United States
Environmental Protection
Agency
Environmental Sciences Research
Laboratory
Research Triangle Park NC 27711
EPA-600 2-80-049
February 1980
Reeeercri and Development
Remote Monitoring of
Gaseous Pollutants by
Differential
Absorption Laser
Techniques
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the ENVIRONMENTAL PROTECTION TECH-
NOLOGY series. This series describes research performed to develop and dem-
onstrate instrumentation, equipment, and methodology to repair or prevent en-
vironmental degradation from point and non-point sources of pollution. This work
provides the new or improved technology required for the control and treatment
of pollution sources to meet environmental quality standards.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/2-80-049
February 1980
REMOTE MONITORING OF GASEOUS POLLUTANTS
BY DIFFERENTIAL ABSORPTION LASER TECHNIQUES
by
S. A. Ahmed, J. S. Gergely, and F. Barone
Electrical Engineering Department
The City College of the City University
140th Street and Convent Avenue
New York, New York 10031
Grant No. 803109
Project Officer
William F. Herget
Emissions Measurement and Characterization Division
Environmental Sciences Research Laboratory
Research Triangle Park, N.C. 27711
ENVIRONMENTAL SCIENCES RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK N.C. 27711
-------
DISCLAIMER
This report has been reviewed by the Environmental Sciences Research
Laboratory, U.S. Environmental Protection Agency, and approved for pub-
lication. Approval does not signify that the contents necessarily re-
flect the views and policies of the U.S. Environmental Protection Agency,
nor does mention of trade names or commercial products constitute
endorsement or recommendation for use.
IX
-------
ABSTRACT
A single-ended laser radar (LIDAR) system was designed, built, and
successfully operated to measure range-resolved concentrations of N02»
and 03 in the atmosphere using a Differential Absorption of Scattered
Energy (DASE) LIDAR technique. The system used a flash-lamp pumped dye
laser as the primary source of laser energy. For the N02 measurements, the
dye laser output was used directly in a novel simultaneous two wavelength
output mode in which two wavelengths, one on and one off the resonance
absorption of N02 molecules are transmitted simultaneously and the relative
attenuation determined for the two backscattered signals detected. This
mode of operation effectively reduces errors due to scintillation and
aerosol drift. For the S02 and 03 measurements, it was necessary to
frequency double the output of the dye laser to match the absorption spectra
of the SC»2 and 03 molecules. Field measurements, which were carried out
over the Upper East Side of Manhattan for all three pollutants, produced
range-resolved concentrations at ranges of over two kilometers. The ambient
pollutant concentrations measured ranged from 0.04 ppm to 0.31 ppm, depending
on location and time of day. In general, these showed reasonably good
correlation with measurements obtained from conventional pollution monitoring
stations in the area and demonstrated the potential of DASE LIDAR systems
for range-resolved ambient pollutant measurements.
iii
-------
CONTENTS
Abstract iii
Figures , f vi
Tables xiii
1. Introduction and Background 1
2. Lidar Techniques 10
3. The Lidar System 45
4. N02 Measurements 118
5. Simultaneous Multiwavelength Outputs From
Energy-Transfer Dye-Mixture Lasers 146
6. Dase Lidar System Applied to Ozone
and Sulfur Dioxide Measurements 180
7. Summary 201
References 206
-------
LIST OF FIGURES
No. Caption page
2.2:1 The Raman Process 12
2.4:1 H02 Absorption Curve 20
2.4:2 Obtainable Depth Resolution for Backscatter LIDAR
Schemes 21
2.4:3 The BASE Scheme 23
2.4:4 The Resonance Fluorescence Scheme 31
2.4:5 The Raman Scheme 35
2.4:6 The Long Path Absorption Scheme 39
3.1:la The LIDAR System for Sequential Operation 46
3.1:lb Photographs of the LIDAR System for Simultaneous
Operation 47
3.2:1 Fresnel Lens 48
3.2:2a The Optical Receiver System for Simultaneous
Operation 51
3.2:2b Photographs of the Optical Receiver System For
Simultaneous Operation 52
3.3:la Coaxial Arrangement of the Flaahlamp-Pumped Dye-Laser
Head 53
3.3:lb Photographs of the Laser Head Situated in the LIDAR
System 54
3.3:2 Typical Flashlamp-Pumped Dye-Laser Output Pulse 55
3.3:3 Energy Levels of Organic Dye Molecules 57
3.3:4 Dispersion of Grating and Effective Aperture Created
by the Laser Head 59
3.4:1 The Overall Transmitting System with Calibration
Setup for Sequential Operation 61
3.5:1 Beam Spreading 63
3.5:2 Beam Steering 63
VI
-------
3.5:3 Beam Scintillation 64
3.5:4a Optical Arrangement for Simultaneous Two-Wavelength
Output 66
3.5:4b Photographs of the Optical Arrangement for
Simultaneous Two-Wavelength Output 67
3.5:5 Dielectric-Interface Polarizing Beam-Splitting Cube 69
3.5:6 Spectrometer Photograph of the Simultaneous Two-
Wavelength Output Utilizing the Dielectric-Interface
Polarizing Beam*Splitting Cube 70
3.5:7 Air-Spaced Glan-Taylor Prism 72
3.5:8 Spectrometer Photograph of the Simultaneous Two-
Wavelength Output Utilizing the Air-Spaced Glan-
Taylor Prism 74
3.5:9 Spectrometer Photograph of the Simultaneous Two-
Wavelength Output Utilizing the Air-Spaced Glan-
Taylor Prism for Equal Output Energies in Each Lasing
Line 76
3.6:la Experimental Setup for Monitoring Laser Frequency
Shifts and Bandwidth Changes 78
3.6:lb Photographs of the Experimental Setup for Monitoring
Laser Frequency Shifts and Bandwidth Changes 79
3.6:2 Spectrometer Photograph of the 1478.5 A Laser Line
for Sequential Operation 81
3.6:3 Spectrometer Photograph of the 4478.5 A Laser Line
for Simultaneous Two-Wavelength Operation 81
Q
3.6:4 Spectrometer Photograph of the 4500 A Laser Line for
Sequential Operation 82
Q
3.6:5 Spectrometer Photograph of the 4500 A Laser Line for
Simultaneous Two-Wavelength Operation 82
3.6:6a Spectrometer Photograph of the Simultaneous Two-
Wavelength Output, Bias Voltage = 20 KV, First Shot 83
3.6:6b Bias Voltage * 20 KV, Second Shot 83
3.6:6c Bias Voltage = 20 KV, Third Shot 84
3.6:6d Bias Voltage = 20 KV, Fourth Shot 84
Vll
-------
3.6:6e Bias Voltage - 19 KV, First Shot 85
3.6:6f Bias Voltage * 19 KV, Second Shot 85
3.6:6g Bias Voltage = 19 KV, Third Shot 86
3.6s6h Bias Voltage = 19 KV, Fourth Shot 86
3.6:6i Bias Voltage = 18 KV, First Shot 87
3.6:6j Bias Voltage » 18 KV, Second Shot 87
3.6:6k Bias Voltage * 18 KV, Third Shot 88
3.6:61 Bias Voltage = 18 KV, Fourth Shot 88
3.6:7a The Calibration Setup 90
3.6:7b Photographs of the Calibration Setup 91
3.6:8a Calibration Signals for Simultaneous Two-Wavelength
Operation for the 4478.5 A Laser Line with N2 in the
Calibration Cell 95
3.6:8b Calibration Signals f oroSimultaneous Two-Wavelength
Operation for the 4500 A Laser Line with N2 in the
Calibration Cell 96
3.6:9a Calibration Signals for Simultaneous Two-Wavelength
Operation for the 4478.5 A Laser Line with N02-»2
Mixture in the Calibration Cell 97
3.6:9b Calibration Signals foroSimultaneous Two-Wavelength
Operation for the 4500 A Laser Line with N02-N2
Mixture in the Calibration Cell 98
3.6:10 Experimental Setup Used for Fine Frequency Tuning 101
3.6:lla Spectrometer Photograph of the Simultaneous Two-
Wavelength Output to Get 5 A Bandwidths in Each
Line—4478.5 A 104
3.6:lib —4500 A 104
3.6:llc —the 4478.5 K and 4500 A Laser Lines 105
3.6:lld —the 4478.5 A and 4500 A* Laser Lines 105
3.6:lie —the 4478.5 A and 4500 A Laser Lines 106
3.6:llf —the 4478.5 A and 4500 A Laser Lines 106
vm
-------
3.6:llg —the 4478.5 A and 4500 A Laser Lines 107
3.6:llh —the 4478.5 A and 4500 A Laser Lines 107
3.6:lli —the 4478.5 A and 4500 A Laser Lines 108
3.6:12a The Initial Fine Tuning Signals for Simultaneous
Two-Wavelength Operation for Obtaining Frequency
Coincidence for the 4478.5 A Laser Line 109
3.6:12b The Final Fine Tuning Signals for Simultaneous Two-
Wavelength Operation for Obtaining Frequency
Coincidence for the 4478.5 A Laser Line 110
3.6:13a The Initial Fine Tuning Signals for Simultaneous
Two-Wavelength Operation for Obtaining Frequency
Coincidence for the 4500 A Laser Line 111
3.6:13b The Final Fine Tuning Signals for Simultaneous Two-
Wavelength Operation for Obtaining Frequency
Coincidence for the 4500 A Laser Line 112
3.6:14 Spectrometer Photograph of the Simultaneous Two-
Wavelength tOutput with Laser Frequencies Fine Tuned
to 4478.5 A and 4500 A with Approximately 5 A
Bandwidths * 113
3.6:15 Spec. Sheet 3.6:1 114
4.2:1 Typical LIDAR Return Signal 120
4.2:2 Typical LIDAR Return Signal with Chimneys Below Beam
Path 120
4.2:3a Sequential LIDAR Return Signals at 4478.5 A and
4500 A 121
4.2:3b Sequential Calibration Signals at 4478.5 A and
4500 A 122
4.2:4a Simultaneous LIDAR Return Signals at 4478.5 A and
4500 A, Time: 8:45 A. M. 125
"+.2:4b Simultaneous Calibration Signals for 4478.5 A and
4500 A, Time: 8:45 A. M. 126
4.2:4c Simultaneous LIDAR Return Signals at 4478.5 A and
4500 A, Time: 9:00 A. M. 127
4.2:4d Simultaneous Calibration Signals for 4478.5 A and
4500 A, Time: 9:00 A. M. 128
IX
-------
4.2:4e Simultaneous LIDAR Return Signals at 4478.5 A and
4500 A, Time: 9:30 A. M. 129
4.2:4f Simultaneous Calibration Signals for 4478.5 A and
4500 X, Time: 9:30 Ai M. !30
4.2:4g Simultaneous LIDAR Return Signals at 1478.5 A and
4500 A, Time: 5:30 P. M. 131
4.2:4h Simultaneous Calibration Signals for 4478.5 A and
4500 A, Time: 5:30 P. M*- 132
4.2:4i Simultaneous LIDAR Return Signals at 4478.5 K and
4500 A, Time: 6:00 P. M. 133
4.2:4j Simultaneous Calibration Signals for 4478.5 A and
4500 A, Time: 6:00 P. M. 134
4.2:4k Simul|aneous LIDAR Return Signals at 4478.5 A and
4500 A, Time: 6:30 P. M. 135
4.2:41 Simultaneous Calibration Signals for 4478.5 A and
4500 A, Time: 6:30 P. M. 136
4.2:4m Simultaneous LIDAR Return Signals at 4478.5 A and
4500 A, Time: 8:00 P. M. 137
4.2:4n Simultaneous Calibration Signals for 4478.5 A" and
4500 A, Time: 8:00 P. M. 138
4,2:4o Simultaneous Calibration Signals for 4478.5 A* and
4500 X for Normalization (with N2 in Cell) 139
4.3:1 Experimental Setup for Determination of LIDAR-
Reading Uncertainities 143
4.3:2 Error Plot for Determination of LIDAR-Reading
Uncertainties 144
5.2:1 Donor-Acceptor System of Two Organic Dyes with
Overlapping Fluorescence and Absorption Bands 148
5.2:2 Donor-Acceptor System of Four Organic Dyes with
Overlapping Fluorescence and Absorption Bands 150
5.2:3 Donor-Acceptor System with Overlapping Absorption
Bands 151
5.2:4 Donor and Acceptor Energy Levels 153
5.3:1 Absorption and Fluorescence Spectra of the Three
Dyes at Optimum Concentrations from Which
Simultaneous Laser Outputs Were Obtained in the
Three Primary Colors 155
-------
5.3:2 Absorption and Fluorescence Spectra of the Three-
Dye Mixture at Optimum Concentrations from Which
Simultaneous Laser Outputs Were Obtained in the
Three Primary Colors 157
5.3:3 Experimental Setup for Obtaining Fluorescence and
Absorption Spectra 158
5.3:4 Optical Arrangement for Obtaining Simultaneous
Three-Color Outputs in the Three Primary Colors 160
5.3:5 Absorption and Fluorescence Spectra for
Simultaneous Two-Wavelength Output 161
5.3:6 Optical Arrangement for Obtaining Two Pairs of
Wavelengths Simultaneously 162
5.4:1 Experimental Setup for Obtaining Fluorescence and
Absorption Spectra 164
5.4:2 Overlap Integral of Dichlorofluorescein (Donor)
and DODC (Acceptor) 166
5.4:3 Overlap Integral of Dichlorofluorescein (Donor)
and Rhodandne B (Acceptor) 167
5.4:4 Fluorescence Quantum Yield of DODC in the Mixture
with Dichlorofluorescein 168
5.4:5 Fluorescence Quantum Yield of Rhodamine B in the
Mixture with Dichlorofluorescein 169
5.4:6 Experimental Setup for Obtaining Fluorescence
Spectra 171
5.4:7 Fluorescence Quantum Yield of 7-Diethylamino-4
Methyl Coumarine Versus Concentration 173
5.4:8 Fluorescence Quantum Yield of 7-Diethylamino-4
Methyl Coumarine (Donor) in Mixtures with DODC
(Acceptor, 1 X 10"1* M/l) Versus Donor Concentration 174
5.4:9 Fluorescence Quantum Yield of DODC (Acceptor,
1 X ID"4 M/l) in Mixtures with 7-Diethylamino-4
Methyl Coumarine (Donor) Versus Donor Concentration 175
5.4:10 Normalized Fluorescence Quantum Yield of DODC
(Acceptor, 1 X 10"1* M/l) in Mixtures with
7-Diethylamino-4 Methyl Coumarine (Donor) with
Respect to the Fluorescence Quantum Yield of Pure
Solution of the Donor Versus Donor Concentration 176
XI
-------
5.^:11 Normalized Fluorescence Quantum Yield of DOEC
(Acceptor, 1 X ICT4 M/l) in Mixtures with
7-Diethylamino-4 Methyl Coumarine (Donor) with
Respect to the Fluorescence Quantum Yield /of the
Donor in the Same Mixture Versus Donor Concentration 177
6.1.1 Absorption coefficients of SO., and ()„ 181
6.2.1 Schematic od Lidar System for S02 and 0 183
6.3.1 Phase Matching for 90° PM 186
6.4.1 Flashlamp Adaptor 189
6.5.1 Lidar Returns, 03 and S02 192
6.5.2 Lidar Returns, 0- and S02 193
6.5.3 Lidar Returns, C>3 and SC>2 194
6.5.4 Lidar Returns, CL and S02 195
6.5.5 Lidar Returns, 0- and S02 196
6.5.6 Lidar Returns, 03 and S02 197
xii
-------
LIST OF TABLES
No. Caption Page
2.4:1 Tabulated Comparison of Monostatic LIDAR Schemes 44
4.2:1 Range, Resolution, NOj Pollution Concentration, and
Uncertainties for Various Times During the Day 140
6.1.1 S02 and 0- Absorption Coefficients
181
xm
-------
CHAPTER 1 INTRODUCTION AND BACKGROUND
1.1 The Need for Effective Pollution Monitoring
The adverse effects of pollution on health are well known, and
hence the need for effectively monitoring such pollution in heavily
populated areas is apparent.
The use of lasers for atmospheric pollution monitoring has re-
2 5 19
ceived considerable attention. • The potential advantages of
using lasers for this purpose are simple in essence.
Existing methods for monitoring air pollution utilize fixed-site
chemical stations. However, to effectively cover a large metropolitan
area, such as that of New York City, very many fixed-site stations are
required with attendant expenses. In addition, such stations only
provide information in their immediate vicinity and not in the air
between or above them.
Optimally, the intelligent planning of selective and more local-
ized pollution control measures, which would be less disruptive than
general ones, would require a three-dimensional pollution map of the
area. Moreover, it should ideally, instantaneously reflect changes in
conditions. An additional desirable goal would be the ability to track
moving clouds of pollutants.
Greatly increasing the number of fixed-site monitoring stations and
using aircraft sample collections would be prohibitively expensive
methods for attaining these goals. It is for these reasons that laser
monitoring schemes, with their potential ability to instantaneously and
remotely determine pollutant concentration and spatial distribution.
-------
1.2 Background
1.2.1 Introduction
LIDAR, an acronym for Light Detection and Ranging, was the name given to
laser ranging systems. The first recorded backscattered echoes from
the atmosphere were reported in 1963 by Ficco and Smullin, and by Ligada,
6 7
who measured backscattered signals from haze. * Since then, research
and development have concentrated on developing LIDAR into an effective
method for air pollution monitoring.
1.2.2 Raman
Raman scattering was first used to detect molecular constituents of
the atmosphere in 1967 by Leonard, Cooney, and Helfi. Later, Inaba
and Kobayashi discussed the possibility of pollution detection by Raman
scattering, and subsequently remotely detected SC>2 and C02 in the atmos-
phere.^1*^2 Unfortunately, the pollution detection sensitivity thai: the
Raman scheme offers is not adequate for trace contaminant measurements
typically found in polluted urban areas. However, considerable effort
has been directed to monitor water vapor and major atmospheric constit-
13—15
uents using the Raman scattering method.
An advantage of the Raman scheme is that a single frequency laSer
can be used as a source to obtain the Raman backscatter from the con-
stituents of interest. On the other hand, Raman schemes are greatly
limited in their attainable sensitivities by the relatively small Raman
scattering cross sections.
Using a nitrogen laser operating at 3371 X at 20 KW peak power 10 ns
halfwidth at 50 HZ, Kobayashi and Inaba (1971) have detected Raman sig-
-------
3
nals from clear air and from an oil smoke plume. The reveiving system
consisted of a 30 cm diameter telescope followed by an f/8.5 half meter
grating monochrometer and spectral interference filter. They were able
to show the presence of H20, CH^, H2S, CO, NO, and S02 in the oil smoke
plume at a distance of 30 m.
Progress in laser sources led to improvement in Raman detection
systems. Thus, in 1973 Hirschfeld described a system using a frequency
doubled Ruby laser transmitter with a 3 m aperture telescope as the
I 1C
receiver. This was used in field tests to detect C02, H20, and S02
over a 200 m range. Concentrations of approximately 10** ppm of water
vapor, 310 ppm of C02, and 300 ppm of S02 were detectable. The S02 was
dispersed in controlled amounts to create detectable plumes for monitor-
ing. The S02 concentration is of course much higher than those present
in ambient conditions.
The 2 j, 2 HZ Ruby laser source was doubled in KDP with 8 percent
efficiency and the returned signal was processed by photon counting
techniques referenced against the nitrogen system and probably repre-
17 II
sents the state of the art for sensitivity and range. '
The advent of tunable dye lasers opened possibilities for other
LIDAR pollution detection methods with inherent advantages over the
o
Raman scheme. The four LIDAR schemes for which tunable dye lasers are
particularly important and that presently appear to be the most prom-
ising for the detection of molecular pollutants with resonant absorption
in the visible and near ultraviolet are: resonance Raman, long path
absorption, resonance fluorescence, and differential absorption of
scattered energy (DASE).
-------
1,2.3 Resonance Raman
Observations of resonance Raman scattering in gases have been very
few. Bernstein (1973) measured resonant Raman enhancements in gaseous
C12, Br2, and I2« He obtained intensities of 10 to 1000 times stronger
72
than the nornal Raman effect*
Rosen (1975) estimated the feasibility of using resonance Raman
73
scattering for the remote detection of pollutants. He estimated that
for 10 pulses of 0.05 raj each, 100 ppm of N02 and 10 ppm of S02 could
be detected at 1 km at night, and 100 ppm of NO could be detected during
daytime. These are for a signal to noise ratio of 10* These concen-
trations are 3 to U orders of magnitude higher than ambient pollution
concentrations encountered.
Long Path Absorption
Hanst (196S) described the measurement of average pollutant concen-
tration along a laser beam path by laser resonance absorption. *
More recently, Zaromb (H69), Nakahara and Ito (1970), Inomota and
Igarashi (1972), and Hodgeson, HcClenney and Hanst (1973), suggested the
22-25
long path absorption scheme for determining average pollutant densities.
Kidal and Byer have given a detailed analysis of the doubled-ended long
2
path absorption method. This analysis was later extended to the single
26
ended absorption method using topographical targets by Byer and Garbuny.
To determine the average pollutant density along a laser beam path,
absorption measurements have been carried out using low power diode
27—29
lasers by Hinkley (1970). Using a pbo.HSn0.12Te diode laser
mounted onto a cold- finger of a cryogenic Dewar, concentrations ranging
-------
from 74 to 1000 ppm of C2H^ in N2 were detected in a laboratory in a
sample chamber 30 cm long. Similar experiments were carried out by
Snowman using a C02 laser.
More recently, O'Shea and Dodge (197»0 measured N02 absorption
31
coefficients for the most prominent argon-ion laser lines. Measure-
ments using the long path absorption of argon-ion lines by N02 in an
urban atmosphere were carried out. A corner cube reflector was posi-
tioned 3.5H km from the laser site to reflect the laser signal back to
the receiver. Integrated pollutant concentrations ranging from 0.05
ppm to 0.15 ppm of N02 were detected.
The topographical single ended absorption method was first demon-
*»*)
strated by Henningsen, Garbuny and Byer in 1974. The transmitter
consisted of a CaLaSOAP:Nd laser source electro-optically Q-switched
and operated in a T£MQQ mode with 10 mj/pulse output energy at 1.06 um.
The 1.06 um output from the laser was frequency doubled in a 2 cm
long CDA crystal with an efficiency of 25 percent. The output from the
frequency doubler was focused into a temperature-tuned LiNbOs parametric
oscillator. An etalon within the laser cavity was fine tuned piezo-
electrically which allowed the infrared frequency to be tuned on and
off a resonance peak of CO.
