• TABLE OF CONTENTS
HIDE
 Title Page
 Dedication
 Acknowledgement
 Table of Contents
 Abstract
 Introduction
 Theoretical considerations
 Experimental
 Results and discussion
 Summary and future work
 Appendix 1: Furnace hardware
 Appendix 2: Furnace software
 Appendix 3: Radiometer system
 Appendix 4: Digital plotting...
 Appendix 5: Diffracted stimulated...
 References
 Biographical sketch














Title: Microcomputer controlled electrothermal atomization using laser atomic fluorescence spectrometry /
CITATION PDF VIEWER THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00099231/00001
 Material Information
Title: Microcomputer controlled electrothermal atomization using laser atomic fluorescence spectrometry /
Physical Description: vii, 157 leaves : ill. ; 28 cm.
Language: English
Creator: Wittman, Philip Kirk, 1956-
Publication Date: 1982
Copyright Date: 1982
 Subjects
Subject: Atomization   ( lcsh )
Laser spectroscopy   ( lcsh )
Fluorescence spectroscopy   ( lcsh )
Atomization -- Data processing   ( lcsh )
Microcomputers   ( lcsh )
Chemistry thesis Ph. D
Dissertations, Academic -- Chemistry -- UF
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
 Notes
Thesis: Thesis (Ph. D.)--University of Florida, 1982.
Bibliography: Bibliography: leaves 152-156.
General Note: Typescript.
General Note: Vita.
Statement of Responsibility: by Philip Kirk Wittman.
 Record Information
Bibliographic ID: UF00099231
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: alephbibnum - 000352519
oclc - 09748587
notis - ABZ0490

Downloads

This item has the following downloads:

PDF ( 5 MBs ) ( PDF )


Table of Contents
    Title Page
        Page i
    Dedication
        Page ii
    Acknowledgement
        Page iii
    Table of Contents
        Page iv
        Page v
    Abstract
        Page vi
        Page vii
    Introduction
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
    Theoretical considerations
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
    Experimental
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
    Results and discussion
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
    Summary and future work
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
        Page 95
    Appendix 1: Furnace hardware
        Page 96
        Page 97
        Page 98
        Page 99
        Page 100
        Page 101
        Page 102
    Appendix 2: Furnace software
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
    Appendix 3: Radiometer system
        Page 111
        Page 112
        Page 113
        Page 114
        Page 115
        Page 116
        Page 117
        Page 118
        Page 119
        Page 120
        Page 121
        Page 122
        Page 123
        Page 124
        Page 125
        Page 126
        Page 127
        Page 128
        Page 129
    Appendix 4: Digital plotting program
        Page 130
        Page 131
        Page 132
        Page 133
        Page 134
        Page 135
        Page 136
        Page 137
        Page 138
        Page 139
        Page 140
        Page 141
        Page 142
        Page 143
    Appendix 5: Diffracted stimulated emission
        Page 144
        Page 145
        Page 146
        Page 147
        Page 148
        Page 149
        Page 150
        Page 151
    References
        Page 152
        Page 153
        Page 154
        Page 155
        Page 156
    Biographical sketch
        Page 157
        Page 158
        Page 159
        Page 160
Full Text












MICROCOMPUTER CONTROLLED ELECTROTHERMAL ATOMIZATION
USING
LASER ATOMIC FLUORESCENCE SPECTROMETRY












BY

PHILIP KIRK WITTMAN


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


UNIVERSITY OF FLORIDA

1982


















This dissertation is dedicated to my closest friends
who most certainly know who they are.
















ACKNOWLEDGEMENTS


I wish to take this opportunity to express my

appreciation to the members of Dr. Winefordner's research

group who, with their discussions and actions, helped make

this time pass. In particular I wish to thank both Dr. Ed

Voigtman and Dr. Ben Smith for their help, guidance and

friendship. I especially wish to express my sincerest

gratitude to Dr. James D. Winefordner, my research

director, for his continual guidance and patient

encouragement during the course of this work. I consider

myself fortunate to have been part of his research efforts.



















TABLE OF CONTENTS


Page

. iii


ACKNOWLEDGEMENTS. . . . . . . . .


ABSTRACT . . . . . . . . . . .

CHAPTER

1 INTRODUCTION. . . . . . . . .

Electrothermal Atomization. . . . .
Laser Atomic Fluorescence Spectrometry . .
Electrothermal Atomization Using
Laser Atomic Fluorescence Spectrometry . .


2 THEORETICAL CONSIDERATIONS. . . .

Electrothermal Atomization. . . .
Types of Atomic Fluorescence. . . .
Atomic Fluorescence Expressions . .
Saturation Spectral Irradiance. . .
Lifetime. . . . . . . .

3 EXPERIMENTAL. . . . . . .

General Comments. . . . . .
Fluorescence Collection . . . .
Temperature Monitoring. . . . .
Detection Electronics . . . . .
Nitrogen Laser . . . . . .
Dye Laser . . . . . . . .
Furnace System . . . . . .
Microcomputer . . . . . . .
Quantum Efficiency. . . . . .
Solutions . . . . . . . .

4 RESULTS AND DISCUSSION. . . . .

Signal To Noise Optimization. . . .
Definitions Of Analytical Parameters.
Limits Of Detection . . . . .
Linear Dynamic Range And Sensitivity.
Precision . . . . . . . .
Sodium Population Profile . . . .
Quantum Efficiency. . . . . .
Saturation Spectral Irradiance. . .


6

6
. . 6


. . 15
. . 11
. . 15
. . 21

. . 21

. . 22

. . 22
. . 22
. . 26
. . 29
. . 352
. . 36
. . 35
. . 36
. . 41
. . 44

. . 44

. . 44
. . 45

. . 53
. . 60
. . 70
. . 73
. . 81












5 SUMMARY AND FUTURE WORK . .

Summary . . . . . .
Future Work . . . . .
Temperature Control . . .
Laser Considerations. . .
Detector Calibration. . .
Sample Types . . . .
Furnace Design . . .

APPENDICES

1 FURNACE HARDWARE . . . .

2 FURNACE SOFTWARE . . . .

3 RADIOMETER SYSTEM . . . .

4 DIGITAL PLOTTING PROGRAM . .

5 DIFFRACTED STIMULATED EMISSION.

REFERENCES . . . . . .

BIOGRAPHICAL SKETCH . . . . .


. . . .. 96

. . . .. 105

. . . 111

. . . . 130

. . . . 144

. . . 152

. . . . 157
















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


MICROCOMPUTER CONTROLLED ELECTROTHERMAL ATOMIZATION
USING
LASER ATOMIC FLUORESCENCE SPECTROMETRY


By

Philip Kirk Wittman

December 1982

Chairman: James Dudley Winefordner
Major Department: Chemistry


Laser atomic fluorescence spectrometry combined with a

microcomputer controlled electrothermal atomizer is eval-

uated with respect to several analytical figures of merit.

Among these are the sensitivity, the limit of detection,

the linear dynamic range and relative standard deviation of

the technique. The atom population distribution as well as

post filter effects are investigated. Additionally the

saturation irradiance curves are determined for three

elements in the electrothermal atomizer. An instrumental

system consisting of a graphite rod furnace, as the

atomizer, a nitrogen laser pumped dye laser as the excita-

tion source and a microcomputer to control the system is

described. Quantum efficiencies are determined by comparing

fluorescence intensities in two sheath gases (argon and

vi










nitrogen) with different quenching cross sections.

Experimental resultss for sodium, tin and manganese are

presented which demonstrate the analytical utility of the

described microcomputer controlled electrothermal

atomization system.
















