close

Вход

Забыли?

вход по аккаунту

?

Investigations of nonmetal charge transfer using helium microwave -induced plasma time of flight mass spectrometry

код для вставкиСкачать
INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI films
the text directly from the original or copy submitted. Thus, some thesis and
dissertation copies are in typewriter face, while others may be from any type of
computer printer.
The quality of this reproduction is dependent upon the quality of the
copy submitted. Broken or indistinct print, colored or poor quality illustrations
and photographs, print bleedthrough, substandard margins, and improper
alignment can adversely affect reproduction.
In the unlikely event that the author did not send UMI a complete manuscript
and there are missing pages, these will be noted. Also, if unauthorized
copyright material had to be removed, a note will indicate the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by
sectioning the original, beginning at the upper left-hand comer and continuing
from left to right in equal sections with small overlaps.
Photographs included in the original manuscript have been reproduced
xerographically in this copy.
Higher quality 6” x 9* black and white
photographic prints are available for any photographs or illustrations appearing
in this copy for an additional charge. Contact UMI directly to order.
ProQuest Information and Learning
300 North Zeeb Road, Ann Arbor, Ml 48106*1346 USA
800-521-0600
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ABSTRACT
Name: Pamela Rose Keating
Department: Chemistry and
Biochemistry
Title: Investigations of Nonmetal Charge Transfer Using Helium Microwave
Induced Plasma Time of Flight Mass Spectrometry
Major: Chemistry
Degree: Doctor of Philosophy
Approved by:
Date:
^
P ft. h
ir a M n r
Jirfsertationn CoDirector
f
/y i& e .^ - 4 u *L A 'p jm ,
Dissertation Co- Director
J
'X- 0 <}!
f
$. ^ 0
0
/
Northern Illinois University
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ABSTRACT
In helium microwave induced plasmas (He-MIP), the +1 ion emission
lines are more intense for most nonmetals than the atom emission lines. The
most likely reason for this behavior is charge transfer (CT) between the
helium ion and the neutral ground state nonmetal, producing an excited state
nonmetal ion. The proposed mechanism is given in the following equation:
He*(24.56 eV) + A°(0 eV) -4 He°(0 eV) + A**(24.56 eV ± AE)
A represents the analyte,0 and * indicate the ionization states, and * indicates
an electronically excited state. AE is the difference between 24.56 eV and
the excited ionization energy of the nonmetal.
The CT phenomenon with nonmetals in He-MIP was studied using
atomic emission spectrometry (AES). AES gives information about
populations of specific atom and ion electronic states. On the other hand,
mass spectrometry monitors the entire population of ion states. Using results
from these techniques, a more complete picture of charge transfer and
nonmetal ionization processes can be produced.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
To examine this mechanism, the helium ion population was reduced by
adding a more easily ionized gas to the plasma. The addition of 0.5% Ar to
the plasma decreases the intensity of the C f* emission line by 60%, while the
Cl atom line signal increases by a factor of 20.
In this work, a time of flight mass spectrometer was constructed and
used to monitor the total chlorine ion signal. Upon addition of 0.8% Ar, the
chlorine total ion signal decreased 63%. A chemical kinetics simulator was
also used in these studies to model the plasma kinetics. Results, based upon
the CT model, match the experiment well. It was only possible to achieve
agreement with experimental results using 8 reactions for chlorine. Similar
results were obtained for bromine.
The simulations verified that CT is the dominant mechanism for
nonmetal ionization in a helium plasma. In the course of modeling the
behaviors of the atoms, additional plasma chemistry was clarified. An argon
metastable state appears to be responsible for excitation of the monitored Cl
atom line.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
NORTHERN ILLINOIS UNIVERSITY
INVESTIGATIONS OF NONMETAL CHARGE TRANSFER
USING HELIUM MICROWAVE INDUCED PLASMA TIME OF
FLIGHT MASS SPECTROMETRY
A DISSERTATION SUBMITTED TO THE GRADUATE SCHOOL
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE
DOCTOR OF PHILOSOPHY
DEPARTMENT OF CHEMISTRY AND BIOCHEMISTRY
BY
PAMELA ROSE KEATING
DEKALB, ILLINOIS
AUGUST 2001
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UMI Number: 3023693
— __
UMI
UMI Microform 3023693
Copyright 2001 by Bell & Howell Information and Learning Company.
All rights reserved. This microform edition is protected against
unauthorized copying under Title 17, United States Code.
Bell & Howell Information and Learning Company
300 North Zeeb Road
P.O. Box 1346
Ann Arbor. Ml 48106-1346
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Certification:
In accordance with departmental and
Graduate School policies, this
dissertation is accepted in partial
fulfillment of degree requirements.
Dissertation Co- Director
Date
Dissertation Co- Director
AutiiAcI' % tkO O /
Date
J
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ACKNOWLEDGMENTS
I would like to thank my dissertation advisors, Dr. Jon W. Carnahan
and Dr. Lee S. Sunderlin, for the knowledge and wisdom that they graciously
shared during this project. I would also like to thank the members of my
committee: Dr. David S. Ballantine, Jr., Dr. W. Roy Mason and Dr. Dennis
Brown.
I would also like to thank Dr. Gary M. Hieftje for allowing me to do
undergraduate research which showed me there was a truly enjoyable side of
chemistry.
I would like to extend my heartfelt gratitude to Debashis Das and Gary
White; without either one I would not have had enough energy to complete
this work. Thank you both.
Lastly, I would like to thank my mom and my brother Kevin, both of
whom believed in me so much that I began to believe in myself.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
DEDICATION
For my mom, thank you
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE OF CONTENTS
Page
LIST OF TABLES.....................................................................................
ix
LIST OF FIGURES...................................................................................
x
Chapter
1. INTRODUCTION...................................................................................
1
1.1 Introduction to Elemental Analysis......................................
1
1.2 Introduction to Plasma Sources...........................................
3
1.3 Monitoring Species with AES and MS...............................
10
1.4 Introduction to Charge Transfer.......................................
11
1.5 Direction of Dissertation...................................................
15
2. CONSTRUCTION OF THE NIU HELIUM MICROWAVEINDUCED PLASMA TIME OF FLIGHT MASS SPECTROMETER.
16
2.1 Introduction to Time of Flight Mass Spectrometry
16
2.2 Principles of Time of Flight Mass Spectrometry......................
19
2.3 Physical Considerations for
Time of Flight Mass Spectrometry.....................................
23
2.4 Time of Flight Instrumental................................................
26
2.4.1 Previous Instrumentation.....................................
26
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Chapter
Page
2.4.2 Current Instrumentation.......................................
28
2.4.2.1 Mass Spectrometry Vacuum System. ..
28
2.4.2.2 Transport of Ions.....................................
30
2.5 Details of the Improvements.............................................
33
2.5.1 Instrumentation....................................................
33
2.5.2 Ion Transport........................................................
39
2.5.3 Microwave Induced Plasma..................................
44
3. ANALYTICAL CHARACTERIZATION OF THE
TOF-MS INSTRUMENT.................................................................
46
3.1 Introduction........................................................................
46
3.2 Reagents...........................................................................
46
3.3 Sample Introduction...........................................................
47
3.3.1 Ultrasonic Nebulization Sample Introduction . . . .
47
3.3.2 Gaseous Sample Introduction..............................
48
3.4 Flow Rate Optimization.....................................................
48
3.5 Actual vs Calculated Flight Times.....................................
49
3.6 Preliminary Mass Calibration............................................
50
3.6.1 Mass Calibration of Methanol and
Deuterated Methanol.............................................
50
3.6.2 Mass Calibration IA and VIIA Elements.................
56
3.7 Resolution and Peak Shape Optimization.........................
59
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
vii
Chapter
Page
3.7.1 Pulse Duration.......................................................
59
3.7.2 Pulse Voltage.........................................................
61
3.7.3 Completed Optimization........................................
63
Conclusions......................................................................
66
4. INVESTIGATION OF THE KINETICS OF CHARGE TRANSFER.
68
3.8
4.1
Introduction.......................................................................
68
4.2 Charge Transfer..............................................................
68
4.3 Sample Introduction..........................................................
70
4.3.1 Sublimation Chamber............................................
70
4.3.2 Gaseous Sample Introduction..............................
71
4.3.3 Simulated Aqueous SampleIntroduction...............
71
4.4 Atomic Emission Spectroscopy Experiments....................
71
4.4.1 Chlorine.. * ..........................................................
72
4.4.2 Bromine.................................................................
74
4.5 Mass Spectrometry Experiments........................................
77
4.5.1 Chlorine.................................................................
77
4.5.2 Bromine.................................................................
79
4.6 Plasma Kinetic Modeling....................................................
83
4.6.1 Simulation Settings...............................................
83
4.6.2 Explanations of Equations.....................................
84
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Chapter
Page
4.6.3 Discussion of Results..........................................
89
4.7 Discussion of Results.......................................................
94
4.7.1 Chlorine................................................................
94
4.7.2 Bromine................................................................
97
5. CONCLUSIONS AND SUGGESTIONS FOR FUTURE WORK. .
100
5.1
General Conclusions.....................................................
100
5.2 Future Work.....................................................................
101
REFERENCES.................................................................................
103
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF TABLES
Table
Page
1.1 Ultraviolet and Visible Detection Limits of
Selected Elements in ICP-AES, ICP-MS and M IP-M S................
8
2.1 Operating conditions of the NIU He-MIP-TOF-MS......................
29
3.1 List of ion optics, voltages and distances for the purposes of
calculating a total flight time for chlorine.......................................
51
3.2 Compilation of the species, mass and times from the methanol,
deuterated methanol spectra........................................................
55
3.3 Compilation of the species, mass and times from the spectra
of solutions in methanol containing the IA and VIIA salts
57
4.1 Listing of the reactions and rate coefficients that were used in the
simulations for chlorine.................................................................
85
4.2 Listing of the reactions and rate coefficients that were used in the
simulations for bromine.................................................................
86
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF FIGURES
Figure
Page
1.1 Energy Diagram of selected nonmetals. Each of the points
represents an excited state of the nonmetal ion with the
lowest point for each element being the first ionization energy.
Each of the states is given in electron volts above the ground
state of that element.....................................................................
13
2.1 Block Diagram of NIU-TOF-MS as of June 1996 ........................
27
2.2 Block Diagram of current NIU-TOF-MS......................................
30
2.3 Diagram of lenses. A, sample cone. B, Skimmer cone. C,
First element of sample introduction einzel lenses, cone. D, E,
second and third elements of sample introduction einzel lenses.
F, final element of sample introduction einzel lenses, barrel.
G, pulse plate. H, extraction grid. I,J,K, elements of flight
tube einzel lenses. N, M, deflectors. 0 , liner...............................
32
3.1 Spectra of methanol (lower trace) and deuterated methanol
(upper trace) with peak assignments............................................
53
3.2 Calibration plot of methanol, CH3OH and deuterated methanol,
CD3OD..........................................................................................
54
3.3 Calibration of the IA in methanol, R2=0.99995 ...........................
58
3.4 Calibration of the halogens in methanol, R2=0.9997 .................
60
3.5 Molecular bromine peaks during optimization of pulse
duration from 600 to 850 ns..........................................................
62
3.6 Chlorine at several voltages during optimization of pulse voltage
64
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure
Page
3.7 Halogen spectra prior to pulse optimization...............................
65
3.8 Halogen spectra after pulse optimization....................................
67
4.1 Chlorine results from UV-Vis Argon doping experiment.............
73
4.2 Chlorine results from VUV Argon doping experiment ...............
75
4.3 Bromine results from UV-Vis Argon doping experiment.............
76
4.4 Dry Chlorine results from MS Argon doping experiment
78
4.5 Wet Chlorine results from MS Argon doping experiment
80
4.6 Dry Bromine results from MS Argon doping experiment
81
4.7 Wet Bromine results from MS Argon doping experiment
82
4.8 AES neutral data of bromine (lower data) and chlorine
(upper data) the UV-Vis data is squares and the VUV data is
the closed triangles.......................................................................
91
4.9 AES data for ionic bromine (upper data) and chlorine
(lower data). The UV-Vis data is represented by open triangles
and the VUV data as squares.......................................................
92
4.10 Mass spectral data of bromine (upper data) and chlorine
(lower data) with the simulation data (solid lines)........................
93
4.11 Chlorine transitions that were monitored in the VUV and
UV-Vis Ar doping experiment.....................................................
95
4.12 Bromine transitions that were monitored in the VUV and UV-Vis
Ar doping experiment...............................................................
98
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Chapter 1
INTRODUCTION
1.1 Introduction to Elemental Analysis
There is growing concern about compounds containing
chlorine and other halogens in drinking water,1the atmosphere,2the water
table3 and in sediment.4 Drinking water is purified in a routine manner for
waste products and a significant amount of contaminants. However, a
portion of pharmaceuticals excreted and put into the sewage system are not
filtered out during the purification process. Pharmaceuticals such as over
the counter diuretics (chlorthalidon), muscle relaxants (enfluran and
isofluran), tranquilizers (diazapam), antihistamines (brompheniramine), and
household cleaning products (chloramine 80) are often present in such
minute amounts as not to be a significant problem in and of themselves.5
However, each of the listed compounds contains halogens that can be
metabolized by algae and halophilic bacteria.6 They are often broken down
into molecular subunits, and are present at concentrations that can
adversely affect humans.7 Halogenated compounds also enter the air often
via exhaust from factories that use compounds containing halogens (HCI,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
HBr, HI and chlorinated fluorocarbons) in the manufacturing process. Once
airborne, these compounds interact with other airborne species. Interaction
with species in the air can lead to degradation of the ozone layer.
Interaction of chlorinated fluorocarbons with ultraviolet radiation produces
atomic chlorine which, is an effective catalyst in the process of 0 3 breaking
down into 0 2 and an oxygen atom with two unpaired electrons, radical
oxygen. Radical oxygen reacts with 0 3 to continue the ozone degradation
process. 0 3 is the main barrier for harmful ultraviolet radiation from the
sun.8 Sulfuric, nitric and hydrochloric acid exhaust from manufacturing can
become part of cloud cover and eventually acid rain. The contaminated rain
water mingles with ground water where water dwellers both consume and
live in the contaminated water. Contaminants can become part of the
sediment and land biota may consume them. Bodies of water, the water
table and sediment can also be contaminated with run off from pesticides
and herbicides (phenoxypropioic acid) or road work (polychlorobiphenyls,
PCBs).9 Therefore, the need for reliable, reproducible and increasingly
more sensitive techniques with the ability to detect lower levels of
contamination is greater than ever before.
For common trace elemental analysis techniques, the most
challenging elements to analyze are nonmetals. Detecting nonmetals
reliably, quickly and cheaply is very desirable. Common optical techniques
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3
such as flame atomic emission and flame atomic absorption spectroscopy
have a variety of drawbacks. The most noteworthy challenge with flames is
that insufficient energy is imparted to the analyte. Without enough energy to
atomize and excite all elements, the dynamic range of the instrument
becomes quite limited.10 These techniques are prone to matrix
interferences, making universal calibration difficult and unreliable. There
are often spectral interferences that give rise to complex backgrounds,
making detection limits in real samples too high for practical value.11,12 Most
techniques do not allow for simultaneous multi-element analysis, making
larger sample sizes necessary and analysis times longer.
However, plasmas possess a number of advantages compared to
flame systems. Not only do plasmas exhibit fewer matrix interferences,
possess a much simpler background spectrum, and have a substantially
longer linear range, they have sufficient energy to produce nonmetal ions as
well as electronically excited atoms and ions.
1.2 Introduction to Plasma Sources
Plasmas are one of many classes of sources for atom ionization and
excitation for mass or atomic spectral determinations. Plasma sources are
advantageous because their associated techniques produce long linear
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ranges, few matrix interferences, simple spectral backgrounds, and plasmas
have sufficient energy to excite or ionize a broad range of species.
A plasma is a superheated, ionized gas, often referred to as the
fourth state of matter.13 In general, states of matter are a function of
temperature. The coolest state is a solid; add heat to most solids and they
become liquids, add heat to most liquids and they become gases. If a
sufficient amount of energy is added to a gas and a significant amount of
ionization occurs, the gas may become a plasma.
An additional parameter describing plasmas is the number density of
electrons and the distance between the electrons and the resultant ions.
When an atom is heated to the point where an electron is liberated, the
atom is ionized, resulting in an ion and a free electron.14 The state is
defined as a plasma when the volume of the ionized gas region exceeds the
distance between oppositely charged species, the Debye length.15 The
Debye length is expressed:
AD= 6.9(TC/Ne)i4
Eq. 1-1
where A0 is the Debye length, Tcis the electron temperature, and Ne is the
electron number density. Because the distance between the ions and
electrons is large, recombination of the ions and electrons is not likely and
the number density of electrons may be large.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Plasmas are a totally destructive technique in that the majority of the
sample is decomposed to its elemental components. Ionized sources of this
nature are known as “hard" ionization sources. A large portion of the
elements are ionized. Plasmas are good sources for both atomic emission
spectroscopy and mass spectrometry.16
There are several types of plasmas used in analytical chemistry.
They include: inductively coupled plasma (ICP), direct current plasma
(DCP), laser induced breakdown plasma (LIB), and microwave induced
plasma (MIP).
Inductively coupled plasmas are the most widely used plasmas in
analytical laboratories. Important characteristics for the ICP include a gas
temperature between 4500 and 8000 K and an electron temperature
between 8000 and 10,000 K.15 The relatively long residence time of the
analyte in the plasma is 2 to 3 ms. This characteristic yields almost
complete vaporization and atomization which, in turn, reduces the chemical
and physical interferences during analysis. The electron number density is
in the neighborhood of 1 to 3 x 1015 e'/cm3, making the ICP a rugged
electrical entity which is not significantly altered by sample introduction or
varying sample compositions. All of these characteristics allow for detection
of a large number of elements in varying solvent environments with a wide
range of concentrations.10,17
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6
Direct current plasma are used primarily for geological samples.18
Figures of merit for the DCP include excitation temperatures of 6000 K and
electron number densities in the neighborhood of 1 to 3 x 1015e*/cm3.10 The
configuration of the DCP, along with the figures of merit, allow for a variety
of sample types. Organic and inorganic samples as well as those with high
solid contents may be analyzed.
While the ICP and DCP have the listed attractive attributes, these
discharges have never been effectively applied to routine nonmetal
analysis.
The MIP produces intense nonmetal ion emissions. Microwave
induced plasmas have been coupled with gas chromatography for the
determination of nonmetals. The figures of merits for MIP are a gas
temperature of about 2000 to 3000 K, electron temperatures of about 4000
to 5000 K and electron number densities of about 5 x 1014e'/cm3.36
Since plasmas are reliable and rugged, they can be operated under a
variety of conditions. However, one mandatory condition is that the sample
must be in the gas phase or introduced as small droplets. Otherwise, the
plasma will be extinguished. The wet gas, or aerosol, comes from the
nebulization of a solution containing the analyte. “Dry” analytes may result
from introduction techniques such as electrothermal vaporization or gas
chromatography. The most pronounced difference is that fewer species are
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
introduced into the plasma with dry gas samples; and usually the only
species in the plasma is the analyte. The fewer species there are in the
plasma, the greater is the ionizing power of the plasma and the more likely
the analyte will exist in ionized or excited state in high populations. Another
major difference is that solvated analytes force some of the plasma energy
to go to stripping the analyte of its solvent before atomization and ionization.
The added detriment of a solvent is that it may reduce both the plasma
excitation and gas temperatures.
Argon is usually the gas of choice for plasmas. It is less expensive
than many gases, is available in high purity, and is easily obtained.19 With
an ionization energy of 15.755 eV, electrons within an argon discharge
posess sufficient energy to excite and ionize most metals, but are too low in
energy to ionize nonmetals significantly. Detection limits in Ar-ICP for
metals and transition metals are quite low.17,19 The reader is referred to
Table 1-1. The table clearly demonstrates that there is tremendous
disparity between the detection limits of metals and nonmetals. An
interesting thing to note is that halogens aren’t even represented on most
lists of detection limits. While Cl, Br, and I may be determined with ICPAES in the 120 to 180 nm spectral regions, these regions are not accessible
with standard instrumentation.
Helium is another choice for plasmas. It is about the same price as
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
8
Table 1-1 Ultraviolet and Visible Detection Limits of Selected Elements in
ICP-AES, ICP-MS and MIP-MS
Element
Ar-ICP-AES
Ar-ICP-MS
He-MIP-AES
He-MIP-MS
(ng/mL)15
(ng/mL)15
(ng/mL)20
(ng/mL)21
As
30
0.01
30
0.1
Br
*
*
★
0.2
Cd
2
0.005
40
0.9
Ca
0.1
0.5
10
2
Cl
*
*
★
39
Cr
5
0.005
1
0.3
Co
3
0.001
150
7
Cu
0.3
0.005
100
0.07
In
100
0.001
★
0.1
I
•k
*
★
0.04
Pb
7
0.001
1
2
K
80
0.5
*
17
Se
50
0.05
40
0.7
S
50
50
*
150
V
2
0.005
80
0.2
* indicates that the detection limit for this element is not listed in cited
source.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
9
Arand the purity equally high. However, helium is not easily obtained in all
countries. Detection limits in He-MIP for nonmetals are low. Metal
detection limits with He-MIP are not quite as low as with Ar-ICP. The
reader is again referred to Table 1-1. It can easily be seen that both
techniques have their virtues and their drawbacks. The MIP is good for
halogens, while the ICP is particularly good for transition metals.
Helium plasmas produce intense nonmetal atom and ion optical line
emissions. These plasmas have been used extensively as elementselective detectors for gas chromatography.22,23,24 Additionally, these
discharges have been applied as detectors for liquid25 and supercritical
fluid26,27chromatographies. Other optical emission applications have
included electrothermal vaporization28 and solution nebulization.29 The
ionizing power of He plasmas has also been utilized for mass spectrometry
ionization.30,31
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
10
1.3 Monitoring Species with AES and MS
With plasma atomic emission detection, light from the plasma is
focused on an entrance slit to a monochromator where a wavelength
corresponding to an analyte emission is selected and monitored. The
wavelength corresponds to a photon being emitted by an excited electron in
an ion or atom as it relaxes to a lower electronic energy level. This
approach has the virtue of being an electronic state specific technique. Not
only does the operator know which elements are present but they also know
which states are being excited by the wavelength of the photon. Since this
technique is state specific, it is extremely sensitive to population changes at
that single state.
Mass spectrometry is a technique that detects ions based on their
mass to charge ratio, ions are drawn out of the plasma into an evacuated
chamber where they are directed to a detector. MS monitors the total ion
population. However, since the ion population as a whole is monitored, the
signal is less susceptible to small population changes of a single ion state.
In terms of mechanistic studies, these techniques are
complementary. AES yields state specific information. MS provides
electronic state independent and total ion population information.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
11
1.4 Introduction to Charge Transfer
Nonmetal optical emission spectra from atmospheric pressure He
microwave-induced plasmas (MIP) are quite different from those of Ar
plasmas. For many nonmetals, the dominant emission lines are those of +1
ions in He plasmas while the most intense emission lines in Ar plasmas are
those of neutral atoms.32,33
Nonmetal ionization potentials are in the range 10-17 e V .34 The
energies required for these species to be ionized and promoted to the upper
state of the observed ion electronic transitions are typically 19-29 eV.
However, atmospheric pressure plasma temperatures are in the
neighborhood of 4600 K,35which correspond to an average energy of only
0.40 eV. It is clear that the states giving rise to nonmetal ion emission in
helium plasmas are overpopulated as judged by Boltzmann and Saha
distributions39 These phenomena suggest that He plasmas possess
different dominant nonmetal ionization pathways than argon discharges.
It is thought that thermal ionization is the primary nonmetai ionization
pathway in Ar discharges, while charge transfer (CT) dominates in the He
plasma.36,37,38 The general CT reaction may be written:
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
12
He+(24.56 eV) + A°(0 eV) - * He°(0 eV) + A+*(24.56 eV) ± AE
Eq. 1-2
where A represents the nonmetal to be ionized,0 and * indicate the
electronic charge, and * indicates an electronically excited state. The
energy defect (AE) is the difference between the 24.56 eV He ionization
energy and the nonmetal ionization energy plus the excitation energy.
Charge transfer is most efficient when nonmetal CT is a one electron
process, where a single electron is removed from the otherwise unperturbed
ground state atomic electron shell. One electron charge transfer is
maximized when the reaction is near resonant and slightly exothermic (AE
is small and slightly negative), as is in the case of chlorine.36
Many nonmetals possess a manifold of excited ion states with
energies near the ionization energy of He, as seen in Figure 1-1. Because
of these CT pathways, it is predicted that nonmetais (Cl, Br, I, etc.) will be
ionized to a much greater degree than predicted by standard Saha-type
calculations, it follows that these nonmetals will produce much more
intense ion emission signals in He discharges than in Ar discharges
because there are few if any resonant ionic states between Ar and most
nonmetals.
This phenomenon is demonstrated with chlorine as an example.38
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
13
60
50
40
30
24.56 eV
20
15.7 eV
10
0
C
N
0
F
P
S
Cl
Br
I
Figure 1-1. Energy Diagram of selected nonmetals. Each of the points
represents an excited state of the nonmetal ion with the lowest point for
each element being the first ionization energy. Each of the states is given in
electron volts above the ground atomic state of that element.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
14
Cl°(3s23p5) + H e^ls1) ->Cr*(3s13p5) + He°(1s2)
Eq. 1-3
Using the measured temperature of 4600 K and electron density of 5x1014
eVcm3 for sets of observable ions and atom emission lines, the ratio of
excited ion states to excited chlorine atom states was calculated to be in the
10‘n to 10'12 range. However, spectroscopic observations indicated that the
ion states were overpopulated by factors of 10n to 1012. These
considerations indicate that thermal ionization, alone, cannot account for
the high degree of chlorine ionization. Observations and calculations show
similar behavior for a number of other nonmetals, including P, S, Br and I.39
In summary, CT is predicted to produce a large overpopulation of
excited state nonmetal ions in helium discharges. While CT may only
slightly enhance the total overall population of nonmetal ions, CT directly to
excited ion states may significantly enhance populations of selected ion
excited states and their associated emission intensities.
Mass spectrometry monitors ions without regard to the electronic
state of the ion and may be used to provide complementary information.
Plasma mass spectrometry has a variety of virtues that have led it to be a
widely used technique in analytical laboratories. Plasmas as ionization
sources for time of flight mass spectrometry have the advantages of being a
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
pulsed techniques with near zero background and with an infinitely long or
short duty cycle.40 TOF allows for simultaneous sampling, which leads to
superior precision from interelement ratioing. TOF-MS has a broad mass
range within a single spectrum and has very good detection limits for
multielement analysis. There are drawbacks as well: the technique requires
complicated and expensive electronics, detection limits for individual
elements are not as good as in plasma quadrupole mass spectrometry, and
the technique requires dauntless fortitude of the operator.16
1.5 Direction of Dissertation
In these studies, mass spectrometry monitored total ion populations.
These results provide additional insight into nonmetal CT in He discharges.
This dissertation describes interfacing a helium microwave induced plasma
with a time of flight mass spectrometer (TOF-MS) for the analysis of plasma
kinetics. The second chapter deals primarily with the redesign and
construction of the TOF-MS. Initial characterizations are discussed in the
third chapter. Preliminary experiments involve the kinetics of the plasma in
connection with halogens and an interferent gas. The fifth chapter deals
with the simulation modeling of the kinetics of the plasma. The sixth and
final chapter will discuss the conclusions and future work of this project.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Chapter 2
CONSTRUCTION OF THE NIU HELIUM MICROWAVE INDUCED
PLASMA TIME OF FLIGHT MASS SPECTROMETER
2.1 Introduction to Time of Flight Mass Spectrometry
Within the governing mathematical principles, TOF-MS has a virtually
unlimited mass range. All masses can be determined, from protons, ionized
hydrogen of one mass unit, to proteins of several hundred thousand mass
units. In the TOF scheme, lighter masses arrive at the detector first
followed by the heavier masses. Since the spectrum starts at time zero and
the operator controls when data collection ends, all masses can be collected
within the same spectrum. Other mass analyzers cannot accommodate
such high masses and/or cannot analyze such a broad mass range in a
single sample. These attributes make TOF-MS a potentially useful
technique for a wide range of applications including quantification, elemental
determination, protein identification, and structure elucidation.4041
Time of flight mass spectrometry was introduced by Wiley and
McLaren in 195542 At the time, signal collection and electronics were too
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
17
slow for TOF-MS to become a viable technique, since flight times for
elements are on the order of tens of microseconds. TOF requires fast data
acquisition and large information storage capacity. With the
commercialization of digitizing oscilloscopes capable of collecting 5000 or
more data points in 200 microseconds and the advent of pulsed laser ion
introduction, TOF-MS found renewed life in the late 1980's.
A large majority of the analytes determined using mass spectrometry
are aqueous inorganic compounds. However, there is growing use for
atomic mass spectrometry for organic and biochemical applications. To use
mass spectrometry for elemental analysis, the sample must undergo several
steps. The sample is typically dissolved in a liquid. The analyte within the
solvent must then be vaporized, desolvated, atomized and ionized.
Inorganic salts in water pose few challenges in terms of solubility and there
are a large variety of nebulizers that will efficiently produce mists of aqueous
solutions. Heating the mist flow stream followed by condensation removes
much of the water from the nonvolatile analyte. However, due to significant
adverse effects on plasmas, organic solvent based samples may require
more desolvation than standard condensers can accommodate. Additional
desolvation may be required to ensure plasma compatibility. Once
desolvated, the sample can be introduced to the plasma for atomization and
ionization.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Mass analyzers measure current at a detector, dictating that the
detected particles must be charged. Most mass spectrometry detectors are
electron multipliers or microchannel plates, both of which collect charged
particles, and produce and amplify a current. In practice, most mass
analyzers are set to select a specific mass to reach the detector and collect
the current for a discrete amount of time to obtain sufficient signal. With
TOF-MS, all of the masses are simultaneously pulsed and directed to the
detector. There is no selective mass discrimination during this sampling
process. Ions within the mass range sequentially arrive at the detector,
each producing a short current burst. Ideally, the duration of the current
burst is equivalent to that of the pulse duration. Because the signal is
discontinuous in nature, a near zero baseline may be produced, allowing for
small sample sizes and low concentrations to be examined.
As discussed, the analyte must be ionized for analysis. There are
many ionization techniques suitable for mass spectrometry. Because of the
basic TOF characteristics, pulsed modes are better suited. Four of the most
widely used ionization approaches for time of flight mass spectrometry
utilized are: plasmas, lasers, electrospray ionization (ESI) and matrix
assisted laser desorption ionzation (MALOI). ESI and MALDl promote “soft”
ionization and are typically used for larger biomolecules. In the ideal case,
the ionized molecule is totally intact and still in its original native form.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
19
MALDl often produces singly ionized molecules, producing a direct measure
of the molecular mass and a very simple spectrum.44 Eiectrospray often
imparts multiple charges on the molecule, producing less direct mass
determinations and a more complicated spectrum.44 However, this
characteristic may work to the analyst's advantage. Multiply ionizing a
protein allows larger molecules to be brought to a mass to charge ratio that
can be determined with most mass spectrometers. Direct ionization by
lasers is a much more destructive ionization technique, but is a natural
match with TOF-MS since both techniques are pulsed. For elemental
analysis, time of flight has been interfaced with both ICP and MIP, both of
which are “hard" ionization techniques.43,44
2.2 Principles of Time of Flight Mass Spectrometry
In the TOF experiment, ions of charge z are directed to a region
where they are accelerated by a pulse of a known voltage (V). Regardless
of mass, each ion attains an equivalent kinetic energy (KE) in accordance
with Equation 2-1
KE = zeV
Eq. 2-1
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
20
where e is the charge of an electron and z is the number of electrons. TOFMS works on the physics principle that the lighter ions will travel at a
greater velocity to a detector than the heavier ions in accordance with
Equation 2-2.
KE=^mv2
Eq. 2-2
where m is the mass of the ion and v is the velocity. Velocity is equal to the
distance traveled, d, within a certain amount of time, t.
v=
Eq 2-3
y
Making this substitution for velocity in equation 2-2 yields:
1
(d\2
KE=zmIt)
Eq. 2-4
Setting equations 2-1 and 2-4 equal and rearranging, a particularly useful
relationship evolves.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Remembering that distance, the charge on the electron and the acceleration
voltage are constant, this equation shows that the time required to travel a
particular distance is proportional to the square root of the mass-to-charge
ratio, m/z, of the ion. In the case of singly charge ions, z is one, making the
mass and the mass-to-charge ratio numerically identical.
This relationship can be better examined with actual values. The
time it takes for the singly positively charged chlorine ion to reach a detector
can be calculated using the following representative values. If the length of
the flight tube is one meter, the pulse is 120 volts and the mass of the
chlorine isotope is 35 grams per mole, one may substitute into Equation 2-5
to obtain:
35x10“ % 0|(1m)“
t=
Eq. 2-6
^
2(120V)[96500 9mol) q v
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Under these conditions 35CI+ should have a flight time of 38.9 us.
Performing this same calculation for 37Cr yields a flight time of 40.0 us, or a
separation of 1.1 us from 35CI+.
The greatest time separation between masses will occur with an
infinitely long flight tube. Of course, there is a practical limit on lab space.
Additionally, because each set of ions at a single mass may not attain
identical kinetic energies, the ion "packet" may broaden while traversing
lengthy flight times. The invention of the reflectron allows roughly doubling
the flight length without increasing the physical space the instrument
occupies. The reflectron, or ion mirror, uses a positive electric field to reflect
the ions back down the length of the flight tube to a detector in the vicinity of
the pulse plate. The positive voltage of the reflectron is about the same
voltage as the acceleration voltage of the pulse plate. The reflectron has an
added benefit in that the kinetic energy distribution in a packet of ions of the
same mass is greatly reduced. This compensation with the reflectron ion
mirror occurs because ions with greater velocities (greater kinetic energy)
will penetrate the reflecting field to a greater depth than those ions with
lesser velocity, giving a longer distance traveled. When the mass packet
reaches the detector, it will arrive in a much more compact form. This effect
results in a narrowing of peak widths and enhanced resolution.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
23
2.3 Physical Considerations for Time of Flight Mass Spectrometry
Mass spectrometers detect ions, which may be short lived and are
very reactive. One way to promote longevity is to reduce the number of
species (including electrons) the ion encounters. The best way to achieve
this is in a reduced pressure environment. The lower the pressure, the
longer will be the distance between collisions. The distance the ions travels
before encountering or colliding with another species is the mean free path.
In plasma-mass spectrometry the ions are typically formed in an
atmospheric pressure plasma and drawn into an evacuated chamber. To
achieve appropriate pressures to ensure a nearly collision free flight path,
two physical items that must be considered are the pumping capacity of the
vacuum system and the diameter of the sampling cone orifice.
Typically in TOF-MS experiments, it is necessary to reduce the
pressure from atmospheric pressure (760 torr) to about 10'6 torr.
Performing such a large pressure reduction in a single step would be both
difficult and costly. Performing the reduction in pressure in two or more
stages is much easier and, in most cases, less costly. The following
equation describes a system at equilibrium with one gas flow in and one
gas flow out of the chamber to be pumped.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
24
^inRn
^oul^out
Eq 2-7
where c,„ is the conductance of the orifice leading from atmospheric
pressure into the chamber to be evacuated, cout is the capacity of the pump
inside the chamber, Pinis atmospheric pressure, and Pout is the pressure
inside the chamber. For viscous flow, the conductance of an orifice is 20
L/sec times the area, nd2/4 (d in cm) where d is the orifice diameter.45
Substituting this relationship in Equation 2-7 and rearranging, the
relationship becomes:
d=
Eq 2-8
This relationship can be better examined with actual values. Stepping
from atmospheric pressure (760 torr) to a pressure of 10'5 torr with a
pumping capacity of 2400 L/sec, cout,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
d =
4*2400 — 10'5torr
sec
tc20
Eq 2-9
- — ^760 torr
sec cm
one would need an orifice no larger than 14 pm. This diameter is too small
for practical sample introduction. Effective sample introduction requires an
orifice diameter of 0.5 to 1.5 mm. Otherwise, an insufficient number of
analyte molecules will be sampled for detection. To allow for a practical
sample orifice, either the initial pump needs to have a much larger pumping
capacity or one can reduce the initial pressure equirements. With a twostage system, it is possible that the pressure drop to the first stage be only
to 10'3torr. Standard vacuum pumps are sufficient for this purpose. From
this reduced pressure, a further pressure reduction to 10'5torr can also be
easily achieved with standard vacuum pumps. The diameter required now
can be as large as 0.1 mm.
d=
4 * 2400 — 10‘3torr
sec
| 7t20
sec cm
Eq 2-10
760torr
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
26
2.4 Time of Flight Instrumental
2 .4 .1
P r e v io u s I n s t r u m e n t a t io n
The NIU time of flight instrument was originally built by Sin and
Schreiner as an interface for MALDl.46 After some characterization and
experiments the instrument was modified to be an atmospheric pressure
interface (API) instrument.46 Details regarding the instrument are provided
in the doctoral dissertation of Alex Schreiner.46 The operating pressure of
the instrument was 10'3torr, which gave a usable signal. The mean free
path with these conditions was 5.91 meters. Since the flight tube is 1 meter
in linear mode and 2 meters in reflectron mode, from 1/6 to 1/3 of the ions
will suffer collisions, which will broaden the ion packet or deflect ions off
course. The design of the pumping system is shown in Figure 2-1. The
pumping system had two stages to go from atmospheric pressure to the
reduced pressure in the ideal range for a mass spectrometer, 10‘5 torr or
lower. The first stage of the differential pumping was the API apparatus and
was maintained by four small mechanical pumps, not shown in the picture,
with an effective pumping capacity of about 10 cfm.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
27
TOFMS
Water
Baffle
Figure 2-1. Block Diagram of NIU-TOF-MS as of June 1996
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
28
2 . 4 . 2 C u r r e n t I n s t r u m e n ta t io n
The TOF-MS used in the experiments described in this dissertation
follows. The optimized operating conditions and manufacturers for the
various components are listed in Table 2-1. A schematic of the instrument
is shown in Figure 2-2.
2 .4 .2 .1
M a s s S p e c tro m e try V a c u u m
S y s te m
An R.M. Jordan Co. (Grass Valley, CA) flight tube was pumped by a
Varian (Wood Dale, IL) VHS-6 diffusion pump at the base of the flight tube.
A Varian VHS-4 diffusion pump was located at the top of the flight tube.
The diffusion pumps were backed by Welch (Chicago, IL) model 1397 and
1402 mechanical pumps respectively. In the sample inlet region, a small
chamber constructed for differential pumping was maintained by a Stokes
(Philadelphia, PA) model 212H-11 pump. The sample cone has an orifice of
0.5 mm and is 5 cm in diameter. The cone rises 7.5 mm from the edge at
an 18.5° angle. Less than 3 mm from the inside of the sample cone, the
orifice of a nickel skimmer cone leads to the flight tube. This skimmer cone
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
29
Table 2-1. Operating conditions of the NIU He-MIP-TOF-MS
2450 MHz. 300-525 W
Plasma generator
Micronow 500W 420B
Plasma Cavity
TM010Beenakker style
Plasma Gas
Helium
13 L/min
Carrier Gas
Helium
1 L/min
USN Heater
140° C
USN Condenser
4° C
Sample cone
Copper
0.5mm
Skimmer cone
Nickel
2mm
1“ element cone
in house
-400 V
2ndelement lens
in house
-5 V
3rt element lens
in house
-12 V
4thelement lens
in house
-60 V
Pulse Conditions
R. M. Jordan Co.
-120 V for 1.75 ps every 920 ps
1st and 3rdelement outer einzel
in house
-3500 V
2ndelement inner einzel
in house
-5 V
Liner
in house
-850
Steering plate
in house
-815 V
MicroChannel plates
Gailileo
-1000 V
Oscilloscope
Tektronics 620B
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
30
Electronics Rack
Flight tube
Sample Inlet
i
j
i
Ii
1200 L/sec Diffusion Pump
i
1 3 o j cfm Mechanics Pump-
5 cfm Mechanical Rump
2400 UseC Diffusion Pump
i
W
17 cfm Mechanical Pump
Figure 2-2. Block Diagram of current NIU-TOF-MS
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
has an orifice of 2 mm and is 2 cm in diameter. The cone rises 10 mm from
the edge at a 65° angle.
When the sample orifice is closed, the pressure in the flight tube is
10-7to 10* torr as read by a Varian 880 Bayard-Alpert ionization gauge.
When sampling from the plasma, the main chamber exhibits a pressure of 2
x 10'5torr, while the pressure directly inside the sample cone is 10* torr.
The actual pressure is probably somewhat higher due to the low He
sensitivity of the Bayard-Alpert gauges.
2 .4 2 .2 .
T ra n s p o rt o f Io n s
The plasma is abutted with the sample cone. Analyte passes
through the copper sample cone and undergoes expansion in the small
chamber, where the analyte passes through the skimmer cone to the main
chamber. Both cones are held at ground potential. In the region beyond
the skimmer cone, the ions undergo less frequent collisions with buffer gas.
The series of focusing lenses depicted in Figure 2-3 aid to direct the
ions from the skimmer cone to the pulse plate. The first lens is a cone with
a significant negative potential to draw the positive ions into the flight tube.
After the cone are two additional focusing lenses, maintained at significantly
less negative voltages. An elongated barrel maintained at a somewhat
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
32
I N °l
AB
M
K.
0
*
2
3
4
5
6
7
8
9
10
Cm
Figure 2*3. Diagram of lenses. A, sample cone. B, Skimmer cone. C,
First element of sample introduction einzel lenses, cone. D, E, second and
third elements of sample introduction einzel lenses. F, final element of
sample introduction einzel lenses, barrel. G, pulse plate. H, extraction grid.
I,J,K, elements of flight tube einzel lenses. N, M, deflectors. 0 , liner.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
larger negative voltage than the previous lenses shapes the ion packet
before entering the pulse plate region. These four potentials are maintained
with an in-house constructed power supply. This high-low-high pattern is
consistent with the three elements of a focusing einzel lens.47 Typical
values are given in Table 2-1.
The pulse plate imparts a +120 V square wave pulse with 1.75 ps
duration every 920 ps, using an R. M. Jordan Company pulser power
supply. The residence time in the 10 mm long pulse region is 0.6ps for the
largest mass species studied in the kinetic experiments(bromine). One cm
above the pulse plate is an extraction grid held at a ground potential. One
cm above the extraction grid is a second set of einzel lenses. The outer
element is maintained at a high voltage and the inner element is held at a
low negative potential. The liner begins 3 cm above the last einzel and is
maintained at a high voltage for the one meter flight tube length. A set of
deflectors is mounted 13 cm from the beginning of the flight tube. The
deflector parallel to the initial path of introduction of the ions is maintained at
a voltage similar to that of the liner while the deflector in the perpendicular to
the initial path of the ions is maintained at a potential 50% more negative.
This means that the deflector is canceling the forward momentum of the
ions imparted by the intiaf set of lenses. The ions reach the linear detector,
a pair of Galileo (part number MCP-18B, Sturbridge, MA) 18 mm
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
34
microchannel plates in a chevron configuration at the manufacturers
suggested voltage. The signal was collected with a Tektronics (Wilsonville,
OR) 620B digital storage oscilloscope.
2.5 Details of the Improvements
This section details the nature and effects of the improvements made
to the TOF-MS system.
2 .5 .1
In s t r u m e n t a t io n
The API port small mechanical pumps with a combined pumping
capacity of 10 cfm were replaced by a single larger rough pump with a
capacity of 130 cfm. The API port was replaced with a sample coneskimmer cone assembly. Instead of the newly installed pump directly
evacuating the main chamber, a smaller chamber containing the sample
cone-skimmer cone assembly was built for the first stage of differential
pumping. With the reduced pressure in the smaller new chamber, it was
easier to obtain the pressure desired in the main chamber. The mean free
path in the main chamber is 7.13 meters.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
35
The large mechanical pump is very good for step down pumping;
however, it is given to significant vibrations. To reduce the vibrations in the
instrument, a flexible hose in an 80° turn was used to connect to the pump.
This greatly reduced the vibrations in the instrument directly from the pump.
To eliminate the vibrations the pump delivered to the floor, a set of four
rubber shock absorbers were bolted onto a plate the size of the base of the
pump and bolted onto the base of the pump. This approach greatly reduced
the vibrations delivered to the floor.
The cryotrap was removed for more efficient pumping of the flight
tube by the large diffusion pump. However, there is a “direct line of sight
path” for the large diffusion pump to the main chamber, meaning that the oil
from the diffusion pump could potentially spray directly into the main
chamber. This process is called backstreaming. The water cooled cryotrap
was reinstalled at a later date to reduce the backstreaming. Although there
was very little change in the operating pressure in the main chamber,
adding the cryotrap increased the time needed to initially “pump down,” or
“rough out" the main chamber. The large gate valve VRC-8, above the
large diffusion pump was replaced with a much more reliable gate valve to
greatly reduce leaking through the shaft seal. The valving system between
the small diffusion pump was streamlined to allow the mechanical pump to
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
36
either back the small diffusion pump or to help rough out the chamber. This
greatly reduced the time necessary to rough out the flight tube.
The diffusion pumps come equipped with thermal snap switches that
turn the pumps off when the pump temperature gets too hot. This happens
when the cooling water is no longer cool enough or a cessation in the water
flow occurs. Also designed, built and installed was an interlock system
connected to the ion gauge. In the event that the pressure measured by the
gauge between the diffusion pump and the mechanical pump got too high
(usually due to an increased load on the diffusion pump caused by a small
leak to the atmosphere), the diffusion pump automatically shut off until the
pressure was returned to the optimum operating pressures. This
configuration helps to avoid back streaming when the diffusion pump
backing pressure got too high. The interlock system cuts off power to the
diffusion pump when the backing pressure reaches a value greater than 100
millitorr. Power is returned to the diffusion pump when the backing pressure
is again reduced to less than 100 millitorr. The reduction in backstreaming
helped avoid oil contamination of the focusing lenses.
A vacuum baffle between the small diffusion pump and the small gate
valve was removed because it was unnecessarily cutting down the pumping
capacity of the small diffusion pump. It was placed there originally to reduce
backstreaming into the flight tube. Since the diffusion pump is at a right
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
37
angle to the flight tube and has no direct line of sight, the backstreaming
was found to be insignificant. Furthermore the original baffles were painted
with latex paint, and any potential benefit of the baffles was outweighed by
the potential of paint outgassing.
The oil in the diffusion pumps was changed from DC-704 to DC-705.
The vapor pressure of the DC-705 is much lower than that of DC-704.
Backstreaming was reduced by the change to the better oil.
There was significant transference of oil between the diffusion pumps
and the mechanical pumps backing them. To reduce the transfer of oil and
increase the lifetime of both the mechanical and diffusion pump oils, an oil
trap was placed between both sets of pumps. The trap consists of copper
turnings held in place by wire mesh. The oil mist from both pumps is
condensed onto the turnings, greatly reducing cross contamination of oils.
The trap was cleaned every time the mechanical oil was changed by rinsing
with hexanes until the rinsing solution runs clear.
The orifice that leads from the atmosphere to the small chamber is
the sample cone. Typically, the sample cone is constructed from nickel
because it has good heat conductance, and can dissipate the heat of the
plasma quicker than it will be destroyed by the heat. However, once the
instrument was interfaced with the plasma, the cone was sampling from the
hottest region of the plasma and was being degraded and the orifice
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
38
enlarged. Several cone designs and materials were tested. Even though
Cu has a lower melting point than Ni, there is better transfer of heat
between the copper cone and the cooling water. As a result, cones made
of copper did not degrade nearly as quickly as a nickel cone of the same
design.
The signal intensity after initial signal optimization was not sufficient
for the intended studies. The orifice of the sample cone could be enlarged
until the signal was sufficiently large and the vacuum was still good enough
for useful work. The initial sample cone orifice was a little bit less than 0.2
mm in diameter. This size was prone to clogging and did not provide
sufficient signal for the proposed kinetics work. The cone was enlarged by
a standard drill bit increment until the pumps were overtaxed. When the
pumps became overtaxed, the previous size orifice was the optimum
diameter. The optimized diameter was determined to be 0.5 mm. With the
diameter 2.5 times larger, the orifice area is 6.25 times larger and the
amount of analyte that can pass through was greatly enhanced.
2.5.2 ton T r a n s p o r t
Ions from the plasma are introduced into the flight tube and directed
to the detector. There are numerous variables to consider. In the TOF,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ions are directed to the detector by a high voltage pulse. When the ions,
separated by mass, reach the detector, it is desirable for all ions of the
same mass to arrive at the same time. However, in practice, there is a
discrete window in time when the ion packet arrives at the detector. Proper
ion focusing helps to minimize the window. The following steps were taken
to reduce the breadth of the ion packet.
When the ions were introduced into the flight tube in the original
instrument, there were 4.5 inches of “dead space" for them to travel before
entering the pulse plate region. This is a long distance to travel without
focusing. To aid in the transport and direction of the ions, an introductory
set of einzel lenses was introduced. Four lenses (a cone, 2 windows and a
barrel) were placed in the 4.5 inch space between the skimmer cone and
the pulse plate. The cone is kept at a fairly high negative potential to draw
the ions into the chamber and to repel the negative ions. Removal of the
negative ions hinders ion recombination and signal reduction. The two
windows are held at low negative potentials, while the barrel is maintained
at a negative potential much greater than the slits but smaller than the
cone. The purpose of the barrel is to shape the packet of ions to enter the
pulse plate region to reduce the spread in the height of the ion packet. The
high-low-high voltage pattern is consistent with a three element focusing
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
40
einzel, which will focus the ions from the expansion into the pulse plate
region.
An analytical signal is most desirable when it is not only very narrow
but very intense as well. A peak that is broad and has low intensity may be
difficult to reliably discern from the background. Several things were done
to obtain a compact ion packet. The first was to end the first lens train with
a cylindrical barrel, which determined the diameter of the packet entering
the pulse plate region. If the diameter is small, the ion furthest away from
the pulse plate receives only slightly less kinetic energy than the ion closest
to the pulse plate.
It is also important that the pulse be long enough to get the heaviest
analyte out of the pulse plate region before the pulse is over so that all ions
are imparted with the same kinetic energy. The pulse plate region is about
1 cm in height and the heaviest ion for the mass calibration is iodine. Using
the equation from the introduction, it should take iodine 740 ns to leave the
pulse plate region. The duration of the pulse was set to 750 ns. The time
in between pulses is important, but not critical. If the time between pulses
is minimized, less time is required to obtain a fixed number of spectra. Due
to the design of the pulser electronics, the fastest pulse cycle which could
be used was 155 ms.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
41
To ensure that the ions do not pass out of the pulse plate region, a
negative bias was applied to the pulse plate while the pulse was off. This
had a dual effect of drawing the ions to the plate to decrease the energy
spread and to slow the ions down to make more of them stay in the pulse
plate region before pulsing to the detector. The bias was optimized and
maintained at about -12 volts. This value is considerably lower than
typically seen in literature.16 There are several possibilities for the different
optimized values. In this case, the pulse is supplied to direct the ions at a
right angle to the sample introduction, making the background near zero so
that there does not need to be a substantial bias to keep the ions from
drifting towards the detector unpulsed. Additionally, the plasma gas in this
case is helium rather than argon. Because helium recombines far more
readily than argon, there is not a steady stream of ionized plasma gas
heading to the detector. Also, the lighter mass of the helium ions will allow
them to be more efficiently removed by a given negative voltage. Finally,
because elemental ions are of the lightest mass, a substantial bias is simply
not required.
The pulsed ions are drawn in the direction of the detector by a mesh
that is held at ground potential. The opening of the extraction grid was
enlarged to allow a larger number of ions to pass, thereby increasing the
intensity of the signal peaks.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
42
Directly after the extraction grid is a set of focusing einzel lenses, in
the high, low, high voltage pattern. The shape of these lenses was such
that the ions experienced the field from the middle einzel for a longer
distance than the other lenses. Thus, focusing was very sensitive to the
setting of this lens. A change of ±10 volts was the difference between an
optimized signal and no signal. The shape of the middle element was
changed from that of a cylinder to an open circle and the distance between
the lenses was shortened. This design enhanced the acceptance
voltage(±50 volts) with an almost linear decay in the signal with changing
voltage.
There were a pair of deflectors immediately after the third element of
the einzel lens. Since the ions were not up to the potential of the liner,
focusing them at this point was ineffective. The deflectors were moved up
into the liner about 23 cm where the sensitivity to the liner voltage was
enhanced. The power supply for the deflectors was changed. Initially,
they were powered with supplies that only allowed incremental 100 volt
changes. The new power supplies have finer control and allow greater
system flexibility. The size of the deflectors was increased from 1 square
inch to 2 square inches; because the ions experience the deflector voltages
after coming to liner velocities and see the deflectors for a longer period of
time, this configuration is much more effective.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
43
From the perspective of the ion, the liner supplies the highest voltage
for the longest period of time. When the flight tube was disassembled, the
connection between the liner and the power supply was tenuous. This
connection was rebuilt making it more reliable.
The ions can either be detected in linear or reflectron mode.
Previously, the instrument had only been operated in the linear mode. The
reflectron was cleaned, including the mesh in the first and last plate. Some
resistors were reconnected and reinstalled. Once the reflectron was
reinstalled, it was connected to the liner potential. Previously, the reflectron
potential was unknown. This modification served to greatly enhance the
signal. When the reflectron was activated, there was signal immediately,
and no further adjustments were needed at that time.
The microchannel plate detectors in both the linear and the reflectron
location were replaced. The signal was greatly enhanced in the linear
detector. Since there was no original reflectron signal, there was nothing to
compare so no enhancement could be established.
The reflectron detector is 40 mm in diameter and the linear detector
is 18 mm in diameter, so there is more surface area in the reflectron
detector for the ion signal. This has two benefits: the packet does not need
to be focused as rigorously, and there is greater likelihood that the entire
ion packet will reach the detector.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
44
2.5.3 M
ic r o w a v e in d u c e d P la s m a
The typical microwave induced plasma cavity used in this research is
the Beenakker cavity.48 The typical cavity is designed with tuning stubs on
one side and a cooling water chamber on the other side. This was not good
for the best sampling of the plasma for the mass spectrometer. The sample
cone could not sample the hottest portion of the plasma. With other cavity
configurations, the cavity could just be flipped over. In this case, the tuning
stubs prevent the plasma from coming into contact with the sampling cone.
The cavity was rebuilt with the microwave focusing portion of the cavity in
contact with the sample cone and the water cooling portion in the back with
the tuning stubs on the water cooled portion. This redesign allowed the
sample cone to come in contact with the hottest portion of the plasma
where the greatest amount of charge transfer will take place.
The electron seed to initiate the plasma was supplied by placing a
copper wire attached to a quartz tube in the quartz tube in the center of the
cavity while applying about 200-300 W power from the generator. The
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
45
optimum flow rate for the plasma support gas is variable; however, the rate
that gave the best signal was 13 L/min.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Chapter 3
ANALYTICAL CHARACTERIZATION OF THE TOF INSTRUMENT
3.1 Introduction
In the previous chapter, details of the construction of the He-MIPTOF-MS instrument that resulted in reliable performance and excellent
operating stability were outlined. In this chapter, initial characterization of
the instrument is described. This characterization included determining the
mass calibration, optimizing the ion focusing optics, and enhancing the
mass spectral resolution. This chapter discusses the evolution of the
viability of the instrument from calculations to experimental data acquisition.
3.2 Reagents
The reagents used in these studies are as follows: 99.995% grade
He was used as the plasma support and analyte transport gas. Solutions
for the initial identification of peaks ranged from 1000 to 40,000 ppm
aqueous and methanolic solutions of RbBr, NaCI, CsCI and KCI. In addition
to the inorganic salts, molecular bromine and iodine solutions of 4% (v/v) in
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
47
reagent grade methanol were used. Spectroscopic grade deuterated
methanol (CD3OD) (MSD Isotopes, Montreal, Canada) was also introduced
to the plasma for mass calibration purposes. Molecular chlorine
introduction was accomplished using 99.995% pure chlorine gas from a
lecture bottle (BOC Gas, St. Charles, IL).
3.3 Sample Introduction
Two types of samples were introduced. Direct gas sampling and
nebulization of liquid samples were utilized.
3.3. 1
U ltr a s o n ic N e b u liz a tio n
S a m p le I n t r o d u c t io n
Aqueous solutions of inorganic salts and methanolic solutions of
RbBr, NaCI, CsCI and KCI were nebulized using a Cetac 5000AT ultrasonic
nebulizer (USN). The USN heated U tube was maintained at 140° C. The
condenser was maintained at 4° C for aqueous samples and at -4° C for
methanolic samples. The transducer frequency was held at 1.36 MHz. The
helium carrier gas flow rate was 1 L/min. Both the aqueous and methanolic
samples were delivered to the USN at a rate of 1 mL/min using a Rainin
peristaltic pump.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
48
3.3.2
G a s e o u s S a m p le I n t r o d u c t io n
Chlorine gas was introduced through the ceramic sampling tube in
the center of the torch. The flow rate of the chlorine gas was less than 1
L/min. Additional ventilation was installed to remove excess chlorine gas
from the plasma area.
3.4 Flow Rate Optimization
Initially, the 20 L/min flow rate utilized by Fu et al.49 was used for the
helium plasma support gas. This flow rate provided a robust and reliable
plasma. A solution of 1000 ppm aqueous RbBr was nebulized into the
plasma with a peristaltic pump uptake rate of 1 mL/min and carrier gas flow
of 1 L/min. The plasma power was 325 W. Without any usable signal, it
was suggested that the optimum atomic emission spectroscopy flow rate
was not appropriate for mass spectrometry.
Plasma gas flow rates of less than 15 L/min produced an appreciable
ion signal. The newly visible peak was tentatively identified as rubidium.
While rubidium has two isotopes, 85 and 87, they were not well resolved.
With only a single mass, a mass calibration could not be obtained.
Optimization of the plasma gas flow rate was based exclusively on the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
49
intensity of the peak. The optimal flow rate for the plasma gas was
determined to be 13 L/min. The optimal flow for the nebulizer carrier gas
was determined to be 1 L/min. Optimization was based on the overall
intensity of the signal, the shape of the peak and the deposition of sample
on both the plasma torch and the sample cone.
3.5 Actual vs Calculated Flight Times
Several mass spectra were obtained for the purposes of
characterization. Identification of individual mass spectral peaks can be
aided with flight time calculations. However, there is usually a discrepancy
between calculated values and experimental flight times. With the voltages
and distances known (given in Table 2-1), the flight times for specific ions
were calculated using the equations from the second chapter.
To determine the flight time of chlorine, 1% methylene chloride was
nebulized into the plasma. The solution was delivered to the nebulizer at
flow rate of 1 mL/min. All other conditions were similar to Table 2-1.
The calculated flight time for 35CP was 38.9 ps. Experimentally, 35CP
reached the detector in 17.25 ps. The discrepancy is due to the equation
only taking into account the initial kinetic energy supplied by the pulse plate
without compensation for energy imparted by ion focusing optics. Additional
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
50
kinetic energy is supplied by einzel lenses. Also, a large increase in the ion
kinetic energy is imparted to the ions upon entering the field free region of
the liner. These energies are significant with respect to the initial energy
supplied by the pulse. The voltages of each element of the TOF are listed
in Table 3-1 with subsequent durations in those regions. The summations
of these durations is the total calculated flight time for 35CI+. Taking these
kinetic energy additions into account yields a flight time of 17.18 ps. This
value is in close agreement with the experimental values.
3.6 Preliminary Mass Calibration
The identification of the mass spectral peaks is a necessary task in
the development of the instrument. This section describes the initial
identification and calibration for spectra with multiple ions.
3.6. 1
M a s s C a lib r a tio n M e th a n o l a n d D e u t e r a t e d M e t h a n o l
Methanol (CH3OH) and deuterated methanol (CD3OD), were
nebulized into the plasma to aid flight time characterization. These
solutions were chosen because the hydrogenated methanol sample
contains no deuterium and the deuterated methanol sample contains no
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
51
Table 3-1. List of ion optics, voltages and distances for the purposes of
calculating a total flight time for chlorine, 35CI+.
Ion optic
length (m)
voltage
time (microseconds)
pulse plate
0.025
125
0.952
extraction grid
0.075
125*
2.86
outer einzel
0.075
4125*
0.509
inner einzel
0.05
127*
1.89
liner
1.25
1925*
12.12
Total time
17.02
* The extraction grid is held at ground potential, however, the ions
have an initial kinetic energy of 125 eV. All of the optics that follow have
the applied potential plus the pulse voltage.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
52
hydrogen. The signal was collected and averaged with a LeCroy
9400A(Chestnut Ridge, NY) digitizing oscilloscope. The electronics were
fixed to settings similar to those given in Table 2-1. Subtle daily
optimization adjustments were never more than 5% from the values given in
the table. The change in the voltages being never more that 5% suggests
that the instrument was far more rugged and reliable than in the past. The
optimal flow rates discussed in Section 3-3, were used for the plasma and
carrier gases. The plasma power was 350 W during these trials.
The signal from both solutions was stable and reproducible. Sample
spectra are given in in Figure 3-1, the spectra are offset for clarification.
The lower spectrum exhibits five easily identifiable peaks. These are ’FT,
12C \ 160 \ ’7[O H f and 18[OH2] \ Four peaks are readily identifiable in the
CD3OD spectrum. These are: 2D", 16CT, 18[OD]* and 20[OD2] \
In both spectra, the peaks are well defined and resolved. The mass
spectral resolution is sufficient for this mass range. With peaks
corresponding to seven different masses at seven different times, a mass
calibration was obtained. The plot of the square root of mass versus time is
shown in Figure 3-2. Corresponding data are listed in Table 3-2. The
points on the plot correspond to the seven unique masses from the spectra
and the line is the linear least squares regression fit. The correlation
coefficient for the plot is 0.99990 and y-intercept (zero mass) is 0.140 |Js.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
53
cr
I6 r \ +
D
OD
OD,"
r
; > i r
«c
'
O
1
10
15
K
V\
OH,
20
25
Mass
Figure 3-1. Spectra of methanol (lower trace) and deuterated methanol
(upper trace) with peak assignments.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Time
54
2
0
0
1
Square root of mass
Figure 3*2. Calibration plot of methanol, CH3OH and deuterated methanol,
CD3OD.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
55
Table 3-2. Compilation of the species, mass and times from the methanol,
deuterated methanol spectra.
Species
Mass (amu)
(Mass)1
Time (ps)
’H
1
1
2.3
2D
2
1.41
4.2
12C
12
3.46
9.0
160
16
4
11.0
17[OH]
17
4.12
11.7
18[OH2]/[OD]
18
4.24
12.2
20o d 2
20
4.47
12.6
Regression zero
0
0
0.140
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
56
This correlation confirms that the mass assignments are correct. For future
reference, it should be noted that the calculated mass spectral resolution is
approximately 10.
3.6.2
M a s s C a lib r a tio n IA a n d V IIA E le m e n ts
Initial mass calibrations were encouraging, but the ion of highest
mass detected in this set of experiments was the 20 amu ion, OD2\ The
majority of species of interest in this study are of a higher mass. To
calibrate the instrument with masses at and above the masses of interest,
IA elements were introduced to the plasma.
Using settings similar to those of the methanol experiment, 1000
ppm potassium, rubidium and cesium were introduced into the plasma as
methanolic solutions. Flight times for the ions are listed in Table 3-3. The
flight times for the four species, 39IC, 85Rb\ 87Rb* and ’^Cs* are plotted on
Figure 3-3. The points correspond to the experimental data and the line is
the linear least squares fit. The data falls on the regression line
consistently. The correlation coefficient for the plot is 0.99995 and a time
for mass of zero is
-0.0886 ps, both of which indicate that the
assignment of masses to the peaks is correct.
Methanolic solutions containing methylene chloride, molecular
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
57
Table 3-3. Compilation of the species, mass and times from the spectra of
solutions in methanol containing the IA and VI IA salts.
Species
Mass (amu)
(Mass)'
Time (ps)
39K
39
6.24
17.41
85Rb
85
9.22
25.73
87Rb
87
9.33
26.12
132Cs
133
11.53
32.35
35CI
35
5.92
16.485
37CI
37
6.08
16.935
79Br
79
8.89
25.035
81Br
81
9
25.435
127l
127
11.27
31.295
IA regression
0
0
-0.0886
VIIA regression
0
0
-0.104
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
58
Time (microseconds)
30
!
I
20 1
I
I
10
I
0
0
2
4
6
8
10
12
Square root of mass
Figure 3-3. Calibration of the IA in methanol, R2=0.99995
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
59
bromine and molecular iodine were nebulized into the plasma using the
same operating conditions. Peaks were observed and flight times are also
listed in Table 3-3. The mass calibration is seen in Figure 3-4. The flight
times and mass calibration were consistent with the work done to date. A
correlation coefficient of 0.9990 and a y-intercept corresponding to -0.104
ps were obtained.
3.7 Resolution and Peak Shape Optimization
The importance of peak shape and resolution optimization cannot be
underestimated. While the aforementioned ion optics voltage settings gave
a reliable mass calibration, the peaks were not of the desired shape. The
element of the ion optics that has the greatest effect on the peak shape is
the pulse plate. There are two aspects of the pulse to optimize: the voltage
and the duration.
3 .7 .1
P u ls e D u r a tio n
The 1% molecular bromine methanolic solution was introduced into
the plasma with the same operating conditions as those used in the
previous halogen experiment. Using a pulse voltage of 400 V, the pulse
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
60
35
Time (microseconds)
30
25
20
15
10
5
0
0
2
4
6
8
10
12
Square root of mass
Figure 3-4. Calibration of the halogens in methanol, R2=0.9990
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
61
duration was varied from 600 to 850 ns. The resultant spectra are shown in
Figure 3-5. The time scale is such that only the bromine peaks are shown
and the spectra are offset for clarity.
As can be seen in the spectra, the peaks became both sharper and
better resolved using the 750 ns pulse duration. The peaks became both
less intense and less resolved at longer and shorter pulse durations. These
results suggest that the pulse duration that is best for these settings is 750
ns.
While 750 ns is the best for bromine, the duration should be long
enough to allow the heaviest ion to leave the region before the pulse ends.
For the purposes of this study, iodine is the heaviest ion. Based on the
calculations described in Chapter 2, Section 2-2, it will take 740 ns for 127r
to leave the pulse plate region. Since a pulse duration of 750 ns was
adequate for the mass range of interest, and provided optimum intensity
and resolution, it was adopted as the permanent pulse duration.
3.7.2
P u ls e
V o lta g e
Chlorine gas was directed into the plasma at a flow rate of less than
1 L/min using the operating conditions described earlier in this chapter.
While the previous settings for the ion optics voltage settings enhanced the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Time (microseconds)
Figure 3*5. Molecular bromine peaks during optimization of pulse duration
from 600 to 850 ns.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
63
resolution the peaks were not of the desired shape. The voltage of the
pulse plate was varied from 30 to 130 V. The resultant spectra are shown
in Figure 3-6. In the lowest trace the pulse voltage is 30 V. In the traces
above, the pulse voltages are 80 V and 130 V. The time scale is such that
only the chlorine peaks are shown and the spectra are offset for clarity.
As can be seen the chlorine peaks become more intense, much
sharper and better resolved with each increase in voltage. Further, the 130
V trace shows the shoulder on the peaks has also become resolved. Once
the optimum voltage was determined, the shoulders on the chlorine peaks
were resolved and determined to be 35CII-r and 37CiFT.
3.7.3
C o m p le t e d O p t im iz a t io n
The modifications to the operating conditions allowed the signal to be
well optimized for peak intensity, peak shape and resolution. 1% molecular
bromine and iodine solutions and gaseous chlorine was introduced into the
plasma with the conditions given in Table 2-2 and the unoptimized
conditions for the pulse plate. The resultant spectra can be seen in Figure
3-7: the halogens, X", are present, but the peak shape and resolution are
not of the desired quality. Using the optimized settings for the pulse plate,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
64
9.0
8.0
7.0
16.0
op
35
I
^ 5.0
0
i
“ C 1H 37C 1H
'■=4.0
(U
130 Volts
£ 3 .0
2.0
,CH37C1
|t
1.0 I
l
35ci 37cT
30 Volts
80 Volts
35£L” Cl
0.0 L
1.6E-051.7E-051.8E-051.9E-05 2E-05 2.1E-05
Time
Figure 3-6. Chlorine at several voltages during optimization of pulse
voltage
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
65
6
Relative Signal
5
3
Chlorine
2
Bromine J \
1
Iodine
10
20
30
40
50
60
70 80
Mass
90
100 110 120 130 140
Figure 3-7. Halogen spectra prior to pulse optimization
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
66
the spectra for the halogens shown in Figure 3-8, can be seen. The peaks
became much better resolved and much more intense.
3.8 Conclusions
The optimized system was produced a reliable mass calibration. A
robust mass calibration was obtained for the mass range of the intended
studies under the optimized conditions. The peaks are well resolved, intense
and of an appropriate peak shape. With these major tasks completed,
examinations of the kinetics of charge transfer could begin.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
g>
Chlorine
<o
a)
>
Bromine
Iodine
0
10
20
30
40
50
60
70
80
90
100 110 120 130
Mass
Figure 3-8. Halogen spectra after pulse optimization
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Chapter 4
INVESTIGATION OF THE KINETICS OF CHARGE TRANSFER
4.1 Introduction
The studies described in this chapter were performed to examine the
charge transfer plasma chemistry of nonmetals in a helium microwaveinduced plasma. Specifically, the effects on charge transfer while doping
argon into the helium plasma were studied. By determining the changes in
the plasma chemistry caused by these manipulations, additional knowledge
of the charge transfer process could be gained.
4.2 Charge Transfer
As was stated in Chapter 1, it is proposed that charge transfer (CT)
dominates nonmetal ion excitation in the He plasma. The general CT
reaction may be written:
He"(24.56 eV) + A°(0 eV)
He°(0 eV) + A**(24.56 eV± AE)
Eq. 4-1
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
69
A represents the analyte,0 and * indicate the ionization states, and *
indicates an electronically excited state. The energy defect (AE) is the
difference between the 24.56 eV He ionization energy and the nonmetal
ionization energy plus the excitation energy of A.
To study the charge transfer process and its extent, a specific set of
experiments was designed to alter the CT rates. Based on Equation 4-1, a
reduction in the He* population should reduce the population of A*'. It
should be possible to affect a He* reduction with the addition of a second,
more easily ionized, plasma gas. Argon meets this criterion. The addition
of Ar to a He discharge will reduce the He* number density and replace it
predominately with Ar*. Because the rate of He* CT with the nonmetal
analyte is proportional to the number density of He*, the overall CT rate
should be reduced. Reduction in the CT rate will also result in the reduction
of ion line intensities and ion populations. Simultaneously, the increase in
the atom population is expected to increase atom line intensities.
Probe species in these experiments are Cl and Br. It is predicted
that these species will have efficient charge transfer with He* and inefficient
CT with Ar*.
If there is a reduction of He* in the plasma as a result of increasing
Ar density, the nonmetal signal should diminish as a function of added Ar.
These experiments are outlined in the following sections of the chapter.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
70
4.3 Sample Introduction
Two methods for introducing gaseous samples were employed. The
methods were direct gas sampling and sampling of the headspace above a
volatile liquid.
4 .3 .1
S u b lim a tio n C h a m b e r
For the kinetic studies involving bromine, molecular bromine was
sampled from the headspace above a solution of bromine and squalane in
a sublimation chamber constructed at NIU. The chamber consisted of a 75
mL well with two ground glass ball and cup hose adapters and a third
sealable opening used to refill the sample. The hose adapters were to
allow access to the helium carrier gas. The well was filled with squalane to
allow for controlled vaporization of the sample. The rate of vaporization
was controlled by heating the chamber over water. Ventilation was installed
to remove excess bromine gas from the plasma area.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
71
4 .3 .2
G a s e o u s S a m p le I n t r o d u c t io n
Chlorine gas was introduced through the ceramic sampling tube in
the center of the torch. The flow rate of 99.995 % pure chlorine (BOC Gas,
St. Charles, IL) was less than 1 L/min and delivered from a lecture bottle.
Ventilation was installed to remove excess chlorine gas from the plasma
area.
4 .3 .3
S im u la t e d A q u e o u s S a m p le I n t r o d u c t io n
The sample gases, chlorine or bromine, were introduced in the
absence of added water. For reference, these experiments are referred to
as “dry." For the wet experiments, the analyte was introduced as described
above and water vapor from the USN was mixed at a quartz Y joint and
introduced to the plasma.
4.4 Atomic Emission Spectroscopy Experiments
With plasma atomic emission detection, light from the plasma is
focused on an entrance slit to a monochromator where a wavelength
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
72
corresponding to an analyte emission is selected and monitored. The
wavelength corresponds to a photon being emitted as an excited electron in
an ion or atom relaxes to a lower electronic energy level. From a
mechanistic standpoint, this approach has the virtue of being an electronic
state specific technique. Based upon the wavelength of the emitted photon,
not only does the operator know which elements are present, the operator
also knows which states are being excited. Since this technique is state
specific, it is extremely sensitive to population changes at that single state.
4 .4 .1
C h lo r in e
The UV-Vis optical spectroscopy apparatus is described in the
masters thesis of Yimin Fu.49 The 500 W discharge was used to observe
UV-Vis atomic emission. Ar was added incrementally from 0 to 0.5% to the
He plasma gas. Aqueous chloride was introduced as 1000 ppm solutions.
The Cl°* was monitored at 452.6 nm and CP* was monitored at 479.5 nm.
Emission intensities as a function of Ar concentration are plotted in Figure
4-1. The addition of 0.5% Ar to the helium discharge decreases the C f line
intensity by 60%. Under the same conditions, the Cl atom line increases by
a factor of 10. Although not shown, similar experiments doping the
discharges with Ne yields a 30% decrease in the CF* signal and a 3-fold
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
73
0.8
1.5
0.6
0.5
0.2
0
0.1
0.2
0.3
0.4
Percent Ar added to Plasma
0.5
Figure 4-1. Chlorine results from UV-Vis Argon doping experiment. The
diamonds are the ionic data and the squares are the atomic data. The lines
are added as visual aids only.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
74
increase in the Cl°* signal. In both sets of experiments, the ion emission
decreases and the atom emission increases, indicating that the Cl+*
populations decreases while the Cl°* population increases.
The VUV optical spectroscopy apparatus is described in the Ph. D.
dissertation of Patricia Brandi.50 The VUV system was used to monitor 1000
ppm aqueous chloride solutions. The Cl°* was monitored at 134.7 nm and
CP* was monitored at 107.1 nm. The results of this experiment are shown
in Figure 4-2. The addition of 1.0% Ar to the He discharge decreases the
CP* intensity by 60%. Under the same conditions, the Cl atom line
remained consistent throughout the range of added Ar. These ion emission
signal decreases and the constant atom emission signal indicates that CT
to the CT* state decreases while the Clc* state for this particular transition
remains relatively unchanged.
4 .4 .2
B r o m in e
Bromine was introduced as a 1000 ppm aqueous salt solution using
the UV-Vis system. Br°* was monitored at 447.8 nm and B f* was
monitored at 478.6 nm as Ar was added incrementally from 0 to 1.0% of the
plasma gas. As can be seen in Figure 4-3, the addition of 1.0% Ar to the
He discharge decreases the Br" line intensity by 60%. Under the same
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
75
0.8
0.8
6 0.6
0.6
0.4
0.2
0.2
0
0.2
0.4
0.6
0.8
1
1.2
Percent Ar added to plasma
1.4
1.6
Figure 4-2. Chlorine results from VUV Argon doping experiment. The
diamonds refer to ionic data and the squares refer to atomic data. The lines
are added as visual aids only.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
76
1
Br+ 478.6
nm
0.8
0.6
Brl°
0.4
0.5
0.2
0
0
0
0.2
0.4
0.6
0.8
1
Percent Ar added to plasma
Figure 4-3. Bromine results from UV-Vis Argon doping experiment. The
diamonds refer to ionic data and the squares refer to atomic data. The lines
are added as visual aids only.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
77
conditions, the Br atom line intensity increases by a factor of 2. This ion
emission decrease and the atom increase indicates that the Br+* population
decreases while Br°* population increases. Although not shown, similar
experiments doping with Ne yields a 30% decrease in the Brf* signal and a
2.5-fold increase in the Br°* signal.
4.5 Mass Spectrometry Experiments
The time of flight mass spectrometer was used to monitor the total
ion signal. Experiments similar to the AES kinetic studies were performed
using TOF-MS.
4 .5 .1
C h lo r in e
Chlorine gas was added to the 500 W plasma and chlorine ions were
monitored with the TOF-MS system using the same operating conditions
listed in Table 2-1. The amount of added argon was varied from 0 to 5 %
under dry chlorine sample introduction conditions. A plot of the relative
chlorine ion signal versus the argon percent is shown in Figure 4-4. The
data correspond to the integrated signal obtained from the mass 35 and 37
ion signals. Upon the addition of 0.8% Ar, the chlorine total ion signal
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
78
Chlorine signal
1.2
0.8
0.6
0.4
0.2
0
1
2
3
4
Percent Ar added to Plasma
Figure 4-4. Dry Chlorine results from MS Argon doping experiment. The
lines are added as visual aids only.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
79
decreased 63%. In the range of the AES experiments (0-1%), the chlorine
signal decreases about 60%. The Cl ion signal depletion is less than was
seen with AES with similar Ar concentrations.
In addition, chlorine gas was introduced to the plasma in the “wet"
mode. The chlorine ion signal was monitored and the results are shown in
Figure 4-5. As can be seen, more Ar is needed to see quenching
comparable to the dry chlorine MS experiment.
4 .5 .2
B r o m in e
Bromine was introduced to the TOF plasma system via head space
sampling of bromine in squalane as described in Section 4.3.3. The
bromine signal was monitored at mass 79 and 81 while 0 to 5 % Arwas
added to the plasma. The integrated signal from the bromine isotopes is
shown in Figure 4-6. The total amount of Ar added to the MS system is an
order of magnitude greater than that of the AES system. To obtain 60% Br*
suppression 0.79% Arwas added to the plasma, this is comparable to the
1% required for the AES experiments.
Bromine gas was introduced to the plasma along with nebulized
water using the same operating conditions. The bromine ion signal was
monitored and the results are shown in Figure 4-7. As can be seen, more
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
80
c 0.8
G)
CO
I 0.6
o
O 0.4
0.2
0
1
2
3
4
Percent Ar Added to Plasma
Figure 4-5. Wet Chlorine results from MS Argon doping experiment. The
lines are added as visual aids only.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
81
Bromine Signal
1.2 -
1.0 ♦
0.8
0.6J
0.41
0.2
0
T
»
T
T
1
2
3
4
5
Percent Ar added to Plasma
Figure 4-6. Dry Bromine results from MS Argon doping experiment. The
lines are added as visual aids only.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
82
Bromine
Signal
1.2
0.8
0.6
0.4
0.2
0
1
2
3
4
5
Percent Ar added to Plasma
Figure 4-7. Wet Bromine results from MS Argon doping experiment. The
lines are added as visual aids only.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
83
Ar is needed to see comparable quenching, as was seen in the wet and dry
chlorine MS experiments.
4.6 Plasma Kinetic Modeling
The proposed mechanism given in Equation 4-1 can be tested
through kinetic modeling of the plasma system. Although too many reaction
steps are involved to solve analytically, numerical modeling is feasible. We
have used the IBM chemical kinetics simulator for this purpose.51 This
program uses numerical, stochastic methods to calculate concentrations in
a kinetics scheme input by the user.
4 .6 .1
S im u la tio n S e ttin g s
Several parameters needed to be set before the simulations could
begin. The total number of molecules in the simulations were two million.
Data was recorded every 5000 reaction events. The total number of events
was not to exceed ten million. Reactions reached equilibrium, as
determined by visual inspection, after 1x104to 1x107 individual reaction
steps. This corresponds to plasma residence times of up to 15 |Js. Since
the average residence time in the plasma is approximately 2.5 ms (with the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
84
optimal flow rates and the known plasma volume), the equilibrium
concentrations are the values used for comparisons to experiment. The
actual computer time for each simulation ranged from three hours to 24
hours.
4.6.2
E x p la n a tio n s o f E q u a tio n s
The reactions included in the kinetics scheme are given in Table 4-1,
and are explained below. It should be emphasized that the actual number
of reactions involved is essentially infinite. For example, over 120 terms
have been experimentally observed for the chlorine atom. Therefore
approximations must be made.
In the following discussion, negative reaction numbers refer to
reverse reactions, “a” refers ro reactions in the chlorine system, and “b"
refers to reactions in the bromine system.
Reactions 1 and 2 represent ionization of the support gas in the
plasma. This ionization actually occurs through the following mechanism,
where Rg is a rare gas atom:
Rg + e' -> Rg* + 2e'
Eq. 4-2
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
85
Table 4-1. Listing of the reactions and rate coefficients used in the
Forward Rate { M
1a
He -►He+
1.5x10s
3x107
2a
Ar -> Ar”
1.5x106
3x108
3a
He” + Ar - He + Ar”
4.0x1 O'2
4a
He” + Cl - He + Cl”*
7.5x10,Q
5a
Cl”* -> Cl”
8.4x107
6a
Cl* - Cl
5.0x106
1.0x10“
7a
5.0x105
1.0x108
8a
Ar + C I - Ar + CI*
*
Reaction
0
1
o
simulations for chlorine.
IL )
Reverse Rate (M/L)
1.0x101°
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
86
Table 4*2. Listing of the reactions and rate coefficients used in the
Reaction
Forward Rate { M
1b
He -» He*
1.5x10s
3x107
2b
Ar -►Ar*
1.5x10s
3x10s
3b
He* + Ar -►He + Ar*
4.0x1012
4b
He* + Br - He + Br**
7.5x1010
5b
Br**
8.4x107
6b
Br* -> Br
5.0x10s
1.0x104
7b
Br - Br*
5.0x10s
1.0x10s
8b
Ar + Br -►Ar + Br*
o
X
— k
o
simulations for bromine
9b
Ar + Br ->■ Ar + Br*
1.0x109
Reverse Rate (M/L)
o
Br*
IL )
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
87
However, since the electron density is approximately constant, the electron
concentrations can be subsumed into the rate coefficients to give a pseudofirst-order description. The ratios of the forward and backwards rate
constants give 0.5% ionization in unperturbed plasmas, in agreement with
previous experiments.6
Reaction 3 does not directly describe a major reaction channel in the
plasma. It represents the preferential ionization of argon over helium, as
discussed in section 4.2, along with the recombination of He" with an
electron. The rate coefficient is set to best match the data, particularly the
decline in excited ion, either Cl"* and Br"*. Direct measurement of the loss
of signal for He" and the rise in signal for Ar" as a function of the argon
concentration would be a useful way to test the accuracy of this
approximation.
These first three reactions are sufficient to give a reasonable and
accurate description of a plasma in the absence of analyte, and should be
independent of the analyte. The remaining reactions are analytedependent. A key step is reaction 4, which for nonmetals is unique to
helium plasmas and the basis for this study. Reaction 4a is almost perfectly
resonant, meaning there is near perfect energy match and therefore the
rate coefficient is near the collision limit. Reaction 4b is less resonant, and
therefore the rate coefficient is lower.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reactions 5 and 6 are the photon emission reactions monitored by
UV-Vis and VUV techniques. The known rate coefficients for these
reactions33 are used in the simulation. The chlorine atom excitation energy
is used to calculate k 6assuming a Boltzmann distribution at the effective
plasma temperature. The reverse of reaction 5 is presumed to be
unimportant because of the relatively low population of the ion ground state.
Reaction 7 accounts for thermal ionization of chlorine atoms, with a
ratio of 200 between the forward and reverse rate coefficients to match the
He and Ar values. The absolute rate coefficients were adjusted to match
the experimental data.
Reaction 8 describes the overall effects of a large number of energy
transfer reactions. These involve the 3s23p54s excited states of Ar, which
are known to be populated in plasmas.15 These states have energy levels
of 11.55,11.62,11.72, and 11.83 eV. The 2P0 upper state of the chlorine
atom transition monitored by Fu et a/.49 is at 11.96 eV, which is near­
resonant with the excited argon states. Thus, reaction 8a has a high rate
coefficient. There are a large number of bromine atom states between
10.63 eV (the energy of the 2PXupper state of the monitored bromine atom
transition) and the ionization potential, 11.84 eV. These states, which span
the energy range of the argon excited state listed above, will in some cases
decay into the 2PXstate. Thus, the excitation transfer to the monitored
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
excited state of bromine will not be as fast as the analogous chlorine
reaction.
Reaction 9b is Penning ionization of Br by the first excited state of
Ar, a reaction that is near-resonant (AE = 0.01 to 0.29 eV). The reaction
rate coefficient is therefore fast. There is no equation for Penning ionization
in the Cl simulations because it is assumed to be negligible since the
ionization energy of Cl is 13.01 eV, giving a minimum energy defect of 1.18
eV.
4.6.3 D i s c u s s i o n
o f R e s u lts
The program does not allow for changing concentrations of a
reactant, in this case Ar. The simulation had to be performed by repeating
the simulations given in Tables 4-1 and 4-2 with varying initial
concentrations of Ar. The line, in the following figures, is a representation
of the simulation data of each species upon achieving equilibrium in the
repeated reactions.
The above description demonstrates that even a scheme with nine
reactions is a significant simplification of the actual situation. The
assumption that the electron density and plasma temperature are
essentially constant is also a possible source of inaccuracy. Nevertheless,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
the model developed here should be able to reproduce most of the features
of the experimental results after adjustment of the unknown rate
coefficients. If it were not possible to match the experimental data, that
would indicate that at least one significant aspect of the reaction scheme is
missing.
Given in Figure 4-8 are the AES experimental data for chlorine and
bromine neutrals with the simulation. Figure 4-9 is the plot of the AES data
from the bromine and chlorine ions, as well as the results of the simulations.
Figure 4-10 is the plot of the data from the MS experiments with the results
of the simulation. The simulated results match the experimental results
reasonably well, reproducing in particular the “elbow" in both total ion
population curves. It is not possible to achieve such agreement with any of
the nine reaction steps missing. The agreement suggests that the reactions
described above are in fact occurring, and that our current understanding of
plasma dynamics is a near match for the experimental observations.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0
0.25
0.5
0.75
1
Percent Ar Added to Plasma
Figure 4-8. AES data for neutral bromine (lower data) and chlorine (upper
data). The UV-Vis data is represented by the squares and the VUV data as
the closed triangles. The lines are from the simulations.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0
0.25
0.5
0.75
1
Percent Ar Added to Plasma
Figure 4-9. AES data for ionic bromine (upper data) and chlorine (lower
data). The UV-Vis data is represented by open triangles and the VUV data
as squares. The lines are from the simulations.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
93
0
1
2
Percent Ar Added to Plasma
3
4
Figure 4*10. Mass spectral data of bromine (upper data) and chlorine
(lower data) with the simulation data (solid lines). The lines are from the
simulations.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
94
4.7 Discussion of Results
4 .7 .1
C h lo r in e
The overall behavior of chlorine can be explained in terms of chlorine
CT with helium ion directly to the 3s3p5 3P° state being a significant
ionization mechanism for Cl. The reader is referred to Figure 4-11. The
3s3p5 3PCstate is the upper state of the 107.105 nm ion line. Reduction of
helium ions, via the addition of Ar, results in a population reduction of this
chlorine state and reduction of the intensity of the emission line arising from
it. Population of the 3s23p34p 5P state, the upper state of the 479.454 nm
emission line, is thought to occur from thermal excitation from the 3s23p4
3P° state. That both the 479.454 and 107.105 nm emission lines are
suppressed to a similar extent supports this argument.
The addition of argon causes a dramatic enhancement of the
452.621 nm chlorine neutral atom emission line. There are two possible
explanations for the large increase in emission intensity. It is possible that
circumvention of the CT process by the addition of argon allows atom
excitation to the upper state of the transition, the 3s23p35p2 2P° state, to
take place. It should be noted that this state is very near the 13.01 eV
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
95
24.56 eV
3s3p5 3P°
3s23p34p 5P
29.00 eV _
26.41 e v E Z Z I I 4 7 9 4 5 4 n m
. . . . . . . . . . . ^ . dP ..s.S .. . . . . . . . . . . . . . He+
107.105 nm
13.01 eV
3s23p43P
9.22 eV
.I1.-.9.6. ?y.. AS.2?P.4.5P.lp
.°................. Q+
9.22 eV
3s23p44s 2P
y * " " " " " 452.621 nm
J S ^ p “4S ^
134.724 nm
0.0 eV
3S23p5 2po
Figure 4-11. Chlorine transitions that were monitored in the VUV and UVVis Ar doping experiment. The light dashed lines are the energies
corresponding to the Ar metastable states from 11.55 to 11.83 eV.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
96
ionization limit of Cl. Enhanced atom signal from this state may indicate an
increased role of thermal ionization and subsequent ion-electron
recombination to high energy atom excited states caused by reduction of
charge transfer. Also, there is a near resonant energy match for this energy
state and a metastable state of Ar, as discussed in the previous section.
With the increase in Ar concentration, there is a greater chance for the
transfer of energy. That the increase in Cl* is nearly linear with Ar
concentration certainly lends validity to this argument. Further, that the
VUV neutral atom emission (134.724 nm) is slightly suppressed, rather than
enhanced, with the addition of argon supports both arguments. This
phenomenon indicates that the overall population of the 3s23p44s 2P state
(9.22 eV above the ground state) is not significantly altered.
The MS experiments fully support the atom and ion emission data.
In contrast to AES, which monitors specific ion or atom electronic states,
MS monitors total ion populations. The AES experiments indicate that the
higher energy ion electronic state populations are dramatically affected by
the addition of Ar; the MS experiments indicate that the addition of Ar in
equivalent amounts reduces the total ion population. These observations
indicate that the intense chlorine emission signals arise from a relatively
small fraction of the total set of chlorine species and that only select
chlorine ion states are overpopulated.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
97
4 .7 .2
B r o m in e
The behavior of Br can be explained by CT with He" directly to the
4s4p5 3P° Br+* state being a significant ionization mechanism. The reader is
referred to Figure 4-12. The addition of Ar reduced the populations of He",
resulting in a population reduction of this bromine ion state and a reduction
in the intensity of the emission line arising from that state. That the 478.6
nm emission line is suppressed supports this argument. While there is
suppression in the bromine total ion signal as seen in the MS experiments,
the signal never approaches zero. This may be explained by the possibility
of Penning ionization from the metastable state at 11.83 eV. Since the
energy match is near-resonant, it is expected that there will be a significant
contribution from this process. While Penning ionization is occurring, the
CT state is being reduced. Perhaps these processes in effect cancel one
another out to explain the plateau, rather than growth or decline of signal
represented in the data. It should be noted that a certain degree of thermal
ionization of Br occurs, regardless of the fraction of Ar. This phenomenon
may contribute to the plateau as well.
The addition of argon causes an enhancement of the 447.8 nm
bromine neutral atom emission line. It is possible that circumvention of the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
98
He+
26.08 eV
— 4s24p35p 5P2
f 475.4 nm
............................ a p a ; w a ..........................
I
7.86 eV *
447.8 nm
4s24p5 2P°
0.0 eV
Figure 4-12. Bromine transitions that were monitored in the UV-Vis Ar
doping experiment. The light dashed lines are the energies corresponding
to the Ar metastable states from 11.55 to 11.83 eV.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CT process by the addition of argon allows atom excitation to the upper
state of the transition, the 4s24p35p22P° state to take place. It should be
noted that this state is near the 11.84 eV ionization limit of Br. Enhanced
atom signal from this state may indicate an increased role of thermal
ionization and subsequent ion-electron recombination to high energy atom
excited states caused by reduction of charge transfer. Also, there is a slight
resonant energy match for this energy state and a metastable state of Ar,
as discussed in the previous section. With the increase in Ar concentration,
there is a greater chance for the transfer of energy. That the increase in Br*
is nearly linear with Ar concentration certainly lends validity to this
argument.
The MS experiments fully support the atom and ion emission data.
In contrast to AES, which monitors specific ion or atom electronic states,
MS monitors total ion populations. The AES experiments indicate that the
higher energy ion electronic state populations are affected by the addition of
Ar; the MS experiments indicate that the addition of Ar in equivalent
amounts reduces the total ion population, but to a much lesser degree than
the fractional reduction of the higher energy state ion populations. These
observations indicate that the intense bromine emission signals arise from a
relatively small fraction of the total set of bromine species and that only
select bromine ion states are overpopulated.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Chapter 5
CONCLUSIONS AND SUGGESTIONS FOR FUTURE WORK
5.1 General Conclusions
In this work, a time of flight mass spectrometer was constructed and
used to monitor the total chlorine and bromine ion signals. To examine the
charge transfer mechanism, the helium ion population was reduced by
adding a more easily ionized gas to the plasma. Upon the addition of 0.8%
Ar, the chlorine total ion signal decreased 63%. Upon addition of 0.8% Ar,
the bromine total ion signal decreased 47%.
In the UV-Vis experiments, the addition of 0.5% Arto the plasma
decreases the intensity of the CP* emission line by 60%, while the Cl atom
line increases by a factor of 20. In the VUV experiments, the addition of 1.0
% Ar to the plasma also decreased the CP* signal 60 %, while the Cl atom
line remained constant. Similar results were obtained with bromine.
A chemical kinetics simulator was also used in these studies to
model the plasma kinetics. Results, based upon the CT model, match the
experiment well, including some of the more interesting features. Similar
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
101
results were obtained for bromine. It was possible to achieve agreement
with 8 critical reactions for chlorine and 9 critical reactions for bromine.
The simulations verified that CT is the dominant mechanism for
nonmetal ionization in a helium plasma. In the course of modeling the
behaviors of the atoms, additional plasma chemistry was clarified. An
argon metastable state appears to be responsible for the increased
excitation of the monitored Cl atom line. During the addition of trace argon,
Penning ionization by the argon metastable state appears to be responsible
for the large bromine ion populations.
5.2 Future Work
To better understand the kinetics of the plasma three areas can be
explored: atomic emission, mass spectrometry and modeling.
Using atomic emission spectroscopy, the number densities for Ar ion
and metastable states can be determined. The number densities of the He
ion can also be determined along with the metastable states to see if there
are contributions from them. Further, other excited states of chlorine atom
and bromine could be monitored to determine if their behavior is consistent
with prior experimental and the modeling results.
Using mass spectrometry, other species, such as iodine, sulfur and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
102
phosphorus, could be determined for consistency with the proposed
nonmetal ionization mechanism, atomic emission results and modeling
results. For sulfur and phosphorus, the resolution of the instrument would
need to be greatly improved in order to achieve differentiation between the
analyte and atmospheric interferant species,
0 2
and HNCT.
The most immediate need for the simulation work is to test the
proposed mechanism with an additional step that includes water, to try to
model the behavior of the wet MS experiments. In addition to this, future
simulation work should include attempts with other species that exhibit or do
not exhibit charge transfer with the plasma species, such as magnesium
and manganese, to compare with AES data.
For the instrument, the detection limits of the TOF-MS should be
improved. With lower detection limits, the instrument could be interfaced
with liquid chromatography. If the resolution of the instrument can be
enhanced, biological compounds, such as amino acids, could be
determined with the LC-He-MIP-TOF-MS. Beyond that, the TOF-MS could
be interfaced with an electrospray ionization source for the purposes of
protein composition determination.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
REFERENCES
1. Cooper, W., Zikka., R., and Steinhauer, M., J .
77,116(1985).
A m . W a te r W o rk s A s s o c .
2. Dudragne, L., Adam, P., and Amaouroux, A p p l .
(1998).
3. Torrades, F., Riva, M., and Perez, M., A n a l .
(1996).
S p e c tro s c .
C h im . A c t a
52,1321
333,139
4. Lansens, P., Meuleman, C., Leemakers, M., and Baeyens, W., A n a l .
C h i m . A c t a 234,417 (1990).
5. Estler, C.,
(1990).
P h a r m a k o lo g ie a n d T o x ik o lo g ie
3ed, Schattauer, NY,NY
6. Coultate, T., F o o d T h e C h e m i s t r y o f i t s C o m
Graham House, Cambridge, UK (1989).
7. Carson, R.,
S ile n t S p r in g ,
8. Manahan, S., E n v i r o n m
Michigan (1991).
p o n e n ts 3 e d ,
Thomas
Houghton Mifflin, Boston MA, (1962).
e n ta l C h e m is tr y 5 e d ,
Lewis Publishers, Chelsea,
9. MacKay, D., M u l t i m e d i a E n v i r o n m e n t a l M o d e l s ,
CRC Press LLC, Boca Raton, FL (1991).
T h e F u g a c ity A p p r o a c h ,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
104
10. Ingle and Crouch, S p e c t r o c h e m
NJ (1988).
ic a l A n a ly s is ,
11. Wendt, R., and Fassel, V., A n a l .
C hem .
38, 337 (1966).
12. Greenfield, S., Salman, M., and Tyson, J.,
1087(1988).
13. Zumdahl, S.,
(1989).
C h e m is tr y ,
S p e c tr o c h im . A c ta
43B,
2ed., D.C. Heath and Co., Lexington, MA
14. Krall, N., and Trivelpiece, P r i n c i p l e s
Press, San Francisco, CA (1986).
15. Montaser, A., I n d u c t i v e l y
NY,NY (1998).
Chapter 8, Prentice Hall,
o f P la s m a
C o u p le d P la s m a ,
16. Myers, D., Li, G., Yang, P., and Hieftje, G.,
5,1008(1994).
P h y s ic s ,
San Francisco
Chapter 6, Wiley-VCH,
J. A m . S oc. M a ss
S p e c tro m .
17. Hill, S., I n d u c t i v e l y C o u p l e d P l a s m a S p e c t r o m e t r y
Chap 1, CRC Press LLC, Boca Raton, FL (1999).
18. Jarvis, I., and Jarvis, K.,
C h e m . G e o l.,
95,1 (1992).
19. Hill, S., I n d u c t i v e l y C o u p l e d P l a s m a S p e c t r o m e t r y
Chap 7, CRC Press LLC, Boca Raton, FL (1999).
20. Zander, A. and Hieftje, G.,
a n d it s A p p lic a tio n s ,
S p e c tr o c h im
A c ta
a n d it s A p p lic a tio n s ,
35B, 357 (1981).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
105
21. Montaser, A., I n d u c t i v e l y
NY.NY (1998).
Chapter 11, Wiley-VCH,
C o u p le d P la s m a ,
22. Estes, S., Uden, P., and Barnes, R., A n a l .
C hem .
23. George, M., Hessler, J., and Carnahan, J., A
(1989).
n a l A to m . S p e c tro m .
24. Quimby, B., Dryden, P., and Sullivan, J., A n a l .
25. Michlewicz, K., and Carnahan, J., A n a l .
L e tt.
53,1829 (1981).
C hem .
4, 51,
62,2509 (1990).
2 0 ,1193 (1987).
26. Luffer, D., Galante, L., David, P., Novotny, M., and Hieftje, G., A n a l .
C h e m . 6 0 , 1365 (1988).
27. Skelton, Jr., R., Farnsworth, P., Markides, K., and Lee, M., A n a l .
6 1 ,1815(1989).
28. Wu, M. and Carnahan, J., A p p l .
29. Osborne, S.,
S p e c tro s c o p y
S p e c tro s c .
C hem .
44, 673 (1990).
7, 37 (1992).
30. Douglas, D., French, J. A n a l .
C hem .
53, 37 (1981).
31. Pack, B., Broekaert, J., Guzowski, J., Poelhman, J., and Hieftje, G.,
A n a l . C h e m . 70, 3957 (1998).
32. Houpt, P. M.,
A n a l. C h im . A c t a
86, 129 (1976).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
106
33. Tanabe, K., Haruguchi, H. and Fuwa, K.,
(1981).
S p e c tr o c h im . A c ta
36B, 633
34. Moore, C ., A t o m i c E n e r g y L e v e l s a s D e r i v e d f r o m O p t i c a l S p e c t r a ,
NSRDS-NBS, Vol. 1(U.S. Department of Commerce, 1949).
Moore, C ., A t o m i c E n e r g y L e v e l s a s D e r i v e d f r o m O p t i c a l S p e c t r a ,
NSRDS-NBS, Vol. 2(U.S. Department of Commerce, 1952).
NIST Atomic Spectra Database, Version 2, March 22,1999
(http://physics.nist.gov/cgi-bin/AtData/main_asd)
35. Wu, M., Doctoral Dissertation, Northern Illinois University, 1990.
36. Carnahan, J. and Hieftje, G.,
S p e c tr o c h im . A c t a
37. Brandi, P. and Carnahan, J.,
S p e c tr o c h im
38. Jones, K. and Carnahan, J.,
S p e c tr o c h im ic a A c t a
39. Brandi, P. and Carnahan, J., A p p l .
40. Wiley, W., and McLaren, I.,
A c ta
S p e c tro s c
47B, 731 (1992).
49B, 105 (1994).
47B, 1229 (1992).
. 49,1781 (1995).
R e v . o fS c i. In s t . 2 6 ,
1150 (1955).
41. Barchick, C., Duckworth, D., and Smith, D., I n o r g a n i c
S p e c t r o m e t r y , Marcel Dekker, Inc., NY, NY (2000).
42. Hoaglund-Hyzer, C., and Clemmer, D., A n a l .
43. Wiley, W., and McLaren, J., R e v .
C hem .
S c i. I n s t r u m . ,
M ass
73,177 (2001).
26,1150 (1955).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
107
44. Myers, D., Yang, G., Hieftje, G., J .
(1994).
A m . S o c . M a s s S p e c tro m .
5,1008
45. de Hoffman, E., Charette, J., and Stroobant, V., M a s s S p e c t r o m e t r y
P r i n c i p l e s a n d A p p l i c a t i o n s , John Wiley and Sons, Inc., NY, NY (1996).
46.
V acuu m
T e c h n o lo g y : I t s F o u n d a t io n s F o r m u la e a n d T a b le s ,
Leybold
Inficon Inc., Export, PA (1995).
47. Schreiner, A., Doctoral Dissertation, Northern Illinois University, 1995.
48. Wollnik, H„
J . M a s s S p e c tro m .
34, 991 (1999).
49. Michelwics, K., and Carnahan, J., A n a l .
C h im . A c t a
183, 275 (1986).
50. Fu, Y., Masters Thesis, Northern Illinois University, 1996.
51. Brandi, P., Doctoral Dissertation, Northern Illinois University, 1996.
52. Hinsberg, W.; Houle, F. Chemical Kinetics Simulator version 1.01, IBM
Almaden Research Center, 1995
(http://www.almaden.ibm.com/st/msim/).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Документ
Категория
Без категории
Просмотров
0
Размер файла
3 314 Кб
Теги
sdewsdweddes
1/--страниц
Пожаловаться на содержимое документа