Using a wooded area as a topographical target a concentration of
47 ppm of CO was detected at a range of 107 m. The CO was released
from containers situated below the laser beam path.
1.2.5 Resonance Fluorescence
In 1969, Bowman, Gibson and'Sandford used a flashlamp-pumped dye
-------
laser tuned by a tilted Fabry-Perot interferometer to obtain nighttime
33
measurements of sodium in the upper atmosphere. The output of the
dye laser was used to induce fluorescence from sodium at altitudes up
to.150 km. Concentrations ranging from l.t X 101** m~2 to 9 X 1012 m'2
were detected at altitudes of 10 to 100 km.
By 1970, regular measurements were being made of the sodium double
layer at 90 km altitude. Since that time, observations of the sodium
layer have been numerous.35"39 In 1972, Gibson and Sandford extended
the use of the resonance fluorescence scheme into daytime operation by
V J *• 1 i W.H1
using a narrow band optical receiver. *
The use of resonance fluorescence backscattering for probing the
troposphere has been limited. However, the method has been extensively
2 W
analyzed by Kidal and Byer, and by C.M. Penny. ' Resonance backscat-
tering also has been discussed by ourselves, Kobayashi and Inaba, and
lib 3 U^
Measures and Pilon. • »
In 1972, Gelbwachs utilized resonance fluorescence of K02 excited by
an argon-ion laser to detect N02 locally with a sensitivity of one part
H-5 9
per billion. The laser excitation was at 1110 A and the fluorescence
o
was monitored at 7000 to 8000 A. Local Los Angeles air, drawn through
filters, was monitored, and»N02 variations measured over a period of a
few hours showed concentrations ranging from 0.03 ppm to 0.1 ppm. Fil-
tering was necessary to eliminate interferring fluorescence from partic-
ulates in the air.
The increased sensitivity of resonance fluorescence backscattering
makes it appear useful as a remote monitoring method compared to the
Raman method. As will be shown later, the sensitivity of the resonance
-------
5
fluorescence scheme is greatly surpassed by| the differential absorption
i
of scattered energy method (DASE).
1.2.i Differential Absorption of Scattered Energy (DASE)
The DASE method was first suggested by Schotland in 1964 using a
searchlight as a source. The method was recently extended to LIDAR
47 50 2
pollution monitoring by Ahmed and Gergely, Igarashi, and others. ' * *
26 ,43,41,49
The earliest experimental work was performed by Schotland who used
a temperature tuned Ruby laser to measure the water vapor vertical pro-
46
file by the DASE method* DASE measurements in the atmosphere were
first reported by ourselves and by Igarashi. * DASE measurements
also were later reported by Rothe, Brinkman and Walther of Cologne,
51 eg
Germany. * Calibration measurements of the differential absorption
52
method were made by Grant et al. Using a flashlamp-pumped dye laser
operating near 4450 A at 4 to I mj pulse energy, they measured N02 of
known concentration in a sample chamber.
The DASE system with its superior operating capabilities is the
topic of interest in this work and a detailed theoretical comparison
V.iX "
between the aforementioned LIDAR schemes ts given in Chp. 2.
1.3 Summary and'Outline of This Study
The tasks accomplished are summarized below:
1. The basic laser radar system was designed built and tested. It
consists of a flashlamp-pumped dye laser and an optical receiver
with a 38cm. HDresnel lens.
-------
2. The optical arrangements of the dye laser system were designed
for two alternate modes of operation. In one mode, outputs at
the two required wavelengths, on and off the resonance absorp-
' ',• *- •*,
tion of,the pollutant, are obtainable in rapid sequence, with
•'-. i •;' "''"'"
frequency selection by an electromagnetically driven grating.
In the second mode of operation, an intracavity polarizing-
prism bean splitter is used to obtain simultaneously a pair
of close-lying independently tunable wavelengths.
3. Field tests of the LIDAR system were carried out from City
College and extended over the upper East Side of Manhattan.
In these tests ambient levels of NC-2 concentrations were meas-
ured. These measurements showed that ambient levels below 0.2
ppm could readily be detected at ranges of more than 1 km.
4. The LIDAR system makes use of a calibration cell containing
the pollutant gas to be monitored, so that a calibration signal
i
is automatically available with each outgoing pulse. This
makes the system independent of the slight variations in laser
output characteristics that occur from pulse to pulse.
5. To improve flashlamp-pump laser output and efficiency, and to
obtain simultaneously laser outputs at wavelengths suitable for
monitoring N02 and other pollutants whose absorptions lie within
the visible band, experiments were also carried out on energy
transfer processes in dye mixtures. In this work, simultaneous
outputs were successfully obtained at desired wavelengths. An
investigation of the excitation energy transfer indicates an opti-
cal process. This work led to the first tunable three-color dye
-------
77
mixture laser spanning the visible spectrum.
The following is a chronological listing of the material presented
in this work:
1. In Chp. 1, the need for an effective method of pollution moni-
toring is explained, and the historical development of LIDAR
systems is given. Finally, an outline of this thesis is pre-
sented.
2. In Chp. 2, atmospheric scattering mechanisms are presented,
first in a qualitative manner, then scattering relationships
are detailed under Atmospheric Parameters. Towards this end,
the four LIDAR schemes, Raman and resonance Raman, long path
absorption, resonance fluorescence, and differential absorption
of scattered energy (DASE) are described, and a comparison of
operating parameters is given in Table 2.4:1.
3. In Chp. 3, the LIDAR system developed in this work, is described
in detail. Sequential mode of operation is described first,
then the necessity and operating characteristics of Simultan-
eous Two-Wavelength operation is presented. Experimental re-
sults showing the fluctuations in output beam quality, including
bandwidth changes, indicate a strong need for calibrated output
signals.
i*. In Chp. 4, results of field tests are given along with uncer-
tainties associated with the LIDAR readings.
5. In Chp. 5, theoretical and experimental results are given on
energy transfer processes in dye mixtures.
6. In Chp. 6, a summary and conclusions are presented on the work
reported.
9
-------
CHAPTER 2 LIDAR TECHNIQUES
2.1 Introduct i on
All laser radar techniques depend upon atmospheric scattering mech-
anisms for their operation. Possible techniques for pollution monitor-
ing include the following:
1. Raman and resonance Raman (Raman backscatter),
2. long path absorption (resonance absorption),
3. resonance fluorescence (resonance fluorescence backscatter),
and
4. differential absorption of scattered energy (DASE) (Rayleigh
and Mie backscatter).
In the following section, the atmospheric scattering mechanisms
are explained qualitatively, then scattering relationships are detailed
in Sect. 2.3 under Atmospheric Parameters. A comparison of the LIDAR
schemes is made in Sect. 2.4 which shows the superior operating char-
acteristics of the DASE system.
2.2 Atmospheric Scattering Hechamisms
2.2.1 Raman Scattering
The Raman effect consists of the appearance of displaced lines in
the spectrum of monochromatic light scattered by molecules. The degree
of displacement, A^, is characteristic of the molecules causing the
scattering. The intensity of the spectral line is proportional to the
density of scattering molecules.
The Raman effect is essentially a detail of the Rayleigh scattering
10
-------
process. In the Rayleigh process, incident light on a molecule is
virtually absorbed and immediately re-emitted. The absorbed light
does not raise the molecule to an excited energy level, as in class-
ical resonance absorption. Instead, after absorption, the molecule
is in a virtual excited state. This has two consequences. First,
1 5
the molecule re-emits the energy in a very short time—about 10
sec. Second, the absorption and scattering cross sections are very
small. It has been estimated that, for a typical liquid irradiated
at §321 A, 1 part in 10** of the incident light will be scattered in
a path length of 1 cm. * Gases have fewer molecules per unit vol-
ume, and the fraction of light will obviously be much less.
—2 —6
A small portion of the molecules excited (10 to 10 or less)
will emit (Raman) light of a slightly different wavelength than the
incident light. * This phenomenon is known as Raman scattering.
The process is shown in Fig. 2.2:1. Light is absorbed by molecules
in the ground state, and for some molecules Raman light is re-emitted,
with these molecules returning not to the ground state, but to an
excited vibrational level. The emitted light is naturally of a lower
energy or frequency than the incident light. The difference in fre-
quency is equal to one of the natural vibrational frequencies of the
molecule. Several such shifted lines are usually observed in a Raman
spectrum. Each molecule has its own set of vibrational frequencies
and therefore its unique Raman spectrum, which can thus be used to
uniquely identify the scattering (or re-emitting) molecule.
A few molecules initially absorb light while they are in an excited
vibrational state and decay to a lower energy level, so that their
Raman scattered light has a higher frequency. Therefore, the complete
11
-------
STOKES
ABSORPTION
VIRTUAL
STKTE
< p«
V A>
n^N /\
A A
&ROUNO
ANT\-STOKES
RAVLE16H
A
STATE
FIGURE 2.2:1 THE RAMAN PROCESS
12
-------
scattering spectrum is made up of the Rayleigh scattering (predominant)
at the same frequency as the incident light and the two components of
the Raman spectrum, the Stokes lines (which are shifted to a lower
frequency), and the anti-Stokes lines (which are shifted to a higher
frequency. The Stokes lines are more intense because under most cir-
cumstances most molecules are in the ground energy level. To summarize,
ii
the typical Rayleigh scattering ratio is 1 in 10 (for a 1 cm liquid
path).60*61 Of this 10~\ only 1 in 102 to 10** of the scatterings
exhibits the displaced wavelengths characteristic of Raman scattering.
For atmospheric gases, the Raman scattering ratio is much less. For
N_ at atmospheric pressure, this scattering ratio is approximately 1
10 If
in 10 for a 1 cm path length.
The Raman cross section determines the fraction of an incident
light beam which will be scattered at Raman wavelengths. It is a crit-
ical piece of information needed for determining pollutant detectabilities
and necessary for quantitative measurements. The cross section is a func-
tion of the fourth power of the frequency of the incident radiation, and
it may be enhanced by resonance effects.
2.2.2 Resonance Raman Scattering
The resonance Raman effect can best be explained if the first step
in scattering is considered to be the quasi-absorption of the incident
light. As the light is scanned in frequency and approaches an absorption
band, the fraction of the light absorbed increases, and the fraction of
the scattered light increases proportionally, thus a larger apparent
cross section results.
13
-------
2.2.3 Absorption and Fluorescence Scattering /
I
Absorption and fluorescence scattering is in Actuality a descrip-
;
tion of the classical resonance absorption and fluorescence mechanisms.
/
As an example, when monochromatic light propagates through the atmos-
phere and encounters a species of molecules whose absorption band coin-
cides with the propagation frequency of the transmitted light, a portion
of the light is absorbed. Subsequently, some of the absorbed monochro-
matic radiation will be re-emitted by the molecules over a wide spectrum
characteristic of the absorbing species. The fluorescence scattered
will be in all directions, shifted toward the longer wavelength end of
the spectrum and will resemble the absorption spectrum of the molecules.
2_.2,» Rayleigh Scattering
Scattering from atoms or molecules for which the particle diameter
is small compared to the wavelength of the incident light is called
Rayleigh scattering. The scattered light is directional and is of the
same frequency of the incident radiation. In the atmosphere, Rayleigh
scattering is due to molecules normally present in the air, such as, N2,
02, H20, and C02.2»li>5g :;
Rayleigh scattering occurs because molecules, for example, react to
nonresonant radiation. When incident radiation passes over molecules,
a general vibration is set up within the molecules, not at the resonance
frequency, W0, but at the frequency, tt, of the field. The vibrating
molecules are the sources of the scattered radiation.
2.2.5 Mie Scattering
Scattering from particles for which the particle diameter is large
14
-------
compared to the wavelength of incident light is called Mie scattering.
in this process, the scattered light is of the sane frequency of the
incident radiation. In the atmosphere, Mie scattering is due to par-
ticulates and aerosols.2*11*51
The DASE scheme makes use of both Rayleigh and Hie scattering. The
next section details scattering relationships which are then used in
Sect. 2.4 in a comparison of the four types of LIDAR schemes that have
potential for use as pollution monitoring systems.
2.3 Atmospheric Parameters
The atmosphere attenuates a transmitted light beam by elastic (Ray-
leigh, Hie) scattering and by absorption.
The propagation of a light beam of intensity 10 through the atmos-
phere may be expressed by:
I - I0 exp(-«*Ar), (2.3:1)
where,
I = the transmitted intensity over a distance r,
MA = the atmospheric volume extinction coefficient which is
composed of a sum of terms:
«A "*R *~M +*ABS' (2'3'2)
where,
-------
in the atmosphere and is highly variable in both particulate size,
wavelength, and particle distribution. For identical molecules of
density N, the volume extinction coefficient can be written in terns
of a cross section <* = otl where,
-------
2 2*2, (2.3:5)
ential cross section is given by: * *56
dORAM = /1_1
dn. \2TT60
where ,
w^ = the pump frequency,
i*2 = the Raman frequency, and
r* = the Raman polarizability.
The two frequencies differ by the vibrational mode frequency, wg, so
that w2 s **1 - *ty»
The large variation in size, density distribution and properties
of atmospheric aerosols preclude an accurate calculation of the Hie
scattering cross section. However, a useful empirical relation relat-
2 57 51
ing the Hie scattering coefficient is given below: ' *
PM a 3-91 I"0'55 °*5I5(V1 3>1 km'1, (2.3:6)
v L x J
where,
X = the wavelength of incident light in urn, and
V - the visibility in km.
This empirical expression is an approximation which has been found
experimentally to hold under normal visibility conditions when ambient
particulate concentrations are relatively low.
The final parameter in the atmospheric volume extinction coeffi-
cient is absorption. Atmospheric absorption is not serious for wave-
lengths longer than 2500 A. Below 2500 X, absorption due to atmospheric
oxygen becomes important with an absorption coefficient reaching 1 km"1
at 2450 A.2
Now, having discussed both the atmospheric scattering mechanisms
17
-------
and atmospheric parameters, a comparative analysis can be made of the
four types of pollution monitoring schemes mentioned in Chp. 1, namely:
1. differential absorption of scattered energy (DASE),
2. resonance fluorescence,
3. Raman and resonance Raman, and
t. long path absorption.
From the discussion which follows, it will be seen that the DASE ap-
proach has inherent advantages over the other pollution monitoring
schemes. '
2.» LIPAR Schemes; A Comparative Analysis
2.1.1 Introduction
For all four schemes, except the Raman, laser sources are needed
at wavelengths that match the resonance absorption of the pollutants
of interest. Dye lasers are now available with outputs tunable over
the entire near ultraviolet and visible spectral ranges. They are
capable of narrow band operation, less than 1 A, with good frequency
59
stability. Pulse energies of over 1 j (per pulse) are typically
available from flashlamp-pumped dye lasers. If coaxial flashlamps
and appropriate discharge circuitry are utilized, these energies are
available in pulses only a few hundred nanoseconds in duration, as is
required for obtaining desired depth resolution in LIDAR applications.
Several of the important commonly encountered pollutants have absorp-
tion spectra in the visible and near ultraviolet due to electronic
transitions. These pollutants include N02, HO, S02, and 03. This
work centers on a pollution monitoring scheme to detect trace contain-
18
-------
inant levels of N02 in the atmosphere,
2,4.2 NO^Absorption Spectrum
As can be seen in Fig. 2.1:1, N02 has a clearly structured absorp-
95
tion spectrum in the blue, with well defined peaks and troughs.
2.1.3 General Formulation
The backscattered signal from a target at range R is given by:
Pr(R) = (£\Kpo4 exp(-2^A(r)dr) , (2.1:1)
where,
Pr(R) = the received power from range R,
Pg * the transmitted power,
K = the optical system efficiency,
A = the area of the receiving telescope,
.p. » the effective reflectivity of the remote target, and
«*A(r) s the volume extinction coefficient at range r.
The transmitted radiation interacts with the pollutant molecules
through the atmospheric volume extinction coefficient, <*A, as defined
in Eq. 2.3:1. The portion of the returned signal due to Raman and
fluorescence scattering is taken into account in the effective re-
flectivity term p/rr. The depth resolution attainable by backscatter
methods is shown schematically in Fig. 2.4:2. The depth resolution is
given by
AR=et, (2.1:2)
2
where,
19
-------
A7 _
k>
u
li-
lt-
|
O
a.
O
Cfl
;.o
o*
o.i .
H as
O
o.i
O./
T—i—r
LINE BROAO6NIN6
TO
T—i—i—i—i—i—r
I 1—r
~f—i—i—i—i—i—'—i—|
4TOO A80O
FIGURE 2.U;1 N02 ABSORPTION CURVE
-------
OUT&OtNG LASER
FIGURE 2.4;2 OBTAINABLE DEPTH RESOLUTION FOR BACKSCATTER LIDAR SCHEMES
-------
t * tp + tD •»• tp, the sum of the laser pulse width tp, the
detector integration time, tp, and the molecular fluo-
rescence tine, tp (tp only cones into effect in the
resonance fluorescence backacatter scheme).
In the discussion that follows, a comparison among the DASE,
resonance fluorescence, Raman and resonance Raman, and long path ab-
sorption schemes is made. For the purpose of comparison the DASE
LIBAR system developed here were used as a basis.
2.**.1* Differential Absorption of Scattered Energy (PASS)
2.4.4.1 Basic Theory
In its simplest form, this scheme determines the concentration of a
pollutant at an arbitrary point, distance r in space, by measuring the
optical resonance absorption due to the pollutant across an incremental
path length, & r. This is shown schematically in Fig. 2.4:3. The ab-
sorption across ^r is obtained from the relative attenuation of two
collinear laser beams at close-lying wavelengths, \i and \2» respect-
ively on and off the resonance absorption of the pollutant molecule in
question. The relative attenuation is determined from comparisons (at
the receiver) of the Rayleigh and Hie atmospheric elastic backscatter
from the two laser beams as they traverse &r. Appropriate temporal
resolution at the receiver permits determination of r and &r, the range
and spatial resolution of the pollutant distribution.
Several features of this scheme should be noted at this stage:
1. As mentioned earlier, dye lasers are now capable of narrow band
operation, less than 1 A, with good frequency stability. Since
22
-------
POLLUTION
DYE
LASER
LKSER OUTPUT
KT
ALTERNATELY OR.
BACKSC^TER.60
POLLUTION DENSITY
DETERMINED ACROSS
SAMPHN6 LENGTH A /t
FIGURE 2.4:3 THE DASE SCHEME
-------
the receiver signal is simply the elastic backscattered compo-
nent of the transmitted signal, this permits the use of very
narrow bandwidth receivers (if possible, down to the bandwidth
of the transmitted signal). This is an important factor in re-
ducing undesirable background noise and the limitations that it
places on LIBAR monitoring schemes.
2. The received signal that is due to the elastic scattering from
the atmosphere is relatively large and thus easily detected.
3. Since this scheme determines the pollutant concentration by an
absorption method, the achievable detection sensitivity is high
even where the required spatial resolution is high (or equiva-
lently, where the sampling lengths are small).
The backscattered power observed by a LIDAR system with collinear
transmitting and receiving optical axes is given by Eq. 2.4:1. For the
DASE case the effective reflectivity of the atmosphere is given by:
(2.4:3)
TT 2
where,
c = the speed of light,
tp = the laser pulse length,
PR(R,X) = the Rayleigh backscatter coefficient, and
PM(R,A) = the Hie backscatter coefficient.
Substituting this relation into Eq. 2.4:1 yields,
R
Pr(R,» * ctp[pR(R,X) + £M(R,X)]KP0(A)AexpI-2$[o
-------
WSC^R»^^ stxR(R»X) +exM^R»^^» tne atmospheric volume scatter-
ing coefficient without absorption,
<*R(R,X) * the Rayleigh volume scattering coefficient, and
<*M(R,X) = the Hie volume scattering coefficient. The other
parameters were previously defined.
An expression for Np averaged over a distance &R can be obtained
from Eq. 2.4:4 by forming the difference of the logarithm of Pr(R) eval-
uated at R and R * &R for both a frequency on the absorption peak
and to the absorption trough (X2). These equations are shown below:
lnPr(R,X,) - lnPr(R + AR) «
(2.4:5)
CR,^) - lnPr(R
PM(R,A2)] -
AR|A2^ + 2«SC(X2)AR t
(2.4:6)
An expression for Nr, the value of Mr averaged for the depth resolu-
i
tion &R at range r can be obtained from Eqs. 2.4:5 and 2.4:6 as:
H - 1 [In P,.(R«Xi) - In Pr(Rt>2) +
2Aff&R Pr(R + AR,Xj.) Pr(R + ^R,X2)
S + S»], (2.4:7)
where,
S * -a^CXi) -SSC(X2)3AR, (2.4:1)
«SC(X) *«R(X) +^M(X), the total volume scattering coeffi-
cient neglecting absorption, averaged over aR, and
ACT * OABS(X!) •
-------
sections.
S' *
ln[pp(R +AR.X,) + PM(R +AR?X?)]. (2.4:f)
>R(R,X2> * PM(R,X2>
In order to evaluate Eq. 2.4:7, it is necessary to know the magnitudes
of S and S1. These quantities are, in general, unknown. However, if
if the absorption peak and trough measurements are taken nearly simul-
taneously, so that the p's andot's remain constant in time, and if
these coefficients do not change significantly over the spectral inter-
49 52 §2
val X^ - X2» then S and S' can be taken to be zero. * * Therefore,
making these assumptions, Eq. 2.4:7 reduces to:
in
2 A OAR
Pr,(R
_Pr(R
(2.4:10)
or,
. (2.4:n)
where,
AFr = the fractional change in the ratio of ratios of the re-
ceived signals.
As can be seen from Eq. 2.4:11, the sensitivity of pollutant detect-
ion is improved by:
1. the ability of the instrumentation to detect smaller changes in
*Fr,
2. increasing the sampling length, &r (at a cost of decreasing spa-
tial resolution), and
3. a large resonance absorption cross section, o^ , and a small off*
26
-------
resonance cross section, flv,
2.4.4.2 Evaluation of Required System Parameters for Detection of N09
2.4.4.2.1 Received Signal Requirements
Assuming a relative change in the ratio of ratios of received sig-
nals of 5 percent, a range of 1000 m, minimum spatial resolution of
100 m, and taking the absorption cross sections from Fig. 2.1:1 as
0XB£(2 a 4^97 * (trough) were chosen. It should be
noted that for narrow filters such as is required in the system describ-
ed, manufacturers encounter difficulty producing filters exactly to the
specifications, and the above was the closest fit that could be obtain-
ed.
