Laser-excited ionic fluorescence of the rare earths in the inductively coupled plasma

MISSING IMAGE

Material Information

Title:
Laser-excited ionic fluorescence of the rare earths in the inductively coupled plasma
Physical Description:
xii, 96 leaves : ill. ; 28 cm.
Language:
English
Creator:
Tremblay, M. E ( Mario Elmen ), 1960-
Publication Date:

Subjects

Subjects / Keywords:
Fluorescence spectroscopy   ( lcsh )
Rare earth metals   ( lcsh )
Plasma spectroscopy   ( lcsh )
Excited state chemistry   ( lcsh )
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1987.
Bibliography:
Includes bibliographical references.
Statement of Responsibility:
by Mario Elmen Tremblay.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001098618
notis - AFJ4463
oclc - 19508072
System ID:
AA00003800:00001


This item is only available as the following downloads:


Full Text
















LASER-EXCITED IONIC FLUORESCENCE OF
THE RARE EARTHS IN THE INDUCTIVELY COUPLED PLASMA






BY






MARIO ELMEN TREMBLAY


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


1987



























To my parents, wife, and daughter,

whose loving support made all this possible.















ACKNOWLEDGMENTS


I wish to thank Dr. James D. Winefordner for the

support and guidance he gave me during my graduate career

at the University of Florida.

I also wish to thank Dr. Ben Smith and Dr. Ed

Voigtman for their guidance and assistance.

Finally, I sincerely thank my parents; my wife,

Beverly; and my daughter, Kimberly, for their love and

encouragement which made all this possible.


iii















TABLE OF CONTENTS



Page

ACKNOWLEDGMENTS........................................iii

LIST OF TABLES..........................................vi

LIST OF FIGURES........................................vii

ABSTRACT................................................ xi

CHAPTERS


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

Statement of the Purpose......................1
Atomic Fluorescence Spectrometry............. 1
Excitation Sources............................2
Atom Reservoir................................7

PRINCIPLES AND NOMENCLATURE.................11

Theoretical Consideration...................11
Types of Fluorescence Transitions...........20

ANALYSIS OF RARE EARTHS.....................23

Introduction................................23
Analytical Measurements...................... 24
Uses and Applications........................26

ANALYTICAL STUDIES OF LASER-EXCITED IONIC
FLUORESCENCE (ONE-STEP).....................28

Introduction................................28
Selection of Transitions for Laser-
Excited Ionic Fluorescence Measurements
in the ICP ................................29
Experimental................................31
Instrumentation........................31
Chemicals ............................ ..34
Results and Discussion......................35
Conclusion.................................. 67


I


II




III





IV











V ANALYTICAL STUDIES OF LASER-EXCITED IONIC
FLUORESCENCE (TWO-STEP)......................69

Introduction. ............................... 69
Experimental. ................................ 72
Results and Discussion...................... 74
Conclusion.................................. 81

VI FUTURE WORK--IMPROVEMENTS...................88

REFERENCES.............................................. 90

APPENDIX GLOSSARY OF ACRONYMS........................95

BIOGRAPHICAL SKETCH ..................................... 96















LIST OF TABLES



Table Page

1 Fluorescence transitions and detection
limits for the rare earth elements in the
inductively coupled plasma (one-step
excitation) ......................................49

2 Fluorescence transitions and detection
limits for the rare earth elements in the
inductively coupled plasma (two-step
excitation) .....................................75














LIST OF FIGURES


Figure Page

1 Cross-sectional areas of the atomizer
including excitation beam and fluorescence
geometry and prefilter and postfilter
effects......................................... 17

2 Theoretical fluorescence curves of growth
for line source excitation including
prefilter, postfilter and self-absorption
effects........................................ 19

3 Schematic diagram indicating basic types of
atomic or ionic fluorescence transitions.......21

4 Block diagram of experimental system for
laser-excited ionic fluorescence of the rare
earth elements in the inductively coupled
plasma (one-step excitation)...................32

5 Analytical calibration curve for ionic
fluorescence of lanthanum (excitation at
407.735 nm and detection at 399.975 nm)........36

6 Analytical calibration curve for ionic
fluorescence of cerium (excitation at
407.585 nm and detection at 401.239 nm)........37

7 Analytical calibration curve for ionic
fluorescence of praseodymium (excitation at
406.282 nm and detection at 406.282 nm)........38

8 Analytical calibration curve for ionic
fluorescence of neodymium (excitation at
406.109 nm and detection at 428.452 nm)........39

9 Analytical calibration curve for ionic
fluorescence of europium (excitation at
299.133 nm and detection at 290.668 nm)........40










10 Analytical calibration curve for ionic
fluorescence of gadolinium (excitation at
407.844 nm and detection at 354.580 nm)........ 41

11 Analytical calibration curve for ionic
fluorescence of dysprosium (excitation at
407.798 nm and detection at 394.470 nm)........42

12 Analytical calibration curve for ionic
fluorescence of erbium (excitation at
410.400 nm and detection at 344.115 nm)........ 43

13 Analytical calibration curve for ionic
fluorescence of thulium (excitation at
301.530 nm and detection at 313.136 nm)........ 44

14 Analytical calibration curve for ionic
fluorescence of ytterbium (excitation at
303.111 nm and detection at 297.056 nm)........45


15 Analytical calibration curve for ionic
fluorescence of lutetium (excitation at
302.054 nm and detection at 296.332 nm)..


......46


16 Analytical calibration curve for ionic
fluorescence of samarium (excitation at
366.136 nm and detection at 373.126 nm)........47


17 Analytical calibration curve for ionic
fluorescence of terbium (excitation at
403.306 nm and detection at 400.557 nm)..


...... 48


18 Partial energy level diagram of lanthanum
ion (one-step excitation). Wavelengths are
also indicated in nm and relative intensities
in parentheses.................................53

19 Partial energy level diagram of cerium ion
(one-step excitation). Wavelengths are also
indicated in nm and relative intensities in
parentheses................................. ... 54

20 Partial energy level diagram of praseodymium
ion (one-step excitation). Wavelengths are
also indicated in nm and relative intensities
in parentheses.................................55

21 Partial energy level diagram of neodymium
ion (one-step excitation). Wavelengths are
also indicated in nm and relative intensities
in parentheses.................................56


viii











22 Partial energy level diagram of europium ion
(one-step excitation). Wavelengths are also
indicated in nm and relative intensities in
parentheses....................................57

23 Partial energy level diagram of gadolinium
ion (one-step excitation). Wavelengths are
also indicated in nm and relative intensities
in parentheses.................................58

24 Partial energy level diagram of dysprosium
ion (one-step excitation). Wavelengths are
also indicated in nm and relative intensities
in parentheses................................ 59

25 Partial energy level diagram of erbium ion
(one-step excitation). Wavelengths are also
indicated in nm and relative intensities in
parentheses.................................... 60

26 Partial energy level diagram of thulium ion
(one-step excitation). Wavelengths are also
indicated in nm and relative intensities in
parentheses....................................61

27 Partial energy level diagram of ytterbium
ion (one-step excitation). Wavelengths are
also indicated in nm and relative intensities
in parentheses.................................62

28 Partial energy level diagram of lutetium
ion (one-step excitation). Wavelengths are
also indicated in nm and relative intensities
in parentheses.................................63

29 Partial energy level diagram of samarium
ion (one-step excitation). Wavelengths are
also indicated in nm and relative intensities
in parentheses................................. 64

30 Multiphoton processes: (a) two-photon,
single wavelength via a virtual intermediate
level; (b) two-step excitation................. 71

31 Block diagram of experimental system for
laser-excited ionic fluorescence of the
rare earth elements in the inductively
coupled plasma (two-step excitation)............73










32 Analytical calibration curve for ionic
fluorescence of lanthanum (excitation at
457.488 nm and 618.809 nm and detection
at 421.756 nm) ................................. 77

33 Analytical calibration curve for ionic
fluorescence of europium (excitation at
443.556 nm and 306.911 nm and detection
at 333.875 nm) ................................. 78

34 Analytical calibration curve for ionic
fluorescence of ytterbium (excitation at
289.139 nm and 506.731 nm and detection
at 366.970 nm) ................................. 79

35 Analytical calibration curve for ionic
fluorescence of lutetium (excitation at
646.312 nm and 571.349 nm and detection
at 290.030 nm) .................................80

36 Partial energy level diagram of lanthanum
ion (two-step excitation). Wavelengths are
also indicated in nm and relative intensities
in parentheses.................................82

37 Partial energy level diagram of europium
ion (two-step excitation). Wavelengths are
also indicated in nm and relative intensities
in parentheses... .............................. 83

38 Partial energy level diagram of ytterbium
ion (two-step excitation). Wavelengths are
also indicated in nm and relative intensities
in parentheses................................. 84

39 Partial energy level diagram of lutetium
ion (two-step excitation). Wavelengths are
also indicated in nm and relative intensities
in parentheses.................................85















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

LASER-EXCITED IONIC FLUORESCENCE OF
THE RARE EARTHS IN THE INDUCTIVELY COUPLED PLASMA

BY

MARIO ELMEN TREMBLAY

December, 1987


Chairman: James D. Winefordner
Major Department: Chemistry


Laser excitation of ionic fluorescence overcomes the

problem of spectral interference encountered when doing

trace analysis of the rare earths by atomic emission

spectrometry in the inductively coupled plasma.

One or two pulsed tunable dye lasers pumped with an

excimer laser are used to excite ionic fluorescence of the

rare earths in the inductively coupled plasma. Because

several fluorescence lines have been observed after laser

excitation, it was possible to draw partial energy level

diagrams for most of the rare earths. Nonresonance

fluorescence lines were used for all measurements in order

to minimize spectral interference. Detection limits at

given excitation wavelength are reported for each

element. This was the first time that ionic fluorescence










was observed for any of the rare earths in the inductively

coupled plasma. Nonresonance as well as resonance

fluorescence was detected for all of the rare earths

studied except for holmium.


xii















CHAPTER I
INTRODUCTION



Statement of the Purpose

The goal of the research presented in this

dissertation is to demonstrate the application of laser-

excited ionic fluorescence in the inductively coupled

plasma (ICP) to the selective determination of trace level

concentration of rare earth (RE) in a mixture of all REs

without spectral interference. This chapter is devoted

to an introduction of atomic fluorescence spectrometry

(AFS) for trace element analysis with special emphasis on

the laser as an excitation source and an ICP as an

atom/ion reservoir.