CHAPTER 1
INTRODUCTION

Electrothermal Atomization

Initially the interest in electrothermal atomization

(1-3) arose mainly in connection with the problem of

developing absolute methods of atomic absorption analysis.

Early publications (2-4) indicated an apparent freedom from

matrix effects on the results of quantitative analyses.

This led to great interest in the technique and

consequently many researchers studied the processes

involved.

Some of this work included characterization of atomic

absorption signals and various methods of their measure-

ment (5-8), the influence of the rate of heating of the

atomizer on the analytical sensitivity (8-10), the effects

of atomizer geometry and construction material on the

signal (11-14), the mechanism of atom formation

(10,12,15-17) and the interference due to compound

formation and composition of the matrix (18-22).

Some of the advantages (23) that electrothermal

atomization (ETA) has over flame techniques are as follows:

1. ETA requires only a few microliters of sample per

analyses.

2. Difficult to nebulize liquids can be conveniently

handled.










3. The efficiency of the vaporization process is

generally better than in a flame, due to the faster

heating rate.

4. The efficiency of the atomization process is

usually better than in a flame, especially in the

cases of those elements which tend to form thermally

stable oxides. This is a consequence of the rapid

heating of the sample in a reducing environment

provided by the argon and hydrogen diffusion flame

sheath gases.

5. Enhancements in the signal to noise ratios in the

electrothermal atomizer are a result of the smaller

sample volume, the absence of analyte dilution by

expanding flame gases and increased lifetime of the

analyte within the analytical volume.

6. The chemical and thermal environment can be better

controlled in ETA.

7. The capability of direct solid sampling exists.

Coupling these advantages with the possibility of

determining a large number of elements with high

sensitivity, selectivity, accuracy and speed, it is little

wonder that electrothermal atomization with atomic

absorption spectrometry has been shown to be of

considerable value for the detection and quantitative

determination of trace amounts of metals in a variety of

matrices (24-26).










Laser Atomic Fluorescence Spectrometry

In 1963, Alkemade (27) reviewed the methods by which

atoms undergo excitation in flames--one of these being

radiational. He described the use of atomic fluorescence

flame spectroscopy for measuring quantum efficiencies and

used this method for measuring the quantum efficiency of

the sodium D-lines.

Atomic fluorescence spectrometry as a method of

chemical analysis was first proposed by Winefordner and

Vickers (28) in 1964. At the same time, Winefordner and

Staab (29) reported the determination of cadmium, mercury,

thallium and zinc in solution by atomic fluorescence flame

spectrometry.

The advent of the tunable dye laser (30,31) and the

power it has to generate atomic fluorescence transitions

has given spectroscopists a valuable tool for analyses.

The general properties of lasers include

1. The laser beam has good directionality (low

divergence).

2. The laser beam is highly monochromatic.

3. The laser beam is spatially and temporally

coherent.

4. The laser beam has a high irradiance.

All of these laser properties have been useful for

analytical atomic fluorescence spectrometry since Fraser

and Winefordner (32) first used a dye laser to excite

atomic fluorescence. Their work covered nine elements in










hydrogen-air and acetylene-air flames. The limits of

detection obtained were within ten to one hundred fold of

those obtained with conventional sources. This lack of

greatly improved sensitivity over conventional sources is

apparently connected with the fact that in flames the

detection limit in most cases is determined by

non-selective scattering of the exciting radiation by small

unburnt particles and by fluctuations in the flame optical

density, as well as by strong molecular fluorescence of

organic species (33). Progress in both theory and

experimental achievements has been rapid in the ensuing

years (34-41).

Electrothermal Atomization Using
Laser Atomic Fluorescence Spectrometry

Patel et al. (42) combined the advantages of graphite

rod ETA with those of atomic fluorescence spectrometry for

the determination of silver, cadmium, copper, mercury,

lead, tin, thallium and zinc in aqueous samples and silver,

lead and tin in oil-based samples. However, this work used

electrodeless discharge lamps as the exciting sources.

This work did point out the usefulness of combining

atomic fluorescence with electrothermal atomization, and so

it was inevitable that the next step would be evaluating

laser excited atomic fluorescence with ETA. The low

scattering level of the exciting radiation in the

analytical zone and the absence of organic compounds, which

quench fluorescence and can produce a considerable

background due to molecular luminescence, permits an







5

appreciable decrease in detection limits over laser excited

atomic fluorescence flame methods.

Laser atomic fluorescence with electrothermal

atomization has produced the lowest detection limits by

spectroscopic methods for several elements (43-46). The

present work was intended to characterize several figures of

merit of such a system and to make analyses easier by the

automation provided by microcompute- control.
















CHAPTER 2
THEORETICAL CONSIDERATIONS

Electrothermal Atomization

Analytical signals obtained with electrothermal

atomizers are, in general, curves having peaks (24), their

exact shape being determined by a number of factors. To

understand why this should be, L'vov (5,47,48), L'vov et

al. (49,50) and Sturgeon and Chakrabarti (23) devised a

mathematical model to describe the time-dependent

characteristics of the atom population within an isothermal

volume. For the sake of simplicity, only the process of the

transfer of sample vapour through the analytical volume

(defined by the beam of excitation radiation and the cell

length) is considered.

Two simplifying assumptions were made in this

mathematical model: the element to be determined is

completely atomized and all of the atoms of the element

enter the analytical volume. The following variables will

be used:

N = number of atoms of the element to be determined
0
in the sample, dimensionless

N(t) = total number of atoms of the element within the

analytical volume at the moment of time, dimensionless

T1 = atomization time--the length of time for the

transfer of all the atoms into the analytical volume, s

6






7

T2 = residence time of atoms within the analytical

volume, s

n ((t) = number of atoms entering the analytical volume
-1
at any time, s

n2(t) = number of atoms escaping from the analytical

volume at any time, s-

The rate of population change within the analytical

volume is given by

dN/dt = nl(t) n2(t).

With a diffusional mechanism of vapour loss, the

function n2(t) may be approximated by

n2(t) = N(t)/T2

Since the introduction of the sample into the analytical

volume takes place via a process of accelerated

vaporization due to the continuously rising temperature of

the eletrothermal atomizer surface, nl(t) can be

approximated by

nl() = Dt

where D is the atomization factor. With normalization D

may be determined:


/ Tl nl(t)dt = No.


Therefore
nl(t) = (2N /T1 =
)1 0
and the rate of population change becomes

dN/dt = (2N /T1) N()/T2
0 1 Nt/2








This is a general linear first order differential equation

which may be solved with an integration factor exp(t/T )

and then integrating and evaluating at the boundary

condition N = 0 at t = 0 yields


N (t) = 2N -1 + e-t/
1 0 1 2


When the sample has been completely atomized (t = T1 ),

N(t) attains a maximum given by

T2
N (t) = 2N 1 + e T1
t= 0 T 2 T2
1

However at t > T1 the rate of population change loses the

term that accounts for the addition of new atoms since they

have now all been atomized. Taking this into account, at

times t > T IN(t) is defined by



N (>T ) = 2N- + e-1/' e(l 2
1 1 2

These last three equations describe the change in the

number of atoms within the analytical volume as a function

of time. From these equations, it can be seen that the

signal grows in an exponential manner and also decays away

exponentially (see Figure 1).

Because of the type of atom population change, it can

be shown that by measuring signal peak areas as opposed to

signal peak heights, one often avoids the problem of


































c-



0 A
-P
SC) -0




N 4 L
Os II
(D o









P-O i
O r--


-P 0
oa T










CL)







ra
O 4-0 0















Ct N


00) a
) r 11

-4




*^o H
ft H I
h~ r^S (<















- CO













L O









z




z


LL1
oum
J m










varying sample composition on the analytical results (23).