The received signal power in this DASE LIDAR system is the elasti-
cally backscattered fraction of the outgoing laser beam. Since this
received power, and therefore the effective range of the system, is
determined by the fraction of the power that is reflected, this aspect
is considered next.
^
2
A detector is dark current limited if
"©«£
(2.4:12)
'd*
where,
27
-------
q = the electronic charge,
B = the electronic bandwidth of the detection system,
n = the number of pulses,
= the minimum acceptable signal to noise ratio, and
•
id = the detector dark current*
For photon limited detection the opposite of Eq. 2.4:12 is true. In-
serting the values of the detector used, Amperex type 5SAVP, (B = 3 MHZ,
id * 15 nA) into Eq. 2.4:12, it is found that the system is dark current
limited for an acceptable signal to noise ratio of 10.
The minimum detectable signal power, Pr|min» for dark current limit-
2
ed detectors is given by:
1/2
pr,min * C2q(B/n)(S/N)mjnidr, (2.4:13)
SD
where,
SD * the detector sensitivity.
-9
For the system described, P_ _*_ is found to be equal to 8.32 X 10 W.
!»•<**• ;
The minimum detectable change, A P^Q, relative to the signal power, Pr,
2
at the detector is given by the following relationship:
*P«M-n * C 4qB/S\ /I + _JLLAiy2 (2.4:14)
P,
Using Eq. 2.4:14 and the parameters for the system (SD = 60 X 10 A/W),
a minimum acceptable signal to noise ratio of 10, and a minimum detect-
able change of 5 percent, it is found that for single pulse operation a
received power, Pr, of 0.25 uW is required.
2.4.4.2.2 Transmitted Power Requirements
It is useful at this stage to get an idea of the power requirements
28
-------
for the laser transmitter for a system of this type.
Assuming an N02-polluted-free atmosphere up to the sampled region,
*R, solving Eq. 2.4:«* for the transmitted power yields:
P0 s MR + & R.X) X
KA expf-2fegf,(X)(R
(R * »D^ L
where,
<* (X) = <* (X) + *U(X), the total atmospheric volume scatter-
ov. K n
ing coefficient without absorption, averager over R +
&R, and
Np = the average pollutant concentration over &R*
An average pollution concentration of 0.315 ppm of NO over the sampled
region of 100 m, a range of 1000 m, a visibility of 10 km, and a receiv-
ed signal power of 0.25 uW for each outgoing wavelength (to give a frac-
tional change of ratio of ratios of received powers of 5 percent), yield
the on-line and off-line transmitted powers as 0.0395 MW (9.11 mj) and
0.0375 MW (9.38 mj) respectively.
Increasing the output energies of each line to 1/2 j, the received
power increases to 15.9 uW off-line and 15.8 uW on-line, which corre-
sponds to a minimum detectable pollution concentration of 0.029 ppm.
2.».*».2.3 Pump Depletion
The foregoing discussion considered the Rayleigh and Hie return
assuming an N02-polluted-free atmosphere up to the sampled region. It
is still necessary to consider the problem of pump depletion if the
29
-------
pollutant is at a density of, say, 0.315 ppm over the entire 1000 m
range* Since 100 m cause approximately a 10 percent reduction in re-
ceived signal power (and approximately a 5 percent reduction in AFr),
1000 m will cause a reduction by a factor of exp(-l.Oi) = 0.34. The
received signal will be down approximately 34 percent of its original
value of 5.41 uW off-line and 5.37 uH on-line, and it would be possi-
ble to detect a minimum concentration of 0.051 ppm, which corresponds
to a 0.45 percent change in ratio of ratios. Thus, a laser pulse
energy of 1/2 j per line is sufficient to detect a change of N02 con-
centration of 0.051 ppm averaged over 100 m at a range of 1000 m even
when pollution concentrations as high as 0.315 ppm exist along the
entire probe beam path.
If the range is decreased to 100 m, a minimum concentration as
low as 1.6 ppb could be detected (corresponding to a change in the
M
ratio of ratios of 2.6 X 10"^ percent), with the same optics and out-
put power described above.
2.4.5 Resonance Fluorescence
2.4.5.1 Basic Theory
This scheme determines the pollution concentration at a distance
r in space by measuring the fluorescence backscatter induced by the
probe beam as it is absorbed by the polluted species. This is shown
schematically in Fig. 2.4:4.
2.4.5.2 Evaluation of Required System Parameters for Detection of NO,
The backscattered fluorescence intensity from range r observed by
30
-------
LASER OUTPUT
X,
POLLUTION
DYE
DETECTOR
MOLECULAR
RETURN
FIGURE 2.*»:H THE RESONANCE FLUORESCENCE SCHEME
-------
a LIDAR system with collinear transmitting and receiving axes is given
2
by Eq. 2.4:1 with the effective reflectivity o given as:
i*—^.
tr r
where,
Op — cr^gc(A)Q, and
Q* 1 t^nrZ. <2.4:li)
fi\ TSP
where,
1 » the probability of quenching per collision,
2?
^COL s t*ie collision time, and
tgp = the spontaneous decay time.
Substituting these terms into Eq. 2.4:1 and assuming an NC^-polluted-
free atmosphere up to the sampled region, AR, yields:
Pr(R) * KPncto-Aagq )NrQAF
2" HIT R2
where,
F s the fraction of the total fluorescence detected by the re-
; •*
(
ceiver,
Mr ^ the average pollutant concentration over the sampled region,
&R, and
<*sc(>0 SO
-------
signal of 6.32 X 10~9 W (the minimum detectable received power for the
detector used) the required transmitted power of 382 MH is obtained.
This power is equivalent to 95.5 j for a dye laser with a pulse length
of 250 ns.
The discussion above considered nighttime operation, that is, the
entire badcscattered spectrum (F - 1) was detected. In daytime opera-
tion the noise attributed to background radiation must be blocked.
This has the effect of reducing the observed fluorescence backscatter-
ed signal. Therefore, to detect 0.029 ppm for daytime operation (F -
0.1) a transmitted energy of 955 j is required.
These energy requirements are, of course, only to be used as a
relative basis for a comparison among the other LIDAR schemes present-
ed. For realistic power requirements, a range of 100 m, a sampled
depth of 100 m, an average pollution concentration of 0.1 ppm of M02,
yields for nighttime operation a required transmitted power of 3.52 X
105 W (I.10 X 10"2 j). For daytime operation, the required transmitt-
ed power increases to 3.52 X 10* W (t.10 X 10'1 j).
To summarize, the resonance fluorescence backscatter scheme has the
following disadvantages:
1. Daytime operation severely limits detection sensitivities, re-
sulting from the blockage of background radiation.
2. The fluorescent backscattered signal at the receiver is depen-
dent upon the concentration of the pollutant in the region
being probed.
2.4.8 Raman and Resonance Raman
33
-------
2.4.6.1 Basic Theory
The Raman scheme determines the concentration of the pollutant at an
arbitrary point in space by measuring the Raman backscattered intensity
induced as the probe beam interacts with the polluted atmosphere. This
is shown schematically in Fig. 2.4:5.
2.4.6.2 Evaluation of Required System Parameters for Detection of N09
The backscattered Raman intensity at the receiver due to the pollu-
tant is of the same form as that for the resonance fluorescence scheme,
but with the effective reflectivity term £ given as:
IT
(2.4:20)
TT 8
where,
°RAM = *k* Raman scattering cross section.
Subsitiuting this relation into Eq. 2.4:1 and assuming an NOj-polluted-
free atmosphere up to the sampled region, &R, yields for the received
power:
Pr(R) = KPQctcrp.M(X)N,.AF exp[-2*sc(X)R -
2 4TT R2
(2.4:21)
For an average pollution concentration of 0.029 ppm across the sampled
region, a range of 1000 m, a depth resolution of 100 m, and a received
power of 8.32 X 10" W (the minimum detectable received power), the re-
quired transmitted power of 1.77 X 101 MW (4.43 X 107 j) is obtained.
To reduce the transmitted power requirements, this LIDAR scheme can
be run in the resonance Raman mode. This has the effect of increasing
the received backscattered Raman intensity by increasing the Raman scat
34
-------
LKSER
CO
m
DETECTOR
OUTPUT
MOLECULAR POLLUTION
RAttAN RETURNS
FIGURE 2.4:5 THE RAMAN SCHEME
-------
2
taring cross section as:
(do1
v-d^ RES x
-------
to be used in a comparison among the LIDAR schemes presented. The
magnitude of the nonresonance Raman backseattering coefficient limits
the range of this type of LIDAR system to less than 50 m for any prac-
tical transmitted energy requirements for monitoring pollution levels
present in ambient conditions (0.01 to 5 ppm), and limits the range
2
to less than 75 m for the resonance Raman scheme*
The disadvantages of the Raman and resonance Raman schemes are
summarized below:
1. The Raman backscattering cross sections are 10 to 103 smaller
than the Rayleigh and Hie backscattering cross sections.
2. The strength of the Raman and resonance Raman returns depend
upon the concentration of the pollutant present in the atmos-
phere. This dependency is the main factor in limiting detec-
tion sensitivities. Even though for N02 detection, the reso-
nance Raman backscattering cross section is larger than the
sum of the Rayleigh and Hie backscattering cross sections
•23 9 —28 2
(1.21 X 10 cm as compared to 3.21 X 10 cm respectively),
the backscattering coefficient which is a function of pollutant
concentration in the resonance Raman (and Raman) case, and a
function of air in the elastic (Rayleigh and Hie) case, deter-
mines the strength of the return signal.
3. The Raman and resonance Raman returns that are anti-Stokes
shifted are interf erred with by the Rayleigh and Mie backscat-
ter. It is therefore necessary to reject that part of the
Raman return as well as the elastically backscattered signal,
which results in a decrease of detected signal power at the
37
-------
receiver.
4. In the resonance Raman scheme, since Raman returns are induced
by probing approximately with an absorption frequency, and the
fluorescence frequency shift for molecules is of the same order
as the Raman shift, discrimination problems exist at the receiv-
er. Finally, the magnitude of the backscattered fluorescence
intensity may be larger by orders of magnitude than the reso-
2
nance Raman intensity.
2.1.7 Long Path Absorption
2.4.7.1 Basic Theory
In this technique, the average pollution concentration along the
probe beam, Fig. 2.1:6, is determined from measurements of the relative
attenuation along the length of the probed path, R, of two collinear
laser beams at close-lying wavelengths, \^ and A2» respectively on and
off a resonance absorption peak of the pollutant in question. This
scheme has the disadvantages of needing a remote detector or reflective
target to receive the transmitted beam, and lacks depth resolved meas-
urements* The advantages of this scheme are its good sensitivity and
the use of low power light sources.
Since monostatic LIDAR schemes are the ones of interest, two types
of long path absorption schemes are considered, namely:
1. long path absorption utilizing a retrorefleetor, and
2. long path absorption utilizing a topographical target.
The received power at the detector is given by Eq. 2.»»:1 with the
appropriate effective reflectivity terms inserted. For absorption meas-
38
-------
DYE
LASER
LASER OUTPUT
AT X,AND\t
ALTERNATELY OR
SIMULTANEOUSLY
MOLECULAR. POLLUTION
<£>
DETECTOR
RETRoREFLECTOR OR.
TOPO&RAPmCAL TARGET
REFLECTED
FIGURE 2.H:6 THE LONG PATH ABSORPTION SCHEME
-------
urements with a re troref lector and a collimated transmitted beam, the
range-squared dependence is effectively cancelled. Thus, all the
transmitted power is collected except for a collection efficiency
factor, |, which is near unity, and the effective target reflectivity
is orders of magnitude better than for a Lambertian scatterer. The
effective reflectivity term for such a case is given by:
E\ " *!l • (2.4:23)
A
For long path absorption by scattering from topographical targets:
i£\ s £ i (2.4:24)
vn/ TOPOG IT
where,
18
p ft* 0.1 for visible and u.v. wavelengths.
An expression for the average pollutant concentration over R can
be obtained by taking the difference between the logarithms of the
received powers from the off-line and on-line wavelengths respective-
ly:
lnPr(R,A2) - InPpCR.X-L) * lnP0U2) - lnP0(Xi) *
(2.4:25)
where,
^SC^) ~ ^R^^ *0
-------
where,
S" = -H^scCX^ -«SC
-------
2.4.7.2 Evaluation of Required System Parameters for Detection of NO/?
For both the long path absorption methods, utilizing a retroreflec-
tor and a topographical target, the received power at the detector is
given by Eq. 2.4:1 with the proper effective reflectivity terms insert-
ed for each case. Solving Eq. 2.4:1 for the transmitted power for the
topographical target case yields:
P0U) *
0.1KA
IT R2
(2.4:29)
Inserting the appropriate values into Eq. 2.1:29 yileds the on-line and
off-line transmitted powers as 1.22 X 105 W (3.05 X 10~2 j) and 1.17 X
105 W (2.93 X 10"2 j) respectively.
The use of a specular retroreflectbr has the effect of cancelling
out the R2 dependence. This is seen by the reduced on-line and off-
line transmitted power requirements, which are 4,44 X 1(T4 tf (1.11 X
10"10 j) and 4.23 X 10"4 W (1.06 X lo"10 j) respectively.
As can be seen from the above results, the long path absorption
scheme offers required sensitivities needed to monitor pollutant con-
centrations found in urban areas (.01 to 5 ppm). This is done with
more than conservative transmitted power requirements. Unfortunately,
this is accomplished at the expense of losing depth-resolved measure-
ments.
2.5 Summary
A tabulated summary showing the relative merits for each LIDAR sys-
tem considered in the comparison is given in Table 2.4:1. The compari-
42
-------
son was made by first calculating the operating characteristics for the
DASE system: an assumed minimum detectable change of received powers
of 5 percent, the received power needed to observe the minimum detect-
able change (0.2S uW for each on-line and off-line wavelengths), an
assumed range of 1000 m, and a depth resolution of 100 m. These con-
ditions led to the attainable sensitivity of 0.029 ppm of NO- with
moderate transmitted energy requirements of 9.It mj (on-line) and 9.3S
raj (off-line). Then, the required received powers were calculated for
the above parameters, assuming output energies of 1/2 j in each lasing
line.
Since the Raman, resonance Raman, and resonance fluorescence schemes
require orders of magnitude higher output energy levels than the long
path absorption and DASE methods, the received power was assumed to be
equal to the minimum detectable power for our system, namely, 6.32 X
-9
10 W. This was done in order to show that under optimum operating
conditions, these systems still fall short of the long path absorption
and DASE output energy requirements to detect the same pollutant concen-
tration of N02 (0.029 ppm).
As can be seen, the long path absorption schemes require the least
output energies to monitor a given pollutant concentration, or equiva-
lently the best sensitivity for a given output energy. Unfortunately,
this is accomplished at the expense of losing depth-resolved measure-
ment capabilities. The DASE scheme, therefore, combines the spatial
resolution offered in the Raman, resonance Raman, and resonance fluo-
rescence schemes, with the greater attainable sensitivities of the long
path absorption schemes.
43
-------
LIDAR
Meth.
Raman
Reson.
Raman
Reson*
Fluor.
(night)
Reson.
Fluor.
(day)
Long
Path
Abs.
Topog.
Long
Path
Abs.
Retro.
DASE
Range
(m)
1000
*
1000
ftft
1000
(100)
1000
(100)
1000
1000
1000
Res.
(m)
100
*
100
ft*
100
(100)
100
(100)
1000
1000
100
Concen.
N02
(ppm)
0.029
ft
0.029
a*
0.029
(0.1)
0.029
(0.1)
0.029
0.029
0.029
Received
Power
On-Line
(W)
*
6.32X10"9
ft*
6.32X10'9
(6.32X10"9)
6.32X10~9
(6.32X10-9)
15.8X10~6
15.8X10"6
15.8X10"6
Off-Line
(W)
6.32X10"9
ft
ftft
-6
15.9X10
15.9X10"6
15.9X10"6
Required Transmitted
Power
On-Line
(MW)
ft
5. 04X10 3
ftft
3.82X102
(3.52X10"1)
3.82X103
(3.52)
-1
1.22X10
U.44X10"10
2.00
Off-Line
(MW)
1.77X109
ft
ftft
-1
1.17X10
-10
»».23X10
2.00
Required Transmitted
Eneri
On-Line
(5)
*
1.26X103
ftft
9.55X101
(8.80X10"2)
9.55X102
(8.80X10"1)
3.05X10*2
1.11X10"10
5.0X10"1
TV
Off-Line
(j)
4.«*3X108
ft
ftft
2.93X10*"2
1.06X10"10
5.0X10'1
( ): For realistic power requirements
*: Limited to less than 50m for detection of ambient pollution levels
**: Limited to less than 75m for detection of ambient pollution levels
TABLE 2.H:1 TABULATED COMPARISON OF MONOSTATIC LIDAR SCHEMES
-------
CHAPTER 3 THE LIDAR SYSTEM
3.1 Introduction
The basic LIDAR system consists of a flashlamp-pumped dye-laser
fi Q
transmitter, and an optical receiver. Figs. 3.1:la and b show a
schematic for the combination. Two arrangements and modes of opera-
tion are possible. In one mode, signals at the "on" and "off" reso-
nance absorption wavelengths were generated and detected sequentially.
In the other mode they were generated and detected simultaneously.
The arrangement shown in Fig. 3.l:la is for sequential operation and
that shown in Fig. 3.1:lb is for simultaneous.
3.2 Optical Receiver
The primary lens of the optical receiver is an acrylic Fresnel
lens (purchased from Fresnel II, Inc.) mounted on a light-tight steel
frame which was designed for the purpose. This lens is basically a flat
thin piece of a plastic on which a series of concentric stepped zones
extending from the center to the edge are molded. Fig. 3.2:1 illus-
trates the geometric relationship between the zones on the Fresnel
lens and the surface of an ordinary lens. Each zone refracts the in-
cident light so that the combined action of each refracting facet fo-
cuses light essentially in the same manner as a conventional lens.
The lens used has a focal length of 2«* inches, a diameter of 15
inches, and a thickness of I/I inch. With concentric zones spaced ap-
proximately 125 grooves per inch, the lens is capable of forming a
sharp image for objects near the optical axis. This is appropriate,
-------
APERTURE
OS
PHOTOMULTlPUER
COLLlMAvTOR
ELECT ROttA6NeTICALl.T'
CONTROU-ABUE
EXPANDER
FIGURE S.ltla THE LIDAR SYSTEM FOR SEQUENTIAL OPERATION
-------
FIGURE 3.1;lb PHOTOGRAPHS OF THE LIDAR SYSTEM FOR SIMULTANEOUS
OPERATION
-------
LENS
-FRESNEL LENS
FIGURE 3.2:1 FRESNEL LENS
48
-------
since a LIDAR receiver does not require a wide field of view for col-
lecting the backscattered light from the region scanned.
For this Fresnel lens, the circle of least confusion for collimated
light directed parallel to the optical axis was measured and found to
be approximately one centimeter. In the plane containing the circle
of least confusion, the spatial distribution of the collected light j.s
a bright spot of the above diameter with the surrounding area weakly
illuminated by light spuriously scattered at the edges of the concen-
tric zones of the Fresnel lens. To eliminate the spurious light, an
iris was placed in the plane containing the circle of least confusion.
Since LIDAR receivers do not require a wide field of view, this remedy
does not pose a problem. The Fresnel lens used in this system costs
far less than a comparable parabolic mirror, yet its quality is certain-
ly acceptable for our LIDAR application.
The backscattered light collected by the Fresnel lens is then col-
limated by a 1.5 inch diameter lens, and for the sequential mode of op-
O
eration, passes through a 58 A bandwidth three-period filter onto an
Amperex 56 AVP photomultiplier. The center wavelength of the filter
and its bandwidth are such that the "on" and "off" resonance absorption
wavelengths are both passed by it. To further reduce spurious light
pick-up, a three foot cylindrical lens hood' is mounted on the front of
the receiver.
In the simultaneous wavelength mode of operation, the collimated
light entering the receiver is split by means of a partial reflector (beam
splitter) into two beams. "Each of these beams is then directed to a. separate
photomultiplier covered with a narrow filter of the appropriate fre-
49
-------
quency. The photomultipliers used are Amperex type 56 AVP. See Figs.
3.2:2a and b. The interference filters have measured half bandwidths
of 9.3 A and 7.7 A centered at 4478.5 A and 4500 A respectively. See
Spec. Sheet 3.6:1.
The photomultiplier outputs, corresponding to the backscatter re-
turn signals are displayed on Tektronix type 549 oscilloscopes and
recorded photographically.
3.3 The Dye Laser
3.3.1 The Laser Head
The basic laser system consists of a flashlarap-pumped organic-dye
laser. The laser head is constructed in a coaxial arrangement in which
the flowing dye is uniformly excited by an annular discharge surround-
ing the active medium. See Figs. 3.3:la and b. The laser head, model
DL2100B, and discharge circuitry was purchased from the Phase-R Company,
and was chosen for maximum energy output and shortest possible pulse
width. Fig. 3.3:lb shows the laser head situated in the LIDAR system,
and Fig. 3.3:2 shows a typical output pulse.
The organic dye, 7-diethylamino-4-methyl coumarine, dissolved in
ethanol was used as the lasing medium. A concentration of 3 X 10"5 M/l
was found to give maximum output energy consistent with a laser line-
width that was less than the halfwidth of the absorption peak of NCu
to be measured, i.e., less than approximately 10 A for the 4478.5 A
absorption peak. In this manner, it was possible to maximize the re-
turn signal, thereby increasing range, and still retain the maximum
difference between absorption peak and trough, and thus attain the sen-
50
-------
INTERFERENCE
FILTERS
FRE.SNEL.
LENS
PHOTON OUT I PLIERS
BENI SPLITTER.
FIGURE 3.2;2a THE OPTICAL RECEIVER SYSTEM FOR SIMULTANEOUS OPERATION
-------
FIGURE 3.2;2b PHOTOGRAPHS OF THE OPTICAL RECEIVER SYSTEM FOR
SIMULTANEOUS OPERATION
52
-------
Ln
CO
INSULATOR
DYE RE61ON
6LECTRODE
COPPER
^^
XENOH-FILLED DISCHARGE /?£G»OA/
END SELL
FIGURE 3.3jla COAXIAL ARRANGEMENT OF THE FLASHLAMP-PUMPED DYE-LASER HEAD
-------
FIGURE 3.3;lb PHOTOGRAPHS OF THE LASER HEAD SITUATED IN THE LIDAR
SYSTEM
-------
Ul
p
UJ
ui
Z
(zoo
Ul
-1
Ul
I
H
4500
TIME (300 /VAWOSffC/0/v)
FIGURE 3.3:2 TYPICAL FLASHLAMP-PUMPED DYE-LASER OUTPUT PULSE
55
-------
sitivities that the system is capable of.