Atomic Fluorescence Spectrometry

Atomic fluorescence in flames was first observed in

1924 by Nichols and Howes (1). However, it was not until

1964 that Winefordner and Vickers (2) first introduced the

use of AFS as an analytical method for the measurement of

a number of metals. In their approach, the analyte

solution is nebulized into a flame where atomization of

the analyte takes place. A light source is then directed

onto the flame where the absorption of radiation of proper










frequency causes some of the analyte atoms to become

excited to an electronic state above the ground state.

Some of these excited atoms may return to a lower

electronic state through radiational deactivation. This

process is also known as atomic fluorescence and is

measured at right angles from the incident excitation

radiation. The frequency of the light emitted through the

fluorescence process is characteristic of the analyte

atoms, and the intensity of the signal is related to the

concentration of the analyte atoms in the sample. Since

then, several reviews (3-5) of AFS have been written in

which extensive discussion of previous analytical studies

are given.



Excitation Sources

The linear dependence of the atomic fluorescence

signal upon the excitation source intensity has led to the

development and use of a variety of high intensity line

and continuum light sources. The continuum source offers

the advantage of providing only one light source for

exciting the atomic fluorescence of several elements.

However, these sources generally have low spectral

irradiance over the spectral region of atomic

absorption. Most of the work in AFS has been done with

intense line sources which are generally useful for only

one or a few elements.









The desirable properties of an ideal excitation

source in AFS include short and long term stability;

simplicity of use; versatility in terms of wavelength

range; and, the most critical of all, high spectral

irradiance over the atomic absorption line width. The

most common types of AFS sources and comments concerning

their use will be discussed.

In 1964, Winefordner and Staab (6) used vapor

discharge lamps as line sources for the atomic

fluorescence determination of cadmium and zinc. However,

because these lamps are characterized by severely

broadened and self-reversed spectral line emission, they

are not used anymore.

In 1966, Armentrout (7) used hollow cathode lamps

(HCLs) for his atomic fluorescence work. Commercially

available HCLs are considered to be low intensity line

sources. They do not possess sufficient intensity to be

of great use in AFS, but it is possible to pulse these

lamps momentarily with higher current, thus increasing

their intensity. High intensity HCLs are also

available. The use of HCLs in AFS has been restricted to

a few studies because of their poor spectral irradiance

over the absorption line width of an atom.

A new boosted-output HCL is now being investigated by

Demers and Skrabak (8) as excitation sources for AFS in

the ICP. Boosted-output lamps are basically HCLs in which









an electron emitter has been added in order to improve the

efficiency of excitation of the sputtered atoms. These

lamps possess the power of an electrodeless discharge lamp

(EDL) with the low background noise of a HCL. Because of

the improved spectral irradiance and low background noise

of the boosted-output HCL, these lamps should become more

useful than HCLs for AFS studies.

The xenon arc lamp was first used for atomic

fluorescence measurements in 1966 by Veillon, Mansfield,

Parsons, and Winefordner (9). Since this type of lamp is

a continuum source, it can be used for multi-element

analysis. However, the spectral irradiance of xenon arc

lamps is low over the atomic absorption line width.

A major breakthrough in AFS occurred in 1966 with the

use of electrodeless discharge lamps (EDLs) for excitation

sources (10,11). This was the first time that the atomic

fluorescence detection limits could compete with atomic

emission and atomic absorption method of analysis. In

1967, Dagnall, Thompson, and West (12) were the first to

report the measurement of nonresonance atomic

fluorescence. By measuring nonresonance fluorescence, it

was possible to minimize scattering which was probably the

most serious limitation in AFS. From the results obtained

in the past, it can be concluded that AFS will continue to

improve as long as excitation sources continue to

improve. These improvements include more intense light









sources because the fluorescence signal is directly

related to the intensity of the excitation radiation and

excitation sources which possess narrower spectral

bandwidth, since this would minimize the possibility of

spectral interference from other elements. These lamps

have higher spectral irradiance than the previous two line

sources. The main drawback with EDLs is the lack of

commercial availability of good lamps for some elements.

The ICP (13) has recently been used as an excitation

source for AFS. The spectral characteristics of the

emission from the ICP, such as high intensity, excellent

short and long term stability, narrow line width, and

freedom from self-reversal of spectral lines, make the ICP

an excellent radiation source for AFS. One advantage

compared to other AFS excitation source is its flexibility

with respect to the availability of intense atomic and

ionic lines for many elements. One can change from one

element to another by aspirating a solution containing the

element of interest into the source ICP.

Finally, in 1971, Fraser and Winefordner (14) and

Denton and Malmstadt (15) described an experiment where

they used a dye laser to excite atomic fluorescence of

several elements. Tunable dye lasers (16) are nearly

ideal sources for AFS (except for their cost). The unique

properties of tunable dye lasers include high spectral









irradiance over a narrow spectral bandwidth and the wide

wavelength tunability over the ultraviolet-visible

range. The high spectral irradiance improves linearity of

the analytical calibration curve to 3-5 orders of

magnitude and may saturate the absorption transition, thus

removing the dependence of the fluorescence signal upon

the source stability. The linearity of the analytical

calibration curve may be improved in two ways due to the

high spectral irradiance of lasers. The most obvious is

that the low end of the calibration curve may be extended

due to lower detection limits, until optical saturation

has been reached. The other way is by eliminating the

prefilter effect which will be discussed in more detail in

the next chapter. The spectral bandwidth of the laser

beam can be made as narrow as 0.0001 nm or less by

inserting a series of tuning elements inside the laser

cavity. By utilization of nonlinear crystals, Second

Harmonic Generation (SHG), of the laser radiation within

birefringent crystals, such as KDP and ADP, results in

tunability as low as 217 nm.

Laser-excited atomic fluorescence spectrometry

(LEAFS) has been shown to be an extremely sensitive method

for trace element analysis (17-20). For the remainder of

this chapter, the use of LEAFS will be emphasized.









Atom Reservoir

It is generally stated that the atom reservoir is the

weakest link of AFS. In this respect, the use of lasers

as an excitation source cannot change the situation. The

ideal atomizer or atom reservoir for AFS must combine the

following characteristics: good atomization/ionization

characteristics, low background emission, freedom from

quenching species, long residence time, good stability,

high reproducibility, and ease of handling. The three

most common atomizers for AFS will be discussed briefly

with special emphasis on the ICP.

Despite their popularity, chemical flames have

serious limitations when applied to the production of

atoms or ions. Hydrogen-based flames generally give low

background emission and relative freedom from quenching

species, but atomization is poor and interference are

severe. The use of these flames has been restricted to

situations in which relatively volatile elements are

contained in essentially matrix-free solutions. For most

practical analyses, hydrogen-based flames are not useful.

Air-acetylene flames (21) are preferred in AFS, but

chemical and ionization interference often occur and for

refractory elements, poor detection limits cannot be

avoided. The nitrous oxide flames (22) have been used for

their superior atomization capabilities due to higher

temperatures. However, the background emission of this









type of flame is higher and can severely degrade the

signal-to-noise ratio of the measurements.

The advantages of the electrothermal atomizers, e.g.,

small sample sizes and low absolute detection limits, are

well known. Besides the high atom concentration that is

produced during the short atomization cycle, electro-

thermal atomizers offer the advantage of an inert gas

atmosphere by controlling the environment within the

atomizer cell. Also, the dilution effect of the analyte

by the internal gases is much less than that of analyte

species in flames. If scattering is eliminated (e.g.,

using nonresonance fluorescence lines) and the intense

background emission from the incandescent graphite is

properly screened, then LEAFS with electrothermal atomizer

may be the most sensitive of all atomic spectroscopic

techniques. Bolshov et al. (23) have demonstrated that

LEAFS with an electrothermal atomizer, which consisted of

a graphite tube, offered very high sensitivity for all the

elements they studied. However, when working with real

samples, Dittrich and Stark (24) have shown that the

electrothermal atomizer often suffers severe matrix

interference. Furthermore, the atomization of refractory

elements may be a problem and memory effects may occur.

In order to inhibit reactions between the analyte and the

graphite, Goforth and Winefordner (25) have utilized









various coatings such as the ones used for graphite

furnace atomic absorption.

The ICP is the best source of multielement excitation

in emission spectrometry (26). Because of the high

excitation temperature of the ICP, emission spectra are

very rich in atomic and ionic lines causing spectral

overlap of many lines to occur. In principle, LEAFS in

the ICP should be effective in avoiding these spectral

interference.

There are several reasons why ICPs have proven to be

very useful as an atomization or ionization cell for

fluorescence spectrometry (27-28). Its high temperature

and relatively long analyte residence time ensure a high

degree of atomization and/or ionization. These factors

increase the freedom from physical and chemical

interference and minimize scattering of the excitation

radiation. The argon coolant and aerosol carrier gas used

in the ICP provide a chemically inert environment which

reduce quenching effects caused by molecular species (as

compared to flames), thus increasing the fluorescence

quantum yields and the fluorescence signal.

The higher excitation temperature in the ICP

increases the population of excited lower levels of any

atom or ion resulting in a larger number of suitable

transitions for excitation. Since the classical

scattering problem which occurs at resonance transitions









can be eliminated by choosing nonresonance transitions, it

may be possible to choose an excitation wavelength which

is different than an intense fluorescence signal.

Therefore, low detection limits will be obtained since the

monochromator will not allow the laser radiation scatter

to reach the detector. The first study using the ICP as

an atomization cell for AFS was reported by Montaser and

Fassel (29).

The advantages of combining laser excitation and the

ICP as an atom or ion reservoir (30-33) are good

sensitivity, low detection limits, and high selectivity.

