Citation
Plasma spectrochemistry with a fourier transform spectrometer

Material Information

Title:
Plasma spectrochemistry with a fourier transform spectrometer
Creator:
Manning, Thomas Joseph John, 1961-
Publication Date:
Language:
English
Physical Description:
ix, 135 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Argon ( jstor )
Fourier transformations ( jstor )
Inductively coupled plasma mass spectrometry ( jstor )
Lighting ( jstor )
Line spectra ( jstor )
Plasmas ( jstor )
Signals ( jstor )
Spectrometers ( jstor )
Spectroscopy ( jstor )
Supernova remnants ( jstor )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1990.
Bibliography:
Includes bibliographical references (leaves 127-133).
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Thomas Joseph John Manning.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright [name of dissertation author]. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
025013958 ( ALEPH )
AHR3706 ( NOTIS )
24160927 ( OCLC )

Downloads

This item has the following downloads:


Full Text










PLASMA SPECTROCHEMISTRY WITH A
FOURIER TRANSFORM SPECTROMETER







BY






THOMAS JOSEPH JOHN MANNING


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY


UNIVERSITY OF FLORIDA


1990


UNLYiRSliY "' '"";L Uil;ii3






















To my family; mother, father, brother, sister,
grandparents; whose support, love, patience,
and caring made this possible.












ACKNOWLEDGEMENTS


I would like to express my extreme gratitude to the

following people whose help dedication, friendship, honesty,

and loyalty made this work possible.

My family played a tremendous role in my recent success.

My mother has been a source of inspiration with her hard work

(teaching, Clemson) and working through her lung problems.

She always asks me the question, "and what does the much

mean," I wish I could say much more to express my gratitude

and love. My father has dedicated his life to his family, and

through his encouragement and prodding, my life has progressed

at a rate which he can take much of the credit for. My

parents have blessed me with good health, a good education,

good morals, and the ability to dream and create with the best

of them.

My sister has long been a stabilizing factor in the

family and her husband, George, a welcome addition. Patrick

is young, cute, impressionable, and has a great arm. Jim, my

big brother, has turned out to be a good brother, someone I

can count on and talk to when needed.

My grandparents, particularly my grandfather who recently

passed away, are sorely missed. I can picture them talking to

iii












friends, on a stoop or in a hall way, in Brooklyn. Bragging

about their grandson, the doctor. I have come to realize that

obtaining a goal of this nature is done as much for others

that you care about, so that they are proud, as it is a

selfish goal to satisfy your curiosity about what you are and

what you can be.

Susan and Jeanne have become friends over the years. I

realize my jokes were not always tactful, but they were meant

to be funny. They were an outlet when the possibility existed

for group politics to get out of hand.

Dr. J. D. Winefordner helped fund my salary for part of

my stay at UF and gave me a the chance to discuss my ideas and

experiments with him and other members of the group.

Doug Hof and Byron Palmer of the Los Alamos Fourier

Transform Facility have become more than just collaborators

since I first started working at LANL in 1987. I would have

left UF out of frustration if it were not for the outlet they

and the LAFTS provided. They let me carry out my ideas and

publish my work so I could establish creditability that group

politics never gave me the chance to establish. Doug and

Byron showed great patience and friendship in allowing me to

come back again and again to the LAFTS and for this I am

truly grateful.






Lots of friends have been made in the lab who have helped

me; Steve L., Tye B., Ali A., Kin N., Paul J., Andres C., Lynn

P., Alicia O., Martin A., Jorge V., Barbara K., Leigh Ann F.,

Ed V., Ben W., Eric G., Tony M., Ramme, Nancy S., Edison B.,

Dennis H., Scott S., Alain B., Wellington M., and En Yu Shoa

are the ones I can name right now. Jill Angus of UNM and LANL

has helped me a great deal. I hope our good relationships

keep up for years to come.

Arlene has recently come to play a major role in my life.

Her love, caring, beauty, and outdoor lifestyle are a welcome

addition to my life. I hope our relationship will be a long,

full one that others will be envious of.

I thank God for Los Alamos, Byron, Doug, and my parents

for sticking by me and helping me work through some tough

situations. Without others, my accomplishments would be

hollow if they could have been achieved at all.













TABLE OF CONTENTS


page

ACKNOWLEDGEMENTS ..........................................iii

ABSTRACT...................................................... viii

CHAPTER

1 INTRODUCTION................................... 1

2 HIGH RESOLUTION FOURIER TRANSFORM SPECTROMETER
TO IDENTIFY THE ROTATIONAL STRUCTURE OF
2 g 2 Zu N2(0,0) IN A HELIUM INDUCTIVELY
COUPLED PLASMA
Introduction.................................. 6
Experimental.................................10
Results and Discussion.......................12
Conclusions............................... ...14

3 A LOW FLOW ATMOSPHERIC PRESSURE
NEON INDUCTIVELY COUPLED PLASMA
Introduction.................................. 31
Experimental....................... ... ........31
Discussion................................... 33
Conclusions.................................... 34

4 A NONFLOWING VARIABLE GAS INDUCTIVELY COUPLED
PLASMA AS A LIGHT SOURCE FOR HIGH
RESOLUTION SPECTROSCOPY
Introduction................ ............. 35
Experimental................. .................... 36
Results......................................37
Discussion....................................38
Conclusions.................................... 41

5 OBSERVATIONS OF LINE SHIFTS AND LINE PROFILES
IN AN INDUCTIVELY COUPLED ARGON PLASMA
Introduction...................................47
Experimental.................................... 50
Results/Discussion..........................52
Conclusions............................... .... 55






6 A VARIABLE BANDPASS FILTER FOR ULTRAVIOLET
AND VISIBLE FOURIER TRANSFORM SPECTROSCOPY
Introduction...... ..........................84
Experimental................................ 86
Results................ ....................... 88
Conclusions.... .............................88


7 AN INTENSITY, WAVENUMBER, AND RESOLUTION
STANDARD FOR VISIBLE FOURIER
TRANSFORM SPECTROSCOPY
Introduction.................................96
Experimental.................................97
Discussion...................................97
Conclusions...................................100

8 A SYSTEM TO STUDY NOISE IN FOURIER
TRANSFORM SPECTROSCOPY
Introduction.................. .................104
Experimental................................ 106
Results......................................107
Conclusions.................................109


9 RELATIONSHIP BETWEEN RESOLUTION AND
SIGNAL-TO-NOISE RATIO'S IN FOURIER
TRANSFORM SPECTROSCOPY ........................119

10 CONCLUSIONS........................ ...........109

REFERENCES.................................................... 128

BIOGRAPHICAL SKETCH.. ..................................135


vii















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy


PLASMA SPECTROCHEMISTRY WITH A
FOURIER TRANSFORM SPECTROMETER


By

Thomas Joseph John Manning

August 1990


Chairman: James D. Winefordner
Major Department: Chemistry


This dissertation can be interpreted as being two

dimensional. The first dimension uses the Los Alamos Fourier

Transform Spectrometer to uncover various physical aspects of

a Inductively Coupled Plasma. The limits of wavenumber

accuracy and resolution are pushed to measure line shifts and

line profiles in an Inductively Coupled Argon Plasma. This is

new physical information that the plasma spectroscopy

community has been seeking for several years. Other plasma

spectroscopy carried out includes line profile studies, plasma

diagnostics, and exact identification of diatomic molecular

spectra.


viii












The second aspect of the dissertation involves studies of

light sources for Fourier Transform Spectroscopy. Sources

developed use an inductively coupled plasma (ICP) power

supply. New sources (neon ICP, closed cell ICP, and helium

ICP) were developed and new methods to enhance the performance

and understand a Fourier Transform Spectrometer were studied

including a novel optical filter, a spectrum analyzer to study

noises, and a standard to calibrate and evaluate a Fourier

Transform Spectrometer.













CHAPTER 1
INTRODUCTION


This dissertation will introduce new applications and

studies that couple a Fourier Transform Spectrometer operating

in the ultraviolet and visible region of the spectrum to

various plasma and discharge sources.

Fourier Transform Spectrometers are built for three

reasons; intensity precision and accuracy, high resolving

power, and wavenumber accuracy and precision. The throughput

advantage and multiplex advantage also make the Fourier

Transform Spectroscopy (FTS) attractive but are not the main

reasons the worlds most powerful ultraviolet and visible

Fourier Transform Spectrometer's have been built.

The first advantage is intensity precision and accuracy

in the line profile. This aspect is utilized less than the

other two in this research but is used in temperature

measurements (e.g. Boltzmann methods involving diatomics

(rotational temperature) and atoms (excitation temperature)),

line profile studies, isotope ratio determinations, etc. The

second advantage is resolution or high resolving power. This

advantage was used to look at rotational levels of molecules,

physically resolve line profiles, and identify isotope shifts

and hyperfine structure of atomic lines. The final advantage






2

is the excellent wavenumber accuracy and precision. This

facet of FT-UV/VIS spectroscopy was utilized to measure line

shifts in an Inductively Coupled Plasma (ICP), and better the

existing wavenumber accuracy of a well studied N2+ transition

in ultraviolet. A summation of each chapter follows.

The second chapter is a continuation of my master's

thesis. The Helium Inductively Coupled Plasma described in

this paper was first developed for nonmetal atomic emission

spectroscopy at Los Alamos National Lab (LANL). The source

proved to be very stable (low flicker noise component)

compared to other flowing systems, had very little background

and is sufficiently energetic to be used as a light source.

This project was the first published work using the Los Alamos

Fourier Transform Spectrometer. It clearly demonstrated that

the instrument, a 13 million dollar national facility, was

functioning at a high level. It also demonstrated the He-ICP

is a good high resolution light source. The N/,(0,0)

transition was resolved with higher wavenumber accuracy (>1

part in 107) than had been previously been reported in the

literature.

This same torch was used to generate a low flow,

atmospheric pressure Neon Inductively Coupled Plasma. Chapter

three is a brief outline of the first Neon Inductively Coupled

Plasma described in the literature. This was originally

generated for inductively coupled plasma-mass spectrometry

studies, but the price of neon proved too expensive to








continue this research.

Chapter four deals with a nonflowing, variable gas

inductively coupled plasma as a light source for high

resolution spectroscopy. This work was recently issued a

patent by the United States Patent office. It generated

plasmas with 10 different gases (He, Ne, Ar, Kr, Xe, H,, N2,

CO2, C2H2, 02) over a wide range of pressure (0.01-100 torr).

Work was also initiated to demonstrate the static, low

pressure source could be interfaced to a mass spectrometer

with several advantages over the traditional ICP-MS. For

Fourier Transform Spectroscopy (FTS) research, the static

environment is important to minimize flicker noise in the

source. Varying the gas allows us to vary the Doppler

temperature and electron density as well as the amount of

background light flux.

Chapter five is a physical analytical study. This

chapter represents the largest project undertaken; the

measurement of line shifts and line profiles of light emitted

from a argon inductively coupled plasma. There are small

shifts in wavenumber position between low pressure sources (no

collisional broadening, no shift) and atmospheric pressure

sources. Line profiles were deconvoluted to measure

broadening contributions from Doppler and collisional

processes in the atmospheric pressure argon ICP.

Approximately 180 lines with a signal-to-noise ratio (SNR)

greater than 100 were measured with the LAFTS. This pushed






4

the wavenumber position and resolution capabilities of the FTS

technique. Argon, iron, calcium, barium, and strontium

emission line profiles from an inductively coupled argon

plasma were recorded at sub-Doppler resolution.

The project involving line shifts concluded with the

development of a optical filter that allowed variance of both

bandpass and wavenumber position. It was designed for

ultraviolet and visible Fourier Transform Spectroscopy in

conjunction with the ICP. Its purpose was to reduce the

amount of unwanted light entering the FTS This filter is

discussed in chapter six.

The Re I (575 nm) line has both hyperfine structure (HFS)

and isotope shifts (IS). This structure was proposed as a

standard to determine the quality of spectra acquired from a

FTS. The reason for doing this is simple. Instrument

companies producing instruments have been selling bad

instruments, that is Fourier Transform Spectrometers that do

not come close to meeting their published specs (often by

orders of magnitude). The people purchasing the instruments

typically do not know much about FTS and are not aware of the

problems. By proposing a standard for wavenumber position,

intensity precision, and resolution, it will allow people to

judge the quality of the instrument they are about to buy or

to judge the performance of the instrument at any given time

during its lifetime. It can also serve as a calibration

standard when fine tuning the instrument. Chapter seven









describes this standard.

Chapter eight covers noise studies in Fourier Transform

Spectroscopy using a relatively simple arrangement. The

output of the light source is chopped by a mechanical

chopper. Chopping simulates the constructive and destructive

interference in an interferometer. When the beam is blocked

by the chopper, this simulates destructive interference, and

when light is passed, this simulates constructive

interference. The beam is chopped at a rate (Hz) similar to

that produced by constructive and destructive interference in

a Michelson interferometer. This work indicates that a shot

noise limited case has not been achieved by analytical

spectroscopists, despite their claims.

Analytical spectroscopists typically use a ratio of

signal and noise intensities to calculate the signal-to-noise

ratio in FTS. This method fails at poor resolutions when the

line profile is broadened by instrumental effects decreasing

the amplitude (peak height) of the line profile. Provided our

integration times are equal, there should be identical signal-

to-noise ratios for high and low resolutions alike. This

research has found this not to be the case. At poor

resolutions, the loss of signal amplitude, caused by

instrumental broadening, negatively affects the SNR. Many

analytical studies have been carried out at poor resolutions

and LODs based on amplitude SNRs giving low values. Many of

these studies claim to have achieved shot noise limited cases.







6
Work in chapter seven gives conclusive evidence that

analytical chemists have not achieved a shot noise limited

spectra.












CHAPTER 2
USE OF A HIGH RESOLUTION FOURIER TRANSFORM SPECTROMETER
TO IDENTIFY THE ROTATIONAL STRUCTURE OF THE B E/ X2Z
TRANSITION OF N/(0,O) IN A HELIUM
INDUCTIVELY COUPLED PLASMA



Introduction

The use of Fourier Transform Spectrometers in the near,

middle and far-infrared spectral region is common in many

analytical labs. Since a wealth of atomic information lies in

the visible and ultraviolet regions, there has been a great

deal of effort by various groups worldwide to build Fourier

Transform Spectrometers that operate over this region of the

spectrum. In 1975 Horlick and Yuen (1) reviewed

spectrochemical analysis with a Fourier Transform Spectrometer

(FTS). More recent reviews by Faires (2) and Thorn (3)

discuss theory, instrumentation, and applications of a FTS in

Atomic Emission Spectroscopy (AES). Stubley and Horlick

(4,5,6), Marra and Horlick (7), Faires (8,9), Faires, Palmer,

Engleman, and Niemczyk (10), and Faires, Palmer, and Brault

(11) carried out pioneering work in FT-ICP-AES. Various

aspects of the LAFTS have been previously reported (12,13).






Los Alamos Fourier Transform Spectrometer

The Los Alamos Fourier Transform Spectrometer (LAFTS),

housed in a 3360 ft2 building, is the highest resolution

visible and ultraviolet FTS in the world. Table 1 lists some

of its characteristics. The instrument is currently operating

with single pass optics allowing for a maximum resolution of

0.0026 cm' (0.023 pm at 300 nm). The interferometer is housed

within a 14 foot by 7 foot vacuum tank (30 mTorr achieved) and

utilizes two entrance ports for experiments. The FTS sits on

a 4 foot by 10 foot NRC vacuum compatible optical table

isolating the instrument from vibration.

Figure 1 shows the folded design of the instrument. The

LAFTS utilizes a double sided interferogram which improves the

signal-to-noise ratio and helps eliminate phase problems. The

A/D converter is composed of seven 16-bit A/D converters,

scale amplifiers, and an 18-bit D/A converter for calibration.

Parsons and Palmer (13) give detailed characteristics and

research plans for the LAFTS.

For inductively coupled plasma-atomic emission studies

(ICP-AES) the advantages of the FTS are intensity precision,

wavenumber accuracy, and high resolution. Provided electronic

noise is minimized, the limiting noise in the FT-ICP system

will be the plasma. Several papers (14-18) have discussed

sources of noise in Argon ICPs.








Helium Inductively Coupled Plasma

Argon is the most common plasma gas used in Inductively

Coupled Plasmas for elemental analysis. Abdallah and Mermet

(19) compared temperatures of helium and argon in ICP and MIP

systems. Abdallah et al. followed with a paper (20) utilizing

a 50 MHz Helium-ICP at atmospheric pressure for analytical

studies. Seliskar's laboratory (21-24) conducted work with a

reduced pressure torch that is adaptable to a commercial ICP

unit. Chan et al. (25-27) developed and tested a tangential

flow, low flow, atmospheric pressure torch that is constructed

entirely of macor and adapts readily to commercial units.

Montaser and Van Hoven (28) give a critical review on mixed

gas, molecular gas and helium ICPs. Due to the high

ionization potential of helium (24.6 eV) versus argon (15.8

eV), it has achieved improved limits of detection for elements

with relatively high ionization potentials (Cl 12.9 eV, Br

11.8 eV) (25). He-ICPs have fewer spectral lines than Ar-

ICPs in the desired regions of AES decreasing the chance for

overlap (26). Tan et al. (29) recently introduced a laminar

flow He-ICP torch and demonstrated its use in the analysis of

some nonmetals.



N,+(0.0)

In 1933 Coster and Brons (30) published detailed work on

the emission spectrum of N2/. Diatomic nitrogen (NN,N2*) has

been utilized as a rotational temperature indicator in RF






10

plasmas (31-39) and various discharges (40-46). Most papers

use the unresolved vibrational bandhead or the partially

resolved rotational structure. The use of Raman scattering

from N2 as a temperature indicator has also been demonstrated

by several groups (47-50). Gottscho et al. (51) give an

account of perturbation effects within the B2Z u XZ,

transition of N2/. Herzberg (52) gives an excellent discussion

of both the theoretical and experimental applications of the

transition. Montaser and Van Hoven (28) give references for

additional uses of N2 and N2+ in plasma diagnostics. Nitrogen

is used in plasmas for metal surface treatment (53), synthesis

of nitrogen oxydes (54), and CO, CO2, and gas lasers (55).

The use of N2 in sub-doppler resolution studies is discussed

in this paper.

The first negative band of N2 has several advantages for

rotational temperature determinations in plasmas. The (0,0)

bandhead (391 nm) is in an accessible region of the spectrum.

The Q branch (AJ=0) is forbidden decreasing the number of

rotational lines present relative to N2(0,0), N2(0,1), and

OH(0,0) (56), which are used as rotational temperature

indicators. At moderate resolution (<0.25 cm') the

rotational structure can be adequately resolved.

This chapter presents a low flow, laminar flow,

atmospheric pressure helium ICP torch. Nitrogen is seeded

into the helium plasma and the emission from N92 is analyzed

using the Los Alamos Fourier Transform Spectrometer.








Experimental

Helium inductively coupled plasma torch

The torch is optimized for low flow (2-6 L/min),

simplicity in design, and adaptability to a commercial unit.

Chan et al. (25) present a low flow (7 L/min), tangential

flow, macor torch. Two problems arise with this design.

First, macor (Corning, Horseheads, NY), a heat resilient,

machinable ceramic, is extremely brittle and time consuming to

work with. Second, the spiral design allows the plasma to

wrap around the torch at low flow rates increasing the chance

for damage to the macor torch. An attempt to run a mixed

argon-helium plasma also proved detrimental to the macor.

These two problems were overcome by a) reducing the amount of

macor and constructing the torch largely from teflon and b)

making the torch laminar flow and keeping the plasma above the

macor tip.

Figure 2 shows the diagram of the torch used in this

study. The plasma gas enters the torch through two side ports

spaced 180 apart. Two ports are utilized to give an even

distribution of gas through the eight symmetrically spaced

channels that carry the plasma gas from the teflon channels

through the bores in the macor tip and into the plasma. The

sample is carried up through the center of the teflon and

macor pieces and enters the plasma region through a 2/1000

inch aperture. The torch handles 10,000 ppm Na solution with

no clogging problems. This system uses a fritted disk






12

nebulizer (11). It has higher efficiency (25% with 0.3 L/min

gas flow), lower cost, and a decreased noise power spectrum

(14) versus the ultrasonic nebulizer. Helium does not work

with a concentric nebulizer due to its relative lack of

viscosity versus argon. An extended quartz tube was used to

keep impurities (N2, H20 etc.) from the atmosphere out of the

plasma. It has one or two wraps of teflon tape around its

base for an airtight fit with the base. A four turn, gold

plated (57) coil was used to generate the 27.12 MHz RF field.

The central teflon piece and the teflon base (2" diameter) fit

tightly together. The torch is supported on a teflon stand in

the ICP box. The tesla coil is attached, via an ignition

wire, to one of the side ports with a copper clip. Quartz

tubing (1/4" diameter) is inserted into the two side ports.

A 12/5 balljoint is inserted into the sample uptake port. The

macor tip has a 0.5 inch diameter and a 0.5 inch height. It

fits tightly with the teflon channel (see Figure 2). Under

normal operating conditions this fit is sufficient to seal the

tip from outside contaminants and keep the sample and plasma

gas separate. As an added precaution, the outside of the

macor-teflon connection is wrapped with teflon tape. The

macor to teflon connection can be fused with chemical etching

techniques or a high temperature quartz based cement can be

used as a sealant. Impedance matching utilized has been

previously described in the literature (58,59).








Inductively coupled plasma

This study employs a 27.12 MHz RF Plasma Products (Cherry

Hill, NJ) HFP 5000 D model Inductively Coupled Plasma unit.

The ICP operating conditions are outlined in table 2. The

nitrogen is 99.9995% certified pure and was obtained from

Union Carbide-Linde Division (Albuquerque, NM).

Los Alamos Fourier transform spectrometer

The parameters of the LAFTS used for this study are as

follows: The spectral output is from 14000 cm1 to 40000 cm".

The detectors utilized are RCA 1P28s. The beamsplitters are

aluminum coated quartz and no optical filters are used. The

resolution chosen (0.09 cm') is sub-Doppler for nitrogen (19).

The number of points taken is 889,219, which is zero filled to

2 (1,048,576) points for transformation. The data

acquisition time for the 20 scans is 1 hour 17 minutes and the

computer transform time is 42 minutes. The spectrum is

unapodized (boxcar). Atmospheric pressure in the LAFTS (7300

ft above sea level) is 593 Torr, and the tank temperature is

21C.



Results and Discussion

Table 3 gives the line positions (0.01 cm') of the P and

R branches observed in the Helium Inductively Coupled Plasma.

The accuracy of the (0,0)band data given by Coster and Brons

(30) is updated by 0.1 to 0.2 cm .

Figure 3 shows part of the (0,0) spectra. The P (1-38)






14

and R(1-10) branches are identified. The pattern is

consistent for larger branch numbers not shown. The doublet

splitting (J-1/2,J+1/2) is evident for both branches (P >16,

R >10). The alternation of line intensity (even > odd) is

unmistakable for both branches. Doublet splitting and even-

odd alternation of line intensity are treated theoretically by

Herzberg (52).

Figure 4 gives the Fortrat diagram for the P and R

branches in the He-ICP. The P branch folds over itself at

25540.10 cm1 giving the vibrational bandhead. The K value

used for all branches that are doublet split is J-1/2. At

higher K values (R >40, P>44) only the even branches are

detectable from this source (He-ICP).

The method to calculate the rotational temperature from

a well resolved structure is taken from Abdallah and Mermet

(19). This method allows for relative line intensity

measurements (as opposed to absolute) for either the P or R

branches. For the P branch,


Log I = f(K"(K"-l)) (1)
K"

and the R branch,

Log gl = f((K"+l)(K"+2)) (2)
K"+1

where a = 1 (even number lines), 2 (odd number
lines)
f = -Bhc/kT = (-2.983/T)
I = experimental line intensity
K" = quantum number of lower state
T = temperature (kelvin)






15

Plotting the left side of equation 1 or 2 (for varying

values of I) versus the right side (increasing values of K")

of the respective equation yields a straight line with a slope

proportional to 1/T. Rotational temperatures for the He-ICP

using the parameters and data listed above are R(odd, 1700 K),

R(even, 1680 K), P(odd, 1830 K), and P(even, 1660 K).

Various papers (35,38,39) outline similar Boltzmann

methods to calculate rotational temperatures in plasmas. This

method allows for four rotational temperatures (P (even,odd),

R(even,odd)). Thorne (60) claims that the Boltzmann method for

N2/(0,O) is suitable for measurements under 5000 K. Cabannes

(39) claims it is allowable for temperatures under 10000 K.

Kornblum and de Galen (61) have interpreted the resulting

measurement as the gas kinetic temperature.



Conclusion

A new He-ICP torch has been demonstrated. The torch has

also been used to generate a neon ICP (62) and is being

investigated as an ionization source for helium and neon

ICP-MS studies. The He-ICP is not being pursued for atomic

emission spectroscopy.

The use of N2 as a diagnostic tool is common. This paper

allows the spectroscopist to easily identify and use the

structure for rotational temperatures. The line position

accuracy of the rotational levels is improved against those

currently in the literature.



























Figure 1. Optical Elements: a) cat's-eye reflectors b) beam
splitter and carousel c) 8-inch turning mirrors d)
collimator optics e) output mirrors Light enters from
either the end or side port and is collimated by d. There it
goes to the upper beam splitter, b, and is divided. These two
beams go to the turning mirrors c and to the primary mirror of
the cat's-eye reflector, a. In the cat's-eye reflector the
beam is focused down onto the secondary mirror and reflected
back to the primary where it is made parallel again. The
effect of the cat's-eye reflector is to reflect the beam back
along a path that is parallel but displaced. From the cat's-
eye reflectors the beams return to the 8-inch mirrors and then
to the lower beam splitter. There the beam is recombined and
interferes with itself. The light then travels to the output
mirrors, e, that focuses the beam and to the flat mirrors that
direct the beam to the detectors.







17





















d
d
end input __--j )) c


e c h~U

~I~KCS .0






















Figure 2. Diagram of low-flow, atmospheric pressure He-ICP
torch. The top inset shows the macor tip and teflon central
channel interface in greater detail. The bottom inset shows
the eight plasma gas holes surrounding the sample gas inlet on
the macor tip.

































DOWNWARD.EXPA~N0DE
VIEW OF MACOR TIP

0
O0
PLASM o O 0
GAS
0 0


SAMPLE
INLET






















Figure 3. The four panels show a portion of the rotational
structure of the N2*(0,O) band. The top x-axis is angstroms
and the bottom is wavenumbers (cm').





























8e+008



6~e008



4e+008



2e+008



0


WAVENUMBER























5e1008


4e+008



2e+008


2e+008



le+008



0


WAVENUMBER


Figure 3--continued



























3e+008



2eM
(O 2e-008
z
I-.
z

le+008


WAVENUMBER


Figure 3--continued


























4e+008



3e+008


2e+008



le+008



0


WAVENUMBER


Figure 3--continued


























Figure 4. Fortrat diagram of P and R branches. At 25540.10
cm the P branch (P12, P13, P14) folds giving the
bandhead.



































*
*
G
G
E
E
Q




.
Q

Q
E

E
Q






* 0
* E


* 0
* 0
* 0
* 0
* 0

40
Qm
B0


25700


13


n BRANCH
* P BRANCH


25900
25900


26100
26100


WAVENUMBER


n


. n .. ..


25500








Table 1. Los Alamos FOurier Transform Spectrometer
Specifications


Spectral Ranges:


Region Range

UV/VIS 200 nm-l.l um


NEAR IR
FAR IR


0.8-5.5 um
4-20 um


Beamsplitters

quartz/aluminum

CaF-GaP
KC1-Ge


Detectors

photomultiplier,
si PIN diode
InSb
HgCdTe


Resolution and Operating Parameters:


Optical Path
Difference
Resolution

Sample Rate



Number of points

Maximum Transform
Size


2.5 meters (single pass)
5.0 meters (double pass)
0.0026 cm (single pass)
0.0013 cm (double pass)
5000 Hz not digitally filtered
40000 Hz with digital filtering
reduced to 5000 Hz throughput

>4,000,000 points

22 (4,194,304)









Table 2. Helium inductively coupled plasma operating
conditions


Source Unit
Frequency
Forward Power
Reflected Power
Plasma Gas Flow
Nebulizer Flow
Sample Uptake Rate
Viewing Height above
load coil
Aperture to FTS


RF Plasma-Therm 5 kW unit
27.12 MHz
1.5 kW
<5 W
4 L min'
0.5 L minm
0.3 mL min"
14 mm

7 mm









Helium inductively coupled plasma.


P BRANCH


K Value
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
16
17
17
18
18
19
19
20
20
21
21
22
22
23
23
24
24
25
25
26
26
27
27
28
28
29
29
30


J(+1/2)
















J-1/2
J+1/2
J-1/2
J+1/2
J-1/2
J+1/2
J-1/2
J+1/2
J-1/2
J+1/2
J-1/2
J+1/2
J-1/2
J+1/2
J-1/2
J+1/2
J-1/2
J+1/2
J-1/2
J+1/2
J-1/2
J+1/2
J-1/2
J+1/2
J-1/2
J+1/2
J-1/2
J+1/2
J-1/2


WAVENUMBER(cm')
25562.36
25558.76
25555.54
25552.62
25549.96
25547.66
25545.67
25543.81
25542.47
25541.37
25540.59
25540.11
25540.11
25540.11
25540.59
25541.15
25541.37
25542.16
25542.47
25543.54
25543.85
25545.14
25545.44
25547.12
25547.35
25549.40
25549.65
25552.20
25552.59
25554.82
25555.09
25557.99
25558.28
25561.46
25561.76
25565.24
25565.56
25569.32
25569.66
25573.71
25574.06
25578.42
25578.76
25583.41









Table 3--continued.


J+1/2
J-1/2
J+1/2
J-1/2
J+1/2
J-1/2
J+1/2
J-1/2
J+1/2
J-1/2
J+1/2
J-1/2
J+1/2
J-1/2
J+1/2
J-1/2
J+1/2
J-1/2
J+1/2
J-1/2


R BRANCH


J(1/2) Wavenumber (cm')
25570.30
25574.77
25579.51
25584.57
25589.93
25595.61
25601.58
25607.85
25614.44
25621.33
J-1/2 25628.41
J+1/2 25628.56
J-1/2 25635.89
J+1/2 25636.12
J-1/2 25643.67
J+1/2 25643.90
J-1/2 25651.76
J+1/2 25651.99
J-1/2 25660.16
J+1/2 25660.39


25583.77
25588.70
25589.06
25594.31
25594.67
25600.21
25600.56
25606.41
25606.76
25612.88
25613.22
25619.59
25619.99
25626.96

25634.12
25634.39
25640.85
25641.80
25648.44
25656.19
25668.43
25676.62
25685.82


K Value
0
1
2
3
4
5
6
7
8
9
10
10
11
11
12
12
13
13
14
14









Table 3--continued.

15 J-1/2 25668.86
15 J+1/2 25669.09
16 J-1/2 25677.79
16 J+1/2 25678.09
17 J-1/2 25687.18
17 J+1/2 25687.40
18 J-1/2 25696.72
18 J+1/2 25697.02
19 J-1/2 25706.62
19 J+1/2 25706.88
20 J-1/2 25716.78
20 J+1/2 25717.08
21 J-1/2 25727.27
21 J+1/2 25727.58
22 J-1/2 25738.05
22 J+1/2 25738.38
23 J-1/2 25749.15
23 J+1/2 25749.53
24 J-1/2 25760.53
24 J+1/2 25760.86
25 J-1/2 25772.21
25 J+1/2 25772.55
26 J-1/2 25784.17
26 J+1/2 25784.54
27 J-1/2 25796.46
27 J+1/2 25796.82
28 J-1/2 25809.02
28 J+1/2 25809.39
29 J-1/2 25821.89
29 J+1/2 25822.26
30 J-1/2 25835.04
30 J+1/2 25835.41
31 J-1/2 25848.48
31 J+1/2 25848.86
32 J-1/2 25862.21
32 J+1/2 25862.58
33 J-1/2 25876.22
33 J+1/2 25876.59
34 J-1/2 25890.44
34 J+1/2 25890.85
35 25905.36
36 J-1/2 25919.96
36 J+1/2 25920.27
37 J-1/2 25934.16
37 J+1/2 25935.11












CHAPTER 3
GENERATION OF A LOW FLOW ATMOSPHERIC PRESSURE NEON ICP


Introduction

A 27.12 MHz Neon Inductively Coupled Plasma (Ne-ICP) has

been generated in this laboratory. Argon has long been used

as the plasma gas for an ICP. Abdallah and Mermet (19) and

Chan et al.(25) have generated atmospheric pressure helium

plasmas in inductively coupled plasma (ICP) units. Walters et

al. (63) generated a reduced pressure (0.5-14 torr) 144 MHz

neon ICP for electrodeless discharge lamp studies. This is

the first known report of an atmospheric Ne-ICP. The torch

design has been used for helium ICP studies and is described

by Manning et al.(64). The neon ICP is being investigated as

a possible atomization source for adaptation to a commercial

inductively coupled plasma-mass spectrometry instrument.



Experimental

The Ne-ICP is generated with a 27.12 MHz RF Plasma

Products HFP 5 kW unit (Cherry Hill, NJ). The forward power

for the ICP is 500 W with less than 5 W reflected (58). The

low flow (3 1/min), laminar flow, atmospheric pressure torch

is described by Manning et al.(64). It utilizes an extended

quartz sleeve to minimize impurities (N2, HO, etc.) from the






33

atmosphere entering the plasma. The nebulization system

employed is the fritted disk nebulizer (11). The sample is

pumped to the fritted disk with a standard peristaltic pump.

The flow rate of the plasma gas is 3 L/min with a minimum of

2 L/min and a maximum of 7 L/min possible. The nebulizer gas

flow rate is 0.3 L/min with a sample uptake rate of 0.3

mL/min. The nitrogen is seeded in at a rate of 5 mL/min. The

neon gas is supplied by Union Carbide-Linde division gas

distributors (Albuquerque, NM).

Los Alamos Fourier Transform Spectrometer

The Los Alamos Fourier Transform Spectrometer (13), a

national facility for high resolution atomic and molecular

spectroscopy, is used in this experiment for medium resolution

(0.07 cm') spectra in the near IR, visible, and ultraviolet

(8500 cm' to 33372 cm) Aluminum coated quartz beamsplitters

are employed and the detectors utilized are silicon PIN

diodes. The number of points taken for each run is 953,863

and the transform size is 220 (1046000) points. The run in

which only water is introduced into the plasma consists of 10

scans (39 minutes data acquisition time) and took 16 scans in

the run introducing both water and N2 (64 minutes data

acquisition time). The number of scans is limited by the

current availability of research grade neon. The cost of neon

is the limiting factor.








Results and Discussion

Neons' ionization potential (21.56 eV) is higher than

that of argon (15.76 eV) but lower than helium (24.587 eV).

This should offer better detection limits for various

nonmetals (eg. F, Cl, Br, etc.). The neon plasma as a source

does not have the luminosity of the argon ICP and is similar

to the helium ICP in this respect. The plasma is stable with

powers as high as 1.5 kW. Introducing sample reduces its

stability. With an aerosol entering the plasma, it could not

operate at powers greater than 600 W. Above this threshold

the plasma either extinguished or formed numerous filaments

and produced a non-homogeneous plasma. The plasma operated

with powers as low as 150 W. Luminosity appears linear with

power.

An attempt to measure the rotational temperature using

N2+(0,0) (391.44 nm) and OH(0,0) (306.4 nm) is made. The N2

bandhead has a SNR of 4.2 and the R branches used in

diagnostics are not discernible making reliable measurements

impossible. The OH(0,0) band is not present at all in the

spectrum. The Hg line is used to measure the electron number

density (65) at 8x10'3 cm The fact that both species

(OH,N2+), which are strong UV emitters in plasmas and

discharges, do not have a readily detectable emission rate,

coupled with the low electron density measurement (8x10'3 cm3)

and the low input power (500 W), give reason to believe the

Ne-ICP in this setup is cool compared to the argon and helium







35

plasmas (19) run at higher powers. Bands from the first

positive system (B3Z, A3.) of N2 are obvious in the near IR

and visible. This transition (10,000 cm'-17,000 cm') requires

less energy for excitation than OH(0,0)

( >31,000 cm').



Conclusion

An atmospheric pressure neon ICP is generated. In the

current setup it operates reliably at 500 W forward power. The

Ne-ICP is being pursued as an atomization source for a

commercial ICP-MS setup. A greater amount of diagnostics is

not feasible due to the expense of running a Ne-ICP. A single

Ne line (585 nm) was used to determine the Doppler

temperature. The value (1058 K) was obtained by deconvoluting

the line profile with a Voigt profile.

Using a Ne-ICP as an ion source for ICP-MS studies should

offer the advantage of "freeing" various mass values that are

not accessible with argon as the plasma gas.












CHAPTER 4
A NON-FLOWING, VARIABLE GAS INDUCTIVELY COUPLED PLASMA AS
A LIGHT SOURCE FOR HIGH RESOLUTION SPECTROSCOPY



Introduction

Ten self-contained 27.12 MHz Inductively Coupled Plasmas

(ICP) are generated in a static torch. The torch offers

spectroscopists a closed system, low pressure (0.01 to 100

Torr), variable power (currently up to 1 kW) plasma. Doppler

temperatures (Na D, 589.0 nm) measured are approximately 5

times that of electrodeless discharge lamps (EDLs) or Hollow

Cathode Lamps (HCLs) operating under normal conditions. It has

improved noise considerations verses the flowing ICP

traditionally used.

Plasmas are generated using helium, neon, argon, xenon,

krypton, hydrogen, nitrogen, oxygen, carbon dioxide, and

ethylene. In some cases, particularly xenon, the closed

system minimizes the consumption of expensive gases. In the

case of hydrogen, it minimizes the potential hazard of working

with the gas in a flowing system. Ideas for absorption,

ICP-MS, and lasing action in a static plasma are discussed.

This torch generates a low noise, high power plasma that can

be used for a variety of spectroscopic studies.








Experimental

Figure 5 shows the torch design utilized in this study.

The quartz tube has a 17 mm OD, 15 mm ID, and a 14 cm length.

A quartz window is fused to the viewing end of the tube. In

some circumstances, using grounded clips helps the plasma

symmetrically fill the entire tube. End on viewing increases

the luminosity of the source. The four turn coil is gold

plated 1/8 inch copper tubing. The ICP instrument is a RF

Plasma Products HFP 5000 D (5 kW) unit. Often the plasmas are

initiated by introducing the field (RF ON) and slowly

increasing the power. In instances where this is not

possible (typically >10 Torr), a tesla coil discharge is sent

through one of the grounded clips to ignite the plasma. A

vacuum rack equipped with a liquid nitrogen cold finger, a

roughing pump, and a diffusion pump is utilized for evacuating

the cell and filling it with the desired gas. The cell is

evacuated to 105 Torr, back filled with the desired plasma

gas, evacuated to 10.5 Torr, and filled to the desired

pressure.

The cell is demountable and easily inserted into the ICP

coil. A 12/5 balljoint connects the rack and the torch. As a

precaution, nitrogen is blown across the quartz torch for

cooling purposes. Pressures given are the fill values at room

temperature. Powers recorded are the forward minus the

reflected values.

The Los Alamos Fourier Transform Spectrometer is used to






38

record the spectra at high (0.039 cm') resolution with 5

scans. The detectors are Silicon PIN diodes, and the

beamsplitters are aluminum coated quartz. The spectral window

transformed is from 8500 cm' to 40000 cm'. 2,052,056 points

are acquired and zero filled to 2,097,152 (221) points.

Additional parameters of the LAFTS are supplied by Manning et

al (64) and Parsons and Palmer (13). All gases used are

certified research grade (>99.995 %).



Results

Figure 6 shows the Na D (589.0 nm) line profile in a Xe-

ICP. Sodium (and lithium) impurities are present in most of

the spectra obtained. They are attributed to impurities in

the quartz. The sodium D line is used for the determination

of the Doppler temperatures in each plasma. Values ranged

from xenon (7 torr, 250 W) at 3700 K to both carbon dioxide

(2.8 torr, 250 W) and oxygen (6 torr, 550 W) at 2400 K. These

values are approximately 5 times that of HCLs or microwave-

EDLs, assuming a normal current or power input. To avoid

having the sample absorb on the cooler ends of the torch, the

length of the torch may have to be reduced so the tube length

fits entirely within the coil. Adding small quantities of a

desired species (e.g., typically in the 1-10 microgram range)

allows the torch to be used as a stable reservoir of atomic

lines. The same is possible for molecules stable in the

plasma environment.






39

The low pressure inert gas (He, Ne, Ar, Kr, Xe) plasmas

seeded with trace amounts of N2 for diagnostic measurements

exhibited varying degrees of continuum emission in the near IR

and visible. Xenon is the most prevalent with approximately

15 prominent continuum emission bands. The signal to

background measurements of the emission bands ranged from 3 to

15 and are typically 2-3 nanometers wide. The bands are found

between 625 and 880 nm. They are semicircular in appearance.



Discussion

The analysis of atomic and molecular species stable in a

static plasma utilizes some of the capabilities of the UV/Vis

Fourier transform spectrometer (particularly the LAFTS). In

particular the intensity precision, wavenumber position

precision and accuracy, and resolution capabilities are

partially realized.

There is no reason to believe this cell could not

generate other plasmas. Atomic species which are gaseous at

low pressure (mercury, radon, volatile salts), diatomic gases

(NO, CO, C12, Fl2, etc.), polyatomic gases (NO,, SO2, C2H,, other

hydrocarbons, etc.) should readily form low pressure plasmas

in this setup. The ethylene-ICP and other hydrocarbons

systems may be of interest to those studying the diamond

synthesis mechanism in a RF-plasma environment (66). Closed

cell hydrogen-ICP minimizes the dangers of working with a

flowing system. This hydrogen-ICP allows scientists to study






40

atomic interactions in a system similar to the surface of

cooler stars.

Fusing quartz windows to both ends of the plasma tube

allows a variety of absorption measurements to be conceived.

For example, intracavity absorption (67) with an ICP is

possible with the advantage of a longer path length, smaller

sample size, and an increased residence time per atom verses

the traditional flowing ICP torch.

A closed system ICP is currently being investigated as an

ion source for mass spectrometry. One end of the tube is

directly interfaced to the mass spectrometer with the flow

regulated by a gated valve. The sample is injected through the

syringe port (0.1 to 1 microliter) before ignition. The

solvent evaporation is enhanced with the vacuum leaving only

sample in the quartz tube. With plasma ignition, the sample

sputters into the plasma. We have found this to work quite

well for the nonmetals and softer metals. The variable gas

static plasma for ICP-MS should offer many advantages over the

flowing argon system. Expensive gases (e.g., neon, krypton,

xenon) can be routinely used due to low gas consumption.

These inert gases may be used to free mass ranges not

accessible with argon (e.g., sample overlaps with ArN ArH*,

ArOH+, etc.). Provided experimental conditions are identical

(power, pressure), the temperature and electron density of the

inert gas plasma (He, Ne, Ar, Kr, Xe) will have an inverse

relationship with the ionization potential. That is, the






41

higher the ionization potential is, the lower the temperature

and electron density will be. Thus by varying the inert gas,

the temperature and electron density range can be varied. The

closed cell allows for small sample sizes. It reduces some of

the hazards associated with radioactive samples in a flowing

system. The pressure differential between the plasma and the

mass spectrometer is reduced by at least two orders of

magnitude. Problems associated with the skimmer/orifice

degradation are reduced. The sample waste associated with a

flowing system is minimized. Nebulization is eliminated. In

some situations it may be possible to use nonpolar samples and

solvents. With evaporation/dehydration techniques, the

formation of hydrides, hydroxides, and oxides will hopefully

be reduced by applying this procedure. The low pressure cell

operated here can be converted to a fireball (68) at higher

powers. The fireball resembles the atmospheric pressure argon

ICP in appearance. This fireball could not be sustained for

extended periods of time or the quartz would glow red and

eventually melt. If a continuous vacuum is applied that

maintains a constant pressure, the fireball can be maintained

for extended periods of time.

Eckert (69) gives an excellent theoretical discussion of

elemental detection in a static induction plasma. His

calculations predict a limit of detection for magnesium (279.6

nm line) of less than 10"15 g in a nonflowing argon-ICP. With

small modifications to the torch, the closed cell ICP may have






42

analytical usefulness. This analytical research would be

similar to work previously carried out with microwave

electrodeless discharge lamps.



Conclusion

The static plasma outlined in this note describes an atom

and molecule reservoir for spectroscopic studies (70). It is

also being pursued as an ion source for mass spectrometry

(70). No graphical data is presented to support the claim

that noise factors in the static plasma are improved over the

flowing system. We hope the reader follows the reasoning that

with the lack of turbulence induced by the gas flow, and no

nebulization system, many of the sources of noise inherent in

a flowing system are either reduced or eliminated in a static

plasma. The intensity variation as a function of time is at

least an order of magnitude improvement over the flowing

system under normal operating conditions.

With various adaptations to the cavity, the cell could be

used for a variety of emission, absorbance, and fluorescence

measurements, both analytical and physical in nature.

























Figure 5. Closed cell ICP torch design.








44





















---- TO1
1FTSt


S-- STOPCOCK~



























Figure 6. Line profile of Sodium-D (589.0 nm) in Xe-ICP.

































2e+009



l. e+009

C-

Z
5-
z



0
o


WAVENUMBER












CHAPTER 5
OBSERVATION OF LINE SHIFTS AND LINE PROFILES
IN AN INDUCTIVELY COUPLED ARGON PLASMA


Introduction

Line shifts in analytical inductively coupled plasma's

(ICP) are a fundamental parameter that has not been studied in

any great detail. This paper will give an introduction to

line shifts measured and trends observed. For more detailed

treatments of pressure effects, including line shifts, the

reader is directed to Thorn (60) (overview), Alkemade et al.

(71) (in flames), Griem (72), and Breene (73) and references

within. Reviews by Konjevic and Weise (74) and by Konjevic

and Roberts (75) give numerous references on Stark broadening

and line shifts from a variety of sources.

Line shifts are important for a variety of reasons.

Analytical determination of elements in a complex sample is

often plagued misidentification of the species being studied.

In physical spectroscopy, ICPs are being used for the

measurement of transition probabilities. Knowledge of line

shift values are important to properly measure energy levels,

transitions, etc.. Whaling and Brault (76) used an ICP in

conjunction with the Kitt Peak Fourier Transform Spectrometer

to measure the transition probabilities of 1200 Mo (I) lines;






48

they introduce Mo as Mo(CO)6 eliminating any effects water has

on the spectrum. They observed line shifts of Mo as great as
-1
0.14 cm .

The Los Alamos Fourier Transform Spectrometer (LAFTS) is

an ideal instrument to measure small spectral shifts. Its

high resolution capabilities easily resolve physical line

widths emitted by hollow cathode lamps (HCL) (0.025 cm1) and

the inductively coupled plasma, (0.07 cm'). The wavenumber

precision of the LAFTS can be as great as 1 part in 108. The

LAFTS can acquire large spectral windows (9000 to 42000 cm')

at the required resolution to physically resolve the ICP

emission.

In studying line shifts, isotope shifts and hyperfine

structure must be considered to properly select a line.

Elements that clearly display either effect are avoided due to

the complexity introduced in attempting line profile studies.

The spectrum of a Pb emission line shows contributions from

three isotopes, 8Pb206 (24.1%), 82Pb207 (22.1%), and 82Pb208 (52.4%).

These isotope shifts are shown in Fig. 7. This spectrum was

taken from a low pressure HCL with a Doppler temperature of

approximately 500 K, resulting in Doppler broadened spectral

lines having negligible pressure (collisional) broadening.

Increasing the temperature to 6500 K and introducing

substantial amounts of collisional broadening, which simulates

ICP conditions, will cause the three isotopic lines to appear

as one asymmetric, broadened line. Measuring the line shift






49
becomes difficult because the individual lines have to be

deconvoluted before spectral shift measurements are possible.

There will be uncertainty introduced into the calculation by

the deconvolution which may affect the validity of the shift

measurement.

Figure 8 shows the hyperfine structure of tantalum. 73Ta 81

is the predominant isotope occurring naturally (99.998%) so

isotopic effects are negligible. With a nuclear spin of 7/2,

73Ta181 clearly exhibits hyperfine structure. Kuhn (77) gives

an excellent treatment of hyperfine structure and isotope

shifts. Ta emission in an ICP has significant Doppler and

pressure broadening of the lines giving asymmetric line

profiles with 8 convoluted hyperfine elements making shift

measurements virtually impossible.

Argon has one predominant isotope (lAr40,99.60%) which has

no nuclear spin making it an ideal choice for line shift

studies. Figure 9 shows a line profile typical of Ar I

emission at 420 nm from the ICP. Iron has one dominant

isotope (26Fe56, 91.72%), which has no hyperfine structure;

Fig. 10 shows a line profile of Fe I at 259.9 nm. The effect

of the lower abundance isotopes of Fe on the line center

position (shift measurement) and the full width at half

maximum (FWHM, a-parameters; where a = 0.86*(I/AD) are

considered negligible for this study.

Kato et al. (78) measured profiles of 16 spectral lines

stemming from 8 elements emitted by an ICP and measured with






50

a pressure-scanning Fabry-Perot interferometer. The results

their study are similar in absolute magnitude of some lines

but occasionally differ in sign. They did not consider the

effects of hyperfine structure and/or isotope shifts as

carefully as we did. Boumans and Vrakking (79) published work

on high-resolution spectroscopy using an echelle spectrometer,

which refers to line shifts in the ICP in connection with the

wavelength calibration of the spectrometer.



Experimental

The Los Alamos Fourier Transform Spectrometer (LAFTS) is

utilized in conjunction with a 27.12 MHz RF Plasma Therm

(Cherry Hill, NJ) ICP. Papers by Manning et al. (64) and

Parsons and Palmer (13) give details of the LAFTS. For all

studies, the viewing height above the coils is 12 mm. The gas

flow rates are 12 L/min (outer), 0.8 L/min (intermediate), and

0.8 L/min (nebulizer). For ICP studies, the LAFTS was operated

at 0.07 cm' resolution with a free spectral range of 33005.2

cm For the HCL studies, the LAFTS was operated at 0.02 cm

resolution with a free spectral range of 40015.2 cm1. The

detectors used were PMTs (1P28) or silicon pin diodes. The

silicon pin diodes were utilized for work involving Ar

emission in the near IR.

The output of the light source (ICP or HCL) is focused

with unity magnification on the FTS spectrometer aperture.

The f number of this spectrometer and the ICP/HCL optical







51
setups are matched. The output of a neon filled Al HCL is

impressed on the ICP emission using a neutral density filter

(0.2) before the focusing optics. Figure 11 shows an outline

of the entrance optics and the light sources. The Ne I line

at 585.2 nm from the HCL is used as a calibration standard for

all runs.

All emission lines used in this study (HCL and ICP) have

signal to noise ratios (SNR) greater than 100. This

requirement reduced the number of lines available between the

two sources. Some strong lines in the HCL were not always as

prevalent in the ICP and vice versa. Ionic lines for iron and

argon detected in the ICP were not observed in HCL emission or

had low signal-to-noise ratios, reducing the accuracy and

precision of the wavenumber measurement.

Six (Fe (1.1, 1.5, 1.9 kW), Ar (1.1, 1.5, 1.9 Kw))

Doppler temperatures were calculated from the ICP emission

line profiles using a Voigt fit. This Doppler temperature was

then used for all calculations involving those particular

conditions. For example, an Fe I line with a high SNR is

chosen from the 1.1 Kw ICP data. Voigt profiles are generated

until one is identical to the Fe I (ICP) profile. This

allowed us to estimate the contributions from Gaussian

(Doppler broadening) and Lorenztian (collisional broadening)

components. The Gaussian component is then used for all line

profile calculations involving Fe emission from the ICP at 1.1

kW. We measured the Gaussian and Lorentzian contributions for






52

one Ar and one Fe spectral line at each power level (1.1, 1.5,

1.9 kW) giving a total of six Doppler temperatures.

The Hp line is used to estimate the electron density in

the analytical zone viewed by the LAFTS. These electron

densities applied only to the analytical zone; the region

where the Fe, Ba, Ca, and Sr emission predominantly occurs is

studied. Ar emission is not predominantly from the analytical

zone but from the outer tangential flow zone. No electron

density calculations were made for the latter zone.



Results and Discussion

The results are separated into three groups. First, Fe

and Ar emission lines in the ultraviolet and visible at three

separate power settings (1.1, 1.5, 1.9 kW) were measured.

Second, Ar emission in the near infrared and far visible was

measured. Third, prominent ion lines of Ca, Sr, and Ba at 1.1

kW were measured.

The first section refers to line shifts as a function of

power for 28 Fe lines and 21 Ar lines in the ICP. The power

dependence (1.1, 1.5, 1.9 Kw) is shown to increase Doppler

temperature, electron density, and shift. The iron data are

refer to the analytical zone of the ICP, whereas the argon

data refers to the outer tangential flow. For the argon and

iron data, the line position is given in Angstroms and

physical parameters (line shift, width, etc.) are given in

wavenumbers (cm'). Line positions are given in Angstroms







53
because the references used for transition assignments list

them in this manner. Most research measuring line shifts

report their results in cm'.

Table 4 gives the Doppler temperature (6500 K),

transitions types, wavelengths, line shifts, Doppler and

Lorentzian contributions to the line profile, and the a-

parameter for 21 argon lines at 1.1 Kw. Transition

assignments for the Ar-term configurations are taken from Li

and Humphrey (80). A single argon line at this power is fitted

with a Voigt profile to approximate the Doppler contribution

to the line profile. The resulting Doppler temperature of

6500 K is then used to calculate the Doppler half-width and a-

parameter for all argon lines at this power and the Lorenztian

contribution is calculated with an iterative method:


Av (1/2) *AV + ( (1/4)*Av + Av ) (3)


where Av, is the total line width, AvL is the Lorentzian

component, and AvD is the Doppler component.

All line shift values given are shifted towards the

red(positive values). Negative values are blue shift. Tables

5 and 6 give similar data for Ar at 1.5 and 1.9 kW

respectively.

Figure 12 shows the relationship between , the average

distance of the electron from the nucleus and the average

shift for the 6p-4s, 5p-4s, and 4p-4s transitions of argon at






54

each of the three powers. The values for 6p, 5p, and 4p

levels are estimated using



= (1/2)a[3n2 1(1+1)] (4)



where a, is the Bohr radius, n is principle quantum number,

and 1 is the azimuthal quantum number.

In Tables 7,8, and 9, give the results for Fe at 1.1,

1.5, and 1.9 kW, respectively. For each table (power level),

a Doppler temperature and electron density are calculated.

The electron density was calculated using the Hp line (486.1

nm). The transition configuration, wavelength of transition,

line shift, FWHM, Doppler and Lorentzian contributions to the

line profile, and the a-parameter for each line profile

studied have been measured. The iron concentration was 1000

gg/ml for all runs. The Fe transition assignments were taken

from Fuhr, Martin, and Wiese (81).

In Table 10, the spectral characteristics of 25

additional argon lines obtained with the ICP (1.1 Kw) in the

near-IR and far visible are also given. A large number of

intense Ar lines emitted from the ICP and the HCL occur in

this region. It contains exclusively 4p-4s transitions.

In Table 11, the final set of data presents line shifts

of calcium, barium, and strontium in a 1.1 Kw ICP are given.

These lines are used to investigate the relevance of line

shifts in analytical spectroscopy. We consider each element







55

to have minimal isotope effects (20Ca4 (96.97%), ,,Sr88 (82.56%),

s6Ba'3 (71.66%)); there are no hyperfine structures on the major

isotopes. The electron density and Doppler temperature are

considered to be the same as those in table 4 (Fe at 1.1 kW).

Emission from these ions are typically blue shifted.



Conclusion

The purpose of this chapter is to report the results of

measurements of shifts in line positions that are a function

of pressure broadening, in particular, Stark broadening. Fe

shows red shifts are on the order of 0.01-0.03 cm'. Line

shift studies on Ca, Sr, and Ba resulted in shifts of the same

absolute value as those of Fe but containing a number of blue

shifts. Line shifts for Ar in the visible and ultraviolet are

equivalent to or greater than those for those for Fe. The

shifts for 6p-4s transition is largest, followed by the 5p-4s

transitions and the 4p-4s transitions. We also show that line

shifts for Ar increase with power. Line shifts are shown to

increase as the average distance of the electron from the

nucleus radius increases provided the lower (ground) state is

the same for all three transitions.

From our measurements, line shifts should have little

consequence to the analytical spectroscopist. The data from

Ca, Ba, Sr, and Fe shows that the average shift is on the

order of 0.01-0.03 cm'. Line positions of this accuracy are

not readily measurable by most analytical systems.

























Figure 7. Only one of Pb's three isotopes has any hyperfine
structure (,2Pb) In the ICP, this isotope structure would
be convoluted into one asymmetric line profile making line
shift measurements of each line extremely difficult. (Top x-
axis is A and the bottom is in wavenumbers, cm').






























1.50



I-
S1.00
Z

Z

0.5








27140 27141

WAVENUMBER





















Figure 8. The hyperfine structure of ,7Ta81 taken from a low
pressure HCL. The higher pressure and temperature of the ICP
would make line selection difficult. (top x-axis is A and the
bottom is cm')










































WAVENUMBER




















Figure 9. Typical line profile of Ar I emission from the ICP.
(top x-axis is A and the bottom is cm )









61












42p2. 42p1 42p0






1.50





1.00




U,
Z 05





0.0(
S .. 0 ..-----------


23795 23800 23805

WAVENUMBER




















Figure 10. Line profile of Fe emission (1000 ppm) from
analytical zone in ICP. (Top axis is in A and the bottom is
cm )




































WAVENUMBER




















Figure 11. Diagram outlining the optical configuration of
light sources (ICP and HCL) and optical components outside
LAFTS.














NEUTRAL
DENS I TY


LENS


0


FH Ce
sT crro4df.L


ICP
HCL
















Figure 12. The relationship between , the average atomic
radius, and the shift in the line position is demonstrated for
1.1 kW (*), 1.5 kW (+), and 1.9 kW (A). The value is for
the upper level and should be multiplied by the constant a,
(Bohr radius) for the correct approximation.







67








1.0



,--0.8


E
, 0.6

7-
U-


LLJ
Z

0.2



0.0
20 30 40 50 60
RELATIVE ENERGY LEVEL RADIUS






68

Table 4. The transition, wavelength of transition, 1, line
shift, Al, full width at half maximum (FWHM), AXT, Doppler
halfwidth, AAD, Lorentzian halfwidth, AXL, and the a- parameter
of Ar emission from a 1.1 kW ICP. The Doppler temperature is
6500K. Positive shift values are red-shifted. Below each
general transition (6p-4s, 5p-4s, 4p-4s) the average shift and
standard deviation is tabulated.



Transition X AXA AXT AAD AL a
(A) (cm) (cm ) (cm') (cm )



6p[1.5]2-4s[1.512 3555. 0.3446 0.6335 0.255 0.53 1.73

6p[0.5]o-4s[1.5], 3607. 0.3136 0.6385 0.251 0.54 1.7

AVERAGE SHIFT (6p-4s) 0.3291 (0.0219)



5p[0.5],-4s[1.5]2 3950. 0.1242 0.4405 0.229 0.32 1.16

5p'[1.5]2-4s[l.5], 4045. 0.1038 0.3972 0.233 0.26 0.93

5p[1.5]2-4s[1.5]2 4159. 0.1154 0.4270 0.218 0.31 1.18

5p[1.5]1-4s[1.5]2 4165. 0.1104 0.4385 0.218 0.33 1.26

5p'[0.5]1-4s'[0.5]o 4183. 0.1383 0.4451 0.217 0.34 1.30

5p[2.5]2-4s[1.512 4191. 0.0957 0.3788 0.216 0.25 0.96

5p'[1.5],-4s'[0.5]0 4192. 0.0984 0.3849 0.216 0.26 0.99

5p[0.5]0-4s[1.5], 4199. 0.1243 0.4011 0.216 0.28 1.08

5p[2.5]3-4s[1.5]2 4201. 0.1063 0.5264 0.216 0.44 1.69

5p[0.51]-4s[1.5]2 4252. 0.0807 0.3280 0.213 0.19 0.74

5p[1.5]2-4s[1.5], 4267. 0.1417 0.3746 0.212 0.25 0.97

5p[1.5],-4s[1.5], 4273. 0.1046 0.3578 0.212 0.23 0.90

5p[2.5]2-4s[1.5], 4301. 0.0976 0.3545 0.211 0.23 0.91


5p'[1.5]2-4s'[0.5], 4334. 0.0972


0.4147 0.209


0.31 1.23









Table 4--continued.

5p'[0.5]i-4s'[0.5], 4336. 0.0773

5p'[1.5],-4s'[0.5]i 4346. 0.1284

5p[0.5]o-4s'[0.5], 4511. 0.1179

AVERAGE SHIFT (5p-4s) 0.1095


0.5441 0.209

0.4080 0.209

0.3613 0.201

(0.0182)


4p'[0.5],-4s[1.5]2 6967. 0.0179 0.1698 0.130

4p'[1.5]2-4s[1.5]2 7069. 0.0176 0.1610 0.128

AVERAGE SHIFT (4p-4s) 0.0177 (0.0002)


0.46

0.30

0.25


0.07

0.06


1.83

1.19

1.03


0.44

0.39






70

Table 5. The transition type, wavelength of transition,l,
line shift, Al,, full width at half maximum (FWHM) ,AlT, Doppler
halfwidth, AAD, Lorentzian halfwidth, AIL, and the a-parameter
of Ar emission from 1.5 kW ICP. The Doppler temperature is
6800 K. Positive shift values are red-shifted. Below each
general transition (6p-4s, 5p-4s, 4p-4s) the average shift and
standard deviation are tabulated.



Transition X AX. AXT AA, AlL a
(A) (cm ) (cm ) (cm ) (cm1)


6p[1.5]2-4s[1.5]2 3555 0.5442 0.7807 0.261 0.69 2.20

6p[0.5]0-4s[1.5], 3607 1.178 0.3333 0.257 0.13 0.42

AVERAGE SHIFT (6p-4s) 0.8610 (0.4481)



5p[0.5]i-4s[1.5]2 3950 0.1622 0.5703 0.235 0.47 1.66

5p'[1.5]2-4s[1.5], 4045 0.1383 0.6131 0.229 0.53 1.92

5p[1.512-4s[1.5]2 4159 0.1703 0.4494 0.223 0.34 1.27

5p[1.5],-4s[1.5]2 4165 0.1357 0.6745 0.222 0.60 2.24

5p' [0.5],-4s' [0.5]o 4183 0.1715 0.5064 0.222 0.41 1.54

5p[2.5]2-4s[1.5]2 4191 0.1700 0.4185 0.221 0.30 1.13

5p' [1.5],-4s' [0.5]o 4192 0.1578 0.4338 0.221 0.32 1.20

5p[0.5]0-4s[1.5]i 4199 0.1970 0.5339 0.221 0.44 1.66

5p[2.5]3-4s[1.5], 4201 0.1528 0.4406 0.221 0.33 1.24

5p[0.5],-4s[1.5]2 4252 0.0811 0.4219 0.218 0.31 1.18

5p[1.5],-4s[1.5], 4267 0.1708 0.4779 0.217 0.38 1.45

5p[1.5],-4s[1.5], 4273 0.1606 0.4202 0.217 0.31 1.19

5p[2.5]2-4s[1.5]1 4301 0.1562 0.5351 0.216 0.45 1.74


0.1219 0.8084 0.214 0.75


5p'[1.512-4s'[0.5], 4334


2.41






71

Table 5--continued.

5p' [0.5],-4s' [0.5], 4336 0.1350 0.5421 0.213 0.44 1.77

5p' [1.5]1-4s'[0.5], 4346 0.1797 0.5264 0.213 0.44 1.71

5p[0.5]0-4s' [0.5], 4511 0.1979 0.4068 0.206 0.30 1.21

AVERAGE SHIFT (5p-4s) 0.1564 (0.0283)



4p' [0.511-4s[1.5]2 6967 0.0204 0.1801 0.133 0.08 0.499

4p' [1.5]2-4s[1.5]2 7069 0.0305 0.1801 0.131 0.09 0.571

AVERAGE SHIFT (4p-4s) 0.0254 (0.0071)






72

Table 6.The transition type, wavelength of transition, l,line
shift, AA,, full width at half maximum (FWHM), AIT, Doppler
contribution, AID, Lorentzian contribution, AIL, and the a-
parameter of Ar emission from a 1.9 kW ICP. The Doppler
temperature is 7100 K. Positive shift values are redshifted.
Below each general transition (6p-4s, 5p-4s, 4p-4s) the
average shift and standard deviation are tabulated.




Transition I Als AIr AID AIL a
(A) (cm') (cm') (cm') (cm')


6p[1.512-4s[1.5]2 3555 0.6432 0.5962 0.266 0.48 1.49

6p[0.5]0-4s[1.5]i 3607 1.165 0.3068 0.263 0.08 0.25

AVERAGE SHIFT (6p-4s) 0.9040 (3680)



5p[0.5],-4s[1.5]2 3950 0.2409 0.5762 0.239 0.48 1.66

5p'[1.5]2-4s[1.5], 4045 0.1869 0.5261 0.234 0.42 1.49

5p[1.5]2-4s[1.5]2 4159 0.2064 0.5011 0.227 0.40 1.46

5p[1.5],-4s[1.5]2 4165 0.1894 0.4608 0.227 0.35 1.28

5p'[0.5],-4s'[0.5]0 4183 0.2152 0.7554 0.226 0.69 2.53

5p[2.5]2-4s[1.5]2 4191 0.1635 0.4865 0.226 0.38 1.40

5p'[1.5]1-4s'[0.5]o 4192 0.1587 0.4689 0.226 0.36 1.32

5p[0.5]0-4s[1.5], 4199 0.2129 0.8689 0.226 0.81 2.98

5p[2.5]3-4s[1.5]2 4201 0.1765 1.2200 0.225 1.18 4.35

5p[0.5],-4s[1.5]2 4252 0.1812 0.4081 0.223 0.29 1.08

5p[1.5]2-4s[1.5]1 4267 0.2300 0.5035 0.222 0.40 1.50

5p[1.5]1-4s[1.5], 4273 0.1630 0.5839 0.222 0.50 1.87

5p[2.512-4s[1.5]i 4301 0.1952 0.4673 0.220 0.36 1.36








Table 6--continued.

5p'[1.5]2-4s'[0.5], 4334 0.1699 0.5565 0.218 0.47 1.79

5p'[0.5]i-4s'[0.5]1 4336 0.2405 0.7283 0.218 0.66 2.51

5p'[1.5],-4s'[0.5], 4346 0.1662 0.4559 0.218 0.35 1.33

5p[0.5]0-4s'[0.5] 4511 0.2759 0.6060 0.210 0.53 2.09

AVERAGE SHIFT (5p-4s) 0.1983 (0.0335)



4p'[0.5],-4s[1.5]2 6967 0.0261 0.1967 0.136 0.10 0.612

4p'[1.5]2-4s[1.5]2 7069 0.0261 0.2098 0.134 0.13 0.807

AVERAGE SHIFT (4p-4s) 0.0261 (0.000)






74

Table 7. The transition type, wavelength of transition, X,
line shift, Als, full width at half maximum (FWHM) AX,
Doppler contribution, AAI, Lorentzian contribution, AAL, and
the a-parameter of Fe emission from a 1.1 kW ICP. The Doppler
temperature is 6000 K and the electron density 1.20 x 10 cm .
Positive shift values are red-shifted. For this table, the
average line shift is 0.0185 cm' with a standard deviation of
0.0062 cm The concentration of iron aspirated into the ICP
was 1000 Ag/ml.



Transition X AIA AXI AAD AAL a
(A) (cm ) (cm ) (cm ) (cm1)


aD z4F II 2755.73 0.0452 0.3394 0.274 0.12 0.364

a5D ySF I 2966.90 0.0181 0.3200 0.254 0.12 0.392

a5D y5D I 2994.43 0.0220 0.3144 0.252 0.11 0.363

a5D ysD I 3021.07 0.0232 0.3091 0.249 0.11 0.366

a5D ySDO I 3047.60 0.0201 0.3265 0.248 0.14 0.470

a5D ySDO I 3059.09 0.0172 0.3474 0.247 0.17 0.573

asD zP I 3440.99 0.0154 0.2646 0.219 0.08 0.303

a5F Z3G I 3565.38 0.0190 0.2605 0.212 0.09 0.353

a5F zSGo I 3581.19 0.0194 0.2601 0.211 0.09 0.355

a F zsGo I 3608.86 0.0168 0.2599 0.209 0.09 0.357

aSF zG" I 3618.77 0.0199 0.2849 0.208 0.13 0.518

a5F z2Go I 3631.46 0.0175 0.2555 0.208 0.09 0.360

a F z'G I 3647.84 0.0183 0.2784 0.207 0.12 0.482

a D zFo I 3719.93 0.0127 0.2447 0.203 0.08 0.328

a D z5F I 3745.56 0.0112 0.2470 0.201 0.08 0.330

a5D z25F I 3748.26 0.0119 0.2472 0.201 0.08 0.330

aSF ySF I 3758.23 0.0187 0.2466 0.201 0.08 0.331









Table 7--continued.

a5F ySF I 3763.79 0.0179 0.2460 0.200 0.08 0.332

a3F y3D I 3815.84 0.0221 0.2477 0.197 0.09 0.378

a5F y5D I 3820.43 0.0184 0.2574 0.197 0.11 0.463

a5F ySD I 3825.88 0.0194 0.2439 0.197 0.09 0.374

a5F y5D0 I 3834.22 0.0165 0.2437 0.197 0.09 0.380

a5D zs5D I 3859.91 0.0122 0.2368 0.196 0.08 0.340

a5D z25D I 3886.28 0.0113 0.2324 0.194 0.07 0.299

a3F y3F I 4045.82 0.0192 0.3011 0.186 0.19 0.847

a3F y3Fo I 4063.60 0.0222 0.2263 0.186 0.08 0.358

a3F z5G" I 4383.55 0.0148 0.2173 0.172 0.08 0.386

a3F zSGO I 4404.75 0.0165 0.2151 0.171 0.08 0.388







76

Table 8. The transition type, wavelength of transition, X,
line shift, AAs, full width at half maximum (FWHM), AXT,
Doppler, AAD, Lorentzian AX,, and the a-parameter of Fe
emission from a 1.5 kW ICP. The Doppler temperature is 6200
K and the electron density is 2.05 x 105 cm Positive shift
values are red-shifted. The average shift value for this
table is 0.0188 cm' with a standard deviation of 0.0066 cm'.
1000 gg/ml Fe solution was aspirated into the ICP.



Transition I AX, AAh AXD AXL a

(A) (cm') (cm") (cm') (cm')



a4D Z4FO II 2755.73 0.0431 0.3615 0.273 0.15 0.455

a5D y5F I 2966.90 0.0173 0.3290 0.254 0.13 0.425

a5D y5Do I 2994.43 0.0255 0.3349 0.252 0.14 0.462

a5D ySD I 3021.07 0.0148 0.3115 0.250 0.11 0.366

a5D ySDo I 3047.60 0.0277 0.3251 0.247 0.14 0.470

a5D y5D I 3059.09 0.0124 0.3262 0.247 0.14 0.472

a5D z5aP I 3440.99 0.0214 0.2827 0.219 0.11 0.417

a5F zSGO I 3565.38 0.0218 0.2766 0.212 0.10 0.384

a5F z5Go I 3581.19 0.0209 0.2731 0.211 0.11 0.434

a5F zSG I 3608.86 0.0199 0.2716 0.209 0.11 0.437

a5F z'5G I 3618.77 0.0198 0.2689 0.209 0.11 0.439

a5F z2Go I 3631.46 0.0188 0.2786 0.208 0.12 0.480

aF z5Go I 3647.84 0.0160 0.2639 0.207 0.10 0.402

a5D Z 5F I 3719.93 0.0134 0.2963 0.203 0.16 0.656

a5D z2Fo I 3745.56 0.0146 0.2497 0.202 0.09 0.371

a5D z5F" I 3748.26 0.0097 0.2575 0.201 0.10 0.431

a*F y5F I 3758.23 0.0232 0.3186 0.201 0.19 0.787









Table 8--continued.

a5F y5F I 3763.79 0.0132 0.2509 0.201 0.09 0.373

a3F y3DO I 3815.84 0.0205 0.2625 0.198 0.11 0.462

a5F y'D I 3820.43 0.0181 0.2820 0.197 0.14 0.589

a5F y5D I 3825.88 0.0191 0.2542 0.197 0.10 0.422

aSF ySD I 3834.22 0.0196 0.2476 0.197 0.09 0.380

ta5D z'5D I 3859.91 0.0108 0.2429 0.195 0.09 0.382

a5D zD O I 3886.28 0.0231 0.2651 0.194 0.12 0.514

a3F y3Fo I 4045.82 0.0137 0.2445 0.186 0.10 0.446

a3F y3Fo I 4063.60 0.0115 0.2707 0.186 0.14 0.627

a3F z25G I 4383.55 0.0146 0.2208 0.172 0.09 0.434

a3F z5Go I 4404.75 0.0211 0.2529 0.171 0.14 0.679







78

Table 9. The transition type, wavelength of transition,l,
line shift, Als, full width at half maximum (FWHM), Al,,
Doppler contribution, AAD, Lorentzian contribution, AAL, and
the a-parameter of Fe emission from a 1.9 kW ICP. The Doppler
temperature is 6500 K and the electron density is 2.85 x 1015
cm Positive shift values are red-shifted. The average shift
for all lines listed in this table is 0.0176 cm with a
standard deviation of 0.0132 cm'. The concentration of Fe
aspirated into the ICP was 1000 Ag/ml.


Transition


a'D z4F I

a5D ySFO I
asD y5Do I


a5D ys5D I
a5D ySDO I

aSD yDo I
a5D zyP0 I


a5F z23G I

aF zSGo I

aF zSGo I

a5F zSG I

a5F z5G I

aF zSGo I


aD zF" I

a5D z5F I

aD zIF" I

a5F y5F" I

a5F y FO I


(A)

2755.73

2966.90

2994.43

3021.07

3047.60

3059.09

3440.99

3565.38

3581.19

3608.86

3618.77

3631.46

3647.84

3719.93

3745.56

3748.26

3758.23

3763.79


AA
(cm)

0.0390

0.0249

0.0544

0.0116

0.0226

0.0157

-0.0109

0.0305

0.0221

0.0188

0.0105

0.0386

0.0101

0.0164

0.0210

0.0140

0.0188

0.0193


AXT
(cm )

0.5696

0.3240

0.4185

0.3210

0.3560

0.3155

0.3246

0.2661

0.2872

0.2592

0.2641

0.2796

0.2652

0.2617

0.3310

0.2507

0.3103

0.2912


A1D
(cm )


AIL
(cm )


0.284 0.42 1.23


0.264

0.262

0.259

0.257

0.256

0.228

0.220

0.219

0.217

0.217

0.216

0.215

0.211

0.209

0.209

0.207

0.208


0.11

0.25

0.11

0.17

0.11

0.16

0.09

0.12

0.08

0.09

0.11

0.09

0.09

0.20

0.08

0.17

0.14


0.36

0.79

0.35

0.55

0.35

0.58

0.34

0.45

0.31

0.34

0.42

0.35

0.35

0.79

0.32

0.68

0.56









Table 9--continued.

a3F y3Do I 3815.84 0.0135 0.3395 0.205 0.21 0.85

a5F y5D I 3820.43 0.0176 0.2575 0.205 0.09 0.36

a5F y'5D I 3825.88 0.0152 0.2599 0.205 0.10 0.41

a5F y5D I 3834.22 0.0266 0.4480 0.204 0.35 1.42

a5D z25D I 3859.91 0.0089 0.2895 0.203 0.15 0.61

a5D z5D" I 3886.28 0.0081 0.2760 0.202 0.13 0.53

a3F y3Fo I 4045.82 0.0118 0.2508 0.194 0.10 0.43

a3F y3Fo I 4063.60 0.0076 0.2509 0.193 0.10 0.43

a3F z2G" I 4383.55 0.0165 0.2426 0.178 0.11 0.51

a3F zG I 4404.75 -0.0102 0.2338 0.178 0.10 0.47







80

Table 10. The transition type, wavelength of transition, 1,
line shift, Als, full width at half maximum (FWHM), A4T,
Doppler contribution, AA,, Lorentzian contribution, AAL, and
the a-parameter of Ar emission from a 1.1 kW ICP. The Doppler
temperature is 6500 K. Positive shift values are red-shifted.


Transition


1 AXA AXT
(A) (cm ) (cm )


A(c AX(c
(cm ) (cm )


4p[0.5],-4s'[0.5]o

4p[2.5]2-4s' [0.5]1

4p[0.5],-4s[1.5],

4p[1.5],-4s'[0.5]1

4p[1.5],-4s'[0.5],

4p[0.5],-4s [1.512

4p[1.5],-4s'[0.5]o

4p' [1.5],-4s' [0.5],

4p[2.512-4s[1.5],

4p' [1.5]2-4s' [0.5],

4p' [0.5],-4s' [0.5],
4p[2.5],-4s[1.512

4p[1.5],-4s[0.5],

4p[2.5]2-4s[1.5]2

4p[1.512-4s[1.5],

4p'[1.5]i-4s'[0.5]o

4p'[0.5],-4s'[0.5]o

4p[1.5]1-4s[1.512

4p[1.5]2-4s[1.5]2

4p[0.5]0-4s[1.5],


10472. 0.0159

9787. 0.0172

9660. 0.0121

9356 0.0050

9227. 0.0189

9125. 0.0042

8670. 0.0268

8523. 0.0151

8427. 0.0016

8410. 0.0164

8266. 0.0183

8117.-0.011

8105. 0.0217

8016. 0.0216

8008. 0.0248

7950. 0.0254

7726. 0.0269

7725. 0.0230

7637. 0.0463

7515. 0.0280


0.1365

0.1934

0.1389

0.1963

0.1999

0.1419

0.1506

0.2092

0.1677

0.2250

0.2125

0.1833

0.1686

0.1602

0.1652

0.1610

0.1658

0.1710

0.1683

0.1851


0.0865

0.0926

0.0938

0.0969

0.0983

0.0994

0.1046

0.1064

0.1076

0.1078

0.1097

0.1117

0.1118

0.1131

0.1132

0.1140

0.1173

0.1174

0.1187

0.1206


0.08

0.15

0.08

0.15

0.15

0.07

0.08

0.16

0.10

0.18

0.16

0.12

0.10

0.08

0.09

0.08

0.08

0.09

0.09

0.11


0.77

1.35

0.71

1.29

1.27

0.59

0.64

1.25

0.77

1.39

1.21

0.89

0.74

0.59

0.66

0.58

0.57

0.64

0.63

0.79







81

Table 10--continued.

4p'[0.5]o-4s'[0.5], 7505. 0.0301 0.2297 0.1208 0.17 1.17

4p'[1.5]2-4s[1.5], 7386. 0.0224 0.1785 0.1227 0.10 0.65

4p'[0.5]1-4s[1.5], 7274. 0.0255 0.1806 0.1246 0.10 0.68

4p'[1.5]2-4s[1.5]2 7069. 0.0241 0.1756 0.1301 0.09 0.57

4p' [0.5],-4s[1.512 6967. 0.0260 0.1783 0.1301 0.09 0.57







82

Table 11. Line shifts, Av, of Ca, Sr, and Ba in 1.1 Kw ICP.
The wavenumber positions, v, for hollow cathode lamp
(standard, nonshifted) verses the ICP emission. The Doppler
temperature is 6000 K and the electron density is 1.20 x 10
cm for the ICP. Wavenumber positions given are calibrated
against a Ne I standard for absolute values. Negative values
are blue-shifted.


ELEMENT


v (HCL)
(cm')


v (ICP)
(cm' )


Avs
(cm )
(cm)


Barium II
Barium II
Calcium II
Calcium II
Strontium II


20261.5370
21952.3718
25191.5197
25414.4307
24516.5792


20261.5479
21952.3975
25191.5241
25414.4147
24516.5870


-0.0109
-0.0257
-0.0044
0.0160
-0.0078












CHAPTER 6
A VARIABLE BANDPASS FILTER FOR ULTRAVIOLET/VISIBLE
FOURIER TRANSFORM SPECTROSCOPY



Introduction

Fourier Transform Spectroscopy (FTS) has several

advantages over grating and prism spectrometers. Included in

these are wavenumber precision and accuracy, intensity

precision, and resolving power. Detrimental to the FT

technique is the multiplex disadvantage (7,82), which arises

from shot and/or flicker noises at detected wavelengths other

than the analytical wavelength hitting the detector causing a

decreased signal-to-noise ratio compared to a single slit

scanning dispersive spectrometer system. This disadvantage is

more prominent in FT-UV/VIS then in FT-IR studies. In FT

studies involving a continuum source, this disadvantage can

become severe. The Inductively Coupled Plasma (ICP), Direct

Current Plasma (DCP), and absorbance techniques utilizing a

white light are examples of light sources that can magnify the

multiplex disadvantage.

In our research (83), the Los Alamos Fourier Transform

Spectrometer is used to measure Stark shifts in wavenumber

positions of atoms and ions from a 27.12 MHz atmospheric

pressure argon Inductively Coupled Plasma (ICP). The line






84

positions of atoms and ions in the ICP are compared to the

unshifted line positions of the same species in a commercial

hollow cathode lamp (HCL). To achieve adequate wavenumber

positions (1 part in 4x10'), the lines analyzed need good

signal-to-noise ratios (SNRs). Both the HCL and ICP emit

numerous spectral lines. The ICP, being a hotter source, has

considerable ion line emission. The HCL, with Doppler

temperatures typically 1/10 that of the ICP (6000 K vs 600 K)

has considerable atom line emission, particularly the fill gas

emission in the near infrared. In such measurements of Stark

shifted line positions, we find that the data set available is

small for two reasons. First, the difference in source

emission intensities between the same lines emitted from the

ICP and the HCL reduces the number of lines available for

comparison. Second, the multiplex disadvantage will

substantially reduce the SNR of all lines.

A single filter with two features is desired. First,

the ability to vary the bandpass of the filter over a wide

range. Second, the ability to select the wavenumber region

observed over a wide range. Optical filters do not meet our

specifications for two reasons; the bandpass allowed rarely

matches the needs desired, and the transmission of these

optical filters often increases in the near IR resulting in

excessive shot noise from spectral lines/bands in this region.

Our sources (HCL, ICP) use inert gases having strong emission

in the near IR region, resulting in a serious deterioration






85

of the SNR for lines observed in the ultraviolet and visible.

Hirschfeld and Chase (84) and Hirschfeld and Milanovich

(85) summarized several techniques that could be used to

overcome problems associated with excess light in FT-Raman

experiments. One of these is the use of a nonscanning

subtractive dispersive double monochromator. This filtering

approach has been shown to be a valuable tool for micro-Raman

work (85). We have adapted this filter for use as a bandpass

filter in FT-continuum source work. We demonstrate the

feasibility of this filter using a white light source.



Experimental

Figure 13 shows the configuration of the experiment. A

Bomem DA-3 is used at low resolution (30 cm') for this study.

The gratings (Gl, G2) are manufactured by Instrumental SA.

Both are blazed for 700 nm (1200 grooves/mm, 25 mm x 25 mm).

F numbers of the input lens (L3) and FTS are matched. The

white light source (W.L.), gratings, lens-1 (LI) and lens-3

(L3) are held constant for all experiments. The mirrors

(M1,M2), lens-2 (L2) aperture-2 (A2) are varied for selective

purposes. L1 serves to collimate the light source. Al (5 mm

diameter) and A3 (3-10 mm diameter) are circular apertures.

For Al, we use an aperture as opposed to a slit to maintain

some of the throughput advantage of the FTS for emission

studies. Since we are looking at relatively large bandpasses

(> 500 cm'), the loss in spectral resolution of an aperture






86

verses a slit is not significant. G1 and G2 are mounted back

to back. A2 serves to block unwanted light and to let through

a preset band. A2 is rectangular; A2 and L2 are moved

laterally to select both wavenumber region and bandpass. Ml

and M2 are matched concave mirrors, where M1 transfer the

range of nearly collimated radiation selected by A2-L2 and M2

transfers that range of nearly collimated light on to the

surface of G2 which further disperses and transfers the band

of nearly collimated radiation into L3-A3 where A3 is the

aperture to the FTS.

The bandpass can be adjusted by changing the diameter of

A2 and/or the distance of A2 from GI. Narrower bandpasses can

be achieved with smaller apertures (Al, A2) at greater

distances (A2 from Gl). Decreasing the apertures (Al, A2)

decreased the photon flux entering the FTS and therefore

decreases the multiplex disadvantage. Moving A2-L2

perpendicular to the nearly collimated light reaching M1

resulted in a change in the central wavelength of the band

passed.

The distances between optical components varied slightly

depending on central wavenumber and bandpass selection.

Approximate dimensions are as follows: L1-A1, 1 cm; Al-Grating

1, 11 cm; Grating 1-A2, 12 cm; M1-M2, 26 cm; Grating 2-L3, 12

cm.








Results

Figure 14 shows two spectra (25 scans) of the filtered

white light source with a 2000 cm1 bandpass. The strong

narrow line in both is the He-Ne (6328 A) laser leakage and is

used as a visual reference for alignment of the optics Al-G1-

A2-L2-Ml-M2-G2-L3-A3. For this demonstration the central

wavenumber shifts from approximately 15000 cm1 to 16000 cm'.

Figure 15 shows a narrower bandpass (approximately 1000

cm') that varied from 16500 to 18700 cm' in four windows. The

gratings performance fell off above 20000 cm' and visual

alignment of the optics in Figure 13 below 13000 cm' becomes

difficult by eye; an infrared sensor is needed to see the

radiation. This justified the range (13000-19000 cm')

demonstrated here.



Conclusion

A bandpass filter which varied both wavenumber position

and bandpass and provide a limited spectral range is

demonstrated. The bandpass filter is being developed the

filter for FT-UV/VIS source studies. The filter removed

unwanted light which affected the quality of broad band

multiple line spectra. Removal of the prominent lines of neon

or argon appearing in the 600-900 nm range in hollow cathode

lamp (HCL) emission is also of interest to us. These lines

are often the most intense in HCL emission studies. Optical

filters have two disadvantages which often arise in studies of






88

this sort. First the bandpass of the optical filter may not

match the desired need, especially in the ultraviolet.

Second, transmission of these filters often increases in the

NIR (above 900 nm) and allows the fill gas emission

(particularly Ar) to enter the FTS. Our aim is to replace

optical filters with a simple filter that allows variation in

both position and bandpass of the filter.

We should note that a variety of derivations of this

filter are possible. For example, replacing M1 and M2 with

parabolic mirrors would allow the removal of L2. Ml and M2

would collimate and reflect the radiation to G2 eliminating

the need for L2.



























Figure 13. Outline of bandpass filter. (W.L. = white light,
A = aperture, G = grating)














I Mirror 2


/2


Grating



Grating 1


White
I Light

Al Lens 1


ljrror 1


Lens 2
Lens 2


Lens 3
Lens 3


Ow





















Figure 14. 2000 cm" bandpass of filtered white light source
centered at 15000 cm1 (top) and 16000 cm' (bottom). Sharp line
(6328 A) is He-Ne laser leakage from Bomem FTS alignment
system.




Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E5ZZ3LHQP_DANQFR INGEST_TIME 2017-07-13T22:03:05Z PACKAGE AA00003751_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES



PLASMA SPECTROCHEMISTRY WITH A
FOURIER TRANSFORM SPECTROMETER
BY
THOMAS JOSEPH JOHN MANNING
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
UÍ
tu
.QilUift USaártlfeS
1990

To my family; mother, father, brother, sister,
grandparents; whose support, love, patience,
and caring made this possible.

ACKNOWLEDGEMENTS
I would like to express my extreme gratitude to the
following people whose help dedication, friendship, honesty,
and loyalty made this work possible.
My family played a tremendous role in my recent success.
My mother has been a source of inspiration with her hard work
(teaching, Clemson) and working through her lung problems.
She always asks me the question, "and what does the much
mean," I wish I could say much more to express my gratitude
and love. My father has dedicated his life to his family, and
through his encouragement and prodding, my life has progressed
at a rate which he can take much of the credit for. My
parents have blessed me with good health, a good education,
good morals, and the ability to dream and create with the best
of them.
My sister has long been a stabilizing factor in the
family and her husband, George, a welcome addition. Patrick
is young, cute, impressionable, and has a great arm. Jim, my
big brother, has turned out to be a good brother, someone I
can count on and talk to when needed.
My grandparents, particularly my grandfather who recently
passed away, are sorely missed. I can picture them talking to
iii

friends, on a stoop or in a hall way, in Brooklyn. Bragging
about their grandson, the doctor. I have come to realize that
obtaining a goal of this nature is done as much for others
that you care about, so that they are proud, as it is a
selfish goal to satisfy your curiosity about what you are and
what you can be.
Susan and Jeanne have become friends over the years. I
realize my jokes were not always tactful, but they were meant
to be funny. They were an outlet when the possibility existed
for group politics to get out of hand.
Dr. J. D. Winefordner helped fund my salary for part of
my stay at UF and gave me a the chance to discuss my ideas and
experiments with him and other members of the group.
Doug Hof and Byron Palmer of the Los Alamos Fourier
Transform Facility have become more than just collaborators
since I first started working at LANL in 1987. I would have
left UF out of frustration if it were not for the outlet they
and the LAFTS provided. They let me carry out my ideas and
publish my work so I could establish creditability that group
politics never gave me the chance to establish. Doug and
Byron showed great patience and friendship in allowing me to
come back again and again to the LAFTS and for this I am
truly grateful.
IV

Lots of friends have been made in the lab who have helped
me; Steve L. , Tye B., Ali A., Kin N., Paul J., Andres C., Lynn
P., Alicia 0., Martin A., Jorge V., Barbara K., Leigh Ann F.,
Ed V. , Ben W., Eric G., Tony M., Ramme, Nancy S., Edison B.,
Dennis H., Scott S., Alain B., Wellington M., and En Yu Shoa
are the ones I can name right now. Jill Angus of UNM and LANL
has helped me a great deal. I hope our good relationships
keep up for years to come.
Arlene has recently come to play a major role in my life.
Her love, caring, beauty, and outdoor lifestyle are a welcome
addition to my life. I hope our relationship will be a long,
full one that others will be envious of.
I thank God for Los Alamos, Byron, Doug, and my parents
for sticking by me and helping me work through some tough
situations. Without others, my accomplishments would be
hollow if they could have been achieved at all.
v

TABLE OF CONTENTS
page
ACKNOWLEDGEMENTS iii
ABSTRACT viii
CHAPTER
1 INTRODUCTION 1
2 HIGH RESOLUTION FOURIER TRANSFORM SPECTROMETER
TO IDENTIFY THE ROTATIONAL STRUCTURE OF
2Z+g - 2X\ N/(0,0) IN A HELIUM INDUCTIVELY
COUPLED PLASMA
Introduction 6
Experimental 10
Results and Discussion 12
Conclusions 14
3 A LOW FLOW ATMOSPHERIC PRESSURE
NEON INDUCTIVELY COUPLED PLASMA
Introduction 31
Experimental 31
Discussion 33
Conclusions 34
4 A NONFLOWING VARIABLE GAS INDUCTIVELY COUPLED
PLASMA AS A LIGHT SOURCE FOR HIGH
RESOLUTION SPECTROSCOPY
Introduction 35
Experimental 3 6
Results 37
Discussion 38
Conclusions 41
5 OBSERVATIONS OF LINE SHIFTS AND LINE PROFILES
IN AN INDUCTIVELY COUPLED ARGON PLASMA
Introduction 47
Experimental 50
Results/Discussion 52
Conclusions 55
vi

6 A VARIABLE BANDPASS FILTER FOR ULTRAVIOLET
AND VISIBLE FOURIER TRANSFORM SPECTROSCOPY
Introduction 84
Experimental 86
Results 88
Conclusions 88
7 AN INTENSITY, WAVENUMBER, AND RESOLUTION
STANDARD FOR VISIBLE FOURIER
TRANSFORM SPECTROSCOPY
Introduction 96
Experimental 97
Discussion 97
Conclusions 100
8 A SYSTEM TO STUDY NOISE IN FOURIER
TRANSFORM SPECTROSCOPY
Introduction 104
Experimental 106
Results 107
Conclusions 109
9 RELATIONSHIP BETWEEN RESOLUTION AND
SIGNAL-TO-NOISE RATIO'S IN FOURIER
TRANSFORM SPECTROSCOPY 119
10 CONCLUSIONS 109
REFERENCES 128
BIOGRAPHICAL SKETCH 135
vil

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
PLASMA SPECTROCHEMISTRY WITH A
FOURIER TRANSFORM SPECTROMETER
By
Thomas Joseph John Manning
August 1990
Chairman: James D. Winefordner
Major Department: Chemistry
This dissertation can be interpreted as being two
dimensional. The first dimension uses the Los Alamos Fourier
Transform Spectrometer to uncover various physical aspects of
a Inductively Coupled Plasma. The limits of wavenumber
accuracy and resolution are pushed to measure line shifts and
line profiles in an Inductively Coupled Argon Plasma. This is
new physical information that the plasma spectroscopy
community has been seeking for several years. Other plasma
spectroscopy carried out includes line profile studies, plasma
diagnostics, and exact identification of diatomic molecular
spectra.
viii

The second aspect of the dissertation involves studies of
light sources for Fourier Transform Spectroscopy. Sources
developed use an inductively coupled plasma (ICP) power
supply. New sources (neon ICP, closed cell ICP, and helium
ICP) were developed and new methods to enhance the performance
and understand a Fourier Transform Spectrometer were studied
including a novel optical filter, a spectrum analyzer to study
noises, and a standard to calibrate and evaluate a Fourier
Transform Spectrometer.
ix

CHAPTER 1
INTRODUCTION
This dissertation will introduce new applications and
studies that couple a Fourier Transform Spectrometer operating
in the ultraviolet and visible region of the spectrum to
various plasma and discharge sources.
Fourier Transform Spectrometers are built for three
reasons; intensity precision and accuracy, high resolving
power, and wavenumber accuracy and precision. The throughput
advantage and multiplex advantage also make the Fourier
Transform Spectroscopy (FTS) attractive but are not the main
reasons the worlds most powerful ultraviolet and visible
Fourier Transform Spectrometer's have been built.
The first advantage is intensity precision and accuracy
in the line profile. This aspect is utilized less than the
other two in this research but is used in temperature
measurements (e.g. Boltzmann methods involving diatomics
(rotational temperature) and atoms (excitation temperature)),
line profile studies, isotope ratio determinations, etc. The
second advantage is resolution or high resolving power. This
advantage was used to look at rotational levels of molecules,
physically resolve line profiles, and identify isotope shifts
and hyperfine structure of atomic lines. The final advantage
1

2
is the excellent wavenumber accuracy and precision. This
facet of FT-UV/VIS spectroscopy was utilized to measure line
shifts in an Inductively Coupled Plasma (ICP), and better the
existing wavenumber accuracy of a well studied N2+ transition
in ultraviolet. A summation of each chapter follows.
The second chapter is a continuation of my master's
thesis. The Helium Inductively Coupled Plasma described in
this paper was first developed for nonmetal atomic emission
spectroscopy at Los Alamos National Lab (LANL). The source
proved to be very stable (low flicker noise component)
compared to other flowing systems, had very little background
and is sufficiently energetic to be used as a light source.
This project was the first published work using the Los Alamos
Fourier Transform Spectrometer. It clearly demonstrated that
the instrument, a 13 million dollar national facility, was
functioning at a high level. It also demonstrated the He-ICP
is a good high resolution light source. The N,+ (0,0)
transition was resolved with higher wavenumber accuracy (>1
part in 10?) than had been previously been reported in the
literature.
This same torch was used to generate a low flow,
atmospheric pressure Neon Inductively Coupled Plasma. Chapter
three is a brief outline of the first Neon Inductively Coupled
Plasma described in the literature. This was originally
generated for inductively coupled plasma-mass spectrometry
studies, but the price of neon proved too expensive to

3
continue this research.
Chapter four deals with a nonflowing, variable gas
inductively coupled plasma as a light source for high
resolution spectroscopy. This work was recently issued a
patent by the United States Patent office. It generated
plasmas with 10 different gases (He, Ne, Ar, Kr, Xe, H2, N2,
CO,, C,H2, O,) over a wide range of pressure (0.01-100 torr) .
Work was also initiated to demonstrate the static, low
pressure source could be interfaced to a mass spectrometer
with several advantages over the traditional ICP-MS. For
Fourier Transform Spectroscopy (FTS) research, the static
environment is important to minimize flicker noise in the
source. Varying the gas allows us to vary the Doppler
temperature and electron density as well as the amount of
background light flux.
Chapter five is a physical analytical study. This
chapter represents the largest project undertaken; the
measurement of line shifts and line profiles of light emitted
from a argon inductively coupled plasma. There are small
shifts in wavenumber position between low pressure sources (no
collisional broadening, no shift) and atmospheric pressure
sources. Line profiles were deconvoluted to measure
broadening contributions from Doppler and collisional
processes in the atmospheric pressure argon ICP.
Approximately 180 lines with a signal-to-noise ratio (SNR)
greater than 100 were measured with the LAFTS. This pushed

4
the wavenumber position and resolution capabilities of the FTS
technique. Argon, iron, calcium, barium, and strontium
emission line profiles from an inductively coupled argon
plasma were recorded at sub-Doppler resolution.
The project involving line shifts concluded with the
development of a optical filter that allowed variance of both
bandpass and wavenumber position. It was designed for
ultraviolet and visible Fourier Transform Spectroscopy in
conjunction with the ICP. Its purpose was to reduce the
amount of unwanted light entering the FTS . This filter is
discussed in chapter six.
The Re I (575 nm) line has both hyperfine structure (HFS)
and isotope shifts (IS). This structure was proposed as a
standard to determine the quality of spectra acquired from a
FTS. The reason for doing this is simple. Instrument
companies producing instruments have been selling bad
instruments, that is Fourier Transform Spectrometers that do
not come close to meeting their published specs (often by
orders of magnitude). The people purchasing the instruments
typically do not know much about FTS and are not aware of the
problems. By proposing a standard for wavenumber position,
intensity precision, and resolution, it will allow people to
judge the quality of the instrument they are about to buy or
to judge the performance of the instrument at any given time
during its lifetime. It can also serve as a calibration
standard when fine tuning the instrument. Chapter seven

5
describes this standard.
Chapter eight covers noise studies in Fourier Transform
Spectroscopy using a relatively simple arrangement. The
output of the light source is chopped by a mechanical
chopper. Chopping simulates the constructive and destructive
interference in an interferometer. When the beam is blocked
by the chopper, this simulates destructive interference, and
when light is passed, this simulates constructive
interference. The beam is chopped at a rate (Hz) similar to
that produced by constructive and destructive interference in
a Michelson interferometer. This work indicates that a shot
noise limited case has not been achieved by analytical
spectroscopists, despite their claims.
Analytical spectroscopists typically use a ratio of
signal and noise intensities to calculate the signal-to-noise
ratio in FTS. This method fails at poor resolutions when the
line profile is broadened by instrumental effects decreasing
the amplitude (peak height) of the line profile. Provided our
integration times are equal, there should be identical signal-
to-noise ratios for high and low resolutions alike. This
research has found this not to be the case. At poor
resolutions, the loss of signal amplitude, caused by
instrumental broadening, negatively affects the SNR. Many
analytical studies have been carried out at poor resolutions
and LODs based on amplitude SNRs giving low values. Many of
these studies claim to have achieved shot noise limited cases.

6
Work in chapter seven gives conclusive evidence that
analytical chemists have not achieved a shot noise limited
spectra.

CHAPTER 2
USE OF A HIGH RESOLUTION FOURIER TRANSFORM SPECTROMETER
TO IDENTIFY THE ROTATIONAL STRUCTURE OF THE B2EU+ - )TZg+
TRANSITION OF N2+ (0,0) IN A HELIUM
INDUCTIVELY COUPLED PLASMA
Introduction
The use of Fourier Transform Spectrometers in the near,
middle and far-infrared spectral region is common in many
analytical labs. Since a wealth of atomic information lies in
the visible and ultraviolet regions, there has been a great
deal of effort by various groups worldwide to build Fourier
Transform Spectrometers that operate over this region of the
spectrum. In 1975 Horlick and Yuen (1) reviewed
spectrochemical analysis with a Fourier Transform Spectrometer
(FTS). More recent reviews by Faires (2) and Thorn (3)
discuss theory, instrumentation, and applications of a FTS in
Atomic Emission Spectroscopy (AES). Stubley and Horlick
(4,5,6), Marra and Horlick (7), Faires (8,9), Faires, Palmer,
Engleman, and Niemczyk (10), and Faires, Palmer, and Brault
(11) carried out pioneering work in FT-ICP-AES. Various
aspects of the LAFTS have been previously reported (12,13).
7

Los Alamos Fourier Transform Spectrometer
The Los Alamos Fourier Transform Spectrometer (LAFTS),
housed in a 3360 ft” building, is the highest resolution
visible and ultraviolet FTS in the world. Table 1 lists some
of its characteristics. The instrument is currently operating
with single pass optics allowing for a maximum resolution of
0.0026 cm’ (0.023 pm at 300 nm) . The interferometer is housed
within a 14 foot by 7 foot vacuum tank (30 mTorr achieved) and
utilizes two entrance ports for experiments. The FTS sits on
a 4 foot by 10 foot NRC vacuum compatible optical table
isolating the instrument from vibration.
Figure 1 shows the folded design of the instrument. The
LAFTS utilizes a double sided interferogram which improves the
signal-to-noise ratio and helps eliminate phase problems. The
A/D converter is composed of seven 16-bit A/D converters,
scale amplifiers, and an 18-bit D/A converter for calibration.
Parsons and Palmer (13) give detailed characteristics and
research plans for the LAFTS.
For inductively coupled plasma-atomic emission studies
(ICP-AES) the advantages of the FTS are intensity precision,
wavenumber accuracy, and high resolution. Provided electronic
noise is minimized, the limiting noise in the FT-ICP system
will be the plasma. Several papers (14-18) have discussed
sources of noise in Argon ICPs.
8

9
Helium Inductively Coupled Plasma
Argon is the most common plasma gas used in Inductively
Coupled Plasmas for elemental analysis. Abdallah and Mermet
(19) compared temperatures of helium and argon in ICP and MIP
systems. Abdallah et al. followed with a paper (20) utilizing
a 50 MHz Helium-ICP at atmospheric pressure for analytical
studies. Seliskar's laboratory (21-24) conducted work with a
reduced pressure torch that is adaptable to a commercial ICP
unit. Chan et al. (25-27) developed and tested a tangential
flow, low flow, atmospheric pressure torch that is constructed
entirely of macor and adapts readily to commercial units.
Montaser and Van Hoven (28) give a critical review on mixed
gas, molecular gas and helium ICPs. Due to the high
ionization potential of helium (24.6 eV) versus argon (15.8
eV), it has achieved improved limits of detection for elements
with relatively high ionization potentials (Cl 12.9 eV, Br
11.8 eV) (25). He-ICPs have fewer spectral lines than Ar-
ICPs in the desired regions of AES decreasing the chance for
overlap (26). Tan et al. (29) recently introduced a laminar
flow He-ICP torch and demonstrated its use in the analysis of
some nonmetals.
N-,+ f o. cn
In 1933 Coster and Brons (30) published detailed work on
the emission spectrum of N,’. Diatomic nitrogen (N2,N2*) has
been utilized as a rotational temperature indicator in RF

10
plasmas (31-39) and various discharges (40-46). Most papers
use the unresolved vibrational bandhead or the partially
resolved rotational structure. The use of Raman scattering
from N2 as a temperature indicator has also been demonstrated
by several groups (47-50). Gottscho et al. (51) give an
account of perturbation effects within the B2£u+ -
transition of N,+. Herzberg (52) gives an excellent discussion
of both the theoretical and experimental applications of the
transition. Montaser and Van Hoven (28) give references for
additional uses of N2 and N,* in plasma diagnostics. Nitrogen
is used in plasmas for metal surface treatment (53) , synthesis
of nitrogen oxydes (54), and CO, CO,, and gas lasers (55).
The use of N,+ in sub-doppler resolution studies is discussed
in this paper.
The first negative band of N,+ has several advantages for
rotational temperature determinations in plasmas. The (0,0)
bandhead (391 nm) is in an accessible region of the spectrum.
The Q branch (AJ=0) is forbidden decreasing the number of
rotational lines present relative to N,(0,0), N,(0,1), and
OH(0,0) (56), which are used as rotational temperature
indicators. At moderate resolution (<0.25 cm'), the
rotational structure can be adequately resolved.
This chapter presents a low flow, laminar flow,
atmospheric pressure helium ICP torch. Nitrogen is seeded
into the helium plasma and the emission from N,+ is analyzed
using the Los Alamos Fourier Transform Spectrometer.

11
Experimental
Helium inductively coupled plasma torch
The torch is optimized for low flow (2-6 L/min),
simplicity in design, and adaptability to a commercial unit.
Chan et al. (25) present a low flow (7 L/min), tangential
flow, macor torch. Two problems arise with this design.
First, macor (Corning, Horseheads, NY) , a heat resilient,
machinable ceramic, is extremely brittle and time consuming to
work with. Second, the spiral design allows the plasma to
wrap around the torch at low flow rates increasing the chance
for damage to the macor torch. An attempt to run a mixed
argon-helium plasma also proved detrimental to the macor.
These two problems were overcome by a) reducing the amount of
macor and constructing the torch largely from teflon and b)
making the torch laminar flow and keeping the plasma above the
macor tip.
Figure 2 shows the diagram of the torch used in this
study. The plasma gas enters the torch through two side ports
spaced 180° apart. Two ports are utilized to give an even
distribution of gas through the eight symmetrically spaced
channels that carry the plasma gas from the teflon channels
through the bores in the macor tip and into the plasma. The
sample is carried up through the center of the teflon and
macor pieces and enters the plasma region through a 2/1000
inch aperture. The torch handles 10,000 ppm Na solution with
no clogging problems. This system uses a fritted disk

12
nebulizer (11). It has higher efficiency (25% with 0.3 L/min
gas flow), lower cost, and a decreased noise power spectrum
(14) versus the ultrasonic nebulizer. Helium does not work
with a concentric nebulizer due to its relative lack of
viscosity versus argon. An extended guartz tube was used to
keep impurities (N,, H20 etc.) from the atmosphere out of the
plasma. It has one or two wraps of teflon tape around its
base for an airtight fit with the base. A four turn, gold
plated (57) coil was used to generate the 27.12 MHz RF field.
The central teflon piece and the teflon base (2" diameter) fit
tightly together. The torch is supported on a teflon stand in
the ICP box. The tesla coil is attached, via an ignition
wire, to one of the side ports with a copper clip. Quartz
tubing (1/4" diameter) is inserted into the two side ports.
A 12/5 balljoint is inserted into the sample uptake port. The
macor tip has a 0.5 inch diameter and a 0.5 inch height. It
fits tightly with the teflon channel (see Figure 2). Under
normal operating conditions this fit is sufficient to seal the
tip from outside contaminants and keep the sample and plasma
gas separate. As an added precaution, the outside of the
macor-teflon connection is wrapped with teflon tape. The
macor to teflon connection can be fused with chemical etching
techniques or a high temperature quartz based cement can be
used as a sealant. Impedance matching utilized has been
previously described in the literature (58,59).

13
Inductively coupled plasma
This study employs a 27.12 MHz RF Plasma Products (Cherry
Hill, NJ) HFP 5000 D model Inductively Coupled Plasma unit.
The ICP operating conditions are outlined in table 2. The
nitrogen is 99.9995% certified pure and was obtained from
Union Carbide-Linde Division (Albuquerque, NM).
Los Alamos Fourier transform spectrometer
The parameters of the LAFTS used for this study are as
follows: The spectral output is from 14000 cm1 to 40000 cm'1.
The detectors utilized are RCA lP28s. The beamsplitters are
aluminum coated quartz and no optical filters are used. The
resolution chosen (0.09 cm1) is sub-Doppler for nitrogen (19).
The number of points taken is 889,219, which is zero filled to
20 • .
2 (1,048,576) points for transformation. The data
acquisition time for the 20 scans is 1 hour 17 minutes and the
computer transform time is 42 minutes. The spectrum is
unapodized (boxcar). Atmospheric pressure in the LAFTS (7300
ft above sea level) is 593 Torr, and the tank temperature is
21°C.
Results and Discussion
Table 3 gives the line positions (±0.01 cm1) of the P and
R branches observed in the Helium Inductively Coupled Plasma.
The accuracy of the (0,0)band data given by Coster and Brons
(30) is updated by 0.1 to 0.2 cm1.
Figure 3 shows part of the (0,0) spectra. The P (1-38)

14
and R(1-10) branches are identified. The pattern is
consistent for larger branch numbers not shown. The doublet
splitting (J-l/2,J+l/2) is evident for both branches (P >16,
R >10). The alternation of line intensity (even > odd) is
unmistakable for both branches. Doublet splitting and even-
odd alternation of line intensity are treated theoretically by
Herzberg (52).
Figure 4 gives the Fortrat diagram for the P and R
branches in the He-ICP. The P branch folds over itself at
25540.10 cm'1 giving the vibrational bandhead. The K value
used for all branches that are doublet split is J-l/2. At
higher K values (R >40, P>44) only the even branches are
detectable from this source (He-ICP).
The method to calculate the rotational temperature from
a well resolved structure is taken from Abdallah and Mermet
(19). This method allows for relative line intensity
measurements (as opposed to absolute) for either the P or R
branches. For the P branch,
Log al = f(K"(K"-l)) (1)
K"
and the R branch,
Loa al = f((K"+l)(K"+2)) (2)
K"+l
where a =1 (even number lines) , 2 (odd number
lines)
f = -Bhc/kT = (-2.983/T)
I = experimental line intensity
K" = quantum number of lower state
T = temperature (kelvin)

15
Plotting the left side of equation 1 or 2 (for varying
values of I) versus the right side (increasing values of K")
of the respective equation yields a straight line with a slope
proportional to 1/T. Rotational temperatures for the He-ICP
using the parameters and data listed above are R(odd, 1700 K),
R(even, 1680 K), P(odd, 1830 K), and P(even, 1660 K).
Various papers (35,38,39) outline similar Boltzmann
methods to calculate rotational temperatures in plasmas. This
method allows for four rotational temperatures (P (even,odd),
R(even,odd)). Thorne (60) claims that the Boltzmann method for
N2+ (0,0) is suitable for measurements under 5000 K. Cabannes
(39) claims it is allowable for temperatures under 10000 K.
Kornblum and de Galen (61) have interpreted the resulting
measurement as the gas kinetic temperature.
Conclusion
A new He-ICP torch has been demonstrated. The torch has
also been used to generate a neon ICP (62) and is being
investigated as an ionization source for helium and neon
ICP-MS studies. The He-ICP is not being pursued for atomic
emission spectroscopy.
The use of N2+ as a diagnostic tool is common. This paper
allows the spectroscopist to easily identify and use the
structure for rotational temperatures. The line position
accuracy of the rotational levels is improved against those
currently in the literature.

Figure 1. Optical Elements: a) cat's-eye reflectors b) beam
splitter and carousel c) 8-inch turning mirrors d)
collimator optics e) output mirrors . Light enters from
either the end or side port and is collimated by d. There it
goes to the upper beam splitter, b, and is divided. These two
beams go to the turning mirrors c and to the primary mirror of
the cat's-eye reflector, a. In the cat's-eye reflector the
beam is focused down onto the secondary mirror and reflected
back to the primary where it is made parallel again. The
effect of the cat's-eye reflector is to reflect the beam back
along a path that is parallel but displaced. From the cat's-
eye reflectors the beams return to the 8-inch mirrors and then
to the lower beam splitter. There the beam is recombined and
interferes with itself. The light then travels to the output
mirrors, e, that focuses the beam and to the flat mirrors that
direct the beam to the detectors.

¿I

Figure 2. Diagram of low-flow, atmospheric pressure He-ICP
torch. The top inset shows the macor tip and teflon central
channel interface in greater detail. The bottom inset shows
the eight plasma gas holes surrounding the sample gas inlet on
the macor tip.


Figure 3. The four panels show a portion of the rotational
structure of the N2+(0,0) band. The top x-axis is angstroms
and the bottom is wavenumbers (cm1) .

INTENSITY
21

INTENSITY
22
Figure 3—continued

INTENSITY
23
Figure 3—continued

INTENSITY
24
Figure 3—continued

Figure 4. Fortrat diagram of P and R branches. At 25540.10
cm1, the P branch (P12, P13, P14) folds giving the
bandhead.

50
40
30
20
10
0
â–¡
R BRANCH
P BRANCH
¡00
25700
1
25900
1
26100
WAVENUMBER

27
Table 1. Los Alamos FOurier Transform Spectrometer
Specifications
Spectral Ranges:
Region Range Beamsplitters Detectors
UV/VIS 200 nm-1.1 um quartz/aluminum
NEAR IR
FAR IR
0.8-5.5 um
4-20 um
CaF-GaP
KCl-Ge
photomultiplier,
si PIN diode
InSb
HgCdTe
Resolution and Operating Parameters:
Optical Path
Difference
Resolution
Sample Rate
Number of points
Maximum Transform
Size
2.5 meters (single pass)
5.0 meters (double pass)
0.0026 cm (single pass)
0.0013 cm’1 (double pass)
5000 Hz not digitally filtered
40000 Hz with digital filtering
reduced to 5000 Hz throughput
>4,000,000 points
222 (4,194,304)

28
Table 2. Helium inductively coupled plasma operating
conditions
Source Unit
Frequency
Forward Power
Reflected Power
Plasma Gas Flow
Nebulizer Flow
Sample Uptake Rate
Viewing Height above
load coil
Aperture to FTS
RF Plasma-Therm 5 kW unit
27.12 MHZ
1.5 kW
<5 W
4 L min
0.5 L min
0.3 mL min
14 mm
7 mm

Helium inductively coupled plasma
P BRANCH
K Value J(±l/2)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
J-l/2
16
J+l/2
17
J-l/2
17
J+l/2
18
J-l/2
18
J+l/2
19
J-l/2
19
J+l/2
20
J-l/2
20
J+l/2
21
J-l/2
21
J+l/2
22
J-l/2
22
J+l/2
23
J-l/2
23
J+l/2
24
J-l/2
24
J+l/2
25
J-l/2
25
J+l/2
26
J-l/2
26
J+l/2
27
J-l/2
27
J+l/2
28
J-l/2
28
J+l/2
29
J-l/2
29
J+l/2
30
J-l/2
WAVENUMBER (cm1)
25562.36
25558.76
25555.54
25552.62
25549.96
25547.66
25545.67
25543.81
25542.47
25541.37
25540.59
25540.11
25540.11
25540.11
25540.59
25541.15
25541.37
25542.16
25542.47
25543.54
25543.85
25545.14
25545.44
25547.12
25547.35
25549.40
25549.65
25552.20
25552.59
25554.82
25555.09
25557.99
25558.28
25561.46
25561.76
25565.24
25565.56
25569.32
25569.66
25573.71
25574.06
25578.42
25578.76
25583.41

30
Table 3—continued.
30
J+l/2
25583.77
31
J-l/2
25588.70
31
J+l/2
25589.06
32
J-l/2
25594.31
32
J+l/2
25594.67
33
J-l/2
25600.21
33
J+l/2
25600.56
34
J-l/2
25606.41
34
J+l/2
25606.76
35
J-l/2
25612.88
35
J+l/2
25613.22
36
J-l/2
25619.59
36
J+l/2
25619.99
37
J-l/2
25626.96
37
J+l/2
38
J-l/2
25634.12
38
J+l/2
25634.39
39
J-l/2
25640.85
39
J+l/2
25641.80
40
J-l/2
25648.44
41
25656.19
42
25668.43
43
25676.62
44
25685.82
R BRANCH
Value J(±1/2)
Wavenumber
0
25570.30
1
25574.77
2
25579.51
3
25584.57
4
25589.93
5
25595.61
6
25601.58
7
25607.85
8
25614.44
9
25621.33
10
J-l/2
25628.41
10
J+l/2
25628.56
11
J-l/2
25635.89
11
J+l/2
25636.12
12
J-l/2
25643.67
12
J+l/2
25643.90
13
J-l/2
25651.76
13
J+l/2
25651.99
14
J-l/2
25660.16
14
J+l/2
25660.39

Table 3—continued
15
J—1/2
25668.86
15
J+l/2
25669.09
16
J-l/2
25677.79
16
J+l/2
25678.09
17
J-l/2
25687.18
17
J+l/2
25687.40
18
J-l/2
25696.72
18
J+l/2
25697.02
19
J-l/2
25706.62
19
J+l/2
25706.88
20
J-l/2
25716.78
20
J+l/2
25717.08
21
J-l/2
25727.27
21
J+l/2
25727.58
22
J-l/2
25738.05
22
J+l/2
25738.38
23
J-l/2
25749.15
23
J+l/2
25749.53
24
J-l/2
25760.53
24
J+l/2
25760.86
25
J-l/2
25772.21
25
J+l/2
25772.55
26
J-l/2
25784.17
26
J+l/2
25784.54
27
J-l/2
25796.46
27
J+l/2
25796.82
28
J-l/2
25809.02
28
J+l/2
25809.39
29
J-l/2
25821.89
29
J+l/2
25822.26
30
J-l/2
25835.04
30
J+l/2
25835.41
31
J-l/2
25848.48
31
J+l/2
25848.86
32
J-l/2
25862.21
32
J+l/2
25862.58
33
J-l/2
25876.22
33
J+l/2
25876.59
34
J-l/2
25890.44
34
J+l/2
25890.85
35
25905.36
36
J-l/2
25919.96
36
J+l/2
25920.27
37
J-l/2
25934.16
37
J+l/2
25935.11

CHAPTER 3
GENERATION OF A LOW FLOW ATMOSPHERIC PRESSURE NEON ICP
Introduction
A 27.12 MHz Neon Inductively Coupled Plasma (Ne-ICP) has
been generated in this laboratory. Argon has long been used
as the plasma gas for an ICP. Abdallah and Mermet (19) and
Chan et al.(25) have generated atmospheric pressure helium
plasmas in inductively coupled plasma (ICP) units. Walters et
al. (63) generated a reduced pressure (0.5-14 torr) 144 MHz
neon ICP for electrodeless discharge lamp studies. This is
the first known report of an atmospheric Ne-ICP. The torch
design has been used for helium ICP studies and is described
by Manning et al.(64). The neon ICP is being investigated as
a possible atomization source for adaptation to a commercial
inductively coupled plasma-mass spectrometry instrument.
Experimental
The Ne-ICP is generated with a 27.12 MHz RF Plasma
Products HFP 5 kW unit (Cherry Hill, NJ). The forward power
for the ICP is 500 W with less than 5 W reflected (58). The
low flow (3 1/min), laminar flow, atmospheric pressure torch
is described by Manning et al.(64). It utilizes an extended
quartz sleeve to minimize impurities (N,, H,0, etc.) from the
32

33
atmosphere entering the plasma. The nebulization system
employed is the fritted disk nebulizer (11). The sample is
pumped to the fritted disk with a standard peristaltic pump.
The flow rate of the plasma gas is 3 L/min with a minimum of
2 L/min and a maximum of 7 L/min possible. The nebulizer gas
flow rate is 0.3 L/min with a sample uptake rate of 0.3
mL/min. The nitrogen is seeded in at a rate of 5 mL/min. The
neon gas is supplied by Union Carbide-Linde division gas
distributors (Albuquerque, NM).
Los Alamos Fourier Transform Spectrometer
The Los Alamos Fourier Transform Spectrometer (13), a
national facility for high resolution atomic and molecular
spectroscopy, is used in this experiment for medium resolution
(0.07 cm') spectra in the near IR, visible, and ultraviolet
(8500 cm'1 to 33372 cm') . Aluminum coated quartz beamsplitters
are employed and the detectors utilized are silicon PIN
diodes. The number of points taken for each run is 953,863
. . °0 , ,
and the transform size is 2“ (1046000) points. The run m
which only water is introduced into the plasma consists of 10
scans (39 minutes data acquisition time) and took 16 scans in
the run introducing both water and N, (64 minutes data
acquisition time). The number of scans is limited by the
current availability of research grade neon. The cost of neon
is the limiting factor.

34
Results and Discussion
Neons' ionization potential (21.56 eV) is higher than
that of argon (15.76 eV) but lower than helium (24.587 eV).
This should offer better detection limits for various
nonmetals (eg. F, Cl, Br, etc.). The neon plasma as a source
does not have the luminosity of the argon ICP and is similar
to the helium ICP in this respect. The plasma is stable with
powers as high as 1.5 kW. Introducing sample reduces its
stability. With an aerosol entering the plasma, it could not
operate at powers greater than 600 W. Above this threshold
the plasma either extinguished or formed numerous filaments
and produced a non-homogeneous plasma. The plasma operated
with powers as low as 150 W. Luminosity appears linear with
power.
An attempt to measure the rotational temperature using
N2* (0,0) (391.44 nm) and OH(0,0) (306.4 nm) is made. The N,+
bandhead has a SNR of 4.2 and the R branches used in
diagnostics are not discernible making reliable measurements
impossible. The OH(0,0) band is not present at all in the
spectrum. The HB line is used to measure the electron number
density (65) at 8xl013 cm'3. The fact that both species
(OH, N2+) , which are strong UV emitters in plasmas and
discharges, do not have a readily detectable emission rate,
coupled with the low electron density measurement (8xl013 cm3)
and the low input power (500 W), give reason to believe the
Ne-ICP in this setup is cool compared to the argon and helium

35
plasmas (19) run at higher powers. Bands from the first
• • 3 3H • •
positive system (BZg - A J of N2 are obvious in the near IR
and visible. This transition (10,000 cm'-17,000 cm'1) requires
less energy for excitation than OH(0,0)
( >31,000 cm 1) .
Conclusion
An atmospheric pressure neon ICP is generated. In the
current setup it operates reliably at 500 W forward power. The
Ne-ICP is being pursued as an atomization source for a
commercial ICP-MS setup. A greater amount of diagnostics is
not feasible due to the expense of running a Ne-ICP. A single
Ne line (585 nm) was used to determine the Doppler
temperature. The value (1058 K) was obtained by deconvoluting
the line profile with a Voigt profile.
Using a Ne-ICP as an ion source for ICP-MS studies should
offer the advantage of "freeing" various mass values that are
not accessible with argon as the plasma gas.

CHAPTER 4
A NON-FLOWING, VARIABLE GAS INDUCTIVELY COUPLED PLASMA AS
A LIGHT SOURCE FOR HIGH RESOLUTION SPECTROSCOPY
Introduction
Ten self-contained 27.12 MHz Inductively Coupled Plasmas
(ICP) are generated in a static torch. The torch offers
spectroscopists a closed system, low pressure (0.01 to 100
Torr), variable power (currently up to 1 kW) plasma. Doppler
temperatures (Na D, 589.0 nm) measured are approximately 5
times that of electrodeless discharge lamps (EDLs) or Hollow
Cathode Lamps (HCLs) operating under normal conditions. It has
improved noise considerations verses the flowing ICP
traditionally used.
Plasmas are generated using helium, neon, argon, xenon,
krypton, hydrogen, nitrogen, oxygen, carbon dioxide, and
ethylene. In some cases, particularly xenon, the closed
system minimizes the consumption of expensive gases. In the
case of hydrogen, it minimizes the potential hazard of working
with the gas in a flowing system. Ideas for absorption,
ICP-MS, and lasing action in a static plasma are discussed.
This torch generates a low noise, high power plasma that can
be used for a variety of spectroscopic studies.
36

37
Experimental
Figure 5 shows the torch design utilized in this study.
The quartz tube has a 17 mm OD, 15 mm ID, and a 14 cm length.
A quartz window is fused to the viewing end of the tube. In
some circumstances, using grounded clips helps the plasma
symmetrically fill the entire tube. End on viewing increases
the luminosity of the source. The four turn coil is gold
plated 1/8 inch copper tubing. The ICP instrument is a RF
Plasma Products HFP 5000 D (5 kW) unit. Often the plasmas are
initiated by introducing the field (RF ON) and slowly
increasing the power. In instances where this is not
possible (typically >10 Torr), a tesla coil discharge is sent
through one of the grounded clips to ignite the plasma. A
vacuum rack equipped with a liquid nitrogen cold finger, a
roughing pump, and a diffusion pump is utilized for evacuating
the cell and filling it with the desired gas. The cell is
evacuated to 10'5 Torr, back filled with the desired plasma
gas, evacuated to 105 Torr, and filled to the desired
pressure.
The cell is demountable and easily inserted into the ICP
coil. A 12/5 balljoint connects the rack and the torch. As a
precaution, nitrogen is blown across the quartz torch for
cooling purposes. Pressures given are the fill values at room
temperature. Powers recorded are the forward minus the
reflected values.
The Los Alamos Fourier Transform Spectrometer is used to

38
record the spectra at high (0.039 cm') resolution with 5
scans. The detectors are Silicon PIN diodes, and the
beamsplitters are aluminum coated quartz. The spectral window
transformed is from 8500 cm' to 40000 cm'1. 2,052,056 points
are acquired and zero filled to 2,097,152 (2"‘) points.
Additional parameters of the LAFTS are supplied by Manning et
al (64) and Parsons and Palmer (13). All gases used are
certified research grade (>99.995 %) .
Results
Figure 6 shows the Na D (589.0 nm) line profile in a Xe-
ICP. Sodium (and lithium) impurities are present in most of
the spectra obtained. They are attributed to impurities in
the quartz. The sodium D line is used for the determination
of the Doppler temperatures in each plasma. Values ranged
from xenon (7 torr, 250 W) at 3700 K to both carbon dioxide
(2.8 torr, 250 W) and oxygen (6 torr, 550 W) at 2400 K. These
values are approximately 5 times that of HCLs or microwave-
EDLs, assuming a normal current or power input. To avoid
having the sample absorb on the cooler ends of the torch, the
length of the torch may have to be reduced so the tube length
fits entirely within the coil. Adding small quantities of a
desired species (e.g., typically in the 1-10 microgram range)
allows the torch to be used as a stable reservoir of atomic
lines. The same is possible for molecules stable in the
plasma environment.

39
The low pressure inert gas (He, Ne, Ar, Kr, Xe) plasmas
seeded with trace amounts of N, for diagnostic measurements
exhibited varying degrees of continuum emission in the near IR
and visible. Xenon is the most prevalent with approximately
15 prominent continuum emission bands. The signal to
background measurements of the emission bands ranged from 3 to
15 and are typically 2-3 nanometers wide. The bands are found
between 625 and 880 nm. They are semicircular in appearance.
Discussion
The analysis of atomic and molecular species stable in a
static plasma utilizes some of the capabilities of the UV/Vis
Fourier transform spectrometer (particularly the LAFTS). In
particular the intensity precision, wavenumber position
precision and accuracy, and resolution capabilities are
partially realized.
There is no reason to believe this cell could not
generate other plasmas. Atomic species which are gaseous at
low pressure (mercury, radon, volatile salts), diatomic gases
(NO, CO, Cl,, FI,, etc.), polyatomic gases (NO,, SO,, C,H,, other
hydrocarbons, etc.) should readily form low pressure plasmas
in this setup. The ethylene-ICP and other hydrocarbons
systems may be of interest to those studying the diamond
synthesis mechanism in a RF-plasma environment (66). Closed
cell hydrogen-ICP minimizes the dangers of working with a
flowing system. This hydrogen-ICP allows scientists to study

40
atomic interactions in a system similar to the surface of
cooler stars.
Fusing quartz windows to both ends of the plasma tube
allows a variety of absorption measurements to be conceived.
For example, intracavity absorption (67) with an ICP is
possible with the advantage of a longer path length, smaller
sample size, and an increased residence time per atom verses
the traditional flowing ICP torch.
A closed system ICP is currently being investigated as an
ion source for mass spectrometry. One end of the tube is
directly interfaced to the mass spectrometer with the flow
regulated by a gated valve. The sample is injected through the
syringe port (0.1 to 1 microliter) before ignition. The
solvent evaporation is enhanced with the vacuum leaving only
sample in the quartz tube. With plasma ignition, the sample
sputters into the plasma. We have found this to work quite
well for the nonmetals and softer metals. The variable gas
static plasma for ICP-MS should offer many advantages over the
flowing argon system. Expensive gases (e.g., neon, krypton,
xenon) can be routinely used due to low gas consumption.
These inert gases may be used to free mass ranges not
accessible with argon (e.g., sample overlaps with ArN+, ArH*,
ArOH+, etc.). Provided experimental conditions are identical
(power, pressure), the temperature and electron density of the
inert gas plasma (He, Ne, Ar, Kr, Xe) will have an inverse
relationship with the ionization potential. That is, the

41
higher the ionization potential is, the lower the temperature
and electron density will be. Thus by varying the inert gas,
the temperature and electron density range can be varied. The
closed cell allows for small sample sizes. It reduces some of
the hazards associated with radioactive samples in a flowing
system. The pressure differential between the plasma and the
mass spectrometer is reduced by at least two orders of
magnitude. Problems associated with the skimmer/orifice
degradation are reduced. The sample waste associated with a
flowing system is minimized. Nebulization is eliminated. In
some situations it may be possible to use nonpolar samples and
solvents. With evaporation/dehydration techniques, the
formation of hydrides, hydroxides, and oxides will hopefully
be reduced by applying this procedure. The low pressure cell
operated here can be converted to a fireball (68) at higher
powers. The fireball resembles the atmospheric pressure argon
ICP in appearance. This fireball could not be sustained for
extended periods of time or the quartz would glow red and
eventually melt. If a continuous vacuum is applied that
maintains a constant pressure, the fireball can be maintained
for extended periods of time.
Eckert (69) gives an excellent theoretical discussion of
elemental detection in a static induction plasma. His
calculations predict a limit of detection for magnesium (279.6
nm line) of less than 1015 g in a nonflowing argon-ICP. With
small modifications to the torch, the closed cell ICP may have

42
analytical usefulness. This analytical research would be
similiar to work previously carried out with microwave
electrodeless discharge lamps.
Conclusion
The static plasma outlined in this note describes an atom
and molecule reservoir for spectroscopic studies (70). It is
also being pursued as an ion source for mass spectrometry
(70) . No graphical data is presented to support the claim
that noise factors in the static plasma are improved over the
flowing system. We hope the reader follows the reasoning that
with the lack of turbulence induced by the gas flow, and no
nebulization system, many of the sources of noise inherent in
a flowing system are either reduced or eliminated in a static
plasma. The intensity variation as a function of time is at
least an order of magnitude improvement over the flowing
system under normal operating conditions.
With various adaptations to the cavity, the cell could be
used for a variety of emission, absorbance, and fluorescence
measurements, both analytical and physical in nature.

Figure 5. Closed cell ICP torch design.

44
TO
FTS

Figure 6. Line profile of Sodium-D (589.0 nm) in Xe-ICP.

INTENSITY
'$ 5 i? ?
o § § § §
G\

CHAPTER 5
OBSERVATION OF LINE SHIFTS AND LINE PROFILES
IN AN INDUCTIVELY COUPLED ARGON PLASMA
Introduction
Line shifts in analytical inductively coupled plasma's
(ICP) are a fundamental parameter that has not been studied in
any great detail. This paper will give an introduction to
line shifts measured and trends observed. For more detailed
treatments of pressure effects, including line shifts, the
reader is directed to Thorn (60) (overview), Alkemade et al.
(71) (in flames), Griem (72), and Breene (73) and references
within. Reviews by Konjevic and Weise (74) and by Konjevic
and Roberts (75) give numerous references on Stark broadening
and line shifts from a variety of sources.
Line shifts are important for a variety of reasons.
Analytical determination of elements in a complex sample is
often plagued misidentification of the species being studied.
In physical spectroscopy, ICPs are being used for the
measurement of transition probabilities. Knowledge of line
shift values are important to properly measure energy levels,
transitions, etc.. Whaling and Brault (76) used an ICP in
conjunction with the Kitt Peak Fourier Transform Spectrometer
to measure the transition probabilities of 1200 Mo (I) lines;
47

48
they introduce Mo as Mo(CO)6 eliminating any effects water has
on the spectrum. They observed line shifts of Mo as great as
0.14 cm1.
The Los Alamos Fourier Transform Spectrometer (LAFTS) is
an ideal instrument to measure small spectral shifts. Its
high resolution capabilities easily resolve physical line
widths emitted by hollow cathode lamps (HCL) (0.025 cm') and
the inductively coupled plasma, (0.07 cm1). The wavenumber
precision of the LAFTS can be as great as 1 part in 10 . The
LAFTS can acquire large spectral windows (9000 to 42000 cm')
at the required resolution to physically resolve the ICP
emission.
In studying line shifts, isotope shifts and hyperfine
structure must be considered to properly select a line.
Elements that clearly display either effect are avoided due to
the complexity introduced in attempting line profile studies.
The spectrum of a Pb emission line shows contributions from
■jnc ?07 on»
three isotopes, 82Pb" (24.1%), 82Pb" (22.1%), and 82Pb~ (52.4%).
These isotope shifts are shown in Fig. 7. This spectrum was
taken from a low pressure HCL with a Doppler temperature of
approximately 500 K, resulting in Doppler broadened spectral
lines having negligible pressure (collisional) broadening.
Increasing the temperature to 6500 K and introducing
substantial amounts of collisional broadening, which simulates
ICP conditions, will cause the three isotopic lines to appear
as one asymmetric, broadened line. Measuring the line shift

49
becomes difficult because the individual lines have to be
deconvoluted before spectral shift measurements are possible.
There will be uncertainty introduced into the calculation by
the deconvolution which may affect the validity of the shift
measurement.
Figure 8 shows the hyperfme structure of tantalum. 73Ta
is the predominant isotope occurring naturally (99.998%) so
isotopic effects are negligible. With a nuclear spin of 7/2,
73Ta clearly exhibits hyperfine structure. Kuhn (77) gives
an excellent treatment of hyperfine structure and isotope
shifts. Ta emission in an ICP has significant Doppler and
pressure broadening of the lines giving asymmetric line
profiles with 8 convoluted hyperfine elements making shift
measurements virtually impossible.
• • 40 .
Argon has one predominant isotope (18Ar ,99.60%) which has
no nuclear spin making it an ideal choice for line shift
studies. Figure 9 shows a line profile typical of Ar I
emission at 420 nm from the ICP. Iron has one dominant
isotope (26Fe56, 91.72%), which has no hyperfine structure;
Fig. 10 shows a line profile of Fe I at 259.9 nm. The effect
of the lower abundance isotopes of Fe on the line center
position (shift measurement) and the full width at half
maximum (FWHM, a-parameters; where a = 0.86* (X¡yXD) are
considered negligible for this study.
Kato et al. (78) measured profiles of 16 spectral lines
stemming from 8 elements emitted by an ICP and measured with

50
a pressure-scanning Fabry-Perot interferometer. The results
their study are similar in absolute magnitude of some lines
but occasionally differ in sign. They did not consider the
effects of hyperfine structure and/or isotope shifts as
carefully as we did. Boumans and Vrakking (79) published work
on high-resolution spectroscopy using an echelle spectrometer,
which refers to line shifts in the ICP in connection with the
wavelength calibration of the spectrometer.
Experimental
The Los Alamos Fourier Transform Spectrometer (LAFTS) is
utilized in conjunction with a 27.12 MHz RF Plasma Therm
(Cherry Hill, NJ) ICP. Papers by Manning et al. (64) and
Parsons and Palmer (13) give details of the LAFTS. For all
studies, the viewing height above the coils is 12 mm. The gas
flow rates are 12 L/min (outer), 0.8 L/min (intermediate), and
0.8 L/min (nebulizer). For ICP studies, the LAFTS was operated
at 0.07 cm1 resolution with a free spectral range of 33005.2
cm1. For the HCL studies, the LAFTS was operated at 0.02 cm1
resolution with a free spectral range of 40015.2 cm'1. The
detectors used were PMTs (1P28) or silicon pin diodes. The
silicon pin diodes were utilized for work involving Ar
emission in the near IR.
The output of the light source (ICP or HCL) is focused
with unity magnification on the FTS spectrometer aperture.
The f number of this spectrometer and the ICP/HCL optical

51
setups are matched. The output of a neon filled Al HCL is
impressed on the ICP emission using a neutral density filter
(0.2) before the focusing optics. Figure 11 shows an outline
of the entrance optics and the light sources. The Ne I line
at 585.2 nm from the HCL is used as a calibration standard for
all runs.
All emission lines used in this study (HCL and ICP) have
signal to noise ratios (SNR) greater than 100. This
requirement reduced the number of lines available between the
two sources. Some strong lines in the HCL were not always as
prevalent in the ICP and vice versa. Ionic lines for iron and
argon detected in the ICP were not observed in HCL emission or
had low signal-to-noise ratios, reducing the accuracy and
precision of the wavenumber measurement.
Six (Fe (1.1, 1.5, 1.9 kW) , Ar (1.1, 1.5, 1.9 Kw) )
Doppler temperatures were calculated from the ICP emission
line profiles using a Voigt fit. This Doppler temperature was
then used for all calculations involving those particular
conditions. For example, an Fe I line with a high SNR is
chosen from the 1.1 Kw ICP data. Voigt profiles are generated
until one is identical to the Fe I (ICP) profile. This
allowed us to estimate the contributions from Gaussian
(Doppler broadening) and Lorenztian (collisional broadening)
components. The Gaussian component is then used for all line
profile calculations involving Fe emission from the ICP at 1.1
kW. We measured the Gaussian and Lorentzian contributions for

52
one Ar and one Fe spectral line at each power level (1.1, 1.5,
1.9 kW) giving a total of six Doppler temperatures.
The line is used to estimate the electron density in
the analytical zone viewed by the LAFTS. These electron
densities applied only to the analytical zone? the region
where the Fe, Ba, Ca, and Sr emission predominantly occurs is
studied. Ar emission is not predominantly from the analytical
zone but from the outer tangential flow zone. No electron
density calculations were made for the latter zone.
Results and Discussion
The results are separated into three groups. First, Fe
and Ar emission lines in the ultraviolet and visible at three
separate power settings (1.1, 1.5, 1.9 kW) were measured.
Second, Ar emission in the near infrared and far visible was
measured. Third, prominent ion lines of Ca, Sr, and Ba at 1.1
kW were measured.
The first section refers to line shifts as a function of
power for 28 Fe lines and 21 Ar lines in the ICP. The power
dependence (1.1, 1.5, 1.9 Kw) is shown to increase Doppler
temperature, electron density, and shift. The iron data are
refer to the analytical zone of the ICP, whereas the argon
data refers to the outer tangential flow. For the argon and
iron data, the line position is given in Angstroms and
physical parameters (line shift, width, etc.) are given in
wavenumbers (cm') . Line positions are given in Angstroms

53
because the references used for transition assignments list
them in this manner. Most research measuring line shifts
report their results in cm'1.
Table 4 gives the Doppler temperature (6500 K),
transitions types, wavelengths, line shifts, Doppler and
Lorentzian contributions to the line profile, and the a-
parameter for 21 argon lines at 1.1 Kw. Transition
assignments for the Ar-term configurations are taken from Li
and Humphrey (80) . A single argon line at this power is fitted
with a Voigt profile to approximate the Doppler contribution
to the line profile. The resulting Doppler temperature of
6500 K is then used to calculate the Doppler half-width and a-
parameter for all argon lines at this power and the Lorenztian
contribution is calculated with an iterative method:
Avv- (1/2) * A v ¿ J ( (1/4) *Av¿ + Av2D)
where Avv is the total line width, AvL is the Lorentzian
component, and AvD is the Doppler component.
All line shift values given are shifted towards the
red(positive values). Negative values are blue shift. Tables
5 and 6 give similar data for Ar at 1.5 and 1.9 kW
respectively.
Figure 12 shows the relationship between , the average
distance of the electron from the nucleus and the average
shift for the 6p-4s, 5p-4s, and 4p-4s transitions of argon at

54
each of the three powers. The values for 6p, 5p, and 4p
levels are estimated using
= (l/2)a0[3n - 1(1+1)] (4)
where a0 is the Bohr radius, n is principle quantum number,
and 1 is the azimuthal quantum number.
In Tables 7,8, and 9, give the results for Fe at 1.1,
1.5, and 1.9 kW, respectively. For each table (power level),
a Doppler temperature and electron density are calculated.
The electron density was calculated using the line (486.1
nm). The transition configuration, wavelength of transition,
line shift, FWHM, Doppler and Lorentzian contributions to the
line profile, and the a-parameter for each line profile
studied have been measured. The iron concentration was 1000
Mg/ml for all runs. The Fe transition assignments were taken
from Fuhr, Martin, and Wiese (81).
In Table 10, the spectral characteristics of 25
additional argon lines obtained with the ICP (1.1 Kw) in the
near-IR and far visible are also given. A large number of
intense Ar lines emitted from the ICP and the HCL occur in
this region. It contains exclusively 4p-4s transitions.
In Table 11, the final set of data presents line shifts
of calcium, barium, and strontium in a 1.1 Kw ICP are given.
These lines are used to investigate the relevance of line
shifts in analytical spectroscopy. We consider each element

55
• • • 40 88
to have minimal isotope effects (20Ca (96.97%), 38Sr (82.56%),
138
56Ba (71.66%)) ; there are no hyperfine structures on the major
isotopes. The electron density and Doppler temperature are
considered to be the same as those in table 4 (Fe at 1.1 kW).
Emission from these ions are typically blue shifted.
Conclusion
The purpose of this chapter is to report the results of
measurements of shifts in line positions that are a function
of pressure broadening, in particular, Stark broadening. Fe
shows red shifts are on the order of 0.01-0.03 cm1. Line
shift studies on Ca, Sr, and Ba resulted in shifts of the same
absolute value as those of Fe but containing a number of blue
shifts. Line shifts for Ar in the visible and ultraviolet are
equivalent to or greater than those for those for Fe. The
shifts for 6p-4s transition is largest, followed by the 5p-4s
transitions and the 4p-4s transitions. We also show that line
shifts for Ar increase with power. Line shifts are shown to
increase as the average distance of the electron from the
nucleus radius increases provided the lower (ground) state is
the same for all three transitions.
From our measurements, line shifts should have little
consequence to the analytical spectroscopist. The data from
Ca, Ba, Sr, and Fe shows that the average shift is on the
order of 0.01-0.03 cm'1. Line positions of this accuracy are
not readily measurable by most analytical systems.

Figure 7. Onl^7one of Pb1s three isotopes has any hyperfine
structure (82Pb~7) . In the ICP, this isotope structure would
be convoluted into one asymmetric line profile making line
shift measurements of each line extremely difficult. (Top x-
axis is Á and the bottom is in wavenumbers, cm1) .

INTENSITY
ui
-j

181
Figure 8. The hyperfine structure of 73Ta taken from a low
pressure HCL. The higher pressure and temperature of the ICP
would make line selection difficult, (top x-axis is Á and the
bottom is cm )

59

Figure 9. Typical line profile of Ar I emission from the ICP.
(top x-axis is Á and the bottom is cm1)

INTENSITY
61

Figure 10. Line profile of Fe emission (1000 ppm) from
analytical zone in ICP. (Top axis is in Á and the bottom is
cm' )

INTENSITY
63

Figure 11. Diagram outlining the optical configuration of
light sources (ICP and HCL) and optical components outside
LAFTS.

65
NEUTRAL
DENS I TV
LENS
HCL
No I .
S T fiNDRRR
: !
I !
L!
i i
t i
AF E ;
URÍ

Figure 12. The relationship between ; the average atomic
radius, and the shift in the line position is demonstrated for
1.1 kW (*), 1.5 kW ( + ) , and 1.9 kW (A). The value is for
the upper level and should be multiplied by the constant aD
(Bohr radius) for the correct approximation.

LINE SHIFT (cm—1)
67

68
Table 4. The transition, wavelength of transition, X, line
shift, AAS , full width at half maximum (FWHM) , AAX, Doppler
halfwidth, AA.D, Lorentzian halfwidth, kXh, and the a- parameter
of Ar emission from a 1.1 kW ICP. The Doppler temperature is
6500K. Positive shift values are red-shifted. Below each
general transition (6p-4s, 5p-4s, 4p-4s) the average shift and
standard deviation is tabulated.
Transition
X
A A.q
AXt
aad
AAl
a
(A)
(cm ) (cm ) (cm )
(cm )
6P [ 1 • 5]2-4s [ 1.5]2
3555.
0.3446
0.6335
0.255
0.53
1.73
6p[O.5]0-4s[1.5]1
3607 .
0.3136
0.6385
0.251
0.54
1.7
AVERAGE SHIFT f6o-
-4s)
0.3291
(±0.0215
')
5p [0.5]j-4s[1.5]2
3950.
0.1242
0.4405
0.229
0.32
1.16
5p- [1.5]2-4s[1.5]1
4045.
0.1038
0.3972
0.233
0.26
0.93
5p [1-5]2-4s[1.5]2
4159.
0.1154
0.4270
0.218
0.31
1.18
5p[1.5]j—4s[1.5]2
4165.
0.1104
0.4385
0.218
0.33
1.26
5p 1 [0.5],-4s' [0.5]
o 4183.
0.1383
0.4451
0.217
0.34
1.30
5P [ 2.5]2-4s [ l. 5 ]2
4191.
0.0957
0.3788
0.216
0.25
0.96
5p' [1.5] [-4s' [0.5]
o 4192.
0.0984
0.3849
0.216
0.26
0.99
5p[O.5]0-4s[1.5],
4199.
0.1243
0.4011
0.216
0.28
1.08
5p [2.5]3-4s [ 1.5 ]2
4201.
0.1063
0.5264
0.216
0.44
1.69
5p [ 0.5 ] ,-4s [ 1.5]2
4252 .
0.0807
0.3280
0.213
0.19
0.74
5p [l*5]2-4s[1.5].
4267 .
0.1417
0.3746
0.212
0.25
0.97
5p [1*5] j-4s [ 1.5 ] j
4273 .
0.1046
0.3578
0.212
0.23
0.90
5p [2.5]2-4s[1.5]j
4301.
0.0976
0.3545
0.211
0.23
0.91
5p1 [1.5] 2-4s'[0.5]
, 4334.
0.0972
0.4147
0.209
0.31
1.23

69
Table 4--continued.
5p 1 [0.5] ,-4s ' [0.5],
4336.
0.0773
0.5441
0.209
0.46
1.83
5p ' [1.5] ,-4s ' [0.5],
4346.
0.1284
0.4080
0.209
0.30
1.19
5p[0.5]„-4s' [0.5],
4511.
0.1179
0.3613
0.201
0.25
1.03
AVERAGE SHIFT f5D-
4s)
0.1095
(±0.0182)
4p' [0.5],—4s[1.5]2
6967 .
0.0179
0.1698
0.130
0.07
0.44
4p' [1.5]2-4s[1.5]2
7069 .
0.0176
0.1610
0.128
0.06
0.39
AVERAGE SHIFT (4p-4s) 0.0177 (±0.0002)

70
Table 5. The transition type, wavelength of transition,!,
line shift, Als, full width at half maximum (FWHM),A1X, Doppler
halfwidth, A1D, Lorentzian halfwidth, AAL, and the a-parameter
of Ar emission from 1.5 kW ICP. The Doppler temperature is
6800 K. Positive shift values are red-shifted. Below each
general transition (6p-4s, 5p-4s, 4p-4s) the average shift and
standard deviation are tabulated.
Transition X AAS Alx AAD AAL a
(Á) (cm1) (cm1) (cm1) (cm'1)
6p[1.5]2-4s[1.5]2
3555
0.5442
0.7807
0.261
0.69
2.20
6p [ 0.5 ] 0-4s [ 1.5 ]!
3607
1.178
0.3333
0.257
0.13
0.42
AVERAGE SHIFT (6d-
-4s)
0.8610
(±0.4481)
5p[0.5]1-4s[1.5],
3950
0.1622
0.5703
0.235
0.47
1.66
5p 1 [ 1 • 5 ]2-4s [1.5],
4045
0.1383
0.6131
0.229
0.53
1.92
5P [1*5]2~4s [ 1.5]2
4159
0.1703
0.4494
0.223
0.34
1.27
5p[l. 5]4s [ 1.5]2
4165
0.1357
0.6745
0.222
0.60
2.24
5p' [0.5] ,-4s'[0.5]
o 4183
0.1715
0.5064
0.222
0.41
1.54
5P[2.5 ]2—4s [ 1.5],
4191
0.1700
0.4185
0.221
0.30
1.13
5p' [1.5] j-4s'[0.5]
o 4192
0.1578
0.4338
0.221
0.32
1.20
5p [ 0.5 ]0-4s [ 1.5 ],
4199
0.1970
0.5339
0.221
0.44
1.66
5p[2.5]3-4s[ 1.5]2
4201
0.1528
0.4406
0.221
0.33
1.24
5p [ 0.5],—4s [ 1.5 ]2
4252
0.0811
0.4219
0.218
0.31
1.18
5p[1.5]2-4s[1.5]1
4267
0.1708
0.4779
0.217
0.38
1.45
5p [ 1.5 ] j-4s [ 1.5] j
4273
0.1606
0.4202
0.217
0.31
1.19
5p [2.5] 2-4s [1.5],
4301
0.1562
0.5351
0.216
0.45
1.74
5p ' [ 1.5 ]2-4s ' [0.5]
, 4334
0.1219
0.8084
0.214
0.75
2.41

71
Table 5—continued.
5p ' [0.5] ,-4s 1 [0.5],
, 4336
0.1350
0.5421
0.213
0.44
1.77
5p 1 [1.5] j—4s 1 [0.5],
, 4346
0.1797
0.5264
0.213
0.44
1.71
5p [0.5] 0-4s ' [0.5],
4511
0.1979
0.4068
0.206
0.30
1.21
AVERAGE SHIFT (5d-
â– 4s)
0.1564
(±0.0283)
4p ' [0.5] ,-4s [ 1.5]2
6967
0.0204
0.1801
0.133
0.08
0.499
4p ' [ 1 • 5]2-4s [ 1.5]2
7069
0.0305
0.1801
0.131
0.09
0.571
AVERAGE SHIFT (4p-4s) 0.0254 (±0.0071)

72
Table 6.The transition type, wavelength of transition, l,line
shift, AAS, full width at half maximum (FWHM) , ^XT, Doppler
contribution, A1D, Lorentzian contribution, hXL, and the a-
parameter of Ar emission from a 1.9 kW ICP. The Doppler
temperature is 7100 K. Positive shift values are redshifted.
Below each general transition (6p-4s, 5p-4s, 4p-4s) the
average shift and standard deviation are tabulated.
Transition X i±Xs kXr A1D kXL a
(Á) (cm1) (cm1) (cm1) (cm1)
6p[1.5]2-4s[1.5]2
3555
0.6432 0.5962
0.266
0.48 1.49
6p[O.5]0-4s[1.5]1
3607
1.165 0.3068
0.263
0.08 0.25
AVERAGE SHIFT Í6D-
-4s)
0.9040 (±3680)
5p[0.5]1-4s[1.5]2
3950
0.2409
0.5762
0.239
0.48
1.66
5p 1 [ 1 • 5 ]2-4s [ 1.5],
4045
0.1869
0.5261
0.234
0.42
1.49
5p [ 1 • 5 )2—4s [ 1.5 ]2
4159
0.2064
0.5011
0.227
0.40
1.46
5p [ 1 ■ 5 ] j—4s [ 1.5 ]2
4165
0.1894
0.4608
0.227
0.35
1.28
5p ' [0.5J.-4S' [O.5]0
4183
0.2152
0.7554
0.226
0.69
2.53
5p [ 2.5 ]2-4s [ 1.5 ] 2
4191
0.1635
0.4865
0.226
0.38
1.40
5p 1 [ 1 • 5 ] ,-4s 1 [ 0.5 ]„
4192
0.1587
0.4689
0.226
0.36
1.32
5p[0.5]„-4s[1.5]1
4199
0.2129
0.8689
0.226
0.81
2.98
5p [ 2.5]3—4s [ 1.5]2
4201
0.1765
1.2200
0.225
1.18
4.35
5p [0.5],-4s[1.5]2
4252
0.1812
0.4081
0.223
0.29
1.08
5p [1.5]2-4s[1.5],
4267
0.2300
0.5035
0.222
0.40
1.50
5p[1.5]1-4s[1.5]1
4273
0.1630
0.5839
0.222
0.50
1.87
5p[2.5]2-4s[1.5]1
4301
0.1952
0.4673
0.220
0.36
1.36

73
Table 6—continued.
5p' [1.5]2-4s* [0.5],
, 4334
0.1699
0.5565
0.218
0.47
1.79
5p ' [0.5] !~4s ' [0.5],
, 4336
0.2405
0.7283
0.218
0.66
2.51
5p ' [1.5] ,-4s ' [0.5],
, 4346
0.1662
0.4559
0.218
0.35
1.33
5p[0.5] 0-4s' [0.5]!
4511
0.2759
0.6060
0.210
0.53
2.09
AVERAGE SHIFT (5x3-
•4s)
0.1983
(±0.0335)
4p' [0.5],-4s[1.5]2
6967
0.0261
0.1967
0.136
0.10
0.612
4p' [1.5]2-4s[1.5]2
7069
0.0261
0.2098
0.134
0.13
0.807
AVERAGE SHIFT Mp-4s) 0.0261 (±0.000)

74
Table 7. The transition type, wavelength of transition, X,
line shift, Als, full width at half maximum (FWHM) , A1T,
Doppler contribution, AA.D, Lorentzian contribution, £±Xh, and
the a-parameter of Fe emission from a 1.1 kW ICP. The Doppler
temperature is 6000 K and the electron density 1.20 x 101 cm .
Positive shift values are red-shifted. For this table, the
average line shift is 0.0185 cm' with a standard deviation of
0.0062 cm’ . The concentration of iron aspirated into the ICP
was 1000 /ng/ml.
Transition
X
(A)
AA.s
(cm )
A If
(cm )
aad
(cm )
AAl
/ 1 \ L
(cm )
a
a4D
-
4 _o
z F
ii
2755.73
0.0452
0.3394
0.274
0.12
0.364
a5D
-
5_o
y f
i
2966.90
0.0181
0.3200
0.254
0.12
0.392
a5D
-
5 _o
y d
i
2994.43
0.0220
0.3144
0.252
0.11
0.363
a5D
-
Cn
D
O
i
3021.07
0.0232
0.3091
0.249
0.11
0.366
a5D
-
5 _o
y d
i
3047.60
0.0201
0.3265
0.248
0.14
0.470
a5D
-
5 _ o
y d
i
3059.09
0.0172
0.3474
0.247
0.17
0.573
a5D
-
5_°
z P
i
3440.99
0.0154
0.2646
0.219
0.08
0.303
a5F
-
3 o
z G
i
3565.38
0.0190
0.2605
0.212
0.09
0.353
a5F
-
5 °
z G
i
3581.19
0.0194
0.2601
0.211
0.09
0.355
a5F
-
5 °
z G
i
3608.86
0.0168
0.2599
0.209
0.09
0.357
a5F
-
5 o
z G
i
3618.77
0.0199
0.2849
0.208
0.13
0.518
a5F
-
5 o
z G
i
3631.46
0.0175
0.2555
0.208
0.09
0.360
a5F
-
5 o
z G
i
3647.84
0.0183
0.2784
0.207
0.12
0.482
a5D
-
5 _o
z F
i
3719.93
0.0127
0.2447
0.203
0.08
0.328
a5D
-
5_°
z F
i
3745.56
0.0112
0.2470
0.201
0.08
0.330
a5D
-
5 _o
z F
i
3748.26
0.0119
0.2472
0.201
0.08
0.330
a5F
-
5 _o
y f
i
3758.23
0.0187
0.2466
0.201
0.08
0.331

75
Table 7—continued.
a5F
-
5 _o
y f
i
3763.79
0.0179
a3F
-
CO
D
o
i
3815.84
0.0221
a5F
-
5 o
y d
i
3820.43
0.0184
a5F
-
5 °
y d
i
3825.88
0.0194
a5F
-
5 o
y d
i
3834.22
0.0165
a5D
-
5_°
z D
i
3859.91
0.0122
a5D
-
5 o
z D
i
3886.28
0.0113
a3F
-
3„o
y f
i
4045.82
0.0192
a3F
-
3_o
y f
i
4063.60
0.0222
a3F
-
5 o
Z G
i
4383.55
0.0148
a3F
5 o
Z G
i
4404.75
0.0165
0.2460
0.200
0.08
0.332
0.2477
0.197
0.09
0.378
0.2574
0.197
0.11
0.463
0.2439
0.197
0.09
0.374
0.2437
0.197
0.09
0.380
0.2368
0.196
0.08
0.340
0.2324
0.194
0.07
0.299
0.3011
0.186
0.19
0.847
0.2263
0.186
0.08
0.358
0.2173
0.172
0.08
0.386
0.2151
0.171
0.08
0.388

76
Table 8. The transition type, wavelength of transition, A,
line shift, AAS, full width at half maximum (FWHM) , AAT,
Doppler, AAD, Lorentzian , AAL, and the a-parameter of Fe
emission from a 1.5 kW ICP. The Doppler temperature is 6200
K and the electron density is 2.05 x 1015 cm'J. Positive shift
values are red-shifted. The average shift value for this
table is 0.0188 cm1 with a standard deviation of 0.0066 cm'1.
1000
¿xg/ml
Fe
solution
was aspirated into the
ICP.
Transition
A
A*s
A At
AAd
AAl
a
(Á)
(cm‘)
(cm1)
(cm1)
(cm1)
a4D
-
4-r-i°
z F
II
2755.73
0.0431
0.3615
0.273
0.15
0.455
a5D
-
5_°
y f
I
2966.90
0.0173
0.3290
0.254
0.13
0.425
a5D
-
5_o
y d
I
2994.43
0.0255
0.3349
0.252
0.14
0.462
a5D
-
5 o
y d
I
3021.07
0.0148
0.3115
0.250
0.11
0.366
a5D
-
5_o
y d
I
3047.60
0.0277
0.3251
0.247
0.14
0.470
a5D
-
5 o
y d
I
3059.09
0.0124
0.3262
0.247
0.14
0.472
a5D
-
5 _o
z P
I
3440.99
0.0214
0.2827
0.219
0.11
0.417
a5F
-
3 °
z G
I
3565.38
0.0218
0.2766
0.212
0.10
0.384
a5F
-
5 o
Z G
I
3581.19
0.0209
0.2731
0.211
0.11
0.434
a5F
-
5 o
z G
I
3608.86
0.0199
0.2716
0.209
0.11
0.437
a5F
-
5 o
z G
I
3618.77
0.0198
0.2689
0.209
0.11
0.439
a5F
-
5 °
z G
I
3631.46
0.0188
0.2786
0.208
0.12
0.480
a5F
-
5 o
z G
I
3647.84
0.0160
0.2639
0.207
0.10
0.402
asD
-
5 o
z F
I
3719.93
0.0134
0.2963
0.203
0.16
0.656
a5D
-
5 _o
z F
I
3745.56
0.0146
0.2497
0.202
0.09
0.371
a5D
-
5 o
z F
I
3748.26
0.0097
0.2575
0.201
0.10
0.431
a5F
-
5 _o
y f
I
3758.23
0.0232
0.3186
0.201
0.19
0.787

77
Table 8—continued.
a5F
-
5 °
y f
i
3763.79
0.0132
a3F
-
3^0
y d
i
3815.84
0.0205
a5F
-
5 °
y d
i
3820.43
0.0181
a5F
-
5 o
y d
i
3825.88
0.0191
a5F
-
Oi
a
o
i
3834.22
0.0196
a6D
-
5 o
Z D
i
3859.91
0.0108
a5D
-
5 o
z D
i
3886.28
0.0231
a3F
-
3t,o
y f
i
4045.82
0.0137
a3F
-
3_o
y f
i
4063.60
0.0115
a3F
-
5 o
z G
i
4383.55
0.0146
a3F
5 o
z G
i
4404.75
0.0211
0.2509
0.201
0.09
0.373
0.2625
0.198
0.11
0.462
0.2820
0.197
0.14
0.589
0.2542
0.197
0.10
0.422
0.2476
0.197
0.09
0.380
0.2429
0.195
0.09
0.382
0.2651
0.194
0.12
0.514
0.2445
0.186
0.10
0.446
0.2707
0.186
0.14
0.627
0.2208
0.172
0.09
0.434
0.2529
0.171
0.14
0.679

78
Table 9. The transition type, wavelength of transition,!,
line shift, AAS, full width at half maximum (FWHM) , AAT,
Doppler contribution, AAD, Lorentzian contribution, AAL, and
the a-parameter of Fe emission from a 1.9 kW ICP. The Doppler
temperature is 6500 K and the electron density is 2.85 x 10J
cm3. Positive shift values are red-shifted. The average shift
for all lines listed in this table is 0.0176 cm' with a
standard deviation of 0.0132 cm'1. The concentration of Fe
aspirated into the ICP was 1000 /¿g/ml.
Transition
A
(A)
A Aa
/ ^ \
(cm )
A A-p
(cm ‘)
aad
(cm ) (cm
AAl
')
a
a4D
-
4_o
z F
ii
2755.73
0.0390
0.5696
0.284
0.42
1.23
a5D
-
5 o
y f
i
2966.90
0.0249
0.3240
0.264
0.11
0.36
a5D
-
5 o
y d
i
2994.43
0.0544
0.4185
0.262
0.25
0.79
a6D
-
5 °
y d
i
3021.07
0.0116
0.3210
0.259
0.11
0.35
asD
-
5 o
y d
i
3047.60
0.0226
0.3560
0.257
0.17
0.55
a5D
-
5 °
y d
i
3059.09
0.0157
0.3155
0.256
0.11
0.35
a5D
-
5 o
z P
i
3440.99
-0.0109
0.3246
0.228
0.16
0.58
a5F
-
3_0
Z G
i
3565.38
0.0305
0.2661
0.220
0.09
0.34
a5F
-
5 o
z G
i
3581.19
0.0221
0.2872
0.219
0.12
0.45
a5F
-
5 °
z G
i
3608.86
0.0188
0.2592
0.217
0.08
0.31
a5F
-
5 o
z G
i
3618.77
0.0105
0.2641
0.217
0.09
0.34
a5F
-
5 °
z G
i
3631.46
0.0386
0.2796
0.216
0.11
0.42
a5F
-
5 o
z G
i
3647.84
0.0101
0.2652
0.215
0.09
0.35
a5D
-
5 o
z F
i
3719.93
0.0164
0.2617
0.211
0.09
0.35
a5D
-
5 o
z F
i
3745.56
0.0210
0.3310
0.209
0.20
0.79
a6D
-
5_°
z F
i
3748.26
0.0140
0.2507
0.209
0.08
0.32
a5F
-
5 o
y f
i
3758.23
0.0188
0.3103
0.207
0.17
0.68
a5F
—
5 o
y f
i
3763.79
0.0193
0.2912
0.208
0.14
0.56

79
Table 9—continued.
a3F
3 — °
- y D
I
3815.84
0.0135
a5F
5 °
- yD
I
3820.43
0.0176
a5F
5_o
- y D
I
3825.88
0.0152
a5F
5 °
- y D
I
3834.22
0.0266
a5D
5 o
- z D
I
3859.91
0.0089
a5D
5 o
- Z D
I
3886.28
0.0081
a3F
3 o
- yF
I
4045.82
0.0118
a3F
3 o
- yF
I
4063.60
0.0076
a3F
5 °
- Z G
I
4383.55
0.0165
a3F
5 o
- z G
I
4404.75
-0.0102
0.3395
0.205
0.21
0.85
0.2575
0.205
0.09
0.36
0.2599
0.205
0.10
0.41
0.4480
0.204
0.35
1.42
0.2895
0.203
0.15
0.61
0.2760
0.202
0.13
0.53
0.2508
0.194
0.10
0.43
0.2509
0.193
0.10
0.43
0.2426
0.178
0.11
0.51
0.2338
0.178
0.10
0.47

80
Table 10. The transition type, wavelength of transition, X,
line shift, Als, full width at half maximum (FWHM) , ^XTI
Doppler contribution, kXDr Lorentzian contribution, ^XLI and
the a-parameter of Ar emission from a 1.1 kW ICP. The Doppler
temperature is 6500 K. Positive shift values are red-shifted.
Transition
X
(A)
Ale
(cm )
AA^ A AD AA
(cm1) (cm1) (cm
*5
a
4p[0.5],-4s' [0.5]0
10472.
0.0159
0.1365
0.0865
0.08
0.77
4p[2.5]2-4s' [0.5],
9787.
0.0172
0.1934
0.0926
0.15
1.35
4p[0.5]1-4s[1.5]1
9660.
0.0121
0.1389
0.0938
0.08
0.71
4p [1.5] ,-4s 1 [0.5]!
9356
0.0050
0.1963
0.0969
0.15
1.29
4p[l»5]2—4s' [0.5]j
9227 .
0.0189
0.1999
0.0983
0.15
1.27
4p[ 0.5] i~4s [1.5],
9125.
0.0042
0.1419
0.0994
0.07
0.59
4p[1.5],-4s' [O.5]0
8670.
0.0268
0.1506
0.1046
0.08
0.64
4p 1 [1.5] [-4s ' [0.5],
8523 .
0.0151
0.2092
0.1064
0.16
1.25
4p[2.5]2—4s[1.5]j
8427 .
0.0016
0.1677
0.1076
0.10
0.77
4p' [1.5],-4s' [0.5],
8410.
0.0164
0.2250
0.1078
0.18
1.39
4p ' [ 0.5 ] ,-4s ' [0.5],
8266.
0.0183
0.2125
0.1097
0.16
1.21
4p[2.5]3—4s[1.5]2
8117.
-0.011
0.1833
0.1117
0.12
0.89
4p[1.5],-4s[1.5]1
8105.
0.0217
0.1686
0.1118
0.10
0.74
4p[2.5]z—4s[1.5]2
8016.
0.0216
0.1602
0.1131
0.08
0.59
4p [ 1 • 5 ]2-4s [1.5],
8008.
0.0248
0.1652
0.1132
0.09
0.66
4p ' [ 1 • 5 ] ,-4s ' [ 0.5 ] 0
7950.
0.0254
0.1610
0.1140
0.08
0.58
4p ' [0.5] ,-4s ' [ 0.5 ]0
7726.
0.0269
0.1658
0.1173
0.08
0.57
4p [ 1.5 ] ,-4 s [ 1.5 ] 2
7725.
0.0230
0.1710
0.1174
0.09
0.64
4p[1.5],-4s[1.5]2
7637 .
0.0463
0.1683
0.1187
0.09
0.63
4p[O.5]0-4s[1.5],
7515.
0.0280
0.1851
0.1206
0.11
0.79

81
Table 10—continued.
4p'
O
in
•
o
-4s' [0.5],
7505.
0.0301
0.2297
0.1208
0.17
1.17
4p 1
[ 1 • 5 ] 2
-4s[1.5],
7386.
0.0224
0.1785
0.1227
0.10
0.65
4p '
[0.5],
-4s[1.5],
7274 .
0.0255
0.1806
0.1246
0.10
0.68
4p'
[ 1 • 5 ]2'
-4s [ 1.5 ],
7069.
0.0241
0.1756
0.1301
0.09
0.57
4p' [0.5],
-4s [ 1.5 ]2
6967 .
0.0260
0.1783
0.1301
0.09
0.57

82
Table 11. Line shifts, Av, of Ca, Sr, and Ba in 1.1 Kw ICP.
The wavenumber positions, v, for hollow cathode lamp
(standard, nonshifted) verses the ICP emission. The Doppler
temperature is 6000 K and the electron density is 1.20 x lo‘°
cm3 for the ICP. Wavenumber positions given are calibrated
against a Ne I standard for absolute values. Negative values
are blue-shifted.
ELEMENT
v (HCL) v (ICP) Avs
(cm1) (cm1) (cm1)
Barium II
Barium II
Calcium II
Calcium II
Strontium II
20261.5370
21952.3718
25191.5197
25414.4307
24516.5792
20261.5479
21952.3975
25191.5241
25414.4147
24516.5870
-0.0109
-0.0257
-0.0044
0.0160
-0.0078

CHAPTER 6
A VARIABLE BANDPASS FILTER FOR ULTRAVIOLET/VISIBLE
FOURIER TRANSFORM SPECTROSCOPY
Introduction
Fourier Transform Spectroscopy (FTS) has several
advantages over grating and prism spectrometers. Included in
these are wavenumber precision and accuracy, intensity
precision, and resolving power. Detrimental to the FT
technique is the multiplex disadvantage (7,82), which arises
from shot and/or flicker noises at detected wavelengths other
than the analytical wavelength hitting the detector causing a
decreased signal-to-noise ratio compared to a single slit
scanning dispersive spectrometer system. This disadvantage is
more prominent in FT-UV/VIS then in FT-IR studies. In FT
studies involving a continuum source, this disadvantage can
become severe. The Inductively Coupled Plasma (ICP), Direct
Current Plasma (DCP), and absorbance techniques utilizing a
white light are examples of light sources that can magnify the
multiplex disadvantage.
In our research (83), the Los Alamos Fourier Transform
Spectrometer is used to measure Stark shifts in wavenumber
positions of atoms and ions from a 27.12 MHz atmospheric
pressure argon Inductively Coupled Plasma (ICP). The line
83

84
positions of atoms and ions in the ICP are compared to the
unshifted line positions of the same species in a commercial
hollow cathode lamp (HCL). To achieve adequate wavenumber
positions (1 part in 4x10 ), the lines analyzed need good
signal-to-noise ratios (SNRs) . Both the HCL and ICP emit
numerous spectral lines. The ICP, being a hotter source, has
considerable ion line emission. The HCL, with Doppler
temperatures typically 1/10 that of the ICP (6000 K vs 600 K)
has considerable atom line emission, particularly the fill gas
emission in the near infrared. In such measurements of Stark
shifted line positions, we find that the data set available is
small for two reasons. First, the difference in source
emission intensities between the same lines emitted from the
ICP and the HCL reduces the number of lines available for
comparison. Second, the multiplex disadvantage will
substantially reduce the SNR of all lines.
A single filter with two features is desired. First,
the ability to vary the bandpass of the filter over a wide
range. Second, the ability to select the wavenumber region
observed over a wide range. Optical filters do not meet our
specifications for two reasons; the bandpass allowed rarely
matches the needs desired, and the transmission of these
optical filters often increases in the near IR resulting in
excessive shot noise from spectral lines/bands in this region.
Our sources (HCL, ICP) use inert gases having strong emission
in the near IR region, resulting in a serious deterioration

85
of the SNR for lines observed in the ultraviolet and visible.
Hirschfeld and Chase (84) and Hirschfeld and Milanovich
(85) summarized several techniques that could be used to
overcome problems associated with excess light in FT-Raman
experiments. One of these is the use of a nonscanning
subtractive dispersive double monochromator. This filtering
approach has been shown to be a valuable tool for micro-Raman
work (85) . We have adapted this filter for use as a bandpass
filter in FT-continuum source work. We demonstrate the
feasibility of this filter using a white light source.
Experimental
Figure 13 shows the configuration of the experiment. A
Bomem DA-3 is used at low resolution (30 cm1) for this study.
The gratings (Gl, G2) are manufactured by Instrumental SA.
Both are blazed for 700 nm (1200 grooves/mm, 25 mm x 25 mm).
F numbers of the input lens (L3) and FTS are matched. The
white light source (W.L.), gratings, lens-1 (LI) and lens-3
(L3) are held constant for all experiments. The mirrors
(Ml,M2), lens-2 (L2) aperture-2 (A2) are varied for selective
purposes. LI serves to collimate the light source. A1 (5 mm
diameter) and A3 (3-10 mm diameter) are circular apertures.
For Al, we use an aperture as opposed to a slit to maintain
some of the throughput advantage of the FTS for emission
studies. Since we are looking at relatively large bandpasses
(> 500 cm1) , the loss in spectral resolution of an aperture

86
verses a slit is not significant. G1 and G2 are mounted back
to back. A2 serves to block unwanted light and to let through
a preset band. A2 is rectangular; A2 and L2 are moved
laterally to select both wavenumber region and bandpass. Ml
and M2 are matched concave mirrors, where Ml transfer the
range of nearly collimated radiation selected by A2-L2 and M2
transfers that range of nearly collimated light on to the
surface of G2 which further disperses and transfers the band
of nearly collimated radiation into L3-A3 where A3 is the
aperture to the FTS.
The bandpass can be adjusted by changing the diameter of
A2 and/or the distance of A2 from Gl. Narrower bandpasses can
be achieved with smaller apertures (Al, A2) at greater
distances (A2 from Gl) . Decreasing the apertures (Al, A2)
decreased the photon flux entering the FTS and therefore
decreases the multiplex disadvantage. Moving A2-L2
perpendicular to the nearly collimated light reaching Ml
resulted in a change in the central wavelength of the band
passed.
The distances between optical components varied slightly
depending on central wavenumber and bandpass selection.
Approximate dimensions are as follows: Ll-Al, 1 cm; Al-Grating
1, 11 cm; Grating 1-A2, 12 cm; M1-M2, 26 cm; Grating 2-L3, 12
cm.

87
Results
Figure 14 shows two spectra (25 scans) of the filtered
white light source with a 2000 cm'1 bandpass. The strong
narrow line in both is the He-Ne (6328 A) laser leakage and is
used as a visual reference for alignment of the optics Al-Gl-
A2-L2-M1-M2-G2-L3-A3. For this demonstration the central
wavenumber shifts from approximately 15000 cm'1 to 16000 cm'1.
Figure 15 shows a narrower bandpass (approximately 1000
cm1) that varied from 16500 to 18700 cm'1 in four windows. The
gratings performance fell off above 20000 cm'1 and visual
alignment of the optics in Figure 13 below 13000 cm'1 becomes
difficult by eye; an infrared sensor is needed to see the
radiation. This justified the range (13000-19000 cm'1)
demonstrated here.
Conclusion
A bandpass filter which varied both wavenumber position
and bandpass and provide a limited spectral range is
demonstrated. The bandpass filter is being developed the
filter for FT-UV/VIS source studies. The filter removed
unwanted light which affected the quality of broad band
multiple line spectra. Removal of the prominent lines of neon
or argon appearing in the 600-900 nm range in hollow cathode
lamp (HCL) emission is also of interest to us. These lines
are often the most intense in HCL emission studies. Optical
filters have two disadvantages which often arise in studies of

88
this sort. First the bandpass of the optical filter may not
match the desired need, especially in the ultraviolet.
Second, transmission of these filters often increases in the
NIR (above 900 nm) and allows the fill gas emission
(particularly Ar) to enter the FTS. Our aim is to replace
optical filters with a simple filter that allows variation in
both position and bandpass of the filter.
We should note that a variety of derivations of this
filter are possible. For example, replacing Ml and M2 with
parabolic mirrors would allow the removal of L2. Ml and M2
would collimate and reflect the radiation to G2 eliminating
the need for L2.

Figure 13. Outline of bandpass filter. (W.L. = white light,
A = aperture, G = grating)

90
Mirror 2
\"
Lens 2
A
-
White
Light
V
Lens 1

Figure 14. 2000 cm1 bandpass of filtered white light source
centered at 15000 cm'1 (top) and 16000 cm' (bottom) . Sharp line
(6328 A) is He-Ne laser leakage from Bomem FTS alignment
system.

INTENSITY
92
WAVENUMBER

Figure 15.
173 00 cm'1 (
1000 cm' bandpass centered at 16600 cm' (1),
2), 17800 cm1 (3), 18750 cm'1 (4).

INTENSITY
vo

CHAPTER 7
A PROPOSED WAVENUMBER, INTENSITY, AND RESOLUTION
STANDARD FOR HIGH RESOLUTION UV/VIS FOURIER
TRANSFORM SPECTROSCOPY
Introduction
In this chapter, using Re I (527.55 nm) as a standard for
measuring the quality of spectra obtained with a Fourier
Transform Spectrometer (FTS) operating in the ultraviolet and
visible is proposed. In addition to the multiplex and
throughput advantages, three additional advantages for FT-
UV/VIS are resolution, intensity precision and accuracy, and
wavenumber precision and accuracy. The Re I (527.55 nm)
line, which has hyperfine structure (HFS) and isotope shifts
(IS) , offers an excellent reference standard for measuring the
above mentioned qualities. Quite often FT-UV/VIS instruments
claim achieving resolutions (Av, cm1) based simply on mirror
displacement (L, cm) where
Av = 1/(2L) (5)
This relationship does not consider factors such as electronic
noise, phase problems, incorrect aperture setting, optical
imperfections and mirror jitter which negatively affect the
resolution claimed by mirror displacement. Imperfections of
95

96
this nature can also affect wavenumber precision and accuracy
and intensity precision and accuracy.
Experimental
The output of a commercially available hollow cathode
lamp (HCL) is used as the standard. An optical filter
recently described (86) is used to narrow the bandpass and
minimize the multiplex disadvantage. The relatively low
pressure, low temperature environment of the HCL served to
minimize collisional and Doppler broadening respectively,
making the details of the transition visible. The Los Alamos
Fourier Transform Spectrometer (64,13) is run at high
resolution for this particular study. The spectra presented
here is acguired at 0.026 cm'1 resolution with 2 scans.
Discussion
Rhenium
Spectroscopists have studied the hyperfine structure and
isotope shifts of Re I (87,88,89). Re has two naturally
occurring isotopes, l85Re, (37.07 %) and 187Re (62.93 %) . Each
of the two isotopes have nuclear spins greater than zero (5/2,
5/2) which led to hyperfine structure. Both have similar
nuclear magnetic moments (3.172, 3.204 nuclear magnetons),
respectively. The clarity and symmetry of the HFS and IS in
this transition make it an excellent selection as a standard.
The line profile of Re I (527.55 nm) is shown in Figure 16.

97
The wavenumbers for this spectrum are listed in Figure 16.
Resolution
The high resolution capabilities of an FTS can be readily
tested by attempting to separate the isotope shifts for each
of the hyperfine components. These isotope splittings are in
the 0.06-0.07 cm' range and serve as a good standard. The
splittings between the hyperfine components are in the 0.2 to
0.6 cm' range and serve as a measure for lower resolutions.
The full width at half maximum (FWHM) of the individual lines
in this study are 0.03-0.04 cm1.
Intensity
The intensity precision and accuracy of the Fourier
transform spectrometer can be measured using the relative
intensities of Re I isotopes. The ratio (37.07% : 62.93%) of
the two naturally occurring isotopes serve as the standard for
intensity precision and accuracy. The relative intensity (I)
of the individual isotope shifts is directly related to the
percent natural abundance.
(I185/I187) = (37.07%/62.93%) (6)
The isotope shifts are separated by only 0.06-0.07 cm':
therefore high resolution capabilities (< 0.05 cm') are
needed. For the hollow cathode lamp, the intensity varies 1-2%
as a function of time, and so errors in this range are
considered acceptable. Sources with smaller signal variations

98
will produce smaller errors. For the two most intense lines
(18950.7397 cm'1, 18950.1715 cm') and the accompanying shifts
for the less abundant isotope (18950.7994 cm'1, 18950.2356 cm
‘), there is an acceptable systematic error of 1%. The LAFTS
has a 0.1% intensity accuracy limit (13).
Wavenumber Positions
Spectroscopists routinely use Fourier Transform
Spectrometers for high wavenumber precision and accuracy.
Several studies (64,83,90) have taken advantage of the high
wavenumber accuracy of the LAFTS. This accuracy has typically
been greater than 1 part in 10?. There are two correction
factors (8) incorporated in these types of studies: a)
aperture correction b) slight misalignment of instrument.
They adhere to the following format:
a' = a [ 1 + (n/47r) + xA ] (7)
where n = 7rD2/4F2 = solid angle (sr)
D = diameter of the aperture (cm)
F = focal length of collimating
optics (cm)
a = measured wavenumber (cm1)
ct'= corrected wavenumber (cm1)
X = instrumental calibration (cm')
e = average wavenumber (cm1)

99
In a previous study (90), the instrumental calibration
(X) was calculated in the following manner. The emission from
a uranium HCL was studied with the LAFTS. Twenty five lines
were measured in the 385-395 nm range and compared against
very accurate (1 part in 10s) wavenumber values from a
uranium atlas (91). With no correction factors, the values
were red-shifted 0.0052 cm1. With only the aperture
correction (fi/47r) applied, this left an average red-shift of
0.0015 cm' with a standard deviation of 0.0013 cm’1. This
value (0.0015 cm') is the instrumental calibration (x)• The
average wavenumber, e, is simply the average of the
wavenumbers being studied. For the LAFTS, the misalignment is
typically smaller than the aperture correction factor. The
correction factors combined are small (roughly 2.2 x 10 7) but
must be considered for high wavenumber accuracy and precision.
Table 1 lists the corrected values for the Re I lines. These
values are considered accurate to better than 1 part in 10?.
Conclusion
An intensity, wavenumber, and resolution standard for
Fourier Transform Spectroscopy in the visible is presented.
It is chosen to evaluate the spectrometer's performance
concerning resolution, intensity precision and accuracy, and
wavenumber precision and accuracy. This Re I transition
provides narrow, intense lines that are in an easily
identifiable pattern. Re HCL's are commercially available and

100
easy to use in conjunction with a FTS. Obviously, many other
standards could be chosen for various regions of the spectra.
This standard is intense, offers symmetry in HFS and IS, and
will clearly tell if the FTS is performing up to claimed
specifications.

Figure 16. Re I line profile clearly shows hyperfine
structure (HFS) and isotope shifts (IS). The corrected
wavenumbers for Re I (18950 cm) transition are: 18950.8035
( Re), 18950.7438 ( Re), 18950.2397 ( Re), 18950.1756 (Re),
18949.8044 ( Re), 18949.7384 (Re), 18949.4877 (Re),
18949.4189 ( Re), 18949.3484 (‘“Re), 18949.2811 (’V) .

INTENSITY
102

CHAPTER 8
UNDERSTANDING NOISE IN FOURIER TRANSFORM SPECTROSCOPY
Introduction
Various aspects of noise studies in Fourier Transform
Spectroscopy have been reviewed by in several books, including
Griffths and deHasth (92), Chamberlain (93), Bell (94), and
Bracewell (95). Theoretical modeling and experimental work
involving noise in FT-UV/VIS consist of work by Horlick and
coworkers (1,4,96), research at Imperial College (97-99), work
by Jim Brault of Kitt Peak National Observatory (100,101),
Voigtman and Winefordner (102), Hirschfeld (103-105), Faires
(2,106,107), Sakai and Murphy (107), Natale and Ventura (108),
Knacke (109), Luc and Gerstenkorn (110), and Emonds
dissertation from the Orsay (France) FT facility (ill). A
complete survey of the literature would envelope many more
papers on various aspects of FTS.
In this paper, a simple system that crudely mimics a
Michelson Interferometer analyzing monochromatic and
dichromatic light for noise studies is described. It is
critical to understand and characterize noise sources in
ultraviolet and visible FTS, whether they be flicker (1/f),
white noise, or interference noise (60 Hz, etc), if they are
to be minimized or eliminated. In infrared FTS (FT-IR), the
103

104
system is typically detector noise limited. In FT-UV/VIS, the
detectors (typically photomultiplier tubes or Si-PIN diodes)
are not the limiting noise source. Commonly used sources in
atomic emission spectroscopy such as flames, plasmas, and
discharges, are source noise limited. For a FTS to show a SNR
gain over a dispersive type system, the FT experiment would
have to be operating near or at the detector noise. If shot
or flicker noises are limiting, the SNR comparison becomes
complex.
In this system, the noise characteristics of any light
source can be studied. Chopping the source at some frequency,
f¡ (Hz), equals the rate of constructive and destructive
interference, R¡ (Hz), which is related to the wavelength
studied (A.¡, cm) and the mirror velocity (v¡, cm/s) by the
relationship
f. = R. = v/A-i (8)
The same detector used in the FTS can be retained. Chopping
at a set frequency allows us to study the magnitude and type
of noise in that region. The effects of low frequency noises
(eg. 1/f, 60 Hz) wrapping around the desired line profile,
external sources producing interferences, etc. can be
characterized. For example, a Cu vapor laser used in our lab
fires at 6 kHz. This produces a strong signal at 6 kHz and
its respective harmonics (12, 18, 24, kHz, etc.). Research

105
buildings typically produce a wide range of interference
noises that can be misidentified if not correctly identified.
Experimental
The experimental setup used in this study is shown in
Figure 17. A Na (589.0 nm) HCL and a Helium Neon laser are
the sources employed for demonstration purposes. The setup
has been used in conjunction with many sources, including
HCL's, diode lasers, and various white light sources. Figure
17 shows the laser light split and the geometric arrangement
which produces frequencies corresponding to different
wavelength's in a Michelson. A photomultiplier tube (1P28, -
760 V) is the detector.
Neutral density filters are used as needed to attenuate
the light flux entering the monochromator. The beam splitters
and mirrors used are optimized for the far visible and near
infrared studies. Mechanical choppers employed here have
maximum chopping rates of 680 and 1100 Hz. Because mirror
velocities for a typical Michelson interferometer range from
0.01 to 4 cm/s, the wavelength range considered (near IR,
visible, UV) and reasonable mirror velocities (v,) , the
maximum rates of constructive and destructive interference
(Rlf 680 and 1100 Hz) are slow but possible. An extreme
range for FT-UV/VIS/NIR modulation rates is 100 HZ (lOOOnm,
0.01 cm/s) to 200,000 Hz (200 nm, 4 cm/s) with 400-20,000 Hz
being reasonable.

106
The 0.25 m monochromator was used to minimize/eliminate
the effects of unwanted light. The output of the PMT was sent
to a Wavetek 5820B Cross Channel Spectrum Analyzer. This data
processor uses a Fast Fourier Transform (FFT) and displays the
• • • 1/2 •
resulting Noise Power Spectra in dBV/(Hz) (x-axis) verses Hz
(y-axis). The Wavetek has a maximum bandwidth of 50 kHz
(0-50 kHz). 64 scans were coadded for all spectra presented.
Results
Figure 18a shows the transform of the PMT dark current
(no light striking it) . For UV/VIS/NIR FTS the region of
interest lies in the 500-20000 Hz range. The exact origin of
the noises at approximately 14 and 41 kHz are not known but
appear to be from an external (not PMT) source. The PMT
voltage was varied from 400-800 V, but the magnitude of the
noises (14, 41 kHz) remained constant. Figure 18b shows the
noise power spectra of the PMT output of unchopped Na HCL
(589.0 nm) emission. It is important to notice the magnitude
of the y-axis (dBV/(Hz) ) in relationship to 18a. The noise
level has been raised substantially. Noise spikes at 14 and
41 kHz that were prominent in 18a are less significant because
of the rising background. Two flicker noise components are
obvious. The low frequency (0-10 Hz) flicker noise component
is in the 0-10 Hz region. The higher frequency flicker noise
component decreases at a slower rate from 0 Hz to 50 kHz, the
upper frequency limit of the Wavetek instrument. The drop

107
off of the high frequency component is not as dramatic as the
low frequency (0-10 Hz) component but can be precisely
measured. Figure 18c shows the HCL emission with sand passing
between the HCL and monochromator (see figure 17). This has
been described as a reliable method to generate a strong 1/f
component on a light source. The increase in NPS is
especially high in the 0-10 kHz range. Recalling that most
Michelson's operating in the UV/VIS have constructive and
destructive interference rates in 400-10,000 Hz range, a
source with a large flicker noise component, such as a flame
or atmospheric plasma, will make achieving a detector shot
noise or a photon shot noise limited case impossible. Figures
18a,b, and c are studies of the noise characteristics of the
PMT that address baseline noises and not the modulated signal.
Figure 19a and b is the measured output of a Na HCL
(589.0 nm) chopped at 680 Hz. The experimental design shown
in figure was utilized with one of the arms blocked so no
light could pass. The bandpass is 0-1 kHz. Decreasing the
frequency range increases the resolution. Figure 19a shows
the modulated HCL emission line. In 19b, sand is used to
generate a flicker noise component on the modulated HCL
emission. This raises the low frequency (0-10 Hz) (1/f) and
the high frequency (0-50 kHz) (1/f) noise magnitudes. In
figure 3a,b, and c the 1/f noise from a HCL decreases from 0-
50 kHz was demonstrated. The important point in 19a and b is
that the system is flicker noise limited but may be perceived

108
as being shot noise limited if a large frequency range is not
examined. The slope of the flicker component is not dramatic
enough to be detected in a small bandpass (eg. 0.5-1 kHz). A
larger frequency range (0-50 kHz, or greater) must be analyzed
to determine if the flicker noise component is limiting. The
background in a narrow frequency range may appear to be a
uniform background that is interpreted to be shot noise
limited.
Figure 20a shows the NPS of the He-Ne laser emission
modulated at 680 Hz with the laser light passing through the
second chopper blocked. Figure 20b shows the same modulated
line (680 Hz) with a 830 Hz signal of equal magnitude also
striking the detector. The average baseline increases (70
dBV/(Hz)!/“ to 60 dBV/(Hz)'/2) while the signal magnitude remains
constant. The He-Ne laser has a 1/f component that is still
decreasing at 50 kHz (upper limit of this system) . The
increase in baseline in this situation can be attributed to an
increase in both the shot and flicker noise caused by the
addition of the second (830 Hz) frequency.
Conclusion
A system that allows one to study the affect of noises in
a Michelson interferometer is outlined. The characterization
of noise possible in a UV/VIS FTS can be made without using a
Michelson interferometer. Some commercially available
interferometers suffer from phase problems, mirror jitter,

109
electronic noise, etc. Problems of this nature, depending on
their severity, can negatively affect the quality of spectra
obtained by the instrument giving false impressions as to the
true potential of the technique. Problems of this nature do
not allow one to study the true effect of source flicker noise
on the spectra or achieve a shot noise limited case.
Flicker (1/f) noise is shown to have two components in
HCL emission. The low frequency 0-10 Hz component and the
higher frequency, lower magnitude component that steadily
decreases out to at least 50 kHz. This high frequency (1/f)
component must be measured and considered if a shot noise
limited case is being claimed. Sources tested (He-Ne laser,
white light sources, other HCL's) had similar NPS; that is the
high frequency flicker component was still decreasing at 50
kHz. Using a spectrum analyzer with a bandwidth extending
into the MHz region would allow the extend of the high
frequency flicker component to be evaluated.

Figure 17. The emission of a He-Ne laser is split, in equal
magnitudes, chopped at different frequencies, recombined, and
analyzed. Numerous other light sources such as a HCL, white
light, and diode laser, have been studied with this design.

Ill
Chopper
B e a ms pi i l l í r

Figure 18. Noise power spectra of phphotomultiplier tube (PMT)
under various conditions of operation.
(a, top) The noise power spectra (NPS) of the dark current and
interferences picked up by the PMT.
(b, middle) The noise power spectra (NPS) of unchopped Na HCL
emission striking the PMT.
(c,bottom) The NPS of Na HCL emission with strong flicker
component induced by passing sand through the light path.

113
j
J
i
~ l?0 1 1 1 ' | I I I I I I 1X1 I I I I I I | I I
0 10000 20000 30000 40000
FREQUENCY (Hz)

Figure 19. Modulated signal of sodium (Na) HCL striking PMT
is transformed revealing its noise power spectra.
(a, top). Na HCL emission modulated at 680 Hz.
(b, bottom) Na HCL emission modulated at 680 Hz with flicker
component induced by sand. Due to small frequency range,
flicker noise may be interpreted as shot noise.

dBV/ s/Hz
115

Figure 20. Modulated He-Ne laser signal is transformed
revealing its noise power spectra.
(a,top) The modulated (680 Hz) signal of He-Ne laser.
(b, bottom) The same modulated line but an additional line
(830 Hz, not shown) of equal magnitude is striking the
detector increasing the baseline noise. The increase is
attributed to a combination of flicker and shot noises.

dBV/ /Hz dBV/ '/Hz
117
| I IT
760

CHAPTER 9
RESOLUTION AND SIGNAL TO NOISE RATIOS IN
FOURIER TRANSFORM SPECTROSCOPY
Hobbs et al. (97) discussed the effect of resolution on
the signal to noise ratio (SNR) in atomic emission (ICP)
experiments. Data concerning the SNR versus resolution were
not presented in a paper primarily concerned with source noise
in the ICP and its effects on the FT spectra. To briefly
review Hobbs et al. (97, p. 543), "There is actually an
apparent gain of (n) in going to higher resolution. This gain
has in fact been earned by increasing the integration time"
(n is the number of spectral data points acguired and can be
directly related to integration time provided other factors
such as mirror velocity and data acquisition rate are
constant) . This is true if the signal constitutes the
integrated area for the emission line profile being studied.
The signal is conserved but will be broadened and with a
decreased in intensity by instrumental effects at poor
resolutions.
Many analytical spectroscopists use the basic
relationship S/N = (average signal amplitude) /(average
noise amplitude) for SNR calculations in Fourier Transform
Spectroscopy (FTS). In FTS, the signal (line profile) can be
118

119
"smeared out" or instrumentally broadened at low resolutions
decreasing the signal amplitude (peak height) but not the
integrated area of the signal.
In this note, the nonlinear gain in SNR with resolution
is demonstrated with a Cu HCL (515.32 nm) emission line, at
five resolutions (2, 0.7, 0.2, 0.07, 0.02 cm') using a signal
amplitude calculation for the SNR. The Los Alamos Fourier
Transform Spectrometer (13,64) (LAFTS) was utilized to
acquired the data. Table 12 presents data (resolution, SNR,
number of points, etc.) for Cu HCL 515.32 nm emission line,
respectively, at five resolutions (2, 0.7, 0.2, 0.07, 0.02
cm') with the same mirror velocity. The signal-to-noise
ratio's were normalized so all resolutions have identical
integration times. The method to calculate signal to noise
ratio's shown in Table 12 is a signal amplitude to noise
amplitude ratio. The best resolution, 0.02 cm'1, is the
physically resolved line profile.
Figure 21 shows the effects resolution has on the same Cu
HCL emission line at three resolutions (2, 0.2, 0.02 cm').
Data in table 12 shows that the line profile is significantly
broadened (FWHM increases) at lower resolutions and the signal
intensity (interpreted from decreasing SNR with a constant
noise level) decreases. The SNR for all resolutions should be
the same since the integration times are identical and other
parameters (eg. mirror velocity, aperture, HCL current, etc.)
are held constant. This is not the case. Comparing the SNR

120
values for 0.02 cm' to 2 cm' resolutions, 1321 vs 31,
respectively, it is obvious that the instrumental broadening
and subsequent decrease in signal amplitude intensity causes
a substantial decrease in SNR. For the same SNR to be
achieved for all resolutions, a hypothetical situation such as
the following would have to be achieved.
Assume we are working at the shot noise limit using a
monochromatic source, data is acquired at two resolutions
(0.02 and 2 cm'1; where 0.02 cm' is physically resolved) with
equal integration times (eg. 0.02 cm', 1 scan? 2 cm', 100
scans). There is a well defined free spectral range (eg.
10,000 cm'). At 0.02 cm' the line profile can be described
with a Gaussian function, and at 2.00 cm'1 the line profile can
be described by the instrumental function. Integrating both
functions over the free spectral range will give essentially
the same integrated signal magnitude. Integrating the noise
over the FSR will also be equal for both resolutions. The
noise per resolution element will be higher at low resolution
but the total noise will remain equal. The energy (light
flux) striking the detector is conserved irrelevant of mirror
displacement. Measuring signal and noise magnitudes in this
manner is, of course, not practical for most analytical
spectroscopists but it shows that signal and noise integrated
magnitudes remain constant for all resolutions with equal
integration times.
Hobbs et al. (97, p. 544) pointed out "Although the

121
signal to noise ratio for an emission spectrum, when properly
evaluated, is independent of resolution until the lines are
actually resolved, there is a real gain to be made in the line
to continuum ratio." These data presented here support their
statement and demonstrate the misinterpretation possible.
When using a ratio method for an SNR determination at low
resolutions, there is a dependence of SNR on resolution even
when integration times are equal. If an analytical
determination of FT-UV/VIS is to be made using relative
intensities, it should be carried out at high resolution
(physically resolved) to maximized signal intensity. We point
out his possible error because several analytical evaluations
of FT-UV/VIS have been carried out at poor resolutions using
a ratio of intensity height to noise for SNR determination.

Ficpare 21. Cu (515.32 nm) emission from a HCL at three (0.02
cmf (top), 0.2 cm1 (middle), 2 cm‘(bottom)) resolutions. Using
poorer resolutions causes the loss of all physical information
involving HFS and IS, as well as a true physical line width.
The top x-axis is in angstroms and the bottom is in
wavenumbers (cm ) .

INTENSITY
123
06 Cu ECL/18 da Res=0.020
WAVENUMBER

124
Table 12. Figures of Merit for Cu (515.32 nm) line from a
Hollow Cathode Lamp measured by
RESOLUTION
cm
0.02
0.07
NUMBER OF
POINTS (N)
3.8xl06
8.6xl05
FSR, cm’1
33357.27
30011.0
SNR
1321.4
1263.6
FWHM, cm'1
0.0714
0.190
N/cm'1
114
28.5
N/FWHM
8.14
5.41
the LAFTS.
o
•
to
0.7
2.0
3.0X105
8.5X104
2.9X104
30011.0
30011.0
30011.0
765.9
207.1
31.3
0.25
1.00
3.87
9.98
2.84
0.998
2.49
2.84
3.86

CHAPTER 10
CONCLUSIONS
A common misnomer among analytical spectroscopists is
that Fourier Transform Spectroscopy in the ultraviolet and
visible does not work. For over a decade, the Fourier
Transform Spectrometer at Kitt Peak National Observatory
(Arizona) has been routinely taking excellent data in uv/vis
for physicists and astronomers. The Los Alamos Fourier
Transform Spectrometer can operate at higher resolution and
higher wavenumber accuracy than its predecessors. During my
tenure at the Los Alamos Fourier Transform Spectrometer, my
work resulted in the first publication from the LAFTS (64),
first patent (70), and the last experimental work (83) to be
carried out and published on the instrument. During its
operation I was the primary user of this national facility.
The Department of Energy spent an estimated 13 million dollars
to build the instrument. Once the instrument was built and
operational, the DOE decided not to fund it anymore. Outside
funding could not be secured, so the instrument now sits in
mothballs with no future ahead of it. To even the most casual
of observers, it is a tremendous waste of financial resources
and brain power. Its full potential had been achieved but was
not allowed to be exploited. The LAFTS would have supplied a
125

126
steady stream of spectroscopic data for scientists worldwide
to use.
Fourier Transform Spectroscopy in the ultraviolet and
visible does work. The ability to exactly measure wavenumber
positions, resolve isotope shifts and collect a large free
spectral is readily demonstrated in this dissertation. Using
a Macintosh computer allowed us to process 4 million point
Fast Fourier Transforms in 30 to 40 minutes. As time goes on,
the computation time will become faster. Advances in the
optics industry brought about by major projects such as the
Hubble Space Telescope, will increase the quality and decrease
the costs of optics routinely available to spectroscopists at
a reasonable costs. The ability to accurately move the mirror
should also advance with technology. I have recently had
contact with Kitt Peak, NIST (Chelsea Instrument) and Imperial
College (Chelsea FTS creator) about future work. This work
will be aimed at supplying data fundamental to plasma
spectroscopy. Work on hyperfine structures and isotope shifts
of metals as well as more detailed work on line shifts is
planned.

127
REFERENCES
(1) Horlick, G.; Yuen K.; Anal. Chem., 47, 775A (1975).
(2) Faires, L.; Anal. Chem. 58, 1023A (1986).
(3) Thorn, A.; J. Anal. Atom. Spectrom., 2, 227 (1987).
(4) Stubley, E.A.; Horlick, G.; Appl. Spectrosc. 39, 800
(1985).
(5) Stubley, E.A.; Horlick, G.; Appl. Spectrosc. 39, 805
(1985).
(6) Stubley, E.A.; Horlick, G.; Appl. Spectrosc. 39, 811
(1985).
(7) Marra, S.; Horlick, G.; Appl. Spectrosc. 40, 804 (1986).
(8) Faires, L.M.? Fourier Transform and Polvchromator Studies
of the Inductively Coupled Plasma. PhD. Thesis,
University of New Mexico, Los Alamos-9888-T (1985).
(9) Faires, L.M.; Spectrochim. Acta. 40B, 1473 (1985).
(10) Faires, L.; Palmer, B.; Engleman, R.; Niemczyk, T.;
Spectrochim. Acta, 40B, 545 (1985).
(11) Faires, L.; Palmer, B.; Brault, J.; Spectrochim. Acta,
4 0B, 135 (1985) .
(12) Palmer, B.A.? Los Alamos Technical Bulletin 576-
111/10060, The Los Alamos Fourier Transform Spectrometer.
LALP-85-16 (1985).
(13) Parsons, M.L.; Palmer, B.; Spectrochim. Acta, 43B, 75
(1988) .
(14) Belchamber, R.M.? Horlick, G.; Spectrochim. Acta, 37B,
17 (1982).
(15) Salin, E.D.; Horlick, G.; Anal. Chem., 52, 1578
(1980).
(16) Walden, G.; Bower, J; Nikdel, S.; Bolton, D.; Winefordner
J.D.; Spectrochim. Acta, 35B, 535 (1980).
(17) Benetti,P.; Bonelli, A.; Cambiaghi, M.; Frigieri, P.;
Spectrochima Acta, 37B, 1047 (1982).
(18) Duursma, R.; Smit, H.; Maessen, F.; Analytica Chimica
Acta, 133, 393 (1981).

128
(19) Abdallah, M.; Mermet,J.? Spectrochim. Acta, 37B, 391
(1982) .
(20) Abdallah, M.; Abdelraiman,A.; Eid, M.; Indian Journal of
Pure and Applied Physics. 22, 458 (1984).
(21) Wolnick,K; Miller,D; Seliskar,C.; Fricke,F.; Appl.
Spectrosc., 39, 930 (1985).
(22) Seliskar, C.; Warner, D.; Appl. Spectrosc. 39, 181
(1985).
(23) Miller, D.; Seliskar, C. ; Davidson, T.; Appl. Spectrosc.
39, 13 (1985).
(24) Fleitz, P.; Seliskar, C., Appl. Spectrosc. 41, 679
(1987) .
(25) Chan, S.? Van Hoven, R.; Montaser, A.; Anal. Chem., 58,
2342 (1986).
(26) Chan, S.; Montaser, A.; Spectrochim. Acta, 4OB, 1467
(1985).
(27) Chan, S.; Montaser, A.; Appl. Spectrosc., 41, 545
(1987).
(28) Montaser, A.; Van Hoven, R.; CRC Critical Reviews in
Analytical Chemistry, 18, 45 (1987).
(29) Tan, H; Chan, S.; Montaser, A.; Anal. Chem., 60, 2542
(1988).
(30) Coster, D.; Brons, H.; Z. Phys., 84, 304 (1933).
(31) Hasegawa, T.; Winefordner, J.D.; Spectrochim. Acta, 42B,
637 (1987).
(32) Seliskar, C.; Miller, D.; Fleitz, P.; Appl. Spectrosc.,
41, 658 (1987).
(33) Tomasik, Z.; Zyrnicki, W.; Spectrosc. Lett., 19, 179
(1986).
(34) Porter R.; Harshbarger, W., J. Electrochem., 126, 460-
464 (1979).
(35) Abdallah, M.; Mermet, J.; J. Quant. Spectrosc. and
Radiat. Transfer, 19, 83-91 (1978).

129
(36) Cramarossa, F.; Benedictis, S.; Ferraro, G. ; J. Quant.
Spectrosc. and Radiat. Transfer, 23, 291-301 (1980).
(37) Bechtel, J.; Appl. Optics, 18, 2100-2106 (1979).
(38) Zeeman, P.; Terblanche, S.; Visser, K.; Hamm, F.; Appl.
Spectrosc., 32, 572-576 (1978).
(39) Cabannes, F.; Pure Appl. Chem., 39, 381 (1974).
(40) Czernichowski, A.; J. Phys. D., 20(5), 559-564 (1987).
(41) Gudeman, C.; Martner, C.; Saykally, R.; Chem. Phys.
Lett., 122, 108-112 (1985).
(42) Bailly, D.; Rossetti, C.; Guelachvili, G.; Chem. Phys.,
100, 101-118 (1985).
(43) Golubovskii, Y.; Telezhko, V. ; J. Appl. Spectrosc., 39,
999-1003 (1983).
(44) Plain, A.; Ricard, A.; Phys. Lett. A, 95A, 235-238
(1983).
(45) Zyrnicki, W. ; Tomasik, Z.; Nowicka, I.; Spectrosc. Lett.,
17(3), 207-217 (1984).
(46) Davis, G.; Gottscho, R.; J. Appl. Phys. 54(6), 3080-
3086 (1983).
(47) Zheng, J.; Snow, J.; Murphy, D.; Leipertz, A.; Chang, R.;
Farrow, R.; Opt. Lett., 9(8), 341-343 (1984).
(48) Burlbaw, E.; Armstrong, R.; Appl. Opt., 22(18), 2860-2863
(1983).
(49) Lapp, M.; Goldman, L.; C.Penney, C.; Science, 175 (4026),
1112-1115 (1972) .
(50) Barrett, J.; Harvey, A.; J. Opt. Soc. Amer., 65, 392-
398 (1975).
(51) Gottscho, R.; Field, R.; Dick, K.; Benesch, W.; J. Mol.
Spectrosc., 74, 435-438 (1979).
(52) Herzberg, G.; Molecular Spectra and Molecular Structure.
(Diatomics) D. van Nostrand, Princeton (1950).
(53) Michel, H.; Gantois, M.; 18th Int. Conf. on Heat
Treatment of Materials, (conference proceedings) Detroit
(1980).

130
(54) Baronnet, J.? et al., I.S.P.C.4, Zurich, 341 (1979).
(55) Svindon, A.; Sobolev, N.; Tselikov, G. ; Zh. Eksp. Teor.
Fiz. Pis'ma, 6, 542 (1967).
(56) Dieke, G.H.? Crosswhite, H.M.;J. Quant. Spectrosc.
Radiat. Transfer, 2, 97-197 (1962).
(57) Klim, M. ; Nikdel, S.; Spectroscopy, 2, 57 (1987).
(58) Schleicher, R.; Barnes, R.; Anal. Chem., 47, 724-728
(1975).
(59) Allemand, C.; Barnes, R.; Spectrochim. Acta, 33B,
513-534 (1978).
(60) Thorn, A.P., Spectrophvsics. Chapman & Hall, London, 1974
(p. 377).
(61) Kornblum, G.; de Galen, L.; Spectrochim. Acta, 32B. 71-
96 (1977).
(62) Manning, T.; Bentley, G.; Palmer, B.; Hof, D., Spect.
Lett., 3, 341-344 (1989).
(63) Walters, P.; Chester, T.; Winefordner, J.; Appl.
Spectros., 31, 1 (1977).
(64) Manning, T.; Palmer, B.; Hof, D. ; Anal. Chem., 1052-1056
(1989).
(65) Huddlestone R.H., Leonard S.L., Plasma Diagnostic
Techniques. Academic Press, New York (1965).
(66) Matsummoto, S.; Hiño, M.; Kobayashi, T.; Appl. Phys.
Lett., 51(10), 737-739 (1987).
(67) Downey, S.; Nogare, N.; Appl. Spectros., 38, 876-79
(1984).
(68) Montaser, A.; Golightly, D.W.; Inductively Coupled
Plasmas in Analytical Atomic Spectrometerv, VCH
Publishers, Inc., Federal Republic of Germany (1987).
(69) Eckert, H.; Spectrochim. Acta, 33B, 591-608 (1978).
(70) Manning, T.; Palmer, B.; Hof, D.; Department of Energy
patent docket S-68,054 (1988)
(71)Alkemade, C.Th.J.; Hollander, Tj.; Snelleman, W;
Zeegers, P.J.Th.; Metal Vapors in Flames. Pergamon Press,
Oxford (1982).

131
(72)
(73)
(74)
(75)
(76)
(77)
(78)
(79)
(80)
Griem, H.R.; Plasma Spectroscopy. McGraw-Hill, New York,
(1964) .
Breene, R.G.; The Shift and Shape of Spectral Lines.
Pergamon, Oxford-New York (1961).
Konjevic, N.; and Weise, W.L.; J.Phys. Chem. Ref. Data
5, 259 (1976).
Konjevic N.; Roberts, J.; J. Phys. Chem. Ref. Data 5,
209 (1976).
Whaling, W.; Brault, J.W.; Physica Scripta, 38, 707
(1988).
Kuhn, H.G.; Atomic Spectra. 2 ed., Longmans, London
(1969) .
Kato, K.; Fukushima, H.; Nakajima, T.; Spectrochim.
Acta, 39b, 979 (1984).
Boumanns, P.W.J.M.; Vrakking, J.J.A.M.; Spectrochim.
Acta 39B, 1239 (1984).
Li, H.H.; Humphreys, C.J.; J. Opt. Soc. Am. 64, 1072
(1974).
(81) Fuhr, J.R.; Martin, G.A.; and Wiese, W.L.; Atomic
Transition Probabilities: Iron through Nickel. J. Phys.
Chem. Ref. Data 17, Supplement 4 (1988).
(82) Voigtman, E.; Winefordner, J.D.; Appl. Spectrosc., 41,
1182 (1987).
(83) Manning, T.J.; Winefordner, J.D.; Palmer, B.A.; Hof,
D.E.; Spectrochim. Acta B, (in press)
(84) Hirschfeld, T.; Chase, B.; Appl. Spectrosc., 40, 133
(1986).
(85) Hirschfeld, T.; Milanovich, F.; Microbeam Analysis.
D. Newbury, Ed., F.J. Heinrich, San Francisco, (1982)
p. 270.
(86) Manning, T.J.; Winefordner, J.D.? Rev. Sci. Instr.,
61(5), 1554 (1990).
(87) Kropp, J.R.; Kronfeldt, H.D.; Lucas, A.; Winkler, R.;
Physica 138C, 335 (1986).
(88) Hohberg, G.; Krebs, K.; Schulz, B.; and Winkler, R.; Z.
Phys. 186, 380 (1965).

132
(89) Burger, K.; Burghardt, B.; Buttgenbach, S.; Harzer, R.;
Hoeffgen, H.; Meisel, G.; Traber, F.; Z. Phys. A307, 201
(1982) .
(90) Manning, T.J.; Winefordner, J.D.; Palmer, B.; J. Quant.
Spec. Rad. Trans., (in press)
(91) Palmer, B.A.; Keller, R.A.; Engleman, R.; An Atlas of
Uranium Emission Intensities in a Hollow Cathode
Discharge. (Los Alamos Report) LA-8251-MS (1980).
(92) Griffiths, P.; deHaseth, J.; Fourier Transform Infrared
Spectrometerv. John Wiley and Sons, New York, (1986).
(93) Chamberlain, J.; The Principles of Interferometeric
Spectroscopy. John Wiley and Sons, New York (1972).
(94) Bell, R.J.; Introductory Fourier Transform Spectroscopy.
Academic Press, New York (1972).
(95) Bracewell, R.; The Fourier Transform and its
Applications. McGraw-Hill, New York (1965) .
(96) Ng, Y.; Horlick, G.; Appl. Spectrosc., 39, 834-841
(1985).
(97) Hobbs, P. ; Spillane, D.E.; Snook, R.D.; Thorne, A.P.;
J.Anal. Atom. Spectrosc., 3, 543 (1988).
(98) Thorne, A.; J. Anal. Atom. Spectrom., 2, 227 (1987).
(99) Irfan, A.Y.; Thorne, A.P.; Bohlander, R.A.; Gebbie,
H.A.; J.Phys. E: Sci. Instrum., 6, 12 (1979).
(100) Brault, J.W.; Phil. Trans. R.Soc.Lond. A 307, 503-511
(1982) .
(101) Brault, J.W.; J.Opt.Soc.Am., 66, 1081 (1976).
(102) Voigtman, E.; Winefordner, J.D.; Appl. Spectrosc., 41,
1182 (1987).
(103) Hirschfeld, T.; Appl. Opt., 16, 3070 (1977).
(104) Hirschfeld, T.; Appl. Spectrosc., 33, 525 (1979).
(105) Hirschfeld, T.; Appl. Spectrosc., 39, 1086 (1985).
(106) Faires, L.M.; J. Anal. Atom. Spectrom., 2, 585 (1987).
(107) Sakai, H.; and Murphy, R.E.; J. Opt. Soc. Am., 60, 422-
423 (1970).

133
(108) Natale, V. ; Ventura, G.; Appl. Opt., 23, 2052
(1984) .
(109) Knacke, R.F.; Appl. Opt., 17, 684 (1978).
(110) Luc, P.; Gerstenkorn, S.; Appl. Opt., 17, 1327 (1978).
(111) Emonds, J.G.M.; A High Resolution Fourier Transform
Spectrometer For the Visible. PhD Dissertation, Utrecht,
(1981).

134
BIOGRAPHICAL SKETCH
Thomas Joseph John Manning was born to James C. and Bette
Anne Manning on January 27, 1961. He is of Irish-Catholic
decent. He was reared on Staten Island, New York, in an area
called Pleasant Plains. He attended St. Joseph and St. Thomas
grammar school.
He attended Monsignor Farrell H.S. (1975-1979), an all
boys high school. In addition to academics, he spent much
time wrestling for his high school and the New York Athletic
Club (winner of many national AAU titles) . He then spent a
semester at Wagner College night school. The next 1.75 years
were spent at Clemson University. Working, wrestling,
studying and growing up all took time. He then returned to
Staten Island and worked at the group home on Roe Street. A
group home for PINS (People in need of supervision) . Sixteen
year old boys who got busted but were not sent to jail. This
was a real experience.
In 1982 Thomas J. Manning moved to Charleston, S.C., and
attended the College of Charleston (8/82-5/85). He obtained
a B.S. in chemistry. He was positively influenced by the
faculty at the College of Charleston (Dr. J. Deavor in
particular) to attend graduate school. He worked at the

135
Medical University of South Carolina as a technician
(producing his first publication), the bookstore with Carl
(RIP) and Spanky1s bar as a bouncer and bartender. He also
spent a summer working with Geoff Coleman, Vince, and Leigh at
the University of Alabama on a chemistry fellowship.
He then entered the University of Florida for graduate
school (1985-1990). For a variety of reasons he became
disenchanted with graduate school. On many occasions politics
and favoritism played the greatest role in success. Ken
Spears proved to be the definitive case. On the verge of
leaving, he spent a summer at Los Alamos National Lab working
with Mike Parsons, Byron Palmer, Doug Hof, and Larry Layman on
the Los Alamos Fourier Transform Facility. This changed his
outlook on science to what it had been during his
undergraduate days. He stuck it out at LANL and UF till he
reached this point. During his stay at UF he became president
of the scuba diving club and ran a lot of long races. Arlene
Reynaldos has recently become an important part of his life.
At the moment of writing this he is headed to West Virginia
University to work with Ray Lovett in the general area of
trace atomic spectroscopy.

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of Doctor
of Philosophy.
£, t<-' ifv- yjfU—*—
fames D. Winefordher, Chairman
Graduate Research Professor
of Chemistry
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of Doctor
of Philosophy.
Richard A. Yost
Professor of Chemis
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of Doctor
of Philosophy.
l/i. -yd?/ L4 '
Vaneica Y. Youi/g/
Associate Professor
â–  0 C-'W'V.,
Chemisi
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of Doctor
of Philosophy.
Anna Braj texf|-Toth
Associate Professor of Chemistry

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of Doctor
of Philosophy.
James H. Hunter
Professor of Astronomy
This dissertation was submitted to the Graduate Faculty of the
Department of Chemistry in the College of Liberal Arts and Sciences and to
the Graduate School and was accepted as partial fulfillment of the
requirements for the degree of Doctor of Philosophy.
August, 1990
Dean, Graduate School

UNIVERSITY OF FLORIDA
3 1262 08553 8311




PAGE 2

3/$60$ 63(&752&+(0,675< :,7+ $ )285,(5 75$16)250 63(&7520(7(5 %< 7+20$6 -26(3+ -2+1 0$11,1* $ ',66(57$7,21 35(6(17(' 72 7+( *5$'8$7( 6&+22/ 2) 7+( 81,9(56,7< 2) )/25,'$ ,1 3$57,$/ )8/),//0(17 2) 7+( 5(48,5(0(176 )25 7+( '(*5(( 2) '2&725 2) 3+,/2623+< 81,9(56,7< 2) )/25,'$ =W 865$0%

PAGE 3

7R P\ IDPLO\ PRWKHU IDWKHU EURWKHU VLVWHU JUDQGSDUHQWV ZKRVH VXSSRUW ORYH SDWLHQFH DQG FDULQJ PDGH WKLV SRVVLEOH

PAGE 4

$&.12:/('*(0(176 ZRXOG OLNH WR H[SUHVV P\ H[WUHPH JUDWLWXGH WR WKH IROORZLQJ SHRSOH ZKRVH KHOS GHGLFDWLRQ IULHQGVKLS KRQHVW\ DQG OR\DOW\ PDGH WKLV ZRUN SRVVLEOH 0\ IDPLO\ SOD\HG D WUHPHQGRXV UROH LQ P\ UHFHQW VXFFHVV 0\ PRWKHU KDV EHHQ D VRXUFH RI LQVSLUDWLRQ ZLWK KHU KDUG ZRUN WHDFKLQJ &OHPVRQf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

PAGE 5

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

PAGE 6

/RWV RI IULHQGV KDYH EHHQ PDGH LQ WKH ODE ZKR KDYH KHOSHG PH 6WHYH / 7\H % $OL $ .LQ 1 3DXO $QGUHV & /\QQ 3 $OLFLD 0DUWLQ $ -RUJH 9 %DUEDUD /HLJK $QQ ) (G 9 %HQ : (ULF 7RQ\ 0 5DPPH 1DQF\ 6 (GLVRQ % 'HQQLV + 6FRWW 6 $ODLQ % :HOOLQJWRQ 0 DQG (Q
PAGE 7

7$%/( 2) &217(176 SDJH $&.12:/('*(0(176 LLL $%675$&7 YLLL &+$37(5 ,1752'8&7,21 +,*+ 5(62/87,21 )285,(5 75$16)250 63(&7520(7(5 72 ,'(17,)< 7+( 527$7,21$/ 6758&785( 2) =J ;? 1f ,1 $ +(/,80 ,1'8&7,9(/< &283/(' 3/$60$ ,QWURGXFWLRQ ([SHULPHQWDO 5HVXOWV DQG 'LVFXVVLRQ &RQFOXVLRQV $ /2: )/2: $70263+(5,& 35(6685( 1(21 ,1'8&7,9(/< &283/(' 3/$60$ ,QWURGXFWLRQ ([SHULPHQWDO 'LVFXVVLRQ &RQFOXVLRQV $ 121)/2:,1* 9$5,$%/( *$6 ,1'8&7,9(/< &283/(' 3/$60$ $6 $ /,*+7 6285&( )25 +,*+ 5(62/87,21 63(&7526&23< ,QWURGXFWLRQ ([SHULPHQWDO 5HVXOWV 'LVFXVVLRQ &RQFOXVLRQV 2%6(59$7,216 2) /,1( 6+,)76 $1' /,1( 352),/(6 ,1 $1 ,1'8&7,9(/< &283/(' $5*21 3/$60$ ,QWURGXFWLRQ ([SHULPHQWDO 5HVXOWV'LVFXVVLRQ &RQFOXVLRQV YL

PAGE 8

$ 9$5,$%/( %$1'3$66 ),/7(5 )25 8/75$9,2/(7 $1' 9,6,%/( )285,(5 75$16)250 63(&7526&23< ,QWURGXFWLRQ ([SHULPHQWDO 5HVXOWV &RQFOXVLRQV $1 ,17(16,7< :$9(180%(5 $1' 5(62/87,21 67$1'$5' )25 9,6,%/( )285,(5 75$16)250 63(&7526&23< ,QWURGXFWLRQ ([SHULPHQWDO 'LVFXVVLRQ &RQFOXVLRQV $ 6<67(0 72 678'< 12,6( ,1 )285,(5 75$16)250 63(&7526&23< ,QWURGXFWLRQ ([SHULPHQWDO 5HVXOWV &RQFOXVLRQV 5(/$7,216+,3 %(7:((1 5(62/87,21 $1' 6,*1$/7212,6( 5$7,2n6 ,1 )285,(5 75$16)250 63(&7526&23< &21&/86,216 5()(5(1&(6 %,2*5$3+,&$/ 6.(7&+ YLO

PAGE 9

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

PAGE 10

7KH VHFRQG DVSHFW RI WKH GLVVHUWDWLRQ LQYROYHV VWXGLHV RI OLJKW VRXUFHV IRU )RXULHU 7UDQVIRUP 6SHFWURVFRS\ 6RXUFHV GHYHORSHG XVH DQ LQGXFWLYHO\ FRXSOHG SODVPD ,&3f SRZHU VXSSO\ 1HZ VRXUFHV QHRQ ,&3 FORVHG FHOO ,&3 DQG KHOLXP ,&3f ZHUH GHYHORSHG DQG QHZ PHWKRGV WR HQKDQFH WKH SHUIRUPDQFH DQG XQGHUVWDQG D )RXULHU 7UDQVIRUP 6SHFWURPHWHU ZHUH VWXGLHG LQFOXGLQJ D QRYHO RSWLFDO ILOWHU D VSHFWUXP DQDO\]HU WR VWXG\ QRLVHV DQG D VWDQGDUG WR FDOLEUDWH DQG HYDOXDWH D )RXULHU 7UDQVIRUP 6SHFWURPHWHU ,;

PAGE 11

&+$37(5 ,1752'8&7,21 7KLV GLVVHUWDWLRQ ZLOO LQWURGXFH QHZ DSSOLFDWLRQV DQG VWXGLHV WKDW FRXSOH D )RXULHU 7UDQVIRUP 6SHFWURPHWHU RSHUDWLQJ LQ WKH XOWUDYLROHW DQG YLVLEOH UHJLRQ RI WKH VSHFWUXP WR YDULRXV SODVPD DQG GLVFKDUJH VRXUFHV )RXULHU 7UDQVIRUP 6SHFWURPHWHUV DUH EXLOW IRU WKUHH UHDVRQV LQWHQVLW\ SUHFLVLRQ DQG DFFXUDF\ KLJK UHVROYLQJ SRZHU DQG ZDYHQXPEHU DFFXUDF\ DQG SUHFLVLRQ 7KH WKURXJKSXW DGYDQWDJH DQG PXOWLSOH[ DGYDQWDJH DOVR PDNH WKH )RXULHU 7UDQVIRUP 6SHFWURVFRS\ )76f DWWUDFWLYH EXW DUH QRW WKH PDLQ UHDVRQV WKH ZRUOGV PRVW SRZHUIXO XOWUDYLROHW DQG YLVLEOH )RXULHU 7UDQVIRUP 6SHFWURPHWHUnV KDYH EHHQ EXLOW 7KH ILUVW DGYDQWDJH LV LQWHQVLW\ SUHFLVLRQ DQG DFFXUDF\ LQ WKH OLQH SURILOH 7KLV DVSHFW LV XWLOL]HG OHVV WKDQ WKH RWKHU WZR LQ WKLV UHVHDUFK EXW LV XVHG LQ WHPSHUDWXUH PHDVXUHPHQWV HJ %ROW]PDQQ PHWKRGV LQYROYLQJ GLDWRPLFV URWDWLRQDO WHPSHUDWXUHf DQG DWRPV H[FLWDWLRQ WHPSHUDWXUHff OLQH SURILOH VWXGLHV LVRWRSH UDWLR GHWHUPLQDWLRQV HWF 7KH VHFRQG DGYDQWDJH LV UHVROXWLRQ RU KLJK UHVROYLQJ SRZHU 7KLV DGYDQWDJH ZDV XVHG WR ORRN DW URWDWLRQDO OHYHOV RI PROHFXOHV SK\VLFDOO\ UHVROYH OLQH SURILOHV DQG LGHQWLI\ LVRWRSH VKLIWV DQG K\SHUILQH VWUXFWXUH RI DWRPLF OLQHV 7KH ILQDO DGYDQWDJH

PAGE 12

LV WKH H[FHOOHQW ZDYHQXPEHU DFFXUDF\ DQG SUHFLVLRQ 7KLV IDFHW RI )7899,6 VSHFWURVFRS\ ZDV XWLOL]HG WR PHDVXUH OLQH VKLIWV LQ DQ ,QGXFWLYHO\ &RXSOHG 3ODVPD ,&3f DQG EHWWHU WKH H[LVWLQJ ZDYHQXPEHU DFFXUDF\ RI D ZHOO VWXGLHG 1 WUDQVLWLRQ LQ XOWUDYLROHW $ VXPPDWLRQ RI HDFK FKDSWHU IROORZV 7KH VHFRQG FKDSWHU LV D FRQWLQXDWLRQ RI P\ PDVWHUnV WKHVLV 7KH +HOLXP ,QGXFWLYHO\ &RXSOHG 3ODVPD GHVFULEHG LQ WKLV SDSHU ZDV ILUVW GHYHORSHG IRU QRQPHWDO DWRPLF HPLVVLRQ VSHFWURVFRS\ DW /RV $ODPRV 1DWLRQDO /DE /$1/f 7KH VRXUFH SURYHG WR EH YHU\ VWDEOH ORZ IOLFNHU QRLVH FRPSRQHQWf FRPSDUHG WR RWKHU IORZLQJ V\VWHPV KDG YHU\ OLWWOH EDFNJURXQG DQG LV VXIILFLHQWO\ HQHUJHWLF WR EH XVHG DV D OLJKW VRXUFH 7KLV SURMHFW ZDV WKH ILUVW SXEOLVKHG ZRUN XVLQJ WKH /RV $ODPRV )RXULHU 7UDQVIRUP 6SHFWURPHWHU ,W FOHDUO\ GHPRQVWUDWHG WKDW WKH LQVWUXPHQW D PLOOLRQ GROODU QDWLRQDO IDFLOLW\ ZDV IXQFWLRQLQJ DW D KLJK OHYHO ,W DOVR GHPRQVWUDWHG WKH +H,&3 LV D JRRG KLJK UHVROXWLRQ OLJKW VRXUFH 7KH 1 f WUDQVLWLRQ ZDV UHVROYHG ZLWK KLJKHU ZDYHQXPEHU DFFXUDF\ SDUW LQ "f WKDQ KDG EHHQ SUHYLRXVO\ EHHQ UHSRUWHG LQ WKH OLWHUDWXUH 7KLV VDPH WRUFK ZDV XVHG WR JHQHUDWH D ORZ IORZ DWPRVSKHULF SUHVVXUH 1HRQ ,QGXFWLYHO\ &RXSOHG 3ODVPD &KDSWHU WKUHH LV D EULHI RXWOLQH RI WKH ILUVW 1HRQ ,QGXFWLYHO\ &RXSOHG 3ODVPD GHVFULEHG LQ WKH OLWHUDWXUH 7KLV ZDV RULJLQDOO\ JHQHUDWHG IRU LQGXFWLYHO\ FRXSOHG SODVPDPDVV VSHFWURPHWU\ VWXGLHV EXW WKH SULFH RI QHRQ SURYHG WRR H[SHQVLYH WR

PAGE 13

FRQWLQXH WKLV UHVHDUFK &KDSWHU IRXU GHDOV ZLWK D QRQIORZLQJ YDULDEOH JDV LQGXFWLYHO\ FRXSOHG SODVPD DV D OLJKW VRXUFH IRU KLJK UHVROXWLRQ VSHFWURVFRS\ 7KLV ZRUN ZDV UHFHQWO\ LVVXHG D SDWHQW E\ WKH 8QLWHG 6WDWHV 3DWHQW RIILFH ,W JHQHUDWHG SODVPDV ZLWK GLIIHUHQW JDVHV +H 1H $U .U ;H + 1 &2 &+ 2f RYHU D ZLGH UDQJH RI SUHVVXUH WRUUf :RUN ZDV DOVR LQLWLDWHG WR GHPRQVWUDWH WKH VWDWLF ORZ SUHVVXUH VRXUFH FRXOG EH LQWHUIDFHG WR D PDVV VSHFWURPHWHU ZLWK VHYHUDO DGYDQWDJHV RYHU WKH WUDGLWLRQDO ,&306 )RU )RXULHU 7UDQVIRUP 6SHFWURVFRS\ )76f UHVHDUFK WKH VWDWLF HQYLURQPHQW LV LPSRUWDQW WR PLQLPL]H IOLFNHU QRLVH LQ WKH VRXUFH 9DU\LQJ WKH JDV DOORZV XV WR YDU\ WKH 'RSSOHU WHPSHUDWXUH DQG HOHFWURQ GHQVLW\ DV ZHOO DV WKH DPRXQW RI EDFNJURXQG OLJKW IOX[ &KDSWHU ILYH LV D SK\VLFDO DQDO\WLFDO VWXG\ 7KLV FKDSWHU UHSUHVHQWV WKH ODUJHVW SURMHFW XQGHUWDNHQ WKH PHDVXUHPHQW RI OLQH VKLIWV DQG OLQH SURILOHV RI OLJKW HPLWWHG IURP D DUJRQ LQGXFWLYHO\ FRXSOHG SODVPD 7KHUH DUH VPDOO VKLIWV LQ ZDYHQXPEHU SRVLWLRQ EHWZHHQ ORZ SUHVVXUH VRXUFHV QR FROOLVLRQDO EURDGHQLQJ QR VKLIWf DQG DWPRVSKHULF SUHVVXUH VRXUFHV /LQH SURILOHV ZHUH GHFRQYROXWHG WR PHDVXUH EURDGHQLQJ FRQWULEXWLRQV IURP 'RSSOHU DQG FROOLVLRQDO SURFHVVHV LQ WKH DWPRVSKHULF SUHVVXUH DUJRQ ,&3 $SSUR[LPDWHO\ OLQHV ZLWK D VLJQDOWRQRLVH UDWLR 615f JUHDWHU WKDQ ZHUH PHDVXUHG ZLWK WKH /$)76 7KLV SXVKHG

PAGE 14

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f OLQH KDV ERWK K\SHUILQH VWUXFWXUH +)6f DQG LVRWRSH VKLIWV ,6f 7KLV VWUXFWXUH ZDV SURSRVHG DV D VWDQGDUG WR GHWHUPLQH WKH TXDOLW\ RI VSHFWUD DFTXLUHG IURP D )76 7KH UHDVRQ IRU GRLQJ WKLV LV VLPSOH ,QVWUXPHQW FRPSDQLHV SURGXFLQJ LQVWUXPHQWV KDYH EHHQ VHOOLQJ EDG LQVWUXPHQWV WKDW LV )RXULHU 7UDQVIRUP 6SHFWURPHWHUV WKDW GR QRW FRPH FORVH WR PHHWLQJ WKHLU SXEOLVKHG VSHFV RIWHQ E\ RUGHUV RI PDJQLWXGHf 7KH SHRSOH SXUFKDVLQJ WKH LQVWUXPHQWV W\SLFDOO\ GR QRW NQRZ PXFK DERXW )76 DQG DUH QRW DZDUH RI WKH SUREOHPV %\ SURSRVLQJ D VWDQGDUG IRU ZDYHQXPEHU SRVLWLRQ LQWHQVLW\ SUHFLVLRQ DQG UHVROXWLRQ LW ZLOO DOORZ SHRSOH WR MXGJH WKH TXDOLW\ RI WKH LQVWUXPHQW WKH\ DUH DERXW WR EX\ RU WR MXGJH WKH SHUIRUPDQFH RI WKH LQVWUXPHQW DW DQ\ JLYHQ WLPH GXULQJ LWV OLIHWLPH ,W FDQ DOVR VHUYH DV D FDOLEUDWLRQ VWDQGDUG ZKHQ ILQH WXQLQJ WKH LQVWUXPHQW &KDSWHU VHYHQ

PAGE 15

GHVFULEHV WKLV VWDQGDUG &KDSWHU HLJKW FRYHUV QRLVH VWXGLHV LQ )RXULHU 7UDQVIRUP 6SHFWURVFRS\ XVLQJ D UHODWLYHO\ VLPSOH DUUDQJHPHQW 7KH RXWSXW RI WKH OLJKW VRXUFH LV FKRSSHG E\ D PHFKDQLFDO FKRSSHU &KRSSLQJ VLPXODWHV WKH FRQVWUXFWLYH DQG GHVWUXFWLYH LQWHUIHUHQFH LQ DQ LQWHUIHURPHWHU :KHQ WKH EHDP LV EORFNHG E\ WKH FKRSSHU WKLV VLPXODWHV GHVWUXFWLYH LQWHUIHUHQFH DQG ZKHQ OLJKW LV SDVVHG WKLV VLPXODWHV FRQVWUXFWLYH LQWHUIHUHQFH 7KH EHDP LV FKRSSHG DW D UDWH +]f VLPLODU WR WKDW SURGXFHG E\ FRQVWUXFWLYH DQG GHVWUXFWLYH LQWHUIHUHQFH LQ D 0LFKHOVRQ LQWHUIHURPHWHU 7KLV ZRUN LQGLFDWHV WKDW D VKRW QRLVH OLPLWHG FDVH KDV QRW EHHQ DFKLHYHG E\ DQDO\WLFDO VSHFWURVFRSLVWV GHVSLWH WKHLU FODLPV $QDO\WLFDO VSHFWURVFRSLVWV W\SLFDOO\ XVH D UDWLR RI VLJQDO DQG QRLVH LQWHQVLWLHV WR FDOFXODWH WKH VLJQDOWRQRLVH UDWLR LQ )76 7KLV PHWKRG IDLOV DW SRRU UHVROXWLRQV ZKHQ WKH OLQH SURILOH LV EURDGHQHG E\ LQVWUXPHQWDO HIIHFWV GHFUHDVLQJ WKH DPSOLWXGH SHDN KHLJKWf RI WKH OLQH SURILOH 3URYLGHG RXU LQWHJUDWLRQ WLPHV DUH HTXDO WKHUH VKRXOG EH LGHQWLFDO VLJQDO WRQRLVH UDWLRV IRU KLJK DQG ORZ UHVROXWLRQV DOLNH 7KLV UHVHDUFK KDV IRXQG WKLV QRW WR EH WKH FDVH $W SRRU UHVROXWLRQV WKH ORVV RI VLJQDO DPSOLWXGH FDXVHG E\ LQVWUXPHQWDO EURDGHQLQJ QHJDWLYHO\ DIIHFWV WKH 615 0DQ\ DQDO\WLFDO VWXGLHV KDYH EHHQ FDUULHG RXW DW SRRU UHVROXWLRQV DQG /2'V EDVHG RQ DPSOLWXGH 615V JLYLQJ ORZ YDOXHV 0DQ\ RI WKHVH VWXGLHV FODLP WR KDYH DFKLHYHG VKRW QRLVH OLPLWHG FDVHV

PAGE 16

:RUN LQ FKDSWHU VHYHQ JLYHV FRQFOXVLYH HYLGHQFH WKDW DQDO\WLFDO FKHPLVWV KDYH QRW DFKLHYHG D VKRW QRLVH OLPLWHG VSHFWUD

PAGE 17

&+$37(5 86( 2) $ +,*+ 5(62/87,21 )285,(5 75$16)250 63(&7520(7(5 72 ,'(17,)< 7+( 527$7,21$/ 6758&785( 2) 7+( %(8 `7=J 75$16,7,21 2) 1 f ,1 $ +(/,80 ,1'8&7,9(/< &283/(' 3/$60$ ,QWURGXFWLRQ 7KH XVH RI )RXULHU 7UDQVIRUP 6SHFWURPHWHUV LQ WKH QHDU PLGGOH DQG IDULQIUDUHG VSHFWUDO UHJLRQ LV FRPPRQ LQ PDQ\ DQDO\WLFDO ODEV 6LQFH D ZHDOWK RI DWRPLF LQIRUPDWLRQ OLHV LQ WKH YLVLEOH DQG XOWUDYLROHW UHJLRQV WKHUH KDV EHHQ D JUHDW GHDO RI HIIRUW E\ YDULRXV JURXSV ZRUOGZLGH WR EXLOG )RXULHU 7UDQVIRUP 6SHFWURPHWHUV WKDW RSHUDWH RYHU WKLV UHJLRQ RI WKH VSHFWUXP ,Q +RUOLFN DQG
PAGE 18

/RV $ODPRV )RXULHU 7UDQVIRUP 6SHFWURPHWHU 7KH /RV $ODPRV )RXULHU 7UDQVIRUP 6SHFWURPHWHU /$)76f KRXVHG LQ D IWf EXLOGLQJ LV WKH KLJKHVW UHVROXWLRQ YLVLEOH DQG XOWUDYLROHW )76 LQ WKH ZRUOG 7DEOH OLVWV VRPH RI LWV FKDUDFWHULVWLFV 7KH LQVWUXPHQW LV FXUUHQWO\ RSHUDWLQJ ZLWK VLQJOH SDVV RSWLFV DOORZLQJ IRU D PD[LPXP UHVROXWLRQ RI FP SP DW QPf 7KH LQWHUIHURPHWHU LV KRXVHG ZLWKLQ D IRRW E\ IRRW YDFXXP WDQN P7RUU DFKLHYHGf DQG XWLOL]HV WZR HQWUDQFH SRUWV IRU H[SHULPHQWV 7KH )76 VLWV RQ D IRRW E\ IRRW 15& YDFXXP FRPSDWLEOH RSWLFDO WDEOH LVRODWLQJ WKH LQVWUXPHQW IURP YLEUDWLRQ )LJXUH VKRZV WKH IROGHG GHVLJQ RI WKH LQVWUXPHQW 7KH /$)76 XWLOL]HV D GRXEOH VLGHG LQWHUIHURJUDP ZKLFK LPSURYHV WKH VLJQDOWRQRLVH UDWLR DQG KHOSV HOLPLQDWH SKDVH SUREOHPV 7KH $' FRQYHUWHU LV FRPSRVHG RI VHYHQ ELW $' FRQYHUWHUV VFDOH DPSOLILHUV DQG DQ ELW '$ FRQYHUWHU IRU FDOLEUDWLRQ 3DUVRQV DQG 3DOPHU f JLYH GHWDLOHG FKDUDFWHULVWLFV DQG UHVHDUFK SODQV IRU WKH /$)76 )RU LQGXFWLYHO\ FRXSOHG SODVPDDWRPLF HPLVVLRQ VWXGLHV ,&3$(6f WKH DGYDQWDJHV RI WKH )76 DUH LQWHQVLW\ SUHFLVLRQ ZDYHQXPEHU DFFXUDF\ DQG KLJK UHVROXWLRQ 3URYLGHG HOHFWURQLF QRLVH LV PLQLPL]HG WKH OLPLWLQJ QRLVH LQ WKH )7,&3 V\VWHP ZLOO EH WKH SODVPD 6HYHUDO SDSHUV f KDYH GLVFXVVHG VRXUFHV RI QRLVH LQ $UJRQ ,&3V

PAGE 19

+HOLXP ,QGXFWLYHO\ &RXSOHG 3ODVPD $UJRQ LV WKH PRVW FRPPRQ SODVPD JDV XVHG LQ ,QGXFWLYHO\ &RXSOHG 3ODVPDV IRU HOHPHQWDO DQDO\VLV $EGDOODK DQG 0HUPHW f FRPSDUHG WHPSHUDWXUHV RI KHOLXP DQG DUJRQ LQ ,&3 DQG 0,3 V\VWHPV $EGDOODK HW DO IROORZHG ZLWK D SDSHU f XWLOL]LQJ D 0+] +HOLXP,&3 DW DWPRVSKHULF SUHVVXUH IRU DQDO\WLFDO VWXGLHV 6HOLVNDUnV ODERUDWRU\ f FRQGXFWHG ZRUN ZLWK D UHGXFHG SUHVVXUH WRUFK WKDW LV DGDSWDEOH WR D FRPPHUFLDO ,&3 XQLW &KDQ HW DO f GHYHORSHG DQG WHVWHG D WDQJHQWLDO IORZ ORZ IORZ DWPRVSKHULF SUHVVXUH WRUFK WKDW LV FRQVWUXFWHG HQWLUHO\ RI PDFRU DQG DGDSWV UHDGLO\ WR FRPPHUFLDO XQLWV 0RQWDVHU DQG 9DQ +RYHQ f JLYH D FULWLFDO UHYLHZ RQ PL[HG JDV PROHFXODU JDV DQG KHOLXP ,&3V 'XH WR WKH KLJK LRQL]DWLRQ SRWHQWLDO RI KHOLXP H9f YHUVXV DUJRQ H9f LW KDV DFKLHYHG LPSURYHG OLPLWV RI GHWHFWLRQ IRU HOHPHQWV ZLWK UHODWLYHO\ KLJK LRQL]DWLRQ SRWHQWLDOV &O H9 %U H9f f +H,&3V KDYH IHZHU VSHFWUDO OLQHV WKDQ $U ,&3V LQ WKH GHVLUHG UHJLRQV RI $(6 GHFUHDVLQJ WKH FKDQFH IRU RYHUODS f 7DQ HW DO f UHFHQWO\ LQWURGXFHG D ODPLQDU IORZ +H,&3 WRUFK DQG GHPRQVWUDWHG LWV XVH LQ WKH DQDO\VLV RI VRPH QRQPHWDOV 1 I R FQ ,Q &RVWHU DQG %URQV f SXEOLVKHG GHWDLOHG ZRUN RQ WKH HPLVVLRQ VSHFWUXP RI 1f 'LDWRPLF QLWURJHQ 11rf KDV EHHQ XWLOL]HG DV D URWDWLRQDO WHPSHUDWXUH LQGLFDWRU LQ 5)

PAGE 20

SODVPDV f DQG YDULRXV GLVFKDUJHV f 0RVW SDSHUV XVH WKH XQUHVROYHG YLEUDWLRQDO EDQGKHDG RU WKH SDUWLDOO\ UHVROYHG URWDWLRQDO VWUXFWXUH 7KH XVH RI 5DPDQ VFDWWHULQJ IURP 1 DV D WHPSHUDWXUH LQGLFDWRU KDV DOVR EHHQ GHPRQVWUDWHG E\ VHYHUDO JURXSV f *RWWVFKR HW DO f JLYH DQ DFFRXQW RI SHUWXUEDWLRQ HIIHFWV ZLWKLQ WKH %eX `76J WUDQVLWLRQ RI 1 +HU]EHUJ f JLYHV DQ H[FHOOHQW GLVFXVVLRQ RI ERWK WKH WKHRUHWLFDO DQG H[SHULPHQWDO DSSOLFDWLRQV RI WKH WUDQVLWLRQ 0RQWDVHU DQG 9DQ +RYHQ f JLYH UHIHUHQFHV IRU DGGLWLRQDO XVHV RI 1 DQG 1r LQ SODVPD GLDJQRVWLFV 1LWURJHQ LV XVHG LQ SODVPDV IRU PHWDO VXUIDFH WUHDWPHQW f V\QWKHVLV RI QLWURJHQ R[\GHV f DQG &2 &2 DQG JDV ODVHUV f 7KH XVH RI 1 LQ VXEGRSSOHU UHVROXWLRQ VWXGLHV LV GLVFXVVHG LQ WKLV SDSHU 7KH ILUVW QHJDWLYH EDQG RI 1 KDV VHYHUDO DGYDQWDJHV IRU URWDWLRQDO WHPSHUDWXUH GHWHUPLQDWLRQV LQ SODVPDV 7KH f EDQGKHDG QPf LV LQ DQ DFFHVVLEOH UHJLRQ RI WKH VSHFWUXP 7KH 4 EUDQFK $f LV IRUELGGHQ GHFUHDVLQJ WKH QXPEHU RI URWDWLRQDO OLQHV SUHVHQW UHODWLYH WR 1f 1f DQG 2+f f ZKLFK DUH XVHG DV URWDWLRQDO WHPSHUDWXUH LQGLFDWRUV $W PRGHUDWH UHVROXWLRQ FPf WKH URWDWLRQDO VWUXFWXUH FDQ EH DGHTXDWHO\ UHVROYHG 7KLV FKDSWHU SUHVHQWV D ORZ IORZ ODPLQDU IORZ DWPRVSKHULF SUHVVXUH KHOLXP ,&3 WRUFK 1LWURJHQ LV VHHGHG LQWR WKH KHOLXP SODVPD DQG WKH HPLVVLRQ IURP 1 LV DQDO\]HG XVLQJ WKH /RV $ODPRV )RXULHU 7UDQVIRUP 6SHFWURPHWHU

PAGE 21

([SHULPHQWDO +HOLXP LQGXFWLYHO\ FRXSOHG SODVPD WRUFK 7KH WRUFK LV RSWLPL]HG IRU ORZ IORZ /PLQf VLPSOLFLW\ LQ GHVLJQ DQG DGDSWDELOLW\ WR D FRPPHUFLDO XQLW &KDQ HW DO f SUHVHQW D ORZ IORZ /PLQf WDQJHQWLDO IORZ PDFRU WRUFK 7ZR SUREOHPV DULVH ZLWK WKLV GHVLJQ )LUVW PDFRU &RUQLQJ +RUVHKHDGV 1
PAGE 22

QHEXOL]HU f ,W KDV KLJKHU HIILFLHQF\ b ZLWK /PLQ JDV IORZf ORZHU FRVW DQG D GHFUHDVHG QRLVH SRZHU VSHFWUXP f YHUVXV WKH XOWUDVRQLF QHEXOL]HU +HOLXP GRHV QRW ZRUN ZLWK D FRQFHQWULF QHEXOL]HU GXH WR LWV UHODWLYH ODFN RI YLVFRVLW\ YHUVXV DUJRQ $Q H[WHQGHG JXDUW] WXEH ZDV XVHG WR NHHS LPSXULWLHV 1 + HWFf IURP WKH DWPRVSKHUH RXW RI WKH SODVPD ,W KDV RQH RU WZR ZUDSV RI WHIORQ WDSH DURXQG LWV EDVH IRU DQ DLUWLJKW ILW ZLWK WKH EDVH $ IRXU WXUQ JROG SODWHG f FRLO ZDV XVHG WR JHQHUDWH WKH 0+] 5) ILHOG 7KH FHQWUDO WHIORQ SLHFH DQG WKH WHIORQ EDVH GLDPHWHUf ILW WLJKWO\ WRJHWKHU 7KH WRUFK LV VXSSRUWHG RQ D WHIORQ VWDQG LQ WKH ,&3 ER[ 7KH WHVOD FRLO LV DWWDFKHG YLD DQ LJQLWLRQ ZLUH WR RQH RI WKH VLGH SRUWV ZLWK D FRSSHU FOLS 4XDUW] WXELQJ GLDPHWHUf LV LQVHUWHG LQWR WKH WZR VLGH SRUWV $ EDOOMRLQW LV LQVHUWHG LQWR WKH VDPSOH XSWDNH SRUW 7KH PDFRU WLS KDV D LQFK GLDPHWHU DQG D LQFK KHLJKW ,W ILWV WLJKWO\ ZLWK WKH WHIORQ FKDQQHO VHH )LJXUH f 8QGHU QRUPDO RSHUDWLQJ FRQGLWLRQV WKLV ILW LV VXIILFLHQW WR VHDO WKH WLS IURP RXWVLGH FRQWDPLQDQWV DQG NHHS WKH VDPSOH DQG SODVPD JDV VHSDUDWH $V DQ DGGHG SUHFDXWLRQ WKH RXWVLGH RI WKH PDFRUWHIORQ FRQQHFWLRQ LV ZUDSSHG ZLWK WHIORQ WDSH 7KH PDFRU WR WHIORQ FRQQHFWLRQ FDQ EH IXVHG ZLWK FKHPLFDO HWFKLQJ WHFKQLTXHV RU D KLJK WHPSHUDWXUH TXDUW] EDVHG FHPHQW FDQ EH XVHG DV D VHDODQW ,PSHGDQFH PDWFKLQJ XWLOL]HG KDV EHHQ SUHYLRXVO\ GHVFULEHG LQ WKH OLWHUDWXUH f

PAGE 23

,QGXFWLYHO\ FRXSOHG SODVPD 7KLV VWXG\ HPSOR\V D 0+] 5) 3ODVPD 3URGXFWV &KHUU\ +LOO 1-f +)3 PRGHO ,QGXFWLYHO\ &RXSOHG 3ODVPD XQLW 7KH ,&3 RSHUDWLQJ FRQGLWLRQV DUH RXWOLQHG LQ WDEOH 7KH QLWURJHQ LV b FHUWLILHG SXUH DQG ZDV REWDLQHG IURP 8QLRQ &DUELGH/LQGH 'LYLVLRQ $OEXTXHUTXH 10f /RV $ODPRV )RXULHU WUDQVIRUP VSHFWURPHWHU 7KH SDUDPHWHUV RI WKH /$)76 XVHG IRU WKLV VWXG\ DUH DV IROORZV 7KH VSHFWUDO RXWSXW LV IURP FP WR FPn 7KH GHWHFWRUV XWLOL]HG DUH 5&$ O3V 7KH EHDPVSOLWWHUV DUH DOXPLQXP FRDWHG TXDUW] DQG QR RSWLFDO ILOWHUV DUH XVHG 7KH UHVROXWLRQ FKRVHQ FPf LV VXE'RSSOHU IRU QLWURJHQ f 7KH QXPEHU RI SRLQWV WDNHQ LV ZKLFK LV ]HUR ILOOHG WR f f SRLQWV IRU WUDQVIRUPDWLRQ 7KH GDWD DFTXLVLWLRQ WLPH IRU WKH VFDQV LV KRXU PLQXWHV DQG WKH FRPSXWHU WUDQVIRUP WLPH LV PLQXWHV 7KH VSHFWUXP LV XQDSRGL]HG ER[FDUf $WPRVSKHULF SUHVVXUH LQ WKH /$)76 IW DERYH VHD OHYHOf LV 7RUU DQG WKH WDQN WHPSHUDWXUH LV r& 5HVXOWV DQG 'LVFXVVLRQ 7DEOH JLYHV WKH OLQH SRVLWLRQV s FPf RI WKH 3 DQG 5 EUDQFKHV REVHUYHG LQ WKH +HOLXP ,QGXFWLYHO\ &RXSOHG 3ODVPD 7KH DFFXUDF\ RI WKH fEDQG GDWD JLYHQ E\ &RVWHU DQG %URQV f LV XSGDWHG E\ WR FPn )LJXUH VKRZV SDUW RI WKH f VSHFWUD 7KH 3 f

PAGE 24

DQG 5f EUDQFKHV DUH LGHQWLILHG 7KH SDWWHUQ LV FRQVLVWHQW IRU ODUJHU EUDQFK QXPEHUV QRW VKRZQ 7KH GRXEOHW VSOLWWLQJ -O-Of LV HYLGHQW IRU ERWK EUDQFKHV 3 5 !f 7KH DOWHUQDWLRQ RI OLQH LQWHQVLW\ HYHQ RGGf LV XQPLVWDNDEOH IRU ERWK EUDQFKHV 'RXEOHW VSOLWWLQJ DQG HYHQ RGG DOWHUQDWLRQ RI OLQH LQWHQVLW\ DUH WUHDWHG WKHRUHWLFDOO\ E\ +HU]EHUJ f )LJXUH JLYHV WKH )RUWUDW GLDJUDP IRU WKH 3 DQG 5 EUDQFKHV LQ WKH +H,&3 7KH 3 EUDQFK IROGV RYHU LWVHOI DW FPn JLYLQJ WKH YLEUDWLRQDO EDQGKHDG 7KH YDOXH XVHG IRU DOO EUDQFKHV WKDW DUH GRXEOHW VSOLW LV -O $W KLJKHU YDOXHV 5 3!f RQO\ WKH HYHQ EUDQFKHV DUH GHWHFWDEOH IURP WKLV VRXUFH +H,&3f 7KH PHWKRG WR FDOFXODWH WKH URWDWLRQDO WHPSHUDWXUH IURP D ZHOO UHVROYHG VWUXFWXUH LV WDNHQ IURP $EGDOODK DQG 0HUPHW f 7KLV PHWKRG DOORZV IRU UHODWLYH OLQH LQWHQVLW\ PHDVXUHPHQWV DV RSSRVHG WR DEVROXWHf IRU HLWKHU WKH 3 RU 5 EUDQFKHV )RU WKH 3 EUDQFK /RJ DO I..Off f DQG WKH 5 EUDQFK /RD DO I.Of.ff f .O ZKHUH D HYHQ QXPEHU OLQHVf RGG QXPEHU OLQHVf I %KFN7 7f H[SHULPHQWDO OLQH LQWHQVLW\ TXDQWXP QXPEHU RI ORZHU VWDWH 7 WHPSHUDWXUH NHOYLQf

PAGE 25

3ORWWLQJ WKH OHIW VLGH RI HTXDWLRQ RU IRU YDU\LQJ YDOXHV RI ,f YHUVXV WKH ULJKW VLGH LQFUHDVLQJ YDOXHV RI .f RI WKH UHVSHFWLYH HTXDWLRQ \LHOGV D VWUDLJKW OLQH ZLWK D VORSH SURSRUWLRQDO WR 7 5RWDWLRQDO WHPSHUDWXUHV IRU WKH +H,&3 XVLQJ WKH SDUDPHWHUV DQG GDWD OLVWHG DERYH DUH 5RGG .f 5HYHQ .f 3RGG .f DQG 3HYHQ .f 9DULRXV SDSHUV f RXWOLQH VLPLODU %ROW]PDQQ PHWKRGV WR FDOFXODWH URWDWLRQDO WHPSHUDWXUHV LQ SODVPDV 7KLV PHWKRG DOORZV IRU IRXU URWDWLRQDO WHPSHUDWXUHV 3 HYHQRGGf 5HYHQRGGff 7KRUQH f FODLPV WKDW WKH %ROW]PDQQ PHWKRG IRU 1 f LV VXLWDEOH IRU PHDVXUHPHQWV XQGHU &DEDQQHV f FODLPV LW LV DOORZDEOH IRU WHPSHUDWXUHV XQGHU .RUQEOXP DQG GH *DOHQ f KDYH LQWHUSUHWHG WKH UHVXOWLQJ PHDVXUHPHQW DV WKH JDV NLQHWLF WHPSHUDWXUH &RQFOXVLRQ $ QHZ +H,&3 WRUFK KDV EHHQ GHPRQVWUDWHG 7KH WRUFK KDV DOVR EHHQ XVHG WR JHQHUDWH D QHRQ ,&3 f DQG LV EHLQJ LQYHVWLJDWHG DV DQ LRQL]DWLRQ VRXUFH IRU KHOLXP DQG QHRQ ,&306 VWXGLHV 7KH +H,&3 LV QRW EHLQJ SXUVXHG IRU DWRPLF HPLVVLRQ VSHFWURVFRS\ 7KH XVH RI 1 DV D GLDJQRVWLF WRRO LV FRPPRQ 7KLV SDSHU DOORZV WKH VSHFWURVFRSLVW WR HDVLO\ LGHQWLI\ DQG XVH WKH VWUXFWXUH IRU URWDWLRQDO WHPSHUDWXUHV 7KH OLQH SRVLWLRQ DFFXUDF\ RI WKH URWDWLRQDO OHYHOV LV LPSURYHG DJDLQVW WKRVH FXUUHQWO\ LQ WKH OLWHUDWXUH

PAGE 26

)LJXUH 2SWLFDO (OHPHQWV Df FDWnVH\H UHIOHFWRUV Ef EHDP VSOLWWHU DQG FDURXVHO Ff LQFK WXUQLQJ PLUURUV Gf FROOLPDWRU RSWLFV Hf RXWSXW PLUURUV /LJKW HQWHUV IURP HLWKHU WKH HQG RU VLGH SRUW DQG LV FROOLPDWHG E\ G 7KHUH LW JRHV WR WKH XSSHU EHDP VSOLWWHU E DQG LV GLYLGHG 7KHVH WZR EHDPV JR WR WKH WXUQLQJ PLUURUV F DQG WR WKH SULPDU\ PLUURU RI WKH FDWnVH\H UHIOHFWRU D ,Q WKH FDWnVH\H UHIOHFWRU WKH EHDP LV IRFXVHG GRZQ RQWR WKH VHFRQGDU\ PLUURU DQG UHIOHFWHG EDFN WR WKH SULPDU\ ZKHUH LW LV PDGH SDUDOOHO DJDLQ 7KH HIIHFW RI WKH FDWnVH\H UHIOHFWRU LV WR UHIOHFW WKH EHDP EDFN DORQJ D SDWK WKDW LV SDUDOOHO EXW GLVSODFHG )URP WKH FDWnV H\H UHIOHFWRUV WKH EHDPV UHWXUQ WR WKH LQFK PLUURUV DQG WKHQ WR WKH ORZHU EHDP VSOLWWHU 7KHUH WKH EHDP LV UHFRPELQHG DQG LQWHUIHUHV ZLWK LWVHOI 7KH OLJKW WKHQ WUDYHOV WR WKH RXWSXW PLUURUV H WKDW IRFXVHV WKH EHDP DQG WR WKH IODW PLUURUV WKDW GLUHFW WKH EHDP WR WKH GHWHFWRUV

PAGE 27

,

PAGE 28

)LJXUH 'LDJUDP RI ORZIORZ DWPRVSKHULF SUHVVXUH +H,&3 WRUFK 7KH WRS LQVHW VKRZV WKH PDFRU WLS DQG WHIORQ FHQWUDO FKDQQHO LQWHUIDFH LQ JUHDWHU GHWDLO 7KH ERWWRP LQVHW VKRZV WKH HLJKW SODVPD JDV KROHV VXUURXQGLQJ WKH VDPSOH JDV LQOHW RQ WKH PDFRU WLS

PAGE 30

)LJXUH 7KH IRXU SDQHOV VKRZ D SRUWLRQ RI WKH VWUXFWXUH RI WKH 1f EDQG 7KH WRS [D[LV LV DQG WKH ERWWRP LV ZDYHQXPEHUV FPf URWDWLRQDO DQJVWURPV

PAGE 31

,17(16,7< 3 -ff 3f

PAGE 32

,17(16,7< )LJXUH FRQWLQXHG

PAGE 33

,17(16,7< )LJXUH f§FRQWLQXHG

PAGE 34

,17(16,7< )LJXUH f§FRQWLQXHG

PAGE 35

)LJXUH )RUWUDW GLDJUDP RI 3 DQG 5 EUDQFKHV $W FP WKH 3 EUDQFK 3 3 3f IROGV JLYLQJ WKH EDQGKHDG

PAGE 36

Â’ 5 %5$1&+ 3 %5$1&+ c :$9(180%(5

PAGE 37

7DEOH /RV $ODPRV )2XULHU 7UDQVIRUP 6SHFWURPHWHU 6SHFLILFDWLRQV 6SHFWUDO 5DQJHV 5HJLRQ 5DQJH %HDPVSOLWWHUV 'HWHFWRUV 899,6 QP XP TXDUW]DOXPLQXP 1($5 ,5 )$5 ,5 XP XP &D)*D3 .&O*H SKRWRPXOWLSOLHU VL 3,1 GLRGH ,Q6E +J&G7H 5HVROXWLRQ DQG 2SHUDWLQJ 3DUDPHWHUV 2SWLFDO 3DWK 'LIIHUHQFH 5HVROXWLRQ 6DPSOH 5DWH 1XPEHU RI SRLQWV 0D[LPXP 7UDQVIRUP 6L]H PHWHUV VLQJOH SDVVf PHWHUV GRXEOH SDVVf FP VLQJOH SDVVf FPf GRXEOH SDVVf +] QRW GLJLWDOO\ ILOWHUHG +] ZLWK GLJLWDO ILOWHULQJ UHGXFHG WR +] WKURXJKSXW SRLQWV f

PAGE 38

7DEOH +HOLXP LQGXFWLYHO\ FRXSOHG SODVPD RSHUDWLQJ FRQGLWLRQV 6RXUFH 8QLW )UHTXHQF\ )RUZDUG 3RZHU 5HIOHFWHG 3RZHU 3ODVPD *DV )ORZ 1HEXOL]HU )ORZ 6DPSOH 8SWDNH 5DWH 9LHZLQJ +HLJKW DERYH ORDG FRLO $SHUWXUH WR )76 5) 3ODVPD7KHUP N: XQLW 0+= N: : / PLQ / PLQ P/ PLQ PP PP

PAGE 39

+HOLXP LQGXFWLYHO\ FRXSOHG SODVPD 3 %5$1&+ 9DOXH -sOf -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O :$9(180%(5 FPf

PAGE 40

7DEOH f§FRQWLQXHG -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O 5 %5$1&+ 9DOXH -sf :DYHQXPEHU -O -O -O -O -O -O -O -O -O -O

PAGE 41

7DEOH f§FRQWLQXHG -f§ -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O -O

PAGE 42

&+$37(5 *(1(5$7,21 2) $ /2: )/2: $70263+(5,& 35(6685( 1(21 ,&3 ,QWURGXFWLRQ $ 0+] 1HRQ ,QGXFWLYHO\ &RXSOHG 3ODVPD 1H,&3f KDV EHHQ JHQHUDWHG LQ WKLV ODERUDWRU\ $UJRQ KDV ORQJ EHHQ XVHG DV WKH SODVPD JDV IRU DQ ,&3 $EGDOODK DQG 0HUPHW f DQG &KDQ HW DOf KDYH JHQHUDWHG DWPRVSKHULF SUHVVXUH KHOLXP SODVPDV LQ LQGXFWLYHO\ FRXSOHG SODVPD ,&3f XQLWV :DOWHUV HW DO f JHQHUDWHG D UHGXFHG SUHVVXUH WRUUf 0+] QHRQ ,&3 IRU HOHFWURGHOHVV GLVFKDUJH ODPS VWXGLHV 7KLV LV WKH ILUVW NQRZQ UHSRUW RI DQ DWPRVSKHULF 1H,&3 7KH WRUFK GHVLJQ KDV EHHQ XVHG IRU KHOLXP ,&3 VWXGLHV DQG LV GHVFULEHG E\ 0DQQLQJ HW DOf 7KH QHRQ ,&3 LV EHLQJ LQYHVWLJDWHG DV D SRVVLEOH DWRPL]DWLRQ VRXUFH IRU DGDSWDWLRQ WR D FRPPHUFLDO LQGXFWLYHO\ FRXSOHG SODVPDPDVV VSHFWURPHWU\ LQVWUXPHQW ([SHULPHQWDO 7KH 1H,&3 LV JHQHUDWHG ZLWK D 0+] 5) 3ODVPD 3URGXFWV +)3 N: XQLW &KHUU\ +LOO 1-f 7KH IRUZDUG SRZHU IRU WKH ,&3 LV : ZLWK OHVV WKDQ : UHIOHFWHG f 7KH ORZ IORZ PLQf ODPLQDU IORZ DWPRVSKHULF SUHVVXUH WRUFK LV GHVFULEHG E\ 0DQQLQJ HW DOf ,W XWLOL]HV DQ H[WHQGHG TXDUW] VOHHYH WR PLQLPL]H LPSXULWLHV 1 + HWFf IURP WKH

PAGE 43

DWPRVSKHUH HQWHULQJ WKH SODVPD 7KH QHEXOL]DWLRQ V\VWHP HPSOR\HG LV WKH IULWWHG GLVN QHEXOL]HU f 7KH VDPSOH LV SXPSHG WR WKH IULWWHG GLVN ZLWK D VWDQGDUG SHULVWDOWLF SXPS 7KH IORZ UDWH RI WKH SODVPD JDV LV /PLQ ZLWK D PLQLPXP RI /PLQ DQG D PD[LPXP RI /PLQ SRVVLEOH 7KH QHEXOL]HU JDV IORZ UDWH LV /PLQ ZLWK D VDPSOH XSWDNH UDWH RI P/PLQ 7KH QLWURJHQ LV VHHGHG LQ DW D UDWH RI P/PLQ 7KH QHRQ JDV LV VXSSOLHG E\ 8QLRQ &DUELGH/LQGH GLYLVLRQ JDV GLVWULEXWRUV $OEXTXHUTXH 10f /RV $ODPRV )RXULHU 7UDQVIRUP 6SHFWURPHWHU 7KH /RV $ODPRV )RXULHU 7UDQVIRUP 6SHFWURPHWHU f D QDWLRQDO IDFLOLW\ IRU KLJK UHVROXWLRQ DWRPLF DQG PROHFXODU VSHFWURVFRS\ LV XVHG LQ WKLV H[SHULPHQW IRU PHGLXP UHVROXWLRQ FPnf VSHFWUD LQ WKH QHDU ,5 YLVLEOH DQG XOWUDYLROHW FPn WR FPnf $OXPLQXP FRDWHG TXDUW] EHDPVSOLWWHUV DUH HPSOR\HG DQG WKH GHWHFWRUV XWLOL]HG DUH VLOLFRQ 3,1 GLRGHV 7KH QXPEHU RI SRLQWV WDNHQ IRU HDFK UXQ LV r DQG WKH WUDQVIRUP VL]H LV f f SRLQWV 7KH UXQ P ZKLFK RQO\ ZDWHU LV LQWURGXFHG LQWR WKH SODVPD FRQVLVWV RI VFDQV PLQXWHV GDWD DFTXLVLWLRQ WLPHf DQG WRRN VFDQV LQ WKH UXQ LQWURGXFLQJ ERWK ZDWHU DQG 1 PLQXWHV GDWD DFTXLVLWLRQ WLPHf 7KH QXPEHU RI VFDQV LV OLPLWHG E\ WKH FXUUHQW DYDLODELOLW\ RI UHVHDUFK JUDGH QHRQ 7KH FRVW RI QHRQ LV WKH OLPLWLQJ IDFWRU

PAGE 44

5HVXOWV DQG 'LVFXVVLRQ 1HRQVn LRQL]DWLRQ SRWHQWLDO H9f LV KLJKHU WKDQ WKDW RI DUJRQ H9f EXW ORZHU WKDQ KHOLXP H9f 7KLV VKRXOG RIIHU EHWWHU GHWHFWLRQ OLPLWV IRU YDULRXV QRQPHWDOV HJ ) &O %U HWFf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r f QPf DQG 2+f QPf LV PDGH 7KH 1 EDQGKHDG KDV D 615 RI DQG WKH 5 EUDQFKHV XVHG LQ GLDJQRVWLFV DUH QRW GLVFHUQLEOH PDNLQJ UHOLDEOH PHDVXUHPHQWV LPSRVVLEOH 7KH 2+f EDQG LV QRW SUHVHQW DW DOO LQ WKH VSHFWUXP 7KH +% OLQH LV XVHG WR PHDVXUH WKH HOHFWURQ QXPEHU GHQVLW\ f DW [O FPn 7KH IDFW WKDW ERWK VSHFLHV 2+ 1f ZKLFK DUH VWURQJ 89 HPLWWHUV LQ SODVPDV DQG GLVFKDUJHV GR QRW KDYH D UHDGLO\ GHWHFWDEOH HPLVVLRQ UDWH FRXSOHG ZLWK WKH ORZ HOHFWURQ GHQVLW\ PHDVXUHPHQW [O FPf DQG WKH ORZ LQSXW SRZHU :f JLYH UHDVRQ WR EHOLHYH WKH 1H,&3 LQ WKLV VHWXS LV FRRO FRPSDUHG WR WKH DUJRQ DQG KHOLXP

PAGE 45

SODVPDV f UXQ DW KLJKHU SRZHUV %DQGV IURP WKH ILUVW f f + f f SRVLWLYH V\VWHP %=J $ RI 1 DUH REYLRXV LQ WKH QHDU ,5 DQG YLVLEOH 7KLV WUDQVLWLRQ FPn FPnf UHTXLUHV OHVV HQHUJ\ IRU H[FLWDWLRQ WKDQ 2+f FP f &RQFOXVLRQ $Q DWPRVSKHULF SUHVVXUH QHRQ ,&3 LV JHQHUDWHG ,Q WKH FXUUHQW VHWXS LW RSHUDWHV UHOLDEO\ DW : IRUZDUG SRZHU 7KH 1H,&3 LV EHLQJ SXUVXHG DV DQ DWRPL]DWLRQ VRXUFH IRU D FRPPHUFLDO ,&306 VHWXS $ JUHDWHU DPRXQW RI GLDJQRVWLFV LV QRW IHDVLEOH GXH WR WKH H[SHQVH RI UXQQLQJ D 1H,&3 $ VLQJOH 1H OLQH QPf ZDV XVHG WR GHWHUPLQH WKH 'RSSOHU WHPSHUDWXUH 7KH YDOXH .f ZDV REWDLQHG E\ GHFRQYROXWLQJ WKH OLQH SURILOH ZLWK D 9RLJW SURILOH 8VLQJ D 1H,&3 DV DQ LRQ VRXUFH IRU ,&306 VWXGLHV VKRXOG RIIHU WKH DGYDQWDJH RI IUHHLQJ YDULRXV PDVV YDOXHV WKDW DUH QRW DFFHVVLEOH ZLWK DUJRQ DV WKH SODVPD JDV

PAGE 46

&+$37(5 $ 121)/2:,1* 9$5,$%/( *$6 ,1'8&7,9(/< &283/(' 3/$60$ $6 $ /,*+7 6285&( )25 +,*+ 5(62/87,21 63(&7526&23< ,QWURGXFWLRQ 7HQ VHOIFRQWDLQHG 0+] ,QGXFWLYHO\ &RXSOHG 3ODVPDV ,&3f DUH JHQHUDWHG LQ D VWDWLF WRUFK 7KH WRUFK RIIHUV VSHFWURVFRSLVWV D FORVHG V\VWHP ORZ SUHVVXUH WR 7RUUf YDULDEOH SRZHU FXUUHQWO\ XS WR N:f SODVPD 'RSSOHU WHPSHUDWXUHV 1D QPf PHDVXUHG DUH DSSUR[LPDWHO\ WLPHV WKDW RI HOHFWURGHOHVV GLVFKDUJH ODPSV ('/Vf RU +ROORZ &DWKRGH /DPSV +&/Vf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

PAGE 47

([SHULPHQWDO )LJXUH VKRZV WKH WRUFK GHVLJQ XWLOL]HG LQ WKLV VWXG\ 7KH TXDUW] WXEH KDV D PP 2' PP ,' DQG D FP OHQJWK $ TXDUW] ZLQGRZ LV IXVHG WR WKH YLHZLQJ HQG RI WKH WXEH ,Q VRPH FLUFXPVWDQFHV XVLQJ JURXQGHG FOLSV KHOSV WKH SODVPD V\PPHWULFDOO\ ILOO WKH HQWLUH WXEH (QG RQ YLHZLQJ LQFUHDVHV WKH OXPLQRVLW\ RI WKH VRXUFH 7KH IRXU WXUQ FRLO LV JROG SODWHG LQFK FRSSHU WXELQJ 7KH ,&3 LQVWUXPHQW LV D 5) 3ODVPD 3URGXFWV +)3 N:f XQLW 2IWHQ WKH SODVPDV DUH LQLWLDWHG E\ LQWURGXFLQJ WKH ILHOG 5) 21f DQG VORZO\ LQFUHDVLQJ WKH SRZHU ,Q LQVWDQFHV ZKHUH WKLV LV QRW SRVVLEOH W\SLFDOO\ 7RUUf D WHVOD FRLO GLVFKDUJH LV VHQW WKURXJK RQH RI WKH JURXQGHG FOLSV WR LJQLWH WKH SODVPD $ YDFXXP UDFN HTXLSSHG ZLWK D OLTXLG QLWURJHQ FROG ILQJHU D URXJKLQJ SXPS DQG D GLIIXVLRQ SXPS LV XWLOL]HG IRU HYDFXDWLQJ WKH FHOO DQG ILOOLQJ LW ZLWK WKH GHVLUHG JDV 7KH FHOO LV HYDFXDWHG WR n 7RUU EDFN ILOOHG ZLWK WKH GHVLUHG SODVPD JDV HYDFXDWHG WR 7RUU DQG ILOOHG WR WKH GHVLUHG SUHVVXUH 7KH FHOO LV GHPRXQWDEOH DQG HDVLO\ LQVHUWHG LQWR WKH ,&3 FRLO $ EDOOMRLQW FRQQHFWV WKH UDFN DQG WKH WRUFK $V D SUHFDXWLRQ QLWURJHQ LV EORZQ DFURVV WKH TXDUW] WRUFK IRU FRROLQJ SXUSRVHV 3UHVVXUHV JLYHQ DUH WKH ILOO YDOXHV DW URRP WHPSHUDWXUH 3RZHUV UHFRUGHG DUH WKH IRUZDUG PLQXV WKH UHIOHFWHG YDOXHV 7KH /RV $ODPRV )RXULHU 7UDQVIRUP 6SHFWURPHWHU LV XVHG WR

PAGE 48

UHFRUG WKH VSHFWUD DW KLJK FPnf UHVROXWLRQ ZLWK VFDQV 7KH GHWHFWRUV DUH 6LOLFRQ 3,1 GLRGHV DQG WKH EHDPVSOLWWHUV DUH DOXPLQXP FRDWHG TXDUW] 7KH VSHFWUDO ZLQGRZ WUDQVIRUPHG LV IURP FPn WR FPn SRLQWV DUH DFTXLUHG DQG ]HUR ILOOHG WR ff SRLQWV $GGLWLRQDO SDUDPHWHUV RI WKH /$)76 DUH VXSSOLHG E\ 0DQQLQJ HW DO f DQG 3DUVRQV DQG 3DOPHU f $OO JDVHV XVHG DUH FHUWLILHG UHVHDUFK JUDGH bf 5HVXOWV )LJXUH VKRZV WKH 1D QPf OLQH SURILOH LQ D ;H ,&3 6RGLXP DQG OLWKLXPf LPSXULWLHV DUH SUHVHQW LQ PRVW RI WKH VSHFWUD REWDLQHG 7KH\ DUH DWWULEXWHG WR LPSXULWLHV LQ WKH TXDUW] 7KH VRGLXP OLQH LV XVHG IRU WKH GHWHUPLQDWLRQ RI WKH 'RSSOHU WHPSHUDWXUHV LQ HDFK SODVPD 9DOXHV UDQJHG IURP [HQRQ WRUU :f DW WR ERWK FDUERQ GLR[LGH WRUU :f DQG R[\JHQ WRUU :f DW 7KHVH YDOXHV DUH DSSUR[LPDWHO\ WLPHV WKDW RI +&/V RU PLFURZDYH ('/V DVVXPLQJ D QRUPDO FXUUHQW RU SRZHU LQSXW 7R DYRLG KDYLQJ WKH VDPSOH DEVRUE RQ WKH FRROHU HQGV RI WKH WRUFK WKH OHQJWK RI WKH WRUFK PD\ KDYH WR EH UHGXFHG VR WKH WXEH OHQJWK ILWV HQWLUHO\ ZLWKLQ WKH FRLO $GGLQJ VPDOO TXDQWLWLHV RI D GHVLUHG VSHFLHV HJ W\SLFDOO\ LQ WKH PLFURJUDP UDQJHf DOORZV WKH WRUFK WR EH XVHG DV D VWDEOH UHVHUYRLU RI DWRPLF OLQHV 7KH VDPH LV SRVVLEOH IRU PROHFXOHV VWDEOH LQ WKH SODVPD HQYLURQPHQW

PAGE 49

7KH ORZ SUHVVXUH LQHUW JDV +H 1H $U .U ;Hf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f ,Q SDUWLFXODU WKH LQWHQVLW\ SUHFLVLRQ ZDYHQXPEHU SRVLWLRQ SUHFLVLRQ DQG DFFXUDF\ DQG UHVROXWLRQ FDSDELOLWLHV DUH SDUWLDOO\ UHDOL]HG 7KHUH LV QR UHDVRQ WR EHOLHYH WKLV FHOO FRXOG QRW JHQHUDWH RWKHU SODVPDV $WRPLF VSHFLHV ZKLFK DUH JDVHRXV DW ORZ SUHVVXUH PHUFXU\ UDGRQ YRODWLOH VDOWVf GLDWRPLF JDVHV 12 &2 &O ), HWFf SRO\DWRPLF JDVHV 12 62 &+ RWKHU K\GURFDUERQV HWFf VKRXOG UHDGLO\ IRUP ORZ SUHVVXUH SODVPDV LQ WKLV VHWXS 7KH HWK\OHQH,&3 DQG RWKHU K\GURFDUERQV V\VWHPV PD\ EH RI LQWHUHVW WR WKRVH VWXG\LQJ WKH GLDPRQG V\QWKHVLV PHFKDQLVP LQ D 5)SODVPD HQYLURQPHQW f &ORVHG FHOO K\GURJHQ,&3 PLQLPL]HV WKH GDQJHUV RI ZRUNLQJ ZLWK D IORZLQJ V\VWHP 7KLV K\GURJHQ,&3 DOORZV VFLHQWLVWV WR VWXG\

PAGE 50

DWRPLF LQWHUDFWLRQV LQ D V\VWHP VLPLODU WR WKH VXUIDFH RI FRROHU VWDUV )XVLQJ TXDUW] ZLQGRZV WR ERWK HQGV RI WKH SODVPD WXEH DOORZV D YDULHW\ RI DEVRUSWLRQ PHDVXUHPHQWV WR EH FRQFHLYHG )RU H[DPSOH LQWUDFDYLW\ DEVRUSWLRQ f ZLWK DQ ,&3 LV SRVVLEOH ZLWK WKH DGYDQWDJH RI D ORQJHU SDWK OHQJWK VPDOOHU VDPSOH VL]H DQG DQ LQFUHDVHG UHVLGHQFH WLPH SHU DWRP YHUVHV WKH WUDGLWLRQDO IORZLQJ ,&3 WRUFK $ FORVHG V\VWHP ,&3 LV FXUUHQWO\ EHLQJ LQYHVWLJDWHG DV DQ LRQ VRXUFH IRU PDVV VSHFWURPHWU\ 2QH HQG RI WKH WXEH LV GLUHFWO\ LQWHUIDFHG WR WKH PDVV VSHFWURPHWHU ZLWK WKH IORZ UHJXODWHG E\ D JDWHG YDOYH 7KH VDPSOH LV LQMHFWHG WKURXJK WKH V\ULQJH SRUW WR PLFUROLWHUf EHIRUH LJQLWLRQ 7KH VROYHQW HYDSRUDWLRQ LV HQKDQFHG ZLWK WKH YDFXXP OHDYLQJ RQO\ VDPSOH LQ WKH TXDUW] WXEH :LWK SODVPD LJQLWLRQ WKH VDPSOH VSXWWHUV LQWR WKH SODVPD :H KDYH IRXQG WKLV WR ZRUN TXLWH ZHOO IRU WKH QRQPHWDOV DQG VRIWHU PHWDOV 7KH YDULDEOH JDV VWDWLF SODVPD IRU ,&306 VKRXOG RIIHU PDQ\ DGYDQWDJHV RYHU WKH IORZLQJ DUJRQ V\VWHP ([SHQVLYH JDVHV HJ QHRQ NU\SWRQ [HQRQf FDQ EH URXWLQHO\ XVHG GXH WR ORZ JDV FRQVXPSWLRQ 7KHVH LQHUW JDVHV PD\ EH XVHG WR IUHH PDVV UDQJHV QRW DFFHVVLEOH ZLWK DUJRQ HJ VDPSOH RYHUODSV ZLWK $U1 $U+r $U2+ HWFf 3URYLGHG H[SHULPHQWDO FRQGLWLRQV DUH LGHQWLFDO SRZHU SUHVVXUHf WKH WHPSHUDWXUH DQG HOHFWURQ GHQVLW\ RI WKH LQHUW JDV SODVPD +H 1H $U .U ;Hf ZLOO KDYH DQ LQYHUVH UHODWLRQVKLS ZLWK WKH LRQL]DWLRQ SRWHQWLDO 7KDW LV WKH

PAGE 51

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f DW KLJKHU SRZHUV 7KH ILUHEDOO UHVHPEOHV WKH DWPRVSKHULF SUHVVXUH DUJRQ ,&3 LQ DSSHDUDQFH 7KLV ILUHEDOO FRXOG QRW EH VXVWDLQHG IRU H[WHQGHG SHULRGV RI WLPH RU WKH TXDUW] ZRXOG JORZ UHG DQG HYHQWXDOO\ PHOW ,I D FRQWLQXRXV YDFXXP LV DSSOLHG WKDW PDLQWDLQV D FRQVWDQW SUHVVXUH WKH ILUHEDOO FDQ EH PDLQWDLQHG IRU H[WHQGHG SHULRGV RI WLPH (FNHUW f JLYHV DQ H[FHOOHQW WKHRUHWLFDO GLVFXVVLRQ RI HOHPHQWDO GHWHFWLRQ LQ D VWDWLF LQGXFWLRQ SODVPD +LV FDOFXODWLRQV SUHGLFW D OLPLW RI GHWHFWLRQ IRU PDJQHVLXP QP OLQHf RI OHVV WKDQ J LQ D QRQIORZLQJ DUJRQ,&3 :LWK VPDOO PRGLILFDWLRQV WR WKH WRUFK WKH FORVHG FHOO ,&3 PD\ KDYH

PAGE 52

DQDO\WLFDO XVHIXOQHVV 7KLV DQDO\WLFDO UHVHDUFK ZRXOG EH VLPLOLDU WR ZRUN SUHYLRXVO\ FDUULHG RXW ZLWK PLFURZDYH HOHFWURGHOHVV GLVFKDUJH ODPSV &RQFOXVLRQ 7KH VWDWLF SODVPD RXWOLQHG LQ WKLV QRWH GHVFULEHV DQ DWRP DQG PROHFXOH UHVHUYRLU IRU VSHFWURVFRSLF VWXGLHV f ,W LV DOVR EHLQJ SXUVXHG DV DQ LRQ VRXUFH IRU PDVV VSHFWURPHWU\ f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

PAGE 53

)LJXUH &ORVHG FHOO ,&3 WRUFK GHVLJQ

PAGE 55

)LJXUH /LQH SURILOH RI 6RGLXP' QPf LQ ;H,&3

PAGE 56

,17(16,7< n L" R i i i i *?

PAGE 57

&+$37(5 2%6(59$7,21 2) /,1( 6+,)76 $1' /,1( 352),/(6 ,1 $1 ,1'8&7,9(/< &283/(' $5*21 3/$60$ ,QWURGXFWLRQ /LQH VKLIWV LQ DQDO\WLFDO LQGXFWLYHO\ FRXSOHG SODVPDnV ,&3f DUH D IXQGDPHQWDO SDUDPHWHU WKDW KDV QRW EHHQ VWXGLHG LQ DQ\ JUHDW GHWDLO 7KLV SDSHU ZLOO JLYH DQ LQWURGXFWLRQ WR OLQH VKLIWV PHDVXUHG DQG WUHQGV REVHUYHG )RU PRUH GHWDLOHG WUHDWPHQWV RI SUHVVXUH HIIHFWV LQFOXGLQJ OLQH VKLIWV WKH UHDGHU LV GLUHFWHG WR 7KRUQ f RYHUYLHZf $ONHPDGH HW DO f LQ IODPHVf *ULHP f DQG %UHHQH f DQG UHIHUHQFHV ZLWKLQ 5HYLHZV E\ .RQMHYLF DQG :HLVH f DQG E\ .RQMHYLF DQG 5REHUWV f JLYH QXPHURXV UHIHUHQFHV RQ 6WDUN EURDGHQLQJ DQG OLQH VKLIWV IURP D YDULHW\ RI VRXUFHV /LQH VKLIWV DUH LPSRUWDQW IRU D YDULHW\ RI UHDVRQV $QDO\WLFDO GHWHUPLQDWLRQ RI HOHPHQWV LQ D FRPSOH[ VDPSOH LV RIWHQ SODJXHG PLVLGHQWLILFDWLRQ RI WKH VSHFLHV EHLQJ VWXGLHG ,Q SK\VLFDO VSHFWURVFRS\ ,&3V DUH EHLQJ XVHG IRU WKH PHDVXUHPHQW RI WUDQVLWLRQ SUREDELOLWLHV .QRZOHGJH RI OLQH VKLIW YDOXHV DUH LPSRUWDQW WR SURSHUO\ PHDVXUH HQHUJ\ OHYHOV WUDQVLWLRQV HWF :KDOLQJ DQG %UDXOW f XVHG DQ ,&3 LQ FRQMXQFWLRQ ZLWK WKH .LWW 3HDN )RXULHU 7UDQVIRUP 6SHFWURPHWHU WR PHDVXUH WKH WUDQVLWLRQ SUREDELOLWLHV RI 0R ,f OLQHV

PAGE 58

WKH\ LQWURGXFH 0R DV 0R&2f HOLPLQDWLQJ DQ\ HIIHFWV ZDWHU KDV RQ WKH VSHFWUXP 7KH\ REVHUYHG OLQH VKLIWV RI 0R DV JUHDW DV FP 7KH /RV $ODPRV )RXULHU 7UDQVIRUP 6SHFWURPHWHU /$)76f LV DQ LGHDO LQVWUXPHQW WR PHDVXUH VPDOO VSHFWUDO VKLIWV ,WV KLJK UHVROXWLRQ FDSDELOLWLHV HDVLO\ UHVROYH SK\VLFDO OLQH ZLGWKV HPLWWHG E\ KROORZ FDWKRGH ODPSV +&/f FPnf DQG WKH LQGXFWLYHO\ FRXSOHG SODVPD FPf 7KH ZDYHQXPEHU SUHFLVLRQ RI WKH /$)76 FDQ EH DV JUHDW DV SDUW LQ 7KH /$)76 FDQ DFTXLUH ODUJH VSHFWUDO ZLQGRZV WR FPnf DW WKH UHTXLUHG UHVROXWLRQ WR SK\VLFDOO\ UHVROYH WKH ,&3 HPLVVLRQ ,Q VWXG\LQJ OLQH VKLIWV LVRWRSH VKLIWV DQG K\SHUILQH VWUXFWXUH PXVW EH FRQVLGHUHG WR SURSHUO\ VHOHFW D OLQH (OHPHQWV WKDW FOHDUO\ GLVSOD\ HLWKHU HIIHFW DUH DYRLGHG GXH WR WKH FRPSOH[LW\ LQWURGXFHG LQ DWWHPSWLQJ OLQH SURILOH VWXGLHV 7KH VSHFWUXP RI D 3E HPLVVLRQ OLQH VKRZV FRQWULEXWLRQV IURP ‘MQH RQ Q WKUHH LVRWRSHV 3E bf 3E bf DQG 3Ea bf 7KHVH LVRWRSH VKLIWV DUH VKRZQ LQ )LJ 7KLV VSHFWUXP ZDV WDNHQ IURP D ORZ SUHVVXUH +&/ ZLWK D 'RSSOHU WHPSHUDWXUH RI DSSUR[LPDWHO\ UHVXOWLQJ LQ 'RSSOHU EURDGHQHG VSHFWUDO OLQHV KDYLQJ QHJOLJLEOH SUHVVXUH FROOLVLRQDOf EURDGHQLQJ ,QFUHDVLQJ WKH WHPSHUDWXUH WR DQG LQWURGXFLQJ VXEVWDQWLDO DPRXQWV RI FROOLVLRQDO EURDGHQLQJ ZKLFK VLPXODWHV ,&3 FRQGLWLRQV ZLOO FDXVH WKH WKUHH LVRWRSLF OLQHV WR DSSHDU DV RQH DV\PPHWULF EURDGHQHG OLQH 0HDVXULQJ WKH OLQH VKLIW

PAGE 59

EHFRPHV GLIILFXOW EHFDXVH WKH LQGLYLGXDO OLQHV KDYH WR EH GHFRQYROXWHG EHIRUH VSHFWUDO VKLIW PHDVXUHPHQWV DUH SRVVLEOH 7KHUH ZLOO EH XQFHUWDLQW\ LQWURGXFHG LQWR WKH FDOFXODWLRQ E\ WKH GHFRQYROXWLRQ ZKLFK PD\ DIIHFW WKH YDOLGLW\ RI WKH VKLIW PHDVXUHPHQW )LJXUH VKRZV WKH K\SHUIPH VWUXFWXUH RI WDQWDOXP 7D LV WKH SUHGRPLQDQW LVRWRSH RFFXUULQJ QDWXUDOO\ bf VR LVRWRSLF HIIHFWV DUH QHJOLJLEOH :LWK D QXFOHDU VSLQ RI 7D FOHDUO\ H[KLELWV K\SHUILQH VWUXFWXUH .XKQ f JLYHV DQ H[FHOOHQW WUHDWPHQW RI K\SHUILQH VWUXFWXUH DQG LVRWRSH VKLIWV 7D HPLVVLRQ LQ DQ ,&3 KDV VLJQLILFDQW 'RSSOHU DQG SUHVVXUH EURDGHQLQJ RI WKH OLQHV JLYLQJ DV\PPHWULF OLQH SURILOHV ZLWK FRQYROXWHG K\SHUILQH HOHPHQWV PDNLQJ VKLIW PHDVXUHPHQWV YLUWXDOO\ LPSRVVLEOH f f $UJRQ KDV RQH SUHGRPLQDQW LVRWRSH $U bf ZKLFK KDV QR QXFOHDU VSLQ PDNLQJ LW DQ LGHDO FKRLFH IRU OLQH VKLIW VWXGLHV )LJXUH VKRZV D OLQH SURILOH W\SLFDO RI $U HPLVVLRQ DW QP IURP WKH ,&3 ,URQ KDV RQH GRPLQDQW LVRWRSH )H bf ZKLFK KDV QR K\SHUILQH VWUXFWXUH )LJ VKRZV D OLQH SURILOH RI )H DW QP 7KH HIIHFW RI WKH ORZHU DEXQGDQFH LVRWRSHV RI )H RQ WKH OLQH FHQWHU SRVLWLRQ VKLIW PHDVXUHPHQWf DQG WKH IXOO ZLGWK DW KDOI PD[LPXP ):+0 DSDUDPHWHUV ZKHUH D r ;c\;'f DUH FRQVLGHUHG QHJOLJLEOH IRU WKLV VWXG\ .DWR HW DO f PHDVXUHG SURILOHV RI VSHFWUDO OLQHV VWHPPLQJ IURP HOHPHQWV HPLWWHG E\ DQ ,&3 DQG PHDVXUHG ZLWK

PAGE 60

D SUHVVXUHVFDQQLQJ )DEU\3HURW LQWHUIHURPHWHU 7KH UHVXOWV WKHLU VWXG\ DUH VLPLODU LQ DEVROXWH PDJQLWXGH RI VRPH OLQHV EXW RFFDVLRQDOO\ GLIIHU LQ VLJQ 7KH\ GLG QRW FRQVLGHU WKH HIIHFWV RI K\SHUILQH VWUXFWXUH DQGRU LVRWRSH VKLIWV DV FDUHIXOO\ DV ZH GLG %RXPDQV DQG 9UDNNLQJ f SXEOLVKHG ZRUN RQ KLJKUHVROXWLRQ VSHFWURVFRS\ XVLQJ DQ HFKHOOH VSHFWURPHWHU ZKLFK UHIHUV WR OLQH VKLIWV LQ WKH ,&3 LQ FRQQHFWLRQ ZLWK WKH ZDYHOHQJWK FDOLEUDWLRQ RI WKH VSHFWURPHWHU ([SHULPHQWDO 7KH /RV $ODPRV )RXULHU 7UDQVIRUP 6SHFWURPHWHU /$)76f LV XWLOL]HG LQ FRQMXQFWLRQ ZLWK D 0+] 5) 3ODVPD 7KHUP &KHUU\ +LOO 1-f ,&3 3DSHUV E\ 0DQQLQJ HW DO f DQG 3DUVRQV DQG 3DOPHU f JLYH GHWDLOV RI WKH /$)76 )RU DOO VWXGLHV WKH YLHZLQJ KHLJKW DERYH WKH FRLOV LV PP 7KH JDV IORZ UDWHV DUH /PLQ RXWHUf /PLQ LQWHUPHGLDWHf DQG /PLQ QHEXOL]HUf )RU ,&3 VWXGLHV WKH /$)76 ZDV RSHUDWHG DW FP UHVROXWLRQ ZLWK D IUHH VSHFWUDO UDQJH RI FPn )RU WKH +&/ VWXGLHV WKH /$)76 ZDV RSHUDWHG DW FPf UHVROXWLRQ ZLWK D IUHH VSHFWUDO UDQJH RI FPn 7KH GHWHFWRUV XVHG ZHUH 307V 3f RU VLOLFRQ SLQ GLRGHV 7KH VLOLFRQ SLQ GLRGHV ZHUH XWLOL]HG IRU ZRUN LQYROYLQJ $U HPLVVLRQ LQ WKH QHDU ,5 7KH RXWSXW RI WKH OLJKW VRXUFH ,&3 RU +&/f LV IRFXVHG ZLWK XQLW\ PDJQLILFDWLRQ RQ WKH )76 VSHFWURPHWHU DSHUWXUH 7KH I QXPEHU RI WKLV VSHFWURPHWHU DQG WKH ,&3+&/ RSWLFDO

PAGE 61

VHWXSV DUH PDWFKHG 7KH RXWSXW RI D QHRQ ILOOHG $O +&/ LV LPSUHVVHG RQ WKH ,&3 HPLVVLRQ XVLQJ D QHXWUDO GHQVLW\ ILOWHU f EHIRUH WKH IRFXVLQJ RSWLFV )LJXUH VKRZV DQ RXWOLQH RI WKH HQWUDQFH RSWLFV DQG WKH OLJKW VRXUFHV 7KH 1H OLQH DW QP IURP WKH +&/ LV XVHG DV D FDOLEUDWLRQ VWDQGDUG IRU DOO UXQV $OO HPLVVLRQ OLQHV XVHG LQ WKLV VWXG\ +&/ DQG ,&3f KDYH VLJQDO WR QRLVH UDWLRV 615f JUHDWHU WKDQ 7KLV UHTXLUHPHQW UHGXFHG WKH QXPEHU RI OLQHV DYDLODEOH EHWZHHQ WKH WZR VRXUFHV 6RPH VWURQJ OLQHV LQ WKH +&/ ZHUH QRW DOZD\V DV SUHYDOHQW LQ WKH ,&3 DQG YLFH YHUVD ,RQLF OLQHV IRU LURQ DQG DUJRQ GHWHFWHG LQ WKH ,&3 ZHUH QRW REVHUYHG LQ +&/ HPLVVLRQ RU KDG ORZ VLJQDOWRQRLVH UDWLRV UHGXFLQJ WKH DFFXUDF\ DQG SUHFLVLRQ RI WKH ZDYHQXPEHU PHDVXUHPHQW 6L[ )H N:f $U .Zf f 'RSSOHU WHPSHUDWXUHV ZHUH FDOFXODWHG IURP WKH ,&3 HPLVVLRQ OLQH SURILOHV XVLQJ D 9RLJW ILW 7KLV 'RSSOHU WHPSHUDWXUH ZDV WKHQ XVHG IRU DOO FDOFXODWLRQV LQYROYLQJ WKRVH SDUWLFXODU FRQGLWLRQV )RU H[DPSOH DQ )H OLQH ZLWK D KLJK 615 LV FKRVHQ IURP WKH .Z ,&3 GDWD 9RLJW SURILOHV DUH JHQHUDWHG XQWLO RQH LV LGHQWLFDO WR WKH )H ,&3f SURILOH 7KLV DOORZHG XV WR HVWLPDWH WKH FRQWULEXWLRQV IURP *DXVVLDQ 'RSSOHU EURDGHQLQJf DQG /RUHQ]WLDQ FROOLVLRQDO EURDGHQLQJf FRPSRQHQWV 7KH *DXVVLDQ FRPSRQHQW LV WKHQ XVHG IRU DOO OLQH SURILOH FDOFXODWLRQV LQYROYLQJ )H HPLVVLRQ IURP WKH ,&3 DW N: :H PHDVXUHG WKH *DXVVLDQ DQG /RUHQW]LDQ FRQWULEXWLRQV IRU

PAGE 62

RQH $U DQG RQH )H VSHFWUDO OLQH DW HDFK SRZHU OHYHO N:f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f ZHUH PHDVXUHG 6HFRQG $U HPLVVLRQ LQ WKH QHDU LQIUDUHG DQG IDU YLVLEOH ZDV PHDVXUHG 7KLUG SURPLQHQW LRQ OLQHV RI &D 6U DQG %D DW N: ZHUH PHDVXUHG 7KH ILUVW VHFWLRQ UHIHUV WR OLQH VKLIWV DV D IXQFWLRQ RI SRZHU IRU )H OLQHV DQG $U OLQHV LQ WKH ,&3 7KH SRZHU GHSHQGHQFH .Zf LV VKRZQ WR LQFUHDVH 'RSSOHU WHPSHUDWXUH HOHFWURQ GHQVLW\ DQG VKLIW 7KH LURQ GDWD DUH UHIHU WR WKH DQDO\WLFDO ]RQH RI WKH ,&3 ZKHUHDV WKH DUJRQ GDWD UHIHUV WR WKH RXWHU WDQJHQWLDO IORZ )RU WKH DUJRQ DQG LURQ GDWD WKH OLQH SRVLWLRQ LV JLYHQ LQ $QJVWURPV DQG SK\VLFDO SDUDPHWHUV OLQH VKLIW ZLGWK HWFf DUH JLYHQ LQ ZDYHQXPEHUV FPnf /LQH SRVLWLRQV DUH JLYHQ LQ $QJVWURPV

PAGE 63

EHFDXVH WKH UHIHUHQFHV XVHG IRU WUDQVLWLRQ DVVLJQPHQWV OLVW WKHP LQ WKLV PDQQHU 0RVW UHVHDUFK PHDVXULQJ OLQH VKLIWV UHSRUW WKHLU UHVXOWV LQ FPn 7DEOH JLYHV WKH 'RSSOHU WHPSHUDWXUH .f WUDQVLWLRQV W\SHV ZDYHOHQJWKV OLQH VKLIWV 'RSSOHU DQG /RUHQW]LDQ FRQWULEXWLRQV WR WKH OLQH SURILOH DQG WKH D SDUDPHWHU IRU DUJRQ OLQHV DW .Z 7UDQVLWLRQ DVVLJQPHQWV IRU WKH $UWHUP FRQILJXUDWLRQV DUH WDNHQ IURP /L DQG +XPSKUH\ f $ VLQJOH DUJRQ OLQH DW WKLV SRZHU LV ILWWHG ZLWK D 9RLJW SURILOH WR DSSUR[LPDWH WKH 'RSSOHU FRQWULEXWLRQ WR WKH OLQH SURILOH 7KH UHVXOWLQJ 'RSSOHU WHPSHUDWXUH RI LV WKHQ XVHG WR FDOFXODWH WKH 'RSSOHU KDOIZLGWK DQG D SDUDPHWHU IRU DOO DUJRQ OLQHV DW WKLV SRZHU DQG WKH /RUHQ]WLDQ FRQWULEXWLRQ LV FDOFXODWHG ZLWK DQ LWHUDWLYH PHWKRG $Y Y f r $ Y  f r$Y $Y'f ZKHUH $YY LV WKH WRWDO OLQH ZLGWK $Y/ LV WKH /RUHQW]LDQ FRPSRQHQW DQG $Y' LV WKH 'RSSOHU FRPSRQHQW $OO OLQH VKLIW YDOXHV JLYHQ DUH VKLIWHG WRZDUGV WKH UHGSRVLWLYH YDOXHVf 1HJDWLYH YDOXHV DUH EOXH VKLIW 7DEOHV DQG JLYH VLPLODU GDWD IRU $U DW DQG N: UHVSHFWLYHO\ )LJXUH VKRZV WKH UHODWLRQVKLS EHWZHHQ U! WKH DYHUDJH GLVWDQFH RI WKH HOHFWURQ IURP WKH QXFOHXV DQG WKH DYHUDJH VKLIW IRU WKH SV SV DQG SV WUDQVLWLRQV RI DUJRQ DW

PAGE 64

HDFK RI WKH WKUHH SRZHUV 7KH U! YDOXHV IRU S S DQG S OHYHOV DUH HVWLPDWHG XVLQJ U! OfD>Q f@ f ZKHUH D LV WKH %RKU UDGLXV Q LV SULQFLSOH TXDQWXP QXPEHU DQG LV WKH D]LPXWKDO TXDQWXP QXPEHU ,Q 7DEOHV DQG JLYH WKH UHVXOWV IRU )H DW DQG N: UHVSHFWLYHO\ )RU HDFK WDEOH SRZHU OHYHOf D 'RSSOHU WHPSHUDWXUH DQG HOHFWURQ GHQVLW\ DUH FDOFXODWHG 7KH HOHFWURQ GHQVLW\ ZDV FDOFXODWHG XVLQJ WKH OLQH QPf 7KH WUDQVLWLRQ FRQILJXUDWLRQ ZDYHOHQJWK RI WUDQVLWLRQ OLQH VKLIW ):+0 'RSSOHU DQG /RUHQW]LDQ FRQWULEXWLRQV WR WKH OLQH SURILOH DQG WKH DSDUDPHWHU IRU HDFK OLQH SURILOH VWXGLHG KDYH EHHQ PHDVXUHG 7KH LURQ FRQFHQWUDWLRQ ZDV 0JPO IRU DOO UXQV 7KH )H WUDQVLWLRQ DVVLJQPHQWV ZHUH WDNHQ IURP )XKU 0DUWLQ DQG :LHVH f ,Q 7DEOH WKH VSHFWUDO FKDUDFWHULVWLFV RI DGGLWLRQDO DUJRQ OLQHV REWDLQHG ZLWK WKH ,&3 .Zf LQ WKH QHDU,5 DQG IDU YLVLEOH DUH DOVR JLYHQ $ ODUJH QXPEHU RI LQWHQVH $U OLQHV HPLWWHG IURP WKH ,&3 DQG WKH +&/ RFFXU LQ WKLV UHJLRQ ,W FRQWDLQV H[FOXVLYHO\ SV WUDQVLWLRQV ,Q 7DEOH WKH ILQDO VHW RI GDWD SUHVHQWV OLQH VKLIWV RI FDOFLXP EDULXP DQG VWURQWLXP LQ D .Z ,&3 DUH JLYHQ 7KHVH OLQHV DUH XVHG WR LQYHVWLJDWH WKH UHOHYDQFH RI OLQH VKLIWV LQ DQDO\WLFDO VSHFWURVFRS\ :H FRQVLGHU HDFK HOHPHQW

PAGE 65

f f f WR KDYH PLQLPDO LVRWRSH HIIHFWV &D bf 6U bf %D bff WKHUH DUH QR K\SHUILQH VWUXFWXUHV RQ WKH PDMRU LVRWRSHV 7KH HOHFWURQ GHQVLW\ DQG 'RSSOHU WHPSHUDWXUH DUH FRQVLGHUHG WR EH WKH VDPH DV WKRVH LQ WDEOH )H DW N:f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f VWDWH LV WKH VDPH IRU DOO WKUHH WUDQVLWLRQV )URP RXU PHDVXUHPHQWV OLQH VKLIWV VKRXOG KDYH OLWWOH FRQVHTXHQFH WR WKH DQDO\WLFDO VSHFWURVFRSLVW 7KH GDWD IURP &D %D 6U DQG )H VKRZV WKDW WKH DYHUDJH VKLIW LV RQ WKH RUGHU RI FPn /LQH SRVLWLRQV RI WKLV DFFXUDF\ DUH QRW UHDGLO\ PHDVXUDEOH E\ PRVW DQDO\WLFDO V\VWHPV

PAGE 66

)LJXUH 2QOARQH RI 3EV WKUHH LVRWRSHV KDV DQ\ K\SHUILQH VWUXFWXUH 3Eaf ,Q WKH ,&3 WKLV LVRWRSH VWUXFWXUH ZRXOG EH FRQYROXWHG LQWR RQH DV\PPHWULF OLQH SURILOH PDNLQJ OLQH VKLIW PHDVXUHPHQWV RI HDFK OLQH H[WUHPHO\ GLIILFXOW 7RS [ D[LV LV ƒ DQG WKH ERWWRP LV LQ ZDYHQXPEHUV FPf

PAGE 67

,17(16,7< LQ -

PAGE 68

)LJXUH 7KH K\SHUILQH VWUXFWXUH RI 7D WDNHQ IURP D ORZ SUHVVXUH +&/ 7KH KLJKHU SUHVVXUH DQG WHPSHUDWXUH RI WKH ,&3 ZRXOG PDNH OLQH VHOHFWLRQ GLIILFXOW WRS [D[LV LV ƒ DQG WKH ERWWRP LV FP f

PAGE 70

)LJXUH 7\SLFDO OLQH SURILOH RI $U HPLVVLRQ IURP WKH ,&3 WRS [D[LV LV ƒ DQG WKH ERWWRP LV FPf

PAGE 71

,17(16,7<

PAGE 72

)LJXUH /LQH SURILOH RI )H HPLVVLRQ SSPf IURP DQDO\WLFDO ]RQH LQ ,&3 7RS D[LV LV LQ ƒ DQG WKH ERWWRP LV FPn f

PAGE 73

,17(16,7<

PAGE 74

)LJXUH 'LDJUDP RXWOLQLQJ WKH RSWLFDO FRQILJXUDWLRQ RI OLJKW VRXUFHV ,&3 DQG +&/f DQG RSWLFDO FRPSRQHQWV RXWVLGH /$)76

PAGE 75

1(875$/ '(16 79 /(16 +&/ 1R 6 7 IL 1 4 65 W L L 8 L L & 7 & U L $3(5785(

PAGE 76

)LJXUH 7KH UHODWLRQVKLS EHWZHHQ U! WKH DYHUDJH DWRPLF UDGLXV DQG WKH VKLIW LQ WKH OLQH SRVLWLRQ LV GHPRQVWUDWHG IRU N: rf N: f DQG N: $f 7KH U! YDOXH LV IRU WKH XSSHU OHYHO DQG VKRXOG EH PXOWLSOLHG E\ WKH FRQVWDQW D' %RKU UDGLXVf IRU WKH FRUUHFW DSSUR[LPDWLRQ

PAGE 77

/,1( 6+,)7 FPf§f

PAGE 78

7DEOH 7KH WUDQVLWLRQ ZDYHOHQJWK RI WUDQVLWLRQ ; OLQH VKLIW $$6 IXOO ZLGWK DW KDOI PD[LPXP ):+0f $$; 'RSSOHU KDOIZLGWK $' /RUHQW]LDQ KDOIZLGWK $$/ DQG WKH D SDUDPHWHU RI $U HPLVVLRQ IURP D N: ,&3 7KH 'RSSOHU WHPSHUDWXUH LV 3RVLWLYH VKLIW YDOXHV DUH UHGVKLIWHG %HORZ HDFK JHQHUDO WUDQVLWLRQ SV SV SVf WKH DYHUDJH VKLIW DQG VWDQGDUG GHYLDWLRQ LV WDEXODWHG 7UDQVLWLRQ ; $ $} A;7 $$G $$O D $f FP f FP f FP f FP f 3 > f @V > @ S>2@V>@ $9(5$*( 6+,)7 IR Vf s6 }f S >@MV>@ S >@V>@ S >@V>@ S > @ Mf§V > @ S >@Vn >@ R S > @V > @ Sn>@ >Vn >@ R S > @ V > @ S >@V > @ S >@MV>@ S > f @V > M M S > f @ MV > @ M S > @f§V > @ M S >@ Vn>@

PAGE 79

7DEOH FRQWLQXHG S >@ V n >@ S n >@ V n >@ S>@fVn >@ $9(5$*( 6+,)7 I' Vf sf Sn >@f§V>@ Sn >@V>@ $9(5$*( 6+,)7 SVf sf

PAGE 80

7DEOH 7KH WUDQVLWLRQ W\SH ZDYHOHQJWK RI WUDQVLWLRQ OLQH VKLIW $OV IXOO ZLGWK DW KDOI PD[LPXP ):+0f$; 'RSSOHU KDOIZLGWK $' /RUHQW]LDQ KDOIZLGWK $$/ DQG WKH DSDUDPHWHU RI $U HPLVVLRQ IURP N: ,&3 7KH 'RSSOHU WHPSHUDWXUH LV 3RVLWLYH VKLIW YDOXHV DUH UHGVKLIWHG %HORZ HDFK JHQHUDO WUDQVLWLRQ SV SV SVf WKH DYHUDJH VKLIW DQG VWDQGDUG GHYLDWLRQ DUH WDEXODWHG 7UDQVLWLRQ ; $$6 $O[ $$' $$/ D ƒf FPf FPf FPf FPnf S>@V>@ S > @ V > @ $9(5$*( 6+,)7 G Vf sf S>@V>@ S > f @V >@ 3 > f @V > @ S>O @V > @ Sn >@ Vn>@ R 3> @f§V > @ Sn >@ Vn>@ R S > @V > @ S>@V> @ S > @V > @ S > f @f§V>O @M S > @ MV > @ M S>@ V>@ S n > @V n >@

PAGE 81

7DEOH f§FRQWLQXHG S n >@ V >@ S >@ Mf§V >@ S >@ V n >@ $9(5$*( 6+,)7 G ‘Vf sf S n >@ V > @ S n > f @V > @ $9(5$*( 6+,)7 SVf sf r

PAGE 82

7DEOH 7KH WUDQVLWLRQ W\SH ZDYHOHQJWK RI WUDQVLWLRQ OOLQH VKLIW $$6 IXOO ZLGWK DW KDOI PD[LPXP ):+0f A;7 'RSSOHU FRQWULEXWLRQ $' /RUHQW]LDQ FRQWULEXWLRQ K;/ DQG WKH D SDUDPHWHU RI $U HPLVVLRQ IURP D N: ,&3 7KH 'RSSOHU WHPSHUDWXUH LV 3RVLWLYH VKLIW YDOXHV DUH UHGVKLIWHG %HORZ HDFK JHQHUDO WUDQVLWLRQ SV SV SVf WKH DYHUDJH VKLIW DQG VWDQGDUG GHYLDWLRQ DUH WDEXODWHG 7UDQVLWLRQ ; Ls;V N;U $' A;/ D ƒf FPf FPf FPf FPf S>@V>@ S>2@V>@ $9(5$*( 6+,)7 IR Vf sf S>@V>@ S n > f @V > @ S>fV>@ S > ‘ @ Mf§V > @ S n >-6n >2@ S > @V > @ S > f @ V > @f S>@V>@ S > @f§V > @ S >@V>@ S >@V>@ S>@V>@ S>@V>@

PAGE 83

7DEOH f§FRQWLQXHG Sn >@Vr >@ S n >@ aV n >@ S n >@ V n >@ S>@ Vn >@ $9(5$*( 6+,)7 [ fVf sf Sn >@V>@ Sn >@V>@ $9(5$*( 6+,)7 0SVf sf

PAGE 84

7DEOH 7KH WUDQVLWLRQ W\SH ZDYHOHQJWK RI WUDQVLWLRQ ; OLQH VKLIW $OV IXOO ZLGWK DW KDOI PD[LPXP ):+0f $7 'RSSOHU FRQWULEXWLRQ $$' /RUHQW]LDQ FRQWULEXWLRQ es;K DQG WKH DSDUDPHWHU RI )H HPLVVLRQ IURP D N: ,&3 7KH 'RSSOHU WHPSHUDWXUH LV DQG WKH HOHFWURQ GHQVLW\ [ FP 3RVLWLYH VKLIW YDOXHV DUH UHGVKLIWHG )RU WKLV WDEOH WKH DYHUDJH OLQH VKLIW LV FPn ZLWK D VWDQGDUG GHYLDWLRQ RI FPf 7KH FRQFHQWUDWLRQ RI LURQ DVSLUDWHG LQWR WKH ,&3 ZDV [JPO 7UDQVLWLRQ ; $f $ $ FP f $ ,I FP f DDG FP f $$O ? / FP f D D' BR ] ) LL D' BR \ I L D' B R \ G L D' &Q 2 L D' BR \ G L D' B R \ G L D' Br ] 3 L D) R ] L D) r ] L D) r ] L D) R ] L D) R ] L D) r ] L D' BR ] ) L D' Br ] ) L D' BR ] ) L D) BR \ I L

PAGE 85

7DEOH f§FRQWLQXHG D) BR \ I L D) BR \ G L D) R \ G L D) R \ G L D) BR \ G L D' BR ] L D' Br ] L D) fR \ I L D) BR \ I L D) R = L D) R = L

PAGE 86

7DEOH 7KH WUDQVLWLRQ W\SH ZDYHOHQJWK RI WUDQVLWLRQ $ OLQH VKLIW $$6 IXOO ZLGWK DW KDOI PD[LPXP ):+0f $$7 'RSSOHU $$G /RUHQW]LDQ $$/ DQG WKH DSDUDPHWHU RI )H HPLVVLRQ IURP D N: ,&3 7KH 'RSSOHU WHPSHUDWXUH LV DQG WKH HOHFWURQ GHQVLW\ LV [ FPn3RVLWLYH VKLIW YDOXHV DUH UHGVKLIWHG 7KH DYHUDJH VKLIW YDOXH IRU WKLV WDEOH LV FP ZLWK D VWDQGDUG GHYLDWLRQ RI FPn [JPO )H VROXWLRQ ZDV DVSLUDWHG LQWR WKH ,&3 7UDQVLWLRQ $ $$6 $ $nMn $$G $$O D ƒf FPff FPf FPf FPnf D' ULr ] ) ,, D' Br \ I D' BR \ G D' BR \ G D' BR \ G D' R \ G D' BR ] 3 D) R ] D) R = D) R ] D) R ] D) R ] D) r ] DV' R ] ) D' BR ] ) D' R ] ) D) BR \ I

PAGE 87

7DEOH f§FRQWLQXHG D) R \ I L D) A \ G L D) r \ G L D) B R \ G L D) 2L D R L D' R = L D' BR ] L D) Wr \ I L D) BR \ I L D) R ] L D) R ] L

PAGE 88

7DEOH 7KH WUDQVLWLRQ W\SH ZDYHOHQJWK RI WUDQVLWLRQ OLQH VKLIW $$6 IXOO ZLGWK DW KDOI PD[LPXP ):+0f $$7 'RSSOHU FRQWULEXWLRQ $$' /RUHQW]LDQ FRQWULEXWLRQ $$/ DQG WKH DSDUDPHWHU RI )H HPLVVLRQ IURP D N: ,&3 7KH 'RSSOHU WHPSHUDWXUH LV DQG WKH HOHFWURQ GHQVLW\ LV [ FP 3RVLWLYH VKLIW YDOXHV DUH UHGVKLIWHG 7KH DYHUDJH VKLIW IRU DOO OLQHV OLVWHG LQ WKLV WDEOH LV FPn ZLWK D VWDQGDUG GHYLDWLRQ RI FPn 7KH FRQFHQWUDWLRQ RI )H DVSLUDWHG LQWR WKH ,&3 ZDV JPO 7UDQVLWLRQ $ $f $ $D A ? FP f $ $S FP ff DDG FP f FP $$O fnf D D' BR ] ) LL D' R \ I L D' R \ G L D' r \ G L DV' R \ G L D' Br \ G L D' R ] 3 L D) B = L D) R ] L D) r ] L D) R ] L D) r ] L D) R ] L D' r ] ) L D' R ] ) L D' Br ] ) L D) R \ I L D) f§ R \ I L

PAGE 89

7DEOH f§FRQWLQXHG D) f§ r \ D) r \' D) BR \ D) r \ D' R ] D' r ] D) R \) D) R \) D) r = D) R ]

PAGE 90

7DEOH 7KH WUDQVLWLRQ W\SH ZDYHOHQJWK RI WUDQVLWLRQ ; OLQH VKLIW $OV IXOO ZLGWK DW KDOI PD[LPXP ):+0f A;7, 'RSSOHU FRQWULEXWLRQ N;'O /RUHQW]LDQ FRQWULEXWLRQ $;/ DQG WKH DSDUDPHWHU RI $U HPLVVLRQ IURP D N: ,&3 7KH 'RSSOHU WHPSHUDWXUH LV 3RVLWLYH VKLIW YDOXHV DUH UHGVKLIWHG 7UDQVLWLRQ ; $f $OH FP f $$A $ $' $$ FPf FPf FP r D S>@Vn >@f S>@Vn >@ S>@V>@ S >@ V >@ 3>Or@Vn >@M S> @ LaV >@ S>@Vn >2@ S >@ aV n >@ 3>@f§V>@M Sn >@Vn >@ S n > @ V n >@ S>@f§V>@ S > f @ V >@ S > @V > @ S > f @V >@ Sn>@Vn>@ S n >@ V n > @ S > @ V > @ S>@V>@ S > @ V >@

PAGE 91

7DEOH f§FRQWLQXHG Sn R LQ f R Vn >@ S > f @ V>@ S n >@ V>@ Sn > f @n V > @ S n > @ V > @

PAGE 92

7DEOH /LQH VKLIWV $Y RI &D 6U DQG %D LQ .Z ,&3 7KH ZDYHQXPEHU SRVLWLRQV Y IRU KROORZ FDWKRGH ODPS VWDQGDUG QRQVKLIWHGf YHUVHV WKH ,&3 HPLVVLRQ 7KH 'RSSOHU WHPSHUDWXUH LV DQG WKH HOHFWURQ GHQVLW\ LV [ ORfr FP IRU WKH ,&3 :DYHQXPEHU SRVLWLRQV JLYHQ DUH FDOLEUDWHG DJDLQVW D 1H VWDQGDUG IRU DEVROXWH YDOXHV 1HJDWLYH YDOXHV DUH EOXHVKLIWHG (/(0(17 Y +&/f Y ,&3f $YV FPf FPf FPf %DULXP ,, %DULXP ,, &DOFLXP ,, &DOFLXP ,, 6WURQWLXP ,,

PAGE 93

&+$37(5 $ 9$5,$%/( %$1'3$66 ),/7(5 )25 8/75$9,2/(79,6,%/( )285,(5 75$16)250 63(&7526&23< ,QWURGXFWLRQ )RXULHU 7UDQVIRUP 6SHFWURVFRS\ )76f KDV VHYHUDO DGYDQWDJHV RYHU JUDWLQJ DQG SULVP VSHFWURPHWHUV ,QFOXGHG LQ WKHVH DUH ZDYHQXPEHU SUHFLVLRQ DQG DFFXUDF\ LQWHQVLW\ SUHFLVLRQ DQG UHVROYLQJ SRZHU 'HWULPHQWDO WR WKH )7 WHFKQLTXH LV WKH PXOWLSOH[ GLVDGYDQWDJH f ZKLFK DULVHV IURP VKRW DQGRU IOLFNHU QRLVHV DW GHWHFWHG ZDYHOHQJWKV RWKHU WKDQ WKH DQDO\WLFDO ZDYHOHQJWK KLWWLQJ WKH GHWHFWRU FDXVLQJ D GHFUHDVHG VLJQDOWRQRLVH UDWLR FRPSDUHG WR D VLQJOH VOLW VFDQQLQJ GLVSHUVLYH VSHFWURPHWHU V\VWHP 7KLV GLVDGYDQWDJH LV PRUH SURPLQHQW LQ )7899,6 WKHQ LQ )7,5 VWXGLHV ,Q )7 VWXGLHV LQYROYLQJ D FRQWLQXXP VRXUFH WKLV GLVDGYDQWDJH FDQ EHFRPH VHYHUH 7KH ,QGXFWLYHO\ &RXSOHG 3ODVPD ,&3f 'LUHFW &XUUHQW 3ODVPD '&3f DQG DEVRUEDQFH WHFKQLTXHV XWLOL]LQJ D ZKLWH OLJKW DUH H[DPSOHV RI OLJKW VRXUFHV WKDW FDQ PDJQLI\ WKH PXOWLSOH[ GLVDGYDQWDJH ,Q RXU UHVHDUFK f WKH /RV $ODPRV )RXULHU 7UDQVIRUP 6SHFWURPHWHU LV XVHG WR PHDVXUH 6WDUN VKLIWV LQ ZDYHQXPEHU SRVLWLRQV RI DWRPV DQG LRQV IURP D 0+] DWPRVSKHULF SUHVVXUH DUJRQ ,QGXFWLYHO\ &RXSOHG 3ODVPD ,&3f 7KH OLQH

PAGE 94

SRVLWLRQV RI DWRPV DQG LRQV LQ WKH ,&3 DUH FRPSDUHG WR WKH XQVKLIWHG OLQH SRVLWLRQV RI WKH VDPH VSHFLHV LQ D FRPPHUFLDO KROORZ FDWKRGH ODPS +&/f 7R DFKLHYH DGHTXDWH ZDYHQXPEHU SRVLWLRQV SDUW LQ [ f WKH OLQHV DQDO\]HG QHHG JRRG VLJQDOWRQRLVH UDWLRV 615Vf %RWK WKH +&/ DQG ,&3 HPLW QXPHURXV VSHFWUDO OLQHV 7KH ,&3 EHLQJ D KRWWHU VRXUFH KDV FRQVLGHUDEOH LRQ OLQH HPLVVLRQ 7KH +&/ ZLWK 'RSSOHU WHPSHUDWXUHV W\SLFDOO\ WKDW RI WKH ,&3 YV .f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f XVH LQHUW JDVHV KDYLQJ VWURQJ HPLVVLRQ LQ WKH QHDU ,5 UHJLRQ UHVXOWLQJ LQ D VHULRXV GHWHULRUDWLRQ

PAGE 95

RI WKH 615 IRU OLQHV REVHUYHG LQ WKH XOWUDYLROHW DQG YLVLEOH +LUVFKIHOG DQG &KDVH f DQG +LUVFKIHOG DQG 0LODQRYLFK f VXPPDUL]HG VHYHUDO WHFKQLTXHV WKDW FRXOG EH XVHG WR RYHUFRPH SUREOHPV DVVRFLDWHG ZLWK H[FHVV OLJKW LQ )75DPDQ H[SHULPHQWV 2QH RI WKHVH LV WKH XVH RI D QRQVFDQQLQJ VXEWUDFWLYH GLVSHUVLYH GRXEOH PRQRFKURPDWRU 7KLV ILOWHULQJ DSSURDFK KDV EHHQ VKRZQ WR EH D YDOXDEOH WRRO IRU PLFUR5DPDQ ZRUN f :H KDYH DGDSWHG WKLV ILOWHU IRU XVH DV D EDQGSDVV ILOWHU LQ )7FRQWLQXXP VRXUFH ZRUN :H GHPRQVWUDWH WKH IHDVLELOLW\ RI WKLV ILOWHU XVLQJ D ZKLWH OLJKW VRXUFH ([SHULPHQWDO )LJXUH VKRZV WKH FRQILJXUDWLRQ RI WKH H[SHULPHQW $ %RPHP '$ LV XVHG DW ORZ UHVROXWLRQ FPf IRU WKLV VWXG\ 7KH JUDWLQJV *O *f DUH PDQXIDFWXUHG E\ ,QVWUXPHQWDO 6$ %RWK DUH EOD]HG IRU QP JURRYHVPP PP [ PPf ) QXPEHUV RI WKH LQSXW OHQV /f DQG )76 DUH PDWFKHG 7KH ZKLWH OLJKW VRXUFH :/f JUDWLQJV OHQV /,f DQG OHQV /f DUH KHOG FRQVWDQW IRU DOO H[SHULPHQWV 7KH PLUURUV 0O0f OHQV /f DSHUWXUH $f DUH YDULHG IRU VHOHFWLYH SXUSRVHV /, VHUYHV WR FROOLPDWH WKH OLJKW VRXUFH $ PP GLDPHWHUf DQG $ PP GLDPHWHUf DUH FLUFXODU DSHUWXUHV )RU $O ZH XVH DQ DSHUWXUH DV RSSRVHG WR D VOLW WR PDLQWDLQ VRPH RI WKH WKURXJKSXW DGYDQWDJH RI WKH )76 IRU HPLVVLRQ VWXGLHV 6LQFH ZH DUH ORRNLQJ DW UHODWLYHO\ ODUJH EDQGSDVVHV FPf WKH ORVV LQ VSHFWUDO UHVROXWLRQ RI DQ DSHUWXUH

PAGE 96

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f DW JUHDWHU GLVWDQFHV $ IURP *Of 'HFUHDVLQJ WKH DSHUWXUHV $O $f GHFUHDVHG WKH SKRWRQ IOX[ HQWHULQJ WKH )76 DQG WKHUHIRUH GHFUHDVHV WKH PXOWLSOH[ GLVDGYDQWDJH 0RYLQJ $/ SHUSHQGLFXODU WR WKH QHDUO\ FROOLPDWHG OLJKW UHDFKLQJ 0O UHVXOWHG LQ D FKDQJH LQ WKH FHQWUDO ZDYHOHQJWK RI WKH EDQG SDVVHG 7KH GLVWDQFHV EHWZHHQ RSWLFDO FRPSRQHQWV YDULHG VOLJKWO\ GHSHQGLQJ RQ FHQWUDO ZDYHQXPEHU DQG EDQGSDVV VHOHFWLRQ $SSUR[LPDWH GLPHQVLRQV DUH DV IROORZV /O$O FP $O*UDWLQJ FP *UDWLQJ $ FP 00 FP *UDWLQJ / FP

PAGE 97

5HVXOWV )LJXUH VKRZV WZR VSHFWUD VFDQVf RI WKH ILOWHUHG ZKLWH OLJKW VRXUFH ZLWK D FPn EDQGSDVV 7KH VWURQJ QDUURZ OLQH LQ ERWK LV WKH +H1H $f ODVHU OHDNDJH DQG LV XVHG DV D YLVXDO UHIHUHQFH IRU DOLJQPHQW RI WKH RSWLFV $O*O $/00*/$ )RU WKLV GHPRQVWUDWLRQ WKH FHQWUDO ZDYHQXPEHU VKLIWV IURP DSSUR[LPDWHO\ FPn WR FPn )LJXUH VKRZV D QDUURZHU EDQGSDVV DSSUR[LPDWHO\ FPf WKDW YDULHG IURP WR FPn LQ IRXU ZLQGRZV 7KH JUDWLQJV SHUIRUPDQFH IHOO RII DERYH FPn DQG YLVXDO DOLJQPHQW RI WKH RSWLFV LQ )LJXUH EHORZ FPn EHFRPHV GLIILFXOW E\ H\H DQ LQIUDUHG VHQVRU LV QHHGHG WR VHH WKH UDGLDWLRQ 7KLV MXVWLILHG WKH UDQJH FPnf GHPRQVWUDWHG KHUH &RQFOXVLRQ $ EDQGSDVV ILOWHU ZKLFK YDULHG ERWK ZDYHQXPEHU SRVLWLRQ DQG EDQGSDVV DQG SURYLGH D OLPLWHG VSHFWUDO UDQJH LV GHPRQVWUDWHG 7KH EDQGSDVV ILOWHU LV EHLQJ GHYHORSHG WKH ILOWHU IRU )7899,6 VRXUFH VWXGLHV 7KH ILOWHU UHPRYHG XQZDQWHG OLJKW ZKLFK DIIHFWHG WKH TXDOLW\ RI EURDG EDQG PXOWLSOH OLQH VSHFWUD 5HPRYDO RI WKH SURPLQHQW OLQHV RI QHRQ RU DUJRQ DSSHDULQJ LQ WKH QP UDQJH LQ KROORZ FDWKRGH ODPS +&/f HPLVVLRQ LV DOVR RI LQWHUHVW WR XV 7KHVH OLQHV DUH RIWHQ WKH PRVW LQWHQVH LQ +&/ HPLVVLRQ VWXGLHV 2SWLFDO ILOWHUV KDYH WZR GLVDGYDQWDJHV ZKLFK RIWHQ DULVH LQ VWXGLHV RI

PAGE 98

WKLV VRUW )LUVW WKH EDQGSDVV RI WKH RSWLFDO ILOWHU PD\ QRW PDWFK WKH GHVLUHG QHHG HVSHFLDOO\ LQ WKH XOWUDYLROHW 6HFRQG WUDQVPLVVLRQ RI WKHVH ILOWHUV RIWHQ LQFUHDVHV LQ WKH 1,5 DERYH QPf DQG DOORZV WKH ILOO JDV HPLVVLRQ SDUWLFXODUO\ $Uf WR HQWHU WKH )76 2XU DLP LV WR UHSODFH RSWLFDO ILOWHUV ZLWK D VLPSOH ILOWHU WKDW DOORZV YDULDWLRQ LQ ERWK SRVLWLRQ DQG EDQGSDVV RI WKH ILOWHU :H VKRXOG QRWH WKDW D YDULHW\ RI GHULYDWLRQV RI WKLV ILOWHU DUH SRVVLEOH )RU H[DPSOH UHSODFLQJ 0O DQG 0 ZLWK SDUDEROLF PLUURUV ZRXOG DOORZ WKH UHPRYDO RI / 0O DQG 0 ZRXOG FROOLPDWH DQG UHIOHFW WKH UDGLDWLRQ WR HOLPLQDWLQJ WKH QHHG IRU /

PAGE 99

)LJXUH 2XWOLQH RI EDQGSDVV ILOWHU :/ ZKLWH OLJKW $ DSHUWXUH JUDWLQJf

PAGE 100

0LUURU ? /HQV $ :KLWH /LJKW 9 /HQV

PAGE 101

)LJXUH FP EDQGSDVV RI ILOWHUHG ZKLWH OLJKW VRXUFH FHQWHUHG DW FPn WRSf DQG FPn ERWWRPf 6KDUS OLQH $f LV +H1H ODVHU OHDNDJH IURP %RPHP )76 DOLJQPHQW V\VWHP

PAGE 102

,17(16,7< :$9(180%(5

PAGE 103

)LJXUH FPn FPn EDQGSDVV FHQWHUHG DW FPn f f FP f FPn f

PAGE 104

,17(16,7< YR

PAGE 105

&+$37(5 $ 352326(' :$9(180%(5 ,17(16,7< $1' 5(62/87,21 67$1'$5' )25 +,*+ 5(62/87,21 899,6 )285,(5 75$16)250 63(&7526&23< ,QWURGXFWLRQ ,Q WKLV FKDSWHU XVLQJ 5H QPf DV D VWDQGDUG IRU PHDVXULQJ WKH TXDOLW\ RI VSHFWUD REWDLQHG ZLWK D )RXULHU 7UDQVIRUP 6SHFWURPHWHU )76f RSHUDWLQJ LQ WKH XOWUDYLROHW DQG YLVLEOH LV SURSRVHG ,Q DGGLWLRQ WR WKH PXOWLSOH[ DQG WKURXJKSXW DGYDQWDJHV WKUHH DGGLWLRQDO DGYDQWDJHV IRU )7 899,6 DUH UHVROXWLRQ LQWHQVLW\ SUHFLVLRQ DQG DFFXUDF\ DQG ZDYHQXPEHU SUHFLVLRQ DQG DFFXUDF\ 7KH 5H QPf OLQH ZKLFK KDV K\SHUILQH VWUXFWXUH +)6f DQG LVRWRSH VKLIWV ,6f RIIHUV DQ H[FHOOHQW UHIHUHQFH VWDQGDUG IRU PHDVXULQJ WKH DERYH PHQWLRQHG TXDOLWLHV 4XLWH RIWHQ )7899,6 LQVWUXPHQWV FODLP DFKLHYLQJ UHVROXWLRQV $Y FPf EDVHG VLPSO\ RQ PLUURU GLVSODFHPHQW / FPf ZKHUH $Y /f f 7KLV UHODWLRQVKLS GRHV QRW FRQVLGHU IDFWRUV VXFK DV HOHFWURQLF QRLVH SKDVH SUREOHPV LQFRUUHFW DSHUWXUH VHWWLQJ RSWLFDO LPSHUIHFWLRQV DQG PLUURU MLWWHU ZKLFK QHJDWLYHO\ DIIHFW WKH UHVROXWLRQ FODLPHG E\ PLUURU GLVSODFHPHQW ,PSHUIHFWLRQV RI

PAGE 106

WKLV QDWXUH FDQ DOVR DIIHFW ZDYHQXPEHU SUHFLVLRQ DQG DFFXUDF\ DQG LQWHQVLW\ SUHFLVLRQ DQG DFFXUDF\ ([SHULPHQWDO 7KH RXWSXW RI D FRPPHUFLDOO\ DYDLODEOH KROORZ FDWKRGH ODPS +&/f LV XVHG DV WKH VWDQGDUG $Q RSWLFDO ILOWHU UHFHQWO\ GHVFULEHG f LV XVHG WR QDUURZ WKH EDQGSDVV DQG PLQLPL]H WKH PXOWLSOH[ GLVDGYDQWDJH 7KH UHODWLYHO\ ORZ SUHVVXUH ORZ WHPSHUDWXUH HQYLURQPHQW RI WKH +&/ VHUYHG WR PLQLPL]H FROOLVLRQDO DQG 'RSSOHU EURDGHQLQJ UHVSHFWLYHO\ PDNLQJ WKH GHWDLOV RI WKH WUDQVLWLRQ YLVLEOH 7KH /RV $ODPRV )RXULHU 7UDQVIRUP 6SHFWURPHWHU f LV UXQ DW KLJK UHVROXWLRQ IRU WKLV SDUWLFXODU VWXG\ 7KH VSHFWUD SUHVHQWHG KHUH LV DFJXLUHG DW FPn UHVROXWLRQ ZLWK VFDQV 'LVFXVVLRQ 5KHQLXP 6SHFWURVFRSLVWV KDYH VWXGLHG WKH K\SHUILQH VWUXFWXUH DQG LVRWRSH VKLIWV RI 5H f 5H KDV WZR QDWXUDOO\ RFFXUULQJ LVRWRSHV 5H bf DQG O5H bf (DFK RI WKH WZR LVRWRSHV KDYH QXFOHDU VSLQV JUHDWHU WKDQ ]HUR f ZKLFK OHG WR K\SHUILQH VWUXFWXUH %RWK KDYH VLPLODU QXFOHDU PDJQHWLF PRPHQWV QXFOHDU PDJQHWRQVf UHVSHFWLYHO\ 7KH FODULW\ DQG V\PPHWU\ RI WKH +)6 DQG ,6 LQ WKLV WUDQVLWLRQ PDNH LW DQ H[FHOOHQW VHOHFWLRQ DV D VWDQGDUG 7KH OLQH SURILOH RI 5H QPf LV VKRZQ LQ )LJXUH

PAGE 107

7KH ZDYHQXPEHUV IRU WKLV VSHFWUXP DUH OLVWHG LQ )LJXUH 5HVROXWLRQ 7KH KLJK UHVROXWLRQ FDSDELOLWLHV RI DQ )76 FDQ EH UHDGLO\ WHVWHG E\ DWWHPSWLQJ WR VHSDUDWH WKH LVRWRSH VKLIWV IRU HDFK RI WKH K\SHUILQH FRPSRQHQWV 7KHVH LVRWRSH VSOLWWLQJV DUH LQ WKH FPn UDQJH DQG VHUYH DV D JRRG VWDQGDUG 7KH VSOLWWLQJV EHWZHHQ WKH K\SHUILQH FRPSRQHQWV DUH LQ WKH WR FPn UDQJH DQG VHUYH DV D PHDVXUH IRU ORZHU UHVROXWLRQV 7KH IXOO ZLGWK DW KDOI PD[LPXP ):+0f RI WKH LQGLYLGXDO OLQHV LQ WKLV VWXG\ DUH FP ,QWHQVLW\ 7KH LQWHQVLW\ SUHFLVLRQ DQG DFFXUDF\ RI WKH )RXULHU WUDQVIRUP VSHFWURPHWHU FDQ EH PHDVXUHG XVLQJ WKH UHODWLYH LQWHQVLWLHV RI 5H LVRWRSHV 7KH UDWLR b bf RI WKH WZR QDWXUDOO\ RFFXUULQJ LVRWRSHV VHUYH DV WKH VWDQGDUG IRU LQWHQVLW\ SUHFLVLRQ DQG DFFXUDF\ 7KH UHODWLYH LQWHQVLW\ ,f RI WKH LQGLYLGXDO LVRWRSH VKLIWV LV GLUHFWO\ UHODWHG WR WKH SHUFHQW QDWXUDO DEXQGDQFH ,,f bbf f 7KH LVRWRSH VKLIWV DUH VHSDUDWHG E\ RQO\ FPn WKHUHIRUH KLJK UHVROXWLRQ FDSDELOLWLHV FPnf DUH QHHGHG )RU WKH KROORZ FDWKRGH ODPS WKH LQWHQVLW\ YDULHV b DV D IXQFWLRQ RI WLPH DQG VR HUURUV LQ WKLV UDQJH DUH FRQVLGHUHG DFFHSWDEOH 6RXUFHV ZLWK VPDOOHU VLJQDO YDULDWLRQV

PAGE 108

ZLOO SURGXFH VPDOOHU HUURUV )RU WKH WZR PRVW LQWHQVH OLQHV FPn FPnf DQG WKH DFFRPSDQ\LQJ VKLIWV IRU WKH OHVV DEXQGDQW LVRWRSH FPn FP ff WKHUH LV DQ DFFHSWDEOH V\VWHPDWLF HUURU RI b 7KH /$)76 KDV D b LQWHQVLW\ DFFXUDF\ OLPLW f :DYHQXPEHU 3RVLWLRQV 6SHFWURVFRSLVWV URXWLQHO\ XVH )RXULHU 7UDQVIRUP 6SHFWURPHWHUV IRU KLJK ZDYHQXPEHU SUHFLVLRQ DQG DFFXUDF\ 6HYHUDO VWXGLHV f KDYH WDNHQ DGYDQWDJH RI WKH KLJK ZDYHQXPEHU DFFXUDF\ RI WKH /$)76 7KLV DFFXUDF\ KDV W\SLFDOO\ EHHQ JUHDWHU WKDQ SDUW LQ 7KHUH DUH WZR FRUUHFWLRQ IDFWRUV f LQFRUSRUDWHG LQ WKHVH W\SHV RI VWXGLHV Df DSHUWXUH FRUUHFWLRQ Ef VOLJKW PLVDOLJQPHQW RI LQVWUXPHQW 7KH\ DGKHUH WR WKH IROORZLQJ IRUPDW Dn D > QUf [H @ f ZKHUH 6, U') VROLG DQJOH VUf GLDPHWHU RI WKH DSHUWXUH FPf ) IRFDO OHQJWK RI FROOLPDWLQJ RSWLFV FPf D PHDVXUHG ZDYHQXPEHU FPf FWn FRUUHFWHG ZDYHQXPEHU FPf ; LQVWUXPHQWDO FDOLEUDWLRQ FPf H DYHUDJH ZDYHQXPEHU FPf

PAGE 109

,Q D SUHYLRXV VWXG\ f WKH LQVWUXPHQWDO FDOLEUDWLRQ ;f ZDV FDOFXODWHG LQ WKH IROORZLQJ PDQQHU 7KH HPLVVLRQ IURP D XUDQLXP +&/ ZDV VWXGLHG ZLWK WKH /$)76 7ZHQW\ ILYH OLQHV ZHUH PHDVXUHG LQ WKH QP UDQJH DQG FRPSDUHG DJDLQVW YHU\ DFFXUDWH SDUW LQ Vf ZDYHQXPEHU YDOXHV IURP D XUDQLXP DWODV f :LWK QR FRUUHFWLRQ IDFWRUV WKH YDOXHV ZHUH UHGVKLIWHG FP :LWK RQO\ WKH DSHUWXUH FRUUHFWLRQ ILUf DSSOLHG WKLV OHIW DQ DYHUDJH UHGVKLIW RI FPn ZLWK D VWDQGDUG GHYLDWLRQ RI FPf 7KLV YDOXH FPnf LV WKH LQVWUXPHQWDO FDOLEUDWLRQ [ff 7KH DYHUDJH ZDYHQXPEHU H LV VLPSO\ WKH DYHUDJH RI WKH ZDYHQXPEHUV EHLQJ VWXGLHG )RU WKH /$)76 WKH PLVDOLJQPHQW LV W\SLFDOO\ VPDOOHU WKDQ WKH DSHUWXUH FRUUHFWLRQ IDFWRU 7KH FRUUHFWLRQ IDFWRUV FRPELQHG DUH VPDOO URXJKO\ [ f EXW PXVW EH FRQVLGHUHG IRU KLJK ZDYHQXPEHU DFFXUDF\ DQG SUHFLVLRQ 7DEOH OLVWV WKH FRUUHFWHG YDOXHV IRU WKH 5H OLQHV 7KHVH YDOXHV DUH FRQVLGHUHG DFFXUDWH WR EHWWHU WKDQ SDUW LQ &RQFOXVLRQ $Q LQWHQVLW\ ZDYHQXPEHU DQG UHVROXWLRQ VWDQGDUG IRU )RXULHU 7UDQVIRUP 6SHFWURVFRS\ LQ WKH YLVLEOH LV SUHVHQWHG ,W LV FKRVHQ WR HYDOXDWH WKH VSHFWURPHWHUnV SHUIRUPDQFH FRQFHUQLQJ UHVROXWLRQ LQWHQVLW\ SUHFLVLRQ DQG DFFXUDF\ DQG ZDYHQXPEHU SUHFLVLRQ DQG DFFXUDF\ 7KLV 5H WUDQVLWLRQ SURYLGHV QDUURZ LQWHQVH OLQHV WKDW DUH LQ DQ HDVLO\ LGHQWLILDEOH SDWWHUQ 5H +&/nV DUH FRPPHUFLDOO\ DYDLODEOH DQG

PAGE 110

HDV\ WR XVH LQ FRQMXQFWLRQ ZLWK D )76 2EYLRXVO\ PDQ\ RWKHU VWDQGDUGV FRXOG EH FKRVHQ IRU YDULRXV UHJLRQV RI WKH VSHFWUD 7KLV VWDQGDUG LV LQWHQVH RIIHUV V\PPHWU\ LQ +)6 DQG ,6 DQG ZLOO FOHDUO\ WHOO LI WKH )76 LV SHUIRUPLQJ XS WR FODLPHG VSHFLILFDWLRQV

PAGE 111

)LJXUH 5H OLQH SURILOH FOHDUO\ VKRZV K\SHUILQH VWUXFWXUH +)6f DQG LVRWRSH VKLIWV ,6f 7KH FRUUHFWHG ZDYHQXPEHUV IRU 5H FPf WUDQVLWLRQ DUH 5Hf 5Hf 5Hf 5Hf 5Hf 5Hf 5Hf "5Hf ff5Hf P5Hf

PAGE 112

,17(16,7<

PAGE 113

&+$37(5 81'(567$1',1* 12,6( ,1 )285,(5 75$16)250 63(&7526&23< ,QWURGXFWLRQ 9DULRXV DVSHFWV RI QRLVH VWXGLHV LQ )RXULHU 7UDQVIRUP 6SHFWURVFRS\ KDYH EHHQ UHYLHZHG E\ LQ VHYHUDO ERRNV LQFOXGLQJ *ULIIWKV DQG GH+DVWK f &KDPEHUODLQ f %HOO f DQG %UDFHZHOO f 7KHRUHWLFDO PRGHOLQJ DQG H[SHULPHQWDO ZRUN LQYROYLQJ QRLVH LQ )7899,6 FRQVLVW RI ZRUN E\ +RUOLFN DQG FRZRUNHUV f UHVHDUFK DW ,PSHULDO &ROOHJH f ZRUN E\ -LP %UDXOW RI .LWW 3HDN 1DWLRQDO 2EVHUYDWRU\ f 9RLJWPDQ DQG :LQHIRUGQHU f +LUVFKIHOG f )DLUHV f 6DNDL DQG 0XUSK\ f 1DWDOH DQG 9HQWXUD f .QDFNH f /XF DQG *HUVWHQNRUQ f DQG (PRQGV GLVVHUWDWLRQ IURP WKH 2UVD\ )UDQFHf )7 IDFLOLW\ LOOf $ FRPSOHWH VXUYH\ RI WKH OLWHUDWXUH ZRXOG HQYHORSH PDQ\ PRUH SDSHUV RQ YDULRXV DVSHFWV RI )76 ,Q WKLV SDSHU D VLPSOH V\VWHP WKDW FUXGHO\ PLPLFV D 0LFKHOVRQ ,QWHUIHURPHWHU DQDO\]LQJ PRQRFKURPDWLF DQG GLFKURPDWLF OLJKW IRU QRLVH VWXGLHV LV GHVFULEHG ,W LV FULWLFDO WR XQGHUVWDQG DQG FKDUDFWHUL]H QRLVH VRXUFHV LQ XOWUDYLROHW DQG YLVLEOH )76 ZKHWKHU WKH\ EH IOLFNHU If ZKLWH QRLVH RU LQWHUIHUHQFH QRLVH +] HWFf LI WKH\ DUH WR EH PLQLPL]HG RU HOLPLQDWHG ,Q LQIUDUHG )76 )7,5f WKH

PAGE 114

V\VWHP LV W\SLFDOO\ GHWHFWRU QRLVH OLPLWHG ,Q )7899,6 WKH GHWHFWRUV W\SLFDOO\ SKRWRPXOWLSOLHU WXEHV RU 6L3,1 GLRGHVf DUH QRW WKH OLPLWLQJ QRLVH VRXUFH &RPPRQO\ XVHG VRXUFHV LQ DWRPLF HPLVVLRQ VSHFWURVFRS\ VXFK DV IODPHV SODVPDV DQG GLVFKDUJHV DUH VRXUFH QRLVH OLPLWHG )RU D )76 WR VKRZ D 615 JDLQ RYHU D GLVSHUVLYH W\SH V\VWHP WKH )7 H[SHULPHQW ZRXOG KDYH WR EH RSHUDWLQJ QHDU RU DW WKH GHWHFWRU QRLVH ,I VKRW RU IOLFNHU QRLVHV DUH OLPLWLQJ WKH 615 FRPSDULVRQ EHFRPHV FRPSOH[ ,Q WKLV V\VWHP WKH QRLVH FKDUDFWHULVWLFV RI DQ\ OLJKW VRXUFH FDQ EH VWXGLHG &KRSSLQJ WKH VRXUFH DW VRPH IUHTXHQF\ Ic +]f HTXDOV WKH UDWH RI FRQVWUXFWLYH DQG GHVWUXFWLYH LQWHUIHUHQFH 5c +]f ZKLFK LV UHODWHG WR WKH ZDYHOHQJWK VWXGLHG $c FPf DQG WKH PLUURU YHORFLW\ Yc FPVf E\ WKH UHODWLRQVKLS I 5 Y$L f 7KH VDPH GHWHFWRU XVHG LQ WKH )76 FDQ EH UHWDLQHG &KRSSLQJ DW D VHW IUHTXHQF\ DOORZV XV WR VWXG\ WKH PDJQLWXGH DQG W\SH RI QRLVH LQ WKDW UHJLRQ 7KH HIIHFWV RI ORZ IUHTXHQF\ QRLVHV HJ I +]f ZUDSSLQJ DURXQG WKH GHVLUHG OLQH SURILOH H[WHUQDO VRXUFHV SURGXFLQJ LQWHUIHUHQFHV HWF FDQ EH FKDUDFWHUL]HG )RU H[DPSOH D &X YDSRU ODVHU XVHG LQ RXU ODE ILUHV DW N+] 7KLV SURGXFHV D VWURQJ VLJQDO DW N+] DQG LWV UHVSHFWLYH KDUPRQLFV N+] HWFf 5HVHDUFK

PAGE 115

EXLOGLQJV W\SLFDOO\ SURGXFH D ZLGH UDQJH RI LQWHUIHUHQFH QRLVHV WKDW FDQ EH PLVLGHQWLILHG LI QRW FRUUHFWO\ LGHQWLILHG ([SHULPHQWDO 7KH H[SHULPHQWDO VHWXS XVHG LQ WKLV VWXG\ LV VKRZQ LQ )LJXUH $ 1D QPf +&/ DQG D +HOLXP 1HRQ ODVHU DUH WKH VRXUFHV HPSOR\HG IRU GHPRQVWUDWLRQ SXUSRVHV 7KH VHWXS KDV EHHQ XVHG LQ FRQMXQFWLRQ ZLWK PDQ\ VRXUFHV LQFOXGLQJ +&/nV GLRGH ODVHUV DQG YDULRXV ZKLWH OLJKW VRXUFHV )LJXUH VKRZV WKH ODVHU OLJKW VSOLW DQG WKH JHRPHWULF DUUDQJHPHQW ZKLFK SURGXFHV IUHTXHQFLHV FRUUHVSRQGLQJ WR GLIIHUHQW ZDYHOHQJWKnV LQ D 0LFKHOVRQ $ SKRWRPXOWLSOLHU WXEH 3 9f LV WKH GHWHFWRU 1HXWUDO GHQVLW\ ILOWHUV DUH XVHG DV QHHGHG WR DWWHQXDWH WKH OLJKW IOX[ HQWHULQJ WKH PRQRFKURPDWRU 7KH EHDP VSOLWWHUV DQG PLUURUV XVHG DUH RSWLPL]HG IRU WKH IDU YLVLEOH DQG QHDU LQIUDUHG VWXGLHV 0HFKDQLFDO FKRSSHUV HPSOR\HG KHUH KDYH PD[LPXP FKRSSLQJ UDWHV RI DQG +] %HFDXVH PLUURU YHORFLWLHV IRU D W\SLFDO 0LFKHOVRQ LQWHUIHURPHWHU UDQJH IURP WR FPV WKH ZDYHOHQJWK UDQJH FRQVLGHUHG QHDU ,5 YLVLEOH 89f DQG UHDVRQDEOH PLUURU YHORFLWLHV Yf WKH PD[LPXP UDWHV RI FRQVWUXFWLYH DQG GHVWUXFWLYH LQWHUIHUHQFH 5OI DQG +]f DUH VORZ EXW SRVVLEOH $Q H[WUHPH UDQJH IRU )7899,61,5 PRGXODWLRQ UDWHV LV += O222QP FPVf WR +] QP FPVf ZLWK +] EHLQJ UHDVRQDEOH

PAGE 116

7KH P PRQRFKURPDWRU ZDV XVHG WR PLQLPL]HHOLPLQDWH WKH HIIHFWV RI XQZDQWHG OLJKW 7KH RXWSXW RI WKH 307 ZDV VHQW WR D :DYHWHN % &URVV &KDQQHO 6SHFWUXP $QDO\]HU 7KLV GDWD SURFHVVRU XVHV D )DVW )RXULHU 7UDQVIRUP ))7f DQG GLVSOD\V WKH f f f f UHVXOWLQJ 1RLVH 3RZHU 6SHFWUD LQ G%9+]f [D[LVf YHUVHV +] \D[LVf 7KH :DYHWHN KDV D PD[LPXP EDQGZLGWK RI N+] N+]f VFDQV ZHUH FRDGGHG IRU DOO VSHFWUD SUHVHQWHG 5HVXOWV )LJXUH D VKRZV WKH WUDQVIRUP RI WKH 307 GDUN FXUUHQW QR OLJKW VWULNLQJ LWf )RU 899,61,5 )76 WKH UHJLRQ RI LQWHUHVW OLHV LQ WKH +] UDQJH 7KH H[DFW RULJLQ RI WKH QRLVHV DW DSSUR[LPDWHO\ DQG N+] DUH QRW NQRZQ EXW DSSHDU WR EH IURP DQ H[WHUQDO QRW 307f VRXUFH 7KH 307 YROWDJH ZDV YDULHG IURP 9 EXW WKH PDJQLWXGH RI WKH QRLVHV N+]f UHPDLQHG FRQVWDQW )LJXUH E VKRZV WKH QRLVH SRZHU VSHFWUD RI WKH 307 RXWSXW RI XQFKRSSHG 1D +&/ QPf HPLVVLRQ ,W LV LPSRUWDQW WR QRWLFH WKH PDJQLWXGH RI WKH \D[LV G%9+]f f LQ UHODWLRQVKLS WR D 7KH QRLVH OHYHO KDV EHHQ UDLVHG VXEVWDQWLDOO\ 1RLVH VSLNHV DW DQG N+] WKDW ZHUH SURPLQHQW LQ D DUH OHVV VLJQLILFDQW EHFDXVH RI WKH ULVLQJ EDFNJURXQG 7ZR IOLFNHU QRLVH FRPSRQHQWV DUH REYLRXV 7KH ORZ IUHTXHQF\ +]f IOLFNHU QRLVH FRPSRQHQW LV LQ WKH +] UHJLRQ 7KH KLJKHU IUHTXHQF\ IOLFNHU QRLVH FRPSRQHQW GHFUHDVHV DW D VORZHU UDWH IURP +] WR N+] WKH XSSHU IUHTXHQF\ OLPLW RI WKH :DYHWHN LQVWUXPHQW 7KH GURS

PAGE 117

RII RI WKH KLJK IUHTXHQF\ FRPSRQHQW LV QRW DV GUDPDWLF DV WKH ORZ IUHTXHQF\ +]f FRPSRQHQW EXW FDQ EH SUHFLVHO\ PHDVXUHG )LJXUH F VKRZV WKH +&/ HPLVVLRQ ZLWK VDQG SDVVLQJ EHWZHHQ WKH +&/ DQG PRQRFKURPDWRU VHH ILJXUH f 7KLV KDV EHHQ GHVFULEHG DV D UHOLDEOH PHWKRG WR JHQHUDWH D VWURQJ I FRPSRQHQW RQ D OLJKW VRXUFH 7KH LQFUHDVH LQ 136 LV HVSHFLDOO\ KLJK LQ WKH N+] UDQJH 5HFDOOLQJ WKDW PRVW 0LFKHOVRQnV RSHUDWLQJ LQ WKH 899,6 KDYH FRQVWUXFWLYH DQG GHVWUXFWLYH LQWHUIHUHQFH UDWHV LQ +] UDQJH D VRXUFH ZLWK D ODUJH IOLFNHU QRLVH FRPSRQHQW VXFK DV D IODPH RU DWPRVSKHULF SODVPD ZLOO PDNH DFKLHYLQJ D GHWHFWRU VKRW QRLVH RU D SKRWRQ VKRW QRLVH OLPLWHG FDVH LPSRVVLEOH )LJXUHV DE DQG F DUH VWXGLHV RI WKH QRLVH FKDUDFWHULVWLFV RI WKH 307 WKDW DGGUHVV EDVHOLQH QRLVHV DQG QRW WKH PRGXODWHG VLJQDO )LJXUH D DQG E LV WKH PHDVXUHG RXWSXW RI D 1D +&/ QPf FKRSSHG DW +] 7KH H[SHULPHQWDO GHVLJQ VKRZQ LQ ILJXUH ZDV XWLOL]HG ZLWK RQH RI WKH DUPV EORFNHG VR QR OLJKW FRXOG SDVV 7KH EDQGSDVV LV N+] 'HFUHDVLQJ WKH IUHTXHQF\ UDQJH LQFUHDVHV WKH UHVROXWLRQ )LJXUH D VKRZV WKH PRGXODWHG +&/ HPLVVLRQ OLQH ,Q E VDQG LV XVHG WR JHQHUDWH D IOLFNHU QRLVH FRPSRQHQW RQ WKH PRGXODWHG +&/ HPLVVLRQ 7KLV UDLVHV WKH ORZ IUHTXHQF\ +]f If DQG WKH KLJK IUHTXHQF\ N+]f If QRLVH PDJQLWXGHV ,Q ILJXUH DE DQG F WKH I QRLVH IURP D +&/ GHFUHDVHV IURP N+] ZDV GHPRQVWUDWHG 7KH LPSRUWDQW SRLQW LQ D DQG E LV WKDW WKH V\VWHP LV IOLFNHU QRLVH OLPLWHG EXW PD\ EH SHUFHLYHG

PAGE 118

DV EHLQJ VKRW QRLVH OLPLWHG LI D ODUJH IUHTXHQF\ UDQJH LV QRW H[DPLQHG 7KH VORSH RI WKH IOLFNHU FRPSRQHQW LV QRW GUDPDWLF HQRXJK WR EH GHWHFWHG LQ D VPDOO EDQGSDVV HJ N+]f $ ODUJHU IUHTXHQF\ UDQJH N+] RU JUHDWHUf PXVW EH DQDO\]HG WR GHWHUPLQH LI WKH IOLFNHU QRLVH FRPSRQHQW LV OLPLWLQJ 7KH EDFNJURXQG LQ D QDUURZ IUHTXHQF\ UDQJH PD\ DSSHDU WR EH D XQLIRUP EDFNJURXQG WKDW LV LQWHUSUHWHG WR EH VKRW QRLVH OLPLWHG )LJXUH D VKRZV WKH 136 RI WKH +H1H ODVHU HPLVVLRQ PRGXODWHG DW +] ZLWK WKH ODVHU OLJKW SDVVLQJ WKURXJK WKH VHFRQG FKRSSHU EORFNHG )LJXUH E VKRZV WKH VDPH PRGXODWHG OLQH +]f ZLWK D +] VLJQDO RI HTXDO PDJQLWXGH DOVR VWULNLQJ WKH GHWHFWRU 7KH DYHUDJH EDVHOLQH LQFUHDVHV G%9+]ff WR G%9+]fnf ZKLOH WKH VLJQDO PDJQLWXGH UHPDLQV FRQVWDQW 7KH +H1H ODVHU KDV D I FRPSRQHQW WKDW LV VWLOO GHFUHDVLQJ DW N+] XSSHU OLPLW RI WKLV V\VWHPf 7KH LQFUHDVH LQ EDVHOLQH LQ WKLV VLWXDWLRQ FDQ EH DWWULEXWHG WR DQ LQFUHDVH LQ ERWK WKH VKRW DQG IOLFNHU QRLVH FDXVHG E\ WKH DGGLWLRQ RI WKH VHFRQG +]f IUHTXHQF\ &RQFOXVLRQ $ V\VWHP WKDW DOORZV RQH WR VWXG\ WKH DIIHFW RI QRLVHV LQ D 0LFKHOVRQ LQWHUIHURPHWHU LV RXWOLQHG 7KH FKDUDFWHUL]DWLRQ RI QRLVH SRVVLEOH LQ D 899,6 )76 FDQ EH PDGH ZLWKRXW XVLQJ D 0LFKHOVRQ LQWHUIHURPHWHU 6RPH FRPPHUFLDOO\ DYDLODEOH LQWHUIHURPHWHUV VXIIHU IURP SKDVH SUREOHPV PLUURU MLWWHU

PAGE 119

HOHFWURQLF QRLVH HWF 3UREOHPV RI WKLV QDWXUH GHSHQGLQJ RQ WKHLU VHYHULW\ FDQ QHJDWLYHO\ DIIHFW WKH TXDOLW\ RI VSHFWUD REWDLQHG E\ WKH LQVWUXPHQW JLYLQJ IDOVH LPSUHVVLRQV DV WR WKH WUXH SRWHQWLDO RI WKH WHFKQLTXH 3UREOHPV RI WKLV QDWXUH GR QRW DOORZ RQH WR VWXG\ WKH WUXH HIIHFW RI VRXUFH IOLFNHU QRLVH RQ WKH VSHFWUD RU DFKLHYH D VKRW QRLVH OLPLWHG FDVH )OLFNHU If QRLVH LV VKRZQ WR KDYH WZR FRPSRQHQWV LQ +&/ HPLVVLRQ 7KH ORZ IUHTXHQF\ +] FRPSRQHQW DQG WKH KLJKHU IUHTXHQF\ ORZHU PDJQLWXGH FRPSRQHQW WKDW VWHDGLO\ GHFUHDVHV RXW WR DW OHDVW N+] 7KLV KLJK IUHTXHQF\ If FRPSRQHQW PXVW EH PHDVXUHG DQG FRQVLGHUHG LI D VKRW QRLVH OLPLWHG FDVH LV EHLQJ FODLPHG 6RXUFHV WHVWHG +H1H ODVHU ZKLWH OLJKW VRXUFHV RWKHU +&/nVf KDG VLPLODU 136 WKDW LV WKH KLJK IUHTXHQF\ IOLFNHU FRPSRQHQW ZDV VWLOO GHFUHDVLQJ DW N+] 8VLQJ D VSHFWUXP DQDO\]HU ZLWK D EDQGZLGWK H[WHQGLQJ LQWR WKH 0+] UHJLRQ ZRXOG DOORZ WKH H[WHQG RI WKH KLJK IUHTXHQF\ IOLFNHU FRPSRQHQW WR EH HYDOXDWHG

PAGE 120

)LJXUH 7KH HPLVVLRQ RI D +H1H ODVHU LV VSOLW LQ HTXDO PDJQLWXGHV FKRSSHG DW GLIIHUHQW IUHTXHQFLHV UHFRPELQHG DQG DQDO\]HG 1XPHURXV RWKHU OLJKW VRXUFHV VXFK DV D +&/ ZKLWH OLJKW DQG GLRGH ODVHU KDYH EHHQ VWXGLHG ZLWK WKLV GHVLJQ

PAGE 121

,OO 0 L &KRSSHU % H D PV SL L O O U

PAGE 122

)LJXUH 1RLVH SRZHU VSHFWUD RI SKSKRWRPXOWLSOLHU WXEH 307f XQGHU YDULRXV FRQGLWLRQV RI RSHUDWLRQ D WRSf 7KH QRLVH SRZHU VSHFWUD 136f RI WKH GDUN FXUUHQW DQG LQWHUIHUHQFHV SLFNHG XS E\ WKH 307 E PLGGOHf 7KH QRLVH SRZHU VSHFWUD 136f RI XQFKRSSHG 1D +&/ HPLVVLRQ VWULNLQJ WKH 307 FERWWRPf 7KH 136 RI 1D +&/ HPLVVLRQ ZLWK VWURQJ IOLFNHU FRPSRQHQW LQGXFHG E\ SDVVLQJ VDQG WKURXJK WKH OLJKW SDWK

PAGE 123

L U )5(48(1&< +]f

PAGE 124

)LJXUH 0RGXODWHG VLJQDO RI VRGLXP 1Df +&/ VWULNLQJ 307 LV WUDQVIRUPHG UHYHDOLQJ LWV QRLVH SRZHU VSHFWUD D WRSf 1D +&/ HPLVVLRQ PRGXODWHG DW +] E ERWWRPf 1D +&/ HPLVVLRQ PRGXODWHG DW +] ZLWK IOLFNHU FRPSRQHQW LQGXFHG E\ VDQG 'XH WR VPDOO IUHTXHQF\ UDQJH IOLFNHU QRLVH PD\ EH LQWHUSUHWHG DV VKRW QRLVH

PAGE 125

G%9 )(

PAGE 126

)LJXUH 0RGXODWHG +H1H ODVHU VLJQDO LV WUDQVIRUPHG UHYHDOLQJ LWV QRLVH SRZHU VSHFWUD DWRSf 7KH PRGXODWHG +]f VLJQDO RI +H1H ODVHU E ERWWRPf 7KH VDPH PRGXODWHG OLQH EXW DQ DGGLWLRQDO OLQH +] QRW VKRZQf RI HTXDO PDJQLWXGH LV VWULNLQJ WKH GHWHFWRU LQFUHDVLQJ WKH EDVHOLQH QRLVH 7KH LQFUHDVH LV DWWULEXWHG WR D FRPELQDWLRQ RI IOLFNHU DQG VKRW QRLVHV

PAGE 127

G%9 +] G%9 n+] \ 7777IUUUUUQf§L aL f§Lf§Lf§UaUaLf§UaLf§?f§Lf§_f§Lf§UaLaUaUa7aaUaUaLf§_f§Lf§}f§U )5(48(1&< +]f

PAGE 128

&+$37(5 5(62/87,21 $1' 6,*1$/ 72 12,6( 5$7,26 ,1 )285,(5 75$16)250 63(&7526&23< +REEV HW DO f GLVFXVVHG WKH HIIHFW RI UHVROXWLRQ RQ WKH VLJQDO WR QRLVH UDWLR 615f LQ DWRPLF HPLVVLRQ ,&3f H[SHULPHQWV 'DWD FRQFHUQLQJ WKH 615 YHUVXV UHVROXWLRQ ZHUH QRW SUHVHQWHG LQ D SDSHU SULPDULO\ FRQFHUQHG ZLWK VRXUFH QRLVH LQ WKH ,&3 DQG LWV HIIHFWV RQ WKH )7 VSHFWUD 7R EULHIO\ UHYLHZ +REEV HW DO S f 7KHUH LV DFWXDOO\ DQ DSSDUHQW JDLQ RI Qf LQ JRLQJ WR KLJKHU UHVROXWLRQ 7KLV JDLQ KDV LQ IDFW EHHQ HDUQHG E\ LQFUHDVLQJ WKH LQWHJUDWLRQ WLPH Q LV WKH QXPEHU RI VSHFWUDO GDWD SRLQWV DFJXLUHG DQG FDQ EH GLUHFWO\ UHODWHG WR LQWHJUDWLRQ WLPH SURYLGHG RWKHU IDFWRUV VXFK DV PLUURU YHORFLW\ DQG GDWD DFTXLVLWLRQ UDWH DUH FRQVWDQWf 7KLV LV WUXH LI WKH VLJQDO FRQVWLWXWHV WKH LQWHJUDWHG DUHD IRU WKH HPLVVLRQ OLQH SURILOH EHLQJ VWXGLHG 7KH VLJQDO LV FRQVHUYHG EXW ZLOO EH EURDGHQHG DQG ZLWK D GHFUHDVHG LQ LQWHQVLW\ E\ LQVWUXPHQWDO HIIHFWV DW SRRU UHVROXWLRQV 0DQ\ DQDO\WLFDO VSHFWURVFRSLVWV XVH WKH EDVLF UHODWLRQVKLS 61 DYHUDJH VLJQDO DPSOLWXGHf DYHUDJH QRLVH DPSOLWXGHf IRU 615 FDOFXODWLRQV LQ )RXULHU 7UDQVIRUP 6SHFWURVFRS\ )76f ,Q )76 WKH VLJQDO OLQH SURILOHf FDQ EH

PAGE 129

VPHDUHG RXW RU LQVWUXPHQWDOO\ EURDGHQHG DW ORZ UHVROXWLRQV GHFUHDVLQJ WKH VLJQDO DPSOLWXGH SHDN KHLJKWf EXW QRW WKH LQWHJUDWHG DUHD RI WKH VLJQDO ,Q WKLV QRWH WKH QRQOLQHDU JDLQ LQ 615 ZLWK UHVROXWLRQ LV GHPRQVWUDWHG ZLWK D &X +&/ QPf HPLVVLRQ OLQH DW ILYH UHVROXWLRQV FPnf XVLQJ D VLJQDO DPSOLWXGH FDOFXODWLRQ IRU WKH 615 7KH /RV $ODPRV )RXULHU 7UDQVIRUP 6SHFWURPHWHU f /$)76f ZDV XWLOL]HG WR DFTXLUHG WKH GDWD 7DEOH SUHVHQWV GDWD UHVROXWLRQ 615 QXPEHU RI SRLQWV HWFf IRU &X +&/ QP HPLVVLRQ OLQH UHVSHFWLYHO\ DW ILYH UHVROXWLRQV FPf ZLWK WKH VDPH PLUURU YHORFLW\ 7KH VLJQDOWRQRLVH UDWLRnV ZHUH QRUPDOL]HG VR DOO UHVROXWLRQV KDYH LGHQWLFDO LQWHJUDWLRQ WLPHV 7KH PHWKRG WR FDOFXODWH VLJQDO WR QRLVH UDWLRnV VKRZQ LQ 7DEOH LV D VLJQDO DPSOLWXGH WR QRLVH DPSOLWXGH UDWLR 7KH EHVW UHVROXWLRQ FPn LV WKH SK\VLFDOO\ UHVROYHG OLQH SURILOH )LJXUH VKRZV WKH HIIHFWV UHVROXWLRQ KDV RQ WKH VDPH &X +&/ HPLVVLRQ OLQH DW WKUHH UHVROXWLRQV FPnf 'DWD LQ WDEOH VKRZV WKDW WKH OLQH SURILOH LV VLJQLILFDQWO\ EURDGHQHG ):+0 LQFUHDVHVf DW ORZHU UHVROXWLRQV DQG WKH VLJQDO LQWHQVLW\ LQWHUSUHWHG IURP GHFUHDVLQJ 615 ZLWK D FRQVWDQW QRLVH OHYHOf GHFUHDVHV 7KH 615 IRU DOO UHVROXWLRQV VKRXOG EH WKH VDPH VLQFH WKH LQWHJUDWLRQ WLPHV DUH LGHQWLFDO DQG RWKHU SDUDPHWHUV HJ PLUURU YHORFLW\ DSHUWXUH +&/ FXUUHQW HWFf DUH KHOG FRQVWDQW 7KLV LV QRW WKH FDVH &RPSDULQJ WKH 615

PAGE 130

YDOXHV IRU FPn WR FPn UHVROXWLRQV YV UHVSHFWLYHO\ LW LV REYLRXV WKDW WKH LQVWUXPHQWDO EURDGHQLQJ DQG VXEVHTXHQW GHFUHDVH LQ VLJQDO DPSOLWXGH LQWHQVLW\ FDXVHV D VXEVWDQWLDO GHFUHDVH LQ 615 )RU WKH VDPH 615 WR EH DFKLHYHG IRU DOO UHVROXWLRQV D K\SRWKHWLFDO VLWXDWLRQ VXFK DV WKH IROORZLQJ ZRXOG KDYH WR EH DFKLHYHG $VVXPH ZH DUH ZRUNLQJ DW WKH VKRW QRLVH OLPLW XVLQJ D PRQRFKURPDWLF VRXUFH GDWD LV DFTXLUHG DW WZR UHVROXWLRQV DQG FPn ZKHUH FPn LV SK\VLFDOO\ UHVROYHGf ZLWK HTXDO LQWHJUDWLRQ WLPHV HJ FPn VFDQ" FPn VFDQVf 7KHUH LV D ZHOO GHILQHG IUHH VSHFWUDO UDQJH HJ FPnf $W FPn WKH OLQH SURILOH FDQ EH GHVFULEHG ZLWK D *DXVVLDQ IXQFWLRQ DQG DW FPn WKH OLQH SURILOH FDQ EH GHVFULEHG E\ WKH LQVWUXPHQWDO IXQFWLRQ ,QWHJUDWLQJ ERWK IXQFWLRQV RYHU WKH IUHH VSHFWUDO UDQJH ZLOO JLYH HVVHQWLDOO\ WKH VDPH LQWHJUDWHG VLJQDO PDJQLWXGH ,QWHJUDWLQJ WKH QRLVH RYHU WKH )65 ZLOO DOVR EH HTXDO IRU ERWK UHVROXWLRQV 7KH QRLVH SHU UHVROXWLRQ HOHPHQW ZLOO EH KLJKHU DW ORZ UHVROXWLRQ EXW WKH WRWDO QRLVH ZLOO UHPDLQ HTXDO 7KH HQHUJ\ OLJKW IOX[f VWULNLQJ WKH GHWHFWRU LV FRQVHUYHG LUUHOHYDQW RI PLUURU GLVSODFHPHQW 0HDVXULQJ VLJQDO DQG QRLVH PDJQLWXGHV LQ WKLV PDQQHU LV RI FRXUVH QRW SUDFWLFDO IRU PRVW DQDO\WLFDO VSHFWURVFRSLVWV EXW LW VKRZV WKDW VLJQDO DQG QRLVH LQWHJUDWHG PDJQLWXGHV UHPDLQ FRQVWDQW IRU DOO UHVROXWLRQV ZLWK HTXDO LQWHJUDWLRQ WLPHV +REEV HW DO S f SRLQWHG RXW $OWKRXJK WKH

PAGE 131

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f WR PD[LPL]HG VLJQDO LQWHQVLW\ :H SRLQW RXW KLV SRVVLEOH HUURU EHFDXVH VHYHUDO DQDO\WLFDO HYDOXDWLRQV RI )7899,6 KDYH EHHQ FDUULHG RXW DW SRRU UHVROXWLRQV XVLQJ D UDWLR RI LQWHQVLW\ KHLJKW WR QRLVH IRU 615 GHWHUPLQDWLRQ

PAGE 132

)LFSDUH &X QPf HPLVVLRQ IURP D +&/ DW WKUHH FPI WRSf FP PLGGOHf FPfERWWRPff UHVROXWLRQV 8VLQJ SRRUHU UHVROXWLRQV FDXVHV WKH ORVV RI DOO SK\VLFDO LQIRUPDWLRQ LQYROYLQJ +)6 DQG ,6 DV ZHOO DV D WUXH SK\VLFDO OLQH ZLGWK 7KH WRS [D[LV LV LQ DQJVWURPV DQG WKH ERWWRP LV LQ ZDYHQXPEHUV FP f

PAGE 133

,17(16,7< &X (&/ G$ 5HV :$9(180%(5

PAGE 134

7DEOH )LJXUHV RI 0HULW IRU &X UXQf OLQH IURP D +ROORZ &DWKRGH /DPS PHDVXUHG E\ 5(62/87,21 FP 180%(5 2) 32,176 1f [O [O )65 FPf 615 ):+0 FPn 1FPn 1):+0 WKH /$)76 R f WR ; ; ;

PAGE 135

&+$37(5 &21&/86,216 $ FRPPRQ PLVQRPHU DPRQJ DQDO\WLFDO VSHFWURVFRSLVWV LV WKDW )RXULHU 7UDQVIRUP 6SHFWURVFRS\ LQ WKH XOWUDYLROHW DQG YLVLEOH GRHV QRW ZRUN )RU RYHU D GHFDGH WKH )RXULHU 7UDQVIRUP 6SHFWURPHWHU DW .LWW 3HDN 1DWLRQDO 2EVHUYDWRU\ $UL]RQDf KDV EHHQ URXWLQHO\ WDNLQJ H[FHOOHQW GDWD LQ XYYLV IRU SK\VLFLVWV DQG DVWURQRPHUV 7KH /RV $ODPRV )RXULHU 7UDQVIRUP 6SHFWURPHWHU FDQ RSHUDWH DW KLJKHU UHVROXWLRQ DQG KLJKHU ZDYHQXPEHU DFFXUDF\ WKDQ LWV SUHGHFHVVRUV 'XULQJ P\ WHQXUH DW WKH /RV $ODPRV )RXULHU 7UDQVIRUP 6SHFWURPHWHU P\ ZRUN UHVXOWHG LQ WKH ILUVW SXEOLFDWLRQ IURP WKH /$)76 f ILUVW SDWHQW f DQG WKH ODVW H[SHULPHQWDO ZRUN f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

PAGE 136

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f DQG ,PSHULDO &ROOHJH &KHOVHD )76 FUHDWRUf DERXW IXWXUH ZRUN 7KLV ZRUN ZLOO EH DLPHG DW VXSSO\LQJ GDWD IXQGDPHQWDO WR SODVPD VSHFWURVFRS\ :RUN RQ K\SHUILQH VWUXFWXUHV DQG LVRWRSH VKLIWV RI PHWDOV DV ZHOO DV PRUH GHWDLOHG ZRUN RQ OLQH VKLIWV LV SODQQHG

PAGE 137

5()(5(1&(6 f +RUOLFN
PAGE 138

f $EGDOODK 0 0HUPHW-" 6SHFWURFKLP $FWD % f f $EGDOODK 0 $EGHOUDLPDQ$ (LG 0 ,QGLDQ -RXUQDO RI 3XUH DQG $SSOLHG 3K\VLFV f f :ROQLFN. 0LOOHU' 6HOLVNDU& )ULFNH) $SSO 6SHFWURVF f f 6HOLVNDU & :DUQHU $SSO 6SHFWURVF f f 0LOOHU 6HOLVNDU & 'DYLGVRQ 7 $SSO 6SHFWURVF f f )OHLW] 3 6HOLVNDU & $SSO 6SHFWURVF f f &KDQ 6" 9DQ +RYHQ 5 0RQWDVHU $ $QDO &KHP f f &KDQ 6 0RQWDVHU $ 6SHFWURFKLP $FWD 2% f f &KDQ 6 0RQWDVHU $ $SSO 6SHFWURVF f f 0RQWDVHU $ 9DQ +RYHQ 5 &5& &ULWLFDO 5HYLHZV LQ $QDO\WLFDO &KHPLVWU\ f f 7DQ + &KDQ 6 0RQWDVHU $ $QDO &KHP f f &RVWHU %URQV + = 3K\V f f +DVHJDZD 7 :LQHIRUGQHU -' 6SHFWURFKLP $FWD % f f 6HOLVNDU & 0LOOHU )OHLW] 3 $SSO 6SHFWURVF f f 7RPDVLN = =\UQLFNL : 6SHFWURVF /HWW f f 3RUWHU 5 +DUVKEDUJHU : (OHFWURFKHP f f $EGDOODK 0 0HUPHW 4XDQW 6SHFWURVF DQG 5DGLDW 7UDQVIHU f

PAGE 139

f &UDPDURVVD ) %HQHGLFWLV 6 )HUUDUR 4XDQW 6SHFWURVF DQG 5DGLDW 7UDQVIHU f f %HFKWHO $SSO 2SWLFV f f =HHPDQ 3 7HUEODQFKH 6 9LVVHU +DPP ) $SSO 6SHFWURVF f f &DEDQQHV ) 3XUH $SSO &KHP f f &]HUQLFKRZVNL $ 3K\V f f f *XGHPDQ & 0DUWQHU & 6D\NDOO\ 5 &KHP 3K\V /HWW f f %DLOO\ 5RVVHWWL & *XHODFKYLOL &KHP 3K\V f f *ROXERYVNLL < 7HOH]KNR 9 $SSO 6SHFWURVF f f 3ODLQ $ 5LFDUG $ 3K\V /HWW $ $ f f =\UQLFNL : 7RPDVLN = 1RZLFND 6SHFWURVF /HWW f f f 'DYLV *RWWVFKR 5 $SSO 3K\V f f f =KHQJ 6QRZ 0XUSK\ /HLSHUW] $ &KDQJ 5 )DUURZ 5 2SW /HWW f f f %XUOEDZ (" $UPVWURQJ 5 $SSO 2SW f f f /DSS 0 *ROGPDQ / &3HQQH\ & 6FLHQFH f f f %DUUHWW +DUYH\ $ 2SW 6RF $PHU f f *RWWVFKR 5 )LHOG 5 'LFN %HQHVFK : 0RO 6SHFWURVF f f +HU]EHUJ 0ROHFXODU 6SHFWUD DQG 0ROHFXODU 6WUXFWXUH 'LDWRPLFVf YDQ 1RVWUDQG 3ULQFHWRQ f f 0LFKHO + *DQWRLV 0 WK ,QW &RQI RQ +HDW 7UHDWPHQW RI 0DWHULDOV FRQIHUHQFH SURFHHGLQJVf 'HWURLW f

PAGE 140

f %DURQQHW -" HW DO ,63& =XULFK f f 6YLQGRQ $ 6REROHY 1 7VHOLNRY =K (NVS 7HRU )L] 3LVnPD f f 'LHNH *+" &URVVZKLWH +04XDQW 6SHFWURVF 5DGLDW 7UDQVIHU f f .OLP 0 1LNGHO 6 6SHFWURVFRS\ f f 6FKOHLFKHU 5 %DUQHV 5 $QDO &KHP f f $OOHPDQG & %DUQHV 5 6SHFWURFKLP $FWD % f f 7KRUQ $3 6SHFWURSKYVLFV &KDSPDQ t +DOO /RQGRQ S f f .RUQEOXP GH *DOHQ / 6SHFWURFKLP $FWD % f f 0DQQLQJ 7 %HQWOH\ 3DOPHU % +RI 6SHFW /HWW f f :DOWHUV 3 &KHVWHU 7 :LQHIRUGQHU $SSO 6SHFWURV f f 0DQQLQJ 7 3DOPHU % +RI $QDO &KHP f f +XGGOHVWRQH 5+ /HRQDUG 6/ 3ODVPD 'LDJQRVWLF 7HFKQLTXHV $FDGHPLF 3UHVV 1HZ
PAGE 141

f f f f f f f f f *ULHP +5 3ODVPD 6SHFWURVFRS\ 0F*UDZ+LOO 1HZ
PAGE 142

f %XUJHU %XUJKDUGW % %XWWJHQEDFK 6 +DU]HU 5 +RHIIJHQ + 0HLVHO 7UDEHU ) = 3K\V $ f f 0DQQLQJ 7:LQHIRUGQHU -' 3DOPHU % 4XDQW 6SHF 5DG 7UDQV LQ SUHVVf f 3DOPHU %$ .HOOHU 5$ (QJOHPDQ 5 $Q $WODV RI 8UDQLXP (PLVVLRQ ,QWHQVLWLHV LQ D +ROORZ &DWKRGH 'LVFKDUJH /RV $ODPRV 5HSRUWf /$06 f f *ULIILWKV 3 GH+DVHWK )RXULHU 7UDQVIRUP ,QIUDUHG 6SHFWURPHWHUY -RKQ :LOH\ DQG 6RQV 1HZ
PAGE 143

f 1DWDOH 9 9HQWXUD $SSO 2SW f f .QDFNH 5) $SSO 2SW f f /XF 3 *HUVWHQNRUQ 6 $SSO 2SW f f (PRQGV -*0 $ +LJK 5HVROXWLRQ )RXULHU 7UDQVIRUP 6SHFWURPHWHU )RU WKH 9LVLEOH 3K' 'LVVHUWDWLRQ 8WUHFKW f

PAGE 144

%,2*5$3+,&$/ 6.(7&+ 7KRPDV -RVHSK -RKQ 0DQQLQJ ZDV ERUQ WR -DPHV & DQG %HWWH $QQH 0DQQLQJ RQ -DQXDU\ +H LV RI ,ULVK&DWKROLF GHFHQW +H ZDV UHDUHG RQ 6WDWHQ ,VODQG 1HZ
PAGE 145

0HGLFDO 8QLYHUVLW\ RI 6RXWK &DUROLQD DV D WHFKQLFLDQ SURGXFLQJ KLV ILUVW SXEOLFDWLRQf WKH ERRNVWRUH ZLWK &DUO 5,3f DQG 6SDQN\V EDU DV D ERXQFHU DQG EDUWHQGHU +H DOVR VSHQW D VXPPHU ZRUNLQJ ZLWK *HRII &ROHPDQ 9LQFH DQG /HLJK DW WKH 8QLYHUVLW\ RI $ODEDPD RQ D FKHPLVWU\ IHOORZVKLS +H WKHQ HQWHUHG WKH 8QLYHUVLW\ RI )ORULGD IRU JUDGXDWH VFKRRO f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

PAGE 146

, FHUWLI\ WKDW KDYH UHDG WKLV VWXG\ DQG WKDW LQ P\ RSLQLRQ LW FRQIRUPV WR DFFHSWDEOH VWDQGDUGV RI VFKRODUO\ SUHVHQWDWLRQ DQG LV IXOO\ DGHTXDWH LQ VFRSH DQG TXDOLW\ DV D GLVVHUWDWLRQ IRU WKH GHJUHH RI 'RFWRU RI 3KLORVRSK\ ^ f§mf§ DPHV :LQHIRUGKHU &KDLUPDQ *UDGXDWH 5HVHDUFK 3URIHVVRU RI &KHPLVWU\ FHUWLI\ WKDW KDYH UHDG WKLV VWXG\ DQG WKDW LQ P\ RSLQLRQ LW FRQIRUPV WR DFFHSWDEOH VWDQGDUGV RI VFKRODUO\ SUHVHQWDWLRQ DQG LV IXOO\ DGHTXDWH LQ VFRSH DQG TXDOLW\ DV D GLVVHUWDWLRQ IRU WKH GHJUHH RI 'RFWRU RI 3KLORVRSK\ 5LFKDUG $
PAGE 147

, FHUWLI\ WKDW KDYH UHDG WKLV VWXG\ DQG WKDW LQ P\ RSLQLRQ LW FRQIRUPV WR DFFHSWDEOH VWDQGDUGV RI VFKRODUO\ SUHVHQWDWLRQ DQG LV IXOO\ DGHTXDWH LQ VFRSH DQG TXDOLW\ DV D GLVVHUWDWLRQ IRU WKH GHJUHH RI 'RFWRU RI 3KLORVRSK\ -DPHV + +XQWHU 3URIHVVRU RI $VWURQRP\ 7KLV GLVVHUWDWLRQ ZDV VXEPLWWHG WR WKH *UDGXDWH )DFXOW\ RI WKH 'HSDUWPHQW RI &KHPLVWU\ LQ WKH &ROOHJH RI /LEHUDO $UWV DQG 6FLHQFHV DQG WR WKH *UDGXDWH 6FKRRO DQG ZDV DFFHSWHG DV SDUWLDO IXOILOOPHQW RI WKH UHTXLUHPHQWV IRU WKH GHJUHH RI 'RFWRU RI 3KLORVRSK\ $XJXVW 'HDQ *UDGXDWH 6FKRRO

PAGE 148

81,9(56,7< 2) )/25,'$