UIUC Physics 436 EM Fields & Sources II Fall Semester, 2015 Lect. Notes 8.5 Prof.

Steven Errede

LECTURE NOTES 8.5

Reflection and Refraction of EM Waves at the
Boundary of a Dispersive/Absorbing/Conducting Medium

Consider a situation where monochromatic plane EM waves are incident on a boundary

between two media {located at z = 0 and lying in the x-y plane} as shown in the figure below.
For the sake of simplicity, the 1st medium (z < 0) is linear/homogeneous/isotropic, non-absorbing
/ non-dispersive and non-magnetic. The 2nd medium is also linear/homogeneous/isotropic and
non-magnetic, but is both absorbing/dispersive and conductive.

Because of the above-stated EM properties of the two media, in medium (1) the incident and

reflected wavevectors kinc and krefl are purely real, whereas in medium (2), the transmitted

wavevector is complex: ktrans ktrans i trans . Note that the monochromatic plane
EM wave(s) have the same frequency , independent of the medium they are propagating in.
THE ELECTRIC FIELDS:

E r, t E r ei kinc r t

Medium (1) inc oinc

kinc ,krefl real, constant wavevectors
(non-absorbing) k r t
Erefl r , t Eorefl r e refl

i ktrans r t

Medium 2) complex
Etrans r , t Eotrans r e ktrans ktrans i trans wavevector
(absorbing /
conducting)

On the boundary/interface (lying in the x-y plane at z = 0) we must have (for arbitrary times, t):

ei k


i kinc r t
i krefl r t
i kinc r t i ktrans r t r t
trans r
e e and: e e trans
e



kinc r krefl r and: kinc r ktrans r ktrans i trans r ktrans r i trans 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 8.5 Prof. Steven Errede

On the interface/boundary lying in the x-y plane at z = 0: n.b. inc , refl and trans

The 1st equation: kinc r krefl r gives usual Law of Reflection: are defined with respect
to the z unit normal of
the interface/boundary.
kinc r sin inc krefl r sin refl

but: kinc v1 krefl v1 because both the incident and reflected waves are in the same
non-dispersive/non-absorbent medium {medium (1)}.

sin inc sin refl inc refl

The 2nd equation: kinc r ktrans r ktrans i trans r ktrans r i trans r ,
after equating the real and imaginary parts, gives:

Re :kinc r ktrans r and Im :0 trans r

In general, ktrans and trans are not parallel to each other!!

i.e. In general, ktrans and trans will point in different directions!! Why/How???

Physically, the requirement that trans r 0 on the interface/boundary {lying in the x-y

plane at z = 0} means that trans Im ktrans must be to the boundary (i.e. trans z ),

since the position vector r {pointing from the origin 0, 0, 0 to an arbitrary point
x, y, z 0 on the boundary} lies in the x-y plane.
Inside Absorbing/Conducting Medium (2) (i.e. z > 0):
i ktrans r t


i k r t

Because ktrans ktrans i trans , then: Etrans z , t Eotrans r e Eotrans r e trans z e trans ;

Thus, we see that:

trans Im ktrans defines planes ( to the boundary/interface) of constant electric field amplitude in medium (2).

Im k

trans trans is the unit normal to the planes of constant electric field amplitude in medium (2).

Furthermore:

ktrans Re ktrans defines planes of constant phase in medium (2)

Re k is the unit normal to the planes of constant phase in medium (2)

ktrans trans

{n.b. in general, planes of constant phase could be in any direction, depending on the material!}

See the following figure for an explicit diagram of what is occurring in this physics problem:

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 8.5 Prof. Steven Errede

On the interface/boundary {lying in the x-y plane at z = 0}, at an arbitrary point x, y, z 0 :

n.b. inc , refl and trans
Re :kinc r ktrans r means: kinc r sin inc ktrans r sin trans are defined with respect
or: kinc sin inc ktrans sin trans to the z unit normal of
the interface/boundary.

Because the wave vector ktrans is complex, we do not have a simple relation between the
wavenumber ktrans and the {angular} frequency in the dispersive, conducting medium (2),
i.e. ktrans v2 as we did for the incident and reflected wavevectors kinc v1 krefl v1
associated with their respective EM waves propagating in the non-dispersive, non-conducting,
non-magnetic medium (1).

In medium (1), the index of refraction n1 is purely real and independent of frequency
(i.e. medium (1) is non-dispersive), thus the {real} relation v1 c n1 is valid in medium (1),
whereas in the dispersive, conductive medium (2), the {frequency-dependent!} complex wavenumber
k2 and index of refraction n2 are related to each other by n2 c k2 , thus the index
of refraction in medium (2) is complex and frequency-dependent n2 n2 i2 , and thus
the speed of propagation in medium (2) v2 c n2 is also complex.

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 8.5 Prof. Steven Errede

Note that we can also determine the relationship between complex wave vector k2 and
complex index of refraction, n2 of the absorbing/dispersive, non-magnetic, conducting

medium (2) from (either of) the wave equation(s) associated with the transmitted E and B -fields

in medium (2), which can be written (e.g. for complex Etrans ) as:

1 2 Etrans r , t n22 2 Etrans r , t
2 Etrans r , t
v22 t 2 c2 t 2

For plane harmonic (i.e. monochromatic) EM waves propagating in absorbing/dispersing non-

magnetic medium (2), noting that: Etrans gives: iktrans and: Etrans t gives: i

Thus, the characteristic equation (aka the dispersion relation) associated with the above
differential equation is:

n 2 n
2
2
2

iktrans iktrans 2 2 i i ktrans 2 2 or: ktrans

2
n22
c c c

2 c
But: ko = vacuum wavenumber = , where: o .
c o f

ktrans
2
n22 n22 ko2 where: ko = purely real quantity.
c c

If we explicitly write out the real and imaginary parts of ktrans ktrans i trans

associated with the above iktrans iktrans term and the real and imaginary parts of
n2 n2 i2 associated with the above n22 term:

k


i trans ktrans i trans n2 i2 n2 i2 ko2
trans

(ktrans ktrans 2iktrans trans trans trans ) n22 2in2 2 22 ko2

ktrans
2 2 iktrans trans costrans trans
2

ktrans
2
trans
2
i 2ktrans trans costrans n22 22 i 2n2 ko2
Equating the real and imaginary parts of the LHS and RHS of the above equation, we see that:

k n k
2
trans
2
trans
2
2
2
2
2
o and: ktrans trans cos trans n2 2 ko2

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 8.5 Prof. Steven Errede

Thus, for ktrans ktrans i trans and n2 n2 i2 we have the complex relations:

2
1.) k n k
2
trans
2
trans
2
2
2
2
2
o with vacuum wavenumber: ko
o

c
2.) ktrans trans cos trans n2 2 k 2
o and vacuum wavelength: o c f , 2 f

We also have the relation:

3.) kinc sin inc ktrans sin trans where: kinc n1 ko n1
c
Inserting relation 3.) into relations 1.) and 2.) above, after some algebra these relations yield the
following relation:

ktrans cos trans i trans n1ko

n 2in sin
2
2
2
2 2 2 2
inc n1ko
n22
sin 2 inc
n12 n12

{n.b. if medium (2) is L/H/I non-conductive/non-magnetic/non-dispersive medium (i.e. like medium (1)), then
trans 2 0 and it is easy to show that this relation then reduces to: kinc sin inc ktrans sin trans fcn }

Let us define: a
n2 n 2
2 2
n 2in
2
2
2
2 2 2
complex!
2
n1 n1 n1

Then: ktrans cos trans i trans n1ko a 2 sin 2 inc

We define the Law of Complex Refraction {for this particular boundary/interface situation} as:

n1 sin inc n2 sin trans

where: trans complex angle: trans trans itrans

with: trans e trans and: trans m trans

Physically, trans e trans has the usual physical meaning (except that it is now

frequency-dependent), whereas trans m trans has no simple/easy physical meaning.

The Law of Complex Refraction can be rewritten as:

n2 n 2 sin inc
a 22
n1 n1 sin trans

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 8.5 Prof. Steven Errede

Then:
a 2 sin 2 trans sin 2 inc a 2 1 cos 2 trans sin 2 inc

1 cos 2 trans sin 2 inc a 2 cos trans 1 sin 2 inc a 2
But:

ktrans cos trans i trans n1ko a 2 sin 2 inc n1koa 1 sin 2 inc a 2

But: cos trans 1 sin 2 inc a 2 {from above}

ktrans cos trans i trans n1koa 1 sin 2 inc a 2 n1koa cos trans

i.e. ktrans cos trans i trans n1koa cos trans

Solve for a :

ktrans cos trans i trans n2 k cos trans i trans

a n2 trans
n1ko cos trans n1 ko cos trans

The {complex} E and B fields involved at the interface are:
i k r t 1
Incident wave: Einc r , t Eoinc r e inc Binc r , t kinc E inc r , t n.b. this form of
B takes care of
i k r t 1
Erefl r , t Eorefl r e refl Brefl r , t krefl E inc r , t
everything!!!
Reflected wave:

i ktrans r t 1


Transmitted wave: Etrans r , t Eotrans r e Btrans r , t ktrans Etrans r , t

1


ktrans Etrans r , t i trans Etrans r , t

The boundary conditions at the interface {lying in the x-y plane at z = 0} are:

BC 1) (normal D continuous): 1 E1 2 E 2 ( free 0 on the interface/boundary)

BC 2) (tangential E continuous): E1 E 2

BC 3) (normal B continuous): B1 B 2
1 1
BC 4) (tangential H continuous): B1 B2 ( K free 0 on the interface/boundary)
1 2
B1 B 2 if 1 2 o

(medium (1) and medium (2) both non-magnetic)

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 8.5 Prof. Steven Errede

On the interface/boundary at z = 0 (for any arbitrary space-point, e.g. (x, y, z) = (0, 0, 0) and time t):
TE Polarization Case:

BC 2) E oinc E orefl E otrans ko c 2 o , o c f

BC 4) Boinc cos inc Borefl cos refl Botrans cos trans kinc n1ko , krefl n1ko and inc refl

= kinc E oinc cos inc krefl E orefl cos refl ktrans E otrans cos trans i trans E otrans

= n1ko E oinc E orefl cos inc ktrans cos trans i trans E otrans

= n1ko cos inc E oinc E orefl ktrans cos trans i trans E otrans

k cos trans i trans

or: Eoinc
E orefl trans
n1ko cos inc
Eotrans

k cos trans i trans

but from BC 2) E otrans E oinc E orefl E oinc E orefl trans

n k
1 o cos inc
Eoinc E orefl


ktrans cos trans i trans n2
Skipping the details of the algebra, but using: a
n1ko cos trans n1
It can be shown that:

E orefl cos inc a cos trans

TE Polarization:
TE cos inc a cos trans
Eo
inc
n.b. these have the identical functional
Similarly, it can also be shown that: forms of those the lossless dielectric case!

E orefl a cos inc cos trans

TM Polarization:
Eo
inc

TM a cos inc cos trans

E o
2

Reflection Coefficient: R refl

E oinc

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 8.5 Prof. Steven Errede

Using the above ratios for TE and TM polarization plus realistic/detailed/full-blown n2

expression for metal, reflection coefficient/reflectance vs angle of incidence for TE and TM
polarized EM waves (in visible light/optical region of EM spectrum) is shown below for a typical
air-metal interface:

For TM polarization, a metal has no Brewster angle where R B 0 , but instead has a dip
(i.e. minima) where B (for a lossless dielectric) used to be. The angular location of this minima
/ dip for TM polarization is known as the principal angle of incidence, 1 .

At normal incidence inc refl trans 0 , both TE and TM polarization give the same ratio:

E orefl 1 a

inc 0 1 a
E o
inc

Thus the reflection coefficient of the metal/conductor at normal incidence, inc 0 is :

E o
2
1 a n2 n 2
2

R inc 0 refl where: a 2 2

Eoinc 1 a n1 n1
inc 0

If (for simplicity) medium 1) is the vacuum, then: n1 1.0 nair

1 n 22
2

R inc 0
2
And:
1 n 22
2
2

1 n
2

For lossless/dispersionless dielectrics 2 0 , then: R inc 0

2
.
1 n
2
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 8.5 Prof. Steven Errede

For metals, the extinction coefficient 2 2 2 is large, e.g. in the visible light range.
R inc 0 unity 85 95% for many metals in visible light range.
C n e2
In the low frequency region, we have shown that: n where: C e
2 o me
2 8 o
Then: Rlow inc 0 1 1 known as the Hagen-Rubens formula
frequency n C

{Works well for metals in the far-infrared portion of the EM spectrum experimentally verified}
The high reflectivity of metals at optical and higher frequencies is caused by (essentially) the
same physics as that for a tenuous plasma!
The complex total electric permittivity for an absorptive/dispersive conducting medium is:

bound

n b 2
e n f osc
P2

Tot bound free o 1 e j
2
o me j 1 1 j i j i o
2 2

Where:
bound n bound
f josc oscillator strength of jth bound resonance, with j 1
f josc 1

1 j 02 j neb e2 3 o me = {angular} frequency of jth resonance of bound valence electrons.

0 j ke me = natural {angular} frequency of jth resonance of bound valence electrons.
j

m = electron mass in medium ( me for electron e.g. in vacuum!)
e

j width/damping constant of jth resonance of bound valence electrons.

neb # density (#/m3) of bound atomic electrons in the valence bands.
o width/damping constant of free/conduction electrons resonance at 0 0 rad/sec
P nef e2 o me plasma frequency associated with free/conduction electrons
nef # density (#/m3) of free/conduction electrons in the metal.

At high frequency, o the total complex permittivity of the metal/conductor takes the
approximate form:

2

Tot bound free bound o P for o

For even higher frequencies, but P , but also where 1 j for {all of} the bound/
valence band resonances in the metal, the complex electric permittivity is given approximately by:

P
2

Tot o 1 for o , 1 j for valence band resonances, but P .

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 8.5 Prof. Steven Errede

Visible light penetrates only a very short distance sc vis 1 vis c P into the metal
and is almost entirely reflected.
When the frequency of the incident EM wave is increased still further, into the UV and x-ray
region then P and the metal suddenly becomes transparent the transmittance T increases
from zero and the reflectance R 1 T therefore decreases.

A Simplified Model of EM Wave Propagation in the Earths Ionosphere and Magnetosphere

Propagation of EM waves in the earths ionosphere is very similar to that in a tenuous plasma,
however, the earths weak DC magnetic dipole field:

Bearth 0.3 Gauss 0.3 104 Tesla 30 Tesla at the earths surface

significantly changes the nature of EM wave propagation in the earths ionosphere, and thus
cannot be neglected in the theory formalism.

Consider a tenuous electronic (i.e. e -only) plasma of uniform number density ne with a

strong, static and uniform magnetic field B Bo with monochromatic plane EM waves

propagating in the direction parallel to B Bo z .

If the {complex} displacement amplitude r of the electronic motion is small and
damping/collisions are neglected, then the approximate equation of motion is given by the
following inhomogeneous 2nd order differential equation:

me r r , t eBo r r , t eE r e it

Note that we can safely neglect the influence of the magnetic Lorentz force term ev B acting

on the electrons associated with the {complex} B -field of the EM wave, as long as BEM Bo .

We specifically/deliberately consider here circularly polarized monochromatic plane EM waves

propagating in the z direction B Bo , which in complex notation can be succinctly written as:
LCP

E r , t 2 E r , t where the polarization vectors are e.g.
1 i 1 x and:
2 y
RCP
If the monochromatic plane EM waves polarization vectors are: 1 x and
2 y and:

B Bo z , then we see that: B 1 x and also that: B 2 y .

The magnetic Lorentz force term eBo r r , t eBo z r r , t can then only have
components in the x-y plane - i.e. it can only have components along the x y or
1
2 axes.

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 8.5 Prof. Steven Errede

A steady-state solution to the above 2nd order inhomogeneous differential equation for the

electrons {complex} displacement amplitude re r at the space point r is:

e e
re r E r i.e. re r , t re r e it E r e it
me B me B

where B eBo me = electron precession frequency spiraling around the magnetic field lines
and the sign depends on the handedness of the circular polarization {TBD, momentarily}.

We can understand this relation better in the rest frame of electrons precessing with frequency

B about the direction of B Bo z (= direction of propagation of the EM wave) the static B -
field is eliminated it is replaced by a rotating electric field of effective frequency B ,
where again the sign depends on the handedness of the circular polarization (see below).

The {complex} harmonic oscillation of each electrons displacement re r , t re r e it

also constitutes a {complex} oscillating electric dipole moment p r , t ere r , t ere r e it ,

and thus results in a corresponding {complex} macroscopic electric polarization r , t

(= electric dipole moment/unit volume) r , t ne p r , t , where ne electron # density

and corresponding {complex} relation r , t o e E r , t and thus has a corresponding
{real!} macroscopic electric permittivity o 1 e .

For circularly-polarized monochromatic plane EM waves propagating parallel to B Bo z ,
the macroscopic electric permittivity is:

P2 n e2 eB
o 1 where: P2 e and: B o
B o me me

where the upper sign () in the denominator is for a LCP EM wave, the lower sign (+) in the
denominator is for a RCP EM wave.

For circularly-polarized monochromatic plane EM waves propagating anti-parallel to B Bo z ,
the macroscopic electric permittivity is:

P2
o 1


B

LCP and RCP monochromatic plane EM waves propagate differently in a tenuous electronic

plasma, depending on whether the EM wave propagation direction is || to (or anti-||) to B .
The earths ionosphere is bi-refringent !!!

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 8.5 Prof. Steven Errede

If the direction of EM wave propagation not perfectly || to (or anti-||) to B , then one simply
replaces B B cos in the above formulae, where opening angle between propagation

o o
wavevector k and B , i.e. k B k B z kB k z kB cos
o

A tenuous electronic plasma is also anisotropic !!!

A typical maximum number density of free electrons in the tenuous electronic plasma of the
earths ionosphere is ne ~ 1010 1012 electrons/m3, which corresponds to a plasma frequency of
p ne e 2 o me 6 106 6 107 (radians/sec).

The precession frequency of electrons in this plasma, in the earths magnetic field is:
B eBo me 5.3 106 (radians/sec) for Bo Bearth 30 Tesla .

P2 P2
k B : o 1 k anti- B : o 1
B B

Note that circularly polarized EM waves with 0 cannot propagate in plasma

because they are exponentially attenuated.

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 8.5 Prof. Steven Errede

An incident monochromatic plane EM wave with circular polarization such that 0

in the tenuous electronic plasma of the earths ionosphere will be totally reflected, the other
circular polarization state (with 0 ) will be partially transmitted/partially reflected.

A linearly-polarized monochromatic plane EM wave incident on the tenuous electronic

plasma of the earths ionosphere will have a reflected wave that is elliptically polarized with
its major axis rotated away from the direction of the polarization of incident wave.
The earths ionosphere has several layers of plasma with electron densities characteristic of
that/each layer, which can also vary in time and space, e.g. depending on the solar wind / solar
storms, as well as earths own weather (thunderstorms, etc.) as well as geological stresses in
earths crust fault lines/earth quakes and volcanic activity.
The number density of free electrons in each ionosphere layer has a maximum at a certain
height inferred from studying reflected pulses of varying frequency, sent vertically upwards
from the ground.
A short EM wave pulse of frequency 1 sent upwards from the ground actually enters the
bottom of the ionospheric layer, because the number density of electrons is small there and also
because the slope dne dh is shallow. However, when the electron number density ne reaches a
critical value for the incident, upward-going EM wave, i.e. 1 P ne e 2 o me , the EM wave
is reflected back, as shown in the figure below:

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 8.5 Prof. Steven Errede

The behavior of at low frequencies is responsible for the magnetospheric propagation

phenomenon known as whistlers. As 0 , (see graph on page 12) because:

P2
o for 0
B

Propagation in the tenuous electronic plasma of the earths ionosphere occurs {because
P
0 } but the wavenumber k corresponds to a highly dispersive medium!
c B
B
Energy transport is governed by the group velocity, here: vg 2v p 2c
P2

Pulses of EM waves (e.g. created in/during a lightning discharge) have frequency

components that propagate in the earths ionosphere at different speeds higher higher
propagation speeds, lower lower propagation speeds.

Spectral Analysis of a Whistler - Frequency vs. Time Plot:

Hear the audio file(s) of whistlers!

If interested in reading more about

whistlers:
See e.g. R. A. Helliwell, Whistlers
& Related Ionospheric
Phenomena, Stanford University
Press, Stanford, CA (1965).
Google whistlers & sferics
there are many websites where you
can hear recordings of them!

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 8.5 Prof. Steven Errede

Finally, we consider the complex index of refraction n n i or equivalently,

the complex wave number, k k i of pure water (H2O): k n
c
The top graph in the figure below shows n f vs. f , the bottom graph shows the absorption

coefficient, 2 2 vs. f , and E eV . Note that both plots are log-log plots!!!
c

1eV 1.6 1019 J

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 8.5 Prof. Steven Errede

Note the following aspects of the above plots for pure H2O:

At low frequencies n f 9 n 2 f f o K e f 81!!! arises from partial

orientation of the permanent electric dipole moment p of the H2O molecule (Langevin

equation) the partial orientation of pH 2O is due to finite-temperature thermal energy
density fluctuations.

The n f vs. f curve falls smoothly through the infrared region some glitches in
n f and f due to molecular vibrational excitations/resonances in infrared region!!

more resonances in the UV region due to excitations in the oxygen atom

The absorption coefficient is very small at low frequencies, but starts to rise steeply at
f 108 Hz . At f 1012 Hz ~ far infrared , ~ 104 m 1 sc 100 m in H2O!!!

In the microwave region, strong absorption by H2O can use for microwave ovens!!!
Strong absorption by H2O limited the trend of RADAR {During WWII} of going to
shorter and shorter wavelengths, to achieve better spatial resolution . . .
In the infrared region, the absorption coefficient for H2O is very large, due to vibrational
resonances of the H2O molecule, 104 m 1 .

In the visible light region, there are no resonances of the H2O molecule, so the absorption
coefficient drops by ~ 7-8 orders of magnitude {!!!} Thus in the visible light region
H2O/water is transparent/invisible.
However, getting into the UV region, oxygen atom resonances (due to inner L, K-shell
electrons), thus rises again dramatically, even higher, 106 m 1 in the UV region.

an absorption window in the visible light region: 4 8 1014 Hz - not very wide!!!

red light
blue/violet light
R 750nm
BV 375nm
The H2O absorption window is of fundamental importance to the evolution of life on earth
Life started off in the water/ocean, aquatic critter vision/sight developed in that
environment and specifically in the H2O absorption window, where significant amounts
of EM energy are present {thanks to the sun!} to be of use/benefit for survival
The co-incidence of the H2O absorption window and our (and other creatures) ability
today to see in the visible light region of the EM spectrum is not a mere coincidence!
Green grass/plants at the center of visible light absorption window! Because green =
reflected light, plants have absorption in both the red and blue/violet regions.
On either side of the H2O absorption window there is not much/very little infrared or UV
radiation in water after ~ few scIR ~ 100 m scUV ~ 1 m - because strongly attenuated !!!

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

2005-2015. All Rights Reserved.