This is due to fact that the integrated signal is neither

dependent on the rate, n ( ), nor the time, of entry of

sample atoms into the analytical volume.

Types of Atomic Fluorescence

There are five basic types of atomic fluorescence

(51-53). These are illustrated in Figure 2. Resonance

fluorescence results when atoms are radiationally excited

to some excited state and then undergo radiational

deactivation back to the ground state, releasing a photon

of light of the same energy as was absorbed. The most

intense resonance fluorescence is typically from the first

excited state. The transition probabilities for resonance

transitions are generally much greater than those for

non-resonance transitions, which then leads to much greater

fluorescence signals. A variation of resonance occurs when

the lower state is not the ground state, but rather a

low-lying, thermally populated state. This is referred to

as thermally assisted resonance fluorescence.

Another type of atomic fluorescence is di-ect-line

fluorescence. This results when an atom is radiationally

excited to a state much above the ground state and

undergoes radiational deactivation to a lower excited state

and emits a photon of light with an energy different from

that which was initially absorbed. This lower state then,

most often, collisionally deactivates to the ground state

(54).
















Figure 2


Types of atomic fluorescence: a--resonance, b--excited
state resonance, c--Stokes direct-line, d--excited state
Stokes direct-line, e--anti-Stokes direct-line, f--excited
state anti-Stokes direct-line, g--Stokes stepwise line,
h--excited state Stokes stepwise line, i--anti-Stokes
stepwise line, j--excited state anti-Stokes stepwise line,
k--thermally assisted Stokes or anti-Stokes stepwise line
(depending upon whether the absorbed radiation has shorter
or longer wavelengths, respectively, than the fluorescence
radiation), 1--excited state thermally assisted Stokes or
anti-Stokes stepwise line (depending upon whether the
absorbed radiation has shorter or longer wavelengths,
respectively, than the fluorescence radiation),
m--sensitized (D = donor, D = excited donor, A =
acceptor, A = excited acceptor, h = exciting
radiation, h = fluorescence radiation), n--two-photon
excitation via a vi-tual level, o--two-photon excitation
via a real level.







13





a b







c d e f







g h i







j k 1








D D h D
n E
D +A A +D

.A A + h
VF
n o










The third type of atomic fluorescence is stepwise line

fluorescence. This occurs when an atom is radiationally

excited to a state considerably above the ground state, is

then collisionally (although it could also be radiationally)

deactivated to some intermediate state from which it then

radiationally deactivates to a still lower electronic

state, emitting a photon of wavelength different from that

which was initially absorbed. Several variations of

stepwise line fluorescence are shown in Figure 2.

The fourth type of atomic fluorescence is multiple-

photon fluorescence. This requires the simultaneous

absorption of two (or more) photons which combined contain

an amount of energy equal to the difference in the energy

levels of the absorbing atom. This is different from

direct-line and stepwise line fluorescence because these

two pass through specific quantized energy levels whereas

multiple-photon absorption does not. However there is also

a two-photon fluorescence which does pass through a

specific energy level. This occurs when an atom is

radiationally excited by a photon of energy, hvl, and then

undergoes additional radiational excitation to a higher

level by the absorption of another photon but of a

different energy, hv2. Two-photon fluorescence then occurs

when the atom undergoes radiational deactivation to a lower

level (see Figure 2).

The last type of atomic fluorescence is called

sensitized fluorescence. This results when donor atoms or








molecules excited by an external light source collide with

the analyte atom transferring energy and exciting the

analyte atom. The analyte atom then undergoes radiational

deactivation resulting in fluorescence.

Atomic Fluorescence Expressions

The basic fluorescence expression (55) is

BF = i E / I2 d\)
F -4--) 21 V 12 v


where

BF = fluorescence radiance, J s m 2sr-
= path length in direction of detection system, m

47 = number of steradians in a sphere (fluorescence is

isotropic), sr

Y21 = fluorescence power (quantum) efficiency, W

fluoresced/W absorbed

E12 = spectral irradiance of exciting radiation
12
at absorption line, 12' W m-2Hz- (1W = J

s- )

~mk dv = integrated absorption coefficient over
absorption line, m-1Hz

The product E k1 dv is the power absorbed from the

source by the analyte atoms per meter cubed of atomic

species. The integrated absorption coefficient is given by

/k dv = n -hv B) 1- [ -
h V e r g2 1

where






16

hu12 = energy of the exciting photon, J
c = speed of light, m s-

B12 = Einstein coefficient of stimulated absorption,
m3 j-1 s-IHz
m J- ls- Hz

g1l g2 = statistical weights of states 1 and 2,
respectively, dimensionless

n n2 = concentration of states 1 and 2,

respectively, m-3 (n. + n2 = n the total

concentration of atoms in all states, assuming a two

level model).

The bracketed quantity corrects for the effective decrease

in absorption caused by stimulated emission from the upper

state.

Assuming steady state, the number of upward

transitions can be set equal to the number of downward

transitions thereby giving


B12EV12 R 21 E12
k12 + I2EI 1 = 21 + A21 + 2--



where

k12, k21 = excitation and deexcitation

non-radiational (collisional) rate constants, s-

A21 = Einstein coefficient of spontaneous emission,

s-1

B21 = Einstein coefficient of stimulated emission,

m3 J s- t Hz

B12 = Einstein coefficient of stimulated absorption,








m3 J s-1 Hz

n n2 = concentration of states 1 and 2,
respectively m-
-1
c = speed of light, m s

The quantum efficiency of a two level transition is

defined as
A
A21
Y21 =A c
A21 21

and A21 is related to B21 and 312 by



A 8whIZ2 B 8xhv12 1
A21 =- 8 ) B2 B 21 = B
c c 92


where h is the Pianck constant. Consolidating expressions,

BF for two level atoms is given by


BF Y21 E 2 c1 ( 12 (f ) *
F 12 12

where E12 is a modified saturation spectral irradiance

(W m-2Hz- ) and is defined as


E, cA21
12 B21Y21


If the modified saturation spectral irradiance is

expressed in terms of the saturation spectral irradiance

E which is that source spectral irradiance that
V12
causes the fluorescence signal to be fifty per cent of the









maximum possible value, the expression

s 2 1
12 12 91

is obtained.

Now substituting for nl in terms of n Bp is

given by




B Y1 EI h 2 812
BF Y21 E nT ) +
V 12
Es
V12

Saturation Spectral Irradiance

Complete saturation means that the populations of an

excited state and the ground state are equalized.

Therefore, radiationless and radiational (spontaneous)

drains on the excited state are neglible with respect to the

optical pumping rate. A further increase in source

irradiance cannot effect an increase in the fluorescence

radiance (56,57).

If the saturation plateau can be reached, i.e., the

plateau resulting when the fluorescence radiance is plotted

versus the source spectral irradiance (See Figure 3),then

the saturation spectral irradiance, Es can be
212
obtained by finding that source spectral irradiance where

3. equals one-half B F Here B x is the
Smax nax
maximum fluorescence radiance (on the plateau).
































a)



OC
0 a
m Cd
co (
*H


t-
C) a
> O
a2 a0

cd aP C
a) C) (2





-H H 'P p




o r-



) -H aO
S n)



y-i (2

a -H

-H I+







0 tl cP
OC C)


V-H m m


































U-j

U






C Q
L-i
cc:

Cr:




-I

N LN O




xI O
X














C'4



~3DN~3DS3an~l


CE
IL
rm









Lifetime

Lifetimes of excited states can be measured by the

pulse method; this is when a pulsed source of excitation is

used and the fluorescence decay is followed upon

termination of the source excitation (58). The decay time

of the fluorescence signal can be measured and if the

temporal influence of the source and detection-measurement

system is known, then these instrumental effects can be

deconvoluted leaving a decay curve indicative of only the

fluorescent species (58,59). The fluorescence lifetime, ,

is given by

1
3 ZA.. + k..
Si 31i 31


where j is the upper state and i represents other states

involved in the deactivation process (generally j > i).