As explained earlier, there is virtually no chance of

spectral interference when laser-excited nonresonance

fluorescence is observed. Therefore, the wavelength

combination yielding the best detection limits can always

be used, whereas the wavelength yielding the best

detection limit cannot always be used in ICP emission

spectrometry (ICP-ES), where spectral interference often

limit the use of the most sensitive lines.

The disadvantages of the laser-ICP fluorescence

system include the complexity of the system, high cost of

the equipment, need for different dyes for different

wavelength regions, and rapid degradation of dyes. The

dye laser can only excite one transition of the analyte

atom at a time; thus it is basically a single element

technique.














CHAPTER II
PRINCIPLES AND NOMENCLATURE



Theoretical Consideration

The dependence of the atomic fluorescence signal upon

the excitation source, experimental configuration, quantum

efficiency, and concentration of analyte atoms has been

treated extensively in the literature (34-43). Only a

brief theoretical treatment applicable to laser excitation

will be given in this chapter.

The general expression for the fluorescence radiance,

BF, of a two-level system is given by (36)



Shv12 glnT
BF (-)Y21E (v2 c )B
12 (g1+g2)(E (v12)/E (v12))+g

(1)


where

E (v12) = spectral radiance of exciting radiation at
-2 -1
the absorption line, v12' Wm Hz

hv12 = energy of the exciting photon, J

c = speed of light, ms-1
c = speed of light, ms









B12 = Einstein coefficient of induced absorption,

m- J s Hz

Y21 = fluorescence power (quantum) efficiency,
dimensionless

9 = fluorescence path length in the direction of
observation, m

47r = number of steradians in a sphere, sr

gl,g2 = statistical weights of levels 1 and 2,
respectively, dimensionless

nT = total concentration of analyte atoms in

states 1 and 2, m-3.

The modified saturation spectral irradiance, E (12), in
-1 -2 -1
J s m Hz is given by

E cA21 2 1
E (v 1) = E (v1 ) (2 ) (2)
v 12 Y21B21 v 12 gl
21 21 1

where Es(v12) is the saturation spectral irradiance.

For a low intensity source E (v12) << E (v12



E (V ) 1
Bh(LO) h( E (v12)( 2 (3)
BFL (4 21V2 E(v12) 1


or substituting for E (v12) from equation (2)
hv12
BF(Lo) = (4-)Y21B12( c )E(12)4)

where A21 is Einstein coefficient for spontaneous

emission, transition s .









For a high intensity source E (v12) >> E (v12)


BF(Hi) = (-I)A2hv1 {f g2I}n
F4r 21 12 91 2 T


(5)


However, if a three-level atomic system is involved,

as would be for any stepwise fluorescence transition, the

general expression for the fluorescence radiance, BF, is

given by


B = (-)A hv n
F A47 32 23 T

1
g E (V13) g2 A +k
[1+ 3(1+ 13-)(1+ g exp(-AE /kT))+( 3232)
3 v 13 1 21
(6)


where level 3 is the upper level reached after excitation

and level 2 is the level reached after radiational decay.

The term E (v13) is defined by


E ) cA31 g3cA31
S 13 Y B31 3 1 31 13


where


B31 = Einstein coefficient of induced emission fo
3 -1 -1
process 3 + 1, m J s Hz;
B13 = Einstein coefficient of induced radiativee)

excitation for process, 1 + 3, m3 Js -1Hz;

Y31 = the fluorescence quantum efficiency for the

radiative process, 3 + 1.


(7)


r









Substitution of E (v13) into equation (7) results in

B (-)A32hv2nT

I. 1


S gl
[1+
g3


cA g232
+( B E(1 )(1+ 2-exp(-AE /kT))+(A32+k32)
31 13 Ev(v) 12 k21


For a low intensity source, E (v13) << E (v13)


g3E (v13)


BF(Lo) = ( I)A32hv23n


g1EV(v3)[1+ -glexp(-AE12/kT)]
v 13 91 1


(9)


Substituting for E (v13)


B (Lo) E(v13)
BF(Lo) ( )A32hv23nTY31 31 cA31


For a high intensity source,


g2
1+ 2 exp(-AE12/kT)
1(10)
(10)


E (v3) >> E(v13
v 13 v 13


BF(Hi) = ('- )A32hv23nTI
F 4 7r 2 23 TI


1+
g3


2 32+k2
+ exp(-AE 2/kT)+( 2)
g3 k21
(11)









According to equations (1) and (6), the most

important parameters affecting the fluorescence radiance

are the intensity of the source radiation, the

concentration of the analyte atoms, the Einstein

coefficient of induced absorption and the efficiency of

conversion of absorbed to emitted radiation. As indicated

by equations (4) and (9), the fluorescence radiance is

dependent upon the source radiance and the fluorescence

quantum efficiency as long as the spectral radiance of

exciting radiation of the absorption line is below the

saturation value. However, according to equations (5) and

(10) the fluorescence radiance becomes independent of

these two parameters when saturation of the upper level is

reached. The saturation effect of pumped transition has

been treated extensively in the literature (44). Due to

the complicated nature of a four-level system, it will not

be included in this discussion.

At low optical densities or low atomic concentration,

the fluorescence radiance is linearly related to the total

concentration of atoms in all states. However, at high

optical densities or high atomic concentration, the

relationship between the fluorescence radiance and the

atomic concentration becomes very complex (45,46).

At high concentration, ambiguities can be introduced

due to the geometry of illumination and observation. In

order to explain the various effects which may affect the









fluorescence signal at high concentration, consider a

square cross-section of an atom reservoir with uniform

atom number density and uniform temperature distribution

and a parallel beam of primary radiation as well as of

observed fluorescence radiation. Such a schematic

representation of the geometry is shown in Figure 1

where L+L is the pathlength transversed by the
pre
excitation beam which passes through the absorbing volume

and Z+.post is the pathlength transversed by the

fluorescence in the direction of the detector. Assuming

that the entire atomic population which is being viewed by

the detector is illuminated by the excitation beam, then

the cross-section monitored will be (1+L). However,

several effects may take place at high atomic

concentrations which may reduce the intensity of the

fluorescence signal.

Assuming that the whole cross-section is examined, a

process known as self absorption may occur. This process

can be described as the reabsorption of the emitted

radiation by the same atomic species which are being

measured. The result is a decrease in the fluorescence

intensity. In addition, a few specific cases may occur as

a function of the illumination and detection geometry. A

prefilter effect may result because of the absorption of

the excitation beam in a region that is not observed by

the detector. Referring to Figure 1, the prefilter effect



















I-
0


I-- I--
z





Sccw LLo
o CE 0 C

w m oc C


LJ

z
--5j
U)
0)
(I

0

3-
Li


t


i_






_-


M




Oc




.)0)








ON
C0)
0


4- a

-4




c Q.
Or









f -4
co




0
0 (0





0o
-4 0




0)
CU)l
~f0)









depends on the distance L pre; thus if Lpre = 0 the

prefilter effect can be eliminated. A postfilter effect

can also occur if there is a region between the

illuminated volume and the detector where absorption can

reduce the fluorescence intensity. Referring again to

Figure 1, the postfilter effect depends on the distance

post and it can also be eliminated if post= 0. The
post post
self absorption of fluorescent radiation by analyte atom

cannot be eliminated by altering the geometry of

illumination and detection. All three of these problems

only occur at high atomic concentrations; thus they can be

avoided by diluting the analyte solution.

Another way to explain the effects which occur at

high atomic concentrations is to use curves of growth. A

curve of growth in AFS is a logarithmic plot of

fluorescence intensity versus atomic concentration or some

related function. Zeegers and Winefordner (47) have

characterized the shapes of those curves both in theory

and experiments. As shown in Figure 2, hypothetical

curves of growth for AFS, may take on a negative slope as

a consequence of any of the three processes described

above. As a consequence of all three processes, the

curves of growth may even reach a point where no

fluorescence is observed by the detector.



















C



co
0
S(0



OO
*)43





I-




o Ow
00
(a (1 o0
1-4 -4 4-)
0 0(o
o u cI











bO
n a 0
w 0.. (- 0
w a: z c *I


















-d
a.
c- 0 a)
-li Li


W 0

--4


(q-


0o 0




0)
*-
CLi
Or-4






Li












lz
N. N. / CJ 0 co~

\. \ V o |
\ \ \ u ^
\ \. \ 'C"
X~~E \ \(J

X \ \ ^> (
X X X J C 0



3 -
^ <^









Types of Fluorescence Transitions

There has been much confusion among authors

concerning the nomenclature of the various types of atomic

fluorescence transitions. In order to solve this problem,

Omenetto and Winefordner (48) have suggested logical

nomenclatures for all of the different types of atomic

fluorescence processes. This becomes very important when

dealing with dye laser excitation because several

different types of atomic fluorescence have been

observed. There are basically three types of atomic

fluorescence, as shown in Figure 3: resonance

fluorescence (RF), in which the same lower and upper level

are involved in the excitation and de-excitation

processes; direct line fluorescence (DLF), in which the

same upper level is involved in the excitation and de-

excitation processes (but not the same lower levels); and

stepwise line fluorescence (SWF), in which different upper

levels are involved in the excitation and de-excitation

processes. Another type of atomic fluorescence which

involves a multi-photon excitation process will be

described in Chapter V.

In addition to these three basic types of atomic

fluorescence, a number of variations can occur. The

fluorescence process is termed Stokes if the excitation

energy is greater than the fluorescence energy and anti-

Stokes if the fluorescence energy is greater than the



















- -


LL























LL


_J

m


0



O


,-4
0


O
'-4



0


C,-
L


-4
E
0



0
C









o
4-
(0
--4

10




CO
0o
-I


-I


C(0






C,



-I


LL

r1l


4


I
m


,


!w






22


excitation energy. If both the excitation and

fluorescence processes involve only excited states, then

it is termed excited-state fluorescence. Finally, if the

excitation process involves a collisional excitation

following the excitation process, then it is termed

thermally assisted. The nomenclature which has been

described for atomic fluorescence also applies to ionic

fluorescence.















CHAPTER III
ANALYSIS OF RARE EARTHS



Introduction

The rare earths (REs) include those elements of

atomic numbers 57 through 71. This group of 15 chemically

similar elements are called "REs" because of their

relative scarcity at the time this name was proposed and

from the fact that their oxides resemble the alkaline-

earth oxides. The REs are widely distributed in nature,

but concentrated mineral deposits are found only in

limited areas. The principal source of the RE material is

the mineral monazite which is found in an alluvial sand in

Brazil, India and Idaho. This is essentially a complex

phosphate, containing as much as 75% REs. The method

selected for the extraction of the RE from the ore depends

on the nature of the ore and the presence of other

elements. They are generally extracted from the minerals

by decomposition in sulfuric acid to give the RE metal

salt solution. The RE metals and associated salts are now

neither rare nor unavailable.

The chemical similarity among each individual RE

element depends upon similarities in electronic










configurations in the atoms and derived ions. Because of

the essential constancy of the resulting outermost

electronic arrangement, beyond lanthanum, it becomes

easier for added electrons to enter a more deeply buried

arrangement (4f) where space for a total of 14 electrons

exists. It is the filling up of this 4f shell which leads

to the RE series. This phenomenon is reflected by

remarkable similarities in crystal structures,

solubilities, and chemical characteristics.



Analytical Measurements

Unfortunately, the REs have been least investigated

by spectroscopists. This is mainly because of the

difficulty in separating and purifying these elements and

especially because of the extreme complexity of some

spectra. The great similarity in the chemical properties

of the REs has made their separation difficult.

The analytical requirements for the determination of

the RE require that the instrument used for analysis must

achieve detection limits below 0.1 yg/g. Before the

introduction of ICP analysis, almost all published RE

determinations were done either by neutron activation

(49,50) or isotope dilution mass spectrometry (51). The

sensitivities obtained by these two methods are good but

they are slow and expensive. Other methods for the

quantitative determination of individual REs include










emission, x-ray, arc and spark spectra, and chromato-

graphy. The application of ion exchange techniques to RE

analysis has proven to be successful for their separa-

tion. Accurate RE analysis by x-ray spectroscopy has

proven to be difficult because of the comparatively large

size of the RE nucleus. Because of the relatively poor

sensitivity and also the great number of lines in arc and

spark spectrometry, it is difficult to measure individual

REs in a mixture. An example of the complexity of their

spectra is indicated by the arc emission spectra of

cerium. Corliss and Bozman (52) have classified almost

2,000 emission lines for this element.

In 1967, Ovchar and Poluektov (53) studied the atomic

emission of the RE in an air-acetylene flame. However,

their best detection limits were approximately 200 ppm.

Then in 1969, a major advance in the spectrometric

analysis of RE resulted from the work of Knisely, Bottler

and Fassel (54). They discovered that the fuel-rich

nitrous oxide-acetylene flame could produce and excite

free atoms of the REs. Their results showed much

improvement in detection limits due to the higher

temperature of this type of flame. Their best detection

limits were 10 ppm. Then in 1972, Crock and Lythe (55)

used the method of ICP-ES for the determination of REs.

Their best detection limit was down to 0.9 ppm. The high

sensitivity as well as rapid and reliable measurement of










REs by ICP-ES (56-58) provides an attractive alternative

to the conventional methods used in the past. It was

determined that when REs are excited by means of an ICP, a

high percentage of ions is achieved because the first

ionization potential ranges from 5 to 7 electron volts.

Therefore, the intensity of the ionic lines are likely to

be much greater than the atomic lines.

In 1980, Houk et al. (59) developed a new technique

called ICP-MS. In this approach, the plasma is not used

as a source of light but as a source of ions. These ions

are almost entirely atomic and are detected with a

quadrupole mass analyzer. This technique provides

excellent sensitivity, a simple spectra and the possi-

bility of measuring isotopes. Detection limits as low as

0.02 ppb have been achieved for the measurement of REs.



Uses and Applications

One of the oldest applications of REs has been to

improve the properties of high temperature alloys. The RE

metals have profound influence on the formation, shape and

distribution of nonmetallic fractions in iron and steel.

For example, in the steel industry, the addition of RE

improves the hot workability and increases resistance to

corrosion and oxidation (60). The addition of RE metals

has also been shown to improve the electrical and thermal

conductivities of copper base alloys (61).










Industrial utilization of REs is very important in

the glass and ceramic industries (62). They are

especially used for polishing glass and for the production

of heat resistant optical glass filters. Because of the

property of ultraviolet ray absorption possessed by the

REs, they have been used in the making of sunglasses and

welder goggles.

The determination of rare earths (REs) in silicate

rocks is of fundamental importance to modern studies of

petrogenesis. Rare earth distributions provide valuable

data in crystal and mantle evolutionary processes thus

helping our fundamental understanding of the processes by

which rocks are formed. The measurement of the

concentrations of these elements in various rock types has

initiated much research by geochemists in recent years

because it can provide them with the origins of these

rocks. Finally, applications of the REs in industrial

chemistry (63) include catalysts, promoters, phosphorus,

activators, and solid state physics and electronics.
















CHAPTER IV
ANALYTICAL STUDIES OF LASER-EXCITED
IONIC FLUORESCENCE (ONE-STEP)



Introduction

The goal of this study was to investigate the

usefulness of a pulsed tunable dye laser for excitation of

fluorescence of the RE in the ICP. It was believed that

laser-excited ionic fluorescence spectrometry (LEIFS) of

RE would overcome the spectral interference obtained when

doing trace elemental analysis by ICP-ES.

Although the emission of the RE in the ICP has been

amply described, the ionic fluorescence characteristics of

the RE in the ICP have never been reported. This chapter

will report on the ionic fluorescence of the RE elements

by pulsed dye laser excitation in the ICP.

Because of the striking simplicity of the RE

fluorescence spectra, which results with narrow bandwidth

dye laser excitation and gated detection, it is possible

to reduce greatly or even eliminate spectral interference

in a mixture of all the REs. Furthermore, detection

limits as low as 25 ppbs were achieved.









Selection of Transitions for Laser-Excited
Ionic Fluorescence Measurements in the ICP

From equation (7) it is possible to give an

expression which expresses the effect of each term upon

the selection of an optimum excitation and fluorescence

transition. Since most excitation and fluorescence

transitions which were investigated involved a three-level

system, only this example will be examined. If all terms

not related to either the excitation or fluorescence

processes are included in a constant K' and it is assumed

g2 AE12
that A31 < (k +k32) and exp(- kT ) is negligible,


then equation (6) is reduced to


E ( )
B = K'[(gA)v][(gA) ] ] (12)
BF K'[(g3A32 32 em g31 3 13 ex 2)
31 ( 31 k32

According to equation (12), it is desirable to select

an excitation wavelength which possesses a high transition

probability value, g3A31. It is also desirable to select

a fluorescence wavelength which possesses a high

transition probability value, g3A32. Therefore, the

selection of excitation and fluorescence transitions for

LEIF measurements in the ICP can be made on the basis of

their gA values. Of course it is also desirable to select

an excitation transition which originates from the ground

ionic state or a level near the ground ionic state.












The next step is to find a probable fluorescence

transition which originates from the upper level of the

excitation transition. If one cannot find such a

transition, then a fluorescence transition from a level

which is close to the upper level of the excitation

transition can be used. Once the excitation and

fluorescence transitions have been chosen, the

monochromator is set to the probable fluorescence

wavelength and the laser is scanned slowly in the region

where excitation is most likely to occur. Once a

fluorescence signal has been observed, then the laser is

tuned to obtain a maximum signal. At this point, the

excitation wavelength has been achieved and the emission

monochromator is scanned from 250 to 700 nm to determine

the most intense fluorescence wavelengths.

Corliss and Bozman (52) have determined experimental

transition probabilities and energy levels for several

spectral lines of the REs. These tabulated values were

very useful in selecting the wavelengths of excitation in

this thesis work. However, it is not possible to predict

exactly the optimum combination of wavelengths for

excitation and fluorescence. The best combinations of

wavelengths must be determined experimentally.










Experimental

Instrumentation

A block diagram of the laser-excited ICP system is

shown in Figure 4. The dye pumping source in the set-up

is a rare gas halide excimer laser (Lumonics, Ontario,

Canada, Model TE-851-S), operated with XeCl at 308 nm. A

maximum energy output of 60 mJ per pulse was obtained at a

repetition rate of 25 Hz. The laser beam energy was

monitored before each measurement with a power meter

(Scientech Inc., Boulder, Colorado, Model 380105).

A frequency doubled (KDP) tunable dye laser

(Lumonics, Ontario, Canada, Model EDP-330) was used to

cover the desired spectral range. One dye laser medium

was made by dissolving 0.199 g of DPS dye (Exciton

Chemical Co., Inc., Dayton, Ohio) in 1.000 L of dioxane.

A fundamental output of 200 to 250 kW peak power was

obtained between 400 and 412 nm. Another dye laser medium

was made by dissolving 0.184 g of rhodamine 610 dye

(Exciton Chemical Co., Inc., Dayton, Ohio) in 1.000 L of

deionized water. The third dye laser medium was made by

dissolving 0.119 g of PBD dye (Exciton Chemical Co., Inc.,

Dayton, Ohio) in 1.000 L of a 1/1 mixture of toluene and

ethanol. A frequency doubled output of 10 to 14 kW peak

power was obtained between 298-306 nm. The pulse width

and spectral bandwidth of the dye laser output was 8 ns

and 0.004 nm, respectively. Wavelength selection was





















0

0 )0
V) '






I I a
-4 E

C 0
I I --
0Oc















Ia
)l .-l































I S
r I 0 *o














cc
.54


I -.a


4- )
I" L _0 0 O










x 0
Id



*~~~ ~ _________X;
** ^ ~ ~ ~ ^ ~ ~' ~ d- I 1-
*~ 0
~ (0I










carried out with an automatic scanning unit (Compuscan,

Lumonics, Model EDP-60).

The ICP was a standard 27 MHz commercial model

(Plasma Therm Inc., Kresson, New Jersey, Model

HFP-1500). Typical operating parameters were incident

power of 0.6 to 1 kW; plasma support argon gas flow rate

of 12 to 15 L min-1; auxiliary argon gas flow rate of 0.5

to 1.1 L min-1; and viewing height above the load coil of

approximately 15 mm. The reflected power was minimized by

the automatic matching network in the power supply. The

sample was introduced into the plasma with a concentric

nebulizer. A nebulizing gas pressure of 32 psig produced

a solution uptake of 1.1 mL/min. The ICP torch was

mounted on an adjustable mount in order to facilitate

optimization of viewing heights without changing the

alignment of the laser and detection system.

As shown in Figure 2, the laser beam was directed

into the ICP by means of two plane mirrors. A third plane

mirror allowed a second pass of the laser radiation when

frequency doubling was used. Two small apertures (2 mm)

were placed in the excitation direction. One was placed

5 cm away from the output of the dye laser and the other

one 40 cm away from the plasma. Because of the slight

divergence of the laser beam, the laser beam diameter in

the ICP was measured to be approximately 3 mm.










The resulting fluorescence from the laser irradiated

volume was focused as a 1:1 image on the entrance slit of

the 0.35 m monochromator (Heath, Benton, Harbor, Michigan

model). A 3 mm slit height and a 150 pm slit width was

used. Neutral density filters were used to increase the

linear dynamic range. A photomultiplier tube (Hamamatsu,

Inc., Middlesex, New Jersey, Model R928) operated at -1000

Volts D.C. was used to measure the intensity of the

fluorescent radiation. The photomultiplier tube base was

modified for fast response (17).

Since the laser output was made up of 8 ns pulses at

a rate of 25 Hz, the output of the PMT was fed into a

gated integrator and boxcar average (Stanford Research

System, Palo Alto, California, Model SR 250) which was

triggered by a photodiode positioned to collect a portion

of the laser radiation. The gate width was set at 200 ns

and the gate delay at 34 ns. The output signals of the

gated integrator and boxcar average were displayed on a

strip chart recorder and measured with a personal

computer.

Chemicals

All RE standard stock solutions were prepared by

dissolving a sufficient amount of the rare earth oxide

(Spex Industries, Inc.) in a minimal volume (3 to 5 mL) of

hot hydrochloric acid and then diluting in enough










deionized water to make 100 mL of 1000 pg mL1 solution

(64). The only exception was for ceric ammonium nitrate

which was dissolved in water. Working standards were made

by serial dilutions from 1000 ppm stock solutions. All

solutions were stored in polyethylene bottles.



Results and Discussion

Analytical calibration curves were constructed for

the laser-excited ionic fluorescence (LEIF) of all the REs

except for promethium, because this chemical was not

available, and holmium, because no ionic fluorescence was

detected for this element. Analytical curves are given

for all the REs in Figures 5-17; detection limit of REs by

LEIF are reported in Table 1. All measurements were

obtained at nonresonance fluorescence lines. The use of

nonresonance instead of resonance fluorescence resulted in

a decrease in the detection limit by a factor of up to 10X

due primarily to the minimization of measuring scattered

radiation from the source. From the results shown in

Table 1, it is observed that for all REs, the ionic

fluorescence LODs are approximately 10X inferior to the

best one reported by ICP-ES (59) with a pneumatic

nebulizer.

In order to reach the wavelengths below 350 nm, it

was necessary to frequency double the output of the dye

























*







o
r-4





0 a
oL









00
C O





-9-'
So 0
c o




LOm



S0 -rc


0=

c,
So



















.-4


rLL
C :4
n^ a) !re
(

^ ^







, T-


0v


^ 0















0
0




0
x





















0 01
-4 1
C)









0
0













o
0











f o




S 0o
O o *





r--4

0 3-0
tC
,-I








CEI
0 0
SLa
r. 4J























--4
to

I CO















-0 C

C~~~~w>~~- insaulau~a~n j(
J"1" ,"" "" L s



2 2 '- ^

(ALU ^SSQ^Uj90U9S9J~|J
C +
< (





















E



0




0.











o 0
a C\O


C






co
00

















,-4 o
0 0























t co
00





r--4-1 4
I-4 0o

C 0
SI0 0













;4















.-I





-4
o r
x *-^ o
\ ~ 0 *- o
\r 4o
^ (O CO














'








*^(




















E

O
o
0







C o
(1) N
02
So0 o



0.
S





o-
0

Z< O 00




0 oo00
-4-






o o






-40


C
S|-0 eai <"













(Aw) I(!sululI 3ouoosJionl
O \. ^ .














0

c,


0
*4









0
-I


v
E
-3

) 0*





o
L



C Lo











0
.--- oC







O














S-4
a o




-:< c
I I I II I I I IIIII I I I
o Ca 0 "-
(Aw) psu :Ul 3 "n~l- h
\O J ^
\7 32
V ~ -
\O 0
X 0 Qf c<
\, *: -? 0 4-







41


















0 0
*o








0
O







C
uo
0)






C 00










c o
Q.



r-t


Cc


0r 0


Sa0x

*4









0
r-
\ ,0. ^
\ -- -. 0
\, 0 o *
\ : -b "^
\ 7- 1 c
\ >co
\ Q? ^
\ 0 :3

\ ^ "cO



\ 2\- *-i o

v +^







42
















S
0)





O0
*0E


e en




0.




0 o
*



-i -s
c






O H
00







CO


o 0 00





O
00




a-I 0
oil- f I

o 4J,
(A ) suu ouosaion cox














d0
Li
X c~
V 0^ ^
r r"1 ^






0,

r-


































































ia11 I I


0

gOUgOSaJOn


0




O*
-4









*
*4-






0
rq





0


o
a,




cL






o
DC
r-4





,-
04








.-1 )
co


0
0 -4









o)
04.-'




C3


-4




-- ---
t3-


r
Q








C
'- I-



"O
0>

CI

Q)
O


ol
C(-)
C

0

20


)

(ALL) TS~ua^u-


_______ ~_____~ _____~~ ~


III I I I I I h i 1 1 I








44







0
'-4


4--)
*-3


o
c)








C-4


0





0)
a)
3






-4E
E




c 0-





0 --4 -
0
oo.





a,

00

C








M- G
1-4 C0




00



*4-3
o m
















A 0u s o




















--4-,
44
n3 r

in ii i -i -- hn i i-i i --- "'' ^_ ^ L























*r-l

-P




0 t
c,
OCE
1Ln



o0 0



0 C1
C> c
00

C
























(A\) uolul *nl *o x
r. a(
bc
O -



CM



\0


ci)

0 1Z 3
~ 0
(~W) ~suaul aua~s~onlC6O



^ 0
c0 *







46
















-4






OE

oc4





0




.4 ,-s
00

.3-.3

no9



C
00



7:C
,ct
o (
C




4 *0






it I I I I I I it I I I I IIt1t a 4
a




-40
*l O c 0















LC'
-s0
\ en 3 ^
X r-1cfl
\ ? ^
\ ^'
'< *t-l *
X c -P
\ E- o o
\ -0.9










(IU1 '- cau~ l ~asao (MX

\ t-i
*- j3 m
*r-







47













E
S
S. *


E

C
a,















.U
0 0 0










oc
0.2 00

cc 00 '0









.,-4
o O










ri
*w 0






(A*) ,uUf C*
III 1 1 > 1 J 11 1 I I *s

o 2 o
(ALU ^ISQ^U)83U8S9Jr|J
*a: '














0
.,-l
4-3









(0
O



















0
4-O

0






C,
0*



Q,--
O(


"o o
O*-l 0





L.o


m

CC n
o




-4





ca 0
islt, I i I III a I I li tI I Ia C
* we
o 0 o
As 1 3








(LUJ) XISUa^U| a0Ua0SSJOn|J (d
a (03







49




4 .,r We O
4 3 I *
So o 0


4- C
C


OQ
4 *
4 -E O O L O, o o O
4 -)0 0 0 0 C0 0 0 L4

i43 C
a a)







x oI
LO





0) 0 co c C/
L0 <0-
0 c0




- CZ
0z r.1







oe i
0-4



a) M0 -M

0 1 c E.l m I0
0 n 0 om o-o .-r Oo N 0 ( Ln

0), ..4-) r_ I C'
.41 a 30 ot I Z; I -: I oI- '.o

0o I m in O o 0


O O
cl as 44 cz C I
E 94a

C cl i
0 -1

4-3

M Q O -

sA3 S = a' <4 m' o
4- 0 C 0 In C m a -

(3) cjo ? n m' Q Co 0


>2 -4 0- 4-) C) -o C0 0 0C ) 0 0 C 0 Qn
a)C) 1 cO m 0 T 0 C





-i 4-3
03) ( o o




3'- La4aS cd > (D S4 .0 z
(a C 4 E1 Q E






. c0) 03 (z CC a r-1
H- U













.0 CO







*. *-4 .r-
o ,) -4




0 e

OM b
4-) .-





o e\a)
Q) 0-/




4 0


0
CO




SOT
3 CL.
1-4
cy


CO
















o'
I




CY,



















o00
-I


















0
CM
o oo

000
in o
*u


in













C-
CM















Lr)
E.





















*COl
0






--cO










.0"
O















CM
^O0




00
0
00


E D 3
E- >4 C


C-







NI











O
























0
(M
































ta
C13

























0o



03'
C I
0















,r\j





ou


0







-r



Co















C's













r





CO
t-"












CY0 t-0
CM f
-(Y)


COS
C-

(0

a).-





OW
00
00

C 0



1-4 :


W -4





C 0
a)





> O
a -

a ..-1
03




a 4
cl z4-















0 0 0
C 0
II COi







a) =
a a) C
0 00

> C/



0 10



0 400a
C13 3 C


(0-




0) hE C






I) C0 0


a) IcO0
CCO
iz1 0I if

) 5


El
O-=O






a --o
z a)) -
,- C >*




U Ho
E-cbt


Cd



E


.o
C z
(0








a) 4.3






LLa
WCW










laser resulting in considerably less power for the

excitation. Whenever a good excitation transition was

available in the region between 400 and 600 nm, it was

used since it was possible to excite with the more

powerful dye laser fundamental.

The detection limit was defined as being 3 times the

standard deviation of the blank solution (water) divided

by the sensitivity (the slope of the analytical

calibration curve). The standard deviation was determined

by collecting 1000 data points at 25 Hz with a personal

computer. The high spectral selectivity is evident since

the blank signal was the same whether water or a

100 Ug mL-1, solution of a mixture of all other REs were

aspirated into the plasma.

In all measurements, no spectral interference

occurred when 100 pg mL~ of all the REs were measured.

In conventional ICP-ES, spectral interference in mixtures

of REs are serious problems, which necessitated some form

of separation prior to the measurements. The unique

spectral selectivity associated with laser excited

fluorescence is due to the rather narrow spectral

bandwidth of the laser, the number densities of the lower

levels of the RE species, and the quantum efficiencies of

the respective fluorescence transitions.

The analytically useful range or linear dynamic range

(LDR) is taken from the detection limit to the upper










concentration where the signal deviates by no more than 5%

from linearity. The LDR covers approximately 3 to 4

orders of magnitude. The strong curvature at approxi-

mately 1000 ppm for all REs, as shown in Figure 5-17, is

probably a result of the combined effects of primary

source absorption, self-absorption of fluorescence and

reabsorption of fluorescence radiation (postfilter

effect).

The pertinent ionic fluorescence transitions for each

RE as well as their relative intensities are shown in

simplified ionic energy level diagram in Figures 18-29.

The transitions shown were selected from NBS Monograph 53

(52). An ionic energy level diagram of terbium is not

included because we were not able to identify the energy

levels of the transitions corresponding to the

fluorescence lines observed. The transitions reported do

not represent the complete fluorescence spectrum within

the wavelength range studied for each RE, because weaker

lines could be observed at higher instrumental gain

settings. Although many other ionic fluorescence spectra

were observed at different excitation wavelengths, only

one ionic energy level diagram per element is given. The

use of transitions not involving the ground ionic state

was not a problem since the lower ionic energy levels

appeared to be highly populated. Excited lower levels may





























-9 -


-'


(8T)


(88)


(8T)


mu as5*886


SLh' Lee


u 6891T'88


m wsc-
cO (dno"
CD 1DO
(U U(UWU

























































0,
(U


(Tha)


SIS )


WU h9T'L8S


Eu S9'GT+il


Vn ) W em-e-
u in @om
(m u -.. ,s


(86) EU S2LS L0


(001)


9S'6688


(COT) *U 866*8ii T


(8 U) m u h *.8 e881i ~

____-_______- .11 .-- -4


(22) Wu


2.9 80+


mu S8L*L24


( D ) R6Jau3


(8e) 'U g886*88


(8Z) *U eS'sthi


(Tf) mU a56*588


0







CO
.-I C
r -4




I -,4

0U"









O .
0 >















Sr-4I
C 4
Cci,












co
0-4 0
S




C C
*r-1
-41
0.',


E-) 0
CU .







--I C
























1-
r4









-4 ^



cdl)c
c ^~
tu~~a


i I


r






















m 0I m0o) 0 0) cr
(0 (U Un (U (UU


san r a sem
i (U ) ( I
o Mn W) 9- 0 1
01 r 3 m cucu


(TI)'"" 8SS e ~


C8T)UU


(5-5)-


mUa- ti


(885I )u


(<8)ju 3 88"'86-;


ITh89L8


(

i:i:i q9sTJ.e 1FF111'


(8T)MU


86S'8h l


- 4-4-4-I-I-~


(51 )Y


BVl -uB


C'Al)"u 898*"hfh


t-1


( w D ) R 6ju3


*
0







-I >
o c
X-4







a)
*4 .
002




0 0


0












60 *O
c-,









S (0
o3)














Tri G
S4 C


1)4
HO

.,- I



-4
0)0
S*1











0 -- 0
co m




a--I





600
(D (0


















0)
t-
(0 < 4-
*^ i-l










*.-l


. .... ..-


IIt )"u s Lai'lh I


496'5Ti


SEa'Iah




















CU(- &S Q- wOC-W Ww
SSON y) ) I ( 0m a 0
c CU (CU U (U (UU
I I I I I I I I i I


M) r- M M

In r mU -


IU 888'868



WU -l




mu -94"868





mu aesgsa'
- -- heetS e -



U U86'SS9








mu 98e'S0


"u Shi'968


L I


ITT


I u 9S+'96e8 -- 1


'm W 8s'90 i


cag
CD Co
(U) (0

( w_ u RFi u3


0
*-4
4-N
co
0




I 0)


--4
a)w



-4
0 0)






o
E.-4

,>.
S C


0 I






La




E4-3
a)




>-0


a)




CO )





0a C
CC C




>Q)

S4
a)mr
ex;
t> <
r- CX
(0 D






















t- V- 0)
(U tI -
S :
0U (U (U


(0 0 (U
(D ( CU
0e S
If) I) m s


I(S'h) u 0899 *3+, I '


(ST) u 8g LS'8+


(6s) u 2s2'ati I-9


(8+h) lu" 8ES' fll








(0ST) mu 6088thS



mu set s6*99


(W_) R6 e u3


0


*4- c
4,-q (D
0


. CO
o





c
o 4o
CO



.- 0
a,
















o>





.-4
CO

t-)










0

-0 0
L*
cn


a)














4.
a)


()0 (
w-4




Cao



bO c

!i o
bOCa


I I(s'e>) u IssG'Ii I I






















wQ o) -y Ma)S
-021 WCUO CUw
Wo w no to ww :r


ccr- t
(W(D m
cJCU cU


+4+ *- r-


'MU GT0BL2


*u 8A61 *0L


"" 8LL4'~L


WU 6i8eI183


~u 8L o0 38I


uU LSE'598


mu 899"862


'MU esI 6s



Mu SL'L08S








'U SET ee S


' u
'MU 6T


S) (S
5) S
co m
en cu


(w u1) A8,auu3


0
*s *5
os















--?
0 0






















C4-4
-I
ca


0 0
c >


















(1)
o c




co







( 1)

















4-4
C\-
Ea







bo
, N




















S.r..i


r 1 r I r 1 I I I ...


'"' '


" "


O3 .1
I f






















0 O)


I I1 *I- I __


(-) u 664'68 *se




(0T) *U Ah8asee-- --


( u e '8 se


(Sh) '"" ihbi8"'L.




(-e) m 619tg 9se




eeOT> uu 88g'hS


(<9) mu h'8Se


S"-=1u h a'--


.o P-
:3- 0
' ) (
-*0
Cn(


ma

cu cu
IM I
0J 0J


0
*-I
C
0 -4

O



.--


S,-I a)
0 0a












,-I
oa)






O 4-










0"
a)


IC









ca

C-4
o a
















o
*> ,-4
03









) o
wco







C m
0
ao 0
o *






coa
a ca
-4 C







S-i
(U O C
C <





cd-id-1
4-30 (


( _w>) R6Jau3































C('S) "u 89h'69s








(IT> "" arm'ese --

-~~~~~ ~ is^w ~-a -
(TT) ,u Ogg-'Ogf








(< E) UIU 3




(88) Uu 886JL'4Oh



(coa) Nu OLPh'hSu



(92) mu hSe'Th



(T<) uu a-8'9ee


mu 8SL'LOh


I IJ L J~


IIi


-4


0


4-
*.-I CO
0 0)

o*-I
0.0




03


CI
OO









0)
O >


(CD

0CL
a?



O C









r*-
o ao











f- I
ca a
r0










U4
0)cP



0*M
Q)







Q) CO C
*-I C
CL% C





SI I


*


Sa uI I
(T-w : ) R80au3





















r- i- u (S 3- "

)D U :" s (o 'n 5 .,
) o) O) ) ) *
N U NU CUN U (cUU i 0
'4- 0)
4 **-I

'L) _u 8_'8_ D o-4




(TIh) Eu 9L .S8 ,*-)
x c


-up
0(1) *
(--) < mu Oah'Otk > o0







0
W--) mu ak8 "8- -l co'
e n)



(II) MU h8'8B





*:0 .
an



0 -0





(8 "h ) lu 83 9 'Tg9- ,_ i
0

(88) T'hE 0
>'-I






*0
4-W


<9) u 68h89, > '
CD CD o


1-40












Cr-
'4
























S (D C3 N o (Dt- o s -in
S (n0) CD S (U S 2 3-

m m m mm 5 5 5 m m A


0.4



*.,q
.S-I











O

4-I
a)
CO



--4
Ca





-4



4-)









-












>4-'






S40
r)o










0)
gu>













a))
3 (0










,-I












0)
-O-
(M rC
>rzl
tO-i

>> Lc
M (0,





<0 < a,
*^ -1
+3 4 >

(0 (0 (
cu;SQ


J CD
"~ (S
CiW) m


( r) R ejAu3




















(U (U (U
0 0) C(
0) (0
CU S 0
0 0 cu


I I I (88e) OU Se8-as a


(ES) *u ITTI *s


<(8TI) Wu LSe'8s -


(88) wu sTh'898


SI u ITTI' Se


c(U


uT- 3


In 2n
r 0m
m m


0
,-4

4-
cr

X)4
4.) 4.








0 >
0







0-4


., -I C





.o -





'C) 1-
ro
a.C





t0
,i




(1) C
Eu


04 >







(CJ


















bL

r-4
clQc


I I -I 1 (913) Mu Ise98' sa


II I I L

FA6 Ju3
























)CIQ Cosn mu -
cu r- 10N mU) 0m
M M in n o-
ic r In m n 1 0-


cu 0)

m- o
M .


'u 6S '-S~


1W)
o 0)
(U.4
-.4 0


mu -g'g*O






wu ~ L'h8














'Mu 38'6Te


-'u 88'96 --


Wu 99L'Le
WU fg'e


3- -
3 -4" -

( wu) ) Ra8u3'q


C
0 r



4-a



0 4
x -4




cbO










0-H
0 oa
SC
















ca,
O 4-




(0














0 0) 0
ao)
r= L








co























4 o>
bo
O C





U0






4-4


r-1




-4L
C%4 s-





















oiot-twwm 110


ai LI r I-=1


I 11( 2 )


MosIna


n; a 0 o:3a


mu gS8'Tth


(89) "u 6389e S
(se) "u" hSe'Shh



(c8) mu 8th'88 -


(08T) mu 9TL'8eLS -


(LS) 'm 98!T'99


(89) Iu 6L".'8e~


(U) Wu 89819G+


(8h) iu hfh'9hh


C(T) wu 88e'hh






(LS) wu 9 I998P
U9~) IUU 98! 998


0 C
4- )


a







C>
0 *-
0 4D











= z
C










0 *)
a4.























bo
-4-'C
























C-I
a)














>c
tC
C"l




002
*r^ (0



















-1I
cv1




CO C
O0 O
-i C





rz.
CD (
1-
(UV






r-l C:
(0 U 4

*^ r-


SI I









in fact be nearly as highly populated as the ground ionic

state as a consequence of the high temperature of the ICP.

In most cases, the sensitivity of the S-DLF or AS-DLF

exceeds that of the S-SWF or AS-SWF. Even though the DLF

signals were usually more intense than the SWF signals,

collisional activation-deactivation caused several SWF

signals to be intense. As indicated in equation (12), the

dominant factor which affects the sensitivity of either

DLF or SWF process appeared to be the transition

probability of both the excitation and fluorescence

line. A second factor which affects the sensitivity of

SWF appears to be the energy difference of the upper

levels between the laser-excited and fluorescence line.

The SWF can be very sensitive when the difference between

the upper levels is small. In other words, the degree of

mixing between upper levels is greater when the energy

difference of the upper levels between the laser-excited

and the upper level of the fluorescence transition is

small.

The degree of mixing between the laser-excited level

and nearby excited levels was examined for dysprosium.

The laser excitation at 407.798 nm corresponds to a

transition from 828 to 25343 cm-. The fluorescence

signal from upper level 25818 cm-1 to the ground ionic

state was three times higher than from upper level









28252 cm-1 to the ground ionic state. Since the gA values

for the two fluorescence transitions are identical

(3.7x10 sec-1), this will not be a significant factor in

the relative fluorescence intensities of these two

transitions. The only explanation for these two

transitions to have different fluorescence intensities

should be that the thermally assisted collisional process

which occurred after primary excitation was higher for one

level than the other. This thermally assisted collisional

process is often referred to as mixing between two

levels. Therefore, as expected, the degree of mixing

between levels 23543 and 25818 cm-1 must be three times

higher than for levels 23543 and 28252 cm-1.

In all cases when the fundamental dye laser

wavelength was used to excite the fluorescence, the pumped

transition did not appear to be optically saturated, i.e.,

decreasing the laser power by a factor of 2 resulted in a

decrease of the nonresonance fluorescence signal by a

factor of 2. Therefore, the fluorescence signal is

dependent upon laser fluctuations.

The precision (%RSD) of the technique at low concen-

trations is approximately 5%. This is determined by

measuring the relative standard deviation (RSD) of 1000

data points at a concentration 10 times greater than the

detection limit. At concentrations much greater than the

detection limit, the %RSD is approximately 3%.












Conclusion

The detection limits obtained by laser-excited ionic

fluorescence spectrometry should be adequate for trace

analysis (<1 pg mL ) of any RE in a complex mixture of

all other REs (100 pg mL-1) without spectral inter-

ferences. This technique did not require isolation of any

of the REs by the use of ion exchange columns or any form

of chemical separation which would most likely be required

for ICP-ES because of severe spectral interference.

Upon consideration of the relative fluorescence

intensities obtained in Figures 5-17 and according to

equation (12), the two most important factors which affect

the intensity of ionic fluorescence are the spectral

irradiance of the laser and the transition probability

value gA. Of course the population density of ions in the

lower level from which the absorption process is initiated

is also important. A second pass of the laser beam

through the plasma was advantageous since it was proven

that optical saturation had not been reached.

The laser-ICP system was effective for avoiding

spectral interference observed in ICP-ES. The spectral

selectivity offered by this method as well as the

tunability of the dye laser makes laser-excited ionic

fluorescence spectrometry a potentially useful method for






68


the analysis of REs in mixtures when high sensitivity, low

detection limits and high selectivity are desired.















CHAPTER V
ANALYTICAL STUDIES OF LASER-EXCITED
IONIC FLUORESCENCE (TWO-STEP)


Introduction

The goal of this study was to investigate the

fluorescence signal obtained when a second laser is

introduced into the analytical system. The fluorescence

signal using two-step excitation is compared to the

fluorescence signal using single-step excitation.

Two-step excitation offers the possibility of

populating high lying atomic and ionic levels which

usually possess more intense fluorescence characteris-

tics. The ability to populate high lying atomic and ionic

levels by two-step excitation is obvious. The use of two-

step excitation can shift the wavelength requirement of

the pump from the low ultraviolet region to wavelengths

typically greater than 300 nm where tunable radiation is

easily achieved. The high lying atomic and ionic levels

usually possess more intense fluorescence characteristics

because it is indicated in equation (12) that the

fluorescence radiance is proportional to the frequency of

emission v32. The fluorescence resulting from the two-

step excitation can be used for analytical purposes in a

similar way that was done for only one laser excitation.









Direct two-photon excited fluorescence using a single

wavelength and proceeding through a virtual intermediate

level, as shown in Figure 30a, have been observed

(65-66). However, the increase in sensitivity resulting

from the higher level of excitation achieved is offset by

the low probability of such transitions. This process

will not be considered in this study.

By using two tunable dye lasers operating at two

different wavelengths, much higher sensitivity can be

obtained since a real intermediate level can be used

rather than a virtual level (see Figure 30b). For this

two-step excitation process, the advantage of high

excitation energy is maintained and the transition

probability is much greater than for two-photon excitation

through a virtual level. Since the two excitation

transitions share a common level, temporal and spatial

coincidence must be assured in the plasma. Furthermore,

optically accessible excited electronic states usually

have short radiative lifetimes. In order to build up a

large enough concentration of the real intermediate level

in the two-step excitation scheme, the rate of pumping

must be large enough to overcome the radiative losses.

The use of pulsed tunable dye lasers should provide high

enough pumping rates to apply this technique to the

measurement of atomic or ionic species in the plasma.



























C
W


(a)


Figure 30.


(b)


Multiphoton processes: (a) two-photon, single
wavelength via a virtual intermediate level;
(b) two-step excitation.


1










Two-step excited fluorescence techniques have not

received much attention in the analytical literature

(67-69). Since there have been only a few studies

performed with the two-step excitation of fluorescence, it

appeared reasonable to explore this technique for the

measurements of REs in the ICP. Thus, one-step excitation

will be compared with two-step excitation.



Experimental

A block diagram of the experimental set-up is shown

in Figure 31. Two-step photoexcitation was achieved by

using two separately tunable dye lasers pumped with a

single excimer laser. The output of the excimer laser was

split between the two dye lasers using a 50% dichroic beam

splitter. Except for the addition of the second dye laser

(Molectron Corporation, Sunnyvale, California, Model

DL-II) and the necessary optics, this system was similar

to the one described in the previous chapter. The two dye

laser beams were directed into the plasma from opposite

directions. Mirrors were adjusted to provide maximum

spatial overlap of the two beams.

In order to achieve temporal coincidence of the two

laser pulses, one laser pulse was delayed with respect to

the other by sliding a retroreflector along an optical

rail. Temporal coincidence of the two laser beams was

then assured by examining their respective temporal















0



0 0
oa







I I e
o 0

--- -- 0








e z



L O.

I I o
r-i





I I (


















I L I a

-4-.
L0 0




Sj m-- 0 ,,
-I Q rO-










o 0




"-* 4 0







C%4
e rc













e-
















bO
*r-l










positions with a fast photodiode (Silicon Photoconductive

Detectors, EG&G Electro-Optics, Salem, Mass., model no.

FND-100Q) and an oscilloscope (Tektronics, model 2465).

The spatial coincidence of the two beams was achieved by a

visual alignment over the ICP torch and then by maximizing

the signal obtained by the scattering of each laser when

water was aspirated through the capillary. Both laser

beams were reduced to approximately 3 mm in diameter upon

entering the plasma by using apertures.

The output of one dye laser was used to generate the

first resonant frequency, X1, and the output of the other

dye laser was used to generate the second resonant

frequency, A2. Fluorescence transitions were measured

using the same detection system as was used for one-step

excitation study. The various laser dyes used for two-

step excitation of REs included coumarin 440 and 481 and

rhodamine 590, 610 and 640. The concentration and

solvents used were those recommended by the Exciton Laser

Dye catalog (Exciton Chemical Company, Inc., Dayton,

Ohio).



Results and Discussion

In two-step laser-excited ionic fluorescence (LEIF),

either of the two wavelengths can be scanned in order to

obtain a fluorescence signal. In practice, it was much

simpler to start by setting the first excitation












,








r-
a(













-4
a,


4-'







S,








O
0
CU






E4-










.l- 0
0





co
O4
Q) 0


























O,---t
*0-



4-) 0.
C.)
4-1






O,

0-'
0a,
0r 0





C\J















-4
0
CU












03


.0 .I



4


C
0









4O 4.- -
02















(0 0
0 4- CA
Ei







S.(D 4

Sr--
o ) 0 I








C -A





S- .E c\j
0 >4- 0 Z
>C












co
SE
E-

(0 4->rl10
*^( 0 >4c r
X (0 -^
M ^ 0
Ca


0

















0-i
0-


-=3
0c'J
0~


o c o (
0 moo mno oCN 0
-.- c.- m c-
.o = \o
PO in Lr


O'

o
o



w-C-
CT


CM
(00





I

(oo






CMl


00
0co

a.
-- .n

0
OcO

CM

(M


CO
0

SN
CM



0
\D


C3-
0



o-
c-
* I


C0


* I


Mn


o--=3

cO I
CM
0


.C 3 .0
^-1 >4i ^-


C\J
-=r


00
CO


c- c\j


C~\J

0

r:-C\j
0-I


(0




C

>
6-0
c -4

> 4-

CU -O
0

C-


-CC
040

a,4 a,
CUcWa)
(U4WS


II


3


C






















4--
0
4-'
(0
4-'



a,



a0

,-.











0)

















CO
t-



-I
-4







0












-14
>-1




















E-O
S0,









m4
c-.



02



0(







-a,





-4
4I


I-4




c-IaC










wavelength by monitoring one of its nonresonance

fluorescence transition. Once this transition was

determined, the second laser was scanned. It was also

much easier to find excitation transitions which did not

require the use of frequency doubling in order to avoid

the difficulty of tracking the angle-tune double crystal

with the wavelength of the dye laser.

Analytical calibration curves were constructed for

the two-step LEIF of lanthanum, europium, ytterbium and

lutetium and are given in Figures 32-35. One can see

that, as expected, the linearity of this technique is

similar to the single-step fluorescence.

No enhancement in the fluorescence signal was

achieved when comparing two-step excitation to one-step

excitation. In fact, the detection limits using single-

step excitation were lower than the ones obtained using

two-step excitation. One explanation may be that when the

two dye lasers were used, the excimer laser beam had to be

split into two beams using a 50% beam splitter. In so

doing, only half of the pump laser beam was directed into

each dye laser. It was experimentally proven that if

intensity of the pump laser was decreased by a factor of

2, the output of the dye laser was decreased by a factor

of 4. Therefore, four times less power was available in

order to excite the various transitions. Since optical

saturation has not been reached, the signal-to-noise ratio







77















*


CO



00
L'-I







oco
*,-4
0 c




0O








0 c co



-, 4

co c.




















0 0
*a
Ex..






e-4 0-b
0 0 0 --
0













m
rz.o
[x







78













LC

*O




0 0
eo




CO .
0 0o






0










C4 C)
q-4
0 0



O







0- C.
c On



Ol e



.0=o1




m
0







C

Ililt I l I I I li lt II I 1 0 4
o 0





W- -4 o


a)
T-'^-"'-'i



0I








79










E




oo
Sr-





















Cc
0








0 o







. O c













oc
S 0















c 4-*C
00



o 0 m






0
*
o












. 0 e
QO o c
0 ~C 0
-, (Jo



















e
C

o
Eo









r-4
OO

eo
= *









E M
O
*-4 0'







CO
o0 0
O o





o,















L
S-r






U
o 0o o






60


















-4 .
M 0 opo
TO0 ^
^"K ^?:
0 ~o ir





0 c <\j
oc ^ '
r ^(^ *^ m
O -^ ~








09









and thus the detection limits would improve by increasing

the output of power of the dye lasers.

Partial energy level diagrams were constructed for

the two-step excitation of lanthanum, europium, ytterbium

and lutetium (Figures 36-39). As shown in Figure 36, the

first laser provided the excitation of lanthanum ions from

1394 cm-1 to 23247 cm-1. The second laser then provided

the excitation of the ions from 23247 cm-1 to 40458 cm-,

and the fluorescence was monitored at several

wavelengths. In most studies concerning two-step

excitation, the level reached by the first laser is the

starting level of the second excitation transition.

However, as shown in Figure 37, the two-step excitation

scheme can be used even when the two excitation

transitions do not share a common level. This scheme was

only applied to lutetium ions but it could be extended to

other REs. As indicated by the higher detection limits,

the lower level of the second excitation which was

indirectly coupled is probably not as populated as if it

had been directly coupled.



Conclusion

The two-step fluorescence technique permits direct

access to highly excited levels whose thermal population

is low even at the elevated temperature of the ICP. It

is, therefore, expected that the emission signal, which in



















0D0
n v
S)0
00)

i i


-1-4-4-4-4 .4......


IT "8)


(8L)


"m 8e8'Se~


*U L8ea'64


*u S60'ee4i


(88T)


e < a-
I I mI
(UBY)<


ITJ)>


n C- 0



-UI-a I I I


mu" 9S.L T1I Ti


---a> e_ se__ s_ I


1J.I '


(G'S)


(9a)


(9'9) wu


m' 66S' h+i


,"" 2881' G


*u L6SS'89


j~gg ee.


(OL) "U 9S'6ih


SI II I 60 8'l9


IVtm Ils84~


( t- wJ ) RBJ u3


CL S) ** 8LE
'
Sth


0
*
0
2 0 ..> .- -

.,-


co
0) .'-



0
















----- *l0
S4-)>
















C 4
ca









0,











>0 r-
.,- 0)









c ,




:>-H
to
S4







M40
( )












*'- C.C


I 1





m
(0


0 w


'H4>0





(b tO n
OL-^ Q


I


-


a r 1 9


-


I I r


I (LI )


q





'


i























"cc .4Wd(U ) ~ Gn


I I II I I I I I IU ~


WWWQ-.4SW
?()OCt-t-(
WWWW.QIU.4-4c


88T) *U"8*'e88


(88) e8 TBI'9s~


Si1)


- -CS)


wTlsBBe'hI


wVg9 *"88


-t--I- -tI .-. .--~_.. .


(il )


*ULTB' gg


(TI) *~ h' -








CS) *"ggT"eee


(9) *u88"888


C<) "u8LLS L a


1 1 1 1 1 1 1 1 1 ** ne-sed I


u 9S'


s
s
a
U
S1


(<- W) fBuiu3


C
*


0 -(
(0 *>



x -'












0 cD
so
O


*CC






02 *-














'4-O
0 0Q
















*-4
Cr














CD boe





-4a
I I I
Y,,




-a
-- >1 s-'







' "* *^ i 1 -1






00~a
(UCQ
t-i~o
3 0a
bOC-
*r -l )



e-,


h





















t o (* o p -i
r-o 10w v ro u
IU) ) I) I I U


(UO)U)WO(UCo (UaWJ( )W

n "m P- M u a c
m m m m m o w N IN w

I l 1 1 I I I I


CU2) '868*


(8) *e88 tse


(Gi) Sueg.L8ee


tT) inuamena,


="Leih'She




(88) *u88*898

81) UVS"G8


(C) *ug989"tl


(9) wU3aB*"hsG


(L) UTg "Svgl


I 1 e* sel I1


4"- 4.-


88
*" 85-B



5
(U
I


0


4-' cI)




0.C
0





0-
O













*4




:0 r-











*T <-
r-4




30
> r











->1
W-I
0)
>> CC






,-
to












1-4
d24
hO*-









(D 4- C





dlC
CO U -


(t- w ) f6 uu3


\ bft


--IPP ~OC


i- "




















m 3-cu oo
t-U)U2
CwuYu,
OD( I) a
GO m


a u

we *.


am OF-
#I 1


(8"G) mu -8h'88 8

CL.'6) uu fi'6 g --



(99) "* s9'*IS



(8t) mu M OO"S"_ -


i 'h88i


(B'i) mu -g6*' g -"--


C*S) mu ggg88g -- -- ---


("i) mu 81~68 -98 -


(TI) mu *sh 8tt;e --
+1 u


I I* 6he *' 1l


*"u 28e*9+1sl I


D -S

0 z

caa
-* *-4 *.-H


*H *I






4-)
0 (D
.t,










40 --f
C




am


03 0
(D *P













a
o CI
































I



.4
4-









to


,-4
Cr-1





E *O
(0 (0Q
D ol~
c&O)^




f-1r
cd)~
>> t-

00 (
s-
0 r
c c<
B b (


r
(t_ ur>) ABJ~u~


C T) "


Cht ) ** 888 988










this case is a source of noise, would be less and thus the

signal-to-noise ratio would improve. The emission signal

for one-step excitation is expected to be higher than for

two-step excitation because the thermal population of the

levels which can be reached by only one laser becomes more

significant. Therefore, the advantage of exciting the

ions to a higher energy level is mostly a decrease in the

background emission noise from the ICP.

It was shown in the Results and Discussion section

that the spectral irradiance of the lasers for the two-

step excitation process is approximately four times less

than the spectral irradiance of the laser used for one-

step excitation. Therefore, the fluorescence radiance is

expected to be four times less, and thus the detection

limits for the two-step process should be four times

higher than for the one-step process. However, the

results obtained indicated that the detection limits for

one or two lasers are similar. Since the gA values for

the excitation and fluorescence processes of the one-step

and two-step technique were similar, the background noise

must have been less for the two-step process than for the

one-step process.

Since it is possible that two elements may be excited

at the same wavelength with only one laser, the addition

of a second laser offers the advantage of increased

selectivity. In other words, the probability of a









contaminant being excited by both dye laser wavelengths is

greatly reduced.

Even though sustained enhancement of the detection

limit was not achieved using two-step excitation as

compared to one-step excitation, this technique may become

important as a tool for the study of the excitation

dynamics in plasmas. However, it is predicted that the

detection limits measured using two lasers could be

improved by increasing the output power of the pump laser.

The use of single-step or two-step LEIF minimizes

some of the severe interference problems which occur by

ICP-ES. A common problem which was eliminated was the

overlap of spectral lines of the various elements present

in the sample. The analytical results obtained for

single-step and two-step LEIF of the REs in the ICP have

indicated that it may be the method of choice for the

analysis of REs in a mixture when high sensitivity as well

as very high selectivity are desired.















CHAPTER VI
FUTURE WORK--IMPROVEMENTS


Future studies which should provide some improvements

in the detection limits obtained by one- or two-step laser

excitation of ionic fluorescence of REs in the ICP include

the use of lasers which possess more power. For example,

the output energy of a dye laser pumped with a Nd:YAG

laser using Rhodamine 6G can approach 50 to 100 mJ per

pulses. However, the output energy of the dye laser used

for this thesis work was approximately 3 mJ per pulses for

the same laser dye.

Furthermore, it would be interesting to try to use a

high intensity continuum source such as a pulsed xenon

flashlamp in order to investigate both one- and two-step

excitation of REs in the ICP. The estimated output energy

of a pulsed xenon flashlamp source over the absorption

linewidth of an atom or ion should approach 1 mJ. It will

also be interesting to find out the results of Demers and

Shrabak (8) for excitation of atoms and ions in the ICP

using the new boosted-output HCLs which are discussed in

the introduction.

Another method which would facilitate detection and

also give much more information for various excitation




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 E7O1T6YAI_7EHVA8 INGEST_TIME 2012-02-17T16:52:32Z PACKAGE AA00003800_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES