Laserby Bblaud - 23

182 LASERS AND NON-LINEAR OPTICS
The DFB lasers are significant advances in the field of integrated optics.
They provide a means of better restricting the laser operation to a single mode.
11.9 GAMMA-RAY LASER

It is generally believed that levels of nuclear excitation that might be efficiently
stimulated in a gamma ray laser are very difficult to pump, because absorption
widths in nuclei are too narrow to permit effective pumping with
X-rays. Similar fears were expressed in atomic physics before Maiman’s
discovery of the ruby laser. But exploration of a broad absorption band linked
through efficient kinetics to the narrow laser level, led to the development of
the ruby laser. Collins et al. [82] proposed an analogous effect at the nuclear
level for pumping a gamma ray laser.
Samples of 79Br and 77Se, both in natural isotopic abundance, were excited
by them with 20 ns pulses of bremsstrahlung radiation produced by an electron
beam machine. The spectral energy density developed at the tungsten converter
was of the order of 1013 keV per pulse. The samples were positioned in a
pneumatic shuttle tube directly behind the target foil and were automatically
transferred to the counting chamber after each shot. The nuclear fluorescence
spectra observed showed that a gamma ray laser is definitely feasible if an
appropriate isotope exists.
Copyright © 2010. New Age International. All rights reserved.
Laud, B.B.. Lasers and Non-Linear Optics, New Age International, 2010. ProQuest Ebook Central, http://ebookcentral.proquest.com/lib/uic/detail.action?docID=3017441.
Created from uic on 2018-07-29 20:19:58.
HOLOGRAPHY 12
The most widespread method of obtaining optical images is the photographic
method. The image of a three dimensional object is recorded on a two dimensional
photographic plate or film. Absorption of light in the light-sensitive layer results
into chemical reactions and in formation of a latent image. This is transformed
into a visible one by process of developing. When we examine a photograph
from various directions, we do not get new angles of approach and we cannot
see what is happening on the other side of the object. This is because, in the
conventional photography, only the distribution of the square of the amplitude
is recorded in a two dimensional projection of the object on the plane of the
photograph.
A fundamentally new method of recording optical images is now available.
This is known as holography. The word “holography” originates from the
Greek word “holos” meaning the whole. Holography means “complete recording”.
In this technique a light wave is a carrier of information and it is recorded in
terms of wave parameters: amplitude and phase components. The technique
was first proposed by Gabor [133] in 1949. Note that Gabor did not have a
laser when he formulated the idea of holography. It was not extant then. He
ran his first experiments with mercury arc lamp as the source of light. The
technique became a practical proposition only after the advent of lasers and
opened the way to major advances in holography.

The process does not record the image of the object but rather records
the reflected light wave. The record is called a “hologram”. It does not bear
any resemblance to the original object but possesses a spectacular property. It
memorizes, so to say, an encoded image of the object and contains much more
information than ordinary photograph can record.
Laser holograms were prepared for the first time by Leith and Upatnicks
[237, 238]. Denisyuk [96, 97], using thick layers of photosensitive material,
recorded three dimensional holograms.
12.1 PRINCIPLE OF HOLOGRAPHY

The basic principle of holography can be explained in two steps:
(i) recording of the hologram, and (ii) reconstructing the image.
12.1.1 Recording of the Hologram

Holography is in principle an interference based technique and, hence, light
waves with a high degree of coherence are required for its realization. The
principle is illustrated in Figs. 12.1 and 12.2.
V irtu a l im a ge R ea l im a ge
P o′ P
B
ve
t wa
ou
ad
A Re
S
Fig. 12.1 Recording of Fig. 12.2 Reconstruction

a hologram of the image
A laser beam is divided by a beam splitter S into two beams A and B. The
transmitted beam B illuminates the object whose hologram is to be recorded
and a part of the light scattered by the object impinges on a photographic plate.
The reflected beam A, which is called the reference beam, also falls onto the
photographic plate. The superposition of these two beams produces an
interference pattern which is recorded on the plate. The pattern is very fine,
the spacing between the fringes being as small as 0.001 mm. The developed
plate is known as the hologram. The hologram is quite unintelligible and gives
no hint of the image embedded in it. It contains, however, enough information
to provide a complete reconstruction of the object. The fine structure of
interference fringes requires photographic emulsion with a high spatial resolution.
12.1.2 Reconstruction of the Image

If now the object is removed and the hologram is put in the place where it was
when formed, the laser beam which is now known as the readout wave,
interacts with the interference pattern on the plate and two images are produced
by the diffracted waves (Fig. 12.2). One of them appears at the original
position occupied by the object (virtual image) and the other (real image) which
can be photographed directly without using a lens.
HOLOGRAPHY 185
The hologram contains in the form of its interference pattern information

in all the geometrical characteristics of the object. The virtual image, which is
seen, by looking through the hologram as if it were a window, appears in
complete three dimensional form. If one moves his eye from the viewing
position, the perspective of the picture changes and it is possible to see the
other sides of the object. The real image has all the aforementioned properties
and is found between the observer and the plate as shown in Fig. 12.2. The
real image will appear inverted in depth, i.e., the relief of the object is reversed,
with the features of the object farther from the viewer appearing closer; but
there will be no lateral inversion. This image is called pseudoscopic image.
Since, how-ever, the real image reverses foreground and background, the
interest of the observer lies in the virtual image.
Although light is commonly used for holography, holograms have been
successfully recorded with acoustic radiation, electron beams, X-rays and
microwaves.
12.2 THEORY
To understand the various characteristics of holograms it is necessary to discuss
the theory on which holography is based.
Let us first consider the case of a small object P. The source illuminates
the object, but most of the light falls undisturbed on a photographic plate
(Fig. 12.3). The light scattered or diffracted by the object also falls on the plate
where it interferes with the direct beam—the reference beam. To find the
intensity at a point O on the plate, we may write the field arriving at O as
E = Er + E0 ...(12.1)
Er
E0
P
Z0 Z
Fig. 12.3 Hologram of a point object

where Er is the field due to the reference beam and E0 is the field scattered
from the object. The scattered field E0 is not simple, both amplitude and phase
vary greatly with position. The reflected wave-fronts are spherical and concentric
around the point of origin. We represent the field of such a wave-front by
A0
E0 = exp (i(kr0 – ω t )) ...(12.2)
r0
and the field Er by the plane wave

Er = Ar exp (i(kz0 – ωt)) ...(12.3)
where r0 = PO and z0 is the distance from P to the plate.
The intensity at O is
2
I = Er + E0
2
2 A0 A0 Ar*
= Ar + + exp (ik (r0 – z0 ))
r02 r0
A0* Ar
+ exp (ik (z0 – r0 )) ...(12.4)
r0
Choosing the constants K and φ suitably, we can combine the last two
terms from the above relation and write it as
2
cos [k (r0 – z0 ) + φ]
2 A0
I = Ar + +K ...(12.5)
r02
r0
Because of the cosine term, the total intensity I as a function of r0 shows
a series of maxima and minima. The interference of the plane wave Er with the
spherical wave E0, thus produces a set of circular interference fringes on the
plate which, when developed, forms the hologram. If we assume that the plate
response is proportional to the intensity I, the power transmission of the plate,
T 2, is given by
T 2 = 1 – αI ...(12.6)
1
or T 1– αI ...(12.7)
2
where α is a constant.
Let us now see what happens when this hologram is illuminated by the
reference beam. The field of the transmitted wave is
 α 
E = TEr = 1 – I  Ar exp (i(kz0 – ωt)
 2 
 2 
α 2 α A0
= 1 – A – · 2  Ar exp (i(kz0 – ωt ))
 2 r 2 r0 
α A0 Ar*
– · exp (ik (r0 – z0 ))Ar exp(i( z0 – ωt ))
2 r0
α A0* Ar
– · exp (ik (z0 – r0 ))Ar exp(i ( z0 – ωt ))
2 r0
 α 2 α A
2 
= 1 – A – · 02  Ar exp (i (kz0 – ωt ))
 2 r 2 r0 
HOLOGRAPHY 187
2
αA0 Ar
– exp (i(kr0 – ωt ))
2r0
α A0* Ar2
· exp (i (2kz0 – kr0 – ωt ))
– ...(12.8)
2 r0
where we have used (12.4).
The first term of (12.8) represents the same plane wave beyond the plate
as the one incident on it, except for the attenuation corresponding to the
average blackening of the plate (see 12.3). The second term represents a
diverging spherical wave surface identical with the wave surface emitted by the
object except for a constant factor (see 12.2). This wave surface when projected
back seems to emanate from an apparent object located at the place where the
original object was located (Fig. 12.4). This is the virtual image of the object.
The third term represents also a spherical wave surface which is a replica of
the original wave but has a conjugate or reverse curvature. It converges at a
point P′ producing a real image at this point which can be photographed
without a lens. The hologram thus produces a virtual image P and a real image P′.
The general theory of holography is too cumbersome to pursue further.
We can, however, generalize the treatment given above for a point object to an
object of finite size. As before the intensity at the point O is
P P′ Z
Z0 Z0
Fig. 12.4 Reconstruction of the image

2 2 2
I = Er + E0 = Er + E0 + E0 E0* + E0* Er ...(12.9)
When the hologram is illuminated by the reference beam, the field on the
other side of the plate is
E = TEr = 1–
α
2
2
Er –
α
2
2
E0 Er { }
α 2 α
– E0 Er – E0* Er2 ...(12.10)
2 2
As before, the first term gives the attenuated reference wave; the second
term produces the virtual image and the third term produces a real image of
the object at a position which is the mirror image of the virtual image, with
respect to the plate.
Equation (12.10) shows how holography allows us to make a complete

record of the wave coming from the object. In the absence of the reference
beam the blackening of the plate would be proportional to | E0 |2 i.e., only the
modulus of E0 would be recorded. This means that only the amplitude
information would be recorded and the phase information is thereby lost.
Because of the presence of the reference wave, the field recorded on the plate
is proportional to E0 as can be seen from the second term of (12.10), i.e., both
amplitude and the phase are recorded thus making the complete reconstruction
of the object possible.
Gas lasers operating in the continuous wave (cw) mode, are often used
for holography because their coherence is high. However, their emitting power
being low, time of exposure has to be large, and, hence, moving objects cannot
be holographed using gas lasers. Pulsed solid state lasers make it possible to
cut down the exposure to about 10–3 sec and, hence, can be used to holograph
moving objects and to record the development of a process in time.
12.3 SOME DISTINGUISHING

CHARACTERISTICS OF HOLOGRAPHS
One may prefer, at first glance, a photograph to a hologram because a photograph
shows everything ‘clearly’, while a hologram is quite unintelligible. However,
holographs have some interesting properties which make it preferable to a
photograph.
(i) The virtual image produced by a hologram appears in complete three
dimensional form. The observer sees something looking very much
like the object did in the process of hologram recording. If he tilts
his head, he notices other objects behind the one in the foreground
or new details that were not noticeable before. The image manifests
vivid realism.
(ii) Destruction of a portion of a photographic image results in an
irreparable loss of information corresponding to a part of the object.
In the case of a hologram the information about a point object is
recorded over the whole area of the hologram as can be seen from
Fig. 12.3, and each such point of a real object is recorded on the
whole hologram and not in one of its points as is the case in a
photograph. Each point on the hologram receives light from all parts
of the object and, therefore, contains information about the geometrical
characteristics of the entire image. Each part of the hologram, no
matter how small, can reproduce the entire image. Consequently,
destruction of a part of a hologram does not erase a specific portion
of the image. It may be noted, however, that the reduction in size
worsens the resolution which is a function of the aperture of the
imaging system. A hologram, thus, is a reliable method of data storage.
HOLOGRAPHY 189
(iii) Another characteristic which proves the supremacy of a hologram

over a photograph is its information capacity. Superposition of several
images on a photographic plate is pointless. A hologram, on the other
hand, may contain a number of consecutively recorded scenes that
can be recorded independently, Different scenes can be recorded on
a hologram, each time changing the angle at which the reference
wave is incident on the holographic plate, which can be done by
rotating the plate. Reconstruction of a specific scene only requires
that the hologram be properly oriented with respect to the read out
wave. It has been estimated that a single hologram with an area of
about 100 square centimeters, may contain at least one volume of
Encyclopaedia Britannica. This indicates the extremely high information
capacity of holograms.
(iv) Yet another curious property of the wave-front reconstruction process
is that it does not produce negatives. One may consider a hologram
itself as a negative. But the image it produces is a positive. A hologram
copied from another by contact printing would, no doubt, be reversed
in the sense that opaque areas now become transparent and vice-
versa; but the image reconstructed from it, would be identical in all
respects to that produced by the original. This is because the
information is recorded on the plate in the form of a modulated spatial
carrier. It may be recalled that the information on the grating carrier
is embodied in the fringe contrast and in the fringe spacing; neither
of this is altered by the reversal of polarity.
12.4 PRACTICAL APPLICATIONS OF

HOLOGRAPHY
Holography has a wide range of applications, particularly in science and
technology. Specialists in the field of data processing devote much of their
attention to holography. The links between computer science and holography
are now well established and are developing fruitfully. An application, remarkable
for its simplicity, is found in an experiment of Brian Thompson and his colleagues
at Technical Operation, Inc. These workers were faced with the problem of
measuring the distribution by size and other properties of floating fog like
particles in a sample volume. Such particles generally do not remain stationary
long enough for the observer to focus on them. It is also desirable to photograph
all the particles in the volume at a given time. The wave-front reconstruction
method offers an ideal solution to this problem. A hologram is made by
illuminating the volume by a short-pulse laser which “freezes” the motion of the
particles. In the reconstruction an image of the entire volume is produced, and
the particle size, distribution and cross-sectional geometry can be measured by
microscopic examination.
It is not likely that holography will replace photographic methods in everyday

situations; its maximum effect is in the scientific and technical domains.
Nevertheless, we may expect holography to come into our daily life in the not
very distant future. It has been experimentally demonstrated that holographic
cinema is a feasible proposition and holographic television is also likely to make
its appearance in the near future. Holography can be useful in practically every
area of human activity, ranging from linguistics to the investigation of processes
occurring in thermonuclear plasma.
12.5 ADVANCES IN HOLOGRAPHY

A major drawback of conventional holographic process is the requirement of
coherent illumination in the image reconstruction. The rainbow (white light
transmission) holography devised by Benton [40] is a major improvement in
holography. Rainbow holograms can be viewed with a white light source and
the image reconstructed is exceptionally bright. However, because of the complex
nature of the two-step process involved, its scientific applications are to some
extent limited. A one-step process demonstrated by Chen and Yu [73] greatly
simplifies the procedure.
An entirely new method of storing confidential data, based on sandwich
holography, has been suggested by Abramson et al. [3].
For details of all these methods, one has to refer to the original papers.
NON-LINEAR
OPTICS
13
Can the optical properties of a medium depend upon the intensity of radiation?
In the pre-laser era, the answer to this question, perhaps, would have been an
emphatic “no”. The basis for this conclusion is that the field strengths of the
conventional light sources used before the advent of lasers, were much smaller
than the field strengths of atomic and inter-atomic fields. The latter are of the
order of 107 to 1010 V/cm; whereas the former would not exceed 103 V/cm.
It is natural that the light wave with such a low intensity is not able to affect
atomic fields to the extent of changing optical parameters. The high degree of
coherence of the laser radiation has made it possible to have extremely high
spatial concentration of light power. It is now possible to generate 1 MW
pulses, lasting a few tenths of nanoseconds, using moderately powerful lasers.
The energy current density in a beam of cross-section 1 mm2 of such a laser
is JE ≈106 MW/m2, which corresponds to peak electric field strength E ≈ 3
× 107 V/m. Due to coherence, the beam can be focussed to an area A ≈ λ2.
If the laser wavelength is assumed to be 1 µ, then A = 10–12 m2 and, hence
JE ≈ 1018 MW/m2 giving E ≈ 3 × 1010 V/m, which is within the range of
atomic fields. At such high fields, the relationship between the electric polarization
P and the field strength E ceases to be linear and some interesting nonlinear
effects come to the fore. Bloembergen played an important role in establishing
the theoretical framework of nonlinear optics. We deal in this chapter some

nonlinear effects describable by semiclassical theory in the electric dipole
approximation.
13.1 HARMONIC GENERATION

Non-linear properties in optical region have been strikingly demonstrated by the
harmonic generation of light observed for the first time by Franken and
coworkers in 1961 [126]. They observed ultraviolet light at twice the frequency
of a ruby laser light (λ = 6493 Å), when the light was made to traverse a quartz
crystal. This experiment attracted widespread attention and marked the beginning
of the experimental and theoretical investigation of nonlinear optical properties.
A simplest scheme for this experiment is shown in Fig. 13.1.
Q ua rtz
R ub y
c rys ta l sla b P ho to c e ll
la se r
ω1
ω2
ω1 ω2 = 2 ω1
F ilte r Q ua rtz UV
fo r ru by la se r c rys ta l Tra n sm iss io n
s lab filte r
Fig. 13.1 Second harmonic generations

A ruby laser beam (λ = 6493 Å) with average power of the order of
10 kW is focussed on a quartz slab. The transmitted light then was passed
through a filter which cuts off the red light and allows uv light to pass through.
The emerging light was incident on a photocell. Radiation with wavelength
λ = 3471 Å and the power of the order of 1 mW was observed in the
transmitted light. How can one account for this change in frequency?
A dielectric medium when placed in an electric field is polarized, if the
medium does not have a transition at the frequency of the field. Each constituent
molecule acts as a dipole, with a dipole moment Pi. The dipole moment vector
per unit volume P is given by
P = ∑ Pi ...(13.1)
i
where the summation is over the dipoles in the unit volume. The orienting
effect of the external field on the molecular dipoles depends both on the
properties of the medium and on the field strength. Thus, we can write [233]
P = ε0χE ...(13.2)
where χ is called the polarizability or dielectric susceptibility of the medium.
This relation is valid for the field strengths of conventional sources. The
quantity χ is a constant only in the sense of being independent of E; its
magnitude is a function of the frequency. With sufficiently intense laser radiation

the relation (13.2) does not hold good and has to be generalized to
P = ε0(χ(1)E + χ(2) E2 + χ(3) E3 +...) ...(13.3)
where χ(1) is the same as χ in (13.2); the coefficients χ(2), χ(3),... define the
degree of non-linearity and are known as nonlinear susceptibilities. If the field
is low, as it is in the case of ordinary light sources, only the first term of (13.3)
can be retained. It is for this reason that the prelaser optics is known as linear
optics. Higher the value of the electric field, more significant become the higher
NON-LINEAR OPTICS 193
order terms. It may be noted that optical characteristics of a medium, such as

dielectric permittivity, refractive index, etc., which depend upon susceptibility,
also become functions of the field strength E, if it is sufficiently high. The
medium of which the polarization is described by a nonlinear relation of the
type (13.3) is called a “non-linear medium.”
Suppose now that the field incident on a medium has the form
E = E0 cos ωt. ...(13.4)
Substituting this in (13.3), we have
P = ε 0 χ(1) E0 cos ωt + ε 0 χ(2) E02 cos 2 ωt + ε 0 χ(3) E03 cos3 ωt + ... ...(13.5)
Using the trigonometric relations
1 + cos 2 θ cos 3 θ + 3 cos θ
cos2 θ = ; cos3 θ = ...(13.6)
2 4
we can transform (13.5) to the form
1  3 
P = ε 0 χ(2) E02 + ε 0  χ (1) + χ (3) E02  E0 cos ωt
2  4 
1 1
ε 0 χ(2) E02 cos 2ωt + ε 0 χ(3) E03 cos 3 ωt + ...
+ ...(13.7)
2 4
The first term is a constant term. It gives rise to a dc field across the
medium, the effect of which is of comparatively little practical importance. The
second follows the external polarization and is called the first or fundamental
harmonic of polarization; the third oscillates at frequency 2ω and is called the
second harmonic of polarization, the fourth is called the third harmonic of
polarization, and so on.
13.2 SECOND HARMONIC GENERATION

A polarization oscillating at frequency 2ω radiates an electromagnetic wave of
the same frequency, which propagates with the same velocity as that of the
incident wave. The wave, thus, produced, has the same characteristics of
directionality and monochromacity as the incident wave and is emitted in the
same direction. This phenomenon is known as the Second Harmonic Generation

(SHG).
In most crystalline materials, the nonlinear polarizability χ(2) depends on
the direction of propagation, polarization of the electric field and the orientation
of the optic axis of the crystal. Since in such crystalline materials the vectors
P and E are not necessarily parallel the coefficients χ must be treated as
tensors. The second order polarization, therefore, may be represented by the
relation of the type
Pi(2) = ε 0 ∑ χijk E j Ek
(2)
...(13.8)
j ,k
where i, j, k represent the coordinates x, y, z,. Most of the coefficients χijk,

however, are usually zero and we have to deal only with one or two components.
It must be mentioned here that the second harmonic generation represented
by (13.8) occurs only in certain type of crystals. Consider, for example, a
crystal that is isotropic. In this case χijk is independent of direction and, hence,
is a constant. If we now reverse, the direction of the axis (x → – x, y →
– y, z → – z) leaving electric field and dipole moment unchanged in direction,
the sign of these two must change.
– Pi (2) = ε 0 ∑ χijk (–E j ) (–Ek ) = + Pi

(2) (2)
∴ ...(13.9)
j ,k
which means Pi(2) = 0 and, hence, χ (2)

ijk
= 0.
Second harmonic generation, therefore, cannot occur in an isotropic medium

such as liquids or gases nor in centro-symmetric crystals (i.e., crystals
symmetrical about a point). Only crystals that lack inversion symmetry exhibit
SHG.
In the case of non-centro-symmetric materials (e.g., anisotropic crystals,
such as uniaxial crystals) both the quadratic and cubic terms are present.
However, generally, the cubic term is sustantially smaller than the second order
term and may be ignored.
For such materials, we can write
P = ε 0 χ(1) E + ε 0 χ(2) E 2 ...(13.10)
and the medium is said to have second order linearity.
13.3 PHASE MATCHING

The intense development of research on the mechanism of generation of optical
harmonics in crystals and media in which such generation is effectively realizable,
has indicated the importance of phase relation between the fundamental and
generated harmonics, as they

propagate in crystals having
optical dispersion [Bloembergen ω1 , k 1 ω1 , k 1
51, Franken and Ward 127]. It 2 ω1 , k 2
was observed that the efficiency
of the generation of harmonics dz
L
depends not only on the intensity
of the exciting radiation, but also
Fig. 13.2 Second harmonic wave propagating
on its direction of propagation
through a material of length L
in crystals.
Suppose a plane wave at frequency ω and the second harmonic wave at

frequency 2ω driven by it are propagating in the z-direction through a material
of length L (Fig. 13.2). Let us find an expression for the intensity of SHG at
the exit surface of the material.
The amount of second harmonic radiation produced within a slab of width
dz located at z will be proportional to the width and to the second harmonic
2
dipole moment per unit volume induced at frequency 2ω, i.e., P( z ) which, in
turn, is proportional to the square of the electric field E, that is
(2) ( )
dE( z ) ∝ P(z2) dz ∝ exp (2i (k1 z – ωt ))dz ( ∵ E ∝ exp (i (k1 z – ωt ))
...(13.11)
where dE22 is the amount of second harmonic radiation within the slab.
We see that the spatial variation of the second harmonic polarization is
characterized by a wave member 2k1.
The second harmonic radiation produced by this slab at the exit surface
of the crystal, i.e., at z = L obviously will be
dE((2) (2)
L) ∝ dE(z ) exp (ik2 (L – z ))dz ...(13.12)
where L – z is the distance from the slab to the end of the crystal and k2 is
the propagation wave number of the second harmonic radiation. In general,
k2 ≠ k1 because of dispersion. (Note k = 2πη/λ.)
(2)
∴ dE(L) ∝ exp (2i(k1z – ωt)) exp (ik2 (L – z))dz
= exp (i(2k1 – k2))z exp (i(k2L – 2ωt))dz ...(13.13)
We have assumed here that the incident power is nearly unchanged as the
beam propagates through the crystal.
Intergrating (13.13)
L
E((2)
L) ∝ Ú exp (i(2k1 – k2 )z) exp (i(k2 L – 2wt ))dz
0
exp(i( k2 L – 2ωt ) [exp(i(2k1 – k2 )L)–1]

= ...(13.14)
i(2k1 – k 2 )
On simplification and some manipulation, we get
Ê 2k – k 2 ˆ
sin Á 1 ˜L
(2) Ë 2 ¯
E(L) ∝ ...(13.15)
Ê 2k1 – k2 ˆ
ÁË ˜
2 ¯
where we have taken only the real part of the proportionality factor.
(2k1 – k2 )L π
This will be maximum when = , that is, the field of the
2 2
second harmonic generation will be maximum when
π λ
L= = ...(13.16)
2k1 – k2 4(ηω – η2ω )
where ηω, η2ω are the refractive indices at ω and 2ω respectively.
Increasing L beyond this value will not result into any increase in E(2). The
magnitude of L given by (13.16) is called the coherence length for the second
harmonic radiation.
The expression for intensity, viz.
 2k – k 2 
sin 2  1 L
I ∝
 2  ...(13.17)
2
 2k1 – k2 
 
 2 
is sharply peaked about
 2k1 – k2 
 L = 0
2 
that is when k 2 = 2k 1 ...(13.18)
For efficient frequency doubling, this relation must be satisfied. This
requirement is known as phase-matching criterion.
Since
2ωη2ω ωηω
k2 = and k1 = ....(13.19)
c c
relation (13.18) reduces to
η2ω = ηω. ...(13.20)
Thus, the phase matching criterion becomes a refractive-index criterion.
It is rather difficult to fulfill this requirement because most materials show
some sort of dispersion in the refractive index.
A satisfactory solution to this problem
would be to use the dependence of the
refractive index on the direction in
anisotropic crystals. A birefringent A′ 2ω A
material has different refractive indices
for different polarization of light. This ω

occurs generally in crystals of low
symmetry. A light wave entering an
anisotropic crystal splits into two waves
travelling at different velocities. In
uniaxial crystals the ray corresponding
A A
to the wave whose refractive index is
independent of the direction of
propagation is called the ordinary ray. Fig. 13.3 Indicatrix for a negative
uniaxial crystal
The ray corresponding to the other light wave whose refractive index depends
upon the direction of propagation is called extraordinary ray. The behaviour of
the refractive index is usually described in terms of refractive index surface,
i.e., the indicatrix or index ellipsoid. In the case of the ordinary ray it is a
sphere; for extraordinary ray it is an ellipsoid. We, therefore, have to choose
a material in which the refractive index for the extraordinary ray at 2ω, is equal
to that of the ordinary ray at ω. This points to the fact that effective frequency
conversion in the second harmonic is possible only in a limited number of
crystals.
Consider a negative uniaxial crystal, i.e., a crystal for which the refractive
index for the ordinary ray is greater than that for the extraordinary ray.
Fig. 13.3 shows a section through the refractive index surfaces (indicatrix) for
one such crystal. The dotted curve represents the surface corresponding to
the frequency 2ω and the solid curve for frequency ω. OX is the optic axis of
the crystal. The refractive index surface of the ordinary wave and that for the
extraordinary wave intersect at A. This means, that for the waves propagating
in the direction OA
η0(ω) = ηe(2ω) ...(13.21)
That is, the incident and the second harmonic waves propagating in this
direction are phase matched. The cone angle θ is the phase matching angle. The
phase matching condition is satisfied for all directions lying on the cone.
As stated in Sec. 13.1, SHG was first realized successfully in quartz. It
was subsequently generated in many other crystals such as: potassium
dihydrophosphate (KDP) [Hagen et al., 157], ammonium dihydrophosphate
(ADP), [Dowling 103], barium titanate [Geusie et al., 143], lithium iodate
[Chesler et al., 74], etc., A SH peak power of 200 kW has been obtained with
only 6 mJ energy in a single pulse [Terhune et al., 375]. Conversion efficiencies
of 15 – 20% have been found at input power densities of the order of 100 MW
cm –2 .
SHG was also successfully realized in gases [Adams et al., 4, Ashkin
et al., 16, White et al., 398] and in semiconductors [Armstrong et al., 14
Garfinkel et al., 135]. It would be interesting to note that SHG has been
observed in calcite as a function of dc electric field, although the crystal
possesses a centre of inversion. This is because the imposed electric field
removes the symmetry [Terhune et al., 374].
The importance of second harmonic generation lies in the fact that it is
one of the principal methods of effective conversion of infrared radiation into
visible and visible into ultraviolet. The mechanism of the second harmonic
generation has been thoroughly investigated theoretically [13, 52, 211, 212,
218, 236, 253].
13.4 THIRD HARMONIC GENERATION

As stated in Sec. 13.2, in the case of centrosymmetric materials, the expression
(13.3) will lack terms in even powers of E and it will reduce to
P = ε0χ(1)E + ε0χ(3)E3 +... ...(13.22)
Or in vector notation
P = ε0χ(1)E + ε0χ(3)E2E + ... ...(13.23)
Third harmonic generation (THG) is, therefore, possible in crystals that
exhibit inversion symmetry. The development of Q-switched lasers had made
it possible to generate third harmonic in crystals [32, 267]. However, the
energy conversion efficiency in such cases is very low. For example, in calcite
the maximum energy conversion efficiency in the third harmonic was 0.01%.
Experiments for observation of the third harmonic were also performed by
Maker and Terhune [268] using giant pulse lasers. Zwernemann and Beeker
[415] have observed experimentally the enhancement of third harmonic generation
(THG) at 9.33 µm in CO by having the interaction take place in a waveguide.
They have presented a theoretical determination of the most suitable waveguide
in which the interaction can take place.
The process of generation of higher order harmonics can be explained on
the same lines.
13.5 OPTICAL MIXING

In the equation for nonlinear polarization (13.3), we have assumed that the
factor E2 in the second term, is the product of the electric field strength with
itself, E.E. However, such a nonpolarization term may also result from the
interaction of two fields with different frequencies. Suppose two coherent light
wave trains of unequal frequencies ω1 and ω2 are traversing the material. The
effective field in the material is
E = E1 cos ω1t + E2 cos ω2t ...(13.24)
Substituting this in (13.3), the second term becomes
P (2) = ε0χ(2) (E1 cos ω1t + E2 cos ω2t)2
= ε0 χ(2) (E12 cos 2 ω1t + E22 cos 2 ω2 t )
+ 2ε 0 χ(2) E1E2 cos ω1t cos ω 2t ...(13.25)

Using the trigonometric relation
2cos α cos β = cos (α + β) + cos (α – β)
We can express the last term as
2ε0χ(2)E1E2 cos ω1t cos ω2t
= ε0χ(2)E1E2[cos (ω1 + ω2)t + cos (ω1 – ω2)t] ...(13.26)
This shows that the non-linear polarization and, therefore, emitted radiation
contains frequencies ω1 + ω2 and ω1 – ω2. The energy conversion between the
beams can take place over significant distances only if the beams travel in the
same direction and at the same velocity.
The sum and difference frequencies can be observed experimentally.
Generation of difference optical frequencies was first observed by mixing a
beam from a ruby laser with an incoherent beam of a mercury lamp
(λ = 3115 Å). The efficiency with which the difference frequency was generated
was negligible; with the power in the mercury beam amounting to about 2 × 10–
4 W, the power radiated at the difference frequency was 10 –10 W. Optical
mixing of the emission, of two ruby lasers with different frequencies was first
observed by Franken and coworkers [127].
The first term of (13.25), besides frequency doubling, also leads to a dc
term (Sec. 13.7). Bass et al., [33] observed a dc pulse of about 200 µV when
1 MW radiation was passed through a KDP crystal. Sum of the frequencies
from two ruby lasers held at different temperatures was observed by Bass
et al., [32] and from a ruby and a neodymium laser by Miller and Savage [275].
We have considered above only the second order term in (13.3). In a more
general case, polarization expression may include terms with E3, E4, etc.,
Substitution of (13.24) in (13.3) with higher order terms, results into an expression
containing terms with frequencies ω = m ωl ± nω2, where m and n are integers.
This shows that in addition to the sum and difference of frequencies, other type
of frequency mixing is possible.
The mixing of optical frequencies in crystals, thus, uncovers additional
possibilities for optical frequency conversion. It has provided a source of
narrow band coherent radiation in various regions including those in which
there are no primary lasers available. Primary laser radiation is available from
the IR through the visible and uv down to ~ 116 nm. Optical frequency mixing
has enabled us to extend the range of wavelength to the XUV reaching almost
to the soft X-ray range. Currently it is the only source of coherent radiation
in XUV.
Incidentally, it may be mentioned here that ω1 – ω2 may fall into the range
of acoustic frequencies. In a sense, therefore, frequency mixing is an optical
method of generating ultrasonic waves.

As in the case of second harmonic generation, phase-matching condition
is also important in frequency mixing. In fact, it is more stringent in the latter
case, because of the number of frequencies involved. In the case of second
harmonic generation, it is necessary to find a direction in crystals such that
k1 = k2. In the case of sum or difference frequencies, three waves must be
matched. If
ω3 = ω1 ± ω2, ...(13.27)
the condition to be satisfied is
k 3 = k 1 ± k2 ...(13.28)
13.6 PARAMETRIC GENERATION OF LIGHT

Parametric phenomena in electronics are widely known. They occur in circuits
involving nonlinear capacitors. Similar processes occur in optics when non-
linear crystals are used as parametric media. The process is known as parametric
generation of light, and is based on “optical mixing” discussed in the preceding
section.
Suppose a powerful signal at frequency ωp (pump frequency) is applied
to a parametric medium and a small signal at frequency ωs (signal frequency)
is introduced at one end. The fields at the original frequencies are regarded as
fixed parameters. “Mixing” of the signal and the pump frequency may result
into a secondary wave at frequency ωi given by
ωi = ωp – ωs ...(13.29)
which is known as ‘idler’ frequency. The corresponding field strength being
proportional to
EpE s = Ei ...(13.30)
as shown by equation (13.26).
In view of the non-linear properties of the medium, further mixing may
occur. In particular, the field generated by the polarization component oscillating
at the idler frequency and the original pump field, when mixed, would make a
contribution to the signal field. Thus
ωp – ωi = ωp – (ωp – ωs) = ωs. ...(13.31)
The strength of the contribution is proportional to
E p E i = E 2p Es ...(13.32)
where we have used (13.30). This shows that the strength is proportional to
Es, in accordance with the usual requirement for parametric amplification.
Thus, the secondary light waves at frequencies ωs and ωi can be excited
parametrically at the expense of the part of energy of the pumping wave. Initial
signals required for triggering the process of parametric generation are always
available in any crystal in the form of spontaneous photons.
13.7 SELF-FOCUSSING OF LIGHT

The refractive index of a material is related to the susceptibility by the relation.
η= 1+χ ...(13.33)
Since the susceptibility χ is a function of the field E, η depends on E. This
dependence of the refractive index on the field strength gives rise to a non-
linear effect: self-focussing of intense light beams.
Self-focussing does not alter the frequency of the light waves. We need,
therefore, consider only the second term in the relation (13.7) which describes
the fundamental harmonic, viz.
3 (2) 2
P (1) = ε 0 (χ(1) + χ E0 ) E. ...(13.34)
4
The expression for the refractive index consequently is
3 (2) 2
η = 1 + (χ(1) + χ E0 ) ...(13.35)
4
We write this as
η = εl + ε nl ...(13.36)
where we have put
εl = 1 + χ(1) ...(13.37)
which gives the dielectric permittivity of the linear medium, and
3 (2) 2
εnl = χ E0 ...(13.38)
4
is a non-linear increment in the expression for dielectric permittivity.
ε nl Ê e ˆ
η = εl 1 + ≈ e l Á1 + nl ˜ ...(13.39)
εl Ë 2e l ¯
(∵ ε nl << εl )
Ê e nl ˆ Ê 3c(2) E02 ˆ
= hl Á1 + = h Á1 + ˜
2hl2 ˜¯
l
Ë Ë 8hl2 ¯
or η = ηl (1 + ηnl E02 ) ...(13.40)
where ηl = εl is the refractive index of the linear medium, and
ε nl 3χ(2) E02
ηl ηnl E02 = = ...(13.41)
2 √εl 8 1 + χ(1)
is the non-linear increment in the expression for the refractive index. We, thus,
see that the refractive index of a non-linear medium is proportional to the
square of the amplitude of the field, that is, to the intensity. Now the intensity
of a laser beam is not constant over its cross-section. It peaks at the axis of
the beam and falls off gradually away from the axis. The velocity of the light
wave is given by v = c/η. Since η decreases owing to the falling of the intensity
of the light beam, the velocity increases with the distance away from the axis.
Consequently, a plane wave-front incident on material becomes concave as it
propagates through the medium and contracts towards the axis (Fig. 13.4). In
other words, it self-focusses, after which it propagates as a narrow light fibre.
Fig. 13.4 Self-focussing

The distance L0 over which the beam self-focusses can be approximately
estimated using the formula [371],
D
L0 = ...(13.42)
ηnl E02
where D is the diameter of the beam. Self-focussing occurs when intensity
reaches a certain limiting value. This threshold value is estimated from the
formula
λ2
I thresh = ...(13.43)
ηl2 ηnl D 2
The formula shows that for higher frequencies and for materials with
greater non-linear susceptibilities, the threshold intensity is lower.
Experimental investigations in self-focussing have been carried out in liquids:
Carbon disulphide, benzene, acetone, etc., For a beam diameters of 0.5 µm, the
self-focussing distance, is about 10 cm and the observed light fibres were 30
to 50 µm, in diameter. It has been further established that the observed light
fibre has still a finer structure; it consists of a number of still thinner filaments
with diameters of about 5 µm [371]. Self-focussing, on the whole, is a
complicated phenomenon.

Laserby Bblaud - 23

Uploaded by

Copyright:

Available Formats

Laserby Bblaud - 23

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Laserby Bblaud - 23

Uploaded by

Copyright:

Available Formats

182 LASERS AND NON-LINEAR OPTICS

11.9 GAMMA-RAY LASER

opened the way to major advances in holography.

12.1 PRINCIPLE OF HOLOGRAPHY

12.1.1 Recording of the Hologram

Fig. 12.1 Recording of Fig. 12.2 Reconstruction

12.1.2 Reconstruction of the Image

The hologram contains in the form of its interference pattern information

Fig. 12.3 Hologram of a point object

and the field Er by the plane wave

Fig. 12.4 Reconstruction of the image

other side of the plate is

Equation (12.10) shows how holography allows us to make a complete

12.3 SOME DISTINGUISHING

(iii) Another characteristic which proves the supremacy of a hologram

12.4 PRACTICAL APPLICATIONS OF

It is not likely that holography will replace photographic methods in everyday

12.5 ADVANCES IN HOLOGRAPHY

the theoretical framework of nonlinear optics. We deal in this chapter some

13.1 HARMONIC GENERATION

Fig. 13.1 Second harmonic generations

magnitude is a function of the frequency. With sufficiently intense laser radiation

order terms. It may be noted that optical characteristics of a medium, such as

13.2 SECOND HARMONIC GENERATION

same direction. This phenomenon is known as the Second Harmonic Generation

where i, j, k represent the coordinates x, y, z,. Most of the coefficients χijk,

– Pi (2) = ε 0 ∑ χijk (–E j ) (–Ek ) = + Pi

which means Pi(2) = 0 and, hence, χ (2)

Second harmonic generation, therefore, cannot occur in an isotropic medium

13.3 PHASE MATCHING

generated harmonics, as they

Suppose a plane wave at frequency ω and the second harmonic wave at

exp(i( k2 L – 2ωt ) [exp(i(2k1 – k2 )L)–1]

for different polarization of light. This ω

13.4 THIRD HARMONIC GENERATION

13.5 OPTICAL MIXING

P (2) = ε0χ(2) (E1 cos ω1t + E2 cos ω2t)2

= ε0 χ(2) (E12 cos 2 ω1t + E22 cos 2 ω2 t )

+ 2ε 0 χ(2) E1E2 cos ω1t cos ω 2t ...(13.25)

= ε0χ(2)E1E2[cos (ω1 + ω2)t + cos (ω1 – ω2)t] ...(13.26)

method of generating ultrasonic waves.

13.6 PARAMETRIC GENERATION OF LIGHT

13.7 SELF-FOCUSSING OF LIGHT

or η = ηl (1 + ηnl E02 ) ...(13.40)

where ηl = εl is the refractive index of the linear medium, and

Fig. 13.4 Self-focussing

You might also like