For a two level atom, T 2 is given by

I
2 21 + k21


In the case whe-e there is no collisional deactivation,

i.e., k21 = O, the lifetime is at its maximum value and

this is referred to as the natural lifetime, T sp This

can be used in the measurement of quantum efficiencies

since the observation of the natural lifetime occurs only

when the quantum efficiency is one (Y21 = 1).
















CHAPTER 3
EXPERIMENTAL


General Comments

The basic setup shown in Figure 4 is the microcomputer

controlled electrothermal atomization system used in this

work. A listing of the equipment used is found in Table

3-1. In order to minimize the fluctuating nature of the

wings of a Gaussian laser beam, the atomizer was placed ten

feet away from the dye laser. By using an iris diaphragm,

the center of the beam was selected and focused over the

graphite rod.

Due to space constrictions a pair of front surface

planar mirrors were used to fold the dye laser beam and

direct it to the atomizer. By defining the path of

interaction of the laser beam with the atomizer as a line

between the aperture and a predefined spot on the beam

stop, it was possible to easily realign the laser beam

after dye changes or equipment sharing.

The collection system was aligned at a 900 angle to

the incoming beam by the use of a helium-neon laser crossed

with the dye laser beam.

Fluorescence Collection

A 0.5 m monochromator (Jarrell-Ash, Newtonville, MA),

with a spectral dispersion of 1.4 nm/mm, was used to
22















Figure 4


Microcomputer controlled electrothermal atomization system
using laser atomic fluorescence spectrometry. Component
symbols are as follows: Al and A2 = diaphragm apertures, L1
and L2 = focussing lenses, Mi and M2 = front surface planar
mirrors, PMT = photomultiplier tube, H.V. = high voltage
power supply.





































































TPTIGER









TABLE 3-1

Microcomputer Controlled Electrothermal Atomization using
Laser Atomic Fluorescence Spectroscopy Equipment List


Item


Manufacturer


Apple II Plus Microcomputer

Video 100 Monitor


Analog/Digital Converter

Digital/Analog Converter

VIA Interface Board

Model UV-24 Nitrogen Laser

Model DL-II Dye Laser

Furnace Assembly

Model 20-250 Power Supply


Gas Solenoid Valves

Graphite Rods

J-A 0.5 m Monochromator


Model R1P28 Photomultiplier
Tube

Model 130 Photomultiplier
Tube Housing

Model 226 High Voltage
Power Supply

Model 162 Boxcar Mainframe


Model 164 Boxcar plug-in
module

Photodiode Trigger


Apple Computer, Cupertino, CA

Leedex Corporation,
Elk Grove Village, IL

Laboratory Built

Laboratory Built

Laboratory Built

Molectron Corp., Sunnyvale, C

Molectron Corp., Sunnyvale, C

Laboratory Built

Electronic Measurements,
Oceanport, MA

Laboratory Built

Poco Graphite, Decatur, TX

Jarrel-Ash Company,
Newtonville, MA

RCA Corp., Lancaster, PA


princeton Applied Research,
Princeton, NJ

Pacific Precision Instr.,
Concord, CA

Princeton Applied Research,
Princeton, NJ

Princeton Applied Research,
Princeton, NJ

Laboratory Built









collect the fluorescence. The slits were curved and

typically set at 100 pm width and 10 mm height. When

setting the wavelength on the monochromator, the final

setting was always approached from a lower setting.

One quartz lens was used to produce an image of unit

magnification at the entrance slit of the monochromator. A

fixed 2 in aperture was coupled to this lens to reduce both

scatter and broadband emission of the heated graphite rod

from entering the monochromator. In addition, curtains of

black felt cloth were used to reduce scatter, which can be

especially important in resonance fluorescence cases. A

baffle was placed horizontally and immediately in front of

the slit to eliminate any broadband emission from directly

entering the monochromator.

Temperature Monitoring

Initially a TIL 67 phototransistor (Texas Instr.,

Dallas, TX) was mounted in a Macor machineable ceramic

block (Accuratus Ceramic, Washington, NJ) viewing the

center of the graphite rod (see Figure 5). The signal

generated was fed into an A/D converter which was then read

by the microcomputer. This was calibrated using an optical

pyrometer (Pyrometer Instr., Northvale, NJ) also focused

at the center of the graphite rod. This temperature

calibration was then compared to the voltage sent out by

the D/A converter to the furnace power supply. This

information was then used to allow a simple selection of

temperature and have the microcomputer calculate and send
































04

-1
0
o r





0 0
4 ->
oo

>











0 0
-P 0)
03
*M 4
4r-l -







CO
s0


OO











r- -
-0-
00
a*















H 0-
.) -


0 cC)
i-i



r-l 0)0




C) 1B-P









28



















Oh
















c
v,






























<^+0









out the control voltage required to produce this

temperature.

Encountering problems with the reproducibility of the

phototransistor after prolonged use, due to excessive

heating and melting of the epoxy dome-shaped lens of the

phototransistor, it was replaced with a tungsten-5%rhenium

vs. tungsten-26%rhenium thermocouple (Omega Engineering,

Stamford, CT). This was also placed at the center of the

rod, but offset to one side by 1 mm. This offset was

necessary because the thermocouple would not function when

in contact with the graphite since it is part of an

electrical circuit. Again using the optical pyrometer, the

thermocouple output was cross calibrated. The thermocouple

output, since it was in the millivolt range, was fed

through a laboratory built scaler set at a gain of 100 fold

so that the output more nearly filled the A/D converter's

input range. It was then sent to channel two of the A/D

converter's four channels. This calibration enabled the

construction of a plot of D/A control voltage versus

furnace temperature (see Figure 6). The temperature

readings were collected within a few hundred microseconds

of each individual laser pulse so that each fluorescence

pulse could be associated with a particular temperature.

Detection Electronics

A PAR Model 162 boxcar average with Model 164 gated

integrator plug-in module was used for the detection and

measurement of the fluorescent pulses. The Model 164










































Q)

43




4-


4-
0
tC


















4-
(d





0
4-
4-P





r-P







0
0 0
H o






H C
H -

S-





























LJ)




U
\ OL
SO







\I-







LU)






o







A) 39y10OA U/o









plug-in offered a fixed sensitivity (100 mV full scale) and

several integration times and gate delay times. The 50 0

input impedance was always used to match the connecting

cables (RG-58U coaxial) and minimize ringing. The gate

most commonly used was 50 ns with an aperture delay of 0.2

vs. This was necessary to avoid the huge ringing RF

interference caused by the electrical breakdown in the

spark gap of the nitrogen laser. The jitter in the time

between triggering and the the laser actually firing was

circumvented by using an optical trigger between the

nitrogen laser and the dye laser. This enabled correct

timing between laser firing and boxcar gating. The output

of the boxcar was fed through a voltage inverter and then

into channel zero of the A/D converter.

Nitrogen Laser

The nitrogen pump laser was run according to the

manufacturer's instructions. The nitrogen flow rate was 26

L/min for all experiments. Operating pressure was

maintained at 60 torr.

The laser repetition rate was controlled by the

microcomputer and for ease of operation only three

repetition rates, 10, 20, and 29 Hz, were user selectable.