3.3.2 Dye Laser Mechanism—Brief Description
1 *
Organic molecules possess a lowest excited singlet state, SQ, and
fluorescsnce when it occurs, originates from this state. In making a
fluorescent transition the molecule reverts back to its ground state,
5 , while simultaneously emitting radiation*
The spectrum of the fluorescent radiation from an organic dye often
has more than one maximum, and it usually spans a region no less than
several hundred angstroms wide. The reason for this large bandwidth
is that the radiation is actually made up of hundreds of components,
corresponding to transitions originating from various sublevels of the
first excited singlet state and terminating at various sublevels of the
ground state. These sublevels are associated with specific vibrations
of the molecule as a whole. See Fig. 3.3:3. Since some of the vibra-
tional sublevels of the ground state may be high enough in energy so
that they are normally unoccupied, a population inversion sufficient
for laser action can be established between the states from which fluo-
rescence originates and some of the higher vibrational levels of the
ground state.
Once threshold conditions are reached, the laser light that is pro-
duced has frequencies centered about one of the broad fluoresence
peaks. The bandwith of such broadband operation can range up to 200
0
A for flashlamp-pumped dye lasers*
To reduce the naturally large bandwidth of the flashlamp-pumped dye
laser output, a tuning element such as a diffraction grating can be
56
-------
o
Of
Ml
COORDINATE
FIGURE 3.3:3 ENERGY LEVELS OF ORGANIC DYE MOLECULES
57
-------
placed in the laser cavity. With the proper choice of grating charac-
teristics, such as blaze wavelength, grooves per millimeter, and power-
density handling capabilities, etc., the desired laser frequency and
bandwidth are attainable, and may be chosen to match the absorption
peak and troughs for N02.
3.3.3 Intra-Cavity Grating for Wavelength Selection
To obtain laser outputs at the required wavelengths, "+478.5 A and
4500 A, and linewidths, a grating with the specifications given in Spec.
Sheet 3*6:1 was chosen, tested, and found suitable. In both sequential
and simultaneous modes of operation, gratings are used in the Littrow
configuration with the Blaze angle in the first order. In this manner,
all the other orders extinguish and the diffracted energy is concentrat-
ed in the first order (at the wavelength desired).
Gratings have different efficiencies (percentage of light incident
on a grating that is diffracted into the desired order) for light that
is polarized perpendicularly or in parallel to the groove profile. For
the gratings used, perpindicularly polarized light gives the better effi-
ciency. See Spec. Sheet 3.6:1. Since a polarizing element is inserted
intra-cavity for the simultaneous mode of operation, this factor must
be taken into account.
For both modes of operation the bandwidth of the laser lines obtain-
ed are determined by the dispersion of the grating and the effective ap-
erture created by the geometry of the lasing medium, which in turn depends
upon the distance of the grating from the laser head. This is readily
seen in Fig. 3.3:'+. The dispersion of the grating is the angular sepa-
58
-------
Cn
VO
DIFFRACTED UGHT
LASER
EFFECTIVE
APER1URB
.OUTPUT n IRROR
FIGURE 3.3:«t DISPERSION OF GRATING AND EFFECTIVE APERTURE CREATED BY THE LASER HEAD
-------
ration obtained for two different radiations. Therefore, the farther
the grating is from the laser head containing the lasing medium the
smaller the lasing bandwidth.
3.** Experimental Arrangement for Sequential Operation
Sequential operation of the dye laser at two desired wavelengths
was achieved with the arrangement shown in Figs. 3.3:1 and 3.4:1.
The grating at one end of the laser cavity is used to obtain the
desired wavelengths of laser oscillation. With a overall cavity
length of approximately 2.5 m, and a 30 percent output mirror, the
desired wavelengths at 147S.5 A and 4500 A (on and off resonance ab-
o
sorption respectively) are obtained with approximately 5 A linewidths.
Mechanical stops are adjusted to define the grating orientation at
these two wavelengths. The grating is then switched from the setting
for one wavelength to the other by means of an electromagnetic drive
after each firing of the laser. The time between shots is approximate-
ly 5 seconds, the time required for the capacitors to recharge after
firing a pulse.
3,5 Simultaneous Two-Wavelength Operation
3.5.1 The Need for Simultaneous Two-Wavelength Operation
Thermal effects and consequent index changes in the atmosphere
cause perturbations in both the outgoing and backscattered signals.
In addition, particulate drift in existing winds can also cause fair-
ly rapid changes of the scene being scanned by the LIDAR beam. Both
these factors argue in favor of the use of simultaneous two-wavelength
60
-------
DETECTORS
FIGURE 3.ttl THE OVERALL TRANSMITTING SYSTEM WITH CALIBRATION SETUP FOR SEQUENTIAL OPERATION
-------
LIDAR operating mode.
51
The thermal factors affecting the signals are summarized below:
1. Bean steering--angular deviation of the beam from the line-of
sight path, causing the beam to miss the receiver. See Fig.
3.5:2.
2. Image dancing—-variations in the beam-arrival angle, causing
the focus to move in the image plane.
3. Beam spreading—small angle scattering, incresing the beam
divergence and causing a decrease in spatial power density at
the receiver. See Fig. 3.5:1.
4. Beam scintillation—small scale destructive interference with-
in the beam cross section, causing variations in the spatial
power density at the receiver. See Fig. 3.5:3.
The effects of scintillation on laser beam propagation have been
£2 64
studied both experimentally and theoretically. ' However, atmos-
pheric scintillation effects involving ItO-degree backscatter have re-
ceived relatively little attention either theoretically or experimen-
tally.
A recent survey article by Lawrence and Strohbehn demonstrates that
under most conditions the power spectrum of intensity fluctuations for
one-way propagation decreases above approximately 100 HZ, and is quite
64
small above 1000 HZ. In the abscence of data for two-way propagation,
it is reasonable to assume that the high-frequency power spectra for
one-way and two-way propagation are the same. Hence, the atmosphere
can perhaps be considered stable insofar as scintillation is concerned
62
for time intervals smaller than 1ms.
62
-------
FIGURE 3.5;1 BEAM SPREADING
FIGURE 3.5:2 BEAM STEERING
63
-------
FIGURE 3.5:3 BEAM SCINTILLATION
58
•
-------
Therefore, to be unaffected by scintillation the differential ab-
sorption measurement must be completed within this time interval. The
optimum manner to meet this requirement is to emit two pulses (of
slightly differing wavelengths) simultaneously. Spatial fluctuations
within each sampled depth will be averaged out for each pulse, and in
any case variations from one sampled depth to another will be the same
for both wavelengths, and therefore win not affect the results.
3.5.2 Cavity Design
3.5.2.1 Introduction
To obtain laser action at the two wavelengths simultaneously, the
arrangement shown in Figs. 3.5:la and b, with a polarizing beam-split-
ting prism in the laser cavity, was evolved and used successfully. *
With the prism splitting the light into two beams polarized at right
angles to each other, each of the two wavelengths lases along one of
the polarizations. Competition, in the form of bandwidth changes, be-
tween the two wavelengths is observed and is discussed in Sect. 3.6.
Two types of polarizing beam splitters were tested for this appli-
cation:
1. a polarizing beam-splitting cube utilizing dielectric coatings,
and
2. an air-spaced Glan-Taylor prism.
While both provided simultaneous laser action for close-lying wave-
lengths in the 4500 A spectral region, it was found that each suffered
some drawbacks. Both types of beam splitters are examined further be-
low.
65
-------
OUTPUT Xi.
OUTPUT
CUBE
FIGURE 3.5:«*a OPTICAL ARRANGEHENT FOR SIMULTANEOUS TWO-WAVELENGTH OUTPUT
-------
FIGURE 3.5;Ub PHOTOGRAPHS OF THE OPTICAL ARRANGEMENT FOR SIMULTANEOUS
TWO-WAVELENGTH OUTPUT
-------
3,5.2.2 Dielectric-Interface Polarizing Beam-Splitting Cube
The principle of this device is that it is always possible to find
an angle of incidence so that the Brewster condition for an interface
between two materials of differing refractive index is satisfied.
When this occurs, the reflectance for the parallel-plane (p) of polar-
ization vanishes. Perpindicularly-polarized (s) light is partially re-
flected and transmitted. To increase the s-reflectance, retaining the
p-transmittance at or very near unity, the two materials are then made
into a multilayer stack. The layer thicknesses are quarter-wave optical
thicknesses at the appropriate angle of incidence.
When the Brewster angle for normal thin-film materials is calculat-
ed, it is found to be greater than 90 degrees referred to air as the
incident medium. In other words, it is; beyond the critical angle for
the materials. This presents a problem which is solved by building the
multilayer filter into a glass prism so that the light can be incident
on the multilayer at an angle greater than critical. The type of ar-
rangement is shown in Fig. 3.5:5.
The dielectric-interface polarizing beam-splitting cube tested was
found to have an acceptance angle of up to 5 degrees, a passband extend-
ing from approximately 2215 A to 6739 X, and whose center wavelength is
approximately 1*190 A. The entrance and both exit faces were polished
to X/10, and were antireflex coated for 1490 A.
Utilizing the dielectric-interface polarizing beam-splitting cube,
simultaneous two-wavelength operation was readily achieved. Fig. 3.5:6
is a picture taken from a 1/2 m Jarrell-Ash scanning spectrometer, and
shows the two beams of close-lying wavelengths (the signal to the right is
68
-------
\
FIGURE 3.5:5 DIELECTRIC-INTERFACE POLARIZING BEAM-SPLITTING CUBE
69
-------
FIGURE 3.5;6 SPECTROMETER PHOTOGRAPH OF THE SIMULTANEOUS
TWO-WAVELENGTH OUTPUT UTILIZING THE DIELECTRIC-
INTERFACE POLARIZING BEAM-SPLITTING CUBE
-------
cantered at W7i.5 A and to the left at 4500 A) operating in the sim-
ultaneous mode, and obtained using the dielectric-interface polarizing
beam-splitting cube. Both beans have output energies of approximately
100 ntj, and bandwidths of 5 A.
3,5.2.3 Air-Spaced 61an-Taylor Prism
The power handling capabilities of the air-spaced Glan-Taylor prism
far exceed that of the dielectric-interface prism, since the medium
sandwitched between the two prisms is air rather than a delicate die-
lectric coating. This prism is in fact comprised of two linearly bire-
fringent calcite (CaO*C02) prisms separated by the air gap.
In each prism the optic axis is perpindicular to the incident beam
and to the lateral faces. The operation of the device may be readily
understood in view of Fig. 3.5:7. This shows the air-spaced prism, its
optic axis, and the incident unpolarized beam. When an unpolarized
laser beam is incident upon the input face, two refracted orthogonally
linearly-polarized beams are produced. The E-ray, the refracted beam
whose polarization direction is parallel to the optic axis, experiences
no change in index as it propagates through the prism-air-gap-prism
combination.
The 0-ray, the refracted beam whose polarization direction is per-
pindicular to the optic axis, experiences a large change in index at
the calcite-air-gap interface and is thus totally internally reflected.
Use is made of both the E and 0-rays to simultaneously obtain laser
action on beams polarized at right angles to each other with gratings
terminating the laser cavities beyond each prism. The optical arrange-
71
-------
CALCITE PRISMS
to
OPTIC (4) ATMS
MR GAP
FIGURE 3.5:7 AIR-SPACED GLAM-TAYLOR PRISM
-------
ment is shown in Fig. 3.5:H.65'6
<-
In the manufacture of the air-spaced prism it was only possible to
polish the input face and the axial output face to within X/tt because
of the crystal structure. The lateral faces are perpindicular to the
optic axis, and since calcite belongs to the hexagonal crystallographic
system, attempts to polish these faces to better than X/4 would result
in hexagonal pitting, giving rise to large scattering losses.
The high losses on the lateral face meant that for geometrically
identically optical arrangements the low-loss axial resonator dominated,
and laser action was restricted to it. When the axial resonator was
blocked however, laser action was also obtainable in the lateral reso-
nator.
By greatly reducing the off-axial resonator length it was possible
to obtain laser action simultaneously in both resonators. Under these
conditions, competition effects were clearly observed between the two
resonators* Fig* 3.5:8 shows simultaneous two-wavelength operation as
displayed in the spectrometer using the air-spaced Glan-Taylor prism.
Thus, when the axial resonator was blocked the output energy of the
lateral resonator increased and vice versa. Similar competition effects
65—69
were previously reported. The interaction between the two laser
beams of perpindicular polarizations in the amplifying medium is pos-
sible because the orientation relaxation time of the dye molecule (ap-
proximately 300 ps in ethanol) in the solvent is much shorter than the
70,71
lifetime of the first excited singlet state (approximately 5 ns).
Because of the high losses associated with the lateral faces, it
was found that when conditions were arranged to obtain equal energies
73
-------
LEFT TO (H6WT • L*>SE(J L |K
-------
and equal linewidths from both resonators simultaneously! the output
energies were too low to be useful. See Fig. 3.5:9.
Thus, in spite of the higher power-density handling capabilities
of the air-spaced prism, it was decided to abandon its use in favor of
the dielectric-interface prism.
With this prism simultaneous two-wavelength outputs of approxi-
mately 100 mj were obtained at W7I.5 A and 4500 A with linewidths of
approximately 5 to 6 X. Fig. 3.5:6 shows the simultaneous outputs
displayed on the spectrometer. It was with the dielectric-coated
prism that the field experiments described in later sections were car-
ried out.
3.6 Calibration
3.6.1 The Need for Calibration
In order to obtain accurate pollution measurements, it is neces-
sary that we know from shot to shot what the absorption cross section
is for a given pollutant for the actual laser bandwidth emitted. In
performing the experiment to achieve simultaneous laser action, it was
noticed that the output bandwidths varied from shot to shot. These
variations in bandwidths depend upon:
1. The bias voltage applied to the charging circuit, i.e., the
energy discharged in each flash.
2. The length of time between shots.
3. The degradation of the dye molecules due to repeated exposure
to the pump source.
Two experiments were performed to determine:
75
-------
FROM LEFT TO Rt&WTl LA* 6R. L I N 6.S , 4 5OO A ,<44-!f.5A ; M0> LIWES, 435?
FIGURE 3.5:9 SPECTROMETER PHOTOGRAPH OF THE SIMULTANEOUS TWO-
WAVELENGTH OUTPUT UTILIZING THE AIR-SPACED CLAN-
TAYLOR PRISM FOR EQUAL OUTPUT ENERGIES IN EACH
LASING LINE
-------
1. If any frequency pulling occurs between the two lasing wave-
lengths.
2. The extent of any laser bandwidth variations and frequency
shifts from shot to shot for the simultaneous mode of opera-
tion* The purpose of this experiment was to determine wheth-
er a calibration system was needed to monitor each of the out-
going wavelengths.
i
The experimental setup is shown in Figs. 3.6:la and b. The laser
is set for simultaneous operation. The two perpindicularly polarized
beams are diverted by a beam-steering mirror mount through a variable
polarizer, a variable neutral density filter, and aim focal length
lens which focuses it onto a Corning 3S50 filter and diffusion plate.
Following the diffusion plate is a 1/2 m Jarrell-Ash scanning spec-
trometer whose input and output slits are set at 10 u each. A camera
attachment is used and the results are recorded photographically.
To observe any frequency shifts, the laser lines were compared
with known Hg lines: 1047 A, 4071 A, and 4351 A. The Hg lines were
initially exposed onto the film by directing the output of an Hg lamp
into the spectrometer through the 3350 Corning filter which blocked
out any unwanted harmonics. The laser lines were then superimposed
onto the film which remained stationary during both exposures.
In the first experiment, an investigation was made of whether fre-
quency shifts occur between the cases where each wavelength lases sep-
arately and when both lase simultaneously (with the same grating set-
tings naturally).
To carry out the experiment, the laser was allowed to lase simul-
77
-------
'/3L-KV SC/SHNVN6
SPECTROMETER:
00
CAMERA.
CORNING 3850 FILTER.
FIGURE 3.6;la EXPERIMENTAL SETUP FOR MONITORING LASER FREQUENCY SHIFTS AND BANDWIDTH CHANGES
-------
FIGURE 3.6;lb PHOTOGRAPHS OF THE EXPERIMENTAL SETUP FOR MONITORING
LASER FREQUENCY SHIFTS AND BANDWIDTH CHANGES
-------
taneously, but only one polarization was allowed to pass through the
variable polarizer and onto the film. The exposed film was then com-
pared to another exposure from a sequential lase at the same wavelength.
Any frequency shifts would be obtained by observing the relative changes,
if any, from the Hg lines to which both films were exposed. Figs. 3.6:2,
3,4, and 5 show respectively the 1471*5 A line lasing sequentially then
simultaneously, and the 4500 A line lasing sequentially then simultane-
ously. There were no frequency shifts noted between the lines when
lasing sequentially as to compared to when they lased simultaneously,
but bandwidth changes did however occur, i.e., there was a change in
laser bandwidth (along with a monotonic change in output energy) when
the system lased sequentially as opposed to simultaneously. As can be
Figs. 3.6:2 through 3.6:5 the simultaneous-mode bandwidths and energies
O o
of the 4478.5 A and 4500 A lines are less than those for the sequential
mode. They are: 4478.5 A: sequential 10 A, 200 mj; simultaneous 7 A,
100 mjj 4500 A: sequential 7 A, 100 mj; simultaneous 4 A, 50 mj.
In the second experiment, lasing occurred simultaneously. Any fre-
quency and bandwidth changes were observed in the same manner as above.
Figs. 3.6:6a through 1 are photographs that show no frequency shifts
existed but bandwidth changes did occur from shot to shot. The band-
width changes depend upon:
1. bias voltage, i.e., the energy discharged in each flash,
2. time between pulses, and
3. degradation of the dye molecules due to repeated exposure to
the pump source.
As can be seen from the photographs the bandwidths varied as much as 43
80
-------
LEFT TO KKHT: ^^n«-5 A, ^358 A, 4cm A,, AND
FIGURE 3.6;2 SPECTROMETER PHOTOGRAPH OF THE U478.5
£ LASER LINE FOR SEQUENTIAL OPERATION
LEFT TO
FIGURE 3.6;3
- 5 A,*35« A, 4OTS , ^KD AOA~> A
SPECTROMETER PHOTOGRAPH OF THE
W78.5 X LASER LINE FOR SIMUL-
TANEOUS TWO-WAVELENGTH OPERATION
-------
LEFT 10 R14HT '. ^$°O A, 43 5SA , 40T ? A, ANQ 4041 A
FIGURE 3.6;4 SPECTROMETER PHOTOGRAPH OF THE U500
A LASER LINE FOR SEQUENTIAL
OPERATION
f O m a
LEFT TO HiGm : ^500 A; 435» A,4oif A, AND 4047 A
FIGURE 3.6;5 SPECTROMETER PHOTOGRAPH OF THE U500
A LASER LINE FOR SIMULTANEOUS TWO-
WAVELENGTH OPERATION
-------
peon LEFT TO UIQKT: ^5ooA.l44i?.SA,*ass ;4
-------
FIGURE 3.6;6c BIAS VOLTAGE = 20 KV, THIRD SHOT FIGURE 3.6;6d BIAS VOLTAGE = 20 KV, FOURTH SHOT
-------
FIGURE 3.6;6e BIAS VOLTAGE = 19 KV, FIRST SHOT FIGURE 3.6;6f BIAS VOLTAGE = 19 KV, SECOND SHOT
-------
•
FIGURE 3.6:6^ BIAS VOLTAGE = 19 KV, THIRD SHOT FIGURE 3.6;6h BIAS VOLTAGE = 19 KV, FOURTH SHOT
-------
FIGURE 3.6;61 BIAS VOLTAGE = 18 KV, FIRST SHOT FIGURE 3.6;6j BIAS VOLTAGE = 18 KV, SECOND SHOT
-------
FIGURE 3.6;6k BIAS VOLTAGE = 18 KV, THIRD SHOT
FIGURE 3.6;61 BIAS VOLTAGE = 18 KV, FOURTH SHOT
-------
percent.
As a result of these experiments, it was therefore concluded that
since significant bandwidth changes did occur from shot to shot during
simultaneous operation, a calibration system must be used in the LIDAR
system in order to allow for these changes. The next two sections de-
scribe the calibration system devised for this purpose, and tuning
techniques used to obtain bandwidths and center-frequency requirements.
3.6.2 The Calibration System
To ensure accurate results which are independent of variations in
laser operation which occur from shot to shot, and which affect band-
width, a calibration system was devised. This is shown schematically
in Figs. 3.6:7a and b. This calibration system analyses both output
beams simultaneously and permits the effective NO absorption cross
sections to be calculated from the data obtained.
Operationally, two beams of mutually-orthogonal polarizations at
the close-lying wavelengths Xj^ (W78.5 A) and X2 («*500 A) enter the
shielded encasement through the variable iris. Both beams impinge up-
on an air-spaced Glan-Taylor prism by which the beams are spatially
separated. The perpindicularly polarized beam, XJL, is directed through
slide #1 by a beam-steering mirror. Four percent of that beam is tapped
off by slide #1 and is focused by the 15-cm focal-length lens #1 onto
an RCA IP39 vacuum-tube photodiode through neutral density filters and
diffusion plate fl« The remaining portion of the beam passes through
a 1/2 m glass cell in which flows 609 ppm of N02 in N2 at just above
atmospheric pressure. The beam is then focused onto RCA IP39 photo-
89
-------
6AS \NLET
SHIELDED
EMCkStMENT
SI&NM.S 1,2,'
FMH IP31 S'
FBPM
USED AS TB.V66ERTO
SCOPES
STOPCOCK.
GAS
TEKTRONl* 5-^^ O£e\ULOSCoPE*
FIGURE 3.6;7a THE CALIBRATION SETUP
90
-------
FIGURE 3.6:7b PHOTOGRAPHS OF THE CALIBRATION SETUP
-------
diod* #2 through neutral density filters and diffusion plate #2.
The same type of configuration applies to the parallel-polarized
beam, X2t which uses component numbers 3 and <*. A portion of the par-
allel-polarized bean is tapped off by slide #3 onto RCA IP39 photovac-
uum diode #5, whose signal acts as a trigger to two Tektronix type 549
storage oscilloscopes. The signals corresponding to the portion of
the beams that propagate through the cell are delayed in time by two
500 ns delay cables. The signals are then displayed on the two stor-
age scopes along with the signals corresponding to the reference beams
that bypassed the cell.
The type of data obtained in this manner and the method of calcu-
lating the effective N02 absorption cross sections for each lasing line
from shot to shot are described below:
1. Four signals, two of which are displayed on each storage scope,
are obtained.
2. The two signals on the first scope correspond to the 1478.5 A
laser line. The first signal on storage scope 1 is used for
normalizing for any power fluctuations that may occur from shot
to shot. This allows relative measurements to be made. The
second signal, which is delayed approximately 500 ns, is par-
tially absorbed by the N02-N2 mixture in the cell.
3. The two signals on storage scope 2 correspond to the 4500 A
laser line. As before, the first signal is used for normali-
zation. The delayed signal is partially absorbed by the N02-
N2 mixture, but not as much as the 1U78.5 A laser line.
«*. To calculate the percent absorption and therefore the effective
92
-------
N02 absorption cross section for each laser pulse for each
laser line, it is necessary to know what losses are attributed to
the N02 in the mixture and what losses are attributed to the
cell itself. It is for this reason that an initial reading
is necessary for each lasing wavelength when only N2 is flow-
ing through the cell.
5. To calculate the percent absorption for each wavelength
(which is related to the effective absorption cross section)
only relative measurements are required. The percent absorp-
tion is calculated using Eq. 3.6:1 below:
A!
= 1 - Blw .
% absorption = 1 - Bhf » (3.6:1)
Al
Bltf/O
where,
A/BJW = the ratio of the signal that propagates through
the cell containing the N02-N2 mixture (the de-
layed signal) divided by the signal that bypass-
es the cell.
A/B|H/O « the ratio of the signal that propagates
through the cell containing only the N2 (the
delayed signal) divided by the signal that by-
passes the cell. This is done for each of the
lasing wavelengths.
6. To obtain the percent transmission of each lasing wavelength
from shot to shot, each A/B|W reading from each scope obtained
during each shot is normalized by the initial A/B|H/O reading.
93
-------
To obtain the percent absorption the ratio of the above ratios
is subtracted from unity.
7. The effective absorption cross section is related to the per-
cent absorption by:
ff--- = -ln(l - % absorption) (cm2), (3.6:2)
EFF _ C
where,
N - the concentration of N02 in the cell (cm**3), and
1 * the length of the cell (cm).
Figs. 3.6:8*a and b, and 3.6:9 a and b show curve traces of photographs of
the signals from the storage scopes in the calibration system. Figs.
3.6:8a and b correspond to the readings without N02 but with N2 in the
cell for the lines centered about 4478.5 A and 4500 A respectively.
Figs. 3.6:9a and b correspond to the readings with the N02-N2 mixture
in the cell for these lines. The effective absorption cross sections
for the lines centered about 4478.5 A and 4500 X for this particular
set of measurements are 6.36 X 10~*9 cm and 3.54 X 10 cm respec-
tively.
Knowing the calibrated values of the effective N02 absorption cross
sections for each of the simultaneous laser wavelengths for each laser
firing, permits allowance to be made for any variations of laser band-
widths which occur from shot to shot.
3.6.3 Tuning Technique
For operation of the LIDAR system, it is necessary that:
1. The center frequencies of both laser lines coincide with the
94
-------
TIME.
FIGURE 3.6;8a CALIBRATION SIGNALS FOR SIMULTANEOUS TWO-WAVELENGTH
OPERATION FOR THE 4478.5 A LASER LINE WITH N, IN THE
CALIBRATION CELL
95
-------
4 S(X> A
56 _
40
t-
-i
(U
z
tu
/
/
/
/
f
1 \
1 \
1
_.,J ,
-too A
\
A
/ i
/ \
/ \
/ i
/ '
/ i
i \
' i
» ' '
! \
\ ;
, ^x •*, to,
1 1 1 1 *' I ^"
oo 600 goo 'ooo /aoo
FIGURE 3.6:8b
CALIBRATION SIGNALS FOR0SIMULTANEOUS TWO-WAVELENGTH
OPERATION FOR THE «»500 A LASER LINE WITH N2 IN THE
CALIBRATION CELL
96
-------
AOO
I I I
ioo /ooO /2oo
FIGURE 3.6;9a CALIBRATION SIGNALS FOR SIMULTANEOUS TWO-WAVELENGTH
OPERATION FOR THE H478..5 A LASER LINE WITH N02-N2
MIXTURE IN THE CALIBRATION CELL
97
-------
Sfe.
41 .
44 .
40
2
3
u>
•z
I
H
ZOO
AsooA.
\ / \
v/ x^.
goo IQQQ
FIGURE 3.6;9b CALIBRATION SIGNALS FOR SIMULTANEOUS TWO-WAVELENGTH
OPERATION FOR THE *500 A LASER LINE WITH N02-N2
MIXTURE IN THE CALIBRATION CELL
98
-------
center frequencies of the two interference filters in the re-
ceiver.
2. The bandwidths of both laser lines must be less than the band-
widths of the respective interference filters.
3. Any increase in laser bandwidths must result in bandwidths
that remain less than the bandpass of the interference filters.
If the laser bandwidths are larger than the bandpass of the filters,
erroneous readings would result, i.e., the data obtained from the cal-
ibration system corresponds to the actual effective absorption cross
sections, whereas the receiver would be detecting only a certain por-
tion of the laser signal spectrum.
The operating characteristics of the receiving system could have
been designed so as to require both lasing bandwidths to be much larger
than the interference filter bandpass. This would eliminate the need
for a calibration system in the following manner. If both laser line
bandwidths were larger than the filter bandpass, then known spectral
quantities would be detected and determined exactly by the filter band-
pass. The exact effective absorption cross sections would then be
known and would not vary since they were determined by the filter char-
acteristics.
There are three reasons why this approach was not used. First,
when operating a LI ADR system, it is desirable to obtain readings from
useful ranges (up to 10 km). When detecting only a portion of the
laser backscatter, the signal is reduced and the attainable range is
decreased. Second, the signal to noise ratio for a given range de-
creases, since the signal is decreased. Third, in dye-laser systems.
99
-------
the dye lasing medium degrades due to repeated exposure to the pump
source. This has the effect of oonotonically decreasing the laser
bandwidth over a period of time. Therefore, there is a difficulty in
maintaining laser output signals whose bandwidths are much larger
than those of the receiver interference filters.
The tuning scheme was therefore developed to achieve:
1. Laser line center frequency and filter bandpass coincidence.
2. Desirable laser bandwidths.
3. Laser bandwidth fluctuations that continue to keep bandwidths
below the bandpass of the interference filters.
Fig. 3.6:10 shows the experimental setup used to check the system.
With the laser operated in the simultaneous mode, fractions of both
\
beams of mutually orthogonal polarizations were tapped off by slide #1
and were directed through variable polarizer #1, variable neutral den-
sity filter #1, and lens #1 (1 m focal length). The beams were then
focused onto diffusion plate #1, neutral density filter #1, and were
detected by an RCA 6217 photomultiplier attached to a 1/2 m Jarrell-
Ash scanning spectrometer. The input and output slits were set at 10/>
o
each to 'give a detectable resolution of 0.16 A. To compensate for
any laser power fluctuations, the signal obtained through the spec-
trometer was normalized by a signal from a RCA IP39 vacuum-tube photo-
diode used as the trigger source in the calibration system. A portion
of the main beam was tapped off with a pellicle and directed into the
calibration system where it was detected by the IP39 photodiode. Sig-
nals from the spectrometer and the IP39 were recorded photographically
on two Tektronix type 549 storage oscilloscopes.
100
-------
SCOPE
PHOTO riuuT \PLtER,.
5k tt
SPECTROMETER,
NORNA.UfcA.TlON
S\G>MAL FROM
CALlfcRATtON
SCT-Up
NEOTRAL
FILTER)
ScoPE
FIGURE 3.6:10 EXPERIMENTAL SETUP USED FOR FINE FREQUENCY TUNING
-------
The spectrometer was initially calibrated using an Hg lamp and was
then manually set to detect the 4478.5 A line. With the laser opera-
ting in the simultaneous mode, both beams were tapped off, but only
the line desired to lase at 4478.5 A was allowed to pass through var-
iable polarizer #1 initially. The laser line was initially sighted by eye
through the view port on the spectrometer. Coarse tuning was accom-
plished by observing the spectral shift while the grating's rotation-
al position was changed. Once having achieved the desired spectral
center line frequency, within visual approximation, the desired laser
bandwidth, approximately 5 X, was sought. As mentioned earlier, the
bandwidths must be less than the bandpass of the interference filters.
If the laser bandwidths were initially set at 5 X with the bias voltage
set at the maximum to be used, and knowing that the bandwidths monoton-
ically decrease due to repeated exposure of the lasing medium to the
pump source, the laser bandwidths would be ensured to always be within
the bandpass of the filters.
Initially, the laser bandwidths were adjusted for maximum width,
approximately 20 A, by setting the grating positions as close as pos-
sible to the polarizing prism. To obtain the desired 5 X bandwidths,
the distance of the gratings need only be increased relative to the pol-
arizing prism. To check the bandwidths each time the-' grating positions
were moved, a camera was attached to the view port on the spectrometer,
and the laser spectral widths were recorded photographically. Initial
exposure of the film to an Hg lamp through a Corning 3850 filter and
the spectrometer served as a calibration check on the bandwidth read-
ings. The resolution obtained on the film was 16 A/mm. Initially, the
102
-------
two laser lines night partially overlap since the bandwidth? were orig-
inally set at approximately 20 X. Therefore, variable polarizer #1 was
set to allow only one polarization to pass into the spectrometer initially.
Through repeated adjustments of the axial positions of the gratings, the de-
sired bandwidths were obtained with the center-line frequencies set vis-
ually at 4478.5 A and 4500 A. Figs.3.6slla through i show the run to
obtain the 5 A bandwidths for each of the simultaneous laser frequencies.
Having obtained the desired bandwidths, the next step was to finely
adjust the laser frequencies using the tuning technique described earlier
in the section* The rotational positions of -the gratings were finely adjusted
i
using micrometers and the normalized readings of the scopes were maxi-
mized. Figs. 3.6:12a.and b and 3.6:13a and b show the first and final scope
readings for fine center line adjustment for the simultaneous laser fre-
quencies 4478.5 A and 4500 A respectively. Fig. 3.6:14 shows the desir-
o
ed simultaneous output with each line approximately 5 A wide and the
O o
line center frequencies set at 4478.5 A and 4500 A.
In conclusion, the results of the experiments to determine whether
frequency shifts and bandwidth changes occur from shot to shot during
LIDAR operation, definitely show the necessity for a calibration system.
If meaningful pollution readings are to be obtained, calibrated output
signals must be used.
In the following chapter, calibrated LIDAR returns are interpreted
and pollution concentrations at various ranges from the LIDAR site are
given.
103
-------
'
I
TO
FIGURE 3.6;lla
SPECTROMETER PHOTOGRAPH OF THE SIMUL-
TANEOUS TWO-WAVELENGTH OUTPUT TO GETQ
5 /( BANDWIDTHS IN EACH LINE— UU78.5 A
FIGURE 3.6:llb —H500 A
-------
e FT To flifcWT'. 45ooi,44it.5AJ435S'A, 40i?A, AND
FIGURE 3.6 tile —THE U478.5 X AND U500 A LASER
LINES
FIGURE 3.6;lid —THE 1478.5 A AND 1500 LASER
LINES
-------
FIGURE 3.6 tile —THE HU78.5 A AND 4500 X LASER
LINE
FIGURE 3.6tllf --THE UU78.5 A AND H500 A LASER
LINE
-------
FIGURE 3.6;llg —THE U478.5 X AND 4500 X LASER
LINE
FIGURE 3.6;llh —THE HU78.H A AND «»500 X LASER
LINE
-------
FIGURE 3.6;lli —THE 4H78.5 A AND 4500 A LASER LINE
108
-------
<
-a
UJ
2
H
L
TIMS. (3.00
3
UJ
=>
£
-.J
£*
to
I
(2.00
FIGURE 3.6:12a
THE INITIAL FINE TUNING SIGNALS FOR SIMULTANEOUS TWO-
WAVELENGTH OPERATION FOR OBTAINING FREQUENCY COINCI-
DENCE FOR THE W78.5 A LASER LINE
109
-------
Ul
-4
ID
10
SKaNM-
L
TYM£ ^ zoo /vA/vosec/ o/^)
p
a
u)
I
U)
Of
z
ui
TIM S (200 /VANfl SEC. 101V)
FIGURE 3.6:12b
THE FINAL FINE TUNING SIGNALS FOR SIMULTANEOUS TWO-
WAVELENGTH OPERATION FOR OBTAINING FREQUENCY COINCI-
DENCE FOR THE 4*78.5 A LASER LINE
110
-------
Itf
7
Tine. (200 N*NO s EC/ON)
UJ
01
Of
I
H
FIGURE 3.6:13a
THE INITIAL FINE TUNING SIGNALS FOR SIMULTANEOUS TWO-
WAVELENGTH OPERATION FOR OBTAINING FREQUENCY COINCI-
DENCE FOR THE 4500 K LASER LINE
111
-------
§
iu
Cf
v*
~z
to
7//1E
e
2
UJ
|
ui
-------
FROrt LEFT TO
'. 4SOO A, 4*1? , ,
A , AM D AOA1
FIGURE 3. 6 tit SPECTROMETER PHOTOGRAPH OF THE SIMULTANEOUS
TWO- WAVELENGTH OUTPUT WITH LASER FREQUENCIES
FINE TUNED TO H478.5 A AND «*500 A WITH APPROX-
IMATELY 5 A BANDWIDTHS
113
-------
EFFICIENCY-PERCENT
100-
50'
MASTER =24 FR
IBOOg/mm
0
0.3
0.4 0.5
WAVELENGTH - MICRONS
0.7
GRATING TYPE
Replica on pyrex
aluminum coating
LASER POWER
CONTINUOUS
50 W/cm2
LASER POWER
PULSED
2 MW/cm2 on
20 - 30 ns
MAXIMUM
TEMPERATURE
100°C
Grating Characteristics
FIGURE 3.6:15 SPEC. SHEET 3.6:1
114
-------
l._.
-------
MPA.NY
BENTON HARBOR. MICHIGAN
CHART NO. 445-19
PRINTED IN U.S.A.
Interference-Filter Characteristics for Simultaneous Two-Wavelength
Operation
Spec. Sheet 3.6:1 Continued
116
-------
HEATH COMPANY
BENTCN HARBOR, MICHIGAN
Interference-Filter Characteristics for Simultaneous Two-Wavelength
Operation
Spec. Sheet 3.6:1 Continued
117
-------
CHAPTER 4 N02 MEASUREMENTS
4.1 Introduction
Measurements on N02 were carried out using laser outputs at 4478.5
A and 4500 A. These were optimized using the procedure described in
Sect. 3.6.3.
Alignment of the outgoing laser beam with the receiver telescope
i'1
was carried out by using a water tower on top of a building over 2 km
away as a target and adjusting the alignment of the receiver telescope
to obtain the maximum return signal from the tower. The outgoing beam goes
out eastwards from a corner window in our laboratory, and passes over
upper Manhattan and the East River.
4.2 Results of Field Tests
4.2.1 Sequential Measurements
Measurements were carried out in the sequential mode of operation
with pulses fired at intervals of approximately 5 seconds. The laser was
fired in trains of three pulses: on-peak, off-peak, and on-peak.
Measurements were only considered valid when the first and the third
pulse returns were nearly identical, indicating that no substantial
scene change had occurred between pulses.
With the LIDAR system pointing out of the laboratory window, meas-
urements of the LIDAR backscatter returns at 4478.5 X and 4500 A were
carried out. The return signal, the output of the photomultiplier
tube in the receiver system, was displayed on the oscilloscope and
recorded photographically.
118
-------
Fig. 4.2:1 shows a curve trace from an oscilloscope photograph of
a typical return signal. The peak occurring at 16 jis represents the
reflection from the water tower on top of a building some 2400 m away
from the LIDAR site.
Fig. 4.2:2 shows the trace of the return signal when a number of
chimneys below the bean path were emitting smoke.
Fig. 4.2:3a and b show respectively two curves plotted from oscil-
loscope photographs of the returns at 4478.5 X and 4500 A, and the cal-
ibration signals obtained for these readings. As can be seen from the
curves, the signal at 4478.5 A, which is on the absorption peak, was
clearly attenuated more than the signal at 4500 A, which is off the ab-
sorption peak.
The different attenuations may be noticed immediately by looking at
the reflection peaks from the building 2400 m (16 us) away.
The relative attenuation is largest over a 450 m region extending
from a range of approximately 1050 to 1500 m. Below the LIDAR beam in
that region are several avenues with heavy traffic. N02 densities over
that region are calculated from the curves to be approximately 0.46 ppm.
The overall average density to the building and back is approximate-
ly 0.09 ppm. Measurements of this type show a correlation within 30
percent of the sample measurements made at a. fixed-site, the Department of
Air Resources monitoring station situated near the area.
4.2.1.1 Factors Affecting Accuracy
In the initial field tests, the laser beam was paraxial with the
optical receiver axis. This arrangement was used to avoid local scat-
119
-------
FIGURE 4.2:1 TYPICAL LIDAR RETURN SIGNAL
%/*
>- t
£ 2
UJ
f\
FIGURE 4.2;2 TYPICAL LIDAR RETURN SIGNAL WITH CHIMNEYS BELOW BEAM
PATH
120
-------
NJ
FIGURE >».2:3a SEQUENTIAL LIDAR RETURN SIGNALS AT H«*78.5 A AND «*500 A
-------
5* .
34 _
U3
at
/
z
UJ
h
M
A
\
I 1
4418.5 A
4500 £
5.11*10
400 bOO gOO 1000 I3.OO
FIGURE »».2;3b SEQUENTIAL CALIBRATION SIGNALS AT 4478.5 A AND 4500 A
122
-------
taring from the outgoing beam into the receiver. This scattering prob-
lem was overcome with a redesigned optical arrangement with the out-
going beam collinear with the optical receiver axis. This permitted
the use of a much more collimated output beam along with a smaller ac-
ceptance angle for the optical receiver field of view. This has result-
ed in improved signal-to-noise ratios as well as reducing the importance
of shot to shot repeatability of the geometric intensity distribution of
the laser output pulse.
Other factors were also found to affect the accuracy of measure-
ments. With sequential operation, changes in the region probed by the
laser beam in the time interval between shots obviously cause errors.
Changes are of two types:
1. Scintillation effects, which cause rapid changes in the trans-
mission of the beam and backscattered signal through the atmos-
phere.
2. More gradual changes due to particulates and pollutants being
blown by wind.
Scintillation effects are particularly bad when there are large
temperature gradients across the beam path.
To check that neither of these effects is causing substantial er-
rors when sequential measurements are being made, the laser is operated
in groups of three pulses, with the first pulse being off-resonance,
the second on-resonance, and the third pulse off-resonance again, to
ensure no substantial changes with the first. It is found that these
conditions prevail particularly when there are no patches of particu-
lates, e.g., smoke drifting across the scene being viewed, and when
123
-------
there are no strong temperature gradients causing scintillation ef-
fects. The curves of Figs. 4.2:1 and 4.2:3a were obtained under such
conditions.
For conditions where the scene or optical path is changing' rapid-
ly, which is frequently the case, there appears to be no remedy but
to use simultaneous wavelength operation, so that both wavelengths are
traversing the same optical path at the same time.
4.2.2 Simultaneous Two-Wavelength Measurements
A series of field tests were carried out operating the laser in
the simultaneous two-wavelength mode of operation. For this purpose,
the arrangement described in Sect. 3.5.2 and Figs. 3.5:la and b was
used, as well as the calibration cell arrangement described in Sect.
3.6.2 and Figs. 3.6:7a and b.
Figs. 4.2:Ha through o each show typical pairs of curves plotted
from oscilloscope photographs of the LIDAR returns and the calibration
signals at 447S.5 & and 4500 A for various times during the day.
During the field tests it was noted that the readings from the
calibration system indicated that there was a monotonic narrowing in
laser bandwidth with the number of laser firings. This can readily be
seen in view of Figs. 4.2:4b, d, f, h, j, 1, and n. These curves show
that the difference in effective absorption cross sections increased
with the number of times the laser was fired.
The effective N02 absorption cross sections therefore varied from
5.43 X 10"19 cm2 and 4.17 X 10~19 cm2 to 6.36 X 10"19 cm2 and 3.54 X
-19 2 oo
10 cm for the 4478.5 A and 4500 A lines, respectively. This effect
was expected, and as explained in Sect* 3.6 is due to solarization of
124
-------
/3J
to
en
b>
x»
-* 1
$ 4
t
z 5
UJ
z ^
3
NORHAUHN6 POINT
T r
~r
3
-I - 1
1 S
nne
r-
IO
-T-
/-*
/7
FIGURE H,2;4a SIMULTANEOUS LIDAR RETURN SIGNALS AT *W78.5 A AND «*500 A, TIME: 8:«*5 A.M.
-------
.5 A
---- 4SOO
200 AQQ 400 600 /OOO 1140
FIGURE «»,2i4b SIMULTANEOUS CALIBRATION SIGNALS FOR 4478.5 A AND 4500
A, TIME: 8:45 A.M.
126
-------
to
-j
/3.
II.
II
10
t ^
2
ui • ^
iu
vn
Z
LU
i
/ _
POINT
T
T T
//
T T
~i 1 r
7 « «? 1C
TIME (/tsec)
FIGURE *.2iHc SIMULTANEOUS LIDAR RETURN SIGNALS AT W78.5 A AND H500 At TIME: 9:00 A.M.
T 1—>
6 II
-------
«,S A
---- 4500 A,
~'
FIGURE 4.2;Ud SIMULTANEOUS CALIBRATION SIGNALS FOR 4478.5 A AND 4500
A, TIME: 9:00 A.M.
128
-------
\O
a.
//.
10
D
£ 9
£
3
I
I
NORnA
-------
4A18.5A
4-i«.S
FIGURE 4.2;4f SIMULTANEOUS CALIBRATION SIGNALS FOR 4478.5 A AND 4500
A, TIME: 9:30 A.M.
130
-------
/3.
10 _
2
t
1
ul
of
til
NORMALHIN6 POINT
-\ 1 1 1 1 1 1 1 1 1 1 1 1 1 r~
/ A 34 S t, 1 8 1 16 II 13. 13 14 IS
Tine
/7
FIGURE *.2iHg SIMULTANEOUS LIDAR RETURN SIGNALS AT 4U78.5 A AND 4500 A. TIME: 5:30 P.M.
-------
ul
§
UJ
a:
Z
UJ
• i i
1OO 400 600 800 /OOO
FIGURE 1.2;*»h SIMULTANEOUS CALIBRATION SIGNALS FOR 4178.5 A AND 4500
A, TIME: 5:30 P.M.
132
-------
CO
/3.
tt.
(i
16
I '
-» 1
ID
UJ
A
3
2
PO\NT
FIGURE <*.2:
-------
SB.
4+ -
U)
s
2* -
3 A?
a _
44T8.SA
A
(T(4300A) » 3.14 x icT'^
i i r
iOO 400 600 850 IQOO 1100
Tin e C/VANO set)
FIGURE f.2;«»j SIMULTANEOUS CALIBRATION SIGNALS FOR W78.5 A AND «*500
A, TIME: 6:00 P.M.
134
-------
u>
ll
It
to
uJ
7 _
32
MJ
t-
NORHAUilNG POINT
II
s <, is 1 10 i/
Tine (/<.sec.)
iz 14 if
FIGURE U.2:
in
-------
36
U)
u
/6
4500 A
400 60O SCO ;ooo /ZOO
FIGURE »*.2:«»1 SIMULTANEOUS CALIBRATION SIGNALS FOR W78.5 A AND 4500
A, TIME: 6:30 P.M.
136
-------
/3
CO
8
MJ
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 , r
I A 34 5 to 7 S 1 to a iZ IT> 14 15 Ib 17
FIGURE «».2:4m SIMULTANEOUS LIDAR RETURN SIGNALS AT «*H78.5 A AND 4500 A, TIME: 8:00 P.M.
-------
44
40
3*
20
Ik
#
*
-foo feoo «oo 1000 /ioo
FIGURE U.2;Hn SIMULTANEOUS CALIBRATION SIGNALS FOR 4478.5 A AND «*500
A, TIME: 8jOO P.M.
138
-------
----- 4500
A
2OO 4OO 60O IOO
FIGURE H.2t*Q SIMULTANEOUS CALIBRATION SIGNALS FOR U«*78.5 A AND «V500
A FOR NORMALIZATION (WITH N2 IN THE CALIBRATION CELL)
139
-------
r
(m)
1050
750
FROM
LIDAR
SITE
TO
BLDG.
T^
(m)
450
300
2400
Time
8:45 A. M.
9:00 A. M.
9:30 A. M.
5:30 P. M.
6:00 P. M.
6:30 P. M.
8:00 P. M.
8:45 A. M.
9:00 A. M.
9:30 A. M.
5:30 P. M.
6:00 P. H.
6:30 P. M.
8:00 P. M.
8:45 A. H.
9:00 A. M.
9:30 A. M.
5:30 P. M.
6:00 P. M.
6:30 P. H.
8:00 P. M.
H02 Concentration
(pptn)
0.29
0.31
0.21
0.26
0.24
0.15
0.14
0.31
0.22
0.22
0.19
0.15
0.16
0.11
0.11
0.11
0.09
0.09
0.08
0.06
0.06
Uncertainity
(ppm)
0.05
0.07
0.009
TABLE 4.2:1 RANGE, RESOLUTION, N02 POLLUTION CONCENTRATION,
AND UNCERTAINTIES FOR VARIOUS TIMES DURING THE DAY
140
-------
the dye and degradation of gain following many laser firings during the
course of the day.
From the traces of the LIDAR returns, Figs. 4.2:4a, c, e, g, i, k,
and mt it is seen that in each case the relative attenuation is largest
over a 450 m region extending from a range of approximately 1050 m to
1500 m. This result corresponds well with the fact that over that re-
gion the LIDAR beam crosses over several avenues with heavy traffic.
Table 4.2:1 lists the pollution concentration measured at various
times during a typical day. The range location of particular interest to ob-
serve pollution changes during rush hours is located approximately 1200
m from the LIDAR site (the region where areas of heavy traffic exist).
Table 4.2:1 shows that the readings obtained at the range of 10SO m and
depth resolution of 450 m do in fact evidence peaks in pollution levels
during the morning and evening rush hours, while dropping lower there-
after. From i:45 a.m. to 9:00 a.m. the pollution level remained approx-
imately the same (0.29 ppm to 0.31 ppm respectively), then declined to
0.21 ppm at 9:30 a.m.. During the evening rush hours the peak N02 con-
centration reached 0.26 ppm at 5:30 p.m. then declined to 0.24 ppm at
6:00 p.m.. Thereafter, the level of M02 pollution reduced steadily from
0.15 ppm at 6:30 p.m. to 0.14 ppm at 8:00 p.m.. The overall average N02
density observed from the building (2400 m or 16 us away) to the LIDAR
site ranged from 0.11 ppm to 0.06 ppm from 8:45 a.m. to 8:00 p.m. respec-
tively.
These average density results were generally found to be within 30
percent of the values measured by the New York City Department of Air
Resources fixed-site monitoring station situated a short distance to the
south of the LIDAR beam trajectory.
141
-------
4.3 Error Analysis
For the practical interpretation of the LIOAR pollution measure-
Dents it is useful to know the uncertainity that is associated with
the measurement. Towards this end, an experiment was performed in
which a known concentration of N02 was monitored by the LIBAR system.
, i
The experimental setup is shown in Fig. 4.3:1. The simultaneous
laser output passed through a 1/2 m sample cell which contained 609
ppm of N02 in H2. The output was then directed into the atmosphere
and the backscattered returns were monitored by the receiving system.
Each output pulse was monitored by the calibration system and the sig-
nals obtained from detectors 1 and 3 in the calibration setup served
as the bef ore-cell signals for both the calibration and sample cells.
The experiment was performed on a windy clear night such that clear-
air conditions existed, and that the output signals would not be fur-
ther attenuated by ambient levels of N02.
By knowing the before cell readings and the signals backscattered
from the atmosphere, the pollutant concentration in ppm*&r(m) was ob-
tained for a series of 5 shots. The mean value of the deviations from
the known concentration in the cell (609 ppm in 1/2 m = 304.5 ppm'&r
(m)), was calculated to be 21 ppm*&r
-------
u>
r
LASER OUTPUTS I C.NM-S
I
FIGURE 4.3:1 EXPERIMENTAL SETUP FOR DETERMINATION OF LIDAR-READING UNCERTAINTIES
-------
HEAN OEVtKTIONi* 20.^ ppM •
400
BdO
100
NC\ CONCENTRATION
IN SAMPLE CELL
FIGURE 4.3;2 ERROR PLOT FOR DETERMINATION OF LIDAR-READING
UNCERTAINTIES
144
-------
hour tiroes and leveled off thereafter. The pollution levels calculat-
ed from the returns are typical of the levels found at intersections
and avenues of heavy traffic*
An experiment was performed to obtain quantitative uncertainties
associated with each pollution reading obtained in the simultaneous
node of operation. The variations in the readings obtained by the
LIDAR system in monitoring the known concentration of N02 were primari-
ly attributed to ambient levels of N02 present during the time of the
experiment.
145
-------
CHAPTER 5 SIMULTANEOUS MULTIWAVELENGTH OUTPUTS FROM ENERGY-TRANSFER
DYE-MIXTURE LASERS
5.1 Introduction
In a second generation LIDAR system it would be desirable to have
the capability of monitoring several pollutants simultaneously. This
can be accomplished by lasing simultaneously in the spectral regions
necessary to monitor the pollutants of interest.
Thus, though the work being reported here centered primarily on the
detection of NO., an experimental program was carried out on dye mix-
tures for the purpose of:
Obtaining simultaneous outputs at two pairs of wavelengths, one
Q
in the 4500 A spectral region (for monitoring N02> and the oth-
e o
er in the 6200 A spectral region. (The outputs in the 6200 A
region could then be used, in conjunction with frequency dou-
bling, for monitoring S02 and 03 in the 3100 A spectral region.)
The program carried out was successful and simultaneous laser opera-
tion was obtained at two and even three wavelengths (in the three pri-
77
mary colors) for the first time.
To achieve these results, use was made of energy transfer processes
in organic dye mixtures. Energy transfer processes in organic dyes have
received some consideration in the past. " The results of the exper-
iments described below conclusively demonstrate that:
1. In a mixture of two or more dyes in a solution, excitation ener-
gy may be very efficiently transferred from an excited molecule
of one species to another, and that the process may be repeated
146
-------
successively to a third dye with the excitation energy being
efficiently cascaded to still longer wavelengths.
2. Furthermore, simultaneous outputs at several widely separated
spectral regions may be obtained from a dye-mixture laser.
3. Dye mixtures can be used to reduce the required concentration
of the lasing dye, while still maintaining effective absorp-
tion of the pump soruce radiation. The reduced concentration
of the lasing dye reduces self-absorption losses at the laser
wavelengths. This in turn reduces threshold upper level popu-
lation density requirements, and therefore the required pump
source intensity.
A second series of experiments was carried out to define the energy
transfer processes involved. Possible processes were either radiative
or nonradiative resonance transfer or a combination of the two. Two
experiments were devised which required only static measurements to be
performed.
The next section considers the theoretical background for energy
transfer processes that could occur at the concentrations found optimum
for simultaneous laser action in dye-mixture lasers.
5.2 Energy Transfer Mechanisms
5.2.1 Introduction
The characteristics of a donor-acceptor system in which intermolec-
ular energy transfer plays the dominant role are illustrated schemati-
cally in Fig. 5.2:1. There, the fluorescence emission spectrum of donor
molecule D overlaps its own absorption spectrum as well as that of the
147
-------
It)
-p-
oo
U)
I-
2:
LENGTH
DONOR-DMORTRKU^E DOMOR-KCCEPTOR TR\US PER
FIGURE 5.2:1 DONOR-ACCEPTOR SYSTEM OF TWO ORGANIC DYES WITH OVERLAPPING FLUORESCENCE AND ABSORPTION
BANDS
-------
acceptor molecule A. These spectra indicate an overlap in the energy
manifolds of the transitions in D and the absorptive transitions in D
and A, thus permitting excitation transfer from D to D and from D to A*
Energy transfer along a chain of donor to donor molecules and from don-
or to acceptor molecules are therefore possible. It is also clear that
such a process can be cascaded down through several dyes, e.g., D to A
to B to C. This is shown in Fig. 5.2:2.
A second type of dye mixture in which simultaneous emission is pos-
sible is illustrated in Fig. 5.2:3. In this case the absorption bands
of dyes D and A overlap, while the emission bands are widely separated.
Thus, excitation of such a mixture will result in simultaneous excita-
tion of, and emission (and possibly laser emission under the appropri-
ate conditions) from both dyes.
The main mechanisms that can be responsible for the intermolecular
energy transfer between the donor-donor pairs and donor-acceptor pair
. , 76,90-92
molecules are: '
1. radiative transfer by means of radiation by the donor dye and
reabsorption by the donor and acceptor dyes.
2. singlet-singlet nonradiative transfer between donor and donor
molecules and between donor and acceptor molecules due to
coulombic interactions (dipole-dipole etc.) and exchange or col-
lision induced nonradiative decay back to S0 (ground state).
5.2.2 Radiative and Nonradiative Energy Transfer
In the case of radiative transfer, excitation transfer rates are re-
lated in a relatively straightforward manner by the values of the absorp-
149
-------
ABSORPTION
Ui
o
C
UJ
fe
-4
U)
a:
IU
V-
FIGURE 5.2{2 DONOR-ACCEPTOR SYSTEM OF FOUR ORGANIC DYES WITH OVERLAPPING FLUORESCENCE AND ABSORPTION
BANDS
-------
z
3
It)
-J
Ul
•
IU
»~
H
\
\
\
FIGURE 5.2t3 DONOR-ACCEPTOR SYSTEM WITH OVERLAPPING ABSORPTION BANDS
-------
tion and emission cross sections of the acceptor and donor dye and
their extent of overlap, and takes place over distances greater than
50 to 100 L
Analysis of these factors show that the probability of radiative
90
transfer, K£T(RAD)f is given by:
Kj^RAD)^ 2.303CNA3l (fn(\))EA(V)dT), (5.2:1)
QD o
where,
1 - the specimen thickness,
f« a the donor normalized fluorescence spectrum,
ED * the acceptor absorption coefficient,
[NA] = the molar concentration of acceptors, and
Q = the quantum efficiency of the donor.
The mechanism of singlet state transfer of energy between donor
93
and acceptor molecules was analyzed by Forster, According to Forster,
the probability of nonradiative transfer from a donor to an acceptor
molecule is proportional to the overlap of the fluorescence band of the
V
donor and the absorption band of the acceptor molecule. This is a reso-
nance transfer mechanism and is illustrated in Fig. 5.2:4. Radiation-
less transfer due to coulombic interaction takes place over distances
less than 50 to 100 A depending upon the molecules involved in the pro-
cess. The probability of nonradiative energy transfer, KET(NOMRAD),
from an excited fluorescent molecule D to an acceptor molecule A is
expressed below:
KET(NOMRAD) ' l/jU6, (5.2:2)
where,
152
-------
s,
U).
s.
K.ST
AA/VWWV**
s,
DONOR
ACCBPTOR
FIGURE 5.2:4 DONOR AND ACCEPTOR ENERGY LEVELS
153
-------
00
Rg6 * (90001nlOfo2(fr(v')Eft(y)dV, (5.2:3)
128(TTpKn'»)N) V*
0
where,
t0 = the fluorescence lifetime of isolated donors,
R * the distance between D and A,
n = the refractive index,
fjj * the normalized fluorescence intensity of D,
E = the absorption coefficient of A,
A
2
•X, * a geometrical factor «X > = 2/3), and
RQ = the critical transfer distance at which there is equal
probability of radiative or nonradiative resonance trans-
fer to occur.
5.3 Multiwavelength Laser Action in Dye Mixtures—Experimental Results
In the first series of experiments carried out, dye mixtures were
chosen from which simultaneous laser outputs at useful spectral combi-
nations could be obtained. A mixture of 7- diethylamino-4 methyl coumarine,
Acriflavine, and Rhodamine'B allowed dye laser outputs to be obtained at
77
three primary colors.
Fig. 5.3:1 shows the absorption and fluorescence spectra of these
three dyes. The absorption bands of Acriflavine and Rhodamine are seen
to overlap the fluorescence bands of the Coumarine and the Acriflavine
respectively. This overlap in the energy gaps of these dyes would per-
mit, in principle, excitation energy transfer from Coumarine to the
Acriflavine and from the Acriflavine to Rhodamine B either by radiative
or nonradiative resonance transfer.
154
-------
Ui
Ln
bSOQ
FIGURE 5.3:1
ABSORPTION AND FLUORESCENCE SPECTRA OF THE THREE DYES AT OPTIMUM CONCENTRATIONS FROM
WHICH SIMULTANEOUS LASER OUTPUTS WERE OBTAINED IN THE THREE PRIMARY COLORS
-------
It was therefore possible to use the 3371 A output of an N2 laser
to excite the Coumarine dye and obtain excitation transfer to, and
simultaneous emission (including laser emission) from* all three dyes
in a mixture with appropriate concentration ratios.
Fig. 5.3:2 shows the absorption and fluorescence spectrum of this
three-dye mixture at concentrations found optimum for three-color
laser operation. The absorption curve shown is for a 1.0 mm optical
path length, and ethanol was the solvent used. Excitation for the
front surface fluorescence measurements depicted in Fig. 5.3:2 was by
0
means of the 3650 A output obtained from an Hg discharge lamp. This
is absorbed by the Coumarine, which transfers excitation to the Acri-
flavine, which in turn transfers excitation to the Rhodamine B. Fig.
5.3:3 shows the experimental setup used to obtain the absorption and
emission curves of the dyes.
The different concentrations used for each of the three dyes arises
from the fact that the excitation source for laser action in the dye
mixture was the 200-kw 30-ns 1-HZ H2 laser output at 3371 A. The ab-
sorption of this pump light was therefore predominantly by the Coumar-
ine dye, hence its large concentration requirements relative to the
other two dyes. Thus, in this case, the type of dye mixture used would
be a combination of the types depicted in Figs. 5.2tl, 2, and 3. With
this three dye mixture, laser action was obtained simultaneously at
three wavelengths centered at 4400 A, 4900 A, and 6200 A. These out-
puts could be individually tuned over a range of approximately 15 A, as
shown in Fig. 5.3:3.
That excitation transfer is the predominant mechanism for exciting
156
-------
Ul
100.
ACRIFLKVINE UKSER (\*IO
-4
RHODMAINE B LKSER. (1-5*10 *n|jQ
35OO
6500
FIGURE 5.3;2
ABSORPTION AND FLUORESCENCE SPECTRA OF THE THREE-DYE MIXTURE AT OPTIMUM CONCENTRATIONS
FROM WHICH SIMULTANEOUS LASER OUTPUTS WERE OBTAINED IN THE THREE PRIMARY COLORS
-------
HG LANP
00
/ \
V ARABLE
NEUTRAL-DEHSITX
X-Y RECORDER /
\
1
»(
L
PH(
4-|
i
3TO
/
/
L
PHOTOMETER
"STRIP-CHART
RecoRDER
uv PINVO
PHOTODlOOe
LENS
-X
PHOTOMULTlPUER Is' \
\
\
KEuTRJkL-DENSITY
SLt DE
\
7
V-
.-ni.
DVECELL /
/ DXE CELL.
-M JARRE LL-
HeNe DETECTOR
FIGURE 5.3:3 EXPERIMENTAL SETUP FOR OBTAINING FLUORESCENCE AND ABSORPTION SPECTRA
-------
laser action in the Acriflavine and Rhodamine B is confirmed by the
fact that it was not possible to obtain laser action from Acriflavine
alone, at any concentration, using the 3371 X N2 laser output for ex-
citation, while laser action by direct excitation at 3371 A of Rho-
damine B alone was found to require an order of magnitude higher con-
centration of Rhodamine B with the same optical arrangement used.
See Fig. 5*3:4.
For two-wavelength operation in the 1500 A and 6200 A region, the
concentration of the center dye, Acriflavine, was reduced, and that
of the Rhodamine B increased. This resulted in effective laser oper-
ation at the desired spectral regions. Fig. 5.3:5 shows the spectrum
of the dye mixture for two-wavelength operation. An optical arrange-
ment suitable for getting simultaneous laser action at two pairs of
wavelengths in each of the spectral regions of interest is shown in
Fig. 5.3:6.
By comparing the fluorescence quantum yield for each of the three
dyes alone with the fluorescence quantum yield for the dye mixtures,
it was found that the fluorescence quantum yield for the single-step
transfer process, Coumarine to Acriflavine, was approximately 90 per-
cent. The yield for the two-step transfer process, Coumarine to Acri-
flavine to Rhodamine B, was approximately 80 percent.
5.U Energy Transfer Mechanisms—Experimental Results
5.4.1 Radiative Energy Transfer
If energy transfer were predominantly a radiative process at the
concentrations required for simultaneous laser action, the energy
159
-------
OUTPUT
CfLlNOR»CKV_ LENS
MIRROR.
FIGURE 5.3;t OPTICAL ARRANGEMENT FOR OBTAINING SIMULTANEOUS THREE-COLOR OUTPUTS IN THE THREE PRIMARY
COLORS
-------
100-
(l*«o
-v
35OO 4OQO ASOQ Sooo 0 SSoO &o'oo &S0O
FIGURE 5.3:5 ABSORPTION AND FLUORESCENCE SPECTRA FOR SIMULTANEOUS TWO-WAVELENGTH OUTPUT
-------
6RKTIN&
to
LASER OUTPUT A.T
cue.es,
FIGURE 5.3:6 OPTICAL ARRANGEMENT FOR OBTAINING TWO PAIRS OF WAVELENGTHS SIMULTANEOUSLY
-------
transfer rate would be given by Eq. 5.2:1. The energy transfer rate
from D to A would be directly proportional to the overlap integral, Pt
as defined in Eq. 5.2:1. Thus, the emission from the acceptor molec-
ules, A, would be proportional to the product of the quantum efficiency
of A and the overlap integral P.
Two pairs of dyes were chosen such that the absorption spectrum of
each acceptor dye overlapped the flourescence spectrum of the same do-
nor dye differently. Relative measurements were made to determine the
dominant mechanism for energy transfer in the following manner:
* Fluorescence of AI in mixture 1 ,
P2&Q2 flourescence of A^ in mixture 2
(5.4:1)
where,
A1 = acceptor dye #1,
&2 ~ acceptor dye #2,
P! 2 = tna overlap integrals of the fluorescence spectrum of
donor dye, D, with the absorption spectrum of acceptor
dy«t Ai,2» and
t
nQl 2 s thc
-------
X.-X RECORDER. /
SPECTROMETER.
FIGURE S.H;1 EXPERIMENTAL SETUP FOR OBTAINING FLUORESCENCE AND ABSORPTION SPECTRA
-------
the same concentration. The first mixture consisted of Dichloroflu-
orescein (2 X 10~* M/l), donor, and DODC (5.52 X 10"5), acceptor, in
ethandl. The second mixture consisted of Dichlorofluorescein (2 X
1
-------
OVeR.\.A,P tNTEGRM. P,
0.1
626?
FIGURE
OVERLAP INTEGRAL OF DICHLOROFLUORESCEIN (DONOR) AND DODC
(ACCEPTOR)
166
-------
Pa.
(R€LCT\ve UNITS)
SSI 5" 5764 SI 33. UOO
6437
FIGURE 5.4;3 OVERLAP INTEGRAL OF DICHLOROFLUORESCEIN (DONOR) AND
RHODAMINE B (ACCEPTOR)
167
-------
l-i-
l.o.
o.f-
a.%.
2
3
ttl
uj
I-
i/j
Z.
0.4
0.3.
T
5764
FLUORESCENCE
DO DC C5.S2.xto~3
AREA* °l~> (RBUMIME UNITS')
QUAjNTUM EFFlClEMtY
I I I I
64 3 T
FIGURE S.Utf FLUORESCENCE QUANTUM YIELD OF DODC IN THE MIXTURE WITH
DICHLOROFLUORESCEIN
168
-------
r i
5411 5515 576?
FIGURE 5,*»tS FLUORESCENCE QUANTUM YIELD OF RHODAMINE B IN THE MIXTURE
WITH DICHLOROFLUORESCEIN
169
-------
ceptor relative to the fluorescence of the donor in the same mixture.
Before this critical concentration, energy transfer would take place
by the radiative process. The emission of the donor would be partial-
ly absorbed by the acceptor. As the donor concentration is increased
and the mean distance between donors decreases (along with a decrease
in donor-acceptor distance), a point will be reached at which the en-
ergy transfer flow would take place from donor to donor to donor to*..
to acceptor. This "hopping" phenomenon is observed during the process
of photosynthesis but at concentrations much higher than those used in
dye-mixture lasers.
Two organic dyes were chosen such that the fluorescence band of
the donor overlapped the absorption band of the acceptor, and that the
acceptor would not absorb at the pump wavelength. 7-diethylamino-*t
methyl coumarine and DODC were chosen as the donor and acceptor dyes
respectively. Ethanol was used as the solvent throughout. The accept-
or concentration was kept constant at 1 X 10" M/l (R = 158 A), while
the donor concentration varied from 1 X 10"1* M/l (R. = 158 A, R = 126 A)
to 5 X 10" M/l CR = 20 A, R * 20 A). Fluorescence curves were obtain-
ed for the pure donor, and the donor and acceptor in the same mixture.
The experimental setup used to obtain the fluorescence curves is shown
in Fig. 5.4:6. The pump light, 3700 A and 50 A bandwidth, was obtain-
ed from a xenon-lamp 1/4-m Jarrell-Ash monochrometer combination. The
output of the monochrometer went through a variable neutral density
filter and was then focused onto a 1 cm quartz cell containing the dye
solutions. The cell was pointed at such an angle that the reflection
of the pump light from the front surface of the cell would not be pick-
ed up by the detection system. The detection system used to monitor the
170
-------
-t RECORDER.
V4-M TARR.6U-ASH
MOHOCHROHBTER
PY6CELU
DISCHKRG.E
FIGURE 5.4:6 EXPERIMENTAL SETUP FOR OBTAINING FLUORESCENCE SPECTRA
-------
front surface fluorescence of the mixtures consisted of a 1/2-m Jar-
rell-Ash scanning spectrometer whose light output was detected by an
RCA 6217 photomultiplier. The output of the phot©multiplier went to
a photometer, the output of which went to an x-y recorder.
The results of the experiment are shown in Figs. 5.4:7 through 11.
Fig. 5.4:7 shows the effect of concentration quenching of pure solu-
90 94
tions of the donor dye, 7-diethylamino-4 methyl coumarine. * Fig.
5.4:8 shows the quantum yield of the donor in mixtures with the accept-
or dye, DODC. The slope in Fig. 5.4:8 is steeper than in Fig. 5.4:7
90 94
indicating energy transfer to the acceptor. * For dilute concen-
trations of the donor, the energy transfer is due to the radiative pro-
90
cess. This can be seen in Fig. 5.4:11, which is the quantum yield of
the acceptor dye normalized with respect to the fluorescence of donor
dye in the same mixtures. The flat portion of the curve extending from
-4 -2
donor concentration of 1 X 10 H/l to 2 X 10 M/l is due predominant-
ly to the radiative process. As the donor-donor distance decreases
still further there is an increase in the relative quantum yield of the
acceptor. If the radiative process were the dominant energy transfer
mechanism in this region, 2 X 10" H/l to 5 X 10~2 M/l, the curve would
have remained flat. The increase in the quantum yield starting at 2 X
-2
10 M/l is therefore due to the nonradiative process. At the donor
f\
concentration of 5 X 10 M/l the nonradiative process is the dominant
energy transfer mechanism.
Hence, the dominant energy transfer mechanisms found for various
regions of donor-donor concentrations are as follows:
1. radiative transfer for the region extending from 1 X 10~4 M/l to
172
-------
1010-
XtEUO
M1tNO-
Puee SOIOTIOM V
couMM»IH€
1000.
Ul
IU
of
i
no.
ISO.
R «1E*,M OISTKMCE
TO-TD
/x 10
I* 10'*
10
-9
ma
LSI
33.
To
DONOR CON CENT* ATI o*J ,
FIGURE 5.«*:7 FLUORESCENCE QUANTUM YIELD OF 7-DIETHYLAMINO-H METHYL COUMARINE VERSUS CONCENTRATION
-------
wo-
1010
1006
i/l
l-
IU
2 lio
y-
K
(2(9
740
13*
It to
5*
"
1 -I
DONOR
(n\JL\t
FIGURE 5.4;8 FLUORESCENCE QUANTUM YIELD OF 7-DIETHYLAMINO-4 METHYL
COUMARINE (DONOR) IN MIXTURES WITH DODC (ACCEPTOR, 1 X 10
M/l) VERSUS DONOR CONCENTRATION
174
-------
2I.S-
3(1.0.
30.0.
IU
11.0.
I9.S.
5*10
I V /<3
-3
-3
5* 10
-2
/X/O
-2
DONOR. coNcfNTRATiON/ (
*
FIGURE 5.1;9 FLUORESCENCE QUANTUM YIELD OF DODC (ACCEPTOR, 1 X 10'** M/l) IN MIXTURES WITH 7-DIETHYL-
AMINO-H METHYL COUMARINE (DONOR) VERSUS DONOR CONCENTRATION
-------
24-
21-
I-
2
i
20-
It-
NORttM.\%fcO NtVK XtEi-O Of OOQC
e
OF
Of n-OlETHUftMlNO-X METWSV.
/o
10
-3
TT-
i _^_ _i
/5fC/Z6.) 3l(fl) 21(7^
CONCENTIZAT/O
FIGURE S.HtlO NORMALIZED FLUORESCENCE QUANTUM YIELD OF DODC (ACCEPTOR, 1 X 10"H M/l) IN MIXTURES WITH
7-DIETHYLAMINO-t METHYL COUMARINE (DONOR) WITH RESPECT TO THE FLUORESCENCE QUANTUM YIELD
OF PURE SOLUTION OF THE DONOR VERSUS DONOR CONCENTRATION
-------
24-
25-
24-
21-
t"
z
Zl
fc
5 ^o.
-A-
/x/o
I
s* 10
-4
I
It 10
-3
-s
/<«T2
10
-i
74
DONOR
iswruwix *>»-'i^'-tr»**-iiiviN'\iii-*.|* i^. i F\ I) r\\nf
FIGURE S.U;11 NORMALIZED FLUORESCENCE QUANTUM YIELD OF DODC (ACCEPTOR, 1 X 10"1* M/l) IN MIXTURES WITH
7-DIETHYLAMINO-H METHYL COUMARINE (DONOR) WITH RESPECT TO THE FLUORESCENCE QUANTUM YIELD
OF THE DONOR IN THE SAME MIXTURE VERSUS DONOR CONCENTRATION
-------
2 X 10"2 M/l,
2. the onset of nonradiative resonance transfer at approximately
2 X 10"2 M/l, and
2
3* nonradiative resonance transfer at 5 X IQ~ M/l.
In conclusion, experiments were designed to obtain simultaneous out-
puts at two (and three) wavelength regions. The experiments were suc-
cessful in that multiwavelength dye-mixture lasers were developed in
spectral regions necessary to monitor the pollutants of interest, namely,
NC>2, SC>2, and 03.
Investigations of the energy transfer processes in dye-mixture
lasers were carried out. Successful determination of the dominant en-
ergy transfer mechanisms was achieved.
To explore the dynamic flow of energy transfer in dye-mixture las-
ers, use would have to be made of time resolved spectrescopy using pico-
second laser techniques. By pumping the donor dye with a picosecond
laser pulse and observing the rise of the acceptor dye's fluorescence
at various acceptor concentrations, the radiative and nonradiative en-
ergy transfer rates could be determined. Only static measurements were
made in this work. This is suggested as another topic for future
studies.
The LIDAR system developed in this work was designed to monitor
ambient levels of HO?. It would be desirable if the LIDAR system had
the capability of simultaneously monitoring several pollutants, for
example, N02, S02, and 03. Use could be made of the results obtained
in achieving simultaneous laser action in dye mixtures and the inves-
tigation of the processes involved. The spectral regions obtained in
178
-------
the dye-mixture lasers that could be used in pollution detection would
Xo o
and 6200 A. The 1500 A region could be used to monitor
N02, and the 6200 A region could be frequency doubled into the ultra-
violet to 3100 A to monitor S02 and 03. This would be an attractive
feature of an integrated LIDAR system, and is suggested as another
topic for future studies in the LIDAR area.
179
-------
CHAPTER 6. DASE LIDAR SYSTEM APPLIED TO OZONE AMD SULFUR
DIOXIDE MEASUREMENTS
6.1 Absorption Spectra of Ozone and. .Sulfur Dioxide
The ozone molecule shows strong absorption in the infrared and the
ultra-violet regions of the spectrum. In the ultra-violet, (Fig. 6.1.1)
the main bands are known as the Huggins band, which begins at 345Q& and
extends to approximately 3000A, and the Hartley band which overlaps the
Huggins band and represents the strongest ozone absorption. The Hartley
band consists of a broad continuum extending between 3000A and 2200A with
o
a high peak near 255OA.
In contrast to ozone, sulfur dioxide has strong, discrete and sharp
absorption bands in the near ultra-violet region, Fig; 6.1.1.
The ozone continuum corresponds to the dissociation of the molecule
while in the sulfur dioxide case a similar dissociation occurs at slightly
higher energies (2280A).
In considering the application of the DASE system to the measurement
of ozone and sulfur dioxide, it is clear that the interference of the two
absorption spectra (of ozone and sulfur dioxide) must be taken into account.
Table 6.1.1 shows the absorption coefficients of ozone and sulfur
dioxide in cm~ atm .
O
For measurements of ozone concentration the wavelength pair at 2923 A
o
and 2940 A were chosen. At these wavelengths, the absorption
due to sulfur dioxide is the same and hence its effect will cancel out
for ozone measurements.
180
-------
E ICO
o
C "°
s-t
-------
For ozone,
^(at 292.3) -
-------
Presnel Lens (15")
00
Mirrc
< \
^s.
\
Mirrc
"^ A
>r ; I ---•••
^ y I :..
Temp . Oven
&, ADA crystal Mirror
On
U
>r Collimator
f
Apertur
f
01
Filter Photomultiplier
Collimating
Lens
Flashlamp Pumped
Dye Laser
Oscilloscope
Disperaive element
(prisro or grating)
Temp. Oven controller
Fig. 6.2,1 Lidar System components
-------
unobtainable directly from flashlamp pumped dye lasers. Because of the
high cost of the temperature tuned ADA frequency doubling element, and
the financial constraints on the project, it was decided to use only one
crystal to carry out the measurements for ozone and sulfur dioxide by
transmitting different required wavelengths sequentially. While not as
desirable as using tow simultaneous near-uv wavelengths which could have
been obtained by using two temperature-tuned frequency-doubling crystals
in conjunction with the beam-splitting prism arrangement used for the
nitrogen dioxide measurements, it was still possible to successfully
demonstrate the viability of the systems for ozone and sulfur dioxide by
averaging series of sequential measurements at the desired wavelengths.
The optical receiver used was the same as that used for nitrogen
dioxide but with a 15" diameter, uv transmitting, acrylic Fresnel lens;
a 1-1/2" uv transmitting collimating lens; together with a uv transmitting
filter and a Centronic photomultiplier.
The center wavelength and bandwidth of the filter are such that all
the ozone and sulfur dioxide probe wavelengths are passed by it.
The same dispersive elements used for the nitrogen dioxide measure-
ments were used to select the wavelengths necessary for the
DASE system and to obtain the narrow linewidth required for efficient
harmonic generation.
The return signal is displayed on an oscilloscope and recorded
photographically as for the nitrogen dioxide measurements.
6.3.1 Harmonic Generation and Choice of Crystal
96
Since the pioneering work of Franken , there has been extensive
research, both theoretically and experimentally, in the field of non-
linear optics, particularly second harmonic generation (SHG).
The appearance of a second harmonic is basically due to the nonlinear
polarization of the crystal. The linear dependence between the polarization
184
-------
and the electric field (the basis of pre-laser optics) is no more than
an approximation. A better picture is:
P - P.+ P ,
1 nl
(2) (3)
where, P^ XE and P^ - x 11 + X ill + ..., with X the linear
susceptibility and x the susceptibility tensor.
The second harmonic is then described by the first term in the non-
linear polarization, Pnl. The electric field is determined by the in-
homogeneous wave equation:
A x A x E - tfeE/ t »
where, e is the permittivity tensor.
The classical model of radiation of slowly moving dipoles
can be used in calculating the ratio of the power densities at the
second harmonic and fundamental. The result is 16:
n - (dL IE,|2/n,n_)(16n3z2/*?) sin*(nz/L )/(nz/L )2 (6.3.1)
JO ' 1' i Z C C
for P(2a))«P(to). Where d_, is the piezoelectric constant, related
36
to the susceptibility by xi:j-2 d±.J nlf n£ are the indices of refraction
at Xj and Xa, respectively; z is the distance into the medium, and Lc
is the coherence length (Lc • X/2|n.- n2| ).
In a dispersive medium, n2 t BI, the intensity of the second harmonic
increases only when zL£ and
oscillates with distance for z L (Maker oscillations). Eq. 6.3.1 is a
when L +0" , (phase matching condition).
This condition can be obtained in some crystals by making use of
the birefringence, that is, the variation of index of refraction with
polarization and direction with respect to the optic axis (z axis).
185
-------
/•>
The direction in which n2(0)=n° is known as the direction
of collinear phase matching. However, the Poynting vectors for the funda-
mental (o wave) and second harmonic (e wave) are not in the same direction.
This has the effect of reducing the interaction between the two waves,
2
thereby decreasing the efficiency by a factor of sin (0). It is difficult
in practice to achieve sufficiently good alignment to avoid losses due to this
factor. These losses can be minimized by choosing a crystal whose phase
matching angle is 90°. Because of the variation of index with temperature,
this can easily be accomplished by temperature tuning of the crystal
The tuning ranges of several crystals for 90° phase matching are
shown in Fig. 6.3. As can be seen, the spectral ranges required for
o
ozone and sulfur dioxide measurements, 2900-3000 A lie at the center
of the tunable band for ADA. In addition, the nm linear coefficient for
ADA is a relatively high 0.75. These factors led to the choice of a
temperature-tuned ADA crystal as the frequency doubling element.
6.4.1 Factors Affecting Second Harmonic Generation Power Saturation
Equation 6.3.1 shows that efficiency of frequency doubling increases
linearly with input power (if beam depletion is neglected). At very
high power densities, saturation effects can be predicted theoretically.
However, the damage threshold of the doubling crystal is generally
reached well below the saturation value. In the experiments carried out,
the input power to the frequency doubling crystal was as large as was
permissible without damaging the crystal.
186
-------
00
o
o
~100- KDP
04
80
-------
Beam Divergence and Diffraction
Beam divergence of the fundamental pump wavelength effectively
causes a mismatch from the optimum 90° phase matched condition and would
therefore reduce the efficiency of the frequency doubling. Among the
factors affecting divergence is diffraction.
For a collinear flashlamp pumped dye laser the diffraction is large
(>10mr) and varies from shot to shot. The shot to shot variation is caused
by the non-uniform heating of the dye which arises because energy is absorbed
non-uniformly and therefore the dye solution closest to the flashlamp is
heated the most, causing refractive variations, radially, and producing a lens
effect. It has also been conjectured that a shock wave is produced within the
dye solution which causes a larger index variation than is produced by non-
uniform heating of the dye.
To minimize these effects, a circular adapter cel^ was fitted around
the dye cell, Fig. 6.4.1. This had the effect of concentrating the same
flashlamp output into a smaller base laser cell which had a smaller beam
divergence and was less susceptible to distorting effects of temperature
gradients and shock waves.
6.5.1 Experimental Results
Field measurements in ozone and sulfur dioxide ware carried out
over the same optical paths as in the nitrogen dioxide experiments described
o o
in previous sections. The probe wavelengths used were: 2923 A and 2940 A
o o
for ozone, and 2923 A and 2933 A for sulfur dioxide.
188
-------
Coolant
In
Fig. 6.4.1
Coolant
Out
v ' v:
Annular Region of Flash
: •x"1:1 ;;x~--—-x— x_',.""• i\-1" s. ""x • "V
Dye
In
Window
Dye Out
Tlashlamp fitted with adaptor to further isolate
dye solution from the annular region of the flash
(Crossectional view). The coolant and dye sollution
temperatures are kept within 1° of each other by
flowing both through the same cool water bath.
ON
oo
-------
As discussed in Section 6.2.1 above, financial constraints permitted
the purchase of only one temperature tuned ADA crystal, consequently
measurements at the three different wavelengths were carried out
sequentially.
The experimental arrangements for carrying out the measurements was
that shown schematically in Fig 6.2.1. The procedure used was essentially
the same as that used earlier to carry out sequential measurements of
nitrogen dioxide which was described previously in Section 4.2.1.
For the measurements at each wavelength at least 6 laser probe
shots were fired and the returns recorded. After each set of measurements
at one wavelength, the grating at the end of the laser cavity was then
rotated, and the ADA crystal temperature changed to obtain the next
wavelength at which measurements were to be taken. The information
required at each wavelength was then obtained by taking the average of
each set of 6 (or more) recorded returns.
o
For the ozone measurements the probe wavelengths at 2923 A and
o
2940 A were used in conjunction with a calibration cell containing
sulfur dioxide.
The use of the calibration cell with sulfur dioxide ensured that
the absorption due to the sulfur dioxide was the same at both these
wavelengths and that its effects were therefore cancelled out as far as
ozone calculations were concerned, as discussed in Section 6.1.1 above.
To obtain the sulfur dioxide concentrations additional measure-
o
ments were made at the 2933 A probe wavelength, again in conjunction
with the sulfur dioxide calibration cell. Along with the returns at the
o o
2923 A and 2940 A wavelengths this permitted the attenuation due to ozone
190
-------
to be allowed for and the sulfur dioxide concentration to be calculated
as discussed in Section 6.1.1.
Several series of tests were carried out in which measurements of
sulfur dioxide and ozone were made. The results of two of these series,
which were typical are presented below.
The two series of measurements were each carried out on summer
afternoons a few days apart. For each of the series twenty five lidar
returns were obtained at each required probe wavelength. Figs. 6.5.1
through 6.5.6 show the first four returns for each of the three probe
wavelengths, for each series.
The relative attenuation was measured in each photograph between
the time intervals of 3-4 u sec and 5-6 u sec. These time intervals
correspond approximately to ranges of 550 m and 850 m respectively, with
a range resolution (or sample length) of 150 m. These attenuation values
were then averaged (over the 25 pictures) and used (in conjunction with
the absorption coefficient obtained from the calibration cell) to
calculate the ozone concentration by means of equation 2.4.11.
These calculations yielded the following ozone concentrations:
First Series
0.09 ppm at 550 m (or 3-4y sec)
and 0.1 ppm at 850 m (or 3-4^ sec)
Second Series
0.08 ppm at 550 m (or 3-4 ju sec)
and
0.09 ppm at 850 m (or 5-6 u sec)
191
-------
Figure 6.5.1 LIDAR returns at 2933A for first series,
192
-------
Figure 6.5.2 LIDAR returns at 29^0A for first series.
193
-------
Figure 6.5.3 LIDAR returns at 2952Afor first series.
-------
Figure 6.5.4 LIDAR returns at 2923A for second series
195
-------
Figure 6.5.5 LIDAR returns at 2933A for second series
196
-------
Itttt
mmvmmmm
Wi^HI^B™ VIHI^MN^RIlHV^^HilNHP ^I^NHNNiiNHNRpi^Hii^HI
jB^yiyj^B ^UBMM|^^Utt|^y»^B^yB«j^^«ay»y|M|^»^£B — -
ri
Figure 6.5.6 LIDAR returns at 29^0A for second series
197
-------
Using these concentrations and averaging the attenuation in
o
the returns for the third wavelength, at 2933 A and making use of the
absorption coefficient obtained from the calibration cell, equation
2.4.11 was again used to obtain the following sulfur dioxide concen-
trations:
First Series
0.05 ppm at 550 m (or 3.4 y sec)
and
0.08 ppm at 850 m (or 5-6 y sec)
Second Series
0.048 ppm at 550 m (or 3-4 y sec)
and
0.047 ppm at 850 m (or 5-6 y sec)
The average concentration over the distance to a water tower
1 km away was also obtained by examining the relative attenuation of
the pulse echo from it (seen at the right hand side of the photographs
of the returns). Making use of the attenuation averaged over the 25
returns, and equation 2.4.28, the following average concentrations
were obtained for ozone and sulfur dioxide over the 1 km path to the
water tower:
First Series
ozone : 0.08 ppm
and
sulfur dioxide : 0.06 ppm
198
-------
Second Series
ozone : 0.06 ppm
and
sulfur dioxide : 0.05 ppm
The results reported by the New York City Department of Air
Resources for the City College - Upper Manhattan area were obtained
and were averaged over the same period that the lidar measurements were
made. This yielded the following average concentrations:
First Series
ozone : 0,07 ppm
and
sulfur dioxide : 0.05 ppm
Second Series
ozone : 0.065 ppm
and
sulfur dioxide : 0»04 ppm
As can be seen the long path average absorption results and the
Department of Air Resources results agree fairly well. The discrepancy
is greater for the DASE measurements at the 550 m and 850 m ranges.
This, however is to be expected since the DASE system measured pollution
at specific points (over 150 m sampling lengths) where local traffic
and other pollution conditions could be expected to vary from the averages
for the Upper Manhattan region.
In contrast the long path absorption method (to the water tower)
can be expected to yield the average concentrations over the Upper
Manhattan region. This is in fact borne out by its closer agreement
with the averaged readings from the Department of Air Resources over
199
-------
approximately the same region for the same period,
200
-------
SUMMARY
Introduction
"""""•"""y™^"'
A single-ended laser radar (LIDAR) system was designed, built and
successfully operated to measure N02, S02 and 0 concentrations in the
atmosphere. The LIDAR technique used was the Differential Absorption of
Scattered Energy (DASE) technique. In its simplest form, the DASE LIDAR
technique determines the concentrations of a pollutant along the
transmitted LIDAR beam by examining the selective attenuation of outgoing
LIDAR signals, transmitted at different wavelengths, by the pollutant
molecules present. More specifically, the concentration of pollutant
present in an incremental path length (or sample length) in space (along
the LIDAR beam path) is obtained by determining the relative attenuation
(across the sample length) of two colinear laser beams at close lying
wavelengths, each respectively on and off the resonance absorption of
the pollutant molecule in question. The relative attenuation is
determined from comparisons, (at the LIDAR receiver) of the Rayleigh and
Mie atmospheric elestic backscatter of the two outgoing laser beams as
they traverse the sample length. Appropriate temporal resolution at the
receiver then permits determination of the range and spatial resolution
of the pollutant distribution.
The LIDAR System
The basic LIDAR system was built around a flashlamp-pumped organic
dye laser as the source of laser energy. For measurements of N02> whose
O
absorption spectrum lies in the 4500 A spectral regions, it was possible
to use the output of the dye laser directly for the transmitted LIDAR
signals. For the measurements of S02 and 03, whose absorption spectra
201
-------
lie in the near ultraviolet (2900 A) spectral regions, it was necessary
to frequency-double the output of the dye laser, by means of a
temperature controlled ADA crystal frequency doubler, to obtain the
transmitted LIDAR wavelenghts required.
The LIDAR system as used for the NO™ measurements was designed to
be capable of operation in two basic modes. In the first mode, the outputs
o o
at the two required wavelengths, 4500 A and 4478.5 A, respectively on and
off the resonance absorption of NOj are obtainable, in rapid sequence with
frequency selection being carried out by an electromechanically-driven
diffraction grating, whose position at one-end of the organic dye laser
may be rapidly and accurately changed between laser shots.
In the second mode of operation, an intra-cavity dielectric-
interface polarizing beam-splitter was used to simultaneously obtain the
o o
pair of close lying wavelengths at 4500 A and 4478.5 A required from
the laser. The use of this system represented the first successful use
of a DASE LIDAR system operating in a simultaneous two wavelength mode.
In this mode of operation, errors due to scintillation effects and
particle and aorosol drift are effectively reduced.
The systems developed for both sequential and simultaneous operation
incorporated an automatic calibration feature which allowed for the
variations in bandwidth which inevitably occur from shot to shot with dye
lasers. Experimental results are presented, which show the necessity and
advantages of this automatic calibration approach. The calibration system
operates by sampling the transmission of part of each outgoing laser pulse
through a test cell containing samples of known concentrations of the
pollutant being measured by the LIDAR system. This approach automatically
allows for the variations in laser batsdwidths and determines the effective
202
-------
pollutant absorption cross-section for each outgoing laser pulse. Again,
this work represented the first reported use of this automatic calibration
approach which we believe to be essential to the successful operation of
LIDAR systems of this type.
The LIDAR system receiver was designed and built using a 15 inch
diameter and 24 inch focal length acrylic Fresnel lens as the primary
lens. This lens is capable of forming a sharp image for objects near the
optical axis. This is appropriate, since LIDAR receivers do not require
a wide field of view for collecting the backscattered light from the
outgoing laser beam. The light from the Fresnel lens is the collected and
collimated then split onto two photomultipliers with appropriate optical
filters to inatch the transmitted wavelengths.
Measurements of NO-, SO- and 03
With the errors in the system minimized through the use of
simultaneous two wavelength operations, NO- measurements were made with
the LIDAR transmitter directed out of a laboratory window across the
upper East Side of Manhattan. Several readings were obtained for various
times during the day to observe changes in pollution levels over areas of
heavy vehicular traffic.
From the LIDAR returns obtained and recorded, range-resolved
calculations were made of pollution concentrations in the atmosphere at
distances of 250 and 1000 meters from the LIDAR receiver. In addition
average pollution concentrations between the LIDAR transmitter and a
fixed target (the water tower on a building) 2400 m away were also
calculated. Concentrations varying from 0.06 to 0.31 ppm were obtained
depending on the time of day. The highest numbers were for the
203
-------
concentrations calculated for the range-resolved returns from 700 m and
1000 m, which are from the atmosphere above heavily trafficked areas.
Measurements of S02 and 0- were carried out using the frequency
doubled output of the dye laser. Because of financial constraints only
one temperature tuned ADA crystal was available for frequency doubling *
Measurements of S02 and 0, were therefore carried out in a sequential
mode of operation. Because of overlap of the absorption spectrum of
S02 and ()„, it is first necessary to determine 0- concentrations then
SO- concentrations. This procedure required LIDAR returns to be
o
obtained at three different wavelengths in the 2900 A spectral region.
These returns were obtained sequentially and results averkged for a
series of (typically 25) laser shots at each wavelength.
From these averaged returns, range resolved atmospheric concentrations
of 0 and S02 were calculated at distances of 550 m and 850 m respectively.
Average concentrations between the LIDAR transmitter and a fixed target
(building) 1000 m distant were also calculated. S02 concentrations varied
from 0.04 to 0.08 ppm, while 0, concentrations varied from 0.07 to 0.1 ppm
depending on the location and time of day (with the highest numbers being
obtained for the 850 m range).
To compare the results of the LIDAR measurements, information was
obtained from the New York City Department of Air Resources on pollution
levels recorded at their various monitoring stations on the upper
East Side, at the same time that the LIDAR measurements were being carried
out. It was found that the averaged readings for these stations, which
cover the area of the LIDAR tests, were generally within 15 percent of the
average concentrations obtained by the LIDAR technique for the same region.
204
-------
Dye Mixture Lasers
To examine the potential for second generation LIDAR systems
capable of operating simultaneously in two and three spectral regions for
the simultaneous monitoring of several pollutants, a program of experiments
on dye-mixture lasers was also carried out. The objective of this was to
use energy transfer mechanisms that were considered feasible in dye
mixtures to obtain simultaneous laser action at widely separated spectral
regions. The program was successful, with simultaneous laser action being
obtained for the first time at widely separated spectral regions,
including simultaneous outputs at the three primary colors. Analysis of
the experimental results showed that radiative energy transfer was dominant
at the concentrations found optimum for simultaneous two and three-
wavelength operation in dye-mixture lasers. Nonradiative resonance energy
transfer was found to dominate at higher concentrations.
Conclusion
The work reported here describes the development, design and
construction of a LIDAR system operating in the visible and near ultra-
violet spectral regions for the measurement of N02, S02 and 0^. Field
tests with the system successfully demonstrated the potential for LIDAR
methods based on the DASE technique to be used for range-resolved
monitoring of ambient N02, S02 and 0^.
205
-------
REFERENCES
1. L. B. Lave and E. P. Seskin, Science, 169, (1970), pp. 723-731.
2. H. Kildal and R. L. Byer, Proc. IEEE, £9_, (1971), pp. 1644-1663.
3. T. Kobayashi and H. Inaba, 'Laser Beam Technology for Remotely
Sensing Invisible Molecules in the Polluted Atmosphere1, lltn
Symposium on Electron, Ion and Laser Beam Technology, ed. R. F.
M. Thornley, San Francisco Press, Inc., (1971), pp. 385-401; see
also S. Nakahara, K. Ito, A. Fuke, S. Koraatsu, H. Inaba and T.
Kobayashi, Opto-electronics, 4, (1972), pp. 169-177.
4. R. T. H. Collins, Appl. Opt., 9_, (1970), pp. 1782-1788.
5. C. S. Cook, Appl. Opt., 11, (1972), p. 1742.
6. G. Fiocco and L. D. Smullin, Nature (London), 109, (1963), pp.
1275-1276.
st
7. M. G. H. Ligada, Proc. Conf. Laser Tech., 1 San Diego,
California, U. S. Navy ONR, (1963), p. 62.
8. D. A. Leonard, Nature, 216, (1967), pp. 142-143.
9. J. A. Cooney, Appl. Phys. Letts., 12, (1968), pp. 40-42; see also
J. Appl. Meteorology, 9_, (1970), pp. 182-184.
10. S. H. Melfi, J. D. Lawrence, Jr. and M. P. McCormick, Appl. Phys.
Letts., 15_, (1969), pp. 295-297.
11. H. Inaba and T. Kobayashi, Nature, 224, (1969), p. 170; see also
Opto-electronics, £, (1970), pp. 45-46.
12. T. Kobayashi and H. Inaba, Appl. Phys. Letts., 17, (1970), pp.
139-141.
13. J. A. Cooney, J. Appl. Meteorology, 10, (1971), pp. 301-308.
14. R. G. Strauch, V. E. Derr and R. E. Cupp, Appl. Opt., 10, (1971),
pp. 2665-2669.
15. S. H. Melfi, Appl. Opt., 11, (1972), pp. 1605-1610.
16 T. Hirschfeld, E. R. Schildleraut, H. Tannenbaum and D. Tannenbaura,
Appl. Phys. Letts., 22, (1973), pp. 38-40; see also T. Hirschfeld
and S, Klainer, Opt. Spect., (1970), pp. 63-66.
17. T. Hirschfeld, Appl. Opt., 13, (1974), pp. 1435-1437.
18. R. L. Byer, Opt. and Quant. Elect., £, (1975), pp. 147-177.
19. G. B. Jacobs and L. R. Snowman, IEEE J. Quant. Elect., QE-3, (1967),
pp. 603-605.
206
-------
20. P. L. Hanst and J. A. Morreal, J. Air Pollut. Cont. Ass.t 18,
(1968), pp. 754-759. —
21. P. L. Hanst, Appl. Spect., 24_. (1970), pp. 161-174.
22. S. Zaromb, Prcc. Electro-Optical Systems Design Symposium, Sept.,
(1969), p. 609.
23. S. Nakahara and K. Ito, International Quantum Electronics Conf.
Digest Tech. Papers, Sept., (1970), p. 226.
24. H. Inotnata and T. Igarashi, Trans. Tech. Group on Quant. Elect, of
the Inst. Elect. Comm. Engr. (IECE) of Japan, QE-70-36. Dec.
25. J. A. Hodgeson, W. A. McClenney and P. L. Hanst, Science, 182,
(1973), p. 2»»8.
26. R. L. Byer and M. Garbuny, Appl. Opt., 12_, (1973), pp. 1496-1505.
27. E. D. Hinkley, Appl. Phys. Letts., 16_, (1970), pp. 351-354; see
also K. W. Nill, F. A. Blum, A. R. Calowa and T. C. Harman, Appl.
Phys. Letts., 19» (1971) pp. 79-82.
28. E. D. Hinkley and P. L. Kelley, Science, 171, (1971), pp. 635-639;
see also I. Melngailis, IEEE Trans, on Geo. Sci. Elect., GE-10,
(1972).
29. E. D. Hinkley, Opto-electronics, j*, (1972), pp. 69-86.
30. L. R. Snowman. Technical Report R72ELS-15, General Electric
Electronic Laboratory, Syracuse, New York, March, (1972).
31. D. O'Shea and L. G. Dodge, Appl. Opt., 13_, (1974), pp. 1481-1486.
32. T. Henningsen, M. Garbuny and R. L. Byer, Appl. Phys. Letts., 24,
(1974), pp. 242-244.
33. M. R. Bowman, A. J. Gibson and M. C. W. Sandford, Nature, 221,
(1969), p. 456; see also A. J. Gibson, J. Sci. Instr., 2_, 7l969),
p. 802.
34. M. C. W. Sandford and A. J. Gibson, J. Atmos. Terr. Phys., 32_,
(1970), pp. 1423-1430.
35. A. J. Gibson and M. C. W. Sandford, J. Atmos. Terr. Phys., 33_,
(1971), p. 1675.
36. C. J. Schuler, C. T. Pike and H. A. Miranda, Appl. Opt., 10, (1971),
pp. 1689-1690.
37. R. D. Hake, D. E. Arnold, D. W. Jackson, W. E. Evans, B. P. Ficklin
and R. A. Song, J. Geophys. Res., 77, (1972), p. 6839.
207
-------
38. J. E. Blamont, M. L. Chanin and 6. Megie, Ann. Geophys., 28,
(1973), pp. 833-838.
39. V. W. J. H. Kirchoff and B. B. Clemesha, J. Atm. Terr. Phys., 35,
(1973), p. 1493.
40. A. J. Gibson and M. C. W. Sandford, Nature, 239, (1972), pp. 509-
511.
41. F. Felix, W. Keenliside, G. Kent and M. C. W. Sandford, Nature,
246, (1973), p. 345.
42. C. M. Penney, W. W. Moray, R. L. ST. Peters, S. D. Silverstein, M.
Lapp and D. R. White, Final Report for NASA CR-132363, prepared by
General Electric Corporate Research and Development, Schenectady,
New York.
43. R. M. Measures and G. Pilon, Opto-electronics, 4, (1972), pp. 141-
153.
44. S. A. Ahmed and J. S. Gergely, 'Atmospheric Pollution Monitoring
Using Tunable Dye Lasers', Proc. Tech. Prog., Electro-Opt. Systems
Design Conf., New York, N. Y., Sept., (1973), pp. 275-282.
45. J. A. Gelbwachs, M. Birnbaum, A. W. Tucker and C. L. Fincher,
Opto-electronics, 4_, (1972), pp. 155-160.
46. R. M, Schotland, E. E. Chermack and D. T. Chang, Proc. First
International Symp. of Humidity and Moisture, p. 569 (Reinhart
Book Div., New York, N. Y., 1964); see also R. M. Schotland, Proc.
Third Symp. on Remote Sensing of the Environment, Oct. 14-16,
(1964), pp. 215-224, Univ. of Michigan, Ann Arbor, Mich.
47. S. A. Ahmed and J. S. Gergely, 'Laser Radar Study Using Resonance
Absorption for Remote Detection of Air Pollutants', Fifth Conf. on
Laser Radar Studies of the Atm., June 4-6, (1973), Williamsburg,
Va.
48. S. A. Ahmed, Appl. Opt. 12_, (1973), pp. 901-903.
49. M. L. Wright, E. K. Proctor, L. S. Gasiorek and E. M. Liston,
Final Report for NASA CR-132724, Contract NA SI-11657, Stanford
Research Institute, Menlo Park, California.
50. T. Igarashi, 'Laser Radar Study Using Resonance Absorption for
Remote Detection of Air Pollutants', Fifth Conf. on Laser Radar
Studies of the Atm., June 4-6, (1973), Williamsburg, Va.
51. K. W. Rothe, U Brinkmann and H. Walther, Proc. VIII ICPEAC, Beograd,
July, (1973); see also Appl. Phys., £, (1974), pp. 115-119.
52. W. B. Grant, R* D. Hake, Jr., E. M. Liston, R. C. Robbins and E. K.
Proctor, Jr., Appl. Phys. Letts., 24, (1974), pp. 550-552
208
-------
53. K. W. Rothe, U. Brinkman and H. Walther, Appl. Phys., to be
published.
54. R. Penndorf, J. Opt. Soc. Am., 47^ (1957), pp. 176-182.
55. R. F. Begley, A. B. Harvey and R. L. Byer, Appl. Phys. Letts.. 25,
(1974), pp. 317-390. —
56. H. Inaba and T. Kobayashi, Opto-electronics, 4, (1972), pp. 101-
123. -
57. P. H. Kruse, L. D. McGlauchlin and R. B. McQuistan, 'Elements of
Infra-Red Technology1, Wiley, New York, (1963).
58. W. K. Pratt, 'Laser Communication Systems1, Wiley, New York, (1969).
59. H. Walther and J. L. Hall, Appl. Phys. Letts., 17, (1970), p. 239.
60. R. C. C. Leite, Phys. Rev. Letts., 14, (1965), p. 7.
61. R. C. Damen, Phys. Rev. Letts., 14, (1965), p. 9.
62. R. M. Schotland, J. Appl. Meteorology, 13, (1974), pp. 71-77.
63. 6. W. Grams and C. M. Wyman, J. Appl. Meteorology, 11, (1972), pp.
1108-1113.
64. R. S. Lawrence and J. W. Strohbehn, Proc. IEEE, 58, (1970).
65. H. S. Pilaff, Appl. Phys. Letts., 21^, (1972), pp. 339-340.
66. P. Dezauzier, A. Eranian and 0. deWitte, Appl. Phys. Letts., 22,
(1973), pp. 664-666.
67. A. A. Friesem, U. Ganiel and G. Neumann, Appl. Phys. Letts., 23,
(1973), pp. 249-251.
68. P. Flamont and Y. H. Meyer, Opt. Comm. 13_, (1975), pp.13-16.
69. H. Furtumoto and H. Ceccon, Appl. Phys. Letts. 13_, (1968), pp. 335-
337.
70. K. B. Eisenthal, Accounts of Chem. Res., £, (1975), pp. 118-124.
71. I. B. Berlman, 'Handbook of Fluorescence Spectra of Aromatic
Molecules', Academic Press, New York, (1965).
72. H. J. Bernstein, Paper presented at the 1973 Spring Meeting of the
Optical Society of America, New York, N. Y., April 11-13.
73. H. P. Rosen, P. Robrish and 0. Chamberlain, 'Feasibility of the
Remote Detection of Pollutants Using Resonance Raman Scattering',
Lawrence Berkeley Lab., Energy and Environment Div., U. C. Berkeley,
California, (1975).
209
-------
7«*. P. Bied-Charreton, C. R. Acad. Sci., Paris, 268, (1969), pp. 1377-
1380.
75. C. E. Moeller, C. M. Verier and A. H. Adelraan, Appl. Phys. Letts.,
18_, (1971), p. 278.
76. T. Urisu and K. Kajiyama, J. Appl. Phys., 47, (1976), p. 3563.
77. S. A. Ahmed, J. S. Gergely and 0. A. Infante, J. Chew, Phys., 61,
(1974), p. 158*.
78. T. Urisu and K. Kajiyaroa, J. Appl. Phys., £7, (1976), p. 3559.
79. J. H. Drake, E. M. Tarn and R. I. Morse, IEEE J. Quant. Elect.
(CorrepJ, QE-8, (1972), pp. 92-94.
80. W. W. Morey, IEEE J. Quant. Elect. (Corresp.), QE-8, (1972), pp.
818-819.
81. D. D. Whawalkar and L. Pandit, IEEE J. Quant Elect. (Corresp.),
QE-9, (1973), pp. 43-46.
82. C. Lin and A. Dienes, J. Appl. Phys., 44, (1973), p. 5050.
83. F. B. Dunning and E. D. Stokes, Opt. Conun., 6_, (1972), p.160.
84. E. D. Stokes, F. B. Dunning, R. F. Sfebbings, G. K. waiters and R.
D. Rundel, Opt. Conan., j5, (1972), p. 267.
85. R. C. Hilborn and H. C. Brayman, J. Appl. Phys., 45, (1974), p.
4912.
86. A. Dienes and M. Madden, J. Appl. Phys., 44, (1973) p. 4161.
87. C. V. Shank A. Dienes and W.T. Silfvast, Appl. Phys. Letts., 17,
(1970), p. 307.
88. 0. G. Peterson and B. B. Snavely, Bull. Am. Phys. Soc., 13, (1968),
p. 397.
89. J. B. Marling, D. W. Gregg and S. J. Thomas, IEEE J. Quant. Elect.,
QE-6, (1970).
90. J. B. Birks, 'Photophysics of Aromatic Molecules', edited by J. B.
Birks (Wiley-Interscience, Hew York, 1970).
91. F. Wilkinson, 'Advances in Photochemistry', edited by W. A. Noyes,
Jr., G. S. Hammond and J. N. Pitts, Jr. (Interscience, New York,
1964), Vol. III.
92. I. B. Berlman, 'Energy Transfer Parameters of Aromatic Molecules',
Academic Press, New York, (1973).
210
-------
93. Th. Forster, 'Light and Organic Crystals', Modern Quantum Chemistry,
Sinanoglu, editor, Academic Press, (1965); Th. Forster, Ann. Phys.,
2_t (1948), p. 55; Z. Naturforsh, A 4^ (1949), p. 32.
94. 6. S. Beddard, S.E. Carlin and G. Porter, Chem. Phys. Letts., 43,
(1976), pp. 27-32.
95. T. D. Wilkerson, B. Ercoli and F. S. Tomkins, 'Absorption Spectra
of Atmospheric Gases', Tech. Note BN-784, Institute for Fluid
Dynamics and Applied Mathematics, Feb., (1974).
211
-------
TECHNICAL REPORT DATA
I Meant read Instructions or. the rcrc.-sc /"Ywv completing)
RE?OF*T NO.
EPA-600/2-80-049
. Till.: ANDSUBTITLE
REMOTE MONITORING OF GASEOUS POLLUTANTS BY
DIFFERENTIAL ABSORPTION LASER TECHNIQUES
6. PERFORMING ORGANIZATION CODE
3. RECIPIENT'S ACCESSION NO.
5. REPORT DATE
February 1980
. AUTHOK(S)
8. PERFORMING ORGANIZATION REPORT NO.
S. A. Ahmed, J. S. Gergeley, and F. Barone
. PERFORMING ORGANIZATION NAMc AND ADDRESS
Electrical Engineering Department
The City College of the City University
140th Street and Convent Avenue
New York, N. Y. 10031
10. PROGRAM ELEMENT NO.
1AD712 BA-041
11. CONTRACT/GRANT NO.
803109
t. SPONSORING AGENCY NAME AND ADDRESS __
nvjronmental Sciences Research Laboratory - RTP, NC
Office of Research and Development
U. S. Environmental Protection Agency
Research Triangle Park, N. C. 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final 6/74-6/78
14. SPONSORING AGENCY CODE
EPA/600/09
5. SUPPLEMENTARY NOTES
6. ABSTRACT
A single-ended laser radar (LIDAR) system was designed, built, and successfully
operated to measure range-resolved concentrations of N0£, SO?, and 03 in the atmos-
phere using a Differential Absorption of Scattered Energy (DASE) LIDAR technique.
The system used a flash-lamp pumped dye laser as the primary source of laser energy.
For the N02 measurements, the dye laser output was used directly in a novel simul-
taneous two wavelength output mode in which two wavelengths, one on and one off the
resonance absorption of N0£ molecules are transmitted simultaneously and the relative
attenuation determined for the two backscattered signals detected. This mode of
operation effectively reduces errors due to scintillation and aerosol drift. For the
S02 and 03 measurements, it was necessary to frequency double the output of the dye
laser to match the absorption spectra of the S0£ and 0? molecules. Field measurements,
which were carried out over the Upper East Side of Manhattan'for all three pollutants,
produced range-resolved concentrations at ranges of over two kilometers. The ambient
pollutant concentrations measured ranged from 0.04 ppm to 0.31 ppm, depending on
location and time of day. In general, these showed reasonably good correlation with
measurements obtained from conventional pollution monitoring stations in the area and
demonstrated the potential of DASE LIDAR systems for range-resolved ambient pollutant
measurements.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS,/OPEN ENDED TERMS
COSATI Field/Group
*Air pollution
Nitrogen dioxide
Sulfur dioxide
Ozone
*Remote sensing
*0ptical radar
Development
T3T
07B
14B
17H
3. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (Tltts Report)
UNCLASSIFIED
21. NO. OF PAGES
226
20. SECURITY CLASS (Thispage)
UNCLASSIFIED
22. PRICE
EPA Form 2220-1 (9-73)
212
------- |