Download High frequency characterization of the Gsanger LM0202P eletro-optic [i.e. electro-optic] modulator

Calhoun: The NPS Institutional Archive
Theses and Dissertations
Thesis and Dissertation Collection
1996-12
High frequency characterization of the Gsanger
LM0202P eletro-optic [i.e. electro-optic] modulator
Tucker, John R.
Monterey, California. Naval Postgraduate School
http://hdl.handle.net/10945/8468
NAVAL POSTGRADUATE SCHOOL
Monterey, California
THESIS
HIGH FREQUENCY CHARACTERIZATION OF THE
GSANGER LM0202P ELECTRO-OPTIC MODULATOR
by
John
R
Tucker
December, 1996
Thesis Advisors:
S.
Gnanalingam
D. Scott Davis
Andres Larraza
Approved
Thesis
T826
for public release; distribution
is
unlimited.
DUDLEY KNOX LIBRARY
NAVAL POSTGRADUATE SCHOOL
MONTEREY CA 9394^5101
REPORT DOCUMENTATION PAGE
Public reporting burden for this collection of information
is
estimated to average
1
OMB
Form Approved
No
0704-01*
hour per response, including the time for reviewing instruction, searching existing data sources, gathering and
maintaining the data needed, and completing and reviewing the collection of information Send comments regarding this burden estimate or any other aspect of this collection of information,
including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite
VA 22202^302,
and to the Office of Management and Budget, Paperwork Reduction Project (0704-01 88) Washington
AGENCY USE ONLY
(Leave blank)
REPORT DATE
2.
December
3
204, Arlington,
REPORT TYPE AND DATES COVERED
.
Master's Thesis
1996.
High Frequency Characterization of the Gsanger LM0202P Electro-Optic Modulator
4.
1
DC 20S03
FUNDING NUMBERS
5.
AUTHOR(S)
Tucker, John R.
PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES)
7.
PERFORMING ORGANIZATION
REPORT NUMBER
Naval Postgraduate School
Monterey
CA 93943-5000
SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES)
1
.
SPONSORING/MONITORING
AGENCY REPORT NUMBER
1 1
.
SUPPLEMENTARY NOTES The
views expressed
in this thesis are those
of the author and do not reflect the
official policy
or position of the Department of Defense or the U.S. Government.
12a.
DISTRIBUTION/AVAILABILrTY STATEMENT
Approved
13.
for public release; distribution
ABSTRACT
is
12b.
(maximum 200 words) This thesis documents experiments conducted with the Gsanger
electro-optic modulator to achieve a high percentage modulation at
produce a single mode, linearly polarized
modulator with no external
spectrum of the source
light at 514.5
electric field applied,
laser.
DISTRIBUTION CODE
unlimited.
When
an
nm. The
125MHz
laser light
of ail argon-ion
was
first
laser.
The
laser
LM0202P
was tuned
to
passed through the electro-optic crystal
and the frequency spectrum was observed
to be the
AC voltage with a frequency of 125 MHz was applied to the
same
as the frequency
modulator sidebands were
observed by using a Fabry-Perot interferometer. Further measurements were taken to determine the suitability of the
LM0202P
modulator over a large frequency range.
14.
SUBJECT TERMS: High Frequency,
Electro-Optic, Laser Modulation
15.
NUMBER OF PAGES
54
17.
SECURITY CLASSIFICATION OF REPORT
Unclassified
NSN 7540-01-280-5500
18.
SECURITY CLASSIFICATION OF THIS PAGE
Unclassified
19.
SECURITY CLASSIFICATION OF ABSTRACT
Unclassified
16.
PRICE CODE
20.
LIMITATION OF
ABSTRACT
UL
Standard Form 298 (Rev. 2-89)
Prescribed by ANSI Std 239-18 298-102
Approved
for public release; distribution
is
unlimited.
HIGH FREQUENCY CHARACTERIZATION OF THE GSANGER
LM0202P ELETRO-OPTIC MODULATOR
John R. Tucker
Lieutenant, United States
B.S., United States Naval
Submitted in
Navy
Academy, 1989
partial fulfillment
of the requirements
for the degree
of
MASTER OF SCIENCE IN APPLIED PHYSICS
from the
NAVAL POSTGRADUATE SCHOOL
S~\
December 1996
DUDLEY KNOX LIBRARY
l
GRADUATE SCHOOL
M^r
MONTEREY CA 93943-5101
^f
ABSTRACT
This thesis documents experiments conducted with the Gsanger
LM0202P
125MHz
electro-optic modulator to achieve a high percentage modulation at
of
an argon-ion
linearly polarized light at
laser.
The
laser
was tuned
514.5 nm. The laser
light
to
was
produce a single mode,
first
passed through the
electro-optic crystal modulator with no external electric field applied,
frequency spectrum was observed
the source laser.
to the
an
AC
be the same as the frequency spectrum of
voltage with a frequency of 125
MHz was
applied
modulator sidebands were observed by using a Fabry-Perot
interferometer.
the
When
to
and the
Further measurements were taken to determine the suitability of
LM0202P modulator
over a large frequency range.
VI
TABLE OF CONTENTS
I.
II.
III.
IV.
INTRODUCTION
ELECTRO-OPTIC MODULATION
A. ELECTRO-OPTIC MODULATION
1
1.
Birefringence
3
3
3
2.
Electro-Optic Crystals and Modulation
4
13
B. INTERFEROMETRIC SPECTROSCOPY
EXPERIMENTAL ARRANGEMENT
17
17
A. THE MODULATOR
19
B. THE LASER
20
C. THE MODULATOR MOUNT
22
D. THE HYBRID CIRCUIT
23
E. THE INTERFEROMETER
24
F. DRIVING AND DETECTIONS SYSTEMS
29
OBSERVATIONS AND MEASUREMENTS
29
A. THE DC CURVE
B. OBTAINING THE SPECTRUM OF THE MODULATED LASER BEAM
34
CONCLUSIONS AND RECOMMENDATIONS
LIST OF REFERENCES
V.
INITIAL
DISTRIBUTION LIST
VI
41
43
45
I.
Imagine a
light is
made
green
of intense
green
will
between red and
blue.
Moving
the
wave
only
light
"AM-FM
when a wave
conversion" effect (so
you
is
will
see
will
light
again see
long,
pattern of behavior
and
that there
continue
will
is
interacts with
field
itself.
Such
amplitude
interaction
is
thus an example of a nonlinear
possible application of the
AM-FM
lasers using fiber optics (Larraza, 1996).
tunable laser
named because
is
the
way
in
which
mechanism
for
normally negligible,
sufficiently high.
AM-FM
effect.
conversion
An
is
is
to
broadband tunable
essential aspect of the
modulating the intensity of
monochromatic source. Electro-optic modulators are one
modulate the
of the
between amplitude and frequency modulation) should occur
unless the electromagnetic
A
the
from a single-frequency source.
alternates
conversion
where you
further along,
fiber, this
If
This mechanism thus allows the possibility of providing tunable
indefinitely.
This
fiber
Providing that the fiber
are minimal energy losses along the
fiber.
theory predicts (Larraza,
be periodic locations down the
light alternating in intensity.
coherent
guided by an optical
light
to alternate in intensity at the source,
1996) that there
alternating
beam
INTRODUCTION
light of
proposed
a
of the different
ways
to
intensity of a source.
This thesis project attempts to build on research conducted previously at
the Naval Postgraduate School by Ladner (1996) and Wallace (1996) regarding
the characteristic behavior of the Gsanger Opto-Elektronics
LM 0202 P
electro-
optic modulator.
We evaluate
argon ion laser beam
The
thesis
is
at
its
performance as an amplitude modulator of an
modulation frequencies
organized
into five chapters.
in
the range of 125
In
Chapter
II,
we
MHz.
give a
general background of the theory of electro-optic modulators and of Fabry-Perot
spectroscopy.
Chapter
IV,
we
In
Chapter
III
we
describe the experimental arrangement.
provide the details of the measurements.
In
In
Chapter V, we
present our conclusions and recommendations for future work.
II.
In this
chapter
we
ELECTRO-OPTIC MODULATION
present the theory for electro-optic modulators,
following closely the treatment by Karim (1990).
modulator
is
Because the output
analyzed by a Fabry-Perot interferometer,
we
of the
also provide an
overview of the basic principles of that type of interferometric spectroscopy.
A.
ELECTRO-OPTIC MODULATION
1.
Birefringence
Light propagates through a transparent crystal by exciting the electrons
that bind the
atoms or molecules which form the
crystal.
The electrons are
driven by the E-field of the electromagnetic wave, and they re-radiate secondary
wavelets which recombine, producing a resultant wave which moves on. The
speed
of the
wave, and therefore the index
of refraction,
is
determined by the
frequency of the E-field and the responses of the inter-atomic or inter-molecular
bonds
to periodic stimulation.
If
the intermolecular binding forces of the crystal
are anisotropic, the indices of refraction along the three principal axes of the
crystal will manifest this anisotropy.
Each
principal axis will
have a
different
index of refraction.
To
illustrate this idea,
of paired springs.
Each
pair
consider a mass suspended
is
in
a box by three sets
oriented along one of three mutually orthogonal
axes. Each set has a different spring constant. Therefore the characteristic
resonant frequency along each axis
is
different.
Now
imagine that the mass-
spring system represents an atom or a molecule
field of
were
a linearly polarized electromagnetic
parallel to the stiffest springs,
in
a crystalline
solid.
wave propagating through
analogous
If
the E-
the crystal
to the direction of strongest
bonding, the natural resonant frequency of the bond would be relatively high, as
compared
to the other directions.
direction of
Similarly,
if
the E-field
were
parallel to the
weakest bonding, the characteristic frequency would be
lower. This
implies that the behavior of a given source of monochromatic, linearly polarized
light
propagating through the crystal would be determined by the orientation of
the principal axes of the crystal relative to the orientation of the plane of
polarization of the
different,
different
2.
In
one
and the
light.
light
That
is,
the index of refraction along each axis
with different polarizations
speeds. This property
is
known as
moves through
is
the crystal at
birefringence.
Electro-optic crystals and modulation
the case of uniaxial crystals, the kind used for electro-optic modulation,
of the principal
axes coincides with the axis of symmetry of the
this case, there are only
plane of symmetry of the
propagation of
refraction n
.
light
In
two indexes of refraction. Taking the xy- plane as a
crystal,
along
and the z-axis as the symmetry
light
along the xy- plane
the extraordinary index of refraction n e
.
will
be characterized by
For an arbitrary orientation of the
an incident beam
orthogonally polarized components.
axis, the
be characterized by the ordinary index of
this axis will
Propagation of
direction of propagation
crystal.
of light
can be decomposed
One component moves
with a
into
speed
two
determined by n and the other component moves with a velocity determined by
n e This can cause the double image characteristic exhibited by
.
some
birefringent crystals.
For a uniaxial
crystal,
if
the linearly polarized lightwave propagates along
the z-axis, the index of refraction
will
not
will
change the observed index
be n
.
Rotating the crystal about the z-axis
propagating along the x-axis, rotating the crystal
refraction
For
between n and n e
some
crystals
However,
of refraction.
will
the
if
vary the observed index of
.
an external
electric field
can
alter the
and hence the propagation speed. Consider the propagation
crystal in the
absence
of
light is
an externally applied
electric field.
index of refraction
of light in
The
such a
refractive
indices along the respective rectangular coordinate axes of the crystal are
related by
an index
occur,
will
if
we
(2.1)
apply an external electric
and here the
effect
induce a change
represented by
described by
\mJ
\ny J
rixJ
However,
ellipsoid
in
may be
of the
field,
first
a change
order
in
the
in
optical
field.
symmetry may
This linear effect
the coefficients of the index ellipsoid which can
be
^)=ZP, E,
(2-2)
I
This
known as the Pockels
is
,2,3,4,5,6;
and =
optic tensor
which
1
k
I
pp k
(pi k
and
k,
ij).
When
The
x, y, z.
j
is
effect.
6x3
The
p^
are the Pockels constants,
electro-optic matrix
a tensor of rank three,
pi kj
,
1 ,2,
3,
known as the
symmetrical
representing the electro-optic tensor
which take on values
is
in
i
in
=
electro-
the suffixes
I
and
matrix form, the indices
are contracted into single index
i
with the
correspondence 1=(11), 2=(22), 3=(33), 4=(23)=(32), 5=(13)=(31), 6=(12)=(21).