The 29 Hz was the upper limit because of the speed at which

the compiled BASIC control program operated. Between

samples the repetition rate would default to 16 Hz, the

slowest that the software independent clock of the

Versatile Interface Adapter could be run; but this still









satisfied the requirement that the nitrogen laser be

triggered at least once every 30 s while in the run mode.

Dye Laser

The dye laser was also operated according to the

manufacturer's instructions. The laser dyes used are

listed in Table 3-2. Adjustments of both the oscillator

dye cell and amplifier dye cell carriage were made after

each new dye cell was introduced into the beam line. Since

the Model DL-II dye laser operates in the third through

seventh order, it is necessary to determine which order you

can best operate in to be able to set correctly the

wavelength dial of the dye laser. Having done this, it is

simply a matter of following the manufacturer's

instructions to maximize the energy output. However some

difficulty was encountered with the frequency doubling

accessory. The doubling crystal and focussing lens assembly

is extremely sensitive to movement which foiled early

attempts to utilize this valuable accessory. Additionally

when doubling to a wavelength below 270 nm the polarization

of the fundamental beam must be vertical. Because the

DL-II laser normally produces a horizontally polarized

beam, an intracavity polarization rotator (half-wave plate)

had to be installed. Once installed, it only takes a flick

of a small lever to rotate the polarization from horizontal

to vertical and back again. In cases where it is not being

used, it can be removed totally in order to achieve the

highest energy output.









Table

Laser


3-2

Dyes


Dye Concentration Solvent Wavelength
(Moles/Liter) Range (nm)


Rhodamine 6G

Rhodamine B

Coumarin 460

PBD


5 x 10-3

5 x 103
1 x 10-2
5 x 10-


1 x 10-3


Ethanol

Ethanol

Ethanol

Ethanol/Toluene
(1/1)

p-dioxane


568--605

594--643

440--478

360--386


390--416









The only problem with dye deterioration occurred with

Coumarin 460, and therefore it was changed frequently in

order to avoid undue energy loss. Although the manufacturer

recommends exchanging the old dye via pipet, rinsing with

fresh dye, and refilling with fresh dye, it was found that

the dyes performed better by flushing the cell with pure

solvent before rinsing and refilling with fresh dye.

Furnace System

The gas flow control system consisted of one solenoid

valve for the sheath and flame diluant gas and another

solenoid valve for the flame gas. Each gas had its own

individual rotameter. The entire system was calibrated by

means of a linear mass flow meter (ALK-50K, Hastings,

Hampton, VA). The flow rate of the sheath gas, the flame

diluant gas and the flame gas was set on their respective

rotameters while the solenoid valves were used to turn this

flow on or off. In this way, it was possible to conserve

the gases when not actually in use. Both the sheath gas

(argon) and flame diluant gas (argon) were used to provide

a sheath of an inert atmosphere around the graphite rod.

The flame gas (hydrogen) was used to provide a reducing

environment and to burn off any entrained oxygen. The

sheath gas flow was optimized at 1.2 L/min, the flame gas

was optimized at 0.50 L/min and the flame gas diluant was

optimized at 0.31 L/min. The flame gas and flame diluant

gas were introduced under the center of the graphite rod

via a multi-capillary tube burner, and this burner was









surrounded by several concentric rings which encased the

length of the rod with sheath gas only (See Figure 7).

The graphite furnace consisted of a cylindrical

bakelite block 8.2 cm in diameter and 2.6 cm high with

water-cooled, copper blocks on either side to support the

graphite rod and provide electrical contact with the SCR

power supply. The SCR Model 20-250 power supply was wired

in the external voltage program control mode to permit the

use of a microcomputer driven D/A converter in controlling

the power supply's output and thus the furnace temperature.

A modification in the graphite rod design was made to

facilitate machining of the rods as well as to reduce the

turbulent flow in the sheath gases caused by sharp edges.

Additionally the small sample hole was eliminated due to

the irreproducibilty in its depth and because the removal

of the graphite by drilling this hole caused the rod to

burn unnecessarily fast at this point. Due to the small

sample volume (pL) used, the sample formed a droplet which

was small enough to place on top of the graphite rod

without any spillage. The dimensions of the simplified rod

design are shown in Figure 8.

The furnace was mounted on a riser capable of

translational and vertical movement. This allowed easy

optimization of rod position under the laser beam.

Microcomputer

The Apple II Plus was used to control as much of the

electrothermal atomization system as possible. The ability




























01
0,C



0




S0-
a,C

SO
-z





C
CC
OD








Cm
(Dro







4fld
-H *H
-^ s






4A
adn


400
,0 -




*H



am



-C
40
ci) o



-PO






-H
c a




0 00
mdc
s:









a,CO




CC
de d
q t-O
0


>laO


























a)


CU
Ia


































C




c

0









a)


o













4C
-4 )
a)

0-r-
r-4

1-C

0-1




"-A














-4c
cC '








40




































C
C C
0 0
c C









t



0
__










of the microcomputer to communicate with the outside world

was made possible by the Versatile Interface Adapter (VIA)

board (see Appendix 1). Through this interface the

nitrogen laser was triggered, the graphite rod temperature

was controlled, the sheath, flame diluant and flame gases

were turned on/off, the boxcar was enabled/cleared, and

both the temperature monitor output and boxcar output were

read. All data as well as the operating parameters could

be stored on disk for later retrieval or dumped to the

video monitor for immediate consideration or to a printer

for a hard copy. The control program was formatted so that

after it was loaded into the microcomputer the data could

be stored on a disk in either drive one or two (see

Appendix 2).

This particular Apple II Plus was outfitted with a

parallel interface board (for printers), a serial interface

board (for the digital plotter), the VIA interface board,

an Integer BASIC ROM board, and a Videoterm 80-column board

(for the wo-d processor). The microcompute" itself was

mounted in a wheeled rack so that it could be used in

various other experiments in the laboratory, such as the

radiometer system (see Appendix 3).

Quantum Efficiency

In order to determine the efficiency of the

fluorescence process in the electrothermal atomizer, two

different sheath gases were used. Argon which has a small

quenching cross section (60) was used in all of the









calibration and fluorescence studies. By replacing argon

with nitrogen, which has a much larger quenching cross

section (60-64), it was possible to obtain relative quantum

efficiencies for the two environments.

To make this relative information more definitive, the

fluorescence lifetime of the sodium 3P-3S transition (65)

in the electrothermal atomizer was measured. To do this a

fast photomultiplier tube with a known rise time had to be

used. The Nitromite nitrogen laser (Photochemical Research

Asc., Ontario, Canada) was used to measure the rise time of

an R1414 photomultiplier tube seated in a fast wired

E850-03 base (Hamamatsu, Middlesex, NJ). The laser pulse is

typically 350 ps (when the spark gap is operated at

atmospheric pressure) and it was scattered into a

monochromator onto which the R1414 tube had been mounted.

The resulting signal was monitored using a Model 7834

storage oscilloscope (Tektronix, Beaverton, OR) with a

Model 7A19 plug-in amplifier (500 MHz bandwidth) which has

a -ise time of 0.9 ns. From this signal, it was determined

that the response time of the photomultiplier tube and base

combination was 1.8 ns. This compares favorably to the

manufacturer's specification of 1.4 ns.

This detector then replaced the RCA 1P28 photo-

multiplier tube used for all the other measurements on the

Curnace system. The Molectron dye laser beam was then

scattered into the monochromator and the photocurrent was

fed di-ectly into the 50 0 input of the Model 7A19






43

amplifier. This yielded a signal with a full-width

half-maximum of 4 ns for the dye laser pulses. A 5 PL

aliquot of 1000 ppm Na was then placed on the graphite rod

and atomized.

