
UIUC Physics 436 EM Fields & Sources II Fall Semester, 2015 Lect. Notes 13 Prof. Steven Errede


Now let us focus our attention on calculating ArE(1) r , t :
q z d 2 sin t r c
ArE(1) r , t o dz z
4 z d 2 r
Because the integration itself introduces a factor of d, then to first order
d
in d r 1 : r r 2 2rz cos z 2 r with: z
2
z d 2 sin t r c
Thus: dz sin t r c d
z d 2 r r
qd 1 r
Then: ArE(1) r , t o sin t z but: p qd
4 r c
p 1 r
ArE(1) r , t o sin t z
4 r c
Thus:


Note that in the static limit, when 0 then ArE(1) r , t 0 as we expect.
Now that we have obtained the (retarded) scalar and vector potentials VrE(1) r , t and

ArE(1) r , t it is a straight forward exercise to compute the associated (retarded) EM fields,



ErE(1) r , t and BrE(1) r , t :

E(1) E(1) ArE(1) r , t
Er r , t Vr r , t and: BrE(1) r , t ArE(1) r , t
t

In spherical coordinates:

n.b. Vr r ,t has no explicit dependence
E (1)


E(1) 1 1 p cos r
Vr r , t r sin t
r r sin 4 o c r c
p 1 r r 1 r
cos 2 sin t cos t r 2 sin sin t
4 o c r c rc c r c
p r 1 r r
cos t cos r cos sin t r sin sin t
4 o cr c c r c c
1
But for far-zone EM radiation, d r we have:
c r
r 1 r r
cos t cos cos sin t sin sin t
c
c r
c c


~ 1 ~ 1 ~ 1

Professor Steven Errede, Department of Physics, University of Illinois at Urbana-Champaign, Illinois 9


2005-2015. All Rights Reserved
UIUC Physics 436 EM Fields & Sources II Fall Semester, 2015 Lect. Notes 13 Prof. Steven Errede

1 r r
So we can neglect/drop the cos sin t r sin sin t terms.
r c c

p 2 cos r
VrE(1) r , t cos t r
4 o c r
2
c

ArE(1) r , t p 1 r o p 2 r
And: o
sin t z
cos t z
t 4 r t c 4 r c

But: z cos r sin in spherical coordinates.



ArE(1) r , t o p 2 r
cos t cos r sin
t 4 r c

E(1) E(1) ArE(1) r , t
Then for far-zone EM radiation, with d r : Er r , t Vr r , t
t

p 2 r o p 2 r
ErE(1) r , t cos t cos r
cos t cos r sin
4 o c r
2
c 4 r c

1 1
But: c2 or: o o
o o c2

p 2 r p 2 r
ErE(1) r , t o cos t cos r o cos t cos r
4 r c 4 r c

o p 2 r
cos t sin
4 r c

p 2 sin r
Or: ErE(1) r , t o cos t
4 r c
E(1) E(1)
Then: Br r , t Ar r , t

p 1 o p 1
with: ArE(1) r , t o r
sin t c z
4 r
r
sin t c cos r sin
4 r

Thus:

0
0

1 0
A 1 1 Ar 0 1 A
BrE(1) r , t sin A r r A rA r
r sin
r sin r
r r

10 Professor Steven Errede, Department of Physics, University of Illinois at Urbana-Champaign, Illinois


2005-2015. All Rights Reserved
UIUC Physics 436 EM Fields & Sources II Fall Semester, 2015 Lect. Notes 13 Prof. Steven Errede

Thus:
1 A
BrE(1) r , t rA r
r r
1 p 1 r 1 r cos
o r sin t sin sin t c
r 4 r r c r
o p r 1 r
cos t sin sin t sin
4 r c c r c
o p r 1 r
cos t sin t sin
4 r c c r c
neglect

1
Again, here, because r , thus
c r

p 2 sin r E(1) o p 2 sin r


BrE(1) r , t o cos t
c
and: E r , t cos t
4 c r 4 r
r
c

E(1) 1 E(1)
Now since r , once again we see that: Br r , t r Er r , t , i.e. B E and B r
c
Note also that:

a.) ErE(1) and BrE(1) both vary as ~ 1 r .

b.) ErE(1) r , t and BrE(1) r , t are in-phase with each other.

c.) ErE(1) r , t and BrE(1) r , t have the same angular dependence ( ~ sin ).

rad
The EM radiation energy density, uE(1) r , t associated with the oscillating E(1) electric dipole
for far-zone EM radiation { d r } is:
Erad Mrad 1 1
rad
uE(1) r , t uE(1) r , t uE(1) r , t o ErE(1) r , t ErE(1) r , t BrE(1) r , t BrE(1) r , t
2 o
1 2 p 2 4 sin 2 2 r o 2 p 2 4 sin 2 2 r Joules
o o 2 2 cos t 2 2 cos t
2 16 r c o 16 c r
2
c m
3

1 o p 2 4 sin 2 2 r o p 2 4 sin 2 2 r
cos t
c cos t
2 16 2 c 2 r 2 16 2 c 2 r 2 c
Erad Mrad
n.b. uE(1) r , t uE(1) r , t using: c 2 1 o o or: o 1 o c 2 .

o p 2 4 sin 2 2 r Joules
u rad
E(1) r , t cos t for: d r far zone limit
16 2 c 2 r 2 c m
3

Professor Steven Errede, Department of Physics, University of Illinois at Urbana-Champaign, Illinois 11


2005-2015. All Rights Reserved
UIUC Physics 436 EM Fields & Sources II Fall Semester, 2015 Lect. Notes 13 Prof. Steven Errede

The EM energy radiated by an oscillating electric dipole, in the far zone { d r } limit is
given by Poyntings vector:
r
rad 1 E(1) E(1)
S E(1) r , t
Er r , t Br r , t
o

r
r
rad 1 o p 2 p 2 2 r
S E(1) r , t o sin cos 2 t
o 4 r 4 rc c
r

rad o p 2 4 sin 2
2 r Watts
Or: S E(1) r , t cos t r
2
16 2 c r 2
c m
Radial outward flow of EM field energy for: d r far zone limit
The EM radiation linear momentum density associated with an oscillating electric dipole, in the
far zone { d r } is given by:

rad 1 rad
rad
E(1) r , t S
o o E(1) r , t SE(1) r , t
c2

rad o p 2 4 r
cos 2 t sin 2 r 2
kg
Or: E(1) r , t
16 r c
2 2 3
c m -sec
Radial outward EM field linear momentum for: d r far zone limit

The EM radiation angular momentum density associated with an oscillating electric dipole, in
the far zone { d r } is given by:
rad rad
E(1) r , t r E(1) r ,t
rad o p 2 4 r
cos 2 t sin 2 r r 0
kg
E(1) r , t
16 r c
2 2 3
c m-sec
No EM field angular momentum for: d r far zone limit
rad
n.b. The exact E(1) r , t 0 i.e. ignore restrictions on far-zone limit, keep all higher-order

terms . . . we have neglected ErE(1) ~ r term which is non-negligible in the near-zone d ~ r
and also in the so-called intermediate, or inductive zone ~ r .

12 Professor Steven Errede, Department of Physics, University of Illinois at Urbana-Champaign, Illinois


2005-2015. All Rights Reserved
UIUC Physics 436 EM Fields & Sources II Fall Semester, 2015 Lect. Notes 13 Prof. Steven Errede

Time-Averaged Quantities for E(1) Radiation from an Oscillating Electric Dipole:

1 t 1 t 1
Recall the definition of time average: A t A t dt

Ao cos 2 t dt Ao
t 0 t 0 2

The time-averaged EM radiation energy density associated with an oscillating electric dipole is:

o p 2 4 sin 2 Joules
rad
uE(1)
r , t 2 2 for: d r far-zone limit
32 c r m
2 3

rad
The time-averaged |Poyntings vector|, which is also the intensity I E(1) of EM radiation
associated with an oscillating electric dipole is:

rad o p 2 4 sin 2 Watts


r SE(1)
1
r , t c o E r , t d r
rad 2
I E(1)
2 2
for:
32 c r m
E(1) r 2
2 far-zone limit

rad
rad Watts
We also see that: I E(1) r S E(1) r , t c uE(1) r , t
rad
2
.
m
The time-averaged EM radiated power associated with an oscillating electric dipole is:

rad
rad o p 2 4 2
PE(1) S E(1)
r , t S r , t da r 2 sin 2 sin
d
d
32 2 c r 2 0 0
d cos

o p
2 4
o p 2 4


16 32 c
2 2
2 0
sin 2 d cos
16 c 2 0
sin 2 d cos

Let: u cos , du d cos , 0 u 1 , u 1 , sin 2 1 cos 2 1 u 2


1
1
1 u du u u 3 1 1 2
1 1 1 2 4
2
1
3 1 3 3 3 3

The time-averaged radiated power associated with an oscillating electric dipole is:
rad
rad

o p 2 4 n.b. PE(1) r , t has
PE(1) r , t (Watts) for: d r far-zone limit
12 c no r-dependence!

Note that time-averaged radiated power varies as the 4th power of frequency!

The time-averaged EM radiation linear momentum density associated with an oscillating electric
dipole is:

rad 1 rad 1 rad p 2 4 sin 2 kg d r


E(1) r , t 2 SE(1) r , t uE(1) r , t r o 2 3 2 r 2 for:
c c 32 c r m -sec far-zone limit

Professor Steven Errede, Department of Physics, University of Illinois at Urbana-Champaign, Illinois 13


2005-2015. All Rights Reserved
UIUC Physics 436 EM Fields & Sources II Fall Semester, 2015 Lect. Notes 13 Prof. Steven Errede

The time-averaged EM radiation angular momentum density associated with an oscillating


electric dipole is:

rad o p 2 4 sin 2 kg
E(1) r , t r rad
E(1) r , t
2 3 r r 0 for:
d r
32 c r m-sec far-zone limit

rad
n.b. The exact E(1) r 0 i.e. ignore restrictions on far-zone limit, keep all higher-order

terms . . . we have neglected the ErE(1) ~ r term which is non-negligible in the near-zone d ~ r
and also in the so-called intermediate, or inductive zone ~ r .

rad
rad o p 2 4 sin 2 Watts
Note that because: E(1)
I r S E(1) r , t
2 2
32 c r m
2

rad rad
SE(1) r , 0, SE(1) r , , 0 since: sin 2 0 sin 2 0
i.e. no EM radiation occurs along the axis of the electric dipole ( z axis)

EM radiation for E(1) electric dipole is peaked/maximum at 2 (then sin 2 1 )


i.e. maximum EM radiation occurs to the axis of the electric dipole:
rad
S E(1) r ,

~ sin 2
o p 2 4
32 2 cr 2
0 2
rad
Thus, the intensity profile I E(1) r in 3-D {for fixed r} for E(1) electric dipole radiation is
donut-shaped - rotationally invariant in , as shown in the figure below:

14 Professor Steven Errede, Department of Physics, University of Illinois at Urbana-Champaign, Illinois


2005-2015. All Rights Reserved
UIUC Physics 436 EM Fields & Sources II Fall Semester, 2015 Lect. Notes 13 Prof. Steven Errede

Griffiths Example 11.1:


o p 2 4
The time-averaged power for E(1) electric dipole radiation is PE(1)
rad
.
12 c
rad
Note that PE(1) ~ 4 (or ~ f 4 , or ~ 4 )
c 3 108
For red light: red 780 nm f red 3.85 1014 Hz
red 780 10 9

c 3 108
For violet light: violet 350 nm f violet 8.57 1014 Hz
violet 350 10 9

PE(1)
violet
f 4
8.57 1014
4

Hence: 2.23 24.67 .


4

f red 3.85 1014


violet
PE(1)
red

PE(1)
rad
~ 4 explains why the sky is blue! Sunlight {unpolarized light} incident on O2 & N2
molecules in the earths atmosphere stimulates the O & N atoms vibrates the {bound} atomic
electrons at {angular} frequency , causing them to oscillate as electric dipoles! Solar EM
radiation at a given angular frequency is thus absorbed and re-emitted in this EM radiation +
atom scattering process.
The above formula for EM power radiated as E(1) electric dipole radiation by such atoms, by
time-reversal invariance of the EM interaction, is also the EM power absorbed by atoms, thus we
see that because of the 4 -dependence of PE(1)
rad
, the higher frequency/shorter wavelength
radiation (i.e. blue/violet light) is preferentially scattered much more so than the lower
frequency/longer wavelength radiation (i.e. red light).

The Earths sky appears blue {e.g. to an observer on the ground, or even e.g. a space shuttle
astronaut in orbit around the earth} because the light from the sky is scattered (i.e. re-radiated)
light, which is preferentially in the blue/violet portion of the visible light EM spectrum. The
scattering of EM radiation off of atoms is known as Rayleigh scattering.
Note that precisely same physics also simultaneously explains why the Sun appears red e.g. to
an observer on the ground at sunrise and sunset because at these times of the day, path that the
sunlight takes through the atmosphere is the longest, relative to that associated e.g. with its
position at {local} noon. If the higher-frequency blue/violet light is preferentially scattered out
of the beam of sunlight, what is left in the beam of sunlight after traversing the entire thickness
of the Earths atmosphere is the lower-frequency, orange-red light.

Professor Steven Errede, Department of Physics, University of Illinois at Urbana-Champaign, Illinois 15


2005-2015. All Rights Reserved
UIUC Physics 436 EM Fields & Sources II Fall Semester, 2015 Lect. Notes 13 Prof. Steven Errede

Note that the Sun is a black-body radiator its EM spectrum peaks in the infra-red region
thus it is NOT flat by any means {also is affected by frequency-dependent absorption in the
atmosphere}:

Note the log scale on the vertical axis! Thus, there is not much violet light in the Suns EM
spectrum, and hence there is a delicate balancing act of flux of EM radiation from the Sun
{convoluted} with its black-body spectrum and the scattering of this radiation by atoms in the
Earths atmosphere thus we see the sky as blue. Thus, if the black-body temperature of the sun
was different, then the color of the Earths sky in the visible portion of the EM spectrum would
also be different compare the black-body spectra of our Sun e.g. with that of Spica (260 ly
away in the Virgo constellation) and Antares (a red giant 600 ly away in the Scorpio
constellation):

16 Professor Steven Errede, Department of Physics, University of Illinois at Urbana-Champaign, Illinois


2005-2015. All Rights Reserved
UIUC Physics 436 EM Fields & Sources II Fall Semester, 2015 Lect. Notes 13 Prof. Steven Errede

Light from the Sun is unpolarized (i.e. it consists of all polarizations, randomly oriented over
time). However, because EM waves are transversely polarized (defined by the orientation of

the E -field vector) an incident EM plane wave from the Sun with polarization in a given

direction ( to k -propagation direction) will (transitorily) induce electric dipole moments in

gas atoms in earths atmosphere, via pmol mol Einc , where mol is the molecular
polarizability at {angular} frequency {see P435 Lect. Notes 12 and P436 Lect. Notes 7.5}.
The axis of induced electric dipole moments will be || to the plane of polarization of incident
wave at that instant, hence the scattered radiation emitted by the atom will be preferentially at
90 2 (i.e. ) to the axis of the (induced) electric dipole of gas atoms in earths

atmosphere. There are two specific/limiting cases to consider (a) when the incident E -field

vector is vertical and (b) when the incident E -field vector is horizontal. Random polarization is
then an arbitrary linear combination of these two limiting cases:

(a.) Einc vertical:

Einc Binc
pinduced

Atom kinc
Note: Escat || pinduced || Einc for scat = 90o
Escat
scat 90 o (max probable direction of emission).
E(1) electric dipoles oscillating to line
-of-sight preferentially tend to radiate
in the line-of-sight direction.
Bscat

k scat = Line of sight

Earth


(b.) Einc horizontal:


Binc

kinc

pinduced
Same atom and same Atom
Einc
observer, but observer
doesnt see this scattered
scat 90o
radiation E(1) electric
dipoles oscillating along
the line-of-sight do not
radiate in that direction. Bscat
= Line of sight
Escat k scat

Earth

Professor Steven Errede, Department of Physics, University of Illinois at Urbana-Champaign, Illinois 17


2005-2015. All Rights Reserved
UIUC Physics 436 EM Fields & Sources II Fall Semester, 2015 Lect. Notes 13 Prof. Steven Errede

Because the blue light an observer sees from a given portion of the sky is due to the
preferential scattering of E(1) electric dipole-type Rayleigh scattering of sunlight/solar EM

radiation off of gas atoms in the Earths atmosphere, with Escat to the line-of-sight, this
radiation has a net polarization i.e. the light from the sky is polarized, especially so away from
the sun, i.e. in the northern portions of the sky {in the northern hemisphere} !!! You can very
easily observe/explicitly verify this using a pair of polaroid sunglasses try it some time!!!
It is beneficial to wear polaroid sunglasses e.g. when out boating on a lake in order to
reduce glare from {polarized} sunlight reflected off of the surface of the water!!!
As mentioned above, at sunrise or sunset, the sun appears red when an observer is looking
directly at the sun, because the blue/violet light is ~ 25 more preferentially scattered out of the
beam of light incident from the sun {per unit thickness of atmosphere} than red light. Thus
sunlight when the sun is near the horizon consists predominantly of what remains red light.
Note that this is also true for moonrise and moonset the moon will {likewise} have a
reddish hue at these times, and note that this is also true e.g. for the case of an eclipse of the
moon by the Earth.
One can also observe this same phenomenon e.g. using a glass pitcher of milk diluted with
water because milk molecules are efficient Rayleigh scatterers of visible light! Heres a simple
experiment that you can carry out at home, e.g. using a flashlight:

18 Professor Steven Errede, Department of Physics, University of Illinois at Urbana-Champaign, Illinois


2005-2015. All Rights Reserved
UIUC Physics 436 EM Fields & Sources II Fall Semester, 2015 Lect. Notes 13 Prof. Steven Errede

The {scalar} EM wave characteristic radiation impedance of an antenna is exactly as we


defined the characteristic impedance of a waveguide; noting here that we are dealing with
manifestly transverse waves for EM wave radiation from an E(1) electric dipole antenna:
rad rad rad
E r E r E r
Z antenna r rad


H r 1 rad 1 rad
B r B r
o o

Lets check the SI units of this definition:

E
Volts
m Volts
m
Volts
Ohms


B o Teslas Henrys

m

N
A-m
N 2
A Amps

For E(1) electric dipole radiation, the EM wave characteristic radiation impedance in the far-
zone limit ( d r ) with c 1 o o is:

E(1) o p 2 sin r E(1) o p 2 sin r


Er r , t cos t and: Br r , t cos t
4 r c 4 c r c

o p 2 sin r
o cos t
4 r c o
(1)
Z antenna r o c Z o 120 377
o p sin
2
r o
cos t
4 c r c

Where:
o 4 107 Henrys/m = magnetic permeability of free space / vacuum
o 8.85 1012 Farads/m = electric permittivity of free space / vacuum

o 4 107 Henrys / m = {scalar} characteristic


And: Zo 120 377 impedance of free
o 8.85 1012 Farads / m space/the vacuum.

Thus we see that E(1) electric dipole antennae (in the far-zone limit ( d b r )) are
perfectly impedance-matched for propagation of E(1) EM waves into free space / vacuum!
Note also that the far-zone ( d r ) EM wave characteristic radiation impedance

Z r has no spatial and/or frequency dependence.
(1)
antenna

Professor Steven Errede, Department of Physics, University of Illinois at Urbana-Champaign, Illinois 19


2005-2015. All Rights Reserved
UIUC Physics 436 EM Fields & Sources II Fall Semester, 2015 Lect. Notes 13 Prof. Steven Errede

The EM Wave Radiation Resistance of an Antenna:


The {scalar} EM wave radiation resistance of an antenna Rrad is defined in terms of the antenna
power Prad and the amplitude of the current I flowing in the antenna:
antenna
Prad I 2 Rrad
antenna
or:
antenna
Rrad Prad
antenna
I2 (Ohms)

For an E(1) electric dipole antenna: I = q = amplitude of current flowing in the dipole.
In the far-zone limit, i.e. d r :
(1)
o p 2 4 o q d o 2 d 2 n.b. Rrad
2 2 4
is
R (1)

12 c3 I 2 12 c q 2 2 12 c
rad
frequency-dependent!

In the far-zone limit, i.e. d r :


o 2 d 2 2d 2 2d 2 o 2 d 2 (1) o
(1)
Rrad o
c Z rad But: Z o (1)
Z rad
12 c 12 c 2 12 c 2 o 12 c 2
o
In the far-zone limit, d r :

2d 2 1 d
2
(1)
R Zo Zo
12 c 12 c
rad 2

d
However, in the far-zone limit, d r we have: 1
c
(1)
Thus, we see that the EM wave radiation resistance Rrad associated with E(1) electric dipole
antenna in the far-zone limit ( d r ) is much less than the EM wave characteristic
(1),M(1)
radiation impedance Z rad Z o 120 377 of an electric dipole antenna:

1 d
2
(1)
R Z o Z o 377
12 c
rad

20 Professor Steven Errede, Department of Physics, University of Illinois at Urbana-Champaign, Illinois


2005-2015. All Rights Reserved

You might also like