close

Вход

Забыли?

вход по аккаунту

?

AN OPTICALLY-PUMPED FREQUENCY STABLE TUNABLE MASER AND INFRARED-MICROWAVE DOUBLE-RESONANCE EXPERIMENTS IN AMMONIA

код для вставкиСкачать
INFORMATION TO USERS
This was produced from a copy of a document sent to us for microfilming. While the
most advanced technological means to photograph and reproduce this document
have been used, the quality is heavdy dependent upon the quality of the material
submitted.
The following explanation of techniques is provided to help you understand
markings or notations which may appear on this reproduction.
1. The sign or "target" for pages apparently lacking from the document
photographed is "Missing Page(s)" If it was possible to obtain the missing
page(s) or section, they are spliced into the film along with adjacent pages.
This may have necessitated cutting through an image and duplicating
adjacent pages to assure you of complete contmuity.
2. When an image on the film is obliterated with a round black mark it is an
indication that the film inspector noticed either blurred copy because of
movement during exposure, or duplicate copy. Unless we meant to delete
copyrighted materials that should not have been filmed, you will find a
good image of the page in the adjacent frame.
3. When a map, drawing or chart, etc , is part of the material being photographed the photographer has followed a definite method in "sectioning"
the material. It is customary to begin filming at the upper left hand corner
of a large sheet and to continue from left to right in equal sections with
small overlaps. If necessary, sectioning is continued again—beginning
below the first row and continuing on until complete.
4. For any illustrations that cannot be reproduced satisfactorily by
xerography, photographic prints can be purchased at additional cost and
tipped into your xerographic copy. Requests can be made to our
Dissertations Customer Services Department.
5. Some pages in any document may have indistinct print. In all cases we
have filmed the best available copy.
University
Microfilms
International
300 N ZEEB ROAD, ANN ARBOR Ml 48106
18 BEDFORD ROW, LONDON WC1R 4EJ, ENGLAND
8026552
MALK, EDWARD GEORGE
AN OPTICALLY-PUMPED FREQUENCY STABLE TUNABLE MASER AND
INFRARED-MICROWAVE DOUBLE-RESONANCE EXPERIMENTS IN
AMMONIA
University of Illinois at Urbana-Champaign
PH.D.
1980
University
Microfilms
IntGrnmlOnSl
300N Zeeb Road, Ann Arbor, MI 48106
18 Bedford Row, London WC1R 4EJ, England
AN OPTICALLY PUMPED FREQUENCY STABLE TUNABLE MASER
AND INFRARED-MICROWAVE DOUBLE-RESONANCE
EXPERIMENTS IN AMMONIA
BY
EDWARD GEORGE MALK
B.S., University of Illinois, 1975
M.S.,.University of Illinois, 1977
THESIS
Submitted m partial fulfillment of the requirements
for the degree of Doctor of Philosophy in Electrical Engineering
in the Graduate College of the
University of Illinois at Urbana-Champaign, 1980
Urbana, Illinois
UNIVERSITY OF ILLINOIS AT URBANA-CHAMPAIGN
THE GRADUATE COLLEGE
March,
1980
WE HEREBY RECOMMEND THAT THE THESIS BY
EDWARD GEORGE MALK
FMTTTT.RD AN OPTICALLY PUMPED FREQUENCY STABLE TUNABLE MASER
AND INFRARED-MICROWAVE DOUBLE-RESONANCE EXPERIMENTS IN AMMONIA
BE ACCEPTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR
t Required for doctor's degree but not for master's
o )17
ill
ACKNOWLEDGEMENT
The author wishes to thank the many people that helped
make this endeavor a success. Professor Paul Coleman has always provided patience, understanding, guidance, and constant
encouragement, during the many trying times of graduate school.
I am proud to call him colleague and friend.
The other members, past and present, of the Electro-Physics
Laboratory deserve special thanks, for always providing the
"loyal opposition" that helped build the author's selfconfidence.
In particular, thanks to Professor Tom DeTemple
for the endless minutes of discussion and the introduction to
the Electro-Physics Lab. The 207 crew of John Leap, Dave Kim,
Rich Adams, Bob Miller, Ray Chu, and Bill Lee (honorary) are
thanked for the many discussions and being able to cope with
the author's possessive nature. Mark Durschlag and Lmdley
Specht are thanked for the experimental help provided on several
occasions.
Jimmie Smith, who has managed to maintain her sanity during
the preparation of this thesis, deserves special thanks. Her
artwork and typing skills are unmatched anywhere.
The technical assistance provided by Don Fulton, Keith
Keuhl, and Ed Boose will always be remembered.
Barney Marshall
deserves more than mere words can express. He has always
IV
managed to finish the endless "just one more thing" jobs in
spite of my efforts.
To A. B. Wilson, unofficial co-author of
this thesis and personal mentor in financial matters,
thanks.
Professors Nick Holonyak, Greg Stillman, Joe Verdeyen and
Oscar Gaddy are graciously thanked for equipment loans and
patient ears.
The primary support for this research was supplied by Grant
AFOSR 76-2988, monitored by Gordon Wepfer and Harold Schlossberg.
Additional funding by the University of Illinois Industrial
Affiliates Program and NSF Grant ENG 79-16455 is greatly appreciated.
To my wife, Joan, and daughter, Chanda, I dedicate this
thesis.
They have suffered the most during this time and de-
serve the thanks that words alone cannot express.
V
TABLE OF CONTENTS
Page
I.
II.
INTRODUCTION
TUNABLE MASER EMISSION FROM
1
14
NH_ OPTICALLY
PUMPED BY A C 0 2 LASER
Introduction
Cell Design
.
Experimental Apparatus
Infrared Pump Laser
Far Infrared Laser
The Microwave Circuit
Time Synchronization
Experimental Results
Microwave Emission Characteristics
Cavity Scanning Results
Microwave Injection Effects
Heterodyne Frequency Measurement of
Tunability and Stability of the Optically
Pumped Microwave Emission
Discussion of Results
Conclusions
III.
3
3
5
6
7
7
8
12
14
15
15
26
29
37
44
INFRARED-MICROWAVE DOUBLE RESONANCE EXPERIMENTS
IN 1 4 NH 3
46
Introduction
46
Cell Designs
48
Waveguide Double-Resonance Cell
49
Resonator Cell
52
Infrared Microwave Double-Resonance
55
Two Photon Absorption
55
Infrared-Microwave Double-Resonance
Frequencies in 14NH3
56
Experimental Apparatus and Data Acquisition ... 60
Reported IMDR Experimental Techniques
61
CW IMDR Experiment
61
Pulsed IMDR Experiment
66
Experimental Results
70
14
C0 2 R(8) 1Q Molecular System in NH 3
70
CW Results
71
Pulsed Results
76
Conclusions
79
VI
Page
IV.
CONCLUSIONS
14
APPENDIX Al: The Spectroscopy of
NH 3 Around 10
Microns: Notation, Energy Levels,
Selection Rules, Transition Moments,
and Laser-Absorption Matches
Energy Levels
Selection Rules - Single Photon
Selection Rules - Multi-Photon
Transition Moments
Laser-Absorption Matches in 14NH 3
Ground State Inversion Transition
Frequencies
APPENDIX A2: The Infrared Pumped Microwave Emission
Cell, Construction and Calibration
Theory
Cell Design
Comparison of Theoretical and Experimental
Frequency Calibration Curves
Absolute Frequency Calibration Using
Microwave Absorption
Conclusions
Suggested Improvements
83
84
84
90
91
95
105
106
108
108
110
115
118
120
120
LIST OF REFERENCES
123
VITA
127
LIST OF FIGURES
Figure
1.
2.
3.
14
Energy level diagram for
NH 7 relevant to the
12 16
C 0 2 R(6), Q laser transition. The numbered
transitions are identified and the frequencies
(cm" ) given
The microwave circuit
Time synchronization block diagram. The system
allows synchronous firing of the T.E.-C02 laser
with the klystron
Time synchronized oscilloscope traces showing
the infrared pump pulse (C02 R(6), 0 ), the optically pumped far infrared (v0:sR(4,4) at 134.86
-1
-1
cm
and v2:sQ(5,4) at 34.34 cm ) emission, and
the resulting microwave emission attributed to
the G:sQ(5,4) at 22.653 GHz
5. Oscilloscope traces of the G:sQ(5,4) optically
pumped emission
Page
4
9
13
4.
6.
7.
8.
9.
16
17
Cavity scans of the G:sQ(5,4) emission (upper
curves) and microwave absorption calibration
scans (lower curves) for the TE.., (n = 67, 68,
and 60) modes. The cavity scan length is
0.0452 inches per division, ET = 60 mJ,
14
NH 3 pressure at 1.1 torr
Cavity scans at different pressures for the
G:sQ(5,4) emission. ^ T E n n ' n = 67 mode cavity
length is 0.0452 mches/div., frequency scan
is 10.8 MHz/div
25
Oscilloscope traces of observed beat notes at
different cavity length settings. G:sQ(5,4)
emission in TE,, (n=68) mode
31
Graph of observed beat frequency for the
G:sQ(5,4) emission versus micrometer cavity
setting. T E l l n (n=68) mode
32
22
Figure
10.
11.
Page
Real-time spectral analysis of the optically
pumped microwave emission block diagram
34
Oscilloscope traces of real-time spectral
analysis of the optically pumped microwave
emission. Each photo represents the tenth
consecutive pulse
36
12.
Cavity scans of the G:sQ(2,2) emission (upper
curves) and microwave absorption calibration
scans (lower curves). The cavity scan length
is 0.0452 m/div. The cavity emission modes are
TE
lln ( n = 7 6 ' 7 7 / 7 8 )• T h e pump laser is C0 2
R(8), Q , E = 80 mJ,
NH 3 pressure is 1.1 torr .. 38
13.
M-splittmgs of g a m frequency peaks with
pump intensity for gain experiment
GrsQfS^) 1
43
14.
Waveguide double resonance cell
50
15.
Resonator cell
53
16.
Two-photon absorption states of definite
parity
57
17.
CW double resonance experiment
62
18.
CW two-photon absorption detection technique ....
65
19.
Pulsed double-resonance experiment
67
20.
Pulsed two-photon absorption detection
technique
14
NH 3 relevant to C0 2 R(8) 1 Q
69
72
CW two-photon absorption vs. pressure for C0 9
R(8) 1 Q + 34.68 GHz
7
73
CW two-photon absorption vs. infrared laser
power
*
75
Pulsed two-photon absorption vs. cell pressure ..
77
21.
22.
23.
24.
IX
Figure
Page
25.
CO- R(8). Q transmission vs. NH 3 pressure
26.
TPA and transmitted
magnetron frequency
TPA and transmitted
magnetron frequency
27.
Al-1.
A2-1.
A2-2.
A2-3.
A2-4.
microwave power vs.
(waveguide cell)
microwave power vs.
(resonator cell)
Transitions allowed in 14NH.,. A linear
(single photon); B, C - non-linear (two and
three photon)
78
80
81
92
Tunable transmission cavity resonator
operating in the linearly polarized TE,,
mode
...
Tunable plunger
112
113
Measured and calculated TE,, guided wavelength versus frequency for the designed
cavity
116
TE
- n n cavity resonant frequency versus cavity
length/micrometer setting for different n
values, measured (•) and calculated (+) -
117
X
LIST OF TABLES
Table
Page
1.
Calculated infrared-microwave doubleno i g
resonance matches for a
C O , laser
14
X4
NH 3
14
Al. 1.
NH 3 energy levels
Al.2.
A1.3.
Al.4.
A2.1
Infrared-microwave two-photon transitions
considered in this investigation
Match routine results for 14NH., and lasers
around 10 microns
14NH ground state inversion
59
86
94
97
3
transition frequencies (GHz)
107
Microwave absorption calibration measurements
and resulting tube diameter calculations
119
1
I.
INTRODUCTION
NH_ is an interesting molecule.
Tunable maser emission
14
from
NH 3 has been observed and investigated.
Infrared-
microwave double-resonance experiments have also been performed.
The subject of this thesis is a detailed report of these two
projects.
Optical pumping and infrared-microwave double-resonance
experiments are not new for NH,. ~
Demonstration of the
tunability and frequency stability of optically pumped NH 3
emission has not been reported in the literature.
The pre-
viously unreported microwave emission observed in this experiment is tunable and frequency stable, as will be demonstrated.
An attempt to generate tunable far infrared (FIR) emission
via the stimulated hyper-Raman scattering (SHRS) effect has
resulted in the performance of infrared-microwave doubleresonance experiments.
The infrared-microwave double-resonance
experiments performed demonstrate the accuracy of the available
spectroscopy in predicting the resulting microwave frequencies.
Two previously unreported double-resonances have been observed.
The pulsed experiments have been performed at higher
microwave powers than previously reported.
The design and frequency calibration of the optically
pumped microwave cell is considered in detail as the subject
2
of Appendix Al. Two cells designed for double-resonance
experiments are also described.
The spectroscopy of 14NH 3 is the subject of Appendix A2.
14
Tables of
NH 3 energy levels, ground state inversion transi-
tion frequencies, and laser line-NH3 absorption matches for
the various 10 micron lasers are presented.
The linear and
non-linear selection rules and associated spectroscopic notation are discussed in detail.
The results presented here are consistent with the reported
literature and theory.
The techniques and cells developed for
this investigation should prove helpful for the study of
higher order non-linear processes in systems involving infrared
and microwave photons.
II.
TUNABLE MASER EMISSION FROM
C0 2 LASER
14
NH 3 OPTICALLY PUMPED BY A
Introduction
Recently Kim and Coleman
presented results of calculated-
experimental evaluation of some optically pumped NH 3 systems.
Gain and emission at microwave frequencies were reported as a
result of infrared pumping with a TEA-C02 laser in a travelingwave cell.
This paper will report observation of distinctly
different results obtained in a high Q (^3000)microwave cavity
using a lower power TE-C02 laser. Tunable microwave emission,
attributed to a linear process, with excellent frequency stability has been observed and studied.
The energy level diagram relevant to this investigation
is shown in Figure 1.
Details regarding spectroscopic notation
and data are the subject of Appendix Al.
The near-resonant
match (i.e., Av = 544 MHz6) of the C0 2 R(6),Q (v = 966.2504
cm"1) laser line and the
cm
14
NH 3 G -»• v2:sQ(5,4) (v = 966.2660
) absorption has been used to produce optically pumped
far infrared (FIR)and infrared emission.
The FIR transitions
2
labeled 3 and 4 were reported by Fetterman, et al. and
3
Gullberg, et al. An additional FIR transition (i.e., G:sR(5,4)
has been reported,
but is not shown in Figure 1.
The infrared
4
transition labeled 4 was observed by Chang and McGee.
The
G:sQ(5,4) transition (labeled 5) occurs in the microwave range
7
of the frequency spectrum at 22.653 GHz.
The availability of
4
S
o 1204.44- »,a(5 t 4)
Ul
UI
£ M70.I
co 1069 5
v 2 s(5 t 4)
«2s(4,4)
up = R ( 6 ) | Q C 0 2
* 966.2504
'£
>
(9
<r
Ui
z
ui
Ia
3578
357.1
Go (6,4)
G3(6,4)
238.9
238 I
Ga£5,4)
Gs(5,4)
z
o
<r
I 3(2(5,4)
966.2660
2. «2«sR(4,4)
134 8594
3. u2'3Q(5,4)
34 3355
4. sP(6,4)
Figure 1.
5 G-sQ(5,4)
6. G aR(5,4)
7 G 3Q(6,4)
7556
I 18 1776
.7003
8 4 7 3328
Energy level diagram for 14NH 3 releveant to the
12 16
C 0 2 R(6) 1 0 laser transition
The numbered
transitions are identified and the frequencies
(cm" ) given.
5
tunable frequency sources in this range has led to investigation of this transition.
Kim and Coleman
recently reported results of the infra-
red-pumped microwave-probed G:sQ(5,4) system.
Microwave fre-
quency gain/absorption spectra were calculated using a twophoton three-level density matrix theory and compared to the
experimentally obtained transmission spectra.
The characteris-
tic shape of the calculated spectra includes two distinct peaks
near 22.653 GHz and 22.109 GHz. These peaks are designated
the "laser" (linear order in field) peak and the "Raman" (third
order) peak, respectively.
Fine structure due to the AC Stark
splitting of the M-level degeneracy is predicted for both
peaks.
The theoretical predictions for the laser and Raman
frequency peaks are in excellent agreement with the experiment.
The measured values of the gain magnitudes are within a factor
of 3 when compared to the theoretical values. These results
prompted further investigation in an experimental cell that
could provide feedback for possible FIR radiation involved in
yet higher order processes.
Cell Design
The cell used in the Kim experiment is a travelling wave
device that provides no intentional feedback for infrared or
FIR radiation.
This design provides a relatively uniform m frared field to pump the 14NH^ gas.
6
The cell designed for this investigation simultaneously
provides feedback for the FIR and microwaves. The cell is a
right cylindrical microwave transmission cavity operating in
the lowest order linearly polarized TE,, mode. The resulting
FIR mode is the circularly polarized TE,Q i. A similar cell
g
has been used by Willenberg, et al. to produce a two-photon
pumped CW laser. Further details regarding cell design and
calibration are contained in Appendix A2. The infrared radiation is focused through a 1.5 mm coupling hole in a 0.1 inch
thick copper mirror. The resulting infrared pump field is
spatially non-uniform which causes the effective pumping field
to be reduced, and the microwave gam fine structure to become
unresolvable.
The compromise in the resonator cell design suggested by
g
Willenberg
that the circularly polarized FIR modes are not
optimally excited by the linearly polarized pump field and
the higher losses for the FIR are of little concern for the
case of pumping with a pulsed high-power C0 2 laser. The point
of resonantly enhancing and confining the microwave to a minimum volume is well taken, and the primary reason for choosing
this design.
Experimental Apparatus
The experimental apparatus consists of four major parts.
These include:
1) the infrared pump laser; 2) the optically
7
pumped FIR laser; 3) the microwave circuit; and 4) the pulse
synchronization circuit.
Each of these parts will be described
in detail, with the experimental cell design features listed
where they are important.
Infrared Pump Laser
The infrared pump laser is a single longitudinal mode
TE-C02 laser capable of energies up to 150 mJ (stronger lines)
at a 0.5 Hz repetition rate. The cavity consists of a goldplated PTR 10m radius of curvature, 100 lines/mm, master diffraction grating blazed at 10.6 pm and a flat germanium partial
reflector (R=75%).
The total cavity length is about 3m. The
transverse discharge cell pressure was maintained at 300 torr
while the longitudinal discharge g a m cell pressure was at 15
torr.
The premixed laser gas is flowed at a moderate rate and
is a mixture of 12% CO,/ 14% N 2 , and 52% He. Pulsed laser
energies were measured with a GENTEC joule meter.
Far Infrared Laser
The FIR laser resonator is the TE.. n microwave translln
mission cavity resonator.
The resonator consists of a non-
precision 32in long 3/8 in. ID copper tube, a fixed circular
(.355 in. OD) copper mirror with a 1.5 mm hole at its center,
and a translatable circular (.355 m . OD) copper mirror.
only alignment adjustment is the cavity length.
The
The remaining
8
alignment is determined by the machined accuracy individual
components. Operating pressures of the 14NH_ gas range from
0.1 to 20 torr and are easily maintained for many hours in
the cell.
The optically pumped FIR radiation is coupled out the 1.5mm
hole and deflected through a TPX vacuum window, similar to the
3
arrangement used by Gullberg, et al. The signal is detected
with a Si:P detector at 4°K.
The detector is also sensitive
to the infrared pump radiation, causing additional filters to
be required for weak FIR signals.
The stable output of the optically pumped FIR laser and
relatively ineffective cavity tuning indicate the FIR losses
are low and the machined alignment is adequate for these experiments .
The Microwave Circuit
The microwave circuit used in the experiment was
changed several times, but two similar circuits produce the
best results. The basic circuit is shown in Figure 2. The
klystron is an OKI 24V10 driven by an FXR Universal Klystron
Power Supply.
(Beam Voltage = 2000V @ lOma, Reflector Voltage
s -250V, Control Grid = -200V).
Swept frequency operation is
obtained by applying a 0-200 Volt ramp to modulate the reflector voltage. Quasi-fixed frequency operation is obtained by
applying a square wave (0-200V) voltage to the reflector.
K
DS2
DSII
>>
0" 5
Al
Al, A2, A3, A4 - VARIABLE
DC
X
<2>
A2
-K+
-\/>Ar
C2
ATTENUATOR
CI - DIRECT READING CAVITY
Dl
WAVEMETER
C 2 - TUNABLE TRANSMISSION
CAVITY FILTER
C3- TUNABLE INTERACTION CELL
DC- 12 db DIRECTIONAL COUPLER
DSI, DS2 - DOUBLE STUB TUNER
01, D2, 0 3 - IN26B DIODE DETECTOR
I - FERRITE ISOLATOR
K- OKI 24VI0 KLYSTRON
T - SERIES T JUNCTION
Figure 2.
The microwave c i r c u i t .
A3
D2
D3
10
DS1, I, and Al serve to pad the Klystron from the rest of
the mostly high Q circuit. DC, CI, and Dl serve to monitor
the reflected microwave signal from C2 and C3. A2 controls
the level of the microwave radiation. T is an E-PLANE (SERIES)
T junction used to split the microwave radiation.
The purpose
of this junction is to facilitate setting of the microwave frequency.
This will be considered in detail shortly.
C2 is a
high Q (s7000) critically coupled tunable transmission cavity
used as a frequency filter. A3 limits the level of microwave
radiation incident on D2, a 1N26B diode. C3 is the designed
cell (design details are in Appendix A2), a transmission cavity
operating with the lowest order TE,, mode. The measured Q
of the cavity is about 3000 and the maximum coupling is less
than 10%. The amount of coupling may be varied by tuning DS2.
The transmitted microwave level is monitored on D3.
The setting of frequency in this circuit is somewhat
tedious.
The best results are obtained in the following manner.
The klystron is operated in a swept frequency mode with the
desired frequency located at or near the center of the sweep,
as monitored with Dl (the maximum safe signal level from the
1N26B into 1MSJ is 0.5V for < 1 msec pulse, for a duty cycle
less than 0.1). DS2 is adjusted until a decrease of 5% or
more is observed at Dl. Careful examination of the change in
signal level will demonstrate frequency sensitivity.
The
11
change should be maximized at the desired frequency.
now be monitored with a high g a m amplifier.
D3 should
If no signal is
present at D3 tune C3 until a signal is observed, taking care
not to burn up D3. Additional adjustment of DS2 may be used
to maximize the transmitted signal. For most applications this
is all that is required.
For more precise frequency settings, C2 is set to a particular frequency.
(high
The critical coupling of C2 puts a sharp
Q and more coupling) dip in the curve monitored at Dl.
This dip represents maximum transmission through C2 and results
in dissipation of the microwave energy in A3 or D2. Following
the procedure outlined previously a signal is then monitored
at D3. When the cavity C3 is tuned such that the frequency of
cavity C2 are the same, the same sharp dip will be observed in
the transmitted peak at D3. This technique is important for
investigating frequency-time behavior.
The second microwave circuit involves setting up a heterodyne detector.
The klystron operating in the quasi-fixed fre-
quency mode is used as the local oscillator.
The basic circuit
is modified as follows. The directional coupler (DC) is reversed so it splits a fraction of the incident microwave radiation to Dl. The tunable transmission cavity filter is completely removed and A2 is placed between T and DS2. D2 now
serves as the mixer for the heterodyne detector.
This circuit
12
has been used to study microwave injection effects, frequency
tunability, and frequency stability of the generated microwave
signals.
Time Synchronization
Pulsed probe experiments require flexible, reliable
triggering if meaningful data is to be obtained.
The require-
ments for the experiments performed resulted in several different techniques being developed for triggering.
The require-
ment peculiar to these experiments are caused by: 1) low repetition rate (^0.5Hz) firing of the C0 2 laser and 2) time
jitter of the reflector modulation power supply.
The low repetition rate firing of the C0 2 laser is required
to get reliably stable pulses of infrared pumping radiation.
The time jitter problem of the klystron reflector modulation
could be solved by repairing or replacing the power supply,
but was not practical at the time.
The only experiments performed where synchronization is
required were microwave injection and heterodyne detection experiments.
In both of these cases the klystron was operated
in the quasi-fixed frequency mode (i.e., square wave modulation
of the klystron reflector voltage).
The triggering technique
is shown in Figure 3.
The modulation output of the FXR universal klystron power
supply is a 0 to 100V square wave with a period of T,.
(T, is
13
UNIVERSAL KLYSTRON
POWER SUPPLY
• MODULATION OUTPUT
fJ+I00V--
ilNPUT
531 OSCILLOSCOPE
GATE OUTPUT
+ 80V-• n
•TRIGGER INPUT
-J
DELAY TRIGGER
— tf
•TRIGGER OUTPUT
200V T
-t
4TRIGGER INPUT
T.E. C0 2 LASER
Figure 3.
Time synchronization block diagram. The system
allows synchronous f i r i n g of the TE-C02 l a s e r
with the klystron.
14
typically 1 msec).
The square wave is monitored by a
Tektronix 531 oscilloscope with a vertical sensitivity of 20
V/div.
The time base is set at 0.2 sec/div and the scope is
internally triggered by the square wave. The resulting gate
output of the oscilloscope has a period determined by the time
base of the oscilloscope (a 2 second period is typical), and
is used to trigger a four channel delay trigger generator.
The delay trigger ensures the laser trigger pulse occurs during
the fixed frequency part of the klystron reflector modulation.
This technique has produced the simplest synchronization of
the infrared and microwave signals, and is least likely to be
affected by jitter m
the reflector modulation power supply.
The experimental details presented thus far are common
to most of the performed experiments.
Specific details pecu-
liar to a given experiment will be discussed with the experiment.
Experimental Results
Optical pumping of 14NH, with a TE-C02 laser has produced
strong, frequency stable microwave emission.
The results to
be presented include characterization of the microwave emission operating conditions, the emission bandwidth dependence
on pressure, effects of microwave injection, and frequency
stability observations.
15
Microwave Emission Characteristics
Microwave emission attributed to the G:sQ(5,4) inversion transition was observed following optical pumping of
the G -*• v2:sQ(5,4) transition with a TE-C02 laser tuned to
the R(6)- n (v
= 966.2504 cm
) line. The time synchronized
photographs shown in Figure 4 show a typical infrared pump laser
pulse (E=60mJ), the resulting FIR laser pulse, and the observed
microwave emission.
The conditions stated on the figure repre-
sent the near optimum conditions for the microwave emission.
The emission is only observed over a relatively narrow pressure
range from 0.3 to 2.2 torr, with stable amplitude behavior
from 0.5 to 1.8 torr. Additional data is given in Figure 5.
The pulse shapes are observed to change with pressure, becoming
shorter with increased pressure.
In all cases shown the cavity
is set to the length that provides the maximum signal level at
1.13 torr.
To provide further understanding of the microwave emission,
the frequency and emission bandwidth required investigation.
The technique employed to obtain this information is performance of cavity tuning scans of the observed emission.
Cavity Scanning Results
A cavity scan is performed in the following manner.
A motor drive is mounted to the micrometer tuning plunger of
the experimental cell.
The microwave emission is monitored
16
INFRARED
CO2 R(6)io
E *60mJ
ShPat4°K
I Volt/div
FAR-INFRARED
,4
NH 3 atl.l3torr
ShPat4°K
I Volt/div
MICROWAVE
NH3at 1.13 torr
IN26B Into IOOOJ2
50mV/dlv
,4
t = l/tsec/div
Figure 4. Time synchronized oscilloscope traces showing the
infrared pump pulse (C02 R(6) 1Q ), the optically
pumped far infrared (v_:sR(4,4) at 134.86 cm
and
-1
v2:sQ(5,4) at 34.34 cm ) emission, and the resulting
microwave emission attributed to the G:sQ(5,4) at
22.653 GHz.
17
P = .53 torr
y = 20mV/div
y\
- j — i — , — | — i — i — | — i — i — * .
P = .71 torr
y = 20mV/div
H
1
1
H
1
\—|
1
1
1
i
1
!
1
1
1
i
h
P = .91 torr
y = 50mV/div
1—K
t = l/isec/dlv
Figure 5. Oscilloscope traces of the G:sQ(5,4) optically
pumped emission.
18
P= l.lltorr
H—I
1
1
\
\
1—I
1
I-
h
P = 1.31 torr
4
1
1
J
1
1
1
\
1
H
1
1
1
1
1
1
1
1—fc
P = 1.5 torr
t = 1/i.sec/dJv
y= 50mV/dlv
Figure 5.
(con't)
Oscilloscope traces of the G:sQ(5,4) optically
pumped emission.
19
P=I.7I torr
i—I
1—I
•I—I
\
1—\—I—I—!—l-
P= 1.90 torr
C02 MISFIRE
\
1
1
1
1
1—b
P= 1.90 torr
UNSTABLE
SIGNAL
•i—1—1—1—1—1—1—1—1—1-
t = l/isec/div
y =50mV/div
Figure 5. Oscilloscope traces of the G:sQ(5,4) optically
(con't)
pumped emission.
20
P=2.ll torr
UNSTABLE
SIGNAL
y\
H
1
1
1
1
1
1-
^
H
t = l/xsec/djv
y = 50mV/dlv
Figure 5.
(con't)
Oscilloscope traces of the G:sQ(5,4) optically
pumped emission.
21
with a 1N26B diode (D3 in Figure 2) and input to a synchronized PAR Boxcar Integrator.
The gate (i.e., integration win-
dow) of the Boxcar is set at 4 usee to allow for time jitter
of the microwave emission pulse. The integration time of the
Boxcar is set to the minimum value (i.e., 0.01 msec).
The
input is AC coupled to the 10kf2 Boxcar input with a shunt lkfl
resistor to decrease the rise time of the detector. The signal output was used as the y input to an x-y recorder. The
x was set to 100 seconds/inch scan rate. The combination of
the motor drive with the x-y recorder produced 0.0452 inches
of cavity scan per inch of graph paper. The initial and
final cavity setting of the scan are placed on the graph for
reference
result
purposes.
The upper curves in Figure 6 are the
of three consecutive cavity scans.
The lower curves give the absolute frequency calibration
data for the upper curves. These curves are considerably more
difficult to record.
Detection of microwave absorption in a
microwave cavity is a standard spectroscopic technique (see
g
Townes and Schawlow, Section 15.11). The result of microwave
absorption is increased loss at the resonant absorption frequency.
The increased loss in the cavity reduces the Q of
the transmitted microwave signal which is readily detectable
at low pressures. At a pressure of 0.1 torr the homogeneous
microwave absorption Imewidth is about 2.5 MHz, thus a sizeable
TRANSMITTED INTENSITY
{ARBITRARY UNITS)
H
iQ
C
CD
en
o n
0> 0
r t B)
<
* .<+ tr rt
01
rt in
H- 0
( + CO
0
h
H
0
0)
.
a a
en
en
0 0
Dl Hi
CD
a
a
rt
tr
to r t
H
— (D
H
o a
t, "
en
CD to
h IO
O
>U
Ol
N)
O Ui
cH
*k
< 'H (D
a B CD
o w g
tr
H
fl> Hi CO
to
O CO
H H•O
O
rt a
cr
CD
H
w -*
ft
<
H
wo
H< I--•
CO H
-1
H - l :3
O
II
O
aS
II <
a
w
fD
H
CM CD
«-J t o
9 ) £D
cr>
O
oo a
ft
3
C4
2
3
ft. H
O
H <x> o
•fc — c
Z
H
u>
13
H
CD
to
CO
Z3
c
H
CD
oi
3 <
O CD
ft.
CD D l
10
•
10
0
a-rt
CD M
0
a
EMITTED INTENSITY
(ARBITRARY UNITS)
23
transmission dip would be observed if the cavity is scanned
during microwave frequency sweep containing the resonant
frequency. The 14NH microwave absorption transitions are
3
tabulated m Appendix Al (see Ref. 7 for a more complete
listing).
An absolute frequency calibration graph is recorded using
the following method.
The cavity wavemeter (CI in Figure 2)
is set to read the desired absorption frequency.
For the
calibration curves in Figure 6 the desired frequency is 22.653
GHz for the G:sQ(5,4) transition.
The evacuated cavity is then
set near the resonant frequency using the technique previously
described (Section: The Microwave Circuit). Approximately 0.1
torr of 14NH3 gas is added to the cell. The cavity tuning
micrometer is adjusted to locate the dip in transmission
monitored at D3 with a Tektronix 1A7A high gain differential
amplifier having a lMfi input impedance. The signal output of
the 1A7A is then input to channel A of a Tektronix 475 oscilloscope and the lOkfi input of the boxcar integrator. The
gate output of the triggered boxcar is input to Channel B of
the 475 oscilloscope. The vertical display is set to "add"
channel A and channel B.
The gate of the boxcar is chosen to
be much shorter in time than the transmission peak of the
swept frequency klystron output. The time base of the boxcar
is set to allow the gate to be delayed to any point of the
frequency sweep. The cavity is adjusted to the desired starting
24
point and the gate output is adjusted to be at the resulting
transmission peak.
The cavity scan is then initiated, and
manual feedback is used to keep the gate, by varying the
delay of the boxcar, at the transmission peak.
If the result-
ing graph exhibits excessive structure (i.e., noise) increase
the integration time and/or aperture time. Care must be exercised to ensure there is no significant detection lag time
in an accurate frequency calibration graph.
The cavity wavemeter (CI) calibration was verified to be
within 5MHz of all the measured absorption frequencies.
(For
more details, see Appendix A2).
Cavity scans performed at different pressures and fixed
input power provide information regarding emission bandwidth,
saturation, and optimal operating conditions.
Figure 7 shows
the resulting cavity scans at various pressures. The emission
bandwidth numbers are obtained by measuring the full width
at half maximum length on the graph and multiplying it by the
length to frequency conversion factor and scan rate factor
f
Af(FWHM) = Lx 0.0452 x
SCAN
FACTOR
2 " fl
L 2 - Lx
LENGTH
FREQUENCY
FACTOR
Af (FWHM) -i- GHz
L -*• GRAPH PAPER INCHES
(1)
25
G-sQ(5,4)
E =60mJ
>-
or
<
H
s
or
<
CO
z
91torr
UI
rZ
z
o
CO
CO
I
LU
P = .7! torr
49 torr
FREQUENCY(GHz)
J
3Q8I25
Figure 7.
30.6638
L___
CAVITY LENGTH (IN.)
Cavity scans a t d i f f e r e n t pressures for the G:sQ(5,4)
emission. (TElln" n=67 mode) cavity length is 0.0452
mches/div., frequency scan is 10.8 MHz/div.
26
0.0452 •*• (CAVITY SCAN INCHES) / (GRAPH PAPER INCHES)
f2, f1 -»• ABSOLUTE FREQUENCY CALIBRATION POINTS
X
g
X
g
2 n —*=•
2 •*• ABSOLUTE LENGTH AT THE
L_, L.. = n -5^,
FREQUENCY CALIBRATION POINTS.
The accuracy of the measured emission bandwidth is limited
by the emission structure observed in the cavity scans. Cavity
scans at higher 14NH_ pressures have been recorded but the
structure observed makes it nearly impossible to measure the
full-width half-maximum with any certainty.
Microwave Injection Effects
Microwave injection effects have been investigated
with very limited success. The experiments were performed in
hopes of producing information regarding magnitude, time, frequency, and saturation behavior of the emission gain. The difficulties in obtaining this information are multiplied by the
high Q frequency characteristics of the interaction cell. To
obtain reliable experimental data it is believed the cell used
in the Kim
experiment would be better.
Several interesting
qualitative observations have been made and will be presented
here.
The qualitative observations resulting from microwave injection mentioned above include:
1) change of the delay time
for the g a m switched microwave emission, 2) increased pressure
27
range operation of emission, 3) observation of the transmitted
signal frequency beating with observed emission frequency, 4)
increased emission bandwidth, 5) the microwave emission dependence on level of injection indicates the emission is saturated,
and 6) time dependent gain is present m
the microwave emission.
The majority of these observations are not particularly surprising. A feature common to all of the observations is the
dependence on injected microwave signal frequency.
Microwave injection effects are only observed when the
injected microwave signal frequency is between 22.645 and
22.725 GHz. This is not symmetric with respect to G:sQ(5,4)
transition at 22.653 GHz, but is consistent with the data observed in the cavity scans.
Injection of signals at and around
the Raman frequency peak at 22.109 GHz produced no obvious
effects.
The most interesting observations are the indication of
g a m saturation and the time dependent gain of the emission.
The g a m saturation is observed by varying the input level of
injection at various pressures. At pressures less than 1 torr
the delay time decreases, the pulse length increases, but the
peak amplitude remains very constant. The level of injection
to produce these effects is not detectable at the transmission
monitor diode (D3 in Figure 2).
Increasing the level of in-
jection an amount so as to be detectable at D3 (^10db in power)
28
produces the identical result. This was observed near the
peak emission cavity tuning.
Tuning the cavity and microwave
frequency to a point beyond the emission bandwidth half-power
points produced similar results with more (but not much) dependence on signal level injection. The most obvious result
is increased amplitude stability of the microwave emission.
Tuning the cavity and injection frequency 15 MHz beyond the
frequency where emission stops produces no effect.
At pressures greater than 1.5 torr, the emission bandwidth
is observed to be decreasing.
This is attributed to higher
losses. Microwave injection can be used to increase the emission bandwidth.
Stable amplitude pulses of emission can be
observed at frequencies 30 MHz beyond the point where selfoscillation stops. At pressures greater than 3 torr no signals
are observed.
The particularly interesting feature of higher
pressure (i.e., > 1.8 torr) injection experiments is the time
behavior of the emitted signals.
The signal is observed to consist of two distinct pulses
separated in time. The multi-pulse observation varies from
pulse to pulse and cannot be reliably reproduced.
The pulse
occurring first is slightly delayed but overlaps in time with
the infrared pump laser pulse. The later pulse amplitude and
delay is relatively constant and is believed to be the usual
microwave emission. The frequency dependence of the earlier
29
pulse on the injected microwave signal was observed to be
shifted toward the higher frequency end (i.e., ^ 22.69 GHz)
of the emission bandwidth.
The observation of the earlier
pulse is consistent with the results of Kim,
and will be
considered in the discussion of results section.
The injection of single frequency microwave radiation into
the optically pumped 14NH 3 has produced results that could be
studied more effectively using a different cell. An investigation of more importance to this study is tunability and stability of the optically pumped microwave emission.
Heterodyne Frequency Measurements of Tunability and Stability
of the Optically Pumped Microwave Emission
Heterodyne frequency measurements provide a straightforward method of determining if a signal is tunable. The
equipment required for a heterodyne detector
a local oscillator.
are a mixer, and
The OKI klystron is used as the local oscil-
lator and a 1N26B diode is used as the mixer (D2 in Figure 2).
The klystron is operated in the fixed frequency mode and the
cavity is tuned until a beat note is observed on the 200 MHz
bandwidth TEKTRONIX 475 oscilloscope.
Care was taken to avoid
injection effects during the heterodyne measurements. This
was achieved by padding the cell with considerable attenuation
A2 and adjusting DS2 for minimal coupling.
fects were observed at D3.
No injection ef-
30
Some observed beat frequencies are shown in Figure 8.
The oscilloscope photographs were taken at different cavity
micrometer settings (X). An interesting feature observed during
the first 0.5 usee is the obviously different period of the
beat frequency.
Careful examination of this is inconclusive
at best, the beat is reproducible but not understood.
Figure 9
is a slightly different presentation of the data in Figure 8.
The number of beat frequency peaks were counted and divided by
the time interval (neglecting the first 0.5 usee) to obtain an
approximate beat frequency and plotted as a function of cavity
setting.
The resulting curve is linear to withm the accuracy
of the measurement in agreement with the expected result.
(See
Appendix Al for details on linearity of cavity tuning).
The method used above is a quick and dirty technique to
demonstrate tunability.
Stability of the microwave emission
was examined and surprisingly stable reproducible beats were
observed.
The stability of the beat note is surprising because
no attempt has been made to frequency stabilize the experiment.
Recently, Fetterman, et al.
performed "real-time spectral
analysis for FIR laser pulses" using a surface acoustic wave
(SAW) reflective array compressor (RAC) device. Bob Miller,
a member of this laboratory, is currently working on a similar
project.
The optically pumped microwave emission observed
and the heterodyne detection system already being used suggested
X = .705
X = .7II
kmkiMi
X= 685
X - .695
X = .675
t = .5/xsec/div
G=sQ(5,4)
,4
X = .655
NH3 at .54 torr
LOCAL OSCILLATOR FREQUENCY 22.670 GHz
Figure 8. Oscilloscope trances of observed beat notes at different cavity length settings. G:sQ(5,4) emission
in TE,, (n=68) mode.
32
.66
.67
68
.69
70
MICROMETER SETTING (INCHES)
f0 = 22.673 GHz
P = 534 TORR
Figure 9. Graph of observed beat frequency for the G:sQ(5,4)
emission versus micrometer cavity setting. TE
lln
(n=68) mode.
33
easy implementation of a spectral analysis system.
The system
used was set up with the help of Bob Miller.
A block diagram of the resulting real-time spectral analysis data acquisition system is shown in Figure 10. The
specifications for the Raytheon RAC device are:
fo = 60 MHz
Bandwidth = 6 MHz = B
Dispersion Time = 20 usee = T
Number of Resolvable Frequencies 11
Frequency Resolution 0.55 MHz
Minimum Pulse Width
0.667 ysec
Maximum Pulse Width
1.217 ysec
The RAC device acts as a linearly dispersive delay line with
the center frequency (f ) being delayed T/2. The bandwidth
divided by the dispersion time (i.e., B/T) gives the dispersion
rate (= 0.3 MHz/ysec).
Thus if one were to apply a single fre-
quency pulse with the correct frequency (f ) and pulse width
(tm m < tp < tmax ) , the Fresnel transform of the pulse
will be
e
the output of the RAC device.
11, 12 and 13).
(For further details, see Refs.
The advantage of using the RAC device to meas-
ure the frequency tuning is the linearity of the delay with
respect to frequency.
Frequency stability measurements become
much more quantitative than the observations mentioned previously.
34
OPTICALLY PUMPED MICROWAVE
NAVAIR
D2 ( -+3-
H P FILTER
^
50fl
>
SYSTRON-DONNER „
PULSER • - 0 U T + « - r
MODULATOR
ZAD
IW
OKI24VI0
RAYTHEON
60 MHz
SAW RAC
HP 462A
••
AMPLIFIER
2 0 - 4 0 db «.
GAIN
^TRIGGER INPUT
466 SCOPE
INPUT*
Figure 10.
Real-time s p e c t r a l analysis of the o p t i c a l l y pumped
microwave emission block diagram.
35
Cavity tuning of the optically pumped emission demonstrated a linear dependence of frequency with length in agreement with Figure 9 and Appendix A2. The study of cavity tuning
length actually requires a much wider bandwidth RAC device
than the 6 MHz bandwidth device used.
A 6 MHz bandwidth cor-
responds to tuning the cavity length approximately 0.025 inches
for the 22.66 GHz TE,,
(n=67) mode. The measured change in
cavity length was 0.028 inches, in good agreement with expected
value.
The most impressive results obtained with the 60 MHz RAC
device are shown in Figure
11. The microwave emission was
gated by a 1 ysec pulse from the triggered SYSTRON-DONNER pulser
applied to the moduator in Figure 10. The gate was delayed
past the initial 0.5 ysec of the pulse. The photographs shown
are the amplified output of the 60 MHz RAC with a local oscillator frequency of 22.605 GHz and a fixed cavity length setting
(micrometer reading = 0.305).
Each photograph in Figure 11 is
a recording of the tenth consecutive pulse.
recorded exhibit the same stability.
The pulses not
The shape of the curves
resembles the (sinx)/x frequency distribution expected for
square wave pulse input.13 The results indicate frequency
stability of the optically pumped microwave emission, to well
within 1.5 MHz over 60 consecutive shots. This is indeed impressive because no attempt to frequency stabilize any part of
36
Y = .l V/div
G sQ(5,4)
t= 5/tsec/div- 1.5 MHz/div
,4
NH 3 at .96 torr
LOCAL OSCILLATOR FREQUENCY 22.605 GHz
X=.305
Figure 11.
Oscilloscope traces of real-time spectral analysis
of the optically pumped microwave emission. Each
photo represents the tenth consecutive pulse.
37
the system, including the OKI 24V10 klystron was made. Changing the pump laser power and cell pressure affected the envelope of the resulting real-time spectra, but did not change
the observed beat frequency.
It was initially hoped that the index of refraction effects could be measured with the RAC device, however, the
index changes are so small that no frequency shift larger than
the possible experimental error could be observed over the
emission operating pressure range. A much longer cell length
may be required to obtain a measurable frequency shift.
The surface acoustic wave RAC device has been used to
demonstrate the high level of frequency stability of the optically pumped microwave emission.
The use of a wider bandwidth
(i.e., 100 MHz bandwidth) RAC would provide more detailed information regarding the emission bandwidth and linearity in
cavity tuning of the pumped emission.
The power of real-time
spectral analysis using this technique cannot be overrated.
Discussion of Results
The results observed in these experiments, present an
alternative method to generate NH, maser emission.
Similar
results have been obtained for the G:sQ(2,2) transition pumped
by the C0 2 R(8), Q laser line. Cavity scans for this system are
shown in Figure 12. The emission is attributed to the same
type of process.
TRANSMITTED INTENSITY
(ARBITRARY UNITS)
H
iQ
C
H
(D
W N
l*
Ul-- ( 0
.b
(0
-J
H
•
„.
ro
Hen
n
o
to
»
00
H r
3 H
^ 0
0. 3
H
tn
Hi
<!
H
rt
u
2
j»
>s
a
H 0
H3 CT £U
6
s 3
U)
o o ro
e
hi
n>
w 1-ru
8 I o>
< o^<
• &H> CO
<—* 5 " H 3
H
0 U> in
o
rt
^ 0U H0 0Hi
W < 3
IT*
H
rt
rt en 3*
II ^< 0 (D
ft)
00 (D 3 Q
o
gH en tn
••
tn - > 0
O H—.
«* C
H 0 l>H 0 S "
» 3 I D Mw
H
z
•0
h
(D
en
in
w
— N
<x
A_
OiC 3
(D H H
CO < to
(D CO
CO H -
0
u
to
i*
—
w
2
T
w to
< o_ - i 5 l
c 0>
ro
tj — 0
(0 • 3
H
~
c
H HH «
3"T3
en H
®*V
3
(D
H
0 n
~
f
t
>
•H 3 < 0
II H C
ft ^J r t K
0
CM>< <
(D
n •^i cn in
n •slQw
* , JU
-J 3 ft)
oo
3
»
w|_.Qi
•
(D
3 3
lQ H
H rt 0
3*3*H
CD
0
H 5
tJ cn ft)
3
<
3 oct>
•0 •
o ft>
H
*".cr
cn cn
pi
CO to 0
ro
h
H
t>
Of
'
W
S
w
p
ro
w
o>
"J
~
n
o
Si-*
o
z
C5
o
EMITTED INTENSITY
(ARBITRARY UNITS)
51
39
The microwave emission is attributed to a linear cascade
lasing process. The data supporting this includes: 1) the
distinctive time delay between the pump laser and the observed
emission; 2) the observed emission bandwidth; 3) the frequency
stability of the microwave emission; and 4) decoupling of the
microwave emission from the FIR emission.
The distinctive delay associated with the initiation of
the gain switched microwave emission is a characteristic of
cascade emission.
The delay has a very weak dependence on
cell pressure as can be observed in the time synchronized
photos (traced) of Figure 5.
The frequency stability of the
emission is partially caused by the decoupling of the infrared
pump laser pulse. The emission center frequency would exhibit
dependence on the infrared pump power. Observations made during
several different power runs indicate decreased amplitude stability by lowering the pump power. The location of the emission center frequency does not change.
The microwave emission and FIR emission were simultaneously
monitored during several cavity scans. No obvious correlation
was observed between the microwave and FIR.
The observed de-
coupling further supports the linearity of the emission.
The bandwidth of the observed microwave emission is
saturated and homogeneously broadened.
The CW two-photon pumped
Q
FIR laser observed by Willenberg
exhibits the Doppler
40
broadened width of the two-photon pump transition.
The same
behavior would be expected for a single photon optically
pumped emission. No evidence of non-linear behavior has been
observed for the G:sQ(5,4) emission so the emission is believed to be linear.
The linear emission is different than that observed by
Kim and Coleman, but their results provide additional understanding needed to explain some of the observed results.
The microwave emission can be theoretically modeled using
a single-photon two-level density matrix formalism.
The den-
sity matrix theory treats the coherence between the energy
levels coupled by the maser field.
For a comprehensive ex-
planation of density matrix theory, see Refs. 14, 15 and 16.
The detailed derivation of the complex susceptibility is outlined m
Section 8.1 of Ref. 16. Two differences in the
G:sQ(5,4) system and the single-photon two-level calculation
assumptions should be noted.
The treatment in reference 16
assumes only two levels are involved in the interaction, the
experimental system actually consists of degenerate energy
levels.
There are eleven degenerate M values (i.e., 2J+1, J=5)
for the upper and lower levels. The presence of the G:sQ(4,3)
transition may also effect the two level assumption. The remaining assumptions are consistent with the experiment. The
imaginary and real parts of the susceptibility are given in
Equations (2) and (3).
41
2
n,
u T-AN rt
,
2
1
.
O
X (w) = — g
o
2~5
1 + (lo-io P T j
y 2 T 2 AN Q
,_.
(2)
5
+
4n
T
2T1
(co-%)T2
X (w) = -r*
T~l
(3)
5
o
AN
T
2
=
T
l
1 + (co-to ) ^ T , + 4£rT~T,
o
z
z i
- Population difference a t zero f i e l d
=
i-326*10"8
sec-torr
J2 = uE /2il = Precession (Rabi) frequency
The absorption/gam can be calculated from the imaginary
part of the susceptibility and the index of refraction may be
calculated from the real part. The importance of equations (2)
and (3) is the resulting Lorentzian line shape of the absorption/gain. This width is 23 MHz/torr for 14NH 3 ground state
inversion transitions. A consequence of saturation is broadening of the Lorentzian line shape given by Equation (4)
^SAT =A V
+
y2E2T9T,
°22
<4>
The problem with using this model is that the frequency shift
of the microwave emission observed in the cavity scans is not
explained.
The emission frequency shift observed in the cavity scans
is believed to be the result of the time dependent g a m observed during the microwave injection experiments.
The
42
results of Kim indicate strong g a m occurring during the
presence of the infrared pump pulse. Figure 13 (reproduced
from Kim's paper) gives the AC Stark shifted laser and Raman
frequencies as a function of the laser power density. The
laser frequency peak is always shifted to a higher frequency
for the G:sQ(5,4) system.
If the emission level occurring
during the laser pulse was lower than the noise level (i.e.,
non-detectable), injection locking of the observed emission by
the non-detectable emission could occur. The large delay and
gain switched pulse shape of the microwave emission would tend
to disprove this explanation.
The cavity scans produce emission bandwidth greater than
the homogeneous Imewidths. Equation (4) can be used to determine the Rabi frequency and hence relative field strengths
of the microwave emission.
PRESS (torr)
AV
H(MHz)
The calculation results are:
^SAT**1^
f
(MHz)
ff
AV
H
.49
11.27
19.6
31.77
2.819
.71
16.33
24.8
38.56
2.36
.91
20.93
31.7
49.23
2.35
Using the measured emission linewidth at 0.91 torr and neglecting the M degeneracy, a circulating power density of 0.54
2
watts/cm
is calculated.
The ratio of the Rabi to the homo-
geneous linewidth is the measure of saturation.
The results
43
•IMh5
4
3
2
FIRST ORDER (LASER) u 2| * 22.653 GHz
y
Pius
R(6)|0 -lOOkW/crn
LASER RAMAN
CAL
22.72 22.05 GHz
EXP
22.72 22 03
THIRD ORDER (RAMAN) u2| - Av = 22 109 GHz
2
3
IMI = 5
100
200
300
400
PEAK C0 2 R(6),0 PUMP INTENSITY \N kW/cm 2
Up II u s
Figure 1 3 . M-splittings of gain frequency peaks with pump
intensity for gain experiment G : s Q ( 5 / 4 ) . 1
44
of microwave injection experiments concerned with saturation
agree with the cavity tuning results in that the degree of
saturation decreases with increasing pressure. The saturation,
tunability, and frequency stability suggest the observed emission may be a valuable tool as a source to study other processes in 14NH.,.
Conclusions
Microwave emission attributed to a linear cascade effect
has been observed from 14NH 3 optically pumped by a TE-C02
laser.
The emission is tunable over the saturated homogeneous
linewidth (i.e., > 24 MHz/torr) and exhibits a high degree of
frequency stability.
The output power density determined from
2
the saturated emission bandwidth is 0.54 watts/cm
torr.
at 0.91
The observed emissions are assigned to G:sQ(5,4) and
G:sQ(2,2) ground state inversion transitions.
There are many other candidate systems for which this type
of microwave emission should be observed. The criterion for
a candidate system is: 1) a laser-14NH, absorption match with
a frequency mismatch less than 1.5 GHz, 2) the lower level of
the NH3 absorption must be the symmetric energy level of the
inversion doublet, 3) the inversion doublet exists (i.e., K^O).
The obvious candidates are the laser-absorption matches that
produce optically pumped FIR emission. Appendix Al includes
45
a "match list" to be examined for possible candidates. Only
14
NH 3 has been considered in this investigation, further study
should produce similar results for other molecules with inversion symmetry.
46
III.
INFRARED-MICROWAVE DOUBLE RESONANCE EXPERIMENTS IN
14
NH 3
Introduction
The generation of tunable far infrared (FIR) radiation is
a difficult problem.
Limited success has been obtained using
high power C0 2 lasers to generate Raman type emission. ' ' 1 7 ' 1 8
Tunability in the optically pumped emission is achieved by frequency tuning of the pump source. The limited tuning bandwidth of a C0 2 laser results in generated off-resonant FIR
emission with a very narrow tuning range. The strong pumping
fields required to pump the non-linear Raman effect also result in A.C. Stark shifting of the emission frequencies. For
these reasons, the realization of a frequency stable, widely
tunable FIR source using an optically pumped Raman effect
seems impractical.
An alternative method for generating the desired tunable
FIR emission is called Stimulated Hyper-Raman Scattering
(SHRS).19 In this fifth order non-linear field process, two
photons are used to pump the resulting FIR emission.
If one
(or both) of the photons is tunable, the resulting FIR emission will demonstrate the tunable bandwidth of the pump
photon.
The higher order A.C. Stark effects are minimal and
20 21 22
a practical source could be produced. ' '
This possibility prompted investigation of infrared-microwave pumped
SHRS.
47
Tunable high power microwave sources (magnetrons) are
commercially available.
The combination of a fixed frequency
C0 2 laser and a tunable magnetron satisfies the SHRS twophoton pump requirement.
The strategy planned for the investi-
gation was divided into four sequential steps: 1) cell design,
2) mfrared-microwave double-resonance (IMDR) experiments, 3)
mfrared-microwave pumped FIR emission, and 4) demonstration
of tunable FIR from mfrared-microwave pumped SHRS.
Satisfactory progress was made in the cell design and
infrared-microwave double-resonance experiments. The pursuit
of infrared-microwave pumped FIR did not meet with the same
success.
The reason attributed with the lack of success for
this experiment is the choice of the wrong molecular system.
The molecular systems chosen for study were dictated by the
availability of microwave components. 14NH 3 was chosen because
of the amount of spectroscopic data available and in particular,
g
the reported mfrared-microwave experiments of Freund and Oka.
The severe limitation imposed by the limited microwave component availability resulted m choice of non-resonantly enhanced
three-photon transitions for SHRS.
Calculations indicate
strong dependence of the resulting SHRS gain on resonance eno
hancement. Willenberg, et al. have produced two-photon
pumped CW FIR emission by choosing a resonantly enhanced system.
This system has an intermediate mismatch of only 133 MHz
48
(i.e., 2.5 Doppler haIf-width).
A typical intermediate mis-
match of 10.8 GHz was common for the systems tried.
The pos-
sibility of success with such large intermediate mismatches
appears unlikely.
The cell designs and resulting mfrared-microwave double
resonance experiments do deserve attention.
The following
material presents the results of this work.
Two working cells
were constructed and used to perform the IMDR experiments.
A
waveguide double-resonance cell, similar to the one presented,
was eventially used by Kim
to probe microwave absorption/gain
in optically pumped NH-, experiments.
reproduced some of the Freund-Oka
The IMDR experiments
results and also produced
two new double-resonance observations.
Cell Designs
A cell design for a SHRS experiment requires the following:
1)
Collinear propagation of infrared, far infrared and
microwave radiations.
2)
Feedback for the FIR.
The non-linear experiments pro-
posed may have relatively small g a m s and hence a FIR resonator
may be required for the desirable emissions.
-4
3)
The cell will be operated at a modest vacuum
(^10
torr) .
4)
quencies.
The molecular system dictates useable microwave fre-
49
The availability of 35 GHz equipment has led to a search
for a molecular system to accommodate this frequency.
The
search for such a molecular system will be described following
discussion of the cell designs used thus far.
Experimental
results using the cells will be presented with the infraredmicrowave double-resonance (IMDR) experiments.
Waveguide Double-Resonance Cell
The waveguide double-resonance cell does not permit
FIR feedback but was constructed to perform IMDR experiments.
A schematic diagram of the cell is shown in Figure 14.
The operating principles of the waveguide double-resonance
cell are straightforward.
The microwaves propagate from the
source in a TE,Q waveguide mode and encounter the 45° H-plane
junction with a similar waveguide.
The direction angle stated
is relative to the propagation direction.
The five wires are
in line with one wall of the inital waveguide and serve to
short the 135° direction path.
The incident radiation is
therefore guided around the 45° corner at relatively low loss
(<0.5db).
A straight path is obtained for the infrared radia-
tion, with the wires serving to scatter some of the incident
infrared radiation.
This design allows for easily obtainable
collinear propagation of the microwaves and infrared.
The
wires present a sizeable diffraction loss for the FIR and are
not desirable in a resonator design.
( 7 ) fiVl POLARIZATION
|
TO SOURCE
F i g u r e 14.
Waveguide d o u b l e - r e s o n a n c e c e l l ,
IR POLARIZATION
TO LOAD
o
51
A microwave Klystron (OKI-35V10) with a slotted line was
attached to the "to source" port and a matched load to the
"to load" port. The measured VSWR was <1.5 in the 33 to 35
GHz range. This was improved (VSWR <1.1) by addition of a
stub tuner before the cell.
The transmission spectrum of the
cell was found to be relatively structureless.
The CO- laser radiation is passed through one of the KC1
windows and detected through the other.
The infrared trans-
mission through the evacuated cell was measured to be about
75%.
The transmission loss is caused by the wires in the wave-
guide scattering the incident radiation, and the normal incidence to the KC1 windows. The transmission loss should also
be a function of the input focusing optics, but has not been
examined.
Microwave radiation leakage through the KC1 windows was
measured to be 30 db down from the incident radiation. Because of the technique used to measure it, the actual leakage
may be much larger.
The desirable characteristics of the cell include: 1)
linear polarization of microwave radiation, 2) cell length is
easily variable, (the length of the cell is determined by a
straight section of the waveguide between the end pieces.
The flanges are not shown in Figure 14 , 3) high vacuum cell
easily obtainable, and 4) collmear infrared-microwave propagation easily obtained.
52
The best IMDR experimental results have been achieved
using the waveguide double-resonance cell, as will be seen later.
Resonator Cell
The resonator cell design offers collinear propagation
of the three radiations with an 8 millimeter free aperture for
the FIR resonator.
This design is considerably more complex
than the waveguide double-resonance cell. A somewhat similar
23
design has been made by H. Jones
for IMDR experiments.
A diagram for half of the cell is shown in Figure 15. The
other half is nearly identical to the half shown. A salt window mounted at Brewster's angle replaces the silicon vacuum
window (B), and vacuum connections are made'to the other mirror mount.
The microwave radiation is introduced through standard
35 GHz waveguide (RG-96) from either port and travels through
a transition to standard 10 GHz waveguide (RG-52).
The change
in boundary conditions causes higher order modes of the 35
GHz to be excited.
The radiation makes a bend and encounters
the 10mm Teflon tubing, again causing higher order modes to
be excited.
A transition from the rectangular waveguide
(0.4"x0.9") to circular (0.4" ID) waveguide forces the microwaves to propagate down the axis of the Teflon tube. Once
the radiation is throug the circular guide the reverse process occurs and the commercial 35 GHz waveguide is terminated
B—
Gold coated hole-coupled glass muror
Silicon vacuum window
Evacuated translatable mirror mount
assembly
TFE Teflon tube 10mm 0D»lmm rfall
Machined brass waveguide u-plane bend
(3 in radiub) with 10 mm hole. IWaveguide dimensions -O.J"NO 9")
Commercial waveguide flanges
Transition - <(0.4"'O 9" to 0.4"ID»*
long)
Circular waveguide - 0.4" ID
Transition - |(0.4"«0.9" to 0. 14'
0 28")<3" long
commercial waveguide (RG-28)
Figure 15. Resonator cell.
Ol
54
with a matched load.
In this configuration the cell is called
the travelling wave cell.
A short can replace the matched
termination to produce a standing wave configuration.
The microwave circuit characteristics of the resonator
cell exhibits strong dependence on microwave frequency. The
measured VSWR was typically in the range of 1.8 to 3. For a
given frequency, a stub tuner was used to reduce the VSWR to
a 1.4 minimum value. The resulting structured microwave power
transmission spectrum has a significant effect on the observed
double resonance experiments as will be discussed later.
The infrared radiation is focused through the salt window and hole-coupled gold-coated glass mirror. The focal
length of the focusinq mirror is approximately equal to the
length of the cell.
The majority of the infrared radiation
overlaps the modes of the FIR resonator formed by the two holecoupled mirrors. The cell has been used without hole-coupled
mirrors to perform IMDR experiments.
The FIR resonator consists of two hole-coupled mirrors
and the 8mm ID aperture dielectric (Teflon) tube. The separation and curvature of the mirrors form a stable open resonator.
The dielectric tube will force the FIR to operate in a waveguide mode.
Several points implemented in the cell design include:
1)
The Teflon tubing was chosen because of its low loss
55
at microwave frequencies.
Polyflo (Eastman Kodak trademark)
tubing may work but has not been tried.
Pyrex tubing has been
used and rejected because of high-loss measured and ease of
breakage.
2)
The transitions were chosen to minimize the reflections
from the Teflon tube and large aperture holes.
3)
The Teflon tube and the two mirror mounts are the only
components requiring evacuation.
4)
A silicon vacuum window is used because of its broad-
band FIR transmission characteristics.
The resonator cell offers the following features:
1)
Collinear propagation of IR, IFR, and microwave radia-
tions,
2)
Feedback for the FIR,
3)
Vacuum readily attainable,
Infrared Microwave Double Resonance
Infrared-microwave double-resonance is a two-photon absorption experiment.
This section will discuss:
general two-
photon theoretical results, application of results to
and calculation of
14
NH3,
14
N H 3 two-photon transitions.
Two-Photon Absorption
Two-photon absorption is a non-linear effect.
An
energy level diagram for two-photon absorption is shown in
56
g
Figure 16. The diagram depicts cases using infrared and
microwave photons, v. and v , to make a transition from state
1 to state 3. All the states are assumed to have definite
parity and transitions must be allowed from state 1 to state
2 and state 2 to state 3 (i.e., u , u ^ 0) . The transition
probability for this process is given in Equation (5).
<l|VEm|2><2|yv.Ej3>
|M 2 | 2
2h2Av
(5)
Equation (5) assumes the two-photon process is resonant (i.e.,
6v = 0 in Figure 1).
moment".
M„ is called the "two-photon transition
The important points of Equation (5) indicate linear
dependence of transition probability on laser and microwave
powers and inverse quadratic behavior with respect to intermediate mismatch, Av. Appendix Al presents further details
regarding application of transition moments to 14NH 3 .
Infrared-Microwave Double-Resonance Frequencies in 14NH 3
The calculation of two-photon transition frequencies
was accomplished by taking the difference between the final
and initial states defined by the selection rules (see Appendix
Al). The energy levels listed in a Ph.D. thesis by J. Curtis 25
have been used for the calculation.
list of the 14NH energy levels).
3
(Appendix Al contains a
57
Mi
jAv
U
M
#L
-t.
TAu
Figure 16. Two-photon absorption states of definite parity.
58
The transition frequencies were sorted in ascending order
and merged with the previously calculated CO- laser fre26
quencies.
The result is a list of two-photon absorption
frequencies and CO- laser frequencies in ascending order.
Taking the difference/sum between/of C0 2 laser and two-photon
transition frequencies yields the required microwave frequency
for a double-resonance condition.
The microwave frequency range of interest is from 33 to 36
GHz.
This further restriction filters many of the possible
matches, until fourteen candidate systems are found.
The re-
sulting molecular systems are listed in Table 1.
Table 1 contains: the calculated microwave frequency for
a double-resonance match (the values in parentheses are the
g
experimentally observed values of Freund and Oka),
the cor-
responding two-photon transition identification (see Appendix
Al for detailed explanation of the spectroscopic notation),
the associated inversion transition frequency,
responding
C
0 2 laser transition.
and the cor-
The microwave fre-
quencies with a M superscript have been observed in this investigation.
The intermediate frequency mismatch (Av in Fig-
ure 16) may be determined by subtracting the inversion frequency
from the microwave double-resonance match frequency.
Evaluation of the two-photon transition moments (Equation (5)) gives a quantitative comparison of the expected
59
Table 1.
Calculated infrared-microwave double-resonance
matches for a
C 0, laser in
NH 3 .
12
C 1 6 0 2 LASER
TRANSITION
MICROWAVE
MATCH FREQ(GHz)
TRANSITION
INVERSION
FREQ(GHz)
M
33.050
aQR(2,l)
23.098
P(40) 9
33.069
sQQ(ll,4)
11.947
P(24) 1 Q
33.593
sQQ(12,2)
9.272
P(20) 1(J
33.644
aQQ(6,3)
19.757
R(6)1Q
33.675
sQQ(12,5)
11.133
P(34) 1 Q
33.695M*
(33.674)
SQQ(2,2)
23.723
P(34) 1 Q
34.038
M
34.139
sQR(7,6)
22.925
R(38) 9
aQR(2,2)
23.723
P(40) 9
34.364
aQR(4,l)
21.134
R(2) 9
34.478 M A
(34.501)
sQQ(6,6)
25.056
P(38) 1 Q
34.645 M *
(34.677)
aQQ(3,3)
23.870
R(8)1Q
35.453
aQR(4,2)
21.703
R(2) 9
35.676
SQR(6,2)
18.885
R(14) g
35.899
aQR(3,l)
22.235
P(20) 9
*
6
Observed by Freund and Oka.
M,Observed this investigation.
60
relative two-photon absorption signal levels. An interesting
feature of Equation (5) is the J and K dependence of the direction-cosine matrix elements (i.e., J and K corrections to the
corresponding dipole moment; namely, u = |y | 6 . . „ . ,
K, M i ) .
For J equal K the direction-cosine matrix element is the maximum value for the inversion transitions.
The direction-cosine
matrix element is always less than 1, and is a quadratic term
in the numerator of Equation (5). Hence for J equal K the
maximum double-resonance signal is expected.
A further dis-
advantage of J^K transitions is the reduced inversion transition frequency which results in larger values of intermediate
mismatch (Av) frequency, making the two-photon transition probability even smaller.
Having determined the microwave match frequencies and expected relative signal levels, the experimental apparatus and
data acquisition techniques will be discussed in detail.
Experimental Apparatus and Data Acquisition
Two types of mfrared-microwave double-resonance (IMDR)
experiments have been performed.
The microwave pumping
source used determines the type of experiment. A CW Elliott
Brothers B579 Klystron and a pulsed SFD 330 tunable magnetron
have been used.
The data acquisition techniques developed for
the experiments will also be detailed.
61
Reported IMDR Experimental Techniques
The design of an experiment to produce tunable FIR
using SHRS is not optimal for performing IMDR experiments.
6 23
The usual IMDR experiment '
consists of a cell containing
the gas of interest through which the microwaves and infrared
are co-propagated.
The microwaves are usually frequency modu-
lated with a square wave while the microwave center frequency
is slowly changed.
A double-resonance signal appears when
the square wave modulation is detected on the transmitted infrared signal. .The modulated infrared signal is easily detected using phase-sensitive detection (PSD) techniques and
usually produces a signal derivative with respect to frequency.
Tremendous improvement of the observed signal-to-noise ratios
are achieved by insertion of the cell into the infrared laser
23
cavity.
The IMDR technique allows for measurement of infrared transition frequencies with microwave frequency measurement accuracy.
CW IMDR Experiment
The CW IMDR experiment was performed with a fixed
frequency Elliott Brothers Klystron.
Only one molecular
system could be studied because of the non-tunability of
the Klystron.
ure 17.
The experimental apparatus is shown in Fig-
62
STABILIZER
I
13 Hz
CHOPPER!
CW co 2
2500 Hz
MOO
D
R
C
E
L
L
8579
KLYS.
MWAVE
LOAD
POWER
METER
•PSD
AT 2 5 0 0 Hz
-REF
F i g u r e 17.
TOUT
/
xz
IN
•
CW double-resonance experiment.
HgCdTe
AT 77°K
63
The CO- laser is approximately 1 meter long, with KC1
Brewster windows. The optics for the laser consist of a 10
meter radius of curvature PTR master grating, blazed at 10.6
microns, ruled in Si and gold-plated, and an 80% Ge partial
reflector.
The cavity is dither stabilized using a Lansing
stabilizer and PZT with a pyroelectric detector used in the
feedback loop. The laser operated at l*s watts on the R(8)
10 micron transition of C0 2 with a flowing gas mix, 1:2:4
(C02:N2:He), at a pressure of 20 torr. The output of the
laser was amplitude modulated by a 13 Hz mechanical chopper.
The laser radiation is passed through the waveguide double resonance cell, and is split by a silicon beam splitter.
Part
of the beam is monitored with a Scientech Power meter and the
rest is incident on the HgCdTe (@77°K) detector.
The microwave source, a B579 Klystron, 9.3 watts at 36.68
GHz, is passed through an isolator and tuning stub into the
cell. After the cell a 30 db coupler in connection with an
1N53 diode are used to monitor the transmitted microwave power.
The remaining microwave power is dissipated in a high power
forced-air cooled dummy load. The microwave radiation is amplitude modulated (m-0.14) at 2500 Hz. The tube is also being
frequency modulated to an unknown degree. This was observed
by beating the amplitude modulated B579 Klystron with a CW
OKI 35V10 Klystron in a 1N53 diode. A periodic chirp was
64
observed while amplitude modulation was applied.
The magni-
tude of the frequency chirp was not measured.
The detection equipment used consists of the HgCdTe detector, two lock-m amplifiers (PAR-HR8) and an oscilloscope
or chart recorder.
The two-photon absorption detection technique is shown in
Figure 18. The time scale shown is arbitrary, as is the relative phase between the infrared and microwave intensities.
The laser intensity is amplitude modulated with a 50% duty
cycle at f_R(=l/TIR)Hz.
similarly at f
The microwave intensity is modulated
(=1/T,,,,) Hz.
The modulation index of the micro-
wave is shown to be less than 1, which is the experimentally
observed case for the Elliott Brothers Klystron.
If two-photon
absorption
occurs then modulation at fuw will be observed on
r
the infrared signal. Phase sensitive detection (PSD) at f
is used to detect the RMS aplitude of this modulation.
If the
double resonance signal is sufficiently large, the output of
the lock-m amplifier may be observed directly on an oscilloscope.
The sensitivity can be greatly increased by connection
of the mixer output of the PSD to a second lock-in amplifier
at f-rp. Care must be taken to avoid saturating the input of
the second lock-in amplifier.
This is essential to obtaining
meaningful data. The technique described is a double modulation technique.
65
LASER INTENSITY
T, R
MICROWAVE INTENSITY
-j |*-T^w
HgCdTe DETECTOR OUTPUT
JTJUinjTJlJTJTiTJl^
^1
AMM?UTAJDEN
JUWU1
LOCK IN [AT f » y—]OUTPUT
MODULATION
AMPLITUDE (V R M § )
TIME (ARBITRARY UNITS)
Figure 18.
CW two-photon absorption detection technique.
66
The apparatus and detection technique described have been
used to obtain CW mfrared-microwave double-resonance data.
The data will be discussed in the Experimental Results section.
Pulsed IMDR Experiment
The pulsed experiment was performed with a tunable
SFD 330 magnetron and a CW C0 2 laser.
The tunable feature of
the magnetron adds tremendous flexibility to the choice of a
suitable molecular system.
The experimental apparatus is
shown in Figure 19.
The C0 2 laser, chopper, detector, and power meter have
already been described in the previous section.
For the pulsed
experiment, the pumping source and double-resonance cell are
changed.
The SFD-330 magnetron is rated at 30-40 kwatts in the 33.5
to 35.5 GHz frequency range. The achieved power is approximately
30 kwatts in the 33.0 to 35.0 GHz range. The spectral purity
of the magnetron is unknown.
The magnetron is operated in a
pulsed mode at 300-400 Hz repetition rate with a 1 ysec pulse
width.
The power is sufficiently strong to "break down" the
inside of the waveguide with a VSWR greater than 2.0. The remainder of the microwave circuit is essentially the same as
before, with the exception of the transmitted power monitor
branch. An addition of 40 db of variable attenuation is added
before the 1N53 diode to prevent destruction of the diode.
67
STABILIZER
t
co2cw
13 Hz
CHOPPER
4C35
MOO
,f
DELAY
TRIG.
GEN.
\
\
D
R
C
E
L
L
SFD 330
MAG
MW
LOAD
POWER
METER
/
/
1
r^
BS
•
I
^
7
HgCdTe
AT 77°K
SCOPE
Figure 19. Pulsed double-resonance experiment.
68
Two cells have been used to perform the double resonance
experiments.
The waveguide double resonance cell and the
resonator cell, without the FIR feedback mirrors, have produced results that will be described in detail.
The detection technique employed for the pulsed experiments are different than the technique previously described.
The low duty cycle (=0.007) of the microwave makes PSD at
the microwave repetition rate frequency less sensitive. The
observed double-resonance signals using the magnetron have
been easily observed using direct detection. A technique for
detection of much smaller signals has been realized.
The detection scheme is shown in Figure 20. The laser
intensity is modulated at fIR(= 1/T IR ).
occur at a f
The microwave pulses
repetition rate. The pulse width, x, of the
magnetron now becomes important in choosing an appropriate
high-pass filter. The signal observed on the HgCdTe detector
has frequency components at f
, f IR , and fD(= 1/x). In the
cases considered, f >> f , fT„. The filter is chosen to
' p
yur IR
pass f and reject fTT, and f . The components at f,, are
e
J
r
p
IR
yw
yw
very small and can be ignored.
The Tektronix 1A7A high gain
differential amplifier is ideal for this purpose, since it
features a variable band pass filter on the input stage. The
low frequency components of the detector signal are filtered
out, and a series of pulses are observed when the laser intensity
69
LASER INTENSITY
T,R
MICROWAVE INTENSITY
SIGNAL OBSERVED USING HgCdTe
IA7A OUTPUTdO KHz - I MHz, 3db POINTS)
BOXCAR OUTPUT
•+•
TIME (ARBITRARY UNITS)
1
1
L
Figure 20. Pulsed two-photon absorption detection technique.
70
is present.
The inverted 1A7A output is shown in the figure.
All the signals observed to data have been detected in this
manner.
A double modulation detection scheme is easily implemented
with the use of a gated high-speed sample and hold circuit and
a lock-in amplifier.
A Boxcar integrator has been used as the
high-speed sample and hold circuit.
The output of the band
pass filter is stored in the sample and hold and reset every
microwave pulse.
The sample and hold effectively stretches
the duty cycle of the microwave modulation to 100%. The output of the sample and hold will appear to be a square wave at
f__, and can be detected using PSD. Again, care must be taken
to avoid saturating any of the amplifiers m
the chain if
meaningful results are to be achieved.
Experimental Results
Infrared microwave double resonance experiments have been
14
performed using four different C0 2 laser transitions in NH3«
The CO- R(8)
+34.68 GHz resonance with the aQQ(3,3) two-photon
transition has been investigated in more detail than the other
systems.
Only the R(8), 0 experiments will be detailed here.
C0 2 R(8) 1 0 Molecular System in
14
NH 3
The decision to study this system was reached for
several reasons. The Gs(3,3) and Ga(3,3) energy levels have
71
the greatest partitioning of population at room temperature.
g
This two-photon transition has been observed by Freund
and
accurate frequency measurements have been obtained for the infrared transition.
The microwave inversion transition has been
studied and the linewidth measured.
An energy level diagram
relevant to the C0 2 R(8), Q transition is depicted in Figure 21.
The two-photon transition is the aQQ(3,3) transition. A
g
microwave frequency of 34.677 GHz
less the C0 2 R(8), Q tran-
sition at 967.707329 produce two-photon resonance.
The inter-
mediate mismatch for the two-photon process is 10.807 GHz. If
two-photon pumping occurs one expects emission on the v2:sQ
(3,3) transition at 35.79 cm
(X = 279 ym). A competing
linear infrared absorption occurs on the sQ(2,2) transition,
with a resonance mismatch of 940.2 MHz.
The Doppler width*
14
of a two-photon transition in
NH 3 at 10 microns is 80 MHz
and a homogeneous broadening rate of ^ 24 MHz/torr.
This
indicates the two-photon transition shoud be Doppler broadened
at pressures below 3 torr.
CW Results
The CW results were obtained with a CW C0 2 laser
operating on the R(8), Q transition and the B579 Klystron operating at 34.68 GHz. Figure 22 shows a pressure scan of the
signal in the waveguide double resonance cell with an interaction length of 1.5 meters. The two-photon absorption signal
72
10603'fi
" 1040
o
cr
ui
jfj 1020
LU
I-
2 " =
>
A= 940.2 MHz
en 1000
3
v
i
E
100-
>
CD
<r
z
3":
80-
=TT. A=10.807 GHz
Vm
LU
LU
CO
60"
Q
3
O
cc
40-
C3
J
K =2
i ^ = 967 707329
Figure 2 1 .
14
J
K=3
v'm
m = 34.677GHz
NH3 relevant to C0 2 R(8). .
C0 2 R(8)I0 + 34.S8 GHz T.P.A
Al
_k~h X I 0 0
Io " Io
P MW =9.3 WATTS
PIR =175 MWATTS
L =1.5 METERS
,04-
2
NH3
3
4
5
PRESSURE (TORR)
CW two-photon absorption vs. pressure for C0 2
R(8) 1() + 34.68 GHz.
74
is defined as AI/I
in Equation (6)
|i = ^ V ± x 100 .
o
(6)
o
Al is the RMS amplitude of the microwave modulation frequency
observed on the infrared signal with the HgCdTe detector. I
is the zero pressure transmitted RMS power of the infrared, I,
is the transmitted RMS power when two-photon absorption occurs.
The signal was detected using the CW double modulation technique.
The significant features of Figure 22 are the linear rise in
signal at pressures less than 1 torr, the peak signal pressure,
and the magnitude of the signals observed.
The linear rise in
the signal at pressures less than one torr indicates the twophoton transition is Doppler broadened.
The magnitude of the
two-photon absorption measured represents a maximum of 0.18 percent modulation of the infrared source, and indicate 0.02 percent modulation is detectable with a reasonable signal-to-noise
ratio using the double modulation detection technique.
The peak pressure point deserves further consideration.
It suggests that saturation of the infrared or microwaves has
occurred; alternatively, the competing linear process may be
depleting the infrared signal. Figure 23 is an investigation
of the possibility of saturating the infrared radiation.
The
two-photon absorption is measured as the input infrared power
is varied.
The two lines were accomplished at different
75
Al
it
VS
^ R ^ M W - ^ WATTS)
14 L
P =1.2 TORR
AI
P = .5 TORR
08
06
.04
.02
00
50
100
150
PIR (MWATTS)
200
250
300
Figure 23. CW two-photon absorption vs. infrared laser power.
76
pressures.
The line P = 1.2 torr corresponds to the peak sig-
nal pressure.
In both cases the magnitude of absorption de-
pendence is linear with infrared power.
This is expected from
Equation (5). This also rules out saturation of the infrared
power.
The transmitted mfrared power was measured to be
approximately 30% of the zero pressure transmitted power at
the peak pressure of the two-photon signal. This partially
explains the peak pressure.
The results observed for the CW experiments agree qualitatively with theory.
A comparison of calculated and measured
values show differences of 200-300%.
Pulsed Results
The pulsed experiments were peformed with a CW COlaser and a tunable pulsed magnetron.
The results here show
larger amounts of two-photon absorption than have ever been
reported before for this molecular system.
Use of higher
power microwaves is the reason for this stronger interaction.
Figure 24 shows a curve similar to that reported in the CW
section.
The cell has been reduced in length to 0.64 meters.
The observed maximum modulation is now up to 40%. A number
that is forty times larger than theoretically predicted value.
The peak pressure has also increased to almost 4 torr.
Figure
25 shows the effect of the competing linear absorption of the
C0 2 R(8) 1 Q photons. The peak pressure now occurs when the C0 2
power is about 50% of the zero pressure power.
77
CO, R(8),rt + 34.68 GHz TP.A.
PMW a 40 KWATTS
P|R = 1.3 WATTS
L = .64 METER
49
42
y^^\
35
Al
Io
/
28
21
14
-
/
-
-
\
\
/
\
/
^
7
i
1
f
i
i
2
3
4
NH 3 PRESSURE (TORR)
i
i
5
Figure 24. Pulsed two-photon absorption vs. cell pressure.
78
80
%T=+170
PIR = 1.3 WATTS
60
50
%T
40
30
20
1
2
3
4
5
iNH3 PRESSURE (TORR)
Figure 25.
CO, R ( 8 ) , n t r a n s m i s s i o n v s . NH? p r e s s u r e .
79
The results discussed thus far have not exploited the
tunability of the microwave source. Figure 26 contains two
curves.
The upper curve is the transmitted microwave power
as a function of magnetron frequency.
The dips at 34.34 and
34.7 GHz represent unstable oscillation of the magnetron.
The
cell used for this graph is the waveguide cell (0.64m) and the
cell pressure is 0.5 torr. The lower curve is the two-photon
absorption curve. The curve is observed to be very broad in
frequency range. The Doppler width of the transition is about
80 MHz, the observed width is significantly greater than this
value.
The structure of the two-photon curve is attributed
to the variation of the transmitted microwave power.
Figure 27 shows a similar set of curves for the resonator
cell.
The observed two-photon absorption is highly structured
and similar in magnitude to the results achieved in the waveguide cell.
The structure is attributed to the microwave
power transmission spectrum of the resonator cell.
The pulsed experiments produce IMDR signals with peak infrared modulations in excess of 40% of the zero pressure power.
The signals exhibit broadening greater than the Doppler width
of the two-photon transition.
Conclusions
Infrared-microwave double-resonance experiments have been
performed in 14NH, with a CO, CW laser. The aQR(2,l)
H
H
(D
ARBITRARY UNITS
OJ
•!>
Ol
Hi 1-3
ro It*
PI
ro
3 Pi
o
rt
H
P> cn
< 3
ro
rt
rt
ro
OiOi
ro
oB.
ro o
M H
M O
~£
<
ro
»o
o
ro
cn
3
Q)
ro
rt
K
O
3
oo
o
81
RESONATOR CELL
PL = 1.3 WATTS
PM a 4 0 KWATTS
P = .5 TORR
Al
.
"J7
^3.64
33.75 3390
34.04 34.19
34.36
MAGNETRON FREQUENCY (GHz)
34.52
34.72
Figure 27. TPA and transmitted microwave power vs. magnetron
frequency (resonator cell).
82
(P(40)9 + 33.05 GHz) and aQR(2,2) (P(40)9 + 34.14 GHz) twophoton transitions are reported for the first time. The microwave frequencies calculated for a double-resonance match are
found to be within ±35 MHz of the experimentally observed
frequencies.
The data presented is a detailed investigation
of the aQQ(3,3) (R(8)1Q + 34.68 GHz) transition.
The cells designed and the data acquisition techniques
developed, have been used to produce results consistent with
two-photon theory.
literature.
The CW results are similar to the reported
The pulsed results have been performed at higher
microwave powers than previously reported.
A perturbation
calculation adequately describes the CW experiment but breaks
down for the pulsed results. The disagreement is not surprising since the maximum modulation of the C0 2 laser is 40%.
The most surprising feature of the pulsed experiments is that
the observed two-photon absorption bandwidth is much broader
than the Doppler width of the two-photon transition.
The
importance of the microwave power transmission spectra is
reflected in the resulting two-photon absorption spectra.
The lack of success in generation of tunable FIR radiation from mfrared-microwave pumped Stimulated Hyper-Raman
Scattering (SHRS) has resulted in cell designs and data acquisition techniques being developed.
Kim and Coleman
used
a waveguide double resonance cell at different frequency to
probe microwave gain/absorption in optically pumped NH, systems.
83
IV.
CONCLUSIONS
Microwave emission attributed to a linear cascade effect
has been observed from 14NH 3 optically pumped by a TE C0 2 laser.
The emission is tunable over the saturated homogeneous linewidth and exhibits a high degree of frequency stability.
The
observed emission frequencies are shifted from the absorpton
line center.
A self-consistent explanation is presented to
explain the unexpected shift. The criterion for additional
microwave emission systems has been presented.
Infrared-microwave double-resonance experiments have
produced two previously unreported double resonances in 14NH 3 .
These are the aQR(2,l) (P(40)g + 33.05 GHz) and aQR(2,2)
(P(40)g + 34.14 GHz) two-photon transitions. The microwave
frequencies may be accurately (±35 MHz) calculated from the
available spectroscopy.
The CW experiments are in good agree-
ment with perturbation calcuations.
The pulsed experiments
have been performed at higher microwave power levels than previously reported.
The pulsed experiments two-photon absorp-
tion Imewidths are much wider than expected.
84
14
APPENDIX Al: The Spectroscopy of
NH 3 Around 10 Microns:
Notation, Energy Levels, Selection Rules,
Transition Moments, and Laser-Absorption Matches
The purpose of this Appendix is to detail the spectroscopic notation and data required for optical pumping and in14
frared-microwave double-resonance experiments in NH3«
Energy Levels
NH_ is an oblate symmetric top molecule which belongs to
the point group C 3 . The symmetrical pyrimidal structure of
point group C 3
molecules is characterized by four fundamental
vibrational modes, two totally symmetric, v . ^ ) , v 2 (A 1 ), and
two doubly degenerate, v (E) and v.(E) modes. 27
3
The NH 3 inversion splittings, resulting from the nitrogen
atom tunneling through the plane of the hydrogen atoms, produce
a rich ground state microwave absorption spectrum.
The fun-
damental v_ vibration-rotation band leads to an infrared absorption band in the 10 micron wavelength range.
These two features
make ammonia an ideal candidate far mfrared-microwave interaction experiments.
The spectroscopic notation used to identify an energy
level in NH 3 is
below:
specified by five sets of characters as shown
85
LEVEL
PARITY
VIBRATIONAL STATE
Gs(J,K,M)
+
GROUND
Ga(J,K,M)
-
GROUND
v2s(J,K,M)
+
v
v2a(J,K,M)
-
v
2v2s(J,K,M)
+
2
2v2
2v2a(J,K,M)
-
2v
2
2
The first characters (G, v 2 , 2v2) designate the vibrational
manifold in the electronic ground state of NH 3 .
The second
character, (s, a ) , gives the symmetric or antisymmetric component of the inversion doublet.
mentum quantum number.
J is the total angular mo-
K(£J) is the projection of the total
angular momentum on the molecular axis quantum number.
M(=-J,..,0..,+J) is the projection of the total angular momentum on a space fixed axis quantum number (i.e., laboratory
reference frame).
The M quantum number is usually not specified
because the M levels are degenerate in the absence of an external perturbing field.
The energy levels for 14NH, are listed in Table Al.l.
The energy levels are taken from a Ph.D. dissertation by J.
25
Curtis.
The accuracy of these energy levels is believed good
to ±0.001 cm
. The selection rules determine the allowed
transitions and are the next topic for discussion.
86
Table Al.l
J
0
1
1
2
2
2
3
3
3
3
4
4
4
4
4
5
5
5
5
5
5
6
6
6
6
6
13
NH-l Energy levels.
GROUND STATE
ANTIK SYMMETRIC SYMMETRIC
0
0
1
0
1
2
0
1
2
3
0
1
2
3
4
0
1
2
3
4
5
0
1
2
3
4
v - STATE
^
ANTISYMMETRIC SYMMETRIC
2v - STATE
ANTISYMMETRIC SYMMETRIC
986. 1220
1882. 1757
.7934
19 8903
16 1408
55. 9075
44. 6674
119. 2411
115. 5069
104. 2948
85. 5750
194. 8789
183 7037
165 0458
138..8562
297 6501
293 9436
282 .8142
264 .2332
238 .1517
204 .5021
412 .6024
401 .5276
383 .0373
357. 0849
-
16 .9311
60 .4146
56 .6779
45 4587
-
116 .2485
105 .0564
86 3712
199 .2994
195 .5838
184 .4276
165 .8025
139 .6614
-
294 .6053
283 .4937
264 .9432
238 .9073
205 .3204
416 .8989
413 .2158
402 .1575
383 .6963
357. 7852
952..5705
948 5628
988. 8132
976. 7920
1053. 1125
1049. 1323
1037. 1516
1017 1316
1129. 4238
1117. 5027
1097 5753
1069 .5583
1233 5460
1229 6008
1217 .7429
1197 .9346
1170 .0822
1134 .0747
1349 .5254
1337 .7510
1318 .0754
1290. 4210
984. 1441
1027. 4325
1023. 6821
1012. 4073
.
1082 9584
1071 7029
1052 9236
1165. 6486
1161
1150
1131
1105
9061
6830
9497
6708
1260 .4757
1249 .2832
1230 .6131
1204 .4177
1170 .6540
1382 2636
1378 .5726
1367 .4263
1348 .8252
1332. 7353
1617. 7694
1613 5736
1654. 2662
1641 5096
1719 7406
1715 4526
1702 .5410
1681 1258
1796 .9556
1784 .0221
1762 .5371
1732 .8186
1903 .1433
1898 .8206
1885 .8273
1864 .2462
1834 .2109
1796 .1793
1898. 0092
1940. 3131
1936. 7686
1926. 1137
1994. 8734
1984. 2257
1966 4653
2075 8869
2072 3394
2*061 6882
2043 9447
2019 .1023
2020 .9692
2007 .9320
1986 .2347
2169 .1052
2158 .4591
2140 .7288
2115 .8922
2083 .9892
2288 .6995
2285 .1604
2274 .5197
2256 .7979
1956. 0344
2231. 9789
87
Table Al.l
J
(con't)
GROUND STATE
ANTIK SYMMETRIC SYMMETRIC
5
6
0
1
2
323..6019
282.,5022
554..4063
550..7389
539..7270
521..3432
495,.5377
462..2470
421..3847
372..8463
8
8
8
8
8
8
8
9
3
4
5
6
7
0
1
2
3
4
5
6
7
8
0
9
9
9
9
9
9
9
9
9
1
2
3
4
5
6
7
8
9
6
6
7
i
7
7
7
7
7
7
7
8
8
708..2189
697.,2780
697..0102
653.,3748
620..3010
579..7065
531..4885
475..5250
888..5101
884..8925
874..0303
855..8998
830..4442
797,.6108
757,.3131
709,.4504
653,.9028
590,.5318
324..3601
283,.3379
551,.3006
540..3037
521,.9441
496..1787
462..9409
422,.1493
373..7040
712..3665
708..7270
697.,7996
679..5590
653..9544
620.,9283
580..3976
532..2634
476,.4095
885,.3465
874..4964
856..3793
830..9620
798,.1711
757,.9301
710,.1420
654,.6919
591,.4483
v 2 STATE
ANTISYMMETRIC SYMMETRIC
2v- STATE
ANTISYMMETRIC SYMMETRIC
1254..6673
1210..6598
1492..9316
1489..0375
1477..3633
2189..1006
1247..8558
1917..5191
1817..3648
2167..7618
2163..3706
2150..2512
1457..8517
1430..4169
1394..9522
1351..3114
1299..3009
1486..4900
1460..5181
1427..0351
1385..9789
1337,.2684
1676..5896
1672..9072
1661..8860
1643..4708
1617..6421
1584.*3457
1543..5010
1495..0411
1438.,8919
1647..9841
1636.,4195
1617.,0744
1589.,8913
1554..7595
1511..5204
1460..0019
1399..9917
1830..0084
1826..1801
1814.,7208
1795..5752
1768.,6591
1733..8833
1691..0862
1640..1130
1580..7400
1512,.7024
1516..0842
1504..9971
1848,,9065
1837..9452
1819..6518
1793..9774
1760..8846
1720..2841
1672..1179
1616,.2903
1552,.7035
2128..4353
2098..0918
2059..3151
2012..3703
1958..3189
2325..9941
2312,.7789
2290.,8320
2260..2713
2221..2616
2174.,0152
2118,.7910
2055..8955
2513..2106
2508..7657
2497..4547
2473..3483
2442..5643
2403..2676
2355..6702
2300,.0313
2236,.6572
2165,.9014
2200.,0829
2161..1286
2420.,4349
2409.,8028
2392.,0904
2367..2945
2335..4277
2296..4908
2250..5051
2578..4401
2574..8996
2564..2789
2546..5802
2541.,8070
2489..9645
2451..0592
2405.,0990
2352.,0934
2748..4923
2337..8813
2720..1999
2695..4505
2663,.6334
2624,.7702
2578..8540
2525,.8991
2465,.9163
88
Table Al.l
T
K
10 0
10 1
10 2
10 3
10 4
10 5
10 6
10 7
10 S
10 9
10 10
11 0
11 1
11 2
11 3
11 4
11 5
11 0
1 1
7
11 3
11 9
11 10
11 11
12 0
12 1
12 2
12 3
12 4
12 5
(con •t)
GROUND STATE
ANTISYMMETRIC SYMMETRIC
-
1080.5944
1069.8131
1051.8209
1026.5785
994.0078
954.0348
906.5603
351.4672
788.6195
717.8625
1298.7031
1295.1453
12*34.4618
1266.6385
1241.5965
1209.3095
1169.6868
1122.6317
1068.0323
1005.7477
935.6349
857.5210
-
1528.3535
1517.7692
1500.0810
1475.3039
1443.3202
1084.5812
1080.9952
1070.2296
1052.2644
1027.0355
994.5022
954.5791
907.1702
852.1627
789.4269
718.8166
-
1295.4949
1284.8207
1266.9899
1241.9950
1209.7405
1170.1612
1123.1631
1068.6381
1006.4505
936.4648
858.5188
1532.1780
1528.6548
1518.0784
1500.4424
1475.6472
1443.6915
v_ STATE
ANTISYMMETRIC SYMMETRIC
-
2023.4185
2012.0826
1993.1381
1966.5186
1932.1102
1889.8026
1839.4061
1780.7237
1713.4979
1637.4329
2243.2070
2239.4865
.2228.2782
2209.5617
2183.2469
2149.2338
2107.4295
2057.6523
1999.6963
1933.3395
1858.2810
1774.1836
2047.5499
2043.9233
2033.0392
2014.8680
1989.3754
1956.4950
1916.1825
1863.3363
1812.8781
1749.7020
1678.7120
-
2474.1616
2463.0830
2444.5805
2257.7899
2246.9874
2228.9654
2203.6630
2171.0232
2131.0123
2083.5305
2028.4767
1965.7683
1395.2353
1816.9277
2493.8938
2490.3242
2479.6061
2461.7209
2418.5853
2384.9908
2436.6261
2404.2532
-
2v- STATE
ANTISYMMETRIC SYMMETRIC
-
2711.6312
2698.2139
2675.9302
2644.8931
2605.2323
2557.2950
2501.1953
2437.2895
2365.9311
2287.5206
2944.6328
2941.1495
2930.5502
2912.9371
2883.1639
2856.3356
2817.5588
2771.6914
2718.7923
2658.8739
2591.9470
89
Table Al.l
12 6
12 7
12 8
12 9
12 10
12 11
12 12
(con't)
GROUND STATE
ANTISYMMETRIC SYMMETRIC
1404.0722 1404.4808
1357.4654 1237.9230
1303.3882 1303.9096
1241.7113 1242.3159
1172.2873 1173.0008
1094.9508 1095.8079
1009.5175 1010.8079
v. STATE
*
ANTISYMMETRIC SYMMETRIC
2343.7258 2364.5848
2294.5968 2317.4996
2237.4194 2262.9086
2171.9554 2200.6989
2097.9451 2130.7738
2015.0729 2053.0306
1922.9511 1967.3582
2v_ STATE
*
ANTISYMMETRIC SYMMETRIC
90
Selection Rules - Single Photon
The selection rules govern the change in quantum numbers
allowed during a transition from one level to another. The
single photon selection rules for the parallel v„ band are:
AJ = J'-J=0, +1
AK = K'-K=0 (±3 perturbation allowed)
AM = M'-M = 0,±1
s •*• a
a •+ s
Av ~ 0, ±1
0 pure rotation transition
±1 vibration-rotation transition
J, K, and M are the initial state quantum numbers. J', K 1 ,
and M* are the final state quantum numbers. The spectroscopic
notation used identifies the inital level and the type of
transition involved.
sitions, respectively.
P, Q, and R designate AJ = -1,0",+1 tranOnly AK = 0 transitions will be con-
sidered in this investigation.
observed by Freund and Oka.
A AK = +3 transition has been
The AV selection rule specifies
either pure rotation or vibration-rotation transitions. The
notation used to specify pure-rotation or vibration-rotation
is demonstrated below.
91
NOTATION
INITIAL LEVEL
FINAL LEVEL
G:sQ(J,K)
gS(J,K)
Ga(J,K)
v2:aR(J,K)
v2a(J,K)
v2s(J+l,K)
sP(J,K)
Gs(J,K)
v2a(J-l,K)
2aQ(J,K)
v2a(J,K)
2v2s(J,K)
In all single-photon cases, the absorption notation is
used. Absorpton notation implies the initial level will always
be lower in energy than the final level. As an example,
G:aQ(J,K) and G:sQ(J,K) are the same transition.
The absorption
notation specifies the G:sQ(J,K) as the correct identification,
because the symmetric component of the inversion doublet is
lower in energy than the antisymmetric component.
Selection Rules - Multi-Photon
The multi-photon selection rules can be obtained by applying the single photon selection rules n sequential times.
(n
is the number of photons involved in the non-lmear transition) .
The absorption notation is not used for multi-photon transitions to avoid unnecessary confusion.
The intermediate levels
can be real, as shown in Figure Al-1, or virtual. A treatment of multi-photon transitions is given in Ref. 28. The
two-photon selection rules are presented as an example.
The two-Photon selection rules for NH 3 are:
2
±
3
+
—4
ui
X
UI
Aa>2
o
a»2
u>2
i<u
a>,
2
+
•>
Aui,
1
c
U>,
B
Figure Al-1.
NH
Transitions allowed in 14.
3«
(two and three-photon).
A-linear (single photon); B,C-nonlinear
vo
to
93
AJ = 0, ±1, ±2
AK = 0
AM = 0, +1
a •*- a
s ->• s
Av = 0, ±1, ±2
The notation for the two-photon transitions is similar but considerably more involved.
Only the infrared-microwave cases
will be considered at this time. The microwave frequencies
will always be interacting with inversion transitions (Q type
rotational) in the ground state. The two-photon selection
rules then reduce to:
AJ = 0, ±1
AK = 0
AM = 0, ±1
a •+• a
s -»- s
Av = ±1
The allowed two-photon transitions considered here are identical
in Table A1.2. Table Al.2 demonstrates the notation used for
multi-photon transitions. The general form for the notation
is
sT x T 2 ... Tn(J,K)
Table Al.2
Infrared-microwave two-photon transitions considered in this investigation.
TRANSITION
IDENTIFICATION
INITIAL
LEVEL
INTERMEDIATE
LEVEL
FINAL
LEVEL
RELATIVE
POLARIZATION
AM=0 AM=±1
FREQUENCY
•
sQP(J,K)
Gs(J,K)
Ga(J,K)
v2s(J-l,K)
aQP(J,K)
Ga(J,K)
Gs(J,K)
v2a(J-l,K)
sQQ(JfK)
Gs(J,K) •
Ga(J,K)
v2s(J,K)
aQQ(J,K)
Ga(J,K)
Gs(J,K)
v2a(J,K)
sQR(J,K)
Gs(J,K)
Ga(J,K)
v2s(J+l,K)
aQR(J,K)
Ga(J,K)
Gs(J,K)
v2a(J+l,K)
1
1
II
II
1
1
II
II
1
1
II
II
V
£+Vm
v
lf V m
V
£+Vm
V
£~ V m
x, m
x. m
vo
.fe.
95
where s, J, and K define the initial state quantum numbers,
and T., T 2 , ... Tn specify the type of AJ transitions involved.
The perpendicular band notation is similar but considerably
more complicated and will not be treated here (See refs. 27
or 29 for perpendicular band selection rules).
Further infor-
mation regarding infrared microwave transitions is contained
in Table Al.2. The relative polarization indicates the orientation of the infrared with respect to the microwave. This
assumes linear polarization of the infrared and microwave.
The orientation listed under AM = 0 is the preferred polarization, as is determined from the direction cosine matrix ele9
ment.
The frequency listed gives the most resonance enhanced
frequency (i.e., Aw is smaller if this condition is true).
This is important because the intermediate mismatch AOJ is large
to begin with for the systems under study.
Transition Moments
The transition moments presented are the result of a perturbation calculation.
They can be used to get a quantitative
feel for the relative strengths of higher order multi-photon
transitions.
The transition moments for the interactions
28
shown in Figure Al-1. are:
MA =
<l|y-E|2>
2^
for A,
(Al.l)
96
M_a =
<l|y,-E, 2><2|y 2 «E-|3>
—
5
—
2fi Aw
for B , and
(Al.2)
<l|ii1 *E, |2><2|u-»E,|3><3|u,-E,|4>
=
—
|—
-—
for C.
4fi Aw, A w 2
M
(A1.3)
The y is the dipole moment and E (w.) is the corresponding
field associated with the transition.
A single-photon transi-
tion moment corresponds to the precession (or Rabi) freqquency.
'
'
The dipole matrix element must b e corrected
for different quantum numbers as indicated in Equation (A1.4)
K
=
lpJ5J,K,MfJ'K'M-
(A1 4)
*
The values of i|u- Iifor 14N H 3 are a function of energy state and
type (i.e., rotation or vibration-rotation) of transition.
These values a r e :
Gs(J,K) ++ Ga(J',K')
v 2 s(J,K) •*-*• v 2 a(J' ,K')
2v-s(J,K) + + 2v 2 a(J',K')
a
G(J,K) +•*• fv_(J' ,K')
gV2(J,K)
ment.
1.25 ±0.01 debye
.83 ±0.02
.23 ±0.02
debye
debye
a. c*
S
The 8j K
1.475 ±0.006 debye
|2v2(J',K')
M . j MK i M i
.27
±0.05
debye
is the direction cosine matrix ele-
These values are determined from Table 4.4 of Townes
and Schawlow.
g
97
Table A1.3
Match routine results for
10 microns.
LASER
LINE
MATCH
cm~l
FREQUENCY
cm"-'-
.0006
.0007
-.0018
-.0031
964 .4612
1072 .0525
924 .0692
.0032
-.0039
.0041
927 .7417
949 .5401
968 0074
-.0046
.0050
.0061
-.0071
.0072
-.0076
-.0082
1054 9155
1046 .3797
1084. 6351
1006 0531
P10(50)
P09( 8)
P10(13)
R10(56)
R10(37)
SEQP09( 7)
SEQP09(17)
R09(30)
P09(14)
962. 5328
944 9970
965. 8905
SEQR10( 8)
P10(28)
R10(34)
-.0085
.0085
1046. 3797
944. 1940
-.0085
-.0092
.0093
-.0093
-.0093
N20
C12018
C12018
ABSORPTION
FREQUENCY TRANSITION
cm--*964 .4606
1072 .0518
924 .0710
1011 .2041
sQ [ 8r
2sR 6 ,
aQ [10 ,
aR 3 r
aQ
2sQ
SQI
aR<
sRI
sRI
C13016
927 .7385
949 .5440
968 .0033
1054 9201
1046 3747
1084. 6290
1006. 0602
C12016
C12018
962 .5256
945. 0046
SEQP09(17)
P10(20)
N20
C12016
C12016
965. 8987
1046. 3882
944. 1855
sQI 11,- 2)
aQ! 12, 2)
SQI 6, 2)
sRI 3, 2)
2sQ| 4, 2)
945. 4071
1012. 4371
R10( 7)
SEQP09(51)
N20
C12016
945. 4155
1012. 4463
2sQ| 6, 5)
aR| 3, 2)
R10(20)
P09(48)
R10(18)
P09(24)
C13018
C13018
C13018
941. 8530
987. 7426
940. 7162
aQI10, 2)
.0097
941. 8623
987. 7333
940. 7069
1043. 1632
C12016
1043. 1535
.0101
-.0103
.0103
944. 1956
939. 5285
951. 2145
HBR10(23)
R10(16)
R10(38)
C12016
C13018
C13018
944 1855
939. 5388
1011 .2010
R10(32)
P09(16)
GAS
14
NH 3 and lasers around
C13016
N20
C13016
N20
C12016
C12016
C12016
951. 2042
8r
4,
1,
5,
2)
2)
9)
3)
7)
4)
1)
1)
3,, 3)
5,, 1)
2sR| 3 r 1 )
2sRI 1( 0)
2sQ| 5, 2)
2sR| 4, 3)
2sQ{ 4, 2)
2sQ| 8, 6)
2sQ| 7, 7)
Table A1.3 (con't)
LASER
LINE
MATCH
cm -1
FREQUENCY
cm"1
-.0107
.0107
-.0107
963 .2631
917 .7417
96317413
RIO(02 )
SEQP10(45 1
RIO(31 1
C12016
.0108
-.0118
945 .1902
947 .9436
1085 .8740
951 .1923
1033 .3031
RIO(26 )
SEQP10(10 1
SEQR09(39 )
P10(12 •
SEQP09(31 1
C13018
.0119
-.0119
-.0122
-.0125
-.0127
-.0127
-.0128
-.0128
.0130
.0142
-.0145
.0148
-.0150
.0151
-.0156
.0157
-.0158
-.0159
.0163
.0169
-.0169
-.0175
-.0175
GAS
C12016
N20
C12016
C12016
C12016
C12016
ABSORPTION
FREQUENCY TRANSITION
cm -1
963.2738
917.7310
963.7520
945.1794
947.9554
1085.8621
951.2042
1033.3153
932.0952
965.3453
945.5068
959.6509
951.2042
957.8407
7 r 6)
2,, 1)
2sR 7 r 1 )
2sQ 7 r 7 )
aR [ 4, 2)
aQ 3 , 2)
sQ 7 r 1 )
aQ< 12,r 1 )
sQI11,,10)
2sQ 7 , 7)
sQ [12 rl2)
sQ 11 r 5 )
2sQ 10 r 6 )
sQ< 10 , 9)
sQ< 11,,11)
sQ< 10,, 3)
sQ 5 , 4)
sQ 3 , 3)
sQI 6 , 1)
2sP< 2,, 1)
2sQ| 8,- 2)
2sQ< 5,, 3)
aQ< 4, 3)
932 .0827
965 .3326
945. 4941
959. 6381
P10(08 •
N20
SEQR10(12] C12016
R10(48] C13016
SEQR10(04] C12016
951 1914
957. 8537
961. 7329
927. 7417
961. 0973
959. 3917
963. 0623
966. 2504
SEQP10(06]
RIO(23
C12016
N20
R10(00]
P10(13]
C12016
N20
C12016
C12016
C12016
C12016
961.7187
927.7562
C12016
C12018
930.3072
SEQR10(06]
P10(02]
967 3643
965. 9544
909. 1801
927. 8758
942. 8111
931. 7559
SEQR10(05)
RIO(06
SEQR10(11] C12016
SEQR10(09 > C12016
SEQP10(53] C12016
SEQP10(35] C12016
SEQP10U9] C12016
SEQP10(31] C12016
930 2897
935. 8901
SEQP10(30]
P10(38]
sQ 9 , 5)
2sQ 10 r 1 )
sQ 9 r 3 )
961.0825
959.4067
963.0471
966.2660
967.3486
965.9702
909.1960
927.8594
942.7942
931.7728
935.9076
2sQ
2sQ
aQ 6 5)
aQ< 7 r 3 )
Table A1.3
(con't)
LASER
LINE
ABSORPTION
FREQUENCY TRANSITION
cm~l
MATCH
cm~l
FREQUENCY
cm"l
-.0176
929 .4882
P10(11
N20
929 .5058
-.0177
1023 .1894
P09(44
C12016
1023 .2071
.0187
966 .2504
RIO (06
C12016
966 .2317
2sQ ( 8r 3)
2sR ( 3r 1)
sP ( 1, 0)
-.0191
964 .7690
RIO (04
C12016
964 .7881
sQ ( 7r 5)
-.0195
1046 .3797
SEQP09(17
C12016
1046 .3992
sR ( 3, 1)
.0197
962 .6872
P10(06
C12018
962 .6675
.0197
1091 .4098
SEQR09(51
C12016
1091 .3901
-.0199
942 .8957
SEQP10(16
C12016
942 .9156
SQ ( 9r 7)
2sR ( 7r 4)
2sQ [ 4, 1)
..0203
1070 .6124
SEQR09(13
C12016
1070. 5921
aR ( 6,
5)
-.0208
949 .3009
RIO (34
C13018
949 .3217
2)
-.0210
965. 3325
SEQR10(12
C12016
965. 3536
-.0215
927 3004
RIO (18
C13016
927 3219
.0216
1084. 6351
R09(30
C12016
1084. 6135
-.0220
1040 .9345
SEQP09(23
C12016
1040. 9564
2sQ [ 2r
sQ ( 6,,
aQ [ 6 r
sR [ 5,
2SR [ 4,,
-.0220
939. 2230
SEQP10(23
C12016
939. 2450
.0221
929. 9932
RIO (22
C13016
929. 9710
.0230
1084 6351
R09(30
C12016
1084. 6121
-.0231
965. 3222
P09(54
C13016
965. 3453
-.0235
RIO(34
C13016
937. 6080
-.0237
937 5845
960 8258
RIO(27
N20
960. 8495
-.0240
931. 3093
RIO(24
C13016
931. 3333
.0241
939. 5285
RIO(16
C13018
939. 5044
.0242
951. 8013
P10(20
C12018
951. 7771
-.0249
933. 6592
SEQP10(29
C12016
933. 6840
-.0252
1023. 1819
SEQP09(41
C12016
1023. 2071
2sR< 3, 1)
-.0252
915. 6396
SEQP10(47
C12016
915. 6648
aQ 11, 11)
-.0255
926. 8635
P10(14
N20
926. 8890
aQI H i
GAS
6)
6)
0)
2)
aQ [12,, 6)
aQ [ 9, 7)
sR [ 5,, 2)
sQ [ 7, 1)
aQ [10, 5)
sQ 10, 10)
aQ [ 2,
2sQ< 5,
aR o,
2sQ [ 9,
2)
1)
0)
6)
9)
100
Table A1.3
(con't)
MATCH
cm~l
FREQUENCY
cm"1
-.0257
.0258
-.0261
-.0263
.0267
.0270
-.0270
.0271
926.8633
.0275
.0275
-.0278
.0283
.0289
.0292
r .0294
915.6390
912.3593
1075.7954
932.1620
1116.0432
908.0855
939.2230
950.4963
937.6355
1046.3797
937.7652
1088.7574
1066.0374
.0297
.0300
.0301
.0302
.0304
-.0306
.0308
963.0177
924.9740
963.9446
921.6753
936.3397
939.5285
992.4848
932.6052
965.1776
.0309
.0310
971.9130
954.5451
-.0310
.0311
932.9604
945.4071
.0297
LASER
LINE
GAS
HBQ10(11 ) C12016
P10(58 ) C12018
P10(30 1
N20
SEQR09(21 I C12016
SEQP10(28 1 C12016
R09(58 1 C12018
SEQP10(52 1 C12016
SEQP10(23 > C12016
R10(58 1 C13016
SEQP10(22
C12016
SEQP09(17
C12016
P10(36] C12018
SEQR09(45 \ C12016
R09(02] C12016
N20
R10 (30]
P10(40] C12016
SEQR10(10] C12016
R10 (10] C13016
P10(03]
N20
RIO (16] C13018
C12016
RIO (50
RIO (26
C12016
RIO (33]
N20
SEQR10(22] C12016
P10(08] C12016
P10(32]
RIO (07]
C12016
N20
ABSORPTION
FREQUENCY
TRANSITION
cm-l
926.8890
915.6648
912.3854
1075.8217
932.1353
1116.0162
908.1125
939.1959
950.4688
937.6080
1046.4075
937.7369
1088.7285
1066.0082
963.0471
924.9443
963.9149
921.6453
936.3096
939.4983
992.4544
932.6358
965.1468
971.8821
954.5141
932.9914
945.3760
aQ (11 r 9)
aQ dl ,11)
aQ [12 rl2)
aR [ 6r 1)
aQ [ 2,
r 1)
aR [ 8r 3)
SP [ 3, 2)
aQ [ 9,
, 3)
aQ [ 6r 6)
aQ 10,- 5)
sR! 3, 0)
aQ! 7, 1)
2sR| 7,, 3)
2sR 5, 5)
sQ! 10, 3)
aQ! 12, 10)
sQ! 9, 2)
2sQ 10, 4)
aQ< 6, 1)
aQ! 11, 5)
aR! 2, 1)
aQ! 6, 4)
sQ! 7, 3)
aRI 1, 1)
2sQ( 10, 10)
aQ! 5, 3)
2sQ|10, 9)
101
Table A1.3 (con't)
LASER
LINE
GAS
MATCH
cm"l
FREQUENCY
cm"1
-.0314
-.0314
-.0315
-.0327
-.0328
.0329
965.3222
919.7176
931.9798
967.7072
P09(54
P10(22
RIO (04
RIO (08
C13016
N20
C13018
C12016
949.3009
949.4793
RIO(34
P10(14
C13018
C12016
.0337
.0338
-.0341
-.0341
-.0343
968.0370
952.1355
944.1040
939.2230
938.6883
.0350
.0352
-.0356
936.8037
949.5792
933.7956
951.7414
SEQR10(16
RIO (40
R10(24]
SEQP10(23
P10(26
P10(28]
C12016
C13018
C13018
C12016
C12016
C12016
C12016
N20
N20
ABSORPTION
FREQUENCY TRANSITION
cm~l
965 .3536
919 .7490
932 .0113
967 7399
949 3337
949 .4464
968 0033
952 .1017
944 1381
939 2571
938 7225
936 7687
949 5440
933 8312
951 7771
962 3269
932 0952
1084 5985
938 7410
940 4755
942 9381
915 6648
941. 2993
962 9710
963. 0471
964. 4221
942.9750
915.6277
941.3366
963.0088
SEQP10(08]
P10(06]
RIO (15
RIO (29]
P10(42]
R09(30]
RIO (36]
R10(01]
R10(04]
HBP10(14]
RIO (02]
P09(56]
-.0383
.0391
963.0088
964.4612
P09(56]
RIO (32]
N20
C12018
C12016
C13016
N20
N20
C12016
N20
C13016
C13016
N20
-.0395
.0396
912.3459
960.0738
SEQP10(48]
SEQR10(01]
C12016
C12016
912 3854
960 0342
-.0400
942.3833
P10(22]
C12016
942 4233
-.0357
-.0363
-.0365
.0366
.0366
.0368
.0369
-.0371
.0373
.0378
962.2906
932.0587
1084.6351
• 938.7777
940.5123
sQ [ 6r 6)
2sQ [10 , 3)
aQ [ 7, 5)
sQ![ 2, 2)
2sQ [ 3r 3 )
2sQ [ 1r 1)
sQI 1,, 1)
2sQ [ 8, 8)
aQ! 12,, 3)
aQ! 8 r 1 )
2sQ!. 6 , 3)
2sQ 6 r 2 )
2sQ 4 , 4)
aQ! 8 , 5)
aR 0 , 0)
sQ! 11 , 3)
aQ! 3 , 2)
sR! 5 , 5)
2sQ( 9, 7)
2sQ 7 , 5)
aQ! 12,, 4)
aQ! 11,,11)
aQ! 12, 5)
sQI 9, 6)
sQI10 , 3)
sQI 7 , 7)
aQ! 12 ,12)
sQI 12 , 7)
aQ 10 r 1)
Table A1.3
102
(con't)
LASER
LASER
ABSORPTION
FREQUENCY TRANSITION
cm"l
MATCH
cm"1
FREQUENCY
cm~l
-.0402
.0408
.0408
957 .8005
933 .8808
887 .9186
P10(04
RIO (28
P10(30
C12016
C13016
C13016
957 .8407
933 .8400
887 .8778
SQ (12 ,12)
aQ ( 4f 1 )
SP ( 4r 3 )
.0420
-.0424
.0426
963 .2631
942 .8957
962 .6872
1085 .8740
1084 .6351
967 .3053
937 .1031
961 .7329
1108 .9226
RIO (02
SEQP10(16
P10(06
SEQR09(39
C12016
C12016
C12018
C12016
963 .2211
942 .9381
962 .6446
1085 .8310
sQ (10 , 2)
aQ (12 r 4 )
R09(30
RIO (36
RIO (12
R10(00
R09(42
C12016
N20
C13018
C12016
C12018
967 3643
SEQRlOdl
C12016
1084 .5920
967 .3486
937 .0596
961 .7765
1108 8789
967 4081
sQ [ 3r
aQ ( 7r
sQ [10 ,
2sR [ 8,
SQ 3 ,
933 .7956
1075 .9878
P10(06
R09(16
N20
C12016
.0454
.0456
.0458
939 5285
962. 5328
1046. 3797
C13018
C12016
C12016
aQ [ 4,
aR [ 6,
aQ 10 sQI 10,,
2sR| 4,
-.0458
.0463
1086. 8698
931. 2212
RIO(16
SEQR10(08
SEQP09(17
R09(34
P10(09
933 8400
1076 0327
939 4831
962. 4872
1046. 3339
C12016
N20
1086. 9156
921. 1749
.0467
-.0468
.0474
.0482
963. 0177
935. 8901
932. 0587
971. 9303
RIO(30
P10(38
P10(42
RIO(14
N20
C12018
C12018
C12016
.0492
930. 3564
933. 8808
923. 1114
938. 0189
P10(10
R10(28
RIO (12
N20
C13016
C12016
962. 9710
935. 9369
932. 0113
971. 8821
930. 3072
933. 8312
923. 1610
2sR| 7,- 2)
aQ! 5, 4)
SQI 9, 6)
aQI 8, 4)
aQ! 7, 5)
aR! 1, 1)
P10(01
932. 9409
P10(07
N20
N20
938. 0691
932. 9914
.0430
.0431
-.0433
.0435
-.0436
.0437
-.0438
-.0444
-.0449
.0496
-.0497
-.0502
-.0505
GAS
sQ (11 r 1 )
2sR ( 6, 6)
sR [ 5r 3)
3)
2)
7)
5)
2)
1)
0)
4)
5)
4)
aQI 6, 5)
aQI 8, 5)
2sQ< 9, 2)
2sQ< 10, 8)
aQI 5, 3)
103
Table A1.3
(con't)
LASER
LINE
FREQUENCY
cm~l
-.0513
935 .8563
R10(10)
C13018
935 9076
.0515
1084 .6351
R09(30)
C12016
1084 5836
aQ [ 7, 3)
sR [ 5,, 4)
-.0520
926 .8635
965 3222
P10(14)
N20
P09(54)
C13016
926 9155
965 2701
2sQ [ 8, 1)
sQ 7, 2)
926 8633
932 .0827
HBQlOdl)
C12016
2sQ< 8, 1)
P10(08)
N20
926 9155
932 1353
P10(26)
C12016
938. 7410
SEQRIO(OI)
C12016
-.0532
938 .6883
960 0738
963 7413
RIO(31)
N20
960 0206
963 7945
.0539
960 .0881
R10(26)
N20
.0541
921. 8703
SEQPIO(41)
C12016
960 0342
921 8162
-.0545
1086 8611
SEQR09(41)
C12016
1086 9156
-.0546
949. 3919
RIO (12)
N20
949 4464
.0552
938. 7777
RIO(36)
C13016
938. 7225
-.0554
935. 5381
SEQPIO(27)
C12016
935 5935
.0556
928. 6166
P10(12)
N20
928. 5610
.0557
965 .9544
SEQR10(09)
C12016
965. 8987
.0559
944. 1940
P10(20)
C12016
944 1381
.0566
RIO(22)
C13018
P10(34)
C12016
.0571
942 9947
931 .0014
932 .9409
P10(07)
N20
942 9381
931 0582
932 .8838
.0572
917 7882
P10(56)
C12018
.0575
948. 0129
SEQPIO(13)
C12016
917 7310
947 9554
.0575
944 1956
HBR10(23)
C12016
944 1381
.0582
949. 3919
RIO (12)
N20
949. 3337
-.0582
1085 .7728
R09(02)
C12018
1085 8310
.0585
959 .7094
PIO(IO)
C12018
959 6509
.0593
957 .9000
R10(54)
C13018
957 8407
.0521
-.0522
-.0526
-.0528
.0532
-.0568
GAS
ABSORPTION
FREQUENCY TRANSITION
cm~l
MATCH
cm"1
aQ![ 2, 1)
2sQ 9, 7)
sQ [11,, 9)
sQ [ 8, 6)
sQ [12,, 7)
aQ 11, 10)
2sR ( 7, 2)
2sQ ( 1 1)
2sQ 6, 3)
aQ [ 6,, 2)
aQ (10, 8)
sQ [ 6,, 2)
aQ (12, 3)
aQ (12,,
aQ (11 ,
aQ [ 3,
2sQ (10
2sQ 2,
4)
8)
1)
1)
1)
aQ (12,, 3)
2sQ ( 3, 3)
2sR ( 6,, 6)
sQ (11 ,10)
sQ (12 ,12)
104
Table A1.3
(con't)
MATCH
cm"l
FREQUENCY
cm~l
.0594
.0595
.0598
-.0599
-.0599
942.9750
926.1076
964.5204
961.5800
959.3468
LASER
LINE
R10(04)
P10(48)
SEQR10(07)
SEQR10(03)
R10(25)
GAS
N20
C12018
C12016
C12016.
N20
ABSORPTION
FREQUENCY TRANSITION
cm~l
942.9156
926.0481
964.4606
961.6399
959.4067
2sQ( 4, 1)
aQ( 9, 8)
sQ( 8, 3)
sQ(12, 3)
sQ(ll,ll)
105
14
Laser-Absorption Matches in
NH-.
Optical pumping experiments require a molecular transition
frequency match with a laser frequency. The absolute value of
the resulting mismatch in frequency may or may not be important
to the interaction of interest. Most optical pumping experiments in 14NH produce far IR emission if the mismatch is less
3
than 1.5 GHz (i 0.05 cm
). The energy levels presented
m
Table Al.l have been used to calculate the normally allowed
transitions in 14NH 3 . These transitions were then merged and
sorted with the frequencies of the lasers around 10 microns.
The results of this matching routine are given m Table A1.3
for mismatches less than 0.06 cm" . The laser frequencies used
for generating this table are the results of heterodyne frequency
measurements and are considerably more accurate than the absorption data. The laser lines used come from the following references.
GAS
BANDS
REFERENCE
c16o2
13 16
c o2
NORMAL
26
NORMAL
31
12
12
c18o2
NORMAL
31
13
c18o2
NORMAL
31
12
c16o2
»c"oa
SEQUENCE
32
HOT BAND
33
N20
NORMAL
34
106
The results contained in Table Al.3 are believed accurate
to ± 0.002 cm" .
Ground State Inversion Transition Frequencies
The ground state inversion transition frequencies are presented in Table A1.4. A more complete listing of the inversion
transitions is presented in Ref. 7.
The combination of Tables
Al.3 and A1.4 can be used to predict candidate systems for
infrared pumped maser emission from 14NH_.
107
Table A1.4
14
NH 3 Ground state inversion transition frequencies (GHz)
J
K
FREQUENCY
(GHz)
J
K
FREQUENCY
(GHz)
J
K
FREQUENCY
(GHz)
1
3
1
1
23.69449
22.23453
2
3
1
2
23.09879
22.83417
2
3
2
3
4
4
1
4
21.13429
24.13941
4
5
2
1
21.70336
19.83826
4
5
3
2
23.72263
23.87013
22.68829
20.37146
5
6
6
7
3
1
4
1
4
7
3
6
1
4
7
1
5
6
6
7
7
8
4
2
5
2
7
7
8
8
9
9
9
10
10
10
10
21.28527
18.39146
20.99461
16.84095
19.21836
25.71517
16.45513
20.71921
13.61208
15.52396
20.73544
12.01702
22.65300
18.88476
22.73243
17.29154
20.80483
15.23312
5
6
6
7
7
8
5
3
6
3
6
2
24.53298
19.75740
25.05602
18.01742
22.92494
15.63984
8
8
9
9
9
10
17.37814
23.23224
13.97454
16.79822
23.65748
12.33648
8
8
9
9
9
7
10
13.70096
18.28555
28.60473
5
8
3
6
9
3
6
9
2
18.80856
26.51891
14.37656
18.49928
27.47810
13.29637
16.31938
24.20529
10.75982
11
11
11
12
3
6
9
1
12
12
12
5
1
10
10
11
4
7
2
5
8
2
5
8
1
14.82270
20.85251
10.48173
10.53630
14.22474
11
11
4
7
11.94714
15.93332
21.07070
9.03281
11
12
10
2
4
7
10.29346
13.71951
12
12
10
21.38155
12
4
10
10
10
11
5
8
12.92310
18.16254
24.88190
9.27210
11
11
11
12
11
3
29.91466
10.83610
5
8
11.13270
15.63288
12
12
6
9
12.25146
18.12732
11
25.69523
12
12
31.42497
108
APPENDIX A2: THE INFRARED PUMPED MICROWAVE EMISSION CELL,
CONSTRUCTION AND CALIBRATION
The experimental cell used in the generation of infrared
pumped maser emission is a high Q right cylinder cavity resonator.
The resonator operates in the lowest order linearly
polarized TE,, mode for the microwaves. The low loss modes for
the FIR are the TE,Q
modes. The cell design discussed in this
appendix has been used to generate both microwave and FIR radiation.
The theory of right cylinder cavity resonators has been
the subject of many investigations.
Three references from the
same book have been used extensively in the design and calibrate "3C 3 "7
tion of the cell. '
'
The notation used in this appendix
closely follows the notation in these references. Only equations
related to the frequency calibration will be considered in detail.
A comparison of the theoretical and experimentally
measured frequency calibration curves will be presented.
(See
Refs. 16, 17 and 18 for a complete treatment of cavity resonators) .35,36,37
Theory
The formula relating the resonant frequency to the mode,
shape and dimensions of the cylinder is given in Equation (A2.1)
2
cr
2
cn
(fD)2 = l-JS)
f ITH ++ |«i|
>
.
{ 2
£
= A + Bn 2
2
Di
(A2.1)
109
f = frequency in MHz
D = diameter of cavity in inches
L = length of cavity in inches
A = mode dependent constant
= .47810xl08
B = a constant dependent upon the velocity of electromagnetic waves in the dielectric
*
= .34799xl08
n = the number of half wavelength along the cylinder
axis (third mode index)
c = speed of light
10
= 1.17981x10
r-
*
inches/second
= m
zero of J,(x) (for TM modes)
= m
zero of J '(x) (for TE modes)
= 1.84118 (TE1;L mode)
* Air at 25°C and 60% relative humidity.
The value of B (and c) was not corrected for changes in the
specified condition. The actual operating conditions for the
14
experiment were
NH 3 at pressures less than 2 torr, and negligible relative humidity.
Simple manipulation of Equation (A2.1) yields (A2.2).
The
guide wavelength may be determined as a function of frequency
110
by e v a l u a t i n g Equation (A2.3).
X (f) = 2 / 9 B ?
9
Vf2-A/D2
(A2.3)
The three equations presented provide adequate design information for the desired cell.
Once the mode and frequency range
choices are made, the only design parameters remaining are the
cavity length and cylinder inside diameter.
Since a tunable
cavity is desired, the length parameter will be variable. The
diameter of the tube is determined by use of a mode chart and
the desired frequency range characteristics. (A mode chart is
2
2
a plot of (fD) vs. (D/L) using Equation (A2.1) evaluated for
the different possible modes).
Cell Design
The linearly polarized lowest order TE,, mode was chosen
because of the relatively small mode volume (i.e., smaller mode
volume implies fewer possible modes).
The choice of the tube
diameter was based on the cutoff frequency for the next lowest
order mode which is the TM nl .
The cutoff wavelength is given
by Equation (A2.4).
TE
^
\
TM
X = -SP° rAm
m
C
= —2- D
r
*m
,
(A2.4)
Ill
For a .375 inch ID tube the cutoff frequency for the TE,,
mode is 18.462 GHz while for the TMQ, mode cutoff frequency
is 24.116 GHz. This result guarantees TE,, mode operation
from 19 to 24 GHz.
The diameter chosen is a standard size available in OFHC
copper.
It was decided that a combination fixed mirror and a
tunable plunger would be used to complete the component parts
of the cavity.
The fixed mirror is a circular copper mirror with a 1.5 mm
hole at its center (item B in Figure A2-1).
The thickness of
the mirror 1.5 mm hole is .1 inches, to reactively attenuate
any microwave radiation leaking through the hole. A half-wave
choke joint is formed at the mirror-tube interface to maintain
a high Q.
(The detail in Figure A2-1 is not correct, the actual
mirror-tube interface occurs at the quarter guide wavelength
point. An additional quarter-wave slot is machined into the
mirror) . The guide wavelength is the coaxial v/aveguide TEM
mode (i.e., free space) wavelength at a frequency of 22.38 GHz.
The tuning plunger was designed using the radial choke
38
design charts by Reed.
The lack of any explanatory informa-
tion regarding the charts produced a design that is operational
but far from optimal. The mistake made in this design resulted
from misleading figure with the associated chart. The same
error can be observed by looking at the tunable choke plunger
^
IT
J1.
c
=r--Ui
—f*H{<«|
f^i
'•
t.//\
A.
3/8" ID OFHC COFPER TUBE
B.
1.5mm HOLE-COUPLED COPPSR MIRROR
C.
COPPER TUNING PLUNGER
D.
3/32" DIAMETER COUPLING HOLE
S.
ZnSe BREWSTER WINDOW
GOLD-PLATED BRASS MIRROR WITH i'a"
MINOR AXIS DIAMETER ELLIPTICAL
HOLE (8 = 45°)
TPX PLASTIC VACUUM WINDOW
Figure A2-1.
Tunable transmission cavity resonator operating in the linearly
polarized TE,, mode.
M
113
(C) in Figure A2-1, and is shown in more detail in Figure A2-2.
COPPER TUBE
.
B
A
•1
, i
C
MICROMETER SHi
A*
"
Figure A2-2. Tunable plunger.
The fundamental principle in a choke design is that it is
easier to short circuit a high impedance than to short circuit
a low impedance. This implies that discontinuities should be
placed at minimum current points, which is the case that occurs
at odd multiples of X /4 away from a short circuit.
In the de-
signed plunger shown in Figure A2-2, the short occurs at point
(C) which is the end of a quarter wavelength section of radial
waveguide, excited by the coaxial guide formed by the plunger
and the copper tube inner wall.
The design charts assume that
the impedance in the coaxial section formed by A' is greater
than the impedance m
the A section.
The designed choke plunger
has equal impedances for the A and A1 section of coaxial
114
waveguide.
This problem could be resolved by increasing A'
to near contact with the copper tube inner wall.
The result
of this error is observation of a lower Q (i.e., ^ 1000)
spurious mode detuned approximately 10 MHz from the principal
mode (the shift and coupling to this mode is a strong function
of frequency.
The maximum coupling observed to this mode is
less than 10% of the principal mode coupling).
The coupling to the transmission cavity is accomplished
by drilling a 3/32" hole through the copper tube and soldering
a short section of copper WR-42 waveguide with a choke flange
to the outside of the tube. The thick-walled tube was machined
to produce a flat short circuit for the rectangular waveguide
and a relatively t h m wall, for the hole drilled in the center
of the resulting flat short circuit.
The resulting cavity has a Q of approximately 3000 with a
coupling near 10%. The cavity designed and produced the desired experimental results, but does not represent the optimal
design.
The remaining components shown in Figure A2-1 are required
to input couple the infrared pump radiation and output couple
the far infrared radiation.
The infrared pump radiation is
focused in through the ZnSe window, the 1/8" (minor axis diameter ) elliptical hole in the gold plated brass mirror, and
finally the 1.5 mm diameter hole in the copper mirror.
The FIR
115
low loss mode radiation is coupled out of the 1.5 mm hole and
deflected, using the gold plated brass mirror, through the TPX
plastic vacuum window to a detector.
Comparison of Theoretical and Experimental Frequency Calibration
Curves
The first comparison of calculated and experimental guide
wavelength produced agreement to within 3%.
was not expected.
to be in error.
The large deviation
The copper tube diameter was then suspected
The guide wavelength measurements could easily
be measured to .001 inches with the micrometer driven tuning
plunger.
(No appreciable micrometer backlash was observed
during the measurements of guide wavelength).
The frequencies
were measured to ±2 MHz using considerable care and a direct
reading cavity wavemeter.
The decision was made to use the measured frequencies and
guide wavelengths to determine the cavity mside diameter using
Equation (A2.3). The results are shown graphically in Figure
A2-3.
The average diameter obtained from the measured values
was found to be approximately 1.3% smaller than the .375"
nominal inside diameter of the tube. The maximum deviation
obtained using the averaged diameter is less than 1% for the
calculated and measured values.
Figure A2-4 is a different form of the cavity mode chart.
This format of presentation aids in locating resonance frequencies and approximate micrometer settings.
The solid dots
116
1.10 r
.00
B
^ • 2 V 7 d A/D*
*
+
.99
*
.98
""
+
97
,„,
co
UI
X
96
z
95
X
.94
o
z
UI
-i
UJ
>
<
f B AD -
+
+
•
-
+
*+
•
+
93
92
FREQUENCY
MODE CONSTANT-0.34799x10*
MODE CONSTANT » 0.47810 x IO8
03702 IN
-
3
Ui
Q
3
<9
91
90
+
•
MEASURED
+
CALCULATED
89
88
-
87
1
86
22 0 .1
•
4
.5
6
.7
.8
J
I
.9
230
l
FREQUENCY (GHz)
Figure A2-3. Measured and calculated TE
guided wavelength
versus frequency for the designed cavity.
n=65
n = 64
4
5
6
7
8
9 10 I I
MICROMETER SETTING (INCHES)
J.
304 5
Figure A2-4.
8
9
a.
310 I
CAVITY LENGTH (INCHES)
_i_
12
-l_
13
_L
8
14
_i_
IS
-J-
J
9 320 I
TE,, cavity resonant frequency versus cavity length/micrometer setting
for different n values, measured (•) and calculated (+).
118
were measured and the crosses are calculated using the averaged
diameter of .3702 inches in Equation(A2.2).
The measured and
calculated results were overlayed and shifted back and forth
until the "best" overlap was observed.
The calibration be-
tween the absolute cell length and the micrometer setting is
now known, since the mode index n is known (i.e., L , =
abs
H(nXg(f))
.
Absolute Frequency Calibration Using Microwave Absorption
The absolute frequency calibration of the cell was desired
to help minimize the experimentally measured frequency errors.
14
This was accomplished using
NH 3 microwave absorption fre-
quencies, thus improving the frequency certainty from ±2 MHz
to ±10 KHz. These results are contained in Table A2.1. The
technique used to measured the cavity resonance lengths was
discussed in the cavity scans section of the main text, and will
not be repeated here.
Three (and sometimes four) consecutive
dips at different cavity settings were observed corresponding
to an absorption.
The cavity wavemeter was used to verify the
approximate frequency.
The distance between two consecutive
dips represents a half guide wavelength move.
The guide wave-
length measurement experimental error is further reduced by
measuring the distance between the first and third or second
and fourth dips.
The measured guide wavelengths were then
used to calculate the diameter of the copper tube. The scatter
119
Table A2.1. Microwave absorption calibration measurements
and resulting tube diameter calculations.
G:sQ(5,4)
Frequency
GHz
22.65300
Guide
Wavelength
(Inches)
.9204
G:sQ(5,4)
22.68829
.9158
.3702
G:sQ(6,5)
22.73243
.9103
.3702
.9108
.3701
.8972
.3704
.8973
.3704
.8869
.3703
.8870
.3703
.8108
.3707
.8102
.3709
.8094
.3701
.8092
.3702
.8062
.3703
.8065
,3703
.7932
.3704
.7933
.3703
Transition
G:SQ(3,2)
G:sQ(7,6)
G:sQ(9,8)
G:sQ(l,l)
G:sQ(2,2)
G:sQ(3,3)
22.83417
22.92494
23.65748
23.69449
23.72263
23.870130
Tube Diameter
(Inches)
.3702
120
obtained in these calculations is considerably smaller than
the results shown in Figure A2-4. The resulting average value
is .37033 inches, which should be used in any further calculations.
The data presented in the m a m body of text used this
value.
Conclusions
The cell frequency-length characteristics can be calculated
with a high degree of accuracy using Equations (A2.1), (A2.2) and
(A2.3).
The parameters used are:
A = .47810X108
B = .34799X108
D = .37033 inches
30.4 S L £ 32.1 inches
f = MHz
The results of the frequency-cavity length mode plots indicate a relatively linear relationship between frequency and
cavity length.
The slope of the frequency tuning versus cavity
length changes with the third mode index (i.e., n ) . These two
results are important for the cavity scan results presented in
the text.
Suggested Improvements
The cell design presented here has been used to produce
optically pumped maser emission.
Optimal design of the cell
121
has not been achieved.
Several suggested modifications are
recommended for improvement.
The problems associated with the design of the tuning
plunger have been mentioned in detail. The recommended improvmeent would entail redesign of the tuning plunger. The
radial line choke design would be abandoned for a series of
quarter wavelength alternate coaxial low impedance - high
impedance sections.
The merits and efficiency of this type
of plunger has been considered by Harvey 39 but no reference
regarding design was located.
The quarter wavelength sections
must be compensated for the resulting additional junction capacitances.
The method employed by Young4^ should be of great
help to determine the length corrections required to compensate
for this capacitance. This would reduce the spurious mode problem and definitely increase the cavity Q.
The thick walled copper tubing should be replaced with
precision bore tubing for two reasons. The cell could be designed and calibrated much more efficiently.
The inside dia-
meter of the tubing used was not concentric with the outside
diameter.
This must have produced some problems, but the re-
sults of these problems are not apparent.
The most confusing design details for the cavity regard
the cavity coupling to the external circuit.
The design of
a tunable cavity further complicates the situation.
sign used by Willenberg
The de-
incorporates two moveable tuning
122
plungers.
The advantage of this design feature is increased
bandwidth coupling, since the standing wave can be moved to
more effectively match the current lines coupling the cavity
to the external circuit.
Optimal coupling is achieved by
matching the current lines in the cavity with the external
circuit.
The resulting coupling produces the smallest reduction
of Q for a given coupling.
The coupling used in this cell, makes microwave probing
of the interaction very difficult.
Changing the coupling to
the critical point would simplify any probe results by increasing the transmitted signal-to-noise ratio. The increased
coupling would undoubtedly decrease the level of saturation
observed in the microwave emission, if it does not suppress
the emission completely.
Adjustment of the coupling level may
be a useful method for characterizing the g a m associated with
the microwave emission.
The designed cavity resonator has produced the desired experimental results. The suggested improvements would further
refine the experimental apparatus but would require a considerable amount of redesign and microwave engineering.
123
LIST OF REFERENCES
K. J. Kim and P. D. Coleman, "Calculated-Experimental
Evaluation of the Gain/Absorption Spectra of Several
Optically Pumped NH_ System", submitted to IEEE J. Quant.
Elec, (1980).
H. R. Fetterman, H. R. Schlossberg and J. Waldman, "Submillimeter Lasers Optically Pumped Off Resonance", Opt.
Commun., 6(2), 156 (1972).
K. Gullberg, B. Hartmann and B. Kleman, "Submillimeter
Emission from Optically Pumped I^NHJ ", Physica Scripta.,
8, 177 (1973).
T. Y. Chang and J. D. McGee, "Off-Resonant Infrared Laser
Action in NH_ and CJH. Without Population Inversion", Appl.
Phys. Lett., 29(11), 725 (1976).
T. Yoshida, N. Yamabayashi, K. Miyazaki and K. Fujisawa,
"Infrared and Far-Infrared Laser Emissions from a T.E. CO~
Pumped NH 3 Gas", Opt. Commun., 26(3), 410 (1978).
S. M. Freund and T. Oka, "Infrared-Mcirwave Two-Photon
Spectroscopy: The v_ Band of NH~",
Phys. Rev. A., 13(6),
J
2178 (1976) .
**
M. S. Cord, J. D. Petersen, M. S. Lojko, R. H. Haas, MicroSpectral Tables, Volume IV. Polyatomic Molecules Without
Internal Rotation, (National Bureau of Standards Monograph
70, Volume IV, 1968), p. 369.
G. D. Willenberg, C. 0. Weiss, H. Jones, "Two-Photon Pumped
CW Laser", submitted to Phys. Rev. Lett. (1979).
C. H. Townes and A. L. Schawlow, Microwave Spectroscopy,
(McGraw-Hill, New York, 1955).
K. K. Clarke and D. T. Hess, "Mixers: RF and IF Amplifiers",
in Communication Circuits: Analysis and Design, (AddisonWesley, Massachusetts, 1971).
H. R. Fetterman, "Real-Time Spectral Analysis for FIR Laser
Pulses", Appl. Phys. Lett., 34, 123 (1979).
R. Miller, Private Communication.
124
P. D. Coleman, "Proposal on Applications of SAW Devices to
the Study of the Spectral Characteristics of Near Millimeter
Wave Sources and Systems", Prepared for U.S. Army Research
Office, University of Illinois, 1979.
R. H. Pantel and H. E. Puthoff, Fundamentals of Quantum
Electronics, (John Wiley and Sons, Inc., New York, 1969).
N. Bloembergen, Nonlinear Optics, (W. A. Benjamin, New York,
1965).
A. Yariv, Quantum Electronics, Second Edition, (John Wiley
and Sons, Inc., New York, 1975).
E. J. Danielewicz, E. G. Malk, and P. D. Coleman, "High
Power Vibration-Rotation Emission from 1 4 NH 3 Optically
Pumped Off Resonance", Appl. Phys. Lett, 29, 557 (1976).
S. J. Petuchowski, A. T. Rqsenberger, and T. A. DeTemple,
"Stimulated Raman Emission in Infrared Excited Gases",
IEEE J. Quant. Elec, QE-13, 476 (1977).
D. Cotter, D. C. Hanna, W.H.W. Tuttlebee and M. A. Yuratich,
"Stimulated Hyper-Raman Emission from Sodium Vapor", Opt.
Comm., 22, 190 (1977).
K. J. Kim and P. D. Coleman, "Stimulated Hyper-Raman
Scattering m a Molecular Gas via a Three-Photon Process
to Obtain IR and FIR Radiation", to appear in March 1980
issue of IEEE J. Quant. Elec.
P. D. Coleman, "Hyper-Raman Scattering and Related Phenomenon ", Proceedings of the International Conference on
Lasers '78, 93 (1978) .
S. J. Petuchowski, J. D. Oberstar and T. A. DeTemple,
"Optical Triple Resonance", Phys. Rev. A, 20(2), 529 (1979).
H. Jones, "Wide-Band Intracavity Microwave Cells for LaserMicrowave Double Resonance Spectroscopy", Appl. Phys., 14,
169 (1977) .
E. J. Danielewicz, "Far Infrared Guided Wave Optics Experiments Using a Waveguide Laser with a Hybrid Output Mirror",
Ph.D. Thesis, University of Illinois, Urbana, Illinois,
1976.
25.
125
J. Curtis, "Vibration-Rotation Bands of NH., in the Region
-1
-1
670 cm -1860 cm ", Ph.D. Dissertation, The Ohio State
University, 1974.
26.
F. R. Petersen, G. D. McDonald, J. D. Culp, and B. L.
Danielson, "Accurate Rotational Constants, Frequencies,
and Wavelengths from 12C160~ Laser Stabilized by Saturated
Absorption", in Laser Spectroscopy, R. G. Brewer and A.
Mooradian, eds., (Plenum, New York, 1977) Vol. 22, pp. 261264.
27.
G. H. Herzberg, Infrared and Raman Spectra of Polyatomic
Molecules, (Van Nostrand, New York, 1945) pp. 221-224.
28.
T. Oka, "Infrared and Radiofrequency Spectroscopy in the
Laser Cavity", in Frontiers in Laser Spectroscopy, Balian,
et al., eds., (North-Holland Publishing Company, 1977).
29.
J. W. Niessen, "A Phosphine Far Infrared Laser Optically
Pumped by a Transverse Excitation Atmospheric Pressure
Carbon Dioxide Laser", Masters Thesis, University of
Illinois, 1978.
30.
W. K. Bischel, P. J. Kelly, C. K. Rhodes, "High Resolution Doppler-Free Two-Photon Spectroscopic Studies of
Molecules II. The v 9 Band of -UNH J " Phys. Rev. A. 13(5)
1829 (1976) .
**
31.
C. Freed, D. L. Spears, and R. G. O'Donnell, "Precision
Heterodyne Calibration", in Laser Spectroscopy, R. G.
Brewer and A. Mooradian, eds., (Plenum, New York, 1974)
pp. 171-191.
32.
K. J. Siemsen and B. G. Whitford, "Heterodyne Frequency
Measurements of CO2 Laser Sequence-Band Transitions", Opt.
Commun., 22, 11 (1977).
33.
B. G. Whitford, K. J. Siemsen and J. Reid, "Heterodyne
Frequency Measurements of CO2 Laser Hot-Band Transitions",
Opt. Commun., 22, 261 (1977).
34.
B. G. Whitford, K. J. Siemsen, H. D. Riccius, and G. R.
Hanes, "Absolute Frequency Measurements of N2O Laser
Transitions", Opt. Commun., 14, 70 (1975).
126
I. G. Wilson, C. W. Schramm, and J. P. Kmzer, "High Q
Resonant Cavities for Microwave Testing", Radar Systems
and Components, M.T.S. of Bell Telephone Laboratories,
(Van Nostrand, New York, 1949) pp. 909-935.
J. P. Kmzer and I. G. Wilson, "End Plate and Side Wall
Currents in Circular Cylinder Cavity Resonator", Radar
Systems and Components, M.T.S. of Bell Telephone Laboratories, (Van Nostand, New York, 1949) pp. 936-984.
J. P. Kinzer and I. G. Wilson, "Some Results on Cylindrical Cavity Resonators", Radar Systems and Components, M.T.S.
of Bell Telephone Laboratories, (Van Nostrand, New York,
1949) pp. 985-1020.
J. Reed, "Radial Line Choke Design Charts", Microwave
Journal Microwave Engineers Technical and Buyers Guide
Edition, 29 (1968).
A. F. Harvey, Microwave Engineering, (Academic Press, New
York, 1963).
L. Young, "Practical Design of a Wide-Band Quarter-Wave
Transformer in Waveguide", Microwave Journal, 6, 76 (1963).
127
VITA
Edward George Malk was born in Long Beach, CA on December 14, 1948. He received a B.S. degree (1975) and his M.S.
degree (1977) from the University of Illinois, ChampaignUrbana, Illinois. He served on the staff of the Electrical
Engineering Department as a Research Assistant in the ElectroPhysics Laboratory of the University of Illinois since 1975.
His publications include:
"Hot Band Lasing in NH 3 " (1979),
"Laser Emission in the 83-223 ym Region from PH3 with
Laser Line Assignments" (1978),
"High Power Vibration-Rotation Emission from
NH3 Opti-
cally Pumped Off Resonance" (1976).
His work entitled, "Two-Photon Pumping of a Four-Level
System in Ammonia to Obtain 12.16 ym Radiation for Isotope
Separation" was presented at the 1978 IEEE MTT International
Microwave Symposium, Ottawa, Canada, June 1978.
"Laser Emis-
sion in the 83-223 ym Region from PH3 with Laser Line Assignments" was presented at the Third International Submillimeter
Wave Conference, Guilford, England, March 1978.
He is a member of Sigma Xi and Eta Kappa Nu.
Документ
Категория
Без категории
Просмотров
0
Размер файла
3 973 Кб
Теги
sdewsdweddes
1/--страниц
Пожаловаться на содержимое документа