The new index
ellipsoid
is
obtained by adding to Eq. (2.1) the tensor contraction
of Eq. (2.2) with the tensor dyadic form from the three principal directions of the
crystal.
Determining the electro-optic effect
characteristics of the crystal
and
in
the modulator involves using the
finding the principal polarization directions for a
given propagation direction. Knowledge of the refractive indices along the
principal directions
can be used
to
decompose
those principal polarization directions.
the incident optical
We can then
wave along
predict the characteristics of
the emergent optical wave.
Consider the case of an
optical
wave propagating through a
potassium dihydrogen phosphate (KH 2 P0 4 ) or KDP.
nonzero Pockels constants are p 4 i=P52, and p 6 3.
field,
the index ellipsoid
is
described by
In
In this
crystal of
case the only
the presence of an external
V
x + y'
with n x =n y =n
reduces
+ 2p4] yzE x + 2p 4] xzE v + 2p 63 xyE =
,
(2.3)
1
:
\n,j
n:
and n z =n e
.
If
the field
is
applied along the z-direction, Eq. (2.3)
to
2
x'
+ y~2
n„
\
f
2
\nj
We can transform
+ 2p 6l xyE z =
(2.4)
\
equation (2.4) to a form of the type
(2.5)
Ux'J
\Ylz'J
KYlv'J
which has no mixed terms. The parameters
major axes of the index ellipsoid
x', y',
and
z'
denote the directions of
the presence of the external
in
field.
The
lengths of the major axes of the index ellipsoid are given by 2n x 2n y and 2n Zl
,
,
respectively.
By comparing equations
to
each
x'
and
y'
other.
The symmetry
in
(2.4)
Eq
and
(2.5)
(2.4) to
we
find that z
an exchange
in
and
z'
are parallel
x and y suggests that
are related to x and y by a rotation of 45 degrees. Thus
r
x"-
+
P63
E
y' 2
:
_/'
=
+
\*h
1
(2.6)
Comparing equations
3
(1/2)n d(1/n
2
),
we
(2.5)
and
(2.6)
n'y-n +-nlpa E
2
n.=n
1/n
2
differential relation
dn = -
find that,
nx'= n o-2 n oP<* Ez
when
and using the
»p63E z
.
The
Z7a
•
(
,
(2.7b)
,
(2.7b)
velocity of propagation of
)
an emerging wave polarized
along the x'-axis differs from that of the emerging wave polarized along the
axis
.
In
in this
a transverse electro-optic modulator, such as the one
thesis, the crystal
incident
light
y'-
beam
lies in
is
oriented such that the polarization vector of the
y'-axis
mounted normal
and an external
to the z-axis.
emerging wave polarized along the
polarized along the z-axis.
the two
waves
characterize
the x'z-plane at 45° degrees from the x'-axis, while the
propagates along the
plate electrodes
we
field is
The
x'-axis differs
applied by parallel
velocity of propagation of
from that of an emerging
The corresponding phase
shift
difference
an
wave
between
(referred to as electro-optic retardation) after having traversed a
crystal of length L
and width d
is
given by
r=
(
27
*
X,
The value
of
I
n
l
\
V
3
- n ej- n 0P63 7T
V for which r
= n
is
(2.8)
called the half-wave voltage,
Vn
.
For a
transverse electro-optic modulator, long crystals with small width have a low
half-wave voltage. Similarly, propagation along the crystal axis with the parallel
plate electrodes
mounted normal
to the z-axis
corresponds to a longitudinal
electro-optic modulator. In this case the relative retardation
is
length and the width of the crystal and the half-wave voltage
independent of the
may be
significantly
higher.
This electrically induced birefringence can cause a
along the y-axis of a crystal of length
L,
with
its
polarization
the x-axis, to acquire a polarization parallel to the z-axis.
polarization
grows
the traveling
wave
at the
expense
cycles through
of the
light
traveling
initially parallel to
The z-component
x-component as the
elliptical polarization,
wave
total polarization of
then circular
polarization, then elliptical polarization again, until the plane of polarization
rotated 90 degrees (Figure 2.1).
of
has
/1\
Figure
As the vertically polarized light wave propagates through the electro-optic crystal the
plane of polarization is rotated 90 degrees
2.1
If
one were
to place a polarizer at the output
perpendicular to the
initial
is
plane of polarization of the incident wave, the optical
beam would be completely
field is applied,
plane of the crystal that
attenuated
when
the electric field
is off.
But
when
the perpendicular component of polarization would increase
in
proportion to the applied voltage. At the half-wave voltage V„, the polarization
rotated 90
degrees and the
the polarizer.
be
of
light exiting
By varying the applied
the
is
the crystal would pass unattenuated by
electric field the flow of optical
energy can
controlled. This serves as the basis of the electro-optic amplitude modulation
light.
The
birefringent
axes x' and
z'
Output
polarizer
INPUT
Figure 2.2
A
typical
BEAM
arrangement
of
an electro-optic modulator.
l()
Consider the arrangement
crystal
bounded
at either
z'.
The
and
z'
which
e. =
ez
.
we
which
in
turn are at
an
induced birefringent axes
Thus
it
has equal in-phase components along
take as,
A cos at
(2.9)
= Acosatt
Using the complex amplitude notation,
E x .(0)
= A
(2.10)
EAO) = A
The
incident intensity
/,
is
thus,
=\Ex .(0f+\E,,(0f=2A 2
When
the optical
wave
acquired a relative phase
(2.11)
.
exits the crystal the
shift of
x'
the input face of the crystal with a vertical
polarization, parallel to the x-axis.
x'
polarizers,
to the electrically
wave reaches
optical
Figure 2.2, which consists of an electro-optic
end by crossed
angle of 45 degrees with respect
and
in
r
radians.
n
We
x'
and
z'
components have
may take them as
ir
E x ,(L) = Ae"
E Z.(L) =
The
total
(2.12)
A
complex
andE 2 ,(L) along
field
emerging from the output polarizer
is
the
sum
of
E X ,(L)
the analyzer, which gives
C-£(.--i)
(2.13)
This corresponds to an output intensity of
'o^^- )^- )]^^^
1
The
ratio of the
K =
—
.
sin
in
(2.14)
^
output intensity to the input intensity
r
—
= sin
.
2
/.
Thus
2
2
1
2
is
thus
7t\V_
(2.15)
2JV,
Eq. (2.15), zero applied voltage corresponds to total extinction
the electro-optic modulator.
As
in
the applied voltage increases to the half-wave
voltage, the transmission (output) intensity increases to
applying a sinusoidal modulating voltage about the
12
50%
100 percent. By
output intensity voltage,
beam can be
1
/4Vn a sinusoidal modulation of the optical
,
induced. This can be
represented mathematically by writing
V
=
V2V n +Vp sinco m
t
For peak voltages smaller than
On
replica of the modulating voltage.
is
comparable
distorted
and
to
will
V£V n
,
it
(2.16)
.
follows from
V2\/n
,
the intensity modulation
the other hand,
Eq
if
is
a linear
the modulating voltage
(2.15) that the intensity variation
is
contain an appreciable amount of odd harmonics of the
fundamental modulation frequency.
B.
INTERFEROMETRIC SPECTROSCOPY
A Tropel 350
Fabry-Perot interferometer
frequency spectrum of the laser beam
parallel glass surfaces, coated with
parallel surfaces are separated
differential
When
to
It
analyze the
consists of two
a highly reflecting dielectric
d.
One
to allow for fine
plate
is
film.
The
mounted on three
adjustment of parallelism.
these two surfaces are held adjusted for parallelism they are referred to
as an etalon. The other plate
moves back and
multiply reflected
set
is
attached to an electrically driven motor and
forth a distance of the order of
portion of the laser
will
experiment.
in this
by a distance
micrometer assemblies
was used
beam
one wavelength
entering the interferometer through
between the two
up a standing wave which
plates.
will
If
the etalon
is
(Fig 2.3).
one
of the plates
aligned properly, this
cause constructive interference
13
A
at the
is
output of the interferometer, and produce a circular interference pattern
consisting of light
and dark
produces a phase
interference.
fringes.
shift of 2n,
The separation
Each
corresponding
in
^
is
of light
By changing the plate separation
power as
125MHz modulation
resolving
power
is
is
of constructive
d,
In this
we change
required.
of the laser with
of refraction of the inter-plate
case the medium
the FSR,
The goal
and
so
is air,
this allows
of this research
was
to
n«
us
beam
to
to
determine the effect that the
light.
14
means
1.
to
observe
an electro-optic modulator. The high
of the Fabry-Perot interferometer provides the
having on the laser
is
related to the etalon spacing by
and index
observe the modulation sidebands, and
modulation
one order
(2.17)
the inter-plate spacing.
adjust the resolving
to
,
where c and n are the speed
medium and d
wavelength of plate travel
frequency between successive spectral orders
called the free spectral range (FSR) and
FSR =
half
Highly reflecting dielectric plates
Incoming beam
Multiple Reflections
Interference fringes
Photo-detector
Etalon spacing = d
Figure 2.3
Fabry-Perot interferometer uses glass plates coated with a highly reflecting
The beam enters and a portion of it is reflected back and forth
between the two etalons, one of which is moving the distance of about one
wavelength. This causes constructive and destructive interference producing an
dielectric film.
interference pattern on the photo-detector.
15
16
I.
In this
chapter
we
EXPERIMENTAL ARRANGEMENT
provide a description of the components used
in
our
experiment. Figure 3.1 shows the schematic of the experimental arrangement.
HP 35670A Dynamic
Signal Analyzer
Microscope slide
Photo-detectors
attenuator
Fabry-Perol Interferometer
moveable
Microscope
1
LEXEL mod
85 Argon-
modulator
Ion Laser
slide
-
-••/
/
AhS
D,
Motorola
AR305
Hybrid
Amplifier
Tektronix 2465 B
PD30
300MHz
Detector 3:
Oscilloscope
photo-detector
and power supply
Mini -Circuits
RF
HP 6205B DC
Amplifier
Power Supply
Tektronix 2445 B
200MHz
HP6205BRF
Oscilloscope
Signal Generator
Figure 3.1.
The modulator, the mirrors, and microscope slides were supported by kinematic
mounts, allowing the beam path to be changed quickly. Detector 1 (D1) was alternated between
the two positions indicated as required by the experiment.
A.
THE MODULATOR
The Gsanger Opto-Elektronics Model LM0202P
(Fig 3.2)
is
the principal
component
electro-optic modulator
of this experimental investigation.
17
The
modulator was designed
through the
cell
electro-optic modulator
5 Watts
at
beam
intensity
was
100% modulation
As
cell.
in
is
the
cell.
laser light passing
designed
According to the manufacturer, the
to
modulate a laser
at
in
the range of 75-100
beam
In
with
an
to
LM0202P
intensity of
our experiment the
mW and our goal was to achieve
125MHz.
The LM0202P transverse electro-optic modulator. Shown in the picture are the
Brewster cube mounted at the exit end and the microdot connectors on the side.
The input aperture is a 3x3 mm square.
stated
in
Chapter
the
II,
LM0202P
is
a transverse electro-optic
modulator, with four electro-optic crystals operating
passes through each
end
The
modulation frequencies up to 150 MHz.
0.1 to
3.2.
as a Pockels
can be amplitude modulated by applying a varying voltage
the birefringent crystals
Figure
to function
crystal
of the modulator there
is
its
polarization
in series.
mode changes.
a Brewster cube that
splits
component passes
18
straight
At the downstream
the laser
two mutually orthogonal polarized components. As depicted
horizontally polarized
As the beam
in
beam
into
its
Figure 3.3, the
through the modulator, while
the vertically polarized component
voltage, the horizontally polarized
wave
is
90 degrees. With no applied
deflected
component
is
voltage the vertically polarized component
extinguished, while at the half-
is
extinguished.
Transmitted
Beam:
Deflected Beam: Vertically Polarized
Horizontally
Input
Beam from
laser
Polarized
:
Vertically Polarized
Figure
Schematic of the modulator.
3.3.
polarized)
beam
is
With no applied voltage, the transmitted (horizontally
When the applied voltage equals the half-wave
extinguished.
voltage, the deflected (vertically polarized)
B.
beam
is
extinguished.
THE LASER
The
laser
was supported
standard lab jacks.
It
was tuned
the two knobs labeled
at the rear
end
iterative
process
tuning
in
one end
of the optical
to optimal single
mode
bench on two
operation by adjusting
HORIZONTAL and VERTICAL & WAVELENGTH
of the laser.
mode whenever
at
is
The
located
laser must be operating in the current-control
performed.
The knobs are used
alternately in
an
order to obtain the desired wavelength and intensity. The
VERTICAL & WAVELENGTH knob
selects the wavelength by adjusting an
etalon within the laser resonant cavity.
A
counterclockwise turn selects a longer
wavelength and a clockwise turn selects a shorter wavelength. The peak power
is
adjusted after each wavelength change by using the
knob and adjusting
for
peak output power.
19
HORIZONTAL
tuning
The
laser frequency spectrum
was measured
with a Fabry-Perot
interferometer along with the dynamic signal analyzer.
optimize single
viewed on the
mode
The
digital signal analyzer.
A
was adjusted
to
The spectrum was
output by observing the spectrum.
HP 35670A
laser
made by
pinhole aperture,
poking a small hole (estimated diameters 100 microns) through a piece of
aluminum
aperture,
foil,
in
was
positioned
in
the
beam
path directly
in front
D1 This
.
conjunction with the other optics, functioned as a spatial
minimize the
field of
filter to
view of the detector.
THE MODULATOR MOUNT
C.
An
effective
means
for
mounting the modulator, shown
devised by using a micro-rotary positioner mounted on a
stage which allows
and
of
limited rotation
tilting
stage
be moved
in
five
degrees of freedom: translation
about the x and y axes
was attached
and out
modulator was
to a kinematic
of the
fitted into
beam
(ie: tilting in
in
Figure 3.4,
New Focus 9082
in
the
pitch
x, y,
tilting
and z axis
and yaw). The
mount which allowed the modulator
path without requiring readjustment.
the micro-rotary positioner by
aluminum adapter.
20
was
means
of a
to
The
home made
Figure
Modulator mounted on the azimuthal and
3.4.
tilting
stages.
This mount allowed micro-rotary adjustment of the modulator
azimuthal direction to micro-radian tolerances.
window
It
in
also positioned the entry
of the modulator over the center of rotation of the tilting stage.
the metal mount exposed more surface area of the
greater passive cooling
variations
in
in
the
cell to
order to minimize possible
Finally,
the air to allow for
drifts
due
to thermal
the modulator.
By positioning the LM0202P modulator
down beam from
the
window end, and
window, the modulator was
the input beam. This
tilted
the
beam path
off
the
was achieved by mounting a two
center of the disk allowed the
A two
beam through
window could be viewed
21
with the analyzer
beam centered on
so the reflection
between the laser and the modulator.
reflection from the
with the
in
the entry
window coincided
inch
with
aluminum disk
millimeter hole drilled through the
the aperture.
relative to the
The
position of the
beam by observing
the
on the aluminum
reflection
made
asymmetrical which
this
procedure yielded
rotated about
transmitted
reflected
its
However, the reflection from the window was
disk.
this
reliable
adjustment more
art
and reproducible
than science. Nevertheless,
results.
The modulator was then
longitudinal axis so the intensity of the horizontally polarized
beam was minimized
beam was
while the intensity of the vertically polarized
maximized. This step ensured that the x and z axes were
properly positioned with respect to the incident laser beam, which
was
vertically
polarized.
D.
THE HYBRID CIRCUIT
When
an
AC
voltage
because the dipoles
in
is
applied, the electro-optical crystal heats up,
the crystal are oscillating back and forth.
As the
frequency of the modulation increases from kilohertz to megahertz,
becomes more pronounced. This thermal
refraction of the birefringent crystals. Therefore
the modulator
equilibrium
needed about 3-5 minutes
whenever the modulation
bias voltage
was then
crossover point as
Fig. 3.5,
signal
was used
was applied
is
applied
in
discussed
may modify
effect
signal
the indices of
when conducting
to stabilize at
a
this effect
new
the experiment
thermal
was changed. A compensating DC
order to keep the modulator centered at the
Chapter
in
to allow for the
DC
IV.
A
hybrid
circuit,
as depicted
biasing of the modulator while the
simultaneously.
22
RF
in
From RF Amplifier
R
C,
From
L
To Modulator
DC
Power Supply
Figure
3.5.
The Hybrid
Circuit:
d= 0.01 ^F, C
2
=
C3 =
0.0047uF,
The Fabry-Perot
interferometer as described
physical arrangement of the overall experiment.
sensitive to the
The
L =
RF choke 1uK
THE INTERFEROMETER
E.
the
R = 100kU
beam
most minute adjustments, and
Chapter
II,
dictated the
The device was extremely
was
absolutely crucial to have
traverse the center of the etalon perpendicularly to the plate surfaces.
interferometer's optical axis
rest of the
it
in
was
the geometric reference around which the
experiment was aligned. Once
fixed.
23
it
was
properly adjusted,
it
was
held
DETECTION AND DRIVING SYSTEMS
F.
The photo-detectors used
the experiment
in
Newport 818 SL photo-detectors, each connected
power meter, were
beam
intensities.
detector 2 (D2)
to
a
DC
primarily
They are
used
interferometer to record the fringes.
HP 35670A dynamic
and D2 are however,
not be useful
An
was used
in
PD30
modulation
specified bandwidth of 1GHz.
an
optical
power
onto
D3
beam
for the
3mW,
Due
(D1) and
of the incident light
Fabry-Perot
was then
routed to the
and recorded the
in
the transmitted
beam because
it
of the
However, the PD30 photo-detector saturates
though
it
intensity with
beam
to the large
was
required to monitor a
slide
was used
beam
it
to reflect a fraction of the
power and
possible to simultaneously monitor
D1 which was mounted
,
slide.
number
Detector
of electronic
limited.
1
was
and
directly in the
main
calibrated to account
optical devices
The modulator,
24
at
intensity that
intensity being reflected into detector 3.
experiment, bench space was
D1
had a
(Fig 3.1). This microscope slide attenuated the optical
beam
data.
photo-detector, identified as detector 3 (D3),
path behind the microscope
10%
signal
1
modulation of the laser beam.
provided the additional benefit of making
the transmitted
at the output of the
The voltage
MHz
0-60mW. A microscope
varied from
beam
of
Series
frequency response to only 180 kHz and would
limited in
Opto-Electronics
own Newport 815
as detector
signal analyzer which monitored
detecting the 125
to detect
its
They both converted the power
D1 was also used
.
to
The two
of two types.
measure the transmitted and deflected
identified respectively
in this thesis.
voltage signal
to
were
used
in
the laser and the
the
interferometer were
realign quickly.
adjusted, they
space and
to
all
very sensitive to small perturbations and
to
be disturbed.
In
minimize the time required
configurations, two different
slides
once they had been aligned and
Efficiency dictated that,
were not
were attached
beam
to kinematic
order to optimize the use of bench
One beam
unmodulated laser
was
between
to transition
optical
paths were used. Mirrors and microscope
mounts which allowed
without requiring constant readjustment. Each
reflectors.
difficult to
beam
for their
path had
rearrangement,
its
own
set of
path went around the modulator so a spectrum of the direct
light
could be measured (Fig 3.6.a).
routed through the modulator so the
AM
The
other
apertures
slide
attenuator
Fabry-Perot
Interferometer
Beam
Path
LEXEL mod 85
Argon-Ion Laser
modulator
(a)
25
path
spectrum could be measured (Fig
3.6.b).
Microscope
beam
apertures
Microscope
slide
attenuator
Beam
Path
LEXELmod85
^>/
1
1
Argon-Ion Laser
7
1
1
(b)
Figure
Beam
3.6. (a)
path for baseline measurements, (b)
Beam
path through modulator.
For both the baseline measurements of the laser
beam coming
used
to
(Fig 3.6. a)
and the
out of the modulator (Fig 3.6b), a microscope slide attenuator
reduce the optical power
microscope
beam
slide
was scratched
into the interferometer.
sandpaper
with
to
The back
defuse the
was
side of the
light reflecting off
the back surface, thus minimizing interference caused by the secondary
reflection.
Two
oscilloscopes, a Tektronix
(300 MHz), were used
2445B (200 MHz) and a Tektronix 2465
to monitor the applied electrical signal
modulation. The reason for using two separate oscilloscopes
and the resultant
was
that internal
coupling between channels, at very high frequencies interfered with the
observed modulation signal when only one multi-channel oscilloscope was used.
The
applied electrical signal (20 -160
V p p was
.
)
several orders of magnitude
greater than the modulation signal (mV) from the photo-detector, hence vastly
different oscilloscope sensitivities
were required.
26
An HP 8640B
modulator
at
Signal Generator provided the
125 MHz. The output
amplified with a Mini-Circuits
up
to
40
Vp. p
.
A second
amplitudes above 40
by an
the
Vp p
.
RF power
up to 160
supply.
Vp p
.
signal for driving the
(maximum
of
3V rms was
),
amplifier providing signal amplification
stage amplifier, a Motorola
HP 6274B DC power
DC
of the generator
RF
.
AR305
The Motorola
RF, was used for
amplifier
was powered
An HP 6209B DC power supply provided
bias to the modulator.
27
28
IV.
A.
OBSERVATIONS AND MEASUREMENTS
THE DC CURVE
To determine the half-wave voltage
DC
shown
is
in
we determined
The experimental arrangement used
characteristic response curve.
purpose
of the modulator,
its
for that
Figure 4.1.
D
Reflected beam:
Vertically
Polarized
LEXEL model
LM0202P E-O
85
H modulator
Argon- 1 on Laser
»D,
Y
Transmitted
beam:
DC
Horizontally
Power
Polarized
Supply
Figure
4.1.
Arrangement
for recording the
The modulator was
beam
aligned for
(horizontal polarization) as
voltage
was
steps of 10
zero.
volts,
The DC
DC
characteristic curve.
maximum
extinction of the transmitted
measured by detector
voltage
Di,
when
was then increased from zero
and the readings
of Di
and D 2 were noted
at
the applied
to
each
330
volts, in
step.
The
results are plotted in Figure 4.2.
At the point where the two curves cross, the modulator behaves as a
quarter-wave
while
50%
is
plate,
and 50%
of the
beam
is
transmitted (horizontally polarized)
reflected (vertically polarized). This point will henceforth
be
referred to as the crossover point. Applying a sinusoidal electrical signal to the
29
modulator modulates the transmitted beam about the crossover
AC
the amplitude of the applied
approximately linearly
signal
the
signal increases the percentage of modulation
the peak-to-peak amplitude of the applied electrical
until
approaches the half-wave voltage
maximum and minimum
of the transmitted
is
no longer
is
exceeded.
AC
————
i
I
i
signal
—
i
i
|
As the modulation
a maximum, the relationship between the
and the percentage modulation
Over-modulation of the
linear.
60
of the modulator. This coincides with
points on the characteristic curve.
beam reaches
amplitude of the applied
point. Increasing
i
i
—j— —r— —i— —
i
i
beam occurs
|
i
i
i
i
i
i
|
i
—
i
if
i
—i— — — —|— —
|
beam
the half-wave voltage
i
i
of the
i
i
i
i-
o-o-cx
DC Voltage
(V)
Figure 4.2.
The DC characteristic curve for the LM0202P electro-optic modulator. The open
diamonds correspond to the transmitted beam measured by D1 (horizontally polarized) and the
stars correspond to the beam measured by D2 (vertically polarized).
30
On
the basis of the
an applied
AC
DC
characteristic of the modulator,
voltage of 120 volts peak-to-peak would result
is
it
in
expected that
about a
90%
modulation, with the modulator operating at the crossover point. However,
this
AC
voltage at 125
MHz was
when
applied to the terminals of the modulator, the
actual percentage modulation observed
in
the optical
beam was
only about
20%.
This clearly confirmed the conclusion previously reached by Ladner (1996), that
only a fraction of the
RF
electro-optic crystal.
A
voltage applied to the modulator terminals reached the
large fraction of the applied signal
was taken up by
the
large high frequency inductive reactance of the internal wire leads to the crystal.
Furthermore,
when
it
was observed
the applied
125MHz AC
peak
to
there
was appreciable
140
frequencies.
capacitance
that the
voltage
volts peak-to-peak.
The
A
in parallel
with a resistance,
to specifically
address
volts peak-to-
possible explanation for this behavior
electrical behavior of the crystal
RF
is
is
that
thus equivalent to that of a
and the value
of the resistance
voltage across the crystal
equivalent circuit of the modulator
will
was increased from 100
dielectric loss in the electro-optic crystal at very high-
decreases steeply when the
designed
percentage modulation hardly increased
is
measure the
shown
in
Figure 4.3.
is
increased.
Future experiments
electrical characteristics of the
this issue.
31
The
modulator
R
"N/N/"
Figure
Equivalent circuit for the modulator. C is the capacitance of the crystal, R is the
resistance equivalent to dielectric loss, L is the inductance of the wire leads
inside the modulator.
4.3.
Inspection of the
DC
characteristic curve, Figure 4.2,
percentage modulation measured by D,
horizontally polarized
(
shows
the intensity of the transmitted
beam) could be increased by lowering the mean value
the reading of Di below the crossover point, while maintaining the
of the reading of Di at a constant amplitude.
though
it
would
envelope.
20%
to
It
80%
We adopted this
result in nonlinear modulation
and a
AC
of
excursion
method, even
slightly distorted
modulation
turned out that the percentage modulation could be increased from
at
125MHz
voltage to lower the
voltage at
that the
60
without serious distortion, by suitably adjusting the
mean
value of the reading of D^ while maintaining the
DC
AC
volts peak-to-peak.
A study
of the modulator's
value of the reading of D,
is
DC
made
characteristic reveals that
when
the
mean
too small, the negative peaks of the
modulation envelope become flattened, and the effective percentage modulation
decreases.
We
performed measurements
to
determine the
mean value
of the
reading of Di that would yield the highest percentage modulation. Figure 4.4
shows
the results of these measurements.
32
0.00
0.05
0.10
0.15
Ratio of the readings of D1 to
Figure
4.4.
D2
Percent Modulation vs ratio of the power output measured by detectors Dt and
D 2 The percentage modulation obtained with an applied AC voltage of 60 volts
peak-to-peak at 125 MHz is plotted against the ratio of the mean value of the
reading of D, to the mean value of the reading of D 2
.
.
The percentage modulation obtained
volts
peak-to-peak
at
125
MHz
is
with
an applied
AC
plotted against the ratio of the
D2 A
voltage of 60
mean
value of
the reading of D^ to the
mean
value of the reading of
modulation against the
mean
value of the reading of D^ would have yielded
.
plot of
percentage
inconsistent results because the absolute values of the readings of D^
were subject
to variability,
D2
and D 2
depending upon the output of the laser and the
percentage transmission of the modulator. The
the readings of
of
that yielded the highest
ratio of the
readings of Di over
percentage modulation was 0.07.
We also performed experiments to determine the variation of the
percentage modulation with the applied
AC
33
voltage at the crossover point and at
a low modulation frequency of 20 kHz, an intermediate frequency of 20 MHz,
and a high frequency
of
percentage modulation
predicted from the
DC
125MHz. Figure 4.5 shows a comparison between the
at
these frequencies and the percentage modulation
characteristic.
The
of the internal wire leads to the crystal,
deleterious effects of the inductance
and the presumed high frequency
dielectric loss in the crystal are both clearly evident.
I
i
I
I
100 -
e
o
jm
80 -
/•
I
I
I
—
^^r"
-
•
W
a
c
o
E
e
o
/•
A
60 -
-
A
A
/»
•
A
40 -
A
o
w
0)
a.
A
20 -
-
T
I
T
20
I
I
40
60
i
I
I
60
100
120
i
160
140
180
Modulation Voltage (peak-to-peak)
Figure
4.5.
Percent modulation vs. modulation voltage for DC (solid squares), 20 kHz
20 MHz (up triangles), and 125 MHz (down triangles).
(solid
circles),
B.
OBTAINING THE SPECTRUM OF THE MODULATED LASER BEAM
During measurements of actual
held constant and
all
AM
modulation, the laser intensity
was
major components were mechanically fixed. Because of
the sensitivity to alignment conditions, the process of taking
organized to be done quickly
in
a consecutive sequence.
34
measurements was
The
first
step required
setting the
beam
measurements
path to route the
beam around
of the laser spectrum could
spectrum of the laser beam, before
was
it
set to 10 cm, corresponding to a
t
—
-
,
the modulator, so baseline
be taken. Figure 4.6 shows the
enters the modulator.
FSR
of
The
1.5GHz.
'
1
i
etalon spacing
1
'
l
I
3.0
GHz
1.5
-
2.5
c
2.0
ui
-
Amplitude
b
"
0.5
J
^
0.0
0.0
0.2
:
^-
1
.
0.4
1
—
.
0.6
-r
1
0.8
,
1.0
Time (sec)
Figure
4.6.
Baseline spectrum. The laser was tuned for single
1.5
mode
operation.
The FSR
is
GHz.
Immediately after obtaining the laser baseline spectrum, the
beam was
routed through the modulator with no voltage applied. Figure 4.7 shows the
resulting spectrum.
reduction
Comparing the Figures 4.6 and
in light intensity in
4.7, indicates
there
the modulator, but no extraneous structure
to the spectrum.
35
is
is
a
added
i
!
i
<
1
1
i
'
1
1
2.2
1.7
>
E,
-
a)
"D
_
1.1
D
_
+±
Q.
"
,
E
<
^
0.6 -
-
A
A
0.0
.
i
0.0
i
.
0.4
0.2
i
.
0.6
.
i
0.8
i
1.0
Time (seconds)
Figure
The unmodulated beam through
4.7.
Having established
that the
spectral lines, the next step
was
to
the modulator.
The FSR
is 1 .5
GHz.
modulator was not adding structure to the
observe the spectral lines while the
modulator operated at 125 MHz.
For
frequency,
AM
modulation, the frequency spectrum consists of a center carrier
co c
with sidebands that are offset on either side of the carrier
,
frequency by the
frequency,
co
m
.
sum and
difference of the carrier with the modulation
The amplitude
of the
sidebands of the
AM
signal
is
one-half the
amplitude of the carrier times the fractional amount of modulation. This
depicted
in
Figure 4.8, and
is
is
mathematically described by Eqs (2.13)-(2.16).
36
co c
-co r
COc+G),,
Frequency
Figure
4.8.
AM
modulation spectrum. The carrier a frequency
frequency
In
© m As
.
our case, the carrier
D1 D2, or D3,
,
optical
is
it
is
is
to the
not feasible to
modulated
to the
it
and the modulation
100 %.
less than
directly
up
is
to
5.8 x 10
14
Hz.
Instead,
D3
frequencies of 300 MHz.
from a Tektronix 2465 300
of modulation
of the
was determined by
MHz
dividing the
modulation by the average voltage.
Because the beam
made
of the carrier
intensity of light
and the percentage
peak voltage due
,
measure the instantaneous amplitude
The readout from D3 was measured
oscilloscope,
is
an argon-ion laser beam. With either detector
wave, because the frequency
can respond
is co c
depicted, the percent modulation
routing
in
the experimental arrangement (Figure 3.1
impossible to simultaneously measure the modulation percentage and to
view the spectrum,
we
first
measured the percentage modulation from the
oscilloscope readings, as stated above, and then inserted a mirror to redirect the
beam
path into the interferometer
in
order to observe and record the spectrum.
This caused a delay of about 10 to 15 seconds between recording the
modulation percentage and recording the frequency spectrum. However,
not appear to be a significant problem.
37
this did
With 40
Vp.p at
125
value for the signal from
MHz
D3
applied to the modulator, the ratio of the peak
to the
a modulation of 60%. Given the
expected
to
equal to the
average value
FSR
see sidebands appear
ratio of
the main fringe.
125
MHz to
in
.5
GHz and 60%
D3
modulation,
predicted
we
the observed spectrum with separation
the FSR, and with an amplitude equal to
of the amplitude of the sidebands
the average value of the central peak
sideband amplitude
D 3 was
read
off
is
30%
is
about 0.02mV, while
about 0.071 mV, corresponding to a
to the central
modulation of 56%. The difference
signal from
1
from
Figure 4.10 shows the spectrum for this level of modulation.
The measured value
ratio of the
of
for the signal
peak amplitude
may be accounted
for
of 0.28, or a
by the fact that the
the oscilloscope, leading to not so precise
determinations of the percent modulation.
38
of
0.08
0.06
1.5
GHz
>
0.04
a
3
Q.
£
0.02
+0.12
GHz
<
0.00
0.0
0.4
0.2
0.6
0.8
1.0
Time(sec)
Figure 4.10.
60%
0.06
modulation with 40
V pp
at
125 MHz.
-
1.5
GHz
0.05
^
wE
>
0.04
0.03
a>
o
3
~
0.02
a.
I—
+0.12
GHz
E
<
0.01
-
0.00
-0.01
0.0
0.4
0.2
0.6
Time(sec)
Figure 4.11.
80%
modulation with 60
Vp p
.
at
125 MHz.
39
0.8
1.0
For an applied signal of 60
the signal from
D3
to the
Vp p
.
at
125 MHz, the
ratio of the
average value of the signal from D 3 predicted
modulation. Therefore the observed amplitude of the sidebands
be 0.4 times of the amplitude
those
in
of the central
peak
is
ratio of 0.40,
modulation.
40
to
peak with the same separation as
The value
about .057mV. The average value of the sideband
.023mV. This yields an observed
of
80%
was expected
Figure 4.10. Figure 4.11 shows the record of this modulation.
of the central
80%
peak value
is
corresponding to the predicted
V.
CONCLUSIONS AND RECOMMENDATIONS
We successfully characterized the
optic modulator
and achieved amplitude modulation
125MHz. However,
in
order to do this
from crossover point on the
intensity of the transmitted
only
7%
of the
When
at
RF
performance of the
beam was
DC
we had
of
LM 0202 P
an argon ion laser
to bias the
A
high modulation of
at
quiescent point away
characteristic curve, which lowered the
beam.
electro-
80% was
mean
achieved, but
actually transmitted through the modulator.
the electric field applied to the
LM 0202P
modulator
was modulated
frequencies the effective parallel resistance of the electro-optic crystal
reduced. This
is
perhaps a
result of dielectric losses in the crystal,
the modulator performance for modulating
Additional research
behavior
may be conducted
profile of the
Nevertheless, there
modulator
may be
to
light
which
near and about 100 MHz.
modulation frequencies.
useful applications for this modulator at lower
41
limits
determine the precise characteristic
at various
frequencies.
is
42
LIST
1
Larraza and Coleman
(1
OF REFERENCES
996), Nonlinear Propagation in Optical Fibers:
Applications to Tunable Lasers, Andres Larraza, paper prepared for thesis
students.
2.
Michael C. Ladner (1 996), Optical Modulator LM 0202 P Characteristics:
Application to Amplitude Modulation of Argon-Ion Laser, Naval Postgraduate
School.
3.
4.
5.
Harlan Wallace
(1
996), Optical Characteristics of the
LEXEL 85 Argon-Ion
AM-FM Light
Laser and Gsanger LM0202P Modulator: Application
Conversion Naval Postgraduate School.
to
Mohammad A. Karim (1990),
KENT Publishing Company.
Electro-Optical Devices
and Systems, PWS-
Amnon
Waves
Yariv
(1
984), Optical
in Crystals:
Laser Radiation, John Wiley and Sons,
43
Inc.
Propagation and Control of
44
INITIAL DISTRIBUTION LIST
Defense Technical Information Center
8725 John J. Kingman Rd., STE 0944
Ft. Belvoir, Virginia 22060-6218
2
2.
Dudley Knox Library
Naval Postgraduate School
411 DyerRd.
Monterey, California 93943-5101
2
3.
Dr. A. A. Atchley,
1.
Code PH/Ay
1
Department of Physics
Naval Postgraduate School
Monterey, California 93943-5002
4.
Professor A. Larraza, Code PH/La
Department of Physics
2
Naval Postgraduate School
Monterey, California 93943-5002
5.
Professor Scott Davis, Code PH/Dv
Department of Physics
2
Naval Postgraduate School
Monterey, California 93943-5002
6.
Professor S. Gnanalingam,
Department of Physics
Naval Postgraduate School
Code PH/Gm
2
Monterey, California 93943-5002
7.
LT. John R. Tucker
5961 N. Brook PI.
Garden City, Idaho 83714
2
45
SwK3KKffi M "'
,,S,yStiYJ<NQXyBRARY
3 2768 00335775 7