Solutions

Stock solutions of 1000 pg/mL for all elements were

prepared from reagent grade chemicals in deionized water as

per Parsons et al. (66). Working solutions were prepared

by serial dilution from the stock solutions on a daily need

basis.















CHAPTER 4
RESULTS AND DISCUSSION

Signal To Noise Optimization

As mentioned in Chapter 3, the intersection of the

center of the excitation optics and the center of the

luminescence optics (L2, L1 and Al, respectively of Figure

4) was determined; this intersection could be located to

better than 1 mm by using graduated markings on the furnace

mount. The graphite furnace was coarsely positioned using

these markers, and then, by means of the translational and

vertical screw adjusts, the fluorescence signal was finely

tuned to a maximum.

Because ideally a fluorescence signal should be

measured against "no" background signal, two apertures (Al

and A2 of Figure 4) were used to minimize the source laser

radiation from being scattered about the room and

eventually finding its way to the photomultiplier tube.

Care was excercised to ensure that the apertures did not

limit the solid angle of fluorescence collection.

No mirrors were used to increase fluorescence

intensity by reflecting the exciting laser beam back upon

the analytical volume or in collecting a large- solid angle

of fluorescence radiation from the analytical volume. An

improvement in SNR of a factor of 3-10 by employing mirrors

in this manner has been reported (41,67,68).

44









The atomic fluorescence SNR of sodium was examined for

graphite rods of two different designs. The SNR for the

round design (see Figure 8) was better than for the

previous rectangular design (69). This is attributable to

the less turbulent gas flow around the rod and a resultant

decrease in analyte transport noise into the analytical

volume of the electrothermal atomizer. Therefore, the round

design was used for all subsequent atomic fluorescence

measurements.

Typical atomic fluorescence data for the three

elements obtained with the microcomputer controlled

electrothermal atomizer are presented in Figures 9, 10 and

11. These data (as all data used in this work) were

reconstructed from data files collected and stored by ETA

PS7.1.0BJ (see Appendix 2) on floppy minidisks. The data

were plotted on a digital plotter by a BASIC plotting

program found in Appendix 4.

The optimum temperatures of the graphite rod were

determined for each element by comparing fluorescence

signal to noise ratios. The graphite rod lifetime was also

taken into account here; higher temperatures led to shorter

lifetimes of the graphite rod.

Definitions of Analytical Parameters

The atomic fluorescence concentration (ng/ml) limit of

detection (LOD) was defined as that concentration (in

ng/ml) giving a signal of 3X the rms noise of the

background (blank) noise. Each LOD was found by preparing






























LC4




-C

00




rd
0 03






> 0



O- 0


0 0
00



,-H 0
*H al
0r> 05





0 0>
H 0 L










O 0
0 E -H

-H *0H










00
0H0
0 S


: -H 0.
0 L*>




P t *






































In-.
In

La
'-







CI,





CN


('n'8) 3DN3DS3om-i7





























* 0


H 0

0cU
O *



0

Ui
> m


O


'40


> -
-HO
cn


0
O
mri
CO O;


> -P
,0
r 0

coo

H 0)


U)
UD>
0 O-

































LO)
LJ


cl i


c'n'f) 3DN3DS3OO71J


































4-'
[H
O








CO


0
me
> 0n
Ccl







Cd



gO
C)O





Q)
> m

0>-'



ai)



C)
a)
a--

SH m










C AP

!* *H
CO





-H L) 4
o-
>,C 0
(DI aio



























UL




i.n
LU

1-


(,n'8) 3DN3353~ON7J









analytical curves for each element and then extrapolating

back to a signal level which corresponded to that of three

times the rms noise. The concentration associated with

this signal was the limit of detection. Atomic

fluorescence measurements of the three elements were

determined for concentrations ranging from 1OOOg/mL to

1 pg/mL. Solutions of lower concentrations were not used

because of the problem associated with accurately making

the required dilutions and concern with how well the small

aliquot (5 uL) would represent the bulk of a dilute

solution. The upper concentration limit was determined by

the beginning of serious curvature (greater than 4 per cent

deviation from linearity) of the analytical curve. The

linear dynamic range (LDR) was the difference between the

upper concentration limit and the limit of detection, and

the sensitivity was the slope of the analytical curve. The

precision of the system (relative standard deviation) was

determined for the lowest concentration available.

Limits of Detection

The resulting LOD's for the three elements are given

in Table 4-1 along with the linear dynamic ranges, relative

standard deviations, sensitivities and the wavelengths

used. Both concentration limits of detection (CLOD--in

ng/ml) and absolute limits of detection (ALOD--in pg) are

listed in Table 4-1. In general, the concentration limits

of detection are about one order of magnitude poorer

(higher) than those reported by Winefordner (70) for laser










excited atomic fluorescence. Compared to the best reported

values to date (46) for furnace laser atomic fluorescence

spectrometry, the concentration limit of detection is 2 or

more orders of magnitude poorer. However, when the aliquot

size is considered, then the absolute limits of detection

can be compared to these other values. In this comparison,

the detection limits are better than those reported by

Browner (71) and within a factor of 5 of those of Bolshov

et al. (46).

It should be noted here that since the LOD depends

upon the signal intensity as well as the background (blank)

noise, it was important to have as pure a graphite rod as

possible. It was noticed that it took as many as three

sequential hearings of a new graphite rod to atomize and

remove any surface contamination. After the third heating

the blank level was a rather flat constant offset with only

the background noise on it. This contamination was

especially evident at the sodium wavelength presumably from

the handling during the final machining of the graphite

rods. The contamination as well as the effect of these

presample hearings are clearly seen in Figures 12, 13 and

14.

Linear Dynamic Range and Sensitivity

The atomic fluorescence analytical calibration curves

were constructed for the resonance fluorescence of

manganese and sodium and for both excited state Stokes

direct-line fluorescence and excited state anti-Stokes





























'0





(D 0 o
4C ,*

-H EH 0

t- 0
a
0 0
0e

1) H H

a 0a







-4-'
0^ do







CO

F, 0 -LC
NO *r

o Ot
0, D f













0 00
0 r 0 -H
Oa) fl a- r-



(O S ,


r0 Orl
Hif3H O-iZ
-H O 3




-P, -4-'



0) a) o
a- -''
































D



Ln






-)




















EI IE


N

(mn*SI) TJNT S3o[T


































C-P






M4- 00
c<0+d
-- CO







S0
en Qa)
















0 PR
:- C d




O












I- C -4 CD
en (D -P e
O? F-


-H C 4S



I-C




C ci



C0 0

n -H a





C C *H

So Co

C, U2 0C
































Ln


u4











N (










) D


rnN) 3DN3DS3Jomj



























,)

)C


, 00
P0 C


M *C






co t
0
~0 0 0-P
H ) -










0 +
0 0 .





0 -P(
- r- O -PC >












--4

3- o


H 0JD







OO O
H 0C 0 )



-P 00







59



















LO





Ln

L/U
I-
















I TTS









clnly) 3DN3S3yonidJ










direct-line fluorescence of tin. These curves are shown in

Figures 15, 16, 17 and 18, respectively. It should be

noted that the excited state anti-Stokes direct-line

fluorescence of tin was not analytically useful in the

concentration range and system conditions investigated. It

is shown here to emphasize that care must be taken in the

selection of atomic transitions to ensure that a linear

relationship exists between the atom concentration and the

fluorescence signal. It can be seen that the curves for

both sodium and manganese do not include the upper

concentration limit--all that can be said for these elements

is that it is above 1000 g/mL. However the linear dynamic

range for both manganese and sodium was over 5.5 orders of

magnitude. The LDR for tin was determined to be only 4

orders of magnitude.

Precision

The precisions for the three elements are listed in

Table 4-1. The precision obtained is competitive with

other atomization devices used in atomic fluorescence. For

the elements examined the relative standard deviation was

0.064 to 0.088 which is typical of graphite rod atomic

fluorescence spectrometry. Two of the factors which

contribute to this high relative standard deviation are

errors in sample placement on the graphite rod and the

fluctuations in the laser beam irradiance from pulse to

pulse.





























0
C



0
t t
C



r-
0


a
rA

0
c)

















t-4
o
a







4-'

oa





C








Sa)



or a)


>ac
lE







62










0o n






00- CN
0




:z
CQ




z
--4

u
z
C)

\z

LO
\ 1-J

0







\ -I-




o





('n') 33ND3SlonJ so




























aB


0
0





C





0




,'1
4B




















O El
co
C
C





















0



0)
**I >



C
as



aH 0



io
HQ
0o --

H *i-



a







64


















C1






\ *-
\ 9-
o-
z


(c




0\


















Tfl' H )JNIDS38Of7A mO7


































0

C).





-a
















0 C
r-3
r
a
0 !



















OH .'
O) 0
I) -I-3










,--- >+



0 I -
0
--0 0D

OH O a)
OF- *H *




o 0-







al Oh-
i nflD *


















(n
O






(N
0-

a-







LU
z
0

.J
z

L.
0
-j


\
I C I I 3I



(Sn'*) 23N3DS3NOf 1 50









































































00

H Ca


H 0


EB


o 'n


0 )
X!N-








68

















O



(L

o-




z
0_





u
z
o
ou


DI
-l

-J







CN
1






r-. u) cn r CN -



'*n') 33N33S3OniJ g0m


























-- C-- D-
Oc C- C- s

0 0 00


O)


0 )

0






0 a


O
S- -
-P0







-a
C )
L (1

0 C)
z 0


o
o oP
0) c




-io
H



0 )
CQ








HoC
Sr-C







o o
1H 0 C.
oHO,


PO r
A-' C t-
-j 0 P
000


>,o Co



d a C

CC)


0 <
o -

Co -
O 0
*H C






S c
S-,0 C)
C C0


CD -0


>,
+3









o
A


-H'
-r-A
CO


o




-C
0
C)





04
C)

0








-o
,0--




o
o
rd ft





0)
o a
0 -


C0
A0




C e
- 00
0 >

H




O -


* O0





Cc
C *C
O ea


-H H H0


-4- M
r. a





-1 a0 -


c -H ,- 0
>P




>4 C- AD C) a) 0




0 .
C o 0) H -3
4-i l 4 ~ 4 d 0



o 0 *r I
o- 0 .H 4- cn
C 0) 0 C 0 CO






0 -P 0
C ) C ) H H

S -- C )



0 H M H
-P 3 4 3 <
o 0) 0
-C0 od O




t v 0 H a0
OH CD EO Cfl + 0
C) > 0
A-' 0 0D C) C)




4o i C0
C *-A C *-C co
0 O *l G CD C C C) C)1

-H H ci A-P o aj
ci CC H)


C 0 > ii C
C) H -1 -H

C) C C) CO H 0

O d *HCD0 H

i (ii ^ H-


O o a 0 C -H

04;13COE-4

MS ord UN


O I







S0 1


C i C
C) :


t I










Sodium Population Profile

The method used to measure the decay of sodium atom

population as a function of height has been previously

described (72). The profile was determined to aid in the

positioning of the electrothermal atomizer in the optical

path of the excitation and emission optics. The same

atomization conditions were used for this study as for the

analytical calibration curve measurements.

The factors contributing to the decrease in the

measured integrated atomic fluorescence as the graphite rod

is lowered include

1. The loss of atoms due to chemical reaction.

2. The transport of atoms outside of the analytical

volume due to diffusion.

3. The decrease of the solid angle subtended by the

atoms in the analytical volume.

4. The loss of excited state atoms due to collisional

quenching.

Even in the absence of condensation, oxide formation and

other possible chemical losses, a gradual decrease of the

fluorescence signal from factors 2 and 3 would be expected.

The decay of atom populations with height above the

electrothermal atomizer was measured by atomic fluorescence

spectrometry, and the results are shown in Figure 19. The

decay of the sodium population was gradual, reaching 50 per

cent at a height of 12 mm above the graphite rod.

































0
0 0
S0 a00







-P0
0H 00

)0 0
-HO C'

0io a a

(0 -P>)0




a)-P F-
F- >cC '



0 0 + -C
0-P ai
- H d aCo )C



a F-)F-
M -H CQ H


0- 0 0


0 K Q 40


*H *H -Cf
0 E-Td a)



-H 0
0 0 4-' (D




a - o -H


a 40 02 -- 0)






72





C4




/


/









LD
>-4




LJ



LO












I I
/D
ED ED FD ED
c Co C -
cn*a) 3DN3S3Jofl1J










In addition to investigating the sodium population as

a function of height it was also studied as a function of

displacement from the center of the graphite rod. This

displacement was investigated to determine what effects, if

any, post-filtering may have. Figures 20, 21 and 22 show

the sodium fluorescence as a function of displacement at

three different heights above the rod. It can be seen that

there is some attenuation of the fluorescence signal when

the path of the exciting laser beam over the graphite rod

is on the opposite side from the monochromator. This shows

up as a skewing of the fluorescence intensity towards the

side of the graphite rod closest to the monochromator; in

the absence of any post-filter effects a totally symmetric

distribution of sodium atoms about the center of the

graphite rod would be expected.

The atom population distribution was adopted as the

general pattern that other atoms would follow;

consequently, the optimum position for sodium fluorescence

was used as the initial position of the furnace for both

manganese and tin.

Quantum Efficiency

In order to obtain an idea of the efficiency of the

fluorescence process in this electrothermal atomizer, two

different sheath gases were (argon and nitrogen) used. By

comparing the fluorescence intensity of a standard solution

in these two gases the relative quenching effects of the

two gases were obtained. Since nitrogen has a much larger





























0

d
E a
4'
B

CM




--40

04


,-C O
R 0
0




4d -

41 t Z


aP CH

0 0 0




-H'(




0 4-

ElO 0
cd




e
0 0 -0



-P *c




0, 0
ao (d
r-l +=

























C(









z


u
0"

Q.
I--






C--





(-
I


I


c n',) 33N3JS3JOu73































ES
E G






c,





0


A-' 0,'
C a--I
0C








o 0..M
-OCO t-
0 0

r-o
Cd




H -H 0






alm
0 0
o a
A C)



o o oP
C ai









C0

cQP 0)l







77


















C'
-1-







Z



cr.
I


--CJ















I i i -
ED C D ED E
a-





















0 E 4 D

cn',N) 3iN33S38onf1j




























0






Lo
o




40
P A







SO-
S0


*0 -P




-H o *Hz

CO 4
*H Ci -
AP -a














A0

4o O








0- 0
Crl )c


oP0 .0

e Oo








.30 0
o0 Ct )
























C14




)-



Z
u-1

EL.
) -J


0-4












CT)
LO L (n C4









(' ) E N3J S j

c'yl ) 3"N3DS30OI7j










quenching cross section, it was expected to have lower

fluorescence signals and this was found to be the case.

The relative quenching ratios (fluorescence in

nitrogen/fluorescence in argon) for a 1000 ppm tin solution

and a 100 ppm tin solution were 0.76 and 0.80,

respectively. A 100 ppm solution of manganese yielded a

quenching ratio of 0.61. The difference in quenching for

the two elements may be attributed to tre different

effective cross sections that these gases have with

different atoms (63).

So that this relative quenching information may be made

more definitive, the fluorescence lifetime of sodium in an

a-gon sheath was measu-ed in the electrothermal atomizer.

The observed lifetime was 6 ns. Comparing this to the

natural lifetime of 16.2 ns (73), it is obvious that a

significant amount of quenching does occur in an argon

sheath. The ratio of observed sodium fluorescence lifetime

to the natural fluorescence lifetime is 0.37 and is

indicative of the magnitude of the radiationless

deactivation processes in the electrothermal atomizer.

Since the quenching processes in the two different sheath

gases would have the same effect on the other elements, the

quenching factor found for sodium can also be applied to

the other elements determined in this system. Therefore,

for manganese and tin, it is implied that thei- respective

quantum efficiencies in an argon sheath are also 0.37

(i.e., reduced from the maximum value of 1 by a factor of










3). The quantum efficiency in a nitrogen sheath is then

found to be 0.23 and 0.30 for manganese and tin,

respectively.

Saturation Spectral Irradiance

The plots of fluorescence intensity versus source

spectral irradiance for sodium, tin and manganese are shown

in Figures 23, 24 and 25, respectively. These were

obtained using an argon sheath with a hydrogen diffusion

flame around the electrothermal atomizer. Of the three

elements examined only manganese clearly showed the

saturation plateau. This indicates that it would be

possible to determine the absolute number density of the

manganese atoms if the detection system were calibrated in

absolute units. This would also then provide a means to

calculate an actual quantum efficiency for manganese in

addition to the relative quantum efficiency determined in

this work.

In order to perform these calculations for sodium and

tin, since they do not reach saturation, it would be

necessary to use the slope method (near saturation

irradiance method) (59). Again, however, this requires the

detection system to be calibrated in absolute units.






































O


mm







C-
-Pa0
0 C4-4
00





m o
0 0





00
t0

00




m o





Co
,C 0



0oZ

o H
0

CQ-

!-0






0 -l
o r- o-i








83
















0



CN

N,
0


\LU
U
z
b-4f

cm




I-







-O















ID
\Q-J



































r0
a)


PC
o -
(^I
CP



2002
0
F- *H


00
Co
0
M (B
> ,

A -P



4,4-
H 02
0 C



>0'
00
4-' 4-'
i -r-4
-H o
x
(00 *
o C
C 0 *r


O-O
LO(a
OM 0
00a0


r^ ffl a,
a o
HOC

0- C22 0
CCO

0-AO
0C00
o c
r -HL
L r-H 0-A















Ln














C








D-4



0


LDi
U
al-


O
(\' .












0
\o
1: .4 u)C

\Rd LIJJ~flJ

































()

4- c
0




o 0

















4
C-40
mO



o0




0 4


So
C)
0 0
00
0(3



00
0 )0

H(





mda



C O
0
0a0



O-C



0
O4
m *



C0



o -



H mO
01~





























Ui






LU







L-J


b-I
CC
u
C)







-i


I _I -L -


cln*N) 33N33538onj so-















CHAPTER 5
SUMMARY AND FUTURE WORK

Summary

A microcomputer controlled electrothermal atomization

system has been described and several analytical parameters

in atomic fluorescence spectrometry have been given.

The theory of the transport of the analyte to the

analytical volume and the resulting fluorescence signal was

reviewed, along with some reasons why non-flame cells in

atomic fluorescence should attain better figures of merit

than flame atomic fluorescence spectrometry. The method

used for obtaining the maximum signal to noise ratio was

described.

The design, construction and evaluation of the new

microcomputer controlled electrothermal atomizer were

discussed. The analytical figures of merit, precision,

sensitivity, linear dynamic range and limits of detection

were examined for three elements. The decay of sodium atom

population above the electrothermal atomizer as well as a

function of horizontal displacement using an argon and

hydrogen diffusion flame sheath were determined, and the

results were discussed.

A major advantage of the present system over some past

electrothermal atomization systems is that the operator has

the power of a microcomputer at his disposal. This allows

88









faster and much more reproducible parameter changes, such

as furnace temperatures and laser repetition rate. This

ease of operation coupled with the long linear dynamic

range and the low limits of detection make this system

attractive for routine use. A more extensive use of the

microcomputer for data reduction and presentation would

only serve to increase the appeal of this system.

Future Work

Temperature Control

An immediate improvement in temperature control would

be realized by adding a real time feedback feature to the

temperature monitor that the controlling microcomputer

uses. By using this feature, it would be possible to

increase the heating rate from the present rate of

approximately 300 K/s to a much higher value (dependent on

the power supply used). This could be accomplished by

operating the power supply at full power until the

temperature monitor senses that the desired temperature has

been reached. At this point, the microcomputer would then

regain control of the power supply and maintain the desired

temperature via the voltage programming mode of the power

supply.

This would also eliminate temperature variations due

to the aging of the graphite rod. This aging results in an

increase in the resistance and a consequential drop in the

current since the voltage is being held constant (voltage

programming mode). This overall drop in power results in a









lower and varying temperature, especially towards the end

of the lifetime of the graphite rod.

Laser Considerations

In the course of this work, many accessories had to be

added to the dye laser in order to obtain the desired laser

wavelength. However, one feature that would have made the

elements requiring frequency doubling easier to determine

was a scan drive for the doubling crystal turret. Since

the angle of incidence of the fundamental laser beam on the

doubling crystal is a function of the wavelength, it is

necessary to change the doubling crystal angle in unison

with the wavelength scanning of the fundamental beam.

Unfortunately, the rate of angle change with respect to the

wavelength change is a non-linear function which is

different for each type of doubling crystal (see Figure

26).

However, it should be possible to use the same

microcomputer that controls the other parts of the system

to change the doubling crystal angle at the proper rate.

This would be done using the stepper motor on the doubling

crystal turret in conjunction with a software controlled

digital output line from the VIA board. Using the

equations that define the curves in Figure 26, the

microcomputer could perform the necessary calculations and

activate the stepper motor to take the required number of

steps at the necessary rate. This of course could only be

of use if the dye laser wavelength scan drive were also
















Figure 26


Tuning curves for the frequency doubling crystals. The
letters designate the different types of doubling crystals.
A--C = KPB crystals which require a vertically polarized
fundamental beam.
D--F = KDP crystals which require a horizontally polarized
fundamental beam.
(from DL-II Series Dye Laser instruction manual, Molectron
Corp., Sunnyvale, CA)



























































FREQUENCY DOUBLER COUNTER










under the microcomputer's control (an external control

connector is provided by the manufacturer).

Improvements in the actual output of the laser would

also improve the analytical parameters of this system.

This would most easily be accomplished by an increase in

the laser repetition rate. For example, it has been

predicted that in the determination of lead, an increase in

the laser repetition rate from 50 Hz to 10 kEz should

result in an lowering of the limit of detection by 15 fold

(74). This, however, would necessitate using a laser other

than a nitrogen laser to pump the dye laser. One of the

most promising in this regard is the Cu laser which can

operate with repetition rates of several kHz. Another

possibility would be using a cavity-dumped continuous-wave

laser to pump a dye laser. In this manner, it would be

possible to obtain repetition rates of several MHz.

Detector Calibration

As mentioned in Chapter 4, if the detection system had

been calibrated in absolute units, it would have been

possible to calculate quantum efficiencies and absolute

number densities. In addition, the bandwidth of the laser

beam needs to be measured (by scanning the monochromator

across the laser wavelength). In further work, this

calibration seems to be of utmost importance to allow a

more definitive characterization of the electrothermal

atomization system.




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs